Altitudinal variation of the gut microbiota in wild house mice

Abstract

The maintenance of oxygen homeostasis in the gut is critical for the maintenance of a healthy gut microbiota. However, few studies have explored how the concentration of atmospheric oxygen affects the gut microbiota in natural populations. High-altitude environments provide an opportunity to study the potential effects of atmospheric oxygen on the composition and function of the gut microbiota. Here, we characterized the caecal microbial communities of wild house mice (Mus musculus domesticus) in two independent altitudinal transects, one in Ecuador and one in Bolivia, from sea level to nearly 4,000 m. First, we found that differences in altitude were associated with differences in the gut microbial community after controlling for the effects of body mass, diet, reproductive status and population of origin. Second, obligate anaerobes tended to show a positive correlation with altitude, while all other microbes tended to show a negative correlation with altitude. These patterns were seen independently in both transects, consistent with the expected effects of atmospheric oxygen on gut microbes. Prevotella was the most-enriched genus at high elevations in both transects, consistent with observations in high-altitude populations of pikas, ruminants and humans, and also consistent with observations of laboratory mice exposed to hypoxic conditions. Lastly, the renin-angiotensin system, a recently proposed microbiota-mediated pathway of blood pressure regulation, was the top predicted metagenomic pathway enriched in high altitudes in both transects. These results suggest that high-altitude environments affect the composition and function of the gut microbiota in wild mammals.

Figure 1.

Effects of altitude on the gut microbiota of wild mice. Wild house mice were collected across two altitudinal gradients from Ecuador transect (n=49) (A) and Bolivia-Brazil transect (n=43) (B). Each bar represents an individual collected from a given elevation and color coded by populations. PCoA plot of Bray-Curtis dissimilarity colored by populations (ADONIS r² = 0.18, p<0.0001) (C) and altitude (ADONIS r² = 0.04, p<0.0001) (D). The colors of Fig.1C correspond to Fig.1A and 1B. Darker colors in Fig.1D correspond to higher altitudes. A correlation between altitude and alpha-diversity measured by phylogenetic diversity (PD) (E). Correlation between altitude and PD of three dominant phyla; PD of Firmicutes (F), PD of Bacteroidetes (G), and PD of Proteobacteria (H). Only PD of Bacteroidetes significantly correlated with altitude among the dominant phyla (rho = 0.41, p < 0.0001). Red and blue colors correspond to individuals from Ecuador transect and Bolivia-Brazil transect, respectively.

Figure 2.

Correlations between altitude and bacterial genera. Bacterial genera were included in the list when (1) the correlation between altitude and relative abundance of genera was in the same direction in both transects based on Spearman’s rho correlation, (2) average relative abundance >0.1% across all samples, and (3) at least named bacterial family was assigned to search for oxygen requirements. The brackets [ ] indicate recommended taxonomy. Red color indicates Ecuador transect and blue color indicate Bolivia-Brazil transect. Filled patterns show obligate anaerobes and open pattern show non-anaerobes (i.e. facultative anaerobes, aerotolerant anaerobes, microaerophiles, and obligate aerobes). Oxygen requirements were assigned to each genus based on Bergey’s Manual of Systematics of Archaea and Bacteria and recent literature. Fisher’s combined p-values of the Spearman’s rho raw p-values are indicated: * p < 0.1, ** p < 0.05, *** p < 0.0001. After Bonferroni correction (alpha = 0.05/17 = 0.003), only the two Prevotella genera were significant.

Figure 3.

Convergent associations between relative abundances of Prevotella and high-altitude environments in different species of mammals. Significant positive correlations were observed between altitude and the relative abundance of Prevotella in wild house mice from Ecuador transect (A) and Bolivia-Brazil transect (B). A similar correlation between altitude and relative abundance of Prevotella was found in Pika (n=102) (Figure generated from data in Li et al. 2016) (C). The relative abundance of Prevotella was higher in Yaks compared to cattle collected from the same farm (elev. 3000m) (D) and in Tibetan sheep (elev. 3000m) compared to sheep (elev. 2200m) (E) (Figure generated from data in Zhang et al. 2016). The relative abundance of Prevotella was higher in Tibetans (3600–4500m) living in high altitudes compared to Han (500–3600m) living in low altitudes (Figure generated from data in Li et al. 2016) (F). In controlled lab settings, intermittent hypoxic exposure in laboratory mice resulted in higher relative abundance of Prevotella compared to controls (Figure generated from data in Moreno-Indias et al. 2015) (G).

Figure 4.

A proposed mechanism of the microbiota-mediated regulation of blood pressure in response to atmospheric oxygen at high altitudes. “Anaerobes” refer to obligate anaerobes and “Aerobes” refer to all other oxygen requirements (i.e. facultative anaerobes, aerotolerant anaerobes, microaerophiles, and obligate aerobes).