text
stringlengths 66
377k
|
|---|
Recognition of 3′-end L1, Alu, processed pseudogenes, and mRNA stem-loops in the human genome using sequence-based and structure-based machine-learning models
The role of 3′-end stem-loops in retrotransposition was experimentally demonstrated for transposons of various species, where LINE-SINE retrotransposons share the same 3′-end sequences, containing a stem-loop. We have discovered that 62-68% of processed pseduogenes and mRNAs also have 3′end stem-loops. We investigated the properties of 3′-end stem-loops of human L1s, Alus, processed pseudogenes and mRNAs that do not share the same sequences, but all have 3′-end stem-loops. We have built sequence-based and structure-based machine-learning models that are able to recognize 3′end L1, Alu, processed pseudogene and mRNA stem-loops with high performance. The sequence-based models use only sequence information and capture compositional bias in 3′-ends. The structure-based models consider physical, chemical and geometrical properties of dinucleotides composing a stem and position-specific nucleotide content of a loop and a bulge. The most important parameters include shift, tilt, rise, and hydrophilicity. The obtained results clearly point to the existence of structural constrains for 3′-end stem-loops of L1 and Alu, which are probably important for transposition, and reveal the potential of mRNAs to be recognized by the L1 machinery. The proposed approach is applicable to a broader task of recognizing RNA (DNA) secondary structures. The constructed models are freely available at github (https://github.com/AlexShein/transposons/). SINE RNA remains unclear 8 . For several species it was experimentally shown that the LINE protein recognizes a secondary structure, such as a stem-loop, at the 3′-end of a transposon RNA. Moreover, it was shown that in some organisms, LINEs and SINEs have identical 3′-end sequences containing a stem-loop structure, which is essential for transposon RNA recognition by transposon proteins 9-11 . In case when transposons do not have an identical 3′-end, it is thought that a poly-A tail serves as a recognition element of SINEs and LINEs. However, almost all mRNAs have poly-A tails and this does not explain the specific recognition of transposon sequences by the LINE-encoded proteins. We discovered a conserved secondary structure at the 3′-end of human L1 (LINE) and Alu (SINE) transposons 12 , as well as in different species across the tree of life (unpublished results). Despite the absence of similarity at the level of sequences, the conserved position of this structure suggests its functionality.Solution 3D-structures of stem-loops for LINE transposons in eel and zebrafish revealed some characteristic structural stem-loop properties that, together with mutagenesis analysis, showed which nucleotide positions are vitally important for retrotransposition. Thus, the solution structure of RNA stem-loop from 3′UTR of eel LINE transposons UnaL2 determined the specific structure of the loop GGAUA 13 . The fourth uridine appeared to be exposed to solvent, but mutation of this residue did not affect retrotransposition. The third and fifth adenosine residues are stacked, and their mutations decreased the activity. Mutation of the second guanosine diminished retrotransposition activity. The authors hypothesized that the second guanosine in the loop is directly recognized by UnaL2 reverse transcriptase (RT).Another study showed that an internal loop is also required for retrotransposition 14 because its deletion completely blocked the activity. With the help of molecular dynamic simulations the authors showed that an internal loop acts as a hinge allowing stem-loop to have conformational flexibility. Mutational analysis showed that this flexibility is somehow required for transposition.Structures of stem-loops of two transposons from zebrafish ZfL2-1 and ZfL2-2 were examined 9 . Stem-loop in one transposon ZfL2-1 has an insertion in the form of the second stem-loop at the place of the 4 th residue in the loop in the ZfL2-2. Thus, the loop of ZfL2-2 became an internal loop in ZfL2-1. The experiments with mutual loop replacements with the stem region showed that retrotransposition was almost reduced to zero in both cases. This points to a specificity in loop recognition. An importance of loop and internal loop in stem-loop recognition for zebrafish ZfL2-1 and ZfL2-2 transposons, and specifically of certain residues, was also demonstrated in this study15.Here, we investigated the structural stem-loop properties of human L1 and Alu transposons with the help of two types of machine-learning models: one is built using information about sequence composition and the other using structural, physical, and chemical properties of ribonucleic acid molecules. Taking into account the importance of position-specific nucleotides in loops and in bulges, we treated structural characteristics of stems and loops separately. For stem characteristics we used dinucleotide properties available from DiProDB database 16 : Gibbs energy, enthalpy, entropy, hydrophilicity, and geometrical structural parameters of RNA molecule: shift, slide, rise, tilt, roll, and twist. For loops and bulges we fixed positions in order to capture position-specific preferences for nucleotides in certain positions.Dinucleotide characteristics were already used in machine-learning models to predict specific genomic sites. For example, protein binding was predicted based on DNA dinucleotide properties 17,18 . Also, using dinucleotide physical and geometrical properties it was possible to build machine-learning models, which recognize recombination sites 19 , splice-sites 20 , regulatory small RNA originated from transposon sequences 21 , DNA-editing sites 18 with high accuracy. Sequence-based models using frequencies of di-and tri-nucleotides were successfully applied to predict quadruplexes 22 .The goal of the present study is to construct machine learning models that can be trained to separate stem-loop structures at the 3′-end of human L1 and Alu transposons from stem-loops from shuffled sequences, and also to test constructed models on mRNAs and processed pseudogenes. We built two types of models -sequence-based models, which use information on nucleotide composition, and structure-based models that consider physical, chemical and geometrical properties of a stem and position-specific nucleotides of a loop and a bulge. Both models were able to recognize L1 and Alu 3′-end stem-loops with high performance (97-99% AUC), and 3′-end stem-loops of L1 and mRNAs or processed pseudogenes with moderate performance. Structure-based models helped to reveal properties that are most important for model prediction power, which appeared to be shift, rise, and hydrophilicity, as well as stem positions such as the dinucleotides adjacent to the loop and the bulge. We discuss the biological significance of top influential characteristics for RNA-protein recognition. The constructed models can be used for de novo discovery of retrotransposon-related stem-loop structures.Recognition of 50 bp 3′-end of Alus and L1s. Before building the model to recognize specifically 3′-end stem-loops, we first built models trained to recognize the last 50 bp of Alu and L1 sequences when compared to shuffled sequences. We built machine learning models based on sequence composition, taking into account di-and trinucleotide frequencies (see Methods). As an alternative class we shuffled sequences using dinucleotide-shuffling algorithm. We implemented six models comparing either L1 or Alu or their joint set with shuffled sequences, and also tested if L1 and Alu 3′-end or L1 3′UTR and 5′UTR can be distinguished. The results of the modeling are presented in theTable 1andFig. 2. All six constructed models achieved high performance with AUC> = 97%. They showed that not only 3′-end sequences of both L1 and Alu transposons could be separated from the other genomic regions, but also L1 3′UTR is distinguishable from 5′UTR.Feature importance analysis(Fig. 2c) revealed important di-and tri-nucleotides for each model. The analysis revealed that different groups of di-and tri-nucleotides are detected as the most influential in recognition of 3′-end 50 bp L1 from shuffled and Alu from shuffled sequences. The special interest of this study is recognizing the combined set of L1 and Alu 3′-end 50 bp and extraction of common features characteristic to this set. For the case of 3′-end 50 bp sequences, these features are mostly trinucleotides rather than dinucleotides, and mostly GC-rich trinucleotides reflecting the existence of compositional bias of 3′-ends of both transposons.
Transposons are pieces of DNA, which are able to multiply and move inside a genome. Transposons have been found in all eukaryotes and usually occupy a considerable part of any genome (46% of a human, 37% of a mouse, 10% of a fruit fly, and 85% of a corn genome) [bib_ref] Active transposition in genomes, Huang [/bib_ref] [bib_ref] Initial sequencing and analysis of the human genome, Lander [/bib_ref]. For a long time, transposons were considered as junk DNA, which do not have any function. However, after discovering haemophilia that was caused by transposon insertion into the gene of a coagulation factor 3 , the pathogenic role of transposons became evident. Nowadays, 96 human diseases caused by a transposon insertion have been discovered [bib_ref] Mobile elements in the human genome: implications for disease, Solyom [/bib_ref]. Many studies were devoted to the role of transposons in cancer formation [bib_ref] Active transposition in genomes, Huang [/bib_ref] [bib_ref] Active human retrotransposons: variation and disease, Hancks [/bib_ref]. Transposons made an important contribution to the variability of genomes of many organs, for example, the brain and the immune system. Transposons started to be considered as a "tool" of evolution, because they cause large-scale genome re-arrangements (for example, recombination between two non-allelic elements in two different chromosomes) as well as minor genome changes (duplications, inversions, deletions) [bib_ref] Mobile elements: drivers of genome evolution, Kazazian [/bib_ref]. Transposons can influence their own expression as well as the expression of the nearby genes [bib_ref] The Influence of LINE-1 and SINE Retrotransposons on, Richardson [/bib_ref].
For a transposition, long interspersed nuclear elements (LINEs) use their own molecular machinery encoded in the LINE sequence. This machinery is employed for copying the sequence and for its subsequent insertion into the genome. Short interspersed nuclear elements (SINEs) are parasites, they do not encode proteins, and they require the machinery of LINE to perform transposition. How LINE proteins recognize their own RNA and
# Results
Recognition of 3′-ends and 3′-end stem-loops of L1s and Alus. We took full-length sequences of 6,622 L1 elements and 12,431 Alu elements (see Methods) from two evolutionary studies identifying active families of transposons in the human genome [bib_ref] Molecular evolution and tempo of amplification of human LINE-1 retrotransposons since the..., Khan [/bib_ref] [bib_ref] Whole-genome analysis of Alu repeat elements reveals complex evolutionary history, Price [/bib_ref]. We annotated all transposons with stem-loop structures (see Methods) and composed sets of stem-loops located at the 3′-end of transposons. Additionally, we composed sets of stem-loops from 5′-end regions of both Alu and L1 transposons and created shuffled sequences as an alternative class (see Methods).
The 3′UTR regions of L1 transposons are poorly conserved regions that lack sequence similarity even in members of one family. [fig_ref] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences [/fig_ref] presents dotplots of 50 bp sequences from 3′-ends of two L1s of L1-Ta family and of representatives from L1-Ta and AluY families. However, as we showed earlier, despite the absence of similarity at the sequence level, all L1 and Alu transposons possess a 3′-end stem-loop structure [bib_ref] Conserved 3′ UTR stem-loop structure in L1 and Alu transposons in human..., Grechishnikova [/bib_ref]. This observation raises a question about the evolutionary importance of this structure in the process of transposition and whether evolutionary structural constrains exist that can be detected by computational methods.
Recognition of 3′-end stem-loops of Alus and L1s with sequence-based models. Next we extracted sequences from 3′-end stem-loops and built two types of models -using only sequence characteristics and only structural, physical and chemical characteristics and position-specific sequence information about loops and bulges (see Methods). As an alternative class we used shuffled sequences.
The results of the modeling using only sequence characteristics to recognize 3′-end stem-loops are presented in and [fig_ref] Figure 3: Recognition of 3′-end stem-loops of L1 and Alu elements with sequence-based models [/fig_ref] ,b. In recognizing L1 3′UTR or Alu 3′-end stem-loops or their combined set from stem-loops from shuffled sequences all models achieved performance AUC> = 0.96. In general, sequence-based models for recognition of stem-loops show the same performance as models recognizing 50 bp-ends of transposons, which are the expected results.
Feature importance analyses [fig_ref] Figure 3: Recognition of 3′-end stem-loops of L1 and Alu elements with sequence-based models [/fig_ref] revealed that top-10 most important features are mostly dinucleotides both for recognizing L1 and Alu 3′-end stem-loops jointly or separately, and these are inverse complementary pairs of dinucleotides such as CA-GT, TC-GA for the experiments L1 vs shuffled and AG-CT, TC-GA for Alu vs shuffled reflecting inner constraints of the analyzed sequences.
Recognition of 3′-end stem-loops of Alus and L1s with structure-based models. Finally, we built machine learning models considering structural and physical properties of a stem and taking into account bulges and loops (see www.nature.com/scientificreports www.nature.com/scientificreports/ Methods). We considered position-specific dinucleotide properties of a stem including geometrical characteristics such as twist, rise, tilt, bent, shift and slide and also physical and chemical characteristics such as enthalpy, entropy, Gibbs free energy, and hydrophilicity. For loops we considered position-specific sequence composition taking into account only the first five positions of the loop creating 20 binary features describing a loop. For bulges, we considered a bulge size of 3 nucleotides from the left and/or from the right part of a stem. All bulge positions also were fixed and 24 binary features were created to characterize a bulge (see Methods). The results of the modelling are presented in .
Feature Importance analysis revealed that L1 and Alu 3′end stem-loops are recognized from shuffled stem-loops based on almost different sets of top-10 features, however the analysis of the joint set of L1 and Alu 3′-end stem-loops captured the properties that are important for both transposons.
The structural characteristics shift, roll, and tilt appeared to be the most important parameters in recognizing L1 3′UTR stem-loop from stem-loops from shuffled sequences (shuffled stem-loops) at two stem positions counting from the loop (positions LS0, close to the loop, and LS3, close to the bulge in [fig_ref] Figure 4: Recognition of 3′-end stem-loops of Alu and L1 elements with structure-based models [/fig_ref]. In distinguishing Alu from shuffled stem-loops, many parameters are identified as important for the first stem position counting from the loop (position LS0 in [fig_ref] Figure 4: Recognition of 3′-end stem-loops of Alu and L1 elements with structure-based models [/fig_ref] , and they include energy characteristics such as enthalpy and free energy; geometrical parameters rise and tilt, and also hydrophilicity. For the combined set of L1 and Alu 3′-end stem-loops, the top-10 most important parameters also include the parameter rise for two stem positions, shift, tilt, roll, and hydrophilicity for four stem positions [fig_ref] Figure 4: Recognition of 3′-end stem-loops of Alu and L1 elements with structure-based models [/fig_ref]. www.nature.com/scientificreports www.nature.com/scientificreports/ When looking only at positions having at least one important parameter, the first position was identified as significant for three parameters: hydrophilicity, rise and tilt (position LS0 in [fig_ref] Figure 4: Recognition of 3′-end stem-loops of Alu and L1 elements with structure-based models [/fig_ref] ; the fourth position (LS3) showed up for two parameters shift and roll. Four out of ten important features for recognition of a joint set L1-Alu 3′-end stem-loops are hydrophilicity at positions close to the loop (LS0-LS1) and at positions close to the base of the stem (LS6-LS7).
The summary of parameters and positions revealed as significant in all six experiments is presented in [fig_ref] Figure 5: The Top-10 features for recognizing L1 and Alu [/fig_ref]. The most influential parameter is parameter hydrophilicity (appeared 11 times) followed by the loop-specific positions (9), rise (9), tilt (6), and shift (5) [fig_ref] Figure 5: The Top-10 features for recognizing L1 and Alu [/fig_ref]. Three stem positions were highlighted as important: first dinucleotide position adjacent to the loop (LS0, 17 times), position close to a bulge (LS3, 10 times), and positions close to a stem base LS6-LS7 (7 times) [fig_ref] Figure 5: The Top-10 features for recognizing L1 and Alu [/fig_ref].
Recognition of 3′-ends and 3′-end stem-loops of processed pseudogenes and mRNAs. Processed pseudogenes are gene copies that result from the process of mRNA retrotransposition. We examined processed pseudogenes and mRNAs for the presence of stem-loops at their 3′-ends (see Methods). The human genome contains around ten thousand processed pseudogenes. We found that 62% (6,309 out of 10,144) of all processed pseudogenes have stem-loops at the 50 bp 3′-ends. Distribution of the number of copies of processed pseudogenes www.nature.com/scientificreports www.nature.com/scientificreports/ in the human genome follows the power-law distribution, when few genes have multiple copies and many genes only few [bib_ref] Millions of years of evolution preserved: a comprehensive catalog of the processed..., Zhang [/bib_ref]. Ribosomal proteins account for 22% of all processed pseudogenes, that is why we additionally checked for ribosomal proteins and found that around 70% of those have stem-loops at their 3′-end (194 out 281). We also examined all cellular mRNAs for the presence of stem-loop structures, and found that 64% of all mRNAs (53,094 out of 82,960) have stem-loops at the 50 bp 3′-ends. When counting only protein-coding mRNAs, their percentage increased up to 68% (20,328 out of 29,636). L1-recognizing model incorporates properties inherent only to L1, and Alu-recognizing model incorporates properties inherent only to Alu; it is evident from Figs 2-4 that they do not overlap. The joint Alu + L1 model incorporates properties inherent to both transposons and we further checked whether the models, trained to recognize the joint set of L1 and Alu 3′-ends or 3′-end stem-loops, are able to recognize 3′-ends or 3′-end stem-loops of processed pseudogenes, ribosomal proteins and mRNAs [fig_ref] Figure 6: Percentage of processed pseudogenes, ribosomal proteins and mRNAs recognized by models trained... [/fig_ref]. It appears that the model, trained to recognize L1-Alu 50 bp 3′-ends, recognizes around 40% of ribosomal proteins and mRNAs and only 17% of processed pseudogenes. However, models trained to recognize L1-Alu 3′-end stem-loops could recognize around 50% of stem-loops of mRNAs and ribosomal proteins, and around 30% of processed pseudogenes, if using the sequence-based model; the structure-based model behaves as the 50 bp-end model. A low percentage of recognition of processed pseudogenes is explained by the fact that the processed pseudogenes are "dead on arrival" and quickly accumulate mutations; some are even 3′-end truncated. What we can conclude from this experiment is www.nature.com/scientificreports www.nature.com/scientificreports/ that at least half of mRNA has common recognizable properties with the joint L1-Alu set, and these properties are shared for both 3′-ends and for 3′-end stem-loop structures.
To obtain information about specific common properties of L1 and mRNA/processed pseudogenes, we repeated the same set of experiments, i.e. those performed for L1s and Alus, for processed pseudogenes and mRNAs. We built models to recognize 50 bp 3′-ends and 3′-end stem-loops of processed pseudogenes and mRNAs together with L1s or separately and extracted the most important parameters of the models. The modeling results are presented in and in Figs 7-9.
In the case of processed pseudogenes and mRNAs we do not observe the same high performance as that of the models built using the same approach for Alus. Characteristically, models trained to recognize 3′-ends or 3′-end stem-loops of processed pseudogenes only have the worst performance: AUC of 0.65 in recognizing 50 bp 3′-ends and AUC of 0.59 for recognizing 3′-end stem-loops. This can be explained by the fast rate of accumulating mutations by the processed pseudogenes. In comparison, 3′-ends of mRNAs can be distinguished with a higher performance (AUC = 0.76), and mRNA 3′-end stem-loops are recognizable even better with structure-based models (AUC = 0.82). The combined set of 50 bp 3′-ends from L1s and processed pseudogenes can be distinguished from shuffled sequences with a high performance (AUC = 0.95). Feature importance analysis revealed [fig_ref] Figure 7: Recognition of 3′-end 50 bp of processed pseudogenes and mRNAs [/fig_ref] that the important parameters are mostly AT-rich di-and trinucleotides, and almost the same set of di-and trinucleotides are identified as significant in the joint set of mRNAs and L1s.
The analysis of 3′-end stem-loops showed that important features are grouped in pairs of processed pseudogenes and mRNAs, both in sequence-based and structure-based models [fig_ref] Figure 8: Recognition of 3′-end stem-loops of of processed pseudogenes and mRNAs with sequence-based... [/fig_ref] : either processed www.nature.com/scientificreports www.nature.com/scientificreports/ pseudogenes and mRNAs are compared to shuffled sequences, L1s, or they are combined in a joint set with L1s. In distinguishing joint sets of mRNAs and L1s the structure-based models showed very high performance (AUC = 0.99) compared to sequence-based models (AUC = 0.85). The difference in the recall is considerably higher (0.93 vs 0.75, see , indicating that the model is well trained to specifically recognize stem-loops of interest. The most important features in recognizing either joint set of L1s-mRNAs or L1s-processed pseudogenes were the shift, roll, hydrophilicity, which appeared to be the same salient features in models for recognizing the joint L1 and Alu sets. Processed pseudogenes have additionally entropy and enthalpy parameters included in the top-10 features, and mRNAs have four loop and bulge positions.
The summary of parameters and positions identified as significant in all experiments with processed pseudogenes and mRNAs is presented in [fig_ref] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences [/fig_ref]. The most influential parameter is hydrophilicity (showed 16 times), followed by shift [bib_ref] Solution structure of an RNA stem-loop derived from the 3′ conserved region..., Baba [/bib_ref] and the loop-specific positions (13) [fig_ref] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences [/fig_ref]. Three stem positions were highlighted as important: position close to a bulge (LS3, 11 times), position close to a loop (LS0, 8 times), and position close to a stem base LS8 (8 times) [fig_ref] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences [/fig_ref]. The same parameters and positions, though in a different order, were identified in the models of L1-Alu recognition [fig_ref] Figure 5: The Top-10 features for recognizing L1 and Alu [/fig_ref].
# Discussion
In the present study, we explored the ability of machine-learning models to recognize 3′-ends and 3′-end stem-loops of active transposons in human genome: L1 and Alu separately and as a combined set as well as L1 and Alu or L1 5′UTR and 3′UTR between each other. We evaluated two types of machine-learning models using different types of features, encoding sequence composition and physical, chemical, and geometrical properties of dinucleotides, calculated from different experiments. Sequence-based models use nucleotide composition of an entire sequence, while structure-based models take into account structural properties of a stem and position-specific properties of loops and bulges.
Here there is a need to clarify that, in essence, the so-called structure-based model also uses sequence information but is mapped to a different alphabet of dinucleotide characteristics. Even though all these characteristics have been obtained from experiments, the experiments were done for a narrow class of RNA duplexes. Thus, to obtain structural properties of RNA dinucleotides X-ray crystallographic images of RNA duplexes were analyzed and later filtered for structures containing proteins, drugs, mismatches, overhanging bases or unusual bases in canonical Watson-Crick pairings [bib_ref] The relative flexibility of B-DNA and A-RNA duplexes: database analysis, Perez [/bib_ref]. The authors performed helical analysis of the remaining structures at the canonical base-pair level. The structures that showed deviations (more than three standard deviations) of helical parameters were also removed from the analyzed dataset [bib_ref] The relative flexibility of B-DNA and A-RNA duplexes: database analysis, Perez [/bib_ref]. As a result, the resulting translational (shift, slide and rise) and rotational (tilt, roll and twist) parameters were shown to correspond to averages for crystallographic static images of naked RNA. As was stated in 26 a sizeable amount of experimental data was removed due to filtering, and it guarantees that perturbations were within the harmonic limit. Also, the filtering removed major part of AT dinucleotides, whose presence corresponds to flexibility. Entropy, enthalpy and free energy used in this study were taken from [bib_ref] Thermodynamic parameters for an expanded nearest-neighbor model for formation of RNA duplexes..., Xia [/bib_ref] , and these parameters are derived from optical melting of RNA duplexes taking into account compositions of ends. Hydrophilicity for 16 dinucleotides were also calculated from experiments as described in [bib_ref] Further studies on the chromatographic behaviour of dinucleoside monophosphates, Barzilay [/bib_ref].
All of the selected features correspond to dinucleotides and, in fact, in the model we use numerical representations of the sequence-based information. However, this approach has two advantages. First, machine-learning algorithms work better with numerical values rather than with binary one-hot encoding or integers resulting from counts of k-mers. Secondly, they can capture some tendencies related to dinucleotides and thus can be interpreted. . Recognition of 3′-ends and 3′-end stem-loops of mRNAs and processed pseudogenes. Note: 50 bp designates models for recognizing 50 bp 3′-ends, SL1 designates models for recognizing 3′-end stem-loops with sequence-based models, and SL2 designates models for recognizing 3′-end stem-loops with structure-based models. www.nature.com/scientificreports www.nature.com/scientificreports/ In a set of L1-Alu experiments, both models showed comparable performance (AUC > 96-99%), however feature importance analysis of the structure-based models allowed revealing properties that are more influential for L1 and Alu 3′-end stem-loops. Among these properties are properties shift, rise, tilt, and hydrophilicity.
We also examined processed pseudogenes and cellular mRNAs for the presence of stem-loop structures at their 3′-ends and constructed the same sets of machine learning models as those for Alus. Those models that were trained to detect 3′-ends or 3′-end stem-loops in processed pseudogenes, achieved performance no higher than 0.65, which can be explained by the fact that the processed pseudogenes are considered "dead on arrival", and accumulate mutations without any selection constrains. The performance of other models ranges from 0.76 for recognizing 50 bp mRNAs to 0.99 for recognizing 3′-end stem-loops of the joint set of L1s and mRNAs with structure-based models, or for distinguishing between L1 3′-end stem-loops from those of mRNA or of processed pseudogenes. Three important parameters in recognizing joint sets of L1s and pseudogenes or mRNAs are similar to those in recognizing joint sets of L1s and Alus: shift, roll, hydrophilicity and specific loop positions. Also the same specific stem positions showed as significant in all sets of experiments, either for L1-Alu, L1-mRNA, or L1-processed pseudogenes: positions close to a bulge, positions close to a loop, and positions close to the stem base.
The importance of stem-loop structural parameters was shown earlier in a series of experiments on transposons of different species. Experimental work on solution structures of the 3′-end stem-loop in eel transposons www.nature.com/scientificreports www.nature.com/scientificreports/ revealed that the sequence GGRNA of the loop forms the structure that is recognized by UnaL2 RT 13 . One of the special characteristics of this structure is that the fourth residue, uracil in the case of the investigated loop, is exposed to the solvent, however, its replacement does not affect retrotransposition. However, mutation of the second residue, guanine, abolished the retrotransposition activity. This led to the conclusion that this residue is required for a specific recognition of the UnaL2 RT. The deletion of the fourth residue results in the sequence of GGAA, which becomes a well-known tetraloop GNRA [bib_ref] A network of heterogeneous hydrogen bonds in GNRA tetraloops, Jucker [/bib_ref]. Solution structures of three tetraloops GAGA, GCAA and GAAA revealed that the three last bases are usually stacked and the first base is located on the other side from three stacked bases. This arrangement creates a sharp turn between the first and the second nucleotides resembling U-turn [bib_ref] GNRA tetraloops make a U-turn, Jucker [/bib_ref] , known as a motif for tertiary interactions.
In zebrafish, 3′-end stem-loops in two transposons ZfL2-1 and ZfL2-2 were investigated which part of stem-loops are important for retrotransposition [bib_ref] Mechanism by which a LINE protein recognizes its 3′ tail RNA, Hayashi [/bib_ref]. Two stem-loops differ from each other by ZfL2-1 having an insertion in the form of another stem-loop in the position of the fourth residue of the loop. First, RT transposes only cognate stem-loop. Experiments with changing loops and stems creating hybrid stem-loops composed of different stems and loops revealed that replacement of a loop and leaving cognate stem leads to complete abolishment of retrotransposition while leaving cognate loop and replacing stem reduced transposition rate only by half. This suggests the loop is important for RT specific recognition. www.nature.com/scientificreports www.nature.com/scientificreports/ The crystal structure of NHL, RNA-binding domain, from fly Brat protein in complex with RNA showed an interaction, where a guanine base is turned in the opposite direction compared to other motif bases, and it protrudes into a binding pocket [bib_ref] The Crystal Structure of the NHL Domain in Complex with RNA Reveals..., Loedige [/bib_ref]. The crystal structure of NHL domain from worm LIN-41 protein revealed 32 that domain binds to stem-loop which shape determines the binding specificity. This stem-loop consists of three bases and all three loop nucleotides −1 and +1 bases of the stem are participating in specific binding. Thus, NHL domain in the fly and in the worm has different binding mechanisms suggesting high evolutionary plasticity.
The crystal structure of stem-loop binding proteins in complex with stem-loop and 3′hExo revealed that the entire stem-loop and its flanking sequences participate in binding with two proteins [bib_ref] Structure of histone mRNA stem-loop, human stem-loop binding protein, and 3′hExo ternary..., Tan [/bib_ref]. It was shown that the proteins recognize the shape of the stem-loop rather than sequence however sequence determines the shape. The specificity of recognition is provided by the nucleotides in the loop: the first (U) and the third (U) nucleotides are highly conserved.
Here, we probed RNA structural parameters characteristic for stem-loops at the 3′-end of L1 and Alu sequences, as well as at the 3′-ends of processed pseudogenes and mRNAs, with machine learning models. The results of the modeling revealed both stem and loop parameters among the top 10 most important features either in recognizing only L1 and only Alu 3′-end stem-loops or a combined set of Alu and L1 elements. The same is true for a joint set of L1s and processed pseudogenes or mRNAs, while in recognizing only processed pseduogenes or only mRNAs most important parameters include only hydrophilicity and loop positions.
The parameter shift as well as rise and slide are translational helical parameters as opposed to rotational (tilt, roll and twist) [bib_ref] The relative flexibility of B-DNA and A-RNA duplexes: database analysis, Perez [/bib_ref]. Shift reflects the translation around X-axis, slide -translation around Y-axis and rise -translation www.nature.com/scientificreports www.nature.com/scientificreports/ around Z-axis [fig_ref] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences [/fig_ref]. They influence the height and width of the double-stranded RNA that would be reflected in the height and width of the stem. All of them contribute to the flexibility of A-RNA, which in case of transposon stem-loops appeared to be a subject for evolutionary selection. The importance of the energy parameters of the dinucleotide adjacent to the loop as in the case of Alu stem-loops reflects structural peculiarity of Alu families since the energetic parameters did not appeared as influential in L1 stem-loops. Cellular mRNA is a class of very diverse objects consisting of protein-encoding RNA, tRNAs, rRNAs and various classes of regulatory RNAs. This explains why the models could not capture any regularities for dinucleotide properties at specific positions of stems. Instead the most important parameters are either hydrophilicity or loop positions.
The conducted experiments cannot not fully answer the question about the role of stem-loop in pseudogenization of mRNA, but they assert such a role exists: first, the 3′-end stem-loop is a property of around 70% of mRNA; secondly, around 50% of mRNA can be recognized by a model based on the common properties of either 50 bp 3′-end or 3′-end stem-loop of the joint L1-Alu set; thirdly, stem-loop structures are recognized better than just 50 bp-ends; fourth, the joint set of L1 and mRNA is recognized with a good performance (AUC > 0.9) reflecting the existence of common properties for that set. The set of ribosomal proteins that account for 22% of all processed pseudogenes percentagewise are recognized no better than mRNA. This probably reflects randomness in selection of those genes that become retropseudogenes.
The trained models have limitations: feature set is preselected and might not be the optimal. As of today there does not exist a standard machine learning approach to predict RNA (DNA) secondary structures. Different machine learning methods have their advantages and disadvantages. For example, quadruplexes were predicted with high performance with eXtreem Gradient Boosting method 22 based on sequence composition only (k-mers frequencies), using experimental data from G4-seq for training [bib_ref] High-throughput sequencing of DNA G-quadruplex structures in the human genome, Chambers [/bib_ref]. Lately, CNN has shown to be slightly in advance as compared to other machine-learning approaches. The advantage of CNN is that there is no need to select features for input data, they are extracted automatically from an input sequence. Usually an input for CNN is a DNA sequence transformed with one-hot encoding to binary matrix. Several studies incorporated RNA structural information into sequence data [bib_ref] Convolutional neural networks for classification of alignments of non-coding RNA sequences, Aoki [/bib_ref] [bib_ref] A deep neural network approach for learning intrinsic protein-RNA binding preferences, Ben-Bassat [/bib_ref]. In recognizing non-coding RNA with CNN, the input data also included probabilities of a nucleotide being either paired or non-paired [bib_ref] Convolutional neural networks for classification of alignments of non-coding RNA sequences, Aoki [/bib_ref]. When predicting RNA-protein binding the probabilities were ascribed to five states: a nucleotide in the hairpin loop, inner loop, multi loop, external region, or in the paired state [bib_ref] A deep neural network approach for learning intrinsic protein-RNA binding preferences, Ben-Bassat [/bib_ref]. In both methods, adding structural information improved CNN performance. However the disadvantage of neural networks is difficulty of interpretation. Some information can be retrieved from CNN through selected filters, but this information is actually sequence logos of common motifs; no information about underlying structures or structural motifs has been obtained so far from CNN filters.
Our approach is structure-centric, based on feature selection and thus having an advantage in interpretation. With representation of RNA structures through averaged properties of dinucleotides and loop and bulge composition, one can capture predominate properties of the analyzed structures -whether geometrical properties or else energy characteristics, or chemical, or loop-specific features are more influential. In our task geometrical properties appeared to be more important for evolutionary young sequences (such as full length copies of L1 and www.nature.com/scientificreports www.nature.com/scientificreports/ Alu), while when comparing the entire range of mRNA or pseudogenes neither structural nor energy features were selected; the selected features included only hydrophilicity and loop positions, reflecting the heterogeneity of objects.
The proposed approach is useful in analysis of RNA (DNA) secondary structures for recognizing different classes of non-coding RNA, and, in general, of functional stem-loops playing important roles as promoters, attenuators, splicing, intermolecular interactions (triplex structures), and others. The method allows detecting different structural constrains inherent to different classes of the analyzed structures.
# Conclusion
We constructed machine learning models that are able to recognize 3′-end L1 and Alu stem-loops, and 3′-ends of processed pseudogenes and mRNAs, based on different types of information. The models of the first type use only information on sequence composition, while models of the second type use physical, chemical and geometrical properties of RNA. We applied these models to a series of experiments on recognizing 50 bp 3′-ends and 3′-end stem-loops of L1s, Alus, processed pseudogenes and mRNAs both jointly and separately. In case of L1-Alu experiments, both types of models showed high performance with AUC > 0.96-98; for experiments with L1-processed pseudogenes and L1-mRNAs, the model performance is lower with AUC in the range of 0.83-0.99.
The feature importance analysis of the structure-based model revealed structurally important characteristics in stem-loop structures of different groups of elements. Thus, the parameters shift, rise and tilt appeared to be more important for recognizing the joint set of L1 and Alu stem-loops, while shift was shown to be more important for recognizing L1 3′-UTR stem-loops and rise was more important for distinguishing between L1 and Alu 3′-end stem-loops. Additionally, the parameters of stem positions adjacent to the loop appeared to be more important for Alu recognition. For recognizing 3′-end stem-loops of mRNAs and pseudogenes, the most important parameters were hydrophilicity and AT-rich loop positions.
The obtained results clearly point to the existence of structural constrains for 3′-end stem-loops of L1 and Alu, which presumably play an important role in recognition of L1-Alu pairs by the L1 machinery. They also highlight the potential of mRNAs to be recognized and retrotransposed by L1 machinery. The constructed machine-learning models are not restricted to recognition of transposon-related stem-loops, but have a broader range of applications in recognizing different classes of functional stem-loops and can be scaled to the task of predicting RNA (DNA) secondary structures with machine-learning methods.
# Methods
## Sets of l1 and alu transposons.
Full-length L1 transposons were taken from an evolutionary study of Khan et al. [bib_ref] Molecular evolution and tempo of amplification of human LINE-1 retrotransposons since the..., Khan [/bib_ref] and it comprises 6622 elements. The selection was done by searching the RepeatMasker annotation of human genome with L1 elements longer than 6 KB.
Full-length Alu transposons were taken from Price et al. [bib_ref] Whole-genome analysis of Alu repeat elements reveals complex evolutionary history, Price [/bib_ref]. AluJ is inactive family and was not taken into consideration [bib_ref] Active Alu retrotransposons in the human genome, Bennett [/bib_ref]. We considered a set of 12,431 AluS (724) and AluY (11,707) sequences (Supplementary in Price et al. [bib_ref] Whole-genome analysis of Alu repeat elements reveals complex evolutionary history, Price [/bib_ref]
## Shuffled sequences.
As an alternative class we used shuffled sequences that were obtained with the dinucleotide shuffling method that preserved dinucleotide frequencies. It prevents the models from relying on low-level statistics of genomic regions [bib_ref] Predicting the sequence specificities of DNA-and RNA-binding proteins by deep learning, Alipanahi [/bib_ref]. We took implementation from https://github.com/wassermanlab/BiasAway/ blob/master/altschulEriksonDinuclShuffle.py and updated it with a newer python version: https://github.com/ AlexShein/transposons/blob/master/src/py_scripts/altschulEriksonDinuclShuffle.py.
## Stem-loop annotations.
Transposon annotation of sequences with stem-loop structures was performed with the program "DNA Punctuation" (www.dnapunctuation.org) with an open source-code (https://github.com/ mariapoptsova/dnapunctuation. The following parameters were used: the stem range of 10-20 bp, the maximum loop length of 10 bp, and the number of mismatches allowed in the stem is 5 bp. For the 3′-end set of stem-loops, we took stem-loops located within the last 50 bp of a transposon sequence, either L1 or Alu. For the 5′-end set of stem-loops, we took stem-loops located within the first 50 bp of transposon sequences. All transposon annotations with stem-loops are available at OSF Data repository: Project " Recognition of 3′-end L1 and Alu stem-loops in human genome using sequence-based and structure-based machine-learning models" https://doi. org/10.17605/OSF.IO/PZKVN.
We also annotated shuffled sequences with stem-loop structures.
Feature vectors for sequence-based and structure-based models. For sequence-based model, we took frequencies of di-and trinucleotides counting occurrences of each k-mer moving with the 1 bp step along the sequence. www.nature.com/scientificreports www.nature.com/scientificreports/ For structure-based models, we considered a stem, a loop, and a bulge. For stem we took RNA dinucleotide properties from DiPRoDb 16 , which include structural parameters shift, slide, rise, tilt, roll, twist and physical and chemical properties such as enthalpy, entropy, free energy, and hydrophilicity.
We took first the 10 bp of a stem counting from the loop and split it into 9 dinucleotides [fig_ref] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences [/fig_ref]. Each dinucleotide was characterized by 10 parameters: 6 structural and 4 physical and chemical. We considered the minimum loop length of 5 bp and created position-specific nucleotide binary vector where each loop position was represented by 4 nucleotides [fig_ref] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences [/fig_ref]. Thus, the loop was represented by 20-dimenisional binary vector. We considered a bulge of a size of 3 bp on each stem thus making 6 bulge positions: 3 on the left stem and 3 on the right stem. Similar to a loop property vector, we characterized each position as a binary vector of 4 nucleotides thus making a 24-dimensional binary vector for bulge positions. If a bulge in a stem-loop structure was less than 3 nucleotides we filled missing positions with zeros. The resulting property vectors consisted of 90 + 20 + 24 = 134 parameters.
Feature vectors compositions are available at github: https://github.com/AlexShein/transposons/.
Machine-learning models. We built Random Forest models with 2000 trees using scikit-learn library. The model implementation and all data analysis is available at github: https://github.com/AlexShein/transposons/.
## Data availability
The model implementations and all data analysis are available at github: https://github.com/AlexShein/transposons/. Transposon sequences of L1s and Alus, processed pseudogenes and mRNAs, annotations with stem-loops, and all the data used in this study are available at OSF Data repository: https://doi.org/10.17605/OSF.IO/PZKVN.
[fig] Figure 1: Dotplots of 3′-end 50 bp L1 and Alu sequences. (a) Two sequences from L1-Ta family. (b) Two sequences from L1-Ta and AluY families. [/fig]
[fig] Figure 2: Recognition of 3′-end 50 bp of L1 and Alu elements. Six classification models were constructed: (1) L1 3′UTR vs shuffled, (2) L1 3′UTR vs L1 5′UTR, (3) L1 (L1 3′UTR + L1 5′UTR) vs shuffled, (4) Alu 3′-end vs shuffled, (5) Alu 3′-end vs L1 3′UTR, (6) Alu 3′-end and L1 3′UTR vs shuffled. (a,b) Models' performance: (a) ROC-curves and (b) precision-recall curves. (c) Feature importance analysis. Top-10 important parameters are coloured. [/fig]
[fig] Figure 3: Recognition of 3′-end stem-loops of L1 and Alu elements with sequence-based models. Six classification models were constructed: (1) L1 3′UTR vs shuffled, (2) L1 3′UTR vs L1 5′UTR, (3) L1 (L1 3′UTR + L1 5′UTR) vs shuffled, (4) Alu 3′-end vs shuffled, (5) Alu 3′-end vs L1 3′UTR, (6) Alu 3′-end and L1 3′UTR vs shuffled. (a,b) Models' performance: (a) ROC-curves and (b) precision-recall curves. (c) Feature importance analysis. Top-10 important parameters are coloured. [/fig]
[fig] Figure 4: Recognition of 3′-end stem-loops of Alu and L1 elements with structure-based models. (1) L1 3′UTR vs shuffled, (2) L1 3′UTR vs L1 5′UTR, (3) L1 (L1 3′UTR + L1 5′UTR) vs shuffled, (4) Alu 3′-end vs shuffled, (5) Alu 3′-end vs L1 3′UTR, (6) Alu 3′-end and L1 3′UTR vs shuffled. (a,b) Models' performance: (a) ROC-curves and (b) precision-recall curves. (c) Feature importance analysis. Top-10 important parameters are coloured. [/fig]
[fig] Figure 5: The Top-10 features for recognizing L1 and Alu. (a) The top-10 important features grouped by structural and physical parameters characterizing a stem, and parameters related to a loop and a bulge. Numbers in rectangles designate how many times these parameters were selected as important totally in different positions. (b) The top-10 important features grouped by positions in a stem (LS0-LS9), loop (LP0-LP4) and bulge (LB0-LB2, RB0-RB2). Numbers in rectangles designate how many times these positions were selected as important totally for different parameters. [/fig]
[fig] Figure 6: Percentage of processed pseudogenes, ribosomal proteins and mRNAs recognized by models trained to recognize L1 3′-ends or 3′-end stem-loops. [/fig]
[fig] Figure 7: Recognition of 3′-end 50 bp of processed pseudogenes and mRNAs. Six classification models were constructed: (1) processed pseudogenes vs shuffled, (2) mRNAs vs shuffled, (3) L1 3′UTR vs processed pseudogenes, (4) L1 3′UTR vs mRNAs, (5) L1 3′UTR and processed pseudogenes vs shuffled, (6) L1 3′UTR and mRNAs vs shuffled. (a,b) Models' performance: (a) ROC-curves and (b) precision-recall curves. (c) Feature importance analysis. Top-10 important parameters are coloured. (2019) 9:7211 | https://doi.org/10.1038/s41598-019-43403-3 [/fig]
[fig] Figure 8: Recognition of 3′-end stem-loops of of processed pseudogenes and mRNAs with sequence-based models. Six classification models were constructed: (1) processed pseudogenes vs shuffled, (2) mRNAs vs shuffled, (3) L1 3′UTR vs processed pseudogenes, (4) L1 3′UTR vs mRNAs, (5) L1 3′UTR and processed pseudogenes vs shuffled, (6) L1 3′UTR and mRNAs vs shuffled. (a,b) Models' performance: (a) ROC-curves and (b) precision-recall curves. (c) Feature importance analysis. Top-10 important parameters are coloured. [/fig]
[fig] Figure 9: Recognition of 3′-end stem-loops of processed pseudogenes and mRNAs with structure-based models. Six classification models were constructed: (1) processed pseudogenes vs shuffled, (2) mRNAs vs shuffled, (3) L1 3′UTR vs processed pseudogenes, (4) L1 3′UTR vs mRNAs, (5) L1 3′UTR and processed pseudogenes vs shuffled, (6) L1 3′UTR and mRNAs vs shuffled.(a,b) Models' performance: (a) ROC-curves and (b) precision-recall curves. (c) Feature importance analysis. Top-10 important parameters are coloured. [/fig]
|
Prevalence of Autism Spectrum Disorder and Co-morbidities in Children and Adolescents: A Systematic Literature Review
Objective: Individuals with autism spectrum disorder often present somatic and/or psychiatric co-morbid disorders. The DSM-5 allows for consideration of additional diagnoses besides ASD and may have impacted the prevalence of co-morbidities as well as being limited in capturing the true differences in prevalence observed between males and females. We describe the prevalence of ASD and frequently observed co-morbidities in children and adolescents (<18 years) in the United States and five European countries.Methods: Two systematic literature reviews were conducted in PubMed and Embase for the period 2014-2019 and focusing on the prevalence of ASD and nine co-morbidities of interest based on their frequency and/or severity: Attention Deficit Hyperactivity Disorder (ADHD), anxiety, depressive disorders, epilepsy, intellectual disability (ID), sleep disorders, sight/hearing impairment/loss, and gastro-intestinal syndromes (GI).Results: Thirteen studies on prevalence of ASD and 33 on prevalence of co-morbidities were included. Prevalence of ASD was 1.70 and 1.85% in US children aged 4 and 8 years respectively, while prevalence in Europe ranged between 0.38 and 1.55%. Additionally, current evidence is supportive of a global increase in ASD prevalence over the past years. Substantial heterogeneity in prevalence of co-morbidities was observed: ADHD (0.00-86.00%), anxiety (0.00-82.20%), depressive disorders (0.00-74.80%), epilepsy (2.80-77.50%), ID (0.00-91.70%), sleep disorders (2.08-72.50%), sight/hearing impairment/loss (0.00-14.90%/0.00-4.90%), and GI syndromes (0.00-67.80%). Studies were heterogeneous in terms of design and method to estimate prevalence. Gender appears to represent a risk factor for co-morbid ADHD (higher in males) and epilepsy/seizure (higher in females) while age is also associated with ADHD and anxiety (increasing until adolescence).Conclusion:Our results provide a descriptive review of the prevalence of ASD and its co-morbidities in children and adolescents. These insights can be valuable for clinicians and parents/guardians of autistic children. Prevalence of ASD has increased over time while co-morbidities bring additional heterogeneity to the clinical presentation, which Bougeard et al.Prevalence of ASD and Co-morbidities further advocates for personalized approaches to treatment and support. Having a clear understanding of the prevalence of ASD and its co-morbidities is important to raise awareness among stakeholders.
# Introduction
Autism Spectrum Disorder (1) is defined as a lifelong neurodevelopmental disorder characterized by two key symptoms: persistent deficits in social communication/interaction and restricted, repetitive patterns of behavior and abnormal sensory responses. The severity of these symptoms varies extensively from one patient to another, leading to a multitude of clinical presentations. Onset of ASD can usually be observed during childhood, with signs detectable as early as 18 months of age. However ASD remains extremely challenging to diagnose due to the diversity of clinical presentations and diagnosis requires both awareness from parents and caregivers to detect signs and an assessment from a multidisciplinary medical/paramedical team to confirm signs. As a consequence, estimating the prevalence of ASD is challenging, as illustrated by who recently conducted a worldwide narrative review on this topic (except for the African and Latin American regions) and confirmed the high variability in reported prevalence across regions. The authors reported high inter and intra variability across regions, with estimates ranging from 0.42 and 3.13% in Europe, 0.11 and 1.53% in Middle-East, 0.08 and 9.3% in Asia, and 0.87 and 1.85% in North America.
Furthermore, individuals with ASD often present co-morbid psychiatric disorders. In a recent review, Hossain et al.reported two studies estimating the prevalence of at least one comorbid psychiatric disorder at 54.8% and up to 94%, with Attention Deficit Hyperactivity Disorder (ADHD), anxiety, depressive disorders and sleep disorders being the most frequent co-morbidities. In addition, individuals with ASD are more likely to experience somatic co-morbidities such as epilepsy, gastrointestinal (GI) disorders or sight/hearing impairments. Both psychiatric and somatic comorbidities further complicate the diagnosis of ASD as they can either exacerbate or mitigate typical symptoms of autism. This can lead to misdiagnosis with inadequate management or delays in diagnosis with missed opportunity for treatment. Additionally, numerous studies have demonstrated the negative impacts of co-morbidities on autistic individuals as well as their surroundings, both in terms of quality of life and economic burden. The latest version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) now allows for consideration of additional diagnoses besides ASD and may have impacted the prevalence of comorbidities. Additionally, the new classification may also have contributed to the limitation of capturing the differences in prevalence observed between males and females for ASD and co-morbidities. Ratto et al.provided evidence of potential sex differences in autistic traits and adaptive skills; however, further investigations on the clinical presentations of ASD in males and females are needed to understand the factors explaining these differences in diagnosis rates. As such, it is of importance to have a clear understanding of the prevalence of ASD and its co-morbidities to raise awareness among stakeholders so both can be identified sooner and managed optimally. The objective of this study was to describe the prevalence and time trends of ASD, as well as frequently observed co-morbidities (psychiatric and somatic) in children and adolescents in the United States (US) and five European countries (France, Germany, Italy, Spain, and the UK). As a secondary objective, this study also looked at how age and gender were linked to the prevalence estimates of ASD and its co-morbidities.
# Materials and methods
A systematic literature review (SLR) was conducted to evaluate the prevalence of ASD in children and adolescents (from 2 to <18 years old) in EU-4 (France, Germany, Spain, and Italy) plus the UK, and the US. This manuscript presents studies reporting the latest prevalence data per country in order to understand the current epidemiological situation of ASD, as well as studies examining the prevalence of ASD overtime. Studies reporting on the prevalence of co-morbidities in ASD were retrieved from a second SLR on the clinical burden of ASD which considered the same population and geographic scope.
## Prevalence of asd
## Study selection
A SLR was conducted on July 24th 2019 in PubMed and Embase for the period 2014-2019, and completed with a gray literature search in December 2019 focused on governmental institutions and research associations such as ASDEU (Autism Spectrum Disorders in the European Union). The search string included the following search terms: "ASD"; "epidemiology" [and related terms]; as MeSH or Emtree terms; or in the title and abstract of articles. The full search strategy for this SLR is presented in Supplementary Material 1. The search focused on retrieving the latest prevalence estimate per country as well as prevalence over time. Study designs were limited to observational studies. Studies with prevalence as primary outcome were included, regardless of the method. Only articles published in English language were considered for review.
## Prevalence of co-morbidities in asd
## Search strategy
The second SLR was conducted on July 24th 2019 in PubMed and Embase for the period 2014-2019. This SLR included several search strings related to the clinical burden of ASD to inform several topics such as risk factors, mortality and prevalence of co-morbidities. The search string included the following search terms: "ASD" [and related terms]; "co-morbidity"; "mortality"; and "risk factor"; as MeSH or Emtree terms; or in the title and abstract of articles. The full search strategy for this SLR is presented in Supplementary Material 2. Only results retrieved from the search string on comorbidity are presented in this paper.
## Study selection criteria
This SLR focused on nine co-morbidities of interest: ADHD, anxiety, depressive disorders, epilepsy, Intellectual Disability (ID), sleep disorders, sight/hearing impairment/loss and GI problems. These co-morbidities were selected as they are frequently associated with ASD in our population of interest and correlated with substantial impairment. The selection of comorbidities was then submitted to clinical experts for approval. A broad definition was considered for each co-morbidity as high heterogeneity in the reporting of outcomes was expected. For example, information on comorbid GI syndromes and hearing impairment was mostly based on reported complaints rather than on medically established diagnoses. Studies reporting on prevalence data for these co-morbidities as primary or secondary outcome were included. Relevant studies from the first search (prevalence of ASD) reporting prevalence of co-morbidities were also included. Geographical scope and population of interest were the same as those stated earlier, and only observational studies were considered.summarizes the PICOS criteria established for this review.
## Study review process
Two researchers independently screened the titles and abstracts of identified studies and then full texts, to assess eligibility. Disagreements were resolved by discussion between the two reviewers and decision was made by a third reviewer if no agreement could be found. Information from included articles was extracted into a predefined data extraction template which included study characteristics, target population details and study outcomes.
Due to the variety of different study types included and the challenges in comparing study quality, studies were included without assessment of their methodological quality. As the study design, participant intervention and reported outcomes measured varied markedly, this research focused on describing the studies, their results, their applicability and their limitations and on qualitative synthesis.
# Results
## Prevalence of asd
We retrieved 13 publications reporting on the latest prevalence data for each country (8 studies) at the time the study was conducted and the evolution of ASD prevalence overtime (6 studies). The PRISMA diagram for this review is presented in Supplementary Material 3.
Prevalence point estimates were found for each country within our scope and ranged from 0.38% in Germany to 1.85% in the US.presents the latest estimates found for each country. All identified studies differed in terms of methodology with differences in age groups, geographical scopes and methods to estimate prevalence. Prevalence trends are summarized in. Time trends by gender were found in Bachman et al.and van Naarden et al.. No time trends were found for Italy or Spain.
In the US, two studies were identified which reported prevalence estimates of 1.70% (CI 95% : 1.61-1.80%) in children aged four year old (15), and 1.85% (CI 95% : 1.80-1.91%) in children aged eight year old. Additionally the eight year old cohort had a four times higher prevalence rate in males compared to females (2.97 vs. 0.69%). Both studies reported data from the Autism and Developmental Disabilities Monitoring network (ADDM), which uses a multiple-source, records-based surveillance method with 13 sites spread across the country and constitutes an active surveillance system. Case ascertainment involves a two-phase process where records from multiple data sources in the community are reviewed and then assessed by trained personnel to determine ASD case status. The ADDM network has reported ASD prevalence in multiple communities every two year since 2000, thus allowing to follow the evolution of the prevalence over time: Christensen et al.reported prevalence rates in children aged four year of 1.34% (CI 95% : 1.25-1.44%), 1.53% (CI 95% : 1.43-1.63%) and 1.70% (CI 95% : 1.61-1.80%), for the years 2010, 2012, and 2014 respectively. Prevalence over time in children aged eight year old was alsoyear and the California Department of Developmental Services collected data for all individuals living in California meeting the DSM diagnostic criteria for autism. The California Department of Developmental Services was the longest running of the three datasets, reporting an increase in ASD prevalence from 0.001% in the cohort born in 1931 to 1.2% among the 5 year-olds born in 2012. All three datasets displayed a consistent upward trend over time.
Prevalence estimates for France and Italy were found in Delobel-Ayoub et al.and Narzisi et al., respectively. Both studies were part of the ASDEU project involving 14 European countries and which aimed at estimating the prevalence of ASD in children aged between 7 and 9 year in 2015. The study by Delobel-Ayoub et al.involved two regional childhood disabilities registries covering the South-West and South-East regions of France. Both registries routinely included children aged eight year old, and cases of ASD were ascertained based on review of medical records from hospitals, psychiatric services and autism referral centers. The authors reported prevalence estimates of 0.48% (CI 95% : 0.40-0.56%) and 0.73% (CI 95% : 0.60-0.87%) for the South-East and South-West region respectively, however the study mentions that 5% of the parents of potential ASD cases refused to participate which slightly impacts the previous estimates. The South-West cohort also reported a higher male/female (M/F) ratio compared to South-East region (5.4 vs. 4.0). In Italy, Narzisi et al.reported prevalence for the metropolitan area of Pisa (Tuscany, Central Italy). The study considered children aged between 7 and 9 years old and followed a two-step process: (1) identification of certified children with a diagnosis of ASD and (2) identification of new cases. Case ascertainment in the first phase was based on the review of medical records by both the ASDEU team and the local child neuropsychiatry service team. The second phase of the study involved the participation of local school teachers who were asked to fill the Teacher Nomination Form (TNF) and distribute the Social Communication Questionnaire (SCQ) to the parents of children identified with the TNF. Children who had a positive TNF and SCQ (≥9) were administered the Autism Diagnostic-Revised (ADI-R) in association with the Autism Diagnostic Observation Schedule-2 (ADOS-2). Prevalence of ASD following the first phase of the study was 0.8% (N = 81/10, 138) (CI 95% : 0.62-0.97%) or 1 in 126 children. Prevalence for the second phase of the study was 0.2% (CI 95% : 0.06-0.33%) and 0.3% (CI 95% : 0.12-0.45%) when adjusting for non-response. Additionally, males had a five time higher prevalence of ASD compared to females. Prevalence over time for France was reported in van Bakel et al.whose study was based on the same two regional registers considered in Delobel-Ayoub et al.and presented an increase in prevalence, from 0.26% (CI 95% : 0.22-0.32%) in 1997 to 0.41% in 2003 (CI 95% : 0.36-0.48%).
A similar study was conducted by Morales-Hidalgo et al.in Spain. The study aimed at estimating the prevalence of ASD, ADHD, and social communication disorder in children aged 4-5 years and 10-11 years in Tarragona (Catalonia, Spain). The research was part of the neurodevelopmental Disorders Epidemiological Research Project and consisted in a screening phase where parents and teachers were administered the Childhood Autism Spectrum Test and the EDUTEA questionnaire respectively. In a second phase, screening was confirmed by interviewing parents of positively screened children using the ADI-R (Autism Diagnostic Interview-Revised) and administering children the ADOS-2 (Autism Diagnostic Observation Schedule) and Wechsler scales. Adjusted prevalence reported by the authors was 1.55% (CI 95% : 0.89-2.20%) in children aged 4-5 years and 1.00% (CI 95% : 0.48-1.51%) in children aged 10-11 years, while direct prevalence was 1.06% (CI 95% : 0.66-1.46%) and 0.78% (CI 95% : 0.44-1.12%) respectively. Additionally, the authors mentioned that between 1.84 and 2.59% of the children exhibited subclinical diagnoses of ASD. Finally, in both cohorts the prevalence was four times higher in males compared to females.
Prevalence rates for the UK were obtained from the Survey of the mental health of children and young people 2017 available on National Health Service digital which aimed at assessing the prevalence of mental health conditions (including ASD) in children and young people aged 2 to 19 years. The method of the survey involved the administration of the Development and Well-Being Assessment questionnaire. Case definition for ASD was based on the ICD-10 classification (International Classification of Diseases, Tenth Revision, Clinical Modification) and encompassed the following codes: F84.0 (childhood autism),
## Prevalence of co-morbidities in asd
A total of 6,094 studies were identified. After reviewing titles and abstracts, 827 full text articles were reviewed for relevance. The 340 articles related to mortality and risk factors were not included in this review. 30 studies met the inclusion criteria and 3 studies from the search on prevalence of ASD were also added as they reported prevalence on co-morbidities of interest. Ultimately, 33 studies were included in this review. A PRISMA diagram for this search is displayed in. Studies that did not provide point prevalence estimates for co-morbidities were excluded.
Of the 33 studies included, 26 (79%) were conducted in the US while seven were conducted in a European country (France [n = 2], Italy [n = 2], UK [n = 3]). Additionally, prevalence of co-morbidities was reported as a primary outcome in 19 (58%) studies and as a secondary outcome in the other 14 publications. Studies were heterogeneous in terms of study design, with a mix of cohort studies (n = 23), cross sectional studies (n = 4) and case-control studies (n = 6). Studies were also heterogeneous regarding the method considered to estimate the prevalence of co-morbidities: tools and scales (± combined with clinical assessment) were used in 18 studies while eight considered database or registries and seven relied on clinical expertise or caregivers report only. All studies considered cohorts with more than 30 autistic individuals, the lowest sample size being retrieved in Mosner et al.with 35 adolescents. In addition, case ascertainment was based on self-reporting in four studiesand sample characteristics were missing in Polyak et al..
The number of studies per co-morbidity was as follows: ADHD (n = 17), anxiety (n = 13), depressive disorders (n = 12), epilepsy/seizure (n = 12), GI syndromes (n = 7), hearing impairment (n = 3), ID (n = 13), sleep disorder (n = 7), and vision impairment (n = 3). Study characteristics are presented in Supplementary Material 4, and a summary of results stratified by age (0-5y; 6-11y and 12-17y) can be found in Supplementary Material 5.
## Adhd
Prevalence for ADHD in ASD children ranged from 0.00% in children aged 3-4y (35) to 86.00%andwith a prevalence of 11.00, 21.00, and 40.00% respectively for the inattentive subtype, 4.00, 1.00, and 23.10% for the impulsive subtype and 62.00, 64.00, and 18.50% for both subtypes combined. In the US, Houghton et al.studied the prevalence of comorbidities in ASD using two insurance claims databases, one commercial database (Truven Health MarketScan R Commercial Database) and the Medicaid database. The authors reported the prevalence of ADHD for three age groups: 6.89/14.19% in 3-4y, 40.46/47.67% in 5-11y, and 47.73/51.08% in 12-17y (commercial/Medicaid). The authors conducted a similar study in the UK and leveraged the Clinical Practice Research Datalink database to estimate the prevalence of multiple co-morbidities while considering the same age groups. Prevalence for ADHD was 0.0% in children aged 3-4y, 10.40% in 5-11y, and 17.7% in the 12-17y group. In a UK study, Salazar et al.reported the prevalence of ADHD to be 59.10% for the total population and by sex 64.40% (CI 95% : 50.00-78.70%) in male children aged 4-8 y and 38.60% (CI 95% : 23.90-53.40%) in female children. The authors also calculated a statistically significant odds ratio (OR) of 2.9 (CI 95% : 1.2-6.9), confirming the disorder was more prevalent in male individuals. In a similar manner, Stacy et al.also reported the prevalence of ADHD by sex, but also further stratified the disorder between "mild" and "moderate/severe" ADHD. Prevalence for mild ADHD in male/female children wasassessed the influence of sex and age on the prevalence rates of comorbidities in ASD leveraging the STRIDE database and reported rates of co-morbid ADHD of 46.39 and 31.25% in males and females aged 0 to 18 years old respectively. Overall, ADHD appears more prevalent in male adolescents as sex was identified as a risk factor in Salazar et al., and also confirmed in Brookman-Frazee et al.who found that boys were 3.37 more likely to score positive for any ADHD disorder on the MINI-KID-P, while Soke et al.found ADHD to be more prevalent in children aged 8y vs. 4y, with an adjusted OR = 4.78 (CI 95% : 3.40-6.73)., although the subtypes considered were different in all three studies. However, all three authors reported prevalence for social phobia and separation anxiety disorder of 24.00, 6.00, 9.20, and 15.00, 1.00, 3.10% respectively. In the US, Houghton et al.reported prevalence based on two cohorts (commercial/Medicaid) of 3.65/2.59% in 3-4y, 18.87/11.50% in 5-11y and 30.49/17.73% in 12-17y. The same study conducted in the UK showed prevalence of 0.00% in children aged 3-4y, 1.70% in 5-11y and 4.8% in the 12-17y group. In Salazar et al., prevalence of any anxiety disorder was 65.90% in female vs. 82.20% in male, and the statistical analysis did not identify sex as a risk factor for this disorder [OR = 2.40 (CI 95% : 0.90-6.00)]. However, an Intellectual Quotient (IQ) above 70 was identified as a risk factor for anxiety in ASD with a statistically significant OR of 2.90 (CI 95% : 1.00-8.10). Salazar et al.also provides prevalence for agoraphobia, specific phobia, panic disorder, generalized anxiety disorder and separation anxiety disorder all stratified by sex, IQ, age at assessment and ASD severity. Similar to ADHD, Stacy et al.reported prevalence rates of comorbid mild and moderate/severe anxiety of 11.70 and 18.60% in females respectively and 11.10 and 24.90% in males respectively (no significant differences). Anxiety appears to be more prevalent in older children and age was identified as a risk factor in Soke et al.[adjusted OR = 2.28 (CI 95% : 1.57-3.39)]. In their paper, Salazar et al.provide a rationale for the higher prevalence of anxiety disorders in older children with higher IQ and hypothesized that both factors expose children to more anxiety-provoking situations and engage in higher-cognitions.
## Depressive disorders
Prevalence for depressive disorders varied from 0.00% in children
## Epilepsy/seizures
Prevalence for epilepsy/seizure in ASD children ranged from 2.80% in children aged 3-11y (35) to 77.50% in a subgroup of ASD individuals characterized by rate of ID in Doshi-Velez (44). In the US, Houghton et al.reported estimates of 3.82/5.41% in 3-4y, 5.21/7.40% in 5-11y and 7.04/10.38% in 12-17y (commercial/Medicaid). In the UK prevalence was 2.80% in children aged 3-11y and 4.10% in 12-17y (35). Soke et al.reported rates of 2.81% and 3.02% in children aged 4y and 8y respectively, but did not identify age as a risk factor (adjusted OR = 0.99 (CI 95% : 0.54-1.84). Superkar et al.reported rates of 40.36% and 43.75% in male and female with ASD aged 0-18y respectively and observed statistically higher rates in females with ASD, however this observation only applied to the whole study population aged 0 to 35y and more. Stacy et al.reported prevalence rates of comorbid mild and moderate/severe epilepsy of 5.10 and 4.60% in females respectively and 7.40 and 3.20% in males respectively (without significant difference). The publication of Ewen et al.specifically studied epilepsy in children with ASD and considered two cohorts from an online research registry (interactive autism network). In one cohort, the child with ASD questionnaire (CAQ) was used to estimate prevalence while the birth and ASD diagnosis history questionnaire (BDQ) was used in the second cohort. The authors reported prevalence of 9.10% in the CAQ cohort and 10.90% in the BDQ cohort and also demonstrated that ID, language deficit, ASD severity, and motor performance abnormalities were risk factors for epilepsy after adjusting for age and sex, with ID being the most impactful In the CAQ cohort, the authors also established that female sex was associated with 40% increased risk for epilepsy compared to males.
## Gastro-intestinal syndromes
Prevalence of GI varied from 0.0% for inflammatory bowel disorders in female with ASD aged 0-18y (41) to 67.80% of children in and Supplementary Material 6). Superkar et al.also reported comorbid rates for bowel disorders of 25.90% and 18.75% in males and females respectively. In Italy, Fulceri et al.estimated the prevalence of seven GI symptoms ("constipated", "diarrhea", "not eat", "nausea", "painful bowel movements", "stomach-aches", "vomiting") in pre-schoolers with ASD compared to a group of typically developed children. Assessment of symptoms was done using the Child Behavior Checklist 1 1/2 -5 (Italian version), and authors reported an overall prevalence of 37.40% (29.6% in males, 7.8% in females) vs. 14.80% in typically developed children (p = 0.0001). "Constipated" and "not eat" were most prevalent subtypes observed in the ASD group with 15.70 and 27.00% respectively, and both rates were significantly higher compared to the typically developed group. Similarly, the research by specifically aimed at assessing the frequency and characteristics of symptoms of GI disorders in 186 children with a median age of 7y (range 2-18y) estimated a prevalence rate of 49.00% for GI symptoms and also for the following subtypes: diarrhea (22%; n = 37), constipation (26%; n = 43), bloating (13%; n = 20) and gastro-enteric reflux/vomiting (10%; n = 16). The authors also demonstrated that the presence of any GI disorder was associated with significantly higher rates of sleep disorders (p = 0.001) and food intolerance (p < 0.001).
## Hearing impairment
Prevalence of hearing impairment ranged from ∼0% (females) in Polyak et al.a study investigating sex bias in ASD comorbidities to 4.90% in Rydzewska et al.. The latter was a Scottish study investigating comorbid conditions in ASD vs. general population (0-24 years). Deafness or partial hearing loss was observed in 2.90% (2.40/4.90% in males/females) vs. 0.50% in the general population for the younger age group (0-15 years), and 3.90% (3.20/6.30% in males/females) vs. 0.80% for the older age group . The study concluded that hearing impairment was nine times more prevalent in the autistic individuals compared to the general population. Finally, Stacy et al.reported prevalence rates of comorbid mild and moderate/severe hearing problem of 2.50 and 4.80% in females respectively and 2.00 and 4.20% in males respectively (without significant difference).
## Intellectual disability
ID was defined by an IQ score below 70 in all studies. Prevalence of ID in ASD children ranged from 0.00% in children aged 3-4y (35) to 91.70% in female ASD aged 0-18yand Supplementary Material 6). Prevalence of ID in Mazurekwas estimated based on telephone surveys aimed at parents of children with ASD and asked the following question: 'have "a doctor or other health care provider ever told [him/her] that [the child] had" other specific co-occurring conditions?'. The study population was stratified by the number of co-morbidities, and the prevalence of 91.7% corresponded to the subpopulation of children who had ASD associated with three co-morbidities. In the US, Christensen et al.and assessed the prevalence of ID in ASD children aged 4y and 8y respectively based on the ADDM network and reported rates of 46.10% and 33.00%. In Scotland, Rydzewska et al.prevalence rates of 11.30% in male and 22.80% in female children aged 0-15y based on data from the Scotland's Census 2011. Additionally, the author reported that ASD was a strong predictor for ID, with an OR of 15.70 (CI 95% : 13.40-18.50). Finally, aimed at assessing the impact of ID on the presence of comorbid symptoms (tantrum behavior, repetitive behavior, worry/depressed, avoidant behavior, undereating, conduct disorder, over-eating) in children with ASD and did not establish any significant impact.
## Sleep disorders
Prevalence for sleep disorders varied from 2.08% in ASD females aged 0-18y (41) to 72.50% in ASD children with a mean age of 9yand Supplementary Material 6) (51). In the US, Houghton et al.reported prevalence of 6.39/12.79% in 3-4y, 5.87/10.22% in 5-11y and 5.14/8.85% in 12-17y (commercial/Medicaid). In the UK, Houghton et al.estimated prevalence at 6.40% in children aged 3-4y, 13.20% in children aged 5-11y and 14.40% in 12-17y. Similar to previously mentioned co-morbidities, age was identified as a risk factor (adjusted OR = 1.34 [CI 95% : 1.15-1.56]) for sleep abnormalities in Soke et al.which estimated prevalence of 26.95% in children aged 4y and 37.12% in children aged 8y. In a retrospective case-cohort study based on the military health system database, Elrod et al.analyzed a cohort of 48,762 children fulfilling the study case definition for ASD, and 31.00% displayed an ICD-9 code for one sleep disorder (vs. 14.00% in the control group). The most prevalent subtypes of sleep disorders were: sleep disorder-not otherwise specified (21.72%; n = 10,593), insomnia (9.52%) and sleep disorder breathing (9.26%). The authors calculated a RR of receiving a sleep disorder diagnosis of 1.97 in children with ASD (CI 95% :
## Vision impairment
Prevalence for vision impairment ranged from ∼0% (males and females) in Polyak et al.study examining the impact of sex bias on comorbidities in ASD, to 14.9% in Aldinger et al.. Aldinger et al. (51) compared a family based cohort that predominantly consists of families with more than one child with ASD (multiplex) vs. a family with one ASD child (simplex). Vision problems were observed in the multiplex cohort (14.9 and 15.3% complete sample). Rydzewska et al.reported estimates of 3.50% (n = 599) in children aged between 0 to 15 years old (males = 2.6%; females = 6.7%) for the outcome "blindness/partial sight loss", and 3.80% in the 16 to 24 year old group (males = 3.0%; females = 6.4%).
# Discussion
The objective of this study was to describe the clinical burden associated with ASD in children and adolescents based on recent prevalence data and trends over time, as well as understanding the prevalence of nine co-morbidities associated with ASD and potential differences by age and gender. Our research focused on children and adolescents with ASD (age < 18y) and considered studies conducted in the US, France, Germany, Italy, Spain, and UK.
Our first review found the latest prevalence data for each country, with 1.70 and 1.85% in the US in children aged 4y and 8y respectively, while prevalence of ASD in Europe ranged from 0.38% in Germany (0-24 years) to 1.55% in Spain. Of note, Fuentes et al.recently published the results of the ASDEU initiative for Spain and reported a prevalence of 0.59% in children aged 7-9y, which is substantially lower than the one reported by Morales-Hidalgo et al.. Although the overall screening and diagnosis processes are similar in both studies, the difference can be explained by the difference in sample sizes vs. 2,827 in Morales-Hidalgo et al.]. Prevalence data for the US is based on the ADDM network, an active and national surveillance system that reported the prevalence of ASD since 2000. However, the methodologies used to estimate the prevalence of ASD in the five European countries were all different, whether in terms of age group considered or method to estimate prevalence. The absence of a global European surveillance network implies that each country individually assesses the prevalence in its own population. As such estimates between European countries are highly heterogeneous, which makes interpretation and comparison of these differences challenging. These differences in prevalence can also be extended to country healthcare systems as most countries in Europe have their own way of diagnosing, apprehending and managing ASD. Our findings are in accordance with those of Chiarotti and Venerosi (5) who conducted a narrative review on the worldwide prevalence of ASD, and identified the same studies. Similarly, Autism Europe estimates between 1.0 and 1.5% the prevalence of ASD in Europe, which is in line with the prevalence of 1.22% reported by ASDEU. The higher prevalence reported in ASDEU compared to the five above mentioned European countries can be explained by the inclusion of countries with high prevalence such as Iceland who reported a prevalence of 2.68%, possibly due to the limited population of the country. Male to female ratios in our review varied between 2.7 in Germany and 6.5 in the UK and is consistent with previous studies that are supportive of a 4:1 ratio. Studies have shown that prevalence is potentially underestimated in females, as autistic females could better escape diagnosis or ASD may only become evident in adolescence when social demands increase. Overall, current evidence is supportive of a global increase in ASD over the past years, while suggesting that the augmentation is not specific to a particular region or healthcare system, but is observed at a broader level. Possible hypothesis behind this progression have been well formulated in who suggested that the increase in prevalence is a result of better professional and public awareness of ASD rather than a true augmentation in the number of cases. Another explanation could be the focus of clinical practice on subtle expression of symptoms in highfunctioning autistic children over the past years. Finally, internal and external causes such as parental age or exposure to drugs like thalidomide or valproic acid during pregnancy could also be associated to this augmentation. However, the current evidence around the role of some of these factors is still unclear and represents active areas of research.
It is possible that prevalence estimates in ASD are underestimated for multiple reasons, the first one lying in complexity of the disorder itself. ASD is a complex disorder associated with a wide array of phenotypes and numerous levels of severity. International guidelines on ASD suggest that each case should be considered individually and diagnosis made by a multidisciplinary team. However, availability of multidisciplinary teams and expert clinicians is rare, leading to extended waiting list and children escaping diagnosis. Underestimation of the prevalence of ASD could also be attributed to the multiplicity of diagnostic tools available. Such tools are commonly used to support diagnosis and complete clinical assessment, but no gold standard officially exists. Although, the combination of the ADOS and ADI-R is commonly reported, both tools require time and skills to administer which can lead to inaccurate evaluations. On the other hand, it is also possible for prevalence to be overestimated in case of misdiagnosis with other disorders and in situations where ASD is associated with services and care that would not be otherwise granted for other disorders. Evidence as to whether the switch from the DSM-IV to the DSM-5 algorithm of ASD has impacted on its prevalence is conflicting.
As for the prevalence of co-morbidities in ASD, our SLR included a total of 33 studies and highlighted wide prevalence ranges for each co-morbidity: 0.0-86% for ADHD (17/33 studies), 0.0-82.2% for anxiety. However both studies considered pediatric and adult populations and different definitions for co-morbidities.
The literature suggests that the prevalence of some comorbidities can be different with age and gender. Prevalence of ADHD and anxiety appear to increase until adolescence, with a decline observed when reaching adulthood. In addition, evidence is supportive of a higher prevalence of ADHD in male compared to female individuals. Epilepsy and depressive disorders were predominant in female individuals and slightly increasing with age. Sleep disorders were more frequently observed in children than adolescents. Evidence related to visual and hearing impairment is scarce and insufficient to formulate hypothesis about the impact of age. ADHD was found to be associated with a greater risk of having other psychiatric comorbidities in addition to lower cognitive functioning, more severe social impairment and greater delays in adaptive functioning. Comorbid epilepsy often occurs in conjunction with ID and was associated with worse behavioral and social outcomes, increased motor difficulties and more challenging behavior compared with ASD alone. Our review also gathered evidence that co-morbidities are more frequently observed in ASD individuals compared to typically developed population, which is in accordance with the current literature (7).
Similar to our findings for the prevalence of ASD, several factors should be considered when interpreting the heterogeneity in co-morbidity prevalence. First, the definition for a given comorbidity differed across studies. This was observed particularly for anxiety disorders where claim database studies such as Houghton et al.considered a wide array of disorders from anxiety states to dissociative disorders, whereas the prevalence reported in Stacy et al.corresponds to "anxiety" without further information. As such, the presence of subtypes within a single co-morbidity is a factor for heterogeneity. Second, study design and methodology should be considered when interpreting our results as our review included a mix of cohort, cross-sectional and case-control studies. Although informative, prevalence rates from surveys should be interpreted carefully as they are subject to multiple bias from parents who often must confirm or infer the presence of a co-morbidity based on a single question. Similarly, claims database analyses do not provide information regarding the diagnostic process undertaken and instrument(s) used to confirm the presence of the comorbidity. Finally, diagnostic ambiguity should be considered when interpreting prevalence of co-morbidities as it is often challenging to distinguish a co-morbidity from the presentation of autism. As an example, deficit in communication/interaction in autism can be interpreted as an anxiety disorder while conversely it can prevent from reporting symptoms potentially attributable to a co-morbidity.
Our study presents several limitations. First, a broad search term for comorbidity was considered and not terms for specific comorbidities. This approach allowed us to identify the most frequently observed comorbidities in ASD whose relevance was further confirmed by clinical experts. The second limitation lies in the decision to include studies reporting prevalence as a secondary outcomes. While these studies did provide relevant information, the rates extracted from these studies might include a selection bias by selecting specific subpopulations of children with ASD. Third, although the purpose of this study was to provide a descriptive review of the prevalence of ASD and its comorbidities, additional research could involve analysis focusing on a single co-morbidity to estimate an overall prevalence rate while accounting for the heterogeneity across studies highlighted in our research. Fourth, our review focused on the most common co-morbidities reported in the literature but other co-morbidities can be observed in association with ASD such as dementia or allergic/autoimmune diseases. Finally, as our search strategy focused on children and adolescents in the US and five European countries, this study did not capture the evidence generated in other geographic regions as well as in adults with ASD.
Our research provides a descriptive review of the prevalence of ASD and its co-morbidities which can be valuable for clinicians as well as parents/guardians of children with ASD. Despite substantial heterogeneity in estimates, our results indicate that co-morbidities are highly prevalent in ASD, and should be kept under consideration when diagnosing and managing individuals with ASD. In addition, co-morbidities should be re-assessed regularly as evidence suggest that prevalence evolves over time, especially during transition ages such as entry to school or adolescence. Also, co-morbidities bring additional heterogeneity to the presentation of ASD, further advocating for personalized and tailored approaches to treatment and support. Finally, co-morbidities in ASD negatively impact the quality of life and life expectancy of individuals as well as representing a substantial economic burden. Results from Hirvikoski et al.reported a 2.56-fold increased odds of mortality in individuals with ASD compared with matched general population while Hedgecock et al.reported a negative correlation between quality of life and behavioral disorders in autistic children. Buescher et al.estimated the mean annual cost of ASD to be $63,292 and $52,205 in children aged 0-5y and 6-17y respectively. However costs increased to $107,863 and $85,690 in children with ASD and co-morbid ID. Similarly, Houghton et al.showed that psychiatric co-morbidities in ASD were associated with increased medication and increased GP interactions, thus leading to higher healthcare resource consumption and potentially higher rates of adverse events from medication.
Our findings have clear implications for the diagnostic process of young people referred for suspicion of ASD and for service organization. The relative high probabilities of psychiatric and somatic co-occurring conditions should be incorporated into clinical guidelines for assessing ASD. The assessment should not only be focused on identifying the core symptoms of ASD but also on the wider range of other psychiatric and somatic problems. To this end, systematic information from parents and teachers should be collected through validated broad band questionnaires that tap internalizing and externalizing behavior problems, such as the ASEBA Questionnaires (62), the Conners Questionnaires (63), or the Strengths and Difficulties Questionnaire (SDQ). The diagnostic work-up should also include an interview about somatic signs and symptoms, and a standard physical exam. Preferably, clinical care should be organized around the patient and provided by a multidisciplinary team of health care professionals with expertise in complementary areas.
These findings are supportive of a global increase in ASD prevalence independent from regions and healthcare systems and call for stronger awareness within populations and healthcare policies. Our review also provides prevalence estimates for nine co-morbidities frequently associated with ASD and highlighted the importance of age and gender in prevalence of both ASD and its co-morbidities. Across Europe, there is still a need for studies applying a similar methodology when estimating prevalence in order to allow comparison across countries. This review highlights the substantial clinical burden associated with ASD, with co-morbidities further complicating evaluation, management and prognosis of ASD. Finally reliable estimates for prevalence of ASD and associated co-morbidities would support economic analysis and further assessment of burden in ASD.
# Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s. |
In vivo dosimetry for total body irradiation: five‐year results and technique comparison
The aim of this work is to establish if the new CT-based total body irradiation (TBI) planning techniques used at University College London Hospital (UCLH) and Royal Free Hospital (RFH) are comparable to the previous technique at the Middlesex Hospital (MXH) by analyzing predicted and measured diode results. TBI aims to deliver a homogeneous dose to the entire body, typically using extended SSD fields with beam modulation to limit doses to organs at risk. In vivo dosimetry is used to verify the accuracy of delivered doses. In 2005, when the Middlesex Hospital was decommissioned and merged with UCLH, both UCLH and the RFH introduced updated CT-planned TBI techniques, based on the old MXH technique. More CT slices and in vivo measurement points were used by both; UCLH introduced a beam modulation technique using MLC segments, while RFH updated to a combination of lead compensators and bolus. Semiconductor diodes were used to measure entrance and exit doses in several anatomical locations along the entire body. Diode results from both centers for over five years of treatments were analyzed and compared to the previous MXH technique for accuracy and precision of delivered doses. The most stable location was the field center with standard deviations of 4.1% (MXH), 3.7% (UCLH), and 1.7% (RFH). The least stable position was the ankles. Mean variation with fraction number was within 1.5% for all three techniques. In vivo dosimetry can be used to verify complex modulated CT-planned TBI, and demonstrate improvements and limitations in techniques. The results show that the new UCLH technique is no worse than the previous MXH one and comparable to the current RFH technique.
In order to deliver a homogeneous dose across the body and to limit doses to organs at risk, the beams are modulated using bolus, compensators or shielding. In vivo dosimetry with thermoluminescent dosimeters (TLDs) or semiconductor diodes allows entry and exit doses to be recorded during treatment and used to calculate delivered midline doses. [bib_ref] Toxicity and dosimetry of fractionated total body irradiation prior to allogeneic bone..., Gerrard [/bib_ref] [bib_ref] Total body irradiation using a modified standing technique: a single institution 7..., Harden [/bib_ref] [bib_ref] An approach to dose measurement for total body irradiation, Greig [/bib_ref] [bib_ref] Midplane dose determination during total body irradiation using in vivo dosimetry, Ribas [/bib_ref] [bib_ref] In-vivo dosimetry by diode semiconductors in combination with portal films during TBI:..., Mangili [/bib_ref] These verify the delivery of the prescribed dose and indirectly check the patient position. Diodes have the advantage of real-time intrafraction feedback and possibility for immediate correction. [bib_ref] Verification of an on line semiconductor dosimetry system for TBI with two..., Sánchez-Doblado [/bib_ref] Calibration of diodes should be performed in TBI treatment conditions, due to the dependence of response on beam spectrum and dose rate. [bib_ref] Calibration of semiconductor detectors for dose assessment in total body irradiation, Jornet [/bib_ref] More recently, treatment planning using computed tomography (CT) scans of the patient have been introduced, using either in-house [bib_ref] A CT-aided PC-based physical treatment planning of TBI: a method for dose..., Sánchez-Nieto [/bib_ref] or commercial software. [bib_ref] Physical aspects of total-body irradiation at the Middlesex Hospital (UCL group of..., Planskoy [/bib_ref] [bib_ref] Physical aspects of total-body irradiation at the Middlesex Hospital (UCL group of..., Planskoy [/bib_ref] [bib_ref] TBI treatment planning using the ADAC Pinnacle treatment planning system, Abraham [/bib_ref] [bib_ref] CT-based analysis of dose homogeneity in total body irradiation using lateral beam, Hui [/bib_ref] [bib_ref] Commissioning and evaluation of an extended SSD photon model for PINNACLE3: an..., Lavallée [/bib_ref] CT data allow accurate estimation of organ-at-risk doses such as the lung, which usually limits the maximum tolerable dose. Placement of bolus and compensation can also be more accurately predicted before the first fraction of a multifraction treatment, reducing the need for test doses or modification later in the course. [bib_ref] In-vivo dosimetry by diode semiconductors in combination with portal films during TBI:..., Mangili [/bib_ref] However, in vivo dosimetry is still essential for the verification of delivered dose accuracy and consistency in comparison to planned values. The accuracy of in vivo measurements should be ± 5%, [bib_ref] An approach to dose measurement for total body irradiation, Greig [/bib_ref] in order to assess the overall aim of total delivered dose being within ± 10% of that prescribed.
CT-planned TBI was introduced clinically at the Middlesex Hospital, London (MXH) in 1989, [bib_ref] Physical aspects of total-body irradiation at the Middlesex Hospital (UCL group of..., Planskoy [/bib_ref] [bib_ref] Physical aspects of total-body irradiation at the Middlesex Hospital (UCL group of..., Planskoy [/bib_ref] with the patient lying supine with their knees bent. Briefly, depth doses and profiles measured under TBI conditions were configured into a commercial planning system, including modifications to allow for large field sizes and SSDs. Compensators were optimized to deliver uniform dose for bilateral fields within nine predetermined CT slices (including shoulders, lungs, and umbilicus). The compensators were then manually interpolated along the whole body by use of wax bolus. In addition, PMMA slabs were used for the head and bolus for the legs, to build up to the effective separation. The plan was normalized by assigning 100% to the highest lung dose. Diodes were used to verify the delivered dose, at five points along the body. [bib_ref] Physical aspects of total-body irradiation at the Middlesex Hospital (UCL group of..., Planskoy [/bib_ref] [bib_ref] Physical aspects of total-body irradiation at the Middlesex Hospital (UCL group of..., Planskoy [/bib_ref] In 2005, when the Middlesex Hospital was decommissioned and merged with UCLH, both UCLH and the RFH introduced updated CT-planned TBI techniques, based on the old MXH technique. More CT slices and in vivo measurement points were used by both; UCLH introduced a beam modulation technique using MLC segments, while RFH updated to a combination of lead compensators and bolus [fig_ref] Figure 1: Beam's eye view [/fig_ref].
The aim of this study was to review the in vivo diode measurements from both centers over five years of treatments, and compare them to the previous MXH technique in terms of the accuracy and consistency of delivered dose. Accuracy, reproducibility between fractions, and variations at different anatomical locations were assessed in terms of percentage dose differences between calculated and measured doses and their standard deviations.
## Ii. materials and methods
## A. treatment planning
## A.1 uclh
In the 'new' UCLH technique, a full CT dataset is acquired, extending from head to thighs, with a slice thickness of 10 mm and no additional bolus, and entered into the standard treatment planning system (Oncentra Master Plan, OMP, Nucletron, Veenendaal, The Netherlands). A TBI-specific 10 MV machine was commissioned in OMP, using a pencil beam algorithm, to reproduce the measured dose distribution at the TBI treatment distance, including the effect of the beam spoiler (see below) on the buildup region. The treatment planning consists of adding MLC segments to shield or boost specific regions in the thorax, abdomen, and pelvis in order to improve the dose homogeneity, ensure the marrow producing bones are covered by 90%-110% of the prescription dose level, and ensure dose limits to lungs and kidneys are met (e.g., 50% and 10% of respective volumes to receive less than 100% of prescription dose) [fig_ref] Figure 1: Beam's eye view [/fig_ref]. The head and neck are compensated with PMMA slabs and bolus, and legs with bolus alone, as in the MXH technique. The plan is normalized to give 100% of the prescription dose to the midpoint at the level of the umbilicus. The accuracy of the TPS calculations was validated through planning a whole body anthropomorphic phantom (RANDO, The Phantom Laboratory, Salem, NY) and comparing predicted doses to TLD readings within the phantom. Expected diode doses are calculated from the final treatment plan at a depth of 1.5 cm from the patient's external contour (corresponding to the inherent buildup of the detectors) at several relevant anatomical positions.
## A.2 rfh
In the RFH technique, a CT dataset is used extending from head to knees with a slice thickness of 10-20 mm. A PMMA headbox, and bolus around the neck, hands, and between the legs are included in the scan. These data are used in Pinnacle treatment planning system (Philips, Amsterdam, The Netherlands) to produce a homogeneous treatment plan by the addition of lead compensators and further bolus, as required. The standard clinical 10 MV treatment beam model (using collapsed cone convolution) is used for calculations; the spoiler screens are simulated using dummy contours within the images to mimic the treatment setup. Typically, several different thicknesses of rectangular lead compensators are used to modulate the incident radiation intensity along the cranial-caudal axis (CCA) of the patient, with the compensator edges aligned in the axial direction [fig_ref] Figure 1: Beam's eye view [/fig_ref]. Extra PMMA may be added to the headbox, and extra bolus to the kidneys, under the arms (to cover the lungs), and to extend the headbox over the shoulders. A homogeneity of ± 5% is aimed for at midline, but ± 10% is acceptable, as long as dose volume constraints to the lungs and kidneys are achieved (e.g., 50% and 10% of respective volumes to receive less than 100% of prescription dose). Expected diode doses are calculated from the final plan at a depth of 2 cm from the patient's external contour, corresponding to the diodes' inherent buildup.
## B. treatment delivery
## B.1 uclh
The patient is set up supine on the free-standing TBI couch and the CCA length is measured in order to obtain the correct knee inflection to ensure feet are well within the field. The patient couch is rotated by 180° half way through the treatment. For each side, a sagittal laser 3.55 m from isocenter (4.55 m from the source) is used to align the patient midline perpendicular to the beam axis. Patient positioning is completed using the lateral tattoos at field center and with measurement of external anatomical landmarks. In addition, a zero dose MLC setup light field is used to check the arm positions and the couch height. A 15 mm thick PMMA beam spoiler screen is then placed in front of the patient at a distance of 0.25 m from the patient skin [fig_ref] Figure 2: The UCLH treatment delivery setup [/fig_ref]. The patient is instructed to keep as still as possible during the treatment and monitored using CCTV cameras from the control area. Treatment fields are delivered at a midline dose rate of about 30 cGy/min at 5 cm deep.
## B.2 rfh
The patient is set up lying supine with bent knees on the dedicated TBI couch with their midline aligned to a laser 3.17 m from the isocenter (4.17 m from the source) [fig_ref] Figure 2: The UCLH treatment delivery setup [/fig_ref]. The large 25 mm thick PMMA spoiler screen is placed abutting the couch. The lead compensators are fixed to the block tray of the linac, using the light field shadow to align the edges. The treatment fields are delivered in four equal parts per side, owing to machine limitations on the maximum monitor units per beam. Diode readings are quickly analyzed after each of the four subfields on each side, and the couch is rotated by 180° half way through the treatment. The midline dose rate is 20-25 cGy/min.
## C. in vivo dosimetry
## C.1 uclh
The in vivo dosimetry system at UCLH consists of a 12 channel electrometer (which displays real-time accumulated dose for each delivered treatment fraction and an interface which allows users to write comments about the treatment) connected to 12 semiconductor diodes with an inherent water-equivalent buildup thickness of 1.5 cm (EDP-15, InViDos Scanditronix Wellhofer AB, version 1.2, IBA Wellhofer, Belgium). The system is calibrated every six months under TBI conditions, and a predefined standard temperature correction is automatically applied by the InViDos software to the diode measurements when used on the patient. The data are then stored on a shared network drive and can be accessed at a later point for retrospective review. During treatment, diodes are placed at anatomical positions such as head, shoulder, thorax, center, knees, and ankles for both the left and right sides of the patient. The locations are chosen at positions easy to set up during treatment, away from MLC and field edges, and in an area of the field that does not have high dose gradients (based on the plan). Measured values are a sum of dose contributions from the entrance field and exit field doses for each fraction.
## C.2 rfh
The RFH in vivo dosimetry system consists of up to 20 semiconductor diodes connected via infrared link to a base station (rf-IVD2, Sun Nuclear, Melbourne, FL). In vivo dosimetry is used on every fraction to provide feedback on the delivered doses. Diodes are placed by measuring distances along the patient from the beam central axis. Each diode contains a temperature sensor to automatically compensate for variations in temperature when placed on the patient's skin. These corrections are checked regularly as part of the periodic quality assurance program. Calibration of the system is performed in TBI conditions (at extended SSD, behind the screen and with 2 mm lead on the treatment head), but constancy is checked against a reference reading collected using these setup conditions: all diodes strapped to a 1 cm thick sheet of PMMA, 100 cm SSD, 100 MU, 40 cm × 40 cm, before each fraction. Any diode with constancy readings differing from the reference value, corrected for daily output, by more than 2% are typically excluded from use for that treatment session. Diode readings are taken on each side of the patient, rotating each fraction through a subset of ten sites from head to ankles. Flat (QED, Sun Nuclear, 1115000, 1.9 g/cm 2 buildup) diodes are used for all sites except hands, where cylindrical (ISORAD, Sun Nuclear, 1163000, 1.6 g/cm 2 buildup) diodes are used owing to oblique entry surfaces. Entrance and exit readings for each subbeam are entered into an Excel spreadsheet (Microsoft, Redmond, WA) to give immediate feedback on doses delivered compared to doses expected at 2 cm depth from the planning system, to within 7%. Exit diode readings are corrected for overresponse, and expected readings for underestimation by the planning system, compared to phantom measurements with an ionization chamber. On the first fraction, TLDs are also used at all potential measurement points, with buildup provided by plastic container discs. Gross errors are investigated immediately, and systematic differences are reviewed after the third fraction, at which point the shielding can be modified, if needed, for the remainder of treatment. For example, additional thickness of lead may be added to the region covering the knees and ankles if readings are high in this area.
# D. data analysis
The data accumulated over the years have led to a sizable dataset for analysis to compare techniques. The UCLH measured doses are stored in the InViDos database (as were the MXH measurements). The database is periodically saved and exported to a Microsoft ACCESS database. The diode data were filtered to remove erroneous readings (typically greater than 20% difference to the expected value), which after further investigation of the comments written by users in the InViDos system were found to be due to malfunction or mispositioning, rather than a true difference in treatment delivery. The datasets amalgamated pediatric and adult patients of all sizes, ages, and gender. Single fraction patients were excluded because they would not be useful when analyzing time trends over several treatment fractions. RFH data were extracted from individual spreadsheets using a custom macro, averaged across both beam directions, and filtered to remove single fraction treatments.
# Iii. results
The total number of patients included in the analysis for the MXH and UCLH techniques were 162 and 135, respectively. The filtering mentioned in the previous section resulted in 1162 (444) data points being removed from a total of 19828 (9793) for the MXH (and UCLH) technique, respectively. This corresponds to about 6% of the points and should not influence the analysis. The number of patients included in the RFH analysis was 66, corresponding to 4106 data points. Erroneous readings were predominantly removed at the time of treatment review. shows the percent variation between measured and expected diode doses, averaged across the whole population and over all fractions, for all monitored anatomical sites, for the MXH, UCLH, and RFH techniques. Variation within the population is indicated by the standard deviation (SD) of these values. The most stable position was the center, with SD of 4.1%, 3.7%, . Percentage differences between measured and expected diode doses, averaged across all patients for each anatomical site, for the three techniques discussed in the text. Error bars represent 1 SD. and 1.7%, respectively, for the three techniques. The least stable position was the ankles, with SD of 6.0%, 5.0%, and 2.9%, respectively. [fig_ref] Figure 4: Percentage differences between expected and measured diode doses, averaged across all patients... [/fig_ref] shows the percent variation between expected and measured diode doses, averaged across the whole population as a function of treatment fraction, at the three most significant anatomical sites (shoulders, thorax, and center). The interfractional variation to these sites was 0.6%, 0.9%, and 1.5% for the three techniques: MXH, UCLH, and RFH, respectively.
## A. accuracy
## B. precision
# Iv. discussion
Accuracy is a measure of how close the delivered dose is to the expected dose. For all three techniques, the in vivo measured doses were within ± 10% of the expected doses, which is clinically acceptable. Compared to the MXH technique, the UCLH technique gave similar values for the planned sites of shoulder, thorax, and center, with average measured doses approximately 1% closer to expected values. Head, knees, and ankles, however, which are still compensated using PMMA and bolus following the MXH technique, showed a decrease of 5%-6% relative to MXH differences. The probable reason for this consistent discrepancy is the different way the plans were normalized. The MXH technique forced 100% of the prescription dose at the maximum of the compensated lung slice. The effective separation to be applied to the head and legs was derived from this normalization. The UCLH technique is normalized to midline at the level of the umbilicus and the actual separation there was applied to the head and legs. These different approaches must have produced a difference of about 2.5 cm in required separation.
RFH values showed similar agreement, within 4%, for all sites including extremities. The variation across the patient population was also lower than the UCLH techniques. This may be a result of scanning down below the knees and including bolus within the computerized treatment plan. Shoulder and thorax (lung) values are systematically lower than those at the center (waist). Shoulder and hip regions are often covered by little or no lead compensation, since they are typically the widest separation. Therefore, the irradiation conditions are most different to how the diodes are calibrated, under 2 mm of lead, as typically used over the center. As expected, the most stable position was the center, while the ankles were the least stable, owing to mobility and difficulties in positioning accurately. The measured diode dose data for all techniques are systematically lower than the doses derived from the TPS, but these are within uncertainties in calculation and measurement, so may not be significant.
Precision is how consistently the measured dose is reproduced, analyzed by considering variations as a function of the treatment fractions. Changes over the course of treatment reflects the complexity of identifying suitable positions for the detectors (regions of low dose gradient) and the challenges of consistent patient and diode positioning between fractions. All three techniques showed similar low variation during the treatment, with no discernible trends with fraction number for the MXH and UCLH techniques. For the RFH approach, some increase is seen for the shoulder and thorax doses, but this is closer to the expected values and probably reflects increased experience with patient setup as the treatment progresses.
Comparison to other studies is dependent on the differences in equipment and techniques used. However, a number of other authors have published cohort analyses using CT planning and diode in vivo dosimetry. Mangili et al. [bib_ref] In-vivo dosimetry by diode semiconductors in combination with portal films during TBI:..., Mangili [/bib_ref] treated 229 patients over five years, including 115 using the CadPlan commercial TPS. Agreement of diode measurements at waist and ankles was 0.0% ± 4.5%, reducing to ± 3.5% with CT planning. Bloemen-van Gurp et al. [bib_ref] Total body irradiation, toward optimal individual delivery: dose evaluation with metal oxide..., Bloemen-Van Gurp [/bib_ref] measured five points from neck to ankle using MOSFETs for ten CT-planned patients. Modulation within the plans was achieved using MLC segments, and mean agreement with planned midline doses were -0.5%. Ramm et al. [bib_ref] In vivo dosimetry with semiconducting diodes for dose verification in totalbody irradiation...., Ramm [/bib_ref] used a couch translation technique with CT planning and diode measurements at four points from head to waist, with 80% of readings within 5% of expected values. Lancaster et al. [bib_ref] ): results and analysis, Lancaster [/bib_ref] treated 86 patients over seven years with a CT-planned technique combining AP/PA and lateral fields. Diodes placed at five points from head to pelvis for the first lateral beam fraction to measure mean exit dose were -4.3% ± 7.8% compared to planned values. Values reported in the current study are at least comparable and, in some cases, better compared to these other authors. Both modern techniques at UCLH and RFH are more closely related to the MXH technique than any other published method; therefore, the comparisons made above between the three are the most useful.
Plans calculated in the Pinnacle TPS using a standard beam model have shown good agreement with doses measured in phantom, including modeling of lung shielding by arms, complex aluminum compensators or MLC blocking of head and lung. [bib_ref] TBI treatment planning using the ADAC Pinnacle treatment planning system, Abraham [/bib_ref] [bib_ref] CT-based analysis of dose homogeneity in total body irradiation using lateral beam, Hui [/bib_ref] Lavallée et al. [bib_ref] Commissioning and evaluation of an extended SSD photon model for PINNACLE3: an..., Lavallée [/bib_ref] found improved accuracy with a specific TBI beam model, but this was mainly in the area of out-of-field dose, which was significant for the AP/PA field translating couch technique being used. In addition, the TBI model was unable to account for the first centimeter of the depth dose curve with the beam spoiler in place, owing to the standard model-based nature of the superposition-convolution algorithms. Out-of-field doses are not significant for the UCLH/RFH techniques of lateral fields at extended SSD, therefore no additional modification of the beam model is warranted.
# V. conclusions
The extensive and consistent in vivo measurements of TBI techniques increase the confidence in the dosimetry and delivery of this complex technique. It also gave good evidence that new techniques were equal or superior to old long-established ones.
[fig] Figure 1: Beam's eye view (BEV) showing (a) UCLH MLC shielding and (b) RFH lead compensator sections. [/fig]
[fig] Figure 2: The UCLH treatment delivery setup (a) showing the spoiler screen and diode assembly; the RFH treatment setup (b) showing bolus around neck, hands, shoulders, and legs and the diode positions. [/fig]
[fig] Figure 4: Percentage differences between expected and measured diode doses, averaged across all patients for each treatment fraction, for three anatomical sites for the three techniques discussed in the text. Error bars represent 1 SD. [/fig]
|
Daily Patterns of Preschoolers’ Objectively Measured Step Counts in Six European Countries: Cross-Sectional Results from the ToyBox-Study
This study is part of the ToyBox-study, which is conducted in six European countries (Belgium, Bulgaria, Germany, Greece, Poland and Spain), aiming to develop a cost-effective kindergarten-based, family-involved intervention to prevent overweight and obesity in four-to six-year-old preschool children. In the current study, we aimed to examine and compare preschoolers' step count patterns, across the six European countries. A sample of 3578 preschoolers (mean age: 4.8 ± 0.4) was included. Multilevel analyses were performed to take clustering of measurements into account. Based on the average hourly steps, step count patterns for the six European countries were created for weekdays and weekend days. The step count patterns during weekdays were related to the daily kindergarten schedules.Step count patterns during weekdays showed several significant peaks and troughs (p < 0.01) and clearly reflected the kindergartens' daily schedules, except for Germany. For example, low numbers of steps were observed during afternoon naptimes and high numbers of steps during recess. In Germany, step count patterns did not show clear peaks and troughs, which can be explained by a less structured kindergarten schedule. On weekend days, differences in step count patterns were observed in the absolute number of steps in the afternoon trough and the period in which the evening peak occurred. Differences in step count patterns across the countries can be explained by differences in (school) policy, lifestyle habits, and culture. Therefore, it might be important to respond to these step count patterns and more specifically to tackle the inactive periods during interventions to promote physical activity in preschoolers.
# Introduction
Already in young children, evidence shows a positive relationship between physical activity (PA) and several health outcomes, such as favorable measures of adiposity, bone and skeletal health, motor skill development, psychosocial health, cognitive development and aspects of cardiometabolic health [bib_ref] Physical Activity Guidelines and Guides for Canadians: Facts and Future, Tremblay [/bib_ref]. To establish good health in young children and to maintain it later in life, specific PA recommendations for preschoolers (aged 3-5 years) have been developed. The most recent guidelines suggest that preschoolers should be physically active for at least 180 min. at any intensity on a daily basis [bib_ref] Physical Activity Guidelines and Guides for Canadians: Facts and Future, Tremblay [/bib_ref] , which corresponds to approximately 11,500 steps per day [bib_ref] The translation of preschoolers' physical activity guidelines into a daily step count..., De Craemer [/bib_ref].
Across studies, large variation exists in the percentages of preschoolers achieving the PA recommendations. While some studies using accelerometers showed that all preschoolers achieved the recommended 180 min of total PA per day [bib_ref] Objectively measured physical activity in four-year-old British children: A cross-sectional analysis of..., Hesketh [/bib_ref] [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref] [bib_ref] Physical activity intensity, sedentary time, and body composition in preschoolers, Collings [/bib_ref] , other research indicated that a high percentage of preschoolers is insufficiently physically active. For example, the results of [bib_ref] Preschoolers' Physical Activity, Screen Time, and Compliance with Recommendations, Hinkley [/bib_ref] showed that only 5.1% of Australian preschoolers achieved the recommendation of 180 min per day [bib_ref] Preschoolers' Physical Activity, Screen Time, and Compliance with Recommendations, Hinkley [/bib_ref]. Additionally, results of the European ToyBox-study indicated that the number of preschoolers who achieved the PA guideline of 11,500 steps per day across six European countries was between 26.5% (Greece) and 60.7% (Spain) on weekdays and between 20.3% (Greece) and 41.8% (Poland) on weekend days [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref]. In addition, in the study of [bib_ref] Weekday-weekend patterns of physical activity and screen time in parents and their..., Sigmundova [/bib_ref] , they found that only 47.4% of the preschool girls and 54.1% of the preschool boys achieved the recommended 11,500 steps per day [bib_ref] Weekday-weekend patterns of physical activity and screen time in parents and their..., Sigmundova [/bib_ref]. These inconsistencies across studies may be due to heterogeneity between samples, differences in monitor type, wear positions and data processing issues, like the cut-point selection [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref] [bib_ref] Calibration and comparison of accelerometer cut points in preschool children, Van Cauwenberghe [/bib_ref]. Still, the main conclusion was that efforts to promote (moderate-to-vigorous) PA in preschoolers are needed. To find out at which time of the day preschoolers' PA levels can be enhanced, it is recommended to describe hour-by-hour patterns of preschoolers' PA during weekdays and weekend days.
Thus far, limited information is available regarding hour-by-hour patterns of preschoolers' PA levels. Only a few studies previously described PA patterns in preschoolers [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref] [bib_ref] Objective assessment of levels and patterns of physical activity in preschool children, Brasholt [/bib_ref] [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] [bib_ref] Patterns of physical activity and sedentary behaviour in preschool children, Van Cauwenberghe [/bib_ref]. In three studies, within-day variability in PA on weekdays and weekend days was examined and showed that less PA-peaks and PA-troughs exist in preschoolers' PA patterns throughout weekend days compared to weekdays [bib_ref] Objective assessment of levels and patterns of physical activity in preschool children, Brasholt [/bib_ref] [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] [bib_ref] Patterns of physical activity and sedentary behaviour in preschool children, Van Cauwenberghe [/bib_ref]. During weekend days, comparable results were found across studies, with the lowest PA levels in the morning until 2-3 p.m. [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] [bib_ref] Patterns of physical activity and sedentary behaviour in preschool children, Van Cauwenberghe [/bib_ref] followed by high PA levels in the afternoon [bib_ref] Objective assessment of levels and patterns of physical activity in preschool children, Brasholt [/bib_ref] [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] , followed by a decline in PA levels around 6 p.m. in preschool boys from Denmark (Europe) [bib_ref] Objective assessment of levels and patterns of physical activity in preschool children, Brasholt [/bib_ref] and in boys and girls from Belgium (Europe) and Melbourne (Australia) [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] [bib_ref] Patterns of physical activity and sedentary behaviour in preschool children, Van Cauwenberghe [/bib_ref]. In contrast, PA patterns on weekdays differed strongly across the studies. The Belgian study of [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] showed that PA levels from Belgian preschoolers (2.5 to 6 years) on weekdays are related with their daily preschool schedule, showing PA-peaks during recess periods (i.e., morning and afternoon recess and recess during lunch break) and low levels of PA during classroom based activities. The PA pattern of preschool children in Melbourne showed less peaks and troughs, which can be explained by the fact that data were partly collected during holidays, by a high variability in frequency and duration of preschoolers' presence at preschool, or a less structured curriculum of Australian preschools [bib_ref] Patterns of physical activity and sedentary behaviour in preschool children, Van Cauwenberghe [/bib_ref]. Finally, in the studies of [bib_ref] Objective assessment of levels and patterns of physical activity in preschool children, Brasholt [/bib_ref] and [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref] , different peaks and troughs were observed throughout the day. However, in the study of [bib_ref] Objective assessment of levels and patterns of physical activity in preschool children, Brasholt [/bib_ref] , no specific link was described between the daily preschool schedule and preschoolers' PA-pattern [bib_ref] Objective assessment of levels and patterns of physical activity in preschool children, Brasholt [/bib_ref]. In the study of [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref] , it was suggested that the "free-flow" policies in the UK childcare, which allows preschoolers to choose their activities, could account for preschoolers' high PA-levels during the school day [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref]. The results of these previous studies give a general idea of the fluctuations of PA in preschool children. However, based on the different periods of data collection and the fact that the curriculum of preschoolers was not taken into account across the studies, results cannot be compared between the studies.
Based on the study of De Craemer et al. , which used the same cross-sectional data of the ToyBox-study as in the current study, it became clear that the number of preschoolers that achieved the PA-recommendations during weekdays and weekend days differs across European countries (see File 2 from Supplementary Materials: Preschoolers' PA levels in each country) [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref]. However, the paper did not address the differences in detail, so in this study we looked further and investigated PA differences in detail. The creation and description of hour-by-hour step count patterns during weekdays and weekend days across different countries may explain why these differences in daily step counts exist between the European countries. More specifically, hour-by-hour step count patterns will provide more detailed information on preschoolers' active and inactive time periods during the day.
The aim of the current study was to examine and compare preschoolers' step count patterns during weekdays (when preschoolers attend preschool) and weekend days, based on hour-by-hour step counts across six European countries participating in the ToyBox-study. Furthermore, the step count patterns during weekdays were related to the country-specific daily kindergarten schedules. We hypothesized that preschoolers' step count patterns during weekdays and weekend days would differ between the countries and that several school elements (e.g., recess, classroom based activities) would determine the weekday PA patterns across all countries, because of the prearranged kindergarten schedules.
A goal of the current study was to try to identify a new window of opportunity to increase preschoolers' PA during the day across different European countries. Additionally, this study aimed to provide recommendations, which could help to develop more efficient interventions.
# Methods
## Sample
For the current study, the baseline measurements of the ToyBox-study were used. The ToyBox-study is a large study conducted in six European countries (Belgium, Bulgaria, Germany, Greece, Poland and Spain), which aimed to develop a cost-effective kindergarten-based, family-involved intervention to prevent overweight and obesity in four-to six-year-old preschool children [bib_ref] A systematic approach for the development of a kindergarten-based intervention for the..., Manios [/bib_ref]. Preschoolers were recruited through kindergartens, daycare centers and preschools in the provinces of West-and East-Flanders (Belgium), Varna (Bulgaria), Bavaria (Germany), Attica (Greece), Warsaw and surroundings (Poland) and Zaragoza (Spain). All these settings are referred to as kindergartens in this study. Across all participating countries, kindergarten is not compulsory below the age of fiveand data on Early Childhood Education and Care in Europe reported that between 74.6% (Poland) and 100% (Spain) of all children between 4 years old and the starting age of compulsory education attend kindergarten [bib_ref] Validity of the Omron pedometer and the actigraph step count function in..., De Craemer [/bib_ref]. Within each province, lists of existing municipalities were created and information on the socio-economic status (SES) was provided (based on years of education for the population of 25-55 years and/or annual income). Based on the SES information, tertiles (low SES, medium SES and high SES) were created and five municipalities were randomly selected from each tertile. Finally, kindergartens within these municipalities were randomly selected. In total, 309 kindergartens and 7056 children (between 3.5 and 5.5 years old) and their families participated in the study, with a participation rate of 63.3% per kindergarten. Preschoolers' parents/caregivers were asked to provide written consent before being enrolled in the study. The baseline measurements were conducted between May and June 2012.
Ethical approval was obtained in all six European countries, in line with national regulations (i.e., Ethical Committee of Ghent University Hospital (Belgium), Committee for the Ethics of the Scientific Studies (KENI) at the Medical University of Varna (Bulgaria), Ethikkommission der Medizinischen Fakultät, Ludwig-Maximilians-Universität München (Germany), the Ethics Committee of Harokopio University of Athens (Greece), Ethical Committee of Children's Memorial Health Institute (Poland), and Comité Ético de Investigación Clinica de Aragón (CEICA) (Spain)).
## Measurements
Hour-by-hour step counts were measured by using Omron Walking Style Pro pedometers (HJ-720IT-E2) in Bulgaria, Germany, Greece, Poland and Spain, and Actigraph (GT1M, GT3X and GT3X+ (Pensacola, FL, USA)) accelerometers in Belgium. Accelerometer-based and pedometer-based step counts are valid estimates of preschoolers' PA and both are comparable (r = 0.92) estimates of preschoolers' PA, which justifies combining pedometer-and accelerometer-based step counts [bib_ref] Validity of the Omron pedometer and the actigraph step count function in..., De Craemer [/bib_ref]. In Bulgaria, Germany, Greece, Poland and Spain, pedometers were used because of budget restrictions.
Furthermore, preschoolers' parents/caregivers were asked to fill out the primary caregiver's questionnaire. Based on this questionnaire, data such as age and sex of the participating child, as well as mothers' educational level were obtained. In line with [bib_ref] Differences in weight status and energy-balance related behaviours according to ethnic background..., Brug [/bib_ref] , a low educational level was defined as ≤14 years of education, a high educational level as >14 years of education [bib_ref] Differences in weight status and energy-balance related behaviours according to ethnic background..., Brug [/bib_ref]. In European education systems more than 14 years of education implies attendance of higher education (e.g., bachelor program).
## Procedure
Preschoolers' parents/caregivers received an informational letter and were instructed to let their child wear the measurement device (pedometer or accelerometer) for six consecutive days, including two weekend days, and to remove the device during water-based activities and sleeping. The devices were worn at the right hip secured with an elastic belt around the waist. To ensure that preschoolers wore the measurement device correctly, parents received an information letter which contained instructions on how to handle the device. The first (fitting day) and the sixth day (collection day) were omitted from analyses because these two days did not have complete data.
Step count data were included if preschoolers had two valid weekdays and two valid weekend days, with a valid day defined as at least 6 h of daily wear time [bib_ref] Monitoring of physical activity in young children: How much is enough?, Penpraze [/bib_ref]. All daily step counts below 1000 and above 30,000 steps were indicated as missing data [bib_ref] Measuring physical activity in children with pedometers: Reliability, reactivity, and replacement of..., Rowe [/bib_ref]. The use of these cutoffs was justified in the study of [bib_ref] Measuring physical activity in children with pedometers: Reliability, reactivity, and replacement of..., Rowe [/bib_ref] based on the best agreement among three selection methods (e.g., selection method 1: elimination of at least the most extreme 1%). Based on these inclusion criteria, 977 preschoolers were excluded. Pedometer data were downloaded by using Omron Health Management Software version E1.1012 (Omron Health Care Inc., Bannockburn, IL, USA) and accelerometer data were downloaded by using ActiLife version 5.5.5-software (Actigraph, Fort Walton Beach, FL, USA).
# Statistical methods
Descriptive analyses were executed using SPSS statistics version 22 (SPSS Inc., Chicago, IL, USA) and were reported as means and standard deviations (SD). An independent samples T-test was conducted to compare the average daily step counts between preschool boys and girls, separately for weekdays and weekend days. Additionally, multilevel repeated measures analyses were performed using MLwiN 2.32 to explore hour-by-hour step count patterns for weekdays and weekend days across the six European countries. MLwiN is a commonly used software package for fitting multilevel models into Windows [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref] [bib_ref] The association between the parental perception of the physical neighborhood environment and..., D'haese [/bib_ref] [bib_ref] De Bourdeaudhuij, I. Associations of neighborhood characteristics with active park use: An..., Van Dyck [/bib_ref] (http://www.bristol.ac.uk/cmm/software/mlwin/). Average hourly step counts were calculated between 6 a.m. and 11 p.m. and the averages per hour were compared with each other (e.g., average step counts at 6 a.m. was compared with average step counts at 7 a.m., etc.). In line with the study of [bib_ref] Physical activity intensity, sedentary time, and body composition in preschoolers, Collings [/bib_ref] , periods before 6 a.m. and after 11 p.m. were not included in the analyses because no step counts were recorded in the majority of preschoolers and because this period also reflected their sleeping time [bib_ref] Physical activity intensity, sedentary time, and body composition in preschoolers, Collings [/bib_ref]. To take clustering of measurements into account, four-level multilevel analyses were performed (time, child, classroom, and kindergarten). All analyses were adjusted for sex, age and mothers' educational level. Statistical significance level was set at p < 0.01 to account for multiple testing.
Based on the average hourly step counts, step count patterns were created for the six European countries, separately for weekdays and weekend days. To describe the fluctuations throughout the day, significant peaks and troughs were indicated. A significant peak in step counts occurred when a significant increase in step counts (p < 0.01) was observed, followed by a significant decrease in step counts (p < 0.01) (indicated on the step count patternsby "↑"). When a significant increase in step counts (p < 0.01) occurred, followed by a stagnation in step counts (i.e., no significant increase or decrease, p < 0.01) this was described as a plateau-peak (indicated on the step count patternsby "→"). Finally, a significant trough occurred when a significant decrease (p < 0.01) was followed by a significant increase in preschoolers' step counts (p < 0.01) (indicated on the step count patternsby "↓"). All significant peaks and troughs during weekdays were related to the daily schedules from the kindergartens per country. In File 1 from Supplementary Materials, the country-specific daily kindergarten schedules are described.
[formula] (A) (B) (C) (D) (E) (F) (G) [/formula]
# Results
In this cross-sectional study, valid data were obtained from 11,500 preschoolers (52.3% boys) between 3.5 and 5.5 years old with a mean age of 4.8 (±0.4) years. Descriptive statistics of the sample, reported per country, can be found in [fig_ref] Table 1: Descriptive statistics of preschoolers across the participating European countries [/fig_ref]. Differences in daily step counts were observed between preschool boys and girls, both on weekdays and on weekend days (respectively: t = 11.265, p < 0.01 and t = 6.95, p < 0.01). On weekdays and weekend days, preschool boys achieved a higher amount of step counts compared to preschool girls. The differences in preschoolers' daily step counts are shown in .
## Step count patterns during weekdays
## Description of significant peaks and troughs related to daily kindergarten schedules
During weekdays, preschoolers' step count patterns showed different significant peaks and troughs in step counts across all countries. The number of step counts in all significant peaks and troughs can be found in [fig_ref] Table 3: Number of steps in significant peaks and troughs during weekdays [/fig_ref]. A distinction was made between general peaks and troughs (i.e., peaks and troughs that could be found across at least five countries) and peaks and troughs in the six European countries. In the section below, only the general peaks and troughs will be presented. The step count pattern during weekdays will be related to kindergartens' daily schedules across the different countries (see File 1 from Supplementary Materials). -: No significant peak or trough/no significant increase or decrease; + : General peak 1; ++ : General peak 2; +++ : General peak 3; − : General trough 1; − − : General trough 2; $ : peak or trough in European country.
In total, three significant general peaks were observed. The first significant general peak (p < 0.01) occurred between 8 and 9 a.m. in all countries, except for Germany where no peak could be observed. Between 8 and 9 a.m. was the starting time of the kindergartens in these countries. In this first peak, step counts from 632 (Greece) to 931 (Belgium) steps per hour were observed. The second significant general peak (p < 0.01) was found in all countries and can be related to a recess period, just before (Bulgaria, Spain, Poland, Germany and Greece: 11-12 p.m.) or after lunchtime (Belgium: 12-1 p.m.). In this general peak around 12 p.m., step counts from 940 (Bulgaria) to 1188 (Poland) were observed. Finally, a third general significant peak (p < 0.01) was observed during the late afternoon and evening in all countries, except for Belgium. In this third general peak, preschoolers' step counts raised up to 529 (Greece)-1306 (Bulgaria) steps per hour. The timing of this third peak clearly differed between countries. In Germany, Greece and Poland, this peak took place at kindergartens' closing time and when preschool children went home (Poland: between 3 and 4 p.m.; Germany: between 4 and 5 p.m.). In Bulgaria and Spain, this last significant general peak occurred during after school hours, between 5 and 6 p.m. In Germany, Greece, Poland and Spain, the last peak was maintained for two hours and is therefore called a plateau-peak. In the step count patterns from Belgian preschoolers no third significant general peak was observed, however the step count pattern showed a relatively high number of steps for two hours after Belgian kindergartens' closing time (3 p.m.).
Furthermore, two significant general troughs were observed during weekdays and were also related to the daily schedule of preschool children. The first significant general trough (p < 0.01) occurred between 9 and 10 a.m., when classroom based activities (Belgium, Greece, Poland and Spain) or breakfast (in Bulgaria) were scheduled. Only in the German step count pattern, no significant trough was observed in this period. In this first trough, preschoolers' step counts decreased to 523 (Poland)-681 (Belgium) steps per hour. The second significant general trough (p < 0.01) was observed in the afternoon (between 12 and 4 p.m.), when preschoolers have an afternoon sleep period (Germany, Greece, Bulgaria, and Poland), and/or classroom based activities (Spain, Germany, and Greece).
Step counts from 145 (Bulgaria) to 670 (Spain) steps per hour were observed. Only in Belgium, no significant trough was observed in this period.
## Step count patterns during weekend days
## Description of significant peaks and troughs
During weekend days, comparable step counts-patterns were observed across the different countries, except for Belgium. In Bulgaria, Germany, Greece, Poland and Spain, a first significant peak occurred, followed by a significant trough, followed by a significant second and final peak. However, differences exist in the number of step counts in peaks and troughs and the period in which peaks and troughs occurred across the different countries. The step counts in the first and second peaks as well as the step counts in the trough can be found in [fig_ref] Table 4: Number of steps in significant peaks 1, 2 and in significant trough... [/fig_ref]. In Belgium, two significant increases in step counts (p < 0.01) were observed (one on the morning and one in the evening), however, no significant trough was observed. Between 11 a.m. and 1 p.m., the first significant peak occurred in the countries (p < 0.01), except for Belgium. This peak was observed between 11 a.m. and 12 p.m. in Bulgaria and Germany, and between 12 and 1 p.m. in Greece, Poland and Spain. Preschoolers reached step counts from 842 (Greece) to 1119 (Spain) steps per hour. A second significant peak occurred between 4 and 9 p.m. (p < 0.01), when preschoolers' number of step counts increased up to 909 (Greece)-1093 (Spain) steps per hour. More specifically, a high number of step counts was observed between 4 and 5 p.m. in Germany, between 5 and 6 p.m. in Poland, between 6 and 7 p.m. in Bulgaria, between 7 and 8 p.m. in Spain and between 8 and 9 p.m. in Greece. Finally, in the afternoon, a significant trough (p < 0.01) was observed between 1 and 5 p.m. across all countries, except for Belgium. The timing of this trough clearly differed between countries, namely the lowest number of steps in this trough was found between 1 and 2 p.m. in Germany; between 3 and 4 p.m. in Bulgaria, Spain and Poland; and between 4 and 5 p.m. in Greece. Across the participating countries, the number of step counts in the trough varied between 417 (Greece) and 933 (Poland) steps.
# Discussion
When comparing the step count patterns across the six European countries, during weekdays, several peaks and troughs were observed and step count patterns clearly differed across the countries. However, when taking into account the country-specific kindergarten schedules (see File 1 from Supplementary Materials), it is clear that kindergartens' scheduling (e.g., recess, classroom based activities, and afternoon sleep period) strongly reflected the number of steps in preschool children during kindergarten hours. German preschoolers' step count pattern showed less significant peaks and troughs during weekdays in comparison to the other participating countries. This can be explained by the fact that German kindergartens can schedule their activities (e.g., recess, lunch break, and afternoon sleep periods) individually, which means that the daily kindergarten schedules differ across the kindergartens in Germany.
A previous study, which also used the cross-sectional data of the ToyBox-study, investigated preschoolers' daily step counts during weekdays and showed that the number of preschoolers that achieve the PA guidelines of 11,500 steps per day [bib_ref] The translation of preschoolers' physical activity guidelines into a daily step count..., De Craemer [/bib_ref] differs between countries (see File 2 from Supplementary Materials) [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref]. De Craemer et al. found that, in Spain, the highest percentage of preschool children achieved the recommended 11,500 steps per day, namely 60.7%, followed by Germany, where 49.9% of the preschoolers achieved the PA recommendations. In the current study, the step count pattern in Spain is characterized by several peaks and troughs throughout the day and higher step counts for about 2 h in the evening. This short retention period of high steps for about 2 h may partly result in a high percentage of preschoolers that achieved the daily step count recommendation in the study of De Craemer et al. (2015) [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref]. In Spain, a number of successful regional and local initiatives exist to encourage after-school physical activities. Thus, the availability of these initiatives could be a possible reason for the short retention period of Spanish preschoolers' high step counts in the evening. However, to the best of our knowledge, no data are available on the amount of preschoolers that participate in these extra-curricular activities. In addition, in the other Southern European countries (i.e., Bulgaria and Greece), higher levels of step counts were observed during after school hours compared to the Central European countries (i.e., Belgium, Germany and Poland). This might be explained by better weather conditions in the Southern countries (i.e., Bulgaria, Greece and Spain), and the fact that Southern European people also have different lifestyles, staying outside for a longer period of time in the evening, compared to Central European people. A possible effective strategy to achieve higher number of step counts during after school hours across all countries could be to stimulate both structured and unstructured PA [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref]. Kindergartens could be encouraged to organize extracurricular activities, during after school hours, in the kindergarten setting [bib_ref] Physical activity levels among children attending after-school programs, Trost [/bib_ref] and in the home environment, preschool children can be encouraged to play actively at home or families can be stimulated to be active with their preschool children. Additionally, it seems to be important to stimulate outdoor play as results showed that time spent outdoors is positively associated with children's PA (aged 3-12 years) [bib_ref] What Is the Relationship between Outdoor Time and Physical Activity, Sedentary Behaviour,..., Gray [/bib_ref].
Next to Spanish preschoolers, a relatively high number of German preschoolers also achieved the recommended step counts during a weekday (i.e., 49.9%) [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref]. Based on this finding, the approach used in Germany (i.e., kindergartens that can schedule their program individually) may be effective to reach a higher number of preschoolers achieving the PA recommendations. This is in line with the findings of the study of [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref] , who showed that all preschoolers in Cambridgeshire (the UK) achieved the PA guideline of 180 min. per day, which may be due to the kindergarten "free-flow" policy allowing preschoolers to choose their activities, freely moving between inside and outside environment for most of the day [bib_ref] UK Preschool-aged children's physical activity levels in childcare and at home: A..., Hesketh [/bib_ref]. In the other countries, clear step count patterns were observed, which could mean that preschoolers' schedules are more fixed in these countries. This shows that preschoolers are almost only able to play at pre-arranged time periods (i.e., during recess), which is comparable with the findings of PA patterns in primary school aged children [bib_ref] Objectively measured physical activity, physical activity related personality and body mass index..., Deforche [/bib_ref]. In Germany, this kindergarten-specific approach may provide more flexibility to teachers to meet the needs of the preschoolers, which may result in more PA during kindergarten hours and consequently, in a higher amount of preschoolers that reach the PA recommendations.
As most troughs in step counts during school hours were probably caused by classroom based activities, this might suggest that those activities are not activity enhancing. Similar results were found in the study of [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] , who indicated that blocks of sedentary classroom based activities are already present in Flemish preschool children [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref]. To enhance preschoolers' PA during school hours, it can be recommended for teachers to integrate more PA (i.e., implement PA in the daily curriculum or introduce movement breaks) into their daily classroom based activities. [bib_ref] Feasibility and efficacy of a "move and learn" physical activity curriculum in..., Trost [/bib_ref] concluded that integrating PA in all aspects of the kindergarten curriculum (e.g., into mathematics, nutrition education) is feasible and a potential effective strategy to increase PA in preschool children [bib_ref] Feasibility and efficacy of a "move and learn" physical activity curriculum in..., Trost [/bib_ref]. In addition, incorporating classroom based PA-breaks seemed to be effective to increase preschoolers' PA levels during the school day.
Across all countries, relatively high numbers of step counts were found during recess periods during the hours at kindergarten (i.e., at kindergartens' starting time, during the recess period just before or after lunch time and during kindergartens' closing time). Similar results were found in the study of [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref] in which preschoolers' recess period was associated with the highest PA levels during the day [bib_ref] Within-and between-day variability of objectively measured physical activity in preschoolers, Verbestel [/bib_ref]. In the study of [bib_ref] Are preschool children active enough? Objectively measured physical activity levels, Cardon [/bib_ref] , results showed that preschoolers' higher step counts per minute were significantly associated with more space per child and with shorter recess time. Therefore, it is recommended for teachers and headmasters to provide sufficient space during recess, and if necessary to split into groups with different recess times [bib_ref] Are preschool children active enough? Objectively measured physical activity levels, Cardon [/bib_ref].
The lowest numbers of preschoolers that achieved the PA recommendations on weekdays were found in Bulgaria and Greece, respectively, 29.3% and 26.5% [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref]. In Bulgaria, this result may be partly explained by the afternoon sleep period that is scheduled in the kindergarten program (between 1 and 4 p.m.). Remarkably, all preschools in Bulgaria are equipped with beds for afternoon naptimes for each preschool child. In addition, in some other countries, namely Germany, Greece, Poland and Spain, preschoolers get the opportunity to nap in the afternoon (see File 1 from Supplementary Materials). Giving the opportunity for a nap may benefit some preschoolers, particularly those who are younger and had less than 10 h of sleep the previous night [bib_ref] Sleep and napping patterns in 3-to-5-year old children attending full-day childcare centers, Ward [/bib_ref]. However, for other preschoolers or preschool children that had sufficient hours of sleep during the night, restlessness, anxiety and aggression during and after the afternoon rest period may occur [bib_ref] Sleep and napping patterns in 3-to-5-year old children attending full-day childcare centers, Ward [/bib_ref]. Therefore, we recommend that kindergartens avoid systematically putting all preschool children in bed in the afternoon and, instead, respond to the needs of each individual preschooler.
During weekend days, a relatively high proportion of Polish and Spanish preschoolers achieved the daily PA recommendations of 11,500 steps (respectively 41.8% and 37.0%) [bib_ref] Differences in energy balance-related behaviours in European preschool children: The ToyBox-study, De Craemer [/bib_ref]. However, this study showed that the step count patterns of Poland and Spain have different fluctuations in step counts over time. These step count patterns differ at two important time points. More specifically, the afternoon trough in Spain is deeper (i.e., lower step counts) compared to Poland, and the late-evening peak (last peak) is higher in Spain compared to Poland. The fluctuation of the PA pattern in Spain (i.e., a deep afternoon trough and late-evening peak) was also found in the other Southern European countries (i.e., Bulgaria and Greece) which might be due to the siesta which takes place during the warmest part of the day. The late-evening peak might be explained by Southern European children staying active for a longer period during the day due to hotter weather conditions, compared to the Central European countries. A possible strategy to promote PA in preschoolers could be to take into account the above-mentioned differences in step count patterns. In Southern European countries, not ignoring the importance of taking a siesta and the hot weather conditions, a possible strategy could be to promote PA, especially in the morning and in the late afternoon or evening. Furthermore, it might be an option to shorten the afternoon trough by promoting PA in more comfortable temperatures (e.g., indoor PA). As the afternoon trough cannot really be explained in the Central European countries, weekend afternoons are a good opportunity to enhance preschoolers PA levels.
A limitation of this study was the use of PA patterns based on step counts instead of accelerations. Due to the high cost of accelerometers (~US $249 or €190 per device) it was not possible to use these devices in all countries (except for Belgium). Based on the literature, it is clear that pedometers are less accurate to measure PA, as they cannot measure PA intensities, compared to accelerometers. However, the study of [bib_ref] Pedometer accuracy in physical activity assessment of preschool children, Oliver [/bib_ref] indicated that pedometers are suitable for situations where a general idea of preschoolers' PA is required, which was the aim of the current study [bib_ref] Pedometer accuracy in physical activity assessment of preschool children, Oliver [/bib_ref]. Furthermore, results on the country-specific level should be interpreted cautiously because in some of the countries (i.e., Bulgaria, Germany and Greece) data originate from only one city. Due to the recruitment of preschoolers in selected areas within each country, the ToyBox-sample is not a fully representative European sample. However, within the ToyBox-sample preschoolers were included from different socio-economic backgrounds (low, medium and high), and in each kindergarten (almost) complete classes were included. Thus, the sample can give a fair approximation of the current situation within Europe. Besides, as already mentioned before, preschoolers were recruited through kindergartens, daycare centers and preschools. However, data regarding the frequencies in each group or how these groups were represented across countries have not been recorded in the ToyBox-study. Finally, all measurements were collected between May and June 2012 across the participating countries, which strengthens the comparability across the countries. However, an overestimation of preschoolers' step counts may have occurred since the colder months (e.g., December, January) were not part of the measurement timeframe. Due to health promotion initiatives in European countries in recent years, PA data, based on data collected in 2012, may slightly differ from the current situation (2018).
There are also some strengths that should be acknowledged. First, a large sample size of 3578 preschoolers was included in this study, although we used a strict number of included valid days (i.e., two valid weekdays and two valid weekend days). Additionally, step count patterns were described across six different countries, which made it possible to compare cross-European step count patterns during weekdays and weekend days. Finally, multilevel analyses were used to take clustering of preschoolers at five levels into account.
# Conclusions
Preschoolers' step count patterns were described and compared across six European countries taking into account the daily step counts during weekdays and weekend days. Weekday step count patterns were related to daily kindergarten schedules across all countries. It became clear that differences in step count patterns were observed across the countries both during weekdays and weekend days, especially between Southern and Central European countries, which may be explained by differences in (school) policy, lifestyle habits or culture. During weekdays, special attention is needed to enhance preschoolers' PA levels during after school hours across the Central European countries. Furthermore, in particular in Bulgaria and Greece, a recommendation might be to avoid putting all preschoolers in bed in the afternoon, but to respond to the needs of each individual preschoolers instead, as low step counts in preschoolers may be partly explained by the afternoon sleep period. During weekend days, preschoolers in Central European countries should be stimulated to be more physically active during the afternoon. On the other hand, taking into account the common habit of taking a siesta in Southern European countries, preschoolers can be encouraged to enhance their PA levels in the morning and in the late afternoon or evening. It is important to take these step count patterns into account in the development of interventions to promote PA in preschoolers. Author Contributions: V.V.S. analyzed and interpreted the data regarding the PA patterns from preschool children across six different countries. All authors read and approved the final manuscript of this research.
## Conflicts of interest:
The authors declare no conflict of interest.
## Availability of data and material:
The datasets analyzed during the current study are available from the corresponding author on reasonable request.
## Abbreviations
## Pa
Physical activity SES Socio-economic status
[fig] Figure 1: (A) Preschoolers' step count patterns during weekdays in Belgium; (B) preschoolers' step count patterns during weekdays in Bulgaria; (C) preschoolers' step count patterns during weekdays in Germany; (D) preschoolers' step count patterns during weekdays in Greece; (E) preschoolers' step count patterns during weekdays in Poland; (F) preschoolers' step count patterns during weekdays in Spain; and (G) preschoolers' step count patterns during weekdays across all participating countries. [/fig]
[fig] Figure 2: (A) Preschoolers' step count patterns during weekend days in Belgium; (B) preschoolers' step count patterns during weekend days in Bulgaria; (C) preschoolers' step count patterns during weekend days in Germany; (D) preschoolers' step count patterns during weekend days in Greece; (E) preschoolers' step count patterns during weekend days in Poland; (F) preschoolers' step count patterns during weekend days in Spain; and (G) preschoolers' step count patterns during weekend days across all participating countries. [/fig]
[fig] Supplementary: Materials: The following are available online at www.mdpi.com/1660-4601/15/2/291/s1, File 1: Description of daily kindergarten schedules in six European countries; File 2: Preschoolers' PA levels in each country. [/fig]
[fig] Funding: The ToyBox-study is funded by the Seventh Framework Programme (CORDIS FP7) of the European Commission under grant agreement No. 245200. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.Ethics Approval and Consent toParticipate: We declare that all applicable institutional regulations pertaining to the ethical use of human volunteers were followed during this research. Ethical approval was obtained in all sex in line with national regulations (i.e., Ethical committee of Ghent University Hospital (Belgium) (2012/110), Committee for the Ethics of the Scientific Studies (KENI) at the Medical University of Varna (Bulgaria) (15/21-07-2011), Ethikkommission der Ludwig-Maximilians-Universität München (Germany) (400-11), the Ethics Committee of Harokopio of Athens (Greece) (28/02-12-2010), Ethical Committee of Children's Memorial Health Institute (Poland) (1/KBE/2012), and CEICA (Comité Etico de Investigacion Clinica de Aragon (Spain) (12/2011)). [/fig]
[table] Table 1: Descriptive statistics of preschoolers across the participating European countries.Table 2. Amount of step counts in preschool boys and girls, across the participating countries. [/table]
[table] Table 3: Number of steps in significant peaks and troughs during weekdays. [/table]
[table] Table 4: Number of steps in significant peaks 1, 2 and in significant trough 1 during weekend days. [/table]
|
No interaction between serotonin transporter gene (5-HTTLPR) polymorphism and adversity on depression among Japanese children and adolescents
Background: Identification of gene × environment interactions (G × E) for depression is a crucial step in ascertaining the mechanisms underpinning the disorder. Earlier studies have indicated strong genetic influences and numerous environmental risk factors. In relation to childhood and adolescent depression, evidence is accumulating that the quality of the parental environment is associated with serotonin biology in children. We hypothesized that maternal depression is a crucial environmental risk factor associated with serotonin-regulating genes.Methods: This study was designed to ascertain the G × E interaction for diagnosis of depression in a Japanese pediatric sample. DNA samples from 55 pediatric patients with depression and 58 healthy schoolchildren were genotyped for the 5-HTT (2 short (S) alleles at the 5-HTT locus) promoter serotonin-transporter-linked polymorphic region (5-HTTLPR) polymorphism. We examined whether an adverse parental environment, operationalized as the mother's history of recurrent major depressive disorder, interacts with 5-HTTLPR polymorphism to predict patients' depression symptoms. Results: Binary logistic regression analyses revealed that maternal depression (adversity), gender, and FSIQ significantly affect the diagnosis of depression among children and adolescents. However, no main effect was found for adversity or genotype. Results of multivariable logistic regression analyses using stepwise procedure have elicited some models with a good fit index, which also suggests no interaction between 5-HTTLPR and adversity on depression.Conclusions: To assess G × E interaction, data obtained from children and adolescents who had been carefully diagnosed categorically and data from age-matched controls were analyzed using logistic regression. Despite an equivocal interaction effect, adversity and gender showed significant main effects.
# Background
Depression, an etiologically heterogeneous group of brain disorders characterized by widely various symptoms that reflect altered cognitive, psychomotor, and emotional processes, strongly affects 3-5% of children and adolescents [bib_ref] The prevalence of DSM-III-R diagnoses in a national sample of Dutch adolescents, Verhulst [/bib_ref]. Depression negatively impacts growth and development, school performance, and peer or family relationships; it can lead to suicide [bib_ref] Psychiatric diagnosis in child and adolescent suicide, Shaffer [/bib_ref] [bib_ref] Incidence of major depressive disorder and dysthymia in young adolescents, Garrison [/bib_ref]. Biomedical and psychosocial risk factors include a family history of depression, female sex, childhood abuse or neglect, stressful life events, and chronic illness [bib_ref] The prevalence of DSM-III-R diagnoses in a national sample of Dutch adolescents, Verhulst [/bib_ref] [bib_ref] Emotional disclosure in school children, Reynolds [/bib_ref] [bib_ref] Childhood and adolescent depression, Bhatia [/bib_ref]. Previous results of studies suggest that exposure to childhood adversity is associated with depression symptoms [bib_ref] Gene-environment interaction and the genetics of depression, Lesch [/bib_ref] [bib_ref] Genetics and genomics of depression, Hamet [/bib_ref] [bib_ref] The interaction of stressful life events and a serotonin transporter polymorphism in..., Kendler [/bib_ref]. Although people at risk tend to have poor life outcomes, some people do well despite adversity. Protective factors such as good intellectual skills, positive temperament, parental warmth, parental involvement, and strong social connections are believed to play an important role in protecting individuals from poor development, psychological problems, and mental illness from the viewpoint of genetic etiology [bib_ref] Exposure to violence during childhood is associated with telomere erosion from 5..., Shalev [/bib_ref].
Results of previous studies suggest that homozygous carriers of the short allele (S/S) of the serotonin transporter gene-linked polymorphic region (5-HTTLPR) might be at increased risk for depression if they have also been exposed to early or current adversity or stress [bib_ref] Influence of life stress on depression: moderation by a polymorphism in the..., Caspi [/bib_ref].
A growing body of evidence suggests that 5-HTTLPR interacts with adverse environmental influences to produce increased risk for the development of depression, although the underlying mechanisms of this association remain largely unexplored [bib_ref] Gene-by-environment (serotonin transporter and childhood maltreatment) interaction for anxiety sensitivity, an intermediate..., Stein [/bib_ref] [bib_ref] Geneenvironment interactions predict cortisol responses after acute stress: implications for the etiology..., Alexander [/bib_ref] [bib_ref] Genetic sensitivity to the environment: the case of the serotonin transporter gene..., Caspi [/bib_ref] [bib_ref] Interaction of serotonin transporter gene-linked polymorphic region and stressful life events predicts..., Mueller [/bib_ref] [bib_ref] Individual differences in neural correlates of fear conditioning as a function of..., Klucken [/bib_ref]. Various reports have described that 5-HTTLPR genotype moderates the relation between stress and depression and interacts to contribute to risk for depression in children [bib_ref] Brain-derived neurotrophic factor-5-HTTLPR gene interactions and environmental modifiers of depression in children, Kaufman [/bib_ref] [bib_ref] Interaction between childhood adversity, brain-derived neurotrophic factor val/met and serotonin transporter promoter..., Nederhof [/bib_ref] [bib_ref] The serotonin transporter promoter variant (5-HTTLPR), stress, and depression meta-analysis revisited: evidence..., Karg [/bib_ref]. However, recent metaanalyses have produced no evidence that the 5-HTTLPR genotype alone or in interaction with stressful life events is associated with an elevated risk of depression [bib_ref] Interaction between the serotonin transporter gene (5-HTTLPR), stressful life events, and risk..., Risch [/bib_ref] , or that the main effect of 5-HTTLPR genotype and the interaction effect between 5-HTTLPR and stressful life events on risk of depression are negligible [bib_ref] Gene X environment interactions at the serotonin transporter locus, Munafo [/bib_ref]. Therefore, we should conduct an examination to incorporate consideration of the possibility of other confounding variables such as gender, age, and socioeconomic status along with stressful life events.
Research in this area can recast our thinking about the role of early experience in psychopathology and genetic interaction. Considering previous research, we hypothesized that interaction between 5-HTTLPR and early adversity is involved in the etiology of childhood depression. No study has tested Japanese children and adolescents. Therefore, the aim of this study is to examine whether opposite G (5-HTTLPR genotype) × E (maternal depression) interactions can be confirmed among Japanese children and adolescents. Our additional research findings will contribute to the resolution of inconsistencies in the literature. One principle aim in the current study was to test whether opposite gene-by-environment (G × E) interactions with maternal depression can be found for Japanese children and adolescents with childhood depression. This study examined G × E interactions in a sample of children with Japanese pediatric depression to test the hypothesis that 5-HTTLPR is involved in the etiology of childhood depression.
# Methods
# Ethics statement
The Committee of Life Ethics, Graduate School of Medicine, Kumamoto University approved the study protocol, Assurance # KUM0313. During the review of this Project, the committee specifically considered (i) the risks and anticipated benefits, if any, to subjects; (ii) the selection of subjects; (iii) the procedures for securing and documenting informed consent; (iv) the safety of subjects; and (v) the privacy of subjects and confidentiality of the data. All potential participants who declined to participate or otherwise did not participate were eligible for treatment (if applicable) and were not disadvantaged in any other way by not participating in the study. According to the Declaration of Helsinki, parents or guardians of all participants provided written informed consent for children to participate in the study after the study procedures had been explained to them.
## Subjects
This study examined 55 unmedicated patients with depression: 21 boys and 34 girls aged 8-16 years (mean age, 14.3 years; standard deviation [SD], 1.9 years) who were referred to our laboratory during 2007-2011 for examination of depression [fig_ref] Table 1: Demographics, early adversity, and clinical characteristics in individuals in patients and healthy... [/fig_ref]. All the patients were referred from their private pediatric clinics. For that reason, there might be referral bias related to severity despite the drug-naïve condition used for this study. All patients satisfied diagnostic criteria for depression of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) (American Psychiatric Association, 2000). To exclude other psychiatric diagnoses, the Mini-International Neuropsychiatric Interview for Children and Adolescents (MINI Kids) was administered by a licensed pediatric-psychiatric clinician. The patient group did not have high levels of suicidality or comorbidity.
To obtain data from normal age-matched healthy controls (HC), healthy schoolchildren and adolescents aged 8-16 years were recruited as subjects from the community; school students were targeted. The control group comprised 40 boys and 18 girls (mean age, 13.8 years; SD, 2.1 years). None had below-average IQ, physical problems, or psychiatric psychopathology.
All participants' race/ethnicity was Japanese. Patients who had undergone treatment with antidepressants or hypotension drugs, or who had diagnoses of neurological illness, migraine, obstructive sleep apnea, below average IQ, or severe psychopathology were excluded from the study. Severe psychopathology was evaluated by referral to at least one pediatric psychiatrist if the patient presented indicative symptoms.
No control subject had any history of DSM-IV-TR Axis I Disorder (based on MINI Kids) including attention deficit hyperactivity disorder (ADHD) or other pervasive developmental disorder, or any history of any form of abuse or drug abuse. [fig_ref] Table 1: Demographics, early adversity, and clinical characteristics in individuals in patients and healthy... [/fig_ref] presents clinical characteristics of the subjects.
As an adult control reference, we also used commercially supplied DNA of Japanese adult subjects (age 44.26 ± 13.58, 50 male and 50 female) supplied by the Health Science Research Resources Bank, Osaka, Japan.
## Assessments
## Depression scales
The severity of depressive symptoms was measured using the Zung self-rating depression scale (SDS) and the Birleson Depression Self-Rating Scale for Children (DSRS-C), which are similar to the children's depression inventory used in pediatric psychiatry. The validity and reliability of Japanese version of these scales were confirmed [bib_ref] A Self-Rating Depression Scale, Zung [/bib_ref] [bib_ref] The validity of depressive disorder in childhood and the development of a..., Birleson [/bib_ref]. Methods used for these analyses were described previously [bib_ref] Event-Related Potentials in Japanese Childhood Chronic Fatigue Syndrome (CCFS), Tomoda [/bib_ref].
## Wechsler intelligence scale for children-third edition (wisc-iii)
An individually administered measure of intelligence intended for children aged 6 years to 16 years and 11 months, WISC-III, was administered by a licensed pediatric-psychological clinician to estimate intellectual capacity. The version is applicable to Japanese children and adolescents.
Child Behavior Checklist Youth Self-Report (CBCL-YSR) and SES Patients with child depression had mean scores of 25 and 14, respectively, for Internalizing and Externalizing Problems on the CBCL Youth Self-Report.
Low income and poverty might be important developmental risk factors related to psychopathology [bib_ref] Pseudohypacusis in childhood and adolescence is associated with increased gray matter volume..., Tomoda [/bib_ref]. The parental socioeconomic status (SES) test was administered as a composite measure of socioeconomic status [bib_ref] Hollingshead two factor index of social position, occupational categories, Hollingshead [/bib_ref] [bib_ref] Social class and mental illness: a community study, Hollingshead [/bib_ref].
Socioeconomic status was classified as follows: annual income below 2 million yen, 1; 2-4 million yen, 2; 4-6 million yen, 3; between 6 and 8 million yen, 4, 8-10 million yen, 5; 10-12 million yen, 6; more than 12 million yen, 7.
## Early adversity
The patients' mothers were also administered a clinical interview by a specialized physician, the Structured Clinical Interview for the DSM-IV (SCID), to assess their Axis-I disorders. They were investigated if they had either (1) no current/past disorder (low early adversity); or (2) at least two major depressive episodes during their children's lifetime but were currently in remission (high early adversity). Approximately 20% (n = 11) of the mothers had recurrent depression; the remaining 80% (n = 44) had no history of mental health disorders. No control subject's mother had high early adversity. Selected from the same community sample, control subjects' mothers with no history of psychiatric problems or exposure to traumatic events served as a no early adversity contrast group.
## Genotyping
Genomic DNA was extracted directly from blood samples of patients using standard techniques (MagNa pure LC350; Roche Diagnostics Corp.). In control subjects, we used a mouth swab sampling technique and extracted the DNA from buccal cells using a standard commercial extraction kit QIAamp DNA Micro Kit (Qiagen Inc., Tokyo, Japan). This study was conducted with ethnically homogeneous individuals (all were of Japanese descent) [bib_ref] Perceived parental rejection mediates the influence of serotonin transporter gene (5-HTTLPR) polymorphisms..., Nishikawa [/bib_ref]. The serotonin transporter gene (5-HTTLPR) of the 5-HTT gene regulatory region was amplified by polymerase chain reaction (PCR) with forward primer (5'-GGCGTT GCCGCTCTGAATGC-3') and reverse primer (5'-GAGG GACTGAGCTGGACAACCAC-3'). For PCR, 10 ng of genomic DNA was used in a 25-μL reaction mixture containing 0.5 U of KOD FX Neo (Toyobo Co. Ltd.) and 10 pmol of each primer in PCR buffer for KOD FX Neo (Toyobo Co. Ltd.). Cycling conditions were the following: denaturation (94°C for 2 min) and 30 cycles of amplification (98°C for 10 s, 63°C for 30 s and 68°C for 30 s). The PCR products were separated using electrophoresis in a 3% agarose gel and visualized by UV after ethidium bromide staining. A 484 bp band was observed for the short (S) allele, and a 528 bp band for the long (L) allele; heterozygous samples showed both the alleles. Two investigators scored allele sizes independently. Any inconsistency was reviewed and procedures were repeated if necessary. Genotyping was conducted blindly without knowledge of the clinical information.
The PCR-based analyses revealed that the distribution of allele frequency in the patients was 76% for the S-allele and 24% for the L-allele [fig_ref] Table 2 5: -HTTLPR genotype distributions and allele frequencies in controls and depressive patients [/fig_ref]. The distribution of the S/S, S/L, and L/L genotypes was, respectively, 60% (n = 33), 33% (n = 18), and 3% (n = 7). In the Japanese population, LL type of 5-HTTLPR has been reported as rare. Therefore, it might be better to examine the 5-HTTLPR proportion using Fisher's exact test or MCMC. However, some previous reports have described the use of Hardy-Weinberg equilibrium examination empirically [bib_ref] Impact of serotonin transporter and catechol-O-methyl transferase genes polymorphism on gastrointestinal dysfunction..., Obayashi [/bib_ref] [bib_ref] Serotonin transporter gene polymorphism may be associated with functional dyspepsia in a..., Toyoshima [/bib_ref]. We also administered Hardy-Weinberg equilibrium examination. The observed genotype distribution was similar to the expected Hardy-Weinberg equilibrium distribution. The genotypes were divided into S/S and S/L + L/L genotypes because only four subjects with the L/L genotype were found in our sample. This genotype is uncommon among Japanese people [bib_ref] Serotonin transporter protein (SLC6A4) allele and haplotype frequencies and linkage disequilibria in..., Gelernter [/bib_ref].
# Data analysis
The clinical variables were compared in both groups using t-tests for quantitative data and Fisher exact tests for qualitative data. Data are expressed as the mean ± SD. To explore the relation between depressive symptoms and the genotypes of the 5-HTTLPR polymorphism, analysis of covariance (ANCOVA) was performed for each of the depression scores (total SDS and DSRS-C scores) using age and gender as covariates because it has been suggested that a variable such as age might affect responses to the depression rating [bib_ref] Pseudohypacusis in childhood and adolescence is associated with increased gray matter volume..., Tomoda [/bib_ref]. To confirm the main effect on the diagnosis of depression, we used binary logistic regression and set genetic, environmental factors, and the effect of the interactions between 5-HTTLPR genotype and adversity as independent variables.
Next, multivariable logistic regression analyses were performed. The dependent variable was the status of diagnosis (0 = no diagnosis, 1 = diagnosed as depression). All predictor variables were entered simultaneously into the model. Some possible models were conducted using stepwise procedures. Finally, two-way analysis of variance (ANOVA) was conducted to assess the effect of G × E interaction in each genotype group.
All statistical tests (t-test, ANCOVA and Fisher exact test, logistic regression analyses, and two-way ANOVA for discrete variables) were two-sided; P values less than .05 were inferred as statistically significant. Statistical analyses were conducted using software (SPSS Statistics 20; SPSS Inc., Chicago, IL).
# Results
Demographics and IQ [fig_ref] Table 1: Demographics, early adversity, and clinical characteristics in individuals in patients and healthy... [/fig_ref] shows that the patient and control groups were well matched in terms of age (F =1.21, P = 0.40) and SES (F = 2.62, P = 0.11). Female predominance was found in the patient group (Fisher exact, P = 0.008). Subjects in the patient group were predominantly male (69%), whereas controls were predominantly female (61%; [fig_ref] Table 1: Demographics, early adversity, and clinical characteristics in individuals in patients and healthy... [/fig_ref]. Of 55 children and adolescents in the patient group, 11 had early adversity, although none had it in the HC group. As expected, the patient group had high SDS scores with a mean of 57.0 ± 6.4 (F = 235.96, P < 0.001) and high DSRS-C scores with a mean of 21.9 ± 6.2 (F = 189.03, P < 0.001). In addition, those who were diagnosed as having depression had a significantly higher total score of CBCL-YSR (F =23.59, P < 0.001) and internalizing score of that (F = 63.91, P < 0.001) than those of the HC group.
The patient group had significantly lower full-scale IQ (FSIQ) without discrepancies between verbal IQ (VIQ) and performance IQ (PIQ), compared to the HC group Similarly, VIQ and PIQ showed significant differences in scores between the two groups (VIQ, F = 9.44, P = 0.003; PIQ, F = 7.67, P = 0.007). The genotypic and allelic frequencies of the 5-HTTLPR variants are presented in [fig_ref] Table 2 5: -HTTLPR genotype distributions and allele frequencies in controls and depressive patients [/fig_ref]. The distribution of allelic frequency of the patients and age-matched controls was found to be almost identical to that previously reported in Japanese people, exhibiting a similar tendency to that of our Japanese adult sample (frequency for the L/L, S/L, and S/S genotypes was: 7, 33, and 60% in the patients and 6, 31, and 63% in the age-matched controls, respectively). The overall pattern of results derived by genetic studies of 5-HTTLPR is most suggestive for a dominant mode of action of the S allele [bib_ref] A polymorphism (5-HTTLPR) in the serotonin transporter promoter gene is associated with..., Willeit [/bib_ref]. We therefore analyzed S/L heterozygous patients grouped with L/L homozygous patients (i.e., L/L + S/L versus S/S). No significant difference was found in the genotypic or allelic frequencies of 5-HTTLPR between both groups, suggesting that no sampling bias exists in the study.
As shown [fig_ref] Table 1: Demographics, early adversity, and clinical characteristics in individuals in patients and healthy... [/fig_ref] , significant differences were found between two groups in FSIQ, depression scores, and scores of CBCL. Because of the significant main effects of FSIQ and gender, we assumed that interaction effects between variables such as SES × age, SES × FSIQ, and gender × age would be large.
## Evaluation of binary logistic regression analyses
In this study, we examined the main effect of diagnosis as depression in childhood and adolescence using binary logistic regression analyses. Adversity, gender, and FSIQ were found to be significantly associated with depression [adversity, P <0.0001, OR = 2129500000; gender, P =0.008, OR (95% CI) = 0.345 (0.158-0.754); FSIQ: P =0.001, OR (95% CI) =0.931 (00.892-0.971)]. A significant G × E interaction was observed (P <0.0001, OR = 1142600000). However no significant main effect in genotype was found [fig_ref] Table 3: Binary logistic regression analyses for variables of genotype, adversity, gender, age, SES,... [/fig_ref].
## Evaluation of multivariable logistic regression analyses
As results of multivariable logistic regression analyses using stepwise procedure, we confirmed three models that have goodness of fit [fig_ref] Table 4: Multivariable logistic regression analyses using stepwise procedure [/fig_ref]. In Model 1, significant effects of G × E interaction, adversity, gender, and SES were found We conducted two-way ANOVA to examine the main and confounding effects on DSRS-C score in all participants. No significant G × E interaction between 5-HTTLPR genotype and adversity or main effects of genotype, adversity, and gender was found from these analyses [fig_ref] Table 5 Two: Dependent variable is Birleson score [/fig_ref]. Additionally, the mean scores of DSRS-C in each genotype group were analyzed. No significant difference was found in any group.
# Discussion
Interaction between adverse parental environment and gender showed a significant main effect despite a lack of G × E interaction in a Japanese pediatric sample. Because all participants in this study were ethnically homogenous, this is an important consideration for generalizing the present findings.
Depression is a critical and common condition found in children and adolescents as well as adults [bib_ref] The prevalence of DSM-III-R diagnoses in a national sample of Dutch adolescents, Verhulst [/bib_ref] [bib_ref] Epidemiology of major depressive disorder: results from the National Epidemiologic Survey on..., Hasin [/bib_ref] [bib_ref] Witnessing violence toward siblings: an understudied but potent form of early adversity, Teicher [/bib_ref]. Early life stress is a risk factor for major depression, posttraumatic stress disorder (PTSD), and drug abuse, among other conditions [bib_ref] Childhood and adolescent depression, Bhatia [/bib_ref] [bib_ref] Adverse childhood experiences and mental health in young adults: a longitudinal survey, Schilling [/bib_ref]. In recent years, a possible association of the 5-HTTLPR with youth depression has been traced in numerous studies [bib_ref] Early adversity and 5-HTT/BDNF genes: new evidence of gene-environment interactions on depressive..., Aguilera [/bib_ref] [bib_ref] Effects of stressful life events, maternal depression and 5-HTTLPR genotype on emotional..., Araya [/bib_ref] [bib_ref] Genetic indeterminism, the 5-HTTLPR, and the paths forward in neuropsychiatric genetics, Blakely [/bib_ref] [bib_ref] Depression and the serotonin transporter 5-HTTLPR polymorphism: a review and a hypothesis..., Brown [/bib_ref] [bib_ref] Serotonin transporter polymorphism predicts waking cortisol in young girls, Chen [/bib_ref] [bib_ref] Interactive effects of corticotropin releasing hormone receptor 1, serotonin transporter linked polymorphic..., Cicchetti [/bib_ref] [bib_ref] Interaction of child maltreatment and 5-HTT polymorphisms: suicidal ideation among children from..., Cicchetti [/bib_ref] [bib_ref] Childhood stress, serotonin transporter gene and brain structures in major depression, Frodl [/bib_ref] [bib_ref] Altered timing of amygdala activation during sad mood elaboration as a function..., Furman [/bib_ref] [bib_ref] Children's attentional biases and 5-HTTLPR genotype: potential mechanisms linking mother and child..., Gibb [/bib_ref] [bib_ref] Brooding rumination and risk for depressive disorders in children of depressed mothers, Gibb [/bib_ref] [bib_ref] Children's inferential styles, 5-HTTLPR genotype, and maternal expressed emotion-criticism: An integrated model..., Gibb [/bib_ref]. Most studies were based on Caucasian populations, except for a few examining the Japanese general population [bib_ref] Short allele of 5-HTTLPR as a risk factor for the development of..., Ezaki [/bib_ref] [bib_ref] Polymorphism of the serotonin transporter gene modulates brain and physiological responses to..., Ohira [/bib_ref]. No significant genotypic association of the 5-HTTLPR polymorphism with depression in a Japanese pediatric population, particularly in patients who have adverse parental environment, was operationalized as the mothers' history of recurrent major depressive disorder. This report is the first of genotypic association of the 5-HTTLPR polymorphism in a Japanese pediatric population. Particularly, this study examined the interactive effect of 5-HTTLPR and history of maternal depression among children and early adolescents with and without current depression. Moreover, few reports in the literature describe studies of G × E interaction in a sample of young people. According to logistic regression analyses, our findings suggest that adversity (maternal depression), gender, and FSIQ had a main effect on current depression. Among them, adversity markedly affected the diagnosis of depression. However interpretation of the effect of FSIQ required careful consideration. A low score of FSIQ might be a cause and an effect [bib_ref] Cognitive improvement associated with tricyclic antidepressant treatment of childhood major depressive illness, Staton [/bib_ref]. More importantly, although G (genotype) × E interaction showed a significant main effect on the diagnosis of depression, it might explain the marked impact of maternal depression mediating the existence of diagnosis. Therefore, no interaction was found between the genotype and environmental factors in this case. Additionally, no significant effect of G × E interaction was found with influences of adversity, gender, FSIQ, and age using multivariable logistic regression analyses. Our findings suggest that depression at an early age was mediated mainly by direct family adversity and gender. These results resemble findings reported previously in longitudinal studies [bib_ref] Adverse childhood experiences and mental health in young adults: a longitudinal survey, Schilling [/bib_ref] [bib_ref] Perceived early-life maternal care and the cortisol response to repeated psychosocial stress, Engert [/bib_ref] [bib_ref] Peer Victimization in Childhood and Internalizing Problems in. A Prospective Longitudinal Study, Zwierzynska [/bib_ref].
Numerous behavioral genetics studies have yielded scientific knowledge related to interaction between genes and the environment [bib_ref] Maternal expressed emotion predicts children's antisocial behavior problems: using monozygotic-twin differences to..., Caspi [/bib_ref]. Most importantly, interactive effects of genes and environmental factors are expected to appear cumulatively over a long period, suggesting that onset of depression results from long-term exposure to stressors. Therefore, an effect of G × E interaction might not be confirmed in this study because the sample comprised children and early adolescents. Our findings from models 1-3 and two-way ANOVA were in concordance with results from prior studies [bib_ref] Interaction between the serotonin transporter gene (5-HTTLPR), stressful life events, and risk..., Risch [/bib_ref] [bib_ref] No association between the serotonin-1A receptor gene single nucleotide polymorphism rs6295C/G and..., Chipman [/bib_ref] [bib_ref] The serotonin transporter gene length polymorphism (5-HTTLPR) and life events: no evidence..., Middeldorp [/bib_ref].
Meta-analyses of the 5-HTTLPR G × E hypothesis have been described in many reports [bib_ref] Genetic sensitivity to the environment: the case of the serotonin transporter gene..., Caspi [/bib_ref] [bib_ref] The serotonin transporter promoter variant (5-HTTLPR), stress, and depression meta-analysis revisited: evidence..., Karg [/bib_ref] [bib_ref] Interaction between the serotonin transporter gene (5-HTTLPR), stressful life events, and risk..., Risch [/bib_ref] [bib_ref] Gene X environment interactions at the serotonin transporter locus, Munafo [/bib_ref] [bib_ref] The moderation by the serotonin transporter gene of environmental adversity in the..., Uher [/bib_ref] [bib_ref] Arguable assumptions, debatable conclusions, Kaufman [/bib_ref]. Sample age and gender composition emerge as important factors [bib_ref] The moderation by the serotonin transporter gene of environmental adversity in the..., Uher [/bib_ref]. Furthermore, the S allele of 5-HTTLPR has been implicated in the pathology of several neuropsychiatric phenotypes including mood and anxiety disorders [bib_ref] A meta-analysis of the association between the serotonin transporter gene polymorphism (5-HTTLPR)..., Schinka [/bib_ref] [bib_ref] Meta-analysis of the association between a serotonin transporter promoter polymorphism (5-HTTLPR) and..., Sen [/bib_ref] and in psychiatric disorders that involve serotonergic dysfunction [bib_ref] Meta-analysis of serotonin transporter polymorphisms and affective disorders, Lotrich [/bib_ref] [bib_ref] Meta-analysis of the association between two polymorphisms in the serotonin transporter gene..., Lasky-Su [/bib_ref]. Furthermore, we reported previously that sensitive periods exist, during which brain structures are most susceptible to exposure to early life events such as childhood maltreatment. These sensitive periods tend to correspond to times of rapid developmental change [bib_ref] Preliminary evidence for sensitive periods in the effect of childhood sexual abuse..., Andersen [/bib_ref] [bib_ref] Childhood sexual abuse is associated with reduced gray matter volume in visual..., Tomoda [/bib_ref]. Therefore, we emphasize that continued monitoring is recommended for children who have early life stress as they pass through puberty. Reasons for the time lag between early life stress and depression are proposed along with potential strategies for early intervention [bib_ref] Length of time between onset of childhood sexual abuse and emergence of..., Teicher [/bib_ref].
Some limitations of this study and its results should be explained. First are the small number of participants and poor statistical power. Second, the covered environmental risk factors in the study were limited. Third, this was not a longitudinal study; no information related to outcomes in late adolescence is available for patients and HC groups. Fourth, the HC group significantly had high IQ than patient group, suggesting difficulty in case control. Fifth, female predominance was found in the patient group (Fisher exact, P = 0.008), which might influence the findings. Given these limitations, it will be important for future studies to ensure an adequate range of environmental variables and populations studied because a restricted range can strongly influence G × E results. Furthermore, to tease apart and thereby discern the respective influences of G × E, G × G, and E × E interactions, employing a molecular framework to extend this research would be beneficial. A follow-up study using both molecular genetics methods and measures of proximal environments is expected to provide a useful extension of the current study.
# Conclusions
In a Japanese pediatric sample, we examined whether adverse parental environment, operationalized as a mother's history of recurrent major depressive disorder, interacts with 5-HTTLPR polymorphism to predict a patient's symptoms of depression. Data obtained from children and adolescents who had been carefully diagnosed categorically and data from age-matched controls were analyzed using logistic regression to assess G × E interaction. Despite an equivocal interaction effect, adversity and gender showed a significant main effect.
[table] Table 1: Demographics, early adversity, and clinical characteristics in individuals in patients and healthy controls * ANCOVA with age and gender as covariates. Values are presented as the mean and SD. DSRS-C, Birleson Depression Self-Rating Scale for Children. [/table]
[table] Table 2 5: -HTTLPR genotype distributions and allele frequencies in controls and depressive patients [/table]
[table] Table 3: Binary logistic regression analyses for variables of genotype, adversity, gender, age, SES, FSIQ, and Genotype (G) × Environment (E) [/table]
[table] Table 4: Multivariable logistic regression analyses using stepwise procedure [/table]
[table] Table 5 Two: Dependent variable is Birleson score. R2 = 0.98 (adjusted R2 = 0.10). note: R = coefficient of determination, age and gender as covariates. [/table]
|
Spatio-Temporal Epidemiology of Human West Nile Virus Disease in South Dakota
Despite a cold temperate climate and low human population density, the Northern Great Plains has become a persistent hot spot for human West Nile virus (WNV) disease in North America. Understanding the spatial and temporal patterns of WNV can provide insights into the epidemiological and ecological factors that influence disease emergence and persistence. We analyzed the 1,962 cases of human WNV disease that occurred in South Dakota from 2002-2012 to identify the geographic distribution, seasonal cycles, and interannual variability of disease risk. The geographic and seasonal patterns of WNV have changed since the invasion and initial epidemic in 2002-2003, with cases shifting toward the eastern portion of South Dakota and occurring earlier in the transmission season in more recent years. WNV cases were temporally autocorrelated at lags of up to six weeks and early season cumulative case numbers were correlated with seasonal totals, indicating the possibility of using these data for short-term early detection of outbreaks. Epidemiological data are likely to be most effective for early warning of WNV virus outbreaks if they are integrated with entomological surveillance and environmental monitoring to leverage the strengths and minimize the weaknesses of each information source.
# Introduction
Emerging infectious diseases, including those that have appeared for the first time, rapidly increased in incidence, or expanded into new geographic areas, are a significant concern in the fields of human and veterinary medicine as well as wildlife conservation [bib_ref] Wildlife ecology-Emerging infectious diseases of wildlife-Threats to biodiversity and human health, Daszak [/bib_ref] [bib_ref] Diseases of humans and their domestic mammals: Pathogen characteristics, host range and..., Cleaveland [/bib_ref]. Disease emergence is often connected with the development of human societies and their interactions with the environment, including the proliferation of transportation networks that facilitate pathogen spread, land use and climate change that affect habitats for arthropod vectors, and animal hosts, and population movements that increase human and domesticated animal contact with wildlife and their pathogens. The successful invasion of North America by West Nile virus (WNV) is a well-known example of infectious disease emergence that was initially caused by long-distance pathogen transport and facilitated by a diversity of suitable environments for transmission [bib_ref] Epidemiology and transmission dynamics of West Nile Virus disease, Hayes [/bib_ref]. WNV is an arbovirus that is maintained in an enzootic cycle with mosquitoes as the primary vectors and birds as the primary reservoir hosts. Humans and horses are incidental hosts; both species can acquire the virus from infected mosquitoes but do not transmit the virus to uninfected mosquitoes and are not necessary for the maintenance of the pathogen.
WNV is indigenous to Africa, Asia, Europe, and Australia and was first identified in North America in the New York City metropolitan area during the summer of 1999 [bib_ref] Epidemiology and transmission dynamics of West Nile Virus disease, Hayes [/bib_ref]. In 2002 widespread WNV cases occurred in the Midwest and south-central states, and the first WNV cases were reported in South Dakota. The epidemic wave moved further west in 2003, with extremely high incidence rates throughout the Great Plains and cases occurring as far west as California. From 2004 onward, WNV has been endemic across much of the western hemisphere with sporadic, localized outbreaks of human disease in a variety of geographic locations [bib_ref] Nile virus: Review of the literature, Petersen [/bib_ref]. Human WNV disease has had a significant public health impact in the United States. A total of 37,088 WNV cases have been reported to the Centers for Disease Control through 2012, including [bib_ref] Landscape-level spatial patterns of West Nile virus risk in the northern Great..., Chuang [/bib_ref] cases of severe neuroinvasive disease and 1,549 deaths. The widespread resurgence of human WNV disease in 2012 following several years of relatively low incidence has highlighted the continued public health hazard posed by WNV and emphasized the need for more accurate predictions of when and where WNV outbreaks will occur.
Disease mapping and exploratory spatial analysis of epidemiological data can help to identify the geographic locations of populations at risk and facilitate the development of hypotheses about the epidemiological and ecological processes that drive these patterns [bib_ref] Geographical epidemiology, spatial analysis and geographical information systems: A multidisciplinary glossary, Rezaeian [/bib_ref]. Time-series analysis can similarly elucidate seasonal cycles, longer-term trends, and deviations from these expected patterns that indicate the times when WNV risk is highest and suggest new hypotheses about the drivers of temporal variability at different scales [bib_ref] On time series analysis of public health and biomedical data, Zeger [/bib_ref] [bib_ref] Syndromic surveillance: STL for modeling, visualizing, and monitoring disease counts, Hafen [/bib_ref]. In particular, early detection of anomalously high human case numbers, particularly during the early part of the WNV season, may serve as a harbinger of developing outbreaks [bib_ref] Detection of epidemics in their early stage through infectious disease surveillance, Hashimoto [/bib_ref]. There are a number of limitations to inferring spatial and temporal patterns of disease risk from human disease cases, particularly in rural areas with low human densities and in situations where time lags in disease case reporting and confirmation delay the availability of these data. However, human disease surveillance can still provide valuable information that complements other sources of information from mosquito surveillance [bib_ref] Seasonal patterns for entomological measures of risk for exposure to Culex vectors..., Bolling [/bib_ref] , dead bird surveillance [bib_ref] Dead bird clusters as an early warning system for West Nile virus..., Mostashari [/bib_ref] , and ecological forecasting based on environmental monitoring data [bib_ref] Local impact of temperature and precipitation on West Nile virus infection in..., Ruiz [/bib_ref].
Since its arrival in 2002, WNV has had a significant public health impact in South Dakota and neighboring Great Plains states. From 1999-2008 South Dakota had the highest average incidence of WNV neuroinvasive disease in the United States [bib_ref] Surveillance for human West Nile virus disease-United States, Lindsey [/bib_ref]. During 2012, a year in which WNV reemerged at a national level, South Dakota once again had the highest incidence rates of both neuroinvasive WNV disease (7.5/100,000), total WNV disease (24.6/100,000 including WNV fever in addition to WNV neuroinvasive disease), and viremic blood donors (5.1/100,000). The high and persistent incidence of WNV in South Dakota and the surrounding region has been hypothesized to reflect the geographic distribution of Culex tarsalis, a highly ornithophilic and particularly efficient amplifying vector of WNV that can also serve as a bridge vector to humans [bib_ref] Ecological niche of the 2003 West Nile virus epidemic in the northern..., Wimberly [/bib_ref]. Previous WNV research in South Dakota found that Culex tarsalis abundance was lowest in urban areas and highest in grass-dominated rural habitats [bib_ref] Weather and land cover influences on mosquito populations in Sioux Falls, South..., Chuang [/bib_ref] , and the incidence of human WNV disease was similarly highest in rural areas with poorly drained soils [bib_ref] Landscape-level spatial patterns of West Nile virus risk in the northern Great..., Chuang [/bib_ref]. Seasonal and interannual variability in Culex tarsalis abundance in South Dakota exhibited much stronger lagged relationships with temperature than with precipitation [bib_ref] Weather and land cover influences on mosquito populations in Sioux Falls, South..., Chuang [/bib_ref] [bib_ref] Satellite microwave remote sensing for environmental modeling of mosquito population dynamics, Chuang [/bib_ref] , and a high number of accumulated degree days during the spring and early summer was found to be an indicator of the risk of human WNV disease during the subsequent main transmission season in mid to late summer [bib_ref] Remote sensing of climatic anomalies and West Nile virus incidence in the..., Chuang [/bib_ref].
The overarching goal of our study was to extend this previous work by characterizing the spatio-temporal epidemiology of WNV disease in South Dakota from 2002-2012. The underlying rationale was the expectation that identifying spatial and temporal patterns of human disease occurrence will provide a clearer understanding of when and where WNV risk is the highest and whether these patterns have changed since the emergence of WNV in South Dakota. We addressed the following specific questions: (1) Where in South Dakota is WNV risk highest and has this geographic pattern changed over time? [bib_ref] Diseases of humans and their domestic mammals: Pathogen characteristics, host range and..., Cleaveland [/bib_ref] During what part of the year is WNV risk highest and has this seasonal pattern changed over time? [bib_ref] Epidemiology and transmission dynamics of West Nile Virus disease, Hayes [/bib_ref] What are the temporal patterns (deviations from expected seasonal cycles and long-term trends) of WNV anomalies and can they be used to predict future WNV risk? (4) Do WNV epidemics arise simultaneously throughout the state or do cases increase earlier in some geographic locations than in others?
# Methods
The study area encompassed the state of South Dakota, which is located in the northern Great Plains of the United States, a region historically dominated by prairie vegetation. Environmental variability across the state is strongly affected by precipitation, which generally decreases along an east-to-west gradient [fig_ref] Figure 1: Environmental variability across South Dakota [/fig_ref]. This precipitation gradient is reflected in the distribution of wetlands, which are highest in the eastern portions of the state [fig_ref] Figure 1: Environmental variability across South Dakota [/fig_ref]. Land cover patterns also vary along this gradient, ranging from a mixture of row crops, pasture, hay, and grasslands in eastern South Dakota to predominantly rangelands in the drier western region [fig_ref] Figure 2: Dot density maps of WNV cases in South Dakota from 2002-2012 [/fig_ref] Epidemiological data on human WNV disease cases were collected by the South Dakota Department of Health. Each case was identified by disease type (WNV fever or neuroinvasive disease), referenced temporally by the reported date of disease onset, and referenced spatially by county, ZIP code, and city of residence. Each case was georeferenced using the 2010 map of ZIP code tabulation areas (ZCTAs) from the U.S. Census Bureau. A total of 23 cases (1.1% of the total number) were missing ZIP codes and were assigned to a ZCTA based on the county and city of residence. In addition, 60 cases (2.8% of the total number) had ZIP codes that were not present in the 2010 ZCTA map and were assigned to a 2010 ZCTA by referencing the 2000 ZCTA map or based on the county and city of residence. The 2010 population data for each ZCTA were also obtained from the U.S. Census Bureau. ZIP codes were developed by the U.S. Postal Service to classify street segments, address ranges and delivery points for mail delivery. ZCTAs were created by the U.S. Census Bureau as spatial units to provide approximate mapping of ZIP code boundaries using aggregated Census blocks. However, it is difficult to map ZIP code boundaries precisely, and ZIP codes are frequently split, discontinued, added, or expanded [bib_ref] On the use of ZIP codes and ZIP code tabulation areas (ZCTAs)..., Grubesic [/bib_ref]. As a result, some ZIP codes may not have a corresponding ZCTA and some addresses may actually be located in different ZCTAs relative to their ZIP code. Despite these limitations, ZIP code level data has the advantages of greater spatial precision than coarser county-level summaries and higher rates of geocoding success than street address-level geocoding [bib_ref] Need for improved methods to collect and present spatial epidemiologic data for..., Eisen [/bib_ref] [bib_ref] Geographic variability in geocoding success for West Nile virus cases in South..., Wey [/bib_ref]. The 2010 ZCTAs in South Dakota had a mean area of 519 km 2 and a mean population of 2,115. The mean population density at the ZCTA level was 34 per km 2 , but individual ZCTAs ranged from 0.05 to 4,628 per km 2 reflecting the variety of settlement patterns from dense cities to rural areas.
Spatial patterns of WNV cases were displayed as dot density maps by selecting a random location for each case within its assigned ZCTA. Spatial patterns of WNV risk were mapped using annual incidence rates and were spatially smoothed by incorporating information from surrounding areas to reduce variability caused by small population sizes at the ZCTA level. For this initial assessment of WNV risk, smoothing was carried out using a simple population-weighted mean smoothing approach as suggested by Waller and Gotway. To generate the smoothed maps, a neighborhood mean of WNV cases was calculated for each ZCTA as the mean number of cases from the focal ZCTA and all ZCTAs with centroids falling within a 75 mile radius of the focal ZCTA. A neighborhood mean of the 2010 population was calculated for each ZCTA using the same method. The annual incidence rate, normalized to a population of 100,000, was calculated for each ZCTA using these smoothed values. To explore changes in the spatial pattern of WNV risk over time, smoothed maps were generated for each of three time periods: . These time periods were subjectively selected to capture three distinctive epochs of WNV transmission in South Dakota: the initial invasion and first large epidemic , a subsequent period of relatively high WNV incidence , and a period of relatively low WNV incidence followed by the resurgence of human WNV cases in 2012 (2008-2012). Separate sets of maps were generated for the total incidence of human WNV disease (including both WNV fever and neuroinvasive disease) and for only WNV neuroinvasive disease.
Seasonal patterns of total human WNV cases were summarized based on a standardized 18-week season in which all cases prior to 1 June were included in week 1; the months of June (weeks 2-5), July (weeks 6-9), August (weeks 10-13), and September (weeks 14-17) were each divided into four weeks; and all cases after September 30th were included in week 18. Weekly cumulative totals of human WNV cases were graphed for each year and for the three time periods used in the mapping . Two-sample Kolmogorov-Smirnov tests were used to test for differences in the seasonal distribution of WNV cases among these three time periods. These tests were carried out using all human WNV cases (including both WNV fever and neuroinvasive disease), only the WNV fever cases, and only the WNV neuroinvasive disease cases.
The 18 weeks of WNV data for each year were concatenated to generate a continuous time series of WNV cases from 2002-2012. This log-transformed time series was then decomposed into seasonal, trend, and remainder components with the seasonal and trend decomposition using loess (STL) procedure [bib_ref] STL: A seasonal-trend decomposition procedure based on loess, Cleveland [/bib_ref]. In this approach, the seasonal component was calculated as the mean number of cases in each week across all years and the long-term trend was estimated using a smoothed locally-weighted regression. The remainder component reflected anomalies of WNV cases relative to these longer term patterns, and was computed by subtracting the seasonal and long-term trends from the raw data. Temporal autocorrelation was assessed by computing autocorrelation functions at time lags ranging from 1 to 18 weeks for both the raw data and the remainders from the STL decomposition.
WNV cases were also analyzed to determine whether the numbers of cases occurring during the early part of the WNV season (June and July) were predictive of the total number of WNV cases occurring throughout the entire season. For this analysis, we summarized the cumulative total number of WNV cases occurring in each year from week 2 (the first week in June) through week 9 (the last week in July). We computed the Spearman rank correlation between each of these cumulative variables and the total number of cases occurring for 2004-2012 (N = 9). We examined the changes in these correlations across the different weeks to determine the point during the season when WNV cases provided a reliable indicator of the total magnitude of WNV cases throughout the season.
We conducted an analysis of space-time interactions during major WNV outbreak years to determine whether there were significant changes in the geographic locations of WNV cases throughout the season. To have sufficient data for the analysis, we focused on major outbreak years during which more than 200 WNV cases occurred in South Dakota . For each of these years, we used a generalized additive model (GAM) to model the day of year (DOY) of each case as the dependent variable using a smoothed functions of the spatial coordinates (eastings and northings) of the ZCTA centroids as independent variables. Statistically significant relationships between DOY and one or both of the independent variables indicated that different geographic locations had different mean dates of WNV onset. For years where there was significant spatial variability in mean DOY, the models were applied across the entire state to produce a continuous map of the mean DOY of WNV cases for each ZCTA.
# Results
The dot density maps emphasized the strong spatial and temporal heterogeneity of human WNV cases within South Dakota [fig_ref] Figure 2: Dot density maps of WNV cases in South Dakota from 2002-2012 [/fig_ref] , [fig_ref] Table 1: Annual cases of human West Nile virus disease in South Dakota [/fig_ref]. During 2002, the first year of WNV occurrence in South Dakota, a relatively small number of WNV cases were concentrated in the eastern portion of the state. In 2003 there was a major increase in both WNV fever and neuroinvasive disease, with several significant clusters occurring in the central and western parts of South Dakota. WNV cases declined in 2004, followed by a three year period (2005-2007) of increased WNV cases. Annual numbers of WNV cases declined steadily over the next four years (2008-2011), reaching their lowest levels in 2011 when only two cases of WNV fever were reported. In 2012, there was a significant resurgence of WNV cases in South Dakota, with a stronger concentration of these cases in the eastern part of the state relative to previous years. Annual, log-transformed fever case counts and neuroinvasive case counts were strongly correlated at the state level (r = 0.96, p < 0.001). [fig_ref] Table 1: Annual cases of human West Nile virus disease in South Dakota [/fig_ref] These patterns were reflected in the seasonal component of the WNV time series decomposition, which showed a strong seasonal peak between the last week of July and the third week of August [fig_ref] Figure 6: Decomposition of the log-transformed 2004-2012 WNV time series into seasonal, trend, and... [/fig_ref]. The smoothed trend component illustrated longer-term variation in human WNV cases, with highest case numbers from 2003 through 2006, declining case numbers after 2007, and an increase driven by the 2012 outbreak. The raw, log-transformed WNV case data exhibited positive temporal autocorrelation at lags from 1 to 4 weeks and from 13 to 18 weeks [fig_ref] Figure 7: Temporal autocorrelation function for [/fig_ref] , reflecting the previously-described seasonal pattern of WNV. After removing the seasonal patterns and long-term trend from the data, the remainder exhibited positive temporal autocorrelation at lags from 1 to 6 weeks [fig_ref] Figure 7: Temporal autocorrelation function for [/fig_ref]. This pattern suggested that anomalously high numbers of WNV cases can serve as indicators of continued high WNV risk up to 6 weeks in the future. Cumulative case counts through the first and second weeks of June were not significantly correlated with annual case counts at an alpha-level of 0.05 [fig_ref] Table 2: Spearman rank correlations between cumulative case numbers through July and August and... [/fig_ref]. However, cumulative case counts from the third week of June were significantly correlated with annual case counts, and the magnitude of this correlation generally increased as more cases were accumulated later into the season. There was statistically significant variability in the onset date of WNV cases in both the north-south and east-west directions in 2003, and in the east-west direction in 2005 and 2007 [fig_ref] Table 3: Results of generalized additive models predicting the day of year of WNV... [/fig_ref]. There were no statistically significant spatial trends in the onset date of WNV cases in 2012. In 2003 and 2005, cases tended to occur earlier in the western part of the state and later in the season in the eastern part of the state [fig_ref] Figure 8: Smoothed maps of the mean day of year of WNV cases for... [/fig_ref]. The reverse pattern occurred in 2007; cases tended to occur earlier in the eastern part of the state and later in the western part of the state. The adjusted R 2 values of the GAMs were all low (<0.08), indicating that these space-time patterns were relatively weak.
# Discussion
There has been a shift in the spatio-temporal niche of human WNV risk in South Dakota from the initial invasion and first [bib_ref] Ecological niche of the 2003 West Nile virus epidemic in the northern..., Wimberly [/bib_ref]. This cluster occupied a climatic niche characterized by higher temperatures and intermediate precipitation, and was hypothesized to reflect the geographic distribution of Culex tarsalis, a particularly efficient WNV vector. Subsequent research showed that interannual variability in temperature and moisture during the spring and early summer was associated with spatial and temporal variability in WNV incidence across the region from 2004-2011 [bib_ref] Remote sensing of climatic anomalies and West Nile virus incidence in the..., Chuang [/bib_ref]. Thus, anomalies in the regional patterns of temperature and precipitation in 2002 and 2003 relative to later years may explain some of the differences in geographic pattern and seasonality. Alternately, changes in geographic and seasonal patterns of WNV cases after 2003 may reflect other ecological and epidemiological factors, such as the development of immunity in both human and avian populations and consequent impacts on virus amplification in avian communities and viral transmission to humans. In addition, significant and persistent declines in the abundances of multiple bird species resulting from WNV-caused mortality have been documented across the United States [bib_ref] Ecology of West Nile virus transmission and its impact on birds in..., Kilpatrick [/bib_ref] [bib_ref] Nile virus emergence and large-scale declines of North American bird populations, Ladeau [/bib_ref]. Resulting changes in bird community composition across South Dakota have the potential to influence the spatial pattern and timing of virus amplification and risk to humans, but have been unexplored to date and merit further investigation.
From 2004 onward, and particularly in 2008-2012, one of the regions of highest WNV risk within South Dakota was concentrated in eastern South Dakota within the James River Valley, which lies between the uplands of the Missouri Couteau to the West and the Prairie Couteau to the East. This broad lowland was created by the James Lobe of the Laurentide ice sheet and occupies approximately 49,000 km 2 . It is characterized by a high density of seasonal and permanent wetlands and a landscape mosaic of row crops and grasslands [fig_ref] Figure 1: Environmental variability across South Dakota [/fig_ref] , and is currently being impacted by land use changes that are resulting in widespread wetland drainage and conversion of grasslands to row crop agriculture [bib_ref] Recent land use change in the Western Corn Belt threatens grasslands and..., Wright [/bib_ref]. Previous studies have highlighted spatial clustering and climatic associations of human WNV cases at national and regional levels [bib_ref] Ecological niche of the 2003 West Nile virus epidemic in the northern..., Wimberly [/bib_ref] [bib_ref] Spatio-temporal cluster analysis of county-based human West Nile virus incidence in the..., Sugumaran [/bib_ref] and finer-scale associations of human WNV disease with land cover, land use, soils, and hydrology at the landscape level [bib_ref] Landscape-level spatial patterns of West Nile virus risk in the northern Great..., Chuang [/bib_ref] [bib_ref] Association of West Nile virus illness and urban landscapes in Chicago and..., Ruiz [/bib_ref] [bib_ref] Factors affecting the geographic distribution of West Nile virus in, Gibbs [/bib_ref]. The geographic patterns highlighted in this study further suggest that broad physiographic features such as the James River Valley can influence WNV risk to humans through their characteristic hydrological processes, land use patterns, and consequent effects on habitat for vector and host populations. More research is needed to better elucidate the proximal determinants of the geographic distribution WNV risk at the state level and to apply this knowledge to improve strategies for WNV prevention and control.
Early detection systems are widely used to identify the onset of mosquito-borne disease outbreaks by tracking human disease cases. Forecasting techniques range from relatively simple methods based on thresholds determined from historical data, to more sophisticated statistical methods for time series forecasting [bib_ref] Remote sensing-based time series models for malaria early warning in the highlands..., Midekisa [/bib_ref]. Our results suggest that anomalously high numbers of WNV cases, relative to seasonal patterns and longer-term trends, provide an indicator of continued high numbers of WNV cases up to 6 weeks in the future. Alternately, the cumulative number of WNV cases through the end of June could be used to make a seasonal forecast of the total number of expected cases during the peak of the WNV season in July and August. Although statistically significant spatio-temporal patterns were detected in the major outbreak years, the relationships were relatively weak and variable from year to year. As a result, there currently appears to be a relatively low potential for using human case data to predict the precise locations of WNV risk within South Dakota at specific times during the WNV season.
Two major limitations to the use of human surveillance data for forecasting disease outbreaks are data quality and data availability. For human WNV disease in the United States, all cases are assessed based on a consistent definition established by the Centers for Disease Control and are subject to laboratory confirmation. However, an unknown degree of spatial uncertainty is introduced because some patients may contract WNV at a location away their residence. In addition, the delay introduced as a result of the time lags between amplification of the virus in mosquito and bird populations, transmission to humans, onset of symptoms, medical diagnosis, and case confirmation mean that human cases are a lagging indicator of WNV risk. As result, WNV amplification will have already occurred and mosquito numbers and infection rates will likely already be high by the time WNV is detected in the human population. In a state like South Dakota that experiences human WNV cases every year, detection of anomalies in early season case data does have some potential to predict the magnitude of WNV in human populations during the peak transmission season. In particular, our results suggest that human surveillance data may be able to indicate the onset of a significant WNV outbreak with several weeks of lead time. This type of alert would not provide enough time to take action to prevent virus amplification, but it might still allow public health officials to take steps such as expanding adult mosquito control and issuing preventive warnings before the period of peak transmission to humans.
One promising approach toward leveraging the strengths and minimizing the limitation of human case surveillance for disease outbreak detection is to use blended approaches that integrate multiple streams of surveillance data. For example, a study of WNV in Colorado combined environmental data, entomological data, and human WNV case data to develop an improved map of disease risk in the Northern Colorado Front Range [bib_ref] Combining mosquito vector and human disease data for improved assessment of spatial..., Winters [/bib_ref]. Research on malaria early warning in the highlands of Ethiopia has found that forecasting models that incorporate both remotely-sensed climatic variables and lagged indicators from human case surveillance are more effective at predicting epidemics than models based on either environmental data alone or human cases alone [bib_ref] Remote sensing-based time series models for malaria early warning in the highlands..., Midekisa [/bib_ref] [bib_ref] A Computer System for Forecasting Malaria Epidemic Risk Using Remotely-Sensed Environmental Data, Wimberly [/bib_ref]. The California Mosquito-Borne Virus Surveillance and Response Plan incorporates multiple sources of information, including meteorological data, vector abundance, vector infection rates, sentinel chicken seroconversion, dead bird reporting and testing, and human surveillance into a single index [bib_ref] California state mosquito-borne virus surveillance and response plan: A retrospective evaluation using..., Barker [/bib_ref] [bib_ref] Comparison of enzootic risk measures for predicting West Nile disease, Kwan [/bib_ref]. This multifactor index was found to be a better early warning indicator for the onset of human WNV disease than predictions based on single sources of information [bib_ref] Comparison of enzootic risk measures for predicting West Nile disease, Kwan [/bib_ref]. Previous research in South Dakota has found influences of climatic variability on the abundance of vector mosquitoes [bib_ref] Weather and land cover influences on mosquito populations in Sioux Falls, South..., Chuang [/bib_ref] [bib_ref] Satellite microwave remote sensing for environmental modeling of mosquito population dynamics, Chuang [/bib_ref] and human disease cases [bib_ref] Remote sensing of climatic anomalies and West Nile virus incidence in the..., Chuang [/bib_ref] and we suggest that combining these types of environmental indicators with mosquito and human case surveillance data offers a strong potential for improving our ability to predict the spatial and temporal dimensions of WNV in South Dakota as well as in other regions where WNV is endemic.
# Conclusions
Human WNV disease exhibits strong patterns of geographic variability, seasonal cycles, and interannual fluctuations in South Dakota. The geographic and seasonal patterns of WNV have changed since the invasion and initial epidemic in 2002-2003, with cases shifting toward the eastern portion of South Dakota and occurring earlier in the transmission season in more recent years. The underlying cause of these shifts is currently unknown, but the results suggest that more recent data on endemic WNV from 2004 onward should be used for risk mapping and assessment of seasonality for public health planning purposes. Temporal analyses indicate that there is a potential for forecasting upcoming large WNV outbreaks with lead times of up to several weeks. Although this time frame is likely insufficient for preventing virus amplification, it could still help reduce transmission to humans by providing time to ramp up emergency mosquito control and disseminate messages to the public to take personal protection measures. Ultimately, information from epidemiological surveillance should be integrated with environmental monitoring, entomological surveillance, and other sources of information to provide more effective predictions of the likelihood of future WNV outbreaks.
[fig] Figure 1: Environmental variability across South Dakota. (a) 1982-2011 mean monthly precipitation (mm) from May-September summarized at the county level (b) Emergent wetlands from the National Wetlands Inventory summarized as proportion of total county area, and (c) 2006 map of major land cover classes from the National [/fig]
[fig] Figure 2: Dot density maps of WNV cases in South Dakota from 2002-2012. (a-k) Each dot represents a single case within the ZIP code tabulation area (ZCTA) of the patient's residence. Red dots represent WNV neuroinvasive disease and blue dots represent WNV fever. (l) Brown dots represent total population with one dot per 100 people. Dots were assigned random locations within the ZCTAs for the purpose of visualization. [/fig]
[fig] Figure 3: Spatially smoothed incidence rates of all human WNV disease (including both fever and neuroinvasive disease) in South Dakota for three time periods. Rates are the annual number of cases per 100,000 population. [/fig]
[fig] Figure 4: Spatially smoothed incidence rates of human WNV neuroinvasive disease in South Dakota for three time periods. Rates are the annual number of cases per 100,000 population. [/fig]
[fig] Figure 5: Weekly cumulative distributions of illness onsets of all human WNV cases (including both fever and neuroinvasive disease) in South Dakota from 2002-2012. (a) Annual curves, (b) Combined curves for three time periods. [/fig]
[fig] Figure 6: Decomposition of the log-transformed 2004-2012 WNV time series into seasonal, trend, and remainder components using the STL procedure. Each "year" in the time series consists of an 18-week WNV season extending from the last week in May through the first week in October. [/fig]
[fig] Figure 7: Temporal autocorrelation function for (a) the raw, log-transformed time series of WNV cases and (b) the remainders from the time series decomposition. Blue lines represent critical values for statistical significance of the autocorrelation statistic at an alpha-level of 0.05. [/fig]
[fig] Figure 8: Smoothed maps of the mean day of year of WNV cases for 2003, 2005, and 2007. Maps were generated by using generalized additive models to predict the DOY as a smoothed function of the spatial coordinates of the ZCTAs. [/fig]
[table] Table 1: Annual cases of human West Nile virus disease in South Dakota. [/table]
[table] Table 2: Spearman rank correlations between cumulative case numbers through July and August and total annual case numbers from 2004-2012 (N = 9). [/table]
[table] Table 3: Results of generalized additive models predicting the day of year of WNV case onset as a function of the eastings and northings of their respective ZCTAs. [/table]
|
Changing treatment paradigms for the management of inflammatory bowel disease
# Introduction
Inflammatory bowel disease (IBD) including ulcerative colitis (UC) and Crohn's disease (CD) is a chronic and progressive inflammatory condition of the gastrointestinal tract. IBD is known to be associated with a genetic predisposition to the disease, environmental exposure, and the subsequent dysregulated immune responses [bib_ref] Navigating the microbial basis of inflammatory bowel diseases: seeing the light at..., Raffals [/bib_ref]. The number of people diagnosed with IBD continues to increase and is the highest in Western countries; however, a significantly increased incidence is being observed in the Asian population secondary to rapid urbanization [bib_ref] Understanding and preventing the global increase of inflammatory bowel disease, Kaplan [/bib_ref] [bib_ref] Changing epidemiological trends of inflammatory bowel disease in Asia, Ng [/bib_ref].
Although there has been much progress in the management of IBD with established and evolving therapies including the use of monoclonal antibodies, most current approaches that are directed only toward symptom control have failed to change the natural course of the Inf lammatory bowel disease (IBD) is a chronic and progressive inf lammatory condition of the gastrointestinal tract causing bowel damage, hospitalizations, surgeries, and disability. Although there has been much progress in the management of IBD with established and evolving therapies, most current approaches have failed to change the natural course. Therefore, the treatment approach and follow-up of patients with IBD have undergone a significant change. Usage of immunosuppressants and/or biologics early during the course of the disease, known as top-down or accelerated step-up approach, was shown to be superior to conventional management in patients who had been recently diagnosed with IBD. This approach can be applied to selected groups based on prognostic factors to control disease activity and prevent progressive disease. Therapeutic targets have been shifted from clinical remission mainly based on symptoms to objective parameters such as endoscopic healing due to the discrepancies observed between symptoms, objectively evaluated inf lammatory activity, and intestinal damage. The concept of treat-to-target in IBD has been supported by population-based cohort studies, post hoc analysis of clinical trials, and meta-analysis, but more evidence is needed to support this concept to be applied to the clinical practice. In addition, individualized approach with tight monitoring of non-invasive biomarker such as C-reactive protein and fecal calprotectin and drug concentration has shown to improve clinical and endoscopic outcomes. An appropriate de-escalation strategy is considered based on patient demographics, disease features, current disease status, and patients' preferences.
www.kjim.org https://doi.org/10.3904/kjim.2017.400 disease [bib_ref] Long-term complications, extraintestinal manifestations, and mortality in adult Crohn's disease in population-based..., Peyrin-Biroulet [/bib_ref] [bib_ref] Ulcerative colitis as a progressive disease: the forgotten evidence, Torres [/bib_ref]. A population-based study has shown that the probability of CD-related surgery was as high as 96.7% in anti-Saccharomyces cerevisiae antibody-positive patients aged < 20 years who had been diagnosed with penetrating disease and had been initially treated with systemic steroids [bib_ref] Risk matrix for prediction of advanced disease in a population-based study of..., Solberg [/bib_ref]. Additionally, disability caused by IBD correlates with disease severity, and it has been observed that improved medication compliance is associated with lower disability and a better quality of life [bib_ref] Disability due to inflammatory bowel disease is correlated with drug compliance, disease..., Yoon [/bib_ref]. The heterogeneity in clinical features and prognosis makes it difficult to use generally effective strategies for the treatment of this condition, and consequently contributes to a wide variation in strategies used for IBD management [bib_ref] Quality of care in inflammatory bowel disease in Asia: the results of..., Song [/bib_ref].
Recently, the treatment approach and follow-up of patients with IBD have undergone a significant change [fig_ref] Table 1: Concept change of inflammatory bowel disease [/fig_ref]. It is now widely acknowledged that IBD is a progressive disease causing intestinal damage and disability, which therefore mandates the institution of prompt individualized treatment during the early stages to prevent irreversible complications and produce better outcomes with reduced rates of hospitalization and surgery [bib_ref] Management strategies to improve outcomes of patients with inflammatory bowel diseases, Colombel [/bib_ref]. This review encompasses the emerging trends in the treatment paradigm of IBD including prognostication and early intervention, treating to target, tight monitoring, and adoption of individualized therapeutic approaches [fig_ref] Figure 1: Current strategies for the management of inflammatory bowel disease [/fig_ref].
## Early intervention based on prognos-tic factors
Recently, top-down or accelerated step-up treatment strategies have been introduced to modify the natural course of IBD aimed at achieving better outcomes. There exists a window of opportunity in the early stages of IBD, particularly in CD patients to reduce bowel damage, hospitalizations, surgeries, and disability as has been observed in patients diagnosed with rheumatoid arthritis [bib_ref] Management strategies to improve outcomes of patients with inflammatory bowel diseases, Colombel [/bib_ref]. Administration of immunosuppressants and/or biologics early during the course of the disease was shown to be superior to conventional management in patients who had been recently diagnosed with IBD in terms of better mucosal healing (MH), induction of steroid-free remission, and prevention of hospitalization [bib_ref] Early combined immunosuppression or conventional management in patients with newly diagnosed Crohn's..., D'haens [/bib_ref] [bib_ref] Early mucosal healing with infliximab is associated with improved long-term clinical outcomes..., Colombel [/bib_ref] [bib_ref] Randomised clinical trial: deep remission in biologic and immunomodulator naive patients with..., Colombel [/bib_ref].
A hospital-based CD cohort study in Korea showed an intestinal resection rate that matched the rate observed in Western countries, and the surgical rate was observed to decrease significantly over 30 years following the early usage of thiopurines [bib_ref] Long-term prognosis of Crohn's disease and its temporal change between 1981 and..., Park [/bib_ref]. This trend of a decreased colectomy rate in Korean UC patients was associated with more frequent and earlier use of thiopurines and anti-tumor necrosis factor (TNF) agents over time [bib_ref] Long-term prognosis of ulcerative colitis and its temporal change between 1977 and..., Lee [/bib_ref].
Despite these benefits, it is unclear whether such aggressive treatment is better than conventional step-up therapy for the prevention of adverse outcomes. Additionally, a significant percentage of patients showed clinical remission over 5 years and remained free from complications after 10 years of diagnosis even during the pre-biologic era [bib_ref] Clinical course in Crohn's disease: results of a Norwegian population-based ten-year follow-up..., Solberg [/bib_ref]. Moreover, drug-related adverse events, high cost, and issues related to generalizability and applicability to clinical practice are concerns associated with the adoption of the top-down or accelerated step-up approach in routine clinical practice [bib_ref] Early combined immunosuppression for the management of Crohn's disease (REACT): a cluster..., Khanna [/bib_ref]. The Randomised Evaluation of an Algorithm for Crohn's Treatment (REACT) study was an open-label cluster-randomized controlled trial that compared the efficacy and safety of early combined immunosuppression (ECI) with the use of anti-TNF agents and antimetabolites versus conventional management of CD among gastroenterology practices in the community. Although ECI did not show greater benefit in terms of clinical remission than that observed with conventional management, ECI reduced major adverse events such as the need for surgery, hospital admissions, or serious CD-associated complications. ECI did not increase the risk of serious treatment-related adverse events or mortality. Despite the limitations of an un-blinded study design and assessment of outcome measures based on evaluation of subjective symptoms, these results suggest that highly effective therapy initiated early in the course of the disease might change the natural course of CD without a significant increase in drug-related risks [bib_ref] Early combined immunosuppression for the management of Crohn's disease (REACT): a cluster..., Khanna [/bib_ref].
The effects of early administration of anti-TNF/immunomodulator (IM) therapy were retrospectively evaluated in Korean CD patients presenting with ≥ 2 poor prognostic factors. Poor prognostic factors were: age < 40 years at the time of diagnosis, early use of systemic steroid, and the presence of a perianal fistula at the time of diagnosis. The early treatment group was defined as patients who were administered anti-TNF and/or IM treatment within 2 years of CD diagnosis. The early treatment group presenting with poor prognostic factors demonstrated better outcomes, which were indicated by a greater time interval until the need for intestinal surgery and fewer complications than those observed in the late treatment group [bib_ref] Early anti-TNF/immunomodulator therapy is associated with better long-term clinical outcomes in Asian..., Oh [/bib_ref].
Prognostic predictors of colectomy in patients presenting with UC are: extensive disease, proximal extension of lesions during the disease course, extra-intestinal manifestations, younger age at the time of disease onset, severity of inflammation, and a poor response to treatment [bib_ref] Ulcerative colitis as a progressive disease: the forgotten evidence, Torres [/bib_ref] [bib_ref] Comparing the clinical outcomes of young-onset and adult-onset ulcerative colitis: a multi-center..., Kim [/bib_ref]. Poor prognostic factors associated with CD include: younger age, presence of strictures and/or penetrating lesions, jejunal or perianal involvement at the time of diagnosis, systemic steroid use, and/or smoking habits/history. The recently developed Korean CD prediction model based on clinical parameters assessed at the time of diagnosis, serves as an accurate predictor of a patient's probability of undergoing CD-related surgery [bib_ref] Development of a novel predictive model for the clinical course of Crohn's..., Park [/bib_ref]. This kind of predictive model based on an assessment of poor clinical factors can be used to stratify patients into subgroups and identify those who would benefit from early individualized intervention among newly diagnosed patients to control disease activity and prevent progressive disease.
## Treat-to-target
Therapeutic targets have been defined for chronic diseases such as diabetes, hypertension, and rheumatoid arthritis. With respect to IBD, the concept of treat-totarget originated from the discrepancies observed between symptoms, objectively evaluated inflammatory activity, and intestinal damage, which is more commonly observed in those presenting with CD than with UC. Various studies have shown that a full Mayo score including the objective endoscopic grade shows a relatively high correlation with disease severity of UC. However, a poor correlation was frequently observed between the endoscopic disease activity and the symptom-based Crohn's disease activity index (CDAI) [bib_ref] Relationships between disease activity and serum and fecal biomarkers in patients with..., Jones [/bib_ref].
Shifting the therapeutic target from clinical remission to objective parameters such as endoscopic healing has been supported by population-based cohort studies, post hoc analysis of clinical trials, and meta-analysis [bib_ref] Early mucosal healing with infliximab is associated with improved long-term clinical outcomes..., Colombel [/bib_ref] [bib_ref] Mucosal healing in inflammatory bowel disease: results from a Norwegian population-based cohort, Froslie [/bib_ref] [bib_ref] Mucosal healing is associated with improved long-term outcomes of patients with ulcerative..., Shah [/bib_ref] [bib_ref] Adalimumab induces deep remission in patients with Crohn's disease, Colombel [/bib_ref]. The Active Ulcerative Colitis Trial (ACT)-1 and ACT-2 have shown that MH observed in patients following 8-week treatment with infliximab demonstrated a positive correlation with better clinical outcomes including lower colectomy rates [bib_ref] Early mucosal healing with infliximab is associated with improved long-term clinical outcomes..., Colombel [/bib_ref]. A meta-analysis involving > 2,000 patients presenting with active UC has shown that patients achieving MH at the time of the first endoscopic evaluation demonstrate long-term clinical remission, avoidance of colectomy, and corticosteroid-free clinical remission [bib_ref] Mucosal healing is associated with improved long-term outcomes of patients with ulcerative..., Shah [/bib_ref]. In patients with CD treated with adalimumab, deep remission at 12 weeks, a composite of clinical and endoscopic remission was significantly associated with fewer dose adjustments, hospitalizations, and CD-related surgeries, as well as an improved quality of life and physical function at 1 year [bib_ref] Adalimumab induces deep remission in patients with Crohn's disease, Colombel [/bib_ref].
Recently, the Selecting Therapeutic Targets in Inflam- [bib_ref] Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE): determining therapeutic goals for..., Peyrin-Biroulet [/bib_ref]. However, the applicability of the treat-to-target approach may be difficult in clinical practice. Most Asian physicians use the partial Mayo score or the Truelove and Witts severity index, which primarily relies upon symptom category to assess the disease activity of UC at the time of diagnosis and during subsequent follow-up. Regarding CD, the CDAI is used by most Asian physicians as the gold standard to determine disease activity; however, recent surveys have shown wide variations among physicians in adopting endoscopic and/or radiological evaluations as objective parameters that accurately reflect inflammatory activity [bib_ref] Diagnosis of inflammatory bowel disease in Asia: the results of a multinational..., Kim [/bib_ref]. Additionally, more conclusive evidence is needed to support this concept, and clinicians should remain mindful of a possible target change occurring in the future and the possibility of overtreatment of low-risk patients [bib_ref] Management strategies to improve outcomes of patients with inflammatory bowel diseases, Colombel [/bib_ref] [bib_ref] Beyond endoscopic mucosal healing in UC: histological remission better predicts corticosteroid use..., Bryant [/bib_ref].
## Individualized approach with tight monitoring
Objective measures to assess inflammatory activity are important to adopt the treat-to-target strategy utilized for IBD management. The STRIDE program recommends endoscopic healing combined with clinical/patient-reported outcomes as therapeutic targets; however, biomarkers such as CRP or FC are used as an adjunctive target [bib_ref] Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE): determining therapeutic goals for..., Peyrin-Biroulet [/bib_ref]. However, endoscopic assessment is usually performed prior to initiating or modifying treatment in patients showing a relapse or showing the development of new symptoms, or those in whom surgery is needed. Moreover, the cost factor and patient compliance are primary limitations associated with regular endoscopic evaluation of MH due to the invasiveness associated with the procedure. Estimation of CRP and FC are well-studied noninvasive biomarkers that are widely used [bib_ref] Advances in use of endoscopy, radiology, and biomarkers to monitor inflammatory bowel..., Panes [/bib_ref].
The Efficacy and Safety of Two Treatment Models in Subjects With Moderate to Severe Crohn's Disease (CALM) study compared the efficacy and safety of a tight control algorithm based on biomarkers such as CRP and FC in addition to clinical symptoms in patients who were naive to IM and/or biologic therapy. A total of 244 patients with CD showing < 6 years of duration were randomly assigned to either a tight control or a clinical management group. The tight control group was monitored with estimation of CRP and FC in addition to clinical symptoms based on the CDAI. After 8 weeks of prednisolone therapy, their treatment regimen was escalated in a stepwise manner including the addition of adalimumab and azathioprine in both groups who met the predefined failure criteria. The primary endpoint defined as MH at 48 weeks was achieved in 46% of the tight control group, which was significantly higher than 30% observed in the clinical management group. This study showed that tight monitoring of objective markers, as well as clinical symptoms helped in effective decision making for the institution of appropriate treatment and therefore achieved superior clinical and endoscopic outcomes compared to the symptom-based approach.
The introduction of IMs and biologic agents has reduced the risk of surgery over time [bib_ref] Long-term prognosis of Crohn's disease and its temporal change between 1981 and..., Park [/bib_ref] [bib_ref] Long-term prognosis of ulcerative colitis and its temporal change between 1977 and..., Lee [/bib_ref]. However, a significant number of patients with IBD, particularly those diagnosed with CD are required to undergo repeat surgery due to relapse, leading to significant morbidity and disability [bib_ref] Surgical management of inflammatory bowel disease in China: a systematic review of..., Yu [/bib_ref]. Thiopurine or anti-TNF therapy is considered in addition to smoking cessation and 5-aminosalicylic acid maintenance therapy to prevent postoperative relapse in patients diagnosed with CD [bib_ref] Second Korean guidelines for the management of Crohn's disease, Park [/bib_ref]. The Post-Operative Crohn's Endoscopic Recurrence (POCER) study showed that individualized management based on the results of 6-month colonoscopic examination combined with an assessment of the clinical risk of recurrence was more effective than standard drug therapy in reducing postoperative recurrence [bib_ref] Crohn's disease management after intestinal resection: a randomised trial, Cruz [/bib_ref]. Estimation www.kjim.org https://doi.org/10.3904/kjim.2017.400
The Korean Journal of Internal Medicine Vol. 33, No. 1, January 2018 of FC at 6, 12, and 18 months after intestinal resection was a good predictor of relapse, and monitoring of FC is a noninvasive and inexpensive technique, which could serve as a useful component of the individualized postoperative management algorithm [bib_ref] Measurement of fecal calprotectin improves monitoring and detection of recurrence of Crohn's..., Wright [/bib_ref].
Despite the proven efficacy of anti-TNF agents in inducing and maintaining remission, approximately 10% to 20% of IBD patients are refractory to induction therapy, and annually 13% to 24% patients report a loss of response (LOR) during maintenance therapy [bib_ref] Is the therapeutic drug monitoring of anti-TNF agents necessary in Korean inflammatory..., Eun [/bib_ref]. Therapeutic drug monitoring (TDM) combined with anti-drug antibodies (ADAs) is an important component of the individualized management strategy used for IBD patients treated with anti-TNF agents. TDM can be performed periodically in patients showing remission, as well as in those showing a primary nonresponse, secondary LOR, or drug-related adverse events to guide decision-making regarding treatment optimization [bib_ref] Review article: consensus statements on therapeutic drug monitoring of anti-tumour necrosis factor..., Mitrev [/bib_ref]. Recently, a proactive approach, targeting a trough level of 5 to 10 µg/mL by measuring the infliximab concentration and ADA, in asymptomatic IBD patients showed better clinical outcomes including lesser need for IBD-related surgery and hospitalization than the benefits noted with reactive monitoring in patients showing symptom aggravation or suspected LOR. Proactive monitoring was also observed to be associated with a reduced risk of ADA development or infusion-related adverse events [bib_ref] Improved long-term outcomes of patients with inflammatory bowel disease receiving proactive compared..., Papamichael [/bib_ref]. Improved clinical outcomes were also observed with the use of adalimumab, fully human anti-TNF agents, and vedolizumab, a humanized monoclonal antibody against the α4β7 integrin in patients in whom doses had been optimized based on the results of pharmacokinetic studies [bib_ref] Development of an algorithm incorporating pharmacokinetics of adalimumab in inflammatory bowel diseases, Roblin [/bib_ref] [bib_ref] Adalimumab induction and maintenance therapy achieve clinical remission and response in Chinese..., Wu [/bib_ref] [bib_ref] apy for inflammatory bowel diseases and need for additional doses within 6..., Williet [/bib_ref].
## De-escalation strategy
Healthcare costs associated with IBD management have shown a significant rise. This is because a greater number of patients are being administered biologic agents alone or concomitantly with IMs earlier in the course of the disease owing to newly emerging paradigms for the management of IBD [bib_ref] The global burden of IBD: from 2015 to 2025, Kaplan [/bib_ref]. Moreover, the risk of tuberculosis (TB), both, de novo infection and reactivation of latent TB, is higher in patients receiving anti-TNF agents, particularly in areas such as Korea, which show a high prevalence of the disease [bib_ref] Clinical features of active tuberculosis that developed during anti-tumor necrosis factor therapy..., Lee [/bib_ref]. Use of thiopurines is associated with significant toxicity such as bone marrow suppression, liver toxicity, opportunistic infections, and malignancies including lymphoma and non-melanoma skin cancers [bib_ref] IBD Study Group of the Korean Association for the Study of Intestinal..., Lee [/bib_ref]. Concerns about cost and toxicities have led physicians and patients to consider safe de-escalation strategies once remission has been achieved.
Several long-term trials have shown that discontinuation of therapy is associated with high relapse rates, suggesting that complete cessation of treatment can be considered and implemented only in selected patients [bib_ref] Systematic review of effects of withdrawal of immunomodulators or biologic agents from..., Torres [/bib_ref]. The infliximab diSconTinuation in CrOhn's disease patients in stable Remission on combined therapy with Immunosuppressors (STORI) trial prospectively examined the risks and predictors of relapse after withdrawal of infliximab in patients receiving combined maintenance therapy with antimetabolites [bib_ref] Maintenance of remission among patients with Crohn's disease on antimetabolite therapy after..., Louis [/bib_ref]. The relapse rate within a 1 year was approximately 50%; however, patients having ≤ 2 risk factors, including male gender, elevated leukocytes, an elevated CRP level, elevated FC level, and decreased hemoglobin showed only a 15% risk of relapse. Several studies have identified complicated or perianal CD, extensive disease, clinical symptoms, elevated biomarkers, endoscopically documented severe lesions, and monotherapy as predictors of relapse following cessation of therapy. However, there is no consensus regarding the optimal time and methodology with respect to cessation of this potent disease-modifying therapy. This decision needs to be individualized based on patient demographics, disease features, current disease status, and patients' preferences [bib_ref] Management strategies to improve outcomes of patients with inflammatory bowel diseases, Colombel [/bib_ref].
# Conclusions
IBD is a disease entity known to cause progressive intestinal damage and disability; thus, early intervention can alter the natural course of the disease and prevent irreversible complications. However, heterogeneity in clinical features and prognosis often prevents the use of early and effective strategies. Recent studies have determined a number of long-term prognostic factors and developed predictive models, which can identify patients who can maximally benefit from early intervention. In patients diagnosed with IBD, clinical symptoms are considered to reflect the tip of the iceberg, and symptom-driven www.kjim.org https://doi.org/10.3904/kjim.2017.400 management often fails to improve long-term outcomes. Lately, therapeutic targets have been shifted from clinical remission to objective parameters such as evidence of endoscopic healing. An individualized treatment approach based on close and timely monitoring using non-invasive tests combined with endoscopic and radiological assessment is essential to ensure better outcomes. The management of IBD patients has greatly improved over the last decade following the introduction of newer classes of drugs, particularly biologic agents. However, earlier administration and wider use of these drugs can lead to significant healthcare costs nationwide, as well as a higher incidence of untoward treatment-related adverse events. Therefore, the use of biologic agents should be optimized with tight monitor, and an appropriate de-escalation strategy is warranted in selected patients. Further studies are needed to validate the efficacy of these newly emerging treatment paradigms to confirm their role in improving long-term prognosis by reducing disability, need for surgery and hospitalization to improve the patient's quality of life.
[fig] Figure 1: Current strategies for the management of inflammatory bowel disease. www.kjim.org https://doi.org/10.3904/kjim. [/fig]
[table] Table 1: Concept change of inflammatory bowel disease [/table]
|
High endothelial venules are associated with microsatellite instability, hereditary background and immune evasion in colorectal cancer
BACKGROUND: Microsatellite-unstable (MSI) tumours show a high load of mutational neoantigens, as a consequence of DNA mismatch repair deficiency. Consequently, MSI tumours commonly present with dense immune infiltration and develop immune evasion mechanisms. Whether improved lymphocyte recruitment contributes to the pronounced immune infiltration in MSI tumours is unknown. We analysed the density of high endothelial venules (HEV) and postcapillary blood vessels specialised for lymphocyte trafficking, in MSI colorectal cancers (CRC). METHODS: HEV density was determined by immunohistochemical staining of FFPE tissue sections from MSI (n = 48) and microsatellite-stable (MSS, n = 35) CRCs. Associations with clinical and pathological variables were analysed. RESULTS: We found elevated HEV densities in MSI compared with MSS CRCs (median 0.049 vs 0.000 counts/mm 2 , respectively, p = 0.0002), with the highest densities in Lynch syndrome MSI CRCs. Dramatically elevated HEV densities were observed in B2M-mutant Lynch syndrome CRCs, pointing towards a link between lymphocyte recruitment and immune evasion (median 0.485 vs 0.0885 counts/mm 2 in B2M-wild-type tumours, p = 0.0237). CONCLUSIONS: Our findings for the first time indicate a significant contribution of lymphocyte trafficking in immune responses against MSI CRC, particularly in the context of Lynch syndrome. High HEV densities in B2M-mutant tumours underline the significance of immunoediting during tumour evolution.
# Background
Genomic instability in cancer can arise through two main mechanisms: chromosomal and microsatellite instability. Microsatellite instability (MSI) is caused by a defect in the DNA mismatch repair (MMR) system, which can occur sporadically by epigenetic alterations in the MLH1 gene, or in the context of the most common inherited cancer syndrome, named Lynch syndrome, by mutational inactivation of one of the four MMR genes (MLH1, MSH2, MSH6 and PMS2).As a consequence, mismatches occurring during DNA replication cannot be corrected, and small insertion/deletion mutations accumulate at short repetitive stretches called microsatellites.
Accumulation of uncorrected mutations at microsatellites leads to shifts of the translational reading frame and generation of long stretches of the so-called frameshift peptides (FSPs), potentially containing highly immunogenic epitopes. [bib_ref] The immune biology of microsatelliteunstable cancer, Kloor [/bib_ref] The high immunogenicity of MSI-induced neoantigens is thought to explain the dense infiltration of tumour tissue with cytotoxic T cells, [bib_ref] High prevalence of activated intraepithelial cytotoxic T lymphocytes and increased neoplastic cell..., Dolcetti [/bib_ref] and hence the favourable prognosis of patients with MSI cancers, in contrast to patients with MSS cancers, [bib_ref] Microsatellite instability in colorectal cancer is associated with local lymphocyte infiltration and..., Buckowitz [/bib_ref] [bib_ref] Systematic review of microsatellite instability and colorectal cancer prognosis, Popat [/bib_ref] [bib_ref] Integrative analyses of colorectal cancer show immunoscore is a stronger predictor of..., Mlecnik [/bib_ref] and the particularly good response to immune checkpoint blockade. [bib_ref] PD-1 blockade in tumors with mismatch-repair deficiency, Le [/bib_ref] In addition, 30% of MSI CRCs present with mutations of the B2M gene encoding for the light chain of MHC class I molecule. [bib_ref] Beta2-microglobulin mutations in microsatellite unstable colorectal tumors, Kloor [/bib_ref] [bib_ref] Complex pattern of immune evasion in MSI colorectal cancer, Ozcan [/bib_ref] Such B2M mutations lead to the synthesis of a truncated B2M protein, thereby completely abrogating MHC class I-associated antigen expression and contributing to immune evasion. The outgrowth of B2Mmutant tumour cell clones represents a manifestation of cancer immunoediting concept [bib_ref] Cancer immunoediting: from immunosurveillance to tumor escape, Dunn [/bib_ref] [bib_ref] Cancer exome analysis reveals a T-cell-dependent mechanism of cancer immunoediting, Matsushita [/bib_ref] [bib_ref] Cancer immunoediting: integrating immunity's roles in cancer suppression and promotion, Schreiber [/bib_ref] and further supports high immunogenicity of MSI CRCs and their active immune surveillance. [bib_ref] Lynch syndrome-cancer pathways, heterogeneity and immune escape, Seth [/bib_ref] However, the mechanisms of lymphocyte recruitment that may contribute to the pronounced immune infiltration of MSI cancers have not yet been determined.
High endothelial venules (HEVs) are postcapillary venules specialised for lymphocyte trafficking. [bib_ref] High endothelial venules (HEVs): specialized endothelium for lymphocyte migration, Girard [/bib_ref] Endothelial cells lining HEVs express on their cell surface peripheral node addressins (PNAd), molecules which are important for lymphocyte homing. HEVs attract naive and central memory lymphocytes from the bloodstream into lymph nodes, where they scan antigen-presenting cells for activating or tolerogenic stimuli. [bib_ref] Understanding high endothelial venules: lessons for cancer immunology, Ager [/bib_ref] Recently, HEVs were also found in tumours of different origin, including colorectal, breast, ovarian, lung cancer and melanomas, [bib_ref] Human solid tumors contain high endothelial venules: association with T-and Blymphocyte infiltration..., Martinet [/bib_ref] [bib_ref] High endothelial venules are rare in colorectal cancers but accumulate in extratumoral..., Bento [/bib_ref] implying the possibility of direct lymphocyte recruitment into cancer tissue. Moreover, the density of HEVs has been shown to correlate with T-cell infiltration in tumours and survival of patients with breast cancer. [bib_ref] Human solid tumors contain high endothelial venules: association with T-and Blymphocyte infiltration..., Martinet [/bib_ref] In this study, we analysed whether HEVs play a role in the pronounced immune cell infiltration of MSI CRCs by assessing their density in MSI and MSS CRC tissues. Moreover, we aimed to identify additional clinicopathological factors potentially associated with the presence of HEVs in colorectal tumours.
# Methods
## Tumour samples
Formalin-fixed paraffin-embedded (FFPE) tumour tissue blocks were collected at the Department of Applied Tumour Biology, Institute of Pathology, University Hospital Heidelberg, as part of the German HNPCC Consortium. Informed and written consent was provided by all patients, and the study approval was obtained from the Institutional Ethics Committee.
Immunohistochemical staining FFPE tissue sections (2 µm) were used for immunohistochemical staining. Deparaffinisation, rehydration and staining were performed according to standard protocols as described previously. [bib_ref] A large MSH2 Alu insertion mutation causes HNPCC in a German kindred, Kloor [/bib_ref] [bib_ref] High numbers of PDCD1 (PD-1)-positive T cells and B2M mutations in microsatellite-unstable..., Janikovits [/bib_ref] For HEV staining, the rat monoclonal antibody MECA-79 (Santa Cruz Biotechnology, Dallas, TX, USA, clone SC-19602) was applied as a primary antibody at 1:750 dilution; as a secondary antibody, the biotinylated goat anti-rat antibody (Vector Laboratories, Burlingame, CA, USA) was used at 1:100 dilution. For PD-L1 staining, anti-PD-L1 rabbit monoclonal antibody (Ventana Medical Systems, Tucson, AZ, USA, clone SP263) was used, CD3 staining was performed using the α-CD3 (Abcam, Cambridge, UK, clone SP7) rabbit monoclonal antibody at 1:200 dilution, PD1 staining was performed using the α-PDCD1 mouse monoclonal antibody (1:50 dilution, Abcam, clone NAT105), CD20 staining was performed using the α-CD20 (Santa Cruz, clone D-10) mouse monoclonal antibody at 1:200 dilution, and cytokeratin staining was performed using mouse monoclonal antibody, clone AE1/AE3 (Dako) at 1:50 dilution; as a secondary antibody in PD-L1, PD1 and CD3 staining, biotinylated horse anti-mouse/anti-rabbit antibody (Vector Laboratories, Burlingame, CA, USA) was used at 1:50 dilution. Staining visualisation was performed using the Vectastain elite ABC detection system (Vector Laboratories, Burlingame, CA, USA) and 3,3′-Diaminobenzidine (Dako North America Inc., Carpinteria, CA, USA) as a chromogen. Finally, slides were counterstained with haematoxylin.
Evaluation of HEV density The HEVs were identified through positive staining with MECA-79 antibody and histomorphological properties characteristic of cells comprising HEVs, such as a cuboidal shape of high endothelial cells. [bib_ref] Understanding high endothelial venules: lessons for cancer immunology, Ager [/bib_ref] [bib_ref] High endothelial venules are rare in colorectal cancers but accumulate in extratumoral..., Bento [/bib_ref] Spatially separated MECA-79-positive HEVs were counted under a light microscope at ×4 objective magnification. Representative MECA-79-positive HEVs are shown in . A randomly selected subset of slides was evaluated by a second observer (r = 0.93, p = 0.00012, Spearman correlation). The observers were blinded for any clinical parameters, such as age, gender, stage, hereditary background, and molecular tumour characteristics. All counts were normalised per mm 2 of tissue. Finally, MECA-79-positive HEVs were calculated for each sample as counts per mm 2 following the formula: overall HEVs counts=mm 2 Â Ã ¼ P counts intratumoural and peritumoural area tumour and peritumoural Intratumoural and peritumoural were calculated separately from each other . Counts in the peritumoural region were calculated as follows: Slides were scanned using a Hamamatsu NanoZoomer Digital Pathology system (Hamamatsu, Hamamatsu City, Japan) and analysed with NDP.view 2 software.
Evaluation of PD-L1 expression Samples were microscopically examined for PD-L1 expression and scored as negative, weak, medium or strong based on the percentage of staining-positive tissue area (0%, 1% to <10%, 10-40% and >40%, respectively). Scoring was performed independently by two observes, and observers were blinded during the scoring process.
Evaluation of intratumoural CD3-positive T-cell density Quantification of CD3 was performed as previously described. [bib_ref] High numbers of PDCD1 (PD-1)-positive T cells and B2M mutations in microsatellite-unstable..., Janikovits [/bib_ref] In digitised images of IHC staining, four random 0.25-mm 2 squares were drawn and positive cells in the tumour region were counted, giving mean values per 0.25 mm 2 . Lymph follicles and germinal centres were identified on the basis of their morphological appearance and positive staining for CD20, and the counts were normalised per mm 2 of the tissue.
Preparation of age-independent cohorts Samples from the MSI sporadic group were age-matched with samples from the hereditary MSI and MSS cohorts. Overall, 20 matches for each cohort could be performed. Age independency was assessed in a Welch two-sample t test [fig_ref] Figure 3: Relation of HEVs and immune infiltration [/fig_ref].
B2M-mutation analysis Mutation status of B2M was determined by Sanger sequencing as described previously. [bib_ref] Beta2-microglobulin mutations in microsatellite unstable colorectal tumors, Kloor [/bib_ref] Briefly, PCR amplification of B2M exons 1 and 2 was performed using primer sequences: Exon 1 For-GGCATTCCTGAAGCTGACA, Exon 1 Rev-AGAGCGGGAGAGGAA GGAC, Exon 2a For-TTTCCCGATATTCCTCAGGTA, Exon 2a Rev-AATTCAGTGTAGTACAAGAG and Exon 2b For-TGTCTTTCA GCAAGGACTGG, Exon 2b Rev-CAAAGTCACATGGTTCACACG. After purification of the obtained PCR products (QIAquick PCR Purification Kit), the sequencing reaction was performed using the BigDye ® Terminator v1.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Wilmington, DE, USA). Precipitated products were dissolved in 12 µl of HiDi Formamide (Thermo Fisher Scientific, Wilmington, DE, USA), sequenced on ABI 3130xl Genetic Analyzer and evaluated using Sequencing Analysis Software (Applied Biosystems).
## Statistical calculations
If not specified otherwise, p-values were calculated using the nonparametric Wilcoxon rank-sum test. For data following t-distribution, p-values were calculated using Welch two-sample t test, and for discrete values, a chi-square test was performed. As a measure of significance, p ≤ 0.05 was applied. All statistical analyses were performed in RStudio 20 (version 1.1.453).
# Results
## Hev density in msi and mss colorectal cancers
We first aimed to analyse whether the density of HEVs in tumours is related to the MSI phenotype. By comparing 48 MSI against 35 MSS CRCs, we found a striking difference in the HEV density, with MSI CRCs having significantly higher HEV density (median MSI = 0.049 vs median MSS = 0.000 counts/mm 2 , p = 0.0002, . HEVs were preferentially located within the peritumoural region (43/54, 79.6% of HEV-positive CRCs) and less frequently found within the intratumoural area (22/54, 40.7% of HEV-positive CRCs); some tumours presented with both intra-and peritumoural HEVs -e, . Intra-and peritumoural did not differ regarding morphology.
Our MSI cohort comprises sporadic and hereditary MSI CRCs. Therefore, we next analysed the density of HEVs in sporadic . There was also a significant increase in the density of peritumoural HEVs: hereditary MSI CRCs presented with significantly elevated HEV densities as compared with sporadic counterparts (median = 0.136 vs median = 0.000 counts/mm 2 , p = 0.0104, . Sporadic MSI CRC samples showed a significantly elevated HEV density (p = 0.0159, , compared with MSS CRC (median MSI = 0.025 vs median MSS = 0.000 counts/mm 2 ), when intra-and peritumoural HEVs were calculated together.
Correlation of HEV density with clinical parameters Unlike sporadic cancer patients, Lynch syndrome patients are diagnosed with cancer at a younger age, due to genetic predisposition. [bib_ref] Microsatellite instability in colorectal cancer, Boland [/bib_ref] [bib_ref] Clinical significance of microsatellite instability in colorectal cancer, Kloor [/bib_ref] We therefore asked whether the strong correlation of high HEV density with hereditary MSI tumour phenotype was a function of the younger age of Lynch syndrome patients, or possibly, whether this phenomenon was related to older age, due to long-term exposure of the immune system of Lynch syndrome patients to FSP neoantigens. [bib_ref] Immune response against frameshift-induced neopeptides in HNPCC patients and healthy HNPCC mutation..., Schwitalle [/bib_ref] [bib_ref] Prevalence of mismatch repair-deficient crypt foci in Lynch syndrome: a pathological study, Kloor [/bib_ref] [bib_ref] Mismatch repair-deficient crypt foci in Lynch syndrome-molecular alterations and association with clinical..., Staffa [/bib_ref] First, we analysed the age distribution in our cohorts: as expected, age at CRC diagnosis differed significantly between three patient groups: while the mean age in the hereditary MSI CRC group was 52.7 years, the mean age of sporadic MSI CRC patients was 70.3 years, and the mean age of MSS CRC patients was 58.9 [fig_ref] Figure 3: Relation of HEVs and immune infiltration [/fig_ref]. However, when we analysed the HEV density solely in relation to patients' age at diagnosis, remarkably, we did not find any differences [fig_ref] Figure 3: Relation of HEVs and immune infiltration [/fig_ref]. Similarly, no correlation was detectable between HEV density and patients' age at diagnosis, upon dissecting the three different patient groups (Supplementary [fig_ref] Figure 3: Relation of HEVs and immune infiltration [/fig_ref].
As a next step, we compiled three age-matched groups with hereditary MSI, sporadic MSI and MSS CRC patients (Supplementary [fig_ref] Figure 3: Relation of HEVs and immune infiltration [/fig_ref]. Comparison of HEV densities between these ageindependent patient groups confirmed the significantly elevated HEV density in hereditary MSI CRC (median = 0.1275 counts/mm 2 ) when compared with sporadic MSI (median = 0.0255 counts/ mm 2 ) and MSS (median = 0.002 counts/mm 2 ) CRCs (p = 0.002 and p = 8.9 × 10 −6 , respectively, .
Furthermore, we analysed the correlation of HEV densities with tumour stage. The distribution of HEV densities among different tumour stages suggested potentially higher HEV counts in tumours with a lower stage; however, statistical significance could not be reached, most likely due to limited sample size for each tumour stage .
Correlation of HEV density with immune infiltration and immune evasion markers To investigate the relation between HEV density and immune infiltration, we first analysed the presence of CD3-and CD20positive immune cells in the vicinity of HEVs. CD3-positive T cells were found to be localised directly around HEVs, whereas CD20positive cells marked the germinal centres of the lymph follicles associated with the HEVs [fig_ref] Figure 3: Relation of HEVs and immune infiltration [/fig_ref]. Next, intratumoural CD3-positive T-cell density was assessed in a subgroup of MECA-79-stained MSI CRCs. Although not reaching significance (p = 0.1637), a trend towards a higher CD3-positive cell infiltration in HEV high tercile was observed, as compared with two terciles presenting with lower HEV densities [fig_ref] Figure 4: Association of HEV densities with immune response and immune evasion markers in... [/fig_ref]. Furthermore, PD1-positive intratumoural T-lymphocyte infiltration was assessed. HEV high MSI CRCs presented with higher PD1positive T-cell infiltration, as compared with HEV low counterparts. However, a significant difference was not observed (p = 0.3361, [fig_ref] Figure 4: Association of HEV densities with immune response and immune evasion markers in... [/fig_ref]. We also analysed the density of lymph follicles and germinal centres in MSI hereditary, MSI sporadic and MSS cohorts and correlated it with HEV densities. As expected, lymph follicle densities were significantly elevated in hereditary and sporadic MSI tumours compared with MSS tumours. Densities of lymph follicles with a germinal centre were also elevated in hereditary MSI tumours compared with MSS tumours. No significant difference in lymph follicle and germinal centre densities was observed between MSI hereditary and MSI sporadic tumours. Importantly, a significant correlation between HEV and lymph follicle densities, as well as between HEV and lymph follicle with germinal centre densities was observed [fig_ref] Figure 5: Graphical summary [/fig_ref] and S6c-f).
Mutations of the B2M gene have been previously reported in 30% of MSI CRCs, with a higher proportion of B2M mutations in Lynch syndrome-associated compared with sporadic MSI CRCs. [bib_ref] Beta2-microglobulin mutations in microsatellite unstable colorectal tumors, Kloor [/bib_ref] We therefore analysed the HEV densities separately in B2M-mutant (n = 21) and B2M-wild-type (n = 26) MSI CRCs. Here, B2M-mutant tumours showed higher HEV densities compared with B2M-wild type tumours (median 0.139 vs 0.052 counts/mm 2 , p = 0.0116, [fig_ref] Figure 4: Association of HEV densities with immune response and immune evasion markers in... [/fig_ref]. We then analysed B2M-mutant hereditary MSI CRCs separately from sporadic ones and compared each of the groups with their B2M-wild-type counterparts. Among hereditary tumours, B2M-mutant MSI CRCs (n = 13) presented with significantly higher HEV density compared with B2M-wild-type (n = 16) tumours (median 0.485 vs 0.089 counts/mm 2 , p = 0.0237, [fig_ref] Figure 4: Association of HEV densities with immune response and immune evasion markers in... [/fig_ref].
In addition to their high HEV density, B2M-mutant (n = 12) tumours also presented with stronger PD-L1 expression (p = 0.0161, Pearson's chi-squared test) in comparison with their B2Mwild-type (n = 14) counterparts [fig_ref] Figure 4: Association of HEV densities with immune response and immune evasion markers in... [/fig_ref].
# Discussion
We observed an increased HEV density in MSI compared with MSS CRCs. To the best of our knowledge, this is the first study showing a relation between HEV density and MSI. A previous study addressing this question did not find a correlation of HEV densities with MSI status, 17 most likely due to the limited number of only four MSI CRC samples analysed. The increased density of HEVs adjacent to MSI CRCs suggests that HEVs may contribute to the pronounced immune infiltration typically observed in MSI CRCs, by attracting circulating lymphocytes from the bloodstream directly to the tumour site. [bib_ref] Microsatellite instability in colorectal cancer is associated with local lymphocyte infiltration and..., Buckowitz [/bib_ref] [bib_ref] Systematic review of microsatellite instability and colorectal cancer prognosis, Popat [/bib_ref] [bib_ref] Tumor-infiltrating lymphocytes are a marker for microsatellite instability in colorectal carcinoma, Smyrk [/bib_ref] The increased density of HEVs in MSI tumours could be due to the high antigenic load and immunogenicity of MSI CRCs. [bib_ref] The immune biology of microsatelliteunstable cancer, Kloor [/bib_ref] However, the induction of HEV neogenesis in tumours is poorly HEV density in sporadic MSI, hereditary MSI and MSS CRC samples. a Overall HEV density. MSI CRCs showed significantly elevated HEV densities as compared with MSS CRCs (p = 0.0002). Comparing MSI hereditary and sporadic MSI or MSS CRCs revealed a significantly elevated HEV density in hereditary MSI CRCs (p = 0.0255 and p = 3.9 × 10 -5 , respectively). MSI sporadic CRCs also showed significantly elevated HEV densities as compared with MSS CRCs (p = 0.0159). b Peritumoural HEV density. Considering only peritumoural HEVs, MSI CRCs with Lynch syndrome background revealed elevated HEV densities within this group, as compared with MSI sporadic and MSS (p = 0.010 and p = 0.0018, respectively). Black lines in boxes indicate the median value; cross indicates the mean value. Significance levels are depicted with asterisks (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001). c Representative image for HEV staining in MSI CRC of hereditary origin, d image for HEV staining in MSI CRC of sporadic origin and e presents HEV staining in MSS CRC. The boundary between the intratumoural (it) and peritumoural (pt) area is marked by a dashed line. Scale bar in c-e is 1 mm and HEVs are indicated by black arrows. All p-values were obtained from a nonparametric Wilcoxon rank-sum test understood. HEV neogenesis has been previously shown to occur under chronic inflammatory conditions, [bib_ref] High endothelial venules (HEVs): specialized endothelium for lymphocyte migration, Girard [/bib_ref] [bib_ref] Lymphatic vessels and tertiary lymphoid organs, Ruddle [/bib_ref] such as the Crohn'slike reaction frequently observed in MSI tumours. [bib_ref] High prevalence of activated intraepithelial cytotoxic T lymphocytes and increased neoplastic cell..., Dolcetti [/bib_ref] [bib_ref] Microsatellite instability in colorectal cancer is associated with local lymphocyte infiltration and..., Buckowitz [/bib_ref] One possible inducer of HEV neogenesis could be inflammatory cytokines, such as lymphotoxins. Lymphotoxin-α has been shown to have a role in HEV neogenesis. Furthermore, the lymphotoxin-αβ complex expressed by dendritic cells has been implicated in HEV neogenesis in mouse models and in human breast cancer cells. [bib_ref] Understanding high endothelial venules: lessons for cancer immunology, Ager [/bib_ref] [bib_ref] High endothelial venule blood vessels for tumor-infiltrating lymphocytes are associated with lymphotoxin..., Martinet [/bib_ref] Recent findings describing HEV neogenesis upon tissue infiltration with activated T cells suggested a role for HEV in tumour destruction. [bib_ref] Treg depletion licenses T cell-driven HEV neogenesis and promotes tumor destruction, Colbeck [/bib_ref] [bib_ref] Tcell trafficking facilitated by high endothelial venules is required for tumor control..., Hindley [/bib_ref] A strong local immune response dependent on the recruitment of circulating lymphocytes by HEV could therefore be important for defeating MSI cancers. Studies analysing the induction of HEV neogenesis in MSI cancers are warranted, in order to identify potential targets for immune therapeutic approaches.
In line with previous publications, HEVs were in general more frequently found in peritumoural areas than intratumourally. [bib_ref] High endothelial venules are rare in colorectal cancers but accumulate in extratumoral..., Bento [/bib_ref] [bib_ref] Ectopic expression of MECA-79 as a novel prognostic indicator in gastric cancer, Okayama [/bib_ref] Recognition by the MECA-79 antibody of a glycoantigen structure also present in a subset of CRCs (about 40%) 15,17,32-34 resulted in positive staining of the tumour epithelium doppelte Beschreibung des subsets. However, this staining was clearly not indicative of HEVs and therefore was not accounted for by the observers .
The comparison of HEV densities between sporadic and hereditary MSI CRCs revealed a significant difference, with higher HEV densities in hereditary MSI CRC. As sporadic MSI CRC patients are commonly of older age than hereditary MSI CRC patients, [bib_ref] Microsatellite instability in colorectal cancer, Boland [/bib_ref] [bib_ref] Clinical significance of microsatellite instability in colorectal cancer, Kloor [/bib_ref] we suspected that the observed differences could be caused by the difference in the average age of patients from these two groups. However, age alone did not have a measurable influence on HEV densities. To exclude a possible [bib_ref] Value of histopathology in predicting microsatellite instability in hereditary nonpolyposis colorectal cancer..., Shia [/bib_ref] [bib_ref] Lynch syndromeassociated neoplasms: a discussion on histopathology and immunohistochemistry, Shia [/bib_ref]. Furthermore, it suggests enhanced lymphocyte trafficking towards manifest tumours in Lynch syndrome individuals compared with sporadic MSI CRC patients and for the first time evaluates the contribution of HEVs to the immune microenvironment of different molecular CRC types, particularly delineating fundamental differences between immune response in hereditary and sporadic MSI CRC patients. Moreover, observation of a stronger immune infiltration in MSI hereditary endometrial tumours compared with sporadic ones, 37 suggests that the difference between sporadic and hereditary MSI tumours may not be restricted to CRCs, but applies to a broad spectrum of Lynch syndrome-associated tumours. As expected, we also observed enhanced lymph follicle counts in MSI vs MSS CRC. However, differentiation between hereditary and sporadic MSI CRCs based on the lymph follicle counts alone was not statistically significant . It is plausible to speculate that, in contrast to lymph follicle counts, which are Lynch syndrome mutation carriers have circulating T cells, specifically recognising MSI-induced neoantigens, which can be detected in peripheral blood even before cancer manifestation. [bib_ref] Immune response against frameshift-induced neopeptides in HNPCC patients and healthy HNPCC mutation..., Schwitalle [/bib_ref] Such immune responses are most likely triggered by MMRdeficient crypt foci harbouring mutations that give rise to FSP neoantigens. [bib_ref] Prevalence of mismatch repair-deficient crypt foci in Lynch syndrome: a pathological study, Kloor [/bib_ref] [bib_ref] Mismatch repair-deficient crypt foci in Lynch syndrome-molecular alterations and association with clinical..., Staffa [/bib_ref] Previous studies have shown the relevance of immune cell recruitment from the bloodstream via HEVs into the site of inflammation for the successful immune response. [bib_ref] Treg depletion licenses T cell-driven HEV neogenesis and promotes tumor destruction, Colbeck [/bib_ref] [bib_ref] CD40L+CD4+memory T cells migrate in a CD62P-dependent fashion into reactive lymph nodes..., Martin-Fontecha [/bib_ref] [bib_ref] Evidence for recruitment of plasmacytoid dendritic cell precursors to inflamed lymph nodes..., Yoneyama [/bib_ref] [bib_ref] Tumor necrosis factor-dependent segmental control of MIG expression by high endothelial venules..., Janatpour [/bib_ref] [bib_ref] Effector lymphocyte-induced lymph node-like vasculature enables naive T-cell entry into tumours and..., Peske [/bib_ref] In Lynch syndrome patients, HEVs may facilitate the recruitment of FSP-specific T cells, including recently primed cytotoxic CD8 T cells 42 or naive T cells, which get activated on-site, from the blood to colon tissue affected by MMR-deficient lesions. This possibly contributes to their elimination by the local immune surveillance [bib_ref] Neogenesis and development of the high endothelial venules that mediate lymphocyte trafficking, Hayasaka [/bib_ref] and may explain the limited penetrance of Lynch syndrome, with only 50% life-time CRC risk, despite the presence of thousands of MMR-deficient crypt foci in Lynch syndrome mutation carriers. [bib_ref] Prevalence of mismatch repair-deficient crypt foci in Lynch syndrome: a pathological study, Kloor [/bib_ref] It has been shown that MMR-deficient crypt foci are present in normal colonic mucosa of Lynch syndrome patients, but not in sporadic MSI CRC patients, with cancers arising from right-sided BRAF-mutant serrated lesions. [bib_ref] Prevalence of mismatch repair-deficient crypt foci in Lynch syndrome: a pathological study, Kloor [/bib_ref] [bib_ref] DNA mismatch repair protein deficient non-neoplastic colonic crypts: a novel indicator of..., Pai [/bib_ref] [bib_ref] Role of the serrated pathway in colorectal cancer pathogenesis, Leggett [/bib_ref] A history of completely different precursor lesions in sporadic and hereditary MSI CRCs could be an explanation for a weaker immune activation in sporadic MSI CRCs compared with hereditary ones. Thus, it is conceivable that the lifelong autoimmunisation that takes place in the colonic mucosa of Lynch syndrome mutation carriers, but presumably not in sporadic MSI CRC patients, is responsible for the activation of the local immune microenvironment triggering HEV neogenesis via immune-activation signalling. The newly formed HEVs in turn recruit more lymphocytes, which may again induce HEV neogenesis, thereby constantly feeding the autoimmunisation loop. Such an auto-immunisation loop suggested to take place in Lynch syndrome-associated CRCs can be well integrated in the previously suggested self-amplifying loop between activated T cells and HEV neogenesis [fig_ref] Figure 5: Graphical summary [/fig_ref] , ultimately leading to possible elimination of precancerous or even cancerous lesions. [bib_ref] Treg depletion licenses T cell-driven HEV neogenesis and promotes tumor destruction, Colbeck [/bib_ref] We found significantly higher HEV densities in tumours with a B2M mutation in comparison with B2M-wild-type MSI CRCs. This association was observed even within the hereditary MSI group, which presented with the highest HEV counts among MSI CRCs. Previously, B2M mutations in MSI CRCs have been shown to correlate with a low number of FOXP3-positive T cells in the normal mucosa adjacent to tumour tissue, as well as with a high number of PD1-positive activated T cells in the tumour tissue, suggesting the role of active immune microenvironment in the induction of immune evasion. [bib_ref] High numbers of PDCD1 (PD-1)-positive T cells and B2M mutations in microsatellite-unstable..., Janikovits [/bib_ref] [bib_ref] Low density of FOXP3-positive T cells in normal colonic mucosa is related..., Echterdiek [/bib_ref] High HEV densities in B2M-mutant tumours corroborate these previous findings and imply that under strong immunoselective pressure created by immune cells recruited via HEVs, tumour cells which have lost MHC class I-associated antigen expression gain growth advantage, according to the immunoediting concept, 12 thus revealing a major role of HEVs in enhancing the immunoselective pressure on highly immunogenic cancers. Taken together with the high HEV counts in Lynch syndrome CRCs, our findings point towards a longer process of immunoediting in Lynch syndrome CRCs, possibly due to pre-existing MMR-deficient crypts, explaining the higher proportion of B2M mutations in the tumours from Lynch syndrome patients compared with the sporadic ones. [bib_ref] Beta2-microglobulin mutations in microsatellite unstable colorectal tumors, Kloor [/bib_ref] Taking into account the strong correlation of B2M mutations with better prognosis and low risk of distant metastasis in MSI CRCs, [bib_ref] Beta2-microglobulin mutations in microsatellite unstable colorectal tumors, Kloor [/bib_ref] [bib_ref] Microsatellite instability and Beta2-Microglobulin mutations as prognostic markers in colon cancer: results..., Tikidzhieva [/bib_ref] most likely explainable by lack of platelet binding essential for metastasis, [bib_ref] Aspirin prevents colorectal cancer metastasis in mice by splitting the crosstalk between..., Guillem-Llobat [/bib_ref] HEV density may represent another prognostic marker for survival of patients with MSI CRC, particularly in Lynch syndrome scenario. In fact, a high density of HEVs may also contribute to local immune surveillance and the limited metastatic potential of MSI cancers. [bib_ref] Microsatellite instability in colorectal cancer is associated with local lymphocyte infiltration and..., Buckowitz [/bib_ref] Studies analysing the functional impact of HEVs on metastasis formation are warranted.
In addition, HEVs could also potentially be of predictive value for treatment response towards immune checkpoint blockers (ICB). Previously, B2M mutations have been suggested to induce acquired resistance to ICB therapy. [bib_ref] Mechanisms of resistance to immune checkpoint inhibitors, Jenkins [/bib_ref] On the contrary, a recent study by Middha et al. demonstrated that patients with B2Mmutant CRCs can still achieve clinical benefit from ICB. [bib_ref] Majority of B2M-mutant and -deficient colorectal carcinomas achieve clinical benefit from immune..., Middha [/bib_ref] Our study did not address the predictive value of HEV densities for ICB therapy. However, B2M-mutant tumours in our study not only had elevated HEV densities, but also higher levels of PD-L1 expression, suggesting secretion of IFN-gamma as a result of a prolonged immune interaction, which in turn favoured the acquisition of B2M mutation, leading to loss of antigen presentation via MHC class I and immune escape. The implications of this finding with regard to the predictive value of HEVs for immunotherapy in MSI cancers should be addressed by future studies. It is important to note that our study has an observational design, and in order to draw functional conclusions regarding HEVs and their role in anticancer immunity, further studies using suitable animal models are required.
In conclusion, our data provide evidence that the elevated HEV densities reflect the active immune milieu in MSI colorectal tumours. HEVs may support recruitment of FSP-specific T cells to MSI cancer tissues, thereby contributing to a strong immuneselective pressure that favours the outgrowth of cell clones, which acquired immune evasion mechanisms. Our findings underline the impact of immune surveillance in cancer and corroborate the concept of immunoediting in tumour evolution.
[fig] Figure 3: Relation of HEVs and immune infiltration. Staining of the same tumour sample for CD20 (left), MECA-79 (middle) and CD3 (right) markers. CD20-positive cells clearly marked the germinal centre of the lymph follicle, whereas CD3-positive cells were co-localised with HEVs. Arrows indicate the examples of positive staining. Scale bars top to bottom CD20: 50 µm, 100 µm, 1 mm; MECA-79: 100 µm, 100 µm, 1 mm, 1 mm; CD3: 100 µm, 100 µm, 1 mm less informative taken alone, HEVs reflect more the functional status of the immune milieu in a tumour. [/fig]
[fig] Figure 4: Association of HEV densities with immune response and immune evasion markers in MSI CRCs. [/fig]
[fig] Figure 5: Graphical summary. Model of auto-immunisation loop in Lynch syndrome. In Lynch syndrome, the host immune system can be primed against FSP neoantigens by precursor lesions, MMR-deficient crypts, existing in the normal colonic mucosa long before tumour manifestation. Recurrent exposure of the immune cells to FSP neoantigens expressed in MMR-deficient crypts leads to the generation of an FSP-specific T-cell pool. When a tumour arises, FSP-specific T cells can promptly recognise the FSP neoantigens expressed on the tumour cells and release cytokines, which could induce neogenesis of HEVs. New HEVs, in turn, can recruit more naive T cells, which would undergo maturation through exposure to the neoantigens into FSP-specific T cells. Newly formed FSP-specific T cells release more cytokines, potentially leading to generation of more HEVs. This self-amplifying loop keeps MSI tumour cells under constant immune pressure and promotes selection and outgrowth of cell clones lacking antigen expression via MHC class I via acquisition of B2M mutations, enabling immune escape. Thereby, HEVs can contribute to immunoediting of MSI tumour cells. [/fig]
|
A Novel V2 Vasopressin Receptor Mutation with X-Linked Nephrogenic Diabetes Insipidus
[bib_ref] Requirement of human renal water channel aquaporin-2 for vasopressin-dependent concentration of urine, Deen [/bib_ref] [bib_ref] Nephrogenic diabetes insipidus, Morello [/bib_ref] [bib_ref] Two novel aquaporin-2 mutations in a sporadic Japanese patient with autosomal recessive..., Tajima [/bib_ref] [bib_ref] Molecular biology of hereditary diabetes insipidus, Fujiwara [/bib_ref] [bib_ref] Nephrogenic diabetes insipidus, Morello [/bib_ref] [bib_ref] Molecular biology of hereditary diabetes insipidus, Fujiwara [/bib_ref] [bib_ref] Three novel AVPR2 mutations in three Japanese families with X-linked nephrogenic diabetes..., Tajima [/bib_ref]
## Patient report
A 14-day-old Japanese boy was referred to our hospital because of low grade fever, failure to thrive, and lethargy. He was born after a 38wk uneventful gestation. His birth weight was 3378 g. The patient was the first child of apparently healthy 25-yr-old parents, although two maternal uncles died in early infancy due to unknown reasons. On admission, his weight was 3672 g. His urine volume was not decreased despite an inadequate fluid intake. Blood test revealed hypernatremia (158 mEq/L), blood urea nitrogen (20 mg/dl) plasma hyperosmolality (306 mOsm/kg) and urine hypoosmolality (91mOsm/ kg). Polyuria continued (2000-3000 ml/m 2 ) and urine specific gravity remained extremely low (1.001-1.002). His plasma ADH was high (38.6 pg/ml, normal range 4-12 pg/ml).
An intramuscular injection of 2.5 units did not increase urine osmolality (before 111 mOsm/kg, after 122 mOsm/kg) nor did it change plasma osmolality (before 312 mOsm/kg, after 324 mOsm/kg). Based on these findings, the boy was diagnosed as having congenital NDI. Treatment with hydrochlorothiazide (1 mg/kg/ day) was initiated and partial improvement in polyuria (a reduction of 20%) was observed.
The parents of the boy gave their informed consent to participate in the study, and genomic DNA from the patient and his mother was extracted from peripheral leukocytes. The polymerase chain reaction for the V2 receptor gene followed by direct sequencing for the gene was performed according to previous reports [bib_ref] Three novel AVPR2 mutations in three Japanese families with X-linked nephrogenic diabetes..., Tajima [/bib_ref].
A four base insertion (ins855/856CGCA) was identified in the V2 receptor gene in the patient [fig_ref] Figure 1: Sequence analysis of the V2 receptor gene [/fig_ref]. His mother was shown to be heterozygous for the mutation [fig_ref] Figure 1: Sequence analysis of the V2 receptor gene [/fig_ref].
# Discussion
We identified one novel insertion mutation of the V2 vasopressin receptor gene in a Japanese boy with X-linked NDI. This insertion mutation changes the open reading frame after the mutation site and causes amino acid substitutions in the C-terminal of the V2 receptor. Thus, the function of the mutant receptor might be impaired.
As mentioned, many mutations have been reported in patients with X-linked NDI. The mutations are scattered throughout the gene and there is no hot spot in the V2 receptor gene . Thus, it is required to search for the molecular defect of the V2 receptor gene in each individual. We did not analyze the patient's grandmother. From family history, her maternal uncles were likely to have been affected and severe dehydration might have been the causes of their deaths in early infancy. This strongly indicates that early diagnosis and adequate therapy are necessary to prevent death and physical and mental retardation in congenital NDI patients [bib_ref] Nephrogenic diabetes insipidus, Morello [/bib_ref] [bib_ref] Molecular biology of hereditary diabetes insipidus, Fujiwara [/bib_ref]. For that purpose, molecular diagnosis is a useful method. Not only can the method confirm the diagnosis in suspected individuals, but also in female carriers. Of particular importance is the diagnostic potential of the method in newborn children who are at risk of inheriting this disease.
In conclusion, a new insertion mutation of the V2 receptor gene was identified. We should keep in mind that congenital NDI causes severe sequelae if not diagnosed properly. Molecular diagnosis is useful for early definitive diagnosis and genetic counseling.
[fig] Figure 1: Sequence analysis of the V2 receptor gene. (a) The patient had a 4 base insertion (855/856insCGCA) denoted by the arrow and underline. (b) His mother had this mutation as a heterozygous state. Note overlapping signals after the insertion site. (c) The sequence of the wild-type gene. The numbers refer to the nucleotides in the genomic sequence in GeneBank accession no L22206; the initial base of the initiation codon is denoted as nucleotides 1. [/fig]
|
Association of Wealth With Longevity in US Adults at Midlife
[bib_ref] Between-within models for survival analysis, Sjölander [/bib_ref] [bib_ref] Regression models for twin studies: A critical review, Carlin [/bib_ref] [bib_ref] Behavior Genetic Research Methods: Testing Quasi-Causal Hypotheses Using Multivariate Twin Data, Turkheimer [/bib_ref] [bib_ref] Causal inference and observational research: The utility of twins, Mcgue [/bib_ref]
## Emethods 2. supplementary sensitivity analyses
Sensitivity analyses were undertaken in the combined sibling and twin subsample to test the robustness of the findings to different model specifications. First, to assess the potential for a non-linear association between net worth and longevity, we estimated a spline model (eTable 1) including two knots: one at the 75 th percentile of net worth ($125,000) and one at the 90 th percentile ($382,500). As shown in eTable 1, HRs for net worth for those below the 75 th percentile (HR=.86, CI=.76-.97, p=.01) and between the 75 th -90 th percentiles (HR=.90, CI=.82-.98, p=.01) were similar, and a subsequent test of the estimates confirmed that their difference was no greater than chance (p=.66). The HR for net worth for those above the 90 th percentile (individuals with >=$680,000; n=182, 45 decedents) was not statistically significant (HR=1.02, 95%CI=.97-1.08, p=.31) and a test of the estimates indicated that the difference between this HR and the HR for those between the 75 th -90 th percentiles was statistically significant (p=.03).
This indicates a possible diminished return on net worth at the very high end of the net worth distribution, though this reflects only approximately 7% of the sample. Collectively, the spline model indicates that among the large majority (93%) of siblings and twins (i.e. those whose net worth was <= $382,000), the association between net worth and survival was approximately linear.
We subsequently reran Model 3 in a subsample of siblings and twins who were in family groups where all members had <= $382,500 in net worth (n=2,110; 321 deaths). As shown in eTable 2 of this supplement, the HR for net worth in this restricted sample (HR=0.89, CI=0.82-0.96, p=0.004) suggested a larger association between net worth and longevity among those with lower family-level wealth. However, these results should be interpreted with a high level of caution due to the restricted sample size.
In another sensitivity analysis, we recoded net worth into ordinal decile groups, given the large positive skew of the net worth distribution. In the combined sibling and twin subsample, between-family (HR = 0.90, 95% CI = 0.85-0.94, p < .001) and within-family (HR = 0.92, 95% CI = 0.87-0.96, p = .001) net worth estimates remained significant predictors of mortality.
Next, to account for the possibility of residual confounding by health status (having a medical problem may both reduce one's ability to accumulate wealth and increase mortality risk), analyses were re-estimated among sibling/twin pairs who were free of previous cancer or heart disease. Among sibling groups with >2 members, only those siblings without heart disease and cancer were compared to one another. Results were largely similar in this restricted sample We also tested the possibility of a nonlinear age trend by including an age 2 term (HR =1.00, 95% CI = 0.99-1.00, p = .65) and also an age*sex interaction term (HR = 1.01, 95% CI = 0.99-1.03, p = 0.14), neither of which was associated with mortality risk nor changed the interpretations of other model estimates. We also tested an interaction between the withinfamily net worth estimate and participant age at MIDUS 1. The rationale being that the withinfamily association between net worth and longevity may vary as a function of age. The interaction term was not statistically significant, HR = 1.00, 95% CI =0.99-1.00, p=0.37,
[formula] © 2021 Finegood ED et al. JAMA Health Forum. [/formula]
suggesting that the within-family association between net worth and longevity did not vary by age at MIDUS 1.
Lastly, as a more conservative test of possible confounding by early experience, we
## Emethods 3. deviations from original preregistered analysis plan
The analyses presented in the main text are consistent with the study rationale, hypotheses, and overall analytic plan outlined in our preregistration (https://osf.io/zyedp). We did, however, deviate from the original analytic plan in some minor ways-these deviations are outlined below.
First, in the original analysis plan, we specified that survival analyses would be estimated as multi-level Cox regression models. Instead, each survival model was estimated as a Cox model with shared frailty term to account for clustering. Both analytic approaches gave very similar results for all models-compare coefficients in Model 1 and Model 3 of Instead, in the main text, we present results from three separate survival models: one among non-twin siblings, one among DZ twins, and one among MZ twins. With either analytic strategy, coefficients were very similar and provided the same interpretations-compare coefficients from Models 4, 5, and 6 in main text to coefficients in eTable 5, which display results from the multilevel mixture model using Cox regression with KNOWNCLASS option in Mplus version 8).
Third, as described in the main text, we undertook a conservative test of whether the within-family net worth effect varied across full sibling, MZ twin, and DZ twin subsamples by including two-way interaction terms between the within-family net worth variable and dummy codes for cluster type and conducting a Wald test to test the equality of the within-family estimates. This test of the equality of within-family net worth effects across sibling subsamples was not described in the original analysis plan.
We also originally proposed an exploratory test of the two-way interaction between parent education and the within-family net worth twin/sibling difference score. We decided to omit this test from the main text because we felt that the rationale for this test was tangential to our primary analyses, which were already many in number. It is also not an optimal test for answering questions related to social mobility, its stated purpose in the preregistration. In any case, we report results from this test here. Among the combined sample of non-twin siblings, DZ twins, and MZ twins (n =2,490) there was no interaction between parents' highest level of education and sibling/twin net worth difference score (HR = .99, 95% CI .98-1.00, p = .30).
Indeed, when the sample is split into subsamples of those at or below (n = 1,333) and above (n = 1,157) the median on parent education, the within-family net worth estimate was significant in both subsamples (at/below median: HR within =.95, 95% CI .91-.99, p = .04; above median: HR within =.94, 95% CI .89-.98, p = .006).
We also specified that full-information maximum likelihood estimation (FIML) would be used to handle missing data. In the preregistration, we specified the criteria by which siblings and twin pairs would be excluded from our analysis sample-resulting in an analytic sample of n=7,017.
[fig] © 2021 Finegood: ED et al. JAMA Health Forum. [/fig]
[fig] 196: deaths): HR between = 0.95, 95% CI= 0.90 -0.99, p = 0.04, and HR within = 0.94, 95% CI = 0.90-0.98, p = 0.01). [/fig]
|
Concise and Straightforward Asymmetric Synthesis of a Cyclic Natural Hydroxy-Amino Acid
An enantioselective total synthesis of the natural amino acid (2S,4R,5R)-4,5-dihydroxy-pipecolic acid starting from D-glucoheptono-1, 4-lactone is presented. The best sequence employed as a key step the intramolecular nucleophilic displacement by an amino function of a 6-O-p-toluene-sulphonyl derivative of a methyl D-arabino-hexonate and involved only 12 steps with an overall yield of 19%. The structures of the compounds synthesized were elucidated on the basis of comprehensive spectroscopic (NMR and MS) and computational analysis.
# Introduction
Cyclic hydroxy-amino acids are adequate building blocks for the development of new therapeutic drug candidates [bib_ref] An efficient synthesis of enantiomerically pure (R)-pipecolic acid, (S)-proline, and their N-alkylated..., Fadel [/bib_ref] [bib_ref] Design of MHC Class II (DR4) ligands using conformationally restricted imino acids..., Hanson [/bib_ref]. These compounds are β-turn inducers in the synthesis of peptides [bib_ref] Modelling and synthesis of conformationally resticted amino acids, Galeazzi [/bib_ref] and good enzyme inhibitors by themselves [bib_ref] Synthesis of all stereoisomers of 3-hydroxypipecolic acid and 3-hydroxy-4,5-dehydropipecolic acid and their..., Ohara [/bib_ref] as are several of their derivatives, such as deoxy-galactono-jirimicin derivatives [bib_ref] DLHex-DGJ, a novel derivative of 1-deoxygalactonojirimycin with pharmacological chaperone activity in human..., Fantur [/bib_ref] [bib_ref] A stereodivergent approach to 1-deoxynojirimycin, 1-deoxygalactonojirimycin and 1-deoxymannojirimycin derivatives, Boucheron [/bib_ref] [bib_ref] 1-Deoxygalactonojirimycin-lysine hybrids as potent d-galactosidase inhibitors, Steiner [/bib_ref]. Indeed, stereoselective synthesis of hydroxy-amino acids is a field that has been growing during the last decades because of the potential biological activity of some representatives of this class of compounds, mainly as enzyme inhibitors [bib_ref] Synthesis of all stereoisomers of 3-hydroxypipecolic acid and 3-hydroxy-4,5-dehydropipecolic acid and their..., Ohara [/bib_ref] [bib_ref] Synthesis of novel potent hepatitis C virus NS3 protease inhibitors: Discovery of..., Thorstensson [/bib_ref] [bib_ref] Novel 3-amino-2-hydroxy acids containing protease inhibitors. Part 1: Synthesis and kinetic characterization..., Stöckel-Maschek [/bib_ref] [bib_ref] A glucuronidase inhibitor from the seeds of Baphia racemosa: Application of fast..., Manning [/bib_ref] [bib_ref] A concise synthesis of (2R,3R)-and (2R,3S)-3-hydroxypipecolic acids, and total synthesis of (-)-deoxoprosopinine..., Mallick [/bib_ref]. Particularly, conformationally constrained hydroxy-amino acids such as the hydroxylated pipecolic acids are very interesting target molecules [bib_ref] A concise synthesis of (2R,3R)-and (2R,3S)-3-hydroxypipecolic acids, and total synthesis of (-)-deoxoprosopinine..., Mallick [/bib_ref] [bib_ref] Synthesis of homochiral β-hydroxy-α-aminoacids, Fleet [/bib_ref]. For instance, the compound (2S,4R)-4-hydroxy-pipecolic acid is a natural product that has been isolated from Calliandra pittieri and Stophantus scandeus [bib_ref] Cis-4-hydroxypipecolic acid and 2,4-cis-4,5-trans-4,5-dihydroxypipecolic acid from Calliandra, Romeo [/bib_ref] , and the cis-5-hydroxy-substituted pipecolic acid skeleton is frequently found in alkaloids from microorganisms and also in plants (febrifugine, pseudoconhidrine) [bib_ref] l-cis-5-hydroxypipecolic acid from seeds of Gymnocladus dioicus, Despontin [/bib_ref] [bib_ref] Cis-5-hydroxy-L-pipecolic acid from Morus alba and Lathyrus japonicus, Hatanaka [/bib_ref]. Furthermore, a tri-hydroxy-pipecolic acid isolated from the seeds of Baphia racemosa proved to have specific human liver beta-glucosidase inhibitory activity [bib_ref] A glucuronidase inhibitor from the seeds of Baphia racemosa: Application of fast..., Manning [/bib_ref]. A couple of years ago a stereoselective synthesis of N-Boc-protected cis-(2R,3S)-3-hydroxy-pipecolic acid, starting from D-glucose, was described [bib_ref] A regioselective reductive cleavage of benzylidene acetal: Stereoselective synthesis of N-Boc-protected cis-(2R,3S)-3-hydroxy..., Kumar [/bib_ref]. In the literature some synthesis of 5-hydroxy-pipecolic acid derivatives and their analogues is also described [bib_ref] Synthesis of enantiomerically pure (2R, 5S)-and (2R, 5R)-5-hydroxypipecolic acid from glycinate schiff..., Hoarau [/bib_ref] [bib_ref] Diastereoselective synthesis of (2S,5R)-5-hydroxypipecolic acid and 6-substituted derivatives, Botman [/bib_ref] [bib_ref] Synthesis of 5-substituted pipecolic acid derivatives as new conformationally constrained ornithine and..., Le Corre [/bib_ref]. A record of an enantioselective synthesis of (2R,4R)-and (2S,4R)-4-hydroxy-pipecolic acids from commercial ethyl (R)-4-cyano-3-hydroxy-butanoate was published [bib_ref] A short and convenient synthesis of enantiopure cis-and trans-4-hydroxypipecolic acid, Occhiato [/bib_ref] , while the synthesis of the latter compound was also performed starting from D-glucoheptono-1,4-lactone [bib_ref] Enantioselective synthesis of (2R,4S)-4-hydroxy-pipecolic acid from D-gluconoheptono-1,4-lactone, Di Nardo [/bib_ref]. Moreover, recently two efficient and complementary routes have been described for the synthesis of 4,5 di-hydroxy-pipecolic acids from cis (4S,5R)-and (4R,5S)-4,5-dihydroxy-γ-valerolactams [bib_ref] Complementary and stereodivergent approaches to the synthesis of 5-Hydroxy-and 4,5-dihydroxypipecolic acids from..., Scarpi [/bib_ref]. In this work we describe another short, convenient and enantioselective synthesis of (2S,4R,5R) 4,5 di-hydroxy-pipecolic acid starting from 2 as chiral template. This target cyclic amino acid is a natural product that has been isolated from the leaves of Julbernardia paniculata [bib_ref] 5-Dihydroxypipecolic acids in the seed of Julbernardia isoberlinia and Brachystegia, Shewry [/bib_ref] and synthesized from the precursor L-baikiaine, isolated from the aerial parts of Derris eliptica [bib_ref] 2S-Carboxy-4R,5S-dihydroxypiperidine et 2S-carboxy-4S,5S-dihydroxypiperidine a partir de Derris elliptica, Marlier [/bib_ref].
Mario Simirgiotis has worked previously on the synthesis of small molecules starting from sugars as chiral templates [bib_ref] Enantiospecific synthesis of the sugar amino acid (2S,5S)-5-(aminomethyl)-tetrahydrofuran-2-carboxylic acid, Uhrig [/bib_ref] and embarked on the synthesis of this compound several years ago (under the mentorship of Dr. Oscar Varela, at the University of Buenos Aires, Argentina), but the results were unpublished.
# Results and discussion
## Retrosynthetic analysis for the synthesis of compound 1
We reasoned (Scheme 1) that 1 can be synthesized by imine hydrogenation after direct carbonyl addition by the amino group at C-2 (synthon I), a strategy that was previously employed for the synthesis of (2S,4R) 4-hydroxypipecolic acid [bib_ref] Enantioselective synthesis of (2R,4S)-4-hydroxy-pipecolic acid from D-gluconoheptono-1,4-lactone, Di Nardo [/bib_ref]. Synthon I can readily be obtained through O-protection of α amino lactone II. Key intermediate II can be in turn obtained from commercially available 2 as a chiral template via a series of elimination of the HO-3 subsequent oxidation, and conversion of the HO-2 into NH2.
## Total synthesis of compound 1
We started with commercially available and inexpensive D-glycero-D-gulo-heptono-1,4-lactone (2) using a previously published methodology [bib_ref] Synthesis of polyhydroxy amino acids based on D-and L-alanine from D-glycero-D-gulo-heptono-1,4-lactone, Gómez [/bib_ref] , which, on peracetylation and β-elimination of the acetate group on C-3, followed by hydrogenation, O-deacetylation, and isopropylidenation of the resulting aldonolactone with 2,2-dimethoxypropane, yielded 3-deoxy methyl ester 3, which has the desired (4R,5R) configurations (Scheme 2). The free hydroxyl group at C-2 of 3 was sulfonylated with tosyl chloride in pyridine to give the tosylate derivative 4 in 12 h with 96% yield. Spectroscopic data for compound 4 resembled the distorted planar zigzag conformational preference in solution as previously reported for compound 3 [bib_ref] Synthesis of polyhydroxy amino acids based on D-and L-alanine from D-glycero-D-gulo-heptono-1,4-lactone, Gómez [/bib_ref] with an anti disposition for H-5 and H-6 (J5, [bib_ref] DLHex-DGJ, a novel derivative of 1-deoxygalactonojirimycin with pharmacological chaperone activity in human..., Fantur [/bib_ref] Treatment of tosylate 4 with sodium azide in DMF produced the desired inversion of C-2 configuration affording azide 5 in 87% yield. Selective hydrolysis of the isopropylidene terminal group of 5 yielded diol 6, which on degradative oxidation of the 1, 2 diol system using sodium periodate produced aldehyde 7. Aminolysis of 7 in a "one pot" procedure (Scheme 3, [fig_ref] Figure 1: Minimum energy conformations, activation energy, and transition state for the intramolecular nucleophilic... [/fig_ref] yielded key compound 8 through a sequence of catalytic hydrogenation of the azide group, nucleophilic attack of the resulting amine on the aldehyde (7a), dehydration of the resulting alcohol 7b to the imine 7c, and reduction of the imine 7c to give the expected piperidine 8, which crystallized from hexane-EtOAc 10:1 (mp 119 °C, D −47.62). In order to demonstrate this reaction and try to understand the low yield (27%) obtained, an Internal Reaction Coordinate (IRC) calculation was performed. An IRC calculation allows one to map out a reaction pathway by integrating the intrinsic reaction coordinate. This method examines the reaction pathway proposed, leading to a transition state in a map of the potential energy surface. The results of this calculation are showed in [fig_ref] Figure 1: Minimum energy conformations, activation energy, and transition state for the intramolecular nucleophilic... [/fig_ref]. The transition state (TS) was found with an imaginary frequency, demonstrating that this is, as expected, a first-order and six-membered cyclic TS. Furthermore, an intense cross-peak observed between H-6' and H-2 confirmed the S configuration for C-2. The correlations were corroborated by HMQC and HMBC spectra. Density Functional Theory (DFT) calculations (Gaussian 9.0)indicated that compound 8 has two minimum-energy conformations (boat and chair) but the boat conformation ( 4 BN) is 2.8 KJ/mol more stable , which is in concordance with the spectroscopic data found for this compound. The isopropylidene group of amino-acid 8 was removed by stirring with trifluoracetic acid-water (1:1) for 8 h at room temperature, while removal of the remaining ester function at C-1 was achieved by subsequent alkaline treatment with NaOH 1 M to yield the target cyclic amino acid 1. Scheme 3. One pot synthesis of piperidine 8 from azide 7.
[formula] H 2 , Pd/C 8. O O NH 2 O O O H N HO O O O O H 2 , Pd/C N O O O O -H 2 O 7a 7b 7c [/formula]
The preparation of key compound 8 from 7 (Scheme 3) resulted in a disappointing 27% yield. We assayed different concentrations of catalyst, solvent, hydrogen pressure, and reaction times [fig_ref] Table 1: Preparation of compound 8 by catalytic hydrogenation of aldehyde 7 [/fig_ref] , but in all cases after consumption of all starting material (monitored by TLC) we obtained compound 8 (yields below 27%) and a mixture of polar compounds, according to 1 NMR a mixture of adducts produced by intermolecular addition of the amine group at C-2 of one molecule of 7 to the aldehyde function at C-6 of another molecule of 7. We assumed that the low yield obtained for this reaction could be attributed in part to the partial conformational impediment [fig_ref] Figure 1: Minimum energy conformations, activation energy, and transition state for the intramolecular nucleophilic... [/fig_ref] produced by the isopropylidene group for the nucleophilic addition of the C-2 amine group to the aldehyde.
## Synthesis of compound 1 starting from azide 5
The low yield obtained for key compound 8 from 7 prompted us to use the alternative synthetic sequence depicted in Scheme 4. Catalytic hydrogenation of 5 in EtOAc generated the amino group that was in turn protected with benzyl chloroformate to obtain carbamate 9 in 88% yield. Furthermore, the benzyloxycarbonyl was a convenient protective group for the amine group of this small class of sugar compounds, since a very stable carbamate derivative can be obtained and the protecting group can be easily removed by hydrogenolysis in high yields and short reaction times [bib_ref] Enantioselective synthesis of (2R,4S)-4-hydroxy-pipecolic acid from D-gluconoheptono-1,4-lactone, Di Nardo [/bib_ref]. We considered that 10 was a convenient precursor of 1 that could be degraded to the aldehyde 11 and this compound could be properly reduced to the primary alcohol, which also has the right configuration and hydroxyl group for the functionalization with a convenient leaving group at C-6 to be displaced by the amino group at C-2, after removal of the benzyloxycarbonyl group. Thus, selective hydrolysis of the terminal isopropylidene group with a mixture of acetic acid-water at 50 °C over 12 h produced 10 in 82% yield, which could be converted to 11 using the procedure applied previously in the synthesis of 7, followed by reductive amination with NaBH3CN at pH = 4 in 4 h with 91% yield. The tosylate 12 was then prepared as explained above to generate the necessary leaving group for the intramolecular nucleophilic displacement by the amino group. Deprotection of the amine group of 12 was performed with 10% palladium on charcoal under hydrogen atmosphere (45 psi) in EtOAc for 12 h, and after filtration and evaporation of the solvent, the isopropylidene and ester groups of the latter compound were removed by hydrolysis with trifluoracetic acid: water (1:1). The resulting crude product, after evaporation of the solvent, was dissolved in water and treated with K2CO3 for 2 h to produce the nucleophilic amine group necessary for the ring closure (Scheme 4 and [fig_ref] Scheme 4: Synthesis of compound 1 from azide 5 [/fig_ref]. As shown in [fig_ref] Scheme 4: Synthesis of compound 1 from azide 5 [/fig_ref] , the activation energy for the closure of intermediate 12a (reactant) into 1 (product) is almost half that for the intramolecular nucleophilic attack of amine on the aldehyde of intermediate 7a to obtain 8 [fig_ref] Figure 1: Minimum energy conformations, activation energy, and transition state for the intramolecular nucleophilic... [/fig_ref]. This can explain in part the different yield obtained for the mentioned reactions.
After 5 h of reaction only one product reactive to ninhydrin was obtained, and the solution was filtrated, acidified to pH = 6, and loaded onto an ion exchange column (Dowex 50W-H + resin), which was rinsed with water and eluted with pyridine 0.1 N to obtain the target syrupy pipecolic amino-acid 1 with an [α]D −5.4, and HRESIMS: [M+H] + = 162.0763, without epimerization in 83% yield. [bib_ref] Synthesis of homochiral β-hydroxy-α-aminoacids, Fleet [/bib_ref] NMR spectra for 1 showed the expected six carbons, including one carboxylic acid (δ 173.35), three oxygenated methynes (C-2, C-4, and C-5 at δ 53.5, 65.4, and 64.9, respectively), and two methylene carbons (C-3 and C-6, at δ 28.4 and 44.2). All assignments were performed by careful analysis of the COSY, NOESY, HMQC, and HMBC spectra. The 1 H-NMR spectral data for this compound was published early in 1976 [bib_ref] 2S-Carboxy-4R,5S-dihydroxypiperidine et 2S-carboxy-4S,5S-dihydroxypiperidine a partir de Derris elliptica, Marlier [/bib_ref] , but recorded using a 100 MHz equipment. As in the 1 H-NMR spectra of pipecolic derivative 8, coupling constants for H-2 (J2,3ax = 11.4, J2,3eq = 4.2 Hz) for H-3ax (J3ax, 3eq = 14.9, J3ax,4 = 2.8 Hz) and H-3eq (J2,3eq = J3eq,4 ≈ 4.2) confirmed the S, R, and R configuration for C-2, C-4, and C-5, respectively, while NOE cross peaks between H-5-H-6', H-2 with H6', and H-3ax with H-5 and H-4 suggested the chair conformation 4C N . The optimized structure of compound 1 [fig_ref] Figure 5: Minimum energy conformation of target compound 1 calculated by DFT [/fig_ref] was built into the Gaussian 9.0 platform and the energy minimized by the DFT method, which coincides with the NMR data obtained and configuration proposed. High resolution NMR (see Experimental section) and ESI-MS spectra reported in this work added more accurate information about this natural product structure, for which the NMR data were published a long time ago [bib_ref] 2S-Carboxy-4R,5S-dihydroxypiperidine et 2S-carboxy-4S,5S-dihydroxypiperidine a partir de Derris elliptica, Marlier [/bib_ref].
## Experimental
## General experimental procedures
Melting points were determined using a Stuart Scientific SMP3 melting point apparatus (Bibby Scientific Ltd, Staffordshire, UK). Analytical thin-layer chromatography (TLC) was performed on silica gel 60 F254 (Merck, Darmstadt, Germany) aluminum-supported plates (layer thickness 0.2 mm). Medium pressure column chromatography was performed with Silica gel (Kieselgel 60 H Merck, Darmstadt, Germany), 55 mm particle size, FMI QG 150 medium pressure lab pumps (Syosset, NY, USA) and Ace Glass Inc. medium pressure columns (Vineland, NJ, USA), using mixtures of n-hexane:ethyl acetate of different polarities as solvent system (flow rate: 5 mL/min.). Optical rotations were measured on an Autopol III automatic polarimeter (Rudolph Research Co., Hackettstown, NJ, USA). The NMR experiments ( 1 H: 400.12 and 600.13 MHz; 13 C: 100.25 and 150.09 MHz) were performed using either a Bruker Avance 400 or Bruker Avance II (Biospin, Rheinstetten, Germany) 600 UltraShield spectrometer with CD3OD or deuterated MeOD as solvent and TMS as internal standard. Optical rotations were measured on an Autopol III automatic polarimeter (Rudolph Research Co., Hackettstown, NJ, USA). IR spectra were measured using a Thermo Nicolet Nexus 470 FT-IR spectrometer (Thermo Nicolet, Madison, WI, USA) with KBr disks. The molecular weight was determined by low and high resolution with a mass spectrometer (Finnigan Mat 900 XLT, Thermo Fisher, Bremen, GmbH, Germany). Another mass spectrometer equipped with electrospray ion source and qToF analyzer MicrOTOF Q II (Bruker Daltonics Inc., Billerica, MA, USA) was used for HR-ESI-MS analysis. All reagents used for reactions were of analytical grade and purchased from Merck (Santiago, Chile) or Sigma-Aldrich (Santiago, Chile).
## Synthesis of methyl 3-deoxy-4,5:6,7-di-o-isopropylidene-2-o p-toluen-sulfonyl-d-gluco-heptonate (4)
To a stirred solution of compound 3 (1.38 g, 4.54 mmol) in anhydrous pyridine (2.
## Synthesis of methyl 2-azido-2,3-dideoxy-4,5:6,7-di-o-isopropylidene-d-manno-heptonate (5)
Compound 5 was obtained with 98% yield (0.68 g) from compound 4 (0.99 g, 2.16 mmol), as previously reported [bib_ref] Synthesis of polyhydroxy amino acids based on D-and L-alanine from D-glycero-D-gulo-heptono-1,4-lactone, Gómez [/bib_ref]. Physical properties and spectroscopic-spectrometric data for compound 5 were identical to those previously reported [bib_ref] Synthesis of polyhydroxy amino acids based on D-and L-alanine from D-glycero-D-gulo-heptono-1,4-lactone, Gómez [/bib_ref].
## Synthesis of methyl 2-azido-2,3-dideoxy-4,5-o-isopropylidene-d-manno-heptonate (6)
A solution of 5 (0.68 g, 2.06 mmol) in acetic acid (25 mL) and water (0.3 mL) was stirred at 50 °C for 12 h. At the end of this time, TLC (1:2, hexane/EtOAc) monitoring showed complete conversion of 5 (Rf = 0.87) into a more polar product (Rf = 0.30). The mixture was concentrated under reduced pressure and the residue purified by medium pressure chromatography, using as eluent mixtures of hexane/EtOAc (from 5:1 to 1:7) to yield 6 (0.56 g, 94%).
## Synthesis of (2s,4r,5r) 4,5-o-isopropylidene-methyl-pipecolate (8)
A solution of 6 (0.56 g, 1.94 mmol) in MeOH (32 mL) was added to sodium periodate (0.48 g, 2.27 mmol) and stirred at room temperature (25 C) for 3 h. At the end of this time, TLC (1:1, hexane/EtOAc) monitoring showed complete conversion of 6 (Rf = 0.12) into a less polar product (7, Rf = 0.52). The reaction mixture was filtered, concentrated, and the syrupy residue dissolved in EtOAc (8 mL) with palladium black (10% Pd on carbon, 0.05 g); the sample was then hydrogenated at 45 psi until TLC (1:1, hexane/EtOAc) revealed total consumption of the starting material. After this time (12 h), the catalyst was removed by filtration with celite (2 g). The reaction mixture was concentrated and the residue purified by medium pressure chromatography using as eluent mixtures of hexane/EtOAc (from 5:1 to 1:7) to yield 8, which crystallized from hexane-EtOAc 10:1 (0.11 g, 27%), mp 119 °C, 3.6. Synthesis of (2S,4R,5R) 4,5 Dihydroxy-pipecolic Acid (1) from 8 A solution of 8 (0.11 g, 0.51 mmol) was stirred into 2 mL trifluoroacetic acid-water (1:1), for 8 h at room temperature, when TLC analysis (1:1, hexane/EtOAc) showed consumption of all starting material and only one product with Rf = 0.4. Afterwards, the crude product was evaporated in vacuo and the residue dissolved in MeOH (1mL), and NaOH 0.1 M (1 mL) was added. TLC analysis (MeCN/EtOH/H2O/AcOH, 13:4:2:1, reactive to ninhidrine) showed only one blue spot with Rf = 0.2 after 5 h. The reaction mixture was concentrated, dissolved in water (1 mL) acidified with trifluoracetic acid (to pH = 6), and submitted to ion exchange chromatography with a Dowex 50W (H + ) resin column (5 cm × 1 cm).The column was rinsed with water (5 mL) and eluted with pyridine 0.1 N (10 mL) to obtain oily 1 (0.06 g, 83%).
## Synthesis of methyl 2-benzyloxycarbonyl-amino-2,3-dideoxy-4,5-o-isopropylidene-d-arabinohexonate (11)
To a solution of 10 (0.86 g, 2.16 mmol) in MeOH (55 mL) was added sodium periodate (0.74 g, 3.46 mmol); this was stirred at room temperature (25 °C) for 4 h. When TLC (1:2, hexane/EtOAc) showed complete consumption of 10 and conversion into a less polar product (Rf = 0.35), the white solid formed was filtered, sodium cyanoborohydride (0.20 g, 3.18 mmol) was added, and a solution of HCl 0.1 N was added dropwise to maintain pH = 5. After 1 h of stirring at 0 °C, a main spot with Rf = 0.45 was detected by TLC (1:2, hexane/EtOAc). The solution was concentrated under reduced pressure, the salts filtere,d and the residue purified by flash chromatography using mixtures of hexane/EtOAc (from 10:1 to 1:2) as eluent to give 11 (0.78 g, 91%). (12) To a stirred solution of compound 11 (0.78 g, 2.13 mmol) in anhydrous pyridine (8.2 mL) was added p-toluen-sulfonyl chloride (0.56 g, 2.94 mmol). After stirring at rt for 12 h, TLC (1:1, hexane/EtOAc) monitoring showed complete conversion of 11 (Rf = 0.45) into a less polar product (Rf = 0.75); the mixture was then concentrated under reduced pressure and the residue purified by flash chromatography (hexane/EtOAc from 20:1 to 5:1), to afford oily 12 (0.97 g, 88%).
## Synthesis of methyl 2-benzyloxycarbonyl-amino-2,3-dideoxy-4,5-o-isopropylidene-6-o-ptoluen-sulfonyl-d-arabino-hexonate
## Synthesis of compound 1 ((2s,4r,5r) 4,5 dihydroxy-pipecolic acid) from 12
A solution of 12 (0.97 g, 1.86 mmol) in EtOAc (17 mL) was added to 10% Pd on carbon (0.1 g) and hydrogenated at 45 psi. When TLC (1:1, hexane/EtOAc) revealed total consumption of the starting material (12 h), the catalyst was removed by filtration with celite (2 g). The crude product was evaporated in vacuo and to the residue we added trifluoracetic acid (3.7 mL) and water (3.7 mL), stirred overnight at rt for 12 h. After this time, the solution was evaporated to dryness, and water (10 mL) and K2CO3 (0.276 g, 2 mmol) were added. After 2 h of stirring, TLC (MeCN/EtOH/H2O/AcOH, 13:4:2:1) revealed total conversion of the previous compound (orange spot, Rf = 0.3) to a more polar one (blue spot reactive to ninhidrine, Rf = 0.12). The solution was evaporated to dryness, diluted with acidified water (to pH ≈ 5), submitted to ion exchange chromatography (Dowex 50W H + resin column, 20 cm × 1 cm); then the column was rinsed with water (15 mL) and eluted with pyridine 0.1 N (20 mL) to obtain syrupy 1 (0.06 g, 88%), Spectroscopic data was identical to that previously published [bib_ref] 5-Dihydroxypipecolic acids in the seed of Julbernardia isoberlinia and Brachystegia, Shewry [/bib_ref].
## Computational details
Calculations were carried out with the B3LYP hybrid functionals and 6-31+G(d) basis set. Full geometry optimizations and transition structure (TS) searches were carried out with the Gaussian 09 package 18. The possibility of different conformational isomers was taken into account for all structures. Frequency analyses were carried out at the same level used in the geometry optimizations, and the nature of the stationary points was determined in each case according to the appropriate number of negative eigenvalues of the Hessian matrix.
# Conclusions
The amino acid (2S,4R,5R) 4,5 dihydroxy-pipecolic acid (1) was synthesized by two enantiospecific sequences, using D-gluco-heptono-1,4-lactone as a chiral template. The key step in the first approach involved the preparation of compound 8 through nucleophilic attack after catalytic hydrogenation of the azide group from aldehyde 7, obtained by degradative oxidation of diol 6. The second strategy enabled the construction of the piperidine ring by intramolecular substitution of the tosylate at C-6 by the amine group at C-2, produced by hydrogenation of carbamate 12, after removal of the isopropylidene group with F3CCO2H. The latter and more efficient sequence involved 12 steps and gave a 19% overall yield, employing a commercially available and inexpensive sugar lactone.
[fig] Figure 1: Minimum energy conformations, activation energy, and transition state for the intramolecular nucleophilic attack of amine on the aldehyde of intermediate 7a (reactant) to obtain 8 (product). (DFT, Gaussian 9.0) [28]. Coupling constant analysis of the 1 H-NMR data of compound 8 suggested anti dispositions between H-4 and H-5 (J4,5 ≈ 6.0), H-5 and H-6 (J5,6 ≈ 6.0, J5,6' = 2.8 Hz), and H-2 and H-3 (J2,3' = 4.4, J2,3 = 8.8 Hz), suggesting a boat ( 4 BN) conformation, which was confirmed by a NOESY experiment (NOE cross peaks between H-3 and H-4, H-3' and H-2, and H-5 with H-3' and H-6' (Figure 2)). [/fig]
[fig] Figure 2, Figure 3: Key NOE correlations for compound 8. Minimum energy conformations of target compound 8 calculated by DFT (Gaussian 9.0)[28]. [/fig]
[fig] Scheme 4: Synthesis of compound 1 from azide 5. [/fig]
[fig] Figure 4: Minimum energy conformations, activation energy, and transition state for the intramolecular ring closure of intermediate 12a to obtain 1 (DFT, Gaussian 9.0)[28]. [/fig]
[fig] Figure 5: Minimum energy conformation of target compound 1 calculated by DFT (Gaussian 9.0)[28]. [/fig]
[fig] 7: mL) was added p-toluen-sulfonyl chloride (0.519 g, 4.54 mmol). After stirring at rt for 12 h, TLC (1:1, hexane/EtOAc) monitoring showed complete conversion of 3 (Rf = 0.58) into a less polar product (Rf = 0.75). The reaction mixture was concentrated and the residue purified by flash chromatography using as eluent mixtures of hexane/EtOAc (from 20:1 to 5:1), to afford crystalline 4 (1.99 g, 96%), mp 109 °C, [α]D: −42.10 (c 1.0, CH3OH). (+)HRESIMS: [M+H] + = 459.1684, [M+Na] + = 481.1489, (calcd for: C21H31O9S = 459.1683, and C21H3ONaO9S = 481.1502). 1 H-NMR (CDCl3, 600 MHz) δ: 7.85 (d, 2H, J = 8.3 Hz, 2 H-Tosyl), 7.35 (d, 2 H, J = 8.3 Hz, 2 H-Tosyl), 5.10 (dd, 1 H, J2,3 = 10.8 J2,3 = 3.1 Hz, H-2) 4.10 (dd, 1 H, J7,7' = 8.3 J6,7 = 5.9 Hz, H-7), 3.88 (dd, 1 H, J7,7' = 8.3 J6,7' = 5.1 Hz, H-7'), 3.92 (ddd, 1 H, J5,6 = 8.3, J6,7' = 5.1, J 6,7 = 5.9 Hz, H-6), 3.79 (ddd, 1 H, J3,4 = 2.2, J3',4= 10.4, J4,5 = 7.8 Hz, H-4), 3.44 (t, 1 H, J4,5 = 7.8, J5,6 = 8.3 Hz, H-5), 3.74 (s, 3 H, OCH3), 2.45 (s, 3 H, CH3-Tosyl), 2.28 (ddd, 1 H, J2,3 = 10.8, J3,4 = 2.2, J3,3' = 14.2 Hz, H-3), 1.97 (ddd, 1 H, J2,3' = 3.1, J3',4 = 10.4, J3,3' = 14.2 Hz, H-3'), 1.36, 1.33, 1.32, 1.16 ((CH3)2C). 13 C-NMR (150.09 MHz, CDCl3) δ: 169 (C-1), 145.0, 133.3, 129.7, 128.2 (C-aromatic), 109.7, 109.3 (Me2C), 81.1 (C-5), 76.7 (C-6), 75.2 (C-4), 75.2 (C-2), 67.7 (C-7), 52.7 (CH3O), 36.0 (C-3), 26.7, 26.6, 26.6, 25.1, ((CH3)2C), 21.7 (CH3-Tosyl). [/fig]
[table] Table 1: Preparation of compound 8 by catalytic hydrogenation of aldehyde 7. [/table]
|
Interactions between reproductive biology and microbiomes in wild animal species
Many parts of the animal body harbor microbial communities, known as animal-associated microbiomes, that affect the regulation of physiological functions. Studies in human and animal models have demonstrated that the reproductive biology and such microbiomes also interact. However, this concept is poorly studied in wild animal species and little is known about the implications to fertility, parental/offspring health, and survival in natural habitats. The objective of this review is to (1) specify the interactions between animals' reproductive biology, including reproductive signaling, pregnancy, and offspring development, and their microbiomes, with an emphasis on wild species and (2) identify important research gaps as well as areas for further studies. While microbiomes present in the reproductive tract play the most direct role, other bodily microbiomes may also contribute to facilitating reproduction. In fish, amphibians, reptiles, birds, and mammals, endogenous processes related to the host physiology and behavior (visual and olfactory reproductive signals, copulation) can both influence and be influenced by the structure and function of microbial communities. In addition, exposures to maternal microbiomes in mammals (through vagina, skin, and milk) shape the offspring microbiomes, which, in turn, affects health later in life. Importantly, for all wild animal species, host-associated microbiomes are also influenced by environmental variations. There is still limited literature on wild animals compared to the large body of research on model species and humans. However, the few studies in wild species clearly highlight the necessity of increased research in rare and endangered animals to optimize conservation efforts in situ and ex situ. Thus, the link between microbiomes and reproduction is an emerging and critical component in wild animal conservation.
# Introduction
Animal evolution has taken place in environments replete with microorganisms. With a few exceptions (in some insects, for example), every animal species is colonized and lives in close association with microbial partners that include viruses, bacteria and microscopic eukaryotes. Animal biology cannot be understood without reference to the microbial communities that have evolved to live on and within hosts, usually referred to as the organism's microbiomes. This has led researchers to propose that the amalgam of host and the microbial community genomes might be a unit of selection. The term microbiome is generally used to refer to both the composition of a microbial community (also termed the microbiota) and the entire microbial genomic content of the community (also termed the metagenome). The expressed microbial genome (also termed the metatranscriptome) is the fundamental biological unit of importance but is more difficult and expensive to determine compared to identifying the microbiota by 16S rRNA gene sequencing. The present review will be using the term 'microbiome' to refer to the microbial community composition.
The microbiome is a valuable proxy for estimating physiological states, including in reproduction, which is
## Open access
Animal Microbiome essential to the survival of species. Balanced interactions within and between host cells and non-host cells, such as the resident microbial community, are essential to overall fitness, including reproductive health. Reproductive microbiomes are defined as microbial communities inhabiting host reproductive tracts in males and females. However, other microbial communities (e.g., gut, milk, and glandular microbiomes) can further influence reproductive success from mate choice to healthy pregnancy and successfully producing offspring. For example, signal-producing microbiomes (in e.g., scent glands) may play a role in pre-copulatory reproductive communication. Similarly, the milk microbiome influences postnatal infant development. Host-associated microbiomes, including reproductive microbiomes, are also affected by both endogenous processes, such as host physiologyand exogenous factors, such as social interactions and environmental variations. Thus, while microbiomes present in the reproductive tract may play the most direct role, other bodily microbiomes may contribute to facilitating reproduction. illustrates the complex interactions between host microbiomes, life cycles, and reproductive success. It is recommended that research on reproductive microbiomes should also be expanded to include microbial effects before copulation/fertilization and the transfer of appropriate microbiomes to the offspring.
Some of the earliest observations on the linkages between host microbiota and host reproduction occur in freshwater hydra, in which bacteria are essential to the production of offspring through asexual budding. Since then, a growing body of literature on the role of microbiota in host reproduction has been produced, with a large focus on invertebrate systems (particularly Wolbachia symbioses) and vertebrate model organisms. This is in part due to the ability to perform controlled laboratory experiments in these systems that allow for inference regarding host reproductive-microbiota interactions. Not surprisingly, we know little about microbiomes in animal species compared to human microbiomes. Most reports on the reproductive effects of microbiomes of non-human species are on domesticated and laboratory animals, mostly mammals.
Although the number of microbiome studies conducted on wildlife remains lower than in other animal species, it has increased over the last decade. Pronounced animal species declines and loss of natural habitats across the globe have led to the establishment of many captive assurance populations in which reproductive success is critical to the persistence of rare and interactions between host microbiomes, life cycles, and reproductive success in wild animal species endangered vertebrate animals. Captive breeding is known to alter host-associated communities and there is increasing support for an associated effect on host reproductive success. As observed in humans, assisted reproductive technologies (from hormonal stimulation to embryo transfer), contraceptive methods, and treatment of pathologies may have an impact on the microbiomes. Our attempts to breed species in captivity (naturally or through assisted reproductive technologies) may have significant effects on the species microbiomes that will influence success or failure of reproduction.
The objective of this review is to (1) specify the interactions between animals' reproductive biology, including reproductive signaling, pregnancy, and offspring development, and their microbiomes, with an emphasis on wild species and (2) identify important research gaps as well as areas for further studies. We will begin by discussing the links between microbiomes, life stage, and reproductive status in males and females. Then, we will move to detailing the link between microbiomes, reproduction, and environmental changes. We will end by discussing future directions in that area. Throughout the manuscript we also iterate the importance of including appropriate environmental controls and experimental criteria to improve the validity of reproductive microbiome research, since these are typically low biomass.
## Links between microbiomes, life stage, and reproductive status in males and females
Endocrine mediation of reproductive microbiomes across reproductive cycles or breeding seasons The host endocrine system affects the structure of reproductive microbiomes in many animal species, particularly through corticosteroid hormones (glucocorticoids and mineralocorticoids). Glucocorticoids, such as cortisol and corticosterone, are associated with stress and can affect reproductive functions through known links to reproductive behavior and endocrine cycling as seen in primates, rhinos, and wild birds.Evidence for bidirectional signaling between the microbiome and the hypothalamic pituitary axis and greater gut-brain axis also suggests that microbiomes can have significant effects on the host endocrine system. Glucocorticoid-related microbiome effects have largely been studied in the gut microbiome with some investigation of reproductive outcomes in rhinos, fish, birds, and pigs. For example, in Eastern black rhinos (Diceros bicornis michaeli), glucocorticoid and progestagen concentrations, as well as the abundance of certain less common gut microbes, differ between female rhinos who produce calves and females that are considered non-breeders. This suggests a relationship between hormone cycling, reproduction, and gut microbiome; however, the direction of these effects is unknown. Bacteria can metabolize steroid hormones such as glucocorticoids and convert them into androgen hormones in humans, highlighting the challenge of disentangling hormone origin and the cause-and-effect relationships between microbiomes and reproduction.
Studies in humans also indicate that increased cortisol associated with prolonged stress can influence the vaginal mucosa, including the deposition of glycogen, which mediates the abundance of Lactobacillus spp. and underpins aspects of human vaginal tract health and copulatory success. More broadly, increased cortisol resulting from prolonged stress is correlated with changes in gut microbiome structure and inflammatory responses, which, in turn, are known to affect pregnancy outcomes in women. Although more research is needed in wild animal species, the relationships between glucocorticoid hormones and microbiomes provide plausible mechanisms by which psycho-social and physiological stress can influence health and reproductive outcomes.
Likewise, sex hormones play an integral part in structuring reproductive microbiomes. Previous research has largely been focused on female mammalian microbiomes and their relationships with ovarian hormones (estradiol and progesterone), reproductive cyclicity, or general variation in vaginal microbiomes, such as in non-human primatesand giant pandas (Ailuropoda melanoleuca). In primates, the increased circulating estradiol associated with sexual maturity in female mammals is linked to changes in vaginal epithelium function (e.g., increased glycogen production) and microbial community structure, further highlighting the complex relationships between endocrine function, host physiology, and microbial communities. In mice models, wild-type and estrogen receptor knock-out mice even indicate an effect of estrogen binding on microbiome structure.
In males, testosterone concentrations are linked to seminal microbiome structure, sperm characteristics (see below), and infertility. However, those studies still need to be confirmed with appropriate controls to discount the influence of DNA contamination on the results as seen in research on other microbiomes. In male birds, cloacal microbial diversity is positively correlated with testosterone levels, which in turn may affect rates of reproductive behaviors-such as extra-pair copulationsthat increase opportunities for microbial transmission. In free-living rufous-collared sparrows (Zonotrichia capensis), cloacal bacterial communities differ between the sexes when they are in breeding condition. In males, but not in females, the bacterial community becomes more diverse with the onset of reproduction, and then decreases in diversity as males transition to non-breeding condition.
Interactions between sex hormones and microbes can have significant metabolic effects on the host. In female Phayre's leaf monkeys (Trachypithecus phayrei) reproductive hormones, specifically progestagens, contribute to the shifts in the gut microbiome during pregnancy and lactation. There also is a shift in the maternal gut microbiome of Tibetan antelope (Pantholops hodgsonii) during the perinatal period, though the effects on reproduction are unknown. Future research over the next decade will likely elucidate how host endocrine-microbiome interactions affect host physiology and reproduction. At present, evidence for sex hormone-microbiome relationships in both sexes largely stem from studies of humans, whose reproductive microbiomes differ drastically even from other closely related primates (e.g., the unique Lactobacillus dominance in many, but not all, human vaginal microbiomes;. The extent to which these patterns of endocrine-microbiome relationships apply to endangered species warrants further investigation, especially given the common use of husbandry practices that alter endocrine function (e.g., hormonal contraceptives; see below).
## Microbiomes and reproductive signaling (visual and olfactory)
Bodily microbiomes could potentially contribute to reproduction by mediating pre-mating reproductive signaling. Bodily microbiomes could potentially contribute to reproduction by mediating pre-mating reproductive signaling (e.g., plumage or bill color in birds. The strongest examples for reproductive signaling mediated by the microbiome stems from studies on olfaction. Olfactory signaling is a mode of reproductive communication that has long been thought to rely on a combination of volatile organic compounds produced by the host and by their commensal microbes (i.e., the fermentation hypothesis;. Olfactory chemical cues can facilitate mate choice and reproductive success through 'honest' signaling of animal health, kinship and genetic compatibility, social status, and reproductive cycle and fertility. Depending on the social organization of the animal species, these cues are relevant to and can be produced by one or both sexes. Although much of the previous research has been focused on male odor production to attract female mates, females of many species (including primates that are female-dominant) also rely on scent signaling to convey socially relevant information. These cues can emanate from multiple body sites (e.g., scent gland and skin surfaces) and waste products (e.g., urine and feces). Scent glands, in particular, provide protected, ideal habitats for microbial growth and specialized glandular tissue can occur at various body sites, including homologous sites in males and females (e.g., labial glands in females and scrotal glands in males).
In multiple mammals, scent gland signals covary with glandular bacterial communities. These olfactory signals convey social and reproductive information that may influence social status and mate choice (for example in giant pandasor European badgers. In male diademed sifakas (Propithecus diadema) sternal gland microbiomes and olfactory chemical signals vary according to the male's breeding vs. non-breeding status. Similarly, in Chinese musk deer (Moschus berezovskii), musk gland bacteria and musk chemical composition reflected breeding status as mated vs. unmated, an important signal for mate choice. Thus, scent gland microbes are associated with the host's ability to advertise reproductive information, which may affect their reproductive success. Outside of mammalian species, bacteria from the uropygial glands of multiple bird species produce volatile compounds that are strongly linked to reproductive successand appear to correlate with major histocompatibility complex diversity and reproductive compatibility. Additionally, in a common South American tree frog (Dendrobates sp.), sex-specific scents can be traced to a frog-specific skin bacterium that produces a class of compounds with strong odor properties. However, investigations of microbial contributions to scent signaling are scarce, particularly in wild animals. If, as mounting evidence suggests, host-associated microbes underpin signals of reproductive status and mate choice, investigation of this microbe-olfactionreproduction axis may be especially relevant to holistic perspectives of reproductive success.
## Influence of mating behavior and copulation on microbiomes
In female mammals, there is a complex vaginal microbiome, which is vital to many aspects of reproduction, including preparing the host for reproductive activity. The disruption of the vaginal microbial community can have severe consequences for health and reproductive success. The vaginal microbiome has been studied in a number domesticated mammal species, nonhuman primates and the giant panda (reviewed in. Vaginal microbiomes vary widely among host taxa, and closely related species will not necessarily have similar vaginal microbiomes. A Lactobacillus dominated vaginal microbiome appears unique to humansand even within humans there are stable, healthy vaginal microbiomes with few Lactobacillus.
In male mammals, semen is a complex biological fluid that contains nutrients for the spermatozoa (fructose, glucose, amino acids, minerals) factors that affect sperm motility and potency, and has a characteristic pH. Many of the characteristics that make semen an adaptive medium for sperm cells also are conducive to microbial organisms. Certainly, pathogens responsible for many sexually transmitted infections (STIs) live in seminal fluid. The existence of microbes in seminal fluid implies that a microbiome exists in the seminal vesicles and/or other parts of the male urogenital tract. Most of the research on seminal fluid microbiomes has been done in humans and rodent biomedical models, and showed microbial-associated sperm dysfunction. Evidence from mice suggests that the seminal fluid microbiome may also have beneficial effects on reproduction. Studies in men have found associations between microbial taxa and sperm concentration, motility, and quality. Some microbes are associated with a reduction in sperm quality, whereas others are associated with higher quality sperm, suggesting that certain microbes may be indicators of sperm quality. A recent study conducted in peccaries (Pecari tajacu) also highlighted the relation between preputial microbiome and semen quality. As mentioned earlier, more studies are warranted to discount the influence of DNA contamination on the results (as seen in research on other microbiomes). Furthermore, the seminal microbiome may also have long term effects on fetal development that may influence later health and reproductive success of offspring. Male mice fed a high-fat diet produced male offspring with poorer sperm quality, suggesting that host metabolism has effects on semen and its microbiome. It is unknown whether these two effects are causally related, but it raises the possibility that the seminal microbiome may have next generational effects on reproductive success.
Copulation is a necessary component of vertebrate reproduction and also presents an opportunity for the transmission and interactions of microbes between host males and females in both directions. The transfer of microbes by copulation-and the associated risk of perturbing the established microbiome or STIs-likely exerts selective pressures on male and female reproductive microbiomes. In non-human primates, peak sexual receptivity and copulation behavior coincide with increased vaginal microbial diversity, which may defend against the increased risk of microbial transmission via copulation, including STIs. In owl monkeys (Aotus sp.), unmated males and females differ from mated animals in preputial and vaginal microbiomes, respectively, which was a result of the transfer of microbes via copulation. Whether the microbiome within semen can benefit the female partner is unknown. Although sexual microbial transfer is most widely studied from the perspective of STIs, copulation may also be an important vector for the transfer of benign or even beneficial microbes. Certainly, the microbial community within seminal fluid will at least briefly inhabit the female reproductive tract after copulation and may potentially influence fertilization and perhaps even implantation. Evidence from human studies suggests that the seminal and vaginal microbiomes of consistent sexual partners exhibit similarities and that the vaginal microbiome changes slightly after intercourse. More numerous or frequent copulations (as seen in promiscuous species) have been linked to greater vaginal or cloacal microbial diversity (in mice:; in primates; in lizards. Greater vaginal microbiome diversity in promiscuous species can aid in creating a community that is more resilient to perturbation or infection and, ultimately, may improve reproductive success. However, in humans, promiscuity has been correlated with dysbiosis of vaginal microbiomes, suggesting that frequent copulations with different partners could also disrupt or destabilize the vaginal microbial community.
## Pregnancy and birth are major colonizing events of microbiomes
The lower mammalian female reproductive tract (vulva, vagina and cervix) harbors complex, dynamic microbiomes which vary by region, hormonal state, and reproductive condition. A lack of stability in the vaginal microbiome in early pregnancy, especially with any significant presence of pathogenic strains, is associated with pregnancy complications including preterm birth and spontaneous abortion. In contrast to the lower tract, the upper mammalian female reproductive tract (uterus, fallopian tubes) is considered a protected body region, with the expectation that it will be difficult for microbes to travel from the lower into the upper reproductive tract and become established. This does not mean that microbes cannot migrate along the tract or colonize these regions, but the diversity and density of the microbes in the upper reproductive tract is generally sparse compared to other body regions in women. The developing fetus, while not in a sterile environment, is in a protected environment, and thus exposed to a lower abundance and diversity of microbes.
There is consistent evidence for a uterine wall microbiome in several species, including humans, horses, cows and rhesus macaques (Macaca mulatta). The uterine microbiome is distinct from and is less densely populated than the vaginal microbiome. There is evidence in both humans and horses for an association between uterine microbiome structure and successful pregnancy outcome. However, more research is required to investigate the robustness of and mechanisms underlying this association.
There is controversy regarding the extent and stability of the microbial communities found in the mammalian placenta. Note that most of the evidence for or against a placental/fetal microbiome again comes from human data. A placental microbial community has consistently been found in human pregnancies that are complicated by pre-eclampsia and preterm birth (reviewed inand microbial invasion of the placenta is considered a causal aspect for these pathologies; however, several studies also reported the existence of a placental microbiome during uncomplicated pregnancy. Studies also found evidence of microbial communities in amniotic fluid. However, subsequent studies have challenged these results, finding no detectable microbiota in amniotic fluid from uncomplicated pregnancies with intact membranes. A recent systematic review concluded there was no evidence for a placental microbiome in uncomplicated human pregnancy. While microbial communities may exist in the uterus, the placenta, and the amniotic sac, these sites are largely sterile in humans. A study of uterine, placental and fetal tissue from four rhesus macaque pregnancies delivered by Cesarean section at approximately 130 days of the normal 166-day gestation found a uterine wall microbiome, but no evidence of placental or fetal microbial communities. Recent studies in horses found evidence for microbiomes in placenta, amniotic fluid and meconium, indicating early establishment of a microbiome within the fetal gut by microbes of placental origin (reviewed in. However, these studies were done on material collected after parturition, the abundance of microbes was low and thus the possibility of contamination cannot be discounted. There is no evidence for any placental microbiome in cattleand bacterial loads of amniotic fluid were no different from negative controls in calves delivered by Cesarean section. However, there are detectable, albeit low abundance, microbes in calf meconium. Thus, colonization of the placenta, amniotic sac, and fetus by maternal microbes may reflect pathology rather than adaptive function. An ascending intrauterine infection is a risk factor for preterm birth and is associated with both maternal and fetal inflammatory response. Growing evidence suggests that the interactions between maternal reproductive tract microbiomes and maternal and fetal immune response are important factors in birth outcome. Prenatal exposure to microbes happens in pregnancies complicated with certain pathologies, but whether microbial communities in placental tissue can inoculate the mammalian fetus before birth remains uncertain.
Regardless of the mode of transmission, early life microbiome may impact later reproduction, as shown in the mouse model. The microbes acquired at birth can affect offspring health later in life (as reviewed by, and health status can have subsequent effects on reproductive success, such as obesity. When the fetus travels through the reproductive tract it will become inoculated with the maternal microbiome. In humans, infants born vaginally vs. by Cesarean section will have different skin and gut microbiomes. For egg laying vertebrates there is less evidence, but both lizard and bird eggs appear to harbor bacterial communities inside the shell. Presumably this inoculation occurs in the upper regions of the oviduct before the shell has formed. The extent to which any maternal microbes transferred after shell formation can persist in the nest and thus be available to inoculate hatchlings is unknown. Similarly, Bornean foam-nest breeding frogs deposit eggs within foam nests with the internal nest colonized by bacteria (McGrath et al. in review for this special issue). The vertical transmission of the maternal microbiome to offspring is a critical aspect of successful reproduction, and the microbiome of the female reproductive tract is the first exposure for eggs and fetuses.
## Post-partum microbiomes and infant development/ reproductive success, from mother to child
In contrast with the often brief exposure to maternal vaginal microbiomes, infant exposure to the maternal skin and oral microbiomes will be chronic, occurring over the much longer time period of neonatal dependence, but also within the context of well-developed (though still developing) neonatal microbiomes that may be capable of resisting colonization by novel microbes. Maternal skin contact may be more important for certain taxa. For example, in southern hairy-nosed wombats (Lasiorhinus latifrons), pouches contain communities of microorganisms that are substantially altered by the host reproductive cycle. More investigations into the roles that pouch microorganisms may play in marsupial reproductive health and joey survival are warranted.
The neonate will also be exposed to a microbiome unique to mammals, the milk microbiome. Milk contains a microbial community that can vary among mammalian species. The origin of the milk microbiome is not well established, but evidence from rodent models suggests that bacteria may be transported from the maternal gut to the mammary gland. Compared to the vaginal microbiome the milk microbiome is depauperate, both in terms of species numbers and total amount of microbial cells. However, a breastfed baby is estimated to consume 10 5 to 10 7 bacteria each day. Suggested biological functions for the milk microbiome include inoculation of the neonatal gut, activation of mammary immune systems and cells, and stimulation of the developing neonatal immune system. Another potential function is to develop immune tolerance in the neonate to commensal and symbiotic maternal microbes. Note that the milk microbiome is not the only aspect of milk that influences the neonatal gut microbiome. Milk contains live maternal immune cells, maternal immune function molecules, antibiotic molecule such as lysozyme, and prebiotic oligosaccharides that are not metabolized by the neonate but rather by the neonate's gut microbes. The relative importance of the milk microbiome compared to these other aspects of milk are not known. The transfer of a gut microbiome from parent to offspring through feeding is not just a mammalian trait. Several avian taxa regurgitate food into their chicks and produce crop secretions, termed 'crop milk' that they feed to chicks. These secretions contain similar bioactive constituents to mammal milk and contain microbes. All these exposures to maternal microbiomes shape the offspring microbiomes, which, in turn, affects later health.
There also are extensive interactions between microbes, their products and the host immune system that are often symbiotic, not antagonistic. The lack of a microbiome can disrupt neonatal development and gnotobiotic rodent models display significant deficiencies in anatomy, physiology, metabolism and immune function. The development of the neonatal immune system appears to be highly influenced by microbiome effects. Captive breeding programs for endangered species need to consider how management procedures may affect early life microbiomes.
## Link between microbiomes, reproduction, and environmental changes
## Captivity, microbiomes, and fertility control
Often attributed to the loss of native or beneficial microbes, the effects of captivity vary across host species but are undeniably a vital aspect of conservation breeding. Moreover, efforts to manage or improve reproduction in captive populations, via contraceptives or assisted reproduction, can shape reproductive microbiomes with largely unknown consequences. Beyond the scope of captivity, human activity, including the alteration and contamination of natural environments, can influence animal microbiomes and affect reproductive success; however, empirical links between these anthropogenic influences and reproductive microbiomes are currently sparse. In the face of extensive and continuing human influence on natural ecosystems and their inhabitants, there has been a recent and urgent call for greater 'microbiome conservation' in order to characterize and preserve global microbial diversity. Here, we reinforce this call and suggest that the study and conservation of reproductive microbiomes are particularly relevant to endangered or rare species in which managed breeding programs are vital resources for species survival.
Understanding the effects of captivity on hostassociated microbiomes is crucial to maintaining and improving the care and management of endangered species. The majority of research to date has focused on gut microbial communitieswith few studies examining captivity-induced changes in reproductive microbiomes. Nevertheless, burgeoning evidence of 'cross-talk' between gut and reproductive microbiomes suggests that many gut-centered outcomes of captivity can have downstream effects on the suite of reproductive microbiomes mentioned above. The factors driving gut microbiome variation between captive and wild animals (e.g. dietary mismatches;may similarly influence reproductive microbiomes by altering host metabolism and dictating available resources. For instance, Williams et al.found that in southern white rhinoceros (Ceratotherium simum simum) fertility varied significantly with phytoestrogen profiles, the abundance of several gut bacterial taxa, and microbially-derived phytoestrogen metabolites. This suggests that diet may drive transformation of dietary phytoestrogens by the gut microbiome which then impacts reproductive outcomes. Even pre-copulation, if the effects of captivity alter scent gland microbial communities similarly to the widely reported effects seen in the gut microbiomes, associated olfactory signals could be altered, resulting in discordant signals and a disruption of reproductive communication.
Medical treatments received by animals in captivity also have the potential to shape the structure and function of reproductive microbiomes. Antibiotic treatment has been strongly linked to dysbiotic microbial communities, and thus antibiotic use in captive endangered animals may structure reproductive microbiomes and influence associated outcomes. For example, antibiotic treatment in captive primates result in alterations of gut microbiomes, with potential long-term consequences for gut health. There also is evidence in lactating women that prenatal antibiotic use changes the milk microbiome, by decreasing the prevalence of beneficial microbes and increasing the proportion of potential pathogens. Although very little is known about the prevalence and impact of antibiotic resistance genes in reproductive microbiomes, they may pose a particular risk to endangered animals in captivity, where the efficacy of antibiotic treatment is paramount to successful care and husbandry.
Additionally, artificial fertility control, via e.g., contraceptives, may further influence the reproductive microbiomes of captive animals. Hormonal contraceptives supply artificial reproductive hormones (e.g., depot medroxyprogesterone acetate) that disrupt natural endocrine processes to minimize fertility and prevent pregnancy. In hormonally contracepted humans, endocrine mediation of microbial communities is altered resulting in changes in the composition of reproductive microbiomes; however, evidence for the effects of contraceptives on the microbiomes of non-human animals is scarce. In humans and non-human primates, hormonal contraception has been linked to the dampening or disruption of olfactory cues used for reproductive communicationpotentially reflecting changes in odor-producing bacteria. Given the propensity for reproductive dysfunction in many captive, endangered species, greater study of reproductive microbiomes in these settings can shed light on potential mechanisms of dysbiosis and, in turn, provide opportunities to integrate microbial ecology into successful husbandry practices.
## Environmental contaminants, microbiomes, and reproductive health
Environmental contaminants, including herbicides, pesticides, and heavy metal pollutants, can cause adverse effects on reproduction in wildlifeand change host microbiomes. The adverse effects on reproduction may be mediated or modulated by changes in the host microbiome. For instance, gut microbes can metabolize environmental chemicals and thus modulate toxicity for the host animal. Likewise, one of the most commonly used herbicides, glyphosate, inhibits the biochemical pathway for aromatic acid synthesis which is found in numerous bacterial species and can subsequently alter host-associated microbiomes. Overall, the interactions among reproduction, contaminants and microbiomes have rarely been examined. One study found that glyphosate-based herbicides altered gut microbiomes and reproductive hormone levels in Japanese quails (Coturnix japonica) in a long-term study, but did not directly affect reproduction in either sex (maturation, testis size or egg production). Future studies on the effects of environmental contaminants on reproductive microbiomes should consider that these interactions may be subtle or potentially delayed in order to fully understand the risks related to contaminant exposure.
In addition, man-made antibiotics that leach into natural ecosystems can alter both environmental and host-associated microbial communities and facilitate the spread of antibiotic resistance genes. In contrast to the effects of antibiotic treatment done under veterinary care, antibiotic contamination of the environment can exert selective, but often non-lethal pressure on microbes, allowing them to evolve or spread resistance genes and expand the 'resistomes' in natural environments. It is now well-known that antibiotic-resistant bacteria in environmental settings (e.g., water, soil, food items) can be picked up by animals and humans, where they can be incorporated into host-associated microbiomes and potentially pose significant health risks. Indeed, in wild and captive populations of endangered ring-tailed lemurs (Lemur catta), antibiotic resistance genes covaried between host gut microbiomes and environmental soil microbiomes, a pattern that also reflected anthropogenic use of antibiotics and environmental contamination.
## Pathogens, microbiomes, and reproductive health
There are well known proximate mechanisms by which pathogens can influence reproduction, usually negatively. When animals mount an immune response due to pathogenic infection, animals can increase, and maintain or decrease their reproductive investment. Such life-history tradeoffs in energy allocation between reproduction and fighting infection are fundamental to fitness. We can broadly consider pathogens as part of the host microbiome as they often colonize and live as part of the microbial community during infection. Most studies to date on reproduction-infection tradeoffs have examined this in the context of single pathogen infections or perceived infection (e.g., vaccination). Evidence shows that tradeoffs between reproduction and fighting infection go in all directions, and likely reflect unique pressures, such as age of individuals in male crickets.
A commonly tested hypothesis is that an individual will increase reproductive investment when immunocompromised. The terminal investment hypothesis predicts that individuals should invest more in their present reproduction if they are less likely to survive, with infection and disease being a potential cue of diminishing reproductive value. A variety of studies provide support for the terminal investment hypothesis in reproductive efforts, including in frog species infected with a fungal pathogen that up-regulate gametogenesis, increase male calling behavior as well as mating success. Likewise, similar support exists in bird species. For example, great tits (Parus major) infected with haemosporidian parasites have increased reproductive success, blue footed boobies (Sula nebouxii) with elicited immune response increase parental care, and vaccinated house sparrows (Passer domesticus) increased egg laying efforts. On the other hand, mounting an immune response can lead to reduction in reproductive investment. Another intriguing hypothesis is the immuno-competence handicap hypothesis which predicts that only high-quality males can maintain high levels of testosterone and afford the physiological cost of hormone-derived immunosuppression, but similar to the terminal investment hypothesis data remains ambiguous to support immunocompetence handicap hypothesis. Future work examining how the larger collective microbial community is involved in these tradeoffs would be useful.
# Conclusions-future directions
While microbiomes present in the reproductive tract play the most direct role, other bodily microbiomes may contribute to facilitating reproduction in wild animal species. Endogenous processes related to host physiology can both influence and be influenced by the structure and function of microbial communities. Importantly, exposures to maternal microbiomes (through the vagina, skin, and milk) shape the offspring microbiomes later in life . In many cases, we do not know if shifts in the microbiome are the cause of changes in the host reproductive status or a by-product of physiological changes.
In addition, host-associated microbiomes are also influenced in wildlife by exogenous factors, such as social interactions, environmental variations, or ex situ breeding programs. However, much more research is warranted in that area in order to optimize conservation efforts. Although our review is biased towards mammals, as there currently is more information regarding reproductive microbiomes, many of the concepts discussed in this review (as well as the next steps proposed below) could be widely applied to other vertebrates.
It is important to remember that there is still a need for more comprehensive microbe taxonomy (for better identification) and more studies about microbial functions. It is critical to culture as many of these 'reproductively relevant' microbes as possible and use standardized 16S protocols so we can compare across studies and find shared microbes. As mentioned earlier, it also is essential to reduce contamination bias, including appropriate environmental controls and experimental criteria to improve the validity of low microbial biomass research. We also want to further highlight the importance of using adequate controls in sampling and sequencing reproductive microbiomes. To improve 'credibility' of reproductive microbiomes -(1) numerous negative control samples should be taken throughout sampling and throughout procedures in the laboratory to be used to identify potential contaminants post sequencing, such as by using the package 'decontam' available in Rand (2) use of positive mock bacterial community controls to verify distribution of communities are not overly skewed and to verify spurious sequences are not being generated during molecular procedures in preparing libraries for sequencing.
To fill the gaps of knowledge in wildlife reproduction and microbiomes, we propose a list of topics that should be explored to better understand and possibly manipulate microbiomes in wild animal species:
- Relationship between hormones, reproduction, and gut microbiome in males and females (including the impact of hormonal contraception or stimulations). - Effect of puberty and aging on microbiomes. - Microbial contributions to scent signaling and reproduction. - Role of male and female microbiomes on sperm competition during and after mating. - Role of seminal microbiome on the reproductive success of offspring (transgenerational effect). - Role of microbiome during conception, early embryo development in the oviduct, implantation, delayed implantation. - Association between uterine microbiome structure and successful pregnancy outcomes (leading to the development of non-invasive microbial markers of early pregnancy and fetal health, prediction of infant health). - Influence of maternal diet on maternal and fetal microbiomes during pregnancy. - Extent and stability of the microbial communities found in the mammalian placenta. - Investigations into the function of the pouch microbiomes in marsupials. - Exploration of the amphibian cloacal and spermic urine microbiome. - Microbiomes and reproduction in invertebrates (e.g., oysters, corals). - Consequences of captive breeding management on early life microbiomes and future reproductive fitness. - Prevalence and impact of antibiotic resistance genes in reproductive microbiomes and their potential risk to endangered animals in captivity). - Links between infertility or reproductive issues and microbiomes. - Impact of artificial insemination or embryo transfer on microbiomes during gestation (does artificial insemination provide the same microbial signals as copulation?). Effects of semen microbiomes on in vitro fertilization.
- Manipulating or bioengineering the microbiomes to enhance or suppress fertility.
Lastly, in situ conservation efforts could benefit from more studies on interactions between reproduction, contaminants, and microbiomes. This could be conducted on sentinel species that could reflect the status of reproductive health in wild species sharing the same habitat. Application of ecological theory and advancement of analytical tools would help to understand what is structuring microbiomes. Collective findings will inform strategies to improve reproductive health in conservation breeding or in the wild. It will also help to develop conservation policy or legislation to include microbiome assessments for translocation or reintroduction of wild animal species in their natural habitats. |
Smoking status and second-hand smoke biomarkers in COPD, asthma and healthy controls
Introduction: Tobacco smoke worsens COPD and asthma. For healthy individuals, quantifying active and second-hand smoke (SHS) exposure clarifies the epidemiology of tobacco consumption and the efficacy of nonsmoking measures. Identifying tobacco exposure biomarkers and cut-offs might allow the creation of sensitive and specific tests. Aim: We describe the state-of-the-art serum, urinary cotinine and exhaled carbon monoxide (CO) cut-offs for assessing smoking status and SHS exposure in adult patients with COPD or asthma, and healthy controls. Methodology: After a keyword research in the PubMed database, we included papers reporting on the cut-offs of the investigated biomarkers in one of the populations of interest. Papers published before 2000, not in English, or reporting only data on nonadult subjects or on pregnant women were excluded from the analysis. 14 papers were included in the final analysis. We summarised diagnostic cut-offs for smoking status or SHS exposure in COPD, asthmatic and healthy control cohorts, reporting sensitivity and specificity when available. Conclusion: Serum and urinary cotinine and exhaled CO are easy-to-standardise, affordable and objective tests for assessing smoking status and SHS exposure. Evidence on cut-offs with good sensitivity and specificity values is available mainly for healthy controls. For COPD and asthmatic patients, most of the currently available evidence focuses on exhaled CO, while studies on the use of cotinine with definite sensitivity and specificity values are still missing. Solid evidence on SHS exposure is available only for healthy controls. An integrated approach with a combination of these markers still needs evaluation.@ERSpublications Reliable cut-off values for smoking status in COPD and asthmatic adults are only available for exhaled CO https://bit.ly/34lsHhDCite this article as: Bradicich M, Schuurmans MM. Smoking status and second-hand smoke biomarkers in COPD, asthma and healthy controls. ERJ Open Res 2020; 6: 00192-2019 [https://
# Introduction
Chronic obstructive pulmonary disease (COPD) and bronchial asthma are obstructive lung diseases that affect together ∼900 million people worldwide [bib_ref] Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and..., Vos [/bib_ref]. Tobacco smoke plays a predominant role in worsening these conditions. Among the other consequences of primary and secondary tobacco exposure, lung function deterioration and exacerbation rate increase represent the two most epidemiologically and socioeconomically relevant effects in these patients. For healthy individuals and patients, the quantification of tobacco exposure represents an important measure for estimating the epidemiological relevance of tobacco consumption and second-hand smoke (SHS) exposure, and the efficacy of nonsmoking measures and campaigns.
A standardised questionnaire is one of the most popular and widely used tools to assess smoking status. Nevertheless, even though a questionnaire is an easy-to-use, affordable and time-efficient tool, the results might be affected by the recall capacity of the patients. In addition to that, patients might be reluctant to give completely correct answers due to the fear of social judgement associated with smoking. For these reasons, the assessment of smoking status using a questionnaire might be easily affected by subjective judgement and therefore not sufficiently reliable. Furthermore, such assessments might not be completely accurate to determine SHS exposure [bib_ref] Overview of cotinine cutoff values for smoking status classification, Kim [/bib_ref].
Thus, the identification of tobacco exposure biomarkers and the consequent definition of cut-offs represent goals of utmost relevance for establishing adequately sensitive and specific tests.
In this article the state of the art of serum cotinine, urinary cotinine and exhaled carbon monoxide (CO) as biomarkers of smoking status and smoking exposure in adult patients with COPD, asthma and healthy controls will be described.
