Melanoma Differentiation−Associated 5 Senses Hepatitis B and Activates Innate Immune Signaling To Suppress Virus Replication This information is current as of September 23, 2021. Hsin-Lin Lu and Fang Liao J Immunol 2013; 191:3264-3276; Prepublished online 7 August 2013; doi: 10.4049/jimmunol.1300512 http://www.jimmunol.org/content/191/6/3264 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2013/08/07/jimmunol.130051

Material 2.DC1 http://www.jimmunol.org/ References This article cites 68 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/191/6/3264.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 23, 2021 • No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Melanoma Differentiation–Associated Gene 5 Senses Hepatitis B Virus and Activates Innate Immune Signaling To Suppress Virus Replication

Hsin-Lin Lu* and Fang Liao*,†

Retinoic acid–inducible gene-I (RIG-I) and melanoma differentiation–associated gene 5 (MDA5) belong to the RIG-I–like recep- tors family of pattern recognition receptors. Both RIG-I and MDA5 have been shown to recognize various viral , but whether they mediate hepatitis B virus (HBV) infection remains unclear. In this study, we demonstrated that the expression of MDA5,butnotRIG-I, was increased in Huh7 cells transfected with the HBV replicative plasmid and in the livers of mice hydrodynamically injected with the HBV replicative plasmid. To further determine the effect of RIG-I–like receptors on HBV replication, we cotransfected the HBV replicative plasmid with RIG-I or MDA5 expression plasmid into Huh7 cells and found that Downloaded from MDA5, but not RIG-I at a similar protein level, significantly inhibited HBV replication. Knockdown of endogenous MDA5, but not RIG-I, in Huh7 cells transfected with the HBV replicative plasmid significantly increased HBV replication. Of particular interest, we found that MDA5, but not RIG-I, was able to associate with HBV-specific nucleic acids, suggesting that MDA5 may sense HBV. Finally, we performed in vivo experiments by hydrodynamic injection of the HBV replicative plasmid into wild-type, MDA52/2, MDA5+/2,orRIG-I+/2 mice, and found that MDA52/2 and MDA5+/2 mice, but not RIG-I+/2 mice, exhibited an increase of HBV replication as compared with wild-type mice. Collectively, our in vitro and in vivo studies both support a critical role for MDA5 in http://www.jimmunol.org/ the innate immune response against HBV infection. The Journal of Immunology, 2013, 191: 3264–3276.

