The Zinc Finger Antiviral Protein Acts Synergistically with an Interferon-Induced Factor for Maximal Activity against Alphaviruses

Cell lines.BHK-21 and Rat2 cell lines transduced with vectors pBabe-HAZ or pBabe-NZAP-Zeo (9) and designated here Rat2/HA-Zeo and Rat2/NZAP-Zeo cells were maintained as previously described (2). The retroviral vector pBabe-HAZ (9) is an MLV-based vector expressing a hemagglutinin (HA)-tagged zeocin resistance gene product under the control of the Simian virus 40 early promoter. Vector pBabe-NZAP-Zeo expresses the amino terminal 254 amino acids of rat ZAP fused in frame with the zeocin resistance gene (9). Immortalized mouse embryo fibroblast (MEF) cells derived from C57BL/6 wild-type (wt) and Stat1^−/− mice (21) were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS). wt and Stat1^−/− MEF cells stably transduced (see below) with pBabe-HAZ or pBabe-NZAP-Zeo (wt MEF/HA-Zeo, wt MEF/NZAP-Zeo, Stat1^−/− MEF/HA-Zeo, and Stat1^−/− MEF/NZAP-Zeo) were maintained in DMEM containing 10% FBS and 200 μg/ml zeocin. wt MEF/HA-Zeo cells transduced with short hairpin RNA (shRNA)-expressing defective lentiviruses (see below) were maintained in DMEM containing 10% FBS, 200 μg/ml zeocin, and 5 μg/ml blasticidin. BHK-21 cells stably transduced with pBabe-HAZ or pBabe-NZAP-Zeo (BHK/HA-Zeo and BHK/NZAP-Zeo) were maintained in minimal essential medium supplemented with 7.5% FBS and 200 μg/ml zeocin.

Production of VSV-G-pseudotyped retroviral particles and cell transductions.MLV particles pseudotyped with VSV protein G (VSV-G) were generated by cotransfection of 293T cells with pBabe-HAZ or pBabe-NZAP-Zeo DNA and DNAs encoding MLV Gag-Pol and VSV-G envelope proteins as described previously (9). Bulk transduced cell lines were obtained by infection of the cells with the pseudotyped viral particles and selection in the presence of 200 μg/ml zeocin. Human immunodeficiency virus particles pseudotyped with VSV-G were obtained by cotransfection of 293T cells with shRNA-expressing derivatives of pLenti-3′-U6-EC-EP7 (see “Silencing of endogenous murine ZAP,” below) and DNAs encoding human immunodeficiency virus Gag-Pol and VSV-G. Bulk transduced cell lines were obtained by infection with the pseudotyped viral particles and selection in the presence of 5 μg/ml blasticidin.