# Search methods
We searched articles published in PubMed (https://www.ncbi.nlm.nih.gov/pubmed) until March 2019 using the keywords "cotinine", "serum cotinine", "urinary cotinine", "exhaled carbon monoxide", "eCO", "carbon monoxide", "smoking", "smoking status", "smoking cut-off", "smoking exposure", "smoke exposure", "asthma", "chronic obstructive pulmonary disease", "COPD" and "healthy". Papers published before 2000, not in English, or reporting only data on nonadult subjects or only data on pregnant women were excluded from the analysis. 14 papers addressing the topic accordingly to the inclusion criteria were included in our final analysis [bib_ref] Overview of cotinine cutoff values for smoking status classification, Kim [/bib_ref] [bib_ref] Assessing secondhand smoke using biological markers, Avila-Tang [/bib_ref] [bib_ref] Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different..., Benowitz [/bib_ref] [bib_ref] Exhaled carbon monoxide levels among tobacco smokers by age, Chatrchaiwiwatana [/bib_ref] [bib_ref] Serum metabolite biomarkers discriminate healthy smokers from COPD Smokers, Chen [/bib_ref] [bib_ref] Environmental tobacco smoke exposure and pulmonary function among adults in NHANES III:..., Eisner [/bib_ref] [bib_ref] Correlation of exhaled carbon monoxide level with disease severity in chronic obstruction..., Ejazi [/bib_ref] [bib_ref] Dual exposure to smoking and household air pollution is associated with an..., Fernandes [/bib_ref] [bib_ref] Comparison of urine cotinine and the tobacco-specific nitrosamine metabolite 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and..., Goniewicz [/bib_ref] [bib_ref] Optimum cutoff value of urinary cotinine distinguishing South Korean adult smokers from..., Kim [/bib_ref] [bib_ref] Breath carbon monoxide as an indication of smoking habit, Middleton [/bib_ref] [bib_ref] Self-reported smoking status and urinary cotinine levels in patients with asthma, Pinheiro [/bib_ref] [bib_ref] Optimal cutoff level of breath carbon monoxide for assessing smoking status in..., Sato [/bib_ref] [bib_ref] Estimation of urinary cotinine cut-off points distinguishing non-smokers, passive and active smokers, Zielinska-Danch [/bib_ref]. For each article, we extracted information from the main texts of those reporting data in the abstracts on the cut-offs of at least one of the investigated biomarkers in at least one of the populations of interest.
# Results
## Cotinine
Nicotine and its metabolites are tobacco-specific biomarkers. Nicotine is essentially metabolised in the liver and its half-life is 2-3 h. The half-life of cotinine, one of the main metabolites of nicotine, is 12-20 h. 10-15% of cotinine is excreted in urine; the rest is metabolised to trans-3′-hydroxycotinine and other byproducts. Because of its pharmacokinetics, cotinine quantification in serum, urine and saliva is commonly used for exposure quantification purposes [bib_ref] Overview of cotinine cutoff values for smoking status classification, Kim [/bib_ref].
## Serum cotinine
Healthy controls A review by KIM [bib_ref] Overview of cotinine cutoff values for smoking status classification, Kim [/bib_ref] shed light on serum cotinine cut-offs, and showed the best specificity and sensitivity among healthy controls. Considering the shift to the low side in serum cotinine levels among nonsmokers after the establishment of effective tobacco control measures, it is crucial to take into consideration the latest studies about this topic. BENOWITZ et al. [bib_ref] Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different..., Benowitz [/bib_ref] pointed out how finding the optimal cut-off differentiating smokers from nonsmokers crucially depends on the distribution of cotinine levels between these two populations. Higher cut-offs should therefore be expected in cohorts being exposed to higher levels of SHS, as it was when indoor smoking limitations had not been developed yet.
Among the studies cited in the review by KIM [bib_ref] Overview of cotinine cutoff values for smoking status classification, Kim [/bib_ref] , a cut-off of 15 ng·mL −1 presented a sensitivity and a specificity >95% in a large population of adult women and men. Nevertheless, the study by BENOWITZ et al. [bib_ref] Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different..., Benowitz [/bib_ref] , conducted on 9901 female and male subjects, showed that a cut-off of 3.08 ng·mL −1 had a sensitivity and a specificity >95%. As for SHS exposure, BENOWITZ et al. [bib_ref] Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different..., Benowitz [/bib_ref] suggested 3 ng·mL −1 as serum cotinine cut-off. Nevertheless, AVILA-TANG et al. [bib_ref] Assessing secondhand smoke using biological markers, Avila-Tang [/bib_ref] observed that heavily exposed individuals presented serum cotinine levels up to 25 ng·mL −1 ; therefore, an overlap might occur.
## Copd patients
SATO et al. [bib_ref] Optimal cutoff level of breath carbon monoxide for assessing smoking status in..., Sato [/bib_ref] reported that serum cotinine concentration correlated positively with the smoking status of COPD patients. Mean±SD serum cotinine was 23.2±69.2 and 191.1±109.8 ng·mL −1 in former and current smokers respectively. Moreover, in the COPD population of current smokers, all patients presented serum cotinine concentrations >50 ng·mL −1 . Nevertheless, a statistically significant cut-off value that may distinguish between COPD smokers and nonsmokers has still to be defined.
## Asthma patients
A study by EISNER [bib_ref] Environmental tobacco smoke exposure and pulmonary function among adults in NHANES III:..., Eisner [/bib_ref] conducted among 440 nonsmoking adult asthmatic patients and 10 581 nonsmoking adult healthy controls showed that there were no statistically significant differences between the median serum cotinine levels of these two groups when exposed to statistically comparable environmental tobacco smoke levels. Both asthmatic and nonasthmatic women in the highest cotinine tertile of the study presented worsened lung function. Even if there were no data about a cut-off that might differentiate between asthmatic smokers and asthmatic nonsmokers, it is plausible that serum cotinine levels were higher in the first group. In the asthmatic population studied by SATO et al. [bib_ref] Optimal cutoff level of breath carbon monoxide for assessing smoking status in..., Sato [/bib_ref] , serum cotinine correlated positively with smoking status, although without statistically significant differences. Mean serum cotinine in nonsmokers, former smokers and current smokers was 6.0±5.2, 12.1±25.0 and 198.3±181.7 ng·mL −1 , respectively. Data concerning serum cotinine are listed in table 1.
## Urinary cotinine healthy controls
Among the studies cited in the review by KIM [bib_ref] Overview of cotinine cutoff values for smoking status classification, Kim [/bib_ref] , two studies by GONIEWICZ et al. [bib_ref] Comparison of urine cotinine and the tobacco-specific nitrosamine metabolite 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and..., Goniewicz [/bib_ref] and KIM and JUNG [bib_ref] Optimum cutoff value of urinary cotinine distinguishing South Korean adult smokers from..., Kim [/bib_ref] showed that a urinary cotinine level between 31.5 and 164 ng·mL −1 correlated with adult healthy controls with a sensitivity >93% and a specificity >95%. The difference in the values might be dependent on the involved population, since the study by KIM and JUNG [bib_ref] Optimum cutoff value of urinary cotinine distinguishing South Korean adult smokers from..., Kim [/bib_ref] involved South Korean patients, and that by GONIEWICZ et al. [bib_ref] Comparison of urine cotinine and the tobacco-specific nitrosamine metabolite 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and..., Goniewicz [/bib_ref] analysed data from American, Polish and Mexican patients, suggesting that some ethnic and metabolic confounding factors might have played a role [bib_ref] Overview of cotinine cutoff values for smoking status classification, Kim [/bib_ref].
BENOWITZ et al. [bib_ref] Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different..., Benowitz [/bib_ref] suggested a urinary cotinine cut-off of 10 ng·mL −1 for SHS exposure among healthy individuals. Furthermore, a study by ZIELINSKA-DANCH et al. [bib_ref] Estimation of urinary cotinine cut-off points distinguishing non-smokers, passive and active smokers, Zielinska-Danch [/bib_ref] estimated more specific urinary cotinine cut-offs for different grades of exposure among smokers and nonsmoker healthy controls. Among smokers, a urinary cotinine level of 2100 ng·mL −1 predicted an elevated daily cigarette consumption. In the nonsmoking population, a urinary cotinine concentration >170 ng·mL −1 correlated with a high SHS exposure. Furthermore, among nonsmokers with a low SHS exposure, nonsmoking individuals who had essentially no SHS exposure presented a urinary cotinine level <50 ng·mL −1 .
COPD patients CHEN et al. [bib_ref] Serum metabolite biomarkers discriminate healthy smokers from COPD Smokers, Chen [/bib_ref] detected no statistically significant difference in urinary cotinine levels between COPD patients and healthy controls when both groups were smokers. Both these groups showed higher urinary cotinine levels when compared to healthy controls who were nonsmokers (table 2). A higher urinary cotinine level was associated with greater COPD severity, worsened physical health status and poorer disease-specific quality of life in another study [bib_ref] Assessing secondhand smoke using biological markers, Avila-Tang [/bib_ref]. Asthma patients FERNANDES et al. [bib_ref] Dual exposure to smoking and household air pollution is associated with an..., Fernandes [/bib_ref] assessed a cut-off value for urinary cotinine that was normalised with serum creatinine, in order to take into consideration the influence of renal function on cotinine urinary concentration. The value of 197 µg·g −1 confirmed the smoking status previously declared by the patients. PINHEIRO et al. [bib_ref] Self-reported smoking status and urinary cotinine levels in patients with asthma, Pinheiro [/bib_ref] showed an increasing trend of creatinine-normalised urinary cotinine values among asthmatic patients with the same smoking status according to the worsening of disease severity. Furthermore, high urinary cotinine concentrations were associated with worsened physical health status and poorer disease-related quality of life [bib_ref] Assessing secondhand smoke using biological markers, Avila-Tang [/bib_ref]. Data concerning urinary cotinine are listed in tables 3 and 4.
## Carbon monoxide
Following smoke inhalation, CO displaces oxygen in red blood cells, forming carboxyhaemoglobin. In this form, CO shows a half-life of about 5-6 h and may be detectable for up to 24 h depending on ventilation rate, physical activity and sex, among other factors. Tobacco smoke is not the only source of CO, since passive smoking, environmental pollution and occupational exposure might raise CO levels. Nevertheless, tobacco smoke is the most likely cause of high CO levels. In a small number of cases, CO poisoning or acute haemolysis might cause elevated CO levels [bib_ref] Breath carbon monoxide as an indication of smoking habit, Middleton [/bib_ref].
## Exhaled co
Despite showing very good sensitivity and specificity values, cotinine measurements in blood or urine do not allow an immediate assessment of the smoking status since the detection methods require time. In addition, the quantification of serum cotinine requires an invasive procedure. The exhaled CO measurement allows a direct assessment of the specific patient's smoking status using a noninvasive method requiring <1 min to perform, with an immediate displayed result.
## Healthy controls
Among healthy controls, the cut-offs for determining the smoking habit of a healthy individual are included in a range that varies from 5 to 6 parts per million ( ppm) [bib_ref] Exhaled carbon monoxide levels among tobacco smokers by age, Chatrchaiwiwatana [/bib_ref] [bib_ref] Breath carbon monoxide as an indication of smoking habit, Middleton [/bib_ref]. The cut-off value that showed the best sensitivity and specificity combination is 6 ppm, with 94% sensitivity and 96% specificity. Moreover, exhaled CO showed a positive correlation with the number of smoked cigarettes per day [bib_ref] Breath carbon monoxide as an indication of smoking habit, Middleton [/bib_ref]. Exhaled CO was ∼1.5 ppm in healthy nonsmokers; nevertheless, no evidence about a precise cut-off for SHS exposure was provided [bib_ref] Correlation of exhaled carbon monoxide level with disease severity in chronic obstruction..., Ejazi [/bib_ref]. COPD patients SATO et al. [bib_ref] Optimal cutoff level of breath carbon monoxide for assessing smoking status in..., Sato [/bib_ref] analysed smoking status in COPD patients who were nonsmokers, former smokers or current smokers, correlating exhaled CO levels with serum cotinine concentrations. The exhaled CO levels in COPD patients were 7.7±4.3 and 13.5±6.5 ppm for former and current smokers respectively (data on nonsmokers were not available). The optimal cut-off to distinguish COPD patients who are current smokers from those who are former smokers or nonsmokers was 11 ppm, with 73.1% sensitivity and 84.7% specificity. The lower sensitivity and specificity values might be explained by the disease-related inflammatory environment, which may have altered such measurements, as confirmed by EJAZI et al. [bib_ref] Correlation of exhaled carbon monoxide level with disease severity in chronic obstruction..., Ejazi [/bib_ref] As for SHS exposure, an exhaled CO cut-off in COPD patients has still to be determined.
Asthma patients SATO et al. [bib_ref] Optimal cutoff level of breath carbon monoxide for assessing smoking status in..., Sato [/bib_ref] compared exhaled CO levels in never-smokers, former smokers and active smokers diagnosed with asthma, correlating these with the serum cotinine concentration. The increase in exhaled CO reflected the smoking status, reaching the highest level in active smokers and the lowest in never-smokers, with intermediate values for former smokers. The exhaled CO levels in asthmatic patients were 6.1±2.4, 7.7±3.2 and 19.9±17.3 ppm, respectively, for never-smokers, former smokers and active smokers. A cut-off for distinguishing current smokers from never-smokers in asthmatic patients was 10 ppm, showing 85.0% sensitivity and 85.8% specificity. The exhaled CO concentration correlated positively with serum cotinine levels. As for COPD, the lower sensitivity and specificity values might have been affected by the inflammatory environment, which may have altered such measurements.
An exhaled CO cut-off related to SHS exposure in asthmatic patients has still to be defined. Data concerning exhaled CO are listed in table 5.
# Discussion
Serum cotinine, urinary cotinine and exhaled CO represent efficient methods to define smoking status and to quantify SHS exposure. Among healthy controls, these three biomarkers can predict smoking status with satisfactory sensitivity and specificity values (sensitivity and specificity both >95% for serum cotinine, ⩾93% sensitivity and specificity for urinary cotinine, and ⩾79% sensitivity and ⩾91% specificity for exhaled CO). The available data on serum cotinine come from studies that involved a solid number of patients, and reaching comparable results in terms of sensitivity and specificity. The difference between the proposed cut-offs, 15 and 3.08 ng·mL −1 , might be due to geographical, ethnic or sex-specific factors, which Heavy and light smokers, heavily and lightly exposed nonsmokers, and lightly exposed and non-exposed non-smokers were compared. SHS: second-hand smoke; NA: not available. Data from [bib_ref] Estimation of urinary cotinine cut-off points distinguishing non-smokers, passive and active smokers, Zielinska-Danch [/bib_ref]. even large patient numbers might have not totally addressed in the cohorts. A similar explanation might be considered for the different urinary cotinine cut-offs. Additionally, urinary cotinine represents an especially flexible and useful tool to assess SHS, since it allows distinction between heavily exposed, lightly exposed and essentially unexposed nonsmokers. The cut-offs for exhaled CO are comparable. The lower sensitivity and specificity values, coming from a 10-fold larger patient cohort [bib_ref] Exhaled carbon monoxide levels among tobacco smokers by age, Chatrchaiwiwatana [/bib_ref] , are reasonably closer to generalisability. SHS can be predicted as well with these quantification methods, even if some overlap between light smokers and heavily exposed nonsmokers might occur using serum cotinine as a biomarker.
Among COPD patients, the only cut-off currently available to distinguish between smokers and nonsmokers is exhaled CO, with a value of 11 ppm defined from a cohort of 170 patients. Sensitivity and specificity values were remarkably lower compared to those calculated from cohorts of healthy individuals, probably due to chronic lung inflammation processes. Although COPD smokers and nonsmokers can be distinguished by means of exhaled CO, neither exhaled CO nor the other investigated biomarkers have been strongly associated with SHS exposure. Nevertheless, a higher urinary cotinine level is associated with greater COPD severity, and poorer physical health status and quality of life.
Among asthmatic patients, an exhaled CO cut-off of 10 ppm determined smoking status with good sensitivity and specificity. In a similar way to COPD patients, these values were lower than the ones resulting from the analysis of healthy individual cohorts, probably due to chronic inflammation. Nevertheless, the proposed cut-off showed both higher sensitivity and specificity compared to COPD patients. A possible explanation of this difference might involve years with disease or the age of the subjects, as both these factors affect the inflammatory environment [bib_ref] Exhaled carbon monoxide levels among tobacco smokers by age, Chatrchaiwiwatana [/bib_ref] and gas exchange, especially for COPD patients. Even though a specific serum cotinine cut-off to distinguish asthmatic smokers from nonsmokers has not been defined, serum cotinine correlates positively with exhaled CO. There is not much evidence yet for using urinary cotinine to assess smoking status among adult asthmatic patients since well-defined sensitivity and specificity values are still lacking. In this context, the use of normalised urinary cotinine levels by basing it on creatinine excretion has been proposed [bib_ref] Optimum cutoff value of urinary cotinine distinguishing South Korean adult smokers from..., Kim [/bib_ref]. A quantification of SHS exposure using these biomarkers is not available among adult asthmatic patients. Nevertheless, high serum and urinary cotinine concentrations correlated with worsened lung function and poorer physical status and quality of life.
# Limitations
The use of these biomarkers has various limitations, mostly concerning metabolising capacities. Different cytochrome P450 (CYP)2A6 expression influences nicotine metabolism and this might affect the smoking status assessment of nonsmoking poor metabolisers and high metabolisers who are current smokers [bib_ref] Biomonitoring of urinary cotinine concentrations associated with plasma levels of nicotine metabolites..., Nagano [/bib_ref]. Moreover, distinct ethnic-specific gene expression might also affect CYP2A6 levels [bib_ref] Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different..., Benowitz [/bib_ref]. Specific modifications of the aforementioned cut-offs might be needed also in certain subpopulations of patients receiving therapies that might induce CYP2A6 (e.g. rifampicin, phenobarbital and dexamethasone). Furthermore, in this study, all results regarding pregnant women were excluded due to foreseeable metabolism variations.
Another limitation of this analysis is intrinsically related to cotinine and exhaled CO measurements that, although more objective than questionnaires, could be influenced by clinical, environmental or social settings that may differ among patient populations. For this reason, different cut-offs for different patient populations with the same clinical condition should not be judged a priori as discordant. The timing of measurement in relation to smoke exposure plays an important role but often cannot be influenced. Further studies should assess these issues and address methods for standardisation.
Moreover, some variability of the data relating to SHS exposure may be due to variable definitions of SHS exposure among the cited studies.
Additionally, further research about SHS exposure in adult asthmatic patients is needed, since most of the excluded studies involved asthmatic, nonadult subjects.
Conclusion and future perspectives Serum cotinine, urinary cotinine and exhaled CO measurements are readily available tests that may be used in clinical routine with affordable costs. They represent an easy-to-standardise, more objective and reliable method to assess smoking status and SHS exposure than a standardised questionnaire. Evidence on cut-offs with good sensitivity and specificity values is available mainly for healthy controls. For COPD and asthma patients, most of the currently available evidence focuses on exhaled CO, while large studies on the use of cotinine with clearly defined sensitivity and specificity values are still missing. Solid evidence on SHS exposure is available only for healthy controls. Cut-off values based on the current available data have been summarised.
[table] TABLE 1: Serum cotinine concentrations in healthy controls (HC), COPD patients and asthma patientsThe study by SATO et al.[14] reported only mean values for each group of patients. Sns: sensitivity; Spc: specificity; SHS: second-hand smoke; NA: not available. [/table]
[table] TABLE 2: Urinary cotinine among nonsmokers, healthy smokers and COPD smokers Data are presented as mean±SD unless otherwise stated. D LCO : diffusing capacity of the lung for carbon monoxide. ND: low or not detected. Reproduced and modified from[6] with permission. [/table]
[table] TABLE 3: Urinary cotinine concentrations in healthy controls (HC), COPD patients and asthma patients [/table]
|
An Iterative Approach for the Optimization of Pavement Maintenance Management at the Network Level
Pavement maintenance is one of the major issues of public agencies. Insufficient investment or inefficient maintenance strategies lead to high economic expenses in the long term. Under budgetary restrictions, the optimal allocation of resources becomes a crucial aspect. Two traditional approaches (sequential and holistic) and four classes of optimization methods (selection based on ranking, mathematical optimization, near optimization, and other methods) have been applied to solve this problem. They vary in the number of alternatives considered and how the selection process is performed. Therefore, a previous understanding of the problem is mandatory to identify the most suitable approach and method for a particular network. This study aims to assist highway agencies, researchers, and practitioners on when and how to apply available methods based on a comparative analysis of the current state of the practice. Holistic approach tackles the problem considering the overall network condition, while the sequential approach is easier to implement and understand, but may lead to solutions far from optimal. Scenarios defining the suitability of these approaches are defined. Finally, an iterative approach gathering the advantages of traditional approaches is proposed and applied in a case study. The proposed approach considers the overall network condition in a simpler and more intuitive manner than the holistic approach.
# Introduction
Pavement management systems (PMS) should assist agencies in the decision making process about which sections of a pavement network should be preserved, maintained, and/or rehabilitated (P + M + R) under budgetary constraints. To address this, a systematic and rational method is needed to ensure an optimal allocation of scarce resources. Studies carried out by the Economic Commission for Latin America and the Caribbean (ECLAC) and the German Agency for Technical Cooperation (GTZ) have shown that between 1% and 3% of gross domestic product is consumed each year unnecessarily due to the lack of road network management [bib_ref] Road Network Management: A New Approach to Road Management and Conservation, Economic..., Schliesser [/bib_ref]. At the same time, users are increasingly demanding in terms of quality, comfort, and safety. Therefore, the design of maintenance programs becomes a crucial aspect when defining the questions about which section to treat, which treatment to apply, and when this treatment must be applied. For this analysis, PMS must integrate three management levels varying in the information detail and complexity of models considered in the decision making: strategic, network, and project level. This study focuses on management at the network level, whose primary purpose is the design of the network maintenance program, within overall budget constraints. In order to evaluate the suitability of maintenance programs at the network level, PMS integrate and coordinate four specific modules [fig_ref] Figure 1: Modules in a PMS used to evaluate the suitability of maintenance programs... [/fig_ref].
(i) Input Data. Data required for the network analysis are inventory data per road, network present condition, performance models (including the evolution of pavement condition and the effect of P + M + R treatments), and strategic level data. Strategic level data include strategic targets (i.e., trigger level of service, environmental policies, institutional scopes, and objectives), available budget, analysis period, and discount rate for long term economic analysis.
(ii) Management System Modules. This module evaluates the suitability of maintenance alternatives. If a sustainable management is implemented, the evaluation has to consider, in an integrated manner, five aspects throughout the pavement life cycle: economic, social, technical, environmental, and political [bib_ref] Mixed transportation network design under a sustainable development perspective, Qin [/bib_ref] [bib_ref] Development of a management framework for rural roads in developing countries: integrating..., Chamorro [/bib_ref] [bib_ref] Optimization of high-performance concrete structures by Variable Neighborhood Search, Torres-Machí [/bib_ref]. Different indicators have been used for assessing these sustainable aspects, for example, present worth cost or cost effectiveness as economic indicators, safety and comfort as social indicators, roughness as an indicator of technical performance, air pollution due to vehicle emissions as environmental indicator, and sections' functional class as a political indicator. Once aspects to consider have been selected, managers at the strategic level have to define the decision making criteria, that is, how these sustainable aspects will be considered for evaluating and comparing the suitability of maintenance alternatives.
(iii) Network Analysis. Based on information from input data and management system modules, this module seeks the optimal maintenance program at the network level.
(iv) Output Data. It mainly consists of the maintenance program at the network level and pavement condition over the analysis period. It could also include recommendations to adjust strategic targets.
Once input data and management system modules are defined, the optimal design of maintenance programs is not straightforward. Indeed, it presents × possible solutions in a network of sections with possible P + M + R treatments over a planning horizon of years [bib_ref] Optimum genetic algorithm structure selection in pavement management, Golroo [/bib_ref]. Given this scenario, two approaches have been identified in the literature to allow the optimization process: sequential and holistic. These approaches differ in how the optimization problem is tackled. Sequential approach deals with the problem in two phases. It first defines the treatment strategy on a section by section analysis. Once the P + M + R treatments and timings are defined for each section, it selects the sections to treat until available budget is exhausted. This sequential approach simplifies the problem by evaluating × possible solutions. On the other hand, holistic approach tackles the problem as a whole, before any specific section or treatment strategy is defined. Therefore, it deals with the × solutions of the problem. In addition to these two possible approaches of the problem, different methods can be applied to look for the optimal maintenance program. These optimization methods vary in the number of evaluated alternatives and how the selection is carried out and they can be applied considering either sequential or holistic approach. There is no specific approach and optimization method appropriate to all possible scenarios; therefore, a previous understanding of the problem within an agency is mandatory to identify the most suitable approach for a particular network.
Previous works in other research areas (i.e., bridge management and construction engineering) have analyzed the use of existing optimization methods in the decision making process [bib_ref] A multimethodology contractor assessment model for facilitating green innovation: the view of..., Hsueh [/bib_ref] [bib_ref] Unsupervised performance evaluation strategy for bridge superstructure based on fuzzy clustering and..., Jiao [/bib_ref]. With respect to pavement management, previous works have analyzed how pavement agencies select the projects to undertake. In 1995, simplistic approaches were detected in most of the American and Canadian agencies. This was mainly due to the lack of computers to undertake more reliable analysis. Recent advances in computer technologies make necessary an update of optimization methods to tackle the problem of designing maintenance programs. A recent study reviews optimization techniques applied to pavement management with special attention to genetic algorithms. However, this review neither compares existing optimization methods nor considers the traditional approaches of the problem (sequential or holistic). Having detected this gap in the literature, the present study aims to assist highway agencies, researchers, and practitioners on when to apply and how to apply available optimization methods for the optimal design of maintenance programs at network level.
The objective of this study is to recommend the most suitable approaches and optimization methods for the design of maintenance programs under different scenarios. Based on an analysis of the current state of the practice, this study proposes an iterative approach that gathers the advantages of traditional approaches (sequential and holistic) by considering the overall network condition in a simpler and more intuitive manner than with a holistic approach.
The study is part of a three-year project developed in Chile by the Pontificia Universidad Católica de Chile (PUC) and named Fondef D09I1018 "Investigación y Desarrollo de Soluciones para la Gestión de Pavimentos Urbanos en Chile" (Research and Development of Solutions for Urban Pavement Management in Chile). The project is being partnered and advised by the Centre for Pavement and Transportation Technology (CPATT) of the University of Waterloo, Canada. The overall project resulted as a cooperative initiative of the PUC and funding partners to accomplish the current and future needs of urban pavements and provide effective The Scientific World Journal 3 management tools to assist agencies that manage urban networks in decision making. Even though the project is being developed in Chile, the expected outcomes, such as technical tools and the resulting Urban Pavement Management System, may be adapted and adopted in other countries for urban pavement management.
# Study methodology
To achieve the proposed objective, a four-step research method is proposed:
(i) review of existing optimization methods applied to pavement management including existing applications in available PMS under traditional approaches (sequential and holistic);
(ii) comparative analysis of reviewed approaches and optimization methods, identifying their advantages and limitations; based on this analysis, recommendations of the most suitable approach and optimization method to implement in future PMS are driven considering different scenarios;
(iii) proposal of an iterative approach gathering the advantages of traditional approaches;
(iv) application of the proposed iterative approach in an illustrative case study and comparison to traditional approaches.
## Selection and optimization of maintenance treatments
As shown in [fig_ref] Table 1: Reviewed optimization methods consider either sequential or holistic approach [/fig_ref] , several optimization methods are available for the design of maintenance programs at the network level. They mainly vary in the number of alternatives considered. Therefore, they are related to the considered approach: sequential approach deals with × alternatives, while holistic approach considers × possible solutions [fig_ref] Table 2: Number of alternatives and type of approach considered in reviewed applications [/fig_ref]. Indeed, the suitability of an optimization method relies on the number of alternatives under evaluation: when the number of alternatives is small, they can be selected based on a ranking. In contrast, when the number of alternatives is large, it becomes necessary to use mathematical or near optimization methods. These optimization methods are reviewed in this section including applications in available PMS.
## Selection based on ranking.
Selection based on ranking is performed by enlisting and rating alternatives based on an indicator. This indicator can be based on judgment, pavement condition, or economic analysis.
When based on judgment, agencies determine from an expert panel a criterion to rate and rank alternatives. Shah et al. [bib_ref] Evaluation of prioritization methods for effective pavement maintenance of urban roads, Shah [/bib_ref] applied this method to select the sections to treat in a road network in India considering criteria such as traffic, connectivity, and road and drainage conditions. Selection based on pavement condition ranks sections to treat considering either a Single or a Composite Condition Index. Single Condition Index is normally based on roughness or structural index, whereas Composite Condition Index often considers pavement condition and functional classification. A Composite Condition Index considering pavement surface distresses, traffic information, and expert opinion is used by Reddy and Veeraragavan [bib_ref] Priority ranking model for managing flexible pavements at network level, Reddy [/bib_ref] for selecting sections to treat on a network of 52 sections.
Ranking based on economic analysis allows a rational comparison among alternatives because it considers costs and benefits. This method was used by Shah et al. [bib_ref] Evaluation of prioritization methods for effective pavement maintenance of urban roads, Shah [/bib_ref] under a sequential approach. Firstly, for each section, four maintenance strategies were ranked based on a benefit cost ratio. This economic indicator was also used to rank sections to treat. Another application of ranking based on economic analysis has been implemented in MicroPaver PMS using the cost effectiveness for selecting the sections to treat [bib_ref] Sensitivity analysis of the application of dynamic programming to pavement management systems, Feighan [/bib_ref].
# Mathematical optimization methods.
Mathematical optimization methods select alternatives maximizing or minimizing an objective function while satisfying some constraints. Objective functions commonly considered are maintenance costs, vehicle operating costs, and effectiveness, among others [bib_ref] Optimum genetic algorithm structure selection in pavement management, Golroo [/bib_ref] [bib_ref] Development of maintenance standards for rural networks based on cost-effectiveness analysis, Chamorro [/bib_ref]. Mathematical programming methods most commonly used for pavement management are linear, nonlinear, integer, and dynamic programming. They are discussed in detail below.
Linear and nonlinear programming seek optimal solutions using continuous variables. The main difference is that the former considers linear functions correlated with time, while the latter may consider curvilinear dependency. These optimization methods have been applied using both holistic [bib_ref] Network-level optimization of pavement maintenance renewal strategies, De La [/bib_ref] [bib_ref] Network-level road pavement maintenance and rehabilitation scheduling for optimal performance improvement and..., Gao [/bib_ref] and sequential approaches [bib_ref] Roads performance modeling and management system from two condition data points: case..., Amador-Jiménez [/bib_ref] [bib_ref] Robust optimization for managing pavement maintenance and rehabilitation, Gao [/bib_ref]. Under the sequential approach, these methods have been used to optimize the treatment strategy at the section level [bib_ref] Robust optimization for managing pavement maintenance and rehabilitation, Gao [/bib_ref] and the sections to treat at the network level [bib_ref] Roads performance modeling and management system from two condition data points: case..., Amador-Jiménez [/bib_ref].
Integer programming simplifies the analysis by considering two variables: a do nothing alternative or to do something. Applications are found using both sequential [bib_ref] The price of uncertainty in pavement infrastructure management planning: an integer programming..., Ng [/bib_ref] [bib_ref] of Highway Development and Management Series, International Study of Highway Development and..., Odoki [/bib_ref] and holistic approaches [bib_ref] Probabilistic segment-linked pavement management optimization model, Ferreira [/bib_ref] [bib_ref] Decision-making problem for managing pavement maintenance and rehabilitation projects, Wang [/bib_ref]. Regarding sequential approach, Ng et al. [bib_ref] The price of uncertainty in pavement infrastructure management planning: an integer programming..., Ng [/bib_ref] optimized the treatment strategy at the section level, while Odoki and Kerali [bib_ref] of Highway Development and Management Series, International Study of Highway Development and..., Odoki [/bib_ref] implemented this procedure in HDM-4 PMS to select sections to treat at the network level. However, due to the complexity of the calculation, HDM-4 limits this analysis to networks with less than 100 sections, 16 treatments, and five years [bib_ref] of Highway Development and Management Series, International Study of Highway Development and..., Odoki [/bib_ref].
Dynamic programming is used in situations that require a number of sequential decisions. This optimization method starts at the desired final solution and works backwards to find the optimal value of variables. Dynamic programming has been applied using holistic [bib_ref] Cost-effective selection and multiperiod scheduling of pavement maintenance and rehabilitation strategies, Yoo [/bib_ref] and sequential approaches for the treatment strategy optimization [bib_ref] Sensitivity analysis of the application of dynamic programming to pavement management systems, Feighan [/bib_ref] and the section selection [bib_ref] Incorporating priority preferences into pavement maintenance programming, Farhan [/bib_ref] [bib_ref] Optimal multiasset maintenance budget allocation in Highway asset management, Fwa [/bib_ref].
## Near optimization or heuristic methods.
Near optimization methods, also called heuristic methods, give solutions that are close approximations to those derived from mathematical optimization. These optimization methods start with an initial solution and look for better solutions within 4 The Scientific World Journal x [bib_ref] Cost-effective selection and multiperiod scheduling of pavement maintenance and rehabilitation strategies, Yoo [/bib_ref] the constraints. They differ in how they search for better solutions: incremental benefit/cost analysis, local search heuristics, and evolutionary algorithms. Incremental benefit/cost analyzes the benefits gained by selecting alternatives with higher costs. This optimization method is often referred to as the efficiency frontier. This frontier is defined in a plot of benefit against cost and gathers the alternatives with higher benefits given a certain cost. Incremental benefit/cost analysis is implemented in HDM-4 PMS under a sequential approach for selecting the maintenance strategy and the sections to treat. However, HDM-4 limits this application to a maximum of 400 sections, The Scientific World Journal 5 17 treatment alternatives, and a 12-year analysis period [bib_ref] of Highway Development and Management Series, International Study of Highway Development and..., Odoki [/bib_ref]. Incremental benefit/cost has also been applied to prioritize sections to treat in unpaved networks.
Local search heuristics start with random initial solution and explore the solution inference space seeking for better feasible solutions. Different local search heuristic can be implemented: gradient search, threshold acceptance, simulated annealing, and so forth. These heuristics have been applied under a sequential approach to optimize a road section treatment strategy: Tsunokawa et al. [bib_ref] True optimization of pavement maintenance options with what-if models, Tsunokawa [/bib_ref] looked for the overlay thickness maximizing benefits, while Chou and Le [bib_ref] Reliability-based performance simulation for optimized pavement maintenance, Chou [/bib_ref] considered two objectives, minimal cost and maximal reliability.
Evolutionary algorithms (EA) mimic the natural evolution guided by learning and adaptation. Among EA, genetic algorithms are one of the most applicable optimization methods in infrastructure management [bib_ref] Soft computing applications in infrastructure management, Flintsch [/bib_ref]. They have been applied using both sequential (optimal treatment strategy [bib_ref] Optimal multiasset maintenance budget allocation in Highway asset management, Fwa [/bib_ref] and section selection [bib_ref] Optimal fund-allocation analysis for multidistrict highway agencies, Chan [/bib_ref] and holistic approaches [bib_ref] Incorporating priority preferences into pavement maintenance programming, Farhan [/bib_ref] [bib_ref] A multi-year pavement maintenance program using a stochastic simulation-based genetic algorithm approach, Chootinan [/bib_ref] [bib_ref] Pavement maintenance programming considering two objectives: maintenance costs and user costs, Meneses [/bib_ref].
## Other optimization
Methods. This section gathers optimization methods that assist decision making in managing pavement maintenance at the network level but they cannot be categorized in the above groups as neural networks and fuzzy logic.
Neural networks are able to learn from examples, enabling these systems to make generalizations and simulate decisions. Fwa and Chan [bib_ref] Priority rating of highway maintenance needs by neural networks, Fwa [/bib_ref] developed a neural network based on the priority ratings awarded by engineers. After the training phase, it provided rating scores to road sections based on their condition.
Fuzzy logic systems incorporate imprecise qualitative data in the decision making. Moazami et al. [bib_ref] Pavement rehabilitation and maintenance prioritization of urban roads using fuzzy logic, Moazami [/bib_ref] applied a fuzzy logic system in a network with 131 sections. In this system, sections were characterized by condition, traffic, road width, and treatment cost. The fuzzy logic system transformed the quantitative values of these parameters into linguistic values that classified priority in three levels: low, medium, or high.
## Comparative analysis and recommendations
This section analyzes the advantages and limitations of reviewed optimization methods and recommends the most suitable methods and approaches for future implementation in PMS under various scenarios. Finally, an iterative approach gathering the advantages of sequential and holistic approaches is proposed.
## Advantages and limitations of the reviewed optimization
Methods. Selection based on ranking is easy to understand but it can only deal with a limited number of alternatives. Ranking based on judgment is the simplest method, but it may be subject to bias and inconsistency resulting in solutions far from optimal. Ranking based on condition is more objective than judgment. However, it does not consider the effect of alternatives on pavement condition over time. Therefore, Amador-Jiménez and Mrawira [bib_ref] Roads performance modeling and management system from two condition data points: case..., Amador-Jiménez [/bib_ref] have defined optimization based on ranking as the least cost effective method. Finally, ranking based on economic analysis allows a rational comparison that includes costs and benefits. Being simple to use, it leads to solutions closer to optimal than other ranking methods. However, it fails to consider externalities that are difficult to monetize. Mathematical optimization methods provide optimal solutions but they are not suited to deal with large networks. Indeed, mathematical optimization methods cannot handle large number of decision variables because this increases the complexity of the problem and requires long computing time.
In contrast, near optimization or heuristic methods provide simpler and more efficient solutions to large optimization problems. They are suitable to tackle the maintenance management at the network level leading to "good/near optimal" solutions [bib_ref] Optimum genetic algorithm structure selection in pavement management, Golroo [/bib_ref]. Heuristic methods, however, should be compared to optimization methods to ensure that they are representing optimal or near optimal solutions.
Regarding other optimization methods, neural networks are useful to replicate a pattern and make generalizations. However, they do not guarantee the suitability of the decision taken and they act as a "black box, " being not possible to easily extract the path followed to explain a solution. Finally, fuzzy logic enable introducing rules from experience or intuition but it has no formal algorithms to learn from existing data [bib_ref] Soft computing applications in infrastructure management, Flintsch [/bib_ref].
## Recommendations of approaches and optimization methods.
Reviewed applications show a greater reliance on mathematical optimization and near optimization rather than on ranking [fig_ref] Table 1: Reviewed optimization methods consider either sequential or holistic approach [/fig_ref]. In addition to this, applications highlight the versatility of the optimization methods for solving the problem. On one hand, an optimization method (dynamic programming, e.g.) can be implemented under a sequential (A7, A8, and A11 in Tables 1 and 2) or a holistic approach (A23 in [fig_ref] Table 1: Reviewed optimization methods consider either sequential or holistic approach [/fig_ref]. On the other hand, different optimization methods can be combined for solving the same problem. For example, Fwa and Farhan [bib_ref] Optimal multiasset maintenance budget allocation in Highway asset management, Fwa [/bib_ref] (A11 in Tables 1 and 2) used a genetic algorithm for optimizing the treatment strategy and subsequently applied dynamic programming for optimizing section selection, while Feighan et al. [bib_ref] Sensitivity analysis of the application of dynamic programming to pavement management systems, Feighan [/bib_ref] (A8 in Tables 1 and 2) combined dynamic programming and selection based on ranking. Therefore, the suitability of an optimization method should be analyzed in conjunction with the approach of the problem.
Sequential approach simplifies the problem making it easier to understand than holistic approach because it defines first the treatment strategy and then selects the sections to treat. Nevertheless, sequential approach ignores the effect on the network as a whole. This may lead to recommending sequential approach for homogeneous or reduced networks in which the overall performance is less compromised by the section by section analysis. Regarding optimization methods, all the analyzed methods (ranking, optimization, and near optimization) have been used under the sequential approach. The recommendation on the optimization method to use would depend on the characteristics of a specific problem. [fig_ref] i: Holistic Approach [/fig_ref] : Decision making process of the proposed iterative approach.
In broad terms, selection based on judgment or condition should be avoided, as they may introduce bias and do not consider the effect of alternatives over time.
Holistic approach enables analyzing network maintenance alternatives as a whole, before any specific treatment strategy or section has been selected. However, this increases the complexity of the problem, making it necessary to use optimization and near optimization methods [fig_ref] Table 1: Reviewed optimization methods consider either sequential or holistic approach [/fig_ref]. Reviewed applications using mathematical optimization methods show a trend of limiting the number of variables considered while near optimization methods are able to handle the problem without sacrificing its complexity. For example, the most complex application using optimization considered 6 108 alternatives [bib_ref] Probabilistic segment-linked pavement management optimization model, Ferreira [/bib_ref] , A9 in [fig_ref] Table 1: Reviewed optimization methods consider either sequential or holistic approach [/fig_ref] , while near optimization has optimized a problem with 7 640 alternatives [bib_ref] Pavement maintenance programming considering two objectives: maintenance costs and user costs, Meneses [/bib_ref] , A14 in [fig_ref] Table 1: Reviewed optimization methods consider either sequential or holistic approach [/fig_ref]. Therefore, holistic approach using near optimization methods may be recommended to deal with large networks. Nevertheless, results are suggested to be periodically compared to mathematical optimization methods.
## Proposed iterative approach.
Gathering the advantages of sequential and holistic approaches, an iterative approach is proposed as shown in [fig_ref] i: Holistic Approach [/fig_ref]. Based on a sequential structure, the iterative approach includes iterations between the selection of treatment strategies and sections looking for a more holistic view of the problem. The proposed approach optimizes first the treatment strategy at the section level. In this optimization, the iterative approach collects a set of optimal and suboptimal solutions. This set of solutions is then considered when selecting the sections to treat at the network level. In this selection, iterations are made looking for the optimal selection of treatment strategies and sections while satisfying budgetary restrictions.
The main difference between sequential and iterative approach is that the latter may select suboptimal treatment strategies for a certain section. On the contrary, reviewed applications considering a sequential approach only consider optimal solutions in the selection of treatment strategies. Therefore, iterative approach enables a deterioration of a solution at the section level if it leads to an improvement of the overall solution at the network level. As a result, the proposed iterative approach considers the overall network condition in a simpler and more intuitively manner than holistic approach.
Several of the reviewed optimization methods (ranking, optimization, and near optimization) may be used considering the proposed iterative approach. As the proposed approach considers two optimizations (treatment strategy and section selection), reviewed optimization methods may be combined. Indeed, iterations are also considered in the reviewed incremental benefit/cost analysis, as shown in the application of Videla and Gaete [bib_ref] Una metodología de optimización económica para la definición de políticas de conservación..., Videla [/bib_ref]. The main difference is that incremental benefit/cost analysis only compares solutions in the efficiency frontier. In contrast, the proposed iterative approach may select suboptimal solutions that are not placed in the efficiency frontier. The recommendation of the most suitable optimization method would depend on the number of alternatives to evaluate. If the number of alternatives is reduced, ranking based on economic analysis may be suitable. In contrast, optimization or near optimization methods would be more suitable when dealing with a large number of alternatives. Finally, as the proposed iterative approach considers two optimizations, it enables considering different criteria or objective functions in the selection of treatment strategy and sections to treat. This would therefore facilitate the sustainable management of pavement networks because sustainable aspects, such as economic, social, technical, environmental, and political, may be considered in different optimizations depending on whether they impact at the network or project level.
## Case study
An illustrative case study is presented to compare the maintenance program obtained under traditional approaches The Scientific World Journal 7 (holistic and sequential) and the proposed iterative approach. The analyzed network, composed of five flexible pavements, is subject to both technical and budgetary restrictions. Each of the sections has a set of six possible rehabilitation treatments and a deterministic deterioration model adopted from Khurshid et al. [bib_ref] Multidimensional benefit-cost evaluation of asphaltic concrete overlays of rigid pavements, Khurshid [/bib_ref]. Pavement condition is assessed in terms of Present Serviceability Index (PSI) using the regressions proposed by Hall et al. [bib_ref] LTPP Data Analysis: Effectiveness of Maintenance and Rehabilitation Options, Hall [/bib_ref]. All the sections present similar characteristics in terms of geometry (1000 m length and 3.7 m width), climate (Average Annual Freeze Index of 400 Celsius days), and traffic (Average Annual Truck Traffic Volume of 0.8 million) while differing on their initial condition [fig_ref] Table 3: Characteristics of sections considered in the case study [/fig_ref]. The maintenance program seeks to maximize long term effectiveness (LTE) over a period of 25 years subject to budgetary restrictions. LTE of maintenance alternatives is assessed by the area bounded by the pavement performance curve (ABPC) and a threshold value (PSI ≥ 2, in this case study) [fig_ref] Figure 3: Long term effectiveness of a maintenance alternative [/fig_ref] , weighted by traffic and section length [bib_ref] Development of maintenance standards for rural networks based on cost-effectiveness analysis, Chamorro [/bib_ref] [bib_ref] Comparison of methods for evaluating pavement interventions: evaluation and case study, Khurshid [/bib_ref] [bib_ref] Determining return on long-life pavement investments, Haas [/bib_ref] (1). The measure of ABPC is based on the fact that a well-maintained pavement (having therefore a larger LTE) provides greater benefits than a poorly maintained pavement [bib_ref] Multidimensional benefit-cost evaluation of asphaltic concrete overlays of rigid pavements, Khurshid [/bib_ref] [bib_ref] Comparison of methods for evaluating pavement interventions: evaluation and case study, Khurshid [/bib_ref] [bib_ref] Determining return on long-life pavement investments, Haas [/bib_ref]. In order to compare alternatives with different costs, the ratio cost effectiveness (C/E) (2) is normally considered [bib_ref] Development of maintenance standards for rural networks based on cost-effectiveness analysis, Chamorro [/bib_ref] [bib_ref] Determining return on long-life pavement investments, Haas [/bib_ref] :
[formula] LTE = ABPC ⋅ ⋅ AADT,(1)= LTE TPWC ,(2) [/formula]
where LTE = long term effectiveness, ABPC = area bounded by performance curve and a threshold value [fig_ref] Figure 3: Long term effectiveness of a maintenance alternative [/fig_ref] , L = section length, AADT = annual average daily traffic, C/E = cost effectiveness, and TPWC = total present worth cost. This case study considers an available budget (in terms of total present worth cost, TPWC) 50% higher than the minimal cost solution that ensures a PSI greater than 2. This minimal cost solution (58 220 C) is taken as a base case to compare solutions obtained using different approaches. Although this case study considers an available budget higher than the minimal cost scenario, the proposed approach could deal with lower budgets. In fact, the ultimate goal of the proposed iterative approach is to assist pavement managers on the optimal design of maintenance programs subject to budgetary restrictions. Therefore, other budgetary scenarios could be similarly considered.
A local search heuristic based on simulated annealing was implemented on Matlab 12 in order to look for optimal solutions. Simulated annealing is based on the analogy of crystal formation from masses melted at high temperature and let to cool slowly [bib_ref] Optimization by simulated annealing, Kirkpatrick [/bib_ref]. This method presents the advantage of escaping from local optima by enabling, under some conditions, the degradation of a solution. This heuristic method, previously implemented by the authors, has led to successful results in large optimization problems [bib_ref] Design of prestressed concrete precast road bridges with hybrid simulated annealing, Martí [/bib_ref] [bib_ref] Heuristic optimization model for infrastructure asset management, Torres-Machí [/bib_ref]. Nevertheless, other optimization methods could be implemented. Indeed, the objective of this application is to analyze the effect of reviewed approaches and not to assess the suitability of the optimization method. The optimization process developed under each approach is the following. (ii) Sequential Approach. It first optimizes the maintenance strategy at the section level by maximizing the incremental cost effectiveness (IC/E, hereafter) (4) while satisfying PSI ≥ 2. Once the optimal maintenance strategy is defined for each pavement, sections with higher IC/E (4) are selected until the budget is depleted.
(iii) Iterative Approach. it considers the set of three best maintenance strategies for each section based on their incremental cost effectiveness (IC/E) (4) assuring that PSI ≥ 2. This set of optimal and suboptimal maintenance strategies is then considered at the network level looking for the maximal LTE (1) while satisfying budget constraint. Therefore, this approach enables the deterioration of a solution at the section level if this would enhance the overall LTE:
[formula] IC = TPWC − TPWC 0 ,(3)IC = LTE − LTE 0 TPWC − TPWC 0 ,(4) [/formula]
where IC = incremental cost of the alternative compared to minimal cost alternative, TPWC = total present worth of the 8 The Scientific World Journal (1)), and LTE 0 = long term effectiveness of minimal cost alternative (assessed by (1)). Sequential and iterative approaches tackle the design of maintenance program by optimizing first the incremental cost effectiveness (IC/E, (4)) of maintenance strategies for each section in the network [fig_ref] Table 4: Optimal and suboptimal treatment strategies considered in the iterative approach [/fig_ref]. The main difference is that, when selecting the sections to treat at the network level, reviewed applications considering a sequential approach only retain optimal maintenance strategies (optimal solutions in [fig_ref] Table 4: Optimal and suboptimal treatment strategies considered in the iterative approach [/fig_ref]. Meanwhile, iterative approach considers optimal and suboptimal solutions.
Considering that the available budget is 50% higher than the minimal cost solution, there is an additional budget of 29 110 C. With this budgetary restriction for improving the network from the minimal cost scenario, sequential approach will solve the optimization problem by only treating Section 2 with the optimal solution [fig_ref] Table 5: Treatment strategies for the different sections of the network under different approaches [/fig_ref]. It is the first ranked solution in terms of IC/E and the cost of the next ranked solution (Section 3: optimal solution) would exceed available budget. However, this solution does not deplete available funds because there is no other optimal solution whose cost does not exceed available budget [fig_ref] i: Holistic Approach [/fig_ref].
Iterative approach, in contrast, enables the selection of suboptimal solutions at the section level looking for an increase in overall performance [fig_ref] i: Holistic Approach [/fig_ref]. Therefore, iterative approach will treat Section 2 with its optimal solution but it will also treat Section 1 with suboptimal solution 2 [fig_ref] Table 5: Treatment strategies for the different sections of the network under different approaches [/fig_ref]. This combination results in a total cost closer to the available budget than sequential approach solution [fig_ref] i: Holistic Approach [/fig_ref]. As a result, a higher performance at the network level in terms of average PSI is obtained under iterative approach [fig_ref] i: Holistic Approach [/fig_ref].
Finally, holistic approach selects a maintenance program based on minimum cost solution except of Section 2, which is treated with a strategy different to those defined as optimal and suboptimal in the section by section analysis. This treatment is referred to as "holistic optimal" [fig_ref] Table 5: Treatment strategies for the different sections of the network under different approaches [/fig_ref]. In relation to costs and effectiveness, holistic approach nearly depletes available budget [fig_ref] i: Holistic Approach [/fig_ref] but it does not necessarily ensure an efficient allocation of funds. Indeed, iterative approach leads to a better solution in terms of average PSI with a lower cost than holistic approach [fig_ref] i: Holistic Approach [/fig_ref].
From this numerical application it can be concluded that the proposed iterative approach leads to more efficient solutions than sequential approach while considering the overall network condition in a simpler and more intuitive manner than holistic approach.
# Conclusions
From the literature review, two approaches are identified in the design of maintenance programs at the network level: holistic and sequential approach. The former tackles the problem as a whole, before any specific section or treatments are defined, dealing with the × possible solutions of the problem. Sequential approach tackles the problem considering a two-step process. It first optimizes the maintenance strategy at the section level. Then, budget is allocated across various sections in the network. This process simplifies the problem to × alternatives. Different optimization methods can be applied in the design of the maintenance programs at the network level considering either sequential or holistic approach: selection based on ranking, mathematical optimization, near optimization, or heuristic methods and other optimization methods.
The Scientific World Journal From the revision of these optimization methods and their applications the following can be concluded.
(i) Ranking systems are easy to understand, but they can only be used when the number of alternatives is limited and they often ignore future needs.
(ii) Mathematical optimization methods provide optimal solutions, but they require long computing time. They may not be feasible for a large network with long period of analysis.
(iii) Near optimization methods give near optimal solutions with less computational effort than mathematical optimization methods. They can handle large number of decision variables and are suitable to solve combinatorial optimization problems.
(iv) Other optimization methods, such as neural networks and fuzzy logic can replicate a pattern, but they do not guarantee the suitability of the decision taken.
Based on the advantages and limitations of the reviewed optimization methods and their applications under holistic and sequential approaches, several recommendations can be driven for future implementation in PMS.
(i) Sequential approach is easy to understand but it fails to consider the effect on the network as a whole. It may be recommended for the analysis of homogeneous or reduced networks, in which the overall performance of the network is less compromised by the section by section analysis.
(ii) Holistic approach analyzes network maintenance alternatives as a whole, before any specific treatment strategy or section has been selected. However, this increases the complexity of the problem, making it necessary to use optimization and near optimization methods. Reviewed applications of mathematical optimization methods using holistic approach show a trend of simplifying the problem or limiting the number of variables (sections, treatments, and/or analysis period). Meanwhile, near optimization methods are able to solve the problem under a holistic approach with no sacrificing of its complexity. Therefore, holistic approach using near optimization methods may be recommended when dealing with large networks.
Finally, an iterative approach is proposed and applied to an illustrative case study. This approach gathers the advantages of sequential and holistic approaches leading to more intuitive and effective design of maintenance programs at the network level. Based on a sequential structure, the proposed iterative approach includes iterations between the selection of treatment strategies and sections to treat looking for a more holistic view of the problem. In this iteration process, the proposed iterative approach may select suboptimal treatment strategies for a certain section if it leads to an improvement of the overall solution at the network level. As a result, the proposed iterative approach considers the overall network condition in a simpler and more intuitive manner than holistic approach.
[fig] Figure 1: Modules in a PMS used to evaluate the suitability of maintenance programs at the network level. [/fig]
[fig] Figure 3: Long term effectiveness of a maintenance alternative. [/fig]
[fig] i: Holistic Approach. It consists of the optimization of LTE (1) satisfying both technical and budgetary restrictions (PSI ≥ 2 and TPWC ≤ 87 330 C, resp.). [/fig]
[table] Table 1: Reviewed optimization methods consider either sequential or holistic approach.Note. Code reference (A1, A2, . . ., A23) is defined inTable 2. [/table]
[table] Table 2: Number of alternatives and type of approach considered in reviewed applications. [/table]
[table] Table 3: Characteristics of sections considered in the case study. [/table]
[table] Table 4: Optimal and suboptimal treatment strategies considered in the iterative approach. [/table]
[table] Table 5: Treatment strategies for the different sections of the network under different approaches.Note. MC corresponds to minimal cost treatment strategy. [/table]
|
BRK phosphorylates SMAD4 for proteasomal degradation and inhibits tumor suppressor FRK to control SNAIL, SLUG, and metastatic potential
The tumor-suppressing function of SMAD4 is frequently subverted during mammary tumorigenesis, leading to cancer growth, invasion, and metastasis. A long-standing concept is that SMAD4 is not regulated by phosphoryl ation but ubiquitination. Our search for signaling pathways regulated by breast tumor kinase (BRK), a nonreceptor protein tyrosine kinase that is up-regulated in ~80% of invasive ductal breast tumors, led us to find that BRK competitively binds and phosphorylates SMAD4 and regulates transforming growth factor-/SMAD4 signaling pathway. A constitutively active BRK (BRK-Y447F) phosphorylates SMAD4, resulting in its recognition by the ubiquitinproteasome system, which accelerates SMAD4 degradation. Activated BRK-mediated degradation of SMAD4 is associated with the repression of tumor suppressor gene FRK and increased expression of mesenchymal markers, SNAIL, and SLUG. Thus, our data suggest that combination therapies targeting activated BRK signaling may have synergized the benefits in the treatment of SMAD4 repressed cancers.
# Introduction
Breast tumor kinase (BRK) is a nonreceptor tyrosine kinase highly expressed in most breast cancer cell lines and tumors [bib_ref] Targeting BRK-positive breast cancers with small-molecule kinase inhibitors, Jiang [/bib_ref]. It displays a similar architecture and 30 to 40% sequence identity with Src family kinases. It is composed of an Src homology three domain (SH3 domain), an SH2 domain, and a catalytic tyrosine kinase domain. Similar to Src, BRK is negatively regulated by phosphorylation of its C-terminal tyrosine-447 and activated by phosphorylation of tyrosine-342 in the catalytic domain. Mutation of tyrosine-447 to phenylalanine substantially enhances the kinase activity of BRK [bib_ref] Constitutive activation of breast tumor kinase accelerates cell migration and tumor growth..., Miah [/bib_ref].
BRK has been implicated in several signaling cascades, notably in mitogenic signaling [bib_ref] Building a better understanding of the intracellular tyrosine kinase PTK6 -BRK by..., Brauer [/bib_ref]. It has been shown to enhance the mitogenic signals of epidermal growth factor (EGF) by promoting the activation of Akt [bib_ref] Expression of the BRK tyrosine kinase in mammary epithelial cells enhances the..., Kamalati [/bib_ref]. Various EGF receptor ligands, including EGF and heregulin, have been shown to stimulate BRK activity, resulting in increased cell proliferation and migration [bib_ref] Breast tumor kinase (protein tyrosine kinase 6) regulates heregulin-induced activation of ERK5..., Ostrander [/bib_ref] [bib_ref] Brk activates Rac1 and promotes cell migration and invasion by phosphorylating paxillin, Chen [/bib_ref]. Consistent with this, BRK has been shown to promote human EGF receptor 2 (HER2)-induced tumorigenesis in orthotropic transplantation-based models [bib_ref] Brk is coamplified with ErbB2 to promote proliferation in breast cancer, Xiang [/bib_ref]. We have also reported that BRK activation significantly enhances tumor formation in xenograft models [bib_ref] Constitutive activation of breast tumor kinase accelerates cell migration and tumor growth..., Miah [/bib_ref]. In addition, BRK is overexpressed in over 80% of breast carcinomas (1) and in many other major cancer types including lung [bib_ref] Detection of Brk expression in non-small cell lung cancer: Clinicopathological relevance, Fan [/bib_ref] , ovarian [bib_ref] The BRK tyrosine kinase is expressed in high-grade serous carcinoma of the..., Schmandt [/bib_ref] , and pancreatic (10) cancers. Although the cellular role of BRK in carcinogenesis has been established, its role in controlling signal transduction pathways is still unclear.
Here, we first used a kinome array [bib_ref] Genome to kinome: Species-specific peptide arrays for kinome analysis, Jalal [/bib_ref] to elucidate the role of BRK in regulating signal transduction pathways and identified components of the transforming growth factor- (TGF-)/SMAD signaling pathway as candidate BRK targets. Although the major molecular components of the TGF-/SMAD signaling pathway are known, the dynamics of TGF-/SMAD signaling remains unclear in many systems, including normal and cancer cells. Current evidence supports that upon TGF-/bone morphogenetic protein receptor activation, SMAD2 and SMAD3 or SMAD1, SMAD5, and SMAD8 bind SMAD4 forming SMAD complexes, which translocate into the nucleus to initiate gene regulation [bib_ref] Mechanistic insight into contextual TGF- signaling, Zhang [/bib_ref] [bib_ref] CBL enhances breast tumor formation by inhibiting tumor suppressive activity of TGF-..., Kang [/bib_ref]. Although TGF-/SMAD signaling networks have played pivotal roles in biological processes, disruption of their signaling has been implicated in several developmental disorders and diseases including cancers [bib_ref] Mechanistic insight into contextual TGF- signaling, Zhang [/bib_ref]. The TGF- signaling pathway can play contradictory roles during tumor development. It can function to suppress tumorigenesis, impeding the proliferation of transformed cells during the early stage of tumorigenesis [bib_ref] Mechanistic insight into contextual TGF- signaling, Zhang [/bib_ref] [bib_ref] TGF- tumor suppression through a lethal EMT, David [/bib_ref]. In contrast, in some advanced cancers, loss-of-function mutations or low expression of SMAD2, SMAD3, and SMAD4 have been observed leading to the suppression of the TGF- signaling pathway [bib_ref] Mechanistic insight into contextual TGF- signaling, Zhang [/bib_ref] [bib_ref] Genetic changes in the transforming growth factor beta (TGF-) type II receptor..., Park [/bib_ref] and uncontrolled cell proliferation [bib_ref] TGF- tumor suppression through a lethal EMT, David [/bib_ref]. Moreover, SMAD2 and SMAD4 are being listed among the 127 most mutated genes in 12 major cancer types [bib_ref] Mutational landscape and significance across 12 major cancer types, Kandoth [/bib_ref].
Given the complexity of TGF-/SMAD signaling networks in biological processes, we set out to investigate how a nonreceptor tyrosine kinase such as BRK may regulate the function and network of the TGF-/SMAD signaling pathway. We used complementary immunological, biochemical, genomics, and proteomics approaches to understand how BRK might regulate TGF-/SMAD protein interaction networks and signal transduction pathways. Collectively, our results provide evidence that BRK regulates the TGF-/SMAD signal transduction pathway in cancer and normal cells. BRK-mediated phosphorylated SMAD4 is targeted for proteasomal degradation, resulting in down-regulation of the tumor suppressor Fyn-related kinase (FRK) and up-regulation of the epithelial-mesenchymal transition (EMT) markers SNAIL and SLUG in cancer cells. Together, http://medicalgenome.kribb.re.kr/GENT/reference.php). We found that the expression of BRK mRNA was significantly higher (P ≤ 0.05) in all five cancer types that we queried compared to their respective noncancerous tissues [fig_ref] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways... [/fig_ref]. Having confirmed that BRK overexpression is prevalent in cancers, we next sought to identify BRK targets.
In this study, we focused on the constitutively active form of BRK, BRK-Y447F (termed BRK-YF from here on). We have previously demonstrated that BRK-YF displayed higher kinase activity than BRK-wild-type (WT) when ectopically and stably expressed in human embryonic kidney (HEK) 293 cells [bib_ref] Constitutive activation of breast tumor kinase accelerates cell migration and tumor growth..., Miah [/bib_ref]. To decipher the role of activated BRK in cellular signal transduction pathways, we expressed SNAP-FLAG tagged BRK-WT (SF-BRK-WT) and BRK-YF (SF-BRK-YF) constructs in HEK293 cells and evaluated their global kinase activity by analyzing cell lysates by Western blotting. When we visualized phosphorylated proteins using the PY20 antiphosphotyrosine antibody, we confirmed that BRK-YF showed higher kinase activity than BRK-WT [fig_ref] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways... [/fig_ref]. Next, we used a kinome peptide array to identify potential signaling pathways regulated by activated BRK in BRK-YF-expressing stable cells. This well-characterized kinome array (11) consisted of 300 distinct target peptides corresponding to different signaling molecules involved in various signal transduction pathways, including the TGF-/ SMAD, mammalian target of rapamycin (mTOR) , phosphatidylinositol 3-kinase, integrin, Janus kinase-signal transducers and activators of transcription, and mitogen-activated protein kinase (MAPK) pathways. We analyzed lysates from stably expressing green fluorescent protein (GFP)-BRK-YF and parental MCF10A and HEK293 cells using this kinome platform [fig_ref] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways... [/fig_ref]. We observed that potential BRK targets were enriched for components of several signaling pathways, notably the TGF-/SMAD signaling pathway (P ≤ 0.05; [fig_ref] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways... [/fig_ref].
## Smad4 is a cytosolic target of brk
As our kinome array data suggested that SMAD family proteins were potential targets for BRK-mediated phosphorylation, we next asked whether SMAD2/3/4 interacted with BRK. First, we expressed Halo-SMAD2/3/4 either alone or with SF-BRK-YF into HEK293 cells, followed by affinity purification using magnetic beads against Halo and SNAP. We found that SF-BRK-YF copurified with either SMAD2, SMAD3, or SMAD4 [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref]. We also observed a reciprocal association when SF-BRK-YF was coexpressed with either Halo-SMAD2/3/4 or affinity purified with Halo magnetic beads [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref]. Since all three of the SMAD proteins (Halo-SMAD2/3/4) interacted with BRK-YF, we next determined which of them, if any, had the strongest interaction with BRK. We coexpressed Halo-SMAD2/3/4, together with SF-BRK-YF in HEK293 cells, and affinity purified proteins from the resulting whole-cell lysates with Halo magnetic beads. We then analyzed these proteins by immunoblotting with specific antibodies against SMAD2, SMAD3, and SMAD4. We detected SMAD4, but neither SMAD2 nor SMAD3 in the SF-BRK purified sample, suggesting that in the presence of all three SMAD proteins, SMAD4 competitively binds SF-BRK-YF, possibly indicating a stronger affinity of SMAD4 toward SF-BRK-YF [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref].
Next, to map the domains of BRK important for interaction with SMAD4, we ectopically expressed five BRK mutants with GFP-SMAD4 in HEK293 cells. These included three mutants affecting the Src homology domains: BRK-W44A; SH2-BRK, which lacks the SH2 domain; and SH3-BRK, which lacks the SH3 domain. In addition, we used two mutants to assess whether BRK activity was necessary for interaction: the kinase-inactive BRK-Y342A and the constitutively active BRK-YF. We observed that BRK-W44A, SH2-BRK, BRK-WT, BRK-Y342A, and BRK-YF all coprecipitated with GFP-SMAD4. In contrast, SH3-BRK did not coprecipitate with GFP-SMAD4, suggesting that the SH3 domain is necessary for BRK/SMAD4 interaction [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref]. In a similar fashion, we mapped domains of SMAD4 essential for interaction with BRK. We expressed BRK-YF in HEK293T cells together with full-length Halo-SMAD4 or with several SMAD4 truncation mutants (described in [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] and captured protein complexes by Halo affinity purification. We found SMAD4 mutants that contained either the MH1 domain or the MH2 domain (mu-tants A, B, D, and E) interacted with BRK-YF. The MH1 domain showed a higher affinity for BRK-YF than the MH2 domain (compare [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] , while the linker region alone showed a very weak affinity for BRK-YF [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] , lane C). Protein interactions were further analyzed by affinity purification followed by mass spectrometry (APMS) of Halo-SMAD4 ectopically expressed by itself or coexpressed with either SF-BRK-WT or SF-BRK-YF in HEK293T cells. Halo affinity purification followed by MudPIT proteomics analyses showed that Halo-SMAD4 associated proteins [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] , A to C, and table S1) and the expression of Halo-SMAD4 in those cells [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] and was able to pull down both BRK-WT and BRK-YF [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] and table S1). The interaction was shown to be reciprocal when SNAP-Flag-BRK-YF copurified with Halo-SMAD4 in the SNAP affinity purification analyzed by MudPIT [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref]. These APMS experiments hence confirmed the conclusion from the coimmunoprecipitation experiments: BRK interacts with SMAD4.
Next, we examined whether Halo-SMAD4 and SF-BRK-YF colocalize in vivo. We used live-cell imaging to assess the localization of ectopically expressed Halo-SMAD4 and SF-BRK-YF in HEK293T cells. Halo-SMAD4 and SF-BRK-YF were transfected in HEK293T cells either alone or together and were imaged by confocal microscopy [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref]. We observed that Halo-SMAD4 predominantly localized to the cytosol [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] , while SNAP-Flag-BRK-YF localized both in the cytosol and nucleus [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] , middle). However, Halo-SMAD4 and SNAP-FLAG-BRK-YF did colocalize in the cytosol [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] , bottom) when cotransfected in HEK293T cells. In addition, immunocytochemistry experiments revealed that the endogenous BRK and SMAD4 colocalized in MCF7 cells [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref]. Together, our observations support that SMAD4 is a cytosolic BRK interaction partner in cells, which is consistent with a potential role of SMAD4 as a target of BRK phosphorylation.
Activated BRK phosphorylates SMAD4 on residues Tyr 353 and Tyr We have shown that activated BRK interacts with SMAD4 and colocalizes with SMAD4 in the cytosol of live cells. Given that BRK is a nonreceptor tyrosine kinase, we asked whether activated BRK phosphorylates SMAD4. To selectively identify BRK-mediated SMAD4 phosphorylation, we expressed Halo-SMAD4 alone or in combination with SF-BRK-YF in HEK293T cells and affinity-purified SMAD4 to analyze posttranslational modifications by mass spectrometry [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. Although prior studies had reported several serine, threonine, and tyrosine phosphorylation sites within SMAD4 [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref] ; source: PhosphoSitePlus), our MudPIT-APMS approach identified several previously unknown phosphorylation sites on SMAD4 in the presence or absence of SNAP-Flag-BRK-YF (table S2). We found that Halo-SMAD4 displayed three unique phosphorylations on serine(S)-344, tyrosine(Y)-353, and tyrosine(Y)-412 in the presence of activated BRK [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. Unexpectedly, one of these phosphorylation sites was S-344, which raises the intriguing possibility that activated BRK may be a dual-specificity kinase, such as MAPK kinase kinases, which are involved in MAP pathways [bib_ref] Signaling by dual specificity kinases, Dhanasekaran [/bib_ref]. However, since BRK is known to potentiate several signaling pathways, including the MAPK signaling pathway, it is possible that the S-344 phosphorylation is an indirect effect. Nonetheless, since BRK is a tyrosine kinase, we focused to further validate the phosphoryl ation of Y353 and Y412 [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. We implemented a multiple-reaction monitoring (MRM) approach to specifically targeting the peptides bearing these phosphorylated and nonphosphorylated tyrosines [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. Halo-SMAD4 or Halo-SMAD4 Y353F and Y412F, with or without SNAP-Flag-BRK-YF were cotransfected into HEK293T cells, and the cell lysates were subjected to affinity purification, followed by immunoblotting (IB) with anti-PY20 or anti-SMAD4 antibodies. The expression of Halo-SMAD4, Halo-SMAD4 Y353F and Y412F, and SNAP-Flag-BRK-YF were analyzed by immunoblotting using anti-Halo and anti-Flag specific antibodies. -Tubulin was used as a loading control. (E) An in vitro kinase assay was performed using the active kinase, Flag-BRK, and the substrate SMAD4, in the presence or absence of adenosine 5′-triphosphate (ATP). Antiphosphotyrosine antibody PY20 was used to detect phosphotyrosine. The blots were reprobed with anti-Flag and anti-SMAD4 antibody (bottom). phosphorylated and nonphosphorylated sites, the transitions of at least four fragment ions bearing the modified/nonmodified residues were targeted for MRM [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref] , left modified and right nonmodified) and were detected in the protein sample from the affinity-purified Halo-SMAD4 or Halo-SMAD4 cotransfected with BRK-YF [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. In addition, the tandem mass spectrometry spectra that were acquired immediately after the MRM spectra mapped to the expected phosphorylated peptides [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref].
To further validate BRK-mediated phosphorylation of SMAD4 Y353 and Y412, we next generated a mutant of SMAD4 lacking these two tyrosine residues (Halo-SMAD4 Y353F and Y412F). We first affinity-purified Halo-SMAD4 from cells coexpressing BRK-YF with Halo-SMAD4 and analyzed the pulled-down proteins with an antibody specific to phosphorylated tyrosines, PY20 [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. We detected phosphorylated SMAD4 in the presence of BRK-YF but not in the absence of BRK-YF [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. However, in the mutant Halo-SMAD4 Y353F and Y412F, we could not detect a similar band indicating the phosphorylation [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref] , lane 5). In light of these findings, we tested whether SMAD4 is a direct substrate of BRK. In an in vitro kinase assay, which was accomplished using BRK-YF and the SMAD4, we observed phosphorylation of SMAD4 in the presence of BRK-YF, confirming that SMAD is a direct substrate of BRK [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref] , lane 6, top). We also observed the kinase activity of autophosphorylated BRK-YF [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref] , lanes 3 and 5, top). Our biochemical and mass spectrometry analysis of phosphorylated peptides from SMAD4 indicates that two previously unidentified tyrosine residues were phosphorylated explicitly by BRK. These sites were confirmed as phosphorylated with an orthogonal MRM mass spectrometry approach and by site-directed mutagenesis combined with Western blotting. These two tyrosines are located within SMAD4 MH2 domain, which is the most frequently mutated in cancers, and are likely involved in regulating the SMAD4 protein interaction network.
## Brk and smad4 protein expression levels are inversely correlated in most breast cancer cells and tissues
Since SMAD4 is a tumor suppressor [bib_ref] TGF- tumor suppression through a lethal EMT, David [/bib_ref] [bib_ref] Mechanistic insight into contextual TGF- signaling, Zhang [/bib_ref] and we and others have previously shown that BRK acts as an oncogene (2, 10, 6), we opted to explore the possible connection between SMAD4 and BRK protein levels in breast cancer cells and tissues. First, we examined the endogenous protein expression of SMAD4 and BRK to determine the expression profiles of these proteins in breast cancer cells. Using antibodies against each protein, we evaluated the expression level of SMAD4 and BRK in a panel of 10 breast cancer cells, two immortalized cell lines commonly used to model nondiseased human mammary epithelial cells (MCF10A and MCF12F), and lastly in HEK293 cells. We detected SMAD4 in MCF10A, MCF12F, and HEK293 and in six breast cancer cell lines MDA231, Hs578T, T47D, BT549, MCF7, and BT474. SMAD4 expression was very low or undetectable in four other breast cancer cell lines: BT20, HCC1428, SKBR3, and HCC1954 . Detectable amounts of BRK were observed in BT20, HCC1428, SKBR3, T47D, MCF7, BT474, and HCC1954, but not in MDA231, Hs578T, BT549, and HEK293 cells, while very low expression was observed in MCF10A and MCF12F . The levels of SMAD4 and BRK were inversely correlated in MCF10A, MCF12F, MDA-MB-231, BT20, HCC1428, SKBR3, Hs578T, BT549, HCC1954, and HEK293. In particular, the expression of SMAD4 and BRK in all of the triple-negative breast cancer (TNBC) cells we tested (MDA231, BT20, Hs578T, BT549, and HCC1954) showed this inverse expression pattern .
These diverse patterns of BRK/SMAD4 protein levels might be explained by differences in BRK or SMAD4 mRNA levels. To compare the gene expression pattern of SMAD4 and BRK, we examined RNA sequencing data from the Cancer Cell Line Encyclopedia (https://portals.broadinstitute.org/ccle) corresponding to the 10 breast cancer cell lines that we had analyzed. Unexpectedly, we observed that SMAD4 and BRK showed similar patterns of mRNA expression in all 10 breast cancer lines including TNBC cells , suggesting that the effects we had initially noticed were likely regulated at the protein level rather than mRNA expression. Next, to reinforce the evidence obtained using cancer cell lines, we mined data from patient tumor samples [bib_ref] Proteomic maps of breast cancer subtypes, Tyanova [/bib_ref] to determine SMAD4 and BRK protein expression patterns in three major breast cancer subtypes: estrogen/progesterone (ER/PR) positive, HER2 positive, and TNBC. Again, we observed that SMAD4 and BRK were inversely expressed in these breast cancer subtypes .
To further interrogate the relationship between SMAD4 and BRK expression, we stably expressed constitutively active BRK (BRK-YF) into three cell lines that expressed SMAD4 but not BRK (HEK293, MDA-MB-231, and MCF10A cell lines). An elevated level of phosphorylation of cellular targets was observed in the cells stably expressing GFP-BRK-YF, as visualized by immunoblotting with an antiphosphotyrosine antibody (4G10) . We therefore examined the expression of SMAD4 in BRK-YF-expressing cells and observed a sharp reduction of endogenous SMAD4 protein in all three of the cell lines expressing activated BRK compared to those parental cells . However, we did not observe any noticeable change in the endogenous protein levels of SMAD2 and SMAD3 , suggesting that the effect that activated BRK had on protein levels was specific to SMAD4. Next, we examined the expression of SMAD4 transcripts in SF-BRK-YF-transduced HEK293 cells by quantitative reverse transcription polymerase chain reaction (qRT-PCR) using SMAD4-specific primers. The qRT-PCR analysis showed no significant difference between the SMAD4 mRNA levels in SF-BRK-YF-expressing cells compared with the control cells . In addition, to interrogate whether SF-BRK-YF regulates SMAD4 via the ubiquitin-proteasome pathway, we tested SMAD4 protein levels in SF-BRK-YF-expressing cells in the presence or absence of the peptide-aldehyde proteasome inhibitor MG132 (carbobenzoxyll-leucyl-l-leucyl-l-leucine). We found that MG132 treatment rescued the SMAD4 protein levels in SF-BRK-YF-expressing HEK293 cells .
Last, since our data showed that SMAD4 and BRK protein levels were inversely correlated, we asked whether knocking down BRK by short hairpin RNA (shRNA) could modulate the levels of SMAD4 in MCF7 cells. MCF7 cells were selected because both SMAD4 and BRK proteins were moderately expressed in these cells . We attained a 70 to 80% reduction of BRK in MCF7 cells . However, we did not observe any noticeable effect on the expression of SMAD4 protein in MCF7 cells with depleted BRK . Since the depletion of BRK did not affect the expression of SMAD4, we aimed to assess the expression of a known target of SMAD4 in GFP-BRK-YF cells. As a proof of principle, we evaluated the expression of p21, a known target of SMAD4 [bib_ref] Regulation of the human p21/WAF1/Cip1 promoter in hepatic cells by functional interactions..., Moustakas [/bib_ref] , in the cells stably expressing GFP-BRK-YF and the parental cell lines. We found that p21 protein levels and mRNA level sharply decreased in the cells expressing activated BRK compared to parental cells . Moreover, reduction of p21 induced by constitutively active BRK-YF could be rescued by ectopically overexpressed GFP-SMAD4 , suggesting that large amounts of the GFP-SMAD4 compensated for the degradation of endogenous SMAD4 mediated by BRK-YF phosphorylation, hence restoring the downstream expression of the p21 protein.
Overall, our data indicate that, compared with control cells, the increased levels of BRK observed in several breast cancer cell types were concomitant with reduced levels of SMAD4 and that changes in protein level did not necessarily result from changes in BRK/SMAD4 mRNA levels. In addition, introducing the active BRK-YF into cells expressing SMAD4 resulted in reduced levels of SMAD4. As BRK and SMAD4 showed similar mRNA expression patterns but differ in protein levels and the reduced SMAD4 protein can be rescued by MG132 treatment, it is possible that BRK suppresses SMAD4 through the ubiquitin/proteasome pathway.
## Smad4 is a target of ubiquitin modification enzymes in the presence of activated brk
Since we had found that BRK phosphorylates SMAD4 [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref] and SMAD4 levels were markedly lower in stably expressing BRK-YF cells , we next examined whether the presence of BRK-WT/BRK-YF made Halo-SMAD4 a potential target of the ubiquitin-proteasome machinery. To interrogate the impact of constitutively active BRK on SMAD4 ubiquitination, we used our established workflowfor Halo/MudPIT APMS analysis . We first expressed Halo-SMAD4 in the presence or absence of SF-BRK-WT or SF-BRK-YF and purified Halo-SMAD4-associated proteins by affinity chromatography. We identified SMAD4-associated proteins by MudPIT mass spectrometry [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] , A to C). We found that Halo-SMAD4 recruited several ubiquitin and deubiquitin ligases in the presence of BRK-WT/BRK-YF and table S1). SMAD4 interacted with the deubiquitin ligases ubiquitin specific peptidase 9 X-linked (USP9X), ubiquitin specific peptidase 32 (USP32), and ubiquitin specific peptidase 7 (USP7) in the presence of both BRK-WT and BRK-YF. However, SMAD4 association with the ubiquitin ligases HECT and RLD domain containing E3 ubiquitin protein ligase 2 (HERC2) and ring finger protein 138 (RNF138) were up-regulated only in the presence of BRK-YF. Since Halo-SMAD4 interacted with ubiquitin ligases in the presence of activated BRK, we then tested the possibility that the BRK-mediated phosphorylation of SMAD4 facilitated its degradation through the ubiquitin/proteasome system [bib_ref] The logic of the 26S proteasome, Collins [/bib_ref]. Halo-SMAD4, BRK-YF, and hemagglutinin (HA)-ubiquitin were expressed in HEK293T cells in the combinations shown in , and cells were treated with or without MG132. We examined SMAD4 protein for ubiquitination both under normal and denaturing conditions by resolving SMAD4 immunoprecipitated samples by SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting using anti-HA antibody. Our data showed a smear of ubiquitinconjugated SMAD4 in the presence of proteasome inhibitor , which was not apparent in controls [fig_ref] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways... [/fig_ref]. Our results led us to conclude that the presence of activated BRK causes the down-regulation of SMAD4 by the ubiquitin/proteasome degradation pathway.
To investigate the rate of proteasomal degradation of SMAD4 in the presence or absence of BRK-YF, we expressed Halo-SMAD4 in HEK293T cells under the control of a tetracycline-inducible promoter, with or without BRK-YF. Twenty-four hours after transfection, tetracycline-containing medium was replaced with regular HEK293T cell culture media, and cells were periodically harvested as indicated in . Immunoblotting of cell lysates showed that Halo-SMAD4 protein levels were markedly reduced in the presence of activated BRK as early as 24 hours , while in the absence of BRK-YF, Halo-SMAD4 protein levels had not significantly decreased even after 72 hours . This difference in protein stability is hence consistent with a role for BRK in regulating SMAD4 degradation. In summary, our data indicate that ubiquitin ligases are recruited to phosphorylated SMAD4 in the presence of BRK-YF, leading to its degradation by the proteasome.
## Brk-mediated phosphorylation of tyrosine-353 and tyrosine-412 of smad4 is required for ubiquitination and degradation of smad4
To gain further mechanistic insight, we again used Halo-MudPIT APMS to explore the BRK-YF-modulated SMAD4 ubiquitination and degradation using a mutant of SMAD4 (Halo-SMAD4 Y353F and Y412F) that could not be phosphorylated on tyrosine residues Y353 and Y412 . As we have described previously with Halo-SMAD4, we analyzed the Halo-SMAD4 Y353F&Y412F-associated proteins with or without BRK-YF and established that baits were present in all affinity purifications , B and C, and table S1). We found that in the presence of the BRK-YF, RNF138 (a ubiquitin ligase) interacted with Halo-SMAD4 but not with SMAD4 Y353F and Y412F, suggesting that the presence of the BRK-YF is necessary for this interaction. Similarly, SMAD4 interaction with the ubiquitin ligase HERC2 was significantly augmented in the presence of activated BRK, and this interaction was completely lost with mutant Halo-SMAD4 Y353F and Y412F [fig_ref] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways... [/fig_ref]. Curiously, itchy E3 ubiquitin protein ligase (ITCH) showed a stronger affinity for Halo-SMAD4 Y353F and Y412F in comparison to Halo-SMAD4 in the presence of activated BRK and did not interact in the absence of BRK. Considering the deubiquitinases, we noticed that USP9X and USP7 showed a higher affinity for Halo-SMAD4 Y353F and Y412F in the presence of activated BRK and that the deubiquitinase USP32 interacted only with Halo-SMAD4 in the presence of activated BRK . Similar to BRK, USP32 is also often overexpressed in breast cancer cells and tumors, and its inhibition reduces cell proliferation, migration, and apoptosis [bib_ref] USP32 is an active, membrane-bound ubiquitin protease overexpressed in breast cancers, Akhavantabasi [/bib_ref].
Since our proteomics data showed that phosphorylation of Y353 and Y412 of SMAD4 is essential for ubiquitin ligase recognition, we tested whether the mutant SMAD4 escaped BRK-regulated proteasomal degradation. Plasmids expressing Halo-SMAD4 Y353F and Y412F, SNAP-Flag-BRK-YF, and HA-ubiquitin were transfected into HEK293T cells. The transfected cells were treated with MG132 for 8 hours to inhibit 26S proteasome. After Halo affinity purification, we analyzed purified samples for ubiquitinated proteins using an anti-HA antibody. Our data showed that ubiquitin was unable to conjugate with mutant Halo-SMAD4 Y353F and Y412F to mark it for proteasomal degradation . Our findings strongly suggest that tyrosine-353 and tyrosine-412 of SMAD4 are essential for SMAD4 ubiquitination.
Since the Halo-SMAD4 Y353F and Y412F was not ubiquitinated, we next examined the stability of SMAD4 Y353F and Y412F in the presence or absence of BRK-YF. To this end, we transfected tetracyclineinducible Halo-SMAD4 Y353F and Y412F alone or with BRK-YF into HEK293T cells. We found that Halo-SMAD4 Y353F and Y412F protein levels remained stable for a longer time in the absence of activated BRK . We also found that the mutant SMAD4 was more stable in the presence of BRK-YF than the WT SMAD4 protein (compare to . These findings are consistent with a model whereby BRK-mediated phosphorylation of SMAD4 accelerates its proteasomal degradation, while the phosphorylationincompetent mutant SMAD4 Y353F and Y412F escapes ubiquitination and subsequent degradation.
## Brk represses the frk tumor suppressor in a smad4dependent manner to induce emt and cell invasion
Since SMAD4 is a transcription factor and phosphorylated SMAD4 interacts with chromatin remodeling complexes [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref] , and table S3), we next investigated genes that might be targeted by SMAD4. We performed an in silico analysis to identify potential SMAD4 binding sites in the genome. We found putative SMAD4binding sites in the promoter of several genes, including tumor suppressor FRK [bib_ref] Understanding the cellular roles of Fyn-related kinase (FRK): Implications in cancer biology, Goel [/bib_ref]. Chromatin immunoprecipitation (ChIP)-sequencing data (Gene Expression Omnibus: GSE96401; www.encodeproject. org/experiments/ENCSR826YMT/) also showed that SMAD4 binds the transcription start sites of FRK in HepG2 cells . This finding spurred our interest to further investigate a potential connection between BRK-mediated SMAD4 regulations of FRK expression. To test our hypothesis that SMAD4 regulates this promoter, we performed a luciferase reporter assay and found that luciferase activity was twofold higher in lysates from cells coexpressing SMAD4 and the FRK promoter, suggesting that SMAD4 positively regulates the promoter activity of FRK [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref] , consistent with the presence of SMAD4 binding sites in the FRK promoter. In addition, we performed ChIP-qPCR experiment and found SMAD4 binds to the transcription start sites of FRK in MCF7 cells [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref]. Next, we compared the mRNA expression of FRK in MDA-MB-231 cells (which express SMAD4, but not BRK-see with FRK expression in MDA-MB-231 cells stably expressing BRK-YF. Consistent with BRK-YF-mediated degradation of SMAD4, FRK mRNA levels were very low in the cells stably expressing BRK-YF. However, overexpressing SMAD4 in the MDA231-BRK-YF cells restored the expression of FRK mRNA [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref]. In addition, we observed that stably expressing BRK-YF decreased the FRK protein levels in MDA231 cells [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref] , compare lanes 1 and 2) and that FRK protein levels were restored by overexpressing SMAD4 [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref].
It has been reported that FRK suppresses EMT and inhibits cancer metastasis [bib_ref] FRK inhibits breast cancer cell migration and invasion by suppressing epithelialmesenchymal transition, Ogunbolude [/bib_ref] , while TGF-/SMAD4 signaling is crucial for EMT and promotes metastasis [bib_ref] Smad transcription factors, Massagué [/bib_ref]. We deduced that the SMAD4-dependent control of FRK expression might consequently be controlling the expression of EMT markers. To this end, we examined the expression of EMT markers (E-cadherin, N-cadherin Twist, vimentin, fibronectin, SLUG, and SNAIL) in parental and BRK-YF-expressing MDA-MB-231 cells. We found that SLUG and SNAIL expression increased in MDA-MB-231 cells stably expressing BRK-YF in comparison to parental cells (six-and fourfolds higher, respectively; [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref] , E and F). In addition, SLUG but not SNAIL was significantly suppressed in stably FRK-YF-expressing MDA-MB-231 cells . Moreover, ectopically expressed SMAD4 suppressed the BRK-YFmediated induction of SNAIL and SLUG [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref]. We did not observe any change in the other EMT markers tested (data not shown). Last, since our proteomics data showed that the presence of BRK-YF reduced the interaction of SMAD4 with cell adhesion molecules (table S1), we examined the cell adhesion properties of the BRK-YF-expressing HEK293 and MCF10A cells. We found that activated BRK reduced the cell adhesion capability of BRK-YF-expressing cells and that adhesion capability could be completely restored by overexpressing SMAD4 [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref]. Overall, our data suggest that activated BRK induces a SMAD4-dependent suppression of tumor suppressor FRK potentially resulting in the stimulation of EMT and, potentially, metastasis [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref].
# Discussion
The cellular role of TGF-/SMAD signaling pathway is a paradox in cancer. On one hand, SMAD4-deficient Kras G12D pancreatic mouse models showed the rapid development of pancreatic tumors [bib_ref] Smad4 is dispensable for normal pancreas development yet critical in progression and..., Bardeesy [/bib_ref] , while restoration of SMAD4 induced apoptosis [bib_ref] Smad4 is dispensable for normal pancreas development yet critical in progression and..., Bardeesy [/bib_ref] and inhibited tumorigenesis in Smad4-defective cancer cells [bib_ref] The role of TGF-/SMAD4 signaling in cancer, Zhao [/bib_ref]. On the other hand, knockdown of SMAD4 significantly reduced liver tumorigenesis in mice [bib_ref] Transforming growth factor- promotes liver tumorigenesis in mice via up-regulation of snail, Moon [/bib_ref]. The molecular mechanism of this duality is yet to be solved. Further, it was thought that ubiquitination, but not phosphorylation, may play a role in the regulation of SMAD4 function [bib_ref] Mutations at a single codon in Mad homology 2 domain of SMAD4..., Goff [/bib_ref]. Our current study demonstrates that activated BRK phosphoryl ates SMAD4 and regulates its stability.
In this study, we have shown that (i) activated BRK regulates TGF-/SMAD signaling by interacting with SMAD2/3 and SMAD4; however, SMAD4 is the preferred target of BRK, as shown by the results of competitive binding [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref] ; (ii) activated BRK phosphorylates SMAD4 on tyrosine-353 and tyrosine-412 [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref] ; (iii) phosphorylated SMAD4 is the target of ubiquitin/deubiquitin ligases and is degraded by the ubiquitin-proteasome pathway ; (iv) the phosphorylation-regulated degradation of SMAD4 correlates well with the observation that BRK and SMAD4 show inverse expression patterns at the protein levels in breast cancer cells and tumors ; and (v) activated BRK modulates SMAD4 to suppress the tumor suppressor FRK, reduces the interaction of SMAD4 with cell adhesion molecules, and induces EMT [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref]. Our study provides experimental evidence that the BRK kinase degrades SMAD4 to suppress tumor suppressor FRK and up-regulate EMT markers SNAIL and SLUG.
BRK is expressed in most of the cancer types. Although the expression of BRK is ubiquitous in most breast cancer cells and tumors, activated BRK was only detected in the plasma of breast tumors [bib_ref] PTK6/BRK is expressed in the normal mammary gland and activated at the..., Peng [/bib_ref]. Thus, in this study, we focused on characterizing activated BRK and its role in signal transduction pathways. A kinome array (11) was used to uncover the signal transduction pathways regulated by activated BRK in cancer and normal cells. Ectopically expressed activated BRK regulates the TGF-/SMAD signaling pathways by interacting with SMAD2/3 and SMAD4. However, SMAD4 outcompetes both SMAD2 and SMAD3 in a binding assay for activated BRK. The BRK modular SH3 domain mediates its protein-protein interactions to regulate signaling [bib_ref] The discovery of modular binding domains: Building blocks of cell signalling, Mayer [/bib_ref]. Notably, the SMAD4 activation domain is proline rich and contains the Pro-X-X-Pro motif [bib_ref] The tumor suppressor Smad4/DPC4 and transcriptional adaptor CBP/p300 are coactivators for Smad3..., Feng [/bib_ref] , which is the recognition site for the SH3 domain [bib_ref] Identification of a ten-amino acid proline-rich SH3 binding site, Ren [/bib_ref]. However, our domain truncation data indicate that activated BRK interacts with an MH1 domain of SMAD4, which does not contain the SH3 recognition motif, suggesting that the Pro-X-X-Pro motif might be dispensable for SH3-mediated protein-protein interactions.
Dupont et al. [bib_ref] FAM/USP9x, a deubiquitinating enzyme essential for TGF signaling, controls Smad4 monoubiquitination, Dupont [/bib_ref] previously found that a cycle of ubiquitination and deubiquitination regulates the function and protein-protein interaction of SMAD4: Ecto/TIF1-mediated monoubiquitination disassembles SMAD4 from the SMAD complex, while deubiquitination by FAM/USP9x allows SMAD4 to return to SMAD signaling pool. Our MudPIT-proteomics data revealed that BRK-phosphorylated SMAD4 interacts with several ubiquitin and deubiquitin ligases such as HERC2, RNF138 (ubiquitin ligases), and USP32, USP7, and USP9X (deubiquitinases). When phosphorylated by activated BRK, SMAD4 becomes a target of ubiquitin ligases (HERC2 and RNF138), which accelerates its degradation by the ubiquitin-proteasome pathway. Recently, SMAD4 has been shown to phosphorylate and degraded (on Thr 277 ) by glycogen synthase kinase 3 [bib_ref] The tumor suppressor Smad4/DPC4 is regulated by phosphorylations that integrate FGF, Wnt,..., Demagny [/bib_ref]. However, phosphorylated SMAD4 also becomes a target of deubiquitin ligases (USP32, USP7, and USP9X). Deubiquitin ligase USP9X shows a stronger affinity for tyrosine phosphorylated SMAD4 than nonphosphorylated SMAD4. In addition, our evidence reveals that BRK or other protein tyrosine kinases [since protein tyrosine kinase are functionally redundant [bib_ref] Non-receptor protein tyrosine kinases, Tsygankov [/bib_ref] ] mediates phosphorylation of SMAD4, which is required for SMAD complexes to interact with chromatin remodelers such as SWI/SNF ((SWItch/Sucrose Non-Fermentable)), mediator, histone acetyltransferesase, or SIN3/histone deacetylase complexes for gene regulation [bib_ref] The role of TGF-/SMAD4 signaling in cancer, Zhao [/bib_ref] [bib_ref] TGF signalling in context, Massagué [/bib_ref]. Notably, in our future study, we opt to explore the impact of SMAD4-chromatin remodeler complexes interaction on gene expression in normal and cancer cells.
Complete loss or mutation in SMAD4 has been reported in several cancer types, including pancreatic, cholangiocarcinoma, and colorectal cancer [bib_ref] The role of TGF-/SMAD4 signaling in cancer, Zhao [/bib_ref]. In addition, SMAD4 protein levels decline concurrently with the cumulative malignancy of the tumor cells [bib_ref] Smad4-expression is decreased in breast cancer tissues: A retrospective study, Stuelten [/bib_ref]. Our data demonstrate differential expression of SMAD4 and BRK in a panel of breast cancer cell lines. An inverse pattern of SMAD4 and BRK protein levels was noticed in HER2 + (HCC1428 and SKBR3) and TNBC (MDA-MB-231, BT20, Hs578T, BT549, and HCC1954) cells but not ER + cells (T47D, MCF7, and BT474). However, there are no discrepancies in the expression of SMAD4 and BRK at the mRNA level in those cell lines, indicating a posttranslational mechanism of regulation of SMAD4. We also found that the inversely correlated pattern of expression between SMAD4 and BRK in patients' breast tumor tissues. In agreement with what is observed in TNBC cells, patients' breast tumor samples also show higher levels of the BRK protein in comparison to SMAD4. Furthermore, cells stably expressing activated BRK show a marked suppression of SMAD4 protein levels. We confirmed that p21, a known downstream target of SMAD4 [bib_ref] The role of TGF-/SMAD4 signaling in cancer, Zhao [/bib_ref] [bib_ref] TGF signalling in context, Massagué [/bib_ref] , is suppressed in these cell lines expressing activated BRK. However, SMAD4 was not restored or up-regulated in MCF7 cells where BRK was knocked down. This suggests that SMAD4 escapes being targeted for proteasomal degradation, which could be due to a mutation or posttranslational modification of SMAD4 or inactivation of BRK in MCF7 cells.
The growth inhibitory signals of TGF-/Smad4 signaling in early stages of carcinogenesis are well established. In particular, the involvement of SMAD4 in the EMT process and in cancer progression in later stages of carcinogenesis is largely unclear [bib_ref] Cytostatic and apoptotic actions of TGF- in homeostasis and cancer, Siegel [/bib_ref]. Moreover, the function of SMAD4 is mostly contextual. Our data indicate that SMAD4 binds in the promoter region and promote FRK expression. However, in the presence of activated BRK, FRK expression is repressed in TNBC cells. Overexpression of SMAD4 restores the BRKinduced suppression of FRK level in the TNBC cells stably expressing BRK. Previous studies have shown that the expression of SNAIL is up-regulated in FRK-knocked down breast cancer cells [bib_ref] FRK inhibits breast cancer cell migration and invasion by suppressing epithelialmesenchymal transition, Ogunbolude [/bib_ref]. We observed that activated BRK induces SLUG and SNAIL expression, while overexpression of FRK significantly suppresses the mesenchymal marker SLUG, suggesting an FRK-dependent mechanism for BRK-induced promotion of EMT.
In summary, we provide additional evidence to counter the longstanding idea that SMAD4 is not regulated by phosphorylation. We have found that activated BRK competitively binds and phosphorylates SMAD4 and regulates TGF-/SMAD signaling pathways. Phosphorylated SMAD4 becomes a target of ubiquitin ligases subsequently degraded through the ubiquitin-proteasome system leading the suppression of tumor suppressor FRK. Activated BRK also reduces cell adhesion ability and induces EMT in a SMAD4-dependent manner. Thus, our data suggest that combination therapies targeting activated BRK signaling may be beneficial in the treatment of SMAD4repressed cancers.
# Material and methods
## Antibodies and reagents
The following antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA): anti-BRK (sc-916 and sc-1188), antiphosphotyrosine PY20 (sc-508), anti-SMAD2/3/4, antitubulin (sc-9104), anti-GFP (sc-8334), and anti--actin (sc-130300). Anti--tubulin mouse monoclonal (T9026) antibody was purchased from Sigma-Aldrich (St. Louis, MO). Anti-Halo rabbit polyclonal antibody (G9281) and Magne HaloTag magnetic affinity beads were purchased from Promega (Madison, WI). Proteasome inhibitor MG132 was obtained from Sigma-Aldrich (St. Louis, MO).
## Construction of expression plasmids in human cells
The pGFP-C1-Smad2, pGFP-C1-Smad3, and pGFP-C1-Smad4 plasmids were a gift from C. Hill, Cancer Research UK. SMAD4 and BRK-YF were subcloned by inserting PCR products (containing SgfI and PmeI restriction sites) to generate pcDNA5-Halo-SMAD4 and pcDNA5-Halo-BRK-YF. Vectors expressing Halo-pcDNA5-Halo-SMAD4_1-140, pcDNA5-Halo-SMAD4_1-320, pcDNA5-Halo-SMAD4_140-320, pcDNA5-Halo-SMAD4_320-552, and pcDNA5-Halo-SMAD4_140-552 were constructed by inserting PCR products between the SgfI and PmeI restriction sites. We further subcloned SMAD4 into pcDNA5/FRT/TO vector (a gift from the Conaway Lab at Stowers Institute) using Gibson Assembly Cloning Kit [New England Biotechnologies (NEB)]. Human SMAD4 double mutant (Tyr 353 Phe and Tyr 412 Phe) was obtained using PCR and Gibson Assembly Cloning Kit (NEB). All plasmid constructs were confirmed by sequencing. The primers were used for cloning, and PCR were listed in table S3.
## Generation of stable cell lines
The construction of cell lines stably expressing GFP-BRK-YF has been previously described [bib_ref] Constitutive activation of breast tumor kinase accelerates cell migration and tumor growth..., Miah [/bib_ref]. Amphotropic HEK293-derived Phoenix packaging cells were used to package pBabe-puro retroviral system. For retrovirus production, packaging cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% bovine calf serum. Transfection with 1% PEI (Polysciences Inc.) was conducted with 10 g of retroviral DNA in 60 l of 1% polyethylenimine plus 430 l of 0.15 M NaCl for the 100-mm culture plates. After 24 and 48 hours, the virus-containing supernatant was collected and filtered through 0.45-m syringe filter and aliquoted and stored at −80°C. To infect MCF10A, MDA-MB-231, and HEK293 cells, virus-containing supernatant was supplemented with bovine calf serum and polybrene (Sigma-Aldrich, St. Louis, MO) and overlaid on the cells. After overnight incubation, the viral supernatant was replaced with fresh culture medium. Pools of GFP-BRK-YF-expressing cells were selected with puromycin (Sigma-Aldrich). Expression of GFP-tagged BRK-YF was detected after 48 to 72 hours of infection by fluorescence microscopy. To produce stable BRK knockdown cell lines, we used BRK-expressing MCF7 parental cell lines. This knockdown experiment was performed according to the manufacturer's protocol using shRNA lentiviral vector plasmids from Santa Cruz Biotechnology. The shRNA plasmids generally consisted of a pool of three to five lentiviral vector plasmids, each encoding target-specific 19-to 25-nt shRNAs designed to knockdown gene expression. As controls, MCF7 cells were infected with a control shRNA and a GFP-control plasmid for transfection efficiency. Transfected cells were selected using puromycin (Sigma-Aldrich).
## Kinome array
High-throughput kinome assay was performed according to the published protocol [bib_ref] Genome to kinome: Species-specific peptide arrays for kinome analysis, Jalal [/bib_ref]. Briefly, the MDA-MB-231, MCF10A, and HEK293 cells stably expressing GFP-BRK-YF were cultured to ~80% confluency in 10-cm culture plates. The cells were harvested and lysed with 100 l of lysis buffer [20 mM tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM Na 3 VO 4 , 1 mM NaF, aprotinin (1 g/ml), leupeptin (1 g/ml), and 1 mM phenylmethylsulphonyl fluoride (PMSF)] and incubated on ice for 10 min, followed by centrifugation at maximum speed in a microcentrifuge for 10 min at 4°C. A 70-l aliquot of clear cells lysate was mixed with 10 l of activation mix [50% glycerol, 50 M adenosine 5′-triphosphate (ATP), 60 mM MgCl 2 , 0.05% (v/v) Brij-35, bovine serum albumin (BSA; 0.25 mg/ml)] and incubated on the peptide array in a humidity chamber for 2 hours at 37°C. Arrays were then washed with phosphate-buffered saline (PBS) containing 1% Triton X-100. Slides were submerged in phosphospecific fluorescent ProQ Diamond Phosphoprotein Stain (Invitrogen) with agitation for 1 hour. Arrays were then washed three times in destain containing 20% acetonitrile (EMD Biosciences, VWR distributor, Mississauga, ON) and 50 mM sodium acetate (Sigma-Aldrich) at pH 4.0 for 10 min. A final wash was performed with distilled deionized H 2 O. Arrays were air dried for 20 min and then centrifuged at 3009g for 2 min to remove any remaining moisture from the array. Arrays were analyzed using a GenePix Professional 4200A microarray scanner (MDS Analytical Technologies, Toronto, ON, Canada) at 532 to 560 nm with a 580-nm filter to detect dye fluorescence. Images were collected using the GenePix 6.0 software (MDS) and the spot intensity signal collected as the mean of pixel intensity using local feature background intensity background calculation with the default scanner saturation level. Data were processed using the PIIKA2 platform (http://saphire.usask.ca/saphire/piika/).
## Preparation of cell lysates
Confluent or subconfluent cells were harvested and washed with ice-cold PBS (twice). The whole procedures were carried out at 4°C (on ice) unless specified otherwise. Cells were resuspended in freshly prepared lysis buffer [20 mM tris (pH 7.5), 1% Triton X-100, 150 mM NaCl, and protease inhibitors aprotinin (5 mg/liter) and 0.1 mM PMSF] and kept on ice for 30 min, followed by centrifugation at 14,000 rpm for 15 min at 4°C. Cells were directly lysed in SDS sample buffer [50 mM tris-HCl (pH 6.8), 2% SDS, 0.1% bromophenol blue, and 10% glycerol] to obtain total cell lysates.
Live-cell imaging HEK293T cells were seeded onto glass-bottom culture dishes (MatTek, Ashland, MA) and transiently transfected with the SNAP-Flag-BRK and Halo-SMAD4 constructs. Affinity-tagged proteins were fluorescently labeled during growth, either with Halo-Tag TMRDirect ligand (Promega) or SNAP-Cell 505-Star (NEB) or with both ligands according to the manufacturer's instructions. Images were taken with a ZEISS LSM 780 confocal microscope with argon laser excitation at 573 to 687 nm for TMRDirect and 499 to 526 nm for SNAP-Cell 505-Star. To limit photobleaching, exposure time and laser power were adjusted to enhance image quality. An alternating excitation mode was adopted to eliminate cross-talk between color channels. HaloTag-SMAD4 or SNAP-Tag BRK-YF were ectopically expressed in HEK293T cells and plated at 20% confluency onto glass-bottom MatTek culture dishes (35 mm, no. 2, 14-mm-diameter glass). To label Halo-SMAD4 proteins, the HaloTag TMRDirect ligand was added in a final concentration of 100 nM and incubated the cells overnight. In addition, the SNAP-Cell 505 ligands were added directly to the cells to label SNAP-Tag BRK-YF in a final concentration of 5 M and incubated the cells for 1 hour at 37°C in 5% CO 2 . For colocalization, Halo-SMAD4 and SNAP-Flag-BRK-YF constructs were cotransfected into HEK293T cells, and the cells were labeled as indicated above. The cultured media were replaced with OptiMEM to remove background fluorescence before imaging. Cells were stained with Hoechst dye to mark nuclei for 30 min before imaging.
## Immunocytochemistry of brk and smad4
MCF7 cells were cultured on MatTek plates before fixation with 4% paraformaldehyde (PFA) and permeabilized with permeabilizing buffer (0.5% Triton X-100) for 15 min. The permeabilized cells incubated in Odyssey blocking buffer for 1 hour. Then, cells were washed and incubated with monoclonal rabbit anti-BRK (1:100) and monoclonal mouse anti-SMAD4 (1:100) antibodies at room temperature for 2 hours, followed by incubation with goat anti-rabbit Alexa Fluor 488 and donkey anti-mouse Alexa Fluor 594 conjugated secondary antibody at room temperature for 1 hour. Immunostained cells were analyzed using a ZEISS LSM 700 confocal microscope with argon laser excitation at 488 nm (green), 561 nm (red), and 405 nm (blue). To remove any background, fluorescence cells were washed four times with TBST (tris-buffered saline and Tween 20) before imaging. Cells were stained with Hoechst dye to mark nuclei for 30 min before imaging.
Halo affinity purification of SMAD4 for proteomic analysis HEK293T cells (1 × 10 7 ) were seeded into 15-cm tissue cultures plates for 24 hours, and then, DNA constructs encoding Halo or SNAP tagged genes of interest were transfected using Lipofectamine LTX (Thermo Fisher Scientific). Forty-eight hours after transfection, cells were harvested and washed twice with ice-cold PBS. The ice-cold PBS washed cells were resuspended in 300 l of mammalian cell lysis buffer (Promega) containing 50 mM tris-HCl (pH 7.5), 150 mM NaCl, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1 mM benzamidine HCl, 55 M phenanthroline, 1 mM PMSF, 10 M bestatin, 5 M pepstatin A, and 20 M leupeptin. Next, the cells were raptured by passing through a 26-gauge needle five to seven times, followed by centrifugation at 21,000g for 30 min at 4°C. The resulting 300 l of cell extracts were collected into a new tube and diluted with 700 l of tris-buffered saline [50 mM tris-HCl (pH 7.4), 137 mM NaCl, and 2.7 mM KCl]. To remove insoluble materials, the diluted cell extracts were further centrifuged at 21,000g for 10 min at 4°C. Next, 1000 l of cell extracts were incubated with magnetic beads prepared from 100 l of Magne HaloTag slurry for overnight at 4°C. Beads were washed four times (750 l of buffer per wash) with wash buffer [50 mM tris-HCl (pH 7.4), 137 mM NaCl, 2.7 mM KCl, and 0.05% NP-40] before elution. Proteins were eluted using elution buffer containing 50 mM tris-HCl (pH 8.0), 0.5 mM EDTA, 0.005 mM dithiothreitol (DTT), and 2 U of AcTEV Protease (Thermo Fisher Scientific) for 2 hours at room temperature. The eluate was further passed through a Micro Bio-Spin column (Bio-Rad, Hercules, CA) to remove any residual particles of beads before proteomic analysis.
## Mudpit analysis for identification of smad4-associated proteins and protein complexes
MudPIT analysis for protein complexes identification was previously described in detail by Banks et al.. Briefly, trichloroacetic acid precipitated purified proteins were proteolytically digested with endoproteinase Lys-C, followed by trypsin digestion, overnight at 37°C. A 10-step MudPIT separation approach was applied, and digested peptides were injected directly into a linear ion trap mass spectrometer, where spectra were collected and identified. Peptide mass spectra were analyzed using the ProLuCID and DTASelect algorithms. Next, we used Contrast and NSAF7 software to rank the putative affinity-purified proteins according to their distributed normalized spectral abundance values. We then used QSPEC software to identify enriched proteins in experimental samples compared to control samples. The Benjamini-Hochberg statistical method was used to calculate false discovery rates from QSPEC parameters suitable for multiple comparisons. Each of the experiments was repeated at least twice unless otherwise stated. All raw mass spectrometry runs are available, as described in table S3.
## In vitro kinase assay
In vitro kinase assays were performed using 20 l of affinity-purified Flag-BRK-YF and a 10-l volume of affinity-purified substrate (SMAD4) in a reaction volume of 50 l, comprising 20 l of kinase buffer [25 mM tris-HCl (pH 7.5), 5 mM -glycerophosphate, 2 mM DTT, 0.1 mM Na 3 VO 4 , and 10 mM MgCl 2 ; kinase buffer (10×) #9802, Cell Signaling Technology] with or without 200 M ATP. The reaction was carried out at 30°C for 30 min and terminated by the addition of Laemmli sample buffer. The samples were then boiled for 10 min and resolved via SDS-PAGE.
## Rna isolation and real-time pcr
Total RNA was isolated from MDA-MB-231 cells using RNeasy Plus Mini Kit (QIAGEN, Mississauga ON). Total RNA (1.0 g) was used to synthesize complementary DNA (cDNA) using Bio-Rad Iscript cDNA Synthesis Kit (Bio-Rad, USA). TaqMan probes Hs00176619_m1, Hs00950344-m1, Hs 00195591_m1, and Hs02758991-g1 were used to quantify the expression of FRK, SLUG, SNAIL, and glyceraldehyde-3-phosphate dehydrogenase, as recommended by the manufacturer (Life Technologies, Burlington, ON, Canada). Briefly, 0.6 l of cDNA, 0.5 l of probes for each target and housekeeping genes, and 5 l of TaqMan(R) Master Mix were added in each well. Distilled H 2 O was added in each well to make the volume of 10 l. Probes for target genes and housekeeping genes were labeled with FAM and VIC dyes, respectively. The expression of both genes was measured within the same well using an Applied Biosystems, Step One Plus qRT-PCR machine (Life Technologies, Burlington, ON, Canada). In addition, total RNA was also prepared from the whole cell extract using Direct-zolTM RNA Miniprep plus kits (Zymo Research, Irvine, CA) according to the manufacturer's instructions. Residual DNA was removed using deoxyribonuclease I. cDNA was synthesized using iScript Reverse Transcription Supermix (Bio-Rad), and the resulting cDNA was analyzed by qPCR using a MyiQ real-time detection system (Bio-Rad). The primers were used for PCR are listed in table S3.
Ubiquitination assay under denaturing conditions HEK293 cells were transfected in combination of Halo-SMAD4, HA-tagged ubiquitin, and/or SNAP-Flag-BRK-YF, and the cells were treated with 10 M MG132 for 6 hours. Denatured cell extracts were prepared using denaturing buffer containing 1% SDS, 50 mM tris-HCl (pH 7.5), 150 mM NaCl, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1 mM benzamidine HCl, 55 M phenanthroline, 10 M bestatin, 20 M leupeptin, 5 M pepstatin A, and 1 mM PMSF. Then, the cell lysates were and boiled for 10 min before incubation with primary mouse anti-SMAD4 antibody, followed by Dynabeads Protein G (Invitrogen) magnetic beads conjugation and immunoblotting with anti-HA antibody to detect ubiquitinated SMAD4.
Luciferase reporter assays FRK promoter and SMAD4 plasmids were cotransfected in HEK293 cells using ViaFect (Promega Corporation, Madison, WI) according to the manufacturer's guidelines. Briefly, 125,000 HEK293 cells were seeded with 500 l of fresh media in each well. The HEK293 cells were cotransfected with FRK promoter (495 ng; firefly luciferase) along with phRL-TK (5 ng; Renilla luciferase) as an internal control. Forty eight hours after transfection, cells were harvested, and luciferase activity was measured using the Dual-Luciferase Assay System with the GloMax 96 Microplate Luminometer (Promega Corporation). To examine the impact of SMAD4 on the FRK promoter, cells were cotransfected with GFP-SMAD4 plasmid (250 ng) and FRK reporters (245 ng), while control cells were cotransfected with FRK promoter construct and an empty vector (pCDNA3, 245 ng per well) and 5 ng of phRL-TK plasmids as an internal control in each well (Invitrogen, Canada).
## Chromatin immunoprecipitation
ChIP was performed as described [bib_ref] Chromatin immunoprecipitation and microarraybased analysis of protein location, Lee [/bib_ref]. Briefly, cells were cross-linked with 1% formaldehyde for 20 min before being sonicated using output power 3 (9-W power) for 10 cycles. Before the immunoprecipitation, Dynabeads Protein G (Invitrogen) magnetic beads were washed and coupled to 10 g of anti-SMAD4 (sc-7966, Santa Cruz Biotechnology), and normal mouse immunoglobulin G (IgG) (sc-2025, Santa Cruz Biotechnology). Cell lysates were incubated with SMAD4 and IgGconjugated magnetic beads for overnight at 4°C. Immunoprecipitated lysates were treated with proteinase K and incubated for overnight at 65°C before the DNA being extracted.
## Chip-qpcr
To determine whether SMAD4 binds to FRK promoter regions, we performed anti-SMAD4 and anti-IgG (control) ChIP using chromatin from MCF7 cells. Immunoprecipitation and input samples were analyzed by qPCR using PerfeCTa SYBR Green FastMix Low ROX (Quantabio) in a QantStudio 7 Flex-Thermal Cycler (Life Technologies-Applied Biosystems). PCR primers were used to amplify the regions at the FRK promoter. Transcription start sites and transcriptionally inactive genomic region are listed in table S3. Enrichment as a percentage of the input was calculated for each ChIP sample using the formula: 100 * 2^[C t Input − log 2 (800/30) − C t IP], where 800/30 is the input dilution factor. For each ChIP, SMAD4 enrichment (ChIP/ input) in the promoter was normalized to the average of the ChIP/input values from a transcriptionally inactive genomic region. In this experiment, C t values of two biological replicates of at least three technical replicates were used for the analysis.
## Cell adhesion assay
Ninety-six-well plates were coated with either fibronectin or collagen I for an hour at 37°C, followed by an hour of incubation with 0.5% BSA-containing blocking buffer. HEK293 or MCF10A cells were seeded at a density of 4 × 10 5 , and cells were allowed to attach for 45 min at 37°C. The cells were washed three times with 0.1% BSA-containing culture media, and then, the cells were fixed with 4% PFA. Next, the cells were stained with crystal violet for 10 min. After staining, the cells were washed 10 times with distilled water. Cells were air-dried for an hour and solubilized with 2% SDS on an agitator. The absorbance of each well was measured at 550 nm to quantify the adhesion properties of cells.
# Statistical analysis
For multiple comparisons (qPCR and adhesion assay), one-way analysis of variance (ANOVA) followed by a post hoc Newman-Keuls test was used using GraphPad Prism version 5.04 (GraphPad Software, San Diego, CA, USA; www.graphpad.com). The results are presented as the means ± SD, n ≥ 3, unless otherwise stated. P ≤ 0.05 was considered statistically significant.
# Supplementary materials
Supplementary material for this article is available at http://advances.sciencemag.org/cgi/ content/full/5/10/eaaw3113/DC1 [fig_ref] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways... [/fig_ref]. The kinase activity of BRK and regulated signaling pathways. [fig_ref] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells [/fig_ref]. Identification of Halo-SMAD4-associated proteins. [fig_ref] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4 [/fig_ref]. Activated BRK phosphorylates tyrosine-353 and tyrosine-412 on SMAD4. . BRK and SMAD4 mRNA and protein expression in different cells. . Halo-SMAD4/Halo-SMAD4 Y353F-and Halo-SMAD4 Y412F-associated proteins in the presence or absence of SF-BRK-YF and their ubiquitination. [fig_ref] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent... [/fig_ref]. Gene ontology analyses for cellular components represented in the proteins associated with Halo-SMAD4 and phosphorylated Halo-SMAD4. . FRK-dependent regulation of EMT markers. . Differential protein interaction of Halo-SMAD4 in the presence of SNAP-F-BRK-WT or SNAP-F-BRK-YF (QSPEC log2 fold change, ≥1; QSPEC false discovery rate, ≤0.05). . Phosphorylation sites on SMAD4 detected by MudPIT analyses of in presence or absence of BRK-YF. . Gene ontology analysis for cellular component of Halo-SMAD4 in the presence of SNAP-F-BRK-WT or SNAP-F-BRK-YF in ClueGO FDR_0.05; Zscore_3: Tab: 3. Primers: Tab: 3. References [bib_ref] ClueGO: A Cytoscape plug-in to decipher functionally grouped gene ontology and pathway..., Bindea [/bib_ref] [bib_ref] Cytoscape: A software environment for integrated models of biomolecular interaction networks, Shannon [/bib_ref] View/request a protocol for this paper from Bio-protocol.
[fig] Figure 1: BRK is overexpressed in several human tumors and regulate different signaling pathways in normal and cancer cells. (A) Differential expression of BRK in five major cancer types. Data obtained from The Cancer Genome Atlas database, median ± one quartile; ***P < 0.001. Tissue samples are denoted N for normal and C for cancer tissue. (B) Activity of BRK-wild-type (WT) and BRK-Y447F (BRK-YF) mutants in transfected human embryonic kidney (HEK) 293 cells. BRK-WT and BRK-YF were transfected in HEK293 cells, and cell lysates were subjected to immunoblot with antiphosphotyrosine antibody (PY20), and anti-BRK and anti--tubulin served as a loading control. (C) Flow diagram of peptide arrays for kinome analysis. (D) Signaling pathways significantly (P < 0.05) affected by activated BRK as identified by kinome analysis in HEK293. [/fig]
[fig] Figure 2: Ectopically expressed BRK and SMAD4 interact and colocalize in HEK293 cells. (A and B)SNAP-FLAG-BRK-YF (SF-BRK-YF) and HALO-SMAD2/3/4 were expressed into HEK293 cells, and cell lysates were subjected to affinity purification (AP) with Halo magnetic beads (A) or SNAP capture magnetic beads (B) antibodies, followed by immunoblotting using anti-Halo and anti-FLAG antibodies. Bottom: The ectopic expression of BRK and SMAD2/SMAD3/SMAD4 as detected by anti-Halo and anti-FLAG antibodies. During affinity purification, either the SNAP_Flag or Halo tags were clipped off using Precision or Tev proteases, respectively. Halo-SMAD4 is ~93 kDa; after Halo-Tag removal, SMAD4 is ~60 kDa. Similarly, SNAP-Flag-BRK is ~73 kDa, and after SNAP-Tag removal, BRK is ~50 kDa. In (B), the three different blots in the middle were probed with the SMAD2, SMAD3, and SMAD4 antibodies. (C) HEK293 cells were cotransfected with SF-BRK-YF, Halo-SMAD2, Halo-SMAD3, and Halo-SMAD4, and the cell lysates were subjected to affinity purification with Halo magnetic beads or SNAP capture magnetic beads, followed by immunoblotting using anti-SMAD2, anti-SMAD3, anti-SMAD4, and anti-Flag antibodies. Total cell lysates were also analyzed by immunoblotting using antibodies against Halo-SMAD2, Halo-SMAD3, Halo-SMAD4, and Flag. (D) GFP-SMAD4 was cotransfected with BRK-W44A, BRK-∆SH2, BRK-∆SH3, BRK-WT, BRK-Y342F, or BRK-YF. The corresponding protein extracts were immunoprecipitated with anti-GFP and mouse immunoglobulin G (IgG; lysates from SMAD4 and BRK-YF cotransfected cells as a representative) and immunoblotted with anti-BRK and anti-GFP antibodies and -actin as a loading control. IP, immunoprecipitation. (E) SF-BRK-YF was expressed either alone or with Halo-SMAD4 [full length (FL)] or SMAD4 deletion mutants (A to E) in HEK293 cells. Total cell lysates were subjected to Halo affinity purification and analyzed by immunoblotting with FLAG and Halo antibodies. (F) Halo-SMAD4 or Halo plasmid alone with SF-BRK-WT or SF-BRK-YF was ectopically expressed in HEK293T cells. SNAP affinity purification followed by MudPIT mass spectrometry analysis showed that Halo-SMAD4 copurified with SNAP-Flag-BRK-WT or SNAP-Flag-BRK-YF but not with Halo alone. (G) Halo-SMAD4 or SF-BRK-YF alone or in combination was transfected into HEK293T cells. Halo-Tag TMRDirect fluorescent ligand (red) and SNAP-Cell 505-Star (green) were used to label Halo-Tag and SNAP-Tag proteins, respectively; DNA was stained with Hoechst dye (blue). DAPI, 4′,6-diamidino-2-phenylindole. [/fig]
[fig] Figure 3: Targeted proteomics reveals BRK-mediated tyrosine phosphorylation of SMAD4. (A) Workflow of global phosphorylation analysis by MudPIT mass spectrometry and targeted proteomics. MS/MS, tandem mass spectrometry. (B) Phosphorylation sites identified in this study (S, serine; T, threonine; Y, tyrosine) tabulated with known phosphorylation sites (www.phosphosite.org/proteinAction.action?id=1845&showAllSites=true on 7 February 2018). The frequency of detection and total spectral counts for the phosphorylated peptides are reported for the Halo-SMAD4 affinity purifications with or without BRK-YF. (C) Validation of previously unidentified tyrosine phosphorylations on Halo-SMAD4 Y353 and Y412 in the presence of SF-BRK-YF by MRM. For each phosphopeptide, at least four fragment ions containing the modified residue were targeted for MRM including respective nonmodified resides. (D) [/fig]
[fig] Figure 4, Figure 5: BRK and SMAD4 are expressed in most breast cancer cells and tissues. (A) BRK expression was detected by immunoblotting in the indicated normal mammary epithelial and breast cancer cell lines (TNBC, HER2, and ER), and -actin was used as a loading control. (B) Differential expression of SMAD4 and BRK in breast cancer cell lines, as obtained from the Cancer Cell Line Encyclopedia. (C) Absolute expression of BRK and SMAD4 in patient-driven tumor tissues of three major breast cancer subtypes: ER/PR, HER2, and TNBC. Super-stable isotope labeling by amino acids-based absolute proteins expression data were obtained from Tyanova et al.(19). (D) Immunoblotting analysis of total cell lysates from HEK293, MDA231, and MCF10A cells with or without ectopically expressed GFP-BRK-YF. Stable cell lysates were analyzed by immunoblotting using SMAD2, SMAD3, SMAD4, antiphosphotyrosine (4G10), and GFP antibodies. -Actin served as a loading control. (E) SF-BRK-YF was ectopically expressed in HEK293 cells and treated with MG132 for 8 hours, followed by immunoblotting analysis using Flag and SMAD4 antibodies. -Tubulin served as a loading control. (F) Total cell lysates from HEK293 and MCF10A; GFP-BRK-YF-expressing HEK293 and MCF10A stable cell lines and GFP-SMAD4-transfected GFP-BRK-YF-expressing stable HEK-293 and MCF10A stable cell lines were analyzed by immunoblotting using p21, SMAD4, and BRK-specific antibodies. -Actin was a loading control. Tyrosine phosphorylated SMAD4 interacts with enzymes of the ubiquitin pathway. (A) A workflow for discovery proteomics using MudPIT mass spectrometry for protein identifications. (B) SMAD4 polyubiquitination under denaturing condition: HEK293T cells were transiently transfected with Halo-SMAD4, SF-BRK-YF, or in combination with HA-ubiquitin plasmid. After 36 hours, the cells were treated with 10 M MG132 for an additional 8 hours. The total cell lysates were subjected to immunoprecipitation, followed by immunoblotting with anti-HA and anti-SMAD4 antibodies. (C and D) SMAD4 stability: Tet-On-inducible Halo-SMAD4 plasmid alone or Tet-Oninducible Halo-SMAD4 plasmid with SNAP-Flag-BRK-YF was transfected into HEK293T cells for the indicated time points and analyzed by immunoblotting with anti-Halo, anti-Flag, and -tubulin antibodies. The protein expression was quantified using ImageJ software and plotted as a bar chart. (E) Interaction with ubiquitin modifying enzymes: Relative QSPEC log 2 fold changes measured for the specified proteins in the APMS analyses of Halo-SMAD4/Halo-SMAD4 Y353F and Y412F with or without BRK-YF were plotted as heat maps (Genesis software package developed by A. Sturn and R. Snajder: http://genome.tugraz.at/genesisclient/genesisclient_description.shtml). (F) SMAD4 Y353F and Y412F polyubiquitination under denaturing condition: HEK293T cells were transiently transfected with Halo-SMAD4 Y353F and Y412F or SF-BRK-YF or in combination with HA-ubiquitin plasmid and analyzed as described above. (G and H) SMAD4 stability: Tet-On-inducible Halo-SMAD4 Y353F and Y412F plasmid with or without SNAP-Flag-BRK-YF was transfected into HEK293T cells and analyzed as described above. a.u., arbitrary units. [/fig]
[fig] Figure 6: BRK regulates tumor suppressor, EMT markers, and metastatic potential in a SMAD4-dependent manner. (A) Luciferase reporter constructs were transfected in HEK293 cells with and without SMAD4 to measure the transcriptional activation of the FRK promoter. (B) ChIP-qPCR experiment shows the relative abundance of SMAD4 in FRK promoter. (C and D) The mRNA levels of FRK were quantified via quantitative RT-PCR and protein levels were analyzed by immunoblotting of the total proteins extracted from parental, stably expressing BRK-YF and SMAD4-transfected BRK-YF expressing MDA-MB-231 stable cell lines. (E and F) The mRNA levels of SNAIL and SLUG were quantified via quantitative RT-PCR in the parental cell line, BRK-YF stable expressing MDA-MB-231 cell line and ectopically expressed SMAD4 in BRK-YF expressing stable cell line. (G) Cell adhesion assay shows the cell adhesion properties of HEK293 and MCF10A, BRK-YF stably expressing HEK293 and MCF10A cell lines and SMAD4-transfected stable cell lines. *P ≤ 0.05; **P ≤ 0.001; ***P ≤ 0.0001. (H) Activated BRK regulates EMT markers (SNAIL and SLUG) and cell adhesion by modulating SMAD4-FRK. [/fig]
|
Noble metal-comparable SERS enhancement from semiconducting metal oxides by making oxygen vacancies
## Supplementary
Calculation of the enhancement factor. The EF was calculated according to the formula:
[formula] EF = (I SERS /N SERS )/(I bulk /N bulk ) (1) N SERS = CVN A A Raman /A Sub (2) N bulk = MρhA Raman N A(3) [/formula]
Where I 0 is the incident laser intensity at frequency ω 0 , and ω IF is a molecular transition frequency between states I and F (presumably two different vibronic levels of the ground electronic state I e ).
The general expression for the polarizability tensor in the molecule-semiconductor system may be shown as:
[formula] σρ = ∑ ( ⟨ | σ | ⟩⟨ | ρ | ⟩ K − I −ℏ 0 + ⟨ | ρ | ⟩⟨ | σ | ⟩ K − F +ℏ 0 ) ≠ ,(5) [/formula]
Where K represents all the other states of the molecule, μ is the dipole moment operator, and σ, ρ are the scattered and incident polarization directions in space (X, Y, Z). Following
Where zero refers to zero-order Born-Oppenheimer states, H eN is the electron-nuclear attraction term in the Hamiltonian, evaluated at the equilibrium nuclear positions (0 For the purely electronic transition moment between states, we write: K = ⟨ e | | e ⟩, S K = ⟨ ′ e | | e ⟩, S = ⟨ e | | e ⟩, V = ⟨ e | | e ⟩, VK = ⟨ e | | e ⟩.
Substituting into Supplementary equation 5 (assuming that ω 0 > ω IF ), the expression of polarizability tensor α σρ in the defect-rich semiconductor-molecule system can be derived, which still involves the sum of three terms：
[formula] σρ = + + (12) = ∑ ∑ [ K σ K ρ ℏ( K − 0 ) + K ρ K σ ℏ( K + 0 ) ] e ≠ e ⟨ | ⟩⟨ | ⟩ (13) = ∑ ∑ ∑ {[ ℎ KS K σ S ρ ℏ( K − 0 ) + ℎ KS K ρ S σ ℏ( K + 0 ) ] ⟨ | ⟩⟨ | | ⟩ ℏ KS e ≠ e e ≠ e + [ ℎ KS S σ K ρ ℏ( K − 0 ) + ℎ KS S ρ K σ ℏ( K + 0 ) ] ⟨ | | ⟩⟨ | ⟩ ℏ KS } + ∑ ∑ ∑ {[ ℎ KV K σ V ρ ℏ( K − 0 ) + ℎ KV K ρ V σ ℏ( K + 0 ) ] ⟨ | ⟩⟨ | | ⟩ ℏ KV e ≠ e e ≠ e + [ ℎ KV V σ K ρ ℏ( K − 0 ) + ℎ KV V ρ K σ ℏ( K + 0 ) ] ⟨ | | ⟩⟨ | ⟩ ℏ KV } (14) = ∑ ∑ ∑ {[ ℎ S K σ KS ρ ℏ( K − 0 ) + ℎ S K ρ KS σ ℏ( K + 0 ) ] ⟨ | ⟩⟨ | | ⟩ ℏ S e ≠ e e ≠ e + [ ℎ S S K σ K ρ ℏ( K − 0 ) + ℎ S S K ρ K σ ℏ( K + 0 ) ] ⟨ | | ⟩⟨ | ⟩ ℏ S } + ∑ ∑ ∑ {[ ℎ V K σ KV ρ ℏ( K − 0 ) + ℎ V K ρ KV σ ℏ( K + 0 ) ] ⟨ | ⟩⟨ | | ⟩ ℏ V e ≠ e e ≠ e + [ ℎ V VK σ K ρ ℏ( K − 0 ) + ℎ V VK ρ K σ ℏ( K + 0 ) ] ⟨ | | ⟩⟨ | ⟩ ℏ V }(15) [/formula]
A represents the contribution of the molecular resonance to the polarizbility tensor via M IK , which is a term insusceptible to the defect states in the semiconductor, as described by
Lombardi and Wang.
## Supplementary references
[fig] I: SERS and bulk are the intensities of the selected Raman peak in the SERS and non-SERS spectra, and N SERS and N bulk are the average number of molecules in scattering area for SERS and non-SERS measurement. The data for R6G (0.05 M) on bare Si/SO 2 substrate were used as non-SERS-active reference. Specifically, the intensity was obtained by taking average from measurements of 30 spots, and the number of analyte molecules was estimated by Supplementary equation 2 on the assumption that the analyte molecules were distributed uniformly on the substrates. C is the molar concentration of the analyte solution, V is the volume of the droplet, N A is Avogadro constant. A Raman is the laser spot area (1 μm in diameter) of Raman scanning. Twenty microliters of the droplet on the substrate was spread into a circle of about 3 mm in diameter after solvent evaporation, from which the effective area of the substrate, A Sub , can be obtained. The confocal depth (h) of the laser beam is 21 μm, 10 and on the basis of molecular weight (M) and density (ρ) of bulk R6G (1.15 g cm -3 ), N bulk is calculated by Supplementary equation 3. The uniformity and reproducibility for statistical analysis. The uniformity was acquired by recording spectra at 30 stochastic spots at different locations across one substrate. The reproducibility between different substrates was acquired by recording 40 spectra from eight different substrates (five stochastic spots per substrate). The peak at 612 cm −1 (P1) and 1360 cm −1 (P3) of R6G Raman spectrum were chosen for quantification because of their relative insensitivity and the representative for different type of vibrations. The calculation of the contribution of photo-induced charge transfer (PCT) to the molecule polarization tensor in the defect-rich semiconductor-molecule system. Based on the Herzberg-Teller theory regarding the vibronic coupling of the zero-order Born-Oppenhermer states, the contribution of photo-induced charge transfer (PCT) to the molecule polarization tensor in the semiconductor-molecule system can be estimated. The calculation in this communication is derived from the treatment in a semiconductor-molecule system reported byLombardi and Wang,9,11 respectively, except for the difference that we include the vibronic coupling of the defect states | 〉 in with the molecular exited state | 〉 and the molecular ground state | 〉 in the semiconductor-molecule system.The intensity of a Raman transition may be obtained from the molecular polarizability tensor by the expression: [/fig]
[fig] B: represents the contribution of photo-induced charge transfer of molecule-tosemiconductor to the polarizability tensor via M IS and M IV . The transition borrow intensity from the allowed transition M IK by means of vibronic coupling between the excited molecular state K and semiconductor conductor band state S through the matrix element h KS , similarly, the transition borrow intensity from the allowed transition M IK by means of vibronic coupling between the excited molecular state K and semiconductor defect state V through the matrix element h KV . C represents the contribution of photo-induced charge transfer of semiconductor-tomolecule to the polarizability tensor via M S'K and M VK . The transition borrow intensity from the allowed transition M IK by means of vibronic coupling between the molecular ground state I and semiconductor valence band state S' through the matrix element h IS' , similarly, the transition borrow intensity from the allowed transition M IK by means of vibronic coupling between the molecular ground state I and semiconductor defect state V through the matrix element h IV . This is illustrated in the scheme of Supplementary Figure 4. [/fig]
|
Towards the best kidney failure prediction tool: a systematic review and selection aid
A B S T R A C TBackground. Prediction tools that identify chronic kidney disease (CKD) patients at a high risk of developing kidney failure have the potential for great clinical value, but limited uptake.The aim of the current study is to systematically review all available models predicting kidney failure in CKD patients, organize empirical evidence on their validity and ultimately provide guidance in the interpretation and uptake of these tools. Methods. PubMed and EMBASE were searched for relevant articles. Titles, abstracts and full-text articles were sequentially screened for inclusion by two independent researchers. Data on study design, model development and performance were extracted. The risk of bias and clinical usefulness were assessed and combined in order to provide recommendations on which models to use. Results. Of 2183 screened studies, a total of 42 studies were included in the current review. Most studies showed high discriminatory capacity and the included predictors had large overlap. Overall, the risk of bias was high. Slightly less than half the studies (48%) presented enough detail for the use of their prediction tool in practice and few models were externally validated. Conclusions. The current systematic review may be used as a tool to select the most appropriate and robust prognostic model for various settings. Although some models showed great potential, many lacked clinical relevance due to being developed in a prevalent patient population with a wide range of disease severity. Future research efforts should focus on external validation and impact assessment in clinically relevant patient populations.
## B a c k g r o u n d
Chronic kidney disease (CKD) may lead to kidney failure, although the rates of progression vary substantially between individuals . Prediction tools that can identify patients at high risk of developing kidney failure could have great clinical value. They could be used to inform individualized decision making, employed in determining the appropriate time for referral to nephrologists and used in the planning and preparation of renal replacement therapy (RRT). Prediction tools might also offer opportunities for risk stratification in research and improvement of health policies [bib_ref] Predicting progression in CKD: perspectives and precautions, Kadatz [/bib_ref].
Multiple prediction models have been developed to identify individuals at high risk of kidney failure and have been previously described in two systematic reviews [bib_ref] Risk prediction models for patients with chronic kidney disease, Tangri [/bib_ref] [bib_ref] Risk models to predict chronic kidney disease and its progression: a systematic..., Echouffo-Tcheugui [/bib_ref]. Many of these models showed good predictive abilities in development. However, despite nephrologists and patients acknowledging a lack of prognosis discussions in practice, clinical uptake of these tools is still limited [bib_ref] Discussions of the kidney disease trajectory by elderly patients and nephrologists: a..., Schell [/bib_ref]. Policymakers also seem hesitant in endorsing prediction tools. The most recent Kidney Disease: Improving Global Outcomes guideline recommends the use of prediction models for timely referral for planning RRT [bib_ref] Chapter 5: referral to specialists and models of care, Kdigo-Group [/bib_ref]. However, the guideline, fails to provide guidance on which risk prediction tool should be used to do so. The lack of uptake by clinicians and policymakers has been partly attributed to substandard methodology, a lack of external validation and a shortage of easy calculation options [bib_ref] A systematic review finds prediction models for chronic kidney disease were poorly..., Collins [/bib_ref].
The last two published reviews in 2012 and 2013 included eight studies each on prediction of kidney failure in CKD patients [bib_ref] Risk prediction models for patients with chronic kidney disease, Tangri [/bib_ref] [bib_ref] Risk models to predict chronic kidney disease and its progression: a systematic..., Echouffo-Tcheugui [/bib_ref]. Since then the number of available models has greatly increased. A new systematic review of the available models is the first step towards the use and recommendation of robust prognostic tools. The aim of the current study is therefore to systematically review all available models predicting kidney failure in CKD patients, organize empirical evidence on their validity and ultimately provide guidance in the selection of the best prediction tool for various settings.
## M a t e r i a l s a n d m e t h o d s
## Data sources and searches
The current review was framed by the search for prognostic prediction models for CKD patients, predicting the future event V C The Author(s) 2019. Published by Oxford University Press on behalf of ERA-EDTA. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] 1527 of kidney failure. To ensure transparent reporting and accurate study appraisal, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) and Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) guidelines were followed where applicable [bib_ref] The PRISMA statement for reporting systematic reviews and meta-analyses of studies that..., Liberati [/bib_ref] [bib_ref] Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis..., Collins [/bib_ref] [bib_ref] Critical appraisal and data extraction for systematic reviews of prediction modelling studies:..., Moons [/bib_ref]. The completed PRISMA checklist is provided as supplementary material. We searched the PubMed and EMBASE databases on 31 December 2017 for English-language studies regarding risk prediction in CKD patients. The search strategies were designed to include relevant development, validation and implementation studies and are provided in Appendix A1.
## Study selection
Titles, abstracts and full-text articles were sequentially screened for inclusion by two independent researchers (C.L.R. and Y.J.). Discrepancies on inclusion of full-text articles were resolved by consulting a third co-author (M.D.). Articles were included if they met the following predefined selection criteria: (i) the study must develop, validate, update or implement a multivariate prognostic prediction model, with a prediction research question as the aim, as opposed to an aetiological or methodological goal; (ii) the study must present at least one measure to assess model performance; (iii) the study population must consist of adult CKD patients and (iv) the study outcome must include kidney failure or end-stage renal disease. The references of included studies and related reviews were manually screened in order to identify additional relevant studies.
## Data extraction and quality assessment
Following selection, two reviewers (C.L.R. and Y.J.) independently conducted the data extraction and quality assessment. Discrepancies were discussed with input from an additional coauthor (M.D.) where necessary. Conforming with CHARMS recommendations, information on the source of the data, population, outcome, sample size, missing data, model development and model performance were extracted and summarized. Additionally, data on external validations of models were extracted. Furthermore, the risk of bias and clinical usefulness were judged by both reviewers independently. In order to facilitate further comparison, studies were grouped by study population, which ranged from very broad (general CKD) to specific CKD subgroups such as immunoglobulin A (IgA) nephropathy or diabetic nephropathy. Quality and risk of bias were assessed in both development and validation studies by making use of a novel tool, the Prediction Study Risk of Bias Assessment Tool (PROBAST). Although this tool has yet to be published in its complete form, there is no other formal risk of bias assessment available that is applicable to prediction studies. The PROBAST is specifically designed for systematic reviews of prediction studies and is used as a domain-based approach with 23 signalling questions that categorize the risk of bias into high, low or unclear for five separate domains: participant selection, predictors, outcome, sample size and missing data, and analysis. It also assesses the usability of a model. It has been used in multiple reviews in the past year and was presented in part at the 2016 Cochrane Colloquia. The final test version of PROBAST was obtained through personal e-mail contact with Dr R.C.G. Wolff.
## Data synthesis
Given the multitude of different models and heterogeneity in study characteristics, we opted for a narrative synthesis of results supported by extensive tables and figures with study characteristics listed per article. Model performance was evaluated by examining the discrimination and calibration of included prediction tools. Discrimination is most often described by the C-statistic and indicates how well the model discriminates between patients with and without the event of interest. It lies between 0.5 and 1, where 0.5 is similar to tossing a coin and 1 indicates perfect discrimination [bib_ref] Prognosis and prognostic research: Developing a prognostic model, Royston [/bib_ref]. Important to take into account is that the C-statistic of the same model can vary greatly, dependent on the population on which the model is tested. When a population is heterogeneous in the predictors that make up the prediction tool, the C-statistic may increase substantially [bib_ref] External validity of risk models: use of benchmark values to disentangle a..., Vergouwe [/bib_ref]. On the other hand, calibration describes the agreement between the absolute number of predicted events and observed events population wide. It is best represented in a plot, wherein the predicted probability of kidney failure is plotted against the observed rate of kidney failure [bib_ref] Prognosis and prognostic research: Developing a prognostic model, Royston [/bib_ref]. To evaluate the sample size and risk of overfitting in development studies, the events per candidate predictor (EPV) were extracted. A minimum of 10 EPV has been suggested as rule of thumb for an acceptable sample size in model development studies [bib_ref] A simulation study of sample size demonstrated the importance of the number..., Wynants [/bib_ref]. For external validation studies, it has been recommended to include a minimum of 100 events in total to obtain a precise estimate of performance [bib_ref] Sample size considerations for the external validation of a multivariable prognostic model:..., Collins [/bib_ref].
## R e s u l t s
## Study selection
The study selection process is described in a flowchart [fig_ref] FIGURE 1: PRISMA flow diagram of study inclusion [/fig_ref]. Overall, 2183 titles were identified, of which 431 abstracts were assessed, and 90 full-text publications were evaluated in depth. From these articles, a final 42 studies met all inclusion criteria and were included in the current review. Most full-text exclusions were due to the predicted outcome not including kidney failure or the lack of a multivariate model. Although prediction research has seen a great surge in nephrology over the last few years, the first included predictive model was published in 1986 for IgA nephropathy patients. Since the beginning of the 2000s, a substantial increase of published models is apparent, as can be seen in [fig_ref] FIGURE 2: Cumulative number of published development and validation studies for models that predict... [/fig_ref]. Although the number of developed models has increased almost every year, the number of validation studies has remained small. Of the 42 included studies, 7 exclusively externally validated already existing models [bib_ref] Estimating time to ESRD using kidney failure risk equations: results from the..., Grams [/bib_ref] [bib_ref] Validation of the kidney failure risk equation in European CKD patients, Peeters [/bib_ref] [bib_ref] Multinational assessment of accuracy of equations for predicting risk of kidney failure:..., Tangri [/bib_ref] [bib_ref] External validation of the kidney failure risk equation and re-calibration with addition..., Lennartz [/bib_ref] [bib_ref] Long-term risk of ESRD in IgAN; validation of Japanese prognostic model in..., Bjorneklett [/bib_ref] [bib_ref] Addition of eGFR and age improves the prognostic absolute renal risk-model in..., Knoop [/bib_ref] [bib_ref] Validation of the absolute renal risk of dialysis/death in adults with IgA..., Mohey [/bib_ref]. Besides development, 10 studies also externally validated their own or previously published models. Disconcertingly, no study assessing the impact of using such a prediction tool was found, which ultimately is the only way of assessing whether the model can improve patient care.
## Characteristics of development studies
A total of 35 studies were published on the development of novel tools to predict kidney failure in CKD patients. Generally, a distinction can be made between models developed for a general CKD patient population (n ¼ 16) and models developed for a population with a specified primary renal disease (n ¼ 19), mainly IgA nephropathy or diabetic nephropathy. The characteristics of all included development studies are described in [fig_ref] Table 1: Baseline characteristics of model development studies [/fig_ref]. Since each study developed between 1 and 12 prediction models, the results presented in [fig_ref] Table 1: Baseline characteristics of model development studies [/fig_ref] concern the final model(s) as selected by the authors or the model with the best performance if no final model was suggested. The population size differed greatly between studies and ranged from 75 to Both studies by Johnson et al. [bib_ref] Predicting the risk of dialysis and transplant among patients with CKD: a..., Johnson [/bib_ref] [bib_ref] Predicting renal replacement therapy and mortality in CKD, Johnson [/bib_ref] overlap in patient population and include the same predictors. The study by Schroeder et al. [bib_ref] Predicting 5-year risk of RRT in stage 3 or 4 CKD: development..., Schroeder [/bib_ref] updates this same model [bib_ref] Predicting the risk of dialysis and transplant among patients with CKD: a..., Johnson [/bib_ref] (the KPNW) by including additional predictors and excluding some original predictors. The formula as provided in the supplement of Schroeder et al.'s article [bib_ref] Predicting 5-year risk of RRT in stage 3 or 4 CKD: development..., Schroeder [/bib_ref] does not provide the knot locations for spline terms. These are available from the authors upon request. c
The study by Xie [bib_ref] Risk prediction to inform surveillance of chronic kidney disease in the US..., Xie [/bib_ref] and Maziarz [bib_ref] Evaluating risk of ESRD in the urban poor, Maziarz [/bib_ref] includes the same patient population. Part of this population is included in Maziarz et al. [bib_ref] Homelessness and risk of end-stage renal disease, Maziarz [/bib_ref]. All three studies include the same predictors in the same four models but re-estimate b-coefficients for different subsets. The population is of underserved/unin- 28 779 patients. A small sample size was a problem in 17/35 studies, as they had <10 EPV, thus running the substantial risk of overfitting their model [bib_ref] A simulation study of sample size demonstrated the importance of the number..., Wynants [/bib_ref]. To assess to what extent these models are overfit, external validation is of key importance. Before the validity of these models is tested, they should not be used in practice. For specific renal diseases, the baseline was almost always the first biopsy (and disease confirmation), providing a clear moment in time for when to use the prognostic model or score. Models developed in general CKD, however, rarely defined the moment in time when their prediction tool should be used, as most of these studies enrolled prevalent CKD patients with a wide range of disease severity. Only two models were developed on incident patients, who were included at the first referral to a nephrologist [bib_ref] A dynamic predictive model for progression of CKD, Tangri [/bib_ref] [bib_ref] A predictive model for progression of chronic kidney disease to kidney failure, Tangri [/bib_ref]. There was some variation in outcome definitions, but for most studies, renal failure was defined as the need for RRT (dialysis start or kidney transplantation). Five studies used estimated glomerular filtration rate (eGFR) or creatinine as a proxy for kidney failure. Two development studies used RRT start or death as a composite outcome measure. A total of four studies did not report their definition of ESRD. The time frame over which the models predict kidney failure ranged from 6 months to 20 years and nine studies failed to define a prediction time frame, presumably using the maximum study follow-up. The specific predictors included per development study are presented in [fig_ref] FIGURE 3: Predictors included in development studies [/fig_ref]. There is a large amount of overlap in final predictors with almost all studies including age, sex, eGFR (or serum creatinine), proteinuria and histological features for IgA nephropathy tools. Concerning the reporting of performance measures, discrimination measures were reported far more often than calibration measures. Discrimination in the form of a C-statistic was reported in 28/35 studies. The C-statistic ranged from 0.72 to 0.96 and was generally high, indicating good to excellent discrimination in most studies. Calibration was presented far less frequently, with only 11 studies presenting a calibration plot, bar chart or test.
In order to calculate an individual's risk, the model constant and hazard ratios (HRs)/regression coefficients per predictor are needed. Many studies only presented HRs per predictor without the constant (intercept or baseline hazard value), and some gave no data on the model equation at all. The full formula for the developed model was presented in only 6/35 studies. Just three studies provided a web calculator for easy use, of which two web calculators are no longer working. A total of 13 studies provided a simplified scoring system. In total, 25 final models were validated in some form, either internally and/or externally. Cross-validation, bootstrapping and random split sample were the most used forms of internal validation.
## Characteristics of external validation studies
A total of 17 studies externally validated one or more of the developed prediction tools. The characteristics of these models and validations can be found in [fig_ref] Table 2: Characteristics of external validation studies and model performance in validations [/fig_ref]. Most validation studies were performed by the same group of researchers who developed the models and were often presented in the same publication as the development. Compared with the development performance, the C-statistic was lower in 68% of the validations. Two studies updated the validated model by recalibrating the baseline hazard and two studies added predictors to the existing model. In total, five risk scores predicting prognosis in IgA nephropathy patients and seven prognostic tools for general CKD patients were externally validated. Only the Absolute Renal Risk (ARR) score, Goto score and Kidney Failure Risk Equation (KFRE) (three, four and eight variables) were validated multiple times. The largest validation study of the KFRE was performed by Tangri et al. [bib_ref] Multinational assessment of accuracy of equations for predicting risk of kidney failure:..., Tangri [/bib_ref] and summarized the validation of the KFRE in >30 countries, including more than half a million patients.
## Risk of bias
Risk of bias was assessed in all 42 included studies, using signalling questions from the PROBAST specified for detecting methodological flaws in both development and validation prediction studies. Overall, the risk of bias was high, as can be seen inand B. Forty-one of 42 studies received a high risk of bias in at least one of the five domains; the only study with an overall low risk of bias was by Schroeder et al. [bib_ref] Predicting 5-year risk of RRT in stage 3 or 4 CKD: development..., Schroeder [/bib_ref]. The majority of studies had a high risk of bias in the domain sample size and missing data. This was often due to the use of complete case analysis, which is generally an inappropriate method of handling missing data. A small sample size was a frequent problem limiting model usage, as a small sample often results in an overfit model and thereby biased results. In the domain statistical analysis, 83% of studies had a high risk of bias. The most common reason was incomplete reporting of performance measures, as few studies reported sufficient calibration results. Also, many studies did not correct their model for overfitting through internal validation. The usability of the model was assessed in a separate domain. If in the publication the full model formula, a calculator or a risk score with absolute risk table was available, then the tool was considered usable. Less than half the studies (48%) presented enough detail for the use of their prediction tool in practice. The usable models that specified a prediction time frame are presented in [fig_ref] FIGURE 5: Model selection guide for CKD patients [/fig_ref] , categorized by the type of patient population and outcome. This figure may be employed as a selection guide when wanting to calculate an individuals' prognosis, taking into account that many of the models have significant shortcomings and may not be ready for clinical use.
## D i s c u s s i o n
This systematic review provides an overview of all development and validation studies of predictive models for progression of CKD to kidney failure. Since the last reviews on this topic, the number of publications has more than doubled [bib_ref] Risk prediction models for patients with chronic kidney disease, Tangri [/bib_ref]. Most included studies report high model performance measures, implying that calculating an individual's risk of renal failure with high accuracy is attainable. This is further emphasized by the similar predictors included in various models. There were, however, substantial shortcomings in many publications. As in many medical prediction studies, aetiological and prediction goals were often confused, limiting interpretability and applicability [bib_ref] A systematic review finds prediction models for chronic kidney disease were poorly..., Collins [/bib_ref] [bib_ref] Prediction versus aetiology: common pitfalls and how to avoid them, Van Diepen [/bib_ref]. First, more than half the tools provided insufficient details to calculate an individual's prognosis of kidney failure, rendering it useless for its intended purpose. Second, the clinical relevance of many models is limited due to the selection of the derivation population. Third, a high risk of bias was observed across studies, mainly due to the high risk of overfitting, inadequate handling of missing data and incomplete reporting of performance measures. Fourth, sufficient validation was largely lacking, increasing research waste and limiting the reliability of models. And finally, not a single impact study on the effect of clinical uptake has been performed. It is therefore not surprising that clinical uptake of models remains sporadic and guidelines on which model to use are lacking.
Providing absolute evidence for the single 'best' prognostic tool to use is complicated by differences between studies, mainly concerning varying study populations, use of different prediction baselines, use of varying time frames and multiple outcome definitions. A selection guide including all usable models is presented that may assist clinicians and patients in choosing the tool appropriate to their setting [fig_ref] FIGURE 5: Model selection guide for CKD patients [/fig_ref]. There are many factors to take into account when selecting the most appropriate model, depending on the user's wishes and specific clinical setting. Users should be wary of overfitting in models developed on a small sample size and we would advise against the use of these models unless validated in a sufficiently large sample. Based on our results, we would advise the use of a tool with an overall low risk of bias that has shown good performance in external validation in a similar population to the 1532 C. L. Ramspek et al. Kidney failure prediction tools 1533 population in which the use is intended and has ideally been assessed in an impact study. For kidney failure prediction in a general CKD cohort with Stages 3-5 patients, we would recommend the four-or eightvariable KFRE, as it has been externally validated extensively for a time frame of 2 and 5 years. Although the development study potentially introduced bias by selecting predictors that were recorded up to 365 days after prediction baseline and by using univariate analysis to select predictors, the model has shown consistently good performance in CKD Stages 3-5 patients from less-biased external validation studies [bib_ref] Multinational assessment of accuracy of equations for predicting risk of kidney failure:..., Tangri [/bib_ref] [bib_ref] A predictive model for progression of chronic kidney disease to kidney failure, Tangri [/bib_ref]. Alternatively, for 5-year predictions, the Kaiser Permanente Northwest (KPNW) model as updated and externally validated by Schroeder et al. [bib_ref] Predicting 5-year risk of RRT in stage 3 or 4 CKD: development..., Schroeder [/bib_ref] also has great potential, mainly due to its methodological rigor and low risk of bias, although it is less easy to use than the KFRE. Various other general CKD models showed promising results in development but should be further externally validated to ensure consistency of performance before clinical use [bib_ref] A dynamic predictive model for progression of CKD, Tangri [/bib_ref] [bib_ref] Looking to the future: predicting renal replacement outcomes in a large community..., Marks [/bib_ref] [bib_ref] A simple tool to predict end-stage renal disease within 1 year in..., Drawz [/bib_ref]. For prediction of disease progression in IgA nephropathy patients, a large number of models are available. However, these models, were generally developed on a small sample size and often had a high risk of bias. The most evidence on validity was found for the risk scores developed by Goto et al. [bib_ref] A scoring system to predict renal outcome in IgA nephropathy: a nationwide..., Goto [/bib_ref] and the ARR (by Berthoux et al. [bib_ref] Predicting the risk for dialysis or death in IgA nephropathy, Berthoux [/bib_ref]. The Goto score contains some risk of bias due to a complete case analysis and univariate selection of predictors, but was developed on a relatively large sample size and has been externally validated twice. Although the ARR score was developed using questionable model building methods and with incomplete reporting of performance, this score has been externally validated the most times, and a recently updated version presented by al. [bib_ref] Addition of eGFR and age improves the prognostic absolute renal risk-model in..., Knoop [/bib_ref] shows great potential. Clinical relevance proved to be largely lacking for many of the included models in the current review. Specifically, models for general CKD patients were often developed on prevalent patients with a wide range of disease severity and did not specify a specific time point when the model should be used. Prediction of renal failure can be extremely accurate when using a population with GFRs ranging from 10 to 60 mL/min/ 1.73 m 2 . However, in practice, such tools would probably be employed for a more homogeneous group of patients in which it is clinically relevant to discuss prognosis. The predictive capacities of the model would be lower in such a population. This is exemplified in the KFRE validation performed by Peeters et al. [bib_ref] Validation of the kidney failure risk equation in European CKD patients, Peeters [/bib_ref] , where the area under the curve of the fourvariable KFRE dramatically decreased from 0.88 in the whole population (CKD Stages 3-5) to 0.71 in the more relevant population of CKD Stage 4 patients. Another factor limiting usability and interpretability is that the number of studies did not define the prediction time frame. Finally, the definition of outcome differs between studies. The use of composite endpoints is particularly problematic, as it limits the value of the model for clinicians, as each separate endpoint requires different interventions. In conclusion, an ideal model is developed for one clearly defined clinically meaningful and objective endpoint in a population for which prediction is clinically relevant. Few models included in this review met these recommendations and this lack of clinical relevance could be a large contributor to the slow uptake seen in practice.
Despite the limited uptake and discussed shortcomings of existing tools, risk prediction models for kidney failure have a great potential for improving patients' decision making, treatment and overall health. In future studies there is the need for improvement of the quality of reporting and methodology used. As the majority of models included had a high risk of bias, these models should not be implemented unless their validity is proven in unbiased external validation studies. Hopefully efforts such as the TRIPOD guidelines will correct these inadequacies and result in more robust, usable and unbiased prognostic tools [bib_ref] Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis..., Collins [/bib_ref]. To limit research waste and improve clinical uptake, it is of crucial importance that development studies provide enough model information (formula/score with absolute risk table) to enable their use. For specific renal diseases and homogeneous patient populations, there certainly appears to be room for improvement in model development. For populations in which multiple models are available, we advise that future research should focus on the updating, validation and implementation of these existing prognostic tools. Previous studies have shown that the combination of well-established clinical risk factors and kidney disease markers can accurately predict renal failure in a general CKD population. Therefore one might advise focusing resources on updating models for more clinically relevant populations in an unbiased fashion. In this step, external validation of multiple models in the same population is of key importance. Additionally, translation of mathematical model formulas to simple tools such as web calculators and enabling automated uptake is of great importance for integration into daily clinical routine. Ultimately, impact studies will be necessary to determine whether the implementation of such tools truly improves patient outcomes. Ideally, such impact studies would be randomized controlled trials and would assess the effect of implementing a prediction model in [bib_ref] Risk prediction models for patients with chronic kidney disease, Tangri [/bib_ref] In this graph, only models that allow calculation of an individual's prognosis and are therefore labelled as usable are included. This entails that these models provide either a full formula, score with absolute risk table or (currently working) web calculator for a specified prediction time frame. For categories containing multiple models, the risk of bias combined with evidence of external validity was weighed in determining the model order, starting with the most valid and least biased models. Nevertheless, many of the models listed have significant shortcomings and should be used with caution.
Kidney failure prediction tools clinical practice. Different outcomes might be considered as endpoints in such studies, partly dependent on the time of prediction. Relevant outcomes might be timely referral to nephrologists, timely placement of vascular access, better informed patients, improved quality of life and possibly even improved survival.
The current review has a number of strengths. First, we expect to have included a complete overview of existing models. Furthermore, this is the first study on kidney failure models to perform a formal risk of bias assessment aimed specifically at prediction research. The study is limited by the inclusion of only English-language articles. Also, the differences in case mix and characteristics of included studies make it difficult to directly compare their performances. Herein we are limited by the lack of validation studies that compare multiple models in the same cohort. Finally, we limited the scope of this review to models predicting kidney failure, although other outcomes such as death or cardiovascular events may also have significant clinical value.
In conclusion, this study provides a systematic overview of existing models for predicting progression to kidney failure in CKD patients. The results may be used as a tool to select the most appropriate and robust prognostic model for various settings. Finally, we hope the current review motivates researchers in this field to decrease the generation of new models and combine efforts to explore, analyse and update existing models in clinically relevant settings to ultimately stimulate clinical uptake and improve patient outcomes.
## S u p p l e m e n t a r y d a t a
Supplementary data are available at ndt online.
## F u n d i n g
The work on this study by M.D. and Y.J. was supported by a grant from the Dutch Kidney Foundation (16OKG12). Patient representatives were involved in framing the research question and gave input on the clinical relevance of this project. The Dutch Kidney Foundation was not involved in the study design, interpretation of results or publication approval.
## A u t h o r s ' c o n t r i b u t i o n s
All authors have made substantial contributions to the conception of the work and the acquisition and interpretation of data. C.L.R. drafted the work and Y.J., F.W.D. and M.D. critically revised the work. The final version of this manuscript was approved by all the authors and all authors agree to be accountable for all aspects of the work.
## C o n f l i c t o f i n t e r e s t s t a t e m e n t
The authors have no competing interests to declare. All authors declare no support from any organization for the submitted work, no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years and no other relationships or activities that could appear to have influenced the submitted work.
[fig] FIGURE 1: PRISMA flow diagram of study inclusion. [/fig]
[fig] FIGURE 2: Cumulative number of published development and validation studies for models that predict kidney failure in CKD patients (N ¼ 42).Kidney failure prediction tools [/fig]
[fig] FIGURE 3: Predictors included in development studies (N ¼ 35). The inclusion of a predictor is shown as 'X'. The subscript under X (e.g. 'X 2 ') indicates the number of predictors included from that category.Kidney failure prediction tools [/fig]
[fig] FIGURE 4: (A) Risk of bias and usability of prediction models (N ¼ 42). Assessed using the PROBAST. The five risk of bias domains were evaluated as low risk (þ), unclear risk (?) or high risk (À). Usability was evaluated as yes (þ) or no (À). (B) PROBAST risk of bias summary for all studies (N ¼ 42). [/fig]
[fig] FIGURE 5: Model selection guide for CKD patients. [/fig]
[table] Table 1: Baseline characteristics of model development studies (N [/table]
[table] Table 2: Characteristics of external validation studies and model performance in validations (N [/table]
|
Biochemical properties of thyroid peroxidase (TPO) expressed in human breast and mammary-derived cell lines
Thyroid peroxidase (TPO) is an enzyme and autoantigen expressed in thyroid and breast tissues. Thyroid TPO undergoes a complex maturation process however, nothing is known about post-translational modifications of breast-expressed TPO. In this study, we have investigated the biochemical properties of TPO expressed in normal and cancerous human breast tissues, and the maturation process and antigenicity of TPO present in a panel of human breast tissue-derived cell lines. We found that the molecular weight of breast TPO was slightly lower than that of thyroid TPO due to decreased glycosylation and as suggest results of Western blot also shorter amino acid chain. Breast TPO exhibit enzymatic activity and isoelectric point comparable to that of thyroid TPO. The biochemical properties of TPO expressed in mammary cell lines and normal thyrocytes are similar regarding glycan content, molecular weight and isoelectric point. However, no peroxidase activity and dimer formation was detected in any of these cell lines since the majority of TPO protein was localized in the cytoplasmic compartment, and the TPO expression at the cell surface was too low to detect its enzymatic activity. Lactoperoxidase, a protein highly homologous to TPO expressed also in breast tissues, does not influence the obtained data. TPO expressed in the cell lines was recognized by a broad panel of TPO-specific antibodies. Although some differences in biochemical properties between thyroid and breast TPO were observed, they do not seem to be critical for the overall three-dimensional structure. This conclusion is supported by the fact that TPO expressed in breast tissues and cell lines reacts well with conformation-sensitive antibodies. Taking into account a close resemblance between both proteins, especially high antigenicity, future studies should investigate the potential immunotherapies directed against breast-expressed TPO and its specific epitopes.
# Introduction
Human thyroid peroxidase (TPO), the crucial enzyme responsible for biosynthesis of hormones by the thyroid gland, catalyzes iodination and coupling of tyrosine residues in thyroglobulin, which leads to the synthesis of triiodothyronine and thyroxine [bib_ref] Philadelphia: Werner and Ingbar's; 2012, Kopp [/bib_ref] [bib_ref] Chemistry and Biology in the Biosynthesis and Action of Thyroid Hormones, Mondal [/bib_ref]. TPO is also a a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 Editor: Salvatore V. Pizzo, Duke University School of Medicine, UNITED STATES major autoantigen in autoimmune thyroid disease (AITD). A majority of polyclonal TPO-specific antibodies (TPOAbs) present in sera of AITD patients react with epitopes located on two discontinuous, three-dimensional integrity-dependent immunodominant regions (IDR) on the surface of the TPO molecule, termed A and B (IDR-A and-B) [bib_ref] Thyroid peroxidase as an autoantigen, Mclachlan [/bib_ref] [bib_ref] Thyroid autoantibodies: thyroid peroxidase and thyroglobulin antibodies, Czarnocka [/bib_ref] [bib_ref] Thyroid autoimmunity, Orgiazzi [/bib_ref]. These regions have been detected both in antigenic competition experiments with a panel of murine monoclonal antibodies (mAbs) [bib_ref] Relationship between immunological structure and biochemical properties of human thyroid peroxidase, Ruf [/bib_ref] and using recombinant human Fab fragments [bib_ref] Immunoglobulin G kappa antithyroid peroxidase antibodies in Hashimoto's thyroiditis: epitope-mapping analysis, Czarnocka [/bib_ref] [bib_ref] Human organ-specific autoimmune disease. Molecular cloning and expression of an autoantibody gene..., Chazenbalk [/bib_ref].
TPO, together with myeloperoxidase (MPO), lactoperoxidase (LPO) and eosinophil peroxidase (EPO), belongs to the family of heme-containing human peroxidases. The human TPO gene is located on chromosome 2 and encodes a 933-amino acid protein. The mature TPO protein has a molecular weight of approximately 100 kDa and consists of a large N-terminal extracellular ectodomain followed by short transmembrane and cytoplasmic regions. The ectodomain, exposed to the lumen of thyroid follicles, is composed of an N-terminal signal peptide, a propeptide, and the following subsequent domains: N-terminal MPO-like domain, complement control protein (CCP)-like domain, and epidermal growth factor (EGF)-like domain. During intracellular trafficking to the cell membrane, TPO undergoes several posttranslational modifications, such as proteolytic trimming, glycosylation, heme fixation, and finally dimerization. Newly synthesized TPO molecules undergo core glycosylation and the heme incorporation in the membrane of the endoplasmic reticulum [bib_ref] Role of heme in intracellular trafficking of thyroperoxidase and involvement of H2O2..., Fayadat [/bib_ref] [bib_ref] Human thyroperoxidase is largely retained and rapidly degraded in the endoplasmic reticulum...., Fayadat [/bib_ref] , and the oligosaccharides of the TPO molecules are further modified while being transported via the secretory pathway [bib_ref] Human thyroperoxidase is largely retained and rapidly degraded in the endoplasmic reticulum...., Fayadat [/bib_ref]. The N-terminal propeptide is removed after exiting the Golgi apparatus complex but before the molecules reach the cell membranes [bib_ref] Endoproteolytic cleavage of human thyroperoxidase: role of the propeptide in the protein..., Le Fourn [/bib_ref]. The processes of TPO dimerization and the homodimer organization are rather poorly understood. However, one molecular modeling study provided structural insight to the dimerization of TPO molecules [bib_ref] Modelling of Thyroid Peroxidase Reveals Insights into Its Enzyme Function and Autoantigenicity, Le [/bib_ref]. Interestingly, it suggested that only TPO dissociated into monomers is fully accessible for autoantibodies [bib_ref] Modelling of Thyroid Peroxidase Reveals Insights into Its Enzyme Function and Autoantigenicity, Le [/bib_ref]. The TPO protein maturation and trafficking require the assistance of thyrocyte endoplasmic reticulum chaperones: calreticulin, calnexin and BiP [bib_ref] Competition between calnexin and BiP in the endoplasmic reticulum can lead to..., Le Fourn [/bib_ref] [bib_ref] Calnexin and calreticulin binding to human thyroperoxidase is required for its first..., Fayadat [/bib_ref].
Several studies have reported increased levels of TPO antibodies in breast carcinoma patients [bib_ref] Thyroid autoimmunity occurs more frequently in women with breast cancer compared to..., Jiskra [/bib_ref] [bib_ref] Thyroid autoimmunity in patients with malignant and benign breast diseases before surgery, Giustarini [/bib_ref] [bib_ref] Serum thyroid peroxidase autoantibodies, thyroid volume, and outcome in breast carcinoma, Smyth [/bib_ref] [bib_ref] Favorable predictive value of thyroid autoimmunity in high aggressive breast cancer, Fiore [/bib_ref] [bib_ref] Breast cancer in association with thyroid disorders, Turken [/bib_ref]. Some authors suggested that patients with high levels of TPO-specific antibodies have a better prognosis [bib_ref] Serum thyroid peroxidase autoantibodies, thyroid volume, and outcome in breast carcinoma, Smyth [/bib_ref] [bib_ref] Favorable predictive value of thyroid autoimmunity in high aggressive breast cancer, Fiore [/bib_ref] [bib_ref] Thyroid function and survival following breast cancer, Brandt [/bib_ref] due to a decreased frequency of distant metastases [bib_ref] Anti-thyroid peroxidase antibodies are associated with the absence of distant metastases in..., Farahati [/bib_ref]. In yet another study, a high level of TPO-directed antibodies has been shown to be associated with a lower risk of breast cancer [bib_ref] Prospectively measured thyroid hormones and thyroid peroxidase antibodies in relation to risk..., Brandt [/bib_ref]. There are also studies in which no influence of TPO antibodies on breast cancer patients' outcome was observed [bib_ref] Thyroid autoimmunity occurs more frequently in women with breast cancer compared to..., Jiskra [/bib_ref]. Recent long-term study conducted on a largest up to date cohort of patients with moderate-to-high-risk early breast cancer did not found the TPOAbs presence as a predictive factor of the long-term recurrence and mortality [bib_ref] TPOAb and Thyroid Function Are Not Associated with Breast Cancer Outcome: Evidence..., Muller [/bib_ref]. The putative protective role of TPO autoantibodies in breast cancer could be explained by the fact that peritumoral and cancerous breast tissue express TPO [bib_ref] Does thyroid peroxidase provide an antigenic link between thyroid autoimmunity and breast..., Muller [/bib_ref] [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref]. Of note, breast TPO is immunologically similar to TPO expressed in the thyroid [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref]. The antigenic TPO activity depends on the proper structure of the TPO molecule. Therefore, in this study we aimed to characterize the biochemical properties of TPO expressed in human cancerous and normal mammary tissues. Additionally, we analyzed basic TPO characteristics in mammary-derived cell lines in order to determine whether these cells may be a suitable model for studies on breast TPO.
# Materials and methods
## Tissue samples
Primary breast tumor samples paired with the adjacent non-cancerous tissues were obtained from patients at different stages of breast cancer hospitalized at the Maria Skłodowska-Curie Memorial Cancer Center and Institute of Oncology (Warsaw, Poland) as described previously [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref]. Archived normal human lung and Graves' disease (G-B) thyroid tissue samples were used as controls. The study and all experimental procedures were approved by the Ethics Committee of Human Studies of the Center of Postgraduate Medical Education and at the Memorial Cancer Center and Institute of Oncology (Warsaw, Poland). Written informed consents were obtained from all patients.
## Cell lines
Chemically immortalized normal human breast epithelial cell line, 184A1, used as a noncancerous control, and breast cancer cell lines, MCF-7 and MDA-MB-231, were obtained from American Type Culture Collection (ATCC). SV40-immortalized human thyroid epithelial line, NTHY-ori 3-1 (NTHY), used as a positive control, was purchased from European Collection of Cell Cultures. Estrogen receptor (ER)-positive MCF-7 cells were maintained in Eagle's Minimum Essential Medium (EMEM; Sigma-Aldrich, Steinheim, Germany) supplemented with 0.01 mg/ml bovine insulin (Sigma-Aldrich) and 10% fetal bovine serum (v/v) (FBS; Roche, Mannheim, Germany). ER-negative MDA-MB-231 cells were cultured in Dulbecco's Modified Eagle Medium (DMEM; Life Technologies, Gibco, NY, USA) supplemented with 10% FBS (Roche, Mannheim, Germany). 184A1 cells were maintained in Mammary Epithelial Cell Growth Medium (MEGM; Lonza, Walkersville, USA) supplemented with 0.005 mg/ml transferrin (Sigma-Aldrich) and 1 ng/ml of cholera toxin (Sigma-Aldrich), while NTHY cells were grown in RPMI-1640 medium (Life Technologies) supplemented with 10% FBS (Roche, Mannheim, Germany). All cell lines were cultivated at 37˚C in humidified atmosphere with 5% CO 2 .
## Antibodies
A panel of four mouse anti-TPO monoclonal antibodies, i.e. mAb 47, mAb 15, mAb 18, mAb 64, were generously provided by Prof. J. Ruf from UMR-MD2, Aix-Marseille University, Marseille Medical School, Marseille, France and had been previously characterized [bib_ref] Relationship between immunological structure and biochemical properties of human thyroid peroxidase, Ruf [/bib_ref] [bib_ref] Localization of key amino acid residues in the dominant conformational epitopes on..., Godlewska [/bib_ref]. TPOspecific mAb A4 was kindly provided by Prof. J. P. Banga [bib_ref] Mapping epitope specificities of monoclonal antibodies to thyroid peroxidase using recombinant antigen..., Ewins [/bib_ref]. Rabbit antibodies to synthetic P14 peptide of TPO (aa 599-617) has been generated as already described [bib_ref] Human thyroid peroxidase: mapping of autoantibodies, conformational epitopes to the enzyme surface, Gardas [/bib_ref] [bib_ref] Identification of an immunodominant region recognized by human autoantibodies in a three-dimensional..., Hobby [/bib_ref] [bib_ref] Evaluation of conformational epitopes on thyroid peroxidase by antipeptide antibody binding and..., Gora [/bib_ref]. TPO-specific mAb ab76935 and β-actin-specific mAb (clone AC-15) were purchased from Abcam. Archival sera were used to prepare a pool of sera with high titers of TPO-specific antibodies (n = 5) and a pool of sera free of TPOAbs (n = 5). Rabbit polyclonal antibodies 10376-1-AP specific to human lactoperoxidase (LPO) were obtained from ProteinTech (Chicago, USA) and the murine isotype IgG1 control was purchased from Dako (Glostrup, Denmark).
## Total protein isolation, thyroid peroxidase immunoprecipitation and western blotting
To extract total proteins, the cells were washed with chilled PBS (pH 7.3) twice, then scraped into PBS and centrifuged (500 x g for 5 min at 4˚C). Next, the pelleted cells were resuspended in RIPA buffer (ThermoScientific, Rochester, USA) containing cOmplete Protease Inhibitor Cocktail (Roche). After incubation on ice for 30 minutes, the samples were centrifuged at 16000 x g for 30 minutes at 4˚C and the supernatants were collected. Protein extracts from human tissues were obtained as previously described [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref]. Protein concentration was determined by the BCA Protein Assay Kit (ThermoScientific). Then, the protein samples were aliquoted and stored at -70˚C. TPO was immunoprecipitated with rabbit antipeptide P14 serum from breast tissue lysates as previously described [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref].
Protein samples were resolved on an 8% SDS-PAGE gel under reducing conditions and electrotransferred to nitrocellulose membranes (Bio-Rad, Richmond, USA). After one-hour blocking in 5% non-fat milk in TBS with 0.1% Tween 20 (TBS-Tween 0.1%), the membranes were probed with the primary antibody at 1 μg/ml (ab76935 and anti-β-actin monoclonal antibodies) or 33 ng/ml (10376-1-AP antibody) in blocking buffer overnight at 4˚C. This was followed by incubation with a HRP-conjugated rabbit anti-mouse IgG (JacksonImmuno Research, West Grove, USA) diluted 1:20000-1:50000 in TBS-Tween 0.1% or with a HRP-conjugated goat anti-rabbit IgG (Dako) diluted 1:20000-1:100000 in the blocking buffer. The signals were then developed using SuperSignal West Pico Chemiluminescent Substrate and SuperSignal West Dura Extended Duration Substrate (ThermoScientific). Protein bands were analyzed using the GelAnalyser 2010a software (http://www.gelanalyzer.com). The anti-TPO ab76935 antibody was preabsorbed as described elsewhere [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref].
## Deglycosylation with peptide-n-glycosidase f (pngase f)
To analyze the asparagine-linked glycan contents in TPO, deglycosylation with PNGase F was performed. To this end, protein samples were denatured for 10 minutes in 0.5% SDS (w/v) and 40 mM dithiothreitol (DTT) at 95˚C. Afterwards, sodium phosphate (pH 7.5) and Nonidet P-40 were added to the final concentrations of 50 mM and 1% (v/v), respectively. Finally, the reaction mixture was supplemented with PNGase F (New England Biolabs, Hitchin, UK) and incubated at 37˚C for 16 hours. Control reactions were performed without PNGase F using the same assay conditions. Finally, the samples were analyzed by Western blotting.
## Two-dimensional electrophoresis (2-de)
Total protein extracts (100 μg) from human tissues or cultured cells in 2-DE rehydration buffer (7 M urea, 2 M thiourea, 4% CHAPS (w/v), 2 mM tributylphosphine (TBP), 0.4% BioLyte 3-10 (v/v), cOmplete Protease Inhibitor Cocktail (Roche)) were isoelectrofocused on 7 cm pH 3-10 IPG strips using PROTEAN IEF Cell (Bio-Rad), according to the manufacturer's instructions. Afterwards, IPG strips were incubated in the equilibration buffer (20% glycerol (v/v), 2% SDS, 6 M urea, and 50 mM Tris; pH 8.8) supplemented with 2% DTT (w/v) for 15 minutes, followed by the incubation with the fresh equilibration buffer supplemented with 2.5% iodoacetamide (w/v) for another 15 minutes. Next, the strips were washed in 25 mM Tris, 1.92 M glycine, 0.1% SDS (pH 8.3), and a standard Western blot analysis was performed. If not indicated otherwise, all equipment and reagents used were purchased from Bio-Rad.
## Detecting tpo enzymatic activity
The enzymatic activity of cell surface-expressed TPO in the studied cell lines was analyzed as we had previously described [bib_ref] A redundant role of human thyroid peroxidase propeptide for cellular, enzymatic, and..., Godlewska [/bib_ref]. Briefly, cells were grown 48 h in a medium supplemented with 20 mM hemin (Sigma-Aldrich). After washing with PBS, cells were incubated with reaction mixture containing Amplex Ultra Red (Life Technologies) in the presence of superoxide dismutase (SOD), KI, and H 2 O 2 (Sigma-Aldrich). Reaction was stopped at 1-minute intervals by addition of catalase and SOD. The fluorescence was measured in the Synergy H4 hybrid multi-mode microplate reader (BioTek, USA) [bib_ref] A redundant role of human thyroid peroxidase propeptide for cellular, enzymatic, and..., Godlewska [/bib_ref]. The enzymatic activity of TPO expressed in tissues was determined using a luminol-based assay described by Jomaa and collaborators [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref] [bib_ref] Simple and rapid in vitro assay for detecting human thyroid peroxidase disruption, Jomaa [/bib_ref] , with some modifications. Briefly, breast tissue lysates (200 μg) were immunoprecipitated with mAb A4 (6 μg) and protein A agarose (Merck Millipore, Darmstadt, Germany). 50 μl of resin-bound TPO in 0.1 M Tris-Cl (pH 8.6) was incubated with 150 μl of 1.3 M glycine-NaOH (pH 9.0), 1.3 mM EDTA in a 96-well plate for 5 minutes. The reaction was initiated by adding 20 μl of 400 μM luminol (Sigma-Aldrich) in 1 M glycine-NaOH (pH 9.0), 1 mM EDTA, followed by the supplementation with 5 μl of 80 mM H 2 O 2 . Luminescence intensities were measured in the Synergy 2 instrument (BioTek).
## Immunocytochemistry
Cells grown on uncoated glass coverslips for two days were washed with cold PBS (pH 7.3), fixed and permeabilized with methanol, then washed again and blocked in 2% goat serum (v/v) and 2% BSA (w/v) in PBS-0.1% Tween 20 (v/v) for one hour. This was followed an overnight incubation with the primary antibody at 4˚C. Monoclonal antibodies were used at the concentration of 14 μg/ml, while the pooled human sera were diluted at 1:10, and the concentration of 10376-1-AP Ab was 0.4 μg/ml. Following the washing, coverslips were incubated with the goat anti-mouse IgG DyLight 549-conjugate (JacksonImmuno Research), goat antihuman IgG fluorescein isothiocyanate-conjugate (Sigma-Aldrich) or goat anti-rabbit IgG Alexa 488-conjugate (Life Technologies). Then, the cells were washed, stained with 4 0 ,6-diamino-2-phenylindole (DAPI), mounted in Fluorescence Mounting Medium (Dako), and finally visualized using the LSM800 confocal microscope with the ZEN 2.1 software (Zeiss, Germany). The reactivity without primary antibody was determined as a negative control. To reduce non-specific binding, the concentration of Tween 20 was increased to 1% (v/v) in all solutions in the experiments with the human sera. If not indicated otherwise, the experimental steps were performed at room temperature.
## Immunohistochemistry
Immunohistochemical staining analysis was performed on 4-μm thick sections of formalinfixed, paraffin-embedded tissues. Dewaxed, deparaffinized and rehydrated sections were incubated with target retrieval solution (pH 9.0) (Dako) at 98˚C for 20 minutes. Following a 15-minute incubation with 0.3% hydrogen peroxide, tissue sections were then incubated with polyclonal anti-LPO antibody (10376-1-AP, 0.3 μg/ml) in the antibody diluent (Dako) for one hour in a humid chamber. This was followed by the incubation with REAL EnVision Detection System (Dako) for 30 minutes. Between all these steps, the sections were intensively washed in TBS-1% Tween 20. The reaction product was developed with 3,3'-diamino-benzidine (DAB; Dako) for 15 minutes. After counterstaining with hematoxylin, the sections were mounted on glass slides and examined under a light microscope (Olympus BX41, Japan). Two independent observers evaluated the immunohistochemical staining and scored it as negative or positive. All steps were performed at room temperature unless stated otherwise.
# Results
## Tpo protein maturation in human breast tissue samples
We analyzed whether breast TPO undergoes similar post-translational modifications as TPO expressed in the thyroid gland. We found that the molecular weight of detected breast tissueexpressed TPO was about 94 kDa. This is consistent with our previous findings [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref] and it is lower than the molecular weight of TPO in the thyroid gland (about 100 kDa). In order to further elucidate the reason for this difference, we removed all N-linked glycans and compared the shift in the migration of digested and non-digested TPO bands [fig_ref] Fig 1: The biochemical properties of TPO protein expressed in breast tissues [/fig_ref]. This revealed that the content of asparagine-linked carbohydrates in breast tissue TPO was lower (3 kDa) than in thyroid TPO (at least 10 kDa). Two-dimensional electrophoresis showed that the band containing breast tissue TPO consisted of two proteins with different isoelectric points (pIs; 6.4 ± 0.2 and 7.1 ± 0.1) which correspond to those previously described for the thyroid TPO [bib_ref] Purification of the human thyroid peroxidase and its identification as the microsomal..., Czarnocka [/bib_ref]. Moreover, we observed the peroxidase activity in crude lysates of breast tissues using the guaiacol oxidation-based method. To determine whether the detected enzymatic activity is specific for TPO, we performed the peroxidase activity measurement using immunoprecipitated TPO, luminol, and H 2 O 2 . Resins incubated with the A4 monoclonal antibody and and breast-derived cell lines (C). (A) N-linked glycan content in TPO expressed in breast tissues. Total cell extract was digested with PNGase F, then subjected to 8% SDS-PAGE, followed by Western blotting, and probing with TPOspecific mAb 47 monoclonal antibody. Controls were processed under the same conditions as the samples except that no enzyme was added. One representative immunoblot out of at least three independent experiments is shown. (B) Enzymatic activity of TPO expressed in breast tissues. Tissue lysate was incubated with TPOspecific mAb A4, then protein A agarose was added to precipitate immune complexes. TPO-antibody complexes bound to agarose were incubated with luminol in the presence of hydrogen peroxide. The intensity of luminescencent signal was measured and results were expressed as relative light units (RLU). As positive control, TPO immunoprecipitated from human thyroid tissue lysate (Graves' disease case) was used to measure luminol oxidation. Agarose A incubated with mAb A4 alone (lysate omitted) was used as negative control. One representative of three independent experiments is shown. protein lysate of human thyroid tissue from a Graves' disease case or resins incubated with the A4 monoclonal antibody alone were used as positive and negative controls, respectively. TPO immunoprecipitated from breast tissue displayed different levels of peroxidase activity [fig_ref] Fig 1: The biochemical properties of TPO protein expressed in breast tissues [/fig_ref]. Interestingly, no dimeric form of breast TPO was observed using the non-reducing SDS-PAGE method.
## Tpo protein maturation and immunological activity in breast-derived cell lines
Then, we analyzed the expression levels and the biochemical properties of TPO expressed in 184A1, MCF-7, and MDA-MB-231 cell lines using described immunological and biochemical methods [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref] [bib_ref] A redundant role of human thyroid peroxidase propeptide for cellular, enzymatic, and..., Godlewska [/bib_ref]. TPO expression was confirmed in all cell lines [fig_ref] Fig 1: The biochemical properties of TPO protein expressed in breast tissues [/fig_ref] left panel, and in S1 The levels of the TPO protein present in the analyzed cell lines were determined semiquantitatively with a standard curve obtained using highly purified hTPO. The average amount of TPO protein expressed in breast-derived cell lines was 6.0 ± 1.7 μg of TPO per 1 mg of crude lysate. The level of the TPO protein expression was stable over time [fig_ref] Fig 1: The biochemical properties of TPO protein expressed in breast tissues [/fig_ref]. We found that the TPO molecular weight was similar between the mammary-derived cell lines and the control NTHY cell line (98.0 ± 0.9 kDa). To confirm the specificity of the observed bands, anti-TPO antibody ab76935 was preabsorbed by highly purified hTPO. As shown in [fig_ref] Fig 1: The biochemical properties of TPO protein expressed in breast tissues [/fig_ref] , preabsorption completely inhibited the signal. We have also analyzed the isoelectric point of the TPO protein using two-dimensional electrophoresis. We found that the TPO band in the extracts obtained from mammary and thyroid gland-derived cells consisted of two proteins with different pIs [fig_ref] Fig 1: The biochemical properties of TPO protein expressed in breast tissues [/fig_ref] The pI values were as follows: 4.9 ± 0.1 and 7.2 ± 0.1 for 184A1 cells, 4.5 ± 0.6 and 7.0 ± 0.1 for MCF-7, and 5.6 ± 0.1 and 7.5 ± 0.1 for MDA-MB-231. Similar results were obtained for the NTHY samples (pI 4.7 ± 0.1 and 7.3 ± 0.3). The N-glycosylation level of TPO expressed in mammary-originated cell lines was comparable to that of TPO in NTHY. To find out whether TPO expressed in breast is enzymatically active, we incubated the cells with a fluorogenic substrate in the presence of hydrogen peroxide and the accumulated fluorescent reaction product was measured in real time. No reactivity was detected in any of the analyzed cell lines, including NTHY. We did not observe dimeric forms of breast TPO in SDS-PAGE under non-reducing conditions, either.
Subsequently, we evaluated the immunoreactivity of TPO expressed in the breast tissuederived cell lines using a panel of monoclonal mouse antibodies recognizing N-(mAb A4) and C-terminal (mAb 47, 15, 18, 64) domains of human thyroid peroxidase. The specificity of mAb and human TPOAb binding was verified using a murine isotype control and a pooled human TPOAb-free serum, respectively. The labeling of TPO present in the breast cancerderived cells, normal breast tissue-derived cells, and normal thyroid follicular cells is shown in No differences in TPO immunoreactivity were observed between the investigated cells in reaction with monoclonal antibodies . Similar TPO staining intensity was observed between different cell lines when pooled human TPOAb-positive serum was used .
## Lpo protein expression in breast-originated cells and human breast tissue samples
Since lactoperoxidase (LPO), another peroxidase from the TPO family, is also expressed in breast tissue, we performed immunochemistry analyses using cancer and normal samples and anti-LPO antibody (10376-1-AP) in order to verify whether co-expression of LPO and TPO did not influence the results obtained for TPO. We found that LPO was expressed in all the tested breast cancer samples [fig_ref] Fig 3: The expression of lactoperoxidase [/fig_ref]. However, the intensity of the staining differed, depending on the tissue. We did not detect any staining in peritumoral (normal) breast tissue [fig_ref] Fig 3: The expression of lactoperoxidase [/fig_ref]. This is consistent with the Western blot results which showed that LPO was expressed in breast cancer tissue and in the human lung tissue used as a positive control [fig_ref] Fig 3: The expression of lactoperoxidase [/fig_ref]. LPO expression was not observed in peritumoral breast tissue and thyroid tissue from the Graves' disease patients [fig_ref] Fig 3: The expression of lactoperoxidase [/fig_ref]. The molecular weight of the LPO band (111 kDa ± 2 kDa) was higher than that of hTPO (about 100 kDa).
We have also performed the immunofluorescent detection of LPO in breast-derived cell lines using the 10376-1-AP antibody [fig_ref] Fig 3: The expression of lactoperoxidase [/fig_ref]. In both breast cancer cell lines, MCF-7 and MDA-MB-231, the LPO protein was found to be localized in the cytoplasm and in the nucleolus. However, no signals were detected in normal breast cells (184A1). No labeling was observed with rabbit IgG which was used as a negative control. Western blot results revealed bands corresponding to the LPO protein in MCF-7 and MDA-MB-231 cells, while no bands were observed for the 184A1 cells [fig_ref] Fig 3: The expression of lactoperoxidase [/fig_ref].
# Discussion
We have previously demonstrated that thyroid peroxidase (TPO) expressed in human mammary tissue is immunologically active [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref]. In this study, we aimed to further determine the biochemical characteristics of breast TPO. We have shown that mammary TPO, like thyroid TPO, is N-glycosylated. However, the amount of carbohydrate residues attached to the breast TPO protein core is lower than in the thyroid TPO molecule. It had been already demonstrated that N-linked oligosaccharides are important in the trafficking of TPO molecule to the
## Fig 2. representative tpo immunostaining results obtained with a panel of monoclonal antibodies (mabs) against human thyroid peroxidase (tpo) (a) and human serum pools (b) in breast epithelial normal (184a1) and cancer cell lines (mcf-7 and mda-mb-231).
A normal human thyroid cell line, NTHY, was used as a positive control. (A) Positive signal (red) was detected with all mAbs except the isotype control (negative control). In the cells incubated with TPOAbs obtained from autoimmune thyroid disease (AITD) patients (TPOAbs(+)), a positive signal (green) was observed but this staining was not observed when TPOAb-free serum (TPOAbs(-)) was used (negative control). Nuclei (blue) were counterstained with DAPI. Magnification: 630×. IIAb: secondary antibody; DAPI: 4 0 ,6-diamino-2-phenylindole; NTHY: NTHY-ori 3-1 cell line. cell surface and in its enzymatic activity. The immunoreactivity with TPO-specific monoclonal and polyclonal antibodies, however, was not markedly affected by a removal of asparaginelinked residues from the enzyme, indicating that carbohydrate residues seem to be rather a maturation signal for TPO as a representative of transmembrane glycoprotein [bib_ref] Human thyroperoxidase is largely retained and rapidly degraded in the endoplasmic reticulum...., Fayadat [/bib_ref] [bib_ref] Effects of deglycosylation of human thyroperoxidase on its enzymatic activity and immunoreactivity, Giraud [/bib_ref] [bib_ref] Analysis of carbohydrate residues on human thyroid peroxidase (TPO) and thyroglobulin (Tg)..., Kiso [/bib_ref]. To check whether an altered protein glycosylation pattern affects mammary TPO functionality, we conducted peroxidase activity assays. The results suggest that at least a part of breast TPO molecules are enzymatically active. Therefore, it could be hypothesized that breast TPO may participate in the regulation of oxidative stress in breast tissue. It has already been shown that LPO, another enzyme from the peroxidase family which is expressed in breast tissue can oxidize natural and synthetic estrogen to catechol estrogens which react with DNA. The mutagenic adducts that are formed during this process can initiate and promote breast cancer development [bib_ref] A kinetic study on the lactoperoxidase catalyzed oxidation of estrogens, Løvstad [/bib_ref] [bib_ref] Lactoperoxidase-catalyzed activation of carcinogenic aromatic and heterocyclic amines, Gorlewska-Roberts [/bib_ref] [bib_ref] The metabolism of 17 beta-estradiol by lactoperoxidase: a possible source of oxidative..., Sipe [/bib_ref]. Furthermore, it has been reported that supplementation with iodine and iodide mixture inhibits the generation of DNA adducts and exerts apoptotic effect on cancerous cells in the rat dimethylbenz[a]anthracene-induced tumor model [bib_ref] Antineoplastic effect of iodine and iodide in dimethylbenz[a]anthracene-induced mammary tumors: association between..., Soriano [/bib_ref]. These effects were attributed to LPO. However, given the results of our current study, the enzymatically active breast TPO might also be considered as a player involved in these processes. The molecular mass of the glycan-deprived breast TPO band is slightly lower than that found for the thyroid TPO. Since in normal and diseased thyroid glands as well as in normal and cancerous breast tissue shorter forms of TPO are expressed, the difference between molecular weights of thyroid-and breast-expressed TPO molecule might be, at least partly, explained by a lack of one exon or several exons [bib_ref] Does thyroid peroxidase provide an antigenic link between thyroid autoimmunity and breast..., Muller [/bib_ref] [bib_ref] Increasing diversity of human thyroperoxidase generated by alternative splicing. Characterized by molecular..., Ferrand [/bib_ref] [bib_ref] Differential expression of thyroperoxidase mRNA splice variants in human thyroid tumors, Le Fourn [/bib_ref] [bib_ref] Expression of tpo mRNA in thyroid tumors: quantitative PCR analysis and correlation..., Cristofaro [/bib_ref]. Nevertheless, splicing out some of the exons does not always lead to significant changes in the thyroid peroxidase structure and function. For example, an isoform abundantly expressed in normal thyroid tissue, TPO4, lacking exon 14, is not only able to acquire a proper three-dimensional structure and enzymatic activity, but also reaches the cellular surface [bib_ref] Increasing diversity of human thyroperoxidase generated by alternative splicing. Characterized by molecular..., Ferrand [/bib_ref]. Therefore, it seems that even if some exon(s) are missing in the breast TPO, its functionality does not seem to be significantly different from that of the full-length thyroid TPO.
Since cell lines originating from human tissues are a good tool for studying and characterizing proteins, we examined whether the breast-derived cell lines can be used as a model to study breast TPO. Our results showed that in normal and breast cancer-derived cell lines (including both the non-metastatic MCF-7 cell line that retains some of differentiated epithelial features and the aggressive triple negative MDA-MB-231 cell line) the TPO protein levels were similar to those detected in the normal immortalized thyrocyte culture. Moreover, the TPO protein expression levels were stable in time. There were no significant differences in the molecular weight, the isoelectric point (pI) value and the N-glycan content between the TPO isolated from the cell lines derived from cancerous and non-cancerous breast tissue and the TPO isolated from normal thyrocytes, either. Altogether, our results suggest that TPO molecules expressed in the thyroid and in the breast have similar biochemical properties. It is also worth noting that all antibodies used in the analysis of antigenic TPO activity have bound the TPO protein in all the tested cell lines with a similar intensity and the obtained immunocytochemical staining was shown to predominantly localize to the cytoplasmic compartment. The fact that antibodies well known to specifically recognize thyroid-expressed TPO bind to the breast TPO surface indicates that the two forms of TPO have a similar conformation. This suggests that they should also have similar functions as the TPO enzymatic activity depends on the integrity of its three-dimensional structure. However, although the conformational structure of the breast TPO seems to be very similar to that of the thyroid TPO, we did not detect any enzymatic activity of the protein in the intact cells. This may be due to the fact that the fluorogenic substrate incubated with the cells was membrane-impermeable and as a consequence only the enzymatic activity of TPO localized and exposed on the cell surface could be detected. The lack of signal for the TPO enzymatic activity is probably due to the fact that the amounts of TPO bound to cell membranes were below the detection threshold of the method. Indeed, using the membrane proteins biotinylation method (non-membrane permeable sulfo-NHS-SSbiotin) which, as we had previously described, significantly enriches this fraction [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref] , we detected a faint TPO band with anti-TPO antibodies in all the tested cell lines, which indicates the occurrence of low levels of cell membrane-located peroxidase. Similar observations were reported in a study conducted on the CHO cells stably transfected with human TPO cDNA, which showed that this enzyme was predominantly detected intracellularly and only a small part reached the cell surface [bib_ref] Human thyroperoxidase is largely retained and rapidly degraded in the endoplasmic reticulum...., Fayadat [/bib_ref]. Furthermore, supplementing the culture medium with the heme precursor (20 mM hemin) which had been shown to increase the enzymatic activity of cell membrane-bound TPO [bib_ref] Role of heme in intracellular trafficking of thyroperoxidase and involvement of H2O2..., Fayadat [/bib_ref] , did not allow to detect higher levels of the protein.
Thyroid peroxidase has a high sequence homology with human LPO (hLPO), an enzyme expressed in lactating breast tissue as well as in salivary, lacrimal, tracheal and bronchial glands [bib_ref] Molecular heterogeneity and alternative splicing of human lactoperoxidase, Fragoso [/bib_ref] [bib_ref] Lactoperoxidase: structural insights into the function,ligand binding and inhibition, Sharma [/bib_ref]. Therefore, we considered the potential influence of LPO detection on the obtained results. We have already shown that all TPO-specific antibodies used in the present study do not cross-react with LPO [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref]. Here we show the LPO protein expression in breast cancer cell lines and breast tissue samples, while the peritumoral sections and the normal breast cell line were not positive for LPO. Low or undetectable levels of LPO in the normal mammary gland and increased expression levels in breast cancer tissue has been already reported by others [bib_ref] Antineoplastic effect of iodine and iodide in dimethylbenz[a]anthracene-induced mammary tumors: association between..., Soriano [/bib_ref] [bib_ref] Uptake and gene expression with antitumoral doses of iodine in thyroid and..., Anguiano [/bib_ref] [bib_ref] Signaling pathways involved in the antiproliferative effect of molecular iodine in normal..., Arroyo-Helguera [/bib_ref]. Furthermore, we found that the molecular weight of human LPO was higher than that observed for breast TPO (approximately 111 kDa versus 100 kDa). The reported hLPO molecular weight measured in human milk is about 80 kDa for the completely proteolytically processed protein, while 90 kDa band corresponds to partly processed molecules [bib_ref] Identification of lactoperoxidase in mature human milk, Shin [/bib_ref]. Therefore, the hLPO band that we detected in crude protein lysates isolated from breast tissues and breastderived cell lines may represent the beginning of the post-translational maturation of hLPO.
In conclusion, we found that the biochemical properties of the TPO protein expressed in cancer and normal breast tissue are similar to those observed for the thyroid TPO. Nevertheless, there are some differences in comparison with the thyroid-tissue expressed protein, such as a lower N-glycan content, a slightly smaller polypeptide length, a decreased enzymatic activity, and undetectable dimer formation. We also show that biochemical characteristics of the TPO protein expressed in normal (184A1) and cancerous cell lines (MCF-7 and MDA-MB-231), independently of the cancer differentiation status, resemble those of TPO present in immortalized normal thyrocytes (NTHY cells). Moreover, we found that the peroxidase activity was below the detection threshold of the used methods as a result of a low protein delivery to the cell surface. The differences in post-translational modifications between breast and thyroid TPO are rather minor, which may be explained by a retained antigenicity of breastexpressed TPO shown here for the mammary-derived cell lines and previously demonstrated for TPO expressed in breast tissue [bib_ref] Thyroid peroxidase (TPO) expressed in thyroid and breast tissues shows similar antigenic..., Godlewska [/bib_ref]. The interaction with TPO-specific antibodies which bind to conformational epitopes strongly depends on intact tree-dimensional structure of the thyroid peroxidase. Our study shows that TPO expressed in breast tissue not only shares similarities with thyroid TPO in biochemical properties (with some minor differences), but also immunodominant regions A and B of breast TPO are specifically recognized by TPOAbs. Therefore, the similarities in biochemical and antigenic properties of breast and thyroid TPO may partially explain the protective role of TPO autoantibodies in breast cancer patients and help further elucidate its mechanism. Future studies should investigate the potential immunotherapies directed against breast-expressed TPO and its specific epitopes.
## Supporting information
[fig] Fig 1: The biochemical properties of TPO protein expressed in breast tissues ( [/fig]
[fig] Fig 3: The expression of lactoperoxidase (LPO) in breast tissues (A and B) and cell lines (C and D) derived from normal (184A1) and cancerous mammary tissues (MCF-7 and MDA-MB-231). (A) Representative immunohistochemical staining of human LPO in breast cancer (upper panel) and peritumoral tissues (lower panel) (magnification: 100×). (B) Human LPO expression in breast cancer and peritumoral tissues detected by immunoblotting with the 10376-1-AP antibody. Human lung and thyroid tissue (Graves' disease case) lysates were used as a negative and positive control, respectively. For each lane, 50 μg of crude protein extract were loaded. A β-actin-specific antibody was used as a loading control. (C) LPO expression in cell lines as shown by immunofluorescent staining. Positive immunofluorescent signal (red) was detected in MDA-MB-231 and, to a lesser extent, in MCF-7 cells. No staining was detected when pre-immune rabbit IgG was used (insets). Nuclei were counterstained with DAPI (blue). Magnification: 630×. (D) LPO protein expression in cell lines analyzed by Western blot. LPO was detected in breast cancer cells, while no band was observed in normal 184A1 cells. For each lane, 10 μg of crude protein extract were loaded. A β-actin-specific antibody was used as a loading control. [/fig]
[fig] S1: Fig. Expression of the TPO protein in breast-derived cell lines. (A) TPO protein expression levels at the indicated time points and (B) its isoelectric point in the analyzed cell lines (B). Normal human thyroid cell line, NTHY, was used as a positive control. (A) TPO was detected with the ab76935 antibody using Western blotting. 20 μg of total protein lysate were loaded on an 8% SDS-polyacrylamide gel. A β-actin-specific antibody was used as a loading control. (B) 100 μg of total protein lysate was subjected to a two-dimensional (2-DE) electrophoresis. TPO was detected with the ab76935 antibody. NTHY: NTHY-ori 3-1 cell line; pI: isoelectric point. (TIF) [/fig]
|
Socioeconomic Deprivation, Sleep Duration, and Mental Health during the First Year of the COVID-19 Pandemic
# Introduction
The coronavirus disease 2019 (COVID-19) pandemic has underscored health disparities in the United States. For example, neighborhoods with lower median family income and higher unemployment rates have higher rates of COVID-19 transmission, hospitalization, and death [bib_ref] Racial, Economic, and Health Inequality and COVID-19 Infection in the United States, Abedi [/bib_ref] [bib_ref] A comprehensive analysis of COVID-19 transmission and mortality rates at the county..., Bhowmik [/bib_ref] [bib_ref] Predictors of COVID-19-confirmed cases and fatalities in 883 US counties with a..., Robertson [/bib_ref]. There is an established relationship between poor health outcomes and socioeconomic deprivation, often assessed via the social deprivation index (SDI), a composite measure of the effects of poverty and deprivation based on income, education, employment, housing, household characteristics, transportation, and demographics [bib_ref] Measures of social deprivation that predict health care access and need within..., Butler [/bib_ref]. The SDI permits the assessment of the individual components to understand the nuanced population effects or socio-economic variability in health outcomes of social determinants of health. The COVID-19 pandemic upended the delicate balance between sleep and mental health [bib_ref] Sleep quality and mental health in coronavirus disease 2019 patients and general..., Alshumrani [/bib_ref]. Evaluating the role of socioeconomic deprivation in sleep and mental health is important to (1) highlight inequality in highly overlooked domains of health behaviors, [bib_ref] A comprehensive analysis of COVID-19 transmission and mortality rates at the county..., Bhowmik [/bib_ref] understand how sleep and mental health are influenced, and (3) target higher-risk groups and areas for future interventions.
Social determinants of health are the social and economic conditions that impact a vast array of health and quality of life outcomes, including sleep and mental health. Social determinants of health are classified into five main areas including healthcare access and quality, education, social and community context, neighborhood and built environment, and economic stability. COVID-19 outcomes have been profoundly impacted by social 2 of 10 determinants of health. For example, homelessness is linked to increased risk of COVID-19 transmission due to crowded living spaces and decreased access to testing facilities [bib_ref] COVID-19: A potential public health problem for homeless populations, Tsai [/bib_ref]. The effects of social determinants of health are far-reaching and have been magnified by the COVID-19 pandemic. Understanding the socioeconomic deprivation within the context of the COVID-19 pandemic can provide important insights into sleep and mental health outcomes.
The National Sleep Foundation recommends that adults sleep 7-9 h per night, and about a third of Americans report insufficient sleep (defined as <7 h) [bib_ref] Epidemiology of insufficient sleep and poor sleep quality, Grandner [/bib_ref]. Sleep of insufficient quality or quantity is associated with higher body mass index (BMI), impaired body weight regulation (lower leptin and higher ghrelin levels), poorer insulin sensitivity, and lower daytime alertness and has been linked to multiple chronic conditions including overweight/obesity, type 2 diabetes, cardiovascular disease, stroke, dementia, and cancer [bib_ref] Biomedical risk factors for decreased cognitive functioning in type 1 diabetes: An..., Jacobson [/bib_ref] [bib_ref] Associations between sleep loss and increased risk of obesity and diabetes, Knutson [/bib_ref] [bib_ref] Sleep duration and body mass index in a rural population, Kohatsu [/bib_ref] [bib_ref] Role of sleep and sleep loss in hormonal release and metabolism, Leproult [/bib_ref] [bib_ref] Impact of sleep and sleep loss on neuroendocrine and metabolic function, Van Cauter [/bib_ref]. During the COVID-19 pandemic years 2019 to 2021, sleep disturbances affected approximately 41% with higher rates during the lockdown compared to no lockdown, 42.49% versus 37.97% in a recent systematic review and meta-analysis of 250 studies (493,475 participants) across 49 countries [bib_ref] Sleep disturbances during the COVID-19 pandemic: A systematic review, meta-analysis, and meta-regression, Jahrami [/bib_ref]. However, there are few studies where the impact of the COVID-19 pandemic on sleep sufficiency has been examined. The COVID-19 impact has varied among populations with individuals reporting adequate sleep duration pre-pandemic and shorter sleep duration and increased sleep complaints during the pandemic [bib_ref] Sleep quality during the COVID-19 pandemic: Not one size fits all, Kocevska [/bib_ref] [bib_ref] Impact of COVID-19 lockdown on sleep-wake schedule and associated lifestyle related behavior:..., Sinha [/bib_ref] [bib_ref] Assessment of Subjective Sleep Problems in Men with Opioid Dependence Maintained on..., Tripathi [/bib_ref]. The effect of sleep behavior on health outcomes has increasingly been studied with an emerging literature on social determinants of sleep [bib_ref] Epidemiology of insufficient sleep and poor sleep quality, Grandner [/bib_ref] [bib_ref] Who gets the best sleep? Ethnic and socioeconomic factors related to sleep..., Grandner [/bib_ref] [bib_ref] Sleep Duration and Diabetes Risk: Population Trends and Potential Mechanisms, Grandner [/bib_ref] [bib_ref] Social and behavioral predictors of insufficient sleep among African Americans and Caucasians, Williams [/bib_ref]. For instance, lower socioeconomic status (individual level) and socioeconomic deprivation (area level) are associated with short or long sleep duration along with sleep disorders (e.g., sleep apnea) which all relate to increased mortality risk [bib_ref] From sleep duration to mortality: Implications of meta-analysis and future directions, Grandner [/bib_ref] [bib_ref] Mortality associated with short sleep duration: The evidence, the possible mechanisms, and..., Grandner [/bib_ref] [bib_ref] Sleep duration in the United States: A cross-sectional population-based study, Krueger [/bib_ref] [bib_ref] Neighborhood socioeconomic deprivation and mortality: NIH-AARP diet and health study, Major [/bib_ref] [bib_ref] The association between area-based deprivation and change in body-mass index over time..., Twaits [/bib_ref]. The COVID-19 pandemic has also magnified preexisting sleep disparities across individuals [bib_ref] Sleep quality and mental health in coronavirus disease 2019 patients and general..., Alshumrani [/bib_ref] [bib_ref] The Magnification of Health Disparities during the COVID-19 Pandemic, Willems [/bib_ref] [bib_ref] Feeling Anxious about Catching COVID": Facilitators and Barriers of Sleep Health among..., Griggs [/bib_ref] [bib_ref] Effects of COVID-19 lockdown on sleep duration, sleep quality and burnout in..., Arrona-Palacios [/bib_ref] [bib_ref] A National Survey of U.S. Adolescent Sleep Duration, Timing, and Social Jetlag..., Wesley [/bib_ref] [bib_ref] Circadian misalignment is associated with Covid-19 infection, Coelho [/bib_ref]. However, less is known about the association between neighborhood socioeconomic characteristics and the nuance of urban-rural classifications on sleep. Further, compared to those with sufficient sleep, individuals with insufficient sleep have nearly threefold increased odds of mental distress [bib_ref] Effect of Inadequate Sleep on Frequent Mental Distress, Blackwelder [/bib_ref].
There is increasing concern about the impact of the pandemic on mental health [bib_ref] The psychological burden of the COVID-19 pandemic and associated lockdown measures: Experience..., Burhamah [/bib_ref] [bib_ref] Longitudinal evaluation of the psychological impact of the COVID-19 crisis in Spain, Planchuelo-Gómez [/bib_ref]. Those with limited economic and social resources are at a higher risk for these impacts [bib_ref] Monthly trends in self-reported health status and depression by race/ethnicity and socioeconomic..., Lee [/bib_ref]. There were considerable increases in anxiety, depressive disorders, and suicidal ideation in mid-March-February 2021 (during the pandemic) compared to the same period in 2019 (pre-pandemic) in the United States. In another national United States survey of 790,633 individuals, there was a significant increase in the percentage of adults with anxiety or depressive disorder symptoms (36.4% to 41.5%) during the past 7 days, as well as an increased need for mental health counseling or therapy during the past 4 weeks (from 9.2% to 11.7%) [bib_ref] Mental Health Treatment among Adults Aged 18-44: United States, Terlizzi [/bib_ref] [bib_ref] Symptoms of anxiety or depressive disorder and use of mental health care..., Vahratian [/bib_ref]. The higher need for mental health counseling persisted after the height of the COVID-19 pandemic in the United States with an increase in the percentage of adults needing treatment within the past 12 months (19.2% to 21.6%) [bib_ref] Mental Health Treatment among Adults Aged 18-44: United States, Terlizzi [/bib_ref] [bib_ref] Symptoms of anxiety or depressive disorder and use of mental health care..., Vahratian [/bib_ref].
The purpose of this study was to quantify levels of socioeconomic deprivation and examine associations among socioeconomic deprivation, sleep duration, and mental health while adjusting for covariates (e.g., age, sex at birth, racial identity, education, body mass index) during the first year of the COVID-19 pandemic. Specifically, through a crosssectional investigation of Ohio residents from the Behavioral Risk Factor Surveillance System (BRFSS) in 2020, we tested the following hypotheses: (1) higher socioeconomic deprivation would be associated with shorter sleep and poorer mental health and (2) those in the highest socioeconomically deprived areas would have higher odds of short sleep (defined as <6 h per night) and poor mental health (defined as >14 days of poor mental health per month).
# Materials and methods
## Data sources and study population
The Ohio BRFSS 2020 data were used for this analysis. The BRFSS is a random-digit dialed state-based telephone survey about health practices, conditions, and risk among non-institutionalized adults ages 18 years and older conducted by the Ohio Department of Health and supported by the Centers for Disease Control and Prevention (CDC). Data are collected from a random probability sample representative of the 14 geographic regions in Ohio. The corresponding author's institution, Case Western Reserve University approved the study (IRB # STUDY20221217). Case Western Reserve University has a federal-wide assurance.
## Variables and measures
Individual demographic variables of interest included age, sex at birth, race/ethnicity, and education. Covariates were selected based on a priori knowledge and included age, biological sex, and race/ethnicity, education, and body mass index. An area-level variable the Social Deprivation Index (SDI) was generated from the zip codes. The SDI is a composite measure of six domains collected in the American Community Survey: income (percent living in poverty), education (percent with less than 12 years of education), housing (percent living in rented housing unit and percent living in overcrowded housing unit), household characteristics (percent single-parent households with dependents <18 years), transportation (percent of households without a car), and employment (percent non-employed adults under 65 years of age) [bib_ref] Measures of social deprivation that predict health care access and need within..., Butler [/bib_ref]. The SDI is an estimate of health care access and health outcomes within a rational primary care service area based on zip code tabulation (ZCTA). The SDI was updated with data from 5-year estimates from 2011-2015 [bib_ref] Measures of social deprivation that predict health care access and need within..., Butler [/bib_ref]. We used ZCTA's in this study based on generalized U.S. Postal Service Zip Codes [bib_ref] Measures of social deprivation that predict health care access and need within..., Butler [/bib_ref].
Self-reported sleep duration was a single-item measure. Respondents were asked "on average, how many hours of sleep do you get in a 24 h period?". Responses range from 1-24 h. We excluded implausible values for sleep duration (>18 h). The significance was not altered with the removal of implausible values. We used hours in the linear regression models. For the logistic regression models, we defined short sleep as ≤6 h consistent with other studies on sleep duration and the National Sleep Foundation Recommendation [bib_ref] Objective short sleep duration increases the risk of all-cause mortality associated with..., Fernandez-Mendoza [/bib_ref].
Mental health was derived from a single item measure. Respondents were asked "Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good?". We used poor mental health days from 0 to 30 in the linear regression models. For logistic regression, poor mental health was defined as 14 or more days of poor mental health in the past 30 days. This approach for defining poor mental health days has been documented in previous research [bib_ref] Mental health burden in a national sample of American Indian and Alaska..., Asdigian [/bib_ref].
# Statistical analysis
A quantitative descriptive approach was used to characterize socioeconomic deprivation, sleep duration, and mental health among the individuals in the study. Bivariate correlations, multivariable linear, and multivariable logistic regression models were used to examine the relationships among socioeconomic deprivation, sleep duration, and mental health.
To evaluate explanatory contributions of socioeconomic deprivation to sleep duration and mental health, we performed a series of multivariable linear regression models. To assess exposure in the logistic regression model, overall SDI was dichotomized into two groups (10% most socially deprived and the remaining 90% as a comparator group). This approach for determining risk in those most socially deprived, the highest 10% of SDI scores, has been documented in previous research [bib_ref] The Association Between Neighborhood Social Vulnerability and Cardiovascular Health Risk Among Black/African..., Basile Ibrahim [/bib_ref]. We defined short sleep as ≤6 h and poor mental health as 14 days or more in the past 30 days consistent with prior research and national recommendations [bib_ref] Objective short sleep duration increases the risk of all-cause mortality associated with..., Fernandez-Mendoza [/bib_ref] [bib_ref] Mental health burden in a national sample of American Indian and Alaska..., Asdigian [/bib_ref]. Statistical significance was set at p < 0.05.
# Results
## Sample characteristics
A population of 14,676 individuals aged 18-99 (mean age 54.9 ± 18.1 years) residing across 1101 ZCTA's were included in the current study. A little over half (54.3%) identified as female and the majority identified as Non-Hispanic White (86.9%), followed by 6.8% Non-Hispanic Black, 4.2% other race, and 2.1% Hispanic. Chronic conditions reported included asthma (13.9%), diabetes (14.9%), heart disease (6.6%), cancer (8.4%), chronic obstructive pulmonary disease (1.92%), and arthritis (1.67%). There was substantial variation in sleep duration and mental health across all ZCTA's. The prevalence of short sleep (<6 h) was 11.9% and insufficient sleep (<7 h) was 34.7% which is comparable to other state data from BRFSS in previous years (2009) [bib_ref] Social and behavioral predictors of insufficient sleep among African Americans and Caucasians, Williams [/bib_ref]. The mean sleep duration was 6.98 ± 1.50 h. Prevalence of poor mental health (>14 days) was 14.5% with a mean of 4.5 ± 8.7. We present demographic and clinical characteristics for the overall sample in [fig_ref] Table 1: Demographic Characteristics of the Sample [/fig_ref].
## Socioeconomic deprivation and sleep duration
We found a higher prevalence of short sleep in areas with higher social deprivation indices [fig_ref] Figure 1: Short sleep and poor mental health prevalence % by social deprivation index... [/fig_ref]. We examined the unadjusted association between SDI and sleep duration in the first set of linear regression models. The unadjusted association between SDI and sleep duration was statistically significant. Specifically, higher socioeconomic deprivation was associated with shorter sleep duration (β = −0.049, p < 0.001). We present the unstandardized coefficient regression coefficient (B), standard error (SE), standardized regression coefficient (β), and coefficient of determination (R 2 ) for each model in [fig_ref] Table 2: Socioeconomic deprivation and covariates to sleep health and mental health outcomes [/fig_ref].
indices [fig_ref] Figure 1: Short sleep and poor mental health prevalence % by social deprivation index... [/fig_ref]. We examined the unadjusted association between SDI and sleep duration in the first set of linear regression models. The unadjusted association between SDI and sleep duration was statistically significant. Specifically, higher socioeconomic deprivation was associated with shorter sleep duration (β = −0.049, p < 0.001). We present the unstandardized coefficient regression coefficient (B), standard error (SE), standardized regression coefficient (β), and coefficient of determination (R 2 ) for each model in [fig_ref] Table 2: Socioeconomic deprivation and covariates to sleep health and mental health outcomes [/fig_ref]. In the second set of models, we examined socioeconomic deprivation and sleep duration after adjusting for covariates (age, sex at birth, racial identity, and education) using multivariable linear regression. The associations between SDI and sleep duration remained statistically significant after adjusting for covariates (p < 0.001) [fig_ref] Table 2: Socioeconomic deprivation and covariates to sleep health and mental health outcomes [/fig_ref].
In the unadjusted logistic regression models, individuals living in the most socially deprived areas had 1.6 times higher odds of short sleep (<6 h) (OR = 1.54, 95% confidence interval [1.34, 1.83]) than those living in less socially deprived areas [fig_ref] Table 3: Socioeconomic deprivation, sleep health, and mental health [/fig_ref]. These associations remained statistically significant after adjusting for covariates (short sleep: aOR = 1.33, 95% confidence interval [1.12, 1.57] [fig_ref] Table 3: Socioeconomic deprivation, sleep health, and mental health [/fig_ref]. In the second set of models, we examined socioeconomic deprivation and sleep duration after adjusting for covariates (age, sex at birth, racial identity, and education) using multivariable linear regression. The associations between SDI and sleep duration remained statistically significant after adjusting for covariates (p < 0.001) [fig_ref] Table 2: Socioeconomic deprivation and covariates to sleep health and mental health outcomes [/fig_ref].
In the unadjusted logistic regression models, individuals living in the most socially deprived areas had 1.6 times higher odds of short sleep (<6 h) (OR = 1.54, 95% confidence interval [1.34, 1.83]) than those living in less socially deprived areas [fig_ref] Table 3: Socioeconomic deprivation, sleep health, and mental health [/fig_ref]. These associations remained statistically significant after adjusting for covariates (short sleep: aOR = 1.33, 95% confidence interval [fig_ref] Table 3: Socioeconomic deprivation, sleep health, and mental health [/fig_ref].
## Socioeconomic deprivation and mental health
We found a higher prevalence of poor mental health in areas with higher social deprivation indices. We examined the unadjusted association between SDI mental health in the first set of linear regression models. The unadjusted association between SDI and mental health was statistically significant (p < 0.001) [fig_ref] Table 2: Socioeconomic deprivation and covariates to sleep health and mental health outcomes [/fig_ref]. Higher socioeconomic deprivation was associated with poorer mental health (β = 0.098, p < 0.001).
In the second set of models, we examined socioeconomic deprivation and mental health after adjusting for covariates (age, sex at birth, racial identity, and education) using multivariable linear regression. The associations between SDI and mental health remained statistically significant after adjusting for covariates (p < 0.001) [fig_ref] Table 2: Socioeconomic deprivation and covariates to sleep health and mental health outcomes [/fig_ref].
In the unadjusted logistic regression models, individuals living in the most socially deprived areas had 1.5 times higher odds of poor mental health (>14 days) (OR = 1.47, 95% confidence interval [1.26, 1.71] than those living in less socially deprived areas [fig_ref] Table 3: Socioeconomic deprivation, sleep health, and mental health [/fig_ref]. These associations remained statistically significant after adjusting for covariates (poor mental health: aOR = 1.18, 95% confidence interval [1.01, 1.39]).
# Discussion
We investigated the associations among socioeconomic deprivation, sleep duration, and mental health status in a large representative sample of Ohio residents during the first year of COVID-19 pandemic. Individuals from the highest socioeconomically deprived areas had an almost 2-fold increase in odds of short sleep and poor mental health after adjusting for covariates (age, sex at birth, racial identity, education, and body mass index). These findings supplement the growing body of evidence on the longstanding sleep and mental health disparities that disproportionately affect those from socioeconomically disadvantaged areas. Sleep and mental health are dynamically interrelated and are essential to socioemotional development, physical health maintenance, and economic advancement. Thus, our findings highlight additional deleterious effects of structural inequities on individual-level outcomes and the disproportionate impact of COVID-19 on disadvantaged areas.
Close to one-third of the individuals in the study reported insufficient sleep (<7 h) and one-quarter short sleep (<6 h). This finding was consistent with other population-based studies of American adults sleeping less than 7-8 h per night [bib_ref] Associations between inadequate sleep and obesity in the US adult population: Analysis..., Jean-Louis [/bib_ref] [bib_ref] Trends in the prevalence of short sleepers in the USA, Knutson [/bib_ref]. Although sleep time remained unchanged from pre-pandemic, increases in short sleepers have been observed in other studies [bib_ref] Time to Sleep?-A Review of the Impact of the COVID-19 Pandemic on..., Neculicioiu [/bib_ref] [bib_ref] Sleep characteristics of U.S. adults before and during the COVID-19 pandemic, Hisler [/bib_ref]. In an international online survey study (ECLB COVID-19), there was a modest increase in self-reported sleep duration as well as increases in poor sleep quality as measured by the Pittsburgh Sleep Quality Index (4.37 ± 2.71 before the lockdown and 5.32 ± 3.23 after the lockdown) [bib_ref] Globally altered sleep patterns and physical activity levels by confinement in 5056..., Trabelsi [/bib_ref]. The public health ramifications of insufficient sleep are under-recognized by society with a conservatively estimated economic burden of USD 107 billion. Higher socioeconomic deprivation was associated with shorter sleep in the present study which is consistent with several studies where individual socioeconomic status was associated with shorter sleep duration or poorer sleep quality [bib_ref] Who has time to sleep?, Hale [/bib_ref] [bib_ref] Objectively measured sleep characteristics among early-middle-aged adults: The CARDIA study, Lauderdale [/bib_ref] [bib_ref] Socioeconomic status predicts objective and subjective sleep quality in aging women, Friedman [/bib_ref] [bib_ref] Socioeconomic position and sleep quantity in UK adults, Adams [/bib_ref] [bib_ref] Racial differences in self-reports of sleep duration in a population-based study, Hale [/bib_ref] [bib_ref] Sleep duration among black and white Americans: Results of the National Health..., Nunes [/bib_ref] [bib_ref] Social and demographic factors related to sleep duration, Patel [/bib_ref] [bib_ref] The differential effects of sleep quality and quantity on the relationship between..., Sickel [/bib_ref] [bib_ref] Associations of frequent sleep insufficiency with health-related quality of life and health..., Strine [/bib_ref].
It is essential to recognize the negative ramifications of the COVID-19 pandemic on mental health to mitigate short-and long-term consequences. One in six (14.5%) individuals in the current study experienced poor mental health (>14 days) with a mean of 4.5 ± 8.7 days within a 30-day timeframe. These findings are consistent with other observational studies [bib_ref] The psychological burden of the COVID-19 pandemic and associated lockdown measures: Experience..., Burhamah [/bib_ref] [bib_ref] Prevalence of Poor Mental Health Days and Adverse Childhood Experience Reporting in..., Kapp [/bib_ref] [bib_ref] Association of Poor Mental-Health Days with COVID-19 Infection Rates in the U.S, Ransome [/bib_ref] [bib_ref] Are adversities and worries during the COVID-19 pandemic related to sleep quality?..., Wright [/bib_ref] [bib_ref] Mental health problems during the COVID-19 pandemics and the mitigation effects of..., Zhang [/bib_ref]. Large disease outbreaks are associated with increased mental health problems due to social disconnection through enforcement of stay-at-home orders, isolation, and unemployment [bib_ref] Psychiatric Manifestations of COVID-19 and Their Social Significance, Ptacek [/bib_ref]. Additionally, social disconnection either real or perceived is a primary risk factor for a suicidal attempt [bib_ref] Social Disconnection in Late Life Mental Illness-Commentary from the National Institute of..., Necka [/bib_ref].
We found a higher prevalence of poor mental health in areas with higher social deprivation indices. This result is consistent with the literature where greater neighborhood deprivation was associated with worse mental health [bib_ref] Area deprivation, perceived neighbourhood cohesion and mental health at older ages: A..., Salvatore [/bib_ref]. Communities with lower socioeconomic status have endured the greatest mental and physical health burden from the pandemic compared to adults with a higher socioeconomic status [bib_ref] Racial, Economic, and Health Inequality and COVID-19 Infection in the United States, Abedi [/bib_ref] [bib_ref] A comprehensive analysis of COVID-19 transmission and mortality rates at the county..., Bhowmik [/bib_ref] [bib_ref] Predictors of COVID-19-confirmed cases and fatalities in 883 US counties with a..., Robertson [/bib_ref]. These results indicate a critical need to understand the relationship between socioeconomic disadvantage and mental health particularly during a global health pandemic in order to mitigate negative short-and long-term consequences. Poor mental health can have profound ramifications including substance use disorders, drug overdose, and lost productivity [bib_ref] Addressing the burden of mental, neurological, and substance use disorders: Key messages..., Patel [/bib_ref]. Some important limitations should be considered when interpreting the results of the current study. First, self-reported sleep duration data were used; hence, estimates may be underestimated based on prior research comparing objective sleep duration with self-report [bib_ref] A systematic review and meta-analysis of behavioral sleep interventions for adolescents and..., Griggs [/bib_ref] [bib_ref] Sleep duration: How well do self-reports reflect objective measures? The CARDIA Sleep..., Lauderdale [/bib_ref]. People often over-estimate their sleep duration by reporting time in bed versus time asleep [bib_ref] Sleep duration: How well do self-reports reflect objective measures? The CARDIA Sleep..., Lauderdale [/bib_ref]. However, in large epidemiological studies, self-reported sleep data have been shown to be reliable [bib_ref] Sleep duration in the United States: A cross-sectional population-based study, Krueger [/bib_ref]. Findings should be corroborated with objective measures of sleep duration to replicate the findings observed in the present study. Second, the use of a cross-sectional design prevents us from investigating changes over time and attributing causality to changes in sleep and mental health trends. Third, respondents may have under-or over-reported mental health symptom frequency leading to increased or decreased estimation of mental health symptom experience in our study. Fourth, our sample was overrepresented in Non-Hispanic White (86.9% vs. 77%), underrepresented in Non-Hispanic Black (6.8% vs. 12.5%), comparable to Asian (0.9% vs. 2.5%), and comparable to Hispanic (2.1% vs. 4.4%) when compared to published 2020 Ohio demographics (census.gov). Finally, there may be societal or cultural variations in how sleep duration and mental health symptoms are reported [bib_ref] Who reports insomnia? Relationships with age, sex, ethnicity, and socioeconomic deprivation, Paine [/bib_ref]. Nonetheless, the large sample size with demographic variability contributes to the literature and generalizability of the findings.
The poor sleep and health outcomes observed in Ohio communities with the worst socioeconomic deprivation might be a common phenomenon in the other parts of the country. Timely and geographically targeted public health interventions from local and national government and health care providers are needed to facilitate access to resources and care in historically underserved areas. In addition, public health initiatives aimed at addressing social determinants of health and developing a safety net system are required to mitigate the worsening effects of the pandemic on sleep and mental health.
We encourage replication of our findings in other areas of the United States and in other countries. Sleep disturbance is linked to the occurrence and worsening of mental health problems, which in turn may affect sleep duration and quality. Investigators should explore the bidirectional relationship between sleep and mental health. Additionally, investigators may consider examining populations at a comparable geographic level like ZCTA or to examine larger areas (e.g., state/country) for broader generalizable findings, or smaller clusters (e.g., neighborhoods) to understand conditional processes that can be used for prescriptive multilevel interventions targeting specific domains of socioeconomic deprivation. Our findings underscore the need for geographically targeted public health interventions that support the sleep and mental health of the most socioeconomically disadvantaged. Ultimately, this work can be used to support local, regional, or national policy initiatives to modify societal infrastructures that enable social inequities to perpetuate ongoing healthcare disparities.
# Conclusions
The underlying mechanisms and trends between socioeconomic disadvantage, sleep duration, and mental health and whether these trends vary by sex, gender identity, or racial identity warrant further investigation. Socioeconomic deprivation characteristics may serve as novel targets to improve sleep and mental health in the United States population. In order to reduce these socioeconomic differences, key stakeholders and policymakers should consider the characteristics of the areas in which people live as well as the residents in those areas. The rising trend in short sleep and mental illness warrants further attention. Further, the current study highlights the need to address how catastrophic events disrupt sleep and mental health particularly in socioeconomically disadvantaged areas.
Author Contributions: According to the author guidelines, each of the authors named on the manuscript fit the criteria for authorship. S.G., principal investigator on the grants (R00NR018886), secured the funding, designed the study, acquired, analyzed, interpreted the data, and wrote the manuscript. C.H.D. contributed to study design, acquired, analyzed, interpreted the data, and wrote the manuscript. Q.H., D.D. and G.P. interpreted the findings and cowrote the manuscript. All authors have read and agreed to the published version of the manuscript.
## Informed consent statement: not applicable.
Data Availability Statement: Data are available upon request from https://odh.ohio.gov/know-ourprograms/behavioral-risk-factor-surveillance-system/data-and-publications (accessed on 1 August 2022) through a data use agreement.
## Conflicts of interest:
The authors declare no conflict of interest.
[fig] Figure 1: Short sleep and poor mental health prevalence % by social deprivation index quartiles. [/fig]
[fig] Funding: The first author, S.G., is funded by the National Institute of Nursing Research (NINR) R00NR018886. The second author, C.H.D., is funded by the National Institute of Nursing Research (NINR) K23NR019744. Q.H. is funded by Frances Payne Bolton School of Nursing Legacy Fellowship Program. G.P. is funded by the National Center for Advancing Translational Sciences (NCATS) KL2TR002547. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NIH or other funders. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of CASE WESTERN RESERVE UNIVERSITY (IRB # STUDY20221217 on 26 September 2022). [/fig]
[table] Table 1: Demographic Characteristics of the Sample. [/table]
[table] Table 2: Socioeconomic deprivation and covariates to sleep health and mental health outcomes (Linear Regression Models). [/table]
[table] Table 3: Socioeconomic deprivation, sleep health, and mental health (Logistic Regression Models). Residence in 10% most-deprived areas. [/table]
|
Phytochemical study of the headspace volatile organic compounds of fresh algae and seagrass from the Adriatic Sea (single point collection)
Performed phytochemical study contributes to the knowledge of volatile organic compounds (VOCs) of Halopteris filicina (Grateloup) Kü tzing, Dictyota dichotoma (Hudson) J. V. Lamouroux, Posidonia oceanica (L.) Delile and Flabellia petiolata (Turra) Nizamuddin from the Adriatic Sea (single point collection). VOCs were investigated by headspace solidphase microextraction (HS-SPME) and analysed by gas chromatography and mass spectrometry (GC-MS/FID). H. filicina headspace contained dimethyl sulfide (DMS; 12.8%), C 8compounds (e.g. fucoserratene (I; 9.5%)), benzaldehyde (II; 8.7%), alkane C 17 , dictyopterene D and C (III, IV), tribromomethane (V), 1-iodopentane, others. F. petiolata headspace was characterized by DMS (22.2%), 6-methylhept-5-en-2-one (9.5%), C 17 (9.1%), II (6.5%), compounds I-V. DMS (59.3%), C 15 (14.5%), C 17 (7.2%) and C 19 (6.3%) dominated in P. oceanica headspace. Sesquiterpenes were found in D. dichotoma, predominantly germacrene D (28.3%) followed by other cadinenyl (abundant), muurolenyl and amorphenyl structures. Determined VOCs may be significant for chemosystematics and chemical communications in marine ecosystem. marine algae were reviewed by Moore [1] including non-isoprenoid C 11 -compounds, acyclic undecapolyenes, less volatile organosulfur compounds, bromine-and iodine-containing haloforms, halogenated compounds, others. The research on the algae VOCs has been continued by different teams[2][3][4]. The production and role of volatile halogenated compounds from marine algae were reviewed in 2011[5]. Recently, more research papers on this topic appeared (e.g.[6,7]). VOCs biogeneration by microalgae, their occurrence, behaviour, ecological implications and industrial applications were described in 2016[8]. The volatile metabolites emission by in vivo microalgae were reviewed in 2017[9]. Tricyclic sesquiterpenes from marine origin were systematically presented in 2017[10].The common VOCs released by terrestrial plants are: ''green leaf volatiles" (GLVs) which consist almost exclusively of C 6 aldehydes and alcohols[11], other aliphatic compounds, monoterpenes, sesquiterpenes, phenylpropane derivatives, others[12]. This is in remarkable contrast to marine VOCs that show much higher structural diversity [13]: e.g. only among marine sesquiterpenes 54 different skeletal types of tricyclic sesquiterpenes were found [10]. In terrestrial ecosystems VOCs are very important group of infochemicals (e.g. plant-herbivore interactions, pheromones). Similarly, VOCs of marine organisms can have multiple functions in ecosystem in intraspecific (pheromones) and interspecific (kairomones) communication and in activated defences[14]. In addition, these compounds (usually present in the extracts) may exhibit other biological activities, e.g. antibacterial, cytotoxic and antitumor [10]. Different methods have been used to investigate VOCs from marine plants such as liquid-liquid extraction (LLE), hydrodistillation (HD), simultaneous distillation-extraction (SDE), vacuum-SDE (V-SDE), static and dynamic headspace extraction techniques (SHE and DHE) as well as headspace solid-phase microextraction (HS-SPME).However, limited number of studies have been carried out on marine algae and seagrass headspace VOCs in general[6,7,15,16]. In addition, to the best of our knowledge, there is only one paper [17] about VOCs (obtained by SDE) of the algae from the Adriatic Sea reporting the chemical composition of Padina pavonia (L.) Gaill. Considering limited data available on the chemical composition of marine plants found in the Adriatic Sea, the focus of this study was on 4 different samples collected from the same location (single point collection): 3 seaweeds (marine macroalgae) -2 brown algae (Halopteris filicina (Grateloup) Kützing, Dictyota dichotoma (Hudson) J. V. Lamouroux) and 1 green alga (Flabellia petiolata (Turra) Nizamuddin) as well as 1 seagrass (Posidonia oceanica (L.) Delile). F. petiolata has been commonly found in Mediterranean basin[18], often in association with other algae (e.g. Dictyopteris spp., Dictyota spp., Dilophus spp.). It is one of the main components of the phytocoenoses associated with the endemic and endangered seagrass P. oceanica[19,20]. The specific goals of present phytochemical study are: 1) to determine and unlock the headspace VOCs composition of targeted marine plants by headspace solid-phase microextraction (HS-SPME)/gas chromatography and mass spectrometry (GC-MS/FID)-first time report; 2) to compare the chemical biodiversity of found VOCs among the samples with other marine plants with same, similar or related constituents; 3) to discuss possible biosynthetic origin of identified VOCs from the literature data.Materials and methodsThe marine plant samplesWe state clearly that no specific permissions were required for the location and collection. There is no need to issue the permission to collect the samples used in this work because the sample were not taken from national park or protected area, but from the public place-sea. The Ministry approved this type of research by approving our research projects: Croatian Science Foundation under the project HRZZ-IP-11-2013-8547 (Research of Natural Products Phytochemical study of fresh algae and seagrass from the Adriatic Sea PLOS ONE | https://doi.
# Introduction
Volatile organic compounds (VOCs) are low-molecular compounds with low to moderate hydrophilicity that can not only be dissolved in water, but also dissipate into the gas phase at air-water interfaces. Prior to 1966, only one volatile substance was identified from wet, undecomposed seaweed and described in the literature as dimethyl sulfide (DMS). Afterwards, the number of identified marine plant VOCs has been constantly growing. In 1976, VOCs from a1111111111 a1111111111 a1111111111 a1111111111 a1111111111
and Flavours: Chemical Fingerprinting and Unlocking the Potential) and by the Croatian Ministry of Science and Education (MZO grant no. KK.01.1.1.01.0002 -The Scientific Center of Excellence for Marine Bioprospecting-BioProCro). We confirm that the field study did not involve endangered or protected species.
The samples of 2 brown algae (Halopteris filicina (Grateloup) Kützing, Dictyota dichotoma (Hudson) J. V. Lamouroux) and 1 green alga (Flabellia petiolata (Turra) Nizamuddin) as well as 1 seagrass (Posidonia oceanica (L.) Delile) were collected in the middle part of the Adriatic Sea, close to Mala Smokvica island, in July 2017 at the same geographic location (43 o 31'04.4"N, 15 o 56'32.6"E). Single point sample collection provided representative robust samples for first time chemical profiling. The samples were collected at 11 a.m. under sunny conditions-constantly illuminated, and area receive direct sunlight for >80% of the time from sunrise to sunset. During the samples collection there was no wind and the sea temperature was 22˚C. The algae were collected from depth of 0.5-3 m and the sea was collected from the same depth (directly into the plastic bag where collected algae were placed). The samples were separately collected and separately placed in air tight plastic bags containing seawater and were immediately transported to the laboratory. The samples were kept in dark at 4˚C until the extractions that were performed within 24 h after the collection. Before HS-SPME, each sample was taken out of the bag, cut into small pieces, and the excess of seawater was removed by placing it between filter paper layers for 2 min (the seawater was not removed completely).
## Headspace solid-phase microextraction (hs-spme)
The headspace extraction was performed with a manual SPME holder using polydimethylsiloxane/divinylbenzene (PDMS/DVB) fibre obtained from Supelco Co. (Bellefonte, PA, USA). The fibre was conditioned prior to the extraction according to the instructions by Supelco Co. For HS-SPME, previously prepared samples (1 g) were placed separately in 5 mL glass vials and hermetically sealed with PTFE/silicone septa. The vials were maintained in a water bath at 60 o C during equilibration (15 min) and HS-SPME (45 min). After sampling, the SPME fibre was withdrawn into the needle, removed from the vial, and inserted into the injector (250 o C) of the GC-FID and GC-MS for 6 min where the extracted volatiles were thermally desorbed directly to the GC column. HS-SPME was done in triplicate for each sample.
## Gas chromatography and mass spectrometry (gc-fid and gc-ms) analyses
The GC-FID analyses were carried out with an Agilent Technologies (Palo Alto, CA, USA) gas chromatograph model 7890A equipped with a flame ionization detector (FID) and a HP-5MS capillary column (5% phenyl-methylpolysiloxane, Agilent J and W). The GC conditions were similar to those described previously [bib_ref] Screening of Satureja subspicata Vis. honey by HPLC-DAD, GC-FID/MS and UV/VIS: prephenate..., Jerković [/bib_ref]. In brief, the oven temp. was programmed isothermal at 70 o C for 2 min, increasing from 70-200 o C at 3 o C min -1 , and held isothermally at 200 o C for 15 min; carrier gas was helium (He 1.0 mL min -1 ). The GC-MS analyses were performed using an Agilent Technologies (Palo Alto, CA, USA) gas chromatograph model 7820A equipped with a mass selective detector (MSD) model 5977E (Agilent Technologies) and a HP-5MS capillary column, under the same conditions as for the GC-FID analysis. The MSD (EI mode) was operated at 70 eV, and the mass range was 30-300 amu.
The identification of VOCs was based on the comparison of their retention indices (RI), determined relative to the retention times of n-alkanes (C 9 -C 25 ), with those reported in the literatureand their mass spectra with the spectra listed in Wiley 9 (Wiley, New York, NY, USA) and NIST 14 (D-Gaithersburg) mass spectral libraries. The percentage composition of the samples was computed from the GC peak areas using the normalization method (without correction factors). The average component percentages in [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] were calculated from GC-FID and GC-MS analyses (triplicates).
# Results and discussion
Headspace solid-phase microextraction (HS-SPME) with the fibre polydimethylsiloxane/ divinylbenzene (PDMS/DVB) was successfully applied first time to investigate VOCs from four fresh marine plants (Halopteris filicina (Grateloup) Kützing, Dictyota dichotoma (Hudson) J. V. Lamouroux, Posidonia oceanica (L.) Delile and Flabellia petiolata (Turra) Nizamuddin) that were collected from the Adriatic Sea from the same location. Single point sample collection provided representative robust samples for first time chemical profiling. Representative chemical profiles of their headspace composition were obtained without the application of heat that can lead to the artefacts (as in other methods for VOCs isolation e.g. hydrodistillation). The first insight into the results [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] reveals striking differences among the obtained chemical profiles. As was expected, significant contrast with common VOCs released by terrestrial plants can also be immediately noticed. The results are further discussed in separate subtitles for each seaweed or seagrass and for possible biosynthetic origin of identified VOCs.
## Halopteris filicina (grateloup) kützing headspace vocs
Halopteris filicina (Grateloup) Kützing is a greenish-brown, fern-like seaweed that belongs to the family Stypocaulaceae and it has been usually found in the Mediterranean and warm seas. Preliminary phytochemical screening of its crude extracts revealed the presence of alkaloids, saponins, flavonoids and terpenes [bib_ref] Screening of antibacterial activity in marine green, red and brown macroalgae from..., Alghazeer [/bib_ref]. Free amino acids, amino sulfonic acids, sugars and sterols have been examined and quantitatively determined [bib_ref] Amino acids, sugars and sterols of some Mediterranean brown algae, Amico [/bib_ref] , but there is no report on its volatile constituents. Its methanolic extract showed inhibition against Staphylococcus aureus, Staphylococcus epidermidis, Bacillus subtilis, Bacillus spp., Salmonella typhi, and Escherichia coli [bib_ref] Screening of antibacterial activity in marine green, red and brown macroalgae from..., Alghazeer [/bib_ref].
The headspace VOCs composition of H. filicina is reported in [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref]. Dimethyl sulfide (DMS) was the most abundant (12.8%). DMS serves as a chemoattractant for phytoplankton, bacteria, zooplankton, fish, and sea birds [bib_ref] Chemoattraction to dimethylsulfoniopropionate throughout the marine microbial food web, Seymour [/bib_ref] [bib_ref] Identification of the algal dimethyl sulfide-releasing enzyme: A missing link in the..., Alcolombri [/bib_ref]. Tribromomethane (2.1%) was found and it can be classified as polyhalomethane. The VOCs of Ascophyllum nodosum contained tribrommethane [bib_ref] A method for quantitative determination of volatile organic compounds in marine macroalgae, Newman [/bib_ref]. One organoiodine was identified: 1-iodopentane (0.8%). Asparagopsis taxiformis, Asparagopsis armata and Falkenbergia rufolanosa synthesized more than 100 different organoiodines including 1-iodopentane [bib_ref] The chemical ionization of organic compounds. 2 nd communication. Linear alkyl halides, Houriet R Gaumann [/bib_ref]. It has also been demonstrated that the volatile halogenated compounds are involved in the defence mechanism (allelopathy) of the algae [bib_ref] Halogenation and oxidation reactions with haloperoxidases, Franssen [/bib_ref] as one of several types of allelochemicals. They are probably important in the global cycling of gaseous organohalogen species [bib_ref] Formation and emission of monohalomethanes from marine algae, Itoh [/bib_ref]. The most widespread were aliphatic compounds, particularly C 8 -compounds (total 9 compounds, [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] such as: octan-1-ol (5.1%), octanal (4.7%), octan-3-one (4.1%), (3E)-octa-1,3-diene (2.3%), (3Z,5E)-octat-1,3,5-riene (9.5%), or oct-1-en-3-ol (1.2%). Two typical marine algae C 11 -hydrocarbons were also present: dictyopterene D (1.9%) and dictyopterene C (0.7%). Most of the chemical signals (specific volatile pheromones) in various genera of macroalgae are unsaturated, acyclic and/or alicyclic C 11 -hydrocarbons of different ring size and different degrees of unsaturation which are active at picomolar concentrations [bib_ref] Ecological functions of volatile organic compounds in aquatic systems, Fink [/bib_ref] [bib_ref] The chemistry of gamete attraction: chemical structure, biosynthesis, and (a)biotic degradation of..., Boland [/bib_ref]. C 8 -Hydrocarbon fucoserratene ((3Z,5E)-octa-1,3,5-triene) is also used for intraspecific chemical communication (pheromone) in Fucales [bib_ref] Ecological functions of volatile organic compounds in aquatic systems, Fink [/bib_ref] [bib_ref] Sexual reproduction and the role of sperm attractants in monoecious species of..., Mtiller [/bib_ref]. Another group of abundant aliphatic compounds were higher acyclic hydrocarbons such as heptadecane (4.0%), pentadecane (2.6%) or pentadec-1-ene (3.3%). Pentadecane was most typical in Ulva spp. and two species of Enterornorpha. Unbranched alkanes were identified in Ulva rigida with heptadecane predominance [bib_ref] Saturated hydrocarbons in marine organisms, Mironov [/bib_ref]. Three C 13 -norisoprenoids were detected: α-ionone (0.4%), β-ionone The symbols in the (1.2%) and β-cyclocitral (0.5%). C 13 -norisoprenoids, such as the potent flavour compound βionone, are produced by a diversity of algae taxa, for example Ulothrix fimbriata and Porphyra tenera [bib_ref] Ecological functions of volatile organic compounds in aquatic systems, Fink [/bib_ref]. Two phenypropane derivatives were found [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] : benzaldehyde (8.7%) and phenylacetladehyde (0.9%). Benzaldehyde was predominant in seaweed Monostroma nitidum [bib_ref] Determination of volatile compounds in four commercial samples of Japanese green algae..., Yamamoto [/bib_ref]. Previous research applying GC-MS on H. filicina reported the amounts of fatty acids (evaluated as the sum of all fatty acids (originally present and those resulting from the alkaline hydrolysis) by GC-MS after derivatization with BF 3 and purification) that ranged from 74 to 1897 mg kg -1 , dry weight [bib_ref] Distinct fatty acid profile of ten brown macroalgae, Silva [/bib_ref]. H. filicina significantly differs from the other species as it shows a much smaller fatty acids total content, although it contains important acids (e.g. hexadecanoic acid (C16:0), octadecanoic acid (C18:0), or eicosapentaenoic acid (EPA, C20:5 ω-3)). In addition, valuable compounds in macroalgae extracts were determined [bib_ref] Valuable compounds in macroalgae extracts, Andrade [/bib_ref] by GC-MS after derivatisation with trimethylsilyltrifluoroacetamide (MSTFA). H filicina extract contained proline, mannitol, hexadecanoic acid, octadecanoic acid, cholestanol and fucosterol. Sterols were identified by comparison of GC retention data of their TMSi ethers with those of authentic compounds and by GC-MS of their acetates [bib_ref] Amino acids, sugars and sterols of some Mediterranean brown algae, Amico [/bib_ref]. The sterol fraction contained fucosterol, cholesterol and 24-methylenecholesterol; minute amounts of 22-dehydrocholesterol and 24-methyl-cholesta-5,22-dien-3β-ol.
[formula] 3. 3-Methylbutanal C 5 H 10 O Aldehyde < 900 0.9 1.0 0.0 0.0 4. 2-Methylbutanal C 5 H 10 O Aldehyde < 900 0.5 0.0 0.0 0.0 5. Benzene C 6 H 6 Aromatic compound < 900 0.0 0.0 0.0 0.1 6. Pent-1-en-3-one C 5 H 8 O Unsaturated ketone < 900 2.6 2.9 0.0 0.2 7. Pentanal C 5 H 10 O Aldehyde < 900 0.0 1.6 0.0 0.1 8. (2E)-Pent-2-enal C 5 H 8 O Aldehyde < 900 0.0 0.0 0.0 0.1 9. Dimethyl disulfide C 2 H 6 S 2 Organosulfur compound < 900 0.0 0.0 0.0 0.1 10. 3-Methylbut-2-enal C 5 H 8 O Aldehyde < 900 1.5 1.6 0.0 0.1 11. (2Z)-Pent-2- [/formula]
## Flabellia petiolata (turra) nizamuddin headspace vocs
Flabellia petiolata (Turra) Nizamuddin is a green alga that belongs to Udoteaceae family (Chlorophyta, Bryopsidales) commonly found in the Mediterranean basin [bib_ref] The culturable mycobiota of Flabellia petiolata: First survey of marine fungi associated..., Gnavi [/bib_ref]. Compared to many other green algae, F. petiolata appears to be a particularly interesting species, since antibacterial, antiviral, antimitotic, antifungal and cytotoxic activities of its raw extract have been observed [bib_ref] Biological-activity of extracts from some Mediterranean macrophytes, Ballesteros [/bib_ref]. Selected constituents from F. petiolata from Turkey were determined [bib_ref] Selected chemical constituents and their seasonal variations in Flabellia petiolata (Turra) Nizam...., Bílgín [/bib_ref] such as: the average cellulose content (18.86±0.69%), the average crude protein content (22.45 ±0.62%) and the crude fat content (1.08±0.24%). There were no available data on its VOCs composition.
DMS, a highly-volatile organosulfur compound was quite abundant in the headspace of F. petiolata (22.2%; [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] and in general serves to the algae as a chemoattractant [bib_ref] Chemoattraction to dimethylsulfoniopropionate throughout the marine microbial food web, Seymour [/bib_ref] [bib_ref] Identification of the algal dimethyl sulfide-releasing enzyme: A missing link in the..., Alcolombri [/bib_ref]. Since DMS distribution in the water column does not correlate well with phytoplankton biomass [bib_ref] The role of the ocean in a regional sulfur cycle, Bates [/bib_ref] , it led to suggestions that only certain groups of marine algae (e.g. Emiliania huxleyi) produce significant amounts of DMS. Polyhalomethane tribromomethane was present with low percentage (1.8%). Variety of aliphatic compounds were found in the headspace [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref]. Low-molecular aliphatic compounds up to 7 carbons, particularly ketones and aldehydes, were quite widespread, e.g. 3-methylbutanal (1.0%), pentanal (1.6%), 3-methylbut-2-enal (1.6%), hexanal (1.5%), or pent-1-en-3-one (2.9%). In distinction from lipid-derived low-molecular aliphatic compounds, 3-methylbutanal is derived from leucine [bib_ref] Optimization of dynamic headspace extraction of the edible red algae Palmaria palmata..., Pape [/bib_ref] and it is generally associated with heated products; however fresh P. petiolata was used for this research. It is frequently found in many sea products, such as crayfish or lobster as well as in red algae Palmaria palmate [bib_ref] Ecological functions of volatile organic compounds in aquatic systems, Fink [/bib_ref] [bib_ref] Selected chemical constituents and their seasonal variations in Flabellia petiolata (Turra) Nizam...., Bílgín [/bib_ref]. 6-Methylhept-5-en-2-one (9.5%) was the most abundant ketone and previously it was found in relatively high amount (along with 6-methylheptan-2-one) in V-SDE extract of the green alga Capsosiphon fulvescens [bib_ref] Volatile compounds of the green alga, Capsosiphon fulvescens, Sun [/bib_ref]. Seven C 8 -compounds were identified with major representatives (3Z,5E)-octat-1,3,5-riene (fucoserratene; 3.2%), oct-1-en-3-ol (2.5%) and oct-1-en-3-one (1.1%). Identified dictyopterene D (7.4%) and dictyopterene C (0.7%) belong to C 11hydrocarbons (volatile pheromones). Heptadecane (9.1%) was the most abundant among alkanes and it was already observed to be one of the most common hydrocarbons in marine algae, detected in relevant amount (26.38%) in Hypnea cornuta [bib_ref] Saturated hydrocarbons in marine organisms, Mironov [/bib_ref]. C 13 -norisoprenoids were represented by α-ionone (0.9%), β-ionone (0.7%) and β-cyclocitral (0.8%). β-Ionone and βcyclocitral were found previously among the major volatile compounds of the essential oil isolated from Capsosiphon fulvescens [bib_ref] Volatile compounds of the green alga, Capsosiphon fulvescens, Sun [/bib_ref] by V-SDE. Benzaldehyde was the most abundant (6.5%) among phenylpropane derivatives. Only one monoterpene was found in low percentage (1,8-cineole; 0.6%).
## Posidonia oceanica (l.) delile headspace vocs
Posidonia oceanica (L.) Delile represents one of the most widespread species endemic to the Mediterranean Sea. Previous phytochemical studies showed that the mature leaves of P. oceanica are rich in amino acids, chicoric acid, p-coumaric acid, vanillin, ferulic acid, gentisic acid, caffeic and cinnamic acids, sterols and phenolic derivatives [bib_ref] HPLC determination of chicoric acid in leaves of Posidonia oceanica, Haznedaroglu [/bib_ref]. In chloroformic extract from P. oceanica a novel methylated sesquiterpene was identified (2E)-3,7,12-trimethyltridec-2-en-1-ol named posidozinol along with β-sitosterol and four fatty acids (palmitic, palmitoleic, oleic and linoleic) [bib_ref] A novel methylated sesquiterpene from seagrass Posidonia oceanica (L.) Delile, Hammami [/bib_ref]. The occurrence of dimethylsulfoniopropionate (DMSP) and its variability during the seasonal cycle in the leaves of P. oceanica was reported [bib_ref] Seasonal and spatial variability of dimethylsulfoniopropionate (DMSP) in the Mediterranean seagrass Posidonia..., Borges [/bib_ref]. However, there is no report on its headspace VOCs.
The main compound of P. oceanica headspace was DMS (59.3%; [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] that was expected since its precursor DMSP was previously identified in the leaves of this seagrass [bib_ref] Seasonal and spatial variability of dimethylsulfoniopropionate (DMSP) in the Mediterranean seagrass Posidonia..., Borges [/bib_ref]. DMSP has been proposed to exhibit physiological roles (e.g. as an intracellular osmolyte and antioxidant), and also serves as a chemoattractant [bib_ref] Identification of the algal dimethyl sulfide-releasing enzyme: A missing link in the..., Alcolombri [/bib_ref] [bib_ref] An antioxidant function for DMSP and DMS in marine algae, Sunda [/bib_ref]. DMS (mediated by both bacterial and algal DMSP lyases) has been generated in oceans at remarkably high amounts (>107 tons per year) and is a key component of the ocean sulfur cycle and has a global role in atmosphere-ocean feedback processes [bib_ref] Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate, Charlson [/bib_ref] [bib_ref] The case against climate regulation via oceanic phytoplankton sulphur emissions, Quinn [/bib_ref]. Chlorophyceae, especially Ulva, Enteromorpha, and Codium, and red alga, Polysiphonia, are capable of producing large amounts of DMS in contrast to Phaeophyceae (brown algae) which produce little [bib_ref] Analysis of dimethyl sulfonium compounds in marine algae, White [/bib_ref] [bib_ref] Measurement and osmotic significance of ß-dimethyl-sulphoniumpropionate in marine macroalgae, Reed [/bib_ref] that is in agreement with present research [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref]. Namely, DMS percentage was the highest in P. oceanica, than in green alga (F. petiolata) and the lowest in brown algae (Cystoseira sp. and D. linearis). Besides higher alkanes pentadecane (14.5%) and heptadecane (7.2%) that were abundant in F. petiolata and Cystoseira spp. [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] , nonadecane was also found (6.3%).
## Dictyota dichotoma (hudson) j. v. lamouroux headspace vocs
The genus Dictyota is represented by more than 40 species, thus being the richest genus of the family Dictyotaceae. In the Mediterranean in general, only the species Dictyota dichotoma and Dictyota linearis are found [bib_ref] Diterpenes from the brown algae Dictyota dichotoma and Dictyota linearis, Siamopoulou [/bib_ref]. Brown algae of the family Dictyotaceae produce a significant number of secondary metabolites, especially diterpenes with three types of carbon skeletons: xenicanes; dolabellanes and ''extended sesquiterpenes". Many members of the family produce cyclic diterpenes, unique in the structural variety of marine natural products [bib_ref] Diterpenes from the brown algae Dictyota dichotoma and Dictyota linearis, Siamopoulou [/bib_ref]. Terpenoid metabolites were identified in crude extract of D. dichotoma [bib_ref] Diterpenes from the brown algae Dictyota dichotoma and Dictyota linearis, Siamopoulou [/bib_ref] , e.g.: bicyclosesquiphellandrene, germacrene D, dictyoxide, pachydictyol A, isopachydictyol A, axenol, acetyldictyolal, dictyol B acetate, isopachydictyolal, 10-acetoxy-18-hydroxy-2,7-dolabelladiene, dictyol E, fucosterol, acetoxycrenulide, hydroxyacetyldictyolal, isodictyohemiacetal, dictyol A, dictyol B, dictyol C, and hydroxycrenulide. In addition, D. linearis contains different natural compounds [bib_ref] Diterpenes from the brown algae Dictyota dichotoma and Dictyota linearis, Siamopoulou [/bib_ref] such as: dictyoxide, isopachydictyol A, pachydictyol A, dictyodial, 18-hydroxy-2,7-dolabelladiene, neodictyolactone, acetylsanadaol, acetyldictyolal, 10-acetoxy-18-hydroxy-2,7-dolabelladiene, dictyol E, 10,18-dihydroxy-2,7-dolabelladiene and dictyol C. There is no report on D. dichotoma headspace VOCs.
In distinction form other investigated samples, the major identified headspace VOCs of D. dichotoma were sesquiterpenes. Predominant sesquiterpene was germacrene D (28.3%; [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref] that was previously found in D. dichotoma [bib_ref] Diterpenes from the brown algae Dictyota dichotoma and Dictyota linearis, Siamopoulou [/bib_ref]. Bicyclogermacrene (4.7%) was also present. Sesquiterpenoids bicyclosesquiphellandrene, germacrene D and axenol were previously isolated from Taonia atomaria [bib_ref] Sesquiterpenes from the brown alga Taonia atomaria, Rosa [/bib_ref]. In this research, an array of sesquiterpenes biosynthetically connected with germacrene D were found [fig_ref] Table 1: VOCs composition determined by headspace solid-phase microextraction [/fig_ref]. They probably derived from protonation of germacrene D resulting in cadinenyl, muurolenyl, and amorphenyl cations that further react to form final products. Cadinenyl type sesquiterpenes were abundant: δ-cadinene (8.3%), γ-cadinene (3.4%), β-cadinene (2.8%), and trans-cadina-1,4-diene (1.2%). Epizonarene was found with 4.3% and it can be formed directly from germacrene D or from other intermediate cadinenes. It has been proposed that cyclization of germacrene D or of its endocyclic double bond isomer could result in the formation of bourbonene and copaene skeletons. β-Bourbonene was identified with (5.1%) and α-copaene with lower percentage. As a consequence of the thermodynamically unfavourable process to cis-decalin systems, the muurolenes and amorphenes are usually formed in smaller quantities that is in agreement with current research since they were found with lower percentages (α-muurolene (2.2%), γ-muurolene (2.1%), and α-amorphene (3.5%)).
## Possible biosynthetic origin of identified vocs
The formation of DMS [bib_ref] Volatile compounds from marine algae, Moore [/bib_ref] results from an enzymatic decomposition of dimethyl-β-propiothetin, a metabolite of methionine that is fairly widespread in marine plants. Formed dimethylsulfoniopropionate (DMSP), a tertiary sulfonium compound involved in osmoregulation in algae, is the precursor of DMS. Recently, the algal enzyme responsible for formation of DMS from DMSP has been identified and characterized in algae Emiliania huxleyi [bib_ref] Identification of the algal dimethyl sulfide-releasing enzyme: A missing link in the..., Alcolombri [/bib_ref].
Marine macroalgae exhibit a high ability to fix halide ions and form a variety of halogenated secondary metabolites [bib_ref] Volatile compounds from marine algae, Moore [/bib_ref]. In present research only tribromomethane and 1-iodopentane were found. A halogenating enzyme, haloperoxidase [bib_ref] Halogenation and oxidation reactions with haloperoxidases, Franssen [/bib_ref] is considered to participate in their synthesis in the presence of halides and hydrogen peroxide. Among them bromoperoxidases were detected in seaweeds (e.g. from Corallina pilulifera or Ascophyllum nodosum). The enzyme produces CHBr 3 by its reaction with ketoacids, halide ions and hydrogen peroxide [bib_ref] Occurrence of bromoperoxidase in the marine green macro-alga, bulvella lens, and emission..., Ohshiro [/bib_ref].
Both in terrestrial and aquatic ecosystems, VOCs of primary producers are usually dominated by lipid degradation products, and the overall mechanism for their enzymatic release is identical as in terrestrial plants [bib_ref] The biogeneration of green odor by green leaves, Hatanaka [/bib_ref] and algae [bib_ref] The oxylipin chemistry of attraction and defense in brown algae and diatoms, Pohnert [/bib_ref]. The enzyme cascade is initiated by activated phospholipase, followed by lipoxygenase and hydroperoxide lyase, which leads to VOCs liberation. However, the particular enzymes are highly species-and sometimes even strainspecific [bib_ref] The oxylipin chemistry of attraction and defense in brown algae and diatoms, Pohnert [/bib_ref] that can explain large biodiversity of volatile lipid degradation products (e.g. carbonyl compounds, alcohols, hydrocarbons). Marine algae contain C 20 , C 22 and C 18 unsaturated fatty acids, and they can produce both plant (C 18 ) and animal type (C 20 and C 22 ) fatty acid hydroperoxides. Short-chain aldehydes (e.g. C 6 , C 9 ) and middle-chain aldehydes (e.g. C 10 ) that were particularly present in H. filicina and F. petiolata are mainly formed from fatty acids (C 20 ) in marine algae (via hydroperoxides), whereas they are formed from C 18 fatty acids in higher plants [bib_ref] Formation of aldehyde flavour (n-hexanal, 3Z-nonenal and 2E-nonenal) in the brown alga,..., Boonprab [/bib_ref] [bib_ref] 4-Decadienals are produced via (R)-11-HPITE from arachidonic acid in marine green alga..., Akakabe [/bib_ref]. For example, the formation of hexanal is proposed via linoleic acid cascade and arachidonic acid cascade through their hydroperoxides as intermediates by the lipoxygenase/fatty acid hydroperoxide lyase pathway. It could also be provided by oxidation from other polyunsaturated fatty acids, as well as heptanal [bib_ref] Gas and liquid chromatography-mass spectrometry of aldehydic products from lipid peroxidation, Enoiu [/bib_ref]. Lower aliphatic alcohols may be formed by decomposition of secondary hydroperoxides of fatty acids by the reduction of the corresponding aldehydes [bib_ref] Volatile flavor components in crayfish waste, Tanchotikul [/bib_ref]. Octanal and nonanal could originate from ω9 mono-unsaturated fatty acids (MUFAs) and also from ω6 PUFAs such as linoleic acid [bib_ref] Volatile flavor components in crayfish waste, Tanchotikul [/bib_ref]. Following this general concept of lipid peroxidation, and subsequent oxidative cleavage of the carbon skeleton, the biosynthesis of C 11 -and C 8 -hydrocarbons could start from a single precursor (e.g. eicosapentaenoic acid). The polyunsaturated fatty acid substrate could be activated [bib_ref] Biosynthesis of the algal pheromone fucoserratene by the fresh-water diatom Asterionella formosa..., Hombeck [/bib_ref] either by 9-lipoxygenase or by 12-lipoxygenase, and resulting 9-or 12-hydroperoxides that cleave oxidatively to produce characteristic C 11 -and C 8 -hydrocarbons (e.g. fucoserratene). However, the origin of 3-methylbutanal is well known. It is obtained from amino acids, and more particularly from leucine, during Maillard reactions by Strecker degradation [bib_ref] Optimization of dynamic headspace extraction of the edible red algae Palmaria palmata..., Pape [/bib_ref].
Terpenoids are enzymatically synthesized from acetyl-CoA (mevalonate (MVA) pathway) and/or pyruvate (deoxyxylulose-5-phosphate (DXP) pathway) via precursor 2-isopentenyl pyrophosphate (2-IPP) and its isomer 3-isopentenyl pyrophosphate (3-IPP). Regular monoterpenes derived from geranyl pyrophosphate (GPP), but are not widespread among marine algae. More often are sesquiterpenes derived metabolically from 300 distinct C 15 -hydrocarbon skeletons, which in turn are produced from farnesyl diphosphate (FPP) by the action of sesquiterpene synthases [bib_ref] Sesquiterpene synthases: passive catalysts or active players?, Miller [/bib_ref]. The structure elucidation, biosynthesis, and biological activity of marine carbotricyclic sesquiterpene compounds were reviewed [bib_ref] Tricyclic sesquiterpenes from marine origin, Bideau [/bib_ref].
Carotenoid cleavage dioxygenases (CCDs) catalyze oxidative cleavage of carotenoids, resulting in the production of norisoprenoids-apocarotenoids [bib_ref] The carotenase AtCCD1 from Arabidopsis thaliana is a dioxygenase, Schmidt [/bib_ref] via glycosylation and breakdown of stored glycosides by glycosidase enzymes. CCDs often exhibit substrate promiscuity, which probably contributes to their natural diversity. Non-enzymatic reactions involving one or several steps of carotenoid degradation, stimulated by light, oxygen, temperature and acid hydrolysis can also occur. The breakdown products of carotenoids found in this research belong to C 13 -norisoprenoids with megastigmane structure. In addition, C 11 -non-isoprenoids were identified. Dictyopterene A ((1R,2S)-cis-1-vinyl-2-(trans-1'-hexenyl)cyclopropane) and dictyoterpene B ((1R,2S)-cis-1-vinyl-2-(trans-1,cis-3-hexadienyl)cyclopropane) are proposed precursors [bib_ref] Volatile compounds from marine algae, Moore [/bib_ref] of dictyopterene C (6-butylcyclohepta-1,4-diene) and dictyopterene D (6-[(1Z)-but-1-enyl]cyclohepta-1,4-diene). Proposed biogenesis of dictyopterenes starts from (3S)-1,cis-5-undecadien-3-ol and (3S)-1,cis-5,cis-8-undecatrien-3-ol by dehydration and cyclization.
Benzenoid and phenylpropanoid volatile compounds, primarily derived from phenylalanine require shortening of the carbon skeleton side chain by a C 2 -unit, which can potentially occur via either the β-oxidative pathway or non-oxidatively [bib_ref] Understanding in vivo benzenoid metabolism in petunia petal tissue, Boatright [/bib_ref].
# Conclusions
Considering limited data available on the chemical composition of marine plants from the Adriatic Sea, the present research is contribution toward their better chemical characterization. Significant differences were found among the headspace VOCs from 3 seaweeds and 1 seagrass. High abundance of DMS was found in P. oceanica followed by F. petiolata and H. filicina indicating those plants as source of sulfur compounds in marine ecosystem. Their headspace contained individually variety of C 8 -compounds (e.g. fucoserratene), benzaldehyde, alkanes C 15 , C 17 and C 19 , dictyopterene D and C, others. Sesquiterpenes were found in D. dichotoma, predominantly germacrene D indicating similarity to terrestrial aromatic plants. Identified VOCs contain different types of organic compounds that may be significant for chemosystematics and ecology (chemical communications in marine ecosystems).
# Author contributions
[fig] Formal analysis: Igor Jerković, Zvonimir Marijanović, Marin Roje, Piotr M. Kuś, Stela Jokić.Funding acquisition: Igor Jerković, Marin Roje. [/fig]
[table] Table 1: VOCs composition determined by headspace solid-phase microextraction (HS-SPME). [/table]
|
Successful repair of an arteriovesical fistula as a complication after coil embolization for right hypogastric artery aneurysm
An 84-year-old man has a history of a right hypogastric artery aneurysm that was excluded with endograft several years ago. His course was complicated by a type II endoleak with an enlarging aneurysm that required multiple attempts of coil embolization. Several years later, he presented with gross hematuria and was found to have embolization coils extruding into his bladder and an arteriovesical fistula. As seen in this case, endovascular intervention for a type II endoleak is not benign. We describe a novel complication of an arteriovesical fistula that necessitated a difficult repair and ultimate revascularization. (J Vasc Surg Cases and Innovative Techniques 2020;6:585-7.)
The high mortality of ruptured isolated iliac artery aneurysms necessitates elective repair for those with diameters of greater than 3 cm, 1,2 although more recent studies suggest surveillance until at least 4 cm. [bib_ref] Few internal iliac artery aneurysms rupture under 4 cm, Laine [/bib_ref] When anatomically feasible, endovascular iliac artery aneurysmal repair (EVIAR) is preferred over an open approach. Type II endoleak is a common complication after EVIAR. [bib_ref] Endovascular management of isolated iliac artery aneurysms, Boules [/bib_ref] [bib_ref] Nature and significance of endoleaks and endotension: summary of opinions expressed at..., Veith [/bib_ref] A type II endoleak can often be closely observed, and intervention is only advised for aneurysms with evidence of growth, although large aneurysms may also be subjected to intervention because they are thought to be more likely to rupture. [bib_ref] Management of type II endoleaks: preoperative versus postoperative versus expectant management, Jonker [/bib_ref] When indicated, a type II endoleak is repaired endovascularly to prevent retrograde collateral flow to the growing aneurysmal sac. However, this approach carries significant odds of failure, high recidivism rates, frequent reinterventions, [bib_ref] Midterm outcomes after treatment of type II endoleaks associated with aneurysm sac..., Gallagher [/bib_ref] and, as seen in this case, are not without complication. We describe a complex open repair of an arteriovesical fistula, a rare complication from multiple attempts of coil embolotherapy of a type II endoleak in a right hypogastric artery aneurysm. The patient provided consent for the publication of this article.
## Case report
An 84-year-old man presented to clinic with gross hematuria.
He was several years status post EVIAR for a right hypogastric artery aneurysm that was excluded with outflow branch coiling and coverage of the hypogastric ostium using an endograft spanning from the right common iliac artery (CIA) to the external iliac artery. However, the coiling of outflow branches proved incomplete. The aneurysm continued to enlarge owing to type II endoleak despite multiple attempts of percutaneous translumbar coil embolizations, one of which was complicated by retroperitoneal hemorrhage requiring admission and multiple transfusions. Four years after the last intervention, he was seen by urology for hematuria, where cystoscopy revealed several embolization coils extruding into the posterior aspect of the bladder [fig_ref] Fig 1: Angiographic coils in urinary bladder, seen on cystoscopy view [/fig_ref]. Visualization on computed tomography angiography imaging was challenging owing to significant coil artifact. Thus, a transfemoral catheter angiography was completed, which confirmed resolution of the endoleak, but showed a large right hypogastric aneurysmal sac that was outlined by the coils [fig_ref] Fig 2: Angiography with patent stent graft from common to external iliac artery on... [/fig_ref]. The patient was afebrile, hemodynamically stable, and without overt signs of infection. Preoperative blood cultures and urine analysis were not drawn. Still, we had high suspicion for an indolent infection of the coils as the cause of the fistula and subsequent extrusion into the bladder. The aneurysm and endograft were assumed to be infected as well.
Thus, we planned for the patient to undergo a complex joint resection and repair with urology. We then sought to explant the iliac endograft, aneurysm, and coils. A midline approach was necessary to ensure adequate control of the left contralateral iliac artery for complete resection of the right iliac endograft. The aneurysm also crossed the midline, and we did not want to risk leaving any of the medial wall through a right retroperitoneal approach. Thus, via a midline laparotomy, we dissected to the retroperitoneum to gain exposure of the aortic bifurcation and bilateral iliacs including endograft. We made a longitudinal arteriotomy in the pulseless right hypogastric aneurysm and large volumes of chronic liquefied thrombus drained out under pressure. Many coils came out freely, although some were adherent to the inner lining of the aneurysm or found extruding through the wall of the aneurysm itself . Individual coils were also found floating freely in the retroperitoneum around the aneurysm. At that point, urology arrived to repair the iliac-cystostomy. The bladder was repaired primarily. We then proceeded with total aneurysmectomy to ensure removal of all coils. We managed the endograft by cross-clamping the distal aorta below the inferior mesenteric artery and the proximal left CIA. We made a transverse arteriotomy and divided the graft at the distal right CIA, then removed both proximal and distal segments in entirety. There was obvious slime on the entire length of the endograft, which was sent for culture.
# Discussion
The majority of isolated iliac artery aneurysms are found incidentally; less commonly, patients may be symptomatic from compression or erosion owing to the growing aneurysm and rupture. [bib_ref] Natural history and management of iliac aneurysms, Richardson [/bib_ref] [bib_ref] Isolated iliac artery aneurysms, Levi [/bib_ref] Historically, open repair was performed with proximal and distal ligation of the aneurysm with bypass revascularization or obliterative aneurysmorrhaphy and graft interposition. [bib_ref] The isolated internal iliac artery aneurysm -a review, Dix [/bib_ref] This approach is complicated by the technical challenge of operating deep in the pelvis, the unique branching pattern of the iliac arteries, and the close proximity to adjacent viscera and nerves. [bib_ref] Outcomes of open and endovascular repair for ruptured and nonruptured internal iliac..., Rana [/bib_ref] Therefore, endovascular repair is preferred.
In several retrospective studies, EVIAR was associated with a lower 30-day mortality rate and similar graft patency rates when compared with open repair. [bib_ref] Open vs. endovascular repair of isolated iliac artery aneurysms: a 12-year experience, Patel [/bib_ref] [bib_ref] Isolated iliac artery aneurysms: endovascular versus open elective repair, Pitoulias [/bib_ref] [bib_ref] Isolated iliac artery aneurysms: a contemporary comparison of endovascular and open repair, Chaer [/bib_ref] Although type II endoleak is frequently benign and expected to thrombose spontaneously, it remains the most common reason for intervention after endovascular repair. [bib_ref] Nature and significance of endoleaks and endotension: summary of opinions expressed at..., Veith [/bib_ref] [bib_ref] Type II endoleaks: challenges and solutions, Brown [/bib_ref] Similar to aortic aneurysms, intervention is offered for iliac aneurysms that are growing or persistent in size as a result of the endoleak. [bib_ref] Management of type II endoleaks: preoperative versus postoperative versus expectant management, Jonker [/bib_ref] [bib_ref] Type II endoleaks: challenges and solutions, Brown [/bib_ref] There are a variety of mechanisms for endovascular embolization of type II endoleaks including the use of coils, thrombin, or other embolic agents. 14-16 However, clinical success, which is defined as no recurrence of endoleak, is very difficult to achieve without open intervention. In a 9-year retrospective review of patients seen at a large tertiary institution, albeit for type II endoleak after abdominal aortic aneurysmal repairs, Gallagher et al 7 showed approximately 20% clinical success after 30-month follow-up from the initial treatment. The majority of these patients required reintervention, which increased their risk for complications including blood transfusions, hematoma, and cardiac arrhythmia. We present a novel complication from an intervention for type II endoleak. Although embolotherapy was ultimately successful after multiple attempts at eliminating the endoleak and preventing exsanguination through the arteriovesicular fistula, primary infection of the coils was the etiology for the fistula. This case should offer consideration of a prosthetic graft infection when a patient with an iliac aneurysmal treatment history presents with hematuria. Direct fistula into the bladder by the native aneurysm itself seems implausible. Bacteriology obtained at the time of explant was consistent with a prosthetic graft infection.
Given the high recurrence rate with current therapies for type II endoleak, if treatment is necessary, the timing of surgical repair as opposed to multiple failed embolotherapy attempts, needs to be addressed.
[fig] Fig 1: Angiographic coils in urinary bladder, seen on cystoscopy view. [/fig]
[fig] Fig 2: Angiography with patent stent graft from common to external iliac artery on the right side (white arrow), no flow though proximal right internal iliac artery, and large coils protruding toward midline (black arrow). [/fig]
|
Discovery of high-performance low-cost n-type Mg3Sb2-based thermoelectric materials with multi-valley conduction bands
## Reviewer 1
Comment 1 'D. Statistic approach does not apply. The study was made on single sample.
E. Concerning the material composition reported in the manuscript I do not have doubts in the reliability of the results. The robustness and validity of the results cannot be judged in general and should be considered with the usual criterions for the thermoelectric measurements made on the single specimen.
F. The report is written quiet fragmentarily and is focused only on one specimen.'
## Reply
We thank the referee for this comment. We agree that in general the robustness and validity of the results cannot be judged on a single sample. To address this concern, we have made more samples with nominal compositions Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x = 0.04, 0.05, 0.08, and 0.20), which have been characterized on our homebuilt systems. The high zT in the high-performance sample with x = 0.05, is measured both on the commercial ZEM-3 setup and the homebuilt system, and the two sets of data show very good agreement with each other (see [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref].
The manuscript has been rewritten to include the additional samples with more compositions. The major revision is summarized below:
(a) [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref] , and their captions have been modified to include different compositions Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x = 0.04, 0.05, 0.08, and 0.20).
(b) One sentence is modified in line 5 of the second paragraph on page 3: "The TE performance of n-type Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x = 0.04, 0.05, and 0.08) is at least 2 times higher than that of p-type Mg 3 Sb 2 -based compounds including Na-doped Mg 3 Sb [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] , a factor of at least 3.7 larger than the mobility 17,18 of the undoped Mg 3 Sb 2 ".
(f) In the beginning of the second paragraph several sentences have been modified as: "The total thermal conductivity values of n-type Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x = 0.04, 0.05, 0.08, and 0.20) samples are low and exhibit decreasing trends with increasing temperature [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref]. The lowest room-temperature thermal conductivity of 0.743 W m -1 K -1 is observed for the sample with x = 0.04. The total thermal conductivity, κ, value decreases to ~0.556 W m -1 K -1 at 725 K."
(g) In sample synthesis of Methods on page 13: "The samples with nominal compositions Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x = 0.04, 0.05, 0.08, and 0.20) were synthesized by combining arc melting and SPS techniques."
(h) In the first paragraph of page 15 one sentence is added: "The Seebeck coefficients of the pellets were then measured from the slope of the thermopower versus temperature gradient using chromel-niobium thermocouples on an in-house system, which is similar to the one reported by Iwanaga et al The composition with Bi 0.5 was chosen as a proof-of-concept, because Mg 3 Sb 1.5 Bi 0.5 possesses a suitable band gap of 0.43 eV as well as ΔE K-ML = 0.12 eV with the ML band as the conduction band minimum. This is good for the electrical transport since the ML band with a high valley degeneracy of 6 can be reached with a low doping concentration. In addition, the formation energy of Te doping on the anion sites was calculated and shown in [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] , which indicates that Te doping on the anion sites of Mg 3 Sb 1.5 Bi 0.5 is easier than that of Mg 3 Sb 2 .
We agree that other compositions with x < 1 and x ≠ 0.5 might also show good performance. However, as a proof-of-concept paper, we have shown a progression to high zT in n-type Mg 3 Sb 1.5 Bi 0.5 samples with different Te compositions. It is an excellent idea to caary out a study of the full range of Bi compositions in a future study. Thanks.
(a) The energy gap E g versus x plot was added in .
(b) Two sentences are added in the beginning of page 9: "The obvious bipolar effect for Mg 3 Sb 2-x Bi x (x > 1) is confirmed in a previous experimental report 17 . Hence, n-type Mg 3 Sb 2-x Bi x (x ≤ 1) compounds are very promising TE candidates if properly doped on the anion sites."
(c) Two sentences have been modified in the second paragraph of page 9: "Achieving n-type Mg 3 Sb 2 by doping tellurium on the anion sites has been attempted and found to be difficult, while it is easier in Mg 3 Sb 2-
x Bi x solid solutions. This is probably due to the lower formation energy of tellurium doping on the anion sites of Mg 3 Sb 2-x Bi x solid solutions (see one example in [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]. From the above theoretical calculation, Mg 3 Sb 1.5 Bi 0.5 possesses a small band gap of 0.43 eV as well as ΔE K-ML = 0.12 eV with the ML band as the conduction band minimum [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] , making it a potential candidate for n-type doping.
The experimental validation is successfully demonstrated in n-type doped Mg 3 Sb 1.5 Bi 0.5 solid solution using tellurium as an effective dopant."
Comment 3
'Justification of the doping: why tellurium and not selenium or sulfur, why is 0.05 an optimum of the doping if it is).'
## Reply
In this work we choose tellurium for n-type doping, because tellurium is the close neighbor of antimony in the Periodic It means that tellurium has a similar ion radius and electronegativity relative to antimony and bismuth, which will possibly make it easier for tellurium to be doped on the Sb or Bi sites. In addition, tellurium has been widely used as an effective n-type dopant in thermoelectric research. We agree that selenium or sulfur doping might also be very interesting due to the abundance of Se and S elements and we believe that it could be a nice future work. Thanks for this comment.
To address the second part of the comment, we have made more samples with nominal compositions Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x = 0.04, 0.05, 0.08, and 0.20). As shown in the updated [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] , x = 0.04 is an optimum of Te doping, and the sample with x = 0.05 shows a comparable performance. Please see the reply to comment 1 for details.
## Reply
We are sorry for the confusion about this point. In [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] ,b, the DFT model is for n-type Mg 3 Sb 2 , which has the ML band and the K band converged. But our experimental data is on Te-doped Mg 3 Sb 1.5 Bi 0.5 which has an energy difference of 0.12 eV between the ML band and K band. This is why the DFT curve looks overestimated.
Unfortunately, it is impossible for semi-classical Boltzman theory to calculate Seebeck coefficient from the spectral-weight representation of the band structure of Mg 3 Sb 1.5 Bi 0.5 . The spectral-weight representation of the band structure of Mg 3 Sb 1.5 Bi 0.5 was used to understand that the ML band moves downward 0.12 eV below the K band and becomes the conduction band minimum. Since the ML band with a high valley degeneracy of 6 is highly desirable for electrical transport, moving the ML band to the conduction band minimum is good and it can be reached with a low doping level.
In order to make the multiple band behavior of Mg 3 Sb 1.5 Bi 0.5 clear, the following revision has been applied:
(a) The second paragraph of page 8: "The results prove that multiple band behavior including the ML band with a six-fold valley degeneracy is indeed preserved in Mg 3 Sb 1.5 Bi 0.5 solid solution and the dispersions and effective masses of the K band and ML band are very similar to those of Mg 3 Sb 2 . The main difference, however, is that the ML band in Mg 3 Sb 1.5 Bi 0.5 is shifted downward 0.12 eV below the K band and therefore becomes the conduction band minimum. This is good for the TE performance of Mg 3 Sb 1.5 Bi 0.5 since the ML band with a high val-ley degeneracy of 6 can be easily reached with a relatively low doping level."
(b) The reason of DFT overestimation is explained in the third paragraph of page 9: "The experimental Seebeck coefficients of n-type Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x samples at 300 K or 725 K are larger than the Seebeck value calculated by a single band model using the DOS effective mass of the ML band [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]. This result not only confirms the theoretical calculation that the ML band dominates the conduction band minimum of Mg 3 Sb 1.5 Bi 0.5 , but also reveals that the K band located 0.12 eV above the conduction band minimum makes a contribution to both the room-temperature and high-temperature electrical transports. Since there is an energy difference of 0.12 eV between the ML band and the K band in n-type Mg 3 Sb 1.5 Bi 0.5 , the experimental Seebeck coefficients of n-type Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x are smaller than the calculated value by DFT for n-type Mg 3 Sb 2 with the effective convergence of the two bands (see Figs 3a,b and 2c)."
(c) The second paragraph of page 10: "The carrier concentration of Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x increases with increasing temperature and reaches 4.83 × 10 19 cm -3 at 725 K in the high-performance sample with x = 0.04 [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] , suggesting that the Fermi level will move upward approaching the K band with rising temperature. Additionally, the broadening of the Fermi distribution makes the Fermi level easier to reach the K band at high temperatures. The temperaturedependent DOS effective mass in Te-doped Mg 3 Sb 1.5 Bi 0.5 are illustrated in . As shown in , the DOS effective mass of Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x=0.04 and 0.05) derived from the experimental Seebeck coefficient increases with increasing temperature at 400-725 K, ruling out the single band behavior at high temperatures. The above results again prove that the multiple band behavior, including the effects from both the ML band and the K band, makes a contribution to the high temperature transport properties."
(d) One sentence is added to the caption of to explain the overestimation of DFT modelling: "Our experimental data lie below the curve by DFT for Mg 3 Sb 2 , which is because there is an energy difference of 0.12 eV between the ML band and K band in n-type Mg 3 Sb 1.5 Bi 0.5 while these two bands are nearly converged in Mg 3 Sb 2 (see [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]."
(e) Two sentences are added to the caption of : "The band structure of Mg 3 Sb 1.5 Bi 0.5 depicts a multiple band behavior similar to that of Mg 3 Sb 2 [fig_ref] f: To the first paragraph of page 16, a few key words are... [/fig_ref] , where the ML band possesses a six-fold valley degeneracy and the K band has a two-fold valley degeneracy. However, the ML band in Mg 3 Sb 1.5 Bi0.5 becomes the conduction band minimum, which is about 0.12 eV below the K band." The same impression makes also the summary. Here, the multi-valley approach is presented as a basis for the materials design, which was not the issue in the presented manuscript.'
## Reply
In the revised manuscript, the overestimation of theoretical calculation is explained. Six-fold valley degeneracy, which is another expression of a valley degeneracy of 6, is added in the caption of . Please see the reply to Comment 4 for details.
Multi-valley band behavior was explained in detail for n-type Mg 3 Sb 2 . Several methods, including ab initio band structure, Fermi surface, and the dependence of Seebeck coefficient and power factor on carrier concentration, have been used to demonstrate the multiple band behavior in n-type Mg 3 Sb 2 . The multiple conduction band feature of Mg 3 Sb 1.5 Bi 0.5 is similar to that of Mg 3 Sb 2 . They have the ML band with a six-fold valley degeneracy and the K band with a two-fold valley degeneracy. The only difference, however, is in Mg 3 Sb 1.5 Bi 0.5 the ML band moves downwards 0.12 eV below the K band, which makes the ML band easier to reach at a low doping level. To make the multiple band behavior clearer, a few modifications are applied. Please see the reply to Comment 4 for details. Thanks.
## Assume that the authors are in the process of preparing more samples for a more complete paper and a communication is warranted with the combination of the theoretical support.'
Reply Thanks for the suggestion. We agree and that is why we originally submitted the paper based on a single sample. However, we also acknowledge that the study is more definitive when based on multiple samples and as explained above we have in the revised manuscript added data for more samples with different compositions. The progression to high zT can be seen in [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] in the revised manuscript. Please see the reply to the Comment 1 of the first reviewer for details.
## Reviewer 3
Comment 1 '1. There are a variety of zT values depending on the method of sample preparation. In this sense, the authors need to show their own data on pristine Mg 3 Sb 2 and Mg 3 Sb 1.5 Bi 0.5 samples, which need to be compared with their own theoretical results for p type cases.'
## Reply
Thank you for the suggestion. We agree that the zT value depends on the method of sample preparation, and this is why we carefully chose the reported data (RSC Adv. 3, 8504-8516 (2013); Acta Mater. 93, 187-193 (2015)) using the similar synthesis method (i.e. Mechanical milling followed by SPS press) for comparison. However, to additionally address the concern, we have prepared one Mg 3 Sb 2 sample using our synthesis method. The obtained zT value by our synthesis method is very similar to the reported data cited in the manuscript (see [fig_ref] f: To the first paragraph of page 16, a few key words are... [/fig_ref].
We would like to stress that the focus of this work is on n-type Mg 3 Sb 2based materials rather than p-type. Therefore, we think using the previous reported data is reasonable and does not affect the present work. [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]. As shown in the figure, doping tellurium on the anion sites (Sb or Bi sites) in Mg 3 Sb 1.5 Bi 0.5 has lower formation energy than that in Mg 3 Sb 2 . We prepared our samples with the nominal compositions Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x (x=0.04, 0.05, 0.08, 0.20) since initially we do not know if Te will prefer to substitute Sb atoms or Bi atoms. In addition, quantitative elemental analysis of the high-performance pellet with x = 0.05 was carried out by SEM-EDS (see . We find that the actual composition is close to the nominal composition, and the actual composition shows slightly less Bi and more Sb relative to the nominal composition. This can be explained by the assumption that more Te has been doped on the Bi sites than on the Sb sites due to the lower formation energy of Te Bi . The above result indicates that in the future we could try synthesis of n-type Mg 3 Sb 1.5 Bi 0.5-x Te x , which might be more effective. We have made the following revisions to address this comment:
(a) A few sentences are added in the beginning of second paragraph on page 9: "Achieving n-type Mg 3 Sb 2 by doping tellurium on the anion sites has been attempted and found to be difficult, while it is easier in Mg 3 Sb 2-x Bi x solid solutions. This is probably due to the lower formation energy of tellurium doping on the anion sites of Mg 3 Sb 2-x Bi x solid solutions (see one example in [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]."
(b) A few sentences are added in the second paragraph on page 14: "Quantitative elemental analysis of the high-performance pellet with x = 0.05 was carried out on FEI Nova Nano SEM 600 equipped with an element EDS X-ray detector. The result shown in was the average value from five randomly selected areas of the pellet."
(c) [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] The actual composition estimated by SEM-EDS analysis for the high-performance Mg 3 Sb 1.5-0.5x Bi 0.5-0.5x Te x sample with x = 0.05. The actual composition by SEM-EDX is close to the nominal composition. The actual composition shows less Bi and more Sb relative to the nominal composition. This is possibly because that more Te has been doped on the Bi sites than on the Sb sites due to the lower formation energy of Te Bi (see [fig_ref] f: To the first paragraph of page 16, a few key words are... [/fig_ref]." '4. I believe that the authors did not consider temperature effect in their DFT calculations. Nevertheless, they compare the theoretical data of ntype Mg 3 Sb 2 with the experimental data taken at 300 K (see Figs. 2a and 2b), and the theoretical data of n-type Mg 3 Sb 1.5 Bi 0.5 with the experimental data taken at 725 K (see . If one considers different temperature, for example, 300 K in and 725 K in Figs. 2a and 2b, the experimental data do not agree with the simulated curves.'
## Reply
This appears to be a misunderstanding. The temperature effect is included in the electrical transport calculation by semi-classical Boltzman transport theory under the constant scattering time approximation (CSTA). Under CSTA, it is assumed that the carrier scattering time τ determining the electrical conductivity will not vary strongly with temperature and doping level. This approach has been successfully applied to predict the Seebeck coefficient and the trend of the electrical conductivity or power factor for a variety of thermoelectric materials (J. Concerning the reason why the DFT curve looks overestimated. This is because the DFT curve is simulated for n-type Mg 3 Sb 2 with nearly converged ML band and K band, while the experimental data is for n-type Mg 3 Sb 1.5 Bi 0.5 , which has the ML band lies 0.12 eV below the K band. Please see the reply to Comment 4 of the first reviewer for details. Thanks.
Comment 5 '5. According to their n_H and mu_H data; when simple looking, n_H is almost 2 times increased with temperature and mu_H is almost 3 times decreased with temperature, the resistivity should be increased about 1.5 times, which seems to be inconsistent with [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref].'
## Reply
The inconsistency is caused by different resistivity measurement methods between our homebuilt setup and the commercial ZEM-3 setup. The mobility was calculated from the Hall coefficient and resistivity measured on the full pellet using four-point Van der Pauw method in our home built setup (Rev. . The resistivity shown in [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref] (a) [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref] in the main text are modified using the data measured by our homebuilt setups on the same pellet. (e) One sentence is added in line 13 of page 15: "The thermoelectric zT, obtained on the high-performance sample with x = 0.05 by the homebuilt system and ZEM-3 setup, shows good consistency between each other [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]." Comment 6 '6. Also, the Seebeck coefficient at 725 K in [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref] seems to be about 280 microV/K, but the data point in is about 300 microV/K. They are not consistent each other.'
## Reply
We have checked the Seebeck coefficients at 725 K in [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref] and in the initial version of manuscript and confirmed that they are consistent with each other (282 microV/K). Please note that the scales of [fig_ref] Comment 4 ': The multi-valley aspect of the band structure [/fig_ref] and are not the same.
As explained above in the revised manuscript, we have used the electrical transport data measured on our homebuilt setups in the main text. The ZEM-3 data have been put in the Supplementary Information for comparison (see the reply to Comment 5).
Comment 7 '7. The authors need to explain the reason on thermal hysteresis of Seebeck coefficient and electrical resistivity (Figs. S7a and S7b). If it originates from annealing effect and/or slight oxidation, the hysteresis may not be reversible.'
## Reply
Although the transport properties show hysteresis, the hysteresis can be repeated very well in two different cycles (see [fig_ref] f: To the first paragraph of page 16, a few key words are... [/fig_ref]. Moreover, the thermoelectric zT obtained on the high-performance sample with x = 0.05 by the home-built system and the commercial ZEM-3 setup shows excellent consistency between each other [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]. The good repeatability and reproducibility of the high zT observed in two different setups indicate that the hysteresis in the sample is likely caused by reversible processes, and that it is not due to an annealing effect or oxidation. If the hysteresis is caused by annealing or oxidation, the transport properties curves will not be repeated in two different cycles and the high zT will not be reproduced by two different setups.
We stress that the focus of the present work is proof-of-concept of multivalley conduction bands in n-type Mg 3 Sb 2 -based compounds. We have demonstrated that the hysteresis does not affect the high performance in the n-type sample, and exploring the exact the reason of thermal hysteresis is beyond the scope of the present work. We are planning to perform in-situ X-ray diffraction on n-type samples for a comprehensive study of structure-property relation, which might explain the hysteresis in the future.
(a) The comparison of zT values measured on ZEM-3 and homebuilt setup is plotted in [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]. [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] is added in the [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] (b) A few sentences are added to the Supplementary Note 2: "The transport properties of the two cycles are consistent upon the repeated heating and cooling measurements. In addition, the thermoelectric zT, obtained on the high-performance sample with x = 0.05 by the home-built system and ZEM-3 setup, shows excellent consistency between each other [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref]. The good repeatability and reproducibility observed in different setups indicate that the hysteresis in the sample is more likely caused by reversible processes."
## Summary of the changes (marked in red in the revised manuscript)
23. A few sentences are modified in the end of page 12 and in the beginning of page 13. 33. One reference is added on page 19.. [fig_ref] Figure A: zT value of p-type Mg 3 Sb 2 by our synthesis method... [/fig_ref] ,c,d, and 4 are modified. The order of and 3d is changed. Once again we would like to thank the reviewers for the time they have devoted to our manuscript. Their very insightful comments have significantly improved our manuscript and we look much forward to your decision.
[fig] Comment 4 ': The multi-valley aspect of the band structure (and the whole calculational part of the work) is not really explained and looks overestimated. There is not clear from the manuscript how from the spectral-weight representation of the band structure single curve-representation for the Seebeck coefficient was obtained. The conclusion which can be drawn from figure 3 are not clear for the reader.' [/fig]
[fig] f: To the first paragraph of page 16, a few key words are added to make it clear: "Electrical transport property calculations of Mg 3 Sb 2 (the DFT curves in Figs 2a,b and 3c) were carried out by combining the ab initio band structure calculations and the Boltzmann transport theory under the constant carrier scattering time approximation as implemented in the BoltzTraP code 34 (Supplementary Note 3 andSupplementary Fig. 10). To calculate the power factor curves of Mg 3 Sb 2 shown inFig. 2b, we need to estimate the carrier scattering time. Calculation details of the constant carrier scattering time τ are provided in In the abstract, the experimental part is suppressed, the calculational aspect is overestimated and presented in the form of the finally proven facts. In the manuscript, the 'six-fold valley degeneracy' is not even explicitly mentioned, e.g. in the comments to the figure 3. [/fig]
|
Host Factors Involved in the Intracellular Movement of Bamboo mosaic virus
Viruses move intracellularly to their replication compartments, and the newly synthesized viral complexes are transported to neighboring cells through hijacking of the host endomembrane systems. During these processes, numerous interactions occur among viral proteins, host proteins, and the cytoskeleton system. This review mainly focuses on the plant endomembrane network, which may be utilized by Bamboo mosaic virus (BaMV) to move to its replication compartment, and summarizes the host factors that may be directly involved in delivering BaMV cargoes during intracellular movement. Accumulating evidence indicates that plant endomembrane systems are highly similar but exhibit significant variations from those of other eukaryotic cells. Several Nicotiana benthamiana host proteins have recently been identified to participate in the intracellular movement of BaMV. Chloroplast phosphoglycerate kinase, a host protein transported to chloroplasts, binds to BaMV RNAs and facilitates BaMV replication. NbRABG3f is a small GTPase that plays an essential role in vesicle transportation and is also involved in BaMV replication. These two host proteins may deliver BaMV to the replication compartment. Rab GTPase activation protein 1, which switches Rab GTPase to the inactive conformation, participates in the cell-to-cell movement of BaMV, possibly by trafficking BaMV cargo to neighboring cells after replication. By analyzing the host factors involved in the intracellular movement of BaMV and the current knowledge of plant endomembrane systems, a tentative model for BaMV transport to its replication site within plant cells is proposed.
# Introduction
Membrane trafficking delivers materials between the endomembrane system and the plasma membrane and therefore plays an essential role in cell survival and development [bib_ref] Membrane trafficking: intracellular highways and country roads, Cheung [/bib_ref]. Many animal microbes that reproduce intracellularly, including viruses and bacteria, have been shown to utilize host endomembrane trafficking and the autophagy system for intracellular transport within the host cells [bib_ref] Endocytosis of viruses and bacteria, Cossart [/bib_ref] [bib_ref] Principles of Virus Uncoating: cues and the Snooker Ball, Yamauchi [/bib_ref] , whereas other microbes alter the degradation pathway used by the host cells to destroy the intruding pathogens [bib_ref] Modification of intracellular membrane structures for virus replication, Miller [/bib_ref] [bib_ref] Principles of Virus Uncoating: cues and the Snooker Ball, Yamauchi [/bib_ref]. Rab proteins are small GTPases involved in membrane trafficking of cells, specifically in vesicle formation and fusion. Recent studies of Salmonella enterica Typhimurium, an animal pathogenic bacterium, have identified several bacterial effectors that target various Rab proteins for replication within the host cells [bib_ref] Proteolytic targeting of Rab29 by an effector protein distinguishes the intracellular compartments..., Spano [/bib_ref] [bib_ref] A bacterial pathogen targets a host rab-family GTPase defense pathway with a..., Spano [/bib_ref] [bib_ref] Salmonella inhibits retrograde trafficking of mannose-6-phosphate receptors and lysosome function, Mcgourty [/bib_ref] [bib_ref] Salmonella disrupts host endocytic trafficking by SopD2-mediated inhibition of Rab7, D'costa [/bib_ref].
Many plant viruses induce membrane remodeling after infecting their host cells. Studies focusing on the modification of endomembrane systems induced by plant viruses have substantially improved our understanding of the intracellular movement of plant viruses. Membrane targeting or recruitment by viral proteins in the viral replication compartment is often the cause of endomembrane remodeling, as has been demonstrated for many viruses [bib_ref] Sequential recruitment of the endoplasmic reticulum and chloroplasts for plant potyvirus replication, Wei [/bib_ref] [bib_ref] The SNARE protein Syp71 is essential for turnip mosaic virus infection by..., Wei [/bib_ref] [bib_ref] Host and viral RNA-binding proteins involved in membrane targeting, replication and intercellular..., Hyodo [/bib_ref] [bib_ref] Expanding use of multi-origin subcellular membranes by positive-strand RNA viruses during replication, Xu [/bib_ref]. Grapevine fanleaf nepovirus (GFLV) infection induces structural changes in the endoplasmic reticulum (ER) membranes to allow virus replication in the ER-derived membrane. Although host factors have not been identified, inhibitor treatment has demonstrated that vesicle transport between Golgi and ER is essential for GFLV replication [bib_ref] Grapevine fanleaf virus replication occurs on endoplasmic reticulumderived membranes, Ritzenthaler [/bib_ref] [bib_ref] A renaissance in nepovirus research provides new insights into their molecular interface..., Fuchs [/bib_ref]. A study on plant potyviruses revealed that the 6k2 viral protein of Turnip mosaic virus (TuMV) induces the formation of vesicles derived from ER and targets them to the chloroplasts for replication [bib_ref] Sequential recruitment of the endoplasmic reticulum and chloroplasts for plant potyvirus replication, Wei [/bib_ref]. Melon necrotic spot virus (MNSV) has been shown to replicate in mitochondria, which is significantly altered by the virus-encoded p29 protein targeting the mitochondria membrane [bib_ref] Induction of necrosis via mitochondrial targeting of Melon necrotic spot virus replication..., Mochizuki [/bib_ref] [bib_ref] Melon necrotic spot virus replication occurs in association with altered mitochondria, Frontiers In Microbiology | Www.Frontiersin [/bib_ref]. ER disorder has also been observed, and the p7B movement protein of MNSV is localized to ER, Golgi, actin filaments, and plasmodesmata. Disruption of the transport between ER and Golgi results in the accumulation of p7B within the ER. Therefore, the ER-to-Golgi secretory pathway could be involved in the intra-and intercellular movement of MNSV [bib_ref] The Intra-and intercellular movement of Melon necrotic spot virus (MNSV) depends on..., Genoves [/bib_ref].
In addition to the viral encoded proteins, host factors are also involved in the membrane remodeling process. Using yeast as a model system and by performing further testing in host plants, several studies on Tomato bushy stunt virus (TBSV) have demonstrated that host factors are responsible for delivering components to remodel the membrane-associated compartment for viral replication [bib_ref] Co-opted oxysterol-binding ORP and VAP proteins channel sterols to RNA virus replication..., Barajas [/bib_ref] [bib_ref] Enrichment of Phosphatidylethanolamine in viral replication compartments via Co-opting the Endosomal Rab5..., Xu [/bib_ref]. Heat shock protein 70 is associated with the replication complex of TBSV and facilitates the insertion of viral replication proteins into the yeast membrane [bib_ref] A key role for heat shock protein 70 in the localization and..., Wang [/bib_ref]. The GTP-bound active form of Rab5-positive endosome is hijacked by TBSV for enrichment of phosphatidylethanolamine at the replication site [bib_ref] Enrichment of Phosphatidylethanolamine in viral replication compartments via Co-opting the Endosomal Rab5..., Xu [/bib_ref]. A host SNARE protein, Syp71, mediates the fusion between the TuMV-induced vesicles and chloroplasts, which is required for TuMV infection [bib_ref] The SNARE protein Syp71 is essential for turnip mosaic virus infection by..., Wei [/bib_ref]. In yeast, membrane-shaping reticulon homology domain proteins are crucial for the formation of the replication compartment induced by the Brome mosaic virus (BMV) 1a protein [bib_ref] Membrane-shaping host reticulon proteins play crucial roles in viral RNA replication compartment..., Diaz [/bib_ref]. These findings indicate that membrane trafficking and targeting are essential processes for plant virus replication.
Several pathways of intracellular movement have been proposed for animal viruses after they enter the host cells. By contrast, the trafficking pathways for the intracellular movement of plant viruses to their replication sites within the host cells remain largely unknown. Particularly, the endomembrane trafficking systems in plants seem to be more complicated and have not been completely revealed [bib_ref] Chapter 4: functions of RAB and SNARE proteins in plant life, Saito [/bib_ref] [bib_ref] Physiological roles of plant post-golgi transport pathways in membrane trafficking, Uemura [/bib_ref]. Moreover, studies of the intracellular movement of plant viruses have mostly focused on cell-to-cell movement through the plasmodesmata; these intracellular movement pathways of plant viruses have been reviewed in several articles [bib_ref] Understanding the intracellular trafficking and intercellular transport of potexviruses in their host..., Park [/bib_ref] [bib_ref] Plant virus replication and movement, Heinlein [/bib_ref] [bib_ref] Viral elements and host cellular proteins in intercellular movement of Bamboo mosaic..., Liou [/bib_ref]. Host factors participating in membrane trafficking or protein targeting may play roles in delivering BaMV or its cargoes to the replication sites. Based on the current knowledge of intracellular trafficking pathways in plants, a model for the intracellular movement of BaMV to its replication compartment is proposed. After replication, plant viruses travel intracellularly to reach plasmodesmata for cell-to-cell movement. A vesicle trafficking-related host protein participates in BaMV cell-to-cell movement; its potential role in BaMV intracellular movement is also discussed in this review.
## Possible replication compartments for bamv
Virus infection commonly induces the formation of dynamic membrane-associated structures that are associated to the virus replication and movement [bib_ref] Host endomembrane recruitment for plant RNA virus replication, Grangeon [/bib_ref] [bib_ref] Plant virus replication and movement, Heinlein [/bib_ref]. Chloroplasts are one of the types of compartments suitable for plant virus replication [bib_ref] Sequential recruitment of the endoplasmic reticulum and chloroplasts for plant potyvirus replication, Wei [/bib_ref] [bib_ref] Chloroplast in plantvirus interaction, Zhao [/bib_ref]. Using BaMV 3 RNA as a probe for in situ hybridization through electronic microscopy, BaMV viral RNAs were detected within several organelles of green bamboo leaf cells, including chloroplasts, mitochondria, and nuclei [bib_ref] Post-embedding in situ hybridization for localization of viral nucleic acid in ultra-thin..., Lin [/bib_ref]. Phage MS2 coat protein can specifically bind to its own RNA sequence, and viral genomic RNA engineered to contain the MS2 sequence can be traced within the cells through the binding of GFPfused MS2 coat protein [bib_ref] A novel procedure for the localization of viral RNAs in protoplasts and..., Zhang [/bib_ref]. Recently, through confocal microscopy, BaMV viral RNA expressing the phage MS2 coat protein binding sequence was found to localize within chloroplasts. Furthermore, negative-strand BaMV RNA was found in the isolated chloroplasts, demonstrating that chloroplasts may be among the BaMV replication compartments within the host cells [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref].
In contrast, Potato virus X (PVX), another potexvirus, has been found to replicate in the ER membrane [bib_ref] Synthesis of potato virus X RNAs by membrane-containing extracts, Doronin [/bib_ref] [bib_ref] Understanding the intracellular trafficking and intercellular transport of potexviruses in their host..., Park [/bib_ref] ; in addition, TGB2/3 has recently been shown to remodel the ER membrane at plasmodesmata, where PVX coupled the replication and movement to the neighboring cells [bib_ref] Replication and trafficking of a plant virus are coupled at the entrances..., Tilsner [/bib_ref].
## Host factors probably involved in the intracellular targeting of bamv to the replication compartment
After entering the host cells, similar to other viruses, BaMV must travel to its replication site intracellularly. Recently, several host factors involved in BaMV replication have been identified through copurification with the BaMV replicase complex [bib_ref] Suppression of Bamboo mosaic virus accumulation by a putative methyltransferase in Nicotiana..., Cheng [/bib_ref] [bib_ref] Glyceraldehyde 3-phosphate dehydrogenase negatively regulates the replication of Bamboo mosaic virus and..., Prasanth [/bib_ref] [bib_ref] Hsp90 interacts specifically with viral RNA and differentially regulates replication initiation of..., Huang [/bib_ref] [bib_ref] Promotion of Bamboo Mosaic Virus accumulation in Nicotiana benthamiana by 5'->3' Exonuclease..., Lee [/bib_ref] , binding to the viral RNAs [bib_ref] Chloroplast phosphoglycerate kinase, a gluconeogenetic enzyme, is required for efficient accumulation of..., Lin [/bib_ref] [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref] , and cDNA-amplified fragment length polymorphism (cDNA-AFLP) analysis [bib_ref] A putative Rab-GTPase activation protein from Nicotiana benthamiana is important for Bamboo..., Huang [/bib_ref] [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref]. Among these host factors, three are possibly involved in the transportation of BaMV or its related cargoes within host cells. Chloroplast phosphoglycerate kinase (chl-PGK) [bib_ref] Chloroplast phosphoglycerate kinase, a gluconeogenetic enzyme, is required for efficient accumulation of..., Lin [/bib_ref] [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref] and NbRABG3f [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref] are related to BaMV replication, whereas NbRabGAP1 is essential for the intercellular movement of BaMV [bib_ref] A putative Rab-GTPase activation protein from Nicotiana benthamiana is important for Bamboo..., Huang [/bib_ref].
In Nicotiana benthamiana, chl-PGK, a nuclear-encoded chloroplast protein, was found to bind to the 3 untranslated region (3 UTR) of BaMV genomic RNAs [bib_ref] Chloroplast phosphoglycerate kinase, a gluconeogenetic enzyme, is required for efficient accumulation of..., Lin [/bib_ref]. Knocking down the expression or mistargeting of chl-PGK reduced BaMV accumulation in N. benthamiana protoplasts, indicating that chl-PGK participates in BaMV replication [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref]. Redirecting a BaMV RNA binding protein, EF1a, to chloroplasts rescued the reduction of BaMV caused by the knock-down of chl-PGK. These results demonstrate that chl-PGK likely assists in the targeting of BaMV to the chloroplasts for replication [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref]. Recently, a chl-PGK from Arabidopsis thaliana was identified as a requirement for efficient Watermelon mosaic virus (WMV) [bib_ref] Cloning of the Arabidopsis rwm1 gene for resistance to Watermelon mosaic virus..., Ouibrahim [/bib_ref] and Plum pox virus [bib_ref] Allelic variation at the rpv1 locus controls partial resistance to Plum pox..., Poque [/bib_ref] infection. Although the replication compartment of WMV is unknown, the 6K protein of other potyviruses induces the formation of mobile vesicles and transports the viruses from the ER to the chloroplast membrane for viral replication [bib_ref] Sequential recruitment of the endoplasmic reticulum and chloroplasts for plant potyvirus replication, Wei [/bib_ref]. Therefore, chl-PGK may be involved in the intracellular transport of different plant viruses to the chloroplasts for replication.
A recent study identified that NbRABG3f, a Rab small GTPase, is involved in BaMV replication [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref]. Rab small GTPases are involved in vesicle trafficking within cells. Rab proteins alternate between a GTP-bound active form and GDPbound inactive form, and this switching is accelerated by their regulatory proteins [bib_ref] Regulation of small GTPases by GEFs, GAPs, and GDIs, Cherfils [/bib_ref]. GTP-bound Rab proteins bud from the donor compartment and fuse with the acceptor compartment, where GTP is hydrolyzed by GTPaseactivating proteins (GAPs) [bib_ref] Regulation of small GTPases by GEFs, GAPs, and GDIs, Cherfils [/bib_ref]. Based on sequence comparison, NbRABG3f is homologous to animal cell Rab7 protein and to A. thaliana RabG3f [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref]. cDNA-AFLP analysis revealed that NbRABG3f expression was upregulated after BaMV inoculation [bib_ref] Identification of differentially expressed genes induced by Bamboo mosaic virus infection in..., Cheng [/bib_ref]. The GDP-bound form of Rab GTPase can be used to trace the donor compartment of Rab proteins [bib_ref] Dissecting virus entry via endocytosis, Sieczkarski [/bib_ref] ; thus, confocal microscopy revealed that the GDP-bound form mutant of NbRABG3f was localized to the Golgi compartment [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref]. In plant endomembrane trafficking systems, endocytotic materials are transferred to the trans-Golgi network for further sorting [bib_ref] Endocytic and autophagic pathways crosstalk in plants, Zhuang [/bib_ref]. As a plant defense mechanism pathway, intruding pathogens are likely delivered to the multi-vesicular body (MVE)/prevacuolar compartment (PVC), or autophagosome and then delivered to the vacuoles for degradation [bib_ref] Membrane trafficking and autophagy in pathogen-triggered cell death and immunity, Teh [/bib_ref]. Successful pathogen infection results from a redirection of the pathway to their replication compartments rather than the degradation pathway [bib_ref] Hijack it, change it: how do plant viruses utilize the host secretory..., Patarroyo [/bib_ref] [bib_ref] Autophagy and viruses: adversaries or allies?, Dong [/bib_ref]. According to the finding that NbRABG3f is derived from the Golgi compartment [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref] and that chloroplasts are likely to be the replication sites of BaMV [bib_ref] Post-embedding in situ hybridization for localization of viral nucleic acid in ultra-thin..., Lin [/bib_ref] [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref] , BaMV may utilize NbRABG3f or NbRABG3f-associated vesicles for transport to chloroplasts. However, the acceptor membrane of NbRABG3f has not yet been identified. Knowledge of the destination of NbRABG3f in the endomembrane systems will verify whether BaMV hijacks NbRABG3f and redirects its route or instead utilizes NbRABG3f-associated vesicle and follows its pathway. Further investigation of the interaction of BaMV with NbRABG3f and other Rab proteins and further evaluation of the colocalization of BaMV and various marker proteins involved in endomembrane trafficking will reveal the intracellular transportation pathway for BaMV.
## Possible model for bamv intracellular movement to its replication compartment
In uninfected N. benthamiana cells, NbRABG3f and its associated vesicles are generated by the Golgi compartment [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref]. The destination of NbRABG3f is currently unknown. According to a putative model proposed for Arabidopsis, the endocytosed materials can be (1) delivered to the MVE/PVC, or autophagosome and then transported to the vacuoles for degradation, or (2) recycled into the plasma membrane [bib_ref] Physiological roles of plant post-golgi transport pathways in membrane trafficking, Uemura [/bib_ref] [bib_ref] Tethering complexes in the, Vukasinovic [/bib_ref]. A previous study demonstrated that Arabidopsis AtRABG3f mediates transport from PVC to the vacuole [bib_ref] Activation of the Rab7 GTPase by the MON1-CCZ1 complex is essential for..., Cui [/bib_ref]. Although the donor membrane of NbRABG3f is different from that of AtRABG3f, possibly because of variation in the C-terminal sequence, based on the fact that NbRABG3f is highly homologous to AtRABG3f [bib_ref] NbRABG3f, a member of Rab GTPase, is involved in Bamboo mosaic virus..., Huang [/bib_ref] , NbRABG3f likely participates in the transport of vesicles to other endomembrane systems rather than in their delivery to the recycling pathway.
According to current knowledge of the intracellular trafficking pathways in plants and studies of NbRABG3f and chl-PGK, BaMV may utilize NbRABG3f-associated vesicles for vesicle trafficking. During vesicle trafficking, chl-PGK is recruited and assists in the targeting of the BaMV complex to the chloroplasts, which are one of the types of BaMV replication compartments [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref]. By contrast, similar to the role of Rab5 protein in TBSV replication [bib_ref] Enrichment of Phosphatidylethanolamine in viral replication compartments via Co-opting the Endosomal Rab5..., Xu [/bib_ref] , NbRABG3f may deliver the materials required for BaMV replication to the chloroplasts, and chl-PGK may bind BaMV RNA and direct the BaMV complex to the chloroplast for replication .
Although the transportation pathway for chloroplasttargeting proteins to the chloroplast is not completely clear, to date, several pathways have been proposed for proteins transported to the chloroplasts after translation [bib_ref] Evidence for an ER to Golgi to chloroplast protein transport pathway, Radhamony [/bib_ref] [bib_ref] The chloroplast protein import system: from algae to trees, Shi [/bib_ref]. Most chloroplast proteins contain an N-terminal cleavable transit peptide that mediates the FIGURE 1 | A possible model for BaMV intracellular movement to the chloroplasts. In the normal membrane-trafficking pathway, NbRABG3f-associated vesicles bud from Golgi compartment. There are two possible pathways that NbRABG3f and chl-PGK may participate in delivering of BaMV cargoes to the chloroplasts. (1) When BaMV infects the cells, it may hijack the NbRABG3f-associated vesicles and redirect the vesicles to the chloroplast which is one of its replication compartments. During the transportation to the chloroplasts, chl-PGK is recruited to the BaMV complex and targets BaMV complex to the chloroplasts stroma (Upper).
(2) Alternatively, NbRABG3f-associated vesicles may deliver materials required for BaMV replication to the chloroplasts and chl-PGK binds to BaMV viral RNA and targets it to the chloroplast stroma for replication (Lower).
targeting of the protein to the chloroplasts [bib_ref] The chloroplast protein import system: from algae to trees, Shi [/bib_ref]. N. benthamiana chl-PGK also contains a putative N-terminal transit peptide [bib_ref] Chloroplast phosphoglycerate kinase is involved in the targeting of Bamboo mosaic virus..., Cheng [/bib_ref] , indicating that it can be transported from the ER to the chloroplast stroma, and that such transport is directed by the N-terminal transit peptide. Whether chl-PGK is recruited to the NbRABG3f-associated vesicle or these two host proteins are involved in separate steps of BaMV transportation is an interesting question that remains to be explored. Future studies should investigate whether chl-PGK interacts with NbRABG3f or NbRABG3f-associated vesicles during the transportation process.
## A host factor possibly involved in bamv vesicle trafficking to the plasmodesmata for intercellular movement
After the replication of plant viruses, virion or virus ribonucleoprotein (vRNP) facilitates their intercellular movement through the plasmodesmata. Unlike PVX, which moves intercellularly as vRNP, BaMV is likely to move as a virion associated with TGBp2 and TGBp3-based ER membrane [bib_ref] The stable association of virion with the triple-gene-block protein 3-based complex of..., Chou [/bib_ref]. Several BaMV viral proteins and N. benthamiana host factors have been demonstrated to facilitate the intercellular movement, but not the replication, of BaMV [bib_ref] Ser/Thr kinase-like protein of Nicotiana benthamiana is involved in the cell-to-cell movement..., Cheng [/bib_ref] [bib_ref] A putative Rab-GTPase activation protein from Nicotiana benthamiana is important for Bamboo..., Huang [/bib_ref]. A recent, thorough review of the intercellular movement of BaMV hypothesizes the possible roles of the viral and host proteins in the intercellular movement of BaMV [bib_ref] Viral elements and host cellular proteins in intercellular movement of Bamboo mosaic..., Liou [/bib_ref]. Among the host proteins involved in the intercellular movement of BaMV, NbRabGAP1, a Rab-GTPase activation protein, has been identified through cDNA-AFLP analysis, and its expression is upregulated after BaMV inoculation [bib_ref] A putative Rab-GTPase activation protein from Nicotiana benthamiana is important for Bamboo..., Huang [/bib_ref]. RabGAP contains the TBC (Tre2/Bub2/Cdc16) catalytic domain that can promote GTP hydrolysis and thus inactivates Rab proteins [bib_ref] Illuminating the functional and structural repertoire of human TBC/RABGAPs, Frasa [/bib_ref]. Rab proteins participate in the regulation of vesicle formation and trafficking in the endomembrane system. In the study of NbRabGAP1, low NbRabGAP1 expression reduced BaMV accumulation in N. benthamiana leaves, but not in protoplasts, whereas overexpression of NbRabGAP1 exerted the opposite effect. Based on these results, it is hypothesized that NbRabGAP1 is involved in the delivery of the BaMV/BaMVrelated cargo from the virus replication complex to neighboring cells [bib_ref] A putative Rab-GTPase activation protein from Nicotiana benthamiana is important for Bamboo..., Huang [/bib_ref] [bib_ref] Viral elements and host cellular proteins in intercellular movement of Bamboo mosaic..., Liou [/bib_ref]. An attempt to examine the interaction between NbRabGAP1 and NbRABG3f was unsuccessful (unpublished data). This outcome was expected because NbRabG3f participates in BaMV replication, whereas NbRabGAP1 assists in the intercellular movement of BaMV. Similar to PVX, the BaMV movement proteins TGBp2 and TGBp3 are ER-targeting membrane proteins [bib_ref] Topological properties of the triple gene block protein 2 of Bamboo mosaic..., Hsu [/bib_ref]. The BaMV infectious complex has been proposed to move from perinuclear ER-derived membrane-bound bodies (MBB) to the plasmodesmata [bib_ref] Viral protein targeting to the cortical endoplasmic reticulum is required for cell-cell..., Wu [/bib_ref] [bib_ref] Viral elements and host cellular proteins in intercellular movement of Bamboo mosaic..., Liou [/bib_ref]. After BaMV replication, NbRabGAP1 possibly participates in the delivery of BaMV cargoes with an unknown Rab protein toward the plasmodesmata through ER and post-ER secretory pathways. Alternatively, NbRabGAP1 may recycle the BaMV movement proteins from plasmodesmata to assist in the next round of BaMV transportation [bib_ref] A putative Rab-GTPase activation protein from Nicotiana benthamiana is important for Bamboo..., Huang [/bib_ref]. However, the connection between chloroplasts and the MBB requires further investigation to unveil the intracellular route after BaMV replication.
Studies in PVX have indicated that its TGB2/3 movement proteins induce ER-derived granular vesicles, which are essential for PVX cell-to-cell movement [bib_ref] Mutations in the central domain of potato virus X TGBp2 eliminate granular..., Ju [/bib_ref] ; therefore, the PVX movement complex may be transported through these vesicles to plasmodesmata [bib_ref] Varied movement strategies employed by triple gene block-encoding viruses, Verchot-Lubicz [/bib_ref]. Accordingly, as yet unidentified host factors may be present that function as Rab or RabGAP proteins to facilitate PVX movement.
The kinases CK2 and NbSTKL are other host factors also involved in the cell-to-cell movement of BaMV [bib_ref] Ser/Thr kinase-like protein of Nicotiana benthamiana is involved in the cell-to-cell movement..., Cheng [/bib_ref]. Knowledge on host factors that affect the intercellular movement of potexviruses have been reviewed recently [bib_ref] Understanding the intracellular trafficking and intercellular transport of potexviruses in their host..., Park [/bib_ref] [bib_ref] Viral elements and host cellular proteins in intercellular movement of Bamboo mosaic..., Liou [/bib_ref]. Their functions in assisting BaMV or other potexvirus movement have also been discussed in a recent review, which reported that they might not directly participate in the transportation of BaMV cargoes [bib_ref] Viral elements and host cellular proteins in intercellular movement of Bamboo mosaic..., Liou [/bib_ref]. Therefore, they are not included in this review.
## Summary
In this review, I summarized the host factors that participate in membrane trafficking and a specific chloroplast targeting protein that may participate in BaMV transportation within the cells. A possible model was proposed to demonstrate how BaMV is delivered to its replication compartment. One of the host factors, NbRABG3f, is a small GTPase that mediates vesicle trafficking and is derived from the Golgi compartment. Both GTPase activity and membrane-targeting ability are essential for BaMV replication. Another host protein, chl-PGK, is a chloroplast-targeting protein, and its targeting ability is required for BaMV replication. Therefore, NbRABG3f and chl-PGK may play roles in delivering BaMV and its related complex to chloroplasts, which are a type of BaMV replication compartment. After completing replication, BaMV cargoes are delivered to the plasmodesmata for intercellular movement. NbRabGAP1, a Rab-associated protein, is not involved in BaMV replication but participates in its intercellular movement. Accordingly, NbRabGAP1 may be involved in the intracellular transport of the BaMV complex to the plasmodesmata after BaMV replication. Additional studies on dissecting the co-localization or interaction between BaMV and other vesicle proteins are required to clarify the intracellular movement pathway for BaMV.
# Author contributions
C-PC wrote and edited the paper.
[fig] FUNDING: This work was financially supported by MOST (grant number MOST 104-2313-B-320-001 and MOST 105-2313-B-320-001). [/fig]
|
Personality traits in patients with myotonic dystrophy type 2
Myotonic dystrophy type 2 (DM2) is a multisystem disorder that affects many organs and systems, including the brain. The objective is to analyze personality patterns in myotonic dystrophy type 2 (DM2) compared to DM1 control group. The study comprised 27 consecutive genetically confirmed DM2 patients and control group of 44 DM1 patients. Personality traits were assessed with the Millon Multiaxial Clinical Inventory III (MMCI III). In DM2 group there were no scale with pathological scores, although compulsive and paranoid traits were the most prominent. DM2 patients had lower scores compared to DM1 patients in almost all scales. Pathological scores on clinical symptom scales were not observed, although anxiety scale almost approached this value. Patients with higher compulsive score had higher level of education (rho = +0.53, p < 0.01). On the other hand, higher paranoid score correlated with younger age at onset (rho = -0.34, p < 0.01) and lower educational level (rho = -0.26, p < 0.05). Our results did not show significant personality impairments in patients with DM2. However, following personality traits were predominant: compulsive (in patients with higher education) and paranoid (in patients with lower education and earlier age at onset). The most common clinical symptoms were anxiety and somatization.
# Introduction
Myotonic dystrophy type 2 (DM2) is a multisystemic disorder that affects many organs and systems, including the brain (1).
One study showed similar histopathological findings in the brain of patients with DM2 and myotonic dystrophy type 1 (DM1) (2), magnetic resonance imaging revealed similar white matter impairments in both diseases [bib_ref] Brain involvement in myotonic dystrophies: neuroimaging and neuropsychological comparative study in DM1..., Romeo [/bib_ref] [bib_ref] Comparative analysis of brain structure, metabolism, and cognition in myotonic dystrophy 1..., Weber [/bib_ref] , and in both diseases dysexecutive impairment has been described, although less severe in DM2 (4-7). On the other hand, previous studies showed less pronounced grey matter loss in DM2 brain [bib_ref] Cerebral and muscle MRI abnormalities in myotonic dystrophy, Franc [/bib_ref].
Reported frequencies of personality disorders in DM1 patients are between 20% and 64%, but some authors did not find significant personality changes [bib_ref] Personality patterns in patients with myotonic dystrophy, Delaporte [/bib_ref] [bib_ref] Temperament and character in patients with classical myotonic dystrophy type 1 (DM-1), Winblad [/bib_ref] [bib_ref] Dependent and paranoid personality patterns in myotonic dystrophy type 1, Peric [/bib_ref] [bib_ref] Cognitive/personality pattern and triplet expansion size in adult myotonic dystrophy type 1..., Sistiaga [/bib_ref] which is far above the prevalence in the general population. However, there is a single previous study that specifically assessed personality pattern in DM2 subjects [bib_ref] Executive dysfunction and avoidant personality trait in myotonic dystrophy type 1 (DM-1)..., Meola [/bib_ref]. Although none of the patients in this study fulfilled the DSM-IV criteria for the diagnosis of personality disorder, significant avoidant behavioural trait was observed in both DM2 and DM1 compared to controls.
The aim of this study was to analyze personality patterns in a cohort of DM2 patients compared to DM1 subjects.
# Materials and methods
The cross-sectional study comprised 25 DM2 patients consecutively recruited during their first hospitalization at the Inpatient Unit of the Neurology Clinic, Clinical Centre of Serbia in the period from March 2013 until January 2014. Genetic diagnosis of CCTG repeats expansion using repeat primed polymerase chain reaction (RP-PCR) was obtained for all patients in addition to typical clinical and electromyographic data [bib_ref] Best practice guidelines and recommendations on the molecular diagnosis of myotonic dystrophy..., Kamsteeg [/bib_ref]. Patients with any other associated somatic and neurological diseases not related to DM2 were excluded. Control group consisted of 44 genetically confirmed DM1 patients examined at the Clinic in the same period. Patients with congenital and childhood-onset DM1 were excluded from the study. During this period no DM1 patients with mild, late-onset phenotype were hospitalized, thus they also were not included in the research. All participants gave informed consent to participate in the study and the study was approved by the Ethical Board of the School of Medicine, University of Belgrade.
Manual muscle testing (0 to 5 scale according to Medical Research Council (MRC) scale) was performed in DM2 patients by experienced clinicians (VRS, SP). We added strength of the weakest muscle of the proximal arms, distal arms, proximal legs and distal legs, with maximum score being 20. Global cognitive status of DM2 and DM1 subjects was assessed using the Addenbrooke's Cognitive Examination -Revised (ACE-R) [bib_ref] The Addenbrooke's Cognitive Examination Revised (ACE-R): a brief cognitive test battery for..., Mioshi [/bib_ref]. Values below 82 were considered indicative of cognitive impairment and these patients were excluded from the study.
Personality traits and psychopathology in our subjects were assessed with the Millon Multiaxial Clinical Inventory (MMCI III). The scores were converted directly into base rate (BR) scores, which take into account the prevalence of a particular characteristic. According to Millon's criteria, BR punctuations > 75 signify the presence of a trait and BR punctuations > 85 are considered as an impairment.
Student t test and Mann Whitney U test were used for group comparisons, as appropriate. Spearman's coefficient was applied for correlation analyses. Significant testing was two-sided, with alpha set at 0.05 for statistical significance and 0.01 for high statistical significance.
# Results
Main sociodemographic, clinical and cognitive features of DM2 and control DM1 patients are presented in [fig_ref] Table 1: Sociodemographic, clinical and cognitive features of investigated patients [/fig_ref].
Mean scores on personality scales are presented in . In DM2 group there were no scale with mean score above 75 although compulsive and paranoid traits were the most prominent. DM2 patients had lower scores compared to DM1 patients in all scales, except for narcissistic and antisocial ones where significant differences were not registered between groups (scores were normal in both groups). Mean score above 75 was not observed in DM2 patients on clinical symptom scales, although anxiety scale almost approached this value. The second highest score was somatization. All scores were lower in DM2 compared to DM1 control group.
We further correlated personality scales with the highest score (compulsive and paranoid) with sociodemographic, clinical and cognitive findings (ACE-R subscores and total score). Patients with higher compul-
# Discussion
Our results did not show significant personality impairments in patients with DM2 and these patients scored better on personality scales compared to DM1 patients. Our DM1 and DM2 groups were well matched regarding gender and education, while difference in age of six years seems to be not relevant regarding personal- ity since they all were in the mature adulthood period. Our results are in line with the study by reported that although significant avoidant behavioural trait was observed in DM2 patients, no one fulfilled the criteria for the diagnosis of personality disorder [bib_ref] Executive dysfunction and avoidant personality trait in myotonic dystrophy type 1 (DM-1)..., Meola [/bib_ref]. On the other hand, reported frequencies of personality disorders in DM1 patients are between 20% and 64% (9-12) which is far above the prevalence in the general population. Bertrand et al reported that 30% of the adult onset DM1 patients were at risk of developing a psychiatric disorder, had high score in paranoid ideation, delusional ideation, psychoticism and phobic-anxiety, lower score in self-esteem, as well as higher score in avoidant behaviour and social withdrawal [bib_ref] Psychological characteristics of patients with myotonic dystrophy type 1, Bertrand [/bib_ref]. Moreover, low self-esteem and anxiety found in the more severe phenotype correlated with a low cognitive profile and with difficulties in executive tasks'Although DM2 affects the brain and although reaction on such a chronic disease is expected, it seems that neither of these two factors is strong enough to cause personality impairment. Possible explanations are that brain seems to be less affected in DM2 compared to DM1 [bib_ref] Brain involvement in myotonic dystrophies: neuroimaging and neuropsychological comparative study in DM1..., Romeo [/bib_ref] [bib_ref] Comparative analysis of brain structure, metabolism, and cognition in myotonic dystrophy 1..., Weber [/bib_ref] [bib_ref] Executive dysfunction and avoidant personality trait in myotonic dystrophy type 1 (DM-1)..., Meola [/bib_ref] [bib_ref] Frontostriatal dysexecutive syndrome: a core cognitive feature of myotonic dystrophy type 2, Peric [/bib_ref] [bib_ref] Cerebral and muscle MRI abnormalities in myotonic dystrophy, Franc [/bib_ref]. Our results also suggest that DM2 patients may adapt and realize themselves better in relation with partners since they were married more frequently compared to DM1 subjects. In our DM2 group the highest score was observed on compulsive scale, although not reaching a pathological level. This finding is in accordance with Meola's study that also showed the highest score for the same scale but without compulsive personality disorder [bib_ref] Executive dysfunction and avoidant personality trait in myotonic dystrophy type 1 (DM-1)..., Meola [/bib_ref]. Compulsive personality is characterized by a general pattern of concern with orderliness, organization, preoccupation with details, mental and interpersonal control and the control of one's environment. They have a tendency to keep control on everything in order to compensate for unpredictable consequences of a chronic disease. These symptoms may cause distress and interfere with a person's occupational and social functioning, but they also might have positive consequences since workaholism is often seen in those with this personality disorder. Accordingly, our results showed higher compulsive score in patients with higher level of education. Conscientiousness, conformation to rules, and wish to confirm himself/ herself are the main features of compulsive personality that might explain its association with educational achievement.
