ABSTRACT
-
Background/Aims
Natural killer (NK) cell function is generally considered dampened in chronic hepatitis B virus (HBV) infection; however, the NK cell pool exhibits phenotypic and functional heterogeneity, and the antibody-mediated effect of NK cells remains less characterized. This study evaluated the dynamic changes in antibody-mediated NK cell responses and the involvement of distinct NK subsets across disease stages and during antiviral treatment.
-
Methods
A T-cell receptor-like antibody specific for the HBV core 18–27 peptide (cTCRL-Ab) was used to determine the antibody-mediated effect of NK cells, and an array of NK cell surface markers were analyzed in cross-sectional and longitudinal cohorts of patients with chronic HBV infection. Single-cell RNA sequencing (scRNA-seq) was performed to identify the heterogeneity of NK subsets.
-
Results
The cTCRL-Ab enabled the detection of NK cell cytolytic activity and IFNγ production. Notably, cTCRL-Ab-mediated NK cell responses were compromised in chronically HBV-infected patients, particularly in those receiving pegylated interferon-α (Peg-IFNα), which was associated with the downregulation of CD16 expression. Correspondingly, Peg-IFNα inhibited cTCRL-Ab-mediated NK cell function by reducing CD16 expression in vitro. scRNA-seq revealed that CD16 downregulation occurred mainly within a dysfunctional CD16hi NK subset exhibiting exhaustion properties. In contrast, an activated CD16hiNK subpopulation (CX3CR1⁺KLRC2–CD16hi) with high cytotoxicity was enriched in patients who experienced favorable treatment responses. Furthermore, the intrahepatic CX3CR1+KLRC2–CD16hi subset tended to exhibit functional restoration in HBsAg-loss individuals.
-
Conclusions
Our data contribute to the understanding of antibody-mediated responses of NK cells in chronic HBV infection, and highlight a previously unappreciated functional CX3CR1+KLRC2–CD16hiNK subset as a potential therapeutic target.
-
Keywords: Hepatitis B virus; Natural killer cell; Antibody-mediated effect; Single-cell RNA-seq
Study Highlights
• A cTCRL-Ab system enables the detection of antibody-mediated NK cell responses.
• The antibody-mediated effects of NK cells are impaired in chronic HBV infection and further suppressed by Peg-IFNα.
• A cytotoxic CX3CR1+KLRC2–CD16hiNK subset is enriched in patients with favorable treatment response.
• Intrahepatic CX3CR1+KLRC2–CD16hiNK cells are partially functionally restored in individuals with HBsAg-loss.
Graphical Abstract
INTRODUCTION
Given that current treatments for chronic hepatitis B (CHB) rarely achieve a functional cure, the development of combination or alternative therapeutic strategies is urgently needed. The clearance of the hepatitis B virus (HBV) requires well-coordinated innate and adaptive immune responses, and a deficiency in either branch may result in persistent infection [
1]. Thus, elucidating the dysfunctional immune response and the switches toward a functional response will provide novel clues for CHB immunotherapies. The immune system employs a cytolytic pathway to eliminate HBV [
1]. Despite being major mediators of cytotoxicity, natural killer (NK) cells have received less attention in chronic HBV infection compared with the extensive research on T cells [
2].
NK cells are a prominent arm of innate immunity and constitute the primary line of defense against viral infections. They are typically categorized into CD56
dimCD16
hi (CD16
hi) and CD56
brightCD16
dim/– (CD16
lo) subsets. The predominant CD16
hiNK subset is endowed with cytotoxic properties through the expression of activating receptor CD16 and cytolytic granules. In contrast, CD16
loNK cells are mainly involved in the production of cytokines [
3]. NK function is mediated by a natural cellular response (without antibodies) or by antibody-dependent effects [
4,
5]. In antibody-mediated immune responses, CD16 expressed on NK cells binds the Fc region of antibodies immobilized on the target cells, triggering activation signals that induce granule release and inflammatory mediator production to eliminate infected cells [
4,
6]. Such antibody-mediated cytolytic responses facilitate early control and resolution of acute HBV infection [
7]. However, the functional competence of NK cells is compromised through the synergistic interaction of multiple receptors during persistent HBV infection [
8-
10]. Worse still, NK cells exert detrimental effects on HBV-specific T cells via interleukin (IL)-10 production, TRAIL and NKG2D expression, or direct killing [
11-
13]. Of note, previous research has focused primarily on the dynamics of the frequencies and mediators of NK cells either
ex vivo or through
in vitro activation. The lack of antibodies specific for HBV-infected hepatocytes and the laborious detection procedures involved have thus far hindered a thorough investigation into whether NK cells can exert antibody-mediated effects against HBV-infected target cells, and whether this function is similarly compromised.
Accumulating studies have shown that first-line antiviral therapy can restore NK function and facilitate HBV clearance [
14,
15]. Following cessation of long-term nucleos(t)ide analogue treatment, NK natural cytotoxicity responses increase and are associated with HBsAg seroclearance [
16]. Similarly, IFNα treatment expands and activates NK cells, elevating IL-15, TRAIL, and interferon-gamma (IFNγ) expression, and is correlated with favorable virological responses [
14,
15,
17]. However, the dynamics of antibody-mediated NK responses and the extent to which heterogeneous functional NK subsets are reshaped in chronic HBV remain unclear.
Here, we utilized a previously reported T-cell receptor-like antibody specific for HBV core18–27 peptides (cTCRL-Ab) [
18,
19] as a tool for investigating the antibody-mediated effects of NK cells in HBV infection. Utilizing the cTCRL-Ab, we discovered that chronically HBV-infected patients exhibited impaired cTCRL-Ab-mediated NK cell responses and that Peg-IFNα inhibited these effects by reducing CD16 expression in NK cells. In combination with scRNA-seq, these results revealed that subset heterogeneity among NK cells is responsible for antibody-mediated responses. Peg-IFNα–induced CD16 downregulation occurred predominantly within a CD16
hiNK cluster exhibiting exhaustion properties, whereas a functionally activated CD16
hiNK subset (CX3CR1
+KLRC2
–CD16
hiNK cells) exhibited high cytotoxicity and was associated with a favorable treatment response. Notably, this subset tended toward functional restoration in individuals with HBsAg-loss. Thus, our data contribute to the understanding of antibody-mediated responses of NK cells and identify a previously unappreciated functional CX3CR1
+KLRC2
–CD16
hiNK subset in chronic HBV infection.