Silencing of endogenous murine ZAP.The pLenti-3′-U6-EC-EP7 plasmid (provided by Daniel Boden, Aaron Diamond AIDS Research Center, New York, NY) was engineered to express an irrelevant shRNA or shRNAs targeting murine ZAP. The shRNA expression vector pLenti-3′-U6-EC-EP7 is a derivative of vector pLenti6/V5-D-TOPO (Invitrogen) and was generated as follows (D. Boden, personal communication): briefly, the Topo binding sites and the CMV promoter of pLenti6/V5-D-TOPO were removed, and a BamHI/HpaI polylinker was introduced into the lentiviral 3′ long terminal repeat. These two restriction sites were used to insert an shRNA expression cassette containing the U6 polymerase III promoter preceded by the enhancer of the CMV immediate-early promoter. Two restriction sites (EcoRV and PstI) downstream of the U6 promoter allow for the introduction of short hairpin DNA sequences. Two repeated stretches of six thymidine residues following the PstI site serve as a polymerase III transcription terminator site. The pLenti-3′-U6-EC-EP7 defective lentiviral vector also expresses the blasticidin resistance gene. The murine ZAP mRNA sequence (NM_028864) was analyzed using a Dharmacon Custom small interfering RNA design tool (www.dharmacon.com ) to locate potential sites for targeting. Sense and antisense oligonucleotides (W. M. Kech Oligonucleotide Synthesis Facility, Yale University) containing inverted repeats of the 23-mer sites to be targeted were annealed and cloned into EcoRV- and PstI-digested pLenti-3′-U6-EC-EP7. The PmeI-NotI fragment of pcDNA4/TO/myc-mouseZAP995, which was provided by Guangxia Gao (Chinese Academy of Sciences, Beijing, China) and contains the murine ZAP cDNA, was cloned into the 3′ untranslated region of the Renilla luciferase gene in the psiCHECK-2 vector (Promega). Efficiency of silencing was assessed as recommended by the manufacturer by cotransfection of the murine ZAP-containing psiCHECK derivative into 293T cells with the various shRNA-expressing plasmids. The two plasmids showing the lowest normalized Renilla luciferase activity, as well as plasmid pLenti-3′-irrelevant-shRNA (obtained from Thomas von Hahn, Rockefeller University, New York, NY), were utilized to transduce wt MEF/HA-Zeo cells to generate constitutively silenced cell lines. Sequences of the sense and antisense oligonucleotides utilized for generation of the two silenced cell lines (wt MEF/HA-Zeo/shRNA-ZAP-1 and wt MEF/HA-Zeo/shRNA-ZAP-5, respectively) were as follows, with the target sequence shown in uppercase letters: ZAP-1 sense, 5′-atcAAGAGAAGAGGTCCAGAGTAAGTaggatcACTTACTCTGGACCTCTTCTCTTttttttctgca-3′; ZAP-1 antisense, 5′-gaaaaaaAAGAGAAGAGGTCCAGAGTAAGTgatcctACTTACTCTGGACCTCTTCTCTTgat-3′; ZAP-5 sense, 5′-atcCACGGCATTCCTGCATAGTAAATaggatcATTTACTATGCAGGAATGCCGTGttttttctgca-3′; and ZAP-5 antisense, 5′-gaaaaaaCACGGCATTCCTGCATAGTAAATgatcctATTTACTATGCAGGAATGCCGTGgat-3′. ZAP-1 and ZAP-5 target sequences start at nucleotides 2731 and 1731 of the murine ZAP mRNA, respectively. Sequences of the sense and antisense oligonucleotides utilized to generate pLenti-3′-irrelevant-shRNA were as follows (target sequences are shown in uppercase letters): sense, 5′-atcAATAGCGACTAAACACATCAATTaggatcAATTGATGTGTTTAGTCGCTATTttttttctgca-3′; and antisense, 5′-gaaaaaaAATAGCGACTAAACACATCAATTgatcctAATTGATGTGTTTAGTCGCTATTgat-3′. pLenti-3′-irrelevant-shRNA, which is not predicted to target any human or murine sequences, was used to generate the control cell line (wt MEF/HA-Zeo/shRNA-irrel).

Real-time quantitative RT-PCR.Total RNA was harvested from cells using an RNeasy Mini Kit (Qiagen) according to the manufacturer's directions. One microgram of total RNA was transcribed into cDNA using a SuperScript III First-Strand Synthesis System for reverse transcription-PCR (RT-PCR) (Invitrogen) according to the manufacturer's instructions, using random hexamers as primers. cDNAs diluted 1:20 were then amplified with a QuantiTect SYBR Green PCR kit (Qiagen) and detected with a LightCycler 480 (Roche). Enzyme activation occurred at 95° for 15 min, followed by 40 cycles of 94° for 15 s, 55° for 20 s, and 72° for 20 s. Primers for amplification of murine ZAP were obtained from Qiagen (QuantiTect primer assay, QT00107100). Samples were normalized based on the amount of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA present, as determined by real-time PCR using Qiagen QuantiTect Primers (QT00309099). Relative levels were determined using samples from duplicate wells, each assayed in triplicate PCRs, using the comparative threshold cycle method (1a). The efficiency of ZAP and GAPDH amplification was tested and found to be approximately equal in the cDNA dilution range utilized in the assay.

IFN treatment.Universal type I IFN (PBL Biomedical Laboratories), a recombinant human IFN-αA and IFN-αD chimeric protein with activity for type I IFN receptors from multiple mammalian species, including hamster and rat, was diluted in medium appropriate for the cell type and utilized for all IFN-α treatments. Duration of treatment is indicated in the figure legends.

Viruses, infections, and titrations.Stocks of wt SINV (Toto1101), SINV expressing firefly luciferase as a fusion with nsP3 (Toto1101/Luc), a temperature-sensitive mutant SINV expressing luciferase (Toto1101/Luc:ts110), and VSV (San Juan strain) were prepared, and titers were determined on BHK-21 cells as described previously (2). Multiplicity of infection (MOI) was determined based on BHK-21-derived titers. Viral infections, growth curves, and virus titrations were performed as previously described (2). The temperature sensitivity property of the Toto1101/Luc:ts110 virus stock was verified by plaque assay titration on BHK/HA-Zeo cells; no plaques were detected at any dilution (including undiluted stock) at 40°C, while the stock titer, determined at 28°C was >10⁸/ml.

