Potential conflict of interest: Nothing to report.
Members of the tripartite motif (TRIM) family are a part of the innate immune system to counter intracellular pathogens. TRIM22 has been reported to possess antiretroviral activity. Here we report that TRIM22 is involved in antiviral immunity against hepatitis B virus (HBV). Our results showed that TRIM22, being a strongly induced gene by interferons in human hepatoma HepG2 cells, could inhibit HBV gene expression and replication in a cell culture system as well as in a mouse model system. Importantly, it was found that TRIM22 could inhibit the activity of HBV core promoter (CP) in a dose-dependent manner. However, TRIM22 lacking the C terminal SPRY domain lost this activity. Further study showed that the SPRY domain deletion mutant was localized exclusively to the cytoplasm of HepG2 cells. In contrast, the wild-type TRIM22 was localized to the nucleus, as expected for a transcriptional suppressor. Interestingly, although RING domain mutants of TRIM22 were localized to the nucleus, they could not inhibit HBV CP activity, indicating that TRIM22-mediated anti-HBV activity was dependent on the nuclear-located RING domain. Conclusion: These findings suggest that TRIM22, which exhibits anti-HBV activity by acting as a transcriptional suppressor, may play an important role in the clearance of HBV. (HEPATOLOGY 2009.)
With over 300 million carriers, hepatitis B virus (HBV) infection remains a major public health problem worldwide.1 Classified in the Hepadnaviridae family, HBV is a small, enveloped DNA virus with a genome size of 3.2 kb. HBV replicates its partially double-stranded DNA genome within core particles by reverse transcription of encapsulated 3.5-kb pregenomic RNA (pgRNA), and thus is related to retroviruses.2 The core promoter (CP) is responsible for the synthesis of pgRNA, and therefore the regulation of this promoter is important in the viral life cycle. It is well established that resolution of HBV infection is critically dependent on adaptive immunity, especially on HBV-specific cytotoxic T lymphocytes response. However, many studies also indicate that an innate immune response is crucial for early clearance of HBV infection.3 For example, activation of Toll-like receptor signaling can inhibit HBV replication in vivo, and overexpression of an innate antiviral molecule, APOBEC3G, has also been shown to interfere with HBV replication efficiently.4, 5
Recent studies show that many members of the tripartite motif (TRIM) superfamily are expressed in response to interferons (IFNs) and display antiviral properties, targeting retroviruses in particular.6, 7 TRIM5α, TRIM19, TRIM22, and TRIM28 were all demonstrated to play important roles in antiretroviral activities.8–12 It has been speculated that the TRIM proteins may represent a new and widespread class of antiviral molecules involved in innate immunity.6, 7 The TRIM family is characterized by a combination of RING, B-Box, and coiled-coil domains, followed by one of several C-terminal domains.13 To date, nearly 70 TRIM family members have been identified, yet the most intensively studied TRIM protein may be TRIM5α, which has been shown to inhibit the infectivity of a range of different retroviruses in a species-specific manner.8, 9
One of the closest paralogs of TRIM5α is TRIM22, which sits directly in the TRIM5 gene cluster. Evolutionary genetic analyses showed that TRIM5 and TRIM22 have evolved under positive selection in a mutually exclusive fashion.14 Several studies indicated that TRIM22 also possessed antiretroviral activity in certain cell types. For example, overexpression of TRIM22 can repress the transcription initiated by the long terminal repeat (LTR) promoter of human immunodeficiency virus (HIV) type-1 and inhibit HIV-1 replication in human monocyte-derived macrophages.15, 16 Recently it was reported that TRIM22 was the key mediator for type I IFN to inhibit HIV-1 replication.10 Additionally, TRIM22 has been implicated in normal hematopoietic differentiation and in diseases such as systemic lupus erythematosus and Wilms tumor.17–19
In the present study we found that TRIM22 was one of the most strongly induced TRIM family molecules in human hepatoma HepG2 cells after treatment with IFNs, which have been demonstrated to be able to inhibit HBV replication noncytopathically.20 Chisari and colleagues21 also reported an association between TRIM22 and HBV clearance by microarray analysis in acutely HBV-infected chimpanzees. We therefore hypothesized that TRIM22 might possess antiviral activity against HBV. Our results showed that TRIM22 could inhibit HBV gene expression and replication in cultured cells and in mice. Importantly, it was shown that TRIM22 could significantly inhibit the activity of HBV CP, which plays a central role in HBV replication. Further study showed that the inhibitory effect of TRIM22 on HBV CP activity was dependent on SPRY domain-mediated nuclear localization and the RING domain.
