Background

Ruminants have a complex rumen microbial community to help in adaptation to high fiber plants and provide energy in the form of VFA for the growth of the host by fermenting nutrients [1]. Rumen microbial community is known to be influenced by various factors such as diet, age, genetics, breed, and geography [2,3,4]. These factors directly or indirectly influence rumen microbiota that responds to variations in the environment and might change the physiological response of the host.

The development of rumen microbiota is linked to the structural variations of the rumen with age. The relationship between host and rumen microbiota occurs at birth as vertical transmission of microbes from the mother and is considered a crucial route for the establishment of microbiota in newborns [55]. The other two genera are known to produce butyrate by fermenting complex polysaccharides fiber and influence the rumen development and health of animal [56]. Overall, we assumed that these genera might be aiding in digestion of high fiber diet and maintaining the health of Mongolian cattle. The networks provide new dimensions to our understanding of the age-dependent variations in rumen bacterial community interactions by identifying keystone taxa.

Although three different age groups received the same diet, the variations in bacterial abundance might indicate that at 5 months of age rumen microbiota undergoes developmental changes independent of diet. The presence of Short-chain fatty acids producing bacteria in RM18 might be linked to its role in providing energy to the growing animals and hel** them to survive on shrubs and herbs during the cold and arid environment. While fiber degrading bacteria in RM36 might indicate the establishment of a stable and mature microbial community. The diet of Mongolian cattle is dominated by a variety of shrubs and halophytes, so rumen microbiota must be adapted to degrade such a recalcitrant diet which is rich in lignocellulosic materials. It is noteworthy that the age gap in this study is quite large and perhaps the analysis of smaller age would reveal more detailed variations in rumen microbiota with development.

Conclusions

In conclusion, the rumen microbiota of Mongolian cattle reached to stability and maturity with age after weaning. The diversity of the rumen bacterial community was lower at a young age which becomes stable with age. Bacteroidetes and Firmicutes were the core phyla in all age groups. We identified functional-specific bacterial genera in three age groups. Genera Prevotella_1, Bacteroides, and Bifidobacterium were abundant in RM05. The Short-chain fatty acids producing bacteria Rikenellaceae_RC9_gut_group showed high abundance in the RM18 group and the fiber degrading genus Alloprevotella was highly abundant in the RM36 group. The genera Ruminococcaceae_UCG-005, Bacteroides, Saccharofermentans, and Fibrobacter in RM05, genera [Eubacterium] coprostanoligenes_group, Erysipelotrichaceae_UCG-004, Helicobacter, Saccharofermentans, Papillibacter, and Turicibacter in RM18, and genera Rikenellaceae_RC9_gut_group, Lachnospiraceae_AC2044_group, and Papillibacter in RM36 showed the top interactions values in the intra-group interaction network. This study provides some preliminary information about the structure and composition of rumen microbiota in Mongolian cattle from weaning to adulthood. Further studies are needed to determine their actual roles and interactions with the host.

Methods

Site description

The trial was conducted at Alashan, which is located in the westernmost part of Inner Mongolia Autonomous Region, bordered in the north by Mongolia, in the south and west by Gansu province, China during the winter of 2020. Alashan has a continental climate, which is dry and windy. Winter is cold and summer is hot. Vegetation in this area is dominated by Salix cupularis, Haloxylon ammodendron, Caragana jubata and Kobresia spp. The precipitation on the Alashan is under 150 mm/year, while annual temperature ranges from 6 °C-8°C on average.

Animals and sample collection

Fifteen Ujumqin Mongolian female cattle from three different age groups i.e., 5 months old (RM05), 18 months old (RM18), and 36 months old (RM36) were randomly selected from a farm in Alashan Mongolia and were individually penned until sampling. The selected animals shared the same raising protocols i.e., all animals were naturally weaned at 5 months of age and before weaning they were purely on mother milk. After weaning, animals were allowed to graze the natural alpine shrub grasslands year round and drank water from the local river. None of the studied animals were pregnant or given birth before. The animals used in this study were not genetically related or receiving antibiotic treatment. All the animals were purely grazing and were not provided with any supplements. The body weight of animals from each age group was measured at the time of sample collection by using electronic weighing scale (Shanghai Yaohua Urban Systems Co., Ltd. Shanghai, China).

Animals from RM18 and RM36 were restrained in a veterinary crush before sampling to ensure the safety of animals, while animals from RM05 group were straddled between the handler’s legs and their shoulders were firmly squeezed between legs to avoid movement and misplacement of the oral tube. Rumen content (70 mL/animal, liquid part) was collected using an oral stomach tube from each animal before morning grazing and snap-frozen in liquid nitrogen and then stored at − 80 °C until use. Polyvinyl chloride oral tube (length = 125 cm) with small side holes (7 mm in diameter) located at the insertion end was used for young animals [57]. The length of the tube to be inserted was measured as the distance from the tip of the calf’s nose to the point of its elbow behind the front leg and marked on the tube with a piece of tape i.e., approximately 45 cm. For adult animals, stainless steel rumen fluid extractor (Chengdu Huazhi Kaiwu Technology Co., Ltd., Chengdu, China) was used for sampling and approximately 200 cm of the tube was inserted to reach the center of the rumen. Each time before taking the new sample, the tube was thoroughly cleaned with fresh warm water and about 10-15 ml of the sample from each cattle was always discarded to prevent saliva contamination [58].