Luciferase assays.Cell monolayers were washed with phosphate-buffered saline, and lysates were prepared and measured using 1× passive lysis buffer and a luciferase assay system (Promega) according to the manufacturer's recommendations. Luciferase activity was measured using a Berthold LB960 luminometer.

Antibody production.A bacterial expression plasmid (pGEX-NZAP) producing glutathione-S-transferase (GST) with the amino-terminal 254 residues of rat ZAP (NZAP) fused to the carboxy terminus was generated in plasmid pGEX-6P-2 (GE Healthcare) using standard techniques. Details of the construction and the sequence of the plasmid are available upon request. Bacterially expressed GST-NZAP was purified on a GSTrap FF (GE Healthcare) column according to the manufacturer's recommendations. For generation of NZAP without the GST tag, GST-NZAP was dialyzed overnight in the presence of PreSission Protease (GE Healthcare), and the GST and protease were removed by passage through the GSTrap FF column. GST-NZAP and NZAP were utilized as the antigen in an alternating schedule to immunize BALB/c mice, and monoclonal antibodies recognizing NZAP were generated by the Monoclonal Antibody Core Facility of Memorial Sloan Kettering Cancer Center using their standard protocols. Antibodies that react with both rat NZAP and full-length rat ZAP were purified from culture supernatants of clone 13B10.1 using a protein A Sepharose CL-4B (GE Healthcare) column.

Antibodies used.Anti-NZAP monoclonal antibody from clone 13B10.1 was utilized for immunofluorescence analysis at 0.5 μg/ml and for Western analysis at 0.26 μg/ml. Anti-β-actin mouse monoclonal antibody AC-15 (Sigma A5441) was utilized at a 1:5,000 dilution. AlexaFluor 594-conjugated goat anti-mouse immunoglobulin G (Molecular Probes) was utilized at a 1:1,000 dilution. Horseradish peroxidase-conjugated rabbit anti-mouse immunoglobulin G (Pierce) was utilized at a 1:10,000 dilution.

Immunofluorescence and image acquisition.Cell were fixed and analyzed by indirect immunofluorescence as previously described (4) using a Nikon Eclipse TE300 inverted microscope with a transmission electron-fluorescence microscopy epifluorescence attachment and mercury lamp power supply. Images were acquired with a 60× oil immersion objective using a Spot RT camera (model 2.2.1; Diagnostic Instruments) using the Spot software. Phase-contrast images were obtained using automatic exposure with either the Spot RT camera or with a Nikon Eclipse TS100 microscope equipped with a Nikon CoolPix 995 camera. Adobe Photoshop or the Spot software was utilized for brightness and contrast adjustment using identical changes to all images within an experiment.

Western blotting.Cells were lysed in 2× Laemmli sample buffer, and the proteins in the lysate were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to Hybond ECL (enhanced chemiluminescence) nitrocellulose membranes (GE Healthcare), and incubated with primary and secondary antibodies as previously described (19). Enhanced chemiluminescence detection was performed with SuperSignal West Pico chemiluminescent substrate (Pierce) according to the manufacturer's recommendation.

Rat ZAP fails to inhibit SINV production in cells defective in type I IFN production.BHK-21 cells (hamster kidney) are frequently utilized for the propagation of a variety of viruses, including SINV. The robust and often persistent growth of various viruses is likely due to a deficiency in IFN production by these cells (1, 3, 24, 27, 30), although the deficiency has not been fully characterized, nor has the molecular mechanism(s) of the defect been elucidated. We utilized these cells to test whether rat ZAP is able to mediate its inhibitory effects in cells with a deficiency in IFN production (and thus likely also with little to no IFN signaling). Using VSV-G-pseudotyped MLV vectors pBabe-HAZ and pBabe-NZAP-Zeo (9), we generated stably transduced BHK-21 cell populations expressing the zeocin resistance gene (BHK/HA-Zeo) or NZAP fused to the zeocin resistance gene (BHK/NZAP-Zeo) and examined the growth of SINV. Virus production at both low (Fig. 1A) and high (Fig. 1B) MOIs was found to be virtually indistinguishable in the two cell lines. SINV-mediated cell death was also similar, although a possible slight delay in cytopathic effect (CPE) was noted in the cells expressing NZAP (Fig. 1C).

• Open in new tab
• Download powerpoint

FIG. 1.