AFP, α-fetoprotein; CP, core promoter; HBcAg, hepatitis B core antigen; HBeAg, hepatitis E antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HIV, human immunodeficiency virus; IFN, interferon; pgRNA, pregenomic RNA; TRIM, tripartite motif.
Materials and Methods
The plasmid THBV-1 (pTHBV-1) contains a head-to-tail dimer of the HBV DNA (ayw1) genome. After transfection into HepG2 cells, pTHBV-1 can express HBV proteins and initiate HBV replication.22 Plasmids expressing myc-tagged wild-type TRIM22, RING domain deletion, and point mutants of (ΔR, C15A) were described in our previous study.23 The SPRY domain deleted TRIM22 or TRIM22 SPRY domain with C-terminal myc epitope was cloned into pcDNA3.1, referred to as pTRIM22-ΔSPRY or pTRIM22-SPRY, respectively. For the construction of the HBV CP-dependent reporter plasmid, one complete HBV genome sequence as described by Raney et al24 (nucleotides 1805 to 3182/1 to 1804) or HBV CP element (nucleotides 1633 to 1804) was cloned into pGL3-Basic (Promega, Madison, WI), so that the expression of the luciferase gene was under the control of the HBV CP.
Cell Culture and Transfection.
HepG2 or Hela cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (Gibco, Grand Island, NY) at 37°C in a 5% CO2 / 95% air mixture. Cells were transfected with the indicated plasmids by the calcium phosphate precipitation method. Precipitates remained on cells for 16 hours, and then were washed two times with phosphate-buffered saline (PBS) before fresh media was added.
Total RNA was isolated from HepG2 cells with TRIzol reagent (Invitrogen, Carlsbad, CA) and was reverse-transcribed (RT) using a complementary DNA (cDNA) synthesis kit (MBI Ferments, St Leon-Roth, Germany) according to the manufacturer's instructions. Subsequently, cDNA was subjected to quantitative real-time PCR using a Lightcycler480 and SYBR Green system (Roche Diagnostics, Mannheim, Germany) following the manufacturer's protocol.
Protein samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) in 12% polyacrylamide gels and transferred onto a nitrocellulose membrane (Amersham Biosciences, Sweden). The membranes were then blocked with nonfat milk and incubated with the indicated antibody. The immunoblot signals were detected by SuperSignal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL).
HBV Protein Analysis.
HepG2 cells (1.5 × 105) were seeded on 12-well dishes and transfected with 1 μg of pTHBV, 0.5 μg of plasmid cytomegalovirus (pCMV)-β-galactosidase (β-gal), and indicated amounts of TRIM22 expression plasmid (0 μg, 0.5 μg, 1.0 μg, 2.0 μg). Where necessary, empty control plasmid was added to ensure that each transfection received the same amount of total DNA. Three days after transfection the cell supernatant was collected to determine levels of hepatitis B surface antigen (HBsAg) and hepatitis E antigen (HBeAg) using AXSYM systems (Abbott Diagnostics, Abbott Park, IL). Transfection efficiency was normalized by detecting the activity of β-gal in cell lysates using the β-Galactosidase Enzyme Assay System (Promega).
Northern Blot and Southern Blot Analysis.