Second highest score was observed on paranoid scale. Similarly, this was the highest personality score in our DM1 cohort [bib_ref] Dependent and paranoid personality patterns in myotonic dystrophy type 1, Peric [/bib_ref]. Fear of being abandoned is the core feature of paranoid personality. It is expressed as an attachment anxiety, so these patients are suspicious and mistrustful of others, think they are in danger and may be hypersensitive and hostile in relations to other people. It is also of note that our DM2 patients and also DM1 patients from our previous study had higher paranoid scores if they were less educated [bib_ref] Dependent and paranoid personality patterns in myotonic dystrophy type 1, Peric [/bib_ref]. Lower educational level means less ability to consider all life circumstances and less flexibility of thoughts. Association between paranoid traits and earlier age at onset of DM2 might be explained with the fact that younger people are more hypersensitive and more vulnerable.
Previous studies showed correlation between personality disorders and cognitive findings in general population [bib_ref] Executive functioning in people with personality disorders, Garcia-Villamisar [/bib_ref]. Also, Meola et al. reported low scores on cognitive tests of frontal lobe function in parallel with avoidant trait personality in both DM1 and DM2 patients. We have also previously reported significant dysexecutive syndrome and certain impairment of episodic verbal memory, while dysexecutive and visuospatial/visuoconstructional deficits predominated in DM1 (7). However, we did not confirm association between cognitive deficit and personality traits when directly comparing neuropsychological and psychological results in the temporary study.
The most prominent clinical symptoms in our DM2 patients was anxiety. Similar was observed in our research on DM1 subjects [bib_ref] Dependent and paranoid personality patterns in myotonic dystrophy type 1, Peric [/bib_ref]. This comorbidity worsens all aspects of patients' life-biological, psychological and mental [bib_ref] Anxiety and depression in patients with medical diseases, Schüssler [/bib_ref]. This was previously confirmed in DM1 patients [bib_ref] Health-related quality of life in patients with myotonic dystrophy type 1 and..., Peric [/bib_ref] [bib_ref] Health-related quality of life in myotonic dystrophy type 1 and its relationship..., Antonini [/bib_ref]. Thus, adequate psychiatric and psychological treatment of anxiety should be considered in these patients.
This study has few limitations. Number of investigated patients is small, but this is a rare disease. Furthermore, correlation with neuroimaging findings would be of interest.
# Conclusions
Our results did not show significant personality impairments in patients with DM2. However, following personality traits were predominant: compulsive (in patients with higher education) and paranoid (in patients with lower education and earlier age at onset). The most common clinical symptoms were anxiety and somatization.
[table] Table 1: Sociodemographic, clinical and cognitive features of investigated patients. [/table]
[table] Table 2: Results on MMCI scales. [/table]
|
Immunologic response in treatment-naïve HIV-2-infected patients: the IeDEA West Africa cohort
Introduction: Response to antiretroviral therapy (ART) among individuals infected with HIV-2 is poorly described. We compared the immunological response among patients treated with three nucleoside reverse-transcriptase inhibitors (NRTIs) to boosted protease inhibitor (PI) and unboosted PI-based regimens in West Africa.Methods: This prospective cohort study enrolled treatment-naïve HIV-2-infected patients within the International Epidemiological Databases to Evaluate AIDS collaboration in West Africa. We used mixed models to compare the CD4 count response to treatment over 12 months between regimens. Results: Of 422 HIV-2-infected patients, 285 (67.5%) were treated with a boosted PI-based regimen, 104 (24.6%) with an unboosted PI-based regimen and 33 (7.8%) with three NRTIs. Treatment groups were comparable with regard to gender (54.5% female) and median age at ART initiation (45.3 years; interquartile range 38.3 to 51.8). Treatment groups differed by clinical stage (21.2%, 16.8% and 17.3% at CDC Stage C or World Health Organization Stage IV for the triple NRTI, boosted PI and unboosted PI groups, respectively, p 00.02), median length of follow-up (12.9, 17.7 and 44.0 months for the triple NRTI, the boosted PI and the unboosted PI groups, respectively, pB0.001) and baseline median CD4 count (192, 173 and 129 cells/ml in the triple NRTI, the boosted PI and the unboosted PI-based regimen groups, respectively, p 00.003). CD4 count recovery at 12 months was higher for patients treated with boosted PI-based regimens than those treated with three NRTIs or with unboosted PI-based regimens (191 cells/ml, 95% CI 142 to 241; 110 cells/ml, 95% CI 29 to 192; 133 cells/ml, 95% CI 80 to 186, respectively, p 00.004). Conclusions: In this observational study using African data, boosted PI-containing regimens had better immunological response compared to triple NRTI combinations and unboosted PI-based regimens at 12 months. A randomized clinical trial is still required to determine the best initial regimen for treating HIV-2 infected patients.
# Introduction
Between one and two million people are estimated to be living with HIV-2 infection in West Africa [bib_ref] Sixteen years of HIV surveillance in a West African research clinic reveals..., Awasana [/bib_ref] , the region that is the epicentre of the HIV-2 epidemic, with prevalence apparently decreasing over time . Although there is limited experience in the management of HIV-2 infection worldwide, it is well known that HIV-2 is naturally resistant to the non-nucleoside reverse-transcriptase inhibitors (NNRTIs) [bib_ref] Structure of HIV-2 reverse transcriptase at 2.35-A resolution and the mechanism of..., Ren [/bib_ref] [bib_ref] Susceptibility of HIV-2, SIV and SHIV to various anti-HIV-1 compounds: implications for..., Witvrouw [/bib_ref] that have been part of the standard first-line antiretroviral therapy (ART) for the treatment of HIV-1 infection in low-income countries for the past decade. In the 2010 treatment guidelines, the World Health Organization (WHO) recommended treating patients living with HIV-2 in limitedresource countries with an ART regimen containing a ritonavir-boosted protease inhibitor (PI) plus two nucleoside reverse-transcriptase inhibitors (NRTIs). Triple NRTI regimens were recommended only in patients with CD4 counts !200 cells/mm 3 . In the 2013 guidelines, WHO recommendations proposed a regimen containing three NRTIs or a ritonavirboosted PI plus two NRTIs. If a PI-based regimen is used, the preferred option for first-line PI is lopinavir. However, this recommendation is based on weak evidence according to WHO grading criteria . Although PIs are active against HIV-2, they show varying degrees of activity due to natural protease polymorphisms. A study that compared the potency of different PIs against HIV-2 showed that lopinavir, saquinavir, tipranavir and darunavir were the most potent [bib_ref] Inhibition of HIV-2 protease by HIV-1 protease inhibitors in clinical use, Brower [/bib_ref]. Another study reported that saquinavir, lopinavir and darunavir are potent inhibitors of HIV-2 isolates [bib_ref] In vitro phenotypic susceptibility of human immunodeficiency virus type 2 clinical isolates..., Desbois [/bib_ref].
In the context of the large and still ongoing scale-up of ART in the West African region, HIV-2 infection causes specific operational, clinical and public health challenges [bib_ref] Antiretroviral therapy for HIV-2 infection: recommendations for management in low-resource settings, Peterson [/bib_ref]. First, the lack of routinely available rapid antibody tests to differentiate between HIV-1 and HIV-2 can lead to delayed diagnoses and thus to initiation of inappropriate and ineffective first-line ART prescriptions [bib_ref] Inadvertent non-nucleoside reverse transcriptase inhibitor (NNRTI)-based antiretroviral therapy in dual HIV-1/2 and..., Sarfo [/bib_ref] [bib_ref] First-year lymphocyte T CD4' response to antiretroviral therapy according to the HIV..., Drylewicz [/bib_ref]. Indeed, a previous study of our group [bib_ref] First-year lymphocyte T CD4' response to antiretroviral therapy according to the HIV..., Drylewicz [/bib_ref] showed that despite international recommendations 17% of HIV-2-infected patients were initially treated with inappropriate NNRTI-containing regimens, resulting in poor immunological response mainly due to late confirmation of HIV-2 infection [bib_ref] First-year lymphocyte T CD4' response to antiretroviral therapy according to the HIV..., Drylewicz [/bib_ref]. Second, the lack of commercial viral load quantification assay [bib_ref] Special aspects of the treatment of HIV-2-infected patients, Camacho [/bib_ref] impacts on the monitoring of HIV-2 infected patients [bib_ref] Antiretroviral therapy for HIV-2 infection: recommendations for management in low-resource settings, Peterson [/bib_ref] [bib_ref] Update on human immunodeficiency virus (HIV)-2 infection, Campbell-Yesufu [/bib_ref].
Owing to its low prevalence and its geographical restriction to West Africa, response to ART in HIV-2 infection is still poorly understood [bib_ref] 20 years of HIV-2 infection in Portugal: trends and changes in epidemiology, Valadas [/bib_ref] [bib_ref] HIV-2 infection: a call for controlled trials, Matheron [/bib_ref]. There have been no randomized trials investigating the response to ART regimens in HIV-2infected patients [bib_ref] A call for randomized controlled trials of antiretroviral therapy for HIV-2 infection..., Gottlieb [/bib_ref]. Only a few observational cohort studies with limited sample size have provided information on HIV-2infected patients on ART in Europe and in the United States [20Á26], as well as in West Africa [13Á15, 27Á31]. Moreover, limited data are available on the response to the different first-line regimens used in HIV-2 patients.
Whereas the data have generally favoured boosted PI regimens over those comprised of triple NRTIs, to our knowledge, only one report from a European collaboration [bib_ref] Immunovirological response to triple nucleotide reverse-transcriptase inhibitors and ritonavir-boosted protease inhibitors in..., Benard [/bib_ref] and two from West Africa [bib_ref] First-year lymphocyte T CD4' response to antiretroviral therapy according to the HIV..., Drylewicz [/bib_ref] [bib_ref] Emergence of multiclass drug-resistance in HIV-2 in antiretroviral-treated individuals in Senegal: implications..., Gottlieb [/bib_ref] explicitly compared the two first-line regimens. Using data from the International Epidemiological Database to Evaluate AIDS West Africa (IeDEA-WA) HIV-2 collaboration [bib_ref] Cohort profile: the international epidemiological databases to evaluate AIDS (IeDEA) in sub-Saharan..., Egger [/bib_ref] , we aimed to investigate the impact of first-line ART regimens on immunological response within the first 12 months of ART, for HIV-2 infected patients in the West Africa region where there are limited options for second-line regimens.
# Methods
Description of the cohort Since 2006 a network of adult and paediatric HIV clinics in West Africa has existed as part of the global IeDEA Collaboration (www.iedea.org/), funded by the US National Institutes of Health [bib_ref] Cohort profile: the international epidemiological databases to evaluate AIDS (IeDEA) in sub-Saharan..., Egger [/bib_ref]. The IeDEA-WA HIV-2 cohort was recently established in order to better understand the epidemiology, care patterns and treatment of HIV-2 infection [bib_ref] Characteristics of HIV-2 and HIV-1/HIV-2 dually seropositive adults in West Africa presenting..., Ekouevi [/bib_ref]. This cohort includes HIV-2 and dually HIV-1-and HIV-2-infected patients, on ART or not, followed in 12 clinics located in five West African countries (six in Côte d'Ivoire, two in Mali, two in Burkina Faso, one in Benin and one in Senegal).
## Data collection
Standardized questionnaires capturing the relevant information on HIV-2 care were developed with an electronic database implemented at the site level. All sites completed the specific questionnaires retrospectively and then prospectively and entered the data into the IeDEA-WA HIV-2 database. The databases from each site are sent every six months to the Regional Centre in Abidjan, Côte d'Ivoire, and Bordeaux, France, using compression/encryption software. Data collected include the following: 1) Baseline demographics and clinical data: birth date, gender, HIV clinical stage (WHO or CDC stage), ART initiated, clinical assessment, medical history; 2) Follow-up: clinical assessment (tuberculosis, other diseases/ infection, HIV clinical stage, weight, height, medications such as antiretroviral drugs and co-trimoxazole); 3) Biological data: CD4 count, haemoglobin, aspartate transaminase, alanine transaminase and plasma HIV RNA viral load (when available); 4) Outcomes: death, loss to follow-up and transferred out.
## Inclusion criteria
Patients aged !17 years at ART initiation and with an HIV-2 infection confirmed by two or three rapid HIV tests based on country-specific national algorithms were eligible for this analysis. Only those patients who initiated ART with three NRTIs or a PI-based regimen were included. Patients without documented CD4 count at ART initiation and/or with unknown gender were excluded.
## Antiretroviral treatment
ART was provided to the HIV-2 infected patients according to the national guidelines. In West Africa, HIV-2 treatment guidelines were based on WHO recommendations. Up to 2013, the recommended first-line regimen contained three NRTIs (tenofovir (TDF)'lamivudine (3TC) or emtricitabine (FTC)'zidovudine (AZT) or AZT'3TC'abacavir (ABC)). Since 2013, a boosted PI-based regimen has been recommended, with lopinavir being the preferred option. In patients for whom boosting the PI is contraindicated or not tolerated, a triple NRTI regimen was recommended (AZT'3TC or FTC'TDF or ABC).
Follow-up and CD4 count measurement After initiation into care, patients were typically followed up every six months or were seen in between scheduled visits if illness occurred. T-CD4 lymphocyte counts were measured every six months. The absolute CD4'/CD8' T-cell counts were performed using standard flow cytometry (FACScan, Becton Dickinson).
# Statistical analysis
CD4 count trajectories over the first 12 months of ART were modelled using linear mixed models (LMMs), with baseline defined as the date of ART initiation. We included all CD4 count measurements between baseline (or the first CD4 count recorded within a window of six months prior to baseline) and 24 months of follow-up (or six months later at the latest). In the main analysis, patients who switched initial treatment were right censored at the date of switching. We modelled the CD4 count changes over time [bib_ref] Modelling the association between patient characteristics and the change over time in..., Harrison [/bib_ref] using fractional polynomials of one and two degrees with powers (2, (1, (0.5, 0 (natural log), 0.5, 1, 2 and 3. The best-fitting fractional polynomial was selected by comparing the deviance of different models and had powers 0.5 and 1. Random effects on the different fractional polynomial terms accounted for the correlation of repeated measurements within each subject. For degree-2 models unstructured and diagonal covariance matrices were compared. Models with an unstructured covariance matrix were selected, as their Akaike information criterion was smaller. Univariable LMMs were used to assess covariate effects, and a manual backwards selection method was used to select significant variables in a multivariable LMM. To confirm the adequacy of the model, residual homoscedasticity and normality were graphically checked.
To address possible informative dropout we performed a sensitivity analysis restricted to patients remaining in care beyond the 12 months of ART (patients deceased or lost to follow-up were excluded). Patients were defined as lost to follow-up if they were not known to have died, not known to have transferred out and not seen at the clinic at least once in the last six months prior to the closure date of the database [bib_ref] Universal definition of loss to follow-up in HIV programs: a statistical analysis..., Chi [/bib_ref].
In a second sensitivity analysis, we performed the multivariable LMM in an intent-to-treat analysis.
# Results
Selection of the study sample As of March 2013 the IeDEA-WA database contained 2005 patients infected with HIV-2 (or dually infected with HIV-1 and HIV-2). [fig_ref] Figure 1: Flow chart of the selection of the study sample of patients infected... [/fig_ref] shows the different steps of the selection of the study sample. We excluded 590 HIV-2 patients and 344 dually infected patients because they were not treated with ART. We excluded 485 patients because they were dually infected, one patient with an unknown first-line ART regimen and 44 (7.5%) HIV-2 patients treated with a non-recommended first-line ART regimen. Among these 44 patients, 41 were initially treated with an NNRTI-based regimen and three with only two drugs. Of these 44 patients, 31 (70.5%) treated with a non-recommended ART regimen had a subsequent drug combination switch reported (28 (90.3%) switched to a PI-based regimen and three patients (9.7%) switched to a triple NRTI regimen), at a median 14 months after ART initiation (interquartile range (IQR) 2.7 to 21.3 months). After exclusion of 118 patients without documented baseline CD4 count measurements, there remained 422 eligible HIV-2-infected patients that initiated ART between 1999 and 2012 who constituted the core sample for this analysis.
## Sample characteristics
Of 422 eligible patients, 285 (67.5%) started a boosted PI-based ART regimen, 104 (24.6%) an unboosted PI-based regimen and 33 (7.8%) a triple NRTI regimen [fig_ref] Table 1: Baseline characteristics of the HIV-2-infected patients on antiretroviral therapy [/fig_ref]. The subjects in the three groups had similar sex and age distributions, mortality and loss to follow-up, but varied markedly in their baseline clinical stage, baseline CD4 count, followup duration and the calendar year of starting ART [fig_ref] Table 1: Baseline characteristics of the HIV-2-infected patients on antiretroviral therapy [/fig_ref]. Of the patients treated with a boosted or unboosted PI-based regimen, 17% had an advanced clinical stage (WHO IV or CDC C) compared to 21.2% of patients treated with three NRTIs (p00.02). Baseline CD4 cell count was lower in the unboosted PI group (median 129 cells/ml; 67 to 206) compared to the boosted PI group (173 cells/ml; 80 to 265) and to the NRTI group (192 cells/ml; 114 to 308; p 00.003). Median followup time was higher for the unboosted PI group (median 44 months; 13.2 to 72.0) compared to 17.7 months (2.4 to 36.6) and 12.9 months (0 to 38.9) for patients treated with a boosted PI-based and a three-NRTI regimen, respectively (pB0.001). Twenty-one (63.6%) patients initially treated with three NRTIs started their first-line ART regimen between 2008 and 2012, compared to 191 (67.0%) patients treated with a boosted PI-based regimen and 9 (8.6%) patients treated with an unboosted PI-based regimen (p B0.001). Of the patients treated with a PI-based regimen (boosted or not) 55% were female, compared to 42.4% of the patients treated with three NRTIs (p 00.35). The median age at ART initiation was 45.5 (38.9 to 52.1), 44.2 (36.9 to 51.2) and 46.1 (40.7 to 51.9) years for patients treated with a boosted PI-based, unboosted PIbased and triple NRTI regimen, respectively (p00.59). Twelve months after starting ART there were 12 (2.8%) deaths in total, with similar proportions in the three treatment groups (3.2%, 1.9% and 3.0% in the boosted PI, unboosted PI and NRTI groups, respectively; p 00.80). Twenty-four percent of patients were lost to follow-up after 12 months (p00.43).
The triple NRTI ART combinations prescribed were AZT'3TC and ABC for 22 patients (66.7%), stavudine'3TC and didanosine for six patients (18.2%), stavudine'3TC and ABC for three patients (9.1%) and AZT'3TC and didanosine for two patients (6.1%). For patients treated with a boosted PI-based regimen, the initial combinations mostly prescribed were AZT'3TC and lopinavir-ritonavir for 130 patients (45.6%), stavudine'3TC and lopinavir-ritonavir for 52 patients (18.3%) and AZT'3TC'indinavir and ritonavir for 34 patients (11.9%). For patients treated with an unboosted PI-based regimen, the initial combinations mostly prescribed were AZT'3TC' indinavir for 61 patients (58.7%), stavudine'3TC'indinavir for 25 patients (24.0%) and AZT'3TC'nelfinavir for 9 patients (8.7%). Lopinavir-ritonavir was the most frequently prescribed PI for 73.7% in the boosted PI group (n0210), followed by Treatment modification during the first 12 months of ART was reported for 11 (33.3%) patients initially treated with three NRTIs: five patients switched to a boosted PI-based regimen, four to another NRTI-based combination and two to an NNRTI-based regimen. Of 45 (15.8%) patients initially treated with a boosted PI-based regimen who had a treatment modification within 12 months, 38 were still treated with a boosted PI-based regimen, three were treated with three NRTIs, two with a NNRTI combination, one with an unboosted PI-based regimen and one with two NRTIs. Of 27 (26.0%) patients initially treated with an unboosted PI-based regimen, 24 were treated with a boosted PI-based regimen, two with an NNRTI combination and one with three NRTIs. The median times from ART initiation to treatment modification were 5.7 (2.9 to 8.9), 4.9 (2.8 to 8.7) and 5.8 (2.0 to 8.8) months for patients initially treated with three-NRTI, unboosted PI-based and boosted PI-based regimens, respectively. Reasons for ART modification were not recorded in the database. [fig_ref] Figure 2: Mean adjusted CD4 count change after antiretroviral treatment initiation according to baseline... [/fig_ref] shows trajectories of CD4 count recovery estimated by the multivariable LMM stratified by CD4 count category at baseline. In the reference group (women treated with a boosted PI-based regimen and with an initial CD4 count B50 cells/ml), the mean CD4 changes were 132 cells/ml (95% CI 89 to 176) and 191 cells/ml (95% CI 142 to 241) at 6 and 12 months after starting ART, respectively. [fig_ref] Table 2: Mean CD4 count changes at 6 and 12 months compared to the... [/fig_ref] shows estimated mean CD4 changes at 6 and 12 months after starting ART. The CD4 count response to ART within the first 12 months of ART was associated with the absolute value of the baseline CD4 count (p 00.0291), with gender (p 00.0302) and with the type of first ART regimen (p00.0045). Compared to patients with a baseline CD4 count B50 cells/ml, patients with baseline CD4 count ]350 had lower CD4 count recovery ( (99 cells/ml ((164 to (34)) 12 months after ART initiation. Compared to women, men had a lower CD4 count recovery after 12 months of ART ( (39 cells/ml ((71 to (8)). We found no association between CD4 response to ART and age, baseline haemoglobinaemia, year of ART initiation, country and initial clinical stage (data not shown).
## Immunologic response to art treatment
There was a strong association between the type of firstline ART regimen and the overall CD4 response trajectory (p00.0045). Compared to patients initially treated with a boosted PI-based regimen, CD4 count recovery was lower for patients treated with three NRTIs or with an unboosted PI-based regimen ((72 cells/ml ((129 to [bib_ref] Special aspects of the treatment of HIV-2-infected patients, Camacho [/bib_ref] ; (42 cells/ml ( (74 to [bib_ref] Inhibition of HIV-2 protease by HIV-1 protease inhibitors in clinical use, Brower [/bib_ref] , respectively) at six months. The difference in CD4 count recovery between the three treatment groups persisted at 12 months [fig_ref] Figure 2: Mean adjusted CD4 count change after antiretroviral treatment initiation according to baseline... [/fig_ref].
The results of the sensitivity analysis restricted to patients remaining in care (n0265) were similar to those from the main analysis. CD4 count changes at 12 months estimated from these two models, according to baseline CD4 count, are shown in [fig_ref] Figure 3: Adjusted mean CD4 count change at M12 [/fig_ref].
The results of the second sensitivity intent-to-treat analysis were similar to those from the main analysis.
# Discussion
Based on a collaborative analysis of data combined from 12 clinics located in five West African countries, we described immunological responses of HIV-2 infected patients in the (14) a Females with initial CD4 count B50 cells/ml treated with boosted PI-based regimen; CI, confidence interval; ART, antiretroviral treatment; PI, protease inhibitor; NRTI, nucleoside reverse-transcriptase inhibitors; b the mean CD4 count change for the reference group at 6 months was 132 cells/ml (95% CI 089; 176); c the mean CD4 count change for the reference group at 12 months was 191 cells/ml (95% CI 0142; 241). largest sample of HIV-2-infected patients followed for up to 12 months on ART. We compared the ART regimens recommended by the WHO for low-and middle-income settings and found CD4 count responses to be higher for patients treated with boosted PI-based regimens than for those treated with triple NRTI or unboosted PI-based regimens 6 and 12 months after ART initiation.
Another study that investigated the immunological responses to the two first-line ART regimens for HIV-2 infected patients, the European cohort study ACHI E V 2E [bib_ref] Immunovirological response to triple nucleotide reverse-transcriptase inhibitors and ritonavir-boosted protease inhibitors in..., Benard [/bib_ref] , showed a lower immunological response for patients initially treated with three NRTIs than those treated with a PI-based regimen. Indeed, a CD4 count decrease was observed for patients treated with three NRTIs ( (60 cells/ml per year) and a CD4 count increase for those treated with a PI-based regimen ('76 cells/ml per year). At 12 months, the CD4 counts for patients treated with three NRTIs and a PI-based regimen were 191 and 327 cells/ml, respectively. The immunological response beyond month 12 was not investigated in that study. Whether the early immunological advantage observed with boosted PI-based ART translates into clinical benefits cannot be answered here and will be difficult to investigate.
In light of the results of the European cohort study [bib_ref] Immunovirological response to triple nucleotide reverse-transcriptase inhibitors and ritonavir-boosted protease inhibitors in..., Benard [/bib_ref] , in 2013 a French group of experts [37] recommended the use of a PI-based regimen as a first-line combination and ceased to recommend the use of a triple NRTI-based regimen for HIV-2 infected patients. The choice between the two first-line regimens is probably more critical in low-income countries: On the one hand, therapies based on a combination of three NRTIs are an interesting alternative in a context of high tuberculosis prevalence because treatment with a rifampinbased regimen requires double boosting of ritonavir in association with lopinavir or saquinavir, which results in an increased risk of hepatotoxicity . On the other hand, the risk of occurrence of drug resistance has to be balanced against this as there are limited options for second-line regimens. We recently reported that, after a median duration of four years on PI-based regimens, 74% of 145 HIV-2-infected patients had suppressed viral loads of B50 copies/ml; however, HIV-2 resistance mutations to NRTIs and PIs were detected in 21 of 25 (84%) and 20 of 29 (69%) samples, respectively, despite adequate antiretroviral plasma concentrations [bib_ref] Genotypic resistance profiles of HIV-2-treated patients in, Charpentier [/bib_ref]. This study clearly showed that in cases of virological failure there is a limited HIV-2 therapeutic arsenal and that cross-resistance dramatically reduced second-line treatment options. Another study reported that 40% of patients with PI resistance mutations appeared to be resistant to darunavir, which is recommended as second-line therapy for HIV-2 patients [bib_ref] Use of observational databases to evaluate the effectiveness of antiretroviral therapy for..., Raugi [/bib_ref]. With regard to the HIV-2 therapeutic arsenal, in cases of NRTI and PI resistance, the only active drug class is integrase inhibitors and possibly the CCR5 inhibitor maraviroc. However, the use of integrase inhibitors may be limited in pretreated patients who require a combination with fully active drugs and who harbour NRTI-resistant viruses because of the low genetic barrier to resistance of this drug class. The potential use of integrase inhibitors and maraviroc is also limited by the high costs of these medicines.
Among patients who initiated triple NRTI regimens, onethird had changed treatment over the period of 12 months. This rate was higher compared to other African public health ART programmes treating HIV-1 patients and needs further exploration to understand the reasons for treatment modification among HIV-2 patients. The most obvious explanation for the high switch rate is the known inferiority of triple NRTI regimens and co-treatment of tuberculosis being the main reason for their use. Side effects, tolerability or adherence could not be investigated as they were not available in our database. In the European HIV-2 cohort study, treatment modification was less common with switching reported in 12% of patients in the PI group and 18% in the NRTI group [bib_ref] Immunovirological response to triple nucleotide reverse-transcriptase inhibitors and ritonavir-boosted protease inhibitors in..., Benard [/bib_ref].
A limitation of our study was the small number and limited follow-up duration of HIV-2 infected patients treated with three NRTIs (17 of the 33 patients had at least one CD4 count documented after six months). Indeed, in the WHO guidelines of 2010this ART regimen was only recommended for patients with a CD4 cell count !200 cells/ml. In the 2013 WHO consolidated guidelinesthis eligibility criteria has been dropped, which could lead to triple NRTI regimens being prescribed more widely.
Another limitation is the use of suboptimal triple NRTI regimens in West Africa. Some combinations used in this study, such as AZT'3TC'ABC, are not well tolerated and are known to be less potent than triple NRTI regimens used elsewhere in Africa.
Because the three groups of patients compared in our study differed in terms of baseline CD4 count, baseline clinical stage and follow-up duration, this analysis was possibly affected by selection bias and the results should be interpreted with caution, as in all observational cohorts comparing treatment regimens . In resource-limited countries, high rates of patients lost to follow-up are frequently observed [bib_ref] Correcting mortality for loss to follow-up: a nomogram applied to antiretroviral treatment..., Egger [/bib_ref] [bib_ref] Samplingbased approach to determining outcomes of patients lost to follow-up in antiretroviral..., Geng [/bib_ref] , which could induce informative dropout . However the results of the sensitivity analysis we conducted on patients remaining in care were similar to those of the main analysis, showing that this informative dropout bias was limited. In conclusion this study further demonstrates the inferiority of unboosted indinavir and triple nucleoside regimens for the treatment of HIV-2 infection. Limitations inherent in this observational design will be addressed by the First-Line Treatment for HIV-2 trial (FIT-2, NCT02150993) already underway in five countries in West Africa and funded by the Agence Nationale de Recherches sur le Sida et les hépatites virales (ANRS). The critical issue of second-line regimens for HIV-2, including the optimal sequence of integrase inhibitors and/ or boosted PIs, could potentially be addressed by a study enrolling the subjects failing therapy in those two arms of the FIT-2 trial.
## Competing interests
The authors declare that they have no conflict of interest.
Authors' contributions EB and MM led the process of data analysis. EB conducted statistical analysis and co-wrote the manuscript. Findings were interpreted by EB, MM, RT, DKE, BT and FD. The manuscript was drafted by EB; DKE, BT, RT and FD. DKE is the coordinator of the HIV-2 cohort within the IeDEA-WA Collaboration; he managed the fieldwork and co-wrote the article. FD is the primary investigator of the IeDEA-WA Collaboration and provided critical revision for important intellectual content. EM, AS and SPE are co-investigators; they managed the fieldwork and commented on the original manuscript. JS is the principal investigator of the ART-CC Collaboration; he commented on the article and provided advice on data interpretation. MM is also a member of the central team of the ART-CC Collaboration and provided substantial advice during statistical analysis and commented the manuscript. All authors read and commented on the original manuscript and all agreed on the version finalized by EB for submission.
[fig] Figure 1: Flow chart of the selection of the study sample of patients infected only with HIV-2 and treated with a recommended antiretroviral treatment regimen, IeDEA West Africa Collaboration. ART, antiretroviral treatment; NNRTI, non-nucleoside reverse-transcriptase inhibitor; PI, protease inhibitor; NRTI, nucleoside reversetranscriptase inhibitors indinavir for 25.3% (n072) and saquinavir for three patients (1.1%). In the unboosted PI group 88 (84.6%) patients were treated with indinavir and 16 (15.4%) with nelfinavir. [/fig]
[fig] Figure 2: Mean adjusted CD4 count change after antiretroviral treatment initiation according to baseline CD4 count (2a Á top panel) and by antiretroviral treatment regimen (2b Á bottom panel), IeDEA West Africa Collaboration. [/fig]
[fig] Figure 3: Adjusted mean CD4 count change at M12 (cells/ml) with 95% confidence interval, by baseline CD4 count (cells/ml) for the reference group (female treated by a first-line antiretroviral treatment regimen including a boosted protease inhibitor). Estimation by multivariable linear mixed model for patients remaining in care (n0265) and for all patients (n0422), IeDEA West Africa Collaboration. [/fig]
[table] Table 1: Baseline characteristics of the HIV-2-infected patients on antiretroviral therapy (ART), IeDEA West Africa Collaboration (n 0422) IQR, interquartile range; PI, protease inhibitor; NRTI, nucleoside reverse-transcriptase inhibitor; WHO, World Health Organization; a chi-square test for qualitative variables and Kruskal-Wallis test for quantitative variables; b during the first 12 months after ART initiation; c Fisher's exact test. [/table]
[table] Table 2: Mean CD4 count changes at 6 and 12 months compared to the reference group a and estimated with multivariable linear mixed model (N 0422; 1341 observations), IeDEA West Africa Collaboration [/table]
|
Assessing the validity and reliability of family factors on physical activity: A case study in Turkey
BackgroundChildhood obesity rates have been rising rapidly in developing countries. A better understanding of the risk factors and social context is necessary to inform public health interventions and policies. This paper describes the validation of several measurement scales for use in Turkey, which relate to child and parent perceptions of physical activity (PA) and enablers and barriers of physical activity in the home environment.MethodThe aim of this study was to assess the validity and reliability of several measurement scales in Turkey using a population sample across three socio-economic strata in the Turkish capital, Ankara. Surveys were conducted in Grade 4 children (mean age = 9.7 years for boys; 9.9 years for girls), and their parents, across 6 randomly selected schools, stratified by SES (n = 641 students, 483 parents). Construct validity of the scales was evaluated through exploratory and confirmatory factor analysis. Internal consistency of scales and test-retest reliability were assessed by Cronbach's alpha and intra-class correlation.ResultsThe scales as a whole were found to have acceptable-to-good model fit statistics (PA Barriers: RMSEA = 0.076, SRMR = 0.0577, AGFI = 0.901; PA Outcome Expectancies: RMSEA = 0.054, SRMR = 0.0545, AGFI = 0.916, and PA Home Environment: RMSEA = 0.038, SRMR = 0.0233, AGFI = 0.976). The PA Barriers subscales showed good internal consistency and poor to fair test-retest reliability (personal α = 0.79, ICC = 0.29, environmental α = 0.73, ICC = 0.59). The PA Outcome Expectancies subscales showed good internal consistency and test-retest reliability (negative α = 0.77, ICC = 0.56; positive α = 0.74, ICC = 0.49). Only the PA Home Environment subscale on support for PA was validated in the final confirmatory model; it showed moderate internal consistency and test-retest reliability (α = 0.61, ICC = 0.48).
[formula] a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 [/formula]
# Introduction
Obesity rates in both children and adults have been rising around the world. The rising levels of obesity in developing countries-some now outpacing those in developed countries-is of particular concern.[1] Middle Eastern and Eastern European countries have been shown to have some of the highest prevalence rates of childhood overweight and obesity among developing nations. [bib_ref] Childhood Overweight, Obesity, and the Metabolic Syndrome in Developing Countries, Kelishadi [/bib_ref] In Saudi Arabia, overweight prevalence in male children (ages 6-18 years) was at 11.2%, and obesity at 15.8%. [bib_ref] Childhood Overweight, Obesity, and the Metabolic Syndrome in Developing Countries, Kelishadi [/bib_ref] In Lebanon boys ages 6-8 years, the prevalence of overweight was 26% and obesity was 7%, while the rates in girls were 25% and 6%, respectively. [bib_ref] Childhood Overweight, Obesity, and the Metabolic Syndrome in Developing Countries, Kelishadi [/bib_ref] In Turkey, recent estimates placed the prevalence of overweight and obesity in youth 10-19 years at 18.3%, [bib_ref] Prevalence of metabolic syndrome in schoolchildren and adolescents in Turkey: a population-based..., Cizmecioglu [/bib_ref] and in certain areas of the country nearly one in four children aged 6-16 years was found to be overweight or obese. [bib_ref] Current Status of Childhood Obesity and its Associated Morbidities in Turkey, Bereket [/bib_ref] Similarly, the Childhood Obesity Surveillance Initiative (COSI) found in a nationally representative sample that the prevalence of overweight and obesity in 7-8-year-old Turkish children was 14.2% and 8.3%, respectively.However, our most recent study among children in Ankara-the second largest city in Turkey-suggests that the prevalence of overweight (21.2%) and obesity (14.6%) may be much higher in large metropolitan regions within Turkey.Current data shows significant differences in the prevalence of adult overweight and obesity between urban and rural areas in Turkey,with urban children having a higher risk of becoming overweight and obese. One study estimated that in Turkish urban children aged 10-19 years, over one in five was obese, which was twice the rate seen in this age group for rural areas. [bib_ref] Current Status of Childhood Obesity and its Associated Morbidities in Turkey, Bereket [/bib_ref] In addition, in COSI, 9.6% of younger urban children aged 7-8 years were obese compared to 3.3% in rural areas.These findings indicate a significant need for studies to improve our understanding of factors that contribute to the high prevalence of childhood obesity as well as potential intervention strategies in urban communities.
Complex behavioral, social, and environmental changes interact to promote the development of obesity, [bib_ref] A systems-oriented multilevel framework for addressing obesity in the 21st century, Huang [/bib_ref] and international childhood obesity research has highlighted the need to address these multiple levels of factors that contribute to the obesity epidemic.The social context surrounding the development of obesity in middle income countries such as Turkey is not well understood, and research in this area is needed to help guide public health interventions and policy. [bib_ref] A systems-oriented multilevel framework for addressing obesity in the 21st century, Huang [/bib_ref] Understanding the socio-cultural environment in which obesity is perceived is essential to designing effective obesity interventions. [bib_ref] A systems-oriented multilevel framework for addressing obesity in the 21st century, Huang [/bib_ref] [bib_ref] Behavioral science at the crossroads in public health: Extending horizons, envisioning the..., Glass [/bib_ref] Childhood obesity has been shown to increase the risk of chronic diseases in adulthood such as cardiovascular disease, type 2 diabetes, and certain cancers.The health behaviors of the parents and the home food and physical activity environment all influence children's lifestyle and habits significantly. [bib_ref] The role of parents in preventing childhood obesity, Lindsay [/bib_ref] In a number of studies, parental overweight or obesity has been shown to be an independent risk factor for child overweight and obesity, likely due to a combination of genetic and environmental factors. [bib_ref] The role of family and maternal factors in childhood obesity, Gibson [/bib_ref] [bib_ref] Influence of diet, physical activity and parents' obesity on children's adiposity: a..., Maffeis [/bib_ref] [bib_ref] Birth weight and parental BMI predict overweight in children from mothers with..., Schaefer-Graf [/bib_ref] [bib_ref] Association between overweight or obesity and household income and parental body mass..., Wang [/bib_ref] [bib_ref] Predicting Obesity in Young Adulthood from Childhood and Parental Obesity, Whitaker [/bib_ref] There are a number of barriers in the home, neighborhood, and school environments that can inhibit physical activity. Perceived barriers such as lack of access, weather, safety, etc. have been shown to reduce the level of physical activity in high school students and adults. [bib_ref] Physical activity and sedentary behavior: a population-based study of barriers, enjoyment, and..., Salmon [/bib_ref] [bib_ref] Perceived Barriers to Physical Activity among High School Students, Allison [/bib_ref] Research on parents shows that similar issues such as lack of social support, competing priorities for time, and financial concerns act as barriers to their ability to promote healthy behaviors and weight at home for their children. [bib_ref] Parental perceptions regarding healthy behaviours for preventing overweight and obesity in young..., Pocock [/bib_ref] In one study, parents who reported a lack of easy access to outdoor play areas for their daughters also reported lower use of active transportation by their daughters (i.e., walking, biking). [bib_ref] Perceptions about the local neighborhood and walking and cycling among children, Timperio [/bib_ref] Outcome expectations are personal factors within Bandura's Social Cognitive Theory,which influence health behaviors in people-the more positive the outcome expectations are, the more likely the person will be to engage in that behavior. [bib_ref] Health Promotion by Social Cognitive Means, Bandura [/bib_ref] In children, their beliefs about the positive or negative results of performing a particular health behavior (outcome expectancies) have been shown to be related to perceived benefits and attitudes,as well as to have the ability to modify self-regulatory skills for maintenance of behavior change. [bib_ref] Social Cognitive Determinants of Nutrition and Physical Activity Among Web-Health Users Enrolling..., Winett [/bib_ref] This is consistent with the review conducted by the World Health Organization (WHO) that reported that correlates of youth physical activity include perceived benefits and attitudes.The home and family environment has also been shown to affect physical activity levels in children. Parenting practices and behaviors related to food and physical activity have been linked to the development and establishment of health behaviors among children, which ultimately contributes to their risk of obesity. [bib_ref] Preventing childhood obesity: what works?, Birch [/bib_ref] Specifically, parental support for physical activity can influence physical activity levels in children. Children who receive more parental support from parents to be physically active (encouragement, transportation, shared activities) reported higher levels of physical activity. [bib_ref] Determinants of exercise in children, Stucky-Ropp [/bib_ref] [bib_ref] Parental behavior in relation to physical activity and fitness in 9-year-old children, Sallis [/bib_ref] [bib_ref] A review of correlates of physical activity of children and adolescents, Sallis [/bib_ref] [bib_ref] Evaluating a model of parental influence on youth physical activity, Trost [/bib_ref] [bib_ref] Parents' activity-related parenting practices predict girls' physical activity, Davison [/bib_ref] Information regarding these factors in Turkey and other middle-income countries is limited, but some evidence exists showing a significant relationship between parental and child obesity. [bib_ref] Effect of activity and television viewing on BMI z-score in early adolescents..., Ö Zmert [/bib_ref] This paper is part of a larger study, the Childhood Obesity Study of Ankara (COSA), a population study across three socio-economic strata in the Turkish capital, Ankara. In this paper, we aim to validate several measurement scales in Turkey that have been previously validated in other countries: the Parent Physical Activity Barriers scale, [bib_ref] Physical activity and sedentary behavior: a population-based study of barriers, enjoyment, and..., Salmon [/bib_ref] the Child Physical Activity Outcome Expectancies Scale, [bib_ref] Measurement characteristics of activity-related psychosocial measures in 8-to 10-year-old African-American girls in..., Sherwood [/bib_ref] and the Child Physical Activity Home Environment scale. [bib_ref] Physical Activity among African-American Girls: The Role of Parents and the Home..., Adkins [/bib_ref] These scales relate to child perceptions of physical activity and enablers and barriers of physical activity in the home environment. The validation of an existing psychological instrument in a new population is a vital step in the process of adaptation. Validation of an existing tool allows the researchers to ensure the tool is culturally appropriate, and that the meaning and difficulty of the items are suitable and conceptually equivalent to the original. This ultimately allows for easier comparisons between populations and a greater ability to generalize findings. [bib_ref] Issues, designs, and technical guidelines for adapting tests into multiple languages and..., Hambleton [/bib_ref] Validation of these scales in Turkey will further research on family factors in obesity and the design and testing of interventions targeting these risk factors.
# Methods
## Partnership
In order to facilitate research in this area, a unique partnership was formed between the University of Nebraska Medical Center in the US and Hacettepe University Institute of Public Health in Ankara, Turkey. A memorandum of understanding was signed by the two public health institutions in the fall of 2013 as part of a broader collaboration agreement between the two institutions. This collaboration promotes the advancement of obesity and health research in Turkey and globally. This project was approved by the Non-interventional Clinical Researches Ethics Board of Hacettepe University, Ankara, Turkey.
## Research setting
In this study, a population-based random stratified survey of 641 students and 537 parents in three socioeconomic strata (SES) in Ankara was conducted that included individual and family psychosocial and behavioral risk factors related to the development of childhood overweight and obesity and that may be associated with parental support.
## Study design
Investigators from the University of Nebraska Medical Center assisted with survey development, followed by survey administration by the investigators at Hacettepe University to parents and children through local schools. Measures from the existing literature were adapted and translated, and then back-translated. The survey instruments were then piloted in a dozen parent-child dyads in a school not part of the study to gauge feasibility and time requirement. Participants were asked to complete the survey and interviewed to determine any issues with the survey translation or adaptation (e.g., is the survey wording clear, are response options compatible with participant experiences, etc.). Subsequently, the surveys were examined by Turkish linguists to fine-tune the language. After the establishment of survey language and feasibility, the surveys were administered across 6 randomly selected schools to children in grade 4 (9-10-years-old) and their parents, stratified by SES (n = 641 students, 537 parents).
## Data collection & measurement
In this study, a stratified random sampling design was used. Stratification of the primary schools in Ankara was achieved by ranking counties according to SES level (low-middle-high), based on previously reported socio-economic indicators and social structures. [bib_ref] Ankara Kentinde Sosyo-mekânsal Farklılaşmanın Ö rüntüleri Ampirik Bir Analiz, Yüceşahin [/bib_ref] The high SES stratum consisted exclusively of the private schools, with the public schools of Cankaya and Yenimahalle counties forming the middle SES, and the lower SES stratum was formed by public schools from Altındağ, Mamak and Sincan counties. The sampling unit within each stratum was 4th grade classrooms. This validation study began with the completion of survey translations and user testing of final survey instruments for parents and Grade 4 children (mean age = 9.7 years for boys; 9.9 years for girls), followed by selection of approximately 650 parent-child dyads from randomly selected schools in order to assess test-retest reliability and validate the scale. Within each school, a minimum of 80-100 students were recruited into the study via the random selection of 2-5 classrooms by taking into account density of classrooms of the school. All classes were included in some schools if the number of Grade 4 students were below 80. The validation surveys were administered to parent-child dyads twice over a 3-week interval to assess test-retest reliability. The surveys were given at six schools, including 2 in each SES category. Each school sent information regarding the study and informed consent to parents, and passive student assent was sought. For both administrations, children were given a packet with the child survey, which was filled out at school, and the parent survey, which was taken home and asked to be returned within 3 days. The surveys were labeled with unique survey numbers, as well as the individual students' identification numbers. No physical measurements were taken in phase I validation surveys. Parent and child surveys were evaluated separately and were not matched for the analysis performed in the current paper. The results from Phase I helped to inform the full implementation (Phase 2) of COSA in 46 schools, in which parent and child data are being matched and analyzed together (not discussed in this paper).
## Measurement of constructs
Physical activity barriers (parent survey). Barriers such as lack of access, weather, safety, etc. have been shown to inhibit physical activity. Timperio et. al. showed that in parents who reported that their daughters were not able to easily access play areas, the girls were less likely to use active transportation (walking, biking) to get to local recreation areas. [bib_ref] Perceptions about the local neighborhood and walking and cycling among children, Timperio [/bib_ref] The validity and reliability of a 16-item scale measuring perceived barriers to physical activity scale in high school students was initially measured by [bib_ref] Perceived Barriers to Physical Activity among High School Students, Allison [/bib_ref]. [bib_ref] Perceived Barriers to Physical Activity among High School Students, Allison [/bib_ref] The scale was found to have a two factor structure-composed of personal/individual barriers, and social/environmental barriers. Salmon et. al also explored the association between physical activity level and perceived barriers, validating a modified 13-item version of the Allison scale in adults (α = 0.73).
[17] While Salmon et al discussed the scale as having two factors, analysis was performed only on the scale as a whole. In the present study, the previously validated 13-item Likert scale [bib_ref] Physical activity and sedentary behavior: a population-based study of barriers, enjoyment, and..., Salmon [/bib_ref] was used to evaluate the parents' perceived barriers to physical activity for their children. Responses ranging from (1) not a barrier to (5) very much a barrier were used to evaluate the following potential personal and environmental physical activity barriers: cost, weather, safety, pollution, no access, no sidewalk, age, disability or injury, tired, lack of time, work commitments, family commitments, and other priorities.
Physical Activity Outcome Expectancies (child survey). The construct of physical activity outcome expectancies refers to the motivational determinants shown to influence physical activity level. The validity and reliability of psychosocial measures examining outcome expectancies for physical activity in 8-11-year-old African American girls was shown in the 17-item Outcome Expectancies Likert scale in the Girls health Enrichment Multisite Study (GEMS). [bib_ref] Measurement characteristics of activity-related psychosocial measures in 8-to 10-year-old African-American girls in..., Sherwood [/bib_ref] This scale was further divided into positive and negative outcome expectancies. For the Positive Outcome expectancy measure, the internal consistency estimate was α = 0.72 and the testretest reliability was r = 0.22. For the Negative Outcome Expectancy measure, the internal consistency estimate was α = 0.68 and the test-retest reliability was r = 0.38. Positive outcome expectancies were measured by participant selection of (1) true of me; (2) sort of true of me; or (3) not true of me as responses to the following questions: "doing physical activity will. . ." make me stronger; keep me from gaining weight; teach me about health and fitness; make me look better; help me to have more energy; make me better at sports; be fun to do with my friends; and be fun. Negative outcome expectancy statements included: make me feel like I am not as good at sports as other kids; make others tease me; make me too tired; make me feel clumsy; be hard because I am often chosen last to be on a team; take too much time; cause me to get hurt; mess up my hair; and make me sweat too much.
Physical activity home environment (child survey). In addition to outcome expectancies, the effect of the home environment on physical activity among 8-11 year old African American girls was also explored in the Girls health Enrichment Multisite Study (GEMS). [bib_ref] Physical Activity among African-American Girls: The Role of Parents and the Home..., Adkins [/bib_ref] Previous studies have shown that girls who lived in more physical activity promoting environments, such as those with access to safe play spaces and sports equipment, reported higher physical activity levels. [bib_ref] A review of correlates of physical activity of children and adolescents, Sallis [/bib_ref] In the present study, the role of the home environment was investigated in students through the students' perception of parent support through a 2-item scale looking at parent permissiveness for sedentary activities (α = 0.86) and a 5-item scale looking at the students' perception of parent support for physical activity at home (α = 0.90). For the subscale on parental permissiveness of sedentary activities, the participants rated their response from (1) almost never to (3)
# Data analysis
Statistical analysis was performed using IBM SPSS v. 23 and IBM AMOS v. 23. For this validation study, initial analysis included descriptive measures (means, frequencies, etc.) for all measures including demographics, variables, and scales. Dependent variables and scales were assessed for outliers and tested for normality using visual assessments and the Kolmogorov-Smirnov test. As needed, measures were transformed to normal distributions. Scale validity was assessed using exploratory and confirmatory factor analysis. The initial round of surveys (test) were used for the exploratory factor analyses, and then the confirmatory factor analyses were done with the second round of surveys (re-test). Varimax rotation was used in exploratory factor analyses for Physical Activity Barriers and Physical Activity Outcome Expectancies scales, but factors were allowed to correlate in confirmatory factor analyses. Rotated factor loadings of at least 0.32 were considered to be significant (using a two-tailed alpha of 0.01). [bib_ref] A Beginner's Guide to Factor Analysis: Focusing on Exploratory Factor Analysis, Yong [/bib_ref] Model fit was determined using the following statistical tests: root mean square error of approximation (RMSEA) for closeness of fit (good fit = <0.06, acceptable = <0.08) [bib_ref] Structural Equation Modelling: Guidelines for Determining Model Fit, Hooper [/bib_ref] [bib_ref] An Empirical Evaluation of the Use of Fixed Cutoff Points in RMSEA..., Chen [/bib_ref] , standardized root mean square residual (SRMR) (good fit = < .05, acceptable = <0.08) [bib_ref] Structural Equation Modelling: Guidelines for Determining Model Fit, Hooper [/bib_ref] to determine the difference between the sample covariance matrix and the model covariance matrix, and adjusted goodness of fit index (AGFI) (good fit = >0.9) [bib_ref] Structural Equation Modelling: Guidelines for Determining Model Fit, Hooper [/bib_ref] to determine the proportion of variance accounted for by the model. Internal consistency of scales and test-retest reliability were assessed by Cronbach's alpha, intra-class correlation (ICC), and Spearman's correlation coefficient.
# Results
The dataset consisted of 641 students (n = 345 boys and 296 girls) and 483 parents (n = 108 male and 375 female) that completed both surveys in the three-week period. The age range for the students was 9-10 years (mean age = 9.7 years for boys; 9.9 years for girls), with 38% coming from low, 32% from middle, and 30% from high SES households. The mean age of parents was 37.6 ± 5.7 years (male: 36.6 ± 5.4, female: 40.8 ± 5.8), with 38% coming from low, 36% from middle, and 25% from high SES households. Descriptive statistics are shown in [fig_ref] Table 1: Demographics of study population [/fig_ref].
## Physical activity barriers (parent)
The construct of physical activity barriers was measured in the parents. The Physical Activity Barriers scale contained 11 items. Exploratory factor analysis yielded two factors: personal Factor loadings are shown in [fig_ref] Table 2: Individual item factor loadings in Turkish and English for the Physical Activity... [/fig_ref]. The confirmatory model [fig_ref] Fig 1: Confirmatory factor analysis of the Physical Activity Barriers Scale [/fig_ref] had a RMSEA value of 0.076, SRMR of 0.0577, and an AGFI of 0.901.
## Physical activity outcome expectancies (child)
Using exploratory factor analysis, the scale of Physical Activity Outcome Expectancies was measured in the students. The scale contained 17 items. Exploratory factor analysis found two factors in this scale: negative outcome expectancies (α = 0.77, ICC = 0.56, ρ = 0.60) and positive outcome expectancies (α = 0.74, ICC = 0.49, ρ = 0.48). [fig_ref] Table 3: Individual item factor loadings in Turkish and English for the Physical Activity... [/fig_ref] shows the factor loadings. The confirmatory model showed
## Physical activity home environment (child)
Using exploratory factor analysis, the construct of Physical Activity Home Environment was evaluated in the students. The scale contained 7 items and was shown to have two factors: support for physical activity (α = 0.65, ICC = 0.48 ρ = 0.49), and permissiveness for sedentary activities (α = 0.55, ICC = 0.55, ρ = 0.40), with good model fit statistics (RMSEA = 0.038, SRMR = 0.0288, AGFI = 0.976). However, the confirmatory model showed that the model fit best with only one factor in terms of support for physical activity, with 5 items (RMSEA value of 0.061, SRMR of 0.0288, and an AGFI of 0.968). The factor loadings are shown in [fig_ref] Table 4: Individual item factor loadings in Turkish and English for the factor loadings... [/fig_ref]. When both factors were included, this resulted in a standardized estimate larger than 1 (1.58) and negative variance in the model for one of the items loading on permissiveness for sedentary activities factor (as a function of the questionnaire item "My father/ mother or other adults let me play video and computer games as much as I want"). The final confirmatory model therefore did not support inclusion of the factor on permissiveness for sedentary activities. When stratified by gender, the data showed similar one-factor results. For boys the confirmatory model showed a RMSEA value of 0.092, SRMR of 0.0414, and an AGFI of 0.934, and for girls, it showed a RMSEA value of 0.023, SRMR of 0.0297, and an AGFI of 0.972.
# Discussion
In light of the rapidly rising childhood obesity rates in developing countries such as Turkey, validated and reliable measures are needed to effectively explore the multiple levels of factors that contribute to the obesity epidemic. The extent of the effect parental and child risk factors have on the development of childhood obesity in Turkey is currently not known. The present study is one of the first to focus on such factors. The main focus of COSA was to increase understanding of the individual and familial factors that are associated with childhood overweight and obesity and that might be relevant to parental support for various interventions in order to help inform future obesity interventions. The Physical Activity Barriers Scale (parent) as a whole was found to have acceptable model fit statistics. Our results support the originally hypothesized two-factor structure-personal Child obesity population survey validation in Turkey barriers and environmental barriers. [bib_ref] Perceived Barriers to Physical Activity among High School Students, Allison [/bib_ref] The Physical Activity Barriers subscales showed good internal consistency and poor-to-fair test-retest reliability (personal α = 0.79, ICC = 0.29, environmental α = 0.73, ICC = 0.59). For the personal barriers subscale, our results suggest some modification may be needed to improve the test-retest reliability in the Turkish context. For the environmental barriers, our results were somewhat stronger than the results other studies have reported for similar scales. In Germany, the relationship of the physical environment to physical activity levels was examined in 9-17 year-old male and female children. [bib_ref] Validity and reliability of a self-report instrument to assess social support and..., Reimers [/bib_ref] Their findings for internal consistency were somewhat lower than ours, ranging from α = 0.42-0.64, while their ICC values for test-retest reliability were somewhat higher than our own (ICC = 0.59-0.74). Similar to our Physical Activity Barriers scale, an Australian study in grade 5 and 6 children evaluated the neighborhood physical environment, looking at the effect of factors such as accessibility, aesthetics, and safety on physical activity in these children. [bib_ref] Development and reliability of a self-report questionnaire to examine children's perceptions of..., Hume [/bib_ref] They found poor to good internal consistency values (ranging from α = 0.43-0.65), which were lower than our findings, and good to excellent test-retest reliability values (ICC = 0.72-0.88), which were significantly higher than our results.
The Physical Activity Outcome Expectancies Scale (child) as a whole was also found to have good model fit statistics (RMSEA = 0.054, SRMR = 0.0545, AGFI = 0.916). Our results support the originally hypothesized two-factor structure-negative outcome expectancies and positive outcome expectancies. [bib_ref] Measurement characteristics of activity-related psychosocial measures in 8-to 10-year-old African-American girls in..., Sherwood [/bib_ref] The Physical Activity Outcome Expectancies subscales showed good internal consistency and fair test-retest reliability (negative outcome expectancies α = 0.77, ICC = 0.49; positive outcome expectancies α = 0.74, ICC = 0.56). Our results were similar to findings from other measures of outcome expectancies. For example, in the original Saunders "Beliefs" scale, which the Sherwood outcome expectancies scale is based on, researchers examined the beliefs of children regarding physical activity outcomes (both physical and social), and found α = 0.75 for the physical outcomes scale and α = 0.58 for the social outcomes scale. [bib_ref] Development of Questionnaires to Measure Psychosocial Influences on Children's Physical Activity, Saunders [/bib_ref] The physical outcomes result is similar to our result, while the social outcomes α is somewhat lower. In Iran, the Trial of Activity in Adolescent Girls (TAAG) study found similar results to those of our study when looking at various psychosocial determinants of physical activity in 10 th grade adolescent girls. [bib_ref] Development and validation of psychosocial determinants measures of physical activity among Iranian..., Pirasteh [/bib_ref] One measure examined was decisional balancechild-perceived pros and cons to physical activity, similar to our Outcome Expectancies scale, and was found to have an α = 0.72. The Physical Activity Home Environment Scale (child) showed good model fit statistics (RMSEA = 0.061, SRMR = 0.02332, and an AGFI = 0.968) as a one-factor model (support for physical activity). Our results did not support the originally hypothesized two-factor structure-support for physical activity and permissiveness for sedentary activities due to negative variance in the model. [bib_ref] Physical Activity among African-American Girls: The Role of Parents and the Home..., Adkins [/bib_ref] Our results suggest that for the Physical Activity Home Environment scale, modification of the scale to reflect cultural differences may be needed to better adapt the constructs to the Turkish context. A one-factor model can be reasonably applied in research. However, it is recommended that future studies collecting data using this tool repeat the factor analysis to verify model structure and provide further evidence for the best model for the Turkish population. Other studies have shown stronger reliability in family social support measurements. In the GEMS study, the scale was validated only in urban African American girls ages 8-9 and showed α>0.86 for both subscales. The relationship of parental support to physical activity levels was also examined in Germany in 9-17-year-old male and female children. [bib_ref] Validity and reliability of a self-report instrument to assess social support and..., Reimers [/bib_ref] Unlike our findings, they showed both strong internal consistency and test-retest reliability (α = 0.78, ICC = 0.83). Also in an Australian study, a home social environment scale was shown to have good internal consistency and excellent test-retest reliability (α = 0.73, ICC = 0.84), again showing stronger results in comparison to our parental support scale. [bib_ref] Development and reliability of a self-report questionnaire to examine children's perceptions of..., Hume [/bib_ref] In Iran, the results from the TAAG study were closer to those of our study, perhaps indicating the presence of regional variability that needs to be further studied. When looking at family support for physical activity, the authors found good internal consistency and fair test-retest reliability (α = 0.72, r = 0.56). [bib_ref] Development and validation of psychosocial determinants measures of physical activity among Iranian..., Pirasteh [/bib_ref] Cultural differences in the Turkish population compared to the population in which the scales were initially validated could have contributed to the lower reliability of both the Physical Activity Barriers and Home Environment scales. For example, cultural norms regarding physical activity or variations in digital access for children can differ between countries. Research has shown that rates of internet access for Turkish youth are still lower than Western levels, and that a significant gender divide in Internet access and use also exists. [bib_ref] Youth of Turkey Online-An Exploratory Study of the Turkish Digital Landscape, Unicef New York D Of [/bib_ref] In addition, the rate of participation in organized sports is low in Turkey, with a significant gender gap existing in this area as well.The Turkish scales may need to be adapted to reflect these cultural differences. Our results provide general support for the direct adaptation and use of the Physical Activity Barriers, Physical Activity Outcome Expectancies, and the Physical Activity Home Environment scales in Turkey for measurement of psychosocial determinants of childhood obesity. Though some variability has been detected in comparison to studies from other countries, our confirmatory factor analysis demonstrates the applicability of these scales in Turkey with some small adaptations. Methodological research on such scales will expand scientific investigations on the contributions of sociocultural factors affecting obesity in Turkey, as well as allow for further comparisons with data from other countries.
Our study population consisted exclusively of fourth grade students and their parents from the city of Ankara, so our results may not be generalizable to other age groups or families living in rural areas of Turkey.
# Conclusions
Research using internationally validated scales will help to broaden our understanding of how social and cultural differences affect nutrition and physical exercise activity behaviors of children and families, which in turn contribute to childhood obesity. This will give rise to a deeper understanding of the obesity issues in Turkey with an eye toward more regionally tailored solutions rather than relying entirely on known evidence from non-Turkish settings.
The present study strengthens the research capacity for addressing obesity in Turkey. Results from future studies utilizing these scales will be able to directly inform intervention design and implementation to prevent or reduce childhood obesity and have the potential to initiate a cohort of families for longitudinal follow-up. This research will further contribute to the body of knowledge on childhood obesity in Turkey and the Eastern Mediterranean region as well as partnerships between researchers in the U.S. and Turkey, enhancing the research collaboration between the two countries and regions.
Supporting information S1 Dataset. (ZIP) Annem/babam veya diğer yetişkinler benimle yürüyüşe çıkarlar My father/mother or other adults take walks with me. .643
Annem/babam veya diğer yetişkinlerin TV seyretmek yerine fiziksel olarak aktif olmam için uğraşırlar My mother/father or other adults encourage me to be physically active instead of watching TV. .636
Annem/babam veya diğer yetişkinler hava güzel olduğu zaman dışarıda oynamam için uğraşırlar My mother/father or other adults encourage me to play outside when the weather is good. .590
Yaşadığım yerin yakınında dışarıda oyun oynamak güvenlidir It is safe to play outside close to my house.
.507
Annem/babam veya diğer yetişkinler istediğim kadar TV seyretmeme izin verirler My father/mother or other adults let me to watch TV as much as I want.
.833
Annem/babam veya diğer yetişkinler video ve bilgisayar oyunlarını istediğim kadar oynamama izin verirler My father/mother or other adults let me play video and computer games as much as I want.
.796
[fig] Fig 1: Confirmatory factor analysis of the Physical Activity Barriers Scale (Parent) showing model fit statistics and individual factor loadings for two dimensions -Personal barriers and environmental barriers.https://doi.org/10.1371/journal.pone.0197920.g001 [/fig]
[table] Table 1: Demographics of study population. [/table]
[table] Table 2: Individual item factor loadings in Turkish and English for the Physical Activity Barriers Scale (Parent)showing two dimensions. [/table]
[table] Table 3: Individual item factor loadings in Turkish and English for the Physical Activity Outcome Expectancies Scale (Child) showing two dimensions. [/table]
[table] Table 4: Individual item factor loadings in Turkish and English for the factor loadings for the Physical Activity Home Environment Scale (Child) showing two dimensions.Ailem genellikle fiziksel olarak etkindirMy family is usually physically active. .683 [/table]
|
Faster Smith-Waterman database searches with inter-sequence SIMD parallelisation
Background: The Smith-Waterman algorithm for local sequence alignment is more sensitive than heuristic methods for database searching, but also more time-consuming. The fastest approach to parallelisation with SIMD technology has previously been described by Farrar in 2007. The aim of this study was to explore whether further speed could be gained by other approaches to parallelisation.Results: A faster approach and implementation is described and benchmarked. In the new tool SWIPE, residues from sixteen different database sequences are compared in parallel to one query residue. Using a 375 residue query sequence a speed of 106 billion cell updates per second (GCUPS) was achieved on a dual Intel Xeon X5650 six-core processor system, which is over six times more rapid than software based on Farrar's 'striped' approach. SWIPE was about 2.5 times faster when the programs used only a single thread. For shorter queries, the increase in speed was larger. SWIPE was about twice as fast as BLAST when using the BLOSUM50 score matrix, while BLAST was about twice as fast as SWIPE for the BLOSUM62 matrix. The software is designed for 64 bit Linux on processors with SSSE3. Source code is available from http://dna.uio.no/swipe/ under the GNU Affero General Public License. Conclusions: Efficient parallelisation using SIMD on standard hardware makes it possible to run Smith-Waterman database searches more than six times faster than before. The approach described here could significantly widen the potential application of Smith-Waterman searches. Other applications that require optimal local alignment scores could also benefit from improved performance.
# Background
The alignment of two biological sequences is a fundamental operation that forms part of many bioinformatics applications, including sequence database searching, multiple sequence alignment, genome assembly, and short read mapping.
Smith and Waterman [bib_ref] Identification of common molecular subsequences, Tf Smith [/bib_ref] described a simple and general algorithm requiring O(N 3 ) time and O(N 2 ) memory to identify the optimal local sequence alignment score using a substitution score matrix and a general gap penalty function. Gotoh [bib_ref] An improved algorithm for matching biological sequences, Gotoh [/bib_ref] showed that with affine gap penalties the optimal local alignment score could be computed in just O(N 2 ) time and O(N) memory.
When the optimal alignment score needs to be computed many times, for example when searching a sequence database, the computation time becomes substantial. Several approaches have been pursued to reduce the time needed. Heuristic approaches like BLAST [bib_ref] Basic local alignment search tool, Sf Altschul [/bib_ref] [bib_ref] Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Sf Altschul [/bib_ref] are considerably faster, but are not guaranteed to discover the optimal alignment.
Reconfigurable hardware in the form of FPGA (Field-Programmable Gate Array) can also accelerate the speed of alignment score computations. Li et al [bib_ref] 160-fold acceleration of the Smith-Waterman algorithm using a field programmable gate array..., Its Li [/bib_ref]. reported speeds equivalent to about 23.8 billion cell updates per second (GCUPS) with DNA sequences, on a state-ofthe-art FPGA board. Search speed is often reported in GCUPS, which indicates the billion (giga) number of cells in the alignment matrix (query sequence length times total number of database residues), processed per second.
The algorithm can also be implemented with various forms of parallelisation in software running on more common hardware. Pairwise alignment of separate sequences is in principle "embarrassingly" parallel because the computations for each pair of sequence are completely independent. Alpern et al [bib_ref] Microparallelism and high performance protein matching, B Alpern [/bib_ref]. suggested improving speed by performing several independent alignment score computations in parallel by dividing the bits of wide registers into several narrower units and using instructions to perform arithmetic operations on these units individually. This form of parallelism within a register was later made much simpler and easier by microprocessor manufacturers with the introduction of technologies like MMX, SSE, SSE2, MAX, MVI, VIS, and AltiVec, which are now generally referred to as SIMD technology. Several implementations take advantage of the SSE2 instructions available on Intel processors. The approach where parallelisation is carried out across multiple database sequences is also known as inter-task parallelisation, in contrast to intra-task parallelisation where the parallelisation occurs within a single pair of sequences.
Efforts have since mostly concentrated on parallelisation within a single alignment of one pair of sequences. illustrates the main approaches. Wozniak [bib_ref] Using video-oriented instructions to speed up sequence comparison, Wozniak [/bib_ref] suggested computing cells along the minor diagonal in the alignment matrix in parallel because these calculations are independent. Rognes and Seeberg [bib_ref] Six-fold speed-up of Smith-Waterman sequence database searches using parallel processing on common..., Rognes [/bib_ref] found that using cells along the query sequence was faster despite some data dependences, because loading values along the minor diagonal was too complicated. Farrar [bib_ref] Striped Smith-Waterman speeds database searches six times over other SIMD implementations, Farrar [/bib_ref] introduced a "striped" approach where computations were carried out in parallel in several separate stripes covering different parts of the query sequence to reduce the impact of some of the computational dependencies. Farrar's striped approach is generally the fastest, and he has reported speeds of more than 11 and 20 GCUPS on four and eight cores, respectively. Szalkowski et al [bib_ref] SWPS3 -fast multithreaded vectorized Smith-Waterman for IBM Cell/B.E. and x86/SSE2, Szalkowski [/bib_ref]. described the SWPS3 implementation which uses the striped approach of Farrar, claiming speeds of up to 15.7 GCUPS on a quad-core processor. The performance was highly dependent on query length though. Using the P07327 query sequence, with a typical protein length of 375 residues, SWPS3 performance was roughly 9 GCUPS.
The Cell processor manufactured by Sony, Toshiba and IBM, have one main core (Power Processing Element, PPE) and 8 minor cores (Synergistic Processing Elements, SPEs). These cores have SIMD vector processing capabilities. Cell processors are found in some IBM servers (QS20) as well as the Sony PlayStation 3 (PS3) (only 6 SPEs available). Several implementations for this processor have been described. SWPS3 by Szalkowski et al [bib_ref] SWPS3 -fast multithreaded vectorized Smith-Waterman for IBM Cell/B.E. and x86/SSE2, Szalkowski [/bib_ref]. used the Farrar approach and claimed speeds on the PS3 of up to 8 GCUPS. Also here the performance was highly dependent on query length. With the P07327 query sequence, SWPS3 performance was less than 2 GCUPS on the PS3. Wirawan et al [bib_ref] CBESW: Sequence Alignment on the Playstation 3, Wirawan [/bib_ref]. also employed the Farrar approach in their implementation called CBESW and claimed speeds on the PS3 of over 3.6 GCUPS, while the performance was 2.2 GCUPS with the P07327 query sequence. Farrar also reported speeds of 15.5 GCUPS on an IBM QS20 and up to 11.6 GCUPS on the PS3 using the same query. Rudnicki et al [bib_ref] The new SIMD Implementation of the Smith-Waterman Algorithm on, Rudnicki [/bib_ref]. described an implementation that used parallelisation over multiple database sequences on the PS3 and reported speeds approaching 9 GCUPS.
Graphics processors (GPUs) can also accelerate alignments. Several implementations have employed the CUDA interface to Nvidia GPUs. The CUDASW++ tool by Liu et al [bib_ref] An efficient implementation of Smith Waterman algorithm on GPU using CUDA, for..., Liu [/bib_ref]. reportedly performed 9.5 GCUPS on the single-GPU GeForce GTX 280 and 14.5 GCUPS on the dual-GPU GeForce GTX 295. Ligowski and Rudnicki [16] reported speeds of up to 14.5 GCUPS on the dual-GPU GeForce 9800 GX2. In 2010, Liu et al [bib_ref] 0: enhanced Smith-Waterman protein database search on CUDA_enabled GPUs based on SIMT..., Liu [/bib_ref]. reported speeds up to 17 and 30 GCUPS by CUDASW+ + 2.0 on the single-GPU GeForce GTX 280 and the dual-GPU GeForce GTX 295, respectively. Recently, Ligowski et al . reported a speed of 42.6 GCUPS on the GeForce GTX 480.
The main advantage of the inter-task parallelisation approach where multiple database sequences are processed in parallel as described by Alpern et al [bib_ref] Microparallelism and high performance protein matching, B Alpern [/bib_ref]. is that it simply avoids all data dependences within the alignment matrix. This approach does not seem to have been explored much in later implementations using SIMD database sequence(s) query sequence A C B D Approaches to vectorisation of Smith-Waterman alignments. Alignment matrices are shown with the elements that form the first five vectors processed indicated in black, blue, red, green and yellow. For simplicity, vectors of only 4 elements are shown, while 16 elements would normally be used. (A) Vectors along the anti-diagonal, described by Wozniak et al [bib_ref] Using video-oriented instructions to speed up sequence comparison, Wozniak [/bib_ref]. (B) Vectors along the query, described by Rognes and Seeberg [bib_ref] Six-fold speed-up of Smith-Waterman sequence database searches using parallel processing on common..., Rognes [/bib_ref]. (C) Striped approach, described by Farrar [bib_ref] Striped Smith-Waterman speeds database searches six times over other SIMD implementations, Farrar [/bib_ref]. (D) Multi-sequence vectors, described by Alpern et al [bib_ref] Microparallelism and high performance protein matching, B Alpern [/bib_ref]. and in this paper.
technology apart from the work by Rudnicki et al [bib_ref] The new SIMD Implementation of the Smith-Waterman Algorithm on, Rudnicki [/bib_ref]. However, in the GPU-based tools [bib_ref] An efficient implementation of Smith Waterman algorithm on GPU using CUDA, for..., Liu [/bib_ref] this approach is common. The aim of the present study was to explore the use of this approach further using SIMD on ordinary CPUs.
Here the algorithm is implemented on Intel processors with SSSE3with parallelisation over multiple database sequences as illustrated in . Instead of aligning one database sequence against the query sequence at a time, residues from multiple database sequences are retrieved and processed in parallel. Rapid extraction and organisation of data from the database sequences have made this approach feasible. The approach has been implemented in a tool called SWIPE. The approach also involves computing four consecutive cells along the database sequences before proceeding to the next query residue in order to reduce the number of memory accesses needed.
The performance of the new implementation has been extensively tested using different scoring matrices, gap penalties, query sequences and number of threads. The speed of SWIPE was almost constant at more than 100 GCUPS on a dual Intel Xeon X5650 six-core processor system for a wide range of query lengths. SWIPE was about six times faster than SWPS3 and Farrar's own implementation for a typical length query, but the factor varied between 2 and 18 depending on query length and number of threads used. Two versions of BLAST were tested and the speed was found to be highly dependent on the score matrix. SWIPE was about twice as fast as BLAST using the BLOSUM50 matrix, while BLAST was about twice as fast as SWIPE using the BLOSUM62 matrix.
# Methods
## Benchmarking
Benchmarking was performed on compute nodes in the Titan high performance cluster at the University of Oslo. Entire nodes were reserved to ensure that no other major processes were running. All data was initially copied to a fast local disk to reduce the influence of the computer networks and minimize file reading time. The output from all programs was redirected to/dev/ null to minimize performance differences due to the amount of output. All combinations of programs, number of threads, query sequences, score matrices, gap open penalties and gap extension penalties were run 15 times and the median total wall clock execution time was recorded. [fig_ref] Table 1: Programs included in performance testing [/fig_ref] lists the software packages that were benchmarked, including their version numbers and command line options used. SWIPE was written mainly in C++ with some parts hand coded in inline assembler and some using SSE2 and SSSE3 intrinsics. It was compiled for 64 bit Linux using the Intel C++ compiler version 11.1. Source code is available at http://dna.uio.no/swipe/ under the GNU Affero General Public License, version 3. An executable binary and score matrix files are also available at the same location. The same files are included in a gzipped tar archive as additional file 1.
## Software
The source code for Farrar's STRIPED software was downloaded from the author's website and compiled with the GNU gcc compiler as specified in the supplied Makefile. The Makefile was also modified to compile STRIPED using the Intel compiler, to see if there were any differences in performance.
The binary executable for the SWPS3 program was downloaded from the authors' website and used directly [bib_ref] SWPS3 -fast multithreaded vectorized Smith-Waterman for IBM Cell/B.E. and x86/SSE2, Szalkowski [/bib_ref].
Precompiled binaries for BLAST and BLAST+ were downloaded from the NCBI FTP site [bib_ref] Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Sf Altschul [/bib_ref].
Graphs were drawn using Gnuplot version 4.5.
## Hardware
Performance tests were carried out on Dell PowerEdge M610 blade servers with 48 GB RAM and dual Intel Xeon X5650 six-core processors running at 2.67 GHz. The X5650 processors have simultaneous multithreading capability also known as hyper-threading (HT). With HT enabled, each of these computers has a total of 24 logical cores.
## Threads
The programs were run using 1 to 24 threads. To take full advantage of the hardware a number of threads equal to the number of logical cores is usually the most appropriate. There are however important differences between software on the effect of HT and in the ability to make efficient use of the cores available. To simplify comparisons, some of the tests were only performed with 1 or 24 threads.
[formula] ./blastall -p blastp -F F -C 0 -b 0 -v 10 -a $T -M $M -G $GO -E $GE -i $Q -d $D BLAST+ 2.2.24+ ./blastp -seg no -comp_based_stats F -num_alignments 0 -num_descriptions 10 -num_threads $T -matrix $M -gapopen $GO -gapextend $GE -query $Q -db $D SWIPE 1.0 ./swipe -v 10 -a $T -M $M.mat -G $GO -E $GE -i $Q -d $D STRIPED ./striped -c 10 -T $T -i -$GO -e -$GE $M.mat $Q $D.fsa SWPS3 20080605 ./swps3 -j $T -i -$GO -e -$GE $M.mat $Q $D.fsa [/formula]
## Database sequences
UniProt Knowledgebase Release 11.0 [bib_ref] The Universal Protein Resource (UniProt) in 2010, Consortium [/bib_ref] consisting of both Swiss-Prot release 53.0 and TrEMBL release 36.0 of 29 May 2007 was used for the performance tests. This database consists of 4 646 608 protein sequences with a total of 1 517 383 530 amino acid residues. The longest sequence contains 36 805 residues. This database was chosen because the size was large enough to be realistic but smaller than the apparent 2 GB file size limit of some software. The database also did not contain any of the special J, O or U amino acid residue symbols that some of the software could not handle. Finally, this release of the database should be available for download in the foreseeable future, making it suitable also for future benchmarking.
The current version of SWIPE will not work with databases split by formatdb into separate volumes. SWIPE therefore cannot search databases larger than about 4 billion amino acids.
The database was converted into FASTA format by a simple Perl script and then formatted with NCBI formatdb version 2.2.24 into the NCBI BLAST binary database format [bib_ref] Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Sf Altschul [/bib_ref]. This binary format was read by SWIPE, BLAST and BLAST+, while STRIPED and SWPS3 read the FASTA-formatted database file directly.
## Query sequences
The 32 query sequences with accession numbers P56980, O29181, P03630, P02232, P01111, P05013, P14942, P00762, P53765, Q8ZGB4, P03989 (replacing the identical but obsolete P10318), P07327, P01008, P10635, P58229, P25705, P03435, P42357, P21177, Q38941, O60341, P27895, P07756, P04775, P19096, P28167, P0C6B8, P20930, P08519, Q7TMA5, P33450 and Q9UKN1, ranging in length from 24 to 5478 residues were retrieved from the UniProt database . Most of them have previously been used several times for performance testing. To simplify comparisons, some of the tests were only performed with the 375 residues long P07327 query, representing a protein of roughly average length.
## Score matrices and gap penalties
All 82 different combinations of amino acid substitution score matrices and gap penalties accepted by BLAST were tested. The matrices used were BLOSUM45, BLO-SUM50, BLOSUM62, BLOSUM80, and BLOSUM90 from the BLOSUM series [bib_ref] Amino acid substitution matrices from protein blocks, Henikoff [/bib_ref] as well as PAM30, PAM70, and PAM250 from the PAM series [bib_ref] SHRiMP: Accurate Mapping of Short Color-space Reads, Mo Dayhoff [/bib_ref]. Matrices were obtained from the NCBI FTP site. Rows and columns for stop codons (*) were removed from the matrices for compatibility with the SWPS3 program. SWPS3 would only run successfully using the BLO-SUM45, 50 and 62 matrices. To simplify comparisons, some of the tests were only performed with the BLO-SUM62 matrix and gap open and extension penalties of 11 and 1, respectively, which is the BLAST default.
# Results
## Algorithm
The optimal local alignment score of two sequences can be computed using a dynamic programming approach. The recurrence relations for the algorithm of Smith and Waterman [bib_ref] Identification of common molecular subsequences, Tf Smith [/bib_ref] with the modifications of Gotoh [bib_ref] An improved algorithm for matching biological sequences, Gotoh [/bib_ref] for affine gap penalty functions are shown below.
[formula] H i,j = ⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ max ⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩ H i−1,j−1 + P[q i , d j ] E i,j F i,j 0 i > 0 ∩ j > 0 0 i = 0 ∪ j = 0 (1) E i,j = ⎧ ⎨ ⎩ max H i,j−1 − Q E i,j−1 − R |j > 0 0 |j = 0 (2) F i,j = ⎧ ⎨ ⎩ max H i−1,j − Q F i−1,j − R |i > 0 0 |i = 0 (3) S = max 1 i m∩1 j n H i,j(4) [/formula]
The query sequence q of length m contains residues q i . The database sequence d of length n contains residues d j . H i,j is the score for aligning the prefixes of q and d ending in the alignment of residues q i and d j . E i,j and F i,j are the scores of aligning the same prefixes of q and d but ending with a gap in the query and database sequence, respectively. P[q i , d j ] is the score of aligning residues q i and d j with each other according to the substitution score matrix P. Q is the sum of gap open and extension penalties while R is the gap extension gap penalty. S is the overall optimal local alignment score.
The calculations are carried out column by column. Only parts of the H, E and F matrices need to be kept in memory: a single element of the F matrix as well as two arrays containing m elements each, corresponding to one column of the H and E matrices.
## Implementation
The main features of the implementation are described below.
## Parallelisation over sixteen database sequences
Residues from 16 different database sequences are processed in parallel as indicated in . These 16 residues are all simultaneously compared to the same query residue. The operations are carried out using vectors consisting of 16 independent bytes. The 16 residues are fed into sixteen independent channels. When the first of these sixteen database sequences ends, the first residue of the next database sequence is loaded into the channel. The databases sequences are read in the order they are found in the original database file. In contrast to the approach by Rudnicki et al. [bib_ref] The new SIMD Implementation of the Smith-Waterman Algorithm on, Rudnicki [/bib_ref] , the database is not sorted by sequence length. illustrates this approach.
## Compact core code of ten instructions
The basis for the computations of the values in each cell in the alignment matrices are the recurrence relations described in the Algorithm section. The computations can be written in as little as ten assembly instructions that constitute the core of the inner loop of the computations, as shown in . These ten instructions compute in parallel the values for each vector of 16 cells in independent alignment matrices. The exact selection of instructions and their order is important; this part of the code was therefore hand coded in assembler to maximise performance. In the figure, H represents the main score vector. The H vector is saved in the N vector for the next cell on the diagonal. E and F represent the score vectors for alignments ending in a gap in the query and database sequence, respectively. P is the vector of substitution scores for the database sequences versus the query residue q (see temporary score profiles below). Q represents the vector of gap open plus gap extension penalty. R represents the gap extension penalty vector. S represents the current best score vector. All vectors, except N are initialised prior to this code.
## Processing four consecutive cells along the database sequences
During the computations of the matrix cells, the values in the two arrays with the H and E values usually have to be read and written once for each matrix cell. These arrays are usually small enough to be cached at a close cache level, so the memory access time should not be a major concern, but they still need to be written and read back for each cell. Since there are sixteen 128-bit xxm registers available and ample space for keeping the H, E and F values of a few cells in the registers, it is possible to reduce running time somewhat by computing a few consecutive cells along the database sequences before moving on to the next query residue. Four consecutive cells was found to perform well. Unrolling the inner loop once along the query sequence was also found to work well. The basic computing blocks then consist of two times four cells that are processed in each inner loop iteration as shown in [fig_ref] Figure 4: Block of cells computed in each iteration [/fig_ref].
## Updating scores and padding blocks
When a new database sequence begins in one of the channels, the score of the previous database sequence must be recorded. In addition, the H, E and S scores from the previous sequence must be reset. When a new column is going to be processed, it is checked whether any database sequence ended in the previous column. If that is the case, special processing is carried out. The score of the sequences that ended are recorded. A mask is created and later used to reset the values of H, E and S in the appropriate channels before the new column is processed. The channels are filled, and one or more new sequences are started. A somewhat simpler and faster processing step is carried out for the new column if no sequences ended in the previous column. In the simpler processing step no scores need to be saved, no new sequences are started, and no mask need to be created or used. Since most columns are of the simpler type the overall performance is mostly dependent on the speed of processing the simple columns. To simplify computations, each channel is padded with 1-3 null symbols after the end of a sequence if its length is not a multiple of 4, to ensure that a new database sequence will only begin at the beginning of a new block of cells. This padding is indicated in pink in . The padding increases the total number of cell a little bit, but allows the checks described above to be carried out only on every fourth residue.
## Computation of temporary score profiles
To make computations fast, it is essential that the vectors of substitution score values can be loaded quickly. Each score vector corresponds to the score of a single query residue against 16 residues from different database sequences. A kind of temporary score profile is created as outlined in . This score profile is valid for matching any query residue with 4 successive residues from 16 database sequences. For every fourth residue in the database sequences a new score profile must be constructed. The temporary score profiles are created from the ordinary substitution score matrices (e.g. BLOSUM62) and the 4 × 16 database sequence residues using a series of packed shuffle instructions (pshufb). The shuffle instruction is only available on Intel processors with Supplemental Streaming SIMD Extensions 3 (SSSE3). On processors without SSSE3 (i.e. AMD processors and older Intel processors), the computations may be replaced by a kind of matrix transpose operation using a series of unpack instructions (punpcklbw, punpckhbw) with only a modest speed penalty.
## Score range and selection of arithmetic instructions
Computations are initially performed using only a 7 bit score range. This allows 16 alignment score matrices to be computed in parallel using SSE2 instructions. Additions and subtractions are performed using signed and saturated arithmetics, while the maximum operations are carried out on unsigned numbers. Only the 7 bit score range from -128 to -1 (signed numbers) or 128 to 255 (unsigned numbers) is used. All scores are biased by an offset of 128. This range of values will ensure that signed saturated addition and subtraction works well on the lower boundary. Also, the unsigned maximum works well in this range. The range enables the use of the packed maximum unsigned byte instruction (pmaxub) and packed add and subtract signed saturated bytes instructions (paddsb and psubsb), which are available on all SSE2 processors and gives the highest speed.
An 8 bit range, which would allow the same number (16) of parallel computations as a 7-bit range, and at the same time allow a wider score range, is not used because it is slower. Either the range from -128 to 127, or the range from 0 to 255 could be used. There are both signed and unsigned versions of the instructions for parallel computation of the maximum of bytes (pmaxub, pmaxsb) and for parallel addition and subtraction of bytes (paddusb, psubusb, paddsb, psubsb), but the pmaxsb instruction for the maximum of signed bytes was only recently available in SSE4.1 and is slower than pmaxub. Unsigned additions and subtractions (paddusb and psubusb) may be used, but the addition of the score matrix vector then requires two instructions. First the score vector including a bias must be added; then the bias must be subtracted.
[formula] M K F L M K F L M A S V M E K K F N S N M V A F M A S K M A R S M S T I M I F D M V L L M F H V M A R V M G R V M A E N M I F T MMMMFMMMMMMMMMMM KKAENVAASIVFAGAI FFSKSASRTFLHRREF LLVKNFKSIDLVVVNT A C D E V W Y A C D E F G H I K L M N P Q R S T V W Y A C D E F G H I K L M N P Q R S T V W Y [/formula]
## Score matrix
Database sequences Score profile Shuffling procedure Creation of a temporary score profile for 64 database sequence residues. A standard score matrix and 4 residues from 16 different database sequences form the basis for a kind of temporary partial database sequences score profile. The score profile enables rapid comparison of any query residue with these 64 database residues.
Versions using 16-bit and 63-bit score ranges are also implemented and used when overflow is detected in computations with lower score ranges. The 16-bit version allows 8 parallel computations. When potential overflow is detected in computations with a narrow score range, the alignment score for that database sequence is recalculated using the next wider score range (first 16-bit and then 63-bit if necessary). Because rather few sequences will usually reach a score that cannot be represented by 7 bits, the additional computation time for wider score ranges are usually negligible.
Recalculations with a wider score range are carried out on a subset of sequences after all sequences in a chunk of database sequences (see below) have been processed using the narrower score range.
## Reading the sequence database
Database sequences were stored in the NCBI BLAST database format, produced by the formatdb tool. This is a binary format where the sequence information is split into at least 3 files: indices (.pin), headers (.phr) and sequences (.psq). The file format allows efficient reading of sequences into memory. Protein sequences are stored using byte values in the range 1-24 and 26-27 representing amino acid residues A-I, K-N, P-T, V-Z, U, O and J, respectively. Sequences are separated by a zero byte, which simplifies the check for sequence ends.
Database sequences are retrieved using memory mapping of the .pin and .psq files. This is an efficient and convenient method of accessing the sequences, in which the operating system manages reading data into memory from disk as necessary, concurrently with program execution. The sequence database is divided into 100 chunks per thread. Each chunk contains approximately the same number of sequences. The sequences in one chunk are mapped into memory and processed before the next chunk is mapped. This results in a small memory footprint of the program.
## Multiple threads
SWIPE uses multiple threads (pthreads) that work on different parts of the sequence database. The number of threads is specified when starting the program and should in general be equal to the number of cores of the computer. For the latest generations of Intel processors with hyper-threading, a number of threads equal to the number of logical cores is usually most effective. Chunks of database sequences are assigned to the threads as they are ready for more work, so the threads may not process exactly the same number of chunks each. Results from the threads are inserted into a common hit list after each chunk is processed.
## Testing
The SWIPE software was benchmarked against BLAST, BLAST+, STRIPED and SWPS3 under many different conditions to measure speed. The performance using a variable number of threads and the effect of query sequence length was studied. Additionally, the impact of different scoring systems, both substitution score matrices and gap open and extension penalties was examined.
Threads [fig_ref] Figure 6: Performance dependency on number of threads and query length [/fig_ref] indicates the performance of the programs running with 1 to 24 threads, the 375 residue long P07327 query sequence, the BLOSUM62 matrix, and with gap open and extension penalties of 11 and 1, respectively. SWIPE runs at 9.1 GCUPS using a single thread and reaches its maximum performance with 19 threads at 106.2 GCUPS, but there is almost no gain from additional threads beyond 12. It scales very well and the maximum speed-up (the ratio of maximum speed to single thread speed) is 11.6. SWPS3 runs at 3.4 GCUPS using a single thread and reaches its best performance at 12 threads with 16.4 GCUPS, but has little gain beyond 9 threads. The maximum speed-up is 4.8. Surprisingly, the speed using two threads is inferior to that with a single thread. STRIPED compiled with the GNU compiler runs at 3.1 GCUPS with a single thread and reaches its maximum performance with 23 threads at 14.7 GCUPS, but gains little beyond 12 threads. The maximum speed-up is 4.8. Compiling STRIPED with the Intel compiler resulted in a 21% speed increase to 3.7 GCUPS when running on a single thread, but just 2% to 15.0 GCUPS with 23 threads. This corresponds to a maximum speed-up of 4.0. BLAST and BLAST+ run at 14.7 and 15.5 GCUPS, with a single thread, respectively, scale very well and reach their maximum performance when 24 threads are running with speeds of 208.4 and 178.9 GCUPS, respectively. The maximum speed-ups of BLAST and BLAST+ are 14.2 and 11.5, respectively. [fig_ref] Figure 6: Performance dependency on number of threads and query length [/fig_ref] and [fig_ref] Figure 6: Performance dependency on number of threads and query length [/fig_ref] illustrates the performance with queries of varying length using either 24 threads (B) or a single thread (C). 32 different sequences with lengths ranging from 24 to 5478 amino acid residues were used as queries.
## Query length
SWIPE had a rather flat performance curve. For queries shorter than about 100 residues there was a gradual loss in performance, especially when running with many threads. The speed was also slightly reduced for very long queries when using many threads. The performance ranged from 23.1 to 110.1 GCUPS for 24 threads and from 3.9 to 9.8 GCUPS for a single thread.
The SWPS3 program was very dependent on query length with speeds ranging from 0.94 to 49.0 GCUPS on 24 threads and between 1.1 and 4.6 GCUPS on a single thread.
The STRIPED program compiled with the Intel compiler was also quite dependent on the query length, in particular when running on 24 threads, with speeds ranging from 1.2 to 46.6 GCUPS. On a single thread, the speed of STRIPED varied between 0.8 and 5.7 GCUPS. STRIPED compiled with the GNU compiler was in general slightly slower, particularly with longer queries.
The speed of the BLAST programs seemed somewhat faster with longer query sequences with speeds ranging from 52.8 to 374.1 and 30.6 to 360.2 GCUPS for BLAST and BLAST+, respectively, but the performance varied a bit from sequence to sequence. There was a noticeable drop in performance with queries shorter than about 100 residues. [fig_ref] Figure 7: Performance with different scoring systems [/fig_ref] shows the performance under different scoring systems. All combinations of matrices and gap penalties allowed by BLAST were tested. The 375 residue long P07327 query sequence and 24 threads were used. The performance of SWIPE is almost constant at about 102-106 GCUPS. The performances of STRIPED and SWPS3 are also almost constant at about 14-15 and 15 GCUPS, respectively. The performance of BLAST was highly dependent on the scoring matrix used. SWIPE was almost twice as fast as ordinary BLAST using the BLOSUM50 matrix. The speeds of the two programs were quite similar with the PAM250 matrix, while BLAST was faster for the other matrices. In general, BLAST+ was about 10-20% slower than ordinary BLAST. Gap penalties had little impact on performance in general, but relatively low gap penalties seemed to reduce the speed of BLAST, BLAST+, and STRIPED in a few cases (e.g. BLOSUM62 with gap penalties 9 and 1), while the impact on SWIPE and SWPS3 was negligible.
## Scoring systems
# Discussion
The SWIPE software greatly increases the speed of sequence database searches based on the Smith-Waterman algorithm compared to earlier SIMD implementations, being more than six times faster in realistic cases. SWIPE was found to be performing at a speed of 106 GCUPS with a 375 residue query sequence on a dual Intel Xeon X5650 six-core processor system. The speed was a bit dependent on the query length, but independent of the scoring system used. The maximum speed corresponds to the processing of more than 3.3 cells per physical core in each clock cycle.
Using GPUs, speeds of up to 42.6 GCUPS have been reported for CUDASW++ 2.0 on a Nvidia GeForce GTX 480 graphics card . This is comparable to the expected performance of SWIPE on a quad-core CPU.
SWIPE scaled almost linearly with the number of threads used up to 12 threads, corresponding to the number of physical cores available. The X5650 processors features hyper-threading which makes it possible to obtain extra performance using more than one thread per physical core, unless all execution units of busy. Almost no increase in performance beyond 12 threads was observed, so apparently the execution units are fairly busy when SWIPE is running on 12 threads. SWIPE scaled much better than SWPS3 and STRIPED with multiple threads. The maximum speed of SWIPE was 6.5 times faster than SWPS3, the fastest of these, using the 375 residue query. When all programs were running on a single thread, SWIPE was about 2.5 times faster. There seems to be two equally important factors responsible for the differences in speed. Based on the single thread performance numbers, it seems like the use of the inter-sequence parallelisation approach instead of the striped approach is responsible for about half of the increase in speed. Efficient thread parallelisation seems responsible for the other half.
In the comparisons with the STRIPED software it should be noted that [bib_ref] Striped Smith-Waterman speeds database searches six times over other SIMD implementations, Farrar [/bib_ref] apparently reported the "scan time" and not the complete running time for the program, excluding the time needed to read the database sequences into memory. The full running time is reported here. For the other programs, it is difficult to separate the scan time from the rest of the time used. STRIPED reads the FASTA formatted files directly and not files formatted by NCBI's formatdb tool. It appears that STRIPED initially parses the entire FASTA file and reads the database into memory using non-threaded code, resulting in low performance when using several threads and measuring the complete running time. Excluding the time for database reading leads to shorter run times and higher performance numbers. On the other hand, it is probably faster to read files in the binary NCBI format than parsing the FASTA text format.
The performance of SWIPE was not very dependent on query length, except for rather short and very long queries. Overhead costs incurred for each database residue probably reduced the performance of SWIPE for the shortest queries. For the longest queries there was a performance decrease when many threads were running. This is probably due to the effects of memory caches, of which some are shared between cores. The programs based on Farrar's approach were considerably more dependent on query length and performed better with increasing query length. The reason for this is probably that as the width of the stripes increase, the relative importance of the dependency between the stripes is reduced.
SWIPE was about twice as fast as BLAST using the BLOSUM50 matrix, while BLAST was twice as fast as SWIPE using the BLOSUM62 matrix. BLAST performance was found to be very dependent on the scoring matrix. The reason may be that BLAST in its heuristics uses an initial hit score threshold (T) that has a fixed default value (11) independent of the score matrix specified. Score matrices with relatively high expected values, e.g. BLOSUM50, will then trigger more initial hits than other matrices, e.g. PAM30.
If one would like to search using a query profile (position-specific scoring matrix) instead of a query sequence, the computation of the temporary score profile need to be carried out for every position of the query, not just for the 20 possible amino acid residues. This has not been implemented, but the resulting reduction in speed has been estimated to 30%.
The software described here should be considered a prototype indicating the performance potential of the approach. A later version that at least computes the actual alignments (not just the alignment score) and the statistical significance of the matches is planned.
# Conclusions
Efficient parallelisation using SIMD on standard hardware now allows Smith-Waterman database searches to run considerably faster than before. In the new tool SWIPE, residues from sixteen different database sequences are compared in parallel to one query residue. Using a 375 residue query sequence a speed of 106 billion cell updates per second (GCUPS) was achieved on a dual Intel Xeon X5650 six-core processor system, which is more than six times faster than software based on Farrar's approach, the previous fastest implementation. Furthermore, for the first time, the speed of a Smith-Waterman based search has been shown to clearly exceed that of BLAST at least with one particular scoring matrix.
Since the slow speed has been the major drawback limiting the usefulness of Smith-Waterman based searches, the approach described here could significantly widen the potential application of such searches. Other applications that require optimal local alignment scores, like short read mapping or genome sequence assembly [bib_ref] Assembly algorithms for next-generation sequencing data, Jr Miller [/bib_ref] could also benefit from improved performance of this method. The approach used here may probably also be applied to HMMbased searches.
# Additional material
Additional file 1: Source code. The source code of SWIPE version 1.0, as well as a binary executable for 64-bit Linux and score matrices are included in this gzipped tar archive file.
[fig] Figure 2, Figure 3: Blocks of database residues processed together. The residues of the first five vectors processed are indicated on grey, blue, red, green and yellow background. Four symbols from each of the database sequences form blocks that are processed as a group. Padding of some sequence blocks are indicated with dashes on a pink background. An additional database sequence change is indicated with a pink vertical bar. Arrows below the sequences indicate positions were new sequences begin. Core instructions. These are the ten core instructions executed for each vector of cells in the alignment matrices. In this code, the first (left) operand is the source and the second (right) operand is the destination. [/fig]
[fig] Figure 4: Block of cells computed in each iteration. In each iteration of the inner loop blocks of eight cells in the alignment matrices computed. The H, E, F and S values are updated for each cell. The computations start in the upper left cell, proceed right three times and then continue from the left on the second row. [/fig]
[fig] Figure 6: Performance dependency on number of threads and query length. The speed in billion cell updates per second (GCUPS) of the BLAST (red), BLAST+ (orange), SWIPE (black), and SWPS3 (green) programs, as well as STRIPED compiled with the GNU (light blue) and Intel (dark blue) compiler, using a variable number of threads and queries of varying length. (A) Number of threads ranging from 1 to 24 and the 375 residue long P07327 query sequence. (B) Query sequences ranging from 24 to 5478 residues in length and 24 threads. (C) Query sequences of varying length and 1 thread. All scales are logarithmic. The BLOSUM62 matrix and gap open and extension penalties of 11 and 1, respectively, were used in all cases. [/fig]
[fig] Figure 7: Performance with different scoring systems. The speed in billion cell updates per second (GCUPS) (logarithmic scale) is shown for the BLAST (red), BLAST+ (orange), SWIPE (black), and SWPS3 (green) programs, as well as STRIPED compiled with the GNU (light blue) and Intel (dark blue) compiler, using different scoring systems. All combinations of scoring matrices and gap penalties accepted by BLAST were tested. The matrix name is indicated above each graph, while the open and extension gap penalties are indicated on the x-axis. The query sequence was P07327 and 24 threads were running. SWPS3 would not run successfully in all cases. [/fig]
[table] Table 1: Programs included in performance testing [/table]
|
Performance of a core of transversal skills: self-perceptions of undergraduate medical students
Background: There is an increasingly growing trend towards integrating scientific research training into undergraduate medical education. Communication, research and organisational/learning skills are core competences acquired by scientific research activity. The aim of this study was to assess the perceived performance of a core of transversal skills, related with scientific research, by Portuguese medical students. Methods: A cross-sectional study was conducted in 611 Portuguese students attending the first, fourth and sixth years of the medical course, during the same academic year. A validated questionnaire was applied for this purpose. Results: Medical students felt confident regarding the majority of the analyzed transversal skills, particularly regarding team work capacity (72.7 % perceived their own capacity as good). On the other hand, the perceived ability to manage information technology, time and to search literature was classified only as sufficient by many of them. The progression over the medical course and participation in research activities were associated with an increasing odds of a good perceived performance of skills such as writing skills (research activity: OR = 2.00; 95 % CI: 1.34-2.97) and English proficiency (research activity: OR = 1.59; 95 % CI: 1.06-2.38/final year medical students: OR = 3.63; 95 % CI: 2.42-5.45). Conclusions: In this line, the early exposure to research activities along undergraduate medical education is an added value for students and the implementation of an integrated research program on medical curriculum should be considered.
# Background
In the 90s of the last century, following the developments in the USA, several Western European universities responded to the need looking for research capacity beyond medical students [bib_ref] Exploring the integration of the biomedical research component in undergraduate medical education, Van Schravendijk [/bib_ref]. Medical doctors are the first to experience the need for more evidence-based knowledge in order to deliver better patient care. However, their capability to participate in medical research programs, and to inspire them, is often limited by both lack of training during their studies and the reduced amount of time available for research, even in university hospitals [bib_ref] Exploring the integration of the biomedical research component in undergraduate medical education, Van Schravendijk [/bib_ref]. Consequently, there is an increasingly growing trend towards integrating scientific research training into undergraduate medical education [bib_ref] Exploring the integration of the biomedical research component in undergraduate medical education, Van Schravendijk [/bib_ref] [bib_ref] Integration of scientific research training into undergraduate medical education: a reminder call, Abu-Zaid [/bib_ref] [bib_ref] Developing research skills in medical students: AMEE Guide No. 69, Laidlaw [/bib_ref].
Medical curriculum in Europe follows the quality assurance frameworks and the portability of qualifications of the directives of Bologna Process [bib_ref] The Bologna process, medical education and integrated learning, Cumming [/bib_ref]. In Portugal, the medical course was reformulated into an Integrated Master Course in Medicine and is currently composed by a first cycle of three years, corresponding to a degree in basic health sciences, and a second cycle also lasting three years of clinical learning and professional clerkship.
At the Faculty of Medicine of the University of Porto (FMUP), along the first cycle of studies all students participate in random research activities in curricular units such as biochemistry, cellular biology and genetics, and some participate in research projects as extra-curricular activities, usually since the fourth year.
Scientific research involves the collection and processing of new information and its interpretation through logic and previous knowledge and experience [bib_ref] Developing research skills in medical students: AMEE Guide No. 69, Laidlaw [/bib_ref]. Critical thinking, communication ability, literature searching, cope with information technology, team work, solving-problems and self-improving learning are scientific skills acquired through research activity [bib_ref] Doing an intercalated BSc can make you a better doctor, Greenhalgh [/bib_ref] [bib_ref] Research and medical students, Remes [/bib_ref] [bib_ref] Skill development in graduate education, Parker [/bib_ref]. Some of these skills are required for effective decision-making being a valuable quality required in any program in higher education [bib_ref] Assessing university students' general and specific critical thinking, Cheung [/bib_ref] [bib_ref] Assessing critical thinking in medical sciences students in two sequential semesters: Does..., Athari [/bib_ref]. However, transversal skills of medical freshman students, as well as their development throughout the medical course, have not yet been assessed, especially considering the recent reforms in medical education. Our research group has been studying several issues related with undergraduate medical education [bib_ref] How students perceive medical competences: a crosscultural study between the Medical Course..., Barbosa [/bib_ref] [bib_ref] Third year medical students perceptions towards learning communication skills: implications for medical..., Loureiro [/bib_ref] [bib_ref] Scientific Skills as Core Competences in Medical Education: What Medical Students think?, Ribeiro [/bib_ref].
The aim of this study, again reflecting our concerns on this matter, was to assess the perceived performance of a core of transversal skills by undergraduate medical students.
# Methods
The specific objectives of this study were: (1) To evaluate if self-perception of medical students about transversal skills execution change over the progression of the medical course and (2) To compare the above objective by gender, type of school attended at high schooling level (public or private) and previous involvement (or not) in research projects.
A cross-sectional study was conducted in order to develop this research. Among the 796 eligible medical students, 611 (76.8 %) have participated in the study. The participants were enrolled in the Master Degree Course in Medicine of FMUP, Portugal. The study included freshman students (first year), those who concluded the first cycle of studies (beginning of the fourth year) and students who completed the entire medical course (end of the sixth year). All students from the first and fourth years answered the questionnaire at once at the beginning, and those from the six year at the end, of the same academic year.
Medical students were invited to fulfill an anonymous self-administered and validated questionnaire: "Importance of Transversal Skills for Clinical Practice" (ITS4CP) developed by the authors [bib_ref] Scientific Skills as Core Competences in Medical Education: What Medical Students think?, Ribeiro [/bib_ref]. The transversal skills included in the questionnaire were adapted from AMEE Medical Education Guide [bib_ref] Developing research skills in medical students: AMEE Guide No. 69, Laidlaw [/bib_ref].
'The perceived ability to perform the transversal skills was assessed through this questionnaire, following the Likert scale: 1 (good), 2 (sufficient), 3 (insufficient) and 4 (poor). This scale showed a good reliability and validity: the reliability was high for the 11 skills and the three factors described below, and the construct validity was set by the Confirmatory Factor Analysis (CFA). The 11 transversal skills were grouped into three factors: communication skills (CS), research skills (RS) and organisational/learning skills (OS). The CS factor included writing communication (WC), oral communication (OC) and visual communication (VC). The RS factor included the ability to perform literature searching (LS), cope with information technology (IT) and data analysis (DA) and English proficiency (EP). The OS factor included team work (TW) capacity, time management (TM), the ability to solve problems (SP) and to self-improve learning (IL). CFA was used to test the fit of the model with 3-factor structure considering the 11 skills as items. CFA fit was assessed using different indexes: (i) the Tucker-Lewis Index (TLI) [bib_ref] A reliability coefficient for maximum likelihood factor analysis, Tucker [/bib_ref] , (ii) the Comparative fit index (CFI) [bib_ref] Comparative fit indexes in structural models, Bentler [/bib_ref] , (iii) the Route mean square error of approximation (RMSEA) [bib_ref] Structural Model Evaluation and Modification -an Interval Estimation Approach, Steiger [/bib_ref] , and (iv) the Standardized root mean square residual (SRMR) [bib_ref] Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus..., Hu [/bib_ref]. A good model fit is indicated by a CFI and TLI values of 0.90 or higher [bib_ref] Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus..., Hu [/bib_ref] and values of RMSEA and SRMR close to 0 [bib_ref] Alternative ways of assessing model fit, Browne [/bib_ref]. The internal reliability of the scale was tested using the Cronbach's alpha coefficient.
The perceived ability to perform the transversal skills was recoded, in comparison with the scale shown in Additional file 1, in two categories sufficient (0) vs. good (1), given that no student perceived his/ her own ability as insufficient or poor. The percentage of medical students who perceived their performance as good was estimated. Odds ratio (OR) was used to estimate the magnitude of the association between the perceived ability to perform each transversal skill and several factors (gender, academic year, participation in research projects and high school). Unconditional logistic regression adjusted for all variables was used to estimate the OR and the respective 95 % confidence interval. All statistics analyses were performed with the R 2.12.1.
The ethical principles of this research followed the guidelines approved by the Ethics Committee of FMUP.
# Results
Characteristics of the sample are summarized in [fig_ref] Table 1: Characteristics of the sample [/fig_ref]. The Cronbach's alpha of the analysis of the 11 transversal skills ranges from 0.651 to 0.818 indicating a moderate to high internal consistency. According to the theoretical model, in the CFA it was assumed that items WC, OC, and VC belong to factor CS (Communication Skills), items LS, IT, DA and EP belong to factor RS (Research Skills), and items TW, TM, SP and IL belong to factor OS (Organizational/learning Skills) and they were correlated with each other. shows the factor loadings supporting these relations. The model global fitness was tested and confirmed. The Cronbach alpha was 0.69, 0.81, 0.80 for factor CS, RS and OS, respectively.
No student perceived his/her own ability to perform the transversal skills as insufficient or poor. More than 50 % of students considered their own ability to manage information technology, of time management and to perform literature searching as sufficient. In relation to the other transversal skills, more than 50 % of students considered their own ability to perform them as good [fig_ref] Figure 2: Percentage of medical students who perceived their performance as good, according to... [/fig_ref]. 72.7 % of medical students perceived their own capacity of team work as good [fig_ref] Figure 2: Percentage of medical students who perceived their performance as good, according to... [/fig_ref].
The progression over the medical course, as well as the participation in research activities, was associated with an increasing odds of a good perceived performance of communication skills [fig_ref] Table 2: Factors associated with communication skills performance Communication skills, OR [/fig_ref]. Regarding writing skills, an approximate 2-fold increase in odds of a good performance was observed in sixth year students and in those who participated in research activities [fig_ref] Table 2: Factors associated with communication skills performance Communication skills, OR [/fig_ref].
The progression over the medical course and the participation in research activities were also associated with an increasing odd of a good perceived performance of research skills and organisational/learning skills [fig_ref] Table 3: Factors associated with research skills performance [/fig_ref]. More than 3-fold increase in odds of a good perceived English proficiency was observed in the students of the sixth year of medical course [fig_ref] Table 3: Factors associated with research skills performance [/fig_ref].
Male students were associated with an increasing odds of a good perceived manage of information technology and a good performance of data analysis, comparatively to female students [fig_ref] Table 3: Factors associated with research skills performance [/fig_ref].
Attending a private school at high school was associated with an increasing odds of a good performance in the majority of skills, although without statistical significance (Tables 2, 3 and 4).
# Discussion
This group of medical students felt confident regarding the majority of the studied transversal skills. The progression along the medical course, as well as the participation in research activities might have contributed to this perception. These findings are in line with what has already been investigated regarding the improvement of transversal skills due to the engagement in research activities [bib_ref] Doing an intercalated BSc can make you a better doctor, Greenhalgh [/bib_ref] [bib_ref] Research and medical students, Remes [/bib_ref] [bib_ref] Skill development in graduate education, Parker [/bib_ref].
On the other hand, the perceived ability to manage information technology, time management and to perform literature searching was classified as sufficient by many of the students. Therefore, improvements in medical education should be further explored in order to develop these competences.
Regarding the information technology, medical educators are rapidly integrating technology into the medical curricula as a necessary intervention [bib_ref] Preparing for the changing role of instructional technologies in medical education, Robin [/bib_ref] [bib_ref] Introducing technology into medical education: Two pilot studies, George [/bib_ref]. According to Ranasinghe et al., there is a need to improve computer literacy among freshman medical students, by increasing computer training in schools, or by introducing computer training in the initial stages of the medical course [bib_ref] Computer literacy among first year medical students in a developing country: a..., Ranasinghe [/bib_ref]. The study by Lim et al. showed that 81.5 % of final year medical students consider their computer skills adequate, despite additional steps need to be taken to increase the use of Internet as a method of instruction [bib_ref] Perceived skill and utilisation of information technology in medical education among final..., Lim [/bib_ref]. Several studies have explored the incorporation of personal digital assistants (PDAs) [bib_ref] The impact of a personal digital assistant (PDA) case log in a..., Ho [/bib_ref] [bib_ref] Personal Digital Assistants: A Review of Current and Potential Utilization Among Medical..., Tempelhof [/bib_ref] , e-learning [bib_ref] Evaluation of different delivery modes of an interactive e-learning programme for teaching..., Hawthorne [/bib_ref] and social networking for interprofessional instruction [bib_ref] Merging social networking environments and formal learning environments to support and facilitate..., King [/bib_ref] in medical education. In order to avoid the workload that medical students usually face, many schools are implementing policies to restrict student work hours [bib_ref] Regulation of medical student work hours: a national survey of deans, Friedman [/bib_ref]. In line with this workload, our students also had difficulties in time management.
Literature searching is an essential competence for both research and clinical activities, and our students could benefit and feel more confident with specific [bib_ref] Is literature search training for medical students and residents effective? A literature..., Just [/bib_ref]. On the other hand, a significant number of medical students perceived their own team work capacity as good. This ability should continue to be explored during the medical course, for instance through students' involvement in small research groups. Indeed, teamwork has become a major focus in healthcare [bib_ref] Teaching teamwork in medical education, Lerner [/bib_ref] , partially, in consequence of the report provided by the Institute of Medicine, which details the high rate of preventable medical errors, mainly originated from dysfunctional or nonexistent teamwork.
In relation to the putative determinants of perceptions about transversal skills, students who participated in research projects revealed an increased odd of good perceived performance in all transversal skills, including those mentioned above. In this analysis, writing ability stood out between communication skills. In health professions education, writing has been used as a tool to promote a variety of learning goals, but students in medical professions often lack the required writing skills during their education and career [bib_ref] Writing-skills development in the health professions, Rawson [/bib_ref] [bib_ref] Writing, self-reflection, and medical school performance: the Human Context of Health Care, Stephens [/bib_ref]. Participation in research activities is therefore essential since students who write concisely and logically, use terminology accurately and justify their assertions, show better academic performance [bib_ref] Writing-skills development in the health professions, Rawson [/bib_ref]. Furthermore, scientific writing for publication is competitive and demanding for researchers [bib_ref] An academic writing needs assessment of Englishas-a-second-language clinical investigators, Wang [/bib_ref] , and training in writing skills along the medical course is associated with a significant improvement in text quality, structure and argumentation [bib_ref] A workshop to improve written communication skills of medical students, Bitran [/bib_ref]. Students who finished the medical course demonstrated a significant increase of odds to have a good perceived English proficiency. Probably, this is explained by the need to use international scientific literature during the study years in medical school. However, students should feel confident in their English proficiency from the beginning of the course in order to have an efficient acquisition of knowledge. For the novice English-as-a-second-language researcher, the pressure to publish compounds the difficulties of mastering the English language [bib_ref] An academic writing needs assessment of Englishas-a-second-language clinical investigators, Wang [/bib_ref]. In addition, in most countries around the world, local medical students outperform international students, in an academic sense, probably because of the acquision of language proficiency skills [bib_ref] The influence of language family on academic performance in Year 1 and..., Mann [/bib_ref]. According to Hays et al. Australian medical students whose primary language is other than English, yet with poor English oral communication skills, should be encouraged to speak English away from the medical school and should be offered additional tuition so that their skills in other languages are not lost to the healthcare system [bib_ref] Language background and communication skills of medical students, Hays [/bib_ref]. On the other hand, Mirza et al. reported that communication skills training purely in English can leave Arab medical students ill equipped to communicate with patients in their own communities and tongue [bib_ref] Communication skills training in English alone can leave Arab medical students unconfident..., Mirza [/bib_ref].
In relation to problem solving skills, fourth year students demonstrated a significant decrease in the odds of good perceived performance, contrary to freshman and final year students. Problem solving skills has been defined as a hypothetical-deductive activity engaged by physicians, in which the early generation of hypotheses influences the subsequent gathering of information [bib_ref] Methods for teaching problem-solving in medical schools, Shumway [/bib_ref]. In the fourth year of the medical course, students begin the clinical practice, hence their perception of a poorer performance can be, partially, explained by the unfulfilled expectations during the first cycle of studies, as well as the fear of becoming incompetent practitioners. Therefore, the need to include instructional activities to promote the development of problem-solving abilities should be asserted in the medical curriculum.
Concerning gender, male students were associated with an increasing odds of a good perceived management of information technology, comparatively to female students. This finding has been reported by some authors who showed that females had more negative attitudes towards computers and Internet than males [bib_ref] Gender and information and communication technologies (ICT) anxiety: male self-assurance and female..., Broos [/bib_ref] [bib_ref] Gender and cultural differences in Internet use: A study of China and..., Li [/bib_ref]. Comparison at high schooling level between private and public schools, showed no significant differences. Nonetheless, further studies are needed to explore whether private schools offer a better support for students' research competences than public schools.
This study assessed the putative performance of a core of transversal skills by Portuguese medical students based on their perceptions. An increased body of evidence shows that student perceptions about several aspects of their competences can be valuable for extracting evaluative conclusions regarding the quality of medical education [bib_ref] University students' perceptions of the learning environment and academic outcomes; implications for..., Lizzio [/bib_ref] [bib_ref] Clinical capabilities of graduates of an outcomes-based integrated medical program, Scicluna [/bib_ref] , despite this approach could be seen as a study limitation.
No student perceived his own ability to execute the transversal skills as insufficient or poor, a finding that may reflect a limitation of the used Likert scale, that require to be revised in future studies. This study has the inherent limitations of a cross-sectional study. Thus, a follow-up study of these students' cohorts could also provide additional information of their transversal skills' perceptions, as well as important data for the design of a medical educational program.
# Conclusions
In conclusion, this group of medical students felt confident regarding the majority of the studied skills, particularly team work capacity. However, conceptual models of medical education modalities should be further explored to improve skills, such as both information technology and time management as well as literature searching.
The progression over the medical course and participation in research activities were associated with an increased odd of a good perceived performance of all transversal skills. Thus, the early exposure to research activities along undergraduate medical education is an added value for students and the implementation of an innovative integrated research program into the medical curriculum should be considered.
## Additional file
Additional file 1: Questionnaire
## Competing interests
The authors declare having no competing interests.
[fig] Figure 2: Percentage of medical students who perceived their performance as good, according to the corresponding transversal skills. Legend: IT, information technology; TM, time management; LS, literature searching; IL, improving learning; OC, oral communication; DA, data analysis; WC, writing communication; VC, visual communication; EP, English proficiency; SP, solving problems; TW, team work [/fig]
[table] Table 1: Characteristics of the sample [/table]
[table] Table 2: Factors associated with communication skills performance Communication skills, OR (95 % CI) [/table]
[table] Table 3: Factors associated with research skills performance [/table]
[table] Table 4: Factors associated with organisational/learning skills performance [/table]
|
Plasmacytoid Dendritic Cells Are Proportionally Expanded at Diagnosis of Type 1 Diabetes and Enhance Islet Autoantigen Presentation to T-Cells Through Immune Complex Capture
OBJECTIVE-Immune-mediated destruction of -cells resulting in type 1 diabetes involves activation of proinflammatory, islet autoreactive T-cells, a process under the control of dendritic cells of the innate immune system. We tested the hypothesis that type 1 diabetes development is associated with disturbance of blood dendritic cell subsets that could enhance islet-specific autoimmunity.RESEARCH DESIGN AND METHODS-We examined blood dendritic cells (plasmacytoid and myeloid) in 40 patients with recent-onset diabetes (median duration 28 days) and matched control subjects. We also examined the relative ability of different dendritic cell subsets to process and present soluble or immune complexed islet cell autoantigen (the islet tyrosine phosphatase IA-2) to responder CD4 T-cells.RESULTS-The balance of blood dendritic cells was profoundly disturbed at diabetes diagnosis, with a significantly elevated proportion of plasmacytoid and reduction of myeloid cells compared with control subjects. Dendritic cell subset distribution was normal in long-standing disease and in patients with type 2 diabetes. Both dendritic cell subsets processed and presented soluble IA-2 to CD4 T-cells after short-term culture, but only plasmacytoid dendritic cells enhanced (by as much as 100%) autoantigen presentation in the presence of IA-2 ϩ autoantibody patient serum.CONCLUSIONS-The plasmacytoid subset of dendritic cells is overrepresented in the blood close to diabetes onset and shows a distinctive ability to capture islet autoantigenic immune complexes and enhance autoantigen-driven CD4 T-cell activation. This suggests a synergistic proinflammatory role for plasmacytoid dendritic cells and islet cell autoantibodies in type 1 diabetes.
T ype 1 diabetes is an autoimmune disease resulting from T-cell-mediated destruction of insulinproducing -cells [bib_ref] Type 1 diabetes: new perspectives on disease pathogenesis and treatment, Atkinson [/bib_ref] [bib_ref] The role of T-cells in the pathogenesis of type 1 diabetes: from..., Roep [/bib_ref]. Although the precise aetiopathogenesis of the disease is unknown, it is apparent that -cell damage involves the generation of activated, proinflammatory, islet-autoreactive, effector CD4 and CD8 T-cells [bib_ref] Translational mini-review series on type 1 diabetes: immune-based therapeutic approaches for type..., Staeva-Vieira [/bib_ref]. The priming, differentiation, and expansion of effector T-cells is largely under the control of a heterogeneous group of immune cells that go under the collective term of dendritic cells, because of their distinctive morphology [bib_ref] Dendritic cells: translating innate to adaptive immunity, Steinman [/bib_ref]. Dendritic cells have numerous specialized forms present in peripheral tissues, lymph nodes, and the blood, and collectively, these cells are responsible for the sensing and ingestion of pathogens and activation of T-cells of relevant specificity. Because activated dendritic cells are a requirement for priming of naïve T-cells, it is likely that a similar process pertains during the development of islet autoreactivity, although the activating stimuli and islet autoantigens involved remain obscure. It is also likely that once this process is initiated, dendritic cell presentation of islet autoantigens remains a feature of the disease, because spreading of the autoimmune response to additional autoantigens and epitopes develops [bib_ref] Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen, Lehmann [/bib_ref].
Given the pivotal role of dendritic cells in the activation of naïve T-cells, there is a strong justification for investigating their activity in type 1 diabetes. Until relatively recently, however, opportunities to study dendritic cells in a human disease setting were limited. In recent years, there has been an increasing recognition that two of the major dendritic cell subsets, the myeloid (myeloid dendritic cell) and plasmacytoid (plasmacytoid dendritic cell) forms are present at low levels in the circulation and can be identified by their expression of distinct lineage and functional markers [bib_ref] BDCA-2, a novel plasmacytoid dendritic cell-specific type II C-type lectin, mediates antigen..., Dzionek [/bib_ref]. Plasmacytoid dendritic cells, also known as the type I interferon (IFN)-producing cells, are of particular interest, being specialized in the sensing of virus infection through selective expression of Toll-like receptors (TLRs) specific for viral single-stranded RNA (TLR7) and double-stranded DNA (TLR9) [bib_ref] IPC: professional type 1 interferon-producing cells and plasmacytoid dendritic cell precursors, Liu [/bib_ref]. Ligation of such viral receptors results in the rapid secretion of type I IFNs, such as IFN-␣, at levels 100-1,000 times more than any other cell type. Serum IFN-␣ levels are elevated in children at diagnosis of type 1 diabetes (10); IFN mRNA subtypes are found in post mortem pancreas samples from type 1 diabetic patients [bib_ref] Interferon expression in the pancreases of patients with type I diabetes, Huang [/bib_ref] ; and critically, IFN-␣ treat-ment for diseases such as chronic viral hepatitis and cancer has precipitated the clinical manifestation of autoimmune disease, including type 1 diabetes, in a number of cases [bib_ref] Type 1 diabetes mellitus in patients with chronic hepatitis C before and..., Fabris [/bib_ref] [bib_ref] Onset of insulin-dependent diabetes mellitus after interferon-alfa therapy for hairy cell leukaemia, Guerci [/bib_ref]. Moreover, there is emerging evidence of a close relationship between plasmacytoid dendritic cells, excessive amounts of type I IFNs, and other autoimmune conditions [bib_ref] Interferon-alpha in systemic lupus erythematosus, Crow [/bib_ref].
We hypothesized that the existence of a relationship between type I IFNs and type 1 diabetes close to the onset of the disease might be reflected in a disturbance in blood dendritic cell subsets. Our study demonstrates a profound disturbance in the normal balance of plasmacytoid dendritic cells and myeloid dendritic cells in peripheral blood in the immediate period after diagnosis. Moreover, we show that plasmacytoid dendritic cells capture islet cell autoantigens via immune complexes and in so doing enhance CD4 T-cell activation, suggesting synergy between elements of the innate and adaptive immune systems, and islet cell autoantibodies in particular, in type 1 diabetes-related autoimmunity.
## Research design and methods
For dendritic cell enumeration studies, fresh heparinized blood was obtained from 40 patients with new-onset type 1 diabetes (mean age 27 Ϯ 7.2 years; disease duration 30 Ϯ 15 days) and 40 nondiabetic control subjects of similar age, sex, and HLA type with no family history of autoimmune disease (mean age 29 Ϯ 5.2 years). Eighteen patients with long-standing type 1 diabetes (median disease duration 19.5 years, range 3-58 years; mean age 44 Ϯ 13.8 years) were also studied along with an additional, appropriately matched nondiabetic control group (n ϭ 18; mean age 34 Ϯ 10 years). In addition, 16 patients with type 2 diabetes (median disease duration 7.5 years, range 2-16 years, mean age 59.8 Ϯ 11.7 years) were also studied along with an appropriately matched nondiabetic control group (n ϭ 16; mean age 51 Ϯ 8.7 years). Blood was also obtained from a further group of patients with new-onset type 1 diabetes and from control subjects of similar age, sex, and HLA-type for detailed analysis of dendritic cell phenotype (n ϭ 13 in each group) and dendritic cell cytokine production after in vitro stimulation (n ϭ 9 in each group). All type 1 diabetic patients gave a history of acute onset of symptoms typical of diabetes and required insulin from diagnosis. Typically, blood from a patient was analyzed alongside that of an appropriate control taken at the same time of day to avoid any potential confounding effects of diurnal rhythm and season. Patients or control subjects reporting recent (up to 2 weeks) symptoms of "virus-like" illnesses were excluded from participation in the study. Metabolic control was assessed by measurement of A1C at the time of blood sampling for dendritic cell studies. The presence of autoantibodies to IA-2 and glutamic acid decarboxylase-65 (GAD65) was detected by radioimmunoassay (RSR, Cardiff, U.K.). Dendritic cell enumeration. Lysed whole blood direct immunofluoresence staining for the quantitation of dendritic cell subsets was performed using the Blood DC Enumeration Kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer's instructions. Briefly, 300 l whole blood was incubated with either the anti-BDCA cocktail, containing monoclonal antibodies (mAbs) specific for the dendritic cell markers BDCA-1 (CD1c for detecting myeloid dendritic cell 1), BDCA-2 (CD303 for detecting plasmacytoid dendritic cells), and BDCA-3 (CD141 for detecting myeloid dendritic cell 2); anti-CD14 (monocytes); and anti-CD19 (B-cells) or the control cocktail (containing appropriate isotype control mAbs). After 10-min photoincubation with these and a dead cell discriminator on ice, erythrocytes were lysed, and remaining cells were washed in PBS containing 1% BSA (Sigma-Aldrich, Poole, U.K.), 0.1% sodium azide, 2% FCS (Invitrogen, Paisley, U.K.), and 1% human AB serum (PAA, Linz, Austria) (flow buffer). Cells were fixed and analyzed by four-color flow cytometry using a FACSCalibur cytometer (BD Biosciences, San Jose, CA). CD14 ϩ , CD19 ϩ , and dead cells were excluded, and dendritic cell subsets were enumerated. A minimum of 3.5 ϫ 10 5 events in the mononuclear (forward and side scatter) gate were collected, of which a mean of 0.8% are dendritic cells (ϳ2,800 events). Gates denoting positivity for markers of myeloid dendritic cell 1, myeloid dendritic cell 2, and plasmacytoid dendritic cell were set using the appropriate isotype control mAbs and the number of dendritic cells in a sample expressed as a percentage of the total number of gated mononuclear cells or the total number of dendritic cells. Dendritic cell isolation, phenotypic analysis, and stimulation. Peripheral blood mononuclear cells (PBMCs) were obtained from whole blood by density gradient centrifugation (Lymphoprep; Axis Shield, Dundee, U.K.). CD14 ϩ and CD19 ϩ cells were depleted from PBMCs using magnetic cell sorting technology (AutoMACS; Miltenyi Biotech). Myeloid dendritic cells were then positively isolated from the remaining PBMCs using BDCA-1 ϩconjugated magnetic microbeads and plasmacytoid dendritic cells then isolated from the remaining cells using BDCA-4 ϩ -conjugated magnetic microbeads. Averaging from 20 separations, mean Ϯ SD purity was 78.8 Ϯ 10.9% for plasmacytoid dendritic cells and 75.0 Ϯ 19.5% for myeloid dendritic cells.
Markers of activation and function were analyzed by four-color flow cytometry using 2 ϫ 10 4 dendritic cells in 100 l fluorescence-activated cell sorting buffer along with the appropriate fluorochrome-labeled mAbs. Samples were incubated on ice in the dark for 30 min and then washed twice in ice-cold flow buffer before immediate analysis. The following mAbs were used: fluorescein isothiocyanate-labeled anti-CD19 (clone LT19), anti-HLA-ABC (clone W6/32), and anti-CD83 (clone HB15e) mAbs; phycoerythrin (PE)-labeled anti-CD62L (clone FMC46) and anti-CD80 (clone MEM-233) mAbs (all from Serotec, Oxford, U.K.); PE-labeled anti-CD4 (clone RPA-T4), peridinin chlorophyll protein-labeled anti-CD14 (clone M P9), and anti-HLA-DR (clone TU36) mAbs; and allophycocyanin-labeled anti-CD3 (clone UCHT1), anti-CD86 (clone 2331FUN-1) (all BD Pharmingen, San Diego, CA), and anti-CD123 (clone AC145) mAbs (Miltenyi Biotec). Antibody concentrations used were based on data supplied by the manufacturers and in-house optimization studies. For these studies, the selected dendritic cell subset was always gated using a specific marker (BDCA1 for myeloid dendritic cell and CD123 for plasmacytoid dendritic cell). Gated cells were then assessed for the mean fluorescence intensity of the specific activation/maturation marker.
In studies on the cytokine potential of isolated blood dendritic cell subsets, myeloid dendritic cells and plasmacytoid dendritic cells were resuspended to 2 ϫ 10 5 per ml in RPMI 1640 supplemented with penicillin, streptomycin, and fungizone (Invitrogen) and 10% human AB serum. Myeloid dendritic cells were stimulated with lipopolysaccharide (Sigma-Aldrich) at concentrations of 0 and 10 ng/ml in a total volume of 200 l in duplicate samples. Plasmacytoid dendritic cells underwent similar treatment in cultures supplemented with 10 ng/ml recombinant human interleukin (IL)-3 (Strathmann Biotec, Hamburg, Germany) and were stimulated with unmethylated oligodeoxynucleotides (ODN 2216, a potent TLR-9 ligand and IFN-␣ secretagog; Autogen Bioclear, Calne, U.K.) at concentrations of 0, 1, 3, 10, and 30 g/ml in a total volume of 200 l in duplicate samples [bib_ref] Virus-stimulated plasmacytoid dendritic cells induce CD4ϩ cytotoxic regulatory T cells, Kawamura [/bib_ref] [bib_ref] Identification of CpG oligonucleotide sequences with high induction of IFN-alpha/beta in plasmacytoid..., Krug [/bib_ref]. After 16 h of stimulation, supernatants were collected and stored at Ϫ80°C for later analysis. IFN-␣ production in plasmacytoid dendritic cell culture supernatants was quantified by specific sandwich ELISA using the IFN-␣ module set (detection range 8 -5,000 pg/ml; Bender MedSystems). There are 14 different IFN-␣ isoforms, of which the IFN-␣ module set detects the majority, with the exception of B and F. Supernatants from myeloid dendritic cell and plasmacytoid dendritic cell cultures were analyzed using the Beadlyte Human Cytokine Detection kit 3 (Upstate, Lake Placid, NY) according to the manufacturer's instructions. Cytokines IL-1, IL-2, IL-4, IL-6, IL-8, IL-10, Il-12p70, tumor necrosis factor-␣ (TNF-␣), and IFN-␥ were measured (detection range 6.9 -5,000 pg/ml). Data were analyzed on the Luminex 100 instrument system using STarstation software. Presentation of autoantigens by dendritic cells. cDNA encoding human IA2-intracellular portion (IA-2ic) was modified to include a coding sequence for the HLA-DR4 restricted influenza hemagglutinin CD4 T-cell epitope (HA 307-319 ), inserted at the NH 2 -terminus of IA-2ic, and recombinant IA-2ic-HA 307-319 protein was prepared as described previously [bib_ref] Autoreactive T cell responses show proinflammatory polarization in diabetes but a regulatory..., Arif [/bib_ref] [bib_ref] Naturally processed and presented epitopes of the islet cell autoantigen IA-2 eluted..., Peakman [/bib_ref]. An HLA-DR4 restricted CD4 T-cell clone specific for HA 307-319 (JNZ-1) was generated by peptide stimulation of PBMC cultures, followed by flow cytometric sorting of cells stained positive with an HLA-DR4 tetramer loaded with HA 307-319 (provided by Dr. G. Nepom, Benaroya Research Institute, Seattle, WA). Plasmacytoid dendritic cells, myeloid dendritic cells, and CD14 ϩ monocytes were isolated from HLA-DR4 ϩ subjects and resuspended to 5 ϫ 10 4 /ml, and 100 l was dispensed into 96-well flat-bottomed plates in serum-free X-VIVO 15 (Cambrex, Charles City, IA). Cultures were supplemented with serum from type 1 diabetic patients positive for either IA-2 or GAD65 autoantibodies, or pooled AB (1-30%) and IA-2ic-HA 307-319 (10 g/ml). In some experiments, the IgG fraction in IA-2 ϩ autoantibody serum was depleted by incubation with protein A matrix (binding capacity 18 -43 mg/ml; Repligen, Waltham, MA) at a 1:1 ratio for 90 min at room temperature according to the manufacturer's instructions. After centrifugation (15 s at 10,000g), unbound serum was carefully removed, and IgG depletion was confirmed by laser nephelometric analysis of IgG levels on samples before and after adsorption. After 24 h of dendritic cell culture with serum under these conditions, JNZ-1 clone cells (10 5 /well) were added, and 48 h later, IFN-␥ levels in supernatants were quantified by ELISA (detection range 2-500 pg/ml; Immunotools, Friesoythe, Germany).
# Statistical analysis.
Dendritic cell subsets, expressed as a percentage of total dendritic cells, were normally distributed in all study groups according to Kolmogorov-Smirnov analysis. Mean levels of dendritic cell subsets were therefore compared between study groups using the Student's t test. Relationships between dendritic cell subset numbers and clinical parameters were compared by calculation of Pearson's correlation coefficient. Differences in distribution were compared by 2 analysis. P values of Ͻ0.05 were considered significant. Differences in IFN-␥ production were compared using Student's t tests.
# Results
## Detection of dendritic cell subsets in whole blood.
Dendritic cell subsets were identified in whole blood using specific mAbs among mononuclear cells identified by their physical side and forward scatter properties after removal of CD14 ϩ , CD19 ϩ , and dead cells by exclusion gating . Dendritic cell numbers were expressed as a percentage of mononuclear cells and dendritic cell subsets as a proportion of total dendritic cells. Representative examples of BDCA-1 ϩ (myeloid dendritic cells), BDCA-2 ϩ (plasmacytoid dendritic cells), and BDCA-3 ϩ (myeloid dendritic cell 2) cells are shown in -E. The majority of events in the CD14 Ϫ /CD19 Ϫ gate in are T-cells, with some natural killer and natural killer T-cells. The relative proportion of this population was not different between type 1 diabetic patients and control subjects (data not shown). Dendritic cell subsets in type 1 diabetic patients. The total number of dendritic cells (i.e., cells expressing BDCA-1, BDCA-2, or BDCA-3) when expressed as a percentage of the total mononuclear cell population was similar in new-onset type 1 diabetic patients and nondiabetic control subjects .
However, within the total dendritic cell population, dendritic cell subsets had a different proportional distribution in the two study groups. The mean percentage of dendritic cells belonging to the plasmacytoid dendritic cell subset was significantly higher in new-onset type 1 diabetic patients when compared with nondiabetic control subjects (P ϭ 0.002; . In contrast, the mean percentage of myeloid dendritic cells was significantly lower in new-onset type 1 diabetic patients compared with the nondiabetic control group (P ϭ 0.003; . These findings were similar regardless of whether dendritic cell subsets were expressed as a percentage of total dendritic cells or of mononuclear cells. There was no difference between the study groups in the proportion of dendritic cells belonging to the BDCA3 ϩ myeloid dendritic cell 2 populations . Among the new-onset group, there was no correlation between percentages of dendritic cell subsets and age or duration of diabetes or A1C levels as a marker of metabolic control. The percentage levels of dendritic cell subsets were not different between patients with and without islet cell autoantibodies and did not differ between patient subgroups divided according to HLA class II genotype.
To assess whether the abnormalities in dendritic cell subsets observed in new-onset type 1 diabetic patients could be genetically determined, we also studied patients in whom type 1 diabetes had been diagnosed for a minimum of 3 years (median duration of disease 19.5 years, maximum 58 years). In these patients, mean levels of total dendritic cells as a percentage of mononuclear cells and the distribution of dendritic cell subsets were similar to those in age-matched non- in patients with type 2 diabetes compared with control subjects .
Although the populations of patients with type 1 diabetes (new-onset and long-standing) were not closely agematched, we also compared the distribution of dendritic cell subsets between them; this comparison seemed justified in the absence of any relationship between age and dendritic cell subset number in the patient and control subject groups. Mean levels of plasmacytoid dendritic cells were higher, and mean levels of myeloid dendritic cells were lower in the new-onset patients (P ϭ 0.0013 and 0.0014, respectively). These data suggested that dendritic cell subset balance is disturbed near to disease diagnosis but gradually normalizes. Further evidence for this was obtained by repeat analysis of blood samples on five patients tested both at diagnosis and 2 years later. Three of the five show evidence of declining plasmacytoid dendritic cell percentages, and four had increased/unchanged percentages of myeloid dendritic cells at year 2 , although with small numbers of subjects, these can only be considered indicative trends (NS). This contrasts with data from a control individual analyzed seven times over the same 2-year period, in whom plasmacytoid dendritic cell and myeloid dendritic cell subsets were remarkably stable . Taken together, these data indicate that type 1 diabetes is characterized by an abnormal distribution of dendritic cells, resulting in high levels of plasmacytoid dendritic cells and low levels of myeloid dendritic cells but that these changes are restricted to the peri-diagnosis period.
## Expression of differentiation and activation markers on plasmacytoid dendritic cells and myeloid dendritic cells in type 1 diabetic patients.
Because the plasmacytoid dendritic cell population appeared to be expanded at diagnosis of type 1 diabetes, we next wished to determine whether this expansion was associated with activation of this subset. Blood plasmacytoid dendritic cells expressed high basal levels of HLA molecules (both class I and class II), IL-3 receptor (CD123), and L-selectin (CD62L) but low levels of the dendritic cell activation marker CD83 and the costimulatory molecules CD80 and CD86, which are typically upregulated on activation. Myeloid dendritic cells were also found to express high levels of HLA class I and class II molecules. Myeloid dendritic cells also expressed detectable levels of CD83, CD80, and CD86 but without any sign of upregulation of these activation molecules in patients (supplementary , available in an online appendix at http://dx.doi.org/10.2337/ db08-0964). Formal comparison of the levels of expression of these markers in new-onset type 1 diabetic patients and nondiabetic control subjects (n ϭ 13 in each group) failed to show any significant differences. Cytokine and chemokine production by plasmacytoid dendritic cells and myeloid dendritic cells. Plasmacytoid dendritic cells are capable of producing very large quantities of type I IFNs after stimulation; and to assess their IFN-␣producing capacity in type 1 diabetes, cells isolated from the study groups were cultured with a TLR9 ligand, and IFN-␣ secretion was measured. Plasmacytoid dendritic cells from both new-onset type 1 diabetic patients and nondiabetic control subjects (n ϭ 9 in each group) did not produce IFN-␣ without stimulation. After stimulation with CpG oligonucleotide, plasmacytoid dendritic cells from patients and control subjects produced similar, large amounts of IFN-␣ in a dose-dependent manner . Thus, per-plasmacytoid dendritic cell IFN-␣ secretion is not abnormal in patients, and there is no evidence of spontaneous ex vivo IFN-␣ secretion by these cells, making it unlikely that they are the source of the elevated serum levels of this cytokine that has been reported in type 1 diabetic patients [bib_ref] Increased level of interferon-alpha in blood of patients with insulindependent diabetes mellitus:..., Chehadeh [/bib_ref]. Using the same samples, we also examined basal and stimulated secretion of IL-1, IL-6, IL-8, IL-10, IL-12p70, TNF-␣, and IFN-␥ by plasmacytoid dendritic cells and myeloid dendritic cells. No differences between patients and control subjects were observed.
## Islet cell autoantibody-positive serum enhances autoantigen presentation to plasmacytoid dendritic cells.
It has recently been reported that plasmacytoid dendritic cell presentation of antigen is enhanced in the presence of immune complexes containing the relevant antigen [bib_ref] Plasmacytoid dendritic cells of melanoma patients present exogenous proteins to CD4ϩ T..., Benitez-Ribas [/bib_ref]. We designed experiments in which plasmacytoid dendritic cells, myeloid dendritic cells, and monocytes presented a hybrid construct of IA-2ic containing the influenza hemagglutinin epitope HA 307-319 . Activation of an HA 307-319 -specific CD4 T-cell clone (JNZ-1) was used as the measure of efficiency of IA-2ic-HA 307-319 presentation. Serum positive or negative for IA-2 autoantibodies was then added to examine whether presentation was enhanced in the presence of relevant autoantibody specificities. Presentation of IA-2ic-HA 307-319 by plasmacytoid dendritic cells was enhanced in the presence of serum containing IA-2, but not GAD65, autoantibodies [fig_ref] FIG. 6: Effect of islet cell autoantibody-positive serum on presentation by plasmacytoid dendritic cells [/fig_ref]. IFN-␥ production by the T-cell clone was more than doubled in the presence of IA-2 autoantibodies (P Ͻ 0.05 compared with control serum for 5 and 10% serum supplementation). In contrast, myeloid dendritic cell and monocyte presentation of IA-2ic-HA 307-319 was largely unaffected by the presence of islet cell autoantibodies [fig_ref] FIG. 6: Effect of islet cell autoantibody-positive serum on presentation by plasmacytoid dendritic cells [/fig_ref]. In the case of myeloid dendritic cells, it is probable that their uptake and presentation of antigen through pinocytosis is highly efficient in the absence of immune complexes and therefore not enhanced to any great degree in their presence.
The effect of IA-2 ϩ serum samples in enhancing antigen presentation of IA-2ic-HA 307-319 was seen in samples from more than one patient with IA-2 autoantibodies and was highly significant [fig_ref] FIG. 6: Effect of islet cell autoantibody-positive serum on presentation by plasmacytoid dendritic cells [/fig_ref]. The enhancing effect was attributable to the IgG fraction in the serum samples, because it was significantly diminished after IgG depletion using a protein A-Sepharose column [fig_ref] FIG. 6: Effect of islet cell autoantibody-positive serum on presentation by plasmacytoid dendritic cells [/fig_ref]. Enhanced presentation of IA-2 by plasmacytoid dendritic cells was not associated with IFN-␣ production.
# Discussion
Our study shows an abnormal distribution of the main blood dendritic cell subsets in type 1 diabetes. IFNproducing plasmacytoid dendritic cells, rather than myeloid dendritic cells, predominate proportionally among dendritic cells in type 1 diabetic patients, and because these changes were only observed at diagnosis of disease, it seems unlikely that the alterations are genetically determined. The increased proportion of plasmacytoid dendritic cells and reduction in myeloid dendritic cells was not accompanied by changes in activation status of these cells, nor, in the case of plasmacytoid dendritic cells, by a change in their capacity for IFN-␣ secretion. There is no immediate explanation for this increased representation of plasmacytoid dendritic cells, but taken together with previous reports of abnormal levels of circulating and tissuedeposited type I IFNs in type 1 diabetes, these findings extend the concept of a relationship between type I IFNs, their major cellular source the plasmacytoid dendritic cell, and the onset of type 1 diabetes. Further weight is given to the potential importance of this relationship by our demonstration that plasmacytoid dendritic cells have enhanced ability to activate primed effector CD4 T-cells in the presence of islet cell autoantibodies, indicating a route through which autoantibodies could promote cell-mediated islet autoimmunity. Reagents for the reliable detection of dendritic cell subsets in peripheral blood have only recently become available, and few investigators have yet had the opportunity to deploy this technology in autoimmune diseases. To the best of our knowledge, this is the first study to demonstrate an increase in plasmacytoid dendritic cells in new-onset type 1 diabetic patients using this approach. These findings are similar to those in the preliminary report of Peng et al. [bib_ref] Abnormal peripheral blood dendritic cell populations in type 1 diabetes, Peng [/bib_ref] , who did not use the BDCA dendritic cell markers but rather detected plasmacytoid dendritic cells as lineage-negative/HLA-DR ϩ /CD11c Ϫ / CD123 ϩ and myeloid dendritic cells as lineage-negative/ HLA-DR ϩ /CD11c ϩ /CD123 Ϫ cells. This group also found a higher proportion of plasmacytoid dendritic cells in type 1 diabetes, although specific results are not reported, limiting the opportunity for direct comparison. A more recent report focuses on patients with type 1 diabetes of at least 5 years duration and finds no abnormality of blood dendritic cell balance, similar to our own analysis of longstanding patients [bib_ref] Reduced IFN-alpha secretion by blood dendritic cells in human diabetes, Summers [/bib_ref]. In contrast, Vuckovic et al. [bib_ref] Decreased blood dendritic cell counts in type 1 diabetic children, Vuckovic [/bib_ref] have reported decreased dendritic cell numbers in children with new-onset type 1 diabetes. However, these authors did not use specific markers of dendritic cell subsets, and they report levels of blood dendritic cells (Ͼ100 cells/l) that are vastly in excess of the known concentration of these cells (ϳ1% of PBMCs, equivalent to 10 -20 cells/l [bib_ref] BDCA-2, BDCA-3, and BDCA-4: three markers for distinct subsets of dendritic cells..., Dzionek [/bib_ref].
Using an in vitro culture system, we show that in the presence of islet cell autoantibodies of the relevant specificity, the ability of plasmacytoid dendritic cells to process and present islet autoantigens to CD4 T-cells is significantly enhanced. This immune complex capture is known to operate through the Fc␥ RII-receptor for IgG (CD32), probably through enhanced antigen uptake [bib_ref] Plasmacytoid dendritic cells of melanoma patients present exogenous proteins to CD4ϩ T..., Benitez-Ribas [/bib_ref]. Antibody-enhanced cross-presentation of an islet antigen has also been shown to break tolerance in a murine model of autoimmune diabetes [bib_ref] Antibody-enhanced cross-presentation of self antigen breaks T cell tolerance, Harbers [/bib_ref]. We consider it unlikely that such a pathway is involved in the initial priming of islet autoimmunity in man, but it could have an important role in the amplification and maintenance of T-cell responses and also in epitope spreading [bib_ref] Determinant spreading and the dynamics of the autoimmune T-cell repertoire, Lehmann [/bib_ref]. The finding that immune complex capture by plasmacytoid dendritic cells enhances autoreactive T-cell activation highlights a potential route through which autoantibodies could participate in cell-mediated autoimmune responses, in addition to that offered by B-cell presentation of autoantigens.
It is unclear why dendritic cell balance should be perturbed near to diagnosis of type 1 diabetes, but our studies exclude any possible effects of glycemic control. One explanation is that selective migration of myeloid dendritic cells into tissues or lymph nodes results in a relative expansion of blood plasmacytoid dendritic cells, especially if accompanied by proliferation of the dendritic cell-precursor pool. The possible effect of dendritic cell migration on blood levels has been noted in diseases in which both inflamed tissues and the circulation have been examined, such as psoriasis [bib_ref] Plasmacytoid predendritic cells initiate psoriasis through interferon-alpha production, Nestle [/bib_ref] , systemic lupus erythematosus [bib_ref] Expression of the markers BDCA-2 and BDCA-4 and production of interferon-alpha by..., Blomberg [/bib_ref] , and respiratory syncytial virus infection [bib_ref] Mobilization of plasmacytoid and myeloid dendritic cells to mucosal sites in children..., Gill [/bib_ref] , each of which are associated with a reduction in plasmacytoid dendritic cells in peripheral blood and infiltration of the tissues. An alternative hypothesis is that plasmacytoid dendritic cell expansion occurs in response to a virus infection that arises toward the end of the islet destructive process. Virus infection at this late stage of disease is known to be capable of accelerating disease onset in animal models [bib_ref] Coxsackievirus infections and NOD mice: relevant models of protection from, and induction..., Tracy [/bib_ref] , and it is tempting to speculate that this induces a shift in the ratio of plasmacytoid and myeloid dendritic cells in the periphery.
At this stage, it is only possible to speculate on the pathological relevance of our findings. Although the possibility of a proinflammatory role is attractive, studies in animal models show that an increase in plasmacytoid dendritic cell/myeloid dendritic cell ratio may prevent antigen-specific inflammatory responses, including diabetes , and a similar role in type 1 diabetes cannot be excluded. Future studies that examine whether dendritic cell subsets are disturbed in the pre-diabetic period and therefore reflect events involved in the chronic autoimmune process of islet destruction or whether plasmacytoid dendritic cell expansion is related solely to peri-diagnosis events may clarify this.
[fig] FIG. 1 FIG. 2 FIG. 3: Example of flow cytometry analysis and gating strategy for identification and quantification of blood dendritic cell subsets. The mononuclear cell population in whole blood is identified by a combination of forward and side scatter properties (A) followed by exclusion gating of live CD14 ؉ and CD19 ؉ mononuclear cells and dead cells (B), before the identification and enumeration of plasmacytoid dendritic cell (pDC), myeloid dendritic cell 1 (mDC1), and myeloid dendritic cell 2 (mDC2) subsets, as shown in C-E with the relevant blood dendritic cell (BDC) marker. F: Total dendritic cell populations as a percentage of mononuclear cells in 40 patients with new-onset type 1 diabetes and matched control subjects. (Please see http://dx.doi.org/10.2337/db08-0964 for a high-quality digital representation of this figure.) diabetic control subjects (Fig. 3A-D). We also examined the potential influence of hyperglycemia per se on dendritic cell subsets by analysis of patients with type 2 diabetes with similar levels of A1C (mean Ϯ SD 7.9 Ϯ 1.9%) compared with the new-onset patients (8.73 Ϯ 2.1%; NS). Chronic hyperglycemia alone could not account for the changes we observed, because proportions of dendritic cell subsets did not differ significantly Dendritic cell subsets in new-onset type 1 diabetic patients and control subjects showing plasmacytoid dendritic cells (A), type 1 myeloid dendritic cells (B), and type 2 myeloid dendritic cells (C). Results are expressed as percentage of total dendritic cell population. Mean values for each group are indicated by a horizontal line, and P values are given where significant. Dendritic cell subsets in long-standing type 1 diabetic patients (LS T1D) and age-matched control subjects showing total dendritic cell population as a percentage of mononuclear cells (A) and the plasmacytoid dendritic cell (B), type 1 myeloid dendritic cell (C), and type 2 myeloid dendritic cell (D) subsets as a percentage of total dendritic cells. Mean values for each group are indicated by a horizontal line and were similar between study groups. Similarly, E-G show plasmacytoid dendritic cell (E), type 1 myeloid dendritic cell (F), and type 2 myeloid dendritic cell (G) subsets as a percentage of total dendritic cells in patients with type 2 diabetes and an appropriately age-matched control group. Mean levels are similar in the two groups (NS). [/fig]
[fig] FIG. 4 FIG. 5: Plasmacytoid (A) and myeloid (B) dendritic cell subsets as a percentage of total dendritic cells in five patients with new-onset type 1 diabetes measured at diagnosis and 2 years later. There is a trend in three of the patients for plasmacytoid dendritic cell levels to decline and in three of the patients for myeloid dendritic cell levels to increase/remain unchanged. C: Dendritic cell subset levels (expressed as a percentage of dendritic cells) in a nondiabetic control individual studied on seven separate occasions over the same 2-year period. IFN-␣ secretion by isolated plasmacytoid dendritic cells in type 1 diabetic patients and control subjects in response to stimulation with increasing concentrations of the unmethylated CpG oligonucleotide 2,216. Each point represents the mean value of IFN-␣ secretion from nine different patients (OE) and nine control subjects (Ⅺ). Vertical bars represent SEs. [/fig]
[fig] FIG. 6: Effect of islet cell autoantibody-positive serum on presentation by plasmacytoid dendritic cells (A), myeloid dendritic cells (B), and CD14 ؉ monocytes (C) of the islet cell autoantigen IA-2ic-HA 307-319 to CD4 T-cells (JNZ-1 clone) specific for the influenza HA 307-319 epitope. CD4 T-cell clone responses are indicated by levels of IFN-␥ produced in the culture supernatants. A: In the presence of soluble 10 g/ml IA-2ic-HA 307-319 without serum supplementation (Ⅺ), IA-2ic is processed and presented with comparable efficiency by plasmacytoid dendritic cells and myeloid dendritic cells, with CD14 ؉ monocytes markedly less proficient. Addition of IA-2 ؉ autoantibody (AAb) serum (3,500 World Health Organization [WHO] units) at 1, 5, and 10% (f) considerably and significantly enhances the presentation of IA-2ic by plasmacytoid dendritic cells but has minimal effects on myeloid dendritic cells and CD14 ؉ monocytes (*P < 0.05 vs. presence of antigen alone; Ⅺ). GAD65 ؉ autoantibody serum (1,700 WHO units; u; 1-10%) has no effect on presentation. Likewise, pooled AB serum at these concentrations has no effect (data not shown). In the presence of 5% serum but no IA-2ic-HA 307-319 , there is no T-cell activation. Data represent one of four experiments using four different cell donors. D: Pooled data from a total of four donors (including the one used in A) for plasmacytoid dendritic cell enhancement of IA-2ic-HA 307-319 (f) presentation to JNZ-1 clone cells (*P < 0.05, **P < 0.01, and ***P < 0.001 vs. presence of antigen alone; Ⅺ). The additional sera contained 3,875.1, 3,263.1, and 3,007.8 WHO units of IA-2 autoantibodies. Bars represent means ؎ SE from quadruplicate wells of individual experiments. E: IA-2 ؉ serum-enhanced IA-2ic-HA 307-319 presentation is dependent on the serum IgG fraction. Serum has been adsorbed using Protein A matrix. Serum IgG level before adsorption, 6.54 g/l; after adsorption, 0.22 g/l. Serum was used at 5% concentration in these experiments. Serum IA-2 autoantibody level before adsorption, 2,152 WHO units; after adsorption, IA-2 autoantibodies were not detectable. *Significant reduction in IFN-␥ production by clone JNZ-1 in the presence of adsorbed compared with nonadsorbed serum (P < 0.05). [/fig]
|
Trends in Binge Drinking and Heavy Alcohol Consumption Among Pregnant Women in the US, 2011 to 2020
# Introduction
Alcohol-related mortality has been increasing among women in the US during the past 2 decades. [bib_ref] Trends in alcohol-induced deaths in the United States, Spillane [/bib_ref] Although no recent studies have found increased alcohol-related mortality specifically among pregnant women, 1 study 2 found increases in combined drug-and alcohol-related mortality. Recent data on binge drinking among pregnant women suggests a modest increase from 2011 to 2018 3 and no increase from 2018 to 2020. [bib_ref] Alcohol consumption and binge drinking during pregnancy among adults aged 18-49 years-United..., Gosdin [/bib_ref] However, little is known about how longer trends in problematic alcohol use may differ between pregnant and nonpregnant women. We sought to compare trends in prevalence of binge drinking and heavy alcohol consumption among pregnant and nonpregnant women from 2011 through 2020.
# Methods
This cross-sectional study obtained public use data from the Behavioral Risk Factor Surveillance System (BRFSS) from January 1, 2011, to December 31, 2020. The BRFSS is a nationally representative, cross-sectional sample of US adults that measures alcohol consumption based on 30-day recall. The BRFSS defines binge drinking as 4 or more drinks during a single occasion and heavy alcohol consumption as 8 or more drinks per week (eMethods and eTable in the Supplement).
Pregnancy status (pregnant vs not pregnant) was the primary exposure; sex was self-reported by all participants in the BRFSS data set, and all pregnant participants identified as women. Race and ethnicity (Hispanic or Latino, non-Hispanic Black, non-Hispanic White, non-Hispanic multiple races or ethnicities, and non-Hispanic other race or ethnicity) are reported in the BRFSS, and the distribution differs significantly by the primary exposure (pregnancy), which should be accounted for when comparing prevalence between pregnant and nonpregnant women. Analyses were adjusted for age group (18-24, 25-29, 30-34, 35-39, and 40-44 years) and race and ethnicity. The University of Texas at San Antonio Institutional Review Board waived the need for approval because the research did not involve human participants. This study followed the STROBE reporting guideline.
Age-and race and ethnicity-adjusted prevalence of binge drinking and heavy alcohol consumption were estimated using logistic regression models adjusted for complex survey design and weighting. Log linear regression models were used to estimate average annual percentage change (AAPC) in prevalence rates with 95% CIs. Two-sided P < .05 was considered statistically significant. Data were analyzed using R, version 4.0.2 (R Foundation for Statistical Computing).
# Results
Among the 49 098 pregnant women included in the study, 28.8% were 18 to 24 years of age and 3.7% were 40 to 44 years of age compared with 26.2% and 19.1%, respectively, for the 1 243 402 nonpregnant women (P < .001). Pregnant women were less likely to be non-Hispanic White (50.5%) compared with nonpregnant women (54.4%) (P < .001) . The prevalence of binge drinking increased from 2.5% (95% CI, 1.6%-3.4%) in 2011 to 6.1% (95% CI, 2.2%-10.0%) in 2020 for pregnant women, an AAPC of 8.9% (95% CI, 4.8%-12.9%; P = .003) . Binge drinking for nonpregnant Data are from the public use Behavioral Risk Factor Surveillance System. A, Estimated average annual percent change for binge drinking was 8.9% (95% CI, 4.8%-12.9%; P = .003) among pregnant women and 0.7% (95% CI, −0.5% to 1.8%; P = .28) among nonpregnant women. B, Estimated average annual percent change for heavy alcohol consumption was 11.6% (95% CI, 4.0%-19.3%; P = .02) among pregnant women and 2.3% (95% CI, 0.9%-3.7%; P = .01) among nonpregnant women.
## Jama network open | obstetrics and gynecology
# Discussion
In this cross-sectional study, we found that binge drinking and heavy alcohol consumption were higher among nonpregnant women than pregnant women, but the AAPC for both behaviors was
[fig] Figure: Prevalence of Binge Drinking and Heavy Alcohol Consumption Among Pregnant and Nonpregnant Women, [/fig]
[table] Table: Descriptive Statistics of Study Characteristics for Binge Drinking and Heavy Alcohol Consumption Samples, Behavioral Risk Factor Surveillance System, 2011 Through 2020 [/table]
|
In-hospital cardiac arrest resuscitation performed by the hospital emergency team: A 6-year retrospective register analysis at Danderyd University Hospital, Sweden
Discuss this article(0) Comments RESEARCH ARTICLE In-hospital cardiac arrest resuscitation performed by the hospital emergency team: A 6-year retrospective register analysis at Danderyd University Hospital, Sweden [version 1; referees: 2 approved] AbstractCardiac arrest requires rapid and effective handling. Huge efforts Background: have been implemented to improve resuscitation of sudden cardiac arrest patients. Guidelines around the various parts of effective management, the , are available. The aim of the present retrospective study was chain of survival to study sudden in-hospital cardiac arrest (IHCA) and the outcomes of emergence team resuscitation in a university hospital in Sweden.The Swedish Cardiopulmonary Resuscitation Registry was used to Methods: access all reported cases of IHCA at Danderyd Hospital from 2012 through 2017. Return of spontaneous circulation (ROSC), discharge alive, 30-day mortality and Cerebral Performance Scales score (CPC) were analysed. 574 patients with cardiac arrests were included in the study: 307 Results: patients (54%) had ROSC; 195 patients (34%) were alive to be discharged from hospital; and 191 patients (33%) were still alive at day-30 after cardiac arrest. Witnessed cardiac arrests, VT/VF as initial rhythm and experiencing cardiac arrest in high monitored wards were factors associated with success. However, 53% of patients' alive at day-30 had a none-shockable rhythm, 16% showed initially a pulseless electrical activity and 37% asystole. CPC score was available for 188 out of the 195 patients that were alive to be discharged: 96.5% of patients where data was available had a favourable neurological outcome, a CPC-score of 1 or 2 at discharge, and only 6 of these patients had a CPC-score of 3 or higher (3%).One third of patients with sudden IHCA were discharged from Conclusions: hospital and alive at day-30, a clear majority without cognitive deficit related to the cardiac arrest. High monitored care, witnessed cardiac arrest and shockable rhythm were factors associated with high success; however, more than half of surviving patients had initially a none-shockable rhythm. PubMed Abstract | Publisher Full Text 3. Ong MEH, Perkins GD, Cariou A: Out-of-hospital cardiac arrest: prehospital management. Lancet. 2018; 391(10124): 980-988. PubMed Abstract | Publisher Full Text 4. Jakobsson J, Dahlqvist M, Rehnqvist N: Resuscitation of hospitalized patients. J Intern Med. 1990; 227(1): 15-8. PubMed Abstract | Publisher Full Text 5. O'Sullivan E, Deasy C: In-hospital Cardiac Arrest at Cork University Hospital. Ir Med J. 2016; 109(1): 335-8. PubMed Abstract 6. Kolte D, Khera S, Aronow WS, et al.: Regional variation in the incidence and outcomes of in-hospital cardiac arrest in the United States. Circulation. 2015; 131(16): 1415-25. PubMed Abstract | Publisher Full Text 7. Tirkkonen J, Hellevuo H, Olkkola KT, et al.: Aetiology of in-hospital cardiac arrest on general wards. Resuscitation. 2016; 107: 19-24. PubMed Abstract | Publisher Full Text 8. Shao F, Li CS, Liang LR, et al.: Incidence and outcome of adult in-hospital cardiac arrest in Beijing, China. Resuscitation. 2016; 102: 51-6. PubMed Abstract | Publisher Full Text 9. Herlitz J, Bång A, Aune S, et al.: Characteristics and outcome among patients suffering in-hospital cardiac arrest in monitored and non-monitored areas. Resuscitation. 2001; 48(2): 125-35. PubMed Abstract | Publisher Full Text 10. Herlitz J, Aune S, Bång A, et al.: Very high survival among patients defibrillated at an early stage after in-hospital ventricular fibrillation on wards with and without monitoring facilities. Resuscitation. 2005; 66(2): 159-66.PubMed Abstract | Publisher Full Text
The importance of prompt recognition of cardiac arrest and initiation of cardio-pulmonary resuscitation has been shown repeatedly [bib_ref] Early cardiopulmonary resuscitation in out-of-hospital cardiac arrest, Hasselqvist-Ax [/bib_ref]. The chain of survival, prompt recognition, early/ bystander cardiopulmonary resuscitation (CPR) and early defibrillation is indeed of outmost importance 3 . Efforts to improve the results from out-of-hospital have been implemented and our hospital has likewise put training efforts into basic and advanced CPR.
The aim of the present retrospective register project was to study sudden in-hospital cardiac arrest (IHCA) and the outcomes of emergence team resuscitation at a hospital in Sweden.
# Methods
This is a retrospective single-centre register study in which the Swedish CPR Registry was used to access all reported cases of IHCA
# Results
A total of 574 patients with sudden IHCA were included in the study: 340 males and 234 females, with a mean age of 73 ± 14 years: 72 ± 13 for males and 75 ± 14 for the females (p<0.05).
The most common place for a sudden cardiac arrest was the Coronary Care Unit (CCU) followed by cardiology and medical wards. A majority (84%) of the cardiac arrests were witnessed, and bystander CPR was initiated within one minute in 96% of cardiac arrest cases. The most common initial rhythm was asystole (n=215) and the least common was VT/VF (n=147). The highest prevalence of VT/VF (57%) was seen in the percutaneous coronary intervention lab followed by the CCU and Intensive Care Unit (ICU) (33% and 27%).
In total, 333 (55.5%) of cardiac arrest patients were successfully resuscitated and had ROSC: 195 patients (34%) were discharged from hospital and 191 (33%) were still alive at day-30 after cardiac arrest (33%), see [fig_ref] Figure 1: Outcome of resuscitation of in-hospital cardiac arrest [/fig_ref].
The highest 30-day survival rate was seen in patients with cardiac arrest in the PCI lab (61%), with the next to highest 30-day survival rate (46%) seen in the CCU.
Shockable rhythm was associated with success: CCU, VT/VF alive at day-30 had 21 out of 28 patients (75%); PCI, VT/VF alive at day-30 had 26 out of 35 patients (74%); and ICU, VT/VF alive at day-30 had 7 out of 12 patients (58%). Overall 89 out of the 167 patients (53%) alive at day-30 had an initial noneshockable rhythm. Age had an impact: patients alive at day-30 were significantly younger than those who were not alive at day-30 (69 vs 75 years; p=0.001) [fig_ref] Table 1: Factors with impact on successful outcome of resuscitation of in-hospital cardiac arrest [/fig_ref].
CPC-score was available for 188 out of the 195 patients that were alive to be discharged (96%). In total, 96.5% of patients where data was available had a favourable neurological outcome after cardiac arrest, i.e. a CPC score of 1 or 2 at discharge.
# Discussion
We found that one third of patients suffering sudden IHCA were alive at day-30 and that patients alive to be discharged did not experience significant impairment of cognitive function. A majority of cardiac arrests were witnessed cardiac arrest and CPR had been initiated within 1-2 minutes. Having VT/VF as an initial rhythm and a lower age of the patient increased the chance of survival. However, it is worth noticing that more than half of the surviving patients had a non-shockable initial rhythm. A previous study at Danderyd Hospital in the late 1980s found only 9 out of 61 IHCA patients were alive to be discharged (15%) 4 . The survival rate seen in our study is higher than that presented from a study in Ireland on in-hospital resuscitation in 2011, one year before the start of our study, which found a 27% survival rate of discharge 5 . The average survival rate in our study is also higher than the survival rate reported from a US survey of in-hospital resuscitation including a total of 838,465 patients 6 . Data analysed from the Nationwide Inpatient Sample databases between 2003 and 2011 showed a 24.7% overall survival to hospital discharge 6 . A study conducted in Finland between 2009 and 2011, including 279 adult IHCA patients attended by the medical emergency team in a university hospital's general wards, found a 180-day survival rate of 19% 7 . They commented on the importance of shockable primary rhythm, monitored/witnessed event and low comorbidity score for survival. One should acknowledge that our study covered the period 2012 to 2017 and all hospital wards, including coronary and general intensive care departments. A study from China revealed a low survival rate where only 9.1% of patients were discharged alive 8 . Our results are however in line with a previous studies from Sweden. Herlitz et al. found a 43% survival rate for discharge among cardiac arrest patients suffering cardiac arrest in wards with monitoring facilities, and a 31% survival rate among cardiac arrest patients in general wards. They also found cerebral function to be favourable in most patients 9,10 .
Our study does have limitations. We did not study the causes of cardiac arrest. It should be acknowledged that cardiac arrest cases throughout the hospital were included in the study, not only on cardiac arrest in high dependency wards and in patients with heart disease. There is missing data for initial rhythm in about 10% of cases, which means that conclusions concerning prevalence of different cardiac rhythms must be performed with caution. It should also be acknowledged that we do not have data on time to defibrillation.
To conclude, one third of IHCA patients resuscitated by the emergency team could be discharged alive and were still alive at day-30 in our study cohort, a majority without signs of cognitive impairment related to cardiac arrest. Most cardiac arrests were witnessed and CPR had been initiated within minutes. We found initial shockable rhythm VT/VF to be a factor related to successful CPR, which is similar to what has been shown for out-of-hospital CA; however, it should be noted that more than half of survivors had a none-shockable initial rhythm.
## Data availability
The data has been retrieved from the Swedish CPR register (https://www.hlr.nu/svenska-hlr-registret/). This is a national database, supported by the Swedish and European Resuscitation Councils. The data can be retrieved by request from CPR register (https://shlrsjh.registercentrum.se/) following Ethical Review board approval on application (https://www.epn.se/en/ start/).
## Competing interests
No competing interests were disclosed.
## Grant information
This study was supported by the Department of Anaesthesia & Intensive Care, Danderyds Hospital. No external funding was provided.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
## Is the work clearly and accurately presented and does it cite the current literature? yes
## Is the study design appropriate and is the work technically sound? partly
## Are sufficient details of methods and analysis provided to allow replication by others? partly
## If applicable, is the statistical analysis and its interpretation appropriate? yes
Are all the source data underlying the results available to ensure full reproducibility? Partly
## Are the conclusions drawn adequately supported by the results? yes
No competing interests were disclosed. The aim of the study was to investigate sudden in-hospital cardiac arrests. The authors report that one third of the patients that suffer in-hospital cardiac arrest are discharged alive form the hospital. The authors describe that the chances of survival increases with shockable rhythm and lower age. Furthermore, and somewhat surprising, the authors report that more than half of the patients initially had a non-shockable rhythm.
## General comments:
The study is a single centre study in which the authors report a survival rate following in-hospital cardiac arrest somewhat higher than described in most other studies. Between the lines, this reviewer gets the notion that the medical emergency team is responsible for the favourable results. The study is, however, devoid of speculations on why this apparent increase in survival is found at Danderyd Hospital compared with other hospitals. This is somewhat disappointing. There is ample literature available to support discussions on the potential benefits of applying medical emergency teams and I feel that a discussion on caused for survival should have been discussed.
The authors report that more than half of the patients that survive to discharge are met with an initial non-shockable rhythm. This is an interesting finding and should have been discussed. As it stands, the paper is just a reporting on rather favourable outcomes following in-hospital cardiac arrest and is not delving deeper into causes and explanations.
## Specific comments:
One sentence is rather difficult to comprehend: "The aim of the present retrospective study was to study sudden in-hospital cardiac arrest (IHCA) and the outcomes of emergence team resuscitation in a university hospital in Sweden." Should the sentence read: "The aim of the present retrospective study was to study sudden in-hospital cardiac arrest (IHCA) and the cerebral outcomes following resuscitation by an emergency team in a university hospital in Sweden." Statistics: When applying means, standard deviations and t-test, the tested variables should follow a normal distribution. Have the authors assured that?
## References:
The list of references is rather small and do not support a deeper discussion of results. Not one reference to the concept of medical emergency teams is made.
to the concept of medical emergency teams is made. Although apparently not the scope of this paper, a more comprehensive discussion of medical emergency teams would have been in order and the list of references should have reflected that.
## Is the work clearly and accurately presented and does it cite the current literature? yes
Is the study design appropriate and is the work technically sound? Yes
## Are sufficient details of methods and analysis provided to allow replication by others? yes
## If applicable, is the statistical analysis and its interpretation appropriate? partly
Are all the source data underlying the results available to ensure full reproducibility? Yes
## Are the conclusions drawn adequately supported by the results? yes
No competing interests were disclosed.
## Competing interests:
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
The benefits of publishing with F1000Research:
Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more
The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact [email protected]
[fig] Figure 1: Outcome of resuscitation of in-hospital cardiac arrest. ROSC, return of spontaneous circulation. [/fig]
[table] Table 1: Factors with impact on successful outcome of resuscitation of in-hospital cardiac arrest. [/table]
|
Primary extra nodal Non-Hodgkin's lymphoma of urinary bladder presenting as a bladder tumor: A case report
Introduction: Primary non-Hodgkin's lymphoma (NHL) of urinary bladder is an exceedingly rare entity accounting for 0.2% of the primary neoplastic lesions. This tumor has female predominance; with most of the cases seen in middle-aged females. It primarily affects urinary bladder without involvement of the surrounding tissues and lymph nodes. The common presentations include hematuria, dysuria, urinary frequency, nocturia, and abdominal or back pain. The clinical, radiological and endoscopic signs are not very specific. It is diagnosed by its characteristic morphology and immunohistochemical characteristics. The various therapeutic options available are chemotherapy; radiotherapy and surgery used either alone or in combination. Presentation of Case: We hereby report a case of 40 years old man who presented with hematuria as the presenting symptom. On radiology, diffuse thickening of bladder wall was noted, which was biopsied. On histopathology, it was NHL, Diffuse large B cell type. He was treated with chemotherapy (6 cycles of CHOP) and radiotherapy for primary NHL of the bladder and now he is in complete remission. Discussion: Primary lymphoma of the urinary bladder is exceedingly rare. Non-specific presentation and rarity of this disease pose a diagnostic challenge for both the clinician and the histopathologist. Diagnosis is based upon the characteristic morphology and is supported by immunohistochemical analysis.Conclusion: All patients with extra-nodal lymphoma need thorough diagnostic work up like nodal lymphomas to arrive at exact staging of the disease to outline subsequent management.
# Introduction
Primary extra nodal non-Hodgkin's lymphomas (NHL) arise from the tissues other than lymph nodes and sites that normally do not contain lymphoid tissues. These comprise about 10-20% of all NHL cases. This entity is heterogeneous and complex in their behavior [bib_ref] Primary lymphoma of the urinary bladder presenting as a large pelvic mass, Sonmezer [/bib_ref]. The most common sites involved are gastrointestinal tract, head and neck, skin, central nervous system, bone, testis, breast and thyroid. Primary lymphomas of the urinary bladder are extremely rare, representing 0.2% of all extranodal lymphomas and < 1% of all bladder tumors [bib_ref] Imaging findings of a primary bladdermaltoma, Maninderpal [/bib_ref]. Eve and Chaffey discovered the disease in 1885 [bib_ref] The first case of vesico-vaginal fistula in a patient with primary lymphoma..., Evans [/bib_ref]. It primarily involves urinary bladder without involving the surrounding tissues, lymph nodes and bone marrow. Females are affected more frequently than males and most of the cases occur in middle-age females. The most common presentations are gross hematuria followed by concomitant urinary tract infection, dysuria, increased urinary frequency and abdominal or back pain. Complications like hydronephrosis, fistulas or involvement of the entire bladder are very rare. The radiological investigations of the urinary bladder reveal submucosal masses: 70% of cases are solitary masses; 20% of cases are multiple masses; and 10% of cases show diffuse bladder wall thickening [bib_ref] Primary lymphoma of the urinary bladder presenting as a large pelvic mass, Sonmezer [/bib_ref].
The lymphoma of the urinary bladder can be classified into three different categories: (i) primary lymphoma localized to the bladder; (ii) lymphoma presenting in the bladder as the first sign of disseminated disease; (iii) recurrent urinary bladder involvement by lymphoma in patients with a history of malignant lymphoma (secondary lymphoma). The majority of primary bladder lymphomas are low-grade tumors with a good prognosis; the most common type being extranodal marginal zone/mucosa-associated lymphoid tissue (MALT) lymphoma and this typically affects adults who are more than 60 years old. The history of https://doi.org/10.1016/j.amsu.2020.05.045 Received 17 April 2020; Received in revised form 27 May 2020; Accepted 27 May 2020 chronic cystitis is commonly associated with this type of tumor. Highgrade tumors are rarer, with the most common type being diffuse large B-cell lymphoma (DLBCL) [bib_ref] Primary bladder lymphoma: management and outcome of 12 patients with a review..., Hughes [/bib_ref]. Usually, the diagnosis of primary lymphoma of the bladder is one of exclusion. It is made on biopsy with immunohistochemical study and after a negative study of disease extension, which includes bone marrow biopsy, computed tomography (CT) and positron emission tomography (PET) to assess for other nodal or extranodal involvement when a possible primary lymphoma of the bladder is suspected. The differential diagnoses of lymphoma of the urinary bladder include urothelial carcinoma with prominent lymphoid infiltrate and undifferentiated carcinoma [bib_ref] The first case of vesico-vaginal fistula in a patient with primary lymphoma..., Evans [/bib_ref]. The prognostic factors of lymphoma of the urinary bladder include histological subtype and the stage of the tumor. Radiotherapy, chemotherapy and surgical excision are various treatment modalities [bib_ref] Primary lymphoma of the urinary bladder presenting as a large pelvic mass, Sonmezer [/bib_ref]. Due to its rarity very few cases have been reported in the literature. Herein we describe a case of primary NHL of urinary bladder diagnosed and managed at our institute. The SCARE criteria were utilized for this case report [bib_ref] OrgillFor the SCARE Group, The SCARE 2018 statement: updating consensus surgical CAse..., Agha [/bib_ref].
## Case report
A 40-years-old male, farmer by occupation, resident of Rohtak, Haryana (India), presented to urology outpatient department with painless gross hematuria for 2 weeks. There was no history of fever, loin pain, dysuria, urgency or increased frequency of urine. He was a nonsmoker. On clinical examination, no pallor or lymphadenopathy was found. Abdominal examination revealed no organomegaly. His routine hematological examination including complete blood counts, coagulation profile, liver function tests, renal function tests and serum electrolytes were within normal limits. Serum biochemical and electrolyte parameters were within normal range. Serology for HIV, HbSAg and HCV were non-reactive. Urine analysis showed hematuria. On ultrasonography and Computed Tomography scan, a diffuse thickening of bladder wall was noted [fig_ref] Figure 1: On CT Scan, diffuse thicking of urinary bladder is evident [/fig_ref]. The patient was planned for cystoscopy and transurethral resection was performed.
The biopsy comprised of multiple grey white to grey brown soft tissue pieces measuring together 3x2x1 cm and weighing 4 gm. Histological examination revealed diffuse infiltration of lamina propria by atypical lymphocytes with pleomorphic and hyperchromatic nuclei (Figs. 2 and 3). On immunohistochemistry the tumor cells were strongly positive for CD 20 [fig_ref] Figure 4: On IHC CD 20 is diffusely and strongly positive [/fig_ref] and were negative for Bcl2, CD 5, CD10, CD 23, cyclin D1 and Tdt. The proliferative fraction of cells, as determined by Ki-67 staining was 40% [fig_ref] Figure 5: Ki67 labelling index of approximately 30% [/fig_ref]. Immunohistochemical analyses were negative for cytokeratin and vimentin. Thus, based on histological and immunohistochemical findings, a diagnosis of B cell lymphoma was made.
Due to the highly invasive nature of the tumor, bone marrow (BM) biopsy was performed which showed no evidence of leukemia or lymphoma. Further investigations including chest radiography and CT scans of his chest, abdomen and pelvis were performed which neither showed any lymphadenopathy nor any organomegaly. No metastatic lesions were found. Thus, the final diagnosis was primary B cell lymphoma of the bladder. The disease was classified stage IE according to the Ann Arbor Staging system and the patient had a score of 0 by using the international prognostic index (IPI). Considering age, general condition of the patient and stage of tumor, the patient underwent six cycles of R-CHOP regimen (Rituximab, Cyclophosphamide, Vincristine, Doxorubicin, Prednisolone) followed by radiotherapy of 30 Gy in 20 fractions 1 month after completion of all cycles of chemotherapy. There were few side effects of chemotherapy in patient which included nausea, vomiting, weight loss and fatigue. He showed good tolerance and initial response to this treatment. Thus far, PET CT scans have shown no abnormal uptake, indicative of effective treatment. Now the patient is in complete remission without evidence of local or distant recurrence on routine follow up since 2 years. His urine cytology, cystoscopy as well as urography were performed 3 months postoperatively and after every 3 months interval for a period of 2 years. All these were negative and a repeat CT scan performed after 2 years was negative too. Currently, patient is free from disease and is doing well.
# Discussion
In 1885, Eve and Chaffey first described primary malignant lymphoma of urinary bladder [bib_ref] The first case of vesico-vaginal fistula in a patient with primary lymphoma..., Evans [/bib_ref]. Usually lymphoid tissue is not observed in urinary bladder, so it is relatively uncommon site for primary extra nodal NHL accounting for less than 1% of neoplasms in urinary bladder and 0.2% of cases with extra nodal lymphoma [bib_ref] Imaging findings of a primary bladdermaltoma, Maninderpal [/bib_ref]. The most common sites involved are gastrointestinal tract, head and neck, skin, central nervous system, bone, testis, breast and thyroid [bib_ref] Primary lymphoma of the urinary bladder presenting as a large pelvic mass, Sonmezer [/bib_ref].
Primary NHL of urinary bladder commonly affects women with female to male ratio of 6.5:1.Various presenting symptoms include intermittent hematuria, dysuria, nocturia, urinary frequency, pain in lower abdomen and frequent or recurrent urinary tract infections [bib_ref] Non-Hodgkin's lymphoma presenting as a primary bladder tumor: a case report, José [/bib_ref]. The etiology of primary NHL of the bladder has not been elucidated, because of the rarity of this neoplasm. According to some authors, lymphoma may be secondary to chronic cystitis. Tumor associated with chronic ystitis may originate in the lymphocytes of the submucosa of bladder because of inflammatory response. Other authors have raised the possibility of a residual embryonic cloaca, source lymphoid proliferation in adulthood [bib_ref] Non-hodgkin lymphoma of the bladder, Antunes [/bib_ref].
Cohen et al. stated that malignant lymphoma of the urinary bladder can be classified into one of three different clinical groups as follows: (i) Primary lymphoma localized to the bladder; (ii) Lymphoma presenting in the bladder as the first sign of disseminated disease (nonlocalized lymphoma); (iii) Recurrent urinary bladder involvement by lymphoma in patients with a history of malignant lymphoma (secondary lymphoma). Primary bladder lymphomas are of two types: Mucosa associated lymphoid tissue (MALT) and Diffuse large B cell type of lymphoma (DLBCL). MALT type lymphoma is more common. Secondary involvement of the bladder by systemic lymphoma has been documented in 10-20% of NHL patients [bib_ref] Primary low-grade B-cell lymphoma of the urinary bladder: case report and literature..., Cohen [/bib_ref].
Bladder lymphoma usually locates at the base and the trigone of the bladder as a sessile mass with normal urothelium. Commonly such lesions are initially diagnosed as urothelial carcinoma of the bladder. CT scans and MRI examination usually do not add much information for diagnosis due to their low sensitivity. But cystoscopy, histological examination and immunohistochemistry play major role for diagnosis of the disease. On cystoscopy, these tumors are visualized as well-defined intravesicular masses typically located at the dome or the lateral walls of the bladder [bib_ref] Imaging findings of a primary bladdermaltoma, Maninderpal [/bib_ref]. The DLBCL is characterized by a proliferation of atypical lymphoid cells. The tumor cells are of large size and often resemble normal centroblasts or immunoblasts. These cells express CD20, CD79a and bcl-2. The positivity of CD10, CD5 and Bcl6 remains variable. The differential diagnosis is essentially a poorly differentiated carcinoma; melanoma, burkitt's and Hodgkin's lymphoma; underlining the interest of an immunohistochemical panel in order to confirm the diagnosis. Immunohistochemical staining for B-cell type lymphomas routinely test positive for CD19, CD20 and CD21. Furthermore, low grade lymphomas routinely test positive for CD20, CD21, and CD43 cell markers, while high grade lymphomas are correlated with CD3, CD20, and CD3 [bib_ref] Malignant lymphoma of the urinary bladder: a clinicopathological study of 11 cases, Bates [/bib_ref]. Once the type of bladder lymphoma has been identified, it is imperative to make the distinction between an indolent, low grade neoplasm versus an aggressive, high grade malignancy in order to initiate the most appropriate treatment regimen for that patient. If determined to be a high grade tumor, it is imperative to exclude systemic involvement; this can be done with a BM biopsy and a PET scan [bib_ref] Malignant lymphoma of the urinary bladder: a clinicopathological study of 11 cases, Bates [/bib_ref].
There are different therapeutic strategies available for primary extra-nodal NHL like radical cystectomy, radiotherapy and chemotherapy depending upon the clinical behavior of tumor, risk of disease progression, life expectancy and general condition of the patient. For MALT lymphoma it is recommended that a less systematically toxic therapy like TURBT be attempted first, followed by chemotherapy, radiation, or combination therapy. No recommendations on lesion size have been published for qualifications for TURBT. Therefore following criteria set for other bladder tumors could be used; TURBT has proven effective in urothelial bladder cancer that has yet to invade the musclaris propria (stage Ta, Tis, and T1). As for the type of chemotherapy, all of the regimens (R-CHOP, CHOP, ChlVP, ChlD) were successful in treating low grade bladder lymphomas; R-CHOP was reported by the literature to be used most frequently. DLBCL is an aggressive lymphoma and without treatment, leaves the patient with a prognosis of a matter of months. Chemotherapy is currently the preferential treatment due to its ability to treat early systematic disease which has not been detected. The chemotherapy regimen of R-CHOP (ritoximab, cyclophosphamide, duanorubacin, vincristine, prednisolone) has documented success in treating both low and high grade primary bladder lymphomas as either solitary treatment or combination therapy. Rituximab has been shown to help overcome the chemotherapy resistance seen in patients with BCL-2 over expression by inducing cell apoptosis through the BCL-2 regulated mitochondrial pathway increasing; the survival rate of patients with DLBCL on CHOP chemotherapy by 10-15% with very little additional toxicity.Surgical techniques such as a cystectomy are recommended for any invasive transitional cell carcinoma that have spread into the muscular layer, however there are no clear surgical recommendation for bladder lymphomas and surgical modalities have been disputed in the literature due to the morbidity and mortality associated with these procedures [bib_ref] Recent advances in the diagnosis and the treatment of bladder cancer, Cheung [/bib_ref].Radiotherapy can be used in first intention especially in low grades or in the adjuvant setting after resection but cannot constitute a standard treatment [bib_ref] Recent advances in the diagnosis and the treatment of bladder cancer, Cheung [/bib_ref] [bib_ref] Primary extra nodal non-Hodgkin's lymphoma of urinary bladder: a case report and..., Unmesh [/bib_ref]. Indeed, its side effects are frequent and managing local recurrence after irradiation is very complicated. The optimum follow-up protocol for patients treated for primary lymphoma of the bladder has yet to be determined. Follow-up evaluations should take place every 3 or 6 months for the first 2 years and yearly thereafter, as is the case for transitional cell carcinoma. It must include at least ultrasound examination and cystoscopy. The prognosis is difficult to be determined given the rarity of primary NHL of the bladder. However, patients with high-grade lymphoma should be considered to have systemic disease and therefore, the IPN can be used to predict prognosis. The factors that should be taken into consideration to outline the management of such patients are age, performance and physiologic status of the patient, stage of tumor and International Prognostic Index (IPN). Calculating IPN that depends on age, stage, serum lactate dehydrogenase, number of extra nodal sites involved and performance status helps to determine the prognosis of the patient. The indexed case has been highlighted for its rarity. Even if the condition is rare, possibility of malignant lymphoma should be always kept in mind for differential diagnosis of bladder tumor. Because once diagnosed there is chance of excellent complete remission of the lymphoma resulting in good prognosis. As there are no clear therapeutic guidelines for managing such cases, present case will add to the present literature.
# Conclusion
Primary lymphoma of the urinary bladder is exceedingly rare. Nonspecific presentation and rarity of this disease pose a diagnostic challenge for both the clinician and the histopathologist. Diagnosis is based upon the characteristic morphology and is supported by immunohistochemical analysis. Radiotherapy and chemotherapy are useful and effective in high grade and locally advanced tumor and it seems to give good results. All patients with extra-nodal lymphoma need thorough diagnostic work up to arrive at exact staging of the disease to outline further management.
## Declaration of competing interestcoi
The manuscript has been read and approved by all the authors, is original and honest work and has not been sent to any other journal for publication and there is no conflict of interests between the authors.
Kindly consider it for publication in your esteemed journal.
## Provenance and peer review
Not commissioned, externally peer reviewed.
[fig] Figure 1: On CT Scan, diffuse thicking of urinary bladder is evident. [/fig]
[fig] Figure 2: On H&E, tumor cells are seen invading the muscle bundles. (100X). [/fig]
[fig] Figure 3: On high magnification, monomorphic population of lymphoid cells (H& E−400X). [/fig]
[fig] Figure 4: On IHC CD 20 is diffusely and strongly positive (100X). [/fig]
[fig] Figure 5: Ki67 labelling index of approximately 30%. (100X). [/fig]
|
STREAM: Static Thermodynamic REgulAtory Model of transcription
Motivation:Understanding the transcriptional regulation of a gene in detail is a crucial step towards uncovering and ultimately utilizing the regulatory grammar of the genome. Modeling transcriptional regulation using thermodynamic equations has become an increasingly important approach towards this goal.Here, we present STREAM, the first publicly available framework for modeling, visualizing and predicting the regulation of the transcription rate of a target gene. Given the concentrations of a set of transcription factors (TFs), the TF binding sites (TFBSs) in a regulatory DNA region, and the transcription rate of the target gene, STREAM will optimize its parameters to generate a model that best fits the input data. This trained model can then be used to (a) validate that the given set of TFs is able to regulate the target gene and (b) to predict the transcription rate under different conditions (e.g. different tissues, knockout/additional TFs or mutated/missing TFBSs). Availability: The platform independent executable of STREAM, as well as a tutorial and the full documentation, are available at
# Introduction
Transcription of a gene can be induced by the binding of specific transcription factors (TFs) to so-called cis-regulatory modules (CRMs). The frequency and duration of the binding events are influenced by the concentrations of the TFs, the binding affinities and location of the TF binding sites (TFBSs) in the CRM and the properties of the TFs themselves (e.g. effectiveness, competitive interaction with other TFs). With the availability of an increasing number of detailed measurements of gene concentrations in different situations (e.g. tissues, developmental time points) as well as TF-DNA binding affinities, it has become possible to build mathematical models for transcriptional regulation. Building mathematical models to associate a specific occupation of a specific CRM with an observed transcriptional response promotes a better understanding of the transcriptional regulation and enables in silico hypothesis testing about postulated regulatory TFs or mechanisms.
An increasingly successful approach to mathematically simulating transcriptional regulation is using thermodynamic models that model the interaction of TFs and DNA using kinetic equations. * To whom correspondence should be addressed.
Several such thermodynamic models have been proposed in the last years [bib_ref] Quantitative and predictive model of transcriptional control of the Drosophila melanogaster even..., Janssens [/bib_ref] [bib_ref] Predicting expression patterns from regulatory sequence in Drosophila segmentation, Segal [/bib_ref] [bib_ref] Computational models for neurogenic gene expression in the Drosophila embryo, Zinzen [/bib_ref]. These models take the CRM sequence, a set of TFs along with their concentrations and predict the transcriptional response of the target gene as mediated by the CRM and the TFs. A training algorithm is used to optimize the model's internal parameters to minimize the difference between the observed and predicted transcriptional response.
## Approach and usage
Here we present STREAM, a Java-implemented framework to calculate and visualize transcriptional regulation using thermodynamic modeling approaches. STREAM currently uses the thermodynamic model introduced by, but the framework is flexible and can be used in conjunction with other models implemented in Java. STREAM offers several optimization methods including gradient descent and simulated annealing for adjusting the internal parameters of the model to best fit the user's input data. To the best of our knowledge, STREAM is to data the only publicly available framework for modeling the regulation of the rate of transcription. STREAM has been tested extensively on the even-skipped gene (eve) in Drosophila melanogaster [bib_ref] Studying the functional conservation of cisregulatory modules and their transcriptional output, Bauer [/bib_ref].
STREAM can be executed using a graphical user interface (GUI) as well as via the command line. The GUI is illustrated in [fig_ref] Figure 1: Screen shot of the GUI of STREAM [/fig_ref]. It offers the same functionality (e.g. multistart options of the optimization or automatic cross-validation evaluation) as the command-line tool, but in an intuitive and dialogue-based fashion. Both the command-line and the GUI version can save the current result and settings of the program into a file, which makes saving and modifying previous experiments simple.
# Methods
In order to generate a model for a particular target gene, the user must identify a set of putative TFs and a putative regulatory region. The suspected role of each TF, x, as an activator x ∈ A or repressor x ∈ B must be specified by the user. The program also requires measurements of the concentrations and binding preferences of those TFs, as well as the transcriptional output of the target gene. In the following section, we introduce the required input data, D, which contains the concentration and rate data, C, and a TFBS map, S.
The concentration and rate data, C, comprises a set of independent data points. Each data point is a pair (V,v t ), with V = [fig_ref] Figure 1: Screen shot of the GUI of STREAM [/fig_ref] , ... ,[a n ]), a vector listing the protein concentrations of the putative regulatory TF proteins © 2008 The Author(s) This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. (a 1 , ..., a n ), and v t giving the corresponding observed transcription rate of the target gene.
The concentration and rate data, C, can be measured by various methods. To measure the TF concentrations, in situ antibody staining can be used to measure V. To obtain the corresponding v t , one can proceed indirectly by measuring the mRNA levels of a reporter gene by staining against the mRNA of the reporter [bib_ref] Dynamical analysis of regulatory interactions in the gap gene system of Drosophila..., Jaeger [/bib_ref].
For visualization purposes, STREAM allows the user to label the data points with up to two 'features' (e.g. the 'condition' under which the measurements were made). Features may be 'continuous' (real numbers as in the example above), or 'categorical' such as cell types, tissue types or experimental treatments. STREAM utilizes the values of the features for visualization only, and the user may interactively select either feature as the X-axis for plots of the data.
Besides the concentration data, the input data also contains the TFBS map, S, corresponding to the regulatory region of interest. The TFBS map, S = (s 1 ,s 2 ,...,s n ), is a list, where each s i represents a TFBS as a triple (a,l,s) giving the name, a, of the binding TF, the position, l, of the binding site and the log-odds score (natural logarithm), s, of the site. The log-odds score is proportional to the binding strength of the TFBS [bib_ref] DNA binding sites: representation and discovery, Stormo [/bib_ref]. S can be constructed from experimentally verified binding sites, in silico predicted sites using a prediction algorithm such as FIMO (http://meme.sdsc.edu), or a combination of both. For more detailed information, see [bib_ref] Studying the functional conservation of cisregulatory modules and their transcriptional output, Bauer [/bib_ref].
The objective of the optimization is to determine the set of model parameters, , that optimally explains the input data, D.
depends on the thermodynamic model. Currently, STREAM implements the model introduced by. This model uses free parameters = (θ 0 ,R 0 ,W), where θ 0 is an energy barrier, R 0 is the maximal transcription rate, and W contains a tuple, (K x ,E x ), for each TF, x, where K x is the association constant of the TF to the DNA and E x is the effectiveness of the TF to activate transcription, if x ∈ A, or to repress, if x ∈ B. For more details seeand [bib_ref] Studying the functional conservation of cisregulatory modules and their transcriptional output, Bauer [/bib_ref].
Four different optimization methods are implemented: simulated annealing (SA), gradient descent (GD), genetic algorithm (GA) and limited-memory quasi-Newton unconstrained optimization (LBFGS) (for a comparison of the optimization methods see D.C.Bauer and T.L.Bailey, manuscript in preparation). All optimization methods seek to optimize the free parameters of the Reinitz model by minimizing the root mean-squared (RMS) error between the known transcription rate and the rate predicted by the Reinitz model, averaged over all input points, D [bib_ref] Studying the functional conservation of cisregulatory modules and their transcriptional output, Bauer [/bib_ref].
## Future developments
Besides the Reinitz model, we plan to provide additional models with enhanced functionality to simulate interacting TFs in a more detailed way, e.g., by incorporating TF-TF cooperation. Furthermore, simplifying models by using discreet TFBSs might introduce artifacts, hence, changing the model approach to use continuous binding gradients deems beneficial. Future research on other CRMs will guide the development of new models and our framework can provide the environment to directly compare them.
In addition, we plan to extend the approach to optimize one model to fit more than one CRM. Being able to fit one model to the data of several CRMs, which are suspected to have the same regulatory TFs, increases the confidence in the produced model.
Finally, we plan to improve the functionality of the GUI in manipulating input data, e.g., by an interactive interface to vary the properties of the TFBS map and to directly observe the changes in the predicted transcription rate.
[fig] Figure 1: Screen shot of the GUI of STREAM. The interface is exemplified on a model trained for the even-skipped gene (eve). [/fig]
|
Effect of faecal consistency on virological diagnosis
A study was set up to investigate the effect of consistency of routine faecal specimens on the diagnostic yield by electron microscopy (EM) and virus isolation. A total of 3078 specimens were characterized as solid, semisolid, or liquid. Of 2568 specimens processed by EM a virus was demonstrated in 8.6% of liquid, 19.9% of semisolid and 25.2% of solid specimens (Chi-squared for linear trend, P value <0.0001). This observation was valid for both adenovirus (2.4% 5.0% and 6.6%) and rotavirus (5.2% 13.6% and 16.6%). Virus isolation was positive in 3.6% of liquid, 17.4% of semisolid and 18.1% of solid specimens. (Chi-squared for linear trend, P value <0.0001).We suggest that solid faecal specimens at the end of an episode of diarrhoea will have a higher diagnostic yield than liquid specimens at the peak of symptoms. Our findings repudiate the commonly held dogma that viruses of gastroenteritis are more likely to be found in liquid than in solid faecal specimens. This finding has important implications for those establishing diagnostic algorithms for the investigation of viral gastroenteritis.
# Introduction
Electron microscopy (EM) on faecal specimens is a timeconsuming and expensive method for diagnosis of viral gastroenteritis. During the investigation of outbreaks of gastroenteritis it may be necessary or desirable to prioritize specimens processed. Traditionally in our laboratory this was done by assessing the consistency of the faeces received and preferentially processing those specimens which were liquid and leaving the more solid specimens to later. A more restrictive approach has been employed in the Public Health Laboratory Service of England and Wales, where the customary practice is to seek only "unformed" faeces for EM (PHLS North West Electron Microscopy Policy, August 1996). The current recommendations from Centers for Disease Control, U.S.A., are that the more formed the stool specimen the lower the diagnostic yield, and therefore liquid specimens only should be collected and examined by EM. 1'2 We set up a study to investigate the validity of this approach to EM and also to look at the relationship between faecal consistency and virus isolation rates.
## Materials, methods and patients
From 1 March 1995 to 30 June 1996 a total of 3078 faecal specimens arriving at the laboratory were examined on receipt and categorized as liquid, semisolid or solid. They were processed according to our laboratory's routine procedures for virus isolation and EM. A 10% suspension of faeces was prepared in 10 ml of Hank's salt solution containing 1000 units of penicillin, 1000 ~tg of streptomycin and 2.5 ~tg of amphotericin B per ml in a glass universal bottle containing sterile glass beads. The suspension was shaken in a mechanical floor mounted shaker for 15 min and centrifuged at 4000 g for I h at 4 °C. All faecal specimens were processed for virus isolation. Those specimens which from the clinical details on the request form were clearly from patients with an illness other than gastroenteritis were not tested by EM. All other specimens were tested by EM, including those with no clinical details on the request form. For the period of the study the consistency category of the faecal specimen was not used to decide whether or not to do EM.
EM was performed using a protocol based on a published methodology. 3 In brief, drops of clarified 10% faecal suspension were placed on a sheet of dental wax. A glow discharged carbon-coated Formvar grid, film side downwards, was placed on each drop of specimen suspension and left at room temperature for 3 h or overnight. Grids were dried and stained with 2% methylamine tungstate pH 6.5 for 5-10 min and examined at a magnification of 34 000.
Virus isolation was carried out using a microtitre plate based system using cells inoculated in suspension. ~ All specimens were inoculated into six cell lines: (HEp2 human epithelial continuous ATCC CCL23), Vero E6 (African green monkey continuous ATCC CRL1586), RD (human rhabdomyosarcoma continuous ATCC CCL136), primary rhesus monkey kidney (supplied by CPHL, Porton Down) and two human fibroblast semicontinuous cell lines (established in the Regional Virus Laboratory (RVL), Belfast, U.K.). The case notes of two inpatients who had a negative followed by a positive result for rotavirus during the same clinical episode were examined to see how the consistency of their faeces varied at the time of these results.
The data was extracted using an in-house developed modular and generic relational data application (PVCS95) constructed using Paradox DOS Version 4.01 (Borland), which serves as the RVL Belfast laboratory computer system. Clinical data accompanying the request was entered as free text and subsequently searched for eight indicator text strings (diarrh, vomit, watery, foul, offensiv, D + V, gastro, enteritis) to determine the number of requests specifically detailing a gastroenteritis-like illness. The text search was not case sensitive and recognized both partial and whole worlds. In order to assess whether group differences (positive vs. negative) were due to the consistency of the faecal sample, a Chi-squared test for linear trend was performed. All statistical analyses were performed using EPI-INFO Version 6 (CDC, Atlanta, Georgia, U.S.A.) and the conventional 5% level of significance was used throughout.
# Results
Of the 3078 faecal samples received, a total of 2568 were processed for EM. A virus was seen in 506 (19.7%).
The positivity rate increased with consistency ( Requests with text strings indicating gastroenteritis accompanying faeces processed for EM were present in 223 (54.8%) of the liquid faeces, 794 (57.4%) of the semisolid faeces and 465 (59.8%) of the solid faeces. These differences were not significant (Chi-squared for linear trend = 2.937, P-value = 0.09). In contrast, text strings indicating gastroenteritis were present in 357 (70.5%) requests accompanying EM positive faeces and in 1125 (54.5%) requests accompanying EM negative faeces (Chi-squared=225.8, P-value <0.0001).
## Case histories
Case A: a lO-month-old female was admitted to the regional infectious diseases unit on day 6 post onset of gastroenteritis, and was discharged on day 12. Faecal consistency and EM results are summarized in [fig_ref] Table I: Electronmiscroscopy results on 2568 faecal specimens categorised by consistency [/fig_ref] Case B: a 6-month-old male was admitted to the regional infectious disease unit on day 3 post onset of gastroenteritis, and was discharged on day 9. Faecal consistency and EM results are summarized in [fig_ref] Table I: Electronmiscroscopy results on 2568 faecal specimens categorised by consistency [/fig_ref].
# Discussion
The finding that the diagnostic yield from faeces examined by EM increases with consistency was unexpected. The most likely explanation for the increase in diagnostic yield with consistency is simply concentration of faeces during resolution of diarrhoea. The two illustrative case histories demonstrate that the finding of a positive EM result may occur at the time of resolution of diarrhoea. The duration of the illness in these two cases is the maximum of what would normally be expected. The were isolated during the course of the study. § All polioviruses were typed as part of an ongoing program (Enteric and Respiratory laboratory CPHL, Colindale) and were all considered to be vaccine strains. proportion of samples with clinical details containing text strings indicating a gastroenteritis type illness did not correlate with the consistency of the faeces, but did correlate with EM positivity. This implies that the more formed faecal specimens were no more or less likely to be from cases of gastroenteritis. It has previously been demonstrated that up to 40% of children who have recovered from viral gastroenteritis will still be positive for a viral agent by EM on the day of discharge from hospital, s Animal studies provide an insight. In suckling rats experimentally infected with group B rotavirus where faeces were assayed by enzyme immunoassay, 100% of the animals assayed at days 1, 4, and 5 tested positive for rotavirus. However, only 70% and 20% of the animals tested positive at days 2 and 3 postinoculation, respectively. 6 This biphasic pattern of viral excretion was confirmed by a dot hybridization assay. It was noted that the second peak coincided with the time when diarrhoea had begun to resolve and the fluid volume in the intestine had decreased. The second peak of viral excretion was shown not to be due to a second round of intestinal replication. A similar biphasic pattern has been also observed in neonatal murine rotavirus models. 7-1° It was noted that the first peak of viral excretion preceded the severe diarrhoea and that the second peak coincided with the faeces becoming solid, a This biphasic pattern was also noted in rotavirus infectivity titre (TCID50) in the mid small intestine of gnotobiotic piglets infected with rotavirus. 11 In this case the first peak of viral excretion was at 19-22 h and the second peak at 61-65 h. The results that we have observed would reflect a similar biphasic pattern of excretion in patients with gastroenteritis. Because the first peak precedes the peak symptomatology, diagnosis is often dependent on the second peak.
The finding that the virus isolation rate increases with faecal consistency was less surprising. As the readily cultivatable enteric viruses are not generally considered to cause gastroenteritis, one would expect them to be most efficiently isolated from solid faeces, where they are the most concentrated. In this study the virus isolation rate was low (3.6%) in liquid faeces, but there was very little difference in the isolation rate between semisolid (17.4%) and solid (18.1%) specimens.
An important lesson from this study is that it is clearly unwise to exclude fully formed faeces in a protocol for rationalizing the investigation of outbreaks, as these were the specimens with the highest yield of positive results. A negative result from faeces during the acute phase of the diarrhoea does not exclude a diagnosis of viral gastroenteritis. Our observations suggest that the optimum specimen is one taken either at the start of symptoms before the faeces have become liquid or at the time the diarrhoea is resolving. The results of this study have important implications for those establishing diagnostic algorithms for use in the investigation of viral gastroenteritis. Further studies looking at the kinetics of viral excretion in humans and the relationship to faecal consistency are required.
[table] Table I: Electronmiscroscopy results on 2568 faecal specimens categorised by consistency. [/table]
|
Epidemiology, classification and treatment of olecranon fractures in adults: an observational study on 2462 fractures from the Swedish Fracture Register
Purpose This nationwide study aims to describe the epidemiology, fracture classification and current treatment regimens of olecranon fractures in adults. Methods We performed a descriptive study based on registered data from the Swedish Fracture Register (SFR). All nonpathological olecranon fractures reported between 1 January 2014 and 31 December 2018 in patients aged ≥ 18 years were included. Data on age, sex, injury mechanism, fracture classification (according to the modified Mayo classification system), primary treatment and seasonal variation were analyzed. We compared patients < 65 with those > 65 years regarding injury mechanism, distribution of fracture types and subsequent treatment. Results In total, 2462 olecranon fractures were identified in the SFR. Median age was 66 years and 65% were women. Of all fractures, 303 (12%) were proximal avulsion, 1044 (42%) simple central, 717 (29%) comminuted central and 398 (16%) distal olecranon fractures. Nonoperative treatment was performed in 21% of the patients < 65 and 35% of the patients > 65 years. Tension band wiring was used for most simple central fractures. Plate fixation was used in almost half of the operatively treated fractures classified as unstable comminuted central and distal olecranon fractures. Men show a higher proportion of high-energy trauma than women in both age groups. Conclusion Isolated fractures of the olecranon occur after a low-energy trauma, especially in older women (> 65 years). Nonoperative treatment is common in uncomplicated fractures and operative treatment in more complex fractures nationwide. A shift to plate fixation in the more unstable fracture patterns is observed. These results may help health care providers and clinicians gain a better understanding of isolated olecranon fractures.
# Introduction
Olecranon fractures can be either isolated fractures of the extensor mechanism in the elbow or of more complex nature, including fracture dislocations. These fractures occur in all age groups but may be an early osteoporotic fracture given the higher incidence in elderly patients [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref]. The injury mechanism is predominately a simple fall and open fractures are rare [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref].
The size of the fractured olecranon tip decides the stability of the ulnohumeral joint, dictating the classification of the fracture and thus treatment. The Mayo classification is the most commonly used classification system dividing the fractures into type I-III, representing undisplaced, displaced and distally displaced with volar ulnar displacement. The type I-III fractures are further divided into A (non-comminuted) and B (comminuted) fractures.
The treatment of olecranon fractures ranges from nonoperative to operative treatment with sutures, tension band wiring (TBW), screw or plate fixation depending on patient factors, fracture configuration and surgeon preference [bib_ref] Olecranon fractures, Brolin [/bib_ref] [bib_ref] Nonoperative management of displaced olecranon fractures in low-demand elderly patients, Duckworth [/bib_ref] [bib_ref] Olecranon mayo IIA fractures treated with transosseous high strength suture: a series..., García-Elvira [/bib_ref] [bib_ref] Surgical and nonoperative management of olecranon fractures in the elderly: a systematic..., Chen [/bib_ref].
Copious reports have been conducted on treatment regimens and smaller randomized trials. However, only one 1 3 small single-center report describes the epidemiology of olecranon fractures [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref].
Therefore, our study aimed to describe the epidemiology of olecranon fractures including injury mechanism, fracture classification, sex and age distribution, primary treatment and seasonal variation using the national Swedish Fracture Register (SFR). By this approach, we aimed to investigate whether there are differences between men and women or younger and older patients regarding injury mechanisms, fracture pattern and subsequent treatment.
# Materials and methods
## Study design and setting
This observational cohort study was designed based on data derived from the SFR.
The SFR, established in 2011, is a national quality register for the management of fractures and treatment. Detailed data on patient and fracture characteristics, injury mechanism and fracture treatment are registered prospectively at each affiliated department via a pre-specified digital form by the treating physician. Only patients with a permanent Swedish personal identity number and fractures that have occurred in Sweden are registered. In the SFR, fractures are mainly classified according to the AO/OTA classification system. Several studies have found the registration in the SFR to have high accuracy and validity [bib_ref] Substantial accuracy of fracture classification in the Swedish Fracture Register: evaluation of..., Juto [/bib_ref] [bib_ref] High reliability in classification of tibia fractures in the Swedish Fracture Register, Wennergren [/bib_ref] [bib_ref] Validity of humerus fracture classification in the Swedish fracture register, Wennergren [/bib_ref]. The proportion of departments affiliated with the SFR has increased gradually, and at the start of this study (January 2014), 40% of affiliated departments were active. By the end of the study (December 2018), participation had risen to 85%, with a coverage of more than 80% of the Swedish population. More than 330,000 fractures had been registered in the SFR by the end of 2018.
The registration of proximal forearm fractures in the SFR includes fracture of the proximal radius and/or proximal ulna, accompanied dislocation and whether the fracture was open. First, the potential fracture of the proximal radius is registered, then the proximal ulna, followed by a question about whether there is an accompanied dislocation or not. Periprosthetic fractures can also be classified separately. Groups of fractures can be distinguished with either isolated proximal radius or ulna fractures or complex fracture configurations. In addition, information on open fractures as described by the Gustilo-Anderson (G-A) classification is available. The olecranon fractures are classified in the SFR according to a modified Mayo classification into four groups: proximal avulsion (Mayo types I A-B), simple central (Mayo type II A), comminuted central (Mayo type II B) and distal olecranon fracture (Mayo types III A-B, [fig_ref] Figure 1: Schematic description of the four fracture subtypes included in this study [/fig_ref]. Treatment is registered with the chosen type of treatment (non-operative or operative). Operative treatment is further specified into TBW, screw fixation, plate fixation and combined method.
## Patient selection
All non-pathological fractures of the olecranon (ICD code S52.00/S52.01 (closed/open)) registered in the SFR between 1 January 2014 and 31 December 2018 in patients aged ≥ 18 years were included. We excluded all isolated fractures of the proximal radius, fractures of the coronoid and proximal ulnar shaft and combinations of more than one fracture, including fracture dislocations.
## Outcome variables
Data on age, sex injury date, mechanism and type (high-or low-energy trauma), fracture classification (type, side, open/ closed fracture) and treatment were analyzed.
The cause of injury was categorized as a simple fall, a fall from a height, an unspecified fall, a transportation accident or any other cause. Transportation accidents were grouped into bicycle, motorbike or other, and fracture types were analyzed according to the modified Mayo classification. Primary treatment was studied in the following groups: TBW, screw fixation, plate fixation or a combined method.
## Statistics
Nominal variables are presented as proportions of all registered fractures and scale variables as means ± standard deviation (± SD) if normally distributed, and as median with inter-quartile range (IQR) for non-normaly distributed data. All statistics, including mean values, percentages, tables and figures, were calculated using the R version 4.0.3 software package.
## Ethics
This study was approved by the Regional Ethical Committee in Uppsala (Dnr: 2015/509 and 2019/011006).
# Results
In total, 3233 proximal forearm fractures were extracted from the SFR (ICD-10 code S 52.00/52.01). After exclusion, 2462 olecranon fractures were included in this study. Twelve patients sustained simultaneous bilateral fractures and 19 patients experienced a re-fracture on the same side during the study period. Of all fractures, 65% occurred in women and the median age was 66 (IQR 50-79) years. Women were older (median age 70 years, IQR 57-81) than men (56 years, IQR 37-72). Only small differences were found between the four subtypes in demographics, injury mechanism and high-or low-energy trauma ; [fig_ref] Figure 2: Distribution of age at injury for A all fractures and B the... [/fig_ref]. Most patients suffered their injury due to a fall, predominantly the samelevel fall ; [fig_ref] Figure 3: Distribution of the injury mechanism for the four fracture subgroups [/fig_ref]. 71% of all patients > 65 sustained their fracture through a simple same-level fall. One in four patients < 65 years sustained their fracture in a bicycle accident. High-energy injuries were rare, causing between 6 and 11% of the all fractures depending on fracture type, with the highest proportion being the distal olecranon fractures. Men sustained high enery injuries to a higher degree than women in both patients younger and older than 65 years [fig_ref] Table 2: Distribution of the type of injury for patients aged 65 years and... [/fig_ref]. 13% of all fractures in patients < 65 were highenergy injuries. The proportion increased with increasing fracture complexity, with 20% of distal olecranon fractures being high-energy injuries in patients < 65 .
## Fracture classification
Of all fractures, 303 (12%) were proximal avulsion, 1044 (42%) simple central, 717 (29%) comminuted central and 398 (16%) distal olecranon fractures. Eighty-nine fractures (3.6%) were classified as open: 28 were Gustilo-Anderson type I, 25 type II, 15 type IIIa, 1 type IIIb, and 4 had a missing G-A classification. Men displayed an even distribution of all four fracture types in all age groups, whereas women sustained all these fractures in older age, mostly the simple or comminuted central fractures [fig_ref] Figure 2: Distribution of age at injury for A all fractures and B the... [/fig_ref].
## Treatment
Most patients suffering a simple avulsion fracture were treated nonoperatively (59%). This proportion dropped for the two central olecranon fractures (32% and 14%) and the more distal olecranon fracture (16%). The method of surgical fixation differed between groups. For the more stable proximal and simple central olecranon fractures, most patients were operated on with TBW. More than half of the operatively treated patients with more unstable fractures were treated with plate fixation [fig_ref] Table 3: Treatment choice for the four fracture subgroups and overall fractures stratified for... [/fig_ref]. Patients > 65 years were more often treated non-operatively for all fracture types.
## Seasonal variation
Both women and men appeared to sustain more fractures in winter (December to January) and there was an observed tendency to a slight increase in summer (May-September) [fig_ref] Figure 4: Seasonal [/fig_ref].
# Discussion
The main finding in this register-based descriptive study of 2462 olecranon fractures registered in the SFR between 2014 and 2018 is that most fractures occurred in elderly women due to low-energy same-level simple falls. Our study is the first to analyze the epidemiology of olecranon fractures for age and sex distribution, injury mechanisms, fracture type and treatment in a nationwide cohort. Previous investigations were restricted to single-center studies [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref] , with limited external validity as reference centers might treat a selected cohort of patients. In the present study of 2462 olecranon fractures, we found a larger share of women (65%) compared to the 55% reported in a previous single-center study describing 64 olecranon fractures [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref]. In addition, the median age (66 years) in our study was higher.
We confirmed the common assumption that an olecranon fracture is considered a fragility fracture [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref] because the leading injury mechanism is a same-level fall in an elderly woman. We found that the four assessed fracture types share roughly the same demographic patterns and injury mechanism but differ in the chosen treatment modality.
The prevalence of simple falls as the injury mechanism for olecranon fractures has been described elsewhere [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref]. Transportation accidents resulting in olecranon fractures occurred mainly in patients < 65 years and most fractures were due to a bicycle accident. Bicycle commuting is a popular way of getting to work and specific bicycle routes are common. High-energy injuries were more common in men, in younger patients and in more compex fractures in our study. These young men with high-energy injuries illustrate a subgroup of patients also seen in other locations with severe injuries sustained by high-energy injury mechanisms [bib_ref] Epidemiology and incidence of tibia fractures in the Swedish Fracture Register, Wennergren [/bib_ref] [bib_ref] Classification and treatment of lateral malleolar fractures-a single-center analysis of 439 ankle..., Rydberg [/bib_ref]. A more detailed investigation of how these accidents occur is warranted to counsel the population and decrease fracture risks. Distribution of sex, age at injury, injury mechanism and type of injury for the four fracture subgroups as well as overall fractures stratified for patients younger ≤ 65 years of age at
## Fracture classification
The distribution of fractures in our study differed compared to the single-center report from Scotland, [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref] where 74% of the fractures were simple central fractures and 81% central fractures. We found a larger share of comminuted central fractures (29%) compared to the study from Scotland. We also observed an almost threefold larger share of distal olecranon fractures in our material than the 6.2% in the Scottish study of 64 olecranon fractures [bib_ref] The epidemiology of fractures of the proximal ulna, Duckworth [/bib_ref]. The different settings of the studies might explain the discrepancy between the two studies: a Scandinavian context with nationwide coverage compared to an urban environment in Scotland. Moreover, the classification of fractures differed between the studies. In our study, a large number of observers (i.e. orthopedic surgeons) with a diverse range There are several different classification systems for olecranon fractures, e.g. Colton, Mayo, Schatzker and AO classications [bib_ref] Agreement of olecranon fractures before and after the exposure to four classification..., Benetton [/bib_ref]. They all have different rationales for grouping the fractures; displacement and fracture pattern [bib_ref] Fractures of the olecranon in adults: classification and management, Colton [/bib_ref] ; stability, displacement and comminution [bib_ref] Current concepts in the treatment of fractures of the radial head, the..., Morrey [/bib_ref] ; fracture morphology and biomechanical stability [bib_ref] Fractures of the olecranon, Schatzker [/bib_ref] ; or as part of elbow fractures with information about extra, partial or complete articular fractures. Further, there is newly updated version of the AO/OTA classification including more subgroups [bib_ref] Fracture and dislocation classification compendium-2018, Meinberg [/bib_ref].
## Treatment
Non-operative treatment was done in decreasing proportion with the increasing complexity of the fracture. A larger share of patients > 65 years were treated non-operatively for all fracture types, which concurs with previous findings of good outcomes after nonoperative treatment of stable fractures in patients with low functional demands [bib_ref] Nonoperative management of displaced olecranon fractures in low-demand elderly patients, Duckworth [/bib_ref] [bib_ref] Surgical and nonoperative management of olecranon fractures in the elderly: a systematic..., Chen [/bib_ref] [bib_ref] Prospective randomised trial of nonoperative versus operative management of olecranon fractures in..., Duckworth [/bib_ref] [bib_ref] Results of non-operative treatment of olecranon fracture in over 75-year-olds, Marot [/bib_ref]. In the more unstable fractures, TBW and plate fixation were equally commonly performed in elderly patients. In a review, implant type was not found to matter regarding risk of reoperation in elderly patients [bib_ref] Surgical and nonoperative management of olecranon fractures in the elderly: a systematic..., Chen [/bib_ref] and this seems not to influence the treatment choice. Plate fixation offers better biochemical properties as well as interfragmentary compression [bib_ref] Biomechanical comparison of interfragmentary compression in transverse fractures of the olecranon, Wilson [/bib_ref]. This seems to influence the choice of treatment in the younger patients with more unstable fractures, i.e. plate fixation was the preferred treatment for younger patients.
In the SFR, there is no information on type of non-operative treatment which can differ from early or direct range of motion to several weeks in cast. In line with previous recommendations, plate fixation was mainly used for unstable fractures [bib_ref] The treatment of olecranon fractures in adults, Powell [/bib_ref]. There is no detailed information on plate type or number of plates in the SFR. This is however a study on recent fractures in the locking plate era, so the majority of plates are presumably locking plates. Recently, some centers have started to use multiple small plates in select cases, sometimes as an adjunct to locking plates. This is not a widespread practice in Sweden. However, identifying and proper handling of intermediate fracture fragments seem to be more important than the implant itself [bib_ref] The pivotal role of the intermediate fragment in initial operative treatment of..., Von Rüden [/bib_ref].
## Seasonal variation
The incidence of all fractures varied little over the year. A slightly higher occurrence of olecranon fractures in both women and men was observed in the winter, which could be linked to icy conditions during this time of year.
# Strength and limitations
Our study has obvious limitations. We used the SFR that had an 85% coverage at the end of the study period to describe the epidemiology of olecranon fractures with the by largest number of olecranon fractures described in the literature.
Regrettably, we cannot identify the overall incidence of olecranon fractures given the stepwise introduction and the present completeness of the SFR in Sweden. Furthermore, the stepwise introduction effectively hindered a comparison with the base population and thus analyses on outcomes of interest.
Another limitation concerns the Mayo classification's questionable reliability [bib_ref] Agreement of olecranon fractures before and after the exposure to four classification..., Benetton [/bib_ref] [bib_ref] Comparison of tension band wiring and precontoured locking compression plate fixation in..., Schliemann [/bib_ref] , making comparisons difficult to establish [bib_ref] Classifications in brief: Mayo classification of olecranon fractures, Sullivan [/bib_ref]. The modified classification of the SFR allows four options and could have a favorable agreement among observers. All treating physicians and orthopedic surgeons performed the registrations and classifications in the SFR. Of note, validation studies in various segments found the classification systems to be as accurate as previous validation studies of fracture classification [bib_ref] Substantial accuracy of fracture classification in the Swedish Fracture Register: evaluation of..., Juto [/bib_ref] [bib_ref] High reliability in classification of tibia fractures in the Swedish Fracture Register, Wennergren [/bib_ref] [bib_ref] Validity of humerus fracture classification in the Swedish fracture register, Wennergren [/bib_ref]. Regarding treatment details, there is no information on type of nonoperative treatment, such as immobilization type or length. Neither is there information on type of and number of plates which would add to the study. However, with the current study, which was based on a nationwide register, we present relevant epidemiological data important for the surgeon (i.e. in preoperative counseling of patients) with information on general type of performed treatment. Reoperation rates in the SFR need to be verified by linking registers or by confirmation through a search in the medical files, which is outside the scope of this study.
# Conclusion
Isolated fractures of the olecranon occur mainly in elderly women after low-energy trauma. Non-operative treatment is common in simple fractures, whereas operative treatment is preferred in more complex fractures. More recently, there has been a shift towards plate fixation in the more unstable fracture patterns. Our results can help health care providers and clinicians understand isolated olecranon fractures.
Funding Open access funding provided by Uppsala University.
Data availability Data can be made available on request to the authors.
Code availability Not applicable.
# Declarations
Conflict of interest All authors report no conflicts of interest relevant to this study.
Ethical approval This study was approved by the Regional Ethical Committee in Uppsala (Dnr: 2015/509 and 2019/011006).
## Consent to participate not applicable.
## Consent for publication not applicable.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
[fig] Figure 1: Schematic description of the four fracture subtypes included in this study [/fig]
[fig] Figure 2: Distribution of age at injury for A all fractures and B the four respective subclassifications of experience from junior doctors to senior consultants performed the classification. [/fig]
[fig] Figure 3: Distribution of the injury mechanism for the four fracture subgroups [/fig]
[fig] Figure 4: Seasonal [/fig]
[table] Table 2: Distribution of the type of injury for patients aged 65 years and below and those older than 65 years, further stratified for sexes [/table]
[table] Table 3: Treatment choice for the four fracture subgroups and overall fractures stratified for patients ≤ 65 years of age at injury and those > 65 years [/table]
|
Sex Differences in the Impact of the Mediterranean Diet on LDL Particle Size Distribution and Oxidation
Sex differences have been previously highlighted in the cardioprotective effects of the Mediterranean diet (MedDiet). The objective of this study was to investigate whether sex differences also exist with regard to LDL particle size distribution and oxidation. Participants were 37 men and 32 premenopausal women (24-53 years) with slightly elevated LDL-C concentrations (3.4-4.9 mmol/L) or total cholesterol/HDL-C ≥5.0. Variables were measured before and after a four-week isoenergetic MedDiet. Sex differences were found in response to the MedDiet for the proportion of medium LDL (255-260 Å) (p for sex-by-time interaction = 0.01) and small, dense LDL (sdLDL; <255 Å) (trend; p for sex-by-time interaction = 0.06), men experiencing an increase in the proportion of medium LDL with a concomitant reduction in the proportion of sdLDL, while an opposite trend was observed in women. A sex difference was also noted for estimated cholesterol concentrations among sdLDL (p for sex-by-time interaction = 0.03), with only men experiencing a reduction in response to the MedDiet. The MedDiet marginally reduced oxidized LDL (oxLDL) concentrations (p = 0.07), with no sex difference. Results suggest that short-term consumption of the MedDiet leads to a favorable redistribution of LDL subclasses from smaller to larger LDL only in men. These results highlight the importance of considering sexOPEN ACCESSNutrients 2015, 7 3706 issues in cardiovascular benefits of the MedDiet.
# Introduction
Lowering LDL-C concentration is the primary target of therapy for the prevention of cardiovascular disease (CVD) [bib_ref] 2012 update of the canadian cardiovascular society guidelines for the diagnosis and..., Anderson [/bib_ref]. However, in addition to LDL-C concentrations, it has been shown that a more detailed analysis of LDL physico-chemical properties (e.g., size and oxidation) provides further insight into individual cardiovascular risk [bib_ref] European panel on low density lipoprotein (LDL) subclasses": A statement on the..., Mikhailidis [/bib_ref] [bib_ref] Circulating oxidized LDL: A biomarker and a pathogenic factor, Ishigaki [/bib_ref]. Individuals characterized by a predominance of small, dense LDL particles (sdLDL) are at increased risk of coronary heart disease compared to those with larger, buoyant LDL particles [bib_ref] Small dense low-density lipoprotein-cholesterol concentrations predict risk for coronary heart disease: The..., Hoogeveen [/bib_ref] [bib_ref] A prospective, population-based study of low density lipoprotein particle size as a..., Lamarche [/bib_ref]. Compared with large LDL, sdLDL possess a lower affinity for the LDL receptor and a longer half-life in plasma [bib_ref] Discrete subspecies of human low density lipoproteins are heterogeneous in their interaction..., Nigon [/bib_ref] , bind more tightly to arterial proteoglycans [bib_ref] Association of apo B lipoproteins with arterial proteoglycans: Pathological significance and molecular..., Camejo [/bib_ref] , penetrate the arterial subendothelial space more easily [bib_ref] Accumulation of lipoprotein fractions and subfractions in the arterial wall, determined in..., Bjornheden [/bib_ref] and are more susceptible to oxidation [bib_ref] Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein..., De Graaf [/bib_ref]. LDL oxidation is another process through which LDL contribute to atherosclerotic plaque formation by favoring endothelial dysfunction, the release of inflammatory cytokines and macrophage transformation into foam cells. Accordingly, oxidized LDL (oxLDL) concentrations have been identified as an important marker of atherosclerotic lesions [bib_ref] Circulating oxidized LDL: A biomarker and a pathogenic factor, Ishigaki [/bib_ref].
Sex disparities have been previously reported for LDL physico-chemical properties, men being characterized by a higher proportion of sdLDL and greater concentrations of oxLDL than premenopausal women [bib_ref] Relationship with serum lipids and LDL size, Vekic [/bib_ref] [bib_ref] Is the gender difference in LDL size explained by the metabolic complications..., Lemieux [/bib_ref] [bib_ref] Effects of gender and menopausal status on plasma lipoprotein subspecies and particle..., Li [/bib_ref] [bib_ref] Women have a larger and less atherogenic low density lipoprotein particle size..., Nikkila [/bib_ref] [bib_ref] Effect of gender and sex hormones on vascular oxidative stress, Miller [/bib_ref] [bib_ref] Is plasma oxidized low-density lipoprotein, measured with the widely used antibody 4E6,..., Wu [/bib_ref]. Given that the presence of both sdLDL and oxLDL are predictive of an increased cardiovascular risk, such sex differences could contribute in part to the higher risk of coronary heart disease found in men compared with premenopausal women.
Adopting the traditional Mediterranean diet (MedDiet) has been identified as a useful strategy in the prevention of cardiovascular events. The Prevención con Dieta Mediterrá nea (PREDIMED) study, consisting of a nutritional intervention among 7447 high-risk individuals, indicates that the adherence to an energy-unrestricted MedDiet supplemented with extra-virgin olive oil or mixed nuts for 4.8 years reduces by approximately 30% the incidence of myocardial infarction, stroke and cardiovascular death compared with a low-fat diet [bib_ref] Mediterranean diet for primary prevention of cardiovascular disease, Estruch [/bib_ref]. Different mechanisms of action have been proposed for the beneficial cardioprotective effects of the MedDiet, one largely documented being its LDL-C lowering effects [bib_ref] Scientific evidence of interventions using the mediterranean diet: A systematic review, Serra-Majem [/bib_ref]. Nevertheless, strong evidence suggests beneficial effects of the MedDiet beyond its impact on LDL-C and other traditional risk factors [bib_ref] Mediterranean diet and cardiovascular risk: Beyond traditional risk factors, Delgado-Lista [/bib_ref].
A shift toward larger LDL particles and a reduction in oxLDL concentrations have been previously observed with the consumption of the MedDiet in some [bib_ref] Effect of mediterranean diet with and without weight loss on apolipoprotein B100..., Richard [/bib_ref] [bib_ref] Mediterranean diet supplemented with nuts reduces waist circumference and shifts lipoprotein subfractions..., Damasceno [/bib_ref] [bib_ref] A nutritional intervention promoting the mediterranean food pattern is associated with a..., Lapointe [/bib_ref] [bib_ref] Effect of a traditional mediterranean diet on lipoprotein oxidation: A randomized controlled..., Fito [/bib_ref] , but not all studies [bib_ref] Effect of a nutritional intervention promoting the mediterranean food pattern on electrophoretic..., Goulet [/bib_ref] , suggesting that some factors, such as the characteristics of participants (e.g., sex, age, genetic predisposition, usual dietary intakes) may modulate these effects. In this sense, previous results suggest that men have greater cardiometabolic changes in response to the MedDiet than premenopausal women [bib_ref] Sex differences in the impact of the mediterranean diet on cardiovascular risk..., Bedard [/bib_ref] [bib_ref] Sex-related differences in the effects of the mediterranean diet on glucose and..., Bedard [/bib_ref] [bib_ref] Effects of the traditional mediterranean diet on adiponectin and leptin concentrations in..., Bedard [/bib_ref]. However, no study has examined whether sex differences exist in the effects of this food pattern on LDL physico-chemical properties. One could speculate that, because of the higher proportion of sdLDL habitually found in men compared with premenopausal women [bib_ref] Relationship with serum lipids and LDL size, Vekic [/bib_ref] [bib_ref] Is the gender difference in LDL size explained by the metabolic complications..., Lemieux [/bib_ref] [bib_ref] Effects of gender and menopausal status on plasma lipoprotein subspecies and particle..., Li [/bib_ref] [bib_ref] Women have a larger and less atherogenic low density lipoprotein particle size..., Nikkila [/bib_ref] , men may experience a more important shift toward larger LDL particles than women in response to the MedDiet.
Consequently, since larger, less dense LDL particles are less susceptible to oxidation than sdLDL [bib_ref] Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein..., De Graaf [/bib_ref] , men may have greater reduction in oxLDL concentrations than women. As estrogens exert beneficial effects on LDL size and oxidation [bib_ref] Effect of gender and sex hormones on vascular oxidative stress, Miller [/bib_ref] [bib_ref] The role of small, dense low density lipoprotein (LDL): A new look, Packard [/bib_ref] [bib_ref] The effects of estrogen administration on plasma lipoprotein metabolism in premenopausal females, Schaefer [/bib_ref] [bib_ref] The influence of sex steroids on adipose tissue growth and function. Horm, Law [/bib_ref] [bib_ref] Estrogen and estrogen receptors in cardiovascular oxidative stress, Arias-Loza [/bib_ref] and the MedDiet has been previously shown to reduce estrogen concentrations in women [bib_ref] A traditional mediterranean diet decreases endogenous estrogens in healthy postmenopausal women, Carruba [/bib_ref] , these facts further support the hypothesis that men may benefit more than premenopausal women from this food pattern.
The aim of the present study was to verify whether the impact of the MedDiet on LDL size distribution, as well as on oxLDL concentrations differs between men and women. Results presented in this paper suggest that adhering to the MedDiet, in addition to a clinically-relevant reduction in LDL-C concentrations, also has additional positive effects on LDL particle size phenotype, leading to a favorable redistribution from smaller to larger LDL in men, but not in women. This study provides new and useful clinical information in order to improve our understanding of the variability in the response to the MedDiet and, ultimately, to further individualize dietary recommendations in the prevention of cardiovascular disease.
## Experimental section
## Subjects
Thirty-eight men and 32 premenopausal women took part to this study. However, one man was excluded from analyses, because of a lack of dietary compliance due to illness during several days just before the end of the controlled MedDiet phase. Analyses were therefore conducted in a sample of 37 men and 32 premenopausal women, aged between 24 and 53 years, who participated in a study that was initially designed to directly document sex differences in the impact of a MedDiet on LDL-C concentrations [bib_ref] Sex differences in the impact of the mediterranean diet on cardiovascular risk..., Bedard [/bib_ref]. The main inclusion criteria were to have slightly elevated LDL-C concentrations (between 3.4 and 4.9 mmol/L) or a total cholesterol to HDL-C ratio ≥5.0, as well as at least one of the four following CVD risk factors: waist circumference >94 cm in men and >80 cm in women; TG ≥1.7 mmol/L; fasting glycemia between 6.1 and 6.9 mmol/L and/or blood pressure levels ≥130/85 mm Hg. No participant had a history of cardiovascular events, and none were taking medication that could affect the dependent variables under study (e.g., lipid-lowering medication). Smokers and pregnant women or those using systemic hormonal contraceptives were excluded from this study. The present study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving human subjects were approved by the Laval University Ethics Committee (#2007-180, 4 October 2007). This clinical trial was registered at www.clinicaltrials.gov as NCT01293344.
Before the intervention (i.e., at screening), no difference between men and women was found for age (mean (SD): men 42.5 (7.3) years; women 41.1 (7.4) years, p for sex difference = 0.45) and BMI (mean (SD): men 28.8 (3.1) kg/m 2 ; women 29.2 (5.6) kg/m 2 , p for sex difference = 0.92). However, men had higher mean values than women for body weight (mean (SD): men 90.7 (13.7) kg; women 77.4 (14.7) kg, p for sex difference <0.001) and waist circumference (mean (SD): men 102.1 (9.9) cm; women 94.3 (10.3) cm, p for sex difference = 0.001).
Power analyses for repeated measures and within-between interactions showed that a total sample size of n = 69 is sufficient to detect significant differences in all outcomes measured with a small effect size estimate (Cohen's d of 0.20 [bib_ref] Confidence intervals for effect sizes in analysis of variance, Bird [/bib_ref] and with an α = 0.05 and a power of 0.8 (G*Power Version 3.0.10, Franz Faul, Universitä t Kiel, Germany).
## Study design
The study protocol consisted of a 4-week run-in period, immediately followed by a 4-week fully-controlled MedDiet phase [bib_ref] Sex differences in the impact of the mediterranean diet on cardiovascular risk..., Bedard [/bib_ref]. During the run-in period, subjects received personalized recommendations by a registered dietitian in order to follow the healthy recommendations of Canada's Food Guide. Briefly, Canada's Food Guide is an educational tool that promotes healthy eating for Canadians in order to reduce the risk of chronic diseases and to achieve overall health. It indicates the recommended number of food guide servings per day for each of the four food groups (vegetables and fruits, grain products, milk and alternatives and meat and alternatives) according to the age and sex of individuals. In addition, more specific recommendations are provided for each food groups (e.g., eat at least one dark green and one orange vegetable each day, make at least half of your grain products whole grain each day, select lower fat milk and alternatives, have meat alternatives, such as beans, lentils and tofu, often and eat at least two servings of fish each week). The 4-week run-in period allowed comparing the effects of the MedDiet between men and women having similar baseline dietary intakes, as reported previously [bib_ref] Sex differences in the impact of the mediterranean diet on cardiovascular risk..., Bedard [/bib_ref]. Moreover, the concordance of men's and women's diet with the traditional MedDiet was similar at the end of the run-in period, as suggested by a Mediterranean score (MedScore; from 0-44 points: 24.8 (5.9) points in men and 24.6 (4.4) points in women, p for sex difference = 0.87) [bib_ref] Effect of a nutritional intervention promoting the mediterranean food pattern on plasma..., Goulet [/bib_ref]. This MedScore has been previously shown to be a valid indicator of MedDiet adherence [bib_ref] Effect of a nutritional intervention promoting the mediterranean food pattern on plasma..., Goulet [/bib_ref] [bib_ref] Gender differences in the long-term effects of a nutritional intervention program promoting..., Leblanc [/bib_ref]. A MedScore of forty-four would imply a food pattern that is perfectly concordant with the traditional MedDiet. No change in body weight was found during the run-in period (+0.02 (0.19) kg in men and −0.01 (0.16) kg in women, respectively p = 0.93 and p = 0.84).
During the 4-week fully-controlled feeding phase, subjects consumed an experimental MedDiet formulated to be concordant with the characteristics of the traditional MedDiet [bib_ref] Mediterranean diet pyramid: A cultural model for healthy eating, Willett [/bib_ref]. More precisely, the experimental MedDiet included an abundance of fruits, vegetables, whole grain cereals, nuts and legumes, moderate amounts of fish, poultry, eggs and low-fat dairy products and low amounts of red meat and sweets. Olive oil was the main source of fat, and wine accompanied meals with moderation. The percentages of energy derived from lipids, carbohydrates, proteins and alcohol were respectively 32% (7.2% SFA, 17.9% MUFA and 4.6% PUFA), 46%, 17% and 5%. Details about the composition of the MedDiet have been already reported [bib_ref] Sex differences in the impact of the mediterranean diet on cardiovascular risk..., Bedard [/bib_ref].
A 4-week controlled phase has been demonstrated as sufficient to induce significant changes in LDL particle size features [bib_ref] Dietary mono-and polyunsaturated fatty acids similarly affect LDL size in healthy men..., Kratz [/bib_ref] [bib_ref] Combined effects of a dietary portfolio of plant sterols, vegetable protein, viscous..., Lamarche [/bib_ref] and oxLDL concentrations [bib_ref] Pistachios increase serum antioxidants and lower serum oxidized-LDL in hypercholesterolemic adults, Kay [/bib_ref] [bib_ref] Effects of moderate sicilian red wine consumption on inflammatory biomarkers of atherosclerosis, Avellone [/bib_ref]. All foods for the MedDiet phase were prepared in a standardized manner in the Clinical Investigation Unit (CIU). Subjects were instructed to consume only the foods and beverages provided to them, which corresponded to 100% of their estimated energy needs. Energy needs were established by averaging energy requirements estimated by a validated food frequency questionnaire (FFQ) [bib_ref] Validity and reproducibility of an interviewer-administered food frequency questionnaire for healthy french-canadian..., Goulet [/bib_ref] administrated at the beginning of the run-in period and energy needs as determined by the Harris-Benedict formula. Participant's body weight was monitored daily before the consumption of the lunch at the CIU, and energy intake was increased or decreased by 250 kcal/day if a subject lost or gained greater than 1 kg and maintained that body weight for at least 3 days. Participants completed a daily checklist confirming the consumption of provided foods and beverages and, if needed, the amount of foods not consumed. Participants were instructed to maintain their usual physical activity level. However, in order to verify whether participants followed this instruction, daily energy expenditure from physical activity participation was evaluated using a validated 3-day activity diary record developed by Bouchard et al. [bib_ref] A method to assess energy expenditure in children and adults, Bouchard [/bib_ref] administrated during the fourth week of both the run-in period and the MedDiet phase. Since some studies have suggested that fluctuations in female hormones influence the lipid-lipoprotein profile [bib_ref] Cyclic changes in lipoprotein and apolipoprotein levels during the menstrual cycle in..., Muesing [/bib_ref] , women's feeding was shortened or prolonged if needed in order to be able to carry out all tests in the early follicular phase of their menstrual cycle (from the third to the ninth day of the menstrual cycle; mean duration of the feeding period in women (SD): 28.8 (4.3) days).
## Laboratory analyses
Blood samples were collected from an antecubital vein into vacutainer tubes after a 12-h overnight fast. Total plasma cholesterol, TG and HDL-C concentrations were measured using commercial reagents on a Modular P chemistry analyzer (Roche Diagnostics, Mannheim, Germany). Apo B was measured by immunoturbidimetry (Roche Diagnostics, Mannheim, Germany). LDL-C was obtained by the equation of Friedewald [bib_ref] Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use..., Friedewald [/bib_ref]. Plasma apo A-1 and apo A-2 concentrations were measured by immunonephelometry. For the measurements of LDL particle size features and oxLDL concentrations, analyses were performed using plasma stored at −80 °C. Non-denaturing 2%-16% polyacrylamide gradient gel electrophoreses were used to characterize LDL particle size distribution, as previously described [bib_ref] Comparison of various electrophoretic characteristics of ldl particles and their relationship to..., St Pierre [/bib_ref]. LDL particle size was computed on the basis of the relative migration of plasma standards of known diameter [bib_ref] The dense LDL phenotype. Association with plasma lipoprotein levels, visceral obesity, and..., Tchernof [/bib_ref]. The LDL peak particle size was computed as the estimated diameter for the major peak of each scan. An integrated (or mean) LDL size, corresponding to the weighed mean size of all LDL subclasses in each individual, was also determined. As revealed by the analysis of pooled plasma standards, measurements of LDL peak particle size and LDL integrated particle size were highly reproducible, considering an inter-assay coefficient of variation of <2%. The relative proportion of sdLDL, characterized by a diameter <255 Å (LDL<255Å), was obtained by computing the relative area of the densitometric scan <255 Å. The absolute concentration of cholesterol in sdLDL particles was estimated by multiplying the total plasma LDL-C concentrations by the relative proportion of LDL<255Å. Similar approaches were used to estimate the relative proportion of medium and large LDL particles and their specific estimated cholesterol concentration, using respectively a diameter between 255 and 260 Å (LDL255-260Å) and >260 Å (LDL>260Å). LDL subclass Pattern A was characterized by an LDL peak particle size ≥255 Å, whereas LDL subclass Pattern B by a LDL peak particle size <255 Å. The measurements of the proportion of medium LDL and large LDL had a coefficient of variation of 12% and 9.3% respectively. oxLDL concentrations were measured using a commercial enzyme-linked immune-sorbent assay (ELISA) kit (Alpco, Salem, NH), with an intra-assay coefficient of variation of 3.9%-5.7% and inter-assay coefficient of variation of 9%-11%.
## Statistical analyses
Statistical analyses were performed using SAS statistical package Version 9.4 (SAS Institute Inc., Cary, NC, USA). All statistical tests were two-sided. p ≤ 0.05 was considered as significant. Data were collected before (i.e., immediately after the run-in period, referred as baseline values) and after the controlled MedDiet phase. Differences in baseline characteristics between men and women were assessed using Student's t-test. Mixed procedures for repeated measurements were used to assess the main effects of time, sex and sex-by-time interaction on LDL particle size features and oxLDL concentrations. When a significant main effect was found, Tukey-Kramer adjusted p-values were used to identify the precise location of differences. Associations between variables were assessed by Pearson's correlation analyses. One woman was excluded from analyses of oxLDL concentrations due to a baseline extreme value (i.e., 3027.8 ng/mL vs. group's mean (SD) of 166.4 (200.6) ng/mL).
Although the MedDiet phase aimed at being isoenergetic, both men and women experienced a small, but significant body weight change (mean (SD): −1.19 (1.23) kg in men, p < 0.001 and −0.55 (0.98) kg in women, p = 0.01). Given that changes in body weight may influence LDL physico-chemical properties (both size and oxidation) [bib_ref] Circulating oxidized LDL: A biomarker and a pathogenic factor, Ishigaki [/bib_ref] [bib_ref] Obesity, weight loss and conditional cardiovascular risk factors, Tzotzas [/bib_ref] , all analyses presented here were adjusted for this small change in body weight. Waist circumference did not statistically change during the MedDiet phase in both men and women (mean (SD): −0.29 (2.68) cm in men, p = 0.56; −0.80 (2.56) cm in women, p = 0.09).
# Results
Even if there were no difference between men and women for LDL-C concentrations before the MedDiet (i.e., immediately after the run-in period; p = 0.36), some sex differences were observed in LDL physico-chemical properties [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref]. In fact, men had a lower proportion of medium LDL(255-260Å) (p = 0.02) and a higher proportion of sdLDL(<255Å) (p = 0.01) than women. Moreover, men had higher estimated cholesterol concentrations among sdLDL(<255Å) than women (p = 0.01). Finally, men were characterized by smaller LDL peak particle size (p = 0.04) and LDL integrated size (p = 0.04) than women. No difference was found between men and women for all of the other variables related to LDL particle size features (p > 0.08). Men and women had similar oxLDL concentrations prior to the controlled MedDiet intervention (p = 0.86).
## Plasma lipids and lipoproteins
As reported previously [bib_ref] Sex differences in the impact of the mediterranean diet on cardiovascular risk..., Bedard [/bib_ref] , reductions in total cholesterol, LDL-C, HDL-C, total cholesterol to HDL-C ratio and apo B were observed in response to the MedDiet, and no difference was found between men and women (p for sex-by-time interaction ≥0.16; [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref]. More precisely, reductions in LDL-C concentrations of 10.4% in men and 7.3% in women were noted (respectively, p = 0.003 and p = 0.04). TG, triglycerides; LDL-C, low-density lipoprotein-cholesterol; HDL-C, high-density lipoprotein-cholesterol; Apo, apolipoprotein; oxLDL, oxidized LDL. a All analyses are adjusted for the body weight change during the MedDiet phase; b baseline values represent those collected after the run-in period, immediately before the 4-week MedDiet; c these data have been previously reported [bib_ref] Sex-related differences in the effects of the mediterranean diet on glucose and..., Bedard [/bib_ref] ; d since these variables were not normally distributed, a transformation was performed in order to obtain a normal distribution. TG, HDL-C and oxLDL were log-transformed; LDL peak particle diameter, LDL integrated size and small LDL-C <255Å were inversed transformed; and large LDL >260Å was square transformed.
## Ldl size distribution
Sex differences were observed in changes in LDL size distribution in response to the MedDiet and more precisely for the proportion of medium LDL(255-260Å) (p for sex-by-time interaction = 0.01) and sdLDL(<255Å) (trend; p for sex-by-time interaction = 0.06) [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref]. Specifically, men experienced an increase in the proportion of medium LDL with a concomitant non-significant decrease in the proportion of sdLDL (respectively, p = 0.03 and p = 0.50), while an opposite non-significant trend was observed in women (respectively, p = 0.74 and p = 0.54). No change was observed for the proportion of large LDL(>260Å) in both men and women [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref]. These results suggest that short-term consumption of the MedDiet leads to a favorable redistribution of LDL subclasses from smaller to larger LDL only in men.
A similar LDL size distribution was found in men and in women after the MedDiet (p for sex difference for each proportion of LDL subclass ≥0.14), suggesting that sex differences at baseline were no longer present after the short-term consumption of the MedDiet.
## Estimated cholesterol concentration among each ldl subclass
A sex difference was noted for cholesterol concentrations among sdLDL(<255Å) (p for sex-by-time interaction = 0.03), with only men experiencing a reduction in response to the MedDiet (p < 0.001 in men and p = 0.88 in women; [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref]. In men, the reduction in cholesterol concentration among sdLDL correlated with concomitant reductions in TG (r = 0.38, p = 0.02), LDL-C (r = 0.41, p = 0.01) and apo B (r = 0.57, p < 0.001). The sex difference in cholesterol concentrations among sdLDL observed at baseline was no longer present after the short-term consumption of the MedDiet (p for sex difference ≥0.11).
Significant reductions in cholesterol concentration among large LDL(>260Å) and a tendency for a decrease in cholesterol concentration among medium LDL(255-260Å) were found, and no sex differences were observed for these variables [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref].
## Ldl peak and integrated (mean) size
Consumption of the MedDiet did not affect LDL peak particle size or the LDL integrated size in both men and women [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref].
However, a three-way sex-by-time-by-LDL subclass pattern (A or B) interaction was found for LDL peak particle size (p for interaction = 0.01). Subgroup analysis indicated that, among men and women with LDL Pattern A at baseline (LDL peak particle size ≥255 Å), only men experienced a decrease in LDL peak particle size in response to the MedDiet (p = 0.03 for men and p = 0.54 for women; p for time effect = 0.004; p for sex-by-time interaction = 0.12) . Moreover, among those with LDL Pattern B at baseline (LDL peak particle size <255 Å), LDL peak particle size was increased in men only (p = 0.003 for men and p = 0.90 for women; p for time effect = 0.007; p for sex-by-time interaction = 0.08). Among each subgroup, no difference between men and women was found for baseline LDL peak particle size (p for sex difference at baseline > 0.99 for Pattern A subgroup and 0.94 for Pattern B subgroup). These results suggest that LDL subclass pattern at baseline influences the LDL peak particle size response to the MedDiet, but only in men. No three-way interaction was found for the LDL integrated size. Figure 1. LDL peak particle size at baseline (i.e., immediately before the MedDiet) and after the four-week MedDiet in men and women according to their initial LDL subclass pattern (Pattern A: LDL peak particle diameter ≥255 Å, men n = 8 and women n = 14, p for sex-by-time interaction = 0.12; Pattern B: LDL peak particle diameter <255 Å, men n = 29 and women n = 18, p for sex-by-time interaction = 0.08). Data are means (SEM). * Different from baseline in men, p < 0.05 by mixed procedure followed by the Tukey-Kramer test.
Since TG concentrations are the main determinant of LDL size [bib_ref] European panel on low density lipoprotein (LDL) subclasses": A statement on the..., Mikhailidis [/bib_ref] and TG reduction in men was twice the reduction in women in response to the MedDiet (respectively −14.6% and −7.7%; [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref] , we thereafter adjusted the analyses for changes in TG concentrations. Sex differences in changes in LDL particle size features in response to the MedDiet were still observed after these adjustments (not shown).
Adjustments for waist circumference at baseline did not influence the results obtained (not shown), suggesting that differences between men and women in abdominal obesity do not explain sex differences observed in LDL size response to the MedDiet.
## Oxidized ldl
A trend for a reduction in oxLDL concentrations was found in response to the MedDiet, and no difference was observed between men and women (p for time effect = 0.07; p for sex-by-time interaction = 0.85; [fig_ref] Table 1: Effects of the four-week Mediterranean diet [/fig_ref].
Diet-induced changes in oxLDL concentrations were not associated with oxLDL concentrations at baseline in either men or women (respectively r = −0.06, p = 0.71 and r = −0.16, p = 0.40). There was no association between changes in oxLDL concentrations and changes in LDL-C concentrations (respectively r = 0.20, p = 0.25 for men and r = 0.13, p = 0.50 for women) and changes in cholesterol concentration among sdLDL(<255Å) (r = 0.11, p = 0.54 for men and r = 0.08, p = 0.67 for women).
Although participants were instructed to maintain their usual physical activity level, a similar decrease in daily energy expenditure from physical activity participation was observed during the MedDiet phase in men and women (−27.8% in men and −27.6% in women, p for sex-by-time interaction = 0.25). Adjustment for changes in physical activity participation did not influence the results obtained (not shown).
# Discussion
Elevated LDL-C concentration has been established as a major risk factor for CVD [bib_ref] 2012 update of the canadian cardiovascular society guidelines for the diagnosis and..., Anderson [/bib_ref]. However, a more detailed analysis of LDL physico-chemical properties is highly relevant given the important atherogenicity of sdLDL particles and of oxLDL [bib_ref] European panel on low density lipoprotein (LDL) subclasses": A statement on the..., Mikhailidis [/bib_ref] [bib_ref] Circulating oxidized LDL: A biomarker and a pathogenic factor, Ishigaki [/bib_ref]. Accordingly, results from the present study highlight sex-specific cardiovascular benefits of the MedDiet, which would not have been observed if only LDL-C concentrations had been measured. Indeed, despite the fact that men and women have similar reductions in LDL-C concentrations in response to the MedDiet, different responses in LDL particle size features to this food pattern are observed. More precisely, men experienced an increase in the proportion of medium LDL particles with a concomitant reduction in the proportion of sdLDL, while an opposite non-significant trend was observed in women. Moreover, a reduction in cholesterol concentrations among sdLDL with MedDiet was noted in men, but not in women. Finally, results showed that the more deteriorated LDL particle size features found in men compared with premenopausal women at baseline were no longer present after the short-term consumption of the MedDiet.
Even if a favorable LDL size redistribution and a reduction in cholesterol concentration among sdLDL were observed in men, one might question the clinical relevance of these changes, since the MedDiet had no global impact on LDL peak particle size, a widely-used surrogate marker of sdLDL. However, a prospective study including 2,034 middle-aged men suggested that LDL peak particle size is a weak predictor of ischemic heart disease risk compared with the proportion and the cholesterol content of sdLDL, which were identified as very powerful and independent risk predictors, even after adjustments for non-lipid and other lipid CVD risk factors, including LDL-C, HDL-C, TG and lipoprotein(a) (Lp(a)) concentrations [bib_ref] Comparison of various electrophoretic characteristics of ldl particles and their relationship to..., St Pierre [/bib_ref]. Therefore, one mechanism behind the reduction in cardiovascular events with the consumption of the MedDiet could be a reduction of sdLDL in dyslipidemic men.
As previously found in the literature, our results highlight that, compared with premenopausal women, age-matched men have a greater proportion of sdLDL [bib_ref] Relationship with serum lipids and LDL size, Vekic [/bib_ref] [bib_ref] Is the gender difference in LDL size explained by the metabolic complications..., Lemieux [/bib_ref] [bib_ref] Effects of gender and menopausal status on plasma lipoprotein subspecies and particle..., Li [/bib_ref] [bib_ref] Women have a larger and less atherogenic low density lipoprotein particle size..., Nikkila [/bib_ref]. Given their greater proportion of sdLDL at baseline, men had more room for improvement, which may explain at least partly why they benefited more from the MedDiet than premenopausal women. Indeed, in previous studies, those including subjects with metabolic syndrome, which is a cluster of metabolic abnormalities characterized by a predominance of sdLDL, found increases in LDL size and/or a favorable redistribution of cholesterol among each LDL subclass with the MedDiet [bib_ref] Effect of mediterranean diet with and without weight loss on apolipoprotein B100..., Richard [/bib_ref] [bib_ref] Mediterranean diet supplemented with nuts reduces waist circumference and shifts lipoprotein subfractions..., Damasceno [/bib_ref] , while one studying the impact of the MedDiet in healthy individuals has observed no effect [bib_ref] Effect of a nutritional intervention promoting the mediterranean food pattern on electrophoretic..., Goulet [/bib_ref]. In line with these previous studies, strong negative correlations were found in the present study between the proportion and the cholesterol content of sdLDL at baseline and changes in these specific variables in response to the MedDiet, meaning that individuals who were characterized by a higher proportion and cholesterol content of sdLDL benefited more from the MedDiet (respectively, r < −0.50 and r < −0.61, p < 0.001). Additional analyses also showed that, when men and women were individually matched for the proportion of sdLDL at baseline (23 men and 23 women), differences between men and women in changes in cholesterol concentration among sdLDL were considerably smaller than the one found with the whole sample (−10.4% in men and −7.9% in women, p for sex-by-time interaction = 0.68 in subjects paired for the proportion of sdLDL at baseline; vs. −14.3% in men and −3.5% in women when the whole sample was considered).
In addition to the difference between men and women in LDL particle size features at baseline, other factors related to sex seem to influence the impact of the MedDiet on LDL size. In fact, in the present study, subjects characterized by the LDL Pattern B phenotype (i.e., predominance of sdLDL) experienced a favorable increase in LDL peak particle size, whereas a reduction was found in those characterized by the LDL Pattern A phenotype (i.e., predominance of large LDL particles). However, this pattern of changes was observed only in men, while women experienced no change, regardless of their baseline LDL peak particle size phenotype. Since these sub-analyses are based on small numbers of subjects in each group, they need to be confirmed.
Although not significant from a statistical perspective, men experienced nearly a two-fold greater reduction in TG concentrations compared with women (14.6% vs. 7.7%), and such a difference could be clinically significant. High concentrations of TG favor the production of large, TG-rich, very low density lipoprotein (VLDL), a precursor of TG-rich LDL. In turn, lipolysis of TG in TG-enriched LDL by hepatic lipase leads to the formation of sdLDL [bib_ref] European panel on low density lipoprotein (LDL) subclasses": A statement on the..., Mikhailidis [/bib_ref]. One could therefore assume that the greater reduction in the concentrations of sdLDL found in men compared with women in response to the MedDiet could also be explained by this greater reduction in TG concentrations in men. In the present study, a strong association between changes in sdLDL and concomitant changes in TG concentrations was found in men. However, sex differences in LDL particle size feature responses to the MedDiet were still observed after adjustments for changes in TG concentrations, suggesting that sex differences observed in the present study were not due only to a greater reduction in TG concentrations in men compared with women.
The reduction in oxLDL concentrations was not associated with decreases in LDL-C concentrations in response to the MedDiet in both men and women, suggesting that oxLDL reduction does not simply reflect the LDL-lowering effects of the MedDiet, but is likely to be due to the antioxidant properties of this food pattern. Since men reduced their proportion and cholesterol content of sdLDL more than women in response to the MedDiet and sdLDL is more susceptible to oxidation compared with larger particles [bib_ref] Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein..., De Graaf [/bib_ref] , it could be expected that men may have reduced their oxLDL concentrations to a greater extent compared with women. However, this was not the case. In fact, no sex difference was found in response to the MedDiet, with a modest reduction in oxLDL concentrations observed in both men and women. One possible explanation is that the consumption of a diet rich in oleic acid, the more abundant fatty acid in the olive oil, may reduce the susceptibility of sdLDL to oxidation [bib_ref] Effects of linoleate-enriched and oleate-enriched diets in combination with alpha-tocopherol on the..., Reaven [/bib_ref] , leading to a reduction in oxLDL, which is independent of the one found in sdLDL. This disassociation between changes in sdLDL and changes in oxLDL concentrations was observed in both men and women. Moreover, the MedDiet is rich in antioxidants, such as carotenoids, tocopherols and phenolic compounds, mainly contained in vegetables, fruits, olive oil and red wine [bib_ref] The antioxidative protecting role of the mediterranean diet, Hadziabdic [/bib_ref] , which may provide an important protection for sdLDL to oxidation. Therefore, these results suggest that the high oleic acid content of the MedDiet, along with its important antioxidant properties, may overshadow the link between sdLDL and LDL oxidation.
One of the strengths of our study is a design that included a highly-controlled dietary phase, which permitted precisely investigating sex differences in response to the MedDiet with a maximum control of possible confounding variables. Moreover, this study included a large number of subjects for a strictly controlled feeding study. However, results should not be generalized, and additional studies are needed. The duration of the controlled phase was appropriate, considering that previous studies of equal or shorter lengths showed significant effects of diet on LDL particle size [bib_ref] Dietary mono-and polyunsaturated fatty acids similarly affect LDL size in healthy men..., Kratz [/bib_ref] [bib_ref] Combined effects of a dietary portfolio of plant sterols, vegetable protein, viscous..., Lamarche [/bib_ref] and oxLDL concentrations [bib_ref] Pistachios increase serum antioxidants and lower serum oxidized-LDL in hypercholesterolemic adults, Kay [/bib_ref] [bib_ref] Effects of moderate sicilian red wine consumption on inflammatory biomarkers of atherosclerosis, Avellone [/bib_ref]. Among the limitations, the study's 'single strand before and after' design does not allow comparisons to a control diet, and therefore, non-specific treatment effects that are not attributable to the MedDiet cannot be ruled out. However, we consider that the absence of a control diet is not a major limitation, since the main objective of this study was to directly compare men's and women's response to the MedDiet.
# Conclusions
In conclusion, data from this strictly-controlled feeding study indicated for the first time the existence of sex differences in the response of the LDL particle size features to the MedDiet in dyslipidemic men and premenopausal women. Our results suggest that adhering to the MedDiet, in addition to a clinically-relevant reduction in LDL-C concentrations (−10.4% in men and −7.3% in women), also has additional positive effects on LDL particle size phenotype, leading to a favorable redistribution from smaller to larger LDL in men, but not in women. Moreover, men, but not women, with smaller LDL peak particle size at baseline (<255 Å) increase LDL peak particle size in response to the MedDiet. Since sdLDL has been suggested as a strong predictor of the progression of atherosclerosis and CVD events [bib_ref] Small dense low-density lipoprotein-cholesterol concentrations predict risk for coronary heart disease: The..., Hoogeveen [/bib_ref] [bib_ref] A prospective, population-based study of low density lipoprotein particle size as a..., Lamarche [/bib_ref] , such findings have implications, both for improved understanding of sex-specific mechanisms behind the beneficial cardiovascular effects of the MedDiet and for the clinical management of dyslipidemic men.
[table] Table 1: Effects of the four-week Mediterranean diet (MedDiet) on the lipid-lipoprotein profile and LDL physical properties in men and women.Table 1. Cont. [/table]
|
Performance Differences of Hexavalent Chromium Adsorbents Caused by Graphene Oxide Drying Process
In this study, the influence of drying conditions on amine (−nH 3 ) functionalization of graphene oxide (GO) was evaluated, and the hexavalent chromium (Cr(VI)) adsorption efficiency of the prepared materials was compared. 3-[2-(2-aminoehtylamino) ethylamino]propyl-trimethoxysilane (3N) was used for amine functionalization. The synthesized materials were analyzed by SEM, BET, TGA, XPS, and EA. TGA results showed that the solution-GO (SGO) was functionalized by more 3N molecules than freezedried GO (FDGO) and oven-dried GO (ODGO). Additionally, XPS analysis also showed that the ratio of N/C and Si/C was relatively high in SGO than FDGO and ODGO. The maximum adsorption capacity of SGO, FDGO, and ODGO for Cr(VI) was 258.48, 212.46, and 173.45 mg g −1 , respectively. These results indicate that it is better to use SGO without drying processes for efficient amine functionalization and Cr(VI) removal. However, when the drying process is required, freeze-drying is better than oven-drying.The active development in various industrial sectors, such as textile, metal finishing, leather tanning, stainless steel, and electroplating 1-7 contribute to the discharge of contaminated wastewater containing inorganic (heavy metals, phosphates, nitrates, sulfates, etc.), organic (phenols, dyes, pesticides, pharmaceutical compounds, etc.), and biological (bacteria, viruses, etc.) compounds. Among these, heavy metals such as chromium (Cr), arsenic (As), copper (Cu), lead (Pb), cadmium (Cd), and mercury (Hg) are typical toxic substances that exist as cations and anions in water 8 . Chromium exists in two stable oxidation states, which are the hexavalent chromium (Cr(VI)) and trivalent chromium (Cr(III)) variants 9 . While Cr(III) is minimally toxic, Cr(VI) is highly toxic and causes significant environmental damage 10 . Further, Cr(VI) can cause carcinogenesis and mutations in humans. For these reasons, the World Health Organization (WHO) regulates the concentration of Cr(VI) in drinking water to less than 50 ng- L −1 11 . To comply with this rigorous and necessary regulation, the application of water treatment techniques that can remove low concentrations of Cr(VI) is required. Various techniques, such as, redox, ion-exchange, adsorption, membrane filtration, and the like 12-16 are applied according to the purpose. However, the removal of low concentrations of Cr(VI) with most of these wastewater treatment systems is highly challenging. Among these approaches, adsorption is more effective for Cr(VI) removal since adsorbents are eco-friendly and not generating by-products 17 . Therefore, we conducted further research on advancing adsorption technology for the removal of Cr(VI) from wastewater. For the first time, in this study, we developed an adsorbent based on graphene oxide (GO) and evaluated its Cr(VI)-adsorption performance.GO has been used in various fields such as electronic, biological, and physical applications 18 owing to its several advantages, such as straightforward synthesis, good solution stability, and the large surface-area-to-volume ratio 19-21 and small weight-to-volume ratio. Particularly, the rich oxygen-containing surface functional groups
(hydroxyl, epoxy, carbonyl and carboxyl groups) of GO facilitate the facile introduction of desired functional groups, which enable adsorbent applications for removal of harmful elements. Due to these properties, many GO composites have been studied for the adsorption of heavy metals . Herein, when focusing on the GO state prior to the functionalization process, there were various GO conditions, including solution, freeze-drying, oven-drying, etc. and some studies had no mention of it.
Recently, we showed that the adsorbents synthesized by chemically bonding GO and amino-silanes have excellent performance in chromium adsorption (260.74 mg g −1 )The minor yet remarkable difference between these two studies, which many researchers overlooked, is the use of different approaches for the preparation of GO. Whileused the oven-drying of GO at 100 °C (oven-dried GO (ODGO)) for preparation of GO,employed the solution form of GO (Solution-GO, SGO), which revealed the importance of adopting the appropriate drying process. As far as we know, there is no study suggesting suitable process of GO for surface functionalization so far. Therefore, with an aim to find improved and effective processes for GO surface functionalization, we compared the differences in the Cr(VI) adsorption capacities caused by the variations in the conditions employed for the preparation of GO surfaces before the functionalization. We are certain that the results of this study will help researchers working with GO for the development of various applications.
# Results and discussion
Cr(VI) adsorption isotherm on GO + 3N adsorbents. The recorded adsorption isotherms describe the physicochemical adsorption resulting from the interaction between Cr(VI) and the adsorbent surfaces. The obtained equilibrium data was fitted to the Langmuir and Freundlich isotherm models. The Langmuir equilibrium adsorption isotherm describes the adsorption reaction of the monolayer on homogenous surfaces as a function of the partial pressure at constant temperature. The equation for the Langmuir adsorption isotherm is shown in Eq. 1.
[formula] Q K Q C K C 1 (1) e L m e L e = + [/formula]
where Q e is the equilibrium adsorption capacity (mg g −1 ); Q m is the maximum adsorption capacity on the adsorbent surface of the monolayer (mg g −1 ); C e is the equilibrium concentration of Cr(VI) (mg L −1 ); and K L is the affinity constant (L mg −1 ) 29-31 . Freundlich equilibrium adsorption isotherm describes the adsorption characteristics of the heterogeneous surface, which comprises the terminal and functional groups that offer stable binding sites. The Freundlich adsorption isotherm is an empirical equation and is represented by Eq. 2.
[formula] = Q K C (2) e F e n 1/ [/formula]
where Q e is the equilibrium adsorption capacity (mg g −1 ); K F is the Freundlich constant related to adsorption capacity (mg g −1 ); n is the adsorption intensity.
The experimental results were fitted using Eqs. 1 and 2 described above. As shown in, the points and nonlinear curves were well matched. The last point in the Langmuir fittingindicated that it was Adsorbents GO drying process Contaminants Ref. www.nature.com/scientificreports www.nature.com/scientificreports/ nearly in equilibrium. A comparison of the two models revealed that the Langmuir model showed relatively high regression coefficients (R square, Rs), whereas the Freundlich model showed low Rs values because of the high concentration results, indicating that the Langmuir model was relatively better in describing the adsorption capacities. These results confirmed that the surfaces of the adsorbents were homogeneous, and adsorption reactions occurred well in the monolayer. On the other hand, the isotherm models for ODGO + 3N did not fit the equilibrium adsorption data well, due to the poor fit of the third point of the result.shows XPS analysis after amine functionalization of the samples. It proved that amino silanes were successfully functionalized on the GO surface regardless of dry process. In adsorbents, the binding energies of Si atom were detected at around 153 eV (Si 2s) and 103 eV (Si 2p). Also, the peak of N 1 s at around 400 eV was confirmed.shows a comparison of the Cr(VI) maximum adsorption capacities of the adsorbents that were prepared by the different functionalization processes and showed a significant difference in adsorption capacities. SGO prepared without the drying process recorded the highest Cr(VI) adsorption capacity of 258.48 mg g −1 , while ODGO dried at 70 °C in air showed the worst capacity of 173.45 mg g −1 . The Cr(VI) adsorption capacity of FDGO was 212.46 mg g −1 and was in-between the values for the other two samples. The values for the 1/n data also showed the orders of adsorption for SGO, FDGO and ODGO. The larger the 1/n values, the greater is the adsorption. The data for the maximum adsorption capacities and 1/n values were therefore in agreement and indicate that the process is critical for imparting adsorption capacity to GO. During the functionalization process, the molecules used, access, react, and bind to the oxygen on the GO surface. The differences in the performances of the GO samples are expected to originate from the differences in the accessibility of the functionalized molecules to the Cr(VI) species; i.e., the overlapping of GO layers could contribute to the differences in the accessibility of the molecules to GO surface. To verify this hypothesis, the following additional analyses were carried out. www.nature.com/scientificreports www.nature.com/scientificreports/show the same amount of GO samples, which were prepared with different processes, and indicate a huge volume difference. FDGO had a much larger volume than ODGO. Further, SEM observations confirmed that the overlapping of GO layers is much more extensive in ODGO than FDGO. The layers of FDGO were well dispersed and enabled the visualization of the transparent GO layers and their wrinkles, whereas the layers of ODGO were multiply stacked and formed particles, which might have resulted in the exposed surface being smaller. The BET analysis results supported this observation. The surface area of ODGO was 2.9587 m 2 g −1 , which is significantly lower than the 24.139 m 2 g −1 of FDGO in Supporting Information). In summary, oven-drying resulted in low dispersion of the GO layer, thereby reducing the accessible surface area. Next, we turned our attention to confirm that the reduction in the accessible surface area influenced the degree of functionalization, which is directly related to the adsorption performance.
## Characterization.
To quantify the degree of functionalization, we conducted a TG analysis. TGA residual after heating up to 800 °C is the amount of SiO 2 , indicates that of functionalizing amino-silane on each GO adsorbent. The residue obtained for SGO + 3N was the largest and was followed by FDGO + 3N and ODGO + 3N. Furthermore, the XPS analysis results were consistent with the TG results. The XPS atomic concentration results of Si/C and N/C indicated that the SGO + 3N contained the highest concentration of the amine functional groups on the surface. The TGA and XPS results have fewer differences than those of the BET surface area because the overlapped GO layers of ODGO might be dispersed further to some extent.
Elemental analysis was conducted to determine if the reduction of the exposed surface area was the only cause of performance degradation. The EA results presented in confirm that the oxygen content was reduced due to partial reduction of the exposed surface area during the drying process, and was significantly lower in ODGO. This means that -COOH and -OH functional groups, which are necessary for functionalization, were reduced during the drying process, thus decreasing the density of silane-functionalization. These studies reveal that despite using the same GO source, variations in the drying processes cause the differences in the density of sites for silane-functionalization, as well as influence the effective surface area. In summary, the degradation of adsorption performance can be attributed to the physical changes, loss of effective GO surface area, the ensuing chemical changes, and the reduction of GO.
# Conclusions
The studies revealed that avoiding drying process is the most effective approach for ensuring suitable surface functionalization of GO. This is not only because the dispersion of GO layers can be maintained and the effective exposure surface area can be extensive by avoiding the drying process, but the reduction of GO can be suppressed. More importantly, the drying method makes a significant difference in the GO surface state. If the drying process is inevitable, freeze-drying is recommended rather than oven-drying. General oven-drying has a negative effect on the GO's performance, caused by the overlapping of GO layers and the reduction of GO. It is confirmed that freeze-drying causes less overlapping and lower loss of oxygen-containing surface functional groups, which act as bonding-sites for silane-functionalization. In this paper, we suggest suitable processes for the treatment of GO with systematic experiments and analysis, which will aid better functionalization and development of GO-based catalysts. Synthesis of 3N-GO-adsorbents. Scheme 1 shows the overall processes used for the adsorbent synthesis and summarizes the differences by the process. GO-V50 was functionalized with AEAEAPTMS using a slight modification of a reported method, and 3 types of GO-V50 were prepared. The first was prepared in the solution form, the second was prepared by freeze-drying, and the final was by oven-drying at 70 °C Each of www.nature.com/scientificreports www.nature.com/scientificreports/ these suspensions (0.5 g of GO/140 mL of methanol) was sonicated for 2 h and refluxed at 60 °C with stirring at 1000 rpm. At 60 °C, AEAEAPTMS (5 mL) was added to each suspension. After a 24 h reaction, the product was washed with ethanol using centrifugation at 13,500 rpm for 12 min three times. Each washed sample of AEAEAPTMS-GO (3N-GO) was dried at 40 °C overnight. Lastly, the products were treated with 0.1 M HCl for 6 h and were collected by centrifugation and dried at 40 °C overnight.
## Equilibrium adsorption of cr(vi) on go-adsorbents.
We acquired the Cr(VI) adsorption isotherms of five samples of Cr(VI) solutions of concentrations 5, 10, 25, 50, 100 mg- L −1 . The three samples (15 mg each) of the adsorbents which were prepared by drying at 70 °C, freeze-drying, and as a GO solution, respectively, and were injected into each of the five Cr(VI) solutions and were reacted for 24 h at 25 rpm using a rotator. Upon reaction completion, the solution was centrifuged at 4200 rpm for solid-liquid separation, and the mixture was filtered using a 0.2 μm PVDF syringe filter for completely separating the solution from the adsorbents.
Scanning electron microscopy (SEM). The morphology of GO samples was analyzed by scanning electron microscopy (SEM, FEI Inspect F50, AP-tech Company).
## Brunauer-emmett-teller (bet).
The specific surface area was determined from the linear portion of BET plots (P/P 0 = 0 to 1), which were acquired using a surface area analyzer (BEL-SORP-max, BEL Japan Inc., Japan). Thermogravimetric analysis (TGA). Thermogravimetric analysis (TGA, N-1500, Scinco) of the GO-adsorbents was performed to confirm the amount of aminosilanes grafted onto the GO surface by heating the adsorbents from 100 to 800 °C at a rate of 10 °C min −1 under air condition. X-ray photoelectron spectroscopy (XPS). X-ray photoelectron spectroscopy (XPS, PHI 5000
VersaProbe Ulvac-PHI with Al X-ray monochromatic source (Al K α source with energy 1486.6 eV at 24.5 W), Physical Electronics Inc.) analysis was performed to determine the surface composition of the adsorbents. The binding energies were referenced to the C 1s line at 284.6 eV from the adventitious carbon.
Elemental analysis (EA). The elemental ratio (O/C ratio) of GO was obtained from elemental analysis (EA, FLASH 2000, Thermo Scientific). The average values of 2~3 measurements were used as the data.
Inductively coupled plasma optical emission spectrometer (ICP-OES). An inductively coupled plasma optical emission spectrometer (ICP-OES, ProdigyPlus, Prodigy) and an autosampler were used for the quantitative analysis of the Cr(VI) solution before and after the reaction. Each sample was analyzed with 4 replicates, and the results were interpreted as the average value.
## Data availability
All data generated and/or analyzed during this study are included in this published article. |
Motor imagery and electrical stimulation reproduce corticospinal excitability at levels similar to voluntary muscle contraction
Background: The combination of voluntary effort and functional electrical stimulation (ES) appears to have a greater potential to induce plasticity in the motor cortex than either electrical stimulation or voluntary training alone. However, it is not clear whether the motor commands from the central nervous system, the afferent input from peripheral organs, or both, are indispensable to induce the facilitative effects on cortical excitability. To clarify whether voluntary motor commands enhance corticospinal tract (CoST) excitability during neuromuscular ES, without producing voluntary muscular contraction (VMC), we examined the effect of a combination of motor imagery (MI) and electrical muscular stimulation on CoST excitability using transcranial magnetic stimulation (TMS).Methods: Eight neurologically healthy male subjects participated in this study. Five conditions (resting, MI, ES, ES + MI [ESMI], and VMC) were established. In the ES condition, a 50-Hz stimulus was applied for 3 to 5 s to the first dorsal interosseous (FDI) while subjects were relaxed. In the MI condition, subjects were instructed to imagine abducting their index finger. In the ESMI condition, ES was applied approximately 1 s after the subject had begun to imagine index finger abduction. In the VMC condition, subjects modulated the force of index finger abduction to match a target level, which was set at the level produced during the ES condition. TMS was applied on the hotspot for FDI, and the amplitude and latency of motor evoked potentials (MEPs) were measured under each condition.Results: MEP amplitudes during VMC and ESMI were significantly larger than those during other conditions; there was no significant difference in MEP amplitude between these 2 conditions. The latency of MEPs evoked during MI and VMC were significantly shorter than were those evoked during rest and ES.Conclusions: MEP acutely reinforced in ESMI may indicate that voluntary motor drive markedly contributes to enhance CoST excitability, without actual muscular contraction.
# Background
Many studies have shown that paired associative stimulation with single transcranial magnetic stimulation (TMS) and ES, repetitive transcranial magnetic stimulation (rTMS), or transcranial direct current stimulation can induce either short-term potentiation or a depressive effect in the motor cortex [bib_ref] Direct current stimulation promotes BDNF-dependent synaptic plasticity: potential implications for motor learning, Fritsch [/bib_ref] [bib_ref] Non-invasive brain stimulation: a new strategy to improve neurorehabilitation after stroke?, Hummel [/bib_ref] [bib_ref] Facilitative effect of high frequency subthreshold repetitive transcranial magnetic stimulation on complex..., Kim [/bib_ref] [bib_ref] Suppression of ipsilateral motor cortex facilitates motor skill learning, Kobayashi [/bib_ref] [bib_ref] Facilitation of implicit motor learning by weak transcranial direct current stimulation of..., Nitsche [/bib_ref] [bib_ref] Modulation of associative human motor cortical plasticity by attention, Stefan [/bib_ref] [bib_ref] A temporally asymmetric Hebbian rule governing plasticity in the human motor cortex, Wolters [/bib_ref]. These facilitatory or suppressive effects are believed to accelerate motor recovery in patients with stroke [bib_ref] Noninvasive cortical stimulation in neurorehabilitation: a review, Harris-Love [/bib_ref] [bib_ref] Repetitive transcranial magnetic stimulation-induced corticomotor excitability and associated motor skill acquisition in..., Kim [/bib_ref] because cortical plasticity plays an important role in motor recovery [bib_ref] Neural correlates of motor recovery after stroke: a longitudinal fMRI study, Ward [/bib_ref] [bib_ref] Mechanisms underlying recovery of motor function after stroke, Ward [/bib_ref] , as well as motor learning in healthy individuals [bib_ref] Rapid plasticity of human cortical movement representation induced by practice, Classen [/bib_ref] [bib_ref] Modulation of muscle responses evoked by transcranial magnetic stimulation during the acquisition..., Pascual-Leone [/bib_ref]. In clinical situations, however, interventions other than transcranial stimulation are needed to facilitate cortical potentiation because of epilepsy, limited resources, or patient preference. In such cases, we attempt to provide some form of sensory input (e.g., visual input) [bib_ref] Kinesthetic illusory feeling induced by a finger movement movie effects on corticomotor..., Kaneko [/bib_ref] or ES, which have been used to improve muscular strength for purposes of rehabilitation.
The effect an exercise has on cortical excitability can be augmented by ES [bib_ref] Cortical excitability changes following grasping exercise augmented with electrical stimulation, Barsi [/bib_ref] [bib_ref] Changes of excitability in M1 induced by neuromuscular electrical stimulation differ between..., Sugawara [/bib_ref]. Therefore, the combination of voluntary effort and ES appears to have a greater potential to induce plasticity in the motor cortex than either electrical stimulation or voluntary training alone. However, it is not clear whether the motor commands produced in the central nervous system, the afferent input from peripheral organs, or both, are indispensable for cortical facilitation, since motor commands and afferent input are summed in the cerebral network during voluntary effort. We hypothesize that motor commands (e.g., those evoked during motor imagery [MI]) independently increase the effect of ES on cortical excitability without afferent input, since the increment of corticomotor excitability during MI is widely accepted [bib_ref] Corticospinal excitability is specifically modulated by motor imagery: a magnetic stimulation study, Fadiga [/bib_ref] [bib_ref] Decreased cortical excitability during motor imagery after disuse of an upper limb..., Kaneko [/bib_ref] [bib_ref] Evidence for facilitation of motor evoked potentials (MEPs) induced by motor imagery, Kasai [/bib_ref] [bib_ref] An increase in cortical excitability with no change in spinal excitability during..., Yahagi [/bib_ref]. We propose that the ES potentially affects reflex gain at the spinal level, and via the higher areas of the central nervous system.
In the current study, we used neuromuscular ES combined with MI to examine the acute effects of a combination of ES and voluntary motor commands on cortical excitability. We aimed to test the hypothesis that a combination of MI and ES could reproduce corticospinal excitation at levels similar to those occurring during voluntary muscular contraction (VMC).
# Methods
## Subjects
Eight neurologically healthy male subjects (mean age, 23.3 ± 1.8 years; range, 21-26 years) participated in this study. Each subject provided his informed consent prior to participating in the experiment, which was approved by the ethics committee of the National Institute of Advanced Industrial Science and Technology (AIST). This experiment was carried out in the period in which the first author was working in AIST. The experimental setup is shown schematically in [fig_ref] Figure 1: The experimental setup [/fig_ref] -A. Subjects were seated comfortably in a chair with their forearm naturally pronated. The left index finger was fixed in the middle position of the full range of abduction, which was the point that subjects felt that they were in the most appropriate position to produce abduction torque.
## Electromyography (emg)
Surface EMG activity was recorded from the left first dorsal interosseous (FDI) using pairs of Ag-AgCl disc electrodes (5-mm diameter) placed on the center of the muscle belly with a 10-mm interelectrode distance. Prior to placing the electrodes, the skin was cleaned with alcohol and abraded with an abrasive skin-prepping gel. EMG signals were amplified (Neuropack MEB2200, Nihon Kohden Co. Ltd., Tokyo, Japan) (×1000) and filtered (5.3-1000 Hz). The isometric maximum voluntary contraction (MVC) force was first measured for index finger abduction using a force transducer (LMR-S-SA2, Kyowa Electronic Instruments Co., LTD., Tokyo, Japan), which also measured the abduction torque during voluntarily index finger abduction. The EMG signals and index finger abduction torque were digitized and recorded at 20 kHz using an A/D converter (Power 1401, Cambridge Electronic Design, Cambridge, UK). The triggers for ES and TMS were generated by the Power 1401 and temporally synchronized with the EMG and torque signals.
TMS TMS (Magstim 200 2 , The Magstim Company Limited, Whitland, UK) was delivered to the right (contralateral) motor cortex, with anteromedial current flow in the motor cortex (perpendicular to the central sulcus), to induce motor evoked potentials (MEPs) recorded from the left FDI. Stimulation was delivered through a figure-eight coil (9-cm diameter loops). Resting motor thresholds were defined as the lowest currents at which MEPs were evoked with a peak-to-peak amplitude greater than 50 μV [bib_ref] A: Safety, ethical considerations, and application guidelines for the use of transcranial..., Rossi [/bib_ref] [bib_ref] Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots:..., Rossini [/bib_ref]. During testing, the TMS stimulus intensity capable of inducing MEPs with average peak-to-peak amplitudes of 0.5-1 mV was applied during the resting condition. Ten MEPs were recorded under each of 5 conditions (Rest, ES, MI, ES with MI [ESMI], and VMC). To allow us to normalize MEP amplitudes between subjects, we measured the gain of the neuromuscular pathway through to the FDI in each subject after TMS testing. A 1-ms rectangular electrical stimulus was applied to the ulnar nerve at the dorsal point of the medial epicondyle, and the resulting supramaximal M-waves were recorded from the FDI using paired electrodes.
## Experimental conditions
Five conditions were established, including the resting condition [fig_ref] Figure 1: The experimental setup [/fig_ref]. In the ES and ESMI conditions [fig_ref] Figure 1: The experimental setup [/fig_ref] , 1-ms rectangular electrical pulses were applied at 50 Hz while subjects were relaxed. The duration of the electrical stimulus trains was precisely varied from 2 to 4 s under the control of the Labview software package (National Instruments Corporation, Austin, TX, USA). The duration was an integral number of seconds, the exact value being randomly set in each trial. The stimulation electrodes were placed on the skin next to the recording electrodes and oriented parallel to the muscle fibers. The intensity of ES was set at a level that induced a small percentage of the FDI MVC, measured before the experiment, 5.8 ± 1.0 mA) without eliciting pain. The stimulus intensity of ES was determined at the beginning of the experiment by increasing the ES intensity until it reached the maximum level without causing discomfort or pain. While determining the ES intensity, index finger abduction torque was measured to determine the target torque for the VMC and ESMI conditions, so that the force produced from the FDI contraction was similar in ES, ESMI, and VMC conditions. TMS was applied 20 ms after the last ES pulse ended to prevent stimulation artifacts from contaminating the MEPs in ES and ESMI. In the MI condition, subjects imagined abducting the left index finger with maximal effort [fig_ref] Figure 1: The experimental setup [/fig_ref]. TMS was applied one second after beginning of MI while subjects imagined performing finger movements and perceiving the kinesthetic feeling of muscular contraction [bib_ref] Decreased cortical excitability during motor imagery after disuse of an upper limb..., Kaneko [/bib_ref]. The background FDI EMG activity was examined by at least 2 investigators to identify voluntary muscle activation. The EMG activity was viewed on 2 computer monitors. The EMG signal was displayed on a scale of 500 μV/division (10 divisions = full scale) on one monitor and on a scale of almost 100 μV/division (maximum scale) on the other monitor [bib_ref] Kinesthetic illusory feeling induced by a finger movement movie effects on corticomotor..., Kaneko [/bib_ref] [bib_ref] The effect of motor imagery on gain modulation of the spinal reflex, Aoyama [/bib_ref]. A trial was rejected if a small level of muscle activation was observed during the testing. Furthermore, each trial was examined off-line, and trials containing EMG signals with amplitudes that exceeded 100 μV were excluded from the data analysis. The EMG data was rectified offline and smoothed with a 1-ms moving average to allow comparisons among the experimental conditions. In the ESMI condition [fig_ref] Figure 1: The experimental setup [/fig_ref] , ES was applied approximately 1 s after the subject had begun to imagine index finger abduction, so that MI was performed for at least 3 s. The subject was instructed that it can be finished to imagine the movement after TMS was applied in a trial during MI and ESMI. Before applying this condition, the ES intensity had been finely modulated to adjust the induced force level to the target level if the force level produced was different from the target level. The abduction torque was examined by 2 experimenters to eliminate trials in which the torque deviated from the target level. In the VMC condition [fig_ref] Figure 1: The experimental setup [/fig_ref] , subjects modulated the force of their index finger abduction to match a target force level, which was presented on a monitor positioned in front of the subject. The target force level was the level produced during the ES condition. On average, the target force level was set at 1.56 ± 0.63% of the MVC of FDI.
# Data analysis
MEP amplitude and latency were measured during each condition. The MEP amplitude was normalized (divided by the supramaximal M-wave [%Mmax] amplitude) for each subject. MEP latency (ms) was defined as the duration between TMS delivery and the onset of an MEP. One-way repeated measures analysis of variance (ANOVA) were used to test the effect of "condition" (Rest, MI, and VMC) on background EMG activity, MEP amplitude, and MEP latency. Post hoc comparisons were made using Tukey's HSD; the threshold for statistical significance was set at p < 0.05.
# Results
Background EMG activity was compared among the resting, MI, and VMC conditions. The main effect of condition was statistically significant (rest: 0.014 ± 0.002 mV, MI: 0.020 ± 0.004 mV, VMC: 0.043 ± 0.013 mV, F = 32.413, p < 0.0001). Tukey's HSD post hoc test indicated that EMG activity was greater during VMC than during rest and MI (p < 0.01). There was no significant difference between background EMG at rest and MI. [fig_ref] Figure 2: Motor evoked potential amplitudes [/fig_ref] -A presents superimposed MEPs recorded during the ES, MI, ESMI, and VMC conditions. The main factor of condition was significant between the 5 different conditions (F = 13.459, p < 0.0001), and Tukey's HSD post hoc test revealed that MEP amplitudes were significantly increased during both the ESMI and VMC conditions, as compared with the resting, ES, and MI conditions. There was no significant difference in MEP amplitudes between the ESMI and VMC conditions, or among the resting condition and the ES or MI conditions [fig_ref] Figure 2: Motor evoked potential amplitudes [/fig_ref]. The average MEP amplitudes recorded during ES and MI conditions were larger than those recorded during the resting condition. However, there were no significant differences among the resting, ES, and MI conditions [fig_ref] Figure 2: Motor evoked potential amplitudes [/fig_ref]. [fig_ref] Figure 3: Mean MEP latencies [/fig_ref] shows the MEP latency for all 5 conditions. The latency in the VMC condition was shorter than those in the resting or ES conditions [fig_ref] Figure 3: Mean MEP latencies [/fig_ref]. The latency was significantly shorter during the MI condition than during the ES condition.
# Discussion
In the present study, TMS revealed a significant facilitation of corticospinal excitability during the ESMI and VMC conditions. More importantly, corticospinal tract excitability was acutely increased in the ESMI condition. MEP amplitude reached a level similar to that measured during brief voluntary muscle contraction (VMC in the present study), even though voluntary muscle contractions were absent during the ESMI condition. This suggests that ESMI intervention may potentiate corticospinal tract excitability. However, the mechanisms through which MEPs were affected in the ESMI and VMC conditions remain unclear, as this factor was not addressed in the current study. This augmented corticospinal tract excitability may be clinically effective for inducing voluntary movement in patients with motor dysfunction, such as patients who have suffered a stroke. Many studies have reported that potentiation after noninvasive cerebral stimuli was effective in motor learning, even if the intended subject had suffered a stroke [bib_ref] Noninvasive cortical stimulation in neurorehabilitation: a review, Harris-Love [/bib_ref] [bib_ref] Repetitive transcranial magnetic stimulation-induced corticomotor excitability and associated motor skill acquisition in..., Kim [/bib_ref] [bib_ref] Rapid plasticity of human cortical movement representation induced by practice, Classen [/bib_ref] [bib_ref] Modulation of muscle responses evoked by transcranial magnetic stimulation during the acquisition..., Pascual-Leone [/bib_ref]. rTMS, which can enhance corticospinal excitability [bib_ref] Transcranial magnetic stimulation and neuroplasticity, Pascual-Leone [/bib_ref] [bib_ref] The effects of motor cortex rTMS on corticospinal descending activity, Lazzaro [/bib_ref] , is not necessarily easy to handle because of its risk of causing epilepsy [bib_ref] A: Safety, ethical considerations, and application guidelines for the use of transcranial..., Rossi [/bib_ref]. If the acutely enhanced excitability shown in the present study were associated with cortical plasticity, then this intervention might be an effective rehabilitation strategy.
The effect on the excitability change during ESMI was clearly different from that seen during either individual ES or MI. The mechanism underlying the reinforcement of corticospinal tract excitability during ESMI may be associated with brain activity that occurs with MI, naturalistic afferent input from the muscle spindles, or spindle and/or cutaneous afferent spiking directly evoked with ES. MI, however, is not a purely psychological intervention. In agreement with several previous studies, the present study confirmed that MI could elicit neural activation in the cortical motor areas [bib_ref] Primary motor and sensory cortex activation during motor performance and motor imagery:..., Porro [/bib_ref] [bib_ref] Supplementary motor area and other cortical areas in organization of voluntary movements..., Roland [/bib_ref] [bib_ref] Functional anatomy of the mental representation of upper extremity movements in healthy..., Stephan [/bib_ref]. We speculate that neural activation was evoked in the cortical motor areas during the MI condition because the average MEP amplitude was greater during the MI condition than during the resting condition, even if the difference between those 2 conditions was not statistically significant. The reinforcement of corticospinal tract excitability does not contradict previous reports [bib_ref] Decreased cortical excitability during motor imagery after disuse of an upper limb..., Kaneko [/bib_ref] [bib_ref] Evidence for facilitation of motor evoked potentials (MEPs) induced by motor imagery, Kasai [/bib_ref] [bib_ref] An increase in cortical excitability with no change in spinal excitability during..., Yahagi [/bib_ref]. In our previous study, we demonstrated the stretch reflex gain increased significantly during MI, without concurrent changes in H-reflex amplitude [bib_ref] The effect of motor imagery on gain modulation of the spinal reflex, Aoyama [/bib_ref]. We interpreted that finding to suggest that gamma motoneurons received input from the motor cortex during MI, which subsequently increased the stretch reflex gain. If this interpretation is correct and intrafusal muscle fibers contracted as a result of gamma motoneuron activity during MI, the afferent input from the muscle spindle during ES was potentially enhanced when MI was performed simultaneously, than during ES performed without MI. Therefore, in the ESMI condition, the ES-evoked muscle contractions may augment the stretch reflex gain. Furthermore, it is well established that the cortical activation that occurs during MI is similar to that occurs during VMC [bib_ref] Motor planning, imagery, and execution in the distributed motor network: a time-course..., Hanakawa [/bib_ref]. Thus, cortical activation is induced by MI, and an enhanced stretch reflex within the spinal circuitry, since the spinal motoneuron pool excitability was enhanced accompanying with the enhanced stretch reflex, may augment corticospinal tract excitability at a level similar to that observed during VMC.
In contrast, ES evokes compound muscular action potentials, and because muscle fibers generally shorten when they contract, the muscle spindles are unloaded during ES. From this perspective, Ia afferent input evoked during ES would differ from that evoked during VMCs. From a different perspective, index finger abduction produced torque during these isometric contractions, and the index finger angle remained unchanged. However, the changes in muscle fiber length in situ that accompanies muscular contraction should occur in synchrony with each electrical pulse; each shortening-lengthening cycle of the muscle must therefore include the muscle extension phase during the relaxation period after contraction. In fact, McKeon and Burke showed that muscle spindles discharge immediately after a single motor unit twitch evoked with an electrical stimulus, even if the muscle was fully relaxed during the stimulation [bib_ref] Muscle spindle discharge in response to contraction of single motor units, Mckeon [/bib_ref]. Consequently, it is difficult to eliminate the possibility that afferent input from the muscle spindle enhanced MEP amplitude during ES. Furthermore, it is possible that the afferent input evoked with ES might be affected at the spinal or cortical level. The mechanisms underlying our findings should be investigated in detail in future studies.
Previous studies have shown that changes in corticospinal excitability were induced with ES applied during VMC [bib_ref] Changes of excitability in M1 induced by neuromuscular electrical stimulation differ between..., Sugawara [/bib_ref] [bib_ref] Real-time changes in corticospinal excitability during voluntary contraction with concurrent electrical stimulation, Yamaguchi [/bib_ref] ; however, whether activity in the motor areas produced during VMC, regardless of the state of muscular contraction is important, has not yet been clarified. A significant point in this study is the suggestion that cerebral activity during MI plays an important role, in addition to the effect of ES, in changing corticospinal excitability, regardless of the VMC in the periphery.
The MEP latency was shorter during MI than during ES. A potential reason for this difference in MEP latency is that MEP transmission from the motor cortex was modulated at different anatomical locations between the 2 conditions. Previous studies have shown that neural activity increases in the motor areas during MI [bib_ref] Corticospinal excitability is specifically modulated by motor imagery: a magnetic stimulation study, Fadiga [/bib_ref] [bib_ref] Decreased cortical excitability during motor imagery after disuse of an upper limb..., Kaneko [/bib_ref] [bib_ref] Evidence for facilitation of motor evoked potentials (MEPs) induced by motor imagery, Kasai [/bib_ref] [bib_ref] An increase in cortical excitability with no change in spinal excitability during..., Yahagi [/bib_ref] [bib_ref] Primary motor and sensory cortex activation during motor performance and motor imagery:..., Porro [/bib_ref] [bib_ref] Supplementary motor area and other cortical areas in organization of voluntary movements..., Roland [/bib_ref] [bib_ref] Functional anatomy of the mental representation of upper extremity movements in healthy..., Stephan [/bib_ref] , suggesting that the MEP increase evoked during MI is likely due to cerebral activation. Furthermore, a shortened latency was noted during VMC; however, the latency in ESMI was unclear, even though MEP amplitude obviously increased. The difference in MEP latency may be due to differences in the anatomical level at which motoneuron excitability was modulated.
One limitation of this study was that voluntary muscle activity during the ESMI condition was not analyzed, whereas the absence of muscular activity was recognized during the MI condition. Furthermore, the absence of EMG was confirmed, at least during the periods when only MI was executed before ES had begun in the ESMI condition. The mechanism of corticospinal excitability reinforcement during ESMI was not investigated in detail. However, it was shown that the combination of intention and ES increased corticospinal excitability, and this may prove to be clinically useful. For example, when functional neuromuscular electrical stimulation would be applied to a patient with stroke to facilitate the cortcicospinal excitability in the involved side, applying it with motor imagery or intension of a movement should be recommended.
# Conclusions
Although the exact mechanisms remain unclear, we successfully reproduced augmented corticospinal excitability during ESMI at a level similar to that observed during VMC. Future studies are necessary to reveal whether sustaining this condition can cause short-term potentiation at the cortical level and is clinically useful in motor rehabilitation.
## Competing interests
This manuscript has not been published or presented elsewhere in part or in entirety, and is not under consideration by another journal. All study participants provided informed consent, and the study design was approved by the appropriate ethics review boards. All the authors have approved the manuscript and agree with submission to your esteemed journal. There are no conflicts of interest to declare.
Authors' contributions FK and TK conceived the study, participated in the study design and coordination, and helped to draft the manuscript. TA and TH participated in the design of the study and performed the statistical analysis. All authors read and approved the final manuscript.
[fig] Figure 1: The experimental setup. Schematic diagram of the experimental setup (A). TMS was applied during (B) rest, (C) electrical stimulation (ES), (D) motor imagery, (E) ES + MI (ESMI), and (F) voluntary muscle contraction (VMC). The spike train, which is shown on the left side of the raw first dorsal interosseous (FDI) EMG, was an artifact of ES (C, E); only the last 12 stimuli are shown. [/fig]
[fig] Figure 2: Motor evoked potential amplitudes. MEP amplitude recorded from FDI during the resting, ES, MI, ESMI, and VMC conditions. (A) MEPs obtained from a single subject in each condition. (B) Mean (±S.D.) MEP amplitude recorded in each condition. *p < 0.05, *p < 0.01. [/fig]
[fig] Figure 3: Mean MEP latencies. Mean (±S.D.) latency of MEP recorded from FDI in each of the 5 conditions. *p < 0.05, **p < 0.01. [/fig]
[fig] Abbreviations: CoST: Corticospinal tract; Rest: Resting condition; MI: Motor imagery; ES: Electrical stimulation; ESMI: Electrical stimulation with MI; VMC: Voluntary muscular contraction; FDI: First dorsal interosseous muscle; MEP: Motor evoked potential; rTMS: Repetitive tanscranial magnetic stimulation; EMG: Electromyography; MVC: Maximum voluntary contraction; %Mmax: MEP amplitude was normalized by the supramaximal M-wave. [/fig]
|
Occurrence and response to treatment of Graves’ disease after COVID vaccination in two male patients
[bib_ref] Two cases of Graves' disease following SARS-CoV-2 vaccination: an autoimmune/ inflammatory syndrome..., Vera-Lastra [/bib_ref] [bib_ref] Effects of methimazole treatment on growth hormone (GH) response to GH-releasing hormone..., Giustina [/bib_ref]
## Case 1
A previously healthy, past-smoker, 32-year-old male patient, without any personal or family history of endocrine and autoimmune diseases, ten days after the second dose of Vaxzevria vaccine developed anxiety, tachycardia and palpitations, without signs of orbitopathy or local pain. These symptoms were not present after the first dose since the patient reported only general short-term self-limited mild side effects including pain and swelling at the injection site, fatigue and low-grade fever which resolved after few days. Thyroidfunction tests revealed decreased TSH with elevated free-T4 and free-T3 levels and elevated thyrotropin receptor-antibodies (TRAbs) [fig_ref] Table 1: Biochemical data of two patients with Graves' disease diagnosis and after 3... [/fig_ref]. Thyroid-ultrasonography revealed gland enlargement with pseudonodules and hypervascularization. He was initially treated with propranolol 40 mg/daily and thiamazole 15 mg/daily. However, he developed rush soon after starting thiamazole and was switched to propylthiouracil 150 mg/daily. Biochemical and clinical response was good with progressive down-titration of treatment reaching rapidly euthyroidism. After three months of therapy his thyroid function is currently normal and TRAbs levels halved on 100 mg/daily dose of propylthiouracil [fig_ref] Table 1: Biochemical data of two patients with Graves' disease diagnosis and after 3... [/fig_ref].
## Case 2
A previously healthy, past-smoker, 35-year-old male patient, without any personal or family history of endocrine and autoimmune disease, 5 days after the first dose of Vaxzevria vaccine developed headache, nausea, asthenia, palpitations, tachycardia, mild eyes-redness and superior palpebral retraction. Thyroid-function tests revealed suppressed TSH with elevated free-T4 and free-T3 levels and elevated TRAbs [fig_ref] Table 1: Biochemical data of two patients with Graves' disease diagnosis and after 3... [/fig_ref]. Thyroid ultrasonography revealed gland enlargement and hypervascularization. He was treated with propranolol 20 mg/daily and thiamazole 15 mg/daily with a good clinical and biochemical response. After three months of therapy his thyroid function and TRAbs levels are currently normal on 5 mg/daily dose of thiamazole [fig_ref] Table 1: Biochemical data of two patients with Graves' disease diagnosis and after 3... [/fig_ref]. Patient was counseled elsewhere not to undergo the second dose of the vaccine. The necessary spread of vaccinations on a population basis may allow observation of their even rare side effects. Autoimmune endocrine diseases after vaccination have been previously reported possibly associated to autoimmune syndrome induced by adjuvants (ASIA). In genetically susceptible and predisposed subjects, particularly in those carrying specific HLA-DRB1 class II typing haplotypes or affected by previous autoimmune/inflammatory diseases and/ or allergic reactions, ASIA is triggered by several vaccine adjuvants and excipients leading to a dysfunctional immune response causing different pathological conditions and endocrinopathies [bib_ref] Anaphylaxis to the first COVID-19 vaccine: is polyethylene glycol (PEG) the culprit?, Garvey [/bib_ref]. Interestingly, our report adds to the current literature in that risk of inducing GD does not seem to be related to a unique mechanism. In fact, if GD after mRNA vaccine could be possibly related to its content in polyethylene glycol (PEG) lipids [bib_ref] Two cases of Graves' disease following SARS-CoV-2 vaccination: an autoimmune/ inflammatory syndrome..., Vera-Lastra [/bib_ref] [bib_ref] Anaphylaxis to the first COVID-19 vaccine: is polyethylene glycol (PEG) the culprit?, Garvey [/bib_ref] , occurrence of GD after modified viral vaccine could be related to oil-in-water emulsion excipients such as polysorbate 80 [bib_ref] Vaccine adjuvants: understanding the structure and mechanism of adjuvanticity, Shi [/bib_ref] , since this latter vaccine does not contain PEG compounds.
To our knowledge, so far six cases including ours of GD after COVID-19 vaccination did occur [bib_ref] Two cases of Graves' disease following SARS-CoV-2 vaccination: an autoimmune/ inflammatory syndrome..., Vera-Lastra [/bib_ref] and overall a 1:1 female male ratio has emerged. Therefore, our observation strengthens the hypothesis of a causal vaccine-GD relationship since males are rarely affected by GD and are known to be less genetically susceptible to develop autoimmune syndromes [bib_ref] Two cases of Graves' disease following SARS-CoV-2 vaccination: an autoimmune/ inflammatory syndrome..., Vera-Lastra [/bib_ref]. To date, no potential strong predictor or risk factor has yet been reported to influence GD occurrence after COVID-19 vaccination. Interestingly, both patients here reported were past-smokers, a factor widely known to affect immune and inflammatory response and autoimmune thyroid diseases occurrence including GD [bib_ref] Graves' disease: Epidemiology, genetic and environmental risk factors and viruses, Antonelli [/bib_ref]. Moreover, hypovitaminosis D is widespread in Italy in both males and females [bib_ref] Sexual dimorphism of coronavirus 19 morbidity and lethality, Brandi [/bib_ref] as well as in many parts of the world. Interestingly, due to the known immunomodulatory role of Vitamin D (VD) [bib_ref] Skeletal and extraskeletal actions of vitamin D: current evidence and outstanding questions, Bouillon [/bib_ref] , hypovitaminosis D has been reported to predispose both to COVID-19 occurrence and severity [bib_ref] Mechanisms in endocrinology: vitamin D and COVID-19, Bilezikian [/bib_ref] and to autoimmune thyroid disease [bib_ref] The role of vitamin D in thyroid diseases, Kim [/bib_ref]. Therefore, it cannot be excluded also a possible influence of VD status on GD occurrence after COVID-19 vaccination, although data on VD status in our patients were not available. Furthermore, information on response to treatment of GD occurring after COVID-19 are scanty. In our two cases hyperthyroidism presented with relatively mildly elevated thyroid hormones and appeared to be well responding to medical treatment (either with thiamazole or propylthiouracil) with rapid restoration of euthyroidism and normalization or remarkable decrease of TRAbs with standard to low dose treatment [bib_ref] Effects of methimazole treatment on growth hormone (GH) response to GH-releasing hormone..., Giustina [/bib_ref].
In conclusion, although COVID-19 vaccination is currently not contraindicated in patients with stable autoimmune endocrine diseases [bib_ref] COVID-19 and endocrine and metabolic diseases. An updated statement from the, Puig-Domingo [/bib_ref] [bib_ref] One year of the pandemic -how European endocrinologists responded to the crisis:..., Giustina [/bib_ref] our data suggest a note of caution in administering a further vaccine dose to patients that developed GD in the first weeks after vaccination. Moreover, we suggest to carefully consider if submitting patients with recent onset or non-stable GD to all types of COVID-19 vaccination.
Finally, it can be worth including also thyroid function monitoring in patients with marked and persistent pathological symptoms occurring after vaccination potentially overlapping with thyrotoxicosis such as fever, palpitations and asthenia particularly in those with personal or family history of autoimmune thyroid and non-thyroidal disease [bib_ref] Development of Graves' hyperthyroidism during the early phase of pregnancy in a..., Gola [/bib_ref] , since development of hyperthyroidism even not severe may have a negative impact on outcomes particularly in the geriatric population [bib_ref] The negative role of subclinical thyrotoxicosis on the outcome of hospitalized geriatric..., Bossoni [/bib_ref].
## Data availability
All authors had full access to all the data in the study and take responsibility for the integrity of the data.
## Compliance with ethical standards
Conflict of interest The authors declare no competing interests.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
[table] Table 1: Biochemical data of two patients with Graves' disease diagnosis and after 3 months of treatment.TSH thyroid stimulating hormone, fT3 free triiodothyronine, fT4 free thyroxine, TRAbs TSH Receptor Antibodies.Reference ranges in parenthesis.Institute of Endocrine and Metabolic Sciences, San Raffaele Vita-Salute University and IRCCS Hospital, Milan, Italy [/table]
[bib_ref] The negative role of subclinical thyrotoxicosis on the outcome of hospitalized geriatric..., Bossoni [/bib_ref] |
Severe hyperkalemia following blood transfusions: Is there a link?
Core tip: Blood transfusion is associated with a wide range of potential complications. Among them, the increase of serum potassium levels is sometimes overlooked. Hyperkalemia is a potential deadly complication, especially when the patient has already increased potassium levels at baseline. A number of pathogenetic mechanisms associated with the development of hyperkalemia in patients receiving transfusions are discussed in the present case report. Moreover, the necessary precautions for minimizing the risk of transfusion-induced hyperkalemia are also presented. Rizos CV, Milionis HJ, Elisaf MS. Severe hyperkalemia World Journal of Nephrology W J N Submit a Manuscript
# Abstract
Patients with gastrointestinal bleeding often require large volume blood transfusion. Among the various side effects of blood transfusion, the increase of potassium levels is a serious one which is often overlooked. We report a case of severe hyperkalemia in a patient with gastric bleeding after large volume transfusion of packed red blood cells. The patient had hyperkalemia at baseline associated with his receiving medication as well as acute renal failure following hypovolemia. The baseline hyperkalemia was further aggravated after massive transfusions of packed red blood cells in a short period of time. The associated pathogenetic mechanisms resulting in the increase of potassium levels are presented. A number of risk factors which increase the risk of hyperkalemia after blood transfusion are discussed. Moreover, appropriate management strategies for the prevention of blood transfusion associated hyperkalemia are also presented. Physicians should always keep in mind the possibility of hyperkalemia in cases of blood transfusion.
# Introduction
Transfusion of red blood cells is associated with a wide range of potential complications . These include mainly febrile non-hemolytic transfusion reactions, allergic reactions, transfusion-associated circulatory overload, infections, acute lung injury and hemolytic reactions [bib_ref] Biomedical Excellence for Safer Transfusion C. Transfusion reactions: prevention, diagnosis, and treatment, Delaney [/bib_ref]. An increase in serum potassium levels associated with blood transfusion is often overlooked. Herein, we describe the case of a patient with massive hematemesis who manifested an aggravation of baseline hyperkalemia after large volume transfusion and discuss the underlying pathophysiology.
## Case report
A 90-year-old patient was admitted to our hospital due to several episodes of hematemesis which were soon followed by hematochezia. He was currently receiving treatment with benazepril (20 mg/d) and hydrochlorothiazide (25 mg/d) for hypertension as well as tamsulosin (0.4 mg/d) for benign prostate hypertrophy. All these medications were discontinued following the patient's hospital admission. He had undergone partial gastrectomy (Billroth Ⅱ) 10 years ago due to a perforated peptic ulcer.
Upon admission he was hemodynamically unstable and normal saline was administered pending transfusion with packed red blood cells (PRBCs). Admission laboratory testing revealed anemia, acute renal failure and hyperkalemia (6.14 mEq/L) without indications of hemolysis [fig_ref] Table 1: Patient's parameters during hospitalization [/fig_ref]. The electrocardiogram showed sinus rhythm tachycardia without changes related to hyperkalemia. The patient was oliguric during the first 15 h of hospitalization. Arterial blood gases were consistent with metabolic acidosis [fig_ref] Table 1: Patient's parameters during hospitalization [/fig_ref]. Monitoring of hematologic, renal, electrolyte and acid base parameters was performed multiple times per day.
Within 12 h the patient required transfusion with 4 PRBC units which were already stored for 30 to 35 d. At the time of admission, the patient had lost and was continuing to lose large volume of blood and was therefore at a critical status. The need to rapidly and efficiently manage the hemorrhagic-hypovolemic shock mandated massive volume blood transfusion using the readily available at the time unwashed PRBC units. Management of hyperkalemia consisted of intravenous administration, in parallel with the transfusions, of a solution made by combining 1 Lt 5% dextrose in water (D/W 5%) together with 15 IU fast acting insulin. The above solution of glucose and insulin was selected in order to both achieve a rapid decrease of potassium levels as well as minimize the risk of hypoglycemia in a patient already at a critical status. When the patient was hemodynamically stable, he underwent an emergency endoscopy which revealed a large ulcer at the site of the anastomosis without active bleeding at the time.
Following blood transfusions, the patient quickly restored normal diuresis and blood pressure. However, potassium further increased to a maximum of 6.97 mmol/L on day 2. This suggests that the aggravation of hyperkalemia was associated with the rapid massive transfusion rather than the initial oliguria and renal impairment. The administration of the D/W 5% with insulin solution was continued and serum potassium levels gradually decreased to normal [fig_ref] Table 1: Patient's parameters during hospitalization [/fig_ref]. As a result, no additional measures to reduce potassium levels were required. On day 3 a relapse of gastric bleeding occurred and the patient required another blood transfusion with a total of 7 PRBC units (which were also already stored for 30 to 35 d) as well as with fresh frozen plasma. However, this time the transfusion of the 7 PRBCs was done over a much greater period (4 d) instead of the initial rapid transfusion of the first 4 PRBCs, which were administered in only 12 h. No further increase in serum potassium was observed. Thereafter, the patient had an uneventful recovery and was discharged after 15 d with normal serum electrolyte concentrations.
# Discussion
There is evidence that transfusion of PRBCs can lead to increased potassium levels, especially in cases of large volume transfusion [bib_ref] Hyperkalemia after packed red blood cell transfusion in trauma patients, Aboudara [/bib_ref]. This was an interesting case of a patient with massive gastric hemorrhage and hyperkalemia at baseline which was aggravated following PRBCs transfusion.
Persistent hyperkalemia is observed only when urinary potassium excretory capacity is reduced due to renal failure, effective circulating volume depletion or more commonly hypoaldosteronism [bib_ref] Disorders of potassium, Schaefer [/bib_ref]. However, increased potassium intake and movement of potassium from cells into extracellular fluid may also play a prominent role in the pathogenesis of hyperkalemia [bib_ref] Disorders of potassium, Schaefer [/bib_ref]. Causes of increased potassium intake include a rapid intravenous infusion, the administration of potassium penicillin mainly as an intravenous bolus, the ingestion of a KCl containing salt substitute or potassium supplements and the use of stored blood for transfusions [bib_ref] Derangements of potassium, Medford-Davis [/bib_ref]. Hyperkalemia evident upon admission could be due to the coexistent prerenal azotemia associated with oliguria and subsequently decreased potassium excretion [bib_ref] Disorders of potassium, Schaefer [/bib_ref]. In fact, volume depletion leads to decreased glomerular filtration rate and increased proximal sodium and water reabsorption. As a result, there is a marked decrease of fluid and sodium delivery to the distal nephron, leading to a decreased potassium secretion [bib_ref] Regulation of renal potassium secretion: molecular mechanisms, Welling [/bib_ref]. This impairment in renal potassium excretion is counterbalanced by the hypovolemic-induced secondary aldosteronism. However, in our case the simultaneous administration of benazepril impairs the synthesis of angiotensin Ⅱ and therefore aldosterone, thus contributing to the profound hyperkalemia observed at patient's admission. It has been suggested that impaired potassium entry into cells observed in patients with hypovolemia can also lead to increased potassium levels [bib_ref] Regulation of renal potassium secretion: molecular mechanisms, Welling [/bib_ref]. Hyperkalemia may contribute to the pathogenesis of the patient's coexistent metabolic acidosis at baseline by suppressing ammoniagenesis in the proximal tubules and decreasing medullary thick ascending limb NH4 + transport, which in turn decreases medullary accumulation of ammonium [bib_ref] Chronic hyperkalemia impairs ammonium transport and accumulation in the inner medulla of..., Dubose [/bib_ref]. As a result, hyperkalemia can decrease ammonium excretion, which is the major component of net acid excretion by the kidneys. Even though acidosis can directly increase potassium levels by both redistribution of potassium out of cells (in cases of mineral acidosis) and by decreased potassium secretion in distal tubules [bib_ref] Disorders of potassium, Schaefer [/bib_ref] , in the present case the mild degree of acidosis is not a major factor for the pathogenesis of hyperkalemia.
The patient had restored normal diuresis and normal blood pressure 15 h after admission. Despite the improvement of blood pressure, diuresis, the continuous administration of D/W 5% plus insulin solution as well as the discontinuation of benazepril, a further increase of potassium levels was observed. Acute renal failure is associated with hyperkalemia when it is accompanied with low urine flow, oliguria and therefore low sodium delivery to the distal tubule, leading to decreased renal potassium excretion. However, in our case the patient had restored normal diuresis. As a result, it is less likely that the observed hyperkalemia is a consequence of acute renal failure. As a result, this aggravation of hyperkalemia can be attributed to the initial rapid transfusion of PRBCs rather than a hypovolemic state. The patient also received transfusion of additional 7 PRBCs between days 2 and 5 of hospitalization. However, these transfusions were not accompanied by further elevation of potassium levels. This can be attributed to the much greater period over which these transfusions took place.
The use of stored blood for transfusions is followed by an increase of serum potassium levels . Indeed, a prospective study showed that potassium levels increase more pronounced in patients who receive blood stored for more than 12 d [bib_ref] A prospective study on red blood cell transfusion related hyperkalemia in critically..., Raza [/bib_ref]. In fact, potassium is gradually released from red blood cells of the stored blood, resulting in an extracellular potassium concentration that by 21 d can reach up to 30 mEq/L in whole blood and up to 90 mEq/L in PRBCs . Furthermore, the lysis and destruction of red blood cells, especially in the transfusion of older PRBCs, can further increase potassium levels. In addition, the significantly decreased blood volume of our patient at baseline in combination with the high volume transfusion would make him more susceptible to potassium levels elevation from the PRBCs transfusion. Indeed, the patient's circulating blood volume had been markedly reduced after the gastric hemorrhage and thus its capacity to dilute exogenously loaded potassium was limited.
The risk of potassium overload can be minimized by selecting only blood collected less than 5 d prior to transfusion and by washing any unit of blood immediately before infusion to remove extracellular potassium [bib_ref] A comparative study of reducing the extracellular potassium concentration in red blood..., Bansal [/bib_ref]. Furthermore, the use of potassium absorption filters during transfusion may also decrease potassium loading [bib_ref] Reduction in potassium concentration of stored red blood cell units using a..., Yamada [/bib_ref] [bib_ref] Transfusion of irradiated red blood cell units with a potassium adsorption filter:..., Cid [/bib_ref]. Moreover, factors that also play a role in the increase of potassium levels are the rate and volume of transfusion as well as the patient's circulating pre-transfusion blood volume.
All things considered, hyperkalemia should be not overlooked in patients receiving large volume blood transfusions. A high level of suspicion and prompt management is warranted by the physician.
## Comments
## Case characteristics
A patient with large volume blood loss due to gastrointestinal bleeding presented requiring emergency mass blood transfusion.
## Clinical diagnosis
An aggravation of baseline hyperkalemia was observed after massive blood To convert hemoglobin from gr/dL to gr/L multiply by 10; to convert serum creatinine from mg/dL to μmol/L multiply with 88.4; to convert serum urea from mg/dL to mmol/L multiply by 0.357; to convert serum glucose from mg/dL to mmol/L divide with 18. Ht: Hematocrit; Hb: Hemoglobin; Cre: Creatinine; Ure: Urea; PRBCs: Packed red blood cells received at the specific day.
## Comments
Rizos CV et al . Hyperkalemia associated with blood transfusion
[table] Table 1: Patient's parameters during hospitalization [/table]
|
Invisible water security: Moisture recycling and water resilience
A B S T R A C TWater security is key to planetary resilience for human society to flourish in the face of global change. Atmospheric moisture recycling -the process of water evaporating from land, flowing through the atmosphere, and falling out again as precipitation over land -is the invisible mechanism by which water influences resilience, that is the capacity to persist, adapt, and transform. Through land-use change, mainly by agricultural expansion, humans are destabilizing and modifying moisture recycling and precipitation patterns across the world. Here, we provide an overview of how moisture recycling changes may threaten tropical forests, dryland ecosystems, agriculture production, river flows, and water supplies in megacities, and review the budding literature that explores possibilities to more consciously manage and govern moisture recycling. Novel concepts such as the precipitationshed allows for the source region of precipitation to be understood, addressed and incorporated in existing water resources tools and sustainability frameworks. We conclude that achieving water security and resilience requires that we understand the implications of human influence on moisture recycling, and that new research is paving the way for future possibilities to manage and mitigate potentially catastrophic effects of land use and water system change.
# Introduction
The movement and availability of water -through surface, soil, and atmosphere -is an important biophysical basis for water security and is crucial for the resilience of human society. Resilience includes the ability to persist in the face of a perturbation; to adapt to change; and to transform, leading to new system configurations of functions and interactions [bib_ref] Resilience: the emergence of a perspective for social-ecological systems analyses, Folke [/bib_ref]. Water resilience can be defined as the role of water in enhancing this persistence, adaptability, and transformability of a desired system state [bib_ref] Understanding of water resilience in the anthropocene, Falkenmark [/bib_ref]. Today, in a new era of social-hydro-ecological dynamics [bib_ref] On the social dynamics of moisture recycling, Keys [/bib_ref] [bib_ref] Progress in socio-hydrology: a meta-analysis of challenges and opportunities: progress in socio-hydrology, Pande [/bib_ref] [bib_ref] Evolving water science in the anthropocene, Savenije [/bib_ref] , significant anthropogenic impacts to the water cycle are re-wiring human-water dynamics and compromising the capacity of the biosphere to support the human endeavor.
While much of the focus in water management is on the visible water flows in rivers, and the use and pollution of this water by humans, most global freshwater flows are largely invisible. Each year 70,000 to 100,000 km 3 of water evaporates from land into the atmosphere where it combines with oceanic evaporation. Then, 100,000 to 130,000 km 3 yr −1 of atmospheric water falls as precipitation on land, providing the water that eventually flows off the land into the rivers, and back to the ocean [bib_ref] Atmospheric moisture transports from ocean to land and global energy flows in..., Trenberth [/bib_ref]. These atmospheric water flows vastly surpass the 45,000 km 3 yr −1 of global river flows [bib_ref] Global hydrological cycles and world water resources, Oki [/bib_ref].
Ocean-to-land moisture transport, notably through atmospheric rivers [bib_ref] Atmospheric rivers: a mini-review, Gimeno [/bib_ref] [bib_ref] Defining "atmospheric river": how the glossary of meteorology helped resolve a debate, Ralph [/bib_ref] and monsoon systems [bib_ref] The Indian Monsoon and its variability, Gadgil [/bib_ref] [bib_ref] Global monsoon dynamics and climate change, Zhisheng [/bib_ref] [bib_ref] Major mechanisms of atmospheric moisture transport and their role in extreme precipitation..., Gimeno [/bib_ref] , provides about 60% of precipitation on land. Conversely, land-to-land moisture transport driven by terrestrial evaporation, accounts for 40% of precipitation over land [bib_ref] Origin and fate of atmospheric moisture over continents, Van Der Ent [/bib_ref]. This process of evaporation from land, traveling through the atmosphere as vapor, and returning to the land surface as precipitation is often referred to as terrestrial moisture recycling [fig_ref] Figure 1: Moisture recycling [/fig_ref] , which is the focus of this paper. For brevity we will simply refer to terrestrial moisture recycling as 'moisture recycling' in this article. Broadly speaking, precipitation occurs under specific circumstances, including the availability of water vapor, proximity to saturation, and a dynamic or thermodynamic process that can drive precipitation [bib_ref] Thermodynamic limits of hydrologic cycling within the earth system: concepts, estimates and..., Kleidon [/bib_ref]. The intensity and pattern of moisture recycling also depends on evaporation and transpiration patterns [bib_ref] Effects of irrigation in india on the atmospheric water budget, Tuinenburg [/bib_ref] , geographic location [bib_ref] Sensitivity of monthly convective precipitation to environmental conditions, Myoung [/bib_ref] , and spatial scale [bib_ref] Observed variations of tropical convective available potential energy, Demott [/bib_ref].
Vegetation plays a particularly important role as it can alter the evaporative flow from land to atmosphere and thus regulates not only regional water availability, but also precipitation elsewhere. The relative role of vegetation varies strongly on a regional basis [fig_ref] Figure 1: Moisture recycling [/fig_ref] ,bc) and with seasons [bib_ref] Contrasting roles of interception and transpiration in the hydrological cycle -part 1:..., Wang-Erlandsson [/bib_ref]. The flow of evaporation can be in the form of T 'fast' fluxes of water vapor (water that evaporates quickly after being intercepted by the land and vegetation surfaces, such as the forest floor and canopy) or 'slow' fluxes (water that evaporates after infiltrating into soil moisture and used for biomass production in plants as transpiration, or that evaporates from open water). While 'fast' fluxes are most important for maintaining a quick turnover of moisture recycling locally, 'slow' fluxes occur long into the dry period and help support more distant precipitation during dry seasons [bib_ref] Forest-rainfall cascades buffer against drought across the amazon, Staal [/bib_ref].
Moisture recycling offers a mechanism through which processes and management of land can affect both the quantity and spatio-temporal distribution of invisible atmospheric flows of water and, consequently, the resilience of the ecological and social-ecological systems that depend on it. Anthropogenic influences on terrestrial evaporation include climate change-induced modification of temperature and water availability [bib_ref] Relative contributions of climate change, stomatal closure, and leaf area index changes..., Alkama [/bib_ref] [bib_ref] Global assessment of trends in wetting and drying over land, Greve [/bib_ref] , carbon dioxide fertilization [bib_ref] The impacts of climate change on water resources and agriculture in China, Piao [/bib_ref] , and land-use changes such as deforestation, irrigated croplands, and reservoirs [bib_ref] Hydroclimatic shifts driven by human water use for food and energy production, Destouni [/bib_ref] [bib_ref] Human modification of global water vapor flows from the land surface, Gordon [/bib_ref] [bib_ref] The impact of global land-cover change on the terrestrial water cycle, Sterling [/bib_ref]. Land-use change further has the ability to modify circulation patterns, such as through the influence of irrigation on monsoon onset in the Indian sub-continent [bib_ref] Effects of irrigation in india on the atmospheric water budget, Tuinenburg [/bib_ref] and afforestation impacts to wind directions in the Amazon [bib_ref] Future deforestation in the amazon and consequences for south american climate, Swann [/bib_ref]. Because rainfall is a key determinant of the local hydroclimate that ecosystems and societies have adapted to, substantial and abrupt changes in moisture recycling can lead to a reduction of local resilience and increase the risk for irreversible regime shifts. Based on a review of water related regime shifts, [bib_ref] Understanding of water resilience in the anthropocene, Falkenmark [/bib_ref] proposed "moisture feedback" as one of eight core resilience functions of water [bib_ref] Understanding of water resilience in the anthropocene, Falkenmark [/bib_ref].
In this paper, we summarise the current state of knowledge on human influences to moisture recycling and on the contribution of atmospheric moisture recycling to water security and the resilience of different ecological and social-ecological systems both locally and at a distance. We then consider implications for governance and the importance of a planetary perspective on building a water resilient future.
## Regime shifts and resilience
A regime shift refers to a large and abrupt system-wide re-configuration of interactions and/or functions, often triggered by an external perturbation or change in internal dynamics [bib_ref] Regime shifts, resilience, and biodiversity in ecosystem management, Folke [/bib_ref] [bib_ref] Early warning of climate tipping points, Lenton [/bib_ref] [bib_ref] Ecosystems and the biosphere as complex adaptive systems, Levin [/bib_ref] [bib_ref] Catastrophic regime shifts in ecosystems: linking theory to observation, Scheffer [/bib_ref]. Regime shifts can be non-linear, with gradual changes leading to abrupt responses in the system. They occur when system thresholds are passed. For some systems they are easily reversible. However, in most regime shifts, internal feedbacks among system drivers change, resulting in hysteresis effects where the external drivers need to be restored beyond the point when the regime shift happened. This may be both very costly and difficult to reverse, or even irreversible. In the context of moisture recycling, the key dimensions are precipitation and vegetative cover. In some cases the transition from low to high vegetation cover is continuous , while in other systems, especially tropical forest systems, a hysteresis may exist between a lower vegetation savanna state and a higher vegetation forest cover state . water that has evaporated from the Earth's surface and falls back as precipitation; while the ocean provides 60% of the precipitation falling globally on land, the orange box denotes the focus of this article, i.e. the land-to-land component, which provides 40% of terrestrial precipitation; (b) the percent of total evaporation that is regulated by vegetation for precipitation on land elsewhere; and (c) the percent of precipitation that is dependent on upwind vegetation regulation [bib_ref] Revealing invisible water: moisture recycling as an ecosystem service, Keys [/bib_ref].
## Fig. 2.
A simple diagram showing how a linear change in precipitation (x-axis) may be associated with a nonlinear change in vegetation cover (y-axis). In some cases, (a) the change can be continuous and reversible, otherwise (b) changes can be discontinuous, leading to a hysteresis in the nonlinearity [bib_ref] Global resilience of tropical forest and savanna to critical transitions, Hirota [/bib_ref] [bib_ref] Self-amplified amazon forest loss due to vegetation-atmosphere feedbacks, Zemp [/bib_ref].
## Evidence of moisture recycling influence on water resilience in various systems
## Tropical rainforests
Forests are disproportionately important to the global water cycle owing to their high and persistent evaporation rates, e.g., tropical, evergreen broadleaf forests currently occupy about 10% of the Earth's land surface, but contribute 22% of land evaporation, and receive 29% of precipitation on land [bib_ref] Contrasting roles of interception and transpiration in the hydrological cycle -part 1:..., Wang-Erlandsson [/bib_ref]. Forests in general, including tropical broadleaf forests, sustain transpiration long into rainless periods through their high soil moisture holding capacity and roots that reach deeper water in the soil, while also supplying high evaporation rates from interception of rainfall in the canopy, forest floor, as well as tree trunk vegetation and lichen [bib_ref] Contrasting roles of interception and transpiration in the hydrological cycle -part 1:..., Wang-Erlandsson [/bib_ref] [bib_ref] Uncertainties in transpiration estimates, Coenders-Gerrits [/bib_ref] [bib_ref] Intercepted by lichens, Savenije [/bib_ref] [bib_ref] Global root zone storage capacity from satellite-based evaporation, Wang-Erlandsson [/bib_ref]. Conversion of tropical forests to agricultural land or pasture generally decreases soil moisture and interception storage. Consequently, this decreases evaporation, the amount of moisture recycled, and precipitation available for downwind regions.
The average moisture recycling over the large, continuous tropical forests of Amazon and Congo is 25-45% depending on moisture tracking methods and study region selection [bib_ref] Recycling of water in the amazon basin: an isotopic study, Salati [/bib_ref] [bib_ref] Congo basin precipitation: assessing seasonality, regional interactions, and sources of moisture, Dyer [/bib_ref]. In parts of the Amazon and Congo, up to 50% of the forest precipitation originates from the forest itself. Internal moisture recycling in forests is especially important during drought years, when, in the Amazon and Congo for instance, recycling increases by 7-15% [bib_ref] Drought and deforestation: has land cover change influenced recent precipitation extremes in..., Bagley [/bib_ref].