MATERIALS AND METHODS
Human subjects
A total of 63 treatment-naïve chronic HBV patients and 17 healthy controls (HCs) were enrolled (
Supplementary Table 1). Chronic HBV patients were classified into those with HBeAg-positive chronic HBV infection (eAg
+Infe, n=15), HBeAg-positive chronic hepatitis (eAg
+Hepa, n=28), severe chronic hepatitis (SHB, n=7) and HBeAg-negative chronic HBV infection (eAg
–Infe, n=13). All enrolled patients underwent liver stiffness assessment using transient elastography to exclude significant hepatic fibrosis. Additionally, HBeAg-positive patients from two clinical trials were included: telbivudine (LDT, β-L-2’-deoxythymidine, n=10, NCT00962533) and PegIFNα-2a (n=10, NCT01086085) (
Supplementary Table 2). Among PegIFNα-2a-treated patients, complete responders (CR, n=5) achieved HBeAg seroconversion and HBV DNA < 300 copies/mL at week 48, whereas non-complete responders (NCR, n=5) remained HBeAg-positive [
20]. For scRNA-seq, 14 HBeAg-positive patients receiving antiviral therapy with different outcomes were longitudinally followed, including PegIFNα-2a–treated NCR (n=3) and CR (n=3), as well as patients achieving HBsAg-loss following PegIFNα-2a (IFNα_Sloss, n=5) or LDT therapy (LDT_Sloss, n=3) (
Supplementary Table 3). Adjacent nontumor liver tissues were collected from 6 patients with HBV-related hepatocellular carcinoma (HCC) who underwent curative hepatectomy. All participants were recruited from Nanfang Hospital (Guangzhou, China). Exclusion criteria included coinfection with HCV, HDV, HIV, autoimmune liver disease, or other severe conditions. The study complied with the Declaration of Helsinki and was approved by the Ethics Committee of Nanfang Hospital (NCT07137247, NCT01086085, and NCT00962533). Written informed consent was obtained.
A comprehensive list of the flow cytometry antibodies, cell lines, deposited datasets, as well as the software and algorithms employed for data analysis is provided in
Supplementary Table 4.
Data are expressed as either the median (range) or the mean±standard error of the mean. SPSS Statistics 26.0 (IBM Co., Armonk, NY, USA) and GraphPad Prism v.8.0.1 (La Jolla, CA, USA) software were used for statistical analysis. Mann–Whitney U-test and Wilcoxon signed-rank test were used when two groups were compared. The Kruskal-Wallis H test or the Friedman test was used for comparisons of more than two groups. Statistical significance for scRNA-seq comparisons was determined using the Kruskal–Wallis test followed by Dunn’s post hoc test for multiple comparisons. Correlations between variables were analyzed with Spearman’s rank-order correlation coefficient. All statistical analyses were based on two-tailed hypothesis tests with a P-value <0.05.
For further details regarding the materials and methods used, please refer to the Supplementary materials.
RESULTS
TCRL-Ab can serve as a research tool to detect the antibody-mediated effects of NK cells
Given its specific recognition of HBV-infected hepatocytes and possession of an Fc domain, we hypothesized that the cTCRL-Ab could be an ideal mediator for detecting antibody-mediated targeting of HBV-infected hepatocytes by NK cells (
Fig. 1A). To test this, we employed T2 cells, which are routinely used to present exogenous peptides, and loaded them with HBc18–27 to simulate HBV-infected hepatocytes (
Fig. 1A). An
in vitro system consisting of cTCRL-Ab, HBc18–27-loaded T2 cells, and NK cells from patients with chronic HBV infection was thus established to mimic antibody-dependent NK cell responses against HBV-infected hepatocytes. As a control, an sTCRL-Ab specific for the HBV surface 183–191 peptide (sTCRL-Ab) was used as an irrelevant antibody in the same HBc18–27-loaded T2 cell system, while a group without TCRL-Ab treatment served as the blank control (
Fig. 1A). As expected, among the subsets of peripheral lymphocytes, NK cells exhibited the strongest IFNγ response (
Fig. 1B). Notably, compared to both the irrelevant and blank controls, the cTCRL-Ab group showed significantly higher IFNγ production and cytotoxic potential, with a strong positive correlation between these two effects (
Fig. 1C). Since the combination of sTCRL-Ab, HBc18–27, and T2 cells can induce NK cell responses through antibody-independent pathways, we defined the antibody-mediated effect of NK cells as the response induced by cTCRL-Ab minus that induced by sTCRL-Ab. We next explored the optimal conditions for cTCRL-Ab-mediated NK cell effects. The application of 1 μg/mL cTCRL-Ab to enable the targeting of 1×10
5 effector cells to 1×10
5 T2 cells (E:T=1:1) loaded with 1 μg/mL HBc18–27 for 5 hours yielded optimal effects of NK cells (
Fig. 1D), and the subsequent assays were conducted under these conditions.
Next, we compared the antibody-dependent NK cell responses among different types of target cells and found that, compared with HepG2 cells, HBV-producing HepG2.215 cells, and HBc18–27-pulsed HepG2 cells, NK cells exhibited the strongest IFNγ production against HBc18–27-pulsed T2 cells (
Fig. 1E). The observed contrastive responses of NK cells were attributed to the differential expression of HBc18–27/MHC complexes on target cells, as evidenced by the positive correlation between the frequencies of IFNγ in NK cells and HBc18–27/MHC complexes on target cells (
Fig. 1E). Further analyzing this cor-relation by using T2 cells or HepG2 cells pulsed with various densities of HBc18–27 and HBV-producing HepG2.215 cells supported our inference (
Fig. 1F). In contrast, this correlation was not observed in HBc18–27-specific CD8
+ T cells (
Supplementary Fig. 1A,
1B).
Taken together, our results demonstrate that TCRL-Ab can serve as a valuable tool for exploring the antibody-mediated effect of NK cells, which is associated with the surface density of peptide/MHC complexes on target cells.
cTCRL-Ab-mediated NK cell function is compromised in chronic HBV infection and further exacerbated after PegIFNα-2a therapy
We then investigated the antibody-mediated NK cell responses using the cTCRL-Ab research system. Although there was no significant difference between the different natural stages of chronically HBV-infected patients, attenuated expression of IFNγ and tumor necrosis factor-alpha (TNFα) in NK cells was observed in chronically HBV-infected patients as compared to HCs (
Fig. 2A), indicating that sustained HBV infection may lead to impaired cTCRL-Ab-mediated function of NK cells. To clarify whether compromised cTCRL-Ab-mediated NK cell response could be rescued by antiviral treatment, we longitudinally assessed the dynamics in chronically HBV-infected patients receiving LDT or PegIFNα-2a therapy at the indicated time points. As shown in
Figure 2A, LDT treatment failed to restore the cTCRL-Ab-mediated NK cell response, as evidenced by the nonsignificant effects on IFNγ, TNFα, and CD107a production during treatment (
Fig. 2A). Unexpectedly, during PegIFNα-2a treatment, the ability of cTCRL-Ab-mediated NK cell to produce IFNγ and TNFα proceeded to decline (
Fig. 2A). The above results demonstrate that cTCRL-Ab-mediated function of NK cell is impaired in chronic HBV infection, and after PegIFNα-2a therapy.
To further understand the characteristics of NK cells in chronic HBV infection, we analyzed an array of surface receptors and cytokine-induced NK cell functions. The frequency of CD16
hiNK cells was decreased in patients with chronic HBV infection and further reduced during PegIFNα-2a therapy (
Fig. 2B). Notably, the expression of inhibitory receptor NKG2A showed no change during chronic HBV infection but gradually increased following PegIFNα-2a therapy (
Fig. 2C). Conversely, the activating receptor NKG2C was significantly decreased in SHB patients and slightly elevated after LDT therapy, yet exhibited a gradual decrease after PegIFNα-2a therapy (
Fig. 2C). Differently, NKG2D and NKp46 demonstrated gradual increases following PegIFNα-2a therapy (
Fig. 2C). Additionally, the cytokines secretion capacity of NK cells in response to IL-12, IL-15 and IL-18 stimulation was diminished in chronic HBV-infected patients, with neither PegIFNα-2a nor LDT treatment reversing IFNγ secretion levels in NK cells (
Fig. 2D). These discordant changes in NK cells indicate a state of functional perturbation in individuals with persistent HBV infection that refractory to anti-HBV therapies.
