Alterations in gut microbiome and metabolomics in chronic hepatitis B infection-associated liver disease and their impact on peripheral immune response

Abstract

Gut dysbiosis has been reported in chronic hepatitis B (CHB) infection, however its role in CHB progression and antiviral treatment remains to be clarified. Herein, the present study aimed to characterize gut microbiota (GM) in patients with chronic hepatitis B virus infection-associated liver diseases (HBV-CLD) by combining microbiome with metabolome analyses and to evaluate their effects on peripheral immunity. Fecal samples from HBV-CLD patients (n = 64) and healthy controls (n = 17) were collected for 16s rRNA sequencing. Fecal metabolomics was measured with untargeted liquid chromatography-mass spectrometry in subgroups of 58 subjects. Lineage changes of peripheral blood mononuclear cells (PBMCs) were determined upon exposure to bacterial extracts (BE) from HBV-CLD patients. Integrated analyses of microbiome with metabolome revealed a remarkable shift of gut microbiota and metabolites in HBV-CLD patients, and disease progression and antiviral treatment were found to be two main contributing factors for the shift. Concordant decreases in Turicibacter with 4-hydroxyretinoic acid were detected to be inversely correlated with serum AST levels through host-microbiota-metabolite interaction analysis in cirrhotic patients. Moreover, depletion of E.hallii group with elevated choline was restored in patients with 5-year antiviral treatment. PBMC exposure to BE from non-cirrhotic patients enhanced expansion of T helper 17 cells; however, BE from cirrhotics attenuated T helper 1 cell count. CHB progression and antiviral treatment are two main factors contributing to the compositional shift in microbiome and metabolome of HBV-CLD patients. Peripheral immunity might be an intermediate link in gut microbe-host interplay underlying CHB pathogenesis.

Keywords: Chronic hepatitis B virus infection-associated liver disease; antiviral treatment; cirrhosis; metabolome; microbiome; peripheral immunity.

Graphical abstract

Figure 1

:Distinct fecal microbiota profiles of HBV-CLD patients and HCs. Flow chart of the study design (a); β diversity (based on Bray-Curtis distances) evaluated by PCoA analysis (b); PLS-DA analysis (c). HBV-CLD, chronic hepatitis B virus infection-associated liver diseases; HC, healthy control; HCC, hepatocellular carcinoma; LC-MS, liquid chromatography–mass spectrometry; T2DM, diabetes mellitus type 2.

Figure 2

:Gut microbiota profile in HBV-CLD patients. The abundance of genera was log-transformed with 0 values assigned with 1e-05. Box plots indicate median (middle line), 25th, 75th percentile (box) and maximum and minimum values (whisker). *p_fdr<0.05, **p_fdr<0.01, ***p_fdr<0.001, ****p_fdr<0.0001. HBV-CLD, chronic hepatitis B virus infection-associated liver diseases; HC, healthy control.

Figure 3

:Distinct gut microbiota profile regarding disease progression and antiviral treatment. Genera correlated with disease progression (a); antiviral treatment (b); Venn diagram outlined the genera associated with disease progression and antiviral treatment (c). The abundance of genera was log-transformed with 0 values assigned with 1e-05. Box plots indicate median (middle line), 25th, 75th percentile (box) and maximum and minimum values (whisker). *p_fdr<0.05, **p_fdr<0.01, ***p_fdr<0.001, ****p_fdr<0.0001. CHB, chronic hepatitis B (treatment naive); Crrh, cirrhosis; ETV, entecavir; HC, healthy control; NC, non-cirrhosis.

Figure 4

:Distinct fecal metabolome signature of HBV-CLD patients and HCs. Representative metabolites that were significantly changed in HBV-CLD (a); correlated with disease progression (b); antiviral treatment (c); Venn diagram outlined the metabolites associated with disease progression and antiviral treatment (d). The abundance of metabolites was log-transformed with 0 values assigned with 1e-05. Box plots indicate median (middle line), 25th, 75th percentile (box) and maximum and minimum values (whisker). *p_fdr<0.05, **p_fdr<0.01, ***p_fdr<0.001, ****p_fdr<0.0001. CHB, chronic hepatitis B (treatment naive); Crrh, cirrhosis; ETV, entecavir; HBV-CLD, hepatitis B virus related chronic liver diseases; HC, healthy control; NC, non-cirrhosis.

Figure 5

:The host-microbiota-metabolite interplay in HBV-CLD patients. The heatmap depicts correlation (partial spearman analysis) of disease-related taxa and metabolites (a); disease-related metabolites and clinical indexes (b); disease-related metabolites and clinical indexes (c). *p_fdr<0.05, **p_fdr<0.01, ***p_fdr<0.001. ALB, albumin; ALT, alanine amino transferase; AST, aspartate aminotransferase; GGT, γ-glutamyl transpeptidase; TB, total bilirubin; DB, direct bilirubin; PT, prolonged prothrombin time.

Figure 6

:Exposure of PBMCs to patient-derived BEs resulted in altered T cell subtype counts. Th17 (CD3⁺CD4⁺IL-17A⁺), Th1 (CD3⁺CD4⁺IFN-γ⁺), monocytes (CD3⁻CD19⁻CD14⁺) in response to BE exposure from HCs and HBV-CLD patients (a). Th17 (b), Th1 (c) and monocytes (d) in response to BE exposure from HCs, non-cirrhotic (NC) and cirrhotic (Crrh) patients. Data are shown as % of viable CD3⁺CD4⁺ lymphocytes for Th17 and Th1 cells, and % of viable singlets for monocytes. Box plots indicate individual values for each sample and median with interquartile range within groups. *p < .05, **p < .01, ***p < .001, ****p < .0001. BE, bacterial extracts; Crrh, cirrhosis; HBV-CLD, hepatitis B virus related chronic liver diseases; HC, healthy control; NC, non-cirrhosis; Th1, T helper 1; Th17, T helper 17.