Chronic hepatitis B infection (CHB) remains a global public health problem. According to the World Health Organization, approximately 2 billion people worldwide have been infected with the hepatitis B virus (HBV), of whom 254 million people are chronically infected [1]. Annually, about 1.1 million deaths are attributable to HBV-related complications, including liver failure, cirrhosis, and hepatocellular carcinoma. In China, the “Guidelines for the Prevention and Treatment of Chronic Hepatitis B (2022 Edition)” reported that HBV affects approximately 6.1% of the general population, corresponding to roughly 86 million cases of HBV infection, including 20–30 million cases of CHB [2].

HBV-specific CD8+ T cells play a crucial role in viral clearance [3]. However, in patients with CHB, the immune system fails to eliminate the virus effectively. The prevailing evidence suggests that HBV-specific CD8+ T cells exist in a state of prolonged immune activation in CHB patients, leading to their dysfunction and exhaustion [4]. Individual patients exhibit heterogeneous HBV peptide-specific CD8+ T cell responses, and this variability has been attributed to multiple factors, including differences in epitopes and antigen-presenting cells (APCs) [5]. The responsiveness of T cells for specific HBV epitopes also differs markedly. For instance, relative to HBV pol455–463 epitope-specific CD8+ T cells, HBV core 18–27 epitope-specific CD8+ T cells exhibit higher frequencies and more robust immune functionality, including higher levels of IFN$\gamma$, TNF-$\alpha$, and CD107a expression [6]. Although there is also variability in the immune responsiveness of individuals with the different human leukocyte antigen (HLA) subtypes [7], this fails to account for the differential immune responses observed among individuals with the same HLA subtype.

In an effort to more fully clarify the immunological basis for inconsistent HBV-specific immunity, we hypothesized that specific CD8+ T cells recognizing a single peptide-major histocompatibility complex (pMHC) complex may differ in their responsiveness such that they can be classified based on their activity into subsets with effective, inhibitory, and hyporeactive responsivity. T cell repertoire heterogeneity is currently considered to be the principal driver of observed differences in viral clearance, whereas T cell exhaustion is thought to be a relatively minor contributing factor. Differences in T cell reactivity are epitope-specific. For example, in the context of cytomegalovirus (CMV) infection, a diverse T cell repertoire enables robust and effective immunity, whereas in CHB infection, the repertoire is often dominated by hyporeactive or inhibitory T cell subsets, leading to an impaired immune response and poor viral clearance [7].

Research focused on the tumor microenvironment has informed the classification of hyporeactive T cells into three categories: exhausted T cells, arising from persistent antigen stimulation; anergic T cells, which lack the first or second signals; and senescent T cells, which are functionally intact but possess limited replicative potential [8]. The common defining characteristic of these three subsets is low proliferative capacity. However, further research is warranted to investigate the presence of these types of T cells in CHB and to clarify how they influence the immune response. As such, proliferation was selected as a core criterion to evaluate T cell specificity in the present study based on analyses of the responsiveness of T cells specific to a single pMHC.

Prior studies of peptide-specific CD8+ T cells have primarily relied on the use of pMHC tetramers or multimers for detection. However, due to low precursor frequencies, these antigen-specific CD8+ T cells are often undetectable or are inaccurately quantified. Although tetramer-associated magnetic enrichment (TAME) has been developed to improve detection accuracy [9], it may still fail to detect some low-affinity cells. Given this limitation, pMHC tetramer staining was used in this study not as a definitive indicator of T cell specificity, but rather as a means of assessing relative pMHC-TCR affinity.

The primary aims of this study were to investigate the differentiation of HBV-specific CD8+ T cells in CHB, and to examine the functional heterogeneity of clonotypes targeting a single pMHC. Through an in vitro culture approach, proliferated T cells were identified as specific CD8+ T cells, enabling the further assessment of their functional profiles. Additionally, tetramer staining was used to identify highly reactive specific CD8+ T cells. The relationship between pMHC-TCR affinity and the differentiation of specific CD8+ T cells was investigated, based on the hypothesis that viral persistence in CHB may be driven in part by the absence of hyporeactivity of specific CD8+ T cells in certain patients. Based on our previous identification of an inhibitory CD8+ T cell subpopulation characterized by high expression of membrane-bound TGF$\beta$ (mTGF$\beta$) among alloreactive T cells [10], we speculated that a similar mTGF$\beta$+ subpopulation may exist in some HBV patients. Accordingly, in this study, T cells expressing IFN$\gamma$ were identified as the effective subset (IFN$\gamma$+), while those expressing mTGF$\beta$ were classified as the inhibitory subset (mTGF$\beta$+), and those expressing neither of these markers were classified as the hyporeactive subset (IFN$\gamma$–/mTGF$\beta$–).

