DNA Sensing-Independent Inhibition of Herpes Simplex Virus 1 Replication by DAI/ZBP1

Cell culture and virus.

HepG2, MEF, Vero, BHK, U2OS, and 293T cells were grown in Dulbecco modified Eagle medium with 10% fetal calf serum, penicillin (100 U/ml), and streptomycin (100 μg/ml) in a 5% CO₂ humidified incubator at 37°C. To induce DAI expression HepG2 cells were treated with IFN-γ (100 ng/ml; Sigma-Aldrich) for 20 h. HSV-1 (strain KOS/tk12) contained a lacZ reporter gene under the control of the ICP4 promoter inserted into the thymidine kinase (TK) gene (20). The mutant virus dlX3.1 has an ICP0 deletion in the KOS background (21). The wild-type virus (KOS/tk12) was grown and titrated in Vero cells by plaque assays. The dlX3.1virus was grown in BHK cells and titrated in U2OS cells, in which ICP0 is not required for efficient replication of HSV-1 (22).

Plasmids.

Plasmids expressing 5′ hemagglutinin (HA)-tagged human DAI (1 to 429 amino acids), its amino-terminal fragment (Zα+Zβ), or an amino-terminal truncated version (Δ1-80) were previously described (23). The ΔExon2 (ΔE2) mutant with a deletion between amino acids 12 and 86 and the ΔExon4/5 (ΔE4/5) mutant with a deletion between amino acids 110 and 223 were generated by PCR in the same pSG5 (24) background. Retroviral vectors, MIN-DAI and MIN-ΔE4/5, were constructed by inserting the corresponding DAI cDNAs into pMIN (25). The cDNA for human RIP1 was provided by Jaehwan Song (Yonsei University, Seoul, Republic of Korea). Plasmid for 5′-HA-tagged RIP1 was produced in pSG5 background using Gateway technology (Invitrogen). Plasmid encoding ΔRHIM mutant RIP1 was generated by PCR in the same background. Plasmid expressing ICP0 was previously described (26) and an ICP0-luciferase reporter gene plasmid was provided by Gary S. Hayward (Johns Hopkins University School of Medicine).

Transfection.

HepG2 and MEF cells were transfected with plasmid DNAs using the Metafectene reagents (Biontex). For cytosolic B-DNA stimulation, HepG2 cells were transfected with poly(dA-dT)-poly(dA-dT) (Sigma-Aldrich) using Lipofectamine 2000 reagents (Invitrogen). 293T cells were transfected via the N,N-bis-(2-hydroxyethyl)-2-aminoethanesulfonic acid-buffered saline (BBS) version of the calcium phosphate method, as described previously (26).

Lentiviral vectors.

Lentiviral vector pLKO.1-TRC control expressing a non-hairpin control RNA (shC) was purchased from Addgene. Lentiviral vectors (on a pLKO.1 background) expressing short hairpin RNAs (shRNAs) against DAI (shDAI-A to E), which correspond to TRCN0000123049 to TRCN0000123053, respectively, were purchased from Open Biosystems. To produce lentiviruses, 293T cells were transfected with lentiviral vectors together with plasmids pCMV-DR8.91 expressing the Gag-Pol, Tat, and Rev proteins of human immunodeficiency virus and pMD-G expressing the envelope G protein of vesicular stomatitis virus (VSV), provided by Kyeong Ho Lee (Yonsei University Wonju Campus, Wonju, Republic of Korea). At 2 days after transfection, the culture media containing the lentiviruses were harvested and stored at −70°C. HepG2 cells were transduced with lentivirus supernatants in the presence of Polybrene (7.5 μg/ml), and then were selected with puromycin (1 μg/ml). The selected cells were maintained in a medium containing puromycin (0.5 μg/ml). Two shRNAs, shDAI-B and shDAI-E, effectively suppressed DAI expression and were referred to as shDAI-1 and shDAI-2.

Retroviral vectors.

Recombinant retroviruses were prepared by cotransfecting 293T cells with retroviral vectors MIN or MIN-DAI (wild-type or mutant) together with packaging plasmids pHIT60 (Gag-Pol) and pMD-G (VSV-G) (27) using Metafectene reagents (Biotex). Viral supernatants were harvested at 48 h after transfection. HepG2 cells were transduced by retroviruses in the presence of Polybrene (7.5 μg/ml). Cells were selected with G418 (0.5 mg/ml; Calbiochem).

