Relationship between rumen microbial differences and traits among Hu sheep, Tan sheep, and Dorper sheep

Abstract

Rumen microbes play an important role in the growth and development of ruminants. Differences in variety will affect the rumen community structure. The three excellent sheep breeds were selected for this study (Hu sheep, Tan sheep, and Dorper sheep) have different uses and origins. The sheep were raised on the same diet to 180 d of age in a consistent environment. 16S rDNA V3 to V4 region sequencing was used to assess the rumen microbes of 180 individuals (60 per breed). There were differences in microbial diversity among different sheep breeds (P < 0.05). Principal coordinate analysis showed that the three varieties were separated, but also partially overlapped. Linear discriminant analysis effect size identified a total of 19 biomarkers in three breeds. Of these biomarkers, five in Hu sheep were significantly negatively correlated with average feed conversion rate (P < 0.05). Six biomarkers were identified in the rumen of Dorper sheep, among which Ruminococcus was significantly positively correlated with body weight at 80 d (P < 0.05). In Tan sheep, Rikenellaceae_RC9_gut_group was significantly positively correlated with meat fat, and significantly positively correlated with volatile fatty acids (VFAs), such as butyric acid and isobutyric acid (P < 0.05). The Rikenellaceae_RC9_gut_group may regulate Tan mutton fat deposition by affecting the concentration of VFAs. Functional prediction revealed enrichment differences of functional pathways among different sheep breeds were small. All were enriched in functions, such as fermentation and chemoheterotrophy. The results show that there are differences in the rumen microorganisms of the different sheep breeds, and that the microorganisms influence the host.

The relationship between trait differences and microbial differences among different sheep breeds was clarified.

Introduction

Sheep (Ovis aries) is one of the earliest domesticated livestock and one of the most important livestock breeds that is widely distributed globally. Differences in the natural environment and social needs in different places have driven the emergence of hundreds of sheep breeds with different characteristics of appearance and physiological habits after long-term selection (Deniskova et al., 2018). Especially in China, there are many excellent subspecies of sheep, including Mongolian sheep, Tibetan sheep, Kazakh sheep, and so on (Alberto et al., 2018).

As a unique digestive organ of ruminants, the rumen plays an important role in the process of digestion and can effectively utilize the indigestible diet, mainly due to the large number of microflora in the rumen (Akin and Borneman, 1990; Matthews et al., 2019). Rumen microbes have long maintained a mutually beneficial and symbiotic relationship with the host. Most studies have shown that changes in rumen microbes can affect host traits (Myer et al., 2015; Bickhart and Weimer, 2018). Rumen microbes decompose diets into volatile fatty acids (VFAs) and other nutrients through degradation and fermentation, which can provide 70% of the energy required by the host (Cholewińska et al., 2020). Bacteria and fungi are effective in manipulating rumen development and function, and can increase ruminants productivity by adjusting the relevant microorganisms (Kmet et al., 1993).

The rumen microorganisms of ruminants are mainly composed of Firmicutes and Bacteroidetes (Xie et al., 2021). However, there are differences in microbial composition among different species. Glendinning et al. (2021) described significant differences in microbial composition in the rumen of cattle, sheep, reindeer, and red deer. The composition of microorganisms is affected mainly by diet, environment, age, and other factors (Eynipour et al., 2019; Wu et al., 2020; Wang et al., 2021). Cholewińska et al. demonstrated significant differences in the microorganisms in the feces of different breeds of sheep (Russell and Hespell, 1981). Therefore, it is necessary to study the differences of rumen microorganisms among different varieties within the same species.

Hu sheep are mainly distributed in southern China and belong to the first-class protected local breeds. Their characteristics include early maturity, rapid estrus growth, and development in four seasons, and their most crucial advantage is the high lambing rate (Li et al., 2021). Tan sheep are distributed in northern China and, like the Hu sheep, were bred from Mongolian sheep. Its meat is tender and fat is evenly distributed, and it is an excellent local breed (Liu et al., 2021). Dorper sheep are native to South Africa and are now widely distributed around the world. They are fast growing, heavy weaning weight, and resistant to stress, and are often used for paternal crossbreeding (Yeaman et al., 2013; Tesema et al., 2020).

In this study, Hu, Tan, and Dorper sheep were used to study the relationship between sheep breeds and rumen microflora structure after eliminating other influencing factors under the same feeding environment and feeding conditions. The findings reveal the key microorganisms in different breeds and will inform the regulation of sheep rumen microorganisms.

