text
stringlengths 1
6.27k
| id
int64 0
3.07M
| raw_id
stringlengths 2
9
| shard_id
int64 0
0
| num_shards
int64 16
16
|
|---|---|---|---|---|
equality of rights and status.In the context of the inheritance of the people of different religions, Asgar Ali Engineer assumed that it is allowed in the present situation based on maslahah.The conception of kāfir (non-Muslim) as formulated by classical scholars is considered irrelevant applied in the present condition. 24 the study of Islamic law, law is divided into two major areas: 'ibādah (worship) and mu'āmalah (social relationship).Ibadah is rules related to the relationship between man and his God, such as prayer, fasting, going on pilgrimage, and others, while mu'āmalah includes such activities as buying and selling, lease, divorce, marriage, inheritance, criminal laws and others. According to Asgar, the verses pertaining to the matter of ibadah can be put in action without reinterpretation in understanding the related verses.In contrast to worship, inheritance is included in mu'āmalah (social relations). Inheritance law in Indonesia generally still uses three legal bases, firstly, inheritance law under Islamic law, customary law, and western law.The absence of nationally binding law resulted in the implementation of inheritance law highly dependent on the choice of its citizens.Although law is not imperative, but the reality showed that the Presidential Instruction-based Compilation of Islamic Law is almost used by Religious Court judges in deciding cases, and is made as a reference by the officials of the Office of Religious Affairs and some members of the community. In the context of Islamic inheritance law, the reform of Islamic family law was first marked by the enactment of Law No. 1 Year 1974 on Marriage.Several years later, Islamic Law | 3,069,700 | 55842779 | 0 | 16 |
Compilation (KHI) was compiled through Presidential Instruction No. 1, Year 1991 which is materially used by the Religious Courts to resolve cases related to marriage, inheritance and endowment. In relation to the rights of inheritance for non-Muslims, KHI refers more to the opinions of the classical "Ulama" who argued that the religious differences between a testator and the heirs become a barrier to the inheritance process.This can be read in article 171, point b, which states "The testator is a person who, at the time of his or her death or declared dead by a judgment of the Islamic Court, leaves the inheritors and the property". In the same article 171 point c states "Heirs shall be those who at the time of death have a blood or marriage relation to the heirs, Muslims, and are not hindered by the law to become the heir".The religiousity of a person may be determined by his/her identities, and this is evident in article 172 which reads "Heirs are considered to be Muslims if they are known from identity card or recognition or practice or testimony, while the religion of newborns or immature children according to their father or environment." The provisions in the Compilation of Islamic Law (KHI) are very clear that the right of inheritance is automatically severed when it comes to religious differences.The rules in KHI are based entirely on the opinions of the 'classical' Muslim scholars, especially al-Shafi'i.Even in the Circular Letter of the Bureau of Religious Justice, 18 February 1958 Number B/ I / | 3,069,701 | 55842779 | 0 | 16 |
735, material law which is made as guidance in the KHI is sourced on 13 (thirteen) Shafi'ite books of Islamic jurisprudence. The dominance of classical scholars' opinion in KHI encouraged some Muslim thinkers in Indonesia to change the material in it because it is considered irrelevant to the this contemporary situation.The thinkers proposed the Counter Legal Draft of the Compilation of Islamic Law (CLD-KHI). 25This text according to the drafting team offers a number of reforms of Islamic family law compiled in the Islamic Marriage Law Bill, the Law of Islamic Inheritance, and the Law of Islamic Endowment. The ideas of Islamic family laws in KHI which has been applied in Indonesia firmly rests its opinion on the classical scholars especially al-Shafi'i.While the ideas in CLD-KHI were firmly inspired by the ideas of contemporary Muslim thinkers such as Abdullah Ahmad An-Na'im, Asgar Ali Engineer, and other Muslim contemporary thinkers. Conclusion From the above discussion, it can be concluded that the controversy on the issue of inheritance of the people of different religions can be seen in three points.First, the classical scholars argued that a Muslim is entitled to inherit non-Muslims, but not vice versa.Second, majority of Muslim jurists represented by four schools of laws-Hanafi, Maliki, Shafi'i, and Hanbali-agreed that Muslims cannot inherit each other with non-Muslims.Third, non-Muslim status does not become a barrier to inheritance because it contradicts the unversal value of the Quran. The differences of opinion about inheritance rights are preceded by disagreements about the terms of kafir (infidels) and murtad (apostates).According to the | 3,069,702 | 55842779 | 0 | 16 |
classical scholars, kufr refers 25 The idea of CLD-KHI arose after 2003. the Ministry of Religious Affairs submitted the draft Law on Applied Religious Court (RUU HTPA) to the House of Representatives (DPR).This This draft law refines the materials of the President Instruction based on KHI and improves its status from Presidential Instruction to Law.In response to the draft law, on 4 October 2004 the Working Group on Gender Mainstreaming of the Ministry of Religious Affairs (Pokja PUG Depag) launched a draft on Islamic law formulation called CLD-KHI. to the act of associating God with things like idolatry, apostasy and atheism.Contemporary thinkers such as An-Na'im and Asgar Ali Engineer said that the pagan conception must be contextualized.An-Na'im said, if there is a content of the shariah which is discriminatory, then its existence must be abrogated (mansūkh) with the more universal shariah provisions. The differences of opinion bewteen classical scholars and contemporary Muslim thinkers have influenced the instrumentation of Islamic family law in Indonesia.Islamic laws expressed by Muslim scholars especially al-Shafi'i became the basis of family law in the Compilation of Islamic Law (KHI) as stated in Presidential Instruction No. 1 of 1991.According to the rules of KHI Article 171 (b, c), religious differences become a barrier to inheritance rights.While the opinion of contemporary Islamic thinkers consider the issue of different religions is not the cause of the inheritance-inheritance process. | 3,069,703 | 55842779 | 0 | 16 |
Effects of water levels on species diversity of silica-scaled chrysophytes in large tributaries of Lake Baikal Large tributaries of Lake Baikal considered as a “hotspot” for silica-scaled chrysophytes diversity. Here we presented the updated species composition of silica-scaled chrysophytes and ecological parameters of their habitat in the Barguzin and Selenga River tributaries and delta in a high water level period. The number of registered taxa was significantly lower compared to the low water conditions (23 versus 66 species) and included the following genera with a given number of species: Chrysosphaerella – 1; Paraphysomonas – 2; Clathromonas – 1; Spiniferomonas – 3; Mallomonas – 9; Synura – 7. Mallomonas guttata and Synura borealis were identified in Russian waters for the first time. Thus, the corrected total list of silica-scaled chrysophytes in the Baikal Region includes 79 taxa. Though, the high water level reduced the total number of silica-scaled chrysophyte taxa, it made the water ecosystem more dynamic by enriching it with the entirely new species for this region. Introduction Lake Baikal is the most ancient and deepest (1637 m) lake in the world (Baikal. Atlas 1993). It has a tectonic origin and lies in a deep depression surrounded by mountain abundance and species diversity were recorded during the flood in July 2013 (Sorokovikova et al. 2017). The aim of our study was to investigate the impact of the high water level in July 2018 on the flora of silica-scaled chrysophytes in the Selenga and Barguzin River with their tributaries. We also compared the species composition in the | 3,069,704 | 218809373 | 0 | 16 |
tributaries under high and low water conditions. Description of the area The studied waterbodies are located in Russia, specifically in the south of East Siberia in the Republic of Buryatia (50°70'-53°82' N and 106°25'-109°90' E). They include the Selenga River, the mouth of the Kharauz Creek of the Selenga Delta, Lake Zavernyaikha, the Dzhida, the Temnik and Chikoy tributaries of the Selenga River, and the upstream portion of the Barguzin River (upward of the Ulyun River) with its mouth (Fig. 1). The catchment area of the Selenga River (the main tributary of Lake Baikal) is primarily located in Mongolia, but its run-off is mainly formed in Russia. It increases three times in size from the Russian-Mongolian border to the mouth. The high altitude of the watershed and its significant slope formed the mountainous character of studied tributaries. The most full-flowing river is the Chikoy River (the right tributary of the Selenga River); its annual average run-off is 267 m3/s. The Dzhida and Temnik Rivers fall from the left bank; their run-off is significantly lower -67.6 and 29.9 m3/s, respectively (Sinyukovich 2005). The Selenga River forms a large delta that includes many creeks, lakes, and former riverbeds (Baikal. Atlas 1993). The Kharauz Creek is one of the largest in the delta. Lake Zavernyaikha is located in the Selenga Delta and is cut off the Kharauz Creek by a sand bar. During the floods and high water conditions, the lake is connected to the creek and thus has a good turnover. In winter and low water conditions, the | 3,069,705 | 218809373 | 0 | 16 |
lake is isolated from the creek (Popovskaya et al. 2011). The Barguzin River is a tributary of Lake Baikal. Its run-off is the third largest by volume (after the Selenga and Upper Angara Rivers). At the upper reaches, this river is an impetuous mountain torrent that flows through a narrow gorge. When it enters into the Barguzin depression, the river flows on a broad valley and becomes a plain. Low parts of the flood plain have plenty of shallow eutrophic lakes and wetlands that are connected by a system of channels that provide the river with organic matter and other substances (Drucker et al. 1997). Specimens were deposited in the following collection: MCTP, Coleção de Aracnídeos, Porto Alegre (curator: Renato Augusto Teixeira) and the Smithsonian Museum of Natural History (SMNH), Arachnida and Myriapoda collection, Washington DC (curator: Hannah Wood). We attempted DNA extraction and amplification of DNA barcodes from legs of borrowed specimens preserved in ethanol, however this yielded no viable DNA. Material and methods We obtained the samples from the Selenga tributaries (Dzhida, Temnik and Chikoy), the Kharauz Creek and Lake Zavernyaikha in the Selenga delta and the Barguzin River (mouth and upward of the Ulyun River) in July 2018. Fourteen samples were used for analysis of the silica-scaled chrysophytes (Fig. 1). We used portable pH meter (IT-1101; Russia) to measure pH, water temperature, and dissolved oxygen concentrations at the sampling sites (Manual for chemical analysis of inland surface waters, 2009). We filtered the samples for chemical analyses through 0.45 μm membrane filters (Advantec, Japan) | 3,069,706 | 218809373 | 0 | 16 |
and measured the conductivity at 25°C with a conductometer DS-12 (Horiba, Japan). We also used colourimetric and dichromate oxidisability (COD -chemical oxygen demand) methods to determine the nutrient concentrations and total organic matter content, respectively (Manual for chemical analysis of inland surface waters 2009); Wetzel and Likens 2003). We took the algal samples from the surface layer of water (0 m) with a 1 L water sampler and fixed with Lugol's solution (1% f.c.). We also took 10-15 mL samples by means of Whatman membrane filters (pore size 1 μm, Whatman, USA). We identify the scaled chrysophytes using scanning and transmission electron microscopy. The samples for SEM analysis were filtered, dried at room temperature, coated with gold and examined using a Quanta 200 (FEI Company, USA) scanning electron microscope. The samples for TEM analysis were taken with water sampler, settled by the sedimentation method (Kuzmin 1975), centrifuged (MiniSpin, Eppendorf, Germany) and washed in deionized water. The washed samples were processed with 30% H2O2 at 75°С for 2 h, than the procedure was repeated and the samples were put on 3-mm-diameter formvar coated grids, dried at room temperature and analyzed by means of a LEO 906E transmission electron microscope (Carl Zeiss, Germany). The scales identified by means of electron microscopy classified the certain species according to their fine structure described and represented by microphotographs (Siver 1988;Hällfors and Hällfors 1988;Siver 1995;Němcová et al. 2012;Škaloud et al. 2012;Scoble and Cavalier-Smith 2014;Siver 2015). We also used the Freshwater Algal Database of Škaloud et al. (2013). Data on river flow rates | 3,069,707 | 218809373 | 0 | 16 |
were retrieved from Hydrometeorological Research Center of Russian Federation (Hydrometcenter of Russia). Physical and chemical characteristics of the studied waterbodies In July 2018, a continuous low water period in the catchment area of Lake Baikal ended when the Selenga and Barguzin Rivers rose and flooded their flood plains. Water discharge of the Barguzin and the Selenga Rivers during the sampling was up to 358 m 3 /s and 1700 m 3 /s, respectively (Table 1). It was 1.5-2 fold higher than in 2016 (Fig. 2). The water temperature in the Selenga and its tributaries was 18.5-20.9°С (Table 1); it varied from 10.2°С in the upper reaches of the Barguzin River (station 7) up to 20.9°С in its mouth (station 6). High pH values were recorded in Lake Zavernyaikha (station 4), the Kharauz Creek (station 5) and at the upper reaches of the Barguzin River (station 7); the lowest pH was recorded in the Chikoy River (station 2). The highest conductivity of 228 µS cm-1 was observed in the Dzhida River (station 1); the lowest conductivity of 58 µS cm-1 was registered in the Chikoy River (station 2; Table 1). The dissolved oxygen content varied over a wide range (Table 1). The water temperature in the Selenga and its tributaries was 18.5-20.9°С (Table 1); it varied from 10.2°С in the upper reaches of the Barguzin River (station 7) up to 20.9°С in its mouth (station 6). High pH values were recorded in Lake Zavernyaikha (station 4), the Kharauz Creek (station 5) and at the upper reaches of | 3,069,708 | 218809373 | 0 | 16 |
the Barguzin River (station 7); the lowest pH was recorded in the Chikoy River (station 2). The highest conductivity of 228 µS cm -1 was observed in the Dzhida River (station 1); the lowest conductivity of 58 µS cm -1 was registered in the Chikoy River (station 2; Table 1). The dissolved oxygen content varied over a wide range (Table 1). The crossflow of water from the Kharauz Creek (station 5) to Lake Zavernyaikha (station 4) caused by the high water level of the Selenga River levelled the oxygen content, water temperature, pH, and conductivity at these stations. The upper reaches of the Barguzin River (station 7) showed an increased oxygen concentration due to its better solubility in cold water and aeration due to the higher stream speed. The lowest oxygen concentration was registered in the river mouth (station 6). The concentration of silicon was high at all stations (Table 1). The inundation of the flood plains enriched the rivers with a large amount of organic matter from the catchment area and increased its water concentration. The highest concentrations were recorded in the mouth of the Barguzin River (station 6), Dzhida River (station 1), and Kharauz Creek (station 5). The total phosphorous values at all stations (except 3 and 7) were typical for polluted eutrophic waters. The crossflow of water from the Kharauz Creek (station 5) to Lake Zavernyaikha (station 4) caused by the high water level of the Selenga River levelled the oxygen content, water temperature, pH, and conductivity at these stations. The upper reaches | 3,069,709 | 218809373 | 0 | 16 |
of the Barguzin River (station 7) showed an increased oxygen concentration due to its better solubility in cold water and aeration due to the higher stream speed. The lowest oxygen concentration was registered in the river mouth (station 6). The concentration of silicon was high at all stations ( Table 1).The inundation of the flood plains enriched the rivers with a large amount of organic matter from the catchment area and increased its water concentration. The highest concentrations were recorded in the mouth of the Barguzin River (station 6), Dzhida River (station 1), and Kharauz Creek (station 5). The total phosphorous values at all stations (except 3 and 7) were typical for polluted eutrophic waters. Only seven species were identified in the Selenga tributaries Chikoy (station 1) and Dzhida (station 2), six species were recorded in the Temnik River (station 3). No chrysophytes were found in Lake Zavernyaikha (station 4) and the upper reaches of the Barguzin River upward of the Ulyun River (station 7; Table 2). During the high water period, the chrysophyte flora was mainly represented by widespread and cosmopolitan species typical in the temperate and subarctic regions of Eurasia and North America. A total of 16 species were identified: Spiniferomonas cornuta, S. serrata, S. trioralis, Mallomonas acaroides, M. akrokomos, M. alpina, M. crassisquama, M. heterospina, M. guttata, M. striata, M. tonsurata, Synura echinulata, S. glabra, S. petersenii, S. spinosa, and S. uvella. Although the species composition varied in all waterbodies, two widespread species M. tonsurata and S. petersenii occurred everywhere. A single scale | 3,069,710 | 218809373 | 0 | 16 |
attributed to the genus Clathromonas was found in the mouth of the Barguzin River (station 6), but it was impossible to identify it to a species (Fig. 4e) (mouth of the Barguzin River, station 6), f -Synura borealis (mouth of the Barguzin River, station 6), g -Synura heteropora (Chikoy River, station 2), h -Synura spinosa (Dzhida River, station 1), I -Synura echinulata (mouth of the Barguzin River, station 6), j -Synura uvella (mouth of the Barguzin River, station 6), k -Synura petersenii (Temnik River, station 3), l -Synura glabra (mouth of the Barguzin River, station 6). Micrographs were obtained with transmission electron microscopy; scale bars are 1 µm. Table 2. List of species and intraspecific taxa of the silica-scaled chrysophytes identified by electron microscopy in the Selenga tributaries and Barguzin River area in July 2018. See Fig. 1 for the location of the stations. A plus (+) indicates the presences of the species at the station. The asterisk (*) indicates the species observed in the Baikal Region for the first time. Discussion Biogeographical structure of silica-scaled chrysophytes of the studied area The flooding on the Selenga and Barguzin Rivers decreased the species diversity of silica-scaled chrysophytes (23 species) versus 66 species observed before this studyduring the low water period (Bessudova et al. 2018a). Although the chrysophyte species diversity decreased almost threefold, their biogeographical distribution changed proportionally (Table 3). Three species identified during this study rarely occur in Russian waters. One of them, C. baicalensis, is endemic to Lake Baikal, and two are arctoboreal species: Paraphysomonas acuminata | 3,069,711 | 218809373 | 0 | 16 |
