Abstract
An intensive feeding system might improve the production cycle of yaks. However, how intensive feeding system contributes to yak growth is unclear. Here, multi-omics, including rumen metagenomics, rumen and plasma metabolomics, were performed to classify the regulatory mechanisms of intensive feeding system on yaks. Increased growth performance were observed. Rumen metagenomics revealed that Clostridium, Methanobrevibacter, Piromyces and Anaeromyces increased in the intensively fed yaks, contributing to amino acid and carbohydrate metabolism. The grazing yaks had more cellulolytic microbes. These microbiomes were correlated with the pathways of “Alanine aspartate and glutamate metabolism” and “Pyruvate metabolism”. Intensive feeding increased methane degradation functions, while grazing yaks had higher methyl metabolites associated with methane production. These rumen microbiomes and their metabolites resulted in changes in plasma metabolome, finally influencing yaks’ growth. Thus, an intensive feeding system altered the rumen microbiome and metabolism as well as host metabolism, resulting in improvements of yak growth.
Similar content being viewed by others
Introduction
The yak (Bos grunniens), inhabiting the Qinghai-Tibetan Plateau (the world’s highest plateau) in China, adapts to the harsh environment with low partial pressure of oxygen, and a high level of ultraviolet radiation1,2. Traditionally, yaks graze yearly on high altitude pastures and experience inadequate nutrient sources in the long-cold winter. Thus, some specific anatomical and physiological traits in the yaks have been developed, such as a giant heart and lung, smaller blood cells, lack of hypoxic pulmonary vasoconstriction, increased ability to graze forage and high energy metabolism3,4. However, some common features between yak and other ruminants are observed. For example, their rumen has a complex microbial community and can ferment fibrous plant materials into volatile fatty acids (VFAs) for the host requirement5. In the meantime, an extraordinary ability to degrade lignocellulose, produce VFAs and metabolize nitrogen was found in the rumen of yak compared to cattle under the same feeding system6,7. Due to malnutrition and the extreme living environment, longer production cycle and poor meat quality in yaks are the major problems for yak industry which need to be addressed as consumer demands increase8. Although hybridization of yaks and other ruminant breeds has been conducted and could improve productivity, improvements of feeding strategy to increase purebred yak productivity may be the best approach for the long-term developments of the yak industry.
An intensive feeding system provides the precisely designed diet to housed animals to maximize the partitioning of feed nutrients into productive outputs and minimize potentially polluting emissions9. Intensive feeding programs have been widely applied to beef and dairy cattle. For yaks, researchers have investigated the effects of the semi-intensive grassland-based systems and obtained improved results of growth performance, rumen fermentation and changes of gut microbiota via supplementation of concentrate diet10,11,12,13. Furthermore, intensive feeding systems improved nutrient digestibility, body metabolism, meat quality and milk yield have been reported14,15,16,17. As we know, the rumen as the most important digestive organ converts plant materials into VFAs and proteins for the host requirement via microbial fermentation18,19. One study found that feedlot fattening changed the rumen fermentation and microbial diversity as well as meat quality in yaks20. Another research observed that warm-grazing and cold-indoor feeding regimes influenced yak rumen microbiome and greenhouse gas emissions compared to traditionally grazing yaks21. However, these studies use 16S rRNA sequencing technology, resulting in less knowledge in low abundance bacteria and other microbiota. Changes in the rumen microbiome and its functions in response to an intensive feeding system and its subsequent influences on the body metabolism in yaks are still unclear. Additionally, some questions need to be addressed, such as whether an intensive feeding system fits the yaks living in a high-altitude area, whether changes in yak rumen microbiome by intensive feeding are similar to other ruminants (e.g., beef and dairy cattle), and how intensive feeding affect the rumen microbiome (composition and functions), microbial metabolites, and the host metabolites in yaks.
Therefore, in this study, we performed rumen metagenomics, rumen metabolomics, and plasma metabolomics to identify the impacts of an intensive feeding system on yaks compared to the traditional grazing system. This study will provide knowledge of the rumen microbiome-dependent mechanism to improve the host phenotypes, and evidence of the advantages of an intensive feeding system for the livestock industry.
Results
Intensive feeding system significantly altered the growth, metabolism and physiology of yaks
As designed, the nutrient level of the diet in the intensive feeding system was significantly greater than forage on the natural grassland. The crude protein and starch of the intensive diet were 16.7% and 36.2% compared to 8.5% and 3.0% in forage (Supplementary Table 1). In the meantime, the NDF and ADF were lower in intensive diet (20.1% and 31.1%) than that in grazed forage (35.0% and 51.5%). Although the forage intake was not measured, significant improvements of the bodyweight of yaks consuming an intensive diet were observed as final body weight and ADG in the intensive group was higher compared to the grazing group (Supplementary Table 2).
Next, serum biochemical parameters associated with the health status of animals and nutrient metabolism of diet in the body were measured in this study (Supplementary Table 3). Alkaline phosphatase (ALP), total bilirubin (TBil), direct bilirubin (Dbil), indirect bilirubin (Ibil) and glucose in the intensive group were significantly higher than those in the grazing group, while concentration of creatinine (Cr) and total cholesterol (TC) were significantly greater in the grazing yaks (P < 0.05). Other parameters, including the total protein (TP), albumin (ALB), globulin (GLB), urea and triglyceride (TG), were not different between the two groups (P > 0.05). These results indicated that the body metabolism, including carbohydrate and lipid, was influenced. In addition, regarding the hash living environment (e.g., high altitude, low oxygen) of Zhongdian yaks, the blood physiological parameters were also determined (Supplementary Table 4). The results showed that red blood cell (RBC) count, hemoglobin (HGB), and mean corpuscular hemoglobin concentration (MCHC) were high in grazing groups. On the contrary, mean corpuscular volume (MCV), mean cellular hemoglobin (MCH), red cell distribution width variance coefficient (RDW-CV), platelet count (PLT) and mean platelet volume (MPV) were increased by the intensive feeding system.
Rumen metagenomics revealed the diversity and structure of the microbiome associated with intensive feeding system
Metagenome sequencing generated a total of 230.5 Gbp data, with 1,026,889,026 reads (73,349,216 reads per sample). After de novo assembly, a total of 2,240,346 contigs were generated (the N50 length of 1,310 bp), with 160,024 per sample. The rumen metagenome across all samples consisted of 97.56% bacteria, 1.75% eukaryotes, 0.49% archaea, and 0.20% viruses (Supplementary Table 5). The rumen microbial domains, including bacteria and viruses, were increased by the intensive feeding system (P < 0.05), while the abundance of eukaryotes was greater in the grazing group. The principal coordinate analysis (PCoA) showed separations between the two groups based on rumen bacterial, eukaryotic and viral species (Fig. 1; Supplementary Table 6). In addition, no differences in the archaeal species were observed. At the same time, decreases in richness at the species level were observed among bacteria, archaea and eukaryote of yaks in the intensive group (Supplementary Figure 1).
Bacterial (A), archaeal (B), eukaryotic (C) and viral (D) compositional structure of the rumen samples from grazing and intensively fed yaks based on species visualized using principal-coordinate analysis (PCoA).
Compositional profiles of the rumen microbiome in yaks altered by intensive feeding system
The dominant rumen bacterial phyla were Bacteroidetes (52.29%), Firmicutes (34.81%) and Proteobacteria (5.41%) across all samples (Supplementary Fig. 2A). In the meantime, we found that the intensive feeding system increased the abundance of Firmicutes (40.85%) and decreased Bacteroidetes (47.67%) compared to the grazing group (28.76% and 56.90%). At the genus level, dominant bacteria included Prevotella (43.96%), Bacteroides (7.94%), Clostridium (4.46%), Alistipes (3.53%), Fibrobacter (2.77%) and Ruminococcus (2.44%) (Supplementary Fig. 3A). At the species level, LEfSe algorithm identified 11 Clostridium sp., 3 Prevotella sp., 3 Ruminobacter sp., 2 Succinatimonas sp., 1 Oscillibacter sp., Succiniclasticum ruminis, Ruminococcus bromii, and Faecalibacterium prausnitzii that were abundant in the intensive group (Fig. 2A). The grazing group had greater abundances of 5 Bacteroides sp., 3 Prevotella sp., 3 Alistipes sp., 3 Fibrobacter sp. and Pseudomonas fluorescens.
