INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Interferons (IFNs) modulate a number of biological functions, namely, virus replication, immune response, and cell growth and differentiation. IFNs exert their actions through species-specific cell surface receptors and induce IFN-stimulated gene (ISG) products, including antiviral products such as double-stranded-RNA-dependent protein kinase (PKR) and 2',5'-oligoadenylate synthetase (2',5'-AS) (28). Recently, IFN-mediated cell signaling has been intensively investigated. Binding of IFN to the cell surface receptor initiates activation of the receptor-associated tyrosine kinases Jak1 and Trk2 (IFN-/) or Jak1 and Jak2 (IFN-). IFN- acts through Stat1/Stat1 homodimers binding to the gamma-activating sequence, and IFN-/ acts through Stat1/Stat2/p48 binding to the IFN-stimulated response element (30).

Several viruses have been shown to inhibit the induction of cellular antiviral resistance by IFN. In our previous study, various cell lines persistently infected with Sendai virus were found to be less susceptible to the antiviral action of IFN than the same cell lines uninfected with Sendai virus (11). On the other hand, when Vero and L929 cells persistently infected with a temperature-sensitive strain of Sendai virus were incubated at 38°C (nonpermissive temperature), they became fully susceptible to IFN, indicating that the lower IFN susceptibility of virus carrier cells is related to the maturation and replication of virus in them (9, 10, 11). It was also found that the low susceptibility of virus carrier cells to IFN was not due to blocked adsorption of IFN or to inability of the cells to respond to IFN and that some step(s) before the synthesis of the mRNAs for the antiviral proteins was blocked.

The C protein in Sendai virus and the V protein in simian virus 5 (SV5) have recently been reported to be responsible for the virus-mediated inhibition of IFN signaling (6, 7). In this study, we analyzed the susceptibility of human parainfluenza type 2 virus (hPIV-2)-infected HeLa cells, hPIV-2 V-protein-expressing HeLa and L929 cells, and SV41 V-protein-expressing HeLa cells to IFN-, IFN-, and IFN-. This study shows that the cells expressing hPIV-2 V protein is highly resistant to the antiviral and anti-cell proliferative activities of IFN-/ and that the cysteine rich V-specific domain is required for the high resistance to IFN.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cells. HeLa and L929 cells were grown in Eagle's minimal essential medium (MEM) supplemented with 5% fetal calf serum.

Viruses. Vesicular stomatitis virus (VSV, New Jersey strain), Sindbis virus, and hPIV-2 (Toshiba and CA strains) were used in this study.

IFNs. Human IFN- (hIFN-; 5 × 10⁶ IU), hIFN- (3 × 10⁶ IU), and hIFN- (1 × 10⁶ internal units) were purchased from Mochida Chemical Industries (Osaka, Japan), Tore Co. Ltd. (Tokyo, Japan), and Shionogi Pharmaceutical Co. Ltd. (Osaka, Japan), respectively. Murine IFN-/ was donated by S. Saito (National Institute of Infectious Disease [NIID], Tokyo, Japan). The IFNs used in this study matched those defined as human leukocyte research reference IFN preparation J/501 (IFN-), human fibroblast research reference IFN preparation J/03 (IFN-), and J/R-8703 (IFN-). One unit of the murine IFN-/ in our system was found to be equivalent to 2.7 internal reference units of murine IFN.

Antibodies. Anti-hPIV-2 P-protein (335A) and V-protein (53-1V) monoclonal antibodies (MAbs) were previously described (25, 32). Anti-Stat1 and anti-Stat2 MAbs and anti-Stat2 and anti-PKR (N-18) rabbit polyclonal antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, Calif.).

Proteasome inhibitors. Proteasome inhibitors MG132 and lactacystin were bought from Peptide Institute (Osaka, Japan).

Titration of IFN sensitivity. IFN sensitivity was determined using various IFNs and VSV or Sindbis virus as a challenge virus by the procedure of Ito et al. (12). The highest dilution of the titrated IFN sample causing at least 50% protection was considered the endpoint. IFN sensitivity was expressed as the minimum international units of IFN causing 50% protection.

Establishment of cell lines which constitutively express virus-specific protein. A cDNA clone of the hPIV-2 V gene or deletion-containing V gene was inserted into plasmid pcDL-SR296 between PstI and KpnI sites. The cDNA fragment was inserted between the ClaI and SalI sites of the vector pkan2. Plasmid pkan2 contains the G418 (Geneticin; GIBCO) resistance gene. The promoter in pkan2 is identical to the SR promoter (SV40+ adult T-cell leukemia virus [ATLV] promoters). HeLa or L929 cells were transfected with each plasmid with Lipofectin (GIBCO). After incubation at 37°C for 8 h, MEM with 10% calf serum was added. After 2 days of further incubation, the culture medium was changed to MEM containing 10% fetal calf serum, 1 mg of Geneticin per ml, and 0.2% agarose, and the cells were cultured for 3 weeks. Expression of the virus-specific protein was detected by enzyme-linked immunosorbent assay with the specific MAb. Plasmids pcDL-SR296 and pkan2 were kindly donated by Y. Takebe (NIID, Tokyo, Japan). Finally, HeLa cells expressing hPIV-2 V protein (HeLa-V cells), HeLa cells expressing the P-V common domain, amino acids (aa) 1 to 164, of hPIV-2 (HeLa-Vn cells), HeLa cells expressing aa 145 to 225, including the hPIV-2 V-specific domain (HeLa-Vc), L929 cells expressing hPIV-2 V protein (L929-V cells), and HeLa cells expressing SV41 V protein (HeLa-SV41V cells) were established. In addition, HeLa cells expressing Vn plus NP of hPIV-2 (HeLa-Vn+NP cells) was also established. HeLa-P and HeLa-SV41V cells which constitutively express hPIV-2 P and SV41 V protein were previously described (23, 34). Several lines of these cells were used in this study.