uman hepatitis B virus (HBV) is a small (3.2-kb), en- RNAs, which are exported to the cytoplasm and used as mRNA veloped, noncytopathic DNA virus characterized by for translating HBV proteins. The largest viral RNA (3.5 kb), H its pronounced species and liver tropism (1). HBV in- known as the viral pregenomic RNA (pgRNA), is assembled with fection is worldwide with a high prevalence in Asia and Africa, HBV polymerase and core proteins to form nucleocapsids, and and it causes a wide spectrum of liver diseases, including acute/ functions as the template for reverse transcription within nucleo- chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma capsids in the cytoplasm, ultimately generating newly synthesized (1, 2). Although HBV vaccine has been available for three deca- rcDNA. The nucleocapsids can be either enveloped during their by guest on September 23, 2021 des, the innate immune response to HBV infection remains to be passage through the endoplasmic reticulum and Golgi complex elucidated. followed by secretion from the cells or retransported into the The genomic arrangement of HBV is unique among . The nucleus for the amplification of the cccDNA pool to generate more HBV genome comprises a relaxed circular partially dsDNA that viral RNAs (1, 2). contains four overlapping reading frames encoding the enve- Hosts infected by viruses usually elicit a rapid and potent in- lope, precore/core, polymerase, and X proteins (2). After entry and nate immune response to produce antiviral molecules to limit viral uncoating of HBV in hepatocytes, the HBV genome is transported replication and to prevent viral spreading before the adaptive into the nucleus, and the relaxed circular DNA (rcDNA) is con- immune response is generated (3–5). Pattern recognition receptors verted into covalently closed circular DNA (cccDNA). The cccDNA (PRRs), which recognize various pathogen-associated molecule serves as a transcriptional template for the synthesis of four viral patterns, have been shown to play a critical role in the innate im- mune response against pathogens (6). Among PRRs, the endo- somal TLRs including TLR3, TLR7/8, and TLR9, and cytosolic *Institute of Microbiology and Immunology, National Yang-Ming University, Taipei RIG-I–like receptors (RLRs) including retinoic acid–inducible † 11221, Taiwan; and Institute of Biomedical Sciences, Academia Sinica, Taipei gene I (RIG-I) and melanoma differentiation–associated gene 5 11529, Taiwan (MDA5) are important for sensing viral RNA during viral infec- Received for publication February 21, 2013. Accepted for publication July 5, 2013. tion (3, 5, 6). After the recognition of virus-associated molecules This work was supported by grants from Academia Sinica in Taiwan. by PRRs, PRRs activate their specific adaptor proteins: TIR Address correspondence and reprint requests to Dr. Fang Liao, Institute of Biomed- domain-containing adapter-inducing IFN-b for TLR3 (7), MyD88 ical Sciences, Academia Sinica, Taipei 11529, Taiwan. E-mail address: fl9z@ibms. b sinica.edu.tw for TLR7/8 and TLR9 (8), and IFN- promoter stimulator 1(IPS- The online version of this article contains supplemental material. 1), also known as mitochondrial antiviral signaling protein, CARD adaptor-inducing IFN-b, and virus-induced signaling adaptor for Abbreviations used in this article: cccDNA, covalently closed circular DNA; DP, deproteinized; HBcAg, HBV core Ag; HBsAg, HBV surface Ag; HBV, hepatitis B RLRs (9–12). The activation of adaptor proteins of PRRs ultimately virus; IPS-1, IFN-b promoter stimulator 1; IRF, IFN regulatory factor; ISG, IFN- activates downstream transcription factors, IFN regulatory factors stimulated gene; MDA5, melanoma differentiation–associated gene 5; NP-40, Non- k idet P-40; PARP, poly(ADP-ribose) polymerase; pgRNA, pregenomic RNA; PRR, (IRFs), and NF- B, to induce that are critical for antiviral pattern recognition receptor; rcDNA, relaxed circular DNA; RIG-I, retinoic acid– functions, as well as the dictation of adaptive immune responses inducible gene-I; RLR, RIG-I–like receptor; siRNA, small interfering RNA; TMB, (3, 5, 6). The inhibition of HBV replication and the induction tetramethylbenzidine; VSV, vesicular stomatitis virus. of antiviral effects by TLR signaling are primarily mediated by Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 nonparenchymal cells, such as dendritic cells, Kupffer cells, and www.jimmunol.org/cgi/doi/10.4049/jimmunol.1300512 The Journal of Immunology 3265 liver sinusoidal endothelial cells (13, 14). Given that TLRs are Committee at Academia Sinica and were performed in accordance with expressed on plasma or endosomal membranes to recognize institutional guidelines. ligands from extracellular compartments, and that HBV is a non- Cell culture cytopathic DNA virus whose nucleic acids may not be present in extracellular compartments, one would expect that TLR signaling Huh7 cells were cultured in DMEM supplemented with 10% FBS (Life Technologies, Grand Island, NY), 2 mM L-glutamine, 1% nonessential may not be so critical for the innate response to HBV in hep- amino acids, and 1% sodium pyruvate (Life Technologies) at 37˚C under atocytes. Because RLRs are cytosolic viral sensors and HBV 5% CO2 in a humidified atmosphere. nucleic acids may be present in the cytosol in addition to the Plasmids nucleus as described in the aforementioned paragraph, HBV is more likely to be recognized by RLRs. Consistent with this no- The pSV2ANeo-HBVx2 plasmid, an HBV ayw dimer DNA containing tion, recent studies have demonstrated that overexpression of IPS- plasmid that has two head-to-tail copies of the HBV genome of ayw 1 in a hepatoma cell line transfected with the HBV replicative subtype (33), and the polymerase-null HBV mutant 2310, a point mutation converting the first ATG codon of HBV pol at nucleotide 2310 into ACG in plasmid significantly suppresses HBV replication (15), and that pSV2ANeo-HBVx2 plasmid (34, 35), were kindly provided by Dr. Chiaho HBV X protein interacts with IPS-1 and disrupts the downstream Shih (Institute of Biomedical Sciences, Academia Sinica). To generate pKRX- signaling of RLRs to prevent the production of type I IFNs in- HBVx2, we subjected pSV2ANeo-HBVx2 to EcoRI partial digestion to duced by Sendai virus, vesicular stomatitis virus (VSV), or poly obtain the HBV dimer DNA fragment of HBV ayw subtype, which was then subcloned into the pKRX vector at the EcoRI site. To construct (dA:dT) (16–19). Furthermore, two studies have shown that HBV pCAGGS-RIG-I-Flag2, we used pEF-BOS-RIG-I plasmid (provided by pol impairs the activation of TBK1/IKKε, the downstream sig- Dr. Takashi Fujita, Department of Molecular Genetics, Institute for Virus naling molecule of IPS-1 in the RLR signaling pathway (20, 21). Research, Kyoto University, Kyoto, Japan) containing the coding region of Downloaded from Altogether, these studies suggest that IPS-1 is likely involved in RIG-I as the template together with a primer set (forward, 59-CCCTCG- 9 9 the innate response against HBV infection. Given that RIG-I and AGATGACCACCGAGCAGCGACGC-3 ; reverse, 5 -CGGGGTACCTT- TGGACATTTCTGCTGGATCA-39) to amplify the fragment of the coding MDA5 are upstream molecules of IPS-1 (22) and that IPS-1 is region of RIG-I by PCR, and the resulting PCR product was purified and involved in the innate response against HBV (15–21), RIG-I and subcloned into pCAGGS-MCS-Flag2 (provided by Dr. Steve S.-L. Chen, MDA5 may play a role in the regulation of HBV infection. Institute of Biomedical Sciences, Academia Sinica) at the XhoI and KpnI sites. To generate pCAGGS-MDA5-Flag2, we performed RT-PCR ampli- In this study, we intended to investigate the involvement of RIG-I http://www.jimmunol.org/ fication of total RNA from HepG2 cells to obtain the full-length MDA5 and MDA5 in HBV infection using both in vitro and in vivo cDNA, which was used as the template together with a primer set (forward, experiments. Experimentally studying HBV infection has been 59-CCCTCGAGATGTCGAATGGGTATTCCACAG-39; reverse, 59- GG- challenging because HBV fails to infect commonly used cell lines GGTACCATCCTCATCACTAAATAAAC-39) to amplify the fragment of hepatocyte origin. Moreover, it becomes difficult to perform of the coding region of MDA5 for subsequent cloning into pCAGGS- in vivo study of the immune response to HBV infection because MCS-Flag2. HBV fails to infect mice, the main animal model widely used for Small interfering RNA studying immune response to pathogens because of the availabil- One nontargeting control small interfering RNA (siRNA), four siRNAs ity of reagents and the similarity of immune response to humans. targeting human MDA5, and four siRNA targeting human RIG-I were Transient transfection of the HBV replicative plasmid into human purchased from Dharmacon (Lafayette, CO). Four sequence-specific by guest on September 23, 2021 hepatoma cell lines has been widely used for in vitro studies of siRNAs targeting human MDA5 were tested for their silence efficacy on HBV replication and of HBV interaction with host (23–25), to the specific inhibition of exogenous and endogenous MDA5 gene, and 9 9 overcome the obstacles, whereas hydrodynamic injection to de- the siRNA (5 -UGACACAAUUCGAAUGAUA-3 ) with the highest si- lence efficacy was used for experiments. Similarly, four sequence-specific liver the HBV replicative plasmid into mouse livers has been siRNAs targeting human RIG-I were also subjected to the determination of developed for the study of the host response to HBV infection the silence efficacy and the siRNA (59-CCACAACACUAGUAAACAA-39) (26–30). For in vitro studies, we transfected Huh7 cells with the with the highest silence efficacy was used for experiments. The negative 9 HBV replicative plasmid and the expression plasmid for MDA5 or control siRNA and the siRNA targeting human MxA (5 -UCCGUUAG- CCGUGGUGAUUUA-39) were purchased from Qiagen (Germantown, RIG-I and found that MDA5, but not RIG-I at a similar protein MD) and used to knock down endogenous MxA. Transfection of plasmids level, significantly activated downstream signaling to inhibit HBV and siRNAs into cells was performed using Lipofectamine 2000 (Invi- replication. In addition, we performed in vivo studies by hydro- trogen) under optimized conditions. dynamic injection of the HBV replicative plasmid into mice to Real-time PCR mimic acute HBV infection (26). In agreement with the in vitro data, the heterozygous and homozygous MDA5 knockout mice, Total RNA was extracted from mouse livers or Huh7 cells by MaestroZol but not the heterozygous RIG-I knockout mice, receiving the HBV RNA Extraction Reagent (Maestrogen, Las Vegas, NV) according to the manufacturer’s instructions. Total RNA (5 mg) was reverse transcribed into replicative plasmid showed increased HBV replication as com- cDNA by SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad, CA) pared with littermate control wild-type mice. Our study clearly according to the manufacturer’s instructions. The cDNA was subjected to demonstrates that MDA5, a well-known cytosolic sensor for RNA real-time PCR analysis. An appropriate amount of the cDNA was mixed viruses, plays a crucial role in the innate immune response against with Maxima SYBR Green/ROX qPCR Master Mix (Fermentas, ON, Canada) supplemented with gene-specific primers. The thermal cycling HBV infection. protocol was 1 cycle at 50˚C for 2 min and then 95˚C for 10 min, followed by 40 cycles of 95˚C for 15 s and 60˚C for 1 min. The resultant PCR products Materials and Methods were analyzed by ABI 7500 software (Applied Biosystems, Foster City, CA). Mice The levels of were analyzed with the DDCt method, and all quantifications were normalized to the level of b-actin (forward, 59-TG- BALB/c mice were purchased from the National Laboratory Animal Center GACTTCGAGCAAGAGATG-39;reverse,59-TTGCTGATCCACATCTGC- (Taipei, Taiwan) and housed under specific pathogen-free conditions at TG-39), MDA5 (forward, 59-AGTTTGGCAGAAGGAAGTGTC-39;reverse, the Institute of Biomedical Sciences, Academia Sinica (Taipei, Taiwan). 59-GGAGTTTTCAAGGATTTGAGC-39), RIG-I (forward, 59-GAATCTG- The original breeders of MDA5+/2 and RIG-I+/2 mice were kindly provided CAAAGACCTCGAA-39;reverse,59-TCTGAGTAAGATCTTGCTCAATC- by Dr. Shizuo Akira (Research Institute for Microbial Diseases, Osaka 39), HBV core gene (forward, 59-CGTTTTTGCCTTCTGACTTCTTTC-39; University, Osaka, Japan) (31, 32). The MDA5+/2 and RIG-I+/2 mice were reverse, 59-ATAGGATAGGGGCATTTGGTGGTC-39), and MxA (forward, crossed with C57BL/6 mice, and the offspring mice were intercrossed and 59-GCTACACACCGTGACGGATATGG-39;reverse,59-CGAGCTGGACTG- maintained under specific pathogen-free conditions. All animal experi- GAAAGCCC-39). The real-time PCRs for OAS1, CXCL10,andIFN-b were ments were approved by the Institutional Animal Care and Utilization performed using the TaqMan probes (Applied Biosystems). The levels of gene 3266 THE MDA5 SIGNALING PATHWAY SUPPRESSES HBV REPLICATION expression were analyzed with the DDCt method, and all quantifications were 1 mM DTT, and 0.5 mM PMSF) and incubated on ice for 15 min. After normalized to the level of GAPDH. incubation, 10% NP-40 was added to the cell suspensions (1/16, v/v) and mixed by vortexing for 20 s. The lysates were then centrifuged at 9000 3 g Western blotting at 4˚C for 2 min to pellet the nucleus, and the supernatants were saved as cytosolic fractions. The nuclear pellets were washed with hypotonic buffer Cells were lysed on ice in lysis buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1 mM aprotinin, 1 mM leupeptin, and 1 mM (20 mM HEPES, pH 7.0, 10 mM KCl, 1 mM DTT, and 0.5 mM PMSF) PMSF), and cell lysates were centrifuged at 14,000 3 g at 4˚C for 15 min. and then resuspended in hypertonic buffer (20 mM HEPES, pH 7.0, 10 The resulting cell lysates were subjected to 8% SDS-PAGE and electro- mM KCl, 0.5 M NaCl, 1 mM DTT, and 0.5 mM PMSF). The nuclear transferred onto nitrocellulose membranes (PerkinElmer, Norwalk, CT) lysates were incubated at 4˚C for 30 min followed by centrifugation at 15,000 3 g for 5 min at 4˚C to pellet the debris. The supernatants were followed by Western blotting. The membranes were blocked with 5% saved as nuclear fractions. nonfat milk in PBS containing 0.1% Tween 20 at room temperature for 1 h, and all incubations and washes were done in the presence of blocking DNA and RNA immunoprecipitation solution. Blots were incubated with specific primary Abs, washed, and incubated with HRP-conjugated secondary Abs (Pierce, Rockford, IL), Cells were harvested, resuspended in PBS, and fixed with 1% formaldehyde washed again, and visualized by chemiluminescence using the SuperSignal in PBS at room temperature for 15 min followed by the addition of 2.5 M West Pico Chemiluminescent Substrate (Pierce). The primary Abs used for glycine (1/10, v/v) to neutralize the cross-linking. The fixed cells were then Western blotting were rabbit polyclonal anti-FLAG (Sigma-Aldrich, St. centrifuged at 800 3 g at 4˚C for 5 min and washed three times with PBS. Louis, MO) to detect FLAG-MDA5 and FLAG–RIG-I, rabbit anti–b-actin The cell pellets were resuspended in ice-cold FA lysis buffer (50 mM polyclonal Ab (Sigma-Aldrich) to detect b-actin, rabbit anti-IRF3 phospho HEPES, pH 7.5, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% (pS386) mAb (Epitomics, Burlingame, CA) to detect phosphorylated sodium deoxycholate, 1 mM leupeptin, 1 mM aprotinin, 1 mM PMSF, and IRF3, rabbit polyclonal anti-IRF3 (Santa Cruz Biotechnology, Santa Cruz, 20 mg/ml RNase inhibitor), and the cells were lysed with repeated freezing CA) to detect total IRF3, rabbit polyclonal anti–NF-kB/p65 (Santa Cruz and thawing. The cell lysates were then centrifuged at 15,000 3 g at 4˚C Downloaded from Biotechnology) to detect NF-kB/p65, rabbit polyclonal anti–poly(ADP- for 15 min, and the supernatants were collected and subjected to immu- ribose) polymerase (PARP) (Cell Signaling, Danvers, MA) to detect the noprecipitation with anti-FLAG M2 affinity gel (Sigma-Aldrich). The nuclear marker PARP, and rabbit monoclonal anti–a-tubulin (Epitomics) to immunoprecipitates were sequentially washed three times with FA lysis detect the cytosolic marker a-tubulin. buffer, once with FA500 (50 mM HEPES, pH 7.5, 500 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 1 mM leupeptin, Northern blotting 1 mM aprotinin, 1 mM PMSF, and 20 mg/ml RNase inhibitor), once with Total RNA (20 mg) was separated in 1% formaldehyde-agarose gels, LiCl wash buffer (10 mM Tris, pH 8.0, 250 mM LiCl, 0.5% NP-40, 0.1% transferred to IMMOBILON NY+ charged nylon membranes (Millipore, sodium deoxycholate, 1 mM EDTA, 1 mM leupeptin, 1 mM aprotinin, http://www.jimmunol.org/ m Billerica, MA), and prehybridized in ULTRAhyb Hybridization buffer 1 mM PMSF, and 20 g/ml RNase inhibitor), and finally once with TE/0.1 (Ambion, Austin, TX) at 42˚C for 1 h. The membranes were then hy- M NaCl (10 mM Tris, pH 8.0, 1 mM EDTA, 100 mM NaCl, 1 mM leu- m bridized with 1 3 106 cpm/ml of [32P]-labeled specific probe at 42˚C peptin, 1 mM aprotinin, 1 mM PMSF, and 20 g/ml RNase inhibitor). The overnight. The radioisotope-labeled probe was generated by labeling 25 ng nucleic acids associated with the immunoprecipitates were eluted from purified full-length 3.2-kb HBV DNA fragment with a-[32P]-dCTP (Per- anti-FLAG M2 affinity gels by elution buffer (100 mM Tris, pH 8.0, kinElmer) using the Amersham Rediprime II DNA labeling System (GE 10 mM EDTA, 1% SDS, 1 mM leupeptin, 1 mM aprotinin, 1 mM PMSF, m Healthcare Life Sciences, Piscataway, NJ), and the specificity of probes and 20 g/ml RNase inhibitor), and the formaldehyde cross-links were re- was ∼1 3 109 cpm/mg. The membrane was washed with 23 SSC/0.1% versed in elution buffer by incubation at 70˚C for 1 h. To detect HBV DNA SDS at room temperature for 1 h followed by three washes with 0.13 SSC/ present in immunoprecipitates, we subjected the eluent to DNA purifica- tion with the DNA Clean/Extraction Kit (GeneMark Technology, Tainan,