Heatmap depicting the significantly different bacterial (A), archaeal (B), eukaryotic (C) and viral (D) species. E Correlation networks showed associations among microbiome species across kingdom. The edge colors (red: positive, blue: negative) are proportional to the correlation strength. Only strong (Spearman R of > 0.7 or <−0.7) and significant (P < 0.05) correlations are displayed. Significant differences were tested by linear discriminant analysis effect size (LEfSe) analysis, with linear discriminant analysis (LDA) score of > 2 and P value of < 0.05.
For archaea, the dominant phyla were Euryarchaeota (92.92%) and Candidatus Heimdallarchaeota (2.53%) across all samples (Supplementary Fig. 2B). The intensive group had a higher abundance of Euryarchaeota but lower abundance of Candidatus Heimdallarchaeota, Candidatus Bathyarchaeota and Candidatus Woesearchaeota. At the genus level, Methanobrevibacter (73.59%), Methanobacterium (3.49%), Methanosarcina (3.47%), Methanothrix (2.86%) and Methanomassiliicoccus (2.31%) were dominant ((Fig. S3B). Except Methanobrevibacter was greater in the rumen of yaks of the intensive group, other dominant archaeal genera, including Methanosarcina, Methanothrix, Methanomassiliicoccus, Methanosphaera and Methanolobus, were higher in the rumen of grazing yaks. The rumen of intensively fed yaks was enriched by the species of Methanobrevibacter curvatus, Methanobrevibacter filiformis, Methanobrevibacter ccuticularis, and Lokiarchaeum sp. GC14_75 (Fig. 2B). However, 4 Euryarchaeota archaeon sp., 2 Methanolobus sp. (M. profundi and M. psychrophilus), 2 Candidatus Heimdallarchaeota archaeon sp. and Methanosaeta harundinacea were over-represented in the rumen of grazing yaks.
In addition, for the rumen Eukaryotic genera, we found that Neocallimastix, Piromyces and Anaeromyces were top taxa and had greater abundances in the intensive group (Supplementary Fig. 4A). The species, including Piromyces sp. E2, Anaeromyces robustus, Neocallimastix californiae, Arthrobotrys oligospora and Cadida auris, were greater in the intensive group (Fig. 2C). Abundances of Rozella allomycis, Rhizophagus irregularis, Basidiobolus meristosporus, Rhizoclosmatium gobosum, Allomyces macrogynus, Batrachochytrium dendrobatidis, Catenaria anguillulae, Conidiobolus coronatus, and Lobosporangium transversale were higher in the grazing group.
Moreover, the intensive feeding system increased the abundances of rumen viral genera, including C5virus, Cp220virus and N4virus (31.83%, 26.20% and 9.95%), compared to the grazing group (20.58%, 7.30% and 1.38%) (Supplementary Fig. 4B). On the contrary, higher abundances of Hokovirus, Alpharetrovirus, Catovirus, Pepy6virus, Bc431virus and Cervidpoxvirus were observed in the rumen of grazing yaks. The species, including Geobacillus virus E3, 2 Campylobacter virus (CP21 and INN35) and Enterobacter phage EcP1, had higher abundances in the intensive group (Fig. 2D). In grazing yaks, their rumens were abundant with Hokovirus HKV1, Avian myeloblastosis virus, Catovirus CTV1, Deerpox virus W-848-83 and Klosneuvirus KNV1.
Additionally, microbiome interactions at the species level were determined using network analysis (Fig. 2E). Complex correlations among bacterial, archaeal, eukaryotic and viral species were observed. Bacterial species were the bridge nodes to connect archaea, eukaryotes and viruses. For instance, Prevotella sp. (brevis, ruminicola) positively correlated with Methanomassiliicoccales archaeon RumEn M1, Methanosaeta harundinacea, Candidatus Heimdallarchaeota archaeon AB_125 and Rhizopus microsporus, but Clostridium sp. CAG-1024 negatively correlated with Catovirus CTV1, and Hokovirus HKV1. Archaeal species interacted with each other. Methanobrevibacter olleyae, Methanobrevibacter ruminantium, Methanobrevibacter wolinii and Methanobrevibacter smithii with positive correlation formed a subunit, while Methanomassiliicoccus luminyensis, Methanobrevibacter millerae, Candidatus Methanomassiliicoccus ntestinalis and Methanobrevibacter sp. A27 shaped like another subunit. The dominant eukaryotic and viral species had complex interactions.
Intensive feeding system changed functional profiles of the rumen microbiome in yaks
The functions of the rumen microbiome were determined by the Kyoto Encyclopedia of Genes and Genomes (KEGG) profiles and genes encoding CAZymes. For KEGG profiles at the first level category, “Metabolism” (52.43%) and “Genetic information processing” (19.50%) were dominant (Supplementary Table 7). When the KEGG pathways were compared, the rumen of intensively fed yaks were significantly enriched with the pathway associated with amino acid metabolism, such as “Alanine aspartate and glutamate metabolism” (ko00250), “Cysteine and methionine metabolism” (ko00270), “Valine, leucine and isoleucine biosynthesis” (ko00290), “Arginine biosynthesis” (ko00220), “Histidine metabolism” (ko00340), and associated with carbohydrate metabolism of “Pyruvate metabolism” (ko00620), “Glycolysis Gluconeogenesis” (ko00010), “Pentose phosphate pathway” (ko00030), “Citrate (TCA) cycle” (ko00020), “Butanoate metabolism” (ko00650), “Propanoate metabolism” (ko00640) (LDA > 2 and P < 0.05; Fig. 3). Moreover, other pathways associated with energy metabolism (e.g., “Methane metabolism” (ko00680), “Carbon fixation in photosynthetic organisms”, “Nitrogen metabolism”) were also higher in the intensive group. However, the grazing group was enriched with some signal transduction pathways, such as “AMPK signaling pathway” (ko04152), “FoxO signaling pathway” (ko04068) and “PI3K-Akt signaling pathway” (ko04151). The intensive group was enriched by 11 of the KEGG modules related to nucleotide and amino acid metabolism, including M00053, M00021, and 21 modules associated with carbohydrate metabolism, such as M00173, M00003, M00620, M00004, M00377, M00308, M00009, and M00374 (Supplementary Fig. 5). Overall, the intensive feeding system increased the function of amino acids and carbohydrate metabolism (Fig. 4).
Significant differences were tested by linear discriminant analysis effect size (LEfSe) analysis, with linear discriminant analysis (LDA) score of > 2 and P value of < 0.05.
Metabolic pathways involved in VFA biosynthesis and methanogenesis were displayed.
For CAZyme profiles, classes of glycoside hydrolases (GHs) (64.68%), glycosyltransferase (GTs) (15.60%), carbohydrate-binding modules (CBMs) (9.87%), carbohydrate esterases (CEs) (7.77%) were dominant in all samples (Supplementary Fig. 6). At the family level, 15 GHs, 7 GTs, 4 CMBs and 2 CEs were identified as signatures for the intensive group, while 11 GHs, 3 GTs and 8 CBMs were enriched in the grazing group (Supplementary Fig. 7). Moreover, the rumen CAZyme activities from the intensive group mainly were under GH13 associated with starch utilization, such as branching enzyme (EC2.4.1.18), cyclomaltodextrinase (EC3.2.1.54), alpha-amylase (EC3.2.1.1), isoamylase (EC3.2.1.68), amylosucrase (EC2.4.1.4), starch synthase (EC2.4.99.16), and neopullulanase (EC3.2.1.135) (Supplementary Fig. 8). The rumen of grazing yaks had a greater abundance of GH3 enzymes to degrade lignocelluloses, such as alpha-L-arabinofuranosidase (EC3.2.1.55), beta-glucosidase (EC3.2.1.21) and beta-xylosidase (EC3.2.1.37).