Given that the functionality of NK cells is modulated by an array of activating (CD16, TRAIL, NKG2C, NKG2D, and NKp46) and inhibitory (NKG2A) surface receptors [
21], we next asked whether the expression of surface molecules affected cTCRL-Ab-mediated NK cell effects by analyzing their correlations. Specifically, cTCRL-Ab-mediated IFNγ and TNFα production by NK cells positively correlated with the expression of activating receptors CD16, NKG2C, and NKG2D and negatively correlated with NKp46 expression in NK cells.
Notably, the frequency of CD16-expressing NK cells was strongly correlated with the production of IFNγ and TNFα (
Fig. 3A), suggesting the potential involvement of CD16 in cTCRL-Ab-mediated NK cell function. Similarly, compared with HCs, patients with chronic HBV infection presented lower expression levels of CD16 in NK cells, which continued to decrease during PegIFNα-2a treatment, which was synchronous with the dynamics of IFNγ production mediated by cTCRL-Ab (
Fig. 3B). We subsequently assessed PBMCs from 6 PegIFNα-2a-treated patients at baseline, 3 days, and 7 days and found that compared with those at baseline, the frequency of CD16 in NK cells substantially decreased after PegIFNα-2a treatment, and correspondingly, the level of IFNγ in NK cells mediated by cTCRL-Ab decreased markedly (
Fig. 3C). Similarly, stimulated PBMCs or sorted NK cells from patients with untreated chronic HBV infection administered PegIFNα-2a
in vitro showed a significant reduction in CD16 expression in NK cells, concurrently with the decline of cTCRL-Ab-mediated IFNγ secretion by NK cells (
Fig. 3C). By comparing the cTCRL-Abmediated effects between the CD16
hi and CD16
lo NK subsets, we found that cTCRL-Ab-mediated IFNγ secretion was strikingly present in CD16
hi NK cells, whereas CD16
lo NK cells produced very little IFNγ (
Fig. 3D). Additionally, the level of cTCRL-Ab-mediated IFNγ secretion significantly correlated with the expression of CD16 in NK cells (
Fig. 3E). We then evaluated the impact of CD16 blockade on cTCRL-Ab-mediated NK cell function and found that as the concentration of anti-CD16 blocking antibody increased, the expression level of CD16 in the NK cells gradually decreased, accompanied by the reduction in cTCRL-Ab-mediated IFNγ production (
Fig. 3F). Collectively, these results indicate a CD16-dependent manner of cTCRL-Ab-mediated NK cell effects, and the reduction in cTCRL-Ab-mediated NK cell function during PegIFNα-2a treatment was associated with the downregulation of CD16 expression in NK cells.
Previous studies have demonstrated the positive effects of IFNα on NK cells during persistent viral infections, which seems inconsistent with the inhibitory effects of IFNα on the TCRL-Ab-mediated function of CD16
hi NK cells. We further analyzed the characteristics of CD16
hi NK cells and found that compared with CD16
lo NK cells, CD16
hi NK cells expressed significantly lower levels of NKp46 and NKG2A and relatively higher levels of TRAIL, CD16, NKG2C and NKG2D, indicating a more active state (
Supplementary Fig. 2A). Of note, we observed that IFNα simultaneously upregulated activation receptors NKp46 and NKG2D in CD16
hiNK subset, indicating distinct CD16
hiNK subpopulations in chronic HBV infection may exhibit contrasting responses to IFNα (
Supplementary Fig. 2B).
To explore the characteristics of distinct NK cell subsets in chronic HBV infection, particularly CD16
hiNK subsets mediating antibody-dependent effects, we next performed scRNA-seq on PBMCs collected at weeks 0, 12, and 48 from 6 HBeAg-positive CHB patient receiving PegIFNα-2a treatment. NK cells were annotated based on canonical markers, and their overall proportion remained unchanged during treatment (
Supplementary Fig. 2C–
2E). NK cells were further categorized into three groups: CD16
loNK (
NACM1+FCGR3A–), CD16
hiNK (
NACM1–FCGR3A+) and proliferation NK (
STMN1+) (
Fig. 4A, left). With the progression of PegIFNα-2a treatment, the proportion of CD16
loNK cells gradually increased, but the proportion of CD16
hiNK cells decreases (
Fig. 4A, right), consistent with previous results. CD16
lo and CD16
hiNK cells were further subdivided into five subsets: CD16
lo_IL7R
hi, CD16
lo_CD69
hi, CD16
hi_ activated1, CD16
hi_activated2 and CD16
hi_dysfunction (
Fig. 4B). CD16
hi_activated2 cells are referred to here as CX3CR1
+KLRC2
–CD16
hiNK cells here based on its marker genes. Interestingly, we observed that the proportions of the CD16
lo_IL7R
hi and CX3CR1
+KLRC2
–CD16
hiNK subsets gradually increased, while the proportion of CD16
hi_dysfunction subset gradually decreased (
Fig. 4C). The CX3CR1
+KLRC2
–CD16
hiNK subset exhibited the strongest cytotoxic, HLA-independent activating and inhibitory receptor scores, whereas CD16
hi_dysfunction subset exhibited high HLA-dependent inhibitory receptor scores (
Fig. 4D,
Supplementary Fig. 2F and
Supplementary Table 5). Pseudotime analysis suggested sequential development from CD16
lo_IL7R
hi subset to CX3CR1
+KLRC2
–CD16
hiNK subset, ultimately leading to the formation of CD16
hi_dysfunction subset (
Fig. 4E). Notably, NK cells showed an initial increase followed by the decrease at the expression levels of cytotoxicity, cytolytic granules (
GZMA,
GZMB, and
PRF1), apoptosis, anti-apoptotic regulators (
MCL1 and
PIM2) along the transition process (
Fig. 4E). Our analyses suggest that the CX3CR1
+KLRC2
–CD16
hiNK cluster is the main cytotoxic NK subgroup during IFNα treatment. Flow cytometry confirmed these cells as a cytotoxic subset with highest CD16 expression and as the predominant subset responsible for cTCRL-Ab-mediated NK responses (
Fig. 4F). To functionally validate the cytotoxic potential of NK subsets, sorted CX3CR1
⁺KLRC2
–CD16
hi and remaining CD16
hi NK cells were separately co-cultured with HBV-producing HepG2.2.15 cells. As shown in
Figure 4G, compared with control or other CD16
hiNK cells, the CX3CR1
+KLRC2
–CD16
hiNK subset induced greater target-cell apoptosis and death, lactate dehydrogenase release, and killing efficiency. This subset also reduced supernatant HBsAg and HBeAg levels, confirming its superior cytotoxic and antiviral capacity consistent with the single-cell transcriptomic find-ings. Taken together, CX3CR1
+KLRC2
–CD16
hiNK subset exhibits the strongest cytotoxic effect in CD16
hiNK population, which may play a potential role during Peg-IFNα treatment.