Plaque assays.

The serially diluted virus samples were used to inoculate Vero or U2OS cells in a six-well plate (10⁶ per well). After 1 h of incubation, the medium was removed, and 2 ml of methylcellulose overlay was added to each well. After incubation for 3 days, the methylcellulose was removed, and the cell monolayer was stained with 1% crystal violet in 50% ethanol for 20 min. The cells were washed with running tap water and dried. The numbers of plaque produced were counted.

Antibodies.

Anti-DAI polyclonal antibody (PAb) (23) was previously described. Anti-HA rat monoclonal antibody (MAb) 3F10 conjugated with peroxidase or labeled with fluorescein isothiocyanate (FITC) and anti-Myc mouse MAb was purchased from Roche. Mouse MAbs for UL42 and IFI16 and rabbit PAb for IRF3 were purchased from Santa Cruz Biotechnology. Mouse MAb for ICP0 was purchased from Virusys. The rabbit PAbs against ICP0 and PML were a gift from Gary S. Hayward (Johns Hopkins University School of Medicine). Rabbit MAb for phosphorylated IRF3 (S396) was obtained from Cell Signaling Technology. Mouse MAbs for IκBα and β-actin were purchased from Young In Frontier (Seoul, Republic of Korea) and Sigma-Aldrich, respectively.

Indirect immunofluorescence assay (IFA).

Cells were fixed for 5 min in cold methanol and rehydrated in cold phosphate-buffered saline (PBS). The cells were incubated with appropriate primary antibody in PBS at 37°C for 1 h and then with secondary antibody in PBS at 37°C for 45 min. To stain nuclei, mounting solution containing DAPI (4′,6′-diamidino-2-phenylindole; Vector Laboratories) or Hoechst strain (Invitrogen) and antifade reagent (Molecular Probes) were used. Slides were examined and photographed with a Carl Zeiss Axiophot microscope or with a Carl Zeiss Axioplan 2 confocal microscope system with LSM510 software.

Immunoblot analysis.

Cells were collected and lysed with an ice-cold RIPA buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 1.0% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS]) supplemented with protease inhibitor cocktail (Santa Cruz). Clarified cell extracts were separated by SDS-polyacrylamide gel electrophoresis (PAGE), and the resolved proteins were transferred onto nitrocellulose membranes (Schleicher & Schuell, Dassel, Germany). All subsequence procedures were performed as previously described (23).

β-Galactosidase reporter assay.

Cells were collected and lysed by three freeze-thaw steps in 100 μl of 0.25 M Tris-HCl (pH 7.9) plus 1 mM dithiothreitol. Cells extracts were clarified in a microcentrifuge. Each reaction mix was prepared in 96-well dishes including 20 μl of extracts plus 100 μl of Z-buffer (0.06 M Na₂HPO₄, 0.04 M NaH₂PO₄, 0.01 M KCl, 0.001 M MgSO₄, 0.05 M β-mercaptoethanol) plus 20 μl of ONPG (o-nitrophenyl-β-d-galactopyranoside [Sigma-Aldrich]; 4 mg/ml in sterile water). Reactions were incubated at 37°C until the yellow color was developed. Reaction was stopped by adding 50 μl of 1 M Na₂CO₃. The β-galactosidase absorbance of each reaction was read at 420 nm.

Cell viability assay.

Cells were seeded in 96-well flat-bottom plates and infected with virus at an MOI of 3. At 24 h after infection, 10 μl of WST-1 solution (EZ-Cytox; Daeil Lab Service) was added to the 96-well cell culture, and the plates were incubated at 37°C in 5% CO₂ for 2 h. The colorimetric determination of WST-1 was measured at 460 nm with a xMARK microplate spectrophotometer plate reader (Bio-Rad).

Reverse transcription-PCR (RT-PCR) and quantitative real-time PCR.

