Induction and Inhibition of Type I Interferon ... - Journal of Virology

1 downloads 0 Views 1MB Size Report
Jan 22, 2010 - All Rights Reserved. Induction and Inhibition of Type I Interferon Responses by Distinct. Components of Lymphocytic Choriomeningitis Virus.
JOURNAL OF VIROLOGY, Sept. 2010, p. 9452–9462 0022-538X/10/$12.00 doi:10.1128/JVI.00155-10 Copyright © 2010, American Society for Microbiology. All Rights Reserved.

Vol. 84, No. 18

Induction and Inhibition of Type I Interferon Responses by Distinct Components of Lymphocytic Choriomeningitis Virus䌤 Shenghua Zhou, Anna M. Cerny, An Zacharia, Katherine A. Fitzgerald, Evelyn A. Kurt-Jones, and Robert W. Finberg* Department of Medicine, University of Massachusetts Medical School, Worcester, Massachusetts Received 22 January 2010/Accepted 22 June 2010

Type I interferons (IFNs) play a critical role in the host defense against viruses. Lymphocytic choriomeningitis virus (LCMV) infection induces robust type I IFN production in its natural host, the mouse. However, the mechanisms underlying the induction of type I IFNs in response to LCMV infection have not yet been clearly defined. In the present study, we demonstrate that IRF7 is required for both the early phase (day 1 postinfection) and the late phase (day 2 postinfection) of the type I IFN response to LCMV, and melanoma differentiation-associated gene 5 (MDA5)/mitochondrial antiviral signaling protein (MAVS) signaling is crucial for the late phase of the type I IFN response to LCMV. We further demonstrate that LCMV genomic RNA itself (without other LCMV components) is able to induce type I IFN responses in various cell types by activation of the RNA helicases retinoic acid-inducible gene I (RIG-I) and MDA5. We also show that expression of the LCMV nucleoprotein (NP) inhibits the type I IFN response induced by LCMV RNA and other RIG-I/MDA5 ligands. These virus-host interactions may play important roles in the pathogeneses of LCMV and other human arenavirus diseases. mately leads to the activation and nuclear translocation of interferon regulatory factors (IRFs) and nuclear factor ␬B (NF-␬B), which, in turn, switches on a cascade of genes controlling the production of both type I IFNs and other proinflammatory cytokines (10, 11, 60). Lymphocytic choriomeningitis virus (LCMV) infection in its natural host, the mouse, is an excellent system to study the impact of virus-host interactions on viral pathogenesis and to address important issues related to human viral diseases (1, 45, 49, 67). LCMV infection induces type I IFNs as well as other proinflammatory chemokines and cytokines (6, 41). Our previous studies have demonstrated that TLR2, TLR6, and CD14 are involved in LCMV-induced proinflammatory chemokines and cytokines (66). The mechanism by which LCMV induces type I IFN responses, however, has not been clearly defined (7, 8, 31, 44). The role of the helicase family members RIG-I and MDA5 in virus-induced type I IFN responses has been recently established. RIG-I has been found to be critical in controlling the production of type I IFN in response to a number of RNA viruses, including influenza virus, rabies virus, Hantaan virus, vesicular stomatitis virus (VSV), Sendai virus (SeV), etc. In contrast, MDA5 is required for responses to picornaviruses (15, 25, 63). In the present study, we demonstrated that LCMV genomic RNA strongly activates type I IFNs through a RIG-I/MDA5dependent signaling pathway. Our present study further demonstrated that the LCMV nucleoprotein (NP) blocks LCMV RNA- and other viral ligand-induced type I IFN responses.

Type I interferons (IFNs), namely, alpha interferon (IFN-␣) and IFN-␤, are not only essential for host innate defense against viral pathogens but also critically modulate the development of virus-specific adaptive immune responses (6, 8, 28, 30, 36, 50, 61). The importance of type I IFNs in host defense has been demonstrated by studying mice deficient in the type I IFN receptor, which are highly susceptible to most viral pathogens (2, 47, 62). Recent studies have suggested that the production of type I IFNs is controlled by different innate pattern recognition receptors (PRRs) (19, 32, 55, 60). There are three major classes of PRRs, including Toll-like receptors (TLRs) (3, 40), retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) (25, 48, 51), and nucleotide oligomerization domain (NOD)-like receptors (9, 22). TLRs are a group of transmembrane proteins expressed on either cell surfaces or endosomal compartments. RLRs localize in the cytosol. Both TLRs and RLRs are involved in detecting viral pathogens and controlling the production of type I IFNs (52, 60). In particular, the endosomelocalized TLRs (TLR3, TLR7/8, and TLR9) play important roles in detecting virus-derived double-stranded RNA (dsRNA), single-stranded RNA (ssRNA), and DNA-containing unmethylated CpG motifs, respectively. In contrast, RIG-I detects virus-derived ssRNA with 5⬘-triphosphates (5⬘-PPPs) or short dsRNA (⬍1 kb), whereas melanoma differentiation-associated gene 5 (MDA5) is responsible for recognizing virusderived long dsRNA as well as a synthetic mimic of viral dsRNA poly(I):poly(C) [poly(I 䡠 C)] (24, 60). Recognition of viral pathogen-associated molecular patterns (PAMPs) ulti-

MATERIALS AND METHODS Virus, cells, mice, and plasmids. LCMV-Armstrong (LCMV-Arm) strain 53b was kindly provided by Raymond Welsh and Liisa K. Selin (University of Massachusetts Medical School, MA) and was propagated on BHK-21 cells (ATCC) at a multiplicity of infection (MOI) of 0.01. Viral titers were determined with an immunological focus assay using antibody against LCMV NP VL4, kindly provided by Demetrius Moskophidis, Medical College of Georgia (5). The VSV-

* Corresponding author. Mailing address: Department of Medicine, University of Massachusetts Medical Center, 364 Plantation Street, Lazare Research Building, Worcester, MA 01605. Phone: (508) 8561886. Fax: (508) 856-6176. E-mail: [email protected]. 䌤 Published ahead of print on 30 June 2010. 9452

