The Ribonuclease L-Dependent Antiviral Roles of Human 2Â€²,5Â€²

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 587 (2013) 156–164

journal homepage: www.FEBSLetters.org

The ribonuclease L-dependent antiviral roles of human 20,50-oligoadenylate synthetase family members against hepatitis C virus ⇑ Young-Chan Kwon a, Ju-Il Kang a, Soon B. Hwang b, Byung-Yoon Ahn a,

a School of Life Sciences and Biotechnology, Korea University, Anam-dong 5-1, Seoul 136-701, Republic of Korea b Ilsong Institute of Life Science, Hallym University, Gwanyang-dong 1605-4, Anyang 431-060, Republic of Korea

article info abstract

Article history: The latent ribonuclease RNase L and the interferon-inducible 20,50-oligoadenylate synthetase (OAS) Received 16 October 2012 have been implicated in the antiviral response against hepatitis C virus (HCV). However, the speciﬁc Accepted 9 November 2012 roles of these enzymes against HCV have not been fully elucidated. In this study, a scarce endoge- Available online 26 November 2012 nous expression and RNA degrading activity of RNase L in human hepatoma Huh7 cells enabled us to demonstrate the antiviral activity of RNase L against HCV replication through the transient Edited by Hans-Dieter Klenk expression of the enzyme. The antiviral potential of speciﬁc members of the OAS family was further examined through overexpression and RNA interference approaches. Our data suggested that Keywords: among the members of the OAS family, OAS1 p46 and OAS3 p100 mediate the RNase L-dependent RNase L, 20,50-Oligoadenylate synthetase antiviral activity against HCV. Hepatitis C virus Ó 2012 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Antiviral protein

1. Introduction often lead to chronic infection, cirrhosis and hepatocellular carcinoma. It is estimated that 170 million individuals around the world RNase L is a latent ribonuclease that is expressed constitutively are currently infected with HCV. Type I IFN in combination with and ubiquitously in almost all mammalian cells. The binding of ribavirin and one of the two recently approved protease inhibitors 20,50-linked oligoadenylate (2–5A) to the N-terminal ankyrin re- is the current standard therapy for chronic hepatitis C [5]. peats of RNase L induces the dimerization of the enzyme and the RNase L and OAS have been implicated in the antiviral response activation of its endonuclease activity. Activated RNase L cleaves against HCV. The two proteins were found to be abundant in the viral and cellular RNAs at single-stranded regions, resulting in the liver tissues of patients with chronic hepatitis C, and they co-local- inhibition of the translation and replication of viral RNA [1]. In addi- ized with HCV NS5A protein, suggesting that they are likely en- tion to the direct antiviral effect of RNase L, an indirect antiviral role gaged in antagonizing interactions with HCV [6]. It has been for this enzyme has been suggested; in particular, it has been found that OAS transcripts are induced in the liver of HCV patients hypothesized that the small RNA fragments that are generated by and model animals [7,8]. HCV patients undergoing IFN therapy RNase L can act as a signal for interferon (IFN) production [2]. demonstrated an elevated level of OAS in sera; this increase in The structurally unique 2–5A, which is a specific activator of OAS was correlated with predicted virological responses in one RNase L, is synthesized through the oligomerization of ATP, which study [9], but they were uncorrelated in another study [10]. is catalyzed by the 20,50-oligoadenylate synthetases (OASs). Despite the strong implications that RNase L and OAS play a role Although the transcription of the OAS gene is induced by IFNs, in the response to HCV infection, their specific roles against HCV OAS proteins are translated as latent enzymes before they are acti- have not yet been elucidated. The cleavage of the HCV genome vated by the binding of dsRNA; this dsRNA is likely of viral origin. RNA by RNase L has been demonstrated in vitro [11]. The RNase Thus, OAS in combination with RNase L constitutes an essential L-mediated cleavage products of transfected HCV RNA triggered antiviral pathway that is induced in response to IFN [3]. IFN producing signals, suggesting an indirect antiviral role of RNase Hepatitis C virus (HCV) is a hepatotrophic member of the flavi- L against HCV [6]. Recently, RNase L was identified along with sev- viruses that possesses a positive-strand RNA genome [4]. HCV-in- eral cellular IFN-stimulated proteins as antiviral proteins that re- fected individuals typically develop persistent infections that strict HCV replicons through an RNAi-based knock-down screening approach [12]. In this study, we demonstrated the antiviral activity of RNase L against HCV replication in human hepa- ⇑ Corresponding author. Fax: +82 2 923 9923. toma Huh7 cells through transient expression of the enzyme. Our E-mail address: [email protected] (B.-Y. Ahn). data further suggested that two members of the OAS family,

0014-5793/$36.00 Ó 2012 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.febslet.2012.11.010 Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164 157