Total RNA was isolated from 2 × 10⁵ cells using TRIzol reagent (Invitrogen). cDNAs were synthesized using the random hexamer primers in the SuperScript III system (Invitrogen). Quantitative real-time PCR was performed using SYBR Green PCR Core Reagents (Applied Biosystems) and ABI Prism SDS software. The following primers were used: human DAI, forward (5′-CAACAACGGGAGGAAGACAT-3′) and reverse (5′-TCATCTCATTGCTGTGTCCC-3′); ICP4, forward (5′-CTCATCGTGGTCAACACCGT-3′) and reverse (5′-CTTCCCAGTCCACAACGTCC-3′); ICP27, forward (5′-AGCCTCTATCGCACTTTTGC-3′) and reverse (5′-ACACGACTCGAACACTCCTG-3′); IFN-β, forward (5′-GCTCTCCTGTTGTGCTTCTCCAC-3′) and reverse (5′-CAATAGTCTCATTCCAGCCAGTGC-3′); and β-actin, forward (5′-AGCGGGAAATCGTGCGTG-3′) and reverse (5′-CAGGGTACATGGTGGTGCC-3′).

Luciferase reporter assay.

Cells were collected and lysed by three freeze-thaw steps in 200 μl of 0.25 M Tris-HCl (pH 7.9) plus 1 mM dithiothreitol. Cell extracts were clarified in a microcentrifuge, and 30-μl portions of the of extracts were incubated with 350 μl of reaction buffer A (25 mM glycyl-glycine [pH 7.8], 15 mM ATP, 4 mM EGTA) and then mixed with 100 μl of 0.25 mM luciferin (Sigma-Aldrich) in reaction buffer A. A TD-20/20 luminometer (Turner Designs) was used for a 10-s assay of the photons produced (measured in relative light units).

CoIP assays.

Cell lysates were prepared by sonication in 1 ml of coimmunoprecipitation (CoIP) buffer (50 mM Tris-Cl [pH 7.4], 50 mM NaF, 5 mM sodium phosphate, and 0.1% Triton X-100 containing protease inhibitors; Sigma-Aldrich) with a Vibra cell microtip probe (Sonics and Materials) for 10 s (pulse on, l s; pulse off, 3 s). Cell lysates were incubated for 16 h with the appropriate antibody at 4°C. A total of 30 μl of a 50% slurry of protein G-Sepharose (Amersham) was added and then absorbed for 1 h at 4°C. The mixture was pelleted and washed several times with CoIP buffer. The beads were resuspended and boiled for 5 min in loading buffer. Each sample was analyzed by SDS-PAGE and immunoblotting with the appropriate antibody.

Statistical analysis.

Statistical significance between samples was determined using the Student t test. P values of <0.05 (*), <0.01 (**), and <0.001 (***) are indicated in the figures.

Depletion of DAI enhances HSV-1 replication.

DAI has been shown to be induced in hepatic cells by interferon treatment and HBV infection (8, 23). In RT-PCR analysis, we found that DAI transcription was induced in HepG2 cells after IFN-γ treatment or HSV-1 infection (Fig. 1A). Human hepatic cells have been used successfully to study intrinsic restriction factors such as PML, Sp100, ATRX, and Daxx in HSV-1 infection (28, 29). To investigate the possible role of DAI in HSV-1 replication, DAI-depleted HepG2 cells were produced by transduction with lentiviral vectors expressing shRNAs for DAI. Transcription of endogenous DAI was suppressed by two different shRNAs (shDAI-1 and shDAI-2) but not by control RNA (shC), and shDAI-1 and shDAI-2 efficiently suppressed DAI transcription in HSV-1-infected cells (Fig. 1B). The expression of DAI protein from transfected plasmid DNA was also efficiently suppressed in these DAI shRNA-expressing HepG2 cells (data not shown). We infected control and DAI-knockdown cells with recombinant HSV-1 (KOS/tk12 strain) at an MOI of 1 or 3 for 24 h and measured the amounts of progeny virions produced in the culture supernatants by plaque assays using Vero cells. Virus titers were higher in shDAI-1 and shDAI-2 cells by about 10- and 15-fold, respectively, than those in normal and control (shC) cells at an MOI of 1 (Fig. 1C). Similarly, a 4- to 6-fold increase of virion production was observed in DAI-knockdown cells at an MOI of 3 (Fig. 1D). Notably, the repressive effect of DAI was more prominent at lower MOI. The plaque sizes were comparable between control and DAI-knockdown cells (data not shown). When the accumulation of viral ICP0 (immediate-early) and UL42 (early) proteins was compared between control and DAI-knockdown cells at 24 h after infection, the levels of viral ICP0 and UL42 proteins were significantly higher in DAI-knockdown cells at MOI of 1, whereas they were slightly increased at MOI of 3 (Fig. 1E and F). The increased ICP0 expression by DAI-knockdown was observed at 3 h after infection (Fig. 1G). Together, these results demonstrate that DAI knockdown in HepG2 cells facilitates HSV-1 replication by promoting viral gene expression and that this effect is more profound at a lower MOI.