VOL. 84, 2010 Indiana serotype was used (58). Virus stocks were prepared on BHK-21 cells infected at a MOI of 0.01. Viral titers were determined by plaque assay on Vero cells (56, 65, 66). MDA5 knockout (KO) and MAVS KO mice were kindly provided by M. Colonna (Washington University) (15) and Z. J. Chen (UT Southwestern Medical Center) (54), respectively. IRF7 KO (19) and IRF3 KO (53) mice were kindly provided by Tadagatsu Taniguchi and Atsushi Yoshiki (Riken BioResource Center, Japan) and Michael David (University of California, San Diego), respectively. 3d mice were kindly provided by B. Beutler (The Scripps Research Institute) (59). All KO mice were backcrossed for more than six generations; MAVS KO, IRF7 KO, and IRF3 KO mice were backcrossed with C57BL/6 mice, and MDA5 KO mice were backcrossed with 129SI/Sv1mJ mice (WT mice were obtained from Jackson Laboratories). Mice were infected intraperitoneally (i.p.) with 2 ⫻ 105 PFU of LCMV-Arm clone 53b. Serum samples were collected at the indicated time points. A mutant RIG-I construct lacking the CARD domains functions as a dominant negative (DN) RIG-I (DN-RIG-C) was kindly provided by Takashi Fujita (Kyoto University, Kyoto, Japan). To construct plasmid-expressing V5/histidinetagged full-length LCMV nucleoprotein (NP), reverse transcription-PCR (RTPCR) was performed. Briefly, LCMV-Arm genomic RNA (LCMV RNA) was extracted using Trizol reagent (Invitrogen) as described below. First-strand cDNA was synthesized using the SuperScript III reverse transcriptase (Invitrogen) and primer GCCCAGAGGGTCTTAGAGTGTCACAAC (corresponding to LCMV small RNA positions 1627 to 1653) (GenBank accession no. M20869.1). To amplify the full-length cDNA encoding LCMV NP, the following primers were designed according to the sequence deposited in GenBank (accession no. M20869.1): forward primer (corresponding to LCMV small RNA positions 3318 to 3298), 5⬘-CTGGATCCCAAGATGTCCTTGTCTAAGGAAG-3⬘ (BamHI restriction site is underlined), and reverse primer (corresponding to LCMV small RNA positions 1627 to 1647), 5⬘-CGGCGATATCTTGCCCAGA GGGTCTTCGAGTGT-3⬘ (EcoRV restriction site is underlined). PCR was performed using GeneAmp PCR system 9700 to amplify the full-length cDNA encoding LCMV NP. PCR products, digested with both BamHI and EcoRV, were ligated directly into the pcDNA3-V5/His vector (Invitrogen) similarly digested with both BamHI and EcoRV, generating plasmid pcDNA3-V5/His-NP, also known as V5-NP. Plasmids were confirmed by restriction analysis and DNA sequencing (Tufts University Core Facility). To generate LCMV NP mutants (amino acid substitution for alanine) at amino acids D382 (D382A) and G385 (G385A), the parent plasmid encoding wild-type (WT) LCMV NP (pcDNA3-NP) and the QuikChange site-directed mutagenesis kit (Stratagene) were used by following the manufacturer’s recommendations. The mutations were verified by sequence analysis. The expression of both the LCMV WT and mutant NP was verified by flow cytometry using anti-LCMV NP antibody VL4 (5) and Western blotting using anti-V5 antibody (Invitrogen). Immortalized wild-type (WT) murine embryonic fibroblasts (MEFs), primary MDA5 KO MEFs, and their WT control MEFs were gifts from Wen-Chen Yeh (University of Toronto, Toronto, Canada) and Shizuo Akira (Osaka University, Japan), respectively. Primary WT, TLR7, 3d, and MyD88 KO MEFs were generated as previously described (29), and low-passage MEFs (⬍8 passages) were used. Preparation of GM-CSF or Flt3L-expanded bone marrow-derived DCs. Bone marrow-derived plasmacytoid dendritic cells (pDCs) and conventional dendritic cells (cDCs) were prepared as previously described (35). Briefly, to generate granulocyte-macrophage colony-stimulating factor (GM-CSF)-expanded cDCs, bone marrow cells were seeded at 2 ⫻ 106 cells per 100-mm dish in 10 ml of RPMI 1640 containing 100 U/ml recombinant mouse GM-CSF (rmGM-CSF; R&D Systems). Half of the medium was removed every 3 days of culture, and fresh culture medium with GM-CSF was added to the cultures. Nonadherent cells were collected at day 9 of incubation and were examined by staining with antibodies specific for CD11c and CD11b. Cells were seeded into 96-well plates at 2 ⫻ 104 per well in triplicate and were transfected with viral RNA or poly(I 䡠 C) or infected with Sendai virus (SeV) (80 hemagglutination [HA] units/ ml). The supernatants were harvested after incubation for 24 h, and the levels of type I IFNs were measured by the bioassay. For the preparation of Fms-related tyrosine kinase 3 ligand (Flt3L)-derived bone marrow pDCs, bone marrow cells were cultured with human recombinant Flt3L (10 ng/ml; R&D Systems). After 7 days, these cultures contained 40% CD11clo, CD11b⫺, B220⫹ pDCs. Nonadherent cells were collected and seeded into 96-well plates at 2 ⫻ 104 cells per well in triplicate. Cells were transfected with viral RNA, challenged with CpG1826 (6 ␮M/ml), or infected with SeV (80 HA units/ml). The supernatants were harvested after incubation for 24 h. The levels of type I IFNs in culture supernatants were measured using the type I IFN bioassay.

RNA AND TYPE I INTERFERONS

9453

Preparation of viral RNAs. LCMV or VSV virions were concentrated from the supernatants collected from virus-infected BHK-21 cells by precipitation with NaCl and polyethylene glycol (PEG) (46). Briefly, PEG 35000 and NaCl were added into virus-containing supernatants at final concentrations of 4% and 6% (wt/vol), respectively, and stirred overnight at 4°C, and then virions were pelleted by centrifugation at 12,000 rpm for 1 h. Pellets were resuspended with phosphate-buffered saline (PBS; pH 7.0) and centrifuged at 12,000 rpm for 1 h. The final viral pellets were resuspended in Trizol reagent (Invitrogen), and viral RNA was extracted and purified by following the manufacturer’s recommendations. RNA was dissolved in RNA storage solution (Ambion), and the concentration was determined by measuring the absorbance at 260 nm (A260) in a spectrophotometer. To verify the identity of LCMV RNA, RT-PCR using primers specific for LCMV-Arm glycoprotein (GP) was conducted by following the same strategy described above. The primers were designed according to the sequence deposited in GenBank (accession no. M20869.1). Primer (corresponding to LCMV small RNA position 3376R) 5⬘-CGCACAGTGGATCCTAGGCATTTG-3⬘ was used for cDNA synthesis. To amplify the full-length LCMV GP-encoding gene, the following primer sets were used: forward primer, 5⬘-GACAAGCTTAGGATG GGTCAGATTGTGACAATG-3⬘, and reverse primer, 5⬘-CTGCGATATCAGTC CGCGTCTTTTCCAGACGG-3⬘. PCR products were cloned into the pcDNA3V5/His vector by following the manufacturer’s recommendations (Invitrogen). Plasmid DNA containing full-length LCMV GP cDNAs was confirmed by both restriction enzyme digestion and sequencing (Tufts University Core Facility). RNA transfection and luciferase assay. HEK293 cells in 96-well plates were transfected with an IFN-␤ promoter-driven luciferase reporter plasmid (IFN␤-luc) using GeneJuice transfection reagent (Novagen). A total of 18 h after transfection, cells were transfected with LCMV RNA, VSV RNA, and poly(I 䡠 C) (TLR3 ligand; GE Healthcare) using Lipofectamine 2000 (Lipo; Invitrogen) or infected with SeV and incubated for an additional 18 h. Cell lysates were prepared, and the activities of the luciferase reporter were determined using the Steady-Glo luciferase assay system (Promega). Results were normalized to those of mock-treated cells and expressed as mean values of normalized data ⫾ standard errors of the means (SEM). To knock down the expression of RIG-I, a mutant RIG-I construct lacking CARD domain function as a dominant negative RIG-I (DN-RIG-C) (kindly provided by Takashi Fujita, Kyoto University, Kyoto, Japan) was used. HEK293 cells in 96-well plates were transfected with DN-RIG-C or control plasmid and incubated for 24 h. Cells were transfected with LCMV RNA, VSV RNA, or poly(I 䡠 C) or infected with SeV and incubated for an additional 18 h. Cell lysates were prepared, and the activities of the luciferase reporter were determined. To determine the ability of LCMV RNA to induce the bioactivity of type I IFNs, WT MEFs were seeded into 96-well plates in triplicate and transfected with LCMV RNA, VSV RNA, or poly(I 䡠 C) or infected with SeV. Cells were incubated for 18 h. The bioactivities of type I IFN in the supernatants were determined by type I IFN bioassay as previously described (66). Type I IFN bioassay. Levels of type I IFN in the supernatants or mouse serum samples were primarily determined using a type I IFN bioassay that measures the inhibition of the cytopathic effect of VSV on NCTC929 cells (20, 66). This assay measures all type I IFN species (IFN-␤ and all species of IFN-␣). Briefly, samples were prediluted by 1/5 (culture supernatants) or 1/50 (sera) and treated with the Stratalinker UV crosslinker for 1 J to inactivate any live virus in the samples. NCTC929 cells (kindly provided by Eva Szomolanyi-Tsuda, University of Massachusetts Medical School, Worcester, MA) were incubated with 2-fold diluted samples for 18 to 24 h at 37°C and were then challenged with VSV. The dilution of samples that resulted in a 50% protection of the VSV-induced cytopathic effect was defined as 1 U/ml of type I IFNs. Recombinant mouse IFN-␣ (PBL Biomedical Laboratories) was included as a positive control. To distinguish between IFN-␣ and IFN-␤, samples were treated with neutralizing antibodies against IFN-␣ (RMMA-1 clone; PBL Biomedical Laboratories) prior to performing the type I IFN bioassay. Immunoprecipitation of MDA5/RNA and RIG-I/RNA complexes from LCMVinfected cells. HEK293T cells were plated in 6-well plates at 2 ⫻ 105 cells/well and allowed to grow overnight. Cells were transfected with plasmids encoding either green fluorescent protein (GFP)-tagged RIG-I or Flag-tagged MDA5. After incubation for 24 h, cells were challenged with LCMV-Arm at an MOI of 0.5 and cultured for an additional 24 h. As a control, cells were transfected but left uninfected. Cell lysates were prepared using the passive lysis buffer (Promega) and subjected to an immunoprecipitation (IP) assay using anti-GFP antibody (Abcam) or anti-Flag antibody (Sigma) together with protein G plus agarose beads (Thermo Scientific) to pull down either RIG-I- or MDA5-containing complexes by following the manufacturer’s instructions. Rabbit IgG or mouse IgG1 isotype control antibodies were included in the IP assay to precipitate either RIG-I- or MDA5-containing complexes. After being washed thor-

9454

ZHOU ET AL.