acuminata and Paraphysomonas vulgaris. P. acuminata acuminata was found and described in one Austrian freshwater lake (Scoble and Cavalier-Smith 2014). We identified the species in the mouth of the guzin River, the creeks of the Selenga Delta (including Lake Zavernyaikha; Bessudova et al. 2018a), Lake Baikal (Bessudova et al. 2017) and in the mouth of the Srednyaya Creek of the Angara-Kichera Delta (Bessudova et al. 2018b). This species occurs in spring, summer and autumn, while most numerous in May. During this study, the species was found in the Dzhida (station 1) and the Chikoy (station 2) Rivers. P. vulgaris was found and described in freshwaters of England (Scoble and Cavalier-Smith 2014). To date, this species has been identified in the mouth of the Barguzin River and the Selenga tributaries (Bessudova et al. 2018a). A spine morphologically similar to those of P. vulgaris was also found in Lake Toko, Japan (see Gusev et al. 2018, fig. 39). During this study, P. vulgaris was registered in the Temnik River (station 3). Two species, Mallomonas trummensis and S. heteropora, have a limited distribution in Russian waters. M. trummensis occurs in waters of the temperate and subarctic zones of Europe (Škaloud et al. 2013). In Russia, the species was identified only in the mouth of the Barguzin River and the creeks of the Selenga Delta (Bessudova et al. 2018a). Here, we found it only at one station, namely the mouth of the Barguzin River (Station 6). S. heteropora occurs in waters of Europe (Škaloud et al. 2014), but in Russia | 3,069,712 | 218809373 | 0 | 16 |
it was previously observed only in Lake Baikal (Bessudova et al. 2017), the mouth of the Barguzin River and the creeks of the Selenga Delta (Bessudova et al. 2018a). During the present study, S. heteropora scales occurred only in the Chikoy River (station 2). Ecology of silica-scaled chrysophytes The studied area is interesting not only to study the effect of floods on species composition of silica-scaled chrysophytes, but also to evaluate the impact on species ecology. The water parameters in study area during the low and high water levels were in a sharp contrast with the optimum for high diversity of silica-scaled chrysophyte (Eloranta 1995;Siver 1995;Siver and Lott 2017). The highest silica-scaled chrysophyte diversity has been recorded in waters with pH low or close to neutral (below 7), low mineralization, conductivity close to or slightly less than 40 µs cm -1 , low nutrient content (oligotrophic to mesotrophic), and moderate quantity of dissolved humic compounds (Eloranta 1995;Němcova et al. 2003;Siver 1995;Siver and Lott 2017). A recent study in Newfoundland Island corroborated high silica-scaled chrysophyte diversity (47 species) in waters with pH 3.9-6.7, high content of humic substances and low nutrient concentration (Siver and Lott 2017). Numerous studies allowed Siver (2015) to divide the silica-scaled chrysophytes into four groups with congruent boundaries along a pH gradient: (1) species that inhabit waters with a low pH (below 6); (2) species that inhabit waters with modern levels of alkalinity (pH below 7 but above 5); (3) species that inhabit waters with a neutral pH (pH-indifferent species), (4) species that | 3,069,713 | 218809373 | 0 | 16 |
inhabit waters with a high pH (above 7). One species, S. echinulata despite their affiliation with the low pH group, was found in waters with high pH (7.87; station 6). S. echinulata was previously observed in the Selenga Delta and Lake Zavernyaikha at pH 8.01 and 8.03, respectively (Bessudova et al. 2018a). We also found two species characteristic to average waters: M. heterospina (station 2) observed at pH 7.39 and S. spinosa (stations 1 and 6) observed at pH 7.87 and 8.0, respectively. M. heterospina was previously identified in the Selenga River and in creeks of its delta at high pH values (7.85 and 7.98), but it was erroneously identified as Mallomonas pugio Bradley (Bessudova et al. 2018a). S. spinosa was also previously observed in the Selenga River and in delta creeks at pH 7.7 and 8.03 (Bessudova et al. 2018a). Only two species characteristic to alkaline waters, namely M. alpina and M. tonsurata, occurred in the study area at pH 7.39-8.15. Němcová et al. (2003) demonstrated that the high diversity of silica-scaled chrysophytes was typical in waters with conductivity close to or below 40 µs cm -1 , whereas conductivity above 200 µs cm -1 reduced the species diversity. The study area had high conductivity values during low and high water levels. The minimum conductivity values of 58 and 92 µs cm-1 were recorded only at two sites, stations 2 and 3, respectively (Table 1). However, this factor did not influence the diversity of silica-scaled chrysophytes; indeed, we observed the opposite situation where it were | 3,069,714 | 218809373 | 0 | 16 |
twice as many species at high conductivity value (174 µs cm -1 ; station 6). The highest silica-scaled chrysophyte diversity, (35 species and intraspecific taxa) was registered during the previous study in the mouth of the Barguzin River at a conductivity of 151 µs cm -1 (Bessudova et al. 2018a). Despite some evidences that silica-scaled chrysophytes prefer oligo-and mesotrophic conditions (Eloranta 1995;Němcová et al. 2003;Siver 1995;Siver and Lott, 2017), high species diversity and biomass were also recorded in eutrophic waters (Cronberg 1996;Kristiansen 1985Siver 2015;Siver and Wujek, 1993), including the mouth of the Barguzin River in the Selenga Delta (Bessudova et al. 2018a). Notably, none of the species that would prefer oligo-and mesotrophic conditions according to Siver (2015) was found in the area studied. At the same time, M. tonsurata and M. alpina, both considered to prefer eutrophic waters, were found during our study in high and low water levels. Furthermore, these species also occurred in the plankton of the oligotrophic Lake Baikal (Bessudova et al. 2017). Overall, the Baikal Region significantly expands the optimal conditions to develop and maintain high diversity of silica-scaled chrysophytes. In large rivers, the flow rates increase proportionally to the rise of water discharge (Fig. 5). Hence, the conditions for phytoplankton growth in 2018 were less favorable compared to 2016. The flood pulse concept elaborated by Junk et al. (1989) stated that a seasonal flood is useful for river ecosystems and could affect their biotic composition, nutrient transport, and sediment distribution. However, violent floods can be destructive for aquatic organisms (Talbot et | 3,069,715 | 218809373 | 0 | 16 |
al. 2018). The seasonal flood in the study area was one of the factors influenced the species composition of silica-scaled chrysophytes in the mouths of the Selenga and Barguzin Rivers. During the flood, we observed a significant depauperisation in the chrysophyte species composition compared to the previous data. Thus, recent studies demonstrated that the beginning of a flood was accompanied by species impoverishment in plankton communities, even if the silicon and nitrogen concentrations were sufficient for their development (Talbot et al. 2018). We suggested that the concentrations of chemical components, including oxygen, silicon, nitrogen, and phosphorous could not limit the development of silica-scaled chrysophytes in studied area. These concentrations were similar or sometimes higher than in low water period (Table 1). However, most species previously observed in the creeks of the Selenga Delta and the mouth of the Barguzin River were absent during the flood. Nevertheless, the number of registered species were proportionally distributed among the genera (Fig. 6). In previous studies, 15 and 20 species of silica-scaled chrysophytes were identified in Lake Zavernyaikha and in the mouth of the Barguzin River, respectively, under the low water level. However, chrysophytes were absent from the mouth of the Kharauz Creek in May, July andSeptember 2016 (Bessudova et al. 2018a). The increased water content in the main tributaries of Lake Baikal favoured significant changes in chrysophyte diversity in these areas. Thus, only 14 species were identified in the mouth of the Barguzin River (station 6) and five species in the Kharauz Creek (station 5), while they were absent | 3,069,716 | 218809373 | 0 | 16 |
in Lake Zavernyaikha (station 4). The high water level connected the waterbodies (stations 4 and 5), which levelled the difference in hydrochemical conditions (oxygen content, water temperature, pH, and conductivity). This phenomenon limited the development of some chrysophytes but did not favour similarity in their species composition. In previous studies, 15 and 20 species of silica-scaled chrysophytes were identified in Lake Zavernyaikha and in the mouth of the Barguzin River, respectively, under the low water level. However, chrysophytes were absent from the mouth of the Kharauz Creek in May, July and September 2016 (Bessudova et al. 2018a). The increased water content in the main tributaries of Lake Baikal favoured significant changes in chrysophyte diversity in these areas. Thus, only 14 species were identified in the mouth of the Barguzin River (station 6) and five species in the Kharauz Creek (station 5), while they were absent in Lake Zavernyaikha (station 4). The high water level connected the waterbodies (stations 4 and 5), which levelled the difference in hydrochemical conditions (oxygen content, water temperature, pH, and conductivity). This phenomenon limited the development of some chrysophytes but did not favour similarity in their species composition. The high diversity of silica-scaled chrysophytes in the mouths of the main tributaries of Lake Baikal, the Selenga, Upper Angara River and Barguzin Rivers in low water conditions could be caused by anterior floods. The inundation of the flood plains leads to integration of small creeks and lakes that enrich their flora due to dissemination of a broad spectrum of species (Fernandes et al. | 3,069,717 | 218809373 | 0 | 16 |
2014;Junk et al. 1989). Water retreats control the bloom of phytoplankton, inter alia chrysophytes, in the warm and shallow waterbodies with high level of biological production, which favours the diversity of silica-scaled chrysophytes in the Baikal Region. Additionally, the climatic changes recorded worldwide during the previous decades, including the Baikal Region, may also underscore the quantitative and qualitative development of silica-scaled chrysophytes (Shimaraev et al. 2002;Sinyukovich et al. 2010;Sorokovikova et al. 2015). Global climate change influences the hydrology of waterbodies (Mushet et al. 2017) that will significantly impact the development of aquatic organisms. The interchange of floods and low water levels created various environmental conditions (Table 1) and stimulated dynamics of the ecosystem allow the formation of a "hotspot" for silica-scaled chrysophytes diversity. | 3,069,718 | 218809373 | 0 | 16 |
Interacting Microbe and Litter Quality Controls on Litter Decomposition: A Modeling Analysis The decomposition of plant litter in soil is a dynamic process during which substrate chemistry and microbial controls interact. We more clearly quantify these controls with a revised version of the Guild-based Decomposition Model (GDM) in which we used a reverse Michaelis-Menten approach to simulate short-term (112 days) decomposition of roots from four genotypes of Zea mays that differed primarily in lignin chemistry. A co-metabolic relationship between the degradation of lignin and holocellulose (cellulose+hemicellulose) fractions of litter showed that the reduction in decay rate with increasing lignin concentration (LCI) was related to the level of arabinan substitutions in arabinoxylan chains (i.e., arabinan to xylan or A∶X ratio) and the extent to which hemicellulose chains are cross-linked with lignin in plant cell walls. This pattern was consistent between genotypes and during progressive decomposition within each genotype. Moreover, decay rates were controlled by these cross-linkages from the start of decomposition. We also discovered it necessary to divide the Van Soest soluble (labile) fraction of litter C into two pools: one that rapidly decomposed and a second that was more persistent. Simulated microbial production was consistent with recent studies suggesting that more rapidly decomposing materials can generate greater amounts of potentially recalcitrant microbial products despite the rapid loss of litter mass. Sensitivity analyses failed to identify any model parameter that consistently explained a large proportion of model variation, suggesting that feedback controls between litter quality and microbial activity in the reverse Michaelis-Menten approach resulted in stable model | 3,069,719 | 15625617 | 0 | 16 |
behavior. Model extrapolations to an independent set of data, derived from the decomposition of 12 different genotypes of maize roots, averaged within <3% of observed respiration rates and total CO2 efflux over 112 days. Introduction Recent studies are challenging the ways that we have traditionally perceived and modeled decomposition. Decomposers were once thought to rapidly degrade labile fractions of plant litter, such as carbohydrates and proteins, leaving more recalcitrant compounds, like lignin, to become the foundation of stabilized soil organic matter [1,2,3]. However, microbial products rather than plant lignin actually may represent the larger fraction of SOM [4,5], and lignin per se may not persist throughout decay [6]. Unfortunately, many mathematical models are based on the traditional view of decomposition. In part this is because the chemical composition of litter has often been evaluated by proximate carbon analysis, typically yielding three, qualitatively different pools of compounds: polar and nonpolar extractives, acid hydrolysable materials, and acid non-hydrolysable materials. These pools provide a convenient structural framework for modeling, but lack the resolution to address finer scale biochemical transformations revealed by more contemporary studies [4,5,6]. Mathematical models must change to reflect these observations. One of the changes needed in traditional decomposition models is to explicitly simulate the activities of microorganisms. If microbial contributions to SOM are more substantial than lignin, then the litter compounds fueling microbial activity may be more important to C stabilization than the plant lignin pool. This point was suggested by Smith et al. [7] who noted that the proportion of initial litter C remaining over | 3,069,720 | 15625617 | 0 | 16 |
time (125 days) was directly related to the initial rate of decomposition prior to lignin decay. In other words, the incorporation of C into microbial biomass increased its overall persistence despite an initially higher fractional loss through respiration [2,8]. A simple explanation is that the higher carbon use efficiency (CUE) of readily decayed materials, like sugars and proteins, transfers a larger fraction of substrate C to microbial biomass than lignin, and that these microbial products are more persistent [4,5]. These results suggest that the early stage of litter decay when microbial activities are greatest should be an integral component of mathematical models. A second change needed in decomposition models is how they respond to differences and interactions between the chemical constituents of decaying litter. Despite the questionable role of plant lignin in decomposition, both the initial lignin content and the lignocellulose index (LCI) of litter are often the best predictors of decay rate [1,3]. The lignocellulose index is normally calculated as the ratio of the acid non-hydrolysable/(non-hydrolysable+ hydrolysable) products of proximate C analysis (see above), and assumed to represent the lignin (non-hydrolysable) and holocellulose (hydrolysable) fractions of litter. Although this assumption may be nearly true for fresh litter, both microbial and lignin degradation products increasingly contribute to proximate C fractions during decay [5,6]. Thus LCI and proximate C fractions are ambiguous metrics of litter chemistry and models using them to regulate decomposition conflate litter quality controls with the decomposition process. In reality, plant cell walls are usually the largest component of plant litter and have a | 3,069,721 | 15625617 | 0 | 16 |
specific biophysical structure of interconnected saccharide and phenolic polymers [9]. This arrangement partly explains the relatively consistent relationship between LCI and decay, but raises questions about the precise relationships between lignin and other litter constituents [10,11,12] needed to more accurately model substrate dynamics during decomposition. Few empirical studies have examined litter chemistry during decay with sufficient resolution to improve models beyond the lignin or LCI controls normally calculated. However, Machinet et al. [11,13,14] examined the detailed litter chemistry of maize (Zea mays L.) roots for naturally occurring genotypes decomposing in laboratory incubations. These genotypes varied primarily in their lignin content. Long-term C losses (.200 days) were most often correlated with compounds associated with lignin and crosslinked between lignin and polysaccharides, whereas short-term (, 200 days) losses were more closely related to soluble compounds and cell wall sugars [11]. Different loss rates of different sugars (e.g., glucans, xylans and arabinans) suggested that the relationship between lignin and decomposition was partly mediated by the specific composition of the polysaccharide fraction of the litter, probably because arabinan substitution in xylan chains interferes with the degradation of hemicellulose [9]. In other words, the sugar composition of cell walls appears to define the transition from short-to long-term patterns of decay, providing a mechanistic explanations for patterns in microbial activity, as well as the empirical relationships between LCI and decay most often used in decomposition models [15]. The overall goal of this study was to simulate the relationships between litter decay, microbial production and litter chemistry at the early stage of decomposition, | 3,069,722 | 15625617 | 0 | 16 |
using data collected by Machinet et al. [11,13] to test and refine the Guild Decomposition Model (GDM) developed by Moorhead and Sinsabaugh [16]. Our two specific objectives were to simulate patterns of microbial productivity during decomposition as litter chemistry changed, and to simulate the relationship between measured changes in cell wall chemistry (sugars and lignin) and decay rate. We chose GDM because it calculates the decay of specific substrate pools as a combined function of both microbial activity and substrate characteristics, using the Michaelis-Menten equation of substrate saturation. The maximum velocity of decomposition (V max ) is thus a function of microbial biomass [16], so that estimated decay rates reflect both the changing chemical composition of the decomposing litter and microbial productivity. Methods Our general approach was to revise GDM to use the empirical data collected by Machinet et al. [11,13,14] to derive parameters needed to simulate decomposition. Then, we analyzed the data from a 112-day laboratory study of four maize genotypes that differed from one another in initial litter chemistry [13,14], to evaluate interactions between litter qualities likely controlling the decomposition process, and to obtain estimates for parameter values used to describe these processes. Next, we selected two key parameters for GDM, which preliminary analyses indicated varied with initial litter chemistry and had large effects on model behavior (below). We then optimized these parameters to produce the best possible fit between simulated and observed patterns of decomposition, with respect to CO 2 efflux and chemical transformations in decaying litter. Best-fit parameter values were in turn | 3,069,723 | 15625617 | 0 | 16 |