Materials and Methods

Ethics approval

The study was carried out as per animal care and experiment procedures in accordance with the regulations and guidelines of the Government of Gansu People’s Congress. The program has been approved by the Animal Conservation and Ethics Committee of Gansu Agricultural University (License No. 2012-2-159).

Animals

The experimental sheep (60 male sheep of each breed) were purchased from modern commercial farms. Hu sheep were from Linqing Runlin Animal Husbandry Co., Ltd (Shandong Province, China), Tan sheep were from Ningxia Zhongmu Yilin Animal Husbandry Co., Ltd (Ningxia Province, China), and Dorper sheep were from Hongsibao District Tianyuan Sheep Breeding and Breeding Co., Ltd (Ningxia Province, China). All experimental sheep were immunized by standardized procedures and weaned at 56 d of age. All experimental sheep were immunized by a standardized procedure, weaned at 56 d of age, and each lamb was transferred to a separate pen for feeding. The lambs after weaning were in the 14-d adaptation period, during which the proportion of silage alfalfa gradually decreased, while the proportion of pellet feed was increased by 7.1% every day until the proportion of pellet feed reached 100%. The pellet feed composition was 27.00% barley straw, 44.00% corn, 2.20% soybean meal, 2.60% rapeseed meal, 4.20% cottonseed meal, and 20.00% concentrate, containing 16.28% crude protein, 28.48% starch, 36.54% neutral detergent fiber, 14.12% acid detergent fiber, 0.60% calcium, and 0.30% phosphorus. The pellet feed was purchased from Gansu Sanyangjinyuan Husbandry Co., Ltd (Gansu Province, China). All animals had free access to food and water. At 80 to 180 d of age a calibrated electronic scale was used to determine body height and body length and chest circumference were determined before feeding in the morning, and measured once every 20 d. During the experiment, the feeding method and the environment were kept the same.

Sample collection and determination

After the measurement period, 180 sheep of the three breeds were slaughtered by carotid bloodletting. The intact rumen of the test animals was obtained, rumen chyme samples were collected at the same location, and the chyme was temporarily stored in liquid nitrogen (−196 °C). After the test, it was transferred to −80 °C ultralow temperature refrigerator for storage. The longissimus dorsi muscle in the carcass was dissected, and meat samples were collected at the same location and stored at −20 °C.

The FoodScan 2 meat quality analyzer (Fosihua Science and Trade Co., Ltd, Beijing, China) was used to determine the content of components in the sample, including fat, moisture, salt, and protein. VFAs in rumen chyme were determined by gas chromatography (Agilent Technologies, Inc., Santa Clara, CA).

DNA extraction and amplification

Cetyltrimethylammonium bromide was used to extract DNA from the samples. Agarose gel electrophoresis was used to detect the purity and concentration of DNA. Samples were diluted to 1 ng/µL with sterile water. The remaining DNA samples were stored at −20 °C.

The extracted DNA was used as a template for PCR amplification. The primers were 341F (CCTAYGGGRBGCASCAG) and 806R (GGACTACNNGGGTATCTAAT). The amplified region was V3 to V4 region of microbial 16S ribosomal RNA (Sun et al., 2013). PCR used a 30-μL volume containing 15 μL Phusion High-Fidelity PCR Master Mix (New England Biolabs, Ipswich, MA), 0.2 μm upstream and downstream primers, 10 ng template DNA, and 2 μL sterile water. The cycling conditions were 1 min at 98 °C; 30 cycles of 10 s at 98 °C, 30 s at 50 °C, and 30 s at 72 °C; and finally, 5 min at 72 °C. The amplified products were detected by 2% agarose gel electrophoresis.

Library construction and data processing

The library was constructed by using the TruSeq DNA PCR-Free Library Preparation Kit (Illumina, San Diego, CA). The library quality was assessed on the Qubit @ 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA) and a Bioanalyzer 2100 system (Agilent, Santa Clara, CA). The library was sequenced on an Illumina NovaSeq platform and 250-bp paired-end reads were generated.

