Circulating HBsAg-specific B cells are partially rescued in chronically HBV-infected patients with functional cure

Abstract

It is well established that humoral immunity targeting hepatitis B virus surface antigen (HBsAg) plays a critical role in viral clearance and clinical cure. However, the functional changes in HBsAg-specific B cells before and after achieving functional cure remain poorly understood. In this study, we characterized circulating HBsAg-specific B cells and identified functional shifts and B-cell epitopes directly associated with HBsAg loss. The phenotypes and functions of HBV-specific B cells in patients with chronic HBV infection were investigated using a dual staining method and the ELISpot assay. Epitope mapping was performed to identify B cell epitopes associated with functional cure. Hyperactivated HBsAg-specific B cells in patients who achieved HBsAg loss were composed of enriched resting memory and contracted atypical memory fractions, accompanied by sustained co-expression of multiple inhibitory receptors and increased IL-6 secretion. The frequency of HBsAb-secreting B cells was significantly increased after achieving a functional cure. The rHBsAg displayed a weaker immunomodulatory effect on B cells than rHBeAg and rHBcAg in vitro. Notably, sera from patients with HBsAg loss reacted mainly with peptides S60, S61, and S76, suggesting that these are dominant linear B-cell epitopes relevant for functional cure. Intriguingly, patients reactive with S76 showed a higher frequency of the HLA class II DQB1*05:01 allele. Taken together, HBsAg-specific B cells were partially restored in patients after achieving a functional cure. Functional cure-related epitopes may be promising targets for developing therapeutic vaccines to treat HBV infection and promote functional cure.

Keywords: B cells; B-cell epitope; Hepatitis B surface antigen; functional cure; hepatitis B surface antigen clearance.

Graphical abstract

Figure 1.

Analyses of the frequency of HBsAg-specific B cells in cross-sectional cohorts. (A) Sequential multiparametric flow cytometric gating strategy used to identify HBsAg-specific B cells and their composition. (B) Frequency of HBsAg-specific B cells out of memory B cells in vaccinated healthy controls (HCs) and different phases of chronic HBV infection: 16 HCs, 20 HBeAg-positive chronic infection (HBeAg⁺CInf), 20 with HBeAg⁺ chronic hepatitis (HBeAg⁺CHep), 21 HBeAg⁻CInf, and 17 HBeAg⁻CHep, 24 HBsAg loss, and 22 HBsAg seroconversion (HBsAg SC). HBsAg^pos group included HBeAg⁺CInf, HBeAg⁺CHep, HBeAg⁻CInf, and HBeAg⁻CHep patients, and HBsAg^neg group included HBsAg loss and HBsAg SC subjects. (C) Frequency of different subsets of memory B cells [resting memory (RM), activated memory (AM), and atypical memory (AtM) B cells] among HBsAg-specific B cells in HCs and patients with chronic HBV infection. No correlation between the frequency of HBsAg-specific B cells among memory B cells with serum HBsAg (D left), HBeAg (D middle), HBV DNA (D right), and ALT levels (E). (F) Correlation of frequency of AtM or RM among HBsAg-specific B cells with frequency of HBsAg-specific B cells or serum HBsAg levels, respectively. (B, C) Mann-Whitney U test or Kruskal-Wallis H test and Dunn's multiple comparisons test. (D-F) Spearman's rank correlation test. *P < 0.05, **P < 0.01.

Figure 2.

Molecular features and functionality of HBsAg-specific B cells in patients with chronic HBV infection. The expression of the indicated inhibitory receptors (A), activated molecules (B left), and the production of cytokines (B right) on CD19⁺ global B and HBsAg-specific B cells. Percentages and cytokines production of CD19⁺ global B cells (C) and HBsAg-specific B cells (D) among HCs, HBsAg^pos patients, and HBsAg^neg patients. The color legend in A could also be applied to B. The color legend in C could also be applied to D. (E) Correlation of expression of BTLA on HBsAg-specific B cells with serum HBeAg levels (E left). Spearman correlation between the proportions of IFN-γ secreting HBsAg-specific B cells with serum HBsAg (E middle) and HBV DNA (E right) levels. (A, B) Wilcoxon signed-rank test. (C, D) Kruskal-Wallis H test and Dunn's multiple comparisons test. (E) Spearman's rank correlation test. *P < 0.05, **P < 0.01.