Associations between rumen microbiome and its functions
As protein and carbohydrate (e.g., fiber and starch) contents are two significant determining nutrient factors of the diet in the intensive feeding and grazing system, we further focused on the functions of amino acid and carbohydrate metabolism in the rumen microbiome. Two most important pathways involved in amino acid and carbohydrate metabolism were “Alanine aspartate and glutamate metabolism” (ko00250) and “Pyruvate metabolism” (ko00620) enriched in the rumen of intensive group yaks (LDA > 2, P < 0.05; Fig. 3). The abundances of genes encoding enzymes involved in these two pathways were also enriched in the rumen of intensive yaks, such as EC 6.3.5.5, EC 6.3.1.2 associated with ko00250 and EC1.2.7.1, EC4.2.1.2 within ko00620 (Supplementary Fig. 9). In addition, “Methane metabolism” (ko00680) and its related enzymes (e.g., EC1.2.7.1, EC 1.1.1.95 and EC 1.1.1.399) were higher in the intensive group.
A network with Spearman’s rank correlation between species and those three pathways was then built to estimate the associations between the rumen microbiome and its functions (Fig. 5). A total of 41 species (17 positive and 24 negative correlation) showed significant correlation with amino acid and carbohydrate metabolism pathways as well as methane metabolism (R > 0.70 and P < 0.05). Bacterial species played key roles in both positive and negative correlations. The species of Clostridium sp. showed the strongest positive relationships, while Bacteroides sp. were prominent members among negative associations. Additionally, archaeal species, including Methanobrevibacter curvatus, Methanobrevibacter filiformis, and Methanobrevibacter ccuticularis, with greater abundance in intensively fed yaks’ rumen were positively correlated with these three pathways. Correspondingly, signature archaea for the intensive group (e.g., Methanosaeta harundinacea, Candidatus Heimdallarchaeota archaeon AB-125) negatively related with “Alanine aspartate and glutamate metabolism”, “Pyruvate metabolism” and “Methane metabolism”. Interestingly, species of Fibrobacter sp. UWB4 and Methanolobus profundi had a negative correlation with “Methane metabolism”.
The edge colors (red: positive, blue: negative) are proportional to the correlation strength. Only strong (Spearman R of > 0.7 or < − 0.7) and significant (P < 0.05) correlations were displayed.
Rumen fermentation parameters and metabolome associated with intensive feeding system
The changes in microbial fermentation products affected by the intensive feeding system were estimated. We observed that the intensive group had significantly higher concentrations of valeric acid and isovaleric acid (Supplementary Table 8). The higher concentration of ammoniacal nitrogen was higher in the grazing group. Thus, the mode of rumen fermentation was altered by the feeding system.
We then analyzed the rumen metabolites using untargeted metabolomics. Orthogonal Partial Least Squares-Discriminant Analysis (OPLS-DA) showed remarkable differences between the intensive and grazing groups (Supplementary Fig. 10A). A total of 256 compounds were identified in the rumen metabolomics. After t-test and variable importance in projection (VIP) filtering for the relative concentrations of rumen metabolites, 42 metabolites were significantly higher in the rumen of intensive feeding yaks, such as L-Glutamic acid, 3-Methyladenine, Nicotinic acid, Gibberellin A97, Pyroglutamic acid, Valyl-Isoleucine, Pyridoxamine and Meta-Tyrosine, while Proline betaine, N-Acetyldopamine, 1-nitroaphthalene and N-Acetyldopamine were abundant in the grazing group (Supplementary Fig. 10B, Fig. 6A). Additionally, Metaorigins analysis was performed to track where the signature metabolites come from the host, microbiome, or both (Supplementary Fig. 11). We found that the signature metabolites differentiating the two groups were from both host and rumen microbiome and related to amino acids metabolism, such as “Histidine metabolism”, “Alanine, aspartate and glutamate metabolism”, “Arginine and proline metabolism”, and “Cysteine and methionine metabolism”. Then, using Metabolic pathway analysis (MetPA), a total of 11 pathways of these metabolites were identified as significantly different pathways (Benjamini-Hochberg false discovery rate (FDR) < 0.01, pathway impact > 0.1), such as “D-Glutamine and D-glutamate metabolism”, “Alanine, aspartate and glutamate metabolism”, “Histidine metabolism”, “Arginine and proline metabolism”, “Vitamin B6 metabolism”, “Glutathione metabolism” and “Arachidonic acid metabolism” (Fig. 6B).
A The Euclidean distance matrix for the quantitative values of differential metabolites was calculated, and a heatmap was selected to display the metabolites between the grazing and intensively fed yaks using a complete linkage method. B Pathway enrichment analysis performed using the significantly different rumen metabolites between grazing and intensively fed yaks.
Plasma metabolome influenced by the intensive feeding system
We also examined plasma metabolomes associated with the intensive feeding system. A significant shift of the structure in the plasma metabolome between the intensive and grazing groups was observed (Supplementary Figure 12). The comparison analysis revealed that the relative concentrations of 26 metabolites (e.g., D-Proline, L-Phenylalanine, Taurine, Graveolinine and Acetone cyanohydrin) were significantly higher in the blood of intensively fed yaks, and the relative concentrations of 118 metabolites (e.g., L-Carnitine, Proline betaine, Creatinine, Betaine, and L-Arginine) were significantly higher in the plasma of grazing yaks (P < 0.05, VIP > 1) (Supplementary Fig. 12, Fig. 7A). Using MetPA, the enrichment of 20 pathways was significantly different between the two groups, including “Taurine and hypotaurine metabolism”, “Phynylalanine, tyrosine and tryptophan biosynthesis”, “Phenylalanine metabolism”, “Glycerophospholipid metabolism”, “Alanine, aspartate and glutamate metabolism” and “Primary bile acid biosynthesis” (FDR < 0.01, pathway impact > 0.1) (Fig. 7B).
A The Euclidean distance matrix for the quantitative values of differential metabolites between grazing and intensive yaks was calculated, and the abundances were changed to logscale for significant metabolites visualization of the radar chart. B Pathway enrichment analysis was performed using the significantly different rumen metabolites between Grazing and Intensive groups.
Relationships between the rumen microbiome, rumen metabolome and plasma metabolome revealing the mechanism of the intensive feeding system to affect yaks
A comprehensive analysis of the rumen microbiome, metabolome and plasma metabolome could reveal how the intensive feeding system affects the microbiome and metabolism of yaks, resulting in an understanding of the overview of the mechanism of the intensive feeding system. Network analysis was performed to determine their associations (Fig. 8). We found that rumen bacteria were tightly associated with rumen metabolites, although other species also contributed to rumen metabolome. The critical microbiotas for each group were associated with their specific metabolites, which formed two modules with negative correlation (Fig. 8, Supplementary Fig. 13A). For example, Clostridium sp. as the signature microbiota for the intensive group was positively correlated with L-Glutamic acid, Pyroglutamic acid, Nitrite and N-Acetylputrescine that were abundant in the intensive group. Similarly, species, including Alistipes sp. and Bacteroide sp., positively correlated with L-Carnitine and Proline betaine abundant in the grazing group. Moreover, archaeal species showed similar patterns, such as Methanobrevibacter filiformis, Methanobrevibacter ccuticularis, M. profundi and M. psychrophilus. Next, similar relationships between rumen microbiome and plasma metabolites were found (Fig. 8, Supplementary Fig. 13B). Eukaryotes and viruses had more correlations with other microbial species and rumen metabolites. In addition, associations between rumen and plasma metabolites existed significant correlations were found, and signature metabolites for the intensive and grazing groups shaped two subunits, respectively (Fig. 8, Supplementary Fig. 13C). Notably, shared metabolites (L-Carnitine and Proline betaine) with higher abundances in both rumen and plasma metabolomics from grazing yaks were positively correlated (Supplementary Fig. 14). Moreover, the common predicted KEGG pathways of rumen and plasma metabolites were observed, including “Alanine aspartate and glutamate metabolism”, “Histidine metabolism”, “Arginine and proline metabolism”, and “D-Glutamine and D-glutamate metabolism” (Supplementary Fig. 15). Thus, the rumen microbiome and metabolites interacted with each other and influenced the host plasma metabolism.