compared to initial litter chemical characteristics to determine possible relationships. These relationships were then used to derive parameter values needed to simulate CO 2 efflux during decomposition of maize roots for 12 additional genotypes [11]. Machinet et al. [13] estimated mass loss of litter based on cumulative CO 2 carbon efflux. Concentrations of litter chemical fractions at each date were multiplied by the estimated mass of remaining litter to estimate pool sizes during decomposition. Decay rate coefficients (k i ) were calculated for chemical constituents C 1 , C 2 and C 3 for all 4 litter types over all 4 periods of observation (days 0-14, 14-36, 36-57 and 57-112), as the difference in the natural log of pool size between observation dates, divided by the time period. Model revisions This model was programmed in MATLAB (The MathWorks Inc., Natick, USA). We revised GDM to use Reverse Michaelis-Menten (RMM) functions to calculate decay rates as functions of microbial activity [18]: dC i /dt = k i ?C i ?B j /(K Bji +B j ), where C i is the amount of substrate i, B j is the amount of microbial biomass in guild j, and K Bji is the half-saturation coefficient of guild j for substrate i. Note that we altered the more common RMM approach by using biomass (B) in place of an enzyme (E) pool, which assumes a constant ratio of E:B resulting from constitutive enzyme production [16,18]. We limited access to substrate pools by guilds: guild 1 was assumed to access only | 3,069,724 | 15625617 | 0 | 16 |
soluble resources (C 1 ); guild 2 was assumed to access both C 1 and C 2 ; guild 3 could access C 1 , C 2 and C 3 ( Figure 1). A final revision to GDM was necessary to capture the dynamics of the soluble pool (C 1 ), a fraction of which persisted throughout the study by Machinet et al. [13]. This persistent fraction (C 1T ) varied between genotypes and was considered to be nondecomposable during the time frame of the study (112 days). Parameters and state variables All parameters and state variables are reported in Table 1. Values of C 1T , the persistent fraction of the initial C 1 pool, were estimated from the average values of C 1 over time for each genotype ( Table 1). We selected a decay rate coefficient (k 1 ) of 0.1 for the decomposable fraction of the C 1 pool (C 1D ), due to its rapid loss. The RMM approach made it possible to simplify the balance of C 2 and C 3 decay as functions of lignocellulose index (LCI) according to Moorhead et al. [19] (Text S1). In brief, the decay rate coefficients (k 2 and k 3 ) were described as linear functions of LCI: k i = m i ?LCI+k imax , given empirically observed maximum values (k imax ) and slopes (m i ). This approach assumes that k 3 = 0 at a threshold level of LCI = LCI T , wherein LCI T = 0.4 [20], | 3,069,725 | 15625617 | 0 | 16 |
which defines the point at which LCI changes from being solely determined by C 2 decay (LCI#LCI T ) to also being determined by C 3 decay (LCI.LCI T ). The value of k 2max (maximum coefficient of C 2 decay) was estimated as the intercept of the linear regression of the observed decay rate coefficients for C 2 (k 2 ) against litter LCI for the four maize genotypes (Figure 2a; Table 1) over days 14-112, excluding values estimated over days 0-14 when we assumed that microorganisms had not begun to fully utilize pool C 2 . Values of k 2 used during simulations were then estimated according to LCI of remaining litter. This revision to GDM provided a closer fit to observed patterns of C 2 decay (Figure 2a; N = 12, R 2 = 0.97, P,0.01) than Moorhead and Sinsabaugh [16]. The value of k 3max was set to 0.001 because there was little evidence of C 3 decomposition for any litter type during incubations [13]. We selected identical values of K B11 , K B12 and K B13 , the halfsaturation coefficients for the utilization of pool C 1 by all three guilds, assuming all microorganisms have similar affinities for soluble substrates (Table 1). We initially set K B22 and K B32 at the same values, again assuming that organisms capable of using the resource would have similar half-saturation coefficients, but higher than those for C 1 , assuming that C 2 was generally less decomposable than C 1 . We then | 3,069,726 | 15625617 | 0 | 16 |
selected an even higher value for K B33 assuming that C 3 was even less decomposable than C 2 . We are unaware of any published values for these parameters, and so followed the rationale of Moorhead and Sinsabaugh [16] in choosing values reflecting relative access to substrates of different qualities. We estimated microbial production as the difference between the quantity of carbon released from decaying substrates and the amount mineralized through both growth-and maintenanceassociated respiration. We assumed that this difference was immobilized in microbial biomass. GDM also calculates microbial turnover necessary to keep total microbial C less than 5% of the total system's organic C (microorganisms+substrates; Table 1). We estimated carbon use efficiency (CUE) as the difference between the amounts of C released from decaying substrates and mineralized through respiration, divided by the amount of C released by decomposition [23]. Optimizing parameter values The values of two key model parameters that showed correlations with initial chemistry, LCI T , and K B22 , were optimized for the four maize genotypes [13] using the fmincon function of the MATLAB Optimization Toolbox (The Math-Works, Inc., Natick, USA), which applies a sequential quadratic programming algorithm. The objective function was the sum of the root mean square errors calculated between the experimental and simulated CO 2 efflux rates, cumulative amount of carbon mineralized over 112-days and chemical evolution in the C 1 and C 2 pools during decomposition, normalized by the means of the observations. We determined the best-fit estimates of parameters K B22 and LCI T that most | 3,069,727 | 15625617 | 0 | 16 |
closely matched simulations to observed patterns of CO 2 efflux and both C 1 and C 2 pools over time (Table 1). We selected these parameters because the dynamics of pool C 2 were most closely related to cumulative CO 2 efflux and thus, estimated mass loss. Sensitivity analysis To evaluate the sensitivity of the model to parameter estimates (Table 1), we randomly varied model parameters e 1 , e 2 , k 1max , BC max and K B11 within 610% of their initial values (Table 1), 100 times for each litter type. We then calculated the differences between simulated and observed values of C 1 , C 2 , cumulative CO 2 , and respiration rates on each date of observation [13] as a relative value = (observation-simulation)/observation. We summed these relative differences for each type of observation over all days of observation (i.e., through day 112), which produced a composite measure of the relative differences between observations and model output for each maize genotype. ANCOVA evaluated the contributions of variations in parameter values to variations in the relative differences between model output and observations by litter type. The type II sums of squares from ANCOVA were interpreted to represent the relative contribution of each parameter to model behavior. Model extrapolation The last set of simulations estimated decomposition for the 12 additional maize genotypes examined by Machinet et al. [13]. Relationships between values of C 1T and best-fit values of LCI T and K B22 (Table 1) and the initial characteristics of litter chemistry | 3,069,728 | 15625617 | 0 | 16 |
were estimated for the four litter types described by Machinet et al. These relationships were then used to estimate parameter values of C 1T , LCI T and K B22 for simulations with each of the 12 additional genotypes reported by Machinet et al. [11], based on their initial chemistry. Principal components analyses evaluated relationships between the relative differences in observed and simulated values of respiration rates and cumulative CO 2 efflux, and initial litter chemistry characteristics for all 12 litters to determine if more detailed litter chemical characteristics than currently used in the model could provide additional insights to litter quality controls on decomposition. Decomposition dynamics We discovered a close correspondence between the arabinan:xylan ratio (A:X) and lignocellulose index (LCI) of the residue (Figure 2b) in the empirical data [11]. Between days 14 and 112, a linear regression of AX over LCI yielded an R 2 = 0.91 (N = 12, P# 0.01); we omitted days 0-14 because we expected decomposition to be limited by microbial activity rather than substrate [16,18]. Perhaps this relationship also explains why both AX and KL were related to best-fit model parameters LCI T and K B22 . Rates of CO 2 efflux for all four litter types rapidly increased to peak values within 14-21 days followed by gradual declines (Figure 3); cumulative CO 2 efflux rose rapidly during the first 36 days and then more slowly until day 112. For 3 of the 4 genotypes, the C 1 pool declined rapidly from day 1 to day 14, and then | 3,069,729 | 15625617 | 0 | 16 |
remained relatively constant during the rest of the incubation (Figure 4). For genotype F292bm3, the pool of C 1 remained essentially unchanged throughout the study. The C 2 pools of all four litters declined throughout the incubations, accounting for most of the litter mass loss over time (Figure 4). The C 3 fraction of the remaining litter showed a slight increase over the first 14 days of incubation (ca. 5-10%) for both F2 genotypes (not shown), but remained roughly constant for the two F292 genotypes [13]. Microbial production Microbial biomass rapidly increased to peak values within 20-40 days, varying among genotypes (Figure 5a), declining most rapidly for those genotypes supporting the most rapid initial growth with the highest decay rates (F292 and F292bm3). Microbial turnover rates (Figure 5b), which should correlate with the generation of microbial products such as cell walls, were similar in shape but had lower peak values and lagged behind the patterns of rising and falling respiration rates (Figure 3). Genotypes with higher turnover rates also showed the greatest declines in biomass by day 112. Litter LCI increased over time in all litter types, coincident with declining biomass and turnover rates (Figure 5c), increasing most rapidly for litters decaying most rapidly (Figure 3). CUE declined over time for all litter types (Figure 5d), coincident with increasing LCI (Figure 5c), but started at higher values in litter types with larger pools of labile C 1 (i.e., C 1D ), which had a higher C-assimilation efficiency (e i ) than C 2 ( Table 1). | 3,069,730 | 15625617 | 0 | 16 |
The total microbial production (C-immobilized into biomass including microbial turnover) by day 112 varied between litter types: 248, 195, 295 and 295 mgC for genotypes F2, F2bm1, F292 and F292bm3, respectively. These values were negatively related to initial litter KL and KL/AX contents, as well as final biomass and best-fit values of K B22 . They were positively related to total cumulative CO 2 efflux by day 112 and best-fit values of LCI T (all N = 4, P#0.05). Model test Our initial model parameter set, including best-fit estimates of K B22 and LCI T , simulated rates of CO 2 efflux closely matching observations (all N = 15; omitting day 0), with R 2 values ranging from 0.87 for genotype F2bm3 to 0.98 for genotype F2bm1 (Fig. 3). Simulated rates peaked before observations for all litter types and at slightly lower values. However, simulated peak rates were within 10% of observed peaks for both F2 genotypes and 20% for both F292 genotypes. Simulated values of cumulative CO 2 efflux (to day 112) were even closer to observations, with R 2 values in excess of 0.99 for all genotypes (Figure 3). All simulations were within 5% of observed cumulative CO 2 efflux values. Simulated values of C 2 seldom differed by more than 5% of the observed values, with all R 2 $0.99 (Figure 3). Finally, our model did not estimate any loss in pool C 3 during simulations and has no mechanism to generate an increase in this pool's size (Figure 1). Sensitivity analysis The | 3,069,731 | 15625617 | 0 | 16 |
relative contributions of individual parameters to explaining variations in model behaviors differed by litter type and observation (Table 2). For example, the efficiency of C assimilation from substrate C 1 (e 1 ) rarely explained more than 1% of the variation in model behavior for any litter type. Moreover, no single parameter made a substantial contribution ($10%) to explaining the variation in any model behavior when all four genotypes were pooled (not shown). When litter types were examined separately, variations in k 1max and K B11 often made their largest contributions to the same output variables for the same litter types. For example both parameters made substantial contributions ($10%) to variations in C 1 for litter types F2, F2bm1 and F292bm3; cumulative CO 2 efflux for litter types F2, F2bm1 and F292; respiration rates for litter types F2bm1 and F292; and overall model fit for litter type F2bm1. Neither parameter made a substantial contribution to variation in C 2 for any litter type. In contrast, parameters e 2 and BC max often made their largest contributions to model behaviors for litter types when k 1max and K B11 did not. For example, e 2 and BC max made substantial contributions to variations in C 2 for litter types F2, F2bm1 and F292bm3; cumulative CO 2 efflux for litters F292 and F292bm1; respiration rates for F2, F292 and F292bm3; and overall model fit for F2. Model extrapolation When we simulated decomposition of the 12 different maize genotypes [11], we found that overall simulated rates of respiration were | 3,069,732 | 15625617 | 0 | 16 |
strongly related to observations (N = 180, R 2 = 0.69, P# 0.01), and that simulated rates of respiration averaged 1617% lower than observed rates (data not shown). However, the relative differences ([observations-simulations]/observations) varied over time (Text S2). The first two axes of the principal components analysis explained 54% of the variation in the relative differences between observed and simulated respiration and litter chemical characteristics (Figure 6a). Differences in rates between days 10-21 were more strongly related to axis 2, along with litter LCI, KL/ AX and gSug. In contrast, differences in respiration for most days $50 were more strongly associated with axis 1, along with several chemical characteristics, the strongest being arabinans, AX, galactose, NDF and pCA (Figure 6a). In contrast to the daily respiration rates, there was no significant relationship between simulated and observed peak respiration rates (means = 17.561.8 and 21.469.9 mgC?kg soil 21 ?d 21 , respectively; not shown). The PCA showed that relative differences between peak rates (PR) were more closely related to axis 2 and opposite those of rates between days 10-21 (Figure 6a), with significant positive correlations with KL and KL/AX. Simulated values of cumulative CO 2 efflux were also strongly related to observations (N = 180, R 2 = 0.923, P#0.01), and averaged only 3618% greater than observations. The first two axes of the PCA explained 69% of the variation in the relative differences between observed and simulated cumulative CO 2 efflux and initial litter chemistry characteristics (Figure 6b). All values for CO 2 efflux were tightly clustered | 3,069,733 | 15625617 | 0 | 16 |
and closely associated with the first axis. Litter chemical characteristics, NDF, arabinans, galactose and AX, were also closely related to the first axis. The relative differences between observations and simulations at day 36 (d36) showed a significant, positive correlation with the initial soluble content of litter (C 1 = C-SOL) and negative correlation with galactan (GAL) content. Differences by day 59 were also related to C 1 and galactan. Although there were no significant differences between simulations and observations at day 112, these variations were significantly related to several aspects of initial litter chemistry, including C 1 . Estimates of microbial production by day 112 for these 12 genotypes ranged 203-303 mgC (mean = 270625 mgC), and were negatively correlated with KL/AX and final microbial biomass, as well as initial LCI. Production was positively related to total cumulative CO 2 efflux by day 112 and best-fit values of LCI T , as well as initial litter concentrations of Glu, gSug, and C 2 (all N = 12, P#0.05). A stepwise regression explained nearly all variation in microbial production as a function of total CO 2 efflux and initial concentrations of xylans (Xyl) and guaiacyl (G) in litter (N = 12, R 2 = 0.998, P#0.01). Lignin-cellulose interaction Simulations provided a close match to observed patterns of holocellulose (C 2 ) decay ( Figure 4). This pool represented the largest fraction of litter (Table 1), also explaining why simulations closely fit patterns of CO 2 efflux (Figure 3). Long-term decomposition (months to years) has often been negatively | 3,069,734 | 15625617 | 0 | 16 |
correlated to initial lignin content of litter [3], partly because lignin decays slowly and partly because some products of decomposition increase the size of the non-hydrolysable pool often interpreted as lignin [6]. However, Machinet et al. [11] also demonstrated the effects of biochemical connections between cell wall polysaccharides and lignin on the pattern of cumulative CO 2 efflux from decomposing maize roots. The fit between k 2 and LCI (Figure 2a) reveals an interaction between C 2 and C 3 in early stages of litter decay [19], well before C 3 begins to decline [13,20]. In other words, there is unlikely to be qualitatively separate pools of ligninshielded and unshielded holocellulose, as is sometimes implied [1,3]. Instead, biochemical linkages between cellulose, hemicellu-lose and lignin components of cell walls influence patterns of decomposition throughout the process. A brief explanation of these relationships is that the structural composition of hemicellulose is a primary chain consisting mainly of xylans with branching arabinan side chains that interact with other cell wall polymers. The level of arabinoxylan substitution (represented by A:X) increases with the progressive enzymatic degradation of plant material, during both digestion in the rumen [9,24] and decomposition in soils [12,13], as the more exposed elements of xylan chains are more readily hydrolyzed than those near arabinan branches. Gunnarsson et al. [25] found that initial xylan and arabinan concentrations in litter were as important as the total amount of hemicellulose in describing C mineralization over the first 9 days of laboratory incubations, and that arabinan was the single most important | 3,069,735 | 15625617 | 0 | 16 |
factor. These results were consistent with those of Machinet et al. [11], who identified initial arabinan content as an early predictor of C mineralization (days 3-7), followed by AX (days [7][8][9][10][11][12][13][14]. In addition, hydroxycinnamic acids (ferulic and p-coumaric acids) play a key role in cross-linking arabinoxylans with lignin and this cohesive network also hampers decomposition [10,11,12]. These cross-linkages explain the Table 2. Results of sensitivity analysis quantifying the relative contributions (%) of random variations in model parameters (column headings) to resulting variations in model behaviors (row labels), based on sums of squares from ANOVA relating model output to parameters. negative effect of lignin on holocellulose decay long reported in the literature [1,3] and also why k 2 decreased as AX increased early in decomposition ( Figure 2). Our model revision approximated this complex control with a straightforward relationship between LCI and k 2 (Figure 2) that included interactions between C 2 and C 3 (based on empirical sugar and lignin determinations). Moreover, key model parameters LCI T and K B22 were most closely related to initial KL/AX ratios of litter, emphasizing the interaction between holocellulose and lignin throughout the decomposition process. Soluble dynamics Data from Machinet et al. [13] also revealed that the soluble C 1 pool included a fraction (C 1T ) that persisted throughout the study (Figure 4). Although the soluble fraction of litter is usually considered to be highly decomposable, the persistence of a relatively large soluble pool is common in both soils and decaying litter [26,27]. The soluble pool usually contains a | 3,069,736 | 15625617 | 0 | 16 |