Immunofluorescent staining. The cells were fixed with 3% paraformaldehyde for 15 min at room temperature and rinsed twice with phosphate-buffered saline (PBS). The cells were permeabilized with PBS- 0.05% Tween-20 (PBS-T)) for 30 min and washed twice with PBS. The cells were then incubated for 60 min with primary antibody and washed three times with PBS. Next, the cells were incubated for 60 min with FITC-labeled secondary antibodies and washed with PBS. Immunofluorescently stained cells were analyzed using a fluorescent microscope.

Western blot assay. Cell extracts were prepared with lysis buffer (150 mM NaCl, 50 mM Tris-HCl [pH 7.5], 0.6% NP-40) containing 4 mM phenylmethylsulfonyl fluoride. The samples were analyzed by sodium dodecyl sulfate-9 to 13% polyacrylamide gel electrophoresis (SDS-9 to 13% PAGE). Electrophoretic transfer from gels onto polyvinylidene difluoride transfer membranes was carried out as described previously. The membranes were blocked with 5% skim milk in PBS, treated with each MAb at room temperature for 1 h, washed three times with PBS-T, and treated with biotinylated secondary antibody for 30 min. After being washed with PBS-T, the membranes were treated with an avidin-biotin-peroxidase complex (Vector Laboratories). After being washed with PBS, one membrane was immersed in methanol-PBS (2:8) containing 4-chloro-1-naphthol (0.3%) and hydrogen peroxide (0.009%), and the other one was immersed in enhanced chemiluminescence (ECL) Western blotting detection reagents (Amersham Pharmacia Biotech, Tokyo, Japan).

Isotopic labeling, radioimmunoprecipitation assay, and SDS-PAGE. Isotopic labeling of cells, radioimmunoprecipitation assay, and SDS-PAGE were done as described elsewhere (32).

RNA isolation and first-strand cDNA synthesis. Total cellular RNA was extracted from 10⁶ cells by the guanidine isothiocyanate-cesium chloride method, as described previously (34). Poly(A)⁺ RNA was purified by oligo(dT)-cellulose chromatography (Pharmacia Biotech). RNA primed with specific primers was reverse transcribed using cloned Moloney murine leukemia virus reverse transcriptase (2.5 U), a 1 mM concentration of each deoxynucleoside triphosphate, 0.2 µg of specific primer, and 1 U of RNase inhibitor in a final volume of 15 µl. The reaction was run at 37°C for 60 min to complete the extension reaction. The reaction mixture was heated to 90°C for 5 min to denature the RNA-cDNA hybrids and quickly chilled on ice.

Reverse transcription-PCR assays (RT-PCR). The first-strand cDNA was subjected to PCR amplification using gene-specific PCR primers as follows: 40-kDa 2',5'-AS (2',5'-AS-40), 5'-TGGCTGAAT TACCCATGCTT-3' and 5'-TGGACAAGGGATGTGAAAAT-3'; 71-kDa 2',5'-AS (2',5'-AS-71), 5'-TTAAATGATAATCCCAGCCC-3' and 5'-AAGATTACTGGCCTCGCTGA-3'; PKR 5'-TTGGCTCAGGTGGATTTGG-3' and 5'-GGCTTTTCTTCCACACAGTC-3'; Stat2 5'-ACAAGGTGCTCATCTACTCTGTGCA-3' and 5'-GAGGAGTAGGAAGGGCAAAGAGATA-3'; and -actin (control), 5'-TGACGGGGTCACCCACACTGTGCCCATCTA-3' and 5'-CTAGAAGCATTTGCGGTGGACGATGGAGGG-3'. PCR was performed in 50-µl reaction mixtures containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, a 0.2 mM concentration of each deoxynucleoside triphosphate, 0.5 µM concentrations of each of the sense and antisense PCR primers, and 2.5 U of Taq DNA polymerase (Perkin-Elmer Cetus, Norwalk, Conn.). The reaction mixture was then subjected to 18, 25, 23, or 30 cycles of amplification in a DNA thermal cycler. Each cycle consisted of a heat denaturation step at 94°C for 1 min, annealing of primers at 50 to 60°C (optimized for each primer pair) for 1 min, and an extension step at 72°C for 1 min. Following completion of 18, 25, 23, or 30 PCR cycles, the mixtures were incubated at 72°C for 5 min. The PCR products were separated by electrophoresis on a 1.5% agarose gel and visualized by ethidium bromide staining with UV illumination. To confirm that the PCR products were derived from target mRNAs, the products were cloned using a TA cloning kit (Invitrogen, San Diego, Calif.) and the nucleotide sequences were analyzed.

Purification of recombinantly expressed protein. The plasmid pCAL-P or pCAL-V, which was inserted into the bacterial expression vector pCAL-n-EK, was transferred to Escherichia coli BL21(DE3), and expression was induced by the addition of 1 mM IPTG (isopropyl--D-thiogalactopyranoside). The proteins were expressed as fusion proteins with calmodulin-binding peptide and purified as described previously (25). The purified fusion proteins were cleaved with the site-specific protease EK to remove the calmodulin-binding peptide tag according to the manufacturer's instructions.