The sequenced DNA fragments were paired-end reads using FLASH (Version 1.2.7) (Magoč and Salzberg, 2011). Paired-end reads were assigned to each sample according to the unique barcodes to obtain Raw Tags and Clean Tags through strict quality control and filtration (Bokulich et al., 2013). The Clean Tags of all samples were clustered using the Uparse algorithm (Version 7.0.1001) (Haas et al., 2011). Sequences were clustered into operational taxonomic units (OTUs) with 97% consistency. The OTUs were annotated and compared and annotated with SILVA138 database by the MOTHUR method (Edgar, 2013). After the taxonomic information was obtained, the community composition of the sample at each classification level was determined (kingdom, phylum, class, order, family, genus, and species). MUSCLE (Version 3.8.31) software was used for fast multi-sequence alignment to obtain the phylogenetic relationship of OTUs (Quast et al., 2013). Finally, the data of each sample were homogenized. The sample with the least amount of data in the sample was used as the standard. Based on the homogenized data, Quantitative Insights Into Microbial Ecology (QIIME) software was used for data analysis. Its internal Perl script was used to analyze the alpha and beta diversity of samples. Intergroup differences were analyzed using IBM SPSS statistics 25 by rank sum test. UniFrac distance was calculated by QIIME software, principal coordinate analysis (PCoA) diagram was drawn by R software (Version 4.0.3). Linear discriminant analysis effect size (LEfSe) software was used to analyze the difference of flora between groups. The linear discriminant analysis (LDA) score was set to 4. R (Version 4.0.3) was used for correlation analysis based on Spearman coefficient.

The specific methods of functional prediction through tax involved extraction of the whole-genome 16S rRNA gene sequence from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, comparison to the SILVA database by the BLASTN algorithm (BLAST bitscore > 1,500), and establishment of the correlation matrix. The whole-genome functional information of prokaryotes in the KEGG database annotated by UProC and PAUDA was mapped to SILVA database to realize the functional annotation of the SILVA database. Samples were annotated through the SILVA database and functional annotation information is obtained. The above microbial functional analysis process was performed using Tax4Fun software.

Measurement of rumen VFAs

The rumen VFAs of sheep was measured by rumen chyme stored at −80 °C using gas chromatography (Erwin et al., 1961). The rumen juice was centrifuged at 5,400 rpm (centrifugal radius: 14.5 cm, relative centrifugal force: 4,731 g) for 10 min. One milliliter was mixed with 0.2 mL of phosphite mixture (2-ethylbutyric acid as the internal standard, 2 g/L), and incubated at 4 °C for 30 min. Following centrifugation at 4 °C for 10 min at 10,000 rpm (centrifugal radius: 5 cm, relative centrifugal force: 5,595 g), the supernatant was filtered by 0.45 µm and injected into a 2-mL gas chromatography bottle for subsequent determination. VFA values were determined using a TRACE-1300 series GC ultra-gas chromatograph (Thermo Fisher Scientific).

Results

Growth traits and meat quality of different varieties

A total of 180 sheep were used in the present cross-sectional study. The characteristics of the test population are shown in Supplementary Table S1. Figure 1A shows that the weights of the Dorper sheep were higher than the weights of Hu and Tan sheep, and that the weights of Hu sheep from 140 to 180 d were higher than the weights of Tan sheep (both P < 0.05). However, the feed conversion ratio (FCR) of Dorper sheep at 80 to 180 d was higher than that of Tan sheep and Hu sheep (P < 0.05) (Supplementary Table S1). Interestingly, the FCR of the three sheep breeds was the highest at the 160- to 180-d stage, indicating that at this stage the energy used for basal metabolism in sheep is higher and that the energy used for development begins to decrease (Figure 1B). The results of the meat nutrition analysis (Figure 1C–F) showed that the fat content of Tan mutton was higher than that of Hu and Dorper sheep (both P < 0.05). The protein and salt of Dorper mutton were higher than those of Hu and Tan sheep (both P < 0.05). The moisture content of Hu mutton was higher than that of Tan and Dorper sheep (both P < 0.05).

Figure 1.

Comparison of economic traits of the three different breeds of sheep. (A) The 80- to 180-d body weight of each variety. Different lowercase letters in the same day indicate significant differences among breeds (P < 0.05). (B) The 80- to 180-d feed conversion ratio (FCR) in each stage. (C–F) Meat parameters, including fat (C), protein (D), moisture (E), and salt (F).

Overview of sequencing data and diversity analysis

We successfully amplified 16S rDNA sequences from rumen digesta samples of three breeds of sheep. All 180 samples were sequenced and 18,020,435 Clean Tags were generated after splicing. By filtering low-quality sequences and chimeras, 11,666,817 effective tags were finally obtained for subsequent analysis. The average length of each sequence was 417 bp (Supplementary Table S2). These sequences were clustered into 7,898 OTUs by MOTHUR. Tan sheep had the most unique OTUs (2,489). There were 2,674 OTUs among the three breeds (Figure 2A). With increasing sequencing depth and sample size, the dilution curve and species accumulation curve tended to be flat (Supplementary Figure S1). These findings indicate that the test conditions met the analysis requirements.