OAS1 p46 and OAS3 p100, mediate the RNase L-dependent antivi- 2.4. The immunoblotting and immunoﬂuorescence assays ral activity against HCV infection. For immunoblotting, cells were lysed in RIPA buffer (10 mM 2. Materials and methods Tris–HCl [pH 7.4], 150 mM NaCl, 1 mM EDTA, 0.5% Triton X-100, 0.1% SDS, 0.25% sodium deoxycholate, and 1 mM PMSF). The pro- 2.1. Cells and viruses teins were separated in a denaturating polyacrylamide gel and transferred onto nitrocellulose membranes (Hybond ECL; GE Human hepatoma Huh7 and Huh7.5 cells were grown in Dul- Healthcare). The blots were incubated with primary and secondary becco’s modiﬁed Eagle’s medium (DMEM) that had been supple- antibodies that had been diluted in TBST buffer (10 mM Tris–HCl mented with 50 lg/ml of gentamicin and 5% fetal bovine serum [pH 7.6], 150 mM NaCl, 0.1% Tween 20) containing 5% skim milk

(Hyclone) at 37 °C in an atmosphere containing 5% CO2. HepG2 and were visualized with PicoWest SuperSignal ECL reagents cells were cultured in the aforementioned medium that had been (Pierce). Polyclonal antibodies to NS5A and MxA were raised in supplemented with 10% FBS. Human lung epithelial carcinoma accordance with previously described procedures [15]. The anti- A549 cells were maintained in RPMI 1640 medium that had been RNase L antibody that was used was purchased from Invitrogen. supplemented with 10% FBS. HCV replicon cells were prepared The anti-tubulin antibody that was used was purchased from Ap- by the electroporation of RNA from the genotype 2a HCV strains plied Biological Materials. The anti-Flag and anti-HA antibodies of JFH-1 [13], J6/JFH-1 [14] or SGR/Luc-JFH1 with a Gene Pulser II were purchased from Sigma, and the anti-core antibody was pur- RF module (Bio-Rad) in accordance with previously described pro- chased from Pierce. cedures [15]. The viral RNA genome was synthesized in vitro from For immunostaining, cells were fixed with 3.7% paraformalde- appropriate templates with the MEGAscript T7 Kit (Ambion). Virus hyde in PHEM buffer (60 mM Pipes, 25 mM HEPES, 10 mM EGTA, titer in the culture supernatant was determined in the focus-form- 2 mM MgSO4, pH 7.0), permeabilized for 10 min in PBS with 0.5% ing assay in accordance with previously characterized procedures Triton X-100, and incubated for 10 min in blocking buffer (PBS con- [16]. As described below, viral replication was assessed by immu- taining 1% BSA and 0.1% Tween 20). The cells were incubated for noblotting, northern blotting or RT-PCR analyses. The luciferase 1 h at room temperature with a 1:1,000 dilution of primary anti- activity for SGR/Luc-JFH1 was measured with the Luciferase Assay bodies in blocking buffer, washed with PBS and incubated for 1 h System (Promega). The recombinant Newcastle disease virus at room temperature in blocking buffer with a 1:1,000 dilution of rNDV-GFP was described in the published literature [17]. Sendai FITC-conjugated sheep anti-mouse IgG (F-3008, Sigma) or Alexa virus was grown in eggs and titrated by a hemagglutination assay. FluorÒ 568 goat-rabbit IgG (Invitrogen Molecular Probes). The cells were stained with 1 lg/ml DAPI in PBS for 1 min, washed with PBS, 2.2. Plasmid construction and transfection and mounted on a slide glass with FluoroGuard antifade reagent (Bio-Rad). Images were obtained with a confocal laser scanning The cDNA for human RNase L, OAS1 isoforms p42, p48 and p52, microscope (LSM 700, Carl Zeiss). OAS2 isoforms p69 and p71, and OAS3 p100 were amplified from the RNA of A549 cells following IFN-a treatment. OAS1 p46 cDNA was prepared from human HT1080 fibrosarcoma cells. The RT-PCR 3. Results was conducted using the ImProm II RT-PCR system (Promega) with oligo(dT) and the specific primers shown in the Supplementary Ta- 3.1. The scarce endogenous expression of RNase L in human hepatoma ble 1. The PCR products were digested with HindIII–BamHI (for cells RNase L) or NotI-XbaI (for OAS1 isoforms) and were then subcloned into p3Flag-CMV10 (Sigma). Amplified OAS2 and OAS3 cDNAs Before investigating the antiviral activity of RNase L against were subcloned into pHM6 (Roche). The nuclease-defective RNase HCV, we examined the expression of RNase L in the human hepa- L constructs, R667A, H672A and R462Q were generated by PCR with toma Huh7 cell line and its subclone Huh7.5, both of which sup- primers shown in Table 1. To construct SGR/Luc-JFH1, the neomycin port HCV replication. Although RNase L (a 80 kDa protein) was resistance gene of pSGR-JFH1 [18] was replaced with the firefly lucif- detected in our immunoblot of the human lung carcinoma A549 erase gene from pGL3-Basic (Promega) in accordance with previ- cell line, it was not detected in Huh7, Huh7.5 or another human ously described procedures [19]. Lipofectamine 2000 (Invitrogen) hepatoma cell line, HepG2 (Fig. 1A). The RNase L expression in was used for the transfection of cells with DNA constructs and poly- A549 cells appeared to be unaffected by IFN-a treatment, whereas ribo(I:C) (Sigma). The siRNAs (GenePharma, sequences shown in Ta- the IFN-inducible protein MxA was markedly induced in all four of ble 1) were co-transfected with JFH-1 RNA by electroporation. the cell lines that we examined. Consistent with the observed expression patterns for RNase L, the specific cleavage of rRNAs, 2.3. RNA analyses the characteristic feature of activated RNase L [20], was observed in the agarose gel electrophoresis of rRNA from the A549 cells that To examine the rRNA cleavage, cellular RNA was prepared with had been either transfected with polyribo(I:C) or infected with the RNAisoPLUS (Takara) and separated in 1.5% agarose gel containing Newcastle disease virus (NDV), whereas no rRNA cleavage was ob- formaldehyde. HCV RNA was analyzed either by northern blotting served in Sendai virus-infected cells (Fig. 1B). In contrast, no rRNA that was performed in accordance with previously described meth- cleavage was detected in the similarly treated Huh7 (Fig. 1C) and ods [15] or by quantitative real-time PCR. For the qPCR amplifica- Huh7.5 cells (data not shown). tions, the ImProm RT-PCR system (Promega) was used to reverse In contrast to the results of our immunoblot data, an antibody transcribe the HCV RNA, and the MyiQ Real-Time PCR system (Bio- had previously been used to detect RNase L in Huh7 and Huh7.5 Rad Laboratories) was used to conduct the subsequent amplification cells [6]. RNase L was also detected in HepG2 cells via the binding with HCV-specific primers (Table 1) under the following conditions: of radiolabeled 2–5A in a previously published study [21]. To ad- 3 min at 95 °C, followed by 40 cycles of 95 °C for 30 s, 55 °C for 30 s dress these differences, we measured RNase L transcripts with and 72 °C for 30 s. The melting curve was obtained through 81 cy- real-time PCR. The RNase L transcript was present in much lower cles; these cycles were each 10 s in length with temperature incre- concentrations in these three cell lines than in A549 cells and ments of 0.5 °C within 55–95 °C. The resulting data were analyzed was barely detectable after amplification (Fig. 1D). These results with the iQ5 optical system software package (Bio-Rad Laboratories). suggested that RNase L might be expressed in these hepatoma cells 158 Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164