Spearman’s rank correlations were calculated, and only strong (Spearman R of > 0.7 or < − 0.7) and significant (P < 0.05) correlations were shown.
Discussion
With increasing demands of animal-based products, intensive feeding practice for finishing cattle, especially beef cattle, has been widely adopted9,22. However, yaks, the vital livestock species of the Qinghai-Tibetan Plateau, are still reared in a wild or semi-wild program, resulting in a lower growth rate and longer production cycle. To address this question, based on traditional feeding practices (grazing pasture yearly), this study house-reared finishing yaks under an intensive feeding system via giving a designed high-grain diet. A significant increase in growth performance of intensively fed yaks was observed as high grain diets with greater protein and starch were consumed. Serum biochemical and physiological parameters were also altered. Based on the rumen metagenome, the community composition of the microbiome and its functions changed by the intensive feeding system. In the meantime, we observed that the changes in rumen fermentation mode and metabolome were correlated with microbial alterations. Moreover, signature plasma metabolites for the intensive group were characterized. Additionally, identifying complex relationships among rumen microbiome, rumen and plasma metabolites provided insight how intensive feeding contributed to the host physiology and metabolism, which benefits into yak production.
The positive effects of the intensive feeding system on the growth performance of yaks were not surprising due to their diet composition with high crude protein and starch but less fibers. Similar results were also reported in yaks3, beef cattle22, sheep23 and goat5. The changes in the serum biochemical parameters reflect the high nutrient utilization efficiency in the intensive group. Notably, for yaks, serum physiological parameters, such as red blood cell count and hemoglobin, associated with high altitude and lower pressure of oxygen were altered. Red blood and hemoglobin were more significant in grazing yaks, which is because grazing yaks exercise more during grazing. On the other hand, intensive feeding systems changed the oxygen delivery model of yaks. Unfortunately, it is unknown whether this condition is suitable for yaks living in high-altitude areas or not. Long-term observation of this aspect should be investigated in future studies.
Rumen microbiome plays a critical role in nutrient utilization, methane production, host growth and food security24,25. Most previous studies investigated yak rumen microbiota or intensive feeding system using 16S rRNA sequencing26,27,28, which restricts our understanding of low abundance bacteria and other microbial species. Based on metagenomics, this study found that the intensive feeding system significantly altered the rumen microbiome, including bacteria, archaea, eukaryotes, and viruses. This might be mostly due to the nutritional factors (proteins and carbohydrate) of the diet of the intensive feeding system29. A previous study found that crude protein level in the supplementary diet resulted in the alteration of rumen bacteria and fermentation in Jersey-yak11, which reflected the effects of crude protein levels in the current study. Our previous research also confirmed that additional high-grain diet supplementation could influence the rumen microbiota5. Bacteria as critical players in diet degradation and fermentation was the major kingdom influenced by the intensive feeding system, which is similar to previous studies30,31. Clostridium sp. involved in amino acid and carbohydrate metabolism were greater in intensive groups, indicating their ability to utilize protein and starch32. Prevotella ruminicola and brevis as well as Bacteroides sp. were more abundant in the rumen from yak grazing on pastures. This might be due to their preferred roles in fiber-based polysaccharides digest33,34. Fibrobacter sp. with cellulolytic properties were higher in grazing groups, which reveals the diet effects further35. Furthermore, high abundance of Bacteroidetes was found in yak rumen community. A previous study observed that grazing yaks had higher abundance of Bacteroidetes than that of feedlot Simmental cattle20. This noted that yak rumen microbiome was different and had its specific characteristics compared to other ruminants. In the meantime, shared rumen microbial species among ruminants reflected their commonalities. Additionally, for archaea, Methanobrevibacter was the only dominant genus in intensive group, and its major species members included M. curvatus, filiformis, and cuticularis. However, the grazing group had too many other archaeal species. The distributions of these methanogenic species suggest that high-grain diet may be associated with slight decreases in methane metabolism and emissions via reducing the number of archaeal species36. Although the abundance and functions of eukaryotes and viruses in the rumen are relatively small and largely unknown, the current study did observe several species were affected by the intensive feeding, such as Neocallimastix californiae and Geobacillus virus E3. Recent studies revealed that eukaryotic and viral members are associated with carbohydrate enzymes or feed digestibility37,38, which is also found in this study. Another study reported that Neocallimastix californiae co-cultured with Clostridium acetobutylicum benefited for butyrate production from cellulose39. All these results of microbiome changes corresponded to the consumption of high grain diet. Further studies should be conducted to deeply explore functions and associations of eukaryotes and viruses with bacteria.
Corresponding to the changes in the rumen microbiome, we observed that the intensive feeding system greatly impacted carbohydrate metabolism. Several pathways enriched in the intensive group based on rumen metagenomic and metabolomics, including “Glycolysis”, “Pentose phosphate pathway”, and “Pyruvate metabolism”, degrade and convert starch into pyruvate and VFAs, revealing that high starch intake altered the yak rumen fermentation and microbiome. Over-representation of starch-related CAZyme (e.g., branching enzyme (EC2.4.1.18), alpha-amylase (EC3.2.1.1) and cyclomaltodextrinase (EC3.2.1.54)) also confirmed this result. Moreover, high abundance of L-Carnitine from the rumen and plasma metabolomics reflected the improvement of carbohydrate metabolism in intensively fed yaks since L-Carnitine was correlated with fat synthesis and accumulation40.
Interestingly, for the grazing yaks, the KEGG pathway of its rumen microbiome contained more signal transduction, and more fiber degradation-related enzymes were found. This study was conducted in winter, and grazing yaks responded to insufficient supply and high fiber proportion of withered and yellow forage41,42. Thus, higher VFA concentrations found in yaks under the intensive feeding system revealed that dietary starch consumption influenced rumen microbiome and metabolism, which finally benefits yak growth.
