Hepatitis B Virus e Antigen Activates the Suppressor of Cytokine Signaling 2 to Repress Interferon Action

Hepatitis B virus (HBV) infection causes acute hepatitis B (AHB), chronic hepatitis B (CHB), liver cirrhosis (LC), and eventually hepatocellular carcinoma (HCC). The presence of hepatitis B e antigen (HBeAg) in the serum generally indicates ongoing viral replication and disease progression. However, the mechanism by which HBeAg regulates HBV infection remains unclear. Interferons (IFNs) are pleiotropic cytokines that participate in host innate immunity. After binding to receptors, IFNs activate the JAK/STAT pathway to stimulate expression of IFN-stimulated genes (ISGs), leading to induction of antiviral responses. Here, we revealed that HBeAg represses IFN/JAK/STAT signaling to facilitate HBV replication. Initially, HBeAg stimulates the expression of suppressor of cytokine signaling 2 (SOCS2). Subsequently, SOCS2 impairs IFN/JAK/STAT signaling through reducing the stability of tyrosine kinase 2 (TYK2), downregulating the expression of type I and III IFN receptors, attenuating the phosphorylation and nucleus translocation of STAT1. Finally, SOCS2 inhibits the expression of ISGs, which leads to the repression of IFN action and facilitation of viral replication. These results demonstrate an important role of HBeAg in the regulation of IFN action, and provide a possible molecular mechanism by which HBV resists the IFN therapy and maintains persistent infection.

suppressing IFN-β and ISG production [22][23][24][25] . Despite the important clinical implications, the function of HBeAg in IFN action and the molecular mechanism by which HBeAg regulates IFN remains largely unknown.
Members of the intracellular suppressor of cytokine signaling (SOCS) family are regulators of cytokine signaling pathways 26,27 . Eight members (SOCS1 to 7 and CIS) are identified, and most SOCSs are induced by cytokines and act in a classical negative-feedback loop to inhibit cytokine signaling 28 . Most SOCS proteins are induced by cytokines and act in a classical negative-feedback loop to inhibit cytokine signaling. SOCS1 and SOCS3 inhibit interferon-mediated antiviral and antiproliferative activities, and are upregulated in brain resident cells in response to virus-induced inflammation of the central nervous system via at least two distinctive pathways 29,30 .
Here, we investigated the mechanism by which HBV resists to IFN action and maintains persistent infection. Our results revealed that HBeAg initially activates SOCS2 through ERK pathway. HBeAg-activated SOCS2 subsequently reduces tyrosine kinase 2 (TYK2) stability, down-regulates IFN receptors expression, represses STAT1 phosphorylation, and finally attenuates ISGs production. Thus, we revealed a novel mechanism by which HBeAg and SOCS2 are coordinated to enhance HBV replication by hijacking the IFN/JAK/STAT pathway and attenuating IFN antiviral action.

Results
HBeAg attenuates STAT1 phosphorylation and nuclear translocation. We initially evaluated the role of HBeAg in the phosphorylation of STAT1 induced by IFN in cells transfected with pCMV-HBeAg or pCMV-Tag2B and treated with recombinant human IFN-α (rhIFN-α) or recombinant human IFN-λ (rhIFN-λ). HBeAg was highly expressed in pCMV-HBeAg transfected cells and mainly secreted to the cell culture supernatant (Fig. S1A). Phosphorylation of STAT1 (p-STAT1) was enhanced by rhIFN-α or rhIFN-λ1 but repressed by HBeAg (Fig. 1A), and p-STAT1 in nucleus was enhanced by rhIFN-α or rhIFN-λ1 but reduced by HBeAg (Fig. 1B), suggesting that HBeAg plays an inhibitory role in IFN-induced phosphorylation and nuclear translocation of STAT1.