HepG2 cells were seeded onto 60-mm dishes (6 × 105 cells per dish) and transiently transfected with pTHBV-1 (5 μg), pcDNA, or pTRIM22 (10 μg) and pCMV-β-gal (2.5 μg). HBV viral RNA and replicative DNA intermediates were detected by northern and Southern blot analysis, respectively. To detect HBV-RNAs, 30 μg of total RNA were separated by electrophoresis on a 1% formaldehyde-agarose gel, transferred to a nylon membrane, and detected with a digoxigenin (DIG)-labeled full-length HBV probe synthesized from the DIG Probe Synthesis Kit (Roche Diagnostics). The method for purification of cytoplasmic core-associated HBV DNA was adapted from the protocol described by Pugh et al.25 Briefly, cells were lysed and nuclei were pelleted by centrifugation. The supernatants were then treated with DNase I to remove retained plasmid. The cytoplasmic HBV DNA was purified by digestion of proteins with 200 μg/mL of proteinase K in the presence of 0.5% of SDS at 37°C for overnight, phenol-chloroform extraction, and ethanol precipitation. Purified DNA was separated on a 1.0% agarose gel, blotted onto a nylon membrane, and probed with a DIG-labeled full-length HBV genome.
Hydrodynamics-Based Transfection in Mice.
For the in vivo experiments, 6 to 8-week-old female Balb/c mice were used. The pTHBV-1 (10 μg), was injected into the tail vein of mice together with pcDNA or pTRIM22 (20 μg) and pCMV-β-gal (5 μg) within 5-7 seconds in a volume of PBS equivalent to 10% of the mouse body weight. Animals were sacrificed after 4 days. Sera were taken for the measurement of HBsAg, HBeAg, and HBV DNA. HBsAg and HBeAg were detected using the AXSYM systems. After removal of retaining plasmid by DNase I in mice sera, HBV DNA was determined by real-time PCR using an HBV diagnostic kit (PG Biotech, Shenzheng, China) as described by the manufacturer's instructions. Formalin-fixed or -unfixed liver sections were processed for immunohistochemical hepatitis B core antigen (HBcAg) detection or for β-gal staining, respectively. All mice were housed in a pathogen-free mouse colony at our institution and the animal experiments were performed according to the Guide for the Care and Use of Medical Laboratory Animals (Ministry of Health, P.R. China, 1998).
Paraffin-embedded liver sections were treated with 3% hydrogen peroxide and blocked with 5% bovine serum albumin. The sections were then incubated sequentially with anti-HBc antibodies (Dako, Glostrup, Denmark), biotin-labeled secondary antibody, and avidin-biotin complex (ABC). Peroxidase stain was developed with 3,3′-diaminobenzidine (DAB) solution and counterstained with hematoxylin.
CP-dependent luciferase report plasmids (CpLUC) was transfected into cells together with plasmids expressing wild-type TRIM22 or its mutants. After transfection for 48 hours, cells were harvested with the addition of cell lysis buffer and the luciferase assays were performed according to the manufacturer's instructions (Promega).
The intracellular localization of TRIM22 or its mutants in HepG2 cells was detected by indirect immunofluorescence staining as described.23
Preparation of Antibodies.
Rabbit polyclonal antibodies to TRIM22 were generated using peptides corresponding to the N-terminal amino acid sequence of human TRIM22 (MDFSVKVDIEKEVTC). The synthetic peptide was linked to the carrier protein KLH through a cysteine added to the C-terminus and used as an immunogen. The specific antibody was purified from immune sera using columns of TRIM22 peptide coupled to Affigel 10 (Bio-Rad, San Diego, CA).
The results are reported as means ± standard deviation (SD). A t test was applied to comparisons between groups; a P-value < 0.05 was considered statistically significant.
Expression of TRIM Family Molecules in IFN-Treated Human Hepatoma HepG2 Cells.