A B C

D E

Fig. 1. A scarce endogenous expression of RNase L. (A) The indicated human cells were immunoblotted for RNase L at 24 h after IFN-a (1,000 U/ml) treatment. The blot was also probed for the IFN-inducible MxA and for tubulin as a loading control. (B) A549 cells were transfected with 1 lg/well of polyribo(I:C) or infected with either Sendai virus (100 HAU/ml) or Newcastle disease virus rNDV-GFP (m.o.i. of 2). Total RNA was extracted at 24 h and separated in 1% agarose gel containing formaldehyde. EtBr-stained 28S and 18S rRNA are shown, and the cleavage products are indicated with arrowheads. Note that no cleavage was observed in the cells that had been infected with Sendai virus. (C) No rRNA cleavage was observed in any of the Huh7 cells that had been treated in the same manner as the A549 cells. (D) RNase L mRNA was amplified and separated on an agarose gel (Top) and the relative amount was quantified by real-time PCR as shown in a bar graph (Bottom). (E) Huh7 cells transiently expressing RNase L were transfected at 24 h with polyribo(I:C), and their rRNA was analyzed at the indicated time. Note that rRNA cleavage (arrowheads) was observed in the RNase L-transfected Huh7 cell. A549 cells were used as controls. but that the amount of RNase L that is present in the hepatoma that had been transfected with the RNase L construct were infected cells is insufficient to induce rRNA cleavage, providing us with an at 48 h with HCV J6/JFH1 (m.o.i. of 0.1). The viral core and NS5A opportunity to address the function of this enzyme through its ec- proteins and viral RNA were reduced by 50% in 48 h in the cells topic expression. For this purpose, the coding sequence of RNase L that had been transfected with functional RNase L, whereas the was amplified from A549 cell mRNA and subcloned into an expres- nuclease-defective enzyme (R667A) [23] had no inhibitory effect sion vector with an N-terminal Flag tag. The transient expression of on the HCV proteins and RNA (Fig. 2A). RNase L caused Huh7 cells to commence rRNA cleavage in few The antiviral effect of RNase L was also demonstrated in HCV hours following polyribo(I:C) transfection, whereas no rRNA cleav- replicon cells. Huh7 cells that had been transfected with the RNA age was observed in the control cells (Fig. 1E). This result indicated genome of HCV JFH-1 were further transfected with RNase L. that the ectopically expressed RNase L was activated at this condi- Northern blot and real-time PCR analyses indicated that viral tion in this cell line. RNA was reduced by 50% in the RNase L-transfected cells, compared with either the control cells or cells expressing the nucle- 3.2. The antiviral activity of RNase L against HCV ase-defective enzyme R667A (Fig. 2B). The concomitant occurrence of rRNA cleavage in the RNase L-expressing cells sug- The overexpression of RNase L in cultured cells has been shown gested that the ectopically expressed enzyme was activated in to inhibit several viruses that are present in these cells [22].To the HCV-replicating cells. The antiviral activity of RNase L was also investigate the antiviral activity of RNase L against HCV, Huh7 cells examined in the cells stably expressing RNase L. In this protocol, Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164 159

