Animals and experimental design
All the experimental protocols were approved by the Animal Care Committee of Anhui Agricultural University. Twelve Hu lambs (local breed, 23 ± 2.3 kg initial body weight) were purchased from Tianchang Zhoushi Sheep Industry Co., Ltd. (Chuzhou, China). All of the lambs were divided into two groups (n=6), were fed basal diet or basal diet supplemented with 80 g/kg DM of gallic skin. The experimental period lasted for 10 weeks, with the first 2 weeks for adaptation. The basal diets were formulated to meet the Feeding Standards of Meat-producing Sheep and Goats in order to achieve a daily gain of 0.250 kg (Ministry of Agriculture of P. R. China, 2004; Table 1). All the animals were housed in individual pen with free access to water and were fed twice daily at 07:00 am and 19:00 pm, with approximately 10 % feed refusal.
Sample collection and measurement
Diet and ort samples were collected daily for 3 consecutive days (1st to 3rd day) every week. Samples were composited per lamb for the whole experimental period, subsampled, and were dried in a baking oven at 65℃ for 48 h, and then were milled through a 1-mm screen for the analysis of dry matter (DM, method 924.05; AOAC,1990), crude protein (method 988.05; AOAC, 1990), ash (method 942.05; AOAC, 1990), neutral detergent fiber (NDF, Van Soest et al., 1991) and acid detergent fiber (ADF, Van Soest et al., 1991) [46-47]. Body weight was measured on d 0, 28, and 56 before morning feeding and ADG was calculated.
Ruminal fluid (approximately 25 mL) was collected using an oral stomach tube immediately prior to feeding on the last day of the trial as described by Shen et al. (2018) . The pH of the ruminal fluid sample was measured immediately after collection using a portable pH meter (Ecoscan pH 5, Eutech Instruments, Singapore). Rumen fluid collections were squeezed through 4 layers of cheese cloth. 1.0 mL subsample of each rumen filtrate sample was acidified by adding 0.2 mL of 25% HPO3 for analysis of volatile fatty acids (VFAs) according to the method described by Hu et al. . Another 2.0 mL subsample was used to determine the ammonia-nitrogen by colorimetric method, as described by Chaney and Marbach . Another two 5 mL subsamples were infused into two10 mL spiral centrifuge tubes and immediately placed into a liquid nitrogen container and then transported to the laboratory, stored at -80 °C for further analysis of microbiota and metabolites. All the lambs were euthanized by an intravenous injection of sodium pentobarbital (100mg/kg) at the end of the study according to the method described by Meling et al. .
Rumen microbiota: 16S rDNA sequencing
Microbial genomic DNA was extracted from rumen fluid (4 mL) using the E.Z.N.A. stool DNA kit (Omega Bio-tek, Norcross, GA, United States) according to the manufacturer's instructions and assessed by 1% (w/v) agarose gel electrophoresis. The concentration of the extracted DNA was measured by a Qubit 2.0 (Life Technologies, Carlsbad, CA, USA). The V3-V4 hypervariable region of 16S ribosomal DNA (rDNA) gene was PCR amplified (first 5 cycles of denaturing at 95 °C for 30 s, annealing at 45 °C for 30 s, elongation at 72 °C for 30 s, then 20 cycles of denaturing at 95 °C for 30 s, annealing at 55 °C for 30 s, elongation at 72 °C for 30 s and a final extension at 72 °C for 5 min) using the forward primer 515F (5’- CCCTACACGACGCTCTTCCGATCTN-3’) and the reverse primer 806R (5’- GACTGGAGTTCCTTGGCACCCGAGAATTCCA-3’) with dual-index barcodes. PCRs were performed in a triplicate 30 μL mixture containing 20 ng of template DNA, 15 μL 2×Taq master Mix (Invitrogen, Carlsbad, CA), and 1.5 μL of each primer (10 μM).
The amplicons were run on a 1% (w/v) agarose gel, and the DNA were extracted from the bands using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, United States) according to the manufacturer’s instructions. The concentration and quality of the eluted amplicons were measured using a Qubit 2.0 (Life Technologies, Carlsbad, CA, USA). Purified amplicons were pooled in equal amounts (10 ng) and sequenced on the Illumina MiSeq platform (Sangon, Shanghai, China). The raw reads were deposited into the NCBI Sequence Read Archive (SRA) database (Accession Number: PRJNA648813).
