We conducted a horizontal comparison of the luminal and mucosal gut microbiomes and metabolomes of oriental rat snakes; we aimed to test our hypothesis that the luminal microbiome was more closely associated with metabolism, whereas the mucosal microbiome was mainly involved in immune function. Although the results demonstrated significant differences in microbial diversity, composition, functions, and associated metabolites, core microbe analysis revealed some instances of co-occurrence in the luminal and mucosal microbiomes.
The mucosal microbiome exhibits greater stability
Within-group stability is a critical factor that influences the analysis of microbiomes in animal gastrointestinal tracts, and lower within-group dissimilarity can help to support robust conclusions. This study revealed that the mucosal microbiome in snakes was significantly more stable than the luminal microbiome, specifically on the basis of within-group β-diversity distance analysis (Fig. 4a) and the related PCoA (Fig. 4c) and NMDS (Fig. 4d) plots. These findings were indirectly supported by the greater number of shared ASVs in the mucosal microbiome (Fig. 2a & b), cumulative bar chart analyses of phyla (Fig. 3a) and genera (Fig. 3b), and PCA plots regarding KEGG function (Fig. 5b). Studies of human microbiomes have shown that the mucosal microbiome is more stable than the fecal microbiome [33]; in swine, Firmicutes abundance was more stable in the mucosal microbiome than in the luminal microbiome [17]. Nevertheless, other studies involving chicken [15], cattle [34], and macaques [14, 35] have not revealed differences in terms of within-group stability between luminal and mucosal/epithelial microbiomes. These results suggest that the mucosal niche supports a more stable microbial community, particularly in snakes, where the luminal microbiome is likely to be strongly affected by the state of absorption and recent diet; their dietary pattern involves intermittent fasting, which can cause dynamic changes in luminal contents [31, 36]. Mucosal microbes colonize the surface of the intestinal tract, where numerous intestinal villi and crypts can shelter microbes from dietary disturbances and other factors.
The mucosal microbiome exhibits greater diversity
The α-diversity of a microbial community is the most intuitive parameter that can indicate its richness, evenness, and abundance. Many ecological indices have been used to explore α-diversity, including Chao1, ACE, Shannon, Simpson, and Faith_pd. In the present study, all four indices (Chao1, ACE, Shannon, and Faith_pd), as well as the number of ASVs, showed significantly greater values in the mucosal microbiome than in the luminal microbiome (Fig. 1a & b, Table S1). These findings strongly indicated greater microbial diversity at mucosal sites in the gut of oriental rat snakes. These results were consistent with previous findings in studies of pigs [37], donkeys [38], and cattle [39], which revealed higher Chao1, Shannon, and ACE values in mucosal scrapings than in luminal contents. However, data from studies of humans [40] and chickens [15] suggest that microbial diversity is greater in intestinal contents than in mucosal samples. A study of horses revealed no differences in α-diversity between mucosal and luminal microbiomes in the cecum and large colon [41]; the authors suspected that the small number of biological replicas prevented detection of such differences. These discrepancies may be related to differences in available oxygen, nutrients, and the microbial environment. Anaerobic bacteria are the most dominant microbes (> 90%) in animal intestinal tracts; intestinal villi and crypts provide an ideal anaerobic environment for these microbes [42]. Importantly, the greater microbial diversity at mucosal sites suggests that these microbes are metabolically and immunologically important for snakes.
Differences in the abundances of Firmicutes, Proteobacteria, Bacteroidetes, and subordinate genera between luminal and mucosal microbiomes
The gut microbiomes of snakes in this experiment were generally dominated by the phyla Firmicutes, Proteobacteria, and Bacteroidetes (Fig. 3a), consistent with the results of previous studies regarding snakes [25, 29, 30]. Moreover, the phyla Actinobacteriota and Fusobacteriota exhibited greater abundance in some samples in the present study (Fig. 3a), and occasionally in other studies [25, 27]. Because snakes are carnivorous, their gut microbiomes included phyla present in other carnivores (e.g., coyotes [43], foxes [44], raccoon dogs [45], and tigers [46]), which highlights the extensive effects of diet on the gut microbiome in animals. Among these five phyla, Bacteroidota, Proteobacteria, and Fusobacteriota showed greater abundance at luminal sites, whereas Firmicutes and Actinobacteriota tended to show greater abundance at mucosal sites (Table 1). Analysis of the human microbiome revealed greater abundances of Firmicutes and Bacteroidota in stool, whereas the abundance of Proteobacteria was greater in mucosa [13]. In macaques, Bacteroidetes was dominant in both luminal and mucosal microbiomes, whereas Firmicutes was more abundant in the luminal microbiome and Proteobacteria was more abundant in the mucosal microbiome [14, 35]. In cattle, Firmicutes and Proteobacteria were more abundant in epimural biopsy samples, whereas Bacteroidetes and Proteobacteria were more abundant in luminal samples [34, 39]. To our knowledge, no study has explored the mucosal microbiome in snakes, but there is evidence that dominant phyla in the snake microbiome are influenced by many factors, including host species [25, 27], venom secretion [29], gastrointestinal biogeography [30], and foraging pattern [31].
