TMEM2 inhibits hepatitis B virus infection in HepG2 and HepG2.2.15 cells by activating the JAK–STAT signaling pathway

We have previously observed the downregulation of TMEM2 in the liver tissue of patients with chronic hepatitis B virus (HBV) infection and in HepG2.2.15 cells with HBV genomic DNA. In the present study, we investigated the role and mechanism of TMEM2 in HepG2 and HepG2.2.15 during HBV infection HepG2 and HepG2.2.15. HepG2 shTMEM2 cells with stable TMEM2 knockdown and HepG2 TMEM2 and HepG2.2.15 TMEM2 cells with stable TMEM2 overexpression were established using lentivirus vectors. We observed reduced expression of TMEM2 in HBV-infected liver tissues and HepG2.2.15 cells. HBsAg, HBcAg, HBV DNA, and HBV cccDNA levels were significantly increased in HepG2 shTMEM2 cells but decreased in HepG2 TMEM2 and HepG2.2.15 TMEM2 cells compared with naive HepG2 cells. On the basis of the western blotting results, the JAK–STAT signaling pathway was inhibited in HepG2 shTMEM2 cells but activated in HepG2 TMEM2 and HepG2.2.15 TMEM2 cells. In addition, reduced and increased expression of the antiviral proteins MxA and OAS1 was observed in TMEM2-silenced cells (HepG2 shTMEM2 cells) and TMEM2-overexpressing cells (HepG2 TMEM2 and HepG2.2.15 TMEM2 cells), respectively. The expression of Interferon regulatory factor 9 (IRF9) was not affected by TMEM2. However, we found that overexpression and knockdown of TMEM2, respectively, promoted and inhibited importation of IRF9 into nuclei. The luciferase reporter assay showed that IRF9 nuclear translocation affected interferon-stimulated response element activities. In addition, the inhibitory effects of TMEM2 on HBV infection in HepG2 shTMEM2 cells was significantly enhanced by pre-treatment with interferon but significantly inhibited in HepG2.2.15 TMEM2 cells by pre-treatment with JAK1 inhibitor. TMEM2 inhibits HBV infection in HepG2 and HepG2.2.15 by activating the JAK–STAT signaling pathway.


TMEM2 inhibits hepatitis B virus infection in HepG2 and
HepG2.2.15 cells by activating the JAK-STAT signaling pathway X Zhu 1,2,3,5 , C Xie 1,2,3,5 , Y-m Li 4 , Z-l Huang 1,2,3 , Q-y Zhao 1,2,3 , Z-x Hu 1,2,3 , P-p Wang 1,2,3 , Y-r Gu 1,2,3 , Z-l Gao* ,1,2,3,5 and L Peng* ,1,2,3, 5 We have previously observed the downregulation of TMEM2 in the liver tissue of patients with chronic hepatitis B virus (HBV) infection and in HepG2.2.15 cells with HBV genomic DNA. In the present study, we investigated the role and mechanism of TMEM2 in HepG2 and HepG2.2.15 during HBV infection HepG2 and HepG2.2.15. HepG2 shTMEM2 cells with stable TMEM2 knockdown and HepG2 TMEM2 and HepG2.2.15 TMEM2 cells with stable TMEM2 overexpression were established using lentivirus vectors. We observed reduced expression of TMEM2 in HBV-infected liver tissues and HepG2.2.15 cells. HBsAg, HBcAg, HBV DNA, and HBV cccDNA levels were significantly increased in HepG2 shTMEM2 cells but decreased in HepG2 TMEM2 and HepG2.2.15 TMEM2 cells compared with naive HepG2 cells. On the basis of the western blotting results, the JAK-STAT signaling pathway was inhibited in HepG2 shTMEM2 cells but activated in HepG2 TMEM2 and HepG2.2.15 TMEM2 cells. In addition, reduced and increased expression of the antiviral proteins MxA and OAS1 was observed in TMEM2-silenced cells (HepG2 shTMEM2 cells) and TMEM2-overexpressing cells (HepG2 TMEM2 and HepG2.2.15 TMEM2 cells), respectively. The expression of Interferon regulatory factor 9 (IRF9) was not affected by TMEM2. However, we found that overexpression and knockdown of TMEM2, respectively, promoted and inhibited importation of IRF9 into nuclei. The luciferase reporter assay showed that IRF9 nuclear translocation affected interferon-stimulated response element activities. In addition, the inhibitory effects of TMEM2 on HBV infection in HepG2 shTMEM2 cells was significantly enhanced by pre-treatment with interferon but significantly inhibited in HepG2. Chronic hepatitis B virus (HBV) infection is a global public health challenge. It has been estimated that two billion people worldwide are infected with HBV 1 and that~350 million people have chronic HBV infection, which is associated with cirrhosis, liver failure, and hepatocellular carcinoma. Up to one million deaths annually are caused by HBV-related diseases. 