CHB infection provides a valuable natural model for studying specific T cell responses. Developing a deeper understanding of the functional diversity among specific CD8+ T cells will help elucidate the mechanisms of viral persistence while also informing the development of novel T cell-based immunotherapies.

Some controversy persists regarding the definition of specific CD8+ T cells. Previous studies have reported a correlation between total T cells and HBV-specific CD8+ T cells in terms of phenotypic and functional characteristics, with total T cells also reflecting the clinical severity of HBV infection [20]. These findings suggest that some specific CD8+ T cells may not yet have been fully characterized. In our study, Tet-HBC+ T cell subsets were detected among proliferated T cells from 21 out of 60 individuals, including 6/16 in the IT group, 11/30 in the IC group, and 4/14 in the R group. Based on previous studies [21, 22], we propose that genetic factors such as HLAA2 polymorphisms, rather than disease status, contribute to the limited detection efficiency of specific CD8+ T cells using a tetramer-based approach. Moreover, a substantial proportion of HBV-specific CD8+ T cells in the IT group exhibited low affinity and diminished immune function, which further limits their detection via conventional tetramer staining. In this study, specific CD8+ T cells were defined through in vitro expansion following peptide stimulation. Although expanded cells show some changes in their phenotype and functionality [23], this approach enabled the detection of a broader spectrum of functional phenotypes, particularly for inhibitory subsets of T cells that are often overlooked in the context of tetramer staining. This provides a valuable complementary strategy for the comprehensive profiling of specific CD8+ T cells, as well as a supplemental strategy that can improve results obtained through tetramer detection.

The HBV core 18–27 epitope is a classical target of HLA-A*0201-restricted CD8+ T-cell responses in HBV infection. However, in Asian populations, HBV genotypes B and C almost exclusively carry this V27I mutation (FLPSDFFPSI) [21], which impairs peptide binding to HLA-A*0201 [24]. Despite this reduced affinity, our data confirmed that both wild-type and variant peptides induce T-cell proliferation and could be detected by HBV core 18–27 Tetramer (Supplementary Fig. 2A), and V27-primed T cells cross-recognize the I27 variant (Supplementary Fig. 3A). These findings are consistent with previous studies demonstrating cross-reactivity between V27 and I27 epitopes [25], suggesting that TCR affinity, rather than MHC binding, dominates the functional response.

Hyporeactive T cells have been conventionally classified into three categories: exhausted T cells exhibiting CD45RO+/CD57+/CD95+/PD-1+/CTLA-4+/Tim-3+/LAG-3+/BTLA+ expression; anergic T cells expressing LAG-3+/PD-1+; and senescent T cells expressing KLRG1+/CD28-/CD57+ [8]. These cells generally exhibit limited proliferative capacity [26]. Although in vitro expansion may partially select against such hypoproliferative cells, some reactive subsets with a distinct expression pattern could nonetheless be detected (Supplementary Fig. 4). In this study, by leveraging the natural model of CHB infection, HBV-specific CD8+ T cells were classified into three subsets: an effector subset (IFN$\gamma$+), an inhibitory subset (mTGF$\beta$+), and a hyporeactive subset (IFN$\gamma$–/mTGF$\beta$–), each demonstrating unique transcriptional and functional profiles.