During HBV infection, seroconversion from HBeAg to anti-HBe may lead to the emergence of replication-competent HBV mutants that are unable to secrete HBeAg, and the most frequent mutation is a guanine (G)-to-adenine (A) change at nt 1896 31,32 . Thus, we evaluated the indispensable role of HBeAg in the activation of STAT1 using HBeAg-1896mut. Cells were transfected with pCMV-Tag2B, pCMV-HBeAg, pCMV-HBeAg-1896mut, and pCMV-HBcAg (expressing hepatitis B core antigen), respectively, and treated with rhIFN-α or rhIFN-λ1. HBeAg was secreted in the cells transfected with pCMV-HBeAg, but not in the cells transfected with pCMV-HBeAg-1896mut or pCMV-HBcAg (Fig. S1B). Phosphorylation of STAT1 was induced by rhIFN-α or rhIFN-λ1 and such activation was repressed by HBeAg, but not affected by HBeAg-1896mut or HBcAg (Fig. 1C). The role of HBeAg in the regulation of STAT1 was also determined using a recombinant HBeAg (rHBeAg). The levels of rHBeAg protein in the cell culture medium were evaluated by ELISA (Fig. S1C). Cells were then treated with anti-HBeAg antibody, incubated with rHBeAg or rHBcAg, and treated with rhIFN-α or rhIFN-λ1. Similarly, p-STAT1 was stimulated by rhIFN-α or rhIFN-λ1 and such activation was repressed by rHBeAg but not by rHBcAg (Fig. 1D). More interestingly, the suppression of p-STAT1 mediated by HBeAg was rescued by anti-HBeAg (Fig. 1D). These results suggested that secreted HBeAg rather than intracellular HBeAg is more relevant for the repression of STAT1.
More importantly, we further confirmed the role of HBeAg in the regulation of STAT1 in a HBV infection context. HBeAg and HBsAg proteins can be detected in the cell culture supenatants 3 days after infection using ELISA, which demonstrated that HepG2-NTCP cells were successfully infected by HBV (Fig. S1D). p-STAT1 stimulated by rhIFN-α or rhIFN-λ1 was decreased in HBV-infected HepG2-NTCP cells than that in mock infected cells. And such suppression of p-STAT1 by HBV was nearly reversed by anti-HBeAg (Fig. 1E). These results demonstrated not only an inhibitory effect of HBeAg on STAT1 activation in an infection system, but also an indispensable role of HBeAg in HBV-mediated antagonism of IFN action.
Furthermore, the specificity of HBeAg inhibitory effect on STAT1 was confirmed. Cells were transfected with pCMV-Tag2B or pCMV-HBeAg and treated with recombinant human IL-6 (rhIL-6) or recombinant human IL-4 (rhIL-4). p-STAT3 was activated by rhIL-6 ( Fig. 1F) and p-STAT6 was activated by rhIL-4 ( Fig. 1G), but such activations were not affected by HBeAg. Taken together, we demonstrated that HBeAg is involved in the regulation of IFN/JAK/STAT signaling by specifically inhibiting IFN-mediated phosphorylation and nuclear translocation of STAT1.
IFN-α and IFN-λ1 activate JAK1 or TYK2 to regulate ISGs expression and immune response. Thus, we evaluated the effect of HBeAg on the regulation of JAK1 and TYK2 in cells treated with rHBeAg, rhIFN-α, and rhIFN-λ, respectively. The levels of HBeAg in the conditioned media were evaluated by ELISA (Fig. S2C). Interestingly, p-TYK2 was up-regulated by rhIFN-α or rhIFN-λ1 and down-regulated by rHBeAg, whereas p-JAK1 was stimulated by rhIFN-α or rhIFN-λ1 but not affected by rHBeAg (Fig. 2B). In addition, cells were transfected with pCMV-tag2B or pCMV-HBeAg, and then treated with rhIFN-α, rhIFN-λ, or anti-HBeAg. p-TYK2 was activated by rhIFN-α or rhIFN-λ, whereas such activation was repressed by HBeAg and restored by HepG2 cells were transfected with pCMV-Tag2B or pCMV-HBeAg for 48 h and then treated with recombinant human IFN-α (rhIFN-α) at 300 U/ml or recombinant human IFN-λ1 (rhIFN-λ1) at 100 ng/ ml for 30 min. Cells were harvested and lysed, and p-STAT1, STAT1, and β-actin proteins in the cell lysates were detected by Western blot analyses (A). Nuclear extracts were prepared from the treated cells, and proteins in nuclear extracts were examined by Western blot analyses using anti-p-STAT1 antibody and anti-Lamin A anti-HBeAg (Fig. 2C). p-JAK1 was also activated by rhIFN-α or rhIFN-λ, but this activation was not affected by HBeAg (Fig. 2C). We also noticed that extracellular adding of rHBeAg and transfection of pCMV-HBeAg could (B) HepG2 cells were incubated with PBS or rHBeAg at 50 ng/ml for 24 h, and treated with rhIFN-α at 300 U/ml or rhIFN-λ1 at 100 ng/ml for 30 min. Cells were harvested and lysed, and p-TYK2, TYK2, p-JAK1, JAK1, and GAPDH proteins in the cell lysates were detected by Western blot analyses. (C) HepG2 cells were transfected with pCMV-Tag2B or pCMV-HBeAg for 12 h, and incubated with anti-HBeAg (10 μg/ml) for another 36 h. Cells were treated with rhIFN-α at 300 U/ml or rhIFN-λ1 at 100 ng/ml for 30 min before harvest. Cells were lysed, and p-TYK2, TYK2, p-JAK1, JAK1, and GAPDH proteins in the cell lysates were detected by Western blot analyses. lead to the reduction of TYK2 protein ( Fig. 2B and C). Taken together, we demonstrated that HBeAg attenuates IFN-α and IFN-λ1 actions by down-regulating IFN-α/βR1 and IL-10Rβ expression, repressing TYK2 activation, and inhibiting STAT1 phosphorylation.