0.1% SDS at 65˚C for 30 min. The signals were detected by autoradiog- by guest on September 23, 2021 raphy film (MIDSCI, St. Louis, MO). Taiwan) followed by the detection of HBV sequence by real-time PCR using a primer set specific for the HBV core gene (forward, 59-CGTT- Southern blotting to detect intracapsid HBV DNA TTTGCCTTCTGACTTCTTTC-39; reverse, 59-ATAGGATAGGGGCATT- TGGTGGTC-39). To detect HBV RNA present in immunoprecipitates, we Viral DNAwas isolated from intracellular viral capsids and detected with pretreated the eluent with DNase I for 30 min at 37˚C, and the RNA was specific isotope-labeled probe as described previously (36). In brief, extracted by MaestroZol RNA Extraction Reagent (Maestrogen). The RNA cells were lysed in lysis buffer (10 mM Tris, pH 7.5, 1 mM EDTA, 50 was then reverse-transcribed into cDNA by SuperScript III Reverse mM NaCl, 0.25% Nonidet P-40 [NP-40], and 8% sucrose) at 37˚C for Transcriptase (Invitrogen), and the cDNA was subjected to the detection of 20 min. The cell lysates were centrifuged at 13,000 3 g for 15 min at HBV sequence by real-time PCR using a primer set specific for the HBV room temperature. The supernatants were collected and brought to a core gene as described earlier. final concentration of 8 mM CaCl2 and 6 mM MgCl2 followed by di- gestion with 30 U/ml micrococcal nuclease (New England Biolabs, In vitro binding of HBV DNA to purified FLAG-tagged RLRs Ipswich, MA) and 3 U/ml RQ1 DNase (Promega, Madison, WI) at 37˚C for 20 min. Then the solution (26% polyethylene glycol 8000, 1.4 M HEK293T cells were transfected with pCAGGS-MDA5-FLAGx2 or NaCl, and 60 mM EDTA) was added into the digested supernatants (1/3, pCAGGS-RIG-I-FLAGx2 by Lipofectamine 2000. At 48 h posttrans- v/v). After incubation at 4˚C for 2 h, the viral core particles were pel- fection, transfectants were lysed on ice in lysis buffer (50 mM Tris, pH leted by centrifugation at 10,000 3 g at 4˚C for 10 min. The pellet was 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1 mM aprotinin, 1 mM leupeptin, and 1 mM PMSF), and cell lysates were centrifuged resuspended in buffer containing 10 mM Tris, pH 7.5, 8 mM CaCl2,and at 14,000 3 g at 4˚C for 15 min. The supernatants were collected and 6mMMgCl2, and subjected to digestion with 30 U/ml micrococcal nuclease and 3 U/ml RQ1 DNase at 37˚C for 20 min. The nuclease- subjected into immunoprecipitation with the M2 anti-FLAG affinity gel. digested core particles were lysed in SDS lysis buffer (25 mM Tris, pH After three washes with lysis buffer, the FLAG-tagged proteins were m 7.5, 10 mM EDTA, and 1% SDS) that contained 400 mg/ml Proteinase eluted from the immunoprecipitates with lysis buffer containing 100 g/ 3 K (MDBio, Taipei, Taiwan) at 50˚C overnight, and the core particle– ml FLAG peptide at 4˚C. After centrifugation at 900 g at 4˚C for associated DNA was released and extracted by phenol/chloroform 5 min, the supernatants containing purified FLAG-tagged proteins were followed by ethanol precipitation. The core particle–associated DNA collected and coated onto Nunc MAXISORP Immuno plate (Nunc, was separated in a 1% agarose gel, transferred onto IMMOBILON NY+ Roskilde, Denmark) at 4˚C overnight. The wells were washed three charged nylon membranes (Millipore), and prehybridized in ULTRAhyb times with PBST (PBS containing 0.1% Tween 20) and then blocked Hybridization buffer (Ambion) at 42˚C for 1 h. The membranes were with 1% BSA in PBS at room temperature for 1 h. Biotinylated then hybridized with 1 3 106 cpm/ml [32P]-labeled specific probes at HBV DNA was generated by PCR with pSV2ANeo-HBVx2 as template 9 9 42˚C overnight. The membranes were washed with 23 SSC/0.1% SDS at and with 5 -biotinylated primer pairs (forward: 5 -AATTCCACAAC- 9 9 room temperature for 1 h, followed by three washes with 0.13 SSC/0.1% CTTCCACCAAACTC-3 ; reverse: 5 -CCACTGCATGGCCTGAGGA- 9 SDS at 65˚C for 30 min. The signals were detected by autoradiography TGAGTG-3 ), and the resultant PCR products were purified using the film. DNA Clean/Extraction Kit. The purified biotinylated HBV DNA was added into the wells coated with or without FLAG-tagged proteins, and Subcellular fractionation incubated at room temperature for 1 h. After three washes with PBST, the biotinylated HBV DNA associated with FLAG-MDA5 or FLAG–RIG-I Cells were harvested and washed twice with PBS. Cell pellets were was detected using HRP-conjugated streptavidin with tetramethylbenzi- resuspended in hypotonic buffer (20 mM HEPES, pH 7.0, 10 mM KCl, dine (TMB) as a substrate. The Journal of Immunology 3267

Detection of HBV surface Ag by ELISA The HBV surface Ag (HBsAg) concentrations in sera or culture super- natants were determined using the SURASE B-96 ELISA kit (General Biologicals, Hsinchu, Taiwan).