Figure 2.

Rumen operational taxonomic units (OTUs) and diversity in sheep breeds. (A) UpSetR figure of rumen OTUs of various varieties. (B) Alpha diversity comparison of rumen microbes. (C) Principal coordinate analysis based on unweighted UniFrac distance.

To investigate the differences in community composition of the rumen microbes of the three breeds of sheep, we assessed the alpha diversity of the rumen microbiota (Figure 2B). The rumen microbial Shannon index of Hu sheep was higher than that of Tan sheep (P < 0.05), while that of Tan sheep was higher than that of Dorper sheep (P < 0.05). These findings suggest that Dorper sheep had the lowest rumen microbial diversity. There were no significant differences in the Chao1 index of rumen microbes among the three breeds (P > 0.05).

In assessing the phylogenetic relationship between OTUs, UniFrac distances were calculated. PCoA based on unweighted UniFrac distances revealed a separation among the three sheep breeds. However, there was also a partial overlap. To clarify the significance of the differences in the rumen microbial community among the three varieties, the Analysis of Similarities (ANOSIM) and Multi-Response Permutation Procedure (MRPP) methods were used to verify the results (Table 1). Significant differences were evident in the structure of the rumen microflora among the three breeds (P < 0.01).

Table 1.

Analysis of differences in community structure among breeds

Method		Hu–Tan	Hu–Dorper	Tan–Dorper
ANOSIM	R-value	0.41	0.33	0.27
	P-valuea	<0.01	<0.01	<0.01
MRPP	Delta	0.47	0.49	0.51
	P-valuea	<0.01	<0.01	<0.01

P < 0.05 indicates a statistically significant difference.

Composition of rumen microbes in the different breeds

The effective tags were annotated through the 16S SILVA database. Among them, 7,220 (91.42%) OTUs could be annotated to the database. The proportions were 83.68%, 83.36%, 81.12%, 72.13%, and 52.85% at the phylum, class, order, family, and genus level, respectively. The phylogenetic tree of genus level species showed that >50% of the genus level species belonged to Firmicutes in the annotated genus level species (Supplementary Figure S2). Annotation results of relative abundance TOP10 were analyzed (Supplementary Table S3). At the phylum level, the rumen composition of the three breeds was similar; Firmicutes and Bacteroidota were the dominant phyla (Figure 3A). At the genus level, the dominant genera were Prevotella, Saccharofermentans, and Treponema. The abundance of Prevotella in the rumen of Hu sheep was higher (24.79%) (Figure 3B). Species abundance clustering revealed that Hu sheep were highly clustered in Prevotella of Bacteroidota, Tan sheep in Methylorubrum of Proteobacteria, and Dorper sheep in Fibrobacter of Fibrobacterota (Figure 3C).

Figure 3.

Rumen microbial annotation results at the phylum level (A) and genus level (B), and (C) clustering heat map of species abundance.

Relationship between biomarkers and sheep traits

LEfSe was used to analyze biomarkers in the rumen of three breeds of sheep with LDA score = 4 (Figure 4A). The biomarkers of rumen microbes in sheep were statistically different between the different breeds. Five biomarkers in Hu sheep included Bacteroidota, Bacteroidia, Bacteroidales, Prevotellaceae, and Prevotella. The eight biomarkers in Tan sheep were Spirochaetota, Spirochaetia, Spirochaetales, Spirochaetaceae, Treponema, Bryantii, and Rikenellaceae–RC9_gut_group. The six biomarkers in Dorper sheep were Firmicutes, unidentified_Clostridia, Hungateiclostridiaceae, Saccharofermentans, and Ruminococcaceae–Ruminococcus. We used these 19 biomarkers to analyze their correlations with key growth traits and meat nutrients in sheep (Figure 4B). The biomarkers of Hu sheep were all significantly positively correlated with meat moisture, and all except Prevotella were significantly correlated with 80- to 180-d average daily gain (P < 0.05). Rikenellaceae–RC9_gut_group was significantly positively correlated with fat in Tan mutton (P < 0.05). Ruminococcus was significantly positively correlated with Dopper sheep 80-d body weight and 80- to 180-d FCR performance (both P < 0.05). The key biomarkers at the genus level were selected for correlation analysis with VFAs (Figure 4C). The Rikenellaceae_RC9_gut_group was positively correlated with isobutyric acid, butyric acid, and isovaleric acid, but negatively correlated with propionic acid (all P < 0.05).