ZAP inhibition of SINV replication and virus-induced cell death in BHK cells. (A and B) BHK/HA-Zeo and BHK/NZAP-Zeo cells seeded at 7 × 10⁵ cells per 35-mm dish the day prior were infected with SINV Toto1101 at low (A, 0.01) and high (B, 5) MOIs, and the amount of virus present in the medium was determined at various times after infection by titration on permissive BHK-21 cells. A separate, single well was utilized for each time point; symbols represent the mean of duplicate titrations. (C) Photographs of mock- or SINV-infected BHK/HA-Zeo and BHK/NZAP-Zeo monolayers (MOI of 5) at the indicated times after infection are shown.

Possible reasons for the failure of ZAP to inhibit SINV growth and virus-mediated CPE in BHK cells include a species incompatibility, the lack of a necessary factor(s), the presence of a factor(s) that interferes with ZAP's action, inadequate levels of ZAP expression from the transducing retroviral vector, or an altered subcellular distribution of ZAP in BHK cells. We felt a species incompatibility was unlikely, given the high similarity in ZAP's amino terminal sequences among different mammalian species (rat NZAP shares 93% and 80% amino acid identity and 94% and 89% similarity with murine and human NZAP, respectively). Additionally, since rat ZAP is able to function in human cells, we felt it was likely to function in the more closely related hamster cells. We tested expression of the NZAP-Zeo protein in the BHK/NZAP-Zeo cells by indirect immunofluorescence (Fig. 2) and Western blotting (see Fig. 5) and found that levels were similar to the level seen in the highly SINV-nonpermissive Rat2/NZAP-Zeo cells (2). Moreover, the diffusely cytoplasmic subcellular distribution (Fig. 2) was similar to that seen in Rat2/NZAP-Zeo cells (not shown).

• Open in new tab
• Download powerpoint

FIG. 2.

Immunofluorescence analysis of NZAP-Zeo expression in BHK/NZAP-Zeo cells. BHK/HA-Zeo and BHK/NZAP-Zeo cells were analyzed by indirect immunofluorescence for the presence of NZAP protein (red) using anti-NZAP monoclonal antibody 13B10.1 and AlexaFluor 594 goat anti-mouse secondary antibody as described in Materials and Methods. The two images shown were acquired with identical exposure conditions under a 60× oil immersion lens. Nuclei are shown in blue.

Pretreatment with type I IFN restores ZAP's inhibitory activity in BHK cells.If, due to defective IFN production (and thus also with reduced constitutive low-level IFN signaling), BHK cells lack an ISG factor necessary for ZAP's function, then treatment of cells with IFN might restore ZAP's inhibitory activity in these cells. We pretreated BHK/HA-Zeo and BHK/NZAP-Zeo cells with various doses of IFN-α and assessed the ability of the cells to support SINV growth (Fig. 3A). In BHK cells expressing vector alone, pretreatment with IFN-α had little effect on SINV growth, with only a mild (less than 1 log) reduction in virion production seen at the highest IFN dose tested (100 U/ml) for both low and high MOIs. In contrast, IFN-α pretreatment of BHK/NZAP-Zeo cells resulted in a dose-dependent reduction of SINV growth at both low and high MOIs. Virion production 24 h after infection at a low MOI was reduced by >2 or >4 logs after pretreatment with 10 and 100 U/ml IFN-α, respectively. Similarly, virion production 12 h after infection at a high MOI was reduced by >1 or >2 logs after pretreatment with 10 and 100 U/ml IFN-α, respectively. IFN-α-treated BHK/NZAP-Zeo cells were also protected from SINV-mediated cell death to a greater degree than BHK/HA-Zeo cells (Fig. 3B).

• Open in new tab
• Download powerpoint

FIG. 3.

Restoration of ZAP's inhibitory function in BHK cells by IFN-α pretreatment. (A) BHK/HA-Zeo and BHK/NZAP-Zeo cells seeded at 3 × 10⁵ (low MOI) or 2.5 × 10⁵ (high MOI) cells per 35-mm dish were allowed to adhere and were then incubated overnight (12 to 14 h) with medium containing the indicated concentration of IFN-α. After infection the next day with SINV Toto1101 at the indicated MOI, the cells were washed, and the medium containing the same IFN-α concentration was added back. The amount of virus present in the medium was determined at the indicated times after infection by titration in duplicate on permissive BHK-21 cells. Two separate wells were utilized for each time point. Symbols represent the mean log titers; error bars, often obscured by the symbol, indicate the SEM. (B) Photographs of the BHK/HA-Zeo and BHK/NZAP-Zeo monolayers from panel A, treated with the indicated concentrations of IFN-α and infected with SINV (MOI of 0.01) for 49 h are shown.