hydrophilic fraction including sugars and amino acids that are readily used by microorganisms [28,29], and a more hydrophobic portion, including soluble polyphenols (e.g., tannins) that are less rapidly utilized [26,30]. This differential use of compounds in the C 1 pool by decomposer microorganisms changes the pool's overall quality and decomposability with time [1]. Although the soluble pool is replenished with degradation products from non-soluble substrates [1,31], products of C 2 hydrolysis would enter the more labile fraction of the soluble pool, which cycles much more rapidly than the more persistent fraction [28,29]. In addition, Machinet et al. [13] found no decrease in the C 3 pool over time (not shown), so that it's degradation products could not have increased the more persistent fraction of the C 1 pool (C 1T ). Simulating the persistence of a sizeable pool of C 1 with GDM required dividing the pool into labile and persistent fractions (Figure 1), with the labile fraction (C 1D ) rapidly declining during decay and the persistent C 1T pool remaining intact (Figure 4). Moorhead and Sinsabaugh [16] found that simply routing the products of C 2 degradation through the whole C 1 pool GDM could not explain the size of this composite pool. Respiration patterns Simulations tended to overestimate rates of CO 2 efflux between days 0-14, and underestimate rates between days 14-36 ( Figure 3). A positive relationship between the most labile components of litter and early respiration (hours to weeks) or decay rate is commonly reported [32,33], and most models, including GDM, | 3,069,737 | 15625617 | 0 | 16 |
assume a greater decay rate for soluble substrates [15,16]. In terms of substrate dynamics, GDM tended to underestimate early losses (day 14) of C 1 and overestimate C 2 losses (Figure 4), suggesting that values of k 1max should be slightly increased and K B1i (i = 1,2,3) reduced (Table 1), to stimulate decay of C 1D and possibly initial CO 2 efflux. However, GDM underestimated respiration to a greater extent for litter types with lower amounts of C 1D , i.e., with higher C 1T (F292 and F292bm3). We found that C 1T was most closely related to the total sugar content of cell walls (gSug; N = 4; R 2 = 0.998; P#0.05) and that xylan, arabinan and glucan concentrations were highly correlated with each other (not shown). Further resolution of the chemical composition of the soluble fraction (C 1D and C 1T ) and its dynamics during decomposition are needed to improve the mathematical descriptions of these relationships. Microbial production Moorhead and Sinsabaugh [16] argued that litter decay is initially limited by microbial action, because there is a time lag in the colonization of fresh litter by decomposer microorganisms (Figure 2). Whether this lag was a numerical (biomass) or functional (physiological) response in the study by Machinet et al. [13] is unknown because they did not monitor microbial biomass. Therefore, the most speculative part of this study was our simulation of microbial dynamics. Nonetheless, the patterns and magnitudes of simulated microbial biomass and turnover rates were consistent with observed patterns of respiration (Figures | 3,069,738 | 15625617 | 0 | 16 |
3, Figure 6. Principal components analysis of variations in initial litter chemistry characteristics and relative differences between observed and simulated rates of microbial respiration and cumulative CO 2 efflux during decomposition of Zea mays roots. Results of principal components analysis of variations in initial litter chemistry characteristics and relative differences = (observationssimulations)/observations from simulated decomposition of 12 novel maize mutants [11] for: a. rates of respiration on days 3-112 (e.g., d3 = rate on day 3) and peak respiration (PR), and b. cumulative CO 2 efflux by days 3-112 (parameter definitions given in text). doi:10.1371/journal.pone.0108769.g006 5b), reported limits to microbial biomass concentration in soils [21,22], and changes in litter chemistry likely controlling microbial activities (Figures 2, 4 and 5c; [1]) as well as CUE (Figure 5d) [23]. Parameter estimates for C-assimilation efficiencies (e i ) for various substrate pools, coefficient of microbial basal respiration rate (g), and maximum biomass:total system C ratio (BC max ), most directly affected the relationships between substrate decomposition and biomass dynamics (Table 1). However, variations in any or all of these parameter values generate similar patterns with respect to different litter chemistry, although amplitudes and temporal regimes vary [16]. Our primary reason for simulating biomass dynamics was to determine if litter quality affected microbial production consistent with the idea that more decomposable chemical fractions of litter not only decay more rapidly but also generate more microbial products likely to enter stable soil organic matter pools [4,5]. In fact, simulated microbial production was significantly and negatively correlated to KL and KL/AX, which | 3,069,739 | 15625617 | 0 | 16 |
are inversely related to decomposability, but production showed no significant positive relationship to any simple measure of litter quality. However, it was positively related to the sum of the two most decomposable carbon pools, i.e., C 2 +C 1D , (N = 4, rho = 0.982, P# 0.05). The contributions of the relatively larger pools of C 1D in litter types F2 and F2bm1 to simulated production were small compared to the larger pools of C 2 in the other litters, despite the lower C assimilation efficiency of C 2 versus C 1 (Table 1). Thus our results were consistent with the observations of Smith et al. [7] and others who found that higher initial rates of C incorporation into biomass coincided with higher losses of litter through respiration [2,8]. Sensitivity analysis The most interesting result of our sensitivity analysis was the lack of any simple, overall interpretation. No single parameter consistently explained.10% of the observed variability in any model behavior (Table 2). Usually a parameter important to explaining one model behavior, such as the contributions of k 1max to simulated C 1 dynamics or cumulative CO 2 efflux (Table 2), made little contribution to other model outputs. The largest discrepancies between simulations and observations were in early respiration and C 1 loss (discussed above). Of the tested parameters, e 2 , the C assimilation efficiency for C 2 , appeared to be most important to respiration rates, which seems reasonable because C 2 was the largest substrate pool and provided most of the C respired | 3,069,740 | 15625617 | 0 | 16 |
(Table 1, Figure 4). As for the dynamics of the C 1 pool, parameter k 1max appeared to be most important, followed by K B11 ; thus parameters controlling the degradation rate of this pool explained differences between simulations and observations ( Figure 4). These results are consistent with our earlier conclusion that greater resolution of the C 1 pool composition and dynamics might provide a better understanding of decomposition. Extrapolations We simulated the decomposition of 12 additional maize genotypes in the second set of incubations conducted by Machinet et al. [11] based on the assumption that the relationships between key parameters, LCI T , K B22 and C 1T , and litter chemical characteristics were consistent with those for the four genotypes examined by Machinet et al. [13]. The assumption seemed reasonable because all 16 genotypes were naturally occurring varieties of Zea mays, and likely to be more similar in chemistry and tissue architecture than unrelated species more commonly used in comparative decomposition experiments [6]. In general, simulations were within a few percent of observed values of respiration rates and cumulative CO 2 efflux over the entire period of incubation. Analyses of these differences between simulations and observations provided relatively little additional insight to patterns of cumulative CO 2 efflux (Figure 6b). The importance of the initial C 1 pool (as both C 1T and C 1D estimates) to cumulative CO 2 efflux through time again emphasized the importance of initial decay rate to longer-term patterns [7,8]. The relationships between CO 2 efflux on day | 3,069,741 | 15625617 | 0 | 16 |
112 and initial arabinan and p-coumaric acid concentrations and AX suggest that the cross-linkages among hemicellulose and lignin became increasingly important with progressive decay. The tight cluster of cumulative CO 2 efflux along the first axis of our PCA also underscored the importance of cross-linkages among cell wall constituents (e.g., arabinan, AX, NDF and p-coumaric acid) to litter decay (Figure 6b), consistent with Machinet et al. [11]. The differences between simulated and observed respiration rates were more variable, suggesting temporally shifting controls on decomposition. Peak rates and those on days 14-28 were related to initial KL and KL/AX, as well as glucan and S:G ( Figure 6a). In contrast, rates on days.42 were more closely associated with arabinan, AX, p-coumaric acid and ester-linked ferrulic acids, perhaps because polysaccharide-ester linked ferulic acids can form ether-links with lignin [34,35] and the syringyl units of lignin can be esterified by p-coumaric acids, which is typical of grass cell walls [11]. However, the biggest differences between simulated and observed rates were on days 36-42, which were most closely related to KL, ether-linked ferrulic acids and galactan (Figure 6b). The frequent importance of KL and AX (or their chemical constituents) to these patterns was surprising, because KL (in LCI) was used to estimate k 2 , and KL/AX to estimate LCI T and K B22 (previously discussed) used in simulations. Clearly, simple linear relationships were insufficient to capture the subtleties of these controls. The relationships between respiration and galactan and C:N ratio (Figure 6b) hint at a microbial control [14], in | 3,069,742 | 15625617 | 0 | 16 |
part because galactan is sometimes used as an index to microbial contributions [36], but it is also a hemicellulosic sugar, along with arabinan, rhamnan, and xylan [37]. Conclusions We found that the level of arabinan substitution in xylan chains (AX) was an important control in early stages of decomposition, and was also linearly related to LCI calculated on the basis of fine scale cell wall chemistry. These relationships provide a plausible, mechanistic explanation for earlier, empirical descriptions of LCI effects on decomposition as a result of biochemical cross-linkages between polysaccharides and lignin. Thus lignin and LCI serve as convenient, negative proxies for the decomposability of litter even at the start of decay. However, additional research is needed to determine the chemical composition and dynamics of the nonhydrolysable product of proximate C analysis that is typically termed ''lignin'' if we are to discover the mechanistic relationships between LCI and latter stages of decomposition. We also found that dividing the soluble pool of litter (C 1 ) into separate persistent (C 1T ) and labile (C 1D ) pools, was necessary to accurately simulate the dynamics of the composite soluble pool during early decomposition because not all soluble compounds are equally decomposable. The finer scale chemical composition and dynamics of the soluble component of litter is needed to determine the possible sources and fates of these compounds. These relationships between simulated patterns of litter decay, litter chemistry (LCI, AX, C 1T and C 1D ) and microbial productivity were consistent with the notion that the more rapid utilization | 3,069,743 | 15625617 | 0 | 16 |
of substrates with high carbon use efficiency generates greater amounts of microbial products that can contribute to stable soil organic matter pools than more persistent substrates with lower C-assimilation, like lignin. Our assumption that lignin decay provides little to no net C-acquisition by microorganisms is also consistent with recent observations that little lignin C enters stable SOC pools. Finally, our extrapolations with litter types that differed in initial chemistry demonstrated that relationships we found between key model parameters, LCI T , K B22 and C 1T , and decomposition were robust across a large range of maize litter types, highlighting the importance of access (K B22 ) to the largest, rapidly decaying pool of substrate (C 2 ) by microorganisms able to use this resource (G 2 ), as well as negative controls imposed by the less-accessible substrate pools (KL, C 1T ). Moreover, differences between simulations and observations indicated that temporal controls on decay rates shifted from relative substrate pool sizes (both accessible and persistent) dominating at the start to factors related to crosslinkages between structural polysaccharides and lignin with progressive decomposition. In closing, our results suggest that interactions between decomposer microorganisms and litter quality characteristics at the earlier stages of decomposition may provide more insights to soil organic C stabilization than later stages dominated by the more persistent chemical characteristics of the cell wall, if indeed microbial products comprise a large fraction of stable soil C pools. Supporting Information Data S1 Data used to test and refine model. Observed respiration rates and chemical composition | 3,069,744 | 15625617 | 0 | 16 |
of residues of maize roots during decomposition [13]. (DOCX) Figure S1 Differences over time between observed and simulated respiration rates and cumulative CO 2 efflux from the decomposition of Zea mays roots. Relative differences between observations and simulations of litter decomposition from 12 novel maize genotypes [11] for: a. respiration rates (mgC?kg soil 21 ?d 21 ) over time (means695% confidence intervals, all N = 12), b. cumulative CO 2 efflux (mgC?kg soil 21 ) over time (means695% confidence intervals, all N = 12). (TIFF) Text S1 Lignocellulose controls. Relationships between decay rate coefficients for polysaccharides (C 2 ) and polyphenolics (C 3 ) as functions of lignocellulose index (LCI = C 3 /[C 2 +C 3 ])). (DOCX) Text S2 Sensitivity analysis. Patterns of differences between observed and modeled patterns of respiration associated with decomposition of decaying maize roots. | 3,069,745 | 15625617 | 0 | 16 |
A comparison of frequentist and Bayesian approaches to the estimation of long-stay per-diems Introduction Within many diagnosis related group (DRG) systems, there is recognition that a single cost weight per DRG is not suitable, and that cost weights should take into account extremely lengthy hospital stays. Long lengths of stay are considered to be due to factors largely beyond the control of the hospital, and a single weight per DRG would potentially place hospitals under financial risk. Introduction Within many diagnosis related group (DRG) systems, there is recognition that a single cost weight per DRG is not suitable, and that cost weights should take into account extremely lengthy hospital stays. Long lengths of stay are considered to be due to factors largely beyond the control of the hospital, and a single weight per DRG would potentially place hospitals under financial risk. Within Canada's acute-care, inpatient grouping methodology -Case Mix Groups (CMG+) -long-stay episodes represent approximately 4.5% of all discharges. Within a CMG (analogous to DRG), the cost weight assigned to long-stay cases consists of the typical cost weight, plus a per diem for each day the case stays beyond the CMG mean. Within a CMG, the volume of long-stay records may be low, and the episode cost data highly variable. This results in per diem estimates of low precision. In this paper, we compare two methods for calculating long-stay per diems. We employ Bayesian methods for sparse data, and compare the results to those of the current frequentist approach. Methods CMG+ uses a two-step, likelihood-based approach | 3,069,746 | 431668 | 0 | 16 |
to estimate long-stay per diems. In the first step, per diems are estimated using a weighted, least-squares regression model fitted separately to each CMG. Only typical cases are used (i.e., deaths, signouts, transfers -and long-stay cases are excluded). The dependent variable in this regression is the cost of the case, while the independent variable is the length of stay. This model provides an estimate of the fixed cost as well as an estimate of the per diem for typical cases. In the second step, a weighted, least-squares regression model is fitted to the long-stay cases. The dependent variable in this regression is the ratio of the actual cost of the case to the predicted cost, where the predicted cost incorporates the typical per diems from the first step. The independent variables are case mix effects. This model provides adjustments to the typical per diems, resulting in per diems for long-stay cases. There is a strong motivation for proposing a Bayesian alternative. First, the current long-stay per diem estimates are susceptible to cost outliers. Second, we have very good information to inform prior distributions based on aggregating information for long-stay episodes across CMG. In our Bayesian alternative, a weighted, least-squares regression model will first estimate the fixed and per diem values across all CMG. In the second step, these estimates will act as prior distributions for the weighted, leastsquares regression models that estimate the typical per diem for each CMG. We will evaluate whether the Bayesian models are sensitive to the values of the prior probability distributions. | 3,069,747 | 431668 | 0 | 16 |
http://www.biomedcentral.com/1472-6963/9/S1/A1 Results We will compare the long-stay per diems, calculated using the current frequentist approach, with those calculated using our Bayesian approach. We will evaluate the magnitude and direction of changes in the per diems, changes in explanatory power of the resulting cost weights, and changes in weighted cases by hospital and stratum of hospitals. Conclusion Hospitals with a disproportionate share of long-stay cases have the most at stake when per diem values are inaccurate. For CMG with large differences, underlying causes will be pursued. We will discuss whether the computing effort associated with implementing Bayesian methods is worthwhile in terms of improvement in the accuracy and precision of per diem estimates. | 3,069,748 | 431668 | 0 | 16 |
BSMBench: a flexible and scalable supercomputer benchmark from computational particle physics Lattice Quantum ChromoDynamics (QCD), and by extension its parent field, Lattice Gauge Theory (LGT), make up a significant fraction of supercomputing cycles worldwide. As such, it would be irresponsible not to evaluate machines' suitability for such applications. To this end, a benchmark has been developed to assess the performance of LGT applications on modern HPC platforms. Distinct from previous QCD-based benchmarks, this allows probing the behaviour of a variety of theories, which allows varying the ratio of demands between on-node computations and inter-node communications. The results of testing this benchmark on various recent HPC platforms are presented, and directions for future development are discussed. I. Introduction Quantum ChromoDynamics (QCD), the theory of the strong interaction of quarks and gluons, is a highly successful theory with high-precision predictive power. However, calculations of physical interest are rarely analytically tractable, instead requiring Monte Carlo simulation of a discretised treatment referred to as Lattice QCD (LQCD). Lattice QCD codes are developed by a number of theoretical particle physics research groups internationally, and these codes use a significant fraction of available supercomputing capacity worldwide-for example, NVIDIA quote that up to 20% of North American supercomputing cycles are used for QCD research [1]. QCD lies in a family of models known as gauge theories, and the numerical techniques developed to study QCD can also be applied to other gauge theories, forming a broader area of research known as Lattice Gauge Theory (LGT). Such theories may differ from QCD in a number | 3,069,749 | 115229961 | 0 | 16 |
