The experimental design and procedures were approved by the Institutional Animal Care and Use Committee of the Gansu Agricultural University under permit No.GSAU-Eth-AST-2022-034. A total of 36 crossbred (small-tailed Han × Hu) female lambs (3 months old, a mean body weight 20.11 ± 0.19 kg) were randomly divided into 4 groups (3 pens, 3 lambs/pen). The CON group was fed a basal diet, the MON group received a basal diet plus 0.5 g/head/day MON. The OEO group received a basal diet with 52 mg/head/day OEO, and the LR group received a basal diet supplemented with 10 g/head/day LR, based on the manufacturers’ recommendation. The MON was supplied by XABC Biotechnology Co. Ltd. (Xian, China) at a 90% concentration. The OEO used to supplement the diet (containing 1.3% OEO from Origanum vulgare subsp. hirtum plants) was obtained from Ralco Inc., Marshall, MN, USA with 98.7% natural feed grade inert carrier. Supplementation with 52 mg/head/day was based on previously reported studies (21). Lyophilized LR (2 × 10⁹ CFU/g) was obtained commercially. The nutrient composition of the basal diet is presented in Table S1 (NRC, 2007). Feed was offered in equal amounts at 07:00 and 18:00 daily. The study period lasted 90 days. All animals were provided ad libitum access to feed and water.

At the 90th day of the trial period, the sheep were slaughtered after 12 h feed restriction. Five milliliters of blood were collected from the jugular vein before slaughter. Serum was collected and stored at -20°C. Jejunal and ileal contents were collected from the mid-jejunum and mid-ileum segments, which were then flushed with PBS. One section was snap-frozen in liquid nitrogen and then stored at -80°C until the time of analysis. Other sections of jejunum and ileum (approximately 2 cm) were fixed in 4% paraformaldehyde for histological analysis. Mucosal samples from the jejunum were scraped and frozen in liquid nitrogen and then stored at -80°C for analysis of mRNA expression.

Serum samples were thawed at 4°C and mixed thoroughly prior to analysis. Complement 3 and C4 were measured using a commercially available kit (Beijing sino-uk institute of Biological Technology, Beijing, China) and analyzed in a semi-automatic biochemical analyzer (A-6, Beijing, China). The concentrations of immunoglobulins (IgA, G, M), IL-1β, IL-2, IL-4, IL-6, TNF-α and TNF-β in serum were detected by ELISA using commercially available kits and according to the manufacturer’s instructions (Beijing sino-uk institute of Biological Technology, Beijing, China).

Intestinal samples were removed from the 4% paraformaldehyde fixative solution and embedded in paraffin. Each segment was sliced to a thickness of 3μm. Sections were stained with hematoxylin and eosin for analysis. Villus height (VH), villus width (VW), crypt depth (CD) and VH/CD ratios were measured. Ten of the most complete and well-oriented villi and their associated crypts from each segment were measured at 40 × magnification by light microscopy (Motic BA 210, Xiamen, China) coupled with an image analyzer (Image-Pro Plus 6.0, Media Cybernetics, Bethesda, MD, USA). Goblet cells were stained using Alcian Blue-periodic acid Schiff (AB-PAS) and counted on 5 morphologically similar villi per sheep.

The ileal contents collected above were homogenized thoroughly in a saline solution, and centrifuged at 2500 × g for 15 min at 4°C. The supernatants were then collected and immediately frozen at -80°C for use in later analyses. The amylase, lipase and trypsin enzyme activities in the samples were determined using standard kits according to the manufacturer’s instructions (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