Diet protein is another essential and critical nutrient for animals. The emission of nitrogenous substances (e.g., ammonia) is related to environmental challenges and feed conversion efficiency43. Thus, the rumen microbial metabolic and metabolomic pathways involved in amino acid degradation and biosynthesis were deeply investigated in this study. For rumen metagenomics, many previous studies focused on carbohydrate metabolism and paid less attention to amino acid metabolism11,31. This study observed consistent changes associated with amino acid metabolism based on rumen metagenomics and metabolomics. For instance, the intensive feeding system increased the KEGG pathways associated with amino acid metabolism, such as “Alanine aspartate and glutamate metabolism”, “Cysteine and methionine metabolism” and “Histidine metabolism”, which indicated that yaks under the intensive feeding system might have a higher efficient of protein utilization Our results also found that the signature microbiomes for the intensive group were correlated with “Alanine aspartate and glutamate metabolism”. The microbiome correlated with amino acid metabolism was also reported in a previous study44. Clostridium sp. as the intensive group classifier might be hydrogen acceptors in amino acid fermentation45. Moreover, the hydrogenotrophic species of Methanobrevibacter with higher abundances in the intensive group may branch amino acids into nitrogen via mutualistic cooperation with bacteria (e.g., Prevotella)46. Other eukaryotic species, including Piromyces sp. E2 and Anaeromyces robustus, were associated with amino acid and carbohydrate degradation39,47,48. Additionally, amino acid metabolism is tightly related to carbohydrates as carbon skeletons derived from deamination could yield VFAs49. Similarly, our results found that the microbiomes associated with “Alanine aspartate and glutamate metabolism” were also related to “Pyruvate metabolism”. However, the cellulolytic microbes (Bacteroides sp. and Fibrobacter sp.) that do not produce hydrogen negatively interacted with amino acid metabolism, which might be due to competition relationships between cellulolytic microorganisms and amino acid producers50. Shared Proline betaine of rumen and plasma metabolomics was higher in grazing yaks, indicating that amino acid metabolism in the rumen is associated with host metabolism. Therefore, hydrogen related microbiotas are critical to rumen microbial amino acid biosynthesis and metabolism, and the intensive feeding system in yaks could promote their amino acid metabolism in the rumen and plasma.
Ruminal methane production is defined as energy loss and related to greenhouse effect. Ruminant researchers found that diet or feeding strategy is associated with reduction of methane emissions. A previous report found greater CH4 production for cattle consuming high-fiber diets than for cattle fed high-grain diets36. In the current study, the intensive feeding system might benefit methane emission reduction in yaks compared to grazing feeding system since more high-grain diet was consumed in yaks. In this study, only three Methanobrevibacter species had higher relative abundances in the rumen of yaks of the intensive group, and abundant “Methane metabolism” of rumen microbiome was also found. However, the rumen of yaks in grazing groups had more archaeal species with relative high abundances. Two major routes for methanogenesis in the rumen include the hydrogenotrophic pathway and methyl groups. Many archaeal, eukaryotic and bacterial species producing hydrogen contribute to methane production, such as Methanobrevibacter sp. and fibrolytic bacteria51. Moreover, methyl groups contained methylamines derived from proline betaine that were abundant in the rumen and plasma metabolomics of grazing yaks. Thus, the intensive feeding system for yaks might improve its methane emissions via manipulation of the rumen hydrogenic microbiome and methyl metabolites. Future investigations should measure the methane emission of the yaks to support this speculation.
In ruminants, although previous studies have started to investigate the relationship between rumen microbiome, rumen metabolome and blood metabolome28,31,52, the detailed information how yaks responded the intensive feeding system is still unclear. A recent study investigated the effects of high-grain diet on the 16S rRNA of the rumen bacteria and their metabolome in adult yaks and found it altered the microbiota and metabolites28. Another study found that house-feeding altered the profile of rumen bacteria based on 16S rRNA sequencing and serum metabolome in Jersey cows27. The house-fattening feeding system affecting rumen bacteria and metabolites in yaks was estimated52,53. In our study, rumen metagenome was determined to excavate more microbial species and their functions, and the associations between multi-omics data and the intensive feeding system were more deeply analyzed. Although individual variation was observed, our results profoundly illustrated the mechanisms of how intensive feeding systems affected the rumen microbiome and metabolites and subsequently host metabolites to improve yak growth performance. Archaea, eukaryotes, and viruses in yak rumen were incorporated into analyses, which unveils their specific roles in nutrient metabolism and interactions with rumen bacteria. One of shortness of this study might be lack of validation trial of signature microbiotas, which will be performed in our future research. Overall, the intensive feeding system influenced multiple aspects of yaks in our results and the intensive feeding system benefits the yak industry.
Our study confirmed that the intensive feeding system significantly improved growth performance, rumen fermentation and serum parameters compared to traditional grazing practices in yaks. Based on multi-omics results, we identified the signature features of the rumen microbiome, microbial functions, metabolites, and their relationships with plasma metabolome which benefits yak growth. High grain diet led to increases of rumen Clostridium sp., Methanobrevibacter sp. (M. curvatus, M. filiformis, M. ccuticularis) and Piromyces sp. E2 that contributed to amino acid and carbohydrate metabolism but decreased methane production. The rumen of grazing yaks with high fibrolytic microorganisms and methylamines might be the major reason for high methane production and lower feed utilization efficiency. The levels of small molecules and their end products in the rumen were absorbed and transported by the host. In summary, the intensive feeding system may improve yak growth and methane emission via rumen microbiome and host metabolites. This study provides evidence that the intensive feeding system could be adopted for the yak industry with increased production efficiency and decreased greenhouse gas emissions.
Methods
Animals and sampling
The animal trial was conducted at the Tiancheng Lun Zhu Agricultural Products Development Co., Ltd. located in the north of Shangri-La County, Yunnan Province, China. The farm belonging to the subtropical and cold temperate monsoon climate was at an altitude of approximately 3600 meters.
A total of 50 healthy Zhongdian yaks aged 3 to 4 years old were selected and divided into two treatments. One treatment group was traditionally grazed all day on the natural ranch (Grazing group), while the other group was reared under an intensive feeding system (Intensive group) that housed and fed a total mixed ration diet to yaks. The trial lasted 60 days. At the end of the trial, fourteen yaks were randomly selected from two treatments to collect samples for downstream analysis (e.g., blood parameters, rumen fermentation parameters and multi-omics), with seven bulls in each group. The ingredients and the chemical composition of the diets are shown in Supplementary Table 1. A total mixed ration comprised whole plant corn silage, beef cattle concentrate and crushed corn. The grassland where yaks grazed contained forage including Blysmus sinocompressus, Carex tristachya, Potentilla fulgens, Spiranthessinensis, Poa pratensis, and Euphorbia jolkinii. The treatment yaks were fed twice at 9:00 a.m. and 4:00 p.m. every day, and the diet intake was recorded daily. The yak’s body weight was measured before early feeding at the beginning and end of the trial using fifty yaks, and average daily gain (ADG) was calculated. Specifically, yaks grazing on the natural pasture were driven to a holding pen at the night for the next day sampling, which guaranteed the same sampling time of two treatment groups. All animals used in the current study were healthy and did not receive any recorded therapeutic or prophylactic antibiotic treatments.
An oral stomach tube was used for the collection of rumen content samples. In brief, at 8 am in the morning, a tube was inserted into the rumen, and a vacuum sampler was used to pump rumen fluid. To avoid saliva contamination, the first 30 mL of rumen digesta sample was discarded. For each animal, 30 ml of rumen fluid was collected, and divided into three parts by placing into 10 ml polypropylene tubes. Then, all samples were snap-frozen in liquid nitrogen, and brought back to the laboratory for long-term storage at −80 °C. The jugular blood samples were also collected at the end of the trial. Before feeding in the morning, blood samples were collected into three 10 mL tubes. One tube without anticoagulant was rested for 1 h and centrifuged (3000 × g, 10 min, 4 °C), and then the aliquots of serum were frozen immediately at −20 °C until later chemical parameters analysis. Another tube containing heparin sodium anticoagulant was used for blood physiological parameter measurement, whereas the remained blood tube containing heparin sodium anticoagulant was centrifuged (3000 × g, 10 min, room temperature), and then the blood plasma was collected and stored at −80 °C for metabolomics.
Chemical analysis
Diet nutrient measurement: The samples of forage and total mixed ration were ground to pass through a 1-mm sieve and dried in an oven at 105 °C for 2 h to measure the dry matter (DM) content. The ash content, neutral detergent fiber (NDF), acid detergent fiber (ADF), calcium, and total phosphorus were determined according to previous methods54. Total nitrogen of diet samples was measured by an Auto‐Analyzer after Kjeldahl digestion and crude protein (CP) of the diet was calculated as 6.25 × N. Ether extract (EE) was determined using a reflux system (ANKOM XT15, America) with petroleum ether at 90 °C for 1 h. Starch was measured by the enzymatic method, involving incubation of the autoclaved sample with amyloglucosidase and spectrophotometric measurement of released glucose.