of ways; computationally, the difference is typically the dimensionality and structure of the sub-matrices related to each point in the discretised space. These differences can have an impact on the demands that LGT code makes of the computer on which it runs-for example, altering the ratio of computations to communications demands. Non-QCD LGT has become of interest recently as a tool for theoretical physicists to probe physics Beyond the Standard Model (BSM), for example relating to recent discoveries at the Large Hadron Collider at CERN. Recent reviews of such techniques include [2], [3]. Benchmarks have previously been developed out of QCD codes, and many of these benchmarks have been adopted in common benchmark suites used for machine evaluation (for example, the NERSC MILC benchmark developed from the MILC research code [4]). However, the QCD codes used for these benchmarks do not have sufficient flexibility to probe BSM theories of physical interest. Thus in order to characterise the diverse performance demands of BSM LGTs, a novel benchmark is necessary, derived from (or at least approximating) a flexible LGT code. In this work, we present BSMBench, a benchmark satisfying this criterion, derived from the HiRep research LGT code [5], [6]. In the remainder of this paper, in section II we will outline the relevant details of LGT (in particular highlighting differences from QCD), then in section III, we will describe the methodology of the benchmark. In section IV we will then present some selected results, characterising some recent machines' performance in the tests set up by the benchmark, | 3,069,750 | 115229961 | 0 | 16 |
before concluding and suggesting future directions our work will take. A. Field content The "lattice" in Lattice Gauge Theory is a hypercubic array of points ("sites"), forming a discretised space (or spacetime). This space is most frequently four-dimensional, and the length in the three spatial directions is generally made to be the same, giving a total number of sites L 3 × T . Each lattice site has eight nearest neighbours (up and down in each of the four dimensions); the lines joining a point to a nearest neighbour is referred to as a "link". The lattice typically has periodic boundary conditions, so the number of links is four times the number of sites (avoiding double counting positive and negative links), with no edge corrections. A gauge theory will typically have one gauge field (the gluon field of QCD), and N f ≥ 1 "flavours" of fermion (the quark fields). On the lattice, the gauge field is an N × N complex-valued matrix on every link, while the fermion fields are M-dimensional complex-valued vectors at every site. N and M are integer-valued tunable parameters of the theory; N ≥ 2 may be freely chosen, while M ≥ N is constrained to certain values allowed by group theory (specifically, the Mvector must transform under some non-trivial representation of the group of which the N × N matrix is an element). Counting up, at each site the gauge field contributes 4 links ×(N × N) elements × 2 real numbers; i.e. 8N 2 real numbers per site. The | 3,069,751 | 115229961 | 0 | 16 |
fermion fields, meanwhile, contributes N f × M × 2 = 2MN f real numbers per site. The contributions to the whole lattice are then multiplied by L 3 T ; i.e. the gauge field comprises 8N 2 L 3 T real numbers, and the fermion fields 2MN f L 3 T . In the case of QCD, N = M = 3 (the three colors of QCD-red, green and blue). B. Dirac operator The physics of the fermions is encoded in the so-called "Dirac operator"; on the lattice this is a matrix relating all elements of the fermion field to all elements of the fermion field-that is to say, it is a (2MN f L 3 T ) × (2MN f L 3 T )-element matrix. The interactions in the fermion fields are taken to be nearest-neighbour, resulting in the Dirac operator being exceedingly sparse, and depend on the values of the gauge field elements. The primary task of the Monte Carlo code is to invert this matrix. This is typically done using a Conjugate Gradient (CG) or related algorithm, and as such the dominant (and to a reasonable approximation, only) contribution to the runtime comes from the routine to multiply the fermion field by the Dirac operator; we will call this Dphi 1 . This overwhelming dominance of execution time by one single subroutine has naturally led to it being the focus for optimization; in QCD applications it is not uncommon for Dphi to be hand-optimized with for example vector intrinsics and manual prefetching, | 3,069,752 | 115229961 | 0 | 16 |
rather than written in a naïve way and relying solely on the compiler. For more general LGT tools this kind of optimization is less practical; the need for generality in N and M precludes us from hard-coding highlyoptimised code in the way that QCD codes can for a fixed N and M. For example, the HiRep research code uses a code generator to produce sets of macros for the matrix-matrix and matrix-vector operations required, which are then called from Dphi. C. Parallelisation The hypercubic geometry and nearest-neighbour interactions found in the problem means that it naturally lends itself to spatial parallelisation, with the lattice being sliced up in each dimension, and each resulting piece of lattice being handled by a dedicated process (with processes generally communicating via MPI, although hybrid OpenMP+MPI approaches exist). The need to store and communicate boundary terms places a lower bound on the piece size that can be efficiently handled, and thus an upper bound on the degree of parallelisation for a particular problem size. III. The Benchmark The priorities when developing BSMBench were to reflect the computational demands and portability of the HiRep research code, to be able to probe more than one theory (i.e. set of values of (N, M) above-since the performance demands change as a function of these parameters), to run in reasonable time, and to spend sufficiently long that the run time is not dominated by startup overheads. Additionally, the test suite should be easily run by non-LGT specialists, so that it may, for example, be used | 3,069,753 | 115229961 | 0 | 16 |
by hardware vendors to quote performance of development machines early on in procurement cycles, without having to grant system access to end-users. The strategy chosen to meet these criteria was based on that of Lüscher [7]. It takes three tasks-two more elementary vector and matrix-vector operations, followed by the full Dphi-and in turn iterates them on randomly-generated fields for a fixed period of time 2 . (The CG inversion is currently not benchmarked, but can be requested as a check on the machine's numerics.) The number of floating point operations for each task has previously been calibrated, and thus the performance can be quantified by FLOPs/s = Number of iterations × FLOPs per iteration / Time Taken. The problem size is fixed, thus the benchmark probes strong scaling behaviour. Since the problem size in production is typically fixed by physical demands, research use of the benchmark is less interested in weak scaling; however, it is possible that it will be added in a future version. Even for the communications-dominated theory, each iteration has a fixed number of FLOPs, and so the FLOP/s rate for the benchmark gives a proxy to the performance. The advantage of using the same measure for all three theories is that the benchmark statistics may then be directly compared between theories. BSMBench was constructed by paring down the HiRep research code to the essential elements, thus the benchmarked code closely reflects the workloads in production runs. Further, optimisations can cross-pollinate between HiRep and BSM-Bench. Owing to the need to be able to | 3,069,754 | 115229961 | 0 | 16 |
adopt new HPC infrastructure as it becomes available, the HiRep research code is highly portable-in general, it can be run on a new machine simply by setting the correct compiler and running make. This property is inherited by BSMBench; in the results shown below, no code changes needed to be made to allow the benchmark to run, and to be reflective of the de factor usage of the research code the only optimisation performed was some tuning of the compiler flags. As mentioned, were system vendors to adopt the benchmark and optimise it more aggressively, the optimisations could be backported to the research code to benefit all users. The flexibility of the benchmark thus manifests in three ways: it is easily portable to many architectures, it can run in reasonable time on a diverse range of machine sizes, and most importantly, it can tune the relative demands on computation and inter-process communication. IV. Results BSMBench has been tested on a variety of HPC platforms, including IBM Blue Gene/P and /Q machines, an SGI ICE XA system with Haswell CPUs, Fujitsu x86 clusters (Westmere and Sandy Bridge-based, at HPC Wales), a Xeon Phibased cluster (at the Hartree Centre), commodity clusters (both Infiniband and gigabit Ethernet setups), and a Mac Pro workstation. 3 Details of MPI libraries, compilers, and compiler flags are shown in Table I; in all cases, the default MPI configuration was used, with no hand-tuning of process placement or run-time flags. Full results of each sub-test on every machine tested would be cumbersome to present | 3,069,755 | 115229961 | 0 | 16 |
here, thus we have chosen an interesting subset of results to highlight. Since the Dphi test is most representative of a typical production workload, it is this test that we focus on in presenting results. Results are plotted on a logarithmic scale, to avoid one or two data dominating the plots; plots are shown both of the total FLOP/s, and also of the FLOP/s normalised by the number of processes. In the case of perfect scaling, the latter plot would be a flat line. A. CPU-based machines On machines with all but the most memory-constrained nodes, all tests may be run on a single core, allowing an accurate look at the scalability of the code. We observe this in Fig. 1, where the three tests start off approximately comparably in performance, but the differing communications demands of the three theories used causes the scaling behaviour to differ. The importance of good interconnects for code of this type is clearly demonstrated by the sharp drop-off in performance once the parallelisation goes beyond a single node (16 cores) and starts hitting the network. Also shown are results for a 12-core Mac Pro workstation; this outperforms the cluster on small core counts, but is outperformed core-for-core once core counts increase. We do not expect HyperThreading to give us any advantage on these workloads, since the code makes heavy use of floating-point units, which are shared between the hardware threads. In Fig. 2, both Blue Gene/P and /Q machines show good scaling behaviour; however, core-for-core, the two machines have very | 3,069,756 | 115229961 | 0 | 16 |
similar performance, despite Blue Gene/Q's higher clock speed. The reason for this is vectorisation; as mentioned above, the code does not vectorise well to vectors longer than 2 double-precision floating-point numbers. This means that the 2-double vector units on Blue Gene/P can be used, but not the 4-double vector units on Blue Gene/Q. The performance on Blue Gene/Q fluctuates more as a function of number of processes than on Blue Gene/P; this illustrates the need to tune the process placement to take advantage of the network topology-in the case of Blue Gene/P, the problem sits advantageously on the network topology without the need for optimisation, whereas on Blue Gene/Q, the default layout is non-optimal for some parallelisations. As we might expect, at low process count the two Westmere clusters perform very similarly, while the Sandy Bridge cluster offers modest (∼ 2×) improvements in performance. At higher process counts, the two Westmere systems diverge somewhat; this is due to a greater freedom in choosing the job layout on this system, with 8 rather than 12 MPI tasks per processor dividing more nicely into the powers of 2 in the spatial parallelisation. Also as expected, the performance at low process count is very close between the three theories, but starts to diverge once communications starts to play more of a role. In Fig. 4 we show results for an SGI ICE XA system with Intel Haswell nodes (24 cores per node). This system has a dual-plane enhanced hypercube interconnect topology (i.e. there are two independent interconnect fabrics, each | 3,069,757 | 115229961 | 0 | 16 |
with their own switches and cables). The figure shows the benchmark results for both the single-plane (one fabric only) and dualplane (both fabrics work cooperatively) cases. At low process counts, the per-core performance is similar to the previousgeneration Intel architectures; however, at higher parallelisa- To briefly summarise these results, all machines tested that have high-speed interconnects (i.e. not Ethernet) show very good strong scaling at small to intermediate parallelisations. Those with more advanced interconnects (Blue Gene) show better strong scaling at the highest parallelisations than machines with simpler Infiniband arrangements. Both of these effects are most pronounced for the more communicationsintensive task. All of the larger machines tested were able to reach between 10 11 and 10 12 FLOP/s. Blue Gene required 4-8 times as many cores to reach comparable performance to x86. Does this mean, then, that any machine with a fast interconnect is suitable or preferable? This depends on a number of factors. At smaller problem sizes, the maximum parallelisation is reached more quickly, and so it would be preferable to minimise the use of the communications links (by using only one or a small number of nodes) rather than needing to procure the fastest available links. Larger problem sizes are only tractable through parallelisation, so the need for the fastest available interconnects becomes more pronounced. This analysis places Blue Gene/P and /Q as similarly desirable on a per-core basis; however, other obvious considerations then come into play-for example, that the footprint and power demands of a Blue Gene/Q would be significantly lower than | 3,069,758 | 115229961 | 0 | 16 |
those of an equivalent number of Blue Gene/P cores. B. Xeon Phi The benchmark has also been tested on a Xeon Phi (Knights Corner) system at the Hartree Centre. No modifications were necessary to allow the code to compile (beyond specifying the compiler). While the Phi needs 240 threads to keep all cores occupied and gain maximum performance, it was impossible to run that many MPI tasks due to the size of the required MPI buffers exceeding the card's memory. It was therefore necessary to use a hybrid MPI+OpenMP approach. Fig. 5 shows the results of these tests; where OpenMP was used, Figure 5: Results of testing the performance of a Xeon Phi node at the Hartree Centre. The horizontal lines are the performance of the host node, with two Xeon sockets, using 24 MPI processes (Per-process performance is not shown, since for the OpenMP runs values for the number of processes should attempt to use the entire card, with only the breakdown between MPI and OpenMP changing.) the number of threads was chosen as Number of threads = 240 Number of MPI tasks . For the comms and balance test, clearly the hybrid approach gives a performance gain over straight MPI; however, maximising the number of MPI tasks also improves performance over using OpenMP only. (The drop in performance of the compute test between 4 and 8 MPI tasks is currently poorly understood.) The performance is at best approximately half that of the two Xeon sockets on the host; one would hope that this relationship | 3,069,759 | 115229961 | 0 | 16 |
could be inverted if the code could be adapted to make use of the 512-bit vector unit in the KNC processor. V. Conclusion We have developed a novel benchmark, BSMBench, based on Beyond the Standard Model Lattice Gauge Theory. Unlike previous benchmarks based on QCD, it has the capacity to adjust the theory under study, and consequently modify the workload's demands in terms of the ratio of computations to communications. Thanks to this, BSMBench could be applied in a variety of user scenarios (e.g. as a monitoring and fault diagnostic tool or as a general-purpose performance evaluation utility) that transcend its original goals. We have tested this benchmark on a variety of recent supercomputing platforms, including CPUs and Xeon Phi coprocessors. Our results show good strong scaling in the presence of a sufficiently fast interconnect, and exhibit the expected splitting between theories under study. One limitation of the benchmark (and the underlying research code) is an inability to make use of vector units wider than two double precision floating-point numbers; work is underway to lift this restriction, which would significantly boost the performance on more modern architectures featuring AVX and QPX vector instructions. Other future improvements to BSMBench will be to reduce the reliance on parameter sets, with the core code instead able to calculate the necessary parameters, and potentially to introduce a weak scaling test. Interesting potential tests of the benchmark currently under investigation include assessing the relative performance of different MPI libraries on the same architecture, and observing how the results compare with those | 3,069,760 | 115229961 | 0 | 16 |
of other benchmarks-for example, those based on QCD (for example, [4]), and those based on Conjugate Gradient solvers (for example, [8]). | 3,069,761 | 115229961 | 0 | 16 |
Combination of rifaximin and lactulose improves clinical efficacy and mortality in patients with hepatic encephalopathy. Background Rifaximin and lactulose are common effective agents for hepatic encephalopathy (HE). Whether a combination of rifaximin and lactulose improves the efficacy and mortality in patients with HE compared with lactulose alone needs to be analyzed. Methods A systematic search was performed in electronic databases and other sources for possible studies focusing on combination therapy of rifaximin and lactulose for HE between January 2000 and February 2018. A meta-analysis was performed by the method recommended by the Cochrane Collaboration, and estimated effect size was presented as risk difference (RD), 95% CI, and the number needed to treat (NNT). Subgroup analysis, sensitivity analysis, and Trial Sequence Analysis were comprehensively performed to indicate the source of heterogeneity and risk of bias. Results Five randomized and five observational studies involving 2,276 patients were included. Combination therapy had a significant advantage in both clinical efficacy increase (RD 0.26, 95% CI 0.19-0.32, NNT 5) and mortality decrease (RD -0.16, 95% CI -0.20-0.11, NNT 9) in overall analysis. In the pooled analysis of randomized studies, combination therapy showed similar results in clinical efficacy (RD 0.25, 95% CI 0.16-0.35, NNT 4) and mortality (RD -0.22, 95% CI -0.33-0.12, NNT 5). Compared with lactulose, hospital stay was also reduced in combination therapy, and there was no significant difference in treatment-related adverse events between the two groups. Conclusion Combination of rifaximin and lactulose has beneficial effects on HE. Compared with lactulose alone, additional rifaximin increases clinical efficacy and decreases mortality. | 3,069,762 | 56894836 | 0 | 16 |