In vitro translation of Stat2 mRNA. Full-length cDNA clones of Stat1 and Stat2 mRNAs were prepared by RT-PCR using specific primers. Luciferase cDNA and an in vitro translation kit were also used. In vitro transcription and translation were carried out using TNT Quick coupled transcription-translation systems (Promega, Madison, Wis.) according to the manufacturer's technical manual. In brief, methionine or [³⁵S]methionine and DNA template were added into T7 Quick master mixture containing rabbit reticulocyte lysate, T7 RNA polymerase, amino acid mixture without methionine, and RNase inhibitor. Subsequently, the mixture was incubated at 30°C for 60 min. The radiolabeled products were analyzed by SDS-PAGE, and nonlabeled products were analyzed by Western blotting.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Suppression of IFN susceptibility by hPIV-2 infection. HeLa cells were infected with hPIV-2 (CA strain, nonfusing type) at a multiplicity of infection (MOI) of 5, and at 5 h postinfection (p.i.), the HeLa-CA cells were further cultured with serial twofold dilutions of hIFN-, hIFN-, or hIFN- (starting titer, 5 × 10⁴ U) for 15 h. At 20 h p.i., the cells were superinfected with 100 50% tissue culture infective doses of VSV or Sindbis virus. On day 2 of superinfection, virus-induced cytopathic effect (CPE) was examined. The highest dilution of the IFN sample giving at least 50% protection was taken as the endpoint. IFN sensitivity was expressed as the minimum amount of IFN causing 50% protection. As shown in Table 1, the hPIV-2-infected HeLa cells show high resistance to hIFN- and hIFN- irrespective of whether VSV or Sindbis virus was used as a challenge virus. In addition, when VSV was used, the hPIV-2-infected HeLa cells were 4 × 10³ times less susceptible to hIFN- than HeLa cells. However, the cells showed moderate susceptibility to hIFN- when Sindbis virus was the challenge virus.

Expression and intracellular localization of the P and V proteins in HeLa cells stably expressing P or V protein of hPIV2. It has recently been reported that the C protein of Sendai virus and the V protein of SV5 are related to anti-IFN action (5-7). Thus, we established HeLa cell lines constitutively expressing hPIV2 P protein (HeLa-P cells) and V protein (HeLa-V cells). The expression of the P and V proteins was analyzed by a flow cytometer (data not shown). In order to determine the intracellular localization of the virus-specific proteins, the cells were fixed, immunostained with MAbs specific for hPIV-2 P or V protein, and then observed with an immunofluorescence microscope (Fig. 1A). The P proteins showed diffuse staining throughout the cytoplasm of HeLa-P cells, whereas almost all of the V proteins was present in the nuclei of HeLa-V cells (Fig. 1A and Table 2). In addition, the V proteins were predominantly detected in the nuclei of HeLa-CA cells (Fig. 1A and Table 2). Subsequently, the expression levels of the virus-specific proteins were studied by Western blotting (Fig. 1B).

View larger version (104K): [in this window] [in a new window]   FIG. 1.   (A) Expression and intracellular localization of viral proteins in HeLa cells stably expressing NP, P, or V protein of hPIV2. HeLa-V (a and b), HeLa-Vc (c), L929-V (d), HeLa-SV41V (e), HeLa-P (f), HeLa-Vn (g), HeLa-Vn+NP (h and i), and HeLa-CA (j and k) cells were fixed, immunostained with MAbs specific for hPIV-2 P (a and f), Vc (b, c, d, and j), Vn (e, g, and h), or NP (i and k), and then observed with an immunofluorescence microscope. The anti-hPIV-2 Vn (P-V common domain) MAb used in this study can react with SV41 V protein (32). (B) Detection of virus-specific proteins by Western blotting in various cells: The lysates of various cells were analyzed by SDS-13% PAGE, and then electrophoretic transfer from gels onto polyvinylidene difluoride transfer membranes was done. Virus-specific proteins were detected by Western blotting using MAbs against Vn and NP of hPIV-2 (a) or Vc (b).

Susceptibility of HeLa-P and HeLa-V cells to IFN. HeLa-P and HeLa-V cells were grown to confluence, the culture fluids were removed, and serial twofold dilutions of hIFN-, hIFN-, or hIFN- were added. IFN titers were determined by the CPE inhibition method using VSV or Sindbis virus as a challenge virus. HeLa-P cells showed moderate susceptibility to all IFNs, while HeLa-V cells showed high resistance to hIFN- and hIFN- irrespective of whether VSV or Sindbis virus was used (Table 1). In addition, when VSV was the challenge virus, HeLa-V cells were about 3 × 10³ times less susceptible to hIFN- than HeLa-P cells (Table 1 and data not shown). However, the cells showed relatively good susceptibility to hIFN- when Sindbis virus was the challenge virus (Table 1). Furthermore, HeLa cells expressing the hPIV-2 NP protein showed moderate sensitivity to all the IFNs (Table 1). In addition, murine L929 cells constitutively expressing hPIV-2 V protein showed high resistance to murine IFN-/ (data not shown).