One possible explanation for the SINV inhibition seen in the NZAP-Zeo-expressing cells treated with IFN is that these cells are inherently more responsive to IFN. To test this, we examined the effects of IFN-α pretreatment of BHK/HA-Zeo and BHK/NZAP-Zeo cells on the growth of VSV, a highly IFN-sensitive virus, whose growth is unaffected by ZAP expression (2). The susceptibility of IFN-α-treated BHK/HA-Zeo and BHK/NZAP-Zeo cells to VSV-mediated CPE was essentially identical (Fig. 4). Over multiple experiments, the BHK/NZAP-Zeo responsiveness to IFN′s effects was always equal to or less than that of the BHK/HA-Zeo cells (not shown). Furthermore, IFN-α pretreatment led to similar reductions in VSV virion production in the two lines, with a slightly reduced effect of IFN-α on VSV growth in the BHK/NZAP-Zeo cells (Table 1). Therefore, the increased ability of IFN-α treatment to inhibit SINV replication in the BHK/NZAP-Zeo cells is not due to an inherent increase in IFN sensitivity.

• Open in new tab
• Download powerpoint

FIG. 4.

IFN-α sensitivity of BHK/HA-Zeo and BHK/NZAP-Zeo cells. BHK/HA-Zeo and BHK/NZAP-Zeo cells were seeded at 1 × 10⁵ cells per well in a 96-well plate in the presence of twofold serial dilutions (500 U/ml to 1 U/ml) of IFN-α as indicated. The next day (28 h later) cells were infected as indicated with VSV (MOI of 0.025), and after 24 h the monolayers were stained with crystal violet as described in Materials and Methods.

Another possible explanation for enhanced SINV inhibition in the BHK/NZAP-Zeo cells is that IFN-α treatment could increase NZAP-Zeo expression from the transducing vector and thus potentially increase the inhibitory activity. To test this, we evaluated NZAP-Zeo levels after IFN-α treatment by Western blotting and found that levels were not altered and were similar to the level seen in Rat2/NZAP-Zeo cells (Fig. 5). Together with the VSV results above, this suggests that the ability of the IFN-α treatment to result in enhanced SINV inhibition in the BHK/NZAP-Zeo cells compared to control cells is due to the expression of NZAP-Zeo, which likely functions in concert with an IFN-induced factor(s).

• Open in new tab
• Download powerpoint

FIG. 5.

Western blot analysis of NZAP-Zeo levels in IFN-α-treated cells. The indicated cell lines were seeded at 7 × 10⁵ cells per 35-mm dish and allowed to adhere. Cells were then treated overnight as indicated with 500 U/ml IFN-α before harvesting. Equal volumes of lysate were loaded onto 10% polyacrylamide gels and subjected to Western blot analysis using anti-NZAP monoclonal antibody 13B10.1 (top panel). A duplicate blot was probed with anti-β-actin antibodies as a loading control (bottom panel). Similar results were obtained in one other independent experiment.

Enhanced activity of ZAP requires signaling through the type I IFN receptor.If ZAP inhibitory activity requires the presence of an IFN-induced factor, then inhibition of IFN signaling should abrogate the ability of IFN-α to restore ZAP-mediated inhibition of SINV. To examine IFN-α effects on ZAP function in the absence of IFN signaling, we utilized immortalized MEFs generated from mice with a targeted disruption of the Stat1 gene (8) (Stat1^−/− mice). Mice deficient in Stat1 have previously been shown to lack responsiveness to IFN-α/β and to be highly susceptible to viral pathogens (8, 22). Using immortalized MEFs from wt mice as well as Stat1^−/− mice, we generated cell lines expressing HA-Zeo (wt MEF/HA-Zeo and Stat1^−/− MEF/HA-Zeo, respectively) or expressing NZAP-Zeo (wt MEF/NZAP-Zeo and Stat1^−/− MEF/NZAP-Zeo, respectively) to examine ZAP's ability to inhibit SINV in the presence or absence of IFN signaling.

As expected, due to the lack of IFN-α/β responsiveness, IFN pretreatment of Stat1^−/− MEFs expressing either HA-Zeo or NZAP-Zeo had no effect on SINV growth (Fig. 6, top). Surprisingly, however, in contrast to the results obtained in BHK cells defective in IFN production, NZAP-Zeo expression resulted in a SINV growth inhibition despite the inability of these cells to respond to IFN. SINV titers were consistently ∼2 logs lower in the NZAP-Zeo-expressing cells for the first 24 h after infection. This suggests that in some cell types, ZAP is able to mediate SINV growth inhibition in the absence of additional IFN-induced factors.

• Open in new tab
• Download powerpoint

FIG. 6.