2 However, the mechanism by which HBV infects HepG2 and HepG2.2.15 cells is not fully understood. Our previous study investigating susceptibility to HBV revealed significant differences in the expression of the p.Ser1254Asn gene between healthy individuals and patients with HBV infection. Expression of the transmembrane protein TMEM2 encoded by p.Ser1254Asn in normal liver tissues and HepG2 cells was significantly higher than that in liver tissues of patients with chronic HBV infection and in HepG2.2.15 cells with HBV genomic DNA, respectively, 3 suggesting that TMEM2 has an important role in inhibiting HBV infection in HepG2 and HepG2.2.15 cells. TMEM2 belongs to the interferon-inducible transmembrane protein superfamily. The biological functions of TMEM2 remain largely unknown. It has been reported that other members of the interferon-inducible transmembrane protein superfamily exhibit interferon-mediated antiviral functions. 4,5 The JAK-STAT signaling pathway regulates cell growth, survival, and differentiation, and is involved in pathogen resistance. Interestingly, Li et al. 6 reported that inhibition of STAT1 methylation led to the resistance of HBV to interferon alpha treatments. On the basis of previous findings, we hypothesized that the JAK-STAT signaling pathway is involved in the anti-HBV activity of TMEM2. In the present study, we investigated the interactions between TMEM2 and the JAK-STAT signaling pathway during HBV infection in HepG2 and HepG2.2.15 cells. First, we performed gain-and loss-of-function assays in HepG2 and HepG2.2.15 cells by stable TMEM2 overexpression or silencing. Second, the expression of a number of key components of the JAK-STAT signaling pathway was evaluated, and the effects of the JAK-STAT signaling pathway on TMEM2-mediated anti-HBV activities were investigated.

Results and Discussion
Expression of TMEM2 in liver tissues and HepG2 and HepG2.2.15 cells. IHC demonstrated strong expression of TMEM2 in the cytoplasm in normal liver tissue and significantly reduced expression of TMEM2 in liver tissues from patients with chronic HBV infection (Figure 1a). The western blot analysis also showed a significant reduction of TMEM2 expression in liver tissues from patients with chronic HBV infection and HepG2.2.15 cells with HBV genomic DNA compared with normal liver tissues and HepG2 cells without HBV genomic DNA, respectively (Figure 1b). In addition, the mRNA level of TMEM2 in liver tissues from patients with chronic HBV infection and HepG2.2.15 cells with HBV genomic DNA was significantly lower than that in normal liver tissues and HepG2 cells without HBV genomic DNA, respectively (Figure 1c; *Po0.05, **Po0.01).  Effects of TMEM2 on the JAK-STAT signaling pathway in HBV-infected HepG2 and HepG2.2.15 cells. To investigate the molecular mechanism underlying the inhibitory effects of TMEM2 on HBV infection in HepG2 and HepG2.2.15 cells, western blot was used to evaluate the expression of a number of key components of the JAK-STAT signaling pathway in HBV-infected HepG2 and HepG2.2.15 cells. We found that the levels of p-Tyk2, p-JAK1, p-STAT1, and p-STAT2 in all HepG2 and HepG2.2.15 cells were reduced by HBV infection. No significant difference in the expression of Tyk2, JAK1, STAT1, and STAT2 was detected among any of the HBV-infected cells (Figure 4a). After HBV infection, the levels of p-Tyk2, p-JAK1, p-STAT1, and p-STAT2 in HepG2 shTMEM2 cells were lower than those in HepG2 cells, and the levels of p-Tyk2, p-JAK1, p-STAT1, and p-STAT2 in HepG2 TMEM2 cells were higher than those in HepG2 cells ( Figure 4a). In addition, the levels of p-Tyk2, p-JAK1, p-STAT1, and p-STAT2 in HepG2.2.15 TMEM2 cells were higher than those in HepG2.2.15 cells following HBV infection ( Figure 4a). After TMEM2 overexpression or knockdown, there were no significant differences in the cellular level of total interferon regulatory factor 9 (IRF9) in HepG2 and HepG2.2.15 cells (Figure 4a).
