Seasonal dietary shifts drive gut microbiome plasticity and metabolic adaptation in wild yaks on the Qinghai-Xizang Plateau Authors: Jiahui Hao, Hongjin Liu, Tongqing Guo, Qian Zhang, Xungang Wang, Lele An, Shixiao Xu Published: 2025 ABSTRACT AbstractThe gut microbiota in different diets helps hosts to obtain sufficient nutrients from food, which is important for wild yaks in the Qinghai-Xizang Plateau to adapt to different seasons. The relationship between the host diet and the gut microbiota in different seasons is important for exploring the adaptation of these wild yaks to their environments. This study used wild yaks as a model organism. Using high-throughput sequencing and liquid chromatography-mass spectrometry, we investigated their seasonal diet, gut microbiota composition and function, and short-chain fatty acid profiles. The inter-season differences in them were compared, and relationships among these differences were explored. Wild yaks in summer had a diet higher in Polygonaceae and lower in Rosaceae compared to their diets in autumn and winter. The broadest dietary niche width (8.8449) was detected in autumn, and the lowest diet niche overlap (0.3751) was found between summer and winter. Co-occurrence network analyses revealed that microbial interactions were more complex in autumn, likely due to transitional dietary adjustments, whereas summer and winter exhibited simpler but more robust interactions. The abundance of Firmicute increased in winter, suggesting enhanced energy extraction from low-quality forage. Specific taxa, such as Alistipes_A in autumn and Romboutsia in winter, were linked to key metabolic pathways, including carbohydrate degradation and short-chain fatty acid production. The study highlights the critical role of gut microbiota plasticity in facilitating wild yaks’ adaptation to the extreme and variable conditions and provides the basis for explaining the harsh environment adaptation of wild herbivores in the Qinghai-Xizang Plateau.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12866-025-04530-2.Keywords: Dietary, Gut microbiota, Wild yak, Adaptability, The Qinghai-Xizang Plateau Background Formed by ongoing tectonic uplift, the Qingha-Xizang Plateau (QTP) exceeds a mean elevation of 4,000 m and is widely regarded as Earth’s “Third Pole” [1]. Its endemic fauna must grapple with a harsh climate and profound seasonal food scarcity, particularly in the long winter months [2, 3]. The wild yak (Bos mutus) exemplifies successful high-altitude adaptation. Recent work on domestic yaks (Bos grunniens) further demonstrates that seasonal dietary shifts drive enterotype transitions associated with enhanced nitrogen and energy efficiency under nutrient restriction [4, 5]. The diet of domestic animals can be controlled, making it convenient to study the feeding habits, gut microbiota, and their interaction in controlled feed lots for domestic animals. But in grazing conditions, it’s difficult to determine the diet of the herbivores, especially the wild ones. While the feeding of wild animals is uncontrollable, wild yaks are known to graze year-round. Previous investigations into wildlife dietary ecology predominantly employed conventional methodologies, including direct field observation and microscopic examination of fecal plant residues, albeit with inherent methodological constraints [6]. In recent years, metabarcoding has enabled high-resolution taxonomic characterization of animal diets [7, 8]. The gut microbiome affects host metabolism, nutrient balance, immune function, and infection responses, and plays a critical role in regulating host health and disease susceptibility [9–11]. The gut microbiota constitutes a complex ecosystem through microbial interactions. Studies demonstrate that its composition is modulated by host genetics, dietary patterns, and physiological variables such as age and sex [12–14]. The gastrointestinal tract of animals must maintain homeostasis to sustain normal growth and metabolism, and intraspecific studies indicate that diet is the primary determinant shaping the gut microbiota. Gut microbiota play a pivotal role in enabling high-energy metabolism and facilitating the extraction of sufficient energy from limited plant resources for plateau-native ungulates, thereby ensuring their adaptation to the environment [15]. Studies of domestic yaks have demonstrated an enhanced adaptability to the plateau environment, achieved through improved rumen microbial fiber degradation and altered fermentation patterns, which facilitate better utilization of autumn and winter dry hay [16]. For herbivores, life-history strategies are determined by dietary choices and foraging behaviour, which respond to seasonal changes in the availability, quality, and ingestibility of forage [3]. Seasonal dietary variation is a primary determinant of gut-microbiome composition [5, 17, 18]. For high-altitude specialists, nutrient acquisition under the QTP’s harsh conditions is a major challenge [19, 20]. Wild yaks are nomadic and primarily forage throughout the day. Although their feeding strategies exhibit seasonal variation, the drivers of these dietary shifts remain unclear. Gut microbiota play a key role in digesting plant material via fermentation, producing metabolites such as short-chain fatty acids (SCFAs) [21]. Understanding microbial responses to dietary shifts is essential for preserving intestinal homeostasis. Studies of diet and gut microbiome differences between seasons in the wild nowadays highlight the role played by specific microorganisms in the corresponding seasons [22–25]. On the QTP, vegetation phenology follows a unimodal trajectory: green-up occurs in late April-early May, peak biomass is reached in late July-late August, senescence begins in early September, and complete dormancy prevails from mid-November to mid-April of the