The association of some TRIM family members with antiviral activities coupled with the fact that many of them are induced by IFNs has led to the hypothesis that TRIM proteins may represent an important class of antiviral molecules involved in innate immunity. To obtain hints of which TRIM molecule may play roles in anti-HBV immunity, real-time PCR was used to determine the expression profile of TRIM molecules in IFN-treated HepG2 cells. The selected TRIM molecules in this study were demonstrated to be crucial for many aspects of resistance to pathogens, especially to retrovirus.6, 7 The results showed that the induction of TRIM22 by IFNs was most significant among these TRIM molecules in HepG2 cells (Fig. 1A). Other notably up-regulated TRIM molecules included TRIM34, TRIM21, TRIM19, and TRIM5α, consistent with previous studies (Fig. 1A).11, 26–28
Table 1. Oligonucleotides Used in the Current Study
To further investigate whether TRIM22 was induced in an IFN-dependent manner, we examined the expression level of TRIM22 protein by western blot using anti-TRIM22 antibody in HepG2 cells stimulated with IFN-α or IFN-γ. As shown in Fig. 1B, TRIM22 protein expression was markedly increased in response to both IFN-α and IFN-γ treatment in HepG2 cells.
Inhibitory Effect of TRIM22 on HBV Gene Expression and Replication in Transfected HepG2 Cells.
To study the role of TRIM22 in anti-HBV activity, the replication competent plasmid (pTHBV-1) was cotransfected with TRIM22 expression plasmid (pTRIM22) into HepG2 cells. The expression of TRIM22 in transfected cells was detected at the messenger RNA (mRNA) level by semiquantitative RT-PCR and at the protein level by Western blot. The results showed that transfection of pTRIM22 could lead to the expression of TRIM22 at both the mRNA and protein level in a dose-dependent manner in transfected cells (Fig. 2A,B).
The antiviral activity of TRIM22 against HBV was investigated by detecting levels of HBV proteins, RNA, and DNA. The results showed that TRIM22 overexpression could result in a dose-dependent reduction of HBsAg (Fig. 2C) and HBeAg (Fig. 2D) titers in supernatants of transfected cells. Consistent with the results of secreted viral proteins in culture supernatants, the HBV RNA level in transfected cells was significantly suppressed by TRIM22 as determined by northern blot (Fig. 2E). The HBV DNA level in transfected cells was examined by Southern blot. As shown in Fig. 2F, cotransfection of pTRIM22 with pTHBV-1 could significantly decrease the HBV DNA replicative intermediates in transfected HepG2 cells (Fig. 2F).
To rule out the possibility that the inhibition of HBV replication is due to the impact of TRIM22 on the growth or apoptosis of HepG2 cells, we examined the concentration of secreted α-fetoprotein (AFP) in the supernatants of transfected cells, which is an important marker of hepatoma cells. We found that there was no difference in the AFP concentration between pCDNA3.1-transfected cells and pTRIM22-transfected cells (Fig. 3A). Furthermore, we determined the viability of transfected HepG2 cells by the MTT method and the apoptosis of transfected HepG2 cells by Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) staining. These results also showed that TRIM22 treatment had no impact on the growth or apoptosis of HepG2 cells in our experimental system (Fig. 3B,C).
Inhibitory Effect of TRIM22 on HBV Gene Expression and Replication in Mice.
Because TRIM22 showed substantial inhibition on HBV gene expression and replication in cell culture model, further experiments were done to examine whether TRIM22 could display anti-HBV activity in vivo. pTHBV-1 and pTRIM22 were codelivered to mouse livers by hydrodynamic injection and the antiviral activity of TRIM22 was evaluated on Day 4 posttransfection. The expression of TRIM22 in transfected mouse liver was detected by Western blot (Fig. 4A). In concordance with findings from cell culture experiments, cotransfection of pTRIM22 with pTHBV-1 could reduce the HBsAg, HBeAg, and HBV-DNA level in mice sera significantly (Fig. 4B,C). Immunohistochemical analysis showed that TRIM22 could inhibit the HBcAg expression in the mouse liver (Fig. 4D). To rule out the possible toxicity contributed by TRIM22, alanine aminotransferase (ALT) levels in sera of mice were determined. No difference in ALT levels was observed between mice treated with pcDNA3.1 and pTRIM22 (data not shown).