We also compared the expression of myxovirus resistance protein A (MxA) and 2ʹ,5ʹ-oligoadenylate synthetase 1 (OAS1), two interferon-inducible proteins in the downstream JAK-STAT signaling pathway, between HepG2 and HepG2.2.15 cells with and without HBV infection using RT-qPCR. We found that HBV infection significantly reduced the expression of MxA in all cells ( Figure 4b). However, in the two groups of cells without HBV infection, the expression of MxA in HepG2 shTMEM2 cells was significantly lower than that in HepG2 cells (Figure 4b, *Po0.01). The expression of MxA in HepG2 TMEM2 cells was significantly higher than that in HepG2 cells (Figures 4b, Po0.01). In addition, overexpression of TMEM2 increased MxA expression in HepG2.2.15 cells irrespective of the presence of HBV infection. A similar expression profile was also obtained for OAS1. HBV infection significantly reduced the expression of OAS1 in HepG2 and HepG2 TMEM2 cells (Figure 4b). The expression of OAS1 in HepG2 shTMEM2 cells was significantly lower than that in HepG2 cells (Figure 4b, Po0.01). The expression of OAS1 in HepG2 TMEM2 cells was significantly higher than that in HepG2 cells (Figure 4b, Po0.05). The expression of OAS1 in HepG2.2.15 TMEM2 cells with or without HBV infection was significantly higher than that in HepG2.2.15 cells (Figure 4b, Po0.01).

TMEM2 inhibits HBV infection
X Zhu et al TMEM2 is a type II transmembrane protein. The biological function of TMEM2 is not yet fully understood. The association of TMEM2 mutations with human deafness was first reported in 2000. 7 Animal experiments have also shown that TMEM2 is involved in cardiac development in zebrafish. 8,9 Expression analyses suggest that TMEM2 is widely expressed in various human tissues, especially liver tissue (http://www.proteinatlas. org/ENSG00000135048-TMEM2/tissue). However, whether TMEM2 is involved in the development of liver diseases is not clear. Therefore, in the present study, we investigated the role of TMEM2 in the pathogenesis of HBV infection in HepG2 and HepG2.2.15 cells.
First, we confirmed that TMEM2 expression in normal liver tissues and HepG2 cells was significantly higher than that in the liver tissues of patients with HBV infection and HepG2.2.15 cells with HBV genomic DNA, respectively. The results suggested that TMEM2 was negatively associated with HBV infection in HepG2 and HepG2.2.15 cells. In addition, we found that, after HBV infection, the levels of HBsAg, HBcAg, HBV DNA, and HBV cccDNA in HepG2 TMEM2 and HepG2.2.15 TMEM2 cells, in which TMEM2 was overexpressed, were significantly lower than those in HepG2 and HepG2.2.15 cells, respectively. In addition, the levels of HBsAg, HBcAg, HBV DNA, and HBV cccDNA in HepG2 shTMEM2 cells, in which the expression of TMEM2 was silenced, were significantly higher than those in HepG2 cells after HBV infection. These results suggested that TMEM2 inhibited HBV infection in HepG2 and HepG2.2.15 cells.