Analysis of rumen volatile fatty acids

The frozen rumen fluid sample was thawed at 4 °C and thoroughly mixed by vortexing. After that, 10 mL of rumen fluid was centrifuged at 3000 g for 10 min, and 1 mL of the supernatant was transferred to a 1.5 mL centrifuge tube, along with 0.2 mL of a metaphosphoric acid solution containing the internal standard 2-ethylbutyric acid. The sample was mixed, placed in an ice-water bath for 30 min, and centrifuged at 10,000 × g at 4 °C. The supernatant was transferred to a new 1.5 mL centrifuge tube and placed at 4 °C for testing. The volatile fatty acids (VFA) concentration was determined by gas chromatography (Agilent Technologies 7820A GC system, Santa Clara, CA) equipped with a 30 m × 0.25 mm × 0.33 μm fused silica column (AE-FFAP, Atech Technologies Co. Ltd., Shanghai, China). The gas chromatographic conditions and subsequent test procedures were conducted as described previously [59].

DNA extraction and sequencing

Genomic deoxyribonucleic acid (DNA) from rumen fluid was extracted by cetyltrimethylammonium bromide method [60] and pure DNA was eluted in 150 µL of elution buffer and stored at − 20 °C until use. DNA quality and quantity were checked by 1.5% agarose gel electrophoresis and NanoPhotometer® spectrophotometer (Implen, Westlake Village, CA, USA), respectively [61]. Polymerase chain reaction (PCR) amplification of the V3-V4 region of the 16S rRNA gene was performed for bacterial analysis by using universal primer pairs (343F (5'-TACGGRAGGCAGCAG-3')-798R (5'-AGGGTATCTAATCCT-3')) with barcodes [62]. PCR amplification was performed by using Phusion® High-Fidelity PCR Master Mix with GC Buffer from New England BioLabs. Briefly, PCR amplifications were done in duplicate with 25 μL reaction mix containing 2X phusion master mix, 0.5 µM forward and reverse primers, and 20 ng of genomic DNA. The thermal cycling procedure consisted of an initial denaturation step at 98 °C for 30 s, followed by 25 cycles of 98 °C for 10 s, 56 °C for 30 s, and 72 °C for 20 s, and a final extension at 72 °C for 10 min. The amplicons were visualized using 1.5% agarose gel electrophoresis and purified with AMPure XP beads (Agencourt) according to the manufacturer's instructions [63]. The purified products were used for second round of PCR for the enrichment of ampilcons having adapters on both sides using TruSeq™ DNA sample preparation kit (Illumina Inc, San Diego, CA) according to the manufacturer’s protocol and quantified using Qubit dsDNA Assay kit (Thermo Fisher) [64]. Paired-end sequencing was carried out according to the standard protocol using the Illumina HiSeq2500 PE250 method by commercial company (Oebiotech, Shanghai, China) [65].

Bioinformatics analysis

After sequencing, barcodes and primer sequences were truncated. The QIIME (Quantitative Insights Into Microbial Ecology, Version 1.9.1) software was used to remove low-quality sequences from the raw data to get clean tags [66]. Chimera sequences were removed from clean tags using UCHIME (version 2.4.2) software to get valid tags [67]. Valid tags were clustered into Operational Taxonomic Unit (OTUs) using Vsearch (version 2.4.2) software according to 97% similarity [68]. The representative sequences of the OTUs were used to classify bacterial taxa using against Silva database (Version 123) (https://www.arb-silva.de/) using RDP Naive Bayesian classifier algorithm [69, 70]. Rarefaction curve was constructed in QIIME software, while bargraphs at phylum, family, and genus levels were created using GraphPad Prism version 8.00 for Windows (www.graphpad.com/). The alpha diversity indices such as Chao1, Shannon, Simpson, and Goods-coverage were calculated using QIIME software (Version 1.9.1). Unweighted and weighted uniFrac distance based principal coordinates analysis (PCoA) plots were drawn in R studio (Version 2.15.3) (http://www.rstudio.com/) using vegan package to demonstrate the difference between samples [71]. A correlation heat map was generated in GraphPad Prism version 8.00 for Windows (www.graphpad.com/).

Statistical analysis

Before any statistical analyses were conducted, all data were checked for normality using Shapiro–Wilk test using SPSS software (Version 20.0, IBM, Armonk, NY, United States). The Kruskal Wallis test was used to compare VFA across different groups using R software (Version 2.15.3) [72]. Analysis of similarities (ANOSIM) analysis was performed by using the ANOSIM function of the R vegan package to confirm statistically significant differences between groups [73]. The linear discriminant analysis effect size (LEfSe) method was used to examine age-dependent variations at phylum and genus levels using a linear discriminant analysis (LDA) score equal to 4 as a thresholds value [74]. Microbial interactions within studied age groups (Intra-group interaction) were determined by Spearman’s correlation coefficient of 40 top rumen bacterial genera to identify keystone species. Only genera showing P < 0.05 were further selected to plot the network by using the cytoscape (Version 3.6.13) [75]. Random forest analysis was performed to identify important age discriminatory bacterial genera by using randomForest function in the R package [76]. All P-values were adjusted using false discovery rate to remove false-positive results and significance was declared at P < 0.05.