Inhibition of HBV CP Activity by TRIM22.
Because HBV CP is responsible for the synthesis of pgRNA, which is crucial for HBV replication, and several TRIM molecules have been demonstrated to perform their antiviral activity through transcriptional suppression,11, 12, 29 we therefore hypothesized that TRIM22 might inhibit HBV replication through down-regulating HBV CP activity. To this end, two (CpLUC) were utilized: one containing the whole HBV genome (nucleotides 1805 to 3182/1 to 1804), and the other containing an HBV CP element (nucleotides 1633 to 3182/1 to 1804) (Fig. 5A). Each CpLUC was cotransfected into HepG2 cells with an increasing dose of pTRIM22, and the luciferase activity was measured on Day 2 after transfection. The results showed that overexpression of TRIM22 could significantly inhibit the HBV CP activity in a dose-dependent manner, and the two report plasmids gave very similar result (Fig. 5B). Considering the construct containing the whole HBV genome is more commonly used,24, 30, 31 we kept using it in the following studies. Interestingly, the inhibitory effect of TRIM22 on HBV CP activity was not observed in the human nonhepatic cell line Hela, indicating TRIM22-mediated suppression of HBV CP activity might be cell-type-specific (Fig. 5C).
Inhibitory Effect of TRIM22 on HBV CP Activity Is Associated with SPRY Domain-Mediated Nuclear Localization.
TRIM22 contains a C-terminal SPRY domain, which is often involved in protein-protein interaction. A recent study showed that the SPRY domain was important for the positive evolutionary selection of TRIM22. To evaluate the role of the SPRY domain in TRIM22-mediated inhibition of HBV CP activity, plasmid-expressing a SPRY domain deletion mutant (TRIM22-ΔSPRY) was constructed (Fig. 6A), and its effect on HBV CP activity was evaluated in HepG2 cells. The results showed that TRIM22-ΔSPRY lost the ability to inhibit HBV CP activity as compared with the wild-type control (Fig. 6B).
To address why TRIM22-ΔSPRY lost the ability to inhibit HBV CP activity, myc-tagged TRIM22-ΔSPRY or wild-type TRIM22 was tested for its subcellular distribution in HepG2 cells by immunofluorescence staining using anti-myc antibody. To our surprise, TRIM22-ΔSPRY was found to be localized exclusively to the cytoplasm of HepG2 cells, whereas wild-type TRIM22 was predominantly localized to the nucleus, as expected for a transcriptional factor (Fig. 6C). To further confirm the nuclear localization of wild-type TRIM22 in HepG2 cells, the cytoplasmic and nuclear extracts of HepG2 cells were prepared and endogenous TRIM22 was detected by western blot using anti-TRIM22 antibody. The results showed that endogenous TRIM22 was also localized to the nuclei of HepG2 cells (Fig. 6D). We tried to perform intracellular localization of endogenous TRIM22 in HepG2 cells through immunofluorescence staining using anti-TRIM22 antibody, but it failed to obtain specific fluorescent signal, which might be due to the low basal expression level of TRIM22 in HepG2 cells.
To further investigate the contribution of the SPRY domain for the intracellular localization and functional activity of TRIM22, the SPRY domain alone was tested for its intracellular localization and transcriptional repression activity. It was found that the SPRY domain alone could localize to the nucleus (Fig. 6B), but failed to suppress HBV CP activity (Fig. 6C). These results indicated that the SPRY domain mediated the nuclear localization of TRIM22, but it might not be responsible for suppressing HBV CP activity directly.
RING Domain of TRIM22 Is Important for the Inhibition of HBV CP Activity.