Decreasing amounts of precipitation (both annual and seasonal) over tropical forests is strongly connected to forest resilience, with a forest-savanna bistability range of 1000-2400 mm yr −1 of precipitation [bib_ref] Global resilience of tropical forest and savanna to critical transitions, Hirota [/bib_ref]. For a savanna to return to a forest requires higher precipitation rates than the rainfall level that caused forest collapse, due to hysteresis induced by moisture recycling and fire feedbacks . Moisture recycling feedbacks are further important in large continuous forest areas, as they allow deforestation-induced declines in evaporation to cascade, leading to self-amplified forest loss when precipitation decreases [bib_ref] Forest-rainfall cascades buffer against drought across the amazon, Staal [/bib_ref] [bib_ref] Self-amplified amazon forest loss due to vegetation-atmosphere feedbacks, Zemp [/bib_ref] [bib_ref] Exploring the likelihood and mechanism of a climate-change-induced dieback of the amazon..., Malhi [/bib_ref] [bib_ref] The impact of amazonian deforestation on amazon basin rainfall, Spracklen [/bib_ref]. The fundamental importance of moisture recycling for both sustaining tropical forest structure, function, and for regulating the flow of moisture to downwind regions, suggests that forests will need to be a key point of consideration in understanding the relationship between water security and moisture recycling.
## Drylands and savannas
Drylands and savannas typically experience water scarcity accompanied by high temperatures. Drylands are not necessarily characterized by low rainfall, but their high mean temperatures and strong seasonality with long drought periods mean that a high proportion of the precipitation is evaporated. Nonetheless, plant dryland ecosystems and biodiversity are extremely well adapted to this characteristic water scarcity and climatic variability. Dryland plant ecosystems show many specific adaptations to water stress, including the ability to limit transpiration by storing water within the plant, to develop deep roots that reach deep soil moisture or groundwater, or to become dormant through long dry spells [bib_ref] Global root zone storage capacity from satellite-based evaporation, Wang-Erlandsson [/bib_ref].
Moisture recycling fosters resilience in drylands through stabilizing vegetation-driven ecohydrological feedbacks [bib_ref] Regime shifts in the sahara and sahel: interactions between ecological and climatic..., Foley [/bib_ref] [bib_ref] Contribution of waterlimited ecoregions to their own supply of rainfall, Miralles [/bib_ref] [fig_ref] Figure 3: Dynamics of dryland moisture recycling [/fig_ref]. These feedbacks include maintenance of plant community structure and the associated plant water storage [bib_ref] Understanding the role of ecohydrological feedbacks in ecosystem state change in drylands, Turnbull [/bib_ref] [bib_ref] Dryland ecohydrology in the anthropocene: taking stock of human-ecological interactions, Wilcox [/bib_ref]. As aridity increases, the bistability between a vegetated and unvegetated state can become more abrupt [bib_ref] Regime shifts in models of dryland vegetation, Zelnik [/bib_ref] , underlining the importance of the interplay among vegetation feedbacks and atmospheric moisture recycling feedbacks. In the extremely arid Kalahari, when woody vegetation with the associated (relatively) higher soil moisture flips to an absence of vegetation and soil moisture, this leads to increased system unpredictability [bib_ref] On soil moisture-vegetation feedbacks and their possible effects on the dynamics of..., Odorico [/bib_ref]. Thus, the delicate ecohydrological balance of dryland vegetation suggests that the functional characteristics of this recycling can change abruptly. Evidence suggests that moderate tree cover in the dry tropics can be beneficial for water availability by maximising groundwater recharge [bib_ref] Intermediate tree cover can maximize groundwater recharge in the seasonally dry tropics, Ilstedt [/bib_ref]. Others show that grasslands can potentially better withstand extreme weather events than forests in semi-arid regions under climate change [bib_ref] Grasslands may be more reliable carbon sinks than forests in california, Dass [/bib_ref]. It remains to be studied how moisture recycling affects hydrologically 'optimal' tree cover in dryland regions under climate change.
## Agriculture
## Rainfed agriculture
Rainfed agriculture comprises between 60% and 80% of global food production [bib_ref] Consumptive water use to feed humanity -curing a blind spot, Falkenmark [/bib_ref] [bib_ref] Agricultural green and blue water consumption and its influence on the global..., Rost [/bib_ref]. Specifically, water productivity in dryland agricultural production systems tends to be extremely low [bib_ref] Future water availability for global food production: The potential of green water..., Rockström [/bib_ref]. Yet, dryland agriculture provides the primary livelihood for about 8% of the world population, many of whom live in extreme poverty [bib_ref] Present and future water requirements for feeding humanity, Falkenmark [/bib_ref]. Supplemental irrigation is often not possible due to the scarcity or inaccessibility of [bib_ref] Future water availability for global food production: The potential of green water..., Rockström [/bib_ref] [bib_ref] The world's road to water scarcity: shortage and stress in the 20th..., Kummu [/bib_ref] surface water resources [bib_ref] Future water availability for global food production: The potential of green water..., Rockström [/bib_ref] [bib_ref] The world's road to water scarcity: shortage and stress in the 20th..., Kummu [/bib_ref]. Moreover, possibilities to secure food supplies through international trade can be limited by a lack of either economic or physical access to imported food [bib_ref] The use of food imports to overcome local limits to growth, Porkka [/bib_ref]. Even small declines in precipitation could therefore have disproportionately large impacts on agricultural yields and food security of dryland communities [bib_ref] Future water availability for global food production: The potential of green water..., Rockström [/bib_ref]. Also, rainfall seasonality (e.g. onset, length, and magnitude of growing season rainfall) has a significant impact on the functioning of dryland rainfed agriculture and ecosystems. Examples from the Sahel suggest that besides sufficient seasonal rainfall, length of the rainy season and steady rainfall after the first wet spells (when seeds tend to be sown) are crucial for successful harvest [bib_ref] The onset of the rainy season and farmers' sowing strategy for pearl..., Marteau [/bib_ref].
Recent studies have emphasised the importance of moisture recycling in the resilience of rainfed agricultural systems. Bagley et al. [bib_ref] Effects of land cover change on moisture availability and potential crop yield..., Bagley [/bib_ref] found that land-use change in the evaporation sources of major global breadbaskets could reduce their moisture availability from between 8% and 17% with significant impacts on grain yields ranging from 1% to 17%, particularly in arid regions such as the Central Asian wheat belt. Keys et al. [bib_ref] Analyzing precipitationsheds to understand the vulnerability of rainfall dependent regions, Keys [/bib_ref] studied the moisture recycling of seven locations where rainfed agriculture is particularly vulnerable to reductions in precipitation, and found that some regions, notably in East Asia, are more vulnerable to land-use change pressure in key evaporation source regions. Expansion of agriculture at the expense of the Amazon has also been shown to be a hydrologic no-win situation, due to decreased crop yields caused by reduction in rainfall supplied by forest evaporation [bib_ref] Large-scale expansion of agriculture in amazonia may be a no-win scenario, Oliveira [/bib_ref].
Besides managing upwind moisture supply regions, rainfed agriculture can also do more to foster resilient and adaptable agricultural systems while maintaining its own moisture recycling self-regulation, and downwind provision of moisture. The primary method here is the "vapor shift", i.e. shifting soil moisture evaporation to agriculturally productive plant transpiration [bib_ref] Future water availability for global food production: The potential of green water..., Rockström [/bib_ref] [bib_ref] Green water and African sustainability, Keys [/bib_ref].
## Irrigated agriculture
Irrigation-induced evaporation is known to have strong impacts on the hydrological cycle globally and regionally. For instance, De Vrese et al. [bib_ref] Asian irrigation, african rain: remote impacts of irrigation, De Vrese [/bib_ref] found that over a third of the annual mean precipitation in some of the arid regions of Eastern Africa can be attributed to moisture recycling from irrigated agriculture in Asia. Evaporation arising from irrigated agriculture contributes significantly to rainfall in Asia as well, comprising between 7 to 15% of Indian, Pakistani, Nepalese, and Tibetan precipitation, and over 20% in parts of northern Pakistan [bib_ref] Where does the irrigation water go? An estimate of the contribution of..., Wei [/bib_ref]. This pattern holds true in North America as well. In the Great Plains of the US, irrigation in Nebraska, Oklahoma, South Dakota, and Kansas has been associated with enhanced precipitation throughout the midwestern states [bib_ref] Evidence of enhanced precipitation due to irrigation over the great plains of..., Deangelis [/bib_ref]. Coupled regional climate modelling shows that irrigation in the California Central Valley can be linked to runoff increases of about 30% in the Colorado River [bib_ref] Irrigation in california's central valley strengthens the southwestern U.S. water cycle, Lo [/bib_ref].
Broadly, increased groundwater extraction to mitigate changes in moisture recycling provides a false sense of water security and instead undermines resilience. A common limitation of current global hydrologic models is their limited capacity to account for decreases in river flow caused by groundwater withdrawal. Withdrawals from the US High Plains aquifer from 1940-1980 outweighed insignificant local precipitation increases, and caused moderate to severe river flow depletion through a decline in groundwater levels [bib_ref] Large-scale water cycle perturbation due to irrigation pumping in the US high..., Kustu [/bib_ref]. In arid European regions, there is a clear inverse relationship between a decrease in moisture recycling and increases in groundwater use [bib_ref] Human water use impacts on the strength of the continental sink for..., Keune [/bib_ref]. As groundwater resources eventually disappear the agricultural systems adapted to this practice may struggle.
## River flows
Moisture recycling studies have often used river basins as their study region to analyse internal basin recycling [bib_ref] Recycling of water in the amazon basin: an isotopic study, Salati [/bib_ref] [bib_ref] A lagrangian perspective of the hydrological cycle in the congo river basin, Sorí [/bib_ref] , the sources of precipitation over a basin [bib_ref] Congo basin precipitation: assessing seasonality, regional interactions, and sources of moisture, Dyer [/bib_ref] [bib_ref] Precipitation, recycling, and land memory: an integrated analysis, Dirmeyer [/bib_ref] , and the fate of evaporation from a basin [bib_ref] Effects of irrigation in india on the atmospheric water budget, Tuinenburg [/bib_ref]. Anthropogenic modifications of river flows, such as construction of reservoirs and water transfers, also greatly affect evaporation [bib_ref] Local flow regulation and irrigation raise global human water consumption and footprint, Jaramillo [/bib_ref] and, presumably, moisture recycling.
Recently, coupled land-atmosphere modelling has more directly studied the effect of deforestation and irrigation on river flows through moisture recycling. Using a coupled land surface-moisture recycling framework, Wang-Erlandsson et al. [bib_ref] Remote land use impacts on river flows through atmospheric teleconnections, Wang-Erlandsson [/bib_ref] show that taking moisture recycling into account (i.e., allowing evaporation change to change precipitation) almost halves the estimate of human land-use change impacts on globally aggregated river flow (from +1200 km 3 yr −1 to +650 km 3 yr −1 ). Their results exhibit considerable spatial heterogeneity: in the Congo, Volga, and Ob basins, river flow increases are reduced by more than half; in the Amazon, river flow increases drop radically from +1630 to +270 m 3 s −1 ; and in the Yenisei, the sign of river flow change is reversed from an increase (+150 m 3 s −1 ) to a decrease (-220 m 3 s −1 ). The ability of land-atmosphere feedbacks to reverse the sign of change in river flow has also been shown under a business-as-usual deforestation scenario in the Xingu River basin in the Amazon, where accounting for land-atmosphere feedbacks switched river flow change estimates from an increase of 10-12% to a decrease of 30-36% [bib_ref] Interactions among amazon land use, forests and climate: prospects for a near-term..., Nepstad [/bib_ref]. Weng et al. [bib_ref] Aerial river management by smart cross-border reforestation, Weng [/bib_ref] further show that the heterogeneity of moisture recycling effects on Amazonian river flows extend to sub-basin levels, and identify sensitive sink and influential source regions.
The combined effect of land-use change on river flows through changes in the water balance, depends on complex interactions among spatial and temporal dynamics among ground, surface, and atmospheric water flows. In addition to the need for saturation and dynamical drivers (e.g. cooling or lifting), it is also important to evaluate what configurations of internal versus external land-use changes lead to persistence in precipitation supply, facilitate adaptation in part of the river basin, or completely transform the function and structure of the river basin. The critical issue of how river resilience is affected by feedbacks between river flow change and moisture recycling remains to be investigated.
## Megacities
Urban regions are rapidly expanding, increasing demands on scarce water resources. The water security of urban areas is a hotly studied topic [bib_ref] Water on an urban planet: urbanization and the reach of urban water..., Mcdonald [/bib_ref] , and the surface water that urban areas depend on is, in many cases, exposed to various moisture recycling pressures. Keys et al. [bib_ref] Megacity precipitation sheds reveal tele-connected water security challenges, Keys [/bib_ref] found that for 29 global megacities, municipal water was sourced from surface water supplies or actively recharged groundwater, which ultimately rely on precipitation falling within the watershed. Factors impacting moisture recycling such as the pace of land-use change, sensitivity to dry vs. wet year dynamics, and pre-existing water scarcity were particularly important. Understanding and managing moisture recycling (see Section 3) may need to be a part of designs to accommodate the needs of growing urban populations.
Urban water infrastructure is commonly designed for robustness to, and recovery from, water stresses and extreme events, with buffering capacity enabled by reservoirs, canals, pipes, treatment, and distribution networks [bib_ref] Urban growth, climate change, and freshwater availability, Mcdonald [/bib_ref]. These engineered approaches to water system resilience contribute to reliability, predictability, and to adaptability, but represent a more limited capacity to transform.
## Management and governance of moisture recycling: tools and concepts
## Basic spatial units of moisture recycling: precipitationshed and evaporationshed
Given the central role that terrestrial moisture recycling plays in water security, it would be useful to have an approach that (a) can assist in the identification of areas that provide rainfall to a given location, and (b) is flexible enough to incorporate multiple datasets, models, and spatial and temporal scales. The precipitationshed serves this purpose, and is defined as the upwind surface (both water and land) that contributes evaporation to a given location's precipitation [fig_ref] Figure 4: Conceptual illustration of the precipitationshed [/fig_ref] ; also see Keys et al. [bib_ref] Analyzing precipitationsheds to understand the vulnerability of rainfall dependent regions, Keys [/bib_ref]. As depicted, evaporation rises up from the surface, flows through the atmosphere, and falls as precipitation downwind.
## Integrating scales and boundaries of moisture recycling, governance, and social-ecological systems
Managing moisture recycling for resilience requires a recognition of the complexity of moisture recycling systems. Keys and Wang-Erlandsson [bib_ref] On the social dynamics of moisture recycling, Keys [/bib_ref] describe this complexity in terms of 'moisture recycling social-ecological systems', that can range along a spectrum from 'isolated', 'regional', to 'telecoupled'. An example of a telecoupled moisture recycling system is the precipitationshed for the country of Bolivia. Moisture recycling provides a biogeophysical connection from the Brazilian and Peruvian Amazon to Bolivia's precipitation. However, socioeconomic infrastructure (e.g. roads, telecoms) and institutions (e.g. biological preserves, international treaties) create telecoupled connections and feedbacks, increasing both system complexity and making deterministic policies or management unsuitable.
In terms of the evidence presented earlier, management can mean a variety of different things. First, rainfed systems (such as forests, drylands, and rainfed agriculture) can be thought of as both providers of evaporation (i.e. sources) as well as beneficiaries (i.e. sinks). Moisture recycling management of these landscapes thus ought to reflect this duality, giving thought to both the key aspects of evaporation flow as well as precipitation. However, some other types of moisture recycling systems are one or the other, such as irrigation systems, which are primarily generators of evaporation to the atmosphere, or urban systems, which are primarily users of runoff that is generated by terrestrial moisture recycling. In these cases, the management might take a different form since it is focused on a specific part of the atmospheric water cycle.
Knowledge of moisture recycling system architecture does not on its own confer an ability to manage moisture recycling. An understanding of system lags and feedbacks is also critical to appreciate the appropriate temporal and spatial scales for measurement, monitoring, and determination of management efficacy. Keys and Falkenmark [bib_ref] Green water and African sustainability, Keys [/bib_ref] suggest that different socio-hydrology systems operate at different timescales, often nested within one another. Additionally, the ability to detect an anthropogenic change in moisture recycling is complicated by internal climate variability [bib_ref] Variability of moisture recycling using a precipitationshed framework, Keys [/bib_ref].
Limited work examines the types of moisture recycling connections that may exist (particularly among nations), and the types of governance appropriate for each [bib_ref] Governance options for addressing changing forestwater relations, Ellison [/bib_ref] [bib_ref] Approaching moisture recycling governance, Keys [/bib_ref]. However, unlike surface water, atmospheric water does not channel into creeks, streams, and rivers, but rather flows continuously with wind currents around the planet. Despite hypothetical examples, intentional management and governance of moisture recycling remains highly theoretical. Yet, substantial evidence relates moisture recycling to the function and resilience of ecological and human systems alike. Consequently, we must understand how changes in land use, modification of surface and groundwater systems, agriculture and urbanization are not only changing terrestrial landscapes, but are altering vital planetary water flows.
## Integration of moisture recycling in existing sustainability management tools and frameworks
Blending moisture recycling into existing sustainability efforts is essential to regional and planetary resilience. Moisture recycling has been scientifically described and evaluated as a key ecosystem service globally, and its inclusion in existing ecosystem service frameworks has been outlined [bib_ref] Revealing invisible water: moisture recycling as an ecosystem service, Keys [/bib_ref]. Likewise, moisture recycling should be included in environmental and natural resource governance efforts, such as international conservation mechanisms and planning methods and potentially in trade and certification programs or other market and multilateral instruments [bib_ref] Approaching moisture recycling governance, Keys [/bib_ref]. Studies are further exploring the possibility to incorporate moisture recycling in water footprint assessments [bib_ref] Water accounting and vulnerability evaluation (WAVE): considering atmospheric evaporation recycling and the..., Berger [/bib_ref] , which otherwise conventionally regard evaporation as a loss [bib_ref] Limits to the world's green water resources for food, feed, fiber, timber,..., Schyns [/bib_ref].
At the largest possible scale, the planetary boundary framework, which is used in sustainability policy and governance, proposes limits to water modifications for Earth to remain habitable for humanity [bib_ref] Others, Planetary boundaries: exploring the safe operating space for humanity, Rockström [/bib_ref]. Moisture recycling is currently only implicitly accounted for in the planetary boundary for freshwater use through the proxy variable 'total blue water consumption' [bib_ref] Others, Planetary boundaries: exploring the safe operating space for humanity, Rockström [/bib_ref] [bib_ref] Towards a revised planetary boundary for consumptive freshwater use: role of environmental..., Gerten [/bib_ref] and in the land system change boundary.To explicitly account for human modifications of different components in the water cycle, a proposed revision of the water planetary boundary suggests six sub-boundaries, one of which concerns the role of moisture recycling for maintaining the stability of the atmospheric circulation system. Ongoing work aims to determine the most important distributed hydrologic units for maintaining the stability of the atmospheric circulation system, such as keystone regions which are disproportionately important to the Earth system.
## Conclusions. understanding moisture recycling can nurture water resilience
Moisture recycling and human interactions are only beginning to be understood. Anthropogenic changes to the land surface in one location have significant potential to impact completely different, disconnected regions. Accordingly, we must develop a better scientific understanding and more suitable models for land-atmosphere interactions and regime shifts. This is particularly urgent, as moisture recycling, through its influence on precipitation, has profound implications for humanityfrom agriculture, forestry, and urban water security to regional climatic stability and Earth system resilience.
Tropical forests serve as one of the most critical biomes for maintaining moisture recycling globally, and are thus critical to water resilience. Drylands and rainfed agriculture both exhibit a dependence on moisture recycling but also a complex self-regulatory role that is still not fully understood. Irrigated systems, especially those driven by groundwater, increase short term moisture recycling and perceived resilience, but threaten long-term resilience when societies overdraw groundwater to meet precipitation deficits. Finally, urban areas depend on their upwind hinterlands for stable and predictable sources of precipitation. In this way, moisture recycling further underscores water's role as a master variable for the resilience of human civilization.
Recent research points at multiple possibilities to address moisture recycling in management and governance: introducing the management unit of precipitionshed, considering vegetation-regulated moisture recycling as an ecosystem service, as well as incorporation of moisture recycling in existing water resources tools and concepts. Existing efforts to consider moisture recycling in water footprinting, river basin management, and planetary boundaries are only at the exploration stage, and future work needs to both dive deeper into local-regional contexts and explore potential synergies with other sustainability management frameworks. Humanity has already unintentionally and substantially engineered precipitation patterns through land-use change, and conscious protection of the terrestrial water cycle is now urgently needed to achieve sustainability and build resilience.
[fig] Figure 1: Moisture recycling: (a) the flow of [/fig]
[fig] Figure 3: Dynamics of dryland moisture recycling: (a) the primary importance of recycling of moisture among water in the root zone, transpiration from grasses, to interception by woody vegetation, to transpiration from the woody vegetation, which ultimately leads to (b) intact atmospheric water recycling feedback. The bottom row demonstrates (c) a broken vegetation feedback, which (d) interrupts the connection to the atmospheric moisture recycling feedback. [/fig]
[fig] Figure 4: Conceptual illustration of the precipitationshed. Originally published in[16], and reproduced here under Creative Commons Attribution 3.0 License. [/fig]
|
Cognition Is Associated With Peripheral Immune Molecules in Healthy Older Adults: A Cross-Sectional Study
Background: Cognition in the elderly is heterogeneous. Senescence of the immune system is increasingly considered as a potential player in cognitive performance. We explored here the interplay between cognitive performance and peripheral immune molecules in healthy older individuals.Methods:A cross-sectional study of clinically well characterized senior healthy individuals (120; 51-87 years old) previously clustered as "Good" and "Poor" performers based on established tests that evaluate memory and executive function. A plasma concentration of 30 immune molecules was assessed by multiplex analysis and correlated with parameters of cognitive performance.Results: Participants with worse cognitive performance ("Poor") exhibited increased concentrations of IL-1β, IL-8, IL-13, and tumor necrosis factor (TNF) when compared to individuals with a better cognitive performance ("Good"). The cognitive dimensions memory and executive function, when considered separately, displayed a negative association with several immune molecules (IL-1β, IL-1RA, IL-6, IL-13, IP-10, and TNF with memory and only IL-1β with executive function), even controlling for age, sex, years of formal education, mood, and use of anti-inflammatory drugs. Regression analysis showed that several of these molecules (IL-1β, IL-6, IL-8, and IL-13) contribute to predicting whether an individual belongs to the "Good" or "Poor" cognitive performance group.Conclusion: These results strengthen the hypothesis that increased concentrations of peripheral immune molecules, like IL-1β, are associated with worse cognitive performance in senior healthy individuals. It further highlights that some poorly studied immune molecules should be considered in the context of cognitive aging, such as IL-13, here revealed as a new player in such interaction.
# Introduction
Understanding how peripheral immune mediators interact with the central nervous system (CNS) and influence cognition is of great relevance to provide clues on strategies to reduce or delay aging-associated cognitive decline in an increasingly aged population.
Several studies associate peripheral T lymphocytes and the cytokines they produce with cognitive performance, both in humans and in animal models. Moreover, it was previously shown, in the same cohort of senior individuals studied here, that those displaying worse cognitive performance have a higher number of circulating effector memory (EM) CD4 + T cells. In fact, using regression models, EM CD4 + T cells were shown to contribute in predicting both memory and executive functions, controlling for age, sex, years of formal education, and mood. Interestingly, among T cells, EM CD4 + T cells are major cytokine producers in the periphery and possibly contribute to the increased inflammatory profile (also known as "inflammaging") usually observed with aging. In accordance, an association between increased peripheral inflammatory profile with worse cognitive performances has been evidenced, with few studies addressing more than three immune molecules in cognitively healthy senior cohorts [IL-6 and C-reactive protein (CRP) are most commonly studied]. Interestingly, studies in rodents revealed that bloodborne factors from old mice are able to impair spatial learning and memory as well as neurogenesis and synaptic plasticity in young animals. By contrast, blood from young animals is able to reverse age-related impairments. Even though the precise bloodborne factors that have the capacity to modulate and influence cognitive function are still under investigation, they likely relate to immune cells and/or immune mediators.
Hereupon, the aim of the present study was to perform an extensive analysis of peripheral immune molecules and explore their association with cognitive performance in a senior cognitively healthy population.
# Materials and methods
## Participant characterization
The sample of this study originates from a larger study in which a cohort representative of the older Portuguese population, in terms of sex and education, was selected from two districts in the North of Portugal (Guimarães and Vizela) (n = 1051). These individuals were extensively clinically and cognitively characterized. Cluster analysis identified neurocognitive/psychological performance patterns, and the two extreme groups were termed "Good" and "Poor" cognitive performers. For further characterization, a total sample of 120 subjects was selected, from both the "Good" and "Poor" cognitive performance groups, balanced for sex and age [sample size estimated assuming a two-tailed sample size, a medium effect size (d = 0.5), an alpha of 0.05, a statistical power of 0.8, and an equal sample size for each group]. The socio-demographic and clinical characterization of the participants is presented in. The recruitment and data acquisition took place between March 2012 and March 2013. Participants with incapacity and/or inability to attend the clinical and neuropsychological sessions, diagnosed with cognitive impairment or dementia and/or unable to understand informed consent, with disorders of the CNS or with overt thyroid pathology were not recruited. One participant was a posteriori excluded from the analysis due to impossibility of conducting blood collection.
Considering the high prevalence of cytomegalovirus (CMV) infection in the Portuguese population and the impact of this chronic viral infection in the immune system, the presence of anti-CMV IgG was determined. Only 7 out of the 119 participants were non-immune to CMV; no correlations were observed between CMV antibody titers and the cognitive performance of the participants (data not shown).
## Cognitive characterization
Trained psychologists evaluated the cognitive and mood profile of the participants as previously described. Briefly, the cognitive profile was established using a battery of neurocognitive and psychological tests selected to evaluate short-term verbal memory, verbal working memory, response inhibition/cognitive flexibility, verbal fluency, multiple trial verbal learning and memory, high-level information processing speed, global cognitive status, and mood. Using a principal component analysis, the neurocognitive/psychological test variables were grouped in three-dimensions: memory (MEM), general and executive function (GENEXEC), and mood (Geriatric Depression Scale-GDS). The cognitive groups, identified using cluster analysis, were classified as "Good" and "Poor" cognitive performers. Descriptive information with respect to scores for GDS, MEM, and GENEXEC of the total participants and of the "Good" and "Poor" cognitive performance groups is described in.
## Peripheral immune molecule quantification
Blood was collected in EDTA tubes and processed for plasma collection and for standard hospital biochemical analysis. Plasma was obtained by gradient centrifugation using Histopaque 1077 (Sigma-Aldrich, United States) for 30 min, at room temperature, according to the manufacturer's instructions, and stored at −80 - C. Prior to use, plasma samples were centrifuged at 10,000 g for 10 min at 4 - C to remove platelets and precipitates. Each participant was assigned a code, and all analyses were assessed blindly. Cytokines, chemokines, and other immune molecules were quantified using multiplex magnetic bead- The plates were read using the Bio-Plex MAGPIX TM Multiplex Reader and the data analyzed using the Bio-Plex Manager MP Software (both from Bio-Rad Laboratories, Lda., United States). The lower limit of quantification (LLOQ) and the percentage of detection for each analyte are listed in. The inter-assay coefficient of variation was lower than 4.5%. Samples below LLOQ were extrapolated from the standard curve. The quantification of high-sensitivity C-reactive protein (hsCRP) was conducted at the certified Pathology Laboratory of Braga's Hospital. Due to technical problems, three samples were not quantified for some of the immune molecules measured (sample size is presented in Supplementary.
# Statistical analysis
Only analytes with a percentage of detection above 50% were considered for analysis. Outliers were defined by mean ± 3 standard deviations (SD) and excluded from the sample (Supplementary. Data for MEM, GENEXEC, and GDS were used in the analysis as z-scores, as previously determined. To evaluate normal distribution of the variables, skewness and kurtosis values were calculated and the approximate normal distribution was defined for variables with absolute values of skewness below 3 and of kurtosis below 8. Considering the overall population, all analytes, except IL-6, followed a normal distribution. IL-6 concentration values were log10 transformed to comply with normality. Considering "Good" and "Poor" cognitive performance groups independently, IL-6 and G-CSF did not follow a normal distribution. Levene's test was used to evaluate equality of variances.
To compare the peripheral concentration of immune molecules between "Good" and "Poor, " an independent-sample t-test was performed for variables with normal distribution and a Mann-Whitney U test for variables with non-normal distribution. Significance was considered for p-values equal or below 0.05. To quantify the strength of the differences, Cohen's d was calculated as a measure of effect size (0.2 considered a small effect size, 0.5 a medium effect size, and 0.8 a large effect size).
Linear regression analyses were performed to explore whether the various immune molecules (independently) were able to predict MEM or GENEXEC dimensions (dependent variables) and a binary logistic regression to explore whether the immune molecules were able predict to which cognitive group each individual belongs to ("Good" vs "Poor"), controlling for socio-demographic (age, sex, and school years) and clinical (GDS and anti-inflammatory drugs) variables. Multicollinearity between variables was assessed by the tolerance values (all the variables had a tolerance >0.5). The statistical procedures were performed in IBM SPSS Version 25 (IBM Corp., United States) and the graphs were designed using Prism7 (GraphPad Software, United States).
## Standard protocol approvals, registrations, and patient consents
The study was performed in accordance with the Declaration of Helsinki and approved by ethics review boards (Hospital de Braga, Centro Hospitalar do Alto Ave, and Unidade Local de Sauìde do Alto Minho) and by the national data protection entity (Comissão Nacional de Proteção de Dados). All study goals and nature of the tests were explained, and informed signed consent was obtained from all participants.
# Results
## "poor" and "good" cognitive performers present a distinct peripheral immune molecule profile
From the 31 analytes measured, 19 were detected in more than 50% of the samples (descriptive statistics and percentages of detection of each analyte for both groups are described in. "Poor" cognitive performers had higher plasma concentrations of IL-1β, IL-8, IL-13, and TNF compared to "Good" cognitive performers and detailed statistics FIGURE 1 | Healthy senior individuals with distinct cognitive performances present differences in the concentration of peripheral immune molecules. The profile of cytokines, chemokines, and other immune molecules in the plasma of "Good" (black circles) and "Poor" (red triangles) cognitive performers. Dashed lines represent LLOQ (lower limit of quantification), and values below LLOQ were extrapolated from the standard curve. Dots represent each participant, columns represent the mean of the group, and bars the standard error of the mean (*p < 0.05, **p < 0.01).
in Supplementary Table 2). No differences were observed in the other analytes measured (Supplementary.
As an internal control, it was validated that all participants presented CRP levels below the limit associated with active inflammation/infection (10 mg/L) (12).
## Il-1β is a significant predictor of executive function, memory, and the cognitive clusters
Cognitive performance has been described to be directly, or indirectly, influenced by socio-demographic variables such as age, sex, and years of formal educationand by emotional health. Considering also that the inflammatory profile can be influenced by age (a process known as "inflammaging"), sex, anti-inflammatory drugs, and mood, we next explored the association between cognitive function and the concentration of immune molecules controlling for these variables. Linear regression models were used to explain the variation of executive function (GENEXEC) and memory (MEM) dimension performances and binary logistic regression to discriminate between "Good" and "Poor" cognitive performers using each of the measured plasma molecules independently.
Regarding GENEXEC, only IL-1β, in addition to age, years of formal education, and mood, as previously reported (2), was found as a significant predictor. The model explained approximately 46% (adjusted R 2 ) of the variance observed in the GENEXEC dimension. Altogether, these observations indicate that, in the GENEXEC, "Good" performers have a lower peripheral concentration of IL-1β and, as shown previously, they are also younger, have more years of formal education, and have a better mood than "Poor" performers.
To infer about the increment in the predictive power of the model due to the immune molecules, a hierarchical regression was performed. The first block was composed by variables needed to control for sex, age, school years, mood, and antiinflammatory drugs. The second block was composed by the immune molecules, independently. Data on the concentration of IL-1β increased the predictive power of the model (R 2 change) by 1.8%. Regarding MEM, IL-1β, IL-1RA, IL-6, IL-13, IP-10, and TNF, in addition to age, mood, and number of years of formal education, were found as significant predictors. Models(a model per molecule) explained between 29 and 32% (adjusted R 2 ) of the variance observed in the MEM dimension. Altogether, these observations indicate that "Good" performers in the MEM dimension have a diminished peripheral concentration of some immune molecules. To infer about the increment due to the immune molecules in the predictive power of the models, a hierarchical regression was performed as described for GENEXEC. The levels of peripheral immune molecules increased the predictive power of the model (R 2 change): IL-13 by 5.6%, followed by IL-1RA (3.2%), IL-1β and TNF (2.9%), IP-10 (2.8%), and IL-6 (2.7%). Body fat content can influence the inflammatory profile; however, in this cohort body mass index, whose values range from 19.5 to 38.2 kg per m 2 , did not influence the association between immune molecules and MEM and GENEXEC (data not shown).
The classification as "Good" or "Poor" performers was defined based on the individual scores in GENEXEC and MEM. To examine whether the immune molecules measured in plasma, together with sex, age, years of formal education, mood, and anti-inflammatory drugs, could account for the group delineation ("Good" vs "Poor"), a binary logistic regression was performed. In addition to years of formal education and mood, IL-1β, IL-6, IL-8, and IL-13 (independently) were found as significant predictors. R 2 Nagelkerke values ranged from 0.348 to 0.407, indicating that about 35 to 41% of the chance to belong to a particular cognitive group could be predicted by the model. The correct classifications of the subjects (% of total hit rates), based on the independent variables added in the model, ranged from 69 to 71%. To infer about the increment of the immune molecules in the predictive power of the models, the same rationale of the linear regression models was followed and a hierarchical regression performed. The levels of peripheral immune molecules increased the predictive power of the model (R 2 Nagelkerke change): IL-13 (10.7%), followed by IL-1β (7.4%), IL-6 (4.4%), and IL-8 (3.5%).
Of notice, inflammasome activation has been associated with cognitive symptoms in several human diseasesand with cognitive deficits in autoimmune experimental models. As such, we further investigated inflammasome activation as a possible mechanism behind the low-grade inflammation observed and the cognitive deterioration. To do so, the expression of five inflammasome activation markers (nlrp3, aim2, pycard, casp1, and nlrc4). No differences were detected in the expression levels of these genes between "Good" and "Poor" cognitive performance groups .
## Was analyzed in peripheral blood mononuclear cells from the same participants (supplementary methods and supplementary
# Discussion
This study shows that, in a healthy senior cohort, the individuals with worse cognitive performance have higher plasma concentrations of IL-1β, IL-8, IL-13, and TNF. In addition, several immune molecules present a negative association with both memory (IL-1β, IL-1RA, IL-6, IL-13, IP-10, and TNF) and executive function (IL-1β), even controlling for age, sex, number of years of formal education, mood, and use of anti-inflammatory drugs. Interestingly, some of the immune molecules (IL-1β, IL-6, IL-8, and IL-13) are able to discriminate between "Good" or "Poor" cognitive performers.
Higher plasma concentrations of IL-1β negatively predict both memory and executive function and increase the odds of belonging to the "Poor" cognitive performance group in the regression models. These observations are in line with several previous studies, as follows: (i) IL-1β expression by activated monocytes predicts, negatively, cognitive performance in working memory in healthy senior individuals (21); (ii) IL-1β administration in rodents (both in the CNS and in the periphery) induces hippocampal-dependent memory deficits [reviewed in]. However, some studies found no association or beneficial effect of IL-1β in cognitive function: (i) the serum concentration of IL-1β failed to demonstrate the association with several cognitive domains in healthy senior individuals (4); (ii) higher serum levels of IL-1β were associated with better semantic memory in older women even controlling for age, education, body mass index, and the presence of disease (23); (iii) blocking IL-1β signaling in rodents, both by administering IL-1RA (an IL-1b receptor antagonist) or by knocking out the IL-1b receptor, impaired memory and learning; (iv) IL-1β administration in rodents had no effect or was beneficial on cognitive function [reviewed in].
In addition to IL-1β, other immune molecules were found to associate with a specific cognitive domain: IL-1RA, IP-10, and TNF negatively associate with memory function; high concentrations of IL-8 increase the odds of belonging to the "Poor" performance group; and IL-6 and IL-13 both negatively associate with memory and increase the odds of belonging to the "Poor" performance group.
The role of IL-1RA on cognition has not been extensively addressed, and most of the available evidence originates from animal studies. To our knowledge, the present report is the first to describe a negative association between peripheral IL-1RA and memory in healthy older adults. Interestingly, the serum concentration of IL-1RA was also inversely correlated with cognitive function in a senior population with bipolar disorder; however, it may not necessarily affect cognition through the same mechanisms. Plus, intracerebral injection of IL-1RA, or its overexpression exclusively in the CNS, was shown to cause learning and memory impairments in various rodent models. The overall evidence favors the hypothesis of a negative role of IL-1RA in the modulation of cognition. Most of the literature support that the proinflammatory profile, usually observed in senior individuals, is associated with a worse cognitive profile (4-6). Interestingly, IL-1RA has mainly an anti-inflammatory action [by blocking one of the IL-1β receptors] and seems to be negatively associated with cognition.
IP-10, also known as IFNγ-IP 10 or CXCL10, is secreted by several cell populations in response to IFNγ. Here we observed that IP-10 concentration negatively associates with working memory performance after controlling for age and other confounders [as also shown by others]. More so, IFNγ was below the detection limit for most individuals in our cohort, but, interestingly, IFNγ was identified as a negative regulator of cognitive functioning in rodents.
Regarding TNF, a negative association with memory is here shown. Another study in which TNF was part of a composite score of various immune biomarkers showed a negative association with processing speed, attention, and executive functioning, but not with memory (5). In addition, most reports in healthy human elders (mostly older than 70 years and based on cognitive assessment diverse from ours) found no association between peripheral TNF concentration and cognitive performance. Discordant findings on the role of TNF in cognition emerge as well from rodent studies. Nevertheless, it is possible that the role of TNF on cognition may be age-dependent. In accordance, McAfoose et al.showed that TNF-/-mice perform worse than WT when they are young and better when they are old. Interleukin-8, although poorly studied in the context of cognition, was one of the immune molecules that increased the odds to belong to the "Poor" cognitive performance group in the regression model, which is in accordance with other studies. Still, in a large sample, another report showed no association between IL-8 and cognitive function in nondemented community-dwelling elderly individuals.
Interleukin-6 is one of the best-studied cytokines in aging. The vast majority of studies report a negative association between peripheral concentration of IL-6 and cognitive performance, mainly with measures of executive function, and not of memory. In this study, we show no association between the concentration of IL-6 and executive function, but higher levels of IL-6 associate with worst memory and increase the odds of belonging to the "Poor" cognitive performance group. Of note, in several other studies it has also been shown to predict cognitive decline. The majority of the literature, including ours, associates an increase in peripheral IL-6 with worst memory function in seniors [although two studies reported no association].
To our knowledge, only one publication indicated that IL-13 deficiency in rodents significantly impairs working and reference memory. Here, we observed the opposite-IL-13 not only negatively associates with memory but also increases the odds of belonging to the "Poor" cognitive performance group. These results highlight the need to perform more studies using cohorts composed by healthy aging individuals and addressing less explored immune molecules like IL-13 and others.
The major study strengths are its extensive evaluation of peripheral immune molecules and the cohort being composed by extremely well-characterized healthy seniors; the comprehensive neuropsychological characterization assessing two distinct cognitive domains (memory and executive function); and the control for active inflammatory responses (through quantification of CRP) and for the use of anti-inflammatory drugs. Nonetheless, the cross-sectional design does not allow causal conclusions about the relationship between peripheral profile of immune molecules and age-associated cognitive decline.
A note on the inconsistencies observed between human studies is warranted. These may possibly occur due to differences in selection, characterization, and inclusion/exclusion criteria. For instance, the age ranges and the sample sizes vary greatly, as well as in the diversity in the cognitive evaluation tools and in the concentration range of the immune molecules measured. More so, not all studies check or control for possible active inflammatory processes or the intake of anti-inflammatory drugs. CRP, an acute phase protein, is vastly studied in geriatric populations and has been associated with an increased risk for cerebrovascular disease, Alzheimer's disease, and vascular dementia. Some studies described no association between levels of CRP and cognitive performance in healthy senior individualsas we do here, while others demonstrate an association between higher levels of CRP and worst cognitive performance. Interestingly, in those studies the CRP range is lower than in this study, indicating that other factors that vary between cohorts may mediate or dilute its impact on cognition.
# Conclusion
In summary, this study presents an extensive evaluation of peripheral immune molecules in a healthy senior population comprehensively characterized for cognitive performance, allowing a broader perspective on the association between the peripheral profile of immune molecules and cognitive performance. The data support the hypothesis that the peripheral profile of immune molecules at older ages is related with cognitive performance. Furthermore, it reinforces the pertinence to explore the possible interactions between cytokines in the modulation of cognitive function. It seems that an overall increased concentration of immune molecules, both pro-and anti-inflammatory, associates with stronger cognitive deterioration during aging. Additionally, a reasonable set of these studies, preferentially performed in distinct geographic regions, are necessary to feed secondary research, like meta-analysis, to be able to integrate the data from independent studies and clearly define the immune molecules that coherently associate with cognitive functions. Furthermore, the precise mechanisms behind the impact of peripheral immune mediators in the cognitive function are still unknown. Animal models are likely to shed some light on those mechanisms.
Understanding how the immune mediators interact with the CNS and influence cognition is of great relevance to provide clues on potential strategies to reduce or delay aging-associated cognitive decline in an increasingly aged population.
# Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation to any qualified investigator.
# Ethics statement
The studies involving human participants were reviewed and approved by ethics review boards (Hospital de Braga, Centro Hospitalar do Alto Ave, and Unidade Local de Sauìde do Alto Minho) and by the national data protection entity (Comissão Nacional de Proteção de Dados). The patients/participants provided their written informed consent to participate in this study.
# Author contributions
CS-M had a major role in acquisition of data, performed the statistical analysis of the data, interpreted the data, and wrote the first draft of the manuscript. SR and NS designed and conceptualized the study and revised the manuscript for intellectual content. NCS established the cohort and revised the manuscript for intellectual content. PC participated in establishment of the cohort, assessment, and statistic advisory. JP and MC-N designed and conceptualized the study, interpreted the data, and revised the manuscript for intellectual content. All authors contributed to the article and approved the submitted version. |
Functional Integration Between the Two Brain Hemispheres: Evidence From the Homotopic Functional Connectivity Under Resting State
Functional integration among neural units is one of the fundamental principles in brain organization that could be examined using resting-state functional connectivity (rs-FC). Interhemispheric functional integration plays a critical role in human cognition. Homotopic functional connectivity (HoFC) under resting state provide an avenue to investigate functional integration between the two brain hemispheres, which can improve the present understanding of how interhemispheric interactions affect cognitive processing. In this review, we summarize the progress of HoFC studies under resting state and highlight how these findings have enhanced our understanding of interhemispheric functional organization of the human brain. Future studies are encouraged to address particular methodological issues and to further ascertain behavioral correlates, brain disease's modulation, task influence, and genetic basis of HoFC.
# Introduction
Brain activity globally interacts at multiple levels with functional integration ranging from neural units to brain regions to overall behavioral output, leading to the emergence of widespread patterns of correlations among neuronal groups [bib_ref] A measure for brain complexity: relating functional segregation and integration in the..., Tononi [/bib_ref]. How the left and right hemispheres are functionally integrated is a crucial issue in neuroscience. For the two hemispheres, functional integration refers to the interaction between specialized regions, enabling interhemispheric longrange synchronization and information flow . Such functional integration has been found to play a vital role in cognitive processing, such as visuospatial attention, speech perception and creative thinking [bib_ref] Two distinct forms of functional lateralization in the human brain, Gotts [/bib_ref] [bib_ref] Modulation of intra-and inter-hemispheric connectivity between primary and premotor cortex during speech..., Nuttall [/bib_ref] [bib_ref] Brain hemispheric involvement in visuospatial and verbal divergent thinking, Chen [/bib_ref].
Various neuroimaging techniques have been applied to investigate functional integration between the two brain hemispheres. For example, [bib_ref] Intercorrelations of glucose metabolic ates between brain regions: application to healthy males..., Horwitz [/bib_ref] used regional metabolic rates acquired by positron emission tomography (PET) to characterize the functional integration among different brain regions. They found that all regions were significantly correlated with their contralateral homologs. Likewise, [bib_ref] Effects of age upon interhemispheric EEG coherence in normal adults, Duffy [/bib_ref] inspected interhemispheric functional integration via resting-state electroencephalogram (EEG) recordings and revealed synchronized electrical activity between the left and right hemispheres. With the rapid development of functional magnetic resonance imaging (fMRI) in recent years, resting-state fMRI (rs-fMRI) has been employed to explore interhemispheric functional integration more frequently, given its advantages such as relatively high spatial and temporal resolution without any invasion over other neuroimaging techniques [bib_ref] Functional connectivity in the motor cortex of resting human brain using echo-planar..., Biswal [/bib_ref]. The rs-fMRI provides a powerful method to investigate functional integration between different brain regions by calculating restingstate functional connectivity (rs-FC) that is deemed to reflect the interactions across different cerebral regions [bib_ref] Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging, Fox [/bib_ref]. By means of the interhemispheric rs-FC, we could directly quantify functional integration between the two brain hemispheres and thus have a great chance to determine how interhemispheric functional integration affects cognitive processing [bib_ref] Low frequency fluctuations reveal integrated and segregated processing among the cerebral hemispheres, Gee [/bib_ref].
rs-FC analyses to detect patterns of functional integration between the two brain hemispheres were firstly applied in the motor system [bib_ref] Functional connectivity in the motor cortex of resting human brain using echo-planar..., Biswal [/bib_ref]. Since then, rs-fMRI studies have revealed whole-brain patterns of synchronized spontaneous activity between homotopic regions in left and right hemispheres, which is referred to as homotopic functional connectivity (HoFC) [bib_ref] Neurophysiological architecture of functional magnetic resonance images of human brain, Salvador [/bib_ref] [bib_ref] Regional variation in interhemispheric coordination of intrinsic hemodynamic fluctuations, Stark [/bib_ref]. As demonstrated, HoFC is a robust feature with a peculiar interior distribution corresponding to the functional hierarchy, which does not follow the general law defined by the spatial distance [bib_ref] Neurophysiological architecture of functional magnetic resonance images of human brain, Salvador [/bib_ref] [bib_ref] Regional variation in interhemispheric coordination of intrinsic hemodynamic fluctuations, Stark [/bib_ref] [bib_ref] A surface-based technique for mapping homotopic interhemispheric connectivity: development, characterization, and clinical..., Tobyne [/bib_ref]. Meanwhile, it has been found to correlate with the performance of cognitive tasks, such as visuospatial attentionand executive function [bib_ref] Agerelated decreases in interhemispheric resting-state functional connectivity and their relationship with executive..., Zhao [/bib_ref]. As summarized in [fig_ref] TABLE 1 |: Basic information for published rs-fMRI studies of HoFC [/fig_ref] , a number of studies have achieved progress in functional integration between the two brain hemispheres by virtue of HoFC measurements under resting state.
This review aims to summarize the studies of functional integration between the two brain hemispheres by using HoFC under resting state and to highlight how the findings have enhanced our understanding of hemispheric functional organization in the human brain. For the summary of existing literatures, we included six sections: (1) Methods of Calculating HoFC, (2) Physiological Effects on HoFC, (3) Structural Foundation of HoFC, (4) Correlation Between HoFC and Cognition, (5) HoFC Abnormalities in Brain Diseases, and (6)
## Hofc and interhemispheric functional integration
In this part, we review relevant rs-fMRI findings of functional integration between the two brain hemispheres by using HoFC.
## Methods of calculating hofc
As shown in , HoFC is calculated by Pearson's correlation between the time series of homotopic locations under resting state, which is the most important way to reflect functional integration between the two brain hemispheres. In early stages, homotopic locations are defined as corresponding anatomical areas or spatial locations in opposite hemispheres [bib_ref] Neurophysiological architecture of functional magnetic resonance images of human brain, Salvador [/bib_ref] [bib_ref] Regional variation in interhemispheric coordination of intrinsic hemodynamic fluctuations, Stark [/bib_ref] [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref]. Recently, researchers proposed that each brain region has a unique homotopic contralateral counterpart with a maximal rs-FC [bib_ref] AICHA: an atlas of intrinsic connectivity of homotopic areas, Joliot [/bib_ref]. Specifically, HoFC is computed by using different methodological definitions with various spatial scales, such as regions of interest (ROIs), voxel-level in volume, or vertex-level on cortical surface. Below, we summarized different methods to obtain HoFC.
## Atlas-based hofc
The majority of early studies adopted anatomical parcellations to define homotopic locations in the left and right hemispheres. Initially, rs-fMRI studies mostly concentrated on a specific cortical or subcortical brain region and have observed a high level of interhemispheric correlation in those regions [bib_ref] Functional connectivity in the motor cortex of resting human brain using echo-planar..., Biswal [/bib_ref] [bib_ref] Functional connectivity in single and multislice echoplanar maging using restingstate fluctuations, Lowe [/bib_ref] [bib_ref] Mapping functionally related regions of brain with functional connectivity MR imaging, Cordes [/bib_ref]. [bib_ref] Neurophysiological architecture of functional magnetic resonance images of human brain, Salvador [/bib_ref] first noticed that synchronous activity across homotopic regions was a ubiquitous phenomenon among the whole brain. This first whole-brain analysis based on human rs-fMRI data calculated the HoFC via the automated anatomical labelling (AAL) atlas [bib_ref] Neurophysiological architecture of functional magnetic resonance images of human brain, Salvador [/bib_ref] , which is proposed by [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref] and comprises 45 corresponding anatomical ROIs in each cerebral hemisphere. Based on AAL atlas as well, [bib_ref] Stable long-range interhemispheric coordination is supported by direct anatomical projections, Shen [/bib_ref] built on their work under resting state and found HoFC was significantly higher than heterotopic or intrahemispheric FC, validated in both human and macaque datasets [bib_ref] Stable long-range interhemispheric coordination is supported by direct anatomical projections, Shen [/bib_ref]. In addition, they discovered HoFC was more stable and resistant across both varied conditions and temporal changes than heterotopic or intrahemispheric FC [fig_ref] FIGURE 2 |: The basic characteristics and structural foundation of HoFC [/fig_ref]. [bib_ref] Regional variation in interhemispheric coordination of intrinsic hemodynamic fluctuations, Stark [/bib_ref] used the Harvard-Oxford Structural Atlas to divide the entire cerebrum into 112 homotopic regions, 56 in each brain hemisphere. They reported a pattern of regional variation in temporally correlated lowfrequency brain activity between the two hemispheres, suggesting that interhemispheric coordination may vary across homotopic regions in the human brain. As shown in [fig_ref] FIGURE 2 |: The basic characteristics and structural foundation of HoFC [/fig_ref] , despite interhemispheric HoFC across the whole brain regions was robust, higher HoFC value was more exhibited in primary sensorimotor cortices than in higher-order association areas. In recent years, studies have suggested that it would be more reasonable and advantageous for the investigation of cerebral functional organization by using homotopic ROIs derived from functional parcellation [bib_ref] AICHA: an atlas of intrinsic connectivity of homotopic areas, Joliot [/bib_ref] [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref]. Therefore, [bib_ref] AICHA: an atlas of intrinsic connectivity of homotopic areas, Joliot [/bib_ref] developed a functional parcellation based on the intrinsic connectivity of homotopic areas, namely the Atlas of Intrinsic Connectivity of Homotopic Areas (AICHA). This atlas covers the entire cerebral gray matter and includes 192 homotopic region pairs, which has already been used in studies related to HoFC [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref].
## Voxel-based hofc
In order to investigate the HoFC on a finer scale, [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref] provided a comprehensive examination of HoFC by using a voxel-wise measure called voxel-mirrored homotopic connectivity (VMHC). Specifically, the individual rs-fMRI data were registered to a symmetrical brain template in Montreal Neurological Institute (MNI) space, which was generated by averaging the group-specific mean normalized T1 image and its left-right mirrored image. This registration step roughly ensures each voxel in one hemisphere has a homotopic voxel in contralateral hemisphere. Then, the Pearson's correlation coefficient between the resting-state time series of each voxel and its symmetrical interhemispheric counterpart was calculated as the VMHC. By using this method, they also verified consistent regional variation of HoFC with [bib_ref] Regional variation in interhemispheric coordination of intrinsic hemodynamic fluctuations, Stark [/bib_ref]. In contrast to the atlas-based HoFC, the VMHC approach allows a greater and finer appreciation of regional variation within larger cortical structures, such as the cingulate, precentral and postcentral gyri [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref].
## Surface-based hofc
To achieve a better alignment of homologous location between the two brain hemispheres, a surface-based HoFC can be used [bib_ref] Quantifying agreement between anatomical and functional interhemispheric correspondences in the resting brain, Jo [/bib_ref]. Using the landmark-based anatomical correspondence, [bib_ref] Quantifying agreement between anatomical and functional interhemispheric correspondences in the resting brain, Jo [/bib_ref] first putatively established vertex-wise homotopic FIGURE 1 | A schematic flowchart of the definition of homotopic functional connectivity (HoFC). The homotopic pair could be mirrored voxels or vertices or from an atlas with defined homotopic regions. For a given homotopic pair, interhemispheric HoFC is calculated as the Pearson's correlation between the time series of this location and its homotopic contralateral counterpart. Finally, a whole-brain vector of HoFC is obtained for each subject.
locations between the two hemispheres on a non-symmetric template brain. They identified the improved precision of this surface-based approach over the previous volume-based flipping method. Following this study, [bib_ref] A surface-based technique for mapping homotopic interhemispheric connectivity: development, characterization, and clinical..., Tobyne [/bib_ref] further introduced a computationally efficient technique, surface-based HoFC, that leveraged surface-based registration to improve the spatial specificity and accuracy of cortical HoFC under resting state. The surface-based HoFC for each pair of homologous vertices was calculated by using the pairwise Pearson's correlation between the extracted time series. This technique is presented as a sensitive measure of HoFC in the human brain and is proved to be reliable both within and across subjects [bib_ref] A surface-based technique for mapping homotopic interhemispheric connectivity: development, characterization, and clinical..., Tobyne [/bib_ref]. Particularly, this method presents a more constrained pattern of HoFC along the medial surface. As noted, either the volumetric smoothing or greater accuracy of spherical registration techniques might lead to discrepancies between volume-based and surface-based HoFC [bib_ref] A surface-based technique for mapping homotopic interhemispheric connectivity: development, characterization, and clinical..., Tobyne [/bib_ref].
## Physiological effects on hofc
The human brain is affected by a number of physiological factors, such as age and sex. [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref] systematically investigated age-related HoFC changes under resting state by using VMHC in 214 healthy individuals (age range: 7-85 years). As shown in [fig_ref] FIGURE 3 |: The physiological effects on HoFC [/fig_ref] , sensorimotor regions displayed a HoFC increase, whereas higher-order association regions showed a HoFC decrease along with age. In addition, they detected complex maturational curves, such as the quadratic trajectories of insula and lingual gyrus as well as the cubic trajectories of putamen and superior frontal gyrus. Furthermore, sex-related differences in HoFC between males and females were discovered, with males having weaker HoFC in the medial and dorsal regions as well as greater HoFC in the parahippocampal and fusiform gyri [fig_ref] FIGURE 3 |: The physiological effects on HoFC [/fig_ref]. Moreover, there were interactions between age and sex on resting-state HoFC in the dorsolateral prefrontal cortex and amygdala. In a recent study, [bib_ref] Agerelated decreases in interhemispheric resting-state functional connectivity and their relationship with executive..., Zhao [/bib_ref] examined VMHC in 93 healthy volunteers between 19 and 85 years of age and observed reductions in VMHC of the ventromedial prefrontal cortex, inferior parietal lobule, dorsal anterior cingulate cortex, hippocampus and insula along with age. In addition to these VMHC approaches, [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref] used the functionally driven homotopic atlas (AICHA) to define the homotopic regions of the human brain. They recruited 291 individuals (age range: 18-57 years; 148 men, 143 women) and found that mean HoFC (averaged across the 36 pairs of lateral homotopic regions) significantly decreased with age and was larger in males than females. Together, HoFC in lifespan indicated a global decrease related with age [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref] [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref]. In specific regions, Zhao's study found negative correlation, while Zuo's study reported various development trajectories [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref] [bib_ref] Agerelated decreases in interhemispheric resting-state functional connectivity and their relationship with executive..., Zhao [/bib_ref]. The discrepancies might be due to different statistical models in their studies, with a simple GLM in Zhao's work and a more complicated one including both quadratic and cubic terms in Zuo's work. In [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref] , the age distribution is skewed, which could affect the results of relationship between age and HoFC. Unlike the three studies above, Gracia-Tabuenca et al. (2018) focused on typically developing children ranging in age from 6 to 10 years. In a sample of 60 participants, significant sex effects were noted in the frontal pole, but no age effects were identified.
In infant group, [bib_ref] Longitudinal study of the emerging functional connectivity asymmetry of primary language regions..., Emerson [/bib_ref] tried to portray the development curve of interhemispheric functional asymmetry during the first 2 years of life, by leveraging a large sample of human infants with longitudinal rs-fMRI scans. The interhemispheric HoFC exhibited three predominant patterns, including one trajectory for sensory regions, one trajectory for higher-order association regions and another trajectory for language-related regions. Concentrated on an earlier stages from 26 weeks postmenstrual age (PMA), [bib_ref] Longitudinal analysis of neural network development in preterm infants, Smyser [/bib_ref] obtained resting-state networks based on rs-FC data of preterm infants, and they found the networks with more mature forms in older infants consisted of localized interhemispheric HoFC between homotopic counterparts, demonstrating a regionally age-specific development pattern.
## Structural foundation of hofc
Like other rs-FC, HoFC is likely constrained by structural connections of the brain [bib_ref] Predicting human resting-state functional connectivity from structural connectivity, Honey [/bib_ref]. Anatomically, the corpus callosum (CC) is the major neural pathway that connects the homotopic regions of the left and right hemispheres [bib_ref] Postnatal development of corticocortical connections, Innocenti [/bib_ref] and putatively plays a pivotal role in mediating HoFC under resting state [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref].
This line of research was explored mainly by using patients with CC defects. In 2003, [bib_ref] Role of the corpus callosum in functional connectivity, Quigley [/bib_ref] found that patients with agenesis of the corpus callosum (AgCC) showed diminished interhemispheric functional connectivity in the motor and auditory cortices compared with healthy participants. [bib_ref] Resting-State Networks and the Functional Connectome of the Human Brain in Agenesis..., Owen [/bib_ref] extracted 17 resting-state networks of patients with AgCC (including partial and complete) and well-matched healthy controls. They detected a striking loss of interhemispheric HoFC of the insularopercular regions, posterior cingulate cortex, and precuneus in AgCC subjects. [bib_ref] Loss of resting interhemispheric functional connectivity after complete section of the corpus..., Johnston [/bib_ref] presented a rs-FC study of a child patient with medically refractory epilepsy both before and after complete section of the CC, which demonstrated reduced interhemispheric HoFC and preserved intrahemispheric connectivity postoperatively. Further, [bib_ref] On the role of the corpus callosum in interhemispheric functional connectivity in..., Roland [/bib_ref] evaluated rs-FC in 22 patients with intractable epilepsy after surgical section of the CC; they found markedly reduced interhemispheric HoFC in the human brain, which was more profound in multimodal associative areas than in primary regions. The existence of polysynaptic FC between distant cerebral regions was inferred by comparison of partial vs. complete callosotomy [fig_ref] FIGURE 2 |: The basic characteristics and structural foundation of HoFC [/fig_ref]. These results together indicated that callosal anatomical connections played a critical role in the existence and maintenance of interhemispheric HoFC, which provided crucial insights into general relationship between structural connection and functional connectivity in the human brain. These approaches, however, were unable to characterize interindividual differences of healthy people because of the particularity of subjects with CC defects, although they did have relatively high biological specificity and interpretability.
A large number of studies have applied diffusion magnetic resonance imaging (dMRI) to study structural connections of the human brain non-invasively. Using dMRI and rs-fMRI data in 11354 subjects from the UK Biobank, [bib_ref] The spatial correspondence and genetic influence of interhemispheric connectivity with white matter..., Mollink [/bib_ref] focused on interhemispheric connectivity and corresponding commissural fibers passing through the CC between pairs of homotopic cerebral regions. They constructed multivariate models to link interhemispheric structural connectivity and functional connectivity, showing that ninety percent of human brain regions were statistically related between HoFC under resting state and white matter microstructure of CC.
In summary, the results both in patients and healthy people provide strong evidence supporting that corticocortical CC connectivity serves as pivotal structural basis underlying the resting-state HoFC between homotopic cortical regions of the two hemispheres.
## Correlation between hofc and cognition
As an important representation of functional integration between the two brain hemispheres, HoFC under resting state was found to be significantly correlated with human's cognitive activities, such as visuospatial attention and executive function.discovered that the VMHC of the occipital fusiform gyrus and the lingual gyrus was negatively associated with the performance in visual long-term memory tasks. In addition, Zhao's study found that the VMHC of the ventromedial prefrontal cortex and insula showed a negative linear relationship with the performance on the color-word inhibition task mainly about executive function [bib_ref] Agerelated decreases in interhemispheric resting-state functional connectivity and their relationship with executive..., Zhao [/bib_ref].
## Hofc abnormalities in brain diseases
A number of studies have made direct comparisons of HoFC between different patient groups and healthy controls, most of which utilized VMHC to represent HoFC.
Compared to healthy controls, some studies have demonstrated increased HoFC in patient groups, such as idiopathic generalized epilepsy [bib_ref] Increased interhemispheric resting-state in idiopathic generalized epilepsy with generalized tonic-clonic seizures: a..., Yang [/bib_ref] and paroxysmal kinesigenic dyskinesia [bib_ref] Increased interhemispheric resting-state functional connectivity in paroxysmal kinesigenic dyskinesia: a resting-state fMRI..., Ren [/bib_ref] , while others reported decreased HoFC, ranging from cocaine addiction , schizophrenia [bib_ref] Decreased interhemispheric coordination in schizophrenia: a resting state fMRI study, Hoptman [/bib_ref] , Parkinson's disease [bib_ref] Decreased resting-state interhemispheric functional connectivity in Parkinson's disease, Luo [/bib_ref] , major depressive disorder (MDD) [bib_ref] Major depressive disorder: findings of reduced homotopic connectivity and investigation of underlying..., Hermesdorf [/bib_ref] , to amnestic mild cognitive impairment [bib_ref] Decreased bilateral FDG-PET uptake and inter-hemispheric connectivity in multi-domain amnestic mild cognitive..., Luo [/bib_ref]. Some brain diseases, including HBV-related cirrhosis [bib_ref] Aberrant interhemispheric coordination characterizes the progression of minimal hepatic encephalopathy in patients..., Ye [/bib_ref] and subacute stroke [bib_ref] Altered static and dynamic voxel-mirrored homotopic connectivity in subacute stroke patients: a..., Chen [/bib_ref] , showed both increases and decreases of HoFC in different brain regions. Although majority of results are consistent across studies, discrepancies exist. For example, several studies have used VMHC under resting state to explore the difference between MDD patients and healthy people [bib_ref] Decreased interhemispheric resting-state functional connectivity in first-episode, drugnaive major depressive disorder, Guo [/bib_ref] [bib_ref] Decreased interhemispheric coordination in the posterior default-mode network and visual regions as..., Guo [/bib_ref] [bib_ref] Major depressive disorder: findings of reduced homotopic connectivity and investigation of underlying..., Hermesdorf [/bib_ref] [bib_ref] Analysis of voxelmirrored homotopic connectivity in medication-free, current major depressive disorder, Fan [/bib_ref]. Compared to healthy controls, most studies found no VMHC increase and a significant VMHC decrease in the cuneus, posterior cingulate cortex and precuneus in MDD patients. However, [bib_ref] Major depressive disorder: findings of reduced homotopic connectivity and investigation of underlying..., Hermesdorf [/bib_ref] reported a VMHC decrease in the putamen, superior temporal gyrus and insula, which exhibited a wider brain region change in the VMHC in patients with MDD than other studies. These discrepancies might be due to the differences among studies in inclusion criteria and the number of participants, fMRI acquisition, analysis methods, and so on. Regarding the relationship with behavior and symptoms, the HoFC under resting state in specific brain regions was found to be associated with scores of disease-related scales in patient groups. For instance, [bib_ref] Reduced interhemispheric resting state functional connectivity in cocaine addiction, Kelly [/bib_ref] revealed a correlation between HoFC within the regions in the dorsal attention network and self-reported lapses of attention in the cocaine-dependent group. For schizophrenia, [bib_ref] Decreased interhemispheric coordination in schizophrenia: a resting state fMRI study, Hoptman [/bib_ref] showed that HoFC in the postcentral gyrus extending into the precentral gyrus was associated with Positive and Negative Syndrome Scale (PANSS) total scores.
## The role of hofc in functional lateralization
In addition to the functional integration between the two brain hemispheres, another key organization principle of the human brain is functional lateralization. Functional lateralization is acquired by comparing the brain activity patterns between a region in one hemisphere and its contralateral homotopic counterpart in the other [bib_ref] Laterality Patterns of Brain Functional Connectivity: Gender Effects, Tomasi [/bib_ref] [bib_ref] An evaluation of the left-brain vs. right-brain hypothesis with resting state functional..., Nielsen [/bib_ref]. Previous studies have reported that HoFC is related to functional lateralization. [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref] investigated this relationship in 297 healthy volunteers, including 153 left-handers (LHs). They found a significant negative correlation between resting-state HoFC and language task-induced functional lateralization, suggesting a translation from enhanced interhemispheric cooperation in the resting state into increased interhemispheric cooperation during the language production. In addition, all the subjects in this study were classified into three different types according to functional lateralization of language task: typicals, strong-atypicals (only LHs) and ambilaterals. A handedness effect on HoFC was also discovered by the lateralization subgroup comparison, with left-handed ambilaterals having significantly greater HoFC than the other groups.also examined these two brain functional properties, both defined based on rs-FC. Their results identified that HoFC was strongly negatively correlated with functional lateralization across the whole human brain.
Combined with previous findings of the fundamental role of CC, HoFC under resting state might reflect the effect of CC on functional lateralization [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref]. In fact, many studies have demonstrated the excitatory role of CC in interhemispheric processing [bib_ref] Individual variability in cortical organization: Its relationship to brain laterality and implications..., Galaburda [/bib_ref]. The CC may enforce cerebral processing integration between the left and right hemispheresand activate the unstimulated hemisphere [bib_ref] Functional significance of individual variations in callosal area, Yazgan [/bib_ref]. According to this hypothesis, more symmetric brains have stronger interhemispheric connections, indicating a correlation between a smaller CC and greater lateralization [bib_ref] The role of the corpus callosum in interhemispheric transfer of information: excitation..., Bloom [/bib_ref]. The significant negative correlations between HoFC and functional lateralization in studies above [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref] therefore provided supporting evidence for the excitatory role of the CC in interhemispheric processing in the human brain.
## Future considerations
We will elaborate several specific aspects about HoFC in the following sections, which deserve particular consideration in future studies.
## Lifespan trajectory of hofc
By means of a range of neuroimaging methods, complex changes in the structure and function of the human brain across the lifespan are becoming accessible for analysis [bib_ref] Topological organization of the human brain functional connectome across the lifespan, Cao [/bib_ref]. As early as the fetal period during the second half of gestation, many human brain networks, involving both corticocortical and thalamocortical connections, are initially appeared and established [bib_ref] The development of cerebral connections during the first 20-45 weeks' gestation, Kostović [/bib_ref]. Findings from infants using HoFC under resting state suggested that the developing human brain might increase its interhemispheric functional symmetry to govern multiple functional systems, which might be a general principle in the first year of life [bib_ref] Longitudinal analysis of neural network development in preterm infants, Smyser [/bib_ref] [bib_ref] Longitudinal study of the emerging functional connectivity asymmetry of primary language regions..., Emerson [/bib_ref]. Therefore, the exploration of functional integration between the two brain hemispheres should also start at earlier stages of brain development such as the fetal or neonatal stage. In most existing lifespan HoFC studies, the age range of subjects were disequilibrium and mostly located at young adults group [bib_ref] Growing together and growing apart: regional and sex differences in the lifespan..., Zuo [/bib_ref] [bib_ref] Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of..., Tzourio-Mazoyer [/bib_ref] [bib_ref] Agerelated decreases in interhemispheric resting-state functional connectivity and their relationship with executive..., Zhao [/bib_ref]. In future studies, it is warranted to map out lifespan trajectories of HoFC for different homologous regions, using a uniformly distributed lifespan cross-sectional or longitudinal data.
## Genetic basis of hofc
Many studies have shown that resting-state FC and networks are considerably heritable, suggesting that HoFC might be influenced by genetic factors [bib_ref] Genetic control over the resting brain, Glahn [/bib_ref] [bib_ref] Heritability analysis with repeat measurements and its application to restingstate functional connectivity, Ge [/bib_ref]. As the structural foundation of HoFC between the two brain hemispheres, the CC is found highly heritable as well. For example, previous studies have confirmed both the size and microstructure of the CC are strongly controlled by genetic factors [bib_ref] Genetics of microstructure of the corpus callosum in older adults, Kanchibhotla [/bib_ref] [bib_ref] Corpus callosum size is highly heritable in humans, and may reflect distinct..., Woldehawariat [/bib_ref] [bib_ref] Heritability of white matter microstructure in late middle age: a twin study..., Vuoksimaa [/bib_ref]. To date, it remains unexplored whether and how genetic factors influence HoFC. In the future, it would be interesting to investigate the heritability of HoFC at multiple levels by employing advanced imaging genetics techniques, such as twin study with ACE model, genome-wide association study (GWAS) or gene set enrichment analysis (GSEA). The findings would help to understand the relationship between interhemispheric functional integration and genetic factors, as well as environmental factors.
## Hofc under tasking state
Compared to rs-fMRI, tasking-state fMRI (ts-fMRI) has been largely used to identify functional activities evoked by specific cognitive tasks. Measurements of brain activity actually reflect a mixture of spontaneous and evoked activities during a specific task. Recently, more and more studies started focusing on tasking-state functional connectivity (ts-FC). Many functional organization properties of the human brain under resting state have been recognized during the tasking state. Across dozens of different tasks, [bib_ref] Intrinsic and task-evoked network architectures of the human brain, Cole [/bib_ref] identified a whole-brain network architecture, which was highly similar to the one presented in resting state. In addition, there was a small but consistent group of changes across different tasks, suggesting the existence of a task-general network architecture distinguished from the resting state. These results indicated that the functional architecture of the human brain under tasking state was likely composed of an intrinsic architecture that appeared under resting state and secondarily shaped by evoked task-general and task-specific changes. In the future, the similarities and differences of HoFC between the resting state and tasking state can be thoroughly explored, which may enhance the understanding about the relationship between the two functional states of the human brain, from an interhemispheric functional integration perspective.
## The functional meaning of hofc to cognition in healthy people
To better understand cognitive consequence of functional integration between the two brain hemispheres, more studies are encouraged to explore the functional meaning of HoFC in human brain. As reviewed above, HoFC has been found to be related with particular cognitions, behaviors and specific symptom in both healthy and diseases. In healthy people, studies revealed only a few cognitions related with regional HoFC under resting state, such as executive function, visuospatial attention, language, etc. The association with cognition was more frequently reported in brain disease than healthy people, which might due to the limited individual difference of HoFC in healthy people.
Currently, some available public data, such as the Human Connectome Project (HCP) or UK biobank, provide both neuroimaging and behavior data in a large sample of normal people, allowing for thorough investigation of brain-cognition relationships with a strong statistical power. More studies are encouraged to directly investigate the HoFC-cognition associations in healthy people. By studying these large sample datasets alone or combining with them, researchers may capture subtle associations between the HoFC and complex cognitions/behaviors. Advanced analysis techniques are encouraged for the investigation of HoFC-cognition relationship as well, such as multivariable correlation or machine learning.
# Conclusion
This article reviewed existing studies of HoFC under resting state that reflects to some degree functional integration between the two brain hemispheres. Future studies are encouraged to better characterize lifespan trajectories, genetic basis, taskingstate modulation, cognitive correlates of HoFC.
[fig] FIGURE 2 |: The basic characteristics and structural foundation of HoFC. (A) The condition variability and temporal stability for different types of functional connectivity, including intrahemispheric (i), heterotopic (he), and homotopic (ho). The figure is adapted from Shen et al. (2015). Permission to reuse Figures 2, 3 from the article was granted by PNAS. (B) Mean interhemispheric HoFC indicated for all cortical regions. The figure is adapted from Stark et al. (2008). Permission to reuseFigure 4from the article was granted by Society for Neuroscience("Copyright [2008] Society for Neuroscience"). (C) HoFC computed as the Fisher Z-transformed VMHC before (first and third rows) and after (second and forth rows) corpus callosotomy (complete and partial). The figure is adapted fromRoland et al. (2017). Permission to reuseFigure 4from the article was granted by PNAS. [/fig]
[fig] FIGURE 3 |: The physiological effects on HoFC. (A) Age effects on HoFC. The linear (Lin), quadratic (Qua), and cubic (Cub) effects of age on VMHC were shown at each voxel, evaluating through multiple linear regressions model. (B) Sex effects of HoFC. One-sample tests on regression coefficients of sex was performed for both positive and negative contrasts. These figures are adapted from Zuo et al. (2010). Permission to reuse Figures 3, 5 from the article was granted by the Creative Commons Attribution-Non-commercial-Share Alike 3.0 Unported License (CC-BY-NC-SA). [/fig]
[table] TABLE 1 |: Basic information for published rs-fMRI studies of HoFC. [/table]
|
The popliteomeniscal fascicles: from diagnosis to surgical repair: a systematic review of current literature
Background: Popliteomeniscal fascicles (PMF) are considered the posterolateral meniscocapsular extensions which connect the lateral meniscus to the edge of the tibia. PMFs disruption leads to hypermobility of the lateral meniscus with pain and locking sensation. Recognition and treatment of PMFs tear remain very challenging. The aim of this systematic review is to collect and analyse the articles concerning popliteomeniscal fascicle disruption from diagnosis to surgical approach. Methods: PubMed, Scopus, Web of Science and EMBASE were searched. Various combinations of the keywords "Popliteomeniscal Fascicles", "Lateral Meniscus", "Popliteal Hiatus", "Posterolateral Corner", "Tear" and "Surgical Repair" were used. The original literature search identified a total of 85 articles comprising of duplicates. The PRISMA guidilines were followed. Studies in English language and published in peer-reviewed journals were included. Articles with level of evidence I to IV were included Results: A total of three articles were included in the qualitative analysis. All the articles included are retrospective case series, with a level of evidence IV. Studies concerning patients with pre-operative imaging MRI and clinical assessment, reporting surgical technique and clinical outcomes assessed by physical examination and/or subjective evaluation scales were analysed. Conclusions: MRI and the Figure-4 test allow to assess PMF tears pre-operatively. Arthroscopic evaluation constitutes the gold standard to confirm the diagnosis. Although surgery is considered resolutive for symptoms, there is still controversy about the most appropriate technique. Further higher quality studies are required.
aponeurosis of the popliteus tendon, and the inferior fascicle is a coronary ligament which connects the meniscus to the edge of the tibia. A third incostant postero-inferior fascicle is sometime present. PMFs, connecting the lateral meniscus at the popliteal hiatus, are thought to provide stability to the lateral meniscus, stabilizing the joint during internal rotation of the tibia and sudden changes of direction. In particular, working in conjunction with the popliteus musculotendinous unit, the PMFs prevent excessive lateral meniscal movement and possible entrapment. For this reason, athletes, such as martial artists, dancers, wrestlers and football players, whose activity is characterised by sudden changes of direction, rotational stresses, repetitive twisting and high jumps, are at higher risk of injuring the PMFs. The injuries which affect the lateral meniscus lead to an increase of contact pressure and rotational instability, predisposing the joint to osteoarthritis as observed on radiographs. A hypermobile lateral meniscus (HLM) may cause knee pain and a locking sensation during deep knee flexion. One of the most frequent cause of a HLM is thought to be a posttraumatic injury of the PMFs.
Clinical and imaging diagnosis of these lesions is always challenging, even though MRI using proton density sequences may be useful. In particular, it has been reported that the detection rate of PMFs on a routine knee MRI on sagittal and coronal plane is approximatively 60%. The clinical diagnosis is even more challenging, as most PMFs, tears occur in multiligamentosus injuries. When a tear of the PMFs contributes to make the lateral meniscus unstable, arthroscopic observation and probing into the popliteomeniscal fascicle area is helpful to identify the damaged fascicle. In particular, many knees with acute and chronic ACL and/or posterolateral injuries have concurrent damage to the popliteomeniscal fascicles. Tears of PMFs are the most frequent lesion occurring in 80% of patients with grade III posterolateral injuries associated with ACL insufficiency. The risk is that an ACL injury could wrongly be identified as the isolated cause of instability and knee pain.
Unrecognised tears to the structure comprising the PLC have been cited as an important factor in postsurgical failure after cruciate ligament reconstruction and in chronic instability and degenerative changes after knee trauma. PMFs tears lead to lateral knee pain, painful squatting and locking sensation. Several surgical options are available: from open surgery to arthroscopic ones. This systematic review identifies and analyses the articles on PMFs tears from diagnosis to surgical approach.
# Methods
The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines were followed to perform this systematic review.
## Search strategy
Four electronic databases (PubMed, Scopus, Web of Science and EMBASE) were used to search the scientific literature using various combination of the keywords "Popliteomeniscal Fascicles", "Lateral Meniscus", "Popliteal Hiatus", "Posterolateral Corner", "Tear" and "Surgical Repair" for the years 1950-2020. The final search was performed on 1 December 2020 by two indipendent investigators. All the resulting titles were organised and screened indipendently. In case of disagreement, a third senior investigator was asked to check and screened the resulting titles.
## Selection criteria
Only studies published in peer-reviewed journals were included. Using the Oxford Center of Evidence-Based Medicine guidelines, level I to IV articles were identified. Studies with patients assessed by pre-operative MRI and then arthroscopically for post-traumatic PMFs tear were included. Furthermore, studies reporting surgical repair of PMFs outcomes assessed by clinical examination and/ or subjective evaluation scales were included. Reviews, metanalyses, cadaveric and animal studies, biomechanical studies, case report, commentaries, expert opinions and operative techniques were excluded. We also decided to exclude studies in which no information about the surgical procedure performed, diagnosis, follow-up, pre-operative imaging, arthroscopic or surgical assessment of PMFs and associated tears, pre-operative clinical examination and clinical postoperative outcomes were recorded.
## Evaluation of the study quality
The methodological quality and bias of each study were evaluated with the Coleman Methodology Score (CMS), which assesses methodology with 10 criteria, giving a total score between 0 and 100. A score of 100 indicates that the study largely avoids chance, various biases, and confounding factors. The subsections that consistute the CMS are based on the subsections of the Consolidated Standards of Reporting Trials (CONSORT) statement (for randomized controlled trials). Each study was scored by two indipendent reviewers and matched each other . In case of mismatch, a third investigator was asked to perform the CMS assessment indipendently. Possible disagreements were resolved by discussion.
## Data extraction
To avoid any bias of selection, the included articles with all the relative list of references, and the articles excluded from the study were reviewed, assessed, and discussed by all the authors. In case of disagreement among the reviewers regarding the selection of articles based on inclusion and exclusion criteria, the senior investigator made the final decision. The following data were independently extracted by all the investigators: demographics, inlcuding mean age, sex, level of activity of population; mean follow-up; timing from symptoms to surgery; pre-operative MRI assessment; associated lesion reported; certainty of diagnosis by pre-operative clinical examination and arthroscopic confirmation; surgical management and technique performed; clinical outcome measurements by post-operative clinical examination and/ or subjective evaluation scales; recurrence of the lateral meniscus instability and/or pain and/or locking sensation, and intra-and/or post-operative complications.
# Results
## Study selection
A total of three articles were included in the qualitative analysis.describes the methodology used for selection and inclusion of articles. The original literature search identified a total of 85 articles comprising of duplicates. Another five articles were identified by other sources. After removal of 41 duplicates, 49 articles were assessed for eligibility. Eighteen articles were removed because they did not concern PMF. Of the remaining 31 full-text articles, 22 articles were removed because do not meet inclusion criteria: 1 was excluded because was a review in German language; 2 articles were excluded because they were cadaveric studies; 4 articles were excluded because they were imaging studies; 9 articles were excluded because they were case reports; 3 were excluded because they were reviews of literature; 1 article was a laboratory study; 1 article was a cadaveric and imaging study; 2 articles were excluded because they described operative techniques without patient outcomes. The most significative data concerning the quality and the study methodology used are reported experienced symptoms such as pain and locking sensation. The pre-operative clinical examination was based on the test, which was positive in all patients in LaPrade et al. study, and in 3 of 6 patients in Simonetta et al. study. Kamiya et al.do not report the results of any clinical examination.
The imaging study to diagnose PMF tears was MRI in for all included articles. Kamiya et al.performed a 3D-virtual load MRI, which allows, under dynamic load, to verify the forward and medial translation of the lateral meniscus during knee flexion. Simonetta et al.imaging study was based on MRI, and the diagnosis of PMF tear was confirmed in the sagittal plane and T2 sequences that demonstrate the disruption of PMFs. LaPrade et al.used plain MRI scan to demonstrate PMF tear. The mean timing of symptoms to surgery, reported in all the articles, was 15.1 months (range 0.3-60).
## Surgical approach and suturing technique
In, the surgical technique is reported. To confirm the diagnosis and assess the disruption of the PMF, arthroscopic evaluation is necessary. In all the articles included, direct arthroscopic visualization confirmed the PMF tear between the posterior horn of the lateral meniscus and the popliteus hiatus.
Further evidence of PMF tear was given by the increased mobility and forward translation of the intact lateral meniscus on arthroscopic probing. Different surgical procedures are reported: both Kamiya et al.and Simonetta et al.performed an arthroscopic technique, while LaPrade et al.performed an open repair. Each article reports a different suturing technique.
Kamiya et al.performed an inside-out repair of injured PMFs using a polyester non-absorbable suture on a 10-inch straight and/or curved cutting needle (Stryker, Kalamazoo, CA, USA). An average of 5.0 (range 2-8) double-stacked vertical stitches connecting the lateral meniscus and the meniscocapsular junction was reported. No other associated lesions were noted at the arthroscopic evaluation.
Simonetta et al.used the FastFix meniscal repair system (Smith & Nephew, Hull, UK) to fix disrupted PMFs with an all-inside technique. Two or three stitches in a vertical fashion on either side of the popliteal hiatus were used. In their case series, 4 of 6 patients, in whom timing from symptoms to surgery was greater than 12 months, presented a chondral lesion of the lateral femoral condyle, but these associated lesions were not addressed.
La Prade et al.performed an open repair. For this purpose, after an antero-lateral capsular arthrotomy, the midthird lateral capsular ligament was retracted and the antero-inferior fascicle tear was exposed. After this, an open repair using horizontal mattress sutures with nonabsorbable 0 sutures was performed.
# Outcome analysis
Clinical outcome with evaluation subjective scales, perioperative complications and post-operative imaging studies are reported in. For all three articles included,
# Discussion
The diagnosis, by imaging and clinical examination or arthroscopically, of disrupted PMFs remains very challenging. Tears of PMFs are present in 80% of patients with an ACL tear or traumatic injury to the PLC. In symptomatic patients, surgery is the gold standard to manage these injuries, although the most appropriate surgical technique is still debated. This systematic review shows surgery for symptomatic patients with PMF tears is effective and safe, with pain relief and resolution of locking sensation with no complications reported. Furthermore, although the diagnosis is not easy and an accurate clinical examination and imaging, by MRI scan, is fundamental to plan the most appropriate treatment, arthroscopic evaluation by probing the lateral meniscus and the intra-articular visualization of disrupted PMFs remain the gold standard to accurately identify these injuries. The lack of high level-of-evidence studies about this topic and the paucity of articles regarding the management of PMFs tears does not allow to fully clarify which could be the most appropriate surgical technique to undertake. According to our results, younger patients with highdemanding activities are most suitable candidates for surgery. Although Kamiya et al.operated on patients older than 50, the mean age among the 32 patients analysed was 26.9. Only for 9 patients the level of activity was specified: 7 of 9 patients were professional or semi-professional athletes. High-demanding athletic patients are susceptible to PMFs tears, with the common injury mechanism being twisting on their knee. Full-contact sports, such as wrestling or soccer, are the most commonly involved. The indications for surgery included symptoms, clinical examination, except for Kamiya et al., and imaging: pain over the lateral aspect of the knee, pain during active joint flexion and a locking sensation were reported in all the patients described in the studies included in the present review. A specific clinical examination was performed by LaPrade et al.and Simonetta et al.: the figure-4 test, first described by LaPrade, was chosen as diagnostic test to check the integrity of PMF. Of 12 patients undergoing the figure-4 test, 9 (75%) resulted positive, indicating that a positive figure-4 test is strongly suggestive of PMF tear. In all the patients reported, the diagnosis of PMF disruption was confirmed arthroscopically. Arthroscopy is crucial to check the lateral meniscus stability . If the PMFs are disrupted, probing the lateral meniscus produces an abnormally large forward translation of the meniscus with a direct visualization of PMF tear on the lateral side of the joint space. In LaPrade et al.study, although they performed an open repair, an arthroscopic check to confirmation the diagnosis was performed. The time from onset of symptoms to surgery suggests that timing does not influence the healing potential of PMF, and that the continuous forward movement of the lateral meniscus prevents the formation of a scar tissue that would stabilise the meniscus. PMF tears are often associated with other intra-articular injuries such as ACL tears, meniscal and chondral tears, and PLC injuries. Furthermore, associated chondral lesions of the lateral femoral condyle were reported in 4 of 32 (12.5%) patients. These four patients had reported symptoms for longer than 12 months. To prevent further damage to the knee joint and protect other structures which have been injured or reconstructed (ACL, meniscal and chondral lesions) or stabilize the lateral compartment, prompt surgery in active patients with symptoms and a positive MRI and clinic examination is mandatory.
A pre-operative MRI is always performed and appears the best imaging modality to demonstrate disruption of PMF on the sagittal plane and in T2 sequences. Kamiya et al.performed a virtual 3D-loading MRI which demonstrated the forward translation of the lateral meniscus that during active knee flexion, a finding subsequently confirmed arthroscopically.
The appropriate surgical technique is still debated. In all the articles included in the present review, the surgical procedure is well described and characterized. While Kamiya et al.and Simonetta et al.performed two different arthroscopic techniques, an inside-out and an all-inside repair respectively, LaPrade et al.choose an open repair, with three different suturing technique. All three articles reported good clinical outcome with resolution of symptoms and no complications: only LaPrade et al.performed a new clinical examination by subjecting patients to a post-operative figure-4 test, which always resulted negative. Kamiya et al.reported subjective evaluation scales and post-operative MRI imaging study: the Tegner and Lysholm evaluations scores showed an improvement between pre-and post-operative values.
# Study limitations
The limitations of the present systematic review are related to the scanty quality of the studies available in the literature. All the articles included were retrospective case series, with limited number of patients. Furthermore, the absence of subjective evaluation scales and no post-operative assessment by MRI imaging, except for Kamiya et al., do not allow complete outcome analysis. The level of evidence for all the articles included was low (IV). According to the Coleman Score, the study quality ranges from fairto poor.
# Conclusion
Although the number of studies available in the literature was limited and the methodological quality of the articles included was questionable, the most relevant evidence arising from this systematic review is that it is important to recognise PMF disruption in patients with pain and locking sensation, and no clear evidence of other causes of internal derangement of the knee. In particular, it is necessary to evaluate PMF tears by pre-oeprative MRI imaging, and clinical examination using the figure-4 test. Arthroscopic confirmation of the diagnosis by direct visualization of the PMF tear and probing of the lateral meniscus is necessary. When recognised, the evidence suggests to operate on the lesion to stabilise the knee joint, especially in young athletic patients. There is no definite evidence about the most appropriate surgical technique to perform, and surgeon experience and confidence should guide the choice of the appropriate technique. Further studies with a higher number of patients, and higher methodological quality should be undertaken to better understand the pathogenesis and the evolution of PMF tears and to clarify the best surgical technique. |
A Phase I study of an intravesically administered immunotoxin targeting EpCAM for the treatment of nonmuscle-invasive bladder cancer in BCGrefractory and BCG-intolerant patients
Purpose:A Phase I dose-escalation study was performed to determine the maximum tolerated dose (MTD) of the immunotoxin VB4-845 in patients with nonmuscle-invasive bladder cancer (NMIBC) refractory to or intolerant of bacillus Calmette-Guerin (BCG). Secondary objectives included evaluation of the safety, tolerability, pharmacokinetics, immunogenicity, and efficacy of VB4-845. Patients and methods: Sixty-four patients with Grade 2 or 3, stage Ta or T1 transitional cell carcinoma or in situ carcinoma, either refractory to or intolerant of BCG therapy, were enrolled. Treatment was administered in ascending dose cohorts ranging from 0.1 to 30.16 mg. After receiving weekly instillations of VB4-845 to the bladder via catheter for 6 consecutive weeks, patients were followed for 4-6 weeks post-therapy and assessed at week 12. Results: An MTD was not determined, as a dose-limiting toxicity was not identified over the dose range tested. VB4-845 therapy was safe and well tolerated with most adverse events reported as mild; as a result, no patients were removed from the study in response to toxicity. By the end of the study, the majority of patients had developed antibodies to the exotoxin portion of VB4-845. A complete response was achieved in 39% of patients at the 12-week time point. Conclusions: VB4-845 dosed on a weekly basis for 6 weeks was very well tolerated at all dose levels. Although an MTD was not determined at the doses administered, VB4-845 showed evidence of an antitumor effect that warrants further clinical investigation for the treatment of NMIBC in this patient population.
# Introduction
Bladder cancer is the fourth most common malignancy in men and the ninth most common in women, with an estimated incidence of 67,160 and 13,750 estimated deaths in the United States. Transitional cell carcinoma (TCC) refers to those bladder cancer tumors derived from urothelial tissue, more than 90% of which originate in the urinary bladder and present as nonmuscle-invasive disease in the majority of patients at diagnosis.Of those patients who respond to current standard of care, approximately two-thirds will have recurrence, and 10%-20% of recurrent tumors will have progressed to muscle-invasive disease. [bib_ref] Patient-recognition data-mining model for BCG-plus interferon immunotherapy bladder cancer treatment, Shah [/bib_ref] Typically, treatment for high-risk, nonmuscle-invasive bladder cancer (NMIBC) is transurethral resection of the bladder tumor and adjuvant therapy with bacillus Calmette-Guerin (BCG). Although BCG treatment can reduce the risk of recurrence and progression, its use is limited by the adverse effect profile and intolerance that occurs in 20% of patients. [bib_ref] Intravesical Bacillus Calmette-Guerin reduces the risk of progression in patients with superficial..., Sylvester [/bib_ref] [bib_ref] High-grade Ta urothelial carcinoma and carcinoma in situ of the bladder, Sylvester [/bib_ref] [bib_ref] Intravesical Bacillus Calmette-Guerin for treating bladder cancer, Boyd [/bib_ref] [bib_ref] Review pathology in a diagnostic bladder cancer trial: effect of patient risk..., Witjes [/bib_ref] Epithelial cell adhesion molecule (EpCAM) is overexpressed in many carcinomas relative to their normal tissue counterparts, as is the case in TCC. [bib_ref] Epithelial cell adhesion molecule (Ep-CAM) modulates cell-cell interactions mediated by classic cadherins, Litvinov [/bib_ref] [bib_ref] The biology of the 17-1A antigen (Ep-CAM), Balzar [/bib_ref] In addition, EpCAM expression increases as these cancers progress from lower to higher grades. [bib_ref] The biology of the 17-1A antigen (Ep-CAM), Balzar [/bib_ref] [bib_ref] Expression of a cell surface antigen recognized by the monoclonal antibody AUA1..., Zorzos [/bib_ref] [bib_ref] EpCAM is predominantly expressed in high grade and advanced stage urothelial carcinoma..., Brunner [/bib_ref] [bib_ref] Immunohistochemical and immunocytochemical study of bladder carcinomas using the epitheliumspecific tumour-associated monoclonal..., Anagnostaki [/bib_ref] Together, these features make EpCAM a clinically relevant antigen for targeted therapy in bladder cancer.
VB4-845 is a recombinant fusion protein that targets EpCAM-positive cancer cells. It consists of an anti-EpCAM humanized single-chain variable fragment (scFv) linked to a truncated form of Pseudomonas exotoxin A (ETA 252-608 ) that lacks the cell-binding domain. [bib_ref] A recombinant immunotoxin derived from a humanized epithelial cell adhesion molecule-specific single-chain..., Paolo [/bib_ref] Once bound to EpCAM on the surface of carcinoma cells, VB4-845 is internalized, whereupon the exotoxin portion of the fusion protein induces apoptosis. [bib_ref] Role of caspases in immunotoxin-induced apoptosis of cancer cells, Keppler-Hafkemeyer [/bib_ref] [bib_ref] Immunotoxin therapy of cancer, Pastan [/bib_ref] One concern of targeted therapies has been the toxicity associated with systemic administration of this class of drug. [bib_ref] Targeted toxins, Frankel [/bib_ref] Moreover, repeated use of therapeutics comprising foreign proteins is limited by their immunogenicity. Therefore, it is desirable to develop therapies designed for local administration, thereby increasing the clinical benefit of these treatments while minimizing any drug-related effects. Accordingly, locoregional delivery of ETA-conjugated antibodies has been demonstrated to be well tolerated and clinically effective in patients with glioblastoma multiforme, ErbB2-expressing breast tumors, and squamous cell carcinoma of the head and neck. [bib_ref] Regression of cutaneous tumor lesions in patients intratumorally injected with a recombinant..., Azemar [/bib_ref] [bib_ref] Direct intracerebral delivery of cintredekin besudotox (IL13-PE38QQR) in recurrent malignant glioma: a..., Kunwar [/bib_ref] [bib_ref] A Phase I clinical study of VB4-845: weekly intratumoral administration of an..., Macdonald [/bib_ref] Nonclinical studies showed a significant reduction in toxicity with locally administered VB4-845. Similarly, local injections of VB4-845 were well tolerated in Cynomolgus monkeys with adverse events (AEs) being mild and easily managed. [bib_ref] Preclinical assessment of an anti-EpCAM Immunotoxin: locoregional delivery provides a safer alternative..., Brown [/bib_ref] In addition to the strong nonclinical safety profile, VB4-845 exhibits highly potent activity against EpCAM-expressing tumor cell lines and has been shown to localize to EpCAM-positive tumor xenografts. [bib_ref] A recombinant immunotoxin derived from a humanized epithelial cell adhesion molecule-specific single-chain..., Paolo [/bib_ref] Based on these preclinical results, a Phase I dose-escalation trial was performed using VB4-845 as an intravesical therapy in BCG-refractory and BCG-intolerant patients with Grade 2 or 3 NMIBC.
# Patients and methods
## Patient selection
Only patients 18 years of age or older with immunohistochemically confirmed EpCAM-positive Grade 2 or 3 NMIBC (Ta, T1, in situ carcinoma [TIS]), either refractory to (recurrence within 2 years following at least one complete cycle of BCG therapy) or intolerant of BCG therapy, were eligible for this study.
Other key inclusion criteria were adequate renal (serum creatinine #1.5 × the upper limit of normal (ULN) or creatinine clearance $60 mLs/min), hepatic (alanine aminotransferase and aspartate aminotransferase #2.5 × ULN and bilirubin levels #1.5 × ULN), and hematological (granulocytes $1500/µL, platelets $100,000/µL, and hemoglobin .8 g/dL) function. Women of child-bearing potential, and all men, must have agreed to use adequate contraception prior to and for the duration of the study.
Key exclusion criteria included patients with muscleinvasive tumors, nodal involvement, or distant metastases; patients with a history of upper tract TCC, adenocarcinoma, or squamous cell carcinoma of the bladder; and patients with disease involving the prostatic ducts or stroma. Moreover, excluded were patients with a history of pelvic malignancy, hydronephrosis, or clinically significant abnormalities of the upper urinary tract and those who had undergone BCG therapy within 6 weeks prior to the start of VB4-845 dosing.
Written informed consent was obtained from each participant before any study-related activity was performed. This study was conducted according to Section C.05.010 of Division 5 of the Food and Drug Regulations of the Government of Canada, ICH Harmonized Tripartite Guidelines for Good Clinical Practice, the Declaration of Helsinki (2002), and all local laws and regulations concerning clinical studies and the protection of study patients. Regulatory clearance for the study was obtained from the Biologics and Genetic Therapies Directorate of Health Canada in Canada.
## Study design and dose escalation
The study was an open-label, multicenter, dose-escalating trial of intravesically administered VB4-845. Eight dose levels were initially evaluated, starting at 0.1 mg once weekly for 6 consecutive weeks and escalating through 0.2, 0.33, 0.66, 1.32, 2.64, 5.28, and 10.56 mg/dose. The maximum tolerated dose (MTD) was not reached; therefore, an additional escalation through 13.73, 17.85, 23.20, and 30.16 mg was undertaken. Each dose was administered to the bladder through a catheter and held for 2 h prior to voiding. Safety data from each dose cohort was evaluated after 3 weeks of treatment before proceeding to the next dose cohort. Dose-limiting toxicity (DLT) was defined as the occurrence of treatment-related AEs, including intractable cystitis persisting for more than 1 week associated with severe pain, urgency, and/or frequency not relieved by measures considered to be routine standard of care according to each clinical site; significant hematuria leading to clot obstruction; Grade 4 flu-like symptoms; Grade 3 or higher hematological toxicity (if it is a two grade increase from baseline); or any Dovepress Dovepress 315 immunotoxin therapy for nMiBc other Grade 3 nonhematological toxicity (with the exception of alopecia). If two patients experienced a DLT at any dose, that dose would be defined as the MTD.
## Patient evaluation
Patients were evaluated at every visit by physical exam and assessment of vital signs. Urinalysis was performed, and blood was collected for biochemistry and hematology. Concomitant medications were recorded. Specific additional evaluations were performed depending on the timing of the visit, as described below.
Pretreatment screening evaluations were conducted within 4 weeks of the baseline visit and included medical history, pregnancy test, 12-lead electrocardiogram, and assessment of Karnofsky performance status. Tumors were evaluated by cystoscopy/photography, cytology, and biopsy. Cytopathology and biopsy pathologic examinations were performed locally and not at a central laboratory. Immunohistochemistry on snap-frozen tissue samples was performed at a single facility to determine whether tumors were EpCAM positive.
In addition to the evaluations performed at every study visit, patients were monitored for AEs for 3 h following drug administration on days 1, 8, 15, 22, 29, and 36. The toxicity grade of any AE was classified according to the National Cancer Institute Common Toxicity Criteria, Version 3. AEs were considered to be treatment related by the investigators if they were possibly, probably, or definitely related to VB4-845 administration. On day 2, AEs and concomitant medications were evaluated by interview.
The last visit occurred 4-6 weeks after the final treatment. Additional assessments performed included pregnancy test, 12-lead electrocardiogram, and Karnofsky performance status. To evaluate tumor response, cystoscopy/photography and cytology were performed. A biopsy was required if the cystoscopy revealed lesions suspicious for malignancy, or cytology results were suspicious or positive for malignancy.
## Pharmacokinetics and immunogenicity
Blood samples for pharmacokinetic analysis were taken on day 1 prior to and 1, 2, and 3 h postdosing. Predose samples were also taken on days 8, 15, 22, 29, and 36. For each sample, 3 mL of venous blood was collected into a tube containing lithium heparinate and placed on ice. VB4-845 plasma levels were measured using a GLP-validated, MTS-based, potency assay. [bib_ref] A Phase I clinical study of VB4-845: weekly intratumoral administration of an..., Macdonald [/bib_ref] The assay detects intact VB4-845 by its ability to kill the EpCAMpositive cell line CAL-27, where the measured IC 50 is directly proportional to the concentration of intact drug. The assay was shown to have a lower limit of detection of 14 pg/mL. Blood samples for assessing humoral immune reactivity to VB4-845 were taken prior to dosing on days [bib_ref] The biology of the 17-1A antigen (Ep-CAM), Balzar [/bib_ref] [bib_ref] Targeted toxins, Frankel [/bib_ref] , and at the final study visit. Samples were collected as for the pharmacokinetic analysis, and antibody titers to the scFv and ETA 252-608 portions of the fusion protein (human antihuman antibodies (HAHA) and human anti-Pseudomonas antibodies (HAPA)) were measured using an enzyme-linked immunosorbent assay. [bib_ref] A Phase I clinical study of VB4-845: weekly intratumoral administration of an..., Macdonald [/bib_ref] Responses were expressed as the geometric mean of measurable titers at each time point.
## Evaluation of efficacy
Tumor response was assessed on evaluable patients 4-6 weeks following the last dose of VB4-845. The evaluation was based on cytology, cystoscopy, and, if required, biopsy. Complete response was defined as nonpositive urinary cytology and either normal cystoscopy or abnormal cystoscopy with negative biopsy.
# Results
## Patient characteristics
Sixty-four patients were enrolled in this study. All participants were Caucasian with a median patient age of 69; 78% were men and 22% were women [fig_ref] Table 1: Patient characteristicsNotes [/fig_ref]. All patients evaluated for efficacy (61/64) had previously received BCG therapy, with 95% of the patients having had two or more bladder cancer recurrences. Based on their previous BCG treatment Of note, as this study was carried out prior to the widespread adoption of re-transurethral resection (re-TUR) of TaT1, high-grade NMIBC in clinical practice, and its inclusion in standard treatment guidelines, re-TUR of T1 tumors was not performed. All tumors showed immunohistological evidence of EpCAM-positive membrane staining [fig_ref] Figure 1: immunohistochemical staining of epcAM on Tcc of the bladder [/fig_ref]. Risk factors for progression present in the patient population are summarized in [fig_ref] Table 2: Disease risk factors at baseline [/fig_ref]. At baseline, 98% of patients had at least one factor associated with disease progression, and 64% of these individuals had three or more risk factors.
## Dose escalation and mtd
VB4-845 was administered to the bladder via catheter once a week for 6 consecutive weeks, followed by 4-6 weeks without treatment. An MTD was not determined from the initial dose escalation as no DLT was reported; therefore, additional dose cohorts were added as described above, and, again, all patients received all six scheduled doses without any DLT. The number of patients per dose cohort is summarized in [fig_ref] Table 3: number of patients per dose cohortNotes [/fig_ref].
safety Forty-one of the 64 patients (64%) experienced an AE during the course of the study. Twenty-one (33%) patients experienced only AEs that were assessed by the investigator as unrelated to study treatment. Classifying the AEs as unrelated to the study drug was decided after considering the temporal relationship of the onset of the event to the administration of the study drug, whether the event could be explained by concomitant medications or concurrent disease, the response to withholding the study drug, and the response to rechallenge with the study drug. These included, but were not limited to, urinary tract infection, incontinence, nocturia, pharyngolaryngeal pain, and bladder spasm. The remaining 20 patients (31%) experienced AEs judged to be related to VB4-845 administration. The most common treatment-related AEs were dysuria and hematuria [fig_ref] Table 4: Adverse events related to VB4-845 administration Note [/fig_ref]. Systemic AEs included fatigue, fever and chills, loss of appetite, myalgia, dizziness, and nausea. The frequency of treatment-related AEs did not increase with dose escalation [fig_ref] Figure 2: Percentage of patients experiencing treatment-related adverse events [/fig_ref]. All treatment-related AEs were Grade 1 or 2, with the exception of one Grade 3 occurrence of hematuria that was reported as possibly related to VB4-845 administration. This event was not reported as a DLT as the investigator judged it to be related to aspirin use, and the patient experienced no AEs with subsequent VB4-845 dosing. There were no cases where the occurrence of AEs required discontinuation of treatment, and no patient experienced a serious AE related to the administration of VB4-845. There was one patient death due to cardiac failure, which occurred 3 weeks after the last dose of VB4-845. This death was assessed by the investigator as unrelated to study treatment and attributed to long-standing coronary artery disease, previous myocardial infarction, and hypertension. immunogenicity
## Pharmacokinetics
The immunogenicity of VB4-845 was examined by analyzing HAPA and HAHA titers in blood samples taken at specified time points during the trial (
## Exploratory efficacy
Sixty-one patients were considered to be evaluable for efficacy; two patients were excluded from the efficacy analysis due to an absence of BCG treatment prior to the study, and there was one study-unrelated death for whom no final tumor assessment was obtained.
Complete response based on tumor classification at baseline and dose group is summarized in [fig_ref] Table 6: Tumor response by stage and dose group Notes [/fig_ref]. Overall, a complete response was achieved by 39% (24/61) of the patients. Of the patients with TIS, 29% achieved a complete response, while complete responses were observed in 44% and 43% of the patients with T1 and Ta, respectively. Of the five patients classified as BCG intolerant, only one (20%) had a complete response.
Given the limited number of patients per dose cohort, it was not possible to make definitive individual dose comparisons. In order to examine a potential dose response, patients were classified into three dose groups: (0.1 to ,1.0 mg) = lowest dose group; (1.0 to ,10.0 mg) = middle dose group; and ($10.0 mg) = highest dose group. A comparison of the response rates in the lowest dose group versus the combined middle and highest dose group revealed a statistically significant P-value of 0.0418.
# Discussion
TCC is the most common cancer of the bladder, and most patients present with nonmuscle-invasive disease. Currently, BCG is the standard of care for this malignancy. Despite its well-documented therapeutic benefit, BCG therapy is delayed or discontinued in a large proportion of patients due to associated toxicity. In rare instances, patients can develop severe conditions associated with BCG infection, in some cases life-threatening, rendering BCG a nonviable treatment option. Termination of treatment due to adverse effects is not uncommon for intravesical therapies. In addition, for those patients who fail to respond to BCG therapy or experience early disease recurrence, alternative therapeutic options are limited. Given the potential for intolerance and the high rate of recurrence in patients with NMIBC, alternative treatments are under investigation.
Targeted therapies limit drug exposure to only diseased cells, thereby minimizing drug-related toxicities. We have described a Phase I study of VB4-845, an anti-EpCAM scFv-ETA, administered intravesically to high-risk patients. Of the 75 patients initially screened for participation, 74 (99%) showed EpCAM-positive disease. This high rate of EpCAM expression suggests that EpCAM represents a significant antigen for VB4-845-targeted therapy in patients with advanced NMIBC. Although not performed in this study, an examination of the expression levels of EpCAM in biopsy samples obtained from patients with residual disease following VB4-845 treatment might be informative and will be considered for future studies.
AEs experienced with VB4-845 were mild and easily managed, and no AE required the discontinuation of treatment, even in the highest dose cohort. Although the presence of pre-existing lower urinary tract symptoms may make a determination of the frequency of AEs difficult, what is evident is that the relatively mild toxicities experienced with VB4-845 are in direct contrast to other intravesical therapies where local and systemic toxicities can be treatment limiting. VB4-845 is a targeted therapy that exerts its effect via specific interaction with the cell surface antigen EpCAM. Although overexpressed on carcinoma cells, EpCAM expression is lower on normal tissue and limited to the basal layers rather than luminal surfaces. These features could minimize drug-related toxicities and may explain why related AEs were predominantly Grade 1 and 2. It is also interesting to note that failure to reach an MTD for VB4-845 over the dose range tested and the apparent absence of any relationship between the appearance of AEs and dose are in keeping with its excellent safety profile. Pharmacokinetic analyses indicated that VB4-845 was not systemically absorbed and remained within the bladder until elimination by voiding. This is anticipated, given the integrity of the bladder and the molecular size of VB4-845 (∼69 kDa).
VB4-845 was shown to be immunogenic with the toxin portion eliciting the more vigorous response. Given the humanization of the scFv antibody fragment and entirely foreign nature of the bacterial toxin, the more intense immune response against the toxin was expected. A similar pattern of immune response was observed with LMB-2, a scFv-ETA immunotoxin, whereby more patients exhibited a stronger response to the ETA moiety than to the scFv portion. [bib_ref] Phase I trial of recombinant immunotoxin anti-Tac (Fv)-PE38 (LMB-2) in patients with..., Kreitman [/bib_ref] Despite the observed immunogenicity, the development of an immune response was not considered to have any bearing on clinical outcome, as there was no apparent relationship between antibody titer and response to treatment. Rather, the presence of circulating antibodies to VB4-845 may, in fact, limit any systemic exposure that may occur by rapidly clearing any drug that may enter circulation. Together, the specificity of VB4-845 and its lack of systemic exposure translate to a very acceptable safety profile.
Although the Phase I trial was primarily designed to determine the MTD, tumor response was assessed as a single efficacy endpoint at 3 months to elucidate any observable trends. Although dosing was not optimized for therapeutic benefit, complete responses were observed across all tumor stages and dose groupings. Although it is not possible to draw definitive conclusions between individual dose and response to treatment, when examined according to dose groups, the data are suggestive of an increased response with higher doses. Although no direct examination of receptor occupancy has been performed, the dose range of VB4-845 administered far exceeds the drug concentration required to achieve .99% cell killing in EpCAM-positive bladder tumor cell lines. Therefore, it is possible that the tumor response may be saturated at the higher doses. It should be noted that the short duration of follow-up is a significant limitation with respect to assessing efficacy; thus, efficacy results reported in this study are considered preliminary and serve only to support further clinical trials with longer follow-up assessments.
# Conclusions
Intravesical therapy with the anti-EpCAM fusion protein VB4-845, for the treatment of NMIBC was extremely well tolerated. AEs were generally mild and very manageable with no patients experiencing any serious AEs related to VB4-845 therapy. The lack of detectable postinstillation plasma levels in the vast majority of patients indicates that the drug is effectively held within the bladder and not systemically absorbed. Although most patients developed antibody titers, the data obtained in this study showed no evidence of detrimental effects on clinical outcome associated with the immune response. Preliminary efficacy results from this highrisk, often treatment-refractory, patient population suggest that VB4-845 can safely inhibit tumor growth in patients with EpCAM-positive, high-grade NMIBC. This promising finding warrants investigation in a larger clinical trial with optimized dosing.
## Drug design, development and therapy
## Publish your work in this journal
Submit your manuscript here: http://www.dovepress.com/drug-design-development-and-therapy-journal Drug Design, Development and Therapy is an international, peerreviewed open-access journal that spans the spectrum of drug design and development through to clinical applications. Clinical outcomes, patient safety, and programs for the development and effective, safe, and sustained use of medicines are a feature of the journal, which has also been accepted for indexing on PubMed Central. The manuscript management system is completely online and includes a very quick and fair peer-review system, which is all easy to use. Visit http://www.dovepress.com/testimonials.php to read real quotes from published authors.
[fig] Figure 1: immunohistochemical staining of epcAM on Tcc of the bladder. Tissue samples were formalin fixed, paraffin embedded, and mounted onto coated microscope slides. Test slides containing A) normal bladder, C) epcAM-negative Tcc, and E) epcAM-positive Tcc tissue sections were incubated with the primary antibody VB4-845 and VB4-845 cell surface binding detected with a secondary antibody, rabbit anti-Pseudomonas eTA (sigma, catalog no. P2318, st. Louis, MO, USA). Specificity of EpCAM staining was demonstrated with the corresponding control slides, B) normal bladder, D) epcAM-negative Tcc, and F) epcAMpositive Tcc, treated only with the secondary antibody. Membranous staining (2+) is only apparent with the epcAM-positive Tcc tissue section E) incubated with both VB4-845 and the secondary antibody. All pictures are shown at 100 × magnification. [/fig]
[fig] Figure 2: Percentage of patients experiencing treatment-related adverse events (Aes) by grade per dose level. Of the 41 patients experiencing Aes, 20 patients experienced Aes related to treatment. Aes at each dose level are proportioned according to the grade level. if a patient experienced an Ae more than once, only the event with the highest grade was included in the calculation. The dose levels and corresponding total number of Aes are as follows: 0.01, n = 9; 0.20, n = 0; 0.33, n = 6; 0.66, n = 12; 1.32, n = 1; 2.64, n = 9; 5.28, n = 6; 10.56, n = 1; 13.73, n = 6; 17.85, n = 1; 23.20, n = 7; and 30.16, n = 2.submit your manuscript | www.dovepress. [/fig]
[table] Table 1: Patient characteristicsNotes: a Percentages are shown in parentheses. The median patient age was 69 years with patient ages ranging 50-87 years; b Two patients had not received Bcg. your manuscript | www.dovepress.com Dovepress Dovepress history, only five patients were considered to have been enrolled as BCG intolerant rather than BCG refractory. All patients had Grade 2 or 3, stage Ta or T1, TCC, and/or TIS, and, except for two cases, all papillary tumors had been previously resected 7-28 days prior to the first dose of study drug. [/table]
[table] Table 2: Disease risk factors at baseline [/table]
[table] Table 3: number of patients per dose cohortNotes: a One treatment-unrelated death. This patient was not included in the assessment for efficacy; b Two patients had not received prior Bcg treatment and were not included in the efficacy assessment. [/table]
[table] Table 4: Adverse events related to VB4-845 administration Note: a lf a patient experienced an adverse event more than once, the event with the highest grade was tabulated. [/table]
[table] Table 5: Antibody responses to the eTA and scFv portions of VB4-845 [/table]
[table] Table 6: Tumor response by stage and dose group Notes: a Percentages are shown in parentheses. P-value for lowest vs middle dose group = 0.1206. P-value for lowest vs highest dose group = 0.0622. P-value for lowest vs middle + highest dose groups combined = 0.0418. [/table]
|
The impact of distractor congruency on stimulus processing in retinotopic visual cortex☆
The brain is frequently confronted with sensory information that elicits conflicting response choices. While much research has addressed the top down control mechanisms associated with detection and resolution of response competition, the effects of response competition on sensory processing in the primary visual cortex remain unclear. To address this question we modified a typical 'flanker task'(Eriksen and Eriksen, 1974)so that the effects of response competition on human early retinotopic visual cortex could be assessed. Healthy human participants were scanned using fMRI while making a speeded choice response that classified a target object image into one of two categories (e.g. fruits, animals). An irrelevant distractor image that was either congruent (same image as target), incongruent (image from opposite category as target), or neutral (image from task-irrelevant category, e.g. household items) was also present on each trial, but in a different quadrant of the visual field relative to the target. Retinotopic V1 areas responding to the target stimuli showed increased response to targets in the presence of response-incongruent (compared to response-neutral) distractors. A negative correlation with behavioral response competition effects indicated that an increased primary visual cortical response to targets in the incongruent (vs. neutral) trials is associated with a reduced response competition effect on behavior. These results suggest a novel conflict resolution mechanism in the primary visual cortex.
# Introduction
Most stimuli encountered in the environment elicit some form of response, related either to previous direct experience or to an indirect association. Coherent goal-directed behavior requires the suppression of responses to stimuli that are irrelevant to the current task in order to prevent response conflicts. This is not always successful; people often fail to ignore irrelevant stimuli and the tendency to respond to them elicits response conflicts, which reduce the efficiency of task performance (e.g. by slowing down task responses).
The neural correlates of response conflict include a network of parietal and prefrontal regions responsible for identifying response conflict, resolving it in favor of the goal-relevant 'target' in accords with current task goals, and redeploying attention accordingly (e.g., [bib_ref] Dissociable contributions of prefrontal and parietal cortices to response selection, Bunge [/bib_ref] [bib_ref] Anterior cingulated cortex, error detection, and the online monitoring of performance, Carter [/bib_ref] [bib_ref] Cognitive and brain consequences of conflict, Fan [/bib_ref] [bib_ref] Neural activation during response competition, Hazeltine [/bib_ref] [bib_ref] Anterior cingulated conflict monitoring and adjustments in control, Kerns [/bib_ref] [bib_ref] Dissociating the role of the dorsolateral prefrontal and anterior cingulated cortex in..., Macdonald [/bib_ref] [bib_ref] Anterior cingulate cortex, conflict monitoring, and levels of processing, Van Veen [/bib_ref]. However, typically this 'resolution' of conflict does not prevent perception of the distracting stimuli (exceptions are cases of high perceptual load in the task, see [bib_ref] Distracted and confused?: selective attention under load, Lavie [/bib_ref] [bib_ref] Attention, distraction and cognitive control under load, Lavie [/bib_ref] or conflict adaptation through sequential repetition of incongruent stimuli, e.g. [bib_ref] Congruency sequence effects and cognitive control, Egner [/bib_ref]. That task-irrelevant stimuli are perceived even in cases of correct response selection (following resolution of the conflict in response tendencies) is clearly evident from typical findings that response times are slower on incongruent compared to congruent or neutral trials. Thus while it is clear that the frontoparietal network controls response selection, it remains unclear whether the identification of conflict and its resolution in terms of response selection has any effect on the sensory processing of target and of the distractor stimuli. Specifically, when people encounter a response conflicting distractor stimulus but successfully select the correct target response, are there any effects on sensory visual processing related to the target or distractor perception?
In the present study we used fMRI to elucidate the effects of response competition on the sensory processing of the target and distractor stimuli. To that purpose we modified a well-established response-competition task [bib_ref] Effects of noise letters upon the identification of a target letter in..., Eriksen [/bib_ref] for an experiment that allowed us to investigate the sensory visual correlates of response competition. Using images of common objects presented in separate visual quadrants, we were able to isolate the early visual cortical response to the target and distractor images under varying conditions of response congruency. We also further analyzed the response in retinotopic cortex relative to the magnitude of behavioral congruency effects.
# Material and methods
## Participants
Seventeen people recruited from the University College London experiment pool participated in this study for monetary compensation. Two people were excluded from the final analysis: one because of excessive head motion during the scanning session, and one because the participant's mean RT and overall accuracy on the behavioral task were more than 2.5 standard deviations away from the group mean. This resulted in a final pool of 15 participants (six females, ages 18-35). All participants provided informed consent in accordance with the UCL ethics committee.
## Stimuli
For the main task, the stimuli consisted of 12 gray scale images of objects spanning three different categories: fruits (strawberry, apple, pineapple, banana), household items (desk, sofa, fan, chair), and animals (cat, bird, bear, turtle). These images fit within a square that measured six degrees of visual angle on a side, and were positioned within the middle of each visual quadrant at a center-to-center distance of six degrees from the fixation point in the middle of the display. The stimuli were presented on a gray background; text and the fixation cross were presented in black.
For a functional localizer, the stimuli consisted of a disk with alternating black and white quarters, presented at the same size and location as the images in the main task. The contrast of the disks reversed at a rate of 5 Hz. The background, fixation cross and text color were the same as in the main task. For the retinotopic mapping runs, the stimuli consisted of pairs of wedges oriented along either the horizontal or vertical midlines and arranged in a "bow-tie" pattern. The interior of the wedges contained a black and white checkerboard pattern whose luminance oscillated at 8 Hz, and were presented on a gray background. Each wedge constituted an arc of 30 radial degrees.
The experiment was run on an Intel-based computer running Windows XP. The stimuli were generated and presented using MATLAB software (MathWorks, Natick, Massachusetts). This experiment was realized using Cogent 2000 developed by the Cogent 2000 team at the WTCN and the ICN and Cogent Graphics developed by John Romaya at the LON at the WTCN.
## Task procedure
Participants were instructed to perform a category judgment task on one member of a pair of briefly presented images. Images were presented in two of the four possible locations (one location in each visual quadrant), both within the same hemifield (upper, lower, left, right). Two of the four possible locations were defined a priori as target locations during the participants' instruction period; these locations were always arranged along the diagonal (i.e. upper-left and lower-right, or lowerleft and upper-right, counterbalanced across participants), such that one (and only one) image of each trial pair was presented in a target location. Participants judged which object category the target image belonged to in a 2AFC task; two of the three object categories (fruit, household item, animal) were defined as target categories at the beginning of the session (counterbalanced across subject, crossed with target locations).
Trials proceeded as follows. The fixation cross appeared in the middle of a blank display. 500 ms later, task images appeared in one of the configurations described above, and were present for 200 ms. [fig_ref] Figure 1: Display example from a single trial [/fig_ref] shows the stimuli from an example trial, where the upper-left and lower-right quadrants are defined as target locations. Participants then had 1.8 s to respond; the fixation cross remained visible during this time. Trials fell into one of three types, based on the identity of the non-target image: congruent (same image as the target), incongruent (image from opposite category as the target), or neutral (image from the task irrelevant category). Although our imaging comparisons involved only the incongruent and neutral conditions, the congruent condition was included to drive conflict in the incongruent condition (otherwise if the response-related distractors were always incongruent, the incongruent condition would become predictive of the (opposite) target). Identical images were used for the congruent condition, rather than different images from the same category, so as to avoid inducing conflict due to condition ambivalence (e.g. .
At the beginning of each experiment session, prior to the start of scanning, participants were given verbal and written instructions, followed by a practice block that was identical in all respects to block of trials during the main task. Participants completed four blocks of trials in the scanner. Each block consisted of 60 trials; trial type, target location and target category were all counterbalanced within each block. Blocks also contained 20 null trials, where the fixation cross appeared alone for 2.5 s and no response was required. Trials were separated by a variable interval (measured from the onset of fixation of one trial to the onset of fixation of the subsequent trial) of 3 to 7 s, to facilitate an event-related analysis. Each block began with a fixation period measuring 22.8 s (10 functional volumes) and ended with a fixation period measuring 11.4 s (5 functional volumes), and lasted~428 s.
Scanning sessions also contained two blocks of retinotopic mapping and two blocks of a functional localizer. Retinotopic mapping scans lasted 296 s, and consisted of alternating periods of stimulation along the horizontal and vertical visual meridians, lasting 18 s each. Participants were instructed to maintain fixation, but given no other task. Functional localizer scans lasted 347 s, and consisted of alternating periods of stimulation in the target and non-target locations lasting 22.8 s. Blocks also contained 22.8 s of fixation at the beginning of the block, 11.4 s of fixation at the end of the block, and two 22.8 s of fixation occurring during the middle of the block. Participants were instructed to maintain fixation, and respond with a button press to a luminance increment in the fixation cross lasting 200 ms (this occurred once during each stimulation period).
The stimuli were presented on a projection screen mounted at the end of the scanner bore, and viewed using a mirror mounted on the head coil. Responses were made using MR compatible fiber-optic button boxes.
## Imaging data collection and analysis
Imaging data were collected at the Wellcome Trust Centre for Neuroimaging using a 3-T Siemens Allegra Scanner with an 8-channel head coil. Anatomical images were acquired using a 3D MDEFT sequence [bib_ref] Optimization of the 3D MDEFT sequence for anatomical brain imaging: technical implications..., Deichmann [/bib_ref] with a sagittal partition direction yielding images with 1-mm isovoxel resolution (time repetition (TR) = 7.92 ms, time echo (TE) = 7.92 ms, flip angle = 15°). Whole-brain echoplanar functional images were acquired in 38 transverse slices (TR = 2280 ms, TE = 30 ms, matrix = 64 × 72, field of view = 216 mm, slice thickness = 2mm, 1 mm gap, descending slice acquisition order). Additionally, a double-echo FLASH field mapping sequence was collected to measure the distribution of field inhomogeneities (TE1 = 10 ms, TE2 = 12.46 ms, 3 × 3 × 2 mm resolution with 1 mm gap).
Data were processed using SPM8 (http://www.fil.ion.ucl.ac.uk/spm). Functional images collected from the scanner were slice-time corrected, aligned with a representative functional volume and motion corrected using an unwarping procedure that employed field inhomogeneity information from the field mapping scans, and spatially smoothed using a 4-mm full-width at half maximum Gaussian kernel. Structural images were also aligned with the representative functional volume. For the main experiment, separate events representing target position, distractor position and distractor congruency were modeled using a general linear model (GLM). These events were modeled as impulses of activity at the onset of the stimuli and were convolved with the canonical hemodynamic response function (HRF) included with SPM8. For the retinotopic and localizer scans, boxcars representing the duration of each block were convolved with the HRF and similarly modeled using a GLM. Results of the analyses, in the form of statistical maps, were overlaid on the structural images collected for each subject. For group-level analyses, these statistical maps were spatially normalized to the standard Montreal Neurological Institute space and entered into a one-way t-test to produce group average statistical maps, which were then overlaid onto an average structural image. To examine which regions of the cortex showed greater blood oxygen-level dependent (BOLD) responses to the various trial types, the group-level maps were generated using an uncorrected voxel-level threshold of p b 0.005 and a contiguous-voxel threshold producing a corrected threshold of p b 0.05.
Data from the retinotopic mapping scans were used to define the borders between areas V1, V2, V3/Vp, and V3a/V4v in each hemisphere [bib_ref] fMRI of human visual cortex, Engel [/bib_ref] [bib_ref] Borders of multiple visual areas in humans revealed by functional magnetic resonance..., Sereno [/bib_ref]. Data from the localizer scans were then used to define the segments of each of these areas that responded most strongly to stimuli in the location of the targets or distractors. To define regions of interest (ROIs) in the early visual cortex, the functional data were projected onto inflated representations of each participant's brain. Cortical reconstruction and volumetric segmentation were performed with the Freesurfer image analysis suite (http:// surfer.nmr.mgh.harvard.edu/). These ROIs were imported into SPM8, and the GLM parameter estimates corresponding to signals elicited by the stimuli appearing in the contralateral visual field for each of the previously described events were extracted from each ROI, allowing for analysis of the response to the different distractor types under the different load conditions. The ROI data were extracted and analyzed separately for each participant, then averaged. [fig_ref] Table 1: Behavioral performance as a function of distractor congruency [/fig_ref] presents participants' response times and accuracies on the main behavioral task. Planned comparisons showed that the incongruent distractors produced significant response competition effects compared to both the neutral (t(14) = 3.31, p b 0.01) and congruent conditions (t(14) = 4.62, p b 0.001). In contrast, congruent distractors produced a facilitation effect compared to the neutral condition (t(14) = 2.79, p b 0.02). As can be seen from the table, accuracy was high and the very small number of errors did not vary between the distractor conditions (t ≤ 1.3).
# Results
## Behavioral data
## Functional data: retinotopic cortices
Retinotopic analyses focused on the comparison of the incongruent and neutral distractor conditions, because these were visually comparable (in both of these conditions the target and the distractor were different to each other, whereas in the congruent condition the target and distractor stimuli were the same). This removed perceptual conflict as a potential confounding factor.
The retinotopic V1 region that responded to the target stimuli showed a significantly greater response to the targets on incongruent compared to neutral distractor trials (t(14) = 2.67, p b .02, see [fig_ref] Figure 2: BOLD responses to targets [/fig_ref]. This difference in V1 BOLD response was also negatively correlated with the difference in behavioral RTs: specifically, participants who had a larger difference between the BOLD response to targets on incongruent vs. neutral trials showed less slowing in RT for incongruent (compared to neutral) trials (r = −0.71, p b 0.001).
There were no other effects of distractor congruency on target responses in areas V2 to V4 (all other t values ≤ 1.4). There were also no effects of distractor congruency on the retinotopic V1-V4 areas that responded to distractors (all ts ≤ 1.2) but V1 and V4 showed a trend for a positive correlation between BOLD response to the incongruent (vs. congruent) distractor and the magnitude of this congruency effect on RTs (V1: r = 0.548, p b 0.05; V4: r = 0.545, p b 0.05; both p-values uncorrected).
## Functional data: full-brain contrasts
Statistical maps were calculated that examined the differences in the BOLD for the different trial types across all subjects; axial slices showing active regions for various contrasts are presented in and summary statistics are shown in [fig_ref] Table 2: Regions showing greater BOLD signals for various full-brain contrasts [/fig_ref]. The contrast of incongruent > congruent revealed a difference in BOLD signals in the right anterior insula. The contrast incongruent > neutral revealed clusters in the bilateral posterior parietal cortex, right superior/middle frontal gyrus, and the right anterior insula.
To determine whether the current task modulated activity in the anterior cingulate cortex (ACC), as shown in previous studies, a region-of-interest analysis was conducted using small volume correction. Based on the results reported in [bib_ref] Conflict monitoring versus selection-for-action in anterior cingulate cortex, Botvinick [/bib_ref] and [bib_ref] Anterior cingulated cortex, error detection, and the online monitoring of performance, Carter [/bib_ref] , we first defined a sphere with radius of 30 mm and centered at the coordinates (0, 23, 34). We then constrained this volume to not extend past x = − 15 or x = 15, to exclude white matter tracks. Finally, we masked the ROI to exclude corpus callosum. Using an uncorrected voxel-level threshold of p b 0.01 and a corrected cluster size threshold of p b 0.05, the contrast incongruent > congruent revealed a cluster in the right cingulate gyrus (x = 9, y = 14, z = 46; BA 32; t = 4.18).
# Discussion
The present findings demonstrate for the first time the effects of response competition on visual cortical responses to the target stimuli. Our study revealed two novel key findings. First, the presence of response competing distractors was associated with a greater V1 response to targets. Note that this comprised a selective increase in the V1 response to targets during incongruent trials compared to response-neutral trials, rather than a general boost in the response to targets in the presence of any distractor. Second, this effect on V1 responses to targets was negatively correlated with the magnitude of response competition effects: an increased V1 response to targets in the incongruent (vs. neutral) trials was associated with a reduced congruency cost to behavioral RT. These findings complement the previous literature, which has focused on the role of parietal and frontal cortex networks in the effects of response competition and their resolution (e.g. [bib_ref] Prefrontal regions involved in keeping information in and out of mind, Bunge [/bib_ref] [bib_ref] Dissociable contributions of prefrontal and parietal cortices to response selection, Bunge [/bib_ref] [bib_ref] Anterior cingulate cortex and conflict detection: an update of theory and data, Carter [/bib_ref] [bib_ref] Anterior cingulated cortex, error detection, and the online monitoring of performance, Carter [/bib_ref] [bib_ref] At the heart of the ventral attention system: the right anterior insula, Eckert [/bib_ref] [bib_ref] Cognitive and brain consequences of conflict, Fan [/bib_ref] [bib_ref] The activation of attentional networks, Fan [/bib_ref] [bib_ref] Neural activation during response competition, Hazeltine [/bib_ref] [bib_ref] Anterior cingulated conflict monitoring and adjustments in control, Kerns [/bib_ref] [bib_ref] Dissociating the role of the dorsolateral prefrontal and anterior cingulated cortex in..., Macdonald [/bib_ref] [bib_ref] The relative involvement of anterior cingulate and prefrontal cortex in attentional control..., Milham [/bib_ref] [bib_ref] Examining a supramodal network for conflict processing: a systematic review and novel..., Roberts [/bib_ref] [bib_ref] Anterior cingulate cortex, conflict monitoring, and levels of processing, Van Veen [/bib_ref]. These studies have identified increased activity in the superior parietal cortex, anterior cingulate cortex (ACC), right anterior insula, and the dorsolateral prefrontal cortex (DLPFC) when the distractor stimuli elicit competing responses. Major parts of this frontoparietal top-down network, including the right anterior insula, ACC, right superior/middle frontal gyrus and bilateral posterior parietal cortex, were also associated with response congruency in the present study, in replication of these previous results. Importantly, our critical new findings of a larger V1 response to targets in the presence of an incongruent distractor that was negatively correlated with the magnitude of distractor response competition effects on behavior offer a new mechanism of competition resolution that operates in the primary visual cortex.
Note that whereas the previous research established a conflict resolution mechanism that is manifested in reduced interference following high conflict trials (e.g. [bib_ref] Conflict monitoring versus selection-for-action in anterior cingulate cortex, Botvinick [/bib_ref] [bib_ref] Conflict monitoring and cognitive control, Botvinick [/bib_ref] [bib_ref] Multiple conflict-driven control mechanisms in the human brain, Egner [/bib_ref] [bib_ref] Optimizing the use of information: strategic control of activation of responses, Gratton [/bib_ref] , our effects are found during the presence of the potential response conflict (i.e. in the comparison of incongruent and neutral trials, rather than on trials following incongruent trials as in the conflict resolution literature e.g. [bib_ref] Conflict monitoring and cognitive control, Botvinick [/bib_ref]. The associated reduction in response competition effects on behavior suggests that the increased V1 response to targets in the face of conflicting distractors reflects an early sensory mechanism that protects the target stimulus from a potential response conflict by boosting its sensory signal in the presence of a response incongruent stimulus. An alternate potential explanation for the difference in activity between neutral and incongruent trials is the presence of one vs. two task-relevant stimuli. However, this is not supported by the data given the lack of difference between the neutral and congruent conditions, as the congruent condition also had two task-relevant stimuli.
Taken together with the findings that response competition resulted in fronto-parietal activation that is associated with top-down control, our research suggests the operation of a rapid top-down control mechanism that serves to control V1 responses to stimuli in accordance with the response-relevance of the context. Given the small receptive fields in V1, and the restricted information coding for visual properties, such a context-based mechanism must result from feedback connections to V1. Indeed, this is suggested by previous demonstrations of rapid response to visual stimuli as early as 45 ms in the parietal cortex and 90 ms in the prefrontal cortex [bib_ref] Flow of activation from V1 to frontal cortex in humans. A framework..., Foxe [/bib_ref] [bib_ref] Feedback to V1: a reverse hierarchy in vision, Juan [/bib_ref] , coupled with demonstrations that V1 has both early and late response periods with the later responses reflecting feedback activation (see also [bib_ref] Long-distance feedback projections to area V1: implications for multisensory integration, spatial awareness,..., Clavagnier [/bib_ref]. Our study thus adds to an increasing body of work that demonstrates a role for the primary visual cortex in attention [bib_ref] Stimulus specific responses from beyond the classical receptive field: neurophysiological mechanisms for..., Allman [/bib_ref] [bib_ref] Bottom-up dependent gating of frontal signals in early visual cortex, Ekstrom [/bib_ref] [bib_ref] Visuotopic cortical connectivity underlying attention revealed with white-matter tractography, Greenberg [/bib_ref] [bib_ref] Mechanisms of visual attention in the human cortex, Kastner [/bib_ref] [bib_ref] Functional MRI reveals spatially specific attentional modulation in human primary visual cortex, Somers [/bib_ref]. The present research also expands on recent work by [bib_ref] Working memory load modulates distractor competition in primary visual cortex, Kelley [/bib_ref] and [bib_ref] Rapid modulation of sensory processing induced by stimulus conflict, Appelbaum [/bib_ref] , who show modulation of visual cortex activity related to distractor congruency. [bib_ref] Working memory load modulates distractor competition in primary visual cortex, Kelley [/bib_ref] used fMRI to examine how working memory load alters the visual response to incongruent distracting images. They found that load interacted with distractor congruency, such that BOLD signals in area V1 were greater to incongruent than congruent distractors under conditions of high (versus low) working memory load. [bib_ref] Rapid modulation of sensory processing induced by stimulus conflict, Appelbaum [/bib_ref] used EEG recording to examine the event-related potentials (ERPs) evoked by lateralized letter flankers. They found a lateralized change in voltage over the occipital cortex in the presence of incongruent flankers, which co-occurred with fronto-parietal voltage changes typically observed in studies of response conflict. These voltages were shown to positively correlate with levels of distractor interference across subjects. The present study fills an important gap in these previous findings by examining target-related (rather than distractor-related) activity in the early visual cortex, specifically retinotopic area V1. Our correlation with response competition effects on behavior also provides insight into how the target related V1 effect may play a functional role on an early sensory resolution of any potential response conflict.
The present findings have implications for how cognitive control is examined in other paradigms and populations. For example, one might predict that in Stroop-type tasks [bib_ref] Half a century of research on the Stroop effect: an integrative review, Macleod [/bib_ref] top-down feedback from control regions would lead to increase activity in visual area V4 to improve color discrimination. Indeed evidence for occipital cortex activity during the Stroop-task being related to compensatory strategies in aging populations [bib_ref] Stroop interference, hemodynamic response and aging: an event-related fMRI study, Zysset [/bib_ref] suggests that this is a promising future research avenue.
One question raised by our study is how feedback to V1 (or input from V1) that allows for resolution of distractor congruency relates to connectivity between ACC, DLPFC and motor control regions. It is possible that, in the presence of an incongruent distractor, visual activity could be modulated in advance of, in concert with, or following modulation of motor cortex activity. One might even predict that two individuals that show the same level of behavioral interference rely differentially on sensory or motor resolution. While the limited temporal resolution of fMRI prevents a close examination of the temporal dynamics of the cortical regions in question, it is possible that an EEG study (investigating, for example, the timing of the lateralized readiness potential, see for example [bib_ref] Subsecond changes in top down control exerted by human medial frontal cortex..., Taylor [/bib_ref] could address this issue. Such a study would provide a more complete picture of how a variety of sensory, motor, attentional and cognitive control regions are recruited to prevent and resolve distractor-induced response competition.
[fig] Figure 1: Display example from a single trial. Solid circles indicate target locations; dashed circles indicate distractor locations. These locations were defined based on instructions to the participant. Circles were not present during the experiment. Target and distractor diagonals varied from participant to participant. [/fig]
[fig] Figure 2: BOLD responses to targets (top panel) and distractors (bottom panel) (measured as difference in GLM parameter estimates comparing stated trial type and null trials) as a function of trial type and region of cortex. Error bars represent SEM. Black bars: incongruent trials; gray bars: neutral trials; white bars: congruent trials. [/fig]
[table] Table 1: Behavioral performance as a function of distractor congruency. SEM listed in parentheses. [/table]
[table] Table 2: Regions showing greater BOLD signals for various full-brain contrasts.Fig. 3. Statistical maps for random effects analysis, showing results of contrasts incongruent > neutral (top row) and incongruent > neutral (bottom row). The maps show clusters of increased BOLD response in the anterior insula (left column, z = 2), superior/middle frontal gyrus (middle column, z = 26), and intraparietal sulcus/precuneus (right column, z = 45). Voxel-level threshold is p b 0.005, uncorrected, with a cluster level threshold of p b 0.05, corrected. Results are overlaid on a mean structural image created by averaging the spatially-normalized brains of all subjects. Slice depth is MNI coordinates. Right side of the image corresponds to right side of the brain. [/table]
|
Endoplasmic reticulum stress signaling and chemotherapy resistance in solid cancers
The unfolded protein response (UPR) is an adaptive cellular program used by eukaryotic cells to cope with protein misfolding stress. During tumor development, cancer cells are facing intrinsic (oncogene activation) and extrinsic (limiting nutrient or oxygen supply) challenges, with which they must cope to survive. Moreover, chemotherapy represents an additional extrinsic challenge that cancer cells are facing and to which they adapt in the case of resistance. As of today, resistance to chemotherapy and targeted therapies is one of the important issues that oncologists have to deal with for treating cancer patients. In this review, we first describe the key molecular mechanisms controlling the UPR and their implication in solid cancers. Then, we review the literature that connects cancer chemotherapy resistance mechanisms and activation of the UPR. Finally, we discuss the possible applications of targeting the UPR to bypass drug resistance.
# Introduction
The endoplasmic reticulum (ER) is the first intracellular compartment of the secretory pathway. It regulates calcium homeostasis, lipid biosynthesis and protein productive folding and quality control. About one-third of all the proteins transit through the ER 1-3 towards their final cellular or extracellular location. The synthesis of these proteins occurs on the cytosolic side of the ER and productive protein folding is orchestrated by elaborated ERresident molecular machines involving chaperones, foldases and quality control proteins. These molecular machines ensure protein biogenesis from their nascent form to their ER exportable form. [bib_ref] Adapting proteostasis for disease intervention, Balch [/bib_ref] However, in the course of this process, a significant proportion of proteins is not properly folded and fails ER protein quality control criteria. [bib_ref] Rapid degradation of a large fraction of newly synthesized proteins by proteasomes, Schubert [/bib_ref] These misfolded proteins are therefore addressed to the ER-associated degradation (ERAD) system that targets them to the cytosol for ubiquitinylation and proteasomal degradation. [bib_ref] Proteostasis control by the unfolded protein response, Hetz [/bib_ref] If the ER faces an important protein folding demand or sees its folding and degradation capacity attenuated, is needed, ER capacity to handle protein biogenesis are overwhelmed, thereby leading to an accumulation of improperly folded proteins in this compartment and to a situation called ER stress. ER stress leads to the activation of an adaptive response, named the unfolded protein response (UPR) that aims at (i) limiting misfolded proteins accumulation in the ER by transiently attenuating protein translation; (ii) augmenting the ER folding capacity by increasing the transcription of ER-resident chaperones proteins; (iii) enhancing protein clearance from the ER by increasing its degradation capacity. If the ER stress persists, the UPR triggers cell death. [bib_ref] UPR activation alters chemosensitivity of tumor cells, Mann [/bib_ref] [bib_ref] When ER stress reaches a dead end, Urra [/bib_ref] During cancer genesis, an acute demand of protein synthesis is needed to support different cellular functions such as tumor proliferation, migration and differentiation, often driven by oncogenic activation. [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] Tumor microenvironment might also provide limited tumor growth/development conditions because of important tumor oxygen and nutrient demands and inadequate vascularization. Therefore, cancer cells have to adapt to such a selective milieu with hypoxia, pH variation and nutrient deprivation that leads to cellular stress, [bib_ref] UPR activation alters chemosensitivity of tumor cells, Mann [/bib_ref] [bib_ref] The role of the unfolded protein response in tumour development: friend or..., Ma [/bib_ref] [bib_ref] Tumorigenic and immunosuppressive effects of endoplasmic reticulum stress in cancer, Cubillos-Ruiz [/bib_ref] [bib_ref] How does protein misfolding in the endoplasmic reticulum affect lipid metabolism in..., Wang [/bib_ref] by activating a range of cellular stress-response pathways including the UPR that will be described in the first part of this review.
Chemotherapy represents an additional source of cellular stress for cancer cells. Indeed, antitumor drugs emphasize the microenvironmental stress acting on the selection of drug-resistant cancer cells. [bib_ref] Molecular targeting therapy of cancer: drug resistance, apoptosis and survival signal, Tsuruo [/bib_ref] Resistance to chemotherapy is a principal problem in treating the most commonly seen solid tumors. Chemotherapy efficacy is indeed exposed to the multiple intrinsic and acquired resistance mechanisms developed by tumor cells that will be presented in the second part of this review. Furthermore, we will discuss the involvement of the ER stress-induced UPR to anticancer drug resistance. Understanding the UPR mechanisms associated with cancer drug resistance will provide insights to open new therapeutic avenues in which the association of standard chemotherapy with drugs targeting the UPR could overtake cancer drug resistance.
(inositol requiring enzyme 1α) [bib_ref] A stress response pathway from the endoplasmic reticulum to the nucleus requires..., Tirasophon [/bib_ref] and PERK (protein kinase RNA-activated-like ER kinase). [bib_ref] Protein translation and folding are coupled by an endoplasmic-reticulum-resident kinase, Harding [/bib_ref] The signaling pathways activated downstream of the three sensors lead to the reduction of protein misfolding, by slowing down de novo protein synthesis on the cytosolic side of the ER and by increasing protein folding and clearance in the ER [fig_ref] 1: CHOP [/fig_ref]. The activation of these three sensors is controlled by the ER-resident chaperone molecule GRP78/BiP (glucose-regulated protein 78/binding immunoglobulin protein). Indeed, under basal conditions, GRP78 constitutively associates with the luminal domains of the sensors through a noncanonical binding, thus preventing their activation. [bib_ref] Proteostasis control by the unfolded protein response, Hetz [/bib_ref] [bib_ref] Endoplasmic reticulum stress-activated cell reprogramming in oncogenesis, Chevet [/bib_ref] Upon accumulation of misfolded proteins, GRP78 dissociates from the sensors when misfolded proteins accumulate in the ER, through mechanism depending on its substrate binding domain. [bib_ref] Noncanonical binding of BiP ATPase domain to Ire1 and Perk is dissociated..., Carrara [/bib_ref] This induces IRE1α and PERK oligomerization and autotransphosphorylation [bib_ref] Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response, Bertolotti [/bib_ref] and the subsequent activation of the downstream signaling cascades. Moreover, BiP dissociation from AFT6α together with protein disulfide isomerase (PDI)-mediated disulfide bond modification [bib_ref] Endoplasmic reticulum stress-activated transcription factor ATF6alpha requires the disulfide isomerase PDIA5 to..., Higa [/bib_ref] [bib_ref] Role of disulfide bridges formed in the luminal domain of ATF6 in..., Nadanaka [/bib_ref] promotes ATF6α export to the Golgi complex. [bib_ref] ER stress regulation of ATF6 localization by dissociation of BiP/GRP78 binding and..., Shen [/bib_ref] [bib_ref] A synthetic biology approach identifies the mammalian UPR RNA ligase RtcB, Lu [/bib_ref] Activating transcription factor 6α. ER stress leads to ATF6α export from the ER to the Golgi apparatus where ATF6α proteolytic cleavage by S1P and S2P proteases releases an active membranefree form ATF6f, which therefore translocates to the nucleus and induces the transcription of genes mainly involved in protein folding and ERAD. [bib_ref] Endoplasmic reticulum stress-activated cell reprogramming in oncogenesis, Chevet [/bib_ref] [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] [bib_ref] Transcriptional induction of mammalian ER quality control proteins is mediated by single..., Yamamoto [/bib_ref] [bib_ref] XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response..., Yoshida [/bib_ref] Inositol requiring enzyme 1α. IRE1α is a type I ER-resident transmembrane protein. Its cytoplasmic domain presents two distinct molecular activities: a serine/threonine kinase and an endoribonuclease (RNase), resembling RNaseL. Upon ER stress, IRE1α dimerizes/oligomerizes and its trans-autophosphorylation induces a conformational change leading to endoribonuclease activation. [bib_ref] Proteostasis control by the unfolded protein response, Hetz [/bib_ref] The first substrate described for IRE1α RNase was X-box binding protein-1 (XBP1) mRNA that is processed together with the t-RNA ligase RTCB (RNA 2′,3′-cyclic phosphate and 5′-OH ligase) leading to a non-conventional mRNA splicing. [bib_ref] The mammalian tRNA ligase complex mediates splicing of XBP1 mRNA and controls..., Jurkin [/bib_ref] The resulting open reading frame is shifted and leads to the translation of a stable and active transcription factor, XBP1s. [bib_ref] IRE1-mediated unconventional mRNA splicing and S2P-mediated ATF6 cleavage merge to regulate XBP1..., Lee [/bib_ref] [bib_ref] IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1..., Calfon [/bib_ref] XBP1s activate the expression of genes involved in protein folding, secretion, ERAD and lipid synthesis. [bib_ref] Endoplasmic reticulum stress-activated cell reprogramming in oncogenesis, Chevet [/bib_ref] [bib_ref] The unfolded protein response: integrating stress signals through the stress sensor IRE1alpha, Hetz [/bib_ref] [bib_ref] XBP1 controls diverse cell type-and condition-specific transcriptional regulatory networks, Acosta-Alvear [/bib_ref] IRE1α RNase is also involved in ER-localized mRNA, ribosomal RNA and microRNAs degradation. [bib_ref] Decay of endoplasmic reticulum-localized mRNAs during the unfolded protein response, Hollien [/bib_ref] [bib_ref] Regulated Ire1-dependent decay of messenger RNAs in mammalian cells, Hollien [/bib_ref] [bib_ref] IRE1alpha kinase activation modes control alternate endoribonuclease outputs to determine divergent cell..., Han [/bib_ref] [bib_ref] Translational control by the ER transmembrane kinase/ribonuclease IRE1 under ER stress, Iwawaki [/bib_ref] [bib_ref] IRE1alpha induces thioredoxin-interacting protein to activate the NLRP3 inflammasome and promote programmed..., Lerner [/bib_ref] [bib_ref] IRE1alpha cleaves select microRNAs during ER stress to derepress translation of proapoptotic..., Upton [/bib_ref] [bib_ref] Getting RIDD of RNA: IRE1 in cell fate regulation, Maurel [/bib_ref] This activity is named regulated IRE1-dependent decay. Importantly, regulated IRE1-dependent decay selectivity is highly dependent on IRE1α oligomerization state and the cell type, the precise mechanisms of regulated IRE1-dependent decay activation are still debated. [bib_ref] Allosteric inhibition of the IRE1alpha RNase preserves cell viability and function during..., Ghosh [/bib_ref] [bib_ref] Ire1 has distinct catalytic mechanisms for XBP1/HAC1 splicing and RIDD, Tam [/bib_ref] [bib_ref] Peptides derived from the bifunctional kinase/RNase enzyme IRE1alpha modulate IRE1alpha activity and..., Bouchecareilh [/bib_ref] [bib_ref] IRE1alpha kinase activation modes control alternate endoribonuclease outputs to determine divergent cell..., Han [/bib_ref] PKR-like ER kinase. As for IRE1α, PERK is a type I ER-resident transmembrane protein. Upon ER stress, PERK trans-autophosphorylates and phosphorylates the translation initiation factor eIF2α (eukaryotic initiation factor 2α) and the transcription factor NRF2 (nuclear respiratory factor 2). Activated eIF2α attenuates global protein translation, reducing the folding demand on the ER 2,3,39,40 whereas activated NRF2 controls the antioxidant response. [bib_ref] Endoplasmic reticulum stress-activated cell reprogramming in oncogenesis, Chevet [/bib_ref] PERK-mediated eIF2α phosphorylation also triggers the translational activation of the transcription factor ATF4 that induces expression of genes involved in protein folding, amino-acid [fig_ref] 1: CHOP [/fig_ref]. The UPR sensors and their downstream partners. During ER stress, GRP78 is released from IRE1α, PERK and ATF6 sensors allowing their dimerization/oligomerization or export to the Golgi apparatus. PERK activation leads to phosphorylation of NRF2 and eIF2α. Phosphorylation of eIF2α induces global translation attenuation and prompts that of AFT4. ATF4 and NRF2 induce expression of genes involved in antioxidant response, protein folding, amino-acid metabolism, autophagy and apoptosis. The negative feedback loop activated downstream of PERK dephosphorylates eIF2α to restore translation. IRE1α activation leads to c-Jun N-terminal protein kinase (JNK) phosphorylation, regulated IRE1-dependent decay (RIDD) activity and XBP1 splicing that induces expression of genes involved protein folding, secretion, ERAD and lipid synthesis. Activation of ATF6 leads to its export in the Golgi apparatus where its cytosolic domain is released to translocate to the nucleus and activate the transcription of genes involved in protein folding and ERAD. Antioxid, antioxidant response; Lipid synth, lipid synthesis; QC, quality control. metabolism, autophagy and apoptosis [bib_ref] Proteostasis control by the unfolded protein response, Hetz [/bib_ref] [bib_ref] Endoplasmic reticulum stress-activated cell reprogramming in oncogenesis, Chevet [/bib_ref] [bib_ref] An integrated stress response regulates amino acid metabolism and resistance to oxidative..., Harding [/bib_ref] [bib_ref] ATF4, an ER stress and hypoxia-inducible transcription factor and its potential role..., Ye [/bib_ref] such as the apoptosis-related gene CEBP (CCAAT/enhancer-binding protein) homologous protein CHOP (CEBP homologous protein/growth arrest and DNA-damaged-inductible protein 153 (GADD153)) that impacts on the control of cell death/survival outputs upon ER stress. [bib_ref] ER-stress-induced transcriptional regulation increases protein synthesis leading to cell death, Han [/bib_ref] Moreover, PERK/eIF2α activation is negatively controlled by a feedback mechanism involving the protein GADD34 induced by this PERK pathway, which, in association with the phosphatase PP1c (protein phosphatase 1c), is responsible for the dephosphorylation of eIF2α. [bib_ref] Feedback inhibition of the unfolded protein response by GADD34-mediated dephosphorylation of eIF2alpha, Novoa [/bib_ref] UPR involvement in cancers The role of ER stress signaling as a key actor in cancer development has been first proposed in 2004 [bib_ref] The role of the unfolded protein response in tumour development: friend or..., Ma [/bib_ref] and is now largely accepted by both the scientific and medical communities. [bib_ref] Endoplasmic reticulum proteostasis in glioblastoma-from molecular mechanisms to therapeutic perspectives, Obacz [/bib_ref] For instance, increased expression levels of major actors of the UPR such as IRE1α, unspliced and spliced XBP1, PERK and ATF6 were observed in tissues sections from a variety of human tumors including brain, breast, gastric, kidney, liver, lung and pancreatic cancers [fig_ref] Table 1: Clinical evidences of UPR involvement in solid cancers [/fig_ref]. Moreover, the chaperone GRP78 is also found overexpressed in many cancers [bib_ref] Induction of the unfolded protein response drives enhanced metabolism and chemoresistance in..., Epple [/bib_ref] [bib_ref] The unfolded protein response regulator GRP78/BiP as a novel target for increasing..., Pyrko [/bib_ref] [bib_ref] GRP78 as a novel predictor of responsiveness to chemotherapy in breast cancer, Lee [/bib_ref] [bib_ref] Overexpression of the glucose-regulated stress gene GRP78 in malignant but not benign..., Fernandez [/bib_ref] [bib_ref] Activation and clinical significance of the unfolded protein response in breast cancer, Scriven [/bib_ref] [bib_ref] Anticipatory estrogen activation of the unfolded protein response is linked to cell..., Andruska [/bib_ref] [bib_ref] Expression of glucose regulated protein 78 (GRP78) determines colorectal cancer response to..., Mhaidat [/bib_ref] [bib_ref] Targeting the endoplasmic reticulum mediates radiation sensitivity in colorectal cancer, Drake [/bib_ref] [bib_ref] GRP78 regulates sensitivity of human colorectal cancer cells to DNA targeting agents, Mhaidat [/bib_ref] [bib_ref] Endoplasmic reticulum stress, unfolded protein response and development of colon adenocarcinoma, Piton [/bib_ref] [bib_ref] Upregulation of GRP78 in renal cell carcinoma and its significance, Fu [/bib_ref] [bib_ref] Unfolded protein response activation contributes to chemoresistance in hepatocellular carcinoma, Al-Rawashdeh [/bib_ref] [bib_ref] Activation of the ATF6, XBP1 and grp78 genes in human hepatocellular carcinoma:..., Shuda [/bib_ref] [bib_ref] CD147 induces UPR to inhibit apoptosis and chemosensitivity by increasing the transcription..., Tang [/bib_ref] [bib_ref] Endoplasmic reticulum ribosome-binding protein 1 (RRBP1) overexpression is frequently found in lung..., Tsai [/bib_ref] [bib_ref] Elevated GRP78 expression is associated with poor prognosis in patients with pancreatic..., Niu [/bib_ref] [bib_ref] A common genetic variation of melanoma inhibitory activity-2 labels a subtype of..., Kong [/bib_ref] and is involved in the dissemination/metastasis of human tumors. GRP78 overexpression is associated with higher tumor grades and reduced patients' survival. [bib_ref] A novel chemical, STF-083010, reverses tamoxifen-related drug resistance in breast cancer by..., Ming [/bib_ref] [bib_ref] GRP78 regulates sensitivity of human colorectal cancer cells to DNA targeting agents, Mhaidat [/bib_ref] [bib_ref] Upregulation of GRP78 in renal cell carcinoma and its significance, Fu [/bib_ref] [bib_ref] Activation of the ATF6, XBP1 and grp78 genes in human hepatocellular carcinoma:..., Shuda [/bib_ref] [bib_ref] Elevated GRP78 expression is associated with poor prognosis in patients with pancreatic..., Niu [/bib_ref] [bib_ref] Synthetic vulnerabilities of mesenchymal subpopulations in pancreatic cancer, Genovese [/bib_ref] In experimental models including tumor cell lines and mouse tumor xenografts, GRP78 was also shown to have an important role in regulating cancer hallmarks [fig_ref] Table 2: Cellular models demonstrating the importance of UPR in solid cancers [/fig_ref]. [bib_ref] Induction of the unfolded protein response drives enhanced metabolism and chemoresistance in..., Epple [/bib_ref] [bib_ref] The unfolded protein response regulator GRP78/BiP as a novel target for increasing..., Pyrko [/bib_ref] [bib_ref] Activation and clinical significance of the unfolded protein response in breast cancer, Scriven [/bib_ref] [bib_ref] Expression of glucose regulated protein 78 (GRP78) determines colorectal cancer response to..., Mhaidat [/bib_ref] [bib_ref] Overexpression of human X-box binding protein 1 (XBP-1) in colorectal adenomas and..., Fujimoto [/bib_ref] [bib_ref] Targeting the endoplasmic reticulum mediates radiation sensitivity in colorectal cancer, Drake [/bib_ref] [bib_ref] GRP78 regulates sensitivity of human colorectal cancer cells to DNA targeting agents, Mhaidat [/bib_ref] [bib_ref] Upregulation of GRP78 in renal cell carcinoma and its significance, Fu [/bib_ref] [bib_ref] Unfolded protein response activation contributes to chemoresistance in hepatocellular carcinoma, Al-Rawashdeh [/bib_ref] [bib_ref] Activation of the ATF6, XBP1 and grp78 genes in human hepatocellular carcinoma:..., Shuda [/bib_ref] [bib_ref] Elevated GRP78 expression is associated with poor prognosis in patients with pancreatic..., Niu [/bib_ref] [bib_ref] A common genetic variation of melanoma inhibitory activity-2 labels a subtype of..., Kong [/bib_ref] [bib_ref] Arginine deprivation induces endoplasmic reticulum stress in human solid cancer cells, Bobak [/bib_ref] [bib_ref] LOXL2 drives epithelial-mesenchymal transition via activation of IRE1-XBP1 signalling pathway, Cuevas [/bib_ref] [bib_ref] The unfolded protein response potentiates epithelial-to-mesenchymal transition (EMT) of gastric cancer cells..., Shen [/bib_ref] [bib_ref] Silencing glucoseregulated protein 78 induced renal cell carcinoma cell line G1 cell-cycle..., Lin [/bib_ref] [bib_ref] RACK1 modulates apoptosis induced by sorafenib in HCC cells by interfering with..., Zhou [/bib_ref] [bib_ref] Compounds triggering ER stress exert anti-melanoma effects and overcome BRAF inhibitor resistance, Cerezo [/bib_ref] For example, GRP78 regulates tumor cell proliferation and migration. [bib_ref] The unfolded protein response regulator GRP78/BiP as a novel target for increasing..., Pyrko [/bib_ref] [bib_ref] Upregulation of GRP78 in renal cell carcinoma and its significance, Fu [/bib_ref] [bib_ref] Elevated GRP78 expression is associated with poor prognosis in patients with pancreatic..., Niu [/bib_ref] Tumor progression is characterized by UPR activation induced by the challenging growth conditions associated with hypoxia and anticancers drugs. [bib_ref] Anticipatory estrogen activation of the unfolded protein response is linked to cell..., Andruska [/bib_ref] Furthermore, tumor cells develop specific metabolic processes to adapt to such environment, [bib_ref] Cellular mechanisms of endoplasmic reticulum stress signaling in health and disease. 3...., Manie [/bib_ref] and examples of highly dynamic network between cancer cells' adaptation and resistance to environmental stresses and UPR signaling pathways will be illustrated in the following section.
UPR linked to cancer initiation. In the normal gastrointestinal tract, a differential expression of GPR78 is observed and is lower in intestinal stem cells and higher in more differentiated transit amplifying cells. [bib_ref] ER stress causes rapid loss of intestinal epithelial stemness through activation of..., Heijmans [/bib_ref] Interestingly, most of the colorectal cancers (CRCs) derive from transformed intestinal stem cell in which activation of the PERK/eIF2α axis is associated with the loss of stemness. [bib_ref] Stem cell dynamics in homeostasis and cancer of the intestine, Vermeulen [/bib_ref] This suggests that cancer initiation might be linked to ER stress in the gastrointestinal tract. [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] Remarkably, in a colitisassociated cancer model, the IRE1α pathway appears to have an important role in mediating ER stress that induces intestinal stem cell expansion. [bib_ref] ER stress transcription factor Xbp1 suppresses intestinal tumorigenesis and directs intestinal stem..., Niederreiter [/bib_ref] Indeed, XBP1 loss in epithelial cells results in intestinal stem cell hyperproliferation, therefore promoting initiating phases of cancer development. [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] UPR linked to tumor quiescence and aggressiveness. Cancer cells must cope with strict growth conditions forced by their intrinsic condition (oncogene expression) but also by the tumor environment including chemotherapy, nutrient starvation and in vivo microenvironmental challenges. They therefore develop adaptive mechanisms such as a metabolic resting state called quiescence/ dormancy. Regulation of tumor cell dormancy has been associated with the activation of both ATF6α and PERK-eIF2α. Both pathways were identified as a survival factors for quiescent but not proliferative squamous carcinoma cells [bib_ref] ATF6alpha-Rheb-mTOR signaling promotes survival of dormant tumor cells in vivo, Schewe [/bib_ref] and under hypoxia, [bib_ref] The PERK/eIF2alpha/ATF4 module of the UPR in hypoxia resistance and tumor growth, Fels [/bib_ref] respectively. In triple-negative breast cancers, the IRE1α/XBP1s axis is found constitutively active, thereby conferring higher aggressiveness due to XBP1s-mediated hypoxia-inducible factor-1α activation. [bib_ref] XBP1 promotes triple-negative breast cancer by controlling the HIF1alpha pathway, Chen [/bib_ref] In glioblastoma (GBM), tumor migration/invasion is associated to aggressiveness. Interestingly, IRE1α endoribonuclease activity regulates the extracellular matrix protein SPARC (secreted protein acidic and rich in cysteine) itself involved in tumor invasion. [bib_ref] Autocrine control of glioma cells adhesion and migration through IRE1alpha-mediated cleavage of..., Dejeans [/bib_ref] UPR-linked 'secretory switch' in cancer cells. To sustain their own important metabolic demands and to adapt to their challenging environment, cancer cells reprogram their secretome and the associated secretory pathway needed to support tumor functions and necessary for cancer progression. [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] [bib_ref] Hallmarks of cancer: the next generation, Hanahan [/bib_ref] For instance, tumor invasion is facilitated by change in secreted extracellular matrix components and matrix metalloproteases. [bib_ref] Why don't we get more cancer? A proposed role of the microenvironment..., Bissell [/bib_ref] [bib_ref] Outgrowth of single oncogene-expressing cells from suppressive epithelial environments, Leung [/bib_ref] Tumor cell proliferation and neoangiogenesis (see below) are sustained through the secretion of growth factors, cytokines and chemokines. [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] As ER is the major site of protein production that also orchestrates their secretion, activation of the UPR strongly modulates tumor cells' secretory switch during cancer development.
UPR linked to tumor epithelial-to-mesenchymal transition. Epithelial-to-mesenchymal transition (EMT) is a physiological process used by cancer cells to acquire critical oncogenic features such as migration/invasion, stemness and drug resistance. [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] EMT is controlled by specific transcription factors involved in these cell functions and the UPR has been often involved in the expression of these transcription factors. For instance, in breast tumors, increased expression of XBP1s is observed in metastatic tumors, which correlates with the EMT inducer SNAIL (snail-related protein). 85 LOXL2 (lysyl oxidase like 2)/GRP78 interaction in the ER also activates the IRE1-XBP1 signaling pathway thereby inducing the expression of several EMT-linked transcription factors including SNAI1 (snail family transcriptional repressor), SNAI2, ZEB2 (zinc-finger E-box-binding homeobox 2) and TCF3 (transcription factor 3). [bib_ref] LOXL2 drives epithelial-mesenchymal transition via activation of IRE1-XBP1 signalling pathway, Cuevas [/bib_ref] Moreover, the overexpression of the TWIST (twist-related protein) transcription factor correlates with PERK constitutive activation. [bib_ref] Epithelial-to-mesenchymal transition activates PERK-eIF2alpha and sensitizes cells to endoplasmic reticulum stress, Feng [/bib_ref] The 'secretory switch' induced by UPR might also contribute to EMT. [bib_ref] Epithelial-to-mesenchymal transition activates PERK-eIF2alpha and sensitizes cells to endoplasmic reticulum stress, Feng [/bib_ref] [bib_ref] ER stress is associated with dedifferentiation and an epithelial-to-mesenchymal transitionlike phenotype in..., Ulianich [/bib_ref] [bib_ref] De-differentiation confers multidrug resistance via noncanonical PERK-Nrf2 signaling, Vecchio [/bib_ref] Indeed, overexpression of Serpin B3, a serine/cysteine protease inhibitor, is associated with chronic UPR induction leading to nuclear factor-κB activation and interleukin-6 production. This results in an EMT-like phenotype in mammary epithelial cells. [bib_ref] SCCA1/SERPINB3 promotes oncogenesis and epithelial-mesenchymal transition via the unfolded protein response and..., Sheshadri [/bib_ref] In GBM, dominant-negative form of IRE1α modulates the expression molecules involved in extracellular matrix structures, angiogenesis and inflammatory chemokines, thus reflecting a mesenchymal drift. [bib_ref] Inositolrequiring enzyme 1alpha is a key regulator of angiogenesis and invasion in..., Auf [/bib_ref] UPR-linked tumor angiogenesis. Expression of proangiogenic factors is affected by the UPR in cancer cells. For instance, vascular endothelial growth factor-A (VEGF-A), interleukin-1β and interleukin-6 are induced downstream of IRE1α signaling in GBM cells. [bib_ref] Inositolrequiring enzyme 1alpha is a key regulator of angiogenesis and invasion in..., Auf [/bib_ref] [bib_ref] IRE1 signaling is essential for ischemia-induced vascular endothelial growth factor-A expression and..., Drogat [/bib_ref] Moreover, IRE1α-mediated mRNA cleavage of the circadian gene PERIOD1, 92 an important mediator of GBM infiltration, also supports tumor angiogenesis through the regulation of the CXCL3 chemokine. [bib_ref] Inositolrequiring enzyme 1alpha is a key regulator of angiogenesis and invasion in..., Auf [/bib_ref] Furthermore, in response to hypoxia, VEGF is also upregulated by the PERK-ATF4 branch of the UPR to induce angiogenesis. [bib_ref] Endoplasmic reticulum stress-activated cell reprogramming in oncogenesis, Chevet [/bib_ref] [bib_ref] Novel roles of the unfolded protein response in the control of tumor..., Dejeans [/bib_ref] [bib_ref] Cellular mechanisms of endoplasmic reticulum stress signaling in health and disease. 3...., Manie [/bib_ref] [bib_ref] Perk-dependent translational regulation promotes tumor cell adaptation and angiogenesis in response to..., Blais [/bib_ref] Interestingly, the UPR-regulated ER chaperone ORP150 (oxygen-regulated protein 150) controls tumor angiogenesis by promoting the secretion of VEGF in prostatic and glioma cancer cells. [bib_ref] Regulation of tumor angiogenesis by oxygen-regulated protein 150, an inducible endoplasmic reticulum..., Ozawa [/bib_ref] [bib_ref] Antitumor effect of reduction of 150-kDa oxygen-regulated protein expression on human prostate..., Miyagi [/bib_ref] UPR-linked tumor metabolic processes. Under nutrient deprivation, cancer cells adapt their metabolic demand in part through activation of the UPR. Downstream of IRE1α, XBP1s activates the expression of key enzymes of the hexosamine biosynthetic pathway that convert glucose to UDP-acetylglucosamine. [bib_ref] Spliced X-box binding protein 1 couples the unfolded protein response to hexosamine..., Wang [/bib_ref] [bib_ref] Hexosamine pathway metabolites enhance protein quality control and prolong life, Denzel [/bib_ref] These are substrates for the O-and N-glycosylation of proteins, thereby improving global proteotasis. In addition, through hypoxia-inducible factor-1α activation, XBP1s also actively promotes glucose uptake in triple-negative breast cancer cells, which in turn upregulates the expression of several proteins involved in glycolytic processes including the glucose transporter 1. 98 ER stress and cancer chemotherapy resistance T Avril et al UPR linked to tumor autophagy. Autophagy is a cellular process that allows cancer cells to generate additional energy supplies through the selective or non-selective degradation of protein aggregates or damaged organelles. Under hypoxia, activation of the PERK/eIF2α/ATF4 pathway is protective for tumor cells through autophagy induction via LC3B (autophagy protein microtubule-associated protein 1 light chain 3b) and ATG5 (autophagy protein 5). [bib_ref] Activating transcription factor 4 is translationally regulated by hypoxic stress, Blais [/bib_ref] [bib_ref] ER stress-regulated translation increases tolerance to extreme hypoxia and promotes tumor growth, Bi [/bib_ref] [bib_ref] The unfolded protein response protects human tumor cells during hypoxia through regulation..., Rouschop [/bib_ref] Similarly, TNF receptor associated factor 2 (TRAF2)/IRE1α activates c-Jun N-terminal protein kinase that also induces autophagy. [bib_ref] Autophagy is activated for cell survival after endoplasmic reticulum stress, Ogata [/bib_ref] CHEMOTHERAPY RESISTANCE INDUCED BY UPR General mechanisms of resistance to chemotherapy in cancer During the past decades, chemotherapy and targeted therapies have become the principal modes of treatment against cancers [fig_ref] Table 3: Standard chemotherapy treatments and their targets in solid DNA-damage repair [/fig_ref] , but their efficacy is confronted to the multiple intrinsic and acquired resistance mechanisms developed by tumor cells before and during the treatment. These resistance mechanisms can include the reduction of drug uptake, the alteration of the drug target, the induction of drug-detoxifying mechanisms, repair of drug-induced damages and insensitivity to drug-induced cell death . [bib_ref] Mechanisms of cancer drug resistance, Gottesman [/bib_ref] [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] Resistance to anticancer drug accumulation. Drugs enter into tumor cells by three main routes: diffusion, active transport and endocytosis. [bib_ref] Mechanisms of cancer drug resistance, Gottesman [/bib_ref] However, tumor cells use several mechanisms to limit this entry by decreasing the uptake or increasing the efflux of the drug. [bib_ref] Mechanisms of cancer drug resistance, Gottesman [/bib_ref] For instance, the family of multidrug resistance proteins, acting as drug efflux pumps (reviewed in Chen and Tiwari 106 and Sodani et al. [bib_ref] Multidrug resistance associated proteins in multidrug resistance, Sodani [/bib_ref] , is the subject of intense research to characterize the role in chemotherapy resistance. [bib_ref] Molecular targeting therapy of cancer: drug resistance, apoptosis and survival signal, Tsuruo [/bib_ref] [bib_ref] Mechanisms of cancer drug resistance, Gottesman [/bib_ref] Expression of these proteins has been reported to correlate with resistance to chemotherapy in vitro. [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] Modulation of their functions is also correlated to in vitro chemosensitivity to drugs such as cisplatin, doxorubicin, paclitaxel and vincristine in several cancer cell lines. [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] [bib_ref] A family of drug transporters: the multidrug resistance-associated proteins, Borst [/bib_ref] In addition, modulation of the expression of cell surface transporters or their mutations can reduce drug uptake. As such, in osteosarcoma, both decreased expression and mutations of the methotrexate transporter reduced folate carrier that reduce their drug affinity have been reported. [bib_ref] Mechanisms of cancer drug resistance, Gottesman [/bib_ref] [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] [bib_ref] Chemotherapy resistance in osteosarcoma: current challenges and future directions, Chou [/bib_ref] Finally, cancer cell mutants that have defective endocytosis are resistant to immunotoxins that enter into tumor cells by endocytosis. [bib_ref] Mechanisms of cancer drug resistance, Gottesman [/bib_ref] Induction of drug-detoxifying mechanisms. Both drug inactivation and the absence of drug activation are specific for given classes of drugs. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] For instance, 5-fluorouracil (5-FU) is catabolized by dihydropyrimidine dehydrogenase that confers in vitro resistance to 5-FU once overexpressed in CRCs. [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] Platinum drugs such as cisplatin, carboplatin and oxaliplatin can also be inactivated after covalent linkage to the thiol glutathione, decreasing the availability of the native drug to bind its target 104,108 and leading to drug efflux by ABC transporter proteins. [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] High levels of glutathione have been found in tumor cells resistant to platinum drugs. Interestingly, expression of glutathione S-transferase-π, a member of the family of glutathione S-transferase that catalyzes glutathione conjugation, is linked to overall survival following cisplatin treatment of head and neck cancers and to cisplatin resistance of ovarian cancers. [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] [bib_ref] Chemotherapy resistance in osteosarcoma: current challenges and future directions, Chou [/bib_ref] Modification of drug targets. Drug sensitivity is affected by alterations of the drug target, such as mutations and/or changes in expression level. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] For instance, 5-FU and pemetrexed treatments inhibit translation of their target mRNA thymidylate synthase (TS), [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] thus leading to increased TS expression level and increased 5-FU resistance. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] Moreover, the overexpression and/or oncogenic mutations in many protein tyrosine kinases have been described in human cancers, rendering difficult the anti-protein tyrosine kinase targeting therapies. Indeed, efficacy of epidermal growth factor receptor (EGFR) inhibitors such as gefitinib and erlotinib is markedly reduced in non-small-cell lung cancers exhibiting the EGFR-T790M mutation. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] Amplification and mutations in anaplastic lymphoma kinase have been [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] Similarly, DNA topoisomerase-II, a target of doxorubicin and etoposide, is mutated in resistant cancer cell lines. [bib_ref] Molecular mechanisms of drug resistance, Longley [/bib_ref] Reduction of DNA topoisomerase-II expression by post-transcriptional modifications such as ubiquitination and sumoylation also leads to drug resistance and reduction of DNA damage. [bib_ref] UPR activation alters chemosensitivity of tumor cells, Mann [/bib_ref] [bib_ref] UPR-induced resistance to etoposide is downstream of PERK and independent of changes..., Mann [/bib_ref] In normal cells, DNA lesions are quickly recognized by DNA-damage response factors, which activate cell cycle checkpoints and direct DNA repair. [bib_ref] The effects of deregulated DNA damage signalling on cancer chemotherapy response and..., Bouwman [/bib_ref] Consequently, the regulation of DNA repair systems in tumor cells is a critical factor for their response to chemotherapeutics. [bib_ref] The effects of deregulated DNA damage signalling on cancer chemotherapy response and..., Bouwman [/bib_ref] For instance platinum-induced DNA damage is repaired by the nucleotide excision repair pathway and in vitro correlation between enhanced nucleotide excision repair and resistance to cisplatin has been reported in many studies. [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] High expression of ERCC1 (excision repair crosscomplementing 1), one of the key components of nucleotide excision repair, is linked to poor response to chemotherapy in numerous cancer types. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] In addition, mutation and/or downregulation of key DNA mismatch repair proteins such as MLH1 (mutL homolog 1) is observed in cisplatin-resistant tumors. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] [bib_ref] Chemotherapy resistance in osteosarcoma: current challenges and future directions, Chou [/bib_ref] Activation of antiapoptotic and prosurvival pathways. Most tumors develop defects in the common cell death pathways that lead to chemotherapy resistance. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] For instance, levels of BIM (Bcl-2 interacting mediator of cell death), a proapoptotic protein of the Bcl-2 (B-cell lymphoma) family, predict clinical responsiveness to EGFR and ERBB2 inhibitors. Moreover, a germline deletion in BIM gene is significantly associated with resistance to protein tyrosine kinase inhibitors in patients with EGFR-mutant lung cancers. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] Expression levels of MCL1, another member of the Bcl-2 family, are important determinant of resistance to Bcl-2 inhibitor ABT-737 and other cytotoxic chemotherapeutics. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] Furthermore, under chemotherapy pressure, tumors develop novel survival signaling pathways that contribute to drug resistance. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] An important number of proteins is involved in these pathways: oncogenes such as RAS and AKT (v-Akt murine thymoma viral oncogene homolog); tumor suppressor genes such as TP53 (tumor protein 53) and PTEN (phosphatase and tensin homolog); and prosurvival factors as nuclear factor-κB and signal transducer and activator of transcription 3. 104,108 Mutations, amplifications, chromosomal translocations and overexpression of these genes are associated with various malignancies and linked to resistance to chemotherapy and targeted therapies. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] Other factors involved in drug resistance. The influence of the local tumor microenvironment is identified as important contributor to chemotherapy resistance. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] For instance, hypoxia enhances drug detoxification by interfering with the generation of oxygen radicals and by increasing hypoxia-inducible factor-1mediated activation of survival signals. [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] Furthermore tumor heterogeneity at the genetic, molecular and cellular levels contributes substantially to chemotherapy resistance. For instance, the presence of cancer stem cells with robust intrinsic drug . General mechanisms involved in chemotherapy resistance. Tumor cells limit chemotherapy drugs accumulation by modifying their membrane composition, reducing drug transporters and increasing efflux pumps. Mechanisms of detoxification lead to drug inactivation. Drug target modification or loss also contributes to chemotherapy resistance. Finally, DNA damage and apoptosis induced by anticancer drugs are prevented by sophisticated DNA repair system and upregulation of prosurvival genes.
resistance capabilities reduces the chemotherapy efficacy. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] In solid tumors, the stroma (extracellular matrix, cancer-associated fibroblasts, immune and inflammatory cells and blood vessels) protects cancer cells from cytotoxic agents, thus allowing them to evade apoptosis and to develop acquired resistance leading to disease relapse. [bib_ref] Molecular targeting therapy of cancer: drug resistance, apoptosis and survival signal, Tsuruo [/bib_ref] [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] Recently, EMT has been associated with chemotherapy and targeted therapy resistance. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] Finally, as most anticancer drugs are primarily targeted against proliferating cancer cells, a significant proportion of cancer cells are in a dormancy/quiescent state, thereby exhibiting a degree of drug resistance linked to their decreased ability to proliferate. [bib_ref] Molecular targeting therapy of cancer: drug resistance, apoptosis and survival signal, Tsuruo [/bib_ref] [bib_ref] Ovarian cancer: strategies for overcoming resistance to chemotherapy, Agarwal [/bib_ref] Chemotherapy resistance induced by the UPR UPR activation is commonly observed in various tumor specimens (see UPR involvement in cancers) and correlates with drug resistance. Clinical evidences and in vitro demonstrations of tight link between UPR activation and drug resistance will be first reviewed in this section. The link between UPR and cellular adaptation of cancer cells including autophagy and hypoxia that also contributes to antidrug resistance will be presented in the next paragraphs .
Clinical relevance of the UPR activation and chemotherapy resistance. Clinical evidences of such phenomenon are almost exclusively limited to breast cancers [fig_ref] Table 4: Clinical evidences of UPR involvement in cancer chemotherapy resistance [/fig_ref]. [bib_ref] GRP78 as a novel predictor of responsiveness to chemotherapy in breast cancer, Lee [/bib_ref] [bib_ref] Anticipatory estrogen activation of the unfolded protein response is linked to cell..., Andruska [/bib_ref] [bib_ref] Expression and splicing of the unfolded protein response gene XBP-1 are significantly..., Davies [/bib_ref] [bib_ref] GRP78 as potential predictor for breast cancer response to adjuvant taxane therapy, Lee [/bib_ref] [bib_ref] Calnexin, an ER-induced protein, is a prognostic marker and potential therapeutic target..., Ryan [/bib_ref] Indeed, expression of the UPR sensors and their downstream partners are correlated with resistance to tamoxifen, thereby leading to decreased time to recurrence and poor survival. [bib_ref] Anticipatory estrogen activation of the unfolded protein response is linked to cell..., Andruska [/bib_ref] Interestingly, opposite effects are observed with the expression of XBP1u and XBP1s. XBP1u is associated with longer survival of breast patients treated with tamoxifen, whereas XBP1s is associated with shorter survival. [bib_ref] Expression and splicing of the unfolded protein response gene XBP-1 are significantly..., Davies [/bib_ref] This underlines IRE1α involvement in tamoxifen resistance. In contrast, GRP78 involvement seems to be more complex. High GRP78 expression in breast cancer specimens predicts a shorter recurrence-free survival in patients who received doxorubicin-based adjuvant chemotherapy. However, the opposite effect is observed in patients treated with doxorubicin and cyclophosphamide, followed by taxane (paclitaxel or docetaxel) on a clinical trial, where GRP78-positive staining predicts a better recurrence-free survival. [bib_ref] GRP78 as potential predictor for breast cancer response to adjuvant taxane therapy, Lee [/bib_ref] These results underline the possibility of use combined anticancer drugs to overcome cancer resistance .
## Induction of upr-dependent chemotherapy resistance in vitro .
Correlations between UPR activation and chemotherapy resistance were mainly demonstrated in cellular models in many types of cancer [fig_ref] Table 5: Cellular models demonstrating the importance of UPR in cancer chemotherapy resistance Abbreviations [/fig_ref]. [bib_ref] Induction of the unfolded protein response drives enhanced metabolism and chemoresistance in..., Epple [/bib_ref] [bib_ref] The unfolded protein response regulator GRP78/BiP as a novel target for increasing..., Pyrko [/bib_ref] [bib_ref] Activation and clinical significance of the unfolded protein response in breast cancer, Scriven [/bib_ref] [bib_ref] Anticipatory estrogen activation of the unfolded protein response is linked to cell..., Andruska [/bib_ref] [bib_ref] A novel chemical, STF-083010, reverses tamoxifen-related drug resistance in breast cancer by..., Ming [/bib_ref] [bib_ref] Expression of glucose regulated protein 78 (GRP78) determines colorectal cancer response to..., Mhaidat [/bib_ref] [bib_ref] GRP78 regulates sensitivity of human colorectal cancer cells to DNA targeting agents, Mhaidat [/bib_ref] [bib_ref] Unfolded protein response activation contributes to chemoresistance in hepatocellular carcinoma, Al-Rawashdeh [/bib_ref] [bib_ref] CD147 induces UPR to inhibit apoptosis and chemosensitivity by increasing the transcription..., Tang [/bib_ref] [bib_ref] Endoplasmic reticulum ribosome-binding protein 1 (RRBP1) overexpression is frequently found in lung..., Tsai [/bib_ref] [bib_ref] Silencing glucoseregulated protein 78 induced renal cell carcinoma cell line G1 cell-cycle..., Lin [/bib_ref] [bib_ref] RACK1 modulates apoptosis induced by sorafenib in HCC cells by interfering with..., Zhou [/bib_ref] [bib_ref] Endoplasmic reticulum chaperone protein GRP78 protects cells from apoptosis induced by topoisomerase..., Reddy [/bib_ref] [bib_ref] Activation of unfolded protein response protects osteosarcoma cells from cisplatin-induced apoptosis through..., Yan [/bib_ref] [bib_ref] Radicol, a novel trinorguaiane-type sesquiterpene, induces temozolomide-resistant glioma cell apoptosis via ER..., Li [/bib_ref] [bib_ref] Nelfinavir targets multiple drug resistance mechanisms to increase the efficacy of doxorubicin..., Chakravarty [/bib_ref] [bib_ref] GRP78/BiP inhibits endoplasmic reticulum BIK and protects human breast cancer cells against..., Fu [/bib_ref] [bib_ref] Human X-box binding protein-1 confers both estrogen independence and antiestrogen resistance in..., Gomez [/bib_ref] [bib_ref] Knockdown of estrogen receptor-alpha induces autophagy and inhibits antiestrogen-mediated unfolded protein response..., Cook [/bib_ref] [bib_ref] NF-kappaB signaling is required for XBP1 (unspliced and spliced)-mediated effects on antiestrogen..., Hu [/bib_ref] [bib_ref] Synergistic promotion of breast cancer cells death by targeting molecular chaperone GRP78..., Li [/bib_ref] [bib_ref] Glucose-regulated protein 78 as a novel effector of BRCA1 for inhibiting stress-induced..., Yeung [/bib_ref] [bib_ref] Lifetime genistein intake increases the response of mammary tumors to tamoxifen in..., Zhang [/bib_ref] [bib_ref] Activation of the unfolded protein response bypasses trastuzumab-mediated inhibition of the PI-3K..., Kumandan [/bib_ref] [bib_ref] Inhibition of endoplasmic reticulum (ER) stress sensors sensitizes cancer stem-like cells to..., Fujimoto [/bib_ref] [bib_ref] Functional coupling of p38-induced up-regulation of BiP and activation of RNA-dependent protein..., Ranganathan [/bib_ref] [bib_ref] Activation of the unfolded protein response is necessary and sufficient for reducing..., Gray [/bib_ref] A vast number of these studies demonstrate the impact of GRP78 expression on drug resistance mainly involving a reduced effect of drug-induced apoptosis. [bib_ref] The unfolded protein response regulator GRP78/BiP as a novel target for increasing..., Pyrko [/bib_ref] [bib_ref] Expression of glucose regulated protein 78 (GRP78) determines colorectal cancer response to..., Mhaidat [/bib_ref] [bib_ref] Unfolded protein response activation contributes to chemoresistance in hepatocellular carcinoma, Al-Rawashdeh [/bib_ref] [bib_ref] Endoplasmic reticulum ribosome-binding protein 1 (RRBP1) overexpression is frequently found in lung..., Tsai [/bib_ref] [bib_ref] Endoplasmic reticulum chaperone protein GRP78 protects cells from apoptosis induced by topoisomerase..., Reddy [/bib_ref] [bib_ref] Activation of unfolded protein response protects osteosarcoma cells from cisplatin-induced apoptosis through..., Yan [/bib_ref] [bib_ref] GRP78/BiP inhibits endoplasmic reticulum BIK and protects human breast cancer cells against..., Fu [/bib_ref] [bib_ref] NF-kappaB signaling is required for XBP1 (unspliced and spliced)-mediated effects on antiestrogen..., Hu [/bib_ref] [bib_ref] Glucose-regulated protein 78 as a novel effector of BRCA1 for inhibiting stress-induced..., Yeung [/bib_ref] [bib_ref] Inhibition of endoplasmic reticulum (ER) stress sensors sensitizes cancer stem-like cells to..., Fujimoto [/bib_ref] [bib_ref] Functional coupling of p38-induced up-regulation of BiP and activation of RNA-dependent protein..., Ranganathan [/bib_ref] However, the precise molecular mechanisms involved remain to be discovered. In chemotherapy-resistant breast cancer cells, GRP78 suppresses doxorubicin-mediated apoptosis in part through inhibition of BAX (Bcl-2-associated X protein) and caspase-7 activation. 49 GRP78 also forms complexes with BIK (Bcl-2-interacting killer), an apoptotic BH3-only protein, and blocks its apoptotic activity under estrogen starvation. [bib_ref] GRP78/BiP inhibits endoplasmic reticulum BIK and protects human breast cancer cells against..., Fu [/bib_ref] Finally, the PDIA5/ATF6α activation loop was described to be essential to confer imatinib resistance in K562 leukemia cells. [bib_ref] Endoplasmic reticulum stress-activated transcription factor ATF6alpha requires the disulfide isomerase PDIA5 to..., Higa [/bib_ref] The direct involvement of the UPR sensors in other mechanisms associated with cancer resistance to chemotherapy (i.e. reduction of anticancer drug accumulation, . The UPR intervention in chemotherapy resistance. UPR activation contributes to chemotherapy drug resistance by increasing drug detoxification and efflux pump expression, by modulating drug targets and activating antiapoptotic and prosurvival genes expression. Examples of anticancer drugs used several cancer types described in the literature are indicated.
drug-detoxifying mechanisms, modification of drug targets and DNA-damage repair) is up to now rather limited. For instance, a role for PERK in chemotherapy-resistant HT29 colon cancer cells has been involved in the upregulation of MDR related protein 1 through the regulation of NRF2. [bib_ref] PERK induces resistance to cell death elicited by endoplasmic reticulum stress and..., Salaroglio [/bib_ref] UPR and cellular adaptation links to cancer chemotherapy resistance. Different anticancer treatments, including those that stimulate ER stress, activate autophagy in tumor cells, which has been proposed to either enhance cancer cell death or act as a mechanism of resistance to chemotherapy. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] [bib_ref] Detecting autophagy in response to ER stress signals in cancer, Salazar [/bib_ref] Indeed, autophagy is a lysosome-dependent degradation pathway that degrades cellular components to maintain cellular biosynthesis and viability during metabolic stresses such as nutrient deprivation. During chemotherapy, autophagy facilitates cancer cell survival to cope with metabolic stresses caused by anticancer drugs. [bib_ref] Cancer drug resistance: an evolving paradigm, Holohan [/bib_ref] For instance, in breast cancer cell models, resistance to endocrine therapy such as tamoxifen and fulvestrant is the result of activation and interactions between different cellular mechanisms including UPR activation, autophagy and apoptosis in breast cancers. [bib_ref] Knockdown of estrogen receptor-alpha induces autophagy and inhibits antiestrogen-mediated unfolded protein response..., Cook [/bib_ref] [bib_ref] NF-kappaB signaling is required for XBP1 (unspliced and spliced)-mediated effects on antiestrogen..., Hu [/bib_ref] [bib_ref] Glucose-regulated protein 78 as a novel effector of BRCA1 for inhibiting stress-induced..., Yeung [/bib_ref] [bib_ref] Lifetime genistein intake increases the response of mammary tumors to tamoxifen in..., Zhang [/bib_ref] [bib_ref] Modelling the effect of GRP78 on anti-oestrogen sensitivity and resistance in breast..., Parmar [/bib_ref] Indeed, antiestrogen-resistant breast cancer cells display higher levels of basal autophagy than sensitive cells. [bib_ref] NF-kappaB signaling is required for XBP1 (unspliced and spliced)-mediated effects on antiestrogen..., Hu [/bib_ref] In addition, XBP1s-overexpressing MCF-7 cells displayed much higher basal levels of autophagy as demonstrated with increased basal LC3II levels and decreased p62 levels. [bib_ref] NF-kappaB signaling is required for XBP1 (unspliced and spliced)-mediated effects on antiestrogen..., Hu [/bib_ref] Autophagy induced by XBP1s overexpression protects the cells against apoptosis. Furthermore, XBP1s-overexpressing cells become sensitive to tamoxifen when autophagy is blocked. [bib_ref] NF-kappaB signaling is required for XBP1 (unspliced and spliced)-mediated effects on antiestrogen..., Hu [/bib_ref] Hypoxia is known to confer cancer cells with resistance to chemotherapy and to modulate UPR during ER stress. [bib_ref] Evidence of galectin-1 involvement in glioma chemoresistance, Mercier [/bib_ref] [bib_ref] Enhancement of cisplatinbased TACE by a hemoglobin-based oxygen carrier in an orthotopic..., Liu [/bib_ref] [bib_ref] Knockdown of glucose-regulated protein 78 abrogates chemoresistance of hypopharyngeal carcinoma cells to..., Pi [/bib_ref] In breast cancers, taxol rapidly induces UPR activation including ATF6α, IRE1α and PERK pathways. However, hypoxia modulates taxol-induced UPR activation acting specifically on the UPR branches PERK, ATF6α and IRE1α. [bib_ref] Taxol-induced unfolded protein response activation in breast cancer cells exposed to hypoxia:..., Notte [/bib_ref] Indeed, ATF4 activation leads to taxol-induced autophagy completion and cell death resistance. Finally, ATF4 expression in association with hypoxia-induced genes, such as adrenomedullin, is a biomarker of a poor prognosis for human breast cancer patients. [bib_ref] Taxol-induced unfolded protein response activation in breast cancer cells exposed to hypoxia:..., Notte [/bib_ref] Intratumoral hypoxia is one predominant feature of GBM and is associated with resistance to temozolomide (TMZ), the standard chemotherapy for GBM. [bib_ref] Hyperoxia resensitizes chemoresistant human glioblastoma cells to temozolomide, Sun [/bib_ref] TMZ sensitivity of both sensitive and resistant GBM cells is significantly enhanced under hyperoxia in vitro through the induction of caspase-dependent pathways. [bib_ref] Hyperoxia resensitizes chemoresistant human glioblastoma cells to temozolomide, Sun [/bib_ref] In addition, elevated PDIA1 expression also occurs in hypoxic brain tumor cells. PDIA1, which belongs to the protein disulfide isomerase superfamily, is the key foldase that has been found to be significantly dysregulated during the development of TMZ resistance in GBM cells. [bib_ref] Hyperoxia resensitizes chemoresistant glioblastoma cells to temozolomide through unfolded protein response, Lee [/bib_ref] Hyperoxia resensitizes TMZ-resistant GBM cells to TMZ by abrogating the hypoxia-induced UPR-related protective mechanisms. Hyperoxia, alone or synergistically with TMZ, activates the UPR in sensitive and resistant cell lines. [bib_ref] Hyperoxia resensitizes chemoresistant glioblastoma cells to temozolomide through unfolded protein response, Lee [/bib_ref] Hyperoxia impairs protein folding that in turn induces UPR-mediated apoptosis. Its reduces survival benefit of cancer cells with PDIA1 overexpression through the UPR by decreasing GRP78 and PDIA1 expression and consequently triggering cell death via downregulation of the ER stress chaperone protectors. [bib_ref] Hyperoxia resensitizes chemoresistant glioblastoma cells to temozolomide through unfolded protein response, Lee [/bib_ref] Interestingly, TMZ increases galectin-1 expression in glioma cells. [bib_ref] Evidence of galectin-1 involvement in glioma chemoresistance, Mercier [/bib_ref] Galectin-1 increases the expression of genes implicated in chemotherapy resistance such as GRP78, ORP150, HERP (homocysteine-induced ER protein), transcription associated factor 1 (TRA1), BNIP3L (Bcl-2/adenovirus E1B 19 kDa protein-interacting protein 3-like), GADD45B and CYR61 (cysteine-rich angiogenic inducer 61), some of which are located in the ER and modified by hypoxia. [bib_ref] Evidence of galectin-1 involvement in glioma chemoresistance, Mercier [/bib_ref] Additionally, under severe hypoxia and chemotherapy, UPR activation occurs in hypopharyngeal carcinomas leading to increased expression of GRP78 associated with hypoxia-induced chemotherapy resistance. [bib_ref] Knockdown of glucose-regulated protein 78 abrogates chemoresistance of hypopharyngeal carcinoma cells to..., Pi [/bib_ref] Diminution of GRP78 inhibits cell proliferation and promotes apoptosis under cisplatin treatment with severely (2) Calnexin(+).
ER stress and cancer chemotherapy resistance T Avril et al ER stress and cancer chemotherapy resistance T Avril et al hypoxic conditions, indicating that GRP78 confers cancer cell resistance to cisplatin in response to severe hypoxia. This phenomenon involves increased CHOP and BAX expression levels and decreased Bcl-2 expression levels with simultaneous increased apoptosis under severely hypoxic conditions. [bib_ref] Knockdown of glucose-regulated protein 78 abrogates chemoresistance of hypopharyngeal carcinoma cells to..., Pi [/bib_ref] A number of studies indicated that improving oxygenation inside the tumor could serve as a potential strategy to target hypoxiainduced chemotherapy resistance. [bib_ref] Enhancement of cisplatinbased TACE by a hemoglobin-based oxygen carrier in an orthotopic..., Liu [/bib_ref] In liver cancers, hypoxia increases cisplatin resistance. The use of a hemoglobin-based oxygen carrier (OC89) enhances the efficacy of cisplatin-based transarterial chemoembolization in rat liver cancer model. OC89 delivery knocks down the balance of UPR pathway by decreasing GRP78 expression and increasing that of CHOP. This leads to increase tumor apoptosis and to inhibit tumor cell proliferation. [bib_ref] Enhancement of cisplatinbased TACE by a hemoglobin-based oxygen carrier in an orthotopic..., Liu [/bib_ref] Interestingly, UPR activation is also observed in non-tumoral cells that compose the tumor microenvironment. [bib_ref] Glucose-regulated protein 78 (Grp78) confers chemoresistance to tumor endothelial cells under acidic..., Visioli [/bib_ref] Indeed, UPR markers GRP78, ATF4 and CHOP are significantly upregulated in endothelial cells from oral squamous cell carcinomas. Furthermore, under severe acidic conditions and hypoxia, which recapitulate the tumor microenvironment, microvascular endothelial cells increase GRP78 expression, acquire antiapoptosis capacities and resist to sunitinib, an antiangiogenic drug. [bib_ref] Glucose-regulated protein 78 (Grp78) confers chemoresistance to tumor endothelial cells under acidic..., Visioli [/bib_ref] GRP78 knockdown resensitizes endothelial cells to drug treatment. [bib_ref] Glucose-regulated protein 78 (Grp78) confers chemoresistance to tumor endothelial cells under acidic..., Visioli [/bib_ref]
## Conclusion and perspectives: targeting the upr to bypass resistance
The UPR is a physiological mechanism developed by cells to cope with misfolded protein accumulation induced by challenging conditions. As observed for other cellular mechanisms, tumor cells hijack the UPR to allow drug resistance, through the activation of the UPR sensors ATF6, IRE1α and PERK, and their master regulator GRP78. As presented above, the involvement of the UPR in chemotherapy resistance is complex and not fully covered yet. This is in part due to the links between the UPR and other tumor adaptive mechanisms as such antiapoptotic mechanisms, autophagy or dormancy. Therefore, a global understanding of the molecular mechanisms controlling UPR-mediated drug resistance is highly needed.
Small-molecule UPR inhibitors that directly target the UPR sensors ATF6α, IRE1α, PERK and their regulators or effectors such as PDIA1 and eIF2α, respectively, have been recently identified. [bib_ref] Control of the unfolded protein response in health and disease, Doultsinos [/bib_ref] Their potential use in combination with chemotherapeutics might greatly improve anticancer drug efficacy. For instance, ISRIB, a drug that reverses the effects of eIF2α phosphorylation, increased gemcitabine-induced death of pancreatic cancer cells. [bib_ref] Integrated stress response is critical for gemcitabine resistance in pancreatic ductal adenocarcinoma, Palam [/bib_ref] Recent evidences have also been provided from leukemic tumors. The PDI inhibitor 16F16 reverses leukemia cell resistance to imatinib linked to the ATF6α pathway most likely by blocking PDIA5. [bib_ref] Endoplasmic reticulum stress-activated transcription factor ATF6alpha requires the disulfide isomerase PDIA5 to..., Higa [/bib_ref] Finally, MKC-3946, an IRE1α RNase inhibitor, synergizes bortezomib or arsenic trioxide induced toxicity of acute myeloid leukemia cells. [bib_ref] Inhibition of IRE1alpha-driven pro-survival pathways is a promising therapeutic application in acute..., Sun [/bib_ref] Alternatively, modulating UPR with pharmacological drugs has shown promising results in vitro. For instance, epigallocatechin gallate, which specifically targets GRP78, resensitizes glioma cells to TMZ. [bib_ref] The unfolded protein response regulator GRP78/BiP as a novel target for increasing..., Pyrko [/bib_ref] [bib_ref] Molecular docking and molecular dynamics studies reveal structural basis of inhibition and..., Bhattacharjee [/bib_ref] Although targeting GRP78 might be an attractive therapeutic approach, the challenge will be to minimize systemic toxicity in normal organs in which GRP78 is essential for the survival and functions of various cellular subtypes. [bib_ref] Glucose-regulated proteins in cancer: molecular mechanisms and therapeutic potential, Lee [/bib_ref] This implies that GRP78-targeting drugs should selectively target tumor cells that require a high level of GRP78 and spare normal organs. Bortezomib, a proteasome inhibitor that amplifies the protein misfolding burden, confers a chemosensitizing effect to cisplatin, doxorubicin or camptothecin in various tumor types including breast, colon pancreatic cancers. [bib_ref] The development of proteasome inhibitors as anticancer drugs, Adams [/bib_ref] Sorafenib, a potent multikinase inhibitor, induces both apoptosis and autophagy in human hepatocellular carcinoma cells through an ER stress-dependent mechanism and the alteration of normal secretory functions. Furthermore, the combination of sorafenib with the autophagy inhibitor chloroquine leads to enhance liver cancer suppression. [bib_ref] Targeting autophagy enhances sorafenib lethality for hepatocellular carcinoma via ER stress-related apoptosis, Shi [/bib_ref] Verteporfin, a YAP1 (Yes-associated protein 1) inhibitor, has been recently involved in the oligomerized protein accumulation in CRC cells, leading in part to tumor apoptosis. Furthermore, hypoxic or nutrient-deprived conditions amplify verteporfin-mediated CRC cell death. [bib_ref] Tumor-selective proteotoxicity of verteporfin inhibits colon cancer progression independently of YAP1, Zhang [/bib_ref] Resistance of melanoma cells to vemurafenib or PLX4032, two BRAFV600E kinase inhibitors, is bypassed in the presence of thapsigargin, an inhibitor of the SERCA pumps or in the presence of HA15, which targets GRP78, respectively, by inducing tumor apoptosis. [bib_ref] Compounds triggering ER stress exert anti-melanoma effects and overcome BRAF inhibitor resistance, Cerezo [/bib_ref] [bib_ref] Vemurafenib potently induces endoplasmic reticulum stress-mediated apoptosis in BRAFV600E melanoma cells, Beck [/bib_ref] In conclusion, future challenges will certainly lead to the development of combined therapeutic approaches with new drugs that specifically target the UPR sensors and downstream partners and will to bypass anticancer drug resistance.
[fig] 1: CHOP(+). (2) calnexin(+), calreticulin(+), CHOP(+), GRP94(+), PDI( − ), phosphorylated IRE1α, PERK and eIF2α(+). (3) CHOP(+), phosphorylated PERK. (4) Decreased CHOP, cleaved ATF6, phosphorylated PERK and eIF2α. (5) DNAJC3, ERO1LB, GRP94. (6) CHOP(+), DNAJC3( − ), ERO1Lb( − ), GADD34(+). (7) CHOP(+), GRP94(+), cleaved ATF6, phosphorylated eIF2α. (8) CHOP (+), phosphorylated eIF2α and JNK. (9) Phosphorylated IRE1α. (10) CHOP(+), LCN2(+). (11) Phosphorylated eIF2α. (12) Calnexin(+). (13) HSP47(+), PDI(+), phosphorylated PERK and eIF2. [/fig]
[table] Table 2: Cellular models demonstrating the importance of UPR in solid cancers [/table]
[table] Table 3: Standard chemotherapy treatments and their targets in solid DNA-damage repair. Most chemotherapeutic drugs drive the induction of DNA damage in tumor cells either directly for platinum-based drugs or indirectly for 5-FU and topoisomerase inhibitors. 104,105 DNA topoisomerase-I mutations have been reported to affect camptothecin sensitivity. [/table]
[table] Table 4: Clinical evidences of UPR involvement in cancer chemotherapy resistance [/table]
[table] Table 5: Cellular models demonstrating the importance of UPR in cancer chemotherapy resistance Abbreviations: ATF, activating transcription factor; BIK, Bcl-2-interacting killer; DTT, dithiothreitol; eIF2α, eukaryotic initiation factor 2α; ERO1L, ER oxidoreduction 1-like; 5-FU, 5-fluorouracil; FRP, glucose-regulated protein; HSP, heat-shock protein; IHC, immunohistochemistry; IRE1α, inositol requiring enzyme 1α; JNK, c-Jun N-terminal protein kinase; LCN2, lipocalin 2; PDI, protein disulfide isomerase; PERK, PKR-like endoplasmic reticulum kinase; Q-PCR, quantitative PCR; RT-PCR, reverse transcriptase-PCR; Tg, thapsigargin; UPR, unfolded protein response; WB, western blot; XBP, X-box binding protein. [/table]
|
A comparison of lumbar transverse pedicle angles between ethnic groups: a retrospective review
Background: Spinal surgery requires an intimate understanding of pedicle morphology to provide safe and effective outcomes. Although current research has attempted to identify morphological vertebral pedicle trends, no study has utilized computed tomography (CT) scans to compare the lumbar transverse pedicle angle (TPA) with patient demographics factors in a diverse population throughout multiple hospital centers. Methods: Analysis of randomly selected CT scans from L1-L5 of 97 individuals who underwent imaging over a two-week period for non-back pain related complaints was conducted. Measuring 970 TPAs in total allowed for comparison of each patients' pedicle angle with important patient specific demographics including ethnicity, age, gender, height and weight. Statistical analysis utilized multiple comparisons of demographics at each level with post-hoc Bonferroni correction analysis to compare demographics at each level. Results: With relation to gender, age, height or weight, no statistically significant differences were identified for TPAs at any vertebral level. However, when stratified by ethnicity, the differences in transverse pedicle angles averages (TPA -Avg) at L2 and L3 were found to be statistically significant (p < 0.05). Conclusion: We have identified a previously unknown and significant relationship between ethnicity and TPA at lumbar vertebral levels. These findings provide critical information that may be added to the operating surgeons' knowledge of pedicle morphology. We hope this novel information can assist in preoperative planning of pedicle screw placement and potentially help improve surgical outcomes.
# Background
Posterior lumbar fusion has been utilized to alleviate pain and instability in patients with spinal injuries and deformities and for nearly 65 years [bib_ref] The treatment of ruptured lumbar intervertebral discs by vertebral body fusion. I...., Cloward [/bib_ref]. When lumbar fusion is indicated, there are multiple different techniques utilized to achieve fusion [bib_ref] Anterior lumbar interbody fusion versus transforaminal lumbar interbody fusion--systematic review and metaanalysis, Phan [/bib_ref] [bib_ref] Comparison of complication rates of minimally invasive transforaminal lumbar interbody fusion and..., Joseph [/bib_ref]. With advancing trends and technologies in surgery, there is an increased impetus to advance patient outcomes by improving operative techniques and lumbar fusion is no exception [bib_ref] Comparison of one-level posterior lumbar interbody fusion performed with a minimally invasive..., Park [/bib_ref] [bib_ref] Clinical and radiological outcomes of minimally invasive versus open transforaminal lumbar interbody..., Peng [/bib_ref]. With the increasing use of posterior lumbar fusion, further elucidation of patient specific variables in relation to vertebral morphometric variation may assist orthopedic surgeons in planning and performance of spinal surgery.
An understanding of the osseous vertebral anatomy and variation between patients is of the utmost importance in spinal fusion, and even more so in with the increasing popularity of minimally invasive spinal fusion as the surgeon may have less visual reference available. Slight deviation in screw trajectory could have devastating outcomes for patients. A better understanding of anatomic variability offers to improve patient safety by increasing the surgeons' precision while performing spinal fusion procedures. Specifically, the transverse pedicle angle (TPA) is utilized by the operating physician to gauge optimal course of pedicle screw placement. The TPA is the angle created between a line drawn from the midline of the spinous process to the anterior vertebral body and the mid-axis of the pedicle [fig_ref] Figure 1: TPA Measurement [/fig_ref]. Identification of this angle is key for guidance of pedicle screw trajectory. Orthopedic surgical textbooks suggest that knowledge of this angle is important for ideal insertion of pedicle screws from the posterior aspect. There is currently a lack of knowledge regarding patient factors that may influence the morphology of the TPA, and thereby the ideal screw trajectory for successful spinal fusion. Utilization of computed tomography (CT) to identify TPA angle in each patient at each vertebral level of fusion is ideal, however, most surgeons do not use this modality to assist planning surgery and may not have this option in emergent situations.
Multiple studies have attempted to utilize diverse modalities to help outline common anatomical parameters of vertebrae and thus help surgeons when planning spinal surgery. Morales-Avalos, et al. [bib_ref] Age and gender related variations in morphometric characteristics of thoracic spine pedicle:..., Morales-Avalos [/bib_ref] have displayed significant correlation in thoracic pedicle variability with age and gender utilizing caliper measurements on dried osseous specimens from a Mexican population. Yu, et al., have shown significant correlation between lumbar pedicle morphometry, gender, height and weight utilizing digital calipers on a population of American human cadavers. Gulec, et al. [bib_ref] Morphometric analysis of the lumbar vertebrae in the Turkish population using three-dimensional..., Gulec [/bib_ref] , have utilized three-dimensional CT to compare gender, age and height with pedicle morphometry in a Turkish only population. Another study utilized electronic calipers to compare the pedicles in a small Greek only cadaveric study. A study utilizing CT reported lumbar pedicle morphometry in a population of Pakistani patients only [bib_ref] Lumbar morphometry: a study of lumbar vertebrae from a pakistani population using..., Alam [/bib_ref]. Mughir, et al. [bib_ref] Morphological comparison between adolescent and adult lumbar pedicles using computerised tomography scanning, Mughir [/bib_ref] , compared pedicle morphology between adults and children in a Malaysian population. In a study of patients with low back problems, lumbar spine morphometry was compared to patient gender on CT [bib_ref] Morphometry of the lumbar spine: anatomical perspectives related to transpedicular fixation, Olsewski [/bib_ref]. Chadha, et al., observed multiple pedicle characteristics in an Indian only population. The authors of that study then reviewed previous literature on TPAs and noted that Indians have different TPAs at some vertebral levels when compared to Western populations. However, this comparison was made between multiple studies utilizing multiple and differing measurement techniques opening the possibility for unreliable correlations. A more recent study undertaken in South Africa has attempted to identify ethnic variation in osseous morphology utilizing 174 dried lumbar vertebral specimens with caliper and goniometer measurements. A meta-analysis study regarding CT analysis of the osseous morphology in the cervical spine undertaken by Marumo, et al. [bib_ref] Ethnic differences in pedicle and bony spinal canal dimensions calculated from computed..., Chazono [/bib_ref] , claims that although there is variation due to ethnicity, there may be a lack of significance.
Still, none of these aforementioned studies provide a large-scale, generalizable and reliable database of CT imaging and measurement of pedicle attributes. A single study utilizing homogenous measurement techniques among a diverse living population without reported back pain is necessary to delineate variations in pedicle morphology related specifically to ethnicity, age, gender, height, and weight. When instrumenting for lumbar fusion, a thorough understanding of the vertebral TPA is integral in safely and precisely placing pedicle screws in the lumbar spine. To our knowledge there has never been a single study of measuring TPA in living adults using CT scan that includes multiple different races and ethnicities.
The specific aim of this research was to create a single study comparing the TPAs of patients in a diverse area of the country. This population allows for analysis of potential trends between multiple ethnicities and other demographic characteristics under the same measurement methodology in order to identify if there are any significant differences in TPA among races.
# Methods
A retrospective review of CT scans of the abdomen and pelvis was performed over a two-week period (between July 1, 2016 and July 14, 2016). The CT scans were performed at seven hospitals within one single health system. We randomly selected 97 CT images of L1-L5 from all scans completed during this time period. Using the CT abdomen and pelvis studies rather than lumbar spine specific CT scans allowed for screening of a population of 97 patients who presented with chief complaints unrelated to back pain. CT scans were reviewed on Carestream PACS and the present "Bone Window" was utilized for evaluation and analysis of the CT scans. From each of the randomly selected CT scans the TPA from L1-L5 were measured. In total, 970 lumbar TPA's were evaluated.
Each lumbar TPA was measured by creating a midline measurement from spinous process to the anterior vertebral body and measuring the angle from that midline to the mid-axis of the pedicle bilaterally [fig_ref] Figure 1: TPA Measurement [/fig_ref]. TPA data was obtained by a single observer and verified by 2 more senior physicians, enhancing interobserver variations. We then compared TPA with multiple patient factors including ethnicity, age, gender, height and weight. Height and weight were directly measured and reported in patient charts. Inclusion criteria for age was 18 through 99 years. Analysis was carried out by a Senior Research Statistics Analyst to determine the significance of the study findings.
Those excluded from the study were patients with evidence of prior lumbar spine surgery on imaging, scans that did not allow analysis of the five lumbar segments and patients with evidence of scoliosis.
# Results
The ethnicities of the patients from which we obtained these scans were: Asian (n = 31), Hispanic (n = 27), Black (n = 27), and White (n = 12). TPA mean, standard deviation (SD) and standard error of the mean (SEM) with respect to each lumbar segment and ethnicity are reported . In all ethnicities an increase in the TPA was appreciated with progression down each lumbar segment from L1 to L5. For each individual lumbar segment, the TPA mean with respect to ethnicity was reported with each corresponding standard error and is displayed graphically [fig_ref] Figure 2: TPA Separated by Race at Lumbar Segments L1-L5 [/fig_ref].
When statistically analyzing TPA with other variables, there were no statistically significant differences found for TPA with relation to gender, age, height or weight. However, when stratified by ethnicity, the TPA averages at individual vertebral levels were found to be statistically significant at the L2 and L3 levels (p < 0.05). No statistically significant findings were found at levels L1, L4, or L5. Multiple comparisons between ethnicities at each level were made followed by post-hoc comparisons utilizing Bonferroni correction indicated that there is statistical Mean TPA Data. Data obtained via CT analysis of mean TPA, standard deviation and standard error of the mean classified by race and individual lumbar level significance at L2 and L3 at the level p < 0.05 [fig_ref] Table 2: Multiple Comparisons [/fig_ref]. When multiple comparisons were made at L2, Asians were found to have significantly larger TPA-Avg angle by 3.11 o (p < 0.0001) when compared to Blacks. Additionally, at L2, the TPA-Avg angle of White individuals is 1.84 o larger than black individuals, which was found to be significant (p = 0.039). When multiple comparisons were made at L3, Asians were found to have a TPA-Avg of 1.94 o (p = 0.002) and 2.91 o (p < 0.0001) larger than Hispanics and Blacks respectively.
# Discussion
Spinal surgery has been depicted to offer multiple benefits to patients. However, there still lacks an in-depth anatomical understanding of a key operative parameter, the TPA. The TPA is of great importance regarding optimal pedicle screw placement. We propose that furthering the understanding of morphological variation in the TPA will assist surgeons in preoperative planning of posterior lumbar fusion. In the current study, we have compared multiple patient factors with TPA including age, height, weight, gender and ethnicity. This study stands do delineate potential significant correlations between TPA and various patient demographics.
We did not find any statistically significant correlations between TPA in L1-L5 when compared to age, height, weight or gender. However, we did find statistically significant relationships between ethnicity and TPA. Here, we present a previously unknown relationship between the TPA of L2 and L3 and ethnicity. Specifically, at vertebral level L2, we have identified that the average TPA of Asian individuals is 3.11 o larger than that of Black individuals (p < 0.0001) and the TPA of White individuals is 1.84 o larger than that of Black individuals (p = 0.039). At L3, we identified that the average TPA of Asian individuals is 1.94 o (p = 0.002) and 2.91 o (p < 0.0001) larger than that of Hispanic and Black individuals, respectively.
The results from this study are of potential value to the orthopedic surgeon performing posterior lumbar fusion techniques in both the preoperative planning stages and intraoperatively. When preparing for instrumentation of the lumbar spine, it is ideal to obtain a CT scan of the lumbar spine to evaluate for possible TPA variation. However, the utilization of CT imaging is not always undertaken in preoperative planning and may expose the patient to undue radiation. Knowing the variation in a given patient ethnicity prior to surgery may lead to faster operative times and more precise pedicle instrumentation leading to lower hardware failure rates and improved patient outcomes. We had the advantage of utilizing CT images from a culturally diverse geographic area served by multiple hospitals. The CT scans evaluated were from living patient without complaints specific to lumbar spine pathology. To further validity of our study, the TPA data was obtained by a single observer and verified by 2 more senior physicians, enhancing interobserver variations.
Limitations of this study include lacking a standardized position of the patient on the CT scanner. Simpson et al. described inaccuracies even by CT when only evaluating on one plane, further emphasizing the need for careful preoperative evaluation of pedicle diameter. Additionally, as race was self-reported with preset options, there was no ability to report multiple races. Finally, for White ethnicity, only 12 patients were included which may be unrepresentative of this population. Future work may be designed to address these limitations.
# Conclusion
We have identified a previously unknown and significant result with respect to TPA and ethnicity. At L2, individuals of Asian ethnicity were found to have larger TPAs than Black individuals and White individuals have larger TPA than Blacks. At L3, Asians were found to have a larger TPA than both Blacks and Hispanics. This result may help guide the orthopedic surgeon when preparing for lumbar instrumentation or any lumbar surgical techniques that require in depth knowledge of lumbar pedicle morphology. Further studies may be done to assess operative duration and clinical outcomes of surgeons who use this information in preoperative planning as well as intraoperative decision making compared to surgeons who are unaware of this ethnicity-TPA relationship.
[fig] Figure 1: TPA Measurement. Each lumbar TPA was measured by creating a midline measurement from spinous process to the anterior vertebral body then measuring the angle from that first central line to the mid-axis of the pedicle bilaterally [/fig]
[fig] Figure 2: TPA Separated by Race at Lumbar Segments L1-L5. This graph depicts the differences seen between race at each lumbar level with respect to mean TPA and race. Error bars indicate standard error of the mean (SEM) for each category [/fig]
[table] Table 2: Multiple Comparisons. Post-Hoc Bonferroni analysis of multiple comparison for each ethnicity at each individual vertebral level was undergone to identify significance [/table]
|
Interleukin-1 receptor associated kinase 1/4 and bromodomain and extra-terminal inhibitions converge on NF-κB blockade and display synergistic antitumoral activity in activated B-cell subset of diffuse large B-cell lymphoma with MYD88 L265P mutation
## Interleukin-1 receptor associated kinase 1/4 and bromodomain and extra-terminal inhibitions converge on nf-κb blockade and display synergistic antitumoral activity in activated b-cell subset of diffuse large b-cell lymphoma with myd88 l265p mutation
The outcome of patients with diffuse large B-cell lymphoma (DLBCL) is very heterogeneous and is most likely dictated by their cell of origin (COO), defining two main molecular subtypes, i.e., germinal center B-cell (GCB) and activated B-cell (ABC). [bib_ref] Molecular pathogenesis of germinal center-derived B cell lymphomas, Pasqualucci [/bib_ref] Upon treatment with multi-agent chemotherapy (cyclophosphamide, doxorubicin, vincristine and prednisone) combined with the monoclonal anti-CD20 antibody rituximab (R-CHOP), almost a third of the patients, corresponding mainly to the ABC subtype of the disease, does not achieve complete remission (CR) or relapses shortly after CR. [bib_ref] Treatment of diffuse large B-cell lymphoma, Miyazaki [/bib_ref] However, the COO does not fully account for the different outcomes. Massive sequencing analyses recently uncovered molecular subtypes of DLBCL with distinct outcomes. In this regard, Chapuy et al. have described five different molecular subtypes with distinct pathogenic mechanisms and prognosis, independently of the COO. Interestingly, the C5 cluster (mostly ABC subtypes), enriched in MYD88 L265P and CD79B mutations, maintained a shorter survival compared to the other ABC cluster. [bib_ref] Molecular subtypes of diffuse large B cell lymphoma are associated with distinct..., Chapuy [/bib_ref] ABC-DLBCL tumors rely almost exclusively on constitutive nuclear transcription factor κB (NF-κB) signaling for their survival, a phenomenon that has been linked to a variety of genetic alterations that aberrantly activate the B-cell receptor (BCR) and the Toll-like receptor (TLR) signaling pathways. [bib_ref] Molecular pathogenesis of germinal center-derived B cell lymphomas, Pasqualucci [/bib_ref] Within the TLR axis, mutations in the gene codifying for the adaptor protein myeloid differentiation primary response gene 88 (MYD88) enhance interleukin-1 receptor-associated kinase 1 and 4 (IRAK1 and IRAK1) activity, providing sustained activation of NF-κB through most of the TLR. The p.L265P mutation, characterized by a change from leucine (CTC) to proline (CCG) in the MYD88 Toll/interleukin (IL)-1 receptor domain, recruits MYD88 to the cytoplasmic tail of TLR to form an active complex. Beside NF-κB, this complex promotes Janus kinase-signal transducer and activator of transcription 3 (JAK-STAT3) signaling through a pathway involving interleukin (IL)-6 and IL-10 secretion. [bib_ref] Oncogenically active MYD88 mutations in human lymphoma, Ngo [/bib_ref] Preclinical data have indicated that MYD88-mutant ABC-DLBCL cells were sensitive to pharmacological blockade of IRAK4 kinase activity, being IRAK4-compromised cells especially responsive to the Bruton's tyrosine kinase (BTK) inhibitor ibrutinib or the BCL-2 antagonist venetoclax, as almost all ABC-DLBCL display BCL2 amplification/overexpression. [bib_ref] Selective interleukin-1 receptorassociated kinase 4 inhibitors for the treatment of autoimmune disorders..., Kelly [/bib_ref] [bib_ref] B-cell-specific conditional expression of Myd88p.L252P leads to the development of diffuse large..., Knittel [/bib_ref] Considering that both IRAK1 and IRAK4 are required for ABC-DLBCL cell survival, [bib_ref] Oncogenically active MYD88 mutations in human lymphoma, Ngo [/bib_ref] we investigated the effect of a 24-72 hour treatment with a selective and orally bioavailable IRAK1/4 inhibitor (IRAKi, Merck), 7 in three well-characterized MYD88-mutated cell lines, OCI-LY3, OCI-LY10, HBL-1, using proliferation as a read out. Three germinal center B-cell (GCB)-DLBCL cell lines (SUDHL-4, SUDHL-8 and OCI-LY8) with wild-type MYD88 (MYD88 wt ) were analyzed in the same settings, as a control. We observed a partial and transitory response to IRAKi in ABC-DLBCL cells only, when using the compound at the physiological dose of 50 mM [fig_ref] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B... [/fig_ref]. Treatment-related cytotoxicity decreased from 25.5% at 24 hours to 19% at 72 hours, respectively, despite an efficient blockade of IRAK1 and IRAK4 phosphorylation at Thr209 and Thr345 residues, in the three MYD88-mutated cell lines [fig_ref] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B... [/fig_ref].
Interestingly, the destabilization of the anti-apoptotic protein and key mediator of IRAKi activity, MCL-1, [bib_ref] Inhibition of IRAK1/4 sensitizes T cell acute lymphoblastic leukemia to chemotherapies, Li [/bib_ref] was not sufficient to confer a significant cytotoxicity to the compound [fig_ref] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B... [/fig_ref]. A gene expression profiling (GEP) analysis in the three MYD88-mutated cell lines exposed for 6 hours to the inhibitor, further showed that IRAK1/4 blockade significantly altered the expression of the top NF-κB gene signatures associated to B-cell lymphoma, [bib_ref] Small molecule inhibitors of IκB kinase are selectively toxic for subgroups of..., Lam [/bib_ref] namely NFKB_ALL_OCI_LY10 and NFKB_BOTH OCILY3ANDLY10, with normalized enrichment score (NES) values reaching 1.8, while in contrast a third gene set, NFKB_OCILY10_ONLY, was slightly upregulated (NES: -1.20), according to GSEA analysis [fig_ref] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B... [/fig_ref] ; Online . In agreement, the transcription of several NF-κB-regulated genes known to promote ABC-DLBCL pathogenesis, including IL6, IL10, IRF4 and CCL3, were either unaffected or even increased after treatment with IRAKi [fig_ref] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B... [/fig_ref] , Online . Consistently, in an OCI-LY3 mouse xenograft model the compound failed to elicit a significant tumor growth inhibition (Online Supplementary [fig_ref] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B... [/fig_ref].
We then considered the possibility to enhance IRAKi activity in MYD88 L265P ABC-DLBCL by combining the compound with the BET bromodomain inhibitor CPI203 (kindly provided by Constellation Pharmaceuticals), as this BRD4 antagonist has been shown to effectively suppress a NF-κB gene signature that includes IL6, IL10 and IRF4, in ABC-DLBCL. [bib_ref] Blockade of oncogenic IκB kinase activity in diffuse large B-cell lymphoma by..., Ceribelli [/bib_ref] After exposing the same ABC-DLBCL cell lines as above to a 50 mM dose of IRAKi, followed by a 24-hour treatment with 0.5 mM CPI203, a new GEP analysis was performed. As shown in [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref] , IRAKi-CPI203 combination induced a significant downregulation of NF-κB-related genes when compared to IRAKi single agent, with NES comprised between 1.46 and 1.99. Of note, the combination therapy allowed to a significant disruption of NFKB_OCILY10_ONLY gene signature with a NES of 1.89. Among the genes included in the NFKB_ALL_OCILY3_LY10 gene set, a selected list of nineteen factors underwent a ≥2-fold increase in their rank metric score between this analysis and the previous one (Online , suggesting that their improved modulation may be associated with the combinational effect of IRAKi and CPI203. From this list, we identified only four genes (LTA, MARCKS, CD44 and HEATR1) that were not included in the core component of the NF-κB target genes affected by either IRAKi or CPI203 as single agents, but which underwent a significant downregulation upon treatment with the drug combination. Among these genes, we were unable to detect significant levels of LTA and HEATR1 transcripts in the three ABC-DLBCL cell lines (data not shown). In contrast, upon exposure of the three MYD88-mutated cell lines to the IRAKi we observed a 1.2-to 2-fold transcriptional increase of MARCKS and CD44, together with IL6 and IL10 used here as hallmarks of NF-kB activation. These genes were all reduced down to 0.5-fold in cells treated with the drug combination [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref]. Accordingly, IRAKi-CPI203 treatment led to the accumulation of the intracellular inhibitor of NF-κB, IκB, and to the consequent reduction in CD44 and MARCKS protein levels, while IRAKi and CPI203 single agents slightly affected the expression of these factors [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref]. As expected, in two out of the three cell lines, CPI203-based treatments led to the decrease in MYC protein and mRNA, used here as hallmarks of BRD4 inhibition [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref]. Also confirming a previous report linking bromodomain inhibitor therapy with IRAK1 downregulation in B-cell lymphoma, 11 IRAK1-pThr209 levels underwent a slight downregulation after CPI203 treatment and this effect was remarkably potentiated upon addition of IRAKi to the cell cultures [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref]. In line with the increased blockade of NF-κB signaling, the addition of CPI203 synergistically improved IRAKi cytostatic effect in the three cell lines, as attested by an 86% blockade in cell proliferation, significantly higher than the 19% activity achieved by IRAKi alone (combination index [CI]: 0.52, [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref]. Importantly, the co-operation between the IRAKi and CPI203 involved a remarkable downregulation of MCL-1 [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref] , which was accompanied by a 36% increase in the relative apoptosis rate when compared with IRAKi and CPI203 used separately [fig_ref] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B... [/fig_ref].
In order to further validate the activity of the drug combination, primary lymph node biopsies from DLBCL patients with either MYD88 wt or MYD88 L265P were cocultured in the presence of a feeding stromal monolayer as previously reported, [bib_ref] The BET bromodomain inhibitor CPI203 overcomes resistance to ABT-199 (venetoclax) by downregulation..., Esteve-Arenys [/bib_ref] and treated with the different drugs as above. While IRAKi-CPI203 was almost inactive in MYD88 wt cells, the combination induced a 16% augmentation in relative apoptotic cell death in the MYD88 L265P primary co-culture [fig_ref] Figure 3: IRAKi and CPI203 combination is active in activated B-cell -diffuse B-cell lymphoma... [/fig_ref] , left panel), which was accompanied by a 12% decrease in the fraction of cells with high contents of IL6 mRNA, a percentage superior to what observed upon treatment with each drug alone [fig_ref] Figure 3: IRAKi and CPI203 combination is active in activated B-cell -diffuse B-cell lymphoma... [/fig_ref] , right panel).
Among the above mentioned genes, CD44 expression and IL-6 serum levels have been described as prognostic markers in DLBCL. [bib_ref] The addition of rituximab to CHOP therapy alters the prognostic significance of..., Wei [/bib_ref] [bib_ref] High serum levels of soluble interleukin-2 receptor (sIL2-R), interleukin-6 (IL-6) and tumor..., Dlouhy [/bib_ref] In order to investigate the role of these two factors in the response of ABC-DLBCL cell lines to IRAKi-based treatment, HBL-1 and OCI-LY3 cells were stimulated with 0.5 mM of the CD44 ligand, hyaluronic acid (HA), or exposed to a 5 mg/mL dose of the IL-6 blocking antibody tocilizumab, prior to a 72-hour treatment with the drugs. In the case of HA, cells were exposed to IRAKi (50 mM) +/-CPI203 (0.5 mM), while effect of tocilizumab pretreatment was evaluated in IRAKi-treated cells. Cell response was determined by fluorescence microscopy recounting of cells with high contents in Factin and by MTT assay, respectively. As shown on [fig_ref] Figure 3: IRAKi and CPI203 combination is active in activated B-cell -diffuse B-cell lymphoma... [/fig_ref] , both IRAKi and CPI203 were able to block actin polymerization by 50.4% and 54.5%, respectively, while the drug combination achieved a total 77.9% decrease in cells with high contents in F-actin following stimulation of Letters to the Editor 2750 haematologica | 2021; 106(10) [fig_ref] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B... [/fig_ref]. Thus, these results suggest a significant activity of the drug combination towards CD44 downstream signaling, while IL-6 expression may not be directly involved in the effect of these agents. Finally, in order to assess the efficacy of the drug combination in vivo, NSG mice were subcutaneously injected with OCI-LY3 cells, and tumor-bearing animals received daily doses of either IRAKi (5 mg/kg, intraperitoneal [i.p.], BID), CPI203 (2.5 mg/kg, i.p., BID), the combination of both agents, or the equivalent volume of vehicle, for 11 days. [fig_ref] Figure 3: IRAKi and CPI203 combination is active in activated B-cell -diffuse B-cell lymphoma... [/fig_ref] shows that CPI203 and IRAKi single agents induced a 31.5% and 46.3% tumor growth inhibition (TGI), respectively, while the combination of both drugs significantly improved this effect with a 65.6% TGI, when compared to vehicle-receiving animals (*P=0.011; **P=0.007). No significant toxicity was observed in any of the treatment arms. Histological analysis of the corresponding tumors revealed an improved reduction of mitotic index together with an accumulation of apoptotic cells by the combination therapy, as assessed by phosphohistone H3 and activated-caspase-3 staining [fig_ref] Figure 3: IRAKi and CPI203 combination is active in activated B-cell -diffuse B-cell lymphoma... [/fig_ref]. In agreement with the in vitro results, an enhanced reduction in the levels of CD44 and MCL-1, and an improved downregulation of nuclear p50 used as a read of NF-κB activity, was observed in the combination group when compared with the other arms [fig_ref] Figure 3: IRAKi and CPI203 combination is active in activated B-cell -diffuse B-cell lymphoma... [/fig_ref].
Collectively, our results suggest that IRAK1/4 inhibition is modestly effective in in vitro and in vivo models of ABC-DLBCL with MYD88 L265P, achieving only a partial inhibition of NF-κB signaling. We confirm that BET inhibition is an efficient strategy to counteract NF-κB over-activation in these models, offering synergistic anti-tumoral and proapoptotic activities with IRAK inhibition, mediated by the downregulation of the NF-κB-regulated factors, CD44 and MCL-1, and the consequent blockade of cell motility and triggering of tumor cell death.
[fig] Figure 1: Limited activity of IRAKi single agent in activated B-cell -diffuse large B cell lymhoma cell lines with MYD88 L265P in relation with incomplete inhibition of NF-κB gene signatures. (A) MTT assay showing that IRAKi (50 mM) elicited a partial and transitory response in activated B-cell -diffuse large B cell lymhoma (ABC-DLBCL) cell lines, while germinal center B-cell (GCB)-DLBCL cell lines were almost completely resistant to the compound. (B) IRAKi efficiently blocked the phosphorylation of IRAK1 at Thr29 and IRAK4 at Thr345, in the three ABC-DLBCL cell lines with MYD88 L265P. b-actin was used as a loading control. (C) Gene expression signatures of NF-κB in HBL-1, OCI-Ly3 and OCI-Ly10 cell lines exposed to IRAKi as above, highlighting that IRAKi treatment slightly affects this pathway (note that only 2 out 3 gene sets showed a false discovery rate [FDR] below 0.05). NES: normalized enrichment score. (D) Quantitative reverse transcriptase polymerare chain reaction (RQ-PCR) analysis of the predominant NF-κB-regulated genes IL6, IL10, CCL3, and IRF4, and the IRAK1/4 target gene MCL-1 in ABC-DLBCL cell treated by IRAKi as before (*P=0.01; **P<0.001). [/fig]
[fig] Figure 2: The BETi CPI203 synergizes with IRAKi in activated B-cell -diffuse large B cell lymphoma mediated by the inhibition of NF-κB downstream pathways. (A) Enrichment plots from gene set enrichment analysis (GSEA) analysis comparing IRAKi single agent vs. IRAKi-CPI203 combo in the 3 cell lines treated for 6 hours (Affymetrix HG-U219; GSEA), showing a significant improvement of NF-κB signature decrease by the addition of CPI203 to IRAKi. (B) Quantitative reverse transcriptase polymerare chain reaction (RQ-PRC) analysis of NF-κB downstream genes in the three cell lines exposed to IRAKi, CPI203 or CPI203-IRAKi combo as before. (*P=0.01; **P<0.001). (C) CPI203-IRAKi combination led to intracellular accumulation of IκB, and subsequent downregulation of IRAK1, MYC, CD44, MARCKS and MCL-1 proteins in activated B-cell -diffuse large B cell lymphoma (ABC-DLBCL) cells with MYD88 L265P . (D)OCI-Ly3, OCI-Ly10, HBL-1 cells were exposed for 24 hours to 0.1-0.5 mM CPI203 and/or 50-500 mM IRAKi. Cytotoxicity was evaluated by MTT assay and combination index (CI) was determined using the Calcusyn software. Shown are the cytotoxicity and the mean CI value calculated for cell treatment with 0.5 mM CPI203 and 50 mM IRAKi. (E) The drug combination led to a synergistic antitumoral effect in vitro in these 3 cell lines, inducing a median 36% increase in apoptosis rate when compared to single agent treatments (*P<0.04). [/fig]
[fig] Figure 3: IRAKi and CPI203 combination is active in activated B-cell -diffuse B-cell lymphoma primary cultures and impairs tumor growth in vivo. (A) Left panel: antitumoral activity of CPI203 (0.5 mM) and/or IRAKi (50 mM) was evaluated after a 24-hour culture of primary lymph node biopsies from activated B-cell -diffuse B-cell lymphoma patients with either MYD88 wt or MYD88 L265P by cytofluorimetric quantification of AnnexinV + cells. Cell of origin (COO) and MYD88 mutational status of the patients were determined by allele-specific polymerase chain reaction and gene expression analysis, as previously.15 Right panel: DLBCL cultures treated as above were labeled with an IL-6 Hu-Cyanine 5 SmartFlare RNA detection probe (Merck Millipore), and percentage of viable cells with high contents in IL6 mRNA was determined by flow cytometry, as previously.16 (B) HBL-1 y OCI-LY3 cell lines were preincubated for 24 hours with 0.5 mM CPI203 and/or 50 mM IRAKi, followed by a 24-hour stimulation with 0.5 mM HA, labeling with 50 mM Phalloidin-TRITC (Sigma-Aldrich) and recounting of red fluorescent cells on a Nikon H5505 microscope by means of a 20X/1.30 NA oil objective (Nikon) with the use of Isis Imaging System v5.3 software (MetaSystems GmbH) (***P<0.001). (C) NOD/SCID IL2Rγ-null (NSG) mice were inoculated subcutaneously with 10 7 OCI-LY3 cells and after 13 days, tumor-bearing animals (n=5 mice per group) received intraperitoneal (i.p.) injection of 2.5 mg/kg CPI203 (BID) and/or i.p. administration of 7.5 mg/kg IRAKi (BID), or an equal volume of vehicle, for 11 days, in a five/two (on/off) schedule. Tumor volumes were measured each 2-3 days with external calipers. (D) Immunohistological analysis of consecutive tumor sections from representative animals reveals a notable decrease in mitotic index and in the NF-κB-regulated CD44, as well as a strong downregulation of MCL-1 and induction of apoptosis in IRAKi-CPI203 combo group. HA. In contrast, the anti-IL-6 antibody failed to sensitize ABC-DLBCL cells to IRAKi-based treatment (Online Supplementary [/fig]
[bib_ref] Blockade of oncogenic IκB kinase activity in diffuse large B-cell lymphoma by..., Ceribelli [/bib_ref] |
Successful Management of Total Knee Replacement in a High Responder Hemophilia Patient With a History of Inhibitor
The development of inhibitors against administered clotting factors may render replacement therapy ineffective for some hemophilia patients. Such patients are therefore at the highest risk of developing arthropathy. Elective orthopedic surgery (EOS) in hemophilic patients having such inhibitors remains a rare, expensive, and difficult surgery, whose management represents a significant challenge. We report the case of a 35-year-old man with a severe form of hemophilia A (factor VIII < 1%), who was suffering from repetitive spontaneous hemarthrosis, especially in his knee joints that had consequently become more susceptible to bleeding. The patient had a history of high levels of factor VIII inhibitor (> 5.0 Bethesda Unit [BU]/ml) as shown by the factor VIII inhibitor assay; therefore, we began treatment with factor VIIa for his mildto-moderate bleeding (90 µg/kg intravenous bolus injections). The interval between injections varied with the severity of the hemorrhage in each bleeding episode. The inhibitor level reduced to 3.1 BU/ml after three months, to 1.6 BU/ml after six months, and disappeared completely after one year of treatment. We administered factor VIII at a dose of 50 IU/kg every eight hours during the first three post-operative days, then continued administration with a dose of 40 IU/kg every 12 hours for another four days, and observed a very good response to treatment with no bleeding. Recombinant activated factor VII (rFVIIa) is not an inhibitor-removal strategy, but an inhibitor-bypassing product. However, in our patient, the treatment of mild-to-moderate bleeding with short-term use of rFVIIa and no exposure to factor VIII caused a gradual reduction in the inhibitor level over a period of 1 year.
# Introduction
Hemophilic arthropathy is caused by recurrent episodes of hemorrhage into the joint, and if left untreated, it can lead to severe chronic pain and permanent functional disability [bib_ref] Haemophilic arthropathy from A to, Negrier [/bib_ref]. The most commonly affected joints are the knee, ankle, and elbow. However, the development of inhibitors against administered clotting factors in some hemophilia patients may render replacement therapy ineffective, and these patients are therefore at the highest risk of developing arthropathy [bib_ref] Haemophilic arthropathy from A to, Negrier [/bib_ref]. Although clotting factors are known to be effective for the treatment of arthropathy, joint bleeding and hence blood-induced joint damage are still commonly observed. There are several reasons for this, one of which is that not all patients (whether in our center or worldwide) have access to sufficient amounts of clotting factor, either due to limited availability or high costs. Although the number of cases and severity of joint bleeding reduces with the administration of clotting factors, such bleeding still occurs in a significant number of cases. Elective orthopedic surgery (EOS) is a last resort for patients with progressive arthropathy and orthopedic complications and end-stage arthropathy of most large joints. It can provide long-term cost savings as it reduces the bleeding frequency and causes a significant drop in pain levels and moreover restore mobility and function with minimal joint damage [bib_ref] Elective orthopaedic surgery for haemophilia patients with inhibitors: single centre experience of..., Caviglia [/bib_ref]. Additionally by and large, favorable functional outcomes following total knee replacement in hemophilic patients have been reported, with a favorable improvement in the clinical scores and range of movements (ROM) of patients. Nevertheless, approximately 20%-25% of hemophilia A patients and 1%-3% of hemophilia B patients have inhibitors for factor VIII and factor IX. Such cases represent a major therapeutic challenge to clinicians [bib_ref] Elective orthopaedic surgery for haemophilia patients with inhibitors: single centre experience of..., Caviglia [/bib_ref]. Arthroplastic surgery in hemophilic patients having such inhibitors remains a rare, expensive, and difficult surgery, whose management represents a significant challenge [bib_ref] Total joint replacement in patients with inhibitors, Solimeno [/bib_ref]. Since 1988, recombinant activated factor VII (rFVIIa) have been indicated for use in surgical prophylaxis, as well as for the treatment of bleeding episodes, in patients with hemophilia and high-responding inhibitors. Many studies support the use of rFVIIa as a first-line therapy in surgery for hemophilia patients with high-responding inhibitors, as it enables a safe surgical procedure without causing anamnesis (7, 8).
## Case report
We report the case of a 35-year-old man with a severe form of hemophilia A (factor VIII < 1%), who was suffering from repetitive spontaneous hemarthrosis, principally in his knee joints that had consequently become more susceptible to bleeding (Target Joint). The patient was diagnosed and treated at our center in 1993. He was first treated with cryoprecipitate, because we did not have factor concentrates at our center. Since 1996, he received factor concentrates (on demand) however, the amounts were insufficient. In addition, he did not receive any prophylaxis treatment. During the last year, repetitive spontaneous hemarthrosis led to chronic synovitis and ultimately to degenerative arthritis, and his arthropathy became more severe. Because of repeated hemarthrosis, he developed synovial hypertrophy, cartilage destruction, bone damage, and disabling arthritis, which is known as chronic hemophilic arthropathy and impairs joint function and inducing pain. The patient was disabled due to persistent pain and dysfunction caused by osteoarthritis secondary to progressive severe hemophilic arthropathy; therefore, the patient was referred to our Orthopedic Consultant. Upon physical examination of both knees, the following findings were noted:
1. Pain-reduced ROM and deformities 2. Grade IV arthropathy in both knees 3. ROM of the knees was 0-90° without contracture (8) Radiological assessment revealed enlargement of the epiphysis, osteoporosis, erosions, osteophyte formation, cartilage damage, and ankylosis. Hemarthrosis led to inflammatory changes in the synovial tissue and degenerative changes in the cartilage as well as affecting the subchondral bone. The patient had tested positive for Hepatitis C Virus, which was treated seven years earlier with interferon and ribavirin, and this treatment showed a favorable response. He also had a history of high levels of factor VIII inhibitor (> 5.0 Bethesda Unit [BU]/ml) as shown by the factor VIII inhibitor assay. We therefore began treatment with factor VIIa via 90 µg/kg intravenous bolus injection for his mild-to-moderate bleeding. The interval between injections varied with the severity of the hemorrhage in each bleeding episode. The inhibitor level reduced to 3.1 BU/ml after three months, to 1.6 BU/ml after 6 months, with complete remission within one year of treatment [fig_ref] Figure 1: The Inhibitor Assay of Patient During Treatment [/fig_ref]. In a recent radiograph of the patient's left knee, we found lateral osteophyte formation in the lateral tibia, fusion of the patella in the distal end of the femur, and a severe reduction in the joint space. The patient was in the care of a multidisciplinary team of specialists including an orthopedic surgeon, a hematologist, and a physiotherapist. Total knee arthroplasty (TKA) was then performed on his right knee. We used factor VIII at a dose of 50 IU/ kg every eight hours during the first three post-operative days, subsequently continued its administration at a dose of 40 IU/kg every 12 hours for another four days, and observed a very good response to treatment with no bleeding.
# Discussion
The interesting feature of this case is that the inhibitor titer of a patient with hemophilia was successfully reduced from > 5.0 BU/ml to approximately 0 BU/ml within one year, without surgery, suggesting that the patient did not require any high-dose or high-costing bypassing agents during the TKA operation. A 90 µg/kg dose of rFVI-Ia has been shown to provide effective hemostasis in over 90% of mild-to-moderate bleeding cases after a mean of 2.2 injections [bib_ref] On-demand treatment of bleeds in haemophilia patients with inhibitors: strategies for securing..., Sorensen [/bib_ref]. Hence, only mild-to-moderate bleeding can be controlled with a low dose (and low cost) of rFVIIa, and the inhibitor level reduced in our patient because he was not exposed to factor VIII. According to Soren et al., the efficacy of on-demand treatment of spontaneous bleeding in inhibitor patients can be optimized by prescribing the highest appropriate dose of rFVIIa based on patients' previous response history. Due to an excellent response of our patient to a 90 µg/kg dose of rFVIIa, we did not consider the initially recommended dose of 270 µg/kg (9). Solimeno et al. reported that a high responder patient had a 1 BU/ml antibody titer at the time of surgery; a high dose of factor VIII (52000 U) was used for the first five days and on the fifth day after surgery, rFVIIa was administered for two days [bib_ref] Total joint replacement in patients with inhibitors, Solimeno [/bib_ref]. In our case, the patient's inhibitor titer was below 1 BU/ml before surgery, and we controlled his bleeding during and after surgery with a high dose of FVIII. Administration of bypassing agents was therefore unnecessary because of the excellent response to factor VIII. In the first randomized clinical trial reported to date on the prophylaxis in hemophilia patients having such inhibitors, rFVIIa reduced the bleeding frequency, duration of hospital stay, and duration of absence from work/school and improved the quality of life as compared to conventional on-demand therapy. In a pharmaco-economic analysis of bypassing agents in the treatment of minor-to-moderate bleeding in hemophilia patients with such inhibitors, the rFVIIa-only strategy was the least expensive [bib_ref] Pharmacoeconomic analysis of recombinant factor VIIa versus APCC in the treatment of..., Joshi [/bib_ref]. It is clear that rFVIIa is not an inhibitor-removal strategy; instead, it is an inhibitorbypassing product. However, in our case, because of the short-term use of rFVIIa for the treatment of mild-tomoderate bleeding and because the patient was not exposed to factor VIII, a gradual reduction of inhibitor level was possible over a period of one year. It is recommended that only in long-term treatments, immune tolerance may be induced, and desensitizing the immune system and eradicating the inhibitor by the infusion of frequent large doses of factor replacement may be possible. However, short-term, cost effective, and safe solutions may also be required, particularly during the management of acute bleeding episodes [bib_ref] Managing acute bleeds in the patient with haemophilia and inhibitors: options, efficacy..., Von Depka [/bib_ref]. EOS in hemophilic patients is a complex procedure that should be performed in a ter-tiary center with support from a dedicated hemophilia team. Further trials on such patients are needed to confirm our findings.
[fig] Figure 1: The Inhibitor Assay of Patient During Treatment [/fig]
|
Diagnostics of COVID-19 Based on CRISPR–Cas Coupled to Isothermal Amplification: A Comparative Analysis and Update
# Introduction
Coronaviruses have caused important outbreaks in recent years, for example, Severe Acute Respiratory Syndrome-related coronavirus (SARS-CoV) in 2002, Middle East Respiratory Syndrome-related coronavirus (MERS-CoV) in 2010, and, most recently, Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2), the etiologic agent of the Corona Virus Disease 2019 . COVID-19 was first reported in December 2019 by the Chinese Center for Disease Control and Prevention attending Wuhan local health facilities [bib_ref] A Novel Coronavirus from Patients with Pneumonia in China, Zhu [/bib_ref]. By the time of writing this review, the World Health Organization has confirmed more than 508 million cases and 6.2 million deaths. The betacoronavirus SARS-CoV-2 virions (60-140 nm in diameter) are composed of a single-stranded positive-sense RNA molecule packed in a coating protein and enveloped into lipids [bib_ref] Molecular Architecture of the SARS-CoV-2 Virus, Yao [/bib_ref]. The virions characteristically display on their surface pendant "spike" proteins which play a key role in the binding and entry to the host human cells [bib_ref] Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation, Wrapp [/bib_ref]. The~29.9 kb SARS-CoV-2 genome encodes for 13-15 Open Reading Frames (ORFs) that express a total of 12 proteins, including the non-structural ORF1a and ORF1b and the structural envelope (E), membrane (M), nucleoprotein (N), and spike (S) proteins [bib_ref] The proximal origin of SARS-CoV-2, Andersen [/bib_ref] [bib_ref] The Architecture of SARS-CoV-2 Transcriptome, Kim [/bib_ref].
The fast propagation and global distribution of COVID-19 have spurred intensive research that aims to develop novel diagnostic methods that could assist in detecting new variants and stopping their propagation. Diagnostic methods based on Clustered Regularly Interspaced Short Palindromic Repeats and CRISPR-associated proteins (CRISPR-Cas variants and stopping their propagation. Diagnostic methods based on Clustered Regularly Interspaced Short Palindromic Repeats and CRISPR-associated proteins (CRISPR-Cas systems) are among the most promising [bib_ref] CRISPR-based diagnostics, Kaminski [/bib_ref] [bib_ref] CRISPR/Cas Systems towards Next-Generation Biosensing, Li [/bib_ref] [bib_ref] Two distinct RNase activities of CRISPR-C2c2 enable guide-RNA processing and RNA detection, East-Seletsky [/bib_ref] [bib_ref] C2c2 is a single-component programmable RNA-guided RNA-targeting CRISPR effector, Abudayyeh [/bib_ref]. This is because they can potentially comply with the "ASSSURED" features: Accurate, Specific, Sensitive, Simple, Rapid, Equipment-free, and Deliverable to end-users [bib_ref] REASSURED diagnostics to inform disease control strategies, strengthen health systems and improve..., Land [/bib_ref]. In addition, CRISPR-Cas systems offer low-cost reactions, are highly versatile and flexible to adapt to new virus variants [bib_ref] CRISPR-Cas12a-Based Detection for the Major SARS-CoV-2 Variants of, Yuanhao [/bib_ref] [bib_ref] Rapid identification and tracking of SARS-CoV-2 variants of concern, Chakraborty [/bib_ref] and future emerging pandemics, and are suitable for large-scale production. A remarkable feature is that CRISPR-Cas systems can be synergically combined with isothermal RNA amplification methods to bypass the disadvantages of using both techniques separately [bib_ref] CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity, Chen [/bib_ref] [bib_ref] CRISPR-Cas12-based detection of SARS-CoV-2, Broughton [/bib_ref]. When the viral RNA is retrotranscribed into DNA and amplified using an isothermal method, the CRISPR-Cas system can detect it with a specificity and sensitivity similar [bib_ref] Clinical validation of a Cas13-based assay for the detection of SARS-CoV-2 RNA, Patchsung [/bib_ref] [bib_ref] Intrinsic signal amplification by type III CRISPR-Cas systems provides a sequence-specific SARS-CoV-2..., Santiago-Frangos [/bib_ref] [bib_ref] Vigilant: An Engineered VirD2-Cas9 Complex for Lateral Flow Assay-Based Detection of SARS-CoV2, Marsic [/bib_ref] [bib_ref] Rapid, Ultrasensitive, and Highly Specific Diagnosis of COVID-19 by CRISPR-Based Detection, Zhu [/bib_ref] [bib_ref] Development and evaluation of a rapid CRISPR-based diagnostic for COVID-19, Hou [/bib_ref] to the benchmark or gold standard: quantitative reverse transcription Polymerase Chain Reaction (RT-qPCR) [bib_ref] First Case of 2019 Novel Coronavirus in the United States, Holshue [/bib_ref] [bib_ref] Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR, Corman [/bib_ref]. CRISPR-Cas methods can also be designed in a portable format without the need for sophisticated instruments and highly skilled personnel, therefore being particularly suitable for deployment in low-resource point-of-care locations (POC). Its great potential is reflected in the steady growth in the number of reports using CRISPR-Cas to diagnose COVID-19 [bib_ref] Enhancement of trans-cleavage activity of Cas12a with engineered crRNA enables amplified nucleic..., Nguyen [/bib_ref] , and 203 references for 2019, 2020, and 2021, respectively, according to PubMed). [bib_ref] Digital CRISPR/Cas-Assisted Assay for Rapid and Sensitive Detection of SARS-CoV-2, Park [/bib_ref] (left) and (C) LbuCas13a (type VI, PDB: 5XWP)(right). Colors represent different domains of Cas proteins. LbCas12a: Wedge I, II, and III (yellow), REC1 (light gray), REC2 (dark gray), PI (wheat), RuvC-I, II, and II (cyan), BH (green lime), and Nuc (magenta). LbuCas13a: NTD (cyan), Helical-1 (wheat), HEPN1-I and II (green lime), Helical-2 (yellow), Linker (orange), and HEPN2 (magenta). Schematics depicting DNA in black and the primers used for isothermal amplification methods in colors: RT-LAMP (D) and RT-RPA (E) and the target sequence for CRISPR-Cas systems (green).
In this review, we comprehensively analyzed and compared more than 50 publicly available reports that correspond to 42 different methods for detection of SARS-CoV-2 that use CRISPR-Cas systems together with isothermal amplification (in this review referred as the "CRISPR-Cas method"). From each of the 42 methods, we extract important technical details . All the compiled information is displayed, sorted, and ranked in [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. After a systematic comparison between the methods, we found the most successful methods in terms of the time to deliver a result (time consumed in the isothermal amplification and CRISPR-based detection), the limit of detection, and the clinical sensitivity and specificity. We also discuss key methodological [bib_ref] Digital CRISPR/Cas-Assisted Assay for Rapid and Sensitive Detection of SARS-CoV-2, Park [/bib_ref] (left) and (C) LbuCas13a (type VI, PDB: 5XWP)(right). Colors represent different domains of Cas proteins. LbCas12a: Wedge I, II, and III (yellow), REC1 (light gray), REC2 (dark gray), PI (wheat), RuvC-I, II, and II (cyan), BH (green lime), and Nuc (magenta). LbuCas13a: NTD (cyan), Helical-1 (wheat), HEPN1-I and II (green lime), Helical-2 (yellow), Linker (orange), and HEPN2 (magenta). Schematics depicting DNA in black and the primers used for isothermal amplification methods in colors: RT-LAMP (D) and RT-RPA (E) and the target sequence for CRISPR-Cas systems (green).
In this review, we comprehensively analyzed and compared more than 50 publicly available reports that correspond to 42 different methods for detection of SARS-CoV-2 that use CRISPR-Cas systems together with isothermal amplification (in this review referred as the "CRISPR-Cas method"). From each of the 42 methods, we extract important technical details (Supplementary . All the compiled information is displayed, sorted, and ranked in [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. After a systematic comparison between the methods, we found the most successful methods in terms of the time to deliver a result (time consumed in the isothermal amplification and CRISPR-based detection), the limit of detection, and the clinical sensitivity and specificity. We also discuss key methodological parameters, such as method of RNA extraction, type of isothermal method, and the CRISPR-Cas system, among others that are related to achieving a successful result. Recognizing key steps and optimizable parameters could help the CRISPR diagnostics (CRISPR-Dx) community to generate optimization and innovations that can contribute to developing more robust methods.
## Crispr-cas in diagnostics
The CRISPR-Cas systems are memory-like defense mechanisms present in bacteria and archaea that prevent the invasion of foreign nucleic acids such as bacteriophages or plasmids [bib_ref] CRISPR-Mediated Adaptive Immune Systems in Bacteria and Archaea, Sorek [/bib_ref]. CRISPR-Cas systems recognize and hydrolyze the foreign nucleic acid through a ribonucleoprotein complex (RNP) composed of Cas proteins with endonucleolytic activity guided by a CRISPR ribonucleic acid (crRNA) [bib_ref] Two distinct RNase activities of CRISPR-C2c2 enable guide-RNA processing and RNA detection, East-Seletsky [/bib_ref] [bib_ref] C2c2 is a single-component programmable RNA-guided RNA-targeting CRISPR effector, Abudayyeh [/bib_ref] [bib_ref] CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity, Chen [/bib_ref] [bib_ref] CRISPR-Mediated Adaptive Immune Systems in Bacteria and Archaea, Sorek [/bib_ref] [bib_ref] DNA targeting specificity of RNA-guided Cas9 nucleases, Hsu [/bib_ref] [bib_ref] CRISPR-Cas12a-assisted nucleic acid detection, Li [/bib_ref]. The class 2 CRISPR-Cas systems, such as CRISPR-Cas9, Cas12, Cas13, and Cas14, are by far the most used in diagnostics because they only need a single Cas protein for the recognition and cleavage of the target nucleic acid sequence.
During the detection, the RNP first scans the DNA/RNA sequence to find a protospacer adjacent motif (PAM) (or a protospacer flanking site, PFS, in the RNA) and then opens the dsDNA to hybridize with the crRNA sequence to form the so-called R-loop [bib_ref] CRISPR-Cas9 Structures and Mechanisms, Jiang [/bib_ref]. Cas12 (type V), Cas13 (type VI), and Cas14 (Type V) differentiate from Cas9 (type II) because after cutting the specific target sequence (cis cleavage), they also cut proximate singlestranded DNA (ssDNA) (Cas12a and Cas14) or ssRNA (Cas13a) through a trans-cleavage activity [bib_ref] Two distinct RNase activities of CRISPR-C2c2 enable guide-RNA processing and RNA detection, East-Seletsky [/bib_ref] [bib_ref] C2c2 is a single-component programmable RNA-guided RNA-targeting CRISPR effector, Abudayyeh [/bib_ref] [bib_ref] CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity, Chen [/bib_ref] [bib_ref] CRISPR-Cas12a-assisted nucleic acid detection, Li [/bib_ref]. This very effective collateral activity is what most CRISPR-based diagnostic systems exploit to generate a fluorescent or detectable signal through the degradation of a nucleic acid probe bi-labelled with a fluorophore and a quencher. This probe can also be adapted to be used in lateral flow strips, which allow direct visual detection.
## Nucleic acid isothermal amplification in diagnostics
The pre-amplification of the target nucleic acid by isothermal methods can increase the analytical sensitivity of detection of CRISPR-Cas systems up to 10 9 times in one hour or less [bib_ref] CRISPR-Cas12a-assisted nucleic acid detection, Li [/bib_ref]. Besides this, isothermal methods are the preferable alternative over Polymerase Chain Reaction (PCR) in the field of diagnostics in low-resource areas because they do not require expensive laboratory equipment such as thermocyclers [bib_ref] Loop-mediated isothermal amplification of DNA, Notomi [/bib_ref]. Since all the reactions occur at a constant temperature, the isothermal amplification of DNA can be performed in common incubators and dry block heaters or even using low-cost hand warmers [bib_ref] Ultrasensitive and visual detection of SARS-CoV-2 using all-in-one dual CRISPR-Cas12a assay, Ding [/bib_ref]. Likewise, it does not need specialized laboratories with specially trained personnel, and it can deliver results in less than one hour, which greatly increases accessibility and allows it to process a higher number of samples. All these advantages have led to the application of isothermal amplification methods to detect multiple pathogens, including SARS-CoV-2 [bib_ref] Development of a reverse transcription-loop-mediated isothermal amplification as a rapid early-detection method..., Baek [/bib_ref] [bib_ref] A Novel Reverse Transcription Loop-Mediated Isothermal Amplification Method for Rapid Detection of..., Lu [/bib_ref] [bib_ref] Karunasagar, I. Isothermal amplification-based assays for rapid and sensitive detection of severe..., Maiti [/bib_ref] [bib_ref] Rapid and visual detection of 2019 novel coronavirus (SARS-CoV-2) by a reverse..., Yan [/bib_ref] [bib_ref] Isothermal Amplification of Nucleic Acids, Zhao [/bib_ref].
In isothermal amplification, instead of using denaturing heating cycles as in the case of PCR, DNA unwinding is achieved by enzymes with strand displacement activity that work at a constant temperature. The enzymes also initiate the amplification by enabling the binding of the primers. RNA detection can be achieved by adding a reverse transcriptase. Once the DNA amplicons accumulate in large amounts in the solution, detection is accomplished by agarose gel electrophoresis, turbidity, colorimetry, or fluorescence [bib_ref] Real-time Detection and Monitoring of Loop Mediated Amplification (LAMP) Reaction Using Self-quenching..., Gadkar [/bib_ref]. Detection by turbidity happens due to the accumulation of magnesium pyrophosphate (Mg 2 P 2 O 7 ). If intercalating dyes, such as SYBR Green I or EvaGreen, or metal ion indicators, such as calcein/Mn 2+ and hydroxynapthol blue dye, are used, then fluorescence or color are detected, respectively.
Some of the isothermal methods most frequently used are Recombinase Polymerase Amplification (RPA) [bib_ref] DNA Detection Using Recombination Proteins, Piepenburg [/bib_ref] and Loop-mediated Isothermal Amplification (LAMP) [bib_ref] Loop-mediated isothermal amplification of DNA, Notomi [/bib_ref]. RPA uses a polymerase with strand displacement activity at a constant temperature between 37 and 42 - C, while LAMP is carried out at a temperature between 60 and 65 - C. RPA amplifies the target sequence by using forward and backward primers, stabilized in a complex formed by a recombinase, and a single-stranded DNA binding protein which stabilizes the interaction and allows the action of the DNA polymerase. Instead, LAMP needs Bst DNA polymerase and four or six specific primers grouped in pairs.
Isothermal amplification methods present the major disadvantages of using nonspecific detection methods (turbidity, fluorescence, or colorimetry), which could lead to an increase in detecting false positives in case of the presence of non-specific amplicons due to cross-contamination during pre-or post-processing [bib_ref] Real-time Detection and Monitoring of Loop Mediated Amplification (LAMP) Reaction Using Self-quenching..., Gadkar [/bib_ref] [bib_ref] Challenges and perspectives in the application of isothermal DNA amplification methods for..., Martzy [/bib_ref]. Therefore, successful isothermal methods need to carefully optimize primer design and concentration, and temperature of the method.
When isothermal amplification methods are coupled to CRISPR-Cas systems, these limitations can be overcome, besides allowing an increase in sensitivity and specificity [bib_ref] CRISPR technology incorporating amplification strategies: Molecular assays for nucleic acids, proteins, and..., Feng [/bib_ref]. In this format, isothermal methods exponentially pre-amplify the targeted sequence, while the detection now is left to a highly specific CRISPR-Cas system. Once the amplicon has been produced, instead of linking its detection to non-specific DNA detection, it is delegated to the programmable CRISPR-Cas system, which detects it more accurately and precisely and generates an exponential signal upon the amplified DNA. Besides, CRISPR-Cas doublechecks the presence of the target sequence (firstly performed by the primers of the isothermal method). This combination has led to the most extended method format for the diagnostic of SARS-CoV-2 based on CRISPR-Cas coupled to an isothermal amplification with a reverse transcriptase (RT-RPA or RT-LAMP). For simplification, here we will refer as "CRISPR-Cas based method" or "CRISPR-diagnostics" to the combination of a CRISPR system with an isothermal method. Here, we are not reviewing isothermal-only [bib_ref] Karunasagar, I. Isothermal amplification-based assays for rapid and sensitive detection of severe..., Maiti [/bib_ref] [bib_ref] Rapid and visual detection of 2019 novel coronavirus (SARS-CoV-2) by a reverse..., Yan [/bib_ref] [bib_ref] Clinical Evaluation of Self-Collected Saliva by Quantitative Reverse Transcription-PCR (RT-qPCR), Direct RT-qPCR,..., Nagura-Ikeda [/bib_ref] [bib_ref] A colorimetric RT-LAMP assay and LAMP-sequencing for detecting SARS-CoV-2 RNA in clinical..., Thi [/bib_ref] , PCR/CRISPR-Cas [bib_ref] Easy-to-Deploy Protocol for Cas13-Based Detection of SARS-CoV-2 Genetic Material, Rauch [/bib_ref] or amplification-free CRISPR-Cas [bib_ref] Application of the amplification-free SERS-based CRISPR/Cas12a platform in the identification of SARS-CoV-2..., Liang [/bib_ref] [bib_ref] Amplification-free RNA detection with CRISPR-Cas13, Shinoda [/bib_ref] [bib_ref] Amplification-free detection of SARS-CoV-2 with CRISPR-Cas13a and mobile phone microscopy, Fozouni [/bib_ref] methods to detect the SARS-CoV-2.
## General procedure to detect sars-cov-2 with crispr-cas
The whole workflow of the CRISPR-based methods comprises five general steps: (1) collection of the clinical sample, (2) preparation of the viral genomic RNA, (3) isothermal amplification of the targeted sequence, (4) target recognition and generation of a molecular signal by the CRISPR-Cas system, and (5) signal read-out [fig_ref] Figure 2: General workflow to detect SARS-CoV-2 with CRISPR-based test includes five general steps [/fig_ref]. ure 1C). RPA uses a polymerase with strand displacement activity at a constant temperature between 37 and 42 °C, while LAMP is carried out at a temperature between 60 and 65 °C. RPA amplifies the target sequence by using forward and backward primers, stabilized in a complex formed by a recombinase, and a single-stranded DNA binding protein which stabilizes the interaction and allows the action of the DNA polymerase. Instead, LAMP needs Bst DNA polymerase and four or six specific primers grouped in pairs.
Isothermal amplification methods present the major disadvantages of using non-specific detection methods (turbidity, fluorescence, or colorimetry), which could lead to an increase in detecting false positives in case of the presence of non-specific amplicons due to cross-contamination during pre-or post-processing [bib_ref] Real-time Detection and Monitoring of Loop Mediated Amplification (LAMP) Reaction Using Self-quenching..., Gadkar [/bib_ref] [bib_ref] Challenges and perspectives in the application of isothermal DNA amplification methods for..., Martzy [/bib_ref]. Therefore, successful isothermal methods need to carefully optimize primer design and concentration, and temperature of the method.
When isothermal amplification methods are coupled to CRISPR-Cas systems, these limitations can be overcome, besides allowing an increase in sensitivity and specificity [bib_ref] CRISPR technology incorporating amplification strategies: Molecular assays for nucleic acids, proteins, and..., Feng [/bib_ref]. In this format, isothermal methods exponentially pre-amplify the targeted sequence, while the detection now is left to a highly specific CRISPR-Cas system. Once the amplicon has been produced, instead of linking its detection to non-specific DNA detection, it is delegated to the programmable CRISPR-Cas system, which detects it more accurately and precisely and generates an exponential signal upon the amplified DNA. Besides, CRISPR-Cas double-checks the presence of the target sequence (firstly performed by the primers of the isothermal method). This combination has led to the most extended method format for the diagnostic of SARS-CoV-2 based on CRISPR-Cas coupled to an isothermal amplification with a reverse transcriptase (RT-RPA or RT-LAMP). For simplification, here we will refer as "CRISPR-Cas based method" or "CRISPR-diagnostics" to the combination of a CRISPR system with an isothermal method. Here, we are not reviewing isothermal-only [bib_ref] Karunasagar, I. Isothermal amplification-based assays for rapid and sensitive detection of severe..., Maiti [/bib_ref] [bib_ref] Rapid and visual detection of 2019 novel coronavirus (SARS-CoV-2) by a reverse..., Yan [/bib_ref] [bib_ref] Clinical Evaluation of Self-Collected Saliva by Quantitative Reverse Transcription-PCR (RT-qPCR), Direct RT-qPCR,..., Nagura-Ikeda [/bib_ref] [bib_ref] A colorimetric RT-LAMP assay and LAMP-sequencing for detecting SARS-CoV-2 RNA in clinical..., Thi [/bib_ref] , PCR/CRISPR-Cas [bib_ref] Easy-to-Deploy Protocol for Cas13-Based Detection of SARS-CoV-2 Genetic Material, Rauch [/bib_ref] or amplification-free CRISPR-Cas [bib_ref] Application of the amplification-free SERS-based CRISPR/Cas12a platform in the identification of SARS-CoV-2..., Liang [/bib_ref] [bib_ref] Amplification-free RNA detection with CRISPR-Cas13, Shinoda [/bib_ref] [bib_ref] Amplification-free detection of SARS-CoV-2 with CRISPR-Cas13a and mobile phone microscopy, Fozouni [/bib_ref] methods to detect the SARS-CoV-2.
## General procedure to detect sars-cov-2 with crispr-cas
The whole workflow of the CRISPR-based methods comprises five general steps: (1) collection of the clinical sample, (2) preparation of the viral genomic RNA, (3) isothermal amplification of the targeted sequence, (4) target recognition and generation of a molecular signal by the CRISPR-Cas system, and (5) signal read-out [fig_ref] Figure 2: General workflow to detect SARS-CoV-2 with CRISPR-based test includes five general steps [/fig_ref].
## Step 1: collection of clinical sample
## Step 1: collection of clinical sample
Sample collection is of the utmost importance to avoid misleading results since it aims to pick up the virus while preserving the integrity of the genomic RNA [bib_ref] SARS-CoV-2 pandemic: A review of molecular diagnostic tools including sample collection and..., Jayamohan [/bib_ref] [bib_ref] Self-sampling versus health care professional-guided swab collection for SARS-CoV-2 testing, Wurstle [/bib_ref]. Furthermore, it needs to be managed without putting at risk the healthcare provider. Since the SARS-CoV-2 infects cells from the upper respiratory tract, most of the samples are taken through nasopharyngeal (NP) and/or oropharyngeal (OP) swabs [bib_ref] Combined throat/nasal swab sampling for SARS-CoV-2 is equivalent to nasopharyngeal sampling, Vlek [/bib_ref] [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. The swabs are commonly stored and transported in Universal or Viral Transport Medium and sent to the analysis lab.
## Step 2: rna preparation
Once the sample has been collected and transported to the analysis location, it needs to be processed in order to make the viral RNA suitable for amplification and detection. The sample can be processed in two ways: (i) extraction to obtain a highly pure RNA or (ii) using a combination of chemical and physical treatments to remove viral components and release an RNA in a less pure form [bib_ref] SARS-CoV-2 RNA Extraction Using Magnetic Beads for Rapid Large-Scale Testing by RT-qPCR..., Klein [/bib_ref] [bib_ref] Direct-RT-qPCR Detection of SARS-CoV-2 without RNA Extraction as Part of a COVID-19..., Kriegova [/bib_ref] [bib_ref] A Saliva-Based RNA Extraction-Free Workflow Integrated With Cas13a for SARS-CoV-2 Detection, Azmi [/bib_ref] [bib_ref] An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid..., Ooi [/bib_ref] [bib_ref] Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2, Arizti-Sanz [/bib_ref] [bib_ref] Rapid and Sensitive Detection of SARS-CoV-2 Using Clustered Regularly Interspaced Short Palindromic..., Tsou [/bib_ref] [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. Extracting methods are typically based on columns and magnetic beads commercialized as kits (e.g., from Qiagen) and yield highly pure RNA [bib_ref] Evaluation of Methods for the Concentration and Extraction of Viruses from Sewage..., Hjelmsø [/bib_ref] [bib_ref] Current Nucleic Acid Extraction Methods and Their Implications to Point-of-Care Diagnostics, Ali [/bib_ref]. They efficiently remove enzyme inhibitors and other contaminants, thus facilitating further downstream steps of the process workflow (RNA amplification and detection and signal generation). On the other hand, "releasing" methods combine chemical (e.g., lysis buffers) with physical treatments (e.g., temperature and centrifugation) to remove viral and patient-derived cell inhibitors, contaminants, and human RNAses [bib_ref] CRISPR-based diagnostics, Kaminski [/bib_ref] [bib_ref] A Saliva-Based RNA Extraction-Free Workflow Integrated With Cas13a for SARS-CoV-2 Detection, Azmi [/bib_ref] [bib_ref] An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid..., Ooi [/bib_ref] [bib_ref] Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2, Arizti-Sanz [/bib_ref] [bib_ref] Current Nucleic Acid Extraction Methods and Their Implications to Point-of-Care Diagnostics, Ali [/bib_ref] [bib_ref] Ultrasensitive CRISPR-based diagnostic for field-applicable detection of Plasmodium species in symptomatic and..., Lee [/bib_ref]. This makes the genomic RNA available for further enzymatic steps while providing suitable conditions to components working in further steps (e.g., polymerase, Cas protein, and gRNA).
## Step 3: isothermal amplification of target sequence
In this step, the viral RNA is retro-transcribed into DNA by a reverse transcriptase (RT), and then the target sequence is exponentially amplified by the chosen isothermal method [bib_ref] Loop-mediated isothermal amplification of DNA, Notomi [/bib_ref] [bib_ref] Development of a reverse transcription-loop-mediated isothermal amplification as a rapid early-detection method..., Baek [/bib_ref] [bib_ref] A Novel Reverse Transcription Loop-Mediated Isothermal Amplification Method for Rapid Detection of..., Lu [/bib_ref] [bib_ref] Isothermal Amplification of Nucleic Acids, Zhao [/bib_ref] [bib_ref] DNA Detection Using Recombination Proteins, Piepenburg [/bib_ref] [bib_ref] Rapid Detection of SARS-CoV-2 by Low Volume Real-Time Single Tube Reverse Transcription..., Behrmann [/bib_ref] [bib_ref] Recent advances and perspectives of nucleic acid detection for coronavirus, Shen [/bib_ref]. This step is critical to achieving a high analytical sensitivity since it quickly makes millions of new copies of the targeted sequence. However, it might also copy incorrect sequences, thus it needs to be properly optimized. The most common isothermal methods are Reverse transcriptase Recombinase Polymerase Amplification (RT-RPA) [bib_ref] DNA Detection Using Recombination Proteins, Piepenburg [/bib_ref] [bib_ref] Rapid Detection of SARS-CoV-2 by Low Volume Real-Time Single Tube Reverse Transcription..., Behrmann [/bib_ref] and reverse transcriptase Loop-mediated Isothermal Amplification (RT-LAMP) [bib_ref] Loop-mediated isothermal amplification of DNA, Notomi [/bib_ref] [bib_ref] Development of a reverse transcription-loop-mediated isothermal amplification as a rapid early-detection method..., Baek [/bib_ref] [bib_ref] A Novel Reverse Transcription Loop-Mediated Isothermal Amplification Method for Rapid Detection of..., Lu [/bib_ref] [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref].
## Step 4: target detection and signal generation by crispr-cas
Here, the RNP complex (Cas protein:crRNA) binds to the target sequence (previously amplified) and cleaves it through its cis nuclease activity [bib_ref] CRISPR-Mediated Adaptive Immune Systems in Bacteria and Archaea, Sorek [/bib_ref] [bib_ref] CRISPR-Cas9 Structures and Mechanisms, Jiang [/bib_ref]. The RNP complex has to be formed previously, usually either in parallel to the isothermal amplification step or beforehand by incubation 10-30 min at 37 - C or even at room temperature. Most of the methods based their detection on CRISPR-Cas12 or Cas13 because they present a trans nuclease activity that hydrolyzes the nucleic acid probe, unleashing the reporter signal to be detected [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref] [bib_ref] CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity, Chen [/bib_ref]. The probe is a short single-stranded oligonucleotide (ssDNA for Cas12 and ssRNA for Cas13) conjugated to a pair of reporter molecules such as a fluorophore/quencher or antigen/biotin [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref].
## Step 5: signal read-out
This final step provides the diagnostic result based on fluorescence or a colorimetric band detection on lateral flow dipsticks (LF) [bib_ref] CRISPR-based diagnostics, Kaminski [/bib_ref] [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. In the first case, probe with a fluorophore and a quencher is used, and the fluorescence is released upon its hydrolysis by CRISPR-Cas [bib_ref] CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity, Chen [/bib_ref]. Fluorescence can be detected by plate readers, real-time thermocyclers, cuvette-based fluorimetry (especially in kinetic studies), or by portable UV/blue transilluminators for naked-eye detection or coupled to cell phone detection. LF requires a probe conjugated to an antigen (usually fluorescein -FAM-) and biotin. Incubating the dipstick in a solution with the hydrolyzed probe, a test red-like color band detectable by the naked-eye appears. LF are simple to incorporate in portable methods deployable in POC. Most of the methods will deliver a qualitative result ("presence" or "no presence" of the targeted viral RNA). However, there are very few CRISPR-Cas methods that allow quantification of the SARS-CoV-2 molecules [bib_ref] Sensitive quantitative detection of SARS-CoV-2 in clinical samples using digital warm-start CRISPR..., Ding [/bib_ref] , although they are complex and difficult to scale-up.
## Key experimental parameters
Within each of the five steps that depict the whole detection process, there is an inherent methodological complexity [fig_ref] Figure 2: General workflow to detect SARS-CoV-2 with CRISPR-based test includes five general steps [/fig_ref]. We found several key components and conditions within each step are related to the observed performance and final result and show significant variability across the analyzed methods . Recognizing the importance of these parameters could guide further optimization, leading to improving the experimental outputs of the detection methods such as total time, the limit of detection, sensitivity, and specificity. . Key parameters and their variability across the CRISPR-based methods to detect SARS-CoV-2.
## Key parameter (condition/component) options
Step 1 Type of sample Nasopharyngeal and Oropharyngeal swabs, viral RNA, saliva, sputum, other
Step 2 Type and time of method of preparation RNA extraction (5-40 min) and RNA release (5-30 min)
Step 3
Targeted Step 4
Type CRISPR
## Type of sample
More than 80% of the reports have validated their methods using some type of clinical sample (Supplementary [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. The most common types are nasopharyngeal (NP) and/or oropharyngeal (OP) swabs. Likewise, saliva and sputum samples (15% of the analyzed methods) [bib_ref] A Saliva-Based RNA Extraction-Free Workflow Integrated With Cas13a for SARS-CoV-2 Detection, Azmi [/bib_ref] [bib_ref] Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2, Arizti-Sanz [/bib_ref] [bib_ref] Sensitive quantitative detection of SARS-CoV-2 in clinical samples using digital warm-start CRISPR..., Ding [/bib_ref] [bib_ref] A one-step, one-pot CRISPR nucleic acid detection platform (CRISPR-top): Application for the..., Li [/bib_ref] [bib_ref] Development of a Broadly Applicable Cas12a-Linked Beam Unlocking Reaction for Sensitive and..., Wang [/bib_ref] [bib_ref] Highly Sensitive and Specific System for COVID-19 Detection, Ma [/bib_ref] , which allow self-sampling directly by the patients [bib_ref] Self-Sampling for SARS-CoV-2 Diagnostic Testing by Using Nasal and Saliva Specimens: Protocol..., Majam [/bib_ref] [bib_ref] Self-collected oral, nasal and saliva samples yield sensitivity comparable to professionally collected..., Gertler [/bib_ref] , have been used successfully to detect the virus. Other less-used types of samples include bronchoalveolar lavage fluid, anal swabs, stool [bib_ref] A one-step, one-pot CRISPR nucleic acid detection platform (CRISPR-top): Application for the..., Li [/bib_ref] [bib_ref] Development of a Broadly Applicable Cas12a-Linked Beam Unlocking Reaction for Sensitive and..., Wang [/bib_ref] , and harvested lentivirus samples.
## Method of preparation: extraction vs. release of dna/rna
The viral RNA "extraction" methods, typically based on kits of columns and magnetic beads, warrant an RNA of high purity. However, using commercial kits is costly, and access to them could be limited when their global demand is high. On the other hand, "release" methods, which combine chemical (e.g., lysis buffers) with physical treatments (e.g., heat inactivation), are simpler but could potentially carry inhibitors and put at risk the optimal progress of the detection method. We found that >70% of all reviewed reports used extraction methods. This probably happens because they are commercially accessible and some of them have been approved for SARS-CoV-2 detection by local or international regulatory organizations. "Release" methods, instead, have been explored in~30% of the reports [bib_ref] Digital CRISPR/Cas-Assisted Assay for Rapid and Sensitive Detection of SARS-CoV-2, Park [/bib_ref] [bib_ref] A Saliva-Based RNA Extraction-Free Workflow Integrated With Cas13a for SARS-CoV-2 Detection, Azmi [/bib_ref] [bib_ref] An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid..., Ooi [/bib_ref] [bib_ref] Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2, Arizti-Sanz [/bib_ref] [bib_ref] Rapid and Sensitive Detection of SARS-CoV-2 Using Clustered Regularly Interspaced Short Palindromic..., Tsou [/bib_ref] [bib_ref] Sensitive quantitative detection of SARS-CoV-2 in clinical samples using digital warm-start CRISPR..., Ding [/bib_ref]. Mostly because they are cheaper than commercial kits. They can be homemade from commercially abundant compounds such as TCEP, DTT, EDTA, Triton X-100, and Proteinase K, helping to reduce dependency on costly commercial extraction kits. Furthermore, they present other advantages such as requiring simple heat-inactivation protocols which usually incubate the sample at different temperatures between 40 and 95 - C for some minutes (5-20 min). Improving "release" methods is a very important step ahead for CRISPR-Dx; however, they need to be carefully validated in the lab and in clinically relevant conditions to confirm their reliability and reproducibility.
Besides using RNA extracted from patients, synthetic DNA-and/or RNA-encoding viral genes are required to use during the characterization and optimization of the diagnostic method, and it is necessary to carry out systematic and controlled-conditions experiments. For example, synthetic nucleic acids alone [bib_ref] CRISPR-Cas12-based detection of SARS-CoV-2, Broughton [/bib_ref] [bib_ref] Ultrasensitive and visual detection of SARS-CoV-2 using all-in-one dual CRISPR-Cas12a assay, Ding [/bib_ref] [bib_ref] Rapid and accurate nucleobase detection using FnCas9 and its application in COVID-19..., Azhar [/bib_ref] or after being spiked into cells [bib_ref] A Saliva-Based RNA Extraction-Free Workflow Integrated With Cas13a for SARS-CoV-2 Detection, Azmi [/bib_ref] , biological matrices, or fluids (e.g., human saliva or sputum)are used to determine the limit of detection (LoD) of the method (analytical sensitivity). Spiked samples with RNA are also used as a surrogate of clinical samples (contrived specimens) [bib_ref] CRISPR-Cas12-based detection of SARS-CoV-2, Broughton [/bib_ref] [bib_ref] A Saliva-Based RNA Extraction-Free Workflow Integrated With Cas13a for SARS-CoV-2 Detection, Azmi [/bib_ref].
## Targeted genes
CRISPR-Cas systems can essentially target across the SARS-CoV-2 genome. There is on average 1 PAM site across the reference SARS-CoV-2 genome every 22.2 and 15 bp for Cas12a (PAM: 5 -TTTV-3 ) and Cas9 (5 -NGG-3 ), respectively. However, the gene that has been detected the most is N in~50% of reports. The large interest in gene N may be due to the fact that it is the gene targeted in the CDC's Diagnostic Test for COVID-19. Other targeted genes include Orf1ab (~21%), S (~20%), E (~10%), RdRp, M, Orf3, and others. Genes E and RdRp are recommended by institutions such as the World Health Organization. Gene S is also very important because it helps to discriminate between variants of interest and concern [7, [bib_ref] CRISPR-Cas12a-Based Detection for the Major SARS-CoV-2 Variants of, Yuanhao [/bib_ref] [bib_ref] Rapid identification and tracking of SARS-CoV-2 variants of concern, Chakraborty [/bib_ref] Thus far, it is not clear how mutations, secondary structures, and architecture of the SARS-CoV-2 genome and its transcriptome [bib_ref] The Architecture of SARS-CoV-2 Transcriptome, Kim [/bib_ref] could affect how the different regions are amplified by the isothermal methods and detected by CRISPR-Cas systems. However, the FDA has reported that some RT-qPCR tests were expected to fail to detect the SARS-CoV-2 Omicron variant due to deletions in the N-gene. By this token, isothermal methods amplifying similar regions could expect to fail as well.
## Type of isothermal amplification method
The most-used isothermal method is RT-RPA (~50% of reports), followed by RT-LAMP (~38%) (Supplementary [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. The former method is clearly the preferred choice; however, alternative methods include Recombinase Aided Amplification assay (RT-RAA), Multiple Cross Displacement Amplification (RT-MCDA), and Dual-Priming Isothermal Amplification (RT-DAMP). Besides, the widely used and available non-isothermal endpoint RT-PCR has also been used successfully together with CRISPR-Cas [bib_ref] Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR, Corman [/bib_ref] [bib_ref] Easy-to-Deploy Protocol for Cas13-Based Detection of SARS-CoV-2 Genetic Material, Rauch [/bib_ref]. Almost all the reports use commercially available isothermal methods or components, either from New England Biolabs© (RT-LAMP) or TwistDx© (RT-RPA). It is important to consider that the intellectual property behind the RPA method belongs to TwistDx Inc., whereas the patent of the LAMP technology has recently expired. Yet, homemade isothermal LAMP enzymes used in the method iScan achieved a significant analytical sensitivity of 10 viral copies per reaction and 100% of clinical specificity and sensitivity [bib_ref] iSCAN: An RT-LAMP-coupled CRISPR-Cas12 module for rapid, sensitive detection of SARS-CoV-2, Ali [/bib_ref] , indicating that homemade productions of isothermal enzymes perform effectively, efficiently, and could reduce costs.
## Time and temperature of isothermal method
The duration time for RT-LAMP was between 20 and 40 min with a median of 30 min (before detection with CRISPR-Cas), while for RT-RPA it was 15 to 30 min with a median of 25 min. The former was incubated between 59 and 65 - C (median of 62 - C), while the latter was between 37 and 42 - C (median of 42 - C). This means that RT-LAMP takes longer than RT-RPA. Furthermore, RT-RPA can be run in parallel (one-pot test) with CRISPR-Cas systems from mesophilic microorganisms. For RT-LAMP, the only way to circumvent the temperature limitation and to develop one-pot methods is using temperature-resistant EnAsCas12a (works at 60 - C) or thermostable AapCas12b, which run simultaneously with the RT-LAMP at 59-62 - C [bib_ref] A one-step, one-pot CRISPR nucleic acid detection platform (CRISPR-top): Application for the..., Li [/bib_ref] [bib_ref] iSCAN: An RT-LAMP-coupled CRISPR-Cas12 module for rapid, sensitive detection of SARS-CoV-2, Ali [/bib_ref] [bib_ref] Detection of SARS-CoV-2 with SHERLOCK One-Pot Testing, Joung [/bib_ref].
## Crispr-cas systems
Most of the used CRISPR-Cas systems present collateral activity (~90%) (Supplementary [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. Approximately~74% use a CRISPR-Cas12 system, either Cas12a (65%) or AapCas12b (10%), while 16% use CRISPR-Cas13 systems. Alternative systems are Cas9 (5%), Cas3 combined with multiprotein Cas complex (Cascade) from E. coli(CONAN method), and Cas10 [bib_ref] Intrinsic signal amplification by type III CRISPR-Cas systems provides a sequence-specific SARS-CoV-2..., Santiago-Frangos [/bib_ref].
The most preferable Cas12a protein comes from Lachnospiraceae bacterium (LbCas12a), followed by Acidaminococcus sp (AsCas12a), and just a few use Enhanced-AsCas12a or from Alicyclobacillus acidiphilus (AapCas12b). Regarding Cas13a, most methods are based on Leptotrichia wadei LwCas13a, followed by Leptotrichia bucalis LbCas13a. The predominance of LbCas12a is based on its commercial accessibility through vendors such as NEB and Integrated DNA Technologies (IDT). Cas13a is usually biosynthesized locally by the lab because it is less commercially available, although recently it has started to be commercialized as well. Furthermore, Cas12a needs a shorter crRNA (~40-44 nt) than Cas13a (~54-58 nt) or Cas9 (~100 nt). Additionally, CRISPR-Cas12 detects DNA targets directly, whereas Cas13 detects RNA targets, and thus it requires the transcription of the pre-amplified DNA into RNA using T7 RNA Polymerase. Another advantage of Cas12a is that it can successfully detect DNA with improved sensitivity using engineered hybrid crRNA-DNAs [bib_ref] An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid..., Ooi [/bib_ref] or Mn 2+ ions [bib_ref] Highly Sensitive and Specific System for COVID-19 Detection, Ma [/bib_ref]. Interestingly, the most frequently used concentrations of Cas and gRNA are 50 and 62.5 nM, while their intervals are 2.2-1000 and 20-1000 nM, respectively. The RNP ratio (Cas:gRNA) interval is 0.07-17.5, with RNP = 1 the most frequently used ratio.
## Type of read-out
Fluorescence is used as a read-out in~88% of the methods, with lateral flow dipsticks in~48% [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. However,~40% used both. Other seldom-reported read-out types include fluorescence anisotropy [bib_ref] Fluorescence polarization system for rapid COVID-19 diagnosis, Lee [/bib_ref] and electrophoresis in gel [bib_ref] Rapid and accurate nucleobase detection using FnCas9 and its application in COVID-19..., Azhar [/bib_ref]. For methods that incorporate Cas9 or systems lacking collateral activity, clever detection methods such as ELISA-like or conjugation of the gRNA and dCas9 with chemical groups detected by lateral flow strips have been applied [bib_ref] Vigilant: An Engineered VirD2-Cas9 Complex for Lateral Flow Assay-Based Detection of SARS-CoV2, Marsic [/bib_ref].
Fluorescence can be detected by dedicated specialized equipment or by low-tech solutions that are suitable for resource-constrained settings. In the first case, there are plate readers, real-time PCR thermocyclers, cuvette-based fluorimetry, or sophisticated high-throughput fluorescence set-ups such as Digital PCR chips [bib_ref] Digital CRISPR/Cas-Assisted Assay for Rapid and Sensitive Detection of SARS-CoV-2, Park [/bib_ref] [bib_ref] Sensitive quantitative detection of SARS-CoV-2 in clinical samples using digital warm-start CRISPR..., Ding [/bib_ref] (especially for kinetic studies and fluorescence quantification). In the second group, we can find UV/LED blue light transilluminators used in portable methods that allow direct detection with the naked-eye [bib_ref] Ultrasensitive and visual detection of SARS-CoV-2 using all-in-one dual CRISPR-Cas12a assay, Ding [/bib_ref] [bib_ref] Development of a Broadly Applicable Cas12a-Linked Beam Unlocking Reaction for Sensitive and..., Wang [/bib_ref] [bib_ref] Highly Sensitive and Specific System for COVID-19 Detection, Ma [/bib_ref] [bib_ref] One-pot visual RT-LAMP-CRISPR platform for SARS-cov-2 detection, Wang [/bib_ref] [bib_ref] Rapid and sensitive detection of COVID-19 using CRISPR/Cas12a-based detection with naked eye..., Wang [/bib_ref] or with cell phones [bib_ref] Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2, Arizti-Sanz [/bib_ref] [bib_ref] Machine Learning-Driven and Smartphone-Based Fluorescence Detection for CRISPR Diagnostic of SARS-CoV-2, Samacoits [/bib_ref].
On the other hand, colorimetric LF assays have been widely explored in diagnostic methods aiming to be fully deployable in points-of-care [bib_ref] CRISPR-based diagnostics, Kaminski [/bib_ref] [bib_ref] Rapid, Sensitive, and Specific Severe Acute Respiratory Syndrome Coronavirus 2 Detection: A..., Brandsma [/bib_ref]. LF assays are also suitable for deployment in resource-constrained settings. The most reported dipsticks are the commercial HybriDetect strips from Milenia Biotech© which are specially designed to work with reporter probes conjugated with FAM and biotin. The almost complete dependence on one provider of the strips is an issue that accounts for the bulk costs and also jeopardizes their availability during high-demand times. The development of simple methods to produce strips cheaply and massively, allowing flexibility to incorporate or modify the test and control lines, would be very beneficial.
## Reporter probe
Most of the CRISPR-Dx methods require an oligonucleotide reporter, although exceptional cases do not [bib_ref] Vigilant: An Engineered VirD2-Cas9 Complex for Lateral Flow Assay-Based Detection of SARS-CoV2, Marsic [/bib_ref]. Reported probes have a length between 5 and 16 nucleotides (nt) (ssDNA for Cas12 and ssRNA for Cas13). Probes used with Cas12a are usually rich in Ts and As. For example the most commonly reported are TTATT, TTATTATT, TATTAT-TAT, and TTATTATTATT [bib_ref] CRISPR-Cas12-based detection of SARS-CoV-2, Broughton [/bib_ref] [bib_ref] Rapid, Ultrasensitive, and Highly Specific Diagnosis of COVID-19 by CRISPR-Based Detection, Zhu [/bib_ref] [bib_ref] Digital CRISPR/Cas-Assisted Assay for Rapid and Sensitive Detection of SARS-CoV-2, Park [/bib_ref] [bib_ref] Ultrasensitive and visual detection of SARS-CoV-2 using all-in-one dual CRISPR-Cas12a assay, Ding [/bib_ref] [bib_ref] iSCAN: An RT-LAMP-coupled CRISPR-Cas12 module for rapid, sensitive detection of SARS-CoV-2, Ali [/bib_ref] [bib_ref] Detection of SARS-CoV-2 with SHERLOCK One-Pot Testing, Joung [/bib_ref] [bib_ref] Fluorescence polarization system for rapid COVID-19 diagnosis, Lee [/bib_ref] [bib_ref] Electric field-driven microfluidics for rapid CRISPR-based diagnostics and its application to detection..., Ramachandran [/bib_ref]. Other reported sequences are T-rich sequences such as TTTTTTT or TTTTT, or CCCCC [bib_ref] Sensitive quantitative detection of SARS-CoV-2 in clinical samples using digital warm-start CRISPR..., Ding [/bib_ref]. Some have used GoTaq ® commercial probes [bib_ref] Rapid and Sensitive Detection of SARS-CoV-2 Using Clustered Regularly Interspaced Short Palindromic..., Tsou [/bib_ref]. Poly-U of 5 to 7 nt length have been reported for Cas13a [bib_ref] Development and evaluation of a rapid CRISPR-based diagnostic for COVID-19, Hou [/bib_ref] [bib_ref] Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2, Arizti-Sanz [/bib_ref] and CUCUCU for Cas10 [bib_ref] Intrinsic signal amplification by type III CRISPR-Cas systems provides a sequence-specific SARS-CoV-2..., Santiago-Frangos [/bib_ref].
Depending on the read-out, the probe is conjugated to diverse chemical groups. When signal read-out is based on fluorescent detection, the probe is conjugated to a fluorophore and a compatible quencher. The most used fluorophore is 5 6-FAM (λ Abs = 495 nm, λ Em = 520 nm), followed by HEX (λ Abs = 533 nm, λ Em = 559 nm) and Alex647N (λ Abs = 650 nm, λ Em = 665 nm). The most used quenchers are Iowa Black FQ ® (IDT) and Black Hole Quencher dye. For lateral flow detection with commercial strips, HybriDetect from Milenia Biotech©, the probe invariably contains FAM and biotin attached to the ends. The probe is used most frequently at a concentration of 500 nM (in 17% of the methods), followed by 1000 nM (12%) and 125 nM (10%). Although the probe is used at the same concentration in most reports that use both fluorescence and LF read-out, sometimes it is used at a higher concentration for LF assays.
## Portable, lyophilized, and one-pot versions
Portability is the ability of the method to execute the analysis outside a central lab. This means that a portable method allows decentralized analysis and can be carried out in a POC. Due to the interest to develop methods suitable for POC, most have developed versions that allow portability (~75%) (Supplementary [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. Around 47% use fluorescence (and portable transilluminators or instruments) and 53% LF assays. We have identified that~10% of the analyzed methods used all or some lyophilized component [bib_ref] Digital CRISPR/Cas-Assisted Assay for Rapid and Sensitive Detection of SARS-CoV-2, Park [/bib_ref] [bib_ref] Autonomous lab-on-paper for multiplexed, CRISPR-based diagnostics of SARS-CoV-2, Yin [/bib_ref]. CRISPRbased lyophilized kits would play a very important role in allowing full decentralization and portability since they do not require maintaining a cold chain. Instead,~24% of the methods report being one-pot, meaning that they combine both steps of isothermal amplification and the CRISPR-Cas detection in one. This has attracted a large interest because it can avoid cross-contamination by reducing vessel opening and pipetting and also reducing the complexity and duration of the whole process.
## Experimental outputs to compare between crispr-based methods
Delivering a correct result rapidly (either positive or negative presence of the SARS-CoV-2 virus) is the aim of any CRISPR-Dx method. Under this consideration, the critical experimental outputs to evaluate the methods are: (1) the overall time to deliver the result, (2) limit of detection (LoD), also called analytical sensitivity, and (3) clinical sensitivity and specificity.
## Total time to deliver a result
Rapid delivery of a diagnostic result makes it possible to act quickly to safeguard the life of the patient and limit the spread of the virus, in case of a positive result. From the compiled methods, we analyzed the total time taken to carry out both target amplification and detection by CRISPR-Cas (steps 3 and 4) (Supplementary [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. We excluded time for sample collection (step 1), RNA preparation (step 2), and read-out (step 5) since their duration time is frequently not reported, however, a typical RNA preparation with a kit for RNA extraction would add an average 20 min to the total time, although some are as short as 5 min, whereas a release method would add on average 15 min (see [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref].
The total time of steps 3 and 4 takes between 15 and 90 min, yet, the most frequently reported time is 45 min, followed by 40 and 60 min. When we break out by step, we find that isothermal amplification (step 3) takes from 15 to 40 min, while detection with CRISPR-Cas (step 4) takes from 2 to 30 min, with 30 min the most recurrent time in both steps. The time consumed is not the same for both isothermal methods. RT-LAMP is used from 20 to 40 min, while RT-RPA from 15 to 30 min. This indicates that step 3 can consume more time than step 4, and RT-LAMP usually takes longer than RT-RPA.
The top methods with the shortest time to carry steps 3 and 4 are listed in [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. We observed they have a duration total time shorter (15-36 min) than the most frequent time (45 min) when all the methods are considered. We noticed, especially at the very top, a few methods which have drastically reduced the time by combining both steps in one (one-pot methods). Likewise, the top two-step methods seem to have achieved a drastic reduction in time in any or both steps. Isothermal amplification and CRISPR-Cas detection last less than the commonly used time of 30 min. For example, some methods run step 3 for 20 min, and step 4 for 1 or 5 min. Furthermore, it seems that incorporation of microfluidic systems, heavy optimization of both amplification and detection steps, and integration of both steps 3 and 4 into a one-pot method, help to shorten the total time by up to more than 50%.
One-pot methods that use RT-RPA can be combined with mesophilic CRISPR-Cas12/13 systems because of their temperature compatibility at 37 - C. Instead, for RT-LAMP, the thermophilic AapCas12b is used, since it is stable at 65 - C. All found one-pot methods are listed in [fig_ref] Table 3: List of one-pot CRISPR-based methods [/fig_ref].
## Limit of detection (lod)
The LoD represents the analytical sensitivity of the method, and it is a very critical parameter that indicates the lowest concentration of the target sequence that can be detected by the method. The LoD of all analyzed methods spans from 2.5 to 5000 viral copies per reaction (vc/rx) or 0.25 to 200 vc/µL and shows the wide diversity of possible CRISPR-based methods [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. In particular, it is related to variations in methodological parameters such as targeted gene, type, and duration of CRISPR-Cas and isothermal steps, concentrations of the components, the use of Mg +2 or Mn +2 , incubation time of lateral flow strips to develop the color bands, and use of dual or modified gRNAs, etc.
For all analyzed methods, their average LoD is 261 vc/rx or 12.9 vc/µL (equivalent to~21 aM) when the volume of the reaction is considered. The median LoD is 50 vc/rx (2.5 vc/µL,~4.1 aM) and the most frequently reported values are 200, 100, and 10 vc/rx. All these values are within LoDs experimentally determined for RT-qPCR [bib_ref] Comparison of Severe Acute Respiratory Syndrome Coronavirus 2 Screening Using Reverse Transcriptase-Quantitative..., Rauch [/bib_ref]. Some methods such as CALIBURN [bib_ref] Development of a Broadly Applicable Cas12a-Linked Beam Unlocking Reaction for Sensitive and..., Wang [/bib_ref] that used saliva or sputum samples report a low LoD of 2.5 vc/rx (0.25 vc/µL). In general, methods that used releasing RNA procedures presented higher median LoD (5 vc/rx) than extraction-based methods (3 vc/rx). However, VaNGuard and deCOVID methods showed good results (2 and 1 vc/µL, respectively) [bib_ref] Digital CRISPR/Cas-Assisted Assay for Rapid and Sensitive Detection of SARS-CoV-2, Park [/bib_ref] [bib_ref] An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid..., Ooi [/bib_ref]. When we consider the type of isothermal method used, RT-LAMP has a median LoD of 2 vc/µL, whereas for RT-RPA it is 2.5 vc/µL. One-pot methods have an average and median LoD of 80.3 and 33 vc/rx, while for two-step methods they are 317 and 100 vc/rx, respectively. Cas12-based methods have a lower median LoD (20 vc/rx or 2 vc/µL) than Cas13a (200 vc/rx or 10 vc/µL). Interestingly, top methods classified by LoD (listed in [fig_ref] Table 4: List of CRISPR-based methods with the lowest limit of detection [/fig_ref] have a LoD < 10 vc/rx, which is equivalent to <1 vc/µL when considering the total volume used in each reaction, and represents a sub-attomolar concentration (<10 −18 M).
According to data compiled by Soroka and collaborators, the average LoD value for RT-LAMP alone for detection of SARS-CoV-2 is 17.5 vc/µL while for CRISPR-based methods it is 12.9 vc/µL [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. This indicates that analytical sensitivity is improved when isothermal methods and CRISPR-Cas systems are combined.
## Clinical sensitivity and specificity
Sensitivity tells us about the ability of the method to differentiate true positives from false positives, while specificity differentiates true negatives from false negatives. In particular, sensitivity is the probability of detecting true positives (sick individuals), and specificity is about detecting true negatives (healthy). It is important to keep in mind that the values of sensitivity and specificity are directly related to the degree of sickness of the patient, the type of RNA extraction method, and the type of isothermal method and Cas, among other things. Analyzed methods have sensitivities ranging from~73 to 100%, with a most frequent value of 100% and a mean value of 95% [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref]. Specificity, on the other hand, has values that span from~71% to 100%, with a most frequent value of 100% and a mean value of 97.5%. There is no difference in sensitivity or specificity when RT-LAMP (95.5% and 100%, respectively) or RT-RPA (96.5% and 100%, respectively) is used. Moreover, Cas12 has the same clinical specificity (100% median) and slightly higher clinical sensitivity (100%) than Cas13 (100 and 95.75%, respectively).
The data suggest that CRISPR-Cas methods are clinically reliable to detect SARS-CoV-2 because they have similar results to the results delivered by RT-qPCR (specificity and sensitivity > 95% or sometimes even 100%) (see [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref] for specific numbers for each method). Indeed, top methods report 100% specificity and sensitivity [fig_ref] Table 5: List of CRISPR-based methods with highest clinical specificity and sensitivity [/fig_ref]. However, clinical validation generally is conducted in academic laboratories on very small cohorts (ranging from 4 to 534 patients, with a most frequent value of 25 samples) (Supplementary [fig_ref] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and... [/fig_ref] and with samples previously known as positive or negative. Thus, sensitivity and specificity need to be confirmed in larger cohorts in clinical settings and with blinded samples. However, some attempts to validate CRISPR-Cas methods in clinical settings using RT-qPCR tests as reference have been carried out with acceptable results (specificity and sensitivity > 95%) [bib_ref] Clinical validation of a Cas13-based assay for the detection of SARS-CoV-2 RNA, Patchsung [/bib_ref] [bib_ref] Comparison of Severe Acute Respiratory Syndrome Coronavirus 2 Screening Using Reverse Transcriptase-Quantitative..., Rauch [/bib_ref] [bib_ref] Comparative performance of CRISPR-Cas12a assays for SARS-CoV-2 detection tested with RNA extracted..., Nimsamer [/bib_ref].
When compared against an isothermal method such as RT-LAMP, the sensitivity and specificity for detecting SARS-CoV-2 are 92-96% and 96-99%, respectively, according to several systematic reviews and meta-analyses [bib_ref] The screening value of RT-LAMP and RT-PCR in the diagnosis of COVID-19:..., Pu [/bib_ref] [bib_ref] Reverse Transcriptase Loop Mediated Isothermal Amplification (RT-LAMP) for COVID-19 diagnosis: A systematic..., Subali [/bib_ref] , which do not differ from the values acquired by CRISPR-diagnostics. However, many RT-LAMP methods have used RT-PCR as a reference, which may exaggerate its performance [bib_ref] The screening value of RT-LAMP and RT-PCR in the diagnosis of COVID-19:..., Pu [/bib_ref].
## Cost and manufacturing
Not many details about the costs of the methods have been reported. We found just six methods that reported a cost estimation [bib_ref] Vigilant: An Engineered VirD2-Cas9 Complex for Lateral Flow Assay-Based Detection of SARS-CoV2, Marsic [/bib_ref] [bib_ref] Ultrasensitive and visual detection of SARS-CoV-2 using all-in-one dual CRISPR-Cas12a assay, Ding [/bib_ref] [bib_ref] An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid..., Ooi [/bib_ref] [bib_ref] A one-step, one-pot CRISPR nucleic acid detection platform (CRISPR-top): Application for the..., Li [/bib_ref] [bib_ref] Detection of SARS-CoV-2 with SHERLOCK One-Pot Testing, Joung [/bib_ref]. The costs range from 1 to 10 USD per reaction (usually only including the cost of the proteins, gRNAs, and other materials). The costs of minor equipment such as micropipettes, thermoblocks, freezers, and lyophilizers (when needed) have to be included in the total cost of implementation and in the costs involved in the regulatory evaluation. Some have reported the preliminary cost for essential apparatus such as water baths (40 USD or less) [bib_ref] Detection of SARS-CoV-2 with SHERLOCK One-Pot Testing, Joung [/bib_ref].
In general, the production of the components is scalable and suitable for mass production; however, since the CRISPR-based methods are quite novel, systematic economic and feasibility analyses must be reported. One main limitation is the elaboration of cheap lateral flow strips. Availability of cheaper options could help to reduce costs significantly, especially if paper-based microfluidics suitable for mass production is involved.
# Conclusions
The onset of COVID-19 spurred intensive research to developing novel diagnostic methods based on CRISPR-Cas systems to detect the SARS-CoV-2 virus. The particular features of CRISPR-Cas systems together with the synergistic work with isothermal amplification such as RT-RPA or RT-LAMP offer many advantages which have found a niche of application with the current fast propagation and global distribution of the COVID-19 pandemic. Scientists have used CRISPR-Cas because it can quickly (in a few days) be adapted with reliability [bib_ref] CRISPR-Cas12-based detection of SARS-CoV-2, Broughton [/bib_ref].
Novel diagnostic methods of SARS-CoV-2 based on CRISPR-Cas and isothermal amplification stand out over other methods because they can clearly compete with the gold-standard RT-qPCR in some features and surpass it in others. In terms of analytical sensitivity, clinical specificity, and sensitivity, the most successful CRISPR-Cas methods can detect down to 2.5 viral copies per reaction (sub-attomolar concentration) in clinical samples with 100% specificity and sensitivity. Furthermore, the methods can deliver a result as soon as in 30-40 min (when considering the RNA preparation step) using a variety of samples such as naso-, oro-pharyngeal, and anal swabs, as well as sputum, stool, and others, in one-pot and decentralized formats. Portable kits based on fluorescence or lateral flow devices have also achieved consistent high standards when critical experimental outputs such as total time, LoD, specificity, and sensitivity are compared [fig_ref] Table 6: Top CRISPR-Cas based methods to detect SARS-CoV-2 [/fig_ref] shows the top-ranked methods that score the highest globally). All this clearly positions CRISPR-Cas and isothermal amplification as an alternative to RT-qPCR, particularly for decentralized detections with shorter waiting times.
Despite the rapid advances, detection of SARS-CoV-2 with CRISPR-Cas still requires much work before mass clinical application. It needs further optimization to develop robust methods and be tested against blinded samples where it is not known a priori if they are positive or negative. Monitoring experimental outputs such as total time, limit of detection, sensitivity, and specificity would be very helpful for guidance. However, more research needs to be conducted to find which experimental parameters are key and how they affect and correlate with experimental outputs. On the other hand, there is plenty of room for innovation. For example, we can foresee the engineering of the predominant Cas12a and Cas13 and the addition of other CRISPR-Cas systems that also present collateral activity, but also the adaptation of Cas systems not based on collateral activity. Furthermore, it would be beneficial to develop multiplex methods to detect several SARS-CoV-2 genes/mutations (related to variants) [bib_ref] CRISPR-Cas12a-Based Detection for the Major SARS-CoV-2 Variants of, Yuanhao [/bib_ref] and other viruses that cause COVID-19-like symptoms (e.g., influenza) in parallel. Furthermore, the incorporation of new formats that include microfluidic systems will help miniaturizing the methods and making them more accessible, cheaper, and more efficient. An important challenge is to develop quantitative methods that continue to be simple and portable. All this will continue making CRISPR-Cas methods very attractive in the current COVID-19 pandemic, and importantly, the lessons learned here will offer advantages for coming pandemics.
[fig] Figure 1: (A) Structural and genome features of SARS-CoV-2 virion. Class 2 CRISPR-Cas proteins extensively used in genetic diagnostics: (B) LbCas12a (type V, PDB: 5XUS) [/fig]
[fig] Figure 2: General workflow to detect SARS-CoV-2 with CRISPR-based test includes five general steps: (1) Clinical sample collection, (2) RNA preparation by extraction or release methods, (3) target sequence amplification, (4) target recognition and generation of molecular signal, and (5) signal read-out using fluorescence or lateral flow strips which could include cell phone detection. [/fig]
[fig] 65: • C (RT-LAMP) and 37-42 • C (RT-RPA) Time 20 to 40 min (RT-LAMP) and 15 to 30 min (RT-RPA) [/fig]
[table] Table 2: List of CRISPR-based methods with shortest total time for steps 3 and 4. [/table]
[table] Table 3: List of one-pot CRISPR-based methods. Fluorescence, FA: Fluorescence Anisotropy, C: Colorimetric, LF: Lateral Flow. * Portable methods demonstrated the use of the method with low-complexity equipment able to work in decentralized settings. [/table]
[table] Table 4: List of CRISPR-based methods with the lowest limit of detection (LoD). [/table]
[table] Table 5: List of CRISPR-based methods with highest clinical specificity and sensitivity. [/table]
[table] Table 6: Top CRISPR-Cas based methods to detect SARS-CoV-2. NP: Nasopharyngeal swabs, BALF: Bronchoalveolar lavage fluid, OP: Oropharyngeal swabs, Surgery: pre-operative samples from patients undergoing surgery. C: Clinical swabs not specified. NA: Not Applicable. R: Release methods, E: Extraction method, NR: not reported, RT: Room temperature, F: Fluorescence, FA: Fluorescence Anisotropy, LF: Lateral Flow. * LoD: Limit of Detection. * We use the name of the article because the method does not have a name. ** Add 20 or 15 min extra considering the RNA extraction or release, respectively. [/table]
|
A Case-Control Study of Involvement of Oxidative DNA Damage and Alteration of Antioxidant Defense System in Patients with Basal Cell Carcinoma: Modulation by Tumor Removal
Oxidative damage has been suggested to play a role in the pathogenesis of basal cell carcinoma (BCC). This study illustrated an involvement of oxidative DNA damage and changes in antioxidant defenses in BCC by conducting a case-control study (24 controls and 24 BCC patients) and assessing urinary 7,8-dihydro-8-oxo-2 -deoxyguanosine (8-oxo-dGuo), plasma antioxidant defenses including catalase (CAT), glutathione peroxidase (GPx), NQO1, and total superoxide dismutase (SOD) activities, and glutathione (GSH) levels before surgery and 1 month after surgery. 8-oxo-dGuo expressions as well as protein and mRNA expressions of DNA repair enzyme hOGG1 and antioxidant defenses (CAT, GCLC, GPx, Nrf2, and MnSOD) in nonneoplastic epidermis of control and BCC tissues were also determined. This study observed induction in urinary 8-oxo-dGuo, increased 8-oxo-dGuo expression, and reduced hOGG1 protein and mRNA in BCC tissues, decreased activities of CAT, GPx, and NQO1, but elevated SOD activities and GSH levels in BCC patients and reduction of all antioxidant proteins and genes studied in BCC tissues. Furthermore, decreased plasma antioxidant activities in BCC patients were restored at 1 month after operation compared with preoperative levels. Herein, we concluded that BCC patients were associated with oxidative DNA damage and depletion of antioxidant defenses and surgical removal of BCC correlated with improved redox status.
# Introduction
Basal cell carcinoma (BCC) is the most common nonmelanoma skin cancer (NMSC) worldwide, in particular, in fair-skinned population and its incidence has been rising over the past several years [bib_ref] Incidence of basal cell and squamous cell carcinomas in a population younger..., Christenson [/bib_ref] [bib_ref] Trends in basal cell carcinoma incidence rates: a 37-year dutch observational study, Flohil [/bib_ref] [bib_ref] A systematic review of worldwide incidence of nonmelanoma skin cancer, Lomas [/bib_ref]. Although BCC can be generally diagnosed and treated by surgical excision, it can be destructive and has a significant impact on patients' quality of life as the tumor removal can cause morbidity including functional disability and disfigurement. Therefore, it is of significance to understand pathogenesis of BCC and identify candidate biomarkers, which may be clinically translated into strategies for its detection and prevention. Photo-damage by chronic exposure to ultraviolet radiation (UVR) has been suggested to play a role in the pathogenesis of BCC, which originates from keratinocytes of the epidermal basal layer [bib_ref] Patterns and timing of sunlight exposure and risk of basal cell and..., Iannacone [/bib_ref] [bib_ref] Incidence of nonmelanoma skin cancer in relation to ambient UV radiation in..., Xiang [/bib_ref].
Excessive reactive oxygen species (ROS) generated by UVR have been shown to contribute to malignant transformation of keratinocytes into cancerous cells including BCC probably through oxidative DNA damage, defects in DNA repair, and interference with cellular signaling [bib_ref] Oxidative stress in malignant melanoma and non-melanoma skin cancer, Sander [/bib_ref] [bib_ref] Molecular aetiology and pathogenesis of basal cell carcinoma, Tilli [/bib_ref]. The oxidative modification of guanine leads to formation of 8-oxo-7,8-dihydro-2 -deoxyguanosine (8-oxo-dGuo), which 2 Oxidative Medicine and Cellular Longevity finally can pair with adenine and cytosine during DNA replication, resulting in GC-TA mutations found to be associated with development of BCC [bib_ref] Increased expression of versican in the inflammatory response to UVB-and reactive oxygen..., Kunisada [/bib_ref]. 8-oxo-dGuo has also been recognized to be the most abundant and potentially mutagenic if not substantially repaired and has thus been developed as a sensitive and stable biomarker for evaluating the degree of oxidative DNA damage. Case-control studies have previously reported that increased urinary 8oxo-7,8-dihydroguanine (8-oxo-Gua) and 8-oxo-dGuo levels were detected in patients with metastatic head and neck cancer, breast cancer, and lung cancer [bib_ref] The level of 8-hydroxyguanine, a possible repair product of oxidative DNA damage,..., Rozalski [/bib_ref] [bib_ref] Urinary 8-hydroxy-2 -deoxyguanosine (8-OHdG) and genetic polymorphisms in breast cancer patients, Kuo [/bib_ref]. Furthermore, impaired DNA repair capacity was suggested to associate with enhanced susceptibility to cancer and deficiency in DNA repair enzyme, human 8-oxoguanine DNA N-glycosylase 1(hOGG1), a key enzyme responsible for 8-oxo-dGua repair, may also be involved in carcinogenesis [bib_ref] The OGG1 gene encodes a repair enzyme for oxidatively damaged DNA and..., Shinmura [/bib_ref]. It has been proposed that promotion of antioxidant defenses including catalase (CAT), glutathione peroxidase (GPx), NAD(P)H: quinone oxidoreductase 1 (NQO1), a crucial detoxifying enzyme, and superoxide dismutase (SOD) as well as redox regulation by nuclear factor-erythroid-2-related factor 2 (Nrf2) that can cope with excessive ROS formation in human body may be useful to prevent ROS-mediated neoplastic transformation of various tissues including the skin [bib_ref] Redox regulation in cancer: a double-edged sword with therapeutic potential, Acharya [/bib_ref] [bib_ref] Oxidative stress in the pathogenesis of skin disease, Bickers [/bib_ref]. A link between increased oxidative DNA damage/decreased antioxidant defense capacity and cancer has been proposed and oxidative/antioxidant defense status has thus been intensively investigated as promising biomarkers in cancer [bib_ref] Use of conventional and -omics based methods for health claims of dietary..., Knasmüller [/bib_ref] [bib_ref] Advances in carcinogenic metal toxicity and potential molecular markers, Koedrith [/bib_ref] [bib_ref] Urinary excretion of 8-oxo-7,8-dihydroguanine as biomarker of oxidative damage to DNA, Loft [/bib_ref]. However, relationship between the oxidative DNA damage/antioxidant defense parameters and BCC and whether surgical removal of tumors affects the redox status in BCC patients have not yet been reported. Here, we conducted a case-control study to investigate whether BCC was associated with the oxidative DNA damage, hOGG1 levels, and antioxidant defense status and whether tumor removal affected the redox status of patients with BCC compared to control subjects with nonmalignant skin diseases. (3) (3 females), normal skin (4) (4 females), and seborrheic keratosis (6) (4 males and 2 females).
# Methods
## Study
## Data
Collection. Data were collected on demographics, clinical characteristics, and lifestyle as shown in [fig_ref] Table 1: Demographics and clinical characteristics of controls subjects and BCC patients [/fig_ref]. Subjects were excluded from the study if they had abnormal clinical characteristics for complete blood count (CBC), blood urea nitrogen (BUN), creatinine (Cr), fasting blood sugar (FBS), cholesterol, high density lipoprotein (HDL), low density lipoprotein (LDL), triglyceride, aspartate transaminase (AST), and alanine transaminase (ALT), had history of chronic diseases such as cancer, metabolic diseases including uncontrolled diabetes mellitus, hypertension, and hyperlipidemia, were under concurrent usage of steroid drugs, had body mass index (BMI) > 30 kg/m 2 , and had history of smoking or drinking alcohol on a regular basis.
All subjects underwent surgical intervention and the collection of blood, urine, and skin tissue samples was done on surgery day prior to operation. At 1 month after surgery, blood samples were collected for evaluation of antioxidant defense parameters and the urinary 8-oxo-dGuo levels of BCC patients were determined at both 1 and 6 months after surgery.
## Blood samples
Processing. Blood samples were collected: 2 mL in sodium fluoride tube, 2 mL in EDTA tube, and 6 mL in lithium heparin tube for clinical chemistry assays and 10 mL in EDTA tube for antioxidant assays. Whole blood was processed according to standard protocols and centrifuged at 1,000 ×g for 10 minutes at 4 ∘ C, and the supernatants were divided into 750 L aliquots. Plasma samples were stored at −80 ∘ C until testing.
## Urine samples
Processing. 20 mL of fresh urine samples was collected in urine container tube: 10 mL for clinical chemistry assays and 10 mL for oxidative DNA damage assay. Urine samples were processed according to standard protocols and centrifuged at 3,000 ×g for 10 min at 4 ∘ C, and the supernatants were collected and stored at −80 ∘ C until testing.
## Tissue samples
Processing. The skin tissue samples were obtained from lesions of BCC and nonmalignant skin diseases. All studied skin lesions in case and control patients were located on the sun-exposed areas and the size of all excised tissues was 0.5 cm in diameter. Tissue samples were divided into 2 parts; the first part was fixed in formalin solution and then embedded in paraffin block for immunohistochemistry and the second part was fixed in liquid nitrogen for RT-PCR until testing.
## Determination of 8-oxo-dguo levels in urine by elisa.
Urinary 8-oxo-dGuo level was quantified using a competitive enzyme immunoassay (STA-320, Cell Biolabs, San Diego, CA). Briefly, urine samples (1 : 20 dilution) or 8-oxo-dGuo standards were first added to an 8-oxo-dGuo/BSA conjugate preabsorbed enzyme immunoassay plate. After incubation, an anti-8-oxo-dGuo monoclonal antibody was added, followed by a secondary reaction with a horseradish peroxidaseconjugated antibody. 8-oxo-dGuo levels in the urine samples were then determined by comparison with the 8-oxo-dGuo Oxidative Medicine and Cellular Longevity 3 standard curve. 8-oxo-dGuo levels in the urine of each subject were adjusted by urinary creatinine level and were measured as ng/mg creatinine.
## Determination of antioxidant defense status in plasma.
Catalase (CAT) activity was determined following the kit protocol from Cayman Chemical (Ann Arbor, MI). The assay was based on the reaction of methanol and the enzyme in the presence of an optimal concentration of H 2 O 2 . The formaldehyde produced was measured colorimetrically at 540 nm using 4-amino-3-hydrazino-5-mercapto-1,2,4-triazole (Purpald) as the chromogen. CAT activity was expressed in unit/mg protein.
Glutathione peroxidase (GPx) activity assay was performed following manufacturer's instruction (Trevigen, Gaithersburg, MD). GPx activity was coupled to glutathione reductase (GR), which catalyzed NADPH-mediated reduction of GSSG back to GSH as previously described [bib_ref] Caffeic acid and ferulic acid inhibit UVA-induced matrix metalloproteinase-1 through regulation of..., Pluemsamran [/bib_ref]. The rate of NADPH oxidation by H 2 O 2 was monitored at 340 nm and GPx activity was expressed in unit/mg protein.
NQO1 activity was evaluated spectrophotometrically as previously described [bib_ref] UNIT 4.22 Biochemical and genetic analysis of NAD(P)H:quinone oxidoreductase 1 (NQO1), Siegel [/bib_ref] using 2,6-dichloroindophenol (DCPIP) as a substrate. The assay was based on the activities for NAD(P)H-dependent reduction of DCPIP at 600 nm and the reaction was specifically inhibited by dicumarol. The NQO1 activity was thus measured as the dicumarolinhibitable reduction in absorbance at 600 nm and was expressed as nmole DCPIP reduced/min/mg protein.
Assay for measurement of total superoxide dismutase (SOD) activity was modified following the method of Johns et al. [bib_ref] Impact of elevated dietary sodium intake on NAD(P)H oxidase and SOD in..., Johns [/bib_ref] [bib_ref] Superoxide dismutase as an anaerobic polypeptide: a key factor in recovery from..., Monk [/bib_ref] and the kit protocol from Cayman Chemical (Ann Arbor, MI). Superoxide anions (O 2
- − ) were generated by a xanthine/xanthine oxidase (XOD) system and were detected using 3 -{1-[(phenylamino)-carbonyl]-3,4-tetrazolium}-bis(4-methoxy-6-nitro) benzenesulfonic acid (XTT) as the chromogen. The total SOD activity was determined by a decrease in XTT reaction rate at 470 nm as a result of superoxide produced by xanthine/XOD system. The SOD activity was calculated using the following equation:
[formula] [{(0.953) × ((Std.conc. 0 − Blank)/(Sample − Blank)) − 0.097}× Dilution factor]. [/formula]
Total SOD activity was expressed in unit/mg protein, where one unit of activity was the amount of protein required for a 50% decrease in the rate of XTT reduction.
GSH assay was carried out by glutathione reductase: DTNB enzymatic recycling method following the kit protocol from Sigma-Aldrich (MO, US). Determination of GSH levels involves GSH oxidation by the sulfhydryl reagent DTNB (5,5 -dithio-bis-2-(nitrobenzoic acid)) to produce the yellow TNB (5 -thio-2-nitrobenzoic acid) measured at 412 nm. The glutathione disulfide (GSSG) formed can be recycled to GSH by glutathione reductase in the presence of NADPH. The rate of TNB production is directly proportional to this recycling reaction in turn directly proportional to the concentration of GSH. The GSH level was expressed in nmol/mg protein.
## Determination of protein content in plasma by
Bradford Assay. Protein concentration was measured using the Bio-Rad Protein Assay Kit (Bio-Rad, Munich, Germany) and bovine serum albumin (BSA) was used as protein standard.
## Immunohistochemical (ihc) determination of oxidative dna damage, antioxidant enzyme, and hogg1 expression in skin
Tissues. Paraffin-embedded tissues were sectioned (2 m thickness) and placed on Super-FrostPlus glass slides fixed at 60 ∘ C overnight. The sections were deparaffinized in xylene and rehydrated in ethanol series and then incubated in citrate buffer, pH 6 (DaKo). Slides were incubated with 3% hydrogen peroxide and then with 2% BSA. Slides were incubated with primary antibodies against 8-oxo-dGuo (ab48508; Abcam, Cambridge, UK) (1 : 50), hOGG1 (NB100-106, Novus IHC staining of all samples was evaluated visually and scored by a pathologist twice in different days. If the IHC staining evaluated in duplicate gave different IHC scores, visual interpretation of the IHC staining would be repeated. The semiquantitative analysis of the stained sections was carried out by light-microscopy, employing the immunoreactive score (IRS) according to the study of Kaemmerer et al. [bib_ref] Comparing of IRS and Her2 as immunohistochemical scoring schemes in gastroenteropancreatic neuroendocrine..., Kaemmerer [/bib_ref]. The level of antibody staining was evaluated by IRS and calculated by multiplying the scores of staining intensity by the percentage of positive cells. Based on the IRS, antibody staining pattern was defined as IRS-classification score.
## Quantitative real-time reverse transcription-polymerase chain
Reaction: Determination of hOGG1, Catalase, GCLC, GCLM, GPx, Nrf2, CuSOD, and MnSOD. Improm-IITM reverse transcriptase (Promega, Madison, USA) was used to synthesize cDNA from total RNA following the manufacturer's protocol. Sequences for PCR primer sets of genes studied were designed using the Primer Express version 3.0 software (Applied Biosystems, USA). Sequences of PCR primer (in 5 → 3 direction) were as follows: hOGG1 (product sizes = 164 bp) sense, TGGAAGAACAGGGCG-GGCTA, and antisense, ATGGACATCCACGGGCACAG; CAT (product sizes = 148 bp) sense, CCTTCGACCCAA-GCAACATG, and antisense, CGAGCACGGTAGGGACAG-TTC; GCLC (product sizes = 160 bp), sense GCTGTCTTG-CAGGGAATGTT, and antisense, ACACACCTTCCTTCC-CATTG; GCLM (product sizes = 200 bp) sense, TTGGAG-TTGCACAGCTGGATT, and antisense, TGGTTTTAC-CTGTGCCCACTG; GPx (product sizes = 94 bp) sense, ACGATGTTGCCTGGAACTTT, and antisense, TCGATG-TCAATGGTCTGGAA; Nrf2 (product sizes = 161 bp) sense, TTCTGTTGCTCAGGTAGCCCCTCA, and antisense, GTTTGGCTTCTGGACTTGG; CuSOD (product sizes = 109 bp) sense, TGCTGGTTTGCGTCGTAGTC, and antisense, ACGCACACGGCCTTCGT; MnSOD (product sizes = 141 bp) sense, TGGCCAAGGGAGATGTTACAG, and antisense, CTTCCAGCAACTCCCCTTTG; GAPDH (product sizes = 150 bp) sense, CCTCCAAAATCAAGT-GGGGCGATG, antisense, CGAACATGGGGGCATCAG-CAGA. Real-time PCR was performed using FastStart universal SYBR Green Master with ROX (Roche diagnostic, USA) and mRNA expression was quantified by real-time PCR using ABI prism 7300 Real-Time PCR System (Applied Biosystems, USA). Melt curve analysis was performed to verify specificity of the amplified product. mRNA expression was normalized to the expression of GAPDH gene. The mean Ct of each gene in each sample was compared with the mean Ct from GAPDH determinations from the same cDNA sample in order to assess mRNA expression. Ct values were then used to calculate fold change in gene expression.
# Statistical analysis.
Descriptive statistics were reported as frequencies and percentage. Categorical variables were analyzed by the chi-square test and performed with SPSS 18.0 software (SPSS, Chicago, IL). Data were expressed as means ± standard deviation (SD). The results were subjected to statistical analysis using Prism 5.0 (GraphPad Software, La Jolla, CA, USA). The Shapiro-Wilk test was employed to test the normal distribution. The statistical significance between nonparametric variables was analyzed by Mann Whitney test or Kruskal-Wallis test and between parametric variables by unpaired Student's -test or one-way analysis of variance (ANOVA) followed by Tukey's post hoc test. values less than 0.05 were considered statistically significant. To investigate systematically similarity of biochemical profiles, we performed the principal component analysis using JMP Pro. Data were then plotted using custom MATLAB code. [fig_ref] Table 1: Demographics and clinical characteristics of controls subjects and BCC patients [/fig_ref] indicate that the demographic and clinical characteristics are similar.
# Results
## Demographic, lifestyle, and clinical characteristics of the study subjects. the results shown in
## The urinary 8-oxo-dguo levels and antioxidant defense
Status in BCC and Control Subjects. The urinary 8-oxo-dGuo levels and all antioxidant defense parameters studied in BCC and control subjects were shown in [fig_ref] Table 2: Comparison of urinary oxidative DNA damage levels and plasma antioxidant defense status... [/fig_ref]. Oxidative DNA damage was significantly higher in BCC patients compared to control subjects [fig_ref] Figure 1: The urinary 8-oxo-dGuo levels [/fig_ref]. After surgery, urinary 8-oxo-dGuo levels of BCC patients were insignificantly altered at 1 month, although they were substantially reduced at 6 months compared to preoperative levels [fig_ref] Figure 1: The urinary 8-oxo-dGuo levels [/fig_ref]. In addition, while preoperative values of plasma CAT, GPx, and NQO1 activities were observed to be lower in BCC patients compared to control [fig_ref] Figure 1: The urinary 8-oxo-dGuo levels [/fig_ref] , plasma total SOD activities and GSH levels were substantially higher in BCC patients compared to control [fig_ref] Figure 1: The urinary 8-oxo-dGuo levels [/fig_ref]. In comparison between preoperative and postoperative levels of plasma antioxidant defense status in BCC patients, CAT, GPx, Oxidative Medicine and Cellular Longevity 5 Results were expressed as mean ± standard deviation (SD). * * * < 0.001 compared to control, before surgery. ### < 0.001 compared to case, before surgery. and NQO1 activities were increased at 1 month postoperatively, although total SOD activities and GSH levels remained unchanged.
In control subjects, urinary 8-oxo-dGuo levels and all antioxidant defense parameters studied were not significantly different before and after surgery.
To compare systematically the DNA damage and antioxidant defense parameters among different treatment conditions, we utilized the principal component analysis. In the PCA space [fig_ref] Figure 1: The urinary 8-oxo-dGuo levels [/fig_ref] , data from all patients will be projected into a 2-dimensional plane that can best demonstrate variation among different treatment groups. Specifically, the first and the second components are linearly combined scores from all measurement parameters that can best explain variances observed from all patients, 38.3% and 18.1%, respectively. Based on this calculation, we clearly observed similarity of PCA scores between the two control groups, either before surgery or at 1 month after surgery. On the other hand, patients with BCC clearly exhibit changes of PCA scores after surgery. Interestingly, half of the BCC patients at 1 month after surgery already exhibit PCA profiles that are similar to those of the control groups while the other half still exhibit PCA scores that do not overlap with those from the control groups. This finding implies the potential heterogeneity of antioxidant defense response among patients but shows that antioxidant defense status in BCC patients can be rescued to the level similar to that of the control group patients as early as 1 month after surgery.
## Ihc determination of oxidative dna damage, antioxidant enzyme, and hogg1 expressions in skin tissues of bcc and control subjects.