The role of HBeAg in the attenuation of antiviral action of IFN-α and IFN-λ was further confirmed in cells co-transfected with pHBV1.3 and pCMV-Tag2B or pCMV-HBeAg and treated with rhIFN-α or rhIFN-λ1. The expression of HBeAg in the cell culture medium was verified by ELISA (Fig. S3A). Interestingly, HBsAg (Fig. 3E) and HBV capsid-associated DNA (Fig. 3F) were significantly repressed by rhIFN-α and rhIFN-λ1 in the absence of HBeAg, but slightly down-regulated by IFN-α or IFN-λ1 in the presence of HBeAg, revealing that HBeAg attenuates the antiviral activities of IFN-α and IFN-λ1. Furthermore, Huh7 cells were transfected with pHBV1.3 or pHBV1.3-1896mut 33 , and treated with rhIFN-α or rhIFN-λ1. HBeAg was detected in the culture medium of cells transfected with pHBV1.3, but not expressed in cells transfected with pHBV1.3-1896mut (Fig. S3B). HBsAg (Fig. 3G) and HBV capsid-associated DNA (Fig. 3H) were slightly down-regulated by IFN-α and IFN-λ1 in the presence of HBV1.3, but significantly attenuated by IFN-α and IFN-λ1 in the presence of HBV1.3-1896mut, indicating that HBeAg is required for the regulation of IFN-α and IFN-λ1. Taken together, we demonstrated that HBeAg attenuates the actions of IFN-α and IFN-λ1.
HBeAg activates SOCS2 expression through ERK signaling. The mechanism by which HBeAg regulates IFN-α and IFN-λ1 was further investigated. Initially, we evaluated the roles of SOCS family members in the regulation of IFN action mediated by HBeAg in cells transfected with pCMV-HBeAg. HBeAg only enhanced the expression of SOCS2, but not SOCS1 and SOCS3, in a dose-dependent fashion (Fig. 4A). The stimulatory effect of HBeAg on SOCS2 expression was further confirmed by the following 4 results. (1) SOCS2 mRNA and protein were activated by rHBeAg in dose-dependent manners (Fig. 4B). (2) SOCS2 mRNA and protein were stimulated by HBeAg in time-dependent fashions (Fig. 4C). (3) SOCS2 mRNA and protein were enhanced by HBeAg, but not by HBeAg-1896mut (Fig. 4D). (4) SOCS2 mRNA and protein were stimulated by rHBeAg, but not by heat-inactivated rHBeAg or rHBcAg (Fig. 4E). Therefore, we revealed that HBeAg plays a specific role in activation of SOCS2.
The effect of HBV on SOCS2 expression was also determined. SOCS2 mRNA and protein were activated in pHBV1.3-transfected cells but not in pBlue-SK-transfected cells, enhanced in HepG2.2.15 cells but not in HepG2 cells (Fig. 4F), and stimulated in pHBV1.3-transfected HepG2 cells in a time-dependent manner (Fig. 4G), demonstrating that HBV stimulates SOCS2 expression in human hepatoma cells. HBV genome contains 4 overlapping open reading frames (ORFs) encoding for 7 proteins. The roles of individual viral proteins in the regulation of SOCS2 were evaluated. SOCS2 mRNA was significantly activated by HBeAg, slightly enhanced by L (PreS1/PreS2/S), but not affected by HBp, HBcAg, HBx, S, or M (PreS2/S) (Fig. 4H), indicating that HBeAg is mainly responsible for the activation of SOCS2. In addition, SOCS2 mRNA and protein were activated by HBV1.3 but not by HBV1.3-1896mut ( Fig. 4I), confirming the essential role of HBeAg in regulating SOCS2.