Immunofluorescence staining Cells were washed three times with PBS, fixed with 4% formaldehyde in PBS at room temperature for 30 min, and permeabilized with 0.1% Triton X-100 in PBS at room temperature for 10 min. The cells were then washed three times with PBS, incubated with 5% BSA in PBS at room temperature for 1 h, and immunostained with specific primary Abs. After three washes with PBS, cells were further incubated with fluorochrome-conjugated secondary Abs for 30 min followed by counterstaining with 0.5 mg/ml DAPI for 10 min. Images were obtained using a confocal microscope (LSM 700; Carl Zeiss AG, Oberkochen, Germany).

Hydrodynamic injection Mice (6–8 wk old) were anesthetized by i.m. injection of 30 ml per mouse of atropine/ketamine/xylazine mixture (1 mg/ml atropine/100 mg/ml ketamine/ 2% xylazine/saline, 2/1/1/1, v/v/v/v). Thirty micrograms of pKRX or pKRX- HBVx2 in a volume of isotonic saline equivalent to 8% of body weight were Downloaded from injected into the tail veins of mice within 8 s. Statistical analysis The unpaired Student t test was used to evaluate the significance of the difference between two experimental results. A p value ,0.05 was con- sidered statistically significant. http://www.jimmunol.org/ FIGURE 1. Mice or cells transfected with the HBV replicative plasmid Results increase MDA5 expression. (A) Huh7 cells were transfected with the HBV Introducing the HBV replicative plasmid into Huh7 cells or replicative plasmid. At 45 h posttransfection, the culture supernatants were into mice increases MDA5 expression collected and subjected to HBsAg measurement by ELISA (left panel), and To investigate the involvement of RLRs in HBV infection, we de- the cells were harvested for total RNA extraction followed by the first- strand cDNA synthesis by reverse transcription. The cDNAs were used to livered the HBV replicative plasmid into Huh7 cells and into mouse determine the expression of MDA5 and RIG-I by real-time PCR (middle liver by transfection and hydrodynamical injection, respectively, and and right panels). (B) The HBV replicative plasmidwasdeliveredinto measured the expression of MDA5 and RIG-I in the Huh7 cells and BALB/c mice by hydrodynamic injection. On day 3 posttransfection, mouse liver. To ensure successful transfection of the HBV replicative HBsAg in the sera was determined by ELISA (left panel), and the ex- by guest on September 23, 2021 plasmid, we measured HBsAg from the culture supernatants of Huh7 pression of MDA5 and RIG-I in the liver was determined by reverse cells and from the sera of mice (Fig. 1A, 1B, left panels). Of interest, transcription followed by real-time PCR (middle and right panels). Data the expression of MDA5, but not RIG-I, was significantly increased are presented as the mean 6 SEM. *p , 0.05. in Huh7 cells transfected with the HBV replicative plasmid as compared with the vector (p , 0.05; Fig. 1A). Likewise, the in vivo experiments also showed significantly increased expression of Fig. 2C). Altogether, MDA5 overexpression significantly reduces MDA5, but not RIG-I, in the livers of mice receiving the HBV the levels of DNA, RNA, and protein of HBV, suggesting that the replicative plasmid as compared with the vector on day 3 postin- MDA5 signaling pathway mediates the suppression of HBV rep- jection (p , 0.05; Fig. 1B). Taken together, both in vitro and in vivo lication. The exogenously expressed MDA5 was knocked down by experiments suggest that MDA5 plays a role in HBV infection. specific siRNA and the level of secreted HBsAg was measured to further confirm that the suppression of HBV replication is medi- MDA5 signaling pathway suppresses HBV replication ated by MDA5 expressed in cells transfected with the HBV rep- Knowing that MDA5 expression is upregulated after HBV trans- licative plasmid. Knockdown of ectopic MDA5 expression by fection (Fig. 1) and that MDA5 signaling pathway is responsible siRNA in cells cotransfected with the HBV replicative plasmid for the antiviral effect (32, 37, 38), we next investigated the effect and MDA5 expression plasmid restored the levels of secreted of MDA5 on HBV replication by gain-of-function experiments. HBsAg (Fig. 2D, 2E). Huh7 cells were cotransfected with the HBV replicative plasmid We next conducted the loss-of-function experiments to confirm and the plasmid encoding MDA5 or RIG-I or the control plasmid. the inhibitory effect of MDA5 on HBV replication. Huh7 cells were With comparable protein levels of MDA5 and RIG-I, only cells cotransfected with the HBV replicative plasmid and MDA5 siRNA transfected with the MDA5 expression plasmid showed significant or RIG-I siRNA followed by the determination of HBV RNA and inhibition of the levels of HBV RNA (3.5-, 2.4-, and 2.1-kb RNA) intracapsid DNA expression, and HBsAg secretion. The siRNA and intracapsid HBV DNA including rcDNA and ssDNA in Huh7 that targeted MDA5 or RIG-I showed ∼40% knockdown efficiency cells (Fig. 2A). Using immunofluorescence staining to detect HBV as evaluated by mRNA expression (Fig. 3A). Knockdown of en- core Ag (HBcAg) expression in cells that showed positive for dogenous MDA5 significantly increased HBsAg secretion (p , MDA5 or RIG-I, we found that the percentage of cells expressing 0.05; Fig. 3B), and substantially increased HBV RNA expression HBcAg in MDA5 transfectants was significantly lower than that and intracapsid DNA expression (Fig. 3C). In contrast, knock- in RIG-I transfectants (14.58 6 2.95 versus 95.27 6 1.63%, p , down of endogenous RIG-I failed to alter the secretion of HBsAg 0.01; Fig. 2B). Consistently, the level of secreted HBsAg in the (Fig. 3B), and the expression of HBV RNA and intracapsid DNA culture supernatants of MDA5 transfectants was significantly re- (Fig. 3C). These results clearly demonstrate that MDA5, but not duced as compared with control or RIG-I transfectants (p , 0.01; RIG-I, mediated the suppression of HBV replication. To further 3268 THE MDA5 SIGNALING PATHWAY SUPPRESSES HBV REPLICATION Downloaded from

FIGURE 2. Overexpression of MDA5 in Huh7 cells leads to the suppression of HBV replication. Huh7 cells were transfected with the HBV replicative plasmid together with either the expression vector encoding FLAG-tagged MDA5 or FLAG-tagged RIG-I or the corresponding control vector. (A)At48h posttransfection, the cells were harvested and the expression of MDA5 and RIG-I were detected by Western blotting using an anti-FLAG Ab (top panel), the levels of HBV RNAs were detected by Northern blotting (middle panel), and the levels of HBV DNAs were detected by Southern blotting (bottom panel). (B) The expression of HBcAg in the transfectants was detected by immunofluorescence staining, and the images were taken by a confocal microscope. The http://www.jimmunol.org/ percentage of HBcAg+ cells in cells expressing exogenous MDA5 or RIG-I was quantified by counting cells from three different fields per sample (bar graph at right). (C) The levels of HBsAg in culture supernatants were determined by ELISA. (D and E) Huh7 cells cotransfected with the HBV replicative plasmid together with either the expression vector encoding FLAG-tagged MDA5 or the corresponding control vector were transfected with either control siRNA or MDA5 siRNA. (D) At 48 h posttransfection, the level of FLAG-MDA5 was detected by Western blotting using the anti-FLAG Ab, and b-actin was used as a loading control. (E) The levels of HBsAg in culture supernatants were determined by ELISA. Data represent the mean 6 SEM from three independent experiments. *p , 0.01. RC, Relaxed circular HBV DNA; SS, single-stranded HBV DNA. demonstrate that the suppression of HBV replication by the MDA5 HepG2 cells. Knockdown of endogenous MDA5 in HepG2 cells signaling pathway was not limited to Huh7 cells, we performed increased HBsAg secretion and the HBV RNA expression in the loss-of-function experiments in another hepatoma cell line, a dose-dependent manner (Fig. 3D–F). by guest on September 23, 2021

FIGURE 3. Knockdown of endogenous MDA5 in Huh7 cells increases HBV replication. (A–C) Huh7 cells were transfected with the HBV replicative plasmid together with control siRNA, MDA5 siRNA, or RIG-I siRNA. At 48 h posttransfection, the culture supernatants were collected and the cells were harvested. (A) The levels of endogenous MDA5 and RIG-I expression were determined by real-time PCR. (B) The levels of HBsAg in culture supernatants were determined by ELISA. (C) The levels of HBV RNAs and HBV DNAs were detected by Northern and Southern blotting, respectively. (D–F) HepG2 cells were cotransfected with the HBV replicative plasmid and MDA5 siRNA. At 48 h posttransfection, the culture supernatants were collected and the cells were harvested. (D) The levels of endogenous MDA5 were determined by real-time PCR. (E) The levels of HBsAg in culture supernatants were determined by ELISA. (F) The levels of HBV RNAs were detected by Northern blotting. Data represent the mean 6 SEM from four independent experiments. *p , 0.05. The Journal of Immunology 3269