View larger version (23K): [in this window] [in a new window]   FIG. 2.   Susceptibility of HeLa and HeLa-V cells to anti-cell proliferative action of IFNs. HeLa cells (A) and HeLa-V cells (B) were dispersed in the growth medium at a concentration of 105/ml. Control culture medium (), hIFN- (), hIFN- (), or hIFN- () (103 U) was added to each well at the time of cell seeding. After various periods of incubation, the cells in each well were washed, detached, resuspended in PBS, and enumerated in a hemocytometer. Vertical bars show ranges.

IFN susceptibility of HeLa cells stably expressing the P-V common domain or cysteine-rich V-specific domain. In the next experiment, we examined whether the site required for resistance to IFN-/ was located at the N terminus (P-V common domain) or C terminus (cysteine-rich V-specific domain). Therefore, we established HeLa cells constitutively expressing the C-terminally truncated V protein, Vn (HeLa-Vn cells) or the N-terminally truncated V protein, Vc (HeLa-Vc cells), and then investigated the IFN susceptibility of these cells. Vc contains 20 amino acids of the P-V common domain plus the V-specific domain. HeLa-Vn+NP cells stably express both hPIV-2 NP and Vn proteins. NP was detected in the cytoplasm of HeLa-Vn+NP and HeLa-CA cells (Fig. 1A and Table 2). As shown in Fig. 1A, Vn is localized in the cytoplasm of HeLa-Vn+NP cells, while it can be detected in both the cytoplasm and nuclei of HeLa-Vn cells. Vc is detected exclusively in the nuclei of HeLa-Vc cells. However, the expression levels of the truncated forms of the proteins (Vn and Vc) were relatively low (Fig. 1B), and the low expression may be subject to increased turnover. As shown in Table 1, HeLa-Vn cells have almost the same susceptibility to IFN-/ as HeLa and HeLa-P cells. In contrast, HeLa-Vc cells showed resistance to IFN-/ and also showed high resistance to IFN- when VSV was used as a challenge virus (Table 1). These findings indicate that the IFN resistance determinant in the V protein maps to the C terminus (cysteine-rich V-specific domain).

High IFN resistance of HeLa cells stably expressing the SV41 V protein. To investigate whether the V proteins of rubulaviruses other than hPIV-2 were also capable of countering IFN action, we established HeLa cells stably expressing the SV41 V protein (HeLa-SV41V cells). The SV41 V protein was also exclusively detected in the nucleus (Fig. 1A and Table 2). As shown in Table 1, HeLa-SV41V cells show complete resistance to hIFN-, hIFN-, and hIFN- when either VSV or Sindbis virus is used as a challenge virus.

Defect of Stat2 in HeLa-CA and HeLa-V cells. Since HeLa-CA cells and HeLa-V cells have high resistance to the anticellular action as well as to the antiviral action of IFN-/, we tried to detect expression of ISG products in HeLa, HeLa-CA, HeLa-P, and HeLa-V cells stimulated with IFN. Stat1 and Stat2, components of ISGF3, were chosen for the ISG products. The cells were treated for 15 h with hIFN- (10³ or 10⁴ U), hIFN- (10³ or 10⁴ U), or hIFN- (10³ U). A small amount of Stat1 was found in HeLa, HeLa-CA, HeLa-P, and HeLa-V cells which were not stimulated with IFN, and the expressed levels of Stat1 were not significantly different among these cells (Fig. 3A and B). In addition, a small amount of Stat2 was also found in unstimulated HeLa and HeLa-P cells, but it was not detected in HeLa-CA and HeLa-V cells even by a highly sensitive method (ECL) (Fig. 3A and B). Treatment of HeLa and HeLa-P cells with all IFNs remarkably increased Stat1 and Stat2 in these cells (Fig. 3A and B). IFN-/ did not stimulate increased expression of Stat1 and Stat2 in HeLa-CA and HeLa-V cells (Fig. 3A and B). Thus, IFN-/ signaling is not stimulated by IFN-/ in HeLa-V and HeLa-CA cells, although there is no defect in the amount of Stat1. On the other hand, hIFN- enhanced expression of Stat1 in HeLa-CA and HeLa-V cells, but Stat2 was not detected in these cells stimulated with hIFN- (Fig. 3A and B). In addition, Stat2 was not found in HeLa-Vc cells stimulated with IFN-/ (data not shown). Therefore, IFN-/ resistance of HeLa-CA and HeLa-V cells is due to a defect of Stat2 in these cells.

View larger version (34K): [in this window] [in a new window]   FIG. 3.   Expression of Stat1 and Stat2 in HeLa-CA, HeLa-P, HeLa-V, and HeLa-SV41V cells. HeLa (A), HeLa-CA (A), HeLa-P (B), HeLa-V (B) and HeLa-SV41V cells (C) were treated for 15 h without or with hIFN- (103 and 104 U), hIFN- (103 and 104 U) or hIFN- (103 U). Stat1 and Stat2 were detected by Western blotting. (D and E) Low labeling of Stat2 in HeLa-V cells with [35S]methionine. HeLa, HeLa-P, and HeLa-V cells were cultured without or with IFN- (103 U) for 15 h, and then the cells were labeled with [35S]methionine (500 µCi/ml) for 30 min. The cell lysates were analyzed by immunoprecipitation using anti-Stat2 MAb (D), anti-Stat2 polyclonal antibody (E), or anti-Stat1 MAb (E) and SDS-PAGE.