However, its effects on different types of HE are still uncertain. Introduction Hepatic encephalopathy (HE) is described as a brain dysfunction caused by liver insufficiency and/or portosystemic shunting; it manifests itself as a wide spectrum of neurological or psychiatric abnormalities ranging from subclinical alterations to coma. 1 HE is a severe complication of acute or chronic liver failure due to cirrhosis; its prevalence and severity is closely related to the underlying liver status. 2 The prevalence of minimal HE in cirrhosis ranged from 20% to 80%. For overt HE, the prevalence was 10%-14% in general cirrhosis, 16%-21% in decompensated cirrhosis, and 10%-50% in patients who adopted transjugular intrahepatic portosystemic, which was a major minimally invasive shunting surgery. 1,3 According to an over-5-year analysis, the mortality of HE in hospital was about 15%. 4 For cirrhosis patients without stable management or cure, 30%-40% of them suffered repeatedly recurrent HE in their survival periods. 5,6 Dovepress Dovepress 2 wang et al Besides intensive care and underlying liver disease treatment, 90% of HE patients can be treated through managing precipitating factors, including gastrointestinal bleeding, excessive protein intake, infection, hypokalemic alkalosis, constipation, hypoxia, or the use of sedatives and tranquilizers. 7 About 60%-80% of HE patients showed an elevated serum ammonia, and thus therapy was recommended, including non-absorbable disaccharides (NAD), rifaximin, L-ornithine-L-aspartate (LOLA), and branched chain amino acids (BCAA), which mostly aim to reduce the level of serum ammonia. 8 Among them, lactulose is the first choice and the most widely adopted NAD, while rifaximin emerged as an effective non-absorbed | 3,069,763 | 56894836 | 0 | 16 |
oral antimicrobial agent in recent years. [9][10][11] However, the two drugs presented similar clinical effects, which were demonstrated by largescale meta-analysis studies. 12 Because rifaximin and lactulose had different pharmacological mechanisms for HE, it is important to understand whether a combination of them would further increase the clinical efficacy compared to lactulose alone. Investigated by a series of small-scale studies, 13-21 the issue was not fully evaluated and has yet to provide a confirmed conclusion. Therefore, we performed a comprehensive systematic review and metaanalysis of published clinical studies aiming to determine the comparative efficacy and safety of combined rifaximin and lactulose with lactulose alone for current HE treatment. Search strategy Current meta-analysis was reported in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses Guideline. Comprehensive searches including electronic databases, clinical register centers, scholar search engine and manual search were completed using modified search strategy. Search terms were "hepatic encephalopathy, rifaximin, lactulose, plus, and combination". Online search was carried out by searching the websites of PubMed, EMBASE, the Cochrane Library, CNKI, Wanfang, ClinicalTrials.gov, Google scholar, and Baidu scholar between January 2000 and February 2018. Manual search was done by screening the references and citations of the similar studies. Publication language was limited to English and Chinese. Study inclusion and data extraction Both randomized and observational controlled studies investigating specific topics were considered in our analysis. Relevant meta-analyses and systematic reviews were also searched. Participants were patients suffering from covert (the West Haven criteria, grade 2) and overt HE (the West Haven criteria, grade | 3,069,764 | 56894836 | 0 | 16 |
2) due to liver cirrhosis. Combination therapy of rifaximin and lactulose was compared with lactulose alone in the treatment of HE, and the specific dose and usage of rifaximin were not restricted; also, there were no restrictions of the control type (bland control or placebo). Outcome measurements were clinical efficacy, mortality, and treatment-related adverse events. Two reviewers independently assessed the eligibility of all potential citations obtained from initial search, and a third reviewer checked the included studies. Data extraction was also completed by two independent reviewers. Baseline study information, patients' characteristic, comparison, type of HE, etiology and severity of HE, treatment duration, and follow-up period for evaluation of clinical outcomes were extracted from each study; the number of events of interest in studies were extracted for further statistical analysis purpose. Discrepancies regarding the extraction of data were resolved by the third reviewer. Outcome measures Primary outcome measures included clinical efficacy and mortality. Clinical efficacy was defined as improvement in the HE clinical syndrome with improved neurological status or a significant decrease in the HE index after treatment. Secondary outcome measure was treatment-related adverse events such as severe diarrhea, episodes of intense abdominal pain, and other gastrointestinal system reactions. Assessment of risk of bias Current analysis included both randomized and observational studies, and risk of bias located in each study was independently assessed by two reviewers by using different scales. Modified JADAD scale was used for randomized controlled trials, which contained items assessing sequence generation for the randomization, allocation of treatment concealment, blinding of participant and outcome | 3,069,765 | 56894836 | 0 | 16 |
measures, follow-up, and drop out. Trials with a score of 3 (total 7 scores) were considered as having high risk of bias and 3 were considered as having low risk of bias. Newcastle-Ottawa Scale (NOS) was used for observational cohort studies, which contained items assessing risk in exposed cohort representativeness, non-exposed cohort selection, ascertainment of exposure, comparability of cohort, interested outcome assessment, and follow-up periods. Studies achieving a score 4 (total 9 scores) were considered to be low risk of bias. Specific checklists for randomized controlled trials and observational Statistical analysis Review Manager (version 5.3, Cochrane Center, Copenhagen, Denmark) was used to perform this meta-analysis. In the software, I 2 and corresponding P-value were used to investigate the significance of statistical heterogeneity. Moderate, considerable, and substantial heterogeneity were set based on the value of I 2 range from 30% to 60%, 60% to 75%, and 75%. Clinical heterogeneity was handled by subgroup analysis according to study design (randomized studies or not). According to I 2 value, random-effects model or fixedeffects model was selected to analyze the data extracted from published reports. For each outcome measure under randomeffects model, the result of fix-effect model was reported only if the difference between the two models existed. Current meta-analysis used risk difference (RD) indicating the effect size of categorical variables and mean difference (MD) indicating the effect size of continuous variables together with their 95% CI. Both of them represented an average MD between the groups, and P-value 0.05 was considered to be statistically significant. For primary outcome measures, | 3,069,766 | 56894836 | 0 | 16 |
the number needed to treat (NNT) was also calculated, which represented the number of patients needed to treat to achieve a different clinical outcome. Based on the data of absolute risk reduction (ARR), experimental event rate (EER), and control event rate (CER) in low risk of bias studies, NNT was obtained as NNT=1/ARR and ARR=CER EER. The 95% CI of NNT was calculated based on the lower limit (LL) and upper limit (UL) of 95% CI of ARR: NNT LL =1/ARR UL and NNT UL =1/ARR LL . Statistical significance was identified by the range of 95% CI that it did not cross the value of 1. We further confirmed reliability of the results in Trial Sequential Analyses (TSA) by using TSA software (Copenhagen Trial Unit, Copenhagen, Denmark). The sequential analyses were based on randomeffects model, with α=5% and a test power of 80%, and model-based heterogeneity (diversity). The analysis results were judged as highly reliable when the z-curve crossed the trial monitoring curve, which indicated that cumulative result of included trials was relative enough to achieve a certain result even if the required sample size was not reached. Study information During the search process, 360 citations including 64 duplicates were obtained. After screening the titles and abstracts, 23 studies were assessed for eligibility by reading full-texts ( Figure 1). Finally, seven studies with full text and two studies with abstracts were included, among them one study divided patients into two groups (with HCC or not), 19 which were regarded as two separate trials. A total | 3,069,767 | 56894836 | 0 | 16 |
of 771 patients were assigned to combination group (rifaximin and lactulose) and 1,505 patients were assigned to control group (lactulose alone). Demographic characteristics are presented in Table 1. Six studies included overt HE patients (the West Haven criteria, grade 2), [15][16][17][19][20][21] and three studies included new onset and recurrent HE patients without detailed grade of the West Haven criteria. 13,14,18 The etiology of cirrhosis included alcohol, hepatic virus infection, and others. The severity of HE was judged by the abovementioned HE grade. The Child Pugh score and model for end-stage liver disease score judged the severity of underlying liver disease. Among them, HE grade 1 was diagnosed as overt HE, which indicated that clinical findings such as lethargy or apathy, disorientation for time, and obvious personality change were reproducible. The dose of rifaximin was fixed in the studies that 1,100 mg therapy was adopted in five studies, 13,14,16-18 and 1,200 mg therapy was adopted in four studies. 15,[19][20][21] The volume of lactulose varied in each study, ranging from 60 mL to 180 mL (the amount was 667 mg in 1 mL). The treatment duration was mainly less than or equal to 10 days, and only one study reported a maximum treatment duration of 15 days. 13 The follow-up period for analysis of clinical efficacy and mortality was similar to treatment duration, as the outcomes were mostly measured during hospital stay, except for two studies that had a follow-up period 180 days. 18,19 Risk of bias Five randomized studies were assessed by JADAD scale; 13,15,16,18,21 three | 3,069,768 | 56894836 | 0 | 16 |
of them had a score 3 15,18,21 and three of them designed a placebo control. 15,16,21 Four observational studies were assessed by NOS scale, 14,17,19,20 and three of them had a score 4. 17,19,20 Both of them were assessed by SIGN scale, where two studies were judged as high quality, 19,20 five studies were judged as acceptable, and only one study was judged as low quality. 14 The assessment result of risk of bias is shown in Table 2. Clinical efficacy Six studies evaluated the effects of the combination of rifaximin and lactulose on HE 13,15-17,20,21 ( Figure 2). The metaanalysis result showed that combination therapy significantly increased clinical efficacy compared with lactulose alone in HE patients (RD 0.19, 95% CI 0.09-0.29, P=0.0002). The heterogeneity value of I 2 was 59%, and one study was identified to be responsible for this moderate heterogeneity through inverted funnel plot. 21 After excluding it, the value of I 2 was reduced to 35%; the sensitivity analysis result was consistent with that of before (RD 0.26, 95% CI 0.19-0.32, P0.00001). Randomized studies were regarded to be better than non-randomized studies in the study design; accordingly, we analyzed the result based on only randomized studies ( Figure 3). In the new pooled analysis, three studies containing 342 HE patients also demonstrated a significant increase in clinical efficacy of combination therapy (RD 0.25, 95% CI 0.16-0.35, P0.00001, I 2 =0%). 13,15,16 The NNT was 5 (95% CI 3.45-11.11) in primary analysis, while it was 4 in sensitivity analysis (95% CI 3.13-5.26) and | 3,069,769 | 56894836 | 0 | 16 |
randomized study analysis (95% CI 2.86-6.25). In TSA analysis, we set α to 5%, test power to 80%, control group incidence to 54%, relative risk reduction to -26%, and heterogeneity correction to 35%. The required sample size was 615 participants. The monitoring boundary was crossed in 2013, which confirmed the reliability of current meta-analysis result with enough required sample size ( Figure 4). Inverted funnel plot indicated a low risk of bias ( Figure 5). In Figure 7). 15,16 The NNT was 9 (95% CI 5.26-33.33) in primary analysis, 6 (95% CI 5.00-9.09) in the sensitivity analysis, and 5 (95% CI 3.03-8.33) in randomized study analysis. In TSA analysis, we set α to 5%, power to 80%, control group incidence to 67%, relative risk reduction to 16%, and heterogeneity correction to 41%. The required sample size was 1,248 participants. The monitoring boundary was crossed in 2014, which confirmed the reliability of current meta-analysis result with enough sample size (Figure 8). Inverted funnel plot indicated low risk of bias ( Figure 9). Hospital stay Three studies reported the data of hospital stay. Two of them were pooled analyzed, and meta-analysis results showed that combination therapy significantly reduced the duration of hospital stay (MD -2.89, 95% CI -3.52 to -2.25, P0.00001). 15,16 The other study only reported the data of median hospital stay between the groups (6/8, P=0.09), 18 which found no significant difference. Adverse events Four studies compared the treatment-related adverse events ( Figure 10). In the pooled analysis, there was no significant difference in combination therapy | 3,069,770 | 56894836 | 0 | 16 |
and lactulose alone (RD -0.06, 95% CI -0.24 to 0.13, P=0.56). One study was found to be responsible for the substantial heterogeneity (I 2 =90%); after excluding it, 14 the I 2 value was obviously reduced (I 2 =0%). The sensitivity result was similar to before (RD 0.0, 95% CI -0.03 to 0.02, P=0.63). Discussion HE is one of the most common causes of death and would cause many precipitating factors in cirrhosis patients. 22 According to the recommendations in the 2014 practice guideline, controlling precipitating factors is of paramount importance in the management of HE. Besides precipitating factor management, lactulose is the first recommended agent. Rifaximin, oral BCAA, intravenous LOLA, neomycin, and metronidazole can also be used as alternative or additional agents. 1 However, high-level evidence such as meta-analysis or systematic review is still lacking. The current meta-analysis was the first to quantitatively evaluate the effective effects of combination therapy of rifaximin and lactulose vs lactulose alone for the management of HE. We included five randomized and five observational controlled studies, with a sample size of 2,276 HE patients, and demonstrated that combination therapy significantly improved the clinical efficacy, mortality, and hospital stay without increasing treatment-related adverse events in HE. As stated within the practice guideline, current definition, diagnosis, classification, or the treatment of HE are not universally accepted, and the complex pathogenesis of HE is difficult to clarify. 1 Currently, clinical observations always exhibit a high average level of ammonia in high grade HE patients; thus, the ammonia hypothesis is a widely accepted premise | 3,069,771 | 56894836 | 0 | 16 |
that leads to frequent assessment of ammonia concentrations. 23 Conventional pharmacologic treatment of HE consisted of NAD since 1966, which mainly included lactulose and lactitol. 9 A meta-analysis including 38 randomized trials published in 2016 had sufficiently confirmed the effectiveness of the routine use of NAD to reduce the production and absorption of gut-derived neurotoxin ammonia in clinical practice, as well as the potential roles of catharsis, ammonia metabolism, and gut microbiome adjustment. 9,24 For a specific HE patient, the serum level of ammonia sometimes did not correlate with the severity of clinical symptoms. Meanwhile, clinical analysis reported that lactulose therapy was related to a non-response rate as high as 22%. 25 Accordingly, improved therapy based on lactulose is warranted since no certain method was introduced to identify a non-responder of lactulose therapy. Confirmed by sensitivity analysis, subgroup analysis, Trial Sequential Analysis, and publication bias analysis, our meta-analysis shows that a combination of rifaximin to lactulose can significantly improve both clinical efficacy of HE manifestations (NNT =4) and mortality (NNT =5) mainly for overt HE patients. Based on limited data in one randomized study, 15 the cause of deaths was significantly reduced only in aspect of sepsis, which indicated that combination therapy may improve clinical outcome by reducing the incidence of severe and systemic infection and inflammation. In past years, multiple factors were included in the explanation of HE progress; among them systemic inflammation and infection burden were originated and induced by gut because of bacterial overgrowth and gut-derived endotoxin seemed to be the most attractive | 3,069,772 | 56894836 | 0 | 16 |
explanation. 19,26,27 Therapies targeting the infection and inflammation deriving from the gut were effective, and a new development of non-absorbed rifaximin with patient tolerance became a promising agent. 11 The therapeutic actions of rifaximin were reported to be twofold in reducing the number 26 Therefore, improved clinical outcomes in HE patients after applying additional rifaximin may be achieved by interrupting the pathway of gut-derived local and systematic infection and inflammations, as well as by inhibiting the potential of involvement and injury in organs including liver, lung, brain, and the body's immune system. All of these would contribute to a significant reduction in length of hospital stay. Lactulose is reported to be associated with increased mild adverse events in the gastrointestinal system, such as diarrhea and abdominal pain. 28 Although it may lead to some temporary discomfort in HE patients, most do not need clinical interventions. Compared with lactulose, rifaximin can be better tolerated with fewer incidences of adverse events. It was reported that compared with other absorbed and systemic antibiotics, non-absorbed rifaximin induced lower risk of bacteria resistance; its plasma concentration was negligible, which indicated a very low risk of bacteria selection outside the gut. 15,29 In the current study, a combination of the two agents did not increase any risk of adverse events related to the treatment. Combined rifaximin and lactulose for hepatic encephalopathy Limitations existing in the meta-analysis mainly included study quality, outcome measures, and targeted population. Randomized studies are regarded as higher level evidence than non-randomized studies. The current meta-analysis included both randomized | 3,069,773 | 56894836 | 0 | 16 |
and non-randomized studies to enable a large enough sample size. For the sensitivity analysis, excluding non-randomized studies did not alter the results despite a small difference in the value of effect size. For outcome measures, the current meta-analysis can only analyze data extracted from original reports. Clinical efficacy was judged by different centers; this difference across the studies may induce moderate heterogeneity although it is comparable between groups within each study. Development of a valid and standard test scale of HE is necessary in future research. Currently, HE treatment is mainly targeting at overt HE, because of lower diagnosis rate in covert HE. Combination therapy achieves better clinical outcomes than lactulose alone. 8,30 However, whether rifaximin and lactulose therapy should be immediately recommended as the first-line treatment for overt HE or an additive therapy of rifaximin for only lactulose-non-response overt HE still needs to be determined. Kang et al performed a cost-analysis study, and showed that the 1-year incremental cost was as high as $85,560 to increase the survival rate in one patient. 19 In another study, Courson et patients receiving combination therapy than lactulose alone. 18 Further cost-effective analyses are needed. Combination therapy of rifaximin and lactulose has beneficial effects on HE. Compared with lactulose alone, combination therapy increases clinical efficacy and decreases mortality in HE patients. However, its effects on different types of HE are still uncertain. Conclusion Combination therapy of rifaximin and lactulose has beneficial effects on HE. Compared with lactulose alone, combination therapy increases clinical efficacy and decreases mortality in HE patients. | 3,069,774 | 56894836 | 0 | 16 |