following year [26–29]. In conclusion, although forage grass biomass remains stable on the QTP during plant mature and dormant stages, domestic yak forage intake declines significantly, likely due to reduced vegetation accessibility or nutritional quality. Here, we explore the adaptive strategies of wild yak gut microbiota to fluctuating food availability, a key mechanism for sustaining host metabolic demands under resource limitation. We collected fecal samples from wild yaks in August and October 2023 and January 2024 to characterize seasonal shifts in their gut microbiota as pasture quality declined. By integrating microbial community data with high-resolution dietary profiles from fecal metabarcoding, investigate the seasonal dynamics of diet-microbiota interactions in wild yaks, aims to provide a new perspective on their high-altitude adaptation. Materials and methods Sample collection of content in wild yak A total of 93 fresh fecal samples were collected from wild yaks during three sampling periods (August 2023, October 2023, and January 2024) in the Three-River-Source National Park, located in western China (35°49′45″ N, 93°58′20″ E). The sample sizes for these periods were 31, 18, and 44, respectively. Because the habitats where wild yaks lived were without roads, we observed the wild yaks from a distance and collected samples immediately after they defecated. Fecal samples were collected from the central portion using a sampling stick and transferred into 2 mL cryovial. None of the wild yaks exhibited any evident signs of illness or disease. All samples were preserved in sterile containers and transported using liquid nitrogen. The collection of samples was carried out in compliance with the Wildlife Protection Law of the People’s Republic of China. The frozen tubes were stored in liquid nitrogen, brought back to the laboratory, and then sent to Nanjing Paisenor Company on dry ice for amplicon high-throughput sequencing analysis. Construction of a plant reference sequence library The plant materials in our study were selected from the Hoh Xil National Nature Reserve, the Yeniugou Valley, and its surrounding areas, where wild yaks are active. For species identification and sequencing, a local plant DNA reference sequence library was constructed. For each species plant, two types of samples were collected: a complete voucher specimen for morphological identification and a separate tissue sample for DNA sequencing. The taxonomic identity determined through morphology was then linked with the corresponding DNA sequence data to establish a local plant DNA database. A total of 113 plant samples widely distributed in the growing region of wild yak were collected, of which 84 were identified to the species level and 29 to the genus level. All the plants have been identified through morphological identification methodology. The specimens were then stored at the herbarium in the Northwest Institute of Plateau Biology, Chinese Academy of Sciences. The obtained sequences were aligned and annotated, and the results showed that the plant DNA sequence library constructed in this study covered more than 90% of the forage fed by most wild yaks (Table S1), so the alignment with other species in the NCBI database was not continued. After rarefaction according to the minimum number of sample sequences, we amplified the trnL (tRNA-Leu (UAA) gene) gene intron, a commonly used DNA barcode for plants, for subsequent analysis. DNA extraction, sequencing, and data processing To analyze the food composition of wild yak, approximately 200 mg of fecal content was used for DNA extraction with hexadecyl trimethyl ammonium bromide (CTAB) [30]. The vegetation community structures (representing the food composition) of the samples were determined by polymerase chain reaction (PCR) amplification of the trnL (UAA) gene using primers trnL (UAA)g: GGGCAATCCTGAGCCAA, and trnL (UAA)h: CCATTGAGTCTCTGCACCTATC [31]. To analyze the fecal microbial composition, genomic DNA was extracted from approximately 200 mg of fecal tract content using the CTAB method. The DNA samples were PCR-amplified using the primers 338 F (5′-CCTAYGGGRBGCASCAG-3′) and 806R (5′-GGACTACNNGGGTATCTAAT-3′), which target the V3V4 region of the bacterial 16S rRNA gene [32]. DNA samples from 93 vegetation community structures and 93 fecal microbial DNA samples were obtained for DNA extraction. The PCR amplifications were performed in a total volume of 25 μL reaction system, made up of 0.25 μL Q5 high-fidelity DNA polymerase (Catalog #M0491, NEB), 5 μL 5*Reaction Buffer, 5 μL 5* High GC Buffer, 2 μL dNTP (10 mM), 2 μL genomic DNA, and 1 μL 10 μM of each primer. The amplification reaction conditions were 5 min at 95 °C, 25 cycles of 30 s at 95 °C, 30 s at 53 °C, 45 s at 72 °C, and finally, 72 °C for 10 min. The PCR products were separated by agarose gel electrophoresis (Catalog # DYY-6C, Liuyi Biotechnology, Beijing) and purified with the Quant-iT PicoGreen dsDNA Assay Kit (Catalog #P7589, Invitrogen, Carlsbad, CA, USA). Equal amounts of PCR products from all samples were sequenced on an Illumina NovaSeq platform with NovaSeq 6000 SP Reagent Kit (500 cycles) at Shanghai Personal Biotechnology Co., Ltd (Shanghai, China). Library preparation was performed using the TruSeq Nano DNA LT Library Prep Kit (Illumina, USA). The joined sequence reads were processed with QIIME (version 2022.11) [33]. The obtained sequences were clustered at 100% similarity to generate amplicon sequence variants (ASVs) and an abundance data table. Singleton sequences were excluded. For trnL (UAA) sequencing, taxonomic assignments were performed by conducting BLAST searches against the local plant DNA reference sequence library. For 16S rRNA gene sequencing, ASV identities were assigned to the Greengenes2 database (release 13.8) [34]. The α-diversity (Chao1, Ace, Shannon, and