Temporal variation selects for diet-microbe co-metabolic traits in the gut of Gorilla spp

Abstract

Although the critical role that our gastrointestinal microbes play in host physiology is now well established, we know little about the factors that influenced the evolution of primate gut microbiomes. To further understand current gut microbiome configurations and diet-microbe co-metabolic fingerprints in primates, from an evolutionary perspective, we characterized fecal bacterial communities and metabolomic profiles in 228 fecal samples of lowland and mountain gorillas (G. g. gorilla and G. b. beringei, respectively), our closest evolutionary relatives after chimpanzees. Our results demonstrate that the gut microbiomes and metabolomes of these two species exhibit significantly different patterns. This is supported by increased abundance of metabolites and bacterial taxa associated with fiber metabolism in mountain gorillas, and enrichment of markers associated with simple sugar, lipid and sterol turnover in the lowland species. However, longitudinal sampling shows that both species' microbiomes and metabolomes converge when hosts face similar dietary constraints, associated with low fruit availability in their habitats. By showing differences and convergence of diet-microbe co-metabolic fingerprints in two geographically isolated primate species, under specific dietary stimuli, we suggest that dietary constraints triggered during their adaptive radiation were potential factors behind the species-specific microbiome patterns observed in primates today.

Figure 1

Gut microbiome composition in lowland and mountain gorillas. (a) Principal coordinates analysis (PCoA) ordination (operational taxonomic units=97% 16S rRNA sequence similarity) based on a Bray–Curtis dissimilarity matrix showing significantly different microbiome composition between the two Gorilla spp., G. g. gorilla and G. b. beringei (permutational multivariate analysis of variance, P<0.001, R²⁼0.17). (b) PCoA ordination showing significantly different microbiome composition between the two Gorilla spp., and across seasons in lowland gorillas. The Blue circle highlights samples from lowland gorillas collected during the low fruit season and a 95% confidence interval. (c) Relative abundance of major phyla. Different letters (a, b, c) denote significant differences in the abundance of taxa between the microbiomes of both gorilla species in the low fruit (LF), transition (T) and high fruit (HF) seasons (P<0.05, Wilcoxon rank sum tests).

Figure 2

Comparison of seasonal gut microbiome traits in lowland (G.g.gorilla) and mountain (G.b.beringei) gorillas. (a) Principal coordinates analysis (PCoA) ordination at genus level based on a Bray–Curtis dissimilarity matrix shows that low fruit (LF) lowland gorilla gut microbiomes are more similar to those of the mountain species. In contrast, lowland gorilla microbiomes in the high fruit (HF) season show the most dissimilarity to those of mountain gorillas. Microbiomes of lowland gorillas in the transition season (T) lie at an intermediate level of similarity between those of mountain gorillas and lowland gorillas in the HF season. Circles represent 95% confidence intervals for mountain and lowland gorilla samples collected during the LF season. Bar plots in b show mean Bray–Curtis dissimilarity between groups and asterisks denote significant differences (***P<0.001, Kruskal-Wallis tests adjusted for multiple comparisons). (c) Taxa that showed either decreasing or increasing abundance between mountain gorillas (G.b) and lowland gorilla microbiomes during LF, T and HF seasons. Different letters (a, b, c, d) denote significant differences according to Kruskal–Wallis tests adjusted for multiple comparisons. Actual abundances of each of these taxa can be seen in Supplementary Table 1. Boxplot showing differential abundance of Clostridiaceae represents the sum of the relative abundances obtained for unclassified Clostridiaceae, Anaerobacter, Clostridium sensu stricto and Sarcina.

Figure 3

Gut metabolomic profiles in lowland (G.g.gorilla) and mountain (G.b.beringei) gorillas. (a) Three-dimensional partial-least squares discriminant analysis (PLS-DA) score plot showing separation of the gut metabolomes of mountain and lowland gorillas in samples collected during low fruit (LF) and high fruit (HF) seasons (permutation test supported model variation P<0.001). The amount of the variation in the metabolite data set explained by the three component model was R²X=95.7%. Predictability of the model and statistical validity was Q²=93.3%. (b) Variables (metabolites) with influence on the PLS-DA projections (VIP) along component 1. The heat map shows the mean normalized abundance of metabolites within groups with VIP values >1.3 (false discovery rate: q<0.05, Kruskal–Wallis multiple comparisons) in mountain gorillas (G.b), LF and HF lowland gorilla samples. Highest VIP values are shown in decreasing order from C17:0 ethyl on top (VIP=2.15) to cholesterol (VIP=1.31) at the bottom. VIP values and normalized metabolite concentrations can be seen in Supplementary Table 2.

Figure 4

Sub-network view of relationships between gut microbiome composition and gut metabolomic profiles in lowland (a) and mountain (b) gorillas. Green and yellow nodes represent bacteria and metabolites respectively. Node size represents the number of connections of a given taxa or metabolite within the network. Edges represent Spearman correlation coefficients (Rho>0.5). Edge thickness shows the strength of the correlation.