DAI expression represses HSV-1 growth and ICP0 promoter activation.

We next investigated the effect of ectopic expression of DAI on HSV-1 replication. Retroviral vectors expressing wild-type or mutant DAI (ΔE2, Zα domain deleted; ΔE4/5, Zβ and D3 domains deleted; and Zα+Zβ, containing only Zα and Zβ domains) were constructed (Fig. 2A). HepG2 cells expressing these DAI proteins were produced by retroviral transduction. These cells were infected with HSV-1 for 24 h and the production of β-galactosidase, which is expressed under the control of the ICP4 promoter in the viral genome, in infected cells was measured. Expression of wild-type and ΔE2 mutant DAI efficiently repressed the β-galactosidase production, whereas ΔE4/5 and Zα+Zβ mutants almost completely lost the repressive activity (Fig. 2B). When the levels of progeny virion production were compared between wild-type and mutant DAI-expressing cells, wild-type DAI significant reduced virus titer, whereas ΔE4/5 mutant had a minimal effect (Fig. 2C). The repressive effect of wild-type DAI on viral replication correlated with the reduced expression of ICP0 and UL42 (Fig. 2C). Both wild-type and ΔE4/5 mutant DAI expressed by retroviral vectors were mainly distributed in the cytoplasm as a mixture of diffuse and punctate forms, with some also being localized in the nucleus as foci (Fig. 2D). When HepG2 cells was transiently transfected with wild-type and ΔE4/5 mutant DAI and then infected with HSV-1 for 24 h, wild-type DAI inhibited the production of β-galactosidase in a dose-dependent manner, whereas ΔE4/5 mutant did not (Fig. 2E). The levels of ICP0 and ICP27 mRNAs were reduced by the expression of wild-type DAI but not ΔE4/5 mutant, suggesting that the anti-HSV-1 effect of DAI expression includes transcriptional control (Fig. 2F). Collectively, these results demonstrate that DAI inhibits HSV-1 growth and that the anti-HSV-1 activity requires the central region encompassing Zβ and D3 domains but not the amino-terminal region containing Zα domain.

To investigate whether DAI expression affects HSV-1 IE gene expression, transient reporter gene assays were performed in HepG2 cells using the ICP0-luciferase reporter gene plasmid. When control and DAI-knockdown cells were transfected with the reporter gene plasmid, the ICP0 promoter activity was higher in DAI-knockdown cells than control cells (Fig. 3A). When cells were cotransfected with the reporter gene plasmid and DAI-expressing plasmid, DAI repressed the ICP0 promoter in a dose-dependent manner (Fig. 3B). Consistent with the data obtained with virus infection (in Fig. 2), DAI mutants without the Zα domain (Δ1-80 and ΔE2) still showed wild-type activity, but ΔE4/5 mutant lost this repressive activity (Fig. 3C). When cells expressing wild-type DAI or ΔE4/5 mutant were transfected with the reporter gene plasmid, the ICP0 promoter was repressed by wild-type DAI, but not by ΔE4/5 mutant (Fig. 3D). In a control experiment, DAI repressed the human T cell leukemia virus (HTLV) long terminal repeat (LTR) but not the cyclin E promoter (Fig. 3E). This result suggests that DAI may be a general repressor, although the activity is found in a promoter-dependent manner. Therefore, these results of reporter gene analysis demonstrate that the anti-HSV-1 activity of DAI at least in part involves inefficient expression of viral ICP0 gene.

DAI is not essential for IFN-β transcription through cytosolic B-DNA sensing in HepG2 cells.