Simpson indices) and β-diversity (Binary-Jaccard distance, Bray–Curtis distance, and UniFrac distance) indices of the composition of communities were calculated. Short-chain fatty acids determination The determination of SCFAs employs an SCFA derivatization method coupled with gas chromatography-mass spectrometry (GC–MS). Samples undergo alkalinization, homogenization, centrifugation, derivatization, and ultrasonic extraction before sample loading and detection. The GC analysis was performed on a Trace 1300 gas chromatograph (Thermo Fisher Scientific, USA). The GC was fitted with a capillary column Agilent HP-INNOWAX (30 m × 0.25 mm ID × 0.25 μm) and helium was used as the carrier gas at 1 mL/min. Injection was made in split mode at 10:1 with an injection volume of 1 μL and an injector temperature of 250℃. The temperature of the ion source and MS transfer line was 300℃ and 250℃, respectively. The column temperature was programmed to increase from an initial temperature of 90℃, followed by an increase to 120℃ at 10℃/min, and to 150℃ at 5℃/min, and finally to 250℃ at 25℃/min, which was maintained for 2 min. Mass spectrometric detection of metabolites was performed on an ISQ 7000 (Thermo Fisher Scientific, USA) with electron impact ionization mode. Single ion monitoring (SIM) mode was used with the electron energy of 70 eV. Statistical analyses The QIIME2 software was utilized to compute the Chao1 index, Good’s coverage index, Shannon index, and Simpson index for each sample group. Boxplots were constructed to assess the seasonal variations in the richness and evenness of food OTUs/microbial ASVs. Based on the Bray–Curtis distance, permutational multivariate analysis of variance (PERMANOVA) was employed to test the overall differences in the food and gut microbiota compositions between seasons. Subsequently, principal coordinates analysis (PCoA) was used to visualize these community structure differences. Linear discriminant analysis Effect Size (LEfSe) was used to determine differences in the gut microbiota between the groups. Bar/pie plot of food and gut microbiota composition between the groups using the online sites (https://www.genescloud.cn/analysis/overView). Pairwise correlations among microbial species were calculated in R to construct a correlation matrix. A threshold for significant interactions was defined, and values below the threshold in the R matrix were set to 0. The filtered matrix was then visualized as a network using Gephi. Redundancy analysis (RDA) was conducted to explore the influence of diet composition on gut microbial structure (based on relative abundance) using Canoco 5 [35]. The metabolic functions of the microbial community were predicted using PICRUSt2 (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States) based on the MetaCyc (https://metacyc.org/) and KEGG (https://www.kegg.jp/) databases [36]. In order to compare the correlations among food, gut microorganisms and SCFAs, the Spearman rank correlation coefficients for pairwise comparisons were calculated, and network analysis was conducted using Cytoscope v3.10.4 software. The food niche breadth is quantified using Simpson's diversity index (B) [37], where Pi represents the proportion of the i-th utilized food item relative to the total consumed food resources. The food niche overlap is measured using Levins' index (Ojk or Okj) [38], where Pij and Pik represent the proportions of the i-th food resource in the total diet of groups j and k, respectively. The results of SCFA were analyzed using one-way ANOVA with GraphPad Prism 8 (GraphPad Software). Spearman correlation analysis between microbiota and SCFAs was performed using the corrplot and psych R packages, with the results presented in a heatmap. Results Seasonal dietary variations in wild yaks revealed by DNA metabarcoding DNA metabarcoding sequences were carried out on fecal samples from 93 wild yaks, which were collected from three stages of the grassland in the QTP, with 4,190 m above sea level (Fig. 1). After quality control, a total of 12,713,598 paired-end reads were generated. These could be identified as dietary sequences represented by 28 plant families, 47 genera, and 56 species. As the number of sampled reads increased, the curve gradually reached a plateau, indicating that the number of libraries collected in this study was sufficient (Fig. S1). The richness of diet species of wild yaks was the highest in winter and the lowest in summer (P < 0.05). The diversity of diet species was the lowest in summer (P < 0.05, Fig. 2A). The PCoA results showed that the differences in the diet of wild yaks between seasons could be distinguished clearly (Fig. 2B). The coverage rate of the top ten diet species of wild yak accounted for more than 75% in each season (Fig. 2C, Table S1). At the family level, Polygonaceae and Rosaceae were the dominant forage in summer, accounting for 84.38%, while Rosaceae, Polygonaceae, Poaceae, Asteraceae, and Fabaceae were the dominant forage in autumn. Winter consists of Rosaceae, Ephedraceae, Poaceae, Asteraceae, and Fabaceae (Fig. 2C). Sibbaldia cuneata was frequently fed by wild yaks throughout the year. The highest relative abundance in summer was found in Bistorta vivipara (46.9%) and Sibbaldia cuneata (27.92%). Ephedra monosperma (23.27%) and Sibbaldia cuneata (22.27%) were predominant in winter (Fig. 2D). Spatial distribution map of wild yak fecal sample collection sites The Diet of wild yak has significant diversity. A Alpha diversity analysis of diet composition in the wild yak. B The PCoA analysis revealed significant differences among diets from different seasons. The two principal component scores accounted for 44.7% (PC1) and 17.9% (PC2) of the total variations, respectively. Each symbol represents an individual sample. C Bar chart depicting the relative abundance of diet in wild yaks. D