ZAP-mediated inhibition in Stat1^−/− and wt MEFs. Stat1^−/− or wt MEFs expressing HA-Zeo or NZAP-Zeo were seeded at 3 × 10⁵ (Stat1^−/−) or 1.2 × 10⁵ (wt) cells per 35-mm dish and were allowed to adhere and then were incubated overnight (15 h) with normal medium (filled symbols) or medium containing 100 U/ml IFN-α (open symbols). The cells were then infected with SINV Toto1101 (MOI of 5), and medium (without IFN-α) was added back. Two separate wells were utilized for each condition, and the medium in each was harvested and replaced at each time point. The amount of virus present in each sample was determined by titration in duplicate on permissive BHK-21 cells, and the cumulative total for each time point is presented. Symbols represent the mean log titers of the duplicate dishes; error bars, often obscured by the symbol, indicate the SEM.

To test whether an IFN-induced factor(s) could further enhance ZAP's inhibitory activity in these cells, we examined the effect of IFN-α on SINV growth in wild-type (IFN-α/β responsive) MEFs expressing HA-Zeo or NZAP-Zeo (Fig. 6, bottom). The growth of SINV in the wt MEF/HA-Zeo cells, which peaked at ∼10⁸ PFU/ml, was reduced compared to the >10⁹ PFU/ml seen in the Stat1^−/− MEF/HA-Zeo cells, presumably due to endogenous IFN production and signaling in the wt cells. As expected, pretreatment with IFN-α resulted in a reduction of SINV growth, with a ∼2 log reduction seen from 6 to 48 h. In the absence of exogenous IFN treatment, the expression of NZAP-Zeo resulted in a ∼2 log reduction of SINV growth from 6 to 48 h. IFN-α pretreatment of wt MEF/NZAP-Zeo cells resulted in enhanced SINV growth inhibition, with a ∼4 log reduction in titers after 12 h compared to that seen in the untreated wt MEF/HA-Zeo cells. Thus, ZAP and an IFN-induced factor(s) are able to mediate synergistic antiviral effects against SINV.

The synergistic antiviral effect of ZAP and IFN-α requires new gene transcription.To further support the idea that the effects of IFN-α on ZAP-mediated SINV inhibition are likely due to production of an ISG factor(s), we examined the effect of transcription inhibition (Fig. 7) on SINV inhibition in wt (IFN responsive) MEFs. Cells expressing HA-Zeo or NZAP-Zeo treated with medium containing vehicle alone (ethanol) showed similar results to the result seen previously (Fig. 6), with a 2-log reduction in viral titers mediated by NZAP expression. Treatment with the cellular transcription inhibitor actinomycin D (ActD) alone had minimal effect on ZAP's inhibitory effect. As expected, treatment of wt MEF/NZAP-Zeo cells with IFN-α resulted in synergistic inhibition of SINV growth (∼3 logs at 24 h compared to that seen in the wt MEF/HA-Zeo cells treated with vehicle alone). The synergistic effect of IFN-α upon ZAP-mediated SINV inhibition was completely abrogated when IFN-α treatment was performed in the presence of ActD. The reduced inhibition seen in IFN-α-treated wt MEF/NZAP-Zeo cells in this experiment (3 logs) compared with the data presented in Fig. 6 (4 logs) is likely due to the short duration of IFN treatment (2 versus 15 h), which was necessary due to toxic effects of ActD (not shown). In the absence of ActD, the effects of IFN-α were time dependent, with a maximal effect requiring 4 to 6 h of treatment in these cells (not shown). Thus, the synergistic inhibitory effect of IFN-α with ZAP on SINV takes time and is dependent on new cellular gene transcription, similar to the production of ISG factors after IFN-α stimulation.

• Open in new tab
• Download powerpoint

FIG. 7.

Effect of transcription inhibition on the ZAP-IFN synergy. wt MEFs expressing HA-Zeo or NZAP-Zeo were seeded at 1.5 × 10⁵ cells per 35-mm dish the day prior to treatment for 2 h with medium containing vehicle alone or with medium containing 500 U/ml IFN-α in the presence or absence of 1 μg/ml ActD. The cells were then washed and infected with SINV Toto1101 (MOI of 5), and medium without additives was added back. Two separate wells were utilized for each condition, and the medium in each was harvested and replaced at each time point. The amount of virus present in each sample was determined by titration in duplicate on permissive BHK-21 cells, and the cumulative total for each time point is presented. Symbols represent the mean log titers of the duplicate dishes; error bars representing the SEM are obscured by the symbols.