To further explore the mechanism of TRIM22 in inhibition of HBV CP activity, the contribution of the N terminal RING domain was then investigated. The results showed that TRIM22 with the RING domain deletion mutant (ΔR) (Fig. 7A) was localized to the nucleus (Fig. 7B), indicating that the RING domain was not implicated in the intracellular localization of TRIM22. However, the functional analysis showed that TRIM22-ΔR failed to inhibit HBV CP activity (Fig. 7C).
The conserved cysteine residue at position 15 is critical for RING domain-mediated biological functions. To further investigate the role of the RING domain in the transcriptional suppression, the 15th cysteine of the RING domain was substituted into the alanine (C15A) (Fig. 7A). Just the same as TRIM22-ΔR, the C15A mutant was also localized to the nuclei of HepG2 cells (Fig. 7B) and lost the ability to inhibit HBV CP activity (Fig. 7C). These data indicated that the nuclear-located RING domain was crucial for the inhibition of HBV CP activity, although the RING domain per se was not required for the intracellular localization of TRIM22.
TRIM family proteins have been involved in the regulation of pathogen recognition and transcriptional pathways in host antiviral defense.6, 7 For example, TRIM25 is essential for RNA helicase RIG-I-mediated antiviral activity32; TRIM19 inhibits the human foamy virus (HFV) transcription by complexing the HFV transactivator Tas11; and TRIM28 can inhibit the replication of murine leukemia viruses (MLVs) or related retroelements in embryonic cells by transcriptional silencing.12, 29 In this study, we found that TRIM22 could significantly inhibit HBV replication both in a cell culture system and in a mouse model system. Importantly, our results showed that TRIM22 could significantly down-regulate the activity of HBV CP, which plays a crucial role in the HBV life cycle, suggesting that TRIM22 acted as a transcriptional suppressor in its anti-HBV process. Because it was also reported that TRIM22 could inhibit the transcription initiated by HIV-LTR,15 we assume that transcriptional repression activity may be important for TRIM22-mediated biological activities. As TRIM22 was identified as being associated with HBV clearance in acutely HBV-infected chimpanzee,21 we therefore speculate that TRIM22 may possess anti-HBV activity under physiological conditions.
The TRIM22 protein contains a C terminal SPRY domain, which is present in a large number of proteins with diverse individual functions in different biological processes. Some studies suggest that the SPRY domain is a protein-interacting module. Deletion of the SPRY domain in TRIM11, TRIM21, and TRIM25 abrogated the interaction of the proteins with their binding partner proteins.26, 32, 33 In this study we found that the SPRY domain was essential for the nuclear localization of TRIM22, and TRIM22 lacking the SPRY domain was localized exclusively to the cytoplasm of HepG2 cells and lost its suppressive activity on HBV CP. As no nuclear localization signal (NLS) is found in the SPRY domain of TRIM22 and the SPRY domain has often been implicated in protein-protein interactions, we suspect that TRIM22 may be transported into the nucleus by binding to other nuclear proteins through its SPRY domain. Further results showed that the SPRY domain alone failed to inhibit HBV CP activity, indicating that there might be other functional domains responsible for TRIM22-mediated transcriptional suppression.
We then focused on the contribution of the N terminal RING domain for the functional activity of TRIM22. Interestingly, we found that the RING domain itself was not required for the nuclear localization of TRIM22, but it was crucial for TRIM22 to inhibit HBV CP activity. In addition, we found that the wild-type TRIM22 could form speckles in the nucleus, consistent with our previous study in COS7 cells,23 whereas the RING domain mutants were uniformly distributed, which might suggest that the RING domain was critical for forming a protein complex through binding its partner proteins. Taken together, we speculate that TRIM22 may serve as a molecule scaffold responsible for forming protein complex in the nucleus, and inhibit HBV CP activity through interacting with the liver transcriptional factor in the nucleus, which deserves further study.