In this study, the effector subset exhibited elevated expression of annotated genes associated with the cytokine-cytokine receptor interaction pathway, consistent with the functional capacity of these cells. Notably, both PDCD1 and FOXP3 were highly expressed in the effector subset, whereas they were expressed at lower levels in the hyporeactive and inhibitory subsets, suggesting that in the context of HBV-specific CD8+ T cells, these markers may not primarily serve as indicators of inhibition, consistent with the recent results from Luo et al. [27]. The hyporeactive IFN$\gamma$–/mTGF$\beta$– subset, in contrast, displayed elevated expression of ENTPD1, CD69, and CD160, consistent with an altered activation state after stimulation. CD160 has been shown to inhibit IL-2 secretion through herpes virus entry mediator (HVEM) [28]. Concurrently, this subset also presented with the upregulation of STAT6, which has been reported to affect the differentiation of CD8+ T cells from Stem-like/Progenitor exhausted T cells (Tprog cells) to Terminally differentiated exhausted T cells (Tterm cells) by antagonizing the TGF$\beta$-BCL6 and IL-2-BLIMP-1 signaling axis [29]. TCF7 was exclusively highly expressed in the IFN$\gamma$–/mTGF$\beta$– subset of CD8+ T cells. In a previous study, TCF7 was shown to directly bind and activate the enhancer regions of EOMES, Bcl-2, and Wnt target genes, suppressing exhaustion-related genes (PDCD1, TOX) while promoting the expression of memory-related genes [30].

As most cells within this subset were negative for tetramer detection, they have likely been overlooked in prior tetramer-based studies focused on specific CD8+ T cells. Collectively, these results suggest that the IFN$\gamma$–/mTGF$\beta$– subset identified here represents a unique population distinct from the conventionally defined exhausted subpopulation, exhibiting properties of both hyporesponsiveness and memory potential. In addition, the ratio of the IFN$\gamma$–/mTGF$\beta$– subset was correlated with HBV viral load in our study, suggesting that there may be other regulatory mechanisms that influence the behavior of this hyporeactive subset and warrant further investigation.

TGF$\beta$ is an important immunomodulatory molecule involved in maintaining immune tolerance, promoting Th17 cell differentiation, and regulating immune homeostasis [31]. Membrane-bound TGF$\beta$ exhibits similar biological activity [32]. It is presented on the cell surface in the form of a latent related peptide (LAP)-TGF$\beta$ complex by binding to latent TGF$\beta$ binding protein (LTBP). It exerts its biological functions via matrix metalloproteinase (MMP) cleavage, integrin-mediated mechanical activation, exposure to an acidic environment or reactive oxygen species, and intercellular contact. Consistent with our previous studies [10], neutralization of TGF-$\beta$ signaling using a specific antibody led to a significant increase in IFN$\gamma$ production in the present study. It has been reported that TGF$\beta$ can upregulate IL-6, activate STAT3, in addition to upregulating VEGF and IL-10, thereby suppressing anti-tumor activity [33]. We also found that the upregulation of the gene encoding TGF$\beta$ was accompanied by the upregulation of IL6. Given its potent immunosuppressive activity and frequent co-expression with other inhibitory molecules [34], mTGF$\beta$ was used as a surface marker in this study to identify the inhibitory subset of CD8+ T cells. Although the upstream regulators of mTGF$\beta$ expression remain incompletely defined, its functional significance supports its utility as a reliable phenotypic marker for an inhibitory subset of cells.

The post-proliferation frequency of specific CD8+ T cells is related to their precursor frequency, as the proliferative capacity (amplification index) remains consistent across different epitopes [35]. In this study, the post-proliferation frequency of HBV core-specific CD8+ T cells did not correlate with the clinical outcome of hepatitis B infection. The variation in precursor frequency of specific CD8+ T cells among individuals is likely attributable to inherent differences in their T cell repertoire. Age has been identified as a contributing factor associated with the precursor frequency of IFN$\gamma$+ peptide-specific CD8+ T cells relative to HBsAg concentration, with a significant decrease in the frequency of HBV env-specific CD8+ T cells after 30 years of age [36].

Studies have also shown that specific T cells responsive to different antigenic epitopes have different reactivity [5]. Recent evidence has further indicated that antigen types influence the aging trajectory of memory T cells [37]. Our findings extend this concept by demonstrating that even among specific T cells responsive to a single epitope, considerable functional heterogeneity exists. Notably, we observed no differences in the proliferation or function of specific CD8+ T cells between the HBC and HBC-PSI groups, suggesting that this particular viral mutation does not substantially alter the phenotype of specific CD8+ T cells, which is consistent with the findings of Hoogeveen et al. [6]. We hypothesize that it is the T cell repertoire in different individuals, rather than any viral mutation, that determines the clinical outcome of HBV infection. However, when combined with different HLA alleles, this mutation will also lead to the impairment of T cell reactivity [38]. This factor should be carefully considered in the application of T cell therapy.