Presence of Stat2 mRNA in HeLa-V cells. The above findings indicate that high IFN resistance of HeLa-V cells is due to a complete defect of Stat2. There are three possibilities for the cause of Stat2's disappearance in HeLa-V cells: (i) block of Stat2 mRNA synthesis; (ii) failure of translation of Stat2 mRNA; and (iii) instability of Stat2. First, for the detection of Stat2 mRNA, RT-PCR was carried out using total RNA isolated from HeLa, HeLa-P and HeLa-V cells. As shown in Fig. 4A, a proper amount of Stat2 mRNA can be detected in these cells. In addition, when we analyzed RNA isolated from the cytoplasmic fraction, Stat2 mRNA was also detected in the cytoplasm of HeLa-V cells (Fig. 4B). These findings indicate that the defect of Stat2 is not caused by suppression of transcription.

View larger version (39K): [in this window] [in a new window]   FIG. 4.   (A) Presence of Stat2 mRNA in HeLa-V cells. HeLa, HeLa-P, and HeLa-V cells were cultured for 15 h, and then mRNAs were isolated. RT-PCR was carried out for the detection of Stat2 mRNA using total RNA (1 µg) isolated from HeLa, HeLa-P, and HeLa-V cells. (B) Presence of Stat2 mRNA in the cytoplasm of HeLa-V cells. HeLa-V, HeLa-P, and HeLa-SV41V cells were cultured for 15 h, and then mRNAs were isolated from cytoplasmic fractions of these cells. RT-PCR (30 cycles) was carried out for the detection of Stat2 mRNA using total RNA (1 µg). (C) Purification of hPIV-2 P and V proteins: Recombinantly expressed P and V proteins of hPIV-2 were analyzed by SDS-10% PAGE and then stained with Coomassie brilliant blue solution. M, marker proteins. (D) Suppression of in vitro translation of Stat2 mRNA by the V protein. Full-length luciferase, Stat1, or Stat2 mRNA was synthesized in vitro by using T7 polymerase. Luciferase cDNA and an in vitro translation kit were used. In vitro translation was carried out using a TNT Quick coupled transcription-translation system (Promega). The reaction (final volume, 50 µl) was performed in the absence (lane 2) or presence of purified P protein (lane 3, 2 µg/ml; lane 4, 0.6 µg/ml) or purified V protein (lane 5, 2 µg/ml; lane 6, 0.6 µg/ml) with [35S]methionine (20 µCi). Lane 1, reaction without template. The products were analyzed by SDS-10% PAGE. (E) Suppression of in vitro translation of Stat2 mRNA by V protein. The specificity of the product was checked by Western blot using specific antibody. The reaction was performed in the absence (lane 2) or presence of purified P protein (lane 3, 20 µg/ml; lane 4, 2 µg/ml) or purified V protein of hPIV2 (lane 5, 20 µg/ml; lane 6, 2 µg/ml). Lane 1, reaction without template. The products were analyzed with Western blotting using anti-Stat2 MAb. (F) Absence of degradation of Stat2 incubated with the purified V protein: In vitro-translated and radioisotope-labeled Stat1 or Stat2 was incubated with either purified P or V protein (20 or 2 µg/ml) for 90 min at 30C. Subsequently, the samples were analyzed by SDS-10% PAGE.

Suppression of in vitro translation of Stat2 mRNA by the V protein. Subsequently, we tested for an inhibitory effect of the V protein on translation of Stat2 mRNA. Luciferase, Stat1, and Stat2 mRNAs were synthesized in vitro by using T7 polymerase, and then the mRNAs were translated in reticulocyte lysates in the presence of [³⁵S]methionine together with purified V or P protein of hPIV-2 (Fig. 4C). As shown in Fig. 4D, the V protein of hPIV-2 suppressed the in vitro translation of Stat2 mRNA more extensively than that of Stat1 mRNA. In contrast, the V protein scarcely suppressed the in vitro translation of luciferase mRNA and the P protein did not suppress the in vitro translation of any mRNA (Fig. 4D). Subsequently, we tried further to detect an inhibitory effect of the V protein on translation of Stat2 mRNA by using Western blotting. As shown in Fig. 4E, the V protein inhibited the in vitro translation of Stat2 mRNA, but the P protein did not. In addition, Stat2 translated in vitro was not degraded by incubation with either the V or P protein (Fig. 4F). These findings suggest that failure of translation is one of mechanisms by which Stat2 expression is abolished in the hPIV-2 V-expressing cells.

Effects of hPIV-2 infection on Stat2 degradation. When HeLa cells were infected with hPIV-2 (CA strain) at a high MOI (MOI of 5), the amount of Stat2 was decreased and a 50% reduction was detected at 3 to 4 h p.i. (Fig. 5A and B). On the other hand, Stat1 levels remained constant throughout the experiment (Fig. 5A and B). Subsequently, HeLa cells were pulse-labeled for 2 h, and then the cells were chased for various times without or with hPIV-2 infection (MOI, 5). During the initial 30 min of chase incubation, labeled Stat2 was increased in uninfected cells, indicating that [³⁵S]methionine remained after chase incubation and the residual radioisotope-labeled methionine was used for protein synthesis in uninfected cells (Fig. 5E). Interestingly, this finding suggests that synthesis of Stat2 is suppressed in hPIV-2-infected HeLa cells, because the degradation rate at the initial stage in uninfected cells is not different from that in hPIV-2-infected cells (Fig. 5E). The half-life of Stat2 was estimated to be 3.5 and 2 h in uninfected and hPIV-2-infected HeLa cells, respectively, under our experimental conditions (Fig. 5C, D, and E). These findings show that Stat2 degradation is enhanced in hPIV-2-infected HeLa cells.