Previous studies have shown that multiple signaling pathways, such as NF-κB, Smad7-TGF-β, REB/PKA, and JAK-STAT, are involved in the pathogenesis of HBV infection. 6,10-12 Among these signaling pathways, the JAK-STAT pathway is involved in the anti-HBV activity of IFN. In addition, HBV and HBV antigens can regulate the expression of JAK-STAT signaling pathway molecules and antiviral proteins in hepatocellular models in vitro. Furthermore, the reduced STAT1 methylation and subsequent increased STAT1-PIAS1 interaction contributed to the antagonistic activity of IFN-alpha against HBV. 6 Therefore, we examined whether the JAK-STAT signaling pathway was involved in the inhibitory effect of TMEM2 on HBV infection. First, we investigated whether TMEM2 could regulate the expression of the JAK-STAT signaling pathway. Our results showed that TMEM2 overexpression activated the JAK-STAT signaling pathway and TMEM2 knockdown decreased the levels of p-Tyk2, p-JAK1, p-STAT1, and p-STAT2 in HepG2 and HepG2.2.15 cells, suggesting that TMEM2 regulates the JAK-STAT signaling pathway. This finding was further supported by the expression levels of MxA and OAS1, two downstream molecules in the JAK-STAT signaling pathway. IFN has been shown to activate the JAK-STAT signaling pathway. We observed that IFN treatment increased the expression of MxA and OAS1 in HepG2 and HepG2.2.15 cells, and this effect was compromised by reduced TMEM2 expression, leading to attenuated inhibitory effects of IFN on HBV production. Furthermore, we explored whether blocking the JAK-STAT signaling pathway could abolish the anti-HBV effects of TMEM2. Our data revealed that pre-treatment with JAK1 inhibitor resulted in elevated levels of HBsAg, HBcAg, HBV DNA, and HBV cccDNA in TMEM2-overexpressing HepG2 and HepG2.2.15 cells. Interestingly, as shown in Figure 8, TMEM2 caused an increase in the translocation of IRF9 into nuclei, but no effect on IRF9 expression was detected. The increased levels of nuclear IRF9 enhanced the expression of a luciferase reporter gene under the control of ISRE motifs. Given that TMEM2 increased the phosphorylation of STAT1 and STAT2, and promoted the translocation of IRF9 into nuclei, it would be interesting to determine whether TMEM2 is involved in the formation of the ISGF3 complex, a key player in type 1 interferon signaling. Interferon regulatory factors (IRFs) are a family of transcription factors that mediate the transcriptional regulation of interferon genes and IFN-inducible genes. 13 IRF9 is an immunomodulatory transcription factor that is important for the response to IFN-α downstream of the JAK/STAT signaling pathway. 14,15 Taken together, our results suggest that TMEM2 inhibits HBV infection in HepG2 and HepG2.2.15 by activating the JAK-STAT signaling pathway.
Bioinformatics analyses have shown that TMEM2 contains a short intracellular region and a long extracellular region that includes the G8 and GG domains. The G8 domain has also been identified in a number of proteins such as PKHD1 and KIAA1109, which are involved in certain human diseases. For example, it has been reported that the transmembrane protein PKHD1 is associated with kidney and liver diseases. 16 Binding to ligands can phosphorylate the intracellular region of PKHD1, which then regulates a variety of cellular activities. 16,17 How TMEM2 regulates the JAK-STAT signaling pathway, that is, promotes the phosphorylation of p-Tyk2, p-JAK1, p-STAT1, and p-STAT2, was not investigated in the present study.