IFN-independent ZAP inhibition of SINV replication has cell-type-specific outcomes on virus production.That ZAP mediates a 2-log SINV reduction in Stat1^−/− MEFs incapable of responding to IFN (Fig. 6) suggests that ZAP has intrinsic antiviral activity which functions independently of the activity of other ISGs. In wt MEFs, the synergistic activities of ZAP and an ISG factor have profound effects on virus production. However, in BHK cells ZAP fails to mediate inhibition of SINV growth in the absence of IFN-α treatment. We wondered, therefore, whether the ISG-independent activity was unable to function in the BHK cellular environment or, alternatively, whether the activity was functional but insufficient to result in significant inhibition of virus production.

Previously, using Rat2 fibroblast cells, we found that overexpression of rat NZAP-Zeo dramatically inhibits SINV virion production (∼6 logs at 24 h after infection at a low MOI) and found that ZAP mediates its inhibition after entry and at or before translation of the incoming genomic RNA (2). The inhibitory effect did not require pretreatment with IFN. Here, we utilized SINV Toto1101/Luc, which expresses luciferase as an in-frame fusion with the nsP3 protein, and monitored luciferase activity after 2 h of infection in order to investigate ZAP's effect on early translation and replication events in MEF and BHK cells. In wt MEFs, either expression of NZAP-Zeo or IFN-α pretreatment (100 U/ml) resulted in a >99% reduction in early replication events, as measured by luciferase activity after 2 h of infection (data not shown). This is consistent with the ability of ZAP or IFN-α to significantly reduce virion production, as seen in Fig. 6. Since the luciferase values obtained in the MEFs with NZAP-Zeo expression or IFN-α pretreatment were at background levels, we were unable to assess for synergistic inhibition of the early viral translation and replication events.

Surprisingly, despite a failure to inhibit SINV production in the BHK cellular environment in the absence of IFN (Fig. 1 and 3), expression of NZAP-Zeo was able to significantly decrease early replication events (Fig. 8), even without IFN-α pretreatment. In nine experiments measuring luciferase activity after 2 to 3 h of infection, expression of NZAP-Zeo resulted in a 92% (standard error of the meant [SEM], 1%) reduction in early replication relative to BHK cells expressing HA-Zeo (data not shown). Pretreatment with IFN-α resulted in a dose-dependent decrease in early viral replication in BHK/NZAP-Zeo cells, with higher doses resulting in complete inhibition, similar to the inhibition seen in the highly nonpermissive Rat2/NZAP-Zeo cells, where inhibition is not dependent on exogenous IFN treatment (Fig. 8). While IFN treatment did not significantly alter the levels of early viral replication in Rat2 cells expressing HA-Zeo, a mild reduction was seen in the BHK/HA-Zeo cells. These findings suggest that, in the absence of additional ISGs, ZAP mediates an antiviral effect on early SINV replication events but that the magnitude of the inhibition in some cell types (BHK) is insufficient to alter virus production, while in other cell types (MEFs and Rat2) virion production is decreased. Amplification of the early inhibition, mediated by the synergistic effects of an ISG factor(s), increases the extent of inhibition such that virion production is reduced.

• Open in new tab
• Download powerpoint

FIG. 8.

Effect of IFN-α pretreatment on ZAP-mediated inhibition of early replication events. The indicated cells were seeded at 5 × 10⁴ cells per well in 96-well plates in the presence of twofold serial dilutions (500 U/ml to 0.06 U/ml) of IFN-α as indicated. The next day (24 h later) cells were infected with SINV Toto1101/Luc (MOI of 5), after which medium was added back. The cells were lysed after 2 h, and firefly luciferase activity was determined. Symbols represent the mean value of triplicate wells; error bars represent the standard deviations. The dashed line represents the mean luciferase value obtained from uninfected cells. RLU, relative light units. Similar results were obtained in four additional independent experiments.

Like ZAP, the IFN-induced factor(s) targets an early step in the SINV life cycle.Since SINV Toto1101/Luc is replication competent, luciferase activity would be expected to increase upon translation of any newly replicated viral RNA. Therefore, reductions in luciferase activity with IFN pretreatment (Fig. 8) could be due to inhibition at any one of several steps of the virus life cycle. We previously determined that ZAP mediates its inhibitory step at or prior to translation of the incoming viral RNA (2). To test if the IFN-induced factor(s) was also affecting a similar step, we infected IFN-α-treated cells with a temperature-sensitive derivative of Toto1101/Luc, Toto1101/Luc:ts110 (2), and measured the luciferase activity under nonpermissive conditions where RNA replication is blocked (Fig. 9). An IFN-α dose-dependent reduction of SINV genome translation was apparent in BHK cells expressing NZAP-Zeo, suggesting that the IFN-induced factor(s) is also inhibiting early events in the SINV life cycle. Since RNA replication and subsequent virus life cycle steps are blocked at the nonpermissive temperature, these results demonstrate that the IFN-induced factor(s) and ZAP both act at or before translation of the incoming viral RNA genome. Additional IFN-induced blocks at later life cycle steps might contribute to the overall inhibition seen in ZAP-expressing cells infected with replication-competent SINV.