The recognition of the TCRs of specific CD8+ T cells is not merely specific to a peptide, but to an intact pMHC complex [17, 39]. This affinity between pMHC complex and TCR serves as a critical indicator of the strength of immune responses against a given antigen, and is influenced collectively by antigen epitopes, MHC type, TCR sequence and expression levels, costimulatory molecules, and the TCR signaling pathway [40, 41]. Recent studies have analyzed the phenotypes and functions of T cells associated with different clinical outcomes by single-cell sequencing and found that there are more exhausted CD8+ T cells in immune-activated and acute convalescent patients [19]. Our single-cell sequencing results focused specifically on proliferating CD8+ T cells also showed that TCR diversity was significantly reduced in the effector subset (IFN$\gamma$+) compared to the inhibitory subset (mTGF$\beta$+) and the hyporeactive subset (IFN$\gamma$–/mTGF$\beta$–). A parallel trend was observed in terms of the expression of TCR signaling pathway-related genes. Furthermore, both the tetramer assay and ERGO-II model predictions supported a higher degree of affinity in the effector subset. Notably, higher pMHC-TCR affinity does not invariably correlate with improved functionality. Recent studies have shown that in tetramer-positive cells, TIM3 and TOX, which are indicative of hyporeactivity, were highly expressed in the highest affinity group [42]. Together, these results demonstrate that the TCR-pMHC affinity serves as a key determinant of the function of these three different subgroups of CD8+ T cells.

Of note, we detected a proportion of TRGV9⁺ and TRDV2⁺$\gamma$$\delta$ T cells following in vitro expansion of specific CD8⁺ T cells. This observation is consistent with previous reports that $\gamma$$\delta$ T cells can non-specifically proliferate in response to cytokines such as IL2 during in vitro culture, even in the absence of specific antigen stimulation [43, 44]. This phenomenon is particularly prominent in cultures enriched for innate-like T cells, where $\gamma$$\delta$ T cells, including a distinct CD8$\alpha$$\beta$⁺ subset, exhibit robust proliferation driven by shared signaling pathways [45]. Mechanistically, activated $\gamma$$\delta$ T cells can function as antigen-presenting cells ($\gamma$$\delta$ TAPCs) and secrete IL4/IL12, thereby directly supporting the expansion and effector differentiation of CD8⁺$\alpha$$\beta$ T cells. Thus, the detection of TRGV/TRDV transcripts does not reflect non-specific contamination but rather indicates a physiologically relevant, cytokine-driven co-expansion that may contribute to the overall potency of the cultured T cell product.

Serum conversion of HBeAg is a clinically important prognostic marker in hepatitis B infection. In CHB patients, HBeAg– individuals exhibit enhanced functionality of HBV core-specific CD8+ T cells as compared to HBeAg+ individuals [6]. Studies employing ELISPOT assays after stimulation with HBV peptide pools have demonstrated that IFN$\gamma$ levels are inversely correlated with HBV DNA and HBsAg concentrations [46]. Some investigators have categorized CHB into HBs^lo (HBsAg 500 IU/mL) and HBs^hi (HBsAg $>$50,000 IU/mL) cases. They further found that the frequency of IFN$\gamma$+/TNF$\alpha$+ HBs^lo HBV-specific CD8+ T cells was 85%, and the frequency of IFN$\gamma$+/IL-2+ CD4+ T cells was 60%, whereas the corresponding frequency for HBs^hi cells were only 33% and 13%, respectively [47]. Our findings align with these observations. The IFN$\gamma$+/mTGF$\beta$+ ratio was used as a functional indicator of the effector and inhibitory subsets of HBV-specific CD8+ T cells. The IFN$\gamma$/mTGF$\beta$ ratio of HBV-specific CD8+ T cells was correlated with the clinical outcome of HBV infection, HBV DNA level, and HBsAg level. We also noted that for the same patients in the R group, the IFN$\gamma$+/mTGF$\beta$+ ratio varied widely among individuals, and the correlation between HBsAg concentration and the specific T cell IFN$\gamma$+/mTGF$\beta$+ ratio was unclear. More precise indicators are thus needed for use in this context, with HBcrAg offering promise [48].