Pie chart depicting the relative abundance of diet in the family. S_WY means wild yak in summer, A_WY means wild yak in autumn, and W_WY means wild yak in winter. ** means p-value between 0.01 and 0.005, and *** means p-value between 0.005 and 0.001 The calculated trophic niche breadth of wild yaks during summer, autumn, and winter was BS_WY = 3.2921, BA_WY = 8.8449, and BW_WY = 7.2235, respectively, with the highest value observed in autumn and the lowest in summer (Table 1). Pairwise comparisons revealed that the niche overlap indices between summer-autumn (OS_WY×A_WY = 0.8406), summer–winter (OS_WY×W_WY = 0.3751), and autumn–winter (OA_WY×W_WY = 0.6815) showed the greatest overlap between summer and autumn, and the least between summer and winter (Table 2). Table 1Food niche breadth of wild yaks in different seasonsSpeciesSummer autumn winter Ephedra monosperma0.0002770.0057790.054153Sibbaldia cuneata0.0779360.0317830.049577Leontopodium pusillum1.78E-050.0064090.019472Littledalea racemosa2.34E-050.0017680.004845Trisetum spicatum3.68E-060.0002630.002773Potentilla multiceps0.0017670.0009270.000445Aster flaccidus9.48E-060.0013540.000187Bistorta vivipara0.2199330.0439910.000178Astragalus membranaceus2.09E-050.0066775.51E-05Rheum spiciforme0.0024221.65E-069.85E-08others0.0040740.0174410.0094160.3064850.1163940.1411 Food niche breadth of wild yaks in different seasons Table 2Food niche overlaps of wild yaks in different seasonsSpeciessummer × autumnsummer × winterautumn × winterPijPikPijPikPijPikSibbaldia cuneata0.049770.062160.039695Potentilla multiceps0.001280.0008870.000643Bistorta vivipara0.0983620.0062540.002797Rheum spiciforme6.32E-051.54E-054.03E-07Littledalea racemosa0.0002040.0003370.002927Trisetum spicatum3.11E-050.0001010.000854Astragalus membranaceus0.0003733.39E-050.000606Ephedra monosperma0.0012650.0038740.01769Leontopodium pusillum0.0003380.0005880.011171Aster flaccidus0.0001134.21E-050.000502others0.0039750.0026250.008371∑PijPik0.155774510.0769170.085257(∑Pij2∑Pik2)1/20.1853168180.2050623980.205062 Food niche overlaps of wild yaks in different seasons Seasonal gut microbiota variations in wild yaks revealed by 16S rRNA gene amplicon sequencing 16S rRNA gene amplicon sequencing was also carried out on fecal samples from the 93 wild yaks above-mentioned. After quality control, a total of 10,434,865 paired-end reads were generated. As the number of sampled reads increased, the curve gradually reached a plateau, indicating that the number of libraries collected in this study was sufficient. Additional data would only reveal a limited number of new species. According to the species accumulation curves estimation based on a random sampling strategy (Fig. S2). A Venn diagram was used to confirm the common gut microbiota in the wild yak in different seasons. The results suggested that 4632 bacterial ASVs were shared by different seasons (Fig. 3A). The result of α-diversity analysis revealed that the gut microbiome of wild yak in winter had the highest richness, while goods_coverage was inverse (P < 0.05, Fig. 3B). PCoA analysis revealed significant differences in the composition of communities in the gut of wild yaks among three seasons (R2 = 0.14, P < 0.001, Fig. 3C). A Venn diagram of the gut microbiome in the wild yak. B Alpha diversity analysis of gut microbiome in wild yak. C The PCoA analysis revealed significant differences among communities from different sampling points. The two principal component scores accounted for 11.3% (PC1) and 4.3% (PC2) of the total variations, respectively. Each symbol represents an individual sample. * means p-value between 0.05 and 0.01, ** means p-value between 0.01 and 0.005, and *** means p-value between 0.005 and 0.001 Seasonal variations in gut microbiota composition of wild yaks In total, 26 gut bacterial phyla were identified, with Firmicutes_A and Bacteroidetes being the most abundant, regardless of season. We calculated the ratio of Firmicutes/Bacteroidetes; the result in winter was significantly higher than in summer and autumn (P < 0.001), and the abundance of Firmicutes was significantly higher in winter than in summer and autumn (Fig. 4A-B, Fig. S3). At the family level, the top 20 microorganisms in abundance were analyzed (Fig. 4C, Fig. S4). The abundances of UBA932, CAG.74, Peptostreptococcaceae, and Ruminococcaceae were higher in winter than in the summer and autumn (P < 0.001), while the abundance of Rikenellaceae was lower (P < 0.001). There was no difference in the abundances of Oscillospiraceae, unclassified_Bacteria, CAG-508, and Akkermansiaceae (P > 0.05). The results of the linear discriminant analysis (LDA) effect size also showed that there were more marker species with significant differences in winter than in summer and autumn (Fig. S5). In addition, we analyzed the intestinal microbiota at the genus level and found that most of the dominant bacteria showed significant differences between seasons (Fig. S6). Bacterial differences among wild Yak in three stages. A Bar chart depicting the relative abundance of gut bacteria in the top 10 phyla. B The ratio of Firmicutes/Bacteroidetes in different seasons. C Bar chart depicting the relative abundance of gut bacteria in the top 20 in the family. D-F Co-occurrence networks of all genera in gut bacteria. Nodes represent genera, and their sizes indicate different relative abundances. Links between the nodes represent a significant and strong correlation between the two genera (Spearman’s correlation greater than 0.5 or lower than − 0.6). *** means p-value between. 