4 | DISCUSSION
In this study, we constructed a gene catalogue of the gut microbiome of wild SNMs for the first time by processing 143 fecal individual samples, which contained 18,169,322 non-redundant genes. The RGC also represents the first gene-set of the gut microbiome from wild NHP populations and provides a comprehensive resource for further investigations of the NHP gut microbiome. Compared with previous studies of the SNMs gut microbiome (V. L. Hale et al., 2018; Trevelline & Moeller, 2022; Xu et al., 2015), the RGC comprehensively characterizes the gut microbiome of SNMs and probably contains most of the gut microbial genes prevalent in wild SNM populations. Metagenomic data from 10 wild Sichuan SNMs researched in a previous study showed a mapping rate of more than 75% in the RGC, but there was relatively poor representation in captive SNMs, with an average mapping rate was 49.83% (Table S8). Previous studies have suggested that the captivity environment has a significant effect on the gut microbiome of SNMs (Hui Zhu, 2018; Mingpu Qi, 2017). Similar results have also been reported in other NHPs (Clayton et al., 2016; Emmanouil Angelakis, 2016; V. L. Hale et al., 2018; Tayte P Campbell, 2020). This proves that the RGC is reliable in describing the gut microbiome of SNMs. Therefore, a comprehensive analysis of the gut microbiome based on the RGC is of practical significance for studying their dietary adaptation, which could further guide the development of better conservation strategies for these endangered species.
We used this catalogue to study the characteristics of the SNMs gut microbiome. In our study, both taxonomic and functional results suggested that the gut microbiome of SNMs was related to structural carbohydrate degradation. Notably, our results were identical to those of previous studies at the phylum level, while the differences were large at the genus level (Wang et al., 2021; Yao et al., 2021). At the phylum level, Firmicutes and Bacteroides were the main groups in the RGC (Figure 2b). This is consistent with the results of previous studies on Sichuan SNMs (Wang et al., 2021; Yao et al., 2021), Yunnan SNMs (Xia et al., 2022) and Guizhou SNMs (V. L. Hale, Tan, C.L., Niu, K., Yang, Y., Zhang, Q., Knight, R., 2019). At the genus level, Clostridium ,Prevotella , Ruminococcus and Bacteroides were the groups with the highest relative abundance in the RGC (Figure 2c), which was different from the results of previous single-species studies (V. L. Hale, Tan, C.L., Niu, K., Yang, Y., Zhang, Q., Knight, R., 2019; Wang et al., 2021). This was likely because previous studies on the gut microbiome of SNMs were mainly based on 16S rRNA analysis and small sample sizes (Xia et al., 2022; Yao et al., 2021), preventing the results from describing the general characteristics of the SNMs gut microbiome. The consistency of the main phyla reflected that the gut microbiome of SNMs was related to their dietary adaptation. Many Firmicutes bacteria, including Clostridium , are able to utilize xylose, xylan and xyloglucan, which are the major hemicellulose components of plant cell walls (Canfora, Meex, Venema, & Blaak, 2019). These results once again verified the reliability and comprehensiveness of the RGC and preliminarily demonstrated that the gut microbiome composition of SNMs plays an important role in their dietary adaptation.
Alpha and beta diversity analyses also revealed the important role of the gut microbiome in the dietary adaptation of SNMs (Figure 3b, 3c and 3d). The comparative analysis of 12 mammals including SNMs, ruminants and monogastric animals, indicated that the gut microbiome of SNMs was clustered more closely to that of ruminants (Figure 3d), which was similar to previous findings (Zhou et al., 2014). As the morphology of mammalian GIT underwent convergent evolution to adapt to herbivory, their microbiome might have developed similar compositional configurations in unrelated hosts with similar gut structures (R. E. Ley et al., 2008). We further found evidence of coevolution between SNMs and ruminants during dietary adaptation. In SNMs, Ruminococcus ,Treponema and Clostridium were significantly enriched at the genus level, and GH78, GH13 and GH109 were significantly enriched in carbohydrate enzymes (GHs). The genera that were significantly enriched in ruminants were Fibrobacter , Butyrivibrio andPrevotella , and the GHs that were significantly enriched were GH25 and GH5 (Figure 4a and 4b; Figure S5). Although the genera and GHs that were significantly enriched in SNMs and ruminants were different, these genera and GHs were mainly associated with the digestion of structural carbohydrates. For example, Ruminococcus ,Clostridium and Butyrivibrio belong to Firmicutes and are related to the degradation of cellulose (Canfora et al., 2019), and GH5 and GH78 both have cellulase activity (Christiane Liers, 2021). However, the genera and GHs that were significantly enriched in monogastric animals were related to the degradation of fat and protein and oligosaccharides, such as Bacteroides (Naofumi Yoshida, 2021) (Figure 4a and 4b). This indicated that the gut microbes of SNMs and ruminants evolved similar digestive strategies in the process of adapting to the plant diet and further suggested that the gut microbiome is an important part of host dietary adaptation.
SNMs and ruminants also display similar physiological adaptation strategies to plant-based diets, and they are both foregut fermentative animals (Liu et al., 2022). The rumen of ruminants and the enlarged saccular stomach of SNMs provide a large space for the bacteria to ferment cellulose and other substances (D., 1998; Karasov & Douglas, 2013; P., 1988), but the fermentation efficiency of SNMs may be lower than that of ruminants, because the sacculated stomach of SNMs does not show as strong functional division as ruminants, and the differentiation is intermediate between those of monogastric animals and ruminants (D. J. Chivers, Hladik, C.M, 1980). Previous studies have reported that the morphology of the GIT has an important impact on the composition and function of bacterial communities (R. E. Ley et al., 2008). In our study, we found that the relative abundance of 1859 genera (over 50%) and 37 GHs in SNMs was higher than that in monogastric animals but lower than that in ruminants (Table S14). Most of these GHs were associated with degradation of structural polysaccharides, such as GH10, GH12, GH9, GH14 and so on (Table S15). We also found that as the host GIT structure adapted to herbivory, the diversity and abundance of GHs related to the fermentation of structural carbohydrates also increased (Table S14), suggesting that the dual effects of GIT morphology and the gut microbiome promoted the dietary adaptive evolution in foregut fermentative animals, which provides a new perspective for exploring the evolutionary path of the gut microbiome in these animals.
In addition, we defined a set of core gut bacterial genera, includingClostridium , Ruminococcus , Prevotella ,Bacteroides , Eubacterium , Parabacteroides ,Roseburia , Alistipes and Oscillibacter , based on the gut microbiome generated by shotgun sequencing of 12 mammalian species (Figure S4). However, the relative abundance of these genera varied profoundly between each species (Figure S3), which might result from the influence of genetic, dietary and physiological characteristics of the host (R. E. Ley et al., 2008). Our core genera reflected, to some extent, the basic composition of the mammalian gut microbiome. Previous studies reported that most of these genera were important keystone bacteria of the gut microbiome, which are closely related to host food degradation, nutrient absorption, health, and intestinal homeostasis (Ruth E Ley, 2016; Petia Kovatcheva-Datchary, 2015). Therefore, it is highly necessary to focus on these core genera in future analyses of the mammalian gut microbiome.