J. VIROL.

oughly with PBS, the precipitates/agarose beads were subjected to RNA isolation using Trizol reagent (Invitrogen) as described above. The concentration of the extracted RNA was determined by measuring the A260 with a spectrophotometer. First-strand cDNA was synthesized using SuperScript III reverse transcriptase (Invitrogen) and an oligo(dT) primer by following the manufacturer’s recommendations. Full-length cDNA encoding LCMV GP was amplified as described above. Anti-LCMV NP pulldown assay and Western blotting. HEK293T cells were plated in 6-well plates at 2 ⫻ 105 cells/well and allowed to grow overnight. Cells were cotransfected with plasmids encoding V5-tagged WT or mutant LCMV NP, GFP-tagged RIG-I, or Flag-tagged MDA5. The expression of GFP is routinely utilized as a reporter molecule to monitor the efficiency of transfection. More than 50% of the cells are usually transfected and express both target proteins at 48 h postcotransfection. After incubation for 40 to 48 h, cell lysates were prepared using passive lysis buffer (Promega) and subjected to an IP assay using anti-V5-conjugated agarose beads (Sigma) to pull down the NP-containing protein complex by following the manufacturer’s instructions. After the final washing, 2⫻ sodium dodecyl sulfate (SDS) sample buffer (Sigma) was added and incubated for 5 min at 100°C. Anti-V5 bead-precipitated proteins were separated by 4 to 12% SDS-polyacrylamide gel electrophoresis (PAGE) and Westernblotted antibody to either GFP (indication of RIG-I) (Abcam) or Flag (indication of MDA5) (M2 clone; Sigma). The same membrane was stripped with Restore Western blotting stripping buffer (Thermo Scientific) and reprobed sequentially with anti-␤-actin (Abcam) and anti-V5 (Invitrogen) (indication of LCMV NP). Statistical analysis. Data were evaluated using the two-tailed Student t test. Results were expressed as means ⫾ standard deviations. P values of ⬍0.05 were regarded as significant.

RESULTS The RIG-I/MDA5-MAVS signaling pathway is involved in LCMV-induced type I IFNs in vivo. To determine whether the RLR-mediated signaling pathway is involved in LCMV-induced type I IFNs, we took advantage of recently developed knockout mice. MDA5 KO and MAVS KO mice were infected with LCMV-Arm. MAVS, also known as IPS-1/VISA/Cardif (27, 42, 64), is a crucial adaptor protein in the RIG-I/MDA5 signaling pathway. Serum levels of type I IFNs were measured using a type I IFN bioassay. Interestingly, we observed that both MDA5 KO (Fig. 1A) and MAVS KO (Fig. 1B) mice produced levels of type I IFNs at day 1 postinfection (p.i.; early phase of type I IFN responses) comparable to those of the WT controls; however, type I IFNs in the sera of both MDA5 and MAVS KO mice dropped to basal levels at day 2 p.i. (second wave of type I IFN responses) as opposed to WT mice, which maintained high levels at day 2 p.i. Recent studies have suggested that the production of type I IFNs is controlled by a group of transcriptional factors, including IRF3 and IRF7 (18, 19, 53). Both IRF3 and IRF7 are the downstream transcriptional factors important for both TLRand RLR-mediated type I IFN production (60). The role of these IRFs in regulating LCMV-induced type I IFN responses remains largely unexplored. To determine whether IRF3 and IRF7 are involved in this response, mice deficient in IRF3 and IRF7 were infected with LCMV-Arm intraperitoneally (i.p.). Both WT control and IRF3 KO mice produced type I IFNs in the sera following LCMV infection (Fig. 1C and D). Strikingly, type I IFNs were absent in IRF7 KO mice following LCMV infection. Of note, when infected with herpes simplex virus type 1 (HSV-1), IRF7 KO mice generated levels of type I IFN comparable to those of WT mice (our unpublished results). To determine whether the endosomal TLRs (TLR3, -7/8, and -9) are involved in LCMV-induced type I IFN production, we utilized a mouse strain (3d mice) with a point mutation

histidine-to-arginine substitution (H412R) in the polytopic membrane protein UNC-93B. UNC-93B is an endoplasmic reticulum protein involved in TLR3, TLR7/8, and TLR9 activation. It has been demonstrated that cells isolated from 3d mice had impaired cytokine production in response to TLR3, -7, -8, and -9 ligands (59). Our result demonstrated that TLR3/ 7/8/9 (3d) do not contribute to LCMV-induced type I IFN production in vivo (Fig. 1E). This is consistent with a recent report (21). To determine whether the RIG-I/MDA5-MAVS signaling pathway affects the production of either IFN-␣ or -␤, serum samples obtained from both MAVS KO and WT mice (day 1 p.i.) were pretreated with a neutralizing rat monoclonal antibody (MAb) against mouse IFN-␣ (RMMA-1 clone; PBL Biomedical Laboratories) for 1 h at 37°C before incubation with NCTC929 cells (65). Type I IFN activity in the sera of WT mice was completely blocked by treatment with neutralizing anti-IFN-␣ antibody (Fig. 1F). In contrast, treatment with neutralizing anti-IFN-␣ only partly blocked the activity of type I IFNs in sera from MAVS KO mice. These results suggest that the RNA helicase signaling pathway does not play a major role in the early phase of type I IFN production, which is predominately IFN-␤ (data not shown). These results are consistent with a recent report (21). Taken together, these results demonstrate that the RLRmediated signaling pathway plays a critical role in regulating the late phase (day 2 p.i.) of type I IFN responses to LCMV infection. LCMV RNA induces a type I IFN response. The mechanisms underlying how LCMV infection induces type I IFNs in the mouse are poorly defined. In particular, it is unknown which LCMV component is responsible for activation of type I IFN responses. The helicases RIG-I and MDA5 are known to detect virus-derived dsRNA or ssRNA and activate type I IFN responses. We first determined whether LCMV RNA alone could induce type I IFN responses. We took advantage of an IFN-␤ promoter-driven luciferase reporter plasmid (IFN-␤-luc). LCMV RNA was extracted from PEG-precipitated LCMV virions. When LCMV RNA was transfected into IFN-␤-luc-expressing HEK cells, LCMV RNA induced IFN-␤-luc activity (Fig. 2A). As controls, cells were also similarly transfected with either VSV RNA or poly(I 䡠 C) or infected with SeV (Fig. 2C). These three ligands have been demonstrated to induce a type I IFN response via either RIG-I (VSV RNA and SeV)- or MDA5 [poly(I 䡠 C)]-dependent signaling pathways (25, 60). Consistent with the published results, negative-control cellular RNA extracted from BHK-21 cells did not induce a type I IFN response (Fig. 2A). The BHK-21 cell line is the parent cell line used to propagate LCMV. Interestingly, challenge with live LCMV did not induce a type I IFN response, in contrast to isolated LCMV RNA (Fig. 2A). These results demonstrate that LCMV RNA itself (without other viral components) is capable of inducing the type I IFN response but suggest that some other LCMV components must turn off the type I IFN response. To further test how LCMV RNA induces type I IFNs, LCMV RNA was transfected into WT MEFs. LCMV RNA induced high levels of type I IFNs in MEFs (Fig. 2D). As positive controls, VSV RNA, SeV, and poly(I 䡠 C) all induced type I IFN production in WT MEFs (Fig. 2E). These studies

VOL. 84, 2010

RNA AND TYPE I INTERFERONS

9455

FIG. 1. LCMV-induced type I IFN production in vivo is dependent on the MDA5-MAVS signaling pathway. (A and B) WT mice and mice deficient in MDA5 (A) and MAVS (B) were infected with LCMV-Arm i.p. Serum samples were collected at days 1 and 2 postinfection (p.i.), and type I IFN levels were determined by bioassay. (C to E) IRF3 and IRF7 (C and D) and 3d (E) were similarly infected with LCMV-Arm i.p. Serum samples were collected at days 1 and 2 p.i., and type I IFN levels were determined by bioassay. (F) To distinguish IFN-␣ from IFN-␤, serum samples obtained from LCMV-infected WT and MAVS KO mice (day 1 p.i.) were treated with the neutralizing anti-IFN-␣ antibody prior to culture with NCTC929 cells to determine the bioactivity of type I IFN. ⴱ, P ⬍ 0.05.

demonstrate the ability of the isolated LCMV RNA to induce strong type I IFN production in multiple cell types. To define the structural basis of the LCMV RNA responsible for the activation of the type I IFN response, LCMV RNA was treated with either the dsRNA-specific RNase III, which cleaves dsRNA, or shrimp alkaline phosphatase (SAP), which efficiently catalyzes the release of 5⬘- and 3⬘-triphosphate groups from RNA. Our results demonstrated that treatment with either RNase III or SAP ablated LCMV RNA-induced activation of the type I IFN response in either the IFN-␤-luc reporter assay in HEK cells or the type I IFN bioassay in MEFs (Fig. 2B and D). Consistent with the published results (25, 51), treatment of VSV RNA with either RNase III or SAP abolished or significantly reduced its ability to induce type I IFN responses in either HEK293 cells or MEFs (Fig. 2C and E).