Figure 4.

Biomarkers and sheep traits. Linear discriminant analysis effect size (LEfSe) analysis (A) and correlation analysis (B) between biomarkers and economic traits of sheep. (C) Correlation analysis between genus level biomarkers and rumen volatile fatty acids (VFAs). *P < 0.05; **P < 0.01.

Functional prediction analysis

The function of sheep rumen microbes was predicted, and 6,523 functional pathways were shared among the three varieties of rumen microbes (Figure 5A). PCoA showed that the rumen microbial function predictions of the three cultivars were similar, and the overlapping area between cultivars was large (Figure 5B). The predicted functional structures of TOP10 in different sheep were consistent (Figure 5C). The main functions were animal_parasites_or_symbionts, chemoheterotrophy, and fermentation. Cluster heat mapping positively correlated Hu sheep microbes with animal_parasites_or_symbionts, Dorper sheep microbes were positively correlated with cellulolysis, and Tan sheep were positively correlated with functions that included xylanolysis (Figure 5D). Interestingly, the microbes of Tan and Dorper sheep were positively correlated with chemoheterotrophy and fermentation, while Hu sheep showed a negative correlation.

Figure 5.

Functional pathway enrichment analysis. (A) Venn diagram of functional pathways. (B) Principal component analysis of functional pathways. (C) Functional annotation. (D) Varietal microbial functional enrichment.

Discussion

The body weight and body size of Dorper sheep at 80 to 180 d were significantly higher than those of Hu and Tan sheep, confirming the superiority of growth traits of Dorper sheep found in previous studies (Zonabend König et al., 2017; de Sousa et al., 2019). However, the daily gain of Dorper sheep during 80 to 180 d was significantly lower than that of Hu sheep, and the FCR during 80 to 180 d was significantly higher than that of Tan and Hu sheep. These findings indicate that Dorper sheep required a higher diet and energy for growth and development during the trial. Appreciable improvement in FCR was obtained after crossing Dorper sheep with Chinese landraces (Di et al., 2012). The significant difference in body weight observed in the present study may be due to higher birth weight and faster growth rate before weaning. Intramuscular fat content reportedly has a positive effect on the sensory quality of meat (Wiseman et al., 2013; Khan et al., 2015), and C8-10 branched-chain fatty acids in mutton fat strongly influence the characteristic odor of cooked mutton (Whitfield, 1992). The results of meat nutrition analysis showed that the fat content of Tan mutton was significantly higher than the other two breeds, indicating that Tan mutton has better flavor, similar to the results of Zhang et al. (2020).

The rumen is an important digestive organ of ruminants. Rumen microbiota are important in the digestion of feed by the host. The rumen microbiota structure in ruminants is similar (Szeligowska et al., 2021). However, differences in species may influence the microbiome (Hernandez-Sanabria et al., 2010; Chang et al., 2020). There have been many reports on rumen microorganisms of sheep (Fan et al., 2021). However, research on the differences of rumen microorganisms among multiple breeds is limited.

In this study, we assessed three important sheep breeds. The Chao1 index did not differ between the breeds. A significant difference in the microbial Shannon index was evident between the different breeds varieties. The gastric microorganism Shannon index of Dorper lambs was significantly lower than the other two breeds, indicating lower biodiversity in Dorper sheep. Aranaz et al. (2021) reported that individual intestinal microbial diversity was low in high-weight individuals compared to normal-weight individuals. PCoA based on UniFrac distance showed that the rumen microflora of the three breeds was partially isolated, which was verified by ANOSIM and MRPP methods. The results revealed significant differences in the microbial community structure among breeds, indicating that different breeds harbor different rumen microorganisms. Similar conclusions have been obtained in cattle (Cersosimo et al., 2016), deer (Zhao et al., 2017), and goats (Gürelli et al., 2016).