To further elucidate the potential role of CX3CR1
+KLRC2
–CD16
hiNK cells in chronic HBV infection, we investigated the dynamic changes in different CD16
hiNK cell subsets at single-cell resolution between the CR and NCR groups of PegIFNα-2a-treated CHB patients. Although there was no difference between the CR and NCR groups in terms of the expression of inhibitory or activating receptors or cytokine production by NK cells (
Supplementary Fig. 3A–
3C), the proportion of CX3CR1
+KLRC2
–CD16
hiNK cells gradually increased, whereas the proportion of CD16
hi_dysfunction cluster decreased, with this trend being more pronounced in the CR group (
Fig. 5A,
5B). Furthermore, the cytotoxicity of NK cells was enhanced in CHB patients post treatment, particularly within the CX3CR1
+KLRC2
–CD16
hiNK subset of the CR group (
Fig. 5C). We further dissected whether similar results could be observed in patients who achieved functional cure. In patients who experienced HBsAg-loss upon PegIFNα-2a treatment, a comparable shift in NK cell subsets was identified, as seen in the CR group, with the CX3CR1
+KLRC2
–CD16
hiNK subset consistently maintaining high cytotoxicity. In contrast, this pattern was absent in patients with LDT treatment-induced HBsAg-loss (
Fig. 5D,
5E). These data underscore the unique involvement of the CX3CR1
+KLRC2
–CD16
hiNK subset in the immune response of CHB patients, particularly in the context of PegIFNα-2a therapy. Interestingly, the expression of
KLRC1, an inhibitory receptor on NK cells, tended to significantly increase in the CX3CR1
+KLRC2
–CD16
hiNK subset following PegIFNα-2a treatment, especially in the NCR group (
Fig. 5F). In summary, the enrichment of CX3CR1
+KLRC2
–CD16
hiNK cells may indicate an effective response to PegIFNα-2a therapy. Conversely, elevated expression of
KLRC1 may be associated with resistance and poor response to Peg-IFNα treatment.
Given that the liver is enriched in NK cells and the target organ for HBV infection, we further analyzed the transcriptional characteristics of intrahepatic CD16
hiNK cell subsets from 6 HCs, 11 chronically HBV-infected patients, and 3 individuals with HBsAg-loss (
Supplementary Fig. 4A). Compared with that in the HCs, the proportions of NK cells decreased in both chronically HBV-infected and HBsAg-loss groups (
Supplementary Fig. 4B). In accordance with the methodology delineated in
Figure 4A–
B, NK cells were classified into CD16
lo, CD16
hi, and proliferation NK subsets (
Supplementary Fig. 4C–
D). The proportion of CD16
hiNK cells decreased sequentially among HCs, chronically HBV-infected, and HBsAg-loss groups (
Supplementary Fig. 4E). To accurately annotate CX3CR1
+KLRC2
–CD16
hiNK cells in this dataset, we used Scibet [
22] to extract the top 50 characteristic genes of CD16
hi_activated1, CX3CR1
+KLRC2
–CD16
hiNK, and CD16
hi_dysfunction cell types from our data and applied them to this dataset (
Fig. 6A,
6B and
Supplementary Fig. 4F). Similar gene expression patterns were observed between the two datasets (
Fig. 6A and
Supplementary Fig. 4G). Notably, HBsAg-loss group had the highest proportion of CX3CR1
+KLRC2
–CD16
hiNK cells (
Fig. 6B, right). Consistency with the functional score in
Figure 4D and
Supplementary Figure 3E further confirmed the accuracy of the annotations (
Fig. 6C). To increase the robustness of the intrahepatic NK cell analysis, we further examined an independent public single-cell RNA-seq dataset [
23] comprising liver samples from six CHB patients and six individuals with HBsAg-loss, and obtained consistent findings (
Supplementary Fig. 4H,
4I). To understand the functional distinctions of CX3CR1
+KLRC2
–CD16
hiNK cells, we conducted enrichment analysis on characteristic genes between groups and found that cytokine and chemokine pathways were specifically enriched in HBsAg-loss individuals (
Fig. 6D,
6E and
Supplementary Fig. 4J). We also found that the scores for IFN and chemokine pathways in CX3CR1
+KLRC2
–CD16
hiNK cells were higher in HBsAg-loss group (
Fig. 6E, right). The expression of
KLRC1 in CX3CR1
+KLRC2
–CD16
hiNK cells was highest in chronically HBV-infected patients, and decreased in HBsAg-loss group (
Fig. 6F, left). In accordance with the expression of
KLRC1, cell–cell interaction analysis between CX3CR1
+KLRC2
– CD16
hiNK cells and other immune cells revealed significant differences in inhibitory ligands, including HLA-E-KLRC1, HLA-E-CD94:NKG2A, and HLA-C-KIR2DL1, in CHB patients; however, these differences were not observed in HCs and HBsAg-loss individuals (
Fig. 6F, right). Consistently, flow cytometry confirmed that intrahepatic CX3CR1
+KLRC2
–CD16
hiNK cells exhibited the highest CD16 expression, along with elevated activation marker SLAMF6 and reduced inhibitory receptor NKG2A expression, displaying the strongest cytotoxic features (
Fig. 6G). These findings suggest that the CX3CR1
+KLRC2
–CD16
hi NK cell subset exhibits functional restoration in patients who have achieved functional cure.
DISCUSSION
The antibody-mediated effect, as a critical component of NK cell function, plays a pivotal role in antiviral responses [
24]. In this study, we address the function of NK cell antibody-mediated effects by employing cTCRL-Ab in the context of chronic HBV infection. Our results demonstrated that IFNα treatment limits cTCRL-Ab-mediated IFNγ production of NK cells by downregulating their expression of CD16. The IFNα-triggered curtailment of CD16 occurs within a CD16
hi_dysfunction subset exhibiting exhausted property, in contrast to a cluster of functionally activated CX3CR1
+KLRC2
–CD16
hiNK cells, which is thriving in both frequency and function as treatment progresses, and is associated with treatment response. Additionally, the intrahepatic CX3CR1
+KLRC2
–CD16
hiNK subset is functionally restored in HBsAg-loss individuals. Our results highlight the differential sculpt of CD16
hiNK subsets and unveil a previously unappreciated functional CX3CR1
+KLRC2
–CD16
hiNK subset, which may facilitate the development of IFNα-based NK cell-targeted therapy for HBV.
Although NK cells, as the leading executors of innate immune function, have been extensively studied [
24], their functional role in chronic HBV infection has not been fully elucidated. In an attempt to assess the antibody-mediated function of NK cells, we utilized the cTCRL-Ab reported by Antonio Bertoletti’s group, initially constructed to visualize peptide/MHC complexes on HBV-infected cells that deliver IFNα specifically to these cells [
18,
19]. The HBV specificity and Fc structure of cTCRL-Ab inspire us to investigate whether it could be used to detect antibody-mediated anti-HBV function of NK cells. The obvious IFNγ production and target cell killing effects detected using cTCRL-Ab reflect the functionality of NK cells, and the CD16-dependent attribute further confirms the feasibility of employing cTCRL-Ab as a research tool for exploring NK cell antibody-mediated effects. Beyond its use as a mechanistic tool, the cTCRL-Ab system may also have translational potential. It enables functional assessment of antibody-dependent NK activation, which could be applied to monitor immune restoration during IFNα therapy or after HBsAg-loss. In addition, this platform could be adapted to screen antibodies or Fc-engineered molecules that enhance FcγRIIIa-mediated antiviral responses, highlighting its potential clinical relevance in chronic HBV infection.