Fig. 2. The antiviral activity of RNase L against HCV. (A) Huh7 cells transfected with RNase L or the nuclease-defective R667A constructs were infected at 48 h with HCV J6/ JFH-1 (m.o.i. 0.1). Viral NS5A and core proteins were immunoblotted at 48 h. The blot was also probed with anti-Flag antibody for RNase L and for tubulin, which was used as a loading control. The viral RNA was analyzed by northern blotting with the intensity of bands indicated. The 28S rRNA was used as a loading control. (B) Huh7 cells that had been transfected with JFH-1 RNA were transfected 4 h later with RNase L. Total RNA was extracted at the indicated time after viral RNA transfection and the rRNA was analyzed by an agarose gel electrophoresis (shown in middle). Viral RNA was measured in northern blotting (shown in top) and real-time PCR as indicated in a bar graph (bottom). (C) Huh7 cells transfected with the wild-type RNase L or R667A constructs were maintained for 3 weeks in the medium containing G418 (300 lg/ml). The polyclonal cells were pooled and electroporated with JFH-1 RNA. At the indicated time, viral NS5A and core proteins were immunoblotted. RNase L expression was conﬁrmed with Flag antibody with Hsp70 protein used as a loading control. (D) RNase L-expressing polyclonal cells were electroporated with SGR/Luc-JFH-1 and the luciferase activity was measured at the indicated times in triplicate. The means and SD were shown in a bar.

Huh7 cells transfected with the RNase L expression plasmid con- The antiviral potential of RNase L was further examined through structs were maintained in a selective medium and a population the immunostaining of JFH-1 replicon cells. RNase L and HCV NS5A of pooled cells were transfected with the HCV JFH-1 genome. displayed a mutually exclusive distribution such that few or no cells Immunoblotting indicated that the viral core and NS5A proteins co-expressed both proteins, whereas the nuclease-defective RNase were reduced by 50% compared to the control cells or cells stably L proteins and NS5A were often found in the same cells (Fig. 3A). expressing R667A (Fig. 2C). The inhibitory effect of RNase L was Counting of the immunostained cells in the microscopic ﬁelds indi- more evident in reducing the expression of a luciferase reporter cated that 10% of cells transfected with the wild-type RNase L that was encoded in the SGR/Luc-JFH1, a replication-competent were positively stained for NS5A, whereas more than 50% of cells RNA subgenome of JFH-1 (Fig. 2D). In combination, these results transfected with each of the three mutant RNase L constructs were demonstrated the antiviral activity of RNase L against HCV and doubly stained (Fig. 3B). Likewise, in the immunoblot for the pro- suggested that RNase L has been activated in this process. tein expression from JFH-1, no inhibitory activity was observed 160 Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164

Fig. 3. The distributions of RNase L and HCV NS5A. (A) Huh7 cells that had been transfected with JFH-1 RNA were transfected 4 h later with wild-type RNase L or each of the nuclease-defective constructs. These cells were immunostained at 48 h with anti-NS5A (red) and anti-Flag (green) antibodies and examined under a confocal microscope. The two images were overlapped with DAPI-stained nuclear images (blue) in the rightmost panels. (B) The number of NS5A-positive cells (indicated with arrowheads) were count among RNase L-positive cells in 5 randomly set microscopic ﬁelds (40–50 cells in a ﬁeld). The means and SDs of the 5 countings were shown in bars. (C) Huh-7 cells that had been transfected with JFH-1 RNA were transfected 4 h later with wild-type RNase L or each of the nuclease-defective RNase L constructs. Immunoblotting was performed at 48 h.

for the three mutants, namely, R462Q, R667A and H672A [24],in that the antiviral effect was conﬁned to the cells that express the contrast to the wild-type RNase L (Fig. 3C). These results suggested RNase L with intact catalytic activity. Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164 161