Total RNA was isolated from the jejunal mucosa samples using Trizol reagent (TransGen Biotech Co., Ltd., Beijing, China) according to the manufacturer’s instructions. The concentration and quality of total RNA were detected using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA). Optical density ratios (260/280 nm) between 1.8 and 2.0 indicated that the RNA was pure and suitable for use in further analyses. The resulting total RNA was used to generate cDNA using the Evo M-MLV RT Kit with gDNA Clean for qPCR II (AG11711, Accurate Biotechnology, Hunan). The cDNA was then used for amplification in a real-time PCR System (LightCycler480, Roche, Basel, Switzerland) using the SYBR Premix Pro Taq HS qPCR Kit (AG11701, Accurate Biotechnology, Hunan). The reaction mixture (20 μL) for qPCR contained 10 μL 2 × SYBR Green Pro Taq HS Premix, 0.8 μL each of the forward and reverse primers (10 μmol/L), 2 μL cDNA and 6.4 μL RNase free water. The following PCR thermocycling profile was used: 30 s at 95°C; and 40 cycles of 95°C for 30 s, 5 s at 60°C. GADPH was used as endogenous control, and the relative expression of genes was calculated using the 2^{−△△ Ct} method (22). The gene specific primers were designed using Primer 3.0 (Applied Biosystems, Foster City, CA, USA) and synthesized by Zhongke Yutong (Shanxi, China). Primer sequences can be found in Table S2 .

Total genomic DNA was extracted from 36 ileal content samples using the E.Z.N.A. stool DNA Kit (Omega Bio-tek, Norcross, GA) according to the manufacturer’s instructions. The concentration and quality of the extracted DNA were assessed by a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA), and 2% agarose gel electrophoresis, respectively. The V3-V4 region of the bacterial 16S rDNA was amplified using the primers 338F (5’-ACTCCTACGGGAGGCAGCAG-3’) and 806R (5’-GGACTACHVGGGTWTCTAAT-3’) in the ABI GeneAmp^® 9700 PCR System (Applied Biosystems Life Technologies, Foster City, CA, USA). The PCR thermocycling profile used was as follows: 95°C for 3 min, followed by 28 cycles at 95°C for 30 s, 55°C for 30 s and 72°C for 45 s, with a final extension at 72 °C for 10 min. Then, the amplicons were purified from 2% agarose gels using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, United States) according to the manufacturer’s instructions. The purified amplicons were then subjected to paired-end sequencing (2 × 300 bp) on the Illumina MiSeq PE300 platform (Illumina, San Diego, USA) according to the standard protocols at Majorbio Bio Technology Co. Ltd. (Shanghai, China). The raw reads were deposited into the NCBI Sequence Read Archive (SRA) database under the accession number PRJNA744748.

The raw paired-end sequences were demultiplexed, quality-filtered with fastp (version 0.19.6) and merged with FLASH (version 1.2.7) (23, 24). High-quality reads were selected using the DADA2 (25) plugin in QIIME2 (version 2020.2) under the recommended parameters to generate amplicon sequence variants (ASVs). The taxonomy of each 16S rRNA gene sequence was analyzed in the RDP Classifier algorithm against the Silva 138 database. A comparison threshold of 70% was used for the analysis.

Approximately 50 mg of ileal content samples were processed with 500 μL methanol/water (ice-cold, 70%, v/v) containing 2-chlorophenylalanine (1 μg/mL, included as an internal standard) and then vortexed for 3 min, followed by sonication in ice water for 10min. After centrifugation at 12,000 × g for 10 min, the supernatants were transferred to autosampler vials for UPLC-MS/MS analysis (metware).

Two microliters of samples were analyzed by UPLC-MS/MS (UPLC, Shim-pack UFLC SHIMADZU CBM A system; MS, QTRAP^® System). In both ESI positive and negative modes, the solvent system was composed of A (0.04% acetic acid in water) and B (0.04% acetic acid in acetonitrile) and was run at a flow rate of 0.4 mL/min using the following elution gradient: 95:5 V/V at 0 min, 5:95 V/V at 11.0 min, 5:95 V/V at 12.0 min, 95:5 V/V at 12.1 min, 95:5 V/V at 14.0 min. The following ESI source conditions were utilized as follows: source temperature 500°C; ion spray voltage (IS) 5500 V or -4500 V; ion source gas I (GSI), gas II (GSII), and curtain gas (CUR) were set at 55, 60, and 25.0 psi, respectively; the collision gas (CAD) was high.

The results of growth performance, serum parameters, digestive enzyme data, histology, and relative gene expression were analyzed between two groups using a Student’s t-test. The analyses were performed using SPSS (version 20.0) and expressed as means ± SEM. Data differences were considered to be statistically significant at a value of P < 0.05 and statistically tendency at 0.05 ≤ P < 0.10.