Intriguingly, when using HepG2 cells as targets, optimal cTCRL-Ab-mediated effects could not be manifested. We speculate that this may be ascribed to the moderate ability of HepG2 cells to present exogenous antigen peptides. Since HepG2 cells are transporters associated with antigen processing (TAP)-competent cells, exogenous peptides need to compete with endogenous peptides for MHC binding, and increasing the concentration of peptides contributes to their competitiveness, although these cells are still significantly inferior to T2 cells, which are professionally used to present exogenous peptides because of their TAP deficiency [
25]. Thus, when cTCRL-Ab is used to assess NK cell function, optimal conditions should be ensured. In contrast, the functionality of HBc18–27-specific CD8
+ T cells is not limited by HBc18–27/MHC levels, indicating their high affinity for HBc18–27/MHC complexes, and even trace amounts of these cells achieve full activation.
Previous studies have demonstrated that NK cells exhibit a functional dichotomy—impaired cytokine production but intact cytotoxicity—during chronic HBV infection, as evidenced by
in vitro cytokine stimulation or direct coculture with target cells [
8,
26,
27]. Here, using cTCRL-Ab, we revealed that NK cells from chronic HBV patients succumbed to compromised antibody-mediated cytokine production while maintaining their degranulation capacity. This phenomenon is reminiscent of the functional dichotomy mentioned above, suggesting that different NK effector functions may follow similarly perturbed patterns during chronic HBV infection.
Of note, IFNα treatment resulted in a stepwise decrease in CD16 expression, leading to a counterintuitive downregulation of cTCRL-Ab-mediated cytokine responses in NK cells, and indiscriminately occurred in both CR and NCR groups. These findings appear paradoxical when compared with previous reports describing the activating effects of IFNα on NK cells during chronic viral infections, including enhanced cytotoxicity, CD107a degranulation, and ADCC activity [
14,
15,
17,
28,
29]. However, accumulating evidence indicates that IFNα exerts receptor-specific and context-dependent regulatory effects rather than uniform activation. Type I IFNs can simultaneously promote activation while constraining certain effector pathways, depending on the cytokine milieu, receptor repertoire, and activation state of NK subsets. For instance, IFNα can upregulate activating receptors such as NKp46 and NKG2D but downregulate FcγRIIIa (CD16), thereby shifting NK cells from antibody-dependent cytotoxicity toward natural cytotoxicity and cytokine-driven effector functions [
14,
15]. Similarly, tumor models have demonstrated bidirectional regulation—enhancing NK-mediated elimination through NKG2D ligand induction in some contexts while reducing NKG2D ligand H60 expression and NK killing efficiency in others [
30,
31]. Moreover, type I IFNs can induce ligands of inhibitory checkpoint pathways such as PD-1/PD-L1 [
32], further underscoring their dual regulatory role. Consistent with these findings, in our study, Peg-IFNα selectively reduced CD16 within the exhausted CD16
hi_dysfunction subset while expanding and functionally reinforcing the cytotoxic CX3CR1
⁺KLRC2
–CD16
hi subset. Together, these results suggest that IFNα reprograms NK subset composition rather than globally inhibiting NK activity, thereby reshaping the immune landscape toward functional rebalancing.
Our scRNA-seq data indeed delineated distinct regulatory patterns of CD16
hiNK subsets by IFNα, with a notable reinvigoration of the CX3CR1
+KLRC2
–CD16
hiNK cells and a suppression of the CD16
hi_dysfunction subset. These discrepancies were more prominent in the treatment-responsive group. Notably, recent studies reported that CX3CR1
⁺ CD8
⁺ T cells positively correlate with HBV clearance [
33] and that dendritic cells secrete high levels of CX3CL1 upon HBV exposure [
34], implying that CX3CL1–CX3CR1 interactions facilitate crosstalk between dendritic cells and cytotoxic effector populations. Within this framework, CX3CR1-expressing NK cells may represent a specialized subset regulated by IFNα through the CX3CL1/CX3CR1 axis to reinforce anti-HBV responses.
Given this complexity, the heterogeneous existence of NK cells should be considered and cautious analysis of each NK subset is warranted when drawing conclusions. Specifically, we identified
KLRC1 expressed in CX3CR1
+KLRC2
–CD16
hiNK cells, as a potential marker associated with IFNα treatment resistance and poor response. The
KLRC1 gene encodes a major NK cell inhibitory receptor, the NKG2A, which has been widely confirmed to impair the antitumor activity of NK cells, and NKG2A blockade enhances the antitumor immunity mediated by NK cells [
35-
37]. Likewise, anti-NKG2A improves the activity of NK cells to facilitate HBV clearance in mice [
38]. Further dissection of the role of NKG2A in CX3CR1
+KLRC2
–CD16
hiNK cells may aid in the development of NK subset-specific immunotherapy strategies.
It is worth noting that the heterogeneous NK subsets identified in our scRNA-seq analysis are based on transcriptome resolution; further dissection at the phenotypic and functional levels is needed to precisely define the role of IFNα in these distinct NK subpopulations. Accordingly, how IFNα differentially manipulates CX3CR1
+KLRC2
–CD16
hi NK cells and the CD16
hi_dysfunction subset warrants mechanistic exploration. In addition, owing to the limited availability of human liver tissue, we did not assess the dynamics of intrahepatic NK subsets throughout the course of IFNα treatment. Further filling the gap in our understanding of intrahepatic NK subsets upon IFNα treatment will contribute to gaining insight into compartmentalized NK cell modulation. In addition, the sex ratio imbalance in our cohort may have influenced NK cell responses. Prior studies have shown that compared with those from males, intrahepatic NK cells from female patients with CHB exhibit stronger degranulation activity, which is correlated with estradiol levels [
39]. Thus, sex-related immune differences should be considered when interpreting our findings.
Collectively, our data complement the understanding of the antibody-mediated responses of NK cells in the context of HBV infection, and provide potential mechanistic evidence of the differential modulation of distinct NK subsets at the single-cell level, which may be exploited in immunotherapeutic strategies to achieve the optimal anti-HBV therapy.
FOOTNOTES
-
Authors’ contributions
LBT, YHW, ZHJ, XYL, YRG, SHZ, and YYL conceived and designed the studies. LBT, YHW, ZHJ, XYL, YRG, ZFZ, LNS, XY, LTZ, WYH, LPW, WXH, XQL and JRS collected samples and laboratory data. LBT, YHW, XYL, YRG, ZHJ, YJZ, ZFZ, LNS, XY, and SHZ performed the experiments and analyses. LBT, YHW, SHZ, and YYL wrote the manuscript. YYL, SHZ, and XYL supervised the study. All authors read and approved the final manuscript.
-
Acknowledgements
The authors express their gratitude to Professor Antonio Bertoletti for generously providing cTCRL-Ab and sTCRL-Ab, and guidance on this work. This work was supported by the Key-Area Research and Development Program of Guangdong Province (2023B1111030005), the Health Care Major Project of Guangzhou (202206080001), the National Natural Science Foundation of China (82170611, 82270647, 82272314 and 82400709), the Postdoctoral Fellowship Program of CPSF (GZC20231079), the Natural Science Foundation of Guangdong (2022A1515012597), the Guangzhou Science and technology planning project (2024A04J6610), and the Guangdong Basic and Applied Basic Research Foundation (2021A1515011726, 2023A1515010227 and 2023A1515110230). Graphical abstract was created with biorender.com.