3.3. The expression of OAS proteins C) at the splicing acceptor site of the exon 7 has been attributed to the expression of the splicing variants p52 from the A allele (due The activation of RNase L is dependent on 2–5A, which is syn- to the shift of splice-acceptor site one base downstream) and p48 thesized by OAS. OAS expression is induced by IFN-a, but its acti- (produced if a splicing acceptor site further downstream is used) in- vation requires dsRNA. Because our results demonstrating the stead of p46 which is produced from a normally spliced transcript of rRNA cleavage in HCV replicon cells and in polyribo(I:C)-treated G allele. The smallest isoform p42 is commonly produced from both Huh7 cells that overexpressed RNase L suggested the activation of these alleles (Supplementary data Fig. S1A). The G allele is known of this RNase L, and thus the presence and activation of OAS, we to be associated with higher enzyme activity, whereas the more sought to examine the expression of OAS and to assess its antiviral common A allele is associated with lower enzyme activity [26].It role against HCV in the studied cell lines. is particularly noteworthy that patients with the homozygous GG In humans, OAS comprises a family of enzymes that are en- genotype demonstrated a higher response to interferon therapy coded by three closely related and linked genes, namely, OAS1, against HCV infection, whereas the AA genotype was associated with OAS2 and OAS3; these genes contain one, two and three repeats, non-response to interferon and higher degrees of liver fibrosis [27]. respectively, of the basic OAS units. Moreover, several isoforms We found by genomic sequence analysis that Huh7 cell line pos- are produced from alternatively spliced transcripts of each OAS sesses the AA genotype and confirmed by RT-PCR the expression gene; in particular, there exist p42 and p46 isoforms of OAS1; of the splicing variants coding for p42, p48 and p52, respectively, p69 and p71 isoforms of OAS2; and p100 isoform of OAS3. These but no expression of p46 transcript (Supplementary data Fig. S1B). various OAS proteins differ in their enzymatic activities, levels of To assess the antiviral potential of each OAS1 isoform, we sub- expression, and subcellular localization, and these differences cloned cDNA for the OAS1 isoforms and examined their antiviral could contribute to different antiviral potentials for each OAS iso- activity. The p46 cDNA was subcloned from human fibrosarcoma form against specific viruses. In addition to the aforementioned HT1080 cell because it was not present in Huh7 or A549 cells OAS genes, an OAS-like (OASL) gene that encodes a catalytically due to their being AA genotype. Although p46 was not expressed inactive protein also exists [25]. in Huh7 cells, it was included in our analysis because it is expected Although OAS expression is considered to be inducible by IFN, to be present in human liver cells of GG or GA genotype. Interest- cell lines vary in their endogenous and IFN-inducible expression ingly, the expression of p46 showed a notable inhibition for the levels of OAS [3]. We found that OAS1 and OAS3 transcripts were protein expression from HCV JFH-1 in an RNase L-dependent man- constitutively expressed in Huh7 cells (and A549 cells) and were ner, whereas no inhibitory effect was produced by the p42, p48 or significantly induced following IFN treatment, whereas OAS2 tran- p52 isoforms of OAS1 (Fig. 5A). The antiviral activity of p46 was as- scripts were not detectable in the absence of IFN treatment. Impor- sessed more quantitatively against SGR/Luc-JFH1. The reporter tantly, the levels of the OAS transcripts of JFH-1 replicon cells were expression was reduced by up to 40% in the Huh7 cells that were no different from the OAS transcript levels of either the parental co-transfected with RNase L and p46 but no other OAS1 isoforms cells or cells that had been transfected with a replication-defective (Fig. 5B). To rule out the possibility that the different inhibitory ef- HCV genome (Fig. 4). These results suggested that no further fect might be caused by the different level of expression (as seen in induction of OAS occurred in HCV replicon cells and that the the immunoblot Fig. 5A), we repeated the inhibition assays with endogenous OAS transcripts that already existed in these cells, various amount of transfected DNA for each OAS1 isoforms. A 2- which were presumably the constitutively expressed isoforms of to 4-fold increase in the amount of transfected DNA of p42, p48 OAS1 and/or OAS3, might be sufficient to produce the 2–5A that and p52, respectively, showed no stimulation on RNase L, whereas is required for the activation of RNase L. p46 showed a DNA dose-responsive inhibitory effect on HCV (Data not shown). These data clearly demonstrated the antiviral activity 3.4. The antiviral activity of OAS1 p46 against HCV of OAS1 p46. However, the functional integrity of OAS activity in Huh7 cells that we observed without OAS overexpression could We first focused on the OAS1 proteins. A distinctive set of OAS1 not be explained because this cell line lacks p46. isoforms has been reported. Most notably, an SNP (AG/GC vs. AAG/ 3.5. The antiviral role of OAS3 p100

Because our OAS mRNA analysis indicated the constitutive expression of OAS1 and OAS3, we next focused on OAS3. As we did for OAS1 isoforms, we examined the effect of OAS3 p100 overexpression on the expression of the transfected HCV JFH-1 genome. Despite being poorly expressed in this protocol, OAS3 p100 expression resulted in a notable reduction in the expression of JFH-1 proteins (Fig. 5C) and that of a subgenomic replicon (Fig. 5D) in an RNase L-dependent manner. This result was contrasted with no effect seen by the expression of OAS2 p69 or that of OAS2 p71 isoforms. We attempted to reduce the expression of OAS1 or OAS3 specifically through the transfection of synthetic siRNA. Whereas the RNase L-mediated inhibition against HCV JFH-1 was not rescued by OAS1-specific siRNA (Fig. 6A), this inhibition was significantly reversed by an OAS3 siRNA (Fig. 6B). This result suggested that at least in Huh7 cells, OAS3 p100 plays an important role against HCV.

