Simulated seasonal diets alter yak rumen microbiota structure and metabolic function

Abstract

Yak is the only ruminant on the Qinghai-Tibetan Plateau that grazes year-round. Although previous research has shown that yak rumen microbiota fluctuates in robust patterns with seasonal foraging, it remains unclear whether these dynamic shifts are driven by changes in environment or nutrient availability. The study examines the response of yak rumen microbiota (bacteria, fungi, and archaea) to simulated seasonal diets, excluding the contribution of environmental factors. A total of 18 adult male yaks were randomly divided into three groups, including a nutrition stress group (NSG, simulating winter pasture), a grazing simulation group (GSG, simulating warm season pasture), and a supplementation group (SG, simulating winter pasture supplemented with feed concentrates). Volatile fatty acids (VFAs) profiling showed that ruminal acetate, propionate and total VFA contents were significantly higher (p < 0.05) in GSG rumen. Metagenomic analysis showed that Bacteroidetes (53.9%) and Firmicutes (37.1%) were the dominant bacterial phyla in yak rumen across dietary treatments. In GSG samples, Actinobacteriota, Succinivibrionaceae_UCG-002, and Ruminococcus albus were the most abundant, while Bacteroides was significantly more abundant in NSG samples (p < 0.05) than that in GSG. The known fiber-degrading fungus, Neocallimastix, was significantly more abundant in NSG and SG samples, while Cyllamyces were more prevalent in NSG rumen than in the SG rumen. These findings imply that a diverse consortium of microbes may cooperate in response to fluctuating nutrient availability, with depletion of known rumen taxa under nutrient deficiency. Archaeal community composition showed less variation between treatments than bacterial and fungal communities. Additionally, Orpinomyces was significantly positively correlated with acetate levels, both of which are prevalent in GSG compared with other groups. Correlation analysis between microbial taxa and VFA production or between specific rumen microbes further illustrated a collective response to nutrient availability by gut microbiota and rumen VFA metabolism. PICRUSt and FUNGuild functional prediction analysis indicated fluctuation response of the function of microbial communities among groups. These results provide a framework for understanding how microbiota participate in seasonal adaptations to forage availability in high-altitude ruminants, and form a basis for future development of probiotic supplements to enhance nutrient utilization in livestock.

Keywords: metagenomics; nutrient simulation; rumen microbiota; season; volatile fatty acid; yak.

Figure 1

Differences in Yak rumen microbial diversity and richness between the GSG, NSG, and SG. Diversity was estimated by Shannon index. Richness estimated by the Chao1 value. (A) Chao 1 index. (B) Shannon index. Asterisks indicate significant difference between the Concentrate Group and the Forage Group (*p ≤ 0.05; **p ≤ 0.01).

Figure 2

Principal coordinate analysis (PCoA) of rumen microbial communities. (A) Bacteria. (B) Fungi. (C) Archaea.

Figure 3

Linear discriminant analysis (LDA) of rumen bacteria. (LDA cutoff of +/−2.0).

Figure 4

Classification of the rumen microbial composition at the genus level across the different nutrient simulations in different grazing patterns. (A) Bacteria. (B) Fungi. (C) Archaea. (D) Bacterial genera with significant changes under various nutrient simulations. (E) Fungal genera with significant changes under various nutrient simulations. (F) Archaeal genera with significant changes under various nutrient simulations. (G) Major fiber-degrading bacteria with significant changes under various nutrient simulations. Asterisks indicate significant difference between the three groups (*p ≤ 0.05; **p ≤ 0.01).

Figure 5

Correlations between the higher abundant genera of rumen microbial and VFAs. Each row in the graph represents a genus, each column represents a metabolite, and each lattice represents a Spearman’s correlation coefficient between a component and a metabolite. Red represents a positive correlation, while blue represents a negative correlation. *p < 0.05, **p < 0.01.

Figure 6

Mutual-correlation among the higher abundant genera of rumen microbial. Red color represents a positive correlation, while blue represents a negative correlation. *p < 0.05, **p < 0.01.