Probiotic Lactobacillus casei Shirota prevents acute liver injury by reshaping the gut microbiota to alleviate excessive inflammation and metabolic disorders

Abstract

Millions of people die from liver diseases annually, and liver failure is one of the three major outcomes of liver disease. The gut microbiota plays a crucial role in liver diseases. This study aimed to explore the effects of Lactobacillus casei strain Shirota (LcS), a probiotics used widely around the world, on acute liver injury (ALI), as well as the underlying mechanism. Sprague Dawley rats were intragastrically administered LcS suspensions or placebo once daily for 7 days before induction of ALI by intraperitoneal injection of D-galactosamine (D-GalN). Histopathological examination and assessments of liver biochemical markers, inflammatory cytokines, and the gut microbiota, metabolome and transcriptome were conducted. Our results showed that pretreatment with LcS reduced hepatic and intestinal damage and reduced the elevation of serum gamma-glutamyltranspeptidase (GGT), total bile acids, IL-5, IL-10, G-CSF and RANTES. The analysis of the gut microbiota, metabolome and transcriptome showed that LcS lowered the ratio of Firmicutes to Bacteroidetes; reduced the enrichment of metabolites such as chenodeoxycholic acid, deoxycholic acid, lithocholic acid, d-talose and N-acetyl-glucosamine, reduce the depletion of d-glucose and l-methionine; and alleviated the downregulation of retinol metabolism and PPAR signalling and the upregulation of the pyruvate metabolism pathway in the liver. These results indicate the promising prospect of using LcS for the treatment of liver diseases, particularly ALI.

Fig. 1

LcS alleviated the hepatic dysfunction, excessive inflammatory response and hepatic and intestinal damage induced by D‐GalN. A. Concentrations of TBA and GGT in serum samples. B. Concentrations of IL‐5, IL‐10, G‐CSF and RANTES in plasma samples. C. HAI score and intestinal injury score based on eight samples in each group. D. Representative images of liver samples stained by H&E and immunohistochemistry. E. Representative images of the terminal ileum samples stained by H&E and the intestinal villus ultrastructure under scanning electron microscopy. (*P < 0.05; **P < 0.01; ***P < 0.001).

Fig. 2

Surveys of the 16S rDNA gene revealed that LcS tended to alleviate D‐GalN‐induced alterations in the gut microbiota. A. Box plot of species richness and flora diversities estimated by the Chao1 index and the Shannon index, respectively. B. Two‐dimensional PCoA plot and NMDS analysis based on an unweighted UniFrac matrix. C. Alterations in the relative abundance of bacterial taxa between the LcS, PC and HC groups. D. The ratio of Firmicutes to Bacteroidetes. (*P _adj < 0.05; **P _adj < 0.01; ***P _adj < 0.001).

Fig. 3

LcS alleviated D‐GalN‐induced metabolic disorder. A. OPLS‐DA illustrated that the metabolic profiles of the LcS, PC and HC groups were clearly separated from each other. B and C. The relative concentrations of thirteen gut metabolites were different between the LcS, PC and HC groups. (*P _adj < 0.05; **P _adj < 0.01; ***P _adj < 0.001).

Fig. 4

LcS partially reverses transcriptional regulatory changes. A. Eighty‐two hepatic genes were differentially transcribed in both HC versus PC and LcS versus PC. Among them, LcS reversed the upregulation of 28 genes and the downregulation of 54 genes induced by D‐GalN. B. Thirty‐two gut genes were differentially transcribed in both HC versus PC and LcS versus PC. Among them, LcS reversed the upregulation of 17 genes and the downregulation of 15 genes induced by D‐GalN. (*P _adj < 0.05; **P _adj < 0.01; ***P _adj < 0.001).

Fig. 5

Function classification and pathway analysis of the differentially transcribed genes. A. Enrichment networks of the hepatic and gut genes whose differential transcription was caused by D‐GalN or LcS based on GO functional classification and clustering analysis. All the enriched functional sets (P _adj < 0.05) are represented by different nodes and clustered by colour. The pairs of nodes are connected by their Kappa similarities. B. KEGG enrichment analysis of hepatic and gut genes whose differential transcription was caused by D‐GalN or LcS. The number of upregulated genes is shown above the zero axis, and the number of downregulated genes is shown below the zero axis. Enrichment significance in LcS versus PC: *P _adj < 0.05, **P _adj < 0.01, ***P _adj < 0.001; enrichment significance in HC versus PC: #P _adj < 0.05, ##P _adj < 0.01, ###P _adj < 0.001.

Fig. 6

Associations between significantly altered variables from liver functions, inflammatory responses, and the gut microbiome, metabolome and transcriptome. A. Correlations of altered gut microbes and altered gut metabolites. B. Correlations of altered gut microbes and metabolites with altered liver function indicators and inflammatory cytokines. C. Correlations of altered gut microbes and metabolites with altered hepatic genes. D. Correlations of altered gut microbes and metabolites with altered gut genes. E. Cross‐talk of 10 differentially expressed gut genes and their associated 26 hepatic differentially expressed genes. Genes are classified by function and designated by areas of different colours. (*P < 0.05; **P < 0.01; ***P < 0.001).