0.005 and 0.001, and **** means p-value less than 0.001 Analysis of the gut microbiome in wild yaks revealed distinct seasonal patterns in microbial interactions at the genus level. Correlation network analysis demonstrated that the primary interactions occurred among Firmicutes_A, Actinobacteriota, Firmicutes_D, Proteobacteria, and Bacteroidota (Fig. 4D-F). The network topology indicated seasonal variations in microbial interactions: summer and winter exhibited weaker microbial interactions with fewer negative correlations, while autumn showed a more complex network characterized by stronger antagonistic effects (27.82%; Fig. 4E). Modularity analysis revealed distinct co-occurrence patterns across seasons. The networks contained 2, 4, and 2 modules with modularity (Q value) exceeding 10% in summer, autumn, and winter, respectively (Fig. S7, Table S2). These modules, characterized by tightly connected internal nodes, likely represent specific functional units or ecological niche partitions. Dietary drivers of gut microbiome distribution in wild yaks: insights from redundancy discriminant analysis To elucidate the dietary drivers shaping the distribution of the gut microbiome, redundancy discriminant analysis (RDA) was employed to analyze the patterns of diet composition (at both family and species levels) across different phylogenetic levels. The highest explanatory power was observed at the class and order levels (Fig. 5, Table S3). In Fig. 5A, the first and second axes collectively accounted for 72.10% of the total variance, with Polygonaceae explaining 26.00% of the variance and exhibiting a significant positive correlation with Bacteroidia. In Fig. 5B, the first and second axes explained 69.28% of the total variance, whereas Polygonaceae contributed 26.20% of the variance and demonstrated a significant positive correlation with Bacteroidales. In Fig. 5C, the first and second axes explained 85.44% of the total variance, with Bistorta vivipara accounting for 25.20% of the variance and showing a significant positive correlation with Bacteroidia. Similarly, in Fig. 5D, the first and second axes explained 77.88% of the total variance, and Bistorta vivipara contributed 25.10% of the variance, displaying a significant positive correlation with Bacteroidales. At the diet species level, Bistorta vivipara emerged as the most influential contributor, while Polygonaceae was the dominant contributor at the family level. Redundancy analysis (RDA) of the associations between dietary composition and gut microbiota at different taxonomic levels. A RDA illustrates the relationship between dietary composition at the family level and bacterial communities at the class level. B RDA shows the correlation between dietary composition at the family level and bacterial communities at the order level. C RDA demonstrates the association between dietary composition at the species level and bacterial communities at the class level. D RDA represents the relationship between dietary composition at the species level and bacterial communities at the order level The relationship between the gut microbiota and short-chain fatty acids Spearman correlation analysis of the wild yak gut microbiota at the family level revealed significant seasonal associations with SCFA concentrations (Fig. S8): In summer (Fig. 6A), CAG.272 exhibited significantly positive correlations with acetate, propionate, isobutyrate, and isovalerate concentrations (P < 0.05). During autumn (Fig. 6 B), Acutalibacteraceae showed negative correlations with all SCFAs, reaching statistical significance for isobutyrate and valerate; conversely, Micrococcaceae demonstrated a significantly positive correlation with caproate (P < 0.05), while Ruminococcaceae correlated positively with butyrate (P < 0.05), with overall stronger microbiota-SCFA correlations observed in this season. In winter (Fig. 6C), significant positive correlations were identified between CAG.74 and valerate, and between Anaerovoracaceae and caproate concentrations (P < 0.05). Relationship between wild yaks’ microbes and SCFA concentration. A Wild yaks in summer; B wild yaks in autumn; C wild yaks in winter. The brown means a positive correlation relationship, and the blue represents a negative correlation relationship. The number on the right represents the correlation relationship ratio r. * means p-value between 0.05 and 0.01, and ** means p-value between 0.01 and 0.005 Correlation analysis was conducted on the food, intestinal microbiota and SCFAs of wild yaks, as shown in Fig. 7A, with a correlation coefficient greater than 0.5. The results of the correlation analysis show that SCFAs have the most nodes around them, in the following order: propionic acid, acetic acid, butyric acid, valeric acid, and isovaleric acid. Secondly, there are intestinal microorganisms, including g__UBA644, g__RUG11942, g__PeH17, g__RF16. A Analysis diagram of the interaction network between gut microbiota, food and short-chain fatty acids. The red line indicates a positive correlation, while the blue line represents a negative correlation. From the outer circle to the inner circle, they are the representative intestinal microorganisms of wild yak, the types of food consumed, and short-chain fatty acids. B Metabolic pathway associated with gut microbiome and SCFAs. Collagen is enriched in the protein metabolism of intestinal microorganisms. During the microbial metabolism of undigested proteins and carbohydrates in the colon, Butyrate is upregulated in summer and winter, while isovalerate and isobutyrate are upregulated in the protein metabolism process Integrated analysis of gut microbial gene function and SCFA-related KEGG pathways revealed seasonal dynamics in wild yaks (Fig. 7B). Specifically, butyrate (C4) biosynthesis pathways under carbohydrate digestion and metabolism were significantly enriched during summer, as demonstrated by SCFA metabolic profiling. The observed enrichment pattern showed a discrepancy with the gene functional analysis of identical metabolic pathways in gut microbiota. SCFAs metabolic profiling demonstrated significant enrichment of butyrate biosynthesis pathways within