Although the TCR-pMHC affinity of specific CD8+ T cells cannot be altered, it can be modified through immune-focused interventions such as therapeutic vaccines, PD-1 blockade, metabolic interventions, TLR antagonists, TCR-T, CAR-T, or soluble TCRS strategies [15, 49, 50, 51, 52], offering potential opportunities to redirect T cell clones such that they differentiate into effector subsets. Hypoxic tumor microenvironments have been reported to promote the generation of CD39+ T cells, and microenvironmental modulation can alter the immunosuppressive status of these exhausted T cells [53]. Similarly, a recombinant fusion protein including IL-10 and the IgG Fc region has been shown to activate the pyruvate/MPC pathway within terminally depleted T cells by reprogramming cellular metabolism, thereby enhancing oxidative phosphorylation levels and the anti-tumor efficacy of these immune cells [54]. In vitro metabolic intervention with MitoQ has also been shown to alter the ratio of effector/inhibitory subsets of specific CD8+ T cells [55]. Wang et al. [10] previously confirmed the efficacy of metabolic interventions by analyzing HBV-specific T cell responses in CHB patients with different disease outcomes, including mitochondria-targeted antioxidants (Mito-TEMPO and MitoQ), polyphenol compounds (resveratrol + oleuropein), and cholesterol acyltransferase inhibitors (AVASIMIBE + K-604), which regulate HBV-specific T cell function in patients with CHB [56]. Notably, these metabolic interventions did not show a substantial influence on HBV core-specific CD8+ T cells despite significantly enhancing IFN$\gamma$ expression among HBV env-specific CD8+ T cells. This differential responsiveness underscores the epitope-dependent nature of metabolic modulation in specific CD8+ T cells.

Several limitations should be acknowledged in the present study. First, this was an observational study lacking any strict control over potential confounding variables, which may have introduced considerable bias. As such, our conclusions require further validation through more rigorous clinical investigation, including prospective studies. Second, HLA-A2 subtypes were not determined for the study participants, highlighting another factor that should be taken into account in future research. Third, although healthy individuals were initially intended as controls, the precursor frequency of HBV-specific T cells was extremely low in this population, making them difficult to detect even after in vitro expansion; we therefore used convalescent patients as a surrogate for individuals with a natural immune response, representing another methodological limitation. Finally, the present study sought to demonstrate that CHB infection is mainly driven by intrinsic interindividual differences in the T cell repertoire, rather than by T cell exhaustion. However, the current evidence is not sufficiently robust to support this conclusion, and in future studies, T cells from HBV-naïve children may be used as controls to further verify this hypothesis.

This study explored the differentiation of T cell subsets and the association of these subsets with TCR-pMHC affinity in the context of CHB infection. Overall, our results support a model in which functionally distinct immune responses can be induced by single peptide-specific CD8+ T cells. These findings provide a theoretical basis to inform the further development of T cell-based immunotherapies. Future studies should focus on approaches to modulating the balance between effector and inhibitory CD8+ T cell subsets through in vitro interventions, thereby enhancing antiviral immune efficacy.

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors. The NGS datasets presented in this study can be found in the NCBI SRA BioProject, accession No.: SUB14183699.

Conceive and design the experiments: XWW, XFW, SL; perform the experiments: ZS, XW, JZ; collect the clinical samples: ZS, LO; analyze the data: MW, TZ, LO; prepare the manuscript: ZS. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was carried out in accordance with the guidelines of the Declaration of Helsinki. It was approved by the Ethics Committee of Hubei No.3 People’s Hospital (Approval No.: BA2025002). All the samples used in the experiment were from discarded samples after clinical examination and applied for exemption from informed consent.

The authors would like to thank director Yuqi Qu in Basic Scientific Research Reagent Department of MBL for kind provision of the expansion protocol of peptide-specific CD8+ T cells. We also thank Dr. Chen Chen in Department of Breast and Thyroid Surgery of Union Hospital of Huazhong University of Science for her extensive revision of the manuscript.

This work was supported by the National Natural Science Foundation, China (81871226, 32170920) and the Research Foundation of the Health Commission of Hubei Province Scientific Research Project, China (WJ2021F134).

The authors declare no conflicts of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/FBL51459.