-
Conflicts of Interest
The authors have no conflicts to disclose.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (
http://www.e-cmh.org).
Supplementary Figure 1.
cTCRL-Ab-mediated IFNγ production by CD8+ T cells and the surface density of peptide/MHC complexes on target cells. (A, B) TCR-transduced CD8+ T cells, which can specially recognize HBc18–27, were cultured with (A) HBc18–27 pulsed T2 cells, (B) HBc18–27 pulsed HepG2 cells and HepG2.215 cells. The IFNγ production from TCR-transduced CD8+ T cells and the density of HBc18–27/MHC complexes on target cells were detected by flow cytometry. cTCRL-Ab, T-cell receptor-like antibody specific for HBV core 18–27 peptide; IFNγ, interferon γ.
Supplementary Figure 2.
The phenotypes of CD16hiNK cells and single-cell transcriptional profiles of PBMCs in CHB patients receiving PegIFNα-2a treatment. (A) The proportions of CD16hi/CD16lo cells in NK cells and the expression of Trail, CD69, Nkp46, NKG2A, NKG2C, and NKG2D were detected in CD16hi/CD16lo NK cells in HCs (n=17), patients with eAg+Infe (n=15), eAg+Hepa (n=28), SHB (n=7), eAg-Infe (n=13). (B) The proportions of CD16hicells in NK cells, the expression of Trail, Nkp46, NKG2A, NKG2C, and NKG2D in CD16hiNK cells and cytotoxicity of CD16hiNK cells after IL-12, IL-15, and IL-18 stimulation were detected in HCs (n=17), patients with chronic HBV infection (n=63), and patients receiving LDT (n=10) or PegIFNα-2a (n=10). (C) UMAP plots showing major cell subsets. (D) Dot plot showing the expression of classic marker genes. (E) The proportion of major cell subsets from the indicated groups. (F) Violin plots showing scores of NK cell-related gene sets between the indicated groups (left); Dot plot showing the expression of TNFRSF9 and PDCD1 (right). eAg+Hepa, HBeAg-positive chronic hepatitis; eAg+Infe, HBeAg-positive chronic HBV infection; eAg-Infe, HBeAg-negative chronic HBV infection; HBV, hepatitis B virus; HCs, healthy controls; IFNγ, interferon γ; LDT, telbivudine; SHB, severe chronic hepatitis. *P<0.05, **P<0.01, ***P<0.001.
Supplementary Figure 3.
(A–C) Dynamics of cytotoxicity, frequencies and phenotypes in NK cells from CR (n=5) and NCR (n=5) patients
receiving PegIFNα-2a. CR, complete response; IFNγ, interferon γ; NCR, non-complete response. *P<0.05, **P<0.01, ***P<0.001.
Supplementary Figure 4.
Single-cell transcriptional profiles of intrahepatic immune cells from HCs, chronically HBV-infected patients and HBsAg-loss individuals. (A) UMAP plots and dot plot showing disease states, major cell type distribution and the expression of classic marker genes. (B) The proportion of major cell subsets from the indicated groups. (C, D) UMAP plots showing major NK cell subsets (C) and the expression of classic marker genes (D). (E) Relative ratios of CD16hiNK subsets among disease states. (F) Dot plot showing the expression of the top 50 characteristic genes of CD16hi_activated1, CD16hi_activated2 and CD16hi_dysfunction cell types in our data. (G) Correlation analysis of multiples with common differentially expressed genes between two sets of data. R and P-values were calculated by Pearson correlation analysis. (H) UMAP plots of disease states and cell type distribution from an independent public single-cell RNA-seq dataset (left). Dot plot of canonical marker genes (middle); Relative ratios of CD16hiNK subsets between HBV and Sloss groups (right). (I) Heatmap of NK-related genes. (J) Violin plots of NK-related gene set scores. HBV, hepatitis B virus; HCs, healthy controls; Sloss, HBsAg-loss individuals.
Figure 1.The utilization of TCRL-Ab for detecting antibody-mediated effect of NK cells. (A) Schematic of cTCRL-Ab and co-culture system to detect NK cytotoxicity and cytokine production against HBc18–27-pulsed T2 cells. (B) Gating strategy and representative IFNγ production data. (C) IFNγ by NK cells (n=64, left), cytolysis of target cells (n=23, middle), and their correlation (n=14, right). (D) Optimization of HBc18–27/cTCRL-Ab concentrations, effector-to-target ratio, and incubation time. (E) IFNγ by NK cells against HepG2 (n=6), HepG2.215 (n=8), HBc18–27-pulsed HepG2 (n=8), and pulsed T2 cells (n=6) (left); HBc18–27/MHC complex density (middle); and their correlation (n=14, right). (F) cTCRL-Ab-mediated IFNγ production by NK cells and the surface density of HBc18–27/MHC complexes on HBc18–27 pulsed T2 cells (left) or HepG2 cells and HepG2.215 cells (right). cTCRL-Ab, T-cell receptor-like antibody specific for HBV core 18–27 peptide; HBc18-27, HBV core18-27 peptides; HBV, hepatitis B virus; IFNγ, interferon γ; NK, natural killer; sTCRL-Ab, T-cell receptor-like antibody specific for HBV surface 183–191 peptide; TCRL-Ab, T-cell receptor-like antibody. *P<0.05, **P<0.01, ***P<0.001.
Figure 2.Characterizations of NK cell phenotype and function in HCs (n=17), patients with eAg+Infe (n=15), eAg+Hepa (n=28), SHB (n=7), eAg-Infe (n=13), and CHB patients treated with LDT (n=10) or PegIFNα-2a (n=10). (A) cTCRL-Ab-mediated IFNγ, TNFα, and CD107a production by NK cells. (B) Frequencies of NK cells and subsets. (C) Surface receptors of NK cells. (D) The production of IFNγ (left), TNFα (middle) and CD107a (right) by NK cells stimulated with IL-12, IL-15, and IL-18. CHB, chronic hepatitis B; cTCRL-Ab, T-cell receptor-like antibody specific for HBV core 18–27 peptide; eAg+Hepa, HBeAg-positive chronic hepatitis; eAg+Infe, HBeAg-positive chronic HBV infection; eAg-Infe, HBeAg-negative chronic HBV infection; HBV, hepatitis B virus; IFNγ, interferon γ; IL, interleukin; LDT, telbivudine; NK, natural killer; SHB, severe chronic hepatitis; TNFα, tumor necrosis factor α. *P<0.05, **P<0.01, ***P<0.001.