Current studies investigating HBV receptors have mainly focused on the sodium taurocholate cotransporting polypeptide (NTCP), which is the receptor of both HBV and HDV. [18][19][20][21] The interaction between the HBV preS1 domain and NTCP mediates the entry of HBV into cells. Our previous in vivo TMEM2 inhibits HBV infection X Zhu et al study confirmed that NTCP is the receptor of HBV and demonstrated that the NTCP mutation (p.Ser267Phe) was associated with a reduced prevalence of acute and chronic liver failure. 22 In particular, we identified five cases of the p. Ser267Phe homozygous mutation among 1899 patients with chronic HBV infection. Given that five homozygous patients were infected with HBV, we speculate that there is more than one route of HBV cellular entry. Characterization of the biological functions of TMEM2 is useful to understand the mechanism of HBV infection. It is not clear whether TMEM2 is a HBV receptor involved in HBV infection through the G8 and GG domains. In addition, the role of TMEM2 in HBV infection should be investigated in vivo. Whether the structure of TMEM2 is altered during HBV infection and whether TMEM2 interacts with NTCP should be investigated in future investigations.  23 Liver tissues were obtained by liver biopsy. Healthy and CHB liver tissues were formalin-fixed, paraffin-embedded and sectioned for immunohistochemistry assays, or immediately frozen for western blot and RT-qPCR assays. Establishment of stable TMEM2-overexpressing or silenced cell lines. Establishment of stable TMEM2-overexpressing or silenced cell lines was performed according to a previous study with slight modifications. 24 To obtain lentiviral expression of TMEM2, TMEM2 cDNA with an in-frame stop codon was amplified by PCR and inserted into vector LV-003 with the CMV promoter (Forevergen Biosciences, Guangzhou, China). Vector LV-008 (Forevergen Biosciences, Guangzhou, China) with the U6 promoter was used to generate small hairpin RNAs. The following oligonucleotides were sub-cloned into the HpaI/XhoI sites: the small hairpin RNA of TMEM2 (shTMEM2) and the TMEM2 shRNA sequences: sense 5′-CACCGCTTGCCCTTTGGGTCCTATACGAATATAGGACCCAA AGGGCAAGC-3′; and antisense 5′-AAAAGCTTGCCCTTTGGGTCCTATATTCGTAT AGGACCCAAAGGGCAAGC-3′. Lentiviral production was performed as previously described. 8 Briefly, HEK293T cells were co-transfected with LV-003-TMEM2 (or LV-008-shTMEM2) and packaging vectors, and the supernatant was collected after 48 and 72 h of incubation. Lentiviruses were recovered after centrifugation at 25 000 r.p.m. for 1.5 h in a Beckman Instrument (Fullerton, CA, USA) and resuspended in PBS. The virus was then used to infect cells in the presence of polybrene (6 mg/ml). After 48 h, the cells were cultured in medium containing puromycin (2 μg/ml) for 10-15 days to generate stable cell lines. Stable TMEM2overexpressing or silenced cell lines were confirmed by western blot, RT-qPCR and the MTT assay.
Immunohistochemistry. Immunohistochemistry was performed using formalin-fixed and paraffin-embedded healthy and CHB liver tissue sections or fixed cultured cells. The tissue sections or cells were incubated with the primary antibodies (TMEM2 antibody 1   HBV DNA and cccDNA quantification. HBV DNA and cccDNA quantification was performed as described previously. 25,29 HBV DNA was isolated according to standard genomic DNA isolation methods. 30 Briefly, HBV-infected cells were lysed at 65°C for 4 h in lysis buffer (50 mM Tris-HCl, pH 8.0, 50 mM EDTA, 100 mM NaCl, 1% SDS) supplemented with proteinase K (200 μg/ml). HBV DNA was isolated from the lysed cells using the QIAGEN QIAamp DNA mini kit (Qiagen, Hilden, Germany). The DNA was quantified using specific primers, 5′-ATGGAGAA CACAACATCAGG-3′ and 5′-GAGGCATAGCAGCAGGATG-3′, by real-time PCR. A total of 1000 ng of isolated HBV DNA was digested with 20-40 units of plasmid-safe DNase (Epicentre Technologies #E3101K, Madison, WI, USA) in a total volume of 20 μl at 37°C for 8 h. The digested HBV DNA was inactivated with DNase at 70°C for 30 min. Two microliter of the 20-μl reaction was then added to a 20-μl real-time PCR reaction. Primers for cccDNA detection (5′-GTCTGTGCCTTCTCATCTG CC-3′; 5′-ACAGCTTGGAGGCTTGAACAG-3′) were validated and used for realtime PCR.