Poly(I 䡠 C), a synthetic dsRNA resembling the RNA of infectious viruses, does not contain the 5⬘-triphosphate structures. Thus, treatment of poly(I 䡠 C) with SAP did not affect its ability to induce the type I IFN response in either HEK293 cells or MEFs, whereas treatment with RNase III completely abolished its ability to induce the type I IFN response (Fig. 2C and E). Thus, these results suggest that both dsRNA and 5⬘triphosphate structures are involved in LCMV RNA-induced type I IFN responses. To further explore the mechanism by which LCMV RNA induces type I IFN responses, we asked whether LCMV RNA associates directly with either MDA5 or RIG-I. We took advantage of the overexpression system to transfect HEK293T cells with plasmids expressing either Flag-tagged MDA5 or GFP-tagged RIG-I, followed by challenge with LCMV-Arm.

9456

ZHOU ET AL.

J. VIROL.

FIG. 2. LCMV RNA activates the type I IFN response and is inactivated by both RNase III and SAP. (A) HEK293 cells were transfected with IFN-␤-luc. At 18 h after transfection, cells were transfected with Lipo as a control, LCMV RNA, or cellular RNA isolated from BHK-21 cells or challenged with LCMV-Arm. Cells were incubated for an additional 18 h. Cell lysates were prepared, and luciferase activity was measured. (B) LCMV RNA extracted from LCMV-Arm was treated with either RNase III or SAP. HEK293 cells were transfected with IFN-␤-luc for 18 h. Cells were then transfected with untreated (⫺) or treated LCMV RNA. Cell lysates were prepared, and luciferase activity was measured. (C) HEK293 cells were similarly cotransfected with IFN-␤-luc and untreated or treated VSV RNA or poly(I 䡠 C) (poly IC). Cells were also infected with SeV, and luciferase activity was measured 18 h later. (D) WT MEFs (immortalized) were transfected with untreated or treated LCMV RNA. At 18 h posttransfection, the bioactivity of type I IFNs in the supernatants was determined by type I IFN bioassay. (E) WT MEFs were transfected with untreated or treated VSV RNA or poly(I 䡠 C). Cells were infected with SeV at 18 h posttransfection/infection, and the bioactivity of type I IFNs in the supernatants was determined by type I IFN bioassay. (F) HEK293T cells in 6-well plates were transfected with plasmids expressing either Flag-tagged MDA5 or GFP-tagged RIG-I. After 24 h, cells were challenged with LCMV-Arm (MOI of 0.5). After culture for an additional 24 h, cell lysates were prepared. MDA5- or RIG-I-containing protein complexes were precipitated using either anti-Flag antibody (MDA5) or anti-GFP antibody (RIG-I) together with the protein G agarose beads. Total RNA were extracted from these precipitated protein complexes. The presence of LCMV RNA in both MDA5- and RIG-I-containing complexes was confirmed by RT-PCR to amplify the LCMV GP gene (1.5 kb). PCR products were analyzed in a 1% agarose gel (with ethidium bromide staining). ⴱ, P ⬍ 0.05.

MDA5- or RIG-I-containing protein complexes were precipitated using protein G-conjugated agarose beads. Total RNA was isolated from these precipitated protein complexes. The presence of LCMV RNA in both MDA5- and RIG-I-contain-

ing complexes was confirmed by RT-PCR to amplify the LCMV glycoprotein (GP) gene (Fig. 2F). These results demonstrated that LCMV RNA is physically associated with both MDA5 and RIG-I in LCMV-infected cells. This physical in-

VOL. 84, 2010

RNA AND TYPE I INTERFERONS

9457

FIG. 3. Both RIG-I and MDA5 are involved in the LCMV RNA-induced type I IFN response. Endosomal TLRs and MyD88 signaling pathways are not involved in the LCMV RNA-induced type I IFN response in fibroblasts. (A) IFN-␤-luc-expressing HEK293 cells in 96-well plates were transfected with either DN-RIG-C or control plasmid pcDNA3 and incubated for 18 h. The next day, cells were transfected with LCMV RNA or other controls. After incubation for an additional 18 h, cell lysates were prepared, and the luciferase reporter activities were measured. (B) MDA5 and WT MEFs in 96-well plates were transfected with LCMV RNA or other controls and incubated for 18 h. The bioactivities of the type I IFNs in the supernatants were measured using a type I IFN bioassay. (C) MEFs in 96-well plates were transfected with LCMV RNA for 18 h. The bioactivities of the type I IFNs in the supernatants were measured using a type I IFN bioassay. Data are representative of results from at least three experiments. ⴱ, P ⬍ 0.05.

teraction provides the molecular basis to trigger type I IFN responses through both the MDA5- and RIG-I-mediated signaling pathways. LCMV RNA activates the RIG-I/MDA5-MAVS-dependent signaling pathway to induce type I IFN responses. To determine whether the RIG-I-mediated signaling pathway is involved in recognizing LCMV RNA and triggering IFN production, we used a mutant RIG-I lacking CARD domain function as a dominant negative (DN) RIG-I (DN-RIG-C). DN-RIG-C is known to be able to specifically block RIG-I-mediated type I IFN responses. HEK293 cells were first cotransfected with IFN-␤-luc and DN-RIG-C plasmids, followed by transfection with LCMV RNA or other controls. Transfection of the dominant negative construct DN-RIG-C significantly reduced the LCMV RNA-induced type I IFN response (Fig. 3A). Consistent with published data, DN-RIG-C also significantly inhibited both VSV RNA- and SeV-induced activation of type I IFN responses but did not affect poly(I 䡠 C)-induced IFN responses (Fig. 3A), because poly(I 䡠 C) activates type I IFN responses through MDA5 rather than through RIG-I. These results suggest that the RIG-I-mediated signaling pathway is involved in LCMV RNA-induced IFN responses. To define a role for other RLR proteins in the LCMV RNA-induced type I IFN response, experiments with MDA5 KO MEFs were carried out. Levels of type I IFNs in the supernatants were measured using a bioassay. LCMV RNA induced significantly lower levels of type I IFNs in MDA5 KO MEFs than in WT MEFs (Fig. 3B). This is consistent with the patterns of LCMV-induced type I IFN production in MDA5 KO mice (Fig. 1A), suggesting that the MDA5-mediated rec-

ognition mechanism is also involved in the LCMV RNA-induced type I IFN response. Although our in vivo studies of mice and in vitro results using the HEK cell system have suggested that a TLR-MyD88independent signaling pathway is involved in LCMV RNAinduced type I IFN responses, we wished to further determine whether the endosomal TLRs (TLR3, -7/8, and -9) are involved in LCMV RNA-induced type I IFN production. We utilized MEFs generated from a mouse strain with a point mutation histidine-to-arginine substitution (H412R) in the polytopic membrane protein UNC-93B (3d mice). 3d, TLR7, MyD88 KO, and WT MEFs were similarly plated and transfected with LCMV RNA or controls. LCMV RNA induced levels of type I IFN in MEFs deficient in TLR7, MyD88, or 3d comparable to those of WT MEFs (Fig. 3C). These results are consistent with the patterns of LCMV-induced type I IFN responses in mice deficient in 3d, TLR7, and MyD88 (Fig. 1E) (66). Taken together, these results demonstrate that neither the endosomal TLR signaling pathway nor MyD88 play a significant role in the LCMV RNA-induced type I IFN response. In contrast, the RIG-I/MDA5/MAVS-mediated signaling pathway is essential for the LCMV RNA-induced type I IFN response. LCMV RNA induces type I IFNs in both pDCs and cDCs, and the MAVS-dependent signaling pathway is essential for type I IFN production from cDCs. Although pDCs have been demonstrated to be the major type I IFN producer in response to selected viral infections, the role of pDCs during LCMV infection is unclear. To define whether LCMV RNA could induce type I IFN production from DCs, bone marrow-derived

9458

ZHOU ET AL.