carbohydrate digestion and metabolism during summer. Notably, this enrichment pattern was not mirrored in gut microbial gene functional analysis of the same pathway. Conversely, pathways associated with SCFAs generation in protein digestion and nutrient absorption showed significant summer enrichment. Comparative analysis indicated winter-specific significant enrichment of butyrogenic pathways relative to autumn. Furthermore, gene functional analysis of gut microbiota for this winter-enriched pathway exhibited an elevated abundance of collagen-related genes. Discussion Mammalian gut microbes are affected by a variety of factors. The gut microbiota of conspecific organisms living in close areas is mainly influenced by diet [39]. The results showed that different diets and seasons had altered gut microbial community composition and function in wild yaks. This study elucidated the changing trends and adaptive mechanisms of gut microbiota during changes in feeding habits. In response to environmental changes, the wild yak adjusts its dietary niche Although the dietary spectrum of wild yaks was similar in each season, the intake showed significant differences. Bistorta vivipara showed a gradual decrease in feed intake from summer to winter, while Ephedra monosperma was more abundant in winter. Bistorta vivipara, a preferred ruminant forage, was the primary summer food source for wild yaks when resources were abundant. It has been found that during adaptation to extreme cold, plants harden at low temperatures, and the lipid metabolism of Ephedra monosperma changes [40]. Shifts in dietary composition and increases in absolute energy intake are believed to be fundamental strategies for coping with extreme environmental conditions [5]. We considered that the nutrients and metabolites of this forage changed during growth and that it was more suitable for wild yaks in winter. Wild yaks are opportunistic foragers, able to adapt to resource dynamics by flexibly adjusting their food choices and home ranges [41]. The broad dietary spectrum was observed in autumn, indicating that individuals expanded their food niche in response to declining summer forage availability. To optimize foraging efficiency, animals exhibit behavioral and physiological adaptations [42]. Employing seasonal niche partitioning and pronounced dietary shifts, wild yaks exhibit marked trophic plasticity to cope with resource limitation in cold climates. Seasonal variation trends in the gut microbiota of wild yaks The rumen and intestines of ruminants harbor a vast number of microorganisms, including bacteria, archaea, fungi, and protozoa, with a wide variety of species and enormous quantities. Among these, bacteria are the most abundant and functionally significant group of microorganisms [43]. Bacteria play a key role in processes such as cellulose breakdown, protein degradation, and energy metabolism [44]. Our alpha and beta diversity analyses reveal that although wild yaks inhabit the same region year-round, their gut microbiome adapts to seasonal variations in dietary intake. The observed seasonal divergence in wild yak gut microbiota with pronounced summer–winter differences and transitional autumn profiles parallels conserved patterns across mammalian species [22, 45]. Our results demonstrate that seasonal dietary shifts induce corresponding compositional changes in gut microbiota. Wild yaks depend on gut microbiota for digestive and metabolic functions. Studies demonstrate that their microbial communities exhibit enhanced cellulose-degrading capacity, facilitating more efficient energy extraction from low-quality forage [46]. During winter, forage quality declines, and gut microbiota undergo structural and functional adaptations to optimize energy harvest from nutrient-depleted diets. During winter, the gut microbiota facilitates wild yaks' adaptation to nutrient scarcity by augmenting energy extraction from fibrous diets, thereby directly countering the seasonal energy deficit. This capability stems in part from the distinct metabolic roles of two key bacterial phyla: Firmicutes, which exhibit superior efficiency in degrading complex carbohydrates, thereby maximizing caloric extraction, and Bacteroidetes, which are adept at catabolizing simpler glycans [47, 48]. An elevated F/B ratio was observed during winter. We propose that this increase constitutes a key microbial adaptation for optimizing energy acquisition during nutritional stress, thereby providing an ecological advantage in the forage-limited QTP winter. Certain strains of Alistipes_A demonstrated seasonal variation in wild yak gut microbiota, with higher abundance observed in autumn samples. Selected Alistipes strains have been reported to function as probiotics through SCFA production (primarily propionate and butyrate) and anti-inflammatory effects in murine models [49]. Faecousia in summer is associated with carbohydrate degradation (Yang et al., 2024), and RF16 is associated with food digestion [50, 51]. Besides, the study of human gut microbiota showed that DTU089 was found to be associated with bone growth, and DTU089 levels were higher in people with less physical activity and lower protein intake [52]. DTU089 abundance in the gut microbiota of wild yaks is significantly lower in autumn and winter, coinciding with reduced dietary protein levels. Seasonal pattern suggests that the spatial distribution of forage, rather than protein content per se, is the primary driver of DTU089 fluctuations. During summer, when high-quality forage is abundant and grazing ranges contract, DTU089 reaches its maximum abundance. Paludibacteraceae (linked to food's thermal effect) and Acutalibacteraceae (involved in SCFA metabolism) [53, 54], with high abundance in autumn, may aid low-nutrient adaptation. In winter, the carbohydrate-utilizing genus Romboutsia and the