Immunofluorescence assay. The IFA was performed to localize IRF9. The cells were fixed with 4% paraformaldehyde in PBS containing 0.5% Triton X-100 at 4°C for 30 min. After three washes with PBS, anti-IRF9 antibody (1:100 dilution) was added to the cells and incubated at room temperature for 1 h. The cells were then washed with PBS and incubated with PE-conjugated secondary antibodies (Invitrogen, 1:1000 dilution). Subsequently, the cells were washed with PBS and mounted using VectaShield with DAPI. IRF9 staining was examined using fluorescence microscopy.
Luciferase reporter assay. A luciferase reporter assay was performed to assess the effects of TMEM2-induced IRF9 translocation into nuclei. Plasmid pISRE-TA-Luc (Clontech) containing five copies of consensus ISRE upstream of the firefly luciferase gene was used for cell transfection. Plasmid pTA-Luc (Clontech) lacking the enhancer element was used for background determination. Plasmid pRL-CMV (Promega) expressing the Renilla luciferase protein was used to normalize the transfection efficiency. To measure IRF9-induced ISRE activities, 5 × 10 4 cells per well were seeded in a 24-well plate 1 day before transfection. Five hundred nanograms of pISRE-TA-Luc or pTA-Luc with 1 ng of pRL-CMV were used for cell transfection with 2 μl of Lipofectamine 2000 according to the manufacturer's instructions. Forty-eight hours after transfection, dual luciferase assays were performed using the cell lysates according to established protocols. Real-time quantitative PCR. Total RNA was isolated from tissue samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. RNA was reverse transcribed using the ReverTra Ace qPCR RT Kit (Toyobo Biochemicals, Osaka, Japan) according to the manufacturer's protocol. The mRNA of the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as an endogenous standard. The primers used for RT-qPCR assay were as follows: TMEM2 forward, 5′-GGAGATATGCTCCGTCTGACC-3′ and TMEM reverse, 5′-CATCTGACTTGCCATACAAGGT-3′; MxA forward, 5′-GTTTCCGAAGTG GACATCGCA-3′ and MxA reverse, 5′-CTGCACAGGTTGTTCTCAGC-3′; OAS1 forward, 5′-CGTGTTTCCGCATGCAAAT-3′ and reverse, 5′-ACCTCGGAAGCACCTT TCCT-3′; GAPDH forward, 5′-GAGTCAACGGATTTGGTCGT-3′ and reverse 5′-GAC AAGCTTCCCGTTCTCAG-3′. All PCR reactions were performed using a SYBR Green Master kit (Roche, Basel, Switzerland) according to the manufacturer's protocol. The threshold cycle (C t ) value of each sample was calculated, and the relative mRNA level was normalized to the GAPDH mRNA value.
Western blot analysis. Total protein was isolated from cells or tissue samples using RIPA lysis buffer. The protein concentration was determined using the Bradford assay. Equal amounts of total protein were separated by SDS-PAGE electrophoresis, transferred to PVDF membranes, and blocked with 5% skim milk powder at room temperature for 1 h. Next, the PVDF membranes were washed with TBST containing NaCl, Tris-HCl, and Tween-20 and incubated with primary antibodies against target proteins at 4°C overnight, followed by two washes with TBST. TMEM2 antibody was purchased from Aviva Systems Biology. Antibodies against p-Tyk2, Tyk2, p-JAK, JAK, p-STAT1, STAT1, p-STAT2, STAT2, IRF9, and GAPDH were obtained from Cell Signaling Technology. PVDF membranes were incubated with the appropriate secondary antibodies at room temperature for 1 h and washed three times with TBST. Protein bands were visualized by chemiluminescence (ECL, Forevergen, Guangzhou, China).
Statistical analyses. Data from all cell culture-based assays are presented as the mean ± S.D. ANOVA tests were used to evaluate differences among groups. A P-value o0.05 was considered statistically significant.

Conflict of Interest
The authors declare no conflict of interest.