View larger version (24K): [in this window] [in a new window]   FIG. 5.   Effects of hPIV-2 infection of Stat2 degradation. (A) HeLa cells were preinfected or infected with hPIV-2 (CA strain) at an MOI of 5 for 30 min or 1, 2, 4, or 8 h, and then expression of Stat1 and Stat2 was analyzed by Western blotting. (B) Plot of the NIH Image analysis of the Stat2 bands in panel A. Data are expressed as percentages of the baseline (the value in the uninfected-cell lane). (C) Pulse-chase experiments with HeLa cells infected with hPIV-2 (CA strain). HeLa cells were labeled with [35S]methionine (250 µCi/ml) for 2 h. Label was removed, and the cells were washed in normal medium and chased in the presence of MEM supplemented with 250 mM methionine for various times without or with hPIV-2 infection (MOI, 5). The cell lysates were analyzed by immunoprecipitation using anti-Stat2 MAb and SDS-PAGE. (D and E) Plot of the NIH Image analysis of the Stat2 bands in panel C. Data are expressed as percentages of the baseline (the value in the prechase lane [D] or the value in the 30-min chase lane [E]).

Effects of proteasome inhibitors on the expression of Stat2 in HeLa-V cells and HeLa-SV41V cells. Since it has recently been reported that the SV5 V protein enhances proteasome-mediated Stat1 degradation (5), we tested the effects of the proteasome inhibitors MG132 and lactacystin on expression of Stat2 in HeLa-V cells under two experimental conditions: (i) HeLa-V cells were incubated with various concentrations (10, 30, or 90 µM) of MG132 for various periods (3, 6, 9, or 18 h), and (ii) HeLa-V cells stimulated with IFN- or IFN- (1,000 U) were incubated with 10 µM MG132 or lactacystin for 18 h.

View larger version (39K): [in this window] [in a new window]   FIG. 6.   Effects of proteasome inhibitors on the expression of Stat2 in HeLa-V and HeLa-SV41V cells. (A) HeLa-V cells were incubated with 0 (0.3% dimethyl sulfoxide), 10, 30, or 90 µM MG132. After 3, 6, 9, or 18 h, the expression of Stat2 was analyzed by Western blotting (ECL). (B and C) HeLa-V (B) and HeLa-SV41V cells (C) were preincubated without or with the proteasome inhibitors MG132 (M; 10 µM) and lactacystin (L; 10 µM) for 3 h. Subsequently, hIFN-, -, or - (103 U) was added to the culture fluids of these cells. After 15 h, the expression of Stat1 and Stat2 was analyzed by Western blotting (ECL).

Expression of PKR and 2',5'-AS in HeLa-V cells stimulated with hIFN-. As described above, when VSV was used as a challenge virus, the hPIV-2-infected HeLa cells and HeLa-V cells were about 10³ times less susceptible to hIFN- than HeLa cells. However, these cells showed moderate susceptibility to hIFN- when Sindbis virus was used as a challenge virus. Consequently, we studied induction of antiviral substances in HeLa-V cells stimulated with hIFN- (Fig. 7). HeLa, HeLa-P, and HeLa-V cells were cultured in the presence or absence of hIFN- for 15 h, and then mRNAs were isolated from these cells. We analyzed 2',5'-AS-71, 2',5'-AS-40, and PKR mRNAs by semiquantitative RT-PCR (Fig. 7). In addition, PKR was also assayed by Western blotting (data not shown). Induction of 2',5'-AS-71 mRNA was observed in HeLa, HeLa-P, and HeLa V cells stimulated with hIFN-, but the amount of induced mRNA was lower in HeLa-V cells than in HeLa and HeLa-P cells. Considerable amounts of PKR, PKR mRNA, and 2',5'-AS-40 mRNA were found in untreated HeLa, HeLa-P, and HeLa-V cells, and treatment of these cells with hIFN- enhanced the expression of 2',5'-AS-40 and PKR mRNAs.

View larger version (62K): [in this window] [in a new window]   FIG. 7.   Expression of PKR and 2',5'-AS mRNAs in HeLa, HeLa-P, and HeLa-V cells stimulated with IFN-. HeLa, HeLa-P, and HeLa-V cells were cultured with or without hIFN- (103 U) for 15 h, and then mRNAs were isolated from the cells. We analyzed 2',5'-AS-71, 2',5'-AS-40, PKR, and -actin mRNAs by semiquantitative RT-PCR (18, 23, and 30 cycles, indicated in parentheses).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

IFN susceptibility and IFN production in virus-infected cells have been studied by many investigators, including Hermodsson (8), who reported that parainfluenza type 3 virus enhanced the growth of superinfecting Newcastle disease virus in calf kidney cells and suggested that this was due to the inhibition by the former virus of the production and antiviral action of IFN. Similar experimental results were also obtained in HeLa cells persistently infected with Sendai virus (16). We previously studied the relationships between virus-infected cells and IFN systems by using temperature-sensitive Sendai virus (Sen_ts) (9-11). L929 cells persistently infected with Sendai virus (L-Sen_ts cells) were less susceptible to both the antiviral action and the anticellular action of IFN (10). Furthermore, baby hamster kidney cells, hamster tumor-derived cells, LLCMK₂ cells, and Vero cells persistently infected with Sendai virus were also less susceptible to the antiviral action of IFN than the same cell lines not infected with Sendai virus. In addition, the IFN-producing capacity of L-Sen_ts cells was suppressed (9). Interestingly, IFN susceptibility and IFN-producing capacity in L-Sen_ts cells were restored shortly by a temperature shift up to 38°C (nonpermissive temperature), and the suppression of IFN production was brought about immediately after a shift down to the permissive temperature (9-11). In addition, IFN sensitivity of Vero-Sen_ts cells was temperature sensitive (11). These findings indicate that the reduced abilities can hardly be due to a factor(s) apart from a virus component blocking the action of IFN and are related to replication and maturation of the virus in virus-infected cells.