Cotransfection of Huh7 cells with the HBV replicative plasmid or the RIG-I expression plasmid alone, or the HBV replicative and MDA5 expression plasmid significantly induces the plasmid plus the RIG-I expression plasmid did not induce nuclear activation of both IRF3 and NF-kB translocation of NF-kB. In contrast, Huh7 cells cotransfected with Given that with similar levels of protein expression of MDA5 and the HBV replicative plasmid and the MDA5 expression plasmid k RIG-I, only MDA5 significantly mediated the suppression of HBV significantly increased the translocation of NF- B into the nu- replication (Figs. 2, 3) and that the key transcription factors ac- cleus. These results indicate that overexpression of MDA5, but not tivated by RLR signaling are IRF3 and NF-kB (3, 5), we then RIG-I, in Huh7 cells transfected with the HBV replicative plasmid triggered RLR downstream signaling pathway. examined whether Huh7 cells cotransfected with the HBV repli- cative plasmid and the MDA5 or RIG-I expression plasmid could Cotransfection of Huh7 cells with the HBV replicative plasmid activate IRF3 and NF-kB. Huh7 cells transfected with the HBV and MDA5 expression plasmid induces IFN-stimulated gene replicative plasmid alone did not significantly increase IRF3 expression phosphorylation (Fig. 4A, lane 4). Overexpression of MDA5 alone We demonstrated that cells cotransfected with MDA5 expression substantially increased IRF3 phosphorylation (Fig. 4A, lane 2), plasmid and the HBV replicative plasmid activated the downstream and IRF3 phosphorylation was further enhanced profoundly in the transcription factors (Fig. 4). We next examined whether activa- presence of the HBV replicative plasmid (Fig. 4A, lane 5). tion of the downstream transcription factors of the MDA5 sig- Overexpression of RIG-I alone barely increased IRF3 phosphor- naling pathway eventually led to the induction of IFN-stimulated ylation (Fig. 4A, lane 3); however, IRF3 phosphorylation was not gene (ISG) expression, transcriptional targets of the RLR signal- significantly increased even in the presence of the HBV replicative ing pathway. Huh7 cells were cotransfected with the HBV repli- Downloaded from plasmid (Fig. 4A, lane 6). IRF3 activation was further confirmed cative plasmid and the MDA5 or RIG-I expression plasmid, and by examining IRF3 nuclear translocation by immunostaining. As total RNAs were extracted from the transfectants 12 and 24 h shown in Fig. 4B, the percentage of cells positive for nuclear IRF3 posttransfection and subjected to the reverse-transcriptase PCR was significantly higher in Huh7 cells cotransfected with the HBV assay. The expression of ISGs, including MxA, OAS1, CXCL10, replicative plasmid and the MDA5 expression plasmid than in and IFN-b, known to be induced by the RLR signaling pathway cells cotransfected with the HBV replicative plasmid and the RIG- was determined (Fig. 5A). The expressions of MxA, OAS1, and http://www.jimmunol.org/ I expression plasmid (89.81 6 13.13 versus 36.86 6 14.42%, p , CXCL10 were upregulated in cells transfected with the MDA5 0.05). To examine the NF-kB activation, we determined the nu- expression plasmid only and further increased in the presence of clear translocation of NF-kB (Fig. 4C). Transfection of Huh7 cells the HBV replicative plasmid; the increases were more evident at with the HBV replicative plasmid, the MDA5 expression plasmid, 24 h posttransfection (Fig. 5A). Interestingly, the level of IFN-b by guest on September 23, 2021

FIGURE 4. Ectopic expression of MDA5 leads to IRF3 and NF-kB activation in HBV-transfected cells. Huh7 cells were cotransfected with the HBV replicative plasmid and empty vector, the MDA5, or the RIG-I expression plasmid. (A) The phosphorylation of IRF3 and the expression of MDA5 and RIG- I were detected by Western blotting. b-actin was used as a loading control. (B) Nuclear translocation of IRF3 in transfectants was detected by immu- nofluorescence staining, and the images were observed by confocal microscopy. The percentage of cells with nuclear IRF3 staining in MDA5- or RIG-I– expressing cells was quantified by counting cells from three different fields per sample, as shown in the bar graph at right. Data are presented as the mean 6 SEM from three individual fields. *p , 0.05. (C) The lysates of transfectants were separated into cytoplasmic and nuclear fractions by subcellular fractionation. Whole-cell lysates, cytoplasmic fractions, and nuclear fractions were run on SDS-PAGE followed by immunoblotting with Abs to NF- kB, PARP, and a-tubulin. PARP and a-tubulin are markers for the nucleus and the cytoplasm, respectively. 3270 THE MDA5 SIGNALING PATHWAY SUPPRESSES HBV REPLICATION Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021

FIGURE 5. Overexpression of MDA5 leads to induction of ISGs in HBV-transfected cells. (A) Huh7 cells were transfected with the HBV replicative plasmid or the corresponding control plasmid together with the empty vector, or the MDA5 or RIG-I expression plasmid. At indicated time points, total RNAs were extracted from the transfectants and reversed transcribed into cDNA followed by quantitative PCR analysis to determine the levels of gene expression. The expression levels were analyzed with the DDCt method, and all quantifications were normalized to the level of GAPDH mRNA. (B) Huh7 cells were transfected with the HBV replicative plasmid together with the empty vector or the MDA5 expression plasmid at one fourth of the plasmid used in (A) and control siRNA or siRNA-specific targeting to MxA. At 24 h posttransfection, cells and culture supernatants were collected. Cells were subjected to total RNA extraction followed by cDNA synthesis and PCR analysis for determining the level of MxA expression. The levels of HBsAg in the culture supernatants were determined by ELISA. expression was slightly higher in Huh7 cells transfected with the slightly increased in Huh7 cells cotransfected with the HBV HBV replicative plasmid and the empty vector as compared with replicative plasmid and MDA5 or RIG-I expression plasmid as cells cotransfected with the HBV replicative plasmid and the compared with cells transfected with the HBV replicative plasmid MDA5 or RIG-I expression plasmid at 12 h posttransfection (Fig. and the empty vector at 24 h posttransfection (Fig. 5A, right 5A, left bottom panel). However, the IFN-b expression was bottom panel). The Journal of Immunology 3271

Given that MxA is known to suppress HBV replication (39–42) and that MxA expression is upregulated in cells cotransfected with the HBV replicative plasmid and the MDA5 expression plasmid (Fig. 5A, top panel), we examined whether MxA contributed to the suppression of HBV replication mediated by the MDA5 sig- naling pathway. Knockdown of MxA in Huh7 cells cotransfected with the HBV replicative plasmid and the MDA5 expression plas- mid restored the suppression of HBV replication mediated by the MDA5 signaling pathway (Fig. 5B). This result indicates that MxA expression induced by the MDA5 signaling pathway participates in the suppression of HBV replication. MDA5, but not RIG-I, is associated with HBV nucleic acids We demonstrated that the MDA5-mediated innate signaling pathway suppressed HBV replication (Figs. 2, 3) and induced the activation of both IRF3 and NF-kB (Fig. 4), which subsequently induced the expression of antiviral ISGs (Fig. 5). We reasoned that HBV might be sensed by MDA5. We next investigated whether

MDA5 was associated with HBV RNA by conducting RNA im- Downloaded from FIGURE 6. MDA5 associates with HBV RNAs. (A) Huh7 cells were munoprecipitation. Huh7 cells were cotransfected with the HBV transfected with the HBV replicative plasmid or the corresponding control replicative plasmid and the plasmid encoding FLAG-MDA5 or plasmid together with either the expression vector encoding FLAG-tagged FLAG-RIG-I, and the transfectants were lysed and subjected to MDA5 or FLAG-tagged RIG-I or the corresponding control vector. At 48 h immunoprecipitation with an anti-FLAG Ab. RNA was extracted posttransfection, cells were cross-linked by formaldehyde followed by cell from RLR immunoprecipitates and subjected to cDNA synthesis lysis, and cell lysates were immunoprecipitated with the anti-FLAG M2