We observed that the sheep rumen is dominated by Firmicutes and Bacteroidetes, consistent with previous findings (Yang et al., 2020). Bacteroidetes was the dominant bacterial phylum in the rumen of Hu sheep, while Firmicutes was the dominant bacterial phylum in the rumen of Dorper and Tan sheep. Cheng et al. (2022) demonstrated a negative relationship of Bacteroidetes with sheep fat deposition traits, while Firmicutes showed a positive correlation, and fat deposition showed a higher correlation with body weight. Therefore, we believe that the high abundance of Firmicutes in the rumen promotes the deposition of fat in Dorper and Tan sheep; in the latter, fat is deposited intramuscularly. At the genus level, Hu sheep are highly enriched in Prevotella. Zhu et al. (2021) reported that by increasing the relative abundance of Prevotella in the rumen, the daily gain of lambs can be effectively increased, consistent with the present results. Tan sheep are mainly enriched in Enterococcus, Dysgonomonas, and Methylobacterium; Enterococcus is considered very significant for flavor development (Abeijón et al., 2006), and the increase of its abundance may improve the flavor of Tan mutton. LEfSe analysis identified 19 biomarkers in the rumen to clarify the key differences between different rumen varieties. Since differences of rumen microbes may affect the phenotypes of hosts (Amabebe et al., 2020), correlation analysis between biomarkers and important economic traits was conducted. Five biomarkers were identified in the rumen of Hu sheep. All belonged to phylum Bacteroidetes, and all were significantly positively correlated with daily gain and meat moisture of Hu sheep, and significantly negatively correlated with FCR. These findings suggest that the high abundance of Bacteroidetes, Bacteroidia, Bacteroidales, Prevotellaceae, and Prevotella in the rumen of Hu sheep can improve the digestion of feed digestion and use of the resulting energy for body development, which in turn improves feed efficiency and daily gain. Previous studies documented that the digestive tract of low FCR and high intraday weight gain individuals showed higher abundance of Bacteroides, and that the abundance of Prevotella was significantly different in sheep with high and low FCR (Wu et al., 2019; McLoughlin et al., 2020). These prior findings support the implications of the present study. A total of eight biomarkers were identified in rumen of Tan sheep, among which only Rikenellaceae and Rikenellaceae_RC9_gut_group were significantly correlated with fat content in meat. High-fat diet can increase the abundance of Rikenellaceae and Rikenellaceae_RC9_gut_group, which can regulate the lipid deposition traits by changing the abundance (Daniel et al., 2014; Wang et al., 2020). Six biomarkers were identified in Dorper sheep, among which Ruminococcus was significantly positively correlated with weight at 80 d. Thus, key bacterial species could affect the weight of Dorper sheep in the early growth stage. Ruminococcus can improve energy absorption by degrading polysaccharides, oligosaccharides, and sugars, thus increasing body weight (Dahl et al., 2020). Correlation analysis was conducted between genus level biomarkers and VFAs in the rumen of the three breeds. The Rikenellaceae_RC9_gut_group biomarker of Tan sheep was significantly negatively correlated with propionic acid, but significantly positively correlated with isobutyric acid, butyric acid, and isovaleric acid. Similar findings have been reported in yaks (Fan et al., 2020). Butyric acid is one of the main products generated by rumen microbial fermentation carbohydrates in sheep. Most is absorbed by the rumen wall and transformed into ketones, mainly β-hydroxybutyric acid, which is used for adipose tissue synthesis (Lane and Jesse, 1997). Therefore, we believe that the high abundance of Rikenellaceae_RC9_gut_group in the rumen of Tan sheep can affect the synthesis of intermuscular fat by regulating the production of VFAs.

Interestingly, although there are certain differences in microbial composition, the predicted functional pathways were similar in the sheep breeds, suggesting that the main function of rumen microorganisms is consistent, and involves energy generation through microbial fermentation for host growth and development. At present, functional enrichment is only predictive. Specific functional pathways need to be further studied.

Conclusions

In this study, three excellent sheep breeds were selected to reveal the differences in rumen microbes among different sheep breeds and the relationship between the differential microbes and host traits. Although gender, diet, and rearing environment were consistent, there were differences in microflora among the different breeds, suggesting that host genetic factors may affect rumen microbial composition of sheep. Exploring and clarifying the microbial composition of different breeds is conducive to the development of different feeding strategies. At the same time, there was a strong correlation between biomarkers and the key economic traits of the breed. This finding will inform efforts to improve sheep traits, which could more efficiently improve the growth traits and meat quality of sheep.

Glossary

Abbreviations

ANOSIM

Analysis of Similarities

FCR

feed conversion ratio

KEGG

Kyoto Encyclopedia of Genes and Genomes

LDA

linear discriminant analysis

LEfSe

linear discriminant analysis effect size

MRPP

Multi-Response Permutation Procedure

OTU

operational taxonomic unit

PCoA

principal coordinate analysis

QIIME

Quantitative Insights Into Microbial Ecology

VFA

volatile fatty acid

Conflict of Interest Statement

The authors declare no real or perceived conflicts of interest.