SOCS2 attenuates the production of IFN receptors.
We evaluated the effect of SOCS2 on the expression of IFN-α and IFN-λ1 receptors by overexpression or knockdown of SOCS2. HepG2 cells were transfected with pcDNA3.1 or pcDNA3.1-SOCS2. A high level of SOCS2 mRNA was detected in pcD-NA3.1-SOCS2-transfected cells, indicating that transfection was efficient and SOCS2 was expressed (Fig. S4A). Flow cytometry analyses indicated that IFN-α/βR1 and IL-10Rβ were significantly reduced by SOCS2, while IFN-α/βR2 and IL-28R1 were relatively unaffected by SOCS2 (Fig. 5A). Western blot analyses confirmed that IFN-α/βR1 (IFNAR1) and IL-10Rβ were significantly reduced by SOCS2, but IFN-α/βR2 (IFNAR2) and IL-28R1 were relatively unaffected by SOCS2 (Fig. S4B). These results suggested that SOCS2 attenuates the expression of IFN-α/βR1 and IL-10Rβ. SOCS2 mRNA was obviously downregulated in siR-SOCS2 transfected cells, indicating that siR-SOCS2 was effective (Fig. S4C). IFN-α/βR1 and IL-10Rβ were up-regulated by siR-SOCS2 ( Fig. 5Ba and Bd), while IFN-α/βR2 and IL-28R1 were relatively unaffected by siR-SOCS2 ( Fig. 5Bb and Bc), suggesting that knockdown of SOCS2 upregulates IFN-α/βR1 and IL-10Rβ. Taken together, we revealed that SOCS2 represses IFN-α and IFN-λ1 action by inhibiting their receptors, IFN-α/βR1 and IL-10Rβ. SOCS2 reduces TYK2 stability and phosphorylation. Since IFN-α/βR1 and IL-10Rβ are associated with TYK2, we determined whether the reduction of IFN-α/βR1 and IL-10Rβ mediated by SOCS2 leads to the dysregulation of TYK2. Cells were transfected with pcDNA3.1 or pcDNA3.1-SOCS2 and treated with rhIFN-α Scientific RepoRts | 7: 1729 | DOI:10.1038/s41598-017-01773-6 or rhIFN-λ1. p-TYK2 was enhanced by rhIFN-α and rhIFN-λ1 but reduced by SOCS2, and total TYK2 was also down-regulated by SOCS2 (Fig. 6A), suggesting that SOCS2 attenuates TYK2 production and activation.  Since SOCS1 inhibits type I IFN by binding to IFN receptor-associated TYK2 34 , it is reasonable for us to speculate that SOCS2 may interact with TYK2. SOCS2 was co-immunoprecipitated with TYK2 in the cells co-transfected with pHA-SOCS2 and pFLAG-TYK2 (Fig. 6B), confirming that SOCS2 interacts with TYK2. We then investigated the role of SOCS2 in the regulation of endogenous TYK2 and demonstrated that the level of TYK2 protein was gradually decreased by SOCS2 in a dose-dependent manner (Fig. 6C). We further evaluated the effect of SOCS2 on TYK2 protein stability in HepG2 cells and confirmed that FLAG-TYK2 protein was gradually degraded as the concentration of SOCS2 increased (Fig. 6D). Taken together, we demonstrated that SOCS2 interacts with TYK2 to attenuate its stability. Knockdown of SOCS2 rescues HBeAg-mediated repression of STAT1. The effect of SOCS2 on IFN-induced activation of STAT1 was further evaluated. The results showed that p-STAT1 was up-regulated by rhIFN-α or rhIFN-λ1, but down-regulated by SOCS2 (Fig. 7A). To determine the role of SOCS2 in the regulation of ISRE-dependent genes, cells were co-transfected with pcDNA3.1-SOCS2 and pISRE-Luc, in which the expression of luciferase (Luc) gene is under the control of ISRE, and treated with rhIFN-α or rhIFN-λ1. ISRE activity was stimulated by rhIFN-α and rhIFN-λ1, but repressed by SOCS2 (Fig. 7B). Therefore, SOCS2 represses the actions of IFN-α and IFN-λ1 through attenuating STAT1 phosphorylation and ISRE activity.