Fig. 4. OAS expression. Transcripts of OAS1, OAS2 and OAS3 were amplified from 1 lg of RNA that had been extracted from A549 or Huh7 cells before or at 24 h after 4. Discussion IFN-a (1000 U/ml) treatment. This figure includes amplified OAS transcripts from Huh7 cells that had been electroporated with JFH-1 RNA or the replication- It was recently demonstrated that the transfection of Huh7 cells defective (GND) RNA. Note that in contrast to the GND RNA, the JFH-1 RNA was amplified. GAPDH mRNA was amplified as a control. The amplified DNA was with 2–5A that had been synthesized in vitro resulted in the inhi- separated in a 1% agarose gel. bition of HCV replication [2]. Although this result demonstrated 162 Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164

Fig. 5. The antiviral activity of OAS1 p46 and OAS3 p100. (A, C) Huh7 cells replicating JFH-1 HCV were co-transfected with RNase L and each of OAS1 (p42, p46, p48 or p52), OAS2 (p69 or p71) or OAS3 p100 constructs. The viral NS5A and core proteins were immunoblotted at 48 h. The blot was also probed with anti-Flag antibody to detect the RNase L and the OAS. Tubulin was probed as a loading control. (B, D) Huh7 cells replicating the subgenome SGR/Luc-JFH1 were co-transfected with RNase L and each of the OAS constructs. The resulting luciferase activity was measured in triplicate at 48 h, as shown in a bar graph. the antiviral role of endogenous RNase L, our analysis indicated which is less active than the other forms of 2–5A for the activation that the endogenous level of RNase L in this hepatoma cell line is of RNase L. Despite these in vitro data, increasing evidence has sug- very low. The additional expression of RNase L enabled us to dem- gested potential antiviral roles for OAS3. For example, the p42 and onstrate this enzyme’s rRNA cleavage and more importantly, the p46 of OAS1 and the p100 of OAS3 inhibited Dengue virus in an antiviral activity against HCV. RNase L most likely has a greater RNase L-dependent manner, whereas the other OAS isoforms did antiviral effect in human liver, given that this enzyme is known not demonstrate this effect [30]. Our data also suggested the to be expressed at a higher level in human liver cells than in the importance of OAS3 p100 against HCV. examined hepatoma cells [21]. The activation of RNase L without the further induction of OAS Our results indicating the anti-HCV activity of p46 but no other expression, as we have shown in JFH-1 replicon cells, is not unique OAS1 isoforms in Huh7 cells suggested that the splicing site alleles to HCV. It was reported that in the coxsackievirus or hepatitis A of OAS1 might be implicated in HCV infection. Genetic analyses virus infection of monkey kidney FRhK-4 cells, RNase L could be have suggested that the A allele of OAS1, which produces p48 activated with no apparent induction of the OAS1 or OAS2 pro- and p52 and is associated with a lower OAS activity, is considered teins; in these FRhK-4 experiments, OAS3 was constitutively ex- to be a risk factor in humans for infection with another ﬂavivirus, pressed at a relatively high level, and this expression level the West Nile virus [28]. It might be interesting to determine the remained unchanged throughout the infection [31]. It appears potential association of this allele with the outcome of HCV infec- likely that only a small quantity of OAS protein is needed as a cat- tions of humans. Another interesting question that also remains alytic enzyme for the activation of RNase L. Thus, constitutively ex- unaddressed is why the isoforms that differ only in their C-termi- pressed OAS1 and/or OAS3 might play a safeguard role through the nal regions demonstrate different activities. activation of RNase L in the early stage of antiviral response against Our data, which indicate the functional integrity of OAS in Huh7 HCV infection. cells despite the absence of OAS1 p46 expression in these cells (and also apparent absence of OAS2 expression), suggested a potential Acknowledgments role of OAS3 against HCV. Earlier studies have established that OAS3 requires much less dsRNA than the other OAS proteins for This work was supported by the Basic Science Research Pro- its activation and mainly produces 2–5A in its dimer form [29], gram of the National Research Foundation (#2012-008126) of the Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164 163

Fig. 6. RNA interference against OAS1 and OAS3. Huh7 cells were co-transfected with JFH-1 RNA and siRNA that was speciﬁc for OAS1 (A) or OAS3 (B); this co-transfection was followed 4 h later by transfection with RNase L. The viral RNA was analyzed at 48 h by northern blotting (Top) and quantiﬁed by real-time PCR (Middle). The expression levels of OAS1 and OAS3 mRNAs were also measured by real-time PCR (Bottom).

