Unraveling the distinctive gut microbiome of khulans (Equus hemionus hemionus) in comparison to their drinking water and closely related equids

Abstract

The microbial composition of host-associated microbiomes is influenced by co-evolutionary interactions, host genetics, domestication, and the environment. This study investigates the contribution of environmental microbiota from freshwater bodies to the gastrointestinal microbiomes of wild khulans (Equus hemionus hemionus, n = 21) and compares them with those of captive khulans (n = 12) and other equids-Przewalski's horse (n = 82) and domestic horse (n = 26). Using PacBio technology and the LotuS pipeline for 16S rRNA gene sequencing, we analyze microbial diversity and conduct differential abundance, alpha, and beta diversity analyses. Results indicate limited microbial sharing between wild khulans and their waterhole environments, suggesting minimal environmental influence on their gut microbiomes and low levels of water contamination by khulans. Wild khulans exhibit greater microbial diversity and richness compared to captive ones, likely due to adaptations to the harsh nutritional conditions of the Gobi desert. Conversely, captive khulans show reduced microbial diversity, potentially affected by dietary changes during captivity. These findings highlight the significant impact of environment and lifestyle on the gut microbiomes of equids.

Keywords: Equus hemionus hemionus; 16S rRNA full length gene sequencing; Freshwater microbiota; Gobi Desert; Microbiome; PacBio.

Fig. 1

Alpha diversity metrics of the microbiome of wild khulan, captive khulan and waterholes in the Gobi. Wild khulan microbiomes are significantly more diverse and rich than waterholes (A,C) and captive khulans (B,D). Wild khulans shared 17 OTUs with environmental samples from waterholes and 726 OTUs with captive khulan.

Fig. 2

Beta diversity analysis shows distinct microbial communities in wild khulans, waterholes, and captive khulans. (A) Principal Coordinates Analysis (PCoA) plots comparing the microbial beta diversity of wild and captive khulans, water and sediments from waterholes. (B) Pairwise comparison of intersample Bray-Curtis (BC) dissimilarity distances shows no difference between wild and captive khulan to water or sediment microbial communities. Every dot represents the distance between a pair of samples. (C) Mean and the 95% confidence interval of a post-hoc Tukey’s HSD test performed on BC dissimilarity distances shows highly significant differences between wild khulan microbiomes, water and sediment microbiomes. (D) The overall comparison of multivariate homogeneity of sample type dispersions for BC dissimilarity distances between samples relative to the centroid shows a higher significant difference between wild khulan microbiomes to waterhole-derived samples. (E) PCoA from waterhole-derived samples confirms that 14% of the variation in PCoA 1 and the observed clustering in microbial composition varies significantly based on the sample origin. (F) Clustering of khulan microbiomes in the PCoA significantly depend on living condition, either wild or captive. Significance codes: ‘****’ 0.0001, ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05.

Fig. 3

Detection of differentially abundant bacteria between sample groups. (A) Among shared OTUs between wild and captive khulans, OTUs belonging to the families Lachnospiraceae, Rikenellaceae and class Kiritimatiellae are significantly more abundant in wild khulan microbiomes, while just Comamonas kerstersii is abundant in captive populations. (B) Water and sediment microbiomes show characteristic bacteria each, while water has significantly more abundant Bacteriovorax sp. and other Proteobacteria OTUs, members of the phyla Firmicutes from the family Clostridiaceae, are more abundant in sediment.

Fig. 4

Alpha diversity metrics between microbiomes from wild, captive khulans and wild, domestic horses. (A) Captive khulan microbiomes are significantly less diverse than those from wild populations of khulans and wild and domestic horses from Mongolia. (B) Venn diagram showing the overlap in OTUs among wild khulans, wild horses, captive khulans, and domestic horses. Wild khulans and wild horses show the highest number of unique OTUs compared to either captive khulans or domestic horses. Wild khulans have 1,651 unique OTUs and wild horses have 2,236, while both domestic horses and captive khulans have fewer than 700 unique OTUs each. A significant overlap of 335 OTUs might indicate a core set of bacteria shared among Mongolian equids.

Fig. 5

Beta diversity analysis shows distinct microbial communities from wild and captive khulans compared to domestic and wild horses from a near region. (A) Principal Coordinates Analysis (PCoA) plots comparing the microbial beta diversity of wild and captive khulans, domestic and wild horses from Metcalf et al., 2017. Both sample populations clustered separately, confirming distinctive compositions between studies. Clustering of khulan and horse microbiomes in the PCoA significantly depends on sample type (condition), either wild, captive kulans or wild or domestic horses. Significance codes: ‘****’ 0.0001, ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 (B) Pairwise comparison of intersample Bray-Curtis (BC) dissimilarity distances shows strong differences within each khulan or horse’s microbial communities. Every dot represents the distance between a pair of samples. Inter host type variability reflects larger microbial dissimilarity between host types than within host types. (C) Mean and the 95% confidence interval of a post-hoc Tukey’s HSD test performed on BC dissimilarity distances shows highly significant differences between wild khulans (WK) and captive khulan (CK) microbiomes to domestic (DH, E. caballus) and Przewalski’s horses (PH, E. ferus przewalskii) microbiomes.