Since HBeAg activates SOCS2 through ERK signaling, and PD98059 (an ERK inhibitor) represses HBeAg-mediated activation of SOCS2 (Fig. 4J), we further determined whether the effect of HBeAg on repression of IFN signaling was due to the activation of SOCS2. Cells were pretreated with PD98059, incubated with rHBeAg, and then treated with IFN-α or IFN-λ1. p-STAT1 was enhanced by rhIFN-α or rhIFN-λ1, but IFN-induced p-STAT1 was repressed by rHBeAg (Fig. 7F). Moreover, rHBeAg-mediated downregulation of p-STAT1 was attenuated by PD98059 (Fig. 7F), indicating that inhibition of SOCS2 can rescue HBeAg-mediated repression of IFN signaling. Taken together, we provided strong evidence to support that SOCS2 plays an important role in HBeAg-mediated repression of IFN-α and IFN-λ1 actions.

Discussion
HBV does not induce a substantial IFN-α/β response in the liver 34 , however, its replication is sensitive to IFN-α/β and IFN-γ produced by NK, NKT, and T cells 35 . IFN-α is used therapeutically to treat HBV infection but has a poor response rate. Thus, HBV must develop strategies to counteract IFN actions and ensure persistent infection. Here, initially, we showed that HBV impairs IFN activity by hijacking the IFN/JAK/STAT pathway through HBeAg. It is known that HBeAg is not required for HBV replication and its exact function is unclear, but may play a role in chronic HBV infection 17,18 . The emergence of HBeAg-negative variants correlates with an exacerbation of liver injury in some patients 20,21 . HBeAg modulates host immune response during CHB progression 22 , suppresses TLR-induced IFN-β and ISG production in liver cells 23 , and inhibits IL-18 signaling and IFN-γ expression in NK and hepatoma cells 36 . In response to HBV infection-established persistent infections, virus-specific CD4 and CD8 T cells are physically deleted or persist in an attenuated (termed exhausted) developmental program unable to proliferate to viral antigens or produce important antiviral and immunostimulatory cytokines (including IFNγ, TNFα, and IL-2) 37 . HBeAg appears to be critical in determining the outcome of immunotherapies in chronic HBV patients. A pDC-based immunotherapeutic approach could be of interest in attempts to Figure 7. The role of SOCS2 in the repression of IFN signaling mediated by HBeAg. (A) HepG2 cells were transfected with pcDNA3.1 or pcDNA3.1-SOCS2 for 48 h and treated with rhIFN-α at 300 U/ml or rhIFN-λ1 at 100 ng/ml for 30 min. Cells were harvested and lysed, and p-STAT1, STAT1, SOCS2, and β-actin proteins were detected by Western blot analyses. (B) HepG2 cells were co-transfected with pISRE-Luc and pcDNA3.1 or pcDNA3.1-SOCS2 for 24 h and treated with rhIFN-α or rhIFN-λ1 for another 12 h. The activity of IFNstimulated response element (ISRE) was measured by luciferase activity assays (upper panel). Data shown were means ± SE; n = 3. *p < 0.05. To confirm the expression of SOCS2 in pcDNA3.1-SOCS2-transfected cells, SOCS2 and β-actin proteins were detected by Western blot analyses (lower panel). (C) Determination of the efficiency of siRNA-SOCS2 in HepG2 cells. HepG2 cells were co-transfected with pCMV-tag2B or pCMV-HBeAg and the siRNA specific to SOCS2 (siR-SOCS2) or its control siRNA (siR-Ctrl) for 48 h. Cells were harvested and lysed, and SOCS2 and β-actin proteins in the cell lysates were detected by Western blot analyses. (D) HepG2 cells were co-transfected with pCMV-Tag2B or pCMV-HBeAg and siR-SOCS2 or siR-Ctrl for 48 h, and treated with rhIFN-α or rhIFN-λ1 for 30 min. (E) HepG2 cells were transfected with siR-Ctrl or siR-SOCS2 for 24 h, and incubated with rHBeAg (50 ng/ml) or PBS for another 24 h, and then treated with rhIFN-α or rhIFN-λ1for 30 min before harvest. (F) HepG2 cells were pretreated with or without PD98059 for 12 h, and incubated with rHBeAg (50 ng/ml) or PBS for 24 h, and then treated with rhIFN-α or rhIFN-λ1 for 30 min before harvest. (D-F) Cells were harvested and lysed, and p-STAT1, STAT1, β-actin, and GAPDH proteins were detected by Western blot analyses. restore functional antiviral immunity, which is critical for the control of the virus in chronic HBV patients 38 . However, despite the important clinical and cellular implications, the molecular mechanism by which HBeAg regulates host immunity remains largely unknown. A previous study provided a molecular mechanism describing HBeAg immunomodulation of innate immune signal transduction pathways via interaction and targeting of TLR-mediated signaling pathways 39 . Here, our results demonstrated that HBeAg represses IFN action and IFN/ JAK/STAT signaling.