J. VIROL.

FIG. 4. LCMV RNA can induce type I IFN in both pDCs and cDCs, and LCMV RNA-induced type I IFN in cDCs is MAVS dependent. Bone marrow-derived pDCs (A) or cDCs (B) were plated in 96-well plates and transfected with purified LCMV RNA or control poly(I 䡠 C) or stimulated with CpG1826. After incubation for 18 h, the bioactivities of the type I IFNs in the supernatants were measured using a type I IFN bioassay. Data are representative of results from two experiments. ⴱ, P ⬍ 0.05.

pDCs and cDCs were prepared. pDCs were defined as CD11clo B220⫹ Gr-1⫹ CD11b⫺, and cDCs were defined as CD11chi B220⫺ Gr-1⫺ CD11b⫹. The purity of both pDCs and cDCs was about 70% based on fluorescence-activated cell sorter (FACS) staining (data not shown). Interestingly, LCMV RNA induced comparable levels of type I IFNs in WT, MyD88 KO, and MAVS KO pDCs (Fig. 4A). As a control, CpG induced type I IFNs in both WT and MAVS KO pDCs but not in MyD88 KO pDCs. In contrast, LCMV RNA did not induce type I IFNs in MAVS KO cDCs but did induce similar high levels of type I IFNs in both WT and MyD88 KO cDCs (Fig. 4B). As a control, transfected poly(I 䡠 C) induced levels of type I IFNs in MAVS KO cDCs that were significantly lower than those in WT cDCs (Fig. 4B). Collectively, these results suggest that a MAVS-mediated signaling pathway is essential for the production of type I IFNs in response to LCMV RNA in cDCs. However, it seems that a signaling pathway independent of both MyD88 and MAVS is involved in the production of type I IFNs in pDCs. LCMV NP targets both RIG-I and MDA5 to inhibit the type I IFN response. We have observed that live LCMV (LCMVArm strain) inhibits LCMV RNA-induced activation of type I IFN (data not shown). Additionally, studies from other groups have shown that LCMV NP inhibits the type I IFN response through targeting the transcriptional factor IRF3, although the details and molecular mechanisms of this inhibition have not been fully defined (37–39). IRF3 is a common downstream transcriptional factor in both the TLR- and RLR-mediated signaling pathways. Since we have demonstrated that LCMV RNA is responsible for triggering a RIG-I/MDA5/MAVS-dependent type I IFN response, we wanted to define whether LCMV NP could target RIG-I/MDA5 or MAVS signaling molecules. To determine the role of LCMV NP in regulating the type I IFN signaling pathway, HEK293T cells were transfected with a plasmid encoding LCMV-Arm NP, followed by transfection with either LCMV RNA (Fig. 5A) or poly(I 䡠 C) (indicator of activation of the MDA5-dependent signaling pathway) (Fig. 5C). Cells were also infected with SeV as a control for RIG-I signaling (Fig. 5B). Expression of LCMV NP was measured by flow cytometric analysis (data not shown). Interestingly, compared to cells transfected with control plasmid pcDNA3, LCMV NP expression dramatically inhibited both LCMV RNA- and SeV-induced activation of the type I

IFN response and modestly affected poly(I 䡠 C)-induced activation (Fig. 5). We have also observed that LCMV infection in HEK cells either failed to induce type I IFN responses (Fig. 2A) or strongly inhibited LCMV RNA- and SeV-induced type I IFN responses (data not shown). Therefore, these results suggest that LCMV NP could target both the RIG-I and MDA5 signaling pathways to inhibit the type I IFN response. Martinez-Sobrido et al. have recently demonstrated that a region in LCMV NP from residues 382 to 386 is critical for LCMV NP-mediated inhibition of type I IFN responses (37). To further study the molecular mechanism by which LCMV NP inhibits type I IFN responses, we generated plasmids encoding LCMV NP mutants carrying either the D382A or G385A mutation. These two mutations did not affect the expression of LCMV NP in HEK cells (data not shown). Interestingly, when cotransfected into HEK cells with IFN-␤-luc

FIG. 5. LCMV NP expression dramatically inhibits the LCMV RNA-induced type I IFN response as well as both the SeV- and poly(I 䡠 C)-induced type I IFN responses. HEK293T cells in 96-well plates were transfected with plasmids encoding either WT NP, D382A mutant NP, or G385A mutant NP together with IFN-␤-luc for 18 h, followed by transfection with LCMV RNA (A), infection with SeV (B), or transfection with poly(I 䡠 C) (C) for an additional 18 h. The activity of luciferase was measured. Data are representative of results from at least four experiments. ⴱ, P ⬍ 0.05.

VOL. 84, 2010

RNA AND TYPE I INTERFERONS

9459

FIG. 6. LCMV NP associates with both RIG-I and MDA5. (A and B) HEK293T cells were transfected with plasmid encoding either WT NP, D382A mutant NP, or G385A mutant NP together with either GFP–RIG-I (A) or Flag-MDA5 (B). At 40 h posttransfection, cell lysates were prepared and subjected to immunoprecipitation (IP) with an anti-V5 antibody (LCMV NP). Eluted proteins were subjected to 4 to 12% SDS-PAGE and immunoblot analysis (IB) with antibodies against V5 (top), GFP (RIG-I) or Flag (MDA5) (middle), and ␤ actin (bottom). Data are representative of results from at least three experiments.

and challenged with either LCMV RNA, SeV, or poly(I 䡠 C), both mutants failed to inhibit activation of IFN-␤-luc induced by LCMV RNA (Fig. 5A), SeV (Fig. 5B), or poly(I 䡠 C) (Fig. 5C), despite the marked inhibition by wild-type LCMV NP. These results are consistent with those of Martinez-Sobrido et al. (38), demonstrate that LCMV NP inhibits both RIG-I and MDA5 pathways, and confirm that both the D382 and G385 residues are critically involved in blocking LCMV RNA- and other virus ligand-induced activation of IFN-␤. LCMV NP physically associates with both RIG-I and MDA5. To define the molecular mechanisms involved in LCMV NP inhibition of the type I IFN response, we asked whether LCMV NP could associate with either RIG-I or MDA5 signaling molecules. Cell lysates prepared from untransfected and GFP–RIG-I/V5-NP (wild-type) plasmid-cotransfected HEK293T cells were subjected to a pulldown assay with anti-V5 antibody (LCMV NP). LCMV NP-associated protein complexes were analyzed by SDS-PAGE and immunoblotted with anti-GFP antibody. Our results demonstrated that an anti-V5 antibody could pull down RIG-I (Fig. 6A). These data indicate that LCMV NP physically associates with RIG-I. The same approach was utilized to define whether LCMV NP also interacts with MDA5. Cell lysates were prepared from FlagMDA5- and V5-NP-cotransfected HEK293T cells. Anti-V5 antibody pulled down MDA5 (Fig. 6B), demonstrating that LCMV NP also associates with MDA5. We further asked whether the association of LCMV NP with either RIG-I or MDA5 explained the inhibitory effect of LCMV NP on IFN induction. Using a similar experimental design, HEK293T cells were transfected with either GFP– RIG-I or Flag-MDA5 together with NP mutants. Interestingly, these two mutations (D382A and G385A in the LCMV NP gene) did not affect the interaction of LCMV NP with either

RIG-I or MDA5 (Fig. 6A and B), although changing these two amino acids in LCMV NP eliminated the ability of this protein to inhibit type I IFN production. DISCUSSION Production of type I IFNs is a critical part of the innate immune response to viral pathogens. Induction of type I IFNs following virus infection is largely controlled by the innate PRRs, which are proteins expressed in cells of the innate immune system. LCMV infection in the mouse, a commonly used model for studying host responses to viruses, has been extensively characterized. However, the molecular mechanisms underlying how LCMV infection induces type I IFNs are poorly defined. In the present study, we have examined the role of LCMV RNA and NP in the activation of the type I IFN response. We demonstrated that LCMV RNA is capable of activating the type I IFN response. Moreover, both dsRNA and 5⬘-triphosphated ssRNA derived from LCMV genomic RNA are responsible for type I IFN production. Second, the RIG-I/MDA5-MAVS-mediated signaling pathway is essential for the LCMV RNA-induced type I IFN responses both in vitro and in vivo. We provided evidence that LCMV RNA is physically associated with both MDA5 and RIG-I. Third, using bone marrow-derived DCs, we demonstrated that both pDCs and cDCs are able to produce type I IFNs in response to LCMV RNA. Finally, we demonstrated that LCMV NP blocks type I IFN induction. In the present study, we characterized the innate immune recognition mechanisms involved in detecting LCMV RNA and triggering the type I IFN response. In LCMV-infected cells, LCMV RNA is associated with both MDA5 and RIG-I (Fig. 2F). This association could account for the activation of

9460

ZHOU ET AL.