Recently, many investigators have focused their attention on the molecular mechanisms involved in the anti-IFN effect mediated by paramyxovirus (4-7, 14, 36, 37). Sendai virus, hPIV-3, SV5, and mumps virus have been found to block both IFN-/ and IFN- signaling, whereas hPIV-2 blocks IFN-/ signaling (37). There was a specific reduction in the level of the serine 727-phosphorylated form of Stat1 in Sendai virus- and hPIV-3-infected cells (37). The C protein of Sendai virus is responsible for preventing the induction of an IFN-induced antiviral state (6, 7). In contrast, the V protein of SV5 targets Stat1 for proteasome-mediated degradation and is thus responsible for the observed block in IFN signaling in SV5-infected human cells, although SV5 does not inhibit the IFN-/-responsive promoter in murine cells (5). A specific loss of Stat2 in hPIV-2-infected cells was reported by Young et al. (37). In various cells persistently infected with mumps virus, Stat1, but not Stat2, disappeared, and no difference between the levels of Stat1 mRNA transcript in the persistently infected cells and uninfected control cells was observed (36). Unexpectedly, the level of Stat1 apparently could not be improved by treating the cells with proteasome inhibitors (36).

HeLa-CA cells and HeLa-V cells showed complete resistance to hIFN- and hIFN- irrespective of whether VSV or Sindbis virus was used as a challenge virus. In addition, when VSV was used, these cells were about 10³ times less susceptible to hIFN- than control HeLa cells and HeLa-P cells. On the other hand, HeLa-SV41V cells showed complete resistance to all IFNs when VSV and Sindbis virus were used. Furthermore, the multiplication of HeLa-V cells was not inhibited by IFN-/, while the multiplication of HeLa-V cells was distinctly suppressed by hIFN-, showing that HeLa-V cells are also resistant to the anti-cell proliferative action of IFN-/.

HeLa cells constitutively expressing the C-terminally truncated V protein had almost the same susceptibility to IFN-/ as HeLa or HeLa-P cells. In contrast, HeLa cells constitutively expressing the N-terminally truncated V protein showed high resistance to IFN-/. We have recently recovered infectious V-knockout hPIV-2 from cDNA clones which possesses a defective V protein that does not have the unique cysteine-rich domain in its carboxyl terminus (13). Interestingly, the V-knockout hPIV-2 is highly sensitive to IFN-/ (13). These findings indicate that the IFN resistance determinant in the V protein maps to the C-terminal half (cysteine-rich V specific domain). Ohgimoto et al. (26) reported that the hPIV-2 V-P gene encoded the V protein and that the P protein mRNA was produced by addition of two nontemplate G residues, which resulted in a frameshift and the expression of the P protein as a fusion protein with the N-terminal 164 aa of V protein. Several properties have been ascribed to the V protein of paramyxovirus. The V protein of hPIV-2 has one binding domain to the NP protein in the P-V common domain, which is located in the N-terminal region, aa 1 to 46 (23). The V proteins in the cells infected with hPIV-2 or transfected with the V-specific cDNA clone are localized in the nucleus of the cells (24, 25, 34). Two noncontiguous regions in the hPIV-2 V protein, aa 1 to 46 and aa 175 to 196 (cysteine-rich V-specific domain), are required for nuclear localization and retention (34). The V proteins of SV5 also interact with both viral NP and cellular proteins (damage-specific DNA binding protein) (15). At present, it is not known whether the V protein directly interacts with any Stat proteins.

In this study, the hPIV-2 V protein blocked IFN-/ signaling, while the SV41 V protein blocked both IFN-/ and IFN- signaling. A complete defect of Stat2 was found in HeLa-CA and HeLa-V cells, whereas no expression of Stat1 was detected in HeLa-SV41V cells treated with or without all IFNs. When HeLa, HeLa-P, and HeLa-V cells were pulse-labeled with a relatively large amount of [³⁵S]methionine, Stat2 was scarcely detected in HeLa-V cells, whereas it was detected in HeLa and HeLa-P cells, indicating that synthesis of Stat2 is suppressed or Stat2 is very rapidly degraded in HeLa-V cells. When HeLa cells were infected with hPIV-2 (CA strain) at a high MOI, the amount of Stat2 was decreased, and a 50% reduction was detected at 3 to 4 h p.i. On the other hand, Stat1 levels remained constant throughout the experiment. The pulse-chase experiment showed estimated half-lives of Stat2 of approximately 3.5 and 2 h in uninfected and hPIV-2 infected HeLa cells, respectively, under our experimental conditions. This finding shows that Stat2 degradation is enhanced in hPIV-2-infected HeLa and HeLa-V cells.