Republic of Korea. Y.C.K and J.I.K were supported by the BK21 Pro- [8] Helbig, K.J., Lau, D.T., Semendric, L., Harley, H.A. and Beard, M.R. (2005) gram from the Ministry of Education, Science & Technology of Analysis of ISG expression in chronic hepatitis C identifies viperin as a potential antiviral effector. Hepatology 42, 702–710. Korea. [9] Kim, K.I., Kim, S.R., Sasase, N., Taniguchi, M., Harada, S., Kinoshita, K., Kim, S.H., Akimoto, Y., Shikata, M., Kimura, N., Izawa, S., Ohtani, A., Nakao, K., Motojima, M., Kinoshita, M., Hirai, M., Ohzu, M., Hirooka, T., Nabeshima, S., Ishii, F., Appendix A. Supplementary data Tanaka, K. and Hotta, H. (2006) 20-,50-Oligoadenylate synthetase response ratio predicting virological response to PEG-interferon-alpha2b plus ribavirin Supplementary data associated with this article can be found, in therapy in patients with chronic hepatitis C. J. Clin. Pharm. Ther. 31, 441–446. [10] Shindo, M., Hamada, K., Morikawa, T., Harano, Y., Nakajima, T. and Okuno, T. the online version, at http://dx.doi.org/10.1016/j.febslet.2012. (2008) In vivo interferon system assessed by 20-50 oligoadenylate synthetase 11.010. activity in chronic hepatitis C virus patients treated with pegylated interferon and ribavirin. Hepatol. Res. 38, 1213–1220. References [11] Han, J.Q., Wroblewski, G., Xu, Z., Silverman, R.H. and Barton, D.J. (2004) Sensitivity of hepatitis C virus RNA to the antiviral enzyme ribonuclease L is determined by a subset of efficient cleavage sites. J. Interferon Cytokine Res. [1] Chakrabarti, A., Jha, B.K. and Silverman, R.H. (2011) New insights into the role 24, 664–676. of RNase L in innate immunity. J. Interferon Cytokine Res. 31, 49–57. [12] Metz, P., Dazert, E., Ruggieri, A., Mazur, J., Kaderali, L., Kaul, A., Zeuge, U., [2] Malathi, K., Dong, B., Gale Jr., M. and Silverman, R.H. (2007) Small self-RNA Trippler, M., Lohmann, V., Binder, M., Frese, M. and Bartenschlager, R. (2012) generated by RNase L amplifies antiviral innate immunity. Nature 448, 816– Identification of type I and type II interferon-induced effectors controlling 819. hepatitis C virus replication. Hepatology, http://dx.doi.org/10.1002/hep. 0 0 [3] Hovanessian, A.G. and Justesen, J. (2007) The human 2 –5 oligoadenylate 25908. 0 0 synthetase family: unique interferon-inducible enzymes catalyzing 2 –5 [13] Wakita, T., Pietschmann, T., Kato, T., Date, T., Miyamoto, M., Zhao, Z., Murthy, 0 0 instead of 3 –5 phosphodiester bond formation. Biochimie 89, 779–788. K., Habermann, A., Krausslich, H.G., Mizokami, M., Bartenschlager, R. and [4] Moradpour, D., Penin, F. and Rice, C.M. (2007) Replication of hepatitis C virus. Liang, T.J. (2005) Production of infectious hepatitis C virus in tissue culture Nat. Rev. Microbio. 5, 453–463. from a cloned viral genome. Nat. Med. 11, 791–796. [5] Pawlotsky, J.M. (2012) New antiviral agents for hepatitis C. F1000 Biology [14] Lindenbach, B.D., Evans, M.J., Syder, A.J., Wolk, B., Tellinghuisen, T.L., Liu, C.C., Reports 4, 5. Maruyama, T., Hynes, R.O., Burton, D.R., McKeating, J.A. and Rice, C.M. (2005) [6] Malathi, K., Saito, T., Crochet, N., Barton, D.J., Gale Jr., M. and Silverman, R.H. Complete replication of hepatitis C virus in cell culture. Science 309, 623–626. (2010) RNase L releases a small RNA from HCV RNA that refolds into a potent [15] Kang, J.I., Kwon, S.N., Park, S.H., Kim, Y.K., Choi, S.Y., Kim, J.P. and Ahn, B.Y. PAMP. RNA 16, 2108–2119. (2009) PKR protein kinase is activated by hepatitis C virus and inhibits viral [7] Su, A.I., Pezacki, J.P., Wodicka, L., Brideau, A.D., Supekova, L., Thimme, R., replication through translational control. Virus Res. 142, 51–56. Wieland, S., Bukh, J., Purcell, R.H., Schultz, P.G. and Chisari, F.V. (2002) [16] Kang, J.I., Kwon, Y.C. and Ahn, B.Y. (2012) Modulation of the type I interferon Genomic analysis of the host response to hepatitis C virus infection. Proc. Natl. pathways by culture-adaptive hepatitis C virus core mutants. FEBS Lett. 586, Acad. Sci. USA 99, 15669–15674. 1272–1278. 164 Y.-C. Kwon et al. / FEBS Letters 587 (2013) 156–164