type I IFN responses. Rehwinkel et al. recently demonstrated that RNAs bearing 5⬘-PPPs derived from single-stranded RNA viruses, like influenza virus, trigger type I IFN induction through association with RIG-I (51a). The cellular source of type I IFNs during LCMV infection is not clear (7, 8, 31). Multiple types of cells might be involved in LCMV-induced type I IFN responses, including spleen marginal zone (MZ) macrophages (34) and DCs (cDCs and pDCs) (7, 8, 31, 44). Depletion of MZ macrophages with clodronate liposomes prior to infection with LCMV in the mouse resulted in dramatic loss of type I IFN production (34). DCs, particularly pDCs, have been demonstrated to play a critical role in producing type I IFN against viral pathogens. Recent studies have also demonstrated that LCMV infection rapidly induces the activation of DCs in the spleen, and this activation is dependent on type I IFN signaling, suggesting a role for both cDCs and pDCs in type I IFN production during LCMV infection. To determine whether both pDCs and cDCs are capable of producing type I IFNs in response to LCMV, we took advantage of bone marrow-derived DCs. It has been demonstrated that cDCs and pDCs utilize distinctive mechanisms to detect viral RNA and to generate type I IFN production. cDCs and other types of cells predominantly use the MAVS-mediated signaling pathway to generate type I IFNs (23, 25, 57). In contrast, expression of TLR7 and TLR9 in pDCs is responsible for ssRNA- and unmethylated CpG motif-induced type I IFN production, respectively (26). In the present study, we demonstrated that LCMV RNA-induced type I IFNs in cDCs are dependent on the MAVS signaling pathway. This is consistent with previous studies. The signaling pathway operating in pDCs responsible for LCMV RNA-induced type I IFN production remains undefined. The TLR family plays a critical role in regulating the type I IFN response as well as other inflammatory cytokine and chemokine responses to viral pathogens. We have previously observed that TLR2 KO mice had a defective type I IFN response during an acute LCMV infection. The mechanism has not been defined. In the present study, we demonstrated that following LCMV infection, wild-type mice had a greater increase in the mRNA levels of genes involved in type I IFN production, including IFN-␣ (IFN-␣4, non-IFN-␣4), IFN-␤, and MDA5, than did TLR2 knockout mice (data not shown). Given that the expression of MDA5 and RIG-I is type I IFN inducible, it is conceivable that TLR2 and other unknown molecules could regulate the expression of MDA5 and RIG-I in certain types of cells through either activation of NF-␬B signals or other unidentified mechanisms. Of note, the role of TLR2 in virus-induced type I IFN production in certain types of cells has also recently been observed by another group (4). The MDA5/RIG-I-MAVS signaling pathway in vivo is critical for the LCMV-induced late phase (day 2 p.i.) of type I IFN responses, while the MDA5/RIG-I-MAVS signaling pathway is dispensable for the LCMV-induced early wave (day 1 p.i.) of type I IFN responses. Mice deficient in either MDA5 or MAVS generated comparable levels of type I IFN in the sera at day 1 p.i., while the type I IFN responses in these mice collapsed on day 2 p.i. It has been reported that MDA5 KO mice are capable of generating type I IFN responses against a number of viruses, including influenza virus (NS1 mutant), SeV, Newcastle disease virus (NDV), and Japanese encepha-

J. VIROL.

FIG. 7. LCMV genomic RNA and NP differentially regulate type I IFN responses. LCMV-derived dsRNA and/or 5⬘-PPP structure activates both the RIG-I- and MDA5-dependent type I IFN responses, whereas LCMV NP inhibits LCMV RNA-induced type I IFN responses through modulating the RIG-I/MDA5-MAVS signaling pathway.

litis virus (JEV) (25). MAVS KO mice have also been demonstrated to be able to produce levels of type I IFN comparable to those of wild-type mice in response to VSV, although these mice are sensitive to VSV infection (57). Interestingly, a recent report demonstrated that in response to SeV infection, while MDA5 KO mice generated comparable levels of type I IFN mRNA at day 2 p.i., the levels of type I IFN at day 5 p.i. were dramatically decreased in MDA5 KO mice (16). Like other viruses, LCMV has also evolved countermechanisms to modulate PRRs and other innate signaling molecules to block type I IFN induction. It has been demonstrated that LCMV NP blocks the type I IFN response by targeting IRF3 (39). IRF3 and IRF7 are the common downstream transcriptional factors involved in both the TLR- and RLR-mediated signaling pathways, and IRF3 has been demonstrated to be a potent transcription factor for activation of the IFN-␤ promoter (19, 60). In the present study, we confirmed and extended these findings by demonstrating that LCMV NP blocks type I IFN production through targeting both the MDA5- and RIG-I mediated signaling pathways. Our experiments demonstrate that LCMV NP physically associates with both RIG-I and MDA5 (Fig. 6). It is not clear whether this association is responsible for the inhibitory effect of NP on the activation of type I IFN responses. The fact that mutant NPs that do not suppress the IFN response still associated with RIG-I and MDA5 indicates that simple association does not lead directly to inhibition. It is possible that the site of association is important or that the interaction of the two proteins leads to changes in the tertiary structure of these proteins or their interactions with additional proteins in the pathway. Other possible mechanisms responsible for the inhibitory effect of NP might include the ability of LCMV NP to bind and sequester viral RNA (14, 17, 43, 51), to degrade the host signaling molecules involved in detection of viral RNA (12, 33, 42), or to inhibit the positive regulatory loop of the RIG-I/MAVS signaling pathway (13). Our data do suggest that these two residues (D382 and G385) in LCMV NP are involved in the inhibitory effect of NP but are not required for the interaction with either RIG-I or MDA5. The fact that LCMV NP-expressing cells are still able to respond to poly(I 䡠 C) suggests that different ligands may interact with the RIG-I/MDA5 pathways in different ways. Further experiments involving more specific localization of bind-

VOL. 84, 2010

RNA AND TYPE I INTERFERONS

ing and analysis of multiple protein complexes may be necessary to define the mechanism of the inhibitory action of NP. Interestingly, Lee et al. have recently observed a similar phenomenon. They noted that mice chronically infected with LCMV had defective type I IFN responses to SeV but were capable of producing type I IFNs in response to other stimuli, including poly(I 䡠 C), albeit in a reduced manner compared to that of uninfected mice (31). In summary, our current studies define a mechanism by which LCMV infection regulates the type I IFN response. Our current understanding of the recognition and regulation of the type I IFN response to LCMV indicates that this response is dependent upon IRF7 but independent of IRF3. There is no role for the RNA-sensing TLRs (neither TLR3 nor TLR7 seems to be important in the response to LCMV). While the “second wave” (day 2 p.i.) of the IFN response to LCMV is dependent on both MAVS and MDA5, the earliest response (day 1 p.i.) is independent of the RIG-I/MDA5 MAVS pathway, and both are independent of the endosomal TLRs (TLR3, -7, and -9). Our current model indicates that expression of the LCMV NP turns off the induction of type I IFN that is initiated by LCMV RNA interaction with RIG-I and MDA5 (Fig. 7). We believe that there are dynamic interactions between LCMV RNA, LCMV NP, and the host type I IFN modulators, including MDA5 and RIG-I. These interactions could determine the pattern of type I IFN responses as well as viral pathogenesis during LCMV infection. These results will aid in understanding LCMV viral pathogenesis and the development of new strategies to treat human arenavirus infections as well as other viral diseases. ACKNOWLEDGMENTS We are grateful to M. Colonna (Washington University) and Z. J. Chen (UT Southwestern Medical Center) for generously providing breeding pairs for MDA5 KO and MAVS KO mice, respectively. We thank T. Taniguchi and A. Yoshiki (Riken BioResource Center, Japan) and M. David (University of California, San Diego) for generously providing breeding pairs for IRF7 KO and IRF3 KO mice, respectively. We thank B. Beutler (The Scripps Research Institute) for kindly providing the breeding pair for 3d mice. We thank S. Akira (Osaka University, Japan) for generously providing the MDA5 KO MEFS. We thank Raymond Welsh for helpful discussions and advice. We thank Jennifer J. P. Wang for critically reading the manuscript and for helpful discussions. Thanks to Jayashree Paranjape for editing the figures and reading the manuscript. This work was supported by NIAID Regional Center of Excellence grant AI 057159, NIH grants R01 AI 49309, P01 AI 0577484, JDRF 24-2008-950, and P01 AI083215-01 (to R.W.F.), and NIH grant AI 51405 (to E.A.K.-J.).

7.

8.

9. 10. 11.

12.

13.

14.

15.

16.

17.

18. 19.

20. 21.

22.

23.

24.

REFERENCES 1. Ahmed, R., A. Salmi, L. D. Butler, J. M. Chiller, and M. B. Oldstone. 1984. Selection of genetic variants of lymphocytic choriomeningitis virus in spleens of persistently infected mice. Role in suppression of cytotoxic T lymphocyte response and viral persistence. J. Exp. Med. 160:521–540. 2. Akira, S. 2001. Toll-like receptors and innate immunity. Adv. Immunol. 78:1–56. 3. Akira, S., and K. Takeda. 2004. Toll-like receptor signalling. Nat. Rev. Immunol. 4:499–511. 4. Barbalat, R., L. Lau, R. M. Locksley, and G. M. Barton. 2009. Toll-like receptor 2 on inflammatory monocytes induces type I interferon in response to viral but not bacterial ligands. Nat. Immunol. 10:1200–1207. 5. Battegay, M., S. Cooper, A. Althage, J. Banziger, H. Hengartner, and R. M. Zinkernagel. 1991. Quantification of lymphocytic choriomeningitis virus with an immunological focus assay in 24- or 96-well plates. J. Virol. Methods 33:191–198. 6. Cousens, L. P., R. Peterson, S. Hsu, A. Dorner, J. D. Altman, R. Ahmed, and

25.