We investigated whether DAI plays an essential role in the type I IFN pathway in HepG2 cells after treatment of immunostimulatory B-DNA. First, we tested induction of IFN-β mRNA synthesis in HepG2 cells after cytosolic B-DNA stimulation. The results of quantitative real-time RT-PCR showed that the IFN-β mRNA level was increased by poly(dA-dT) stimulation in a dose-dependent manner; the maximum increase in this experiment was observed with stimulation of cells with 3 μg of poly(dA-dT) for 12 h (Fig. 4A). We found that the B-DNA-mediated induction of IFN-β transcription in this condition was similar between control and DAI-knockdown cells (Fig. 4B). Since IFN-β transcription is activated by IRF3 and NF-κB, we further examined the effect of DAI-knockdown on activation of IRF3 and NF-κB. Consistent with the data for IFN-β mRNA levels, both phosphorylation of IRF3 at serine 396 and degradation of IκBα after poly(dA-dT) stimulation were comparable between control and DAI-knockdown HepG2 cells (Fig. 4C). Taken together, these results suggest that DAI is not essential for the IFN-β production through cytosolic B-DNA sensing in HepG2 cells.

To further investigate the effect of DAI expression on NF-κB activation in HepG2 cells, we examined the effect of DAI-RIP1 interaction on the plasmid DNA transfection-dependent DAI-mediated activation of NF-κB (12, 13). As observed in these previous reports, human DAI interacted with wild-type RIP1 but not with ΔRHIM mutant in cotransfected 293T and HepG2 cells (Fig. 5A and B). Consistently, RIP1was relocalized to cytoplasmic DAI foci in cotransfected cells in a RHIM-dependent manner (Fig. 5C). In reporter assays, a synergistic effect of DAI and RIP1 expression on NF-κB activation was observed in 293T cells (Fig. 5D), and this was also consistent with previous reports (12, 13). However, cotransfection of DAI and RIP1 did not enhance NF-κB activation in HepG2 cells (Fig. 5E). Furthermore, the B-DNA-mediated activation of NF-κB was normally observed in HepG2 cells, but this was not augmented by the overexpression of DAI and RIP1 (Fig. 5F). These results indicated that DAI interacted with RIP1, but their overexpression did not enhance the NF-κB activation by B-DNA treatment in HepG2 cells. Collectively, our analysis demonstrates that DAI is not essential for the IFN-β production through IRF3 and NF-κB activation upon stimulation by cytosolic B-DNAs in HepG2 cells.

Interaction of DAI with ICP0.

We examined the localization pattern of DAI in HSV-1-infected HepG2. In normal HepG2 cells, the endogenous DAI proteins were detected at very low level throughout the cells (Fig. 6Aa to c), although a few cytoplasmic DAI foci were detected in only a minority of cells (data not shown). After HSV-1 infection, the number of cells showing the DAI signals in the cytoplasm and often in the nucleus was increased at 3 h (Fig. 6Ad to i). ICP0 localizes to the nucleus at the early phase of infection and then is also distributed in the cytoplasm as diffuse and punctate forms at the late phase (30–33). We found that some DAI induced in HSV-1-infected cells colocalized with ICP0 in the nuclear and cytoplasmic foci (Fig. 6Ad to j, note the closed barbed arrowheads for nuclear colocalization and open arrowheads for cytoplasmic colocalization in panel j).

We previously demonstrated that transfected DAI was localized largely in the cytoplasm, but also in the nucleus in several cell types, including HepG2 (23). When HA-DAI-transfected HepG2 cells were infected with HSV-1 for 8 h, we found that HA-DAI effectively colocalized with ICP0 in the nucleus and the cytoplasm (Fig. 6B). Notably, the cytoplasmic ICP0 foci was less efficiently found in HA-DAI-expressing cells, suggesting that overexpression of DAI by transfection may delay the relocalization of ICP0 into the cytoplasm (Fig. 6B a to c). Collectively, the results of the IFA suggest a specific association of DAI with ICP0 during HSV-1 infection.