followed by real-time PCR analysis using primers specific to the affinity gel. The immunoprecipitates were washed with stringent condition, http://www.jimmunol.org/ HBV core gene. The amplification of core gene was found only in and the cross-linking was reversed by heating. Total RNAs extracted from cDNAs that were reverse transcribed from the RNAs extracted the immunoprecipitates were reverse transcribed into cDNAs. The cDNAs from MDA5 immunoprecipitates, but not from RIG-I immuno- were used as templates to amplify HBV DNA by real-time PCR using precipitates (Fig. 6A). To show the RNA purified from RLR primers specific for the HBV core gene. The copy number reflected the amount of the HBV nucleic acid in the immunoprecipitates from the cells immunoprecipitates was not contaminated with HBV genomic cultured in a 6-cm dish. Data are presented as the mean 6 SEM and are DNA, we also conducted RT-PCR in the absence of reverse representative of three independent experiments. (B) Similar to (A) except transcriptase. No PCR product was detected in the absence of that additional experiments in the absence of reverse transcriptase (RT) reverse transcriptase, indicating no HBV genomic DNA contam- were performed to rule out possible HBV genomic DNA contamination in ination in the purified RNA (Fig. 6B). These results demonstrate the RNA samples. The resultant PCR products were visualized with by guest on September 23, 2021 that MDA5, but not RIG-I, is associated with HBV RNAs, sug- ethidium bromide staining. gesting that MDA5 may sense HBV RNA. To ensure the lack of association of HBV RNA with RIG-I in RNA immunoprecipita- be sufficient to activate MDA5-mediated signaling, which subse- tion assay was not due to an experimental artifact, we performed quently suppresses HBV replication. RNA immunoprecipitation assay simultaneously with cells trans- Although MDA5 is recognized as a cytosolic RNA sensor, some fected with the RIG-I or MDA5 expression plasmid followed by studies showed that MDA5 interacts with DNA (44, 45). We sought infection with VSV, known to be sensed by RIG-I (31, 43). As to determine whether HBV DNA was associated with RLR im- expected, the result showed that RIG-I, but not MDA5, was as- munoprecipitates by performing DNA immunoprecipitation. Huh7 sociated with VSV RNA (Supplemental Fig. 1). To further confirm cells were cotransfected with the HBV replicative plasmid and that HBV RNAs indeed activated MDA5 signaling, we cotrans- the plasmid encoding FLAG-MDA5 or FLAG-RIG-I, and the fected Huh7 cells with the plasmid encoding MDA5 or RIG-I transfectants were lysed and subjected to immunoprecipitation together with the HBV replicative plasmid or the polymerase- with an anti-FLAG Ab. DNAs associated with the RLR immu- mutated HBV replicative plasmid (pol-null) that does not ex- noprecipitates were extracted and subjected to real-time PCR press HBV polymerase and only produce HBV RNAs. Over- analysis using primers specific to the HBV core gene. Interest- expression of MDA5 modestly increased the level of IRF3 ingly, HBV DNAs were detected in MDA5, but not RIG-I, phosphorylation (Fig. 7A, lane 2), and the level of IRF3 phos- immunoprecipitates (Fig. 8A). We also developed a plate-based phorylation was further increased in cells cotransfected with the in vitro binding assay to further confirm the association of HBV HBV replicative plasmid (Fig. 7A, lane 5). Of interest, replacing DNA with MDA5. FLAG-MDA5 or FLAG–RIG-I purified from the HBV replicative plasmid with the polymerase-mutated HBV HEK293T cells overexpressing the protein was coated onto ELISA replicative plasmid, which produced no HBV DNA, induced plates. The biotinylated HBV DNA generated from PCR was then comparable levels of IRF3 phosphorylation (Fig. 7A, lane 8). added into the coated plates. After incubation followed by sev- Cotransfection of Huh7 cells with the RIG-I expression plasmid eral washes, the biotinylated HBV DNA associated with FLAG- together with the HBV replicative plasmid or the polymerase- tagged RLRs was detected using HRP-conjugated streptavidin mutated HBV replicative plasmid slightly increased the levels of with TMB as a substrate. Consistent with the result obtained from IRF3 phosphorylation (Fig. 7A, lanes 6, 9). The supernatants DNA immunoprecipitation, the plate-based in vitro binding assay of these transfectants were collected for determining the levels of showed that only purified MDA5, but not RIG-I, was associated HBsAg. As expected, HBsAg secretion was significantly inhibited with HBV DNA (Fig. 8B, 8C). Based on the results obtained from in cells cotransfected with the MDA5 expression plasmid together both assays, we conclude that MDA5 is associated with HBV with the HBV replicative plasmid or the polymerase-mutated DNA. Collectively, the results shown in Figs. 6–8 demonstrate that plasmid (Fig. 7B). These results demonstrate that HBV RNAs may MDA5, but not RIG-I, is associated with not only HBV RNA but 3272 THE MDA5 SIGNALING PATHWAY SUPPRESSES HBV REPLICATION Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021

FIGURE 7. HBV RNAs activate MDA5 signaling. Huh7 cells were transfected with the HBV replicative plasmid (HBV wild-type) or the polymerase-mutated HBV replicative plasmid (Pol-null), or the corre- sponding control plasmid together with either the expression vector encoding FLAG-tagged MDA5 or the FLAG-tagged RIG-I or the corre- FIGURE 8. MDA5 associates with HBV DNAs. (A) Huh7 cells were sponding control vector. (A) The phosphorylation of IRF3 and the ex- transfected with the HBV replicative plasmid or the corresponding control pression of MDA5 or RIG-I were detected by Western blotting. b-Actin plasmid together with either the expression vector encoding FLAG-tagged was used as a loading control. (B) The cultured supernatants from (A) were MDA5 or the FLAG-tagged RIG-I or the corresponding control vector. collected and subjected to the determination of HBsAg by ELISA. Cells were cross-linked by formaldehyde followed by cell lysis, and cell lysates were immunoprecipitated with anti-FLAG M2 affinity gel. The immunoprecipitates were washed, and the cross-linking was reversed. also HBV DNA, suggesting that MDA5 may sense HBV nucleic Total DNAs were extracted from the immunoprecipitates and used as acids. templates to amplify HBV DNAs by real-time PCR using primers specific for the HBV core gene. The copy number reflected the amount of the HBV MDA5 is required to control HBV replication in vivo nucleic acid in the immunoprecipitates from the cells cultured in a 6-cm 6 To investigate the physiological importance of RLRs in HBV in- dish. Data are presented as the mean SEM and are representative of B fection, we conducted in vivo experiments by inducing acute HBV three independent experiments. ( ) HEK293T cells were transfected with FLAG-MDA5 or FLAG–RIG-I expression plasmid. The FLAG-MDA5 or infection using hydrodynamic injection to deliver the HBV rep- +/2 2/2 FLAG–RIG-I proteins were purified from transfectants using M2 anti- licative plasmid into the livers of MDA5 , MDA5 , and RIG- FLAG affinity gel followed by the elution of bound proteins with FLAG +/2 I mice and their wild-type littermate controls. The mouse sera peptides. The purity of the purified proteins was determined by Coomassie on days 1, 3, and 5 postinjection were collected to monitor the brilliant blue staining (left panel) and Western blotting (right panel). (C) serum levels of HBsAg. The serum levels of HBsAg in acute HBV Purified FLAG-MDA5 or FLAG–RIG-I was coated onto ELISA plates, and infection induced by hydrodynamical injection of the HBV rep- increased concentrations of biotinylated HBV DNA were added into the licative plasmid into livers consistently peaked on day 3 postin- coated plates. After incubation followed by several washes, the HBV DNA jection and diminished within a week. Interestingly, the serum that remained on the plates was detected using HRP-conjugated strepta- levels of HBsAg in MDA5+/2 and MDA52/2 mice were signifi- vidin with TMB as a substrate. cantly higher than those in littermate control mice on day 3 postinjection (Fig. 9A). Consistent with the result of serum RIG-I2/2 mice rarely survive and show massive liver degeneration HBsAg, the levels of HBV RNAs and intracapsid HBV DNAs in (31), we had difficulties obtaining enough RIG-I2/2 mice for the 2 2 2 livers from MDA5+/ and MDA5 / mice were markedly increased experiments. We then compared wild-type littermate control mice as compared with littermate control mice (Fig. 9B). Because with RIG-I+/2 mice and found their serum levels of HBsAg to be The Journal of Immunology 3273

