Multi-omics analysis reveals the mechanism of rosemary extract supplementation in increasing milk production in Sanhe dairy cows via the “rumen-serum-milk” metabolic pathway

Figures

• 

  Fig. 1. Impact of rosemary extract (RE) supplementation on the rumen metabolic profile of Sanhe dairy cows. (Aand B) Partial least squares discriminant analysis (PLS-DA) and fitting models of rumen metabolites in negative ion mode for the RE and CON groups. (C and D) PLS-DA analysis and fitting models of rumen metabolites in positive ion mode for the RE and CON groups. (E) Volcano plot depicting differential metabolites in the rumen between the RE and CON groups. Red dots denote differentially expressed metabolites significantly upregulated in the RE group compared to the CON group; blue dots denote those significantly downregulated; grey dots denote metabolites showing no significant change between the two groups. (F) Cluster heatmap of differential metabolites in the rumen between the RE and CON groups. (G) Metabolic pathways significantly influenced by differentially enriched metabolites in the rumen of the RE and CON groups. RE, RE group, n = 15, the basal diet supplemented with 20 g RE/d; CON, control group, n = 15, the basal diet without supplementation of RE. Sig.-Up = significantly up – regulated; Sig.-Down: significantly down – regulated; Ns(N): not significant; PC = principal…

• 

  Fig. 2. Impact of rosemary extract (RE) supplementation on the serum metabolic profile of Sanhe dairy cows. (A and B) Partial least squares discriminant analysis (PLS-DA) and fitting models of serum metabolites in negative ion mode for the RE group and CON group. (C and D) PLS-DA analysis and fitting models of serum metabolites in positive ion mode for the RE group and CON group. (E) Volcano plot depicting differential metabolites in the serum between the RE and CON groups. Red dots denote differentially expressed metabolites significantly upregulated in the RE group compared to the CON group; blue dots denote those significantly downregulated; grey dots denote metabolites showing no significant change between the two groups. (F) Cluster heatmap of differential metabolites in the serum between the RE and CON groups. (G) Metabolic pathways significantly influenced by differentially enriched metabolites in the serum of the RE and CON groups. RE, RE group, n = 15, the basal diet supplemented with 20 g RE/d; CON, control group, n = 15, the basal diet without supplementation of RE. Sig.-Up = significantly up – regulated; Sig.-Down: significantly down – regulated; Ns(N): not significant; PC =…

• 

  Fig. 3. Impact of rosemary extract (RE) supplementation on the milk metabolic profile of Sanhe dairy cows. (A and B) Partial least squares discriminant analysis (PLS-DA) and fitting models of milk metabolites in negative ion mode for the RE group and CON group. (C and D) PLS-DA analysis and fitting models of milk metabolites in positive ion mode for the RE group and CON group. (E) Volcano plot depicting differential metabolites in the milk between the RE and CON groups. Red dots denote differentially expressed metabolites significantly upregulated in the RE group compared to the CON group; blue dots denote those significantly downregulated; grey dots denote metabolites showing no significant change between the two groups. (F) Cluster heatmap of differential metabolites in the milk between the RE and CON groups. (G) Metabolic pathways significantly influenced by differentially enriched metabolites in the milk of the RE and CON groups. RE, RE group, n = 15, the basal diet supplemented with 20 g RE/d; CON, control group, n = 15, the basal diet without supplementation of RE. Sig.-Up = significantly up – regulated; Sig.-Down: significantly down – regulated; Ns(N): not significant; PC =…

• 

  Fig. 4. Comparative analysis of metabolic differences among the rumen, serum, and milk of Sanhe dairy cows in response to rosemary extract (RE) supplementation. (A) UpSet diagram illustrating common differential metabolites in the rumen, serum, and milk after comparison between the RE and CON groups. (B) Cluster analysis of common differential metabolites in the rumen, serum, and milk. (C) Fold change (FC) (RE group vs. CON group) of common differential metabolites in the rumen and serum. (D) Receiver operating characteristic (ROC) analysis of common differential metabolites between serum and other body fluids, highlighting metabolites with an area under the curve (AUC) > 0.9000. (E) Classification and cluster analysis of differential lipid metabolites in milk between the RE and CON groups, with inter-group fold change (RE group vs. CON group). RE, RE group, n = 15, the basal diet supplemented with 20 g RE/d; CON, control group, n = 15, the basal diet without supplementation of RE.