At the genus level, the most dominant bacteria were Bacteroides, Agathobacter, Escherichia-Shigella, Blautia, Prevotella, Roseburia, Romboutsia, Collinsella, and Faecalibacterium (Fig. 3b). Furthermore, Bacteroides, Escherichia-Shigella, and Romboutsia (although not significant) showedtended to show greater abundance in the luminal microbiome, whereas the Agathobacter, Blautia, Prevotella, Roseburia, Collinsella, and Faecalibacterium showed greater abundance in the mucosal microbiome (Table 2). Bacteroides and Prevotella belong to the phylum Bacteroidota; Escherichia-Shigella belongs to the phylum Proteobacteria; Faecalibacterium, Agathobacter, Blautia, Roseburia, and Romboutsia belong to the phylum Firmicutes; and Collinsella belongs to the phylum Actinobacteriota. Thus, the findings at the genus level are consistent with results at the phylum level. The physicochemical conditions and substrate availability at mucosal and luminal sites create diverse microenvironments that support distinct microbial populations [47]. Previous studies showed that the gut microbiome of snakes was dominated by diverse genera, whereas Bacteroides was the only taxon shared among all samples; thus, Bacteroides is a core member of the microbiome in snakes [25, 27, 29, 30]. Bacteroides is the important keystone bacteria in animal gut; it primarily participates in the digestion and utilization of complex carbohydrates, indigestible polysaccharides, fat, and proteins [48]. This is consistent with our finding that Bacteroides was dominant in the snake gut microbiome; its abundance was greater at luminal sites, where the main stage of digestion occurs. Notably, the opportunistic pathogen genus Escherichia-Shigella exhibited high abundance in all 10 luminal samples, but it was rarely present in mucosal samples (Fig. 3b). Conversely, the pathogenic bacteria Escherichia coli and Enterococcus faecalis exhibited greater abundances at mucosal sites than at luminal sites [15]. These findings are consistent with the gut lumen environment serving as a location for dietary intake to interact with environmental microbes; thus, luminal microbes are more likely to include pathogenic bacteria. Our results have significant implications for snake health, considering that members of the mucosal microbiome are important for mucosal barrier function.
Differences in microbial metabolites and KEGG functions between luminal and mucosal microbiomes
Metabolites are key mediators between microbiome and host, and metabolite pathways are directly involved in host function. First, consistent with the above microbial analysis, sample clustering based on metabolite abundance revealed distinct groups (Fig. 6c–f). The metabolites were mainly lipids and lipid-like molecules, organic acids and derivatives, and carboxylic acids and derivatives (Fig. S2a & b). The RDA analysis (Fig. 8a) revealed that the top 10 metabolites with significant differences were mainly produced by Agathobacter, Blautia, Prevotella, Roseburia, Collinsella, and Faecalibacterium at mucosal sites. Correlation heatmap analysis (Fig. 8b) furtherly revealed that the top 30 metabolites with significant differences were positively correlated with the above six genera, which dominated the mucosal microbiome, rather than the luminal microbiome. SCFAss are the most important microbial metabolites, but their abundance was low in this study, presumably because of the absence of SCFA-producing microbes (e.g., UCG-005, Treponema, and Rikenellaceae_RC9_gut_group) [49]. Among the main upregulated metabolites (Fig. 6a & b), most were cholinergic substances (glycerophosphocholine, Sn-glycerol-3-phosphoethanolamine, phosphocholine, and cytidine 5’-diphosphocholine), followed by nucleic acid substances (cytidine, uridine 5’-monophosphate, and inosine 5’-monophosphate); these substances comprise important substrates in the synthesis of nucleic acids and flow of information between cells [50, 51]. Among the top 30 metabolites with significant differences, nine were cholinergic substances, including phosphocholine, cytidine 5’-diphosphocholine, glycerophosphocholine, Sn-glycerol-3-phosphoethanolamine, securinine, and acetylcarnitine. Seven nucleic acid substances were among the top 30 metabolites with significant differences, including adenosine 2’-monophosphate, adenosine 3’-monophosphate, uridine 5’-diphosphate, and beta-nicotinamide adenine dinucleotide.
The KEGG functions of microbial communities (Fig. 7a) and metabolites (Fig. 7b) exhibited changes associated with aminoacyl-tRNA biosynthesis, ABC transporters, purine metabolism, and pyrimidine metabolism. These greater functional abundances coupled with the above metabolites indicated that the mucosal microbiome participated in genetic information processing and cellular processes, whereas the luminal microbiome participated in metabolic regulation (Table 3). The mucosal microbiome is located at the interface with the gut epithelium; therefore, it participates in immune processes, mucosal barrier function, and gut–brain communication [52, 53]. Notably, the metabolite fenfluramine was abundantly produced by the mucosal microbiome; this metabolite is an important regulator of circulating triglyceride levels [54]. Snakes accumulate fat depots during active periods and use these stored lipids to support survival during hibernation [55]. Although snakes become obese on a seasonal basis, they remain metabolically healthy, partly because of regulatory factors such as fenfluramine.