FIGURE 9. MDA5 deficiency increases HBV replication in vivo. (A) MDA5+/2, MDA52/2 mice, and wild-type littermate controls were hydro- dynamically injected with 30 mg of the HBV replicative plasmid. The mouse sera were col- Downloaded from lected on days 1, 3, and 5 postinjection, and the levels of HBsAg were determined by ELISA. The numbers in parentheses indicate the num- ber of mice in each group. Data are presented as the mean 6 SEM. *p , 0.05. (B)Onday3post- injection, the livers from MDA5+/2, MDA52/2 mice, and wild-type littermate controls were http://www.jimmunol.org/ harvested and subjected to the detection of HBV RNA and DNA levels by Northern and Southern blotting, respectively. (C and D) Similar to (A) and (B) except that RIG-I+/2 mice and their lit- termate controls were used. by guest on September 23, 2021

comparable (Fig. 9C). Consistent with the result of serum HBsAg, Recognition of viruses by PRRs is essential to initiate the innate the levels of HBV RNAs and intracapsid HBV DNAs were similar antiviral immune response and, subsequently, to dictate the in livers between RIG-I +/2 mice and littermate control mice (Fig. adaptive immune response. RIG-I and MDA5 are cytoplasmic viral 9D). Altogether, the in vivo findings indicate that MDA5, but not sensors that play important roles in host defense against viral RIG-I, senses HBV and subsequently activates the innate immune infection, particularly RNA virus infection (31, 37, 38, 46). Al- signaling pathway to suppress HBV replication. though MDA5 and RIG-I have similar structures and share high , they recognize distinct RNA species (32). It Discussion is generally thought that RIG-I recognizes RNAs with uncapped In this study, both loss-of-function and gain-of-function experi- 59-triphosphate structures (47, 48), short dsRNA molecules (49, ments demonstrated that MDA5, but not RIG-I, activates the innate 50), and RNAs with panhandle structures (50), whereas MDA5 immune signaling pathway to suppress HBV replication. Impor- recognizes long dsRNAs (49) and high m.w. RNAs (51). RIG-I tantly, hydrodynamic injection of the HBV replicative plasmid into recognizes a wide variety of RNA viruses, whereas MDA5 rec- homozygous and heterozygous MDA5 knockout mice, but not in ognizes mainly picornaviruses whose RNAs do not bear 59-tri- heterozygous RIG-I knockout mice, significantly increased HBV phosphates (32, 37). Intriguingly, several recent studies have replication as compared with wild-type mice. Thus, our in vitro revealed that RIG-I and MDA5 are able to induce an antiviral and in vivo studies clearly indicate a critical role of the MDA5- response to viruses containing dsDNA genomes, such as EBV (52, mediated innate immune signaling in the suppression of HBV in- 53), vaccinia virus (51, 54), and HSV (55, 56). RIG-I is proposed fection. to induce an antiviral response to DNA viruses via RNA poly- 3274 THE MDA5 SIGNALING PATHWAY SUPPRESSES HBV REPLICATION merase III–mediated conversion of microbial DNA into 59-tri- fact that type I IFN is not significantly induced in HBV infection phosphate dsRNA (57, 58). Recognition of the DNAs of HSV and may suggest that the ISGs induction by HBV-mediated activation EBV by RIG-I is mediated via this pathway (52, 55, 58). Of note, of the MDA5 signaling pathway may be independent of IFN-b. one report showed that innate recognition of HSV in macro- Supporting this notion, a recent study demonstrated that IPS-1 is phages is mediated via an MDA5-dependent and RNA polymerase located on peroxisomes and mitochondria (64), and that peroxi- III–independent pathway (56). In addition to HSV, MDA5 also somal IPS-1 induces IFN-independent expression of ISGs,whereas mediates an antiviral response to another DNA virus, vaccinia vi- mitochondrial IPS-1 activates an IFN-dependent expression of ISGs rus. RNAs generated during vaccinia virus infection have higher- (64). Whether the peroxisomal IPS-1–mediated MDA5 signaling is order RNA structures, which activate an MDA5-dependent anti- more favorable than the mitochondrial IPS-1–mediated MDA5 sig- viral response (51). Of interest, this study demonstrated that naling in HBV-infected hepatocytes requires further investigation. immunoprecipitates of MDA5, but not RIG-I, were associated Intriguingly, although cells cotransfected with the HBV repli- with HBV RNA, suggesting that MDA5, but not RIG-I, acts as the cative plasmid and RIG-I expression plasmid slightly activate cytoplasmic sensor for HBV RNA and triggers downstream sig- IRF3 and NF-kB, this RIG-I–mediated signaling pathway failed to naling to suppress HBV replication. Given that the 3.5-kb HBV suppress HBV replication (Fig. 7). Given that HBV DNA directs pgRNA does not bear a 59-triphosphate and that pgRNA is pre- the synthesis of a 700-base RNA (HBV 700) by RNA polymerase dicted to have a complicated secondary structure (59), it is plau- III (65, 66), that RNA polymerase III–transcribed RNAs contain sible that HBV pgRNA may be a potential ligand for MDA5. a59-triphosphate, and that RIG-I recognizes 59-triphosphate However, we cannot exclude the possibility that the 2.4-, 2.1-, and RNAs and subsequently activates downstream signaling (47, 48),

0.7-kb HBV RNAs are also MDA5 ligands because these viral it is possible that RIG-I might sense HBV 700 and activate IRF3 Downloaded from RNA sequences are also part of HBV pgRNA. and NF-kB. Of note, HBV 700 is transcribed by RNA polymerase It is of interest that we found MDA5, but not RIG-I, is associated III at a very low level (65); therefore, HBV 700 may slightly in- with HBV DNA. Although some studies mentioned the interaction duce the activation of RIG-I signaling, which is not efficient of MDA5 with DNA (44, 45), whether the association of MDA5 enough to suppress HBV replication. Alternatively, IRF3 phos- with DNA is required for MDA5-mediated signaling remains phorylation alone may not be sufficient to inhibit HBV replication;

unknown. HBV cccDNA, a transcriptional template for the syn- other molecules triggered by the MDA5 signaling pathway are http://www.jimmunol.org/ thesis of viral RNA during HBV replication, may not be the can- required to work in concert with activated IRF3 to effectively didate DNA ligand for MDA5 because cccDNA is localized to the inhibit HBV replication. nucleus and is absent from the cytosol. However, a recent study We have proved the physiological importance of MDA5 in innate showed that an intracellular deproteinized rcDNA (DP-rcDNA) of immunity to HBV infection by in vivo experiments. Of interest, HBV is present in both the cytoplasm and the nucleus (60). The both MDA5+/2 and MDA52/2 mice hydrodynamically injected presence of some naked DP-rcDNA in the cytosol raises the pos- with the HBV replicative plasmid to mimic acute HBV infection sibility that DP-rcDNA may be sensed by MDA5. had comparable increases in HBV replication (Fig. 9A, 9B), sug- Guo et al. (15) demonstrated that overexpression of IPS-1 gesting that a single functional copy of MDA5 is not sufficient for suppresses HBV replication, and our study showed that MDA5, suppressing HBV replication. The haploinsufficiency of MDA5 by guest on September 23, 2021 an upstream molecule of IPS-1, also suppresses HBV replication. further strengthens the notion that MDA5 plays a critical role in Together, these results provide the complete RLR pathway me- innate immunity to HBV replication. Given that innate immunity diating the innate response to HBV infection. Importantly, our dictates adaptive immunity and that our study shows the impor- data showing that MDA5 selectively associates with both HBV tance of MDA5 in innate immunity to HBV infection, MDA5 RNAs and DNAs, and subsequently initiates the downstream likely plays a role in regulating adaptive immunity to HBV infec- signaling pathway provide one more piece of evidence on the tion as well. Supporting this notion, MDA52/2 mice have a re- nonredundant role of MDA5 and RIG-I in virus recognition. duced number of Ag-specific CD8+ cells during acute infection of In this study, we demonstrated that MDA5 not only recognizes lymphocytic choriomeningitis virus, resulting in persistent infec- HBV nucleic acids, but also initiates a signaling pathway that leads tion (67). The expression and/or signaling of MDA5 may be to the activation of transcription factors IRF3 and NF-kB to induce critical for determining the outcome of HBV infection. When the expression of ISGs-MxA, OAS1, and CXCL10. Among these MDA5 expression and/or signaling are robust upon HBV infec- ISGs, MxA and OAS1 are antiviral genes, whereas CXCL10 is tion, hosts should be able to efficiently eliminate the virus. In a chemokine gene responsible for the recruitment of effector CD8 contrast, hosts may not efficiently clear the virus if MDA5 ex- T cells and NK cells to the site of viral infection. MxA likely plays pression and/or signaling are attenuated, eventually leading to an important role in HBV infection because it has been shown chronic HBV infection. In this regard, genetic polymorphism of to inhibit HBV replication (39, 41, 42). Our results show that MDA5 may be one of the factors accounting for the variation in an knockdown of endogenous MxA reversed the inhibition of HBV individual’s susceptibility to HBV infection and outcome of HBV replication mediated by MDA5 signaling (Fig. 5B). It is plausible infection. Some MDA5 variants that affect MDA5 expression are that induction of MxA expression during HBV infection is likely significantly associated with (68, 69). Because via the MDA5-mediated innate signaling pathway. this study demonstrates the importance of MDA5 in suppressing Interestingly, we found that unlike other ISGs (MxA, OAS1, and HBV replication, one would expect that MDA5 variants with high CXCL10), the induction of IFN-b seems not evident in cells MDA5 expression may control HBV infection, whereas MDA5 cotransfected with the HBV replicative plasmid and the MDA5 or variants with low MDA5 expression may lead to chronic HBV RIG-I expression plasmid at 24 h posttransfection, although the infection. Thus, the association of MDA5 genetic polymorphisms IFN-b induction in cells cotransfected with the HBV replicative with the outcome of HBV infection should be investigated. plasmid and RIG-I expression plasmid is somewhat increased as compared with cells cotransfected with the HBV replicative Acknowledgments plasmid and MDA5 expression plasmid. This observation is We are grateful to Dr. Chiaho Shih for providing pSV2ANeo-HBVx2 plas- reminiscent of the clinical observation that type I IFN induction is mid and pSV2ANeo-HBV mutant 2310, Dr. Steve S.-L. Chen for providing not always observed in patients infected with HBV (61–63). The pGAGGS-MCS-Flag2 plasmid, Dr. Takashi Fujita for providing pEF-BOS- The Journal of Immunology 3275