The IHC staining was classified as negative, weak positive, mild positive, and strong positive and the data were presented as average IRS. H&E staining identified structures of skin sections of case and control subjects as shown in [fig_ref] Figure 2: Continued. [/fig_ref]. [fig_ref] Table 3: Expressions of oxidative DNA damage, DNA repair enzyme, and antioxidant proteins in... [/fig_ref] demonstrated that expressions of nuclear 8-oxo-dGuo [fig_ref] Figure 2: Continued. [/fig_ref] , 2(e), and 2(g)) were higher and hOGG1 proteins [fig_ref] Figure 2: Continued. [/fig_ref] , 2(i), and 2(k)) in nucleus and cytoplasm were lower in BCC tissues than those in the epidermis of control subjects. Additionally, Results were expressed as mean ± standard deviation (SD). * < 0.05, * * < 0.01, * * * < 0.001 compared to epidermis of control. and GSH levels (f)] in control subjects and BCC patients before and after surgery. Values given are mean ± SD. The statistical significance of differences between the control and case was evaluated by one-way ANOVA followed by Tukey's post hoc test. * * * < 0.001 compared to preoperative values in control subjects prior to surgery; ### < 0.001 compared to preoperative values in BCC patients prior to surgery. Principal component analysis was performed to systematically investigate similarity of DNA damage and antioxidant defense parameters from different groups of patients (g). Factor analysis is overlaid on top of the patient scores, both before surgery (case: triangle; control: square) and 1 month after surgery (case: circle; control: plus). The dashed and dotted ovals delineate the approximated distribution of each group with 95% confidence intervals. ) and IHC staining for oxidative DNA damage, 8-oxo-dGuo ((d)-(g)), DNA repair enzyme, hOGG1 ((h)-(k)), and antioxidant proteins, CAT ((l)-(o)), GCLC ((p)-(s)), GPx ((t)-(w)), Nrf2 ((x)-(aa)), and MnSOD ((ab)-(ae)), in control subjects and tumor and nontumor lesions of BCC patients. Values given are mean ± SD. The statistical significance of differences between the control and case and between adjacent epidermis and tumor lesions of BCC patients was evaluated by nonparametric variables with Kruskal-Wallis test followed by Dunnett's post hoc test. * < 0.05, * * < 0.01, * * * < 0.001 compared to control; ## < 0.01, ### < 0.001 compared to tumor lesions of BCC patients.
[formula] (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (l) (m) (n) [/formula]
## Mrna expression of hogg1 and antioxidant defense system in skin tissues of bcc and control subjects.
In agreement with protein expression data, [fig_ref] Figure 3: DNA repair and antioxidant gene expression in nonmalignant skin tissues of control... [/fig_ref] demonstrated that mRNA expressions of hOGG1 (0.85 ± 0.12-fold decrease, < 0.001), CAT (0.88 ± 0.03-fold decrease, < 0.001), GCLC (0.66 ± 0.09-fold decrease, < 0.001), GCLM (0.40 ± 0.17-fold decrease, < 0.001), GPx (0.76 ± 0.07-fold decrease, < 0.001), Nrf2 (0.63 ± 0.02-fold decrease, < 0.001), CuSOD (0.47 ± 0.13-fold decrease, < 0.001), and MnSOD (0.65 ± 0.09-fold decrease, < 0.001) were substantially lower in BCC tissues than in skin tissues of control subjects.
# Discussion
An involvement of oxidative damage in the pathogenesis of BCC has been widely discussed since several studies have shown possible mechanisms through which excessive ROS generation and antioxidant defense impairment may play a role in malignant transformation to NMSC or keratinocytic cancer including BCC [bib_ref] Oxidative stress in malignant melanoma and non-melanoma skin cancer, Sander [/bib_ref] [bib_ref] Interaction between functional polymorphic variants in cytokine genes, established risk factors and..., Rizzato [/bib_ref]. Potential mechanisms of carcinogenesis may involve oxidative DNA damage accountable for genomic mutations because attacks on DNA by ROS result in considerable DNA lesions including strand breaks and DNA base oxidation products, in particular 8oxo-dGuo, considered the most potentially mutagenic [bib_ref] Role of oxidative stress and DNA damage in human carcinogenesis, Kryston [/bib_ref]. hOGG1 is highly specific for the removal and repair of 8oxo-dGuo from oxidatively damaged DNA. Increased 8-oxo-dGuo formation and/or loss of hOGG1's expression and function were reported to play a role in the development and progression of skin cancers including BCC [bib_ref] Human 8-oxoguanine-DNA glycosylase-1 is downregulated in human basal cell carcinoma, Huang [/bib_ref] [bib_ref] 8-Oxoguanine formation induced by chronic UVB exposure makes Ogg1 knockout mice susceptible..., Kunisada [/bib_ref] [bib_ref] Nuclear 8-hydroxy-2 -deoxyguanosine as survival biomarker in patients with cutaneous melanoma, Murtas [/bib_ref]. In this study, we observed elevation of preoperative 8-oxo-dGuo levels in urine and its expression in BCC tissues in comparison with epidermis of control subjects. At 1 month after operation, there were no significant changes in urinary 8-oxo-dGuo levels in BCC patients compared with preoperative levels, although the postoperative 8-oxo-dGuo levels in BCC patients were substantially reduced at 6 months. It is possible that greater urinary 8-oxo-dGuo in BCC patients than in control subjects observed in our study could be attributed to oxidative DNA damage in neoplastic cells of BCC tissues. Changes in system redox status in patients subjected to surgical removal of primary cutaneous tumor were also reported. Gadjeva et al. observed higher malondialdehyde (MDA) levels, lipid peroxidation products as an oxidative stress marker, in plasma of patients with melanoma compared with control subjects and reduction of the MDA levels at 20 days after surgery compared with preoperative values [bib_ref] Influence of therapy on the antioxidant status in patients with melanoma, Gadjeva [/bib_ref].
In addition, urinary 8-oxo-dGuo levels in control subjects in our study were observed to be higher than those of healthy subjects in previous studies employing ELISA [bib_ref] 8-Oxoguanine formation induced by chronic UVB exposure makes Ogg1 knockout mice susceptible..., Muzembo [/bib_ref] [bib_ref] The cost of liver disease in Korea: methodology, data, and evidence, Chung [/bib_ref] , although the levels found in this study were comparable to those reported by Ksiazek et al. [bib_ref] Impaired response to oxidative stress in senescent cells may lead to accumulation..., Ksiazek [/bib_ref]. The urinary 8-oxo-dGuo levels were usually higher and variable when detected by ELISA compared with chromatographic methods. Thus, further studies using chromatographic techniques including chromatographic techniques with either mass spectrometric or electrochemical detection having higher sensitivity and specificity for determination of urinary 8-oxo-dGuo levels need to be done with a larger sample size in order to confirm an association between oxidative DNA damage and BCC.
We also observed decreased protein and mRNA expressions of hOGG1 in BCC tissues in comparison with epidermis of control subjects. Furthermore, the intrasubject comparison of nonneoplastic epidermis adjacent to BCC and lesional BCC skin demonstrated a higher expression of 8-oxo-dGuo and a lower expression of hOGG1 in BCC tissues compared with the adjacent epidermis. Our results are consistent with those from previous case-control studies suggesting a contribution of oxidative DNA damage and impaired hOGG1 to the development of various cancers including breast, pancreatic, gastric, and lung cancers [bib_ref] The level of 8-hydroxyguanine, a possible repair product of oxidative DNA damage,..., Rozalski [/bib_ref] [bib_ref] Interactive effect of cigarette smoking with human 8-oxoguanine DNA N-glycosylase 1 (hOGG1)..., Chang [/bib_ref] [bib_ref] Contribution of environment and genetics to pancreatic cancer susceptibility, Hocevar [/bib_ref] [bib_ref] Relation between gastric cancer and protein oxidation, DNA damage, and lipid peroxidation, Ma [/bib_ref]. Hence, augmented levels of 8-oxo-dGuo could be related to a defect in hOGG1 expression in BCC patients that may be deficient in the repair of 8-oxo-dGuo, although the mechanism by which hOGG1 impairment contributes to the development of BCC needs further investigation.
A disturbance in redox homeostasis probably contributed to development of multiple tumors including NMSC which can be attributed to not only increased oxidative DNA damage but also impaired antioxidant defense capacity [bib_ref] Oxidative stress in malignant melanoma and non-melanoma skin cancer, Sander [/bib_ref] [bib_ref] Redox regulation of cell survival, Trachootham [/bib_ref]. Major antioxidant defenses in the human body include GSH and enzymatic antioxidants including GPx and CAT, which neutralize H 2 O 2 , NQO1, which catalyzes detoxification of various electrophilic toxicants and oxidants and SOD, which dismutates the superoxide radical (O 2 - − ) to H 2 O 2 . Thus, pre-and postoperative plasma levels of GSH levels and antioxidant enzyme (CAT, GPx, NQO1, and SOD) activities as well as protein and mRNA expressions of related antioxidant enzymes (GCLC, CAT, GPx, and MnSOD) and Nrf2, a well-known transcription factor that regulates expression of detoxifying enzyme genes, in skin lesions were determined in BCC patients compared with control subjects. Our results indicated that while decreases in CAT, GPx, and NQO1 activities in BCC patients and in protein expressions of CAT, GCLC, GPx, Nrf2, and MnSOD in BCC tissues compared to epidermis of control subjects were observed, plasma SOD activities and GSH levels were higher in BCC patients than in control subjects. In correlation with IHC findings showing diminished expressions of all antioxidant proteins in BCC tissues, mRNA levels of the corresponding genes including CAT, GCLC, GCLM, NQO1, Nrf2, CuSOD, and MnSOD were also reduced in BCC tissues compared to epidermis of control subjects.
While CAT, GPx, and NQO1 activities markedly declined in BCC patients compared to control subjects, elevation of SOD activities and GSH levels was observed in association with upregulated protein expressions of GCLC, a rate-limiting enzyme in GSH synthesis, and MnSOD in the adjacent nonneoplastic tissues of BCC patients. In agreement with previous studies, lower activities of GPx and CAT as well as higher activities of SOD were observed in patients with oesophageal, gastric, and colorectal cancers compared with control subjects [bib_ref] Antioxidant enzyme activities and lipid peroxidation levels in erythrocytes of patients with..., Dursun [/bib_ref] [bib_ref] Plasma antioxidant enzymes and clastogenic factors as possible biomarkers of colorectal cancer..., Maffei [/bib_ref] and low NQO1 activity was linked to increased risk of acute leukemia [bib_ref] Low NAD(P)H:quinone oxidoreductase 1 activity is associated with increased risk of acute..., Smith [/bib_ref]. NQO1 polymorphism, which can cause reduction of its activity, was reported to affect susceptibility to lung, bladder, and colorectal cancers that could either increase or decrease cancer risks associated with ethnicity and exposure to carcinogens [bib_ref] NAD(P)H:quinone oxidoreductase 1 (NQO1) Pro187Ser polymorphism and the risk of lung, bladder,..., Chao [/bib_ref]. In addition, this study demonstrated decreased Nrf2 expression in BCC tissues compared to both adjacent nonneoplastic tissues of BCC patient and noncancerous tissues of control subjects which was in agreement with previous reports showing downregulation of Nrf2 and its target genes including NQO1 was associated with malignant transformation and upregulation of antioxidant mRNA including CAT, GCLC, GCLM, GPx, and NQO1 genes by Nrf2 overexpression was able to delay tumor growth [bib_ref] Oncogenic transformation of mesenchymal stem cells decreases Nrf2 expression favoring in vivo..., Funes [/bib_ref]. Nevertheless, the role of Nrf2 in cancer is complex as it can play a dual role in both cancer prevention and promotion depending on cellular environment [bib_ref] Dual roles of NRF2 in tumor prevention and progression: possible implications in..., Moon [/bib_ref]. In addition, differential expression of antioxidant proteins was found in several cancers. Reduction of CAT and GPx protein expressions was demonstrated in neoplastic tissues of patients with esophageal, colorectal, and lung cancers, although both downregulation and overexpression of MnSOD protein in lung and esophageal cancers, respectively, were reported [bib_ref] Differential expression of manganese superoxide dismutase and catalase in lung cancer, Ho [/bib_ref] [bib_ref] Aberrant expression of selenoproteins in the progression of colorectal cancer, Murawaki [/bib_ref] [bib_ref] Using proteomic approach to identify tumor-associated proteins as biomarkers in human esophageal..., Zhang [/bib_ref].
Enhancement of plasma SOD activities and GSH levels in correlation to increased protein expressions of MnSOD and GCLC in the adjacent nonneoplastic tissues of BCC patients may be due to an adaptive response of normal skin cells to persistent elevation of oxidative stress and damage in cancer patients. It has been suggested that upregulation of antioxidant defenses including GSH and MnSOD may serve as the defense mechanisms for cell survival against stress and inflammatory insults, which can take place during cancer initiation and progression [bib_ref] Oxidative stress, inflammation, and cancer: how are they linked, Reuter [/bib_ref]. Accumulation of ROS and oxidative damage during malignant transformation was found to be accountable for transcriptional downregulation of antioxidant defense via Nrf2 signaling [bib_ref] Oncogenic transformation of mesenchymal stem cells decreases Nrf2 expression favoring in vivo..., Funes [/bib_ref]. Moreover, previous reports suggested that the presence of a tumor could have the systemic impact on distant tissues by generating oxidative stress and releasing inflammatory mediators including cytokines, which in turn can promote oxidative damage [bib_ref] Systemic DNA damage accumulation under in vivo tumor growth can be inhibited..., Georgakilas [/bib_ref] [bib_ref] Chronic inflammation and cytokines in the tumor microenvironment, Landskron [/bib_ref]. Thus, increased oxidative damage and changes in antioxidant defense system may associate with initiation of carcinogenesis and/or occur as a consequence of alterations in cellular homeostasis and antioxidative metabolism or of inflammation in tumor growth, progression, and microenvironment. Case-control studies of genetic polymorphisms in DNA repair and antioxidant enzymes as well as changes in inflammatory parameters in association with BCC risk with a large number of subjects should also be done in future studies.
This study showed that patients with BCC, a locally invasive malignant skin cancer, could exhibit systemic disturbance in redox status. Previous in vivo and clinical studies have reported the impact of skin damage on systemic oxidative parameter. Acute exposure of mice skin to UVA and UVB irradiation was shown to affect markers of oxidative damage and antioxidant defenses in plasma and nonskin tissues including erythrocytes and liver [bib_ref] Acute exposure to solar simulated ultraviolet radiation affects oxidative stress-related biomarkers in..., Svobodová [/bib_ref]. Moreover, lower CAT activity and higher SOD activity and MDA levels were observed in plasma of patients with acne vulgaris [bib_ref] Oxidative stress in patients with acne vulgaris, Arican [/bib_ref] and lower GSH levels and greater MDA levels in plasma of cutaneous melanoma patients compared with control subjects [bib_ref] Systemic oxidative profile after tumor removal and the tumor microenvironment in melanoma..., Bernardes [/bib_ref]. Furthermore, our findings suggested that surgical intervention could influence oxidative DNA damage and antioxidant defense status only in case patients with BCC but not control subjects with nonmalignant skin diseases because tumor removal was observed to be related to improvement of redox status by reducing 8-oxo-dGuo levels at 6 months postoperatively and enhancing plasma antioxidant defenses including CAT, GPx, and NQO1 activities at 1 month postoperatively.
# Conclusions
Patients with BCC may be under oxidative stress associated with induction of oxidative DNA damage, defects in DNA repair hOGG1 at protein and mRNA levels, and reduction of plasma CAT, GPx, and NQO1 activities and of all antioxidant proteins and genes studied in the BCC tissues. Surgical removal of BCC tissues correlated with improved redox status. An elevation of plasma total SOD activities and GSH levels as well as protein expressions of MnSOD and GCLC in nonneoplastic tissues of BCC patients may indicate an adaptive response to oxidative stress. Whether oxidative DNA damage and antioxidant defense parameters can serve as biomarkers of oxidative stress to predict development and progression of BCC needs further studies.
[fig] Figure 1: The urinary 8-oxo-dGuo levels (a) and plasma antioxidant defense status [CAT (b), GPx (c), NQO1 (d), and total SOD (e) activities [/fig]
[fig] Figure 2: The H&E staining ((a)-(c) [/fig]
[fig] Figure 3: DNA repair and antioxidant gene expression in nonmalignant skin tissues of control subjects and BCC tissues. Gene expression was evaluated by real-time PCR with the 2 −ΔΔCt method. The data are presented as the fold change in gene expression normalized to GAPDH. Values given are mean ± SD. The statistical significance of differences between the control and case was evaluated by nonparametric variables with Mann Whitney test. * * * < 0.001 compared to control. [/fig]
[table] Table 1: Demographics and clinical characteristics of controls subjects and BCC patients. [/table]
[table] Table 2: Comparison of urinary oxidative DNA damage levels and plasma antioxidant defense status between control subjects and BCC patients. [/table]
[table] Table 3: Expressions of oxidative DNA damage, DNA repair enzyme, and antioxidant proteins in skin tissues of control subjects and BCC patients by IRS. [/table]
|
Clinicians’ attitudes towards clinical trials of cancer therapy
BACKGROUND: Patient accrual into cancer clinical trials remains at low levels. This survey elicited attitudes and practices of cancer clinicians towards clinical trials. METHOD: The 43-item Clinicians Attitudes to Clinical Trials Questionnaire was completed by participants in an intervention study aimed at improving multi-disciplinary involvement in randomised trials. Responses from 13 items were summed to form a researchorientation score. RESULTS: Eighty-seven clinicians (78%) returned questionnaires. Physicians, more often than surgeons, chose to prioritise prolonging a patient's life, recruited X50% of patients into trials and attended more research-focussed conferences. Clinicians at specialist centres were more positive about trials with no-treatment arms than those at district general hospitals, more likely to believe clinician, rather than patient reluctance to participate was the greater obstacle to trial accrual, and preferred national and international to local recognition. Clinicians belonging to breast and colorectal teams were less disappointed about not enrolling patients in trials and more accepting of no-treatment arm trials. Research orientation was higher in physicians than surgeons and higher in specialist centres than district hospitals. CONCLUSIONS: This study provides greater understanding of clinicians' attitudes to trials. Results have been used to inform training interventions for clinicians targeting the problem of low and selective accrual.
The randomised controlled trial (RCT) remains the gold-standard by which new cancer treatments are assessed and through which therapeutic progress is made. Despite the need to develop new cancer therapies, accrual of patients into such trials remains low with fewer than 5% of cancer patients participating [bib_ref] Representation of African-americans, hispanics, and whites in national cancer institute cancer treatment..., Tejeda [/bib_ref] [bib_ref] Increasing participation of physicians and patients from underrepresented racial and ethnic groups..., Christian [/bib_ref] [bib_ref] Participation in cancer clinical trials: Race-, Sex-, and age-based disparities, Murthy [/bib_ref] [bib_ref] Organizational barriers to physician participation in cancer clinical trials, Somkin [/bib_ref] [bib_ref] Generalizability of cancer clinical trial results, Elting [/bib_ref]. Even in centres running trials suitable for a particular patient group such as women with limited, local breast cancer or other solid tumours, participation of those eligible is only 12 -30% [bib_ref] Generalizability of cancer clinical trial results, Elting [/bib_ref] [bib_ref] The role of the multidisciplinary team in recruiting to cancer clinical trials, Maslin-Prothero [/bib_ref]. As well as increasing the time needed to complete studies this low recruitment means that randomised patients may differ significantly from those who do not take part. This risks undermining the study results by making it difficult to draw generally applicable conclusions [bib_ref] Generalizability of cancer clinical trial results, Elting [/bib_ref]. Several US and European organisations acknowledge that slow accrual hampers the timely release of new therapies into the clinic. For example, only about 60% of US National Cancer Institute (NCI)-sponsored trials are completed and published [bib_ref] Impact of selection process on response rate and long-term survival of potential..., Nass [/bib_ref]. In addition, certain patient populations are under-represented, resulting in the recent development of several national and international programmes to address the problem of slow and selective recruitment [bib_ref] Applying community-based participatory research principles and approaches in clinical trials: forging a..., Seifer [/bib_ref].
Studies have found that the majority of clinicians favour clinical trials, viewing them as a source of high-quality patient care, and as a benefit to themselves, their institutions, and to society [bib_ref] Attitudes of primary care physicians and specialists about cancer clinical trials: a..., Weinberg [/bib_ref] [bib_ref] Organizational barriers to physician participation in cancer clinical trials, Somkin [/bib_ref]. However, there are two main problems needing attention if trial participation is to be increased: first that clinicians do not offer trials to all eligible patients [bib_ref] The impact of the physician on the accrual to randomized clinical trials..., Kaas [/bib_ref] , and second that some patients offered seemingly appropriate trials refuse participation [bib_ref] Factors that predict the referral of breast cancer patients onto clinical trials..., Siminoff [/bib_ref]. Reducing clinician gate-keeping and broadening the range of patients to whom trials are offered is challenging, but improving clinicians' communication skills when explaining trials has the potential to resolve patients' concerns and increase the likelihood of their participation. A review of NCI-sponsored trials noted the importance of changing physicians' perspectives so that they will be more likely to encourage their patients to participate in clinical trials [bib_ref] Impact of selection process on response rate and long-term survival of potential..., Nass [/bib_ref]. Understanding more about clinicians' attitudes and practices in this area is necessary and timely, if recruitment is to be improved.
Research shows that clinicians often adopt stringent, idiosyncratic criteria when selecting the patients to whom they offer trials, over and above the criteria delineated in trial protocols. In particular, some clinicians only select patients with even better health status and prognosis than demanded by the protocol [bib_ref] Selection bias in clinical-trials, Antman [/bib_ref] [bib_ref] Impact of selection process on response rate and long-term survival of potential..., Nass [/bib_ref] [bib_ref] Small-cell lung cancer: Patients included in clinical trials are not representative of..., Cottin [/bib_ref] [bib_ref] Factors that predict the referral of breast cancer patients onto clinical trials..., Siminoff [/bib_ref] or choose not to offer trial participation on putative compassionate grounds [bib_ref] A systematic review highlights threats to validity in studies of barriers to..., Fayter [/bib_ref]. Clinicians may hesitate to inform patients of trials, based on their own attitudes and beliefs about a patient's willingness to participate, ability to understand the trial, or adhere to the protocol . This is despite 83% of patients being potentially willing to participate if given full trial information [bib_ref] The attitudes of 1066 patients with cancer towards participation in randomised clinical..., Jenkins [/bib_ref]. Other clinicians are reluctant to enrol patients in trials if the treatment concerned may result in extra side effects [bib_ref] A survey of the views of palliative care healthcare professionals towards referring..., White [/bib_ref]. If older and sicker patients are not included in clinical trials then trial participants are likely to have better outcomes than would be seen were the new treatments offered in normal clinical practice.
Perhaps unsurprisingly, assessments of clinicians' attitudes show that many tend to be clinically, rather than research oriented, believing that individual benefit for their patient is more important than improving future therapy [bib_ref] Barriers to Australian physicians' and paediatricians' involvement in randomised controlled trials, Caldwell [/bib_ref]. Some indicate that the main reason for including their patients in trials is so that they receive state of the art treatment [bib_ref] Views of American oncologists about the purposes of clinical trials, Joffe [/bib_ref] , and trials focussing on quality of life outcomes tend to have better accrual rates [bib_ref] Physicians' attitudes toward clinical trials and their relationship to patient accrual in..., Hjorth [/bib_ref].
A lack of confidence in explaining clinical trials to patients and obtaining informed consent is a barrier to some clinicians [bib_ref] Factors influencing inclusion of patients with malignancies in clinical trials: a review, Tournoux [/bib_ref] [bib_ref] A systematic review highlights threats to validity in studies of barriers to..., Fayter [/bib_ref]. Clinicians are reticent to enrol patients who may have difficulty understanding what the trial involves [bib_ref] Attitudes of primary care physicians and specialists about cancer clinical trials: a..., Weinberg [/bib_ref]. If the trial arms are not equally attractive or there is a no-treatment arm, this too may prove a disincentive [bib_ref] The role of the multidisciplinary team in recruiting to cancer clinical trials, Maslin-Prothero [/bib_ref]. Furthermore, if the protocol is overly arduous to implement (e.g., requiring substantial time or resource commitments from the clinician or their organisation) clinicians are unlikely to recruit a large number of patients [bib_ref] Physicians' attitudes toward clinical trials and their relationship to patient accrual in..., Hjorth [/bib_ref] [bib_ref] Barriers to participation in randomized clinical trials for early breast cancer among..., Ellis [/bib_ref] [bib_ref] Barriers and facilitators to enrollment in cancer clinical trials: qualitative study of..., Grunfeld [/bib_ref] [bib_ref] Attitudes of primary care physicians and specialists about cancer clinical trials: a..., Weinberg [/bib_ref] [bib_ref] Factors influencing inclusion of patients with malignancies in clinical trials: a review, Tournoux [/bib_ref] [bib_ref] A systematic review highlights threats to validity in studies of barriers to..., Fayter [/bib_ref]. Some patients also may have a problem with understanding the concept of randomisation and may struggle to accept clinical equipoise, for example, in the case of a placebo or no-treatment arm trial [bib_ref] Lay public's understanding of equipoise and randomisation in randomised controlled trials, Robinson [/bib_ref]. Furthermore, clinicians may explain these concepts in ways that patients may find hard to understand or even find threatening .
Despite these barriers, the majority of medical and clinical/ radiation oncologists do report enrolling patients into clinical trials [bib_ref] Clinician's attitudes to clinical trials of cancer therapy, Fallowfield [/bib_ref]. Interestingly, different specialties within oncology have different accrual rates, although which specialties accrue more is not consistent. In one survey of UK clinicians, medical oncologists generally placed more emphasis on research than did surgical or clinical oncologists, and felt greater pressure to participate in trials [bib_ref] Clinician's attitudes to clinical trials of cancer therapy, Fallowfield [/bib_ref]. In Australia, medical and surgical oncologists participated more in clinical trials than radiation oncologists [bib_ref] Barriers to participation in randomized clinical trials for early breast cancer among..., Ellis [/bib_ref]. In the United States, medical and paediatric oncologists reported the lowest rates of patient enrolment in trials [bib_ref] Views of American oncologists about the purposes of clinical trials, Joffe [/bib_ref] , although a more recent study showed greater recruitment among medical and radiation oncologists than among surgeons [bib_ref] A population-based assessment of specialty physician involvement in cancer clinical trials, Klabunde [/bib_ref]. A further study found no differences in attitudes towards clinical trials between surgeons and physicians, although this was not limited to oncologists [bib_ref] Tolerance of uncertainty, extroversion, neuroticism and attitudes to randomized controlled trials among..., Mcculloch [/bib_ref]. An understanding of how differences between specialists affect trial recruitment may help promote accrual across specialties. Some suggested reasons for differences are type of education, concept of medicine, individual or professional aims, affiliation with academic or research groups, or particular working environment [bib_ref] Why don't cancer patients enter clinical trials? A review, Castel [/bib_ref] , while [bib_ref] A population-based assessment of specialty physician involvement in cancer clinical trials, Klabunde [/bib_ref] suggest that specialty type, involvement in teaching, affiliation with a Community Clinical Oncology Programme or with a National Cancer Institute-designated cancer centre were important.
Greater understanding of clinicians' attitudes to clinical trials is needed to target the problem of low and selective accrual [bib_ref] Why don't cancer patients enter clinical trials? A review, Castel [/bib_ref]. The present survey elicited clinician attitudes towards clinical trials in their practice using a modified version of the Physician Orientation Profile [bib_ref] Interpreting physician participation in randomised clinical trials: the physician orientation profile, Taylor [/bib_ref]. The aims were to assess clinician responses for differences by specialty, type of hospital, type of team, and geographic region and to estimate a research-orientation score for each clinician, again looking for differences between groups. The data used for this study are a component of a large Cancer Research UK funded prospective study examining multidisciplinary team members' communication skills and involvement in clinical trials; primarily randomised, controlled, phase 3 trials. The main study examined different aspects of trial recruitment, including involvement of individual team members in clinical trials; assessment of the clarity of health professionals' communication by patients recruited into clinical trials, and attitudes towards RCTs . The attitudes of patients to RCTs [bib_ref] The attitudes of 1066 patients with cancer towards participation in randomised clinical..., Jenkins [/bib_ref] and that of clinicians to RCTs were collected for each multi-disciplinary team (MDT) to provide an evidence-based argument for encouraging clinicians to consider approaching more patients about trials. The clinician data are presented here.
# Materials and methods
## Sample
Recruitment of clinicians (senior doctors) within oncology MDTs (also known as interdisciplinary tumour boards) was a joint effort between the Cancer Research UK Psychosocial Oncology Group and the Wales Cancer Trials Network. Twenty-two MDTs throughout Wales participated in the larger communication study. The study had ethical approval from the South East Wales Local Research Ethics Committee (ref: 07/WSE03/17). Teams were randomised to receive communication training or to go on a waiting list. Consultant (attending) surgeons, oncologists, chest physicians, and haematologists were asked to complete questionnaires examining their attitudes to trials and their involvement in trials before randomisation.
# Materials
The Clinicians' Attitudes to Clinical Trials of Cancer Therapy Questionnaire is a modified version of the 30-item, binary option, Physician's Orientation Profile [bib_ref] Interpreting physician participation in randomised clinical trials: the physician orientation profile, Taylor [/bib_ref] and has been used before by this research group [bib_ref] Clinician's attitudes to clinical trials of cancer therapy, Fallowfield [/bib_ref]. Items on this questionnaire are classified into five subscales assessing various aspects of clinicians' attitudes towards their clinical and scientific work, specifically primary allegiance, professional activities, decision-making under uncertainty, perceived professional rewards, and peer group influence. The questionnaire used in this study had 43 items and is very similar to the 45-item questionnaire used by [bib_ref] Clinician's attitudes to clinical trials of cancer therapy, Fallowfield [/bib_ref]. Professional information (specialty, MDT cancer site, etc.) was also collected for each clinician.
## Derivation of research-orientation score
Two senior researchers (LF and VJ) independently chose items from the questionnaire on the basis of their face validity for the construct 'Research Orientation'. Each researcher independently chose 18 items from the original 43 with 100% consistency. The items were recoded where necessary to provide binary responses (0 out of 1) with 1 indicative of a research orientation. The chosen items were subjected to statistical scrutiny to assess their factorial validity and reliability. Classical principal components analysis (PCA) was undertaken based on the binary responses to these 18 questions (D'Agostino [bib_ref] The EDICT Project: Policy Recommendations to Eliminate Disparities in, Snr [/bib_ref] with the pragmatic aim of identifying items, which were well correlated and could be hypothesised to represent one or more underlying factors or constructs. Following PCA, a scree plot suggested a single component solution was most suitable for these items (first component eigenvalue 5.79, 32.15% of variance; second component 2.03, 11.29% of variance). , and 40 loaded onto the first component with loadings from 0.47 to 0.83. Item 11 loaded at 0.32 and items 5, 22, 32, 43 loaded at o0.30.
All 18 items were then examined for internal consistency. Cronbach's a was found to be 0.83, and low item-total correlations were found for items 5 (0.02), 11 (0.27), 22 (0.13), 32 (0.13), and 43 (0.20), suggesting that a would increase by removing these items. Items 5, 11, 22, 32, and 43 were removed and Cronbach's a recalculated for the 13 items (a ¼ 0. 89). A second PCA on these remaining 13 items showed again that these items loaded onto one component explaining 44.4% of the variance (loadings 0.49 -0.84). The responses from the remaining 13 items were summed to form a research-orientation score.
# Analysis
Question response was linked to possible explanatory variables through logistic regression. Random effects were used to account for within-team correlations, although in general these correlations were low. As the number of clinicians returning questionnaires was relatively small compared with the number of questions, and with conservatism introduced because of within-team correlations, there is little scope for multiplicity adjustments. Thus, significance levels related to individual questions should be treated with caution and seen as primarily providing an ordering of observed discriminatory ability. The analysis of the derived researchorientation score, which represented the sum of 13 questions, was based on continuation ratio ordinal regression analysis, with use of a complementary log -log link. The more flexible continuation ratio model was preferred to a binomial regression because of the variability between questions in the proportion of 'research oriented' answers.
# Results
Out of the 111 clinicians approached, 87 (78.4%) returned questionnaires. The sample was formed of 47 surgeons, 28 oncologists, 9 haematologists, and 3 chest physicians. These clinicians belonged to MDTs for the following cancers: breast (25; 31%), colorectal (15; 17%), gynaecology (8; 9%), haematology (10; 11.5%), lung (6; 7%), lymphoma (3; 3%), upper gastrointestinal (10; 11.5%), and urology (10; 11.5%). Responses to each item on the questionnaire, and differences in responses by separate groupings (physician vs surgeon; specialist hospital (large teaching or city hospital) vs district general hospital (DGH) (community hospital); breast and colorectal teams vs others) are shown in [fig_ref] Table 1: Differences in response in physicians vs surgeons, specialist hospital vs district general... [/fig_ref]. [fig_ref] Table 2: Items distinguishing between physicians and surgeons [/fig_ref] shows the questionnaire items that best distinguished between physicians (oncologists, haematologists, and chest physicians) and surgeons. Physicians more frequently endorsed prolonging a patient's life over improving quality of life. They were also more likely to report recruiting 50% or more of their patients into clinical trials and attending conferences focussed on research rather than clinical issues. These differences suggest physicians may be more research focussed than their surgical counterparts. [fig_ref] Table 3: Items distinguishing between clinicians in specialist centres and those in district general... [/fig_ref] shows the questionnaire items best differentiating clinicians who worked in a specialist cancer centre from those who worked in a DGH. Those working at a specialist centre were less reluctant to randomise patients into trials with no-treatment arms, more likely to believe that clinician reluctance (as opposed to patient reluctance) to participate was a major obstacle to trial accrual, and were more likely to wish to be well known among Audit: Overall 'Teams Talking About Trials' project. *Data presented in this paper. national or international, rather than local, colleagues. These differences suggest that those at specialist centres may be more research oriented than those at DGHs.
## Differences by specialty
## Differences by centre
## Differences by geographical region
There were some regional differences in answers to questionnaire items, but after exploration of confounders, it was found that the effect of region (SE vs N and SW) disappeared when specialist centre was added to the model.
## Differences by type of mdt
Multi-disciplinary teams were grouped into breast and colorectal teams vs the other teams . This grouping was chosen because breast and colorectal cancer are areas where there are typically many clinical trials and where recruitment has been common . We hypothesised that MDTs treating such patients may be more favourable to clinical research. [fig_ref] Table 4: Items distinguishing between MDTsQ15 [/fig_ref] shows the questionnaire items which best differentiate clinicians by their MDT specialty. These results suggest that breast and colorectal teams may be less disappointed about not enrolling patients in a trial and more accepting of trials with no-treatment arms. Interestingly, clinicians belonging to other types of teams reported stronger publications records and were more desirous of international rather than local recognition.
## Research orientation
Binary responses to questions , and 40 (highlighted in bold in [fig_ref] Table 1: Differences in response in physicians vs surgeons, specialist hospital vs district general... [/fig_ref] were summed to form the research-orientation scale. Full data were available from 74 participants and scores on the scale ranged from 0 to 12. The numbers of observations with these 13 values were 18, 23, 7, 8, 4, 4, 1, 1, 1, 2, 3, 1, and 1, respectively, showing a skew towards the nonresearch orientation end of the scale. Ordinal regression analysis demonstrated that the probability of a high research orientation was greater in physicians vs surgeons (Po0.001) and in respondents from a specialist centre (Po0.001) but little demonstrable effect was associated with MDT type (breast/colorectal vs rest: P ¼ 0.06). Both the physician/surgeon and specialist/DGH centre classifications retained their significance in a multivariate analysis but MDT type demonstrated no relationship with research orientation [fig_ref] Table 5: Multivariate ordinal regression analysis of clinical/researchorientation scale Abbreviation [/fig_ref].
# Discussion
## Recruitment of patients into clinical trials is essential if progress is
to be made in improving cancer care. Despite this, recruitment is often poor with studies slow to complete and with only a small proportion of cancer patients being enrolled in studies. The process of recruitment involves discussion between the patient and their doctor but is preceded by identification of suitable patients fulfilling eligibility criteria and then the doctor's decision to offer the patient an appropriate trial. The attitude of doctors is therefore a critical part of the enrolment process. This study evaluated doctors' attitudes to clinical research and to clinical trial recruitment. Using the Clinicians' Attitudes to Clinical Trials of Cancer Therapy Questionnaire the attitudes of 87 clinicians with a range of specialties, tumour site interests, geographical location, and affiliations were assessed. The questionnaire also allowed clinicians to be allocated a research orientation score.
The results indicate that research orientation was greater in physicians than surgeons, with physicians more likely to enter patients into trials and more likely to attend research-focussed conferences. The lesser research orientation found in surgeons may reflect the paucity of surgical trials available and the difficulties encountered recruiting patients into a surgical trial. One study of US breast cancer surgeons showed that 26% did not discuss trials with any breast cancer patients, and a further 39% discussed trials with fewer than 10%. Surgeons identified inadequate infrastructure and lack of time as significant deterrents in trial participation [bib_ref] Clinical trial priorities among surgeons caring for breast cancer patients, Schroen [/bib_ref]. [bib_ref] A population-based assessment of specialty physician involvement in cancer clinical trials, Klabunde [/bib_ref] also show significantly lower clinical trial recruitment among surgeons compared with oncologists although surgical oncologists were more likely to participate in clinical trials than general surgeons. Furthermore, surgeons showed the same pattern as physicians with increased trial participation being associated with academic affiliation, teaching of medical students, having a more specialist clinical practice, and more frequent attendance at multidisciplinary meetings.
It may be that although surgeons are a crucial part of a cancer team, they feel they have a remote role in recruiting patients to medical/oncological trials. The increasing number of neo-adjuvant and peri-operative trials makes this untenable. Furthermore, it can be argued that the surgeon has a pivotal role in influencing patients' expectations about the cancer treatments on offer following surgery. The way the surgeon introduces the idea of trials in general can facilitate any discussions, which follow with the oncologist. If trials are treated as a 'team business', all team members should be made aware of what trials are available for patients in that tumour site and have the skills and knowledge necessary to discuss these trials with patients.
The finding that clinicians at specialist centres are more research oriented than those in DGHs is not surprising. Specialist centres have more staff and resources available for trials and are more likely to be set up to facilitate the running of trials. For example, in a DGH there may be only one histopathologist dealing with several cancer sites and who does not have any protected time to deal with 'trial' tissue blocks in the tight timelines that some trials have. Specialist centres may also be attached to research institutions and therefore their staff is more likely to have a teaching or research focus. Research-oriented clinicians may selfselect into posts within specialist centres or teaching hospitals. It is interesting that clinicians working in specialist centres seem more aware that clinician reluctance could be a barrier to the successful completion of a clinical trial. This may result in clinicians in specialist centres being more aware of their communication training needs and more likely to address their own personal barriers to trial recruitment.
Furthermore, ability to find trial information, and the familiarity of investigators with trials, may differ in different types of hospital. According to one survey, the most common reason cited for physician non-participation in trials was a lack of knowledge about available trials [bib_ref] Most physicians do not participate in clinical trials because of lack of..., Taylor [/bib_ref]. A review of US publicly funded trials suggested that 'encouraging the development of a user-friendly, transparent, up-to-date, and easily accessible centralised registry could improve both physician and patient awareness of the available trials. In combination with electronic tools, such as, clinical decision support software, a centralised registry could cue physicians to important, applicable clinical trials at the point of care' . A centralised system of trial information available to all clinicians, whatever type of institution they worked in, would potentially overcome some barriers to recruitment in DGHs. Lack of clinical trial support has been frequently cited as a barrier to clinical trial involvement [bib_ref] Organizational barriers to physician participation in cancer clinical trials, Somkin [/bib_ref] [bib_ref] Clinical trial priorities among surgeons caring for breast cancer patients, Schroen [/bib_ref] but this issue does not explain the findings in this study as all the MDTs involved in the study were supported by the Wales Cancer Trials Network whose infrastructure (research nurse time, data management support, etc.) was available to all team members. Other papers refer to lack of reimbursement for clinical trial work and other financial constraints as a barrier to recruitment [bib_ref] A population-based assessment of specialty physician involvement in cancer clinical trials, Klabunde [/bib_ref]. Again this was not relevant to this study, which took place in a health service (NHS) where all staff are salaried.
Our hypothesis that clinicians working in breast and colorectal cancer teams would be more research oriented was not borne out by the data. Although these clinicians were more accepting of notreatment arms and more sanguine about patients declining trial entry, they were less likely to seek national or international recognition and had fewer publications compared with clinicians treating other cancer sites. It appears then that specialty by cancer site is not a marker of research orientation. As new therapies are required across all types of cancer, it may be that clinicians specialising in all tumour sites have a broadly equal interest in research, because of the constantly emerging treatments in their field.
The research-orientation scale showed a strong skew towards the non-research orientation end of the scale, with the majority of participants scoring 0 or 1 out of a possible 13. However, despite the skew, in ordinal analyses scores on this scale were significantly different between surgeons and physicians, and between those working in specialist centres and in DGHs, confirming our findings from the single question analyses. This scale can therefore be used in future studies to investigate clinician orientation towards research in general and trials in particular, and could potentially be used to evaluate training materials designed to increase research interest and participation among clinicians.
There are several limitations to this study. The clinicians assessed were all members of recognised cancer MDTs and were involved in regular discussion of patient management. In Wales, all clinical trial recruitment is recorded centrally and MDTs with no recognised trial recruitment were excluded as this was an explicit entry criterion for the associated trial. It is therefore possible that the attitudes expressed are more favourable towards research than would be seen in the general clinical population, as MDTs with no trial experience were excluded.
Second, the questions on attitudes in this study asked primarily about attitudes to RCTs and not to earlier, phase 1 and 2 trials.
Clinicians may have different attitudes to early phase trials, which have different aims and require specialist infrastructure. Third, only a small range of clinician characteristics, such as specialty, were available. Clinician attitudes may also vary by other clinician characteristics, such as number of patients seen per month, clinician sex, training status, previous experience as principal investigator, time since graduation, or academic or industry affiliations [bib_ref] Views of American oncologists about the purposes of clinical trials, Joffe [/bib_ref]. In addition, the clinician sample in this study was culturally homogenous and therefore cultural or ethnic differences in attitudes could not be explored. Findings may be different in a more multicultural sample. It is also possible that clinician attitudes may depend on the sociodemographic characteristics of their patient population. Further research in this area would be interesting and may be useful in increasing understanding about current recruitment practices.
In conclusion, this study gives us greater understanding of clinicians' attitudes to clinical trials, in order to target the problem of low and selective accrual. Despite an increase in overall recruitment to trials over the past 5 years, patients' understanding of them and health professionals' explanations about them are still problematic. There is a crucial need for all cancer team members to be fully supportive and aware of the trials that are available in their centres, and for all to receive and deliver consistent information to patients and their relatives. The results of this study are part of a larger project, in which members from participating MDTs have taken part in training. The training has addressed problems such as team dynamics acting as a barrier to trial recruitment, inconsistent team communication, and the difficulty of talking about trials where patients are unenthusiastic about joining for different reasons. Previous research has shown that communication skills training can also alter the attitudes and beliefs of clinicians . The overall aims of this work are to increase the number of patients offered a trial by clinicians, and to increase the quality of communication about trials when they are discussed.
[table] Table 1: Differences in response in physicians vs surgeons, specialist hospital vs district general and breast and colorectal teams vs other teams Q1. Ideally, clinicians are able to increase survival and improve the quality of patients' lives. In cases where only one can be achieved at the cost of the other, I feel more satisfied when I can:Q2. If a patient refuses to participate in a randomised clinical trial that I suggest, I would:Q10. When a patient on a protocol relapses or progresses and the protocol dictates additional treatment that the patient does not want, I: [/table]
[table] Table 2: Items distinguishing between physicians and surgeons [/table]
[table] Table 3: Items distinguishing between clinicians in specialist centres and those in district general hospitals [/table]
[table] Table 4: Items distinguishing between MDTsQ15. When a potentially eligible patient chooses not to enrol on a trial that I have suggested I: Q19. I am reluctant to participate in a trial that may randomise a patient to a 'no treatment' group.Abbreviation: MDT ¼ multi-disciplinary team. [/table]
[table] Table 5: Multivariate ordinal regression analysis of clinical/researchorientation scale Abbreviation: DGH ¼ district general hospital. [/table]
|
Glucose-oxidase like catalytic mechanism of noble metal nanozymes
[fig_ref] Figure 2d: Mechanism of glucose oxidation catalyzed by GOD [/fig_ref]
## Reviewer #1 (remarks to the author):
The authors have investigated the mechanism for the catalytic reactions between glucose and O2 (ABTS + ), with noble metal nanoparticles (NPs) as the catalysts, which they call nanozymes. They have found that the most probable mechanism for this catalytic process is first the reduction of the NPs by the glucose to form the intermediate like NP-H and then is the transfer of the H from NP-H to the oxidant O2 (ABTS + ). I feel this work could be published after the authors address my following concerns. Response. Thank you very much for your positive comments. Comment 1. Because the proposed catalytic mechanism involves mainly the transfer of hydrogen atoms rather than electrons and because electron transfer is a concept that already exists, the title of this manuscript is misleading. Response 1. Thank you for this suggestion. The title of this manuscript has been changed to "Bionic catalytic pathway enlightened noble metal nanozymes" Comment 2. The mechanism proposed by the authors for the catalytic process (glucose + O2) is not complete. It could not explain how H2O2 is formed and why Au has the largest catalytic activity with the activity order Au Pt Ru Ir Pd Rh. Response 2. Thank you for this suggestion. The specific catalytic process is shown in (* denotes a bare surface active site). Glucose oxidation can be divided into two steps: the first step is dehydrogenation of glucose, and the second step is oxygen reduction. For dehydrogenation of glucose, the O-H bond in the hydroxyl group breaks and the hydrogen protons are removed. Then the C-H bond breaks and the proton and two electrons leave. The removed protons and electrons were transferred to oxygen. In the oxygen reduction reaction, O2 can be reduced to H2O by 4epathways or to H2O2 by 2epathways. The difference between the two paths is that the O = O double bond in the 2epath is not completely broken, so the product is H2O2, while the O = O double bond in the 4epath is completely broken, resulting in water.
For Au catalyzed oxygen reduction, the energy barrier of O = O double bond breaking is higher . Therefore, Au catalyzed oxygen reduction mainly takes the 2epath, and the product is H2O2.
In previous reports, theoretical calculation shows that the rate determining step of alcohol oxidation is the breaking of C-H bond, and Au has the strongest effect on reducing the breaking energy of C-H bond (Science 330 (6000),74-78). Therefore, Au has the highest catalytic activity.
We have added some content in the revised manuscript to discuss the specific catalytic process of glucose oxidation.
This The difference between 2e-path and 4e-path in oxygen reduction reaction.
Comment 3. It seems that the oxidations of TMB in presence of noble metals NPs and O2 follow a totally different mechanism, because this series of reactions have a totally different activity order (Pt Ir Pd Rh Ru Au vs. Au Pt Ru Ir Pd Rh). The authors may want to explain the reason for the different activity order when the reactant is changed from glucose to TMB. Response 3. The oxidations of TMB in presence of noble metals NPs and O2 follow totally different mechanism. The oxidation potential of TMB is 1.13 V, which is only slightly lower than that of oxygen (1.23 V). Therefore, it is generally believed that the oxidation of TMB requires the production of reactive oxygen species with higher activity.
In previous studies, researchers reported that the oxidase-like activity of a metal is closely related to the dissociative adsorption of the O2 on the respective metal surface (Nat Commun 9, 129 (2018)): O2 = 2 O According to this mechanism, the metal surface catalyzes the dissociation of the O2 molecule to yield single O atoms, which are responsible for the subsequent oxidation of substrates.
Because Au is difficult to catalyze the break of double bonds in O2, it is difficult for Au to catalyze TMB oxidation with O2 as oxidant.
In the revised manuscript, some discussion was added to speculate the reason of differences between glucose oxidation and TMB oxidation.
## Minor points
1. Line 56 in Page 2, the abbreviation "GOH" is not defined before use.
Response. Thank you very much for pointing this out. The "GDH" was defined before use in the revised manuscript. 2. in Page 6, what does "C1" and "N5" mean? Please label them in [fig_ref] Figure 2d: Mechanism of glucose oxidation catalyzed by GOD [/fig_ref]. Response. Thank you for your suggestion. We have noted "C1" and "N5" in [fig_ref] Figure 2d: Mechanism of glucose oxidation catalyzed by GOD [/fig_ref]. Reviewer #2 (Remarks to the Author): In the manuscript titled "Open a way to explore nanozymes by following the electron transport pathway in the catalytic process of natural enzymes", the authors studied the glucose oxidase mimics of Au and other noble metal NPs. They found that all the NPs can catalyze and dehydrogenation of glucose, and only Au ones reduce O2 to H2O2 instead of H2O by two electrons transfer. Overall, this work is interesting and the experiments are rather comprehensive. So, it can be accepted for the publication on this Journal after concerning the following comments. Response. We are very grateful to your valuable comments. were dehydrogenized and formed aldehyde groups in the first step reaction. It seems reasonable. However, in my opinion, to solid their contribution, the authors should design and present some experiment results to demonstrate the generation of aldehyde products. Response 1. Thank you for your suggestion. We chose benzyl alcohol and cyclohexanol as substrates and detected the products by GC-MS. In the typical procedure, 500 μL of as-prepared Au NPs and 20 μL of benzyl alcohol or cyclohexanol were sequentially added into a vial containing 10 mL of 100 mM PBS (pH = 8). After reaction for 2 hours, 5 mL of ethyl acetate was added into the mixture solution to extract organic components. The products were detected by GC-MS. As shown in , benzaldehyde can be detected in the presence of Au, which proves that Au can catalyzed the oxidation of alcohol hydroxyl to aldehyde group. Besides, when the substrate is cyclohexanol, the oxidation product is cyclohexanone.
## Comment 1. the authors have considered that glucose and other substrates
Supplementary . Chromatogram of benzyl alcohol before (a) and after (b) Au catalyzed oxidation reaction.
Supplementary . Chromatogram of cyclohexanol before (a) and after (b) Au catalyzed oxidation reaction.
## Comment 2.
The authors shown that their proposed catalytic systems can be employed for glucose detection by a colorimetric signal readout. In view of "practical" applications, for example blood sugar assay, two points should be well concerned. The first one is the catalytic performances of Au and other metal NPs glucose oxidase mimics. Based on previous reports (for example , the catalytic activity of Au NPs can be substantially quenched by thiol molecules due to strong S-Au bond. It is known many biological samples contain lots of thiol biomolecules (cysteine, glutathione, etc.). So, whether can these metal NPs well hold their activity in some complex environments? Second, as the observation by the authors, lots of hydroxyl-containing substances can be catalyzed and present similar reactions. Obviously, how to resolve the selectivity? Response 2. Thank you for your insightful comments. In our early study, we also found that Au is easy to inactivate. At the beginning of this work, we synthesized citric acid protected gold nanoparticles (Cit-Au). The size of Cit-Au is about 13 nm, but its catalytic activity is higher than that of PVP-Au (5 nm) in pure water. We think that the binding force of citric acid on Au surface is weak, therefore, Cit-Au has higher catalytic activity than PVP-Au. This phenomenon is consist with previous report that the catalytic activity of "Naked" Gold particles was very high (Angew. Chem. . However, the Cit-Au are easy to inactivate, even in PBS. Although the activity of PVP-Au is low, it is more stable than Cit-Au. So we choose PVP-Au as catalyst to carry out the following research.
We further compare the stability of Au NPs under complex conditions. When 20 μL of serum was added to the reaction solution, the catalytic activity of Cit-Au disappeared completely, while that of PVP-Au remained about 10%. PVP has a good protective effect on nanoparticles, so PVP-Au can still retain a certain catalytic ability. According to our experiences, it is believe that even the same kind of nanoparticles may show different stability and catalytic ability(we can even think they are different catalysts, as you may see, different literatures with same kind of nanoparticles may give contradictive results). Our work tried to balance the activity and stability of the nanoparticles because PVP capped Au NPs are not easily quenched.
Improving the selectivity of substrate is always the challenge task for the development of pure nanozymes. Taking Au NPs for example, many works reported the applications of Au nanozymes for the detection of glucose, while these work hardly achieve good selectivity because nanozyme are essentially nanoparticles that may not provide a specific active site to bind the substrate. To fulfill the selectivity of the nanozyme, we can usually use them with further modifications, or innovatively use them in specific cascade reactions.
In this work, we proposed the protocol of synthesizing 6 kinds of metal nanozymes by a simple and cheap method, followed by exploring their catalytic mechanism, studying their catalytic ability and developing the cascade reactions for nanozyme catalysis. We also think the stability and selectivity of Au nanoparticles are important for bioassay, which may need systematic study to address. Moreover, we believe that revealing the mechanism of glucose oxidation catalyzed by Au NPs is of great significance for guiding studies to improve the catalytic selectivity in the future.
## Reviewer #3 (remarks to the author):
This work by Chen et al. systematically demonstrated the glucose oxidase-like catalytic mechanism of noble metal nanozymes from the electron transfer pathway. They found that the glucose oxidation process catalyzed by Au NPs is the same as that of natural glucose oxidase, that is, a two-step reaction, including glucose dehydrogenation and subsequent two electrons to reduce O2 to H2O2. Moreover, Pt, Pd, Ru, Rh and Ir NPs can also catalyze the dehydrogenation of glucose, but O2 tends to be reduced to H2O. Using the electron transfer properties of noble metal nanoparticles, they overcome the limitation of the traditional two-step glucose analysis that must generate H2O2, and achieve rapid one-step colorimetric detection of glucose. Inspired by the electron transport pathway in the process of natural enzyme catalysis, it was also found that noble metal nanoparticles can mimic various enzymatic electron transfer reactions of cytochrome c, coenzyme and nitrobenzene reduction. This work explained the similarity of the catalytic mechanism between nanozymes and natural enzymes through simple and ingenious design, and further laid the foundation for the research of the enzyme-like catalytic mechanism of noble metal nanozymes. Although there have been many studies on the catalytic mechanism of nanozymes, this study is one of the rare studies that confirms the similarity between nanozymes and natural enzymes from the electron transfer pathway. This research will have important reference significance for subsequent research on the catalytic mechanism of nanozymes. Through an in-depth understanding of the electron transfer path in the catalysis of noble metal nanozymes, this research subtly improved the detection method of glucose and extended the application scope to various enzymatic electron transfer reactions. This research will promote the research and application of nanozymes in the field of molecular detection. Therefore, I suggest that this manuscript be considered for publication by Nature Communications after the following revisions. Response. We are very grateful to your encouraging comments. Comment 1. The molecular formulas of hydroxyl radicals and superoxide radicals were not standardly written. Response 1. Thank you very much for the comments. These errors have been corrected.
Comment 2. Some abbreviations in the text lack full explanations, such as GOH, NAD(P), etc. Response 2. Thank you very much for pointing this out. The "GDH" and "NAD(P)"were defined before use in the revised manuscript.
Comment 3. In line 146, the authors defined Au NPs as formate oxidase mimics and alcohol oxidase mimics. However, showed that the oxidation rate of ethanol catalyzed by Au NPs was very low, so it is not appropriate to call it an alcohol oxidase mimic. Response 3. According to the advice, the language about alcohol oxidase mimic have been removed from the manuscript.
Comment 4. In line 179, it is mentioned that Au NPs did not exhibit catalase-like activity. However, many previous studies have shown that Au NPs have considerable catalase-like activity : 765-773.). How to explain the contradiction between this study and previous studies? Response 4. Thank you for your insightful question. We have read these papers carefully, and thought over the point you mentioned. First, the catalytic performance of nanoparticles prepared by different synthesis methods will be different. Changing the synthesis methods is a common way to improve the activity or selectivity of catalysts. Second, at present, there is no standard to measure the catalytic activities of nanozymes. Due to the lack of comparison with representative materials, it is difficult to accurately measure the catalytic activities of Au NPs in these articles. For a catalyst with poor activity, the catalytic results can be trickily obtained by increasing the amount of catalyst and prolonging the reaction time. Third, a previous report 37-44, DOI:10.1016/j.biomaterials.2015.01.012) confirms our results and the two points mentioned above. The researcher compared the catalytic activities of Pt, Pd, Ag and Au for the decomposition of H2O2, and found that Au and Ag had no detectable catalytic effect compared with Pt and Pd. Therefore, we conclude that Au is not a good catalyst for the decomposition of H2O2 in our work. Comment 5. The Method section lacks specific experimental methods for detecting catalase activity. Response 5. For detecting catalase-like activity, 50 μL of as-prepared noble metal NPs and 50 μL of H2O2 (1 M) were sequentially added into a vial containing 900 μL of 100 mM buffer solution: acetate buffer (pH = 5-6), PBS (pH = 7-8), carbonate buffer (pH = 9-10). The decomposition of H2O2 was measured by the decrease of absorbance at 240 nm. We have added this information to the revised Method section.
[fig] Figure 2d: Mechanism of glucose oxidation catalyzed by GOD. [/fig]
[fig] Figure response: The catalase-like activities of Pt, Pd, Ag, and Au at pH = 7.4. (from Biomaterials 48 (2015) 37-44). [/fig]
|
The Effect of Psyllium Husk on Intestinal Microbiota in Constipated Patients and Healthy Controls
Psyllium is a widely used treatment for constipation. It traps water in the intestine increasing stool water, easing defaecation and altering the colonic environment. We aimed to assess the impact of psyllium on faecal microbiota, whose key role in gut physiology is being increasingly recognised. We performed two randomised, placebo-controlled, double-blinded trials comparing 7 days of psyllium with a placebo (maltodextrin) in 8 healthy volunteers and 16 constipated patients respectively. We measured the patients' gastrointestnal (GI) transit, faecal water content, short-chain fatty acid (SCFA) and the stool microbiota composition. While psyllium supplement had a small but significant effect on the microbial composition of healthy adults (increasing Veillonella and decreasing Subdoligranulum), in constipated subjects there were greater effects on the microbial composition (increased Lachnospira, Faecalibacterium, Phascolarctobacterium, Veillonella and Sutterella and decreased uncultured Coriobacteria and Christensenella) and alterations in the levels of acetate and propionate. We found several taxa to be associated with altered GI transit, SCFAs and faecal water content in these patients. Significant increases in three genera known to produce butyrate, Lachnospira, Roseburia and Faecalibacterium, correlated with increased faecal water. In summary, psyllium supplementation increased stool water and this was associated with significant changes in microbiota, most marked in constipated patients. their constituent sugars as an energy source[1][2][3][4][5], and therefore, psyllium can be considered to have prebiotic potential. In general, the health-promoting effects of prebiotics include supporting the growth of bacteria beneficial to the host and increasing the production of short-chain fatty acids (SCFA) such as butyrate and propionate previously shown to be positive for colonic health[6]. Another property of psyllium is that it is capable of retaining water in the small intestine, and thereby, increasing water flow into the ascending colon. The resulting increase in the fluidity of colonic content may explain the success of psyllium husk in treating constipation[7][8][9]. In addition to the relief of symptoms through softening of stool and increasing stool frequency, the increase of free water alters the environmental conditions of the colon.The effects of psyllium husk on host physiology have already been described, however, the effect that it has on the microbial ecosystem has not been well characterised. The effect of psyllium on microbial composition has been studied in an in vitro model mimicking the healthy intestine where it was shown that the microbial composition was not strongly affected.[10] However, there was an indication that the bacterial fermentation of fibres increased colonic levels of SCFAs. Moreover, a recent trial observed that psyllium supplementation protected mice from colitis by reducing the inflammatory response. This response was found to be largely microbiota dependent[11].In this study, we aimed to determine the effect of psyllium husk supplementation on the intestinal microbiota of both healthy subjects and patients with chronic idiopathic constipation using two separate randomised, placebo-controlled intervention studies. The samples for this study come from two linked, published clinical trials[12]where the effect of psyllium husk on colonic volume and intestinal water content was tested on first healthy adult subjects and then adults with chronic constipation. Here, we built on the results reported previously by concentrating on the microbial composition alterations introduced by psyllium supplementation. We aimed to understand how altered gut environmental conditions induced by psyllium husk affected the microbiota composition.Here, we showed that in healthy controls there was a very small change in the microbiota, however, there was a larger change in the microbial composition of the constipated patients.ResultsDifferences Between Healthy and Constipated Patients at BaselineThe demographics of the participants (study 1: heathy subjects, study 2: adults with chronic constipation) in the two already published clinical trials[12]are shown inTable 1. There were significant differences in age and gender distribution of the participants. Therefore, the effects of psyllium husk on the intestinal microbiota in the two studies were addressed separately and microbial composition differences between the studies were only compared at baseline.
# Introduction
Psyllium husk, derived from the seeds of Plantago ovata, consists of highly branched and gel-forming arabinoxylan, a polymer rich in arabinose and xylose which has limited digestibility in humans. However, several members of the intestinal microbiota can utilize these oligosaccharides and The microbial composition of healthy subjects and constipated patients differed at baseline. The health status accounted for 12% of the total microbial variation [fig_ref] Figure 1: Principal co-ordinate analysis [/fig_ref] , p = 0.006). There were 10 taxa contributing to this difference, including 14.0-fold higher abundance of Desulfovibrio and 9.4-fold higher levels of Adlercreutzia in constipated subjects compared to the healthy controls. Moreover, the levels of Sutterella and Lachnospira were significantly decreased at baseline in constipated patients (6.7 and 4.6-fold respectively; [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref]. Moreover, there were significant differences in the amount of faecal SCFAs where the constipated subjects had 2.2-fold higher level of faecal acetate than the healthy controls [fig_ref] Table 1: The patient demographics and amount of short chain fatty acids and microbial... [/fig_ref]. The microbial composition of healthy subjects and constipated patients differed at baseline. The health status accounted for 12% of the total microbial variation [fig_ref] Figure 1: Principal co-ordinate analysis [/fig_ref] , p = 0.006). There were 10 taxa contributing to this difference, including 14.0-fold higher abundance of Desulfovibrio and 9.4-fold higher levels of Adlercreutzia in constipated subjects compared to the healthy controls. Moreover, the levels of Sutterella and Lachnospira were significantly decreased at baseline in constipated patients (6.7 and 4.6-fold respectively; [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref]. Moreover, there were significant differences in the amount of faecal SCFAs where the constipated subjects had 2.2-fold higher level of faecal acetate than the healthy controls [fig_ref] Table 1: The patient demographics and amount of short chain fatty acids and microbial... [/fig_ref]. The temporal stability and resistance to altered environmental conditions are key characteristics of a healthy intestinal microbiota [bib_ref] Intestinal microbiota in healthy adults: Temporal analysis reveals individual and common core..., Jalanka-Tuovinen [/bib_ref]. Dietary supplementation with the carbohydrate inulin has been shown to promote expansion of Bifidobacteria in the stool in two recent studies [bib_ref] A low FODMAP diet is associated with changes in the microbiota and..., Sloan [/bib_ref]. Inulin has a structure more readily accessible to bacteria for digestion than psyllium, which in comparative studies tends to be less fermentable than gums but more fermentable that cellulose [bib_ref] In vitro degradation and fermentation of three dietary fiber sources by human..., Bliss [/bib_ref]. Therefore, it was not surprising that the composition of relatively few taxa was affected by the psyllium supplementation. The only detected change was a significant increase of Veillonella (fold change (fc) = 3.7, q = 0.02) and significant decrease of Subdoligranulum (fc = −2, q = 0.01, [fig_ref] Figure 2: Microbial changes introduced by the psyllium supplementation to the intestinal microbiota of... [/fig_ref] with a higher dose of psyllium. There was no significant change from baseline in the levels of SCFAs measured at each study point [fig_ref] Table 1: The patient demographics and amount of short chain fatty acids and microbial... [/fig_ref]. The temporal stability and resistance to altered environmental conditions are key characteristics of a healthy intestinal microbiota [bib_ref] Intestinal microbiota in healthy adults: Temporal analysis reveals individual and common core..., Jalanka-Tuovinen [/bib_ref]. Dietary supplementation with the carbohydrate inulin has been shown to promote expansion of Bifidobacteria in the stool in two recent studies [bib_ref] A low FODMAP diet is associated with changes in the microbiota and..., Sloan [/bib_ref]. Inulin has a structure more readily accessible to bacteria for digestion than psyllium, which in comparative studies tends to be less fermentable than gums but more fermentable that cellulose [bib_ref] In vitro degradation and fermentation of three dietary fiber sources by human..., Bliss [/bib_ref]. Therefore, it was not surprising that the composition of relatively few taxa was affected by the psyllium supplementation. The only detected change was a significant increase of Veillonella (fold change (fc) = 3.7, q = 0.02) and significant decrease of Subdoligranulum (fc = −2, q = 0.01, [fig_ref] Figure 2: Microbial changes introduced by the psyllium supplementation to the intestinal microbiota of... [/fig_ref] with a higher dose of psyllium. There was no significant change from baseline in the levels of SCFAs measured at each study point [fig_ref] Table 1: The patient demographics and amount of short chain fatty acids and microbial... [/fig_ref].
## Intestinal microbiota and scfas of constipated patients were altered with psyllium supplementation
In constipated patients, the effect of psyllium supplementation was greater than in healthy controls. There were seven bacterial genera with significantly altered abundance when comparing the time point of psyllium supplementation to other time points (detailed in [fig_ref] Table 3: Bacterial taxa affected by the psyllium treatment of constipated patients [/fig_ref]. The psyllium husk increased the levels of Lachnospira (fc = 1.71), Faecalibacterium (fc = 2.71), Phascolarctobacterium (fc = 3.62), Veillonella (fc = 2.30) and Sutterella (fc = 2.13) when compared to baseline, whereas the levels of uncultured Coriobacteria (fc = −2.36) and Christensenella (fc = −1.83) decreased. Additionally, we detected a statistically significant difference in the amount of acetate and propionate when psyllium was consumed when compared to the baseline levels [fig_ref] Table 1: The patient demographics and amount of short chain fatty acids and microbial... [/fig_ref].
## Intestinal microbiota and scfas of constipated patients were altered with psyllium supplementation
In constipated patients, the effect of psyllium supplementation was greater than in healthy controls. There were seven bacterial genera with significantly altered abundance when comparing the time point of psyllium supplementation to other time points (detailed in [fig_ref] Table 3: Bacterial taxa affected by the psyllium treatment of constipated patients [/fig_ref]. The psyllium husk increased the levels of Lachnospira (fc = 1.71), Faecalibacterium (fc = 2.71), Phascolarctobacterium (fc = 3.62), Veillonella (fc = 2.30) and Sutterella (fc = 2.13) when compared to baseline, whereas the levels of uncultured Coriobacteria (fc = −2.36) and Christensenella (fc = −1.83) decreased. Additionally, we detected a statistically significant difference in the amount of acetate and propionate when psyllium was consumed when compared to the baseline levels [fig_ref] Table 1: The patient demographics and amount of short chain fatty acids and microbial... [/fig_ref].
## The impact of increased transit on intestinal microbiota
One aim of the study was to find associations between altered microbial abundance and whole gut transit time introduced by psyllium consumption. These associations were calculated with linear mixed models, taking into account that there were multiple samples from one subject (detailed in materials and methods). In both study groups of healthy and constipated subjects, an increased amount of Sutterella genera was associated with faster transit [fig_ref] Figure 3: Associations between whole gut transit time and bacterial abundance measured from all... [/fig_ref]. Additionally, in the constipated patients, there were altogether 10 additional taxa associated with transit time [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref] ; these included Odoribacter (q = 0.03) and Christensenella (q = 0.05) with negative and positive associations to increased transit time, respectively.
## The impact of increased transit on intestinal microbiota
One aim of the study was to find associations between altered microbial abundance and whole gut transit time introduced by psyllium consumption. These associations were calculated with linear mixed models, taking into account that there were multiple samples from one subject (detailed in materials and methods). In both study groups of healthy and constipated subjects, an increased amount of Sutterella genera was associated with faster transit [fig_ref] Figure 3: Associations between whole gut transit time and bacterial abundance measured from all... [/fig_ref]. Additionally, in the constipated patients, there were altogether 10 additional taxa associated with transit time [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref] ; these included Odoribacter (q = 0.03) and Christensenella (q = 0.05) with negative and positive associations to increased transit time, respectively.
## Associations between intestinal microbiota and faecal water content and scfas
In line with previous reports, psyllium supplementation increased the participants' stool water content in both studies [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref]. Additionally, we showed that this same measure could be associated with changes in the intestinal microbiota in both study groups. The detailed associations can be found in [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref] and 3 for constipated and healthy subjects respectively. In the constipated patient cohort, we found that several Actinobacteria taxa (genera Actinomyces, Bifidobacterium, Asaccharobacter and uncultured Coriobacter) showed significant negative associations with increased stool water content. In addition, three genera known to produce butyrate, Lachnospira, Roseburia and Faecalibacterium, were significantly increased with increased faecal water content [fig_ref] Figure 4: Associations between faecal water content and intestinal genera in constipated subjects at... [/fig_ref]. In the healthy subjects, only Anaerococcus showed a negative association to faecal water content.
## Associations between intestinal microbiota and faecal water content and scfas
In line with previous reports, psyllium supplementation increased the participants' stool water content in both studies [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref]. Additionally, we showed that this same measure could be associated with changes in the intestinal microbiota in both study groups. The detailed associations can be found in [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref] for constipated and healthy subjects respectively. In the constipated patient cohort, we found that several Actinobacteria taxa (genera Actinomyces, Bifidobacterium, Asaccharobacter and uncultured Coriobacter) showed significant negative associations with increased stool water content. In addition, three genera known to produce butyrate, Lachnospira, Roseburia and Faecalibacterium, were significantly increased with increased faecal water content [fig_ref] Figure 4: Associations between faecal water content and intestinal genera in constipated subjects at... [/fig_ref]. In the healthy subjects, only Anaerococcus showed a negative association to faecal water content.
Short-chain fatty acids are the end products of bacterial fermentation of dietary fibres. We found a significant association between longer transit time and lower levels of SCFAs. Additionally, we noted that the abundance of 16 bacterial genera were associated with the amount of faecal SCFAs in the constipated patients [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref]. A higher amount of acetate was associated with increased levels of two genera from Lachnospiraceaea family, Roseburia (q = 0.03) and Lachnospira (q = 0.03) as well as Faecalibacterium (q = 0.0001; [fig_ref] Figure 5: Significant associations between acetate and the abundance of microbial taxa in constipated... [/fig_ref]. Moreover, the latter two showed positive associations between the levels of propionate (Lachnospira q = 0.002, Faecalibacterium q = 0.04). In healthy subjects, there were 10 bacterial taxa with associations to the altered levels of faecal SCFAs [fig_ref] Table 3: Bacterial taxa affected by the psyllium treatment of constipated patients [/fig_ref]. These included the positive associations of Leuconostoc (q = 0.03) with butyrate levels and the positive association of Blautia (q = 0.07) with acetate levels.
# Discussion
In this study, we utilized two clinical trials to investigate the effects of psyllium husk ingestion on the faecal microbiota composition of both healthy controls and patients with idiopathic constipation [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref]. We showed that the environmental changes (change in transit time, faecal water content and the concentration of SCFAs) introduced by the psyllium husk were substantial in both patient groups. However, the healthy subjects showed very little change in their faecal microbial composition, whereas several key members of the microbial ecosystem were altered in the constipated patients. Additionally, we noted that the abundance of 16 bacterial genera were associated with the amount of faecal SCFAs in the constipated patients [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref]. A higher amount of acetate was associated with increased levels of two genera from Lachnospiraceaea family, Roseburia (q = 0.03) and Lachnospira (q = 0.03) as well as Faecalibacterium (q = 0.0001; [fig_ref] Figure 5: Significant associations between acetate and the abundance of microbial taxa in constipated... [/fig_ref]. Moreover, the latter two showed positive associations between the levels of propionate (Lachnospira q = 0.002, Faecalibacterium q = 0.04). In healthy subjects, there were 10 bacterial taxa with associations to the altered levels of faecal SCFAs [fig_ref] Table 3: Bacterial taxa affected by the psyllium treatment of constipated patients [/fig_ref]. These included the positive associations of Leuconostoc (q = 0.03) with butyrate levels and the positive association of Blautia (q = 0.07) with acetate levels. Additionally, we noted that the abundance of 16 bacterial genera were associated with the amount of faecal SCFAs in the constipated patients [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref]. A higher amount of acetate was associated with increased levels of two genera from Lachnospiraceaea family, Roseburia (q = 0.03) and Lachnospira (q = 0.03) as well as Faecalibacterium (q = 0.0001; [fig_ref] Figure 5: Significant associations between acetate and the abundance of microbial taxa in constipated... [/fig_ref]. Moreover, the latter two showed positive associations between the levels of propionate (Lachnospira q = 0.002, Faecalibacterium q = 0.04). In healthy subjects, there were 10 bacterial taxa with associations to the altered levels of faecal SCFAs [fig_ref] Table 3: Bacterial taxa affected by the psyllium treatment of constipated patients [/fig_ref]. These included the positive associations of Leuconostoc (q = 0.03) with butyrate levels and the positive association of Blautia (q = 0.07) with acetate levels.
# Discussion
In this study, we utilized two clinical trials to investigate the effects of psyllium husk ingestion on the faecal microbiota composition of both healthy controls and patients with idiopathic constipation [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref]. We showed that the environmental changes (change in transit time, faecal water content and the concentration of SCFAs) introduced by the psyllium husk were substantial in both patient groups. However, the healthy subjects showed very little change in their faecal microbial composition, whereas several key members of the microbial ecosystem were altered in the constipated patients.
# Discussion
In this study, we utilized two clinical trials to investigate the effects of psyllium husk ingestion on the faecal microbiota composition of both healthy controls and patients with idiopathic constipation [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref]. We showed that the environmental changes (change in transit time, faecal water content and the concentration of SCFAs) introduced by the psyllium husk were substantial in both patient groups. However, the healthy subjects showed very little change in their faecal microbial composition, whereas several key members of the microbial ecosystem were altered in the constipated patients.
There were detectable microbial differences between healthy and constipated patients at baseline, including significantly higher levels of Desulfovibrio ssp. in the constipated patients. Interestingly, in a recent study, a member of the Desulfovibrio genus was shown to reduce transit in an in vivo model [bib_ref] Sulfate-reducing bacteria slow intestinal transit in a bismuth-reversible fashion in mice, Ritz [/bib_ref]. It has also been shown that this taxon produces hydrogen sulphide, which is known to decrease inflammation in the GI tract and to inhibit motility [bib_ref] Physiological implications of hydrogen sulfide: A whiff exploration that blossomed, Wang [/bib_ref] [bib_ref] Hydrogen sulphide as a signalling molecule regulating physiopathological processes in gastrointestinal motility, Jimenez [/bib_ref]. The association of Desulfovibrio spp. with constipated patients is a new finding that may be followed up by studies on the relevance of this organism to GI health and its effect on orchestrating motility.
Another genus level taxon that showed significant difference between the study groups was Lachnospira, with an over four-fold higher level in the healthy than in the constipated subjects. Additionally, we detected a significant increase of this taxon in these patients after psyllium supplementation and we found positive associations between increased faecal water content and acetate levels. It is known that several members from Lachnospira genus are capable of producing lactate and acetate. Lactate may be further metabolized into butyrate or propionate. It has been shown that low concentration of butyrate contributes to constipation by inhibiting mucin secretion [bib_ref] Potential beneficial effects of butyrate in intestinal and extraintestinal diseases, Canani [/bib_ref]. Therefore, decreased levels of these SCFAs produces could contribute to the disease state of constipation.
Psyllium husk has long been used as a treatment for constipation; however, the changes it produces in the composition of the intestinal microbiota have so far not been addressed. We showed that the intestinal microbiota composition of healthy individuals was less affected by the psyllium supplementation than those of the constipated patients. This is in line with previous findings on psyllium supplementation [bib_ref] Starch-entrapped microsphere fibers improve bowel habit but do not exhibit prebiotic capacity..., Rasmussen [/bib_ref] as well as the general understanding where the microbial composition of healthy subjects is regarded to be resistant to environmental changes [bib_ref] Diversity, stability and resilience of the human gut microbiota, Lozupone [/bib_ref]. It should be noted that the constipated patients had a larger environmental change, as reflected by the change in colonic T1, an Magnetic Resonance Imaging (MRI) time constant reflecting fluidity of the colonic chyme [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref] , than the healthy subjects; the degree of microbial change may be a reflection of this. We showed that in both studies there was an increase of Veillonella, an organism typically more abundant in the small intestine [bib_ref] Diversity of human small intestinal Streptococcus and Veillonella populations, Van Den Bogert [/bib_ref]. There were also changes in other small intestinal taxa [bib_ref] Duodenal microbiota composition and mucosal homeostasis in pediatric celiac disease, Cheng [/bib_ref] , such as Sutterella (belonging to the Proteobacteria phylum), which increased in abundance after psyllium supplementation in the constipated subjects. Sutterella bacteria are slow growing organisms [bib_ref] Mucosal Prevalence and Interactions with the Epithelium Indicate Commensalism of Sutterella spp, Hiippala [/bib_ref] , and therefore, their higher abundance with faster transit could be an indication that these bacteria are washed out of the small intestine and become detectable in greater numbers in faeces. Therefore, we hypothesize that the increase in both of these bacterial taxa is secondary to the supplementation of psyllium and change in the colonic environment rather than increased colonic growth of the bacteria belonging to these taxa.
The supplementation of psyllium significantly increased the numbers of commensals such as Phascolarctobacteriumand Faecalibacterium [bib_ref] Action and function of Faecalibacterium prausnitzii in health and disease, Ferreira-Halder [/bib_ref] in the constipated patients. These same organisms have also been shown to have increased abundance after consumption of a polydextrose fibre [bib_ref] 454 pyrosequencing reveals a shift in fecal microbiota of healthy adult men..., Hooda [/bib_ref] ; Faecalibacterium and Roseburia have also been previously associated with faster colonic transit [bib_ref] Relationship Between Microbiota of the Colonic Mucosa vs Feces and Symptoms, Colonic..., Parthasarathy [/bib_ref] and Faecalibacterium with loose stools [bib_ref] Gut microbiota composition associated with stool consistency, Tigchelaar [/bib_ref]. In our study, the increase in Faecalibacterium was also associated with higher stool water content and the increase of both Phascolarctobacterium and Faecalibacterium with larger amounts of acetate in stools. Interestingly, Faecalibacterium converts sugars and acetate into butyrate-an important molecule for intestinal health used by the colonocytes as an energy resource. The favourable change in the acetate concentration may be explained by the increase in abundance of these genera.
Another bacterial genus affected by the psyllium supplementation in the constipated patients was Christensenella, which decreased almost two-fold in abundance after psyllium treatment. Previously, Christensenella had been associated with hard stools [bib_ref] Gut microbiota composition associated with stool consistency, Tigchelaar [/bib_ref] , and interestingly, we were able to show a similar trend, since in our study, this genus was negatively associated with increased faecal water content.
In conclusion, we showed that consumption of psyllium husks introduced small but significant changes in the intestinal microbiota of both healthy and constipated patients. This change was more pronounced in the constipated patients. Organisms associated with intestinal short-chain fatty acid production such as Faecalibacterium ssp. were increased, an effect that suggests a potential health benefit from psyllium supplementation. Several candidate mechanisms for such an effect were present, such as changes in gut transit and increased intestinal water content. These data support potential health benefits for such mechanisms in addition to their role in the relief of constipation symptoms.
# Materials and methods
## Study design
This study consisted of two separate, published clinical trials [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref] where the effect on psyllium husk on MRI parameters of gastrointestinal physiology was tested on adults without digestive disorders (study 1) and adults with chronic constipation (study 2). The protocols were approved by institutional and national review boards respectively and registered on www.clinicaltrials.gov (NCT01805999 and NCT02144376). All participants gave written informed consent.
A detailed description of these studies has been published previously [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref] , thus, the methodology is here only given in brief. The study 1, in adults without digestive disorders (n = 9; one subject failed to provide stool samples), was a double-blinded, placebo-controlled three-period, three-treatment crossover trial. Treatments were taken in a random order, each taken for a six-day period. Subjects took 14 g of powder three times daily, either 14 g maltodextrin, 14 g Metamucil (providing 7 g psyllium), or a 50:50 mixture 7 g maltodextrin and 7 g Metamucil (providing 3.5 g psyllium). Treatments were separated by washout periods of one week. Four faecal samples were collected, one at baseline and one after each treatment period [fig_ref] Figure 6: Study outline [/fig_ref].
The study 2 included patients (n = 16) suffering from chronic constipation, which was defined by meeting the Rome III criteria for either functional constipation or constipation-predominant irritable bowel syndrome. This study was a double-blind, placebo-controlled, two-period, two-treatment study, using only maltodextrin (placebo) and the higher dose (21 g/day) of psyllium. In order to avoid any carryover effects both treatment periods were preceded with ≥10 days of usual laxative use; then, 8 days without laxatives other than rescue therapy. Rescue therapy (oral bisacodyl 5 mg) was permitted in patients who had not opened their bowels for 3 days and were experiencing distressing symptoms, but not in the 48 h before stool sample collection. Four faecal samples were collected: at baseline, after each of the treatments and after the washout period in between treatments [fig_ref] Figure 6: Study outline [/fig_ref].
This study consisted of two separate, published clinical trials [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref] where the effect on psyllium husk on MRI parameters of gastrointestinal physiology was tested on adults without digestive disorders (study 1) and adults with chronic constipation (study 2). The protocols were approved by institutional and national review boards respectively and registered on www.clinicaltrials.gov (NCT01805999 and NCT02144376). All participants gave written informed consent.
A detailed description of these studies has been published previously [bib_ref] Demonstration of differences in colonic volumes, transit, chyme consistency, and response to..., Major [/bib_ref] , thus, the methodology is here only given in brief. The study 1, in adults without digestive disorders (n = 9; one subject failed to provide stool samples), was a double-blinded, placebo-controlled three-period, three-treatment crossover trial. Treatments were taken in a random order, each taken for a six-day period. Subjects took 14 g of powder three times daily, either 14 g maltodextrin, 14 g Metamucil (providing 7 g psyllium), or a 50:50 mixture 7 g maltodextrin and 7 g Metamucil (providing 3.5 g psyllium). Treatments were separated by washout periods of one week. Four faecal samples were collected, one at baseline and one after each treatment period [fig_ref] Figure 6: Study outline [/fig_ref].
The study 2 included patients (n = 16) suffering from chronic constipation, which was defined by meeting the Rome III criteria for either functional constipation or constipation-predominant irritable bowel syndrome. This study was a double-blind, placebo-controlled, two-period, twotreatment study, using only maltodextrin (placebo) and the higher dose (21 g/day) of psyllium. In order to avoid any carryover effects both treatment periods were preceded with ≥10 days of usual laxative use; then, 8 days without laxatives other than rescue therapy. Rescue therapy (oral bisacodyl 5 mg) was permitted in patients who had not opened their bowels for 3 days and were experiencing distressing symptoms, but not in the 48 h before stool sample collection. Four faecal samples were collected: at baseline, after each of the treatments and after the washout period in between treatments [fig_ref] Figure 6: Study outline [/fig_ref].
## Measurements of transit (waps)
Subjects swallowed five identical transit markers: cylinder-shaped inert capsules containing 0.4 mL 15 µM gadoteric acid, a positive MRI contrast agent. Ingestion was confirmed in patients by direct observation or via a time-stamped video. The weighted average position score of the transit markers on an MRI scan 24 h after ingestion (WAPS24) was calculated using the formula (sum of the segment
## Measurements of transit (waps)
Subjects swallowed five identical transit markers: cylinder-shaped inert capsules containing 0.4 mL 15 µM gadoteric acid, a positive MRI contrast agent. Ingestion was confirmed in patients by direct observation or via a time-stamped video. The weighted average position score of the transit markers on an MRI scan 24 h after ingestion (WAPS24) was calculated using the formula (sum of the segment number X the number of markers in each segment divided by the total number of segments), as described previously [bib_ref] Novel MRI tests of orocecal transit time and whole gut transit time:..., Chaddock [/bib_ref]. A higher score denotes slower whole gut transit. WAPS48 was based on the same calculation 48 h after marker ingestion. For healthy controls, transit was measured at the end of each of the three treatment periods and for constipated patients at the end of each of the two treatments.
## Analysis from faecal samples
## Quantification of scfa by gas chromatography-mass spectrometry (gc-ms)
0.5 g of faecal sample was suspended in 5 mL 0.5% phosphoric acid (H 3 PO 4 ), mixed and centrifuged at 25,000× g for 10 min. The SCFA-containing supernatant was filtered through cellulose acetate membrane with pore size 0.2 µm and 700 µL of this supernatant was mixed with 700 µL of ethyl acetate centrifuged 25,000× g for 10 min. 300 µL of the supernatant (ethyl acetate layer) was transferred to a clean tube and stored at −80 - C until analysis.
Separation and detection of the short chain fatty acids of interest was achieved using a Trace GC Ultra (Thermo Scientific, Manchester, UK) coupled with a DSQII mass spectrometer (Thermo Scientific). Separations were performed on a Zebron ZB-FFAP column (length 30 m, inner diameter 0.25 mm, and film thickness 0.25 µm; Phenomenex Inc., Macclesfield, UK). The initial oven temperature was set at 60 - C for 1 min then increased at 8 - C min −1 to 180 - C. Compound identification was achieved by matching with database mass spectra (NIST/EPA/NIH Mass Spectral Library, Version 2.0d, NIST, Gaithersburg, MD, USA). Identification was further verified by comparing with the retention times and mass spectra of authentic standards. Concentrations of analyte were calculated using 'Xcalibur' software (Thermo Scientific, UK). Retention time and specific ions used for quantification are detailed below for each analyte: acetic acid (7.10 min, m/z 60); propanoic acid (8.41 min, m/z 57); butyric acid (9.75 min, m/z 88). The recovery levels were determined on few selected samples; acetic acid 62% (sd = 3.9), propanoic acid 84% (sd = 4.2), butyric acid 95% (sd = 3.5).
## Faecal water
## Analysis of the intestinal microbiota
The V4-V5 region of the 16S rRNA gene was analysed with Illumina MiSeq platform. The obtained reads were pre-processed according to the Mare pipeline. In short, the forward reads were trimmed to 150 bps and those with bad quality were discarded. The resultant reads (in average 84,242 reads ranging from 8474 to 189,270 per sample) were clustered to OTUs with UPARSE and chimeras were removed [bib_ref] UPARSE: Highly accurate OTU sequences from microbial amplicon reads, Edgar [/bib_ref]. These were clustered against Silva database [bib_ref] The SILVA ribosomal RNA gene database project: Improved data processing and web-based..., Quast [/bib_ref] at assigning OTUs with 97% similarity. The sequencing data is publicly available in the European Nucleotide Archive (ENA, acc.no. PRJEB29397).
The total microbial load [bib_ref] Determination of bacterial load by real-time PCR using a broad-range (universal) probe..., Nadkarni [/bib_ref] and the abundance of methanogenic archaea [bib_ref] A real-time qPCR assay for the detection of the nifH gene of..., Johnston [/bib_ref] determined with qPCR as described previously.
# Statistical analysis
The effect of psyllium was calculated for both trials separately, since there were notable demographic differences between trial populations. Comparisons between subjects from the two trials were only undertaken at baseline to elucidate differences between the groups. The baseline differences in SCFA, stool water content, qPCR results and transit times between the patient groups were calculated either with t-test (parametric data) or Wilcox test (nonparametric data). For microbial diversity, we used the inverse Simpson's diversity index. Correlations between these data was analysed by using Spearman correlation.
All of the 16S rRNA gene sequencing data were analysed with in-house script from the Mare r-packageusing R (version 3.2.3). For visualization of the overall microbial composition differences at baseline we performed PCoA using Bray-Curtis dissimilarities. For analysing the effect of health status on the overall microbiota composition, we used a MANOVA test. Differences in bacterial abundances and associations between bacterial taxa and SCFAs, stool water content and transit times within each study were estimated with generalized linear mixed models from all available data points. In both models, the obtained read counts were modelled as a function of time point, using subject ID as a random factor and the total number of reads per sample as an offset, assuming negative binomial distribution. The resultant p-values were corrected for multiple comparisons with false discovery rate (FDR), and subsequent q-values bellow 0.1 were considered to be significant [bib_ref] A direct approach to false discovery rates, Storey [/bib_ref].
Supplementary Materials: Supplementary materials can be found at http://www.mdpi.com/1422-0067/20/2/ 433/s1. [fig_ref] Table 1: The patient demographics and amount of short chain fatty acids and microbial... [/fig_ref]. Differences in the amount of short-chain fatty acids (SCFAs) during the intervention in both healthy and constipated. Significant difference against the baseline level (t-test, p < 0.05) indicated with an asterisk, standard deviation in brackets. [fig_ref] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and... [/fig_ref]. Associations between the intestinal microbiota composition and whole gut transit time (Weighted Average Position Score after 24 h (WAPS-24) and 48 h (WAPS-48)), relative stool water content and SCFAs, introduced by psyllium consumption in constipated patients. Each cell contains information of the direction of the association (+/-) and the FDR corrected p-value for this association. If the cell is empty, no significant association was found. [fig_ref] Table 3: Bacterial taxa affected by the psyllium treatment of constipated patients [/fig_ref]. Associations between the intestinal microbiota composition and whole gut transit time (WAPS-24), relative stool water content and SCFAs, introduced by psyllium consumption in healthy subjects. Each cell contains information of the direction of the association (+/-) and the FDR corrected p-value for this association. If the cell is empty, no significant association was found.
[fig] Figure 1: Principal co-ordinate analysis (PCoA) of the baseline microbial differences between the two study populations. [/fig]
[fig] Figure 2: Microbial changes introduced by the psyllium supplementation to the intestinal microbiota of healthy adults. [/fig]
[fig] Figure 3: Associations between whole gut transit time and bacterial abundance measured from all available data points: (A) relative abundance of genus Sutterella and transit time measure (weighted average position score, Weighted Average Position Score (WAPS) measure 24 h after ingestion of transit markers, smaller number indicates longer transit) in healthy subjects; (B) relative abundance of genus Sutterella and transit (WAPS) measured 48 h after ingestion of transit markers in constipated subjects. [/fig]
[fig] Figure 4: Associations between faecal water content and intestinal genera in constipated subjects at all data points. Short-chain fatty acids are the end products of bacterial fermentation of dietary fibres. We found a significant association between longer transit time and lower levels of SCFAs. The amounts of acetate (cor = −0.59, p = 0.0002), butyrate (cor = −0.42, p = 0.015) and propionate (cor = −0.59, p = 0.0002) all showed a significant correlation with transit time (measured with Weighted Average Position Score (WAPS) after 48 h) in constipated patients. There were no significant correlations between SCFAs and transit times in healthy subjects. [/fig]
[fig] Figure 5: Significant associations between acetate and the abundance of microbial taxa in constipated patients. [/fig]
[fig] Figure 6: Study outline. The arrows indicate faecal sample collection points. [/fig]
[fig] Author: Contributions: Conceptualization, G.M., R.S.; methodology, G.M., R.S.; investigation, J.J., G.M., G.S., A.N., C.K., I.S.S., J.M.J.; resources, W.M.d.V., R.S.; data curation, J.J., G.M.; writing-original draft preparation, J.J., W.M.d.V., R.S.; writing-review and editing, J.J., G.M., K.M., G.S., A.N., C.K., I.S.S., J.M.J., W.M.d.V., R.S.; supervision, R.S.; project administration, G.M., A.N., R.S.; funding acquisition, W.M.d.V., R.S. Funding: This research was funded by the Finnish Academy (grant no. 0313471-7 for J.J.). The clinical trials were collaborative studies undertaken by the Nottingham Digestive Diseases Centre and NIHR Nottingham Biomedical Research Centre at Nottingham University Hospitals NHS Trust, the University of Nottingham, funded by Ironwood Pharmaceuticals Inc., Cambridge, MA, USA. [/fig]
[table] Table 1: The patient demographics and amount of short chain fatty acids and microbial content at baseline. The standard deviations for each number show in brackets. [/table]
[table] Table 2: Relative abundance of bacterial taxa showing a statistical difference between healthy and constipated patients at baseline. Fc-fold change of taxa in constipated versus healthy subjects. [/table]
[table] Table 3: Bacterial taxa affected by the psyllium treatment of constipated patients. Entries represent average relative abundance. Those time points with statistically significantly different bacterial abundance from the psyllium supplementation are indicated in bold (q-values below 0.05). [/table]
|
Recent advances in M13 bacteriophage-based optical sensing applications
Recently, M13 bacteriophage has started to be widely used as a functional nanomaterial for various electrical, chemical, or optical applications, such as battery components, photovoltaic cells, sensors, and optics. In addition, the use of M13 bacteriophage has expanded into novel research, such as exciton transporting. In these applications, the versatility of M13 phage is a result of its nontoxic, self-assembling, and specific binding properties. For these reasons, M13 phage is the most powerful candidate as a receptor for transducing chemical or optical phenomena of various analytes into electrical or optical signal. In this review, we will overview the recent progress in optical sensing applications of M13 phage. The structural and functional characters of M13 phage will be described and the recent results in optical sensing application using fluorescence, surface plasmon resonance, Förster resonance energy transfer, and surface enhanced Raman scattering will be outlined. which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
# Introduction
In general, a bio-chemical or a bio-optical sensor is a device to detect chemical or optical change in its biological medium. When the chemical or physical phenomena are measured, the quantitative information for the chemical and/or physical state within medium are provided by converting that into electrical or optical signal. Thus, a bio-chemical or a bio-optical sensor is a kind of a transducer using biomaterials that is used as a receptor for detection or measurement. Recently, bio-chemical or -optical sensors monitoring various substance or chemical constituent that is of interest such as nitro compound, peptide, nucleic acid, polymer, toxin, neurochemicals have been developed. Alongside with these advances in biosensor applications, receptor materials that make it possible to transduce biological phenomena have also been extensively investigated in many fields. In these sensors, receptors are either integrated within or closely associated with a transducer interface providing the corresponding output. In addition, transducers normally do not have specificity against a target analyte. For that reason, the development of a sensor that is able to selectively detect target analytes is required. Moreover, target-specificity is also an essential factor in the novel sensor development.
Recently, due to its nontoxic, self-assembling and specific binding properties, bacteriophage has been proven to be useful for the detection of target analytes in biomaterials by bio-chemical or bio-optical sensing applications. In addition, surface properties of bacteriophage, which is suitable as a receptor for the development of target-specific bio-sensor, can be controlled through genetic engineering. Employing this advantage, M13 bacteriophage (M13 phage) has expanded its use into novel research area, such as electrode, solar cell, environmental monitoring, plasmonics, cancer diagnosis, cell imaging, and functional device. In this review, we will focus on the application of M13 phage as bio-optical sensor. Initially, we will explain the structural and functional characteristics of M13 phage. Next, the recent progress in optical sensing application of M13 phage, such as immunofluorescence assay, surface plasmon resonance (SPR), Förster resonance energy transfer (FRET), and surface enhanced Raman scattering (SERS) will be introduced. The simple physical and/or optical concept of the corresponding sensing application will also be outlined.
## M13 phage for the immunofluorescence assay application
As mentioned above, M13 phage has often been used to manage various functional nanomaterials. In particular, since the shape of M13 phage is welldefined and can be modified genetically and chemically through the phage display technique to reveal functional peptides, M13 phage is particularly useful for various optical applications. Structurally, M13 phage has a cylindrical shape, 880 nm in length and 6.6 nm in diameter (see . The single stranded DNA is covered with cylindrical coat made up of 2700 copies of the major coat protein (pVIII), and there are five copies each of the minor protein (pIII and pVI) and other minor proteins (pVII and pIX) at both ends of the cylinder, respectively. Physically, pVIII protein has a helical structure and the shape of phage body is the fivefold rotational and twofold screw symmetry. This precisely defined structure can provide regular molecular spacing. According to recent results, the average distances between two neighboring N-termini of pVIII proteins are 3.2 and 2.4 nm. M13 phage can be used as a platform for functional materials (see , c). It is in fact a nanofiber whose surface chemistry can be site-specifically modified. By genetic engineering, a peptide or protein of M13 phage can be site-specifically displayed at the tip and/or along the body of the phage. This DNA engineering can leads to the single, double, or triple display with the combination of different peptides or proteins on a single phage. The genetically engineered phage can be used as novel phage-based array chips that are optically readable for cell proliferation and morphology assays. In practice, Yoo et al. have implemented phage-chip arrays using M13 phage that was engineered to display integrinbinding peptide (RGD) on its major coat proteins and/ or immobilize FGFb on its minor coat proteins. The authors also monitored cellular proliferation using this phage-chips array with engineered phage to display integrin-binding peptide RGD, a control RGE peptide, or no peptide (wild-type phage). In addition, the fluorescent labelling by conjugation between peptides and different nanomaterials, such as semiconductor nanocrystals and metal complexes, is possible (see [fig_ref] Figure 2 a: Fluorescent labelling of phage through conjugation with ruthenium complexes [/fig_ref] , b). This labelling of the phage envelope by assembling a characteristic protein with fluorescent material is particularly useful in chemical or biological applications. In addition, the labelling can overcome the photo-bleaching and selfquenching effects suppressing the fluorescence intensity of fluorophores. Recently, Ghosh et al. conducted the imaging using M13 phage in the detection of small, deep tumors for early diagnosis and surgical interventions (see [fig_ref] Figure 3 a: Schematic diagram for tumor-targeting of SBP-M13-SWNT probe [/fig_ref]. Through vivo fluorescence imaging, the authors have detected tumors such as ovarian cancer in mouse. For this implementation, Ghosh and co-authors have synthesized M13-stabilized SWNT (SBP (SPARC binding peptide)-M13-SWNT). Their SBP-M13-SWNT selectively detects secreted protein, acidic and rich in cysteines (SPARC)-expressing tumor nodules in ovarian cancer. In particular, as compared to fluorescein isothiocyanate (FITC), no noticeable loss of fluorescence of SBP-M13-SWNTs by photo-bleaching was observed during this period.
## M1bacteriophage based spr sensor applications
Recently, the plasmonic effect in a metallic surface has been widely studied as the method to enhance the optical signal. Surface plasmonic effect is known to enhance optical processes, such as the change in the reflective index of a molecule in a metal surface. The origin of this phenomenon is attributed to the oscillation of the free conduction electrons, that is, surface plasmon (SP), induced by the interaction with the external electromagnetic field. This unique phenomenon provides information for the fundamental interaction, such as the molecular binding between an antibody and a receptor. SPR is a surface field by the charge density oscillation in a metallic surface, where the free electrons of a conductor are responded by oscillating in resonance with the external electromagnetic wave. In the dielectric/ metal interface, the wave equation leads to a dispersion relation for SP which depends on the dielectric constant. Solving the Maxwell equations with a frequency dependent dielectric constant of metal and a dielectric material, the dispersion relation for SP is given by Eq. (1):
where k sp is the wave vector, ω is the optical frequency, c is the speed of light, ε d and ε m are the dielectric constants of dielectric material and metal, respectively. Since metals usually have negative dielectric constants in the UV/VIS range, this equation represents that the momentum of the SP mode is greater than that of a freespace photon at the same frequency: dielectric materials have the positive dielectric constant. This charge density wave is known as the surface plasmon polariton (SPP), which is a non-radiative electromagnetic wave that
[formula] (1) k sp = ω c ε d ε m (ω) ε d + ε m (ω) [/formula]
propagates in the direction parallel to dielectric material interface and a transverse magnetic (TM)-polarized that magnetic field vector is perpendicular to the direction of propagation. Since the SPR frequency is sensitive to the values of the dielectric constant of a dielectric material, the SP oscillation is very sensitive to any change in this interface. For example, the resonant coupling with SP and electromagnetic field gives rise to the noble phenomena, such as the enhancement of molecular absorption.
The commonly used SPR setups are shown in [fig_ref] Figure 4: Commonly used configurations of SPR sensors [/fig_ref]. In the Kretschmann configuration, the external electromagnetic field (optical wave) is generally incident a Schematic for structure of M13 phage. Reprinted with permission from Macmillan Publishers Limited and part of Springer Nature © 2016. b Fluorescent image of phage proteins. Reprinted with permission from Royal Society of Chemistry © 2016. c Fluorescent labelling of various genetic engineered M13 phage (RGD, RGE, and Wild). Reproduced with permission from American Chemical Society © 2016 through the prism coupler above the metal layer. Then, the photons induce an evanescent wave into the metal layer. While no transport of photons normally occurs through this wave, photons incident at a certain angle are able to pass through the field and excite surface plasmons (SPs) on the opposite side of the metal layer. Whenever analyte absorbs photon, a dip appears in the spectrum of the reflected light at that specific angle. This angle depends on the refractive index of the analyte and, is measured by a spectrometer.
In biosensor application, SPR characterizes ultrathin organic and biopolymer films at metal interfaces in a spectrally resolved manner by using its high surface sensitivity. SPR is particularly outstanding in biosensor applications, since SPR can measure the interaction of unlabeled molecules with surface-bound species in real time. In addition, it has advantages of the high specificity and affinity of antibodies to directly detect analytes without additional treatments, such as sample purification or enrichment, competitive immunoassay set-ups, or the use of labeled reagents. Recently, the use of the SPR imaging has been successfully demonstrated in biosensor applications of antigen-antibody reaction, annealing complementary pairs of oligonucleotides, and regulating biological function of DNA interactions.
M13 phage is also used as an antibody for various SPR applications. Since the production of antibody allowing the specific detection of a target material is time-consuming and expensive, M13 phage, which enables a specific recognition of target, can be used as an alternative. Recently, Karoonuthaisiri et al. reported the SPR assay based on M13 phage. For this application, M13 phage expressing 12-mer peptides was employed as a Salmonella-specific bacteriophage to detect the foodborne bacterium Salmonella. This Salmonella-specific phagebased SPR assay has a very low cross reactivity with other non-target foodborne pathogens and detection limits of 8.0 × 10 7 and 1.3 × 10 7 CFU/mL for one-time and five-time immobilized sensors, respectively. The phage-based SPR technique can be used to detect glyphosate (see . Glyphosate (N-(phosphonomethyl)glycine) is a herbicide that is used to remove weeds in farms, parks, and gardens. However, glyphosate as a potential endocrine disruptor can cause several environmental problems. Ding et al. reported a SPR biosensor enabling the detection of glyphosate in real time. The oligopeptide, which is prepared by phage display and has a sequence of TPFDLRPSSDTR, is used as a sensing element. It shows the high binding specificity for glyphosate (K D = 8.6 μM). For the SPR measurement, modified oligopeptide (TPFDLRPSSDTRGGGC) is immobilized on the gold sensor chip. In the detection of glyphosate, the oligopeptide-based SPR biosensor shows the sensitivity of 1.02 RU/μM and has the limit of detection of 0.58 μM. This SPR biosensor is also compatible to other analytes, such as glycine, thiacloprid, and imidacloprid.
These noble applications are attributed to the chemical and biological properties of phages suitable for a realtime sensor development. Therefore, M13 bacteriophage can be used as an attractive platform for the SPR sensing application.
While previous M13 phage-based studies of the SPR measurement focused on the specific binding property of the functionalized M13 phage, some groups integrated self-assembled property of M13 phage to SPR-sensing applications. Recently, Yoo et al. reported phage-arrays composed of self-assembled structures. To make a uniform phage film, the authors used the pulling up Commonly used configurations of SPR sensors: a prism coupler-based SPR system (the Kretschmann configuration), b grating coupler-based SPR system, c optical waveguide-based SPR system. Reprinted with permission from Elsevier B.V © 2016
## Fig. 5
Glyphosate-binding oligopeptide (TPFDLRPSSDTR) and glyphosate concentration dependence of SPR responses. Reproduced with permission from American Chemical Society © 2016 method enabling for the ordered patterning due to the evaporation-induced spontaneous reorganization of phages. In their contribution, phages modified to display peptides of RGD, RGE and HPQ successfully detected NIH3T3 mouse fibroblast cells through a spectral shift in the SPR spectrum. Their phage-chip array shows the phage concentration-and cell numbers-dependent SPR shift. As phage concentration increases from 0.3 to 1 mg/ mL, the SPR spectrum is red-shifted due to the increase in sample thickness. Similarly, the increase in numbers of cells influences the spectral shift to a longer wavelength (see [fig_ref] Figure 6 a: M13 phage [/fig_ref]. In addition, the authors observed that controlling of cellular direction and morphology by self-assembled monolayer is effective in guiding cell growth.
Oh et al. have applied the self-assembled property of M13 phage on the SPR sensing application. The authors reported M13 phage based on the novel SPR sensing system including high selectivity for streptavidin. As a sensing material, M13 phage incorporated with specific binding peptide (His-Pro-Gln: HPQ) has been prepared through the phage display technique. The nematic M13 phage matrices have been fabricated on the gold films with the thickness of ~50 nm deposited on glass substrates by a simple pulling technique, which is commonly used for the self-assembly process of liquid particles. Through this fabrication process, Oh et al. have implemented an anisotropic nanostructure by mimicking the 3D photonic crystal structure of Morpho didius. Their system has demonstrated excellent selectivity and sensitivity in the SPR signal by ~2700 copies of genetically expressed peptide on the pVIII major coat protein. Their system has also exhibited the sensitivity dependence for the alignment of receptor matrix in the specific direction. As shown in [fig_ref] Figure 7 a: The spectral shift of resonance peak for different types of phage films [/fig_ref] , different spectral shift of resonance peak is observed for three types of M13 phage films (isotropic, nematic horizontally (nematic 0°), and nematic perpendicularly oriented (nematic 90°) phage films). Since the confinement of the near field differs depending on the orientation due to the anisotropic nature of the self-assembled M13 phage, these results imply that the detecting efficiency of the phage based on the SPR signal can be maximized by analyte concentrations in real time. These applications are attributed to the chemical and biological properties of M13 phage suitable for a real-time sensor development. Therefore, M13 phage can be used as an attractive platform for the SPR sensing application.
## M13 phage-based fret applications
FRET has been most widely investigated in various applications of fluorescence, including medical diagnostics, DNA analysis, and bio-optical imaging. After this phenomenon, named by Theodor Förster, was initially described in 1948, FRET-based studies have expanded into other research areas with the help of advances in the fluorescence detection technique by the improved spectral resolution and high sensitivity. A typical aspect of these applications involves the selection of probe materials suitable for the optimization of the energy transfer. For this reason, many studies and developments associated with fluorescent materials applicable to FRET, such as organic dyes, conjugated polymers, semiconductor nanocrystals, and quantum dots (QDs), have been performed. For example, due to their electron affinity and high quantum efficiency, organics dyes are most commonly used as efficient fluorescent materials in FRET based on optical detection. For decades, it has been proven that organic dyes offer several unique advantages in FRET-based biomolecular imaging application. Furthermore, due to their unique electrical and optical properties, conjugated polymers have also received more attention as probe materials in the investigation of the FRET mechanism. Conjugated polymers have a unique structure characterized by a π-orbital enabling exciton hopping along their backbone. In addition to the use of organic molecules, recent research has suggested that colloidal semiconductor nanocrystals or QDs are also useful for FRET applications, because they have many advantages as compared to conventional organic fluorophores, such as high extinction coefficient and size tenability.
Although the fluorescent materials for FRET have noticeable advantages, such as high quantum efficiency and electron affinity, the optical properties of these materials do not guarantee the optimal energy transfer, because the carrier relaxation is affected by the quenching process which diminishes fluorescence intensity or by trapping excited carriers. This quenching process is originally caused by molecular contact, and the common molecular system generally has many quenchers. Thus, the appropriate work of the system design is very important for efficient FRET. Recently, in situ FRET based on the optical DNA detection scheme using conjugated polymer has been demonstrated by Bazan and many other researchers. They suggested the excellent FRET design enabling the amplification of fluorescence signal through the fine tuning of the intermolecular distance by the electrostatic interaction between optical platforms. Thus, their result also implies that the optimal FRET is possible through a careful system design.
In M13 phage based on sensor applications, M13 phage is commonly used as an alternatives to classical 1D nanoscaffolds, such as carbon nanotubes, while providing suitable constructs serving as heterogeneous supports of nanoparticles (NPs), a high surface area template for the co-anchoring of photo-activated molecular donors/ acceptors, a spacer element to funnel and direct the sequential electron-transfers. In particular, due to its due specific binding property and well-defined shape, M13 phage is useful as a platform or scaffold. Thus, we can easily expect that M13 phages can be used as optical platforms for FRET.
In this section, we will introduce the use and potential of M13 phage in FRET-based optical sensor application. For this purpose, we will first explain the basic equations of FRET for understanding the essential concept of FRET. Then, we will account for the recent progress in FRET application based on M13 phage.
Theoretically, FRET is the excitation energy transfer process from the excited donor molecule to an acceptor molecule by the dipole-dipole coupling and it is observable at the range of 10-100 Å as shown in the Jablonski diagram (see [fig_ref] Figure 8 a: Jablonski diagram of FRET process[104] [/fig_ref]. When a donor molecule is excited by incident light, the excited state energy of donor can be transferred to an acceptor molecule which is in close proximity. Then, this leads to a decrease in the donor's fluorescence intensity and an increase in the acceptor's emission intensity. Interestingly, the resonance energy transfer process non-radiatively occurs without the involvement of a photon, although the emission spectrum of donor molecules overlaps with the acceptor's absorption spectrum. In addition, the energy transfer is a through-space interaction which is mostly independent from the intervening solvent and/or macromolecule.
In FRET, the energy transfer process is based on the concept of coupled dipoles, which can exchange energy with another dipole with a similar resonance frequency. Thus, the strength is determined by the relative orientation and distance between two dipoles. Assuming that two molecules (donor and acceptor) are separated by a distance R, the FRET rate is inversely proportional to sixth power of R. The rate equation is given by Eq. (2).
where Q D is the quantum yield of the donor in the absence of acceptor, τ D is the decay time of donor's fluorescence in the absence of acceptor, κ 2 is a factor considering the relative orientation of two dipoles; κ 2 = 2/3 considering molecular averaging at liquid solution, N A is Avogadro's number, and n is the refractive index of medium. In case of solution, it corresponds to the refractive index of the solvent. F D (λ) is the donor's fluorescence intensity in the absence of acceptor and α A (λ) is the extinction coefficient of the acceptor at wavelength λ. The integral part represents the degree of spectral overlap between the donor's emission and the acceptor's absorption. [fig_ref] Figure 8 a: Jablonski diagram of FRET process[104] [/fig_ref] schematically shows the distance dependence of FRET and spectral overlap . This rate The Förster distance (R 0 ) of [fig_ref] Figure 8 a: Jablonski diagram of FRET process[104] [/fig_ref] is the intermolecular distance between two molecules when the energy transfer efficiency is 50 % and this can be simply calculated from the spectral properties of the donor and the acceptor [105]. The distance-dependent nature of FRET is very useful in chemical or biological applications, because this process occurs over distances comparable to the dimensions of biological macromolecules. Therefore, FRET is suitable for quantitatively detecting the conformational change in the orientation of fluorescent molecules and obtaining structural information about the macromolecule. For this reason, FRET is described as "a spectroscopic ruler" to be a proximity indicator.
[formula] (2) k FRET (R) = Q D κ 2 τ D R 6 9000 ln (10) 128π 5 N A n 4 ∫ ∞ 0 F D ( )α A ( ) 4 d ∫ ∞ 0 F D ( )d [/formula]
Recently, Chen et al. have reported FRET based on ratiometric fluorescent nanosensors using M13 phage. They have used M13 phage as a scaffold to construct FRET-based ratiometric fluorescent nanoprobes. As a FRET donor and an acceptor, fluorescein isothiocyanate (FITC) and rhodamine B (RhB) are used, respectively. Fluorescent dyes are conjugated to the N-terminus at the exterior surface of M13 phage using β-Cyclodextrin (β-CD) as a molecular linker (see . In general, there are the proper spectral overlaps between FITC and RhB, where the FRET process from FITC to RhB would occur. Thus, the changes in the fluorescence spectra by FRET can be expected, if intermolecular distance between dyes is sufficiently close for FRET. In the presence of M13β-CD, the significant increase in the emission intensity of RhB (580 nm) was observed. Considering that the average distances between two neighboring N-termini of pVIII proteins are 3.2 and 2.4 nm, this result clearly indicates that peptides of M13 phage provide a proper molecular spacing for efficient FRET. In addition, Chen and co-authors have implemented M13 phage-based ratiometric sensor using the sensitivity of dyes for acidity: FITC is pH sensitive, while RhB is not. Therefore, M13 phage is a suitable optical platform for the FRET-based applications.
As other optical platform application of M13 bacteriophage, it can be used as a template for light harvesting or exciton transporting. Recently, Nam et al. have demonstrated a light-harvesting antenna system using M13 phage. Since further modifications and genetic engineering over phage was pivotal in terms of tuning the assembly geometry and chromophore distances, the authors have noticed that the ordered coat protein
[formula] (3) k FRET (R) = 1 τ D R 0 R 6 [/formula]
of M13 phage can serve as a template guiding the interaction between pigments. The authors used Zn (II) deuteroporphyrin IX 2,4-bis (ethylene glycol) (ZnDPEG) as a model pigment and synthesized two samples, ZP-M13-1 (1564 porphyrins) and ZP-M13-2 (2900 porphyrins) with different numbers of zinc porphyrins. In this light harvesting system, Nam and co-authors observed the temporal migration of carriers (excitons) along the pigments assembled on the virus using the transient absorption measurement. In the presence of M13 phage, ZP-M13 rapidly decays as compared to ZnDPEG. The lifetime of ZP-M13 is two times shorter than that of ZnDPEG (see [fig_ref] Figure 10 a: Tryptophan fluorescence emission spectra of native M13 phage and ZP-M13 under 295... [/fig_ref]. This change results in the delocalization of the excitons driven by FRET, since the modification of the site energy (spectral broadening of absorption spectrum) by intermolecular interaction between protein and porphyrins has clearly influenced the pathway of excitation energy transfer. Park et al. have also demonstrated the exciton transporting mechanism in M13 phage using organics dyes. They have fabricated a light harvesting system enabling molecular wire effect along to the coat protein of M13 phage. As a scaffold for the fluorophores, M13 phage (M13CF, ADSPHTELPDPAK) engineered by modifying the amino acid sequence of the major coat pVIII protein were used as shown in [fig_ref] Figure 11 a: Schematics for exciton transporting mechanism in the genetically engineered M13 phage based... [/fig_ref]. However, unlike in a previous study, Park et al. have directly controlled molecular spacing using M13 phage with an additional binding site of 10 Å distance (M13SF, AENKVEDPAK). This binding site contributed to the occupation probability of donor molecules over the possible site per M13 phage. This led to the enhancement of the coupling strength of the bound fluorophores and suppression of the fluorescence quenching via shortrange Dexter exchange interaction. In the presence of M13 phage, the fluorescence intensity of Alexa Fluor 488 (donor) is significantly quenched by FRET, while the yellow emission of free donor has no change (see [fig_ref] Figure 11 a: Schematics for exciton transporting mechanism in the genetically engineered M13 phage based... [/fig_ref]. In addition, due to the strong electronic coupling and the effective energy transport by subsequently highly linked network between the binding sites, the exciton lifetime (~422 ps) of donor is significantly faster than that of free donor (~4 ns). Therefore, the controlling of specific binding sites by inserting or deleting amino acids on pVIII of M13 phage is very useful for FRET based on sensing or higher-level exciton transporting.
## M13 phage-based sers applications
SERS is a powerful spectroscopic technique enabling for a highly sensitive detection due to the significant amplification of Raman signal from molecules attached to the metallic surface with nanometer size. It has been widely used in various bio-or chemical sensing applications for probing of single molecules, molecular analysis, bio markers, and environmental monitoring. It has even been used in forensic science for detection of explosives, drugs, blood, DNA, and fingerprints, since the Raman spectra of pyridine on silver electrode were first observed in 1974and their mechanism was analyzed in 1977. In particular, the SERS field has dramatically progressed by virtue of a high sensitivity with theoretical signal enhancement factor of 10 11 in an optimized substrate.
Before discussing M13 phage-based SERS applications, we first explain the Raman effect and the theoretical mechanism of SERS. The Raman effect is an inelastic scattering process between a photon and a molecule. When light is incident in a molecule, some of the incoming photons are inelastically scattered from a molecule. These scattered photons have a different energy (frequency) as compared to that of incident photons by interaction with molecular vibrational energy states in molecules. This phenomenon appears to frequencyshifted signal (spectral shift to lower or higher energy) in the spectrum. It is called the Stokes scattering when scattered photons lose energy and the reverse process where photons gain the energy due to interacting with excited vibrational states is called anti-Stokes scattering. In comparison to the Raman scattering or Raman, SERS is incorporated with metal nanostructures.
In the presence of nanostructures with coinage metals, the excitation of localized surface plasmon resonance (LSPR) can induce the amplification of scattered electromagnetic field. By |E 4 | approximation, the SERS intensity is proportional to the fourth power of incident electric fields (E(ω inc )): I SERS ≈ |E(ω inc )| 4. In SERS, the enhancement of the Raman signal is explained by two mechanisms of electromagneticand chemical theories. The electromagnetic approach describes signal enhancement with the amplification in the electric field due to roughness or shape of the metallic surface. In general, an incident electromagnetic plane wave can excite localized surface plasmons confined to the metal surface. Then, the plasmon oscillation perpendicular to the surface can occur, scattering of photons by the resonance with an incident electromagnetic wave. Therefore, SERS is strongly influenced by the systems which can absorb the photon and store the electromagnetic energy into the surface plasmon, i.e., small metal features and gratings. The theoretical electromagnetic enhancement factor is in the order of 10 10 -10 11. In the chemical approach, the enhancement of the Raman signal results in the transient charge transfer by the electronic interaction between metal and adsorbate, since a new electronic state by chemisorption can serve as a resonant intermediate state in the Raman scattering. In this case, theoretical chemical enhancement factors are in the order of 10 3.
Due to the specific binding property of M13 phage, it often used as an analytic material in the SERS applications for biosensing. Recently, Lee et al. reported a biosensor system that integrates SERS-active metal nanoparticles with DNA-modified M13 bacteriophage. In this application, M13 phage is used as a platform for signal enhancement. The authors observed that the high capturing of SERS-active Au@Ag core-shell nanoparticles by single phage leads to the exponential enhancement in the Raman signal. Therefore, they the DNA-phage system shows a 75-fold increase in the Raman signal as compared to that of DNA-antibody due to the high surface area of the phage.
M13 phage can be used as a signal reporter for the SERS-based medical applications; for example, Nguyen et al. reported a new mesoporous SERS substrate using M13 phage modified with cysteine-rich peptides on the pVIII major coat protein for sepsis diagnosis. In their new system, the authors prepared silica mesoporous templates for SERS through polymerization by mixing [fig_ref] Figure 2 a: Fluorescent labelling of phage through conjugation with ruthenium complexes [/fig_ref] a SERS spectra for three sepsis-specific biomarkers (C-reactive protein, procalcitonin, and sTREM-1). b Detection of unspecific and cross-talk assay for the biomarker-based sepsis. c The detection limit based on concentration of each biomarkers is estimated by Linear fitting of peak intensities.. Reprinted with permission from Elsevier B.V. © 2016 between M13 phage displaying cysteine-rich peptide and silica precursors. The authors carried out detection for three typical sepsis-specific biomarkers, including C-reactive protein, procalcitonin, and sTREM-1 based on principles of immunoassays (see [fig_ref] Figure 2 a: Fluorescent labelling of phage through conjugation with ruthenium complexes [/fig_ref]. Therefore, they observed that the SERS spectrum shows distinct peaks for each tags and has the detection limit of 27, 103, and 78 pM for each sepsis-specific biomarker. Furthermore, Lentini et al. reported the SERS material based on Phage-Ag nanoparticles for identification of Histiocytic lymphoma cell line (U937). U937 is an in vitro model cell line for cancer diagnosis in biomedical research. In their contribution, the phage display technique of a 9-mer pVIII M13 phage is used to screen over U937 and silver nanoparticles are incubated with phage clones to acquire the SERS signal. Therefore, the authors found that assembled network between phage-displayed peptides with EIII1 alignment and Ag nanoparticles shows new Raman scattering peaks at 862.6, 1132, and 1154 cm −1 as compared to the fundamental feature of U937, as well as signal enhancement.
## Summary
In the above-mentioned M13 phage-based optical sensing applications, there are difficulties to directly compare the sensing performance. Specifically, intrinsically different sensing mechanisms and techniques are used, while sensitivity, selectivity, pros and cons of a biosensor are crucial factors. In addition, these techniques were applied to the different sensing ranges, for example, pg/ mL to ng/mL level for SPR and ng/mL to μg/mL level for FRET. To provide a more comprehensive picture of prior research, sensing parameters used in previous studies are summarized in [fig_ref] Table 1: M13 phage-based optical sensors [/fig_ref].
# Conclusions
In recent years, M13 bacteriophage has expanded its use into various novel research areas, such as fluorescence, SPR, and exciton transporting network. In these sensing applications, versatility of M13 phage is attributed to its nontoxic, well-defined shape, self-assembling, and specific binding properties. Considering the recent trend of the use of M13 phage in bio-optical applications, M13 phage is consequently the most powerful candidates that corresponds to the necessity of the development of sensor enabling detection of various analytes, such as explosives, proteins, DNA, cancers, bacteria, toxins, and metal ions. In addition, since M13 phage is robust, thermally and chemically stable, as well as easy to incorporate with other motifs, such as biomolecules or nanoparticles, via genetic engineering, it is very useful as a functional nanomaterial for more diverse applications. For this reason, M13 phage based on sensor will be consistently developed and lead to new optical sensing strategies for the rapid, accurate, selective, and sensitive detection of analytes incorporated with various spectroscopic methods.
[fig] Figure 2 a: Fluorescent labelling of phage through conjugation with ruthenium complexes. b Various time-lapse confocal microscopy images of phage conjugated fluorescent materals. Red [Ru(phen)2(dppz)]2+ , green GFP, yellow colocalization of red and green fluorescence [37]. Reprinted with permission from Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim © 2016 [/fig]
[fig] Figure 3 a: Schematic diagram for tumor-targeting of SBP-M13-SWNT probe. b Absorbance spectrum of SWNTs in sodium cholate and SBP-M13-SWNT probe, respectively. c Fluorescence intensity of SBP-M13-SWNT in ovarian cancer cell culture. d Photobleaching fluorescence decay of FITC and SBP-M13-SWNTs, respectively. e Pharmacokinetic circulation study of SBP-M13-SWNT[38]. Reprinted with permission from National Academy of Sciences © 2016 [/fig]
[fig] Figure 4: Commonly used configurations of SPR sensors: a prism coupler-based SPR system (the Kretschmann configuration), b grating coupler-based SPR system, c optical waveguide-based SPR system [45]. Reprinted with permission from Elsevier B.V © 2016 [/fig]
[fig] Figure 6 a: M13 phage (RGD) concentration and b Number of NIH3T3 mouse fibroblast cells dependent SPR spectral shift[36]. Reproduced with permission from American Chemical Society © 2016 [/fig]
[fig] Figure 7 a: The spectral shift of resonance peak for different types of phage films. b The SPR sensitivity comparison between phage films. c The selectivity of HPQ phage film for streptavidin FITC[64]. Reprinted with permission from Elsevier B.V. © 2016 [/fig]
[fig] Figure 8 a: Jablonski diagram of FRET process[104]. Reproduced with permission from MDPI AG © 2016, b The schematics for distance (r) dependence of FRET and spectral overlap. c Förster distance and FRET efficiency[105]. Reprinted with permission from Royal Society of Chemistry © 2016 equation can be modified by the Förster characteristic distance (R 0 ) as given by Eq. (3)[65]. [/fig]
[fig] Figure 10 a: Tryptophan fluorescence emission spectra of native M13 phage and ZP-M13 under 295 nm excitation. b Molar extinction coefficient (absorption), c transient absorption at 400 nm, and d fluorescence spectra of ZnDPEG, ZP-M13-1, and ZP-M13-2 under 400 nm excitation. The insert in b shows a magnified emission spectrum around 550 nm, and the insert in d shows a magnified emission spectra of ZP-M13-2 [108]. Reproduced with permission from American Chemical Society © 2016 (See figure on previous page.)Fig. 9a The structures of M13-β-CD, Ada-FITC, and Ada-RhB. b Schematic of the FRET based ratiometric fluorescent pH nanosensor. c Normalized absorption spectra of Ada-FITC (green solid) and Ada-RhB (red solid), normalized fluorescence spectra of Ada-FITC (green dot) and Ada-RhB (red dot), respectively. d Fluorescence spectra of the mixture of Ada-FITC and Ada-RhB (green) or M13-b-CD/Ada-FITC/Ada-(red) under 450 nm excitation. The inset shows the image of solution in the absence (green) and presence (yellow) of M13 phage. e pH dependence of M13-β-CD/Ada-FITC/Ada-RhB complex. f The peak emission ratio between Ada-FITC (515 nm) and Ada-RhB (580 nm) shows linear pH dependence[34]. Reprinted with permission from Royal Society of Chemistry © 2016 [/fig]
[fig] Figure 11 a: Schematics for exciton transporting mechanism in the genetically engineered M13 phage based light harvesting system. Magnified M13 phage surfaces show the energy-transfer networks between chromophore-binding sites for M13CF (left) and M13SF (right), respectively. Insets show schematic networks of energy transport by exciton hopping in the M13CF and the M13SF, respectively. b Absorption and emission spectra of free donor (DN) and free acceptor (AC). c Fluorescence spectra of donor under 495 nm. Significant fluorescence quenching is observed in the presence of M13 phage. The upper inset is a fluorescence microscope image of the M13SF-DN. The bottom inset shows the change in emission colour of donor molecule as FRET occurs[109]. Reprinted with permission from Macmillan Publishers Limited, part of Springer Nature © 2016 [/fig]
[table] Table 1: M13 phage-based optical sensors [/table]
|
The study of brain functional connectivity in Parkinson’s disease
Parkinson's disease (PD) is a neurodegenerative disorder primarily affecting the aging population. The neurophysiological mechanisms underlying parkinsonian symptoms remain unclear. PD affects extensive neural networks and a more thorough understanding of network disruption will help bridge the gap between known pathological changes and observed clinical presentations in PD. Development of neuroimaging techniques, especially functional magnetic resonance imaging, allows for detection of the functional connectivity of neural networks in patients with PD. This review aims to provide an overview of current research involving functional network disruption in PD relating to motor and non-motor symptoms. Investigations into functional network connectivity will further our understanding of the mechanisms underlying the effectiveness of clinical interventions, such as levodopa and deep brain stimulation treatment. In addition, identification of PD-specific neural network patterns has the potential to aid in the development of a definitive diagnosis of PD.
# Background
Parkinson's disease (PD) is the second most common neurodegenerative disorder in the aging population. PD is characterized by progressive deterioration of motor function, such as bradykinesia, rigidity, resting tremor, gait disturbance, and postural instability. Patients with PD also experience non-motor symptoms such as cognitive deficits, anxiety, apathy, hallucination, and depression. The pathological hallmark of PD is progressive decrease in dopamine concentrations and neuronal cell loss within the substantia nigra and other brain structures combined with the appearance of intracytoplasmic inclusions composed of α-synuclein aggregates known as Lewy bodies [bib_ref] Parkinson's disease: clinical features and diagnosis, Jankovic [/bib_ref]. However, the precise mechanism by which the pathological changes in the brain result in the described clinical symptoms is unknown. It is well known that PD affects a large scale of neural networks. For example, dysfunction of cortico-basal gangliathalamo-cortical pathway is well known to be critical for the development of parkinsonian symptoms [bib_ref] Circuits and circuit disorders of the basal ganglia, Delong [/bib_ref].
Therefore, further examination of neuronal network integrity may provide more valuable information for understanding the pathophysiological changes of PD than investigations of local brain activity, and may be helpful to bridge the gap between pathological changes and clinical presentations in PD.
The development of techniques such as functional magnetic resonance imaging (fMRI), electroencephalography (EEG), magnetoencephalography MEG), and transcranial magnetic stimulation (TMS) has greatly enhanced the ability to evaluate functional network integrity in vivo [bib_ref] Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging, Fox [/bib_ref] [bib_ref] Regional homogeneity changes in patients with Parkinson's disease, Wu [/bib_ref]. In recent years, extensive studies have investigated PD-related disruption of functional networks, and have provided useful information regarding neurophysiological mechanisms underlying parkinsonian symptoms. In addition, these studies have served to identify mechanisms of anti-parkinsonian interventions, and suggest that brain networks have the potential to be developed as a biomarker for the diagnosis of PD. The aim of this review is to provide a comprehensive overview of the application of functional network connectivity in investigating neural mechanisms underlying parkinsonian symptoms and interventions, and as a potential biomarker in patients with PD. The reviewed publications were selected by the authors on the basis of relevance to the topic. The functional connectivity studies included in the current study is summarized in [fig_ref] Table 1: Summary of the characteristics of the reviewed studies [/fig_ref].
Techniques to assess network integrity EEG, fMRI, and MEG are the most widely used techniques that enable researchers to assess large-scale neural networks at different spatial and temporal resolutions. With the advantages of being noninvasive and having high spatial resolution, fMRI is now the most used method to investigate functional integrity of networks in PD. In broad definition, fMRI includes all magnetic resonance imaging (MRI) methods that detect neural functional changes, such as blood oxygen leveldependent (BOLD) contrast imaging, perfusion, or diffusion. However, fMRI typically refers to BOLD fMRI, which detects changes in oxygen saturation levels of the blood [bib_ref] Brain magnetic resonance imaging with contrast dependent on blood oxygenation, Ogawa [/bib_ref]. In this review, the applications of BOLD fMRI on network integrity in PD will be discussed. The methods used to explore network integration involve the analysis of functional or effective connectivity [bib_ref] Organization, development and function of complex brain networks, Sporns [/bib_ref] [bib_ref] The elusive concept of brain connectivity, Horwitz [/bib_ref] [bib_ref] Characterization of anatomical and functional connectivity in the brain: a complex networks..., Stam [/bib_ref] [bib_ref] Complex brain networks: graph theoretical analysis of structural and functional systems, Bullmore [/bib_ref] [bib_ref] Graph theoretical analysis of complex networks in the brain, Stam [/bib_ref]. Functional connectivity is defined as a temporal correlation between spatially remote neurophysiological events, whereas effective connectivity is defined as the influence that one neuronal system exerts over another [bib_ref] Functional and effective connectivity in neuroimaging: A synthesis, Friston [/bib_ref]. Findings from both methods will be presented in this review.
## Motor symptoms-related network changes
## Bradykinesia
Bradykinesia is an important feature contributing to motor difficulties in PD. In this review, we use bradykinesia to describe bradykinesia (slowness of movement), hypokinesia (smallness of movement), and akinesia (lack of movement). Although extensive research has been conducted in this area, the pathophysiological mechanisms underlying bradykinesia remain unclear. Several neuroimaging studies have investigated network connectivity during performance of various motor tasks in patients with PD. During the performance of selfinitiated movement, the functional connectivity between the striatum and cortical motor areas, i.e., primary motor cortex (M1), premotor cortex (PMC), and supplementary motor area (SMA), is weakened in PD [bib_ref] Effective connectivity of brain networks during self-initiated movement in Parkinson's disease, Wu [/bib_ref]. In addition, the connectivity between the prefrontal cortex, PMC and SMA is disrupted [bib_ref] Effective connectivity of brain networks during self-initiated movement in Parkinson's disease, Wu [/bib_ref] [bib_ref] Attention to action in Parkinson's disease: impaired effective connectivity among frontal cortical..., Rowe [/bib_ref] [bib_ref] Effective connectivity of neural networks in automatic movements in Parkinson's disease, Wu [/bib_ref] [bib_ref] Dynamic causal modelling of effective connectivity from fMRI: are results reproducible and..., Rowe [/bib_ref]. The SMA is critical in planning and decision of movements and plays a primary role in the preparation of self-initiated movements [bib_ref] Cortical areas and the selection of movement: a study with positron emission..., Deiber [/bib_ref] [bib_ref] The preparation and execution of self-initiated and externally-triggered movement: a study of..., Cunnington [/bib_ref] [bib_ref] Selfinitiated versus externally triggered movements. II. The effect of movement predictability on..., Jenkins [/bib_ref]. The SMA is one of the main receiving regions of the basal ganglia motor circuit [bib_ref] The origin of thalamic inputs to the arcuate premotor and supplementary motor..., Schell [/bib_ref]. The dysfunction of the SMA has been correlated with motor difficulty, and the administration of levodopa has been shown to relatively normalize the function of the SMA in patients with PD resulting in improved motor performance [bib_ref] Event-related functional magnetic resonance imaging in Parkinson's disease before and after levodopa, Haslinger [/bib_ref] [bib_ref] Pharmacologically modulated fMRI-cortical responsiveness to levodopa in drug-naive hemiparkinsonian patients, Buhmann [/bib_ref]. Thus, the disconnection of the striato-SMA pathway due to the deficit of the nigrostriatal dopamine system is likely to be an important factor contributing to bradykinesia in PD.
Motor automaticity has been proposed as a possible mechanism underlying bradykinesia [bib_ref] Motor automaticity in Parkinson's disease, Wu [/bib_ref]. Automaticity is the ability to perform movements without attention directed toward the details of the movement [bib_ref] The influence of a new chlordiazepoxide analogue on human mental and motor..., Bernstein [/bib_ref]. In healthy people, the processing of motor automaticity is accompanied by the more efficiency of neural network and less significant of attentional network. The automated motor program is likely stored in the sensorimotor striatum (posterior putamen), and is resistant to interference [bib_ref] Motor automaticity in Parkinson's disease, Wu [/bib_ref] [bib_ref] Goal-directed and habitual control in the basal ganglia: implications for Parkinson's disease, Redgrave [/bib_ref] [bib_ref] How self-initiated memorized movements become automatic: a functional MRI study, Wu [/bib_ref] [bib_ref] Neural correlates of dual task performance in patients with Parkinson's disease, Wu [/bib_ref]. Most bradykinesia-related motor problems are associated with deterioration of motor automaticity, as PD patients tend to perform all daily behaviors slower or with smaller amplitude, e.g., akinesia, reduced arm swing, freezing of gait (FOG), and micrographia [bib_ref] Neural correlates underlying micrographia in Parkinson's disease, Wu [/bib_ref]. Motor automaticity dysfunction is already apparent in the early stages of PD [bib_ref] A functional MRI study of automatic movements in patients with Parkinson's disease, Wu [/bib_ref] [bib_ref] Attention to Automatic Movements in Parkinson's Disease: Modified Automatic Mode in the..., Wu [/bib_ref]. During automatic processing, the connectivity of striatocortical motor pathways is decreased, the activity in the sensorimotor striatum is not enhanced, and the attentional networks remain active in PD compared to controls [bib_ref] Motor automaticity in Parkinson's disease, Wu [/bib_ref] [bib_ref] Attention to Automatic Movements in Parkinson's Disease: Modified Automatic Mode in the..., Wu [/bib_ref] [bib_ref] Restingstate functional connectivity of dentate nucleus is associated with tremor in Parkinson's..., Ma [/bib_ref].
Based on these studies of neural networks, neural mechanisms for impaired motor automaticity in PD includes less efficient neural coding of movement, failure to shift automated motor skills to the sensorimotor striatum, instability of the automatic mode within the striatum, and use of attentional control efforts to execute movements usually performed automatically in healthy people [bib_ref] Motor automaticity in Parkinson's disease, Wu [/bib_ref]. As a consequence, PD patients lose previously acquired automatic skills and have difficulty in acquiring new automatic skills or restoring lost motor skills, which results in bradykinesia.
## Tremor
As tremor may disturb fMRI signals, tremor-related network connectivity has been much less investigated compared to bradykineisa. In an elegant study, Helmich and colleagues described the use of electromyography to monitor tremor during fMRI scanning, and measured functional connectivity between basal ganglia nuclei and the cerebellothalamic circuit [bib_ref] Intact limbic-prefrontal connections and reduced amygdala volumes in Parkinson's disease with mild..., Surdhar [/bib_ref]. The authors reported that the basal ganglia nuclei were transiently activated at the onset of tremor, while activity in the cerebellothalamic circuit correlated with tremor amplitude. The internal globus pallidus and putamen had increased functional connectivity with the cerebellothalamic circuit. These findings suggest that parkinsonian tremor may result from a pathological interaction between the basal ganglia and the cerebellothalamic circuit, which is supported by the following studies [bib_ref] Restingstate functional connectivity of dentate nucleus is associated with tremor in Parkinson's..., Ma [/bib_ref].
Functional connectivity experiments have also been used to explore the underlying mechanisms for several [bib_ref] Resting-state brain connectivity in patients with Parkinson's disease and freezing of gait, Tessitore [/bib_ref]. Functional neuroimaging studies suggest that the disturbances in frontal cortical regions, the basal ganglia, and the midbrain locomotor region are possibly the origins of FoG [bib_ref] Freezing of gait: moving forward on a mysterious clinical phenomenon, Nutt [/bib_ref]. Network connectivity also can be used to identify the neural characters in different subtypes of PD [bib_ref] Circuits and circuit disorders of the basal ganglia, Delong [/bib_ref] [bib_ref] Altered restingstate functional connectivity of the dentate nucleus in Parkinson's disease, Liu [/bib_ref] [bib_ref] Decreased interhemispheric functional connectivity in subtypes of Parkinson's disease, Hu [/bib_ref].
## Non-motor symptoms
In addition to symptoms related to motor function, most PD patients present with some non-motor symptoms such as cognitive, emotional, or olfactory impairments.
In recent years, more focus has been applied to characterizing the neural network of these non-motor symptoms. Cognitive deficits are common in PD patients. PD with dementia is associated with selective disruption of corticostriatal connectivity [bib_ref] Interregional correlations in Parkinson disease and Parkinson-related dementia with resting functional MR..., Seibert [/bib_ref]. Moreover, it has been shown that the connectivity of the so called "default Resting state RFCS (regional functional connectivity strength)
The PD patients showed significant RFCS increases in the left parahippocampal gyrus, left angular gyrus and right middle temporal gyrus mode network" (DMN) is disrupted in PD patients with cognitive deficits [bib_ref] To rise and to fall: functional connectivity in cognitively normal and cognitively..., Gorges [/bib_ref] [bib_ref] Resting state functional connectivity is associated with cognitive dysfunction in non-demented people..., Disbrow [/bib_ref]. The DMN is a network showing consistent task-related deactivations, and includes the medial prefrontal cortex, anterior cingulate cortex, posterior cingulate cortex, precuneus, and inferior parietal lobe [bib_ref] Neurophysiological architecture of functional magnetic resonance images of human brain, Salvador [/bib_ref] [bib_ref] Functional connectivity in the resting brain: a network analysis of the default..., Greicius [/bib_ref]. The DMN is thought to facilitate cognitive performance by allocating neural resources to critical brain regions. The disruption of the DMN was associated with the progress of cognitive decline [bib_ref] To rise and to fall: functional connectivity in cognitively normal and cognitively..., Gorges [/bib_ref] , while the decline in cognitive function, particularly in the memory and visuospatial domains, was associated with stronger coupling between the dorsal caudate and the rostral anterior cingulate cortex [bib_ref] Resting-state functional connectivity of the striatum in early-stage Parkinson's disease: Cognitive decline..., Manza [/bib_ref]. These findings suggest that malfunctioning of the DMN may contribute to the executive function deficits in PD. Depression is the most frequent psychiatric disorder reported in patients with PD. Abnormal prefrontallimbic network connectivity has been demonstrated in depressed PD patients [bib_ref] Intact limbic-prefrontal connections and reduced amygdala volumes in Parkinson's disease with mild..., Surdhar [/bib_ref] [bib_ref] Resting-state fMRI study on drug-naive patients with Parkinson's disease and with depression, Luo [/bib_ref]. PD patients with depression are associated with disrupted functional connectivity between the median cingulate cortex and precuneus, prefrontal cortex, and cerebellum [bib_ref] Altered Resting-State Brain Activity and Connectivity in Depressed Parkinson's Disease, Hu [/bib_ref]. The cingulate cortex plays key roles in integrating multimodal information that is important for emotional, sensorimotor, and cognitive functions [bib_ref] Two systems of resting state connectivity between the insula and cingulate cortex, Taylor [/bib_ref]. The median cingulate cortex also appears to be involved in many emotion-related cognitive processes such as meditation, self-related rumination, aversive conditioning, and the anticipation and perception of pain [bib_ref] Structural and functional dichotomy of human midcingulate cortex, Vogt [/bib_ref]. The impaired median cingulate cortex-related networks may play a role in depression experienced by patients with PD.
Network connectivity in some other non-motor symptoms, such as olfactory impairment, apathy, and hallucination has also been investigated. PD patients with olfactory impairment had decreased connectivity between the posterior cingulate cortex and bilateral primary sensory areas, right frontal areas, and right parietal areas, and had an enhancement of striatocortical connectivity compared to PD patients with normal olfaction [bib_ref] Olfactory performance and resting state functional connectivity in nondemented drug naive patients..., Sunwoo [/bib_ref]. Apathetic PD patients showed reduced functional connectivity mainly involving limbic striatal and frontal territories. In addition, the limbic division of the left striatum showed reduced connectivity with the ipsilateral frontal cortex and with the rest of the left striatum [bib_ref] Resting-state frontostriatal functional connectivity in Parkinson's disease-related apathy, Baggio [/bib_ref]. In PD patients with visual hallucinations, occipitalcortico-striatal connectivity was significantly higher than in patients without hallucinations [bib_ref] Resting activity in visual and corticostriatal pathways in Parkinson's disease with hallucinations, Yao [/bib_ref]. Hallucinations have been associated with functional abnormalities in primary visual cortex and visual associative cortices [bib_ref] Impaired visual processing preceding image recognition in Parkinson's disease patients with visual..., Meppelink [/bib_ref].
# Intervention-related network changes
Functional connectivity can be also used to investigate neural mechanisms underlying anti-parkinsonian interventions. Levodopa treatment has been reported to normalize the function of the basal ganglia motor pathways (e.g., by enhancing neural activity in the SMA and striatum) and restore striato-cortical motor pathway connectivity [bib_ref] Regional homogeneity changes in patients with Parkinson's disease, Wu [/bib_ref] [bib_ref] Altered resting state cortico-striatal connectivity in mild to moderate stage Parkinson's disease, Kwak [/bib_ref] [bib_ref] Cortico-striatal-thalamic network functional connectivity in hemiparkinsonism, Agosta [/bib_ref] [bib_ref] Dopaminergic basis for impairments in functional connectivity across subdivisions of the striatum..., Bell [/bib_ref] [bib_ref] Functional connectivity in the basal ganglia network differentiates PD patients from controls, Szewczyk-Krolikowski [/bib_ref] in a manner associated with improvements in motor function.
Although levodopa remains the most effective medication for the management of PD symptoms, many PD patients develop daily fluctuations in mobility and involuntary movements known as levodopa-induced dyskinesias (LID). The neural correlates in the genesis of LID remain poorly understood. A recent study found an increase in connectivity between the putamen and M1 after levodopa intake in patients developed LID [bib_ref] Abnormal dopaminergic modulation of striato-cortical networks underlies levodopa-induced dyskinesias in humans, Herz [/bib_ref]. This excessive striato-cortical connectivity in response to levodopa may play a role in the pathophysiology of LID [bib_ref] Abnormal dopaminergic modulation of striato-cortical networks underlies levodopa-induced dyskinesias in humans, Herz [/bib_ref] [bib_ref] Resting-state connectivity predicts levodopa-induced dyskinesias in Parkinson's disease, Herz [/bib_ref]. Another study showed that the connectivity of inferior frontal cortex was decreased with M1 and increased with the putamen in patients with LID [bib_ref] The motor inhibition system in Parkinson's disease with levodopa-induced dyskinesias, Cerasa [/bib_ref]. This finding suggests that the neural network centered on the inferior frontal cortex may also involve in the pathophysiological mechanisms underlying LID.
Deep brain stimulation (DBS) is another effective therapy for PD, but the neural mechanism underlying therapeutic effects of DBS remain unclear. It has been shown that DBS on the subthalamic nucleus (STN) can modulate the connectivity of striato-thalamo-cortical and STN-cortical pathways in association with symptom improvements [bib_ref] Resting state functional MRI in Parkinson's disease: the impact of deep brain..., Kahan [/bib_ref]. The pedunculopontine nucleus (PPN) is a target in treating primarily gait and posture symptoms. PPN-DBS has been reported to normalize pathological PPN connectivity [bib_ref] Chronic pedunculopontine nucleus stimulation restores functional connectivity, Schweder [/bib_ref].
## Diagnosis
The diagnosis of PD is based mainly on clinical assessments. Some studies have combined fMRI and various pattern analysis methods to try to establish an imaging methodology for PD diagnosis [bib_ref] Functional connectivity in the basal ganglia network differentiates PD patients from controls, Szewczyk-Krolikowski [/bib_ref] [bib_ref] Automatic classification of early Parkinson's disease with multi-modal MR imaging, Long [/bib_ref] [bib_ref] Reliability analysis of the resting state can sensitively and specifically identify the..., Skidmore [/bib_ref] [bib_ref] Parkinson's disease-related spatial covariance pattern identified with resting-state functional MRI, Wu [/bib_ref]. In a recent study, the authors of this review have identified a PDrelated spatial covariance pattern that was characterized by decreased activity in the striatum, supplementary motor area, middle frontal gyrus, and occipital cortex, and also by increased activity in the thalamus, cerebellum, precuneus, superior parietal lobule, and temporal cortex. This pattern had a high accuracy (90 %) to discriminate PD patients from healthy controls [bib_ref] Automatic classification of early Parkinson's disease with multi-modal MR imaging, Long [/bib_ref]. These studies have proven that network connectivity approach can identify characteristic PD-specific neural changes, and has the potential of network pattern as a biomarker for PD diagnosis. However, functional connectivity cannot directly reveal pathological changes in PD, therefore, whether this method can be applied in clinical practice needs further investigation.
# Conclusions
## Future directions
Studies on functional connectivity have provided important information on PD-related functional and pathophysiological changes. At present, functional connectivity is primarily used to further understand how pathological changes lead to parkinsonian symptoms, and is far from being a method in routine clinical investigations. As shown in the [fig_ref] Table 1: Summary of the characteristics of the reviewed studies [/fig_ref] , the analytic methods of "functional connectivity" studies vary a lot from study to study. Few studies have used the same "functional connectivity" procedure. Therefore, it is hard to perform meta-analysis on these studies. In contrast, voxel-level analysis, like most PET studies and some resting-state fMRI studies focusing the local activity support coordinate-based meta-analysis and, hence, are more helpful to clinical studies. Additional research should focus on increased efforts to develop neural network pattern as a neuroimaging biomarker for early diagnosis of PD; this might well require further technical and methodological improvements. These developments will improve early diagnosis, better evaluate disease progression, differentiate PD from other parkinsonisms on an individual basis, and may guide novel targets for future therapies.
[table] Table 1: Summary of the characteristics of the reviewed studies [/table]
|
Psychiatric Symptoms in Patients with Shiga Toxin-Producing E. coli O104:H4 Induced Haemolytic-Uraemic Syndrome
Background: In May 2011 an outbreak of Shiga toxin-producing enterohaemorrhagic E. coli (STEC) O104:H4 in Northern Germany led to a high number of in-patients, suffering from post-enteritis haemolytic-uraemic syndrome (HUS) and often severe affection of the central nervous system. To our knowledge so far only neurological manifestations have been described systematically in literature.Aim: To examine psychiatric symptoms over time and search for specific symptom clusters in affected patients.Methods: 31 in-patients suffering from E. coli O104:H4 associated HUS, were examined and followed up a week during the acute hospital stay. Psychopathology was assessed by clinical interview based on the AMDP Scale, the Brief Symptom Inventory and the Clinical Global Impressions Scale.Results: At baseline mental disorder due to known physiological condition (ICD-10 F06.8) was present in 58% of the examined patients. Patients suffered from various manifestations of cognitive impairment (n = 27) and hallucinations (n = 4). Disturbances of affect (n = 28) included severe panic attacks (n = 9). Psychiatric disorder was significantly associated with higher age (p,0.0001), higher levels of C-reactive protein (p,0.05), and positive family history of heart disease (p,0.05). Even within the acute hospital stay with a median follow up of 7 days, symptoms improved markedly over time (p ,0.0001).Conclusions: Aside from severe neurological symptoms the pathology in E.coli O104:H4 associated HUS frequently includes particular psychiatric disturbances. Long term follow up has to clarify whether or not these symptoms subside.
# Introduction
From May to July 2011 an outbreak of Shiga toxin-producing enterohaemorrhagic Escherichia coli (STEC) O104:H4 infections in Northern Germany occurred. Of the 3.186 reported STECpatients 845 (22%) presented with haemolytic-uraemic syndrome. Ninety percent of the affected patients were adults [bib_ref] Epidemic profile of shiga-toxin producing Escherichia coli O104:H4 outbreak in Germany, Frank [/bib_ref]. According to the German STEC-HUS registry 69% of all HUS patients presented with neurological symptoms like headache or visual disturbances or neurological signs ranging from tremor and aphasia to delirium, coma and seizures [bib_ref] Best supportive care and therapeutic plasma exchange with or without eculizumab in..., Kielstein [/bib_ref]. One explanation for this unusual high incidence of neurological symptoms/signs was the increased virulence of the outbreak-strain O104:H4, which combined virulence factors of typical enteroaggregative and Shiga toxin-producing E. coli and showed an unusual age distribution of the affected patients as STEC-HUS usually occurs in pre-school children, were it shows lower rates of central nervous system (CNS) involvement. [bib_ref] Acute neurological involvement in diarrhea-associated hemolytic uremic syndrome, Nathanson [/bib_ref] In the majority of cases infection with Shiga toxin-producing Escherichia coli leads to watery diarrhoea followed by bloody diarrhoea and abdominal cramps due to colitis [bib_ref] Hemolytic uremic syndrome: Epidemiology, pathophysiology, and therapy, Andreoli [/bib_ref]. These symptoms can be followed by the haemolytic-uraemic syndrome, which is defined by the sudden onset of microangiopathic haemolytic anaemia, thrombocytopenia and acute kidney injury [bib_ref] Pathogenesis of shiga toxinassociated hemolytic uremic syndrome, Proulx [/bib_ref]. In severe cases of HUS patients can also show CNS pathology associated with worse outcome. CNS pathology sometimes occurs before the onset of the haemolytic-uraemic syndrome and shows different manifestations that range from lethargy to seizures and coma [bib_ref] Neurological manifestations of hemorrhagic colitis in the outbreak of Escherichia coli O157:H7..., Hamano [/bib_ref] [bib_ref] The neurological syndrome in adults during the 2011 northern German E. coli..., Magnus [/bib_ref].
Little is known about the mechanisms causing CNS pathology. In infection with STEC different cytotoxins, Shiga toxin 1 and Shiga toxin 2, play a crucial role in systemic disease. The subunit StxB of Shiga toxin 2 is able to bind to glycolipid receptor, globotriaosylceramide (Gb3), on the cell membrane of target cells. Subunit StxA inhibits protein synthesis within the target cell, leading to its apoptosis [bib_ref] Direct pathway from early/recycling endosomes to the golgi apparatus revealed through the..., Mallard [/bib_ref]. Gb3 expressing cells have been found in the glomerular endothelium and neurons of different animals, in rabbits they determine the localization of pathological lesions [bib_ref] Verotoxin glycolipid receptors determine the localization of microangiopathic process in rabbits given..., Zoja [/bib_ref]. Obata et al localized Gb3 on neurons and vascular endothelial cells within the human CNS [bib_ref] Shiga toxin 2 affects the central nervous system through receptor globotriaosylceramide localized..., Obata [/bib_ref]. Autopsy of the human CNS showed oedema, hypoxic-ischemic changes, microthrombosis and microhaemorrhages [bib_ref] A 20-year population-based study of postdiarrheal hemolytic uremic syndrome in Utah, Siegler [/bib_ref]. Former MRI studies in affected humans showed abnormal findings in up to 50% of patients, especially in the basal ganglia, but also in almost every other structure of the central nervous system. These lesions could not be correlated with the variety of clinical symptoms, severity of disorder or outcome, suggesting that the Shiga toxin could also have a direct effect on the central nervous system [bib_ref] Acute neurological involvement in diarrhea-associated hemolytic uremic syndrome, Nathanson [/bib_ref] [bib_ref] Patterns in early diffusion-weighted MRI in children with haemolytic uraemic syndrome and..., Donnerstag [/bib_ref].
In all previous outbreak reports on STEC-HUS specific psychiatric symptoms were not reported. None of the reports used a structured psychiatric evaluation of patients, neither during the acute phase nor during follow up.
The aim of our analysis was to conduct a structured psychiatric evaluation of patients suffering from STEC-HUS amidst the 2011 outbreak in our tertiary care center. Furthermore we were trying to substantiate suspected specific psychiatric symptom clusters suggested by the treating physicians and whenever possible perform a short term follow up during the hospital stay.
# Materials and methods
# Ethics statement
This study was performed in accordance with the Declaration of Helsinki.
Data were obtained during psychiatric consultation service and have been analysed retrospectively and anonymously. Results of blood samples, obtained in clinical indication, have been analysed retrospectively as well. The examinations were part of clinical routine treatment and patients gave their verbal informed consent to be examined. Therefore no written informed consent was given by the participants and all data were anonymized and deidentified prior to analysis. All 52 STEC-HUS in-patients treated at the Hannover Medical School participated in a study on the German outbreak of E. Coli O104:H4 [bib_ref] Best supportive care and therapeutic plasma exchange with or without eculizumab in..., Kielstein [/bib_ref]. Approval for this analysis was obtained from the Ethics Committee of the Hannover Medical School, Permit Number: 1106-2011.
## Study design and patients
During the German outbreak of E. Coli O104:H4 from May to July 2011 32 out of 52 STEC-HUS in-patients at our tertiary care hospital received psychiatric consultation service examinations. Systematic psychiatric consultation was initiated after various psychiatric symptoms were noted in many patients by their treating doctors. Due to the lack of data on psychiatric symptoms in infection with E. Coli O104:H4 all patients received a structured psychiatric consultation as well as follow up examinations.
One of the 32 screened patients was excluded from the analysis as criteria for HUS were not met. All interviews were conducted by experienced psychiatrists, trained for the assessment of the study. The first interview with the patient was performed after a median of 7 days (range 2 to 23 days) after the onset of diarrhoea.
The second interview was performed 7 days later (range 7 to 9 days).
Assessment of baseline data included demographic data like qualification and marital status, psychiatric and medical history, family medical history, present medication, treatment with plasma exchange or dialysis and laboratory results. Afterwards psychopathology was rated using the 115-item Association for Methodology and Documentation in Psychiatry (AMDP) scale. Assessment of global psychological distress and psychopathology with respect to the past seven days was performed using the Brief Symptom Inventory (BSI). Severity of psychiatric symptoms was conducted using the Clinical Global Impressions Scale (CGI). At follow up examination present medication and again psychopathology, severity of psychiatric symptoms and global psychopathological distress was examined using the AMDP scale, the CGI and the BSI (past seven days). Additionally patients' charts, clinical progress documentation and third party informations (physicians, nurses) were collected at baseline and follow up to gain more information about psychopathological abnormalities. 22 Patients completed both assessments, 9 only participated at baseline visit and were discharged before the conduction of the second interview.
## Definition of stec-hus and severity of illness
Infection with E. coli O104:H4 was confirmed by either positive stool cultures and/or presence of Shiga toxin, confirmed either by ELISA or PCR. HUS was defined according to the Robert-Koch-Institute (the national level health authority) definition if at least two out of three criteria were fulfilled: (i) thrombocytopenia (, 150610 9 /L), (ii) microangiopathic haemolytic anaemia lactate dehydrogenase (LDH.240 U/L and haemoglobin ,12.0 g/dL), and (iii) acute kidney injury according to the AKIN definition [13] [bib_ref] Acute kidney injury network: Report of an initiative to improve outcomes in..., Mehta [/bib_ref].
Severity of illness was assessed through lowest platelet count, highest peak of C-reactive Protein (CRP) and haemolytic activity, which was defined through highest peak of lactate dehydrogenase (LDH) and minimum haemoglobin level.
Psychiatric Assessment AMDP scale. The AMDP scale was developed in Europe by the Association for Methodology and Documentation in Psychiatry (AMDP) in order to standardize the assessment of psychopathological symptoms [bib_ref] The syndrome scales in the AMDP-system, Pietzcker [/bib_ref]. It is based on a semi-structured interview method and consists of 115 items. Each psychiatric symptom is scored from 0 (not at all), one (mild), two (moderate) to three (severe). The AMDP system is the most commonly used and best known psychiatric documentation system in the Germanspeaking area and has been translated into many other languages [bib_ref] Standardised rating scales in psychiatry: Methodological basis, their possibilities and limitations and..., Moller [/bib_ref]. Several studies reported that it has moderate to high interrater reliability for most symptoms and that it can be considered a well-established test with good to very good reliability and validity [bib_ref] Particular aspects of the interrater reliability of the AMDP psychopathology scale, Renfordt [/bib_ref].
Brief Symptom Inventory. The BSI was derived from the Symptom Check-List-90-revised (SCL-90-R) and has been used extensively in diverse populations as a screening instrument for global psychological distress [bib_ref] The brief symptom inventory: An introductory report, Derogatis [/bib_ref]. It is a 53-item self-administrated questionnaire that uses five-point Likert scales and measures nine dimensions: somatisation (SOM), obsessive-compulsive (O-C), depression (DEP), anxiety (ANX), interpersonal sensitivity (I-S), hostility (HOS), phobic anxiety (PHO), paranoid ideation (PAR), and psychoticism (PSY). Items of the BSI are assessed using a 5point scale ranging from 'not at all' (0) to 'extremely' [bib_ref] Hemolytic uremic syndrome: Epidemiology, pathophysiology, and therapy, Andreoli [/bib_ref]. Patients rate their symptoms with respect to the last seven days. In addition, the inventory comprises three general distress measures:
The General Severity Index (GSI) quantifies severity of illness and is used for outcome measurements. The Positive Symptom Distress Index (PSDI) is designed to measure the intensity of symptoms and the Positive Symptom Total (PST) reports the number of symptoms.
Scores are interpreted by comparison to age-appropriate norms. Normative data are available for both clinical and non-clinical samples of adolescents (over 13 years) and adults. The BSI demonstrated strong internal consistency reliability and a strong correlation with the SCL-90-R [bib_ref] The brief symptom inventory (BSI) as an outcome measure for adult psychiatric..., Piersma [/bib_ref] [bib_ref] Screening childhood cancer survivors with the brief symptom inventory-18: Classification agreement with..., Recklitis [/bib_ref]. Convergent validity was examined in various studies. Studies also showed a good reliability for the German version of the BSI which was used in this study.
Clinical Global Impressions scale. The Clinical Global Impressions scale is commonly used to measure symptom severity (CGI-S), improvement (CGI-I) and efficacy of treatments in treatment studies of patients with mental disorders. It is based on external rating. The severity of mental illness is rated on the following seven-point scale: 1 = normal, not at all ill; 2 = borderline mentally ill; 3 = mildly ill; 4 moderately ill; 5 markedly ill; 6 severely ill; 7 among the most extremely ill patients. Improvement, in this study assessed after one week, also can be rated on a sevenpoint scale: 1 = very much improved; 2 = much improved; 3 = minimally improved; 4 = no change; 5 = minimally worse; 6 = much worse; 7 = very much worse. In this observational study the Clinical Global Impression-Efficacy Index has not been used.
# Statistical analysis
Deviation from normal distribution was tested using the Kolmogorov-Smirnov test. Correlation analysis between continuous data was then performed using Pearson's test. Between group comparisons (presence of psychiatric disorder vs. healthy) were performed using t-tests for continuous data and Chi squared or Fisher's exact test for categorical data. For not normally distributed data non-parametric Mann-Whitney U-test was used.
BSI data were compared with a healthy control group (German reference population). For comparement scores from all scales were transformed into t-values (mean = 50; SD = 10). These were calculated according to age and gender specific German norm groups. Between group analyses were performed using two way Analysis of Variance (ANOVA) and Bonferroni's post-hoc test when appropriate. Considering the significant correlation between age and presence of psychiatric disorder (p,0.0001) a receiver operating characteristic (ROC) -analysis and calculation of Youden's Index (Sensitivity + Specificity -1) has been performed.
Results are presented as means (SD). P-values of less than 0.05 (two-tailed) were considered significant. Analysis was performed using the Statistical Package for the Social Sciences (SPSS TM ) for Windows 16.0.2 (SPSSInc., Chicago,IL) and most data are presented using Graph Pad Prism 5 (Graph Pad Inc., San Diego, CA).
# Results
## Sample characteristics
All 31 patients at baseline and 22 of them at follow up have been included in analysis. The following data describe the sample characteristics at baseline: Women comprised 71% of the sample. Median age was 40 years (SD 19.2) and ranged from 18 to 81 years. Median duration of diarrhoea was 7 days (SD 5) and of associated HUS 8 days (SD 4.5).
Assessed previous illnesses included prior renal disease (n = 0), smoking (n = 1; type 2), diabetes (n = 1), hyperlipidemia (n = 2) and elevated arterial blood pressure (n = 5). Only one patient reported a history of psychiatric disorder (Panic disorder). [fig_ref] Table 1: Frequency of abnormalities within the AMDP-Scale and other psychiatric findings [/fig_ref] shows frequency of psychopathological abnormalities on the AMDP scale at baseline and follow up. Aggregation of collected data and clinical psychiatric examination revealed that an organic psychiatric disorder (ICD-10 F06.8) was confirmed in 58% of patients. On the CGI only 7 patients (22.6%) were rated as not psychiatric ill, meaning that even more patients suffered from psychopathological abnormalities, though diagnosis criteria for a psychiatric disorder were not fulfilled. At baseline 88% of patients showed disturbances of affect. Mostly feelings of anxiety and panic attacks were reported. Some patients suffered from 10 panic attacks daily. Anxiety was often reported as generalized without obvious specific causes. The second most observed abnormalities were disturbances of attention and memory (88%). These mostly included concentration and attention deficits which were associated with slow and stiff thinking (formal disorders of thought: 78.2%). Four patients suffered from vivid hallucinations. As an illustrating example a patient reported he was seeing different people in his room and described the surrounding as if he was travelling in a train. He reported knowing that these scenic hallucinations could not be real and memorized them in detail after recovery. No patient reported or showed signs of suicidal tendencies.
## Frequency and characterization of psychiatric symptoms
## Brief symptom inventory
To confirm the findings in the AMDP-Scale in a self-reported instrument, patients completed the BSI. As some patients could not read due to visual disturbances or poor concentration in a few cases questions were read to them. [fig_ref] Figure 1: BSI t-values at baseline [/fig_ref] shows the results of the two way ANOVA analysis comparing patients with an age and gender matched healthy control group (t = 50). Post-hoc analyses show that the clinical sample significantly differs from the healthy controls.
# Laboratory results
There was no statistically significant difference in the levels of lowest platelet count or minimum haemoglobin between healthy and psychiatric affected patients. Presence of psychiatric disorder at baseline and follow-up was significantly associated with higher peak CRP (p,0.05). Presence of psychiatric disorder at follow up was also associated with higher levels of peak LDH (p,0.05).
## Other correlations and significant findings
Psychopathological abnormalities started between 0 to 10 days after onset of HUS (median 1.5 d, SD 2.9), 50% of patients developed symptoms within 24 h of HUS onset. [fig_ref] Table 2: Sample characteristics and differences between patients with or without organic psychiatric disorder... [/fig_ref] test showed statistically significant results according to age (p, 0.0001) and family history of heart disease (p,0.05). ROCanalysis and calculation of Youdens's Index for age younger/older than 30.5 years is shown in [fig_ref] Table 4: Age and organic psychiatric disorder [/fig_ref]. The observed significance with higher age was also found investigating single items like anxiety (p,0.0001), panic attacks (p,0.005) or slow thinking (p, 0.0001). Statistical analysis showed no significant difference between treatment with different types of medication like antibiotics (n = 6), eculizumab (n = 12), anticoagulants (n = 24) or antiepileptics (n = 19) and severity, outcome or presence of psychiatric disorder at baseline or in the completer-group.
## Course of the disease
According to the CGI scale patients significantly improved over time (p,0.0001). Evaluation of the CGI revealed that at follow up 50% of patients scored 1, meaning ''much improved''. In only one patient the condition remained unchanged and none showed signs of worsening. However, 9 patients remained with the diagnosis of an organic psychiatric disorder. [fig_ref] Figure 2: BSI t-values at follow up compared to baseline [/fig_ref] shows the post hoc analysis of the BSI with all completers between baseline and follow-up.
# Discussion
To our knowledge, we here report the first systematic psychometric analysis of CNS involvement in adult patients with STEC-HUS. Thirty-one in-patients with E.coli O104:H4 STEC-HUS during the 2011 German outbreak were studied. The majority of patients (58%) met the ICD-10 criteria for a mental disorder due to a known physical condition. Given that nine patients were already discharged at the follow up exam the frequency of psychiatric symptoms/diagnosis could be even higher.
Presence of psychiatric disorder at baseline was related to age or family history of heart disease, as well as higher CRP levels. CRP levels are known to be increased in patients suffering from depression or anxiety disorders [bib_ref] Differential association of somatic and cognitive symptoms of depression and anxiety with..., Duivis [/bib_ref]. The strong bidirectional connection between cardiovascular diseases and depression has been examined in various studies, but there are no previous data on susceptibility for psychiatric disorders concerning family history of heart disease [bib_ref] Heartache and heartbreak-the link between depression and cardiovascular disease, Nemeroff [/bib_ref].
Interestingly neither laboratory and other demographic data than age, nor the prescribed medication were related to the psychiatric symptoms.
The rapid onset of psychiatric symptoms, which occurred within 24 h after diagnosis of HUS in half of the patients, as well as their improvement over time correspond with recently published data from neurological examinations [bib_ref] Treatment of severe neurological deficits with IgG depletion through immunoadsorption in patients..., Greinacher [/bib_ref]. Our observation that higher age could be a risk factor for development of psychiatric symptoms is in line with a reported significance between higher age and a low score in the Mini-Mental State Examination of affected patients with HUS [bib_ref] Neurologic manifestations of E coli infection-induced hemolytic-uremic syndrome in adults, Weissenborn [/bib_ref]. Interestingly, this association was not found for other neurological deficits but is stable in our exact analysis of different symptoms (anxiety only, panic attacks only etc.). From the view of a toxin-mediated hypothesis of CNS involvement in STEC-HUS older patients could be more often affected due to a higher vulnerability or weakening of the blood-brain barrier [bib_ref] Age-and anatomy-related values of blood-brain barrier permeability measured by perfusion-CT in non-stroke..., Dankbaar [/bib_ref]. The ability of Shiga toxin to cross the blood-brain barrier has already been shown in the rat [bib_ref] Intraperitoneal administration of shiga toxin 2 induced neuronal alterations and reduced the..., Lucero [/bib_ref]. A recent publication focussing on the neurologic examination reported an elevated albumin ratio in the cerebrospinal fluid (CSF) of four out of five STEC-HUS affected patients. The authors interpreted these findings as a possible sign of blood-CSF barrier dysfunction. Additionally age was also reported to be the overriding risk factor for death in the German STEC-HUS registry [bib_ref] The German 2011 epidemic of shiga toxin-producing E. coli-the nephrological view, German [/bib_ref]. The significant correlation of psychiatric disorder in patients with higher peak CRP but not with lowest haemoglobin or lowest platelet count contributes to the thesis that CRP could be a better fitting laboratory marker of CNS complications in STEC-HUS then the usual indicators for the degree of progression in HUS [bib_ref] Escherichia coli O-157-induced hemolytic uremic syndrome: Usefulness of SCWP score for the..., Teramoto [/bib_ref]. Also, an interesting recently published laboratory finding is the alteration of circulating microRNAs (mi-R 126) in STEC-HUS patients with moderate to severe CNS involvement compared to patients with HUS only [bib_ref] Circulating microRNAs in patients with shiga-toxin-producing E. coli O104:H4 induced hemolytic uremic..., Lorenzen [/bib_ref]. The majority of affected patients suffered from anxiety symptoms and/or cognitive dysfunction. At first sight the anxiety syndrome of affected patients could be interpreted as a reaction due to suffering from a life-threatening epidemic disease. Clinical interview on the other hand revealed that most patients were not able to indicate the reason of their anxiety. Anxiety was described as a generalized state of affect, mixed with free floating anxiety and panic attacks up to 10 times daily. The mixture of generalized anxiety and panic attacks is typical for organic anxiety syndromes (ICD-10 F06.4). Interestingly, some of the patients suffering from panic attacks also reported recurring chest pain. The phobic scale within the BSI could represent sickness behavior and impairment due to diarrhoea and HUS. Specific phobias were negated.
Dysfunction of memory and disturbances of thought mainly included slow thinking and poor concentration. For instance, some patients were not able to read more than a few sentences. Seven patients reported severe dyscalculia.
In contrast to the various neurological findings in CNS pathology, observed psychiatric disturbances were relatively specific. Still, making a precise distinction between neurological and psychiatric symptoms is complicated as both are a sign of CNS affection and long term observations could help in making a more precise distinction.
Focussing on the psychiatric part, the question if a special circuit or anatomical structure is affected predominantly should be considered. Tironi-Farinati et al. showed that intraventricular injection of Shiga toxin 2 causes dendritic abnormalities in rat brains and increases the expression of Gb3 [bib_ref] Intracerebroventricular shiga toxin 2 increases the expression of its receptor globotriaosylceramide and..., Tironi-Farinati [/bib_ref]. They also reported a localization of Gb3 expressing cells in the CA1 region of the hippocampus and the striatum of rats. Another recent study conducted by Meuth et al. showed neuronal expression of Gb3 and dose dependent neuronal cytotoxity of Shiga toxin 2 on thalamic neurons and astrocytes of female rats, fitting to recently published research showing symmetric MRI alterations in the lateral thalamus and brainstem (50%) of affected patients [bib_ref] The neurological syndrome in adults during the 2011 northern German E. coli..., Magnus [/bib_ref] [bib_ref] Thalamic involvement in patients with neurologic impairment due to shiga toxin 2, Meuth [/bib_ref].
Involvement of the hippocampus and/or thalamus could possibly explain anxiety, panic attacks and disturbances of memory and attention [bib_ref] Memory-a century of consolidation, Mcgaugh [/bib_ref] [bib_ref] The many paths to fear, Gross [/bib_ref]. These questions remain unanswered, as our report only allows a description of psychiatric symptoms, but is not suitable to establish causality.
Beside the possible scientific value of accurate psychiatric examinations, these findings could also be clinically important, as only few patients received a sufficient symptomatic treatment such as benzodiazepines to lower severe anxiety. Though the course of the disease seems to be benign, due to lack of long-term data we cannot exclude the possibility of vulnerable patients developing disorders afterwards. In our outpatient clinic two patients asked for consultancy weeks after discharge.
# Limitations
First this study is limited by its open design and the single center setting, both limitations owed to the acute onset of the crisis making a proper planned randomized multi-center evaluation impossible. Given that the so far largest report on STEC-HUS in adults before the described outbreak involved 22 patients the assessment of 31 patients is rather high [bib_ref] Effectiveness of therapeutic plasma exchange in the 1996 Lanarkshire escherichia coli O157:H7..., Dundas [/bib_ref]. Still, the sample size is small and does not allow interpretations beyond a descriptive level. It is possible that moderate effects of medication or correlation with more laboratory findings only gain statistical significance in a larger cohort.
Second a major limitation of this study is the lack of long-term follow up data and its retrospective design. Though our analysis showed healing tendency for psychiatric disorders it is possible that patients suffer from residual symptoms or new psychiatric disturbances. Thus, there is a great need for further studies on long-term outcome. Third, continuous observation could reveal more data about correlations with other systemical abnormalities, especially neurological deficits, impact on outcome and more specific laboratory results.
## Implications
CNS involvement in patients with Escherichia coli 0104:H4 associated haemolytic uraemic syndrome shows high rates of psychiatric affection with a probably benign course of psychiatric disorder. A higher sensitivity for psychiatric disturbances could be important for an optimized disease severity based treatment approach of Escherichia coli 0104:H4-associated haemolytic uraemic syndrome. Further studies are needed to assess potential long-term sequelae of STEC-HUS and investigate the underlying mechanisms of these symptoms.
# Author contributions
[fig] Figure 1: BSI t-values at baseline.Figure 1 shows the results of the Brief Symptom Inventory at baseline as boxplots. Results are compared to a healthy cohort which is marked by horizontal lines (t = 50, SD = 10). The horizontal line inside the ''box'' represents the median. The boundaries of the ''box'' indicate the 25th and 75th percentiles; the whiskers indicate the largest and smallest observed values. P-values given in the figure are derived from Bonferroni's post hoc test after two way ANOVA (cases vs controls): **P,0.01 ***P,0.001. doi:10.1371/journal.pone.0101839.g001 [/fig]
[fig] Figure 2: BSI t-values at follow up compared to baseline. BSI t-values of completers at follow up (t = 1) compared to baseline (t = 0). Healthy cohort is marked by horizontal lines (t = 50, SD = 10). Numbers given in the figure represent mean (SD). P-values given in the figure are derived from Bonferroni's post hoc test after two way ANOVA (follow up vs baseline): *P,0.05 **P,0.01. doi:10.1371/journal.pone.0101839.g002 [/fig]
[table] Table 1: Frequency of abnormalities within the AMDP-Scale and other psychiatric findings. [/table]
[table] Table 2: Sample characteristics and differences between patients with or without organic psychiatric disorder at baseline. [/table]
[table] Table 3: Family history of illness and risk factors in patients with or without organic psychiatric disorder. [/table]
[table] Table 4: Age and organic psychiatric disorder. [/table]
|
The acquisition of clinically relevant amoxicillin resistance in Streptococcus pneumoniae requires ordered horizontal gene transfer of four loci
Understanding how antimicrobial resistance spreads is critical for optimal application of new treatments. In the naturally competent human pathogen Streptococcus pneumoniae, resistance to β-lactam antibiotics is mediated by recombination events in genes encoding the target proteins, resulting in reduced drug binding affinity. However, for the front-line antibiotic amoxicillin, the exact mechanism of resistance still needs to be elucidated. Through successive rounds of transformation with genomic DNA from a clinically resistant isolate, we followed amoxicillin resistance development. Using whole genome sequencing, we showed that multiple recombination events occurred at different loci during one round of transformation. We found examples of non-contiguous recombination, and demonstrated that this could occur either through multiple D-loop formation from one donor DNA molecule, or by the integration of multiple DNA fragments. We also show that the final minimum inhibitory concentration (MIC) differs depending on recipient genome, explained by differences in the extent of recombination at key loci. Finally, through back transformations of mutant alleles and fluorescently labelled penicillin (bocillin-FL) binding assays, we confirm that pbp1a, pbp2b, pbp2x, and murM are the main resistance determinants for amoxicillin resistance, and that the order of allele uptake is important for successful resistance evolution. We conclude that recombination events are complex, and that this complexity contributes to the highly diverse genotypes of amoxicillin-resistant pneumococcal isolates.Author summaryStreptococcus pneumoniae (the pneumococcus) is an asymptomatic coloniser of the human nasopharynx and a common cause of middle ear infections and pneumonia. The frontline treatment for bacterial respiratory tract infections is amoxicillin, a β-lactam antibiotic that inhibits cell wall synthesis. The pneumococcus is naturally competent and β-PLOS PATHOGENSPLOS Pathogens | https://doi.
[formula] a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 [/formula]
# Introduction
Genomic plasticity through frequent and large-scale recombination drives the evolution of antibiotic resistance in Streptococcus pneumoniae (the pneumococcus). This naturally competent member of the human nasopharyngeal commensal microbiota opportunistically causes otitis media as well as severe invasive diseases such as pneumonia, bacteremia, and meningitis [bib_ref] Influence of bacterial interactions on pneumococcal colonization of the nasopharynx, Shak [/bib_ref]. Despite conjugate vaccine introduction, the species remains an important human pathogen, as vaccine-escape and antibiotic resistant variants constantly arise [bib_ref] Rapid pneumococcal evolution in response to clinical interventions, Croucher [/bib_ref] [bib_ref] Pneumococcal capsular switching: A historical perspective, Wyres [/bib_ref].
Variants occur primarily through natural transformation and homologous recombination of DNA from strains or closely related species that occupy the same niche, such as S. mitis and S. oralis . One case study showed more than 7% of the genome had been transferred between two pneumococcal strains during the polyclonal infection of a pediatric patient over three months [bib_ref] Generation of Genetic Diversity among Streptococcus pneumoniae Strains via Horizontal Gene Transfer..., Hiller [/bib_ref]. Early studies of transformation in the pneumococcus showed uptake of fragments ranging from 2-6 Kb [bib_ref] Physical and genetic hybrids formed in bacterial transformation, Gurney [/bib_ref] , but cell-to-cell contact was found to facilitate transfer of long DNA fragments [bib_ref] Evolution via recombination: Cell-to-cell contact facilitates larger recombination events in Streptococcus pneumoniae...., Cowley [/bib_ref] , and recombinant regions as large as 30-50 Kb in length have been observed in clinically relevant lineages such as PMEN-1 [bib_ref] Rapid pneumococcal evolution in response to clinical interventions, Croucher [/bib_ref] [bib_ref] Pneumococcal capsular switching: A historical perspective, Wyres [/bib_ref] [bib_ref] The multidrugresistant PMEN1 pneumococcus is a paradigm for genetic success, Wyres [/bib_ref]. Events such as these have played a significant role in shaping the pneumococcal pangenome [bib_ref] Structure and dynamics of the pan-genome of Streptococcus pneumoniae and closely related..., Donati [/bib_ref].
Natural competence is activated when the competence stimulating peptide (CSP) interacts with the ComDE two-component system [bib_ref] Natural transformation and genome evolution in Streptococcus pneumoniae, Straume [/bib_ref] , where phosphorylated ComE activates transcription of the genes encoding the alternative sigma factor comX [bib_ref] Identification of DNA binding sites for ComE, a key regulator of natural..., Ween [/bib_ref] [bib_ref] Refining the Pneumococcal Competence Regulon by RNA Sequencing, Slager [/bib_ref]. ComX then activates transcription of the late com genes necessary for DNA uptake and recombination. This includes the type IV-like pilus (ComGC), endonuclease EndA, and transport proteins ComFA, and ComEC [bib_ref] Complex structure of the membrane nuclease of Streptococcus pneumoniae revealed by two-dimensional..., Rosenthal [/bib_ref] [bib_ref] Membrane location of a deoxyribonuclease implicated in the genetic transformation of Diplococcus..., Lacks [/bib_ref] [bib_ref] Single-strand breakage on binding of DNA to cells in the genetic transformation..., Lacks [/bib_ref] [bib_ref] Midcell Recruitment of the DNA Uptake and Virulence Nuclease, EndA, for Pneumococcal..., Bergé [/bib_ref]. RecA and the transformation-dedicated DNA processing protein A (DprA) are essential for single-stranded DNA (ssDNA) uptake without degradation [bib_ref] Role of the singlestranded dna-binding protein Ssbb in pneumococcal transformation: Maintenance of..., Attaiech [/bib_ref]. Although not fully understood, DprA and RecA are thought to polymerize on ssDNA upon cell entry, initiating the homology search, while additional DNA molecules are then bound by multiple SsbB proteins protecting them from degradation by endogenous nucleases [bib_ref] Role of the singlestranded dna-binding protein Ssbb in pneumococcal transformation: Maintenance of..., Attaiech [/bib_ref] [bib_ref] Identification of the major protein component of the pneumococcal eclipse complex, Morrison [/bib_ref]. It has been hypothesized that the accumulation of stable ssDNA-SsbB eclipse complexes in the cytoplasm enables successive recombination events during one competence window, increasing genetic plasticity [bib_ref] Role of the singlestranded dna-binding protein Ssbb in pneumococcal transformation: Maintenance of..., Attaiech [/bib_ref]. Sequestered ssDNA can then be accessed by DprA, which mediates the loading of RecA onto the pre-synaptic filament [bib_ref] Molecular determinants of the DprA−RecA interaction for nucleation on ssDNA, Lisboa [/bib_ref] [bib_ref] Structurefunction analysis of pneumococcal DprA protein reveals that dimerization is crucial for..., Quevillon-Cheruel [/bib_ref]. DprA is also involved in competence shut-off, along with ComX and RpoD sigma factor competition, and CSP degradation by HtrA [bib_ref] Direct involvement of DprA, the transformation-dedicated RecA loader, in the shut-off of..., Mirouze [/bib_ref] [bib_ref] Competence for Genetic Transformation in Streptococcus pneumoniae: Termination of Activity of the..., Piotrowski [/bib_ref] [bib_ref] Exit from Competence for Genetic Transformation in Streptococcus pneumoniae Is Regulated at..., Weng [/bib_ref] [bib_ref] The HtrA protease from Streptococcus pneumoniae digests both denatured proteins and the..., Cassone [/bib_ref].
To study recombination and the development of AMX resistance in the pneumococcus in more detail and identify possible evolutionary routes towards AMX resistance, we sequentially transformed genomic DNA from a resistant clinical isolate (serotype 11A, ST6521, German National Reference Centre for Streptococci 2017, SN75752) into two different recipient strains coming from different clonal complexes (serotype 2 D39V and serotype 4 TIGR4). Whole genome sequencing showed recipient-dependent, genome-wide mutation uptake. In addition to identifying the necessary determinants for AMX resistance, we found a strong tendency towards the order of pbp and murM allele uptake. Long read Pacbio sequencing excluded a role for DNA methylation or genomic rearrangements or movement of genetic mobile elements as resistance determinants. We also investigated non-contiguous recombination, previously observed in S. pneumoniae, Haemophilus influenzae, and Helicobacter pylori [bib_ref] Mosaic DNA Imports with Interspersions of Recipient Sequence after Natural Transformation of..., Kulick [/bib_ref] [bib_ref] Transformation of natural genetic variation into haemophilus influenzae genomes, Mell [/bib_ref] , and propose a model for how these complex events occur. To conclude, we show that the uptake of four alleles from a resistant strain (pbp2x, pbp2b, pbp1A and murM) is required and sufficient for clinically relevant amoxicillin resistance development in sensitive S. pneumoniae, and that necessary fitness compensation may define the order of allele uptake.
# Results
## Serial transformation with genomic dna from an amx resistant strain increased amx mic
In order to understand the sequence of recombination events leading to the development of AMX resistance, a susceptible S. pneumoniae strain D39V (minimum inhibitory concentration (MIC) 0.01 μg/mL) was transformed in two successive rounds with genomic DNA originating from AMX resistant clinical isolate 11A (MIC 4 μg/mL) followed by selection on a range of AMX concentrations above MIC [fig_ref] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX... [/fig_ref] [fig_ref] Table 1: Key strains used in this study [/fig_ref]. In the first round, ten colonies were picked at random from the highest AMX concentration with growth (0.03 μg/mL AMX). The MICs of these strains ranged from 0.047-0.125 μg/mL, 4-10 times that of wild-type D39V [fig_ref] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX... [/fig_ref].
The second round of transformation resulted in 100 strains with MICs from 0.25 to 2 μg/ mL [fig_ref] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX... [/fig_ref] , lower than the donor strain, and a third round of transformation did not increase the MIC further. Interestingly, five of the ten lineages had seven or more strains with MICs of 1 μg/mL or higher. In contrast, the AMR37 lineage resulted in MICs from 0.25-0.5 μg/mL, with one exception at 2 μg/mL. These lineage-dependent differences in AMX MIC range (p = 5 x 10 −4 , Fisher's Exact Test) suggested that mutations acquired in the first round contributed to the recombination events which occurred in the second round, and thus to the MICs of the resulting strains.
Similar lineage-dependent MIC patterns were seen when TIGR4 was used as the recipient strain (p = 1.00x10 -3 , Fisher's Exact Test) [fig_ref] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX... [/fig_ref]. Here, the AMX MICs of the first-generation ranged from 0.023-0.5 μg/mL, with one strain showing a more than 30-fold increase from wild-type TIGR4. In the second-generation, three recombinant strains reached the same MIC as the donor strain (4 μg/mL), suggesting potential key differences in the recipient strain genomes, or deleterious epistatic effects of AMX resistance mutations which affected resistance development through recombination.
## Cell lysis upon amx treatment correlated with pbp affinities
To test whether transformed strains phenocopy the donor with regards to AMX resistance, phase contrast microscopy with an AMX concentration above the recipient MICs (1 μg/mL) was performed. A complete stall in growth was observed for both recipient strains and firstgeneration recombinant AMR38, with mild to severe lysis after 6 hrs of treatment . In
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae contrast, donor strain 11A grew to form a microcolony in this AMX concentration, while second-generation AMR53 grew more slowly and with abnormally elongated cells .
To test whether the increased resistance of AMR53 is related to alterations in the affinity of the PBPs towards AMX, we performed Bocillin-Fl labelling. Bocillin-Fl is a fluorescently labelled penicillin derivative that binds all six PBPs of S. pneumoniae [bib_ref] BOCILLIN FL, a sensitive and commercially available reagent for detection of penicillin-binding..., Zhao [/bib_ref] , and the intensity of labelling can be used as a proxy for the affinity of the PBP for β-lactam antibiotics [bib_ref] Profiling of β-lactam selectivity for penicillin-binding proteins in Streptococcus pneumoniae D39, Kocaoglu [/bib_ref]. All six PBPs of D39V and TIGR4 had a high affinity for the fluorescently labelled penicillin . Interestingly, TIGR4 showed some evidence of Pbp1a degradation products that bound Bocillin-Fl with high affinity even though it is 99.7% identical to D39V Pbp1a. In contrast, 11A showed an altered PBP pattern, where Pbp2b and Ppb1a showed complete loss of affinity, and Pbp2x was labelled more faintly and migrated more slowly .
Both first-and second-generation recombinants demonstrated a loss of affinity of Pbp2b and Pbp2x for Bocillin-FL, with Pbp1a in AMR53 also showing reduced affinity, and potentially some degradation . Together, these characterizations demonstrate that transformed recipients phenocopy the donor AMX resistant strain and have acquired PBPs with reduced affinity towards bocillin-FL. were transformed with genomic DNA from an AMX resistant clinical isolate then selected on a range of AMX concentrations. Ten colonies were picked at random from the highest concentration with growth (first round on 0.03 μg/mL and second round on 0.17-0.3 μg/mL, depending on MIC of the recipient strain). Isolated transformants with increased MICs were then subjected to a second round of transformation. (B) Pedigree chart of recombinant strains derived from D39V, and (C) from TIGR4. Color represents AMX MIC. Strains which were sequenced and used throughout the study are labelled.
https://doi.org/10.1371/journal.ppat.1010727.g002
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae
## Recombination events clustered around resistance-associated loci
In order to examine the genetic differences in recombinant strains that may explain the differences in MIC, we selected eight recombinant strains from each experiment, and Illumina sequenced to~600-fold mean coverage. For the D39V experiment, we chose AMR38 and three of its descent strains, as this lineage had the highest MICs. To contrast, we also chose AMR37 and three descendent strains, which had the lowest and most variable MICs in the second generation [fig_ref] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX... [/fig_ref]. We used similar selection criteria for the TIGR4 experiment, taking strains from the high MIC EB7 lineage, and from the low MIC EB3 lineage [fig_ref] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX... [/fig_ref].
To be able to accurately map recipient to donor, we also performed long-read PacBio sequencing followed by short-read polishing on the 11A donor and TIGR4 recipient (an accurate genome map was already available for D39V [bib_ref] Deep genome annotation of the opportunistic human pathogen Streptococcus pneumoniae D39, Slager [/bib_ref]. Recombination was detected using SNPs, as described in [bib_ref] Evolution via recombination: Cell-to-cell contact facilitates larger recombination events in Streptococcus pneumoniae...., Cowley [/bib_ref]. On average 0.38% ± 0.22% of the donor genome was transferred during a single transformation event, and the total percentage transferred after two rounds was 0.80 ± 0.05%. Although higher numbers of recombination events were detected in the D39V dataset, the average event length was larger in the TIGR4 dataset (D39V 0.79 Kb and TIGR4 2.29 Kb, S3 and S4 Tables), resulting in similar total amounts of donor genome transfer after two rounds of transformation (D39V 0.79 ± 0.23%, TIGR4 0.80 ± 0.16%, S3 and S4 Tables). Of note, the longest recorded recombination fragment was 4.00 Kb for D39V and 12.52 Kb for TIGR4.
Recombination events clustered around key loci for pneumococcal β-lactam resistance in both datasets; pbp2x, pbp2b, pbp1a and murM (Figs 4A and 5C). SNPs were also found at closely located loci, likely due to linkage. In the D39V strains, there was extensive recombination at distantly located sites. Examples include ptvABC, associated with vancomycin tolerance [bib_ref] Transcriptional repressor PtvR regulates phenotypic tolerance to vancomycin in Streptococcus pneumoniae, Liu [/bib_ref] , and murT, which performs an amidation step essential for peptidoglycan crosslinking [bib_ref] Identification and in vitro Analysis of the GatD/MurT Enzyme-Complex Catalyzing Lipid II..., Mü Nch [/bib_ref] [bib_ref] Identification of Genetic Determinants and Enzymes Involved with the Amidation of Glutamic..., Figueiredo [/bib_ref].
In contrast, of the six total recombination loci in the TIGR4 dataset, only three were located distally from the pbp and murM genes, and all were detected based on the presence of very few SNPs.
We hypothesised that differences in recombination patterns could have arisen from reduced similarity with the donor genome, however comparing total shared and unique SNPs did not suggest major differences in variability between the two recipient genomes [fig_ref] Fig 5: Uptake of four donor fragments is sufficient for AMX resistance in TIGR4 [/fig_ref]. When considering variation around key loci, we saw that the TIGR4 pbp2b region was more divergent from the donor than D39V. The quality of donor DNA used may have also played a role. Together, these results demonstrate that regardless of the recipient strain, four loci are consistently acquired in high AMX resistant recombinants, namely pbp2x, pbp2b, pbp1a and murM [fig_ref] Fig 5: Uptake of four donor fragments is sufficient for AMX resistance in TIGR4 [/fig_ref].
Structural variation and base modification detection with PacBio sequencing revealed no major structural changes or differences in methylated motifs in the D39V recombinant strains. However, a local rearrangement was identified at the hsdS-creX locus in EB7, EB28, and EB29. In this region, spontaneous reshuffling of hsdS genes by CreX results in different methyltransferase specificity of the type I restriction modification system SpnD39III [bib_ref] Deep genome annotation of the opportunistic human pathogen Streptococcus pneumoniae D39, Slager [/bib_ref] [bib_ref] A random six-phase switch regulates pneumococcal virulence via global epigenetic changes, Manso [/bib_ref] [bib_ref] A statistical model-based tool to reveal intercellular heterogeneity of DNA modification from..., Feng [/bib_ref]. TIGR4 was found to be predominantly in the B-configuration, EB7 and EB29 in the D-configuration, and EB28 in the C-configuration, while all D39V-derived strains and 11A were mostly in the F-configuration. Differences in methylated motifs have transcriptomic consequences [bib_ref] A random six-phase switch regulates pneumococcal virulence via global epigenetic changes, Manso [/bib_ref] , however MIC differences among these recombinants can largely be explained by SNPs in key loci for β-lactam resistance.
## Non-contiguous recombination events likely result from a single donor molecule
requirement for homology is no longer met, at which point the interaction ends. However, we detected recombination events from one round of transformation located very close together on the chromosome, sometimes within the same gene but with separating recipient SNPs, a motif reminiscent of clinically resistant mosaic pbp alleles. This has also been noted previously in both in vitro pneumococcal and H. influenzae recombination studies and was hypothesized to result from a single donor molecule [bib_ref] Transformation of natural genetic variation into haemophilus influenzae genomes, Mell [/bib_ref]. To determine if events were significantly close together, forming a single non-contiguous recombination event, a bootstrapping approach was applied to the distances between novel recombination events within the same strain . In this approach, the shortest distances between all recombination events for all strains were pooled to form the test population. Bootstrap tests were performed 1000 times, where the test population was sampled with replacement, and the resulting distribution compared to the shortest recombination distance for each strain (d test ). If d test was below the 0.05 quantile value in at least 95% of bootstrap tests, then the events were considered to be linked, and the next closest pairing was tested. We found 37 events across all strains that could be amalgamated into single events under the assumptions of this approach (S2 [fig_ref] Fig 4: Widespread transformation and uptake of 11A donor DNA in recipient D39V upon... [/fig_ref].
In addition, using PCR amplified fragments of the pbp2x-mraY locus from 11A, we were able to test, for the first time, whether non-contiguous recombination events occurred more frequently when the template was provided as a single molecule, or as two overlapping fragments. By ligating additional regions of non-homologous DNA, we also tested the effect of molecule length on recombination. a list of all detected events). Recombination events determined to be non-contiguous are marked with ( � ). Non-contiguous events detected in the first-generation were removed from the data set on the second-generation strains, so are not marked. (B) Coverage of reads mapped competitively to the donor genome and plotted by coordinate beginning at the origin. Recombination events cluster at known β-lactam resistance determinants, pbp2x, pbp2b, murM, and pbp1a.
https://doi.org/10.1371/journal.ppat.1010727.g004
## Plos pathogens
## Origins of amoxicillin resistance in streptococcus pneumoniae
Transformation efficiency increased with length of both the overall molecule, and the homologous region [fig_ref] Fig 6: Fragmented transformation of donor DNA predominantly originates from non-contiguous recombination events [/fig_ref]. This is in line with previous studies showing the importance of molecule length and homology in pneumococcal transformation [bib_ref] Transformation and deoxyribonucleic acid size: extent of degradation on entry varies with..., Morrison [/bib_ref]. We hypothesize that increased length may increase the likelihood of spontaneous contact with the ComGC pilus, while also decreasing the likelihood of homologous region fragmentation by EndA, prior to entry. The efficiencies of the short overlapping fragments were lower than that of the long fragments, although increased slightly when transformed in combination.
Recombination occurred in all colonies sequenced. When two overlapping fragments were provided, uptake was often unequal, with either the upstream or downstream fragment incorporated more often. Recombination of both fragments was observed in 1/20 cases and was not observed when fragments were lengthened by non-homologous DNA (S3-S5 Figs and S7 .
Non-contiguous recombination was detected in all conditions, although the frequency of occurrence increased with length of homology (Figs 6C and S3, S4, S5 and S7 . It was observed both when the template was given as a single long fragment, and as two overlapping fragments. a list of all detected events). Only six loci with recombination were identified in the entire data set, four of which are known β-lactam resistance determinants, pbp2x, pbp2b, murM, pbp1a.
https://doi.org/10.1371/journal.ppat.1010727.g005
## Plos pathogens
## Origins of amoxicillin resistance in streptococcus pneumoniae
Overall, the results suggested that the formation of multiple D-loops from either a single template molecule, or from multiple molecules, could lead to non-contiguous recombination events [fig_ref] Fig 6: Fragmented transformation of donor DNA predominantly originates from non-contiguous recombination events [/fig_ref]. However, given that longer homology and total molecule length increased the efficiency of transformation partly by reducing strand degradation [bib_ref] Transformation and deoxyribonucleic acid size: extent of degradation on entry varies with..., Morrison [/bib_ref] , there may have been more opportunity for complex recombination patterns to occur when longer fragments were taken up.
## Amx resistant pbp and murm alleles are sufficient to explain amx mics
Substitutions in all four proteins (Pbp1a, Pbp2b, Pbp2x and MurM) were required to reach an MIC close or equal to that of the donor strain (S6 [fig_ref] Fig 5: Uptake of four donor fragments is sufficient for AMX resistance in TIGR4 [/fig_ref] , but the order of uptake
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae varied both stochastically, and depending on the recipient genome. The correlation between allele uptake and MIC increase was consistent in the D39V background. First-generation strains had mutations in pbp2b and pbp2x, the two essential PBPs for cell growth [bib_ref] Effects of low PBP2b levels on cell morphology and peptidoglycan composition in..., Berg [/bib_ref]. In the second-generation, all strains acquired mutations in murM, and those with MICs higher than 1 μg/mL also had mutations in pbp1a. In TIGR4, pbp2x alleles were also acquired in the firstgeneration, however pbp2b mutations only appeared after the second round of transformation. Mutations in murM were recombined in EB7 in the first round of transformation, then were inherited vertically in that lineage, but were not acquired by EB3 or any descendant strains [fig_ref] Fig 5: Uptake of four donor fragments is sufficient for AMX resistance in TIGR4 [/fig_ref]. Although murM mutations were not present in all strains with pbp1a modifications, strains with recombination events at all four loci had higher MICs [fig_ref] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX... [/fig_ref]. This indicated that murM mutations may be less important for successful integration and selection for pbp1a mutations but are key for expression of the AMX resistant phenotype of the donor [bib_ref] A biological price of antibiotic resistance: major changes in the peptidoglycan structure..., Garcia-Bustos [/bib_ref] [bib_ref] Alterations in MurM, a cell wall muropeptide branching enzyme, increase high-level penicillin..., Smith [/bib_ref] [bib_ref] Inhibition of the expression of penicillin resistance in Streptococcus pneumoniae by inactivation..., Filipe [/bib_ref] [bib_ref] The murMN operon: A functional link between antibiotic resistance and antibiotic tolerance..., Filipe [/bib_ref].
An experiment was designed to test whether the order of allele uptake in D39V recombinant strains was crucial for AMX resistance development, and if PBP and MurM substitutions alone were sufficient to explain AMX MICs. Alleles of pbp2x and pbp2b from AMR37 (pbp2x37, pbp2b37) and AMR38 (pbp2x38, pbp2b38) were transformed into D39V and selected on AMX. Strains carrying either a recombinant pbp2x or pbp2b allele had E-test MICs up to 0.064 μg/mL. However, when both pbp alleles were integrated (D39V 2x38-2b38 ), the E-test result equalled that of the recombinant allele donor [fig_ref] Fig 7: Uptake of mosaic pbp1a, pbp2b, pbp2x and murM is required and sufficient... [/fig_ref] , suggesting that mutations outside these two loci were not involved in the observed reduction in AMX susceptibility in the first-generation strains.
To assess whether mutated pbp2x and pbp2b were critical for murM mutation uptake, D39V was transformed with murM from both AMR53 (murM53) and AMR95 (murM95) and MICs determined. Surprisingly, we were able to select for colonies with integration of the AMR53 murM allele (D39V M53 ) resulting in an E-test result of 0.023 μg/mL, despite this protein not being a direct target of AMX (Figs 7 and 8A and 8B). The genome was sequenced, and no spontaneous point mutations were identified in previously described resistance determinants outside murM (S8 . When transformed into the D39V 2x38-2b38 , the E-test MICs reached 0.5 μg/mL and 0.38 μg/mL for AMR95 and AMR53 murM alleles (D39V 2x38-2b38-M95 , D39V 2x38-2b38-M53 ), respectively. AMX susceptibility decreased substantially when all three genes were mutated, however failed to reach the final MIC of the recombinant allele donor, indicating the necessity of another resistance determinant [fig_ref] Fig 7: Uptake of mosaic pbp1a, pbp2b, pbp2x and murM is required and sufficient... [/fig_ref].
pbp1a alleles from AMR53 and AMR95 (pbp1a53, pbp1a95) could not be successfully selected on AMX when transformed into D39V, either alone or in conjunction with a murM allele from the same donor. Indeed, Pbp1a is not a key target of AMX, and this result was not unexpected. We instead transformed a construct where a kanamycin resistance cassette was cloned behind pbp1a, with 1 Kb of homology up-and downstream. Growth in liquid culture of the resulting strains at AMX 0.01 μg/mL was similar to D39V [fig_ref] Fig 8: Recombinant PBP proteins visualized by PBP-bocillin binding affinities [/fig_ref]. Interestingly, successful integration occurred for both pbp1a53 and pbp1a95 when transformed into D39V 2x38-2b38 (D39V 2x38-2b38-1a53 , D39V 2x38-2b38-1a95 ), suggesting that a mutated murM allele was not required for mutations in pbp1a to cause a selectable increase in AMX MIC. However, substitutions in Pbp2x, Pbp2b, and Pbp1a alone were insufficient to replicate the E-test results of the allele source strains [fig_ref] Fig 7: Uptake of mosaic pbp1a, pbp2b, pbp2x and murM is required and sufficient... [/fig_ref].
We therefore transformed murM and pbp1a alleles from AMR53 and AMR95 into D39V 2x38-2b38 simultaneously, and found that selection was possible on higher concentrations of AMX (1 μg/mL) than for either gene alone (0.2 μg/mL). The resulting strains (D39V 2x38-2b38-M53-1a53 , D39V 2x38-2b38-M95-1a95 ) had AMX resistance phenotypes matching those of AMR53 and AMR95, with E-test results of 2 μg/mL and 1.5 μg/mL respectively [fig_ref] Fig 7: Uptake of mosaic pbp1a, pbp2b, pbp2x and murM is required and sufficient... [/fig_ref].
The strains containing all four genes from AMR53 and AMR95 were whole genome sequenced to check for spontaneous point mutations outside these loci. One non-synonymous
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae mutation in glnA (glutamine synthase type I) was found in both strains. Downregulation of glnA has been linked to the pneumococcal penicillin stress response [bib_ref] Penicillin induces alterations in glutamine metabolism in Streptococcus pneumoniae, Khoury [/bib_ref] , but no differences in AMX survival were observed in these strains compared to the allele donors lacking the SNP.
To quantify the affinities of resistant PBP alleles to β-lactam antibiotics, Bocillin-FL binding patterns were determined for allele swap strains and the source strains [fig_ref] Fig 8: Recombinant PBP proteins visualized by PBP-bocillin binding affinities [/fig_ref]. This confirmed that the recombinant PBPs had reduced penicillin binding affinities. For Pbp2b, both the AMR37 and AMR38 alleles resulted in almost complete loss of binding, and a band could not be quantified [fig_ref] Fig 8: Recombinant PBP proteins visualized by PBP-bocillin binding affinities [/fig_ref]. A large reduction in affinity was observed for Pbp2x38, larger than for Pbp2x37, which correlated to a small difference in growth at 0.1 μg/mL AMX between the two source strains [fig_ref] Fig 8: Recombinant PBP proteins visualized by PBP-bocillin binding affinities [/fig_ref]. This could be explained by 13 amino acid substitutions including S389L and T338A which were not present in the AMR37 allele but have previously been linked to reduced susceptibility to β-lactams [bib_ref] Profiling of β-lactam selectivity for penicillin-binding proteins in Streptococcus pneumoniae D39, Kocaoglu [/bib_ref].
Perhaps most interesting were the two low affinity Pbp1a alleles tested. In all strains containing pbp1a95 (AMR95 derived), reduced band intensity was observed in the Pbp1a/Pbp1b size range, however, a faster migrating species was also detected. The pbp1a95 allele contained a spontaneous point mutation (not present in the 11A donor) causing a premature stop
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae codon. This resulted in the loss of 65 amino acids (of a total of 720) and put the predicted size at 73.1 kDa instead of 79.6 kDa. Reduced intensity at the expected Pbp1a/Pbp1b size range was also observed for the AMR53 allele (Pbp1a53) however, an additional Bocillin-FL-tagged species which migrated slightly faster than expected was also observed. The predicted protein length of Pbp1a53 is identical to D39V and the donor, with no premature stop codons identified. The additional band may indicate some degradation of this protein species, although the TP domain must still be available for Bocillin-FL binding. Together, these experiments demonstrate that transfer of the pbp2x, pbp2b, pbp1a and murM alleles of 11A into strain D39V is sufficient to explain the observed AMX resistance in transformants obtained using chromosomal DNA.
## High amx resistance in the d39v comes at a significant fitness loss
Of the 16 recombinant strains sequenced, only EB31 and EB28 from the TIGR4 collection had the same AMX MIC as the donor strain, and a third round of transformation with gDNA into AMR53 failed to increase the MIC. The only region of mutations that EB28 and EB31 had in common that was not present in the highest MIC AMR strains, was a block of 10 mutations in the 590-641 region of Pbp2b [fig_ref] Fig 5: Uptake of four donor fragments is sufficient for AMX resistance in TIGR4 [/fig_ref]. To test whether these substitutions were
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae sufficient to explain the MIC difference, the pbp2b allele from EB31 (pbp2b31) was transformed into AMR53 and AMR95. The AMX E-test scores for the resulting strains were both 4-6 μg/mL. However, growth curves revealed a significant fitness loss, with a mean maximum specific growth rate of 0.11 μ/h for AMR95 decreasing almost 100-fold to 0.0036 μ/h for AMR95 2b31 , which was exacerbated by the presence of AMX . In contrast, AMR53 2b31 grew almost 3-fold slower than AMR53 (AMR53 0.12 μ/h, AMR53 2b31 0.046 μ/h, , but showed a lytic phenotype in stationary phase. Interestingly, this strain outgrew the 11A donor in sublethal AMX concentrations . The difference in growth rate between AMR95 2b31 and AMR53 2b31 was likely explained by the truncated Pbp1a95 protein. Significant modifications to the active sites of essential proteins often need to be compensated for, which may be impacted by the sub-optimal performance of other proteins in the system. Although the TP and TG domains of Pbp1a95 were intact, it is possible that the loss of 65 amino acids from the C-terminal resulted in an overall reduction in function, unrelated to its AMX binding affinity. This correlates with previous studies showing that compensation by substitutions in Pbp1a are required to mitigate fitness loss resulting from modifications to the Pbp2b active site . Together, these experiments show that D39V is capable of reaching high AMX resistance, but that the absence of compensation by Pbp1a substitutions may result in significant fitness costs.
# Discussion
Natural competence and homologous recombination in S. pneumoniae allow for enormous genomic plasticity which contributes to the rapid development of vaccine-escape and antibiotic resistant isolates. In one round of transformation, more than 29 recombination events can occur in a single cell [bib_ref] Evolution via recombination: Cell-to-cell contact facilitates larger recombination events in Streptococcus pneumoniae...., Cowley [/bib_ref] , while events up to 50 Kb in length have been identified in clinical isolates [bib_ref] Pneumococcal capsular switching: A historical perspective, Wyres [/bib_ref] , allowing for significant remodelling at multiple loci simultaneously. Indeed, here we show a twostep sequential transformation experiment with up to 20 individual recombination events around the genome and lengths up to 13 Kb, resulting in an average donor genome transfer per step of 0.38%. This allows for huge genomic diversity, which, when placed under selective pressure, can lead to the rapid spread of antibiotic resistance-associated genotypes [bib_ref] Dense genomic sampling identifies highways of pneumococcal recombination, Chewapreecha [/bib_ref].
β-lactam resistance in the pneumococcus relies on DNA movement between closely related strains and species within the same niche, allowing tens of substitutions to accumulate in the target proteins [48]. It has previously been shown through in vitro transformation experiments into a laboratory strain that the main determinants of β-lactam resistance are modified alleles of pbp2x, pbp2b, pbp1a, and murM [bib_ref] Alterations in MurM, a cell wall muropeptide branching enzyme, increase high-level penicillin..., Smith [/bib_ref] [bib_ref] Single-step capsular transformation and acquisition of penicillin resistance in Streptococcus pneumoniae, Trzciński [/bib_ref]. For the front-line antibiotic AMX, resistance has also been previously associated with mutations in these alleles although the relative importance of the changes in each locus is not well understood [54-57,60,61]. We confirmed the requirement for mutations in all four genes and showed that subtle differences between alleles could cause noticeable changes in both AMX-dependent fitness and the penicillin-binding affinity of PBPs.
In addition, a specific order of mutation uptake for penicillin resistance has been demonstrated in both Staphylococcus aureus and the pneumococcus [bib_ref] Stepwise introduction of transformable penicillin resistance in Pneumococcus, Shockley [/bib_ref] [bib_ref] Transfer of penicillin resistance in pneumococci by the desoxyribonucleate derived from resistant..., Hotchkiss [/bib_ref] , and for AMX this order was found to be pbp2x, pbp2b, then pb1a, with the role of murM unable to be confirmed in that experimental system . Here, we confirm that substitutions in Pbp2x or Pbp2b are a critical first step for AMX resistance, perhaps reflecting the high affinity of Pbp2x for the antibiotic [bib_ref] Profiling of β-lactam selectivity for penicillin-binding proteins in Streptococcus pneumoniae D39, Kocaoglu [/bib_ref] [fig_ref] Fig 7: Uptake of mosaic pbp1a, pbp2b, pbp2x and murM is required and sufficient... [/fig_ref] , and that the T338A substitution is required for resistance to fully develop. We also find that murM mutations appear to be an intermediary step, and that a modified pbp1a allele is important for high-level AMX resistance. It is however important to note that these experiments were performed using high concentrations of naked DNA as a model, while in the nasopharynx genetic exchange likely occurs via fratricide or from small quantities of
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae free environmental DNA. It may be that lower donor DNA concentrations cause slower resistance allele acquisition, and result in lower fitness costs. In addition, previous studies have strongly implied the existence of multiple mechanisms of AMX resistance in S. pneumoniae, associated with different pbp2b and murM alleles [52,54,56], and that by using a single donor strain we were limited to the study of only one. However, we confirmed that a previously described block of ten substitutions in the 590-641 region of the Pbp2b TP domain [54-56] was essential to reconstruct the AMX MIC of the serotype 11A ST6521 donor strain in D39V and TIGR4. Perhaps most interestingly, we also show that the loss of 65 amino acids from the C-terminus of Pbp1a results in a significant fitness cost in the presence of these additional Pbp2b substitutions, highlighting the importance of interactions and fitness loss compensation between essential cell-wall synthesis proteins .
In addition to these definite experiments demonstrating the evolutionary route to clinical AMX resistance, we show that non-contiguous recombination could result from either a single, or multiple, donor DNA molecules. It is easy to imagine a long ssDNA-RecA presynaptic filament interacting with the chromosome at multiple points in the 3-dimensional cellular space [bib_ref] Single-molecule imaging of DNA pairing by RecA reveals a threedimensional homology search, Forget [/bib_ref] [bib_ref] Mechanism of strand exchange from RecA-DNA synaptic and D-loop structures, Yang [/bib_ref] , which then initiates the formation of several D-loops simultaneously as homologous DNA-DNA interactions stabilize [bib_ref] Regulation and Mechanism of a, Bell [/bib_ref] [bib_ref] The synapsis event in the homologous pairing of DNAs: RecA recognizes and..., Hsieh [/bib_ref] [bib_ref] The specificity of the secondary DNA binding site of RecA protein defines..., Mazin [/bib_ref]. On the other hand, SsbB coated ssDNA sequestered in the cytoplasm could provide a pool from which multiple donor molecules homologous for the same chromosomal region could be sourced [bib_ref] Role of the singlestranded dna-binding protein Ssbb in pneumococcal transformation: Maintenance of..., Attaiech [/bib_ref]. However, we also observed that both transformation efficiency and non-contiguous event occurrence increased with length of homology and total length of the donor molecule, suggesting that single donor molecule events may simply be more likely. This concurs with the previous observation of transforming DNA concentration-independent differences in recombination event density, which would not have been expected under a multiple donor molecule hypothesis . Recombination structures such as these have also been observed in H. influenzae, with the authors noting that localized clustering of long tracts of donor sequence appeared too close to have occurred by chance [bib_ref] Transformation of natural genetic variation into haemophilus influenzae genomes, Mell [/bib_ref]. They hypothesized incoming DNA degradation via cytosolic or translocation endonucleases resulted in these clusters. In addition, mismatch repair via the Hex system could in principle be responsible for the reversion of small numbers of point mutations between non-contiguous recombination segments, although the system was likely saturated in our model . To untangle the different possible mechanisms of non-contiguous recombination from other effects on transformation efficiency, testing mixtures of different donor fragment lengths would be necessary.
Altering the active sites of essential proteins in cell division is risky and requires a delicate balance between antibiotic avoidance and fitness loss. Accumulation of sufficient mutations in the appropriate order is aided both by the simultaneous recombination of multiple donor fragments distally around the genome, and by the occurrence of complex non-contiguous events. The experimental transformation experiments combined with whole genome sequencing as performed here provide valuable insights into viable and non-viable evolutionary paths toward AMX resistance development in the pneumococcus. The strong co-dependence among different proteins and their alleles creates significant challenges for complete understanding of resistance mechanisms, and emphasizes the importance of combining data from phenotypic, molecular, genomic, and genome-wide association studies.
# Methods
## Bacterial strains, growth conditions, and antibiotics
S. pneumoniae strains are shown in Tables 1 and S1. Bacteria were cultured in liquid C+Y medium with no shaking at 37˚C. C+Y media was adapted from Adams and Roe.
## Plos pathogens
## Origins of amoxicillin resistance in streptococcus pneumoniae
For transformation, S. pneumoniae was grown in C+Y (pH 6.8) at 37˚C to OD 595 0.1. Competence was induced by addition of 100 ng/ml synthetic CSP-1 (D39V) or CSP-2 (TIGR4) and 12 min incubation at 37˚C. DNA uptake occurred during 20 min at 30˚C, followed by dilution and recovery for 1.5 hr. Transformants were selected by plating inside Columbia agar supplemented with 3% defibrinated sheep blood (CBA, Thermo Scientific) containing the appropriate antibiotic and incubated at 37˚C in 5% CO 2 overnight. Successful transformants were confirmed by PCR and Sanger sequencing (Microsynth).
Strains were stocked at OD 595 0.3 in C + Y with 15% glycerol at -80˚C. AMX (Sigma Aldrich) powder was dissolved initially in 100% DMSO. This solution was diluted in molecular grade water with 4% DMSO to two concentrations, 100 μg/mL and 1 mg/ mL, and stored at -80˚C.
## Minimum inhibitory concentrations (mics)
Bacteria were grown in C+Y broth to OD 595 0.1, then AMX MIC was determined using E-tests (Biomeriaux, Lyon) on Mueller Hinton agar containing 5% defibrinated sheep's blood. MICs were read after 18 h incubation at 37˚C and 5% CO 2 .
Strains with MICs below the limit of the E-tests (0.016 μg/mL) were tested in Mueller Hinton cation-adjusted broth with 5% defibrinated sheep's blood using the broth micro-dilution method. Briefly, bacteria were streaked onto MH2-blood plates and grown overnight. Colonies were picked from plates and ressupended in PBS (pH 7.4) to McFarlane standard 0.5. This inoculum was diluted 100-fold into 96-well plates containing MH2-blood broth and AMX (in 2-fold concentration steps) then plated on agar plates as a control for contamination and cell density. Plates were incubated for 18 hours at 37˚C and 5% CO 2 . MIC was defined as the lowest concentration of antibiotic where visible α-haemolysis occupied approximately less than 10% of the surface area of the well. Due to a high level of subjectivity, MIC determination was repeated twice on different days, and MIC calls were confirmed by a second person in each case.
## Growth curves and analysis
Strains were precultured in C+Y broth to OD 0.1, then diluted 100-fold into 96-well microtiter plates, containing fresh C+Y and AMX. The OD 595 was measured every 10 min for 12 hr in a plate reader (Infinite F200, Tecan) with incubation at 37˚C. All strains were tested in triplicate
The area under the curve (AUC) of each replicate was divided by that of the no antibiotic control for each strain, to normalize for strain-dependent differences in growth. The mean and standard error of the mean (SEM) of the normalized AUC was determined for three biological replicates.
Statistical significance of maximum specific growth rate (μ/h) was determined using oneway ANOVA with a Tukey correction for multiple testing.
## Bocillin-fl binding assay
Bocillin-FL is a fluorescently labelled penicillin molecule which can be used to visualise and quantify PBP-penicillin binding affinity in vitro. Strains were grown to OD 0.2 in 4 mL C+Y, washed and resuspended in phosphate buffered saline (PBS), then incubated for 30 min at 37C in 5 μg/mL bocillin-FL (Invitrogen). Cells were washed in ice cold PBS, then lysed by sonication (10 x 1s pulses at 30% power). The cell lysate was spun down and supernatant removed to collect the membrane fraction, which was homogenized by sonication (1s pulse at 10%). Total protein concentration was determined by nanodrop, then samples were diluted in NuPAGE LDS sample buffer (Life Technologies) with 10 mM DTT and incubated at 95˚C for 5 min.
## Plos pathogens
## Origins of amoxicillin resistance in streptococcus pneumoniae
Samples were loaded into two 15-well Novex WedgeWell 10% Tris-Glycine 1.0 mm Mini Protein Gels (Invitrogen), to a total protein amount of approximately 1.8 mg in a maximum volume of 30 μL. Where necessary, gels were imaged simultaneously on an Amersham Typhoon (GE Healthcare) with Cy2 DIGE filter setup. Bands were quantified using ImageQuant (GE Healthcare). As all strains tested had identical Pbp3 proteins, the intensities of the Pbp1a/b, Pbp2x/a, and Pbp2b bands were divided by the PBP3 band intensity of the corresponding sample for normalisation. Imaging and quantification were performed in duplicate. Coomassie Blue staining was also performed, to confirm equal sample loading.
## Phase-contrast time-lapse microscopy
For time-lapse microscopy, cells were pre-cultured in liquid C + Y medium at 37˚C and spotted on a low-melting 1.2% C + Y agarose patch, with or without AMX at 1 μg/mL. Phase-contrast pictures were taken every 10 min using a Leica DMi8 with a 100x phase contrast objective.
## Isolation of genomic dna
10 mL of OD 595 0.2 culture was pelleted by centrifugation, then resuspended in Nuclei Lysis solution (Promega) supplemented with 0.05% SDS, 0.025% deoxycholate (DOC), and 200 μg/ mL RNase A. This was incubated at 37˚C for 20 minutes, 80˚C for 5 minutes, then 37˚C for 10 minutes. Protein Precipitation Solution (Promega) was added to the lysate, vortexed vigorously, then incubated on ice for 10 minutes. The precipitated protein was pelleted by centrifugation and the supernatant transferred into isopropanol to precipitate the DNA, which was then collected by centrifugation. DNA was washed once in 70% ethanol before air drying and resuspension in the appropriate buffer (either TE buffer or molecular grade water). DNA was run on a 1% agarose gel to confirm no significant degradation had occurred, and stored at 4˚C.
## Serial transformation of s. pneumoniae with genomic dna
AMX susceptible recipient strains were transformed with 100 μg/mL genomic DNA isolated from AMX resistant strain 11A [fig_ref] Table 1: Key strains used in this study [/fig_ref] as above, with the following modifications. DNA uptake occurred during 30 min, followed by recovery for 2 hrs. Transformants were selected by plating inside Columbia agar supplemented with 5% defibrinated sheep blood and a range of AMX (0.03 μg/mL). Ten colonies were then picked at random and streaked onto CBA plates containing AMX. The next day, single colonies were grown up in C+Y with AMX and stocked. MICs were determined for these 10 strains, which were then subjected to another round of transformation and selection on a range of AMX concentrations (0.17-0.5 μg/mL), and 5-10 single colonies were picked for each of the 10 strains transformed. These were isolated, and MICs determined, as above. To test for significant differences between MICs of second-generation strains in different lineages, a Fisher's Exact Test with Monte Carlo p-value simulation was used.
To control for spontaneous mutations arising during transformation and recombination, cells transformed with either no DNA, or with D39V genomic DNA, were also plated in CBA with AMX (0.03). Neither resulted in visible colonies.
## Whole genome sequencing and quality control
Genomic DNA from recombinant strains from each recipient background, as well as 11A and TIGR4 were Illumina sequenced (PE150, Novogene and Eurofins Genomics), with a mean
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae coverage of 600-fold. Reads were trimmed using Trimmomatic, then assembled using SPAdes [bib_ref] Assembling Genomes and Mini-metagenomes from Highly Chimeric Reads, Nurk [/bib_ref]. Assembly quality was assessed with Quast [bib_ref] QUAST: quality assessment tool for genome assemblies, Gurevich [/bib_ref] , and read depth was determined in samtools [bib_ref] The Sequence Alignment/Map format and SAMtools, Li [/bib_ref]. Previously published D39V Illumina reads were filtered as above and used where necessary (GEO accessions GSE54199 and GSE69729) [bib_ref] Antibiotic-induced replication stress triggers bacterial competence by increasing gene dosage near the..., Slager [/bib_ref]. Allelic exchange strains (D39V M53 , D39V 2x38-2b38-M53-1a53 , D39V 1a53 , D39V 2x38-2b38-M95-1a95 ) were treated in the same way, then variants were called using bbmap, Freebayes, and vcftools [bib_ref] The variant call format and VCFtools, Danecek [/bib_ref] , in order to check for random mutations arising elsewhere in the genome.
11A, TIGR4, and a subset of recombinant strains were sent for PacBio sequencing (Sequel I and II, Lausanne Genomic Technologies Facility). Demultiplexing and quality control were performed using SMRTlink (PacBio), then reads shorter than 1000 bp were removed with filtlong. Recombinant strain reads were mapped to the recipient reference genome using NGMLR and structural variants (SVs) called with sniffles with default settings, except with read mapping quality > 40 and read depth > 50 [bib_ref] Accurate detection of complex structural variations using single-molecule sequencing, Sedlazeck [/bib_ref]. Imprecise SVs were then filtered out. DNA methylation detection was performed using motifMaker [bib_ref] Characterization of DNA methyltransferase specificities using single-molecule, real-time DNA sequencing, Clark [/bib_ref].
Raw read files are deposited with NCBI Sequence Read Archive under BioProject PRJNA789167 (Individual SRA accession numbers for recombinant strains can be found in S4 .
## Hybrid assemblies of reference genomes
As PacBio sequencing depth was greater than 200x, genomes were assembled in a long-read only manner using Trycycler [bib_ref] Trycycler: consensus long-read assemblies for bacterial genomes, Wick [/bib_ref] , followed by short-read polishing.
Briefly, filtered reads were subsampled 12 times and de novo assembled in four different assemblers, flye [bib_ref] Assembly of long, error-prone reads using repeat graphs, Kolmogorov [/bib_ref] , raven [bib_ref] RAVEN 2.0: A versatile toolbox for metabolic network reconstruction and a case..., Wang [/bib_ref] , wtdbg2 [bib_ref] Fast and accurate long-read assembly with wtdbg2, Ruan [/bib_ref] , and miniasm plus minipolish [bib_ref] Minimap2: pairwise alignment for nucleotide sequences, Li [/bib_ref]. Some assemblies were removed during the Trycycler cluster and reconcile steps, leaving the final consensus sequence to be built from seven to eight assemblies. The consensus assembly was subjected to rounds of polishing with PacBio reads using pbmm2 and pbgcpp (PacBio Toolkit), followed by one round of short-read polishing with Pilon [bib_ref] An Integrated Tool for Comprehensive Microbial Variant Detection and Genome Assembly Improvement, Walker [/bib_ref].
Genomes were annotated in Prokka [bib_ref] Prokka: Rapid prokaryotic genome annotation, Seemann [/bib_ref] using Barrnap and Aragorn [bib_ref] ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide..., Laslett [/bib_ref] and are deposited at NCBI under BioProject PRJNA789167.
## Recombination detection with ngs
Competitive mapping of filtered illumina reads onto a pseudogenome containing both recipient and donor chromosomes was used for crude visualization of recombination loci (bbmap) [bib_ref] Neighbor predation linked to natural competence fosters the transfer of large genomic..., Matthey [/bib_ref]. In order to characterize these loci more accurately, filtered reads of recombinant and recipient strains were mapped onto the 11A hybrid genome assembly using smalt, then variants were called with bcftools [bib_ref] A statistical framework for SNP calling, mutation discovery, association mapping and population..., Li [/bib_ref]. SNPs were filtered as described previously [bib_ref] Evolution via recombination: Cell-to-cell contact facilitates larger recombination events in Streptococcus pneumoniae...., Cowley [/bib_ref] , and used to build a consensus alignment of donor, recipient, and recombinant strains. The consensus alignment was run through a python script designed to identify SNPs between recombinant and donor which are not shared with the recipient genome. Once identified, the match was extended until the donor and recombinant sequences are no longer identical, then the start, end, and length of detected recombination events are extracted [bib_ref] Evolution via recombination: Cell-to-cell contact facilitates larger recombination events in Streptococcus pneumoniae...., Cowley [/bib_ref].
## Identifying non-contiguous recombination events
Many recombination events were located very close together, with sometimes only a few recipient allele SNPs separating the two, suggesting a single donor DNA molecule as the source. To determine if events were significantly close together, a bootstrapping approach was applied . Firstly, the distances between one recombination event and each of the other recombination events in a single genome were calculated. The circular chromosome gives two values, and the shortest distance between any two events was added to the test population. This was calculated for every sequenced strain in the D39V and TIGR4 recombinants, and for secondgeneration strains any prior events carried through from the first-generation were removed from the dataset. The shortest recombination distance within a strain became the test value (d test ), and then the test population was sampled (with replacement).
The distribution of a bootstrapped sample (same size as the test population) was used to test the hypothesis that d test was significantly shorter than expected under the null hypothesis that recombination events are located at random around the chromosome. A Holm correction was applied to account for multiple testing, based on the number of recombination events in the strain. Bootstrap tests were performed one thousand times for each d test value and considered significantly short if the null hypothesis was rejected at least 95% of the time. If a d test value was significant, then the distance between the next shortest pairing within the strain became d test , and was tested in the same fashion. This continued until a d test where the null hypothesis could not be rejected.
When recombination events were considered significantly close together, they were linked together into non-contiguous recombination events. Kb region of the 11A genome spanning pbp2x was used to test the origin of non-contiguous recombination. This region included an 846 bp stretch with SNPs conferring a small decrease in AMX susceptibility sufficient for selection. One long fragment spanning the entire region was amplified with OVL5375 and OVL5378. Two shorter fragments were then amplified, an upstream fragment (Up) with OVL5379 and OVL5375 (2.84 Kb) and a downstream fragment (Down) with OVL5376 and OVL5378 , which overlapped at the SNPs determined to be necessary for selection (S2 and S5 Tables). The three donor DNA templates used in the experiment and their SNPs are shown in [fig_ref] Fig 6: Fragmented transformation of donor DNA predominantly originates from non-contiguous recombination events [/fig_ref].
## Confirming the origins of non-contiguous recombination events in vivo
D39V was transformed as above, with equimolar quantities of each fragment separately, as well as a combination of the two short fragments, and a negative control with no DNA. Transformants were selected on AMX (0.015 μg/mL), and the transformation efficiency was determined by dividing by the total viable count. Individual colonies were restreaked on AMX, then single colonies were picked and the 4.25 Kb region of interest was amplified and Sanger sequenced (Microsynth, S2 . Nucleotide sequences were aligned in Mega [bib_ref] Molecular evolutionary genetics analysis across computing platforms, Kumar [/bib_ref].
As molecule length affects transformation efficiency [bib_ref] Three new integration vectors and fluorescent proteins for use in the opportunistic..., Keller [/bib_ref] , an additional experiment was performed, where the amplified fragments were all extended to 6 Kb by the addition of non-homologous DNA amplified from E. coli and ligated to one end using Golden Gate assembly with BsmBI and T4 DNA ligase (NEB, Vazyme). All oligos and the corresponding templates used in the cloning for these experiments are outlined in S6 and a comprehensive oligo list can be found in S2 [fig_ref] Table: Recombinant strains isolated in sequential rounds of transformation with AMX resistant DNA... [/fig_ref] The assemblies were then amplified to ensure high quality of transforming fragments, resulting in three fragments of equal size, with differing lengths of homologous regions to the pbp2x locus. These were then transformed, amplified, and Sanger sequenced as above, and the transformation efficiency determined.
## Gene and protein alignments
Nucleotide sequences of pbp and murM genes were translated using Emboss:TransSeq [bib_ref] The EMBL-EBI search and sequence analysis tools APIs in 2019, Madeira [/bib_ref]. Alignments were performed with Muscle [bib_ref] Multiple sequence alignment with high accuracy and high throughput, Edgar [/bib_ref]. Phylogenetic trees were produced using concatenated protein alignments of Pbp2x, Pbp2b, Pbp1a, and MurM as input for FastTree (default settings), and visualised in ITOL [bib_ref] Computing large minimum evolution trees with profiles instead of a distance matrix, Price [/bib_ref] [bib_ref] Interactive tree of life (iTOL) v5: An online tool for phylogenetic tree..., Letunic [/bib_ref].
## Amplification and transformation of pbp and murm loci for allelic exchange experiments
Genes for pbps were amplified with Phanta Max Super-Fidelity DNA Polymerase (Vazyme) and primers ordered from Sigma (S2 . Equimolar quantities of PCR products were used for transformation. To avoid transforming mutations in murN alongside those in murM, alleles at this locus were cloned into a construct with D39V flanking sequences. The upstream and downstream fragments were amplified from D39V with OVL5540/OVL5779 and OVL5780/OVL5541 respectively, and the murM allele from AMR53 or AMR95 with OVL5777/OVL5778. Fragments were then assembled using BsmBI and T4 DNA ligase.
pbp2x and pbp2b alleles from AMR37 and AMR38 were transformed into D39V as above, selected on 0.012 μg/mL, 0.015 μg/mL, 0.02 μg/mL AMX, and confirmed by Sanger sequencing. D39V 2x38-2b38 [fig_ref] Table 1: Key strains used in this study [/fig_ref] was then used as the recipient for murM and pbp1a alleles from AMR53 and AMR95, with selection on 0.2 μg/mL, 0.5 μg/mL and 1 μg/mL AMX.
Supporting information S1
## Plos pathogens
Origins of amoxicillin resistance in Streptococcus pneumoniae
[fig] Fig 2: Transformation as an experimental tool to track the evolutionary path to AMX resistance. (A) Experimental design. Susceptible D39V and TIGR4 [/fig]
[fig] Fig 4: Widespread transformation and uptake of 11A donor DNA in recipient D39V upon AMX selection. (A) Circos plot with detected recombination events in D39V derived strains, arranged from lowest to highest MIC from the center. AMR38 and all derived strains are underlined, AMR37 and all derived strains are not. Grey indicates the sequences matches that of the recipient D39V, while recombination events are colored. Events acquired in the first round of transformation are shown in green, in the second round in blue, and lengths were increased to aid in visualization (see [/fig]
[fig] Fig 5: Uptake of four donor fragments is sufficient for AMX resistance in TIGR4. (A) Upset plot of shared and unique SNPs between each recipient and the donor. Bars on the left show the total number of SNPs for each recipient genome compared to the donor. In the central bar plot, from left to right, SNPs shared between D39V and TIGR4, SNPs found only in D39V, and SNPs found only in TIGR4. Although the two recipients differ from the donor strain by similar amounts of SNPs, only about half are unique. (B) SNP counts in pbp/murM genes, including 5 Kb up-and downstream. Number of SNPs unique to each recipient are shown in blue, and shared in light grey. (C) Circos plot showing detected recombination events in TIGR4 derived strains, arranged from lowest to highest MIC from inside to out. EB7 and all derived strains are underlined, EB3 and all derived strains are not. Grey indicates the sequences matches that of the recipient D39V, while recombination events are colored. Events acquired in the first round of transformation are shown in green, and in the second round in blue, and lengths were increased to aid in visualization (see [/fig]
[fig] Fig 6: Fragmented transformation of donor DNA predominantly originates from non-contiguous recombination events. Three fragments were PCR amplified from the 11A strain pbp2x locus, one long 4.2 Kb fragment, and two overlapping short fragments (up 2837 bp, and down 2260 bp). In addition, the same fragments were ligated to stretches of non-homologous DNA, taking all total lengths to 6 Kb (long-NH, NH-up, and down-NH). (A) Efficiencies of transformation of fragments into D39V with selection on AMX (0.015 μg/mL). Statistically significant differences in efficiency compared to the long fragment was determined by t test with BH correction for multiple testing. NS not significant, � indicates p < 0.05.(B) Alignment of the homologous region of template fragments used in the experiment. Donor SNPs are shown in blue, while bases matching the recipient (D39V) are shown in grey. (C) Alignment of 4.2 Kb locus from colonies transformed with different fragments. Bases shared with donor but not with recipient are shown in blue. (D) Schematic showing hypothesized molecular mechanisms for non-contiguous and contiguous recombination. As shown in panel B, both events can take place, but based on transformation efficiencies, the process on the left with long donor DNA is more efficient. https://doi.org/10.1371/journal.ppat.1010727.g006 [/fig]
[fig] Fig 7: Uptake of mosaic pbp1a, pbp2b, pbp2x and murM is required and sufficient for clinically relevant amoxicillin resistance in S. pneumoniae. (A) AMX MICs determined by E-test for D39V transformant strains (solid circles) where pbp and murM alleles from different source strains (empty circles) were transformed in combination. Color shows allele source. (B) Schematic overview of the optimal road to AMX resistance development through horizontal gene transfer. Low affinity pbp2x was taken up in the first round of transformation in both datasets, whereas donor murM and pbp2b alleles were acquired in the first or second round, depending on recipient and lineage. Low affinity pbp1a was recombined in the second round, and only found in strains with AMX MICs above 0.75 μg/mL. In principle uptake of low affinity and modified alleles could be achieved in a single (unlikely) or double step.https://doi.org/10.1371/journal.ppat.1010727.g007 [/fig]
[fig] Fig 8: Recombinant PBP proteins visualized by PBP-bocillin binding affinities. Transformed AMX resistant strains produce mosaic PBPs with reduced affinity for Bocillin-FL. Each column shows the phenotypic characterization of one pbp and murM allele swap strain. (A) Table showing the presence (+) and absence (-) of alleles from different source strains (colour). Allele source strains are also included and are indicated by the unfilled circles. (B) Survival in different AMX concentrations quantified by determining the area under the growth curve (AUC) after 12 hours growth in [/fig]
[fig] S1: Fig. Fluorescence and coomassie stained images of Bocillin-FL gels. (A) Bocillin-FL labelled cell extracts imaged in an Amersham Typhoon with Cy2 filter setup showing PBP binding affinities, from Fig 3B Coomassie stain of Bocillin-FL gel from Fig 3C Bocillin-FL labelled cell extracts imaged in an Amersham Typhoon Cy2 filter setup from Fig 8D Coomassie stain of Bocillin-FL gel from Fig 8. (TIF) S2 Fig. Identification of non-contiguous recombination from sequenced recombinant strains using a bootstrapping method (Croucher et al. 2012). (A) Distances between recombination events which were found to shorter than expected under a null hypothesis where events occur randomly around the chromosome. (B) Schematic showing how events were treated if they were found to be closer to each other than expected. (C) Histogram of recombination events length across all sequenced recombinant strains if non-contiguous recombination is not taken into account. (D) Histogram of recombination events length across all sequenced recombinant strains if non-contiguous recombination is accounted for. (TIF) S3 Fig. Alignments of pbp2x-mraY from colonies isolated after transformation with different length and combinations of donor fragments to explain the occurrence of non-contiguous recombination. Facets show colonies selected from different donor templates which correspond to Fig 6. Bases which match the recipient (D39V) are coloured grey, those which match the donor (11A) are shown in blue. (A) Long, (B) Up, (C) Down. (TIF) S4 Fig. Alignments of pbp2x-mraY from colonies isolated after transformation with different length and combinations of donor fragments to explain the occurrence of non-contiguous recombination. Facets show colonies selected from different donor templates which correspond to Fig 6. Bases which match the recipient (D39V) are coloured grey, those which match the donor (11A) are shown in blue. (A) Up & Down, (B) Long-NH, (C) NH-Up. (TIF) S5 Fig. Alignments of pbp2x-mraY from colonies isolated after transformation with different length and combinations of donor fragments to explain the occurrence of non-contiguous recombination. Facets show colonies selected from different donor templates which correspond to Fig 6. Bases which match the recipient (D39V) are coloured grey, those which match the donor (11A) are shown in blue. (A) Down-NH, and (B) NH-Up and Down-NH. (TIF) S6 Fig. Dendrogram of strain relationships built from amino acid sequences of Pbp and murM proteins. Alignments of Pbp1a, Pbp2b, Pbp2x, and MurM amino acid sequences were used to construct a phylogenetic tree of donor, recipient, and recombinant strains. AMX MICs are shown on the right, color scale represents MIC and corresponds to Fig 2, where yellow is an AMX MIC of 0.01 μg/mL and dark purple is an MIC of 4 μg/mL. (TIF) [/fig]
[table] Table 1: Key strains used in this study. The full collection of recombinant strains and their AMX MIC can be found in S1 Table.Fig 3. Phenotypic characterization of key strains used in this study. (A)Phase-contrast snapshots of recipient strains D39V and TIGR4, donor strain 11A, and recombinant strains AMR38 (generation 1) and AMR53 (generation 2) on C+Y agarose pads after 30 minutes and 6 hours of treatment with 1 μg/mL AMX. This concentration was above MIC for D39V, TIGR4, and AMR38, and lead to a complete stall in growth and partial lysis. Donor strain 11A grew well in this AMX concentration, although some heterogeneity in cell shape could be observed. Although lower than the AMX MIC for AMR53, this strain showed decreased growth and abnormally elongated cells, indicating cells were stressed. Scale bar is 5 μm. (B) Bocillin-FL binding patterns for the same five strains. Expected PBP band sizes for D39V are shown on the left. Image of Coomassie stained gel can be found in S1Fig). [/table]
[table] Table: Recombinant strains isolated in sequential rounds of transformation with AMX resistant DNA and their associated MICs. (XLSX) S2Table. Primers used in this study. (XLSX) S3 Table. List of all recombination events identified in transformants using SNPs. (XLSX)S4 Table. Summary statistics for the characterisation of SNP and recombination results for sequenced recombinant strains. (A) All predicted events detected when compared to parent strain genome. Potential non-contiguity not accounted for. (B) Total DNA transferred during the final round of transformation for that strain. ( � ) All DNA or SNPs transferred into the original recipient (D39V or TIGR4), through one or two rounds of transformation (depending on the generation). SRA accession numbers for raw Illumina and PacBio reads used for analysis are also provided. (XLSX) S5 Table. Key amino acid substitutions in PBPs and MurM of recombinant strains. Substitutions listed have been associated with β-lactam resistance in the literature previously. Cells coloured green have the same allele as D39V, those coloured blue have the same allele as 11A. Strain AMX MICs are also indicated for reference. (XLSX) S6Table. Cloning design for fragments used to test for single-molecule non-contiguous recombination at the pbp2x-mraY locus. (XLSX) S7 Table. Summary statistics from colonies isolated after transformation with different length and combinations of donor fragments to explain the occurrence of non-contiguous recombination. (XLSX) S8 Table. SNPs detected in D39V M53 . (XLSX)PLOS PATHOGENSOrigins of amoxicillin resistance in Streptococcus pneumoniae [/table]
[table] 45: Gasc AM, Sicard AM, Claverys JP. Repair of single-and multiple-substitution mismatches during recombination in Streptococcus pneumoniae. Genetics. 1989; 121(1):29-36. https://doi.org/10.1093/ genetics/121.1.29 PMID: 2645195 46. Humbert O, Prudhomme M, Hakenbeck R, Dowson CG, Claverys JP. Homeologous recombination and mismatch repair during transformation in Streptococcus pneumoniae: Saturation of the Hex mismatch repair system. Proc Natl Acad Sci USA. 1995; 92(20):9052-6. Available from: https://www. jstor.org/stable/2368408?seq=1#metadata_info_tab_contents. PMID: 7568071 47. Scheffers DJ, Pinho MG. Bacterial Cell Wall Synthesis: New Insights from Localization Studies. Microbiology and Molecular Biology Reviews. 2005 Dec; 69(4):585-607. Available from: https://pubmed. ncbi.nlm.nih.gov/16339737/. PMID: 16339737 48. Hakenbeck R, Brü ckner R, Denapaite D, Maurer P. Molecular mechanisms of β-lactam resistance in Streptococcus pneumoniae. Vol. 7, Future Microbiology. Future Medicine Ltd London, UK; 2012. p. 395-410. Available from: https://www.futuremedicine.com/doi/10.2217/fmb.12.2. PMID: 22393892 49. Kalizang'oma A, Chaguza C, Gori A, Davison C, Beleza S, Antonio M, et al. Streptococcus pneumoniae serotypes that frequently colonise the human nasopharynx are common recipients of penicillinbinding protein gene fragments from Streptococcus mitis. Microbial Genomics. 2021 Sep 22; 7 (9):000622. Available from: https://pubmed.ncbi.nlm.nih.gov/34550067/. https://doi.org/10.1099/ mgen.0.000622 PMID: 34550067 50. Doern G V, Brueggemann A, Holley HP, Rauch AM. Antimicrobial resistance of Streptococcus pneumoniae recovered from outpatients in the United States during the winter months of 1994 to 1995: results of a 30-center national surveillance study. Antimicrobial Agents Chemotherapy. 1996 May 1; 40(5):1208-13. Available from: http://www.ncbi.nlm.nih.gov/pubmed/8723468. https://doi.org/10. 1128/AAC.40.5.1208 PMID: 8723468 51. Doit C, Loukil C, Fitoussi F, Geslin P, Bingen E. Emergence in France of Multiple Clones of Clinical Streptococcus pneumoniae Isolates with High-Level Resistance to Amoxicillin. Antimicrobial Agents and Chemotherapy. 1999 Jun 1; 43(6):1480-3. Available from: http://www.ncbi.nlm.nih.gov/pubmed/ 10348775. https://doi.org/10.1128/AAC.43.6.1480 PMID: 10348775 52. Càmara J, Cubero M, Martín-Galiano AJ, García E, Grau I, Nielsen JB, et al. Evolution of the β-lactam-resistant Streptococcus pneumoniae PMEN3 clone over a 30 year period in Barcelona, Spain. Journal of Antimicrobial Chemotherapy. 2018; 73(11):2941-51. Available from: https://academic.oup. com/jac/advance-article/doi/10.1093/jac/dky305/5079858. PMID: 30165641 53. Stanhope MJ, Walsh SL, Becker JA, Miller LA, Lefé bure T, Lang P, et al. The relative frequency of intraspecific lateral gene transfer of penicillin binding proteins 1a, 2b, and 2x, in amoxicillin resistant Streptococcus pneumoniae. Infection, Genetics and Evolution. 2007 Jul 1; 7(4):520-34. Available from: https://www.sciencedirect.com/science/article/pii/S1567134807000408. https://doi.org/10.1016/ j.meegid.2007.03.004 PMID: 17475572 54. Cafini F, del Campo R, Alou L, Sevillano D, Morosini MI, Baquero F, et al. Alterations of the penicillinbinding proteins and murM alleles of clinical Streptococcus pneumoniae isolates with high-level resistance to amoxicillin in Spain. Journal of Antimicrobial Chemotherapy. 2006 Feb 1; 57(2):224-9. Available from: http://academic.oup.com/jac/article/57/2/224/804622/Alterations-of-the-penicillinbindingproteins-and. https://doi.org/10.1093/jac/dki442 PMID: 16368701 55. Kosowska K, Jacobs MR, Bajaksouzian S, Koeth L, Appelbaum PC. Alterations of Penicillin-Binding Proteins 1A, 2X, and 2B in Streptococcus pneumoniae Isolates for Which Amoxicillin MICs Are Higher than Penicillin MICs. 2004 Oct 1; 48(10). Available from: http://www.ncbi.nlm.nih.gov/pubmed/ 15388470. https://doi.org/10.1128/AAC.48.10.4020-4022.2004 PMID: 15388470 56. du Plessis M, Bingen E, Klugman KP, Plessis M du, Bingen E, Klugman KP, et al. Analysis of Penicillin-Binding Protein Genes of Clinical Isolates of Streptococcus pneumoniae with Reduced Susceptibility to Amoxicillin. Antimicrobial Agents and Chemotherapy. 2002 Aug 1; 46(8):2349-57. Available from: http://www.ncbi.nlm.nih.gov/pubmed/12121904. https://doi.org/10.1128/AAC.46.8.2349-2357. 2002 PMID: 12121904 57. Chesnel L, Carapito R, Croizé J, Dideberg O, Vernet T, Zapun A. Identical penicillin-binding domains in penicillin-binding proteins of Streptococcus pneumoniae clinical isolates with different levels of betalactam resistance. Antimicrob Agents Chemother. 2005 Jul 1; 49(7):2895-902. Available from: http:// www.ncbi.nlm.nih.gov/pubmed/15980366. https://doi.org/10.1128/AAC.49.7.2895-2902.2005 PMID: 15980366 58. Hakenbeck R, Nig AK, Kern I, Van Der Linden M, Keck W, Billot-Klein D, et al. Acquisition of Five High-M Penicillin-Binding Protein Variants during Transfer of High-Level β-Lactam Resistance from Streptococcus mitis to Streptococcus pneumoniae. Journal of Bacteriology. 1998; 180(7):1831-40. https://doi.org/10.1128/JB.180.7.1831-1840.1998 PMID: 9537382 59. Albarracín Orio AG, Piñas GE, Cortes PR, Cian MB, Echenique J. Compensatory Evolution of pbp Mutations Restores the Fitness Cost Imposed by β-Lactam Resistance in Streptococcus pneumoniae. [/table]
|
Effects of ankle eversion taping using kinesiology tape in a patient with ankle inversion sprain
Purpose] The aim of this study was to report the effects of ankle eversion taping using kinesiology tape on ankle inversion sprain.[Subject] The subject was a 21-year-old woman with Grade 2 ankle inversion sprain.[Methods] Ankle eversion taping was applied to the sprained left ankle using kinesiology tape for 4 weeks (average, 15 h/day).[Results] Ankle instability and pain were reduced, and functional dynamic balance was improved after ankle eversion taping for 4 weeks. The Cumberland Ankle Instability Tool score and reach distances in the Y-Balance and lunge tests were increased.[Conclusion] Repeated ankle eversion taping may be an effective treatment intervention for ankle inversion sprain.
# Introduction
Ankle inversion sprain often occurs during sports-related activity, landing on an inverted and plantar flexed foot after jumping, and running on uneven surfaces [bib_ref] Evaluation and treatment of ankle sprains: clinical recommendations for a positive outcome, Hockenbury [/bib_ref]. Approximately 85% of ankle sprain injuries are related to the lateral ligament [bib_ref] Predictive factors for lateral ankle sprains: a literature review, Beynnon [/bib_ref]. It involves pain, lateral ligamentous injury, excessive ankle inversion, swelling, and limitations in ankle range of motion [bib_ref] Does talocrural joint-thrust manipulation improve outcomes after inversion ankle sprain?, Krueger [/bib_ref]. Some patients with ankle inversion sprain experience continuous pain and ankle instability at long-term follow-up [bib_ref] Persistent disability associated with ankle sprains: a prospective examination of an athletic..., Gerber [/bib_ref]. Additionally, the reinjury rate of ankle inversion sprain may be as high as 80% [bib_ref] Treatment of ankle sprains in young athletes, Smith [/bib_ref]. Therefore, utilizing the most effective intervention for ankle inversion sprain is important.
Here, we report the effects of repeated application of ankle eversion taping (AET) using kinesiology tape in a patient with ankle inversion sprain.
# Subjects and methods
A 21 year-old woman complained of lateral left ankle pain after experiencing a Grade 2 ankle inversion sprain during walking one week prior to treatment. She had been diagnosed with left ankle inversion sprain 2 years prior, for which she had received manual physical therapy for 3 weeks. However, she continued to experience occasional ankle sprain. She complained of painful ankle inversion and instability when descending stairs, ascending an oblique surface, and standing on unstable ground and severe pain during running, jumping, and squatting. Full weight bearing on the sprained leg was especially difficult. Prior to study participation, she provided a written informed consent document that was approved by the Ethics Committee of Dong-Eui University. This study was conducted in accordance with the ethical standards of the Declaration of Helsinki.
AET using kinesiology tape (BB Tape, WETAPE Inc, Seoul, Republic of Korea) was applied daily to the left ankle for 4 weeks (average, 15 h/day). First, an I-shaped tape was applied from the talus to the calcaneus in a mild dorsiflexion position for posterior talar glide [fig_ref] Figure 1: Ankle eversion taping using kinesiology tape [/fig_ref]. Second, an I-shaped tape was applied from 5 cm above the lateral malleolus, over the medial calcaneus, to the inside of the instep of the foot in an eversion position to prevent painful inversion [fig_ref] Figure 1: Ankle eversion taping using kinesiology tape [/fig_ref]. Third, an I-shaped tape was applied using the same method as that used for the second tape to reinforce prevention of painful inversion and allow ankle eversion [fig_ref] Figure 1: Ankle eversion taping using kinesiology tape [/fig_ref]. Fourth, an I-shaped tape was applied using the same method as that used for the first tape to reinforce posterior talar glide and provide ankle support [fig_ref] Figure 1: Ankle eversion taping using kinesiology tape [/fig_ref]. We applied AET to the sprained left ankle daily after removal of the AET applied the previous day, even though the patient did not complain of itchiness. Additionally, the start and end points of the tape (approximately 2-3 cm) were applied without stretching to prevent skin problems [bib_ref] Effects of scapula-upward taping using kinesiology tape in a patient with shoulder..., Kim [/bib_ref]. No other treatment intervention for ankle inversion sprain was used.
# Results
In the initial assessment, the Numeric Pain Rating Scale (NPRS) score (0, no pain, and 10, the worst pain) of the lateral ankle at rest was 4 and that on application of 3 kg pressure on the most tender area of the lateral ankle using an algometer (Pain Test Model FPK; Wagner Instruments, Greenwich, CT, USA) was 7. Her left ankle score on the Cumberland Ankle Instability Tool (CAIT) questionnaire (9 items scored on a 30-point scale for measurement of functional ankle instability 7) ) was 2/30. The Y-Balance test was used to assess functional dynamic balance with ankle instability [bib_ref] Correlation of the Y-balance test with lower-limb strength of adult women, Lee [/bib_ref]. The maximal reach distances of the opposite leg in 3 directions (anterior, posteromedial, and posterolateral) while in a sprained left leg stance were 35 cm, 55 cm, and 54 cm, respectively. The lunge test (distance from the wall to the big toe in ankle dorsiflexion without lifting the heel and with knee flexion) was used to assess ankle flexibility [bib_ref] Intra-rater and inter-rater reliability of a weight-bearing lunge measure of ankle dorsiflexion, Bennell [/bib_ref]. The initial distance of the left toe was 5 cm.
Following AET application for 4 weeks, the NPRS score for the lateral ankle at rest decreased from 4 to 0 and that after application of 3 kg pressure to the most tender lateral ankle area using an algometer decreased from 6 to 1; the CAIT score increased from 2/30 to 29/30. The reach distances in the anterior, posteromedial, and posterolateral directions increased from 35 cm to 46 cm, from 55 cm to 69 cm, and from 54 cm to 63 cm, respectively. The lunge test distance increased from 5 cm to 9.5 cm. She no longer experienced ankle pain and instability when descending stairs, ascending an oblique surface, standing on unstable ground, or when jumping, running, or squatting. She was able to fully bear weight on the left leg without pain.
# Discussion
This case study showed that repeated AET application for 4 weeks reduced lateral ankle pain and improved ankle flexibility and functional dynamic balance. Painful plantar flexion and inversion of the sprained ankle were prevented through a more everted ankle with AET application. In addition, the sprained ankle was protected from reinjury through the mechanical effects of AET application. Previous studies have reported that kinesiology tape application to an ankle with instability 10) and eversion sprain 11) reduced pain and improved stability. Therefore, healing of ankle inversion sprain was possible.
Kinesiology tape may provide support to the joint structure [bib_ref] Efficacy of kinesiology taping for recovery from occupational wrist disorders experienced by..., Kim [/bib_ref] and improve joint position sense [bib_ref] Effect of kinesio taping on proprioception in the ankle, Murray [/bib_ref]. Activation of proprioceptors through kinesiology tape application may increase the ankle joint position sense. Therefore, maintaining the neutral ankle position using repeated AET application may increase ankle stability, the reach distances in the Y-Balance and lunge tests, and the CAIT score. Future studies on the clinical effects of AET in a larger number of patients with ankle inversion sprain are required.
[fig] Figure 1: Ankle eversion taping using kinesiology tape [/fig]
|
Lactation mastitis: Promising alternative indicators for early diagnosis
Although lactation mastitis (LM) has been extensively researched, the incidence rate of LM remains a salient clinical problem. To reduce this incidence rate and achieve a better prognosis, early and specific quantitative indicators are particularly important. It has been found that milk electrolyte concentrations (chloride, potassium, and sodium) and electrical conductivity (EC) significantly change in the early stages of LM in an animal model. Several studies have evaluated EC for the detection of subclinical mastitis in cows. EC, chloride, and sodium content of milk were more accurate for predicting infection status than were other variables. In the early stages of LM, lactic sodium, chloride, and EC increase, but potassium decreases. However, these indicators have not been reported in the diagnosis of LM in humans. This review summarizes the pathogenesis and the mechanism of LM in terms of milk electrolyte concentration and EC, and aim to provide new ideas for the detection of sub-clinical mastitis in humans.
# Introduction
Lactation mastitis (LM), which is universally described as a suite of breast conditions that present with local, and often systemic, inflammatory symptoms and signs during lactation . It is an enervating and common disease that affects up to 33% of lactating women . LM has a negative impact on both the mother and the baby. Women with LM may develop pain, localized skin redness, and can have associated systemic symptoms, including fever. However, breast pain is the most common and the most distressing symptom for mothers . The symptoms of LM can lead to a compromised maternal psychological state. In addition, approximately 3% of women with LM will develop a breast abscess , and an incidence rate of up to 11% has been reported , which may cause permanent damage to the shape of the breast. In order to minimize the detrimental effects of LM, researchers have explored various strategies, including the associated risk factors and etiology . However, none of these are known to improve clinical prognosis.
To date, the clinical diagnosis of LM has mostly relied on empirical diagnoses, such as a tender, hot, swollen, wedge-shaped area of redness on the affected breast which is associated with an elevated temperature and systemic symptoms, such as a temperature of 38.5 °C or higher, chills, flu-like aching, and systemic illness . Due to the lack of early standardized diagnostic criteria, this causes repeated outbreaks and even progresses to a breast abscess, causing great physical and mental pain in mothers with LM. In some Western countries, for example, 73% of children born in Sweden in 1996 were breastfed for 6 mo[12], but LM frequently results in the cessation of exclusive breastfeeding in the absence of appropriate treatment. As a result, it frequently reduces the protective effect of breastfeeding in mothers. For example, related data show that breastfeeding is associated with a 24% lower risk of invasive ovarian cancer , and aggregate results indicate that breastfeeding is inversely associated with the risk of breast cancer and reduces the incidence of osteoporosis and type 2 diabetes . Furthermore, LM is linked to lower levels of fat, carbohydrate, and energy in breast milk , which may lead to nutritional deficiencies, lowered immunity, and mental effects in infants.
Several studies have evaluated electrical conductivity (EC) for the detection of subclinical mastitis in cows, some have even identified mastitis causing pathogens using EC . EC, chloride (Cl -), and sodium (Na + ) content in milk were more accurate in predicting infection status than other variables. The electrolyte concentration and EC of milk are the physicochemical properties of milk, and EC measurements were used as an experimental screening indicator for LM in animals as early as the 1990s . Paudyal et al showed that absolute changes in cow's milk electrolyte concentrations and EC can be used to screen breast milk samples for LM. A previous study provided evidence that when an inflammatory response occurs in the breast tissue of animals, breast permeability increases, and as a result, the potassium (K + ) concentration reduces, but the Na+, Clconcentrations, and EC increase . Singh et al recently conducted an observational animal study and concluded that the magnitude of changes in the milk electrolyte concentration and EC may have diagnostic and prognostic values. Due to a lack of research, more data are needed to identify the variation in human milk electrolyte concentration and EC during LM.
However, it is uncertain whether human milk electrolyte concentrations and EC have a positive effect on the early diagnosis of LM. To enable more patients with LM to be effectively diagnosed and treated at an early stage of the disease, the pathogenesis of LM and the relationship between LM and electrolyte concentrations and EC [fig_ref] Figure 1: The mechanism of lactation mastitis about milk electrolyte concentration and electrical conductivity... [/fig_ref] are discussed in this review. We investigate the changes in electrolyte concentrations and EC from two perspectives: Altered cell membrane permeability and osmotic pressure level. This will provide new ideas for the detection of subclinical mastitis in humans. November 6, 2022
Volume 10 Issue 31
## Literature search
A descriptive review was conducted on the mechanism of changes in milk electrolyte concentration and EC during LM. Pub-Med was searched for articles published between July 1966 and February 2022. The following Medical Subject Headings or free-text terms were used in the search: LM, milk electrolyte concentration, milk EC, early diagnosis, pathogenesis, mechanism, inflammatory injury, and altered cell membrane permeability. The search was limited to papers written in English, with no restrictions on the type of article.
## Pathogenesis of lm
The susceptibility and severity of LM are positively correlated with inflammatory factors. Some researchers consider LM to be an infectious disorder[19,24,25], but the etiology of LM has now shifted from infection to inflammation.
## Infectious mechanism
Due to the characteristics of LM, it was previously considered to be an infectious disease.
## Inflammatory mechanism
Recently, in-depth studies on LM have revealed that the presence of infectious agents is not positively associated with the occurrence of LM . In addition, there is no positive correlation between the severity of LM and the bacterial count in milk . Several studies showed that the susceptibility and severity of LM are positively correlated with inflammatory factors, including C-reactive protein, interleukin (IL)-1, IL-6, IL-8, and tumor necrosis factor-α (TNF-α) [bib_ref] Inflammatory mediators in mastitis and lactation insufficiency, Ingman [/bib_ref]. Elevated serum cytokines IL-1, IL-6, IL-8, and TNF-α reveal the activation of transcription factor nuclear factor-kappa B (NF-κB) in the host [bib_ref] LPS induces pro-inflammatory response in mastitis mice and mammary epithelial cells: Possible..., Xiao [/bib_ref]. Furthermore, activation of the NF-κB pathway in LM hosts has been demonstrated in numerous animal studies, in vitro experiments, and genetic studies [bib_ref] CD36 regulates lipopolysaccharide-induced signaling pathways and mediates the internalization of Escherichia coli..., Cao [/bib_ref] [bib_ref] Lingual antimicrobial peptide and IL-8 expression are oppositely regulated by the antagonistic..., Liu [/bib_ref] [bib_ref] Altered molecular expression of the TLR4/NF-κB signaling pathway in mammary tissue of..., Wu [/bib_ref] [bib_ref] Pathogen recognition receptors: ligands and signaling pathways by Toll-like receptors, Sasai [/bib_ref]. November 6, 2022 Volume 10 Issue 31
As upstream candidate receptors of NF-κB, Toll-like receptors (TLRs) are important inflammatory mediators and their isoforms include TLR1-TLR11 [bib_ref] From inflammasomes to fevers, crystals and hypertension: how basic research explains inflammatory..., Mcdermott [/bib_ref] , which are recognized as pattern recognition receptors (PRRs) by microbe-associated molecular patterns (MAMPs) or danger-associated molecular patterns (DAMPs) and activate downstream NF-κB signaling pathways[34].
## Mamp-mediated tlr signaling pathway
Of all the TLRs, TLR2, TLR4, and TLR5 bind to bacterial byproducts in MAMPs, such as lipopolysaccharide, phospholipid wall acids, bacterial lipopeptides, and flagellin, activate the corresponding TLRs, which activate transcription, translation, and the release of a series of inflammatory factors, chemokines, and adhesion molecules, and recruit other molecules involved in the innate immune response (neutrophils, etc.) to the site of infection . Pathogenic byproducts bind to TLRs on the surface of PRRs, and the signal is transmitted from extracellular to intracellular, activating TLRs, followed by signal transduction, and activation of the transcription factor NF-κB. This is the pathogenesis of infectious LM via the inflammatory mechanism.
## Damp-mediated tlr signaling pathway
TLR activation via the DAMP pathway explains how LM can lead to disease in the absence of infectious agents. DAMPs are endogenous proteins and can activate TLRs in a sterile environment [bib_ref] Endogenous signals released from necrotic cells augment inflammatory responses to bacterial endotoxin, El Mezayen [/bib_ref]. DAMPs can activate TLRs in two ways: first, some inflammatory mediators activated by DAMPs activate TLRs and downstream NF-κB [bib_ref] Lactate boosts TLR4 signaling and NF-kappaB pathway-mediated gene transcription in macrophages via..., Samuvel [/bib_ref] ; second, DAMPs can enhance the susceptibility of the TLR immune response in a sterile environment, thus activating TLRs and downstream NF-κB leading to the development of noninfectious LM [bib_ref] Effect of milking throughout pregnancy on milk yield in the succeeding lactation, Smith [/bib_ref].
## Mechanism of lm-induced changes in milk electrolyte concentration and ec
## Mechanism of cell membrane permeability alteration
Changes in milk electrolyte concentration and EC are linked to the inflammatory response during LM, and pathological changes in breast tissues caused by inflammatory factors result in changes in milk composition. Our understanding of the changes in milk electrolyte concentration and EC during LM is linked to the research by Smith et al [bib_ref] Breast Abscess, and Granulomatous Mastitis, Omranipour [/bib_ref] in the 1960s.The essence of LM is the inflammatory response of the breast tissue [bib_ref] Recent advances in the study of monovalent ions movements across the mammary..., Peaker [/bib_ref] , and inflammatory factors cause increased epithelial permeability and both vascular and parenchyma damage .
The sodium pump on the basolateral membrane, which keeps intracellular K + high and Na + low, and the distribution of these ions according to the gradient of electric potential across the luminal membranes are the most important characteristics in terms of Na + and K + . The Na + /K + ratio of the intracellular fluid is maintained at approximately 1:3 and the ratio of the extracellular fluid is maintained at approximately 3:1 by active transport of the sodium-potassium ion pump. As milk is electrically positive compared to the interior of the cell, the concentrations of these ions are lower in milk. However, because milk is nearly isosmotic to plasma[50], the ratio between them is similar, with a Na + /K + ratio of about 1:3[51, [bib_ref] Mastitis in rural Gambian mothers and the protection of the breast by..., Prentice [/bib_ref].
When an inflammatory response occurs in the breast tissue, all of the inflammatory factors produced damage the ducts and secretary epithelial cells, disrupt the tight junctions between secretary cells, and increase capillary permeability. As a result, higher levels of Na + and Clin the extracellular fluid enter the mammary gland alveolar lumen through the tight junctions that are opened, while the K + concentration decreases in order to maintain the osmotic pressure of milk in the alveolar lumen . The increase in ionic concentrations in breast milk in the presence of mastitis, such as Na + and Cl -, is consistent with other animal studies[54-56].
## Mechanism of the milk osmolality level
Milk is rich in lactose. Lactose excretion may provide a reliable basis for fateful changes in breast permeability as lactose in food contributes minimally to the circulation[54]. Fetherston et al conducted relevant studies, and the 24-h urinary excretion of lactose during LM has been extensively discussed. The correspondingly low steady-state urinary excretion of lactose demonstrates that these variations are not the result of increased paracellular pathway permeability. It indicates that the higher than normal concentrations of milk Na + and Clobserved in "normal breasts" could be a normal physiologic response to a lower concentration of lactose, ensuring that the osmolality of milk remains isotonic with plasma[20]. This has significant implications for the supposition that a raised Na + concentration is an outcome of subclinical LM. November 6, 2022
Volume 10 Issue 31
## Additional diagnostic prospects: milk electrolyte concentration and ec
The electrolyte concentration and EC are physical properties of body fluids that are used to assess human health status and disease severity [bib_ref] Body Fluid Compartments, Cell Membrane Ion Transport, Electrolyte Concentrations, and Acid-Base Balance, Seifter [/bib_ref]. Previous animal studies have indicated that they can be used as an alternative method for the early diagnosis of LM[60]. In addition, demonstrated that the measurements of electrolyte concentration and EC in milk samples for the diagnosis of LM in animals were comparable to other diagnostic methods. Similar to the pathological response in animals with LM[62,63], when a woman is diagnosed with LM, inflammatory factors, such as procalcitonin, IL-1, IL-6, IL-8, and TNF-α [bib_ref] LPS induces pro-inflammatory response in mastitis mice and mammary epithelial cells: Possible..., Xiao [/bib_ref] , cause disruption of the tight junctions between cells. Furthermore, as a result of both the decrease in available glucose, particularly during severe symptoms, and the damage or death of lactocytes due to inflammation, lactose synthesis decreases . Both of these factors can lead to changes in the electrolyte concentration and EC of milk. As mentioned earlier, numerous animal studies have indicated that milk electrolyte concentration and EC can be used in the early diagnosis of LM; therefore, based on the similar pathological responses, it is feasible to use milk electrolyte concentration and EC to diagnose early LM in humans.
In a recent study, we collected milk specimens from approximately 30 healthy women and 15 patients diagnosed with LM, and measured the electrolyte concentration (including Na + , K + , and Cl -) and EC of bilateral breast milk in all patients. The final test results revealed that the concentrations of Na + , Cland EC in breast milk were significantly higher in women with LM than in healthy women, and in some cases several times higher. In addition, in women with LM, the concentrations of Na + , Cland EC in breast milk were markedly higher on the affected side than on the healthy side. Furthermore, we also found that the concentrations of Na + , Cland EC in milk were increased to varying degrees in patients who had symptoms but were not diagnosed with LM.
# Conclusion
The early diagnosis and prevention of LM still face many challenges. We summarized the pathogenesis of LM, the mechanism of LM-induced changes in milk electrolyte concentration and EC, and found that changes in milk electrolyte concentration and EC in humans were primarily correlated with LM. It is clear from animal studies that there is a correlation between LM and milk electrolyte concentration and EC, and from the significant changes in these indicators, LM can be diagnosed at an early stage and thus achieve a better prognosis. However, there have been few studies carried out on this topic. As a result, more data are required to verify these findings. If the changes in milk electrolyte concentration and EC are beneficial, these will have an enormous impact on clinical practice.
## Footnotes
Author contributions: Huang Q, Zheng XM and Zhang ML contributed equally to this work; Ning P, Wu MJ designed the research study; Huang Q, Zheng XM collected data from the literature; Huang Q, Zheng XM and Zhang ML wrote the manuscript; Ning P, Wu MJ conducted a review of the manuscript; all authors have read and approve the final manuscript.
# Conflict-of-interest statement:
The authors have no potential conflicts of interest to disclose.
## Open-access:
[fig] Figure 1: The mechanism of lactation mastitis about milk electrolyte concentration and electrical conductivity in animals. DAMPs: Dangerassociated molecular patterns; LPS: Lipopolysaccharide; MAMPs: Microbe-associated molecular patterns; TLR: Toll-like receptors; EC: Electrical conductivity. [/fig]
|
Cyrcadian Rhythm, Mood, and Temporal Patterns of Eating Chocolate: A Scoping Review of Physiology, Findings, and Future Directions
## Introduction: life, health, and circadian rhythms
The Earth's rotation around its axis causes periodic light and dark variations in the environment over a 24-h period. These predictable variations in the light environment allow organisms to optimally organize their physiology and activity-rest rhythms to specific periods of the day-night cycle [bib_ref] Light as a central modulator of circadian rhythms, sleep and affect, Legates [/bib_ref]. Indeed, there is a clear selective advantage in anticipating changes in the environment, particularly light-dark, by adapting their activities. Therefore, evolved organisms have an internal biological clock that functions with a period close to 24 h, which is circadian, in the absence of environmental influences such as light [bib_ref] Coordination of circadian timing in mammals, Reppert [/bib_ref]. Anticipation of daily events organized in this way allows optimal management of time and available energy, giving those organisms a significant advantage. The correct alignment between the light-dark rhythm, the circadian clock, and behavior determines a temporal order in organisms that is of fundamental importance for ensuring survival [bib_ref] A clockwork web: Circadian timing in brain and periphery, in health and..., Hastings [/bib_ref].
The circadian clock drives many outputs, including sleep-wakefulness, hormonal, and metabolic rhythm. Locomotor activity and sleep are segregated into specific phases of the light-dark cycle [bib_ref] Mood, the Circadian System, and Melanopsin Retinal Ganglion Cells, Ospri [/bib_ref]. Endocrine processes are also regulated by a rhythmic periodicity: the cortisol concentration in circulation reaches a peak level in the morning, and unlike many hormones involved in metabolism, such as leptin and ghrelin, reach their peak at night [bib_ref] Circadian rhythms, sleep, and metabolism, Huang [/bib_ref]. Urine production has a circadian nature [bib_ref] Circadian Rhythms in Urinary Functions: Possible Roles of Circadian Clocks?, Noh [/bib_ref] , as well as perception-nociception [bib_ref] Circadian rhythm of nociception in the golden hamster, Pickard [/bib_ref] , The reciprocal influences of circadian clocks and energy metabolism suggest that feeding time has a critical impact on metabolism. The quality and quantity as well as the timing of food intake are all important for nutrition: this is referred to as 'chrononutrition'. Indeed, food intake has a knock-on effect on peripheral clocks, mainly the liver, but not on the SCN, so that food intake outside the physiologically designated 'time' window induces desynchronization between peripheral clocks and SCN [bib_ref] Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker..., Damiola [/bib_ref].
Tissue-specific gene expression of peripheral clocks is regulated by food intake, whereas the central gene expression of the SCN itself is protected by nutrient deregulation [bib_ref] Lifestyle and Circadian Health: Where the Challenges Lie?, Gupta [/bib_ref]. Timing of food intake also influences body weight and obesity risk in humans [bib_ref] Timing of food intake predicts weight loss effectiveness, Garaulet [/bib_ref] [bib_ref] Timing of energy intake during the day is associated with the risk..., Wang [/bib_ref] : a two-group study evaluating weight loss during an isocaloric diet showed improvements in several metabolic indicators such as body weight, fasting blood glucose, insulin, ghrelin, average hunger, and satiety scores in the group eating a larger breakfast and lighter dinner compared to the other way around [bib_ref] High Caloric intake at breakfast vs. dinner differentially influences weight loss of..., Jakubowicz [/bib_ref]. In both humans and experimental animals, eating dinner late at night and skipping breakfast leads to increased body weight and obesity.
With regard to individual food components associated differently with different eating times, it has been shown that eating a diet rich in fat ad libitum attenuates the amplitude of the clocks, while eating a diet limited in time restores their amplitude by regulating the circadian cycle. A combination of carbohydrates and protein also appears to be essential for proper resetting of peripheral clocks such as the liver clock [bib_ref] A Balanced Diet Is Necessary for Proper Entrainment Signals of the Mouse..., Hirao [/bib_ref].
In contrast, the intake of protein nutrients at times of phase-shifting hepatic cell clocks correlates with blood glucose absorption, and readily digestible starches with a high glycemic index have a powerful entrainment effect on peripheral liver clocks [bib_ref] Time-restricted feeding of rapidly digested starches causes stronger entrainment of the liver..., Itokawa [/bib_ref].
The action of specific nutrients occurs through the different expression of clockcontrolled genes [bib_ref] Nutrigenetics and Nutrimiromics of the Circadian System: The Time for Human Health, Micó [/bib_ref]. Glucose is able to regulate BMAL1 and period expression [bib_ref] Nutrients and Circadian Rhythms in Mammals, Wu [/bib_ref] as demonstrated, for example, in a study of mice fed a high-fat diet in which an alteration of lipid metabolism genes was documented [bib_ref] Impact of nutrients on circadian rhythmicity, Oosterman [/bib_ref]. Indeed, this finding is further corroborated by a study conducted by Eckel-Mahn et al., which illustrated that most hepatic metabolites also follow changes according to a circadian rhythm, and that these variations are regulated by both the clock transcriptome and the feeding-fasting cycles that favor the maintenance of hepatic homeostasis [bib_ref] Coordination of the transcriptome and metabolome by the circadian clock, Eckel-Mahan [/bib_ref].
A high-fat diet, such as the ketogenic diet with a high fat content and low carbohydrate content, has the greatest effect on the central clock and/or eating behavior: a high fat content reduces the duration of the circadian rhythm of locomotor activity under conditions of constant darkness, whereas under normal conditions of alternating light-dark, there is an advancement in the rhythm of clock expression and clock-controlled genes in peripheral tissues [bib_ref] Ketogenic Diet Disrupts the Circadian Clock and Increases Hypofibrinolytic Risk by Inducing..., Oishi [/bib_ref]. In recent years, the field of nutraceuticals and functional foods has developed alongside interest in the potential modulatory effects of food constituents on human health. In this context, mention should be made of caffeine/theophylline, which could influence our circadian system, and flavonoids. Caffeine in food and drink has been shown to prolong circadian locomotor rhythms in Drosophila and mice [bib_ref] The Effects of Caffeine on Sleep in Drosophila Require PKA Activity, But..., Wu [/bib_ref] [bib_ref] Caffeine lengthens circadian rhythms in mice, Oike [/bib_ref]. The dose of caffeine needed to influence circadian rhythms appears to be low, at 0.05%, equivalent to the dose contained in coffee. Indeed, consumption of non-decaffeinated coffee prolongs the rhythm of circadian activity in mice under constant dark conditions [bib_ref] Caffeine lengthens circadian rhythms in mice, Oike [/bib_ref]. Interestingly, both caffeine and theophylline also prolong the circadian period in cultured cells, mouse tissues, Neurospora, Chlamydomonas, Drosophila, and even higher plants [bib_ref] Nutrients and Circadian Rhythms in Mammals, Wu [/bib_ref] [bib_ref] Circadian Periodicity in Neurospora: Alteration by Inhibitors of Cyclic AMP Phosphodiesterase, Feldman [/bib_ref] [bib_ref] The effects of caffeine and theophylline on the phototactic rhythm of Chlamydomonas..., Goodenough [/bib_ref] [bib_ref] Effects of cAMP, theophylline, imidazole, and 4-(3,4-dimethoxybenzyl)-2-imidazolidone on the leaf movement rhythm..., Bollig [/bib_ref].
Among the dietary constituents, flavonoids, a class of polyphenolic compounds, have received particular interest for their various beneficial biological actions, such as neurological and cardiological protection and neuromodulation. In particular, a rich source of flavonoids has been detected in the cocoa bean, especially the subclass of flavanols in the form of epicatechin and catechin [bib_ref] Flavonoids, cognition, and dementia: Actions, mechanisms, and potential therapeutic utility for Alzheimer..., Williams [/bib_ref].
## Chocolate
Cocoa products and especially chocolate are foodstuffs originating from South America. Cocoa is obtained from the seeds of the Theobroma cacao tree, which are then dried, shelled, fermented, and ground with other substances such as sugar, fat, and other flavorings to produce the wide variety of chocolate available on the market, from dark to milk variations.
Chocolate consists of a combination of several ingredients, the main ones being cocoa, cocoa butter, and sugar, which make up the solid food product. Cocoa beans, along with foods such as tea, red wine, and fruit, are a rich source of flavanols, a subgroup of natural flavonoids that are bioactive plant compounds.
There have been reports of the health benefits of chocolate since ancient times, the earliest dating back to Aztec and Mayan medical practice [bib_ref] Cacao usage by the earliest Maya civilization, Hurst [/bib_ref]. It was not until the end of the 20th century, however, that claims about the alleged health benefits of chocolate attracted more and more scientific interest. Flavonoids, contained in cocoa, have been approved by the European Food Safety Agency as being beneficial to health.
Due to dark chocolate's high concentrations of flavanols, a subgroup of flavonoids, especially epicatechin, [bib_ref] Natural phenolic compounds 1900-2000: A bird's eye view of a century's chemistry, Whiting [/bib_ref] known as powerful antioxidant agents [bib_ref] Chocolate and the brain: Neurobiological impact of cocoa flavanols on cognition and..., Sokolov [/bib_ref] , have received a health claim in relation to their impact on 'maintaining normal endothelium-dependent vasodilation'.
Many studies conducted so far focus on the effects of chocolate on the cardiovascular system, skin, cholesterol levels, the release of the neurotransmitters anandamide and serotonin, and on the properties of specific stimulating constituents such as theobromine and caffeine [bib_ref] Cocoa and Chocolate in Human Health and Disease, Katz [/bib_ref].
## Biochemical components and neurobiological impact
Chocolate contains more than 300-500 known chemicals, some of which also act on brain cells and modulate mood. Of these 300 to 500 chemicals in chocolate, some play an important role in humans, influencing neurocognitive functions.
The main psychoactive components of chocolate are [bib_ref] Effects of chocolate on cognitive function and mood: A systematic review, Scholey [/bib_ref] [bib_ref] The neuroprotective effects of cocoa flavanol and its influence on cognitive performance, Nehlig [/bib_ref] as follows.
-Carbohydrates, which have known behavioral effects. -Flavanols, which are ubiquitous in the plant kingdom. In foods normally consumed in the diet, high levels of flavonoids can be found in green and black tea, grapes, red wine, apples, and especially in cocoa and cocoa-containing products. In fact, cocoa is particularly rich in flavonoids and contains a distinct complement of flavanols (a subclass of flavonoids), flavan-3-ols, mainly present in the form of epicatechin and catechin [bib_ref] Natural phenolic compounds 1900-2000: A bird's eye view of a century's chemistry, Whiting [/bib_ref] , and their derivatives in high concentrations [bib_ref] Concentrations of Proanthocyanidins in Common Foods and Estimations of Normal Consumption, Gu [/bib_ref]. Flavan-3-ols are the building blocks for polymeric procyanidin type B-2. -Methylxanthines (MX), such as caffeine and its highly fat-soluble derivative and metabolite theobromine, which have peak plasma levels 60-120 min after ingestion. Like caffeine, theobromine binds to adenosine receptors, exhibiting its psychoactive potential similar to that of caffeine. However, these two MX have distinct functional binding properties. -Biogenic amines, such as serotonin, tryptophan, phenylethylamine, tyrosine, tryptamine, and tyramine, have a concentration that increases during fermentation and decreases during roasting and alkalinization. -Anandamide, an endogenous ligand for the cannabinoid receptor that is found in low quantities, such as 0.5 mg g −1 , salsolinol, and tetrahydro-b-carboline.
The most studied substances are flavanols and their metabolites: it has been shown in animal studies that these products can pass through the blood-brain barrier (BBB), with positive effects on brain tissue, vessels, and function (angio-and neurogenesis, changes in neuron morphology), stimulating cerebral blood circulation [bib_ref] The neuroprotective potential of flavonoids: A multiplicity of effects, Vauzour [/bib_ref]. Epicatechin [bib_ref] Natural phenolic compounds 1900-2000: A bird's eye view of a century's chemistry, Whiting [/bib_ref] , the most common flavanol found in cocoa, is rapidly absorbed by humans. Thirty min after intake it is detectable in blood plasma, and its peak is reached after 2-3 h, showing a concentration directly related to the dose of chocolate ingested [bib_ref] Plasma kinetics in man of epicatechin from black chocolate, Richelle [/bib_ref] , and then it returns to basal level within 6-8 h after consumption.
## Chocolate and brain functions
In order to exert effects in the brain, flavanols cross the BBB by a process that is not only time-dependent but also stereoselective, which, thus, favors the passage of epicatechin more than catechin, as has been demonstrated in two cell lines, one of rat and one of human origin [bib_ref] Insights into the putative catechin and epicatechin transport across blood-brain barrier, Faria [/bib_ref] ; the permeability is proportional to the degree of lipophilicity and inversely proportional to the degree of polarity.
In humans, the main polyphenols increase cerebral blood flow (CBF) and are derived from nutrients such as cocoa, wine, grape seeds, berries, tea, tomatoes, and soy [bib_ref] Vascular action of polyphenols, Ghosh [/bib_ref].
Optimal brain function requires an adequate CBF that promotes the correct and constant supply of oxygen and glucose to neurons and the excretion of deposited waste products. Increasing CBF is also a potential tool for enhancing brain function. Flavanols and their metabolites have the ability to reach and accumulate in the brain regions mainly involved in learning processes and memory, and are therefore thought to exert a direct brain action on cognitive function and neuroprotection [bib_ref] The neuroprotective effects of cocoa flavanol and its influence on cognitive performance, Nehlig [/bib_ref]. After chronic administration of chocolate, high concentrations of tangeretin were found in the striatum nucleus, hypothalamus, and hippocampus of the rat [bib_ref] Tissue distribution and neuroprotective effects of citrus flavonoid tangeretin in a rat..., Datla [/bib_ref].
The neurobiological impact of flavanols on the brain in the areas of learning, memory, cognitive function, and mood is thought to occur mainly in two ways. The effects of flavonoids on the brain are mediated by their function of neuroprotecting vulnerable neurons in particular, improving neuronal function and stimulating regeneration (neurogenesis) [bib_ref] Food for thought: The role of dietary flavonoids in enhancing human memory,..., Spencer [/bib_ref] by interacting with intracellular neuronal signaling pathways that control neuronal survival and differentiation, long-term potentiation (LTP), and memory.
First, flavonoids interact with a number of cellular signaling pathways by activating gene expression and protein synthesis for the maintenance of LTP, and the stabilization of long-term memories [bib_ref] Translational Control by MAPK Signaling in Long-Term Synaptic Plasticity and Memory, Kelleher [/bib_ref] are critical for neurogenesis, synaptic growth, and survival of neurons, mainly in the brain hippocampus and subventricular area related to learning and memory [bib_ref] Effect of the flavonoid, oroxylin A, on transient cerebral hypoperfusion-induced memory impairment..., Kim [/bib_ref] [bib_ref] A Diet Enriched in Polyphenols and Polyunsaturated Fatty Acids, LMN Diet, Induces..., Valente [/bib_ref].
Secondly, flavonoids induce vasodilation through nitric oxide at both cardiovascular and peripheral levels through the production of nitric oxide (NO), a key regulator of vascular function, which acts as a signaling molecule by inhibiting the action of adhesion molecules in atheromatous plaque that cause inflammation [bib_ref] Anti-inflammatory properties of dietary flavonoids, González-Gallego [/bib_ref] , and, most importantly, promote and improve the function of the vascular endothelium by acting, with dilating action, on the smooth muscle tissue of blood vessels [bib_ref] Epicatechin mediates beneficial effects of flavanol-rich cocoa on vascular function in humans, Schroeter [/bib_ref]. This effect on the vascular system with endothelium-dependent vasodilation, contributes to the maintenance of normal blood flow and improvement of blood pressure; it also induces a reduction in platelet aggregation [bib_ref] Dark Chocolate Improves Coronary Vasomotion and Reduces Platelet Reactivity, Flammer [/bib_ref] [bib_ref] Cocoa Reduces Blood Pressure and Insulin Resistance and Improves Endothelium-Dependent Vasodilation in..., Grassi [/bib_ref] [bib_ref] Blood Pressure Is Reduced and Insulin Sensitivity Increased in Glucose-Intolerant, Hypertensive Subjects..., Grassi [/bib_ref] [bib_ref] Protective effects of dark chocolate on endothelial function and diabetes, Grassi [/bib_ref] [bib_ref] Cocoa consumption dose-dependently improves flow-mediated dilation and arterial stiffness decreasing blood pressure..., Grassi [/bib_ref] [bib_ref] Flavonoid-Rich Cocoa Consumption Affects Multiple Cardiovascular Risk Factors in a Meta-Analysis of..., Shrime [/bib_ref]. This. in turn. results in increased CBF and blood perfusion throughout the central and peripheral nervous system [bib_ref] Flavanols, the Kuna, cocoa consumption, and nitric oxide, Hollenberg [/bib_ref] , allowing better oxygen and glucose delivery to neurons and removal of waste metabolites in the nervous systems [bib_ref] Research on Polyphenolics in Vision and Eye Health, Kalt [/bib_ref]. Among flavanols, epicatechin has the greatest ability to increase nitric oxide (NO) bioavailability, leading to improvements in vascular tone and blood pressure regulation [bib_ref] Brain Protection and Cognitive Function: Cocoa Flavonoids as Nutraceuticals, Grassi [/bib_ref]. These vascular changes occurring at the peripheral level may also extend to cerebral perfusion, leading to optimized cerebrovascular integration during neuronal activation phases, a mechanism considered crucial for the functional and structural integrity of the brain and for promoting adult neurogenesis in the hippocampus.
Administration of cocoa flavonoids, therefore, also stimulates angiogenesis in the hippocampus [bib_ref] Chocolate and the brain: Neurobiological impact of cocoa flavanols on cognition and..., Sokolov [/bib_ref] [bib_ref] Plant-Derived Flavanol (-)Epicatechin Enhances Angiogenesis and Retention of Spatial Memory in Mice, Van Praag [/bib_ref] , as demonstrated by administering epicatechin, given to mice at a dose of 500 mg g −1 (daily supply of 2.5 mg). Combining epicatechin with exercise also improved the consolidation of spatial memory and the density of dendritic spines in the dentate gyrus of the hippocampus. In the same study, epicatechin treatment was shown to increase learning-associated gene expression in the hippocampus, while it did not appear to influence neurogenesis in the adult hippocampus [bib_ref] Plant-Derived Flavanol (-)Epicatechin Enhances Angiogenesis and Retention of Spatial Memory in Mice, Van Praag [/bib_ref]. Thus, flavonoids could also exert their neurocognitive effects both directly and indirectly by interacting with the cellular network and molecular system deputed to memory acquisition, storage, retrieval, and learning [bib_ref] Food for thought: The role of dietary flavonoids in enhancing human memory,..., Spencer [/bib_ref] , also through long-term potentiation, synaptic plasticity [bib_ref] The interactions of flavonoids within neuronal signalling pathways, Spencer [/bib_ref] , enhanced neuronal connection, and communication.
Epidemiological studies suggest that regular flavonoid intake may be associated with better cognitive function [bib_ref] Chocolate intake is associated with better cognitive function: The Maine-Syracuse Longitudinal Study, Crichton [/bib_ref] , to decreased risk of dementia and cognitive decline [bib_ref] Intake of flavonoids and risk of dementia, Commenges [/bib_ref] [bib_ref] Chocolate Consumption is Associated with a Lower Risk of Cognitive Decline, Moreira [/bib_ref] [bib_ref] Green tea consumption and cognitive function: A cross-sectional study from the Tsurugaya..., Kuriyama [/bib_ref] , better cognitive development over a 10-year period [bib_ref] Flavonoid Intake and Cognitive Decline over a 10-Year Period, Letenneur [/bib_ref] , and improved dose-dependent cognitive performance in physiological aging [bib_ref] Intake of Flavonoid-Rich Wine, Tea, and Chocolate by Elderly Men and Women..., Nurk [/bib_ref].
## Chocolate and mood
A lot of data from the literature support the hypothesis of the influence of theobromine and caffeine on mood and cognitive function [bib_ref] Effects of low doses of caffeine on cognitive performance, mood and thirst..., Smit [/bib_ref] [bib_ref] Methylxanthines are the psycho-pharmacologically active constituents of chocolate, Smit [/bib_ref] [bib_ref] Reinforcing effects of caffeine and theobromine as found in chocolate, Smit [/bib_ref] , but the impact and mechanism by which flavanols affect mood remains unclear.
It is commonly believed that eating chocolate improves mood and rapidly induces a sense of well-being in people. An initial rapid effect of chocolate on emotional comfort appears to be related to the ability of the carbohydrates it contains, to promote such positive feelings through the release of several gut and brain peptides [bib_ref] Atypical Depression: A Reappraisal, Parker [/bib_ref].
In rats, the intake of cocoa-extracted polyphenols, while significantly reducing the duration of immobility in a forced swimming test, had no effect on locomotor activity in the open field, confirming its specific antidepressant effect [bib_ref] Antidepressant-like effects of a cocoa polyphenolic extract in Wistar-Unilever rats, Messaoudi [/bib_ref]. The most likely basis for this effect may be attributable to endorphin release. Indeed, sweet food intake is increased by opioid agonists and decreased by opioid antagonists [bib_ref] Endogenous opioid peptides and regulation of drinking and feeding, Reid [/bib_ref] [bib_ref] Naloxone's anorectic effect is dependant upon the relative palatability of food, Giraudo [/bib_ref]. The effect of chocolate is also exerted through interaction with neurotransmitters such as dopamine (tyrosine contained in chocolate is the precursor to dopamine), serotonin, and endorphins, which contribute to appetite, reward, and mood regulation. The dopaminergic system contributes to the desire to consume chocolate, probably by acting mainly non-specifically towards food. After carbohydrate ingestion, only when the protein component of the meal is less than 2% does it induce an increase in serotonin concentrations in the brain [bib_ref] Atypical Depression: A Reappraisal, Parker [/bib_ref]. It should be noted that chocolate contains 5% of its caloric content in the form of protein, which would cancel out any effect of serotonin. Moreover, manipulations of tryptophan, the precursor of serotonin, also cause physiological changes that are too slow to explain the mood effects described during or immediately after eating chocolate [bib_ref] Tryptophan depletion causes a rapid lowering of mood in normal males, Young [/bib_ref]. Another area where chocolate might act could be in the area of opioids, which are known to play a role in the palatability of preferred foods [bib_ref] Opioid and non-opioid components of insulin-induced feeding, Si [/bib_ref] , releasing endorphins during food intake, and, thus, justifying the increase in pleasure during food intake [bib_ref] Food and mood, Ottley [/bib_ref]. The mood effects of cocoa may also be partly due to opioids released in response to the ingestion of sweets and other pleasantly palatable foods [bib_ref] Mood state effects of chocolate, Parker [/bib_ref] [bib_ref] Sugar, opioids and binge eating, Fullerton [/bib_ref]. The increase in central opioidergic activity, in turn, stimulates the immediate release of beta-endorphin in the hypothalamus, which exerts an analgesic effect. Bad mood stimulates consumption of comfort foods such as chocolate in two different ways [bib_ref] Depression" increases "craving" for sweet rewards in animal and human models of..., Willner [/bib_ref]. The former is called craving and is associated with an impulsive desire for chocolate, and its compulsive consumption occurs especially when under high emotional stress, showing a clear link between the perception of a negative mood and the intense desire to consume chocolate [bib_ref] Chocolate Addiction": A Preliminary Study of its Description and its Relationship to..., Hetherington [/bib_ref]. The association between chocolate craving and consumption under emotional stress was demonstrated in a study in which subjects had to listen to music that induced a happy or sad mood. Chocolate consumption was increased by listening to sad music [bib_ref] Depression" increases "craving" for sweet rewards in animal and human models of..., Willner [/bib_ref].
The second modality to be considered is the palatability of the food. The pleasure induced by palatable food is regulated by endogenous opioids that stimulate food intake in rats. The pleasure induced by palatable foods is regulated by endogenous opioids that stimulate their intake in rats. In humans, however, the critical factor in satisfying chocolate craving appears to be taste and mouthfeel [bib_ref] Chocolate craving and liking, Rozin [/bib_ref]. Females, mainly in the perimenstrual period, seems more sensitive to chocolate. The response to satiety seems to vary by gender [bib_ref] Effect of satiety on brain activation during chocolate tasting in men and..., Smeets [/bib_ref].
It is more conceivable that an important role in liking or craving chocolate is due more to the composite sensory properties of chocolate than to its role in appetite and satiety [bib_ref] Pharmacological versus sensory factors in the satiation of chocolate craving, Michener [/bib_ref]. During the consumption of chocolate, different brain areas are also activated depending on the motivation to eat chocolate, based on positive/appetitive stimuli or associated with negative/adverse stimuli. Modulation of brain activity has been observed in chemosensory cortical areas such as the insula, prefrontal regions, and caudomedial and caudolateral orbitofrontal cortex, with overlapping and co-activation under contrasting motivational conditions [bib_ref] A 3 T event-related functional magnetic resonance imaging (fMRI) study of primary..., Smits [/bib_ref]. The ability to activate images of appetizing foods involved in food motivation and hedonism in a fronto-striatal-amygdala-midbrain network appears to be dependent on individual variability in reward sensitivity. If this same neuronal circuit is stimulated in the animal, it may result in the cancellation of the sense of satiety and cause overeating of highly palatable foods [bib_ref] Individual Differences in Reward Drive Predict Neural Responses to Images of Food, Beaver [/bib_ref].
The smell of chocolate itself is sufficient to modulate brain activity recorded on the electroencephalography (EEG). The smell of chocolate induces a significant reduction in theta activity compared to any other stimulus. Theta activity is considered to be closely related to attentional level, cognitive load in general, and, in this specific case, to olfactory perception, so a reduction in theta activity could be indicative of a reduced level of attention and an increased propensity to distraction [bib_ref] Human electroencephalographic (EEG) response to olfactory stimulation: Two experiments using the aroma..., Martin [/bib_ref]. In addition to olfaction, the sight of chocolate also evokes activations in the brain and especially in the medial orbitofrontal cortex and ventral striatum, particularly in subjects who crave chocolate compared to those who do not: the combination of an image of chocolate with chocolate in the mouth evoked greater brain activation than the sight of the sum of the different components in the medial orbitofrontal cortex and cingulate cortex [bib_ref] Enhanced affective brain representations of chocolate in cravers vs. non-cravers, Rolls [/bib_ref].
The motivation for chocolate preference appears to be primarily, if not entirely, sensory. The origin of the liking of its sensory properties is unclear; it could be innate or acquired based on the sweetness, texture, and aroma characteristics of chocolate, or it could depend on the interaction between a person's state and the post-gastronomic effects of chocolate. Surprisingly, there is little evidence of a relationship between chocolate addiction and chocolate liking [bib_ref] Chocolate craving and liking, Rozin [/bib_ref]. However, chocolate consumption fails to activate the key structure for drug addiction, the nucleus accumbens [bib_ref] Are we dependent upon coffee and caffeine? A review on human and..., Nehlig [/bib_ref] [bib_ref] Nucleus accumbens shell and core dopamine: Differential role in behavior and addiction, Di Chiara [/bib_ref] [bib_ref] A common profile of prefrontal cortical activation following exposure to nicotine-or chocolate-associated..., Schroeder [/bib_ref].
The effect of chocolate on mood may be attributed to the affinity for adenosine and benzodiazepine (GABAa) receptors of polyphenolic compounds, which means that their ingestion may have a soothing effect [bib_ref] Overview-Flavonoids: A New Family of Benzodiazepine Receptor Ligands, Medina [/bib_ref]. Some polyphenolic compounds indeed have anxiolytic properties [bib_ref] Anxiolytic properties of green tea polyphenol (−)-epigallocatechin gallate (EGCG), Vignes [/bib_ref]. A small randomized controlled pilot study in humans with chronic fatigue reported a reduction in anxiety-related symptoms after eating polyphenolrich chocolate, compared to polyphenol-poor chocolate [bib_ref] High cocoa polyphenol rich chocolate may reduce the burden of the symptoms..., Sathyapalan [/bib_ref].
Pase et al. [bib_ref] Cocoa polyphenols enhance positive mood states but not cognitive performance: A randomized,..., Pase [/bib_ref] investigated the acute and subchronic effects of polyphenol supplementation on mood and cognitive performance in a randomized, placebo-controlled, double-blind study. Thirty days of treatment with a high dose of cocoa polyphenols reduced self-rated anxiety and contentment. No significant effect on cognitive performance was recorded with either the high or low dose in either the acute or chronic phase.
The optimal dosage of cocoa polyphenols needed to improve cognitive function and mood remains unclear.
## Chocolate, sleep and circadian rhythms
Our modern society functions at a hectic pace of activity 24/7, which leads individuals to sacrifice sleep hours and disregard daily sleep-wake rhythms. As experienced daily by shift workers, jet-lagged travelers, or those with so-called social jet-lag syndrome, disturbed sleep-wake rhythms create a conflict, a temporal mismatch between the circadian system and temporal signals derived from the cyclical environmental changes, such as the lightdark cycle, [bib_ref] The Circadian System: A Regulatory Feedback Network of Periphery and Brain, Buijs [/bib_ref] or a desynchronization. This condition, if prolonged over time, leads to chronic sleep disorders (CSD) that result in deficits to health and psychophysical well-being. The most frequent symptoms belong to the neurobehavioral sphere and are often associated with mood changes such as a tendency to depression and impairment of cognitive functions, especially executive functions, as well as cardio-and cerebrovascular disorders, stroke, hypertension, obesity, and diabetes [bib_ref] The metabolic consequences of sleep deprivation, Knutson [/bib_ref]. CSD are a socio-economic and public health issue due to their high prevalence in the general and working population, their impact on health, and work output given the higher incidence of absenteeism, and increased rates of errors and accidents at work [bib_ref] Sleep problems and work injuries: A systematic review and meta-analysis, Uehli [/bib_ref]. In addition, CSD often have a bidirectional relationship with stress [bib_ref] The impact of stress on sleep: Pathogenic sleep reactivity as a vulnerability..., Kalmbach [/bib_ref].
Jet lag results from a sudden change in the light-dark cycle due to trans-south travel or social life (in the case of social jet lag), which leads to a misalignment between internal circadian rhythms, mainly but not limited to SCN, and the external rhythmic time-cues (Zeitgeber), mainly light, for the day-night cycle. The days required for the process of coordinating the internal circadian clock to external rhythmic are often associated with behavioral and physiological discomfort.
Escobar C. et al.observed that chocolate administration resulted in a faster rate of realignment and synchronization between activity-rest cycle and circadian temperature rhythms. Programmed access to chocolate activates brain areas involved in motivation and metabolic response to food [bib_ref] Expectancy for food or expectancy for chocolate reveals timing systems for metabolism..., Ángeles-Castellanos [/bib_ref] as well as the circadian system by improving neuronal activation in the SCN [bib_ref] A daily palatable meal without food deprivation entrains the suprachiasmatic nucleus of..., Mendoza [/bib_ref]. Other studies have also reported a similar effect on the speed of realignment to feeding schedules in peripheral oscillatory clocks [bib_ref] Effects of light cues on re-entrainment of the food-dominated peripheral clocks in..., Wu [/bib_ref] [bib_ref] Restricted feeding entrains liver clock without participation of the suprachiasmatic nucleus, Hara [/bib_ref] [bib_ref] Resetting Central and Peripheral Circadian Oscillators in Transgenic Rats, Yamazaki [/bib_ref].
However, it is not only the type of food that determines the ability to realign and re-synchronize internal clocks with external signals but also the time of intake of food, especially for palatable food.
Due to this complex interaction between external Zeitgeber and internal circadian rhythms, the greatest beneficial effects of entrainment on circadian function are seen when food intake coincides with the activity phase [bib_ref] Shift Work or Food Intake during the Rest Phase Promotes Metabolic Disruption..., Salgado-Delgado [/bib_ref] [bib_ref] Time-restricted feeding improves adaptation to chronically alternating light-dark cycles, Schilperoort [/bib_ref] , whereas an inhibitory effect occurs when food is taken during the rest phase [bib_ref] Shifting eating to the circadian rest phase misaligns the peripheral clocks with..., Mukherji [/bib_ref] [fig_ref] Figure 1: Schematic illustration of chrononutrition with chocolate for breakfast and components of chocolate... [/fig_ref]. by shift workers, jet-lagged travelers, or those with so-called social jet-lag syndrome, disturbed sleep-wake rhythms create a conflict, a temporal mismatch between the circadian system and temporal signals derived from the cyclical environmental changes, such as the light-dark cycle, [bib_ref] The Circadian System: A Regulatory Feedback Network of Periphery and Brain, Buijs [/bib_ref] or a desynchronization. This condition, if prolonged over time, leads to chronic sleep disorders (CSD) that result in deficits to health and psychophysical well-being. The most frequent symptoms belong to the neurobehavioral sphere and are often associated with mood changes such as a tendency to depression and impairment of cognitive functions, especially executive functions, as well as cardio-and cerebrovascular disorders, stroke, hypertension, obesity, and diabetes [bib_ref] The metabolic consequences of sleep deprivation, Knutson [/bib_ref]. CSD are a socio-economic and public health issue due to their high prevalence in the general and working population, their impact on health, and work output given the higher incidence of absenteeism, and increased rates of errors and accidents at work [bib_ref] Sleep problems and work injuries: A systematic review and meta-analysis, Uehli [/bib_ref]. In addition, CSD often have a bidirectional relationship with stress [bib_ref] The impact of stress on sleep: Pathogenic sleep reactivity as a vulnerability..., Kalmbach [/bib_ref].
Jet lag results from a sudden change in the light-dark cycle due to trans-south travel or social life (in the case of social jet lag), which leads to a misalignment between internal circadian rhythms, mainly but not limited to SCN, and the external rhythmic time-cues (Zeitgeber), mainly light, for the day-night cycle. The days required for the process of coordinating the internal circadian clock to external rhythmic are often associated with behavioral and physiological discomfort.
Escobar C. et al.observed that chocolate administration resulted in a faster rate of realignment and synchronization between activity-rest cycle and circadian temperature rhythms. Programmed access to chocolate activates brain areas involved in motivation and metabolic response to food [bib_ref] Expectancy for food or expectancy for chocolate reveals timing systems for metabolism..., Ángeles-Castellanos [/bib_ref] as well as the circadian system by improving neuronal activation in the SCN [bib_ref] A daily palatable meal without food deprivation entrains the suprachiasmatic nucleus of..., Mendoza [/bib_ref]. Other studies have also reported a similar effect on the speed of realignment to feeding schedules in peripheral oscillatory clocks [bib_ref] Effects of light cues on re-entrainment of the food-dominated peripheral clocks in..., Wu [/bib_ref] [bib_ref] Restricted feeding entrains liver clock without participation of the suprachiasmatic nucleus, Hara [/bib_ref] [bib_ref] Resetting Central and Peripheral Circadian Oscillators in Transgenic Rats, Yamazaki [/bib_ref].
However, it is not only the type of food that determines the ability to realign and resynchronize internal clocks with external signals but also the time of intake of food, especially for palatable food.
Due to this complex interaction between external Zeitgeber and internal circadian rhythms, the greatest beneficial effects of entrainment on circadian function are seen when food intake coincides with the activity phase [bib_ref] Shift Work or Food Intake during the Rest Phase Promotes Metabolic Disruption..., Salgado-Delgado [/bib_ref] [bib_ref] Time-restricted feeding improves adaptation to chronically alternating light-dark cycles, Schilperoort [/bib_ref] , whereas an inhibitory effect occurs when food is taken during the rest phase [bib_ref] Shifting eating to the circadian rest phase misaligns the peripheral clocks with..., Mukherji [/bib_ref] [fig_ref] Figure 1: Schematic illustration of chrononutrition with chocolate for breakfast and components of chocolate... [/fig_ref]. Furthermore, to maintain a coordinated and synchronized circadian function, food intake must be phased with the light-dark cycle. The main effect seems to be due to a direct synchronization action on brain oscillators and central and peripheral clocks [bib_ref] Effects of light cues on re-entrainment of the food-dominated peripheral clocks in..., Wu [/bib_ref] [bib_ref] Chronobiological aspects of food intake and metabolism and their relevance on energy..., Ekmekcioglu [/bib_ref].
Under normal light and dark (LD) conditions, programmed food intake does not shift the SCN phase [bib_ref] Restricted feeding entrains liver clock without participation of the suprachiasmatic nucleus, Hara [/bib_ref] [bib_ref] Restricted-feeding-induced anticipatory activity rhythm is associated with a phase-shift of the expression..., Wakamatsu [/bib_ref]. Other studies indicate that SCN is inhibited during food anticipation and fasting as observed with c-Fos, a major early gene that is activated by external signals [bib_ref] Activation of c-fos in the brain, Herrera [/bib_ref] [bib_ref] Circadian blueprint of metabolic pathways in the brain, Greco [/bib_ref] [bib_ref] Interaction between hypothalamic dorsomedial nucleus and the suprachiasmatic nucleus determines intensity of..., Acosta-Galvan [/bib_ref] or electrophysiological recordings, whereas the ventral SCN is activated both after re-feeding and with light [bib_ref] The NPY intergeniculate leaflet projections to the suprachiasmatic nucleus transmit metabolic conditions, Saderi [/bib_ref].
Recent results indicate that the SCN may also respond to palatable food construed as hedonic information, via dopaminergic projections from the ventral tegmental area [bib_ref] Direct Midbrain Dopamine Input to the Suprachiasmatic Nucleus Accelerates Circadian Entrainment, Grippo [/bib_ref]. The rapid achievement of synchronization with limited daily chocolate intake may also be partly due to the increase in arousal induced by chocolate intake as a hedonic effect.
When planned for breakfast, an appetizing food, such as chocolate, can influence activation in the SCN, at the level of the dorsomedial region. This rhythmic pattern in the dorsal SCN may promote faster re-entrainment [bib_ref] Scheduled meal accelerates entrainment to a 6-h phase advance by shifting central..., Ubaldo [/bib_ref] when bounded by a time window in which chocolate was administered during the active phase, whereas chocolate did not promote re-entrainment when administered during the resting phase.
Scheduled feeding has been shown to be a strong entrainment signal for circadian rhythm; especially when food intake is in phase with the period of activity. This exerts beneficial effects on the circadian system by favoring its synchronization and activation of the metabolism [bib_ref] Scheduled Food Hastens Re-Entrainment More Than Melatonin Does after a 6-h Phase..., Ángeles-Castellanos [/bib_ref] [bib_ref] Food Intake during the Normal Activity Phase Prevents Obesity and Circadian Desynchrony..., Salgado-Delgado [/bib_ref] , as demonstrated in experimental studies on shift-worker models. Time-limited access to food accelerates resynchronization in a jet-lag model, prevents circadian desynchronization in a shift-work model, and induces positive effects in metabolism [bib_ref] Shift-work: Is time of eating determining metabolic health? Evidence from animal models, Guerrero-Vargas [/bib_ref] [bib_ref] Time-restricted feeding improves glucose tolerance in men at risk for type 2..., Hutchison [/bib_ref]. In contrast, food scheduled in the sleep-rest phase slows circadian synchronization and metabolism and alters behavior [bib_ref] Shift-work: Is time of eating determining metabolic health? Evidence from animal models, Guerrero-Vargas [/bib_ref] [bib_ref] Time-restricted feeding improves glucose tolerance in men at risk for type 2..., Hutchison [/bib_ref] [bib_ref] Time-Restricted Eating to Prevent and Manage Chronic Metabolic Diseases, Chaix [/bib_ref].
Recently, Oishi et al. [bib_ref] Dietary natural cocoa ameliorates disrupted circadian rhythms in locomotor activity and sleep-wake..., Oishi [/bib_ref] confirmed the positive action of cocoa on sleep disturbance induced by psychophysiological stress in mice using EEG. Cocoa intake attenuated the alteration of circadian sleep-wake rhythms. The EEG revealed that cocoa significantly improved both the increase in the level of alertness during the first half of the light period and the increase in NREM sleep during the first half of the dark period in mice with CSD. Under non-CSD conditions, cocoa does not appear to influence either the rhythms of run-rest activity or sleep-wake cycles. It is hypothesized that this positive action may be attributable to high concentrations of flavonoids in cocoa (epicatechin, catechin, and procyanidins), which improve blood flow and have antioxidant and neuroprotective properties. Indeed, in experimental animals, sleep deprivation and CSDs in general have been shown to increase oxidative stress levels in specific brain regions such as the hypothalamus, hippocampus, and thalamus [bib_ref] Sleep Deprivation and Oxidative Stress in Animal Models: A Systematic Review, Villafuerte [/bib_ref].
Flavanols acting on endothelial function could also play a role in insomnia, as it is just endothelial dysfunction that appears to be responsible for some insomnia-related symptoms and the association between insomnia and cardiovascular disease [bib_ref] Insomnia Symptoms Are Associated With Abnormal Endothelial Function, Routledge [/bib_ref]. Cocoa flavanols, by facilitating nitric oxide production, improve vascular endothelial function due to their vasodilatory effect [bib_ref] Chocolate and the brain: Neurobiological impact of cocoa flavanols on cognition and..., Sokolov [/bib_ref]. CBF also appears to play an important role in sleep regulation [bib_ref] Regional cerebral blood flow throughout the sleep-wake cycle. An H2(15)O PET study, Braun [/bib_ref] as well as cognitive and emotional processes, although it is not known how cerebral blood flow varies during alternating sleep-wake cycles [bib_ref] Cerebral blood flow changes after a day of wake, sleep, and sleep..., Elvsåshagen [/bib_ref]. Flavanolrich cocoa significantly increases cerebral blood flow in humans [bib_ref] Chocolate and the brain: Neurobiological impact of cocoa flavanols on cognition and..., Sokolov [/bib_ref] and attenuates the CSD-induced disturbance of circadian activity rhythm, sleep-wake cycles, cognitive functions by improving cerebral endothelial cell function, and blood flow, as demonstrated by Grassi et al. [bib_ref] Brain Protection and Cognitive Function: Cocoa Flavonoids as Nutraceuticals, Grassi [/bib_ref].
In addition, another pathway through which cocoa has a protective effect on the synchronous maintenance of the sleep-wake rhythm in subjects with CSD is the modulated neurotransmission of serotonin [bib_ref] Effects of indole-pyruvic acid on sleep and food intake in the rat, Pich [/bib_ref] [bib_ref] Indole-pyruvic acid, a tryptophan ketoanalogue, antagonizes the endocrine but not the behavioral..., Biagini [/bib_ref]. In fact, regular cocoa consumption has been shown to increase serotonin concentrations in the brain [bib_ref] Anxiolytic effects of short-and long-term administration of cacao mass on rat elevated..., Yamada [/bib_ref].
Acute administration of flavanol-rich chocolate can counteract the negative effects of total sleep deprivation both on working memory performance in healthy young people [bib_ref] Brain Protection and Cognitive Function: Cocoa Flavonoids as Nutraceuticals, Grassi [/bib_ref] and on endothelial and arterial function and, thus, on blood pressure [fig_ref] Figure 1: Schematic illustration of chrononutrition with chocolate for breakfast and components of chocolate... [/fig_ref]. Natural cocoa seems to be an ideal nutrient for ameliorating stress-induced psychophysiological sleep disturbance without distorting behavioral or sleep regulation under normal conditions.
# Conclusions
Modern society imposes increasingly stressful rhythms of life and work, and technological innovations have led us to live an active life 24/7. The immediate effect is on the quality and quantity of sleep, which in turn has repercussions on our lower tolerance to stress, alterations in mood, and greater susceptibility to infectious, metabolic, and cardiovascular diseases. Lifestyles are undoubtedly the main targets to work on to maintain a harmonious synchronization between our central and peripheral endogenous clocks and external synchronizers. Of course, the circadian rhythm of light and dark is the most powerful synchronizer, but we must not forget the other fundamental synchronizer: food and fasting.
Chocolate is a food that, due to its high flavonoid content, when taken in the morning, can have positive effects on our mental and cognitive well-being, our cardiovascular system, and our metabolism. Chocolate also induces positive effects on mood and is, therefore, often spontaneously consumed under conditions of emotional stress. In addition to the beneficial effects on the vascular system and cerebral blood flow, flavonoids have a protective function on the neuronal cell by inhibiting neuronal death by apoptosis induced by neurotoxicants, such as oxygen radicals through interaction with signaling cascades involving proteins and lipid kinases. They also promote neuronal survival, neurogenesis, and synaptic plasticity, preserving cognitive abilities during the stages of aging. Crossing the BBB, they act on both the vascular and neurocellular sides of the brain, on one hand, by stimulating cerebral perfusion and promoting angiogenesis, and, on the other, by modulating changes in the morphology of the neurons participating in learning and memory processes. All these properties are of great interest, but it is currently unclear when consumption of flavonoid-rich cocoa and chocolate should be initiated to achieve beneficial and protective effects against the mechanisms underlying cognitive decline and age-dependent neurodegenerative diseases [bib_ref] Cocoa and Chocolate in Human Health and Disease, Katz [/bib_ref]. Many studies are still needed to understand the mechanisms of action and the timing of the neuroprotective activity of cocoa and chocolate.
Although the dosages to be taken in order to achieve the beneficial effects are still unclear, the literature data support the suggestion that the beneficial effect is obtained by eating cocoa only in modest amounts, preferably dark chocolate, during the activity hours of the day and in the early part of the day to avoid problems of weight gain. Frequent consumption of chocolate managed in this way might actually be associated with a lower body mass index [bib_ref] Association between more frequent chocolate consumption and lower body mass index, Golomb [/bib_ref].
Time of food intake is now suggested as a chronotherapeutic strategy that may help speed up the time needed to resynchronize biological clocks and, thus, reduce circadian disruption caused by shift work, jet lag, and social jet lag [bib_ref] Scheduled meal accelerates entrainment to a 6-h phase advance by shifting central..., Ubaldo [/bib_ref] [bib_ref] Food Intake during the Normal Activity Phase Prevents Obesity and Circadian Desynchrony..., Salgado-Delgado [/bib_ref] [bib_ref] Shift-work: Is time of eating determining metabolic health? Evidence from animal models, Guerrero-Vargas [/bib_ref]. Palatable food, such as chocolate, scheduled for breakfast is a valuable aid in maintaining circadian synchrony and improving body weight. The present data agree with the previous literature and indicate that the thermogenic effect of a high-calorie food such as chocolate is actually different depending on the time of intake in relation to the activity-rest phase: high postprandial thermogenic response takes place during the active phase, not during the rest phase. Breakfast induces a strong postprandial thermogenesis, which induces an increase in energy expenditure [bib_ref] High-energy breakfast based on whey protein reduces body weight, postprandial glycemia and..., Jakubowicz [/bib_ref] [bib_ref] The causal role of breakfast in energy balance and health: A randomized..., Betts [/bib_ref] [bib_ref] Circadian variation of diet-induced thermogenesis, Romon [/bib_ref].
A piece of chocolate a day may also modulate circadian oscillations in brain areas involved in the reward system [bib_ref] Bidirectional interactions between the circadian and reward systems: Is restricted food access..., Webb [/bib_ref] [bib_ref] Restricted feeding schedules phase shift daily rhythms of c-Fos and protein Per1..., Ángeles-Castellanos [/bib_ref]. It may also have effects on behavior, mood, and cognitive functions.
In a socio-historic period such as the current one, when psychophysical distress is significantly increasing, due in part to the pandemic that has led to social distancing, significant repercussions on economic stability and the ability to envision future prospects are to be expected. The intake of chocolate rich in flavanols at breakfast and during the active phase, in accordance with the rules of chrononutrition, could provide a new tool that is economical, quick, and easily applicable for all ages to support cognitive performance during periods of sleep deprivation and psychophysical stress as well as during phases of desynchronization of circadian rhythms, with a positive effect on mood, while also helping to protect our cardiovascular system and regulate our metabolism.
Future studies in selected populations will be essential to establish the correct ways of taking chocolate by defining dosages as well as timing during the day and over a lifetime.
Author Contributions: Conceptualization, S.G. and P.L.; methodology, S.G. and P.L.; data curation, S.G., E.G. and P.L.; writing-original draft preparation, S.G., E.G. and P.L.; writing-review and editing, S.G., E.G. and P.L.; supervision, S.G., E.G. and P.L. All authors have read and agreed to the published version of the manuscript.
## Conflicts of interest:
The authors declare no conflict of interest.
[fig] Figure 1: Schematic illustration of chrononutrition with chocolate for breakfast and components of chocolate with the main mechanisms of action in both the central nervous system and peripheral organs and their actions. [/fig]
[fig] Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. [/fig]
|
Cardiorespiratory responses to exercise related to post-stroke fatigue severity
Physical deconditioning after stroke may induce post-stroke fatigue. However, research on this association is limited. Our primary objective was to investigate the associations of post-stroke fatigue severity with oxygen uptake ( V O 2 ) at peak exercise and the time constant of V O 2 kinetics (τV O 2 ) at exercise onset. The secondary objective was to examine the associations between fatigue and cardiorespiratory variables potentially affecting V O 2 during exercise. Twenty-three inpatients from a subacute rehabilitation ward were enrolled in this study. The median (interquartile range) Fatigue Severity Scale (FSS) score, as a measure of fatigue, was 32 (range 27-42) points. The FSS score was not associated with V O 2 at peak exercise during a symptom-limited graded exercise test (rho = − 0.264; p = 0.224), whereas it was significantly associated with τ V O 2 during a submaximal constant-load exercise test (rho = 0.530; p = 0.009). A higher FSS score also significantly correlated with a longer time constant of cardiac output (CO) kinetics (rho = 0.476; p = 0.022). Our findings suggest that severe poststroke fatigue is associated with delayed increases in V O 2 and CO at the onset of exercise. Our findings can contribute to the development of an appropriate rehabilitation programme for individuals with post-stroke fatigue.Post-stroke fatigue is defined as 'a subjective lack of physical and/or mental energy that is perceived by the individual to interfere with usual or desired activities' 1,2 . A systematic review reported that the prevalence of post-stroke fatigue ranged between 25 and 85% 2 . Post-stroke fatigue is associated with various factors, such as depressive symptoms and functional disability 1 , and individuals with post-stroke fatigue are reported to have poor recovery in terms of activities of daily living 3,4 , a lower rate of returning to work 5 , reduced health-related quality of life 6 , and increased mortality 4,7 . The underlying pathophysiology of post-stroke fatigue is not completely understood 1 . Although exercise training can improve fitness, balance, mobility, and activities of daily living in individuals with stroke 8 , there is insufficient evidence regarding the effectiveness of rehabilitative exercise programmes for improving post-stroke fatigue 1,9 .It has been suggested that post-stroke fatigue is triggered through physical deconditioning, which may lead to the avoidance of physical activities and further deconditioning 10 . Oxygen uptake ( V O 2 ) at peak exercise, measured during a symptom-limited graded exercise test, is widely accepted as an indicator of cardiorespiratory capacity in individuals with stroke 8,11,12 . V O 2 at peak exercise in individuals with stroke is 26-87% of that in healthy age-and sex-matched individuals 11 . Moreover, the assessment of V O 2 kinetics at the onset of submaximal exercise has also been shown to provide objective information on the cardiorespiratory fitness of individuals with stroke 13,14 . Transient measurements of V O 2 in a constant-load exercise at an intensity below the ventilatory threshold are classified into three phases. Time constant of V O 2 in phase II (τVO 2 ) has often been used to assess V O 2 kinetics at the onset of exercise, which reflects the ability of the cardiorespiratory system to adapt from rest to a new steady-state during submaximal exercise 15 . A longer τ V O 2 is associated with poorer health status, ageing, and a sedentary lifestyle15,16 . Tomczak et al. 13 reported that τ V O 2 was greater in individuals with stroke than in age-, sex-, and activity-matched healthy adults. However, there is limited evidence concerning the associations between post-stroke fatigue and these cardiorespiratory fitness variables; therefore, this study aimed to investigate these associations.The variables involved in post-stroke fatigue may differ based on the severity/stage of stroke. Wu et al. 17 proposed a conceptual model of post-stroke fatigue, wherein biological factors are expected to trigger fatigue at the OPEN
www.nature.com/scientificreports/ early stage after stroke (usually within the first three months after stroke), whereas fatigue at the later stage after stroke (usually > 1 year after stroke) is attributed to psychological and behavioural factors. A cross-sectional study on individuals with chronic stroke (4.1 ± 3.5 years post-stroke) reported that post-stroke fatigue was associated with depressive symptoms but not with V O 2 at peak exercise [bib_ref] Exertion fatigue and chronic fatigue are two distinct constructs in people post-stroke, Tseng [/bib_ref]. However, the association between post-stroke fatigue at the early stage after stroke and cardiorespiratory fitness has not been reported. The primary objective of this study was to examine the associations of the severity of post-stroke fatigue with V O 2 at peak exercise obtained during a symptom-limited graded exercise test and τ V O 2 at the onset of exercise measured during a submaximal constant-load exercise test in inpatients at a subacute rehabilitation ward. In previous studies, τ V O 2 has been reported to be more sensitive to changes in the levels of physical activity compared with V O 2 at peak exercise [bib_ref] Progressive effect of endurance training on VO 2 kinetics at the onset..., Phillips [/bib_ref] [bib_ref] Changes in pulmonary oxygen uptake and muscle deoxygenation kinetics during cycling exercise..., Hamasaki [/bib_ref] [bib_ref] Early effects of exercise training on on-and off-kinetics in 50-year-old subjects, Fukuoka [/bib_ref]. Moreover, post-stroke fatigue has been found to be related to lower levels of physical activity [bib_ref] Factors associated with post-stroke physical activity: A systematic review and meta-analysis, Thilarajah [/bib_ref]. Based on these findings, we hypothesised that the severity of post-stroke fatigue would likely be more strongly associated with τ V O 2 than V O 2 at peak exercise. In addition, respiratory and cardiac function impairment in relation to supplying oxygen and an inability of skeletal muscles to extract oxygen may limit the increase in V O 2 during exercise in individuals with stroke [bib_ref] Cardiac reserve and pulmonary gas exchange kinetics in patients with stroke, Tomczak [/bib_ref] [bib_ref] Decreased tidal volume may limit cardiopulmonary performance during exercise in subacute stroke, Sisante [/bib_ref] [bib_ref] Cardiorespiratory factors related to the increase in oxygen consumption during exercise in..., Oyake [/bib_ref]. Therefore, our secondary objective was to identify associations between post-stroke fatigue and cardiorespiratory variables potentially affecting V O 2 during exercise, such as the oxygen uptake efficiency slope (OUES), cardiac output (CO), and arterial-venous oxygen difference (AVO 2 diff). Ventilatory efficiency and muscle oxygen extraction, measured using OUES and AVO 2 diff, respectively, have been reported to be lower in individuals with stroke than in healthy adults [bib_ref] Evolution of cardiorespiratory fitness after stroke: A 1-year follow-up study. Influence of..., Baert [/bib_ref] [bib_ref] Discrepancy between cardiac and physical functional reserves in stroke, Jakovljevic [/bib_ref]. Therefore, we hypothesised that impairment of these variables would also be associated with post-stroke fatigue. Elucidating cardiorespiratory factors associated with post-stroke fatigue could contribute to the development of an appropriate rehabilitation programme for individuals with post-stroke fatigue.
# Results
## Participants.
A flow chart of participants enrolled in the study is shown in [fig_ref] Figure 1: Flow diagram of study participants [/fig_ref]. Thirty individuals with stroke provided informed consent. However, two participants declined to perform exercise tests. Furthermore, in five of 28 participants who performed the submaximal constant-load exercise test, cardiorespiratory data during the test could not be measured because of technical difficulties. Consequently, data concerning 23 participants were included in the analysis. Although all participants were recruited from a subacute rehabilitation ward, five participants were in the chronic phase of stroke recovery (≥ 3 months after stroke) [bib_ref] Facilitating recovery: Evidence for organized stroke care, Kalra [/bib_ref]. [fig_ref] Table 1: Associations between the Fatigue Severity Scale score and participants' characteristics [/fig_ref] shows the participants' characteristics. Regarding cardiorespiratory variables measured during the submaximal constant-load exercise test, the mean ± standard deviation (SD) coefficients of determination of the kinetics of V O 2 , CO, AVO 2 diff, and minute ventilation ( V E) were 0.99 ± 0.01, 0.96 ± 0.02, 0.93 ± 0.03, and 0.98 ± 0.01, respectively. In addition, the mean ± SD ratio of the time constant of CO (τCO) to τ V O 2 was 1.19 ± 0.56. In 15 of 23 (65.2%) participants, the ratio of τCO to τ V O 2 was > 1.00.
Measurement values obtained during the symptom-limited graded and submaximal constant-load exercise tests are shown in [fig_ref] Table 2: Correlations between the Fatigue Severity Scale score and cardiorespiratory variables during a... [/fig_ref].
Correlations between the Fatigue Severity Scale score and cardiorespiratory variables measured during the symptom-limited graded and submaximal constant-load exercise tests [fig_ref] Table 2: Correlations between the Fatigue Severity Scale score and cardiorespiratory variables during a... [/fig_ref]. The median (interquartile range) Fatigue Severity Scale (FSS) score was 32 (range [bib_ref] Facilitating recovery: Evidence for organized stroke care, Kalra [/bib_ref] [bib_ref] Reduced cardiorespiratory fitness after stroke: Biological consequences and exercise-induced adaptations, Billinger [/bib_ref] [bib_ref] Effects of circuit training as alternative to usual physiotherapy after stroke: Randomised..., Van De Port [/bib_ref] [bib_ref] Hypertension and daytime hypotension found on ambulatory blood pressure is associated with..., Harbison [/bib_ref] [bib_ref] Experiences of poststroke fatigue: Qualitative meta-synthesis, Eilertsen [/bib_ref] [bib_ref] Physical activity and exercise after stroke: Review of multiple meaningful benefits, Saunders [/bib_ref] [bib_ref] Reproducibility of onset and recovery oxygen uptake kinetics in moderately impaired patients..., Kemps [/bib_ref] [bib_ref] Are oxygen uptake kinetics in chronic heart failure limited by oxygen delivery..., Kemps [/bib_ref] [bib_ref] Oxygen deficits and oxygen delivery kinetics during submaximal intensity exercise in humans..., Capelli [/bib_ref] [bib_ref] Cardiac atrophy after bed rest and spaceflight, Perhonen [/bib_ref] [bib_ref] Left ventricular remodeling during and after 60 days of sedentary head-down bed..., Westby [/bib_ref] [bib_ref] The metabolic and hemodynamic effects of prolonged bed rest in normal subjects, Chobanian [/bib_ref] [bib_ref] Cardiovascular consequences of bed rest: effect on maximal oxygen uptake, Convertino [/bib_ref] [bib_ref] Speeding of VO 2 kinetics in response to endurance-training in older and..., Murias [/bib_ref] [bib_ref] Speeding of VO 2 kinetics with endurance training in old and young..., Murias [/bib_ref] [bib_ref] Cognitive and graded activity training can alleviate persistent fatigue after stroke: A..., Zedlitz [/bib_ref]. The mean ± SD V O 2 values at peak exercise and τ V O 2 at the onset of exercise were 18.0 ± 4.2 mL kg −1 min −1 and 38.6 ± 10.1 s, respectively.
The FSS score was not significantly correlated with V O 2 at peak exercise (rho = − 0.264; p = 0.224, [fig_ref] Figure 2: Correlations of the Fatigue Severity Scale score with [/fig_ref] , whereas a higher FSS score significantly correlated with a longer τ V O 2 (rho = 0.530; p = 0.009, [fig_ref] Figure 2: Correlations of the Fatigue Severity Scale score with [/fig_ref]. Although the FSS score was not associated with the other cardiorespiratory variables at peak exercise and at the ventilatory threshold measured during the symptom-limited exercise test, a higher FSS score was significantly associated with a longer τCO (rho = 0.476; p = 0.022, [fig_ref] Figure 3: Correlation between the Fatigue Severity Scale score and the time constant of... [/fig_ref] during the submaximal constant-load exercise test. Changes in V O 2 and CO at the onset of exercise in representative participants with different fatigue levels (low and high) are shown in Supplementary Figs. S1 and S2, respectively, online.
The association between the FSS score and τ V O 2 after adjusting for participants' characteristics. The FSS score was not found to be significantly associated with participants' characteristics, including age, sex, height, weight, body mass index, type of stroke, side of motor paresis, time since stroke, presence of hypertension and diabetes mellitus, Mini-Mental State Examination (MMSE) score, presence of depressive symptoms, Stroke Impairment Assessment Set motor score, and Functional Independence Measure score in the motor and cognition items (p > 0.05, [fig_ref] Table 1: Associations between the Fatigue Severity Scale score and participants' characteristics [/fig_ref]. τ V O 2 also showed no significant associations with participants' characteristics (p > 0.05, [fig_ref] Table 3: Associations between the time constant of oxygen uptake and participants' characteristics [/fig_ref]. Nevertheless, we additionally performed multiple regression analyses to confirm whether the association between the FSS score and τ V O 2 remained significant, even when adjusting for the logically confounding variables such as age, sex, type of stroke, time since stroke, presence of depressive symptoms, and Functional Independence Measure motor score [bib_ref] Poststroke fatigue: Emerging evidence and approaches to management: A scientific statement for..., Hinkle [/bib_ref] [bib_ref] Oxygen uptake kinetics, Poole [/bib_ref] [bib_ref] Reduced cardiorespiratory fitness after stroke: Biological consequences and exercise-induced adaptations, Billinger [/bib_ref]. These variables were entered into the regression model one by one. The regression models were significant when adjusting for sex or type of stroke, and the [fig_ref] Table 1: Associations between the Fatigue Severity Scale score and participants' characteristics [/fig_ref].
# Discussion
The primary objective of this study was to examine associations of the severity of post-stroke fatigue with V O 2 at peak exercise obtained during a symptom-limited graded exercise test and τ V O 2 at the onset of exercise measured during a submaximal constant-load exercise test in inpatients at a subacute rehabilitation ward. The results of post-stroke fatigue assessment in this study are similar to those of previous studies 29, [bib_ref] Hypertension and daytime hypotension found on ambulatory blood pressure is associated with..., Harbison [/bib_ref]. This study showed that a higher FSS score was associated with a longer τ V O 2 at the onset of exercise, but not with V O 2 at peak exercise. Our secondary objective was to identify the associations between post-stroke fatigue and cardiorespiratory variables potentially affecting V O 2 during exercise, such as the OUES, CO, and AVO 2 diff. Our findings showed that a higher FSS score was associated with a longer τCO at the onset of exercise, which suggests that the severity of post-stroke fatigue is related to delayed increases in V O 2 and CO at the onset of exercise. Additionally, the FSS score and τ V O 2 had no significant associations with age, sex, height, weight, body mass index, type of stroke, side of motor paresis, time since stroke, presence of hypertension and diabetes mellitus, MMSE score, presence of depressive symptoms, Stroke Impairment Assessment Set motor score, and Functional Independence Measure score in the motor and cognition items. Moreover, the association between the FSS score and τ V O 2 remained significant, even when we adjusted for sex or type of stroke. Therefore, the confounding effects of the participants' characteristics on the association between the FSS score and τ V O 2 appear to be limited. Individuals with post-stroke fatigue lack the energy necessary to perform activities, are more easily tired due to activity, experience unpredictable and unexplainable feelings of fatigue, and have increased stress sensitivity and an increased need for longer sleep durations, naps, or rest 31 . Thus, it is plausible that post-stroke fatigue might be associated with decreased cardiorespiratory fitness and reduced physical activity [bib_ref] Fatigue after stroke: A systematic review of associations with impaired physical fitness, Duncan [/bib_ref] [bib_ref] Physical activity and exercise after stroke: Review of multiple meaningful benefits, Saunders [/bib_ref]. However, no statistically significant correlation between the FSS score and V O 2 at peak exercise has been shown in individuals with chronic stroke [bib_ref] Exertion fatigue and chronic fatigue are two distinct constructs in people post-stroke, Tseng [/bib_ref]. To our knowledge, this study is the first to investigate association between fatigue at an early post-stroke stage and cardiorespiratory fitness. Our finding that the association between the FSS score and V O 2 at peak exercise was not significant was consistent with that of a previous study involving patients with chronic stroke [bib_ref] Exertion fatigue and chronic fatigue are two distinct constructs in people post-stroke, Tseng [/bib_ref] , whereas the association between the FSS score and τ V O 2 measured during the submaximal constantload exercise test was significant. We observed that the mean coefficient of determination of cardiorespiratory kinetics was > 0.85, which indicated that the fitting procedures were acceptable [bib_ref] Reproducibility of onset and recovery oxygen uptake kinetics in moderately impaired patients..., Kemps [/bib_ref]. The mean value of τ V O 2 in this study was similar to the value obtained in a previous study that found that individuals with stroke had a longer τ V O 2 than age-, sex-, and activity-matched healthy adults [bib_ref] Cardiac reserve and pulmonary gas exchange kinetics in patients with stroke, Tomczak [/bib_ref]. In healthy adults, the acceleration of τ V O 2 has been reported to occur in the early period of endurance training, and V O 2 at peak exercise subsequently increases [bib_ref] Progressive effect of endurance training on VO 2 kinetics at the onset..., Phillips [/bib_ref] [bib_ref] Early effects of exercise training on on-and off-kinetics in 50-year-old subjects, Fukuoka [/bib_ref]. Additionally, a previous study reported that τ V O 2 was shorter in a recreationally active group than in an inactive group; however, there was no significant difference in V O 2 at peak exercise between the two groups [bib_ref] Fitness level and not aging per se, determines the oxygen uptake kinetics..., George [/bib_ref]. Based on these findings, τ V O 2 may be more sensitive to changes in the levels of physical activity compared with V O 2 at www.nature.com/scientificreports/ peak exercise. Therefore, the associations of the FSS score with τ V O 2 and V O 2 at peak exercise observed in this study suggest that τ V O 2 more sensitively reflects lower levels of physical activity in individuals with post-stroke fatigue than V O 2 at peak exercise. We found that the mean ratio of τCO to τ V O 2 was > 1.00, indicating that oxygen delivery did not exceed the metabolic demand during exercise onset and that V O 2 kinetics at exercise onset were limited owing to a delayed increase in CO [bib_ref] Are oxygen uptake kinetics in chronic heart failure limited by oxygen delivery..., Kemps [/bib_ref]. However, it is unclear why participants with severe post-stroke fatigue showed a delayed increase in CO at the onset of exercise. A prompt increase in CO at the onset of exercise is compatible with the notion of immediate vagal withdrawal. Capelli et al. [bib_ref] Oxygen deficits and oxygen delivery kinetics during submaximal intensity exercise in humans..., Capelli [/bib_ref] reported that the increase in CO at the onset of exercise slowed after than before prolonged bed rest in healthy adults, because of a decrease in vagal activity at rest and elimination of vagal withdrawal during exercise. In addition, reduced cardiac mass and function, plasma volume, and venous return after prolonged bed rest 36-39 may also negatively affect the increase in CO during exercise onset. Furthermore, given that post-stroke fatigue is associated with low physical activity [bib_ref] Factors associated with post-stroke physical activity: A systematic review and meta-analysis, Thilarajah [/bib_ref] , an inactive lifestyle may act as a confounder in relation to severe fatigue and delayed increases in V O 2 and CO at the onset of exercise. An assessment of physical activity is needed in future studies to determine the reasons for these associations of the FSS score with τ V O 2 and τCO observed in this study.