VFAs determination: The rumen fluid samples were thawed at 4 °C and then centrifuged at 2500 × g at room temperature. Next, 1 ml of the supernatant per sample was transferred into a 1.5 ml centrifuge tube which contained 0.2 ml of metaphosphoric acid solution (25% w/v). Then, the mixture was centrifuged at 10,000 × g at 4 °C after it was placed in an ice water set at 4 °C for 30 min. The collected supernatant was transferred to a solution containing an internal standard (2-Ethylbutyric acid, 2 EB) and a concentration of 25% metaphosphoric acid. The VFA concentration was detected using gas chromatography (GC–6800, Beijing Beifen Tianpu Instrument Technology, Co., Ltd., China).
Serum parameters analyses: The serum analysis was performed at Beijing Jinhai Keyu Biotechnology Development Co., Ltd. (Beijing, China). The concentration of serum biochemical parameters of alkaline phosphatase (ALP), total bilirubin (TBil), Direct Bilirubin (DBil), Indirect Bilirubin (IBil), Total Protein (TP), albumin (ALB), globulin (GLB), creatinine (Cr); triglyceride (TG) and total cholesterol (TC) was determined using A Kehua ZY KHB-1280 automatic biochemical analyzer (Shanghai Kehua Biological Engineering Co., Ltd., Shanghai, China) according to the manufacturer’s instructions of the standard commercial kits. Blood physiological indicators, including Red Blood Cell (RBC) Count, Hemoglobin (HGB), hematocrit (HCT), Mean corpuscular volume (MCV), mean corpuscular hemoglobin content (MCH), mean corpuscular hemoglobin concentration (MCHC), coefficient of variation (RDW), platelet number (PLT), mean platelet volume (MPV) and platelet distribution width (PDW), were measured using automatic blood analyzer (BC-2800Vet, Tokyo, Japan).
Shotgun metagenomic sequencing of rumen microbiome
Rumen samples were subjected to metagenomic sequencing. All experimental procedures were executed in a stringent clean and sterile class II B2 biosafety chamber to lower the contamination. Total genomic DNA was extracted using a DNeasy PowerSoil Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Negative controls during extraction were included. DNA integrity was estimated using 1% agarose gel electrophoresis. A Qubit 2.0 Fluorimeter (Invitrogen, Carlsbad, California, USA) was employed to measure the DNA concentration of samples. The metagenomic library was built using a TruSeq DNA PCR-Free Library Preparation Kit (Illumina, San Diego, California, USA). Metagenomic sequencing was conducted using an Illumina HiSeq 4000 platform with 150 bp paired end reads at the Beijing Headquarters of Novogene Biotech Co., Ltd (Beijing, China).
The metagenomic raw data obtained from the Illumina HiSeq sequencing platform were processed using Trimmomatic (version 0.33) to remove low-quality base with quality scores < 20, short reads (< 40 bp) and “N” base. By aligning to the yak genome (BosGru3.1), host contaminations were removed using Bowtie software (version 2.2.4) to filter the reads that were of host origin, with the setting of –end-to-end, –sensitive, -I 200, -X 40055. Next, the clean data were assembled for each sample using SOAPdenovo software (V2.04)56, and the parameters were as follows: -d 1, -M 3, -R, -u, -F, -K 55. MetaGene (http://metagene.cb.k.u-tokyo.ac.jp/) was used to predict open reading frames (ORFs) from the assembled contigs (length >= 500 bp)57. Assembled contigs were then pooled to remove non-redundancies using CD-HIT with 95% identity and 90% coverage for the selection of the representative sequences58.
Rumen metagenomic bioinformatics
Regarding taxonomic prediction of rumen microbiome, DIAMOND software59 (V0.9.9) was used to blast the contigs to the sequences of bacteria, fungi, archaea and viruses which were all extracted from the nr database (Version: 2023-07-28) of NCBI with the parameter setting of blastp -e 1e-5. Rumen microbiome taxonomic profiles were analyzed at the domain, phylum, genus and species levels to identify the changes of microbiome composition. Beta diversity based on Bray-Curtis distances was visualized on PCoA plots. Microbial taxa with a relative abundance > 0.1% in at least 50% of yaks within each group were used for downstream analysis.
For functional analysis, DIAMOND was used to annotate non-redundant genes against the KEGG (Kyoto Encyclopedia of Genes and Genomes) database with an E value of 1e-560. Abundances of KEGG Orthology (KO), pathways, KEGG enzymes, and Modules were normalized for the subsequent statistics. Moreover, the Carbohydrate-Active EnZymes database (CAZy)61 were annotated by aligning genes to dbCAN database (HMMdb V8) with hmmscan program in HMMER (v3.1b2). The abundance values of the CAZymes were assigned by mapping the clean reads from each sample to the corresponding unigenes obtained after CD-HIT clustering. For all functional annotations, the annotated hit(s) with the highest score was used for the subsequent analyses.
Determination and analysis of rumen and plasma metabolome
The rumen and plasma metabolome were analyzed using LC-MS/MS system combined quadrupole Orbitrap mass spectrometer (Q Exactive Orbitrap, Thermo Fisher Scientific, USA)62. In detail, 100 μL of the sample was transferred to an EP tube. After the addition of 400 μL of extract solution (acetonitrile/methanol = 1; containing isotopically labeled internal standard mixture), the tube was vortexed for 30 s, sonicated for 5 min in an ice-water bath, and incubated for 1 h at -40 °C to precipitate proteins. Then, the sample was centrifuged at 12000 rpm for 15 min at 4 °C. The resulting supernatant was transferred to a fresh glass vial for downstream analysis. The quality control (QC) sample was prepared by mixing an equal aliquot of the supernatants from all the samples. The mobile phase consisted of 25 mmol/L ammonium acetate and 25 ammonia hydroxides in water (pH = 9.75) (A) and acetonitrile (B). The analysis was carried with elution gradient as follows: 0 ~ 0.5 min, 95% B; 0.5 ~ 7.0 min, 95% ~ 65% B; 7.0 ~ 8.0 min, 65% ~ 40% B; 8.0 ~ 9.0 min, 40% B; 9.0 ~ 9.1 min, 40% ~ 95% B; 9.1 ~ 12.0 min, 95% B. The column temperature was 25 °C. The auto-sampler temperature was 4 °C, and the injection volume was 2 μL. The QE HFX mass spectrometer was used for its ability to acquire MS/MS spectra on information-dependent acquisition (IDA) mode in the control of the acquisition software (Xcalibur, Thermo). The acquisition software continuously evaluates the full scan MS spectrum in this mode. The ESI source conditions were set as follows: sheath gas flow rate as 50 Arb, Aux gas flow rate as 10 Arb, capillary temperature 320 °C, full MS resolution as 60000, MS/MS resolution as 7500, collision energy as 10/30/60 in NCE mode, spray Voltage as 3.5 kV (positive) or -3.2 kV (negative), respectively.
The raw data of metabolomics was converted to the mzXML format using ProteoWizard and processed with an in-house program, which was developed using R and based on XCMS, for peak detection, extraction, alignment, and integration. Then, an in-house MS2 database (BiotreeDB) was applied to metabolite annotation. The cutoff for annotation was set at 0.3. In total, 263 rumen metabolites and 177 serum metabolites were identified and transformed to have a zero mean and a unit variance for downstream analysis. An online platform, MetaboAnalyst 5.063, was used for the metabolomic analysis based on targeted metabolites utilizing the library of Bos Taurus (cow)64. Metabolite set enrichment analysis (MESA) was performed using MetaboAnalyst 5.0, based on the pathway-associated metabolite sets library65.