Subsequently, we revealed that SOCS2 is required for the function of HBeAg in the repression of IFN activity. Members of the SOCS family are negative regulators of cytokine signaling pathway [26][27][28] . SOCS1 inhibits IFN-γ signaling via direct interaction with IFNGR1 or JAK kinases 40 , and represses type I IFN signaling via the interferon alpha receptor (IFNAR1)-associated TYK2 41 . SOCS3 attenuates cell signaling via binding directly to the cytokine receptor subunit gp130 42 . HBV regulates the expression SOCS1 and SOCS3 in mouse liver 43 . We noticed that our result revealing HBeAg activates SOCS2 expression in human hepatoma cells is contrary to the previous report showing HBV had not effect on SOSC-2 expression in mouse liver 43 . We speculated that the discrepancy between these results may be due to the differences in cell phenotype, cellular responses, surface receptors, and cellular functions between human hematoma cells and mouse liver cells.
Moreover, we demonstrated that SOCS2 subsequently interacts with TYK2 to reduce the protein stability. In addition to the facilitation of IFN signaling by enhancing the interaction between STAT1 and IFN receptor, TYK2 is also required for maintaining IFN receptors on the cell membrane. Thus, it is reasonable for us to speculate that interaction of SOCS2 with TYK2 may result in the dissociation of TYK2 from IFN receptor, leading to a reduction in the receptors. Interestingly, we confirmed that SOCS2 indeed reduces IFN-α/βR1 production, suppresses TYK2 phosphorylation, and attenuates STAT1 nuclear translocation, which lead to the repression of ISGs production.
The effectiveness of IFN-α treatment between HBeAg-positive and HBeAg-negative patients remains uncertain. Most studies have shown that approximately 30-40% of HBeAg-positive patients respond to IFN-α therapy. However, the rate to IFN-α therapy for HBeAg-negative patients is divergent: some studies reported that IFN-α response rate is up to 50% [44][45][46] , whereas others showed long-term IFN-α response rate is less than 10% [47][48][49][50] . Thus, we could not neglect the roles of other viral proteins, including HBsAg and HBp 23,51,52 , in the regulation of IFN-α actions. In contrast to type I IFNs, IFN-λ is elevated in PBMCs of patients with CHB, and HBV is sensitive to IFN-λ in cell culture models 8,[53][54][55] . The weaker and prolonger antiviral responses induced by IFN-λ may have implications for the therapeutic use of IFN-λ 56,57 . Low transaminase levels, high viral replication, long duration of disease, and low inflammatory score in liver histology have been reported to be associated with low response rates to IFN therapy [58][59][60][61] . We demonstrated that SOCS2 also inhibits the expression of IFN-λ receptor and IFN-λ activity, providing a better understanding how HBV resists IFN-α and IFN-λ treatment.
In conclusion, we reveal a novel mechanism by which HBeAg and SOCS2 are coordinated to enhance HBV infection by hijacking the IFN/JAK/STAT pathway and attenuating IFN action (Fig. 8). HBeAg initially activates SOCS2 that subsequently hijacks the IFN/JAK/STAT signaling to reduce TYK2 stability and phosphorylation, down-regulate IFN receptors production, attenuate STAT1 phosphorylation and nucleus translocation, and finally block ISGs production, which results in the facilitation of HBV immune evasion and persistent infection.

Statistical analysis.
All experiments were reproducible and carried out in duplicate or quadruplicate. Each set of experiments was repeated at least three times with similar results, and representative experiments were shown. The results were presented as means. Student's t-test for paired samples was used to determine statistical significance. Differences were considered statistically significant at a p value of ≤0.05.