Figure 3.Relationship between cTCRL-Ab-mediated NK effects and CD16 expression. (A) Correlation of CD16, Trail, NKG2A, NKG2C, NKG2D, and NKp46 expression with cTCRL-Ab-mediated IFNγ, TNFα, and CD107a production (n=90). (B) CD16 expression (left) and cTCRL-Ab-mediated IFNγ production (right) in NK cells from HCs (n=17), chronic HBV patients (n=63), and PegIFNα-2a-treated patients (n=10). (C) CD16 expression and cTCRL-Ab-mediated IFNγ production in NK cells from PegIFNα-2a-treated patients at baseline, day 3, and day 7 (left, n=6); PBMCs (middle, n=6) or purified NK cells (right, n=6) stimulated with PegIFNα-2a for 0, 3, and 7 days. (D) cTCRL-Ab-mediated IFNγ production in CD16hiNK cells or CD16loNK cells (n=14). (E) Correlation between NK-derived IFNγ and CD16 expression (n=21). (F) PBMCs from 3 volunteers treated with anti-CD16 antibody at different concentrations to assess cTCRL-Ab-mediated IFNγ production and CD16 expression. cTCRL-Ab, T-cell receptor-like antibody specific for HBV core 18–27 peptide; HBV, hepatitis B virus; IFNγ, interferon γ; NK, natural killer; PBMCs, peripheral blood mononuclear cells; TNFα, tumor necrosis factor α; TCRL-Ab, T-cell receptor-like antibody. *P<0.05, **P<0.01, ***P<0.001.
Figure 4.Single-cell transcriptional profiles of peripheral NK cells from CHB patients receiving PegIFNα-2a. (A) UMAP plots of major NK subsets and classic marker gene expression and proportions of major subsets at weeks 0, 12, and 48. (B) UMAP of specific NK subsets (left) and dot plot of marker gene expression (right). (C) Proportions of specific subsets at weeks 0, 12, and 48. (D) Violin plots of NK-related gene set scores (left) and heatmap of NK-related genes (right). (E) Pseudotime trajectories colored by value and subsets (left), with signature scores and gene expression along pseudotime (right). GZMA, GZMB, and PRF1 represent cytolytic granules, while MCL1 and PIM2 denote anti-apoptotic regulators. (F) Expression of granzyme B, perforin 1 (left) and CD16 (middle) in each subset (n=7). cTCRL-Ab-mediated IFNγ production in each subset (n=26, right). (G) Sorted CX3CR1⁺KLRC2-CD16hi and remaining CD16hiNK cells were separately co-cultured with HBV-producing HepG2.2.15 cells. Bar plots show the proportions of live, apoptotic, and dead target cells, killing efficiency, and LDH release, as well as supernatant HBsAg and HBeAg levels after co-culture (n=6). CHB, chronic hepatitis B; HBV, hepatitis B virus; LDH, lactate dehydrogenase; NK, natural killer; UMAP, Uniform Manifold Approximation and Projection reduction. CD16hi_activated1, CX3CR1-KLRC2-CD16hiNK cells; CD16hi_activated2, CX3CR1+KLRC2-CD16hiNK cells; CD16hi_dysfunction, CX3CR1-KLRC2+CD16hiNK cells. *P<0.05, **P<0.01, ***P<0.001.
Figure 5.Single-cell analysis of peripheral CD16hiNK subset signatures in CR and NCR groups receiving PegIFNα-2a. (A) Relative ratios of CD16hiNK subsets in CR and NCR groups at each time point (left); UMAP and density plots of CD16hiNK subsets in CR and NCR (right). (B) Relative ratios of CD16hiNK subsets in CR and NCR. (C) Heatmap of the signature scores. (D) Relative ratios of CD16hiNK subsets in patients achieving functional cure following LDT or PegIFNα-2a treatment. (E) Heatmap of the signature scores. (F) KLRC1 expression in CD16hi_activated2 NK cells. CR, complete response, HBeAg seroconversion; LDT, telbivudine; NCR, non-complete response, HBeAg non-seroconversion; NK, natural killer; UMAP, Uniform Manifold Approximation and Projection reduction. CD16hi_activated1, CX3CR1-KLRC2-CD16hiNK cells; CD16hi_activated2, CX3CR1+KLRC2-CD16hiNK cells; CD16hi_dysfunction, CX3CR1-KLRC2+CD16hiNK cells.
Figure 6.Single-cell analysis of intrahepatic CD16hiNK cells from HCs, chronically HBV-infected patients and HBsAg-loss individuals. (A) UMAP plots of disease states and cell type distribution. (B) Dot plot of canonical marker genes (left); Relative ratios of CD16hiNK subsets across groups (right). (C) Violin plots of signature scores. (D) Gene expression trends in CD16hi_activated2 NK cells. (E) GO enrichment of specific genes (left); Heatmap of signature scores (right). (F) KLRC1 expression in CD16hi_activated2 NK cells (left); Significant ligand-receptor pairs toward CD16hi_activated2 NK cells via CellChat (right). (G) Expression of CD16 (left), activation and inhibitory markers SLAMF6 and NKG2A (middle), and cytotoxic molecules granzyme B, perforin 1, and IFNγ (right) in intrahepatic CD16hiNK subsets (n=6). HBV, hepatitis B virus; HCs, healthy controls; IFNγ, interferon γ; NK, natural killer; UMAP, Uniform Manifold Approximation and Projection reduction. CD16hi_activated1, CX3CR1-KLRC2-CD16hiNK cells; CD16hi_activated2, CX3CR1+KLRC2-CD16hiNK cells; CD16hi_dysfunction, CX3CR1-KLRC2+CD16hiNK cells.
Abbreviations
T-cell receptor-like antibody specific for HBV core 18-27 epitope
HBeAg-positive chronic hepatitis
HBeAg-positive chronic HBV infection
HBeAg-negative chronic HBV infection
peripheral blood mononuclear cells
single-cell RNA sequencing
transporter associated with antigen processing
T-cell receptor-like antibody
REFERENCES
- 1. Bertoletti A, Ferrari C. Innate and adaptive immune responses in chronic hepatitis B virus infections: towards restoration of immune control of viral infection. Gut 2012;61:1754-1764.
- 2. Marotel M, Villard M, Drouillard A, Tout I, Besson L, Allatif O, et al. Peripheral natural killer cells in chronic hepatitis B patients display multiple molecular features of T cell exhaustion. Elife 2021;10:e60095.
- 3. Cooper MA, Fehniger TA, Turner SC, Chen KS, Ghaheri BA, Ghayur T, et al. Human natural killer cells: a unique innate immunoregulatory role for the CD56(bright) subset. Blood 2001;97:3146-3151.
- 4. Caligiuri MA. Human natural killer cells. Blood 2008;112:461-469.
- 5. Vivier E, Raulet DH, Moretta A, Caligiuri MA, Zitvogel L, Lanier LL, et al. Innate or adaptive immunity? The example of natural killer cells. Science 2011;331:44-49.
- 6. Nimmerjahn F, Ravetch JV. Fcgamma receptors as regulators of immune responses. Nat Rev Immunol 2008;8:34-47.
- 7. Yu WH, Cosgrove C, Berger CT, Cheney PC, Krykbaeva M, Kim AY, et al. ADCC-mediated CD56(DIM) NK cell responses are associated with early HBsAg clearance in acute HBV infection. Pathog Immun 2018;3:2-18.
- 8. Tjwa ET, van Oord GW, Hegmans JP, Janssen HL, Woltman AM. Viral load reduction improves activation and function of natural killer cells in patients with chronic hepatitis B. J Hepatol 2011;54:209-218.
- 9. Lunemann S, Malone DF, Hengst J, Port K, Grabowski J, Deterding K, et al. Compromised function of natural killer cells in acute and chronic viral hepatitis. J Infect Dis 2014;209:1362-1373.