However, its effects on different types of HE is still uncertain. 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 the features 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. | 3,069,775 | 56894836 | 0 | 16 |
Biomarker Amplification by Serum Carrier Protein Binding Mass spectroscopic analysis of the low molecular mass (LMM) range of the serum/plasma proteome is a rapidly emerging frontier for biomarker discovery. This study examined the proportion of LMM biomarkers, which are bound to circulating carrier proteins. Mass spectroscopic analysis of human serum following molecular mass fractionation, demonstrated that the majority of LMM biomarkers exist bound to carrier proteins. Moreover, the pattern of LMM biomarkers bound specifically to albumin is distinct from those bound to non-albumin carriers. Prominent SELDI-TOF ionic species (m/z 6631.7043) identified to correlate with the presence of ovarian cancer were amplified by albumin capture. Several insights emerged: a) Accumulation of LMM biomarkers on circulating carrier proteins greatly amplifies the total serum/plasma concentration of the measurable biomarker, b) The total serum/plasma biomarker concentration is largely determined by the carrier protein clearance rate, not the unbound biomarker clearance rate itself, and c) Examination of the LMM species bound to a specific carrier protein may contain important diagnostic information. These findings shift the focus of biomarker detection to the carrier protein and its biomarker content. Traditionally, mass spectrometry analysis of complex protein mixtures involves an upfront chromatographic separation step [4][5][6], followed by an enzymatic fragmentation of the separated proteins for direct MS-MS identification [14][15][16]. In contrast to such traditional approaches, recent applications of mass spectrometry biomarker analysis may have been successful because no enzymatic treatment was conducted and have utilized the native undigested serum proteome as a launch point for biomarker discovery [9][10][11][12][13]. Small molecular mass biomarkers may in | 3,069,776 | 3788714 | 0 | 16 |
fact be created by specific disease related enzymatic cleav-age or posttranslational modification of larger proteins. Enzymatic treatment prior to analysis may destroy or mask this information content by cleavage of disease biomarkers and by creating large quantities of enzymatic fragments from high abundance proteins. Several groups of investigators have reported the discovery of low molecular mass diagnostic biomarkers using direct mass spectrometry analysis of nonenzymatically treated serum or plasma [8][9][10][11][12][13]. Since the vast majority of the proteins in the test sample are above the range of accurate detection (> 10 kDa) by the mass spectrometer, the low molecular mass biomarkers that emerge from the analysis must be derived from two possible sources: a) free in solution or b) bound to larger carrier molecules. SELDI-TOF and MALDI-TOF mass spectrometry analysis involves the laser-induced ionization of dried mixtures of molecules [17,18] adherent to a surface. The generated ions can represent biomarkers originally existing in the free unbound phase or existing in a bound state with larger proteins. Since the low molecular mass serum/plasma proteome is largely uncharacterized, there has been no previous analytical or experimental estimate of the relative proportion of small molecular mass biomarkers that exist in the free versus the bound phase. Under the assumption that the low molecular mass biomarkers contain important diagnostic information and that protein biomarkers useful in disease detection are of very low abundance, the search for biomarkers usually begins with a separation step to remove the abundant high molecular mass "contaminating" proteins such as albumin, thyroglobulin, and immunoglobulins so that the | 3,069,777 | 3788714 | 0 | 16 |
analysis can focus on the lower molecular mass region [14][15][16]. The purpose of the present study is to examine the proportion of the low molecular mass species bound to the high molecular mass fraction of the serum/plasma proteome. From a physiologic perspective, free phase low molecular mass molecules (< 30 kDa) should be rapidly cleared through the kidney [19][20]. Consequently such rapid physiologic excretion may significantly reduce the concentration of free phase low molecular mass species to a level below detection. In the face of the vast excess of high molecular mass serum proteins, it may be likely that low abundance and low mass species will tend to bind large carrier proteins. The abundant high molecular mass carrier proteins exist above the cut-off for kidney clearance [19,20], and hence possess a half-life that is many orders of magnitude larger than small molecules. Circulating carrier proteins may thus become the reser-voir for the accumulation and amplification of bound low mass biomarkers. Non-covalent association with albumin has been shown to extend the half-life of shortlived proteins introduced into the circulation [21][22][23]. The fact that many investigators now are employing and/or developing methods by which the higher abundance proteins above 30 kDa are specifically subtracted from native serum/plasma prior to analysis [4][5][6] may dramatically diminish the chances of finding the important low abundance and low molecular mass disease biomarkers. We examined the proportion of low molecular mass species detectable by SELDI (surface enhanced laser desorption and ionization) that are associated with the higher molecular mass serum proteome. Human serum | 3,069,778 | 3788714 | 0 | 16 |
was fractionated into high molecular mass and low molecular mass native fractions. Each fraction was assayed by SELDI to assess whether the preponderance of low molecular mass ions is found in the low or the high molecular mass fraction. We further examined the subpopulation of molecular species bound to albumin compared to the total carrier protein fraction. We explored whether SELDI-TOF identified biomarkers correlating with presence of ovarian cancer were associated with high molecular mass carrier proteins. Finally, we explored the theoretical implications of biomarker amplification due to carrier protein binding. The overall aim was to evaluate the use of carrier proteins as an affinity capture means for disease relevant biomarkers. Serum samples Serum samples were derived from the ovarian cancer clinical study set of the National Ovarian Cancer Early Detection Program, Northwestern University. The full characteristics of this study set have been described previously [9], and are posted in detail on http://clinicalproteomics.steem.com. Bioinformatic serum proteomic pattern analysis was conducted as described previously [9]. Further details are provided on http://clinicalproteomics.steem.com. Mass spectrometry Surface Enhanced Laser Desorption and Ionization (SELDI) was conducted using a PBSII (Ciphergen Systems) as described previously [17,18]. Human serum was collected and anonymized as previously reported [17,18]. Analysis was conducted on a WCX2 (weak cationic exchange) chip. The serum was fractionated into molecular mass classes under native conditions. Mass fractionation Thirty microliters of unfractionated human native serum was introduced into a Sephadex G-25 or a Sephadex G-50 molecular sieve spin column according to the manufacturer's instructions. The column was centrifuged at 3000 x | 3,069,779 | 3788714 | 0 | 16 |
g for three minutes, and approximately 30 microliters of eluate containing the high molecular mass fraction was collected. The eluate was treated with 50% acetonitrile (w/w in water) to dissociate bound molecules for 30 minutes and was transferred to the inlet of a molecular filtration microcolumn. (Microcon YM-30 Millipore Centrifugal Filter Device) The column was centrifuged at 1000 x g. The eluate containing the low molecular weight fraction was collected. All fractions at each stage were sampled and one microliter was analyzed by SELDI. Albumin separation Segregation of albumin and its low molecular mass binding constituents was conducted using the Montage Albumin Deplete Kit. 100 µL of human serum was diluted one to one with Equilibration Buffer provided with the kit for a final volume of 200 µL, and vortexed. The column was rehydrated twice with 400 µL of Equilibration Buffer and centrifuged through the column insert for 2 minutes at 2,000 rpm. 200 µL of diluted serum was introduced into the rehydrated albumin column and centrifuged for 2 minutes at 2,000 rpm. The eluate from the column contained the serum without albumin. The bound fraction contained the albumin and the low molecular weight species bound to the albumin. We then added 400 µL EAM solution composed of 50% acetonitrile and 0.1% TFA to the column to strip the column and dissociate albumin from its bound species. After 30 minutes, EAM solution was centrifuged through the column at 2,000 rpm for 3 minutes. The eluate contained the dissociated albumin and low molecular weight species that bind | 3,069,780 | 3788714 | 0 | 16 |
to albumin. Analysis of the proteins bound to the column using ion trap mass spectrometry was performed in line with an LCQ Classic MS (ThermoFinnigan, San Jose, CA) with a modified nanospray source. Dynamic exclusion of the three most abundant peptide hits from a full MS scan were selected for MS/MS analysis by collision induced dissociation with normalized collision energy of 35% and an activation time of 30 ms. Ion spray voltage was 2.00 kV with a capillary voltage of 26.20 V and a capillary temperature of 160 • C. Results for MS/MS scans were searched and compared with theoretical spectra in the Sequest Browser database specified for human proteins. SELDI/TOF WCX2 protein arrays were processed in a bioprocessor (Ciphergen Biosystems, Inc). 100 µl of 10 mM HCl was applied to the protein arrays in the bioprocessor and allowed to incubate for 5 minutes. The HCl wash was aspirated and discarded and 100 µl of H 2 O was applied and allowed to incubate for one minute. The H 2 O was aspirated and discarded, then reapplied for another minute. 100 µl of 10 mM ammonium acetate with 0.1% TritonX was applied to the surface and allowed to incubate for 5 minutes. The ammonium acetate was aspirated and discarded. A second application of ammonium acetate was applied and allowed to incubate for 5 minutes. The chip surfaces were then dried using a vacuum to remove any excess amount of liquid. Five microliters of raw sera, or molecular mass fraction, or eluate, was then applied to each | 3,069,781 | 3788714 | 0 | 16 |
chip surface and allowed to incubate for 55 minutes. Each protein chip was washed six times with 150 µl of PBS and H 2 O and then vacuum dried. Cross contamination was eliminated between spots by using a bioprocessor gasket. The gasket was removed and 1.0 µl of a saturated solution of the Energy Absorbing Molecule cinnamic acid (25% saturation) in 50% (v/v) acetonitrile, 0.5% trifluoroacetic acid was applied to each spot on the protein array twice allowing the solution to dry between applications. Mathematical modeling The kinetics of biomarker production, carrier protein(s) binding, and clearance, was modeled as a deterministic compartmental model with first order kinetics. Association of LMW species with HMW carrier proteins Analysis of native human serum fractionated into high and low molecular mass fractions revealed that the vast majority of low molecular mass serum / plasma biomarkers detectable as MS ions, exist bound to large carrier proteins. SELDI analysis of native serum frac-tions of high and low molecular mass, shown in Fig. 1, demonstrate that virtually all of the detectable ions are derived from molecular species bound to large carrier proteins. In fact, removal of the high molecular mass proteins under native conditions ( Fig. 1(B)), a common method used for biomarker discovery [14][15][16], removes a significant proportion of the ions generated by SELDI-TOF. Comparing the spectra of Figs 1(A) to 1(C) indicates that the majority of ions generated from unfractionated serum are derived from species associated with larger carrier proteins. Figure 1(D), displays the ion spectra of species previously bound, and | 3,069,782 | 3788714 | 0 | 16 |
then dissociated and separated from the higher molecular mass fraction. The intensity and number of many ion species is augmented comparing Figs 1(A), and 1(C). Populations of LMW species associated specifically with albumin Figure 1(E) displays the ions associated with the nonalbumin carrier proteins, and Figure 1F displays the ions generated from species bound only to albumin. A significant proportion of the ions in the spectra appear to be derived from species associated with albumin, compared to non-albumin carrier proteins. We verified through microcapillary LC MS/MS that our albumin bound fraction acquired through stripping the Montage Albumin Deplete Column was entirely albumin and its bound low molecular mass species. Since there was no indication of a significant proportion of other high molecular mass proteins bound to the albumin specific column, we can assume that the low molecular mass species detected were derived from a specific association with albumin, or at least aggregated with and co-separated with albumin. Furthermore, we have positively identified hundreds of low molecular mass species after dissociation from their albumin carrier. Additional studies which will be submitted for publication elsewhere involve characterizing the entire repertoire of low molecular mass species bound to individual serum carrier proteins by LC MS/MS. We next addressed the question as to whether the ions generated from the species bound to albumin contained disease biomarker information. SELDI-TOF ion patterns, generated on the WCX2 chips, correlating with ovarian cancer were identified by methods described previously [9] (http://clinicalproteomics.steem.com). Two clinical sera data sets were employed and all spectra are provided on | 3,069,783 | 3788714 | 0 | 16 |
the website for downloading as follows: Dataset 8-7-02: Ovarian Dataset 8-7-02.zip, Ovarian Sample Info 8-7-02.xls. The sample set included 91 unaffected controls and 162 ovarian cancers. The following selection of ion mass/charge (m/z) values generated a pattern that was 100% predictive in the training and blinded testing -2760. 6685 Randomly selected representative serum samples from this study set were analyzed by MALDI-TOF comparing the spectra generated by unfractionated sera to the spectra generated only from the species bound to albumin. As demonstrated in Fig. 2 the spectra generated from the species bound to albumin is complex and exhibits a number of differences between the cancer and the unaffected ("normal") cases shown in the example. Comparing the peak intensities between the unfractionated serum (containing all the carrier proteins with their associated or bound species), and the albumin-bound fraction (Fig. 2) indicates that a significant proportion of putative disease biomarkers may be associated with albumin. Figure 3 is an example of ion 6631.7043, a member of the ion pattern 100% correlated with ovarian cancer in this clinical study set. Matched for dilution and amplitude, the predicate ion is highly associated with albumin, and the ionization intensity is augmented in the albumin bound fraction. This demonstrates the albumin binding selectivity of a specific SELDI-TOF ion associated with ovarian cancer. Figure 4 displays the ion spectra for a pooled ovarian cancer serum sample in which the ion species bound only to albumin are compared for different amounts of albumin captured on the column. The captured albumin with its associated species | 3,069,784 | 3788714 | 0 | 16 |
was denatured and its binding partners were dissociated. When a higher number of albumin molecules were stripped of their associated species, the amplitude and complexity of the LMW species, including those in the region of putative ovarian cancer biomarkers, were augmented. These data indicate that albumin capture is a feasible method for biomarker enrichment. Carrier protein concentration, C r (t): Dependency on clearance rate At any point in time, the total concentration of the biomarker is dependent upon the biomarker production rate, the biomarker clearance / excretion rate, the binding of the biomarker to a circulating carrier protein, and the clearance / excretion rate of the carrier protein. We can view the blood intravascular space as a single compartment with volume V . We define the concentration of the carrier protein r as Cr, where the rate of carrier protein production is k in,r , and the rate of its elimination or removal is k out,r . Then the change in the carrier concentration can be expressed as: Using the LaPlace Transform, the concentration of the carrier protein, at time t is: The initial conditions are C r (t) = 0 at t = 0. Amplification of biomarker concentration in the presence of the carrier We assume that a biomarker is continuously produced or shed from the tissue source over time. As shown by the experimental data, biomarker molecules can accumulate over time in a carrier-bound form. At steady state, the total concentration of a biomarker measured in a blood sample can therefore become elevated due | 3,069,785 | 3788714 | 0 | 16 |
to its association with the carrier protein. The level of amplification (A) of the biomarker concentration at steady state, due to the presence of the carrier protein can be defined as the following ratio where C B is the concentration of the biomarker. A = (3) C B in the presence of carrier protein(t) C B in the absence of carrier protein(t) . Plasma biomarkers reflecting a physiologic or disease state of perfused tissues are expected to exist at concentrations many orders of magnitude below the concentration of large carrier proteins such as albumin and immunoglobulins. The experimental findings are logical because, even with a low affinity for the carrier protein, the majority of the biomarker molecules will tend to be associated with the vast excess of circulating carrier protein. Consequently, it is also logical that the biomarker will take on the clearance rate of the carrier protein it is associated with. The concentration of the biomarker C B (t) as a function of time can therefore be described as the balance between the biomarker input production rate k in,b distributed in the blood volume V , and the loss or clearance of the biomarker bound to the carrier protein (k out,br C br (t)). If k in,b is assumed to be a simple constant production rate, a linear function of time, similar to the assumption for the carrier protein (Eq. (1)) assuming first order kinetics, where the clearance rate is a constant proportion of the carrier bound biomarker, k out,br C br (t), where "br" | 3,069,786 | 3788714 | 0 | 16 |
refers to bound biomarker, then Initial conditions C B (t) = 0 at t = 0. At t → ∞ or steady state, we have Because the biomarker bound to the carrier protein acquires the clearance rate of the carrier protein, k out,br = k out,r , the total steady state concentration of biomarker in the plasma becomes a simple function of the biomarker production rate and the clearance, excretion rate of the carrier protein. The results of this analysis reveal that the final total concentration of the biomarker measurable in a blood sample is inversely proportional to the clearance rate of the carrier protein to which the biomarker is bound. Table 1 is a series of computed solutions to Eq. (6) for a range of hypothetical biomarker production rates, and for a series of different named carrier proteins. The clearance rates for serum carrier proteins listed in Table 1 are known [21]. The clearance and excretion rate for free biomarkers was chosen to span the known range for small molecules [19][20][21]27]. For a carrier protein such as albumin, with a long half-life, the resulting amplification (Table 1) can be several orders of magnitude. Carrier protein amplification thus becomes a major factor determining whether a low abundance biomarker can reach a threshold of concentration that Table 1 Theoretical prediction based on Eq. (6) of measured total biomarker concentration for selected high abundance serum carrier proteins, as a function of biomarker production rate, and carrier protein half-life. The experimental results indicated that the majority of the biomarker | 3,069,787 | 3788714 | 0 | 16 |
