Molecular characterization of gut microbiome in weaning pigs supplemented with multi-strain probiotics using metagenomic, culturomic, and metabolomic approaches

Abstract

Background: Probiotics have been reported to exhibit positive effects on host health, including improved intestinal barrier function, preventing pathogenic infection, and promoting nutrient digestion efficiency. These internal changes are reflected to the fecal microbiota composition and, bacterial metabolites production. In accordance, the application of probiotics has been broadened to industrial animals, including swine, which makes people to pursue better knowledge of the correlation between changes in the fecal microbiota and metabolites. Therefore, this study evaluated the effect of multi-strain probiotics (MSP) supplementation to piglets utilizing multiomics analytical approaches including metagenomics, culturomics, and metabolomics.

Results: Six-week-old piglets were supplemented with MSP composed of Lactobacillus isolated from the feces of healthy piglets. To examine the effect of MSP supplement, piglets of the same age were selected and divided into two groups; one with MSP supplement (MSP group) and the other one without MSP supplement (Control group). MSP feeding altered the composition of the fecal microbiota, as demonstrated by metagenomics analysis. The abundance of commensal Lactobacillus was increased by 2.39%, while Clostridium was decreased, which revealed the similar pattern to the culturomic approach. Next, we investigated the microbial metabolite profiles, specifically SCFAs using HPLC-MS/MS and others using GC-MS, respectively. MSP supplement elevated the abundance of amino acids, including valine, isoleucine and proline as well as the concentration of acetic acid. According to the correlation analyses, these alterations were found out to be crucial in energy synthesizing metabolism, such as branched-chain amino acid (BCAA) metabolism and coenzyme A biosynthesis. Furthermore, we isolated commensal Lactobacillus strains enriched by MSP supplement, and analyzed the metabolites and evaluated the functional improvement, related to tight junction from intestinal porcine enterocyte cell line (IPEC-J2).

Conclusions: In conclusion, MSP administration to piglets altered their fecal microbiota, by enriching commensal Lactobacillus strains. This change contributed amino acid, acetic acid, and BCAA concentrations to be increased, and energy metabolism pathway was also increased at in vivo and in vitro. These changes produced by MSP supplement suggests the correlation between the various physiological energy metabolism functions induced by health-promoting Lactobacillus and the growth performance of piglets.

Keywords: Gut microbiome; Metabolite; Multi-strain probiotics; Multiomics; Weaning pigs.

Fig. 1

Comparison of microbial population of piglet feces within MSP and control group. A Quantitative analysis of total aerobic bacteria between MSP and control groups, B yeast/mold, C coliform, D lactic acid bacteria and E Bifidobacterium (**p < 0.01)

Fig. 2

Microbial Characterization of piglet fecal microbiota utilizing metagenomics. Differences in microbial communities in MSP and control group at 6 weeks (n = 9 per group). Violin plots reflect differences in bacterial diversity in fecal microbiota according to the Shannon index and Chao richness. A Shannon diversity representing alpha diversity and B Chao diversity. C, D Nonmetric multidimensional scaling analysis of weighted and unweighted UniFrac distance. Relative abundance of 10 dominant OTUs in MSP group piglet fecal microbiota at family level E and genus level F. Relative abundance of 10 dominant OTUs in control group piglet fecal microbiota at family level G and genus level H

Fig. 3

Microbial characterization of MSP group utilizing culturomics and comparison analysis with metagenomics. A Identification of MSP group piglet fecal microbiota through culturomics approaches. B 10 dominant bacteria genus of MSP piglet fecal microbiota in culturomics approaches. C bacteria species. D Composition of Lactobacillus spp. detected from MSP piglet fecal microbiota through culturomics approaches. E Relative abundance of OTUs identifying Lactobacillus spp. F Venn diagram depicting the number of bacterial species identified by culturomics (yellow) and metagenomics (purple), along with the proportion of species detected by both approaches (gray)

Fig. 4

Difference concentration and variation of SCFAs and metabolites compared with MSP and control piglets. Metabolomic analysis of piglet feces at 6 weeks. A SCFA profiling using HPLC–MS/MS. B The results of the volcano plot analysis showed that 7 peak pairs (FC > 1, p ≤ 0.05) were increased, whereas 3 peak pairs (FC < 1, p ≤ 0.05) were reduced. All metabolites that were positively and putatively identified with high confidence were matched using the NIST library. C Principal component analysis (PCA) score plots of metabolites in the feces of piglets between the MSP and control groups. D Selected metabolite and bacterial boxplots showed differences between the MSP and control groups. Mean (SD) values of each group are provided by Student’s t test. Median (IQR) values are provided by the Mann–Whitney U test

Fig. 5

Correlation analysis between fecal microbiota and metabolites. The depiction of similarity analysis and SparCC correlation analysis results. A PA, the length of the lines connecting two locations reflects how well two datasets samples agree. The correlation coefficient R ranges from 0 to 1, and the closer it is to 1, the more similar the two datasets are. B Correlation between the functional metabolites and fecal microbiota. The pink nodes represent the functional metabolites and the other nodes of bacteria were grouped with different colors (functional false discovery rate (FDR) < 0.05). C SparCC correlations between bacteria and metabolites are depicted in a Circos plot. Red or green lines indicate positive or negative correlations. D Heatmap represent metabolites concentrations correlated with each group

Fig. 6

Metabolic pathway enrichment analysis was performed on differential metabolites. A KEGG pathways predicted in the fecal microbiota of the control and MSP groups using PICRUSt. Statistical analysis was carried out using STAMP software. B Differential metabolic pathways depicted up-regulated or down re-regulated metabolites involved in each metabolic pathway. p-value < 0.05 was considered statistically significant

Fig. 7

Functional analysis of commensal Lactobacillus strains enriched by MSP supplementation. A Heatmap provides intuitive visualization of metabolites. Each colored cell on the map corresponds to a concentration. B The sparse PLS-DA (sPLS-DA) algorithm represent selected metabolites for given component. Metabolites are ranked by the absolute values of their loadings. C–E Tight junction gene expression of enriched commensal Lactobacillus strains by MSP

Fig. 8

Schematic diagram. Schematic showing the results of integrated metagenomic, culturomics, and metabolomic analyses of the beneficial roles of MSP in weaning pigs. MSP increased the relative abundance of commensal Lactobacillus strains as well as improvement of BCAA production and energy metabolism. Furthermore, enhanced concentration of BCAA coupled with intestinal mucosal barrier integrity are considered to benefit piglets digest nutrients more efficiently, resulting in improved growth performance