Our findings suggest that individuals with severe post-stroke fatigue need to improve τ V O 2 at the onset of exercise. Previous studies have shown that aerobic exercise training was effective in the improvement of τ V O 2 in older individuals [bib_ref] Changes in pulmonary oxygen uptake and muscle deoxygenation kinetics during cycling exercise..., Hamasaki [/bib_ref] [bib_ref] Speeding of VO 2 kinetics in response to endurance-training in older and..., Murias [/bib_ref] [bib_ref] Speeding of VO 2 kinetics with endurance training in old and young..., Murias [/bib_ref]. A randomised controlled trial reported that a combination of cognitive-behavioural therapy and graded activity training was more effective than cognitive-behavioural therapy alone in treating post-stroke fatigue [bib_ref] Cognitive and graded activity training can alleviate persistent fatigue after stroke: A..., Zedlitz [/bib_ref]. Although post-stroke fatigue has a negative effect on recovery of activities of daily living 3,4 , one systematic review reported that aerobic exercise can improve functional ability in individuals with stroke 8 . Furthermore, in the subacute phase of stroke recovery, several studies have shown the effectiveness of exercise in improving health outcomes, including cardiovascular, functional, and mobility outcomes, after stroke [bib_ref] Effects of cardiovascular exercise early after stroke: Systematic review and meta-analysis, Stoller [/bib_ref] [bib_ref] Aerobic exercise in subacute stroke improves cardiovascular health and physical performance, Billinger [/bib_ref]. Therefore, rehabilitative exercise programmes may be beneficial for individuals with post-stroke fatigue.
The use of exercise testing for clinical assessment and exercise prescription is limited in stroke rehabilitation settings [bib_ref] Barriers and facilitators to aerobic exercise implementation in stroke rehabilitation: A scoping..., Moncion [/bib_ref] [bib_ref] Aerobic exercise prescription in stroke rehabilitation: A web-based survey of US physical..., Boyne [/bib_ref] , thus limiting the clinical applicability of our findings. A lack of exercise equipment, time, space, and support staff have also been reported as barriers to exercise testing [bib_ref] Barriers and facilitators to aerobic exercise implementation in stroke rehabilitation: A scoping..., Moncion [/bib_ref]. Moreover, cardiac, cognitive, functional, and physical impairments in individuals with stroke may make it difficult to perform exercise testing safely [bib_ref] Barriers and facilitators to aerobic exercise implementation in stroke rehabilitation: A scoping..., Moncion [/bib_ref]. More specific clinical guidelines for post-stroke exercise testing, educational training associated with exercise testing, and greater collaboration between stroke and cardiac rehabilitation teams could help to implement exercise testing more effectively in stroke rehabilitation settings [bib_ref] Barriers and facilitators to aerobic exercise implementation in stroke rehabilitation: A scoping..., Moncion [/bib_ref] [bib_ref] Aerobic exercise prescription in stroke rehabilitation: A web-based survey of US physical..., Boyne [/bib_ref].
This study had some limitations. First, the sample size was relatively small because we only calculated the sample size required for a bivariate correlation analysis. Post-stroke fatigue has been found to be associated with older age, female sex, depressive symptoms, and functional disability 1 . Changes in the brain after a stroke may also affect post-stroke fatigue and cardiorespiratory control during exercise [bib_ref] Poststroke fatigue: Emerging evidence and approaches to management: A scientific statement for..., Hinkle [/bib_ref] [bib_ref] Reduced cardiorespiratory fitness after stroke: Biological consequences and exercise-induced adaptations, Billinger [/bib_ref]. Even though our participants were recruited from a subacute rehabilitation ward, five participants were in the chronic phase of stroke recovery. In addition, 12 participants with ischaemic stroke and 11 with haemorrhagic stroke were included in this study. Functional recovery differs between individuals with ischaemic and haemorrhagic strokes. Stroke severity is higher in haemorrhagic stroke than in ischaemic stroke, while individuals with haemorrhagic stroke have been shown to have a higher therapeutic response to rehabilitation than those with ischaemic stroke [bib_ref] Hemorrhagic and ischemic strokes compared: Stroke severity, mortality, and risk factors, Andersen [/bib_ref] [bib_ref] Functional outcome of ischemic and hemorrhagic stroke patients after inpatient rehabilitation: A..., Paolucci [/bib_ref]. www.nature.com/scientificreports/ However, our study findings indicated that these variables were not associated with the FSS score and τ V O 2 . Further studies using multivariate analysis and a sufficiently large sample size are warranted to confirm the robustness of our findings. Second, most participants were in the subacute phase of stroke recovery. Because fatigue at a later stage after stroke may be associated more with psychological and behavioural factors than with biological factors 17 , generalising our findings to individuals in the later stage after stroke should be made with caution.
Third, many individuals with stroke (n = 424) were excluded from the study. Many of them were excluded due to being > 80 years of age, having an MMSE score of ≤ 24 points 49 , and/or having unstable medical conditions, as shown in [fig_ref] Figure 1: Flow diagram of study participants [/fig_ref]. This may limit the generalisability of our findings in relation to individuals with these conditions. Finally, because this study used a cross-sectional observational design, the cardiorespiratory variables associated with temporal changes in post-stroke fatigue could not be examined. Thus, further longitudinal studies are needed to investigate the temporal association between post-stroke fatigue and cardiorespiratory fitness variables.
In summary, a higher FSS score statistically significantly correlated with longer τ V O 2 at the onset of exercise measured during a submaximal constant-load exercise test, but not with V O 2 at peak exercise obtained during a symptom-limited graded exercise test. In addition, a higher FSS score was associated with a longer τCO at the onset of exercise. These results suggest that severe post-stroke fatigue is related to delayed increases in V O 2 and CO at the onset of exercise. Collectively, our findings can contribute to the development of an appropriate rehabilitation programme for individuals with post-stroke fatigue.
# Methods
Study design. This cross-sectional study's protocol was approved by the appropriate ethics committees of Tokyo Bay Rehabilitation Hospital (approval number, 172-2) and Shinshu University (approval number: 3813), and conducted according to the Declaration of Helsinki of 1964 as revised in 2013. All participants provided written informed consent before enrolment.
## Participants. participants were recruited from a subacute rehabilitation ward between november 2017 and
March 2020. Inclusion criteria comprised the following: age 40-80 years, within 180 days of the initial stroke, an ability to maintain a target cadence of 50 rpm during exercise, and an MMSE score of > 24 [bib_ref] Mini-mental state". A practical method for grading the cognitive state of patients..., Folstein [/bib_ref]. Exclusion criteria comprised the following: limited range of motion and/or pain that could affect the exercise test; unstable medical conditions, such as unstable angina, uncontrolled hypertension, or tachycardia; the use of beta-blocker medication; and any comorbid neurological disorders. Demographic and clinical data, such as age and type of stroke, were obtained from patient medical records.
Procedure. Data collection was completed within a week from the start of the procedure. On day 1, we assessed post-stroke fatigue, depressive symptoms, and functional outcomes. On day 2, participants performed a symptom-limited graded exercise test to determine the workload for their submaximal exercise test. On day 3, three repetitions of the submaximal constant-load exercise test were performed at 80% of the workload corresponding to the ventilatory threshold to assess the kinetics of cardiorespiratory variables [bib_ref] Are the parameters of VO 2 , heart rate and muscle deoxygenation..., Spencer [/bib_ref].
Assessments of post-stroke fatigue, depressive symptoms, and functional outcomes. Post-stroke fatigue was assessed using the 9-item FSS with each item rated on a 7-point Likert scale that ranged from 1 to 7 (1, strongly disagree; 7, strongly agree) [bib_ref] The fatigue severity scale. Application to patients with multiple sclerosis and systemic..., Krupp [/bib_ref]. The FSS score was calculated as the sum of the scores of the 9 items. A high score indicated a greater effect of fatigue on daily activities.
The 15-item Geriatric Depression Scale (GDS) 52 was used to assess depressive symptoms. A GDS score of ≥ 5 denoted the presence of depressive symptoms. Motor function and independence in performing daily activities were assessed as functional outcomes. The total Stroke Impairment Assessment Set motor function score was measured to assess motor impairments in the paretic upper and lower extremities [bib_ref] Stroke Impairment Assessment Set (SIAS) A new evaluation instrument for stroke patients, Chino [/bib_ref]. The Functional Independence Measure score was used to evaluate the degree of independence in activities of daily living 54 .
Exercise testing. Participants were instructed to refrain from food consumption for 3 h, caffeine intake for at least 6 h, and vigorous physical activity for 24 h prior to undertaking the symptom-limited graded and submaximal constant-load exercise tests [bib_ref] Systematic review of cardiopulmonary exercise testing post stroke: Are we adhering to..., Van De Port [/bib_ref]. The tests were performed on a recumbent cycle ergometer (Strength Ergo 240; Mitsubishi Electric Engineering Co., Ltd., Tokyo, Japan) that could be precisely load-controlled (coefficient of variation, 5%) over a wide range of pedalling resistance (0-400 W). Participants were instructed to maintain a target cadence of 50 rpm in all exercise phases [bib_ref] Systematic review of cardiopulmonary exercise testing post stroke: Are we adhering to..., Van De Port [/bib_ref]. Expired gas was measured on a breath-by-breath basis during the exercise test using an expired gas analyser (Aerosonic AT-1100; ANIMA Corp., Tokyo, Japan). Before data collection, the analyser was calibrated using gas mixtures with accurately known concentrations of oxygen and carbon dioxide. CO was measured on a beat-by-beat basis using a non-invasive impedance cardiography device (Task Force Monitor model 3040i; CN Systems Medizintechnik GmbH., Graz, Austria), as previously described [bib_ref] Non-invasive beat-to-beat cardiac output monitoring by an improved method of transthoracic bioimpedance..., Fortin [/bib_ref]. Three short band electrodes, one on the neck and two below the thorax, were placed on the participants. Stroke volume was calculated using the following equation:
where V th is the electrical participating thoracic volume, LVET is the left ventricular ejection time, (dZ/dt) max is the maximal rate of decrease in impedance for a given heartbeat, and Z 0 is the base impedance. CO was calculated as the product of stroke volume and heart rate. Impedance cardiography is a valid and reliable method for measuring cardiac haemodynamics at rest and during exercise [bib_ref] Non-invasive beat-to-beat cardiac output monitoring by an improved method of transthoracic bioimpedance..., Fortin [/bib_ref] www.nature.com/scientificreports/ variables were interpolated to 1 s intervals, time-aligned, and averaged into 5 s bins to derive the AVO 2 diff on a second-by-second basis [bib_ref] Cardiac reserve and pulmonary gas exchange kinetics in patients with stroke, Tomczak [/bib_ref] [bib_ref] Cardiorespiratory factors related to the increase in oxygen consumption during exercise in..., Oyake [/bib_ref] calculated based on Fick's equation [bib_ref] Clinician's Guide to cardiopulmonary exercise testing in adults: A scientific statement from..., Balady [/bib_ref] , as follows:
All participants rested for 5 min before taking the tests. The symptom-limited graded exercise test started with a warm-up at 0 W for 3 min followed by a 10 W increment every minute [bib_ref] Systematic review of cardiopulmonary exercise testing post stroke: Are we adhering to..., Van De Port [/bib_ref]. The test was terminated if the participant showed signs of angina, dyspnoea, inability to maintain a cycling cadence of > 40 rpm, hypertension (> 250 mmHg systolic or > 115 mmHg diastolic pressure), or a drop in systolic blood pressure of > 10 mmHg, despite an increase in workload [bib_ref] Cardiorespiratory factors related to the increase in oxygen consumption during exercise in..., Oyake [/bib_ref]. To identify whether the maximal effort was reached during the exercise test, at least one of the following criteria had to be met: a < 150 mL min −1 increase in V O 2 for > 1 min despite increased work rate, respiratory exchange ratio of > 1.10, or heart rate that was 85% of the age-predicted maximal heart rate calculated as 220 minus age 12 . V O 2 , CO, AVO 2 diff, V E, and the respiratory exchange ratio at peak exercise were defined as the average values obtained during the last 30 s of the exercise test [bib_ref] Systematic review of cardiopulmonary exercise testing post stroke: Are we adhering to..., Van De Port [/bib_ref]. In addition, the OUES was determined through calculating the slope of the regression line between V O 2 and the log transformation of V E during the whole exercise period using the following equation:
where the constant a is the OUES 58 . A low OUES represents a high amount of ventilation required in response to a given oxygen uptake, which indicates ventilatory inefficiency during exercise.
The ventilatory threshold was determined using a combination of the following criteria: the point where the ventilatory equivalent of oxygen reaches its minimum or starts to increase, without an increase in the ventilatory equivalent of carbon dioxide; the point at which the end-tidal oxygen fraction reaches a minimum or starts to increase, without a decline in the end-tidal carbon dioxide fraction; and the point of deflection of carbon dioxide output versus V O 2 (the V-slope method), as previously described [bib_ref] Cardiorespiratory factors related to the increase in oxygen consumption during exercise in..., Oyake [/bib_ref] [bib_ref] Methodological approach to the first and second lactate threshold in incremental cardiopulmonary..., Binder [/bib_ref]. V O 2 , CO, AVO 2 diff, V E, and the respiratory exchange ratio at the ventilatory threshold were obtained.
The submaximal constant-load exercise test started with resting on the cycle ergometer for 3 min, followed by performing the exercise at 80% of the workload corresponding to the ventilatory threshold for 6 min [bib_ref] Fitness level and not aging per se, determines the oxygen uptake kinetics..., George [/bib_ref] [bib_ref] Are oxygen uptake kinetics in chronic heart failure limited by oxygen delivery..., Kemps [/bib_ref]. The protocol was repeated three times, with a rest between each repetition. Data concerning the kinetics of the V O 2 , CO, AVO 2 diff, and V E at exercise onset were obtained through averaging the three repeats. Additionally, before modelling, we eliminated the first 20 s of data after exercise onset because the increase in V O 2 during this period reflects merely an increase in the pulmonary blood flow rather than changes in tissue gas exchange [bib_ref] Oxygen uptake kinetics, Poole [/bib_ref]. To calculate the time constants of V O 2 , CO, AVO 2 diff, and V E at exercise onset, a non-linear least squares regression procedure (GraphPad Prism version 7.00 for Windows; GraphPad Software, CA, USA) was applied to the onset phase, using the following equation:
where Y (t) represents V O 2 , CO, AVO 2 diff, or V E at a given time (t); TD is the time delay; and τ is the time constant. Y baseline and Y steady-state are the average values of Y during the last minute of the resting period and exercise, respectively. Fit quality was determined using a coefficient of determination. The fitting procedure was considered acceptable if the coefficient of determination was > 0.85 [bib_ref] Reproducibility of onset and recovery oxygen uptake kinetics in moderately impaired patients..., Kemps [/bib_ref]. In addition, the ratio of τCO to τ V O 2 > 1.00 indicated a slow increase in CO relative to V O 2 at exercise onset.
Statistical analyses. The sample size for examining the correlations between the FSS score and cardiorespiratory variables was computed at alpha = 0.05 and power = 0.80 using G Power software version 3.1.9.2 (Heinrich Heine University, Dusseldorf, Germany). Tseng et al. [bib_ref] Exertion fatigue and chronic fatigue are two distinct constructs in people post-stroke, Tseng [/bib_ref] reported that the FSS score significantly correlated with the GDS score (r = 0.639) but not with V O 2 at peak exercise (r = − 0.125) in 21 people with chronic stroke. Therefore, we calculated the required sample size to detect only a large effect size for correlation (0.50). Consequently, a minimum sample size of 26 participants was required. Assuming that 10% of the participants could be excluded, we aimed to recruit 30 participants.
The results are presented as medians (interquartile ranges) or means ± SDs. We examined the associations between the FSS score and cardiorespiratory variables using the Spearman's rank correlation coefficient. To identify the potential confounding variables, we also determined the associations of participant's characteristics with the FSS score and cardiorespiratory fitness variables that significantly correlated with the FSS score using the Pearson's product-moment correlation coefficient, Spearman's rank correlation coefficient, and unpaired t-test based on variable types. Additionally, we performed multiple regression analysis with forced entry to confirm whether the associations between the FSS score and cardiorespiratory fitness variables observed in the correlation analysis remained significant, even when adjusting for potential confounding variables. Statistical analyses were performed using Statistical Package for the Social Sciences software version 27.0 (International Business Machines Corp., NY, USA). Poisson (p) values < 0.05 were considered statistically significant.
## Data availability
The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.
[fig] Figure 1: Flow diagram of study participants. MMSE Mini-Mental State Examination. [/fig]
[fig] Figure 2: Correlations of the Fatigue Severity Scale score with (a) oxygen uptake at peak exercise and (b) the time constant of oxygen uptake kinetics. [/fig]
[fig] Figure 3: Correlation between the Fatigue Severity Scale score and the time constant of cardiac output kinetics. Scientific Reports | (2021) 11:12780 | https://doi.org/10.1038/s41598-021-92127-w [/fig]
[fig] 2: = alogV E + b, [/fig]
[fig] 4: Y (t) = Y baseline + Y steady−state − Y baseline × 1 − exp −(t − TD)/ τ , [/fig]
[table] Table 1: Associations between the Fatigue Severity Scale score and participants' characteristics. Values are presented as mean ± standard deviation, median (interquartile range), or number. rho Spearman's rank correlation coefficient, FSS Fatigue Severity Scale, GDS Geriatric Depression Scale, NA not available. [/table]
[table] Table 3: Associations between the time constant of oxygen uptake and participants' characteristics. τVO 2 time constant of oxygen uptake kinetics, r Pearson's product-moment correlation coefficient, rho Spearman's rank correlation coefficient, GDS Geriatric Depression Scale, NA not available. [/table]
|
Challenges in detecting and quantifying intron retention from next generation sequencing data
a b s t r a c tIntron retention (IR) occurs when an intron is transcribed into pre-mRNA and remains in the final mRNA. An increasing body of literature has demonstrated a major role for IR in numerous biological functions and in disease. Here we give an overview of the different computational approaches for detecting IR events from sequencing data. We show that these are based on different biological and computational assumptions that may lead to dramatically different results. We describe the various approaches for mitigating errors in detecting intron retention and for discovering IR signatures between different conditions.
# Introduction
Amongst the three major types of alternative splicing (AS) that include exon skipping/inclusion, alternative 5 0 and 3 0 splice-site selection and intron retention (IR), the latter has until recently been regarded as an oddity in mammals; IR was often added to the list for no other intent than to be exhaustive. However, recent discoveries about the role IR can play in fine-tuning gene expression [bib_ref] Orchestrated intron retention regulates normal granulocyte differentiation, Wong [/bib_ref] [bib_ref] A dynamic intron retention program in the mammalian megakaryocyte and erythrocyte lineages, Edwards [/bib_ref] [bib_ref] Widespread intron retention in mammals functionally tunes transcriptomes, Braunschweig [/bib_ref] as well as the observation of characteristic IR patterns [bib_ref] Gene-specific intron retention serves as molecular signature that distinguishes melanoma from non-melanoma..., Giannopoulou [/bib_ref] [bib_ref] Coordinated splicing of regulatory detained introns within oncogenic transcripts creates an exploitable..., Braun [/bib_ref] [bib_ref] SMN deficiency in severe models of spinal muscular atrophy causes widespread intron..., Jangi [/bib_ref] [bib_ref] Intronretained transcripts of the spinal muscular atrophy genes, SMN1 and SMN2, Harahap [/bib_ref] [bib_ref] Increased intron retention is a post-transcriptional signature associated with progressive aging and..., Adusumalli [/bib_ref] highlight the value of investigating IR in transcriptomic studies. Numerous reports have demonstrated a regulatory role for IR in hematopoiesis [bib_ref] Orchestrated intron retention regulates normal granulocyte differentiation, Wong [/bib_ref] [bib_ref] A dynamic intron retention program in the mammalian megakaryocyte and erythrocyte lineages, Edwards [/bib_ref] , neuronal differentiation [bib_ref] Targeted intron retention and excision for rapid gene regulation in response to..., Mauger [/bib_ref] , germ cell differentiation [bib_ref] Functional interaction between U1snRNP and Sam68 insures proper 3 0 end Pre-mRNA..., Naro [/bib_ref] and CD4+ T cell activation [bib_ref] Global intron retention mediated gene regulation during CD4+ T cell activation, Ni [/bib_ref] amongst others. In addition, a recent analysis of 1812 cancer patient samples showed that over 18% of splicing-associated single nucleotide variants caused IR and most of these events affected tumor suppressor genes [bib_ref] Detained introns are a novel, widespread class of post-transcriptionally spliced introns, Boutz [/bib_ref]. Finally, the analysis of 2573 samples showed that IR occurs in all tissues analyzed and can affect over 80% of all coding genes. Measuring IR can help decipher gene-level variations and inter-connections between transcriptional loads, structural variations and phenotypes [bib_ref] Profiling RNA-Seq at multiple resolutions markedly increases the number of causal eQTLs..., Odhams [/bib_ref].
IR has been to demonstrated to downregulate gene expression in numerous systems by triggering the nonsense mediated mRNA decay (NMD) pathway. NMD recognizes transcripts with premature stop codons (PTC) that could potentially generate C-terminal truncated proteins and degrades them. This surveillance mechanism can thus rapidly degrade IR transcripts if they harbor a PTC. Given that introns are much longer than exons and under less selective pressure to conserve open reading frames, the probability that an IR event harbors a PTC is high and IR transcripts are thus good candidates for degradation via NMD. Initially, transcriptomic and bioinformatic analyses of NMD concluded that it was not coupled with mRNA splicing and that most PTC containing transcripts do not have major functional roles [bib_ref] Quantitative microarray profiling provides evidence against widespread coupling of alternative splicing with..., Pan [/bib_ref] , thus relegating NMD to the role of scavenger [bib_ref] The functional consequences of intron retention: alternative splicing coupled to NMD as..., Ge [/bib_ref]. The same team however recently revised their view on the functional importance of NMD [bib_ref] Widespread intron retention in mammals functionally tunes transcriptomes, Braunschweig [/bib_ref] and the current consensus is that it couples with IR (and other forms of AS) to regulate gene expression in numerous systems [bib_ref] Orchestrated intron retention regulates normal granulocyte differentiation, Wong [/bib_ref] [bib_ref] Widespread intron retention in mammals functionally tunes transcriptomes, Braunschweig [/bib_ref] [bib_ref] Intron retention as a component of regulated gene expression programs, Jacob [/bib_ref] [bib_ref] Genome-wide characterization of the routes to pluripotency, Hussein [/bib_ref] [bib_ref] Intron retention in mRNA: no longer nonsense, Wong [/bib_ref]. The importance of NMD is further underscored by the fact that deletion of its core components result in embryonic lethality [bib_ref] Rent1, a trans-effector of nonsense-mediated mRNA decay, is essential for mammalian embryonic..., Medghalchi [/bib_ref].
RNA-seq data is well suited for resolving local exon connectivity because sequencing reads are sufficiently long to cover exon-exon junctions. It is also well suited for measuring global gene expression because the high number of reads that map to genes generally enables the use of powerful statistical models. Detecting and measuring IR with RNA-seq is more complex. The technical biases that are known to distort gene expression levels (eg: GC content, amplification biases) and other types of splicing, also affect IR measurement. In addition, measurement of intronic expression is challenged by numerous factors. Within introns, highly expressed features such as small nucleolar RNAs, microRNAs or unannotated exons may erroneously inflate count-based measures of intronic expression. Conversely, low complexity regions, common in introns, prevent unique mapping of reads. Because retained introns are generally expressed at a fraction of their flanking exons, uncorrected biases can massively disrupt IR estimation.
Transcriptome-wide evaluation of IR by computational means is still a budding field of investigation. In many studies, IR was assessed via custom and briefly detailed procedures, which is probably due, in part, to the fact that few dedicated and comprehensive tools have been published so far. In the following, we give a survey of the technical biases that confound IR detection and available computational methods to tackle IR screening. We emphasize three crucial steps which are: the preparation and quality control of the sequencing data and the reference transcriptome; generat-ing metrics that reflect the biological signal of IR transcripts and using a model to discover condition-specific IR events.
## Ir detection
## Filtering sequencing data
Unlike sequencing reads that map to exon-exon junctions, reads that map to introns can originate from DNA contamination caused by ineffective DNase treatment or from pre-mature mRNA. One method to detect DNA contamination is to measure reads across splice sites and check that the majority of introns display high splicing efficiency (above 90%) [bib_ref] Nascentseq indicates widespread cotranscriptional pre-mRNA splicing in Drosophila, Khodor [/bib_ref]. Another approach, implemented inis to verify that the ratio of the number of reads that map to intergenic regions to the number that maps to coding regions is less than 10%. Another source of bias is that intronic reads may originate from nascent and pre-mature RNAs [bib_ref] Challenges in defining the role of intron retention in normal biology and..., Vanichkina [/bib_ref]. So as to lessen signals due to unprocessed transcripts and overlapping antisense transcripts, it is recommended to use Poly-A enriched RNA-seq (or cytoplasmic fractionation) and strand-specific protocols.
Regarding library size, IR occurs at relatively low frequency in mammals, and introns tend to be substantially longer than exons. Determining an optimal library size obviously depends on many experiment-specific factors and on the IR effect size considered as biologically significant, 35 millions mapped reads for a oneversus-one experiment was suggested as an optimum for detection of differential intron usage (based on a resampling approach, [bib_ref] IntEREst: intron-exon retention estimator, Oghabian [/bib_ref]. In order to bypass intronic alignment biases, it has been suggested to consider only splice junction reads or similarly to focus only on a window centered on splice sites. Nonetheless, those junction-only analyses are likely to be more affected by splicing variations in flanking exons and lead to even more unstable estimates. Accordingly, previous studies pointed out that they require higher sequencing-depth (at least 70 million reads per sample, ideally more than 150 million reads) [bib_ref] Challenges in defining the role of intron retention in normal biology and..., Vanichkina [/bib_ref].
## Defining reference intronic sequences
Sequencing reads that map to intronic intervals may originate from several different sources such as overlapping genes. These confound measurements of the magnitude of true IR. It is therefore crucial to correctly define the intronic intervals that will be used to measure IR. Here, two main approaches have been adopted (cf: [fig_ref] Figure 1: Fig [/fig_ref] , each calling for precautions for interpretation and specific processing to avoid false positive detection.
## The all-introns view
A first possibility is to analyze all intronic intervals present in at least one annotated transcript model. Although this allows to screen the largest set of candidates, this comes at the expense of having to deal with peaks of intronic alignments caused by expressed alternative exons and redundant IR calls due to overlapping introns.
## The measurable introns view
Measurable (or independant) introns are (parts of) introns that do not overlap with any annotated exon [bib_ref] IntEREst: intron-exon retention estimator, Oghabian [/bib_ref] [bib_ref] Detecting differential usage of exons from RNAseq data, Anders [/bib_ref]. They are obtained by subtracting merged exons from genes. This comes with the advantage to simplify the analysis of introns flanked by exons with known alternative donor sites, but ignores introns fully overlapped by annotated exons.
## Overcoming sequencing and alignment artifacts
Even when appropriate sequencing protocols have been used, several sources of confusion remain that can only be overcome with computational means. These are detailed below and in .
## Overlapping features
First, the sequencing protocol may capture molecules overlapping or mapping within introns, such as small nucleolar RNAs, micro RNAs, unannotated exons or alternative 5 0 and 3 0 splicesites. They form characteristic peaks of reads within intronic regions that may induce false IR detections and result in inaccurate quantification if not properly identified and filtered.
## Repeated regions
Relatively to exons, introns are long and poorly conserved, often contain low mappability regions (eg: duplicated regions like transposons [bib_ref] An intron with a constitutive transport element is retained in a Tap..., Li [/bib_ref] [bib_ref] Intron retention in viruses and cellular genes: detention, border controls and passports, Rekosh [/bib_ref] [bib_ref] An NXF1 mRNA with a retained intron is expressed in hippocampal and..., Li [/bib_ref] or repeated regions such as microsatellites [bib_ref] Intron retention induced by microsatellite expansions as a disease biomarker, Sznajder [/bib_ref] which impair correct IR level estimations.
## Low coverage of flanking exons
Thirdly, sequencing and alignment artifacts may occur in flanking exons. For example, GC rich exons are under-covered and very small exons are more difficult to map. This may perturb or inflate IR measures, especially those that only measure reads directly surrounding the intron. They may also lead to missed IR events as the junctions are weakly supported.
## 3 0 coverage bias
PolyA-enriched RNA-seq data usually display a marked 3 0 coverage bias, so that most 3 0 introns are likely to be more covered and thus more easily captured than 5 0 ones. This should be kept in mind for any inference regarding any positional bias of intron retention.
In practice, the prevailing strategy for classifying introns relies essentially on user-defined thresholds. Here, we summarize the different strategies used by IR detection tools, and indicate the parameter values suggested by their respective authors when they exist [fig_ref] Figure 3: Standard implementation of computational detection of IR events [/fig_ref] and [fig_ref] Table 1: Computational tools available to perform IR detection and their main features [/fig_ref].
## Implemented strategies to pinpoint reliable ir events
Although numerous computational methods have been developed to estimate splicing efficiency and to model sequencing errors that may affect their estimation, we have decided to list here those approaches that specifically cater to the difficulties of detecting IR. Defining intronic intervals to be analyzed. Comprehensiveness of transcript annotation and the selection of reference intronic sequences have a major impact on IR detection. In the example, we consider a gene having three possible isoforms (A, B and C). Exons are represented as plain rectangles and introns as thick black lines. If only Isoforms B and C were annotated, the starred interval (*) would not be defined as an intron and most likely not detected as retained. Colored boxes indicate whether the annotated introns match the ''all introns" or ''independent/measurable intron" criteria used by current algorithms. . Potential sources of bias and confusion: a very unfortunate gene. Only intron 3 is retained in this example. In intron 1: expression of an overlapping feature causes a peak in alignments, which can artificially inflate the estimation of IR. Intronic alignments in intron 2 originate from an unanottated exon. Intron 3 is retained but it's detection is hampered by multiple biases. First, the presence of a low mappability region (repeated A sequence in red) would result either in a gap or in high uncertainty in read alignments in that region. Secondly, high GC content in the 5 0 exon explains the lack of exon-exon junctions and 3 0 exon-intron reads and may affect filtering and IR metrics based on them. Thirdly, due to its long length, it tends to be more sparsely covered. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Keep Me Around (KMA)uses the measurable introns approach. Transcripts are quantified using eXpress [bib_ref] Streaming fragment assignment for real-time analysis of sequencing experiments, Roberts [/bib_ref] or Kallisto [bib_ref] Near-optimal probabilistic RNA-seq quantification, Bray [/bib_ref] [bib_ref] Intron retention is a source of neoepitopes in cancer, Smart [/bib_ref] and a PSI value is computed to evaluate IR levels (cf: Quantifying IR levels).
Spurious intronic signals are spotted by finding the longest alignment gap in an intron and calculating the probability of observing such a long gap given the intron's expression and if the distribution were uniform.
The authors selected introns with at least three uniquely mapped reads, cumulated TPM values for non-IR transcripts greater than 1 and introns with zero-coverage regions longer than 20% of the intron length [bib_ref] Distributions of exons and introns in the human genome, Sakharkar [/bib_ref].
IRFindermakes the choice to screen all introns derived from the annotation. Introns that overlap with any known exon or RNA molecule are marked in the output.
A procedure is implemented to identify low mappability regions and exclude them and their reads from the subsequent calculation. Potential other artifacts are handled by discarding bases with outlier read depth value compared to the average intron depth.
The IRratio and several other complementary metrics are then computed to evaluate support for IR.
Suggested parameters values for IRFinder are: IRratio > 0.1, and at least 3 reads supporting intron exclusion on both sides.
In iRead, reference introns are provided by the user. Suspect cases are likened to non-uniform coverage. Read coverage uniformity is quantified by the Shannon's entropy of read distribution along the intron and low entropy is associated to a nonuniform read coverage in the intron.
By default, iRead will select intron having FPKM > 3, at least one exon-intron junction read and the normalized entropy-score > 0.9.
IntEREst [bib_ref] IntEREst: intron-exon retention estimator, Oghabian [/bib_ref] computes FPKM and PSI values for independent introns. Optionally, low mappability regions can be excluded from the calculations. No specific guidelines are provided to select IR events, but data are formatted for some methods for performing differential analysis.
It is worth emphasizing that each approach makes use of prefixed threshold values for all intronic regions. Most of these values are defined according to the coverage profile expected for (wellbehaved) short length medium-coverage introns, and are maybe the most straightforward way to guard from the bulk of artefactual detections. However, introns form a highly heterogeneous set of regions, hugely differing in length, inner and flanking coverage and sequence feature. For example, on genes having sparse coverage, chances to observe counts on a very small pre-specified area of the genome are quite low. Therefore, filters on the number of junction reads are likely to exclude most of their introns from further analysis. Very long introns are especially problematic and in practice these regions have little chance to be covered at their fullextent, and well generally fail on the hard cutoffs set by these algorithms. It is thus clear that no universal threshold can be convenient in all cases, and that any rigid thresholding is likely to introduce a severe selection bias. We thus argue that a sensible choice for the various parameters must be intron-specific and encourage the development and use of models that account explicitly for sequence features and coverage variations.
## Quantifying ir levels
Though essential, devising a computable and robust metric that reflects ''splicing efficiency" or oppositely the level of IR can be difficult. Three types of alignments should be taken into account [bib_ref] Intron retention as a component of regulated gene expression programs, Jacob [/bib_ref]. Intronic reads and reads that span the flanking exons are informative of the level of IR. In addition, all the remaining alignments, can indicate how reliable the sequencing data is and what degree of confidence we may have in each IR event. The most commonly used ratios to quantify IR are the percentage spliced in and the IR-ratio, both described below.
## Percentage spliced-in
Alternative splicing event frequencies are commonly quantified by the percentage spliced-in (PSI) ratio [bib_ref] JUM is a computational method for comprehensive annotation-free analysis of alternative pre-mRNA..., Wang [/bib_ref]. An intronic version has been suggested [bib_ref] Genome-wide characterization of the routes to pluripotency, Hussein [/bib_ref] as the number of reads supporting the retention of the intron against the number of reads supporting its exclusion. In practice, a transcript-level quantification is performed using an annotation of IR-free isoforms augmented with independent introns (taken as dummy transcripts). The PSI for a given intron can be formulated as:
[formula] PSI ¼ IntronTPM IntronTPM þ P transcriptsTPM [/formula]
where the sum is performed across all annotated transcripts of the same gene not retaining an intron.
## Ir ratio
This metric is to reflect splicing efficiency as the portion of informative reads which come from a transcript retaining the intron, that is:
[formula] IRratio ¼ IntronicAbundance IntronicAbundance þ NormalSplicingAbundance [/formula]
where Intronic abundance is measured by the medianor average [bib_ref] Abiotic stresses modulate landscape of poplar transcriptome via alternative splicing, differential intron..., Filichkin [/bib_ref] intron depth. The abundance of normal splicing is taken as the number of reads spliced across the intron.
These ratios tend to show high fluctuations and their behavior is difficult to model. This may explain why, so far, no approach has been developed to estimate dispersions and confidence intervals. Importantly, this hinders the identification of robust and reproducible patterns based on their observed values. Although these metrics can be employed, as a proxy for splicing efficiency, to call manifest IR events, additional statistics are required to infer intra-and cross-sample variation levels. One of the major difficulties for quantifying splicing efficiency is due to the fact that the exons flanking an intron may connect not only to each other but with other exons from the same gene to form different isoforms. This hampers the estimation of the portion of reads to attribute to the transcripts in which the intron is spliced. The two measures presented above (PSI and IRratio) address this problem differently. So as to overcome global variations in gene coverage caused by alternative exon usage, the strategy behind the IRratio is to only make use of the junction-crossing reads that hit any one of the two exons flanking the intron. The maximum value between the left and right quantities is then taken as a means to mitigate against the existence of multiple isoforms that connect to the flanking exons. However, the number of junction reads tend to be highly dispersed with high coverage, and to take zero values when the coverage is low. This may incidentally affect estimation accuracy. On the other hand, by using information across the whole transcript to evaluate the gene coverage, the PSI estimator might be more resistant to these local variations. Nonetheless, it would be of interest to assess the quality of the PSI estimates on genes which undergo manifold alternative splicing events [bib_ref] A benchmark for RNA-seq quantification pipelines, Teng [/bib_ref] [bib_ref] Computational approaches for isoform detection and estimation: good and bad news, Angelini [/bib_ref].
## Cross-sample comparison
Inferring differences in IR between conditions necessitates a statistical framework to combine biological replicates, assess dispersion of IR level estimates and control the false positive rate. Moreover, sample read abundances need to be normalized to account for variations in library size [bib_ref] A comprehensive evaluation of normalization methods for Illumina highthroughput RNA sequencing data..., Dillies [/bib_ref]. Moreover, the coverage depth of an intron is correlated to its gene coverage; sample comparison thus necessitates strategies to control for differences in gene expression [bib_ref] Challenges in defining the role of intron retention in normal biology and..., Vanichkina [/bib_ref].
To our knowledge, currently four implemented frameworks fulfill these requirements (cf: . In regards of their statistical methodology, we split them into three families of approaches.
# Intron-bin count-based methods
The first two approaches re-use existing methods that were primarily devised either for gene expression [bib_ref] Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2, Love [/bib_ref] [bib_ref] edgeR: a Bioconductor package for differential expression analysis of digital gene expression..., Robinson [/bib_ref] or exon usage [bib_ref] Detecting differential usage of exons from RNAseq data, Anders [/bib_ref] [bib_ref] RNA-Seq analysis of differential splice junction usage and intron retentions by DEXSeq, Li [/bib_ref] analyses, after some reworking of the data to adapt them to IR. They are count-based. To adjust for differences in library size, a gene-wise normalization factor is determined and applied to all (exon and intron) bins, as in usual gene expression differential analyses.
The authors of ASpli 1 suggest to adjust each intron bin counts B_ {i}, in each sample s, by biological condition through:
[formula] B i G Condition à G À [/formula]
where G_{Condition} is the average gene count in condition and bar {G} the average gene count across all samples. Classical testing procedures (either of edgeR or DESeq2) are then applied to these adjusted counts to infer a set of differential introns.
In the DEXSeq-IR method, for each intron, two count bins are considered: the intron bin and the union of all the remaining bins. The average bin count is modeled via a negative binomial generalized linear model with interaction term.
In more details, for an intron indexed by i in a sample j:
[formula] log 2 mean Intron ijl ¼ b Sample ij þ lb Intron i þ b IntronxCondition iC [/formula]
where l = 1 for the intron bin and 0 otherwise. The sample parameter b_{Sample} adjusts internally for the gene expression level and differences in intron usage between conditions is inferred by testing whether the interaction term is significantly different from zero 2 .
As described previously, so as to limit spurious noise in intronic read counts, it may be worth refining further intron bin counts by removing reads that map to artifact-prone intervals. Several detection softwares already output these corrected read counts [bib_ref] IntEREst: intron-exon retention estimator, Oghabian [/bib_ref].
## Average intron coverage method
The third and most recent approach, iDiffIR (implemented but not yet published, https://bitbucket.org/comp_bio/idiffir, [bib_ref] Abiotic stresses modulate landscape of poplar transcriptome via alternative splicing, differential intron..., Filichkin [/bib_ref] [bib_ref] Common and distinct transcriptional signatures of mammalian embryonic lethality, Collins [/bib_ref] is primarily designed for IR.
IR levels are quantified, from genomic unique alignments, by the average per base read coverage (over the intron interval). To account for library size, TMM normalization [bib_ref] A scaling normalization method for differential expression analysis of RNA-seq data, Robinson [/bib_ref] is applied, separately, for intron and exon per base read counts. Per base counts are further normalized to force overall gene coverage to be equal across conditions. The test statistic is a corrected log fold change between the average read coverage in each of the conditions compared:
[formula] logFC I ð Þ ¼ log 2 a þ mean I1 ð Þ a þ mean I2 ð Þ [/formula]
The correction parameter a is a pseudo-count whose value is chosen to minimize the log fold-change and control large values caused by lowly covered introns.
The biological intuition of splicing efficiency translates quantitatively as the proportion of emitted transcripts which retain an intron. However, owing to short read size and rarity of IR events, this frequency cannot be reliably estimated from RNA-seq data. Available computational methods to perform IR differential analysis.
# Method
Year Metrics that can actually be computed (eg: PSI, IRratio) would only be proxys, and their properties and meaning are still poorly understood. Importantly, it is not sure whether intronic expression can be compared based on these metrics. All the problems discussed previously in the measurement of IR levels will have direct repercussions on the detection of alternate IR levels between samples. Ironically, despite numerous efforts to quantify IR events, the cases best suited to most of these approaches remains introns with high coverage and short length, exon-like introns.
# Discussion
The recent interest in expressed introns has led to a flourishing number of examples of regulation through intron retention. The accurate detection of retained introns and precise measurement of intronic expression are crucial to these studies. Numerous factors impede the detection of IR from next generation sequencing data. Introns are much longer than exons and thus have a much higher probability of containing overlapping features that may confound the estimation of intronic expression. In addition, introns are enriched in low complexity and repeat sequences that may prevent sequencing data from being uniquely mapped. These factors must be accounted for when detecting IR events. Most computational approaches however will introduce a selection bias as only introns with sufficient coverage can be detected and the statistical power required to detect differences between conditions increases with coverage depth and the read count [bib_ref] Transcript length bias in RNA-seq data confounds systems biology, Oshlack [/bib_ref] [bib_ref] Gene ontology analysis for RNA-seq: accounting for selection bias, Young [/bib_ref]. As a result of this bias, gene enrichment tests of genes derived from IR signatures [bib_ref] Multiple sources of bias confound functional enrichment analysis of global-omics data, Timmons [/bib_ref] are heavily skewed towards the more expressed genes and towards introns that do not contain these confounding features.
Despite the recent results that demonstrate a crucial role for IR, very few IR events have been validated in the wetlab and amongst these an even smaller portion have been investigated for their functional impact. As a consequence, no reliable benchmark of IR detection or differential intronic expression has been published. This lack of reliable controls is however temporary because long read technologies capable of sequencing entire IR transcripts help resolve most of the detection problems. However, due to their low coverage, these technologies are far from allowing a comprehensive detection of IR events and even further from allowing a reliable quantification of IR levels between different tissues.
## Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
[fig] Figure 1: Fig. 1. Defining intronic intervals to be analyzed. Comprehensiveness of transcript annotation and the selection of reference intronic sequences have a major impact on IR detection. In the example, we consider a gene having three possible isoforms (A, B and C). Exons are represented as plain rectangles and introns as thick black lines. If only Isoforms B and C were annotated, the starred interval (*) would not be defined as an intron and most likely not detected as retained. Colored boxes indicate whether the annotated introns match the ''all introns" or ''independent/measurable intron" criteria used by current algorithms. [/fig]
[fig] Figure 3: Standard implementation of computational detection of IR events. [/fig]
[table] Table 1: Computational tools available to perform IR detection and their main features. [/table]
|
Palliative care for people who use substances during communicable disease epidemics and pandemics: a scoping review protocol
# Introduction
Communicable disease epidemics and pandemics magnify the healthcare inequities encountered by people who use substances. [bib_ref] Statement by the un expert on the right to health* on the..., Dainius [/bib_ref] The United Nations 2019 World Drug Report estimated that 271 million people used substances in the previous year. Among this population, 13% suffered from substance use disorders (SUDs), 2 which are most commonly related to alcohol, cannabis and opioids. [bib_ref] The global burden of disease attributable to alcohol and drug use in..., Degenhardt [/bib_ref] In 2017, there were 585 000 deaths and 42 million years of healthy life lost globally as a result of substance use. 2 People who use substances are at high risk of life-limiting illnesses such as HIV, cancers, strokes and liver cirrhosis. [bib_ref] The global burden of disease attributable to alcohol and drug use in..., Degenhardt [/bib_ref] The most common causes of non-overdose-related deaths among people with opioid use disorders are cancers (17%-29%), cerebrovascular diseases (17%-22%), pulmonary diseases (20%) and infectious diseases (14%). The novel coronavirus pandemic has increased the risks associated with substance use, and vice versa. For instance, the pandemic has created immense challenges of providing care for people who use substances, [bib_ref] The COVID-19 pandemic and its impact on substance use: implications for prevention..., Ornell [/bib_ref] and loss of traditional in-person clinical and support interventions may exacerbate substance use and lead to worse health outcomes, particularly
Strengths and limitations of this study ► This study will be the first scoping review of peerreviewed and grey literature on providing palliative care to people who use substances during communicable disease epidemics and pandemics. ► Our scoping review will follow the established Open access those who are structurally disadvantaged. [bib_ref] COVID-19 and persons with substance use disorders: inequities and mitigation strategies, Melamed [/bib_ref] An analysis of electronic health records in the USA found that people recently diagnosed with a SUD were at an increased risk of COVID-19 and had a higher prevalence of comorbid chronic diseases affecting the kidneys, liver, lung, as well as cardiovascular diseases, type 2 diabetes, obesity and cancer. [bib_ref] COVID-19 risk and outcomes in patients with substance use disorders: analyses from..., Wang [/bib_ref] The overdose crisis has worsened during COVID-19, with some scholars referring to 'an epidemic within a pandemic'. [bib_ref] An epidemic in the midst of a pandemic: opioid use disorder and..., Alexander [/bib_ref] Given the high morbidity and mortality associated with substance use, the frequent and concurrent negative impacts of social and structural determinants of health, and the resulting increased risk during communicable disease emergencies including the current COVID-19 pandemic, these individuals should be a priority population to receive palliative care. Palliative care addresses the suffering experienced by people with life-limiting illnesses through symptom management, psychosocial support and advance care planning.Demand for palliative care is likely to increase during a pandemic due to increased mortality, with the additional possibility of deprioritising individuals with a lower chance of survival for critical care if available resources are overwhelmed. [bib_ref] Pandemic palliative care: beyond ventilators and saving lives, Arya [/bib_ref] Many people who use substances encounter barriers to receiving palliative care, including structural inequities resulting in fewer social supports and a lack of financial resources. [bib_ref] Illicit drug use as a challenge to the delivery of end-of-life care..., Mcneil [/bib_ref] Moreover, delivery of community-based health services, including palliative care, may be restricted due to concerns that the settings are risky or unsafe. [bib_ref] Illicit drug use as a challenge to the delivery of end-of-life care..., Mcneil [/bib_ref] Zero-tolerance policies toward non-medical use of substances may also restrict access to palliative care units and hospices.
For example, in response to the high transmissibility and mortality of SARS-CoV-2, healthcare has been limited to essential services.These services include the provision of substance use treatment and outreach, but many programmes, such as supervised consumption sites, are closed or have reduced hours of operation. 14 Similarly, palliative care providers are considered essential but their care has transitioned to primarily virtual care for clinically stable patients to reduce physical meetings. [bib_ref] Palliative care provision at a tertiary cancer center during a global pandemic, Hannon [/bib_ref] However, people may not have access to the necessary technologies to receive virtual care (eg, smartphones, internet), especially those who have severe mental health disorders, poor health literacy and socioeconomic disadvantages. 14 These changes are necessary to combat the COVID-19 pandemic but may have disproportionate negative consequences for people who use substances. [bib_ref] Collision of the COVID-19 and addiction epidemics, Volkow [/bib_ref] In combination with the overdose crisis, restricted access to mental health and substance use services can increase the risk of relapse and overdoses. Additionally, physical distancing is an important public health measure to mitigate the spread of COVID-19 but this measure may not be possible for some people who use substances, and especially for individuals who live in congregate settings such as shelters.
Physical isolation can reduce their support system and further increase their risk of relapse and overdose.Given the pre-existing inequities to palliative care access, it is important to understand the impact of communicable disease epidemics and pandemics on people with lifelimiting illnesses who use substances.
## Rationale
Preliminary database searches (PubMed, Web of Science) in May 2020 did not identify any reviews that address the palliative care needs for people who use substances during communicable disease outbreaks, such as severe acute respiratory syndrome (SARS), influenza A (H1N1), Ebola and COVID-19. Therefore, we propose a knowledge synthesis project that will allow us to identify and map available and emerging evidence about palliative care for patients who use substances during communicable disease epidemics and pandemics. The findings from our scoping review will provide accessible and relevant evidence for healthcare professionals and decisionmakers (eg, policymakers, administrators) that can be applied to the COVID-19 pandemic response efforts.
## Review objective
The primary objective of this scoping review is to answer the research question, 'what is known about communicable disease epidemics and pandemics, palliative care and people who use substances?' Specifically, the focus of our review is on people with life-limiting illnesses who use substances. Scoping review research questions are intended to be broad, as comprehensiveness and breadth are important to map a literature. [bib_ref] Scoping reviews: time for clarity in definition, methods, and reporting, Colquhoun [/bib_ref]
# Methodology
This article describes the protocol for a scoping review that is currently being conducted. Our scoping review is guided by Levac et al's six-stage framework, [bib_ref] Scoping studies: advancing the methodology, Levac [/bib_ref] which builds on the contributions of Arksey and O'Malley, [bib_ref] Scoping studies: towards a methodological framework, Arksey [/bib_ref] Joanna Briggs Institute guidance document for the conduct of scoping reviews, 21 and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) [bib_ref] PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation, Tricco [/bib_ref] reporting guidelines.
## Eligibility criteria
We will use the PECOS (Population, Exposure, Comparison or Control, Outcomes and Characteristics) framework [bib_ref] Identifying the PECO: a framework for formulating good questions to explore the..., Morgan [/bib_ref] to structure our evidence eligibility criteria. For 'Population', we will include adults (defined as ≥18 years of age) who use substances and have life-limiting illnesses (eg, advanced cancer, end-stage organ failure), including those who receive treatment for SUDs and use prescription drugs and/or unregulated substances. In our study, we will acknowledge that substance use occurs on a spectrum [bib_ref] A concept analysis of substance misuse to inform contemporary terminology, Mahmoud [/bib_ref] and that for some people, their substance use may not be a 'problem' in their lives although they might meet the diagnostic criteria for SUD as per the Diagnostic and Statistical Manual of Mental Disorders, fifth edition. We will focus on the most used substances globally: alcohol, tobacco, cannabis, opioids (eg, heroin), cocaine, central nervous system depressants (eg, benzodiazepines) and stimulants (eg, amphetamines).The 'Exposure' will be Open access communicable disease outbreaks, specifically epidemics and pandemics (eg, COVID-19, SARS). We will include literature with and without 'Comparison or Control' groups, including grey literature (eg, government reports, policy statements). Our 'Outcomes' of interest will be quantitative outcomes and qualitative themes related to palliative care; specifically, interventions, access to services, policies/guidelines and clinical programmes that address physical, psychological, social and/or spiritual issues faced by patients with life-limiting illnesses and their families. We will include topics such as education outreach to family members and harm reduction and overdose prevention if they occur within a palliative care framework. Lastly, our study 'Characteristics' will include research conducted in any country and published in English or French. We will restrict our studies to these languages due to limited available resources and the rapid nature of our review. We aim to include peer-reviewed and non-peer-reviewed literature including letters to the editors, theses and book chapters. Eligible studies will not be limited to specific publication years.
## Information sources and search strategy
An information specialist, in collaboration with our team of technical and subject matter experts in scoping reviews, palliative care and substance use, developed a comprehensive search strategy to identify studies in the following bibliographical databases: Medline ALL (Medline and Epub Ahead of Print and In-Process and Other Non-Indexed Citations), Embase Classic+Embase, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trial, PsycInfo all from the OvidSP platform and Scopus from Elsevier. Where available, both controlled vocabulary terms and text words were used. There were no date or publication type restrictions. Where applicable, the search was limited to adults and the English/French language. The online supplemental appendix presents our Ovid Medline (R) ALL 1946 to 28 August 2020, search as an example of our search strategy. We are using EndNote, a reference management software, to store data from these database searches.We also conducted a grey literature search in August 2020 to identify published literature that is not indexed in the bibliographical databases, including but not limited to guidelines, conference and meeting abstracts, theses and reports. We searched grey literature resources, such as TRIP, Google Scholar, Google, WHO, websites of prominent health organisations and associations of palliative care, public health (communicable disease), substance use, individual conference, theses and guidelines. Search strategies were customised based on each resource's available searching feature. For instance, Google limits the number of queries; therefore, individual searches were conducted for each communicable disease listed in the study's scope combining with palliative care and substance use. Each search was further limited to a specific country to retrieve more focused results.
All the searches from the bibliographical databases and grey literature resources will be rerun in April 2021 to retrieve recently published literature. Furthermore, we will hand search reference lists of all potentially eligible full-text publications to identify additional relevant studies.
# Methods limitations
Our search may be limited because the scoping review cannot include all possible literature, as new information regarding treatment changes daily and will not include literature published in languages other than English and French. Since our review explores different types of communicable diseases (eg, COVID-19, HIV, tuberculosis), specific disease may have different palliative care resource implications at the health system level and may look different depending on the type of substance in question. It is beyond the scope of this protocol to explore all possible permutations and combinations.
## Selection of sources of evidence
From our database search from inception to 28 August 2020, we imported 5939 records obtained from the database searches into Covidence, a web-based literature review software.We will use this program to detect and remove duplicates. We will screen all the record titles and abstracts in duplicate according to our eligibility criteria. The same process will be applied to the full-text articles. If there is a disagreement, we will consult a third party to act as a tiebreaker. Records with no abstract will be screened based only on their title. If there are conference proceedings or abstracts without full text, we will exclude these records due to the limited information available. These three main reasons will count toward exclusion: wrong population (eg, people with curative diagnoses), wrong exposure (eg, non-communicable disease public health emergencies) and wrong outcomes (eg, not related to palliative care).
We have imported the 2889 records identified in the initial grey literature search into Google Sheets. We will manually assess each record's eligibility by applying our PECOS framework criteria. If there are disagreements between the two investigators assessing the records, they will be resolved by consulting a third member as a tiebreaker.
Data charting process For our research from August 2020, we developed a standardised data charting form using Google Sheets to extract the following variables: author/organisation, publication year, journal, country of the corresponding author, article type (commentary/opinion article, conference/workshop, guideline, news article, programme report, other), publication language, countries and settings in which studies were conducted (acute care, inpatient palliative care, long-term care, community, home care, infectious diseases clinic, palliative care clinic), populations (lifelimiting diagnoses, substances used, sex and gender, age, Open access ethnicities), target audiences (healthcare professionals, researchers, policymakers, governments), communicable diseases and palliative care outcomes (interventions, access to services, policies/guidelines and clinical programmes).
# Synthesis of results
We will develop a PRISMA-ScR flow diagram to map out the review process including the number of identified records, studies included and excluded, and the reasons for exclusion. We will also provide a narrative description of the search decision process. Quantitative data (eg, study characteristics documented in our data charting process) will be analysed using descriptive statistics and summarised in tabular and graphical forms. We will summarise the findings in text using a narrative approach. Knowledge gaps will be identified through comparative analysis of the extracted data, particularly the palliative care outcomes.
We will analyse the qualitative data thematically to help us make sense of the large amounts of data. Based on the priorities of our knowledge users, who are experts in palliative care and addictions, we have developed a codebook with the following four palliative care-related codes that will be applied to our findings: interventions, clinical programmes, policies/guidelines and access (please see table 1). We will code each record in duplicate using the codebook and the qualitative analysis software, NVivo.During the coding process, we will use the constant comparative method to inductively identify additional codes that capture common themes and differences within and between the records. [bib_ref] The constant comparative analysis method outside of grounded theory, Fram [/bib_ref] Based on our inductive thematic analysis, we will likely identify subthemes within each main code. Unlike the deductive component of our analysis, the subthemes cannot be identified a priori. We will report our findings in narrative form, and our synthesis will explore relationships such as the type of substance use and life-limiting illness, as well as geographical, cultural factors, as well as epidemic or pandemic context in which the palliative care is provided.
## Patient and public involvement
There were no patients involved in the design of this protocol. The research team includes healthcare professionals who provide palliative care for people who use substances and several team members who work closely with people who use substances in the context of hospital and community-based work. Our research team also includes a principal knowledge user, Canadian Virtual Hospice, that provides comprehensive online knowledge about palliative and end-oflife care. These members have provided guidance on Open access designing the scoping review and will help lead our knowledge dissemination strategy.
## Ethics and dissemination
We did not require approval from a research ethics board for this study as it does not involve human participants or unpublished secondary data. We will refine our research question and interpret our results in consultation with our content experts and knowledge users. Our principal knowledge user, Canadian Virtual Hospice, is an award-winning digital health palliative care organisation that provides comprehensive online knowledge about palliative and end-of-life care. Their website has more than 25 000 visits per month.They will lead our integrated and end-of-grant knowledge mobilisation strategy. Our target audiences will be healthcare professionals, decision-makers, patients and their families. The findings from this scoping review will be disseminated through professional networks, digital communications using social media platforms, conference presentations and scientific journal publications. Our knowledge dissemination plan includes development of infographics, digital communications (eg, Twitter) and online news features.
## Implications
Our scoping review will provide important information to inform palliative care practice and policy. We aim to rapidly synthesise available evidence to identify strengths and gaps in the existing knowledge about palliative care for people who use substances during communicable disease epidemics and pandemics.
Our goal is to provide information that may be used to minimise inequities in palliative care access for this population, especially during times of health system strain and resource scarcity.
[table] Table 1: Palliative care-related codes used for deductive analysis Policies/guideline '…decisions, plans and actions that are undertaken to achieve specific health care goals within a society.' 'An explicit health policy can achieve several things: it defines a vision for the future which in turn helps to establish targets and points of reference for the short and medium term. It outlines priorities and the expected roles of different groups; and it builds consensus and informs people.'Access '…systems, processes and resources [that] are in place so that people with identified palliative care needs can receive palliative care support whenever they need it' 'Access to appropriate palliative care means that it meets the needs of those requiring care at the right time, in the right place and by the right provider.' WHO's definition of access for universal health coverage: '…availability of good health services… and other aspects of service organization and delivery that allow people to obtain the services when they need them'; '… measure of people's ability to pay for services without financial hardship'; and '… people's willingness to seek services' clinical-guide-en.pdf https://www.cihi.ca/sites/default/ files/document/access-palliativecare-2018-en-web.pdf https://www.who.int/bulletin/ volumes/91/8/13-125450.pdf [/table]
|
Cancer: An unknown territory; rethinking before going ahead
Cancer is a disease of altered signaling and metabolism, causing uncontrolled division and survival of transformed cells. A host of molecules, factors, and conditions have been designated as underlying causes for the inception and progression of the disease. An enormous amount of data is available, system-wide interaction networks of the genes and proteins are generated over the years and have now reached up to a level of saturation, where we need to shift our focus to the more advanced and comprehensive methods and approaches of data analysis and visualization. Even with the availability of enormous literature on this one of the most pressing pathological conditions, a successful cure of the disease seems to be obscure. New treatment plans, like immunotherapy and precision medicine, are being employed for different studies. Nevertheless, their actual benefits to the patients would be known only after the evaluation of clinical data over the next few years. Therefore, we need to look at few fundamental challenges that should be addressed in more depth before we could devise better, rigorous, and comprehensive treatment plans and may successfully reach a possible cure of the disease. This article aims at bringing attention towards some fundamental gaps in our approach towards the disease that leads to failure in devising successful therapeutics.
# Introduction
Cancers are one of the most devastating classes of human pathologies, presenting the versatile range of hallmark clinical features and leading to millions of deaths each year around the globe. These groups of maladies constitute more than a hundred genetically diverse conditions sharing several commonalities in the molecular mechanisms and metabolic alterations among themselves. [bib_ref] Understanding the intersections between metabolism and cancer biology, Vander Heiden [/bib_ref] [bib_ref] Emerging biological principles of metastasis, Lambert [/bib_ref] The direct involvement of the tissue microenvironment and inflammatory changes on the tumor growth survival is wellestablished. [bib_ref] Microenvironmental regulation of tumor progression and metastasis, Quail [/bib_ref] [bib_ref] Cancer-related inflammation, Mantovani [/bib_ref] However, a clear understanding of the underlying causes and factors is still elusive and requires more research. A plethora of genetic mutations has been reported that could end up in the transformation of the normal human cells leading to the genesis of tumor and development of cancer. [bib_ref] Genetic diversity, inbreeding and cancer, Ujvari [/bib_ref] [bib_ref] The challenges of tumor genetic diversity, Mroz [/bib_ref] In the past, a variety of scientific and technological approaches have been tried in order to understand, define, investigate, and challenge these extensively dreadful forms of diseases. Various approaches applied in oncological research include genetic, molecular, biochemical, biophysical, immunological, genomic, proteomic, systems and computational biology, etc. However, none of these have sufficed the needs to devise successful treatment strategies so far. [bib_ref] A broad-spectrum integrative design for cancer prevention and therapy: the challenge ahead, Bishayee [/bib_ref] The area of the oncological research is well-documented but evolving in a way that researchers and clinicians find it difficult to get updated and informed of the new information and advancements most of the time. Therefore, the field requires a continuous assessment of the new developments and analyses of the lacunae present within persisting knowledge.
Understanding cancer: do we need to change the approach?
Although a vast volume of data has accumulated over the decades of research, we at the present stage of scientific and technological advancements still lack a sound understanding of these debilitating conditions that leads to millions of deaths every year. In fact, before classifying the cancers as classical disease conditions, we need to understand that these could be considered as a phenomenon of sequential alteration at molecular, cellular, tissue, and organ levels affecting the whole physiology of the organism. We cannot ignore the fact that the inception of the disease is neither caused necessarily by some external agents nor is potentially associated with the dysfunction of any specific organ, as occurs in many other disease classes. In simpler terms, cancer could be caused by multiple kinds of disorders in the highly ordered cellular physiological systems. [bib_ref] Origin of cancer: an information, energy, and matter disease, Hanselmann [/bib_ref] The onset of the disease is still the most intriguing part of the oncology, that we are far from understanding. [bib_ref] Cells of origin in cancer, Visvader [/bib_ref] Enormous data has accumulated showing multiple kinds of physical, chemical, biological, genetic, and environmental factors leading to the transformation phenomenon. However, unfortunately, all these reports have just led us to the tip of the iceberg. We are yet to get a proper understanding of the onset of the disease, and until we get the insights of the initiation, we could not find the actual targets for delivering the treatments.
The past century of scientific research has given us a detailed picture of primary, secondary, and tertiary changes associated with these conditions. However, the hunt to get point zero is still on, and more efforts are needed to reach right at the origin of the disease. The reason behind raising the issue of the origin of cancer is that the primary reasons behind not being able to crack the codes of this century-old problem could be our 'top-down' approach that works on providing symptomatic relieves from the disease conditions. We have confined our research most of the time in designing and improving the methods and strategies of arresting the growth of transformed tumor cell mass. Unfortunately, we have shifted our objectives primarily towards devising newer preventive measures rather than looking for the origin of the disease. The reason behind this shift could also be due to the continuous failure of scientists in delivering the results in 'bottom-up' approaches where more focus should be given to the genesis of cancer, instead of ways to cut down resources for its sustenance or forcibly inducing the cell death pathways. We must also look for the evolutionary perspectives of the origin, sustenance, and development of cancer-like conditions over the eons to understand why and how selection forces have let these processes to settle down with our current set of physiological paradigms. In fact, we need a comprehensive multi-dimensional approach to clearly investigate and understand the disease and thus to deliver more effective and successful treatment strategies.
Can we look at cancer through an evolutionists prism?
In the views of an evolutionist, the development of cancers seems to be driven by the continuous acquisition of numerous somatic mutations to best fit into a continuously changing and challenging microenvironment. [bib_ref] Evolutionary dynamics of neoantigens in growing tumours, Lakatos [/bib_ref] Spontaneous mutation-driven evolution of tumor cells itself is one significant barrier before us that tends to provide them the resistance against the drugs and treatment plans. [bib_ref] Evolutionary determinants of cancer, Greaves [/bib_ref] The dialectical interrelations among different physiological pathways and their dynamic interaction networks have further complexicized our molecular understanding of the disease. The uniqueness of each cancer type and diversity among the tumors additionally raise multifold challenges in understanding this terrible and complicated disease and devising treatment strategies to fight against it. Additional adversities are added by their unique and yet to be understood capabilities of evading the host immune responses, thereby limiting the inherent tendency of the body to fight back with these odious masses of tissues. In-depth knowledge of underlying pathways and mechanisms behind the origin, transformation, and development of cancer may provide us therapeutic advantages in switching to a 'bottom-up' approach that will hand us an edge in targeting the most underlying molecular pathways. The past century has seen tremendous methodological and technological advancements in the diagnosis, and treatment of various types of cancers.
Where do we stand?
Cancer cells are a kind of parasitic population of our own cells that resides in our body, utilizing the nutrition and supplementation of normal cells. They have learned to hijack the metabolic pathways and exploit the tissue microenvironment for their growth, sustenance, and progression. They have acquired the art of camouflage and thus evade our army of immune cells of various kinds. Their ability to mutate and evolve within the population further complicates the condition, which makes it hard for researchers and scientists to understand how to target these unwelcome populations of our cells, which otherwise may lead to enormous complications in the end. One school of thought finds possible parallelism between developmental processes and the origin of cancer; however, more support on the notion is needed. Thousands of studies are conducted every year with new findings and solutions of the disease advancing our present knowledge about the disease. However, the disease is still growing with an increased pace and turns out to be the leading cause of mortality in many developed countries. Despite acquiring enormous knowledge on the etiology, causes, and effects of cancer inside the body, unfortunately, we could not have successfully devised too many methods to curb the disease condition and provide proper relief to the patients suffering from this one of the most challenging puzzles of the medical sciences. [bib_ref] Tumo evolution as a therapeutic target, Amirouchene-Angelozzi [/bib_ref] In most cases, a treatment plan for the majority of the cancer types remains limited to surgical removal, chemotherapeutic targeting, high-intensity photon-beam radiotherapy, hormonal therapy, and modern-day immunotherapy depending upon their complexity, stage, and localization. 14 All patients suffering from one type of cancer, despite being genetically diverse, typically receive a similar treatment plan. This diversity among the cell population of the same tumor mass leads to the attainment of the drug resistance in the transformed cells. Also, the lack of real-time monitoring of the ongoing treatment further worsens the situation. Unfortunately, due to the limitations of resources and diagnostic tools and a lack of personalized treatment strategies, in addition to the highly diverse genetic makeup of tumor populations, many times, treatments do not meet the expectations. Here, a few major characteristic features of the disease, which mostly intrigues our understanding and raise new challenges in therapeutics, are discussed in detail.
## Extreme genetic diversity
Intratumor heterogeneity, i.e., extreme genetic diversity among cell populations residing in the same tumor mass, leads to Darwinian principles and natural selection forces to work on it and establish a more robust variety of cancer cells. [bib_ref] Clonal heterogeneity and tumor evolution: past, present, and the future, Mcgranahan [/bib_ref] This heterogeneity increases drug resistance and thus poses a great therapeutic challenge before clinicians in developing cancer treatment plans, which therefore led to the evolution of the idea of personalized medicines. [bib_ref] Personalized medicine in cancer: where are we today?, Turnbull [/bib_ref] [bib_ref] The causes and consequences of genetic heterogeneity in cancer evolution, Burrell [/bib_ref] Increased application of pharmacogenetics, i.e., understanding the genetics of cancer by molecular profiling of the disease cells, identification of the mutation associated with the given cancer type in an individual patient is nowadays used for targeting specific genes or proteins driving the growth of cancer. [bib_ref] Cancer pharmacogenomics: strategies and challenges, Wheeler [/bib_ref] The advent of nextgeneration sequencing, array methods, and other mathematical or computational tools have tremendously driven the progress of this revolutionary approach to target the cancer cells. 19e21 However, the application of personalized medicines is very limited at present due to multiple reasons. The high cost of cancer genomics is one major reason behind not opting for precision medicine at a larger scale. [bib_ref] Predictive, personalized, preventive, participatory (P4) cancer medicine, Hood [/bib_ref] [bib_ref] Cancer genomics: from discovery science to personalized medicine, Chin [/bib_ref] Possibly, the diversity at the genetic level in these diseases and the inability to afford for personalized treatment strategies is one of the major underlying causes of failures in the treatment of cancer. Therefore, more efforts are required in this direction so that precision medicines can be used in other cases also.
Looking at various genes independently may not be an ideal way to understand the plethora of changes occurring while a cell transforms, and the tumor is formed; therefore, a broad landscape of genome-wide alteration may provide a better understanding of the genetics of cancer.The increase in genetic diversity, in recent years, has also been linked with a phenomenon called genome or karyotype chaos, primarily caused due to largescale changes creating new genomic codes for the systemic inheritance, which in turn provide increased chances of adaptation and survival at the cellular and organismal level. [bib_ref] Debating Cancer: The Paradox in Cancer Research, Heng [/bib_ref] [bib_ref] Micronuclei and genome chaos: changing the system inheritance, Ye [/bib_ref] An increasing number of sequencing-based studies have hypothesized for the possibility of a more pivotal influence of the genome topology over the tumor diversity in comparison to the independent gene profiles. [bib_ref] Tracking the evolution of nonesmall-cell lung cancer, Jamal-Hanjani [/bib_ref] 'Chromothripsis' is an umbrella term used to indicate multiple kinds of chromosomal rearrangements happening under crisis, which may include: chromosomal aneuploidy, chromoplexy, chromoanagenesis, etc. [bib_ref] Genome chaos: survival strategy during crisis, Liu [/bib_ref] Clonal or constitutional aneuploidy is another less understood phenomenon reported in many cancer types, which may also act as a heterogeneous agent affecting the emergence and evolution of cancer. [bib_ref] Aneuploidy and cancer: from correlation to causation, Duesberg [/bib_ref] [bib_ref] Understanding aneuploidy in cancer through the lens of system inheritance, fuzzy inheritance..., Ye [/bib_ref] These terms are not very familiar to oncology research as most of these processes were overlooked for years, while scientists were more inclined towards the genetic theory of cancer, which has largely failed to address many fundamental questions of evolution, development, and diversity of cancer. [bib_ref] Aneuploidy and cancer: from correlation to causation, Duesberg [/bib_ref] An opaque connection between developmental biology and cancer Do cancers have any developmental correlation; and if the disease could be transferred from one generation to the other? These are a few more questions, which are not understood in detail. At face values, development entails order, whereas cancer represents an example of extreme disorder. Interestingly, such a distinctive origin and progression of the two processes define their dichotomous correlation. When Prof. Mintz referred cancer as an error of development, possibly she took into account the proliferative capabilities of the cells, which could be compared with that of early-stage stem cell populations during organism development. [bib_ref] Gene expression in neoplasia and differentiation, Mintz [/bib_ref] However, the two processes are at considerably far distant ends of the same spectrum of diversification. On the one hand, development is a process that originates from a single zygotic cell that diversifies into multiple cell types via a tightly controlled epigenetic regulation. Contrarily, cancer is a slow and continuous progression towards a highly similar cell population at late malignancy stages from the early benign tissue masses, which are highly diverse. [bib_ref] Developmental disease and cancer: biological and clinical overlaps, Bellacosa [/bib_ref] Many similar epigenetic signatures could be found among both the developmental stem cells and cancer progenitors, marking the parallelism between both processes. [bib_ref] Epigenetic stem cell signature in cancer, Widschwendter [/bib_ref] There is also an ambiguity upon the idea of inheritance of cancer-causing disease mutations, which may or may not be transferred to progenies. Possibly, some of the mutations are inherited, while many of the cases are acquired because of replication errors during the multiplication of cells.
Many overlapping clinical features among multiple developmental disorders and cancer predisposition could be another possible connecting link between the two processes, putting forward an idea of cancer being a disease of accumulating developmental errors. [bib_ref] Developmental disease and cancer: biological and clinical overlaps, Bellacosa [/bib_ref] [bib_ref] Cancer as an embryological phenomenon and its developmental pathways: a hypothesis regarding..., Cofre [/bib_ref] Highly precise control over spatial and temporal switching ON/OFF of a well-orchestrated gene network is another binding feature that connects the two distinct biological phenomena. [bib_ref] Rethinking "cancer as a dynamic developmental disorder" a quarter century later, Rubin [/bib_ref] Owing to these similarities, many scientists are now exploiting both types of disease models to understand each other. [bib_ref] Echoes of the embryo: using the developmental biology toolkit to study cancer, Aiello [/bib_ref] In the coming years, a more rigorous shift in approach towards studying cancer as a disease of erroneous development is needed. This change of approach may educate us more about the origin and development of this highly diverse disease.
## Hiding away from innate immunity
Another major challenge that scientists have faced over the years is the extreme disguise or hiding of cancer cells from our immune system that are expected to identify and kill the cancer cells. [bib_ref] Immune evasion in cancer: mechanistic basis and therapeutic strategies, Vinay [/bib_ref] The specific mechanisms of how precisely cancer cells evade our immune system and what could be the possible strategies to target these aberrant cells of our own body is still an enigma of modern-day biology. [bib_ref] Mechanisms of tumor celleintrinsic immune evasion, Spranger [/bib_ref] However, targeting neoantigens is a possible strategy that is under consideration with huge expectations, as these newly evolved peptides provide immune cells an opportunity to target cells, which otherwise remain in disguise. Some scientists believe that it is a dynamic head to head clash between neoantigen expressing cancer cells and the immune surveillance that poses a quest of survival before cancer cells. [bib_ref] Cancer immunotherapy targeting neoantigens, Lu [/bib_ref] [bib_ref] Immune targets and neoantigens for cancer immunotherapy and precision medicine, Wang [/bib_ref] The selection pressure mounted, therefore, leads to mutations in such a way that they start tricking and escaping our own immune cells. [bib_ref] Neoantigens generated by individual mutations and their role in cancer immunity and..., Efremova [/bib_ref] Additionally, the disease cells, by expressing some specific antigens or modifying the tumor environment, acquire: the tendency or ability to overpower the normal immune responses inside the body. [bib_ref] Remodeling the tumor microenvironment with emerging nanotherapeutics, Chen [/bib_ref] [bib_ref] Immune responses to tumour antigens: implications for antigen specific immunotherapy of cancer, Jäger [/bib_ref] Cancer immune surveillance has remained a longstanding topic of debate that has both proponents and opponents. It is hard to answer whether our immune system promotes or suppresses, or it directly ignores the mutated or transformed cells. [bib_ref] Immuno-oncology: understanding the function and dysfunction of the immune system in cancer, Finn [/bib_ref] [bib_ref] The immune system in cancer metastasis: friend or foe?, Janssen [/bib_ref] Interestingly, over the years, it has been suggested that transformed cells themselves sculpt the immune cells via a process called immuneediting in such a way that they help them in skipping their molecular identification and elimination and establishing a Darwinian selection. [bib_ref] Cancer immunoediting: from immunosurveillance to tumor escape, Dunn [/bib_ref] Few reports indicating latent metastasis after decades of surgical removal of the primary tumor is another question that, if answered, may probably help us better understand how the cells conceal themselves from possible immune attacks. [bib_ref] Latent bone metastasis in breast cancer tied to Src-dependent survival signals, Zhang [/bib_ref] Currently, we are quickly moving towards modern immunotherapy-based treatment methods, which are providing very positive results indicating a better future of these strategies towards cancer therapeutics. [bib_ref] The future of immunotherapy: a 20-year perspective, Wraith [/bib_ref] [bib_ref] Cancer immunotherapy: the beginning of the end of cancer, Farkona [/bib_ref] Chimeric antigen receptor (CAR)-T cells were the first line of treatment based on immune modulation, which successfully initiates a ferocious assault on cancer cells. The clinical trials have shown enthusiastic results in the past, but at this point of time, we cannot predict the future of these treatment plans.We should not underestimate the art of camouflage acquired by cancer cells over a long period of evolution. Instead, we must undermine the evolutionary relations between cancer cells and different types of immune cells.
We need to find out the possible diversions of the eukaryotic transformed cells when in the history of evolution, they have started deceiving our own immunity. We need to investigate the history of life to find out the answers to the present-day questions in order to find solutions for the future. Most of the physiological hallmarks defined for the cancers have their own evolutionary significance, and their retention by the selection forces might have their own significance and implications in the process of evolution.
A controversy: one target multiple bullets; or multiple targets with a common bullet?
A large fraction of research to date has focused on identifying the multimodal driving factors of the disease. Nevertheless, when the causative factors, genes, and pathways are subjected to the possibilities of druggability, most of these have shown limitations of multiple types. One major problem with most of the molecular targets is the toxicity generated due to their modulations, as most of these are somehow part of one or multiple molecular pathways. [bib_ref] Targeted therapies: a new generation of cancer treatments, Gerber [/bib_ref] Several articles have been published in recent years covering various aspects of the disease in detail. Thus, here in the present article, the primary focus is to highlight the gaps in postulation and understanding of the disease. Cancer has remained an unsolved puzzle over the years, despite having an extensive knowledge base of genes, proteins, and pathways affected during the pathogenesis. The primary cause of failure for most of the hypotheses could be overlooking many vital topics and defining factors in the disease etiology. What remains crucial is a rethinking of the already available knowledge and revisiting the persisting treatment methodologies and therapeutic approaches (see . It is obvious to say that an overwhelming amount of data is present and is enough to spend days or months to understand the complicated networks and models that can be generated using various available tools out of this vast literature available over online resources.
The exploitation of the available knowledge needs multimodal approaches, forming multidisciplinary teams with experts from different fields collaborating with and complementing each other. The use of computational tools and system biology approaches may further benefit the field with faster and more accurate calculations and predictions, saving lots of money and time that is spent in a large number of unsuccessful clinical trials. A large amount of biochemical and genomic data is present; however, the focus should be shifted now towards the transformation of these research into clinical applications. The majority of the data remain limited to the publications only and could never generate any therapeutic benefit. Many drugs proposed through these studies fail in successive phases of clinical trials, either because of their limited effects on curbing the disease or due to high toxicity profiles. In the end, a tiny fraction of drugs proposed every year get into clinics and could be utilized for treatment purposes. The primary reasons behind the failure of many drugs are because of the negligence of specific facts associated with this multifactorial disease.
Regulating the cell cycle progression, inducing apoptosis, managing the tissue microenvironment, modulation of the immune system, and cutting the sources of nutrition and growth, all at a time could be near to impossible for most of the drugs, which are currently in practice or under clinical trials. Even if we consider any drug, with the least of the possibilities to have most of the characteristics mentioned above, the more significant challenge would be its delivery to the specific tissues. Targeted delivery of drugs is another highly frustrating challenge faced by researchers and clinicians. None of the approaches proposed to date for the drug delivery to the affected site has stood with the promises and leads to the toxicity and collateral adversities, as is seen in patients undergoing chemotherapy. Due to this multifaceted nature of cancer development and progression, we ultimately land in an unfortunate situation of firing multiple rounds of bullets (chemotherapy, radiation, surgical removal, and immunotherapy, etc.) all at a time. The lack of a broader understanding of the pathological condition leads to overall physiological and psychological distress to the patients.
Surgical removal of affected body parts or heavy doses of chemotherapeutic drugs and high-energy radiations may cause physical disfigurements, followed by weakening of the immunity and deterioration of the body mass. All these combinedly affects the psychology of the patients with the highest of the will power. Family members of the patients may also get frustrated due to the very long treatment process and follow-ups. Also, most of the newly developed methodologies, like immunotherapy, cancer genomics, and precision medicine, are too costly to be afforded by most of the patients in underdeveloped and developing countries. Most of the cases of these diseases do not present any noticeable symptoms and remain undetected in earlier stages, and when detected, leave patients and doctors with minimal palliative options to choose and start the treatment. Diagnosis of the pathological symptoms at a very late stage is another major challenge that needs to be addressed with high priority. In fact, many things are possibly required for the consideration of the researchers, clinical scientists, and doctors.
## Conclusions: the road ahead
Cancer has remained one of the significant health issues for long. Description of the disease could be found in some of the ancient literature of Indian and Chinese medicines. The ancient scholars and clinicians suggested different treatment strategies. It could not be accurately predicted how An overview of known, unknown, available therapeutics, and future directions of the cancer therapeutics. The top panel shows the fundamental underlying causes of cancer, including major risk factors, involved mechanisms and currently available strategies to devise treatments and therapeutic strategies. The central part shows a few representative types of cancer among many occurring around the globe and leading to a very high number of deaths. The lower subsections show the treatment methodologies adopted around the world to curb the disease progression. In contrast, the few most challenging and lessunderstood complexities of tumor biology are indicated in the lowermost subsection of the figure. was prepared using a few templates from Servier Medical Art by Servier ( http://smart.servier.com/), which is licensed under a Creative Commons Attribution 3.0 Unported License. effective these treatment plans were; although their applications in the modern age medicines cannot be neglected. In the past century, efforts have intensified exponentially in all parts of the world to address the common health problems and diseases associated with such conditions. Many of the endemics have been cured completely. Most of the deadly infectious diseases leading to mass death have been understood and have possible cures around the corner. Several are still under investigation and may get some solutions in the past. In between all these advancements, one major class of human diseases, cancer that is neither acute in origin nor contagious in the spread, remain uncured up to a more considerable extent. Enormous hopes have been ignited from time to time by the development of multiple promising drugs, methods of their targeted delivery, and the evolution of novel treatment strategies, like immunotherapy, in the past.
All the strategies developed so far, and the drugs approved until now have their own limitations and toxicities associated with their use. Nevertheless, due to the lack of more sophisticated ways of treatments, the patients suffering from the advanced terminal stages of the disease usually are left with very few options to live with. Several new treatment plans are underway and may need a decade or more to be available for patients of developing countries at an affordable cost. Most of the research conducted around the world is now limited to finding ways of trimming away large branches (symptoms) of a tree (cancer), which may further grow and proliferate at new sites. The knowledge of the point of origin of the tree and ways to cut it from the bottom is still missing and needs extensive work. The field may need a complete turnaround in the approaches of looking at the disease in order to find proper solutions. It is hard at present to predict how much time we still require in reaching a definitive solution. However, yes, there always remains a hope, and we must follow it.
AcknowledgementsThe author would like to thank the Central University of Rajasthan for providing basic requirements during the preparation of the manuscript. The author also appreciates Servier (http://smart.servier.com/) for providing templates for the preparation of medical illustrations under Creative Commons Attribution 3.0 Unported License.Conflict of InterestsThe author declares no competing interests. |
Characteristics of tertiary lymphoid structures in primary cancers
© 2013 Landes Bioscience reVIeW TLSs have first been described in humans in many non-neoplastic inflammatory contexts, such as autoimmune diseases, viral and bacterial infections, graft rejection, and several idiopathic diseases. 5-7 The inducible nature of TLSs was first highlighted by Kratz and colleagues in a transgenic mouse model expressing
# Introduction
A tumor is a complex network of interacting cellular compartments, comprising malignant cells as well as endothelial, stromal, and immune cells. This so-called "tumor interactome," which differs from patient to patient, is intimately associated with tumor progression and disease outcome. [bib_ref] Hallmarks of cancer: the next generation, Hanahan [/bib_ref] Tumors have been shown to establish an immunosuppressive microenvironment that robustly inhibits, if not completely neutralizes, immune effector cells. This is instrumental for malignant cells to replicate uncontrolled and spread throughout the entire body. Several lines of evidence from mouse and human studies have reported that the immune system can control oncogenesis and tumor progression in spite of such an immunosuppressive pressure, at least under selected circumstances. [bib_ref] The immune contexture in human tumours: impact on clinical outcome, Fridman [/bib_ref] [bib_ref] Adaptive immunity maintains occult cancer in an equilibrium state, Koebel [/bib_ref] More particularly, an association between an abundant infiltration of neoplastic lesions by T cells with improved clinical outcome has been reported in patients affected by multiple types of solid neoplasms. [bib_ref] The immune contexture in human tumours: impact on clinical outcome, Fridman [/bib_ref] However, the mechanisms that govern the shaping of an efficient immune contexture within the tumor remain poorly understood. Moreover, the site(s) at which the activation of tumor-specific effector cells takes place remain(s) unknown. The dogma states that professional antigen-presenting cells (APCs) sample and process antigens within the tumor, then migrate through lymphatic and/or blood vessels to tumor-draining lymph nodes, where they present processed peptides to antigen-specific T and B cells. [bib_ref] Cancer immunotherapy comes of age, Mellman [/bib_ref] However, an increasing body of evidence has demonstrated that adaptive immune responses that efficiently protect the host against infectious agents can be initiated independently of secondary lymphoid organs (SLOs). [bib_ref] Lymphoid organ development: from ontogeny to neogenesis, Drayton [/bib_ref] [bib_ref] Tertiary lymphoid organs in infection and autoimmunity, Neyt [/bib_ref] In this setting, antigen presentation appears to take place in ectopic lymph node-like structures that have been named tertiary lymphoid structures (TLSs). TLSs share many structural characteristics with SLOs. In particular, T and B lymphocytes are segregated into 2 distinct and adjacent regions surrounded by specialized blood vessels called high endothelial venules (HEVs), which allows the extravasation of lymphocytes from the blood to the tissue. The T-cell rich area comprises clusters of T cells and mature dendritic cells (DCs). A B-cell follicle is also present, encompassing a ring of naïve B cells around a germinal center that mainly contains B cells, but also T cells, follicular DCs, and macrophages.
The immunological functions of TLSs in cancer have been poorly investigated. Here, we will review novel insights on the impact of cancer-associated TLSs on the micro-architecture of the tumor microenvironment, i.e., its vasculature and global immune contexture, and how the presence of TLSs ultimately affects patient survival.
## The tumor microenvironment: a permissive milieu for lymphoid neogenesis?
Tumors are sustained by complex networks of interactions between malignant cells , stromal cells and tumor-infiltrating immune cells. These networks differ from patient to patient in terms of nature, composition and organization as well as with regard to the precise localization of tumor-infiltrating cells. Of note, the heterogeneity of the immunological component of the tumor microenvironment, as opposed to its mere abundance, has been shown to influence disease outcome. However, a key question remains: where does the activation of tumor-specific T cells take place? The recently described, tumor-associated lymph node-like entities termed "tertiary lymphoid structures" exhibit a structural organization that is reminiscent of secondary lymphoid organs, and thus may imprint the local immune contexture. Here, we discuss how cancer-associated tertiary lymphoid structures impact on the tumor micro-architecture, immune microenvironment, and ultimately, patient survival. lymphotoxin (LT)α in pancreatic islets (RIP-LTα model). These authors proposed the term "lymphoid neogenesis" for the process governing the development of TLSs. [bib_ref] Chronic inflammation caused by lymphotoxin is lymphoid neogenesis, Kratz [/bib_ref] In contrast to Kratz and colleagues, Moyron-Quiroz et al. observed the genesis of TLSs in a LTα-deficient mice, [bib_ref] Role of inducible bronchus associated lymphoid tissue (iBALT) in respiratory immunity, Moyron-Quiroz [/bib_ref] although these structures appeared less organized than in wild-type animals, primarily consisting of B cells but containing few T cells, no follicular DCs, and no HEVs. These data suggest that LTα participates in the organization of TLSs but it is not an absolute requirement for lymphoid neogenesis. In the RIP-LTα model, LTα induced the expression of multiple lymphoid chemokines, including chemokine (C-C motif) ligand 19 (CCL19), CCL21, and chemokine (C-X-C motif) ligand 13 (CXCL13). [bib_ref] Lymphoid tissue homing chemokines are expressed in chronic inflammation, Hjelmström [/bib_ref] In addition, the ectopic expression of CCL19, CCL21, and CXCL13 promoted the local recruitment of T and B cells into TLS, [bib_ref] Lymphoid neogenesis in chronic inflammatory diseases, Aloisi [/bib_ref] [bib_ref] Ectopic lymphoid tissues and local immunity, Carragher [/bib_ref] and the absence of one of these chemokines was sufficient to abrogate the development of TLSs in the lungs of mice infected by the influenza virus. [bib_ref] Pulmonary expression of CXC chemokine ligand 13, CC chemokine ligand 19, and..., Rangel-Moreno [/bib_ref] Thus, even if each chemokine can act as a TLS inducer, their concomitant expression seems mandatory for a complete program of lymphoid neogenesis. Subsequently, lymphoid chemokines were shown to sustain lymphoid neogenesis by participating in a self-amplificatory loop involving their own expression and that of LTα, as clearly demonstrated in SLOs.
It is well known that lymphoid tissue inducer (LTi) cells play a critical role in the initial phases of lymphoid organogenesis. Thus, many studies have investigated the role of LTi cells in TLS neogenesis. It seems that TLS induction can be initiated independently from LTi cells, as TLSs can emerge in their absence. [bib_ref] Lymphoid neogenesis in chronic inflammatory diseases, Aloisi [/bib_ref] [bib_ref] Role of chemokines in the development of secondary and tertiary lymphoid tissues, Cupedo [/bib_ref] Thus, lymphoid neogenesis is likely to involve additional players that can substitute for LTi cells. Some immune cells have indeed the capacity to activate local resident mesenchymal stromal cells through LTα and tumor necrosis factor α (TNFα) receptors. Moreover, the expression of LTα 1 β 2 by T cells can be induced by CCL21 and CCL19, whereas B cells can be stimulated to secrete LTα 1 β 2 by CXCL13. [bib_ref] A chemokine-driven positive feedback loop organizes lymphoid follicles, Ansel [/bib_ref] [bib_ref] BLC expression in pancreatic islets causes B cell recruitment and lymphotoxin-dependent lymphoid..., Luther [/bib_ref] [bib_ref] Differing activities of homeostatic chemokines CCL19, CCL21, and CXCL12 in lymphocyte and..., Luther [/bib_ref] Nevertheless, the precise stimulus initiating the development of TLSs, in particular cancer-associated TLSs, has not been clearly identified to date.
Several studies showed that TLSs are induced in a chronic inflammatory environment and their presence is associated with the exacerbation of local immune responses. Although the tumor microenvironment presents many similarities with such a setting (including the chronicity of the inflammation and the presence of functional APCs), it differs by at least one major aspect, its immunosuppressive nature. [bib_ref] The three Es of cancer immunoediting, Dunn [/bib_ref] [bib_ref] Cellular constituents of immune escape within the tumor microenvironment, Kerkar [/bib_ref] One could suspect that this immunosuppressive microenvironment would represent a major obstacle for the emergence of TLSs and/or for their functionality.
The presence of intratumoral lymphoid aggregates with varying levels of organization (i.e., comprising or not a B-cell follicle and/or HEVs) has been described in patients affected by primary colorectal cancer, 20-24 breast carcinoma, 24-30 melanoma, 31-34 mucosal-associated lymphoid tissue (MALT) lymphoma, [bib_ref] CXCL13, CCL21, and CXCL12 expression in salivary glands of patients with Sjogren's..., Barone [/bib_ref] ,36 ovarian carcinoma, [bib_ref] Human solid tumors contain high endothelial venules: association with T-and B-lymphocyte infiltration..., Martinet [/bib_ref] and lung cancer 24,37-39 [fig_ref] Table 1: Studies reporting the presence of tertiary lymphoid structures in human neoplasms [/fig_ref]. TLSs have been observed even in the context of metastatic disease, especially in colorectal carcinoma, [bib_ref] Characteristics and clinical impacts of the immune environments in colorectal and renal..., Remark [/bib_ref] renal cell carcinoma, [bib_ref] Characteristics and clinical impacts of the immune environments in colorectal and renal..., Remark [/bib_ref] and melanoma patients, [bib_ref] Neogenesis of lymphoid structures and antibody responses occur in human melanoma metastases, Cipponi [/bib_ref] [bib_ref] 12-Chemokine gene signature identifies lymph node-like structures in melanoma: potential for patient..., Messina [/bib_ref] demonstrating that lymphoid neogenesis can occur and/or be maintained at all disease stages. However, distinct levels of TLS organization have been observed in metastatic melanoma [bib_ref] Neogenesis of lymphoid structures and antibody responses occur in human melanoma metastases, Cipponi [/bib_ref] and primary lung cancer (unpublished data from our laboratory). Such differences may reflect different stages in the program of lymphoid neogenesis, as previously suggested in non-transformed inflammatory settings. 10, [bib_ref] Chronic rejection triggers the development of an aggressive intragraft immune response through..., Thaunat [/bib_ref] Finally, the density of TLSs is highly heterogeneous even among tumors of a given type. This heterogeneity might also reflect the diversity of individuals tumor microenvironments, being more or less permissive to lymphoid neogenesis.
## Tlss and hevs crosstalk in a finely regulated manner within the tumor microenvironment
In many tumors, one of the major hurdles against the efficacy of chemo-and immunotherapy (including anticancer vaccines) is the limited accessibility of immune cells, notably T cells, to the neoplastic lesion. Several mechanisms have been proposed to underlie this phenomenon including the disruption of the architecture and structure of intratumoral blood vessels. The presence of an abnormal vasculature, associated with reduced vessel density, has been reported in multiple cancers, suggesting some sort of disturbance in the neoangiogenetic program. Two main types of blood vessels co-exist in tumors, CD34 + (or CD31 + ) blood vessels and peripheral node addressin (PNAd) + HEVs, each of which can differentiate into the other under physiological and pathophysiological conditions. One study suggested that no correlation exists between the densities of blood vessels and that of HEVs in breast carcinoma, suggesting that their developmental programs can proceed independently from each other. [bib_ref] Human solid tumors contain high endothelial venules: association with T-and B-lymphocyte infiltration..., Martinet [/bib_ref] Strikingly, the intratumoral density of these 2 different types of vessels has an opposite prognostic value in cancer patients. Indeed, high densities of CD34 + (or CD31 + ) blood vessels are associated with tumor growth and poor clinical outcome in breast, [bib_ref] Microvessel density as a prognostic factor in women with breast cancer: a..., Uzzan [/bib_ref] colorectal, [bib_ref] Microvessel density and VEGF expression are prognostic factors in colorectal cancer. Meta-analysis..., Guetz [/bib_ref] and lung [bib_ref] Inverse prognostic impact of angiogenic marker expression in tumor cells versus stromal..., Donnem [/bib_ref] [bib_ref] The clinical significance of lymphangiogenesis and angiogenesis in non-small cell lung cancer..., Kadota [/bib_ref] [bib_ref] Prognostic impact of VEGF, CD31, CD34, and CD105 expression and tumour vessel..., Mineo [/bib_ref] cancer patients. These data support a model according to which neoangiogenesis is stimulated by the tumor microenvironment, promoting the formation of vascular emboli and support the metastatic spread of neoplastic cells to distant sites. In contrast, the presence of HEVs, which have been observed in melanoma [bib_ref] Neogenesis of lymphoid structures and antibody responses occur in human melanoma metastases, Cipponi [/bib_ref] [bib_ref] High endothelial venules (HEVs) in human melanoma lesions: Major gateways for tumor-infiltrating..., Martinet [/bib_ref] and breast cancer [bib_ref] Human solid tumors contain high endothelial venules: association with T-and B-lymphocyte infiltration..., Martinet [/bib_ref] [bib_ref] High endothelial venule blood vessels for tumor-infiltrating lymphocytes are associated with lymphotoxin..., Martinet [/bib_ref] lesions, correlated with long-term patient survival. [bib_ref] Human solid tumors contain high endothelial venules: association with T-and B-lymphocyte infiltration..., Martinet [/bib_ref] [bib_ref] High endothelial venules (HEVs) in human melanoma lesions: Major gateways for tumor-infiltrating..., Martinet [/bib_ref] [bib_ref] Tumor high endothelial venules (HEVs) predict lymphocyte infiltration and favorable prognosis in..., Martinet [/bib_ref] Moreover, the density of HEVs closely correlated with that of tumor-infiltrating T and B cells, [bib_ref] Human solid tumors contain high endothelial venules: association with T-and B-lymphocyte infiltration..., Martinet [/bib_ref] [bib_ref] High endothelial venules (HEVs) in human melanoma lesions: Major gateways for tumor-infiltrating..., Martinet [/bib_ref] [bib_ref] Tumor high endothelial venules (HEVs) predict lymphocyte infiltration and favorable prognosis in..., Martinet [/bib_ref] and HEVs were shown to co-localize with cancer-associated TLSs. [bib_ref] Human solid tumors contain high endothelial venules: association with T-and B-lymphocyte infiltration..., Martinet [/bib_ref] [bib_ref] High endothelial venule blood vessels for tumor-infiltrating lymphocytes are associated with lymphotoxin..., Martinet [/bib_ref] [bib_ref] Neogenesis of lymphoid structures and antibody responses occur in human melanoma metastases, Cipponi [/bib_ref] [bib_ref] High endothelial venules (HEVs) in human melanoma lesions: Major gateways for tumor-infiltrating..., Martinet [/bib_ref] [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] In lung cancer, the PNAd ligand CD62L, is expressed by most lymphocytes found within TLSs (with the exception of germinal center B cells), but not by lymphocytes present in the other areas of the neoplastic lesion. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] HEVs are vessels specialized in the extravasation of lymphocytes from the bloodstream to SLOs. Given the strong analogies between SLOs and TLSs, we suggest that HEVs may actively participate in the recruitment of circulating lymphocytes to intratumoral TLSs. Recently, Girard and colleagues demonstrated that DCs play a key role in the regulation of the entry of lymphocyte into SLOs as they maintain a fully mature HEV endothelium in a LTβ receptor-dependent manner. [bib_ref] Dendritic cells control lymphocyte entry to lymph nodes through high endothelial venules, Moussion [/bib_ref] [bib_ref] HEVs, lymphatics and homeostatic immune cell trafficking in lymph nodes, Girard [/bib_ref] Mature DCs are exclusively detected in the T cell-rich areas of TLSs, adjacent to HEVs. [bib_ref] High endothelial venule blood vessels for tumor-infiltrating lymphocytes are associated with lymphotoxin..., Martinet [/bib_ref] [bib_ref] Neogenesis of lymphoid structures and antibody responses occur in human melanoma metastases, Cipponi [/bib_ref] [bib_ref] High endothelial venules (HEVs) in human melanoma lesions: Major gateways for tumor-infiltrating..., Martinet [/bib_ref] [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] Thus, it is tempting to speculate, but remains to be formally demonstrated, that DCs may regulate lymphocyte trafficking not only in SLOs but also in cancer-associated TLSs.
The mechanisms governing the extravasation of lymphocytes into SLOs are well characterized, and involve 3 main families of molecules: adhesion molecules, chemokines, and integrins. In lung cancer patients, a specific set of adhesion molecules expressed by HEVs have been selectively associated with the presence of TLS: intercellular adhesion molecule (ICAM)2, ICAM3, vascular cell adhesion molecule 1 (VCAM1), mucosal vascular addressin cell adhesion molecule 1 (MAdCAM1) and PNAd. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] In mouse and rat models, intratumoral endothelial cells were shown not to express ICAM1 and VCAM1, as opposed to endothelial cells in distant tissues, [bib_ref] Tumor microvasculature as a barrier to antitumor immunity, Chen [/bib_ref] [bib_ref] Diminished leukocyte-endothelium interaction in tumor microvessels, Wu [/bib_ref] in accordance with the downregulation of VCAM1 on the surface of endothelial cells cultured in presence of neoplastic cells. [bib_ref] Endothelial vascular cell adhesion molecule 1 expression is suppressed by melanoma and..., Piali [/bib_ref] HEVs expressing ICAM1, VCAM1, MAdCAM1, and PNAd have been also observed in the close proximity of lymphoid aggregates in LT-induced chronic inflammatory lesions. [bib_ref] Chronic inflammation caused by lymphotoxin is lymphoid neogenesis, Kratz [/bib_ref] Moreover, the expression of ICAM2, VCAM1, and PNAd has also been observed in lymphoid aggregates found in non-inflamed human lungs, [bib_ref] Expression of endothelia and lymphocyte adhesion molecules in bronchus-associated lymphoid tissue (BALT)..., Kawamata [/bib_ref] [bib_ref] Lymphocyte homing to bronchus-associated lymphoid tissue (BALT) is mediated by L-selectin/PNAd, alpha4beta1..., Xu [/bib_ref] suggesting that TLSs could represent a favorable microenvironment for the differentiation and maintenance of blood vessels into functional HEVs. ICAM2 is involved in the arrest of circulating T cells along the HEV endothelium of SLOs, [bib_ref] Lymphocyte trafficking across high endothelial venules: dogmas and enigmas, Miyasaka [/bib_ref] and may have the same function within TLSs. VCAM1 is induced upon inflammation, and contributes to the migration of circulating memory T cells to tissues. [bib_ref] Integrin function in T-cell homing to lymphoid and nonlymphoid sites: getting there..., Denucci [/bib_ref] Thus, VCAM1 may represent a key molecule for the recruitment of memory T cells, which are the main T-cell components of lung cancer-associated TLSs. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] In contrast, MAdCAM1 is a gut-homing molecule that regulates T-cell migration, [bib_ref] T-cell recruitment to the intestinal mucosa, Agace [/bib_ref] and has not previously shown to contribute to TLS. This suggests that HEVs from lung cancer-associated TLSs exhibit an adhesion molecule profile that differ from that of TLSs surging in non-neoplastic settings.
Altogether, these observations corroborate the hypothesis that TLSs may represent a privileged site for the maturation and maintenance of functional HEVs, unlike the other areas of the tumor microenvironment. TLS-serving HEVs may provide a gateway for the recruitment of circulating T cells into the tumor, and thus represent an interesting target for anticancer immunotherapy [fig_ref] Figure 1: role of tertiary lymphoid structures in the initiation of local and systemic... [/fig_ref].
## Chemokines, an orchestra conductor of tls organization
The migration of lymphocytes into SLOs is dependent on the expression of various lymphoid chemokines, including CCL19, CCL21, and CXCL13. [bib_ref] Ectopic lymphoid tissues and local immunity, Carragher [/bib_ref] [bib_ref] Role of chemokines in the development of secondary and tertiary lymphoid tissues, Cupedo [/bib_ref] [bib_ref] Chemokines and cell migration in secondary lymphoid organs, Cyster [/bib_ref] The overexpression of these chemokines has been documented in the TLSs of lung and breast carcinoma patients. [bib_ref] CD4 + follicular helper T cell infiltration predicts breast cancer survival, Gu-Trantien [/bib_ref] [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] Moreover, in several murine models, the injection of cancer cells engineered to express either CCL19 and/or CCL21 has been shown to induce a robust infiltration of developing neoplastic lesions by T cells and DCs. [bib_ref] Activated local immunity by CC chemokine ligand 19-transduced embryonic endothelial progenitor cells..., Hamanishi [/bib_ref] [bib_ref] Local expression of secondary lymphoid tissue chemokine delivered by adeno-associated virus within..., Liang [/bib_ref] [bib_ref] Antitumor effects of the mouse chemokine 6Ckine/SLC through angiostatic and immunological mechanisms, Vicari [/bib_ref] [bib_ref] Intratumoral administration of dendritic cells overexpressing CCL21 generates systemic antitumor responses and..., Yang [/bib_ref] [bib_ref] Intrapulmonary administration of CCL21 genemodified dendritic cells reduces tumor burden in spontaneous..., Yang [/bib_ref] This has been associated with a reduction in tumor growth and, at least in some cases, with the induction of a systemic antitumor immune response linked to the control of metastatic dissemination. [bib_ref] Activated local immunity by CC chemokine ligand 19-transduced embryonic endothelial progenitor cells..., Hamanishi [/bib_ref] [bib_ref] Regulated expression of CCL21 in the prostate tumor microenvironment inhibits tumor growth..., Yousefieh [/bib_ref] In human lung cancers, CCL19 + cells are predominantly mature DCs, which are selectively present in TLS. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] In the lung of mice infected by the influenza virus, the depletion of DCs correlated with a dramatic decrease in TLS density, and the few remaining TLSs were significantly disorganized. [bib_ref] Dendritic cells are crucial for maintenance of tertiary lymphoid structures in the..., Geurtsvankessel [/bib_ref] [bib_ref] Induced bronchus-associated lymphoid tissue serves as a general priming site for T..., Halle [/bib_ref] The maintenance of TLSs may be partly mediated by DCs via the recruitment of CCR7 + cells, including newly activated DCs and naïve and central-memory T cells, as we observed in lung cancer-associated TLSs. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] Recently, DCs have also been shown to play a critical role in the induction of lymphangiogenesis within TLSs in a LT-and CCL21-dependent manner. [bib_ref] A critical role for dendritic cells in the formation of lymphatic vessels..., Muniz [/bib_ref] As observed in various inflammatory diseases, lung cancer-associated lymphatic vessels express CCL21. Thus, it is tempting to speculate that CCL21 + lymphatic vessels in the proximity of TLS may represent a major path for immune cells (e.g., activated DCs and central-memory T cells) to reach tumor-draining lymph nodes and establish a systemic protection against metastatic dissemination.
In human lung cancers, CXCL13 was detected on follicular DCs that homed specifically to the germinal centers of B-cell areas within TLSs. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] Therein, follicular DCs co-localized with a subset of TLS T cells expressing CD4 and chemokine (C-X-C motif) receptor 5 (CXCR5), hence displaying a phenotype of follicular helper T cells. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] Interestingly, CXCL13 + follicular helper T cells have recently been detected within breast carcinomaassociated TLSs, and their density was associated with long-term patient survival. [bib_ref] CD4 + follicular helper T cell infiltration predicts breast cancer survival, Gu-Trantien [/bib_ref] The density of CXCR5 + follicular B cells also appears to correlated with a favorable clinical outcome among lung cancer patients (Dieu-Nosjean et al., unpublished data) suggesting that the expression of CXCL13 in cancer-associated TLSs may promote the local initiation of a protective humoral immune response through the recruitment of CXCR5 + cells into the germinal center.
CCL17 and CCL22 are also overexpressed in lung cancerassociated TLSs, in agreement with the higher proportion of CCR4 + T cells among TLS-associated T cells as compared with T cells found in other areas of the neoplastic lesion. 38 CCL17 and CCL22 operate by a shared receptor (CCR4), which is expressed by many T-cell subpopulations including T H 2, T H 17, T H 22 T cells, regulatory T cells (Tregs), and effector T cells under inflammatory conditions. [bib_ref] Dissecting the human immunologic memory for pathogens, Zielinski [/bib_ref] Nonetheless, the role of the CCL17-CCL22/CCR4 in cancer remains controversial. In some murine models, the intratumoral injection of CCR4 ligands induces tumor regression with a concomitant recruitment of effector T cells, [bib_ref] TARC and RANTES enhance antitumor immunity induced by the GM-CSFtransduced tumor vaccine..., Inoue [/bib_ref] [bib_ref] CC-chemokine ligand 17 gene therapy induces tumor regression through augmentation of tumor-infiltrating..., Kanagawa [/bib_ref] [bib_ref] Targeted knock down of CCL22 and CCL17 by siRNA during DC differentiation..., Kang [/bib_ref] [bib_ref] Tumor suppressive efficacy through augmentation of tumor-infiltrating immune cells by intratumoral injection..., Okada [/bib_ref] whereas in others studies a deleterious effect associated with a strong influx of Tregs has been reported. [bib_ref] Targeted knock down of CCL22 and CCL17 by siRNA during DC differentiation..., Kang [/bib_ref] [bib_ref] CCL17 and CCL22 chemokines within tumor microenvironment are related to accumulation of..., Mizukami [/bib_ref] [bib_ref] CCL22 recruits CD4-positive CD25-positive regulatory T cells into malignant pleural effusion, Qin [/bib_ref] However, it should be noted that the prognostic value of tumor-infiltrating Tregs in cancer patients is also a matter of debate. 2 This is likely due to the limited specificity of biomarkers for human Tregs, such as the transcription factor forkhead box P3 (FOXP3), which is also an indicator of the T-cell activation status. shown that Tregs are present in the stroma of breast tumors as well as within intratumoral lymphoid aggregates. [bib_ref] Regulatory T cells recruited through CCL22/CCR4 are selectively activated in lymphoid infiltrates..., Gobert [/bib_ref] The migration of Tregs to lymphoid structures is mediated by the CCL22/ CCR4 axis. Tregs can be activated by mature DCs, and their high density in lymphoid aggregates is correlated with short-term survival of breast cancer patients. Murine models have illustrated that a high density of intratumoral Tregs correlates with low densities of HEVs, suggesting that HEV neogenesis is inhibited by Tregs. [bib_ref] T-cell trafficking facilitated by high endothelial venules is required for tumor control..., Hindley [/bib_ref] [bib_ref] Limited tumor infiltration by activated T effector cells restricts the therapeutic activity..., Quezada [/bib_ref] However, in melanoma patients, no correlation was observed between the density of HEVs and that of Tregs. [bib_ref] High endothelial venules (HEVs) in human melanoma lesions: Major gateways for tumor-infiltrating..., Martinet [/bib_ref] This apparent discrepancy suggest that further investigations are required in order to better define the role of TLS-associated Tregs in oncogenesis and tumor progression.
Interleukin (IL)-16 is also expressed by T and B cells infiltrating lung cancer-associated TLSs. [bib_ref] Characterization of chemokines and adhesion molecules associated with T cell presence in..., De Chaisemartin [/bib_ref] In an asthma model, IL-16 has been shown to operate as a chemoattractant for various CD4 + cells, including CD4 + T cells, DCs and myeloid cells. [bib_ref] B lymphocyte-derived IL-16 attracts dendritic cells and Th cells, Kaser [/bib_ref] can also prime CD4 + T cells to respond to IL-2 by proliferating, becoming activated and/or differentiating into memory cells. [bib_ref] Synergistic activation of CD4+ T cells by IL-16 and IL-2, Parada [/bib_ref] Thus, the role of IL-16 within TLSs may be of particular importance for the functionality of tumor-infiltrating T lymphocytes.
Thus, the expression of a defined set of chemokines by cancerassociated TLSs strongly suggests that some immune cell populations could be selectively recruited into TLSs with a putative key impact on the cell-to-cell cross-talk and global intratumoral immune contexture.
## The presence of tls impacts on the local immune microenvironment
The selective recruitment of immune cells, especially naïve and central-memory CD62L + T cells, to TLSs led us to hypothesize that TLSs may represent a privileged site for the differentiation and activation of tumor-infiltrating lymphocytes. To obtain further insights into these issues, we have compared the characteristics of the local immune microenvironment with the density of TLSs in patients with lung cancer (Goc et al., manuscript submitted). Patients were stratified into two groups according to the presence of a high or a low density of mature DCs, a cell population that is selectively detected within TLSs and hence is used as a specific TLS marker. Tumor bearing high densities of mature DCs (DC hi tumors) contained increased amounts of CD4 + and CD8 + T cells (encompassing naïve, central-memory and effector-memory cells) as well as of activated T cells, as compared with tumors scarcely infiltrated by mature DCs (DC lo tumors). Elevated TLS densities also correlated with the overexpression as well as the coordination of genes linked to T H 1 polarization, cytotoxicity, activation, and immune effector functions. In contrast, genes involved in T H 2 polarization, inflammation, immunosuppression, and angiogenesis were not differentially expressed by DC hi and DC lo tumors. These data indicate that high TLS densities are associated with a specific intratumoral contexture that is characterized by the coordination of adaptive cellular immune responses [fig_ref] Figure 1: role of tertiary lymphoid structures in the initiation of local and systemic... [/fig_ref]. This suggests that TLS might directly impact on the intratumoral immune contexture in lung cancer patients. Similar findings have been reported in lung metastases of colorectal carcinoma but not renal cell carcinoma, in agreement with the differential density of TLSs found in these cancer types. [bib_ref] Characteristics and clinical impacts of the immune environments in colorectal and renal..., Remark [/bib_ref] Of note, the protective effects of a T H 1 and cytotoxic immune signature have been widely demonstrated in mice, [bib_ref] Cancer immunoediting: integrating immunity's roles in cancer suppression and promotion, Schreiber [/bib_ref] and correlate with a favorable disease outcome in patients affected by most solid cancers (exception made for renal cell carcinoma and ocular melanoma 2 ). However, the possibility that such immune signature may be associated with the density of TLSs has never been reported.
One of the remaining key questions is whether TLSs are directly involved in the induction of antitumor immune responses, or, vice versa, whether they relay locally an intense immune reaction initiated in SLOs. Several lines of evidence support the former possibility. First, it has been demonstrated that antitumor immune responses can develop within neoplastic lesions in mice lacking SLOs. This demonstrates that effective T-cell priming can be achieved within intratumoral TLSs independently of SLOs. [bib_ref] The dynamics of the T-cell antitumor response: chemokine-secreting dendritic cells can prime..., Kirk [/bib_ref] [bib_ref] Immunological tumor destruction in a murine melanoma model by targeted LTalpha independent..., Schrama [/bib_ref] Second, the presence of TLSs has been correlated with the generation of tumor-specific T cells, suggesting that TLSs could act as a functional site for the induction of protective antitumor immunity. [bib_ref] Immunological tumor destruction in a murine melanoma model by targeted LTalpha independent..., Schrama [/bib_ref] [bib_ref] Targeting of lymphotoxin-alpha to the tumor elicits an efficient immune response associated..., Schrama [/bib_ref] Similar findings have also been obtained in models of infectious diseases, [bib_ref] Role of inducible bronchus associated lymphoid tissue (iBALT) in respiratory immunity, Moyron-Quiroz [/bib_ref] [bib_ref] Induced bronchus-associated lymphoid tissue serves as a general priming site for T..., Halle [/bib_ref] indicating that TLSs may represent an active site for the induction of adaptive immune responses in multiple settings.
In summary, it seems that the presence of a favorable (or at least a permissive) tumor microenvironment allows for the creation of TLSs, which is critical for the local coordination and polarization of protective antitumor immune responses.
## Tls as a novel biomarker in cancer
The majority of prognostic studies has highlighted a positive impact of intratumoral T cells on the survival of patients affected by solid tumors. [bib_ref] The immune contexture in human tumours: impact on clinical outcome, Fridman [/bib_ref] [bib_ref] Trial watch: Prognostic and predictive value of the immune infiltrate in cancer, Senovilla [/bib_ref] The high reproducibility of these data strongly support the idea that one should take into account immune criteria for the prognostic staging of neoplastic lesions. [bib_ref] From the immune contexture to the Immunoscore: the role of prognostic and..., Angell [/bib_ref] Several studies have indeed reported a favorable clinical outcome for lung cancers patients whose neoplastic lesions contain a high density of CD8 + T cells (Refs. 86,87 and Goc et al., submitted manuscript). Moreover, a high density of mature DCs (a specific marker of TLS) has been correlated with the long-term survival of patients affected by early-stage, advanced and metastatic lung cancer (Refs. 37,40 and Goc et al., submitted manuscript).
Thus the survival of lung cancer patients appears to be positively influenced by the abundance of tumor-infiltrating mature DCs and CD8 + T cells. This raises the question of evaluating the clinical impact of these tumor-infiltrating cells in combination. The stratification of lung cancer patients (at all disease stages) according to low or high densities of intratumoral mature DCs and CD8 + T cells (DC lo /CD8 lo , DC lo /CD8 hi , DC hi /CD8 lo , and DC hi /CD8 hi patients) provided several interesting hints (Goc et al., submitted manuscript). First, patients with a high density of mature DCs, regardless of the density of CD8 + T cells (DC hi / CD8 lo and DC hi /CD8 hi groups), were at lower risk of death than their DC lo counterparts, indicating that elevated amounts of intratumoral mature DCs are sufficient to favorably impact on clinical outcome. This said, it should be noted that most of DC hi tumors are also highly infiltrated by CD8 + T cells. In this context, the scarcity of DC hi /CD8 lo tumors suggests a causal link between the presence of TLSs and that of tumor-infiltrating CD8 + T cells. This observation further supports the hypothesis that intratumoral TLSs represent an active site for the recruitment, activation, proliferation, and priming of tumor-infiltrating T cells. Second, patients with low intratumoral amounts of both mature DCs and CD8 + T cells (DC lo /CD8 lo group) turned out to have a poor clinical outcome as compared with the three other groups of patients (DC lo /CD8 hi , DC hi /CD8 lo , and DC hi /CD8 hi patients). This demonstrates that the DC/CD8 score allows for the identification of a subgroup of patients with a very high risk of death. Finally, patients with a high density of tumor-infiltrating CD8 + T cells but a low density of mature DCs (DC lo /CD8 hi individuals) exhibited an intermediate risk of death. These data suggest that a high density of tumor-infiltrating CD8 + T cells in the absence of a robust tumor infiltration by mature DCs is not sufficient to predict favorable disease outcome among lung cancer patients. The presence of mature DCs appears therefore to be required to license the positive prognostic value of tumorinfiltrating CD8 + T cells. In absence of TLSs, high density of intratumoral CD8 + T cells may be less efficient in maintaining long-term protective antitumor immunity. Similar results have been obtained with regard to tumor-infiltrating B cells and CD8 + T cells in ovarian cancer patients. [bib_ref] CD20+ tumor-infiltrating lymphocytes have an atypical CD27-memory phenotype and together with CD8+..., Nielsen [/bib_ref] [bib_ref] Tumor-infiltrating B cells and T cells: Working together to promote patient survival, Nielsen [/bib_ref] The presence of TLSs may imprint the behavior of tumorinfiltrating CD8 + T cells. This assumption is in agreement with the involvement of TLSs in the local activation of CD8 + T cells, which has been demonstrated in different murine models. [bib_ref] Induced bronchus-associated lymphoid tissue serves as a general priming site for T..., Halle [/bib_ref] [bib_ref] Targeting of lymphotoxin-alpha to the tumor elicits an efficient immune response associated..., Schrama [/bib_ref] [bib_ref] Tumor masses support naive T cell infiltration, activation, and differentiation into effectors, Thompson [/bib_ref] Several mechanisms might explain the survival benefit related to elevated densities of cancer-associated TLSs. Mature DCs, which may have engulfed a large spectrum of tumor-associated antigens (TAAs), are very efficient at presenting processed peptides to antigen-specific T cells [fig_ref] Figure 1: role of tertiary lymphoid structures in the initiation of local and systemic... [/fig_ref]. In addition, the recruitment of naïve T cells into TLSs is highly favorable for the continuous activation of new T-cell clones specific for newly arising TAAs [fig_ref] Figure 1: role of tertiary lymphoid structures in the initiation of local and systemic... [/fig_ref]. Moreover, the local stimulation of intratumoral T cells by mature DCs might potentiate their effector functions and limit their sensitivity to anergy and/or exhaustion. [bib_ref] T cell exhaustion, Wherry [/bib_ref] [bib_ref] Antigen-specific CD4 T-cell help rescues exhausted CD8 T cells during chronic viral..., Aubert [/bib_ref] Finally, CD4 + T cells, which represent the major T-cell subset of TLSs, could deliver important help signals to CD8 + T cells. In a murine model of pancreatic cancer, it has been illustrated that the "help" coming from CD4 + T cells can also limit the induction of tolerance and the depletion of CD8 + T cells, promote the survival of effector-memory T cells and optimize their effector activity. [bib_ref] Tumor-infiltrating B cells and T cells: Working together to promote patient survival, Nielsen [/bib_ref] [bib_ref] CD4+ T-cell help in the tumor milieu is required for recruitment and..., Bos [/bib_ref] It should be noted that the improved prognosis associated with a high density of tumor-infiltrating mature DCs was observed among lung cancer patients who underwent surgical tumor resection, and hence who were deprived of all tumorinfiltrating immune cells. This suggests that the prognostic value of intratumoral mature DCs is most likely related to the development of systemic antitumor immunity, in turn originating from local adaptive immune responses orchestrated within TLSs. As such, the presence of lymphatic vessels close to TLSs may participate in the emigration of some immune cell populations, including central-memory T cells, to tumor-draining lymph nodes, a key pathway for the establishment of robust systemic immunity against micro-metastasis [fig_ref] Figure 1: role of tertiary lymphoid structures in the initiation of local and systemic... [/fig_ref].
# Conclusions
The presence of TLSs has been documented in many solid tumors, in both mice and humans. Interestingly, TLSs manifest a huge variability in terms of density and cellular organization, indicating that the nature of tumor microenvironment is critical for lymphoid neogenesis. In this context, one of the challenges for the future is to precisely determine the conditions that provide an optimal environment for the induction and maintenance of cancer-associated TLSs.
In addition, the presence of particular vessels, namely, HEVs, around TLSs strongly suggests that this microenvironment represents an ideal gateway for the entry of circulating lymphocytes into TLSs, a step that is critical for the initiation of adaptive immune responses, as demonstrated in SLOs. Moreover, TLSs can imprint the local immune contexture, hence influencing disease outcome in cancer patients. Thus, the density of CD8 + T cells predicts long-term survival among lung cancer patients only in presence of TLS-associated DCs. These data need to be confirmed in patients affected by other solid tumors. Nonetheless, the presence of TLSs provides per se a survival benefit to cancer patients. We speculate that lymphoid neogenesis might represent an interesting target for eliciting and/or boosting antitumor immunity in cancer patients. Moreover, TLSs may be used in the future as a novel biomarker for the identification of cancer patients with a high risk of relapse. A deeper comprehension of the cellular and molecular mechanisms whereby TLSs confer long-term protection to cancer patients is essential to develop efficient strategies for their therapeutic manipulation.
## Disclosure of potential conflicts of interest
No potential conflicts of interest were disclosed.
[fig] Figure 1: role of tertiary lymphoid structures in the initiation of local and systemic protective immune responses against primary neoplastic lesions and metastases. DC, dendritic cell; HeV, high endothelial venule; SLO, secondary lymphoid organ; TLS, tertiary lymphoid structure. [/fig]
[table] Table 1: Studies reporting the presence of tertiary lymphoid structures in human neoplasms [/table]
|
Alcohol and Cardiovascular Diseases—Do the Consumption Pattern and Dose Make the Difference?
# Introduction
Alcohol is one of the most important risk factors for mortality. Recent studies conducted in 189 countries for the purposes of the Global Burden of Disease project show that the three main factors of health loss (the so-called DALY indicator-the disabilityadjusted life years) are, respectively: hypertension (7%), smoking (6%) and alcohol drinking (5.5%) [bib_ref] Alcohol Collaborators. Alcohol use and burden for 195 countries and territories, 1990-2016:..., Griswold [/bib_ref]. Alcohol consumption is a causal factor in more than 200 diseases (especially non-infectious diseases) and injuries. The risk for disease development increases in proportion to the dose of alcoholic beverage consumed. On a global scale, global alcohol intake is showing a steady upward trend-from 5.9 L of pure alcohol per year per adult in 1990 to 6.5 L in 2017, and it is expected to increase to 7.6 L by 2030 [bib_ref] Global alcohol exposure between 1990 and 2017 and forecasts until 2030: A..., Manthey [/bib_ref]. Estimates suggest that by 2030, half of adults will drink alcohol on a regular basis and almost a quarter (23%) will drink at least once a month. In most regions of the world, especially in low-income countries, the amount of alcohol consumed is currently growing faster than the number of drinkers, which not only increases the percentage of heavy drinking episodes, but also inevitably leads to an increasing burden of alcohol-related diseases [bib_ref] Global alcohol exposure between 1990 and 2017 and forecasts until 2030: A..., Manthey [/bib_ref]. Although Mendelian randomized, controlled trials show that abstainers have the lowest risk of cardiovascular events and that any dose of alcohol may be harmful, the above data show that alcohol has deeply embedded in the lifestyle of many societies and there is no indication that it will change even in the distant future [bib_ref] Conventional and genetic evidence on alcohol and vascular disease aetiology: A prospective..., Millwood [/bib_ref] [bib_ref] InterAct Consortium. Association between alcohol and cardiovascular disease: Mendelian randomisation analysis based..., Holmes [/bib_ref]. If this phenomenon has to be accepted, then it may be worthwhile to have a closer look at the results of many clinical and observational studies on what dose and pattern of alcohol consumption are most harmful to health and emphasize that the types of alcohol are unequal in terms of their cardiovascular effects.
## Wine in moderation-the joy of the soul and heart
This sentence expressed in the biblical Book of Sirach is still valid in spite of the passage of several thousand years. In 1926, American biologist Raymond Perl showed that people who drink moderate amounts of alcohol have a longer lifespan than both non-drinkers and heavy drinkers. He was the first researcher to observe the U-shaped relationship between mortality and the amount of alcohol consumed.
The published results of a 13-year prospective observational study involving a group of 12,000 British physicians clearly indicated the beneficial effects of moderate alcohol consumption on the cardiovascular system [bib_ref] Mortality in relation to consumption of alcohol: 13 years' observation on male..., Doll [/bib_ref]. These observations were confirmed by a 7-year observation of the German population that participated in the German National Health Surveys. The lowest risk of death from cardiovascular causes and death from all causes was observed in the group of men consuming up to 20 g of alcohol per day. Compared with abstainers, the risk was lower by nearly 50% [bib_ref] The relationship between alcohol consumption, health indicators and mortality in the German..., Hoffmeister [/bib_ref].
Part of the observational Seven Countries study of the Italian population living in rural regions also showed a link between moderate, regular alcohol consumption and life expectancy [bib_ref] InterAct Consortium. Association between alcohol and cardiovascular disease: Mendelian randomisation analysis based..., Holmes [/bib_ref]. In the group of men aged 45-64 consuming up to 60 g of alcohol per day, mainly wine, the expected life expectancy from the time of inclusion in the study was over 21 years and was higher by an average of 2 years, compared to abstainers and alcohol abusers (>10 "drinks" per day). In men (1536 Italian males, aged 30-69, 30 years-1096 deaths), the model assessing the number of deaths related to coronary heart disease, which were prevented by moderate consumption of wine and the number of coronary deaths caused by consumption of excessive amounts of ethanol. This model showed that low-to-moderate amounts of alcohol prevented about 400 deaths due to coronary heart disease each year, which resulted in a 14% reduction in mortality. During the same period, 20 coronary deaths were associated with alcohol abuse [bib_ref] Alcohol and survival in the Italian rural cohorts of the Seven Countries..., Farchi [/bib_ref].
In the majority of observational prospective studies and meta-analyses, it was shown that the relationship between alcohol consumption and the incidence of coronary artery disease and stroke took the shape of the letters U or J on the graph. The lowest risk of cardiovascular disease and coronary artery disease was observed in people consuming moderate amounts of alcohol. This risk increased especially in people abusing alcohol, but also in non-drinkers. There was a higher risk of falling ill and dying, mainly due to injuries, poisoning, suicide, liver cirrhosis, cancer and stroke observed in heavy drinkers [bib_ref] Alcohol consumption and mortality among middle-aged and elderly U.S. adults, Thun [/bib_ref] [bib_ref] Alcohol consumption, heavy drinking, and mortality: Rethinking the j-shaped curve, Plunk [/bib_ref] [bib_ref] Association of alcohol consumption with selected cardiovascular disease outcomes: A systematic review..., Ronksley [/bib_ref] [bib_ref] Alcohol and coronary heart disease: A meta-analysis, Corrao [/bib_ref]. This relationship was well-illustrated by the results of the American Cancer Prevention Study II, which evaluated drinking habits and other lifestyle-related factors in a group of nearly half a million American adults [bib_ref] Alcohol consumption and mortality among middle-aged and elderly U.S. adults, Thun [/bib_ref]. At the same time, the causes of deaths in this group were analyzed during the next nine years of observation. The lowest risk of death was observed in persons of both sexes who consumed 1-2 portions of alcohol daily [bib_ref] The relationship between alcohol consumption, health indicators and mortality in the German..., Hoffmeister [/bib_ref]. In the meta-analysis by Ronksley et al., the authors showed an increased risk of death and a higher incidence of stroke with daily alcohol consumption in the range of 50-60 g and more, whereas lower doses of alcohol had a beneficial effect in virtually all studies analyzed [bib_ref] Association of alcohol consumption with selected cardiovascular disease outcomes: A systematic review..., Ronksley [/bib_ref].
Through the analyses of data from nine representative surveys from the districts keeping the National Health Interview Survey (NHIS) in the United States in 1987-2000 and including data of 245,207 people, it was found that low and moderate alcohol consumption (slightly more than one drink daily) may be a factor that reduces the risk of cardiovascular death. Among those who were lifelong abstainers or had been rare-drinkers in the past and those who stopped drinking, the risk of dying from cardiovascular disease was higher than in moderate drinkers [bib_ref] Alcohol consumption and cardiovascular mortality among U.S. adults, Mukamal [/bib_ref].
One of the effects of alcohol consumed in low doses may be also a reduced risk of sudden cardiac death. The US Physicians Health Study, which evaluated patients without diagnosed coronary artery disease, showed that the relative risk of sudden cardiac death was 0.4 and 0.2 in men consuming 2-4 or 5-6 portions of alcohol per week compared to non-drinkers; this risk was higher and approached unity when consuming two portions of alcohol a day. A typical U-shaped curve of incident dependence on alcohol consumption was observed here [bib_ref] Moderate alcohol consumption and the risk of sudden cardiac death among US..., Albert [/bib_ref]. The mechanism of alcohol action in sudden cardiac death remains incompletely understood, but it is thought to be associated with a reduction in the risk of ventricular arrhythmia induced by acute coronary episodes [bib_ref] Light-to-moderate alcohol consumption and risk of sudden cardiac death in women, Chiuve [/bib_ref]. In spite of the accepted knowledge that ethanol acutely modulates numerous targets in cardiomyocytes, including ion channels, Ca 2+ -handling proteins and gap junctions, the actual impact of ethanol on cardiac electrophysiology and reentrant arrhythmias is still unclear. Sutanto et al. postulate ethanol concentration-dependent electrophysiologic phenomenon eliciting both the inducibility and stability of reentrant arrhythmias, particularly in the presence of disease-associated remodeling. Another interesting observation was an anti-fibrillatory effect of low levels of ethanol within atrial muscle [bib_ref] Acute effects of alcohol on cardiac electrophysiology and arrhythmogenesis: Insights from multiscale..., Sutanto [/bib_ref].
It is difficult to estimate the level of alcohol consumption that would be associated with the strongest protective effect against coronary artery disease, as different measures and definitions of alcohol consumption were used in different studies. According to Mukamal et al., about 30 g of ethanol consumed on a regular basis can contribute to reducing the risk of death, coronary heart disease or stroke [bib_ref] Alcohol consumption and cardiovascular mortality among U.S. adults, Mukamal [/bib_ref]. A meta-analysis of 51 studies published by showed that the risk of coronary heart disease was the lowest when consuming 20-25 g of ethanol per day [bib_ref] Alcohol and coronary heart disease: A meta-analysis, Corrao [/bib_ref].
Most researchers claim that the strongest protective effect is associated with the consumption of 20-30 g of alcohol per day, which corresponds to 2-3 portions of alcoholic beverages, the so-called drinks. Usually one unit of alcohol (10 g) is taken as 240-300 mL 5% alcohol by volume of beer, 100-125 mL 12% alcohol by volume wine or 30-37.5 mL 40% alcohol by volume spirits. Heavy drinking is defined as a long-term, high-dose intake of >60 g/day in men and >40 g/day in women [bib_ref] Cardiovascular risks and benefits of moderate and heavy alcohol consumption, Fernández-Solà [/bib_ref] [bib_ref] Consensus and Controversy in the Debate over the Biphasic Impact of Alcohol..., Stătescu [/bib_ref]. "Standard drink" varies considerably in different countries. For example, in the UK, one standard unit equals 8 g of pure alcohol, in the USA it amounts to 14 g per standard drink and in Canada it is 3.6 g [bib_ref] Governmental standard drink definitions and low-risk. Alcohol consumption guidelines in 37 countries, Kalinowski [/bib_ref] [bib_ref] Alcohol's Impact on the Cardiovascular System, Roerecke [/bib_ref].
Based on a meta-analysis of 42 studies, Rimm et al. showed a beneficial relationship between the consumption of 30 g of ethanol and changes in high-density lipoprotein (HDL) cholesterol, apolipoprotein I, triglycerides and selected hemostatic risk factors that can reduce the risk of coronary heart disease by nearly 25% [bib_ref] Moderate alcohol intake and lower risk of coronary heart disease: Meta-analysis of..., Rimm [/bib_ref]. The beneficial effects of moderate doses of alcohol on the cardiovascular system in people with a cardiovascular history were confirmed by the results of a large meta-analysis from 54 publications by Costanzo et al. Eight prospective studies on patients with a history of cardiovascular diseases (CVD) were selected from them. Alcohol was shown to be protective in patients with documented cardiovascular disease, as alcohol consumption at low and moderate doses (5-26 g/day) was clearly associated with a reduction in both general and cardiovascular mortality. The curve of this relationship was shaped like the letter J [bib_ref] Alcohol consumption and mortality in patients with cardiovascular disease: A meta-analysis, Costanzo [/bib_ref]. The same authors in the metaanalysis of 16 prospective studies on the relationship between wine consumption and the occurrence of cardiovascular events reported the most beneficial effect for the dose of 21 g ethanol/d, which remained up to 72 g/d. The risk of death from cardiovascular causes was the lowest with a daily consumption of wine containing 24 g of ethanol, and the reduction in the risk of death from other causes was the highest for 10 g/d. The analysis of spirit drinks did not confirm the existence of a beneficial relationship [bib_ref] Wine, beer or spirit drinking in relation to fatal and non-fatal cardiovascular..., Costanzo [/bib_ref].
On the contrary, Wood et al. analyzed the individual-participant data from 599,912 current drinkers without previous cardiovascular diseases from three large-scale data sources in 19 high-income countries. They found that the threshold for lowest risk of all-cause mortality was about 100 g of alcohol/week. For cardiovascular diseases other than myocardial infarction (MI), there were no clear risk thresholds below which lower alcohol consumption stopped being associated with lower disease risk. This analysis shows the lower limits for alcohol consumption than those which are recommended in most current guidelines [bib_ref] Risk thresholds for alcohol consumption: Combined analysis of individual-participant data for 599,912..., Wood [/bib_ref].
Although the interconnection between alcohol use and cardiovascular disease is affected by several behavioral, genetic and biological factors, the dose and frequency of alcohol consumption seem to have the greatest influence. Nevertheless, any positive aspects of low-to-moderate drinking must be balanced against significant physiological consequences [bib_ref] Alcohol's Effects on the Cardiovascular System, Piano [/bib_ref].
Optimal, light/moderate alcohol consumption means that you need to remain at the lowest point of the J-shaped curve for alcohol use in order to take advantage of potential cardioprotection, which may appear difficult, since alcohol is the most popular addictive.
So, is the type of alcohol consumed important? A Scottish prospective cohort study including 5766 men aged 35-64 during a 21-year follow-up, after considering the effects of classic risk factors, showed no association between high-percentage alcohol consumption and reduced ischemic mortality. The risk of death from all causes was greatest for people consuming over 22 units of alcohol per week. The study also showed a doubling of the risk of death from stroke in a group of participants consuming over 35 units of alcohol per week. It should be emphasized that in the analyzed population the main sources of ethanol were whiskey and gin, in contrast to the population of France and southern Europe, where red wine occupies the dominant position among alcoholic beverages and in which the lowest cardiovascular mortality in Europe is observed [bib_ref] Alcohol consumption and mortality from all causes, coronary heart disease, and stroke:..., Hart [/bib_ref] [bib_ref] The French paradox: Lessons for other countries, Ferrière [/bib_ref].
Interpreting the results of observational studies, it should be borne in mind that it is impossible to define a clear cause and effect relationship between alcohol consumption and reduction in cardiovascular morbidity and mortality. Observational studies are mainly based on data from questionnaires, through which participants independently assess the amount or frequency of alcoholic beverages consumed [bib_ref] Alcohol volume, drinking pattern, and cardiovascular disease morbidity and mortality: Is there..., Murray [/bib_ref]. This may be associated with an underestimation of consumption in the group of people who actually abuse alcohol. In addition, abstainers, who in most epidemiological studies appear to be at greater risk of coronary incidents and death from cardiovascular causes than those who consume moderate amounts of alcohol, are also a group of non-drinkers based on the disease history.
## Holiday heart syndrome-alcohol and atrial fibrillation
In 1987, Ettinger et al. designed the study to assess the impact of weekend and holiday alcohol consumption on the risk of atrial fibrillation (AF) and other arrythmiasthe phenomenon currently known as Holiday Heart Syndrome (HHS). Patients with HHS are asymptomatic (without ischemic heart disease or heart failure symptoms) and usually have normal laboratory test results [bib_ref] Arrhythmias associated and the "Holiday Heart": Cardiac rhythm disorders, Ettinger [/bib_ref]. In 1984, Thorton et al. revised the accepted notion that HHS can only occur in chronic alcohol consumers. They presented a couple of cases with acute AF associated with sporadic large amounts of alcohol consumption. In all patients, spontaneous cardioversion was observed within 24 h [bib_ref] Atrial Fibrillation in Healthy Non-Alcoholic People After an Alcoholic Binge, Thornton [/bib_ref]. A characteristic feature of HHS is the absence of new arrhythmic episodes during abstinence and recurrence after alcohol consumption. In Samokhvalov et al. meta-analysis, increased risk of AF was observed in females consuming >24 g pure ethanol daily and in males ingesting more than 36 g. Lower amounts of alcohol did not raise the risk of arrhythmia [bib_ref] Alcohol Consumption as a Risk Factor for Atrial Fibrillation: A Systematic Review..., Samokhvalov [/bib_ref]. In Gallagher et al. meta-analysis, 5-10% participants consuming excessive amounts of alcohol experienced AF, whereas low levels of ethanol consumption did not significantly influence the frequency of AF [bib_ref] Alcohol and incident atrial fibrillation-A systematic review and meta-analysis, Gallagher [/bib_ref]. Klein et al. proved that an ethanol concentration of 2% and higher inhibits cardiac sodium channels leading to a sodium-calcium exchanger stimulation that, in turn, prolongs action potential together with a repolarization period, resulting in the increased risk of arrhythmia [bib_ref] Effect of ethanol on cardiac single sodium channel gating, Klein [/bib_ref]. Cardy et al. investigated 13 individuals aged 23-27 years old, consuming 0.95 g of ethanol per body mass kilogram. P wave and QRS complex widening reflecting conduction disturbances were observed in every study participant [bib_ref] Acute effects of ethanol ingestion on signal-averaged electrocardiograms, Cardy [/bib_ref]. T Mäki et al. showed an increase in beta-adrenoceptors density in lymphocytes during the alcohol drinking period in patients with a history of alcohol-associated AF, but no change occurred in the density in control subjects (age-matched controls without history of alcoholassociated AF). The observed increased density of beta-adrenoreceptors might be associated with an augmented reaction to adrenergic stimuli. The study also revealed a significantly higher catecholamines concentration in patients with AF than in the control group [bib_ref] Effect of ethanol drinking, hangover, and exercise on adrenergic activity and heart..., Mäki [/bib_ref].
## Deleterious cardiovascular effects of alcohol abuse-"binge and compulsive drinking"
While it is acknowledged that regular consumption of moderate doses of alcohol can protect against cardiovascular disease, the issue of the effects of alcohol abuse-regular or episodic-has not yet been clearly clarified, especially the phenomenon of "binge drinking", i.e., drinking large amounts of alcohol in the short term, e.g., drinking ≥ five drinks within one day of the week, with abstinence on the following days. People who admit to this "way of drinking" experience the adverse effects of such alcohol consumption [bib_ref] Association of alcohol consumption with selected cardiovascular disease outcomes: A systematic review..., Ronksley [/bib_ref]. Based on nearly 8 years of observation, Murray et al. found that occasional consumption of large amounts of alcoholic beverages increases the risk of coronary heart disease in men and women and hypertension in men. On the other hand, regular drinking of small doses of alcohol had a statistically significant cardioprotective effect in both sexes [bib_ref] Alcohol volume, drinking pattern, and cardiovascular disease morbidity and mortality: Is there..., Murray [/bib_ref]. The theory of the adverse effects of irregular consumption of relatively large amounts of alcohol was also confirmed by studies by Mukamal et al., indicating a twofold increase in the risk of death in people after MI who reported episodes of consumption of three or more "drinks" in a short period of 1-2 h [bib_ref] Binge drinking and mortality after acute myocardial infarction, Mukamal [/bib_ref]. Nearly 10 years of observation of almost 10,000 men without coronary artery disease at the beginning of the research in the prospective PRIME study showed that binge drinking increases the risk of coronary heart disease and hypertension, while regular consumption of low doses of alcohol had a statistically significant cardioprotective effect in both sexes [bib_ref] Patterns of alcohol consumption and ischaemic heart disease in culturally divergent countries:..., Ruidavets [/bib_ref]. In the review by Piano et al., the data from human experimental, prospective cross-sectional and cohort epidemiological studies show that binge drinking is associated with a higher risk of pre-hypertension, hypertension, MI and stroke in middleaged and older adults. Moreover, it may have adverse cardiovascular effects in young adults aged 18 to 30. It is suggested that binge drinking may cause oxidative stress, vascular injury and be proatherogenic [bib_ref] Cardiovascular Consequences of Binge Drinking: An Integrative Review with Implications for Advocacy,..., Piano [/bib_ref].
The results of prospective studies have largely confirmed the results of clinical studies. Alcohol abusers have a higher risk of mortality from all causes [bib_ref] Average volume of alcohol consumption, patterns of drinking, and all-cause mortality: Results..., Rehm [/bib_ref]. It was observed that people who had sudden cardiac death more often consumed alcohol in the last hours of life, compared with patients who died of other cardiovascular diseases [bib_ref] Alcohol and sudden cardiac death, Wannamethee [/bib_ref]. This regularity was confirmed by the results of the British Regional Heart Study, which showed that the incidence of sudden cardiac death in the group of alcohol abusers was about twice as high as among persons who did not abuse alcohol. The increase in the incidence of sudden deaths was most pronounced in older men, even if coronary artery disease was not diagnosed [bib_ref] Alcohol-use disorders, Schuckit [/bib_ref].
The relationship between compulsive drinking, i.e., a strong irrational need, even hunger for drinking, and cardiovascular diseases is expected to be systematically analyzed. Compulsive drinkers are thought to be at an increased risk of sudden cardiac death (most commonly in the mechanism of ventricular arrhythmia) and the development of cardiomyopathy. There are also data suggesting that compulsive drinking may increase the risk of MI. It is believed that the dominant drinking pattern-compulsive drinking-may underlie the dramatic increase in mortality from cardiovascular disease in the former Soviet Union, and currently in the Russian Federation [bib_ref] Global burden of disease and injury and economic cost attributable to alcohol..., Rehm [/bib_ref]. In some regions (e.g., some areas in Russia and Eastern Europe), even every second death among men before 54 is due to alcohol addiction and abuse [bib_ref] Global burden of disease and injury and economic cost attributable to alcohol..., Rehm [/bib_ref] [bib_ref] The relation between alcohol and cardiovascular disease in Eastern Europe: Explaining the..., Britton [/bib_ref]. A study conducted in the United States showed that mortality was almost twice as high in people drinking small amounts of alcohol, who had compulsive drinking episodes than in those drinking the same amounts of alcohol without compulsive episodes [bib_ref] Average volume of alcohol consumption, patterns of drinking, and all-cause mortality: Results..., Rehm [/bib_ref]. A causal connection between the pattern of alcohol consumption and adverse epidemiological phenomena was also confirmed by other types of observation. Both in Russia in 1985 and Poland in the years 1981-1982, a decline in death rate from cardiovascular causes was observed. In both countries, these periods coincided with the reduced availability of high-percentage alcohol in Russia resulting from the anti-alcohol policy of Mikhail Gorbachev under the so-called perestroika, and in Poland, it was associated with the limited distribution of alcohol during martial law [bib_ref] The relation between alcohol and cardiovascular disease in Eastern Europe: Explaining the..., Britton [/bib_ref] [bib_ref] Cardiovascular disease risk factors and mortality in Russia: Challenges and barriers, Petrukhin [/bib_ref].
It has been shown that the effect of alcohol on mortality varies depending on the age group. In younger age groups (under 50 years of age), the effects of alcohol are clearly unfavorable mainly due to the increased risk of death from injuries and accidents, especially in the group of people aged 15-29. On the other hand, among the elderly, the number of deaths, mainly due to cardiovascular diseases, which could be prevented by moderate alcohol consumption, is higher than the number of deaths caused by drinking [bib_ref] Global mortality, disability, and the contribution of risk factors: Global Burden of..., Murray [/bib_ref]. It should be noted that most studies conducted to date on the relationship between alcohol consumption and risk of death included middle-aged and older people, excluding young people. Studies that also included the younger population indicate that there is no protective effect associated with moderate alcohol consumption in young adults [bib_ref] Alcohol and mortality among young men: Longitudinal study of Swedish conscripts, Andreasson [/bib_ref]. This is probably due to the fact that in older age groups, where cardiovascular diseases are the predominant cause of death, the benefits of drinking alcohol are more pronounced than in people in the age group of 15-29 who more often die from injuries and accidents, and alcohol may even increase the risk of death from these causes [bib_ref] Underage drinking: An evolutionary concept analysis, Jones [/bib_ref].
It should also be borne in mind that the conclusion on the protective effect of small doses of alcohol could also result from a sample selection error because people with health problems could have stopped drinking, which would artificially increase mortality in the non-drinkers group [bib_ref] Risk of cardiovascular mortality in alcohol drinkers, ex-drinkers and non-drinkers, Klatsky [/bib_ref]. However, the available data do not support this hypothesis. In several studies, among non-alcohol drinkers, there were identified individuals who stopped drinking for various reasons and abstainers. Compared to moderate alcohol drinkers, both abstainers and former drinkers had a higher risk of dying from cardiovascular disease, while no significant differences were found between these groups [bib_ref] Alcohol and mortality among young men: Longitudinal study of Swedish conscripts, Andreasson [/bib_ref]. Moreover, the results of most studies did not change after excluding from the analysis, deaths that occurred during the first years of observation (it was assumed that these deaths could have been a consequence of previous diseases) [bib_ref] Effects on mortality of alcohol consumption, smoking, physical activity, and close personal..., Rehm [/bib_ref] [fig_ref] Table 1: Cont. [/fig_ref]. [fig_ref] Table 1: Cont. [/fig_ref]. Summary of selected studies.
## Authors/year design/number of participants the most important outcome
Millwood et al. 2019 [bib_ref] Conventional and genetic evidence on alcohol and vascular disease aetiology: A prospective..., Millwood [/bib_ref] mendelian randomization, prospective study/512,715
Protective effects of moderate alcohol intake against stroke are non-causative. Alcohol consumption uniformly increases blood pressure and stroke risk.
Holmes et al. 2014 [bib_ref] InterAct Consortium. Association between alcohol and cardiovascular disease: Mendelian randomisation analysis based..., Holmes [/bib_ref] mendelian randomization, meta-analysis of 56 studies/261,991
Individuals with a genetic variant associated with non-drinking and lower alcohol consumption had a more favourable cardiovascular profile and a reduced risk of coronary heart disease than those without the genetic variant. Reduction in alcohol consumption is beneficial for cardiovascular health.
Doll et al. 1994 [bib_ref] Mortality in relation to consumption of alcohol: 13 years' observation on male..., Doll [/bib_ref] prospective study/ 12,321
The consumption of alcohol appeared to reduce the risk of ischemic heart disease, largely irrespective of amount. Among regular drinkers, mortality from all causes combined increased progressively with amount drunk above 168 g of pure alcohol a week. Among British men in middle or older age, the consumption of an average of one or two units of alcohol a day is associated with significantly lower all-cause mortality vs. non-drinkers.
## Authors/year design/number of participants the most important outcome
Hoffmeister et al. 1999 [bib_ref] The relationship between alcohol consumption, health indicators and mortality in the German..., Hoffmeister [/bib_ref] prospective study Higher serum HDL cholesterol levels were observed with increasing alcohol intake. Men who consumed 1-20 g alcohol/day had a significantly lower all-cause and cardiovascular mortality vs. non-drinkers.
Ronksley et al. 2011 [bib_ref] Association of alcohol consumption with selected cardiovascular disease outcomes: A systematic review..., Ronksley [/bib_ref] meta-analysis of 84 studies Dose-response analysis revealed that the lowest risk of coronary heart disease mortality occurred with one to two drinks a day. Secondary analysis of mortality from all causes showed lower risk for drinkers compared with non-drinkers.
Mukamal et al. 2010 [bib_ref] Alcohol consumption and cardiovascular mortality among U.S. adults, Mukamal [/bib_ref] retrospective study/245,207 Light and moderate volumes of alcohol consumption were inversely associated with cardiovascular mortality.
Biddinger et al. 2022 [bib_ref] Association of Habitual Alcohol Intake With Risk of Cardiovascular Disease, Biddinger [/bib_ref] prospective study/371,463
Genetic epidemiology suggested that alcohol consumption of all amounts was associated with increased cardiovascular risk, but marked risk differences exist across levels of intake, including those accepted by current national guidelines.
Hu et al. 2022 [bib_ref] Causal associations of alcohol consumption with cardiovascular diseases and all-cause mortality among..., Hu [/bib_ref] prospective study/ 40,386
The risk of incident CVD and all-cause mortality was increased by 27% and 20% per standard drink increment of genetically predicted alcohol consumption, respectively. The authors show the potential health benefits of lowering alcohol consumption, even among light-to-moderate male drinkers.
## Are moderate alcohol consumers simply living a healthier life?
Some authors suggest that regular and moderate alcohol consumption while maintaining a healthy lifestyle (physical activity, a proper diet with large vegetable intake and no addictions) can significantly reduce the risk of cardiovascular disease and death, especially heart attack, and even prolong life expectancy [bib_ref] Effect of potentially modifiable risk factors associated with myocardial infarction in 52..., Yusuf [/bib_ref] [bib_ref] Wine, alcohol and cardiovascular diseases, Sinkiewicz [/bib_ref] [bib_ref] Relationship of Alcohol Consumption to All-Cause, Cardiovascular, and Cancer-Related Mortality in U.S...., Xi [/bib_ref] [bib_ref] Long-term wine consumption is related to cardiovascular mortality and life expectancy independently..., Streppel [/bib_ref]. On the contrary, some recent analyses indicate that association of low-to-moderate alcohol consumption with the reduction in cardiovascular risk is a result of lifestyle changes and that any reduction in alcohol consumption is in fact beneficial in terms of general health [bib_ref] Association of Habitual Alcohol Intake With Risk of Cardiovascular Disease, Biddinger [/bib_ref] [bib_ref] Causal associations of alcohol consumption with cardiovascular diseases and all-cause mortality among..., Hu [/bib_ref].
Biddinger et al. conducted an observational study in the group of 371,463 participants to assess the relationship between habitual alcohol consumption and the risk of cardiovascular diseases (hypertension, CAD, MI, stroke, HF and AF). Simultaneously, authors also aimed to evaluate the influence of six other lifestyle factors (smoking frequency, normalized BMI, self-reported physical activity, cooked vegetable intake, red meat consumption and self-reported health). Drinking groups were defined as abstainers (0 drinks/week), light (>0-8.4 drinks/week), moderate (>8.4-15.4 drinks/week), heavy (>15.4-24.5 drinks/week) and abusive (>24.5 drinks/week). [A standard drink is any drink that contains about 14 g of pure alcohol.] Alcohol intake comprised 38% beer, 29% red wine, 24% champagne or white wine, 6% spirits, 3% fortified wine and 0.2% other alcoholic beverages. Investigators found that light-to-moderate drinkers had the lowest heart disease risk, followed by people who abstained from drinking. People who drank heavily had the highest risk. They also found that light-to-moderate drinkers tended to have healthier lifestyles than abstainers; moreover, taking just a few lifestyle factors into account significantly lowered any benefit associated with alcohol consumption. The genetic analyses also revealed differences in cardiovascular risk across the spectrum of alcohol consumption, with minimal increases in risk when going from zero to seven drinks/week, much higher risk increases when progressing from seven to fourteen drinks/week (in both men and women) and very high risk-twenty-one or more drinks/week. Authors suggest a rise in cardiovascular risk even at levels deemed low-risk-below two drinks per day for men and one drink per day for women [bib_ref] Association of Habitual Alcohol Intake With Risk of Cardiovascular Disease, Biddinger [/bib_ref].
Hu et al., in the group of 40,386 males, proved the harmful effect of alcohol consumption on both cardiovascular and all-cause mortality, underlining potential health benefits associated with the reduction in alcohol intake even in the group of low-to-moderate consumption [bib_ref] Causal associations of alcohol consumption with cardiovascular diseases and all-cause mortality among..., Hu [/bib_ref].
Evidences from different experimental studies have suggested that these beneficial effects are due to polyphenols found in red wine, especially resveratrol. Therefore, there is a need for further clinical studies, especially randomized, double-blind, placebo-controlled trials, to objectively confirm the possible health-promoting effects of this substance and to determine both the efficacy and safety and possible therapeutic potential [bib_ref] Resveratrol and cardiovascular system-the unfulfilled hopes, Chudzińska [/bib_ref] [bib_ref] Resveratrol: An effective pharmacological agent to prevent inflammation-induced atrial fibrillation?, Sutanto [/bib_ref]. Moreover, resveratrol can potentially increase plasma concentration of direct oral anticoagulants such as dabigatran, edoxaban and betrixaban [bib_ref] The Clinical Significance of Drug-Food Interactions of Direct Oral Anticoagulants, Grześk [/bib_ref].
On the contrary, obesity, poor diet, low physical activity, smoking and excessive alcohol consumption are correlated with a higher incidence of CVD, dementia and diabetes [bib_ref] The combined effects of healthy lifestyle behaviors on all-cause mortality: A systematic..., Loef [/bib_ref] [bib_ref] Healthy lifestyle behaviors and all-cause mortality among adults in the United States, Ford [/bib_ref].
# Conclusions
The possibility of the protective effect of small and moderate doses of alcohol on cardiovascular diseases is significant from the point of view of public health. All over the world, cardiovascular diseases are the main cause of death, so even a minimal reduction in their relative risk can translate in absolute terms into a significantly lower number of incidence and death from cardiovascular diseases. However, due to the fact that alcohol consumption is associated with many health hazards, it is not recommended to consume it as a preventive action of cardiovascular diseases [bib_ref] ESC Guidelines on cardiovascular disease prevention in clinical practice: Developed by the..., Visseren [/bib_ref].
Funding: This research received no external funding.
## Institutional review board statement: not applicable.
Informed Consent Statement: Not applicable.
# Data availability statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.
## Conflicts of interest:
The authors declare no conflict of interest.
[table] Table 1: Cont. [/table]
|
Early‐onset leukoencephalomyelopathy due to a biallelic NDUFV1 variant in a mid‐forties patient
We present a patient who developed, after an early-onset, a stable course of spastic paraplegia and ataxia for 4 decades and eventually succumbed to two episodes of postinfectious lactic acidosis. Diagnostic workup including muscle biopsy and postmortem analysis, oxymetric analysis, spectrophotometric enzyme analysis, and MitoExome sequencing revealed a necrotizing leukoencephalomyelopathy due to the so far unreported biallelic variant of the NDUFV1 gene (p.(Pro122Leu)). This case extends our understanding of NDUFV1 variants with a 14-fold longer lifetime than so far reported cases, and will foster sensitivity toward respiratory chain disease also in adult patients with sudden deteriorating neurological deficits.CaseAn institutionalized man in his mid-forties was admitted twice in our neurology department following new episodes of mutism, coma, and flaccid tetraparesis, 6 months apart.888
# Background
Complex I deficiency (OMIM #252010) is the most common biochemical defect among the genetic mitochondrial disorders. However, its genetic basis is heterogeneous, and so is the clinical spectrum that ranges from lethal neonatal disease to adult-onset conditions. Among the frequent presentations are neonatal cardiomyopathy, leukoencephalomyelopathy, and fatal lactic acidosis. [bib_ref] Natural disease course and genotype-phenotype correlations in Complex I deficiency caused by..., Koene [/bib_ref] Here, we describe an adult individual in whom the diagnosis of a mitochondrial disorder has been made only post-mortem (with the first biallelic NDUFV1 variant c.365C>T p.(Pro122Leu)), in spite of having presented since early childhood with developmental delay, progressive spastic paraplegia, and cerebellar ataxia. In his midforties, he suddenly developed a postinfectious metabolic acidosis with muscle weakness and rapidly progressive leukoencephalomyelopathy with a fatal course over a few months.
Medical history revealed that he was the second child of consanguineous parents at the 3 rd degree. At the age of 4 months, motor development became delayed; generalized hypotonia and brisk reflexes were noticed, followed by slowly appearing progressive spastic paraplegia and signs of cerebellar ataxia during his first two decades. At the age of 20, the extent of his intellectual disability was considered significant (IQ around 60), he could walk only with assistance or drive his electric wheelchair. He talked in simple grammar and participated in the institution's activities. There were no dysmorphic features and no seizures. During the next two decades the only medical contact was the orthopedic surgeon for bilateral pes equinovarus treatment. His spasticity was regularly treated with oral baclofen.
The first admission was 2 weeks after recovery from a viral gastroenteritis with fever, after he suddenly had presented with fatigue, dysarthria, difficulties lifting his upper limbs, and drowsiness. On clinical examination, excessive sweating, severe dysarthria, generalized muscle weakness, and areflexia were observed. He then received iv methylprednisone 1 g/d for 6 days for a suspected acute disseminated encephalomyelitis, but recovered only very slowly.
The second admission was 6 months later, 1 week after a urinary infection. He presented with abdominal pain, confusion, anarthria, and finally coma. Again generalized muscle weakness and areflexia with bilateral Babinski sign were the prominent signs. Laboratory workup showed leukocytosis at 22.7 G/L with polymorphonuclear leukocytes, lactic acidosis (pH 7.3, lactate 14.3 mmol/L, pCO2 18.4 mmHg, bicarbonate 8.9 mmol/L). Lactate/ pyruvate ratio, amino acids profile, and urine organic acids were normal. The CSF protein content and lactate were elevated to 2 g/L (normal <0.45), and to 7.9 mmol/L (normal <1.8) respectively, but no pleocytosis was found.
MRI of the head and neck during the first episode. a-e) showed in the T2-weighted image an inhomogeneous hypersignal in the posterior cervical region, extending up to the anterior horn. The brain was of normal size and not atrophic, however all ventricles were dilated symmetrically. The white matter showed diffuse T1-hypointensities and T2-hyperintensities, mostly in the frontal periventricular regions, sparing gray matter, and U fibers. Cysts with partially enhancing walls and hyperintense areas gave a heterogeneous appearance to the thinned diseased white matter.
During the second episode 6 months later. ab), the lesions of the periventricular white matter in the frontal and occipital regions were confluent, with new T2-hyperintensities in the medulla, pons, the posterior limb of the internal capsule, and both thalami. The frontal white matter and centrum semiovale appeared to be swollen and contrast enhancement was more prominent and with new irregularities in the frontal periventricular white matter and the internal capsule. After corticoid treatment, contrast enhancement disappeared. c). MR spectroscopy found highly elevated lactate peak within the frontal white matter lesions, as well as a moderately decreased peak of N-acetylaspartate (NAA).
The comatose patient continued to deteriorate slowly. He was finally transferred to palliative care, and passed away 4 months later from acute pneumonia.
Postmortem findings are described inand D. Extra-cerebral autopsy showed multiple lesions of acute bronchopneumonia that was mostly responsible for the acute death, as well as hepatic microvacuolar steatosis. The other organs, including the heart, were unremarkable. The brain's external examination was unremarkable, it weighted 1280 gr after formalin fixation. Coronal sections revealed large, bilateral, and symmetrical lesion occupying the entire bilateral centrum semiovale. These lesions were characterized by a marked yellowish discoloration of the white matter with pronounced secondary cavitation. The surrounding cortex was unremarkable as were the basal ganglia nuclei. The cerebellum, the brainstem, and the upper cervical cord showed no macroscopic abnormality.
Microscopically, lesions of the centrum semiovale were infarct-like, characterized by large areas of axonal and myelin damage, and numerous resorptive vacuolated macrophages. A scattered mild perivascular lymphocytic infiltrate was observed. Similar but much less extensive infarct-like lesion were occasionally found in the basal ganglia nuclei. These lesions were not clearly delineated and did not follow a specific vascular pattern.
The brainstem and upper cervical cord were included in paraffin in totality. The midbrain and the pons were unremarkable. The medulla showed axonal and myelin damage in the anterior and lateral spinothalamic tracts. Also, the upper cervical cord showed axonal and myelin damage associated with foci of vacuolated macrophages especially in the anterior, lateral, and posterior tracts.
In view of the clinical picture of the patient with central and peripheral neurological involvement, lactic acidosis and lactate peak as well as a decreased peak of NAA on MR spectroscopy, mitochondrial disease was suspected.
On a biopsy of the vastus lateralis muscle, oxymetric analysis, and spectrophotometric enzyme analysis of the respiratory chain (RC) in fibroblasts were performed as previously described. [bib_ref] Mutations in SDHD lead to autosomal recessive encephalomyopathy and isolated mitochondrial complex..., Jackson [/bib_ref] In summary, analysis in fibroblasts was performed in isolated mitochondria, and in skeletal muscle homogenates (600 g supernatants) as described. [bib_ref] Biochemical and molecular investigations in respiratory chain deficiencies, Rustin [/bib_ref] [bib_ref] Assaying mitochondrial respiratory complex activity in mitochondria isolated from human cells and..., Birch-Machin [/bib_ref] The activities of the individual respiratory chain complexes and the mitochondrial matrix marker enzyme citrate synthase were measured spectrophotometrically with an UV-1601 spectrophotometer (Shimadzu) thermostatically maintained at 30°C. Values are estimated by the difference in activity levels measured in the presence and absence of specific inhibitors, and are expressed as ratio to the mitochondrial marker enzyme citrate synthase (mU/mU citrate synthase). The complex I (CI) activity was rotenone sensitive and the enzymatic activities of isolated complex II, IV, and V were normal. In fibroblasts d and e: FLAIR and T1 post-gadolinium of the brain showed markedly dilated lateral ventricles, especially in the posterior regions, due to white matter volume loss. In the remaining white matter extensive diffuse T1-hypointense and T2-hyperintense abnormalities were found, most prominent in the frontal periventricular regions, sparing gray matter and U fibers. Cysts with partially enhancing walls and hyperintense areas gave an heterogenous appearance to the thinned diseased white matter. This picture was interpreted as a mix of old and more recent lesions. The corpus callosus was thinned, brainstem and cerebellum were normal. (B) MRI during second episode of degradation, 6 months later. a: New brain MRI revealed progression of the confluent lesions of the periventricular white matter in the frontal and occipital regions. The frontal white matter and centrum semiovale appeared to be more swollen than 6 months before.b: Contrast enhancement was much more prominent with new irregularities in the frontal periventricular white matter and in the internal capsule.c: Two weeks later the confluent cystic lesions were even more advanced.d: MR point resolved spectroscopy showed a highly elevated lactate level and a moderately decreased peak of N-acetylaspartate (NAA) in frontal white matter lesions. (C) Postmortem brain: Coronal brain sections showed a severe leukoencephalopathy affecting both hemispheres with relative sparing of the temporal lobes (a-c). Large infarct-like lesions were found with axonmyelinic loss, numerous vacuolated macrophages, and swelling of the vascular endothelium all the CI-related enzymatic activity ratios were normal, as well as the CI depended and succinate-related pyruvate ratio (SRPR) analyzed with high-resolution respirometric analysis using an OROBOROS system. In muscle homogenates, all activity ratios were however clearly indicative of an isolated CI defect.
Complete mtDNA sequencing from the skeletal muscle biopsy revealed no pathogenic variants. Subsequently, MitoExome sequencing comprised of 1476 nuclear genes (including 1013 genes coding for mitochondrial proteins according to the MitoCarta supplemented with genes from extensive literature search) using an in-solution hybridization capture method (NimbleGen Madison, WI, USA) was performed with paired-end sequencing on a HiSeq2500 (Illumina) to an average 178x coverage. Sequence alignment and variant calling were done with CLC Workbench v.7.0.4. CLC bio, Aarhus, Denmark). [bib_ref] Deficiency of ECHS1 causes mitochondrial encephalopathy with cardiac involvement, Haack [/bib_ref] Variants with a higher frequency than 0.1% (gnomAD) were excluded from the analysis. Only the homozygous variant in NDUFV1 (exon 4) c.365C > T p.(Pro122Leu) remained to be compatible with an autosomal-recessive pattern of inheritance and a pathogenic prediction (SIFT, Mutationtaster, and PolyPhen2). Additionally, according to the ACMG-guidelines, [bib_ref] Standards and guidelines for the interpretation of sequence variants: a joint consensus..., Richards [/bib_ref] this variant is being classified as likely pathogenic. Both, the patient's father and his mother were identified as being healthy heterozygous carriers of this variant.
# Conclusions
Mitochondrial complex I deficiency is a relatively nonspecific finding that has both a variable clinical presentation and a large genetic heterogeneity. The clinician must think about the possibility of mitochondrial disease; and the task of the laboratory is difficult because of the many different genes involved. Thus, securing a diagnosis still remains challenging. Of note, the diagnosis in our patient was suspected only during the final stage of his clinical course, and confirmed only by molecular analysis after biochemical analysis was doubtful.
The NADH Dehydrogenase (Ubiquinone) Flavoprotein 1 (NDUFV1) is a core subunit of the mitochondrial membrane respiratory chain NADH dehydrogenase (Complex I), which binds to the flavin mononucleotide using ubiquinone as an electron acceptor. It is a nuclear gene that encodes the 51 kDa subunit and forms the NADH-and FMN-binding site of complex I. [bib_ref] The human mitochondrial NADH: ubiquinone oxidoreductase 51-kDa subunit maps adjacent to the..., Spencer [/bib_ref] The NDUFV1 variant found in the homozygous/biallelic state in our patient has previously been reported at the compound heterozygous state in four patients: In one patient 8 (the variant is referred to as p.(Pro113Leu) with reference to transcript variant 2 (NM_001166102), onset was at age 1 year, with stroke, exercise intolerance, and leukodystrophy. Two others patients were sisters, 9 15 and 13 years of age, who showed mild learning disability and decreased visual acuity due to optic nerve atrophy. The fourth patient presented with early-onset (3-month) and progressively deteriorating leukoencephalopathy. [bib_ref] Genotypic spectrum and natural history of cavitating leukoencephalopathies in childhood, Zhang [/bib_ref] Our patient had an early-onset, but the course was considered as stationary for over two decades, finally terminated by a rapidly evolving leukoencephalopathy apparently triggered by recurrent infectious episodes with metabolic decompensations.
Our patient extends substantially the clinical profile of known pathogenic NDUFV1 variants including a detailed description of the biochemical, radiological, tissue histopathology, and postmortem autopsy. According to the HGMD database (HGM Professional 2021.4, total), there are 47 NDUFV1 variants reported until now, 33 of which are missense or nonsense, five splicing, seven small deletions, one small insertion/duplication, and at least one large deletion. No clear genotype-phenotype correlation has been observed. [bib_ref] Natural disease course and genotype-phenotype correlations in Complex I deficiency caused by..., Koene [/bib_ref] A non-exhaustive review of the literature on PubMed revealed at least 43 reported patients with NDUFV1 variants description . The median age of onset was 9 months (range 0-18 years), nine of them have already died at an average age of 14 months (range 4-36 month), and the remaining patients have a current median age of 45.5 months (range 1-480 months). Eighteen of the 43 reported patients were homozygous, with a median onset age of 8 months, and only three already died from infantile-onset leukoencephalopathy (median onset age 3.5 months). Therefore, the life span of more than 4 decades in our patient following the early childhood-onset of leukoencephalopathy is very unusual in patients with NDUFV1 variants . The next oldest surviving patient, aged 40 at the time of publication, presented not earlier than at young adult age with spasticity and ataxia, and was compound heterozygous c.1268C>T/c.175C>T for variants in NDUFV1. [bib_ref] A Heterozygous NDUFV1 variant aggravates mitochondrial complex I deficiency in a family..., Baertling [/bib_ref] To conclude, it is therefore important to consider complex I deficiency not only in young children, but also in adult patients with leukoencephalomyelopathies. Because of the large number of candidate genes, whole exome sequencing is currently widely used as a tool to determine the molecular diagnosis of complex I deficient patients. Muscle biopsy and biochemical testing in tissues analysis should be considered when genetic testing cannot confirm a diagnosis in clinically suspected complex I deficiency. In our case, an earlier diagnostic may have had an impact on the final (fatal) outcome, as preventive measure aimed at limiting catabolic triggers and antioxidant therapies such as multiple vitamins, co-enzyme Q10, and carnitine are sometimes clinically effective. [bib_ref] Mitochondrial complex I mutations in Caenorhabditis elegans produce cytochrome c oxidase deficiency,..., Grad [/bib_ref]
[fig] Figure 1: (A) MRI of the first episode of degradation. a: T2 of spinal cord showed a hypersignal of the posterior two thirds of the spinal cord over its entire extent, but predominantly in the cervical region.b: Axial plane of the cervical MRI showing that the lesions were inhomogenous and affected the anterior horn. -3 weeks later, a second MRI was done.c: T2 of the spinal cord showed that the abnormalities regressing only partially after 3 weeks. [/fig]
|
ComplexContact: a web server for inter-protein contact prediction using deep learning
ComplexContact(http://raptorx2.uchicago.edu/ ComplexContact/) is a web server for sequencebased interfacial residue-residue contact prediction of a putative protein complex. Interfacial residueresidue contacts are critical for understanding how proteins form complex and interact at residue level. When receiving a pair of protein sequences, ComplexContact first searches for their sequence homologs and builds two paired multiple sequence alignments (MSA), then it applies co-evolution analysis and a CASP-winning deep learning (DL) method to predict interfacial contacts from paired MSAs and visualizes the prediction as an image. The DL method was originally developed for intraprotein contact prediction and performed the best in CASP12. Our large-scale experimental test further shows that ComplexContact greatly outperforms pure co-evolution methods for inter-protein contact prediction, regardless of the species.
# Introduction
Most proteins function by interacting with others to form complexes and/or protein-protein interaction (PPI) networks [bib_ref] Principles of protein-protein interactions, Jones [/bib_ref]. Solving the structures of protein complexes by experimental techniques is very challenging [bib_ref] Modeling protein-protein and protein-peptide complexes: CAPRI, Lensink [/bib_ref]. For example, there is little structural information for ∼80% of currently known protein interactions in bacteria, yeast or human [bib_ref] Interactome3D: adding structural details to protein networks, Mosca [/bib_ref]. Computational prediction is an alternative way to elucidate the structure of a complex of interacting proteins. Inter-protein contact prediction is becoming an important intermediate step for such a task [bib_ref] Sequence co-evolution gives 3D contacts and structures of protein complexes, Hopf [/bib_ref].
Due to the evolution pressure, co-evolved residues are often found to be spatially proximal within the protein structure [bib_ref] Protein structure prediction from sequence variation, Marks [/bib_ref] or upon the protein-protein interface [bib_ref] Identification of direct residue contacts in protein-protein interaction by message passing, Weigt [/bib_ref]. As such, co-evolution analysis or more specifically direct-coupling analysis (DCA) (e.g. EVcomplex [bib_ref] Sequence co-evolution gives 3D contacts and structures of protein complexes, Hopf [/bib_ref] and Gremlin-Complex (4)) is widely used to identify co-evolved residues and predict inter-residue contacts from multiple sequence alignments (MSA) [bib_ref] Identification of direct residue contacts in protein-protein interaction by message passing, Weigt [/bib_ref] [bib_ref] Emerging methods in protein co-evolution, De Juan [/bib_ref] [bib_ref] Protein contact prediction by integrating joint evolutionary coupling analysis and supervised learning, Ma [/bib_ref] [bib_ref] PSICOV: precise structural contact prediction using sparse inverse covariance estimation on large..., Jones [/bib_ref] [bib_ref] CCMpred--fast and precise prediction of protein residue-residue contacts from correlated mutations, Seemayer [/bib_ref] [bib_ref] CoinFold: a web server for protein contact prediction and contact-assisted protein folding, Wang [/bib_ref] [bib_ref] Simultaneous identification of specifically interacting paralogs and interprotein contacts by direct coupling..., Gueudré [/bib_ref] [bib_ref] FreeContact: fast and free software for protein contact prediction from residue co-evolution, Kaján [/bib_ref] [bib_ref] Direct-coupling analysis of residue coevolution captures native contacts across many protein families, Morcos [/bib_ref]. Although popular, DCA has low accuracy when a protein under prediction does not have many sequence homologs in MSA [bib_ref] Accurate de novo prediction of protein contact map by ultra-deep learning model, Wang [/bib_ref] [bib_ref] DNCON2: Improved protein contact prediction using two-level deep convolutional neural networks, Adhikari [/bib_ref] [bib_ref] Protein contact prediction by integrating deep multiple sequence alignments, coevolution and machine..., Adhikari [/bib_ref] [bib_ref] EPSILON-CP: using deep learning to combine information from multiple sources for protein..., Stahl [/bib_ref] [bib_ref] A deep learning framework for improving long-range residue-residue contact prediction using a..., Xiong [/bib_ref] [bib_ref] Prediction of residue-residue contact matrix for protein-protein interaction with Fisher score features..., Du [/bib_ref] [bib_ref] NeBcon: protein contact map prediction using neural network training coupled with naïve..., He [/bib_ref] [bib_ref] Improved contact predictions using the recognition of protein like contact patterns, Skwark [/bib_ref]. This problem becomes even more serious for inter-protein contact prediction since it is challenging to find so many interlogs (i.e. interacting homologs) for an interacting protein pair, especially for eukaryotic species [bib_ref] Sequence co-evolution gives 3D contacts and structures of protein complexes, Hopf [/bib_ref].
We present ComplexContact, a web server that predicts inter-protein residue-residue contacts without using any structural templates. The underlying algorithm of this server is a Deep Learning (DL) model which has won intraprotein contact prediction in CASP12 [bib_ref] Analysis of deep learning methods for blind protein contact prediction in CASP12, Wang [/bib_ref] [bib_ref] Assessment of contact predictions in CASP12: co-evolution and deep learning coming of..., Schaarschmidt [/bib_ref]. In addition to co-evolution information, our DL method makes use of sequential features and contact occurrence patterns to dramatically reduce the requirement of sequence homologs and greatly improve accuracy [bib_ref] Accurate de novo prediction of protein contact map by ultra-deep learning model, Wang [/bib_ref]. This server also applies a phylogeny-based method to identify interlogs and build better MSAs for a protein pair from eukaryotes. Our experimental results show that ComplexContact outperforms pure DCA for inter-protein contact prediction for both prokaryotes and eukaryotes.
# Materials and methods
The detailed description of the DL algorithm underlying ComplexContact is described in [bib_ref] Accurate de novo prediction of protein contact map by ultra-deep learning model, Wang [/bib_ref] and the detailed experimental results for inter-protein contact prediction is available at [bib_ref] Deep learning reveals many more inter-protein residue-residue contacts than direct coupling analysis, Zhou [/bib_ref]. Here, we briefly summarize the method.
## Overall flowchart for inter-protein contact prediction
As shown in [fig_ref] Figure 1: Illustration of ComplexContact workflow [/fig_ref] , given a pair of putative interacting protein sequences A and B for which users would like to predict inter-protein contacts, our method first employs HHblits to search for sequence homologs and build MSAs for A and B, respectively. Then we employ two strategies (i.e. genome-and phylogeny-based) to concatenate MSA A and MSA B into two paired MSAs consisting of only interlogs, denoted as MSA g and MSA p, from which we may predict two inter-protein contact maps using our DL model (trained from single-chain proteins), and calculate their average as the final prediction.
## Concatenate two multiple sequence alignments
Concatenating MSAs by genomic distance. In prokaryotes and some eukaryotes, interacting genes are often co-located on the chromosome into operons [bib_ref] Inter-protein sequence co-evolution predicts known physical interactions in bacterial ribosomes and the..., Feinauer [/bib_ref] , so we may assume two proteins forming an interacting pair if their intergenic distance is less than a threshold [bib_ref] Group role assignment via a Kuhn-Munkres algorithm-based solution, Zhu [/bib_ref]. A similar approach is used in EVcomplex (5) and Gremlin-Complex (4).
Concatenating MSAs by phylogeny. In most eukaryotes, it is challenging to concatenate two individual MSAs since an individual MSA may contain abundant paralogs and two genes may interact even if they are not close by genomic distance [bib_ref] Simultaneous identification of specifically interacting paralogs and interprotein contacts by direct coupling..., Gueudré [/bib_ref]. Here, we group proteins in each MSA by their species (or sub-species if possible) according to the phylogeny tree in the Taxonomy Database [bib_ref] The NCBI taxonomy database, Federhen [/bib_ref]. Then we sort proteins of a specific species/subspecies in each MSA by their sequence similarity (from high to low) to their respective query proteins. Let p 1 , p 2 , . . . , p m and q 1 , q 2 , . . . , q n be the sorted proteins of a specific species in two MSAs, respectively. Then we pair p i and q i together where i ranges from 1 to the minimum of m and n.
On average for eukaryotes, the phylogeny-based method works better while for prokaryotes, the genomic-based method works better. Combining them can improve performance on eukaryotes. For some protein pairs, neither method can identify many sequence homologs, and the resultant interfacial contacts may have low accuracy.
## Deep learning for inter-protein contact prediction
Our DL model is formed mainly by two deep residual neural networks (ResNet) [bib_ref] Deep residual learning for image recognition, He [/bib_ref]. One is used to handle sequential features and the other pairwise features. The first ResNet conducts 1-dimensional (1D) convolutional transformation of sequential features to capture long-range sequential context of each residue in the query proteins. Its output is converted to a 2-dimensional (2D) matrix and then fed into the 2nd ResNet together with the original pairwise features. The second ResNet conducts 2D convolutional transformation of its input to capture long-range 2D context of a residue pair. Finally, the output of the 2nd ResNet is fed into logistic regression, which predicts the probability of any two residues forming a contact.
The sequential features include protein sequence profile, predicted 3-state secondary structure (31) and 3-state sol- vent accessibility [bib_ref] AcconPred: Predicting solvent accessibility and contact number simultaneously by a multitask learning..., Ma [/bib_ref] [bib_ref] RaptorX-Property: a web server for protein structure property prediction, Wang [/bib_ref]. The pairwise features include the direct information produced by CCMpred (11) and mutual information derived from paired MSAs.
The DL model underlying ComplexContact was originally developed for intra-protein contact prediction, which has been officially ranked No. 1 in CASP12 [bib_ref] Analysis of deep learning methods for blind protein contact prediction in CASP12, Wang [/bib_ref] [bib_ref] Assessment of contact predictions in CASP12: co-evolution and deep learning coming of..., Schaarschmidt [/bib_ref]. That is, our DL model was trained by single-chain proteins instead of protein complexes, so there is no overlap between our training and test sets. We doubled check this by BLAST [bib_ref] Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Altschul [/bib_ref] , which shows that none of the test protein pairs can be simultaneously aligned to a training protein.
# Result
## Evaluation and metrics
We compare ComplexContact with pure DCA methods such as CCMpred (11), Gremlin [bib_ref] Assessing the utility of coevolution-based residue-residue contact predictions in a sequence-and structure-rich..., Kamisetty [/bib_ref] and EVfold [bib_ref] Protein 3D structure computed from evolutionary sequence variation, Marks [/bib_ref] and two related web servers GremlinComplex and EVcomplex. CCMpred, EVfold and Gremlin are pure co-evolution methods and their accuracy is not very different.
We evaluate the accuracy of the top L/5, 50, 20, 10, 5 predicted inter-protein contacts. The top-k accuracy is defined as the percentage of correct predictions among the top k predicted contacts. When the number of native contacts is smaller than k, we still use k as denominator in calculating accuracy, which may make the accuracy look small when k is big. More experimental results are available at the online documentation page http://raptorx2.uchicago. edu/ComplexContact/documentation/#6.
## Performance on baker and 3dcomplex datasets
As shown in , tested on the Baker's dataset (of 32 protein pairs), ComplexContact greatly outperforms EVComplex (EVfold), GremlinComplex and CCMpred regardless of how many predicted contacts are evaluated.
As shown in [fig_ref] Figure 2: The top-50 prediction accuracy by ComplexContact and Gremlin on the protein pairs... [/fig_ref] , on a much larger benchmark (of 4479 heterodimers) extracted from 3D complex (37), Com-plexContact outperforms Gremlin by a large margin regardless of the species of the test dimers. The prediction accuracy depends on two main factors: the number of non-redundant sequence homologs in the multiple sequence alignment (MSA) and the interfacial contact density (measured by the number of interfacial contacts divided by sequence length sum). The former determines the quality of co-evolution signal (one of the input features of our DL method). The latter impacts prediction accuracy because our DL method makes use of contact occurrence patterns. When contact density is low, it is hard to identify reliable contact patterns.
## Quality assessment of the predicted probability
ComplexContact predicts the probability of any two residues forming a contact. Here we assess the quality of the top 50 predicted probability values of the heterodimers extracted from 3Dcomplex.
As shown in [fig_ref] Figure 3: Quality assessment of the top 50 predicted interfacial contacts for 4479 heterodimers... [/fig_ref] , ComplexContact has much better AUC . Gremlin has an AUC <0.5, which implies that its contact selection is even worse than random guess. [fig_ref] Table 2: Precision and recall of the top 50 predicted interfacial contacts by ComplexContact... [/fig_ref] shows the precision and recall for a list of probability values produced by ComplexContact. For example, when the predicted probability is >0.90, the precision is 0.57.
## Server implementation
## Overall description
Input. As shown in [fig_ref] Figure 4: ComplexContact server job submission [/fig_ref] , users may submit a single sequence pair, a pair of multiple sequence alignments or a batch of 20 sequence pairs by copying and pasting to the input text field or uploading files. A jobname and an email address are optional, but they can facilitate job retrieval.
Job retrieval. ComplexContact assigns one unique job ID and one URL to each submission for job retrieval. When an email is provided in submission, users will be notified by email once a batch of jobs are done; users may also retrieve their jobs by the 'My Jobs' link at the top right of the web page and by the 'Job Status' link, through which users may find a job by one of its submitted sequences.
Output. In addition to the original input sequences, the result web page has three result sections (see [fig_ref] Figure 5: ComplexContact server result page [/fig_ref]. The first section visualizes the predicted contact map, which can be zoomed in and dragged around to facilitate detailed examination. Hovering mouse over the contact image will display the predicted contact probability value at a specific residue pair. The second section includes three panels: a panel for contact image zooming and dragging, a panel for downloading the predicted complex contact map, and a panel for downloading the detailed prediction results. The third section displays the two paired MSAs generated by genome-and phylogeny-based methods as well as the number of sequence homologs in each MSA.
Nucleic Acids Research, 2018, Vol. 46, Web Server issue W435 (1), optional user email address [bib_ref] Modeling protein-protein and protein-peptide complexes: CAPRI, Lensink [/bib_ref] , and a pair of sequences (or multiple sequence alignments) [bib_ref] Interactome3D: adding structural details to protein networks, Mosca [/bib_ref]. The sequences shall be in FASTA format and can also be submitted in a file. (B) Users may also submit a job by a publicly available program Curl without using the web interface. In this command, Job name and Email address are optional. A job URL will be returned on screen after submission. Curl allows users to submit a large number of jobs quickly. Predictors ending with (s) are a web server. EVfold is same as EVcomplex, but runs locally with our MSAs. Columns 2-5 show accuracy of top predicted contacts. (1), where the predicted probability is displayed in greyscale, with a darker color indicating a larger value. The middle part shows three panels. The first one is used to zoom and drag contact images [bib_ref] Modeling protein-protein and protein-peptide complexes: CAPRI, Lensink [/bib_ref]. The second panel is for downloading the predicted contact map (3), and the third panel is for downloading the detailed prediction results (4). The right part shows two paired MSAs generated by genome-based method (5) and phylogeny-based method (6). Processing time. The running time depends on the length of the two input sequences and the number of sequence ho-mologs detected by the server. For a protein pair of ∼250 residues, it takes about one hour to finish one job after it is scheduled to run. When there are many waiting jobs (or jobs of long sequences) in the queue, it may take a few hours for a job to be scheduled to run. Nevertheless, most jobs can be done within one day after submission.
Documentation. The documentation of ComplexContact is available by the 'Docs' link at the web page. It includes some details about the server, descriptions of input and output, explanations of prediction results, a sample prediction result and more experimental results.
## Conclusion and future work
We have presented ComplexContact, a web server for sequence-based interfacial residue-residue contact prediction using deep learning and residue co-variation. Com-plexContact outperforms similar servers by a large margin regardless of the species. However, it shall be noted that complex contact prediction is a very challenging problem and there is still a large room for improvement. Currently our DL model was trained using only single-chain proteins.
[fig] Figure 1: Illustration of ComplexContact workflow. Given a pair of putative interacting proteins A and B, ComplexContact first uses HHblits(38) to search for sequence homologs and build an MSA for each protein. Then ComplexContact constructs two paired MSAs using genome and phylogeny information. Finally, ComplexContact applies deep learning to predict two inter-protein contact maps from the two paired MSAs and calculates their average as the final contact prediction. The top half of this figure is inspired byFig. S2in(40). [/fig]
[fig] Figure 2: The top-50 prediction accuracy by ComplexContact and Gremlin on the protein pairs extracted from 3DComplex. Each dot represents one protein pair and is colored by its species. A dot below the diagonal line indicates that ComplexContact has a better accuracy. [/fig]
[fig] Figure 3: Quality assessment of the top 50 predicted interfacial contacts for 4479 heterodimers extracted from 3Dcomplex. (A) and (B) show the precisionrecall (in red) and ROC (in blue) curves generated by Gremlin and ComplexContact, respectively. AUC: Area Under the ROC curve; AUPRC: Area Under the precision-recall curve. [/fig]
[fig] Figure 4: ComplexContact server job submission. (A) Users may submit a job by a web interface, which has fields for job name [/fig]
[fig] Figure 5: ComplexContact server result page. The left part shows the predicted complex contact map [/fig]
[table] Table 2: Precision and recall of the top 50 predicted interfacial contacts by ComplexContact on the 3Dcomplex data Detailed prediction results. The downloadable file contains the followings: (a) the two input protein sequences in FASTA format; (b) MSAs generated by HHblits (38) for each input sequence; (c) two paired MSAs; (d) one predicted complex contact map for each paired MSA; (e) the final predicted contact map. [/table]
|
Combinatorial patterns of gene expression changes contribute to variable expressivity of the developmental delay-associated 16p12.1 deletion
[fig] Figure S2, Figure S3: Identification of rare gene-disrupting variants from WGS data. Schematic shows the genomic locations in relation to a mock gene for 25 classes of rare coding and non-coding SNVs/indels, copy-number variants (duplications and deletions), and short tandem repeat (STR) expansions identified from sequencing data. Each variant class is color-coded and labeled by letter (A-Y). The number of rare variants in each class per individual is presented in Additional File 2: Table S4. Quality control of RNA sequencing data. (A) Scatter plot shows moderate correlation between sample RIN scores and median TIN scores (r=0.38, p=1.0×10 -4 , Pearson correlation) of the 96 RNA-Seq replicates. (B) Heatmap shows Spearman correlation coefficients between pairs of 96 RNA-sequencing sample replicates. Pairs of replicates corresponding to the same individual showed higher correlation coefficients (orange) than replicates from different individuals. [/fig]
[fig] Figure S4: Genes expressed in LCL samples by disease relevance. (A) Percentage of genes in candidate disease gene databases that are expressed in the 32 LCL samples (>0.2 TPM in all replicates of ≥1 sample). "OMIM", Online Mendelian Inheritance in Man; "DDD", Deciphering Developmental Disorders (ID/DD); "DBD", Geisinger Developmental Brain Disorder; "DDDG2P", Developmental Disorders Genotype-Phenotype Database (ID/DD); "SFARI", Simons Foundation Autism Research Initiative; "Purcell", GeneBook from Purcell and colleagues (schizophrenia) (1). (B) Percentage of genes expressed in GTEx adult brain tissues that are expressed in the 32 LCL samples. GTEx expression was defined as genes that showed >0.5 TPM in 80% of samples for a particular tissue. (C) Heatmap depicts Spearman correlation coefficients between the median expression of adult brain tissues and lymphocyte samples from GTEx datasets (2). [/fig]
[fig] Figure S5: Family-based differential gene expression analysis. (A)To identify family-specific patterns for gene expression changes in 13 trios among the five families, we performed differential expression analysis between offspring and their carrier parents, and separately identified differentially expressed genes between offspring and noncarrier parents. After overlapping these two gene sets, we assigned gene expression changes in the offspring as unique (different in offspring than both parents), shared with the carrier parent (different in offspring than noncarrier parent), or shared with the noncarrier parent (different in offspring than carrier parent).(B) Carrier children (n=10) have a non-significant trend towards a higher burden of rare deleterious variants compared with carrier parents (n=6, p=0.071, one-tailed Mann-Whitney test), mirroring trends previously observed in a larger cohort of 26 deletion families (3). Boxplot indicates median (center line), 25th and 75th percentiles (bounds of box), and minimum and maximum (whiskers). (C) Line plots show family-specific patterns for differentially expressed genes in the child samples from 13 trios with carrier offspring (including carrier children compared to parents and carrier parents compared to grandparents) and four trios with noncarrier offspring in each of the five 16p12.1 deletion families. Individual samples are labeled by sample ID (Additional File 2: [/fig]
[fig] Figure S6, Figure S7: Outlier gene expression analysis in the 16p12.1 deletion cohort. (A) Scatter plot shows first two principal components (PC) of normalized expression values for RNA-seq replicates (n=96) before PEER correction. (B) Scatter plot shows first two PCs of PEER-corrected expression values for each RNA-Seq replicate (n=96). (C) Boxplot shows the number of outlier genes per individual (n=32) for different outlier expression z-score thresholds. (D) Boxplot shows that carrier children (n=10) have trends towards smaller numbers of genes with outlier expression (|z|>2) than carrier and noncarrier parents (n=6, p<0.073, two-tailed Mann-Whitney test). All boxplots indicate median (center line), 25 th and 75 th percentiles (bounds of box), and minimum and maximum (whiskers). Differentially expressed genes between 16p12.1 deletion carriers and noncarriers. (A) Venn diagram shows the overlap of differentially expressed genes identified in the main analysis (using edgeR pipeline) and with relatedness correction (using PQLSeq pipeline). (B) Bar plot shows Gene Ontology terms whose genes are significantly up-or down-regulated in deletion carriers compared with non-carriers (PAGE analysis; FDR<0.05). Positive z-scores indicate upregulated biological functions, and negative z-scores indicate downregulated biological functions. (C) Diagram shows GTEx brain tissues with preferentially expressed genes that are enriched (FDR<0.05) among genes differentially expressed in 16p12.1 deletion carriers. GTEx preferential tissue expression was defined as expression >2 standard deviations than the median expression across all tissues for that gene. Brain diagram modified from WikiCommons open-source image Enrichment of genes disrupted by the 16p12.1 deletion in GTEx brain tissues behavioral fear response non-canonical Wnt pathway for heart development canonical Wnt pathway for cardiac muscle cell fate commitment regulation of heart morphogenesis cardiac muscle cell proliferation urinary tract smooth muscle contraction negative regulation of muscle tissue development skeletal muscle tissue growth forebrain generation of neurons cranial ganglion development neuron fate determination negative regulation of cell proliferation 16p12.1 deletion carriers (PAGE analysis) Fig. S8. Weighted gene co-expression network analysis. (A) Heatmap showing co-expression between pairs of genes expressed in LCL samples derived from all individuals within the five 16p12.1 deletion families (n=32 samples). Yellow cells indicate pairs of genes that are co-expressed across samples, and colored bars on the side of the heatmap represent assignment of genes to one of 35 WGCNA modules. (B) Heatmap showing expression fold-change differences of genes in the 35 WGCNA modules between LCL samples of carriers (n=57 replicates of 19 samples) andnoncarriers (n=39 replicates of 13 samples). Two modules showed sex-specific differences and were excluded for further analysis. Six modules showed significant expression differences between carriers and noncarriers (*FDR<0.05, two-tailed T-test with Benjamini-Hochberg correction). (C) [/fig]
[fig] FigFigure S11, Figure S12, Figure S13, Figure S14, Figure S15: S10. Enrichment of differentially expressed genes for rare variants. (A) Forest plot shows enrichment (Fisher's exact test, **=FDR<0.05, *=uncorrected p<0.05) of differentially expressed genes in the carrier offspring of 13 trios for proximal coding and non-coding rare variants, including single-nucleotide variants (SNVs) and insertions/deletions (indels) with CADD scores >10 (4), structural variants (SVs), and short tandem repeats (STRs). (B) Forest plot shows classes of rare variants with significant enrichment (Fisher's exact test, **=FDR<0.05, *=uncorrected p<0.05) towards differentially expressed genes in carrier children (n=9) or carrier parents (n=4). (C) Forest plot shows classes of rare variants with significant enrichment (Fisher's exact test, **=FDR<0.05, *=uncorrected p<0.05) towards differentially expressed genes in carrier children (n=9) that are shared with either carrier or noncarrier parents. All forest plots show log-odds ratios (dots) and 95% confidence intervals (whiskers). Odds ratios, confidence intervals, p-values, and Benjamini-Hochberg corrected FDRs for comparisons with all classes of "second-hit" variants are listed in Additional File 4Rare variant burden of outlier genes in the 16p12.1 deletion cohort. (A) Outlier expression genes in all individuals (n=32) have a higher average burden of rare variants than non-outlier genes (p=1.01×10 -3 , one-tailed t-test). (B) Plots show individual classes of rare variants with increased average burden (FDR, one-tailed t-test with Benjamini-Hochberg correction) in outlier genes across all samples (n=32) compared with nonoutlier genes. (C) Outlier genes with preferential expression in GTEx brain tissues (2), RVIS scores <20 th percentile (5), or pLI scores >0.9 (6) have a higher average burden of rare variants than outlier genes not in those categories (p<0.05, two-tailed t-test) across all samples (n=32).All plots show mean +/-SD of the average burden of rare variants in each gene class. Examples of synergistic effects between 16p12.1 deletion and rare variants on expression. Scatter plots show normalized expression values (TPM) of genes with synergistic effects due to the 16p12.1 deletion and inherited "second-hit" variants (see Methods). Blue circles indicate expression values for samples from deletion carriers and family members with rare "second-hit" variants, orange circles indicate expression values for samples from family members without the rare variant, and green circles indicate expression values of samples from other deletion carriers and noncarriers in the cohort. Black lines denote median gene expression for LCL replicates of each individual used to identify genes with outlier expression in individual deletion carriers. Family-based alternative splicing analysis. (A) Boxplot shows the distribution of alternative splicing events in the carrier offspring of 13 trios within 16p12.1 deletion families based on family-specific patterns. Unique events in offspring occurred at a lower frequency (p=2.44×10 -4 , twotailed paired Mann-Whitney test) than those shared with either parent. Boxplot indicates median (center line), 25th and 75th percentiles (bounds of box), and minimum and maximum (whiskers). (B) Forest plot shows enrichment (Fisher's exact test, **=FDR<0.05, *=uncorrected p<0.05) of alternate splicing events in the carrier offspring of 13 trios for rare variant classes that could putatively affect gene splicing. (C) Forest plot shows classes of rare variants with significant enrichment (Fisher's exact test, **=FDR<0.05, *=uncorrected p<0.05) towards genes with alternative splicing in carrier children (n=9) or carrier parents (n=4). Forest plots show log-odds ratios (dots) and 95% confidence intervals (whiskers). Odds ratios, confidence intervals, p-values, and Benjamini-Hochberg corrected FDRs for comparisons with all classes of "second-hit" variants are listed in Additional File 4: Data S2. Allele-specific expression analysis. (A) Boxplot shows the distribution of allele-specific expression (ASE) events by family-specific pattern in the carrier offspring of 13 trios within the deletion families. Unique events in the offspring occurred at a higher frequency (p=2.44×10 -4 , two-tailed paired Mann-Whitney test) than events shared with either or both parents. (B) Boxplot shows the number of ASE events that lead to overexpression of a deleterious (CADD score>25) coding variant per individual. There were no significant differences (p>0.05, two-tailed Mann-Whitney test) between carrier children (n=10) and their carrier (n=6) or noncarrier parents (n=6). All boxplots indicate median (center line), 25th and 75th percentiles (bounds of box), and minimum and maximum (whiskers). Expression quantitative trait loci (eQTL) analysis. (A) Bar plot shows the genomic characteristics of the 21 discovered eQTL variants, including genomic location, location within non-coding regulatory features, and chromatin state annotations in the adult or fetal brain. (B) Boxplotshows distribution of eQTL minor alleles observed in carrier children (n=10), carrier parents (n=6), and noncarrier parents (n=6). A non-significant trend (p=0.107, two-tailed Mann-Whitney test) towards higher minor allele burden was observed in carrier children compared with carrier parents.Boxplot indicates median (center line), 25th and 75th percentiles (bounds of box), and minimum and maximum (whiskers). [/fig]
[table] Table showing: clusters of enriched (FDR<0.05) Gene Ontology (GO) Biological Process terms in the six WGCNA modules with significant carriernoncarrier differences. The dark orange module did not have any significant enrichments.Fig. S9. Functional enrichments of differentially expressed genes in carrier children. Table shows clusters of Gene Ontology (GO) biological function terms enriched (FDR<0.05) among differentially expressed genes in carrier children with the 16p12.1 deletion (n=9), colored by familyspecific patterns for the differentially expressed gene sets (unique, shared with the carrier or noncarrier parent, or both unique and shared with either parent). [/table]
|
On the Diseases and Injuries of the Joints—Clinical and Pathological Observations
## 83
doubt" in the minds of surgeons about opening large joints in a state of suppuration, we can testify from personal observation that there is often much hesitation) and a good series of cases in which knee-joints had been saved from amputation or resection by timely incisions, would have done much to encourage the reader in the adoption of a course which is certainly recommended by reason, and we believe sanctioned by great success. Further, it is a point of much interest to determine the general consequences of such incisions. In children they are undoubtedly often followed by complete cure, and m adults it is not at all clear that incurable anchylosis usually follows. The subject is dismissed by Mr. Bryant in a single sentence, and without any allusion to the important difference in prognosis occasioned by the age of the pahent^e gti(m of the existence of synovial membrane over the articular cartilages in the natural condition, Mr. Bryant's opinion is in the affirmative :
"In a case " lie says, "where the ankle had evidently but recently become inflamed, and presented a synovial membrane which was most exquisitely iniected films of recent but firm fibrinous material were poured out over the surface of the cartilage, and beneath this were fine radiating capillary vessels proceeding from the margin; in one spot I carefully raised the deposit of lymph, The observations of Mr. Bryant on the subject of chronic degeneration of the synovial membrane lead him to conclude, with most pathologists of the present day, that the disease is a mere result of chronic inflammation, and that there is nothing peculiar about it He describes, however, two forms of the disease-the ' gdatimform and the " nulDV " of which the inflammatory origin of the former he thinks more clearly proved than of the latter; but we cannot say that he has established any rational ground for the division which he makes, and to us they seem merely two different appearances of the same thing.
Mr. Bryant's views on the pathology of the articular cartilages are probably known to most of our readers, by a paper read before the Medico-Chirurgical Society, and of which an abstract will be found m their ? Proceedings,' vol. i. p. 70. They may be thus summed np in Mr. Bryant's own words:
."The diseases to which articular cartilages are liable may be thus classed.
Like other tissues, they may undergo hypertrophy or atrophy, using the latter in its simplest sense. Inflame and ulcerate they cannot, as the presence of vessels in the tissue is generally considered necessary for such processes. But to granular, fatty, and fibrous degeneration they are peculiarly liable; and in these forms of degeneration may be included the processes winch have been so variously described by different authors." (p. 42.) 84 Reviews.
[Jan.
We will not quote more largely from Mr. Bryant's work on this head, as no doubt his views are familiar to most of our readers. The principal fact on which Mr. Bryant dwells is, that as cartilages derive their nutrition from other structures (either the bones or synovial membranes, as he believes), so their diseases are secondary to the affections of these structures. But this order of events is, according to his own showing, not universal. " There is a form of disease which approaches the nearest to primary disease of the cartilages, in which this structure rapidly disappears, followed by, or connected with, suppuration of the joint; it is seen in patients of middle age who have had an attack of rheumatism, which at last settled in one joint. . . .
The disease is at first confined entirely to the cartilage, involving the synovial membranes in a secondary manner. The cartilage may disappear, and in healthy subjects ankylosis may rapidly take place," &c. (p. G3.)
In these cases, then, we have, according to Mr. Bryant's statement, an acute disease attended with all the symptoms of inflammation, commencing in the cartilage, and producing its absorption. Whether the old term " ulceration" be retained for this, or the new one proposed by Mr. Bryant, "granular degeneration," be substituted, seems to us more a verbal than a practical question. For ourselves, as the old term at any rate keeps alive in the minds of those who use it the essential idea of the disease?viz., that it is one of the sequelse of acute inflammation, while the new name rather keeps this connexion out of sight, we should prefer the old term, bearing in mind, of course, that inflammation and ulceration of non-vascular tissues differ in certain easily-described particulars from the usual instances of those processes. This subject has been most ably treated by Mr. Barwell, in the last number of this Review, and to that paper we would refer our readers for a proof that the terms " inflammation" and "ulceration" may be as rationally used of cartilages as of any other structures. We would refer to the section on " Treatment" appended to Mr. Bryant's account of the diseased condition of the cartilages, as a specimen of the ill effects of the subordinate importance which he has attached to these structures. It has seldom been our lot to see anything in a practical surgical work more meagre and unsatisfactory. In fact, all the chapters on treatment are vei-y slight. This perhaps is intentional in a work which professes to be merely a collection of " observations," but it is not the less to be regretted. We believe a student might read the work from beginning to end, without being aware that any one had ever made an issue for the cure of a diseased joint; and in the only passage, as far as we can discover, where mercury is distinctly spoken of, its use to salivation is unhesitatingly condemned, and perhaps rightly enough in the particular affection which Mr. Bryant is there speaking of?viz., chronic affection of the articular extremities; but surely some opportunity should have been afforded to the reader of learning his author's opinions on two such very important points as the use of issues and salivation in treating the disease which, if Mr. Bryant will not permit us to call it acute inflammation of the cartilages, we hardly know what he would have us denominate it.
## I860.]
Bryant on Diseases and Injuries of the Joints.
## 85
On the subject of diseases of the articular ends of the bone, Mr. Bryant is, we think, more happy, and puts in a strong light the important fact that " the majority of the cases which are described by surgeons as strumous or scrofulous disease of a joint depends upon a chronic inflammation in the bone. The whole of this section on inflammation of the articular extremities of bone is well worthy of perusal, and its defects only lead us the more to regret that Mr. Bryant did not delay the publication of his book until he had worked it up into a complete treatise on the whole subject. No one who reads this chapter attentively can doubt that Mr. Bryant's experience has been as ample as his industry has been untiring But, on the other hand, no one can say that the diagnosis is clearly laid down, or the treatment fully and logically ^ On^the^'subj ect of amputation or excision, Mr. Bryant has a very interesting chapter. He speaks strongly, perhaps too strongly, against excision of the hip; and is hardly more favourable to excision of the knee On the subject of the comparative mortality of excision of the knee'and amputation he makes the following rather startling observations :?
"In a paper which I have read before the Medical and Chirurgical Society, I demonstrated that the deaths from amputation of the thigh including these two last specified classes of cases-i.e, if we understand Mr Bryants fnr diseases, or 'pathological, as he terms them], was Mr. Butcher and Mr. Price, avowed advocates for excision, give the fatality [a term which Mr Bryant always uses instead of mortality ] of excision ot the i ? ti,is shows the most favourable aspect. In the hands been more than half Unless therefore the advantages to be gained by preserving the leg can be proved to be so great as to counterbalance this great difference in the risk of the operation of amputation, as performed in a large London hospital, I think it will be difficult to prove that the operation ot excision should, as a rule, be preferred." (p. 143.)
"We think this argument is pressed rather too far, but the fact is a strikin" one, and the data on which it is asserted are deserving of very attentive sifting. If the cases on which the returns were made are fairly co in parabfe with each other, one great argument for their favourite operation will be taken out of the mouths of the part.sa.is of resection, and it will be restricted to exceptional cases. Mr. Bryant is not, however, by any means an opponent of conservative surgery and speaks in terms of just praise of resection of the elbow and shoulder.
A few pages on loose bodies in the joints, and a valuable chapter on inflammation external to the joints, and "bursitis (a barbarous word, which we trust will not pass into common use), conclude the portion of the work which treats of diseases of the joints, the whole being summed up in a statistical table (p. 174), which shows indeed the large materials from which Mr. Bryant has worked, since it includes 974 cases, but is useless for drawing any practical deduction as to the success of treatment, the majority of the cases (538) being returned as "relieved," a vague way of saying that the case was left incomplete, jReviews.
[Jan.
A. large part of the book consists of a collection of cases of injuries of the joints. It is very practical and good; but we fear that our space will not permit of our noticing it more particularly.
In taking leave of Mr. Bryant we would express a sincere hope that we have not seen the last of him. We have not been able to speak in terms of unqualified praise of this work; but we have no doubt that if Mr. Bryant will bestow rather more pains on the style of his writings, and if he will avoid the error committed here, of overlaying his matter with a heap of material,' his great experience and diligent observation will secure to him deserved success as a surgical writer. |
Nanomaterials-Based Electrochemical Immunosensors
With the development of nanomaterials and sensor technology, nanomaterials-based electrochemical immunosensors have been widely employed in various fields. Nanomaterials for electrode modification are emerging one after another in order to improve the performance of electrochemical immunosensors. When compared with traditional detection methods, electrochemical immunosensors have the advantages of simplicity, real-time analysis, high sensitivity, miniaturization, rapid detection time, and low cost. Here, we summarize recent developments in electrochemical immunosensors based on nanomaterials, including carbon nanomaterials, metal nanomaterials, and quantum dots. Additionally, we discuss research challenges and future prospects for this field of study.of the nanoparticles near the Fermi surface will change from the quasi-continuous energy level to the discrete energy level, or the energy gap becomes wider after the size of the nanoparticle is as small as a certain value, which results in the thermal, electrical, optical, acoustic, magnetic, and superconducting properties of the particles are significantly different from conventional materials[10]. In addition, nanomaterials also have volume effects[11], dielectric effects[12,13], and so on.Nanomaterials have multiple methods for classification, depending on the discipline and perspective. Nanomaterials can be divided into three categories, according to the dimensions of the basic units. The first category is zero-dimensional nanomaterial, such as quantum dots and atomic clusters, all of which are in the order of nanometers[14,15]. The second is one-dimensional nanometers nanomaterials, such as nanorods and nanowires, which have two dimensions in the three-dimensional space[16,17]. The last category is two-dimensional nanomaterials, such as superlattices and nano-films[18,19]. The methods for preparing nanomaterials have been continuously developed and enriched since the successful development of metal nanoparticles in the 1970s. Owing to the great application potential of nanomaterials, how to prepare nanomaterials with excellent performance, high efficiency, and low cost is still a hot spot for international researchers[20,21]. At present, the main preparation methods of nanomaterials include physical methods, comprehensive methods, and chemical methods. The physical methods include mechanical grinding, evaporation condensation, laser beam, and ion sputtering[22][23][24]. The chemical methods include microemulsion methods, electrochemical deposition method, complex decomposition method, and hydrothermal method[25][26][27]. The comprehensive method mainly contains the laser gas phase synthesis method, ultrasonic chemical method, and plasma enhanced chemical deposition method[28,29].The electrochemical immunosensor based on antigen-antibody immunoreactivity is the largest class of electrochemical methods being used for protein analysis in biological research and clinical testing. Traditional immunoassays include enzyme-linked immunosorbent assays [30], immunoblot analysis [31], radio immunoassays, and immune electrophoresis[32,33], which are mainly based on the changes of signal that are generated by specific binding of target proteins to known antibodies[34]. Electrochemical immunoassay, as a currently mature technique, has displayed advantages with high sensitivity of electrochemical technology and high specificity of immune recognition reaction, so it can be used in many fields[35,36]. Electrochemical immunoassays can be broadly classified into label-free sensor and labeled sensor. The signal amplification is the core of the preparation electrochemical biosensor, which was realized by nanomaterials modification on the surface of electrode. The immunosensor can selectively identify the analyte by the capture antibody. After that, the label-free electrochemical immunoassay determines the concentration of the analyte by directly measuring the antigen-antibody specific recognition of the change in the electrochemical signal that is generated after binding. The electrochemical impedance signal can effectively reflect various physical and chemical processes that occur on the surface of the electrode, and thus it is often applied for the detection of proteins. The sandwich-type electrochemical immunosensor also used the Ab2, which is often labeled by electrochemical probe and nanomaterials. Last, as the main body of the signal converter, the electrode can derive the identification signal that is generated on the surface of the electrode and convert it into an electrical signal, including current, voltage, and resistance, which can be measured and analyzed in order to achieve the qualitative or quantitative analysis of the analysis target.Figure 1shows the operating principle of electrochemical immunosensor.
# Introduction
Nanotechnology has gradually become an independent and comprehensive research field since the late 1980s, including microscopy, microelectronics, bioanalytical technology, nanomechanics, and electronics. The rapid development of many disciplines, such as biology, has also enabled researchers to obtain deeper understanding regarding the macro and micro worlds. Materials whose structural units are in the nanometer scale (1-100 nm) in at least one dimension or assembled in this range are called nanomaterials or nanostructured materials. When compared with traditional materials, the structure and properties of nanomaterials have changed in essence. At present, nanomaterials have been widely recognized as "the most promising materials in the 21st century" and they are widely used in various fields, such as catalysts, biomedical materials, luminescent materials, insulating materials, and building materials .
When compared with conventional materials, nanomaterials exhibit special properties, including: (1) surface effect: the smaller the diameter of the nanoparticles will have the larger the ratio of the number of surface atoms to the total number of atoms, which in turn causes a sudden change in the properties of the nanoparticles. The concentration of atoms on the surface of the particles will increase the surface energy, as well as the dangling bonds, and can cause the insufficient coordination of the surface atoms, making it easy to combine with other atoms and enhance chemical activity; (2) the macroscopic quantum tunneling effect: according to the classical mechanical principle, microscopic nanoparticles can pass through the barrier, because the total kinetic energy of nanoparticles is less than the barrier height. The ability of the nanoparticle to penetrate the barrier is called macroscopic quantum tunneling, which is the basis of future microelectronic devices, including electrochemical immunosensor. When microelectronic devices are further miniaturized, the quantum effects must be considered, because it establishes the limit of size; (3) quantum size effect: the electron energy level
## Nanomaterials based electrochemical immunosensor
The use of nanomaterials to modify the electrode surface can not only significantly enhance the ability of the electrode to transport electrons, but also improve the adsorption capacity of the electrode surface to bioactive substances, and then shorten the detection time thanks to the advantages of good biocompatibility, high surface reactivity, and large specific surface area of nanomaterials. There are currently three types of nanomaterials that are mainly used in the modification of electrode of electrochemical immunosensor: carbon nanomaterials, metal nanomaterials, and quantum dots.
## Electrochemical immunosensor based on carbon nanomaterials
Adams first proposed carbon-based electrodes in 1958. In addition to the reproducibility of working electrodes by simple sanding, carbon or carbon-based materials have many other advantages, including ease of preparation and catalysts. Carbon nanomaterials can be uniformly dispersed in aqueous solution with good stability and low electrical resistance. With the use of carbon nanomaterials (graphene, carbon nanotubes, fullerenes, etc.) in the electrochemical immunosensor, the signal-to-noise ratio of the reaction occurring between the interfaces and the sensitivity of the biosensor is improved.summarizes the carbon nanomaterials-based electrochemical immunosensors.
Graphene, which is a single atomic thick sheet of graphite composed of sp 2 bonded carbon, has been widely used in electrochemical immunosensor since its first discovery in 2004. Pham et al.developed a label-free electrochemical immunosensor for the detection of microRNAs (miRNA) while using a conducting polymer/reduced graphene oxide (CP/RGO)-modified electrode to detect miR-141 (a prostate biomarker) and miR-29b-1 (a lung cancer biomarker). They employed two specific RNA−DNA antibodies to recognize miRNA−DNA heteroduplexes and square wave voltammetry to detect the redox signal in order to verify its selectivity. Whereafter, this group reported another electrochemical immunosensor for the determination of these two miRNAs based on RGO-carbon nanotubes modified gold electrode. The secondary antibody was labeled by horseradish peroxidase (HRP) and hydroquinone was used as signal molecular. The performance of the prepared immunosensor was more excellent than classical optical detection. The study of the performance of immunosensors in the above papers was focused on selectivity and sensitivity, so the reproducibility and stability were not discussed. Hu et al.used streptavidin-functionalized nitrogen-doped graphene (NG) for the first time to fabricate a highly sensitive electrochemical immunosensor for the detection of tumor markers. The bio-functionalized NG not only showed excellent electrochemical performance, but it also adsorbed more antibodies. Combining the secondary antibody labeled by HRP, the prepared CEA immunosensor revealed satisfactory results when being applied to the detection in human serum samples. Recently, Sanati-Nezhad et al.reported an immunosensor that was based on polyethylenimine (PEI) coated graphene screen-
## Nanomaterials based electrochemical immunosensor
The use of nanomaterials to modify the electrode surface can not only significantly enhance the ability of the electrode to transport electrons, but also improve the adsorption capacity of the electrode surface to bioactive substances, and then shorten the detection time thanks to the advantages of good biocompatibility, high surface reactivity, and large specific surface area of nanomaterials. There are currently three types of nanomaterials that are mainly used in the modification of electrode of electrochemical immunosensor: carbon nanomaterials, metal nanomaterials, and quantum dots.
## Electrochemical immunosensor based on carbon nanomaterials
Adams first proposed carbon-based electrodes in 1958. In addition to the reproducibility of working electrodes by simple sanding, carbon or carbon-based materials have many other advantages, including ease of preparation and catalysts. Carbon nanomaterials can be uniformly dispersed in aqueous solution with good stability and low electrical resistance. With the use of carbon nanomaterials (graphene, carbon nanotubes, fullerenes, etc.) in the electrochemical immunosensor, the signal-to-noise ratio of the reaction occurring between the interfaces and the sensitivity of the biosensor is improved.summarizes the carbon nanomaterials-based electrochemical immunosensors.
Graphene, which is a single atomic thick sheet of graphite composed of sp 2 bonded carbon, has been widely used in electrochemical immunosensor since its first discovery in 2004. Pham et al.developed a label-free electrochemical immunosensor for the detection of microRNAs (miRNA) while using a conducting polymer/reduced graphene oxide (CP/RGO)-modified electrode to detect miR-141 (a prostate biomarker) and miR-29b-1 (a lung cancer biomarker). They employed two specific RNA−DNA antibodies to recognize miRNA−DNA heteroduplexes and square wave voltammetry to detect the redox signal in order to verify its selectivity. Whereafter, this group reported another electrochemical immunosensor for the determination of these two miRNAs based on RGO-carbon nanotubes modified gold electrode. The secondary antibody was labeled by horseradish peroxidase (HRP) and hydroquinone was used as signal molecular. The performance of the prepared immunosensor was more excellent than classical optical detection. The study of the performance of immunosensors in the above papers was focused on selectivity and sensitivity, so the reproducibility and stability were not discussed. Hu et al.used streptavidin-functionalized nitrogen-doped graphene (NG) for the first time to fabricate a highly sensitive electrochemical immunosensor for the detection of tumor markers. The bio-functionalized NG not only showed excellent electrochemical performance, but it also adsorbed more antibodies. Combining the secondary antibody labeled by HRP, the prepared CEA immunosensor revealed satisfactory results when being applied to the detection in human serum samples. Recently, Sanati-Nezhad et al.reported an immunosensor that was based on polyethylenimine (PEI) coated graphene screen-printed electrode for the highly sensitive determination for glial fibrillary acidic protein (GFAP), which is a marker of central nervous system injury. The label-free and quantitative detection of GFAP was realized through the method of electrochemical impedance spectroscopy. This prepared immunosensor could be used for the rapid monitoring of central nervous system (CNS) injury in clinic. printed electrode for the highly sensitive determination for glial fibrillary acidic protein (GFAP), which is a marker of central nervous system injury. The label-free and quantitative detection of GFAP was realized through the method of electrochemical impedance spectroscopy. This prepared immunosensor could be used for the rapid monitoring of central nervous system (CNS) injury in clinic. Carbon nanotubes (CNTs), which are also known as Buckytubes, were composed of wellordered cylinders of sp 2 -hybridized carbon atoms that were first discovered by S.Iijima in 1991 and belong to the fullerene carbon system. The hexagonal carbon atoms were first to constitute the graphene, which is curled into several layers at a certain spiral angle of coaxial nano-scale round tubes. The two ends of the CNTs are basically sealed, whose diameter and length are generally in the range of 2 to 100 nm and micrometers, respectively. According to the number of layers constituting the sheet structure of CNT, the CNTs mainly include two types of single-walled carbon nanotubes (SWCNTs) and multi-walled carbon nanotubes (MWCNTs). SWCNTs, which are also known as fullerenes tubes, are hollow cylinders that crimped from a layer of graphene sheets. The MWCNTs are composed of two or more layers of coaxial circular tubular graphene sheets, and a fixed distance of about 0.34 nm that was caused by van der Waals forces is maintained between the layers. F. Rusling et al.reported an electrochemical sandwich immunosensor that was based on single wall carbon nanotube (SWNT) forests with attached capture antibodies (Ab1) and multiwall carbon nanotubes-HRP labeled second antibody to detect IL-6. The prepared immunosensor could accurately measure the secreted IL-6 from a wide range of HNSCC cells, which is in agreement with standard enzyme linked immunosorbent assays (ELISA). Krishnan et al.designed a voltammetric immunosensor for the diagnosis of type 1 and type 2 diabetic disorders that were based on multi-walled carbon nanotube-pyrenebutyric acid frameworks on edge plane pyrolytic graphite electrodes (PGE/MWNT/Py) attached anti-insulin antibody for the first time. Poly(acrylic acid)-functionalized magnetite nanoparticles (MNP, 100 nm hydrodynamic diameter) was applied to label Ab2. The electrochemical signal of the immunosensor was decreased when it was taken in the electrolyte solution. Carbon nanotubes (CNTs), which are also known as Buckytubes, were composed of well-ordered cylinders of sp 2 -hybridized carbon atoms that were first discovered by S.Iijima in 1991 and belong to the fullerene carbon system. The hexagonal carbon atoms were first to constitute the graphene, which is curled into several layers at a certain spiral angle of coaxial nano-scale round tubes. The two ends of the CNTs are basically sealed, whose diameter and length are generally in the range of 2 to 100 nm and micrometers, respectively. According to the number of layers constituting the sheet structure of CNT, the CNTs mainly include two types of single-walled carbon nanotubes (SWCNTs) and multi-walled carbon nanotubes (MWCNTs). SWCNTs, which are also known as fullerenes tubes, are hollow cylinders that crimped from a layer of graphene sheets. The MWCNTs are composed of two or more layers of coaxial circular tubular graphene sheets, and a fixed distance of about 0.34 nm that was caused by van der Waals forces is maintained between the layers. F. Rusling et al.reported an electrochemical sandwich immunosensor that was based on single wall carbon nanotube (SWNT) forests with attached capture antibodies (Ab1) and multiwall carbon nanotubes-HRP labeled second antibody to detect IL-6. The prepared immunosensor could accurately measure the secreted IL-6 from a wide range of HNSCC cells, which is in agreement with standard enzyme linked immunosorbent assays (ELISA). Krishnan et al.designed a voltammetric immunosensor for the diagnosis of type 1 and type 2 diabetic disorders that were based on multi-walled carbon nanotube-pyrenebutyric acid frameworks on edge plane pyrolytic graphite electrodes (PGE/MWNT/Py) attached anti-insulin antibody for the first time. Poly(acrylic acid)-functionalized magnetite nanoparticles (MNP, 100 nm hydrodynamic diameter) was applied to label Ab2. The electrochemical signal of the immunosensor was decreased when it was taken in the electrolyte solution. Bian et al.developed a novel electrochemiluminescence (ECL) immunosensor for the detection of N-terminal pro-brain natriuretic peptide (NT-proBNP) based on N-(aminobutyl)-N-(ethylisoluminol) (ABEI)-functionalized MWCNTs/gold nanodots (ABEI/COOH-MWCNTs/chitosan/GNDs) hybrid modified ITO electrode, as shown in. The principle of the immnosensor was also that the signal of ECL would decrease when NT-proBNP was captured on the surface of electrode by its antibody. The prepared immunosensor also displayed satisfactory performance when quantifying the NT-proBNP in practical plasma samples, which revealed that the ABEI/COOH-MWCNTs/chitosan/GNDs nanomaterial was a superior electrochemical sensing platform. Recently, Omidi et al.constructed a simple electrochemical immunosensor for the determination of carcinoma antigen 125 (CA125) based on chitosan-gold nanoparticle/multiwall carbon nanotube/graphene oxide (CS-AuNP/MWCNT/GO) platform. The lactate oxidase is applied as the single-enzyme label in electrochemical immunosensor for the first time. The immunosensor was realized in the detection of CA125 in human serum.
Yuan et al.employed the inner redox activity of fullerene (C 60 ) to construct an electrochemical immunoassay for doping detection. They first decorated the C 60 nanoparticles by polyamidoamine (C 60 NPs-PAMAM) to enhance its hydrophilicity and modification site. After that, the AuNPs were linked on C60NPs-PAMAM and used as nanoprobe to label antibodied. CNPs-Au-PAMAM would produce redox electrochemical signal when tetraoctylammonium bromide (TOAB) aroused the inner redox activity of C 60 . Recently, Kumar et al.designed a novel electrochemical redox platform that was based on sesamol-quinone/carbon nanoblack modified platform (GCE/CB@Ses-Qn). In this work, they compared the cyclic voltammetry curve (CV) responses of different carbon nanomaterials modified electrodes, including multi-walled carbon nanotube (MWCNT), activated charcoal, double-walled carbon nanotube, GO, single-walled carbon nanotube, carboxylic acid-functionalized MWCNT, and CB. The results demonstrated that the sensor that was based on CB displayed the highest current signal and excess surface values of sesamol loading. In the end, the platform was used to fabricate an electrochemical immunosensor for the detection of white spot syndrome virus while using HRP-linked secondary antibody. detection of N-terminal pro-brain natriuretic peptide (NT-proBNP) based on N-(aminobutyl)-N-(ethylisoluminol) (ABEI)-functionalized MWCNTs/gold nanodots (ABEI/COOH-MWCNTs/chitosan/GNDs) hybrid modified ITO electrode, as shown in. The principle of the immnosensor was also that the signal of ECL would decrease when NT-proBNP was captured on the surface of electrode by its antibody. The prepared immunosensor also displayed satisfactory performance when quantifying the NT-proBNP in practical plasma samples, which revealed that the ABEI/COOH-MWCNTs/chitosan/GNDs nanomaterial was a superior electrochemical sensing platform. Recently, Omidi et al.constructed a simple electrochemical immunosensor for the determination of carcinoma antigen 125 (CA125) based on chitosan-gold nanoparticle/multiwall carbon nanotube/graphene oxide (CS-AuNP/MWCNT/GO) platform. The lactate oxidase is applied as the single-enzyme label in electrochemical immunosensor for the first time. The immunosensor was realized in the detection of CA125 in human serum. Yuan et al.employed the inner redox activity of fullerene (C60) to construct an electrochemical immunoassay for doping detection. They first decorated the C60 nanoparticles by polyamidoamine (C60NPs-PAMAM) to enhance its hydrophilicity and modification site. After that, the AuNPs were linked on C60NPs-PAMAM and used as nanoprobe to label antibodied. C60NPs-Au-PAMAM would produce redox electrochemical signal when tetraoctylammonium bromide (TOAB) aroused the inner redox activity of C60. Recently, Kumar et al.designed a novel electrochemical redox platform that was based on sesamol-quinone/carbon nanoblack modified platform (GCE/CB@Ses-Qn). In this work, they compared the cyclic voltammetry curve (CV) responses of different carbon nanomaterials modified electrodes, including multi-walled carbon nanotube (MWCNT), activated charcoal, double-walled carbon nanotube, GO, single-walled carbon nanotube, carboxylic acid-functionalized MWCNT, and CB. The results demonstrated that the sensor that was based on CB displayed the highest current signal and excess surface values of sesamol loading. In the end, the platform was used to fabricate an electrochemical immunosensor for the detection of white spot syndrome virus while using HRP-linked secondary antibody. As an important part of nanomaterials, metal nanomaterials have the characteristics and properties of nanomaterials, such as macroscopic quantum tunneling, quantum size effects, and surface effects. It has been used in various fields, such as biosensors and electronic products. The main preparation methods of metal nanomaterials include physical preparation methods and chemical preparation methods. The physical methods mainly include evaporation condensation, sputtering, and mixed plasma method; the chemical methods mainly include phase chemical synthesis, chemical reduction method, electrochemical method, and liquid phase chemical reduction method.summarizes the metal nanomaterials-based electrochemical immunosensors.
At present, gold (Au), platinum (Pt), silver (Ag), palladium (Pd), titanium (Ti), and other metal nanomaterials have been widely used in the active modification of the electrode surface to prepare immunosensor. Main functions of metal nanomaterials can be summarized, as the following. The first function is that metal nanomaterial can be used as a signal molecule to label the bioactive substances. In addition, metal nanomaterial can be used as an excellent carrier to adsorb bioactive As an important part of nanomaterials, metal nanomaterials have the characteristics and properties of nanomaterials, such as macroscopic quantum tunneling, quantum size effects, and surface effects. It has been used in various fields, such as biosensors and electronic products. The main preparation methods of metal nanomaterials include physical preparation methods and chemical preparation methods. The physical methods mainly include evaporation condensation, sputtering, and mixed plasma method; the chemical methods mainly include phase chemical synthesis, chemical reduction method, electrochemical method, and liquid phase chemical reduction method.summarizes the metal nanomaterials-based electrochemical immunosensors.
At present, gold (Au), platinum (Pt), silver (Ag), palladium (Pd), titanium (Ti), and other metal nanomaterials have been widely used in the active modification of the electrode surface to prepare immunosensor. Main functions of metal nanomaterials can be summarized, as the following. The first function is that metal nanomaterial can be used as a signal molecule to label the bioactive substances. In addition, metal nanomaterial can be used as an excellent carrier to adsorb bioactive substances because of its large specific surface area and biocompatibility. Metal nanomaterials can also directly participate in the reaction of the electrode surface as a reactant or catalyst. Many metal nanomaterials (such as Cu and Pt) are themselves high-quality electrochemical reaction substrates or catalysts, which not only increase the peak value of the reaction current, but also reduce the potential of the reaction and amplify the signal intensity. Last, the metal nanomaterials can enhance electron transport as an enzyme mediator for electron transport. For example, when the enzyme sensor is working, the active center of the enzyme is often buried inside the enzyme, which is not conducive to the electron transfer between the catalytic active center and the electrode surface, which thereby affects the sensitivity and detection time of the sensor. The metal nanoparticles have strong electrical conductivity and it can serve as an electron mediator between the enzyme and the electrode surface. The experiments have shown that metal nanoparticle-modified electrodes can significantly accelerate the electron transport speed.
Bai et al.completed the simultaneous detection of prostate specific antigen (PSA) and a-fetoprotein (AFP) based on Au nanoparticles modified polymer brush (poly (acrylonitrile-g-glycidyl methacrylate)) (AuNPs/PGMA-g-PAN). The antibodies of PSA and AFP were labeled by anthraquinone-2-carboxylic acid (Aq) and ferrocenecarboxylic acid (FeC-COOH), respectively, and the developed immunosensor demonstrated good performance, even if in clinical serum analysis. Wei et al.developed an immunoassay for the detection of carcinoembryonic antigen (CEA) based on Ag nanoparticles-molybdenum disulfide-reduced graphene oxide (Ag/MoS 2 /rGO) nanocomposites platform, which could catalyze hydrogen peroxide (H 2 O 2 ) and produce the electrochemical signal. The prepared immunosensor displayed a low detection limit (1.6 fg/mL) while using amperometric i-t curve with acceptable reproducibility, selectivity, and excellent stability. Following the report, Wei et al.employed Pd nanocubes functionalized magnetic graphene sheet (Pd-Fe 3 O 4 -GS) and silicon dioxide (SiO 2 ) to label the secondary antibodies (Ab2) of the target detector. The developed immunosensor displayed an ultrasensitive and specific detection of human immunoglobulin G (IgG), owing to the superior electrochemical catalytic ability to H 2 O 2 of the matrix composed by which used to immobilized the primary antibodies (Ab1).
Bimetallic or polymetallic nanomaterials are increasingly popular among researchers for preparation of immunosensor, owing to the synergistic effects of metal composite nanomaterials. For example, Ju et al.synthesized β-CD functionalized gold−palladium bimetallic nanoparticles (AuPd−CD) in aqueous solution and used it as a platform to detect the adamantine, which was labeled with antibody (ADA-Ab), as shown in. AuPd nanoparticles could catalyzed NaBH 4 and produce an ultrasensitive response to the target chloramphenicol (CAP). The host−guest interaction strategy between CD and ADA provided a universal labeling approach for the ultrasensitive detection of small molecule targets. Recently, Wei et al.fabricated dendritic platinum−copper nanoparticles (PtCu NPs), which were loaded on titanium dioxide octahedral composites (Cu 2 O@TiO 2 -NH 2 ) in order to realize the signal amplification of the insulin electrochemical immunosensor. The PtCu NPs-Cu 2 O@TiO 2 was used to label Ab2. The immunosensor acquired satisfactory results when detecting insulin in human serum based on the platform of AuNPs /MoS 2 loaded Ab1.
ultrasensitive detection of small molecule targets. Recently, Wei et al.fabricated dendritic platinum−copper nanoparticles (PtCu NPs), which were loaded on titanium dioxide octahedral composites (Cu2O@TiO2-NH2) in order to realize the signal amplification of the insulin electrochemical immunosensor. The PtCu NPs-Cu2O@TiO2 was used to label Ab2. The immunosensor acquired satisfactory results when detecting insulin in human serum based on the platform of AuNPs /MoS2 loaded Ab1. ultrasensitive detection of small molecule targets. Recently, Wei et al.fabricated dendritic platinum−copper nanoparticles (PtCu NPs), which were loaded on titanium dioxide octahedral composites (Cu2O@TiO2-NH2) in order to realize the signal amplification of the insulin electrochemical immunosensor. The PtCu NPs-Cu2O@TiO2 was used to label Ab2. The immunosensor acquired satisfactory results when detecting insulin in human serum based on the platform of AuNPs /MoS2 loaded Ab1. In addition to metal nanoparticles, metal oxides are also a large class of excellent nanomaterials for the construction of electrochemical immunosensors. For example, Song et al.fabricated a label-free immunosensor for the detection of cancer biomarker α-fetoprotein (AFP), while taking advantage of iridium oxide (IrOx, 0 ≤ x ≤ 2) nanofibers that were prepared by a simple one-spinneret electrospinning method. The IrOx nanofibers displayed superior performance and provided a highly stable matrix for the conjugation of chitosan (CS), which facilitate the immobilization of antibody. The immunosensor has also shown satisfactory results when determining AFP in human serum. Wei et al.chose the amino functionalized Co 3 O 4 @MnO 2 -thionine (Co 3 O 4 @MnO 2 -Th) as secondary label, which could greatly enhance the electrochemical response signal in order to develop a sandwich-type electrochemical immunosensor for the determination of alpha fetoprotein (AFP). Titanium oxide nanoclusters functionalized nitrogen-doped reduced graphene oxide (TiO 2 -NGO) decorated by Au@Pd holothurian-shaped nanoparticles (Au@Pd HSs) were used to modify the surface of electrode for the detection of human epididymis specific protein 4 antigen (HE4 Ag). The perfect catalytical activity to H 2 O 2 of Au@Pd HSs could endow the superior performance to the developed electrochemical immunosensor. Dendritic gold-PCEPy ALP CTX LOD: 1 ng/mL --Researchers have found that the morphology and structure of metal nanomaterials also have an important impact on their performance with the wide application of metal nanomaterials in electrode modification. Therefore, how to realize the controllability of the size, structure, and morphology of metal nanomaterials, as well as the application of metal nanomaterials with special structural morphology to the field of biosensing, has attracted great interest from researchers in recent years. Most metal nanomaterials with special morphology in the fabrication of a sensor used its own catalytic capacity to the targets, for example, hydrogen peroxide, glucose, and ascorbic acid. There are also have several reports regarding electrochemical immunosensors that are based on mental nanomaterials with special morphology. For example, Yuan et al.synthesized netlike Au nanostructure using β-cyclodextrins and poly(amidoamine) as platform. The prepared Fc-Fc was absorbed by the β-CD to form Fc-Fc/β-CD/PAMAM−Au, which was attached to the Ab2. The netlike nanostructure could attach large amounts of the β-CD and Ab2; meanwhile, the nanomaterial will amplify the current signal. The constructed immunosensor was used to detect the procalcitonin (PCT) with perfect performance. Researchers have found that the morphology and structure of metal nanomaterials also have an important impact on their performance with the wide application of metal nanomaterials in electrode modification. Therefore, how to realize the controllability of the size, structure, and morphology of metal nanomaterials, as well as the application of metal nanomaterials with special structural morphology to the field of biosensing, has attracted great interest from researchers in recent years. Most metal nanomaterials with special morphology in the fabrication of a sensor used its own catalytic capacity to the targets, for example, hydrogen peroxide, glucose, and ascorbic acid. There are also have several reports regarding electrochemical immunosensors that are based on mental nanomaterials with special morphology. For example, Yuan et al.synthesized netlike Au nanostructure using β-cyclodextrins and poly(amidoamine) as platform. The prepared Fc-Fc was absorbed by the β-CD to form Fc-Fc/β-CD/PAMAM−Au, which was attached to the Ab2. The netlike nanostructure could attach large amounts of the β-CD and Ab2; meanwhile, the nanomaterial will amplify the current signal. The constructed immunosensor was used to detect the procalcitonin (PCT) with perfect performance. Furthermore, metal nanomaterials with the morphology of nanowires, nanorod, or dendritic structure as one-dimensional nanometers nanomaterials are applied in the construction of electrochemical immunosensor. For example, the Ag nanowires were prepared for capturing Ab1 and thionine(TH)-doped mesoporous ZnO nanostrawberries (MP-ZnO) were used to immobilize the HRP-anti-IgG by Wang et al.. The super conductivity of silver nanowires enhanced the electrochemical signal, which demonstrated the advantage of onedimensional (1D) nanomaterial. Zhu et al.used platinum nanowire inlaid globular SBA-15 (Pt NWs@g-SBA-15/Thi) as the signal probe and double-deck gold film (D-Au film) as electrode modified nanomaterials to develop an electrochemical for detecting hepatitis B surface antigen (HBs Ag). The nanomaterial could accelerate the electron transfer on the electrode interface, owing to the tunneling effect between the two Au films. The advantage of Pt NWs@g-SBA-15/Thi is that the nanocomposite could reduce the spatial limitation and then load more Ab2 and provide abundant catalytically active sites for analyst. Recently, Chiles et al.reported an onchip electrochemical immunosenor in order to test the cholera toxin subunit B(CTX) based on a dendritic gold architecture that was modified through poly(2-cyanoethyl)pyrrole (PCEPy), which was synthesized by electrodeposition for 20 min. in a solution of 30 mM HAuCl4. The surface area of dendritic gold electrode was 18× greater than a planar gold surface which allowing for a higher sensitivity. The prepared immunosensor meted the demand for point-ofcare for the diagnosis of cholera, which demonstrated the value of the reported dendritic gold architecture. Furthermore, metal nanomaterials with the morphology of nanowires, nanorod, or dendritic structure as one-dimensional nanometers nanomaterials are applied in the construction of electrochemical immunosensor. For example, the Ag nanowires were prepared for capturing Ab1 and thionine(TH)-doped mesoporous ZnO nanostrawberries (MP-ZnO) were used to immobilize the HRP-anti-IgG by Wang et al.. The super conductivity of silver nanowires enhanced the electrochemical signal, which demonstrated the advantage of one-dimensional (1D) nanomaterial. Zhu et al.used platinum nanowire inlaid globular SBA-15 (Pt NWs@g-SBA-15/Thi) as the signal probe and double-deck gold film (D-Au film) as electrode modified nanomaterials to develop an electrochemical for detecting hepatitis B surface antigen (HBs Ag). The nanomaterial could accelerate the electron transfer on the electrode interface, owing to the tunneling effect between the two Au films. The advantage of Pt NWs@g-SBA-15/Thi is that the nanocomposite could reduce the spatial limitation and then load more Ab2 and provide abundant catalytically active sites for analyst. Recently, Chiles et al.reported an on-chip electrochemical immunosenor in order to test the cholera toxin subunit B(CTX) based on a dendritic gold architecture that was modified through poly(2-cyanoethyl)pyrrole (PCEPy), which was synthesized by electrodeposition for 20 min. in a solution of 30 mM HAuCl 4. The surface area of dendritic gold electrode was 18× greater than a planar gold surface which allowing for a higher sensitivity. The prepared immunosensor meted the demand for point-of-care for the diagnosis of cholera, which demonstrated the value of the reported dendritic gold architecture.
## Electrochemical immunosensor based on quantum dots
Quantum dot (QD) that was composed of III-V atoms (such as GaAs, InP) or II-VI atoms (such as CdTe, CdS), also known as semiconductor nanocrystal, is a uniform inorganic nanoscale particle that is generally within 10 nm in diameter with a core/shell structure. QD is an excellent tool to prepare an electrochemical immunosensor, because it has all of the special properties of nanomaterials, including quantum size effects, surface effects, and unique characteristic of high electron density. Several researchers have reported the use of voltammetry to detect QDs as electrochemical labels, taking advantage of their redox properties.summaries the QD nanomaterials-based electrochemical immunosensors.
## Electrochemical immunosensor based on quantum dots
Quantum dot (QD) that was composed of III-V atoms (such as GaAs, InP) or II-VI atoms (such as CdTe, CdS), also known as semiconductor nanocrystal, is a uniform inorganic nanoscale particle that is generally within 10 nm in diameter with a core/shell structure. QD is an excellent tool to prepare an electrochemical immunosensor, because it has all of the special properties of nanomaterials, including quantum size effects, surface effects, and unique characteristic of high electron density. Several researchers have reported the use of voltammetry to detect QDs as electrochemical labels, taking advantage of their redox properties.summaries the QD nanomaterials-based electrochemical immunosensors. A novel magneto-controlled electrochemical immunosensor was constructed for the sensitive detection of low-abundance protein (IgG1) with a sandwich-type assay that was based on CdS QD-doped bovine serum albumin (QD-BSA) and IgG1-functionalized magnetic bead. The electrochemical signal was studied while using anodic stripping voltammetric analysis of cadmium ion released by acid from quantum dot. Lin et al.developed an electrochemical immunosensor for the detection of organophosphorylated butyrylcholinesterase (OP-BChE), which is a specific biomarker for exposure to toxic organophosphorus agents. QD was employed to tag anti-BChE antibody in order to amplify the signal. The prepared immnosensor displayed a highly selective and sensitive response to the target. Gong et al.reported electrochemical immunosensors for the detection of prostate specific antigen (PSA) that was based on QD functionalized graphene sheets (GS-QD). The immunosensor was capable of detecting PSA in serum samples. Employing carbon QDs-graphene oxide-PEI-Au nanohybrid (CQDs-PEI-GO/AuNPs) as a probe to sense carbohydrate antigen 15-3 (CA15-3) was used to develop a novel electrochemiluminescence immunosensor. The Ab1 was linked on Ag nanoparticles and polydopamine (AgNPs-PDA), which had large surface area. The immusensor exhibited excellent performance, owing to the synergistic effect of the nanocomposite materials. A miniaturized electrochemical immunosensor was established by using 8-channel screen-printed carbon arrays and IgG labeled with CdSe/ZnS QDs for the detection of anti-transglutaminase antibodies (a celiac disease biomarker, tTG) in human sera. The developed miniaturized electrochemical immunosensor is easier to use for clinic detection. Several electrochemical immunosensors while using QD are the focus on the detection of marker proteins on the surface of cell. For example, aiming to realize the simultaneous determination of EpCAM and GPC3 antigens on the surface of the human liver cancer cell line, M. Hui et al.took advantage of CdTe QD-coated silica nanoparticles and ZnSe QD-coated silica nanoparticles to link the antibodies of EpCAM and GPC3, respectively. Chitosan-electrochemically reduced graphene oxide film (CS-GO) modified glassy carbon electrode (GCE) was used to immobilize the Ab1. The immunosensor showed superior reproducibility, accuracy, and stability. Similarly, Zhu et al.reported a dual-signal-marked electrochemical immunosensor for the simultaneous detection of Bcell lymphoma 2 (Bcl-2) and Bcl-2-associated X protein (Bax), which are often used to monitor the apoptosis of tumor cells and to evaluate the cancer drug effect. CdSeTe@CdS QDs and Ag Several electrochemical immunosensors while using QD are the focus on the detection of marker proteins on the surface of cell. For example, aiming to realize the simultaneous determination of EpCAM and GPC3 antigens on the surface of the human liver cancer cell line, M. Hui et al.took advantage of CdTe QD-coated silica nanoparticles and ZnSe QD-coated silica nanoparticles to link the antibodies of EpCAM and GPC3, respectively. Chitosan-electrochemically reduced graphene oxide film (CS-GO) modified glassy carbon electrode (GCE) was used to immobilize the Ab1. The immunosensor showed superior reproducibility, accuracy, and stability. Similarly, Zhu et al.reported a dual-signal-marked electrochemical immunosensor for the simultaneous detection of B-cell lymphoma 2 (Bcl-2) and Bcl-2-associated X protein (Bax), which are often used to monitor the apoptosis of tumor cells and to evaluate the cancer drug effect. CdSeTe@CdS QDs and Ag nanoclusters were used to connect with Ab2 of Bcl-2 and Bax, respectively. The electrochemical signal was amplified by mesoporous silica and graphene, which could provide abundant surface area for absorbance of probes and Ab1. Under anodic stripping voltammetry detection, Cd and Ag would be oxidized and produce currents under different potential, which represents the concentration of Bcl-2 and Bax, respectively.
Micromachines 2019, 10, x 15 of 21. Fabrication of the dual-signal-marked electrochemical immunosensor for the detection of Bcl-2 and Bax. First, reduced graphene oxide (RGO) was used to modify glassy carbon electrode (GCE) to increase surface area and then linked antibodies I of Bcl-2, Bax. After bovine serum albumin (BSA) protein block the nonspecific adsorption sites, the electrode was immersed into the mixed antigens to capture the active Bcl-2 and Bax proteins. Finally, the electrode was reacted with antibody Π-targeted QDs and Ag nanoclusters as the signal probes, which were proportional to targets on the electrode. Reproduced with permission from.
# Conclusion and outlook
In conclusion, we reviewed the recent developments in nanomaterials-based electrochemical immunosensor, mainly including carbon, metal, and QD nanomaterials. With the continuous advancement of nanotechnology, the electrochemical immunosensor has been greatly developed. A compound to improve their water solubility, biocompatibility, and the ability to attach antibody, which were applied to modify the surface of electrode, mainly modified these nanomaterials. In addition, the nanomaterials are also used to label Ab2, which could enhance the sensitivity of electrochemical signal. Most reports have been realized the analysis in real sample while using electrochemical immunosensors. In clinical applications, electrochemical immunosensors could be utilized in the early detection of tumor and cancer biomarkers.
Currently, researchers in the field seem to be focusing on the implementation of optimal performance limits for sensors and the lowest detection threshold at the small molecule level. However, we should also pay attention to electrochemical immunosensor not yet becoming the mainstream of practical application for some technical and commercial reasons. Long-time reproducibility and stability are needed to improve, because the electrode interface system is relatively complex. Fromwe could demonstrate that the reproducibility and stability of many prepared electrochemical immunosensors were not mentioned. It is a challenge in the. Fabrication of the dual-signal-marked electrochemical immunosensor for the detection of Bcl-2 and Bax. First, reduced graphene oxide (RGO) was used to modify glassy carbon electrode (GCE) to increase surface area and then linked antibodies I of Bcl-2, Bax. After bovine serum albumin (BSA) protein block the nonspecific adsorption sites, the electrode was immersed into the mixed antigens to capture the active Bcl-2 and Bax proteins. Finally, the electrode was reacted with antibody Π-targeted QDs and Ag nanoclusters as the signal probes, which were proportional to targets on the electrode. Reproduced with permission from.
# Conclusions and outlook
In conclusion, we reviewed the recent developments in nanomaterials-based electrochemical immunosensor, mainly including carbon, metal, and QD nanomaterials. With the continuous advancement of nanotechnology, the electrochemical immunosensor has been greatly developed. A compound to improve their water solubility, biocompatibility, and the ability to attach antibody, which were applied to modify the surface of electrode, mainly modified these nanomaterials. In addition, the nanomaterials are also used to label Ab2, which could enhance the sensitivity of electrochemical signal. Most reports have been realized the analysis in real sample while using electrochemical immunosensors. In clinical applications, electrochemical immunosensors could be utilized in the early detection of tumor and cancer biomarkers.
Currently, researchers in the field seem to be focusing on the implementation of optimal performance limits for sensors and the lowest detection threshold at the small molecule level. However, we should also pay attention to electrochemical immunosensor not yet becoming the mainstream of practical application for some technical and commercial reasons. Long-time reproducibility and stability are needed to improve, because the electrode interface system is relatively complex. Fromwe could demonstrate that the reproducibility and stability of many prepared electrochemical immunosensors were not mentioned. It is a challenge in the functionalization of the electrode interface to improve the reproducibility and stability. Therefore, we need to consider at least three aspects when preparing an electrochemical immunosensor to overcome the problem of stability: (1) the immobilization of antibody is a pivotal step, because the antibody acts as the recognition element for antibody-antigen reaction. Selecting appropriate binding method is important, for example, the oriented antibody molecular layers are often used to improve the efficiency of binding. Huang et al.reported an immunosensor for AFP detection, the current response of which could maintain about 96.2% of the original signal after 40 days. The good stability of the immunosensor was owed to the microenvironment that was provided by the layer-by-layer assembly film and the polymer PEDOT offered a stable substrate for the immobilization of antibodies. (2) Maintaining excellent stability of the probe is also vital for the electrochemical immunosensor. The functionalized performance of the nanomaterials that were used for adsorbing probe or the label of Ab2 should be stable and the method for linking is also required to be steady. For example, Zhang et al.constructed an electrochemical sensor with good stability by using the Thi-CNTs sensing platform, which could prevent the leak of hydrophilic thionine from the electrode and avoid the addition of mediator to the solution, so the activity of the antibody could be kept for a long time. (3) How to actually apply the test results under ideal laboratory conditions to the detection of real samples and realize the demand of market for portable rapid detection technology.
Furthermore, the deep integration of electrochemical detection technology with nanotechnology, microfluidic technology, and other related technologies, including screen-printing and patterning, will definitely improve the performance of electrochemical immunosensors. The development of electrochemical immunosensor may present the trends of miniaturization and portability. Small and low-cost nanomaterials-based electrochemical sensing devices will be more popular in the near future. |
Dominant-negative ATF5 rapidly depletes survivin in tumor cells
Survivin (BIRC5, product of the BIRC5 gene) is highly expressed in many tumor types and has been widely identified as a potential target for cancer therapy. However, effective anti-survivin drugs remain to be developed. Here we report that both vector-delivered and cell-penetrating dominant-negative (dn) forms of the transcription factor ATF5 that promote selective death of cancer cells in vitro and in vivo cause survivin depletion in tumor cell lines of varying origins. dn-ATF5 decreases levels of both survivin mRNA and protein. The depletion of survivin protein appears to be driven at least in part by enhanced proteasomal turnover and depletion of the deubiquitinase USP9X. Survivin loss is rapid and precedes the onset of cell death triggered by dn-ATF5. Although survivin downregulation is sufficient to drive tumor cell death, survivin over-expression does not rescue cancer cells from dn-ATF5-promoted apoptosis. This indicates that dn-ATF5 kills malignant cells by multiple mechanisms that include, but are not limited to, survivin depletion. Cell-penetrating forms of dn-ATF5 are currently being developed for potential therapeutic use and the present findings suggest that they may pose an advantage over treatments that target only survivin.
# Introduction
The transcription factor ATF5 is over-expressed in a variety of cancer types and such over-expression correlates with poor prognosis and treatment resistance [bib_ref] Histology-based expression profiling yields novel prognostic markers in human glioblastoma, Dong [/bib_ref] [bib_ref] Selective destruction of glioblastoma cells by interference with the activity or expression..., Angelastro [/bib_ref] [bib_ref] The transcription factor ATF5 is widely expressed in carcinomas, and interference with..., Monaco [/bib_ref] [bib_ref] The transcription factor ATF5: role in neurodevelopment and neural tumors, Greene [/bib_ref] [bib_ref] A genome-wide RNA interference screen reveals an essential CREB3L2-ATF5-MCL1 survival pathway in..., Sheng [/bib_ref] [bib_ref] An activating transcription factor 5-mediated survival pathway as a target for cancer..., Sheng [/bib_ref] [bib_ref] HSP70 protein promotes survival of C6 and U87 glioma cells by inhibition..., Li [/bib_ref] [bib_ref] Overexpression of activating transcription factor 5 in human rectal cancer, Kong [/bib_ref] [bib_ref] Interference with ATF5 function enhances the sensitivity of human pancreatic cancer cells..., Hu [/bib_ref] [bib_ref] ATF5 is overexpressed in epithelial ovarian carcinomas and interference with its function..., Chen [/bib_ref] [bib_ref] Activating transcription factor 5 enhances radioresistance and malignancy in cancer cells, Ishihara [/bib_ref] [bib_ref] Role of ATF5 in the invasive potential of diverse human cancer cell..., Nukuda [/bib_ref] [bib_ref] Targeting ATF5 in Cancer, Angelastro [/bib_ref] [bib_ref] miR-141-3p functions as a tumor suppressor modulating activating transcription factor 5 in..., Wang [/bib_ref] [bib_ref] Mitochondrial dysfunction in cancer: Potential roles of ATF5 and the mitochondrial UPR, Deng [/bib_ref] [bib_ref] Human cytomegalovirus immediate-early protein promotes survival of glioma cells through interacting and..., Hu [/bib_ref] [bib_ref] Activating Transcription Factor-5 Knockdown Reduces Aggressiveness of Mammary Tumor Cells and Attenuates..., Ben-Shmuel [/bib_ref] [bib_ref] Expression of activating transcription factor 5 (ATF5) is increased in astrocytomas of..., Feldheim [/bib_ref]. ATF5 is a member of the activating transcription factor family and possesses a basic leucine zipper domain that is required for association with selected partners as well as DNA binding and transcriptional activity [bib_ref] Selective destruction of glioblastoma cells by interference with the activity or expression..., Angelastro [/bib_ref] [bib_ref] The transcription factor ATF5 is widely expressed in carcinomas, and interference with..., Monaco [/bib_ref] [bib_ref] The transcription factor ATF5: role in neurodevelopment and neural tumors, Greene [/bib_ref] [bib_ref] Mouse Atf5: molecular cloning of two novel mRNAs, genomic organization, and odorant..., Hansen [/bib_ref] [bib_ref] Regulation of asparagine synthetase gene transcription by the basic region leucine zipper..., Sarraj [/bib_ref] [bib_ref] Data-driven prediction and design of bZIP coiled-coil interactions, Potapov [/bib_ref] [bib_ref] The transcription factor ATF5: role in cellular differentiation, stress responses, and cancer, Sears [/bib_ref]. To interfere with ATF5 function, we developed a dominantnegative (dn) form of this protein that retains the native leucine zipper, but is N-terminally truncated and modified in the basic domain to abolish DNA binding and to potentially extend the leucine zipper [bib_ref] Regulated expression of ATF5 is required for the progression of neural progenitor..., Angelastro [/bib_ref]. When expressed in tumor cells in vitro or in vivo, dn-ATF5 promotes their apoptotic death and causes tumor regression [bib_ref] Selective destruction of glioblastoma cells by interference with the activity or expression..., Angelastro [/bib_ref] [bib_ref] The transcription factor ATF5 is widely expressed in carcinomas, and interference with..., Monaco [/bib_ref] [bib_ref] The transcription factor ATF5: role in neurodevelopment and neural tumors, Greene [/bib_ref] [bib_ref] Interference with ATF5 function enhances the sensitivity of human pancreatic cancer cells..., Hu [/bib_ref] [bib_ref] ATF5 is overexpressed in epithelial ovarian carcinomas and interference with its function..., Chen [/bib_ref] [bib_ref] Targeting ATF5 in Cancer, Angelastro [/bib_ref] [bib_ref] Regulated ATF5 loss-of-function in adult mice blocks formation and causes regression/eradication of..., Arias [/bib_ref]. To generate a dn-ATF5 with therapeutic potential, we designed a form fused to a cell-penetrating "penetratin" [bib_ref] Penetratin story: an overview, Dupont [/bib_ref] domain that can be produced either as a recombinant protein [bib_ref] Regression/eradication of gliomas in mice by a systemicallydeliverable ATF5 dominant-negative peptide, Cates [/bib_ref] or by synthesis [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref]. The presence of the cellpenetrating domain permits the penetratin-dn-ATF5 fusion peptide (CP-dn-ATF5) to rapidly pass through tissue barriers and to enter cells [bib_ref] Regression/eradication of gliomas in mice by a systemicallydeliverable ATF5 dominant-negative peptide, Cates [/bib_ref]. Both recombinant and synthetic forms of CP-dn-ATF5 effectively promote apoptosis of a wide variety of tumor cells in vitro and in vivo [bib_ref] Regression/eradication of gliomas in mice by a systemicallydeliverable ATF5 dominant-negative peptide, Cates [/bib_ref] [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref]. In contrast, neither plasmid nor penetratin delivered dn-ATF5 appears to affect survival of nontransformed cells and there have been no evident sideeffects in mice treated with either the recombinant or synthetic forms of CP-dn-ATF5 at doses that suppress tumor growth [bib_ref] Selective destruction of glioblastoma cells by interference with the activity or expression..., Angelastro [/bib_ref] [bib_ref] Regulated ATF5 loss-of-function in adult mice blocks formation and causes regression/eradication of..., Arias [/bib_ref] [bib_ref] Regression/eradication of gliomas in mice by a systemicallydeliverable ATF5 dominant-negative peptide, Cates [/bib_ref] [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref].
The mechanisms by which dn-ATF5 promotes tumor cell death are only partially understood. dn-ATF5 decreases levels of anti-apoptotic proteins BCL2 and MCL1 in tumor cells by what appears to be both decreased transcription and elevated protein destabilization [bib_ref] A genome-wide RNA interference screen reveals an essential CREB3L2-ATF5-MCL1 survival pathway in..., Sheng [/bib_ref] [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref] [bib_ref] BCL-2 is a downstream target of ATF5 that mediates the prosurvival function..., Dluzen [/bib_ref]. A preliminary survey of additional proteins that regulate cell survival and death indicated that dn-ATF5 may also affect expression of the inhibitor-of-apoptosis protein (IAP) family member survivin (BIRC5, product of the BIRC5 gene). Like ATF5, survivin is highly expressed in multiple tumor types with little expression in most non-transformed cells [bib_ref] A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma, Ambrosini [/bib_ref]. High survivin expression in tumors is correlated with metastasis, resistance to treatment and poor prognosis [bib_ref] Targeting survivin in cancer, Altieri [/bib_ref] [bib_ref] Survivin-The inconvenient IAP. Semin, Altieri [/bib_ref]. In addition to its action as an inhibitor of apoptosis, biological roles for survivin that also appear to contribute to its actions in tumors include regulation of cell cycle and promotion of mitochondrial function [bib_ref] Survivin-The inconvenient IAP. Semin, Altieri [/bib_ref]. Agents that directly or indirectly downregulate survivin levels interfere with the proliferation of cancer cells and promote their apoptotic death and thus, given survivin's absence from most non-transformed cells, it has been widely considered as an attractive potential target for cancer treatment [bib_ref] Targeting survivin in cancer, Altieri [/bib_ref] [bib_ref] Survivin-The inconvenient IAP. Semin, Altieri [/bib_ref] [bib_ref] Survivin-biology and potential as a therapeutic target in oncology, Cheung [/bib_ref] [bib_ref] Survivin beyond physiology: orchestration of multistep carcinogenesis and therapeutic potentials, Athanasoula [/bib_ref] [bib_ref] Targeting survivin for therapeutic discovery: past, present, and future promises, Peery [/bib_ref] [bib_ref] Survivin as a novel target protein for reducing the proliferation of cancer..., Li [/bib_ref] [bib_ref] Therapeutic strategies involving survivin inhibition in cancer, Martinez-Garcia [/bib_ref]. Consequently, there has been substantial effort to identify/generate agents that suppress survivin expression in neoplasias [bib_ref] Survivin-The inconvenient IAP. Semin, Altieri [/bib_ref] [bib_ref] Survivin beyond physiology: orchestration of multistep carcinogenesis and therapeutic potentials, Athanasoula [/bib_ref] [bib_ref] Targeting survivin for therapeutic discovery: past, present, and future promises, Peery [/bib_ref] [bib_ref] Survivin as a novel target protein for reducing the proliferation of cancer..., Li [/bib_ref] [bib_ref] Therapeutic strategies involving survivin inhibition in cancer, Martinez-Garcia [/bib_ref]. To date, no such drug has reached clinical use beyond trials, neither as a mono-or combination therapy. Thus there is a continued need to identify agents that affect survivin expression and that have the potential to be used as safe cancer therapeutics.
# Materials and methods
Cells culture and transfection GBM12 cells were kindly supplied by Dr. Jann Sarkaria (Mayo Clinic). All other cell lines were obtained from the ATCC and authenticated by the supplier. All lines were grown in DMEM supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 U/ml streptomycin. siUSP9X (#6308 S, Cell Signaling Technology, Danvers MA), siSurvivin (#6351, Cell Signaling Technology; (#4390824, Silencer Select S1458, Ambion), siRNA CTR (#6568, Cell Signaling Technology; Silencer TM Select Negative Control, #4390843, Ambion) were transfected into cells using Oligofectamine™ Transfection Reagent (Invitrogen, Waltham MA) following the supplier's protocols. All plasmids were transfected by using Lipofectamine™ 3000 (Invitrogen) following the supplier's protocols.
## Plasmids
FLAG-tagged human survivin cDNA cloned into a pCMV6-entry vector was obtained from Origene, Rockville MD (#RC205935). The plasmid used for FLAGsurvivin over-expression was pLVX-EF1α-IRES-mCherry (#631987, Takara Bio USA, Mountainview CA), a bicistronic lentiviviral vector allowing the expression of the transgene and mCherry under the control of the EF1-α promoter. FLAG-survivin was generated and cloned in the pLVX vector using primers AAGAATTC (EcoRI) ATGGGTGCCCCGACGTTG and AATCTAGA(XbaI) TTACTTATCGTCGTCATC. GFP-BCL2 37 was a gift from Clark Distelhorst (Addgene plasmid # 17999; http:// n2t.net/addgene:17999; RRID:Addgene_17999).
Indicated experiments employed wild-type and mutant pCMV-1A-3xFLAG-dn-ATF5. To generate these constructs, DNA optimized for human codon usage with a 5'-BamHI site and a 3'-XhoI site were synthesized as gBlock fragments (Integrated DNA Technologies Inc, Skokie IL) encoding the wild-type dn-ATF5 sequence, MASMTGG QQMGRDPDLEQRAEELARENEELLEKEAEELEQENAE LEGECQGLEARNRELRERAESVEREIQYVKDLLIEVYK ARSQRTRSA, or encoding a mutant form of dn-ATF5, MASMTGGQQMGRDPDGEQRAEEGARENEEGGEKE AEEGEQENAEGEGECQGGEARNREGRERAESVEREIQ YVKDGGIEVYKARSQRTRSA in which the indicated (bolded) leucines were replaced with glycines to inactivate leucine zipper activity. The fragments were subcloned into the BamH1 and XhoI site of pCMV-3Tag-1A (Agilent Technologies Inc, Santa Clara CA) plasmid for in frame N-terminal 3XFlag-tagged expression of dn-ATF5 or mutant dn-ATF5. Where indicated, experiments additional employed pLe-FLAG-GFP-dn-ATF5 as previously described [bib_ref] Regulated expression of ATF5 is required for the progression of neural progenitor..., Angelastro [/bib_ref].
## Lentivirus preparation
Lentivirus were prepared in HEK293 cells by cotransfecting pLVX expression plasmids along with second generation lentiviral packaging plasmids using the calcium phosphate transfection method as previously described. Lentiviral particles were collected, then concentrated using Lenti-X concentrator (#631231, Takara), resuspended in PBS, and stored at −80°C.
For lentiviral infection, 0.1 up to 5 × 10 7 viral particles were added per cm 2 of culture area, directly in the culture medium. The transduced cells were analyzed by qPCR, Western blot or flow cytometry after 3-5 days.
## Flow cytometry
Cells were dissociated with trypsin, and washed with cold PBS twice, then resuspended in 1 X binding buffer at a concentration of 1 × 10 6 cells/ml. About 100 µl of the solution (1 × 10 5 cells) were incubated with 5 µl of FITC Annexin V and 5 µl PI (#556547, BD, Franklin Lakes NJ) for 15 min at room temperature in the dark. The cell suspension was mixed with 400 µl of 1X binding buffer and analyzed by flow cytometry.
qPCR Cells were lysed and total RNA was purified using TRI regent (Molecular Research Center, Cincinnati OH) following the manufacturer's protocol. 1 µg of mRNA was used for the synthesis of cDNA using the first-strand cDNA synthesis kit (Origene). qPCR was performed using FastStart SYBR Green Master Mix (Roche, Indianapolis IN) by using the following primer pairs: h-Survivin, CCACTGAGAACGAGCCAGACTT and GTATTACAG GCGTAAGCCACCG; h-18S RNA, AGTCCCTGCCCT TTGTACACA and GATCCGAGGGCCTCACTAAAC. The relative mRNAs levels for genes of interest were normalized to 18 S RNA.
## Western immunoblotting
Cells were homogenized in cell lysis buffer (#9803, Cell Signaling Technology, Danvers MA) with protease inhibitor mixture (#11836170001, Roche). Protein samples were prepared in X2 loading buffer (#161-0737, BioRad, Portland ME) with βME following the manufacturer's instruction. Protein was loaded and separated by electrophoresis and then transferred onto PVDF membranes (Bio-Rad). Blots were probed with the following primary antibodies: rabbit anti-survivin (#2808, Cell Signaling Technology), anti-FLAG (#8146, Cell Signaling Technology), mouse anti-ACTIN (#3700, Cell Signaling Technology), rabbit anti-USP9X (#5751, Cell Signaling Technology). All band intensities were determined using ImageJ and normalized to ACTIN signal.
## Immunofluorescence
Cells were fixed for 10 min in 4% PFA, washed 3 times with PBS, and blocked with Superblock (#37515 Thermo Fisher Scientific, Waltham MA) with 0.3% Triton-X for 30 min at room temperature. The following primary antibodies were used for immunofluorescence: rabbit anti-FLAG (#14793, Cell Signaling Technology), rabbit anti-GFP (# PA5-22688, Invitrogen), rabbit anti-survivin (#2808, Cell Signaling Technology). Hoechst 33328 was used to stain nuclei. Images were acquired using a Zeiss epifluorescence microscope equipped with a digital camera and Axiovision software.
# Results
Expression of dn-ATF5 depletes nuclear survivin expression in tumor cells prior to promotion of apoptotic cell death
We previously reported that expression of dn-ATF5encoding plasmids promotes apoptotic death of a wide range of tumor cell types, but not of non-transformed cells 2,3,24 . As an initial approach to assessing the potential effect of dn-ATF5 on survivin expression, we transfected PC3 prostate tumor cells with a previously described construct expressing GFP-FLAG-dn-ATF5 23 and determined total survivin levels by western immunoblotting after 3 days [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref]. This revealed an approximate 40% loss of survivin protein expression (not accounting for a transfection efficiency of 50-70%).
Similar results were achieved by transfecting T98G glioblastoma cells with 3xFLAG-dn-ATF5 (Supplementary [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref]. We also carried out a parallel transfection with a mutated form of this construct in which the critical leucine residues in the leucine zipper were replaced with glycine residues. We previously found that such replacements in a dn form of ATF5 compromises its capacity to promote cell death [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref]. In contrast to the nonmutated construct, the mutant construct failed to downregulate survivin [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref]. Taken together, these findings indicate that dn-ATF5 reduces survivin expression in tumor cells and that this requires a functional leucine zipper domain.
To further assess effects of dn-ATF5 on survivin expression and localization at the cellular level, we transfected T98G and LN229 glioblastoma cells with a GFP-FLAG-dn-ATF5 construct or with a control plasmid expressing only GFP. One day after transfection, transfected (GFP+) cells were immunostained for survivin expression and blindly scored for proportions with either no, cytoplasmic/nuclear (cells with evident survivin expression in both nuclei and cytoplasm) or nuclear-only localized survivin [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref]. In both lines, about 60% of control cells showed nuclear-only localization, about 25% cytoplasmic/nuclear staining and about 15% no detectable survivin staining [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref]. In contrast, there was an approximate 3-fold increase in proportion of cells without detectable survivin expression (to almost 50% of the population) [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref] after transfection with GFP-FLAGdn-ATF5. This appeared to come mainly at the expense of nuclear survivin for which there was a corresponding loss in proportion of cells with immunostaining in this compartment [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref]. These effects were sustained in that they were also evident 3-5 days after transfection [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref].
As noted above, dn-ATF5 expression triggers extensive apoptotic death of tumor cells. This raised the question as to whether survivin depletion might either precede or be a result of the death process. We therefore monitored the proportions of cells with apoptotic nuclei at 1-5 days after transfection. In agreement with prior findings 2 , this revealed no increase in apoptosis compared to controls at 24 h after transfection [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref] , a time when survivin expression was already diminished. This contrasted with later times when both loss of survivin and increased apoptotic death were apparent [fig_ref] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death [/fig_ref]. Thus, survivin appears to be lost prior to the appearance of apoptotic nuclei and to be sustained during the death process.
## Cell penetrant synthetic cp-dn-atf5 depletes survivin protein in multiple tumor cell lines
We previously described a synthetic version CP-dn-ATF5 that contains a penetratin domain that permits rapid cellular uptake of the peptide and that promotes apoptotic death of a wide variety of tumor types in vitro and in vivo [bib_ref] Regression/eradication of gliomas in mice by a systemicallydeliverable ATF5 dominant-negative peptide, Cates [/bib_ref] [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref]. We therefore next tested whether this potential therapeutic prototype form of dn-ATF5 would also affect tumor cell survivin protein levels. CP-dn-ATF5 has the advantage that it rapidly reaches all cells in cultures and that the effects on survivin can therefore be readily quantified. We initially treated T98G, U87 and LN229 cells with 50, 100, and 200 µM CP-dn-ATF5 for 24 h and assessed relative survivin levels. These concentrations of CP-dn-ATF5 have little effect on cell survival or numbers at 24 h, and trigger apoptotic death in a dose-responsive manner beginning at 2 days and strongly manifest at 3 days of treatment [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref]. The data reveal a consistent loss of survivin protein within 24 h with some degree of variability from experiment to experiment [fig_ref] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in... [/fig_ref]. survivin depletion of over 90% in some cases [fig_ref] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in... [/fig_ref] and [fig_ref] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in... [/fig_ref].
We also examined the onset time of survivin protein loss with CP-dn-ATF5 at 100 µM in T98G cultures. There was no significant effect at 4 h, but reduction of survivin levels was evident at 8 h and beyond [fig_ref] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in... [/fig_ref]. To determine whether this depletion of survivin might be the result of cell death, we also assessed cell viability over time and confirmed no loss of cells at 24 h with death evident only at 48 and 72 h [fig_ref] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in... [/fig_ref]. Taken together, our data indicate that CP-dn-ATF5 depletes survivin levels in multiple tumor lines and appears to do so well before the appearance of cell death. To determine whether CP-dn-ATF5 affects levels of survivin mRNA as well as protein, we carried out qPCR on the 12 cell lines listed above after a single dose exposure to 50, 100, or 200 µM peptide for 48 h. In each case, there was substantial depletion of survivin mRNA, with a range of responsiveness among the lines [fig_ref] Figure 3: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin mRNA in... [/fig_ref].
Additional analyses of survivin mRNA in multiple lines showed time and dose-dependent loss of survivin mRNA at 24 and 72 h of treatment with changes apparent within 24 h [fig_ref] Figure 3: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin mRNA in... [/fig_ref]. A time course study of survivin mRNA levels in T98G cells [fig_ref] Figure 3: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin mRNA in... [/fig_ref] parallel to that described above for survivin protein [fig_ref] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in... [/fig_ref] revealed that survivin transcripts were already substantially decreased by 4 h of CP-dn-ATF5 exposure. This well precedes the onset of cell death in such cells and occurs before depletion of survivin protein [fig_ref] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in... [/fig_ref].
## Cp-dn-atf5 depletes virally-expressed exogenous survivin protein in multiple tumor cell lines
While our results indicate that CP-dn-ATF5 causes survivin mRNA depletion, the possibility remained that CP-dn-ATF5 also causes survivin protein loss by promoting its turnover. To assess this, we infected multiple tumor lines with lentivirus expressing FLAG-tagged survivin driven by a CMV promoter. Despite expression of the protein from a heterologous promoter, exposure to 100 µM CP-dn-ATF5 for 3 d caused substantial reduction in expression of the exogenous survivin protein in all lines tested [fig_ref] Figure 4: CP-dn-ATF5 promotes the turnover of survivin protein [/fig_ref]. This observation suggests that CP-dn-ATF5 causes depletion of survivin protein at least partly by mechanisms independent from its effects on survivin mRNA.
## Cp-dn-atf5 destabilizes cellular survivin protein and promotes survivin loss via proteasomal degradation
We next examined whether, as appears for overexpressed survivin, CP-dn-ATF5-promoted reduction of endogenous survivin also involves a non-transcriptional mechanism. Prior reports indicate that cellular survivin levels are subject to post-transcriptional regulation by degradation [bib_ref] The ubiquitinproteasome pathway regulates survivin degradation in a cell cycle-dependent manner, Zhao [/bib_ref] [bib_ref] The proapoptotic F-box protein Fbxl7 regulates mitochondrial function by mediating the ubiquitylation..., Liu [/bib_ref]. To assess whether CP-dn-ATF5 affects survivin stability, we pretreated T98G cells with or without this peptide for 24 h and then exposed them to the translational inhibitor cycloheximide for various times before determining relative survivin protein levels by western immunoblotting. As shown in [fig_ref] Figure 4: CP-dn-ATF5 promotes the turnover of survivin protein [/fig_ref] , c, this revealed that CP-dn-ATF5 significantly accelerates survivin turnover under conditions when survivin synthesis is To approach the issue of the mechanism by which dn-ATF5 affects survivin stability, we generated T98G cells stably over-expressing FLAG-survivin on a heterologous promoter. To determine whether the post-translational component of the effect of CP-dn-ATF5 on survivin is mediated by the proteasome, we exposed the survivin over-expressing cells to 100 µM CP-dn-ATF5 for 24 h in presence or absence of the proteasomal inhibitor epoxomicin [fig_ref] Figure 5: Proteasomal turnover and USP9X contribute to the depletion of survivin protein by... [/fig_ref]. Under such circumstances, FLAGsurvivin levels decreased with CP-dn-ATF5 alone by about 50% compared with untreated controls; with epoxomicin alone, FLAG-survivin levels rose by about three-fold and this level was not reduced by CP-dn-ATF5 co-treatment. Taken together, these data thus indicate that CP-dn-ATF5 decreases survivin stability and support the idea that enhanced proteasomal turnover contributes to survivin depletion caused by CP-dn-ATF5.
## Usp9x depletion contributes to the cp-dn-atf5-mediated decrease in survivin stability
We next explored the mechanism by which CP-dn-ATF5 treatment leads to survivin protein destabilization. The deubiquitinase USP9X has been shown to deubiquinate survivin and thus potentially to inhibit its proteasomal degradation [bib_ref] Chromosome alignment and segregation regulated by ubiquitination of survivin, Vong [/bib_ref]. Although USP9X downregulation in HeLa cells was reported without effect on survivin protein levels [bib_ref] Chromosome alignment and segregation regulated by ubiquitination of survivin, Vong [/bib_ref] , similar downregulation of USP9X in SF188 and Panc-1 cells resulted in survivin protein depletion Our past work has shown that CP-dn-ATF5 dramatically reduces USP9X levels in multiple tumor cell lines including T98G cells [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref]. We therefore asked whether such an effect of CP-dn-ATF5 on USP9X in T98G cells could contribute to the depletion of survivin seen with this agent. Accordingly, we used a previously described USP9X-directed siRNA [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref] to knockdown USP9X and compared survivin levels with those in cells receiving a control siRNA. This revealed a significant loss of survivin protein [fig_ref] Figure 5: Proteasomal turnover and USP9X contribute to the depletion of survivin protein by... [/fig_ref] , but not survivin mRNA [fig_ref] Figure 5: Proteasomal turnover and USP9X contribute to the depletion of survivin protein by... [/fig_ref] , thus indicating that the reduction of USP9X caused by CP-dn-ATF5 treatment likely contributes to survivin protein depletion.
## Is survivin depletion by cp-dn-atf5 sufficient to account for the anti-tumor actions of this agent?
Numerous studies have documented that reduction of survivin levels is sufficient to promote death of various tumor cell types [bib_ref] Targeting survivin in cancer, Altieri [/bib_ref] [bib_ref] Survivin-The inconvenient IAP. Semin, Altieri [/bib_ref] [bib_ref] Survivin-biology and potential as a therapeutic target in oncology, Cheung [/bib_ref] [bib_ref] Targeting survivin for therapeutic discovery: past, present, and future promises, Peery [/bib_ref]. We confirmed this with an siRNA targeting survivin in T98G, HCT116, MCF7, and MDA-MB-468 cells that induced apoptotic death in each case [fig_ref] Figure 5: Proteasomal turnover and USP9X contribute to the depletion of survivin protein by... [/fig_ref]. However, while direct targeting of survivin expression has been raised as a potential treatment for a variety of cancers, to this point, no successful anti-survivin monotherapy has emerged. In this context, we thus asked whether the capacity of CP-dn-ATF5 to deplete survivin is necessary for its ability to promote tumor cell death or whether this agent also has additional death-promoting activities. To address this, we first overexpressed FLAG-survivin in T98G tumor cells and then assessed the capacity of CP-dn-ATF5 to promote their death. As shown in [fig_ref] Figure 6: Survivin over-expression is not sufficient to rescue tumor cells from apoptotic death... [/fig_ref] , although CP-dn-ATF5 b CP-dn-ATF5 accelerates the turnover of survivin protein in T98G cells. Replicate cultures were treated for 24 h with or without (Control) 100 µM CP-dn-ATF5 and then exposed to 50 µM cycloheximide for the indicated times in the continued presence or absence of CP-dn-ATF5 and then assessed for relative survivin levels by western immunoblotting. c CP-dn-ATF5 accelerates the turnover of survivin protein in T98G cells. Cultures were treated as in b and relative survivin expression determined vs ACTIN and normalized to the zero time value in each independent experiment. Numbers of independent experiments for various points are as follows: Control, 1 h, n = 5; 2 h, n = 6; 3 h, n = 5; 4 h, n = 6; 12 h, n = 1; 24 h, = 5. CP-dn-ATF5: 1 h, n = 5; 2 h, n = 6; 3 h, n = 5; 4 h, n = 5; 12 h, n = 1, 24 h, n = 6. Mean data points were fitted to an exponential curve reduced the expression of the FLAG-survivin as anticipated, even under these conditions exogenous survivin was expressed at far higher levels than the corresponding endogenous protein. Despite such high survivin overexpression, it failed to significantly protect T98G cells from death caused by 3 days of CP-dn-ATF5 treatment as indicated by morphology [fig_ref] Figure 6: Survivin over-expression is not sufficient to rescue tumor cells from apoptotic death... [/fig_ref] , percentage of apoptotic cells as determined by flow analysis (6 C), or cell numbers (6D). We also observed similar results by flow analysis of U87 cells and by cell counts of LN229, GBM12, MDA-MB-231, U251, and DU145 cultures transfected to over-express survivin [fig_ref] Figure 6: Survivin over-expression is not sufficient to rescue tumor cells from apoptotic death... [/fig_ref]. Finally, to control for the possibility that CP-dn-ATF5 kills survivin over-expressing tumor cells by a mechanism apart from its dn-ATF5 activity, we also asked whether survivin over-expression would rescue cells transfected with the pLe-GFP-FLAG-dn-ATF5 plasmid. As shown in [fig_ref] Figure 6: Survivin over-expression is not sufficient to rescue tumor cells from apoptotic death... [/fig_ref] , over-expressed survivin also failed to protect from plasmid-delivered dn-ATF5. These findings thus support the conclusion that while CP-dn-ATF5 rapidly depletes cells of survivin, it has additional actions that can promote cell death that cannot be rescued by survivin overexpression.
The inability of survivin over-expression to rescue tumor cells from the lethal effects of dn-ATF5 exposure raised the question of whether CP-dn-ATF5 might kill tumor cells by a non-apoptotic mechanism. Our past work showed that tumor cell death promoted by transfected dn-ATF5 plasmid is apoptotic in nature and is blocked by inhibition of caspase activity 2 . We confirmed this for CP-dn-ATF5 in the present study in which the caspase inhibitor zVAD rescued cell numbers and the occurrence of apoptotic nuclei from the effects of 3 days of peptide exposure . Past work has shown that dn-ATF5 expression leads to downregulation of pro-survival BCL2 9,27,28 . We therefore additionally assessed whether BCL2 over-expression would protect T98G cells from 3 days of CP-dn-ATF5 treatment. While Bcl2 over-expression reduced the presence of apoptotic nuclei, such protection from CP-dn-ATF5 was only partial . Taken together, these findings support the idea that dn-ATF5 promotes caspase- or pLVX-EF1α-FLAG-survivin-IRES-mCherry (pLV-survivin in as indicated and 24 h later treated with or without 100 µM CP-dn-ATF5 for 3 d and assessed for exogenous and endogenous survivin by western immunoblotting with anti-survivin. Long exposure shows levels of endogenous survivin. b Survivin over-expression does not rescue the effects of CP-dn-ATF5 on number or appearance of T98G cells. Cultures were infected with lentiviruses described in panel a as indicated and 24 h later were treated with 100 µM CP-dn-ATF5 for 3 days. Panels show immunofluoresence for mcherry (red) or phase contrast images. Scale bars = 50 µm. c Survivin over-expression does not rescue T98G cells from apoptotic death promoted by CP-dn-ATF5. Cultures were infected with lentivirus described in panel a and 24 h later were treated with 100 µM CP-dn-ATF5 for 3 days. Cultures were harvested and analyzed for proportion of apoptotic cells by flow cytometry. Left panel shows the representative cytometry data and right panel shows quantitative results from three independent experiments, each in triplicate. d Survivin overexpression does not rescue cell number in T98G cultures treated with CP-dn-ATF5. Cultures were infected with above described lentiviruses and 24 h later were treated with or without 100 µM CP-dn-ATF5 as indicated for 3 days. Cultures were harvested and analyzed for total cell numbers. Data are from three independent experiments, each in triplicate. e Survivin over-expression does not rescue promotion of apoptosis by FLAG-GFP-dn-ATF5 in multiple cancer cell lines. Indicated cell lines were co-transfected as shown with a ratio of pLVX:pLE constructs of 3:1 and transfected (GFP+) cells were assessed for proportion with apoptotic nuclei 3 days later. Data are from three independent experiments dependent apoptotic death of cancer cells and that it does so by multiple pro-apoptotic actions including downregulation of survivin and BCL2. Moreover, owing to the multiplicity of these actions, replacement of a single prosurvival protein such as survivin is insufficient to maintain viability.
# Discussion
The present findings establish that dn-ATF5 consistently depletes survivin expression in a wide variety of human tumor cell lines. This appears to occur by both decreases in levels of survivin mRNA and enhanced survivin turnover. These decreases occur rapidly and are durable. Given survivin's roles in cancer cell proliferation, survival, metastasis, and therapeutic resistance [bib_ref] Targeting survivin in cancer, Altieri [/bib_ref] [bib_ref] Survivin-The inconvenient IAP. Semin, Altieri [/bib_ref] [bib_ref] Survivin-biology and potential as a therapeutic target in oncology, Cheung [/bib_ref] [bib_ref] Targeting survivin for therapeutic discovery: past, present, and future promises, Peery [/bib_ref] [bib_ref] Survivin as a novel target protein for reducing the proliferation of cancer..., Li [/bib_ref] , dn-ATF5-induced survivin depletion is highly likely to contribute to dn-ATF5's anti-tumor actions.
We observed that both plasmid-encoded and a penetratin-linked synthetic forms of dn-ATF5-promoted survivin depletion. The plasmid-encoded form has Nterminal GFP and FLAG tags followed by a linker region and retains the 25 amino acids C-terminal to the leucine zipper domain of ATF5. These sequences are not present in CP-dn-ATF5 [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref] , thus indicating that they are neither required for, nor interfere with, the capacity of dn-ATF5 to deplete survivin. Likewise, addition of the N-terminal penetratin domain in the synthetic form does not appear to interfere with its capacity to regulate survivin expression.
It was also conceivable that plasma-encoded and penetratin-linked dn-ATF5 might be present in different subcellular compartments that would result in distinct effects on survivin levels. This did not appear to be the case, indicating that both forms have similar activities. We also evaluated a form of 3xFLAG-dn-ATF5 mutated in the leucine zipper domain. This mutant form failed to deplete survivin, demonstrating that leucine zipper function is indispensable for dn-ATF5's capacity to affect survivin expression.
One potential issue about the effects of dn-ATF5 on survivin expression was whether this was a response to, rather than contributor to, apoptotic death. Our findings with both transfected and penetratin-linked dn-ATF5 showed depletion of survivin mRNA and protein well before onset of cell death, thus supporting a role for survivin depletion as a primary contributor rather than responder to cell death triggered by dn-ATF5.
Our findings indicate that dn-ATF5 selectively depletes nuclear-localized survivin. It has been reported that survivin turns over more efficiently in nuclei [bib_ref] Nuclear survivin has reduced stability and is not cytoprotective, Connell [/bib_ref] [bib_ref] Survivin withdrawal by nuclear export failure as a physiological switch to commit..., Chan [/bib_ref] [bib_ref] Mitotic activity of survivin is regulated by acetylation at K129, Aljaberi [/bib_ref]. In this light, given that dn-ATF5 rapidly diminishes survivin mRNA levels, loss of nuclear survivin may reflect decreased survivin synthesis that cannot keep up with the level of degradation in this compartment. Additionally, or alternatively, dn-ATF5 may affect survivin movement between the cytoplasmic and nuclear compartments, so that nuclear import is reduced or export is increased. In either case, loss of nuclear localization would suggest that at least one consequence of survivin depletion by dn-ATF5 is defective cell cycle progression.
The mechanisms that underlie the depletion of survivin by dn-ATF5 appear to be complex with both reduction of mRNA levels and post-translational degradation playing roles . It is presently unknown whether survivin is a direct or indirect transcriptional target of ATF5 or whether dn-ATF5 might act by binding and sequestering another partner that influences survivin mRNA or protein expression. Our findings that dn-ATF5 reduces expression of survivin protein encoded by plasmids with heterologous promoters and that it accelerates turnover of endogenous survivin establish a role for a posttranslational mechanism in survivin depletion. Prior work shows that survivin is subject to proteasomal degradation [bib_ref] The ubiquitinproteasome pathway regulates survivin degradation in a cell cycle-dependent manner, Zhao [/bib_ref] [bib_ref] The proapoptotic F-box protein Fbxl7 regulates mitochondrial function by mediating the ubiquitylation..., Liu [/bib_ref]. Consistent with this, we found that the proteasomal inhibitor epoxomicin elevated expression of transfected FLAG-survivin. We also observed that epoxomicin blocked the depletion of FLAG-survivin by CP-dn-ATF5, a finding that supports the idea that one mechanism by which dn-ATF5 reduces survivin is by promoting its proteasomal destruction. Survivin appears to be protected from proteasomal degradation by interaction with partners such as HSP90 [bib_ref] Regulation of survivin function by Hsp90, Fortugno [/bib_ref] and by deubiquitination by enzymes including USP9X [bib_ref] Inhibition of deubiquitinases primes glioblastoma cells to apoptosis in vitro and in..., Karpel-Massler [/bib_ref] [bib_ref] Deubiquitinase USP9X promotes cell migration, invasion and inhibits apoptosis of human pancreatic..., Liu [/bib_ref]. We previously reported that CP-dn-ATF5 substantially reduces expression of USP9X in a variety of tumor cell lines [bib_ref] A synthetic cell-penetrating dominant-negative ATF5 peptide exerts anticancer activity against a broad..., Karpel-Massler [/bib_ref]. This raised the possibility that enhanced survivin turnover by dn-ATF5 is mediated at least in part by USP9X depletion and in support of this, we found that USP9X knockdown in T98G cells results in a significant reduction of survivin protein levels.
Among the issues explored here was whether survivin depletion fully accounts for the apoptotic activity of dn-ATF5. The degree of survivin depletion in many cell lines achieved with dn-ATF5 (in some cases, >90%) was on the order reached by survivin si-or sh-RNA treatments that Proposed role of survivin regulation in the mechanism by which dn-ATF5 kills tumor cells elicit apoptotic tumor cell death. Thus, it appears that the depletion of survivin caused by dn-ATF5 is likely to be sufficient to cause death. On the other hand, our experiments indicated that death of tumor cells promoted by CP-dn-ATF5 was not rescued by survivin overexpression. This suggests that dn-ATF5 triggers proapoptotic events in addition to survivin depletion that are sufficient to kill cancer cells. Thus far, additional identified actions of dn-ATF5 that could contribute to its apoptotic actions include depletion of MCL1 and BCL2 5,10,27,28 and it is highly likely that more remain to be identified. The potential capacity of dn-ATF5 therapeutics to deplete survivin as well as other survival-promoting proteins may represent an advantage over drugs such as siRNAs that target only survivin.
In addition to promoting cancer cell death as monotherapies, agents that reduce survivin expression or activity are reported to sensitize tumor cells to chemotherapeutics or radiation [bib_ref] Targeting survivin in cancer, Altieri [/bib_ref] [bib_ref] Survivin-The inconvenient IAP. Semin, Altieri [/bib_ref] [bib_ref] Survivin-biology and potential as a therapeutic target in oncology, Cheung [/bib_ref] [bib_ref] Survivin beyond physiology: orchestration of multistep carcinogenesis and therapeutic potentials, Athanasoula [/bib_ref] [bib_ref] Targeting survivin for therapeutic discovery: past, present, and future promises, Peery [/bib_ref] [bib_ref] Survivin as a novel target protein for reducing the proliferation of cancer..., Li [/bib_ref]. Because dn-ATF5 depletes survivin in tumor cells, this suggests that it may also serve in combination therapies to overcome therapeutic resistance. This possibility is supported by reports that dn-ATF5 sensitizes pancreatic cancer cells to paclitaxel [bib_ref] Interference with ATF5 function enhances the sensitivity of human pancreatic cancer cells..., Hu [/bib_ref] and that ATF5 promotes radioresistance of lung cancer cells [bib_ref] Activating transcription factor 5 enhances radioresistance and malignancy in cancer cells, Ishihara [/bib_ref].
[fig] Figure 1: Transfection with GFP-FLAG-dn-ATF5 depletes survivin protein prior to onset of cell death. a Transfection with GFP-FLAG-dn-ATF5 depletes survivin in PC3 prostate tumor cells. Cells were transfected as described in methods and assessed 3 days later for relative survivin levels (normalized to ACTIN) by western immunoblotting. Left panel shows representative blot, right panel shows quantification from 3 independent experiments. b Examples of survivin localization in T98G cells. Transfected cultures were immunostained for survivin and co-stained with Hoechst 33828 to visualize nuclear DNA. Blue/white arrowhead shows a cell without significant survivin expression. Red and blue arrowheads show cells with both cytoplasmic (red arrowheads) and nuclear (blue arrowheads) staining. Yellow arrowhead shows a cell with nuclear-only staining. Scale bar = 10 µm. c, d Transfection with GFP-FLAG-dn-ATF5 depletes nuclear-only localized survivin and increases the proportion of cells without detectable survivin expression in T89G and LN229 cultures at 24 h (C) or at 3-5 d (D). Transfected cells (GFP+) were blindly scored for survivin localization using criteria described in panel b. Data are from three independent experiments, each carried out in triplicate. In each experiment about 200-300 cells were evaluated. e Transfection with GFP-FLAG-dn-ATF5 does not increase the incidence of apoptosis in cultures of T89G and LN229 cultures at 24 h, a time when survivin is already depleted. Transfected cells (GFP+) in cultures described in panels c and d were scored for proportion with apoptotic nuclei. Data are from three independent experiments, each carried out in triplicate. In each experiment about 200-300 cells were evaluated. f Transfection with GFP-FLAG-dn-ATF5 promotes cell death at 3-5 days. As in e, but at 5 days for T98G and 3 days for LN229 cells. [/fig]
[fig] Figure 2: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin protein in multiple cancer cell lines prior to loss of cell viability. a, b CP-dn-ATF5 depletes survivin in T98G, U87 and LN229 glioblastoma cell lines at 24 h after treatment. a Representative western blots. b Data are from three independent experiments. c CP-dn-ATF5 depletes survivin in T98G, U87 and LN229 glioblastoma cell lines at 48 and 72 h after treatment. Normalized survivin to ACTIN ratios are shown for each lane. d Treatment with CP-dn-ATF5 causes a rapid depletion of survivin protein in T98G cultures. Normalized survivin to ACTIN ratios are shown for each lane. e Cell death caused by CP-dn-ATF5 (100 µM) in T98G cultures occurs after survivin depletion. Relative numbers of cells per culture were determined at the indicated times. Three replicate cultures were evaluated at each time point CP-dn-ATF5 depletes survivin mRNA in multiple tumor cell lines [/fig]
[fig] Figure 3: Treatment with CP-dn-ATF5 causes a dose-and time-dependent depletion of survivin mRNA in multiple cancer cell lines. a-d CP-dn-ATF5 depletes survivin mRNA in multiple cancer cell lines at 48 h of treatment. Data are from three independent experiments. e CP-dn-ATF5 (100 µM) causes rapid depletion of survivin mRNA in T98G cells. Values are from one experiment carried out in triplicate blocked. These findings thus indicate that survivin depletion caused by dn-ATF5 likely reflects decreased survivin protein stability. [/fig]
[fig] Figure 4: CP-dn-ATF5 promotes the turnover of survivin protein. a CP-dn-ATF5 decreases the expression of exogenous FLAG-survivin. Indicated cell lines were infected with lentivirus expressing FLAGsurvivin and 48 h later were treated with 100 µM CP-dn-ATF5 for 72 h and then assessed for relative (to ACTIN) FLAG-survivin levels by western immunoblotting. Normalized survivin to ACTIN ratios are shown for each lane. [/fig]
[fig] Figure 5: Proteasomal turnover and USP9X contribute to the depletion of survivin protein by CP-dn-ATF5. a Proteasomal inhibitor epoxomicin suppresses loss of survivin expression promoted by CP-dn-ATF5. T98G cells stably expressing FLAG-survivin were treated for 100 µM CP-dn-ATF5 for 24 h in presence or absence (Control, CTR) of 10 nM epoxomicin (EPOX) and assessed by western immunoblotting for levels of FLAG-survivin and ACTIN. Left panel shows representative Western immunoblot. Right panel shows quantification of relative FLAG-survivin levels under each condition. Data are from three independent experiments. b Knockdown of USP9X reduces levels of survivin protein in T98G cells. Cultures were transfected with control (CTR) or USP9X siRNA for 72 h and assessed 3 days later for USP9X, survivin and ACTIN protein levels by western immunoblotting (left panel). Right panel shows relative survivin protein levels for five independent experiments. c USP9X knockdown does not affect survivin mRNA levels in T98G cells. Cultures were treated as in b and assessed for relative survivin mRNA levels. Data are from 3 independent experiments [/fig]
[fig] Figure 6: Survivin over-expression is not sufficient to rescue tumor cells from apoptotic death promoted by dn-ATF5. a CP-dn-ATF5 reduces expression of over-expressed FLAG-survivin, but to levels far in excess of endogenous survivin. T98G cultures were infected with lentivirus expressing either pLVX-EF1α-IRES-mCherry (pLV in [/fig]
|
A Rare Ectopic Ovary Mimicking Colon Sigmoideum Mesenchymoma Presenting as an Intestinal Mesenchymoma
Ectopic ovaries are a rare occurrence. A 33-year-old woman presented to our unit for evaluation of a 2-year history of sporadic abdominal pain that was becoming sharp and frequent. Computed tomography (CT) suggested a gastrointestinal tract mesenchymoma. An abdominal laparotomy was performed and the tumor was excised for pathologic evaluation. A rapid frozen section pathologic examination showed a solitary fibrous tumor (SFT). The final pathology report was an ectopic ovary with corpora lutea bleeding. Ectopic ovaries are benign and the present case is the first report involving an ectopic ovary mimicking a gastrointestinal stromal tumor (GIST). The patient recovered well after surgery. Maldevelopment of the genital tract can lead to ectopic ovaries and surgery is a good management choice. The present case provides a possible differential diagnosis for GISTs.
# Introduction
Ectopic ovaries are rare embryologic abnormalities with an estimated prevalence between 1:29,000 and 1:93,000 gynecologic admissions. Because patients are asymptomatic, it is difficult to diagnose ectopic ovaries. Gastrointestinal stromal tumors (GISTs) are rare tumors which can arise anywhere within the GI tract. Herein, we report the first case of a patient with an ectopic ovary presenting as a GIST and provide the differential diagnosis for GIST.
## Case presentation
A 33-year-old female sought evaluation in our Department of General Surgery with a 2-year history of sporadic abdominal pain that had become aggravated during the past week. The character of pain became sharp and frequent. The pain was localized to the left lower abdomen. There was no nausea and vomiting. There was no history of abdominal trauma. The patient had a congenital anomaly of the kidneys and uterus; there was no menstruation. The patient had undergone an appendectomy in the past. On physical examination, the patient was afebrile. The abdominal examination revealed pain and a mass in the left lower quadrant area upon palpation. The mass was approximately 4 × 5 cm in diameters and was not circumscribed. The patient had no rebound tenderness and muscle rigidity. Laboratory testing revealed the following: white blood cell count, 7.13 × 10 9 /L; neutrophilic granulocytes, 76.8%; hemoglobin, 120 g/L; and platelet count, 322 × 10 9 /L. Computed tomography (CT) revealed an intestinal stromal tumorand pelvic kidneys. Digestive tract radiography showed possible extraintestinal involvement. An intestinal stromal tumor was diagnosed and an abdominal laparotomy was performed; however, the intestinal tract was normal and a mass was noted in the sigmoid flexure. The tumor exhibited exophytic growth without infiltration and was 6.0 × 5.0 × 3.0 cm in size. The tumor and colon (proximal and distal length, 10 cm; ∼25 cm) were excised. A rapid frozen section pathologic examination revealed a solitary fibrous tumor (SFT). A colon anastomosis was performed and the patient had fully recovered 7-days post-operatively. The final diagnosis was an ectopic ovary with corpora lutea bleeding. The patient recovered well after surgery and there were no post-operative complications. The patient was doing well at the 11-month follow-up visit. Written informed consent was obtained from the patient and The Third People's Hospital of Dalian had approved the study (NO. 2018-LW-001).
# Discussion
The patient presented to the hospital for evaluation of aggravated abdominal pain, and the CT scan revealed an intestinal stromal tumor. Intra-operatively, a mass located in the colon was thought to be a colon stromal tumor; however, the final pathologic diagnosis was an ectopic ovary with corpora lutea bleeding.
Ectopic ovaries can be classified as congenital and acquired. The present case belongs to the congenital type. A developmental error occurring during the formation of genital canals and external genitalia in women may induce ectopic ovaries.
The methods by which ectopic ovaries are diagnosed include MRI and surgery; however, surgery is the gold standard (3, 8). MRI can be used to diagnose genital tract and renal system abnormalities. Controlled ovarian stimulation (COH) is thought to aid in the diagnosis of ectopic ovaries; magnetic resonance imaging (MRI) more accurately identifies undescended ovaries in the upper abdomen after COH. In the present study, because the CT scan revealed an intestinal stromal tumor, an MRI was not performed. Ectopic ovaries are usually accompanied by maldevelopment of the genital system and renal tract. The present case had similar maldevelopments: congenital abnormal development of the ovaries and ectopic kidneys. Ectopic ovaries may lead to menstrual disorders, infertility, or abdominal pain. In the present case, because of uterine dysfunction and amenorrhea, an ectopic ovary was not suspected. Ectopic ovaries can be found in the upper abdomen, near the pelvic brim or neighboring inguinal canal. The location of the ovary in the current case was the colon, which is the first such reported case.
GISTs are gastrointestinal mesenchymal tumors accounting for 0.2% of all gastrointestinal tumors. GISTs can originate anywhere in the gastrointestinal tract. Therefore, the present case was initially suspected to be a GIST.
Ectopic ovaries can present as primary infertility (13), a hernia or cyst in the inguinal canal, acute appendicitis (3), ovarian malignancy (15), a Brenner tumor (16), a Wilms' tumor, as well as a GIST. Ectopic ovaries can cause irregular menses and pain, and are often accompanied by an abnormal urinary systemor a mature teratoma.
Patients with developmental anomalies need close attention. An abnormal urinary system is usually accompanied by an abnormal genital system. Although it is difficult for the diagnosis of ectopic ovaries pre-operatively, additional examinations should be performed, such as ultrasonography, MRI, and endoscopy. A multiple disciplinary team (MDT) is also advised.
# Conclusion
The present case is an ectopic ovary mimicking a GIST. Maldevelopment of the genital tract can lead to an ectopic ovary and surgery is a good management choice. We have shared our clinical experience to help guide the management of similar cases and offer a differential diagnosis of GISTs.
## Data availability
All datasets generated for this study are included in the manuscript and/or the supplementary files.
## Consent
Written informed consent was obtained from the patient for publication of this case report and the accompanying images.
# Author contributions
JP and SW: conceptualization. HW: data curation. ZF: investigation, validation, and writing of the original draft. |
Elucidation of a protein-protein interaction network involved in Corynebacterium glutamicum cell wall biosynthesis as determined by bacterial two-hybrid analysis
Mycobacterium species have a highly complex and unique cell wall that consists of a large macromolecular structure termed the mycolyl-arabinogalactan-peptidoglycan (mAGP) complex. This complex is essential for growth, survival and virulence of the human pathogen Mycobacterium tuberculosis, and is the target of several anti-tubercular drugs. The closely related species Corynebacterium glutamicum has proven useful in the study of orthologous M. tuberculosis genes and proteins involved in mAGP synthesis. This study examines the construction of a protein-protein interaction network for the major cell wall component arabinogalactan in C. glutamicum based on the use of a bacterial two-hybrid system. We have identified twenty-four putative homotypic and heterotypic protein interactions in vivo. Our results demonstrate an association between glycosyltransferases, GlfT1 and AftB, and interaction between the sub-units of decaprenylphosphoribose epimerase, DprE1 and DprE2. These analyses have also shown that AftB interacts with AftA, which catalyzes the addition of the first three arabinose units onto the galactan chain. Both AftA and AftB associate with other arabinofuranosyltransferases, including Emb and AftC, that elongate and branch the arabinan domain. Moreover, a number of proteins involved in arabinogalactan biosynthesis were shown to form dimers or multimers. These findings provide a useful recourse for understanding the biosynthesis and function of the mycobacterial cell wall, as well as providing new therapeutic targets.
# Introduction
Mycobacterium tuberculosis the causative agent of tuberculosis (TB), remains a major cause of mortality and morbidity from a single infectious organism. In 2012, approximately 8.6 million people developed TB and 1.3 million died from the disease. Emergence of multidrug-resistant [bib_ref] WHO progress report 2011: towards universal access to diagnosis and treatment of..., Who [/bib_ref] , extensively drug resistant [bib_ref] WHO progress report 2011: towards universal access to diagnosis and treatment of..., Who [/bib_ref] and recently reported totally drug-resistant [bib_ref] Emergence of new forms of totally drug-resistant tuberculosis bacilli: super extensively drug-resistant..., Velayati [/bib_ref] [bib_ref] Totally drug-resistant tuberculosis in India, Udwadia [/bib_ref] [bib_ref] First tuberculosis cases in Italy resistant to all tested drugs, Migliori [/bib_ref] clinical isolates has prompted the need for new drugs and drug targets. M. tuberculosis and other bacteria in the suborder of Corynebacterineae are characterized by a highly complex cell envelope. This cell wall is comprised of a cross-linked peptidoglycan (PG), covalently linked to arabinogalactan (AG) chains, which are further esterified by mycolic acids [bib_ref] A new interpretation of the structure of the mycolylarabinogalactan complex of Mycobacterium..., Besra [/bib_ref] [bib_ref] Evidence for the nature of the link between the arabinogalactan and peptidoglycan..., Mcneil [/bib_ref] [bib_ref] Location of the mycolyl ester substituents in the cell walls of mycobacteria, Mcneil [/bib_ref]. This macromolecular structure is often referred to as the mycolyl-arabinogalactan-peptidoglycan (mAGP) complex [bib_ref] Arabinogalactan and lipoarabinomannan biosynthesis: structure, biogenesis and their potential as drug targets, Jankute [/bib_ref].
AG is composed predominantly of arabinofuranosyl (Araf) and galactofuranosyl (Galf) residues [bib_ref] Demonstration that the galactosyl and arabinosyl residues in the cell-wall arabinogalactan of..., Mcneil [/bib_ref] and is covalently attached to PG via a specialized linker unit, L-Rhap-(1→ 4)-α-D-GlcNAc [bib_ref] Evidence for the nature of the link between the arabinogalactan and peptidoglycan..., Mcneil [/bib_ref]. The galactan domain of AG is composed of approximately 30 alternating β (1→5) and β (1→6) Galf residues connected in a linear fashion [bib_ref] Predominant structural features of the cell wall arabinogalactan of Mycobacterium tuberculosis as..., Daffe [/bib_ref]. Three similar Darabinan chains comprising roughly 30 Araf residues each are attached to the galactan chain [bib_ref] Identification of a novel arabinofuranosyltransferase (AftA) involved in cell wall arabinan biosynthesis..., Alderwick [/bib_ref]. Since the AG structure is essential to M. tuberculosis, many gene deletion studies investigating AG have been performed in the closely related Corynebacterium genus, where aspects of AG biosynthesis are non-essential. Deletion studies in C. glutamicum demonstrated that the arabinan chains of AG are attached distinctively to the 8th, 10th, and 12th residue of the linear galactan chain [bib_ref] Identification of a novel arabinofuranosyltransferase (AftA) involved in cell wall arabinan biosynthesis..., Alderwick [/bib_ref]. Unlike most bacterial polysaccharides, AG lacks repeating units and is composed of a few distinct structural motifs, notably the terminal Ara 6 motif, with the 5-OH of the t-Araf and 2-Araf residues representing sites for mycolylation [bib_ref] Predominant structural features of the cell wall arabinogalactan of Mycobacterium tuberculosis as..., Daffe [/bib_ref] [bib_ref] Location of the mycolyl ester substituents in the cell walls of mycobacteria, Mcneil [/bib_ref]. Collectively, AG, PG, and mycolic acids with additional outer layer lipids result in an exceptionally robust and hydrophobic cell wall structure. Importantly, a number of anti-TB drugs, such as ethambutol [bib_ref] Inhibition of synthesis of arabinogalactan by ethambutol in Mycobacterium smegmatis, Takayama [/bib_ref] [bib_ref] The embAB genes of Mycobacterium avium encode an arabinosyl transferase involved in..., Belanger [/bib_ref] [bib_ref] The emb operon, a gene cluster of Mycobacterium tuberculosis involved in resistance..., Telenti [/bib_ref] and isoniazid [bib_ref] Inhibition by isoniazid of synthesis of mycolic acids in Mycobacterium tuberculosis, Winder [/bib_ref] [bib_ref] inhA, a gene encoding a target for isoniazid and ethionamide in Mycobacterium..., Banerjee [/bib_ref] , target the biosynthesis of the mAGP complex.
The biosynthesis of AG involves the formation of the linkage unit synthesized on a decaprenyl phosphate lipid carrier (DP). Firstly, WecA transfers GlcNAc-1-P from the substrate UDP-GlcNAc-1-P onto the DP carrier [bib_ref] Biosynthesis of the linkage region of the mycobacterial cell wall, Mikusova [/bib_ref] [bib_ref] Mycobacterium tuberculosis Rv1302 and Mycobacterium smegmatis MSMEG_4947 have WecA function and MSMEG_4947..., Jin [/bib_ref]. The rhamnosyltransferase WbbL then attaches the rhamnosyl residue to the DP-P-P-GlcNAc forming the full linker unit of AG, DP-P-P-GlcNAc-Rha [bib_ref] Biosynthesis of the linkage region of the mycobacterial cell wall, Mikusova [/bib_ref] [bib_ref] Inactivation of the mycobacterial rhamnosyltransferase, which is needed for the formation of..., Mills [/bib_ref]. The linker unit serves as an acceptor for the sequential addition of roughly 30 Galf residues. Bifunctional galactofuranosyltransferase (GalfT) GlfT1 recognizes the linkage unit and transfers two Galf residues to DP-P-P-GlcNAc-Rha yielding DP-P-P-GlcNAc-Rha-Galf 2 [bib_ref] Galactosyl transferases in mycobacterial cell wall synthesis, Belanova [/bib_ref]. GlfT2 then attaches further Galf residues acting both as a UDP-Galf: β-D-(1→5) GalfT and a UDP-Galf:β-D-(1→6) Galf T [bib_ref] Tetrameric structure of the GlfT2 galactofuranosyltransferase reveals a scaffold for the assembly..., Wheatley [/bib_ref] [bib_ref] Synthetic UDPgalactofuranose analogs reveal critical enzyme-substrate interactions in GlfT2-catalyzed mycobacterial galactan assembly, Poulin [/bib_ref]. Ar abinan biosynthesis employs decaprenylphosphoryl-D-arabinofuranose (DPA), the only known donor of Araf residues in AG biosynthesis. The assembly of DPA has been recently investigated in detail [bib_ref] Biochemical characterization of the Mycobacterium tuberculosis phosphoribosyl-1-pyrophosphate synthetase, Alderwick [/bib_ref] [bib_ref] Deletion of Cg-emb in corynebacterianeae leads to a novel truncated cell wall..., Alderwick [/bib_ref]. DPA biogenesis begins with UbiA transferring 5phosphoribosyl-1-pyrophosphate to a DP forming decaprenylphosphoryl-5-phosphoribose (DPPR) [bib_ref] Deletion of Cg-emb in corynebacterianeae leads to a novel truncated cell wall..., Alderwick [/bib_ref]. DPPR is then dephosphorylated to decaprenyl-5-phosphoribose (DPR) by a putative phospholipid phosphatase [bib_ref] The effect of MSMEG_6402 gene disruption on the cell wall structure of..., Jiang [/bib_ref]. DprE1 and DprE2 then catalyze the epimerization of DPR to DPA, consequently forming the essential sugar donor DPA [bib_ref] Partial redundancy in the synthesis of the D-arabinose incorporated in the cell..., Meniche [/bib_ref] [bib_ref] Decaprenylphosphoryl-beta-D-ribose 2'-epimerase, the target of benzothiazinones and dinitrobenzamides, is an essential enzyme..., C R E L L I N [/bib_ref] [bib_ref] Decaprenylphosphoryl arabinofuranose, the donor of the Darabinofuranosyl residues of mycobacterial arabinan, is..., Mikusova [/bib_ref]. A specialized arabinofuranosyltransferase (ArafT) AftA transfers the first Araf residue from the substrate molecule DPA onto the 8th, 10th and 12th Galf residues of the galactan chain [bib_ref] Identification of a novel arabinofuranosyltransferase (AftA) involved in cell wall arabinan biosynthesis..., Alderwick [/bib_ref]. Further α (1→5)-linked Araf residues are added by EmbA and EmbB in M. tuberculosis [bib_ref] The role of the embA and embB gene products in the biosynthesis..., Escuyer [/bib_ref] or Emb in C. glutamicum [bib_ref] Deletion of Cg-emb in corynebacterianeae leads to a novel truncated cell wall..., Alderwick [/bib_ref]. Branching α (1→3) ArafTs, AftC and AftD, are responsible for α (1→3)-linked Araf residues of the arabinan domain [bib_ref] A truncated lipoglycan from mycobacteria with altered immunological properties, Birch [/bib_ref] [bib_ref] Biosynthesis of mycobacterial arabinogalactan: identification of a novel alpha (1→3) arabinofuranosyltransferase, Birch [/bib_ref] [bib_ref] Reconstitution of functional mycobacterial arabinosyltransferase AftC proteoliposome and assessment of decaprenylphosphorylarabinose analogues..., Zhang [/bib_ref]. Finally, the terminal β (1→2) Araf residues are transferred from DPA onto the arabinan domain by AftB [bib_ref] Role of the major antigen of Mycobacterium tuberculosis in cell wall biogenesis, Belisle [/bib_ref] [bib_ref] Identification of a novel arabinofuranosyltransferase AftB involved in a terminal step of..., Seidel [/bib_ref].
The structure and biogenesis of AG has been fairly well described, however, certain aspects of its biosynthesis remain poorly understood. For instance, the characterization of multi-protein complexes has been extremely limited, perhaps due to a number of cell wall biosynthesis proteins being transmembrane or membrane bound. In this study, we attempted to investigate the associations between C. glutamicum proteins involved in the assembly of AG by using the bacterial adenylate cyclase two-hybrid (BACTH) system [bib_ref] A bacterial twohybrid system based on a reconstituted signal transduction pathway, Karimova [/bib_ref]. This system is based on the functional complementation between two fragments of the adenylate cyclase to restore a cAMP signaling cascade in Escherichia coli. Importantly, BACTH is able to detect physical interactions between both cytoplasmic as well as membrane proteins [bib_ref] Interaction network among Escherichia coli membrane proteins involved in cell division as..., Karimova [/bib_ref] [bib_ref] Characterization of YmgF, a 72-residue inner membrane protein that associates with the..., Karimova [/bib_ref] [bib_ref] PspB and PspC of Yersinia enterocolitica are dual function proteins: regulators and..., Maxson [/bib_ref] [bib_ref] Involvement of PlsX and the acyl-phosphate dependent sn-glycerol-3-phosphate acyltransferase PlsY in the..., Hara [/bib_ref] [bib_ref] Large-scale study of the interactions between proteins involved in type IV pilus..., Georgiadou [/bib_ref].
Our data supports interactions between various proteins involved in AG biosynthesis. Moreover, we demonstrate a number of novel interactions between these proteins. Altogether, these results suggest that proteins involved in AG assembly are associated to one another through multiple interactions.
# Materials and methods
## Bacterial strains and growth conditions
All cloning steps were performed in E. coli XL-1 Blue cells (Invitrogen). The E. coli cya strain BTH101 ((F − , cya-99, araD139, galE15, galK16, rpsL1 (Str r ), hsdR2, mcrA1, mcrB1) was used for the bacterial two-hybrid screen (Euromedex). E. coli strains were grown in Luria-Bertani (LB) medium at 30°C or 37°C as specified in the text. Plasmids were maintained with ampicillin (100 μg/ml) or kanamycin (50 μg/ml). LB agar reporter plates contained streptomycin (100 μg/ml), ampicillin (100 μg/ml), kanamycin ( 5 0 μ g / m l ) , 5 -b r o m o -4 -c h l o r o -3 -i n d o l y l -β -Dgalactopyranoside (X-gal; 40 μg/ml) and isopropyl β-D-1thiogalactopyranoside (IPTG; 0.5 mM). MacConkey plates (Difco TM ) contained streptomycin (100 μg/ml), ampicillin (100 μg/ml), kanamycin (50 μg/ml), IPTG (0.5 mM) and maltose (1 %). M63 minimal media plates [bib_ref] The bacterial two-hybrid system based on adenylate cyclase reconstitution in Escherichia coli, Battesti [/bib_ref] were supplemented with streptomycin (50 μg/ml), ampicillin (50 μg/ml), kanamycin (25 μg/ml), X-gal (40 μg/ml), IPTG (0.5 mM) and maltose (0.2 %).
## Plasmid construction
All recombinant DNA methods were performed using standard protocols. Briefly, the genes involved in C. glutamicum AG biosynthesis were amplified from genomic DNA of C. glutamicum ATCC 13032. The plasmids have been constructed by inserting gene sequences of interest in pKT25 (T25 fusion at N-terminus), pKNT25 (T25 fusion at C-terminus), pUT18 (T18 fusion at C-terminus) and pUT18c (T18 fusion at N-terminus) [bib_ref] A bacterial twohybrid system based on a reconstituted signal transduction pathway, Karimova [/bib_ref] , using oligonucleotides provided in [fig_ref] Table 1: Predicted topology and function of C [/fig_ref]. The bacterial BACTH system kit was obtained from Euromedex and contained empty vectors together with positive control plasmids pKT25-zip and pUT18c-zip.
Bacterial two-hybrid system Two plasmids expressing recombinant proteins bearing N-or C-terminal T25 and T18 fusions were co-transformed into E. coli BTH101 cells . Cells were spread on LB plates containing streptomycin (100 μg/ml), ampicillin (100 μg/ml), kanamycin (50 μg/ml) and incubated at 30°C for 48 h. Several colonies were picked and used to inoculate 3 ml of LB supplemented with appropriate antibiotics and 0.5 mM IPTG. Cultures were grown overnight at 30°C with shaking. Overnight cultures were washed three times in minimal media and spotted (2 μl) onto LB, MacConkey or M63 minimal media agar plates supplemented with appropriate antibiotics and nutrients. The β-galactosidase assay was performed as described elsewhere [bib_ref] The bacterial two-hybrid system based on adenylate cyclase reconstitution in Escherichia coli, Battesti [/bib_ref]. The values presented are the mean of 3 independent activity assays.
# Statistical analysis
The results are expressed as the means±S.D. and were analyzed using a Student's t-test to determine significant differences (p<0.01) between samples.
# Results and discussion
## Network analysis of ag biosynthetic proteins
We initially aimed to identify whether any of the proteins involved in cell wall assembly have been predicted or demonstrated to make a functional network. Focusing on the list of proteins associated with AG biosynthesis we used the STRING database of interactions [bib_ref] The STRING database in 2011: functional interaction networks of proteins, globally integrated..., Szklarczyk [/bib_ref] to reveal a putative protein association network, with GlfT2 chosen as the network node . The interaction patterns of proteins had a high confidence score (>0.7) and served as a basis for selection of C. glutamicum proteins that were further analyzed using the in vivo BACTH system. The generated network contained ABC family transporters (RfbD and RfbE), GT-A type GalfTs GlfT1, and proteins involved in rhamnose sugar donor formation, all centered on GlfT2. Transmembrane ArafT Emb showed a strong evidence for interaction with AftA and AftB, as well as the uncharacterized protein NCgl2596. The network also contained a putative phospholipid phosphatase NCgl2782 and proteins involved in DPA synthesis: DprE1, DprE2 and UbiA.
Bacterial two-hybrid analysis of AG proteins To characterize the physical interactions between components of the C. glutamicum cell wall biosynthetic machinery, the following full-length proteins WecA, WbbL, GlfT1, GlfT2, AftA, AftB, AftC, AftD, Emb, UbiA, DprE1, and DprE2, were tested systematically for pair-wise interactions using BACTH. The known or predicted function of proteins is shown in [fig_ref] Table 1: Predicted topology and function of C [/fig_ref]. Each protein was fused to the fragment of the catalytic domain of chimeric adenylate cyclase (T25 or T18) of Bordetella pertussis at either the C-or N-terminus . Interaction between two hybrid proteins leads to reconstitution of the fragments of adenylate cyclase resulting in restoration of cAMP production in a E. coli cya mutant [bib_ref] A bacterial twohybrid system based on a reconstituted signal transduction pathway, Karimova [/bib_ref]. The resulting cAMP forms a complex with the catabolite activator protein and binds to various promoters, thus regulating transcription of several genes, including the lactose and maltose operons. The activation of these operons can be detected on selective agar plates or using a β-galactosidase assay. Importantly, this bacterial two-hybrid system was shown to be suitable to detect interactions between cytoplasmic, as well as transmembrane or membrane associated proteins [bib_ref] Coordination of polymerization, chain termination, and export in assembly of the Escherichia..., Clarke [/bib_ref] [bib_ref] In vivo interaction between the polyprenol phosphate mannose synthase Ppm1 and the..., Baulard [/bib_ref].
Despite several attempts, we did not succeed in constructing pKT25 and pUT18 derivatives expressing UbiA and AftD proteins, respectively. This is probably due to the toxicity of hybrid proteins to bacterial cells when expressed at high levels, which is especially true of membrane proteins. Moreover, the UbiA-T18 N and UbiA-T18 C hybrid proteins, when co-expressed with several other hybrid proteins, appeared to reduce down bacterial cell growth suggesting that overproduction of UbiA is toxic to E. coli cells.
To examine putative interactions between the hybrid proteins, E. coli BTH101 cells were co-transformed with pairs of recombinant plasmids . In total, 577 pairs were screened for protein-protein interactions in vivo. All cotransformants, together with the positive and negative controls, containing either pKT25-zip/pUT18c-zip or empty pKT25/pUT18, were then spotted onto selective agar plates and the coloration of the colonies observed after 48 h of growth at 30°C. In the absence of association between T25 and T18 fragments colonies appear white, whereas they are blue or red when functional complementation occurs. Representative plates from the screening are shown in and -S13. Efficiency of the functional complementation between T25 and T18 domains was quantified by measuring β-galactosidase activity . Ultimately, 50 pairs of hybrid proteins resulted in a positive signal representing 24 putative homotypic and heterotypic proteinprotein interactions. The interaction results from the assays are summarized in .
Self-association of C. glutamicum cell wall biosynthesis proteins Among all the tested proteins, dimerization or multimerization of WecA, GlfT1, GlfT2, AftA, AftB, AftC, DprE1, and DprE2 have been demonstrated employing BACTH. Co-expression of transmembrane WecA-T25 C and WecA-T18 C hybrid proteins restored a cya + phenotype and synthesis of cAMP in the E. coli cells, resulting in blue and red colonies on LB/M63-Xgal and MacConkey media, respectively . The β-galactosidase assay revealed a significant increase in βgalactosidase activity (487±47 Miller units) when compared Network of C. glutamicum proteins found to be important for cell wall assembly round the GlfT2 protein as determined by STRING analysis. Lines connecting the nodes indicate various interaction data supporting the network, colored by evidence type . Importantly, the transmembrane fusions have to be correctly inserted into the plasma membrane with the T18 and T25 domains facing the cytoplasm in order an interaction to be detected. Therefore, suggesting that the C-terminus of GlcNAc-1-phosphate transferase WecA is cytoplasmic. This is in agreement with the predicted topology of WecA [bib_ref] Conserved amino acid residues found in a predicted cytosolic domain of the..., Amer [/bib_ref]. Physical self-dimerization or multimerization in vivo was also demonstrated for GlfT1 and GlfT2. Consistently, GlfT1-T25 N and GlfT1-T18 C , GlfT1-T25 N and GlfT1-T18 N , GlfT1-T25 C and GlfT1-T18 N , GlfT2-T25 N and GlfT2-T18 C , and GlfT2-T25 C and GlfT2-T18 N hybrids restored lac + and mal + phenotypes and resulted in significant β-galactosidase activity
[formula] WecA ✓ WbbL - - GlfT1 - - ✓ GlfT2 - - - ✓ AftA - - - - ✓ AftB ✓ - ✓ - ✓ ✓ AftC ✓ - - - ✓ ✓ ✓ AftD - - - - - - - - UbiA ✓ ✓ - - ✓ ✓ ✓ - - D p r E 1 - - - - - - - - - ✓ DprE2 - - ✓ - ✓ ✓ ✓ - - ✓ ✓ Emb ✓ - - - ✓ ✓ - - - - - - [/formula]
The positive interaction is indicated as (✓), whereas the lack of interaction is marked as (−) ranging from 266±69 to 679±118 Miller units .
Recently, the structure of the polymerizing GlfT2 orthologue in M. tuberculosis has been solved revealing its assembly as a homotetramer [bib_ref] Tetrameric structure of the GlfT2 galactofuranosyltransferase reveals a scaffold for the assembly..., Wheatley [/bib_ref] , thus supporting the results obtained in this BACTH study. Transmembrane AftA, AftB and AftC, proteins also tested positive for self-association. Co-expression of AftA-T25 N and AftA-T18 N , AftB-T25 N and AftB-T18 N , AftC-T25 C and AftC-T18 C combinations yielded β-galactosidase activity of 714±92, 1185±265, and 398±23 Miller units, respectively . The C-terminal region of AftA and AftB is predicted to be directed towards the periplasm [bib_ref] Identification of a novel arabinofuranosyltransferase AftB involved in a terminal step of..., Seidel [/bib_ref] [bib_ref] Identification of a novel arabinofuranosyltransferase (AftA) involved in cell wall arabinan biosynthesis..., Alderwick [/bib_ref] , therefore the lack of interaction between fusion pairs carrying C-terminal T25 or T18 fragment was expected. In addition, BACTH experiments propose that the N-termini of AftA and AftB are cytoplasmic. In contrast to AftA and AftB, AftC is characterized by the absence of a periplasmic C-terminal extension [bib_ref] Biosynthesis of mycobacterial arabinogalactan: identification of a novel alpha (1→3) arabinofuranosyltransferase, Birch [/bib_ref]. Hence it is unsurprising that multimerization of AftC is observed with the fusion proteins tagged at the C -t e r m i n u s . I n t e r e s t i n g l y, n o e v i d e n c e f o r homodimerization could be obtained for Emb and AftD. Finally, DprE1 and DprE2, both involved in DPA synthesis, appeared positive for self-interaction. DprE1-T25 N and DprE1-T18 C , DprE1-T25 C and DprE1-T18 N fusions, as well as all four pairs of hybrid proteins co-expressing DprE2 led to a strong lacZ induction (ranged between 291±33 and 1156±54 Miller units), significantly exceeding the negative control .
In vivo interaction network among AG proteins Next, we examined the interactions between different proteins involved in AG biosynthesis. Our results indicate that in addition to homodimerization, WecA is also able to interact with multiple partners of AG biosynthesis . WecA-T18 C hybrid, when co-expressed with AftB-T25 N , AftC-T25 C , Emb-T25 N , and UbiA-T25 C yielded significant β-galactosidase activities 1082 ± 268, 1047 ± 186, 1346 ± 217, and 1018 ± 137 Miller units, respectively . BACTH also revealed an interaction with the rhamnosyltransferase WbbL, when UbiA hybrids were used as the bait. Cotransformation of WbbL-T18 C or WbbL-T18 N together with UbiA-T25 C led to a restoration of cAMP cascade with β-galactosidase activities of 1094 ±93 and 1195± 78 Miller units, respectively . Our studies have demonstrated the physical interaction between GlfT1 and AftB , as well as the DprE2 involved in DPA formation . Recent studies reported the physical interaction between GlfT1 and Rv3789, a small multidrug resistance-like transporter [bib_ref] A small multidrug resistance-like transporter involved in the arabinosylation of arabinogalactan and..., Larrouy-Maumus [/bib_ref]. Rv3789 was proposed to target and stabilize the membrane associated GlfT1 [bib_ref] A small multidrug resistance-like transporter involved in the arabinosylation of arabinogalactan and..., Larrouy-Maumus [/bib_ref]. Further experiments demonstrated evidence for a physical interaction between UbiA and AftA-T25 N (387 ± 22 Miller units) , AftB-T25 N (1015 ± 185 Miller units) and AftC-T25 C (755 ± 118 Miller untis) , responsible for the biosynthesis of An interaction network of C. glutamicum proteins involved in AG biosynthesis generated using yEd graph editor software. The circular arrows indicate selfassociation the arabinan domain of AG. Most of these ArafTs could also establish multiple interactions with each other. AftA, which primes the galactan chain of AG, associated with Emb, AftC and AftB . In addition, AftB also interacted with Emb and AftC hybrid proteins . Finally, C. glutamicum DprE1 was also found to strongly interact with DprE2. DprE1-DprE2 association has been identified with seven different plasmid combinations resulting in a significant β-galactosidase activity ranging between 637± 52 and 1027± 86 Miller units . Previous studies reported that orthologues of DprE1 and DprE2 in M. tuberculosis were able to catalyze the epimerization reaction in vitro, however, neither protein alone was sufficient to support this activity [bib_ref] A small multidrug resistance-like transporter involved in the arabinosylation of arabinogalactan and..., Larrouy-Maumus [/bib_ref]. Thus, strongly suggesting that DprE1 and DprE2 work in concert to catalyze the conversion of DPR to DPA. However, when the same M. tuberculosis orthologues were experimentally tested for interaction using BACTH, co-transformants yielded negative results [bib_ref] A small multidrug resistance-like transporter involved in the arabinosylation of arabinogalactan and..., Larrouy-Maumus [/bib_ref].
# Conclusions
The majority of bacterial cell wall surface polysaccharides are built on a carrier lipid in the cytosolic side of the plasma membrane. Although it is not fully clear how and when these polymers are translocated to the periplasm, one could speculate that anchoring these macromolecules to the membrane positions them closely to the transporters and glycosyltransferases, therefore, promoting productive export across the plasma membrane. Formation of multi-protein complexes, that contain glycosyltransferases, enzymes forming its sugar nucleotides and transporters, is expected to be beneficial for the bacterial cell, since the tight arrangement of the biosynthetic reactions would retain productivity and accuracy of the polymerization process. We have demonstrated that the proteins responsible for the formation of the AG linker unit, W e c A a n d W b b L , f o r m a c o m p l e x w i t h decaprenylphosphoryl-5-phosphoribose synthase UbiA at the cytoplasmic membrane . WecA and UbiA directly employ DP for the linker unit and DPPR formation, respectively [bib_ref] Biosynthesis of the linkage region of the mycobacterial cell wall, Mikusova [/bib_ref] [bib_ref] Decaprenylphosphoryl arabinofuranose, the donor of the Darabinofuranosyl residues of mycobacterial arabinan, is..., Mikusova [/bib_ref] [bib_ref] Identification of amino acids and domains required for catalytic activity of DPPR..., Huang [/bib_ref]. Proximal interactions between WecA, WbbL and UbiA could perhaps facilitate synchronized utilization of DP for coordinated AG biosynthesis. In addition, UbiA show evidence for physical interaction between AftA, AftB and AftC proteins, which utilize DPA as a substrate . It is possible that this multi-protein complex formation assists a mechanism similar to substrate channeling, where intermediary metabolic products of one enzyme are passed directly to another enzyme. Other DPA forming proteins, DprE1 and DprE2, showed evidence for a physical interaction. Interestingly, while both DprE1 and DprE2 are required for the epimerization reaction, there is evidence that C. glutamicum NCgl1429 may play a similar function to DprE2 [bib_ref] Partial redundancy in the synthesis of the D-arabinose incorporated in the cell..., Meniche [/bib_ref]. Investigation into potential DprE1-NCgl1429 complexes could provide insight into this gene redundancy. Notably, GT-A glycosyltransferases GlfT1 and GlfT2 showed evidence for homodimerization using BACTH. GlfT1 transfers the first two Galf residues to the linker unit, whereas GlfT2 is responsible for addition of approximately 30 Galf residues in a linear chain. The recent crystal structure of M. tuberculosis GlfT2 in its apo-form and in complex with UDP, established its homotetrameric architecture [bib_ref] Tetrameric structure of the GlfT2 galactofuranosyltransferase reveals a scaffold for the assembly..., Wheatley [/bib_ref]. Finally, AftA, AftB, AftC and Emb proteins involved in the assembly of arabinan domain in AG, indeed form a multi-protein complex at the inner membrane . One could speculate that such a sophisticated complex would maintain the efficiency and fidelity of AG polymerization.
