Blood donors were routinely screened between June 2016 and September 2017 at Guangzhou Blood Center for HBsAg and antibodies to hepatitis C virus (HCV), human immunodeficiency virus (HIV), and treponema pallidum by two different enzyme immunoassays (EIAs), respectively. The blood donors were further tested for HBV, HCV, and HIV genomes by nucleic acid testing (NAT) [Procleix Ultrio, individual donation NAT, Grifols Diagnostic Solutions, San Diego, CA, United States; limit of detection (LOD) HBV DNA 10.4 IU/mL] (Tang et al., 2018). The HBsAg and OBI carriers were tested by follow-up for HBsAg, anti-HBs, HBeAg, anti-HBe, and anti-HBc with chemiluminescent microparticle immunoassay (CMIA) in the Abbott Architect i2000SR analyzer (Abbott Diagnostic, Chicago, IL, United States). HBsAg and anti-HBs values <0.05 IU/mL or 10 IU/L, respectively, are considered negative. HBeAg or anti-HBc S/CO value <1 is considered negative, while anti-HBe S/CO value >1 is negative. HBV DNA was extracted from 1 to 2.5 ml of plasma with a large volume High Pure Viral Nucleic Acid Extraction kit (Roche Diagnostics, Mannheim, Germany). Viral load was quantified by real-time PCR (qPCR) as previously described (LOD HBV DNA 5 IU/mL) (Zhang et al., 2018a). OBI carriers were identified by the follow-up detection of different viral genomic regions at a minimum interval of 6 months (Tang et al., 2018; Raimondo et al., 2019).

This study was approved by the Medical Ethics Committee of Guangzhou Blood Center and followed the ethical guidelines of the 1975 Declaration of Helsinki. All study subjects signed an informed consent.

The levels of biochemistry parameters in blood were measured by an automatic biochemistry analyzer (Cobas c 702, Roche Diagnostics, Mannheim, Germany), including alanine transaminase (ALT), aspartate transaminase (AST), total bilirubin (TBIL), direct bilirubin (DBIL), total bile acid (TBA), albumin (ALB), adenosine deaminase (ADA), cholinesterase (CHE), γ-glutamyl transpeptidase (γ-GT), and total protein (TP).

Peripheral blood mononuclear cells (PBMCs) were isolated from EDTA anti-coagulated blood within 4 h of collection by Ficoll-Hypaque density gradient centrifugation (LSM, MP Biomedicals, LLC, United States). Freshly isolated PBMCs were used for testing of T-cell responses to HBV.

Hepatitis B virus core and polymerase (pol) peptides (genotype: B2; GenBank accession number: ACO05370.1) were synthesized commercially, including 9 HLA-I and 8 HLA-II restrictive core epitope peptides and 15 HLA-I and 10 HLA-II restrictive pol epitope peptides (Supplementary Table 1) (Brinck-Jensen et al., 2015; Zhang et al., 2018b; Tan et al., 2019). These core or pol peptides were prepared in a pool with 5 μg/mL of each peptide.

Peripheral blood mononuclear cells were cultured in RPMI 1640 (GIBCO) supplemented with 10% fetal bovine serum (FBS), 1% HEPES, 100 U/mL penicillin, and 100 μg/mL streptomycin.

Freshly isolated PBMCs (1 × 10⁶ per well) were stained with 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE), and then seeded in 24-well round bottom culture plates incubated for 6 days with HBV core/pol peptides (5 μg/mL each). Cells were stained with anti-CD3, anti-CD4, and anti-CD8 fluorescent antibodies (BD Pharmingen), and then analyzed by flow cytometry (BD FACSCanto II, BD Biosciences).

Triplicates of 2 × 10⁵ PBMCs per well were inoculated in 96-well plates pre-coated with anti-IFN-γ monoclonal antibody (Dakewe Biotech Co., Ltd., China), and incubated for 20 h with the pooled HBV core/pol peptides (5 μg/mL each). ELISpot was performed according to the manufacturer’s protocol. The spots were counted using an ELISpot-reader (Bio-Sys GmbH, Karben, Germany) and results were given as the mean number of spot-forming cells (SFCs) per 1 × 10⁶ cells ± SD.

Peripheral blood mononuclear cells (1 × 10⁶ per well) were stimulated by HBV core/pol peptides (5 μg/mL each), and phorbol myristate acetate/ionomycin (PMA; positive control) or buffer control (negative control). After 2 h of incubation, 1 μl Brefeldin A protein transport inhibitor (BD GolgiPlug) was added and continued for incubation with an additional 12 h. At completion of incubation, cells were stained with anti-CD4, anti-CD8, and anti-CD3 antibodies (BD Pharmingen), then fixed and permeabilized. Intracellular cytokines were stained separately for interferon-γ (IFN-γ), tumor necrosis factor α (TNF-α), (interleukin-2) IL-2, IL-17A, IL-21, IL-10, and tumor growth factor-β (TGF-β) (BD Pharmingen). Last, cells were washed and analyzed by flow cytometry.