species exists in a state of association with carrier proteins, whereby the clearance rate of the biomarker takes on the clearance rate of the carrier protein is above the lower limits of detection. Discussion A growing body of scientific studies supports the importance of the low molecular mass region of the serum proteome as an uncharted resource for biomarker discovery [9,28]. The experimental data of the present study supports the concept that the vast majority of small mass ions detected by mass spectrometry of native human serum exist in association with circulating carrier proteins of higher molecular weight. This conclusion has several important implications for biomarker physiology and biomarker measurement technology. Experimental data presented in Figs 1-4 reveal that the majority of ions generated by SELDI-TOF analysis are found to be associated with carrier proteins, rather than free in solution phase (1B versus 1C-F). Moreover, as shown in Figs 2 and 3, ion species altered in disease study sets may be those specifically captured on a single carrier protein. In the example, the carrier protein is albumin. In the past, extensive effort has been placed on separating and discarding the high abundance large carrier proteins in the native plasma so that the remaining low abundance, diseaserelated markers could be discovered. The present results demonstrate that the search for biomarkers must be directed to those molecules bound to the carrier proteins. Removal of high abundance serum/plasma proteins prior to proteomic analysis should be conducted after, not before, dissociation from binding partners. This separation approach has been attempted | 3,069,788 | 3788714 | 0 | 16 |
for 2-D gel analysis [4,5]. As shown in Fig. 4, albumin capture can be used as a means to enrich for disease relevant biomarkers specifically associated with albumin. This now provides a novel method to harvest the necessary quantities of biomarker species required for sequencing and identification. The first implication from this study is that the concentration of a biomarker measured in serum or plasma is directly related to the clearance rate or half life of the carrier protein, not the biomarker clearance rate itself. As shown in Eq. (6), the concentration of the biomarker is a function of the ratio between the biomarker production rate from the tissue and the clearance rate of the carrier protein. This means that carrier protein binding amplifies the total biomarker concentration levels measured in serum or plasma. Amplification occurs because the carrier protein acts as a reservoir to accumulate the biomarker over time, as the tissue is continuously producing the biomarker. Thus a biomarker produced by a small volume of tissue such as the ovary [9,10], prostate [8,12,13], or breast [11], at a low concentration (e.g. one femtomole per day) can accumulate to a concentration of one picomole in the serum because it binds with a carrier protein with a much longer half-life. In this example the existence of the carrier protein can raise the concentration of the biomarker to a range detectable by conventional assay technology [29]. Without the carrier protein, the free biomarker would be rapidly cleared by the kidney and would therefore reside at a steady | 3,069,789 | 3788714 | 0 | 16 |
state concentration many fold below the detection limits of assay technology. The impact of this conclusion extends beyond current mass spectrometry detection technology. Small biomarkers are commonly not the province of two-site sandwich immunoassays [30,31]. This is because it is difficult to develop two antibody-binding sites on the same small molecule. In contrast, if the first half of the immunoassay sandwich was the carrier protein and the second half was the small biomarker, a sandwich immunoassay could be achieved. It is logical that the biomarker clearance rate becomes the carrier protein clearance rate because the carrier protein, even if it has low affinity for the biomarker, is in vast excess. This means that if we know the clearance rate of a given carrier protein, and we know the serum/plasma concentration of the biomarker bound to that carrier protein; we can estimate a lower limit for the continuous production rate by the tissue (Eq. (6)). Thus, if the concentration of a generic, bound biomarker α is 3.8 ng/mL (42.22 fmol/mL, where the biomarker α with its carrier protein have a combined molecular weight of 90 kDa) and the carrier protein half-life is 2.43 days, then the production rate of α from the tissue is at least 45,000 femtomoles per day [21,25,27,32,33]. If α is produced by a one cubic centimeter tumor composed of 10 9 cells, then each cell would produce approximately 16,000 molecules per day. This approximation is consistent with previous experimental findings [32]. These data indicate that the low molecular mass proteome, existing within the | 3,069,790 | 3788714 | 0 | 16 |
range detectable by MALDI-TOF, exists predominately in the bound phase. We propose that technologies that focus on efficient capture of the carrier proteins and specific elution of the low molecular weight biomarkers will yield the greatest amount of diagnostic information. The bound biomarkers may exist in concentrations ten to 500 times greater compared to their free counterparts. Since the carrier proteins exist in vast excess compared to the biomarkers, it is unlikely that the carrier proteins will become saturated with bound biomarkers. Moreover, based on its unique affinity topology [34], each carrier protein may have its own constellation of bound biomarkers. Indeed, the distribution of biomarkers among specific plasma/ serum carrier proteins may have important diagnostic information. Finally, these findings lead to the concept of artificial carrier molecules designed to harvest specific populations of biomarkers associated with target organs or diseases. | 3,069,791 | 3788714 | 0 | 16 |
The IL-33 Receptor ST2 Regulates Pulmonary Inflammation and Fibrosis to Bleomycin Idiopathic pulmonary fibrosis is a progressive, devastating, and yet untreatable fibrotic disease of unknown origin. Interleukin-33 (IL-33), an IL-1 family member acts as an alarmin with pro-inflammatory properties when released after stress or cell death. Here, we investigated the role of IL-33 in the bleomycin (BLM)-induced inflammation and fibrosis model using mice IL-33 receptor [chain suppression of tumorigenicity 2 (ST2)] mice compared with C57BL/6 wild-type mice. Unexpectedly, 24 h post-BLM treatment ST2-deficient mice displayed augmented inflammatory cell recruitment, in particular by neutrophils, together with enhanced levels of chemokines and remodeling factors in the bronchoalveolar space and/or the lungs. At 11 days, lung remodeling and fibrosis were decreased with reduced M2 macrophages in the lung associated with M2-like cytokine profile in ST2-deficient mice, while lung cellular inflammation was decreased but with fluid retention (edema) increased. In vivo magnetic resonance imaging (MRI) analysis demonstrates a rapid development of edema detectable at day 7, which was increased in the absence of ST2. Our results demonstrate that acute neutrophilic pulmonary inflammation leads to the development of an IL-33/ST2-dependent lung fibrosis associated with the production of M2-like polarization. In addition, non-invasive MRI revealed enhanced inflammation with lung edema during the development of pulmonary inflammation and fibrosis in absence of ST2. respiratory diseases are promoted, at least initially by a strong inflammatory response (1). Studies delineating the precise role of inflammation and immunity are needed to better characterize the mechanisms involved. We previously showed that bleomycin (BLM) induces uric acid and | 3,069,792 | 49396821 | 0 | 16 |
ATP release that act as a danger signals, ATP acting through its receptor P2X7, both leading to NLR pyrin domain containing 3 inflammasome-dependent IL-1β secretion and lung fibrosis (5)(6)(7)(8). However, the role of IL-1 family member interleukin-33 (IL-33) and its receptor suppression of tumorigenicity 2 (ST2) in pulmonary inflammation and fibrosis is unclear. IL-33 acts as a dual-function protein, with both nuclear and extracellular effects when released as a danger signal upon cellular damage (9,10). IL-33 is constitutively expressed in the nucleus of endothelial and/or epithelial cells where it associates with chromatin (11)(12)(13)(14) and is involved in maintaining barriers (15). Extracellular IL-33 interacts with the ST2 receptor which is either expressed on the cell surface (membrane-bound ST2L) or shed from these cells [soluble ST2 (sST2)], thereby functioning as a "decoy" receptor to bind and efficiently inhibit IL-33 activity (16). ST2L is closely related to the IL-1 receptor 1 (IL-1R1) and binding of IL-33 on ST2L activates NF-κB pathway (17), suggesting that it regulates the response (16,18). IL-33 is produced as a precursor or in full-length form (266 amino acids in mice) with the typical IL-1-like cytokine domain localized at the C terminal (19). Unlike IL-1β, full-length IL-33 is bioactive and may be processed by serine proteases secreted by activated neutrophils, generating 20-to 30-fold more active forms (20). ST2 expression was shown to be increased in the mouse model after BLM administration or in patients upon acute exacerbation of pulmonary fibrosis (21,22). Moreover, recent studies showed that IL-33 potentiates BLM-induced lung injury (23). Another study showed that | 3,069,793 | 49396821 | 0 | 16 |
a treatment with a lentivirus expressing sST2 improved survival rate, reduced weight loss, and profoundly attenuated pulmonary inflammatory cell infiltration, fibrotic changes, and levels of IL-33 and TGF-β1 levels in the airways after BLM (24). By contrast, following transient ST2 overexpression before BLM administration, ST2 was shown to dampen the initial stage of acute lung injury (25) and to promote lung fibrosis in a ST2-dependent manner through the induction of alternatively activated macrophages and innate lymphoid cells (26) while sST2 suppressed the initial stage of BLM-induced lung injury (25). The increased concentration of sST2 in serum may be a biomarker of IPF (21). Full-length IL-33 may be pro-inflammatory and profibrotic effects through its intracellular form, IL-33 remaining predominantly intracellular (27). However, the role of IL-33 and IL-33/ST2 signaling in establishment of pulmonary inflammation and fibrosis is not well understood. The IL-33/ST2 axis was shown to have an anti-inflammatory effect (26). Here, we revisited the role of the IL-33/ST2 in the BLM model of pulmonary inflammation and fibrosis by both classical immunologic methods and magnetic resonance imaging (MRI). The results indicate that MRI provides important information to monitor the evolution of edema in the mouse lung. We show that the absence of ST2 results in increased early inflammation with fluid retention, but decreased fibrosis. Our data reveal that the IL-33 pathway leads to shift from acute pulmonary inflammation and remodeling induced by lung damage to an excessive lung repair response with fibrosis through the production of M2-like polarization. Regulation of this IL-33/ST2 axis may attenuate pulmonary fibrosis | 3,069,794 | 49396821 | 0 | 16 |
and enhance recovery. In addition, we show that MRI allows a rapid non-invasive detection of lung edema during the development of pulmonary fibrosis. resUlTs sT2-Deficient Mice Display exacerbated airway inflammation to BlM 24 h after BlM exposure We observed that IL-33 was expressed in lung homogenates of naïve C57BL/6 wild-type (WT) mice and was significantly increased after BLM instillation ( Figure 1A) whereas IL-33 remained undetectable in bronchoalveolar lavage fluid (BALF) (data not shown). In order to decipher the role of the IL-33/ ST2 axis in the pulmonary inflammation induced by BLM, we performed saline or BLM intranasal instillation in WT and ST2deficient (ST2 −/− ) mice. Lung inflammation was characterized by an increased recruitment of total cells (Figure 1B), primarily neutrophils ( Figure 1C) in the airway of BLM-treated WT mice, which were drastically augmented in BLM-treated ST2 −/− mice. Exacerbated neutrophil recruitment in ST2 −/− mice was associated with increased myeloperoxidase (MPO) activity ( Figure 1D) and neutrophilic chemokine CXCL1/KC levels in the lungs (Figure 1E). In addition, monocyte chemokine CCL2/ MCP-1 levels were higher in the BALF and the lungs of BLMtreated ST2 −/− mice (Figures 1F,G). By contrast, levels of the interleukin-6 (IL-6) known to induce the expression of the IL-4 receptor on macrophages (28) were decreased in the BALF and the lungs of BLM-treated ST2 −/− mice (Figures 1H,I). Finally, expression of the remodeling factors matrix metalloproteinase (MMP)-9 ( Figures 1J,K) and of the tissue inhibitor of metalloproteinase (TIMP)-1 (Figures 1L,M) upregulated by BLM in BALF and lung of WT mice | 3,069,795 | 49396821 | 0 | 16 |
were higher in ST2-deficient mice. In line with the above findings, lung histological analysis showed increased inflammatory cell recruitment into the lung in BLM ST2 −/− mice when compared with WT counterparts (Figures 1N,O). sT2-Deficient Mice Display no Difference of Total cell, but reduced recruitment of alternative Macrophage 11 Days after BlM exposure To investigate the role of the IL-33 receptor ST2 in chronic inflammation and fibrosis, WT and ST2 −/− mice were instillated with BLM (3 mg/kg) or NaCl as control and inflammation was analyzed 11 days post-BLM treatments. Lung IL-33 levels remained elevated in BLM-instillated WT mice in comparison to NaCl mice (Figure 2A) and were slightly higher when compared with day 1 post-BLM ( Figure 1A), but were not detected in BALF (data not shown). When compared with WT mice, BLM-treated ST2 −/− mice displayed a trend, but not significant increase of total cell (Figure 2B), neutrophils (Figure 2C), macrophages 1 | ST2-deficient mice display an exacerbated inflammatory response to bleomycin (BLM) 24 h after exposure. Wild-type (WT) mice and ST2-deficient mice (ST2 −/− ) were instilled with 7.5 mg/kg of BLM or saline and inflammation parameters were assayed at day 1. Interleukin-33 (IL-33) contents in the lungs (a), total cell (B), and neutrophil (c) numbers in the bronchoalveolar lavage fluid (BALF) were significantly increased in BLM-treated ST2 −/− when compared with WT mice. Cell influx was in correlation with increased lung myeloperoxidase activity (D) and enhanced levels the neutrophil chemoattractant chemokine, CXCL1/KC in lung (e), of the monocyte chemoattractant CCL2/MCP-1 levels in | 3,069,796 | 49396821 | 0 | 16 |
BALF (F) and lung (g) and the interleukin-6 (IL-6) cytokine in BALF (h) and lung (i). The tissue remodeling factors matrix metalloproteinase (MMP)-9 and tissue inhibitor of metalloproteinase (TIMP)-1 were higher in BALF and lung (J-M). Histological lung sections of 5-µm were stained with picrosirius red (original magnification 200×). Histological micrographs showed increased parenchyma infiltrating cells (n) and increased inflammation score (O) after BLM instillation. Data are representative of three independent experiments and are expressed as mean values ± SEM (n = 4-6 mice per group, *p < 0.05, **p < 0.01, ***p < 0.001). ( Figure 2D), and lymphocytes ( Figure 2E) recruited into the BALF. In addition, using flow cytometry analysis (FACS) of cells recruited into the lung parenchyma, we observed not difference in total cell ( Figure 2F), neutrophil (Figures 2G,H), and lymphocyte (Figures 2I,J) numbers and percentages between WT and ST2 −/− mice. By contrast, we report a significant decreased in the number of CD11b + FA/80 + lung macrophage but only a trend of decreased frequency (Figures 2K,N) sT2-Deficient Mice Display Decreased expression of M2 Macrophage Mediators We then investigated expression of inflammatory mediators. Analyzing expression of the CCL17/TARC chemokine characteristic M2 macrophage profile, we report decreased levels of the CCL17/TARC chemokine in lung of ST2-deficient mice ( Figure 3A) in contrast to unchanged levels of the chemokine CCL5/RANTES ( Figure 3B). Moreover, performing multiplex analysis, we analyzed the expression of the characteristic M2 cytokines IL-4 and IL-5. BLM airway instillation induced IL-4 expression in lung ( Figure 3C) and IL-5 | 3,069,797 | 49396821 | 0 | 16 |
expression in BALF and lung (Figures 3D,E) in WT mice. We observed significant reduction of IL-4 levels in lung ( Figure 3C) and IL-5 levels in BALF and lung (Figures 3D,E) in ST2 −/− mice indicating that production of cytokines representative of alternative M2 macrophages and T helper type 2 (Th2) cells depends on ST2 signaling. In addition, we report reduction of the M2 polarizing cytokine IL-6 in BALF and lung (Figures 3F,G) in ST2 −/− mice in comparison to WT mice in response to BLM. On the other hand, we observed no effect of BLM instillation on the expression of the anti-inflammatory cytokine IL-10 upon airway BLM in lung of WT or ST2 −/− mice ( Figure 3H). In addition, the expression of the IL-13 cytokine known to be produced by innate lymphoid cell type 2 (ILC2) and Th2 cells, but not by M2 macrophages, was not significantly changed by BLM and between WT and ST2 −/− mice ( Figure 3I). Finally, the expression of the T helper1 (Th1)-like cytokine IFN-γ in lung was reduce after BLM instillation in both WT and ST2 −/− mice ( Figure 3J). These results indicates that the IL-33/ST2 pathway leads to a shift from M1 to M2 macrophage differentiation and suggest that M2 macrophages are the main immune cells involved in inflammation resolution and promoting tissue repair. reduced Pulmonary Fibrosis in sT2-Deficient Mice 11 Days after BlM exposure Interestingly, we observed reduced body weight loss in ST2 −/− mice in comparison to WT mice (data not shown). Moreover, the | 3,069,798 | 49396821 | 0 | 16 |
level of the remodeling factor TIMP-1, a marker of evolution toward fibrosis, was significantly lower in lung of ST2 −/− mice in comparison to WT mice at day 11 ( Figure 4A). In addition, total collagen content in the lungs ( Figure 4B) was lower in ST2 −/− mice. Finally, histological analysis revealed a significant reduction of lung fibrosis in the absence of ST2 as shown by representative lung sections and semi-quantitative severity scores (Figures 4C,D) and inflammation score ( Figure 4E). Moreover, inflammation in the airways was associated with enhanced extracellular fluid retention expressed as edema upon BLM instillation which was increased in ST2-deficient mice ( Figure 4F). These results suggest that ST2 deficiency impairs the development of lung fibrosis, but enhances edema in response to BLM. augmented extracellular Fluid retention in the airways assessed by Mri Non-invasive in vivo imaging by MRI was used to evaluate the effects of BLM instillation on fluid retention and inflammation in WT and ST2 −/− mice. MRI recording were performed at two anatomical levels, corresponding to slices 1 and 2 as shown in Figure 5A which represent axial slices through the chest of allowed to assess the structure of the lung and neighboring organs ( Figure 5B). Typical MRI images of healthy lungs appeared predominantly dark because of the low signal associated with air space of the lung parenchyma. Importantly, MRI baseline measurements before BLM administration (day 0) allowed the use of each animal as its own control. There was no significant difference in signal intensity in the | 3,069,799 | 49396821 | 0 | 16 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.