Therefore, we further investigated the interaction of DAI with ICP0 using CoIP assays. When 293T cells were cotransfected with myc-DAI and ICP0, immunoprecipitation of myc-DAI coprecipitated ICP0 and, in reciprocal assays, immunoprecipitation of ICP0 coprecipitated myc-DAI (Fig. 7A). DAI ΔE4/5 mutant did not interact with ICP0, as well as wild-type DAI, indicating that the central region encompassing Zβ and D3 domains is required for efficient ICP0 binding (Fig. 7B). We also examined the DAI-ICP0 interaction in HSV-1-infected cells. Since the DAI antibody used could not detect endogenous low-level DAI in immunoblot analysis, we could not detect the association of endogenous DAI with ICP0 in HSV-1-infected HepG2 cells. However, when HepG2 cells transiently transfected with HA-DAI were infected with HSV-1, ICP0 was coprecipitated with DAI (Fig. 7C). The association of DAI with ICP0 during virus infection was further assessed using wild-type or ΔE4/5 mutant DAI-expressing HepG2 cells. When these control (MIN) and DAI-expressing (MIN-DAI or MIN-DAI-ΔE4/5) cells were infected with HSV-1, immunoprecipitation of wild-type DAI, but not ΔE4/5 mutant, coprecipitated ICP0 (Fig. 7D). These results confirm that DAI indeed interacts with ICP0.

ICP0 is involved in DAI-mediated repression of HSV-1 growth.

We demonstrated that DAI expression more significantly inhibits ICP0 expression at lower MOI (Fig. 1). In addition to the effect of DAI on viral IE gene expression, the interaction of DAI with ICP0 may also affect viral growth. To test it, we compared the effects of DAI knockdown and overexpression on the growth of wild-type and ICP0-deleted mutant viruses. The dlX3.1 virus contained a 3.1-kb deletion in both copies of the ICP0 gene in the HSV-1 genome and, therefore, completely loses all ICP0 functions (21). We first infected normal and DAI-knockdown HepG2 cells at an MOI of 3 and then compared the growth of wild-type and dlX3.1 mutant viruses by measuring the level of progeny virion production. In normal and control (shC) HepG2 cells, the dlX3.1 virus growth was reduced by >10-fold compared to the wild-type virus growth (Fig. 8A). Importantly, we found that, unlike wild-type virus, the growth of dlX3.1 virus was not affected by DAI knockdown (Fig. 8A). The growth of wild-type and dlX3.1 mutant viruses was also compared between control (MIN) and DAI-expressing [MIN-DAI and MIN-DAI(ΔE4/5)] cells. The results showed that, unlike wild-type virus, the growth of dlX3.1 virus was not repressed by wild-type DAI expression (Fig. 8B).

Since the high-level expression of DAI mediates virus-induced programmed necrosis in certain cell types, in which RIP3 was also expressed at high level (14), we investigated whether the DAI-mediated inhibition of HSV-1 growth is related to increased cell death. However, the levels of cell viability after infection with HSV-1 were similar among control and DAI-expressing cells, indicating that the anti-HSV-1 activity of DAI in HepG2 cells may be independent of virus-induced necrosis (Fig. 8C).

DAI is not required for the DNA-mediated production of IFN-β in MEFs (11, 16). Therefore, we also tested whether the expression of DAI inhibits HSV-1 replication in MEFs in an ICP0-dependent manner. MEFs expressing wild-type or ΔE4/5 mutant DAI were produced by retroviral vectors and infected with wild-type or ICP0 mutant virus. Like in HepG2 cells, the expression of wild-type DAI, but not ΔE4/5 mutant, repressed HSV-1 replication in MEFs, and that this repressive effect of DAI was not observed in dlX3.1 virus infection (Fig. 8D). Collectively, the results with the ICP0-deleted mutant virus demonstrated that the repression of HSV-1 growth by DAI in HepG2 cells and MEFs requires the presence of ICP0 protein, suggesting that the DAI-ICP0 interaction may contribute to anti-HSV-1 activity of DAI.

We further investigated whether DAI expression affects the ICP0's Ub E3 ligase activity. It was recently shown that another DNA sensor IFI16 is degraded by ICP0 during HSV-1 infection in human fibroblasts (34). We also observed that IFI16 was degraded in HepG2 cells after infection with wild-type HSV-1 but not with dlX3.1 mutant virus (Fig. 9A). When compared between control and DAI-knockdown cells, the loss of IFI16 and PML (another substrate of ICP0) during virus infection was increased by DAI knockdown (Fig. 9B). These results indicate that DAI expression downregulates the ICP0's Ub E3 ligase activity, suggesting that the DAI-ICP0 interaction may affect the ICP0 function.