• 

  Fig. 5. Impact of rosemary extract (RE) supplementation on the rumen microbial composition of Sanhe dairy cows. (A) Comparison of microbial kingdom between the RE and CON groups. "∗" indicates significantly different kingdoms tested by the Wilcoxon rank-sum test with an adjusted P-value of <0.05. (B) Alpha diversity analysis based on species abundance at the genus level for the RE and CON groups. ∗ indicates a significant difference (P < 0.05), and NS indicates a non-significant difference (P＞0.05). (C) Principal coordinates analysis (PCoA) based on species abundance at the genus level. (D) Anosim analysis based on species abundance at the genus level. (E) Sample clustering based on species abundance at the genus level using Bray–Curtis distance and relative abundance analysis. (F) Random Forest test for species at the genus level, selecting the top 10 microorganisms as biomarkers. RE, RE group, n = 8, the basal diet supplemented with 20 g RE/d; CON, control group, n = 8, the basal diet without supplementation of RE.

• 

  Fig. 6. Differential Kyoto Encyclopedia of Genes and Genomes (KEGG) functions between the RE group and the CON group. (A) Differential KEGG modules enriched by microorganisms in the rumen. The black font represents the names of different KEGG pathways at level 3, with their affiliation to the KEGG pathway level 2 indicated in gray font on the left. The different-colored blocks beneath the names represent their respective affiliations to KEGG pathway level 1. (B) Differential microbial metabolic pathways identified in the RE and CON groups involved in fatty acid biosynthesis. RE, RE group, n = 8, the basal diet supplemented with 20 g RE/d; CON, control group, n = 8, the basal diet without supplementation of RE. RE = rosemary extract.

• 

  Fig. 7. Effects of rosemary extract (RE) supplementation on the correlation network and differential genera of rumen microbiota in sanhe dairy cows. (A) Correlation network analysis of genus-level microbes identified in the rumen, with different colors representing modules at the phylum level. (B) Volcano plot visualizing the results of rank-sum tests on genus-level microbes in the rumen after filtering (RE group compared to CON group, significance criteria: |FC| ≥ 2 and P < 0.05). Red dots denote differentially expressed metabolites significantly upregulated in the RE group compared to the CON group; blue dots denote those significantly downregulated; grey dots denote metabolites showing no significant change between the two groups. (C) Cluster heatmap analysis of genus-level differentially abundant microbes in the rumen. RE, RE group, n = 8, the basal diet supplemented with 20 g RE/d; CON, control group, n = 8, the basal diet without supplementation of RE. Sig.-Up = significantly up – regulated; Sig.-Down: significantly down – regulated; Ns(N): not significant; RE = rosemary extract; FC = fold change.

• 

  Fig. 8. Correlation network between differentially abundant microbes at the genus level in the rumen and differentially regulated metabolites from rumen, serum, and milk metabolomics. (A) Correlation network analysis between genus-level differentially abundant microbes and differentially regulated metabolites in the rumen. (B) Correlation network analysis between genus-level differentially abundant microbes in the rumen and differentially regulated metabolites in serum. (C) Correlation network analysis between genus-level differentially abundant microbes in the rumen and differentially regulated metabolites in milk. Each displayed node (colored circle) represents a genus or a differentially regulated metabolite, with larger node areas indicating more connections with others. The thickness of the lines connecting nodes represents the strength of correlation. The criteria for correlation with others are a Spearman correlation coefficient (r) > 0.7 and P < 0.05. RE, RE group, n = 8, the basal diet supplemented with 20 g RE/d; CON, control group, n = 8, the basal diet without supplementation of RE. RE = rosemary extract.

• 

  Fig. 9. Correlation analysis of key microbiota with differential indicators. Biomarkers selected in the RE group are highlighted in red, those in the CON group in blue, and the top ten genera in abundance in the rumen in yellow. (A) Spearman's correlation analysis of key microbiota with shared differential microbiota. After pairwise comparisons between the RE and CON groups in the three body fluids, the x-axis represents shared differential microbiota between two body fluids. The left side shows the correlation between the abundance of shared differential microbiota in the rumen, serum, or milk and the abundance of key microbiota. The right side shows the correlation between the abundance of shared differential microbiota in serum and the rumen or milk and the abundance of key microbiota. (B) Mantel's and Pearson's correlation analyses of key microbiota with apparent differential indicators. Mantel's correlation tests were used to determine the correlation between indicators with significant differences in the rumen, serum, and milk between the RE and CON groups and key microbiota. The thickness of the lines represents the correlation coefficient, with bold lines indicating Mantel's r ≥…