[17] Park, M.S., Garcia-Sastre, A., Cros, J.F., Basler, C.F. and Palese, P. (2003) [26] Bonnevie-Nielsen, V., Field, L.L., Lu, S., Zheng, D.J., Li, M., Martensen, P.M., Newcastle disease virus V protein is a determinant of host range restriction. J. Nielsen, T.B., Beck-Nielsen, H., Lau, Y.L. and Pociot, F. (2005) Variation in Virol. 77, 9522–9532. antiviral 20,50-oligoadenylate synthetase (2’5’AS) enzyme activity is controlled [18] Kato, T., Date, T., Miyamoto, M., Furusaka, A., Tokushige, K., Mizokami, M. and by a single-nucleotide polymorphism at a splice-acceptor site in the OAS1 Wakita, T. (2003) Efﬁcient replication of the genotype 2a hepatitis C virus gene. Am J. Hum. Genet. 76, 623–633. subgenomic replicon. Gastroenterology 125, 1808–1817. [27] El Awady, M.K., Anany, M.A., Esmat, G., Zayed, N., Tabll, A.A., Helmy, A., El [19] Targett-Adams, P. and McLauchlan, J. (2005) Development and Zayady, A.R., Abdalla, M.S., Sharada, H.M., El Raziky, M., El Akel, W., Abdalla, S. characterization of a transient-replication assay for the genotype 2a and Bader El Din, N.G. (2011) Single nucleotide polymorphism at exon 7 splice hepatitis C virus subgenomic replicon. J. Gen. Viol. 86, 3075–3080. acceptor site of OAS1 gene determines response of hepatitis C virus patients to [20] Wreschner, D.H., James, T.C., Silverman, R.H. and Kerr, I.M. (1981) Ribosomal interferon therapy. J. Gastroenterol. Hepatol. 26, 843–850. RNA cleavage, nuclease activation and 2–5A(ppp(A20p)nA) in interferon- [28] Lim, J.K., Lisco, A., McDermott, D.H., Huynh, L., Ward, J.M., Johnson, B., Johnson, treated cells. Nucleic Acids Res. 9, 1571–1581. H., Pape, J., Foster, G.A., Krysztof, D., Follmann, D., Stramer, S.L., Margolis, L.B. [21] Zhou, A., Molinaro, R.J., Malathi, K. and Silverman, R.H. (2005) Mapping of the and Murphy, P.M. (2009) Genetic variation in OAS1 is a risk factor for initial human RNASEL promoter and expression in cancer and normal cells. J. infection with West Nile virus in man. PLoS pathog. 5, e1000321. Interferon Cytokine Res. 25, 595–603. [29] Marie, I., Blanco, J., Rebouillat, D. and Hovanessian, A.G. (1997) 69-kDa and [22] Zhou, A., Paranjape, J.M., Hassel, B.A., Nie, H., Shah, S., Galinski, B. and 100-kDa isoforms of interferon-induced (20-50)oligoadenylate synthetase Silverman, R.H. (1998) Impact of RNase L overexpression on viral and cellular exhibit differential catalytic parameters. Eur. J. Biochem. 248, 558–566. growth and death. J. Interferon Cytokine Res. 18, 953–961. [30] Lin, R.J., Yu, H.P., Chang, B.L., Tang, W.C., Liao, C.L. and Lin, Y.L. (2009) Distinct [23] Dong, B., Niwa, M., Walter, P. and Silverman, R.H. (2001) Basis for regulated antiviral roles for human 20,50-oligoadenylate synthetase family members RNA cleavage by functional analysis of RNase L and Ire1p. RNA 7, 361–373. against dengue virus infection. J. Immunol. 183, 8035–8043. [24] Xiang, Y., Wang, Z., Murakami, J., Plummer, S., Klein, E.A., Carpten, J.D., Trent, [31] Kulka, M., Calvo, M.S., Ngo, D.T., Wales, S.Q. and Goswami, B.B. (2009) J.M., Isaacs, W.B., Casey, G. and Silverman, R.H. (2003) Effects of RNase L Activation of the 2–5OAS/RNase L pathway in CVB1 or HAV/18f infected FRhK- mutations associated with prostate cancer on apoptosis induced by 20,50- 4 cells does not require induction of OAS1 or OAS2 expression. Virology 388, oligoadenylates. Cancer Res. 63, 6795–6801. 169–184. [25] Kristiansen, H., Gad, H.H., Eskildsen-Larsen, S., Despres, P. and Hartmann, R. (2011) The oligoadenylate synthetase family: an ancient protein family with multiple antiviral activities. J. Interferon Cytokine Res. 31, 41–47.