Peripheral blood mononuclear cells (1 × 10⁶ per well) were stimulated with HBV core/pol peptides, positive and negative controls as above. After 5 days of incubation at 37°C in 5% CO₂, the cytokines IFN-γ, TNF-α, IL-2, IL-17A, IL-21, IL-10, and TGF-β in the supernatants of cell cultures were quantified by a cytometric bead array (CBA, Becton-Dickinson). Data were analyzed with CBA software, and cytokine concentration was determined by comparison with standard curves.

Data were presented as the mean ± SD or the median within a range. The normal distribution of measurement data was used for two independent sample t-tests; multiple comparisons between groups were analyzed by the one-way ANOVA test; the set comparison between groups in non-parametric data was analyzed by the Mann–Whitney U test. Statistically significant differences are indicated with asterisks (*P < 0.05, ^**P < 0.01, and ^***P < 0.001). All graphs are generated with GraphPad Prism 7 software.

A total of 193 blood donors were successfully recruited for participating in the follow-up detection of specific cellular immune response to HBV antigens. Among these participants, 37 donors HBsAg-/HBV DNA+/anti-HBc+ were classified as OBI carriers, 53 HBsAg+ as CHB carriers (1 with anti-HBs+), 47 anti-HBc+ and anti-HBs+ as spontaneously resolved HBV infection, 56 anti-HBs+ only (33 vaccinated) or HBV serological/genomic negative (23 no marker) as HBV non-infected individuals, respectively. None of the OBI or CHB carriers received any antiviral treatment for at least 6 months prior to blood donation. OBI carriers ranged in age between 19 and 55 years and all except one had normal ALT level. OBI carriers were all anti-HBc+ but HBeAg−, and 20 of them (54.1%) were anti-HBs+. Viral load quantified by qPCR ranged between <5 and 548 IU/mL, which was significantly lower than observed in the CHB group (median 81 vs. 8,600 IU/mL, P < 0.001). The clinical information and HBV infection markers from the four groups of blood donors are presented in Supplementary Table 2. There was no significant difference in gender between the groups, while age was significantly higher in OBI than in other groups (42.4 ± 10.8, P < 0.001). Liver function parameters level of TBIL and DBIL were higher in the OBI group than in other groups, but not statistically different (P = 0.203 and P = 0.294, respectively).

The proliferation rates of CD4⁺ and CD8⁺ T lymphocytes stimulated with HBV core peptides were quantified by flow cytometry, which varied significantly between the four groups of blood donors with different HBV infection status (Figures 1A,B; P < 0.01 and P < 0.05). A specific CD4⁺ T cell proliferation rate was higher in OBI carriers (3.0%) and CHB carriers (3.2%) than in donors with resolved HBV infection (2%) or non-infected individuals (1.6%) (OBI vs. resolver or non-infection, P < 0.01; CHB vs. resolver or non-infection, P < 0.001; Figure 1A). Similar profiles were observed for specific CD8⁺ T cell proliferation rates between groups of blood donors, in which a significantly higher proliferation rate was observed in OBI and CHB carriers (both 1.8%) compared with resolved HBV infection (1.5%) and non-infected individuals (1.6%) (OBI vs. resolver or non-infection, P < 0.05; CHB vs. resolver or non-infection, P < 0.05; Figure 1B). The proliferation rates of CD4⁺ or CD8⁺ T cells between OBI and CHB carriers were not statistically different (P > 0.05). The median CD4⁺ and CD8⁺ T cell proliferation rates in each study group indicated that the proliferation rate of CD4⁺ T cells was significantly higher than that of CD8⁺ T cells in all four groups (P < 0.05; Figure 1C).

A specific IFN-γ secretion T-cell response between groups of blood donors with various HBV infection outcomes was found significantly different by ELISpot after stimulation with HBV core peptides (Figure 2A). The number of specific IFN-γ secretion T cells was significantly higher in OBI carriers (25 SFC/10⁶ PBMCs), followed by CHB carriers (20 SFC/10⁶ PBMCs), than in HBV resolved infections (10 SFC/10⁶ PBMCs, P ≤ 0.004) or non-infected individuals (5 SFC/10⁶ PBMCs, P < 0.001), but not statistically different between OBI and CHB carriers (P > 0.05; Figure 2B). Individuals with resolved HBV infection had relatively higher specific IFN-γ-secreting T-cell response than HBV non-infected individuals (P = 0.03; Figure 2B).