The raw reads containing ambiguous bases and any longer than 480 base pairs (bp) were dislodged and those with a maximum homopolymer length of 6 bp were allowed, and sequence short than 200 bp were removed . Paired-end raw reads were merged as raw tags using FLSAH (v 1.2.3) with a minimum overlap of 10 bp . Noisy sequences of raw tags were filtered by the Pre.cluster tool to obtain clean tags. Chimeras were detected by using Chimera UCHIME. The effective reads were clustered into operational taxonomic units (OTUs) of ≥ 97% similarity using the UPARSE pipeline , and the representative sequence of each OUT was classified as organisms (phylum, class, order, family, and genus) by Naïve Bayesian assignment using an RDP classifier (version 2.12) . Species richness and diversity statistics including coverage, rarefaction, Chao1, ACE, Simpson, and Shannon were calculated using Mothur (version 1.30.1). A weighted and unweighted UniFrac distance matrix were also generated by Mothur (version 1.30.1), and a PCoA was conducted based on the unweighted Unifrac distance method .
Metabolomics profiling: ultra-high-performance liquid chromatography-tandem quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS)
Rumen samples were prepared according to the previously described procedures . UPLC-TOF/MS analysis was performed using an Agilent 1290 (Agilent, USA) liquid chromatograph system equipped with a UPLC BEH Amide column (2.1 * 100 mm, 1.7 μm, Waters) and coupled with TripleTOF 6600 mass spectrometry (AB Sciex) with an electrospray ionization (ESI) source. The mobile phase consisted of 25 mM ammonium acetate and 25 mM ammonia hydroxide in water（pH = 9.75, solvent A) and acetonitrile (solvent 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 flow rate was 0.5 mL/min. The column temperature was 25 °C. The auto-sampler temperature was 4 °C. The ESI-Q-TOF-MS was applied on an information-dependent basis in both positive and negative ion modes with the conditions as follows: Gas 1 as 60 psi, Gas 2 as 60 psi, Curtain Gas as 35 psi, Source Temperature as 600 °C, Decluttering potential as 60 V, Ion Spray Voltage Floating as 5000 V or -4000 V in positive or negative modes, respectively.
Metabolomics data analysis
The injection volume was 1 μL (positive) or 1 μL (negative), respectively. During the operation, the acquisition software (Analyst TF 1.7, AB Sciex) continuously evaluates the full scan survey MS data as it collects and triggers the acquisition of MS/MS spectra depending on preselected criteria. In each cycle, the most intensive 12 precursor ions with intensity above 100 were chosen for MS/MS at collision energy (CE) of 30 eV. The cycle time was 0.56 s. MS raw data were converted to the mz XML format by Proteo Wizard and processed by R package XCMS (version 3.2). The process includes peak deconvolution, alignment and integration. Minfrac and cut off are set as 0.5 and 0.3 respectively. In-house MS2 database was applied for metabolites identification. A total of 1,812 resulting peaks with the peak numbers, sample names, and normalized peak areas were import to the SIMCA software package (V15.0.2, Sartorius Stedim Data Analytics AB, Umea, Sweden) for PCA and OPLS-DA. Additionally, 7-fold permutation tests in the OPLS-DA model were used to verify model validity and robustness. Encyclopedia of Genes and Genomes (KEGG, http://www.genome.jp/kegg/) was utilized for metabolite identification and confirmation, and MetaboAnalyst 4.0 was used for the pathway topology analysis.
The matter intake (DMI), growth performance, and rumen fermentation characteristics were analyzed as a randomized block design using PROC MIXED of SAS software (version 9.3, SAS Institute Inc., Cary, NC, United States). Dietary treatment was included as a fixed effect, and lamb was included as a random effect. Means were separated using the PDIFF option in the LSMEANS statement.
Statistical calculations rumen bacterial community, metabolomic, and their correlation data were carried out by conducting tests using the SPSS software package (SPSS version 23.0; SPSS Inc., Chicago, IL, United States). The Kruskal-Wallis sum-rank test was used to select and demonstrate differentially abundant taxa between the groups. For metabolite identification, the first principal component with a VIP value >1.0 and a P value < 0.05 in Student’s t test were considered significantly different.
The correlations between different rumen microbial genera (P < 0.05 and relative abundance > 0.05% in at least one of the samples) and varied altered rumen metabolites (VIP > 1.0 and P < 0.05) were assessed by Spearman’s correlation test. Significance was declared at P < 0.05, and a tendency was declared at 0.05 ≤ P < 0.10.