26. 27.

28.

29.

9461

C. A. Biron. 1999. Two roads diverged: interferon alpha/beta- and interleukin 12-mediated pathways in promoting T cell interferon gamma responses during viral infection. J. Exp. Med. 189:1315–1328. Dalod, M., T. P. Salazar-Mather, L. Malmgaard, C. Lewis, C. AsselinPaturel, F. Briere, G. Trinchieri, and C. A. Biron. 2002. Interferon alpha/ beta and interleukin 12 responses to viral infections: pathways regulating dendritic cell cytokine expression in vivo. J. Exp. Med. 195:517–528. Diebold, S. S., M. Montoya, H. Unger, L. Alexopoulou, P. Roy, L. E. Haswell, A. Al-Shamkhani, R. Flavell, P. Borrow, and C. Reis e Sousa. 2003. Viral infection switches non-plasmacytoid dendritic cells into high interferon producers. Nature 424:324–328. Eisenbarth, S. C., and R. A. Flavell. 2009. Innate instruction of adaptive immunity revisited: the inflammasome. EMBO Mol. Med. 1:92–98. Finberg, R. W., J. P. Wang, and E. A. Kurt-Jones. 2007. Toll like receptors and viruses. Rev. Med. Virol. 17:35–43. Fitzgerald, K. A., S. M. McWhirter, K. L. Faia, D. C. Rowe, E. Latz, D. T. Golenbock, A. J. Coyle, S. M. Liao, and T. Maniatis. 2003. IKKepsilon and TBK1 are essential components of the IRF3 signaling pathway. Nat. Immunol. 4:491–496. Foy, E., K. Li, R. Sumpter, Jr., Y. M. Loo, C. L. Johnson, C. Wang, P. M. Fish, M. Yoneyama, T. Fujita, S. M. Lemon, and M. Gale, Jr. 2005. Control of antiviral defenses through hepatitis C virus disruption of retinoic acidinducible gene-I signaling. Proc. Natl. Acad. Sci. U. S. A. 102:2986–2991. Gack, M. U., Y. C. Shin, C. H. Joo, T. Urano, C. Liang, L. Sun, O. Takeuchi, S. Akira, Z. Chen, S. Inoue, and J. U. Jung. 2007. TRIM25 RING-finger E3 ubiquitin ligase is essential for RIG-I-mediated antiviral activity. Nature 446:916–920. Garcia-Sastre, A. 2001. Inhibition of interferon-mediated antiviral responses by influenza A viruses and other negative-strand RNA viruses. Virology 279:375–384. Gitlin, L., W. Barchet, S. Gilfillan, M. Cella, B. Beutler, R. A. Flavell, M. S. Diamond, and M. Colonna. 2006. Essential role of mda-5 in type I IFN responses to polyriboinosinic:polyribocytidylic acid and encephalomyocarditis picornavirus. Proc. Natl. Acad. Sci. U. S. A. 103:8459–8464. Gitlin, L., L. Benoit, C. Song, M. Cella, S. Gilfillan, M. J. Holtzman, and M. Colonna. 2010. Melanoma differentiation-associated gene 5 (MDA5) is involved in the innate immune response to Paramyxoviridae infection in vivo. PLoS Pathog. 6:e1000734. Guo, Z., L. M. Chen, H. Zeng, J. A. Gomez, J. Plowden, T. Fujita, J. M. Katz, R. O. Donis, and S. Sambhara. 2007. NS1 protein of influenza A virus inhibits the function of intracytoplasmic pathogen sensor, RIG-I. Am. J. Respir. Cell Mol. Biol. 36:263–269. Hiscott, J. 2007. Triggering the innate antiviral response through IRF-3 activation. J. Biol. Chem. 282:15325–15329. Honda, K., H. Yanai, H. Negishi, M. Asagiri, M. Sato, T. Mizutani, N. Shimada, Y. Ohba, A. Takaoka, N. Yoshida, and T. Taniguchi. 2005. IRF-7 is the master regulator of type-I interferon-dependent immune responses. Nature 434:772–777. Ishikawa, R., and C. A. Biron. 1993. IFN induction and associated changes in splenic leukocyte distribution. J. Immunol. 150:3713–3727. Jung, A., H. Kato, Y. Kumagai, H. Kumar, T. Kawai, O. Takeuchi, and S. Akira. 2008. Lymphocytoid choriomeningitis virus activates plasmacytoid dendritic cells and induces a cytotoxic T-cell response via MyD88. J. Virol. 82:196–206. Kanneganti, T. D., M. Body-Malapel, A. Amer, J. H. Park, J. Whitfield, L. Franchi, Z. F. Taraporewala, D. Miller, J. T. Patton, N. Inohara, and G. Nunez. 2006. Critical role for Cryopyrin/Nalp3 in activation of caspase-1 in response to viral infection and double-stranded RNA. J. Biol. Chem. 281: 36560–36568. Kato, H., S. Sato, M. Yoneyama, M. Yamamoto, S. Uematsu, K. Matsui, T. Tsujimura, K. Takeda, T. Fujita, O. Takeuchi, and S. Akira. 2005. Cell type-specific involvement of RIG-I in antiviral response. Immunity 23:19–28. Kato, H., O. Takeuchi, E. Mikamo-Satoh, R. Hirai, T. Kawai, K. Matsushita, A. Hiiragi, T. S. Dermody, T. Fujita, and S. Akira. 2008. Length-dependent recognition of double-stranded ribonucleic acids by retinoic acid-inducible gene-I and melanoma differentiation-associated gene 5. J. Exp. Med. 205: 1601–1610. Kato, H., O. Takeuchi, S. Sato, M. Yoneyama, M. Yamamoto, K. Matsui, S. Uematsu, A. Jung, T. Kawai, K. J. Ishii, O. Yamaguchi, K. Otsu, T. Tsujimura, C. S. Koh, C. Reis e Sousa, Y. Matsuura, T. Fujita, and S. Akira. 2006. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 441:101–105. Kawai, T., and S. Akira. 2009. The roles of TLRs, RLRs and NLRs in pathogen recognition. Int. Immunol. 21:317–337. Kawai, T., K. Takahashi, S. Sato, C. Coban, H. Kumar, H. Kato, K. J. Ishii, O. Takeuchi, and S. Akira. 2005. IPS-1, an adaptor triggering RIG-I- and Mda5-mediated type I interferon induction. Nat. Immunol. 6:981–988. Kolumam, G. A., S. Thomas, L. J. Thompson, J. Sprent, and K. MuraliKrishna. 2005. Type I interferons act directly on CD8 T cells to allow clonal expansion and memory formation in response to viral infection. J. Exp. Med. 202:637–650. Kurt-Jones, E. A., F. Sandor, Y. Ortiz, G. N. Bowen, S. L. Counter, T. C.

9462

30.

31.

32.

33.

34.

35.

36. 37.

38.

39.

40. 41.

42.

43.

44.

45.

46.

47.

48.

ZHOU ET AL.