SCFA-producing family Peptostreptococcaceae both increased in the gut microbiome, indicating a shift toward fat and energy metabolism [55, 56]. In summer, microbial diversity increased with a rise in carbohydrate-metabolizing taxa, likely aiding in digesting sugar-rich forage. This shifted in autumn to a marked increase in immunomodulatory microbes, potentially reflecting heightened gut barrier activity. Winter microbiota shifted toward energy-efficient taxa, coinciding with expanded foraging ranges, suggesting adaptive responses to low-resource conditions. As wild yaks extend their winter foraging range, their physical activity levels rise, and both metabolic efficiency and microbial metabolic-gene abundance rise significantly. The complex interactions of wild yak gut microbiota in autumn The autumn gut microbiota network exhibited multiple key nodes, including the probiotic genus Akkermansia, which is functionally linked to intestinal barrier enhancement through mucin degradation and tight junction modulation [57, 58]. Previous studies on domestic yaks have demonstrated that Akkermansia contributes to nutritional homeostasis maintenance in high-altitude environments [5]. Ventrisoma (family Lachnospiraceae) comprises anaerobic, Gram-positive gut bacteria. Like other Lachnospiraceae members, it likely ferments complex polysaccharides into SCFAs. These SCFAs modulate host physiology through immunoregulatory and anti-inflammatory mechanisms [59–62]. Within the autumn gut microbiota ecosystem, putative probiotic taxa, including Ruminococcus_D and Clostridium_P, emerge as keystone nodes in microbial co-occurrence networks. Their topological centrality suggests critical roles in maintaining network stability, likely through polysaccharide fermentation (yielding SCFAs) and immunomodulatory interactions [59, 61]. Conversely, heavy gut colonization by Enterococcus downregulates the mucosal vascular addressin MAdCAM-1 and correlates inversely with the abundance of other commensals [63]. We therefore propose that the autumn gut environment of wild yaks selects for a microbiota that integrates digestion, metabolism, and immune regulation. Rather than relying on a few exclusive pairwise relationships, this community is shaped by extensive [64], function-based interactions that enhance seasonal adaptability and host homeostasis. Although wild-animal gut microbiota is shaped by multiple factors, dietary variation is the principal determinant [65–67]. Seasonal shifts significantly altered the dominant genera of the wild yak gut microbiota. Seasonal variation significantly reshaped the dominant bacterial genera in wild yak gut microbiota. Network analysis revealed limited antagonistic interactions during summer (forage peak) and winter (forage dormancy), whereas autumn (forage senescence) exhibited dense negative correlations (27.8%) and higher functional modularity, indicating disrupted intestinal homeostasis. Thus, the autumn gut microbiota of wild yaks exhibits pronounced functional niche partitioning and enhanced synergistic interactions [68], enabling rapid adaptation to environmental fluctuations while sustaining host homeostasis. Dietary shifts reshape the gut microbiota community structure and composition of wild yaks To quantify diet-microbiota relationships, we performed redundancy analysis (RDA) using dietary variables and gut microbiota profiles at family and species levels. Diet explained the largest proportion of variance at class and order levels (P < 0.01), suggesting that these higher taxonomic ranks are sufficiently stable to serve as robust indicators of host dietary intake [69]. At the class and order levels, microbial taxa most faithfully tracked seasonal shifts in the wild yak’s diet and associated metabolic demands. Pronounced fluctuations in Polygonaceae (particularly Bistorta vivipara) intake served as the primary driver, with this plant group showing a strong positive correlation with both Bacteroidia (class) and Bacteroidales (order). Functionally, Bacteroidia depolymerize recalcitrant plant polysaccharides indigestible by the host and ferment the resulting oligosaccharides into SCFAs, providing an additional energy source [70]. Given the marked decline of Bacteroidetes in winter, which is driven by the concomitant reduction in dietary plant diversity, we hypothesize that seasonal dietary shifts are the primary determinant of gut-microbiota composition, contingent upon substantial phylum-level fluctuations. From a community-engineering standpoint, negative and weak microbial interactions jointly enhance gut-microbiota stability [71]. As autumn entails a rapid decline in food availability, wild yaks are forced to adopt novel foraging strategies. Consequently, their gut microbiota, now recognized as a sensitive biosensor, rapidly restructures in response to these environmental perturbations, supporting its utility for ecological monitoring [72, 73]. Wild yaks optimize their winter diet by balancing nutritional gain against foraging costs, resulting in a relatively stable dietary profile. During dietary transitions in autumn, fluctuating substrate availability destabilizes the intestinal milieu, eliciting rapid microbial reconfiguration. Consequently, the microbial community simplifies its interactions in summer and winter to maintain homeostasis, but employs negative feedback through metabolites and immune signals in autumn to buffer environmental fluctuations. Short-chain fatty acids: gut microbiota fermentation and intestinal homeostasis regulation During summer, abundant nutrients enhance microbial fermentation and the metabolism of SCFAs in wild yaks. In autumn and winter, reduced forage quality coincides with lower energetic efficiency. Although dietary collagen is negligible, metagenomic profiling revealed significant enrichment of collagen-degrading genes