Stat2 mRNA was detected in the cytoplasm of HeLa-V cells. An extremely small amount of Stat2 was detected in the HeLa-V cells treated with the proteasome inhibitors MG132 and lactacystin (31), suggesting that Stat2 is degraded by the proteasome-mediated process in the HeLa-V cells. However, IFN-/ did not enhance the expression of Stat1 and Stat2 in the presence of the proteasome inhibitors. In addition, the disappearance of Stat1 in HeLa-SV41V cells was not blocked by the proteasome inhibitors. Thus, whether a complete defect of Stat2 in the HeLa-V cells can be explained by the enhanced proteasome-dependent degradation remains to be investigated. In other words, proteasome-independent degradation might contribute to the V protein-mediated degradation. One of the most interesting findings in this study is that the hPIV2 V protein suppresses, but the P protein shows no effect on, in vitro translation of Stat2 mRNA in a reticulocyte lysate system. The V protein showed little influence on in vitro translation of luciferase mRNAs and did not directly degrade Stat2. Thus, it is inferred that a failure of translation may be one of mechanisms by which Stat2 expression is abolished in the hPIV-2 V-expressing cells. However, since the V protein is an RNA binding protein and is capable of binding to various proteins, the V protein may also exert some general effects on the ability of a ribosome to assemble in vitro. Thus, there is a possibility that suppression of in vitro translation of Stat2 mRNA by the V protein does not work in vivo. Further investigation is required for clarification of the involvement of the translation suppression in the Stat2 defect in hPIV-2 V-expressing cells. There is no difference in IFN sensitivity between wild-type measles virus and V-deficient measles virus (27). Therefore, it is possible that the V protein of different members of the Paramyxoviridae function differently to overcome the antiviral effect of the immune system (20).

When VSV was used as a challenge virus, the hPIV-2-infected HeLa cells and HeLa-V cells were about 10³ times less susceptible to hIFN- than HeLa cells. However, these cells showed moderate susceptibility to hIFN- when Sindbis virus was used as a challenge virus. The antiviral activity of IFNs is mediated by multiple cellular proteins. Among the IFN-induced cellular proteins with antiviral activity are PKR, the enzymes of the 2',5'-oligoadenylate pathway, and the Mx proteins (28). These proteins show selective antiviral activities. Thus, the apparent molecular mechanism which is primary responsible for the inhibition of virus replication may differ considerably between virus types and even host cells. IFN-/ and IFN- mainly act through Stat1/Stat2/p48 (ISGF3) binding to the IFN-stimulated response element and through Stat1/Stat1 homodimers binding to the gamma-activating sequence, respectively (30). It has recently been reported that IFN- can also involve the activation of ISGF3 in mouse primary embryonic fibroblasts, melanoma cells, and endothelial cells (19, 22, 35). Very low protection against VSV in HeLa-V cells stimulated by IFN- may be due to a lack of activation of ISGF3 resulting from a complete defect of Stat2.

It is not clear which pathway plays a dominant role in preventing the replication of various types of viruses. Chebath et al. (2) studied Chinese hamster ovary cell clones expressing high constitutive levels of 2',5'-AS as a result of transfection with the cDNA encoding 2',5'-AS-40. Elevated enzyme levels correlated directly with resistance to infection by a picornavirus such as Mengo virus but did not make the cells resistant to VSV. Coccia et al. (3) confirmed these results; that is, in the 2',5'-AS full-length cDNA transfected cell clones, where 2',5'-AS accumulated in the absence of IFN treatment, inhibition of encephalomyocarditis virus (EMCV) replication was observed. However, the constitutive expression of this enzyme did not protect cells against VSV replication. These findings indicate that the 2',5'-AS pathway seems to be directly involved in the inhibition of EMCV but not of VSV replication. The inhibitory action of PKR is also restricted to EMCV, as no reduction is observed for the growth of a rhabdovirus, VSV (21). In addition, when PKR-defective HeLa cells were treated with IFN, these cells remained antiviral for VSV but not for EMCV (21). Taken together, these observations indicate that the two IFN-induced double-stranded-RNA-activated enzymes, PKR and 2',5'-AS, restrict their antiviral action to certain viruses, and the antiviral action of IFN against VSV is mediated by mechanisms distinct from 2',5'-AS or PKR pathways (28). This effect is probably a consequence of the replication cycle of each virus. The primary site of the antiviral action of IFN on EMCV is at the translational level, whereas that on VSV may occur at the level of primary transcription (1, 18) and/or viral protein synthesis (33). VSV maturation may also be affected by IFN treatment (17, 29). Therefore, hPIV-2 V-expressing cells stimulated by IFN- offer a good system for investigating the molecular mechanism of antiviral action of IFN against VSV.

Induction of 2',5'-AS-71 mRNA was observed in HeLa, HeLa-P, and HeLa-V cells stimulated with hIFN-, although its induction is lower in HeLa-V cells than in HeLa and HeLa-P cells. Considerable amounts of PKR, PKR mRNA, and 2',5'-AS-40 mRNA were found in untreated HeLa, HeLa-P, and HeLa-V cells, and treatment of these cells with hIFN- enhanced the expression of PKR and 2',5'-AS-40 mRNAs. However, these results cannot explain the molecular mechanism(s) by which HeLa-V cells show different IFN- susceptibilities dependent on types of challenge virus.

The antagonistic relationship between viruses and the IFN system is universal. Just as the IFN system tries to block the replication of viruses, many viruses have evolved mechanisms to counteract the IFN system. Studies on the anti-IFN action of viruses are important for understanding viral pathogenesis and antiviral mechanisms of IFN.