For microbial community profiling, the alpha diversity analysis included richness estimator (ACE and Chao 1) and diversity indices (Shannon and Simpson) were calculated based on ASV-level bacterial taxa with a Wilcoxon rank-sum test. The results were presented as box-and-whisker plots using GraphPad Prism (version 8.0). Principal coordinates analysis (PCoA) based on binary-hamming distances and the analysis of similarities (ANOSIM) were performed to analyze the similarity or difference of the compositions of the bacterial communities. Taxa abundances at the phylum and genus levels were statistically compared among groups using python 2.7. Linear discriminant analysis (LDA) coupled with effect size (LEfSe) was performed to determine the microbial differences among treatments and LDA > 2.5. The Wilcoxon rank-sum test was used to identify differences in the most abundant genera between groups. False discovery rate (FDR) was used to correct p-values.

The Analyst 1.6.3 software was used to process mass spectrometric data. Orthogonal projections to latent structures-discriminate analysis (OPLS-DA) were used to determine metabolic differences between the two groups. The parameters of R² and Q² were used to evaluate the OPLS-DA model validity to avoid over-fitting. The fold change (FC) was calculated as the ratio of mean peak values between each of the two groups. The significantly different metabolites were selected based on the variable important in projection (VIP) ≥ 1.00 and FC ≥ 2 or ≤ 0.5 between the two groups. The differentially expressed metabolites were represented as the log of peak area using box plots. The Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used for annotation of the results, as well as for enrichment and classification for identification of pathway enrichment patterns. The correlations between the ileum bacterial and digestive enzyme, inflammatory cytokines, and barrier functions related gene expression, and ileum microflora-related metabolites were analyzed using the Spearman’s correlation test in R (version 3.1.1).

Compared to the CON group, OEO and LR supplementation significantly increased ADG and decreased the observed feed/gain (F/G) ratio (P < 0.001) ( Figures 1A, B ). Analysis of the serum immune related proteins showed that serum C3 was significantly decreased in the MON group (P < 0.01), OEO significantly decreased the IL-2 and TNF-β concentration (P < 0.05) and tended to increase IgA (P = 0.076) and IgG (P = 0.078) concentrations of serum, LR enhanced the IgA concentration (P < 0.01) relative to the CON group ( Figures 1C, D ). These results suggested that OEO improved growth performance and may exert anti-inflammatory effects.

Dietary supplementation with OEO alleviated the villus destruction and severe desquamation in the jejunum and ileum ( Figure 2A ). The VW of the ileum in the LR group was higher than that in the CON group (P < 0.05). However, no significant differences were observed for VH, CD, VH/CD among groups in the jejunum and ileum (P > 0.05) ( Figure 2B ). Inclusion of OEO and MON to the diets significantly increased the number of goblet cells in jejunal villi compared to the CON group (P < 0.01) ( Figures 2C, D ). In the ileum, the lipase activity was significantly decreased in the LR group (P < 0.01). Amylase activity was significantly increased in the OEO group (P < 0.01) when compared with CON group ( Figure 3A ). Overall, these results suggested that OEO supplementation provided benefits to intestinal morphology and digestive enzyme activity.

To assess the influence of OEO on intestinal barrier, the expression of tight junction proteins, mucins, and cytokines in the jejunal mucosa were measured by qRT-PCR. The mRNA expression of OCLN (P < 0.01) and MUC2 (P < 0.05) increased, while the mRNA expression of CLDN1 (P < 0.05) and TGF-β1(P < 0.01) decreased in the MON group compared to the CON group. Supplementation of OEO increased mRNA expression of MUC1, MUC2, MUC13, and MUC20 (P < 0.01), but decreased the expression of nuclear factor kappa B (NF-κB) p65 (P < 0.01), toll-like receptor-4 (TLR-4) (P < 0.001) and IL-6 (P < 0.05), as compared with the CON group. The mRNA expression of NF-κB p65 (P = 0.092) and IL-6 (P < 0.05) decreased, and no significant differences were observed for the mRNA expression of MUC1, MUC2, MUC13, and MUC20 in the LR group, as compared with the CON group ( Figures 3B–I ). These results indicated that OEO enhanced the mucosal barrier function.