within the autumn and winter microbiota, suggesting endogenous or microbially derived collagen metabolism. The gut microbiota of herbivores modulates mucus-layer permeability and immune signaling, thereby maintaining host–microbe homeostasis [74]. To compensate for seasonal nutrient scarcity, wild yaks appear to exploit microbial collagenolytic pathways that release amino acids from host or dietary collagen. Concurrently, the microbiota-derived metabolite butyrate serves as a central energy source and reinforces colonic barrier integrity, immune modulation, and epithelial homeostasis. The underlying mechanism entails concentration-dependent diffusion, inter-microbial signaling, and host-receptor activation [75]. Autumn and winter severe nutrient restriction typically lowers colonic butyrate; however, our data indicate a winter rebound in butyrate concentration. Network analyses coupled with metabolite profiles reveal that autumn is characterized by microbial dysbiosis, coinciding with reduced butyrate and heightened immune signaling. Within this context, Clostridium emerges as a keystone taxon whose interactions may stabilize community function. Conversely, winter recovery is driven by Ruminococcaceae, whose cat1-encoded CoA-transferase pathway efficiently converts dietary fiber to butyrate. Consistent with its role as a sentinel of microbial balance, the marked increase in Ruminococcaceae abundance from autumn to winter parallels the restoration of intestinal homeostasis and highlights its functional significance during seasonal dietary transitions. On the QTP, the gut microbiota enables wild yaks to adapt to seasonal diets. As forage changes, the microbiome restructures its function: it up-regulates carbohydrate-degrading enzymes and short-chain fatty acid pathways in winter to maximize energy extraction, and enhances host-microbe interactions in autumn to maintain mucosal integrity and immune homeostasis. This gradual dietary transition provides structural stability while allowing functional fine-tuning, demonstrating the microbiota's adaptive capacity—a trait vital for survival in this extreme environment. Conclusion Our findings demonstrate that seasonal dietary shifts are a key driver of taxonomic, functional, and metabolomic reconfiguration in the wild yak gut microbiota. SCFA-producing taxa were enriched in winter, whereas potentially pathogenic genera linked to intestinal inflammation peaked in autumn. Winter microbiomes exhibited enhanced capacity to degrade recalcitrant plant polymers, elevating digestive efficiency and energy yield. Conversely, autumnal dysbiosis was mitigated by up-regulated immunomodulatory pathways, indicating that microbial functional plasticity buffers the adverse effects of dietary instability. These patterns were corroborated by targeted SCFA metabolomics. Collectively, the study provides mechanistic evidence for dietary adaptation in high-altitude herbivores and extends optimal-foraging theory to extreme environments. The data also offer a theoretical basis for conservation strategies and caution against abrupt dietary modifications in captive husbandry of plateau ungulates. One limitation of this study lies in the use of fecal samples instead of rumen fluid samples to characterize the microbiome. Although the fecal microbiome is easily accessible and can provide valuable information about the overall intestinal ecology, it differs from the rumen microbiome, which is mainly responsible for fiber degradation. Future research, by directly analyzing the rumen microbiome, will be able to more directly reveal its functional mechanisms in digestive adaptation. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements The authors express their gratitude to the Northwest Institute of Plateau Biology, Chinese Academy of Sciences. Thanks to the support of the Chief Scientist Program of Qinghai Province (2024-SF-102), the Joint Fund Project of NSFC (U21A20250), and the Youth Innovation Promotion Association, Chinese Academy of Sciences. Authors’ contributions Jiahui Hao conceived the project. Hongjin Liu and Shixiao Xu designed the project. Xungang Wang and Lele An analyzed experimental data. Jiahui Hao, Tongqing Guo, and Qian Zhang discussed the results. Jiahui Hao wrote the manuscript. Hongjin Liu and Shixiao Xu review and edit. All the authors contributed to the manuscript revision and approved the final manuscript. Funding This work was supported by the Chief Scientist Program of Qinghai Province (2024-SF-102), the Joint Fund Project of NSFC (U21A20250), and the Youth Innovation Promotion Association, Chinese Academy of Sciences. Data availability The raw sequencing data for this study have been deposited in the Sequence Read Archive (SRA) database under the accession numbers PRJNA1244544 (diet) and PRJNA1244101 (gut microbiome). Declarations Ethics approval and consent to participate This experimental protocol was reviewed and approved by the Institution of Animal Care and Ethics Committee of the Northwest Institute of Plateau Biology, Chinese Academy of Sciences. The authors declared that experimental research on the plants described in this paper complies with institutional, national, and international guidelines. Field studies were conducted following local legislation and obtained permissions from the provincial department of forest and grass of Qinghai. The specimens were then stored at the herbarium in the Northwest Institute of Plateau Biology, Chinese Academy of Sciences. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Hongjin Liu, Email: liuhj@nwipb.cas.cn. Shixiao Xu, Email: sxxu@nwipb.cas.cn. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement The raw sequencing data for this study have been deposited in the Sequence Read Archive (SRA) database under the accession numbers PRJNA1244544 (diet) and PRJNA1244101 (gut microbiome).