Statistics
Growth performance, serum biochemical and physiological parameters, and rumen VFAs concentrations were compared using t test in R (v4.2.0) “stats” package. Rumen microbiome at the domain, phylum, and genus levels were compared using the Wilcoxon rank-sum test, with the Benjamini–Hochberg FDR adjusted P value < 0.05 being considered as significantly different. The dissimilarity of bacterial, archaeal, eukaryotic and viral composition was tested using Analysis Of Similarity (ANOSIM) performed on the R “vegan” package. The linear discriminant analysis effect size (LEfSe) algorithm66 based on an online Galaxy server was used to detect the features differentiating grazing and intensive groups, such as rumen microbial species, KEGG pathways and modules, and CAZyme families. The cutoff for significant differences of LEfSe was LDA score > 2 and P value < 0.05.
MetaboAnalyst 5.0 was employed to perform the statistical analysis for both rumen and plasma metabolomic data. The orthogonal projections to latent structures- discriminant analysis (OPLS-DA). The t test was performed between the two groups, with the FDR adjusted P value of Benjamini–Hochberg and the Variable Importance in the Projection (VIP) > 1 being considered as significantly different metabolites. Additionally, ‘pheatmap’ package in R was used to display the signature metabolites in rumen and plasma differentiating two groups. The analysis of traceability and enrichment of rumen metabolites was performed using MetOrigin67.
Spearman’s analysis was performed to calculate the correlations between signature microbiota and critical metabolites, using ‘psych’ package in R, and the network was visualized using the Cytoscape (version 3.10.1). Only significant connections were shown in the network (|r | > 0.7, P < 0.05).
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Data availability
Sequencing data generated from shotgun metagenomes in this study have been deposited with the NCBI SRA (PRJNA795434) and are publicly available.
References
Qiu, Q. et al. Yak whole-genome resequencing reveals domestication signatures and prehistoric population expansions. Nat. Commun. 6, 10283 (2015).
Ge, R. et al. Draft genome sequence of the Tibetan antelope. Nat. Commun. 4, 1858 (2013).
Zhao, C. et al. Yak rumen microbiome elevates fiber degradation ability and alters rumen fermentation pattern to increase feed efficiency. Anim. Nutr. 11, 201–214 (2022).
Guo, W. et al. Survey of rumen microbiota of domestic grazing yak during different growth stages revealed novel maturation patterns of four key microbial groups and their dynamic interactions. Anim. Microbiome 2, 1–20 (2020).
Chai, J. et al. Solid diet manipulates rumen epithelial microbiota and its interactions with host transcriptomic in young ruminants. Environ. Microbiol 23, 6557–6568 (2021).
Zhou, J. et al. Comparison of nitrogen utilization and urea kinetics between yaks (Bos grunniens) and indigenous cattle (Bos taurus). J. Anim. Sci. 95, 4600–4612 (2017).
Zhou, J. et al. Apparent digestibility, rumen fermentation, digestive enzymes and urinary purine derivatives in yaks and Qaidam cattle offered forage-concentrate diets differing in nitrogen concentration. Livest. Sci. 208, 14–21 (2018).
Du, M. et al. Dietary energy levels affect carbohydrate metabolism-related bacteria and improve meat quality in the longissimus thoracis muscle of yak (Bos grunniens). Front Vet. Sci. 8, 718036 (2021).
Moorby, J. & Fraser, M. New feeds and new feeding systems in intensive and semi-intensive forage-fed ruminant livestock systems. Animal 15, 100297 (2021).
Dai, D. et al. Effects of concentrate supplementation on growth performance, rumen fermentation, and bacterial community composition in grazing yaks during the warm season. Animals 12, 1398 (2022).
Dai, R. et al. Effects of dietary crude protein levels in the concentrate supplement after grazing on rumen microbiota and metabolites by using metagenomics and metabolomics in Jersey-yak. Front Microbiol 14, 1124917 (2023).
Liu, H. et al. Effects of supplementary concentrate and/or rumen-protected lysine plus methionine on productive performance, milk composition, rumen fermentation, and bacterial population in Grazing, Lactating Yaks. Anim. Feed Sci. Tech. 297, 115591 (2023).
Ma, J. et al. Glutamine supplementation affected the gut bacterial community and fermentation leading to improved nutrient digestibility in growth-retarded yaks. FEMS Microbiol Ecol. 97, fiab084 (2021).
Zhang, Y. et al. Influences of dietary nitrogen and non-fiber carbohydrate levels on apparent digestibility, rumen fermentation and nitrogen utilization in growing yaks fed low quality forage based-diet. Livest. Sci. 147, 139–147 (2012).
Xiong, L. et al. Lipidomics and transcriptome reveal the effects of feeding systems on fatty acids in yak’s meat. J. Foods 11, 2582 (2022).
Zhu, M. et al. Milk production and quality of lactating yak fed oat silage prepared with a low-temperature-tolerant lactic acid bacteria inoculant. Foods 10, 2437 (2021).
Liu, Y. et al. Effects of intensive fattening with total mixed rations on carcass characteristics, meat quality, and meat chemical composition of yak and mechanism based on serum and transcriptomic profiles. Front Vet. Sci. 7, 599418 (2021).
Lv, X. et al. The signature microbiota drive rumen function shifts in goat kids introduced to solid diet regimes. Microorganisms 7, 516 (2019).
Godfray, H. C. et al. Food security: the challenge of feeding 9 billion people. Science 327, 812–818 (2010).
Hu, C. et al. Effects of management, dietary intake, and genotype on rumen morphology, fermentation, and microbiota, and on meat quality in yaks and cattle. Front Nutr. 8, 755255 (2021).
Zhang, Q. et al. Rumen microbiome reveals the differential response of co2 and ch4 emissions of yaks to feeding regimes on the Qinghai-Tibet plateau. Animals 12, 2991 (2022).
Costantini, M., Vázquez-Rowe, I., Manzardo, A. & Bacenetti, J. Environmental impact assessment of beef cattle production in semi-intensive systems in Paraguay. Sustain Prod. Consum 27, 269–281 (2021).
Sultana, N., Hassan, N., Ershaduzzaman, M., Talukder, M. & Iqbal, A. Effect of intensive and semi-intensive feeding system on productive and reproductive performances of native sheep. J. Sci. Res 3, 693–698 (2011).
Mizrahi, I., Wallace, R. J. & Moraïs, S. The rumen microbiome: balancing food security and environmental impacts. Nat. Rev. Microbiol 19, 553–566 (2021).
Zhao, W., Abdelsattar, M. M., Wang, X., Zhang, N. & Chai, J. In vitro modulation of rumen fermentation by microbiota from the recombination of rumen fluid and solid phases. Microbiol Spectr. 11, e03387–03322 (2023).
Pang, K. et al. Ruminal microbiota and muscle metabolome characteristics of Tibetan plateau yaks fed different dietary protein levels. Front Microbiol 15, 1275865 (2024).
Yuan, N. et al. From the perspective of rumen microbiome and host metabolome, revealing the effects of feeding strategies on Jersey Cows on the Tibetan Plateau. PeerJ 11, e16010 (2023).
Dai, D. et al. High-grain diet feeding altered blood metabolites, rumen microbiome, and metabolomics of yaks. Fermentation 9, 215 (2023).
Chai, J., Zhuang, Y., Cui, K., Bi, Y. & Zhang, N. Metagenomics reveals the temporal dynamics of the rumen resistome and microbiome in goat kids. Microbiome 12, 14 (2024).
Weimer, P. J. Symposium review: Host-rumen microbe interactions may be leveraged to improve the productivity of dairy cows. J. Dairy Sci. 101, 7680–7689 (2018).