Figure 3.

Antibodies secretion potential of HBV-specific B cells in patients with various phases and HCs. (A) Image showing the results of representative ELISpot assay. (B) Detection of total IgG-secreting B cells (upper left), HBsAb-secreting B cells (upper right), HBeAb-secreting B cells (bottom left), and HBcAb-secreting B cells (bottom right) in chronic HBV infected patients and HCs by ELISpot assay. Correlation of HBcAb-secreting B cells with serum virological (C) and biochemical (D) parameters. (E) Comparing serum HBcAb levels within patients with chronic HBV infection. (B, E) Mann-Whitney U test or Kruskal-Wallis H test and Dunn's multiple comparisons test. (C, D) Spearman's rank correlation test. *P < 0.05, **P < 0.01.

Figure 4.

Phenotypic and functional changes of B cells in different HBV antigens stimulation. (A) Gating strategy for CD19⁺ global B cells and their distribution into memory B cells (CD19⁺CD27⁺CD38⁻) and plasmablasts (CD19⁺CD27⁺CD38⁺) B-cell subsets. (B) The effect of different HBV antigens on the frequency of B-cell subsets (n = 16). The color legend in B could also be applied to C, D, and E. The effect of different HBV antigens on the expression of costimulatory and inhibitory molecules and functional capacity in CD19⁺ B cells (C), memory B cells (D), and plasmablasts (E) (n = 16). (B-E) Friedman test and Dunn's multiple comparisons test. *P < 0.05, **P < 0.01.

Figure 5.

General recognition of linear B-cell epitopes by screening sera from patients with chronic HBV infection and vaccinated HCs. (A) Schematic of the large surface protein: PreS₁, PreS₂, and S. In the diagram of S, the “a determinant region” is in red. Design of peptide array. 15-mer overlapped peptides covering the entire S protein were shown. According to the classical regions of large S protein, these peptides were classified into 8 subpartitions: PreS₁ (S1-31), PreS₁-S₂ (S28-31), PreS₂ (S28-44), PreS₂-S (S41-44), S (S41-98), Pre “a” (S72-75), “a” (S72-80), and Suf “a” (S78-80). (B) ELISA array was performed a screening on B-cell linear epitopes with sera from 13 HCs, 11 HBeAg-positive chronic infection (HBeAg⁺CInf), 13 with HBeAg⁺ chronic hepatitis (HBeAg⁺CHep), 14 HBeAg⁻CInf, and 7 HBsAg⁺HBsAb⁺, 20 HBsAg loss, and 27 HBsAg seroconversion (HBsAg SC). The positive peptides for each subject were plotted (green bar in HCs, blue bar in HBsAg^pos patients, and red bar in HBsAg^neg patients). (C) Comparing subpartition recognition ratio (a set of recognition ratios in the subpartition of each patient) or positive peptide coverage among vaccinated HCs, HBsAg^pos patients, and HBsAg^neg patients. (D) Comparing the recognition ratios of each peptide in HBsAg^pos and HBsAg^neg patients with those in HCs. (C) Kruskal-Wallis H test and Dunn's multiple comparisons test (left) and Chi-square test (right). (D) Chi-square test. *P < 0.05, **P < 0.01.

Figure 6.

Distribution of dominant linear B-cell epitopes on “a determinant region”. (A) The amino acid sequence of “a determinant region”. The grey lines indicate the corresponding peptides. The “a determinant region” was in red. (B) The recognition ratio of each peptide in “a determinant region”. The proportion of patients with detectable HBsAg or HBsAg loss in S76^pos/neg groups. (D) Frequency of HBsAg-specific B cells out of memory B cells and their expression of PD1 in S76^pos and S76^neg patients. Comparison of the frequencies of HLA alleles between patients with HBsAg loss and HBsAg positive (E), and patients with dominant epitope S76 and those without (F). (B, C, E, and F) Chi-square test. (D) Mann-Whitney U test. *P < 0.05, **P < 0.01.