The frequency of positive T-cell response for IFN-γ secretion was significantly higher in OBI carriers (64.9%) than in resolved HBV infection (53.2%), or in CHB carriers (45.3%) or non-infected individuals (23.2%), which was statistically different between groups (P < 0.05; Figure 2C).

After stimulation with HBV core peptides, the intracellular cytokines expressing CD4⁺ effective T-cell responses were individually quantified (Figure 3A). The frequency of intracellular IFN-γ (0.26%), IL-17A (0.48%), and IL-21 (0.15%) expressing CD4⁺ T cells was significantly higher in resolved HBV infection than in OBI (0.17, 0.22, and 0.09%) or CHB carriage (0.19, 0.27, and 0.10%), respectively (Figures 3B,E,F; P < 0.05). Any group of HBV infected blood donors had higher intracellular IL-2 expressing CD4⁺ T-cell response than HBV non-infected individuals (Figure 3D; P < 0.05), although there was no statistical difference in TNF-α⁺/CD4⁺ T-cell response between groups (Figure 3C; P > 0.05). Similarly, the frequency of intracellular cytokines expressing CD8⁺ effective T cells was tested (Figure 4A). The frequency of IFN-γ and IL-17A expressing CD8⁺ cells was significantly higher in resolved HBV infections (0.23 and 0.31%) or non-infected individuals (0.21 and 0.29%) than in those with OBI (0.12 and 0.16%) or CHB carriers (0.12 and 0.20%) (Figures 4B,E; P < 0.05), while the frequency of TNF-α, IL-2, and IL-21 expressing CD8⁺ T cells was not significantly different between groups (Figures 4C,D,F; P > 0.05).

The frequency of intracellular IL-10 expressing CD4⁺ suppressive T cells to HBV core peptides was higher in HBV infected individuals with OBI (0.12%), CHB (0.11%), or resolved infection (0.1%), respectively, than in HBV non-infected individuals (0.06%) (Figures 5A,B; P < 0.05). The frequency of intracellular IL-10 expressing CD8+ suppressive T cells was higher in CHB carriers (0.09%) than in HBV non-infected individuals (0.055%), resolved infections (0.06%), or OBI carriers (0.07%) (Figures 5A,C; P < 0.05). The frequency of intracellular TGF-β expressing CD4⁺ and CD8⁺ T cells was significantly higher in CHB carriers (both 0.16%) than in OBI carriers (0.10 and 0.11%) (Figures 5A,D,E; P < 0.05).

Secretion of cytokines IFN-γ, TNF-α, IL-2, IL-17A, IL-21, IL-10, and TGF-β in culture supernatants of PBMCs was quantified after stimulation with HBV core peptides by cytometric bead array (Figure 6A). The level of TNF-α secreted by OBI (4.41 pg/mL) or CHB carriers (3.41 pg/mL) was lower than that obtained in HBV non-infected individuals (7.12 pg/mL) (Figure 6B; P < 0.05). The level of IL-2 was significantly higher in OBI (3.39 pg/mL) or CHB carrier cultures (3.40 pg/mL) than in resolved HBV infection (2.76 pg/mL) or non-infected individuals (2.76 pg/mL) (Figure 6C; P < 0.05). The level of IL-17A observed with OBI (11.82 pg/mL) or CHB carriers (10.12 pg/mL) was higher than with resolved HBV infection (4.57 pg/mL) or non-infected individuals (5.62 pg/mL) (Figure 6D; P < 0.05). The level of IL-21 varied insignificantly between the four groups of blood donors (Figure 6E; P > 0.05). The level of IL-10 in CHB carriers (33.62 pg/mL) was significantly higher than observed in resolved HBV infection (24.47 pg/mL) and non-infected individuals (22.4 pg/mL, P < 0.01), but was not statistically different from OBI carriers (25.2 pg/mL) (Figure 6F; P = 0.355). The level of secreting TGF-β between our four groups of blood donors was not statistically different (Figure 6G; P > 0.05).

By using stimuli of HBV pol peptides, the specific cellular immune response was measured from the above four groups of blood donors with different HBV infection status (Supplementary Figures 1–5). The data showed that proliferation of CD4⁺ and CD8⁺ T cells was higher in OBI and CHB carriers (Supplementary Figure 1), IFN-γ secreting T-cell response was higher in CHB carriers (Supplementary Figure 2), intracellular IFN-γ, IL-17A, and IL-21 in CD4⁺ T-cell responses were lower (Supplementary Figure 3) but IL-10/CD8⁺ T-cell response (Supplementary Figure 4) and IL-10 secretion (Supplementary Figure 5) were higher in CHB carriers. These results were in a manner consistent with the cellular responses to HBV core peptides among these four groups of blood donors with various HBV infections.