Wang, and R. W. Finberg. 2004. Use of murine embryonic fibroblasts to define Toll-like receptor activation and specificity. J. Endotoxin Res. 10:419– 424. Le Bon, A., V. Durand, E. Kamphuis, C. Thompson, S. Bulfone-Paus, C. Rossmann, U. Kalinke, and D. F. Tough. 2006. Direct stimulation of T cells by type I IFN enhances the CD8⫹ T cell response during cross-priming. J. Immunol. 176:4682–4689. Lee, L. N., S. Burke, M. Montoya, and P. Borrow. 2009. Multiple mechanisms contribute to impairment of type I interferon production during chronic lymphocytic choriomeningitis virus infection of mice. J. Immunol. 182:7178–7189. Loo, Y. M., J. Fornek, N. Crochet, G. Bajwa, O. Perwitasari, L. MartinezSobrido, S. Akira, M. A. Gill, A. Garcia-Sastre, M. G. Katze, and M. Gale, Jr. 2008. Distinct RIG-I and MDA5 signaling by RNA viruses in innate immunity. J. Virol. 82:335–345. Loo, Y. M., D. M. Owen, K. Li, A. K. Erickson, C. L. Johnson, P. M. Fish, D. S. Carney, T. Wang, H. Ishida, M. Yoneyama, T. Fujita, T. Saito, W. M. Lee, C. H. Hagedorn, D. T. Lau, S. A. Weinman, S. M. Lemon, and M. Gale, Jr. 2006. Viral and therapeutic control of IFN-beta promoter stimulator 1 during hepatitis C virus infection. Proc. Natl. Acad. Sci. U. S. A. 103:6001– 6006. Louten, J., N. van Rooijen, and C. A. Biron. 2006. Type 1 IFN deficiency in the absence of normal splenic architecture during lymphocytic choriomeningitis virus infection. J. Immunol. 177:3266–3272. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and G. Schuler. 1999. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223:77–92. Marrack, P., J. Kappler, and T. Mitchell. 1999. Type I interferons keep activated T cells alive. J. Exp. Med. 189:521–530. Martinez-Sobrido, L., S. Emonet, P. Giannakas, B. Cubitt, A. Garcia-Sastre, and J. C. de la Torre. 2009. Identification of amino acid residues critical for the anti-interferon activity of the nucleoprotein of the prototypic arenavirus lymphocytic choriomeningitis virus. J. Virol. 83:11330–11340. Martinez-Sobrido, L., P. Giannakas, B. Cubitt, A. Garcia-Sastre, and J. C. de la Torre. 2007. Differential inhibition of type I interferon induction by arenavirus nucleoproteins. J. Virol. 81:12696–12703. Martinez-Sobrido, L., E. I. Zuniga, D. Rosario, A. Garcia-Sastre, and J. C. de la Torre. 2006. Inhibition of the type I interferon response by the nucleoprotein of the prototypic arenavirus lymphocytic choriomeningitis virus. J. Virol. 80:9192–9199. Medzhitov, R. 2001. Toll-like receptors and innate immunity. Nat. Rev. Immunol. 1:135–145. Merigan, T. C., M. B. Oldstone, and R. M. Welsh. 1977. Interferon production during lymphocytic choriomeningitis virus infection of nude and normal mice. Nature 268:67–68. Meylan, E., J. Curran, K. Hofmann, D. Moradpour, M. Binder, R. Bartenschlager, and J. Tschopp. 2005. Cardif is an adaptor protein in the RIG-I antiviral pathway and is targeted by hepatitis C virus. Nature 437:1167–1172. Mibayashi, M., L. Martinez-Sobrido, Y.-M. Loo, W. B. Cardenas, M. Gale, Jr., and A. Garcia-Sastre. 2007. Inhibition of retinoic acid-inducible gene I-mediated induction of beta interferon by the NS1 protein of influenza A virus. J. Virol. 81:514–524. Montoya, M., M. J. Edwards, D. M. Reid, and P. Borrow. 2005. Rapid activation of spleen dendritic cell subsets following lymphocytic choriomeningitis virus infection of mice: analysis of the involvement of type 1 IFN. J. Immunol. 174:1851–1861. Moskophidis, D., F. Lechner, H. Pircher, and R. M. Zinkernagel. 1993. Virus persistence in acutely infected immunocompetent mice by exhaustion of antiviral cytotoxic effector T cells. Nature 362:758–761. Moskophidis, D., and F. Lehmann-Grube. 1984. The immune response of the mouse to lymphocytic choriomeningitis virus. IV. Enumeration of antibody-producing cells in spleens during acute and persistent infection. J. Immunol. 133:3366–3370. Muller, U., U. Steinhoff, L. F. Reis, S. Hemmi, J. Pavlovic, R. M. Zinkernagel, and M. Aguet. 1994. Functional role of type I and type II interferons in antiviral defense. Science 264:1918–1921. Nakhaei, P., P. Genin, A. Civas, and J. Hiscott. 2009. RIG-I-like receptors:

J. VIROL. sensing and responding to RNA virus infection. Semin. Immunol. 21:215– 222. 49. Oldstone, M. B., R. Ahmed, J. Byrne, M. J. Buchmeier, Y. Riviere, and P. Southern. 1985. Virus and immune responses: lymphocytic choriomeningitis virus as a prototype model of viral pathogenesis. Br. Med. Bull. 41:70–74. 50. Ou, R., S. Zhou, L. Huang, and D. Moskophidis. 2001. Critical role for alpha/beta and gamma interferons in persistence of lymphocytic choriomeningitis virus by clonal exhaustion of cytotoxic T cells. J. Virol. 75:8407–8423. 51. Pichlmair, A., O. Schulz, C. P. Tan, T. I. Naslund, P. Liljestrom, F. Weber, and C. Reis e Sousa. 2006. RIG-I-mediated antiviral responses to singlestranded RNA bearing 5⬘-phosphates. Science 314:997–1001. 51a.Rehwinkel, J., C. P. Tan, D. Goubau, O. Schulz, A. Pichlmair, K. Bier, N. Robb, F. Vreede, W. Barclay, E. Fodor, and C. Reis e Sousa. 2010. RIG-I detects viral genomic RNA during negative-strand RNA virus infection. Cell 140:397–408. 52. Rothenfusser, S., N. Goutagny, G. DiPerna, M. Gong, B. G. Monks, A. Schoenemeyer, M. Yamamoto, S. Akira, and K. A. Fitzgerald. 2005. The RNA helicase Lgp2 inhibits TLR-independent sensing of viral replication by retinoic acid-inducible gene-I. J. Immunol. 175:5260–5268. 53. Sato, M., H. Suemori, N. Hata, M. Asagiri, K. Ogasawara, K. Nakao, T. Nakaya, M. Katsuki, S. Noguchi, N. Tanaka, and T. Taniguchi. 2000. Distinct and essential roles of transcription factors IRF-3 and IRF-7 in response to viruses for IFN-alpha/beta gene induction. Immunity 13:539–548. 54. Seth, R. B., L. Sun, C.-K. Ea, and Z. J. Chen. 2005. Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-kappaB and IRF3. Cell 122:669–682. 55. Sharma, S., B. R. tenOever, N. Grandvaux, G. P. Zhou, R. Lin, and J. Hiscott. 2003. Triggering the interferon antiviral response through an IKKrelated pathway. Science 300:1148–1151. 56. Steinhoff, U., U. Muller, A. Schertler, H. Hengartner, M. Aguet, and R. M. Zinkernagel. 1995. Antiviral protection by vesicular stomatitis virus-specific antibodies in alpha/beta interferon receptor-deficient mice. J. Virol. 69: 2153–2158. 57. Sun, Q., L. Sun, H. H. Liu, X. Chen, R. B. Seth, J. Forman, and Z. J. Chen. 2006. The specific and essential role of MAVS in antiviral innate immune responses. Immunity 24:633–642. 58. Sy, M. S., M. Tsurufuji, R. Finberg, and B. Benacerraf. 1983. Effect of vesicular stomatitis virus (VSV) infection on the development and regulation of T cell-mediated immune responses. J. Immunol. 131:30–36. 59. Tabeta, K., K. Hoebe, E. M. Janssen, X. Du, P. Georgel, K. Crozat, S. Mudd, N. Mann, S. Sovath, J. Goode, L. Shamel, A. A. Herskovits, D. A. Portnoy, M. Cooke, L. M. Tarantino, T. Wiltshire, B. E. Steinberg, S. Grinstein, and B. Beutler. 2006. The Unc93b1 mutation 3d disrupts exogenous antigen presentation and signaling via Toll-like receptors 3, 7 and 9. Nat. Immunol. 7:156–164. 60. Takeuchi, O., and S. Akira. 2009. Innate immunity to virus infection. Immunol. Rev. 227:75–86. 61. Tough, D. F. 2004. Type I interferon as a link between innate and adaptive immunity through dendritic cell stimulation. Leuk. Lymphoma 45:257–264. 62. van den Broek, M. F., U. Muller, S. Huang, M. Aguet, and R. M. Zinkernagel. 1995. Antiviral defense in mice lacking both alpha/beta and gamma interferon receptors. J. Virol. 69:4792–4796. 63. Wang, J. P., A. Cerny, D. R. Asher, E. A. Kurt-Jones, R. T. Bronson, and R. W. Finberg. 2010. MDA5 and MAVS mediate type I interferon responses to coxsackie B virus. J. Virol. 84:254–260. 64. Xu, L.-G., Y.-Y. Wang, K.-J. Han, L.-Y. Li, Z. Zhai, and H.-B. Shu. 2005. VISA is an adapter protein required for virus-triggered IFN-beta signaling. Mol. Cell 19:727–740. 65. Zhou, S., E. A. Kurt-Jones, K. A. Fitzgerald, J. P. Wang, A. M. Cerny, M. Chan, and R. W. Finberg. 2007. Role of MyD88 in route-dependent susceptibility to vesicular stomatitis virus infection. J. Immunol. 178:5173–5181. 66. Zhou, S., E. A. Kurt-Jones, L. Mandell, A. Cerny, M. Chan, D. T. Golenbock, and R. W. Finberg. 2005. MyD88 is critical for the development of innate and adaptive immunity during acute lymphocytic choriomeningitis virus infection. Eur. J. Immunol. 35:822–830. 67. Zinkernagel, R. M. 1996. Immunology taught by viruses. Science 271:173– 178.