• Open in new tab
• Download powerpoint

FIG. 9.

The IFN-α-induced factor(s) blocks translation of incoming SINV RNA. The indicated cells were seeded at 5 × 10⁴ cells per well in 96-well plates in the presence of twofold serial dilutions (500 U/ml to 0.06 U/ml) of IFN-α as indicated. The next day (∼17 h later) cells were infected at 40°C with SINV Toto1101/Luc:ts110 (MOI of 17), after which medium was added back. The cells were incubated at 40°C for 5 h, and firefly luciferase activity was determined. Symbols represent the mean value of triplicate wells; error bars represent the standard deviations. The dashed line represents the mean luciferase value obtained from uninfected cells. RLU, relative light units. Similar results were obtained in one other independent experiment.

Silencing of endogenous ZAP reduces the antiviral efficacy of IFN-α.While our studies utilizing overexpressed ZAP demonstrate anti-SINV activity, we wondered if ZAP induced by IFN-α treatment might contribute to the antiviral state of the cell. To examine ZAP's role in the effects of IFN-α, we generated derivatives of wt MEF/HA-Zeo cells that stably express ZAP-specific shRNAs (wt MEF/HA-Zeo/shRNA-ZAP-1 or -5) or an irrelevant (wt MEF/HA-Zeo/shRNA-irrel) shRNA. Treatment of wt MEF/HA-Zeo/shRNA-irrel cells with IFN-α (100 U/ml for 6 h) resulted in a 4.7-fold increase in ZAP RNA levels (range, 4.6 to 4.8 in duplicate samples), consistent with previous work showing that ZAP is an ISG (20, 25, 31). Treatment of cells expressing ZAP-specific shRNAs with various doses of IFN-α revealed a decreased IFN-mediated antiviral efficacy (Fig. 10). While treatment with a low dose (6.25 U/ml) of IFN-α reduced SINV Toto1101/Luc replication in cells expressing the irrelevant shRNA to less than 5% of that seen in untreated cells, significant replication (∼30% of that seen in untreated cells) was apparent in the two cell lines expressing ZAP-specific shRNAs. The level of ZAP RNA induced by treatment of cells expressing the irrelevant shRNA with 6.25 U/ml IFN-α was assessed by quantitative RT-PCR and showed a ∼1.7-fold induction (range, 1.67 to 1.76 in duplicate samples) compared to untreated cells. RT-PCR analysis confirmed that silencing of ZAP in the ZAP-1 and ZAP-5 shRNA-expressing cells was efficient. Compared to untreated wt MEF/HA-Zeo/shRNA-irrel cells, the levels of ZAP RNA in IFN-treated wt MEF/HA-Zeo/shRNA-ZAP-1 and MEF/HA-Zeo/shRNA-ZAP-5 cells were reduced, with relative levels of 0.8 (range, 0.71 to 0.84) and 0.2 (range, 0.22 to 0.23), respectively. Even at higher doses of IFN-α, the effect of ZAP silencing was apparent, with higher levels of viral replication occurring in the cells silenced for ZAP expression than in cells expressing the irrelevant shRNA after treatment with 12.5 or 25 U/ml IFN-α (Fig. 10). Treatment with 50 U/ml resulted in minimal virus replication in all three lines, likely due to the induction of multiple ISGs with antiviral activity. That silencing of endogenous ZAP reduces the anti-SINV efficacy of IFN treatment was also recently reported by others (31).

• Open in new tab
• Download powerpoint

FIG. 10.

Effect of ZAP silencing on antiviral activity of IFN. The indicated cells were seeded at 2.4 × 10⁵ cells per well in 12-well plates the day prior to addition of twofold serial dilutions (50 U/ml to 6.25 U/ml) of IFN-α as indicated. After 6 h, cells were infected with SINV Toto1101/Luc (MOI of 10), after which medium (without IFN-α) was added back. The cells were lysed after 3 h, and firefly luciferase activity was determined. For each cell type, mean luciferase values obtained from mock-infected cells were subtracted from values of infected samples. The bars indicate the relative luciferase activity from IFN-treated cells as a percentage of the mean luciferase values obtained from infected cells without IFN pretreatment (set at 100%). Error bars represent the range of values from duplicate wells.