- 10. Heiberg IL, Pallett LJ, Winther TN, Høgh B, Maini MK, Peppa D. Defective natural killer cell anti-viral capacity in paediatric HBV infection. Clin Exp Immunol 2015;179:466-476.
- 11. Boni C, Lampertico P, Talamona L, Giuberti T, Invernizzi F, Barili V, et al. Natural killer cell phenotype modulation and natural killer/T-cell interplay in nucleos(t)ide analogue-treated hepatitis e antigen-negative patients with chronic hepatitis B. Hepatology 2015;62:1697-1709.
- 12. Peppa D, Gill US, Reynolds G, Easom NJ, Pallett LJ, Schurich A, et al. Up-regulation of a death receptor renders antiviral T cells susceptible to NK cell-mediated deletion. J Exp Med 2013;210:99-114.
- 13. Li H, Zhai N, Wang Z, Song H, Yang Y, Cui A, et al. Regulatory NK cells mediated between immunosuppressive monocytes and dysfunctional T cells in chronic HBV infection. Gut 2018;67:2035-2044.
- 14. Cao W, Li M, Zhang L, Lu Y, Wu S, Shen G, et al. The characteristics of natural killer cells in chronic hepatitis B patients who received PEGylated-interferon versus entecavir therapy. Biomed Res Int 2021;2021:2178143.
- 15. Gill US, Peppa D, Micco L, Singh HD, Carey I, Foster GR, et al. Interferon alpha induces sustained changes in NK cell responsiveness to hepatitis B viral load suppression in vivo. PLoS Pathog 2016;12:e1005788.
- 16. Zimmer CL, Rinker F, Höner Zu Siederdissen C, Manns MP, Wedemeyer H, Cornberg M, et al. Increased NK cell function after cessation of long-term nucleos(t)ide analogue treatment in chronic hepatitis B is associated with liver damage and HBsAg-loss. J Infect Dis 2018;217:1656-1666.
- 17. Micco L, Peppa D, Loggi E, Schurich A, Jefferson L, Cursaro C, et al. Differential boosting of innate and adaptive antiviral responses during pegylated-interferon-alpha therapy of chronic hepatitis B. J Hepatol 2013;58:225-233.
- 18. Ji C, Sastry KS, Tiefenthaler G, Cano J, Tang T, Ho ZZ, et al. Targeted delivery of interferon-α to hepatitis B virus-infected cells using T-cell receptor-like antibodies. Hepatology 2012;56:2027-2038.
- 19. Sastry KS, Too CT, Kaur K, Gehring AJ, Low L, Javiad A, et al. Targeting hepatitis B virus-infected cells with a T-cell receptor-like antibody. J Virol 2011;85:1935-1942.
- 20. European Association for the Study of the Liver. EASL 2017 Clinical Practice Guidelines on the management of hepatitis B virus infection. J Hepatol 2017;67:370-398.
- 21. Lanier LL. NK cell recognition. Annu Rev Immunol 2005;23:225-274.
- 22. Li C, Liu B, Kang B, Liu Z, Liu Y, Chen C, et al. SciBet as a portable and fast single cell type identifier. Nat Commun 2020;11:1818.
- 23. Narmada BC, Khakpoor A, Shirgaonkar N, Narayanan S, Aw PPK, Singh M, et al. Single-cell landscape of functionally cured chronic hepatitis B patients reveals activation of innate and altered CD4-CTL-driven adaptive immunity. J Hepatol 2024;81:42-61.
- 24. Björkström NK, Strunz B, Ljunggren HG. Natural killer cells in antiviral immunity. Nat Rev Immunol 2022;22:112-123.
- 25. Luft T, Rizkalla M, Tai TY, Chen Q, MacFarlan RI, Davis ID, et al. Exogenous peptides presented by transporter associated with antigen processing (TAP)-deficient and TAP-competent cells: intracellular loading and kinetics of presentation. J Immunol 2001;167:2529-2537.
- 26. Oliviero B, Varchetta S, Paudice E, Michelone G, Zaramella M, Mavilio D, et al. Natural killer cell functional dichotomy in chronic hepatitis B and chronic hepatitis C virus infections. Gastroenterology 2009;137:1151-1160 1160.e1151-1157.
- 27. Peppa D, Micco L, Javaid A, Kennedy PT, Schurich A, Dunn C, et al. Blockade of immunosuppressive cytokines restores NK cell antiviral function in chronic hepatitis B virus infection. PLoS Pathog 2010;6:e1001227.
- 28. Tomescu C, Tebas P, Montaner LJ. IFN-α augments natural killer-mediated antibody-dependent cellular cytotoxicity of HIV-1-infected autologous CD4+ T cells regardless of major histocompatibility complex class 1 downregulation. Aids 2017;31:613-622.
- 29. Kwaa AKR, Talana CAG, Blankson JN. Interferon alpha enhances NK cell function and the suppressive capacity of HIV-specific CD8(+) T cells. J Virol 2019;93:e01541-18.
- 30. Katlinskaya YV, Carbone CJ, Yu Q, Fuchs SY. Type 1 interferons contribute to the clearance of senescent cell. Cancer Biol Ther 2015;16:1214-1219.
- 31. Bui JD, Carayannopoulos LN, Lanier LL, Yokoyama WM, Schreiber RD. IFN-dependent down-regulation of the NKG2D ligand H60 on tumors. J Immunol 2006;176:905-913.
- 32. Müller L, Aigner P, Stoiber D. Type I interferons and natural killer cell regulation in cancer. Front Immunol 2017;8:304.
- 33. Zhang C, Li J, Cheng Y, Meng F, Song JW, Fan X, et al. Single-cell RNA sequencing reveals intrahepatic and peripheral immune characteristics related to disease phases in HBV-infected patients. Gut 2023;72:153-167.
- 34. Dumolard L, Gerster T, Chuffart F, Decaens T, Hilleret MN, Larrat S, et al. HBV and HBsAg strongly reshape the phenotype, function, and metabolism of DCs according to patients’ clinical stage. Hepatol Commun 2025;9:e0625.
- 35. Sun C, Xu J, Huang Q, Huang M, Wen H, Zhang C, et al. High NKG2A expression contributes to NK cell exhaustion and predicts a poor prognosis of patients with liver cancer. Oncoimmunology 2017;6:e1264562.
- 36. André P, Denis C, Soulas C, Bourbon-Caillet C, Lopez J, Arnoux T, et al. Anti-NKG2A mAb is a checkpoint inhibitor that promotes anti-tumor immunity by unleashing both T and NK cells. Cell 2018;175:1731-1743.e1713.
- 37. Kamiya T, Seow SV, Wong D, Robinson M, Campana D. Blocking expression of inhibitory receptor NKG2A overcomes tumor resistance to NK cells. J Clin Invest 2019;129:2094-2106.
- 38. Li F, Wei H, Wei H, Gao Y, Xu L, Yin W, et al. Blocking the natural killer cell inhibitory receptor NKG2A increases activity of human natural killer cells and clears hepatitis B virus infection in mice. Gastroenterology 2013;144:392-401.
- 39. Macek Jilkova Z, Decaens T, Marlu A, Marche H, Jouvin-Marche E, Marche PN. Sex differences in spontaneous degranulation activity of intrahepatic natural killer cells during chronic hepatitis B: association with estradiol levels. Mediators Inflamm 2017;2017:3214917.
, Shihong Zhong1