To further study whether OEO could affect the intestinal microbiota, the composition of the ileal microbiota was analyzed via sequencing of the 16S rRNA gene. A total of 1,965,934 V3–V4 16S rRNA effective sequences were obtained from the 36 samples, with an average of 54,609 sequences per sample. The α-diversity analysis revealed that ACE and Chao1 indices were significantly decreased in the trial group, and the Shannon index significantly decreased in the MON and LR groups. However, the Simpson index significantly increased in LR group ( Figure 4A ) ( Supplementary Figures S1A–C ). The PCoA plots using the binary-hamming method revealed that there were differences in microbiota between groups (ANOSIM, R = 0.114, P = 0.003) ( Figure 4B ).

At the phylum level, Firmicutes, Actinobacteriota, and Proteobacteria were the predominant bacteria ( Figure 4C ). At the genus level, the dominant genera were Romboutsia and Aeriscardovia ( Figure 4D ). According to the statistical analysis, OEO increased the relative abundance of Ruminococcus (4.98 ± 1.52 vs 2.20 ± 0.51), Bifidobacterium (4.47 ± 4.32 vs 0.13 ± 0.08), Enterococcus (P < 0.05), but decreased the relative abundance of Eubacterium_saphenum_group relative to CON (P < 0.05). Conversely, LR increased the relative abundance of Streptococcus (14.64 ± 9.78 vs 0.57 ± 0.31) and Lactobacillus (0.35 ± 0.83 vs 0.02 ± 0.03), but decreased the relative abundance of Turicibacter (P < 0.01), Lysinibacillus (P < 0.001), Clostridium_sensu_stricto_1, and Eubacterium_saphenum_group (P < 0.05). It was observed that MON decreased the relative abundance of Lysinibacillus (P < 0.05) and Eubacterium_saphenum_group (P < 0.05). Compared with the CON group, the relative abundances of norank_f_norank_o_Saccharimonadales was significantly higher in the MON group, whereas Enterococcus was significantly higher in the OEO group and Marvinbryantia and Succiniclasticum were significantly higher in the LR group ( Figure 4E ) ( Supplementary Figures S1D–F ).

The different bacteria from phylum to genus level that were specific between the trial and CON groups were identified by LEfSe analysis ( Figure 4F ). There were 23, 2, 8, 4 dominant taxa in ileal content samples of sheep fed CON, MON, OEO and LR diets, respectively. Anaerovoracaceae, Family_XIII_AD3011_group, Mogibacterium, Family_XIII_UCG-001, Eubacterium_saphenum_group, unclassified_f_Anaerovoracaceae, Eggerthellaceae, DNF00809, unclassified_f_Eggerthellaceae, Planococcaceae, Lysinibacillus, Eubacterium_hallii_group, Lachnospiraceae_UCG-002, Lachnospiraceae_FE2018_group, Atopobium, norank_f_Atopobiaceae, unclassified_f_Atopobiaceae, UCG-010 and norank_f_UCG-010 were enriched in the CON group. Peptococcus and Solobacterium were enriched in the MON group, whereas Erysipelotrichales, Erysipelotrichaceae, Turicibacter, Enterococcaceae, Enterococcus, Rikenellaceae, norank_f_Eggerthellaceae and Coriobacteriaceae_UCG-003 were enriched in the OEO group. Marvinbryantia, Acidaminococcales, Acidaminococcaceae and Succiniclasticum were enriched in the LR group. Collectively, these results demonstrated that multiple beneficial bacteria were significantly enriched in the OEO supplemented group.

A Spearman correlation analysis was employed to investigate the correlations among intestinal barrier functions, mucin gene expression, inflammatory cytokines, digestive enzyme activity, and the predominant microbial genera. In this study, the abundance of Bifidobacterium (P < 0.01) and Lactobacillus (P < 0.05) were positively correlated with MUC2 expression, and Lactobacillus (P < 0.01) abundance was negatively correlated with IL-6 expression. Ruminococcus (P < 0.001) and Enterococcus (P < 0.05) were positively correlated with MUC20 expression. Moreover, the relative abundance of Enterococcus was positively correlated with amylase activity (P < 0.05). The lipase activity was positively correlated with Turicibacter (P < 0.01) and Lysinibacillus (P < 0.001) ( Figure 4G ). These results further indicated that the effects of OEO regulated intestinal microbiota may contribute to improve the intestinal function.