Xue, M. Y., Sun, H. Z., Wu, X. H., Liu, J. X. & Guan, L. L. Multi-omics reveals that the rumen microbiome and its metabolome together with the host metabolome contribute to individualized dairy cow performance. Microbiome 8, 020–00819 (2020).
Hu, D. et al. Metagenomic analysis of fecal archaea, bacteria, eukaryota, and virus in przewalski’s horses following anthelmintic treatment. Front Vet. Sci. 8, 708512 (2021).
Guo, T., Wang, Z. L., Guo, L., Li, F. & Li, F. Effects of supplementation of nonforage fiber source in diets with different starch levels on growth performance, rumen fermentation, nutrient digestion, and microbial flora of Hu lambs. Transl Anim Sci. 5, txab065 (2021).
Xue, M.-Y. et al. Investigation of fiber utilization in the rumen of dairy cows based on metagenome-assembled genomes and single-cell RNA sequencing. Microbiome 10, 11 (2022).
Fakih, I. et al. Dynamic genome-based metabolic modeling of the predominant cellulolytic rumen bacterium Fibrobacter succinogenes S85. Msystems 8, e01027–01022 (2023).
Harper, L. A., Denmead, O. T., Freney, J. R. & Byers, F. M. Direct measurements of methane emissions from grazing and feedlot cattle. J. Anim. Sci. 77, 1392–1401 (1999).
Yan, M. et al. Interrogating the viral dark matter of the rumen ecosystem with a global virome database. Nat. Commun. 14, 5254 (2023).
Li, Z. et al. Genomic insights into the phylogeny and biomass-degrading enzymes of rumen ciliates. ISME J. 16, 2775–2787 (2022).
Brown, J. L. et al. Co-cultivation of anaerobic fungi with Clostridium acetobutylicum bolsters butyrate and butanol production from cellulose and lignocellulose. J. Ind. Microbiol. Biotechnol. 49, kuac024 (2023).
Yin, J. et al. Obese Ningxiang pig-derived microbiota rewires carnitine metabolism to promote muscle fatty acid deposition in lean DLY pigs. Innovation 4, 100486 (2023).
Fu, H. et al. Domestication Shapes the Community Structure and Functional Metagenomic Content of the Yak Fecal Microbiota. Front Microbiol 12, 594075 (2021).
Gharechahi, J. et al. Metagenomic analysis reveals a dynamic microbiome with diversified adaptive functions to utilize high lignocellulosic forages in the cattle rumen. Isme j. 15, 1108–1120 (2021).
Tan, P. et al. Amino acids metabolism by rumen microorganisms: Nutrition and ecology strategies to reduce nitrogen emissions from the inside to the outside. Sci. Total Environ. 800, 149596 (2021).
Deng, F. et al. The unique gut microbiome of giant pandas involved in protein metabolism contributes to the host’s dietary adaption to bamboo. Microbiome 11, 180 (2023).
Lewis, T. & Emery, R. Intermediate products in the catabolism of amino acids by rumen microorganisms. J. Dairy Sci. 45, 1363–1368 (1962).
Forwood, D. L. et al. Feeding an unsalable carrot total-mixed ration altered bacterial amino acid degradation in the rumen of lambs. Sci. Rep. 13, 6942 (2023).
Steenbakkers, P. J. et al. A serpin in the cellulosome of the anaerobic fungus Piromyces sp. strain E2. Mycol. Res 112, 999–1006 (2008).
Hagen, L. H. et al. Proteome specialization of anaerobic fungi during ruminal degradation of recalcitrant plant fiber. ISME J. 15, 421–434 (2021).
Bento, C. B. P., de Azevedo, A. C., Detmann, E. & Mantovani, H. C. Biochemical and genetic diversity of carbohydrate-fermenting and obligate amino acid-fermenting hyper-ammonia-producing bacteria from Nellore steers fed tropical forages and supplemented with casein. BMC Microbiol 15, 28 (2015).
Hassan, F. et al. Effect of methionine supplementation on rumen microbiota, fermentation, and amino acid metabolism in in vitro cultures containing nitrate. Microorganisms 9, 1717 (2021).
Tapio, I., Snelling, T. J., Strozzi, F. & Wallace, R. J. The ruminal microbiome associated with methane emissions from ruminant livestock. J. Anim. Sci. Biotechnol. 8, 7 (2017).
Xu, C. et al. Multi-omics analysis reveals a dependent relationship between rumen bacteria and diet of grass- and grain-fed yaks. Front Microbiol 12, 642959 (2021).
Huang, C. et al. Microbiome and metabolomics reveal the effects of different feeding systems on the growth and ruminal development of yaks. Front Microbiol 12, 682989 (2021).
Van Soest, P. J., Robertson, J. B. & Lewis, B. A. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74, 3583–3597 (1991).
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Luo, R. et al. SOAPdenovo2: an empirically improved memory-efficient short-read de novo assembler. GigaScience 1, 18 (2012).
Noguchi, H., Park, J. & Takagi, T. MetaGene: prokaryotic gene finding from environmental genome shotgun sequences. Nucleic Acids Res 34, 5623–5630 (2006).
Fu, L., Niu, B., Zhu, Z., Wu, S. & Li, W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28, 3150–3152 (2012).
Buchfink, B., Xie, C. & Huson, D. H. Fast and sensitive protein alignment using DIAMOND. Nat. Methods 12, 59–60 (2015).
Kanehisa, M. & Goto, S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res 28, 27–30 (2000).
Lombard, V., Golaconda Ramulu, H., Drula, E., Coutinho, P. M. & Henrissat, B. The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res 42, D490–D495 (2014).
Wang, J. et al. Metabolomic profiling of anionic metabolites in head and neck cancer cells by capillary ion chromatography with Orbitrap mass spectrometry. Anal. Chem. 86, 5116–5124 (2014).
Chong, J. et al. MetaboAnalyst 4.0: towards more transparent and integrative metabolomics analysis. Nucleic Acids Res 46, W486–W494 (2018).
Xia, J. & Wishart, D. S. MetPA: a web-based metabolomics tool for pathway analysis and visualization. Bioinformatics 26, 2342–2344 (2010).
Xia, J. & Wishart, D. S. MSEA: a web-based tool to identify biologically meaningful patterns in quantitative metabolomic data. Nucleic Acids Res 38, W71–W77 (2010).
Segata, N. et al. Metagenomic biomarker discovery and explanation. Genome Biol. 12, R60 (2011).
Yu, G., Xu, C., Zhang, D., Ju, F. & Ni, Y. MetOrigin: Discriminating the origins of microbial metabolites for integrative analysis of the gut microbiome and metabolome. IMeta 1, e10 (2022).
Acknowledgements
We thank the farm for their cooperation in animal handling. This study was funded by the National Natural Science Foundation of China (32060762, 32260823), Yunnan Province agricultural basic research joint special grant (202301BD070001-095), and Key Research and Development Plan Project of the Tibet Autonomous Region (XZ202301ZY0008N).
Author information
Authors and Affiliations
Contributions
S.Y. and D.W. contributed to animal trials, sample collection. S.Y. and J.C. contributed to data analysis, visualization, primary drafting, and proof. J.Z. and H.M. helped with sample collection and investigation. P.V., D.W. and J.C. contributed to the experimental design, visualization, project management and draft proof. All authors have read and agreed to the published version of the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics approval
The current study was performed according to the guidance of the Animal Ethics Committee of the Yunnan Agricultural University (Protocol Number: 202009006).
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Yang, S., Zheng, J., Mao, H. et al. Multiomics of yaks reveals significant contribution of microbiome into host metabolism. npj Biofilms Microbiomes 10, 133 (2024). https://doi.org/10.1038/s41522-024-00609-2
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41522-024-00609-2