BACTH is a powerful technique for the investigation of protein-protein associations, however, several important notes should be highlighted regarding the significance of the interaction data obtained from BACTH. Firstly, the lack of lacZ induction might be a result of plasmid instability, insoluble or dissipating fusions, and not the lack of direct physical interaction. Therefore, the hybrid proteins that test negative for interactions may still interact in vivo. Moreover, since the output of the interaction -cAMPrequires to be generated in the cytoplasm, these negative results may also result from the incorrect topological orientation of functional T25 and T18 domains into plasma membrane. In addition, using BACTH, the fusion proteins are overexpressed when compared to the expression levels of native cells. Under these conditions, BACTH could have revealed a number of weak interactions between AG biosynthetic proteins. Although such associations would not take place at low protein concentrations, they can still occur when AG is being synthesized, where the local concentrations of proteins should be significantly higher. Finally, it is possible that some of the indentified interactions are a consequence of non-specific interactions initiated by endogenous E. coli host proteins that act as a tethering agent. These indirect associations, caused by a third protein, cannot be simply rejected.
In conclusion, our findings here suggest that enzymes involved in C. glutamicum cell wall assembly and precursor formation form complicated multi-protein complexes. We have identified 24 interactions in vivo between 12 proteins responsible for AG biosynthesis using BACTH. The challenge for the future will be to discover precisely how each of these multi-protein complexes form and function to synthesize and translocate AG.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
[table] Table 1: Predicted topology and function of C. glutamicum proteins described in this study [/table]
|
Tolerability and efficacy of glycemic control with saxagliptin in older patients (aged ≥ 65 years) with inadequately controlled type 2 diabetes mellitus
Purpose:To assess safety and efficacy of saxagliptin in older patients with type 2 diabetes mellitus (T2DM). Patients and methods: This was a post hoc analysis of pooled data from older patients ($65 years of age) from five 24-week phase III trials: three studies of saxagliptin versus placebo as an add-on therapy to metformin, glyburide, or a thiazolidinedione; and two studies of saxagliptin versus placebo as monotherapy in drug-naïve patients. Separate analyses were conducted on one study of initial combination therapy with saxagliptin plus metformin versus metformin monotherapy in drug-naïve patients. The safety analysis population for the fivestudy pool included 428 patients $ 65 years of age with baseline glycated hemoglobin (HbA 1c ) 7.0% to 10.5% who received saxagliptin 2.5 or 5 mg or placebo, and for the study of initial combination therapy included 69 patients $ 65 years of age with baseline HbA 1c 8.0% to 12.0% who received saxagliptin 5 mg in combination with metformin or metformin monotherapy. The primary efficacy endpoint was change from baseline HbA 1c . Results: In the five-study pool, the differences in the adjusted mean change from baseline HbA 1c among older patients receiving saxagliptin versus placebo were −0.60% (95% confidence interval [CI], −0.99% to −0.21%) for saxagliptin 2.5 mg and −0.55% (−0.97% to −0.14%) for saxagliptin 5 mg; in the initial combination study, the difference was −1.22% (−2.27% to −0.17%) among older patients receiving saxagliptin 5 mg plus metformin versus metformin monotherapy. The results were generally similar in older and younger patients. Saxagliptin was well tolerated; the incidence and types of adverse events were similar for saxagliptin and comparators. Hypoglycemia was reported in 3.0% to 9.4% of patients receiving saxagliptin (0%-8.0% for comparators) and was confirmed (finger stick glucose # 50 mg/dL, with associated symptoms) in 0% to 0.7% (0%-0.7% for comparators); hypoglycemic episodes did not vary by age category and did not require medical intervention. Conclusion: Saxagliptin was effective and well tolerated, with a low risk of hypoglycemia, when used as monotherapy, add-on therapy, or initial combination therapy with metformin in older patients with T2DM.
# Introduction
Elderly adults are becoming an increasingly large segment of the US population,and those aged $ 65 years are almost seven times more likely to have diabetes compared with adults 20 to 44 years of age (27% vs 4%).Elderly adults with type 2 diabetes mellitus (T2DM) are more likely to have comorbid conditions, such as hypertension and coronary heart disease, than those without T2DM 3 and are also likely to experience complications due to polypharmacy, mobility impairment and falls, cognitive impairment, chronic pain, and depression. [bib_ref] Impact of diabetes on physical function in older people, Sinclair [/bib_ref] [bib_ref] Health, Aging, and Body Composition Study. Diabetes-related complications, glycemic control, and falls..., Schwartz [/bib_ref] [bib_ref] Diabetes mellitus, glycemic control, and incident depressive symptoms among 70-to 79-year-old persons:..., Maraldi [/bib_ref] [bib_ref] Cognitive dysfunction is associated with poor diabetes control in older adults, Munshi [/bib_ref] These factors present challenges for glycemic control and successful management of T2DM in elderly patients.
There remains a need for treatment options that are well tolerated and efficacious in the geriatric population, especially because of the more challenging needs and higher risk of comorbid complications in these patients.Currently available and commonly employed therapies include the biguanide metformin, insulin, insulin secretagogues (sulfonylureas and meglitinides), thiazolidinediones, and incretin-based therapies (long-acting glucagon-like protein-1 [GLP-1] agonists and dipeptidyl peptidase-4 [DPP-4] inhibitors).
DPP-4 inhibitors have become widely accepted in clinical practice possibly because of their low risk of hypoglycemia, favorable adverse event (AE) profile, and once-daily oral administration. [bib_ref] Dipeptidyl peptidase-4 inhibitors for the treatment of type 2 diabetes mellitus, Neumiller [/bib_ref] The general glycemic goal recommended by the American Diabetes Association and the European Association for the Study of Diabetes for most patients is HbA 1c , 7%, and the goal recommended by the American Association of Clinical Endocrinologists and American College of Endocrinology is HbA 1c # 6.5%, [bib_ref] American Association of Clinical Endocrinologists medical guidelines for clinical practice for developing..., Handelsman [/bib_ref] but the goal for each patient should be individualized, based on such factors as life expectancy, comorbidities, and a history of severe hypoglycemia. [bib_ref] American Association of Clinical Endocrinologists medical guidelines for clinical practice for developing..., Handelsman [/bib_ref] Accordingly, the American Diabetes Association and European Association for the Study of Diabetes also recommend treatment options that avoid hypoglycemia, particularly in the elderly,given that conditions associated with age, such as renal impairment, polypharmacy, and cognitive dysfunction, may contribute to an increased risk of hypoglycemia. 4,6-8 DPP-4 inhibitors are considered promising options in the elderly population. [bib_ref] Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for..., Nathan [/bib_ref] In recent analyses of patients $ 65 years of age, DPP-4 inhibitors achieved glycemic control without increasing the risk of hypoglycemia or other AEs. [bib_ref] Alogliptin use in elderly people: a pooled analysis from phase 2 and..., Pratley [/bib_ref] [bib_ref] Management of type 2 diabetes in treatment-naive elderly patients: benefits and risks..., Pratley [/bib_ref] [bib_ref] Role of vildagliptin in managing type 2 diabetes mellitus in the elderly, Halimi [/bib_ref] [bib_ref] Clinical experience with vildagliptin in the management of type 2 diabetes in..., Schweizer [/bib_ref] [bib_ref] Efficacy and tolerability of sitagliptin monotherapy in elderly patients with type 2..., Barzilai [/bib_ref] Saxagliptin is a potent, selective DPP-4 inhibitor approved in 2009 by the US Food and Drug Administration (FDA) as an adjunct to diet and exercise, to improve glycemic control in adults with T2DM.Saxagliptin significantly improves glycemic control, as demonstrated by decreases in HbA 1c , fasting plasma glucose (FPG), and postprandial glucose (PPG). [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] Saxagliptin: the evidence for its place in the treatment of type 2..., Kulasa [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] In placebo-controlled phase III trials, saxagliptin has been assessed as monotherapy in drug-naïve patients, [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] as an add-on therapy in patients who did not achieve adequate glycemic control on monotherapy with other oral antidiabetic drugs, [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] and as initial combination therapy with metformin in drug-naïve patients.Each saxagliptin regimen in these trials was well tolerated; there was minimal risk of hypoglycemia with saxagliptin versus comparators except for a numerically higher incidence in the study of saxagliptin used as add-on to a sulfonylurea.Data on the safety and efficacy of saxagliptin in older patients are limited. One previous retrospective subgroup analysis of pooled data from five core placebo-controlled, phase III saxagliptin studies in patients aged $ 65 years receiving saxagliptin 5 mg has been reported. [bib_ref] Efficacy and safety of saxagliptin in older patients with type 2 diabetes..., Doucet [/bib_ref] Saxagliptin is approved to be used at doses of 2.5 and 5 mg in the United States and at 5 mg in many other countries. In addition, saxagliptin is approved for use as an add-on agent or for first-line therapy.To provide relevant clinical-use data, the current post hoc analysis assessed efficacy and safety outcomes with saxagliptin 2.5 mg and 5 mg, in the subgroup of patients aged $ 65 years, from the five placebo-controlled core saxagliptin studies as well as an additional core study evaluating saxagliptin used as initial combination therapy with metformin versus metformin monotherapy in drug-naïve patients. The five placebo-controlled trials, which shared similar methodology, were pooled to achieve a large sample of older patients for statistical analysis. The remaining study was assessed separately due to differences in methodology (including the lack of a placebo control); it will be referred to subsequently as the initial combination study.
# Material and methods
## Study designs
The core saxagliptin phase III program consisted of a set of multicenter, randomized, double-blind, 24-week trials [fig_ref] Table 1: Pivotal phase III Studies [/fig_ref]. Saxagliptin, at daily doses of 2.5 and 5 mg, was assessed in two placebo-controlled trials of monotherapy in drug-naïve patients (ClinicalTrials.gov identifiers NCT00121641, NCT00316082) [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] and in three placebocontrolled trials of add-on therapy in patients who did not achieve adequate glycemic control on monotherapy with metformin (immediate-release [IR] formulation) 500 to 2500 mg (NCT00121667), a thiazolidinedione (pioglitazone 30 or 45 mg or rosiglitazone 4 or 8 mg) (NCT00295633), or a sulfonylurea (glyburide at a submaximal dose of 7.5 mg, uptitrated to 15 mg per protocol in the placebo arm as needed) (NCT00313313). [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] In the add-on studies, patients remained on their initial metformin, thiazolidinedione, or sulfonylurea therapy as saxagliptin or placebo was initiated. In the initial combination study, saxagliptin 5 mg was assessed as initial combination therapy with metformin IR 500-2000 mg submit your manuscript | www.dovepress.com versus metformin monotherapy in drug-naïve patients (NCT00327015).One of the monotherapy studies 24 also included a treatment arm in which saxagliptin was titrated from 2.5 to 5 mg; the other monotherapy study, 18 the metformin add-on study, 20 and the initial combination studyalso included treatment arms in which saxagliptin was given at 10 mg.
## Dovepress
## Dovepress
Individual study methodology has been previously reported. [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] Briefly, the study populations included men and women aged 18 to 77 years with T2DM and inadequate glycemic control, a fasting C-peptide concentration $ 1 ng/mL, and a body mass index (BMI) #40 kg/m 2 (#45 kg/m 2 in the study of saxagliptin added on to thiazolidinedione [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref]. Patients were treated on an outpatient basis. Exclusion criteria were similar across the studies. [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] Rescue therapy was initiated if patients did not meet prespecified and progressively stringent glycemic goals while on study medication (FPG . 240 mg/dL at week 4 or 6; .220 mg/dL at week 8; .200 mg/dL at week . The study protocols and patient informed consent were approved by the respective institutional review board. The studies were conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki.
## Efficacy and safety parameters
Efficacy results were assessed across all six studies. [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] The primary efficacy endpoint was change in HbA 1c from baseline to week 24. The secondary efficacy endpoints included changes in FPG from baseline to week 24, in PPG at 120 minutes (PPG-120) following a 75 g oral glucose tolerance test (OGTT), and in PPG area under the curve from 0 to 180 minutes (PPG-AUC 0−180 ) in the OGTT; the proportion of patients achieving a therapeutic glycemic target (HbA 1c , 7.0%) was also reported. The safety and tolerability assessments included overall AEs, serious AEs (SAEs), and discontinuations due to AEs. Hypoglycemic AEs, including all reported cases of hypoglycemia and confirmed hypoglycemia (defined as finger stick glucose value # 50 mg/dL associated with symptoms), were recorded.
# Statistical methods
For the present post hoc analysis, the subgroup analysis of these six core studies, efficacy and safety outcomes were analyzed in older patients with T2DM ($65 years of age); the data for younger patients (,65 years of age) are presented for reference. The age-categorized data from the placebocontrolled trials of saxagliptin as monotherapy and as add-on therapy are pooled; data from the initial combination study are presented separately.
The overall efficacy and safety results have previously been reported for all six studies. [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] The present post hoc analysis of outcomes in older patients included only data from patients receiving saxagliptin at the currently approved doses of 2.5 or 5 mg. Data from patients randomized to treatment with saxagliptin 2.5 mg with possible titration to 5 mg 24 (n = 12) were excluded from the efficacy but not the safety analyses because the doses were not constant across the evaluation period; data from patients receiving saxagliptin 10 mg [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] were excluded from the efficacy and safety analyses because 10 mg is not an FDA-approved dose. A baseline plus one or more postbaseline measurement(s) were required for inclusion in the efficacy analysis. Safety analyses included patients who received one or more dose(s) of study medication.
In an intent-to-treat analysis of the five pooled studies, the evaluation for change from baseline to week 24 in HbA 1c , FPG, PPG-AUC 0−180 , and PPG-120 used analysis of covariance (ANCOVA), with treatment, subgroup, study, and treatmentby-subgroup as factors and baseline as a covariate. A nominal test was performed for the treatment-by-subgroup interactions including treatment-by-age interaction. Two-sided 95% confidence intervals (CIs) are presented for the within-group adjusted mean change from baseline to week 24 and for the difference between the adjusted mean change from baseline to week 24 with the saxagliptin regimen minus that of the comparator regimen. For the proportion of patients who achieved HbA 1c , 7.0%, 95% CIs are presented for the difference versus comparators, using an exact CI for the initial combination study and the Mantel-Haenszel CI for the risk difference for the five pooled studies. Missing data were imputed using last observation carried forward (LOCF) methodology. In patients who required rescue medication, the last value before rescue medication was used in the analysis.
# Results
## Study population
Patient disposition for each of the individual studies has been described previously. [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] There were a total of 468 older patients (16.5% of the overall study population) in the five pooled placebo-controlled studies of saxagliptin as monotherapy and add-on therapy, and 166 older patients (12.7% of the overall study population) in the study of saxagliptin as initial combination therapy with metformin. In the pooled studies and initial combination study, the majority of older patients were aged 65 to 75 years with only 23 and 16 patients, respectively, aged $ 75 years. Because of treatment arm exclusions, as detailed in the Methods, the safety analysis included 428 older patients from the five pooled studies and 69 older patients from the initial combination study; 416 older patients in the five pooled studies and 69 from the initial combination study were included in the efficacy analyses. In all, 341/428 (79.7%) older patients in the pooled studies and 58/69 (84.1%) in the initial combination study completed 24 weeks of treatment.
Baseline demographic and clinical characteristics are shown in Tables 2A and B. In the five-study pool and the initial combination study, the mean age of the older patients was 69 to 70 years. The mean duration of T2DM was 7.2 to 7.3 years in the pooled studies and 0.8 to 1.4 years in the initial combination study. Patients in the initial combination study had higher glycemic values than those in the pooled studies (HbA 1c : 8.9%-9.5% vs 7.9%-8.2%; FPG: 192.5-207.2 mg/dL vs 161.2-171.9 mg/dL). In the pooled population, mean creatinine clearance rates were 80 to 84 mL/min in older patients and 118 to 119 mL/min in younger patients. In the initial combination study, the mean creatinine clearance rates were 82 to 83 mL/min in older patients and 119 to 120 mL/min in younger patients. Concomitant medications were widely used at baseline. In the pooled population, 58% to 69% of older patients and 40% to 41% of younger patients were studied on a background of $5 medications. In the initial combination study, the majority of patients in both age groups were studied on a background of $2 concomitant medications. Other demographic and baseline characteristics were similar across treatment groups.
## Efficacy change in glycated hemoglobin
In the five-study pooled analysis, the mean baseline HbA 1c ranged from 7.91% to 8.24% and was similar between the saxagliptin 2.5 mg and 5 mg groups and between age groups . The week 24 adjusted mean changes from baseline HbA 1c were −0.78% and −0.73% for older patients taking saxagliptin 2.5 and 5 mg and were slightly smaller in the younger subgroup. The differences in the adjusted mean changes from baseline HbA 1c with saxagliptin minus that of placebo were −0.60% (95% CI, −0.99% to −0.21%) and −0.55% (95% CI, −0.97% to −0.14%) for older patients taking saxagliptin 2.5 and 5 mg. There was no evidence for a treatment-by-age interaction for HbA 1c in the five-study pool (P = 0.88).
In the initial combination study, the adjusted mean changes from baseline HbA 1c at week 24 for saxagliptin 5 mg plus metformin were similar in the older (−2.48%) and younger (−2.55%) groups. The metforminsubtracted adjusted mean change from baseline with saxagliptin 5 mg plus metformin was numerically larger in older versus younger participants (treatment-by-age interaction, P = 0.14).
## Other glycemic outcome measures
For both older and younger subjects, reductions from baseline to week 24 in PPG-AUC 0−180 and PPG-120 and the percentage of patients achieving HbA 1c , 7% at week 24 were greater with saxagliptin than with placebo, in the pooled studies. However, the adjusted mean change from baseline FPG for older patients receiving saxagliptin 2.5 mg (−7.6 mg/dL) was associated with a 95% CI that spanned zero (−17.4, 2.2), although it was directionally consistent with that in younger patients (−13.1 mg/dL) . In the initial combination study, the metformin-subtracted reductions from baseline to week 24 in FPG, PPG-AUC 0−180 , and PPG-120 were numerically greater with saxagliptin plus metformin in the older population than those observed in the younger patients. The difference in the proportions of patients achieving HbA 1c , 7% with saxagliptin plus metformin versus metformin alone was similar in magnitude in older (18.7%) and younger (19.2%) patients, but the older subgroup had a 95% CI which spanned zero (−5.7%, 40.6%).
## Safety and tolerability
Overall, saxagliptin was generally well tolerated as monotherapy or combination therapy in older patients. In both the five-study pool and the initial combination study, the incidence and types of AEs were similar for saxagliptin and the placebo and metformin comparators. The overall incidence of AEs and the specific AEs occurring in $5% of patients in any treatment group in the pooled studies and in the initial combination study are shown in [fig_ref] Table 4B: Adverse events in older and younger patients with type 2 diabetes mellitus... [/fig_ref] , respectively.
In the pooled analysis, no SAE in the older subgroup was considered by the investigator to be related to saxagliptin treatment. In the initial combination study, one saxagliptin patient in the older subgroup had a treatment-related SAE (presumed overdose, without hypoglycemia or other symptoms), which led to discontinuation from the study on grounds of poor compliance. There were two deaths in the older subgroup (one patient receiving saxagliptin 2.5 mg in the pooled analysis [car accident owing to submit your manuscript | www.dovepress.com weather/road conditions]; one patient receiving metformin monotherapy in the initial combination study [apparent congestive heart failure]); neither event was considered treatment-related.
## Dovepress
## Dovepress
In the pooled studies, the incidence of confirmed hypoglycemia was 0.7% and 0% with saxagliptin 2.5 and 5 mg, and 0.7% with placebo in older patients. The incidence of all reported hypoglycemia in the older subgroup was 9.4% and 6.3% with saxagliptin 2.5 and 5 mg, respectively, and 8.0% with placebo. In a separate pooled analysis excluding the study of add-on to the sulfonylurea glyburide, which carries a known risk of hypoglycemia, the incidence of reported hypoglycemia in the older subgroup dropped to 7.5% and 4.0% for saxagliptin 2.5 mg and 5 mg, respectively, and to 5.9% for placebo.
In the initial combination study, reported hypoglycemia was infrequent among older patients (3.0% with saxagliptin 5 mg plus metformin; 0% with metformin monotherapy), and there were no cases of confirmed hypoglycemia. No hypoglycemic event required medical assistance. The incidence of hypoglycemic events was similar in older and younger patients.
# Discussion
Decisions regarding the use of antihyperglycemic therapies in the geriatric population should be made with consideration of safety profiles and ease of use. Specifically, sulfonylureas carry a greater risk of hypoglycemia, 27 whereas thiazolidinediones may increase the risk of peripheral edema, congestive heart failure, [bib_ref] Thiazolidinediones and the risk of incident congestive heart failure among patients with..., Filion [/bib_ref] and bladder cancer, [bib_ref] Risk of bladder cancer among diabetic patients treated with pioglitazone: interim report..., Lewis [/bib_ref] and GLP-1 agonists require subcutaneous injection and are associated with nausea. [bib_ref] Benefit-risk assessment of exenatide in the therapy of type 2 diabetes mellitus, Gallwitz [/bib_ref] The presence of comorbidities and polypharmacy are additional important considerations and potentially limiting factors in the selection of antihyperglycemic therapy for older patients with diabetes.DPP-4 inhibitors enhance natural glucoregulatory physiology by preventing the rapid degradation of endogenous incretin hormones, GLP-1 and glucose-dependent insulinotropic polypeptide (GIP), that is normally mediated submit your manuscript | www.dovepress.com
## Dovepress
## Dovepress
## Table 3a
Glycemic efficacy at 24 weeks in older and younger patients with type 2 diabetes mellitus from five pooled studies of saxagliptin vs placebo as monotherapy and as add-on therapy [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] by the enzyme DPP-4. These incretin hormones, released from the gut upon food intake, 31,32 lower blood glucose by stimulating the pancreas to synthesize and secrete insulin while decreasing glucagon release. Because these actions are glucose-dependent and their effects diminish as postprandial glucose drops back toward normal levels, 31,33 the risk of hypoglycemia with DPP-4 inhibitors is minimal. Together with the fact that they are orally administered, effective, and well tolerated, this makes DPP-4 inhibitors a potentially useful therapeutic option in elderly patients with T2DM. [bib_ref] Dipeptidyl peptidase-4 inhibitors for the treatment of type 2 diabetes mellitus, Neumiller [/bib_ref] The joint American Association of Clinical Endocrinologists and American College of Endocrinology treatment algorithm recommends DPP-4 inhibitors as monotherapy in patients with HbA 1c levels of 6.5% to 7.5% or as a preferred combination therapy with metformin for patients with HbA 1c levels $ 7.6%. [bib_ref] Statement by an American Association of Clinical Endocrinologists/American College of Endocrinology consensus..., Rodbard [/bib_ref] In the present post hoc analysis of the outcomes in older patients with T2DM, the DPP-4 inhibitor saxagliptin was effective in improving glycemic control, as shown by reductions in HbA 1c , FPG, PPG-AUC 0−180 , and PPG-120 and by an increased proportion of patients achieving the glycemic goal of HbA 1c , 7.0% with saxagliptin versus the comparators (placebo in the five-study pooled analysis or metformin monotherapy in the initial combination study). In the five-study pooled analysis and the initial combination study, patients older and younger than 65 years showed greater improvements in all glycemic parameters with saxagliptin than with comparators. The results were generally similar in older and younger patients except in the initial combination study, wherein the difference in adjusted mean change in HbA 1c with saxagliptin plus metformin versus metformin monotherapy was greater in older patients (−1.22%) submit your manuscript | www.dovepress.com than in younger patients (−0.53%); there were no significant treatment-by-age interactions. Saxagliptin treatment was well tolerated and was associated with a low incidence of hypoglycemic events. These results are important given that the Action to Control Cardiovascular Risk in Diabetes (ACCORD) [bib_ref] ACCORD Investigators. The effects of baseline characteristics, glycaemia treatment approach, and glycated..., Miller [/bib_ref] and Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) [bib_ref] The efficacy of lowering glycated haemoglobin with a gliclazide modified release-based intensive..., Zoungas [/bib_ref] trials showed an association between hypoglycemia (glucose , 50 mg/dL) and age, and ACCORD 37 linked hypoglycemia to an increased risk of death, prompting the American Diabetes Association and the European Association for the Study of Diabetes to note that it is especially important to avoid treatment-related hypoglycemia in elderly patients.Hypoglycemia can impair judgment, behavior, and performance of physical tasks; 38 such adverse effects can interfere with treatment adherence, especially in elderly patients.
## Dovepress
## Dovepress
Concern about iatrogenic hypoglycemia can be a barrier to good glycemic control. Because the antihyperglycemic effects of the DPP-4 inhibitors are glucose dependent, the risk of hypoglycemia is low. [bib_ref] Enhancing incretin action for the treatment of type 2 diabetes, Drucker [/bib_ref] [bib_ref] The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors in..., Drucker [/bib_ref] [bib_ref] Changes in prandial glucagon levels after a 2-year treatment with vildagliptin or..., Ahrén [/bib_ref] In contrast, sulfonylureas and meglitinides stimulate insulin secretion independent of glucose concentrations, and sulfonylureas in particular are associated with a higher risk of hypoglycemia, [bib_ref] Effect of noninsulin antidiabetic drugs added to metformin therapy on glycemic control,..., Phung [/bib_ref] including use in combination with a DPP-4 inhibitor.Thus, sulfonylureas should be used with particular caution in elderly patients.
The results from the present post hoc analysis are generally consistent with the results of another pooled analysis of data on saxagliptin in older patients that was limited to patients treated with saxagliptin 5 mg in the five submit your manuscript | www.dovepress.com
## Dovepress
## Dovepress
## Table 4a
Adverse events in older and younger patients with type 2 diabetes mellitus from five pooled studies of saxagliptin vs placebo as monotherapy and as add-on therapy [bib_ref] Saxagliptin added to a thiazolidinedione improves glycemic control in patients with type..., Hollander [/bib_ref] [bib_ref] Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin when added..., Defronzo [/bib_ref] [bib_ref] clinical-interventions-in-aging-journal Clinical Interventions in Aging is an international, peer-reviewed journal focusing on..., Frederich [/bib_ref] placebo-controlled core saxagliptin studies. [bib_ref] Efficacy and safety of saxagliptin in older patients with type 2 diabetes..., Doucet [/bib_ref] The previous study reported a change in HbA 1c of −0.73% (from baseline mean 8.1%) and a low incidence of AEs comparable with that of placebo. The present results are also comparable with outcomes reported in studies of other DPP-4 inhibitors. In a 24-week study, the adjusted mean change from baseline HbA 1c was −0.64% in drug-naïve elderly patients treated with vildagliptin. [bib_ref] Comparison of vildagliptin and metformin monotherapy in elderly patients with type 2..., Schweizer [/bib_ref] As add-on to metformin in adult patients (all ages), vildagliptin yielded a change from baseline HbA 1c of −0.44%. [bib_ref] Fifty-two-week efficacy and safety of vildagliptin vs. glimepiride in patients with type..., Ferrannini [/bib_ref] The pooled data from five earlier trials of vildagliptin monotherapy showed a mean change from baseline HbA 1c of −1.2% in drug-naïve patients $ 65 years, which was comparable with the reduction of 1.0% in patients , 65 years. [bib_ref] Management of type 2 diabetes in treatment-naive elderly patients: benefits and risks..., Pratley [/bib_ref] In a 24-week trial of sitagliptin monotherapy in elderly patients with T2DM, the mean change from baseline in HbA 1c was −0.7%, the incidence of AEs was comparable with that of a placebo, and there were no episodes of hypoglycemia. [bib_ref] Efficacy and tolerability of sitagliptin monotherapy in elderly patients with type 2..., Barzilai [/bib_ref] Pooled data from the trials of the investigational DPP-4 inhibitor alogliptin, used as monotherapy or dual therapy, show changes from baseline in HbA 1c of −0.7% to −0.8% in elderly patients and −0.5% to −0.6% in younger patients. [bib_ref] Alogliptin use in elderly people: a pooled analysis from phase 2 and..., Pratley [/bib_ref] For linagliptin, no differences in safety and effectiveness were reported in elderly versus younger patients.In perspective, the present analysis suggests that saxagliptin as monotherapy, combination therapy, and initial combination therapy with metformin improves glycemic control in older patients to an extent comparable with that seen with other DDP-4 inhibitors.
With respect to safety and tolerability, the present analysis demonstrated that saxagliptin was generally well tolerated in older patients; the overall incidence of AEs was similar in the older and younger subgroups in the pooled submit your manuscript | www.dovepress.com analysis and also in the initial combination study, with no notable differences between the subgroups and the overall populations of the individual studies. A particular safety concern in elderly patients with T2DM is renal impairment, which may develop independently or secondarily to diabetes. A study of saxagliptin in adult patients (all ages) with T2DM and renal impairment revealed that the use of 2.5 mg saxagliptin resulted in no treatment-related decline in renal function. [bib_ref] Gause-Nilsson I; D1680C00007 Investigators. Saxagliptin improves glycaemic control and is well tolerated..., Nowicki [/bib_ref] For patients with mild and moderate renal impairment, there is no need for dose adjustment. For patients with severe renal impairment or end-stage renal disease, the 2.5 mg dose of saxagliptin is recommended.In the present subanalysis, the mean baseline creatinine clearance rate of the older patients ranged between 80 and 84 mL/min, and no safety and tolerability issues were noted with the 2.5-mg or 5-mg dose. Further efficacy and tolerability data are expected in 2013 from an ongoing study of saxagliptin versus the sulfonylurea glimepiride in elderly patients with T2DM inadequately controlled on metformin monotherapy (ClinicalTrials.gov identifier: NCT01006603).
## Dovepress
## Dovepress
Certain statistical limitations should be considered when assessing the results of this analysis. Although outcomes with saxagliptin appeared similar for patients older and younger than 65 years, the fact that the older subset contained notably fewer patients than the younger subset and that the older subset mainly comprised patients aged 65 to 75 years of age limits the ability to draw conclusions about possible age-related effects.
# Conclusion
In T2DM patients aged $ 65 years, the majority of whom reported a background of polypharmacy, saxagliptin was superior to placebo when used as monotherapy in treatmentnaïve patients and as add-on to another oral antidiabetic drug. The improvement in glycemic control was similar to that seen in younger patients. Saxagliptin used in initial combination therapy with metformin was also superior to metformin monotherapy in antidiabetic drug-naïve older patients, many of whom were taking multiple concomitant medications, and the improvement in glycemic measures with saxagliptin was numerically greater in the older group than in submit your manuscript | www.dovepress.com Dovepress Dovepress the younger group. In view of its favorable profile of efficacy and tolerability and low risk of hypoglycemia, saxagliptin is an attractive treatment option to improve glycemic control in elderly patients with T2DM.
[table] Table 1: Pivotal phase III Studies (24 Weeks) of saxagliptin in patients with type 2 diabetes mellitus [/table]
[table] Table 2A: Baseline demographic and clinical characteristics of older and younger patients with type 2 diabetes mellitus from five pooled studies of saxagliptin vs placebo as monotherapy and as add-on therapy[18][19][20][21]24 [/table]
[table] Table 2B: Baseline demographic and clinical characteristics of older and younger patients with type 2 diabetes mellitus from a study of saxagliptin 5 mg as initial combination therapy with metformin vs metformin monotherapy22 [/table]
[table] Table 4B: Adverse events in older and younger patients with type 2 diabetes mellitus from a study of saxagliptin 5 mg as initial combination therapy with metformin vs metformin monotherapy 22 AEs a (excluding hypoglycemia) occurring in $5% of patients in any treatment group, n (%) Notes: a Presented by preferred term in order of system order class, in the older ($65 years) subgroup; b signs or symptoms of hypoglycemia, with or without documented blood glucose levels; c finger stick glucose # 50 mg/dL, with associated symptoms. Abbreviations: AEs, adverse events; MET, metformin; PBO, placebo; SAEs, serious AEs; SAXA, saxagliptin. [/table]
|
Anæsthesia, or the Employment of Chloroform and Ether in Surgery, Midwifery, &c.
On the Inhalation of the Vapour of Ether in Surgical Operations; con-taining a Description of the various stages of Etherization
M.DProfessorM.DM.DProfessorOn the Inhalation of the Vapour of Ether in Surgical Operations; con-taining a Description of the various stages of Etherization
9 Des Effets Physiologiques et Tlierapeutiques des Ethers Par H
the limbs in animals poisoned by chloroform, never in those killed by ether.* We now proceed to the topics more immediately before us.
1. Mode of Administering Anaesthetic Agents.?When ether-inhalation was first introduced, no one seems to have thought of using the agent otherwise than through an apparatus. To describe all the forms that were invented, would far exceed the limits of this article j and indeed many of them could not be understood without illustrations, their arrangements being so excessively complicated. But as there are still advocates for the use of apparatus, we shall refer to some of the more conspicuous of those inventions. In most of them, the ether and chloroform were contained in pieces of sponge. Mr. Morton first used a spherical reservoir, with two wide and short tubulures; one of these had a valve, permitting the entrance of atmospheric air ; the other was connected with a tube having two valves, one near the reservoir, opening on inspiration, and another on the side, which opened on expiration. Then came the more elaborate apparatus of Mr. Robinson, of London: here there was a reservoir with two tubulures, one of which was connected with a long flexible tube, supplied with valves similar to those of Morton, and a stop-cock in addition, so that the communication between the reservoir and the patient might be instantly cut off.
Besides this, there was an apparatus for compressing the nostrils. Then Dr. Snow invented an apparatus, nearly similar in principle with the others, as far as its stop-cock and valves were concerned; but different as regarded the construction of the reservoir, which consisted of a spiral chamber in a tin box The portion of the box not occupied by the spiral chamber was filled with water, never quite approaching 65?. The effect of this apparatus was to keep the air which passed through the spiral chamber saturated with ether, and at the same time to mix the ether vapour with a sufficient quantity of air. And as Dr. Snow finds, that at the temperature of 64? air will take up about an equal quantity of ether vapour, or rather more than its own bulk, the temperature offers a guide to the amount of air with which the agent is mixed. Thus from an apparatus of this kind, or on this principle, if we know the temperature, we can estimate the extent to which this or any volatile body of the kind is mixed with air; and so knowing the diffusibility of the agent from the quantity of it removed during inhalation, we may estimate the amount of air which has been respired with it.t The face-piece of Dr. Snow's apparatus was so constructed as to allow of respiration by both nose and mouth. Afterwards Dr. Snow rendered his apparatus more portable.
These instruments were made still more complicated in Paris. Of those of MM. Charriere and Luer, only pictorial representations could give an adequate idea, with such a number of tubes, stop-cocks, and valves are they provided. The constitution of the mixture of air and ether thus inhaled, is such as * Medical Gazette, vol. vii. 1848, p. 413. t The researches of Dr. Snow on the amount of air and ether-vapour which mix at various temperatures do not correspond with those of MM. Doy^re, Lassaigne, and other foreign experimentalists. Thus, according to M. Doy&re, the quantity of ether-vapour whiph the atmospheric air may take up, may fall in winter below 17 per cent., and in summer may ascend to 30; and while, according to Snow, 100 parts of atmospheric air, at the temperature of 70?, will absorb no less than 138 measures of ether-vapour, Lassaigne, at 59? Fahrenheit, only found 35 4 per cent., whereas the proportion, according to Snow, should be 80 of ether to 100 of air.
1S-1X.
-11
Practical Applications of Aneesthesia.
[April, might render it inflammable on the approach of a light; and even the presence of aqueous vapour is not sufficient to destroy the inflammability: but when a light is brought near the mouth of an animal whose breath is so constituted, immediately after the withdrawal of an ether apparatus, the only effect of the explosion which takes place is to singe slightly the hair about the mouth; the flame does not pass into the lungs.
Similar instruments to those invented for ether have been used for chloroform, although they have generally been of the simplest description; and yet, perhaps, of the two agents, there is more reason for using an apparatus in the case of chloroform; but Simpson having from the first set the example of dispensing with the use of any instrumental aid for inhaling the new agent, an idea came to prevail that chloroform was more independent of such assistance than ether. Dr. Snow has particularly advocated the use of an inhaling apparatus for chloroform. His argument is, that the fatal cases have arisen in great part from the use of the chloroform vapour in a too concentrated state. His chloroform inhaler is constructed upon a similar principle to the one already described for ether, only it is of a different form; this principle is, as we have just seen, that air at a certain temperature will dissolve or mix with a certain volume of chloroform; and the object is sought to be obtained by surrounding the chloroform reservoir with a bath kept at a certain temperature.* An apparatus a good deal used on the Continent, and partially, we think, in this country, and much praised by M. Sedillot of Strasburgh,+ for the inhalation of chloroform, consists of a round hollow vessel of wood, pierced with holes for the admission of air, and having a tube for the introduction of chloroform, and connected with a caoutchouc mouth-piece, furnished with a valve for the egress of air in expiration. Similar instruments are much praised by Murphy in midwifery,also by Clianning; by Clendon and others in dental surgery. ? The various inventions which they approve of are all simple, and vary a little in form, but are similar in principle, and of the same kind as the only form of chloroform-inhaler now generally used in our hospitals ; consisting of a small tin reservoir, simple means, is the occasional contact of chloroform with the lips and nostrils, whereby irritation and sometimes vesication of these tender parts is produced. This can generally be guarded against with care; and if the precaution is deemed necessary, it may always be prevented by previously rubbing the parts exposed with a little olive oil, or other oily or unctuous substance.
The remarkable success of Professor Simpson is sufficient to show that instruments are not necessary. Some difference of opinion prevails, as to whether it is better to bring the patient rapidly under the anaesthetic influence, or to accomplish this more slowly. Professor Simpson is an advocate for the former mode. A somewhat interesting case occurred lately at Strasburgh, where one moro death from chloroform subjected the medical man who operated to an accusation before the correctional police. The operation was tooth-drawing on a married female. The quantity of chloroform used was small, and the death sudden, almost before the completion of the operation. M. Sedillot, who gave evidence, stated, that in his opinion sufficient precautions had been taken to ensure the entrance of a due amount of air into the lungs; but he condemned the inducing of anaesthesia with chloroform in a shorter period than ten or twelve minutes.t
In surgery, some prefer placing the patient in an ansesthetio state before his removal from the bed to the operating table. Such appears to be the Practical Applications of Ancesthesia.
[April, common practice in tlie Edinburgh and Aberdeen Hospitals; and it is strongly advocated by Professor Miller, (p. 25.) In midwifery, Simpson's mode is to render the anaesthesia as deep as in surgery, at the time when operative or surgical measures are necessary in parturition; but in ordinary labour he induces a much less profound degree of insensibility, increasing, however, its intensity as the pains become more severe, and keeping the patient constantly asleep towards the last and otherwise most painful part of the progress. He believes that the great secret of its exhibition in midwifery, is to give it instantly as each uterine contraction commences, and to withdraw it entirely during the intervals between the pains, during which a state resembling natural sleep will usually continue. In dentistry, Mr. Imlach, who has extracted several thousand teeth under its influence, has given some excellent directions for the use of chloroform. He always uses a large dose, and judges by the effect, not by the quantity used; a cotton handkerchief is his only inhaler; and to prevent the patient clinching his teeth when insensible, he places in the mouth a gag of ivory or gutta percha, which causes no inconvenience, (p. 5.)
The great points in the administration of chloroform are, to watch the pulse, and especially the respiration.
2. Applications of Anaesthesia to Surgery, Midwifery, and Dentistry.? (A.) Of the desirableness of the subjugation or annihilation of pain in surgical operations, considered in itself, we cannot, on the whole, for a moment doubt; to disarm the operating table of a great portion of its terrors, is indeed a triumph of which our age may be justly proud. Not only is the actual pain of an operation thus removed, but also, in great part, that indescribable horror which often torments the patient for some time previously. Men of the greatest courage in other respects, and who have faced danger and death in many forms, have yet shrunk from the prospect of the slow and cold-blooded torture they had before them from the knife of the surgeon. Indeed, however man may summon his fortitude to meet physical pain, or any other dire misfortune, we must all bow to the laws of humanity, and feel the severity of fate, in spite of the efforts of our moral nature to rise above it.
But when we find that this great relief which anaesthetic agents afford, is to be obtained almost without risk, and on the whole with very beneficial results otherwise, we should receive this great discovery with gratitude and exultation. So complete is the general use of anaesthetic agents, that the element of pain as an obstacle or source of danger or of terror in surgery, is for ever almost destroyed. Manual and instrumental therapeutics, as a branch of materia medica, now proceeds, says M. Bouisson, " in the silence, as it were, of vegetative life, and its salutary mutilations are only made known by changes of form without any painful sensation having been experienced by the organism." True it is, as has been already shown, there are exceptions to the general rule; but they are so rare as not to militate practically against it.
Besides the applications of anaesthesia to operations, it may be employed in surgery most usefully to favour the diagnosis of some cases.
Professor Miller recommends its use in the examination of some female diseases, to save the delicacy and modesty of the patients. There are Practical Applications of Ancesthesia. 449 some diseases which cannot otherwise he diagnosed without pain. For instance, in some diseases of the eye there is such intense photophobia, with spasmodic contraction of the eyelids on the entrance of light into the eye, that it is very difficult to open the eyelids, so as to make a proper examination of the organ itself.
In such cases, a moderate degree of anaesthesia will often overcome the resistance of the orbicularis, and destroy for a time the sensibility of the retina, so as to allow of the examination being made. In many accidents the pain renders it difficult to allow of the garments of the patient being removed, and the parts injured being properly examined. In burns it is often difficult to remove the scorched clothes, burned as it were into the skin. In many painful affections of the vagina, accompanied by constriction, it is often hardly possible to use the speculum. Cases of painful catheterism may also be adduced, and necessary exploration of the urinary canal and bladder.* In such cases, and in others which can easily be imagined, as in affections of children, when the struggles of the patient afford an obstacle, and in many instances of feigned disease, as we shall see under our fourth head, the services which anaesthesia may render to diagnosis are considerable. For instance, Simpson, in his letter to Dr. Meigs, says, " My friend Dr. Andrew Wood has just made a beautiful application of it. A boy fell from a height and severely injured his thigh. It was so painful that he shrieked when Dr. Wood tried to handle the limb, and would not allow of a proper examination. Dr. Wood immediately chloroformed him?at once ascertained that the femur was fractured?kept him anaesthetic till he sent for his splints?and did not allow his patient to awake till his limb was all properly set, bandaged, and adjusted."
The surgical applications of anaesthesia may now be considered under the following heads (a) the general indications and contra-indications to its use; (6) the prognosis of operations performed under its influence; and (c) the applications to some special groups of operations. (a.) Anaesthesia is generally indicated wherever there is much pain or great muscular resistance to be overcome. By overcoming pain, it has caused many operations which used to be as rarely performed as possible, to come more in the way of the surgeon: the removal of nails, and the operation of the actual cautery, need no longer inspire horror to the operator or to the patient. With the view of removing muscular resistance, its use has become general in the reduction of dislocations and the operation of the taxis.
The observations of M. Bouisson appear to us here, on the whole, so just and so truly eclectic, worthy in that respect of the reputation for critical talent which the school of Montpellier has long enjoyed, that we * Cases of painful catheterism hardly come under the applications of anesthesia to diagnosis; but of course the same rule applies to them as to the diagnosis of urinary affections. Professor Miller quotes an interesting case of a naval officer who had long suffered under a stricture, where many attempts at cure by bougies had been stopped by the extreme sensibility of the urethra, which always on any attempt to pass these instruments occasioned intense suffering and spasm, long-continued after-pain, and often fever. Hearing of chloroform, he came to Edinburgh, and placed himself under the care of Professor Miller. Under the influence of anaesthesia, the usual treatment could be adopted, and was attended with success. The same author has a very high opinion of the utility of anaesthesia in sounding for stone, especially in children, whose struggles and shouting often interfere seriously with the diagnosis. He mentions an instance where in sounding a boy under the influence of chloroform, he encountered a projecting portion of the bladder, coated probably by some sabulous deposit, and where, but for the facility which the anaesthesia gave to the examination, he might have wrongly inferred the presence of a calculus. Practical Applications of Anaesthesia.
[April, sliall follow his exposition pretty closely. He commences by remarking, that the sources of the contra-indications to the use of anaesthetics are very numerous and various, although the sum of cases in which these agents ought not to be used, is very insignificant in proportion to the immense number in Avhich they can be employed with perfect safety and advantage. He recommends that anaesthetic agents should not be used before the expiring of the first six months of life; and that only ether should be given up to the second year, reserving chloroform for the epoch when the vital powers have acquired more energy ; and for a similar reason, he prefers the former agent .after the seventieth year. He would, then, only employ ether up to the eightieth year, at which lie considers that anaesthesia ceases to be applicable in any shape; not because there is absolute danger in departing from these general rules, but more prudence in obeying them. With regard to morbid states: " Diseases of the nervous centres, and of the lungs and heart, contra-indicate the employment of anaesthetic inhalations, when the physical or functional lesion is advanced to a certain degree. This degree can only be determined by the sagacity of the practitioner. A simple catarrh would not oppose the practice; but it would be dangerous to have recourse to it in phthisis, with spitting of blood; a slight hypertrophy of the heart would not be a sufficient contra-indication; but anaesthesia should be renounced if the lieart-affection were accompanied with irregularity and intermittence of the pulse. A recent nervous affection, far from being aggravated by etherization, might be ameliorated; an old nervous affection with a tendency to syncope, might be a source of the greatest danger. It must be understood that the appreciation of these differences is the province of the operator. Here we can only establish some general principles. " Ether, then, and especially chloroform, should be renounced in the following cases: " 1st. In patients extremely weakened by haemorrhage, or by a spontaneous chlorotic anaemia. " 2nd. In epileptic patients, and hysterical persons with weakened constitutions. " 3rd. In individuals who appear very much disposed to cerebral congestions, or who have softening of the brain. " 4th. In those who are very subject to fainting fits. " 5th. In such as are disposed to haemoptysis, or to pulmonary apoplexy.
For example, in several operations on tlie anus and rectum, tlie surgeon is assisted by the efforts which the patient is desired to make, and by the attitude which he maintains, as in excising or tying internal piles. In the extraction of foreign bodies, also, which have penetrated into the tissues, it is often useful for the patient to take a posture similar to that which he had at the time of the accident.
3rd. Operations where the sensibility serves as a guide to the operator.
Some objections have been made under this head without sufficient foundation. M. Lallemand, for instance, expressed his fears that in tying arteries a neighbouring nerve might be included in the ligature, on account of the absence of the usual sensation during the anaesthetic state. But there are very few arteries where such a mistake is likely to occur; perhaps the only one being the subclavian, on account of the numerous nerves of the brachial plexus surrounding it, and its liability to anomalies of position.
Litliotrity has with more reason been enumerated as subject to the objection, on account of the possibility of piercing the walls of the bladder without the consciousness of the patient. On the continent, the opinions of surgeons have been much divided on this point. According to M. Leroy d'Etiolles, etherization is especially beneficial in lithotrity, when the calculus is contained in a bladder whose walls are enlarged, and whose muscular fibres are thickened, and, as it were, embrace the stone. Here anesthesia, he thinks, promotes a relaxation of the walls, and, by disengaging the stone from pressure, renders it more easily seized. But in such bladders we are also more liable to pierce the walls by the instruments used. M. Serre, of Montpellier, for this reason and others, opposes the practice of anaesthesia in this operation; and relates the case of a customhouse-officer, who had already been subjected to lithotrity, upon whom it was impossible to operate under the influence of ether, but who was successfully operated on afterwards when not etherized. On the other hand, M. Bouisson mentions the case of a female, in whom there was much general nervous excitement and irritation of the urinary passages, and in whom anaesthesia was employed during the operation of lithotrity with obvious benefit. And M. Amussat has given the cases of several old men who had severe attacks of cystitis, after lithotrity had been performed without ether, and were free from any such affection after having been etherized in other operations. In this country, also, opinions are divided on this subject. Keith, in a communication in the ' Edinburgh Monthly Journal,' where he reports an interesting case of lithotrity, says? "The influence exercised by the chloroform was altogether beneficial. Usually the sphincter of the bladder is so tightened by the patient's fears and struggles, that the movements of the instruments are very much impeded, the delicacy of touch needful in detecting small fragments quite destroyed, and a painful resistance offered to the withdrawal of the instrument, especially if that chance to be a scoop well filled with the debris of the broken stone. But under chloroform, fears are at rest, the bladder insensible, and the sphincter quite relaxed." * He proceeds to refer to other advantages. At a meeting of the Medico-Chirurgical Society, when Dr. Keith read his paper previous to its publication, Mr. Syme is reported to have said, that he " feared the consequences likely to result from the performance of lithotrity on insensible patients by * Edinburgh Monthly Journal of Medical Science, April, 1848. Practical Applications of Ancesthesia.
[April, operators less expert tlian Dr. Keith;" and from a recent communication in the ' Lancet,'* we find that he still retains this opinion. In general, it may be remarked, lithotrity is not a very painful operation; sometimes anaesthesia may be useful in injecting the bladder, and introducing the apparatus where there is much irritation, but the chief pain is encountered during the expulsion of the fragments. 4th. There are some few operations where the object is the production of pain. Such are the cases in which moxa or other means of counterirritation are used, in order to rouse the sensibility of the spinal cord or of the limbs, not simply to produce inflammation.
5tli. Operations performed in cases where there exist previous causes of torpor and insensibility. In the operation of trepanning, for instance, the patients are often plunged in coma, or nearly insensible to pain; in such cases it would be, to say the least, superfluous to have recourse to anaesthesia.
(b.) On the first discovery of etherization, fears were expressed that the state of anaesthesia might have a bad effect on the after-results of operations; that the healing of wounds might take place less readily, and that other accidents might supervene. But experience has proved the reverse; the general opinion of surgeons is as favourable, as regards the after-consequences of etherization in surgery, as of the immediate effects. The diminution of the shock to the nervous system seems to favour the healing of wounds, and altogether increases the chances of recovery. The shivering and re-active fever which often follow operations are greatly diminished by etherization; there is generally more sleep, and a more complete feeling of comfort. Contrary to what some have asserted, the pain which follows or accompanies the return of consciousness after an operation, is less intense after etherization; although, perhaps, in some few cases, the absence of pain during the operation may render the patient comparatively more sensitive to the after-pains. The most common sources of danger which follow surgical operations, are the nervous phenomena, violent inflammatory affections of the wounded surfaces, gangrene and purulent absorption, and haemorrhages. Both the first and second of these are obviously diminished; indeed, the diminution of nervous irritation has a direct tendency to moderate inflammatory action. A few examples of hospital gangrene, which occurred in the practice of M. Roux, were made the most of by the opponents of etherization. When men are to be found capable of maintaining such a monstrous doctrine as that pain itself was beneficial in surgical operations, it is not surprising that every little instance of misfortune, from whatever cause arising, should be attributed to the influence of etherization.
Professor Miller states, that at one time he was inclined to think a tendency to erysipelas, which existed in the Edinburgh Infirmary, was due to this cause; but a more extensive experience convinced him of his error.
It is evident that purulent absorption cannot be promoted by etherization.
The only evil consequences, in our opinion, which can fairly result from anaesthesia in the ordinary way, are certain nervous phenomena, either of an hysterical character, or a kind of nervous asthenia or prostration, headache, and cerebral congestion, which may perhaps, in some rare cases, proceed to assume an inflammatory character; and irritation of the air passages, which may likewise, in some rare cases, pass into a species of pneumonia. Of the fatal cases we have already sufficiently spoken; but in considering what weight should be ascribed to them, and to the ill consequences which may occasionally result from etherization, we should not forget the ill consequences which may result from pain itself, and the fright attending its mere expectation. Ranking has related a case where death itself appears clearly traceable to mere pain.* We ourselves, on one occasion, had occasion to experiment on a Russian poodle. On merely cutting the skin of the thigh, the animal, apparently in perfect health, suddenly expired, without other cause, as we believe, except pain and fright. But it is not by quoting one or two isolated cases, or trusting to mere opinions, that we can form a satisfactory conclusion on this subject. Statistical results are our best guides, and we have them to a sufficient extent to enable us to form a clear judgment. Finally, Simpson himself, from statistics of English hospitals, makes out 183 fatal cases in 618 amputations, or 29 per cent. The same author has collected the results of 302 operations of the same class, performed in English and Scottish hospitals, on etherized patients, and the result is only 71 deaths, or 23 per cent. But this is not all. Of the particular operations collected, the most numerous are those of amputations of the thigh. At the same time, there are no operations in surgery of a more fatal description. According to Mr. Syme, the mean mortality is 60 or 70 per cent. Of 987 amputations of the thigh, collected by Mr. Phillips, 435 died, or 44 per cent. Mr. Curling, in an account of the amputations practised from 1837 to 1843 in the London hospitals, finds a proportion of 41 deaths in 100 cases. In short, to conclude, and omitting details, Malgaigne gives in the Paris hospitals a mortality of 62 per cent, in cases of this description; Peacock, in the Edinburgh Infirmary, 49; Phillips, in his general table, 44; Lawrie, at Glasgow, 36; and Simpson, taking the whole reports of English and Scottish hospitals, 38; but in etherized patients only 25 per cent., collected from the same sources.
Numerous statistical tables of the kind could be collected from the various minor works published on the subject of anaesthesia in different countries, most of which we have now before us; but we shall content ourselves with remarking, that they all tend to the same end?namely, to Practical Applications of Anaesthesia.
[April, prove, not merely the general liarmlessness, but tlie positive benefits of anaes- (c.) There are some special surgical operations, respecting which a few words may be said; since it is not in our power to pass in review the various considerations which belong to the application of anaesthesia to the varied phases that surgery may offer. It is iu amputations, that the discovery of anaesthesia has achieved its highest triumph. It is curious, as M. Bouisson remarks, that Professor Burns should, before this discovery took place, have called attention, after speaking of the shock to the system produced by the removal of a member, or of a considerable portion of the living mass, to the benefits likely to accrue from the ?partial benumbing of the nervous system, so as to destroy, more or less completely, the sympathetic relation of parts.
Among special operations in which doubtful benefit is derived from anaesthesia, may be reckoned, 1st, those called for in several affections of the eyes.
In cataract, the excitement which some etherized patients undergo, and that manifested on the return of consciousness, may interfere with the success of the operations; but in the great majority of operations on the organ, anaesthesia may be used with great benefit. In operations practised on the parts connected with the posterior fauces, there has been much fear of the consequences of the bleeding in etherized patients. We shall confine ourselves to stating, that in these operations extreme caution should be used; and that this caution should be increased as the principal seat of the disease to be removed, or of the operation, approaches the glottis. Nevertheless, Professor Miller has shown how, by careful management, the dangers to be encountered in cases of this kind can be considerably diminished.
It is not necessary to point out the occasional benefit which may be derived from the use of anaesthesia in the taxis, in the reduction of dislocations, and setting of fractures. In all this class of surgical operations, cases must be continually occurring, proving to the surgeon the immense results which he may reap from the new discovery. There remains one important operation, in which, at first, considerable doubts were apprehended with regard to the application of anaesthesia?viz., lithotomy. For instance, it was supposed that the relaxation of the vesical walls which occurs in profound etherization, would prevent the bladder from contracting after the incision of the neck, and so allow the calculus to escape. Other objections of as little value were made, which experience has fully demolished.
Anaesthesia is now used in private and in public practice, as regularly in lithotomy as in other operations; and it is unnecessary to recall arguments which have almost been forgotten. It seems almost unpliilosophical to conclude this brief summary of the results of anaesthesia in surgery in a voice of triumph. Yet who that feels a pride in the progress of his profession, can refrain from rejoicing in the great benefit, which, within so short a space of time, it has conferred on suffering humanity 1 Henceforth, even the cock-pit of a man-of-war, and the hospital after a field of battle, will be disarmed of half their terrors.
Henceforth the medical profession, as a body, cannot be accused of an unwillingness to adopt new discoveries; as the invention, labour, and zeal, and even the critical care, bestowed on this subject by the profession abundantly testify. (B.) Let us now proceed to consider the application of etherization to Midwifery, and the alleged advantages of the practice with the principal arguments brought forward in opposition.
We cannot do better than commence this part of our subject with the first recorded case of anaesthesia in midwifery. Professor Simpson says? " The first case in which I employed the ether vapour occurred on the 3 9th of January The pelvis of the mother was greatly contracted in its conjugate diameter, from the projection forwards and downwards of the promontory of the sacrum; the lumbar portion of the spine was distorted, and she walked very lamely. The present was her second confinement. Her first labour had been long and difficult; she began to suffer on a Monday, and after a protracted trial of the long forceps, was at last delivered by craniotomy, late _on the subsequent Thursday night."
As, in her present confinement, the medical attendant had not been made aware of the fact until nearly the full term, it was too late to have recourse to the induction of premature labour. " The pains of her second labour commenced in the forenoon of the 19th. I saw her with Mr. Pigg, at five o'clock in the afternoon, and again at seven. The os uteri was pretty well dilated, the liquor amnii not evacuated, the presenting head very high, mobile, and difficult to touch; and a pulsating loop of the umbilical cord was felt floating below it in the unruptured bag of membranes. Prom five to nine o'clock, the pains seemed only to push the circle of the os uteri further downwards, without increasing its dilatation, or making the head in any degree enter into the pelvic brim I, shortly after nine o'clock, made the patient inhale the ether vapour. She was soon under the influence of the vapour, which was continued for about twenty minutes, when the operation of turning was performed. The extremities and trunk were readily extracted; but great difficulty was experienced in extracting the head. With extreme exertion it at length passed the contracted brim with the anterior part of its right parietal bone deeply indented by pressure against the projecting promontory of the sacrum, and the whole cranium flattened and compressed lateially." The child gasped, but soon died. The transverse or biparietal measurement of its head, at the site of the indentation, was, in its compressed state, not more than two inches and a half. Hence, we judged the conjugate diameter of the pelvic brim not to exceed this. The infant, was large, and rather above the usual size."
The mother was insensible during all the time, until she heard, but did not feel, the head of the infant jerk from her. She rapidly recovered, expressed her gratitude for the exemption she had enjoyed, and was up and dressed on the fifth day after her delivery. This case was, it must be admitted, sufficiently encouraging; and accordingly Professor Simpson pushed the practice, both before and after his introduction of chloroform, with his well-known energy and resolution. Previous to this it was known that the motor nervous powers concerned in parturition belong to the ganglionic and to the spinal systems, and are not necessarily dependent on the brain or the state of mind. Cases of perfect paraplegia had been published, where the act of parturition had proceeded regularly without any consciousness of pain. Haller long ago adduced the .authority of Harvey, Smellie, Lamotte, &c., to prove that uterine contractions and labour may go on with the mother ignara stupida Amiens, while in a state of unconsciousness from drunkenness, and where, on coming to her senses, she was so much delighted at the result, as to declare her intention of making herself drunk in future should she have the same process to undergo.
In considering the propriety of using etherization in ordinary midwifery, the first consideration is the amount of pain to be relieved. Some regard the general amount of pain as so bearable, as not to warrant the use of anesthesia; while others even maintain the pain in parturition to be of a conservative character. Thus Professor Meigs of Philadelphia, says, " I have been accustomed to look upon the sensation of pain in labour as a physiological relative of the power or force; and notwithstanding I have seen so many women in the throes of labour, I have always regarded a labour-pain as a most desirable, salutary, and conservative manifestation of life-force."* And he maintains that women, if sufficiently sustained and comforted, may in general be made to support the amount of pain, incidental to parturition, without much suffering. On the other hand, the advocates of anesthesia, with Professor Simpson at their head, maintain both that the pain is often so excessive as to warrant the use of means to annul it; and that there is nothing about the character of pain in ordinary labour, to authorize it being called natural and conservative. Professor Simpson quotes numerous authorities to show that the amount of suffering is often enormous.t
Here we perfectly agree with him; and as this suffering does not occur with all races of mankind, but seems especially an attribute of high civilization, and perhaps of the increased dimensions of the foetal head due to high civilization, certainly what civilization has brought about cannot be termed natural, least of all ought that method which takes it away to have the epithet unnatural applied to it, as by Asliwell and others. We wholly agree with the arguments of Professor Simpson on this point. The ridicule with which he has overwhelmed his opponents is as well merited as it is severe. This cuckoo-cry is precisely similar to that which was got up against vaccination, and might with equal sense be applied to the invention of steamboats or railroads. It is a very striking fact, that the benevolent Dr. Push should have expressed a wish, so many years ago, that some means could be devised for freeing women from the torture they experience in labour. He first mentions the exemption which women have in some parts of the world, and then goes on as follows, " I have expressed a hope in another place, that a medicine would be discovered that should suspend sensibility altogether, and leave irritability, or the powers of motion, unimpaired, and thereby destroy labour-pains altogether. to cherish this hope, by having known delivery to take place, in one instance, during a paroxysm of epilepsy, and having heard of another, during a fit of drunkenness in a woman attended by Church, in both of which there was neither consciousness nor recollection of pain."* The sole question we have to consider is, whether the relief of pain by etherization is beneficial or injurious to mother and child; common sense teaches us that pain is an evil, and if we can remove it without injury, it is vain to plead any fantastic notions.
Some have committed the very grave error of confounding the sensation of pain with labour-pains, or uterine contractions. That the contractions are essential to delivery is notorious ; but the question is, whether they are necessarily attended with pain to the mother 1 It is true, in some cases, the uterine contractions have appeared to be lessened or suspended during the state of anaesthesia; but in others they have even been thought to be increased.
It has appeared to many, that owing to the action of the will being suspended, and no voluntary movement made to impede the labour, this increase of the contractions is directly due to the ansestliesia. Dr. Channing thus expresses himself on this head: c: The whole uterine power is exerted regularly and uniformly, and progress almost at once declares itself, dilatation of the passage, as we have before seen, very soon becomes more perfect and determines the amount of contractions. The secretions are increased. In short, a state most favourable to easy, rapid, and safe delivery is produced and sustained, how successful both to mother and child the tables hereafter will show."f As opinions are divided with regard to the precise degree of influence of etherization on the uterine contractions, it may be well to give the opinions of some of the authors whose publications are lying before us. Dr. Protlieroe Smith says,?" It may momentarily suspend the natural contractions; but it does not suspend them, nor the contractions of the abdominal muscles, when energetically set up. "J Professor Murphy, after referring to a case related by Dr. Shekleton of Dublin, where the administration of chloroform seemed to suspend the contractions, and to two similar related by Denliam, comes to the following very important conclusions: "1st. That chloroform does not impair the contractile power of the uterus, neither does it interfere with the action of the uterus, unless it be given in very large doses, or the patient be highly susceptible of its influence.?2nd. That the full anaesthetic effect of chloroform may be produced without paralyzing the uterus.
The intervals between the pains may be lengthened, or the pains be s uspended, without any loss of power. On the contrary, in such instances the moment that chloroform is withdrawn, the uterine contractions instantly return with increased force and efficiency. The renewal of the uterine action may occur when the patient is under the influence of chloroform.?3rd. Whether the action of the uterus be temporarily suspended by chloroform, whether it be increased or not interfered with, in every case where the patient had previously suffered agonizing pains, and her labour was making an unfavourable progress, chloroform has produced a most salutary change in restoring the proper action of the uterus, by which labour was brought to a happy conclusion.?4th. Those efforts that indicate want of power in the uterus?viz., great protraction of labour, haemorrhage, retained placenta, &c.?have not been proved to he the results of | chloroform; on the contrary, where some of these conditions have been observed, The action exerted by anaesthetic vapours on the contractility of the uterus is not so easily verified; so that it has given rise to contradictory opinions, to which practical consequences have come to be attached.
The general fact most easily appreciated is, that under the influence of a moderate etherization the uterus can expel the product of conception even when its sensibility is completely annulled. But does the contraction remain in its normal limits ? The reply should be in the affirmative, if we refer to the observations of Messrs. Simpson and Paul Dubois. It would experience, on the contrary, notable modifications from the evidence of some other accoucheurs. M. Stoltz, for example, has cited facts from which it results that the womb Ins a tendency to contract more energetically under the influence of ether, and that this increase ctf activity, by narrowing the cavity of the organ, hinders the obstetric operations which may be had recourse to. According to MM. Bouvin and Sieboldt, a contrary action is exerted by anaesthetic agents."** * Edinburgh Monthly Journal of Medical Science, opinion is, that on the whole labours are somewhat prolonged under the anaesthetic state, but not to such an extent as to interfere seriously with the use of the agents.
Let us now proceed to consider the principal arguments adduced in favour of anaesthesia in midwifery, and those on the other side; since we have cleared the ground, as it were, and shown that pain can be removed without any serious interruption of the natural process of parturition, if not with actual benefit to it.
The arguments in favour of the employment of anaesthesia may be summed up as follows:
1st. The removal of the pain is beneficial to the mother by preventing the nervous excitement and shock which physical pain is apt to excite, and the nervous and inflammatory reaction which, in some constitutions, is apt to result.
2nd. It renders many operations, requisite in complicated labours, easier of performance, and more beneficial to the patient.
3rd. Statistics prove the practice of anaesthesia to be beneficial to the mothers, and nowise dangerous to the children.
The first proposition appears almost self-evident, unless the opponents of anaesthesia could prove specific objections which we shall consider. In many delicate females, the excessive shock to the nervous system during delivery is a most serious source of injury. Most practitioners must have Practical Applications of Anaesthesia. 461 showed great improvement, and what I have not observed before, consciousness returned. But so deeply impressed was the whole system by the disease, so powerful was the action of its exciting causes, and so strongly marked were the symptoms of predisposition to it as they existed during pregnancy, that convulsions recurred before delivery, and though this was accomplished with great ease, were fatal. If the question should be asked, why was not etherization continued during the three hours of absence of convulsions ? I answer, that there was nothing in that condition which seemed to me to require or to warrant its employment ; and further, had fatal convulsions followed its use, would not this have been regarded as the direct result and elFect of needless etherization ? But it was not to escape any such possible issue as this, that the use of ether was omitted. It was discontinued, simply and wholly, because it was not thought necessary. This statement enables us to avoid erroneous inferences, and thereby a groundless objection to etherization is removed."
Dr. Channing, in one of bis tables, presents a series of 28 cases of instrumental, preternatural, and complicated labours, in which etherization was not used. Of these 15 died! Of the 7 cases of convulsions, 6 were fatal, and there were 1G still-births! While of 51 cases of the kind, where etherization Avas employed, only 4 were fatal, and there were 19 still-births! Had any practitioner met with such results as the former of these tables exhibits, at the outset of the practice of etherization, doubtless the practice itself would have been to blame in his eyes.
For the statistics which have been published of the results of etherization in midwifery, we must refer principally to the works of Dr. Simpson and Dr. Charming. Channing's tables are too extensive to be transferred to our pages. The works of these accoucheurs contain, also, a large correspondence with numerous practitioners, which proves, not merely the safety, but the benefits, of the new practice. Simpson, in 1848, communicated the results obtained in 1519 cases, and in our opinion established the utility of anaesthesia upon incontestable grounds. But let us consider some of the objections which have been made.
These are chiefly; 1st, the so-called religious objections; 2nd, the assertion that etherization in pregnancy is immoral, and apt to excite improper feelings in the patient; 3rd, that in some cases it renders the use of the forceps dangerous; 4th, that it is injurious or dangerous to the patient; and lastly, that it is injurious to the child.
Our readers are well aware of the mode in which Professor Simpson encountered the argument from the third chapter of Genesis?namely, by showing that the true meaning of the Hebrew word used to express sorrow, did not offer any obstacle to the practice of amesthesia. We should Ihave preferred to have encountered the objection in another manner.
[Surely if the Almighty had intended woman, or rather ordained her, to suffer physical pain in giving birth, what effort of man could have availed to mbdue his will? But it must be remembered, that there are certain preudices, which, however we may be disposed to regard them as prejudices, ought still to be respected on account of their origin. And therefore perhaps Professor Simpson acted wisely in taking up the ground he did.
Of the second objection very much lias been made.3" Undoubtedly, Tnsesthesia does occasionally give rise to erotic sensations, but these are ;o very rare as not to amount to an objection. to be shown, whether the occurrence of the emotions referred to be more common in obstetrical than in surgical practice; so that if there be anything in the objection, it is applicable to the entire use of anaesthesia. On this point Simpson says?
" I have never once witnessed any trace of indecency, either in word or action, in any obstetric patient under the use of chloroform; and the evidence of one and all of my obstetric brethren of whom I have inquired on the subject, is to the same effect." (p. 148.)
The third objection has been principally urged by Meigs, who maintains that the sensations of the patient afford us our best guide for the introduction of instruments. He says, in order to introduce the forceps with the greatest safety to the mother, " the best guide of the accoucheur is the reply of the patient to his interrogatory 'does it hurt youV The patient's reply, yes' or ' no,' are worth a thousand dogmas and precepts." Simpson answers the argument of Dr. Meigs by asserting, that every accoucheur possesses a much greater anatomical knowledge than is here supposed. Certainly the argument does look exceedingly like a plea for ignorance; and its only value arises from the very high authority from which it emanates. We disposed of Lallemand's analogous argument about the danger of including nerve and artery in the same ligature, in a similar manner.
The arguments under the fourth head, if these were true, would be fair; but the vast amount of testimony of a favourable character now before the world, renders it unnecessary to examine them in detail. The most has been said, that could be said on the subject; and time, if it has not fully resolved the question, is rapidly disposing of it. With regard to the alleged injurious effects on the children, Channing observes, "Etherization, we are told, may mentally or physically affect the infant. It; may be weakened in body or mind; and particular diseases are named, which may be looked for from the agency of the remedy of pain. This objection rests upon hypothesis alone. It has no facts for its origin, and none for its support."
Nevertheless, M. Amussat, in some experiments which he made on the females of animals in a state of gestation, observed, when etherization had been carried so far as to produce a commencement of asphyxia?the last stage of Dr. Snow?that the foetuses were stupified to some extent, and their blood of a darker colour than usual.* M. Dubois, also, has noticed some acceleration of the pulse in new-born infants. He has observed an increase to the extent of 160 or 170 beats, instead of 130 or 140 ordinarily observed.
M. Bouisson regards these facts as exceptions, and concludes, generally that etherization exercises no deleterious influence over th-b health of the child.?This, moreover, does not constitute a valid objection*, against chloroform; for we have been informed by obstetric practitioners' who had made trial of both agents, that while the influence of ether was frequently exerted upon the infants, as shown by their anaesthetic conditio!1 at birth, no such result has ever been witnessed by them since they have employed chloroform.
The same general rules which are applicable, as contra-indications to the occasional use of anaesthesia in surgery, hold good in obstetrics. In th] latter branch prodigious quantities of chloroform have been given, and th * Bouisson, p. 4S4. In all cases of forceps-application, frequently in difficult contraction of the placenta, in turning, in serious obstetrical operations, such as the Cesarean section, symphyseotomy, craniotomy, and such like, and generally in all cases of painful labour, he recommends the use of anaesthetics.
But in cases of ordinary labour, where the pain experienced is slight, he hesitates.
Dr. Channing has given some excellent general rules,* the principal of which is, not to employ anaesthesia in forceps cases until the instrument is applied, so as to mitigate the objection of Dr. Meigs ; but this rule necessarily exposes the patient to considerable suffering during the previous part of the labour, and should not, therefore, be practised by one sufficiently conscious of his own resources.
We cannot quit this part of our subject without paying the illustrious professor of Edinburgh that tribute which he deserves for the courage, energy, and perseverance, with which he has carried out the application of anaesthetics to his own peculiar branch of our art. Nor is less credit due to him for the moderation and patience with which he has encountered some most harsh and wholly undeserved attacks. If he has sometimes used ridicule, it was where ridicule was the most efficient argument* Practical Applications of Ancesthesia.
[April,
3. Therapeutical Applications of Ancesthesia.?It is now evident that the use of anaesthetic agents is capable of an extension beyond the bounds of merely operative medicine. It has been transported into medicine itself; and perhaps this circumstance may attract the attention of the profession to the advantages which may accrue from the use of other remedies in the form of inhalations. When we consider the great extent of the pulmonary mucous membrane, and the facility with which vapours may be introduced through the respiration into the blood, it seems extraordinary that this mode of administering medicines has not been more exactly studied. Dr. Beddoes, as is well known, established an institution at Bristol for the purpose of investigating the therapeutical effects of gases; and, independent of other results, Ave have seen that to this very institution Ave OAve, through the genius of Davy, the first hint of anaesthesia by means of nitrous oxide.
Pneumatic medicine may, perhaps, noAV receive a fresh development. Pain exists in a vast number of diseases, Avliere even opium is insufficient to afford relief; in such cases a field is opened to the use of anaesthetic agents.
The first class of diseases Avhich present themselves to our notice, are neuralgic affections. In several cases of intense facial neuralgia, benefit has been obtained from the inhalation of chloroform, when all other remedies have failed.
In pains of the boAvels, gastralgia, and in nervous colics, similar results have followed. Many observers have pointed out the benefit derived from doses of chloroform in the liquid form, in relieving the pain in the early stages of cholera. We can ourselves confirm this, with the addition that the doses ordinarily administered of chloroform in the liquid form have been much too small.
We have ourselves given ten or twelve drops for a dose, or even more, instead of the doses of three or four drops recommended by some. Frictions of chloroform along the vertebral column have also been recommended. Ether at one time was announced, before the discovery of anaesthesia, as a specific in cholera; but, in fact, remedies of this kind can only be considered as palliates. It has been said that the spasms of angina pectoris have been averted by anaesthesia. A M. Aubrun " Avas able by employing it to lengthen to eight days the intervals betAveen attacks (of the kind) which A\Tere formerly reneAved several times in the twenty-four hours." M. Bouisson particularly points out the benefit Avhich has been derived from anaesthesia in nephritic colic; and the probability of its being equally useful in relieving the torture Avhich is occasionally felt during the passage of a biliary calculus. The relief experienced from the internal administration of ether in such cases had been before noticed, and ascribed to the solvent action of the ether on the fatty constituents of the calculi.
There are many spasmodic affections likewise, closely allied to the class of Avhich Ave have been speaking, where the benefits of anaesthesia have either been verified, or are likely to be so; although in some such affections there have been very different results.
For instance, in some cases of hysteria, etherization has produced a calmative effect?in others the reverse. In some patients, it has even appeared to bring on a fit of the disease. In epilepsy the evidence is, on the Avhole, unfavourable to the use of anaesthetic agents, and especially of chloroform. Where the symptoms seem to be accompanied Avith organic disease of the brain, this, of course, is particularly Practical Applications of Ancesthesia. 465 the case; so much so, that in such cases the practice should be abandoned altogether. Hooping-cough, nervous asthma, chorea, and hydrophobia, may be particularized as diseases in which much was hoped from the inhalation of these agents, and where the results as yet have been insignificant. The cases in which most benefit may be expected, are, as Dr. Simpson has remarked, those in which the spasmodic affection results simply from undue excitability or super-polarity of the spinal centre. He has recently published a very interesting case of convulsions in an infant of only a fortnight old, in which the very best effects accrued from the use of chloroform, after all other remedies had failed, and the child seemed rapidly sinking; and he refers to the recent experience of this agent in pneumonia, as indicating that it may be freely used in the spasmodic stage of hoopingcough, without fear of inducing or aggravating pneumonic inflammation.* In tetanus many favourable cases have been reported; although one of the first trials terminated fatally ; and the patient was supposed to have perished partly from the effects of ether. This was in the practice of M. Roux. M. Bouisson states that he has been able to collect 15 reported cases of etherization in tetanus; and of these 10 terminated favourably. In 7 of these cases, ether was the agent employed; and in 3, chloroform. It is worthy of observation, that ether having been employed in the liquid form, administered by the stomach in the treatment of tetanus, we have thus a means of comparing the respective energies of the gastric and pulmonary mode of administration; and the result shows that the former is decidedly less efficacious than the latter.
M. Hutin, surgeon-in-chief to the Hospital of Invalids, says, that in Africa he had observed 60 cases of traumatic tetanus; and that of these a third had been treated by ether in the liquid form, administered by injections and frictions along the vertebral column, without favourable results. However, it might be well, in the treatment of such a disease as tetanus, not to rest content with one mode of applying ether or chloroform, but to use these bodies in various ways, by way of inhalation, friction, or otherwise.
In mental alienation, anaesthesia has been a good deal used. Mac-Gavin of Montrose, has especially pointed out its beneficial effects in cases of madness with much excitement.
One great use of anaesthesia in lunatics would be, that by means of it their indocility could be overcome, and thus they might be brought to undergo operations to which otherwise they would be refractory. It has been thus used to administer food through tubes in the oesophagus, to patients who were resolved not to swallow, but rather to die of hunger.
In meningitis and delirium tremens, the use of anaesthetic agents is of very doubtful advantage. [April, the hands of some it has been found exceedingly successful in relieving pain and subduing inflammation, and in the dressing of ulcers. The first effect of the local application is to produce some degree of redness; but this soon subsides; and the secondary local action is decidedly anaesthetic. It is very plain to us that we are only at the beginning of the medicinal use of these agents; and we would refer in illustration to the remarkable results of the use of chloroform in the treatment of pneumonia, which have been obtained by Wacliern, Baumgartner, Hebling, Schmidt, and Varrentrapp, who have altogether treated 216 cases in this mode.* 4. Applications to Legal Medicine.? In order to show the use which may be made of anaesthesia in feigned diseases, Ave extract the following from the work of M. Bouisson. He had for a long time thought of employing this means in the diagnosis of deafness supposed to be feigned: " Chance," he says, " had not furnished me with a case of the kind; hut having in 1848 to treat a soldier whom I suspected of counterfeiting deafness, and who showed himself very clever at his part, I chloroformed him slightly, and on the first appearance of anaesthetic sleep I asked him questions, in a low tone, to which he replied. This result may be explained by reflecting on the effect of these stupefying vapours on the understanding. The unconnected chain of ideas, as in drunkenness, do not admit of those precautions which are requisite for the success of the pretender." Simulated dumbness and stammering can be detected in a similar way; and in pretended muscular contractions, the method is of easy application, and has been found very successful.
It is very obvious that anaesthesia might render very much easier such a plot as the celebrated one in which the Countess of Saint Geran was the victim; for whatever opium or any other poison of the kind could accomplish, ether or chloroform could certainly effect more surely. The substitution also of one child for another could thus be rendered easy. Such circumstances are no objections to the practice of anaesthesia. Every great discovery has some inconvenience inseparable from the lot of mankind.
In the same way, no doubt, etherization might be employed by the libertine to accomplish criminal purposes; and young or feeble persons might be made to perish by means of ether, or especially chloroform. We have already spoken of the post-mortem appearances to be expected in such cases. We shall conclude, however, by quoting on this head the following observations of M. Bouisson. He says, "Although etherized or chloroformed animals which have been killed by this form of intoxication, are exempt from constant and characteristic anatomical lesions, they often present the greater part of the appearances of asphyxia. The heart is distended with blood, the lungs are coloured of a deep red; there exist sometimes apoplectic spots iu these organs, ecchymoses beneath the pleura, or traces of emphysema. The liver, of the colour of wine-lees, is gorged with black blood; the kidneys are of a violet-colour, owing to sanguineous congestion. The vessels of the meningeal membranes are distended; the pia mater is especially injected towards the lower part of the brain and the annular protuberance. The cerebral pulp is in general spotted with blood. The blood itself is blacker and more fluid than is natural. Bubbles of air are often remarked. But the truly important character, and which can distinguish anaesthetic from ordinary asphyxia, is the characteristic odour of ether and chloroform preserved by the blood, the fluids, and the tissues. M. Flandin, by analyzing the blood, has extracted ether from it; and M. Lassaigne, while declaring that it is difficult to obtain much, does not the less point out the decisive marks of its presence."
We observe that Dr. Glover, in his 1 Experiments on Chloroform,' published in 1842, remarks upon the power which chloroform and its allied bodies appear to possess of penetrating the tissues. In speaking of one experiment, he observes, " the smell of the chloroform was distinct in the chest, and in the urine passed a few minutes before death."
We conclude by congratulating the profession on the amount of zeal and energy which has been displayed in this interesting subject; and by hoping that anaesthesia will be carried to the furthest extent of which it is susceptible. We are no Sybarites; but unfortunately, with all our efforts, there will still remain in the world a sufficient amount of misery. Let us do what we can to mitigate human suffering. |
Historical ESWT Paradigms Are Overcome: A Narrative Review
Extracorporeal Shock Wave Therapy (ESWT) is a conservative treatment modality with still growing interest in musculoskeletal disorders. This narrative review aims to present an overview covering 20-year development in the field of musculoskeletal ESWT. Eight historical paradigms have been identified and put under question from a current perspective: energy intensity, focus size, anesthesia, imaging, growth plates, acuteness, calcifications, and number of sessions. All paradigms as set in a historical consensus meeting in 1995 are to be revised. First, modern musculoskeletal ESWT is divided into focused and radial technology and the physical differences are about 100-fold with respect to the applied energy. Most lesions to be treated are easy to reach and clinical focusing plays a major role today. Lesion size is no longer a matter of concern. With the exception of nonunion fractures full, regional, or even local anesthesia is not helpful in musculoskeletal indications. Juvenile patients can also effectively be treated without risk of epiphyseal damage. Further research is needed to answer the question about if and which acute injuries can be managed effectively. Treatment parameters like the number of sessions are still relying on empirical data and have to be further elucidated.
# Introduction
Explosive events in nature (e.g., lightning stroke) and technics (e.g., airplanes breaking through the sound barrier) create shock waves. In principle, these acoustic waves transmit energy "from the point of generation to remote regions." The principle of this natural phenomenon has been transferred to medical application. "Shock and pressure waves are pulses, while ultrasound is a continuous oscillation". Shockwaves are generated extracorporeally (electrohydraulic, piezoelectric, or electromagnetic). The resulting energy is focused by concentrating reflectors and is noninvasively transmitted inside the body to induce therapeutic effects at a target area. So-called radial shockwaves have different physical characteristics. They are pressure waves and not real shockwaves. Different tissues possess different acoustic impedance. At the interface between these tissues, acoustic energy is released and transformed into mechanical energy.
Starting in 1980, extracorporeal shockwaves were applied transcutaneously for the first time in medicine to destroy a kidney stone in a human [bib_ref] Extracorporeally induced destruction of kidney stones by shock waves, Chaussy [/bib_ref]. Since then, several million people have benefited from this noninvasive method. As a result of the high energy applied in Extracorporeal Shock Wave Lithotripsy, much research has been performed to investigate possible side effects on tissues which are penetrated by the shockwaves on their way from the skin to the stone. By doing this, attention was paid not only to the focus zone where the highest energy is delivered but also to the surrounding area where lower energy is released. In consequence both destructive and regenerative effects were seen in bony tissues [bib_ref] Effect of high energy shock waves on bony tissue, Graff [/bib_ref]. A dose-dependent effect was detected with high energy leading to more destructive effects and lower energy leading to more regenerative effects on the treated tissue [bib_ref] Use of extracorporeal shock waves in the treatment of pseudarthrosis, tendinopathy and..., Haupt [/bib_ref] [bib_ref] Dose-related effects of shock waves on rabbit tendo Achillis. A sonographic and..., Rompe [/bib_ref]. In the early 1990s, extracorporeal shockwave effects on bone and soft tissues have led to indicating this treatment also for musculoskeletal disorders [bib_ref] Use of extracorporeal shock waves in the treatment of pseudarthrosis, tendinopathy and..., Haupt [/bib_ref]. Consequently, specific devices for musculoskeletal focused Extracorporeal Shock Wave Therapy (fESWT) were introduced into the market. These devices focus the shock waves to a point which is approximately 4-6 cm apart from the application to the skin. Compared with the urologic lithotripters which recommended immersion of the patient in a water bath, this first generation of orthopedic devices had reduced and adjustable energy release. Coupling to the patient's body was performed by ultrasound gel and aiming was realized by ultrasound. In a consensus meeting in 1995, instructions were established for the use of extracorporeal shock waves in musculoskeletal indications: (a) high energy only, (b) small "focus," (c) anesthesia, (d) imaging guided application, (e) avoiding growth plates, (f) no acute injuries, (g) soft tissue pain in the proximity to bones (insertional tendinopathy), and (h) tendinopathies with extraosseous calcification.
In the early 2000s, devices featuring ballistic pressure waves were introduced into the Extracorporeal Shock Wave Therapy (ESWT) market. These waves are produced mechanically by a compressed air driven projectile which hits the applicator. This technology is since named radial ESWT (rESWT). The respective devices are much cheaper, smaller, and easier to handle. However, the maximum rESWT energy is delivered at the applicator to skin interface and focused shock waves peak pressure is about 100 times higher while the pulse duration is 1000 times shorter. The clinical effect of rESWT could soon be demonstrated [bib_ref] Radiale stoßwellentherapie beim fersensporn (Fasciitis plantaris), Haupt [/bib_ref] and today rESWT is a widely accepted method with comparable results specifically for superficial musculoskeletal disorders [bib_ref] Clinical application of extracorporeal shock wave therapy in orthopedics: focused versus unfocused..., Foldager [/bib_ref] [bib_ref] A systematic review of shockwave therapies in soft tissue conditions: focusing on..., Speed [/bib_ref].
This review paper updates the current knowledge with respect to the historical paradigms as set in 1995.
# Materials and methods
This narrative review presents eight different ESWT paradigms which were extracted from a historical German consensus meeting held in 1995. We evaluated if these paradigms are still true after 20 years of further development of the method.
Historically, most research related to musculoskeletal ESWT literature was published in German language and in books or journals which are not referenced in Medline. Therefore, a systematic search was judged not to be a reasonable approach.
The bases for the current investigation are the authors' databases, containing both historical Medline listed papers on ESWT and also historical ESWT articles which were published in German language. The content of these articles is further reported.
For each of the eight individual paradigms, the historical background is addressed. Developments over time and recent perspectives to these topics were analysed also from the authors' literature databases.
# Results
## High energy only?
Historically, the companies provided the users with different specifications of the used energy levels, some of them used the applied energy flux density (ED), and others used the voltage (kV) led into the device to produce the shock waves. In particular, the description of the voltage is device depending and therefore a comparison between different technologies (devices from different producers) is meaningless. So the convention was made to use ED (mJ/mm 2 ) as the comparable parameter. It turned out that it is not enough to look at only one parameter. So it is no wonder that there are many conflicting publications due to the different energy descriptions [bib_ref] Destruction of gallstones and model stones by extracorporeal shock waves, Delius [/bib_ref].
Beside the well-known shock wave effect of disintegration of concrements, a stimulation of fibrous tissue could be demonstrated to occur and this different biologic mechanism was dose-dependent [bib_ref] Effect of shock waves on the healing of partial-thickness wounds in piglets, Haupt [/bib_ref].
Consequently and already in the early 1990s musculoskeletal ESWT was divided into "low" (0.08-0.23 mJ/mm 2 ) and "medium" (14-18 kV) energy applications. Not concordant with the former, a classification of low (<0.08 mJ/mm 2 ), medium (0.08-0.28 mJ/mm 2 ), and high (>0.28 mJ/mm 2 ) energy was introduced and established [bib_ref] Dose-related effects of shock waves on rabbit tendo Achillis. A sonographic and..., Rompe [/bib_ref]. Evidence was obtained from an experimental study, demonstrating that "energy flux densities of over 0.28 mJ/mm should not be used clinically in the treatment of tendon disorders" [bib_ref] Dose-related effects of shock waves on rabbit tendo Achillis. A sonographic and..., Rompe [/bib_ref]. Initially, low energy ESWT was called "pain therapy" and anesthesia was not considered a "conditio sine qua non". Early reports demonstrated promising results with low energy ESWT for soft tissue injuries like tennis elbow and plantar fasciitis. Meanwhile, soft tissue indications were equally established for low energy fESWT and also rESWT. Comparable results are published regarding plantar fasciitis [bib_ref] Clinically relevant effectiveness of focused extracorporeal shock wave therapy in the treatment..., Gollwitzer [/bib_ref] [bib_ref] Radial extracorporeal shock wave therapy is safe and effective in the treatment..., Gerdesmeyer [/bib_ref] [bib_ref] Comparison of radial versus focused extracorporeal shock waves in plantar fasciitis using..., Lohrer [/bib_ref] , Achilles [bib_ref] Results of radial shockwave therapy in patients with unsuccessfully treated tendinoses, Lohrer [/bib_ref] [bib_ref] Eccentric loading, shock-wave treatment, or a wait-and-see policy for tendinopathy of the..., Rompe [/bib_ref] , and patellar tendinopathy [bib_ref] Results of radial shockwave therapy in patients with unsuccessfully treated tendinoses, Lohrer [/bib_ref] [bib_ref] Crosssectional outcome analysis of athletes with chronic patellar tendinopathy treated surgically and..., Peers [/bib_ref] [bib_ref] No difference in effectiveness between focused and radial shockwave therapy for treating..., Van Der Worp [/bib_ref]. A recent systematic review, respectively, confirms that "there is no scientific evidence in favor of either rESWT or fESWT with respect to treatment outcome" [bib_ref] Efficacy and safety of extracorporeal shock wave therapy for orthopedic conditions: a..., Schmitz [/bib_ref].
## Small "focus"
Only? Historically, ESWT was performed with lithotripters and also the first generation of musculoskeletal ESWT devices was based on the focused technology. Respectively, maximum energy was applied to a small area 5-10 cm below the applicator and this energy was concentrated in an area with a diameter of 5-10 mm. Therefore, painful syndromes originating from a larger area were not considered as an indication for ESWT . Similarly, radiating or referred pain syndromes without a clear anatomic substrate were not regarded suitable for ESWT .
At that time, the fact that relevant energy is also measurable peripherally to the focal zone was neglected. Accepted indications were nonunion fractures, plantar fasciitis, tennis elbow, and calcific shoulder tendinopathy. The "small focus only" statement was held until the invention of the radial technology [bib_ref] Results of radial shockwave therapy in patients with unsuccessfully treated tendinoses, Lohrer [/bib_ref] , with the maximum energy delivered at the tip of the applicator. Due to the smaller sizes and lower costs of the devices, rESWT has increasingly been used all over the world. Even if the applied energy diminishes by square relative to the penetration depth, also this method has meanwhile clearly demonstrated its effectiveness for soft tissue injuries in level 1 studies [bib_ref] Radial extracorporeal shock wave therapy is safe and effective in the treatment..., Gerdesmeyer [/bib_ref] [bib_ref] Comparison of radial versus focused extracorporeal shock waves in plantar fasciitis using..., Lohrer [/bib_ref].
In a next step, rESWT was applied to treat more complex musculoskeletal symptoms associated with trigger points. The underlying mechanism of action is explained by the concept of myofascial pain [bib_ref] Trigger points-diagnosis and treatment concepts with special reference to extracorporeal shockwaves, Gleitz [/bib_ref]. Recently and inspired by traditional Chinese medicine, ESWT acupuncture has been invented [bib_ref] 10 years of experience of acupuncture with shock waves, Everke [/bib_ref].
## (local) anesthesia?
Anesthesia allows applying shockwaves with higher intensities. Derived from kidney stone and nonunion fracture experience, high energy was proposed for orthopedic ESWT indications. Consequently, in the early 1990s it was suggested to adapt anesthesia (full, regional, or local) according to the applied energy level. As a result of analgesia or anesthesia, several randomized controlled studies failed to demonstrate a significant advantage of ESWT against sham treatment [bib_ref] Extracorporeal shock wave therapy for plantar fasciitis: randomised controlled multicentre trial, Haake [/bib_ref] [bib_ref] Ultrasound-guided extracorporeal shock wave therapy for plantar fasciitis: a randomized controlled trial, Buchbinder [/bib_ref]. It was in 2005 when two randomized controlled studies revealed that local anesthesia at least reduces the effect of ESWT for plantar fasciitis [bib_ref] Repetitive low-energy shock wave application without local anesthesia is more efficient than..., Rompe [/bib_ref] [bib_ref] Influence of local anesthesia and energy level on the clinical outcome of..., Labek [/bib_ref] and this effect was only partly compensated by applying higher energy levels under local anesthesia [bib_ref] Influence of local anesthesia and energy level on the clinical outcome of..., Labek [/bib_ref]. Comparable negative local anesthesia effects were demonstrated for insertional Achilles tendinopathy [bib_ref] High-energy extracorporeal shock wave therapy as a treatment for insertional achilles tendinopathy, Furia [/bib_ref].
Nowadays, (local) anesthesia is still regarded as helpful for bone indications [bib_ref] Extracorporeal shockwave therapy (ESWT)-first choice treatment of fracture non-unions?, Schaden [/bib_ref] but is not recommended for soft tissue ESWT [bib_ref] Efficacy and safety of extracorporeal shock wave therapy for orthopedic conditions: a..., Schmitz [/bib_ref].
Meanwhile, there is evidence from experimental research that the pain producing effect of ESWT is responsible for the release of neuropeptides (like substance P) initiating both central and local trophic effects to increase metabolism in bradytrophic tissues [bib_ref] Efficacy and safety of extracorporeal shock wave therapy for orthopedic conditions: a..., Schmitz [/bib_ref] [bib_ref] Application of local anesthesia inhibits effects of low-energy Extracorporeal Shock Wave Treatment..., Klonschinski [/bib_ref]. It was experimentally demonstrated that ". . . ESWT dose-dependently activates and sensitizes primary afferent nociceptive C-fibers, and that both activation and sensitization were prevented if local anesthesia was applied" [bib_ref] Application of local anesthesia inhibits effects of low-energy Extracorporeal Shock Wave Treatment..., Klonschinski [/bib_ref].
## Imaging guided application?
At the beginning of the orthopedic shock wave era, it was generally agreed that focal degenerative lesions within the injured tissues are responsible for the painful syndromes and should be exactly targeted by ESWT. Therefore, visualizing aiming devices were demanded. Fluoroscopy was already integrated in all urologic fESWT devices which were used also for the initial years to treat orthopedic injuries. However, visualization of soft tissues was not possible. In 1995, in our center, the first fESWT device was installed to specifically treat sport orthopedic soft tissue indications. Most importantly, it was radiation free. An inline sonography system was incorporated in order to aim exactly the shock waves at the structure of interest. In 1996, this machine was available for the German team athletes during the Atlanta Olympic Games [fig_ref] Figure 1: Initiation of the ESWT technology to treat Olympic athletes during the 1996... [/fig_ref]. Even if it was not stationary and its volume, weight, and price were considerably reduced compared with the lithotripters, transportation was a logistic effort. Therefore, really small and mobile ESWT devices were requested. Again, urologists took this next step and applied the principle and the technology of an already existing device for intracorporeal ballistic lithotripsy to treat orthopedic soft tissue indications percutaneously.
That new technology produced pressure waves and not real shock waves, but the term radial shock wave was generally agreed upon and is used since [bib_ref] Clinical application of extracorporeal shock wave therapy in orthopedics: focused versus unfocused..., Foldager [/bib_ref]. Nowadays, ballistic devices have been developed with electromagnetic working mechanisms.
Users and investigators found out that aiming at the most painful area was sufficient or even superior to aiming just at an anatomically given landmark which was identified by imaging. This procedure has consequently been demonstrated to be superior and was termed "biofeedback" [bib_ref] Extracorporeal shock wave therapy in the treatment of chronic tendinopathies, Sems [/bib_ref]. One well-known example is a double-blind, randomized placebo-controlled study on ultrasound-guided fESWT for plantar fasciitis [bib_ref] Ultrasound-guided extracorporeal shock wave therapy for plantar fasciitis: a randomized controlled trial, Buchbinder [/bib_ref].
Actually, focusing by biofeedback is also the cornerstone for myofascial trigger point ESWT [bib_ref] Trigger points-diagnosis and treatment concepts with special reference to extracorporeal shockwaves, Gleitz [/bib_ref]. However, the treatment of bone lesions like nonunions and osteochondrosis dissecans still needs image guided application, for example, by fluoroscopy.
## Growth plates?
In an experimental study on proximal rat tibiae, dysplastic lesions could be identified following high energy fESWT (20 kV, 1500 shock waves) [bib_ref] Effects of shock waves on the structure and growth of the immature..., Yeaman [/bib_ref]. As a consequence from this study, ESWT was regarded to be contraindicated in a juvenile population .
Only two years later, another animal study was published demonstrating no negative histological differences comparing fESWT effect with the untreated contralateral femoral head of immature rabbits [bib_ref] Effect of lithotripsy on immature rabbit bone and kidney development, Van Arsdalen [/bib_ref]. Another experimental rabbit study was published in German language. The investigators applied 800 focused impulses (0.32 mJ/mm 2 ) which is comparable to a high fESWT in a human bone application. Obviously, these two papers were underestimated in the scientific world [bib_ref] Effects of high-energy shock waves on the structure of the immature epiphysis-a..., Nassenstein [/bib_ref]. For rESWT, a recently published rat experiment could detect "no negative effects" when 1500 or 3000 impulses of 4 bar were applied to the immature rat knees [bib_ref] The long-term effects of extracorporeal shock waves on the epiphysis of the..., Oztemur [/bib_ref].
Even if initially mentioned anecdotally already in 1995it was only recently when the first clinical case series reported both safety and effectiveness when Osgood Schlatter or Sever's diseases were treated by using rESWT [bib_ref] Extracorporeal shock wave therapy for patients suffering from recalcitrant osgood-schlatter disease, Lohrer [/bib_ref] [bib_ref] Radial shock wave therapy for patients with apophysitis calcanei, Lohrer [/bib_ref].
3.6. Acute Injuries? When introducing musculoskeletal ESWT, it was declared to be indicated for chronic injuries. The reason for this was that in general a new treatment modality should provide evidence before being spread out to the public, and, as long as the evidence is missing, it should be recommended only for patients, who already have been treated by other options. This means that three months of conservative treatment should have been performed without success before ESWT is indicated as an alternative to operative treatment . Extensive technical, manpower, and time requirements have been advocated as rationales for this limitation. Additionally, economic factors limited the musculoskeletal ESWT application. Consequently, most research was traditionally made for conservatively pretreated injuries with a history of more than three months. International shock wave societies still consider only nonacute pathologies (http://www.digest-ev.de/leitlinien/). With the advent of cheaper and more flexible ESWT devices, this rule has been broken. For instance, in acute and operatively treated long bone fractures high energy fESWT effectively reduced the number of nonunions [bib_ref] The effects of extracorporeal shockwave on acute high-energy long bone fractures of..., Wang [/bib_ref]. Contrary to this, in a randomized controlled study rESWT treatment was inferior to stretching for plantar fasciitis patients when patients were not pretreated and complained about symptoms under six weeks [bib_ref] Plantar fascia-specific stretching versus radial shock-wave therapy as initial treatment of plantar..., Rompe [/bib_ref].
If ESWT can be relevant to effectively treat acute muscular or tendon strains is currently not known and respective research is needed.
## Tendinopathies with extraosseous calcification.
Historically, only mechanical (and not biologic) ESWT effects were regarded as relevant in medicine. At the transmission through tissues with similar acoustic properties (soft tissue) a minor amount of energy is released. It was assumed that the resulting mechanical effect is negligible. In contrast, high acoustic impedance differences exist between cortical bone (6.12 × 10 6 kg/m 2 s) and soft tissue (e.g., muscle = 1.66 × 10 6 kg/m 2 s). ESWT consequently releases a large amount of mechanical energy at the interface. This concept was the rationale not only to treat kidney stones but also to treat soft tissue calcifications . Initially, a real destruction of bone was not detected as a result from ESWT , but later experimental research demonstrated a dose-dependent induction of cortical fractures and periosteal detachment [bib_ref] New bone formation by extracorporeal shock waves. Dependence of induction on energy..., Maier [/bib_ref]. Relevant acoustic impedance differences exist also at the interface between tendon and bone. Therefore, well-defined insertional tendinopathies like tennis elbow and plantar fasciitis were thought to be also eligible for ESWT treatment specifically when combined with a spur.
These treatment principles were held until the invention of the rESWT with a completely different technology. Historically, the main differences between fESWT and rESWT are as follows: (a) principle of generation = pneumatic rESWT versus electrohydraulic, piezoelectric, or electromagnetic fESWT, (b) wavelength = 0.15 to 1.5 m (rESWT) versus 1.5 mm (fESWT), and (c) maximum pressure = 1 (rESWT) versus 10-100 (fESWT) MPa and penetration depth = 2-5 cm (rESWT) versus 5-20 cm (fESWT). Nowadays, there are also ballistic devices available with electromagnetic working mechanisms accelerating the projectile to hit the applicator. Clinically most important thing is that the maximum energy in rESWT is delivered at the interface between the applicator head of the device and the skin and diminishes its energy inside the treated tissue by the square of the penetration depth.
As a result, rESWT was applied to tendon lesions, featured by their immediately subcutaneous localization and by a large area of injured tissue. Midportion Achilles tendinopathy and patellar tendinopathy fulfil these criteria and have been demonstrated to be an indication for rESWT [bib_ref] Results of radial shockwave therapy in patients with unsuccessfully treated tendinoses, Lohrer [/bib_ref] [bib_ref] Eccentric loading versus eccentric loading plus shock-wave treatment for midportion achilles tendinopathy:..., Rompe [/bib_ref] [bib_ref] The effectiveness of extracorporeal shock wave therapy in lower limb tendinopathy: a..., Mani-Babu [/bib_ref]. Based on current evidence, we are unable to prefer fESWT or rESWT for musculoskeletal soft tissue injuries [bib_ref] Clinical application of extracorporeal shock wave therapy in orthopedics: focused versus unfocused..., Foldager [/bib_ref] [bib_ref] No difference in effectiveness between focused and radial shockwave therapy for treating..., Van Der Worp [/bib_ref]. Conflicting evidence exists from the results of two studies that directly compared fESWT to rESWT in plantar fasciitis and patellar tendinopathy patients [bib_ref] Comparison of radial versus focused extracorporeal shock waves in plantar fasciitis using..., Lohrer [/bib_ref] [bib_ref] No difference in effectiveness between focused and radial shockwave therapy for treating..., Van Der Worp [/bib_ref]. FESWT revealed moderately superior results compared to rESWT in plantar fasciitis, while no difference was demonstrated between the two applications regarding patellar tendinopathy [bib_ref] Comparison of radial versus focused extracorporeal shock waves in plantar fasciitis using..., Lohrer [/bib_ref] [bib_ref] No difference in effectiveness between focused and radial shockwave therapy for treating..., Van Der Worp [/bib_ref].
## Three sessions only?
The number of required treatment sessions is a relevant parameter in principle. Recently, systematic research recommends "three treatment sessions at 1-week intervals, with 2000 impulses per session and the highest energy flux density the patient can tolerate" [bib_ref] Efficacy and safety of extracorporeal shock wave therapy for orthopedic conditions: a..., Schmitz [/bib_ref]. However, historical reports do not adequately address that detail [bib_ref] Use of extracorporeal shock waves in the treatment of pseudarthrosis, tendinopathy and..., Haupt [/bib_ref]. Analogue to and derived from the lithotripsy nonunion fractures have been treated with high energy predominantly in one session. The reason for this procedure is most probably based upon the intensive effort required by anesthesia and fluoroscopy. For the soft tissue conditions a wide range (1 to 10) of treatment sessions was initially instructed [bib_ref] Use of extracorporeal shock waves in the treatment of pseudarthrosis, tendinopathy and..., Haupt [/bib_ref]. The need of standardization of treatment regimens in randomized controlled trials established one to three ESWT sessions at weekly intervals as a standard clinical practice regardless of the underlying pathology [bib_ref] A systematic review of shockwave therapies in soft tissue conditions: focusing on..., Speed [/bib_ref] [bib_ref] Clinically relevant effectiveness of focused extracorporeal shock wave therapy in the treatment..., Gollwitzer [/bib_ref] [bib_ref] Radial extracorporeal shock wave therapy is safe and effective in the treatment..., Gerdesmeyer [/bib_ref] [bib_ref] No difference in effectiveness between focused and radial shockwave therapy for treating..., Van Der Worp [/bib_ref] [bib_ref] Eccentric loading versus eccentric loading plus shock-wave treatment for midportion achilles tendinopathy:..., Rompe [/bib_ref] [bib_ref] A single application of low-energy radial extracorporeal shock wave therapy is effective..., Furia [/bib_ref] [bib_ref] Eccentric loading compared with shock wave treatment for chronic insertional achilles tendinopathy...., Rompe [/bib_ref] [bib_ref] Lowenergy extracorporeal shock wave therapy as a treatment for medial tibial stress..., Rompe [/bib_ref] [bib_ref] SWAAT study: extracorporeal shock wave therapy and arginine supplementation and other nutraceuticals..., Notarnicola [/bib_ref] [bib_ref] Extracorporeal shockwave for chronic patellar tendinopathy, Wang [/bib_ref] [bib_ref] Patient guided Piezoelectric extracorporeal shockwave therapy as treatment for chronic severe patellar..., Zwerver [/bib_ref].
Recently, there have been a few reports which retrospectively addressed the number of rESWT sessions needed to treat soft tissue pathologies such as trigger digits, symptomatic calcified shoulder tendinopathy, and plantar fasciitis. These studies revealed that pretreatment symptom duration was significantly correlated with the number of rESWT sessions applied [bib_ref] The recurrence rate of plantar fasciitis after Radial Extracorporeal Shock Wave Therapy..., Malliaropoulos [/bib_ref]. Additionally, there is evidence that there is a dose-related ESWT effect with lower energy flux densities [mJ/mm 2 ] requiring more treatment sessions to obtain the same effect [bib_ref] Dose-related effect of extracorporeal shock wave therapy for plantar fasciitis, Lee [/bib_ref].
Discussion is still going on about which parameters or which combination of parameters should be used to maximize the effect of ESWT treatment for a specific indication. In this context, it has to be mentioned that comparability of studies should not be reduced on one single parameter (e.g., energy flux density).
In clinical practice, ESWT is rarely used as a monotherapy. Strategic loading and/or exercises are usually prescribed in addition to shock waves, a fact that in general RCTs have not adequately addressed. An individualized intervention should be considered depending initially on the type and characteristics of the pathology [bib_ref] Is tendon pathology a continuum? A pathology model to explain the clinical..., Cook [/bib_ref].
# Discussion
The most important finding of this review is that all historical paradigms as set for musculoskeletal ESWT in 1995 did not withstand the technical and clinical developments over the last 20 years. The initial phase of the musculoskeletal ESWT was driven by side effect research in context with lithotripsy investigation and the first orthopedic applications have been performed by urologists [bib_ref] Use of extracorporeal shock waves in the treatment of pseudarthrosis, tendinopathy and..., Haupt [/bib_ref]. Principles which were already known from more than two million lithotripsies in men and from respective animal studies were transferred and adapted to musculoskeletal indications.
At the beginning of the musculoskeletal shock wave age it was thought that the higher the energy is, the better the outcome would be. For soft tissue pathologies it was early realized that lower ESWT intensities are able to induce tissue regeneration instead of necrotic reactions [bib_ref] Dose-related effects of shock waves on rabbit tendo Achillis. A sonographic and..., Rompe [/bib_ref]. The pain resulting from the ESWT is clearly depending on the energy intensity [bib_ref] Application of local anesthesia inhibits effects of low-energy Extracorporeal Shock Wave Treatment..., Klonschinski [/bib_ref] but clinical focusing was shown to improve the treatment results especially when performed without local anesthesia [bib_ref] Repetitive low-energy shock wave application without local anesthesia is more efficient than..., Rompe [/bib_ref] [bib_ref] Application of local anesthesia inhibits effects of low-energy Extracorporeal Shock Wave Treatment..., Klonschinski [/bib_ref]. Specific ESWT devices for musculoskeletal conditions were produced. Further reduction of the applied energy was delivered with the rESWT technology. So and over the years, devices became much more flexible/mobile and had reduced volume, weight, and costs.
There are an increasing number of high quality ESWT studies for musculoskeletal conditions published in the literature. It can be summarized without exaggeration that ESWT is the best analyzed treatment modality in the orthopedic field. This statement includes also operative interventions. A recent systematic musculoskeletal ESWT review concludes that there is more need for high level studies [bib_ref] A systematic review of shockwave therapies in soft tissue conditions: focusing on..., Speed [/bib_ref]. But the question to be answered in future is not if ESWT works but rather which treatment protocol and parameters are the best for specific and well described conditions [bib_ref] The effectiveness of extracorporeal shock wave therapy in lower limb tendinopathy: a..., Mani-Babu [/bib_ref]. Research finally has to follow clinical practice, where treatment protocols are individualized.
Until now, clinical ESWT research is aiming exclusively at detecting the success of ESWT applied following a standardized protocol. The question, however, if ESWT is similarly effective in each stage of a given musculoskeletal indication is completely unanswered up to date. For instance, a "tendon pathology continuum model" has been described [bib_ref] Is tendon pathology a continuum? A pathology model to explain the clinical..., Cook [/bib_ref]. Derived from this, tendinopathy is "no longer a 'one size fits all' diagnosis" [bib_ref] Tendinopathy: no longer a 'one size fits all' diagnosis, Cook [/bib_ref]. It is to expect that different stages of a given pathology will respond differently to ESWT. Moreover, monotherapies are rarely used in clinical practice. Given the former, future randomized controlled work should focus on assessing and comparing more realistic treatment protocols.
# Conclusion
With the exception of bone related conditions, modern musculoskeletal ESWT is performed with energy below 0.28 mJ/mm 2 and without anesthesia. The size of the tissue area to be treated can be small or large. "Biofeedback" is superior to imaging guided focusing. ESWT application in apophyseal osteochondral lesions in patients with open growth plates seems to be promising and safe. ESWT protocols should be adapted to the stage and chronicity of the treated pathology.
[fig] Figure 1: Initiation of the ESWT technology to treat Olympic athletes during the 1996 Olympic Games in Atlanta. [/fig]
|
Validation of the ITS2 Region as a Novel DNA Barcode for Identifying Medicinal Plant Species
Background: The plant working group of the Consortium for the Barcode of Life recommended the two-locus combination of rbcL + matK as the plant barcode, yet the combination was shown to successfully discriminate among 907 samples from 550 species at the species level with a probability of 72%. The group admits that the two-locus barcode is far from perfect due to the low identification rate, and the search is not over.Methodology/Principal Findings: Here, we compared seven candidate DNA barcodes (psbA-trnH, matK, rbcL, rpoC1, ycf5, ITS2, and ITS) from medicinal plant species. Our ranking criteria included PCR amplification efficiency, differential intra-and inter-specific divergences, and the DNA barcoding gap. Our data suggest that the second internal transcribed spacer (ITS2) of nuclear ribosomal DNA represents the most suitable region for DNA barcoding applications. Furthermore, we tested the discrimination ability of ITS2 in more than 6600 plant samples belonging to 4800 species from 753 distinct genera and found that the rate of successful identification with the ITS2 was 92.7% at the species level.Conclusions: The ITS2 region can be potentially used as a standard DNA barcode to identify medicinal plants and their closely related species. We also propose that ITS2 can serve as a novel universal barcode for the identification of a broader range of plant taxa.
# Introduction
The World Health Organization estimates that 80 percent of the world's population utilizes traditional medicines for healing and curing diseases (http://www.worldwildlife.org/what/globalmarkets/ wildlifetrade/faqs-medicinalplant.html). There is an increasing international market for medicinal plants, which are used both for herbal medicine and for pharmaceutical products. Medicinal plants cover a wide range of plant taxa and closely related species. According to surveys in China, medicinal plants belong to 11,146 species from 2,309 genera of 383 families, representing a rich biodiversity. Accurate and rapid authentication of these plants and their adulterants is difficult to achieve at the scale of international trade in medicinal plants. We aim to provide a practical and powerful tool for identifying medicinal plants and their adulterants in trade and for ensuring safety in their use.
The term ''DNA barcode'' for global species identification was first coined by Hebert in 2003 [bib_ref] Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related..., Hebert [/bib_ref] [bib_ref] Biological identifications through DNA barcodes, Hebert [/bib_ref] and has gained worldwide attention in the scientific community [bib_ref] Molecular systematics: counting angels with DNA, Blaxter [/bib_ref] [bib_ref] DNA barcoding does not compete with taxonomy, Gregory [/bib_ref] [bib_ref] DNA barcoding a useful tool for taxonomists, Schindel [/bib_ref] [bib_ref] DNA barcoding and the renaissance of taxonomy, Miller [/bib_ref] [bib_ref] Will DNA bar codes breathe life into classification?, Marshall [/bib_ref]. Recognition of animals, plants and fungi has been performed using this technique [bib_ref] DNA barcodes distinguish species of tropical Lepidoptera, Hajibabaei [/bib_ref] [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] Prospects for fungus identification using CO1 DNA barcodes, with Penicillium as a..., Seifert [/bib_ref] [bib_ref] Assessing the effect of varying sequence length on DNA barcoding of fungi, Min [/bib_ref] [bib_ref] Testing the reliability of genetic methods of species identification via simulation, Ross [/bib_ref]. Most researchers agree that the mitochondrial gene encoding cytochrome c oxidase subunit 1 is a favorable region for use as a DNA barcode in most animal species and even in some fungal species, including those of the groups Ascomycota, Basidiomycota and Chytridiomycota. However, the CO1 gene and other mitochondrial genes from plants have limited usefulness for identifying plant species across a wide range of taxa due to the low amounts of variation in the genes, as well as the variable structure of the mitochondrial genome [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] Wanted: a barcode for plants, Pennisi [/bib_ref] [bib_ref] Land plants and DNA barcodes: short-term and long-term goals, Chase [/bib_ref] [bib_ref] Multiple multilocus DNA barcodes from the plastid genome discriminate plant species equally..., Fazekas [/bib_ref] [bib_ref] A proposal for a standardised protocol to barcode all land plants, Chase [/bib_ref]. Thus, screening for single or multiple regions appropriate for DNA barcoding studies in nuclear and plastid genomes in plants has been an important research focus [fig_ref] Figure 1: Genes from three genomes in plants that are candidate barcodes [/fig_ref].
First, most single-copy genes in the nuclear genome, as well as their introns, have been excluded as barcode candidates because of the lack of universal primers for their amplification [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref]. However, with the exception of 5.8S, the internal transcribed spacer (ITS) of nuclear ribosomal DNA and regions of the ITS could be potential barcodes [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] Land plants and DNA barcodes: short-term and long-term goals, Chase [/bib_ref] [fig_ref] Figure 1: Genes from three genomes in plants that are candidate barcodes [/fig_ref]. Second, extensive studies have focused on genes and introns of the chloroplast genome. For example, Kress et al. [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] compared 10 loci for authenticating closely related species in 7 plant families and 99 species belonging to 88 genera in 53 families, and they reported that the psbA-trnH spacer and the internal transcribed spacer could be used as a pair of potential barcodes for identifying widely divergent angiosperm taxa. However, this pair failed to distinguish members of the order Cycadales [bib_ref] DNA barcoding in the Cycadales: testing the potential of proposed barcoding markers..., Sass [/bib_ref]. Lahaye et al. [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref] analyzed 1084 plant species (nearly 96% orchid species) and demonstrated that a portion of the plastid matK gene could be a universal DNA barcode for flowering plants. In another study, two separate regions (matK and psbA-trnH) had significant sequence variations and correctly discriminated 95% of 40 nutmeg samples representing 10 species in the Myristicaceae family [bib_ref] Testing candidate plant barcode regions in the Myristicaceae, Newmaster [/bib_ref]. Newmaster et al. [bib_ref] DNA barcoding in land plants: evaluation of rbcL in a multigene tiered..., Newmaster [/bib_ref] analyzed over 10,000 rbcL sequences from GenBank and found that rbcL could discriminate samples in approximately 85% of pairwise comparisons of congeneric species, whereas the discrimination efficiency was about 88% when a combination of psbA-trnH and rbcL was used across 96 diverse species of 48 genera from 43 families [bib_ref] A two-locus global DNA barcode for land plants: the coding rbcL gene..., Kress [/bib_ref]. Altogether, although no single plant marker has been found that works as well as the COI in animals, several markers of the plastid genome, such as psbA-trnH, matK, rbcL and rpoC1, have shown superior qualities as DNA barcodes relative to accD, ndhJ and rpoB [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] Wanted: a barcode for plants, Pennisi [/bib_ref] [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref] [bib_ref] DNA barcoding plants in biodiversity hot spots: progress and outstanding questions, Hollingsworth [/bib_ref] [bib_ref] Selecting barcoding loci for plants: evaluation of seven candidate loci with species-level..., Hollingsworth [/bib_ref] [fig_ref] Figure 1: Genes from three genomes in plants that are candidate barcodes [/fig_ref].
There are two groups of potential users of DNA barcodes: plant taxonomists/systematists and scientists in other fields [bib_ref] Land plants and DNA barcodes: short-term and long-term goals, Chase [/bib_ref] [bib_ref] DNA barcodes: genes, genomics, and bioinformatics, Kress [/bib_ref]. In addition, DNA barcodes will be a useful and powerful tool for nonprofessional users such as customs officers, traditional drug producers and managers and forensic specialists. Therefore, a rapid and simple DNA barcoding identification system, even an imperfect one, is likely to be welcomed.
In this study, we tested seven potential DNA regions (psbA-trnH, matK, rbcL, rpoC1, ycf5, ITS2 and ITS) for their suitability as DNA barcodes across 8557 medicinal plants and closely related samples belonging to 5905 species from 1010 diverse genera of 219 families in 7 phyla (Angiosperms, Gymnosperms, Ferns, Mosses, Liver-worts, Algae and Fungi). These plants have a long history of use in traditional herbal medicines; included here are species from the Chinese and the Japanese Pharmacopoeias, as well as a few from the European Pharmacopoeia. The seven candidate DNA barcodes were compared using several criteria. Four of the loci (rbcL, rpoC1, matK and ycf5) were proposed by the Plant Working Group (www.kew.org/barcoding). As for the other three, psbA-trnH and nrITS were recommended by Kress et al. [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] and ITS2 was first described by Chiou et al. [bib_ref] Authentication of medicinal herbs using PCR-amplified ITS2 with specific primers, Chiou [/bib_ref]. The ITS2 region was selected as a barcode candidate because ITS2 sequences are potential general phylogenetic markers and are widely used for phylogenetic reconstructions at both the genus and species levels [bib_ref] A common core of secondary structure of the internal transcribed spacer 2..., Schultz [/bib_ref] [bib_ref] ITS2 is a double-edged tool for eukaryote evolutionary comparisons, Coleman [/bib_ref] [bib_ref] Analysis of the internal transcribed spacer 2 (ITS2) region of Scuticociliates and..., Miao [/bib_ref] [bib_ref] Pan-eukaryote ITS2 homologies revealed by RNA secondary structure, Coleman [/bib_ref] [bib_ref] ITS2 sequence-structure analysis in phylogenetics: a how-to manual for molecular systematics, Schultz [/bib_ref]. The search for and development of herbal medicines is rapidly increasing worldwide, so practical and accurate authentication resources are urgently needed [bib_ref] Genome-based approaches to the authentication of medicinal plants, Sucher [/bib_ref] [bib_ref] Identification of Dendrobium species by a candidate DNA barcode sequence: the chloroplast..., Yao [/bib_ref] [bib_ref] Authentication of the family Polygonaceae in Chinese pharmacopoeia by DNA barcoding technique, Song [/bib_ref] [bib_ref] Traditional remedies and food supplements: a 5-year toxicological study (1991-1995), Shaw [/bib_ref]. Our study shows the potential for a DNA barcoding technique to become a standard for the authentication of medicinal plants and their adulterants.
# Results
## Efficiency of pcr amplification
The success rate of PCR amplification with four pairs of primers for ITS1 was poor in our pilot study, so ITS1 was not included in subsequent experiments. Two pairs of ITS2 primers designed by Chiou et al. [bib_ref] Authentication of medicinal herbs using PCR-amplified ITS2 with specific primers, Chiou [/bib_ref] and one designed for this study failed to amplify the sequences in gymnosperms and ferns. The primer pairs with the highest success rate for each barcode are listed in [fig_ref] Table 1: Efficiency of PCR amplification of three potential barcodes in a wide range... [/fig_ref]. These rates were obtained in our pilot study. For example, we compared the PCR amplification efficiency of psbA-trnH, ITS2 and ITS sequences across 400 samples belonging to 326 species in 98 families including dicots, monocots, gymnosperms and ferns. The success rates for psbA-trnH and ITS2 sequences were 92.8% and 93.8%, respectively, while ITS fragments were only successfully amplified in 42.3% of the experiments [fig_ref] Table 1: Efficiency of PCR amplification of three potential barcodes in a wide range... [/fig_ref]. Furthermore, we also calculated the efficiency of PCR amplification in total number of samples [fig_ref] Table 2: Analysis of inter-specific divergence between congeneric species and intra-specific variation of candidate... [/fig_ref]. Results showed that rpoC1 provided the highest rate, followed by psbA-trnH and ITS2.
## Determination of genetic divergence using six parameters
First, three parameters were used to characterize inter-specific divergence [bib_ref] DNA barcoding: error rates based on comprehensive sampling, Meyer [/bib_ref] [bib_ref] The use of mean instead of smallest interspecific distances exaggerates the size..., Meier [/bib_ref] : (i) average inter-specific distance (K2P distance) between all species in each genus with at least two species; (ii) average theta prime (h9), where theta prime is the mean pairwise distance within each genus with more than one species, thus eliminating biases associated with different numbers of species among genera; and (iii) smallest inter-specific distance, i.e., the minimum inter-specific distance within each genus with at least two species. A favorable barcode should possess high inter-specific divergence in order to distinguish different species. In comparisons of inter-specific genetic distances among congeneric species using six candidate barcodes, the ITS2 region exhibited the highest inter-specific divergence according to all three parameters, followed by psbA-trnH [fig_ref] Figure 2: Inter-specific divergence between congeneric species and intra-specific variation of paired loci [/fig_ref] and [fig_ref] Table 2: Analysis of inter-specific divergence between congeneric species and intra-specific variation of candidate... [/fig_ref] , while rpoC1 provided the lowest. Moreover, Wilcoxon signed rank tests confirmed that ITS2 and psbA-trnH provided the highest inter-specific divergence between congeneric species, whereas the lowest belonged to rpoC1 [fig_ref] Table 3: Identification efficiency for ITS2 and psbA-trnH loci using different methods of species... [/fig_ref].
Second, three additional parameters were used to determine intra-specific variation [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref] [bib_ref] DNA barcoding: error rates based on comprehensive sampling, Meyer [/bib_ref] : (i) average intra-specific difference (K2P distance), that between all samples collected within each species with more than one individual; (ii) theta (h), where theta is the mean pairwise distance within each species with at least two representatives; h eliminates biases associated with unequal sampling among a species; and (iii) average coalescent depth, which is the maximum intra-specific distance within each species with at least two individuals. Here, rbcL showed the lowest level of intra-specific variation with all three parameters, while ITS2 still exhibited the highest level of variation with all three parameters, followed by psbA-trnH [fig_ref] Figure 2: Inter-specific divergence between congeneric species and intra-specific variation of paired loci [/fig_ref] and [fig_ref] Table 2: Analysis of inter-specific divergence between congeneric species and intra-specific variation of candidate... [/fig_ref]. For intra-specific divergence, Wilcoxon signed rank tests indicated that rbcL, rpoC1 and matK showed the lowest variation between conspecific individuals, whereas ITS2 showed the highest [fig_ref] Table S4: Wilcoxon signed rank tests for intra-specific variation [/fig_ref].
Similarly, the candidate DNA barcodes ITS2 and psbA-trnH were found to have high inter-specific divergence and high intraspecific variation using the six parameters and statistical tests described above. This analysis demonstrated that ITS2 and psbA-trnH sequences represent the most suitable DNA barcodes to meet our goal. Their further evaluation was then assessed using two other criteria: DNA barcoding gap and authentication ability.
## Assessment of barcoding gap
In an ideal situation, genetic variation of a DNA barcode should demonstrate separate, non-overlapping distributions between intra-and inter-specific samples. Meyer et al. and Moritz et al. [bib_ref] DNA barcoding: error rates based on comprehensive sampling, Meyer [/bib_ref] [bib_ref] DNA barcoding: promise and pitfalls, Moritz [/bib_ref] demonstrated that when the number of closely related species is increased, the overlap of genetic variation without barcoding gaps significantly increases. Our results demonstrated that the distributions of intra-and inter-specific variation of psbA-trnH and ITS2 exhibited distinct gaps, but when intra-specific variation between conspecific individuals and inter-specific divergence between all hetero-specifics were calculated using matK, rbcL, ycf5 and rpoC1, there was significant overlap without gaps [fig_ref] Figure 1: Genes from three genomes in plants that are candidate barcodes [/fig_ref]. However, when intra-specific variation between conspecific individuals and inter-specific divergence between congeneric species were computed, i.e., the proportion of closely related species was enhanced, none of the barcodes revealed large gaps [fig_ref] Figure 2: Inter-specific divergence between congeneric species and intra-specific variation of paired loci [/fig_ref]. Furthermore, Wilcoxon's two-sample tests showed that, for six barcodes (psbA-trnH, ITS2, matK, ycf5, rbcL and rpoC1), the mean of the inter-specific divergences was significantly higher than that of the corresponding intra-specific variations . Therefore, psbA-trnH and ITS2 pass this test, as they possess intra-and inter-specific variation gaps.
## Evaluation of species authentication capability of barcodes
To estimate the reliability of species identification using a DNA barcoding technique, two methods (BLAST1 and the nearest genetic distance) were used [bib_ref] Testing the reliability of genetic methods of species identification via simulation, Ross [/bib_ref]. The BLAST1 method determines the identity of a sample based on the best hit of the query sequence and the E-value for the match must be less than a cutoff value. In comparison, the nearest genetic distance method determines the identity of a sample based on which sequence in the database has the smallest genetic distance from the query sequence; this distance must be less than a distance threshold. We first retrieved all ITS2 and psbA-trnH sequences and constructed a reference sequence library. We then searched the database with the sequences generated in this study from samples with proven taxonomic identity. The species identities of these query sequences were then determined using the BLAST1 or nearest genetic distance method.
The results indicated that, using the BLAST1 method, ITS2 correctly identified 92.7% and 99.8% of the samples at the species and genus level, respectively. When the nearest genetic distance method was used, ITS2 correctly identified 90.3% and 99.7% of the samples at the species and genus level, respectively [fig_ref] Table 3: Identification efficiency for ITS2 and psbA-trnH loci using different methods of species... [/fig_ref]. In contrast to ITS2, the success rates of psbA-trnH were much lower (67.6% and 72.8%) at the species level using the BLAST1 and nearest genetic distance methods, respectively. However, psbA-trnH had a .95% success rate of identification at the genus level [fig_ref] Table 3: Identification efficiency for ITS2 and psbA-trnH loci using different methods of species... [/fig_ref]. Our data demonstrate that, when samples from a wide range of plant taxa are tested, ITS2 possesses a higher success rate in species and genus identification compared to psbA-trnH. Therefore, the ITS2 region is a powerful universal barcode, and as such, it is a promising candidate for authenticating all major plant taxa used in traditional herbal medicine. The psbA-trnH spacer may be used as a complementary barcode.
# Discussion
In contrast to other studies, our study presents a strong case for the ITS2 region being the most promising universal DNA barcode for authenticating medicinal plants, as assessed against several criteria. First, at 160-320 base pairs, the ITS2 region is short. As a result, ITS2 sequences are relatively easy to amplify using one pair of universal primers selected by our group [fig_ref] Table 1: Efficiency of PCR amplification of three potential barcodes in a wide range... [/fig_ref]. Second, determination of genetic divergence using six parameters and statistical tests confirmed that the ITS2 region possesses high interspecific divergence [fig_ref] Figure 2: Inter-specific divergence between congeneric species and intra-specific variation of paired loci [/fig_ref] , S3) and is well separated. Analyses of the DNA barcoding gap and Wilcoxon two-sample tests support the notion that the mean inter-specific divergence of the ITS2 region is significantly higher than its mean intra-specific variation , S1-S3). Third, according to the BLAST1 method, for 6685 samples from 4800 species in 753 genera of 193 families, identification accuracies using the ITS2 region were 92.7% and 99.8% at the species and genus level, respectively. The plant samples represented lower and higher plants (Angiosperms, Gymnosperms, Ferns, Mosses, Liverworts, Algae and Fungi), along with a series of closely related species. To our knowledge, this is by far one of most comprehensive samples of plants reported in the literature. The inclusion of many closely related species supports the notion that the ITS2 region is not only capable of discriminating plant taxa from different plant families but is also able to distinguish closely related taxa at the genus and species levels. This finding suggests that, similar to CO1 in animals, the ITS2 region in plants is a suitable DNA barcode for authenticating taxa at different taxonomic levels.
Although the ITS2 region possesses many advantages compared to plastid genomic fragments and other nuclear genomic regions, including ITS, other researchers have not given sufficient attention to this region. Previous studies have suggested that ITS1 and ITS exhibit higher inter-specific divergence relative to ITS2 and psbA-trnH, which were used for further testing [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] A two-locus global DNA barcode for land plants: the coding rbcL gene..., Kress [/bib_ref]. However, universal primers for ITS1 and ITS have not been identified for broad taxonomic use, leading to low amplification efficiency and the need for specific PCR conditions and additives [bib_ref] A proposal for a standardised protocol to barcode all land plants, Chase [/bib_ref] [bib_ref] A two-locus global DNA barcode for land plants: the coding rbcL gene..., Kress [/bib_ref]. Our results confirmed these previous observations. Nevertheless, the potential of ITS2 as a suitable marker applicable for taxonomic classification and phylogenetic reconstructions has already been demonstrated using Eukaryota [bib_ref] A common core of secondary structure of the internal transcribed spacer 2..., Schultz [/bib_ref] [bib_ref] ITS2 is a double-edged tool for eukaryote evolutionary comparisons, Coleman [/bib_ref] [bib_ref] Analysis of the internal transcribed spacer 2 (ITS2) region of Scuticociliates and..., Miao [/bib_ref] [bib_ref] Pan-eukaryote ITS2 homologies revealed by RNA secondary structure, Coleman [/bib_ref] [bib_ref] DNA barcoding: how it complements taxonomy, molecular phylogenetics and population genetics, Hajibabaei [/bib_ref] [bib_ref] Distinguishing species, Müller [/bib_ref]. This finding contributed to the discovery of a conserved core of the secondary structure of ITS2 in green and brown algae, land plants and most animals [bib_ref] ITS2 is a double-edged tool for eukaryote evolutionary comparisons, Coleman [/bib_ref]. Coleman argued that ITS2 has many advantages: a size of a few hundred nucleotides, comparison of relationships from the subspecies to the order levels, double-checking possible sequence errors in alignments directed by secondary structure, etc. Based on this evidence and our own findings, we propose that ITS2 should be a gold standard barcode for identifying plants and fungi [bib_ref] Distinguishing species, Müller [/bib_ref]. At the recent Barcode Conference in Mexico City, it was reported that a significant portion of the ITS2 GenBank records from plants are likely to represent fungal sequences from endophytes. We checked the plant ITS2 sequences in our experiments using BLAST analysis (e-value,0.001) and Hidden-Markov-Model (HMM)-based ITS2 annotation methods (fungal model, e-value,0.001) and did not find any fungal sequences. Further, we also checked 6022 plant ITS2 sequences from GenBank used for our analysis. Indeed, 5 plant ITS2 sequences (Accession Numbers: AM920396, AM920397, AM920401, AM920402, and AM920403) may represent fungal sequences. Although the overall ratio is less than one in one thousand, it is very important for researchers to verify the sequences from GenBank.
The present study also evaluated a chloroplast non-coding region, psbA-trnH, and compared to the ITS2 region, it also demonstrated excellent reliability for species authentication. For 2108 plant samples representing 1433 species of 551 genera in 135 families from 4 phyla (Angiosperms, Gymnosperms, Ferns and . Distribution of inter-specific divergence between congeneric species and intra-specific variation for paired loci. Two color bars in each box represent inter-specific (above) and intra-specific (below) genetic distances. doi:10.1371/journal.pone.0008613.g003
Mosses), the identification rate of the psbA-trnH region was 96.5% at the genus level using the nearest distance method; however, this rate was lower, 72.8%, at the species level. In previous studies, most researchers accepted psbA-trnH as a potential plant barcode [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] Multiple multilocus DNA barcodes from the plastid genome discriminate plant species equally..., Fazekas [/bib_ref] [bib_ref] A proposal for a standardised protocol to barcode all land plants, Chase [/bib_ref] [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref] [bib_ref] Testing candidate plant barcode regions in the Myristicaceae, Newmaster [/bib_ref] [bib_ref] A two-locus global DNA barcode for land plants: the coding rbcL gene..., Kress [/bib_ref] [bib_ref] Authentication of the family Polygonaceae in Chinese pharmacopoeia by DNA barcoding technique, Song [/bib_ref]. Similar to our findings [fig_ref] Table 2: Analysis of inter-specific divergence between congeneric species and intra-specific variation of candidate... [/fig_ref] , the inter-specific divergence of the psbA-trnH locus is higher than that of other plastid loci investigated [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] A two-locus global DNA barcode for land plants: the coding rbcL gene..., Kress [/bib_ref] , even though the matK locus only demonstrated half of the inter-specific divergence of the psbA-trnH locus [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref]. Therefore, we strongly recommend psbA-trnH as a complementary barcode to ITS2 for a broad series of plant taxa.
Comparing to ITS2 and psbA-trnH, ITS was rejected as a universal barcode due to the low PCR efficiency [fig_ref] Table 1: Efficiency of PCR amplification of three potential barcodes in a wide range... [/fig_ref] , while rpoC1 showed the lowest inter-specific divergence [fig_ref] Table 2: Analysis of inter-specific divergence between congeneric species and intra-specific variation of candidate... [/fig_ref] , S3, [fig_ref] Figure 2: Inter-specific divergence between congeneric species and intra-specific variation of paired loci [/fig_ref] , thus not all samples was amplified for rpoC1 despite the highest PCR efficiency [fig_ref] Table 2: Analysis of inter-specific divergence between congeneric species and intra-specific variation of candidate... [/fig_ref]. In our experiments, ITS2, psbA-trnH, rpoC1, and rbcL provided not bad PCR efficiency (80%-96%) and not satisfactory sequencing efficiency (63%-73%), because AT-rich or homologous sequences existed, or concentration of PCR products was not high enough. Thus sequencing technology should be improved to obtain more sequences with high quality. Anyway, the fact that ITS2 region is not a coding region but possesses a conserved core of the secondary structure promotes establishment of data handling systems [bib_ref] ITS2 sequence-structure analysis in phylogenetics: a how-to manual for molecular systematics, Schultz [/bib_ref]. Recently, the CBOL (Consortium for the Barcode of Life) plant working group recommended using the 2-locus combination of rbcL + matK as a plant barcode, yet the barcode was shown to successfully discriminate among 907 samples from 550 species at the species level with a probability of 72%. The group admits that the two-locus barcode is far from perfect, and the search is not over [bib_ref] Plant bar code soon to become reality, Thomas [/bib_ref]. In our study, for ITS2 data, it is convenient (90%-93%) to identify more than 6600 samples from 4800 species using BLAST1 and the nearest genetic distance methods. We believe that the ITS2 region should be a standard barcode applied to international trade and safe use of medicinal plants.
# Materials and methods
## Taxon sampling
To select the most suitable DNA barcoding fragments, a total of 8557 medicinal plants and closely related samples belonging to 5905 species from 1010 diverse genera of 219 families in 7 phyla (Angiosperms, Gymnosperms, Ferns, Mosses, Liverworts, Algae and Fungi) were used [fig_ref] Table S5: Samples for testing potential barcodes and accession numbers in GenBank [/fig_ref] , S6, S7). A first set of plant samples collected in nine provinces of China (Beijing, Guangxi, Yunnan, Hainan, Sichuan, Fujian, Chongqing, Jilin and Hubei) was used to test seven potential DNA barcode regions [fig_ref] Table S5: Samples for testing potential barcodes and accession numbers in GenBank [/fig_ref]. All corresponding voucher samples are curated in the Herbarium of the Institute of Medicinal Plant Development, Chinese Academy of Medicinal Sciences. A second set of plant samples was used for testing the selected potential barcodes [fig_ref] Table S5: Samples for testing potential barcodes and accession numbers in GenBank [/fig_ref] , including a broader range of plant taxa with emphasis on closely related species. The selection of the first dataset was made mainly according to the Chinese Pharmacopoeia and Flora of China. This set of samples is of great medical and economic importance. The second sample set was selected to represent lower and higher plants (Angiosperms, Gymnosperms, Ferns, Mosses, Liverworts, Algae and Fungi).
## Pcr amplification and sequencing of candidate dna barcodes
Leaf tissues were first dried in silica gel. Ten milligrams of each of the dried tissues was rubbed for one minute at a frequency of 30 times/second in a FastPrep bead mill (Retsch MM400, Germany). DNA extractions were performed using the Plant Genomic DNA Kit (Tiangen Biotech Co., China) according to the manufacturer's instructions. The sequences of the universal primers for the DNA barcode to be tested, including those for psbA-trnH, matK, rbcL, rpoC1, ycf5 and ITS, and general PCR reaction conditions were obtained from previous studies [bib_ref] Use of DNA barcodes to identify flowering plants, Kress [/bib_ref] [bib_ref] DNA barcoding in the Cycadales: testing the potential of proposed barcoding markers..., Sass [/bib_ref] [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref] [bib_ref] A two-locus global DNA barcode for land plants: the coding rbcL gene..., Kress [/bib_ref]. Based on the conserved regions of 18S and 5.8S, we designed four pairs of primers for ITS1. Similarly, according to a previous study [bib_ref] Authentication of medicinal herbs using PCR-amplified ITS2 with specific primers, Chiou [/bib_ref] and the conserved regions of 5.8S and 26S, we also designed four pairs of primers for ITS2. PCR amplification was performed in 25ml reaction mixtures containing approximately 30 ng of genomic DNA template, 1 X PCR buffer without MgCl 2 , 2.0 mM MgCl 2 , 0.2 mM of each dNTP, 0.1 mM of each primer (synthesized by Sangon Co., China) and 1.0 U Taq DNA Polymerase (Biocolor BioScience & Technology Co., China), with a Peltier Thermal Cycler PTC0200 (BioRad Lab, Inc., USA). Purified PCR products were sequenced in both directions with the primers used for PCR amplification on a 3730XL sequencer (Applied Biosystems, USA).
To estimate the quality of the generated sequence traces, the original forward and reverse sequences were assembled using CodonCode Aligner 3.0 (CodonCode Co., USA). Base calling was carried out using the Phred program (version no. 0.020425.c). The quality values were defined for three levels: low quality (0 to 19 QV), medium quality (20 to 30 QV) and high quality (higher than 30 QV). The sequences showing .2 bases with a quality value below 20 QV in a 20-base window were trimmed. The forward and reverse reads have a minimum length of 100 bp, a minimum average QV of 30, and the post-trim lengths should be .50% of the original read length. In addition, the assembled contig should have a minimum average QV score of 40 and .50% overlap in the alignment of the forward and reverse reads. All sequences of the second set of plant samples containing the ''internal transcribed spacer 2''or ''psbA-trnH'' were retrieved according to Keller et al. [bib_ref] 8S-28S rRNA interaction and HMM-based ITS2 annotation, Keller [/bib_ref] and GenBank annotations. Subsequences marked as ITS2 or psbA-trnH intergenic spacer were recovered after deleting sequences with ambiguous nucleotides and those shorter than 100 bp.
## Sequence alignment, genetic analysis and species identification
Candidate DNA barcodes were aligned by Clustal W and Kimura 2-Parameter (K2P) distances were computed with PAUP4b10 (Florida State University, USA). Average intra-specific distances, theta and coalescent depth were calculated to determine intra-specific variation using a K2P distance matrix [bib_ref] DNA barcoding: error rates based on comprehensive sampling, Meyer [/bib_ref]. Average inter-specific distance, theta prime and smallest inter-specific distance were used to characterize inter-specific divergence [bib_ref] DNA barcoding: error rates based on comprehensive sampling, Meyer [/bib_ref] [bib_ref] The use of mean instead of smallest interspecific distances exaggerates the size..., Meier [/bib_ref]. Wilcoxon signed rank tests were performed as described previously [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref] [bib_ref] A two-locus global DNA barcode for land plants: the coding rbcL gene..., Kress [/bib_ref]. The distribution of intra-versus inter-specific variability was compared using DNA barcoding gaps [bib_ref] DNA barcoding the floras of biodiversity hotspots, Lahaye [/bib_ref] [bib_ref] DNA barcoding: error rates based on comprehensive sampling, Meyer [/bib_ref]. Two methods of species identification, including BLAST1 and the nearest distance method, were performed as described previously [bib_ref] Testing the reliability of genetic methods of species identification via simulation, Ross [/bib_ref]. [fig_ref] Figure 1: Genes from three genomes in plants that are candidate barcodes [/fig_ref] The barcoding gap between inter-specific and intraspecific divergences for six candidate barcodes. [fig_ref] Figure 2: Inter-specific divergence between congeneric species and intra-specific variation of paired loci [/fig_ref] The presence/absence of barcode gaps. The percentage of species pairs with dintra/dinter ratios ,1 were determined for six candidate regions including ITS2, psbA-trnH, matK, rbcL, ycf5, and rpoC1 to be 73.3%, 73.7%, 47.4%, 69.0%, 60.0%, and 35.7%, respectively. Therefore, ITS2 and psbA-trnH have significant barcode gaps. Found at: doi:10.1371/journal.pone.0008613.s002 (0.58 MB TIF) Wilcoxon two-sample tests for the divergences of paired loci with the same set of samples. Inter and Intra mean number of inter-specific distances and number of intra-specific distances, respectively. Found at: doi:10.1371/journal.pone.0008613.s003 (0.55 MB TIF)
## Supporting information
[fig] Figure 1: Genes from three genomes in plants that are candidate barcodes. Green markers are potential barcodes, red markers are poor candidates and yellow markers are pending to be investigated. doi:10.1371/journal.pone.0008613.g001 [/fig]
[fig] Figure 2: Inter-specific divergence between congeneric species and intra-specific variation of paired loci. doi:10.1371/journal.pone.0008613.g002 [/fig]
[table] Table 1: Efficiency of PCR amplification of three potential barcodes in a wide range of plant taxa. [/table]
[table] Table 2: Analysis of inter-specific divergence between congeneric species and intra-specific variation of candidate barcodes. [/table]
[table] Table 3: Identification efficiency for ITS2 and psbA-trnH loci using different methods of species identification. [/table]
[table] Table S1: List of universal primers and reaction conditions for candidate barcodes. Found at: doi:10.1371/journal.pone.0008613.s004 (0.06 MB DOC) Table S2 Efficiency of PCR amplification of potential barcodes in total number of samples. Found at: doi:10.1371/journal.pone.0008613.s005 (0.03 MB DOC) Table S3 Wilcoxon signed rank tests for inter-specific divergence. Found at: doi:10.1371/journal.pone.0008613.s006 (0.05 MB DOC) [/table]
[table] Table S4: Wilcoxon signed rank tests for intra-specific variation. Found at: doi:10.1371/journal.pone.0008613.s007 (0.04 MB DOC) [/table]
[table] Table S5: Samples for testing potential barcodes and accession numbers in GenBank. Found at: doi:10.1371/journal.pone.0008613.s008 (1.23 MB DOC) [/table]
[table] Table S6: Samples for determining the ability of the ITS2 barcode to identify species and accession numbers in GenBank. Found at: doi:10.1371/journal.pone.0008613.s009 (5.25 MB DOC) Table S7 Samples for determining the ability of the psbA-trnH barcode to identify species and accession numbers in GenBank. Found at: doi:10.1371/journal.pone.0008613.s010 (1.28 MB DOC) [/table]
|
Neuromodulation of the subthalamic nucleus in Parkinson’s disease: the effect of fiber tract stimulation on tremor control
Background Therapeutic effects of deep brain stimulation (DBS) of the subthalamic nucleus (STN) in Parkinson's disease (PD) may in parts be attributed to the stimulation of white matter near the targeted structure. The dentato-rubro-thalamic (DRT) tract supposed to improve tremor control in patients with essential tremor could be one candidate structure. The aim of this study was to investigate the effect of stimulation proximity to the DRT on tremor control in PD patients treated with STN-DBS. Methods For this retrospective analysis, we included 36 consecutive patients (median age 65.5 years) treated with STN-DBS for disabling motor symptoms including tremor. Stereotactic implantation of DBS electrodes into the motor area of the STN was performed using direct MRI-based targeting and intraoperative microelectrode recording. Tremor severity was assessed preoperatively and at regular intervals postoperatively (Unified Parkinson's Disease Rating Scale III). The DRT was visualized in 60 hemispheres after probabilistic fiber tracking (3-T MRI). The position of active electrode contacts was verified on intraoperative stereotactic X-rays and postoperative CT images after co-registration with 3D treatment planning MRI/CT images. We determined the shortest distance of active contacts to the ipsilateral DRT tracts on perpendicular view slices and correlated this value with tremor change percentage. Results Twelve patients had unilateral tremor only, and accordingly, 12 hemispheres were excluded from further imaging analysis. The remaining 60 hemispheres were associated with contralateral resting tremor. Active brain electrode contacts leading to resting tremor improvement (46 hemispheres) had a significantly shorter distance to the DRT (1.6 mm (0.9-2.1) [median (25th-75th percentiles)]) compared with contacts of non-responders (14 hemispheres, distance: 2.8 mm (2-4.6), p < 0.001). Conclusion This retrospective analysis suggests that in STN-DBS, better tremor control in PD patients correlates with the distance of active electrode contacts to the DRT. Tractography may optimize both individually DBS targeting and postoperative adjustment of stimulation parameters.
# Introduction
The predominant aspects of the motor symptoms associated with Parkinson's disease (PD) classify the patients into an akinetic-rigid, tremor-dominant, and balanced motor PD subtype. Patients with tremor-dominant PD present typically with rest tremor at a frequency of 4-6 Hz, which can be activated by mental stress and is temporarily suppressed during voluntary movements of the extremities.
Pathophysiological mechanisms supposed to be involved in tremor genesis are tremor of the extremity, tremor mediated through reflexes of the central nervous system (CNS), malfunction of feed forward loops within the CNS, and central oscillation. The latter is the result of rhythmic activity of a group of neurons within a nucleus or within neuronal networks, composed of loops consisting of neurons and their axons. Chronic high-frequency electrical stimulation (deep brain stimulation (DBS)) of the subthalamic nucleus (STN), the ventro-postero-lateral part of the globus pallidus internus (VPL-GPi), and the nucleus ventralis intermedius of the thalamus (V.im.) according to Hassler's classification) reduces effectively Parkinsonian rest tremor. These anatomical structures are relay nuclei of the basal ganglia loop and/or the cerebello-thalamic loop. Even though the mode of action is not completely unraveled to date, one important therapeutic mechanism of DBS is the disruption and overriding of network-wide pathological signals leading to electrical stabilization of diseased neuronal networks. In this context, myelinated axons and hence white matter are the assumed target structures for DBS effects. in 2001 the direct electrical stimulation of the prelemniscal radiation (RAPRL), a white matter tract in the subthalamic area, for DBS therapy of both tremor and rigidity in PD patients. Others linked the motor improvement in PD patients following STN-DBS more generally to the white matter surrounding this small nucleuscomprising the zona incerta (ZI) (anterodorsomedial to the STN), Forel's field H, the lenticular fascicle (Forel's H2 field anterodorsal to the STN), or the thalamic fascicle (H1 field dorsomedial to the STN). The RAPRL is located together with these anatomical structures, which, in essence, constitute the pallidothalamic bundle, in the posterior subthalamic area (PSA).
Another larger group of fibers inside the PSA belong to the dentato-rubro-thalamic (DRT) tract and represent ascending projections originating in the dentate (DN), emboliform, and globose nuclei of the cerebellum according to tracer studies in the brain of nonhuman primates. The DRT fibers constitute the main fiber bundle of the superior cerebellar peduncle. Approximately 90% of DRT fibers decussate at the pontine region. The minority of these projections enters the contralateral RN, and the majority terminates in the thalamic ventralis oralis posterior nucleus (V.o.p.) and the V.im. These nuclei, in turn, project onto the primary motor cortex.
In the past, several groups proposed the PSA as probably optimum target area for DBS treatment of several tremor types. Coenen et al. introduced specifically the DRT as candidate structure for effective antitremor neuromodulation.
Two studies investigated systematically the antitremor effect of DRT neuromodulation on Parkinsonian tremor to date. Endpoints of these case series studies were tremor reduction either during 14 days postoperatively or immediately intraoperatively. At this early time point, however, tremor improvement resulting from micro-lesional effects cannot be excluded. Thus, we analyzed retrospectively in 36 consecutively operated PD patients the effects of STN-DBS on tremor at longer follow-up intervals and determined the spatial position of active electrode contacts relative to the DRT and hence a possible electrical co-stimulation of DRT fibers.
# Methods
## Patients
For this retrospective analysis, we included 36 consecutive patients (median age at surgery: 65.5 years (range 27-77 years)) treated with STN-DBS for PD. All patients suffered in the medication-off state from resting tremor. Other inclusion criteria were (i) treatment planning MRI examination with a 3-T scanner including diffusion-weighted imaging (DWI), (ii) at least one complete follow-up examination within the first year after surgery, and (iii) a score of ≥ 2 out of 4 in any resting tremor components of the Unified Parkinson's Disease Rating Scale (UPDRS) III.lists the main patients' characteristics.
The clinical data and the tremor assessments were taken from the patients' medical records. For the imaging analysis, we used the existing treatment planning MRI files stored in our database. All these patients gave their informed consent that their clinical and imaging data can be used in a strictly pseudonymous form for retrospective evaluations and publications.
## Surgical procedure
The standardized treatment planning MRI protocols for a 3-T scanner comprised T1-, T2-, and PD-weighted series and DWI. To obtain contrast-enhanced T1-weighted images, the patients received an intravenous contrast infusion.
STN targets were primarily defined using brain atlas coordinates (1.7 mm posterior, 4 mm ventral, and 10.5 mm lateral to mid-commissural point (MCP)). Then, we aligned axial T2weighted images parallel to the AC-PC line (anterior-commissure-posterior commissure) and divided 4 mm below the intercommissural plane the hypointense STN signal into four quadrants. The definite target was placed at the intersection of these quadrants. If necessary, we adjusted the position in anterior-posterior direction in line with the rostral border of the nucleus ruber. The surgery for stereotactic implantation of brain electrodes in local anesthesia (28 patients) or general anesthesia (8 patients) was basically performed as described previously. Briefly, after mounting a modified Riechert-Mundinger stereotactic frame on the patient's head, we performed an intraoperative contrast-enhanced stereotactic CT examination (intravenous infusion of 1.53 g imeprol/kg BW). Stereotactic CT scans were co-registered with images from preoperative MRI examinations using dedicated software (PraezisPlus, Heidelberg, Germany).
For intraoperative registration of local field potentials (LFPs), we used a microprobe positioning device with five parallel arranged tubes (MicroDrive, inomed, Emmendingen, Germany) and 3-5 micro−/macroelectrodes per patient. Microelectrode recordings (MER) started 5 mm above the planned target until reaching a position where the LFP pattern indicated the transition zone from the STN into surrounding white matter and/or substantia nigra (SN). In awake patients, targeting was completed by intraoperative neurological examination determining the minimum voltage required for maximal tremor suppression and/or the improvement of rigor and the threshold for stimulation side effects using trajectories with a MER signal typical for the STN motor area. In patients operated under general anesthesia, the propofol medication was stopped, prior to MER. Under continuous remifentanil medication, characteristic bursting activity of STN neurons started 10-15 min after that time point.
The definite brain electrodes were implanted under fluoroscopic guidance using the trajectory with the longest course in the STN motor area according to MER recordings and/or the largest therapeutic window (voltage for the induction of side effects, voltage for motor improvement). Fourteen patients received brain electrodes with four non-segmented contacts (ACTIVA model 3389, Medtronic Inc., MN, USA) and 22 patients quadrupolar brain electrodes with two segmented contacts ((Boston Scientific, MA, USA; Cartesia, directional lead, DB 2202 (n = 14)) or (St. Jude Medical Neuromodulation Division, Plano, TX, USA; Infinity™ DBS System, directional lead, 6170 and 6172 (n = 8))). In all brain elect r o d e s , t h e d i s t a n c e b e t w e e n c o n t a c t s w a s 0.5 mm. Electrodes were placed at the ventral border of the motor STN (reference: most distal contact in electrodes with non-segmented contacts and most distal segmented contacted (contact two out of four) in directional leads). We confirmed the electrode position intraoperatively on stereotactic X-ray images (anterior-posterior and lateral direction) using stationary biplanar X-ray tubes and additionally by macrostimulation and clinical examination.
Extension cables and impulse generators (ACTIVA-PC: 9 patients, ACTIVA-RC: 3 patients (Medtronic Inc., MN, USA), Vercise Gevia: 14 patients (Boston Scientific, MA, USA), Infinity 7: 8 p atients (St. Jude Medical Neuromodulation Division, Plano, TX, USA)) were implanted at the same day under general anesthesia. All patients underwent postoperative non-stereotactic CT examinations. These images were co-registered with the stereotactic CT and preoperative MRI data.
# Tremor analysis
For the purpose of this study, we used in particular the resting tremor score of the UPDRS III motor score. The same expert movement disorders neurologist (I.G.) performed the examinations at baseline and postoperatively. We compared the preoperative OFF-medication state (OFF-MED) with the OFFmedication/ON-stimulation state (OFF-MED/ON-STIM) at follow-up visits. The global resting tremor score of the UPDRS III motor score rating all extremities plus lips and jaw can achieve a maximum value of 20 points. If resting tremor is registered separately for the right or left limbs, the maximum score is 8. For the actual tremor analysis, we defined a percentage value by dividing the patient's individual score by 20 in the case of global resting tremor scores and by 8 in the case of limb resting tremor scores. Tremor response to DBS was defined as tremor percentage difference (subtraction of postoperative tremor percentage scores from corresponding preoperative values). Negative percentage values or 0% values were classified as "No Improvement." Finally, the improvement tremor percentage was correlated with the distance to the DRT tract in the hemisphere contralateral to the affected upper limb. In 12 patients with only unilateral tremor, the hemispheres contralateral to the non-tremor body half were excluded from further analysis.
## Mri sequences, preprocessing, and fiber tracking
The imaging data were acquired on a Siemens MAGNETOM Verio 3-T MRI scanner (Siemens, Erlangen, Germany) using a 32-channel head coil and Syngo MR B19 software. The MR protocol includes a high-resolution, T1-weighted structural scan for anatomical reference using a 3D magnetizationprepared rapid acquisition gradient echo (MPRAGE) sequence (TE = 7.21 ms, TR = 2700 ms, TI = 1100 ms, flip angle = 7°, voxel size = 1 × 1 × 1 mm 3 , 176 axial slices, field of view = 256 × 192 mm 2 , scan time = 7 min:34 s). Moreover, a T2-weighted sequence (TE = 80 ms, TR = 6950 ms, bandwidth = 252 Hz/pixel, field of view = 256 × 192 mm 2 , 80 axial slices aligned with the anterior commissure-posterior commissure plane, slice thickness = 2 mm, scan time = 7 min:11 s) was acquired. DWI images were obtained using a twice refocused, single shot, echo planar imaging (EPI) pulse sequenceusing the following parameters: TE/TR = 86/11900 ms, matrix size = 128 × 128, 80 contiguous axial slices, isotropic resolution = 2 × 2 × 2 mm 3 , receiver bandwidth = 1698 Hz/pixel, and an echo spacing = 0.69 ms. Diffusion-weighted volumes were acquired along 30 non-collinear diffusion directions with a b value b = 1000 s/mm 2 and one volume without diffusion weighting (b = 0 s/mm 2 ). We allowed for parallel acquisition of independently reconstructed images using generalized auto calibrating, partially parallel acquisitions or GRAPPA, with acceleration factor of 3 and 57 reference lines. The total acquisition time was 7 min:08 s. For correction of geometric distortions in EPI caused by magnetic field inhomogeneities, a B0 field map was acquired prior to the EPI sequence using a double-echo gradient recalled echo (GRE) sequence (TE ½ = 4.92 ms/7.38 ms, TR = 514 ms, flip angle = 60°, voxel size = 2.7 × 2.7 × 3.2 mm 3 , FoV = 256 × 256 mm 2 , 50 axial slices).
For preprocessing, we used the FMRIB software library (FSL, University of Oxford, https://fsl.fmrib.ox.ac.uk) version 5.0.9. To correct for eddy-current-induced distortions, the DWI images were registered to a corresponding non-diffusion-weighted volume based on a 12-dof affine transformation using eddy_correct with spline interpolation. Geometric distortions induced by magnetic field inhomogeneities were corrected based on the GRE field map, and the diffusion data were registered to the corresponding structural scan. These steps (EPI distortion correction and EPI-to-MPRAGE registration) were performed simultaneously using epi_reg. Diffusion tensors were fitted with dtifit to obtain the eigenvalues and eigenvectors for each voxel.
The DRT reconstruction was carried out using a probabilistic approachimplemented in MATLAB (MathWorks, Natick, MA) in the native diffusion data space. Seed and filter regions were manually delineated based on structural T1-and T2-weighted images according to Kwon et al. and Yamada et al.. Conducting the stored spatial transformations, the region of interests (ROIs) were transformed in the native diffusion space. Briefly, DRT fibers started in the primary motor cortex as a seed region (precentral gyrus, 100,000 starts per voxel) and had to cross the following filter regions: ipsilateral red nucleus and contralateral dentate nucleus. Hit maps were created by incrementation of each voxel (starting at zero) if it crossed a filtered path. The hit maps were transformed into the space of the T1 anatomy, and DICOM files were created using the Mathematica software package (Wolfram Research, Oxfordshire, UK).
To delineate the DRT, we used a method based on P-FT at which the likelihood of water molecules' diffusion at multiple directions is determinedin comparison with only one direction per voxel as the case with deterministic fiber tracking (D-FT). Accordingly, the limitation of not being able to accurately delineate the white matter tracts in dense areas of crossing, kissing, or branching fibers was prevented. In accordance with groups that used P-FT, we were able to show the tract's decussation instead of ipsilateral tracts resulted from D-FT. In fact, studies comparing P-FT and D-FT showed higher anatomical precision using P-FT not only with tracking the DRT but also other white matter fibers such as the corticospinal tract and fornix.
# Image analysis
Intraoperative stereotactic X-ray images (AP and lateral view) were co-registered with preoperative MRI images and intraoperative stereotactic CT images . On the intraoperative stereotactic X-ray images, we registered for each active electrode contact X, Y, and Z coordinates, referring to the center of these contacts (monopolar cathodic stimulation). In the case of bipolar stimulation, we used the center of the distance between the contacts as reference. The obtained coordinates were transferred into the preoperative MRI images, which were co-registered with stereotactic CT images.
Segmentation of the dentate nucleus was not very specific. As a result, the generated ROI included other cerebellar nuclei, contributing also fibers to the origin of the DRT. This wide segmentation resulted on DRT tracts which became visible on tractographic MRI images for each individual hemisphere with different contrast and consecutively different diameters. To standardize the contrast among study patients, we applied the following semiquantitative procedure: First, in each brain hemisphere, we defined on axial T2-weighted MRI images 4 mm ventral to the intercommissural line a straight line along the medial and posterior borders of the STN, respectively. Then, we adjusted the image contrast as long as the DRT tract signal did no more cross these lines. Finally, we measured the shortest distance between the active contact and the border of the DRT tract on images sectioned perpendicular to the long axis of the brain electrode. Values were given in millimeters.
# Statistical analysis
Pairwise comparison between resting tremor percentage before and after DBS surgery was calculated using the Wilcoxon signed-rank test since the data did not pass normality tests. According to normality tests, independent t test or Mann-Whitney test was used to investigate difference of UPDRS III score between tremor responders group and tremor nonresponders group. Mann-Whitney test was also used to determine significant difference of distances between the active electrode contacts and the DRT tract in both groups. Spearman's correlation coefficient (Rs) was used to investigate correlation between resting tremor improvement percentage and distance to the DRT. The variances explained by the correlations (Rs 2 ) were also calculated and reported. Significance was considered if the p value was ≤ 0.05. The statistical analysis was performed using SPSS 26 (statistical Package for the Social Sciences, IBM).
# Results
## General outcome
Implantation of DBS systems was performed without intraoperative or immediate postoperative complications. Three patients presented postoperatively with the following problems: suture granuloma in one case requiring wound revision and infection in two patients with subsequent removal of extension cables and impulse generator. The DBS systems were reimplanted 2 months later, which was in one case combined with repositioning of one brain electrode. The remaining 33 patients had no complications due to DBS surgery. The median postoperative follow-up time was 3 months (range, 1-7 months).
## Motor response-tremor
Compared with baseline, DBS reduced the resting tremor percentage significantly in the whole study population. The median global tremor percentage decreased from 20% [25th-75th percentiles] preoperatively to 5% (0-15) postoperatively (p < 0.0001). Moreover, the median tremor percentage of right and left limbs decreased from 25% (12.5-46.9) [25th-75th percentiles] at baseline to 0% (0-12.5) under STIM-ON conditions (p < 0.0001) and from 37.5% [25th-75th percentiles] to 0% (0-25) (p < 0.0001), respectively . All differences were statistically significant.
## Motor response-updrs iii scores stratified for patients with tremor improvement and patients with non-tremor improvement
At baseline, the UPDRS III motor scores in the OFFmedication state (OFF-MED) did not differ significantly between patients, with tremor improvement under DBS (tremor responders, N = 25) and those with no tremor improvement (tremor non-responders, N = 11) (tremor responder 31.88 ± 8.64 (mean ± SD) vs. tremor non-responder 28.73 ± 18.40 (mean ± SD), p = 0.09). When UPDRS III tremor scores were subtracted from the total UPDRS III score, the difference between the groups was also not statistically significant (p = 0.11). However, at follow-up visits, the UPDRS III motor score differed statistically significant between tremor responders (13.84 ± 6.49 (mean ± SD)) and tremor nonresponders (19.27 ± 8.51 (mean ± SD), p = 0.04). After subtracting the resting tremor score from the follow-up total UPDRS III motor score (OFF-MED/ON-STIM), the group difference was again not statistically significant (tremor responder: 12.92 ± 6.06 (mean ± SD) vs. 14.91 ± 6.86 (mean ± SD) tremor non-responder, p = 0.39).
## Distance of active contacts the drt and tremor response
Probabilistic fiber tracking of the DRT tract was performed in 60 hemispheres. Twelve hemispheres were excluded from imaging analysis because of absent tremor in the corresponding, contralateral body half at baseline. In 46 out of 60 hemispheres, DBS was associated with contralateral tremor improvement. In the remaining 14 hemispheres, electrical stimulation did not reduce contralateral tremor.
The distance between the active electrode contacts and the DRT tracts was shorter in tremor responders (1.6 mm (0.9-2.1) [median (25th-75th percentiles)]) compared with patients without tremor response (2.8 mm (2-4.6)). This Whiskers represent range (minimum-maximum). Wilcoxon signed-rank test was used to determine significance difference was statistically significant (p < 0.001). Spearman's correlation revealed a significant inverse correlation between resting tremor improvement percentage and distance to the DRT tract (R S = − 0.6), p < 0.0001) .
# Discussion
The main results of the actual study are (i) STN-DBS reduced resting tremor significantly in the whole study population. (ii) The measured distance was significantly shorter in brain hemispheres that corresponded to tremor improvement in the contralateral body half. (iii) Tremor improvement was significantly correlated to a shorter distance to the DRT tract. (iv) The antitremor effect seemed to be determined rather by the distance to white matter than by the position of active electrode contacts inside the motor area of the STN. When we subtracted the tremor subscore of the UPDRS III from the total UPDRS III score, there was no more a significant difference in the motor outcome between patients with tremor reduction after STN-DBS and those classified as tremor non-responders. We took this observation as an indirect argument that all patients underwent basically effective electrical stimulation in the STN motor part with the exception of tremor.summarizes the key findings of published studies addressing the role of electrical DRT stimulation and tremor control. In five of these case series, patients with ET were treated. With one exemption, the results confirmed the hypothesis that a shorter distance between active electrode contacts and the DRT may improve tremor suppression. Fenoy et al., who enrolled 20 ET patients in a prospective observational study, targeted the DRT directly. The volume of activated tissue of active contacts covered the DRT fibers partially or completely in all cases, and the tremor suppression under STIM-ON conditions was significant compared to baseline.
Coenen et al. analyzed in a recently published retrospective study the relationship of symptom improvement after DBS and distance to the DRT in a heterogeneous group of patients with tremor of various origin (ET, PD, multiple sclerosis, dystonic head tremor, tardive dystonia). The authors used an efficacy measure, the tremor improvement per current ratio (TiCR = improvement on a four-point scale divided by the current applied in mA). The antitremor effect of DBS was determined intraoperatively at stimulation points with different distances to the target and hence to the DRT (251 points in total). The outcome measure TiCR deteriorated significantly if the distance to both the border and the center of the DRT increased.
Other studies listed inaimed at the DBS treatment of Parkinsonian tremor only. In three of these studies, the patients underwent STN-DBS. The study from Sweet and coworkers aimed primarily at the tractographic visualization of fiber tracts connecting the basal ganglia with the cerebellum in 14 PD patients. In accordance with results from labeling studies performed in nonhuman primates, the group characterized in patients a descending subthalamo-ponto-cerebellar tract and the ascending dendato-thalamic tract. Nine out of 14 patients were treated with STN-DBS, and eight of nine presented with tremor. In these cases, the authors defined the proximity of active contacts and of the VTA of active electrode contacts to each of the fiber tracts. The relationship between motor outcome in general (tremor response was not specifically reported) and the distance from the active contact to the DRT was not significant. However, this group observed a nonsignificant trend toward better outcome when the VTA was closer to the fiber tract.
Prent et al. included in their retrospective study 35 patients treated with bilateral STN-DBS for PD. Twenty out of 35 patients were suffering from tremor. In these cases, the authors Resting tremor score (RTS) components were subtracted from the total UPDRS III score at the different examinations time points. § Mann-Whitney test and § § independent t test were performed to investigate statistical difference Relation between resting tremor improvement and distance of active electrode contacts to the DRT in mm. R s, Spearman's correlation coefficient; R s 2 , coefficient of determination observed a statistically significant negative correlation between the proximity of active contacts to the DRT and tremor control 14 days after DBS surgery. The distance of electrode contacts associated with tremor improvement was located 1.13 mm closer to the center of the DRT compared with noneffective contacts. This difference is approximately in the same range as in the actual study where effective electrode contacts were 1.7 mm closer to the border of the DRT compared with electrode contacts without antitremor effect. Moreover, the tremor improvement observed 14 days after implantation of brain electrodes by Prent et al. was not a result of microlesion effects but a DBS effect, because the actual study with longer follow-up intervals (range 1-7 months) came to the same conclusion. As mentioned, the results of some studies on this subject were statistically not significant. Three possible reasons can probably explain this discrepancy. (i) The first possibility is the low sample size in many studies. The fact that electrical activation of the intended target area-the motor thalamus or the PSA-is supposed to have an antitremor effect leads a priori to a high number of tremor responders if electrical stimulation is performed. Consequently, differences of the strength of tremor reduction among the responding patients will be small which may complicate statistical comparison tests in small patient cohorts. Larger patient cohorts with a size comparable with the actual analysis may lead then to statistically significant results. (ii) The segmentation technique applied to tract the DRT can also potentially influence the measured distances. As noticed from, the definition of ROIs is heterogeneous among different study groups, limiting the direct comparison and assessment of the results published to date. (iii) Another possibility is that in contrast to the action tremor of ET patients, the proximity to the DRT will not exclusively determine the antitremor effect of DBS in resting tremor of PD patients. Helmich and others developed a dimmer-switch hypothesis of Parkinsonian tremor, which, in essence, has two driving components. One is located in the basal ganglia and represents the trigger of tremor. This part would explain tremor improvement following GPi-DBS. Referred to the subthalamic white matter tracts, candidate structures mediating antitremor effects of DBS could be the pallidothalamic projections (ansa lenticularis, lenticular fascicle). Activity inside the cerebello-thalamo-cortical circuit, on the contrary, is linked to tremor power and is supposed to act as tremor maintainer favoring the V.im. or the DRT as target. In addition, the role of the STN in Parkinsonian tremor is not completely explained by the dimmer-switch model. Anatomically, the STN has connections to the GPi, is directly connected to the motor cortex (hyperdirect path), and projects via the pons onto the cerebellum. Thus, according to its particular position in the tremor circuitry, the STN could potentially be involved in triggering tremor, maintenance of the tremor rhythm, or in both.
Other white matter tracts inside the PSA with antitremor effects when electrically stimulated in PD patients are the caudal zona incerta (cZI) or the RAPRL located posterior and medial to the STN. Interestingly, three studies reported also a positive effect of DBS on other PD motor symptoms than tremor. Blomstedt et al. randomized in their prospective study PD patients either to cZI-DBS or best medical treatment. The stimulation effect was clearly superior to medication, and the UPDRS III improvement of 41% (OFF-MED/ON-STIM) compared with baseline in the cZI-DBS group was on the order of other clinical studies randomizing patients to either STN-DBS or best medical treatment. However, in contrast to STN-DBS in patients with cZI-DBS, the dopaminergic medication could not be reduced significantly indicating probably a different mode of action.
Limitations of the actual study include the retrospective design and a tremor analysis performed unblinded and at different follow-up time points (range 1-7 month). Another critical point may be that our study cohort included patients with tremor-dominant PD and others with the balanced motor PD subtype. Patients with a tremor-dominant subtype have often a more benign disease course than patients without tremor. Even though there is, to our knowledge, no study addressing specifically the characteristics of tremor-dominant and balanced subtypes in PD, it cannot be excluded that the tremor of patients presenting with the balanced subtype may respond to STN-DBS differently compared with tremor-dominant patients.
In conclusion, this retrospective analysis is in line with the observation of others that in STN-DBS, a shorter distance of active electrode contracts to the DRT improves tremor control in PD patients. In three studies including the actual analysis, this correlation was statistically significant. As a consequence, the use of individualized tractography and identification of the distance between a planned trajectory and the DRT could potentially improve the outcome in PD patients treated with STN-DBS for tremor. In addition, further research combining advanced imaging and clinical data is necessary taking also candidate structures of the PSA other than the DRT into consideration.
Funding Open Access funding enabled and organized by Projekt DEAL.
## Compliance with ethical standards
Conflict of interest None.
## Author
Category of disclosure Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent All included patients gave their informed consent that their clinical and imaging data can be used in a strictly pseudonymous form for retrospective evaluations and publications.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. |
Households Health Expenditure in Interannual Correlation With Public Health Expenditure in Greece
The aim of this article is to investigate the relationship between the public and the private health expenditure (macroeconomic and microeconomic approach) over time and within the recession and austerity period in Greece, in order to find out whether the strict Memorandum health policies pass, influence, or go along with the health expenditure to the final consumer, i.e., the health services user. In this context, by using econometric tools, such as multiple regression and cointegration analysis on the raw microdata of Household Budget Surveys from 1987/88 up to 2018, as well as by using data of public expenditure of Organization for Economic Co-operation and Development-Health Statistics 2019 in the Stata version 13, the study compares the variation of household and public health expenditures before and after the financial crisis in Greece and also examines the correlation between the two variables. The analysis demonstrated that the Greek household health expenditure (HHE) was rapidly increasing during the period 1988-2008, and afterward, it started decreasing. Moreover, the total private and the total public health expenditures seem to have a bidirectional long-run relationship and significant cointegration. The same was observed regarding the public expenditure and household medical services expenditure, as well as pharmaceuticals. Furthermore, the results indicate that over the years of recession, the monthly HHE decreases, due to the confiscation of middle-class income, which led to consumerism restrictions. However, as households are now spending a bigger portion of their shrunken income for health (as health is an inelastic commodity), HHE, as a proportion of total private expenditure, has eventually risen.
# Introduction
The study of health expenditure over time is a difficult task, because it has to be considered the fact that health is a product of inelastic demand and also that the public sector has a role of "payer, " and therefore, it has the power to regulate the market prices. Financial fluctuations can affect the providers, the users, and, ultimately, the population's health.
The health sector in Greece, after a period of growth within the first decade of the millennium, when total expenditure on health [% gross domestic product (GDP), (1)] exceeded 10% (above the European Union average), the following years it began to moderate. After signing the Memorandum of Understanding in 2010 (2), a number of urgent expenditure restricting measures and structural reforms were imposed to the Greek health sector and, in particular, to pharmaceutical sector. The last one seems to significantly contribute (negatively) to both deficit and public debt, due to the excessive public spending and the lack of control in both the volume and the cost of prescriptions.
Therefore, in May 2010, pharmaceutical industry became the focus of fiscal consolidation, and it was one of the main areas of intervention, in order to reduce public pharmaceutical expenditure to 1% of GDP, thus approaching the European average. As a result, public pharmaceutical expenditure decreased by 60.8% between 2009 and 2014, reaching e2billion in 2014 against e5.1 billion in 2009 (4).
Coupled with the economy contraction, the health expenditure in Greece has also been reduced proportionally from 9.8% of GDP in 2008 to 8.1% in 2014. Reduced health costs, which were required under the austerity program, have been criticized, because they do not contain specific provisions to safeguard the National Health Service (NHS). The NHS was established in the 1980s, as part of the national compulsory social security program, through which most of the Greek inhabitants have been cared. However, as mentioned above, Greece has failed to control health spending between 2000 and 2009, and the health budget deficit of the country reached e50 billion. Consequently, at the beginning of the crisis, the health sector was set to be a priority by the Troika, having contributed significantly to the economic derailing of the country.
On top of that, there is a "hidden" financial burden of at least 27%, according to Souliotis et al., which influences negatively the living conditions of the households, which is not reported as purchasing ability or cost of living: the informal (under-thetable) payments, as a common reimbursement method for the health care services in order to gain access, to reduce waiting time, to receive a higher quality of services or out of gratitude, and so on. This survey also reveals that, because of severe financial pressure, there is a growing unwillingness of citizens to pay informally and that there is an increasing demand for these payments as a prerequisite for access to services or to redeem services provided. Apart from this, a more recent survey concludes that more than 60% of the health care incidents involved informal payments in Greece. Despite near-universal coverage of the population by public health insurance, informal payments are widespread and a major source of inequity and inefficiency in the Greek health care system. The present study examined the level of the public health expenditure and the private Greek household health expenditure (HHE) within the period 1988-2018. Additionally, the existence of interaction between the public and the private expenditure was examined by using the method of cointegration, in order to determine whether a change in the Greek public health spending has ultimately the same behavior with the variation of the private HHE. The survey results will provide an enlightening insight into the evolution of health expenditure in Greece for the past 30 years, and the impact of the financial crisis will be determined, through the evolution of the costs.
# Materials and methods
In this study, preform (meta) data were derived from the Greek Statistical Authority (HEL.STAT.), concerning the private health expenditure from the Household Budget Survey (HBS) for the , as well as annually data from 2008 up to 2018. It should be borne in mind that HBSs are carried out randomly and at a regular basis 1 , by HELSTAT, throughout Greek territory. Through these surveys, information on the consumer spending, income, housing facilities, consumer durables and household socioeconomic characteristics, and household members was collected. In regard to the health expenditure, data collected referred to the HBS records, inter alia, households' responses to pharmaceuticals expendituretreatment-equipment devices and non-hospital medical and hospital care services, not individually, but on household level, because the household in-charge person reflects the costs and consumer behaviors of the entire household. Additionally, data were drawn from the Organization for Economic Co-operation and Development (10) on the level of public expenditure over time as a percentage of GDP. Health spending refers to the final consumption of goods and the use of health services (i.e., current health expenditure), including the personal therapeutic care (curative care, rehabilitation, long-term nursing care, ancillary health care services and medical products, prevention, and public health services, as well as health care administrative costs, excluding investment and research-educational costs). It should be noted that the HBS results are produced in accordance with the relevant International Classification Systems, as seen below, for the category of Health. Also, HBSs for the covered all the areas of the country (urban, semiurban, and rural) with a sample size from 6,000 to 6,800 households. The duration of the aforementioned surveys was 1 year. From the year 2008, taking into consideration the national needs for the Consumer Price Index compilation, because the main purpose of the HBS is to determine in detail the HE pattern in order to revise the Consumer Price Index and in order to have higher reliability for being able to produce comparable statistics used by the National Accounts Division, the annual and continual conduct of the survey was decided with a sample of ∼4,000 households in the whole Greek territory.
As regards the accuracy of the data, sampling and nonsampling errors occurred, due to the fact that the HBS is a sampling survey. The sampling errors are depicted by estimating the coefficient of variation for the main survey variables and their values proved to be within the acceptance limits. The non-sampling errors were divided into non-response errors, elaboration errors, and measurement errors. During the previous years, the overall accuracy of the survey was good enough.
Data analysis was performed via Stata software, version 13, and for the research purposes, the technique of cointegration analysis was used.
The vast majority of chronological series of economic variables are not characterized by stagnation. Applying a simple regression, just to identify the possible correlation between two or more variables, can lead to the phenomenon of malicious or apparent or spurious regression, as proposed by . It is very difficult to analyze time-series data. Spuriosity is an issue. A correlation may arise between them (apparent), which, as the authors argued, may be due to the existence of short-term trends, which is treated by applying first differences in time-series to turn them into time-series characterized by stagnation. The concept of integration came to eliminate this phenomenon. In non-stationary time-series that show the same trend or otherwise "move together, " the results of regression may not be fictitious, so in this case the usual conclusions based on statistics t and F may apply. When there is a causal relationship, such as expected, for example, between variables of income and health expenditure, the two variables will not be deviated for a long time, even though they both increase; i.e., they have a tendency and are therefore non-stationary. This "synchronization" of non-stationary timeseries is the basic idea behind the concept of integration, where two or more variables move in the same direction in the long run, i.e., there is a long-term equilibrium relationship among variables, without necessarily to apply also in the short term.
According to the aforementioned author, two or more non-stationary variables are linked to a longterm relationship, in the sense that they "move together in time" showing the same trend. For example, if the probabilistic linear relationship of two non-stationary variables with a degree of completion of 1 is characterized by a degree of completion of 0 (stationary), then the two time-series are cointegrated, in a sense that the variable describing their relationship moves long around a point of equilibrium.
All prices in the tables/graphs hereof are in euros (e), unless otherwise defined. It should be mentioned that Greece adopted the euros in 2001, after irrevocably fixing the conversion rate on June 2000 at e1 = 340.75 drachmas. Then, on January 1, 2002, euro banknotes and coins were launched and put into circulation within the transition period at the end of February 2002 and onward. All data of HBS before this currency changeover, that is, HBS 1988, HBS 1994, and HBS 1999, for the purpose hereof, were officially given by HEL.STAT. converted to euros.
Furthermore, HBS depicts the average monthly expenditure of the whole household, so prices in the tables/graphs hereof refer to average monthly health expenditure of the whole household. There was no year chosen as a reference year for inflation observations, because the research was based in current prices. It should be noticed that the inflation rates (average consumer prices) during the period 1988-2018 ranged between 20.3 (1990) and −1.4 (2014), and more specifically during the crisis period 2008-2018 the annual percent change ranged between 4.7 (2010) and −1.4 (2014).
## Cointegration hypothesis
A set of time-series cointegrates, when there is a linear combination of these time series, which are stationary, cannot present a stochastic trend. The linear time-series combination is the equation of cointegration that represents the longterm equilibrium relationship among these sequences. For the purposes of a cointegration test, variables must be stationary at the same level. A cointegration test was performed via the Johansen test, in order to examine whether there is a cointegration between the variables. The Johansen test was chosen because many authors agree that it is an improvement over the Engle-Granger test and Stock & Watson's test (in Introduction to Econometrics). Furthermore, it avoids the issue of choosing a dependent variable, as well as issues created when errors are carried from one step to the next. As such, the test can detect multiple cointegrating vectors, and it is more appropriate than Engle-Granger for multivariate analysis. Another desirable property is that the Johansen test treats every test variable as endogenous variable.
We therefore define the following hypothesis:
- Ho: There is no cointegration between variables.
- Ha: There is cointegration.
If time series are co-integrated, it means that, in the longterm, variables move together on the same direction without however, meaning that in the short-term the same phenomenon will occur. The test shows if there are, and how many, linear cointegration relations between the controlled variables.
The link between the short-and long-term relationships of variables is performed by the existence of cointegration, as a prerequisite for assessing the Vector Error Correction (VEC) model. If the findings show that in our model there is, indeed, a cointegration relationship, it allows the assessment of the VEC model. Hence, VEC model is used, because we want to simultaneously examine the results of the regressions relating to the linear equations. All assays were performed in 5% level of significance (α = 0.05).portray the longitudinal data on the average monthly household expenditure for medicines, therapeutic appliances, and equipment. The results are given both for the total average cost and their subcategories. All prices are in euros (e). Furthermore, it should be noted that until 2008 no data were available regarding these three cost categories separately.shows that the average total household expenditure in medicines, therapeutic apparatus, and equipment increases rapidly from the year 1988 to the year 2018. The increase reaches 994.38%, from e3.56 in 1988 to e38.96 in 2018. Furthermore, the graph shows that within 2008-2011 (financial crisis), the average monthly household spending had a downward trend compared to previous years, whereas in the period from 2012 to 2018, it began to rise again at considerable pace. Additionally, the results showed that the biggest proportion of the category "medicines, therapeutic apparatus, and equipment" expenditure comes from the medicines expenditure. In contrast, the "other therapeutic apparatus and equipment" household expenditures seem to be stagnant and declining from 2004 to 2018.
# Results
## Descriptive presentation of the hhe categories
Inand, the temporal data on the average monthly household expenditures on medical services (outpatient) are presented. The results show both the total average cost of medical services and its subcategories. Furthermore, it should be noted that there are overall figures with respect to the classification of expenditure in subcategories only after the year 2004.shows that the average monthly total household expenditure on medical services was increasing from 1988 to 2008 (growth rate of 525.29%, from e14 in 1988 to e87.54 in 2008). In contrast, within the period 2008-2018, the average monthly total household expenditure on medical services appeared to have a declining trend from e87.54 in 2008, to e33.18 in 2018, which is a 62.1% decrease. Additionally, the results showed that, since the beginning of the financial crisis in 2008, the average household expenditure for medical services had a downward trend compared to the previous years. Moreover, the biggest part of expenditure for medical services derives from the dental services. In contrast, the paramedical services expenditure seems to have been stagnant and declining from 2004 to 2018, while it happens to be the smallest percentage among three subcategories of the medical expenses.
As regards the paramedical Greek household services, particularly, from HBS data for the years 1987, 1994, 1999, and 2004 and the period from 2008 to 2018, it is observed that the biggest portion of the expenditure for paramedical services is related to the microbiological laboratory services and the radiology centers, followed by the paramedical staff services, while a very small percentage is attributable to the other non-hospital services.andindicate a common and constant reduction of costs in all three subcategories over the years. The next category of health expenditure refers to the hospital care, and the analytical results are presented inandbelow. The results show that the average monthly expenditure of the Greek households for the hospital care increased steadily over the period 1988-2018. Specifically, the average cost of hospitalization increased from e3.1 in 1988 to e35.85 in 2018, which is equal to an increase of 1.056.45% (∼10-fold average expenditure). As regards the subcategories of the hospital care, it is noticed that the highest average contribution to the total hospital care was the private hospital care, where it is impossible to separate accommodation services and health services (medical or paramedical) followed by the public hospital and the private hospital care (accommodation, nutrition, etc.). Regarding the subcategories of the hospital care expenditure, the most important finding is that the private hospital care expenditure, where it is impossible to separate the accommodation services and the health services (medical or paramedical), increases considerably over the years, whereas the other categories show either stagnation or decline.
It is also noticed that within the period 2008-2018, the average private hospital care, where it is impossible to separate the accommodation services and the health services (medical or paramedical) expenditure, increased from e6.34 to e24.66 (growth rate 288.96%), whereas the average monthly expenditure for the private hospital care (accommodation, food, etc.) decreased from e6.08 to e3.03 (reduction rate 50.16%), and the average monthly expenditure for the private hospital care (doctors' services and fees) decreased from e2.91 to e1.54 (reduction rate 47.08%).
## Descriptive presentation of the total health expenditure
Findings concerning the total HHE for the period 1988-2018 and its three main subcategories of expenditure (medical, pharmaceutical, hospital) are shown below.shows that the total monthly HHE in Greece had an upward trend during the period 1988-2008. Specifically, the total monthly expenditures from e20.66 in 1988 reached e142.1 in 2008, which refers to an increase of 587.8% (5-fold increase on expenditure). On the contrary, from the beginning of the financial crisis up to 2018, the average monthly total household expenditure declined from e142.1 in 2008 to e107.99 in 2018, which is a reduction of 24%.
## Descriptive presentation of the public health expenditure as a percentage of gdp
At this point, we present the findings regarding the public health expenditure as a percentage of GDP in Greece. The results of this study are illustrated in, proving that public spending had an increasing trend within the period 1988 to 2010, when the percentage of the health expenditure to GDP rose from 3 to 6.6%, whereas from 2010 up to 2018, we observed a sharp reduction in the public spending, due to the implementation of fiscal adjustment strict measures, from 6.6 to 4.7% of GDP.shows both, simultaneously, the HHE and the public health expenditure development. According to this graph, the private household expenditure seems to go "hand in hand" with the public expenditure, because, as mentioned previously, the total household spending appears to have a strong growth during
## Joint presentation of the hhe and the public health expenditure development (multiple cointegration)
## Cointegration analysis results between the public health spending and the greek hhe
In the last section of results, data are presented regarding the integration of the HHE to the public expenditure as a percentage of GDP. Because of few observations (which did not allow us to draw reliable conclusions), Johansen test was not expected to yield results in the total data set, and therefore, integration test between the public health expenditure and the total private health expenditure was applied, as well as between the public health expenditure and the two main categories of the private health expenses (medical and pharmaceutical).shows that for r = 0, the null hypothesis is rejected (because trace statistic > critical value), and this means that we reject the hypothesis of cointegration equation absence. Conversely, for r = 1, the null hypothesis is not rejected (2.16 < critical value = 3.74), and we do not reject the null hypothesis that there is only one cointegration equation. This means that there is a cointegration relationship between the public health spending and the private health expenditure.
Subsequently, cointegration equation was evaluated and the multivariate model/error correction model VEC model (or otherwise ECVAR). The results are detailed in.
The cointegration equation is as follows:
ECT t−1 = 4.635 + Household HE t−1 − 22.832 · Public HE t−1 VEC indicates that long-run equilibrium equations are as follows: The results suggest that there is a statistically significant cointegration between the public expenditure and the private household spending. The public expenditure (GDP) coefficient in the cointegration equation is statistically significant, so are the adjustment parameters. The adjustment parameters in this bilateral example are easy to be interpreted and one can notice that the estimates have the correct signs (negative) and indicate rapid adjustment toward equilibrium, showing that as the public health expenditures increase, the private expenditures are reduced (negative: b = −22.832, p < 0.05). The estimation of the coefficient D_V1 (ce1) equals −0.718 (p < 0.05). Therefore, when the average rate of private expenditure is very high, a sharp fall is followed to the level of the public expenditure. The estimated coefficient D_GDP (ce1) equals 0.068 (p < 0.05), indicating that the average rate of the public spending is quickly adapted to the levels of the private spending. This means that ∼71.8% of the imbalances between the private and the public expenditure are corrected by changing the private spending, whereas only 6.8% of the imbalances are corrected by changing the public spending.
[formula] Public HE t = −0.718 · ECT t−1 + 1.826 · Public HE t−1 +2.218 · Household HE t−1 − 0.0002 Household HE t = 0.681·ECT t−1 −0.681·Public HE t−1 −0.104·Household HE t−1 −0.002 [/formula]
The results clearly indicate a bi-long relationship between the private and the public expenditure, where the private spending seems to be adapted more rapidly to the public spending, while there are those who mainly correct the imbalance between the private and the public expenditure.
# Discussion
It should be noted here that health expenditures in Greece are financed through a combination of funding agencies, including the public spending and the mandatory health insurance ("government/mandatory") and the private health insurance, and the private funds, such as the out-of-pocket household payments, non-governmental organizations expenditures, and so on.
However, the Greek health care system faces a public funding gap, resulting from the current financial crisis and the relevant austerity measures being forced. Official data have shown that Greece's GDP has been contracted by 25% within the last decade, and this has had an impact on the health expenditure of the Greek households (of already high out-of-pocket payment), as well. Based on the literature review, it is clear enough that most OECD countries followed, more or less, the same route: increase in spending during pre-crisis period and declining afterward. The austerity program, in which Greece entered due to the financial crisis, has indeed significantly affected the HHE, as well. Both microeconomics and macroeconomics research lead to the same conclusion.
As mentioned before, in the research approach of the whole project, part of which is hereof, both microeconomics and macroeconomics were used. Studying the microeconomic data in the health care market of a country may provide very important information in detail. One can easily find where individuals purchase medical products or services; the decisions they make, which are based on price; quantity and quality; and how these three factors interact with each other. Then, health policy makers can perform or adjust more effectively their health care planning into the community.
Other people gloat over the fact that macroeconomic approach is always the best to adopt, as it refers to the "sum total of economic activity, dealing with the issues of growth, inflation and unemployment". Macroeconomics helps countries to understand their development and growth, and in terms of health care, it influences policy making and planning in macro perspective.
The only thing beyond doubt is that the combination of macroeconomic and microeconomic studies is a complementary perspective on the overall subject of the health care economy and therefore should both be considered on decision making by health care policy planners.
# Limitations
Perhaps, the most important limitation of any study, concerning the private health expenditure, is the lack of data on the informal payments (under-the-table payments to medical staff, etc., as mentioned previously), which constitute the phenomenon of the shadow economy and make the already enormous private health costs skyrocket. Although the present study is based on the official published metadata (HBS) and not on the original data collected by the authors, while part of these informal payments is included in HBS, thus making HBS more reliable than in other surveys, in no way were they considered to be representative of the actual size of informal payments. Furthermore, presented data in this study may not fully reflect the Greek health expenditure, given the relatively short period of studying. Further research and comparative analysis are needed, to identify the impact of Greece's financial crisis on the HHE, whereas the effect of the economic crisis appears clearer over the longer periods of time. However, the statistical approach and the validation of hypothesis testing remain the strongest point of this study.
# Conclusions
The austerity program, in which Greece entered because of the financial crisis, significantly affected the health expenditure. The results of this study showed that the average total household expenditure increased 5-fold from 1988 to 2008, and from 2008 to 2018, the average total expenditure of households on health decreased by 24%.
Regarding subsets, it was shown that the average household expenditure on medicines, therapeutic appliances, and equipment were considerably increased from 1988 to 2018, while it seemed that it was the only category that was not affected significantly by the financial crisis. Similarly, the average household spending on the hospital care increased notably from 1988 to 2018, and it was affected very slightly by the economic crisis. Conversely, the average costs for medical services showed an upward trend from 1988 to 2008 and then a significant decrease from 2008 to 2018.
As regards the public expenditure, the results showed that there was an increase during the period 1988-2010 and a decrease during the period 2010-2018. In fact, one can say that the households at the onset of the financial crisis have had reduced the health expenditure, while the state has adapted to this situation 2 years later.
Lastly, the cointegration analysis showed significant cointegration between the public health expenditure and the total household expenditure and significant cointegration between the public health expenditure and the medical and pharmaceutical household expenditure.
In a review of the literature, the expansion of reliance of the health care funding on out-of-pocket payment has increased the financial risk and the hardship of the Greek households, which may disrupt their living conditions and create barriers to health care access. Cost-sharing policies should recognize the different social protection needs of the households. Another study also reported that the Greek HHE became more sensitive to the income changes after the introduction of the economic adjustment period.
Taking into account the fragile household health care behavior described herewith, on the basis that the health expenditure will sparingly begin to rise in the foreseeable future due to the severe impact of population aging, with simultaneous smaller share of active population, which means significant revenue shortfalls, it is recommended that the health policy planning should consider and obviate all these challenges by financing and designing social safety nets. |
Dietary Protein Consumption and the Risk of Type 2 Diabetes: A Systematic Review and Meta-Analysis of Cohort Studies
Recently, some studies have focused on the relationship between dietary protein intake and the risk of type 2 diabetes mellitus (T2DM), but the conclusions have been inconsistent. Therefore, in this paper, a systematic review and meta-analysis of cohort studies regarding protein consumption and T2DM risk are conducted in order to present the association between them. We searched the PubMed and Embase databases for cohort studies on dietary protein, high-protein food consumption and risk of T2DM, up to July 2017. A summary of relative risks was compiled by the fixed-effect model or random-effect model. Eleven cohort studies regarded protein intake and T2DM (52,637 cases among 483,174 participants). The summary RR and 95% CI (Confidence Interval) of T2DM was 1.12 (1.08-1.17) in all subjects, 1.13 (1.04-1.24) in men, and 1.09 (1.04-1.15) in women for total protein; 1.14 (1.09-1.19) in all subjects, 1.23 (1.09-1.38) in men, and 1.11 (1.03-1.19) in women for animal protein; 0.96 (0.88-1.06) in all subjects, 0.98 (0.72-1.34) in men, and 0.92 (0.86-0.98) in women for plant protein.We also compared the association between different food sources of protein and the risk of T2DM. The summary RR (Relative Risk) and 95% CI of T2DM was 1.22 (1.09-1.36) for red meat, 1.39 (1.29-1.49) for processed meat, 1.03 (0.89-1.17) for fish, 1.03 (0.64-1.67) for egg, 0.89 (0.84-0.94) for total dairy products, 0.87 (0.78-0.96) for whole milk, 0.83 (0.70-0.98) for yogurt, 0.74 (0.59-0.93) in women for soy. This meta-analysis shows that total protein and animal protein could increase the risk of T2DM in both males and females, and plant protein decreases the risk of T2DM in females. The association between high-protein food types and T2DM are also different. Red meat and processed meat are risk factors of T2DM, and soy, dairy and dairy products are the protective factors of T2DM. Egg and fish intake are not associated with a decreased risk of T2DM. This research indicates the type of dietary protein and food sources of protein that should be considered for the prevention of diabetes.Nutrients 2017, 9, 982 2 of 17 between dietary protein intake and the risk of T2DM. High-protein diets have shown beneficial effects on glucose homeostasis in some short-term trials[7,8]. Subsequent longitudinal studies have evaluated the associations between dietary protein intake and risk of T2DM as well as the type of dietary protein and risk of T2DM.A study focusing on the Mediterranean islands showed that animal protein consumption was associated with a higher prevalence of diabetes among the elderly, and a recommended range of protein from plant sources appears was seen to be considerably protective[9]. Some other publications have also reported an increased risk of T2DM with a high intake of total protein[10][11][12][13][14]and animal protein[10,[13][14][15], and a decreased risk of T2DM with a high intake of plant protein[14,15]. However, there have been no reports that how an association between T2DM and total protein[15][16][17][18], animal protein [16-18] and plant protein intake or a high risk of T2DM with a high intake of plant protein[10,13].The association between dietary protein and T2DM is still debated.In the present study, we conducted a systematic review and meta-analysis of cohort studies to clarify the association of protein consumption with risk of T2DM. In order to provide a better dietary instruction for the lay public, we also conducted a systematic review and meta-analysis of cohort studies to study the association between different kinds of high-protein food and the risk of T2DM.
# Introduction
Type 2 diabetes is rapidly increasing in the world. Ninety percent of patients with diabetes have type 2 diabetes. Type 2 diabetes patients are at increased risk of cardiovascular disease, neuropathies, nephropathies, gangrene, and leg ulcers. From the date of diagnosis, patients with T2DM have to face at least eight years of economic burden.
Evidence suggests that dietary factors may influence the risk of T2DM. Many previous studies have focused on dietary macronutrient intake associated with diabetes risk, but the main content of those studies was carbohydrates and fats. Recently, some studies focused on the relationship
# Results
## Dietary protein intake and risk of t2dm
We included 11 cohort studies in the analysis, Tables S1 and S2). Six of the studies were from the Unite States, three from Europe, one from Asia, one from Melboume, one from Finland.
Nutrients 2017, 3 of 17 follow-up, the methods of diet exposure assessment, the RRs and 95% CI, and the adjustment factors (Tables S1-S12).
# Statistical methods
We used effect models to calculate the summary RRs and 95% confidence to compare the highest dietary protein consumption with the lowest dietary protein consumption. Two-sided p < 0.05 was considered statistically significant. The black square in the forest plots represents the weight contribution of every study. In order to evaluate the extent of variability, the I 2 -test statistic was adapted to estimate the heterogeneity.When I 2 < 50% and p > 0.05, there was no heterogeneity; we used the fixed-effect model. The random-effect model was selected when I 2 > 50% or p < 0.05. We used the Egger linear regression test and Begg rank correlation test to search for publication bias. When p > 0.05, there was no publication bias. Comprehensive Meta-Analysis V2 was employed for data analysis.
# Results
## Dietary protein intake and risk of t2dm
We included 11 cohort studies in the analysis, Tables S1 and S2). Six of the studies were from the Unite States, three from Europe, one from Asia, one from Melboume, one from Finland. Total Protein Eleven cohort studiesresearched the association between total protein intake and the risk of T2DM, included 52,637 cases among 483,174 participants. The summary RR and 95% CI for high vs. low values of all studies was 1.12 (1.08-1.17)Total Protein Eleven cohort studiesresearched the association between total protein intake and the risk of T2DM, included 52,637 cases among 483,174 participants. The summary RR and 95% CI for high vs. low values of all studies was 1.12 (1.08-1.17)Total protein and type 2 diabetes relative risks(RRs) for (a) the highest vs. the lowest intake in all subjects and (b) the highest vs. the lowest intake in men (c) the highest vs. the lowest intake in women. The RR of each study is represented by a square, 95% CI are represented by the horizontal lines, and the diamond represents the estimate and its 95% CI.
Animal Protein Nine cohort studiesresearched the association between animal protein intake and the risk of T2DM. This included 31,557 cases among 380,689 participants. The summary RR and 95% CI for high vs. low values of all studies was 1.14 (1.09-1.19) ( Plant Protein Nine cohort studiesresearched the association between plant protein and the risk of T2DM, included 31,817 cases among 381,879 participants. The summary RR and 95% CI for high vs. low values of all studies was 0.96 (0.88-1.06), (I 2 = 59.01, p = 0.07) in all subjects. We used a sensitivity analysis to exclude the most influential studies: the summary RR and 95% CI ranged from 0.92 (0.87-0.98) when the European men's studywas excluded to 0.98 (0.87-1.10) when the USA men's studywas excluded. The heterogeneity was partly because of the European men's studyand when we excluded this study, there was moderate heterogeneity (I 2 = 14.85, p = 0.31). The summary RR and 95% CI was 0.98 (0.72-1.34)(I 2 = 78.57, p = 0.003) in men. A sensitivity analysis was used to exclude the most influential studies: the summary RR and 95% CI ranged from 0.88 (0.76-1.02) when the European studywas excluded to 1.07 (0.75-1.52) when the Finnish men's studywas excluded. The heterogeneity in men still existed Total protein and type 2 diabetes relative risks(RRs) for (a) the highest vs. the lowest intake in all subjects and (b) the highest vs. the lowest intake in men (c) the highest vs. the lowest intake in women. The RR of each study is represented by a square, 95% CI are represented by the horizontal lines, and the diamond represents the estimate and its 95% CI.
Animal Protein Nine cohort studiesresearched the association between animal protein intake and the risk of T2DM. This included 31,557 cases among 380,689 participants. The summary RR and 95% CI for high vs. low values of all studies was 1.14 (1.09-1.19) ( Plant Protein Nine cohort studiesresearched the association between plant protein and the risk of T2DM, included 31,817 cases among 381,879 participants. The summary RR and 95% CI for high vs. low values of all studies was 0.96 (0.88-1.06), (I 2 = 59.01, p = 0.07) in all subjects. We used a sensitivity analysis to exclude the most influential studies: the summary RR and 95% CI ranged from 0.92 (0.87-0.98) when the European men's studywas excluded to 0.98 (0.87-1.10) when the USA men's studywas excluded. The heterogeneity was partly because of the European men's studyand when we excluded this study, there was moderate heterogeneity (I 2 = 14.85, p = 0.31). The summary RR and 95% CI was 0.98 (0.72-1.34)(I 2 = 78.57, p = 0.003) in men. A sensitivity analysis was used to exclude the most influential studies: the summary RR and 95% CI ranged from 0.88 (0.76-1.02) when the European studywas excluded to 1.07 (0.75-1.52) when the Finnish men's studywas excluded. The heterogeneity in men still existed after sensitivity analysis, partly because there were only four cohort studies about the association between plant protein intake and the risk of T2DM in men. The summary RR and 95% CI was 0.92 (0.
## High-protein food and risk of t2dm
We also conducted a systematic review and meta-analysis of cohort studies to clarify the association with high dietary protein food consumption and the risk of T2DM.
Red Meat Thirteen cohort studiesresearched the association between red meat consumption and the risk of T2DM, Tables S3 and S4).The summary RR and 95% CI for high vs. low red meat consumption was 1.22 (1.09-1.36)(I 2 = 51.11, p = 0.01). We used a sensitivity analysis to exclude the most influential studies: the summary RR and 95% CI ranged from1.20 (1.07-1.35) when the U.S. Studywas excluded to 1.26 (1.15-1.38) when the Chinese studywas excluded. The heterogeneity was partly because of the Chinese study, and when we excluded this study, there was moderate heterogeneity (I 2 = 27.68, p = 0.17). . Plant protein and type 2 diabetes RRs for (a) the highest vs. the lowest intake in all subjects and (b) the highest vs. the lowest intake in men (c) the highest vs. the lowest intake in women. The RR of each study is represented by a square, 95% CI are represented by the horizontal lines, and the diamond represents the estimate and its 95% CI.
## High-protein food and risk of t2dm
We also conducted a systematic review and meta-analysis of cohort studies to clarify the association with high dietary protein food consumption and the risk of T2DM.
Red Meat Thirteen cohort studiesresearched the association between red meat consumption and the risk of T2DM, Tables S3 and S4).The summary RR and 95% CI for high vs. low red meat consumption was 1.22 (1.09-1.36) (Figure5a) (I 2 = 51.11, p = 0.01). We used a sensitivity analysis to exclude the most influential studies: the summary RR and 95% CI ranged from1.20 (1.07-1.35) when the U.S. Studywas excluded to 1.26 (1.15-1.38) when the Chinese studywas excluded. The heterogeneity was partly because of the Chinese study, and when we excluded this study, there was moderate heterogeneity (I 2 = 27.68, p = 0.17). Fish Nine cohort studiesresearched the association between fish consumption and the risk of T2DM, Tables S5 and S6), the summary RR and 95% CI for high vs. low values of fish consumption was 1.03 (0.89-1.17)(I 2 = 79.71, p < 0.001). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from0.99 (0.89-1.10) when the American studywas excluded to 1.50 (0.92-1.20) when the Japanese studywas excluded. The heterogeneity was partly because of the Japanese study, and when we excluded this study, there was moderate heterogeneity (I 2 = 79.41, p < 0.001). processed meat and type 2 diabetes RRs for the highest vs. the lowest intake in all subjects. The RR of each study is represented by a square, 95% CI are represented by the horizontal lines, and the diamond represents the estimate and its 95% CI.
Processed Meat Eleven cohort studiesresearched the association between processed meat consumption and the risk of T2DM, Tables S3 and S4), the summary RR and 95% CI for high vs. low of processed meat consumption was 1. Fish Nine cohort studiesresearched the association between fish consumption and the risk of T2DM, Tables S5 and S6), the summary RR and 95% CI for high vs. low values of fish consumption was 1.03 (0.89-1.17)(I 2 = 79.71, p < 0.001). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from0.99 (0.89-1.10) when the American studywas excluded to 1.50 (0.92-1.20) when the Japanese studywas excluded. The heterogeneity was partly because of the Japanese study, and when we excluded this study, there was moderate heterogeneity (I 2 = 79.41, p < 0.001). Egg Five cohort studiesresearched the association between egg consumption and the risk of T2DM, Tables S7 and S8), the summary RR and 95% CI for high vs. low of egg consumption was 1.03 (0.64-1.67)(I 2 = 91.12, p < 0.001). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from 0.97 (0.49-1.88) when the Lithuanian studywas excluded to 1.57 (1.30-1.89) when the Finnish studywas excluded. The heterogeneity was partly because of the Finnish study, and when we excluded this study, there was moderate heterogeneity (I 2 = 69.43, p = 0.06).
## Figure 7.
Egg and type 2 diabetes RRs for the highest vs. the lowest intake in all subjects. The RR of each study is represented by a square, 95% CI are represented by the horizontal lines, and the diamond represents the estimate and its 95% CI.
## Total dairy product consumption
Eleven cohort studiesresearched the association between total dairy product consumption and the risk of T2DM, Tables S9 and S10), the summary RR and 95% CI for high vs. low total dairy product consumption was 0. Egg Five cohort studiesresearched the association between egg consumption and the risk of T2DM, Tables S7 and S8), the summary RR and 95% CI for high vs. low of egg consumption was 1.03 (0.64-1.67)(I 2 = 91.12, p < 0.001). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from 0.97 (0.49-1.88) when the Lithuanian studywas excluded to 1.57 (1.30-1.89) when the Finnish studywas excluded. The heterogeneity was partly because of the Finnish study, and when we excluded this study, there was moderate heterogeneity (I 2 = 69.43, p = 0.06). Egg Five cohort studiesresearched the association between egg consumption and the risk of T2DM, Tables S7 and S8), the summary RR and 95% CI for high vs. low of egg consumption was 1.03 (0.64-1.67)(I 2 = 91.12, p < 0.001). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from 0.97 (0.49-1.88) when the Lithuanian studywas excluded to 1.57 (1.30-1.89) when the Finnish studywas excluded. The heterogeneity was partly because of the Finnish study, and when we excluded this study, there was moderate heterogeneity (I 2 = 69.43, p = 0.06).
## Total dairy product consumption
Eleven cohort studiesresearched the association between total dairy product consumption and the risk of T2DM, Tables S9 and S10), the summary RR and 95% CI for high vs. low total dairy product consumption was 0. TotalDairyProduct Consumption Eleven cohort studiesresearched the association between total dairy product consumption and the risk of T2DM, Tables S9 and S10), the summary RR and 95% CI for high vs. low total dairy product consumption was 0. Whole Milk Seven cohort studiesresearched the association between whole milk consumption and risk of T2DM, Tables S9 and S10), the summary RR and 95% CI for high vs. low of fat dairy consumption was 0.87 (0.78-0.96)(I 2 = 52.20, p = 0.01). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from 0.85 (0.76-0.94) when the American studywas excluded to 0.89 (0.80-0.99) when the Japanese studywas excluded. The heterogeneity was partly because of the American study, and when we excluded this study, there was moderate heterogeneity (I 2 = 46.87, p = 0.04).
Yogurt Seven cohort studiesresearched the association between yogurt consumption and the risk of T2DM (Figure1, Tables S9 and S10).The summary RR for high vs. low yogurt consumption was 0.83 (0.70-0.98)(I 2 = 62.06,p = 0.01). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from0.81 (0.67-0.97) when the Japanese studywas excluded to 0.88 (0.76-1.00) when USA studywas excluded. The heterogeneity was partly because of the Japanese study, and when we excluded this study there was moderate heterogeneity (I 2 = 40.61, p = 0.11).
Soy Eight cohort studiesresearched the association between legume consumption and the risk of T2DM, Tables S11 and S12), the summary RR for high vs. low soy consumption was 0.87 (0.74-1.01)(I 2 = 86.57, p < 0.01). We used a sensitivity analysis to exclude the Whole Milk Seven cohort studiesresearched the association between whole milk consumption and risk of T2DM, Tables S9 and S10), the summary RR and 95% CI for high vs. low of fat dairy consumption was 0.87 (0.78-0.96)(I 2 = 52.20, p = 0.01). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from 0.85 (0.76-0.94) when the American studywas excluded to 0.89 (0.80-0.99) when the Japanese studywas excluded. The heterogeneity was partly because of the American study, and when we excluded this study, there was moderate heterogeneity (I 2 = 46.87, p = 0.04).
Yogurt Seven cohort studiesresearched the association between yogurt consumption and the risk of T2DM, Tables S9 and S10).The summary RR for high vs. low yogurt consumption was 0.83 (0.70-0.98)(I 2 = 62.06, p = 0.01). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from 0.81 (0.67-0.97) when the Japanese studywas excluded to 0.88 (0.76-1.00) when USA studywas excluded. The heterogeneity was partly because of the Japanese study, and when we excluded this study there was moderate heterogeneity (I 2 = 40.61, p = 0.11).
Soy Eight cohort studiesresearched the association between legume consumption and the risk of T2DM, Tables S11 and S12), the summary RR for high vs. low soy consumption was 0.87 (0.74-1.01)(I 2 = 86.57, p < 0.01). We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from0.82 (0.68-0.98) when the American studywas excluded to 0.93 (0.81-1.06) when the Chinese studywas excluded. The heterogeneity was partly because of the American study, and when we excluded this study, there was moderate heterogeneity (I 2 = 69.34, p < 0.01). However, the summary RR for high vs. low soy was 0.74 (0.59-0.93), (I 2 = 82.09, p < 0.001) in women. We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from0.66 (0.49-0.90) when the American studywas excluded to 0.81 (0.65-1.00) when the Chinese studywas excluded. The heterogeneity was partly because of the American study, and when we excluded this study, there was moderate heterogeneity (I 2 = 0.00, p = 0.67).
Nutrients 2017, 9, 982 9 of 17 most influential studies: the summary RR ranged from0.82 (0.68-0.98) when the American studywas excluded to 0.93 (0.81-1.06) when the Chinese studywas excluded. The heterogeneity was partly because of the American study, and when we excluded this study, there was moderate heterogeneity (I 2 = 69.34, p < 0.01). However, the summary RR for high vs. low soy was 0.74 (0.59-0.93), (I 2 = 82.09, p < 0.001) in women. We used a sensitivity analysis to exclude the most influential studies: the summary RR ranged from0.66 (0.49-0.90) when the American studywas excluded to 0.81 (0.65-1.00) when the Chinese studywas excluded. The heterogeneity was partly because of the American study, and when we excluded this study, there was moderate heterogeneity (I 2 = 0.00, p = 0.67). the highest vs. the lowest intake in women. The RR of each study is represented by a square, 95% CI are represented by the horizontal lines, and the diamond represents the estimate and its 95% CI.
## Publication bias
No significant publication bias was detected in the Begg-Mazumdar's test and Egger's test of our meta-analyses, provided in.. Soy and type 2 diabetes RRs for (a) the highest vs. the lowest intake in all subjects and (b) the highest vs. the lowest intake in women. The RR of each study is represented by a square, 95% CI are represented by the horizontal lines, and the diamond represents the estimate and its 95% CI.
## Publication bias
No significant publication bias was detected in the Begg-Mazumdar's test and Egger's test of our meta-analyses, provided in.
# Discussion
According to the results of this meta-analysis, the intake of total protein and animal protein was associated with a high risk of T2DM both in males and females. The intake of plant protein was associated with low risk of T2DM in females, but not in males. In high animal protein food, red meat, and processed meat were associated with a high risk of T2DM in all subjects, while total dairy products, low-fat dairy, and yogurt were associated with a low risk of T2DM in all subjects, and egg and fish were not associated with a decreased risk of T2DM. In high plant protein food, soy was associated with a low risk of T2DM in females.
Higher intake of dietary protein is often associated with lifestyles, including physical activity, body weight, smoking, drinking. For example, we already know that overweight and obesity are risk factors for T2DM, and a meta-analysis showed that each unit increase of BMI would increase the risk for T2DM by approximately 20%. In our meta-analysis, most studies were adjusted for known influencing factors, including age, BMI (Body Mass Index), smoking, physical activity, alcohol consumption, energy intake, family history of T2DM and menopausal status (among women).
The statistical power of the results could be significantly increased as the number of studies and the sample size increase, but it could also lead to heterogeneity. Some heterogeneity was due to different participants' characteristics and regions, and different dietary assessment methods. Thus, heterogeneity is usually used to explain the study characteristics and is difficult to interpret. In our study, the heterogeneities of total protein and animal protein were in the acceptable range, but the heterogeneity of plant protein was outside of the range. We found that the European studycontributed to the heterogeneity. When this study was excluded, the heterogeneities of both the overall and subgroup analyses were much lower. The reason that the European studyhad inconsistent results compared with other studies was not clear, but it could partly be due to the participants in this study having a lower plant protein intake than other studies. For red meat and processed meat, we found that the Chinese studycontributed to the heterogeneity. When this study is excluded, the heterogeneities were much lower. The reason that the Chinese studyhad inconsistent results may be due to the participants in this study having a lower meat intake than other participants. For fish, we found that the Japanese studycontributed to the heterogeneity, but the heterogeneity was not moderate when it was excluded, which means the association between fish and T2DM needs further refinement. For egg, we found that the Finnish studycontributed to the heterogeneity. When this study is excluded, the heterogeneity was much lower. The reason why the Finnish studyhad inconsistent results compared with other studies was not clear, but it could partly be due to the participants in this study being older than the other studies. For whole milk, we found that the American studycontributed to the heterogeneity, and the heterogeneity was moderate when it was excluded. The reason why the American studyhad heterogeneity may be due to the milk intake of black women in America being different from the participants of other studies'. For yogurt, the Japanese studycontributed to the heterogeneity. The reason why the Japanese studyhad inconsistent results was not clear, but it could partly be due to the female participants in this study being older than other studies.
As meta-analysis is based on published studies, the publication bias affect is inevitable. It is particularly important to evaluate publication bias. In our meta-analysis, we used the Egger linear regression test and Begg rank correlation test to determine publication bias, and we found that there was no publication bias in our study.
Dietary protein and amino acids are involved in the modulations of insulin sensitivity and glucose metabolism. However, the results from human studies were still inconsistent. Some studies showed that high intake of dietary protein had negative effects on glucose homeostasis by facilitating insulin resistance and increasing gluconeogenesis. Amino acid signaling may facilitate insulin resistance, by activation of the mammalian target of rapamycin (mTOR), a nutrient sensor that operates a detrimental feedback loop toward insulin receptor substrate 1 signaling. Moreover, amino acids may also inhibit glucose uptake through phosphorylation of downstream factors of the insulin signaling cascade by the translation initiation factor serine-kinase-6-1. On the other hand, in vivo and in vitro studies also demonstrated that amino acids play a beneficial role in glucose homeostasis by modulating insulin action on hepatic glucose production and muscle glucose transport, secretion of glucagon and insulin, as well as various tissues gene and protein expression. One of the possible mechanisms that might explain this was that higher protein reduced the amount of carbohydrate intake under isoenergy conditions and thus a smaller amount of glucose was absorbed after ingestion of the meals, with the consequence of a reduced store of glycogen and, thus, a decrease in glycogenolysis rate. The other possible causal mechanism was amino acids stimulating the insulin secretion to intervene in the glucose metabolism and serve as substrates for gluconeogenesis; thus, increased gluconeogenesis could stimulate insulin secretion, which might prevent hyperglycemia.Additionally, some scientists think that different qualities rather than quantities of proteins play a more important role in insulin resistance. Our study supported the hypothesis that animal proteins caused a high risk of T2DM in males and females, and plant proteins were protection factors of T2DM in females. We can find some support for this from the literature. The abundance in certain amino acids are different between animal proteins and plant proteins. This may contribute to the different effects between them on the risk of T2DM. Typically, plant protein contains lower levels of the branched chain amino acids leucine, isoleucine and valine and of the sulfur amino acid methionine as compared with animal proteins. Branched chain amino acids and higher methionine intake have been associated with insulin resistance and type 2 diabetes. In addition, dietary glycine is also mainly consumed from animal-based foods and some cohort studies have shown that glycine was positively associated with T2DM, and hypertension. On the other hand, dietary glutamic acid, an amino acid that is mainly consumed from plant protein was found to be inversely associated with risks of hypertension and arterial stiffness. So far, three intervention studies have compared the effects of animal protein with plant protein meals on glycaemic variables in people with T2DM, but they were obtained from three different results. This should be further investigated in future studies.
From the current literature available, the inconsistency in association between plant protein and T2DM was probably due to gender difference. The negative association between plant protein or soy and T2DM was observed mainly in women, while most of the studies in men found null results. Therefore, results from all subjects without considering gender difference were different and the proportion of women in the study may influence the results. However, the exact mechanism is unclear.
In order to further refine and compare the association between different food sources of protein and T2DM, and facilitate dietary guidance, we have analyzed the relationship between different high-protein food and the risk of type 2 diabetes. We found that different high-protein foods play different roles in T2DM, even if they are all animal-based foods. Our results indicated that the intake of red meat and processed meat are risk factors for T2DM. They are also positively associated with weight gain, stroke, coronary heart diseaseand mortality.First, the increased meat protein may increase iron load, which was associated with the increased risk of T2DM. Moreover, the other nutrients in red and processed meat, including nitrites and advanced glycation end products, were also thought to mediate the association between meat intake and the risk of T2DM.The relationship between egg consumption and T2DM was not clear. Some studies have shown that egg intake was associated with a lower risk of T2DM.Some research showed that egg consumption was positively associated with the risk of T2DM in our study, and this result was supported by the AHA dietary guidelines which advise restricted egg consumption in adults for preventing cardiometabolic diseases.We found that total dairy products, whole milk, and yogurt intake were protective factors for T2DM. Some studies showed that milk proteins, like whey protein, may enhance satiety and reduce risk factors for T2DM. The calcium and vitamin D in milk and its products may also contribute to its beneficial effects on T2DM.In our study, fish consumption was not associated with decreased risk of T2DM. This result may partly relate on the increase in plasma selenium level with the increment of fish intake, which may increase the risk of diabetes.The relationship between fish intake and the risk of T2DM needs further refinement. For high plant protein-based foods, there was a negative association between soy consumption and the risk of T2DM in this study. Soy protein may inhibit insulin secretion from pancreatic β cells or inhibit lipogenesis and enhance lipolysis in the adipose and liver to reduce adiposity.This protective effect may also be associated with biologically active ingredients such as phytoestrogen in soybeans.
Our meta-analysis also had limitations. First, publications into our research were adjusted for BMI, but some studies had measurement errors because of self-reporting of height and weight, resulting in the BMI relying on self-reporting, which could lead to confounding results. Second, some important factors that influence T2DM such as fiber, lipids, and carbohydrates, were only adjusted in some of these studies, which may also lead to confounding results. Additionally, limitations might be due to temporal bias. Studies with longer follow-up might beless influenced by temporal bias. In our studies, the follow-up period of each research study was different, so the temporal bias might impact the association between dietary protein intake and the risk of T2DM.
# Conclusions
In summary, we found that total protein and animal protein consumption were the risk factors for T2DM, and plant protein was the protective factor for T2DM in women, but not in men. We also found that different high-protein foods have a different effect on T2DM risk, even if they all belong to animal proteins. These results underline the significance of taking into account what kind of dietary protein and food sources of protein are recommended for the prevention of diabetes.
Supplementary Materials: The following are available online at www.mdpi.com/2072-6643/9/9/982/s1,: Characteristics of participants and follow-up in included studies of protein consumption in relation to the risk of T2DM, : Prospective cohort studies of protein consumption and type 2 diabetes, : Characteristics of participants and follow-up in included studies of meat consumption in relation to the risk of T2DM, : Prospective cohort studies of meat consumption and type 2 diabetes, : Characteristics of participants and follow-up in included studies of fish consumption in relation to the risk of T2DM, : Prospective cohort studies of fish consumption and type 2 diabetes, : Characteristics of participants and follow-up in included studies of egg consumption in relation to risk of T2DM, :Prospective cohort studies of egg consumption and type 2 diabetes, : Characteristics of participants and follow-up in included studies of dairy consumption in relation to the risk of T2DM,: Prospective cohort studies of dairy consumption and type 2 diabetes, Table S11: Characteristics of participants and follow-up in included studies of soy consumption in relation to the risk of T2DM,: Prospective cohort studies of soy consumption and type 2 diabetes. |
Emergence of self-affine surfaces during adhesive wear
Friction and wear depend critically on surface roughness and its evolution with time. An accurate control of roughness is essential to the performance and durability of virtually all engineering applications. At geological scales, roughness along tectonic faults is intimately linked to stick-slip behaviour as experienced during earthquakes. While numerous experiments on natural, fractured, and frictional sliding surfaces have shown that roughness has self-affine fractal properties, much less is known about the mechanisms controlling the origins and the evolution of roughness. Here, by performing long-timescale molecular dynamics simulations and tracking the roughness evolution in time, we reveal that the emergence of self-affine surfaces is governed by the interplay between the ductile and brittle mechanisms of adhesive wear in three-body contact, and is independent of the initial state.
T he roughness of surfaces is a key parameter for all tribology-related phenomena, namely friction, wear, and lubrication. This was already clear to the pioneers of tribology, from Da Vinci 1 to Coulomb 2 , who linked friction and surface morphology. Their findings were generalized in the past century by Bowden and Tabor 3 , who studied the effects of adhesion and introduced the concept of real contact area. More recently, experimental evidence has shown that both natural and manufactured surfaces are self-affine over many scales [bib_ref] Constant dimensionality of fault roughness from the scale of micro-fractures to the..., Renard [/bib_ref] [bib_ref] Surface topography as a nonstationary random process, Sayles [/bib_ref] [bib_ref] Fractal characterization and simulation of rough surfaces, Majumdar [/bib_ref] [bib_ref] On the nature of surface roughness with application to contact mechanics, sealing,..., Persson [/bib_ref] [bib_ref] Fractal character of fracture surfaces of metals, Mandelbrot [/bib_ref] [bib_ref] Failure of heterogeneous materials: a dynamic phase transition?, Bonamy [/bib_ref]. In geophysics, the fault roughness decreases with slip [bib_ref] Faults smooth gradually as a function of slip, Brodsky [/bib_ref] [bib_ref] Evolution of slip surface roughness through shear, Davidesko [/bib_ref] and is related to the fault's strength [bib_ref] Constraints from fault roughness on the scale-dependent strength of rocks, Brodsky [/bib_ref] [bib_ref] Nanoscale roughness of natural fault surfaces controlled by scale-dependent yield strength, Thom [/bib_ref] and deformation mechanisms [bib_ref] The minimum scale of grooving on faults, Candela [/bib_ref]. Also, for various engineering surfaces, the roughness is found to converge upon rubbing to a steady-state value. New surfaces generated by tensile fracture are well known to be self-affine, too [bib_ref] Fractal character of fracture surfaces of metals, Mandelbrot [/bib_ref] [bib_ref] Failure of heterogeneous materials: a dynamic phase transition?, Bonamy [/bib_ref] , and different universal Hurst exponents have been linked to different damage mechanisms [bib_ref] Failure of heterogeneous materials: a dynamic phase transition?, Bonamy [/bib_ref] [bib_ref] Turbulent fracture surfaces: a footprint of damage percolation?, Vernède [/bib_ref]. However, the physical origins of these observations are still unclear [bib_ref] Research focus: connections between fault roughness, dynamic weakening, and fault zone structure, Marone [/bib_ref].
Several theoretical surface growth models have been developed to explain roughness evolution [bib_ref] Universality classes in constrained crack growth, Gjerden [/bib_ref]. Nonetheless, the generalization of simple diffusion models to the complex case of rubbing surfaces [bib_ref] Diffusion as a model of formation and development of surface topography, Schargott [/bib_ref] still misses significant mechanisms, like the concurrence of ductile and brittle mechanisms when working the surface. Continuum numerical models are also limited, as they struggle to capture all the several intertwined non-linearities, such as contact, adhesion, plasticity, and fracture. Molecular dynamics simulations can capture the aforementioned non-linearities and atomic-scale mechanisms, but the computational cost is high [bib_ref] Probing the limits of metal plasticity with molecular dynamics simulations, Zepeda-Ruiz [/bib_ref] [bib_ref] Trillion-atom molecular dynamics becomes a reality, Germann [/bib_ref] and mechanisms that take place at long time and length scales cannot be modelled.
To overcome the scale limitations in atomistic simulations, simple model potentials have recently been developed [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] , which have proved to be able to capture the ductile-to-brittle transition taking place upon collision of surface asperities in adhesive wear processes [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref]. When two asperities come into contact, three possible mechanisms can take place [bib_ref] Nanoscale wear as a stress-assisted chemical reaction, Jacobs [/bib_ref] : atom by atom removal in the light load and low adhesion limit [bib_ref] Nanoscale wear as a stress-assisted chemical reaction, Jacobs [/bib_ref] [bib_ref] Ultralow nanoscale wear through atom-by-atom attrition in silicon-containing diamond-like carbon, Bhaskaran [/bib_ref] [bib_ref] Adhesion suppresses atomic wear in singleasperity sliding, Yang [/bib_ref] , and alternatively ductile deformation [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] [bib_ref] Liquid-like tribology of gold studied by in situ TEM, Merkle [/bib_ref] or brittle fracture [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] [bib_ref] Contact and rubbing of flat surfaces, Archard [/bib_ref] [bib_ref] Method for characterizing nanoscale wear of atomic force microscope tips, Liu [/bib_ref] [bib_ref] Preventing nanoscale wear of atomic force microscopy tips through the use of..., Liu [/bib_ref] of the asperities at larger loads and moderate-to-high adhesion. Our investigations are conducted in the latter conditions, where the mechanism depends on the size of the junction formed by the two contacting asperities [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] [bib_ref] On the debris-level origins of adhesive wear, Aghababaei [/bib_ref] [bib_ref] Asperity-level origins of transition from mild to severe wear, Aghababaei [/bib_ref] [bib_ref] A mechanistic understanding of the wear coefficient: from single to multiple asperities..., Frérot [/bib_ref].
Here, we perform long-time molecular dynamics simulations of rubbing surfaces, investigating different initial conditions. Thanks to the adopted method, the ductile-to-brittle transition occurs spontaneously (that is, it is not enforced), permitting us to explicitly capture the debris particle formation [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] [bib_ref] On the debris-level origins of adhesive wear, Aghababaei [/bib_ref] [bib_ref] Asperity-level origins of transition from mild to severe wear, Aghababaei [/bib_ref] and the subsequent transition to a three-body configuration. We thus have a competition between the brittle fracture mechanism that roughens the surfaces and the ductile one that flattens them. Once the debris particle is formed, these mechanisms take place at the contact interface between the particle and the surfaces. We find that the resulting worn surfaces are self-affine and characterized by the same statistics independently of the initial state, within the investigated range. Our results also show that the debris particle wear rate is lower in the three-body configuration, i.e. after running-in.
# Results
Simulation setup. The investigation consists of a large set of twodimensional (2D) molecular dynamics simulations over a long duration to maximize the chances of observing a steady state for some surface feature, as observed experimentally. The analysed condition is dry sliding of one surface on top of the other, both described by the same model interaction potentials, at constant temperature and constant sliding velocity (see and Methods). The simulations differ from one another in bulk material properties, surface topography, temperature, and system size (see and Methods for the full details). Throughout the article, quantities are measured in reduced units, the fundamental quantities being the equilibrium bond length r 0 , the bond energy ε at 0 K, and the particle mass m.
Recently, a critical length scale d * was found to govern the ductile-to-brittle transition in adhesive wear [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] : when two asperities collide, if the junction size d formed by the asperities is larger than d * , the asperities break and a debris particle is formed, otherwise the asperities deform plastically . The critical length scale is expressed as d à ¼ λ Á ΔwG=τ 2 j , where τ j is the junction shear strength, G is the shear modulus of the material, Δw is the fracture energy, and λ is a geometrical factor. In our simulations, the materials are described by interaction potentials, which we characterize by the maximum stress τ sf on the generalized stacking fault curve at zero temperature [bib_ref] How to affect stacking fault energy and structure by atom relaxation, Cai [/bib_ref] [bib_ref] Stacking fault energies and slip in nanocrystalline metals, Van Swygenhoven [/bib_ref] : the lower τ sf , the more ductile the material is and the larger its critical length scale d * . We adopt two different initial surface morphologies: single asperity against single asperity and selfaffine surface against self-affine surface . In all cases, the initial contact takes place in a two-body configuration, that is the two surfaces come directly into contact with one another. In the single-asperity setup -d, simulations S1-S7, where S stands for single asperity, cf. , each surface is atomistically flat, except for one semicircular asperity. By sliding the top surface, its asperity collides with the asperity of the opposing surface and forms a junction. The size of the initial asperities is chosen large enough for the junction size d to be greater than the critical size d * and to create a debris particle , which is then constrained to roll between the two surfaces. In the case of initially self-affine surfaces rubbing against one another -h, i-l, m-p, simulations R1 to R3, H1, G1 and G2, where R stands for rough, H for heterogeneous material, and G for grain boundaries, cf. , we construct the two surfaces with the same Hurst exponent and same root mean square roughness, but we do not control the position of the first contact, nor its size. The first contact involves several small asperities in both surfaces, which deform plastically until they form a junction of size d > d * and the surfaces break . In this case, several asperities come into contact with the newly formed debris particle and interact with it in a brittle or ductile fashion, according to the size of the contact that is developed each time. The initial stage is even more complex when the material is heterogeneous -l, m-p, simulations H1, G1, and G2, cf. . We prepared such a case by geometrically dividing the material into irregularly shaped sub-regions (randomly distributed both in size and in position), and randomly assigning one of the two potentials within each sub-region. The two potentials differ in their critical size d * , so that two different critical length scales coexist in the system. As a result of this mixture, the overall surfaces are heterogeneous in terms of inelastic behaviour (with half of the tiles being relatively more brittle than the other half). Upon asperity collision, the critical length scale for the ductile-tobrittle transition in the case of the mixture is then no longer well defined. In this case, the surfaces favour cracks within the least tough material (or at weak, heterogeneous interfaces), and plastic deformation within the toughest one (cf. Supplementary Movie 1).
After this initial stage, and independently of the original geometrical setup and of the heterogeneity of the material, the surfaces reach a state where the debris particle continuously rolls between them and works them , h, l, p). The contact now takes place in a three-body configuration, the third body being the debris particle that separates the two surfaces (i.e. the first bodies). The surfaces undergo both brittle and ductile deformation and material transfer takes place both ways: from the debris particle to the surfaces and vice versa. This interplay with the third body allows for a continuous reworking of the surfaces, which appear to be self-affine whenever a steady-state roughness is reached. Remarkably, as we explain below, this self-affine morphology is independent of the initial conditions investigated. The case of heterogeneous materials is particular: each of the two materials is characterized by a different critical length scale and the effect of this on the critical length scale of the mixture is still unknown. From our results, no trace of this heterogeneity is found in the scaling of the self-affine morphology of the worn surfaces. In all cases, the inclusion of the ductile-to-brittle transition within the modelled length scales is fundamental to capture the self-affine nature of the surfaces. When it is not included, asperity collision never leads to the formation of loose debris particles and surfaces always smoothen [bib_ref] Nanoscale sliding friction phenomena at the interface of diamond-like carbon and tungsten, Stoyanov [/bib_ref] [bib_ref] Anisotropic mechanical amorphization drives wear in diamond, Pastewka [/bib_ref] [bib_ref] Energy filtering transmission electron microscopy and atomistic simulations of tribo-induced hybridization change..., De Barros [/bib_ref] [bib_ref] Molecular dynamics simulation of severe adhesive wear on a rough aluminum substrate, Zhong [/bib_ref] [bib_ref] Large-scale molecular dynamics simulations of wear in diamond-like carbon at the nanoscale, Sha [/bib_ref] [bib_ref] Dry sliding contact between rough surfaces at the atomistic scale, Spijker [/bib_ref] [bib_ref] Simulations of atomic-scale sliding friction, Sorensen [/bib_ref].
Self-affine description. For a complete description of the surfaces, we investigate their power spectral density (PSD) Φ per unit length, as it contains information about the contribution of all the length scales involved. Self-affine surfaces are in fact characterized by an anisotropic scale transformation. This means that their heights h(x) scale differently than the horizontal position x, and the scaling relation is
[formula] h(λx)~λ H h(x), [/formula]
where λ is the scaling factor, and 0 < H < 1 is the Hurst exponent [bib_ref] Self-affine fractals and fractal dimension, Mandelbrot [/bib_ref] (see Methods for a more detailed discussion about H). In other words, magnifying the x axis by a factor λ will produce a magnification of the heights h(x) by a factor λ H . An important consequence of this relation is that if the statistics of a self-affine surface are known at a given scale, they can be extended to all the other scales by means of the Hurst exponent H. The PSD of self-affine surfaces displays a power-law behaviour Φ(q)~q −α (where q is the wavevector) and the Hurst exponent can be expressed in terms of the power-law exponent α 46-48 : in the case of one-dimensional (1D) surfaces, H = (α − 1)/2 (see Methods for more details).
The results of the analysis are displayed in and , which show the normalized PSD and the height-height correlation function averaged over several time steps for the worn top and bottom surfaces of different simulations, respectively. The surfaces are sampled during the steady state, where their roughness (expressed as the root mean square of heights) fluctuates around a stabilized value and their profile can be assumed to be stationary. This is the first time, to the best of our knowledge, that such a trend in the roughness evolution is numerically reproduced. It is in fact known from experimental evidence that surfaces undergo large roughness variations in the early stage of the wear process, before settling around a steady-state value [bib_ref] Transient wear of machine parts, Queener [/bib_ref] , as reproduced by our simulations (cf. . The PSDs in and collapse around an average value of the Hurst exponent H = 0.7, the lowest value being H = 0.6 and the highest H = 0.8, irrespective of the initial geometry, the material, the heterogeneity of the material, or the system size, within the range of conditions investigated. The scaling of the roughness in terms of root mean square of heights σ with the system size is also consistent with a self-affine morphology (cf. , which shows σ for two surfaces of different size). The estimation obtained for H is in good agreement with the values found for natural faults over a broad range of length scales 4,13 (H = 0.77 ± 0.23), shear experiments in limestone blocks [bib_ref] Fault-surface geometry controlled by faulting mechanisms: experimental observations in limestone faults, Sagy [/bib_ref] The fact that no change in the statistics is observed in correspondence to d * and that the surfaces are rough at all scales is possibly due to the change in the contact configuration. After the debris particle is formed, the system transitions in fact to a three-body contact configuration. The loading conditions are then changed and the critical length scale d * might thus assume a different value upon rolling contact. Furthermore, atoms can be removed at the interface by attrition, and plastic deformation also contributes to the change of the surface morphology. Another important mechanism, first put forward to explain fault roughness [bib_ref] Faults smooth gradually as a function of slip, Brodsky [/bib_ref] [bib_ref] Smoothing and re-roughening processes: the geometric evolution of a single fault zone, Shervais [/bib_ref] , is the smoothing and re-roughening of the surface by the removal of fragments from it. According to this mechanism, a fragment removed from the surface roughens it at the scale of the fragment and smooths it at larger wavelengths. In our simulations, the strong interfacial adhesion allows for this mechanism to happen ideally at all the modelled length scales. An upper length scale for the fragments is set by the current contact size between the debris particle and the surfaces. According to the described picture, larger wavelengths should be smoothed more than shorter ones, which we observe in the initial stage of the process (see Supplementary Discussion). We also remark that, beside the ductile and brittle mechanisms, surface diffusion takes place in our simulations and, without sliding, at infinite time the equilibrium surfaces would be close to atomistically flat. The Ductile-to-brittle transition, simulation setup, and evolution. a Upon collision between two asperities (1), two possible mechanisms can take place depending on the junction size d: if it is larger than the critical, material-dependent value d * , surfaces break and a debris particle is formed (2), else the asperities deform plastically. Solid red lines represent the junction of size d and dotted red lines represent the crack path. b Setup: the two bodies are pressed together by a normal force f y , while the sliding velocity is imposed on the top layer of atoms of the upper body. The bottom layer of atoms is fixed horizontally. A thermostat is applied on the layers next to the fixed boundaries. The box width l x is fixed and periodic boundary conditions are applied along x; the simulation cell, with initial vertical size l y , is allowed to expand/shrink vertically. b-d Single-asperity setup, example frames from simulation S1. The point of first contact in the two-body configuration is well defined and a debris particle is formed upon asperity collision (c); in the three-body configuration, the debris particle wears away the surfaces while increasing in volume (d). e-h Setup with self-affine homogeneous surfaces, example frames from simulation R2. The asperities are present at all modelled scales and deform plastically upon contact in the two-body configuration (f) until a junction size is large enough to favour debris particle formation (g) and the transition to the three-body configuration (h). i-l Setup with heterogeneous self-affine surfaces: harder material is depicted in red and dark blue, softer material in yellow and light blue, example frames from simulation H1. m-p Setup with heterogeneous self-affine surfaces with grain boundaries, example frames from simulation G1. The steady-state surface appears rougher (p). In all figures, colours distinguish particles originally belonging to the top (dark and light blue) and bottom (yellow and red) surfaces; in b-p, black lines represent simulation box boundaries and s is the sliding distance expressed in units of r 0 Summary of the simulations S indicates simulations with initial geometry described by a single asperity on each surface. R indicates simulations with surfaces that are initially self-affine. H indicates simulations with surfaces that are initially self-affine and with heterogeneous materials without grain boundaries, modelled by even shares of the potential characterized by the two values of τ sf reported. G indicates simulations with surfaces that are initially self-affine and with heterogeneous materials with grain boundaries, modelled by even shares of the potentials characterized by the values of τ sf reported (see also Methods section). l x and T indicate the horizontal resolution and the temperature, respectively. H is the initial surface Hurst exponent for initially self-affine surfaces.
[formula] Name τ sf εr À2 0 À Á l x (r 0 ) T (ε) Initial H (−) Micro-structureS1 [/formula]
Temperatures are expressed in terms of equivalent kinetic energy per atom; n/a: field not applicable to that simulation deformation mechanisms, though, are fast enough to counteract diffusion, and contribute to a rich distribution of the surface heights. The lattice planes are in fact not aligned across the sample surface (cf. . Furthermore, when grain boundaries are modelled, each grain is initially assigned a random rotation and can rotate during the sliding process, providing an additional mechanism for the surface roughening and leading to a larger spread in the height distribution (cf. .
A theoretical value of H = 0.5 (i.e. random correlation) for wear processes was proposed by means of a diffusion model with random deposition [bib_ref] Diffusion as a model of formation and development of surface topography, Schargott [/bib_ref]. On the other hand, both experimental and numerical results suggest that adhesive wear is not a random Gaussian process, and that H > 0.5. Thus, more ingredients are needed in theoretical models that aim at describing the surface evolution during adhesive wear processes, including plastic deformation and brittle fracture. The debris particle is in contact with only a small part of the surface at every instant, localizing the deformation and the material transfer, and the sliding direction breaks down the symmetry in the evolution. This provides some similarities with the gradient percolation models used in fracture front propagation: in this class of models, the self-affine fractal front propagates towards a preferred direction (providing asymmetry) and the predicted Hurst exponent H = 2/3 (when small-scale effects prevail over large-scale elasticity) is consistent with our findings 20 . Models for directed polymers in a random medium also exhibit H = 2/3 and may provide further insights. Finally, the inclusion of a scale-dependent material strength is likely another fundamental ingredient needed in theoretical models to capture a persistent Hurst exponent (i.e. H > 0.5) [bib_ref] Nanoscale roughness of natural fault surfaces controlled by scale-dependent yield strength, Thom [/bib_ref]. There is in fact evidence 13 that mechanical behaviour underlies a Hurst exponent H = 0.75 ± 0.05 in rocks at the nanoscale.
Evolution of a debris particle. A particular feature of our simulations is that the two surfaces are worn mostly during threebody contact, which is relevant for the overall wear formation in both natural and industrial sliding processes [bib_ref] Faults smooth gradually as a function of slip, Brodsky [/bib_ref] [bib_ref] The third-body approach: a mechanical view of wear, Godet [/bib_ref]. The presence of third bodies clearly plays a key role in the emergence of the selfaffine morphology and therefore we now analyse the life of a debris particle once it is formed. Details on how the debris particle is born have already been addressed elsewhere [bib_ref] On the debris-level origins of adhesive wear, Aghababaei [/bib_ref].
Two different geometrical setups are adopted for the simulations shown in : single asperity on single asperity at three different temperatures (S1-S3) and self-affine surface on selfaffine surface at two different temperatures (R1 and R2). The evolution of the wear volume in shows that in all cases our simulations capture both the severe wear running-in phase (formation of the debris particle, i.e. the non-zero initial wear volume) and the mild wear steady-state phase that follows. This matches experimental observations [bib_ref] Transient wear of machine parts, Queener [/bib_ref] [bib_ref] Wear processes during frictional sliding of rock: a theoretical and experimental study, Wang [/bib_ref] , as suggested by the evolution of the equivalent roughness σ eq and Supplementary Discussion).
It can be observed that the evolution of the three parameters (surface roughness σ eq , tangential work W t , and wear-particle volume V) exhibits common features among the simulations. In each simulation, a sharp increase in σ eq is observed upon formation of the debris particle, which corresponds to a sudden increase in the frictional work (inset of . In this initial stage, the wear volume has been found to be proportional to the work done by the frictional force [bib_ref] On the debris-level origins of adhesive wear, Aghababaei [/bib_ref] : V = W t /τ j , with τ j being the junction shear strength and W t being the integral of the frictional force over the sliding distance. This relation applies during the particle formation, where the frictional force reaches its maximum and then slowly decreases towards small values. Over long timescales, the integral of these small values leads to a significant amount of work, which partly contributes to the particle growth. The inset of and c indeed shows that the initial volume of the debris particle is larger when the frictional work is larger, and for the single-asperity setup, the frictional work appears to reach a plateau. When the original surfaces are self-affine and more collisions take place at running-in, the plateau is not clearly defined. Looking at the evolution of W t in , it is clear that the frictional work is not constant over the investigated timescale, but that it keeps increasing at an Steady-state surface morphology analysis. a PSD per unit length Φ as a function of the wavevector q and the wavelength λ, the relation between the two being q = 2π/λ. b Height-height correlation function
[formula] H = 0.7 H = 1.0 S1 t S2 t S3 t S4 t S5 t R2 t R3 tH = 0.7 H = 1.0 S1 t S2 t S3 t S4 t S5 t R2 t R3 t H1 t G1 t G2 t Φ (rΔhðδxÞ ¼ hðx þ δxÞ À hðxÞ ½ 2 D E 1=2 [/formula]
. The surfaces are taken from ten different simulations (see for details), the subscript indicates the top surface for each simulation. Bottom surfaces for the same simulations are reported in approximately constant rate (consistently with a constant tangential force F t , cf. . The rate is nevertheless lower than the average rate displayed in the running-in stage, which governs the initial debris particle size (cf. . This suggests a change in the mechanism of wear, as supported by the loss of the proportionality of the wear volume to 1/τ j after running-in and by the change in the wear rate . We can then split the debris particle life into two distinct phases: particle formation and particle evolution. The two situations are characterized by different wear rates (cf. Supplementary , the one corresponding to particle formation being much larger. This result is consistent with decreasing gouge-formation rates observed for natural faults [bib_ref] Faults smooth gradually as a function of slip, Brodsky [/bib_ref]. The reduction in the friction coefficient with the sliding distance in fault lubrication processes 56 provides another consistent observation, under the reasonable assumption that wear rate and friction variations are similar under those conditions [bib_ref] Fault weakening and earthquake instability by powder lubrication, Reches [/bib_ref] [bib_ref] Frictional strength and wear-rate of carbonate faults during high-velocity, steady-state sliding, Boneh [/bib_ref]. Likewise, shear experiments on rocks have shown vanishing wear rates in a threebody steady-state configuration [bib_ref] Frictional strength and wear-rate of carbonate faults during high-velocity, steady-state sliding, Boneh [/bib_ref] , and wear experiments have displayed a reduction in the wear rate if the wear debris is not evacuated [bib_ref] The third-body approach: a mechanical view of wear, Godet [/bib_ref] [bib_ref] Third particle ejection effects on wear with quenched and tempered steel fretting..., Hintikka [/bib_ref] [bib_ref] Wear debris mobility, aligned surface roughness, and the low wear behavior of..., Harris [/bib_ref]. We ascribe the change in the wear rate, i.e. high to low, to the different contact configurations. In the initial phase, asperities belonging to both surfaces collide continuously upon sliding in a two-body configuration, and wear debris particles are repeatedly formed at a constant rate, as in Archard's picture [bib_ref] Contact and rubbing of flat surfaces, Archard [/bib_ref]. As wear experiments are usually performed in open systems, where the wear debris particles are regularly evacuated, the formation of a full three-body contact configuration is avoided and the rate of wear particle creation is constant (due to local collisions of asperities between the two surfaces). If a three-body contact configuration develops instead (for example in faults), the third body behaves similarly to a lubricant 53 , separating the first bodies. As a consequence, the two surfaces are not directly in contact with one another, but with the third body only. Wear can then take place exclusively at the interface between the third body and the surfaces. Both our simulations and the cited observations include three-body contact conditions and predict a decrease in the wear rate compatible with the introduction of a lubricating effect in the tribosystem. The rate of evacuation of debris particles from the system is thus fundamental in the evolution of surface roughness. A low evacuation rate is expected to reduce asperity collisions, which are responsible for creating roughness at scales larger than the critical junction size [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref]. The mechanical behaviour of the wear debris is also expected to affect the wear rate [bib_ref] The third-body approach: a mechanical view of wear, Godet [/bib_ref]. While in our case the debris particle has the same mechanical behaviour as the first bodies, the presence of chemical processes that harden the wear debris might actually increase the wear rate. If the debris particles favour rigid rolling over frictional sliding, though, this increase might not occur [bib_ref] The third-body approach: a mechanical view of wear, Godet [/bib_ref].
Influence of heterogeneity. The case of heterogeneous materials with grain boundaries (simulations G1 and G2) yields further observations. The inclusion of two different species and the presence of grains, instead of a perfect lattice, provide favourable crack propagation directions that are not present in the homogeneous cases. Cracks propagate more easily along the grain boundaries-where the mismatch of the lattices reduces the overall bond strength between the atoms at the interface-and within the bulk of the least tough species. Additionally, cracks are more likely to stop propagating when the tip meets the tougher material (cf. . Furthermore, the heterogeneous grains modelled in simulations G1 and G2 can be seen as inclusions of one of the two species in a matrix made of the other one, in the extreme case where the two species have the same phase fraction. The same mechanisms affecting the crack path (because of the lattice mismatch and/or different material properties) are thus expected to occur-although more localizedwhen the inclusions' phase fraction is smaller than the matrix phase fraction. As the latter is a case in between the homogeneous and heterogeneous bulk cases investigated here (i.e. simulations R and G, respectively), the findings of this work suggest that a similar self-affine surface morphology should be recovered. Real materials might also contain pre-existing dislocations or point defects, which are not considered in the present set of simulations. In these cases, we would expect the defects to affect the plastic response of the bulk and, thus, its critical length scale and the debris formation process. Moreover, while we restricted ourselves to one possible crystal structure, materials can also be amorphous. Assuming homogeneous materials, we would then expect the crack path to have no favourable direction because of the isotropic structure, contrary to the hexagonal lattice modelled here. Finally, surfaces undergoing continuous reworking are also known to exhibit hardening, which most likely leads to a change of the critical length scale over time, affecting in turn the wear rate.
It should be also noted that our simulations are restricted to 2D systems by the large computational demands of long-timescale simulations in three dimensions (3D). One relevant difference Evolution of equivalent roughness σ eq , frictional work W t , and wear volume V. See for further simulations. a Evolution of σ eq . While for most simulations the value of σ eq stabilizes, cold temperatures (S3) and debris particle shape (R1) can inhibit this stabilization. b Evolution of the tangential work W t with the sliding distance. The work W t exhibits a sharp increase upon formation of the debris particle 33 (inset: W t for sliding distances up to 1000 r 0 ), after which it grows at smaller rates. c Evolution of the wear volume of the rolling debris particle, as defined only after its formation. In all simulated conditions, the wear rate after the debris particle formation is small compared to the ratio of the initial particle size over the sliding distance necessary to form the particle (cf. , consistent with the transition from severe to mild wear 49 between 2D and 3D systems is that in the first case, the lattices of the two bodies are more likely to locally match, allowing full, bulk-like adhesion to develop. Furthermore, in 3D, the debris particle could roll at some angle with respect to the sliding direction of the top body, as a consequence of local peaks and valleys due to the roughness along the additional dimension. It is also known that the steady-state roughness normal to the sliding direction is different than that parallel to the sliding direction [bib_ref] Evolution of fault-surface roughness with slip, Sagy [/bib_ref] , which cannot be captured by 2D simulations. Also, in 3D, the debris particles are not forced to pass over the same track again and again. Despite these differences, which might result in a slowdown of the roughness evolution process, we believe that the same mechanisms should be recovered in 3D systems, too, as the aforementioned evidence from experiments and field observation of faults suggests. Short timescale 3D simulations have also confirmed 2D observations for the early stages of the sliding process [bib_ref] On the debris-level origins of adhesive wear, Aghababaei [/bib_ref]. Further investigations are necessary in any case to address the effects of 3D geometry and of the more complex micro-structures.
# Discussion
Our molecular dynamics simulations of rubbing surfaces highlight the importance of including both ductile and brittle deformation mechanisms in the modelling of adhesive wear processes. This allows to explicitly capture the transition to the three-body configuration [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] [bib_ref] On the debris-level origins of adhesive wear, Aghababaei [/bib_ref] , where the surfaces are worn away by the debris particle.
The approach leads us to two major results. The first is the ability to track the evolution of rubbed surfaces into self-affine fractals characterized by a persistent Hurst exponent. We argue that the development of the self-affine morphology is due to smoothing and re-roughening mechanisms and that these mechanisms take place mostly in a three-body configuration, as in natural faults [bib_ref] Faults smooth gradually as a function of slip, Brodsky [/bib_ref] [bib_ref] Smoothing and re-roughening processes: the geometric evolution of a single fault zone, Shervais [/bib_ref]. The second result is that the wear rate is lower once the system has transitioned to a three-body configuration. We ascribe this to the different contact configuration: two-body contact at running-in and three-body contact later, the latter having a lubricating effect. This unveils the role of the debris evacuation-rate in wear experiments. We conclude that accounting for the ductile-to-brittle transition in wear mechanisms is fundamental when investigating the physics of adhesive wear, from the nano-to the geological-scale.
# Methods
Interaction potentials. The model pair potentials used for this study belong to the family of potentials first introduced in ref. [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] and also discussed in ref. [bib_ref] On the debris-level origins of adhesive wear, Aghababaei [/bib_ref]. This class of model potentials allows for a critical junction size small enough to observe the ductile-to-brittle transition in adhesive wear with molecular dynamics simulations. Their main feature is to share the same elastic properties up to a bond stretch of 10% and to have a controllable yield strength by modifying the potential tail. This is made possible by modification of the Morse potential 62 , leading to the expression:
[formula] VðrÞ ε ¼ 1 À e Àα rÀr 0 ð Þ À Á 2 À1 r < 1:1r 0 c 1 r 3 6 þ c 2 r 2 2 þ c 3 r þ c 4 1:1r 0 r <r cut 0 r cut r 8 > < > : ;ð1Þ [/formula]
where r is the interparticle distance, ε is the bond energy at 0 K, r 0 is the equilibrium bond length, and α = 3.93 r À1 0 governs the bond stiffness. The value of r cut controls the cut-off distance, governing the inelastic behaviour, and the c i coefficients are chosen to ensure the potential continuity in energy and force. With respect to the values in , the potential with τ sf = 3.96 εr À2 0 corresponds to the potential named P6 in ref. [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] , the potential with τ sf = 3.52 εr À2 0 corresponds to the potential named P4 in ref. [bib_ref] Critical length scale controls adhesive wear mechanisms, Aghababaei [/bib_ref] , and the potential with τ sf = 3.42 εr À2 0 is more ductile than potential P4 but more brittle than potential P3.
Simulation geometry and boundary conditions. All simulations were performed in 2D using the molecular dynamics simulator LAMMPS [bib_ref] Fast parallel algorithms for short-range molecular dynamics, Plimpton [/bib_ref]. A simple scheme of the simulation setup is shown in . Two different horizontal box sizes have been adopted, i.e. l x = 336 r 0 and l x = 673 r 0 , and periodic boundary conditions are applied in the horizontal direction. The initial vertical box size is constant (l y = 392 r 0 ) and the box is allowed to expand vertically, e.g. upon debris particle formation. A constant force (f y = 0.02 εr À1 0 ) is applied on both horizontal boundaries of the material to press the surfaces together. A constant horizontal velocity v = 0.01 ffiffiffiffiffiffiffiffiffiffi ffi εm À1 p is imposed on the first layer of atoms of the top surface. The bottom layer of atoms of the bottom surface is fixed. Temperatures are enforced by means of Langevin thermostats with a damping parameter of 0.05 r 0 = ffiffiffiffiffiffiffiffiffiffi ffi εm À1 p . On each body, the thermostats are applied to the three layers of atoms next to the external layer where the fixed displacement or velocity is imposed. Temperature values provided in are expressed in terms of equivalent kinetic energy per atom. The integration is performed with a time step of 0.005 r 0 = ffiffiffiffiffiffiffiffiffiffi ffi εm À1 p for a large number of steps, i.e. 1 billion for the shortest simulations and 2.6 billion for the longest one. summarizes the main features of the simulations. Simulation names reflect the initial geometry or material heterogeneity: S stands for single asperity, R for rough self-affine surface, H for heterogeneous materials (without grain boundaries), and G for heterogeneous materials with grain boundaries. In the singleasperity setups S1-S7, we chose to model semicircular asperities, but different shapes (e.g. square or sinusoidal) are not expected to alter our findings 24 . In simulations H1, G1, and G2, both bodies are modelled by two phases (using two different potentials), present in equal shares and randomly distributed in 500 tiles generated from 500 points randomly distributed in the simulation cell by means of Voronoi tessellation. No grain boundaries between the tiles are modelled in simulation H1. Grain boundaries are modelled in simulations G1 and G2: each of the 500 tiles is a grain, whose orientation is determined by a random rotation, leading to lattice mismatch. In simulations G1 and G2, one of the two potentials is the same used for simulations S1, S2, S3, S4, R1, R2, and H1, and it is characterized by τ sf = 3.52 εr À2 0 (cf. . The other adopted potential has the same equilibrium energy ε, but its equilibrium and cut-off distances are scaled by a factor 0.90 in simulation G1 (resulting in τ sf = 4.35 εr À2 0 ) and 0.95 in simulation G2 (τ sf = 3.90 εr À2 0 ). The potential well of the cross-terms between the two species is 0.9ε; its equilibrium and cut-off lengths are given by the average of the respective lengths of the two species. The starting system is obtained by constructing a bulk microstructure, heating it up and then annealing it at the target temperature, allowing the grains to reach an equilibrium configuration. Atoms are then removed from the system based on a purely geometric criterion to obtain two distinct rough surfaces.
Self-affine surfaces. Fractal surfaces whose heights h(x) scale differently than the horizontal distance x are self-affine fractals, and they obey the scaling relationh(λx)~λ H h(x), where λ is the scaling factor and H is the Hurst exponent, which is 0 < H < 1 for fractional Brownian motion (fBm) [bib_ref] Self-affine fractals and fractal dimension, Mandelbrot [/bib_ref]. The Hurst exponent describes the correlation between two increments in the surface. Assuming x 1 < x 2 < x 3 < x 4 , let us consider the two height increments Δh 1 = h(x 2 ) − h(x 1 ) and Δh 2 = h(x 4 ) − h (x 3 ). For H = 0.5, Δh 1 and Δh 2 are randomly correlated (i.e. standard Brownian motion), which means that Δh 2 has a 50% probability of having the same sign of Δh 1 . For 0 < H < 0.5, the increments Δh 1 and Δh 2 are negatively correlated, that is Δh 2 is more likely to have the opposite sign of Δh 1 (the motion is anti-persistent: a positive increment is more likely to be followed by a negative one). Finally, for 0.5 < H < 1, the increments are positively correlated, that is Δh 2 is more likely to have the same sign of Δh 1 (the motion is persistent: a positive increment is more likely to be followed by another positive one). The generation of engineering surfaces, that is surfaces that are manufactured, is known to be non-stationary and random 5 , and it can be described as a non-stationary process with stationary increments, which allows to relate the fractal dimension D of the surface with its Hurst exponent H through its Euclidean dimension n 46 : D + H = n + 1. Moreover, the statistics of this class of surfaces have been investigated studying the Weierstrass-Mandelbrot function [bib_ref] Generalized fractal analysis and its applications to engineering surfaces, Ganti [/bib_ref] , allowing to relate the fractal dimension D and the power law exponent α of the PSD of a 1D surface profile 47,48 as α = 5 − 2D. Under these assumptions, H = (α − 1)/2 for a 1D surface (n = 1).
The latter allows to estimate the Hurst exponent H from a linear fit of the data in the log-log plane, discarding the tail (λ < 4 r 0 ), where the definition of surface for a discrete system breaks down and data are inevitably polluted by the numerical surface reconstruction. This method has been shown 64 to be accurate but also to be affected by a systematic error that can possibly lead to an underestimation of H. This underestimation decreases by increasing the system size, and for system sizes of the order of magnitude investigated in this paper it is at most 0.1. The range of values of H in our study would then be H = 0.7-0.9. We refer throughout the text to the values found with the fit, without correcting for the underestimation, for consistency with the methods adopted in the literature (e.g. refs. [bib_ref] Constant dimensionality of fault roughness from the scale of micro-fractures to the..., Renard [/bib_ref] [bib_ref] Fractal characterization and simulation of rough surfaces, Majumdar [/bib_ref] [bib_ref] Nanoscale roughness of natural fault surfaces controlled by scale-dependent yield strength, Thom [/bib_ref] , where no correction is applied.
Spectral analysis. Let us consider a surface of length L, whose height is defined by the continuous function h(x), where x is a spatial coordinate. We refer to the PSD of such a surface in terms of PSD per unit length Φ h (q), q being the wavevector, defined as 65 where P h (q n ) is the classical periodogram [bib_ref] Understanding the Lomb-Scargle periodogram, Vanderplas [/bib_ref] :
[formula] Φ h q ð Þ 1 L Z L h xðP h q n ð Þ ¼ 1 N X NÀ1 k¼0 h k e Àiq n x k 2 ;ð4Þ [/formula]
the summation being the discrete Fourier transform of the surface. In fact, h(x) is known only at a discrete set of N points x k (k = 0, 1, …, N − 1), regularly sampled at an interval Δx, such that h k = h(kΔx) are the known values of h(x). In our case, Δx = L/N, N being the number of atoms belonging to a surface of length L, and is approximately 1 r 0 . It can be shown that both the PSD Φ h per unit length and the periodogram P h are normalized such that R q Φ h q ð Þdq and P n P h q n ð Þ are equal to the mean squared amplitude σ 2 of h(x) and h k , respectively.
Note that the derivative of both P h and Φ h is the same in the log-log plane, as their estimation differs only by a multiplicative factor Δx (see Eq. (3)): the estimated self-affine exponent does not change if one or the other is considered.
Height-height correlation analysis. Another suitable method to estimate the Hurst exponent H is to investigate the height-height correlation function, which describes the change of heights Δh between two points at distance δx horizontally:
[formula] ΔhðδxÞ ¼ hðx þ δxÞ À hðxÞ ½ 2 1=2 ;ð5Þ [/formula]
where the angle brackets indicate spatial average. The height-height correlation function scales as Δh(δx)~δx H : it is therefore possible to determine the Hurst exponent H from its log-log plot. It is also possible to observe potential upper and lower cut-off values of δx limiting the scaling regime.
Average roughness quantification. The surface roughness is defined as the variations in height of the surface profile with respect to an arbitrary plane of reference, which in our case is always taken to be the mid-plane of the surface heights. To quantify the surface roughness, several parameters are used-in particular in engineering practice-but we limit our discussion to the root mean square of heights σ, which in the case of a surface discretized in a set of N points is given by
[formula] σ ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi 1 N X N k¼1 h 2 k v u u t ;ð6Þ [/formula]
where h k is the distance of the point k from the plane of reference. When the system as a whole is investigated, the equivalent σ of the composite surface (given by the two surfaces of the top and bottom materials) becomes 55
[formula] σ eq ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi σ 2 top þ σ 2 bottom q :ð7Þ [/formula]
The root mean square of heights σ can also be expressed as a function of the zeroth moment of the PSD per unit length Φ h , assuming that the surface profile is continuous 6,65 :
[formula] σ 2 ¼ Z q l q h Φ h q ð Þdq;ð8Þ [/formula]
where q l and q h are the lowest and highest wavevectors modelled in the system. If q h ) q l , it has been shown 6 that σ 2 / q Àð4À2DÞ l . Data analysis. All frames are visualized with OVITO 68 . Due to the long duration of the simulations, frames are stored every 1,000,000 steps. At first, a surface is reconstructed by identifying atoms with low coordination number. This preliminary surface includes (i) the top and bottom surface portions that are not in contact with the debris particle and (ii) the debris particle surface portion that is not in contact with any surface. The two top and bottom surfaces are then reconstructed, their portion in contact with the debris particles being approximated by the closest group of atoms belonging to a straight segment. This approximation to a straight segment gives no loss of information for the spectral analysis by means of the Fast Fourier Transform up to high wavevectors [bib_ref] Effect of data gaps: comparison of different spectral analysis methods, Munteanu [/bib_ref]. The atoms contained between the two segments and the previously identified debris particle surface identify the debris particle and are thus discarded in any top and bottom surface analysis. Each reconstructed surface now consists of irregularly spaced atoms and is processed independently. The atom positions are linearly interpolated to recreate a bijective profile that is then discretized in N points, evenly spaced along the horizontal x-axis, where N is the number of unevenly spaced atoms belonging to the surface prior to the linear interpolation. This step allows to proceed with an analysis of the surface by means of the classical periodogram through a Fast Fourier Transform algorithm, as the Lomb-Scargle periodogram for unevenly spaced data is known to provide poorer results in the spectral analysis of surface morphologies [bib_ref] Effect of data gaps: comparison of different spectral analysis methods, Munteanu [/bib_ref] [bib_ref] Fault roughness at seismogenic depths from LIDAR and photogrammetric analysis, Bistacchi [/bib_ref]. The horizontal spacing of the re-sampled surface is approximately 1 r 0 . Forcing a bijective profile results in a noisy geometry where overhangs are present, affecting the large wavevector amplitudes, but does not alter the data at lower wavevectors, where the self-affine morphology is observed. The interval between two consecutive time steps used for averaging the PSD is large enough for the particle to have rolled over the whole surface at least one time.
The data for σ and σ eq are averaged over 10 consecutive data points. The tangential work W t is computed as the integral of the tangential force F t over the sliding distance s: R F t ds; the tangential force values are stored every 5000 steps and are averaged over 2000 data points. The debris particle volume V is computed by multiplying the number of atoms belonging to the debris particle and the atomic volume. The latter is computed as the lattice unit cell volume at the temperature T, divided by the number of atoms in the unit cell. As the detection code is designed for the most common situation of the particle being in contact with both surfaces at the same time, due to the particular geometry at some time steps, e.g., when the particle is only in contact with one of the two surfaces and is not rolling, the data for the surface roughness σ and σ eq , the wear volume V, and the linear fit exponent α may be unevenly spaced locally around some time steps.
## Data availability
All raw data supporting the findings of this study are available from the authors upon request.
Received: 14 June 2018 Accepted: 8 February 2019 |
Peer support for patients with type 2 diabetes: cluster randomised controlled trial
Objective To test the effectiveness of peer support for patients with type 2 diabetes. Design Cluster randomised controlled. Setting 20 general practices in the east of the Republic of Ireland.Participants 395 patients (192 in intervention group, 203 in control group) and 29 peer supporters with type 2 diabetes. Intervention All practices introduced a standardised diabetes care system. The peer support intervention ran over a two year period and contained four elements: the recruitment and training of peer supporters, nine group meetings led by peer supporters in participant's own general practice, and a retention plan for the peer supporters. Main outcome measures HbA 1c ; cholesterol concentration; systolic blood pressure; and wellbeing score. Results There was no difference between intervention and control patients at baseline. All practices and 85% (337) of patients were followed up. At two year follow-up, there were no significant differences in HbA 1c (mean difference −0.08%, 95% confidence interval −0.35% to 0.18%), systolic blood pressure (−3.9 mm Hg, −8.9 to 1.1 mm Hg), total cholesterol concentration (−0.03 mmol/L, −0.28 to 0.22 mmol/L), or wellbeing scores (−0.7, −2.3 to 0.8). While there was a trend towards decreases in the proportion of patients with poorly controlled risk factors at follow-up, particularly for systolic blood pressure (52% (87/166) >130 mm Hg in intervention v 61% (103/169) >130 mm Hg in control), these changes were not significant. The process evaluation indicated that the intervention was generally delivered as intended, though 18% (35) of patients in the intervention group never attended any group meetings.Conclusions A group based peer support intervention is feasible in general practice settings, but the intervention was not effective when targeted at all patients with type 2 diabetes. While there was a trend towards improvements of clinical outcomes, the results do not support the widespread adoption of peer support. Trial registration Current Controlled Trials ISRCTN42541690.
# Introduction
There has been an increasing focus on involvement of patients in chronic disease care, and the World Health Organization's action plan for chronic disease management encourages governments to take action to help people manage their own chronic conditions better by providing education, incentives, and tools for self management and care. 1 Type 2 diabetes is a chronic disease that is rising in prevalence across the world and placing increasing demands on healthcare systems. WHO has suggested that peer support is a promising approach for diabetes care as it harnesses the ability of patients with diabetes to support each other in managing their everyday lives. [bib_ref] A solution might be within people with diabetes themselves, Caro [/bib_ref] Peer support has been defined as the provision of support from an individual with experiential knowledge based on a sharing of similar life experiences. [bib_ref] Peer support within a health care context: a concept analysis, Dennis [/bib_ref] It is usually provided within a volunteering framework and can be delivered in many ways, including group or individual support or through more remote formats such as telephone or internet based support.
Peer support has been used in various conditions with varied results, 4-10 but there is limited evidence to support its effectiveness, particularly for people with type 2 diabetes. There is also substantial variation in the degree of training and level of involvement of the peer supporters or community health workers in these studies, with many having a predominantly educational focus and peer groups being facilitated by health professionals rather than peers themselves. [bib_ref] Effectiveness of community health workers in the care of persons with diabetes, Norris [/bib_ref] We report the results of a pragmatic cluster randomised controlled trial examining the effectiveness of peer support in improving biophysical and psychosocial outcomes for people with type 2 diabetes. The intervention was based on social support theory and was delivered in groups based in the general practices of participating patients.
# Methods
The methods and intervention development have been reported in detail previously. In brief, this was a cluster randomised controlled trial set in general practices in the Republic of Ireland. Diabetes care in Ireland has generally been unstructured, with more than half of patients still attending specialist diabetes clinics for annual reviews. [bib_ref] Primary care diabetes in the Republic of Ireland, Smith [/bib_ref] Thirty two practices from the Trinity College Dublin network of teaching practices were invited to participate, and 20 accepted. Practices were eligible to participate if they had a practice nurse; had computerised records; did not participate in the pilot study; had a minimum of about 1000 patients with GMS cards (eligible for free healthcare) or had more than 50 patients on their register of those with type 2 diabetes, or both; and were not participating in an existing shared care diabetes programme involving structured care between the general practices and hospital diabetes clinics. Participating practices were given a grant of €5000 (£4200, $6900) a year for three years. This grant reflected the fact that data collection and intervention delivery were carried out at practice level with the support of the research team.
Practices were stratified by practice size and the presence of existing structured diabetes care and were then allocated to control or intervention group by an independent statistician using minimisation. 14 All practices introduced a structured diabetes care system as only two of them had such systems in place before the study. This involved regular recall of patients every three to six months with an annual audit of risk factors. This was done to standardise delivery of diabetes care across all study practices so that the peer support intervention would be an addition to diabetes care. All practices created or, in the case of two practices, updated their diabetes registers. Intervention practices then recruited three individuals as peer supporters.
The box outlines the eligibility criteria for peer supporters. They attended two evening training sessions each lasting about 90 minutes and delivered by the research team. Appendix 1 on bmj.com describes the content of the training sessions. Peer supporters were advised that they were not being asked to provide healthcare or education and that the emphasis was on social support. There was also a strong emphasis on the importance of confidentiality, and peer supporters were given specific advice as to how they could seek support from their general practitioner and practice nurse or the study manager if needed. The general practitioners and practice nurses also received training in relation to the study protocol and delivering standardised diabetes care (one author (SMS) delivered a 90 minute session in each practice with an academic detailing approach). Further details of the training are provided in the study protocol. [bib_ref] Peer support in type 2 diabetes: a randomised controlled trial in primary..., Paul [/bib_ref] We used a random number list to select patients randomly from the practice diabetes register until a total of 21 patients were recruited in each practice. To be eligible patients had to be aged over 18, have type 2 diabetes, be able to participate in a group, and be attending one of the participating practices. Those who agreed to participate were invited to attend their practice to complete a questionnaire containing demographic information and study outcome measures. The practice nurse then collected biophysical outcome data.
The intervention was developed with the MRC Framework for the evaluation of complex interventions to improve health. [bib_ref] Designing and evaluating complex interventions to improve health care, Campbell [/bib_ref] The underlying theoretical framework was social support, and the full development of the intervention has been described in detail previously. [bib_ref] Development of a complex intervention to test the effectiveness of peer support..., Paul [/bib_ref] The intervention was delivered over a two year period, from May 2007 to April 2009. The box describes the components of the intervention.
The primary outcomes were HbA 1c (measured with reversed phase cation exchange liquid chromatography, with an automatic glycol-haemoglobin analyser, DCCT compatible); blood pressure (measured with an OMRON M5-1 Automatic BP monitor); cholesterol concentration (analysed in local hospital laboratories with automated clinical chemistry analysers), and wellbeing.Secondary outcomes were body mass index (BMI), diabetes self care activities, 17 self efficacy, [bib_ref] Validation of the Australian/English version of the diabetes management selfefficacy scale, Mcdowell [/bib_ref] adherence to medications, [bib_ref] Patients' beliefs about prescribed medicines and their role in adherence to treatment..., Horne [/bib_ref] family and friends subscale of the chronic illness resources survey, 20 smoking (self reported), prescriptions (aspirin, antihypertensive drugs, and cholesterol lowering agents), and measures of the process of care (visits
## The peer support intervention
The peer support intervention had the following components:
Peer supporters Peer supporters were identified by general practitioners and practice nurses and were trained at a ratio of about one peer supporter to seven or eight patients with type 2 diabetes. The criteria for eligibility were:
Having had type 2 diabetes for at least one year Participation in preventive treatments and judged by the practice team as being generally adherent to treatment and behaviour change regimens Capacity and commitment to undergo the training required A full understanding of the importance of patients' confidentiality Undertaking to liaise with the practice nurse or general practitioner if unanticipated problems arose during the course of their peer support activity Peer supporter training The peer supporters attended two evening training sessions, which were conducted by the research team. These sessions focused on the basics of type 2 diabetes and issues relating to working with groups and confidentiality.
Peer support meetings Peer support meetings were held in the general practice premises at a convenient time for practice staff, peer supporters, and participants. Practices offered various daytime or early evening sessions, depending on patients' preference. There were nine peer support sessions over two years; at month one, month two, and every three months thereafter. Each meeting was facilitated by the peer supporter, and there were no health professionals present in the meeting room though they were available on site, if needed. Each meeting had a suggested theme and a small structured component. The contents of the meetings were recorded (see appendix 2 on bmj.com). There was also a "frequently asked questions" (FAQs) system-that is, at the end of each session the group fed back questions to the research team who compiled written answers based on the feedback from all groups. The FAQs from all groups were combined and sent back to the groups for the next session.
Retention and support of peer supporters Formal structures were put in place to ensure peer support workers were supported in their role, including telephone calls from the project manager before and after meetings; a course handbook and resource pack; an annual social or educational event; a protocol to follow if a peer supporter resigned; and travel and related expenses (this was given in the form of a general shopping voucher at the end of each year with a value of €300).
to general practitioner, practice nurse, hospital diabetes outpatients department, and hospital diabetes centre and admissions to hospital). Demographic details, including measures of socioeconomic status (medical card status) and educational attainment, were also collected. Participants completed the questionnaire while waiting to see the practice nurse, who then checked that the questionnaire was filled in and completed the biophysical measures. The nurses extracted data relating to the process of care from the patients' records and entered all data into a FileMaker Pro database.The project manager doubled checked the data. Full details of data collection and storage procedures are outlined in the study protocol. [bib_ref] Peer support in type 2 diabetes: a randomised controlled trial in primary..., Paul [/bib_ref] We could not blind participants or practice nurses collecting data to group allocation because of the nature of the intervention. Three of the four primary outcomes, however, were collected with automated tests.
We undertook a process evaluation of the trial and measured treatment fidelity using a framework developed by Bellg et al. [bib_ref] Enhancing treatment fidelity in health behaviour change studies: best practices and recommendations..., Bellg [/bib_ref] This involved consideration of five elements: treatment design, training procedures, delivery of treatment, receipt of treatment, and enactment of treatment skills. This information was collected with peer supporter log diaries, the study managers' contact records, and focus groups with practice staff, peer supporters, and participants. Parallel qualitative and economic analyses are ongoing and will be reported elsewhere.
## Sample size calculation
We aimed to achieve a sample of 400 patients from 20 practices. This incorporated the effect of cluster randomisation and allowed for 80-85% follow-up of patients and a 15% rate of practice attrition. Sample size calculations also incorporated a 20% improvement from Recruitment and follow-up of practices and participants. *Does not include participants lost to follow-up RESEARCH baseline in the control group and 50% improvement in the intervention group. All calculations were two sided and based on an α of 5% and a power of 80%. Full details are reported in the study protocol. [bib_ref] Peer support in type 2 diabetes: a randomised controlled trial in primary..., Paul [/bib_ref] HbA 1c -We needed 130 patients from eight practices to show a clinically significant difference in mean HbA 1c between intervention and control groups (that is, a difference of 0.9% 23 ; SD 1.6, intracluster coefficient 0.001 [bib_ref] The North Dublin randomized controlled trial of structured diabetes shared care, Smith [/bib_ref].
Systolic blood pressure-We needed 400 patients from 20 practices to show a significant improvement in the proportion of patients with a systolic blood pressure below 160 mm Hg. This was in the context of a treatment target of 135 mm Hg,based on locally available data indicating that 46% of patients in a previous study in Dublin had a systolic blood pressure >160 mm Hg and the intracluster coefficient was 0.001. [bib_ref] The North Dublin randomized controlled trial of structured diabetes shared care, Smith [/bib_ref] Cholesterol-We needed 410 patients from 20 practices to show a significant improvement in the proportion of patients with a cholesterol concentration <5 mol/L, which was the treatment target for cholesterol at that time.Locally available data indicated that 57% of patients had a concentration >5 mmol/L and the intracluster coefficient was 0.06. [bib_ref] Effectiveness of community health workers in the care of persons with diabetes, Norris [/bib_ref] Wellbeing scores-We needed 221 patients from 12 practices to show a clinically significant difference in wellbeing scores between intervention and control groups (that is, a mean difference of 5 points, SD 10.3; intracluster coefficient 0.07 [bib_ref] The North Dublin randomized controlled trial of structured diabetes shared care, Smith [/bib_ref].
# Statistical analysis
The analyses were based on intention to treat and are reported according to the CONSORT guidelines for the reporting of cluster randomised controlled trials. [bib_ref] CONSORT statement: extension to cluster randomised trials, Campbell [/bib_ref] We also undertook a sensitivity analysis and a per protocol analysis to estimate the effect of exposure to the intervention or group attendance on outcomes. Preplanned analyses were also conducted to examine those participants whose risk factors were above target ranges at baseline-that is, HbA 1c >7%, systolic blood pressure >130 mm Hg, and cholesterol concentration >4.8 mmol/L. These were the targets presented to the clinicians involved in delivering diabetes care across both intervention and control practices.
We used multilevel linear or logistic regression models with random effects of patients nested within practices. In these models, the primary fixed effect of interest is the differential effect of intervention versus control over time. For the subgroup analysis, we selected individuals deemed out of control at baseline for modelling with practice as a random effect. The effect size in this instance refers to the contrast between the intervention versus the control group at follow-up. For all models, we included additional patient specific (such as age and sex) and practice specific (such as type of practice) covariates. The multilevel analysis was conducted with R (2.11).Analysis of secondary outcomes was limited to an intention to treat cluster level analysis apart from analysis of BMI, which achieved a significant effect in this preliminary analysis so was then entered into the multilevel model analysis.
# Results
The figure shows the flow of practices and patients through the study. We could not carry out baseline data collection before randomisation of practices as we had to first identify intervention practices so that we could recruit and train the peer supporters, which took place over a six month period. As a result, there was a difference in recruitment rates of patients between intervention and control practice. More potential participants had to be invited to participate in the study in intervention practices. Recruitment of participants took place between November 2006 and April 2007 and the intervention ran from May 2007 until April 2009. Non-participants in both the intervention and control practices were similar in terms of sex, age, and socioeconomic status (as measured by eligibility for free primary health care). Follow-up data collection took place between May and June 2009. [fig_ref] Table 1 |: Baseline characteristics of participants with type 2 diabetes allocated to peer support [/fig_ref] gives the characteristics of patients in the intervention and control group and the peer supporters. Tables 2 and 3 present the primary and secondary outcomes. They indicate better than anticipated control of risk factors at baseline, though a considerable proportion of patients still had risk factors above target levels. Of the participants, 163/388 (42%) had an HbA 1c above 7%; 291/394 (74%) had a systolic blood pressure above 130 mm Hg, and 89/387 (23%) had a total cholesterol concentration above 4.8 mmol/L. These treatment targets for risk factors differed from those used for the original power calculation as there had been changes in the target levels in the intervening years.
# Follow-up results
At follow-up we found no significant improvements in any primary or secondary outcomes when we used multi-level modelling that accounted for clustering and other confounding variables (see tables 2 and 3). We also carried out multi-level modelling to examine the subgroups of patients with poorly controlled risk factors at baseline, and, while there was a trend towards clinically relevant improvements in proportions with better blood pressure control, these effects were not significant [fig_ref] Table 4 |: Risk factors above target at baseline and follow-up in participants with type... [/fig_ref].
We carried out additional per protocol analyses as planned to test whether there were any links between group attendance and outcomes. We found nothing relating to attendance versus non-attendance and to numbers of groups attended by participants that could be regarded as a dose effect.
## Peer supporters
At baseline data from peer supporters and participants were similar (table 1), though peer supporters had attained a higher level of education. Primary outcomes at baseline for the peer supporters were mean HbA 1c 6.8%, mean total cholesterol 4.3 mmol/L, mean systolic blood pressure 140 mm Hg, and mean wellbeing score 27. Primary outcomes at follow-up were mean HbA 1c 6.9%, mean total cholesterol 3.7 mmol/L, mean systolic blood pressure 139 mm Hg, and mean wellbeing score 24. Secondary outcomes at baseline and follow-up were also similar in peer supporters and participants. Twenty nine (97%) peer supporters were followed up. The descriptive analysis of the peer supporters at follow-up indicated no significant changes over time apart from some decline in wellbeing (mean score 27 at baseline; 24.1 at follow-up).
## Process evaluation
The training and intervention were delivered as planned for the general practitioners, practice nurses, and peer supporters in the protocol. All intervention and control practices implemented structured diabetes care as planned. All the practices and 28 out of the 29 peer supporters were followed up, though only 23 of the peer supporters were retained in their role. The main concern regarding the delivery and receipt of treatment-that is, the intervention-was the low attendance at the group meetings. Participants in the intervention group attended a mean of five peer support meetings, and 18% never attended a meeting and therefore had no exposure to the intervention. This was despite repeated phone calls from practice nurses and a call from the study manager to all non-attenders after the third round of meetings.
Peer supporters were contacted after each meeting and also kept diaries. Appendix 2 on bmj.com provides data collected about the content of peer support meetings. In general, the groups followed and discussed the planned topics.
The process evaluation also highlighted the heavy workload involved in delivering a peer support intervention over two years. There was a mean of 15 contacts between the study manager and the intervention practices relating specifically to the peer support intervention rather than the research process. There was a mean of 25 contacts with the peer supporters during the two year period. These contacts included training sessions, meetings, telephone calls, and letters and indicate that as an intervention peer support requires substantial clinical and administrative input. Most of these contacts related to running the intervention rather than collection of research data.
# Discussion
For people with diabetes a group based peer support intervention is feasible in general practice settings. While there was a trend towards improved management of clinical risk factors, however, peer support did not significantly improve physical and psychosocial outcomes for people with type 2 diabetes. One RESEARCH explanation for this is that, despite the hope that peers could be harnessed to improve outcomes for patients, peer support is ineffective when delivered as a structured intervention. There are some features of this study, however, that suggest we do not yet have a definitive answer about the potential effectiveness of peer support in type 2 diabetes. In particular, while the study achieved its proposed sample size, baseline levels of risk factors for diabetes had improved in the intervening years, leaving little room for improvement in mean HbA 1c , systolic blood pressure, and total cholesterol concentration. The treatment targets had also changed, however, becoming tighter over time, and there were still significant proportions of patients not meeting these targets, particularly for systolic blood pressure. Secondary outcome measures were also no different between intervention and control groups. These indicated high levels of adherence to treatment drugs and self efficacy and medium level scores for diabetes self care activities and social support at baseline and followup. The intervention might not have been intensive enough to affect broader outcomes. We were, however, trying to achieve a balance between overburdening the peer supporters with frequent meetings and having sufficient exposure to the intervention in terms of meeting intensity.
There was a non-significant reduction in wellbeing in the intervention group, and, while this is also clinically relevant, it is important to consider that peer support could have a detrimental impact on wellbeing if groups focused on negative experiences. In relation to social support, it was difficult to find a measure that reflected the type of social support that a group based intervention might provide. The measure used might have failed to detect the type of social support that peers provide compared with the support provided by family and friends, which existing measures consider. This is possible as the qualitative analysis indicated that participants valued the meetings and were positive about the support they had received from their leaders and fellow group members (G Paul, personal communication). An alternative social support outcome measure would be the Lubben social network scale.While this scale does focus on broader social, less disease oriented social support networks, it has been used only in older patients.
Comparison with other studies Our results are consistent with those of other studies published on peer support for type 2 diabetes, which have also failed, in general, to show a significant impact on glycaemic control as measured by HbA 1c . 9 10 29-32 Lorig et al recently published a randomised controlled trial of their chronic disease self management programme adapted for people with diabetes. [bib_ref] Community-based peer-led diabetes self-management: a randomized trial, Lorig [/bib_ref] While they found improvements in depression and in healthy eating, there was no effect on HbA 1c . Studies examining peer support have not specifically targeted patients with poorly controlled type 2 diabetes, and such patients might benefit most from a peer support intervention. Our qualitative analysis also indicated that participants thought they would have benefited from peer support around the time of diagnosis (G Paul, personal communication). The logistics of running a trial of a peer support intervention for people with a new diagnosis would be more challenging but perhaps worth pursuing.
While most of the studies of peer support for type 2 diabetes to date have shown no benefit for participants in terms of glycaemic control, benefits in terms of personal gains (training and satisfaction of helping people) for peer supporters have been reported. [bib_ref] The experience of native peer facilitators in the campaign against type 2..., Struthers [/bib_ref] In this study, however, the peer supporters showed some decline in wellbeing at follow-up, though this might be a chance finding as numbers were small. This raises concerns that the role of peer supporter could be demanding and stressful for the peer supporters themselves, particularly if participants fail to attend group meetings. This has been described previously in other peer support settings. 34
# Strengths and limitations
This cluster randomised controlled trial of group peer support for patients with type 2 diabetes included detailed analysis of the peer supporters as well as participants. Follow-up of practices and participants was high, and there was clearly defined structured diabetes care across both intervention and control practices so that changes in diabetes care could not have been attributed to differences in healthcare delivery as we were trying to assess the potential effectiveness of the peer support intervention itself. A full process evaluation was carried out that indicated the intervention was largely delivered as planned. It also highlighted the considerable workload involved in supporting the intervention. The differential recruitment rate between intervention and control practices and the lower than expected attendance rates indicate that group based peer support is not attractive to all patients with type 2 diabetes and reduced the external validity of this study. [bib_ref] Internal and external validity of cluster randomised trials: systematic review of recent..., Eldridge [/bib_ref] Other modes of provision of peer support might need to be offered in parallel with group meetings. It might also be important to consider flexible approaches to providing peer support, such as having drop-in groups available when patients need support rather than providing scheduled courses that take no account of individual needs. [bib_ref] Peer-based behavioural strategies to improve chronic disease self-management and clinical outcomes: evidence,..., Funnell [/bib_ref] Other issues relating to external generalisability include resourcing of practices. We provided a grant to practices to recognise the work involved in setting up peer support groups and providing ongoing informal support to the peer supporters. The workload of the study manager was considerable and that role is essential to the running of a peer support intervention and needs to be considered if peer support is to be introduced on a larger scale.
One limitation was our inability to conceal allocation. Practices had to be randomised before we collected baseline data so that peer supporters could be identified and trained in intervention practices. This reduced the internal validity of the study as did the lack of blinding of outcome data. Practice nurses collected outcome data and so were not blind to group allocation. For three of the four primary outcomes, however, data were collected with automated tests or devices so this minimised the risk of detection bias. The study performed well in relation to the other features of internal validity for cluster randomised trials described by Eldridge et al. [bib_ref] Internal and external validity of cluster randomised trials: systematic review of recent..., Eldridge [/bib_ref] A further potential limitation is that the presence of a peer support intervention within a practice could have motivated the entire team to provide better diabetes care. The process evaluation and the data relating to prescribing, however, did not suggest differences in delivery of diabetes care between intervention and control practices.
# Conclusion
This cluster randomised controlled trial indicates that it is feasible to implement a peer support system for patients with type 2 diabetes attending general practices, though not all patients will be interested in participating. The intervention was not effective in improving biophysical and psychosocial outcomes for individuals with type 2 diabetes, when targeted at all such patients. While there was a trend towards improvements of clinical outcomes, particularly for systolic blood pressure, our results suggest that peer support should not be widely adopted in clinical practice until further research is carried out. The Peers for Progress organisation is currently carrying out several trials and demonstration projects of peer support for type 2 diabetes across the world.Future research could focus on alternative models of delivering support or targeting support to those with poorly controlled risk factors.
The peer support study team also includes D Handy, K Keogh, M D'Eath, P Gillespie, and F O'Kelly. We thank the peer supporters, patients, and the general practitioners and practice nurses from all the participating practices. We also thank Scott Walkin, who contributed to the initial design of the study, and Graham Watt, who acted as an external reviewer for our grant and provided valuable feedback throughout. Contributors: SMS, DLW, EO'S, and TO'D conceived the study and together with GP and AK participated in the design of the trial and intervention. All authors participated in the acquisition and analysis of
## What is already known on this topic
Peer support could be a promising approach for diabetes care as it harnesses the ability of patients with diabetes to support each other in managing their everyday lives There is limited evidence to date supporting its effectiveness
## What this study adds
Though peer support group meetings can be introduced in general practice settings, many patients were not interested in participating, and 18% of those who agreed to participate never attended any meetings
There was a trend towards improvements in clinical care but no significant improvements in diabetes or psychosocial outcomes Details of the role of study sponsors: The study sponsors funded the study and carried out annual and mid-term reviews of the study progress. They had no part in the design of the study; the collection, analysis, and interpretation of the data; the writing of the report; and the decision to submit the article for publication.
Competing interests: All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years, no other relationships or activities that could appear to have influenced the submitted work. Ethical approval: This study was approved by the research ethics committee of the Irish College of General Practitioners and informed written consent was given by all patients and peer supporters. Data sharing: Additional data can be obtained from the corresponding author for the purposes of secondary research.
[table] Table 1 |: Baseline characteristics of participants with type 2 diabetes allocated to peer support (intervention) or no peer support (control) and peer supporters. Figures are numbers (percentage) of participants unless stated otherwise [/table]
[table] Table 2 |: Primary outcomes at baseline and follow-up in participants with type 2 diabetes allocated to peer support (intervention) or no peer support (control) [/table]
[table] Table 3 |: Secondary outcomes at baseline and follow-up in participants with type 2 diabetes allocated to peer support (intervention) or no peer support (control). Figures are mean (SD) scores except where indicated [/table]
[table] Table 4 |: Risk factors above target at baseline and follow-up in participants with type 2 diabetes allocated to peer support (intervention) or no peer support (control) [/table]
|
Antimicrobial Peptide Induced-Stress Renders Staphylococcus aureus Susceptible to Toxic Nucleoside Analogs
Cationic antimicrobial peptides (AMPs) are active immune effectors of multicellular organisms and are also considered as new antimicrobial drug candidates. One of the problems encountered when developing AMPs as drugs is the difficulty of reaching sufficient killing concentrations under physiological conditions. Here, using pexiganan, a cationic peptide derived from a host defense peptide of the African clawed frog and the first AMP developed into an antibacterial drug, we studied whether sub-lethal effects of AMPs can be harnessed to devise treatment combinations. We studied the pexiganan stress response of Staphylococcus aureus at sub-lethal concentrations using quantitative proteomics. Several proteins involved in nucleotide metabolism were elevated, suggesting a metabolic demand. We then show that Staphylococcus aureus is highly susceptible to antimetabolite nucleoside analogs when exposed to pexiganan, even at sub-inhibitory concentrations. These findings could be used to enhance pexiganan potency while decreasing the risk of resistance emergence, and our findings can likely be extended to other antimicrobial peptides.
# Introduction
Antimicrobial peptides (AMPs, we use AMPs here as synonymous with host defense peptides) are immune effector molecules used by multicellular organisms to control infections [bib_ref] Antimicrobial peptides of multicellular organisms, Zasloff [/bib_ref] [bib_ref] Antimicrobial peptides: application informed by evolution, Lazzaro [/bib_ref]. These peptides are usually active against a broad spectrum of bacterial pathogens and some display activity against antibiotic-resistant bacteria. Thus, antimicrobial peptides are considered a promising source of new antibacterial drugs [bib_ref] Alternatives to antibiotics-a pipeline portfolio review, Czaplewski [/bib_ref] to tackle the current antibiotic crisis [bib_ref] A return to the pre-antimicrobial era?, Baker [/bib_ref].
Some of the factors that make AMPs attractive are their high diversity across the tree of life [bib_ref] APD: the Antimicrobial peptide database, Wang [/bib_ref] and the finding that, although drug resistance also evolves against AMPs [bib_ref] Mechanisms and fitness costs of resistance to antimicrobial peptides LL-37, CNY100HL and..., Lofton [/bib_ref] [bib_ref] Genomic signatures of experimental adaptation to antimicrobial peptides in Staphylococcus aureus, Johnston [/bib_ref] [bib_ref] Genomics of experimental adaptation of Staphylococcus aureus to a natural combination of..., Makarova [/bib_ref] [bib_ref] Therapeutic antimicrobial peptides may compromise natural immunity, Habets [/bib_ref] , it evolves at a much lower probability in comparison to conventional antibiotics [bib_ref] Predicting drug resistance evolution: insights from antimicrobial peptides and antibiotics, Yu [/bib_ref] [bib_ref] Integrated evolutionary analysis reveals antimicrobial peptides with limited resistance, Spohn [/bib_ref]. One common problem with the development of AMPs as drugs is that, under physiological conditions, their antimicrobial activity cannot be easily recaptured and the required dosage is extremely high [bib_ref] Antimicrobial host defence peptides: functions and clinical potential, Mookherjee [/bib_ref]. This dosage issue can be addressed by making use of synergistic combinations of AMPs [bib_ref] Combination effects of antimicrobial peptides, Yu [/bib_ref] , a property common in natural defense cocktails [bib_ref] Magainins and the disruption of membrane-linked free-energy transduction, Westerhoff [/bib_ref] [bib_ref] Synergistic interactions between mammalian antimicrobial defense peptides, Yan [/bib_ref].
While the mode of action on bacterial membranes has been worked out for some AMPs [bib_ref] Influence of hydrophobic residues on the activity of the antimicrobial peptide magainin..., Strandberg [/bib_ref] , the consequence of AMP-induced stress on bacterial physiology is less studied. The first goal of this study, therefore, is to understand how the pathogen Staphylococcus aureus reacts to different doses of pexiganan at the minimum inhibitory concentration (MIC). Pexiganan is a drug that was mostly developed against this bacterium [bib_ref] In vitro antibacterial properties of pexiganan, an analog of magainin, Ge [/bib_ref]. This molecule is a 22-amino-acid peptide (Gly-Ile-Gly-Lys-Phe-Leu-Lys-Lys-Ala-Lys-Lys-Phe-Gly-Lys-Ala-Phe-Val-Lys-Ile-Leu-Lys-Lys-NH2); molecular weight, 2478 Da [free peptide base] and has cationic nature. It is a derivative analog of the magainin II peptide isolated from the skin of the African clawed frog Xenopus laevis. Pexiganan exhibited broadspectrum antibacterial activity in vitro when tested against 3,109 clinical isolates of gram-positive and gram-negative, anaerobic and aerobic bacteria [bib_ref] In vitro antibacterial properties of pexiganan, an analog of magainin, Ge [/bib_ref]. Pexiganan shows a barrel-stave type mechanism of membrane disruption (or channel formation). The consensus is that Pexiganan exerts its antibacterial effect by forming toroidal pores in the bacterial membranes. Pexiganan effectively induced the uptake and leakage of small compounds from both bacterial membranes and in vitro assembled lipid vesicles [bib_ref] Structure, membrane orientation, mechanism, and function of pexiganan -a highly potent antimicrobial..., Gottler [/bib_ref].
Using pexiganan as an example, we found that different concentrations induce the upregulation of several genes depending on nucleotides or related to nucleotide metabolism. Based on these results, we hypothesized that this would lead us to identify phenotypic collateral sensitivity. We hypothesized that the response to pexiganan sensitizes S. aureus against certain nucleoside antimetabolites or toxic nucleoside analogs. Interestingly, these analogs have been proposed as an alternative to antibiotics as a consequence of resistance emergence [bib_ref] Nucleoside analogues as antibacterial agents, Thomson [/bib_ref]. Nucleoside analogs have the advantage of being clinically approved for cancer therapies, but also as antiviral and antifungal treatments [bib_ref] Nucleoside analogues as antibacterial agents, Thomson [/bib_ref]. Pyrimidine and purine analogs, as we use here, showed potent antimicrobial activity against S. aureus in the past [bib_ref] Nucleoside analogues as antibacterial agents, Thomson [/bib_ref] [bib_ref] On the initial stage in peptidoglycan synthesis. Effect of 5-fluorouracil substitution on..., Stickgold [/bib_ref] [bib_ref] Gemcitabine is active against clinical multiresistant Staphylococcus aureus strains and is synergistic..., Jordheim [/bib_ref] [bib_ref] 5-Fluorouracil and mucopeptide biosynthesis by Staphylococcus aureus, Rogers [/bib_ref].
In this study, we show how proteomic changes in S. aureus in response to low-dose pexiganan uncover cellular soft spots that help to identify intervention opportunities. In addition, our findings contribute to the understanding of the early stages of resistance evolution to antimicrobial peptides. Here, we first study the global proteomic response of S. aureus to the cationic antimicrobial peptide pexiganan at concentrations similar to and below MIC. This helps us to detect the possible metabolic changes that open the path to collateral sensitivity to nucleoside analogs. We then confirm that these treatments sensitize S. aureus to antimetabolite purine and pyrimidines analogs.
# Materials and methods
## Bacteria and growth conditions
We used S. aureus SH1000 [bib_ref] δb modulates virulence determinant expression and stress resistance: characterization of a functional..., Horsburgh [/bib_ref] for all experiments. Bacteria were cultured in non-cation-adjusted (unsupplemented) Mueller-Hinton broth (MHB) as recommended for antimicrobial peptides susceptibility testing [bib_ref] In vitro susceptibility tests for cationic peptides: comparison of broth microdilution methods..., Giacometti [/bib_ref].
## Global proteomics by lc-mass spectrometry
Staphylococcus aureus strain SH1000 was grown in non-cationadjusted MHB to the mid-exponential-phase (OD 600 0.5) at 37 - C with vigorous shaking. The cultures were diluted 100 times in fresh MHB in a separate tube to a final volume of 5 ml. Pexiganan was added to tubes for a final concentration of 0.5, 1, 2 and 4 µg/ml (1/8,1/4, 1/2, 1x MIC, respectively) in a final culture volume of 10 ml per tube. Non-treated samples were used as controls. After the addition of pexiganan, all tubes were incubated for 30 min with moderate shaking at 37 - C. The pellets were collected by centrifugation at 10,000 × g for 5 min and the supernatant was removed by aspiration using a sterile vacuum line. Fifty microlitre of denaturation urea buffer (6 M urea/2 M thiourea/10 mM HEPES, pH 8.0) were then added to each pellet. The resulting suspensions were transferred to new 1.5 ml Eppendorf tubes and exposed to 5 freeze-thawing cycles alternating between liquid nitrogen and a 37 - C water bath. The tubes were centrifuged at 20,000 × g for 10 min and the resulting supernatants were transferred to fresh tubes and used as starting protein material for digestion. Each experimental condition had six independent biological replicates. Approximately 50 µg proteins were processed per sample and were in-solution digested as described elsewhere [bib_ref] Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using..., Rappsilber [/bib_ref]. Denaturation buffer-containing protein solutions were reduced by adding 1 µl of 10 mM DTT (final concentration) and incubated for 30 min. The reactions were then alkylated by adding 1 µl of 55 mM iodoacetamide and incubated for 20 min in the dark. Lysyl endopeptidase (LysC, Wako, Japan) resuspended in 50 mM ABC was added to the digestion reaction in a proportion of 1 µg per 50 µg of total sample protein and incubated for 3 h. The samples were diluted with four volumes of 50 mM ammonium bicarbonate (ABC) and digested overnight with 1 µg of sequencing grade modified trypsin (Promega, USA). All digestion steps were performed at room temperature. The next day, the digestions were stopped by adding final concentrations of 5% acetonitrile and 0.3% trifluoroacetic acid (TFA). The samples were desalted using the Stage-tip protocol as described previously [bib_ref] Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using..., Rappsilber [/bib_ref] , and the eluates were vacuum-dried. Peptides were reconstituted in 10 µl of 0.05% TFA, 2% acetonitrile, and 6.4 µl were analyzed by a reversed-phase capillary nano liquid chromatography system (Ultimate 3000, Thermo Scientific) connected to an Orbitrap Velos mass spectrometer (Thermo Scientific). Samples were injected and concentrated on a trap column (PepMap100 C18, 3 µm, 100 Å, 75 µm i.d. × 2 cm, Thermo Scientific) equilibrated with 0.05% TFA, 2% acetonitrile in water. After switching the trap column inline, LC separations were performed on a capillary column (Acclaim PepMap100 C18, 2 µm, 100 Å, 75 µm i.d. × 25 cm, Thermo Scientific) at an eluent flow rate of 300 nl/min. Mobile phase A contained 0.1% formic acid in water, and mobile phase B contained 0.1% formic acid in acetonitrile. The column was pre-equilibrated with 3% mobile phase B followed by an increase of 3-50% mobile phase B in 50 min. Mass spectra were acquired in a data-dependent mode utilizing a single MS survey scan (m/z 350-1,500) with a resolution of 60,000 in the Orbitrap, and MS/MS scans of the 20 most intense precursor ions in the linear trap quadrupole. The dynamic exclusion time was set to 60 s and automatic gain control was set to 1 × 10 6 and 5,000 for Orbitrap-MS and LTQ-MS/MS scans, respectively.
MS and MS/MS raw data were analyzed using the MaxQuant software package (version 1.6.4.0) with an implemented Andromeda peptide search engine [bib_ref] The MaxQuant computational platform for mass spectrometry-based shotgun proteomics, Tyanova [/bib_ref]. Data were searched against the reference proteome of S. aureus strain NCTC 8352 downloaded from Uniprot (2,889 proteins, taxonomy 93061, last modified September 2017) using label-free quantification and the match between runs option was enabled. Filtering and statistical analysis was carried out using the software Perseus [bib_ref] The Perseus computational platform for comprehensive analysis of (prote)omics data, Tyanova [/bib_ref]. Only proteins with intensity values from at least 3 out of 6 replicates were used for downstream analysis. Missing values were replaced from normal distribution (imputation) using the default settings (width 0.3, down shift 1.8). Student's T-tests were performed using permutation-based FDR of 0.05.
## Antimetabolite nucleosides
In this study, we used four nucleoside analogs. We used the pyrimidine analogs 6-azauracil, gemcitabine, 5-fluorouracil and the purine analog 6-thioguanine. All drugs were purchased from Sigma Aldrich (Germany). 6-azauracil is used as a growth inhibitor of various microorganisms via depletion of intracellular GTP and UTP nucleotide pools [bib_ref] The effect of 6-azauracil on microorganisms inhibited by chloramphenicol, Habermann [/bib_ref]. Gemcitabine is a chemotherapy medication used to treat different types of cancer. Gemcitabine is a synthetic pyrimidine nucleoside analog in which the hydrogen atoms on the 2 ′ carbon of deoxycytidine are replaced by fluorine atoms and competitively takes part and disrupts several pathways where pyrimidines are needed [bib_ref] Gemcitabine is active against clinical multiresistant Staphylococcus aureus strains and is synergistic..., Jordheim [/bib_ref]. 5-Fluorouracil is also used as an anticancer treatment and it works by inhibiting cell metabolism by blocking many pathways, but its major action is the inhibition of the thymidylate synthase. By doing so, the synthesis of the pyrimidine thymidine is stalled, which is an essential nucleoside required for DNA replication [bib_ref] The complex mechanism of antimycobacterial action of 5-fluorouracil, Singh [/bib_ref]. 5-Fluorouracil causes a drop on dTMP, causing cells to undergo cell death via thymineless death [bib_ref] The complex mechanism of antimycobacterial action of 5-fluorouracil, Singh [/bib_ref] [bib_ref] Thymineless death lives on: new insights into a classic phenomenon, Khodursky [/bib_ref].
## Pexiganan and antimetabolite nucleosides susceptibility testing
Minimal inhibitory concentration (MIC) was determined by broth micro-dilution method modified for cationic antimicrobial peptides [bib_ref] Agar and broth dilution methods to determine the minimal inhibitory concentration (MIC)..., Wiegand [/bib_ref]. Briefly, 2 µl of the mid-exponential phase culture diluted 1:100 (around 10 5 bacteria) were inoculated into each well of a polypropylene V-bottom 96-well plates with antievaporation ring lids (Greiner Bio-One GmbH, Germany). Prior to inoculation, pexiganan and the analogs (a kind gift from Dr. Michael A. Zasloff, Georgetown University) were 2-fold serially diluted in a final volume of 100 µl MHB per well using 32 µg/ml as starting concentration. Each assay was performed with eight replications and plates were incubated at 37 - C in a humid chamber. The MIC was defined as the lowest concentration where no visible bacterial growth was observed after 24 h.
## Isobologram assay
The combined activity and interactions between peptides, pexiganan, purine and pyrimidine analogs against S. aureus in MHB were determined using isobolographic combinations, also called the checkerboard assay method, (8 × 8 matrix of concentrations combinations) [bib_ref] An overview of drug combination analysis with isobolograms, Tallarida [/bib_ref]. In a 96-well plate, 50 µl of pexiganan at 4x MIC concentration was 2-fold serially diluted ranging from 32 to 0.25 µg/ml in the direction of the columns from 1 to 8. In another 96-well plate, 100 µl of nucleoside analogs at 8x MIC concentrations were prepared identically to the previous plate, but diluted in the direction of the rows from A to H. Half of the content (50 µl) of each well from the analog drug plate was transferred to the corresponding well of the plate containing pexiganan in an equal 1:1 mix fashion, halving the concentration of both compounds. In the same plate, the columns 9 and 10 were used to serially dilute both the peptide and the analog drug in the same concentrations that were present in the combination to compare single compounds vs. combination. Columns 11 and 12 were used as a control, by inoculating column 11 wells with bacteria without any drug and leaving columns 12 only with the same volume of MHB as a media contamination control. Each plate was prepared in triplicates to check for consistency. The bacterial suspension was prepared by growing S. aureus SH1000 to mid-exponential phase (2.5 h, with moderate shaking at 37 - C) in MHB to an OD 600 between 0.3 and 0.5. The bacterial suspension was diluted in MHB and ∼1 × 10 6 bacteria were inoculated in each well. After 24 h of incubation at 37 - C in a humid chamber, the plates were visually examined for growth. The Fractional inhibitory concentration (FIC index) for a combination of pexiganan and each antimetabolite drug was calculated as [(MIC of the peptide in combination with a given analog)/(MIC of peptide alone)] + [(MIC of analog in combination with peptide)/(MIC of analog alone)]. The interpretation of the results was as follow: FIC ≤ 0.5, synergistic; 0.5 < FIC ≤ 1, additive; 1 < FIC ≤ 4, indifferent; FIC > 4, antagonistic [bib_ref] Rational design and synthesis of modified teixobactin analogues: in vitro antibacterial activity..., Ng [/bib_ref]. To ensure that bacteria lost viability while reading MIC values for pexiganan-analog combinations, we used the resazurin colorimetric assay as described previously with minor modifications [bib_ref] Resazurin-based 96-well plate microdilution method for the determination of minimum inhibitory concentration..., Elshikh [/bib_ref]. Resazurin (THK, Germany) was prepared at 0.015 % in distilled water and sterilized by filtration. It was stored at 4 - C for a maximum of 1 week after preparation. Resazurin (0.015%) was added to each well (10 µl per well, 1/3 of the original described quantity) and further incubated for 3 h for the observation of color change. Columns with no color change (blue resazurin) were scored as dead culture. In contrast, color change to purple (reduced resazurin) was considered as a sign of viability.
## Time-kill experiments
Starting from early mid-exponential phase cultures (1 × 10 7 CFU/ml), bacteria were exposed to growing concentrations of pexiganan ranging from 1 to 8x MIC or pexiganan combined with the nucleoside analogs 6-azauracil, gemcitabine, 5-fluorouracil and 6-thioguanine at their respective 1/2x MIC values. The cultures were incubated with soft shaking at 37 - C for 2 h. Samples from each culture (1 ml) were taken at 20min time-point intervals. The samples were diluted and plated to determine cell viability. The experiments consisted of five independent replicates. Non-treated cells were used as a control.
# Statistical analysis
The effect of treatments on bacterial killing was analyzed using R package nparLD which is designed to perform non-parametric analysis of longitudinal data in factorial experiments modeling the variation over time [bib_ref] nparLD: an R software package for the nonparametric analysis of longitudinal data..., Noguchi [/bib_ref]. P values of ≤0.05, after correction, if needed, were considered statistically significant. All tests were performed with the statistic software R [bib_ref] R: A Language and Environment for Statistical Computing, R Core Team [/bib_ref].
# Results
## Changes in protein profiles after pexiganan treatment
We examined S. aureus exposed to pexiganan by studying proteome-wide changes after a 30-min treatment with different pexiganan concentrations (0.125, 0.25, 0.5, and 1x MIC, . Overall, 1160 proteins were identified at a 1% or less false discovery rate (FDR) among which 968 proteins were quantified in at least 3 out of 6 replicates and used for downstream analysis. All identified proteins, their quantification and statistical tests are provided in . A global overview shows a proteome-wide perturbation induced by pexiganan stress for all concentrations compared to control. Many proteins were significantly differentially expressed [fig_ref] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid... [/fig_ref]. It is noticeable that as long as the dose increases, the level of expression (foldchange) of both overexpressed and suppressed genes, decreases, making the dot scattering of the volcano plot less disperse [fig_ref] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid... [/fig_ref]. This indicates a decrease in the ability of the cell to react with increasing peptide concentration.
## Envelope stress response to amps
Within the upregulated proteome fraction [fig_ref] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid... [/fig_ref] and [fig_ref] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid... [/fig_ref] , a group of proteins related to osmotic stress response shows up. The multi-peptide resistance factor MprF, a protein associated with cationic peptide resistance, which is conserved among many bacterial species [bib_ref] MprFmediated lysinylation of phospholipids in Staphylococcus aureus leads to protection against oxygen-independent..., Kristian [/bib_ref] [bib_ref] Bacterial resistance to antimicrobial host defenses -an emerging target for novel antiinfective..., Weidenmaier [/bib_ref] is upregulated in all pexiganan doses except in the lowest dose (1/8x MIC). MprF catalyzes the transfer of a lysyl group from L-lysyl-tRNA(Lys) to membrane-bound phosphatidylglycerol producing lysyl-phosphatidylglycerol, a major component of the bacterial membrane with a net positive charge. Hence, a modification of the anionic phosphatidylglycerol with positively charged L-lysine results in the repulsion of the peptides. Changes in the membrane charge is a per se resistance mechanism against cationic antimicrobial peptides (42). Thus, MprF increases resistance to moenomycin and vancomycin but also to human defensins (HNP1-3) and contributes to the evasion of oxygen-independent neutrophil killing and other AMPs and antibiotics [bib_ref] Characterization of the Staphylococcus aureus mprF gene, involved in lysinylation of phosphatidylglycerol, Oku [/bib_ref] [bib_ref] MprFmediated biosynthesis of lysylphosphatidylglycerol, an important determinant in staphylococcal defensin resistance, Staubitz [/bib_ref]. Another highly expressed protein is CapF, which is involved in the pathway capsule polysaccharide biosynthesis, a mucous layer on the surface of the bacterium that facilitates immune evasion and infection. CapF is an important virulence factor during infections by S. aureus. The enzyme CapF is considered a therapeutic candidate to disrupt the capsule polysaccharide biosynthesis [bib_ref] Dynamic elements govern the catalytic activity of CapE, a capsular polysaccharide-synthesizing enzyme..., Miyafusa [/bib_ref]. Another protein that contributes toward modifying the bacterial envelope and has a significant higher expression is TagG. This protein is part of the wall teichoic acid synthesis during the final steps of the pathway. Wall teichoic acids are important in pathogenesis and play key roles in antimicrobial resistance [bib_ref] Bacterial resistance to antimicrobial host defenses -an emerging target for novel antiinfective..., Weidenmaier [/bib_ref] [bib_ref] Wall teichoic acids of gram-positive bacteria, Brown [/bib_ref].
## Proteases and chaperones proteins
The chaperones/proteases ClpL and TreP are among the fifty upregulated genes for the dose corresponding to the MIC (8 µg/ml). ClpL is an ATP-dependent Clp protease. Clp proteases play a central role in stress survival, virulence and antibiotic resistance of S. aureus [bib_ref] Clp chaperones and proteases are central in stress survival, virulence and antibiotic..., Frees [/bib_ref]. Another protease induced by pexiganan is PepT, also known as Staphopain A. This enzyme is a cysteine protease that plays an important role in the inhibition of host innate immune response. It cleaves host elastins from connective tissues, pulmonary surfactant protein A in the lungs, and the chemokine receptor CXCR2 on leukocytes [bib_ref] Staphylococcus aureus proteases degrade lung surfactant protein A potentially impairing innate immunity..., Kantyka [/bib_ref]. Proteolytic cleavage of surfactant protein A impairs bacterial phagocytosis by neutrophils while CXCR2 degradation blocks neutrophil activation and chemotaxis [bib_ref] Staphylococcus aureus proteases degrade lung surfactant protein A potentially impairing innate immunity..., Kantyka [/bib_ref] [bib_ref] Degradation of elastin by a cysteine proteinase from Staphylococcus aureus, Potempa [/bib_ref]. Additionally, PepT promotes vascular leakage by activating the plasma kallikerin/kinin system, resulting in patient hypotension [bib_ref] Induction of vascular leakage through release of bradykinin and a novel kinin..., Imamura [/bib_ref].
## Alteration of metabolism
For example, NptA, a phosphate transporter, usually induced by phosphate limitation, is highly abundant in post-pexiganan treatment. Inorganic phosphate acquisition via NptA is particularly important for the pathogenesis of S. aureus. NptA homologs are widely distributed among bacteria and closely related less pathogenic staphylococcal species do not possess this importer. Another phosphate metabolism-related gene with high expression is SAOUHSC_00480, that codes for a putative nucleoside triphosphate pyrophosphohydrolase. Also related to phosphate metabolism, we observed a high level of FruA in different pexiganan concentrations. This protein is a phosphoenolpyruvate-dependent sugar phosphotransferase system (a PTS system) is a major carbohydrate active transport system, which catalyzes the phosphorylation of incoming sugar substrates concomitantly with their translocation across the cell membrane and are potentially important for survival in the respiratory tract of the host [bib_ref] Fructose transport-deficient Staphylococcus aureus reveals important role of epithelial glucose transporters in..., Garnett [/bib_ref]. GlcB, another PTS system is a phosphoenolpyruvate-dependent sugar phosphotransferase system. This protein catalyzes the phosphorylation of incoming sugar substrates and their translocation across the cell membrane [bib_ref] Expanded glucose import capability affords Staphylococcus aureus optimized glycolytic flux during infection, Vitko [/bib_ref]. Another two phosphate metabolism related proteins, CarA and CarB, which participate in the L-arginine biosynthesis, were induced. They are involved in the first step of the sub-pathway that synthesizes carbamoyl phosphate from bicarbonate. The elevation of these enzymes could indicate that pexiganan stress may be involved in amino acid depletion.
The gene SAOUHSC_00456 that codes for YabA is significantly increased as well by pexiganan. YabA is involved in the initiation of chromosome replication and is a negative controller of DNA replication initiation in Bacillus subtillis. YabA and DnaD inhibit helix assembly of the DNA replication initiation protein DnaA [bib_ref] YabA and DnaD inhibit helix assembly of the DNA replication initiation protein..., Scholefield [/bib_ref]. The elevated concentration of YabA could stall the cell division while the bacteria is under severe stress. S. aureus upregulates Spermidine/putrescine import ATP-binding protein PotA. This protein is part of the ABC transporter complex PotABCD and responsible for Frontiers in Immunology | www.frontiersin.org 5 September 2020 | Volume 11 | Article 1686 energy coupling to the transport system. Spermidine and putrescine are polyamines whose role in S. aureus is not well-defined [bib_ref] Polyamines: Emerging players in bacteria-host interactions, Martino [/bib_ref]. There are also a set of up-regulated proteins coded by the genes SAOUHSC_01717, SAOUHSC_02581, and SAOUHSC_02581 whose functions remain unknown as described in Uniprot database and showed no homology with any known sequence (51).
One of the hallmarks of our proteomic dataset is that we found a higher level of expression, compared to controls, for proteins related with nucleotide metabolism [fig_ref] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid... [/fig_ref] , which is directly connected to the upregulation of phosphate metabolism proteins described above. GmK, for example, is an essential protein for recycling GMP and indirectly, cGMP Guanylate kinase is near four times more abundant than in the control group. GMK is an essential enzyme and a potential antimicrobial drug target owing to its role in supplying DNA and RNA precursors [bib_ref] Structural biology and crystallization communications structure of Staphylococcus aureus guanylate monophosphate kinase, Omari [/bib_ref]. Another nucleobase metabolism-related protein having or exhibiting a higher expression for the 1x MIC treated cells is PyrG. This enzyme catalyzes the ATP-dependent amination of UTP to CTP with either L-glutamine or ammonia as the source of nitrogen. It also regulates intracellular CTP levels through interactions with the four ribonucleotide triphosphates.
## Downregulation response to pexiganan
Pexiganan also negatively impacted proteome-wide gene expression [fig_ref] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid... [/fig_ref] and . Among the most affected gene expressions throughout all concentrations are genes such as dps (coding for a known iron storage protein), hld, copZ, cspC, metQ, sceD, isaA csoB/C, dltC, adsA and sasG, SAOUHSCA_01134 and SAOUHSCA_02576. The gene cspB codes for the downregulated protein CspD, a cold shock protein that accumulates during low temperature or cold shock. This gene is also a component of the stringent response, indicating that it could be a general stress response gene [bib_ref] Characterization of the Staphylococcus aureus heat shock, cold shock, stringent, and SOS..., Anderson [/bib_ref]. Other genes showing a differentially low level of expression are SAOUHSC_01986, SAOUHSC_01986, SAOUHSC_008020, SAOUHSC_02093, SAOUHSC_02535, and SAOUHSC_01414 which code for uncharacterized proteins. SAOUHSC_01030 is a putative glutaredoxin domain-containing protein but it is not characterized either. The gene SAOUHSC_02576 codes for a putative secretory antigen SsaA, identified in S. epidermidis but its function is also unknown.
In contrast to the upregulation of peptidoglycan synthesis, we observed that putative peptidoglycan hydrolases and probable lytic transglycosylases IsaA and SceD were downregulated. Interestingly, the isaA sceD double mutant is attenuated for virulence, while SceD is essential for nasal colonization in cotton rats [bib_ref] Characterization of IsaA and SceD, two putative lytic transglycosylases of Staphylococcus aureus, Stapleton [/bib_ref]. The gene moaD shows also a reduced level of expression and it codes for a molybdopterin converting factor subunit 1. Molybdopterins are a class of cofactors found in most molybdenum-containing and all tungstencontaining enzymes. Molybdopterin pathway reactions consume guanosine triphosphate that is converted into the cyclic phosphate of pyranopterin [bib_ref] The biosynthesis of the molybdenum cofactors, Mendel [/bib_ref]. Another metabolic enzyme, AldA, aldehyde dehydrogenase central carbohydrate metabolism is downregulated in all doses of pexiganan. This is also the case of CopZ, a chaperone that serves for the intracellular sequestration and transport of copper, delivering it to the copper-exporting P-type ATPase A (CopA) [bib_ref] Molecular characterization of the copper transport system in Staphylococcus aureus, Sitthisak [/bib_ref].
## Pexiganan stress has a strong impact on the essential proteome
We visualized the global impact of pexiganan stress (at 1x MIC) on bacterial physiology by using a network analysis based on protein-protein interactions and the function (61) of S. aureus essential genes [fig_ref] FIGURE 3 |: Isobolographical response of pexiganan combination with different antimetabolite nucleosides [/fig_ref]. This network analysis provides global view information on protein level alterations and integrates protein-protein interactions, including indirect functional and direct physical associations [bib_ref] STRING v10: protein-protein interaction networks, integrated over the tree of life, Szklarczyk [/bib_ref]. At this concentration, it is noticeable that the majority of the essential genes are downregulated, and it is possible that this pattern has a strong influence on pexiganan lethality. The majority of upregulated proteins are ribosomal components.
## Gene ontology analysis points to an upregulation of nucleotide metabolism
The signature of pexiganan stress on S. aureus in the upregulated fraction points to nucleotide metabolism-related genes. GO annotation allows enrichment analysis providing global information based on the gene expression levels by proteomics or transcriptomics or other gene expression datasets [bib_ref] PANTHER version 14: more genomes, a new PANTHER GO-slim and improvements in..., Mi [/bib_ref]. We focus this comparative analysis on the protein expression levels of the 100 most upregulated proteins of every pexiganan dosage. We focussed on categorizing by pathways. Some of the upregulated pathways involved genes related to oxidative stress, peptidoglycan synthesis, and Nacetylglucosamine that are expected from cationic antimicrobial peptides since they attack the cell envelopes. In addition, there is a reactivation of the central metabolism by the upregulation of genes from glycolysis, TCA cycle, arginine, and thiamine synthesis. However, the most enriched pathways in the GO analysis for all pexiganan concentrations were related to nucleotide metabolism [fig_ref] FIGURE 2 |: Functional characterization of pathways of up-regulated proteins in S [/fig_ref]. The nucleotide upregulated pathways include ATP synthesis, Adenine and hypoxanthine salvage pathways, de novo synthesis of purines and pyrimidines and S-adenosylmethionine. This result confirms that pexiganan stress induces a scarcity of these metabolites within the cell. Taking into account the previous results, we hypothesized that upregulation of nucleotide-dependent and related genes could create a collateral sensitivity.
## Nucleoside analogs antimetabolites act synergistically with pexiganan
We designed a simple drug interaction experiment between pexiganan and some nucleoside analogs including the purine and pyrimidine antimetabolites: 6-azauracil, gemcitabine, 5-fluorouracil, and 6-thioguanine [fig_ref] FIGURE 3 |: Isobolographical response of pexiganan combination with different antimetabolite nucleosides [/fig_ref] and . This experiment is the classic isobologram, also known as a checkerboard assay [bib_ref] An overview of drug combination analysis with isobolograms, Tallarida [/bib_ref]. All analogs showed a synergistic activity when combined with pexiganan . The most active ones were 5-fluorouracil and gemcitabine and, while the 6-azauracil and 6-thiogunine showed a milder effect according to their respective fractional inhibitory concentration index . All the combinations managed to decrease of the minimal inhibitory concentration for each drug when compared to the respective drug alone. These results indicate that pexiganan induces a strong collateral sensitivity to nucleoside analogs.
To study the influence of nucleoside analogs on the killing by pexiganan, we carried out a time-kill experiment combining each of 6-azauracil, gemcitabine, 5-fluorouracil, and 6-thioguanine with pexiganan. We assayed all drugs using half of the minimal inhibitory concentration. We exposed mid-exponential phase S. aureus cells to these combinations and sampled the viability of the cultures every 20 min [fig_ref] FIGURE 4 |: Pexiganan-nucleoside antimetabolite combinations drastically increases the killing capacity of pexiganan [/fig_ref]. All compounds significantly increased the killing ability of pexiganan, gemcitabine and 5fluorouracil being the most active drugs. The killing rate was increased by a few orders of magnitude in all combinations. The killing by the combination of gemcitabine or 5-fluorouracil with pexiganan, at their corresponding half MIC values, was more efficient than 8x MIC concentration of pexiganan alone. The viability was assessed not only by the absence of growth but also by the addition of resazurin, a reagent that turns from blue to purple when it is reduced by microbial enzymes that only work within living bacteria (37).
# Discussion
We have found that pexiganan, a cationic antimicrobial peptide, can induce a stress response in S. aureus that results in a proteome-wide impact. Pexiganan treatment upregulates known virulence factors such as MprF, the capsule synthesis protein CapF, a wall teichoic acid TagG, the proteases ClpL and PepT and other proteins important for the interactions with the hosts. This could lead to a phenotypic cross-tolerance of other immune effectors of hosts and possibly complicate the bacterial infection in case of inefficient treatment where bacteria could be exposed to sub-lethal concentrations. This is a legitimate concern since AMP-resistant variants have been reported to have evolved which have shown some cross-resistance with immune system effectors [bib_ref] Arming the enemy: the evolution of resistance to self-proteins, Bell [/bib_ref] [bib_ref] Induced bacterial cross-resistance toward host antimicrobial peptides: a worrying phenomenon, Fleitas [/bib_ref]. This risk has been shown for pexiganan as well [bib_ref] Therapeutic antimicrobial peptides may compromise natural immunity, Habets [/bib_ref]. Our data also provides input about possible induced physiological changes that would help S. aureus to adapt to the intra-host environment during its interaction with specific immune system effectors.
It is important to note that, given the coverage of the proteomic data and range of pexiganan doses, we did not find evidence of activation of mutagenic stress pathways or recombination. This indicates that the mode of killing by cationic antimicrobial peptide does not increase genome instability as is typical for classic antibiotics [bib_ref] Antimicrobials as promoters of genetic variation, Blázquez [/bib_ref]. We have previously shown and proposed that antimicrobial peptides, including pexiganan, do not increase the rate of either mutagenesis [bib_ref] Antimicrobials, stress and mutagenesis, Rodríguez-Rojas [/bib_ref] or recombination [bib_ref] Cationic antimicrobial peptides do not change recombination frequency in Escherichia coli, Rodríguez-Rojas [/bib_ref] in Gram-negative bacteria. Our findings here are consistent with these observations in the Gram-positive model bacterium S. aureus.
The elevated level of expression of proteins such as GmK, PyrG, NptA and some amino acids-biosynthesis enzymes such as CarA and CarB that participate in the biosynthesis of L-arginine could be explained by changes in permeability. Amino acids, nucleobases and nucleotides are small molecules that could easily escape from the cellular compartment in case of membrane damage. This is a well-known property of cationic agents, including AMPs [bib_ref] Dissection of antibacterial and toxic activity of melittin: a leucine zipper motif..., Asthana [/bib_ref] [bib_ref] Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria?, Brogden [/bib_ref] [bib_ref] Cationic peptides facilitate iron-induced mutagenesis in bacteria, Rodríguez-Rojas [/bib_ref]. The fact that only a few proteins from the amino acids biosynthesis pathways are upregulated could be explained because the experiments were carried out in a complex medium like MHB that contains several amino acids and bacteria would upregulate only necessary pathways. A similar situation might be expected within a host.
The upregulation of the phosphate and nucleotide-related proteins provides a direction to investigate drug susceptibilities created by pexiganan stress. Although the antimetabolites used in this work have good antibacterial activity, if they are used in monotherapy they are also prone to generate resistance [bib_ref] Nucleoside analogues as antibacterial agents, Thomson [/bib_ref] [bib_ref] Gemcitabine is active against clinical multiresistant Staphylococcus aureus strains and is synergistic..., Jordheim [/bib_ref]. Thus, their use in combination could possibly help to prevent resistance [bib_ref] Combination effects of antimicrobial peptides, Yu [/bib_ref] [bib_ref] Drug combinations: a strategy to extend the life of antibiotics in the..., Tyers [/bib_ref].
The synergistic combined action of pexiganan with nucleoside antimetabolites could be potentially explained by two underlying mechanisms. First, pexiganan stress forces a response by S. aureus that upregulates nucleobase salvage pathways and other nucleotide-dependent metabolic pathways. Second, pexiganan has the potential to change membrane permeability and induce the uptake of such metabolites even at sublethal concentrations possibly leading to much higher intracellular concentrations . We have shown previously that cationic antimicrobial peptides can mediate the uptake of small molecules due to changes in permeability at sublethal concentrations [bib_ref] Cationic peptides facilitate iron-induced mutagenesis in bacteria, Rodríguez-Rojas [/bib_ref]. The more potent activity of gemcitabine and 5-fluorouracil could be explained because they act on the cell walls as previously reported [bib_ref] Gemcitabine is active against clinical multiresistant Staphylococcus aureus strains and is synergistic..., Jordheim [/bib_ref] [bib_ref] Effect of 5-fluorouracil, mitoxantrone, methotrexate, and vincristine on the antibacterial activity of..., Gieringer [/bib_ref]. An additional potential therapeutic advantage of the nucleoside analogs studied here is that all the clinical properties of these drugs are well-known, including toxicological profile, pharmacological activities and metabolizing properties [bib_ref] Nucleoside analogues as antibacterial agents, Thomson [/bib_ref] [bib_ref] Repurposing screen identifies unconventional drugs with activity against multidrug resistant Acinetobacter baumannii, Cheng [/bib_ref]. All of them are approved drugs, which should facilitate the introduction of such combinations in clinical practices.
We have shown recently that antimicrobial peptides, including pexiganan, can induce priming in bacteria, an enhanced response to the peptides when bacteria are preexposed to low concentrations. We defined the priming response as the ability of bacteria to have better survival to the peptide when it has been exposed to sub-inhibitory concentration in advance. The consequence of priming is not only survival but an increase in tolerance and persistence. Tolerance and persistence are a non-genetic path that increase survival to antimicrobials and lead to infection control failure [bib_ref] Definitions and guidelines for research on antibiotic persistence, Balaban [/bib_ref]. It has been proposed that evolution of tolerance in response to sub-inhibitory concentrations of antibiotics precedes or enhances the emergence of resistance [bib_ref] Antibiotic tolerance facilitates the evolution of resistance, Levin-Reisman [/bib_ref]. The use of antimetabolites could potentially abolish this property in therapeutic usage.
# Conclusions
The analysis of the pexiganan stress response by S. aureus has shown a global response involving several proteins known for their role in the development of resistance against FIGURE 5 | A general model illustrating the positive interaction between pexiganan and nucleoside antimetabolites against S. aureus. The interactions of pexiganan with the membrane at sub-inhibitory concentrations lead to transient permeability changes in the envelope that promote leakage of small molecules such as nucleotides, nucleobases or nucleosides. Simultaneously, other small molecules such as toxic nucleoside analogs can increase the diffusion rate toward the intracellular compartment. This stress is sensed by the cell that responds by activating nucleoside metabolism creating an intervention opportunity. In this situation, toxic nucleoside antimetabolites are more efficiently incorporated into RNA, DNA, and other nucleotide depending reactions that may include envelope synthesis, enhancing toxicity and leading to faster cell killing.
antimicrobial peptides and other immune system effectors. Pexiganan has also shown a synergistic increase of antibacterial activity when it is combined with nucleoside antimetabolites. Taken together, our results suggest that pexiganan renders S. aureus susceptible to purine and pyrimidine analogs, which are traditionally used for cancer treatment. These antimetabolites can enhance the bactericidal activity of pexiganan against S. aureus under the tested conditions. The significant potentiation of the pexiganan bactericidal activity and the decrease of minimal inhibitory concentrations when compared with pexiganan alone could be the basis for new formulations of pexiganan. These results are probably extendable to other antimicrobial peptides and other bacterial pathogens. Thus, the leakage of nucleotides and intermediate small metabolites or cofactors caused by cationic peptides and nucleotide metabolic pathways are common traits of bacteria-peptide interactions as proposed for the symbionthost interface [bib_ref] Metabolic integration of bacterial endosymbionts through antimicrobial peptides, Mergaert [/bib_ref]. Our results also show that understanding how antimicrobials operate and how pathogens respond to them is important to guide the design of new effective therapies. Physiological response by bacteria is informative or suggestive about additional drug combinations that can limit the chances of pathogens to evolve resistance while increasing pathogen clearance and decreasing toxicity. This approach should be exploited to rationally design new antimicrobial combinations.
# Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.
# Author contributions
AR-R and JR conceived the study. AR-R, AN, BE, GS, and BK performed the experiments and collected the data. AR-R, GS, JK, BK, and CW analyzed the data. AR-R and JR wrote the manuscript and revised the final document. All authors agree to be held accountable for the content therein and approved the final version.
# Acknowledgments
We would like to thank Dr. Dan Roizman from Freie Universität Berlin for help with Resazurin assay and critical reading of the manuscript. We would like to also thank Dr. Michael A. Zasloff from Georgetown University for kindly providing pexiganan.
# Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fimmu. 2020.01686/full#supplementary-material [fig_ref] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid... [/fig_ref] | Volcano plots -log q values vs. log2 fold change of protein intensity measured by LC-MS of pexiganan treated cells with different fractions of the MIC, each compared to an untreated control). Black dots represent not significant expressed proteins while green dots show the upregulated portions and red ones represent the down-regulated fraction. [fig_ref] FIGURE 2 |: Functional characterization of pathways of up-regulated proteins in S [/fig_ref] | Heatmap of relative protein expression based on label-free quantification by liquid chromatography/mass spectrometry (LC-MS). Only the 50 most statistically significant down-regulated proteins are shown, taking as a reference the ones from 1xMIC pexiganan concentration (0.125, 0.25, 0.5, and 1 fractions of the minimal inhibitory concentration). Intensity ranges of the log2 fold-changes are given from highest intensity (green) to lowest (red) sorted by their values for the 1x MIC. [fig_ref] FIGURE 3 |: Isobolographical response of pexiganan combination with different antimetabolite nucleosides [/fig_ref] | Network analysis of pexiganan stress (1x MIC) on essential genes interactome of S. aureus SH1000. Pale green nodes indicate upregulated proteins while pale red ones represent down-regulated ones. Gray nodes correspond with genes with a high degree of connectivity with this essential network, but they were not labeled. Note the higher proportion of downregulated genes among essential proteome while the majority of unregulated proteins are ribosomal components and thus they aggregate due to physical interaction. The interaction among nodes shows the proteome-wide impact of pexiganan stress at an inhibitory concentration. [fig_ref] FIGURE 4 |: Pexiganan-nucleoside antimetabolite combinations drastically increases the killing capacity of pexiganan [/fig_ref] | Isobologram showing the synergistic activity of pexiganan and different nucleotide antimetabolite combinations against S. aureus SH1000. Columns with no color change (blue resazurin) indicate no viable bacteria while color change to purple (reduced resazurin) was considered as a sign of bacterial growth. Please note that the red rectangle indicates the well of the plates used for the drug combination (8 × 8 wells) while the double arrows indicate the alongside single drug MIC determination. A typical plate is shown from three repetitions. The blue rectangles show the positive (same bacterial inoculum in LB medium) and negative (LB medium alone) controls.
[fig] FIGURE 1 |: Heatmap of relative protein expression based on label-free quantification detected by liquid chromatography-mass spectrometry (LC-MS). Only the 50 most significantly up-regulated proteins compared to the control are shown (log2 fold-change). Red rectangle highlights proteins that participate in or depend on nucleotide metabolism. Proteins were extracted after 30 min of pexiganan addition (0.125, 0.25, 0.5 and 1 fractions of the minimal inhibitory concentration). log2 fold-changes are given from highest (green) to lowest (red). [/fig]
[fig] FIGURE 2 |: Functional characterization of pathways of up-regulated proteins in S. aureus SH1000 at different concentrations of pexiganan (0.125, 0.25, 0.5, and 1 fractions of the minimal inhibitory concentration, MIC). For this analysis, only the 100 most highly differentially expressed proteins for each concentration of pexiganan were used. On top of each chart the number of identified genes and the number of pathway hits (number of genes used for the enrichment analysis) is visible. The analysis was carried out using the online gene ontology analysis software PANTHER(62). [/fig]
[fig] FIGURE 3 |: Isobolographical response of pexiganan combination with different antimetabolite nucleosides. The crosses indicate the presence of bacterial growth in the unique concentration combinations of each well. Blue rectangles indicate the MIC value for single drug situations (pexiganan or antimetabolites) and is marked as a reference to visually compare with the actual level of inhibition for each pexiganan-antimetabolite combination. [/fig]
[fig] FIGURE 4 |: Pexiganan-nucleoside antimetabolite combinations drastically increases the killing capacity of pexiganan. (A) Killing dynamic of S. aureus SH1000 at different concentrations of pexiganan using the MIC as the starting point. (B) Example data of time-kill experiment exposing mid-exponential phase bacterial cultures to pexiganan-nucleoside antimetabolite combinations (both at 1/2x MIC concentrations). The combination has a dramatic effect on the killing ability of pexiganan. Data points were determined by counting colony-forming units (CFU) at different time points. Mean ± SDM, n = 5. Asterisks represent significant differences (R package nparLD, one asterisk for p < 0.05 and two asterisks for p < 0.01 and three asterisks for p < 0.001). Only comparisons between pexiganan (1/2x MIC) and pexiganan-analogs combinations are shown. [/fig]
[fig] FUNDING: AR-R and JR were supported by SFB 973 (Deutsche Forschungsgemeinschaft, project C5). We acknowledge support by the German Research Foundation and the Open Access Publication Fund of Freie Universität Berlin. For mass spectrometry (BK and CW) we would like to acknowledge the assistance of the Core Facility BioSupraMol supported by the Deutsche Forschungsgemeinschaft (DFG). [/fig]
|
Pooled-Peptide Epitope Mapping Strategies Are Efficient and Highly Sensitive: An Evaluation of Methods for Identifying Human T Cell Epitope Specificities in Large-Scale HIV Vaccine Efficacy Trials
The interferon gamma, enzyme-linked immunospot (IFN-γ ELISpot) assay is widely used to identify viral antigen-specific T cells is frequently employed to quantify T cell responses in HIV vaccine studies. It can be used to define T cell epitope specificities using panels of peptide antigens, but with sample and cost constraints there is a critical need to improve the efficiency of epitope mapping for large and variable pathogens. We evaluated two epitope mapping strategies, based on group testing, for their ability to identify vaccineinduced T-cells from participants in the Step HIV-1 vaccine efficacy trial, and compared the findings to an approach of assaying each peptide individually. The group testing strategies reduced the number of assays required by >7-fold without significantly altering the accuracy of T-cell breadth estimates. Assays of small pools containing 7-30 peptides were highly sensitive and effective at detecting single positive peptides as well as summating responses to multiple peptides. Also, assays with a single 15-mer peptide, containing an identified epitope, did not always elicit a response providing validation that 15-mer peptides are not optimal antigens for detecting CD8+ T cells. Our findings further validate pooling-based epitope mapping strategies, which are critical for characterizing vaccineinduced T-cell responses and more broadly for informing iterative vaccine design. We also show ways to improve their application with computational peptide:MHC binding predictors that can accurately identify the optimal epitope within a 15-mer peptide and within a pool of 15-mer peptides.
# Introduction
Measures of immunogenicity are an integral part of iterative and rational vaccine design [bib_ref] The rational design of an AIDS vaccine, Douek [/bib_ref] [bib_ref] Statistical evaluation of HIV vaccines in early clinical trials, Moodie [/bib_ref].
The enzyme-linked immunospot (ELISpot) assay can be used to quantify the frequency of antigen-specific T cells via the secretion of interferon gamma (IFN-γ) following brief ex vivo stimulation with one or multiple peptide antigens [bib_ref] Detection and quantification of blood-derived CD8+ T lymphocytes secreting tumor necrosis factor..., Herr [/bib_ref] [bib_ref] Evaluation of the interferon-gamma ELISPOT-assay for quantification of peptide specific T lymphocytes..., Schmittel [/bib_ref]. The magnitude readout of the assay is the frequency of cytokine-secreting cells (spot-forming cells per million [SFC/M]), and although IFN-γ itself is not necessarily linked to CD8+/CD4+ T cell cytolytic activity [bib_ref] Correlation between interferon-gamma secretion and cytotoxicity, in virus-specific memory T cells, Horton [/bib_ref] , ELI-Spot responses identify antigen-specific T cells and have previously been associated with endpoints in vaccine clinical trials [bib_ref] Durable Human Memory T Cells Quantifiable by Cultured Enzyme-Linked Immunospot Assays Are..., Keating [/bib_ref] [bib_ref] Randomized, double-blind, phase 2a trial of falciparum malaria vaccines RTS,S/AS01B and RTS,S/AS02A..., Kester [/bib_ref] [bib_ref] A CD4(+) T-cell immune response to a conserved epitope in the circumsporozoite..., Reece [/bib_ref] [bib_ref] Correlation of cellular immune responses with protection against culture-confirmed influenza virus in..., Forrest [/bib_ref] [bib_ref] Correlation of antiviral T-cell responses with suppression of viral rebound in chronic..., Yang [/bib_ref] [bib_ref] Vaccine-induced gag-specific T cells are associated with reduced viremia after HIV-1 infection, Janes [/bib_ref]. One of the main advantages of the ELISpot assay over flow-cytometric assays (e.g., intracellular cytokine staining [ICS] [bib_ref] Comparison of the ELI-SPOT and cytokine flow cytometry assays for the enumeration..., Karlsson [/bib_ref] or tetramer sorting) is the ability to efficiently screen a wide array of peptide antigens covering the entire set of vaccine immunogens, thus effectively mapping the specificity of T-cell responses [bib_ref] T-cell epitope mapping using the ELISPOT approach, Anthony [/bib_ref]. Another benefit is that extensive effort has gone towards standardization of the assay across labs to reduce variability [bib_ref] Equivalence of ELISpot Assays Demonstrated between Major HIV Network Laboratories, Gill [/bib_ref] [bib_ref] Design and validation of an enzyme-linked immunospot assay for use in clinical..., Mwau [/bib_ref] [bib_ref] Moving to human immunodeficiency virus type 1 vaccine efficacy trials: defining T..., Russell [/bib_ref] , and towards statistical method development, which has improved positivity calls [bib_ref] Statistical evaluation of HIV vaccines in early clinical trials, Moodie [/bib_ref] [bib_ref] Statistical positivity criteria for the analysis of ELISpot assay data in HIV-1..., Hudgens [/bib_ref] [bib_ref] Mixture Models for Single-Cell Assays with Application to Vaccine Studies, Finak [/bib_ref] [bib_ref] Statistical positivity criteria for the analysis of ELISpot assay data in HIV-1..., Moodie [/bib_ref] [bib_ref] Response definition criteria for ELISPOT assays revisited, Moodie [/bib_ref] [bib_ref] Response determination criteria for ELISPOT: toward a standard that can be applied..., Moodie [/bib_ref] [bib_ref] Statistical considerations for the design and analysis of the ELISpot assay in..., Hudgens [/bib_ref].
Much of the work to develop and optimize the assay was motivated by a need to quantify vaccine-induced T-cell responses in HIV vaccine trials. Based on studies establishing associations between HLA class I alleles and slower progression to AIDS [bib_ref] Widespread impact of HLA restriction on immune control and escape pathways of..., Carlson [/bib_ref] [bib_ref] Influence of combinations of human major histocompatibility complex genes on the course..., Kaslow [/bib_ref] [bib_ref] Association of HLA profiles with early plasma viral load, CD4+ cell count..., Saah [/bib_ref] [bib_ref] Additive contribution of HLA class I alleles in the immune control of..., Leslie [/bib_ref] [bib_ref] Host Genetics of HIV Acquisition and Viral Control, Shea [/bib_ref] , it is hypothesized that vaccine-primed T cells may slow disease progression if breakthrough infection occurs [bib_ref] The quest for an AIDS vaccine: is the CD8+ T-cell approach feasible?, Mcmichael [/bib_ref]. To date, most candidate HIV vaccines have elicited HIV-specific T-cell responses [bib_ref] Vaccination with ALVAC and ADISVAX to Prevent HIV-1 Infection in Thailand, Rerks-Ngarm [/bib_ref] [bib_ref] HIV-1 vaccine-induced immunity in the test-of-concept Step Study: a case-cohort analysis, Mcelrath [/bib_ref] [bib_ref] Efficacy trial of a DNA/rAd5 HIV-1 preventive vaccine, Hammer [/bib_ref] [bib_ref] Expanded breadth of the T-cell response to mosaic human immunodeficiency virus type..., Kong [/bib_ref] [bib_ref] Induction of immunity to human immunodeficiency virus type-1 by vaccination, Mcelrath [/bib_ref] including the recombinant adenovirus-vectored vaccine tested in the Step Study, in which vaccine recipient responders targeting at least three Gag epitopes had lower viral load compared to non-responders [bib_ref] Vaccine-induced gag-specific T cells are associated with reduced viremia after HIV-1 infection, Janes [/bib_ref].
Mapping the specificities of vaccine-induced responses is necessary for a rational approach to vaccine design, since T-cell responses vary in their effect on HIV disease progression [bib_ref] Preservation of T Cell Proliferation Restricted by Protective HLA Alleles Is Critical..., Horton [/bib_ref] [bib_ref] Control of human immunodeficiency virus type 1 is associated with HLA-B*13 and..., Honeyborne [/bib_ref] [bib_ref] Compensatory mutation partially restores fitness and delays reversion of escape mutation within..., Crawford [/bib_ref] [bib_ref] Mutually exclusive T-cell receptor induction and differential susceptibility to human immunodeficiency virus..., Yu [/bib_ref]. Identifying vaccine-induced T-cell epitopes is made easier by the fact that the sequence of the vaccine immunogen is known. This greatly restricts the number of antigens that need to be tested as potential epitopes in "epitope mapping" assays. However, it is also of great importance to measure the responses of vaccine-primed T cells to circulating viral variants (i.e., response "depth"). This requires testing of many additional potential antigens [bib_ref] Peptide selection for human immunodeficiency virus type 1 CTL-based vaccine evaluation, Li [/bib_ref] and will be made more complicated as HIV vaccines are designed to provide coverage of multiple subtypes [bib_ref] Polyvalent vaccines for optimal coverage of potential T-cell epitopes in global HIV-1..., Fischer [/bib_ref]. Epitope mapping of vaccine-induced responses can be more challenging as they are typically lower in magnitude than those observed in natural infection [bib_ref] Moving to human immunodeficiency virus type 1 vaccine efficacy trials: defining T..., Russell [/bib_ref]. However, recently Borthwick et al. showed that prime-boost regimens of adenovirus-and pox virus-vectored vaccines containing conserved elements of HIV elicited broad and high magnitude T-cell responses, demonstrating that it is possible to elicit robust cellular responses with a vaccine and that they can be readily detected using ELISpot [bib_ref] Vaccine-elicited Human T Cells Recognizing Conserved Protein Regions Inhibit HIV-1, Borthwick [/bib_ref].
Several methods for epitope mapping have been devised to efficiently utilize participant samples and other lab resources. Since typically only a few peptides elicit a T-cell response in vaccine studies, "group testing" approaches have been developed where pools of peptides are evaluated in one assay; peptides in pools that don't elicit a response can be quickly eliminated en masse [bib_ref] Moving to human immunodeficiency virus type 1 vaccine efficacy trials: defining T..., Russell [/bib_ref] [bib_ref] Optimal configuration of a square array group testing algorithm, Hudgens [/bib_ref] [bib_ref] Group testing for case identification with correlated responses, Lendle [/bib_ref] [bib_ref] Optimizing peptide matrices for identifying T-cell antigens, Precopio [/bib_ref]. Pools can be designed in several hierarchical stages of decreasing size until individual peptides are identified. Pools can also be used in a matrix-based strategy such that each column and row in a matrix of peptides is tested together. This specific arrangement, such that each peptide is in exactly one row-pool and one column-pool, allows for more efficient identification of responses at the "intersection" of positive pools. Ultimately, in any strategy a positive response must be confirmed using single peptide stimulation. The sizes and design of peptide pools can be optimized based on the expected number of responses per participant and the expected covariation of response across the cohort [bib_ref] Group testing for case identification with correlated responses, Lendle [/bib_ref] [bib_ref] Optimizing peptide matrices for identifying T-cell antigens, Precopio [/bib_ref] [bib_ref] Optimized determination of T cell epitope responses, Roederer [/bib_ref]. Though the theoretical efficiencies, sensitivities and specificities have been computed for many group testing algorithms, the calculations require assumptions about the sensitivity and specificity of the assay that may not hold in practice. For example, specificity for any single antigen may decrease for pools with a large number of peptides, or pool sensitivity may increase when they contain multiple positive peptides. For this reason it is important to study the operating characteristics of group testing procedures experimentally.
In this study we evaluated two common group testing based strategies for epitope mapping using previously cryopreserved PBMC samples from participants in the Step Study. We compared these methods to each other and to a "Test-all" method in which every peptide was tested individually. Though the Test-all method would be too costly in practice, by testing individual peptides it identified the greatest number of epitopes and provided a standard by which to evaluate the two group testing strategies. We found that the group testing strategies reduced the number of ELISpot assays that were required by as much as 7.7-fold and did not result in significantly lower accuracy in estimating breadth. Assays using small pools of peptides (7-30 peptides) were highly sensitive and were able to detect responses to single positive peptides, as well as summate responses to multiple peptides. The overlapping peptides were an efficient way to map T-cell epitopes; however, some HLA class I restricted T cells may require shorter optimal-length peptides for robust stimulation. These findings provide experimental validation of group testing-based strategies for identifying vaccine-induced T-cell epitopes. The results also provide further motivation to continue optimization and standardization of epitope mapping methods in vaccine trials, which will be critical for the evaluation and comparison of novel vaccine regimens.
# Materials and methods
## Study cohort
The Step Study (Merck V520-023/HVTN 502) was a multicenter, double-blind, randomized, placebo-controlled phase IIb proof-of-concept study to evaluate the safety and efficacy of the MRKAd5 HIV Gag/Pol/Nef trivalent vaccine in 3,000 HIV-negative adults at high risk of HIV infection (ClinicalTrials.gov Identifier: NCT00095576) [bib_ref] Efficacy assessment of a cell-mediated immunity HIV-1 vaccine (the Step Study): a..., Buchbinder [/bib_ref]. Participants were randomized to vaccine or placebo and followed for incident HIV infection. Vaccine recipients were given three IM doses of 1.5×10 10 virus genomes at weeks 0, 4, and 26. All participants signed informed consent and the following list of institutional human subjects review committees approved the protocol prior to study initiation: Participants selected for the epitope mapping study cohort (n = 20) received all 3 injections, were HIV-1 uninfected throughout the follow-up period, and had peripheral blood mononuclear cell (PBMC) samples available from the Week 8 (post-enrollment) time point (Primary immunogenicity time point, 4 weeks after the second vaccination). Participants were stratified based on either a positive (n = 15) or negative (n = 5) T-cell response in a previous IFN-γ ELISpot assay at the same time point. We also included 5 placebo recipients as negative controls. All participants were previously typed for their HLA class I alleles using sequence-based typing.
## Elispot assay and epitope mapping strategies
Validated IFN-γ ELISpot assays were performed on previously cryopreserved PBMCs using previously detailed methods [bib_ref] Human adenovirus-specific T cells modulate HIV-specific T cell responses to an Ad5-vectored..., Frahm [/bib_ref]. Upon thawing, the PBMC viability in all samples was above 80%, and the majority were above 90% after an overnight rest. Briefly, 100,000 ex vivo PBMCs were stimulated overnight in triplicate for all test conditions in the presence of individual or pools of peptides at a final concentration of 1 μg/mL each. Peptides and pools were initially reconstituted in DMSO (50 mg/mL) from individually lyophilized 15-mer peptides. For each participant, 6 negative control wells (PBMCs with media alone) and 3 positive control wells (PBMCs incubated with PHA) were also tested. Responses are reported in spot-forming cells per million (SFC/M) as the mean of three replicate wells minus the mean of six replicate negative control wells.
Three epitope mapping strategies were evaluated concurrently: 1) "Test all", 2) "Mini-pool", and 3) "Matrix-pool" (details available in S1 File). ELISpot assays were performed using individual and pooled peptides, all 15 amino acids in length and overlapping by 11 amino acids; these were taken from a panel of 134 15-mers spanning the HIV-1 Gag Consensus B protein (Table A in S1 File; Gag ConB peptides, BioSyn; GenBank accession: AAS19377.1). The consensus sequence differed from the vaccine insert sequence at 7 amino acid residues. None of these differences were in peptides that elicited responses.
## Epitope identification
Participants often responded to two overlapping 15-mers, though the responses were likely specific to a single underlying epitope contained within the overlap. To determine the minimal set of epitopes that best accounted for the observed responses we applied the following criteria: if two 15-mers share a region of at least 11 residues then they can be explained by a single epitope, and that epitope is determined to be the entire region of overlap. Rarely, multiple consecutive overlapping 15-mers elicited responses. For this case epitopes were determined by iteratively applying the overlap criteria for each pair of overlapping peptides.
# Statistical analysis
All summary statistics are reported either with the standard error or the 95% confidence interval as noted by either a ± or "[LL, UL]", respectively. Standard errors, confidence intervals and P-values were computed using bootstrap methods with participants as the independent unit of analysis, providing valid intervals and inference that account for within-vaccinee correlations of multiple assay readouts. Correlations were computed using Spearman's rank correlation ρ. Wilcoxon rank sum tests were used for the comparison of two independent groups of response magnitudes, while signed-rank tests were used for comparing two groups of paired samples. All data available in Supporting Information (S2 File).
# Results
## Vaccine-primed t cells recognize hiv-1 gag in master peptide pools
We selected [bib_ref] Response definition criteria for ELISPOT assays revisited, Moodie [/bib_ref] Step study participants [bib_ref] Efficacy assessment of a cell-mediated immunity HIV-1 vaccine (the Step Study): a..., Buchbinder [/bib_ref] who received all 3 injections of the Ad5-vectored Gag/Pol/Nef HIV-1 candidate vaccine; 15 had a positive and 5 had a negative response in a prior IFN-γ ELISpot assay evaluating cryopreserved PBMC and using a pool of overlapping peptides spanning the Gag protein. Five placebo recipient samples served as negative controls. T-cell responses four weeks after the second vaccination, were measured to both individual and pooled Gag 15-mers. Assay positivity was determined using MIMOSA [bib_ref] Mixture Models for Single-Cell Assays with Application to Vaccine Studies, Finak [/bib_ref] , a statistical test that controls the false-discovery rate (FDR) at 0.1% using assay replicates and negative controls.
The largest peptide pool, the Gag "master" pool, contained 134 15-mers spanning the entire Gag protein. As anticipated, the master pool elicited Gag-specific T-cell responses in all 15 vaccine responders, with a mean magnitude of 571 ± 205 SFC/M. The magnitudes of the master pool responses correlated strongly with the summed magnitude of all individual positive peptides within the pool (ρ = 0.92 [0.71, 0.99]), though the pool magnitude was consistently lower than that of the summed peptides (mean difference 694 ± 406 SFC/M, p = 0.018). No positive responses to the master pool were detected among placebo recipients or the five vaccine nonresponders (n = 10, mean 17 ± 5 SFC/M). The master pool responses were used to differentiate responders from non-responders in all other analyses.
## T cells of vaccine recipients respond to individual 15-mer peptides
We assessed the responses of each participant to each of the 134 15-mers in the Gag master pool in S1 File), and detected 37 positive responses (mean magnitude 513 ± 209 SFC/ M; [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref]. For positive responses the median coefficient of variation across replicates was 0.19 [0.13, 0.25]. This is equivalent to an 82 SFC/M standard deviation for a 500 SFC/M response. Among the responders, the mean number of responses was 2.5 ± 0.5 [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref]. There were no positive responses to individual peptides among the placebo recipients or the vaccine nonresponders.
Depending on sample availability, responders were tested using individual 8, 9, 10, 11, and 12 amino acid "optimal" Gag peptides previously found to elicit T-cell responses in HIV infected patients in S1 File). These were individually selected based on the restricting HLA allele and the HLA alleles of each responder. In total we tested 334 optimal peptides and detected 27 positive responses [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref]. The magnitudes of responses to optimal peptides (293 ± 47 SFC/million) was not significantly different from those to 15-mers (p = 0.50).
To evaluate mapping strategies it was important to validate each response with at least one related response. The response was related either as: 1) an overlapping 15-mer sharing at least 8 amino acids, 2) an overlapping optimal peptide sharing at least 8 amino acids, or 3) a pool containing the peptide and no other positive peptides. Based on these criteria all 37 responses were validated. All but two responses were validated by overlapping peptides, suggesting that the epitopes were HLA class I restricted. The remaining two responses contained known HLA class I restricted epitopes. This is consistent with previous observations that while the Step vaccine primed both CD4+ and CD8+ T cells, CD8+ T cells were the primary producers of IFN-γ [bib_ref] HIV-1 vaccine-induced immunity in the test-of-concept Step Study: a case-cohort analysis, Mcelrath [/bib_ref]. Given that 2010 total 15-mers were tested and all responses were validated, the false discovery rate is less than 0.05%, suggesting that we are using an appropriate (if slightly conservative) false-discovery rate cutoff for assay positivity.
Overall there was strong correlation between the magnitudes of responses to the positive 15-mers and those of the peptides or pools that validated them, including optimal peptides (Spearman's ρ = 0.76 [0.49, 0.90]; n = 25), overlapping 15-mers (ρ = 0.75 [0.24, 0.98]; n = 32) and pools (ρ = 0.92 [0.72, 0.98]; n = 32) [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref]. Additionally, the differences in magnitude between the responses to positive 15-mers and their validating pools or overlapping peptides were not significantly different from zero, suggesting that they were adequate proxies for the 15-mer response [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref] Optimal length peptides were also used to further define the epitopes. Negative peptides containing the sequence of a positive optimal peptide were labeled as false-negative. Breadth was the minimal number of epitopes that explained the responses of each participant (B). Breadth was occasionally greater than the number of 15-mer responses when responses were detected to optimal peptides, but not the parent 15-mer peptide. The magnitudes of responses to 15-mer peptides and those that confirmed their positivity were plotted to assess their similarity and their correlation (C; dashes illustrate line of equality). Magnitude of response to a 15-mer peptide differed depending on the relative position of the epitope containing optimal peptide (D). Responses to optimal peptides identify false negative 15-mer responses
We identified 17 15-mers which tested negative despite containing the sequence of a positive optimal peptide (0.8% of all negative 15-mer tests) [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref]. Many of these negative 15-mers also overlapped positive 15-mers (n = 9) or were included in positive peptide pools that did not contain any other positive 15-mers (n = 5), thus meeting at least one of the other criteria that was used for validation of positive responses. Furthermore, their mean magnitude was 24 ] SFC/M, which is greater than negative assays overall (6.4 ] SFC/M). However, four of these putatively false negative 15-mers (two overlapping pairs each sharing an optimal) could not be confirmed by any other criteria, yet the response to the optimal peptide was well above background (78 and 537 SFC/M). In the subsequent evaluation of epitope mapping strategies, these responses were counted as two epitopes that were not detected by any of the strategies.
One possible explanation for a negative 15-mer that contains an optimal epitope is that the response depends on the location of the epitope within the 15-mer. To address this possibility, we examined the 22 optimal peptides that elicited positive responses and that were contained by two 15-mer parent peptides. Responses were significantly higher when the epitope was located toward the N-terminus compared to the C-terminus, with a mean difference of 77.5 SFC/M [17.6, 154.7, p = 0.014] SFC/M [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref].
## Underlying epitopes deduced from overlap of positive 15-mer and optimal peptides
We hypothesized that responses to overlapping peptides shared a single underlying epitope. Based on a simple set of criteria requiring at least 8 overlapping residues (see Materials and Methods for details) we determined that the 37 responses to 15-mers and the 27 responses to optimal peptides could be explained by 27 underlying epitopes. This corresponded to a breadth of 1.8 [1.4, 2.3] Gag epitopes per responder [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref]. With the exception of two epitopes identified only by an optimal peptide, all of the epitopes could be confirmed by positive responses to at least two overlapping peptides. These 27 epitopes were used to evaluate the epitope mapping strategies.
## Small pools of peptides elicit robust t-cell responses
We tested for T-cell responses to pools of 15-mers that varied in their size and makeup. Each pool type was a stage in one of the epitope mapping strategies being evaluated. The positivity of responses to pools was determined using the MIMOSA method, controlling the false discovery rate at 0.1% within each pool type (i.e. master, sub, matrix and mini pools). There were no positive responses to any peptide pools among the placebo recipients and vaccine non-responders. Among the responders, 125 of 855 pools elicited a detectable positive response [fig_ref] Fig 2: T-cell responses to small pools of 15-mer peptides [/fig_ref]. Of the positive pools, 75 ± 9% contained at least one 15-mer that elicited a positive response when tested individually, and as such were referred to as "true positive" pools [fig_ref] Fig 2: T-cell responses to small pools of 15-mer peptides [/fig_ref]. Magnitudes of false positive pools correlate weakly with those of the negative 15-mers they contain There were 31 positive pools whose peptides did not elicit responses when tested individually [fig_ref] Fig 2: T-cell responses to small pools of 15-mer peptides [/fig_ref]. This corresponded to a false positive rate of 31/125 = 25 ± 9%. One possible explanation is that these putative false positive pools are due to decreasing assay specificity when pools of peptides are used for stimulation. Alternatively, ELISpot assays with pools of peptides may be more sensitive than those with single peptides, as they could summate multiple low magnitude responses. Several of these pools contained a 15-mer that assayed negative despite containing the sequence of a positive optimal peptide (i.e., false negative 15-mer). This may indicate that there were T-cell responses to an epitope that was not detected by the individual 15-mer, but was detected by the pool of 15-mers and by the optimal peptide. We False negative pools are the source of potential errors in epitope mapping strategies
The negative responses to pools of peptides were classified into two groups based on whether or not they contained a 15-mer that elicited a positive response [fig_ref] Fig 2: T-cell responses to small pools of 15-mer peptides [/fig_ref]. Most negative pools did not contain a positive 15-mer (99% of all negative pools). However, 7 negative matrix-pools (1.3%), 3 negative mini-pools (1.6%) and zero negative sub-pools (0%) contained at least one positive 15-mer. These "false negative" pools were the source of errors in epitope mapping strategies. There are two possible explanations for the disagreement between the positive 15-mer and the negative pool containing that 15-mer: 1) the 15-mer elicits a response that is not detectable in the pooled peptide assay (e.g. peptide competition or antagonism in the pool); or 2) despite testing positive, neither the 15-mer nor the pool elicit true positive T-cell responses. Notably, the mean magnitude of these pools is greater than that of the other negative pools (36 [bib_ref] Response definition criteria for ELISPOT assays revisited, Moodie [/bib_ref] [bib_ref] Vaccine-elicited Human T Cells Recognizing Conserved Protein Regions Inhibit HIV-1, Borthwick [/bib_ref]
## Evaluation of the epitope mapping strategies
With the results of the single peptide and pooled peptide ELISpot assays, we evaluated the efficiency and sensitivity of three T-cell epitope mapping strategies: 1) Test-all, 2) Mini-pool, and 3) Matrix-pool (details available in S1 File). The Test-all strategy is the least complex: for each participant, test for a response to each of the 134 Gag 15-mers individually. Despite the false negative 15-mers described above, the Test-all strategy successfully identified 23 of the 27 validated epitopes [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. However, the Test-all strategy required the greatest number of assays-134 single peptide assays per donor [fig_ref] Table 1: Number of assays required to identify epitopes using each of the strategies... [/fig_ref].
The Mini-pool strategy is a two-staged, pooling approach in which 15-mer peptides are not tested individually if they fail to elicit a response as part of a pool of peptides. Since the minipool strategy depends on testing individual 15-mers in the final step, in practice it cannot identify any epitopes that are not also identified by the Test-all strategy [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. However, in our evaluation of the strategy we included as successes those epitopes that were identified using optimal peptides and whose sequence was contained by a 15-mer within a positive mini pool, even if the parent 15-mer did not elicit a response [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. By this measure, 6 epitopes went undetected by the mini-pool strategy, 2 of which were undetected by all strategies (including Test-all). The Mini-pool strategy required 29.6 ± 2 tests per participant, substantially lower than that for the Test-all approach. By definition, all 15-mers were identified by the Test-all strategy. Epitopes detected using an optimal peptide were not necessarily identified by the Test-all strategy, but may have been detected by a pool strategy if all pools containing the epitope were positive. Potential benefits were explored of using computational HLA binding predictors to identify the epitope-containing 15-mer within a positive mini-pool (C, n = 20 pools) and to identify the optimal peptide within a positive 15-mer (D, n = 25 peptides). Ranks of the 15-mers in the mini-pool or the optimal peptides within the 15-mer were based on predicted HLA binding with each participant's alleles. Observed average (line) and 95% confidence interval (dashed lines) are illustrated relative to a distribution of averages computed using random rankings. The matrix-pool strategy is a three-staged, pooling approach which seeks to further reduce the number of necessary tests with a "matrix" of pools that identify the positive 15-mer in the final stage. This strategy successfully identified 21 of the 27 epitopes. Unidentified epitopes were the direct result of false-negative matrix pools-not false-negative sub-pools-as no falsenegative sub-pools were observed. One epitope identified by an optimal peptide was also identified by intersecting matrix pools, but not by its parent 15-mer or sub-pool. Due to the efficient design of the matrix pools the strategy required significantly fewer tests than the minipool strategy (17.4 ± 3 tests per participant, n = 15 responders; p < 0.001).
Although the number of assays for each strategy is a good proxy for efficiency in general, the number of cryovials of PBMC required by each strategy is also relevant. In general, a single cryovial of PBMC can be used for a 96-well plate of ELISpot assays, though this can depend on cell recovery. A single plate allows for~30 assays in triplicate in addition to negative and positive controls. Based on these numbers the Test-all strategy requires 5 cryovials per participant, independent of the number of responses. The Mini-pool strategy requires 1 cryovial for the pools in the first stage and 1 cryovial for the peptides in the second stage, per person. Theoretically more cryovials would be required if the response breadth was greater, but in this study all participants would have required two. The Matrix-pool strategy requires 1 cryovial for the subpools in the first stage, 1 or more cryovials for the matrix-pools in the second stage and 1 cryovial for the "intersection" peptides in the final stage. Therefore, based on this study, the Matrixpool strategy would require at least one additional cryovial per participant and possibly two, compared to the Mini-pool strategy, depending on the number of positive sub-pools. However, since there are only three Gag subpools in the Matrix-pool strategy, this difference could be mitigated in practice by testing subpools for several HIV proteins on a single plate.
HLA binding predictors could be used to further increase the efficiency of epitope mapping strategies HLA binding predictors are computational tools that estimate the binding affinity of a short peptide with a specific HLA allele based on a large database of experimental data [bib_ref] NetMHCpan, a method for quantitative predictions of peptide binding to any HLA-A..., Nielsen [/bib_ref]. Though HLA binding is not the only factor that determines whether a vaccine peptide is targeted by T cells, predicted binding affinity has been used extensively to help identify vaccine-induced Tcell epitopes [bib_ref] A consensus epitope prediction approach identifies the breadth of murine T(CD8+)-cell responses..., Moutaftsi [/bib_ref]. Given the class I HLA genotypes of the Step participants, we hypothesized that HLA binding prediction could be used to reduce the number of tests needed to identify the positive 15-mers in positive mini-pools. To assess the usefulness of HLA binding predictors for this purpose, we determined the binding affinity of each 15-mer, based on the minimum IC 50 of all 8, 9 and 10-mers it contained with each of the participant's four HLA-A and -B alleles (HLA-C and class II HLA alleles were excluded due to poor predictive performance). For each positive mini-pool, we determined the rank of the positive 15-mer in the pool. The ranks [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. This demonstrated that within this cohort, a mini-pool strategy in which only the top 4 ranked peptides were tested within each mini-pool could have successfully identified all positive 15-mers and would have saved 144 negative 15-mer assays (9.6 per participant). As a comparison, we repeated the procedure 1,000 times assigning ranks randomly within each mini-pool. Under these conditions the average rankings of all the peptides formed a distribution centered at 5.5 [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. None of the random iterations yielded a mean ranking that was as low as we observed, which provides a demonstration of the information added by the HLA binding predictions. HLA binding predictors could also be used to identify an optimal epitope within a positive 15-mer [bib_ref] HLArestrictor-a tool for patient-specific predictions of HLA restriction elements and optimal epitopes..., Larsen [/bib_ref]. To test this idea we used the 25 positive 15-mers, which contained the sequence of an optimal peptide that also tested positive, to assess the predictive performance of HLA binding predictors at identifying the optimal peptide within a 15-mer. For each 15-mer we ranked the optimal peptide and all 8, 9 and 10-mer peptides (21 k-mers) contained by the 15-mer, based on its minimum predicted IC 50 across the responder's four HLA-A and -B alleles. The rank of the optimal peptide among all k-mers ranged from 1 st to 5 th with an average ranking of 2.5 [2.0, 3.0] [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. As a reference, to show the information added by using the HLA binding predictors, we repeated this procedure 1,000 times using random rankings of the peptides within each 15-mer [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. This demonstrated that HLA binding predictors could be used to reliably identify optimal peptides restricted by HLA-A and HLA-B when 15-mer peptides have been used for epitope mapping. As the performance of HLA-Cw and class II HLA binding predictors improves, and with more rigorous validation of their application, these techniques could be used more broadly to increase the efficiency and accuracy of epitope mapping strategies in vaccine trials.
# Discussion
We have comprehensively evaluated two epitope mapping strategies that are commonly employed for the identification of vaccine-induced T-cell epitopes. Compared to a approach of testing all the overlapping 15-mer peptides in Gag, the matrix-pool and mini-pool strategies each identified all but three of the detectable T-cell epitopes, while benefitting from the efficiency gains of a group testing design. The estimated mean breadth for both of the group testing strategies was 1.4 epitopes per participant compared to a breadth of 1.5 estimated using the Test-all strategy. Though the Test-all strategy is simple in design, as it lacks peptide pools and conditional testing, the efficiency gains of the group testing strategies (4.5-fold and 7.7-fold for the mini-pool and matrix-pool strategies, respectively) make it possible to use epitope mapping as a variable in large HIV pathogenesis studies [bib_ref] Simultaneous assessment of cytotoxic T lymphocyte responses against multiple viral infections by..., Bihl [/bib_ref] [bib_ref] Consistent cytotoxic-T-lymphocyte targeting of immunodominant regions in human immunodeficiency virus across multiple..., Frahm [/bib_ref] [bib_ref] CD8+ T-cell responses to different HIV proteins have discordant associations with viral..., Kiepiela [/bib_ref] , and as an immunogenicity endpoint in vaccine efficacy trials [bib_ref] HIV-1 vaccine-induced immunity in the test-of-concept Step Study: a case-cohort analysis, Mcelrath [/bib_ref] [bib_ref] Mapping HIV-1 vaccine induced T-cell responses: bias towards less-conserved regions and potential..., Li [/bib_ref] [bib_ref] The Thai phase III trial (RV144) vaccine regimen induces T cell responses..., De Souza [/bib_ref] for which sample volume and lab resources may be limiting. The efficiency advantage of both pooling strategies over the Test-all strategy was also evident in the reduced usage of PBMC cryovials, although the difference between the two pooling strategies might not be significant in the context of mapping T-cell responses in multiple proteins. It might be possible to forego the sub-pool stage altogether, using a two-stage matrix-pool approach that would further reduce the number of assays and cryovial requirement. However, reduced assays and sample usage at the sub-pool level would increase assays and sample usage at the matrix pool level, as negative sub-pools help to screen-out peptides that do not contain epitopes. In theory, eliminating the sub-pool stage may also increase the sensitivity of the matrix-pool strategy, however in practice there were no false-negative sub-pools and therefore no responses went undetected due to the sub-pool level of assays.
The increased efficiency of group testing strategies may also be particularly important for identifying epitopes in the pathogens that cause TB and malaria as their proteomes are much larger than HIV and there is evidence that a T-cell response may be necessary for an effective vaccine [bib_ref] Randomized, double-blind, phase 2a trial of falciparum malaria vaccines RTS,S/AS01B and RTS,S/AS02A..., Kester [/bib_ref] [bib_ref] Immunological mechanisms underlying protection mediated by RTS,S: a review of the available..., Moorthy [/bib_ref] [bib_ref] Ag85B-ESAT-6 adjuvanted with IC31 1 promotes strong and long-lived Mycobacterium tuberculosis specific..., Van Dissel [/bib_ref] [bib_ref] The Immunodominant T-Cell Epitopes of the Mycolyl-Transferases of the Antigen 85 Complex..., Huygen [/bib_ref]. An epitope mapping strategy could potentially be further optimized either for greater efficiency or higher sensitivity, for example, by decreasing or increasing, respectively, the number of technical replicates of each assay. The efficiency and sensitivity we observe is based on three replicates for all pooled and single peptide assays. Others have demonstrated increased efficiency in sample usage by using a cultured ELISpot assay [bib_ref] Vaccine-elicited Human T Cells Recognizing Conserved Protein Regions Inhibit HIV-1, Borthwick [/bib_ref] ; though the expansion of antigen-specific T cells may increase an assay's sensitivity for detecting a response, the response may not mimic the phenotype or functionality of an ex vivo response.
One complexity in this evaluation was that the T-cell epitopes were not known with absolute certainty; we were limited by the individual 15-mer assays. This issue was mitigated by validating the responses using a positive overlapping peptide or positive parent pool. With this approach we were able to validate the true positivity of all the 15-mer responses. However, there were several "false negative" 15-mers that did not elicit responses despite containing the sequence of an optimal peptide that did. It is possible there were additional false negative 15-mers that we did not identify because we did not exhaustively test all optimal peptides for all negative 15-mers.
A potential pitfall of any group testing procedure is that the pooled assay may be less sensitive to any single positive signal within the pool. In the context of natural HIV-1 infection Addo et al. showed that matrices of small pools of peptides (<12) had sensitivities relative to individual peptide testing that ranged from 93 to 97%. In this study we identified 10 instances in which a pool did not elicit a response, despite containing at least one peptide that did individually. The impact of these false negative pools was that the matrix-pool and minipool strategies each failed to identify 6 T-cell epitopes. Relative to the test-all strategy, the sensitivity of the pooling strategies was 91% (21 of 23 epitopes; [fig_ref] Fig 4: Evaluation of epitope mapping strategies [/fig_ref]. Responses to the false negative pools tended to be higher in magnitude than to negative pools overall, which suggested that there may have been low-level responses that were below the level of detection for the assay. Smaller pool sizes may help to minimize the problem by reducing the likelihood for antigen competition for MHC binding. However, larger pools are also the source of efficiency gains, since negative pools eliminate the need for further testing of the peptides contained by the pool.
Our findings suggest that a benefit of pooled assays may be increased sensitivity in detecting low level responses. We hypothesize that an ELISpot assay with a pool of 15-mers will be more sensitive than an assay of any one of the peptides tested individually, if there are multiple 15-mers that stimulate different subsets of T cells. This is supported by the strong correlation between the magnitudes of pools containing more than one positive 15-mer with the summed magnitude of the positive peptides within the pool (ρ = 0.97; [fig_ref] Fig 3: Correlation of responses to peptide pools and the peptides they contain [/fig_ref] , which shows that pools summate multiple responses and suggests that the false positive pools may actually be detecting low-level responses. Of the 31 putatively false positive pools, 7 included a 15-mer that contained the sequence of a positive optimal peptide, suggesting that there was indeed a T-cell response detected by the pooled assay [fig_ref] Fig 2: T-cell responses to small pools of 15-mer peptides [/fig_ref]. Response summation is enabled by the design of pools that contain overlapping peptides, which also reduces the number of total tests required compared to other pool designs [bib_ref] Group testing for case identification with correlated responses, Lendle [/bib_ref] [bib_ref] Optimizing peptide matrices for identifying T-cell antigens, Precopio [/bib_ref].
Our experiments support previous observations that 15-mers are not the ideal antigen for class I restricted T cells. Of the 22 positive optimal peptides that were contained by two parent 15-mer peptides, 15 had at least one parent that assayed negative. The position of an optimal peptide within a 15-mer has been shown to impact the magnitude of the response [bib_ref] Impact of intrapeptide epitope location on CD8 T cell recognition: implications for..., Draenert [/bib_ref] ; out data support this as well, although in contrast we find that the 15-mer magnitude is greater when the optimal is towards the N-terminus. In addition, when the 15-mer parent peptide did elicit a response, it was not significantly different from the response to the optimal peptide and the two were highly correlated [fig_ref] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides [/fig_ref]. Together this suggests that the impact of epitope location within a parent peptide may be epitope dependent. We suggest that it may be beneficial to increase the number of replicates for the assays of individual 15-mers, since this would help to increase the overall sensitivity to low level responses. The 19% coefficient of variation we observed is comparable to the 20% that was observed in another study of a validated ELISpot assay [bib_ref] A dual color ELI-SPOT method for the simultaneous detection of IL-2 and..., Boulet [/bib_ref].
Though we have discussed strategies to detect T-cell epitopes near the limit of detection, some may question the protective or therapeutic benefits of such low-level responses. In the future it may be important to optimize for the efficient detection of only the robust immunodominant responses or for the detection of polyfunctional responses using dual-color ELISpot assays [bib_ref] A dual color ELI-SPOT method for the simultaneous detection of IL-2 and..., Boulet [/bib_ref] or gene expression markers. Such optimization could allow for fewer tests. However, it may also be important to increase the sensitivity of the ELISpot assay, if only to reduce the minimum PBMC and peptide requirements. Until the sample requirements and sensitivities of other stimulated assays like ICS can better those of the ELISpot assay, it will remain the standard for mapping and quantifying vaccine-induced epitope specificities.
## Supporting information
[fig] Fig 1: Vaccine primed IFN-γ ELISpot responses to HIV Gag peptides. Peptides derived from a consensus clade B Gag sequence were used individually to detect T-cell responses in participants (n = 15 vaccine responders) of the Step HIV vaccine trial. Responses were quantified in units of spot forming cells per million (SFC/M) using an IFN-γ ELISpot assay (A; one dot indicates the mean of a triplicate assay). [/fig]
[fig] Fig 2: T-cell responses to small pools of 15-mer peptides. ELISpot responses to peptide pools were categorized based first on their positivity (A, positive; B, negative; C, response magnitudes, one dot indicates the mean of a triplicate assay) and subsequently on whether or not they contained a positive 15-mer or a 15-mer containing the sequence of a positive optimal peptide. doi:10.1371/journal.pone.0147812.g002 [/fig]
[fig] Fig 3: Correlation of responses to peptide pools and the peptides they contain. Rank-based correlations were assessed between the magnitudes of T-cell responses to positive peptide pools that contain a single positive peptide and the individual response to the positive peptide (A, one dot per pool, dashes indicate line of unity) and positive pools containing >1 positive peptide and the sum of the individual responses to the positive peptides they contain (B). Positive pools containing no positive 15-mer and >1 false negative 15-mer were plotted against the sum of the negative peptides they contained (C). 95% confidence intervals and p-value were computed using a participant-based bootstrap (n = 15 responders). doi:10.1371/journal.pone.0147812.g003 [/fig]
[fig] Fig 4: Evaluation of epitope mapping strategies. Epitope mapping strategies were assessed based on the number of positive 15-mers (A) and the number of epitopes (B) that were identified. [/fig]
[fig] S1: File. Supporting methods, tables and figures. (DOCX) S2 File. Tables of comma-separated values (CSV) containing ELISpot magnitudes and response calls. (CSV) [/fig]
[table] Table 1: Number of assays required to identify epitopes using each of the strategies † . [/table]
|
A Year of Living Dangerously: Challenges and Recommendations for Safely Performing Ophthalmic Surgery During the COVID-19 Pandemic, from Start to Finish
The COVID-19 pandemic has forced all nations to take an active role in infection control incorporating recommendations and measures to control viral dissemination. The epidemiological impact is very diverse and dynamic, even within the same region. Scientific knowledge regarding SARS-CoV-2 continues to improve every day with protocols needing to be updated and adjusted on a regular basis. Ophthalmology is a medical specialty identified to be at high risk for several reasons: it has very close doctor-patient contact, the virus has been detected in tears, and the ocular surface serves as a gateway to developing the infection. We have reviewed the current information on SARS-CoV-2 in the ophthalmologic field and provide up-to-date recommendations to help create protocols that can adapt to the dynamic situation of ophthalmologic institutions, patient cases, economic situations and access to diagnostic tests. This paper outlines the main recommendations regarding the initial consultation and outpatient clinics, measures to apply in the operating room (OR), and suggestions for post-surgical controls. Triage, according to the patient's conditions and eye pathology, reduction of the time the patient is at the institution, social distancing, correct use of personal protective equipment (PPE), barrier methods, hygiene, as well as other recommendations mentioned in this document, will allow physicians to take care of the visual health of the patients while reducing the impact of the COVID-19 pandemic.
# Introduction
The world is about to reach 1 year dealing with the COVID-19 pandemic, offering a mixed picture, with countries that are dealing with the second wave of this disease, some are just getting out of the first wave, while a minor sector is still waiting for the hit after travelling restrictions quit.
The future in the medium term is promising with hundreds of vaccines under research, with a few already in phase 3 accounting for over 90% of efficacy. In the short term, this pandemic is still forcing all nations to take an active role in infection control incorporating recommendations and measures to limit economic, social and health consequences related to the virus.
Some patients with ophthalmic pathologies (urgent and chronic) still postpone consultations, occasionally due to risk factors for COVID-19, but most frequently because of lack of information and fear of being exposed to the virus.
Ophthalmologists, on the other side, evidence the consequences of the absence of appropriate follow-ups and consultations as well as the updated descriptions of the impact of SARS-CoV-2 and the new lifestyle on visual health.
In this review, we examine the latest recommendations and main challenges in the ophthalmic field to reduce COVID-19's impact learnt during the first year of the pandemic, to be applied at the present time or to future waves. Recommendations might be useful for other viral pandemics in the future.
# Methods
A comprehensive review of the literature was carried out through PubMed beginning in April 2020. This search was conducted on regular basis due to a large number of articles being published. The amount of new information published in the short window shows the importance of this topic to health professionals, health systems, and government decision-makers.
The following PubMed terms in diverse combinations were used to search for related articles in this review: SARS-CoV-2, COVID-19, transmission, risk factors, eye, ophthalmology, ophthalmologist, surgery, operating room, conjunctivitis, personal protective equipment (PPE). A subsequent search was conducted when the terms were relevant. Publications cited in the articles selected by the search were also retrieved. Furthermore, special links related to COVID-19 on the National Center for Biotechnology Information (NCBI) site were checked, as well as public health information from the Centers for Disease Control and Prevention (CDC) and the National Institute of Health (NIH). Also, the World Health Organization (WHO) site was consulted. Previous to the publication's date there were no restrictions applied.
## Covid-19 pandemic
Coronavirus is a family of enveloped positive-sense, single-stranded RNA viruses associated with a range of digestive and respiratory diseases: common cold, bronchitis, bronchiolitis and mild to severe pneumonia. Two members of this family have caused epidemic disease in the past: Middle East Respiratory Syndrome-related Coronavirus (MERS-CoV) and Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV).SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2) is the name that the World Health Organization (WHO) assigned to a novel coronavirus responsible for the outbreak of COVID-19 (Coronavirus Disease 2019) in Wuhan city, province of Hubei, China on 31 December 2019.One month later, the Emergency Committee of the WHO declared the disease outbreak a Public Health Emergency of International Concern.COVID-19 has a global impact due to its rapid viral spread aided by an international travel sector, which has led to even more rapid dissemination of the disease and to the health system's collapse, even of those that were thought to be robust.Up to now, counting with treatments without demonstrated efficacy and some promising vaccines currently in phase 3 of clinical trials to limit the COVID-19 pandemic, a preventative strategy focused on proper epidemiologic measures and maximizing infection control are warranted. [bib_ref] COVID-19 and ophthalmology: a new chapter in an old story, Reviglio [/bib_ref] In order to develop a program to control and prevent the spread of the disease in healthcare facilities, it is important to understand the various ways in which pathogens are transmitted: through direct contact, fomites, oral (ingestion), vectorborne, and droplets (respiratory droplets: >5 to 10 um, and airborne or droplets nuclei: <5 um).Airborne droplets may remain in the air for long periods of time and be transmitted to others over distances greater than one meter.The main vehicle for the transmission of SARS-CoV-2 is respiratory droplets and direct contact. [bib_ref] Community transmission of severe acute respiratory syndrome Coronavirus 2, Liu [/bib_ref] The mechanism of SARS-CoV-2 is presumed to infect the human cells by spike glycoprotein binding to its cellular receptor, angiotensinconverting enzyme 2 (ACE2). Thus, the main virus entry occurs through nose, mouth and eyes cells that have the ACE2. [bib_ref] The ocular surface and the coronavirus disease 2019: does a dual 'ocular..., Napoli [/bib_ref] [bib_ref] Cell entry mechanisms of SARS-CoV-2, Shang [/bib_ref] [bib_ref] Ocular conjunctival inoculation of SARS-CoV-2 can cause mild COVID-19 in rhesus macaques, Deng [/bib_ref] Even though there is evidence of viral shedding in the excreta of COVID-19 patients, currently, there are no reports of faecal−oral transmission. [bib_ref] The presence of SARS-CoV-2 RNA in the feces of COVID-19 patients, Chen [/bib_ref] Similarly, there is no evidence that SARS-CoV-2 undergoes intrauterine or transplacental transmission, but some recent case reports have suggested this might be possible. [bib_ref] An analysis of 38 pregnant women with COVID-19, their newborn infants, and..., Schwartz [/bib_ref] Cross-sectional and experimental studies indicate that the virus can remain viable and infectious for hours or even days depending on the material where the inoculum is placed. [bib_ref] Detection of air and surface contamination by SARS-CoV-2 in hospital rooms of..., Chia [/bib_ref] [bib_ref] COVID-19 vulnerability: the potential impact of genetic susceptibility and airborne transmission, Pollitt [/bib_ref] In the context of COVID-19, to protect patients and health care providers, infection control has to focus on direct contact, and the airborne droplet transmission in ophthalmology may be possible in specific circumstances and settings in which aerosolizing procedures are performed.
## Sars-cov-2 and eye
The human ocular surface epithelium provides an additional entry portal for SARS-CoV-2. 262 some reports that indicate that the eyes were the suspected gateway for the virus. [bib_ref] Ocular conjunctival inoculation of SARS-CoV-2 can cause mild COVID-19 in rhesus macaques, Deng [/bib_ref] [bib_ref] Coronavirus disease 2019 presenting as conjunctivitis, Ying [/bib_ref] Several studies have documented the presence of SARS-CoV-2 genetic material on the ocular surface which was detected in conjunctival swabs and tears after performing a reverse-transcription polymerase chain reaction (RT-PCR) assay for the detection of viral RNA in COVID-19 positive patients. A meta-analysis showed that this situation is still controversial and not very frequent, with estimated positive swabs in 1.95% of the cases (95% C.I. 0.74 to 4.11). [bib_ref] The evidence of SARS-CoV-2 infection on ocular surface, Zhang [/bib_ref] [bib_ref] Presence of viral RNA of SARS-CoV-2 in conjunctival swab specimens of COVID-19..., Kumar [/bib_ref] [bib_ref] A comprehensive Chinese experience against SARS-CoV-2 in ophthalmology, Yu [/bib_ref] At the moment of analysing the reason for this low percentage, one has to consider the effect of broad-spectrum antivirals in topical ophthalmic medications, a fact not reported in most publications. [bib_ref] Panel of broad-spectrum antivirals in topical ophthalmic medications from the drug repurposing..., Napoli [/bib_ref] Viral RNA of SARS-CoV-2 was also detected in the retina of COVID-19 patients.Also, in a cross-sectional study, retinal and vasculature alterations were detected correlating with COVID-19 disease severity in 54 patients.Recent evidence demonstrated the presence of SARS-CoV-2 in the vitreous and cornea, highlighting the criticality of donor screening guidelines and the plausibility of human eye tissue or eye fluids to be a source of infection. [bib_ref] There is no intraocular affection on a SARS-CoV-2 -Infected ocular surface, Lauermann [/bib_ref] The ocular manifestations of COVID-19 include mainly signs and symptoms compatible with follicular conjunctivitis. The incidence of conjunctivitis is estimated around 1% to 3% and is related in a meta-analysis to more severe forms of COVID-19. [bib_ref] The evidence of SARS-CoV-2 infection on ocular surface, Zhang [/bib_ref] [bib_ref] Retinal findings in patients with COVID-19, Marinho [/bib_ref] Few articles have reported keratoconjunctivitis as the first symptom of COVID-19. [bib_ref] Keratoconjunctivitis as the initial medical presentation of the novel coronavirus disease 2019..., Cheema [/bib_ref] [bib_ref] Ocular manifestations and tear or conjunctival Swab PCR positivity for 2019-nCoV in..., Sarma [/bib_ref] Findings on the retina of COVID-19 positive patients include cotton wool spots and microhemorrhage at fundus examination. OCT findings were hyperreflective lesions at the level of ganglion cell and inner plexiform layers more prominently at the papillomacular bundle in both eyes. Results of OCT-angiography and ganglion cells complex analysis appeared normal. [bib_ref] Swabs collected by patients or health care workers for SARS-CoV-2 testing, Tu [/bib_ref] Some authors reported an increase in dry eye syndrome symptoms along with this pandemic. This might be due to a double mechanism: the mass use of face mask (incorrect fitting of the mask could disperse air around the eyes) and the increase in exposure to computer screens and cell phones (prolonged blinking intervals).Another finding, suspected as a possible COVID-19 repercussion, is the Sanitizer aerosol-driven ocular surface disease (SADOSD). 32
## Sars-cov-2 and eye care
At the beginning of the COVID-19 pandemic, the initial recommendation the population heard was: "stay at home". This was strongly remarked for elderly people and especially with comorbidities. Scientific associations communicate to postpone non-essential outpatient visits and elective surgical procedures. Soon expert panels defined what should be considered urgent to take care of, and once protocols were applied to control viral dissemination, eye care institutions reopened making a big effort to work in the new normal.
There was an evident delay in visits to the eye doctor, which still persists, turning acute and chronic pathologies to be misattended. This is frequent in patients with risk factors, due to fear and lack of correct information. At the top rank of eye, pathologies are diabetic retinopathy, glaucoma and maculopathies. As we all know, they are sightthreatening diseases that may have a devastating impact. Patients and ophthalmologists are witnessing this trend that needs to revert; otherwise, this process could lead to a dramatic increase in disability, with an increase in social and government costs. [bib_ref] Early impact of COVID-19 outbreak on eye care: insights from EUROCOVCAT group, Toro [/bib_ref] [bib_ref] Safety recommendations and medical liability in ocular surgery during the COVID-19 pandemic:..., Napoli [/bib_ref]
## Outpatient clinic and general presurgical consideration
## Consideration in the outpatient clinic
Measures to lower the risk of exposure to SARS-CoV-2 should be bidirectional, they aim to take care of the patients and the staff.To plan a first contact with the patient in a remote way, promoting telephone consultation and/or telemedicine care, might be the most logical way to start. It is favourable to 1) give instructions about the visit,make a triage, 3) advance with interrogation, and to reduce the time of the visit.
A triage aims to detect possible cases of COVID-19 and refer them to receive specialized attention, to differ the consultation whenever possible, to solve the patient's problem via telemedicine care, and to define the circuit the patient should go for the personal consultation 37 [fig_ref] Figure 1: Suggested flow-chart to deliver ophthalmologic assistance during COVID-19 pandemic [/fig_ref].
It may be useful to give informed consent to the patient before the appointment. It should be adapted to local law and it should basically inform the patient about the risks during the pandemic, while at the same time, it may serve as a written declaration of symptoms compatible with COVID-19 or of contact with confirmed cases.Appointments should be organized in such a way that the patients do not overlap in common areas. The patient should be indicated to come alone or explained that their
## Dovepress
## 263
companion should wait outside the facility if possible. It is also good to remind them to use a mask and eye protection. [bib_ref] Facing COVID-19 in Ophthalmology Department, Romano [/bib_ref] Once the patient arrives, triage should be repeated. Provide alcohol-based hand sanitizer. [bib_ref] Virucidal activity of world health organization-recommended formulations against enveloped viruses, including Zika,..., Siddharta [/bib_ref] Once the patient enters the examination room, the ophthalmologist should wear PPE according to the exposure planned. It is not common that ophthalmologists perform aerosol-generating procedures at the praxis. The risk 264 comes mainly because of the very close working distance between doctor and patient and when getting in touch with the mucosae, ocular secretions, and tears. For a safe working environment, it is important to train the healthcare team in the proper use of PPE. [bib_ref] Physical distancing, face masks, and eye protection to prevent person-to-person transmission of..., Chu [/bib_ref]
## Ppe considerations in ophthalmology respiratory protection
There is consensus that the use of N95 respirators and face masks is indicated when facing patients who are COVID-19 positive or suspected to be infected during the aerosolgenerating procedure. [bib_ref] All India ophthalmological society -Indian journal of ophthalmology consensus statement on preferred..., Sengupta [/bib_ref] [bib_ref] Medical masks vs N95 respirators for preventing COVID-19 in healthcare workers: a..., Bartoszko [/bib_ref] Nevertheless, during a simulated clinical activity, airborne aerosol generation was evidenced during speech and sneezing. N95 respirators mitigated airborne aerosol transmission while standard surgical masks did not. [bib_ref] Airborne aerosol generation during endonasal procedures in the era of COVID-19: risks..., Workman [/bib_ref] Given this evidence, N95 respirators would provide physicians and healthcare workers with major protection.
When the supply of N95 respirators is short, they should be kept for high-risk use and aerosol-generating procedures only. [bib_ref] A comprehensive Chinese experience against SARS-CoV-2 in ophthalmology, Yu [/bib_ref] Evaluate contingency measures for reprocessing N95. [bib_ref] Sterilization of N95 respirators: the time for action is upon us!, Prakash [/bib_ref] [bib_ref] Proposal for a EN 149 acceptable reprocessing method for FFP2 respirators in..., Widmer [/bib_ref] [bib_ref] Microwave-generated steam decontamination of N95 respirators utilizing universally accessible materials, Zulauf [/bib_ref] The National Institute for Occupational Safety and Health (NIOSH) published a recommended guidance for extended use and limited reuse of N95 filtering facepiece respirators in healthcare settings, which emphasizes there is no determined limited time of use when a shortage of supplies.
## Eye protection
When close contact is required, eye protection equipment should be used as part of a full PPE.Goggles should be worn for aerosolizing procedures, while glasses with an enveloping design may be easier to wear at the consultation.
## Hand protection
One of the most important prevention methods is hand hygiene.Alcohol-based formulations proposed by the World Health Organization (WHO) are proved to be highly efficient. [bib_ref] Virucidal activity of world health organization-recommended formulations against enveloped viruses, including Zika,..., Siddharta [/bib_ref] The use of gloves is important when having direct contact with the patient. Gloves are subject to single-use. Ensure their removal does not cause additional contamination of hands. 17
## Body protection
The physicians should protect their personal clothes with a white coat, disposable fluid repellent coveralls, or longsleeved disposable fluid repellent gowns. These last two options will be required for aerosol-generating procedures.
The clothing can be worn for a single session. This means it should be changed every time that a health worker leaves a specific clinical care setting or exposure environment. 17
## Eye examination considerations to reduce viral transmission
- Breath shields or slit lamp barriers are a must at this moment. 37 - When testing for visual acuity, a phoropter is easier to sanitize. - A hypoallergenic tape can be placed on the nose bridge of the surgical mask or cloth mask of the patient, to avoid blurring of the glasses with breathing vapour. This may be useful as well for other proceedings. - Binocular indirect ophthalmoscope delivery system is preferred for fundus examination at the slit lamp. - Determine intraocular pressure only when necessary in this transition time of the pandemic. Disposable tonometry devices are preferred. Avoid non-contact "air-puff" tonometry, since it generates aerosol. [bib_ref] Microaerosol formation in noncontact 'air-puff' tonometry, Britt [/bib_ref] Perkins tonometer use is also discouraged since it requires extremely close working distance. The use of alcohol 70% should be effective at disinfecting tonometer tips from SARS-CoV-2, always remembering it would not be enough to inactivate adenovirus. In this last case, sodium hypochlorite dilution would be effective. - Keep communication between patients and physician as minimum as possible during the visit. Avoid talking during a close examination. 55 - Strive to perform all the studies needed during the same visit. - When multidose eye drops are required, evaluate fractioning for daily disposal. Try not to touch the ocular surface nor the eyelashes with the tip of the eye drop bottle at the moment of instilling it as it should be. 56 - Disinfect your hands after having any contact with the patient. - Take notice of the instruments and medications' placement in the consulting room, avoid their contamination and any source of transmission. - Try to reduce ophthalmic diagnostic testing mainly when they will not change treatment criteria. - Companies are developing and already offering ophthalmic equipment with a remote command to preserve social distancing.
## 265
- Manufacturers are providing guidance on their websites on the disinfection process of specific equipment. 57
## General considerations for ophthalmologic institutions
Ophthalmologist and ophthalmic administrators should introduce structural and organizational changes to their institutions [fig_ref] Table 1: General Recommendations for Institutions and Health Workforce Institution • It is wise... [/fig_ref].
## Defining cases, triage and screening process defining criteria for appropriate surgical cases
The crucial step is to define the epidemiological situation the ophthalmic institution is immersed in, and its real capacity to handle urgent or elective surgeries in this pandemic.
In order to create a formal frame to list urgencies that should be attended, the Argentinean Council of Ophthalmology created a consensus by an expert panel, conformed by representatives of each ophthalmic subspecialties societies of Argentina, the Committee for Prevention of Ocular Infections and the Chairs of the Ophthalmology Department of all the Universities of Argentina.In its turn, different societies worldwide gave similar suggestions.It is important to highlight that the medical criteria to decide what case should be operated on will always be fundamental.
Before performing eye surgery for a given case, the ophthalmologist has to consider and evaluate risks and benefits very thoroughly. As stated by the Centers for Disease Control and Prevention, USA (CDC), people at a higher risk are older adults (with 65 years old or above) and people with underlying medical conditions (such as chronic lung disease or moderate to severe asthma, a serious heart condition, severe obesity, diabetes, chronic kidney disease undergoing dialysis, liver disease and those immunocompromised). Every mentioned situation could represent serious disadvantages when performing surgery. [bib_ref] Mortality and pulmonary complications in patients undergoing surgery with perioperative SARS-CoV-2 infection:..., Archer [/bib_ref] A retrospective study on patients that developed COVID-19 pneumonia shortly after general surgery were older, more likely to have underlying comorbidities and underwent more difficult surgeries. [bib_ref] Mortality and pulmonary complications in patients undergoing surgery with perioperative SARS-CoV-2 infection:..., Archer [/bib_ref] This impact was not studied on ophthalmic operated patients, ocular surgery should not have the same negative repercussions on COVID-19 evolution, since it does not have an important hemodynamic impact, the area of compromised tissue is small and anaesthesia is usually done locally.
In this pandemic context, surgeries have been curtailed to preserve resources to face special medical needs raised by COVID-19. Elective surgery could be understood as optional in this scenario. Nevertheless, if surgery was consent between the patient and the physician, except limited cases, has implicit an important rationale backingup the indication. If the pandemic prolongs, the elective surgeries need to come to the agenda [fig_ref] Table 2: Facts to Consider in the Decision-Making Process Regarding the Necessity of the... [/fig_ref]. - Use a signalling system to indicate circulation, where people should stand, where they should sit, to remind them to keep social distancing. [bib_ref] COVID-19: guidance on social distancing and for vulnerable people, Health England [/bib_ref] - Make sure to install desks and countertop barriers for employee protection.
- As a way to reduce the use of PPE and avoid viral transmission, physical barriers, such as acrylic or glass windows, could be installed. 60
- Ensure adequate ventilation and air exchange in all areas. High-efficiency particulate air (HEPA) filters in air conditioning systems will reduce the possible dissemination of the virus.
- Reduce unnecessary objects on surfaces to facilitate the cleaning and disinfection process.
- Implement the digital signature in as many processes as possible (eg at the registration desk, in informed consents, prescriptions, and medical records).
## Staff
- The staff could be divided into two or more teams, avoiding contact between them, to reduce chances of interpersonal dissemination of the virus.
- Protect all the staff that may be at higher risk for COVID-19. (Assign roles that are out of exposure: telemedicine, counselling, writing, research, etc.).
- Contemplate the introduction of infection control by checking temperature and symptoms to staff daily. A group of the University of Chicago Medicine and Biological Sciences proposed a scoring system to ethically and efficiently manage resource scarcity and provider risk during the COVID-19 pandemic called Medically Necessary, Time-Sensitive Procedures (MeNTS). They considered 21 factors divided into three main areas: procedure factors, disease factors and patient factors. This system offers an objective way to ponder the indication of surgery. It helps to consider the major variables that may have an impact on the risk of the surgery. [bib_ref] Medically necessary, time-sensitive procedures: scoring system to ethically and efficiently manage resource..., Prachand [/bib_ref] The Spanish Association of Surgeons published a dynamic scale of five possible scenarios/alerts that take into account the percentage of beds occupied by COVID-19 patients in ICU, the compromise of resources, and the surgical procedures that should be performed.Taking this scale into account and adapting these criteria to ophthalmology, we could plan our surgical indications in five different scenarios given mainly by the curve of the infection and its direct relationship to the percentage of ICU occupancy [fig_ref] Table 3: Comparative Chart [/fig_ref].
Along stages 1 to 4, ophthalmologists should always consider proceeding with promptness to operate in urgent cases. Close to scenario 5, it is interesting to have in mind that it may be important to increase the critical care capacity and, since elective surgeries are postponed, ORs and postanesthesia care units (PACUs) could be converted into ICUs. [bib_ref] Transforming ORS into ICUs, Peters [/bib_ref] Preoperative Screening Process to Rule Out COVID-19
The main objective of making a COVID-19 screening is to avoid transmission of the illness along the surgical process, to keep a circuit that is free of COVID-19 and to decrease risk in presymptomatic patients who could be operated while infected with SARS-CoV-2.
Since the beginning of the outbreak of the virus in Argentina, as well as in many other countries, the health system has had significant limitations of their diagnostic testing capabilities. Moreover, such limitations became much clearer when the restrictive criteria for testing control were removed by the CDC. [bib_ref] Co-infection with SARS-CoV-2 and influenza a virus in patient with pneumonia, Wu [/bib_ref] Until today physicians are recommended to use their judgment for screening patients and setting up biosecurity protocols. For this matter, we have included a variety of screening methods reported in multiple publications stating their results and limitations.
The preoperative screening will vary depending on each facility's capacities, its epidemiological context, the type of patient, and the availability of diagnostic tests.
## Epidemiologic and symptomatic triage
This evaluation aims to identify patients who are highly suspected of SARS-CoV-2 infection, regardless of any diagnostic test results [fig_ref] Table 4: Facts to Inquire in the Triage Process During COVID-19 Pandemic Epidemiological •... [/fig_ref]. [bib_ref] Clinical characteristics of COVID-19 patients with digestive symptoms in Hubei, China: A..., Pan [/bib_ref] The evaluation is suggested to be made using telemedicine within 72 h before the surgery. It should be repeated in person on the day of the surgery.
Another approach recommends to carry it out 14 days before the proceeding, recommending the patient to avoid exposure to possible sources of contamination in their community.Triage results should be included in the patient's medical history.
## Diagnostic tests for detection of sars-cov-2
Nasopharyngeal RT-PCR Real-time reverse-transcriptase polymerase chain reactionbased assays performed in a laboratory are the reference standard for COVID-19 diagnostics. There are protocols that include this exam as a routine previous to any ophthalmic surgical procedure.
It is the most reliable test that confirms active infection because it has a high specificity to detect the virus (low rate of false positive).
Nevertheless, its sensitivity is estimated to be 70%. This means a high rate of false negative. [bib_ref] The potential transmission of SARS-CoV-2 from patients with negative RT-PCR swab tests..., Cao [/bib_ref] This estimate will be modified mainly by the chances of infection of the patient's community (pretest probability) and the quality
## Diagnostic tests for detection of antibodies against sars-cov-2: igm and igg
The tests that detect antibodies against SARS-CoV-2 are not the gold standard to diagnose an active infection. They are tools that help to know, still with an important degree of uncertainty, if the patient has been in contact with the virus and may give a notion of where the patient might be situated in the natural history of the disease.
These tests could give more information regarding the cases suspected to have the infection, with repeated negative PCR results. It was proved that this group of patients can increase the risk of nosocomial transmission. [bib_ref] Diagnostic testing for severe acute respiratory syndrome-related Coronavirus 2: a narrative review, Cheng [/bib_ref] Negative results would not exclude SARS-CoV-2 infection.IgM represents the humoral immune reaction to the virus. The antibody detection is performed by the enzyme-linked immunosorbent assay (ELISA). Immunoreactivity begins around the first week of symptoms onset, it reaches a peak at week three, and it lowers at week four to five. IgG can be detected after the second week and remains positive for several months. It still needs to be clarified to what extent a positive IgG response confers resistance to reinfection with SARS-CoV-2 and for how long [fig_ref] Figure 2: Chronological estimate of diagnostic test probability to detect SARS-CoV-2 in the COVID-19... [/fig_ref]. [bib_ref] Interpreting diagnostic tests for SARS-CoV-2, Sethuraman [/bib_ref] Thoracic Imaging
## Role of thoracic imaging in preoperative assessment
With RT-PCR as a gold standard, chest computedThis statement contrasts with some guidelines that recommend chest X-rays as a preoperative exam. [bib_ref] All India ophthalmological society -Indian journal of ophthalmology consensus statement on preferred..., Sengupta [/bib_ref] Taking into account the last considerations, among the difficulties we still encounter, the most important ones are the gaps that remain in screening asymptomatic persons during the incubation phase.The New Normal: Considerations to Perform an Elective Surgery on a Recovered COVID-19 Patient
In order to perform programmed eye surgery in a patient that had the infection, the advice of a specialist in COVID-19 should be followed since guidelines are frequently changing. Time since diagnosis and recovery should be confirmed to uphold the decision as to whether clearance testing is required or not.
Ideally, the patient should already be asymptomatic for more than 14 days and count with two negative PCRs results separated by 24 h.Additionally, IgM should be negative and IgG, already with detectable titles.
## Operating room considerations
## Sars-cov-2 and eye surgical procedures
During eye surgery, the main factors that may impact viral transmission are 1-procedures that have the potential to generate aerosol, 2-working distance between ophthalmologist and patient, 3-prolonged time of surgery, 4-air flow, and 5-use of barrier methods and personal protective equipment (PPE). Ophthalmic procedures and the potential to generate aerosol in the operating room (OR):
-Phacoemulsification 91 -Cautery use [bib_ref] Airborne aerosol generation during endonasal procedures in the era of COVID-19: risks..., Workman [/bib_ref] -Excimer laser [bib_ref] Respirable particles in the excimer laser plume, Taravella [/bib_ref] -Vitrectomy 93 -Oculoplastic, orbit, strabismus, tumoral and nasolacrimal duct surgery: Given that common ophthalmic procedures do not generate an important amount of aerosols, the main source of transmission continues to be the patient's respiratory tract. During this transition of the COVID-19 pandemic, general anaesthesia should be replaced, if the case allows it, to avoid aerosolization during intubation.
-YAG and Argon laser: They are not supposed to produce aerosol patients that need to be treated with panphotocoagulation; they are mostly diabetic and immunocompromised.
## Considerations for ophthalmological surgical procedures to reduce viral transmission
Topical povidone-iodine use should be considered prior to any procedure to lower the risk of any unknown or uncertain evidence of virus transmission.Standard surgical antisepsis using povidone-iodine 5% would reduce the viral load on the conjunctiva. Topical povidone-iodine is virucidal with in vitro testing of 0.23% povidone-iodine producing a 10,000-fold reduction in the viral load of SARS-CoV and MERS-CoV within 15 seconds. [bib_ref] The outbreak of COVID-19: an overview, Wu [/bib_ref] [bib_ref] In vitro bactericidal and virucidal efficacy of povidone-iodine gargle/ mouthwash against respiratory..., Eggers [/bib_ref]
## Phacoemulsification considerations and aerosolization during cataract surgery
In the phacoemulsification procedure, the majority of the aqueous humour is visco-expressed before capsulorhexis which creates a replacement of the concerning infected intraocular fluid. Performing irrigation/aspiration for a minimum of 6 seconds prior to starting active phacoemulsification will significantly decrease the proportion of any remaining intraocular fluid (and consequently any theoretical virus load) in any aerosol produced. As a result, the aerosolization created during phacoemulsification is generated mainly from BSS (the surgery is performed in a closed intraocular system which allows this dilution process to occur). [bib_ref] Reducing visible aerosol generation during phacoemulsification in the era of Covid-19, Darcy [/bib_ref] Administrate hydroxypropyl methylcellulose 2% (HPMC) every minute during phacoemulsification to coat the corneal surface, including the main incision. HPMC is a low viscosity and moderate viscoelastic compound, that acts as a barrier for aerosol formation and dispersion. [bib_ref] The use of hydroxypropyl methylcellulose in extracapsular cataract extraction with intraocular lens..., Liesegang [/bib_ref] A smaller incision size reduces aerosol leakage. This statement was noted in a comparison between a 2.75 mm tip with a 2.2 mm tip. [bib_ref] Corneal wound architecture and integrity after phacoemulsification. Evaluation of coaxial, microincision coaxial,..., Berdahl [/bib_ref] Applying the measures mentioned above, the risk of aerosolized virus during phacoemulsification is expected to be greatly reduced.
## Pars plana vitrectomy (ppv) and other posterior segment procedures
The PPV is expected to involve a low risk of virus transmission due to the fact that it is performed in a closed surgical system with valved trocar cannulas. Aerosol generated during any procedure is expected to be contained within the eye. However, external cautery should be minimized and performed with vigorous irrigation.The intravitreal injections procedure has no additional risk of transmission from the ocular surface. However, given that diabetic and elderly patients are at a higher risk for COVID-19 complications, it is important to consider prioritizing those at the greatest risk of irreversible vision loss. The Royal College of Ophthalmologists has prepared guidance as a temporary response to the clinical management of patients during the ongoing COVID-19 pandemic.
# Cautery-related procedures
Oculofacial plastics, pediatric ophthalmology, ocular surface tumours and reconstruction.
Cautery implementation should be reduced to a minimum or avoided, if possible since it has been associated with significant airborne particulate matter production. [bib_ref] Detection of air and surface contamination by SARS-CoV-2 in hospital rooms of..., Chia [/bib_ref] During cautery-related procedures, it is recommended to wear N95 masks or equivalent. [bib_ref] Detection of air and surface contamination by SARS-CoV-2 in hospital rooms of..., Chia [/bib_ref] One option to reduce cautery plume is to apply generous irrigation. Another option to consider is to perform active air suction right next to the area where coagulation is intended.
## Pan-photocoagulation treatment
For this intervention, the physician should consider the recommendations given for eye examination. For instance, it is suggested the use of breath shield on a laser slit lamp, the hypoallergenic tape to secure the patient's surgical mask on the nasal bridge, to avoid talking, and to use proper PPE (gloves, long sleeves gowns, and respirators, if available). Instilling a drop of iodopovidone 5% after proparacaine may be considered to reduce the possible submit your manuscript | www.dovepress.com
## Dovepress
Clinical Ophthalmology 2021:15 presence of virus on the eye surface. Panphotocoagulation contact lenses should be disinfected according to the manufacturer's recommendations. Washing the contact lenses with detergent and water proved to be effective to control viruses and bacteria as well as disinfecting them with bleach. [bib_ref] Efficacy of detergent and water versus bleach for disinfection of direct contact..., Abbey [/bib_ref] General Considerations for the OR Each institution should adapt the recommendations according to its capabilities. It is important to critically evaluate if the institution can responsibly face the performance of the surgery of a suspected or confirmed COVID-19 patient. When it is not convenient to operate this type of patient in the institution, you should have the patient referral ready.
## Staff assignments
Reduce the number of personnel and circulation within the operating room. It is important to plan the tasks to be carried out, the minimum necessary elements inside the OR and the assignment of roles. The training of all the members of the staff that work in the OR should be mandatory. It aims at avoiding cross-infection. If necessary, the staff could be trained using simulation [fig_ref] Table 5: Staff Assignments in the Operating Room for an Ophthalmic Surgery During COVID-19... [/fig_ref].
## Or cleaning and disinfection considerations
Aerosol and surface stability of SARS-CoV-2 has been proven in multiple studies. The proper cleaning and disinfection in the OR are crucial for both patient and healthcare staff according to the procedures that were followed. [bib_ref] Protection and disinfection policies against SARS-CoV-2 (COVID-19), Fathizadeh [/bib_ref] Improve the organization in the OR to increase the frequency of cleaning and ensure a thorough cleaning. The time between surgeries will be compromised and should be scheduled accordingly.
These interventions require proper compliance. Evidence-based surveillance can be implemented optimizing sustainability. 107
## Considerations for the cleaning personnel
Health care workers' training, practice, and observation for infection prevention are of great importance, as well as guidance on the type of PPE, and how the donning and doffing process should be made. [bib_ref] Perioperative COVID-19 defense: an evidence-based approach for optimization of infection control and..., Dexter [/bib_ref] Before entering the OR, air contamination should be addressed. [bib_ref] Detection of air and surface contamination by SARS-CoV-2 in hospital rooms of..., Chia [/bib_ref] A recent study showed a 3-hour persistence of SARS-CoV-2 in aerosols.
## 271
When entering the OR after surgery of a COVID-19 positive patient, the use of PPE is mandatory (face shield or goggles, N95 or higher respirator, isolation gown, a pair of non-sterile gloves).
## Disinfectant solutions
Disinfectant solutions proven to eliminate the virus are sodium hypochlorite 0.21% [fig_ref] Figure 3: Formula to dilute sodium hypochlorite at desired concentration [/fig_ref] , alcohol solutions at 70-95%, and hydrogen peroxide at 0.5%, glutaraldehyde 0.5-2.5%, formaldehyde 0.7-1% and quaternary ammonium. [bib_ref] COVID-19 vulnerability: the potential impact of genetic susceptibility and airborne transmission, Pollitt [/bib_ref] [bib_ref] Efficacy of various disinfectants against SARS coronavirus, Rabenau [/bib_ref] [bib_ref] Disinfection with sodium hypochlorite in hospital environmental surfaces in the reduction of..., Pereira [/bib_ref] The specific disinfectant solution should be used as instructed by the manufacturer.
A simple approach consists in wet cleaning with two buckets: cleaning with detergent (10 ml of detergent diluted in 10 litres of warm water), then disinfecting all the surfaces with sodium hypochlorite at 0.5%.The cleaning sequence should remain the same. Decontaminate all surfaces, considering screens, keyboard, mouse, cables, monitors, and anaesthesia machines. For proper inactivation of the coronavirus, surfaces should be exposed to solutions for at least 1 minute. [bib_ref] COVID-19 vulnerability: the potential impact of genetic susceptibility and airborne transmission, Pollitt [/bib_ref] Some protocols consider the use of hydrogen peroxide vaporizers, UV-C lights and chlorine dioxide solution to decontaminate the OR. More evidence-based research should be done to adopt these recommendations.
## Considerations after surgery, follow-ups, and complications
The same criteria used for previous processes should be applied here: reduce patient time spent at the institution and employ proper use of hygiene and barrier methods. For surgical patients, one needs to consider several factors before proceeding, such as anaesthesia, operating room workflow/staffing, and how nursing preoperative and postoperative care will be managed. For example, if the patient requires close follow-up or may be at higher risk for postoperative complications, unless urgency or emergency, this surgical patient should be deferred so as not to make them run the unnecessary risk of contracting the virus with a potentially unfavourable outcome. Virtual visits: try to make follow-ups using telemedicine.
- Reduce the face-to-face visit rate. - Provide the patient with written instructions and alarm symptoms, as well as ways of seeking medical assistance. - The postoperative exam is hard to be replaced.
However, if required, try to streamline the examination by testing what is essential and hold the rest of the consultation by phone or using telemedicine.
Proper documentation of the steps aimed at mitigating the transmission of the virus and keeping patients and staff
## Surgical assistant
- Properly plan the material that will enter the OR. Any unnecessary material will be assumed as contaminated and should be discarded.
## Surgeon
- The operation should be led by an experienced surgeon to minimize complications, risks, and time of exposure inside the OR.
- Avoid risky manoeuvres during surgery.
Aim to perform safe surgeries.
- Contemplate those surgical approaches that have the lowest probability of postoperative complications.
## Anesthesiologist
Since this is one of the most exposed specialities, generally anaesthetists have very clear protocols to follow to reduce aerosolization and extreme air contamination during urgent and elective surgeries.- Try to avoid general anaesthesia unless mandatory.
- Topical anaesthesia should be preferred to local anaesthesia whenever possible.
- The OR staff must not access the OR until the patient is intubated.
- N95 or OR PAPR (powered air-purifying respirator) must be used for all aerosolgenerating procedures.
- If intubation is required for the surgery, recommend intubation in a negative pressure room prior to OR, as well as extubation. Consider using an aerosol box to reduce aerosolization in the OR. [bib_ref] Maximising application of the aerosol box in protecting healthcare workers during the..., Malik [/bib_ref] [bib_ref] Protecting against COVID-19 aerosol infection during intubation, Tseng [/bib_ref] - Use an extra heat and moisture exchanger (HME) filter and viral filter on the expiratory limb of the anaesthesia machine circuit. [bib_ref] Protecting against COVID-19 aerosol infection during intubation, Tseng [/bib_ref] Clamp endotracheal tube before any disconnection is planned. safe is prudent, safeguarding yourself and the institution against legal action. We encourage ophthalmologists to take special consideration not only to actively work on prevention but also to document proof of all the measures and special precautions taken. Thus, it is recommended that the medical record reflects good rationale as to why the patient was operated during this pandemic and that it includes the patient's consent to the possibility of increased morbidity. During the COVID-19 pandemic, it is important to demonstrate that the institution has implemented measures to avoid viral dissemination and has taken special care to address patients' concerns. The lack of adequate legal protection for surgeons seems to have an impact in some institutions, reducing the volume of surgical interventions during a pandemic era and the immediately following period. This needs to shift, in order to provide efficient health care to all patients. [bib_ref] Safety recommendations and medical liability in ocular surgery during the COVID-19 pandemic:..., Napoli [/bib_ref]
# Discussion
In this review, we aim to provide information as well as several recommendations for ophthalmologists and healthcare institutions to better adapt to the new situation created by the COVID-19 pandemic.
First, regarding hospital triage: a reasonable approach to help standardize indications and preserve resources for more urgent cases according to ICU occupancy has been adapted following, in part, the proposals by the Spanish Association of Surgeons. Although this model has not yet been validated, its usefulness seems reasonable. In addition, we have outlined factors for the surgeon to keep in mind when judging surgical indications.
Second, the eye as a source of viral transmission: the presence of SARS-CoV-2 on the ocular surface and its infectious capacity is an obvious major concern in the ophthalmologic community. Although this remains controversial, it should soon be clarified by evidence. After writing this review, we realize that considering the eye as a source of viral dissemination for the ophthalmologist may not have the expected relevance. The reason for this is that tears and ocular secretions are body fluids we should always be aware of and not only during the COVID-19 pandemic. Taking the precautions mentioned above, the risk of transmission can be significantly lowered while providing care of ophthalmic patients. The focus of concern should continue to be the respiratory tract as the main source of viral spread.
Third, viral screening and testing: there are multiple approaches in the screening process to rule out COVID-19; up to now, they have shown to be expensive, extensive and non-conclusive. Significant work has gone into making diagnostic tests more reliable and accessible. However, beyond this, given the possibility that a COVID-19 patient may not be detected by screening, precautions and protections should never be avoided.
Fourth, viral transmission: the airborne route of transmission is still controversial, and this explains the main differences in PPE recommendations. On the one hand, the World Health Organization recommends droplet and contact precautions for those people caring for COVID-19 patients. Their argument is that airborne transmission remains unproven in clinical situations since it was not reported in the analysis of 75,465 COVID-19 cases in China. Furthermore, airborne evidence was studied in simulated conditions and was largely not peer-reviewed.
On the other hand, the US CDC and the European Centre for Disease Prevention and Control recommend airborne precautions for any situation involving the care of COVID-19 patients and consider the use of medical masks as an acceptable option in case of respirator shortages. Moreover, it is unanimously recommended to use full protection for aerosol-generating procedures.
Fifth, following all the recommendations to take care of the health of patients and workforce at ophthalmic institutions becomes unproductive when 24-hour personal care is not taken.
Until the COVID-19 pandemic reaches its end, it is crucial to monitor new publications and recommendations, remaining flexible in order to actively make appropriate modifications during this pandemic.
# Conclusion
When facing a viral pandemic, our suggestion is that a designated group or committee summarizes preferred practices and tailors them to the particular institution. They should take into account the epidemiologic context, the local regulations, the patient's needs, the economic situation, the access to diagnostic tests, and the services offered.
Making a triage according to the patient's conditions and eye pathology, reducing the time the patient is at the institution, keeping social distancing, accurately using personal protective equipment (PPE), placing barrier methods, taking care of hygiene, as well as following other recommendations mentioned in this review, will allow physicians to take care of the visual health of the patients while reducing the impact of the COVID-19 pandemic.
## Disclaimer
The authors would like to declare that this document contains recommendations based on the best evidence found at the moment of writing, and also contains personal analysis considered important for ophthalmology.
This document does not have the intention to set a standard of care and cannot ensure a successful outcome in every situation. These guidelines aim to help adapt protocols according to the regional, local and individual hospital policies and expertise. New information regarding this topic may change the recommendations in this review in a significant way; and therefore, we cannot warrant the completeness of the guidance.
[fig] Figure 1: Suggested flow-chart to deliver ophthalmologic assistance during COVID-19 pandemic. submit your manuscript | www.dovepress.com DovePress Clinical Ophthalmology 2021:15 [/fig]
[fig] •: Assign inspection roles. Check and recheck processes and implementation of protective measures. submit your manuscript | www.dovepress.com DovePress Clinical Ophthalmology 2021:15 [/fig]
[fig] Figure 2: Chronological estimate of diagnostic test probability to detect SARS-CoV-2 in the COVID-19 natural history. Estimated time intervals should be considered approximations. Publications are not consistent with this information since many factors have a significant influence on the shift of the curves. 1-PCR indicates real-time polymerase chain reaction. 2-IgM, Immunoglobulin M. 3-IgG, Immunoglobulin G. Clinical Ophthalmology 2021:15 submit your manuscript | www.dovepress.com DovePress 269 [/fig]
[fig] Figure 3: Formula to dilute sodium hypochlorite at desired concentration. Clinical Ophthalmology 2021:15 submit your manuscript | www.dovepress.com DovePress 273 [/fig]
[table] Table 1: General Recommendations for Institutions and Health Workforce Institution • It is wise to have two different clear circuits. One for patients who have tested negative for COVID-19 and one for patients with symptoms or suspected epidemiological background positive to COVID-19. [/table]
[table] Table 2: Facts to Consider in the Decision-Making Process Regarding the Necessity of the Elective Eye Surgery During COVID-19 Pandemic Institution • Resource limitations (PPE, health workforce, OR necessity for other urgent procedures) • COVID-19 transmission risk • Ophthalmic Institution Only/General Hospital • OR facility and installation [/table]
[table] Table 3: Comparative Chart: Different Possible Scenarios According to the Percentage of Intensive Care Unit Occupancy [/table]
[table] Table 4: Facts to Inquire in the Triage Process During COVID-19 Pandemic Epidemiological • Close contact with a person known or suspected to have coronavirus. The patient has been in an area where ongoing COVID-19 transmission has been confirmed. The patient has ever been diagnosed with COVID-19 infection Date of diagnosis Time Despite the sensitivity levels of chest imaging, the CDC does not currently recommend to diagnose COVID-19. This health institute states that viral testing remains the only specific method of diagnosis. Confirmation with the viral test is required even if radiologic findings are suggestive of COVID-19 on radiograph or CT. Similarly, the American College of Radiology (ACR) has recently communicated CT should not be used to screen for or as a first-line test to diagnose COVID-19. CT should be used sparingly and reserved for hospitalized, symptomatic patients with specific clinical indications for CT. [/table]
[table] Table 5: Staff Assignments in the Operating Room for an Ophthalmic Surgery During COVID-19 Pandemic Member of the Surgical Team Assignments in the Operating Room Circulating Nurse • Have a preoperational checklist • In charge of bringing materials and instruments into the OR as the surgery develops. [/table]
|
Integrated Digital and Conventional Treatment Workflow in Guided Complete Mouth Implant Rehabilitation: A Clinical Case Report
The introduction of digital dentistry and CAD/CAM technology has redefined treatment concepts in implant dentistry-computer guided implant placement has become routine practice, and CAD/CAM prostheses are now commonplace. These advances in treatment options and modalities has led to a paradigm shift in the workflow of surgical and restorative treatments. This case report presents a customized staged treatment protocol that involves the strategic retention of teeth to serve as transitional abutments, which will support a computer guided implant surgical guide as well as a fixed interim prosthesis. The treatment protocol also describes an integrated digital and conventional workflow for full mouth implant-supported fixed prosthetic rehabilitations to provide improved patient care with more predictable outcomes and fewer complications.The overall expected prognosis of the presented treatment is considered good with patient compliance in the wearing of an occlusal guard and following a proper recall and maintenance regimen.ConclusionsWith adequate planning and proper execution, a phased computer-guided implant surgical protocol and CAD/CAM technology can bring accuracy, efficiency, and predictability to complex full-mouth implant rehabilitation cases. Understanding the limitations of conventional and digital dentistry will allow the clinician to create a synergistic integration of processes from both realms to optimize the management of complex and challenging treatments.
# Introduction
The challenges involved in the treatment of completely edentulous patients with fixed implant rehabilitations have been reported in the literature. Inadequate surgical and restorative planning can result in both biological and prosthetic failures. A thorough analysis of the available prosthetic space, as well as an evaluation of the functional and esthetic parameters through a diagnostic wax up are critical steps in planning the position and angulation of implants. Advances in three-dimensional (3D) imaging technology and digital scanners have allowed for better visualization of anatomic structures and improved precision and accuracy for implant placement. By merging the DICOM data from CT scans of edentulous arches and STL files from a digital scan of a diagnostic prototype, implant placement can be virtually planned in relation to the final prosthesis in 3D software. A surgical template is then fabricated according to the virtual plan by the placement of surgical guide sleeves. The clinical viability of using computer-guided surgery has been investigated and has become a routine practice in implant dentistry. The advantages of a computer guided surgical protocol are manifold: It serves as a tool for preoperative analysis, provides for predictable implant positioning, allows optimal utilization of existing bone with minimum bone augmentation, and facilitates the possibility of a flapless approach.
Once the implants are placed in prosthetically idealized positions based on the guided surgical planning, the restorative phase is greatly simplified as the appropriate prosthetic space as well as implant access hole locations have been virtually pre-verified. The combination of CAD-CAM technology with computer guided implant surgery not only simplifies the treatment protocol, but can also enhance the final prosthetic outcome.
The transitional period from a provisional to a definitive prosthesis during a complete mouth implant reconstruction presents a challenge for clinicians and patients, alike. Unexpected procedural complexities may hamper the clinician's progress, and patients may experience difficulty in accepting and adapting to prosthetic changes. Various techniques have been applied to facilitate this transition including the use of removable dentures and provisional mini-implants, as well as immediate implant loading protocols to reduce the treatment duration and enhance the patient experience. A staged approach, in which questionable or hopeless dentition are retained as transitional abutments, has been considered beneficial for easing the transition from the provisional stage to the final restoration. This approach involves strategic extractions, implant placement, tooth-supported fixed interim prostheses, and strategic timing of implant loading and restoration. One major advantage of employing a staged approach is the presence of a fixed provisional throughout the entire treatment period; hence, implants are protected from premature loading and patient acceptance is increased. An additional clinical benefit of a staged approach is that the abutments can serve to support both the radiographic and surgical templates in a stable and repeatable position-a simpler alternative to the traditional horizontal fixation pins used to stabilize soft/hard tissue-borne templates.
The aim of this clinical report is to describe an integrated digital and conventional treatment workflow in the management of a fixed implant-supported complete mouth rehabilitation. A staged treatment approach was utilized in conjunction with computer guided implant placement, and definitive prostheses were fabricated using CAD/CAM titanium abutments and monolithic zirconia prostheses. Before starting the treatment, the subject gave her informed consent that records of the case would be available for teaching purposes, including scientific publication.
## Case presentation
## Diagnostic phase
A 77-year-old female patient presented to the Advanced Graduate Education Prosthodontics clinic at Harvard Dental Center with the chief complaint of "I would like to chew and smile." The patient requested fixed restorations and expressed a desire to avoid wearing removable prostheses during her treatment. The patient's medical history was non-contributory to any dental treatment with normal vital signs and a reported allergy to penicillin. The initial comprehensive clinicaland radiographicexamination revealed partial edentulism with defective restorations, inadequate occlusal plane and loss of unilateral posterior support. The patient's remaining teeth presented with poor to guarded prognoses involving advanced periodontal disease, caries, and Miller's class III mobility.
## Foundational phase: elimination of active disease, extraction and fixed provisional prosthesis
The first phase of treatment consisted of oral hygiene instruction and elimination of active disease, strategic extraction of teeth and fixed provisionalization. Eight implants were planned in the maxilla at the positions of the lateral incisors, canines, first premolars and first molars, and six Following a discussion of the benefits, risks and alternative treatment options, the patient consented to maxillary and mandibular fixed implant-supported rehabilitation. Based on the patient's existing anatomy, the planned prosthetic design, and biomechanical prosthodontic principles, eight implants were planned in the maxillary arch, and six were planned in the mandibular arch. The treatment goals were to restore form, function, and esthetics while maintaining the patient in fixed prostheses throughout the entire treatment period.
## Foundational phase: elimination of active disease, extraction and fixed provisional prosthesis
The first phase of treatment consisted of oral hygiene instruction and elimination of active disease, strategic extraction of teeth and fixed provisionalization. Eight implants were planned in the maxilla at the positions of the lateral incisors, canines, first premolars and first molars, and six implants were planned in the mandible in the areas of the canines, first premolars and first molars. Extraction sites were determined based on the planned implant positions while the remaining abutment teeth were kept to support a fixed provisional as well as the computer guided surgical template during the implant surgical phase
## Foundational phase: elimination of active disease, extraction and fixed provisional prosthesis
The first phase of treatment consisted of oral hygiene instruction and elimination of active disease, strategic extraction of teeth and fixed provisionalization. Eight implants were planned in the maxilla at the positions of the lateral incisors, canines, first premolars and first molars, and six implants were planned in the mandible in the areas of the canines, first premolars and first molars. Extraction sites were determined based on the planned implant positions while the remaining abutment teeth were kept to support a fixed provisional as well as the computer guided surgical template during the implant surgical phase
## Surgical phase: virtual planning of implant placement and computer guided surgery
The second phase of treatment involved an integrated digital and conventional workflow for guided implant surgical intervention. Eight weeks after the extractions, radiographic guides were fabricated by duplicating the provisional restorations in a radiopaque acrylic and affixing them to templiX reference plates (Straumann USA, LLC, Andover, MA, USA). Computerized tomography (CT) scans were taken with the patient wearing the radiographic guides stabilized on the transitional abutments. The implant positions were virtually planned in relation to the hard and soft tissue
## Surgical phase: virtual planning of implant placement and computer guided surgery
The second phase of treatment involved an integrated digital and conventional workflow for guided implant surgical intervention. Eight weeks after the extractions, radiographic guides were fabricated by duplicating the provisional restorations in a radiopaque acrylic and affixing them to templiX reference plates (Straumann USA, LLC, Andover, MA, USA). Computerized tomography (CT) scans were taken with the patient wearing the radiographic guides stabilized on the transitional abutments. The implant positions were virtually planned in relation to the hard and soft tissue architecture and the planned prostheseswith virtual implant planning software (coDiagnostiX. Straumann USA, LLC, Andover, MA, USA). Based on the virtual planning, the radiographic guides were converted into surgical templates (Straumann USA, LLC, Andover, MA, USA). During the pre-operative visit, the accurate seating of the surgical templates on the transitional abutment teeth was confirmed and irrigation holes were made. The implants were placed as planned following the manufacturer's recommended surgical protocol and guided surgery instrumentation (Straumann USA, LLC, Andover, MA, USA).
architecture and the planned prostheseswith virtual implant planning software (coDiagnostiX. Straumann USA, LLC, Andover, MA, USA). Based on the virtual planning, the radiographic guides were converted into surgical templates (Straumann USA, LLC, Andover, MA, USA). During the pre-operative visit, the accurate seating of the surgical templates on the transitional abutment teeth was confirmed and irrigation holes were made. The implants were placed as planned following the manufacturer's recommended surgical protocol and guided surgery instrumentation (Straumann USA, LLC, Andover, MA, USA).
[formula] (A) (B) [/formula]
## Restorative phase: second strategic extraction and conversion of provisional prostheses with integrated digital workflow
After a three-month healing period, the transitional teeth abutments were extracted. The tooth-supported provisional restorations were converted into screw-retained implant-supported provisional prostheses and the soft tissue around the pontic areas were contoured with an ovate pontic design.
## Restorative phase: second strategic extraction and conversion of provisional prostheses with integrated digital workflow
After a three-month healing period, the transitional teeth abutments were extracted. The tooth-supported provisional restorations were converted into screw-retained implant-supported provisional prostheses and the soft tissue around the pontic areas were contoured with an ovate pontic design.
## Restorative phase: second strategic extraction and conversion of provisional prostheses with integrated digital workflow
After a three-month healing period, the transitional teeth abutments were extracted. The tooth-supported provisional restorations were converted into screw-retained implant-supported provisional prostheses and the soft tissue around the pontic areas were contoured with an ovate pontic design. The prototypes were used to evaluate marginal fit and passivity, as well as to verify proper occlusion. Occlusal interferences were eliminated to achieve mutually protected occlusion. Gingival embrasures and pontic areas were adjusted to ensure adequate accessibility and contours for cleansability. . The verified prototypes were re-scanned and served as a blueprint for fabrication of the definitive anterior zirconia frameworks (Katana Zirconia, Kuraray Noritake Dental Inc. Miyoshi, Aichi, , Japan) as well as the posterior full contour monolithic zirconia restorations (Katana Zirconia HT, Kuraray Noritake Dental Inc. Miyoshi, Aichi, Japan). Once fabricated, the anterior zirconia frameworks and posterior full contour crowns were clinically and radiographically evaluated to confirm acceptable fit . The anterior zirconia frameworks were veneered with porcelain (Cerabien ZR (CZR), Kuraray Noritake Dental Inc. Miyoshi, Aichi, Japan). The segmented definitive prostheses were delivered with self-adhesive universal resin cement (Rely X Unicem, 3M ESPE, St. Paul, MN, USA). The patient was educated in proper hygiene protocols for the final prostheses, and an occlusal appliance was delivered to minimize attritive wear of the prostheses. The prototypes were used to evaluate marginal fit and passivity, as well as to verify proper occlusion. Occlusal interferences were eliminated to achieve mutually protected occlusion. Gingival embrasures and pontic areas were adjusted to ensure adequate accessibility and contours for cleansability. . The verified prototypes were re-scanned and served as a blueprint for fabrication of the definitive anterior zirconia frameworks (Katana Zirconia, Kuraray Noritake Dental Inc. Miyoshi, Aichi, Japan) as well as the posterior full contour monolithic zirconia restorations (Katana Zirconia HT, Kuraray Noritake Dental Inc. Miyoshi, Aichi, Japan). Once fabricated, the anterior zirconia frameworks and posterior full contour crowns were clinically and radiographically evaluated to confirm acceptable fit . The anterior zirconia frameworks were veneered with porcelain (Cerabien ZR (CZR), Kuraray Noritake Dental Inc. Miyoshi, Aichi, Japan). The segmented definitive prostheses were delivered with self-adhesive universal resin cement (Rely X Unicem, 3M ESPE, St. Paul, MN, USA). The patient was educated in proper hygiene protocols for the final prostheses, and an occlusal appliance was delivered to minimize attritive wear of the prostheses. The prototypes were used to evaluate marginal fit and passivity, as well as to verify proper occlusion. Occlusal interferences were eliminated to achieve mutually protected occlusion. Gingival embrasures and pontic areas were adjusted to ensure adequate accessibility and contours for cleansability. . The verified prototypes were re-scanned and served as a blueprint for fabrication of the definitive anterior zirconia frameworks (Katana Zirconia, Kuraray Noritake Dental Inc. Miyoshi, Aichi, , Japan) as well as the posterior full contour monolithic zirconia restorations (Katana Zirconia HT, Kuraray Noritake Dental Inc. Miyoshi, Aichi, Japan). Once fabricated, the anterior zirconia frameworks and posterior full contour crowns were clinically and radiographically evaluated to confirm acceptable fit . The anterior zirconia frameworks were veneered with porcelain (Cerabien ZR (CZR), Kuraray Noritake Dental Inc. Miyoshi, Aichi, Japan). The segmented definitive prostheses were delivered with self-adhesive universal resin cement (Rely X Unicem, 3M ESPE, St. Paul, MN, USA). The patient was educated in proper hygiene protocols for the final prostheses, and an occlusal appliance was delivered to minimize attritive wear of the prostheses.
# Discussion
In this case, the transition from failing dentition to a fixed implant-supported complete rehabilitation was accomplished in several phases: strategic extraction, virtual implant planning, computer guided surgery, and CAD/CAM prosthesis fabrication. Despite advances in the computer guided surgical technology, inaccuracies can accumulate from all stages of the workflow-from CT data acquisition, and data manipulation, to imprecision in the seating of the surgical template. Horizontal fixation pins are commonly used in computer guided implant systems treating edentulous arches. In such cases, however, the reproducible seating of the radiographic guide and surgical template can be a challenge due to the nature of mucosa-borne guided surgery. In the present case, the utilization of transitional tooth abutments allowed for reproducible and accurate positioning of the radiographic and surgical guides. Another benefit of strategically keeping transitional abutments is that quadrilateral positioning of transitional tooth abutments allows for fixed provisional prostheses and uninterrupted healing of the implants.
In the final restorative phase, the integration of digital technology with conventional prosthodontic techniques allowed for a customized treatment protocol. The use of a digital impression system for a single implant restoration has been found to be more efficient than
# Discussion
In this case, the transition from failing dentition to a fixed implant-supported complete rehabilitation was accomplished in several phases: strategic extraction, virtual implant planning, computer guided surgery, and CAD/CAM prosthesis fabrication. Despite advances in the computer guided surgical technology, inaccuracies can accumulate from all stages of the workflow-from CT data acquisition, and data manipulation, to imprecision in the seating of the surgical template. Horizontal fixation pins are commonly used in computer guided implant systems treating edentulous arches. In such cases, however, the reproducible seating of the radiographic guide and surgical template can be a challenge due to the nature of mucosa-borne guided surgery. In the present case, the utilization of transitional tooth abutments allowed for reproducible and accurate positioning of the radiographic and surgical guides. Another benefit of strategically keeping transitional abutments is that quadrilateral positioning of transitional tooth abutments allows for fixed provisional prostheses and uninterrupted healing of the implants.
In the final restorative phase, the integration of digital technology with conventional prosthodontic techniques allowed for a customized treatment protocol. The use of a digital impression system for a single implant restoration has been found to be more efficient than
# Discussion
In this case, the transition from failing dentition to a fixed implant-supported complete rehabilitation was accomplished in several phases: strategic extraction, virtual implant planning, computer guided surgery, and CAD/CAM prosthesis fabrication. Despite advances in the computer guided surgical technology, inaccuracies can accumulate from all stages of the workflow-from CT data acquisition, and data manipulation, to imprecision in the seating of the surgical template. Horizontal fixation pins are commonly used in computer guided implant systems treating edentulous arches. In such cases, however, the reproducible seating of the radiographic guide and surgical template can be a challenge due to the nature of mucosa-borne guided surgery. In the present case, the utilization of transitional tooth abutments allowed for reproducible and accurate positioning of the radiographic and surgical guides. Another benefit of strategically keeping transitional abutments is that quadrilateral positioning of transitional tooth abutments allows for fixed provisional prostheses and uninterrupted healing of the implants.
In the final restorative phase, the integration of digital technology with conventional prosthodontic techniques allowed for a customized treatment protocol. The use of a digital impression system for a single implant restoration has been found to be more efficient than
# Discussion
In this case, the transition from failing dentition to a fixed implant-supported complete rehabilitation was accomplished in several phases: Strategic extraction, virtual implant planning, computer guided surgery, and CAD/CAM prosthesis fabrication. Despite advances in the computer guided surgical technology, inaccuracies can accumulate from all stages of the workflow-from CT data acquisition, and data manipulation, to imprecision in the seating of the surgical template. Horizontal fixation pins are commonly used in computer guided implant systems treating edentulous arches. In such cases, however, the reproducible seating of the radiographic guide and surgical template can be a challenge due to the nature of mucosa-borne guided surgery. In the present case, the utilization of transitional tooth abutments allowed for reproducible and accurate positioning of the radiographic and surgical guides. Another benefit of strategically keeping transitional abutments is that quadrilateral positioning of transitional tooth abutments allows for fixed provisional prostheses and uninterrupted healing of the implants.
In the final restorative phase, the integration of digital technology with conventional prosthodontic techniques allowed for a customized treatment protocol. The use of a digital impression system for a single implant restoration has been found to be more efficient than conventional impression techniques. However, a conventional impression technique was utilized in this case because the accuracy and precision of intraoral digital scanning for full arch fixture level impressions has not been demonstrated. CAD/CAM technology was utilized to design the abutments and fabricate the final prostheses. Virtual superimposition and duplication of an idealized prototype facilitates predictable outcomes by improving prosthesis accuracy, ensuring passive fit, and transferring a verified occlusal scheme from the provisional stage to the final restorative phase.
With the introduction of novel digital technology into the field of dentistry, it is necessary to determine the ideal combination of a digital and conventional workflow based on the stages of treatment. In our proposed protocol, the overall workflow can be largely divided into three phases consisting of integrated conventional and digital techniques. Phase 1 is a diagnostic and preparatory phase wherein conventional techniques are used to set the foundation for future surgical and restorative interventions. The end result of this phase is an idealized provisional restoration on transitional tooth abutments. Phase 2 consists of a conventional conversion of the provisional to a radiographic stent, and a digital planning and subsequent guided placement of the implant fixtures. Finally, in phase 3, a conventional technique is utilized for the final impressions of the implants, while a predominantly digital protocol is utilized to arrive at the final prostheses. conventional impression techniques. However, a conventional impression technique was utilized in this case because the accuracy and precision of intraoral digital scanning for full arch fixture level impressions has not been demonstrated. CAD/CAM technology was utilized to design the abutments and fabricate the final prostheses. Virtual superimposition and duplication of an idealized prototype facilitates predictable outcomes by improving prosthesis accuracy, ensuring passive fit, and transferring a verified occlusal scheme from the provisional stage to the final restorative phase.
With the introduction of novel digital technology into the field of dentistry, it is necessary to determine the ideal combination of a digital and conventional workflow based on the stages of treatment. In our proposed protocol, the overall workflow can be largely divided into three phases consisting of integrated conventional and digital techniques. Phase 1 is a diagnostic and preparatory phase wherein conventional techniques are used to set the foundation for future surgical and restorative interventions. The end result of this phase is an idealized provisional restoration on transitional tooth abutments. Phase 2 consists of a conventional conversion of the provisional to a radiographic stent, and a digital planning and subsequent guided placement of the implant fixtures. Finally, in phase 3, a conventional technique is utilized for the final impressions of the implants, while a predominantly digital protocol is utilized to arrive at the final prostheses. A systematic review of implant supported fixed prostheses reported complications such as the fracture of resin teeth or frameworks, material wear, and veneer chipping or fracture. The use of monolithic zirconia eliminates the problem of veneer chipping and reduces the risk of fracture. The wear properties of CAD/CAM monolithic zirconia materials have proven to be more favorable than those of feldspathic veneering porcelain, which has been traditionally used in dentistry. However, chairside adjustments with diamond burs may introduce microfractures and roughen the surface of zirconia restorations. In the present case report, CAD/CAM technology was utilized in the duplication of an idealized acrylic prototype into the definitive zirconia restorations, which minimized chair-side adjustments of the final prostheses and allowed a direct transfer of the verified occlusal scheme developed in the provisional phase.
The overall expected prognosis of the presented treatment is considered good with patient compliance in the wearing of an occlusal guard and following a proper recall and maintenance regimen. A systematic review of implant supported fixed prostheses reported complications such as the fracture of resin teeth or frameworks, material wear, and veneer chipping or fracture. The use of monolithic zirconia eliminates the problem of veneer chipping and reduces the risk of fracture. The wear properties of CAD/CAM monolithic zirconia materials have proven to be more favorable than those of feldspathic veneering porcelain, which has been traditionally used in dentistry. However, chairside adjustments with diamond burs may introduce microfractures and roughen the surface of zirconia restorations. In the present case report, CAD/CAM technology was utilized in the duplication of an idealized acrylic prototype into the definitive zirconia restorations, which minimized chair-side adjustments of the final prostheses and allowed a direct transfer of the verified occlusal scheme developed in the provisional phase. |
Diet and Sleep Physiology: Public Health and Clinical Implications
This mini-review examines the complex relationship between diet and sleep and explores the clinical and public health implications of the current evidence. Dietary quality and intake of specific nutrients can impact regulatory hormonal pathways to alter sleep quantity and quality. Sleep, in turn, affects the intake of total energy, as well as of specific foods and nutrients, through biological and behavioral mechanisms. Initial research in this field focused primarily on the effects of short sleep duration on nutritional quality. However, more recent studies have explored the dynamic relationship between long sleep duration and diet. Current evidence suggests that extremes of sleep duration alter sleep patterns, hormonal levels, and circadian rhythms, which contribute to weight-related outcomes and obesity, and other risk factors for the development of chronic disease such as type 2 diabetes and cardiovascular disease. These patterns may begin as early as childhood and have impacts throughout the life course. Given that non-communicable diseases are among the leading causes of death globally, deeper understanding of the interactions between sleep and nutrition has implications for both public health and clinical practice.
This mini-review examines the complex relationship between diet and sleep and explores the clinical and public health implications of the current evidence. Dietary quality and intake of specific nutrients can impact regulatory hormonal pathways to alter sleep quantity and quality. Sleep, in turn, affects the intake of total energy, as well as of specific foods and nutrients, through biological and behavioral mechanisms. Initial research in this field focused primarily on the effects of short sleep duration on nutritional quality. However, more recent studies have explored the dynamic relationship between long sleep duration and diet. Current evidence suggests that extremes of sleep duration alter sleep patterns, hormonal levels, and circadian rhythms, which contribute to weight-related outcomes and obesity, and other risk factors for the development of chronic disease such as type 2 diabetes and cardiovascular disease. These patterns may begin as early as childhood and have impacts throughout the life course. Given that non-communicable diseases are among the leading causes of death globally, deeper understanding of the interactions between sleep and nutrition has implications for both public health and clinical practice.
Keywords: diet and sleep, nutrition and sleep, sleep quantity, sleep quality, sleep physiology inTRODUCTiOn In contrast to other lifestyle risk factors of chronic disease, sleep has not been accorded the same amount of attention in public health or clinical research and practice until recently. There exist complex processes linking sleep duration, quality, and behaviors to both nutrition and risk of chronic disease [fig_ref] FigURe 1 |: Conceptual framework for the interconnections between dietary factors, sleep, and disease [/fig_ref]. Currently, the evidence suggests a bidirectional relationship between sleep quality and duration and diet. These sleep components and their interactions with diet subsequently affect the risk of developing chronic disease. Here, we summarize the main evidence for these complex relationships. We also review the evidence for the effects of diet quality, sleeppromoting foods, and dietary composition on sleep outcomes. We then consider the influence of sleep quality and quantity on risk of chronic disease, such as obesity and weight-related outcomes, type 2 diabetes, and cardiovascular disease (CVD). We extend this discussion by examining the relationships between sleep and diet throughout the life course. Finally, we discuss the clinical and public health implications of this evidence and suggest possible directions for future research. Of note, this mini-review summarizes the literature on human studies, while recognizing that there are data from animal experimental studies that have explored potential mechanisms relating diet and sleep. SCienTiFiC eviDenCe
## Diet and sleep
Dietary composition, with a focus on specific dietary components, has been shown to influence sleep duration, quality, and behaviors [bib_ref] Sleep patterns, diet quality and energy balance, Chaput [/bib_ref] [bib_ref] Short sleep duration and dietary intake: epidemiologic evidence, mechanisms, and health implications, Dashti [/bib_ref] [bib_ref] Dietary nutrients associated with short and long sleep duration. Data from a..., Grandner [/bib_ref]. We cite studies analyzing the role of macronutrients, micronutrients, and whole foods on sleep [fig_ref] Table 1 |: Example of studies linking dietary components and sleep outcomes [/fig_ref] ; most have been small trials or cross-sectional studies conducted in healthy adults. For macronutrients, low fiber, high-saturated fat, and sugar intake were associated with lighter, less restorative sleep, over ad libitum diet, in an inpatient sleep crossover study [bib_ref] Fiber and saturated fat are associated with sleep arousals and slow wave..., St-Onge [/bib_ref]. In another study, a high-carbohydrate/low-fat diet was associated with poorer sleep quality vs. a normal balanced diet or a lowcarbohydrate/high-fat diet [bib_ref] Isocaloric diet changes and electroencephalographic sleep, Phillips [/bib_ref]. Protein and carbohydrate deficiencies have also been associated with shorter sleep duration (4). A study found that consuming high-glycemic index carbohydrate meals approximately 4 h before bedtime decreased sleep latency, or the time to sleep initiation, which was attributed to an increase in tryptophan after carbohydrate consumption [bib_ref] High-glycemic-index carbohydrate meals shorten sleep onset, Afaghi [/bib_ref]. Tryptophan is an amino acid precursor to the sleep-regulating hormone serotonin and is often mentioned as important in the proposed relationships between diet and sleep [bib_ref] Diet promotes sleep duration and quality, Peuhkuri [/bib_ref]. Furthermore, another study found that evening dietary increases in tryptophan intake improved sleep in adults with sleep disturbances and enhanced alertness in the morning, most likely as a result of improved sleep quality [bib_ref] Effects of tryptophan loading on human cognition, mood, and sleep, Silber [/bib_ref]. Micronutrients intake have also been suggested to affect sleep patterns. For example, associations have been reported for deficiencies in vitamin B1, folate, phosphorus, magnesium, iron, zinc, and selenium with shorter sleep duration (4), lack of alpha-carotene, selenium, and calcium with difficulty falling asleep, low intake of vitamin D and lycopene with sleep maintenance, and low intake of calcium and vitamin C with nonrestorative sleep (11). Short-term trials have shown that nightly intake of melatonin, magnesium, or zinc improved sleep quality in long-term care facility residents with insomnia [bib_ref] The effect of melatonin, magnesium, and zinc on primary insomnia in long-term..., Rondanelli [/bib_ref] , and use of vitamin D supplement resulted in better outcomes for sleep quality, sleep latency, and sleep duration in adults with a sleep disorder (13). The evidence for zinc is further supported by another randomized, double-blinded, placebo-controlled trial in healthy adults that showed that zinc-rich foods improved sleep onset latency and sleep efficiency over placebo [bib_ref] Zinc-rich oysters as well as zinc-yeast-and astaxanthin-enriched food improved sleep efficiency and..., Saito [/bib_ref]. Moreover, optimal magnesium levels, rather than levels above or below the clinically recommended range, have been found in a mice study to be required for normal sleep regulation. The exact mechanisms through which these micronutrients may
affect sleep remains unclear [bib_ref] Magnesium involvement in sleep: genetic and nutritional models, Chollet [/bib_ref]. Micronutrients have also been proposed as mediators of diet-disease associations; for example, carotenoids and vitamin D mediate associations between sleep duration and waist circumference or systolic blood pressure, and vitamin C mediates the sleep duration-diastolic blood pressure relationship [bib_ref] Contribution of inflammation, oxidative stress, and antioxidants to the relationship between sleep..., Kanagasabai [/bib_ref]. Intake of stimulant-containing foods and beverages similarly affects elements of sleep. Caffeine and theobromine are competitive antagonists to adenosine, a hormone that regulates sleep-wake cycles [bib_ref] Caffeine and adenosine, Ribeiro [/bib_ref]. While caffeine and theobromine provide immediate energy after consumption, there are also longer lasting effects consequences that alter sleep patterns for many hours after intake, including prolonged sleep latency, reduced total sleep time, sleep inefficiency, worsened perceived sleep quality, and REM sleep behavior disorder [bib_ref] Exacerbation of REM sleep behavior disorder by chocolate ingestion: a case report, Vorona [/bib_ref] [bib_ref] Coffee, caffeine, and sleep: a systematic review of epidemiological studies and randomized..., Clark [/bib_ref]. In addition, alcohol, which is often regarded as a sedative, has a nuanced impact on sleep. Consumption of alcohol decreases sleep latency and may disrupt sleep later due to its ability to influence levels of serotonin and norepinephrine [bib_ref] Sleep, sleepiness, sleep disorders and alcohol use and abuse, Roehrs [/bib_ref].
There is evidence that particular whole foods affect sleep. For example, milk, fatty fish, cherries, and kiwis have been associated with beneficial effects on sleep outcomes [bib_ref] The effect of Lactobacillus helveticus fermented milk on sleep and health perception..., Yamamura [/bib_ref] [bib_ref] Fish consumption, sleep, daily functioning, and heart rate variability, Hansen [/bib_ref] [bib_ref] Dietary fish intake and sleep quality: a population-based study, Brutto [/bib_ref] [bib_ref] Effects of a tart cherry juice beverage on the sleep of older..., Pigeon [/bib_ref] [bib_ref] Jerte valley cherry-enriched diets improve nocturnal rest and increase 6-sulfatoxymelatonin and total..., Garrido [/bib_ref] [bib_ref] Effect of kiwifruit consumption on sleep quality in adults with sleep problems, Lin [/bib_ref]. The relatively high content of tryptophan found in some of these specific foods may be responsible for these observed associations [bib_ref] Diet promotes sleep duration and quality, Peuhkuri [/bib_ref]. Intake of bread, pulses, and fish and shellfish has been positively correlated with sleep duration in men [bib_ref] Relationship between self-reported dietary nutrient intake and selfreported sleep duration among Japanese..., Komada [/bib_ref]. Finally, there is evidence to suggest that changes in daily dietary composition and eating behaviors can subsequently affect elements of sleep [bib_ref] Sleep disturbances, body fat distribution, food intake and/or energy expenditure: pathophysiological aspects, St-Onge [/bib_ref].
## Sleep and chronic disease
Growing evidence suggests that sleep patterns, such as short (<7 h) and long (>9 h) sleep duration, can impact risk of chronic disease. Short sleep has received more scrutiny, and it has been associated with an increased risk of obesity, type 2 diabetes, and CVD; it also impairs glucose metabolism, which aggravates the risk of type 2 diabetes [bib_ref] The relationships among sleep, nutrition, and obesity, Arora [/bib_ref]. Researchers hypothesize that short sleep duration may interfere with the body's restorative processes that occur during sleep, leading to biological and behavioral risk factors for chronic disease development [bib_ref] The impact of sleep disturbances on adipocyte function and lipid metabolism, Broussard [/bib_ref].
Biologically, sleep influences the circulating levels of the hunger signaling hormones ghrelin and leptin. Ghrelin indicates hunger and leptin signals satiety; sleep deprivation causes high levels of ghrelin and low levels of leptin. Therefore, the hormonal imbalance of ghrelin and leptin may induce overeating behaviors [bib_ref] The relationships among sleep, nutrition, and obesity, Arora [/bib_ref] [bib_ref] Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased..., Taheri [/bib_ref]. Furthermore, adiponectin, a secretory product of adipose tissue that is found in lower levels in the plasma of individuals with obesity, is inversely associated with sleep duration in teenage girls [bib_ref] Subjective sleep duration and quality influence diet composition and circulating adipocytokines and..., Al-Disi [/bib_ref]. Sleep deprivation also causes greater neuronal activation in response to food stimuli, which results in increased motivation to seek food with high energy intake, particularly energy-dense foods high in fat and sugar [bib_ref] Effects of diet on sleep quality, St-Onge [/bib_ref].
Another pathway by which sleep deprivation and disorders contribute to metabolic dysregulation is through activation of the hypothalamic-pituitary-adrenal (HPA) axis, which deregulates neuroendocrine parameters such as cortisol, leading to downstream increases in glucose and insulin and decreases in adiponectin levels [bib_ref] Interactions between sleep, stress, and metabolism: from physiological to pathological conditions, Hirotsu [/bib_ref]. The HPA-axis pathway, along with increased sympathetic nervous system activity and inflammatory responses, has been implicated in the relationship between short sleep duration and increased risk for hypertension, coronary heart disease, recurrent acute coronary syndrome, and heart failure.
Circadian disturbances are another possible mechanism linking short sleep behaviors (such as shift work and sleep deprivation) and dietary behaviors and intake (meals irregularity and infrequency) to weight-related outcomes [bib_ref] Chronobiological aspects of food intake and metabolism and their relevance on energy..., Ekmekcioglu [/bib_ref]. Small laboratory studies conducted with healthy adults have established the presence of intrinsic circadian rhythms regulating hunger, satiety, and food-specific appetite [bib_ref] Daily rhythms of hunger and satiety in healthy men during one week..., Sargent [/bib_ref] [bib_ref] The internal circadian clock increases hunger and appetite in the evening independent..., Scheer [/bib_ref] , as well as increases in energy expenditure with sleep deprivation or decreases with wakefulness and recovery events [bib_ref] Energy expenditure during sleep, sleep deprivation and sleep following sleep deprivation in..., Jung [/bib_ref]. Moreover, it has been reported that men have higher energy intake during sleep restriction and late-night hours, making them more susceptible to weight gain during sleep loss [bib_ref] Sex and race differences in caloric intake during sleep restriction in healthy..., Spaeth [/bib_ref].
As for behavioral mechanisms, both short sleep duration and poor sleep quality are correlated with increased energy intake, poor diet quality, and dysregulated dietary patterns, which can lead to weight gain [bib_ref] Short sleep duration and dietary intake: epidemiologic evidence, mechanisms, and health implications, Dashti [/bib_ref] [bib_ref] Dietary fat and sleep duration in Chinese men and women, Shi [/bib_ref]. Other proposed behaviors include more time and opportunities for eating, psychological distress, greater sensitivity to food reward, disinhibited eating, more energy needed to sustain extended wakefulness, and changes in appetite hormones [bib_ref] Sleep patterns, diet quality and energy balance, Chaput [/bib_ref]. Indeed, those who experience short sleep duration demonstrate more irregular eating patterns, including more frequent, smaller, and energy-dense foods during nonregular mealtimes (3).
Sleep deficiency may thus promote excess energy intake by affecting both eating behaviors and dietary composition (2). There is an established association between short sleep duration and higher total energy and fat intake (3). For example, individuals sleeping less than 7 h per night have a significantly higher proportion of total energy intake from fat, compared to those sleeping the recommended 7-9 h per night. In fact, the evidence suggests an inverse dose-response relationship between sleep and fat intake [bib_ref] Dietary fat and sleep duration in Chinese men and women, Shi [/bib_ref].
While the association between sleep and weight gain has been well studied and appears robust [bib_ref] Short sleep duration and weight gain: a systematic review, Patel [/bib_ref] , the effect on weight loss remains unclear. Randomized clinical trials have suggested that weight loss and weight maintenance through diet and lifestyle interventions can contribute to sleep improvements [bib_ref] Effect of a very low energy diet on moderate and severe obstructive..., Johansson [/bib_ref] [bib_ref] A randomized study on the effect of weight loss on obstructive sleep..., Foster [/bib_ref] [bib_ref] Weight reduction improves nocturnal respiration in obese sleep apnoea patients-A randomized controlled..., Nerfeldt [/bib_ref] [bib_ref] Lifestyle intervention with weight reduction: first-line treatment in mild obstructive sleep apnea, Tuomilehto [/bib_ref] [bib_ref] Effectiveness of lifestyle interventions on obstructive sleep apnea (OSA): systematic review and..., Araghi [/bib_ref] , although these have been mostly conducted in patients with preexisting sleep disorders. One clinical trial reported a direct association between sleep duration and successful weight loss [bib_ref] Impact of sleep, screen time, depression, and stress on weight change in..., Elder [/bib_ref]. However, further research is needed to elucidate the directionality of these associations, and whether sleep improvement could lead to weight loss in the general population.
Finally, there is emerging evidence regarding the association between long sleep duration, usually defined as more than 9 h of sleep, and chronic disease [bib_ref] Is long sleep duration a new risk factor for obesity?, Furuncuoğlu [/bib_ref] [bib_ref] Long sleep duration associated with a higher risk of increased arterial stiffness..., Tsai [/bib_ref] [bib_ref] Short and long sleep are positively associated with obesity, diabetes, hypertension, and..., Buxton [/bib_ref]. Long sleep duration has been associated with higher risk of CVD, type 2 diabetes, depression, obesity, and chronic kidney disease in trials and observational studies [bib_ref] Is long sleep duration a new risk factor for obesity?, Furuncuoğlu [/bib_ref] [bib_ref] Long sleep duration associated with a higher risk of increased arterial stiffness..., Tsai [/bib_ref] [bib_ref] Short and long sleep are positively associated with obesity, diabetes, hypertension, and..., Buxton [/bib_ref] [bib_ref] Sleep duration and cardiovascular disease: results from the National Health Interview Survey, Sabanayagam [/bib_ref] [bib_ref] Sleep duration and chronic kidney disease: the Korean genome and epidemiology study..., Choi [/bib_ref]. However, the mechanisms for these relationships are not clear [bib_ref] Is long sleep duration a new risk factor for obesity?, Furuncuoğlu [/bib_ref] [bib_ref] Long sleep duration associated with a higher risk of increased arterial stiffness..., Tsai [/bib_ref]. There may be unobservable reverse causation bias, as these conditions may cause sleep disruptions, such as sleep apnea and sleep fragmentation, that are then associated with long sleep duration. Coexisting poor sleep quality and longer sleep duration have also been shown to be associated with higher cortisol reactivity in adolescents in a cross-sectional study [bib_ref] Sleep problems predict cortisol reactivity to stress in urban adolescents, Mrug [/bib_ref] and higher incident coronary heart disease in a cohort of women [bib_ref] Sleep duration, insomnia, and coronary heart disease among postmenopausal women in the..., Sands-Lincoln [/bib_ref]. Moreover, evidence from prospective studies suggests that changes in sleep duration may increase the risk of metabolic syndrome, type 2 diabetes, and CVD mortality as well as of all-cause mortality [bib_ref] Changes in sleep duration and risk of metabolic syndrome: the Kailuan prospective..., Song [/bib_ref] [bib_ref] Short-term changes in sleep duration and risk of type 2 diabetes: Kailuan..., Song [/bib_ref] [bib_ref] A prospective study of change in sleep duration: associations with mortality in..., Ferrie [/bib_ref]. Studies have shown that workers with habitual changes in their shift rotations, and thus in their sleep patterns, have physical inactivity, overweight, sleep deprivation, increased cortisol secretion, and higher inflammation [bib_ref] Quality of life, sleep, and health of air traffic controllers with rapid..., Sonati [/bib_ref] [bib_ref] The effect of shift rotation on employee cortisol profile, sleep quality, fatigue,..., Niu [/bib_ref] [bib_ref] Shift rotation and age -interactions with sleep-wakefulness and inflammation, Viitasalo [/bib_ref]. Therefore, current evidence suggests that the relationship between sleep duration and risk of chronic disease is J-shaped, such that "extremes of sleep duration" may better predict elevated chronic disease risk [bib_ref] Addressing sleep disturbances: an opportunity to prevent cardiometabolic diseases?, Grandner [/bib_ref].
## Diet and sleep throughout the life course
The complex relationship between diet, sleep, and risk factors for chronic disease becomes evident early in life and continues throughout the life course. The impact of dietary composition on sleep patterns has been observed in early childhood. In a cohort of 1-and 2-year-old children, higher energy intake during the evening meal was associated with longer sleep duration [bib_ref] Associations between the macronutrient composition of the evening meal and average daily..., Diethelm [/bib_ref]. Studies of sleep patterns in children also highlight the impact on biological risk factors-particularly the dysregulation of ghrelin and leptin-that contribute to chronic disease risk. A longitudinal study that evaluated sleep duration and leptin found that chronic short sleep duration was associated with lower levels of leptin later in childhood and was exacerbated in girls with greater body adiposity [bib_ref] Childhood sleep duration and quality in relation to leptin concentration in two..., Boeke [/bib_ref]. There is also evidence that eating behaviors affect sleep in children. A cross-sectional study found that children who snack in-between meals or after dinner demonstrate decreased sleep duration and quality [bib_ref] Is it nutrients, food items, diet quality or eating behaviours that are..., Khan [/bib_ref]. Furthermore, decreased sleep duration in children has been associated with higher obesity risk [bib_ref] Sleep duration from infancy to adolescence: reference values and generational trends, Iglowstein [/bib_ref].
Similar to the pattern observed in young children, adolescents that report short sleep duration have elevated ghrelin and relatively low leptin levels [bib_ref] Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased..., Taheri [/bib_ref]. Adolescents are at particularly high risk for short sleep duration and currently report the highest prevalence of insufficient sleep (68.9%) [bib_ref] Prevalence of insufficient, borderline, and optimal hours of sleep among high school..., Eaton [/bib_ref]. There are several unique behavioral risk factors that may account for this finding, including increased electronic device use and unhealthy diets [bib_ref] Higher screen time is associated with overweight, poor dietary habits and physical..., Christofaro [/bib_ref] [bib_ref] Do sleep-deprived adolescents make less-healthy food choices?, Kruger [/bib_ref] [bib_ref] The association of sleep duration with adolescents' fat and carbohydrate consumption, Weiss [/bib_ref]. Nearly one-quarter of adolescents report using an electronic device "constantly, " and 72% report bringing their cellphones into their bedrooms and using them while trying to fall asleep. Increased screen time has been associated with poorer sleep quality, unhealthy eating behaviors, and decreased physical activity [bib_ref] Higher screen time is associated with overweight, poor dietary habits and physical..., Christofaro [/bib_ref]. In addition, adolescents with short sleep duration eat significantly fewer servings of fruits and vegetables and have increased odds of fast food consumption, relative to adolescents who sleep for 8 h per night [bib_ref] Do sleep-deprived adolescents make less-healthy food choices?, Kruger [/bib_ref]. Similarly, adolescents sleeping fewer than 8 h per night consumed a higher proportion of energy from fat and a lower proportion of energy from carbohydrates compared with adolescents sleeping more than 8 h [bib_ref] The association of sleep duration with adolescents' fat and carbohydrate consumption, Weiss [/bib_ref]. Compounding these risk factors, adolescents are frequently the target audience for marketing of energy-dense, nutrient-poor food, and beverage products. The combination of short sleep duration and an obesogenic environment among adolescents may contribute to developing eating behaviors and dietary choices that increase the risk of further sleep disturbances and chronic disease. In fact, sleep deprivation in adolescents has subsequently been associated with a higher risk for obesity, decreased insulin sensitivity, and hypertension [bib_ref] Short sleep duration is associated with increased obesity markers in European adolescents:..., Garaulet [/bib_ref] [bib_ref] Association of short and long sleep durations with insulin sensitivity in adolescents, Javaheri [/bib_ref].
Much of the research regarding the relationship between diet and sleep has been conducted in healthy, young or middle-aged adult populations. There is less evidence regarding the relationship between sleep and nutrition in the elderly, or in subgroups with preexisting conditions. The limited research conducted with the elderly echoes that found in younger populations. Among the elderly, poor sleep quality has been associated with obesity, hypertension, metabolic syndrome, and type 2 diabetes [bib_ref] Sleep duration, general and abdominal obesity, and weight change among the older..., López-García [/bib_ref] [bib_ref] Sleep complaints and metabolic syndrome in an elderly population: the three-city study, Akbaraly [/bib_ref] [bib_ref] Type 1 diabetes through the life span: a position statement of the..., Chiang [/bib_ref]. However, many of these studies are cross-sectional, and issues of reverse causality are especially relevant in elderly populations, who are already at elevated risk of developing these conditions.
## Implications for public health and clinical practice
According to the National Sleep Foundation, both diet quality and sleep duration are poor and have been declining steadily in the U.S. population [bib_ref] Sleep duration from infancy to adolescence: reference values and generational trends, Iglowstein [/bib_ref]. In 2010, 42% of U.S. adults and more than 50% of U.S. children reported insufficient sleep (80). Concurrently, lifestyle risk factors that may be influenced by sleep and are protective against chronic disease have been on the decline: the percent of the U.S. population adhering to five healthy lifestyle habits decreased from 15% in 1988 to 8% in 2006 [bib_ref] Adherence to healthy lifestyle habits in US adults, King [/bib_ref]. The changes in the U.S. reflect global patterns of reported sleep disturbances and shifts toward unhealthier lifestyle behaviors, especially in low-income Asian and African countries [bib_ref] Sleep problems: an emerging global epidemic? Findings from the INDEPTH WHO-SAGE study..., Stranges [/bib_ref]. These trends highlight the importance of translating the existing scientific evidence focusing on the relationship between sleep patterns and nutrition into messages, programs, and interventions that the public can easily understand and utilize to prevent chronic disease.
Nevertheless, there are few official sleep recommendations available to guide health practitioners and the general population. In the U.S., the National Sleep Foundation has published agespecific, evidence-based recommendations for sleep duration to lower the risk of chronic disease [bib_ref] National Sleep Foundation's sleep time duration recommendations: methodology and results summary, Hirshkowitz [/bib_ref]. The American Academy of Sleep Medicine has also provided pediatric and adult recommendations for sleep duration [bib_ref] Recommended amount of sleep for pediatric populations: a consensus statement of the..., Paruthi [/bib_ref] [bib_ref] Recommended amount of sleep for a healthy adult: a joint consensus statement..., Watson [/bib_ref]. The 2015 Dietary Guidelines for Americans include recommendations about physical activity and other aspects of a healthy lifestyle; however, they do not include recommendations on the integral relationship between diet and sleep. Considering the mounting literature, information on sleep should be incorporated into future iterations of the Dietary Guidelines for Americans to further enhance healthy lifestyle recommendations. Similarly, global agencies and other countries could dedicate more resources to this topic to provide population-wide recommendations on sleep and nutrition.
In addition, there should be efforts to incorporate sleeprelated content into existing interdisciplinary programs that target nutrition and other relevant aspects of health. Initial work in this area has been promising. For example, in a communitybased intervention focused on wellness, participants experienced improvements in dietary quality, sleep duration, and indicators of obesity [bib_ref] Healthy children, strong families 2: a randomized controlled trial of a healthy..., Tomayko [/bib_ref]. An intervention to prevent adverse sleep behaviors among pregnant women demonstrated a protective effect against obesity in the child's first 2 years of life [bib_ref] Adherence to healthy lifestyle habits in US adults, King [/bib_ref]. Most encouraging, these improvements have been observed in individuals from a variety of cultural and socioeconomic backgrounds [bib_ref] Healthy children, strong families 2: a randomized controlled trial of a healthy..., Tomayko [/bib_ref] [bib_ref] Targeting sleep, food, and activity in infants for obesity prevention: an RCT, Taylor [/bib_ref].
Additional strategies could help improve the approach to sleep and nutrition in the clinical setting. We recommend the following actions: (1) train and educate health-care professionals on the relationship between diet and sleep, particularly those caring for at-risk groups; (2) develop and apply rapid, validated screeners to assess diet composition, eating behaviors, and sleep patterns, to help identify and counsel at-risk patients; and (3) develop new and integrative therapies that account for the critical associations between diet and sleep.
# Future research
Despite the evidence and public health recommendations presented here, there is still a gap in understanding the complex relationship between sleep, diet, and nutrition, and risk for chronic disease. Much of the evidence to date has been done in cross-sectional studies or small trials, making it difficult to define causal pathways between various dietary components and sleep or to generalize results. More research is needed to understand the mechanisms by which specific nutrients, foods, and eating behaviors impact quality and quantity of sleep. This could be achieved through laboratory studies, larger randomized clinical trials, and longitudinal analyses of diet and sleep outcomes. Similarly, these studies can help identify potential mechanisms that mediate the relationship between sleep, diet, and risk for chronic disease. Future research should seek to strengthen the evidence linking short and long sleep duration and risk of chronic disease. Such research would inform clinical and public health recommendations regarding the specific dietary and sleep behaviors associated with the lowest risk of developing chronic conditions.
At the population level, emerging research has explored the social determinants of sleep in the U.S. A recent study found that poor sleep quality was directly related to indicators of socioeconomic status and race/ethnicity, with African-American and Hispanic/Latino populations reporting the poorest sleep quality (90). Addressing the many barriers to optimal sleep and nutrition in underserved and minority populations is crucial to improving health on a population level. Research is needed to identify feasible, culturally appropriate interventions that target sleep-and nutrition-related health gaps. Finally, given the emergence of sleep disturbances as a global epidemic [bib_ref] Sleep problems: an emerging global epidemic? Findings from the INDEPTH WHO-SAGE study..., Stranges [/bib_ref] , research in other countries, especially low-income countries, is needed. Such research could be used to guide future sleep recommendations about sleep practices and nutrition in high-risk populations across the globe.
aUTHOR COnTRibUTiOnS SF, KG, LL-A, and MY conceptualized the topic, researched and analyzed the background literature, and wrote the manuscript, including interpretations and conclusions. JM analyzed the content to prepare the table and portions of the figure and manuscript. MT and JM provided substantial scholarly guidance on the conception of the topic, manuscript draft and interpretation, and revised the manuscript critically for intellectual content. All the authors approved the final version of the manuscript, ensured the accuracy and integrity of the work, and agreed to be accountable for all aspects of the work.
## References
[fig] FigURe 1 |: Conceptual framework for the interconnections between dietary factors, sleep, and disease. The complex, bidirectional, relationship between sleep components, dietary composition, behavioral factors, and biological factors are theorized to affect the development of chronic diseases and other health outcomes. Intake of nutrients and foods and dietary behaviors are correlated with components of sleep quality and quantity. Dietary components and eating behaviors are regulated by biological factors, which in turn may impact nutritional status. Similarly, sleep quantity and quality result in biological factors being modulated, and these biological factors control sleep factors in return. Dietary components, sleep components, and biological factors have subsequent independent as well as interactive effects on downstream health outcomes. Notably, this multifaceted interaction is evident early in life and continues throughout the life course. [/fig]
[fig] FUnDing: KG was support by a Harvard ERC/NIOSH training grant (grant number T42 OH008416). MT was supported by the National Council of Science and Technology (CONACyT, Mexico). JM was supported by a NIH-NHLBI Mentored Career Development Award to Promote Faculty Diversity in Biomedical Research (grant number K01-HL120951). [/fig]
[table] Table 1 |: Example of studies linking dietary components and sleep outcomes. [/table]
|
Prevalence and Clinical Characteristics of Mycobacterial Diseases in the Barletta-Andria-Trani Province, Italy (2005–2013)
Tuberculosis remains one of the major worldwide problems regarding public health. This study evaluates the burden of this disease in the BAT Province of the Apulia region (Italy); 12,295 patients were studied, including 310 immigrants. Tubercular disease and mycobacteriosis were found in 129 patients. The number of new TB cases/year ranged from three in 2005 to 12 in 2009. TB was more frequently localized in the lung (70.5%). 14.4% of cases were institutionalized patients for severe neurological and/or psychiatric disease. The database evidenced certain aspects of our study population: the large number of TB patients institutionalized between natives, but no larger presence of TB among HIV-positive patients in immigrants compared to Italians. Our findings should help to redefine the alarm regarding the spread of an epidemical form of TB but also to present certain criticisms regarding patient management (especially immigrants) regarding costs, hospitalization, and difficulty of reinstating the patient in the community. Further our data underscore the importance of prevalence of TB in bedridden, institutionalized patients.
# Introduction
In spite of the fact that it is a preventable and curable disease, TB represents one of the major global health problems [1]; in fact in the 2013 Global Tuberculosis Report, more than 8.5 million cases were identified, 1,450 million of which were fatal, thus indicating that TB is the most important infectious disease and cause of death after HIV. In certain countries of Southern Africa (South African Republic, Lesotho, and Swaziland), the incidence of TB is greater than 1000 cases/100,000 inhabitants. In 2006, the World Health Organization (WHO) initiated the "Stop TB Strategy" with the objective of reducing, before 2015, both the prevalence and 50% mortality rate reported in the 1990 data and the global incidence of active TB, aiming to reach a goal of 1 case/100,000 million inhabitants by 2050 [bib_ref] Global tuberculosis incidence and mortality during, Dolin [/bib_ref].
In Italy, TB is a relatively rare pathology and the 1998 incidence has been constantly less than 10 cases/100,000 inhabitants (considered by WHO as a low prevalence), which actually has even been reduced to less than 5 cases/100,000 . However, beginning in the middle 1980s, an increase in TB has been reported, as in all the Western countries, due to the increase in risk categories (immunodepressed subjects, alcoholics, indigents, institutionalized patients, or those subjected to chemotherapy) and an increase of resistant strains to pharmacological therapies [bib_ref] Risk assessment of tuberculosis in immunocompromised patients. A TBNET study, Sester [/bib_ref] [bib_ref] Tuberculosis from head to toe, Harisinghani [/bib_ref]. In particular, the number of TB cases among immigrants has currently reached that of native subjects, especially in large cities .
Although most of the migrant population in Italy is concentrated in the north (about 60%), the increase of foreigners during 2011 was greater in the islands (Sardinia +13.7%) and Southern Italy (Puglia +13.5%)than in the previous year. The most numerous foreigners are from Eastern Europe (Albania, Bulgaria, Ukraine, and Poland) and the Indian subcontinent (India, Bangladesh, and Pakistan), countries in which the epidemiological situation is complicated by the diffusion of mycobacteria resistant to at least two firstline antitubercular drugs (multidrug-resistant-TB, MDR-TB) or injected second-line drugs (extensively drug-resistant-TB, XDR-TB).
In this study, we have attempted to quantify the presence of TB in the Italian BAT Province (Barletta-Andria-Trani) which has a population of about 400,000 inhabitants with a medium yearly increase of 0.1-0.3%. In recent years, a gradual yearly increase of 0.9% to 2.1% [9] in migrants from 2005 to 2010, respectively, has been observed; in fact, the number of non-Italian residents has more than doubled in 5 years (from about 3,600 to 8,440) compared to a steady decrease in the Italian population [10]. On Dec. 31, 2013, a total of 8,700 immigrants with regular permit to stay were present in the BAT territory (50% males) [9], equal to 2.1% of the entire population. Moreover, the economy of this province is based on agriculture, and it is reasonable to imagine the same number of migrants who were without regular documents (500 according to reliable estimates) was working in the fields.
The BAT Province includes 10 towns, three hospitals, and one Department of Infectious Disease in the Bisceglie Hospital. It is important to emphasize that Bisceglie is also the seat of a large private hospital (House of Divine Providence) with about 1,500 beds subdivided into an Alzheimer Unit, Rehabilitation Unit (Neurological and Cardiorespiratory), residence for cancer patients, and Psychiatric Clinic (in addition to Clinics for Acute, Neurological, Cardiac and Lung Patients). This Orthophrenic Institute (ex. Psychiatric Hospital of Bisceglie) which in the past provided 2,000 beds for institutionalized patients (from all of Italy) with cognitivemotorial problems either congenital or arising in childhood now includes 700 beds.
Infectious disease clinics have also become important observers of most TB cases notified for a series of reasons (possibility of isolation, association with HIV and immunosuppression). The Clinic of Infectious Diseases of the Bisceglie Hospital is one of the largest units providing the largest number of beds throughout Italy.
# Materials and methods
Our study includes all hospitalized patients at the Unit of Infectious Diseases of the Bisceglie Hospital between Jan. 1, 2005, and Dec. 31, 2013, subjected to a retrospective analysis. The database includes demographic (age, sex, residence, and nationality) and clinical (diagnosis at admission and dismissal, length of hospitalization, and mode of dismissal) data in addition to fiscal code, temporary presence code, or nonregistered Europeancode. Regarding dismissal diagnosis, the database also provides the principal and secondary diagnoses (up to five) reported in the hospital dismissal form (SDO).
Italy legally requires all citizens to be inscribed in the National Health System or to be classified as a foreigner temporarily present or nonregistered European, thus guaranteeing health assistance also for foreigners without a regular residence permit. This is a constitutional privilege and the legislative degree number 286 of July 25, 1998, regards regulations for foreigners.
Selection of cases was based on the presence of a TB diagnosis (and/or mycobacteriosis) according to the "International Classification of Diseases-IX-Clinical Modification" (ICD-IX-CM). In the case of mycobacteriosis disease, the code defined the site of disease (between 0010 and 0018 for TB and 031.0 and 031.9 for nontubercular mycobacteriosis, NTM).
# Statistical analysis.
Data were processed using the software package STATA 11.0. 2 -test was used to compare proportions. For small samples, Fisher's exact test was used.
-test was used to compare means. If a value was below 0.05, the difference between proportions was considered to be statistically significant. The nosological code referring to TB and mycobacteriosis was found 159 times, that is, 124 times as the principle diagnosis and [bib_ref] Tuberculosis from head to toe, Harisinghani [/bib_ref] , and 0 times for the five fields of secondary diagnosis in the dismissal form (SDO). TB was diagnosed in 149 patients (93.7%) and the remaining 10 cases were NTM. In thirty cases the same patients were admitted several times in our unit and, therefore, these cases were not considered.
# Results
Of the 129 TB cases, 85 were Italians and 44 were immigrants. The median age of TB patients was 48.3 years (range 16-88) and 70% were males. The average age of immigrant TB patients was 36 years (standard deviation ±12.9, range 18-82) and was significantly reduced ( < 0.0001) compared to that of the TB Italian patients (51 years, SD ±19.01). The 66% of immigrant patients with TB were males (29/44). The Italian TB patients residing in the BAT Province were 65/85 (76%), in particular Bisceglie had 33, Barletta 11, Andria 10, Trani 8, Canosa 2, and Trinitapoli 1, while those living in the other provinces included Bari (13 patients, 15.2%), Foggia (4 patients, 4.7%), and Lecce, Brindisi, and Rome (1 patient each).
The migrants with TB living in the BAT Province were 25/44 (56%), in particular Andria had 10, Bisceglie, Barletta, and Trani 4 each, Canosa 2, and San Ferdinando di Puglia 1, followed by 1 each for the Province of Foggia (9/44), Bari (8/44), Brindisi, and Lecce. [fig_ref] Table 1: Epidemiological and clinical differences between native and immigrants patients with TBC [/fig_ref] shows the comparison [fig_ref] Figure 1: Patients admitted with tuberculosis in the Infectious Disease Unit of Bisceglie [/fig_ref]. The prevalence of disease in the entire BAT Province was of a median of 1.8/100,000 inhabitants/year (range 0.75 to 3/100,000 inhabitants/year). Regarding TB clinical characteristics, the most common site was the lung (91/129, 70.5% of cases) followed by lymphoglandular (8 cases), disseminated, and bone (4) sites. Also in immigrants, the most frequent site was the lungs (38/44, 86%), followed by the disseminated, bone (2), and lymphoglandular (1) forms; lastly, one case of pulmonary mycobacteriosis was found in a Chinese patient. Among the TB patients, 20% presented a comorbidity involving the respiratory system (chronic bronchial obstruction, pulmonary fibrosis, and pulmonary hypertension), 13.8% regarded the cardiovascular system, and 7.5% regarded the gastrointestinal apparatus. A total of 8.5% of patients were HIV positive (11 patients, only 3 immigrants) while 14.4% (22 Italian patients) were hospitalized due to serious neurological and/or psychiatric problems (mental retardation, schizophrenia, results of cerebral ictus). [fig_ref] Figure 3: Comparison of comorbidities [/fig_ref] presents the comorbidities of the Italian and immigrant TB patients in a more detailed manner. The Italian TB patients were hospitalized for a median of 29 days (range 1-213); those dismissed in a suitably improved condition were 105/129 (61.4%) while 3.1% were transferred to a ward for acute patients and 9.3% (12/129) voluntarily left the hospital against the advice of the physician, 3/12 of them within 14 days after recovery; 6% of cases died, equal to 0.22/100.000 inhabitants/year. The length of recovery among immigrants ranged from 4 to 120 days with a median of 29 for legal and 31 days for illegally present immigrants. Those dismissed in a better state of health were 79.5% (35/44) while 4.5% were transferred to other wards and 13.6% of patients (6/44) left the hospital against the advice of the physician.
# Discussion
Our data confirm certain characteristics previously reported in the literature (prevalence of TB localized in the lung in males and in young immigrants) [bib_ref] Molecular epidemiology of tuberculosis among immigrants in, Diel [/bib_ref] [bib_ref] Tuberculosis surveillance in Sicily, Italy, Mammina [/bib_ref] which will not be repeated here. The total number of cases demonstrates that, even today, tuberculosis is a disease present in Italy, even if not frequently. If we consider that the prevalence curve of this disease began to increase in Western countries over a period of hundred years, it then seemed to return to zero. However, new cases of TB then reappeared again often due to an incorrect management of health services (e.g., in the 1980s it was not only AIDS to determine this increase in TB, but also the lack of prevention campaigns and elimination of public health structures [bib_ref] Tuberculosis in New York city: recent lessons and a look ahead, Paolo [/bib_ref]. In the BAT Province, 1.2% of all recoveries in the Infectious Disease Unit regard mycobacteria. This is apparently a small number with a tendency, however, to increase due to both the migratory phenomenon (as in all provinces with a large presence of immigrants) and the increase of pharmacologically induced immunodeficiencies. Considering that the population of our province reaches 400,000, our data indicates a TB prevalence of 1.8 cases/100,000 inhabitants, in accordance with data previously reported for the entire country. At present, in Italy, the cases of TB among immigrants have continually increased; in 1995, 525 cases were noted (10% of total amount) which expanded to 2,026 cases in 2008 (46% of total) [9, 10]. This variation appears to be due to the numerical increase of the foreign population (from 700,000 in 2002 to more than 5 million in 2013). Therefore, the TB epidemic appears to be changing even in our relatively small province and is beginning to assume the characteristics described on the national level.
The inversion evidenced in 2012 regarding the increase in number of TB diagnoses in immigrants compared to that found for Italians can be explained by the fact that from the year of establishing this province (2005) no increase in Italian residents has been noted while there was a threefold increase in the migrant population; in fact, Puglia has become the region with the highest increase (15%) between 2012 and 2013.
When considering comorbidity, certain interesting characteristics also appeared. The 8.8% prevalence of AIDS found in our cases was comparable to that of Europeanand Italian [17] data even if, surprisingly, it was found to be less among immigrants who also demonstrated a lesser dependence on alcohol and presence of neuropsychiatric problems. It must be remembered that psychic disturbances (also posttraumatic problems related to stress) are well known events which are quite important according to the ISTAT annual reports (10% of foreigners) . Only disturbances of respiratory (65%), gastrointestinal (20%), osteomuscular (17%), and nervous (20%) systems present greater percentages, while only 2.6% of migrants were also affected by infectious diseases. These results can be partially explained both by the fact that migration to the BAT Province is a relatively recent phenomenon and by the fact that diagnosis of HIV/AIDS in migrants (as in the rest of the country) is generally delayed, resulting in higher frequency of late presenters in this group [bib_ref] Late presentation and loss to follow-up of immigrants newly diagnosed with HIV..., Saracino [/bib_ref] [bib_ref] Late presenters among persons with a new HIV diagnosis in Italy, Camoni [/bib_ref].
The greater prevalence of patients with serious disabilities and those who were hospitalized merits a closer analysis. Our study included 22 hospitalized patients (all Italian), eight of them with serious mental retardation, five with epilepsy and five with Down syndrome, three with schizophrenia or serious psychoses, two with damage from cerebral ictus, and one with congenital tetraplegia.
It is well known that hospitalization presents a risk factor for TB and the Ministry Guidelines indicate that all elderly subjects living in rest homes or submitted to lengthy recoveries are classified as risk categories. However, our study emphasizes that the young age of our patients recovered at Bisceglie Hospital was most likely due to the fact that these patients had been hospitalized, even at a young age, in the Divine Providence hospital (median age 62,5 years; 12/22 were younger than 65), and the prolonged presence in this hospital was the risk factor for them.
The length of hospitalization (a median of 29 days for TB) was much longer compared to the median of other hospital recoveries in the same department (9 days). However, due to the number of voluntary dismissals, especially among immigrants, the length of time was probably less. In fact, the cases of patient dismissal within the first 15 days against the doctor's opinion were 5.6%, a high percentage when considering that these are all contagious patients. Moreover, the length of hospitalization also has repercussions on the cost as TB patients require up to three times that of most patients (up to a maximum of 20-25 times) and the length of isolation must also be taken into consideration. The medium length of recovery, which may extend up to a maximum of 213 days, not only reflects the natural history of the disease, but also is due to the caution of the physician when the patient lacks familial and social assistance during the induction and maintenance phase of therapy at home. This data is important for indigents, homeless immigrants, and institutionalized patients (for whom the importance of directly observed therapy and entrance in a secure residence is easily understood); these data determine a boomerang effect (for first-aid emergency workers and those in other hospital wards) not only regarding health costs but also due to the difficulty to find beds "in isolation" when a suspected case is discovered.
The Guidelines proposed by the Ministry of Health state that "the TB patient is often hospitalized without motivation and for longer periods of time than absolutely necessary. . ." and that ". . . hospitalization is indicated in case of: extensive disease or tubercular meningitis, compromised clinical conditions, immune compromised patients, positive bacteriological expectorants when a safe isolated home is not available, and presence or suspected polyresistence [sic] . . .". However, often these indications clash with daily life and it is impossible for the physician to be sure about proper isolation from those with whom they mix (the homeless, indigents, and alcoholics).
# Conclusions
(1) Our data should help to place the alarm regarding the spread of TB in the mass media into its proper prospective indicating several crucial aspects concerning the access and use of health services by legal and illegal immigrants and how these factors determine correct patient management. (2) The promotion of research programs would be extremely useful to identify symptomatic TB cases, not only in countries with a high incidence of disease as previously stated [bib_ref] Access to health services for undocumented immigrants in Apulia, Brindicci [/bib_ref] and indicated by WHO. Further research is needed to also identify those categories of patients at risk and rarely encountered, thus reaching an early diagnosis in a higher percentage of patients and providing an improvement in health for the general population and a reduction of costs for the national health system.
[fig] Figure 1: Patients admitted with tuberculosis in the Infectious Disease Unit of Bisceglie (2005-2013). [/fig]
[fig] Figure 2: Countries of origin of the 44 immigrants with TBC. [/fig]
[fig] Figure 3: Comparison of comorbidities (%) between natives and immigrants. [/fig]
[table] Table 1: Epidemiological and clinical differences between native and immigrants patients with TBC. [/table]
|
The interaction network of the YidC insertase with the SecYEG translocon, SRP and the SRP receptor FtsY
## Supplementary
[fig] Figure S1: YidC makes multiple contacts to SecY. The in vivo photo-cross-linking material produced forfigure 2a, 2b and 2cwas analysed on western blot with α-SecY antibodies. [/fig]
[fig] Figure S2: YidC makes multiple contacts to SecE and SecG. In vivo photo-crosslinking was performed with the co-expression system using YidC variants which contained pBpa at the indicated positions. Cross-linked material was subsequently purified via the His-tag on SecY. The analysis of the purified material on western blot with SecE antibodies (a and b) or SecG antibodies (c) revealed YidC-SecE and YidC-SecG cross-links migrating at about 65kDa. * indicates a possible proteolysis product of the YidC-SecG cross-link. Supplementary Figure S3: TM1 is not required for the interaction with SecY. (a) Expression [/fig]
[fig] SupplementaryFigure S4: Uncropped images of all western blots and LB plates used in this study. Labelling of the panels corresponds to the figure labeling in the manuscript and the supplementary information. The numbers in the right margin of each panel corresponds to the molecular weight marker.Supplementary [/fig]
[table] Table S1: Oligonucleotides used in this study [/table]
|
Pediatric Acute Severe Hepatitis of Unknown Origin: What is New?
Globally, there are emerging cases of acute severe hepatitis of unknown origin in children. These cases have gathered increasing attention, owing to the development of acute liver failure in some cases that resulted in liver transplantation. This review briefly summarizes the outbreak and diagnostic criteria of the disease. We further discuss the possible causes and related mechanisms underlying its occurrence and progression, and analyze the challenges in management. Finally, this review emphasizes patient management in clinical settings and a combination of efforts to unmask the disease.
On March 31, 2022, five children (aged 3-5 years) in Scotland were the first to be diagnosed with acute severe hepatitis without an identified origin. As of April 5, more than 10 cases of acute severe hepatitis in children under 10 years of age were retrospectively reported in Scotland, with the earliest being January 1, 2022. These emerging cases raised concerns since they outnumbered the annual cases reported in this region of the UK. [bib_ref] Investigation into cases of hepatitis of unknown aetiology among young children, Marsh [/bib_ref] Globally, reports of children with acute hepatitis continue to increase [fig_ref] Figure 1: Timeline of major events in acute severe hepatitis children with unknown origin... [/fig_ref]. As per the latest reports from the World Health Organization (WHO) and European Centre for Disease Prevention and Control (ECDC) on May 18, 2,3 the number of cases worldwide has increased to 576 from 28 countries. Among these countries, the UK reported 176 cases and the USA reported 180 cases. In Asia, cases have been reported from Indonesia (21 cases), Israel (12 cases), and Japan . The distribution of reported cases is shown in [fig_ref] Figure 2: Distribution of reported cases with acute severe hepatitis of unknown origin worldwide... [/fig_ref].
## Clinical manifestation and laboratory testing
The clinical characteristics of the reported cases mainly include (1) age at presentation (of 1 month to 16 years), with most affected children under the age of 10 years; (2) a significant elevation of serum alanine aminotransferase (ALT) or aspartate transaminase (AST) >500 U/L; and (3) symptoms including jaundice, nausea, abdominal pain, fatigue, lethargy, and gastrointestinal manifestations, such as diarrhea and vomiting. Fever was reported less frequently; in particular, it was reported in Alabama (5/9 cases) and England (24/81) but not in Scotland. The most important laboratory finding was the absence of known hepatitis viruses (hepatitis A, B, C, D, and E viruses) in any of the reported cases. Preliminary epidemiological data have shown that there is no association between the disease and the use of drugs, exposure to toxins, COVID-19 vaccination, nor environmental factors.On April 23, 2022, the WHO issued a communiqué statement that adenovirus infection was detected in at least 74 cases. Among samples from 18 cases analyzed through adenovirus genotyping, adenovirus type 41 was identified. Three separate studies have reported the rate of adenovirus infection, as follows: Alabama (100%, 9/9), England (75.5%, 40/53), and Scotland (38.5%, 5/13) [fig_ref] Table 1: Summary of acute severe hepatitis children with unknown origin from the USA,... [/fig_ref]. [bib_ref] Investigation into cases of hepatitis of unknown aetiology among young children, Marsh [/bib_ref] [bib_ref] Acute Hepatitis and Adenovirus Infection Among Children -Alabama, Baker [/bib_ref] In addition to adenovirus, systemic acute respiratory syndrome (SARS)-coronavirus-2 (CoV-2) monoinfection was detected in 20 cases via real-time PCR testing. Notably, coinfection of adenovirus and SARS-CoV-2 was detected in 19 cases. Among the nine patients in the USA, two recovered after liver transplantation and seven recovered without liver transplantation; there were no reports of mortality. Among the 43 patients in England, 3 had a full recovery after liver transplantation. By , at least 11 children died outside England, including five children in Indonesia, 5 in the Americas, and 1 in Palestine.The latest data jointly released by the WHO and ECDC show that the proportion of diagnosed children with acute liver failure who required liver transplantation was approximately 6%.In general, epidemiological features revealed that most patients were not vaccinated against SARS-CoV-2. Except for 2 of the 13 patients in Scotland, who had close contact with the other two affected children, 1 an obvious epidemiologic association was lacking in most cases reported in the Americas and England.
## Differential diagnosis and treatment strategies
Differential diagnosis should be accomplished with the exclusion of other diseases, including (1) hepatophilic viral infections, such as hepatitis A, B, C, D, and E viruses;
(2) non-hepatitis infections, such as those caused by rhinovirus, Epstein-Barr virus, cytomegalovirus, human herpes viruses, microvirus B19, and human immunodeficiency virus-1 (HIV-1); (3) non-viral pathogens that cause acute hepatitis, such as meningococcus and leptospira; (4) drugs or toxins, such as azithromycin and erythromycin; (5) immune-mediated liver injury, such as autoimmune liver diseases, hemophagocytic lymph histiocytosis, and gestational alloimmune liver disease; and (6) inherited metabolic liver diseases, such as hepatolenticular degeneration in children older than 5 years and in adolescents.
Children with acute severe hepatitis are at risk of disease exacerbation. Currently, there is a lack of consensus on the treatment regimen for this disease. Dynamically monitor-ing the biochemical parameters in the blood and clinically assessing the condition of affected children can help physicians assess disease progression promptly to modify the therapeutic strategies as needed.
Among the current treatment strategies, symptomatic therapy, such as the use of hepatoprotective drugs, is experience-based in clinic and needs to be further studied in the acute severe hepatitis with unknown origin. With this approach, in particular, physicians should be cautious when treating severe cases at risk of severe complications, such as hepatic encephalopathy, hepatorenal syndrome, secondary infection, and sepsis. Antiviral therapy should be initiated to control viral replication if the virus is identified. Liver transplantation is recommended in severe cases of acute liver failure. Whether artificial liver supportive therapy can be administered to children requires further study. If immune hyperactivation is involved in disease progression, immunomodulatory therapy, such as glucocorticoid therapy, is worth considering and may be beneficial in reducing liver injury in children.
## Disease etiology: induced by pathogens or non-infectious factors?
The primary priority was to identify the cause of acute se-vere hepatitis in children. Although hepatitis A, B, C, D, and E viruses were excluded, other pathogens or non-infectious factors may have contributed to or resulted in the observed hepatitis cases.
Since adenovirus was most frequently detected in these patients, it is considered a suspected pathogen. Three studies from Scotland, UK and Alabama, USA reported adenovirus positivity rates in children ranging from 39% to 100%. 1,4,5 Furthermore, the two cohorts of children (severe acute hepatitis of unknown origin or adenovirus infection) had a similar age range. [bib_ref] Molecular Study of Adenovirus Genotypes 40 and 41 in Children with Acute..., Essa [/bib_ref] In addition, the England-based patients who received liver transplantation had approximately 12-fold adenovirus DNA levels than those who did not receive liver transplantation.However, more robust data are still needed to support this notion. Although more than 25% of children tested positive for adenovirus infection in their respiratory, serum, or stool samples, according to a report from the ECDC,there are some concerns with this hypothesis. Adenovirus positivity in throat swab samples was found in 11% of healthy children. [bib_ref] Diagnosis of Pediatric Acute Adenovirus Infections: Is a Positive PCR Sufficient?, Song [/bib_ref] [bib_ref] Prevalence of rotavirus, adenovirus and enteric parasites among pediatric patients attending Saint..., Djeneba [/bib_ref] [bib_ref] Adenovirus as an emerging pathogen in immunocompromised patients, Leen [/bib_ref] However, acute severe hepatitis induced by adenovirus infection is rarely observed in healthy children. The gold standard method for the diagnosis of adenoviral hepatitis is the detection of adenovirus inclusion bodies in liver biopsy samples. [bib_ref] Adenovirus Hepatitis: Clinicopathologic Analysis of 12 Consecutive Cases From a Single Institution, Schaberg [/bib_ref] However, adenovirus inclusion bodies were not observed in the current study. In addition, a high serum adenoviral load was absent in the reported cases. Furthermore, it is documented that ADV41 infection mainly causes gastrointestinal symptoms, such as diarrhea, and rarely causes acute hepatitis.
In clinical settings, 72% of patients with adenoviral hepatitis were mainly infected with strains ADV2 and ADV5. Thus, whether adenovirus infection is the cause of the current disease needs to be further clarified. SARS-CoV-2 is also considered a suspected (contagious pathogen) contributor. Among the 12 cases reported in Israel, 11 had a history of SARS-CoV-2 infection.Studies have shown that viral RNA of SARS-CoV-2 remains in the gastrointestinal tract of affected children longer than in the respiratory tract. [bib_ref] Persistence of SARS-CoV-2 virus RNA in feces: A case series of children, Du [/bib_ref] [bib_ref] Prolonged viral shedding in feces of pediatric patients with coronavirus disease 2019, Xing [/bib_ref] Therefore, the superantigen motif in the spike protein of SARS-CoV-2, structurally similar to staphylococcal enterotoxin B, could trigger broad activation of non-specific T cells, which may lead to a multisystem inflammatory syndrome in children. [bib_ref] COVID-19-associated multisystem inflammatory syndrome in children (MIS-C): A novel disease that mimics..., Rivas [/bib_ref] [bib_ref] Super(antigen) target for SARS-CoV-2, Brown [/bib_ref] [bib_ref] Immune dysregulation and autoreactivity correlate with disease severity in SARS-CoV-2-associated multisystem inflammatory..., Ramaswamy [/bib_ref] [bib_ref] Immunopathological signatures in multisystem inflammatory syndrome in children and pediatric COVID-19, Sacco [/bib_ref] Alternatively, similar to patients with HIV-1, 19 the children previously infected by SARS-CoV-2 may have a repetitive immune activation caused by the comparatively long-term existence of SARS-CoV-2 in the gastrointestinal tract. [bib_ref] SARS-CoV-2 infections in children: Understanding diverse outcomes, Brodin [/bib_ref] Under these conditions, children may be prone to infections by other viruses, which would contribute to the development of acute hepatitis. Previous animal studies reported that adenovirus infection significantly enhanced T cell activation mediated by superantigens. This increased the risk of toxic shock with liver damage. [bib_ref] Severe acute hepatitis in children: investigate SARS-CoV-2 superantigens, Brodin [/bib_ref] [bib_ref] Increased sensitivity to staphylococcal enterotoxin B following adenoviral infection, Yarovinsky [/bib_ref] Thus, continuous monitoring of SARS-CoV-2 superantigen is strongly recommended for affected children to clarify whether SARS-CoV-2 superantigen, together with infection with adenovirus or other viruses, caused the disease.
In addition to the above-mentioned viral infection, bacterial and/or fungal infection may induce hepatitis, particularly under conditions of systematic infection. However, there is no clinical evidence to date that supports this. Other noninfectious factors may be responsible for the incidence or contribute to the progression of acute severe hepatitis of unknown etiology, including genetic susceptibility, unknown toxin exposure, hepatotoxic drugs, or environmental exposure factors. Unlike highly transmissible viral infections, these factors are not contagious and cannot induce a global pandemic. Thus, there are no concerns regarding biosafety issues.
## Challenges and perspectives
During the last several months, there have been increasing reports not only from the literature, WHO, and Centers for Disease Control (CDCs) of different countries worldwide but also from social/public media and many We-Media, which may overstate the risk of the disease and its potential for a global pandemic. Therefore, hepatologists should pay attention to the potential biosafety issues involved in this disease.
Some challenges must be addressed immediately. It is crucial to document unmet issues of acute severe hepatitis of unknown etiologies in children. Further epidemiological studies and whole-genome sequencing of biological samples may help identify the pathogen responsible for the disease. Close collaboration between physicians and scientists is critical to conduct basic, translational, and clinical studies to highlight epidemiology. Rapid diagnosis in potential cases is necessary for better disease management. Furthermore, it is necessary to clarify viral, immunological, and pathological mechanisms underlying acute severe hepatitis. In summary, complete control of the disease may require significant public health and medical efforts.
It may be helpful to consider the current progress in the management of adults with acute severe hepatitis when taking care of affected children. Although there are few reports, acute severe hepatitis of unknown origin was previously investigated in adult patients. Cytopenia is one of the most frequently observed clinical features. Reportedly, 49% of adult patients required liver transplantation, and 24% of them died from acute liver failure. [bib_ref] Acute severe hepatitis in adult-onset Still's disease: case report and comprehensive review..., Muller [/bib_ref] Steroid treatment may be effective for some patients. The characteristics of pediatric patients are distinct from adult patients, suggesting that clinical experience drawn from adult patients can only be used as a general reference for pediatric patients. There is an urgent need to develop effective treatment strategies for pediatric patients.
As recently recommended by the WHO, acute severe hepatitis of unknown origin in children should be actively monitored and extensively studied worldwide. Therefore, it is also necessary to investigate and assess the risk of the disease and disease progression. Moreover, routine preventive measures, such as careful hand-washing and maskwearing, should be advocated for all children and their caregivers to avoid potential pathogenic infections. Furthermore, professional training and awareness among pediatricians and other clinicians along with development of disease management guidelines are crucial to improve the diagnosis and treatment of acute severe hepatitis of unknown origin in children.
# Funding
Innovation Groups of the National Natural Science Foundation of China (No. 81721002), China National Natural Science Foundation (No. 82130019)
## Conflict of interest
FSW has been an executive associate editor of Journal of Clinical and Translational Hepatology since 2022. The other authors have no conflict of interests related to this publication.
# Author contributions
Study concept and design (FSW, JL and JYZ), literature search and drafting of the manuscript (JL, WH and JY), critical revision of manuscript for important intellectual content, senior authorship guidance and supervision (FSW). All authors agreed with the final version of the manuscript.
[fig] Figure 1: Timeline of major events in acute severe hepatitis children with unknown origin from March 31 to May 18, 2022. Journal of Clinical and Translational Hepatology 2022 vol. 10(3) | 509-514 511 Li J. et al: Acute severe hepatitis in children [/fig]
[fig] Figure 2: Distribution of reported cases with acute severe hepatitis of unknown origin worldwide until May 18, 2022. (A) Number of reported cases up to 18 th May, 2022; (B) Distribution of diagnosed cases worldwide. [/fig]
[table] Table 1: Summary of acute severe hepatitis children with unknown origin from the USA, England and Scotland respectivelyFor England cases: outcome at 28 days after presentation. *Including one patient who successfully underwent liver transplantation but remains in hospital. [/table]
|
Complementary Treatment of the Common Cold and Flu with Medicinal Plants – Results from Two Samples of Pharmacy Customers in Estonia
The aim of the current survey was to investigate the complementary self-treatment of the common cold and flu with medicinal plants among pharmacy customers in Estonia. A multiple-choice questionnaire listing 10 plants and posing questions on the perceived characteristics of cold and flu, the effectiveness of plants, help-seeking behaviour, self-treatment and sources of information, was distributed to a sample of participants in two medium size pharmacies. The participants were pharmacy customers: 150 in Tallinn (mostly Russian speaking) and 150 in Kuressaare (mostly Estonian speaking). The mean number of plants used by participants was 4.1. Of the respondents, 69% self-treated the common cold and flu and 28% consulted with a general practitioner. In general, medicinal plants were considered effective in the treatment of the above-mentioned illnesses and 56% of the respondents had used exclusively medicinal plants or their combination with OTC medicines and other means of folk medicine for treatment. The use of medicinal plants increased with age and was more frequent among female than male respondents. Among Estonian-speaking customers lime flowers, blackcurrant and camomile were more frequently used, and among Russian speaking customers raspberry and lemon fruits. Regardless of some statistically significant differences in preferred species among different age, education, sex and nationality groups, the general attitude towards medicinal plants for self-treatment of the common cold and flu in Estonia was very favourable.
# Introduction
The common cold is one of the most frequent minor illnesses in the world. Caused by 200 identified types of viruses, the common cold is primarily associated with rhinoviruses [bib_ref] Rhinovirus infections: more than a common cold, Hershenson [/bib_ref]. Common cold symptoms may include nasal congestion, an acute cough and/or sore throat [bib_ref] Cough and the common cold: ACCP evidence-based clinical practice guidelines, Pratter [/bib_ref]. The common cold is distinct from flu, associated with the influenza virus. A fever greater than 38uC and generalized aches and pains are the best predictors of a diagnosis of flu [bib_ref] The pathology of influenza virus infections, Taubenberger [/bib_ref]. In the popular approach, both the described illnesses are perceived as similar conditions [bib_ref] Clinical diagnosis of influenza virus infection: evaluation of diagnostic tools in general..., Van Elden [/bib_ref]. The common cold and flu are predominantly self-diagnosed and self-medicated with OTC medicines and/or herbal products [bib_ref] Understanding the symptoms of the common cold and influenza, Eccles [/bib_ref] [bib_ref] Herbs and Other Dietary Supplements: Current Regulations and Recommendations for Use to..., Mathens [/bib_ref].
Nowadays complementary medicine, including herbal medicine, is becoming an increasingly popular health care approach, which has been used for both general maintenance of health and for treatment of minor illnesses. In Europe several herbs of different plant species have been used against flu and the common cold: lime and elder flowers, meadowsweet flowers and herb, purple echinacea aerial parts and roots, wild rose, blackcurrant and sea buckthorn fruits, lemon juice, etc. In Russia and Estonia garlic, ramson, onion, raspberry, cranberry, Iceland moss, cowslip primrose aerial parts and roots, plantain species leaves, yarrow, oregano, aerial parts of common and creeping thyme, coltsfoot leaves, willow bark, etc. have also been used for centuries.
Estonia is a small country in Northern Europe, one of the three Baltic States, bordering with Finland, Russia and Latvia. About 3/ 4 of its 1.3 mil population are Estonian speaking, while the remaining population mostly speaks Russian as their mothertongue. Estonian people have long traditions in the use of medicinal plants and, due to neighbouring Russia, the traditional medicinal plants of both Europe and Russia have been consumed.
Of almost 1500 native and over 700 introduced plant species [bib_ref] Eestitaimestik. [Vascular plant flora of Estonia, Kukk [/bib_ref] , Estonians utilized over 500 taxa in the period 1888-1994. Although within the century the proportion of medicinal plants requiring human intervention increased and the use of taxa unfavourable to human impact decreased, overall knowledge on the use of medicinal plants remained high among the Estonian speaking population [bib_ref] Change in medical plant use in Estonian ethnomedicine: a historical comparison between..., Sõukand [/bib_ref]. Currently in Estonian alternative medicine there are two main means of obtaining information on the use of medicinal plants: traditions or literary sources [bib_ref] Plant as object within herbal landscape: different kinds of perception, Sõukand [/bib_ref]. Also, some people rely on previous personal experience or information provided by a health care professional, usually a pharmacist [bib_ref] How well informed are pharmacy customers in Estonia about minor illnesses and..., Volmer [/bib_ref] [bib_ref] Self-reported competence of Estoniancommunity pharmacists in relation to herbal products: Findings from..., Volmer [/bib_ref].
The aim of this study was to evaluate the complementary selftreatment of the common cold and flu with medicinal plants among pharmacy customers in Estonia. We argue that different demographic groups prefer to use different medicinal plants for self-treatment of the common cold and flu.
# Methods
## Survey design
The study was undertaken in two different regions of Estoniain the capital city of the biggest island of Estonia, Kuressaare, where the majority of permanent residents are Estonian speaking, and in the capital city of Estonia, Tallinn, where the number of Russian speaking permanent residents is higher. In both towns the survey was undertaken in one community pharmacy classified as a medium size pharmacy (approx. 30,000 prescriptions per year) by an independent researcher. From June to August 2011, a sample of pharmacy customers participated in the survey. In both towns the pharmacy customers who had bought prescription or OTC medicine were invited to participate in the survey, the main difference being that in Kuressaare the majority of survey participants were Estonian speaking and in Tallinn, Russian speaking. The research assistant explained the purposes of the study and oral informed consent was received from all interviewed pharmacy customers. In Kuressaare 185, and in Tallinn 174 survey instruments were distributed, with a request to complete the questionnaire at the pharmacy or return it within one week. Due to the lack of time to complete the survey instruments at the pharmacy and to unreturned survey instruments, 150 completed survey instruments were collected from each pharmacy.
## Survey instrument
The survey instrument was developed and discussed by a panel of researchers with a pharmaceutical background and extended knowledge in phytotherapy, ethnobotany and social pharmacy. A copy of the survey instrument can be obtained from the corresponding author.
The survey instrument consisted of multiple-choice items related to (1) To compile the list of medicinal plants included in the survey instrument, a literature search was undertaken and a pilot test was run among 13 randomly selected community pharmacists, who selected ten of the more frequently suggested plants against the common cold and flu out of twenty proposed medicinal plants. The content validity and comprehensibility of the survey instrument was pre-tested by ten randomly selected pharmacy customers of participating community pharmacies.
Minor changes to the wording of the items were made, based on feedback received. No samples of medicinal plants were used, but the names of medicinal plants included in the survey were well known and univocal for the speakers of both languages.
The following ten most popular herbs were included in the survey instrument: Allium sativum L.
# Statistical analysis
Initial data were coded, inserted and stored in an MS Excel database. Most data analysis were performed in SPSS (Statistical Package for Social Sciences, Chicago, IL), while the analysis of the means of used medicinal plants was performed in statistical program R. Frequencies were calculated and cross-tabulation was performed to evaluate the correlation between the use and perception of medicinal plants and demographic characteristics (spoken language, age, gender and education). Linear regression analysis was used; level of statistical significance was set at p#0.05.
# Results
## Symptoms of and action taken to treat the common cold and flu
Respondents (for details on demographic characteristics see [fig_ref] Table 1: Socio-demographic characteristics of study population [/fig_ref] perceived the common cold and flu as a combination of several symptoms. The most frequent ones were tiredness, weakness, muscle and joint pain (17%), followed by fever, headache and cough with rhinitis (10%). The rest of the respondents presented other different combinations of described symptoms.
Female respondents mentioned the whole complex of symptoms more often, and stressed fever with tiredness, while male respondents described fever, headache and cough with rhinitis more frequently (p = 0,004). Both Estonian and Russian speaking respondents presented equally general symptoms of the common cold and flu. However, Estonians listed tiredness as a symptom more frequently, whereas Russians listed headaches (p = 0,004).
Among 97% of the respondents, action was taken to treat the common cold and flu -69% used self-treatment and 28% consulted a general practitioner. In the case of self-treatment, single use of home made remedies and medicinal plants was described as the most common mode of treatment by 56% of the respondents, followed by the combined use of herbal products and OTC medicines. Statistically significant differences were identified regarding education and treatment activities. Respondents with higher education used more self-treatment with medicinal plants than respondents with different education levels (p = 0.044).
Knowledge about medicinal plants was mostly received from family and friends or literature (49%). Only a few respondents (2%) used general practitioners or pharmacists as an information source about medicinal plants for treatment of the common cold and flu. Respondents have used a combination of Estonian native and non-native species of medicinal plants. Across all the groups of respondents the most used herbs were lemon fruits (73%) and garlic (70%), followed by raspberry (51%), chamomile (48%) and blackcurrant (44%). To some extent ginger (37%), lime flowers (35%) and onion (32%) were used. Only a few people mentioned the use of Iceland moss (11%) and echinacea (9%).
## Use of medicinal plants
## Age, gender and education
In general, consumption of medicinal plants increased with age (p = 0.01). To be specific, the use of garlic (p,0.01), onion, lime flowers, blackcurrant and raspberry increased with age [fig_ref] Figure 2: The use of medicinal plants in different age groups [/fig_ref]. The use of ginger (p = 0.000), lime flowers (p = ,0.013) and blackcurrant (p = 0.005) was more popular among women than men. Respondents with higher education used more lemon (p = 0.005), raspberry and lime flowers (p = 0.011).
## Language spoken
According to the results of the current survey, respondents speaking Estonian were more eager to use lime flowers and chamomile than those speaking Russian. On contrary, the Russian speaking respondents reported more frequent use of raspberry and lemon [fig_ref] Figure 3: The use of medicinal plants in different language groups [/fig_ref].
# Discussion
This was the first study in Estonia to describe the use of medicinal plants for the treatment of the common cold and flu by different demographic groups among pharmacy customers. Respondents demonstrated good knowledge in describing the symptoms of the common cold and flu, and a responsible attitude towards the cure of the described minor illnesses. Self-treatment with medicinal plants was widespread, as every respondent used several different species of medicinal plants. Alongside the single use of medicinal plants, home made remedies and OTC medicines were used in combination to relieve the discomfort related to illnesses. The described results support a contemporary approach to healthcare, where complementary medicine and biomedicine have not been contrasted, they further knowledge in biomedicine (e.g. about medicinal products) and even serve as a promoting factor to learn more about medicinal plants and herbal products [bib_ref] Do pharmaceuticals displace local knowledge and use of medicinal plants? Estimates from..., Giovannini [/bib_ref].
The hypothesis set for our survey about differences in the use of medicinal plants by different demographic groups was not always supported by the survey results.
More frequent use of medicinal plants and herbal products at older ages could be connected not only with the insufficient knowledge about medicinal plants of younger generations. Older people suffer more frequently from different illnesses and have the time and motivation to look for different approaches towards relieving their condition. In contrast, younger people look for fast treatment options and combine medicinal plants with OTC medicines.
The findings of our study do not support the results of previous international research where schooling had a negative correlation with the use of traditional medicines [bib_ref] Do pharmaceuticals displace local knowledge and use of medicinal plants? Estimates from..., Giovannini [/bib_ref]. In contrast in our study the respondents with higher education were more frequent users of medicinal plants than the other age groups. The reason for such differences could be connected not with the education itself, but with the environment in which people live and obtain theoretical knowledge and practical skills. Estonia is a small country, not yet strongly influenced by urbanization. Over the last few years there has been a trend towards a more traditional way of living, and respondents with higher education, more knowledgeable in health care issues and in better financial situations, probably practice complementary medicine more frequently.
Nationally based differences in the use of medicinal plants did not always support the information that was previously known. The preference for lemon and garlic contradicts the traditional approach, as in Estonian historical herbal folklore (covering 1888-1994) lemon is never mentioned and garlic is mentioned only few times. Although the medicinal use of garlic was already promoted by some authors in Estonia at the beginning of 20th century, its wider culinary use was established considerably later and its healing properties were finally acknowledged by a popular herbaland its future editions [bib_ref] Change in medical plant use in Estonian ethnomedicine: a historical comparison between..., Sõukand [/bib_ref].
In our survey the use of garlic depended only on the age of the respondents, contradicting the common myth in Estonia of Russians using this medicinal plant traditionally. Still, the use of lemon depends on the language respondents use and is probably caused by the habit of using lemon in the traditional Russian tea ceremony, which makes lemon more often available in households with a Russian background.
The use of lime flowers was unknown in Estonia until the first decades of the 20th century, a historical study of the regional use of plants in the 1930s reports only one use of lime used against colds, among 175 use-reports on the treatment of various diseases [bib_ref] Personal and shared: the reach of different herbal landscapes, Sõukand [/bib_ref]. Later, the use of lime became very popular, it became the most used herb against flu and the common cold in the latter half of the 20th century [bib_ref] The use of teetaimed in Estonia, 1880s-1990s, Soukand [/bib_ref].
Despite the modern media emphasizing the use of echinacea for the prevention of flu and the common cold, supported by evidence based data, this herb is the least used by all demographic groups, probably because of its perceived expensiveness: although it is a popular plant in gardens, there is no tradition of making homemade extract, and the respondents used preparations purchased from community pharmacies.
Respondents of the current study used mostly non-professional information sources to obtain knowledge on how to use medicinal plants in the treatment of the common cold and flu. The results are partly supported by earlier research, where previous experience was considered important in decision making about which treatment option to select [bib_ref] How well informed are pharmacy customers in Estonia about minor illnesses and..., Volmer [/bib_ref]. However, in Estonia, with its long tradition of the sale of medicinal plants and herbal products at community pharmacies, customers were more likely to employ pharmacists as a professional source of information.
# Survey limitations
The survey was undertaken in only two regions of Estonia, and although an appropriate sample of pharmacy customers was used, the presented results are not generalizable across the whole country or for different demographic groups. The low number of male respondents who participated in the survey could be connected with the lesser frequency of their visits to community pharmacies. In the future, the survey setting could be changed.
# Conclusion
Regardless of some statistically significant differences in preferred species among different age, education, sex and language groups, the general attitude towards medicinal plants for selftreatment of the common cold and flu in Estonia was very favourable. Estonia serves as an example of a country with long traditions of the use of medicinal plants in the treatment of minor illnesses. Due to its geographical location, the traditions of different nations are mixed. In the future, the empirical use of complementary medicine could be more frequently combined with evidence based knowledge about the efficacy of medicinal plants. Future research should evaluate the effectiveness of the most used plants. Moreover, it is important to assess the factors affecting the selection of specific plants, for better understanding of the selftreatment behaviour of the population.
[fig] Figure 1: Mean number of used medicinal plants with reference to different demographic characteristics. Abbreviations: age groups: 15-35, 36-50, 51-65, 65+; education: ELM-elementary, SEC -secondary, VOC -vocational, HGH -higher; languages spoken at home: EST -Estonian, RUS -Russian, TWO -both languages; gender: F -female, M -male. doi:10.1371/journal.pone.0058642.g001 [/fig]
[fig] Figure 2: The use of medicinal plants in different age groups. * -p,0.01. ** p,0.05. doi:10.1371/journal.pone.0058642.g002 among male rather than female respondents (mean 3.4 medicinal plants, p = 0.022). However, the use of medicinal plants increased with age (p = 0.01). [/fig]
[fig] Figure 3: The use of medicinal plants in different language groups. * -p,0.01. doi:10.1371/journal.pone.0058642.g003 [/fig]
[table] Table 1: Socio-demographic characteristics of study population. [/table]
|
Languages cool as they expand: Allometric scaling and the decreasing need for new words
We analyze the occurrence frequencies of over 15 million words recorded in millions of books published during the past two centuries in seven different languages. For all languages and chronological subsets of the data we confirm that two scaling regimes characterize the word frequency distributions, with only the more common words obeying the classic Zipf law. Using corpora of unprecedented size, we test the allometric scaling relation between the corpus size and the vocabulary size of growing languages to demonstrate a decreasing marginal need for new words, a feature that is likely related to the underlying correlations between words. We calculate the annual growth fluctuations of word use which has a decreasing trend as the corpus size increases, indicating a slowdown in linguistic evolution following language expansion. This ''cooling pattern'' forms the basis of a third statistical regularity, which unlike the Zipf and the Heaps law, is dynamical in nature.
B ooks in libraries and attics around the world constitute an immense ''crowd-sourced'' historical record that traces the evolution of culture back beyond the limits of oral history. However, the disaggregation of written language into individual books makes the longitudinal analysis of language a difficult open problem. To this end, the book digitization project at Google Inc. presents a monumental step forward providing an enormous, publicly accessible, collection of written language in the form of the Google Books Ngram Viewer web application. Approximately 4% of all books ever published have been scanned, making available over 10 7 occurrence time series (word-use trajectories) that archive cultural dynamics in seven different languages over a period of more than two centuries. This dataset highlights the utility of open ''Big Data,'' which is the gateway to ''metaknowledge'' 2 , the knowledge about knowledge. A digital data deluge is sustaining extensive interdisciplinary research efforts towards quantitative insights into the social and natural sciences ''Culturomics,'' the use of high-throughput data for the purpose of studying human culture, is a promising new empirical platform for gaining insight into subjects ranging from political history to epidemiology [bib_ref] Quantitative analysis of culture using millions of digitized books, Michel [/bib_ref]. As first demonstrated by Michel et al. [bib_ref] Quantitative analysis of culture using millions of digitized books, Michel [/bib_ref] , the Google n-gram dataset is well-suited for examining the microscopic properties of an entire language ecosystem. Using this dataset to analyze the growth patterns of individual word frequencies, Petersen et al. [bib_ref] Statistical laws governing fluctuations in word use from word birth to word..., Petersen [/bib_ref] recently identified tipping points in the life trajectory of new words, statistical patterns that govern the fluctuations in word use, and quantitative measures for cultural memory. The statistical properties of cultural memory, derived from the quantitative analysis of individual word-use trajectories, were also investigated by Gao et al. [bib_ref] Culturomics meets random fractal theory: Insights into long-range correlations of social and..., Gao [/bib_ref] , who found that words describing social phenomena tend to have different long-range correlations than words describing natural phenomena.
Here we study the growth and evolution of written language by analyzing the macroscopic scaling patterns that characterize word-use. Using the Google 1-gram data collected at the 1-year time resolution over the period 1800-2008, we quantify the annual fluctuation scale of words within a given corpora and show that languages can be said to ''cool by expansion.'' This effect constitutes a dynamic law, in contrast to the static laws of Zipf and Heaps which are founded upon snapshots of single texts. The Zipf law [bib_ref] Zipf's law and the structure and evolution of languages, Tsonis [/bib_ref] [bib_ref] Modeling statistical properties of written text, Serrano [/bib_ref] [bib_ref] Two regimes in the frequency of words and the origin of complex..., Ferrer I Cancho [/bib_ref] [bib_ref] The variation of Zipf's law in human language, Ferrer I Cancho [/bib_ref] [bib_ref] Least effort and the origins of scaling in human language, Ferrer I Cancho [/bib_ref] [bib_ref] Zipf's law unzipped, Baek [/bib_ref] , quantifying the distribution of word frequencies, and the Heaps law 13,18-20 , relating the size of a corpus to the vocabulary size of that corpus, are classic paradigms that capture many complexities of language in remarkably simple statistical patterns. While these laws have been exhaustively tested on relatively small snapshots of empirical data, here we test the validity of these laws using extremely large corpora.
Interestingly, we observe two scaling regimes in the probability density functions of word usage, with the Zipf law holding only for the set of more frequently used words, referred to as the ''kernel lexicon'' by Ferrer i Cancho et al. [bib_ref] Two regimes in the frequency of words and the origin of complex..., Ferrer I Cancho [/bib_ref]. The word frequency distribution for the rarely used words constituting the ''unlimited lexicon'' 14 obeys a distinct scaling law, suggesting that rare words belong to a distinct class. This ''unlimited lexicon'' is populated by highly technical words, new words, numbers, spelling variants of kernel words, and optical character recognition (OCR) errors.
Many new words start in relative obscurity, and their eventual importance can be under-appreciated by their initial frequency. This fact is closely related to the information cost of introducing new words and concepts. For single topical texts, Heaps observed that the vocabulary size exhibits sub-linear growth with document size. Extending this concept to entire corpora, we find a scaling relation that indicates a decreasing ''marginal need'' for new words which are the manifestation of cultural evolution and the seeds for language growth. We introduce a pruning method to study the role of infrequent words on the allometric scaling properties of language. By studying progressively smaller sets of the kernel lexicon we can better understand the marginal utility of the core words. The pattern that arises for all languages analyzed provides insight into the intrinsic dependency structure between words.
The correlations in word use can also be author and topic dependent. Bernhardsson et al. recently introduced the ''metabook'' concept [bib_ref] The meta book and size-dependent properties of written language, Bernhardsson [/bib_ref] [bib_ref] Size-dependent word frequencies and translational invariance of books, Bernhardsson [/bib_ref] , according to which word-frequency structures are author-specific: the word-frequency characteristics of a random excerpt from a compilation of everything that a specific author could ever conceivably write (his/her ''metabook'') should accurately match those of the author's actual writings. It is not immediately obvious whether a compilation of all the metabooks of all authors would still conform to the Zipf law and the Heaps law. The immense size and time span of the Google n-gram dataset allows us to examine this question in detail.
# Results
Longitudinal analysis of written language. Allometric scaling analysis 21 is used to quantify the role of system size on general phenomena characterizing a system, and has been applied to systems as diverse as the metabolic rate of mitochondria [bib_ref] Allometric scaling of metabolic rate from molecules and mitochondria to cells and..., West [/bib_ref] and city growth [bib_ref] Modelling urban growth patterns, Makse [/bib_ref] [bib_ref] Modeling urban growth patterns with correlated percolation, Makse [/bib_ref] [bib_ref] Laws of population growth, Rozenfeld [/bib_ref] [bib_ref] Zipf's law for cities: An explanation, Gabaix [/bib_ref] [bib_ref] Growth, innovation, scaling, and the pace of life in cities, Bettencourt [/bib_ref] [bib_ref] The size, scale, and shape of cities, Batty [/bib_ref] [bib_ref] The area and population of cities: New insights from a different perspective..., Rozenfeld [/bib_ref]. Indeed, city growth shares two common features with the growth of written text: (i) the Zipf law is able to describe the distribution of city sizes regardless of country or the time period of the data [bib_ref] Zipf's law for cities: An explanation, Gabaix [/bib_ref] , and (ii) city growth has inherent constraints due to geography, changing labor markets and their effects on opportunities for innovation and wealth creation [bib_ref] Growth, innovation, scaling, and the pace of life in cities, Bettencourt [/bib_ref] [bib_ref] The size, scale, and shape of cities, Batty [/bib_ref] , just as vocabulary growth is constrained by human brain capacity and the varying utilities of new words across users [bib_ref] Two regimes in the frequency of words and the origin of complex..., Ferrer I Cancho [/bib_ref].
We construct a word counting framework by first defining the quantity u i (t) as the number of times word i is used in year t. Since the number of books and the number of distinct words grow dramatically over time, we define the relative word use, f i (t), as the fraction of the total body of text occupied by word i in the same year
[formula] f i t ð Þ:u i t ð Þ=N u t ð Þ,ð1Þ [/formula]
where the quantity N u t ð Þ:
[formula] X Nw t ð Þ [/formula]
i~1 u i t ð Þ is the total number of indistinct word uses while N w (t) is the total number of distinct words digitized from books printed in year t. Both the N w (''types'' giving the vocabulary size) and the N u (''tokens'' giving the size of the body of text) are generally increasing over time.
The Zipf law and the two scaling regimes. Zipf investigated a number of bodies of literature and observed that the frequency of any given word is roughly inversely proportional to its rank, with the frequency of the z-ranked word given by the relation
[formula] f z ð Þ*z {f ,ð2Þ [/formula]
with a scaling exponent f < 1. This empirical law has been confirmed for a broad range of data, ranging from income rankings, city populations, and the varying sizes of avalanches, forest firesand firm size [bib_ref] Zipf plots and the size distribution of firms, Stanley [/bib_ref] to the linguistic features of nonconding DNA 32 . The Zipf law can be derived through the ''principle of least effort,'' which minimizes the communication noise between speakers (writers) and listeners (readers) [bib_ref] Least effort and the origins of scaling in human language, Ferrer I Cancho [/bib_ref]. The Zipf law has been found to hold for a large dataset of English text [bib_ref] Two regimes in the frequency of words and the origin of complex..., Ferrer I Cancho [/bib_ref] , but there are interesting deviations observed in the lexicon of individuals diagnosed with schizophrenia [bib_ref] The variation of Zipf's law in human language, Ferrer I Cancho [/bib_ref]. Here, we also find statistical regularity in the distribution of relative word use for 11 different datasets, each comprising more than half a million distinct words taken from millions of books [bib_ref] Quantitative analysis of culture using millions of digitized books, Michel [/bib_ref]. [fig_ref] Figure 1 |: Two-regime scaling distribution of word frequency [/fig_ref] shows the probability density functions P(f) resulting from data aggregated over all the years (A,B) as well as over 1-year periods as demonstrated for the year t 5 2000 (C,D). Regardless of the language and the considered time span, the probability density functions are characterized by a striking two-regime scaling, which was first noted by Ferrer i Cancho and Solé 14 , and can be quantified as
[formula] P f ð Þ* f {a{ , if f vf | 00 unlimited lexicon 00 ½ f {az , if f wf | 00 kernel lexicon 00 ½ :ð3Þ [/formula]
These two regimes, designated ''kernel lexicon'' and ''unlimited lexicon,'' are thought to reflect the cognitive constraints of the brain's finite vocabulary [bib_ref] Two regimes in the frequency of words and the origin of complex..., Ferrer I Cancho [/bib_ref]. The specialized words found in the unlimited lexicon are not universally shared and are used significantly less frequently than the words in the kernel lexicon. This is reflected in the kink in the probability density functions and gives rise to the anomalous two-scaling distribution shown in [fig_ref] Figure 1 |: Two-regime scaling distribution of word frequency [/fig_ref].
The exponent a 1 and the corresponding rank-frequency scaling exponent f in Eq. (2) are related asymptotically by [bib_ref] Two regimes in the frequency of words and the origin of complex..., Ferrer I Cancho [/bib_ref] a
[formula] z <1z1=f,ð4Þ [/formula]
with no analogous relationship for the unlimited lexicon values a 2 and f 2 . lists the average a 1 and a 2 values calculated by aggregating a 6 values for each year using a maximum likelihood estimator for the power-law distribution [bib_ref] Power-law distributions in empirical data, Clauset [/bib_ref]. We characterize the two scaling regimes using a crossover region around f 3 < 10 25 to distinguish between a 2 and a 1 : (i) 10 28 # f # 10 26 corresponds to a 2 and (ii) 10 24 # f # 10 21 corresponds to a 1 . For the words that satisfy f > f 3 that comprise the kernel lexicon, we verify the Zipf scaling law f < 1 (corresponding to a < 2) for all corpora analyzed. For the unlimited lexicon regime f = f 3 , however, the Zipf law is not obeyed, as we find a 2 < 1.7. Note that a 2 is significantly smaller in the Hebrew, Chinese, and the Russian corpora, which suggests that a more generalized version of the Zipf law 14 may be needed, one which is slightly language-dependent, especially when taking into account the usage of specialized words from the unlimited lexicon. For individual books, the empirically-observed scaling relation between N u and N w obeys
[formula] N w * N u ð Þ b ,ð5Þ [/formula]
with b , 1, with Eq. (5) referred to as ''the Heaps law''. It has subsequently been found that Heaps' law emerges naturally in systems that can be described as sampling from an underlying Zipf distribution. In an information theoretic formulation of the the abstract concept of word cost, B. Mandelbrot predicted the relation b 5 1/f in 1961 [bib_ref] On the theory of word frequencies and on related Markovian models of..., Mandelbrot [/bib_ref] , where f is the scaling exponent corresponding to a 1 , as in Eqs. (3) and (4). This prediction is limited to relatively small texts where the unlimited lexicon, which manifests in the a 2 regime, does not play a significant role. A mathematical extension of this result for general underlying rank-distributions is also provided by using an infinite urn scheme, and extended to broader classes of heavy-tailed distributions recently by Gnedin et al [bib_ref] Notes on the occupancy problem with infinitely many boxes: general asymptotics and..., Gnedin [/bib_ref].
Recent research efforts using stochastic master equation techniques providing evidence that the distribution is stable even over shorter time frames and likely emerges in corpora that are sufficiently large to be comprehensive of the language studied. For details concerning the scaling exponents we refer to www.nature.com/scientificreports to model the growth of a book have also predicted this intrinsic relation between Zipf's law and Heaps' law [bib_ref] Modeling statistical properties of written text, Serrano [/bib_ref] [bib_ref] A formal derivation of Heaps' Law, Van Leijenhorst [/bib_ref] [bib_ref] Zipf's law leads to Heaps' law: Analyzing their relation in finite-size systems, Lü [/bib_ref]. [fig_ref] Figure 2 |: Allometric scaling of language [/fig_ref] confirms a sub-linear scaling (b , 1) between N u and N w for each corpora analyzed. These results show how the marginal returns of new words are given by
[formula] LN u LN w * N w ð Þ 1{b ð Þ =b ,ð6Þ [/formula]
which is an increasing function of N w for b , 1. Thus, the relative increase in the induced volume of written languages is larger for new words than for old words. This is likely due to the fact that new words are typically technical in nature, requiring additional explanations that put the word into context with pre-existing words. Specifically, a new word requires the additional use of preexisting words as a result of both (i) the explanation of the content of the new word using existing technical terms, and (ii) the grammatical infrastructure necessary for that explanation. Hence, there are large spillovers in the size of the written corpus that follow from the intricate dependency structure of language stemming from the various grammatical roles [bib_ref] The large-scale structure of semantic networks: Statistical analyses and a model of..., Steyvers [/bib_ref] [bib_ref] Network model of human language, Markosova [/bib_ref].
In order to investigate the role of rare and new words, we calculate N u and N w using only words that have appeared at least U c times. We select the absolute number of uses as a word use threshold because a word in a given year can not appear with a frequency less than 1/N u , hence any criteria using relative frequency would necessarily introduce a bias for small corpora samples. This choice also eliminates words that can spuriously arise from Optical Character Recognition (OCR) errors in the digitization process and also from intrinsic spelling errors and orthographic spelling variations. [fig_ref] Figure 3 |: Pruning reveals the variable marginal return of words [/fig_ref] show the relational dependence of N u and N w on the exclusion of low-frequency words using a variable cutoff U c 5 2 n with n 5 0 … 11. As U c increases the Heaps scaling exponent increases from b < 0.5, approaching b < 1, indicating that core words are structurally integrated into language as a proportional background. Interestingly, Altmann et al. [bib_ref] Niche as a determinant of word fate in online groups, Altmann [/bib_ref] recently showed that ''word niche'' can be an essential factor in modeling word use dynamics. New niche words, though they are marginal increases to a language's lexicon, are themselves anything but ''marginal'' -they are core words within a subset of the language. This is particularly the case in online communities in which individuals strive to distinguish themselves on short timescales by developing stylistic jargon, highlighting how language patterns can be context dependent.
We now return to the relation between Heaps' law and Zipf's law. summarizes the b values calculated by means of ordinary least squares regression using U c 5 0 to relate N u (t) to N w (t). For U c 5 1 we find that b < 0.5 for all languages analyzed, as expected from Heaps law, but for U c > 8 the b value significantly deviates from 0.5, and for U c > 1000 the b value begins to saturate approaching unity. Considering that a 1 < 2 implies f < 1 for all corpora, [fig_ref] Figure 3 |: Pruning reveals the variable marginal return of words [/fig_ref] shows that we can confirm the relation b(U c ) < 1/f only for the more pruned corpora that require relatively large U c . This hidden feature of the scaling relation highlights the underlying structure of language, which forms a dependency network between the common words of the kernel lexicon and their more esoteric counterparts in the unlimited lexicon. Moreover, the function hN w /hN u , (N u ) b21 is a monotonically decreasing function for b , 1, demonstrating the decreasing marginal need for additional words as a corpora grows. In other words, since we get more and more ''mileage'' out of new words in an already large language, additional words are needed less and less.
Corpora size and word-use fluctuations. Lastly, it is instructive to examine how vocabulary size N w and the overall size of the corpora N u affect fluctuations in word use. [fig_ref] Figure 5 |: Literary productivity and vocabulary size in the Google Inc [/fig_ref] shows how N w (t) and N u (t) vary over time over the past two centuries. Note that, apart from the periods during the two World Wars, the number of words printed, which we will refer to as the ''literary productivity'', has been increasing over time. The number of distinct words (vocabulary size) has also increased reflecting basic social and technological advancement [bib_ref] Quantitative analysis of culture using millions of digitized books, Michel [/bib_ref].
To investigate the role of fluctuations, we focus on the logarithmic growth rate, commonly used in finance and economics
[formula] r i t ð Þ:ln f i tzDt ð Þ{ln f i t ð Þ~ln f i tzDt ð Þ f i t ð Þ ,ð7Þ [/formula]
to measure the relative growth of word use over 1-year periods, Dt ; 1 year. Recent quantitative analysis on the distribution P(r) of word use growth rates r i (t) indicates that annual fluctuations in word use deviates significantly from the predictions of null models for language evolution [bib_ref] Statistical laws governing fluctuations in word use from word birth to word..., Petersen [/bib_ref]. We define an aggregate fluctuation scale, s r (tjf c ), using a frequency cutoff f c / 1/Min[N u (t)] to eliminate infrequently used words. The quantity Min[N u (t)] is the minimum corpora size over the period of analysis, and so 1/Min[N u (t)] is an upper bound for the minimum observed frequency for words in the corpora. [fig_ref] Figure 6 |: Non-stationarity in the characteristic growth fluctuation of word use [/fig_ref] shows s r (tjf c ), the standard deviation of r i (t) calculated across all words that satisfy the condition AEf i ae $ f c for words with lifetime T i $ 10 years, using f c 5 1/Min[N u (t)]. Visual inspection suggests a general decrease in s r (tjf c ) over time, marked by sudden increases during times of political conflict. Hence, the persistent increase in the volume of written language is correlated with a persistent downward trend what could be thought of as the ''system temperature'' s r (tjf c ): as a language grows and matures it also ''cools off''.
Since this cooling pattern could arise as a simple artifact of an independent identically distributed (i.i.d) sampling from an increasingly large dataset, we test the scaling of s r (tjf c ) with corpora size. [fig_ref] Figure 7 |: Growth fluctuation of word use scale with the size of the corpora [/fig_ref] shows that for large N u (t), each language is characterized by a scaling relation
[formula] s r tjf c ð Þ*N u tjf c ð Þ {b ,ð8Þ [/formula]
with language-dependent scaling exponent b < 0.08-0. [bib_ref] Central limit theorems for certain infinite urn schemes, Karlin [/bib_ref]. We use f c 5 10/Min[N u (t)], which defines the frequency threshold for the inclusion of a given word in our analysis. There are two candidate null models which give insight into the limiting behavior of b. The Gibrat proportional growth model predicts b 5 0 and the Yule-Simon urn model predicts b 5 1/2 42 . We observe b , 1/2, which indicates that the fluctuation scale decreases more slowly with increasing corpora size than would be expected from the Yule-Simon urn model prediction, deducible via the ''delta method'' for determining the approximate scaling of a distribution and its standard deviation s 43 .
To further compare the roles of the kernel lexicon versus the unlimited lexicon, we apply our pruning method to quantify the dependence of the scaling exponent b on the fluctuations arising from rare words. We omit words from our calculation of s r (tjU c ) if their use u i (t) in year t falls below the word-use threshold U c . [fig_ref] Figure 7 |: Growth fluctuation of word use scale with the size of the corpora [/fig_ref] shows that b(U c ) increases from values close to 0 to values less than 1/2 as U c increases exponentially. An increasing b(U c ) confirms our conjecture that rare words are largely responsible for the fluctuations in a language. However, because of the dependency structure between words, there are residual fluctuation spillovers into the kernel lexicon likely accounting for the fact that b , 1/2 even when the fluctuations from the unlimited lexicon are removed. For a given corpora and U c value we make a scatter plot between N w (t | U c ) and N u (t | U c ) using words with u i (t) $ U c , using the same data color-U c correspondence as in [fig_ref] Figure 3 |: Pruning reveals the variable marginal return of words [/fig_ref]. (Panel Inset) We use OLS estimation to estimate the scaling exponent b(U c ) for the model
[formula] N w (t | U c ) , [N u (t | U c )] b to show that b(U c ) [/formula]
increases from approximately 0.5 towards unity as we prune the corpora of extremely rare words. Our longitudinal language analysis provides insight into the structural importance of the most frequent words which are used more times per appearance and which play a crucial role in the usage of new and rare words.
## Www.nature.com/scientificreports
A size-variance relation showing that larger entities have smaller characteristic fluctuations was also demonstrated at the scale of individual words using the same Google n-gram dataset [bib_ref] Statistical laws governing fluctuations in word use from word birth to word..., Petersen [/bib_ref]. Moreover, this size-variance relation is strikingly analogous to the decreasing growth rate volatility observed as complex economic entities (i.e. firms or countries) increase in size [bib_ref] The size variance relationship of business firm growth rates, Riccaboni [/bib_ref] [bib_ref] Scaling Behavior in Economics: I. Empirical Results for Company Growth, Amaral [/bib_ref] [bib_ref] Power Law Scaling for a System of Interacting Units with Complex Internal..., Amaral [/bib_ref] [bib_ref] The growth of business firms: Theoretical framework and empirical evidence, Fu [/bib_ref] [bib_ref] Quantitative relations between risk, return, and firm size, Podobnik [/bib_ref] [bib_ref] Common scaling behavior in finance and macroeconomics, Podobnik [/bib_ref] , which strengthens the analogy of language as a complex ecosystem of words governed by competitive forces.
Further possible explanations for b , 1/2 is that language growth is counteracted by the influx of new words which tend to have growth-spurts around 30-50 years following their birth in the written corpora [bib_ref] Statistical laws governing fluctuations in word use from word birth to word..., Petersen [/bib_ref]. Moreover, the fluctuation scale s r (tjf c ) is positively influenced by adverse conditions such as wars and revolutions, since a decrease in N u (t) may decrease the competitive advantage that old words have over new words, allowing new words to break through. The globalization effect, manifesting from increased human mobility presented for all examined corpora. There is an overall decreasing trend arising from the increasing size of the corpora, as depicted in [fig_ref] Figure 5 |: Literary productivity and vocabulary size in the Google Inc [/fig_ref]. On the other hand, the steady production of new words, as depicted in [fig_ref] Figure 5 |: Literary productivity and vocabulary size in the Google Inc [/fig_ref] counteracts this effect. We calculate s r (t | f c ) using the relatively common words that meet the criterion that their average word use AEf i ae over the entire word history T i (using words with lifetime T i $ 10 years) is larger than a threshold f c ; 1/Min[N u (t)] (see .
www.nature.com/scientificreports SCIENTIFIC REPORTS | 2 : 943 | DOI: 10.1038/srep00943 during periods of conflict, is also responsible for the emergence of new words within a language.
# Discussion
A coevolutionary description of language and culture requires many factors and much consideration. While scientific and technological advances are largely responsible for written language growth as well as the birth of many new words 9 , socio-political factors also play a strong role. For example, the sexual revolution of the 1960s triggered the sudden emergence of the words ''girlfriend'' and ''boyfriend'' in the English corpora 1 , illustrating the evolving culture of romantic courting. Such technological and socio-political perturbations require case-by-case analysis for any deeper understanding, as demonstrated comprehensively by Michel et al. [bib_ref] Quantitative analysis of culture using millions of digitized books, Michel [/bib_ref].
Here we analyzed the macroscopic properties of written language using the Google Books database 1 . We find that the word frequency distribution P(f) is characterized by two scaling regimes. While frequently used words that constitute the kernel lexicon follow the Zipf law, the distribution has a less-steep scaling regime quantifying the rarer words constituting the unlimited lexicon. Our result is robust across languages as well as across other data subsets, thus extending the validity of the seminal observation by Ferrer i Cancho and Solé 14 , who first reported it for a large body of English text. The kink in the slope preceding the entry into the unlimited lexicon is a likely consequence of the limits of human mental ability that force the individual to optimize the usage of frequently used words and forget specialized words that are seldom used. This hypothesis agrees with the ''principle of least effort'' that minimizes communication noise between speakers (writers) and listeners (readers), which in turn may lead to the emergence of the Zipf law [bib_ref] Least effort and the origins of scaling in human language, Ferrer I Cancho [/bib_ref].
Using an extremely large written corpora that documents the profound expansion of language over centuries, we analyzed the dependence of vocabulary growth on corpus growth and validate the Heaps law scaling relation given by Eq. 5. Furthermore we systematically prune the corpora data using a word occurrence threshold U c , and comparing the resulting b(U c ) value to the f < 1 value, which is stable since it is derived from the ''kernel'' lexicon. We conditionally confirm the theoretical prediction f < 1/b 13,34-38 , which we validate only in the case that the extremely rare ''unlimited'' lexicon words are not included in the data sample (see [fig_ref] Figure 3 |: Pruning reveals the variable marginal return of words [/fig_ref].
The economies of scale (b , 1) indicate that there is an increasing marginal return for new words, or alternatively, a decreasing marginal need for new words, as evidenced by allometric scaling. This can intuitively be understood in terms of the increasing complexities and combinations of words that become available as more words are added to a language, lessening the need for lexical expansion. However, a relationship between new words and existing words is retained. Every introduction of a word, from an informal setting (e.g. an expository text) to a formal setting (e.g. a dictionary) is yet another chance for the more common describing words to play out their respective frequencies, underscoring the hierarchy of words. This can be demonstrated quite instructively from Eq. (6) which implies that for b~1=2 that LN u LN w !N w , meaning that it requires a quantity proportional to the vocabulary size N w to introduce a new word, or alternatively, that a quantity proportional to N w necessarily results from the addition. Though new words are needed less and less, the expansion of language continues, doing so with marked characteristics. Taking the growth rate fluctuations of word use to be a kind of temperature, we note that like an ideal gas, most languages ''cool'' when they expand. The fact that the relationship between the temperature and corpus volume is a power law, one may, loosely speaking, liken language growth to the expansion of a gas or the growth of a company [bib_ref] The size variance relationship of business firm growth rates, Riccaboni [/bib_ref] [bib_ref] Scaling Behavior in Economics: I. Empirical Results for Company Growth, Amaral [/bib_ref] [bib_ref] Power Law Scaling for a System of Interacting Units with Complex Internal..., Amaral [/bib_ref] [bib_ref] The growth of business firms: Theoretical framework and empirical evidence, Fu [/bib_ref] [bib_ref] Quantitative relations between risk, return, and firm size, Podobnik [/bib_ref] [bib_ref] Common scaling behavior in finance and macroeconomics, Podobnik [/bib_ref]. In contrast to the static laws of Zipf and Heaps, we note that this finding is of a dynamical nature.
Other aspects of language growth may also be understood in terms of expansion of a gas. Since larger literary productivity imposes a downward trend on growth rate fluctuations -which also implies that the ranking of the top words and phases becomes more stable [bib_ref] Evolution of the most common English words and phrases over the centuries, Perc [/bib_ref] -productivity itself can be thought of as a kind of inverse pressure in that highly productive years are observed to ''cool'' a language off. 8) between s r (t | f c ) and the corpus size N u (t | f c ). We calculate s r (t | f c ) using the relatively common words that meet the criterion that their average word use AEf i ae over the entire word history (using words with lifetime T i $ 10 years) is larger than a threshold f c ; 10/Min[N u (t)] (see . We show the language-dependent scaling value b < 0.08-0.35 in each panel. For each language we show the value of the ordinary least squares best-fit b value with the standard error in parentheses. (B) Summary of b(U c ) exponents calculated using a use-threshold U c , instead of a frequency threshold f c as used in (A). Error bars indicate the standard error in the OLS regression. We perform this additional analysis in order to provide alternative insight into the role of extremely rare words. For increasing U c the b(U c ) value for each corpora increases from b < 0.05 to b , 0.25. This language pruning method quantifies the role of new rare words (also including OCR errors, spelling and other orthographic variants), which are the significant components of language volatility. www.nature.com/scientificreports SCIENTIFIC REPORTS | 2 : 943 | DOI: 10.1038/srep00943
[fig] The: Heaps law and the increasing marginal returns of new words. Heaps observed that vocabulary size, i.e. the number of distinct words, exhibits a sub-linear growth with document size 18 . This observation has important implications for the ''return on investment'' of a new word as it is established and becomes disseminated throughout the literature of a given language. As a proxy for this return, Heaps studied how often new words are invoked in lieu of preexisting competitors and examined the linguistic value of new words and ideas by analyzing the relation between the total number of words printed in a body of text N u , and the number of these which are distinct N w , i.e. the vocabulary www.nature.com/scientificreports SCIENTIFIC REPORTS | 2 : 943 | DOI: 10.1038/srep00943 size 18 . marginal returns of new words, hN u /hN w quantifies the impact of the addition of a single word to the vocabulary of a corpus on the aggregate output (corpus size). [/fig]
[fig] Figure 1 |: Two-regime scaling distribution of word frequency. The kink in the probability density functions P(f) occurs around f 3 < 10 25 for each corpora analyzed (see legend). (A,B) Data from all years are aggregated into a single distribution. (C,D) P(f) comprising data from only year t 5 2000 [/fig]
[fig] Figure 2 |: Allometric scaling of language. Scatter plots of the output corpora size N u given the empirical vocabulary size N w using all data (U c 5 0) over the 209-year period 1800-2008. Shown are OLS estimation of the exponent b quantifying the Heaps' law relation N w , [N u ] b . [/fig]
[fig] Figure 3 |: Pruning reveals the variable marginal return of words. The Heaps scaling exponent b depends on the extent of the inclusion of the rarest words.For a given corpora and U c value we make a scatter plot between N w (t| U c ) and N u (t| U c ) using words with u i (t) $ U c . (Panel Inset) We use OLS estimation to estimate the scaling exponent b(U c ) for the model N w (t| U c ) ,[N u (t| U c )] b to show that b(U c ) increases from approximately 0.5 towards unity as we prune the corpora of extremely rare words. Our longitudinal language analysis provides insight into the structural importance of the most frequent words which are used more times per appearance and which play a crucial role in the usage of new and rare words. [/fig]
[fig] Figure 4 |: Pruning reveals the variable marginal return of words. The Heaps scaling exponent b depends on the extent of the inclusion of the rarest words. [/fig]
[fig] Figure 5 |: Literary productivity and vocabulary size in the Google Inc. 1-gram dataset over the past two centuries. (A) Total size of the different corpora N u (t | U c ) over time, calculated by using words that satisfy u i (t) $ U c ; 16 to eliminate extremely rare 1-grams. (B) Size of the written vocabulary N w (t | U c ) over time, calculated under the same conditions as (A). [/fig]
[fig] Figure 6 |: Non-stationarity in the characteristic growth fluctuation of word use. The standard deviation s r (t | f c ) of the logarithmic growth rate r i (t) is [/fig]
[fig] Figure 7 |: Growth fluctuation of word use scale with the size of the corpora. (A) Depicted is the quantitative relation in Eq.( [/fig]
|
An emerging paradigms on cervical cancer screening methods and devices for clinical trails
# Introduction
Every year around 5,70,000 women are affected with cervical cancer and over 3,11,000 women die from the disease [bib_ref] Estimates of incidence and mortality of cervical cancer in 2018: a worldwide..., Arbyn [/bib_ref]. Although there are techniques of screening in various forms and types around the world, most of the knowledge or the technique does not reach the interior parts of the world, like that which are in developing countries. Most of the rural areas either lack good health care support systems or high-level screening equipment, especially when it comes to cancer screening. Most women notice changes in their body only when the symptoms are severe or close to higher rates of malignancies. Fortunately, different studies have come up with various techniques that are cost effective, simple and efficient. In the following sections, major types of screening and the subcategories of testing are described.
Cervical cancer is curable, unlike the majority of malignancies. There are two methods of prevention. First, through immunization, and second, by routine screening that can find HPV infection or aberrant cells before they become malignant. Though infections may be cured within 2 years, 10% of the infections may last longer than 2 years. A chronic infection raises the possibility of getting precancerous or, ultimately, aggressive cancer. However, there is a safe and effective vaccine that can stop HPV 16 and HPV 18 infections. Starting at age nine, vaccinations are preferred for young girls. If a high-grade precancerous disease manifests, it must be surgically removed before developing into cervical cancer.
. /fpubh. .
## Literature survey
Van Baars, studied that the primary screening with high-risk human papillomavirus (hrHPV) detection has been advocated to prevent cervical cancer. While given the chance to selfsample for hrHPV testing, women who are not already attending screening (non-responders) are more likely to participate. Dry Evalyn Brush system is as good for self-sampling compared to physician-taken sample for hrHPV detection and is highly acceptable to women. The fact that this study was conducted in a hospital setting is a drawback. Self-samples were always taken prior to practitioner smears, which is another theoretical restriction (3). Parashari et al. found that the Magnivisualizer has an increased identification rate of early malignant tumors from 60 to 95% when compared to unaided visual inspection. It also has allowed for the detection of 58 percent of low-grade dysplasia cases and 83 percent of high-grade dysplasia cases that would not have been detected by simple visual assessment. The Magnivisualizer has a poorer sensitivity for detecting low-grade dysplasias, although this may not be a severe drawback because most low-grade dysplasias tend to regress even in the absence of treatment (4).
Veena Singh et al., showed that, in resource poor environments where colposcopic services are not offered on a local level, a cost efficient, handheld instrument called magnivisualizer is a useful for identifying cervical precancerous and cancerous lesions. In comparison to Visual inspection with acetic acid (VIA), this device demonstrated higher sensitivity (83 vs. 54%) without sacrificing specificity in the detection of severe precancerous lesions of the cervix. Due to the standard of colposcopy has limited specificity, it leads to unnecessary biopsies, therefore it cannot be used as a substitute (5).
Saleh, found that, in comparison to a Pap smear, VIA is an effective screening tool because it is a simple test with a low cost and great sensitivity. It can be therefore used in low-resource locations as an alternative cervical cancer screening method. The sensitivity of Pap smear was 50.1%, specificity was 93.1%, and its negative and positive predictive values were 89.3 and 65.6%, respectively. VIA's sensitivity was 90%, specificity was 37%, and its prediction accuracy was positive. Fifty-two percent and an 81% negative predictive value. Because of the less PPV of VIA, the issue of multiple false positives, discourages the see-andtreat strategy. Although, PPV linked to incidence, the VIA test's capabilities might increase if a see-and-treat approach were used in a high incidence of cervical cancer in a high-risk area [bib_ref] Can visual inspection with acetic acid be used as an alternative to..., Saleh [/bib_ref].
The findings of Emre Ozgu et al. suggest that TruScreen, has 86.1% of sensitivity, and can be used as a cervical cancer screening test that offers quick results without the requirement for a professional. Because it eliminates the need for pathologists and subjectivity in Pap smear interpretation, Cervical cancer screening is possible with TruScreen, particularly in nations with low socioeconomic level. The effectiveness of screening did not significantly enhance when TruScreen and HPV testing were combined (7).
Muszynski et al. performed a study where Colposcopy alone showed 61% of sensitivity and 80% specificity for identifying high-grade lesions. Zedscan and colposcope together exhibited a sensitivity of 93%−100%, and between a range of 91 and 100% negative predictive value [bib_ref] The impact of using electrical impedance spectroscopy (ZedScan) on the performance of..., Muszynski [/bib_ref]. Based on the above literatures, there are certain methodologies and techniques with which cervical cancer screening is done. A detailed explanation of the various methods is discussed in methodology. From the literatures it is also observed that each of the techniques has their own advantages and disadvantages.
# Methodology
There are various ways of screening, testing, and diagnosing cervical cancer. The below mentioned are mostly used for cervical cancer and these are as follows:
- Screening using Tissue Scrapping.
- Screening using Visual inspection with acetic acid (VIA).
- Screening using Devices.
- Screening using artificial intelligence (AI) and machine learning (ML) techniques.
- Screening using Mobile technology.
## Screening using tissue scrapping
Cervical screening checks the health of the cervix. It helps to prevent cancer or treat them if any abnormality is found. In this method, a small portion of the cervical tissues are smeared using swab test brushes and are tested in laboratories for traces of HPV infections. There are two major methods through which this screening takes place, one of the methods is the Pap smear test and the other is HPV-DNA test. The Pap smear test is considered as the golden standard for cervical cancer screening (9).
## Pap smear
Typically, a pelvic exam is performed in addition to the Pap smear (10, 11). In some circumstances, HPV test may be administered to females older than 30 in place of a Pap smear. Based on the type of test, the doctor either places the cell sample obtained from the woman' cervix onto a glass slide (conventional) or place it in a container containing a specific liquid to preserve the sample (liquid-based) [bib_ref] Persistent disparities in cervical cancer screening uptake: knowledge and socio demographic determinants..., Johnson [/bib_ref].
Then the samples are then taken to a lab where they are examined under a microscope for cell features that might point . /fpubh. . to cervical cancer or a precancerous condition. shows the Pap smear test kit and shows the procedure depiction.
## Hpv-dna test
It is common to have HPV infection around the genitals. Cervical cancer and other malignancies are caused by specific high-risk type of HPV. Low-risk types of HPV may cause genital warts in the vagina, cervix, and on the skin. In general, it is not advised to use the HPV-DNA test to identify low-risk HPV infections. This is because majority of low-risk lesions are physically recognizable. The medical professional inserts a device known as a speculum into the vagina, opens it slightly and gently collects the cells from the area around the cervix (14). shows the HPV-DNA test kit.
The cells are delivered to a lab where a microscope examination will take place. This examiner tests the cells to see if they contain genetic material (referred to as DNA) from cancer-causing HPV strains. To identify the exact type of HPV, further testing may be conducted. A Pap smear may be .
/fpubh. . substituted with the HPV DNA test. Co-testing is the term used when they are carried out together.
## Screening using via
VIA is a screening method in which the cervix is observed after the application of 3%−5% of acetic acid in the cervix region which results in acetowhite lesions. shows the result of before and after applying acetic acid. VIA offers the advantages of being simple to use, affordable [bib_ref] Comparative analysis of visual inspection with acetic acid and Lugol's iodine and..., Poli [/bib_ref] , and sensitive when compared to Pap smear, and quick results assessment (17). As a result, VIA is a good way of cervical cancer screening in many regions of the world, particularly in areas with limited resources. Variations in sensitivity and specificity could be caused by a variety of factors, including the following:
- Expertise training - Light source variation, and - The procedure for making a 4%−5% acetic acid solution and storing it.
In poor countries with limited resources, VIA can be utilized as a mass screening method for cervical cancer. It was reported that at the low grade squamous intraepithelial lesion (LSIL) threshold, VIA was less sensitive i.e., is 86.7% which is lesser than that of clinical cytology with a sensitivity of 91.4%, but the difference was not statistically significant (18). HPV testing outperformed cytology in terms of sensitivity, but there was no significant reduction in specificity (84.2 vs. 86.6%).
In addition to VIA tests there are methods that gives importanceto a white light visual inspection of the cervix; white light enables the correct site of biopsy to be selected. The majority of rural clinics utilize a torch or a regular tungsten bulb, which misses many severe lesions. Through this study, the usage of white light is highly advisable for screening purpose. When compared to Pap smear, VIA has a high sensitivity.
## Screening using devices
From the previous studies, it is understood that a clinical test includes the collection of tissues and is slower when compared to other methods. Although the success rates of cancer screening ishigh, the scrapping method may disturb the patient's convenience. In order to be more efficient, cost effective and quick, current studies have come up with techniques that does not involve scrapping of tissues and does not infuse any discomfort. This section will discuss about the modern cervical cancer screening devices and their efficiency.
## Av magnivisualizer
It is a low-cost technology for screening uterine cervical cancer using magnivisualizer. It increases the identification rate of early malignant tumors from 60 to 95% when compared to single-handed visual inspection. It also allows the detection of 58% of low-grade dysplasia cases and 83% of high-grade dysplasia cases that would not have been detected by simple visual assessment. The magnivisualizer is highly sensitive, with a sensitivity of around 57.5% in detecting low-grade dysplasia, when compared to 75.3% of cytological evaluation [bib_ref] AV Magnivisualizer: a low-cost screening technology for early detection of precancerous and..., Parashari [/bib_ref]. For higher degrees of lesions, however, the two approaches had equivalent sensitivity. The magnivisualizer had a 94.3% specificity, while cytology had a 99% specificity.
The AV Magnivisualizer, has a complete spectrum of visible light (white light) and interchangeable magnification. In the AV Magnivisualizer is shown.
The sole accessible light source in primary health Center outdoor situations is usually a tungsten bulb providing yellow light attached to a torch or examination light. On lesions with a pinkish mucosal background, this type of light has a masking effect. The handheld Magnivisualizer can be considered a suitable tool for identification of cervical precancerous and cancerous lesions in low-resource settings where colposcopic services are not available at the community level. However, due to its low specificity, it cannot replace colposcopy, which results in numerous needless biopsies.
. /fpubh. .
## Pocket
Point of Care Tampon (POCkeT) is a Novel Low-Cost device that can capture images and can be used to diagnose cervical lesions. By delegating cervical cancer screening to community health workers, the portable, low-cost method has the potential to enhance access to cervical cancer screening in low-resource settings. Women who enter the screening cascade for the first time are usually not familiar with the procedure of having a speculum and are intimidated by the idea of having a cold metal object inside their bodies. This barrier was the reason that ultimately led to the conceptualization of the pocket colposcope [bib_ref] International image concordance study to compare a point of care tampon colposcope..., Mueller [/bib_ref].
The pocket colposcope can be inserted through the speculum to provide a close-up view of the cervix to take a picture. When the colposcope is close to the cervix, a set of high-quality pictures are obtained that are better than that of colposcopes on the market, and are both effective in cost and size. As seen in there are two versions of the pocket colposcope (20), one with a 5-mega pixel camera that can be used to obtain images via insertion through a speculum and one with a 2-mega pixel camera that is more slender and can be inserted into a tampon-like introducer called the Calla scope, to enable speculum-free visualization of the cervix.
## Truscreen
TruScreen is a unique, proprietary Opto-Electrical technology to evaluate the tissue of the cervix. Unlike cytology, TruScreen does not only examine surface epithelial cells, it produces specific frequencies of light transmitted through the cervical tissue to identify changes in the basal and stromal layers. There are four LEDs that sequentially emit light at three wavelengths, distant red, infrared and green. Electrical measurements test the cell's resistance to current to characterize the tissue. This characterization of these tissues is called electrical impedance spectroscopy (21). As seen in , the TruScreen system consists of a disposable Single Use Sensor (SUS), a Handheld Device (HHD), and an Intelligent Cradle (IC) that work in concert to detect and classify the cancerous and precancerous changes in the cervix.
First, many areas on the cervix are gently touched using a pen-like wand wrapped in a SUS. The SUS has electrodes and a precision lensthat interact with the cervix. During this process, it transmits and receives low-level optical and electrical information from the cervical tissue.
The signals are then analyzed by an integrated AI-enabled algorithm on the TruScreen Handheld Device, which compares them to a database of 2,000 patients from various ethnic and geographic backgrounds who have different histology diagnoses. Physicians receive immediate results from this analysis, which detects the presence of abnormal (cancerous and precancerous) cells in the cervix. In contrast to traditional Pap tests, which can take weeks or even months to provide a result in some countries, each TruScreen examination produces results in 1-2 min.
## Zedscan
ZedScan is a diagnostic gadget that makes use of an accessory to conventional colposcopy to offer an evaluation of the cervical epithelial tissue in real time. Electrical Impedance Spectroscopy (EIS) is a scientifically-proven technique to distinguish among normal, pre-cancerous and cancerous tissues (neoplasias) [bib_ref] The impact of using electrical impedance spectroscopy (ZedScan) on the performance of..., Muszynski [/bib_ref]. This technique is likewise suitable for the prognosis of diverse cancers and pre-cancerous conditions. shows the ZedScan cervical probe.
. /fpubh. . ZedScan makes use of EIS technique to distinguish among normal, pre-cancerous and cancerous tissue at the cervix based on electrical properties. When used along with colposcopy, ZedScan has established extra accuracy in detecting cervical disease.
## Cervastra
There are many devices that detects and screens cervical cancer and Pap smear is one of the predominant ways. One of the greatest disadvantages of the Pap smear is that it takes a long time to receive the test results. In order to cater to this problem CervAstra was invented [bib_ref] Artificial intelligence in clinical diagnostics-an Indian perspective, Roy [/bib_ref]. CervAstra is a device to detect Cervical Cancer using a Computational Pathology platform. shows the device.
CervAstra analyzes Pap smear samples at the Point-of-Care to state normal or abnormal in few hours compared to a longer duration depending on the location of sample collection.
## Luviva
It is a Hyperspectral Imaging Spectroscopy (HIS) technology based non-invasive scanning device that includes a base unit and a single-patient-use disposable probe. It is useful to scan the cervix with light source to detect the cancerous and precancerous cells [bib_ref] Optical imaging for cervical cancer detection: solutions for a continuing global problem, Thekkek [/bib_ref]. Light reflected from the cervix is analyzed through a spectrometer. shows the LuViva Scan device.
Based on the information from the spectrometer, an image of the cervix will be generated which distinguishes the healthy from diseased tissue. The development of this technology has yielded seventeen patents.
## Screening using ai and ml based applications
Numerous automatic and semi-automatic techniques have been developed as a result of the automatic analysis of colposcopy for the diagnosis of precancerous lesions. Neoplasia can be divided into several categories, and acetowhite zones can be classified as high-or low-risk, malignant or non-cancerous, normal or aberrant.
Many research works have been carried out to detect the cancer from Pap smear images and colposcopy images using ML and Deep Learning (DL) techniques. Many research works used Support Vector Machine (SVM) (24, 25), Adaptive Neuro Fuzzy Inference System (ANFIS) based classifier [bib_ref] Multi-feature based benchmark for cervical dysplasia classification evaluation, Xu [/bib_ref] [bib_ref] RamaPraba PS, Ranganathan H. Computerized lesion detection in colposcopy cervix images based..., Jaya [/bib_ref] , Bayesian classifier (27, 28) for cervical cancer detection and classification into cancerous or noncancerous. Many other ML techniques such K-Nearest Neighbor (KNN), Neural Networks, Adaboost classifier have been explored for the detection purpose.
By analyzing digitalized Papanicolaou-smear images with a primary training dataset, 15 different machine learning algorithms were built for the detection of cervical cancer. Almost all algorithms successfully identified the cancer cells. Although multilayer perceptrons are the highest among all the algorithms used in recent times multiple back propagation neural networks had a higher level of efficiency, whereas the other algorithms has a lower level of efficiency. The findings show that techniques based on AI can be utilized to develop tools for widespread cervical cancer screening [bib_ref] Screening of cervical cancer by artificial intelligence based analysis of digitized papanicolaou-smear..., Gupta [/bib_ref].
Using colposcopy images, the traditional methods based on image processing and machine learning produced good results. However, these techniques require manual skill for feature extraction. The features can be automatically extracted from the data by deep learning. Apart from conventional ML techniques, existing DL architectures such as LeNet, VGG16/19, ResNet, MobileNet, Long Short-Term Memory (LSTM) and many proposed convolutional neural networks (CNN) have been used for the classification purpose.
The three most prominent Deep CNNs (ResNet-50, MobileNet, and NasNet) have been configured for training to Although AI is an advantage of the present digital era it also has major processing problems that might not give a complete solution for cancer screening techniques. It can be used for primary level of screening through which the presence of cervical cancer can be monitored. The future awaits for more development in AI algorithms that would facilitate in secondary level of cervical cancer screening.
## Screening using mobile
From the different methodologies and studies done this far, it also required to explore the techniques based on mobile screening. With a growing technology based on mobile and smartphone, there are two major mobile based techniques in order to screen cervical cancer through smartphones, these are discussed in detail as follows.
## Gynocular
The monocular colposcope called the Gynocular as shown in is a device that has optical capabilities when compared to basic colposcopes. This device screens the cancer using high resolution lens and is almost a smaller version of the traditional colposcope [bib_ref] Evaluation of stationary colposcope and the gynocular, by the Swede score systematic..., Nessa [/bib_ref].
Using this device diagnostic forecasts from distant assessment were revealed to be equivalent to estimates from actual colposcopy evaluation for the diagnosis of CIN2+ lesions.
## Mobile odt
This device uses a method called the Enhanced Visual Assessment (EVA) Colpo, which is made up of a Mobile phone attached with magnifying lenses, a number of rechargeable batteries and LEDs for illumination. Mobile ODT has also developed a mobile app that retains patient information, preserve cervical pictures, and keeps track of biopsies along with other clinical findings (35). shows the Mobile ODT device, this device has also proved to reduce false positive as well as false negative rates when compared to Pap tests, and the AI created in-house has been demonstrating good results.
. /fpubh. .
## Smartscopy
The ideology of smartscopy is that, instead of using an external flash and a device separately, the smartphones are used. This method is done after the application of a 3% solution of acetic acid to the cervix for 1 min, once the application is over a gynecologist inspects the cervix using Smartscopy with the activated flash mode pictures of the cervix (36). The recorded prominent areas revealed abnormal epithelium. Subsequently, the smartscopy findings, and the histological diagnosis was evaluated and was relatively successful. Although results were successful by using the iPhone 5S to inspect the uterine cervix for cervical cytology is welcoming, it might not be always always useful for screening cervical cancer.
# Conclusion
This paper intended to study various techniques that were found to be successful, simple and cost effective when it came to screening cervical cancer. In equipments like digital colposcope and LuViva which are high in cost and sensitive in hardware are difficult to be moved to rural areas. This is a disadvantage caused due to whichtraditional ways have to be followed in rural areas. Though the knowledge of how different techniques and methods have been of great use in cervical cancer screening was studied. In future, studies may have to come up methods where all of the possible screening methods are put under one roof. A cost-effective method with the application of Machine learning techniques collaborated with a mobile application would be of great use, such a device would be both cost, time efficient compared to the effectiveness of other devices. When such applications become a reality, it would be of great use in remote sectors of many developing countries.
# Author contributions
KR carried out the conceptualization of the research idea and supervised the study. GR carried out the analysis of the screening methods and verified the manuscript. SJ validated the systematic review. ST did a detailed survey on the screening methods for cervical cancer and drafted the manuscript. PE carried out the literature survey of screening techniques and methods. LM performed the study and drafted the manuscript.
[fig] FIGURE: A) The Pap smear test kit. (B) The procedure depiction. [/fig]
|
Seroprevalence of SARS-CoV-2 following the largest initial epidemic wave in the United States: Findings from New York City, May 13-July 21, 2020
Background: New York City (NYC) was the U.S. epicenter of the Spring 2020 COVID-19 pandemic. We present seroprevalence of SARS-CoV-2 infection and correlates of seropositivity immediately after the first wave.Methods: From a serosurvey of adult NYC residents (May 13-July 21, 2020), we calculated the prevalence of SARS-CoV-2 antibodies stratified by participant demographics, symptom history, health status, and employment industry. We used multivariable regression models to assess associations between participant characteristics and seropositivity.Results: Seroprevalence among 45,367 participants was 23.6% (95% CI, 23.2%-24.0%). High seroprevalence (>30%) was observed among Black and Hispanic individuals, people from high poverty neighborhoods, and people in health care or essential worker industry sectors. COVID-19 symptom history was associated with seropositivity (adjusted relative risk=2.76; 95% CI, 2.65-2.88).Other risk factors included sex, age, race/ethnicity, residential area, employment sector, working outside the home, contact with a COVID-19 case, obesity, and increasing numbers of household members.Conclusions:Based on a large serosurvey in a single U.S. jurisdiction, we estimate that just under one-quarter of NYC adults were infected in the first few months of the COVID-19 epidemic. Given disparities in infection risk, effective interventions for at-risk groups are needed during ongoing transmission.
A c c e p t e d M a n u s c r i p t 3
# Background
New York City (NYC) was the earliest U.S. epicenter of the coronavirus disease 2019 (COVID- [bib_ref] The Advisory Committee on Immunization Practices' Updated Interim Recommendation for Allocation of..., Dooling [/bib_ref] pandemic. The first known COVID-19 case in a NYC resident was reported on March 1, 2020. NYC schools closed on , and all non-essential businesses statewide closed on Governor Cuomo announced the "New York State on PAUSE" executive order. However, daily case counts rapidly rose in NYC, reaching a peak of 6,365 reported cases (76 per 100,000) on April 6, and returning to fewer than 500 cases (<6 per 100,000) in early June 2020. By July 21, there were 219,128 NYC residents diagnosed using nucleic acid testing for SARS-CoV-2 (the virus that causes , yielding a cumulative incidence of approximately 2,628 per 100,000 for confirmed COVID-19 infections among NYC residents. This estimate did not include people who were infected but unable or chose not to seek testing (e.g., those with asymptomatic or mildly symptomatic infections), and thus was not reflective of true cumulative COVID-19 incidence.
Compared with studies that use diagnostic tests, seroprevalence surveys can provide more complete estimates of the burden of infection by identifying people who were infected with SARS-CoV-2 but not reported as COVID-19 cases. If antibodies are a marker of total or partial immunity, they may also provide information on the proportion and characteristics of people who remain susceptible to the virus. Early reported seroprevalence estimates for communities in several geographic areas within the U.S. ranged from 1-20% [bib_ref] Performance characteristics of the Abbott Architect SARS-CoV-2 IgG assay and seroprevalence in, Bryan [/bib_ref] [bib_ref] Population point prevalence of SARS-CoV-2 infection based on a statewide random sample-Indiana, Menachemi [/bib_ref] [bib_ref] CDC Field Surveyor Team. Estimated community seroprevalence of SARS-CoV-2 antibodies-two Georgia counties, Biggs [/bib_ref] [bib_ref] Seroprevalence of SARS-CoV-2-Specific Antibodies Among Adults, Sood [/bib_ref] [bib_ref] Cumulative incidence and diagnosis of SARS-CoV-2 infection in New York, Rosenberg [/bib_ref]. Most published studies lack details on the characteristics of tested persons useful for determining risk factors for infection and seroconversion.
From mid-May through late-July 2020, widespread SARS-CoV-2 immunoglobulin G (IgG) antibody testing was offered to all adult NYC residents. We present the seroprevalence of SARS-CoV-2 infection and correlates of seropositivity among a large sample of the City's population.
A c c e p t e d M a n u s c r i p t 4
# Methods
## Specimen and data collection
Opportunities for no-cost antibody testing were made available for NYC residents aged >18 years during May 13-July 21, 2020. BioReference Laboratories, a large commercial laboratory, conducted specimen collection and testing. SARS-CoV-2 IgG testing was conducted using the Liaison SARS-CoV-2 S1/S2 assay (DiaSorin, Saluggia, Italy; 97.6% sensitivity and 99.3% specificity), which had received Emergency Use Authorization from the U.S. Food and Drug Administration.
Media, the internet, and local advertisement were used to disseminate information to the public on testing locations and registration procedures. Serosurvey participants made appointments for testing using BioReference Laboratories' online platform, where a survey form was available for them to complete at the time of booking an appointment. Testing sites were set up at City-operated facilities, one in each of the five boroughs (counties) of NYC. The testing sites were rotated to expand catchment areas, such that five specific sites were in operation from May 13-June 2 and a different set of five sites was set up for collections during June 26-July 21.
## Serosurvey eligibility and data elements
During online self-registration, required screening questions identified persons who may have had very recent infections and were either potentially infectious or had insufficient time to develop antibodies to SARS-CoV-2. Those who self-reported that they tested positive for SARS-CoV-2 by nasal or throat swab, or had fever, new onset or worsening cough, shortness of breath, or loss of taste (ageusia) or smell (anosmia) in the prior two weeks were not eligible for serosurvey participation.
Results of antibody testing were paired with participant characteristics and potential COVID-19 exposures from the self-administered survey. Survey data included demographics; industry and occupation; whether or not participants worked outside of the home during the PAUSE period (March 23-June 7); if they had exposure to a person diagnosed with COVID-19 (within 6 feet for >10 minutes); symptoms and symptom onset date; health care seeking and hospitalization for COVID-like A c c e p t e d M a n u s c r i p t 5 illness; whether they had a history of a prior SARS-CoV-2 polymerase chain reaction (PCR)-positive test; housing type and number of household members (measure of crowding); and health status (height, weight, and chronic underlying medical conditions).
Age and address were required for obtaining serology testing; completion of other survey questions was voluntary. Overall, 14.3% of participants (n=7,574) did not complete surveys. Among those who provided survey data, missing data across 12 categories of questions ranged from 2.7% to 14.0%.
# Data analysis
Participants' ZIP codes of residence were mapped to neighborhood poverty levels.
Neighborhood poverty was defined as the percent of a ZIP code's population with household incomes <100% of the Federal Poverty Level, per the 2013-2017 American Community Survey (low poverty: <10%, medium: 10%-19.9%, high: 20%-29.9%, very high: >30%).
We collected information on 9 underlying chronic conditions: diabetes, hypertension, heart disease, kidney disease, liver disease, asthma, chronic obstructive pulmonary disease/emphysema/chronic bronchitis, immunosuppressive condition, and immunosuppressive therapy. Reported height and weight were used to calculate participants' body mass index (BMI), (kilogram/meter 2 ); weight status categories were defined as underweight or normal (BMI<25), overweight (25≤BMI<30), obesity (30≤BMI<40), and severe obesity (BMI≥40). Participants with implausible weight or height values were excluded from BMI calculation (n=357).
We collected self-reported information on 11 symptoms of COVID-like illness (CLI) and categorized them into three groups. They were 1) those meeting the current Council of State and Territorial Epidemiologists (CSTE) COVID-19 case definition, which includes cough, shortness of breath/difficulty breathing, loss of taste, loss of smell, or any two of: chills, fever, headache, diarrhea, vomiting, sore throat, or body aches; 2) non-CSTE: one or more of the following: chills, fever, headache, diarrhea, vomiting, sore throat, or body aches; and 3) asymptomatic: report of no symptoms. Occupational Classification (SOC). Free text responses that autocoded below a predetermined probability threshold were reviewed by professionally trained coders.
In analyses, we combined participant NAICS sectors to examine seroprevalence for 5 broad categories: 1) health care and social assistance (NAICS Sector 62); 2) essential workers, considered for our local context to comprise individuals in sectors that were largely operational during New York For participants who tested more than once (n=200), we included test and survey data from the first testing event. Using population denominator estimates, we calculated testing rates per 100,000 NYC residents, stratified by age group, race/ethnicity, borough of residence, and neighborhood poverty level. We calculated the prevalence of SARS-CoV-2 antibodies among those with available survey data (85.7% of all participants), stratified by participant characteristics. Finally, we used robust Poisson regression to examine associations between seropositivity and correlates of interest for the entire sample, as well as subgroups of health care workers and essential workers.
Significant variables in bivariate analyses (p<0.05) were included in separate multivariable regression models that adjusted for sex (male, female), age group
# Results
## Serosurvey sample
The seroprevalence among 52,941 serosurvey participants was 24.8%; 7,574 did not complete surveys and were excluded from further analysis. While seroprevalence among everyone tested was similar to that among those included in the final analytic sample, participants with missing survey data were older than those with any survey data (44% vs 39% aged >45 years) and higher proportions of them lived in Brooklyn, Queens, or the Bronx (versus in Manhattan or Staten Island).
There were 45,367 participants in the final sample, representing an overall testing rate of 680 per 100,000 adult NYC residents. Testing rates varied by key demographics, with substantially higher testing levels among non-Hispanic White individuals (908/100,000) and those residing in Staten Island (2,512/100,000), and very low testing rates among people aged >65 years (260/100,000), non-Hispanic Black individuals (236/100,000), and those residing in Brooklyn, the Bronx, and in areas of very high poverty (all below 460/100,000) . [fig_ref] Table 1: Characteristics of participants and proportion with antibodies to SARS-CoV-2 [/fig_ref]. Approximately one-third of non-Hispanic Black and Hispanic participants were seropositive (33.5% and 35.3%, respectively). There was no appreciable variability in seroprevalence according to participant sex at birth, gender identity, sexual orientation, or age. Geographically, the highest seroprevalence was observed among residents in the Bronx (37.0%) and very high poverty neighborhoods (35.3%). Lower than average seroprevalence (<20%) was noted among residents in Manhattan, Staten Island, and low poverty neighborhoods.
## Seroprevalence estimates
One-quarter of participants (10,455/41,037) with comorbidity information reported at least one chronic condition, most commonly hypertension (47%), followed by asthma (38%) and diabetes (18%). Seroprevalence did not vary by chronic disease status shows a detailed breakdown of seroprevalence by industry within the health care and social assistance and essential worker categories.
## Correlates of seropositivity
Of 13 variables we examined, only two -underlying conditions and type of housing -were not statistically significantly associated with seropositivity in bivariate analyses [fig_ref] Table 3: Factors associated with seropositivity [/fig_ref]. Male sex, age 44-64 years, non-White race/ethnicity, living in a borough other than Manhattan or Staten Island, and living in neighborhoods with high or very high poverty levels were significantly associated with seropositivity in a multivariable regression model that included sex, age group, race/ethnicity, borough, and poverty. Adjusting for these 5 demographic variables in separate multivariable models, the following factors were associated with seropositivity: employment in a health care or essential worker category, or being unemployed at the time of the serosurvey; working outside of the home during PAUSE; having close contact with someone with COVID-19; having had CLI symptoms (the factor most strongly associated with seropositivity; adjusted relative risk 2.76 (95% CI, 2.65-2.88); being overweight, obese, or severely obese; and increasing numbers of household members.
A c c e p t e d M a n u s c r i p t 10 For healthcare workers and essential workers specifically, correlates of seropositivity were largely the same as those observed for the entire serosurvey sample. Key differences for health care workers were that seropositivity was not associated with sex or living in very high poverty. For essential workers, seropositivity was not associated with: age, living in Brooklyn, the Bronx, Manhattan, or Staten Island, or with working outside the home during PAUSE.
# Discussion
From the largest SARS-CoV-2 serosurvey in a single U.S. jurisdiction to date, we estimate that the number of persons infected with SARS-CoV-2 may have been as much as 7 times higher than the number of reported cases in the first five months of the pandemic. Almost one-quarter of people tested had evidence of acquired SARS-CoV-2 in the initial period of the pandemic. Amid continued SARS-CoV-2 transmission, layered interventions including rigorous and extensive monitoring, testing, contact tracing, promotion of individual prevention measures (e.g., face coverings, social distancing, frequent handwashing), and community restrictions on indoor activities have been implemented as an attempt to slow its spread.
Geographic and demographic characteristics of serosurvey participants with SARS-CoV-2 antibody tracked with the epidemiology of reported COVID-19 cases in NYC [bib_ref] COVID-19 Outbreak, Thompson [/bib_ref]. The highest seroprevalence was observed among Black and Hispanic people, and those living in Queens and the Bronx. Communities with higher poverty levels were disproportionately affected; more than onethird of participants living in ZIP codes where >30% of the population was living below the federal poverty level had antibodies to the SARS-CoV-2 virus. Elevated risks of testing seropositive persisted for these groups after accounting for other individual characteristics. Drivers of COVID-19 risk in urban areas include population density, transportation, employment with frequent public contact, crowded housing, and other socioeconomic and environmental factors [bib_ref] High population densities catalyse the spread of COVID-19, Rocklov [/bib_ref] [bib_ref] Racial health disparities and Covid-19 d Caution and Context, Chowkwanyun [/bib_ref] [bib_ref] Assessing differential impacts of COVID-19 on black communities, Millett [/bib_ref]. Differential exposure to various forms of structural oppression, including structural racism -centuries of racist policies and discriminatory practices across institutions, including government agencies, and society -also negatively affects the overall health and well-being of Black and Hispanic individuals [bib_ref] Structural racism and health inequities in the USA: evidence and interventions, Bailey [/bib_ref]. While [bib_ref] Prevalence of SARS-CoV-2 Antibodies in Health Care Personnel in the New York..., Moscola [/bib_ref] [bib_ref] Seroprevalence of SARS-CoV-2 Among Frontline Health Care Personnel in a Multistate Hospital..., Self [/bib_ref] , which was found to be as high as 31% in one NYC medical center in late Spring-Summer 2020 [bib_ref] Seroprevalence of SARS-CoV-2 Among Frontline Health Care Personnel in a Multistate Hospital..., Self [/bib_ref]. We report a similar prevalence among serosurvey participants who worked in the health care and social assistance sector (30%), and further found 45-50% seropositivity among subsets in nursing home or home health care services. Essential workers with exposure to the public, as defined for this study, had seropositivity over 30%. Our data support vaccine prioritization for workers in industries of accommodations and food services (e.g., restaurant workers), transportation and warehousing (e.g., public transit, drivers, postal workers, couriers), retail trade (e.g., grocery store workers), and those of other essential workers [bib_ref] The Advisory Committee on Immunization Practices' Updated Interim Recommendation for Allocation of..., Dooling [/bib_ref]. Until vaccines are widely available and taken, preventive measures to reduce workplace exposures to SARS-CoV-2 should continue to be emphasized.
While it has been suggested that 40-45% of SARS-CoV-2 infections are asymptomatic [bib_ref] Prevalence of Asymptomatic SARS-CoV-2 Infection: A Narrative Review, Oran [/bib_ref] , we found that 21% of serosurvey participants with antibodies to SARS-CoV-2 did not report a COVID-19-like illness. More than one in ten participants with either no symptoms or symptoms not meeting the clinical case definition for COVID-19 were seropositive. While symptoms meeting the CSTE COVID-19 definition captured three-quarters of antibody-positive people and had suboptimal prediction (positive predictive value: 34%), there was an almost three-fold risk of testing seropositive among participants with CSTE-consistent symptoms compared to those without such A c c e p t e d M a n u s c r i p t 12 symptoms. Among the CSTE symptoms, loss of taste and/or smell as sole symptom(s) was more predictive of seropositivity compared with only fever or only cough. Our experience with investigating SARS-CoV-2 clusters has revealed presentations with few and mild symptoms, making early recognition and control a significant challenge.
There are limitations to our analysis. We used a convenience sample, therefore it is possible that people who sought out testing perceived themselves to have been more likely to have been exposed to SARS-CoV-2; indeed, 60% of participants reported a history of CLI symptoms. A considerably biased sample could overestimate seroprevalence. On the other hand, lower representation of certain subgroups with higher infection rates -such as Black and Hispanic individuals -could underestimate true seroprevalence. However, our overall seroprevalence estimate was very similar to published estimates derived from a number of other NYC-based convenience samples during roughly the same time period; these include a New York State serosurvey of almost 6,000 NYC residents recruited at grocery stores (seroprevalence: 22.7%) [bib_ref] Cumulative incidence and diagnosis of SARS-CoV-2 infection in New York, Rosenberg [/bib_ref] , testing of residual clinical specimens obtained from two commercial laboratories that serve NYC (ranging from 17.6%-23.2% depending on the week of collection between April and July) [bib_ref] Seroprevalence of antibodies to SARS-CoV-2 in 10 sites in the United States, Havers [/bib_ref] , and a serosurvey of NYC first responders and public safety personnel (seroprevalence: 22.5%).
Our estimate also aligns well with the overall percent of NYC residents who tested positive for SARS-CoV-2 antibodies across NYC health care facilities, which was roughly 20% as of the third week of July 2020. Population-based, representative serosurveys will yield the most accurate measures of seroprevalence for NYC; results from a series of citywide population-based serosurvey being conducted by the DOHMH since early in the pandemic are forthcoming. The earliest of those serosurveys (June-October 2020) has found a weighted seroprevalence of 24.3%, a very similar estimate to ours [Manuscript submitted for publication].
Complete survey data were missing for approximately 15% of records and there were additional, partially missing data across survey fields. We chose not to impute missing values as we could not know the source and/or extent of any biases related to missing data and did not want to A c c e p t e d M a n u s c r i p t 13 risk magnifying them. We did not pre-test questions and there was the potential for issues related to comprehension. For example, approximately 20% of participants who reported having tested positive using a swab or saliva test for SARS-CoV-2 did not have detectable antibodies. Some people, particularly those with mild or clinically inapparent infection, may not develop detectable serum antibodies following infection or have antibodies that persist [bib_ref] Symptoms and immunoglobulin development in hospital staff exposed to a SARS-CoV-2 outbreak, Brandstetter [/bib_ref] [bib_ref] Antibody detection and dynamic characteristics in patients with COVID-19, Xiang [/bib_ref] , perhaps accounting for no IgG antibody detection by the time the serosurvey took place. It is possible though that some of these participants reported having received a diagnostic test, and not necessarily a positive test result, leading to our misclassification of their prior diagnosis.
Seroprevalence studies are useful for understanding the true prevalence of infection. We found that after the first several months of the pandemic, a substantial proportion of people in NYC had evidence of prior infection with SARS-CoV-2. However, it was less than what would lead to herd immunity needed to prevent future large outbreaks of COVID-19, which is estimated to be as much as 70% assuming that: 1) the basic reproductive number for SARS-CoV-2 is between 2 and 3. A c c e p t e d M a n u s c r i p t 25
[table] Table 1: Characteristics of participants and proportion with antibodies to SARS-CoV-2.CI: confidence interval a Council for State and Territorial Epidemiologists definition: cough, shortness of breath/difficulty breathing, loss of taste, or loss of smell, or any two of the following captured in serosurvey: chills, fever, headache, diarrhea, vomiting, sore throat, or body aches b Includes workers considered essential during the period of New York on PAUSE (March 23-June 7, 2020).Includes food services, administrative and support and waste management and remediation services, construction, retail trade (grocery, pharmacy/drug stores only), transportation and warehousing c Includes one or more of the following: diabetes, hypertension, chronic heart disease, chronic kidney disease, chronic liver disease, asthma, chronic obstructive pulmonary disease/emphysema/chronic bronchitis, immunosuppressive condition (e.g. HIV, autoimmune disease), immunosuppressive therapy (e.g. cancer treatment) d Exposure within 6 feet for >10 minutes to a person diagnosed with COVID-19 [/table]
[table] Table 2: Seroprevalence among participants with symptoms meeting the Council for State and TerritorialEpidemiologists clinical case definition.New loss of smell and/or taste (n=13,224) Symptoms include one or more of: cough, shortness of breath/difficulty breathing, loss of taste, loss of smell, chills, fever, headache, diarrhea, vomiting, sore throat, body aches [/table]
[table] Table 3: Factors associated with seropositivity.No./total (%) with SARS-CoV-2 antibodies [/table]
|
Molecular Aspects of Mycotoxins—A Serious Problem for Human Health
Mycotoxins are toxic fungal secondary metabolities formed by a variety of fungi (moulds) species. Hundreds of potentially toxic mycotoxins have been already identified and are considered a serious problem in agriculture, animal husbandry, and public health. A large number of food-related products and beverages are yearly contaminated by mycotoxins, resulting in economic welfare losses. Mycotoxin indoor environment contamination is a global problem especially in less technologically developed countries. There is an ongoing effort in prevention of mould growth in the field and decontamination of contaminated food and feed in order to protect human and animal health. It should be emphasized that the mycotoxins production by fungi (moulds) species is unavoidable and that they are more toxic than pesticides. Human and animals are exposed to mycotoxin via food, inhalation, or contact which can result in many building-related illnesses including kidney and neurological diseases and cancer. In this review, we described in detail the molecular aspects of main representatives of mycotoxins, which are serious problems for global health, such as aflatoxins, ochratoxin A, T-2 toxin, deoxynivalenol, patulin, and zearalenone.
# Introduction
Mycotoxins are secondary metabolites synthesized by a variety of fungi (moulds) species such as Fusarium, Aspergillus, Penicillium, Alternaria, and Claviceps. They constitute a structurally diverse group of toxic and low molecular weight compound, which is generally less than 1000 Da. Approximately 400 potentially toxic mycotoxins produced by more than 100 fungi species have been identified and it is considered that the most toxigenic to agriculture, animal husbandry, and public health are trichothecenes, ochratoxins, aflatoxins, zearalenone, fumonisins, patulin, and citrinin. Diseases caused by mycotoxins are called mycotoxicoses. Mycotoxins can enter human and animal bodies via food and feed. They contaminate many agricultural commodities. According to the Food and Agriculture Organization of the United Nations (FAO) report, 25% of global agricultural products are contaminated by mycotoxins each year, resulting in economic losses.
Nephrotoxicity, hepatotoxicity, genotoxicity, teratogenicity, immunotoxicity neurotoxicity EU: 3 µg/kg (cereal products), 5 µg/kg (unprocessed cereal), 10 µg/kg (dried fruits),µg/kg (spices) European Commission (EC): 5 ng/kg (coffee beans), 10 ng/kg (instant coffee), 0.5 µg/kg (cereal-based food) 2 µg/kg (wines) Joint FAO/WHO Expert Committee on Food Additives (JECFA): 0.1 µg/kg b. w. per weekT-2 toxin
## Aflatoxins
Aflatoxins are produced by Aspergillus species such as A. flavus, A. parasiticus, and rarely A. nomius. Among 18 different types of aflatoxins, there are four commonly occurring: B 1 (AFB 1 ), B 2 (AFB 2 ), G 1 (AFG 1 ), and G 2 (AFG 2 ). The AFB 1 is defined as the most common contaminant of foods and the most carcinogen and mutagen potent. According to the International Agency for Research on Cancer (IARC), AFB 1 is classified as group 1 carcinogen (carcinogenic to humans). Human exposure to aflatoxin B 1 is especially dangerous in populations with a high rate of hepatitis B virus (HBV), because it is estimated that the risk of liver cancer from AFB 1 exposure in HBV-positive people is 30 times higher than in the HBV-negative. It is considered that AFB 1 causes up to 28% of worldwide cases of hepatocellular carcinoma (HCC), which is the most frequent form of liver cancer. Chemically, aflatoxins are coumarin derivates, containing a fused dihydrofurofuran moiety. The structure of AFB 1 is distinguished by the cyclopentenone ring fusion to the coumarin lactone ring. Moreover, AFB 1 is freely soluble in polar organic solvents, slightly soluble in water, and insoluble in nonpolar solvents. This toxin is also characterized by instability to pH conditions, such as <3 or >10, and UV light. Aflatoxicosis is a result of consumption of contaminated food, which can cause immunosuppression, stunting in children and cancer. Tropical and subtropical countries are more exposed to aflatoxicosis, because the level of food contamination with mycotoxins is not sufficiently monitored. Aflatoxins contribute to the contamination of maize, peanuts, wheat, barley, oilseeds, and spices, but also milk, dairy products, meat, and eggs as a consequence of mouldy feed consumed by livestock. Aspergillus contamination can occur at pre-and post-harvest stages. Fungal growth can occur also in unsuitable storage conditions. Optimum conditions for Aspergillus growth in peanut kernels and on polished rice are 28-40 - C and a w 0.94-0.99, and for aflatoxins production they are 25-33 - C and a w 0.92-0.. The maximum amounts of aflatoxin B 1 in peanut kernels is at 28 - C, a w 0.96, and on polished rice optimal values are 33 - C and a w 0.96. AFB 1 is absorbed in the small intestine, transferred to blood stream, and then transported by plasma proteins and red blood cells to the liver. In liver cells, toxin is metabolized by microsomal-mixed function oxidase (MFO) enzymes, which belong to the Cytochrome P450 (CYP450) superfamily. AFB 1 is converted to a more toxic, highly reactive 8,9-exo-epoxide and 8,9-endo-epoxide metabolites. The 8,9-exo-epoxide has a high affinity to bind to the DNA and then form the 8,9-dihydro-8-(N'-guanyl)-9-hydroxy-AFB1 (AFB1-N-Gua) adduct, the formation of which leads to DNA mutations. The 8,9-exo-epoxide also binds to other macromolecules like RNA and proteins, which leads to inhibition of RNA, DNA, protein syntheses, and cellular function disorders. This epoxide can be involved in enzymatic and non-enzymatic conversion in AFB1-8,9-dixydroxydiol that can bind protein such as albumin, or can be converted in the aflatoxin dialdehyde and excreted via urine as a result of aflatoxin aldehyde reductase action. In the endoplasmic reticulum (ER) of liver cells, AFB 1 is also hydroxylated to fewer toxic metabolites: aflatoxin M 1 (AFM 1 ), aflatoxin Q 1 (AFQ 1 ), and aflatoxin P 1 (AFP 1 ). In ruminants fed with contaminated feed, a part of the AFB 1 is degraded by ruminal fluid microbiota and then transformed into an 18-times less toxic metabolite called aflatoxicol. About 1-2% of the toxin is absorbed in the gastrointestinal tract, next, by passive diffusion, it is hydroxylated in the liver and then it forms AFM 1 metabolite, which occurs in blood and is finally secreted in milk. Ruminants, as a consequence of their characteristic four-chamber stomachs, are considered to be less susceptible to mycotoxins than monogastric species, because mycotoxins can be partially eliminated by the rumen microbiota. However, the ruminal microbial detoxification varies depending on dietary changes or as a result of metabolic diseases. The maximum limit of AFB 1 has been set by the European Union at 2 µg/kg in all cereals and all cereal-derived products. China also has regulated the maximum limit of aflatoxin B 1 at 5 µg/kg in barley, wheat, and their products, 20 µg/kg in corn and corn products. The Food and Drug Administration (FDA) has set a maximum level for aflatoxins in dairy animal feed at 20 µg/kg.
## Ochratoxin a
Ochratoxin A (OTA) is produced by several fungal species including Aspergillus ochraceus, A. carbonarius, A. niger, Penicillium verrucosum, P. nordicum, and P. viridicatum. The optimal conditions for growth of OTA-producing fungi range from 15 to 40 °C and 0.77 to 0.99 water activity. The optimum temperature for toxin production for A. niger on maize kernels is 15-40 °C and for A. carbonarius, it is 15-35 °C. The conditions under which A. ochraceus can produce ochratoxin A in coffee beans are aw 0.97-0.99 and 25-30 °C. OTA is more common and the most toxic among ochratoxins, like ochratoxin B (OTB) or ochratoxin C (OTC). Different types of food are sources of exposure to OTA, including cereals (oats, maize, wheat, barley), cereal products, coffee beans, dried fruits, beer, grape juice, wine, as well as nuts, cacao products, and spices. A. carbonarius is the main OTA producer in wine and wine-dried fruits. In addition, this toxin is found in milk products, eggs, and meat. The chemical name of OTA is L-phenylalanine-N-[(5-chloro-3,4-dihydro-8-hydroxy-3-methyl-1-oxo-1H-2-benzopyrane-7-yl)carbo nyl]-(R)-isocoumarin. It is a pentaketide derived from the dihyrdocoumarins family, linked by a peptide bond to β-phenylalanine. It is a weak organic acid, which in neutral and acid conditions are soluble in polar organic solvents (chloroform, alcohols). In alkaline pH, it is soluble in aqueous sodium bicarbonate solution. The crystalline structure of OTA varies from white to colorless, and this toxin exhibits blue fluorescence in alkaline conditions and green fluorescence in acidic environment.
## Ochratoxin a
Ochratoxin A (OTA) is produced by several fungal species including Aspergillus ochraceus, A. carbonarius, A. niger, Penicillium verrucosum, P. nordicum, and P. viridicatum. The optimal conditions for growth of OTA-producing fungi range from 15 to 40 - C and 0.77 to 0.99 water activity. The optimum temperature for toxin production for A. niger on maize kernels is 15-40 - C and for A. carbonarius, it is 15-35 - C. The conditions under which A. ochraceus can produce ochratoxin A in coffee beans are a w 0.97-0.99 and 25-30 - C. OTA is more common and the most toxic among ochratoxins, like ochratoxin B (OTB) or ochratoxin C (OTC). Different types of food are sources of exposure to OTA, including cereals (oats, maize, wheat, barley), cereal products, coffee beans, dried fruits, beer, grape juice, wine, as well as nuts, cacao products, and spices. A. carbonarius is the main OTA producer in wine and wine-dried fruits. In addition, this toxin is found in milk products, eggs, and meat. The chemical name of OTA is L-phenylalanine-N-[(5-chloro-3,4-dihydro-8-hydroxy-3-methyl-1-oxo-1H-2-benzopyrane-7-yl) carbonyl]-(R)-isocoumarin. It is a pentaketide derived from the dihyrdocoumarins family, linked by a peptide bond to β-phenylalanine. It is a weak organic acid, which in neutral and acid conditions are soluble in polar organic solvents (chloroform, alcohols). In alkaline pH, it is soluble in aqueous sodium bicarbonate solution. The crystalline structure of OTA varies from white to colorless, and this toxin exhibits blue fluorescence in alkaline conditions and green fluorescence in acidic environment. Many studies have shown negative effects of OTA on human and animal health. It induces several toxic effects like nephrotoxicity, hepatotoxicity, genotoxicity, teratogenicity, immunotoxicity, and neurotoxicity. The toxin has a half-life of 840 h in blood after oral ingestion. OTA is absorbed from the small intestine, enters the circulation, and binds to the serum albumin in plasma. In the human circulatory system, 99.8% of ochratoxin A is present in the albumin-bond form. Erythrocytes contain insignificant amounts of OTA. Then, it is distributed to kidneys, liver, muscle, brain, and fat. Kidneys are the primary targetin which using transmission electron microscopy focal tubular cell proliferation, multiple adenoma-like structures in layers of the renal papilla and in convoluted tubules were observed. However, the presence of OTA has also been found in bone marrow, skin, adrenal medulla, and cortex or myocardium. Different mechanisms of OTA action have been observed. The toxin causes G2/M phase cell cycle arrest, apoptosis, necrosis, inhibition of microtubule assembly, cell division processes, or protein syntheses. Several studies have shown that OTA induces reactive oxygen species (ROS) generation leading to oxidative stress and ROS-mediated apoptosis, as well as DNA adducts and DNA single-strand breaks. have demonstrated that OTA causes a decrease in glutathione (GSH), which is an important antioxidant. OTA treatment causes also ROS, lipid peroxides, and nitric oxide (NO) generation. Another study has shown that exposure of chickens to OTA decreased their number of lymphocytes, leukocytes, and throbocytes, modified the mucosal architecture of duodenum, jejunum, and also ileoceacal junction, caused a reduction in the intestinal TCR1, TCR2, CD4+, CD8+ lymphocyte population, and led to enterocyte apoptosis. The neurotoxic properties of OTA are the result of inhibition of human astrocyte cell line proliferation and differentiation. An increase in mitochondrial and cytosolic calcium concentration also appears as a consequence of calcium overload-cells' death occurs. OTA inhibits expression of glutamate transporter 1 (GLT1), glutamate aspartate transporter (GLAST), that leads to reducing glutamate absorption by astrocytes, which play a potential role in stability or induction of neurodegenerative diseases like Parkinson's and Alzheimer's, as well as neuron motor degenerations. shown that in neuronal (Neuro-2a) cell line, OTA causes ROS generation resulting in c-jun amino-terminal-kinase (JNK)-mediated caspase-dependent apoptosis. The IARC has classified ochratoxin as a group 2B carcinogen (possible human carcinogen). Various institutes and committees have developed limits on the OTA content in different types of food. The European Union has set a maximum limit of ochratoxin A at 3 μg/kg for cereal products, 5 μg/kg for unprocessed cereals, 10 μg/kg for dried fruits, and 15 μg/kg for spices. The European Commission established a maximum limit of OTA at 5 ng/kg in coffee beans, 10 ng/kg in instant coffee, 0.5 μg/kg in cereal-based food for infants and children, and 2 μg/kg in wines. Furthermore, the Joint FAO/WHO Expert Committee on Many studies have shown negative effects of OTA on human and animal health. It induces several toxic effects like nephrotoxicity, hepatotoxicity, genotoxicity, teratogenicity, immunotoxicity, and neurotoxicity. The toxin has a half-life of 840 h in blood after oral ingestion. OTA is absorbed from the small intestine, enters the circulation, and binds to the serum albumin in plasma. In the human circulatory system, 99.8% of ochratoxin A is present in the albumin-bond form. Erythrocytes contain insignificant amounts of OTA. Then, it is distributed to kidneys, liver, muscle, brain, and fat. Kidneys are the primary targetin which using transmission electron microscopy focal tubular cell proliferation, multiple adenoma-like structures in layers of the renal papilla and in convoluted tubules were observed. However, the presence of OTA has also been found in bone marrow, skin, adrenal medulla, and cortex or myocardium. Different mechanisms of OTA action have been observed. The toxin causes G2/M phase cell cycle arrest, apoptosis, necrosis, inhibition of microtubule assembly, cell division processes, or protein syntheses. Several studies have shown that OTA induces reactive oxygen species (ROS) generation leading to oxidative stress and ROS-mediated apoptosis, as well as DNA adducts and DNA single-strand breaks. Shin et al. have demonstrated that OTA causes a decrease in glutathione (GSH), which is an important antioxidant. OTA treatment causes also ROS, lipid peroxides, and nitric oxide (NO) generation. Another study has shown that exposure of chickens to OTA decreased their number of lymphocytes, leukocytes, and throbocytes, modified the mucosal architecture of duodenum, jejunum, and also ileoceacal junction, caused a reduction in the intestinal TCR1, TCR2, CD4+, CD8+ lymphocyte population, and led to enterocyte apoptosis. The neurotoxic properties of OTA are the result of inhibition of human astrocyte cell line proliferation and differentiation. An increase in mitochondrial and cytosolic calcium concentration also appears as a consequence of calcium overload-cells' death occurs. OTA inhibits expression of glutamate transporter 1 (GLT1), glutamate aspartate transporter (GLAST), that leads to reducing glutamate absorption by astrocytes, which play a potential role in stability or induction of neurodegenerative diseases like Parkinson's and Alzheimer's, as well as neuron motor degenerations. Bhat et al. have shown that in neuronal (Neuro-2a) cell line, OTA causes ROS generation resulting in c-jun amino-terminal-kinase (JNK)-mediated caspase-dependent apoptosis. The IARC has classified ochratoxin as a group 2B carcinogen (possible human carcinogen). Various institutes and committees have developed limits on the OTA content in different types of food. The European Union has set a maximum limit of ochratoxin A at 3 µg/kg for cereal products, 5 µg/kg for unprocessed cereals, 10 µg/kg for dried fruits, and 15 µg/kg for spices. The European Commission established a maximum limit of OTA at 5 ng/kg in coffee beans, 10 ng/kg in instant coffee, 0.5 µg/kg in cereal-based food for infants and children, and 2 µg/kg in wines. Furthermore, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) has set the provisional maximum endurable consumption of OTA at 0.1 µg/kg b. w. per week.
## T-2 toxin
T-2 toxin is one of the most toxic mycotoxins, mainly produced by Fusarium sporotrichoides, F. poae, F. acuminatum, and F. equiseti, which are mainly found in regions with cold climate and wet storage conditions. The World Health Organization categorized T-2 toxin as an unavoidable contaminant in agricultural products, human food, and animal feed as early as in 1973. T-2 toxin naturally occurs in cereals, especially in wheat, oats, barley, and also in cereal-based products. It makes this toxin harmful to human and animal health. Contamination of overwintered wheat caused an outbreak of alimentary toxic aleukia (ATA) in the 1930s in the former Soviet Union and was related with other gastrointestinal problems. The etiology of Kashin-Beck disease (KBD) is still unclear, but it can be suspected that T-2 mycotoxin is the cause of this disease. In Chinese villages, which are endemic for KBD, the presence of T-2 toxin is relatively high with an average range of 78.91 µg/kg in wheat and 47.47 µg/kg in flour. Pleadin et al. have presented that in unprocessed cereals and cereal-based products coming from Croatia and Bosnia and Herzegovina, the proportion of contamination with T-2 ranged from 26.9% to 81.6%. T-2 mycotoxin molecular weight is 466.51 and it occurs as white, needle-like crystals. It has a distinctive tetracyclic sesquiterpenoid 12,13-epoxytrichothec-9-ene ring in common, and a 12,13-epoxy ring, which has a crucial function for the toxicity. The chemical structure is characterized by a hydroxyl (OH) group at the C- Food Additives (JECFA) has set the provisional maximum endurable consumption of OTA at 0.1 μg/kg b. w. per week.
## T-2 toxin
T-2 toxin is one of the most toxic mycotoxins, mainly produced by Fusarium sporotrichoides, F. poae, F. acuminatum, and F. equiseti, which are mainly found in regions with cold climate and wet storage conditions. The World Health Organization categorized T-2 toxin as an unavoidable contaminant in agricultural products, human food, and animal feed as early as in 1973. T-2 toxin naturally occurs in cereals, especially in wheat, oats, barley, and also in cereal-based products. It makes this toxin harmful to human and animal health. Contamination of overwintered wheat caused an outbreak of alimentary toxic aleukia (ATA) in the 1930s in the former Soviet Union and was related with other gastrointestinal problems. The etiology of Kashin-Beck disease (KBD) is still unclear, but it can be suspected that T-2 mycotoxin is the cause of this disease. In Chinese villages, which are endemic for KBD, the presence of T-2 toxin is relatively high with an average range of 78.91 μg/kg in wheat and 47.47 μg/kg in flour. Pleadin et al. have presented that in unprocessed cereals and cereal-based products coming from Croatia and Bosnia and Herzegovina, the proportion of contamination with T-2 ranged from 26.9% to 81.6%. T-2 mycotoxin molecular weight is 466.51 and it occurs as white, needle-like crystals. It has a distinctive tetracyclic sesquiterpenoid 12,13-epoxytrichothec-9-ene ring in common, and a 12,13-epoxy ring, which has a crucial function for the toxicity. The chemical structure is characterized by a hydroxyl (OH) group at the C-3 position, acetyloxy (-OCOCH3) groups at C-4 and C-15 positions, an atom of hydrogen at C-7 position, and an ester-linked isovaleryl [OCOCH2CH(CH3)2] group at the C-8 position. The metabolism of T-2 mycotoxin has been studied in various in vitro and in vivo experiments. The main biotransformation pathway of T-2 toxin is deacetylation of the C-4 acetyl group, leading to a conversion to HT-2 toxin. HT-2 toxin has been detected in isolated kidney microsomes, liver, and spleen of various animal models as the sole metabolite of T-2 toxin. Other reactions usually connected with metabolism of this toxin in mammals are oxidation (e.g., 3′-hydroxy-HT2 and 3′-hydroxy-T2), hydrolysis (e.g., neosolaniol, T2-triol, and T2-tetraol), de-epoxidation (e.g., de-epoxy-3′-hydroxy-HT2, de-epoxy-T2-triol, and de-epoxy-HT2), and glucuronide conjugation of biotransformation products such as HT-2 and neosolaniol. De-epoxidation is an essential detoxification mechanism and, with metabolic changes (e.g., conjugation) of the hydroxyl group at C-3, it has an effect on reducing the toxicity of trichothecenes. In vitro studies with African green monkey kidney cell line (VERO) and Chinese hamster ovary cell line (CHO) identified traces of T-2 triol, T-2 tetraol, and several other unknown metabolites. Studies with human blood cells have The metabolism of T-2 mycotoxin has been studied in various in vitro and in vivo experiments. The main biotransformation pathway of T-2 toxin is deacetylation of the C-4 acetyl group, leading to a conversion to HT-2 toxin. HT-2 toxin has been detected in isolated kidney microsomes, liver, and spleen of various animal models as the sole metabolite of T-2 toxin. Other reactions usually connected with metabolism of this toxin in mammals are oxidation (e.g., 3 -hydroxy-HT2 and 3 -hydroxy-T2), hydrolysis (e.g., neosolaniol, T2-triol, and T2-tetraol), de-epoxidation (e.g., de-epoxy-3 -hydroxy-HT2, de-epoxy-T2-triol, and de-epoxy-HT2), and glucuronide conjugation of biotransformation products such as HT-2 and neosolaniol. De-epoxidation is an essential detoxification mechanism and, with metabolic changes (e.g., conjugation) of the hydroxyl group at C-3, it has an effect on reducing the toxicity of trichothecenes. In vitro studies with African green monkey kidney cell line (VERO) and Chinese hamster ovary cell line (CHO) identified traces of T-2 triol, T-2 tetraol, and several other unknown metabolites. Studies with human blood cells have shown the metabolism of T-2 toxin to HT-2 toxin and neosolaniol as second metabolite by carboxylesterases activity. The amount of both metabolites was equal. In contrast, experiments with human liver homogenates have demonstrated HT-2 toxin as the only metabolite. The lipophilic nature of this toxin implies that it is easily absorbed through the gut, skin, and pulmonary mucosa. T-2 toxin impacts the vascular system, leading to dilation and swelling of micro vessels, as well as damage of the plasma membrane and the blood vessel wall. Central nervous system disorders caused by T-2 toxin poisoning cause lethargy, ataxia, and emesis in humans and animals. Various reports suggest that exposure to T-2 toxin can change the concentration of neurotransmitters in the brain. Dermal exposure to this toxin in mice increases the blood-brain barrier (BBB) permeability and it is related to the activation of matrix metallopeptidase 9 (MMP-9) and proinflammatory cytokines IL-1 α, IL-1β, and TNFα in periphery and in the brain. Chaudhary et al. have showed that T-2 mycotoxin exposure via percutaneous and subcutaneous route causes notable oxidative brain damage as a consequence of increased lipid peroxidation, depletion of hepatic glutathione, changes in antioxidant enzymes activity, and protein oxidation. have demonstrated that mice exposure to T-2 toxin during pregnancy and lactation can lead to an increase in the lipid content in young mice's liver tissues and serum. Furthermore, disruptions in bile acid metabolism may lead to lipid accumulation in the liver and as a consequence result in young mice's liver disfunction. The significant toxicological effects of T-2 mycotoxin is dermal irritation and wound forming. Skin exposure to this toxin induces a spectrum of damages from erythema to necrosis. Skin inflammation, fibroblast cell destruction in skin, and skin damages are similar to the detrimental effects of radiation. Inhibition of the DNA, RNA, and protein synthesis are considered to be the main cause of dermal damages. T-2 toxin has a toxic effect on the immune system. It decreases the production of IL-2 and the expression of plasma IFN-γ and can upregulate the mRNA expression of IL-1β, IL-6, and TNF-α, depending on the dose in RAW264.7 cell line, which is a model of mice macrophage cells. The European Commission has established a tolerable daily intake of of 100 ng/kg body weight for T-2 toxin.
## Deoxynivalenol
Deoxynivalenol (DON), also known as vomitoxin, is mainly produced by Fusarium graminearum and F. culmorum. These fungi are essential plant pathogens, which grow on field crops and cause a disease called Fusarium head blight (FHB). DON can contaminate various types of food or feed and unprocessed grains, especially in temperate regions. DON is one of the most frequently occurring mycotoxins in European food and feed. In addition, 25% of global crops production is considered to be contaminated with this toxin. The highest level of this toxin is observed in maize, wheat, and derived products. The chemical name of DON is 12,3-epoxy-3α,7α,15-trihydroxytrichothec-9-en-8-on. The molecular structure contains 3 free hydroxy groups (-OH), which are associated with its toxicity. It resembles colorless, fine needles, soluble in water and polar organic solvents (ethanol, methanol, chloroform, acetonitrile). DON remains stable in high temperatures, and at 150-170 - C, the toxin is not eliminated.
considered to be contaminated with this toxin. The highest level of this toxin is observed in maize, wheat, and derived products. The chemical name of DON is 12,3-epoxy-3α,7α,15-trihydroxytrichothec-9-en-8-on. The molecular structure contains 3 free hydroxy groups (-OH), which are associated with its toxicity. It resembles colorless, fine needles, soluble in water and polar organic solvents (ethanol, methanol, chloroform, acetonitrile). DON remains stable in high temperatures, and at 150-170 °C, the toxin is not eliminated. DON is called vomitoxin. The strong emetic/anorectic effects of DON are associated with two major mediators: proinflammatory cytokines and secretion of satiety hormones, which can activate receptors in the abdominal vagal afferent. The emetic effect was first observed in contaminated barley in 1972 in Japan. Biological toxicity of DON was described in various in vitro and in vivo studies. Vomitoxin induces physiological irregularities, including digestive disorders as well as reproductive and endocrine disruptions. The mechanism of toxicity of DON involves the inhibition of protein synthesis. The toxin interacts with peptidyl transferase enzyme, binding the 60S ribosomal subunit and then causes translation inhibition as well as ribotoxic stress. Another mechanism of toxicity is associated with the activation of a number of mitogen-activated protein kinases (MAPK), which are responsible for many effects of the toxin, including oxidative stress, apoptosis, inflammatory response, and endocrine disorders. It has been shown that DON-induced toxicity in intestinal epithelial cells is based on MAPK activation in tandem with a decreased expression of the tight junction proteins (TJP) and the loss of barrier function. Behrens et al.'s in vitro studies have shown that DON causes cytotoxic effect at the blood-brain barrier. A putative mechanism of action is based on increasing the cellular inflammation markers like MAPK and reducing the expression of claudins, which are significant to maintain the performance of TJP. It leads to reduced barrier integrity and, consequtively, increased permeability. A different study has demonstrated that DON causes DNA damage as a result of direct action of toxin and by mechanisms like formation of DNA adducts from free radicals. DON causes lipid peroxidation and malondialdehyde (MDA) formation. MDA reacts with deoxy guanosine and deoxyadenosine in DNA, later forming DNA adducts, mainly the mutagenic M1G (pyrimidol(1,2-a]purin-10(3H)-one). Additionally, it has been shown that DON exposition can lead to toxin accumulation in various products of animal origin such as eggs, milk, fat, and muscle. To reduce the risk of DON-induced effects, the JECFA has established a maximum daily limit of toxin intake at 1 µg/kg body weight. DON has been classified by IARC in group 3 as it did not show carcinogenic effects in humans.
## Patulin
Patulin is produced by fungi of the genera Aspergillus, Penicillium, Byssochlamys, and Paecilomyces. Thirteen species of Penicillium synthesize patulin among which are P. expansum, P. carneum, P. coprobium, P. clavigerum, P. dipodomyicola, P. glandicola, P. concentricum, P. gladioli, and P. griseofulvum. Aspergillus species like A. clavatus, A. longivesica, and A. giganteus are also patulin producers. Amongst Byssochlamys and Paecilomyces species, only B. nivea and P. saturatus can produce this mycotoxin. Aspergillus species produce patulin in warm and humid environments (tropical and subtropical areas), while Penicillium species are responsible for toxin secretion in lower temperatures. P. expansum is a potential toxin producer in an apple's pre-and post-harvest stage, whilst Byssochlamys nivea is related to patulin contamination in pasteurized fruit juices. Patulin is a natural contaminant of fruits and vegetables, including apples, apple-derived products, plums, grapes, pears, pineapples, peaches, and tomatoes. Predominantly, patulin contamination is associated with blue and soft rot, mainly caused by P. expansum. Humans are exposed to this toxin through consumption of contaminated food and beverages. Many countries have done research and investigated into contamination related to patulin in apple and apple products. Studies have shown that in Belgium organic apple juice reveals higher levels of toxin than conventional juices. In Portugal, 23% of apple-derived products and 69% of rotten apples are contaminated with patulin. In India, investigations have shown that patulin is noticeable in 24% of apple juice samples, and in 16% of samples, more than 100 µg/L of toxin content was present, while the maximum level of patulin in apple juice is set by the WHO at 50 µg/L. Patulin (4-hydroxy-4H-furo[3,2-c]pyran-2(6H)-one) is a polyketide lactone with low molecular weight. It is a white powder soluble in water and polar organic solvents (ethanol, methanol, acetone). It is also stable at high temperatures. Chemical studies have reported that the toxin is not eliminated from apple juice exposed to 80 - C for 20 min. (ethanol, methanol, acetone). It is also stable at high temperatures. Chemical studies have reported that the toxin is not eliminated from apple juice exposed to 80 °C for 20 min. Initially, patulin was described as an antibiotic, but its toxic effects on animals were later observed. Exposure to patulin results in various acute and chronic health effects, including agitation, pulmonary congestion, hyperemia, dyspnea, edema, ulceration, and intestinal inflammation. Patulin is highly toxic to the liver, gastrointestinal tract, kidneys, nervous system, and immune system. The toxicity of patulin induces cell damage and cellular processes disruption through electrophilic reactivity, resulting in formation of adducts with nucleophiles like amino (-NH2) and sulfhydryl (-SH) groups. The adducts created with GSH, lysine-, cysteine-, histidine-, or α-amino-acid containing proteins are covalently cross-linked compounds. Studies on kidney and intestinal cell lines have shown that patulin increases the level of intracellular ROS and also induces mitochondrial anion superoxide generation. Patulin is also able to inhibit the activity of different enzymes, like aminoacyl-tRNA synthetases and RNA-polymerases. Song et al. have demonstrated hepato-and genotoxic properties of patulin. The study has shown that mycotoxin causes serum alanine transaminase (ALT) and aspartate transaminase (AST) activity, induces lipid peroxidation, ROS generation, and decreases the GSH level in mice. Moreover, in bone marrow, patulin causes micronucleus and chromosomal aberration formation. Other in vitro studies have shown nephrotoxic properties of patulin by activating p38 and JUN kinase in the HEK293 cell line. Immunotoxicity of patulin has been confirmed in different studies. In mice, patulin increases a number of monocytes, NK cells and cytotoxic T cells, and decreases a number of lymphocytes and peripheral blood leukocytes. According to IARC, patulin is a group 3 carcinogen as it is not carcinogenic to humans. International agencies and institutions have introduced limits for patulin in various food products. WHO estimated the maximum limit of patulin at 50 μg/kg in apples, 50 μg/L in apple juice, and 10 μg/L in young children and infants' apple-based food. Initially, patulin was described as an antibiotic, but its toxic effects on animals were later observed. Exposure to patulin results in various acute and chronic health effects, including agitation, pulmonary congestion, hyperemia, dyspnea, edema, ulceration, and intestinal inflammation. Patulin is highly toxic to the liver, gastrointestinal tract, kidneys, nervous system, and immune system. The toxicity of patulin induces cell damage and cellular processes disruption through electrophilic reactivity, resulting in formation of adducts with nucleophiles like amino (-NH 2 ) and sulfhydryl (-SH) groups. The adducts created with GSH, lysine-, cysteine-, histidine-, or α-amino-acid containing proteins are covalently cross-linked compounds. Studies on kidney and intestinal cell lines have shown that patulin increases the level of intracellular ROS and also induces mitochondrial anion superoxide generation. Patulin is also able to inhibit the activity of different enzymes, like aminoacyl-tRNA synthetases and RNA-polymerases. Song et al. have demonstrated hepato-and genotoxic properties of patulin. The study has shown that mycotoxin causes serum alanine transaminase (ALT) and aspartate transaminase (AST) activity, induces lipid peroxidation, ROS generation, and decreases the GSH level in mice. Moreover, in bone marrow, patulin causes micronucleus and chromosomal aberration formation. Other in vitro studies have shown nephrotoxic properties of patulin by activating p38 and JUN kinase in the HEK293 cell line. Immunotoxicity of patulin has been confirmed in different studies. In mice, patulin increases a number of monocytes, NK cells and cytotoxic T cells, and decreases a number of lymphocytes and peripheral blood leukocytes. According to IARC, patulin is a group 3 carcinogen as it is not carcinogenic to humans. International agencies and institutions have introduced limits for patulin in various food products. WHO estimated the maximum limit of patulin at 50 µg/kg in apples, 50 µg/L in apple juice, and 10 µg/L in young children and infants' apple-based food.
## Zearalenone
## Zearalenone
Zearalenone (ZEA) is a nonsteroidal estrogenic mycotoxin, produced by Fusarium species such as F. graminearum, F. cerealis, F. culmorum, and F. equiseti. The main contamination source of ZEA are cereals, including maize, barley, oats, sorghum, and wheat, but also spices, milk, and beer. The toxin is synthesized at diverse stages of food processing, like vegetation, harvesting, and storage. Chemically, zearalenone is a macrolide, comprising a fourteen-membered lactone fused to 1,3-dihydroxybenzene. ZEA is resistant to UV light and stable in high temperatures. A heat study has shown that toxin decomposes by 3.2% at 100 °C for 15 min and 28.5% at 150 °C for 60 min. Two major pathways of ZEA biotransformation in animals are known. The first one is hydroxylation resulting in the formation of two stereoisomers-α-zearalenol (α-ZOL) and β-zearalenol (β-ZOL). The conversion occurs in different parts of the organism including porcine and bovine granulosa cells, swine intestinal mucosa, human intestinal Caco-2 cell line, and rat erythrocytes. The estrogenic potential of stereoisomers is different. α-ZOL is characterized by a high affinity for estrogen receptors and is more toxic than ZEA. Form β has much lower affinity for estrogen receptors and thus is nearly harmless. The next pathway is conjugation of ZEA and its metabolites with glucuronic acid. This process is catalyzed by uridine diphosphate glucuronyl transferases (UDPGT). These glucuronides are excreted into the bile and eliminated from the organism through urine and feces. Zearalenone is a phenolic resorcylic acid lactone and its chemical structure shows a resemblance to endogenous estrogen (17β-estradiol (E2)). As a result, ZEA has estrogen-like activity and is also able to competitively bind to the related receptors. As a consequence, the toxin causes estrogenic effects and induces reproductive disorders in livestock. Pigs are the more susceptible species among all domestic animals. In humans, the toxin can cause hyper estrogenic syndromes. Furthermore, ZEA possesses hepatotoxic, immunotoxic, and genotoxic properties. In animals, ZEA induces oocytes' death in the follicles and a lack of ovulation. The toxin inhibits the secretion of steroid hormones, disturbs estrogenic response on the preovulatory stage, and represses the maturation of mammalian ovarian follicles. In vivo studies have demonstrated that ZEA inhibits the growth of beneficial gastrointestinal microbiota. Additionally, the toxin induces an intestinal mucosal immune response, thus causing mucosal inflammation. Zearalenone also induces modifications in DNA methylation and expression of genes related to nuclear receptors and metabolic pathways like IGF1, HK2, PXR, and PPARγ in the breast cancer cell line. The JECFA has established a maximum tolerable daily intake (TDI) for ZEA at 0.5 μg/kg body weight. According to the International Agency for Research on Cancer (IARC), zearalenone is classified in group 3 as not carcinogenic to humans. ZEA is resistant to UV light and stable in high temperatures. A heat study has shown that toxin decomposes by 3.2% at 100 - C for 15 min and 28.5% at 150 - C for 60 min. Two major pathways of ZEA biotransformation in animals are known. The first one is hydroxylation resulting in the formation of two stereoisomers-α-zearalenol (α-ZOL) and β-zearalenol (β-ZOL). The conversion occurs in different parts of the organism including porcine and bovine granulosa cells, swine intestinal mucosa, human intestinal Caco-2 cell line, and rat erythrocytes. The estrogenic potential of stereoisomers is different. α-ZOL is characterized by a high affinity for estrogen receptors and is more toxic than ZEA. Form β has much lower affinity for estrogen receptors and thus is nearly harmless. The next pathway is conjugation of ZEA and its metabolites with glucuronic acid. This process is catalyzed by uridine diphosphate glucuronyl transferases (UDPGT). These glucuronides are excreted into the bile and eliminated from the organism through urine and feces. Zearalenone is a phenolic resorcylic acid lactone and its chemical structure shows a resemblance to endogenous estrogen (17β-estradiol (E2)). As a result, ZEA has estrogen-like activity and is also able to competitively bind to the related receptors. As a consequence, the toxin causes estrogenic effects and induces reproductive disorders in livestock. Pigs are the more susceptible species among all domestic animals. In humans, the toxin can cause hyper estrogenic syndromes. Furthermore, ZEA possesses hepatotoxic, immunotoxic, and genotoxic properties. In animals, ZEA induces oocytes' death in the follicles and a lack of ovulation. The toxin inhibits the secretion of steroid hormones, disturbs estrogenic response on the preovulatory stage, and represses the maturation of mammalian ovarian follicles. In vivo studies have demonstrated that ZEA inhibits the growth of beneficial gastrointestinal microbiota. Additionally, the toxin induces an intestinal mucosal immune response, thus causing mucosal inflammation. Zearalenone also induces modifications in DNA methylation and expression of genes related to nuclear receptors and metabolic pathways like IGF1, HK2, PXR, and PPARγ in the breast cancer cell line. The JECFA has established a maximum tolerable daily intake (TDI) for ZEA at 0.5 µg/kg body weight. According to the International Agency for Research on Cancer (IARC), zearalenone is classified in group 3 as not carcinogenic to humans.
# Conclusions
# Conclusions
Mycotoxins are poisonous, ubiquitous in chemical nature compounds, produced by various fungi species, whose occurrence in the food chain is inevitable and poses a serious problem on a global scale. Human exposure to mycotoxins is common due to food and feed contamination.. Mycotoxins contamination can result from poor hygienic conditions during harvest, transport, processing, or storage as well as infavorable climate. In addition to using good sanitation measures, it would be good practice to create awareness of the toxic effects of mycotoxin poisoning in humans and livestock. Fungal contamination poses a serious threat to human and animal health, which, depending on dose and time of exposure, can lead to various ailments. The intestine is the first barrier to food contaminants and the gastrointestinal tract is the first target of mycotoxins. Numerous studies focusing on the toxic actions of mycotoxins have shown that ingestion of fungal toxins may result in a variety of effects. It has been reported that mycotoxins are toxic to the nervous, immune, and reproductive systems. A summary of mycotoxin mechanisms of action on the human organism is presented in. Apart from human and animal health threat, contamination of agricultural crops with mycotoxins contributes to significant economic losses. The European Commission has estimated that 5-10% of global crop losses are caused by mycotoxins, causing the loss of 2.4 billion Euro in Europe. Future research should focus on generating data on epidemiological effects and long-term toxicity, especially in humans. The development of inexpensive mycotoxin detection instruments that are portable, reliable, and easy to use in the field is also an aspect to be noted. An interseting solution may be the development of new, genetically modified plants that can be resistant to fungal invasion. To maintain economic stability and agriculture, it may be beneficial to develop new protocols and strategies to compare the costs and benefits of different measures to combat fungal pathogens. which, depending on dose and time of exposure, can lead to various ailments. The intestine is the first barrier to food contaminants and the gastrointestinal tract is the first target of mycotoxins. Numerous studies focusing on the toxic actions of mycotoxins have shown that ingestion of fungal toxins may result in a variety of effects. It has been reported that mycotoxins are toxic to the nervous, immune, and reproductive systems. A summary of mycotoxin mechanisms of action on the human organism is presented in. Apart from human and animal health threat, contamination of agricultural crops with mycotoxins contributes to significant economic losses.
The European Commission has estimated that 5-10% of global crop losses are caused by mycotoxins, causing the loss of 2.4 billion Euro in Europe. Future research should focus on generating data on epidemiological effects and long-term toxicity, especially in humans. The development of inexpensive mycotoxin detection instruments that are portable, reliable, and easy to use in the field is also an aspect to be noted. An interseting solution may be the development of new, genetically modified plants that can be resistant to fungal invasion. To maintain economic stability and agriculture, it may be beneficial to develop new protocols and strategies to compare the costs and benefits of different measures to combat fungal pathogens. Another important issue in relation to mycotoxins is a detailed characterization of their molecular mechanism of action and its effect on animal and human health, which is necessary for the creation of strategy for prevention and therapy after poisoning.
Author Contributions: All the authors wrote the manuscript, read and approved the final draft of the manuscript, and agree to be personally accountable for their own contributions. Another important issue in relation to mycotoxins is a detailed characterization of their molecular mechanism of action and its effect on animal and human health, which is necessary for the creation of strategy for prevention and therapy after poisoning.
Author Contributions: All the authors wrote the manuscript, read and approved the final draft of the manuscript, and agree to be personally accountable for their own contributions. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
## Conflicts of interest:
The authors declare no conflict of interest.
## Abbreaviations
|
Toxic Environmental Factors and their Association with the Development of Dementia: a Mini Review on Heavy Metals and Ambient Particulate Matter
# Introduction
Dementia is the disease that has taken on unprecedented dimensions in recent years. The number of patients suffering from dementia is expected to increase dramatically. According to the World Health Organization, the number of people suffering from dementia is estimated to be tripled over the next 30 years (1). In the USA, in 2010, Alzheimer's disease was the fifth leading cause of death in the elderly. This increase is expected mainly in developing countries. Already, 58% of people with dementia live in low-and middle-income countries, but, by 2050 this will increase to 68%. Dementia affects the elderly, although the number of cases starting before the age of 65 is constantly increasing. In Europe, the number of people suffering from dementia is estimated to rise from 7.7 million in 2001 to
## Million in 2040.
A similar increase is projected in terms of the European cost for tackling the disease, which is expected to soar to 43% between 2008 and 2030, amounting to more than € 250 billion. Dementia is the third most expensive disease after cardiovascular disease and cancer, due to the increased costs required for its prevention, diagnosis, treatment and management.Laboratory of Histology and Embryology, Aristotle University of Thessaloniki, Thessaloniki, Greece;
The most distinct characteristic of dementia is a gradual decline in mental capacity, with memory loss appearing as the most severe symptom either in the early stages or in advanced stages of the disorder. Memory function has a catalytic role for the formation of personality, since it allows communication, the acquisition of skills and survival in the best possible way. Depending on the degree of disturbance of the patient's daily life, the disease is divided into three degrees of severity. In the initial stages, the mild degree that limits the daily life of the patient, his domestic and social activities. The degree of moderate severity, that patients are in need of intermittent supervision. In the final stage, the severity of the disease requires constant care and supervision of the patient. The most common form of dementia is Alzheimer's disease (65-70% of all cases), followed by Lewy body dementia and vascular dementia.
The environment in which we live is considered responsible for many diseases that appear from time to time and afflict all societies. The first reports of the connection between the environment and the health of individuals have been made by the father of medicine, Hippocrates.
Hippocrates was the first to deal on how the natural environment affects human health. Studying his work, the reader realizes that Hippocrates considers the human body an integral part of the environment.
Environmental risk factors may have a key role in accelerating or slowing the onset and progression of dementia. Environmental risk factors include mostly air pollution and chronic exposure to various metals, as the aggressive rate of anthropogenic activity releases excess amount of pollutants into the environment. As a result, a large group of the population is exposed not only to basic metals, such as copper and aluminum, but also to toxic metals, more specific, mercury and lead which disrupt cellular homeostasis. This study is structured in a specific way in order to present the association of dementia with these heavy metals. The implication of ambient particulate matter to Dementia is also analyzed.
## Aim
The purpose of this bibliographic work is «ecological consciousness» of modern societies to be awaken, to identify the harmful environmental factors and to highlight their involvement in the causal pathogenesis of the most debated disease, dementia.
# Methods
During the review of literature, studies were selected by searching in reputable scientific databases such as PubMed BioMed Central, Science Direct, NCBI.
The eligibility criteria required a recent publication year, namely publications of the last 5 years, with the exception of a few previous publications of key information. The status of the first author was taken into account, as well the duration and follow-up of the clinical study, the geographical location that took place, its design, the size of the sample and the year in which it was conducted. Full-text publications were preferred.
The exclusion criteria were the year of the study, the size and the monitoring time of the sample. Posts that re-stricted the ability to display full text were also excluded. However, the restriction on English language publications may have led to the exclusion of possible relevant studies in other languages, which contained important information.
The aim of the study is not to describe dementia and brain diseases, but to cite the findings regarding the environmental risk factors. An additional goal is to study the most recent research on the aforementioned study topics and compare them with the knowledge gathered so far, which has been collected by experts, in order to arrive at a poised conclusion in regards to brain toxicity of heavy metals.
## Toxic environmental factors
Exposure of individuals to heavy metals such as aluminum, copper, mercury and lead has been shown to have serious effects on brain function. Below we will study the effect of the main environmental toxic factors which appear to have a direct effect on the occurrence of different types of dementia.
## Aluminum (al)
Aluminum is not found on a pure form in nature, but in the form of minerals from which it is extracted. It is the third most abundant element in the earth's crust and the first of the minerals. The British Davy, in 1807, was the first to support the existence of aluminum, while the Dane Oerstead managed to isolate the metal. Although aluminum is abundant in the earth, its bioavailability is limited due to its insoluble nature. The content of aluminum that can be found in water and underground is extremely low. The rare exposure to aluminum maintains a low concentration load in the biosystem, however in pathological conditions, its concentration is increased.
From a toxicological aspect, the main aluminum compounds are considered to be aluminum oxide Al 2 O 3 , aluminum hydroxide Al(OH)and KOH and NaOH. Aluminum is absorbed primarily by the lungs and to a lesser extent by the gastrointestinal tract. After its absorption it is found in the blood stream, bound to proteins, in a percentage of 50% and mainly with albumin and transferrin. Insoluble aluminum salts do not display toxicity in contrast to soluble aluminum salts which show significant toxicity.
Exposure to high levels of aluminum leads to neuronal degeneration. In Alzheimer's disease, a high concentration of aluminum is observed in the degenerate neurons. Injection of aluminum into the cerebrospinal fluid of experimental animals has been found to cause progressive encephalopathy and degenerative lesions of nerve fibers, similar to those observed in people with Alzheimer's disease. This poses a strong evidence that aluminum is one of the causes of neurodegenerative disorders. Aluminum compounds combined with mercury compounds, can act synergistically in the degeneration of glial cells.
The majority of human exposure to aluminum derives from food, drinking water and alcoholic beverages. The use of certain pharmaceutical and cosmetic products, is the main source of aluminum since Al is one of the key ingredients of deodorants, and various personal hygiene products (10). Its intake is estimated at 10-100 mg per day, mainly from meat, cereals, dairy products and fish. Concen-trations have also been reported in tea leaves, with higher concentration levels in young leaves, in the autumn and summer. Other sources of Al are food additives, containers, cookware and food packaging. Dietary intake of aluminum is minimal compared to amounts consumed through other sources, such as the use of aluminum cookware and utensils for cooking and storing food.
Aluminum is responsible for Neurochemical changes. Some of the cellular processes involved in neurodegeneration are oxidative stress, apoptosis and the formation of neurofibrillary tangles (NFT). A possible mechanism is the increase in the nitrification of tyrosine residues in cytoskeletal proteins such as tau proteins, mediated by peroxide dissolution, leading to the formation of neurofibrillary tulips. The presence of nitrotyrosine has been demonstrated in Alzheimer's disease proving its involvement in oxidative damage. Through oxidative stress it leads to neurodegeneration, and also causes the apoptosis of astrocytes, which contributes to the transportation of fluids and ions from the extracellular space to the blood vessels. With the fluids' loss, the neurotrophic support is reduced. This consequents to neurotic death. Apoptosis is considered to be the main toxicity mechanism of aluminum in nerve cells. Mitochondrial dysfunction marks the initial stage of aluminum neurotoxicity and is a prerequisite for this vicious cycle.
A possible mechanism for the initiation and development of these diseases could be the disturbed homeostasis of basic metals and the appearance of unfolded or incorrectly folded proteins which are basically the main culprits to several diseases. Aluminum has the ability to bind to negatively charged phospholipids in the brain, which contain polyunsaturated fatty acids and make them vulnerable to reactive oxygen species (ROS). Aluminum stimulates lipid peroxidation mediated by iron ions in the Fenton reaction, which causes ROS production, formation of trivalent iron ions, loss of cellular homeostasis and oxidative stress. The above disorders are observed in glial, astrocyte and microglial cells. Aluminum is an inhibitor of the cells' antioxidant enzymes which support the defense. The charge of the peroxide is neutralized by trivalent aluminum ions Al 3+ forming the Al-O 2 complex which leads to the increase of the oxidizing capacity of O 2.
Regarding inflammatory reactions, aluminum induces the expression of transcription factor NF-κB, the precursor of interleukin-1β and phospholipase A2, which are involved in proinflammatory and pro-apoptotic signaling mechanisms. Aluminum increases pro-inflammatory cytokine levels such as TNF-a and IL-1a.
In addition to the above, aluminum is also involved in the apoptosis of nerve cells, as mentioned. It releases cytochrome c from mitochondria, reduces an anti-apoptotic molecule of Bcl-2 nerve cells in the mitochondria and endoplasmic reticulum, induces the displacement of the Bax pro-apoptotic protein and the activation of the executive caspase-3. These events mark the beginning of apoptosis. Aluminum has also been shown to lead to apoptosis by activating the SAPK / JNK kinase signal transduction pathway (stress-activated protein kinase or c-jun N-terminal kinase).
Aluminum, in its oxidizing state as Al 3+ , has a high affinity for organic and inorganic phosphates, carboxyls and deprotonated groups. Al 3+ ions display a strong positive charge and a relatively small ionic radius, compared to other metal ions such as Ca 2+ , Zn 2+ and Na + . Thus it is bound strongly to metal-binding amino acids, such as histidine (His), tyrosine (Tyr), arginine (Arg), or phosphorylated amino acids and acts as a crosslinking agent. The binding of Al 3+ to phosphorylated amino acids is strong, thus the accumulation of highly phosphorylated cytoskeletal proteins, including neurofibrillary and microtubule-associated proteins (MAPs).
Aluminum has also been shown to accumulate in neurons, where it inhibits sodium (Na) + / calcium (Ca) 2+ exchange mechanisms and thus intervenes to the overload of the mitochondria with cytoplasmic Ca 2+ . Increased levels of intra-mitochondrial Ca 2+ leads to the opening of Mitochondrial Trifunctional Protein (MTP), which is considered to be the main trigger for neuronal damage. This is followed by the release of cytochrome c, which is a mitochondrial protein and an important apoptotic factor. 4.1.1. Recent research on Aluminum Based on the above, aluminum is fairly regarded to be a neurotoxic agent. However, the association of aluminum with the etiology of various serious neurological disorders, such as Alzheimer's disease, remains unclear. Despite this uncertainty, there are several epidemiological reports in the scientific literature regarding exposure to aluminum and the risks of neurological disorders. An important reason for this uncertainty is the small number of examinees, due to the ethical concerns of the tests performed on humans.
There have been many studies in mammals, especially to those that have been exposed to aluminum throughout their lives, so that the effects of aluminum can be fully observed. Exley C, et al, in their recent study on human brain tissue, argue that high aluminum concentration in the brain is not an inevitable result during the process of aging. Wang, Z., et al, in a meta-analysis of 10,567 people, found that chronic aluminum exposure was associated with an increased risk of Alzheimer's disease. In contrast, an earlier study by Forster DP, et al, found no significant association between aluminum exposure and the incidence of Alzheimer's disease. Also, in a meta-analysis, Virk SA, et al, do not support the causal role of aluminum in the pathogenesis of Alzheimer's disease.
## Copper (cu)
As the brain displays low levels of antioxidant activity and becomes vulnerable to oxidative stress, free copper concentration levels must be kept minimum in order for a neuronal integrity to be sustained.
Copper contributes to the maintenance of life, the development and function of the nervous system, the formation and maintenance of myelin and finally is actively involved in various biochemical cycles, as it forms a building block of several proteins and enzymes. Its concentration in high quantities causes toxicity. The organs of copper accumulation are mainly the liver and the brain. Copper toxicity has an important role in the pathogenesis of Alzheimer's disease and in the general loss of cognitive function. It has also been shown that people who consume high-fat foods in combination with taking copper supplements are at risk of memory impairment.
Since the beginning of the 20th century, there has been a significant increase in the use of copper for the construction of water pipes in developed countries. During World War I and World War II, its use was limited. After 1950 the use of copper pipes took off, reaching 90% of the plumbing of homes in the United States. At the same time, the Alzheimer's disease epidemic began in developed countries.
Another source of inorganic copper that is common in the developed world is the intaking copper through dietary supplements. A high-fat diet combined with elevated copper levels is considered to be a significant risk factor. Copper has been linked to oxidative stress and the destruction of nerve cells. Elevated homocysteine levels are also a risk factor for Alzheimer's disease; homocysteine interacts with copper during the oxidization of cholesterol and results to neurons' destruction as well. The apolipoprotein E4 allele poses a risk factor for Alzheimer's disease, while apoE2 is a protective factor and apoE3 remains neutral. Certain alleles of the ATP7B protein gene, also known as the Wilson disease gene, increase the risk of Alzheimer's disease. In Wilson's disease the cytotoxic effect of copper is achieved by the production of free radicals, which destroy cell membranes by fat peroxidation. ATP7B gene controls free copper levels, which have been shown to be elevated in people with Alzheimer's disease.
A high concentration of copper and zinc has been found at amyloid plaques deposited in the brain tissue. There is also a high concentration of copper and zinc in the cerebrospinal fluid in people with Alzheimer's and Parkinson's disease. Maintaining copper homeostasis in the brain at normal levels is important, since copper can catalyze O2 activation by creating reactive oxygen species (ROS). The result of these potent oxidizing agents is a variety of Aluminum (Al) Exley C, et alfamilial Alzheimer's disease, autism spectrum disorder and multiple sclerosis. Detailed statistical analyses showed that aluminium was significantly increased in each of these disease groups compared to control tissues. We have confirmed previous conclusions that the aluminium content of brain tissue in Alzheimer's disease, autism spectrum disorder and multiple sclerosis is significantly elevated. Further research is required to understand the role played by high levels of aluminium in the aetiology of human neurodegenerative and neurodevelopmental disease.
## 2020
"Al is not an inevitable aging factor for human brain" NO oxidative damage including lipid peroxidation, DNA oxidation, protein oxidation, and advanced glycosylation of end products (AGEs). Neurodegenerative disorders exacerbate the already increased oxidative stress, which is an early event in the development of Alzheimer's disease and is associated with normal aging. Oxidative stress is enhanced by the pathological increase in the interactions of divalent copper oxide with the Aβ peptide, and then the produced interactions promote the production of reactive oxygen species. Studies focusing on the interaction of divalent copper oxide with t-protein or phosphorylated t-protein and their portions are rather limited. According to in vitro studies, the ability of t-protein to bind to copper also contributes to oxidative stress, as does the binding of β-amyloid peptide to copper.
A high-fat diet, along with inorganic copper intake, is considered an important risk factor for Alzheimer's disease. This hypothesis is difficult to be studied in humans. The studies are only applied experimentally to animal models, on the permitted extension, and therefore they are limited. The degree of absorption of copper depends mainly on its chemical state but also on the presence of compounds such as zinc and iron. Zinc has a competitive effect against monovalent copper. Zinc intake seems to have a beneficial effect, as its deficiency poses a risk factor for the disease (21).
## Recent research on copper
Regarding exposure to copper, several studies highlighted the role of copper as a cofactor in increasing the activity of β-amyloid protein toxic plaques and tau protein deposits. Disorder of copper homeostasis seems to promote the onset of Alzheimer's disease, as in its occurrence, copper levels are found to be elevated. At normal levels of nutritional intake, copper, as an essential trace element, helps to enhance brain function.
Yu F, et al, observed that divalent copper enhances the effect of β-amyloid protein on microglial activation and subsequent neurotoxicity. The study of Xu J, et al, reinforces the view that low copper levels are consistent with the cause of neurodegeneration and may well contribute to the pathogenesis of Alzheimer's disease. They also claim that copper therapy could help treat Alzheimer's disease. In conclusion, the question of whether excess or insufficiency of copper levels is responsible for the pathogenesis of Alzheimer's disease remains controversial (24).
## Mercury (hg)
Mercury is one of the most toxic natural elements, it is ten times more neurotoxic from lead and causes a numerous health problem. There is no safe threshold level of exposure and ideally it should not be detected as it does not provide anything beneficial to the human body. The half-life of mercury is 30-60 days, while in the brain it reaches about 20 years. The binding of elemental mercury to the brain after oxidation is due to selenium, which probably contributes to the retention of mercury for a longer period of time.
Studies have shown an association between mercury and Alzheimer's disease. Neurotransmitters such as acetylcholine, serotonin, dopamine, glutamate-one of the most important excitatory neurotransmitters of the central nervous system-and norepinephrine are inhibited in patients with Alzheimer's disease. The same inhibition occurs in conditions of mercury toxicity. In Alzheimer's disease, there is a malfunction in the production of enzymes β-secretase, γ-secretase, cyclooxygenase-2, cytochrome-o-oxidase, protein kinases, monoamine oxidase, nitric oxide synthetase, acetylcholine transferase and caspases. These dysfunctions can be attributed to mercury toxicity. Immune and inflammatory responses to Alzheimer's disease also occur when cells are exposed to mercury. Mercury can inhibit DNA synthesis in the hippocampus, and has been linked to genetic mutations in the presenilin 1 and 2 genes found in Alzheimer's disease. Mercury toxicity causes disorders of minerals and vitamins, such as calcium, copper, iron, magnesium, selenium, zinc and vitamins B1, B12, E and C, a situation encountered in patients with Alzheimer's disease. Aluminum seems to increase mercury's toxicity.
Organic forms of mercury can easily penetrate bloodbrain barrier and enter the neurons, resulting in neuronal loss and free entry of toxic metals and substances into the brain. Methylmercury, alongside with elemental mercury, leads to the extravasation of plasma proteins and to a greater extent of degenerative lesions of the sensory nerves, compared to motor nerves, possibly due to the disturbance of cellular metabolism. Exposure to relatively low concentrations of mercury in the air as vapor, has been shown to inhibit the polymerization of tubulin in the brain, which is a necessary protein for the formation of microtubules.
Mercury accumulates in all nerve cells, has a devastating effect on energy production and can damage the detoxification process, causing either death or chronic malnutrition in the cells. Exposure of cells to mercury leads to changes in membrane's permeability, changes in the macromolecular structure and DNA destruction. It is also blamed for causing oxidative stress and mitochondrial dysfunction, which can lead to changes in calcium homeostasis and increased lipid peroxidation. Evidence suggests that mercury could be a causative agent of Alzheimer's disease. Further investigation is needed.
## Recent research on mercury
Geier DA, et al, in their study provide epidemiological data, which linked the increase of ethyl-mercury levels in the blood with the cognitive impairment of the elderly. Also, Sun YH, et al, conducted a study in which they discovered that individuals who were exposed to mercury via dental amalgam, had an increased ratio of 1,132 on developing Alzheimer's disease, in comparison to those who were not exposed at all.
## Lead (pb)
Various epidemiological researches and studies in animal models prove the etiological association between Alzheimer's disease and lead exposure. Lead appears to be one of the environmental factors causing the levels of the precursor amyloid protein APP and β-amyloid Aβ to increase. Exposure to lead in prenatal and early childhood has been shown to increase biomarkers involved in Alzheimer's disease in adulthood. In particular, exposure at an early age is associated with a decrease in H3K9Ac histone over time.
A possible mechanism by which environmental factors can alter the expression of Alzheimer's predisposing genes, is via epigenetic routes. Epigenetic mechanisms refer to genome change, affecting gene expression without altering the underlying DNA sequence. Animal studies showed that lead may be associated with epigenetic changes including DNA methylation and histone modifications.
Lead is highly toxic and bioaccumulative. Even exposure to acceptable lead levels could cause neurophysiological dysfunction and reduce peripheral nerve conduction rate. Exposure to lead is associated with severe cognitive and learning deficits. Lead is a risk factor for increased hippocampal gliosis, a state associated with the development of Alzheimer's disease. Rats exposed to lead during their early life showed increased amyloidogenesis and senescent plaque accumulation.
Inorganic lead behaves like calcium, mimicking its action in various systems. It replaces calcium ions and ensures the ability to penetrate the blood-brain barrier, resulting to accumulation in astroglial cells. It destroys immature nerve cells and prevents the formation of myelin. In addition, it disrupts the release of neurotransmitters, such as protein kinase C, which, among other basic functions, is involved in learning and memory processes. Lead also causes iron deficiency and anemia, exacerbates calcium and copper deficiency and competes with zinc, mainly in its enzymatic functions.
## Recent research on lead
The findings show that lead exposure has a long-term effect and may increase the risk of Alzheimer's disease. The presence of lead in the blood, results in its intervention in many organs and functions of the body, but also to the central nervous system, which is by far the most vulnerable. Zhao ZH, et al, reported in their study that rats exposed to lead showed deficits in work, as well as impaired memory and learning. In an earlier study, Schafer JH, et al, reported that in a large, general population sample, high levels of lead in the blood were found to be related to homocysteine levels. According to the data, both lead and homocysteine are associated with an increased risk of dementia.
## Ambient particulate matter (pm)
The air pollution particles, which belong to the category of submetric particles (UFT, Ultrafine Particles) with a diameter of less than 0.1 μm, display high velocity and penetrate the capillaries and the brain. Chronic exposure of the individual to these particulate matter, leads to inflammation and causes cell damage.
Inhalation of gases and airborne particles causes inflammatory responses, increasing microglycemic activity and production of active oxygen forms. Particles less than 0.1 μm in diameter reach the brain directly through the olfactory bulb to the cortical and subcortical centers of the frontal lobe and cause damage. Inevitably, a vicious cycle of inflammation, capillary damage and further silent inflammation is developed, and an increase in pro-inflammatory cytokines is observed, which promotes astrogliosis and neuronal death.
## Recent research on ambient particulate matter
Various studies have shown that PM2.5 particulate matter, in addition to being associated with a wide range of diseases, may also be responsible for a variety of neurode-generative diseases, such as Alzheimer's disease. Chen H, et al, in a cohort study of a large population group, found that residents living near busy highways had a higher incidence of dementia. reported strong correlations between long-term exposure to PM2.5 and neurodegenerative diseases such as Alzheimer's disease. Finally, Wu CH, et al, found that long-term exposure to PM10 or ozone was significantly associated with an increased risk of Alzheimer's disease and Vascular Dementia (2).
## Perspectives and challenges of future research
Dementia is a degenerative disease that could be treated, if not avoided, by minimizing exposure to various risk factors. From this standpoint we made an attempt to study some of the toxic agents which pose a great threat for developing neurodegenerative diseases. More specific, we put "under the microspore" heavy metals and ambient air pollutants by comparing old and new studies.
Due to the unprecedented acceleration of urban development, heavy metals are the main culprits of toxicity in the environmental systems. Their impact on public health goes hand to hand to the probable sources of pollution, especially in urban areas. Future research should emphasize on the connection among the prevalence of dementia and the probable sites of heavy metal contamination. Zhao L., et al, observed that industrial activity in Southeastern China led to an increased level of heavy metals in the soil-rice system. Farmland soils are displaying an alarming accumulation of Cu, Zn and Pb, due to fertilizers and urban pollution, and that could lead to hazardous levels of heavy metals uptake in biological systems. An intriguing study found out that atmospheric dust carries heavy metal particles at various sizes, and this kind of contamination could be detected by measuring the contamination in urban road-deposited sediments. These observations prove that future studies should focus on the source of contamination and the manifestation of dementia. A distinct well-designed questionnaire might be helpful in order to clarify whether or not the participants have been exposed to heavy metals and if this kind of exposure is the main reason for developing the disease. As there are innumerous sources of contamination, future researchers should try to evaluate the participants by their work and living places, their everyday habits, the sources of food, vitamins uptake, probable hazardous spots and many other factors.
Ambient air pollutants are also a major problem regarding public health and a significant risk factor for developing dementia. Human activities are responsible for the soaring levels of air pollution, especially in low-income countries, which suffer severe consequences. It has been calculated that air pollution's death toll rises up to seven million deaths annually. Similarly to heavy metals, future researches should establish an undeniable connection among the prevalence of dementia and polluted sites. Governments and scientists should collaborate in order to achieve a declining rate of air pollution. Society should also be educated on their everyday habits in order to minimize the pollution. Meta-analyses and epidemiological studies of large scales, might be able to persuade the public for the danger of developing dementia due to atmospheric contamination. This should motivate each one to adopt environmental consciousness, in an attempt to protect public health. Apart from review studies and surveys, a more direct approach to this investigation could be the processing of tissue and blood samples of individuals, from a designated group of patients. However, this approach displays a limitation in regards with the number of participants, since it is rather challenging to gather a significant amount of people and receive samples for further study.
A concise presentation of the findings regarding the relation among Dementia and the highlighted toxic factors is presented at Table 1. In the case of Mercury, Lead and Ambient Particulate Matter, there is no doubt that they pose a serious threat for developing dementia. Future studies should focus on presenting further evidence on the matter. Potential findings could be beneficial not only to health specialists and researchers who want to treat and cure these kinds of diseases, but also to the society, since they could act as a protective shield against various risk factors that threaten public health. Aluminum and Cooper on the other hand, displayed a contradictive behavior in regards with their effects on dementia. The scientific community has studied thoroughly the effects of Aluminum and Cooper on humans but their relation to these neurodegenerative diseases remains obscure. Hopefully, future studies will shed light on this matter and add extra knowledge on this long-lasting arduous effort of scientists.
# Conclusion
Given the enormous dimensions of the socio-economic impact caused by the increased amount of people suffering from dementia, some of the basic environmental risk factors were investigated. The aim of the investigation was to avoid or minimize the unpleasant effects on human health by analyzing in depth those environmental culprits who are responsible for the incidences of this disease.
The data collected suggest that elevated levels of particulate matter in the atmosphere and the accumulation of heavy metals in the body disrupt cellular metabolism, antioxidant defenses and immune responses, leading to the onset and progression of the disease. However, in some cases, writers found rather contradictive results. These were the cases of aluminum and copper. Especially for aluminum, regardless its conspicuous negative effects on brain and neuron system, many researchers could not find enough evidence to support its correlation to Dementia. For copper, results were similarly contradictive. On the other hand, mercury, lead and ambient air pollutants were clearly associated to Dementia, as was expected due to their neurotoxicity. However, further robust, time-consuming studies with repeated environmental exposure evaluation are needed in order to confirm the stated evidence and clarify the picture regarding aluminum and copper.
- Acknowledgements: This research was part of an MSc thesis that approved by Aristotle University of Thessaloniki, Medical School.
- Authors contribution: Antoniadou F (writing, data acquisition, design), Papamitsou T (conceptualization, analysis, writing, approving the final manuscript), Kapoukranidou D (conceptualization, analysis, critically revised), Kavvadas D (writing, data acquisition, drafting), Sioga A (analysis, critically revised), Papaliagkas V (analysis, critically revised).
- Conflict of interest: The authors declare no conflict of interest.
- Financial support and sponsorship: None received. |
microRNA Expression Profile in the Vitreous of Proliferative Diabetic Retinopathy Patients and Differences from Patients Treated with Anti-VEGF Therapy
[formula] miRNA Control PDR hsa-miR-142-3p 30% 80% hsa-miR-149-5p 70% 60% hsa-miR-185-5p 20% 70% hsa-miR-197-3p 60% 70% hsa-miR-199a-5p 70% 20% hsa-miR-204-5p 70% 70% hsa-miR-20a-5p 10% 70% hsa-miR-21-5p 80% 80% hsa-miR-222-3p 90% 80% hsa-miR-223-3p 50% 80% hsa-miR-23a-3p 50% 80% hsa-miR-23b-3p 20% 70% hsa-miR-320a 70% 80% hsa-miR-324-3p 90% 90% hsa-miR-326 100% 60% hsa-miR-329-3p 100% 100% hsa-miR-345-5p 90% 90% hsa-miR-361-5p 100% 100% hsa-miR-362-5p 100% 100% hsa-miR-373-3p 70% 70% hsa-miR-376c-3p 80% 50% hsa-miR-454-3p 80% 80% hsa-miR-491-5p 80% 100% hsa-miR-502-5p 80% 80% hsa-miR-509-3p 80% 60% hsa-miR-595 70% 70% hsa-miR-598-3p 100% 100% hsa-miR-615-3p 80% 70% hsa-miR-662 0% 70% hsa-miR-663a 90% 90% hsa-miR-760 70% 100% hsa-miR-765 60% 50% hsa-miR-877-5p 80% 60% Supplemental [/formula]
[table] Table 2: miRNAs regularly expressed in the vitreous of control and PDR patients. miRNAs found to be expressed in ≥70% of all samples in one condition were considered to be regularly expressed and are marked in bold (miRNAs regularly expressed in Control: 24; miRNAs regularly expressed in PDR: 26). [/table]
[table] Table 3: miRNA levels in vitreous of control (CON) and PDR patients in the three cohorts. Data are presented in absolute amounts with 1st and 3rd quartile values. The Mann-Whitney U test was used to assess statistical differences between groups. [/table]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.