The UPLC-MS/MS platform was used to analyze the changes of ileal content metabolite profiles in sheep provided either standard or supplemented diets. A total of 465 metabolites were detected. The OPLS-DA indicated a clear separation between the CON and OEO groups (R²X = 0.590, R²Y = 0.996, Q² = 0.567, Figure 5A ), the CON and MON groups (R²X = 0.518, R²Y = 0.990, Q² = 0.718, Supplementary Figure S2A ), and the CON and LR groups (R²X = 0.650, R²Y = 0.995, Q² = 0.795, Supplementary Figure S2B ), which suggested that these groups had differential metabolites in the contents. In addition, the value of Q² > 0.5 suggests that the OPLS-DA models were reliable.

The parameters of the variable importance in the project (VIP ≥ 1.00 and FC ≥ 2 or ≤ 0.5) were used to screen out differentially expressed metabolites ( Supplementary Table S3 , Figures 5B,C , Supplementary Figures S2C–E ). When comparing the CON and OEO groups, 31 metabolites with significant differences were observed, 21 being up-regulated and 10 down-regulated. Levels of indole acetaldehyde (IAAD), indole-3-acetic acid (IAA) and N’-formylkynurenine (NFK) were significantly increased in response to OEO supplementation, while 2-(formylamino) benzoic acid was significantly decreased. When comparing the CON and MON groups, 31 metabolites with significant differences were observed, with 15 up-regulated and 16 down-regulated. Levels of taurochenodesoxycholic acid (TCDCA), spermidine, IAAD, IAA, and NFK were significantly increased by MON, while taurocholic acid (TCA) was significantly decreased. Between the CON and LR groups, 63 metabolites with significant differences were observed, with 24 being up-regulated and 39 down-regulated. Levels of IAAD and IAA were significantly increased by LR, while tryptamine and 5-hydroxyindole-3-acetic acid (5-HIAA) decreased significantly. Additionally, compared with the CON group, levels of lysope 16:0, lysope 18:0, and pantothenol were significantly decreased in all experimental groups.

Compared with the CON group, sheep receiving the OEO supplement mainly exhibited alterations in pyrimidine metabolism and tryptophan metabolism ( Figure 5D ), whereas the MON group exhibited a greater impact on 4 metabolic pathways including cholesterol metabolism, primary bile acid biosynthesis, bile secretion and tryptophan metabolism ( Supplementary Figure S2F ). The LR group was enriched in the pathway of tryptophan and phenylalanine metabolism ( Supplementary Figure S2G ). Thus, dietary supplementation with OEO may have important modulatory effects on tryptophan metabolism.

The Spearman correlation coefficient was used to investigate the potential associations between the ileum microbiota and differential metabolites observed in the same animals (|R| > 0.6, P < 0.05). When a comparison was made between the CON and OEO groups, Eubacterium_saphenum_group was significantly positively correlated with 2-(formylamino) benzoic acid (R = 0.733, P = 0.033) and m-coumaric acid (R = 0.721, P = 0.034), but was negatively correlated with NFK (R = -0.673, P = 0.038), IAAD (R = -0.696, P = 0.035) and IAA (R = -0.735, P = 0.034). Enterococcus was negatively correlated with kinic acid (R = -0.709, P = 0.035, Figure 5E ). Between the CON and MON groups, Eubacterium_saphenum_group was significantly negatively correlated with TCDCA (R = -0.718, P = 0.048) and spermidine (R = -0.776, P = 0.026), whereas Bacteroides was significantly positively correlated with NFK (R = 0.759, P = 0.029) and IAAD (R = 0.752, P = 0.030, Supplementary Figure S2H ). Between the CON and LR groups, Eubacterium_saphenum_group was significantly positively correlated with lysope 16:0 (R = 0.667, P = 0.045) and pantothenol (R = 0.740, P = 0.024). Lysinibacillus was negatively correlated with IAAD (R = -0.770, P = 0.019), NFK (R = -0.785, P = 0.014) and IAA (R = -0.710, P = 0.036, SupplementaryFigure S2I ).