2 RIG-I plasmid, Dr. Shizuo Akira for providing breeding pairs of MDA5+/ 26. Yang, P. L., A. Althage, J. Chung, and F. V. Chisari. 2002. Hydrodynamic in- mice and RIG-I+/2 mice, Drs. Mei-Yi Lu and Kuo-Ming Lee for technical jection of viral DNA: a mouse model of acute hepatitis B virus infection. Proc. Natl. Acad. Sci. USA 99: 13825–13830. consulting, and Dr. Pauline Yen for critical reading of the manuscript. 27. Huang, L. R., H. L. Wu, P. J. Chen, and D. S. Chen. 2006. An immunocompetent mouse model for the tolerance of human chronic hepatitis B virus infection. Proc. Natl. Acad. Sci. USA 103: 17862–17867. Disclosures 28. Ketzinel-Gilad, M., A. Zauberman, O. Nussbaum, Y. Shoshany, O. Ben-Moshe, The authors have no financial conflicts of interest. O. Pappo, Y. Felig, E. Ilan, H. Wald, S. Dagan, and E. Galun. 2006. The use of the hydrodynamic HBV animal model to study HBV biology and anti-viral therapy. Hepatol. Res. 34: 228–237. References 29. Tian, Y., W. L. Chen, and J. H. Ou. 2011. Effects of -a/b on HBV replication determined by viral load. PLoS Pathog. 7: e1002159. 1. Guidotti, L. G., and F. V. Chisari. 2006. Immunobiology and pathogenesis of 30. Huang, L. R., Y. A. Gabel, S. Graf, S. Arzberger, C. Kurts, M. Heikenwalder, P. viral hepatitis. Annu. Rev. Pathol. 1: 23–61. A. Knolle, and U. Protzer. 2012. Transfer of HBV genomes using low doses of 2. Rehermann, B., and M. Nascimbeni. 2005. Immunology of hepatitis B virus and adenovirus vectors leads to persistent infection in immune competent mice. hepatitis C virus infection. Nat. Rev. Immunol. 5: 215–229. Gastroenterology 142:1447–1450.e3. 3. Beutler, B., C. Eidenschenk, K. Crozat, J. L. Imler, O. Takeuchi, J. A. Hoffmann, 31. Kato, H., S. Sato, M. Yoneyama, M. Yamamoto, S. Uematsu, K. Matsui, and S. Akira. 2007. Genetic analysis of resistance to viral infection. Nat. Rev. T. Tsujimura, K. Takeda, T. Fujita, O. Takeuchi, and S. Akira. 2005. Cell type- Immunol. 7: 753–766. specific involvement of RIG-I in antiviral response. Immunity 23: 19–28. 4. Stetson, D. B., and R. Medzhitov. 2006. Type I in host defense. 32. Kato, H., O. Takeuchi, S. Sato, M. Yoneyama, M. Yamamoto, K. Matsui, Immunity 25: 373–381. S. Uematsu, A. Jung, T. Kawai, K. J. Ishii, et al. 2006. Differential roles of 5. Takeuchi, O., and S. Akira. 2009. Innate immunity to virus infection. Immunol. MDA5 and RIG-I in the recognition of RNA viruses. Nature 441: 101– Rev. 227: 75–86. 105. 6. Akira, S., S. Uematsu, and O. Takeuchi. 2006. Pathogen recognition and innate 33. Shih, C. H., L. S. Li, S. Roychoudhury, and M. H. Ho. 1989. In vitro propagation immunity. Cell 124: 783–801. of human hepatitis B virus in a rat hepatoma cell line. Proc. Natl. Acad. Sci. USA 7. Yamamoto, M., S. Sato, H. Hemmi, K. Hoshino, T. Kaisho, H. Sanjo, 86: 6323–6327. Downloaded from O. Takeuchi, M. Sugiyama, M. Okabe, K. Takeda, and S. Akira. 2003. Role of 34. Roychoudhury, S., and C. Shih. 1990. cis rescue of a mutated reverse tran- adaptor TRIF in the MyD88-independent toll-like receptor signaling pathway. scriptase gene of human hepatitis B virus by creation of an internal ATG. J. Science 301: 640–643. Virol. 64: 1063–1069. 8. Beutler, B., Z. Jiang, P. Georgel, K. Crozat, B. Croker, S. Rutschmann, X. Du, 35. Roychoudhury, S., A. F. Faruqi, and C. Shih. 1991. Pregenomic RNA encapsi- and K. Hoebe. 2006. Genetic analysis of host resistance: Toll-like receptor dation analysis of eleven missense and nonsense polymerase mutants of human signaling and immunity at large. Annu. Rev. Immunol. 24: 353–389. hepatitis B virus. J. Virol. 65: 3617–3624. 9. Kawai, T., K. Takahashi, S. Sato, C. Coban, H. Kumar, H. Kato, K. J. Ishii, 36. Yuan, T. T., A. Faruqi, J. W. Shih, and C. Shih. 1995. The mechanism of natural

O. Takeuchi, and S. Akira. 2005. IPS-1, an adaptor triggering RIG-I- and Mda5- occurrence of two closely linked HBV precore predominant mutations. Virology http://www.jimmunol.org/ mediated type I interferon induction. Nat. Immunol. 6: 981–988. 211: 144–156. 10. Seth, R. B., L. Sun, C.-K. Ea, and Z. J. Chen. 2005. Identification and charac- 37. Gitlin, L., W. Barchet, S. Gilfillan, M. Cella, B. Beutler, R. A. Flavell, terization of MAVS, a mitochondrial antiviral signaling protein that activates M. S. Diamond, and M. Colonna. 2006. Essential role of mda-5 in type I IFN NF-kappaB and IRF 3. Cell 122: 669–682. responses to polyriboinosinic:polyribocytidylic acid and encephalomyocarditis 11. Meylan, E., J. Curran, K. Hofmann, D. Moradpour, M. Binder, R. Bartenschlager, picornavirus. Proc. Natl. Acad. Sci. USA 103: 8459–8464. and J. Tschopp. 2005. Cardif is an adaptor protein in the RIG-I antiviral pathway 38. Yoneyama, M., M. Kikuchi, K. Matsumoto, T. Imaizumi, M. Miyagishi, K. Taira, and is targeted by hepatitis C virus. Nature 437: 1167–1172. E. Foy, Y. M. Loo, M. Gale, Jr., S. Akira, et al. 2005. Shared and unique 12. Xu, L. G., Y. Y. Wang, K. J. Han, L. Y. Li, Z. Zhai, and H. B. Shu. 2005. VISA is functions of the DExD/H-box helicases RIG-I, MDA5, and LGP2 in antiviral an adapter protein required for virus-triggered IFN-beta signaling. Mol. Cell 19: innate immunity. J. Immunol. 175: 2851–2858. 727–740. 39. Li, N., L. Zhang, L. Chen, W. Feng, Y. Xu, F. Chen, X. Liu, Z. Chen, and W. Liu. 13. Isogawa, M., M. D. Robek, Y. Furuichi, and F. V. Chisari. 2005. Toll-like re- 2012. MxA inhibits hepatitis B virus replication by interaction with hepatitis B ceptor signaling inhibits hepatitis B virus replication in vivo. J. Virol. 79: 7269– core antigen. Hepatology 56: 803–811. by guest on September 23, 2021 7272. 40. Meng, Z., S. Qiu, X. Zhang, J. Wu, T. Schreiter, Y. Xu, D. Yang, M. Roggendorf, 14. Wu, J., M. Lu, Z. Meng, M. Trippler, R. Broering, A. Szczeponek, F. Krux, J. Schlaak, and M. Lu. 2009. Inhibition of woodchuck hepatitis virus gene ex- U. Dittmer, M. Roggendorf, G. Gerken, and J. F. Schlaak. 2007. Toll-like pression in primary hepatocytes by siRNA enhances the cellular gene expression. receptor-mediated control of HBV replication by nonparenchymal liver cells Virology 384: 88–96. in mice. Hepatology 46: 1769–1778. 41. Peltekian, C., E. Gordien, F. Garreau, V. Meas-Yedid, P. Soussan, V. Willams, 15. Guo, H., D. Jiang, D. Ma, J. Chang, A. M. Dougherty, A. Cuconati, T. M. Block, M. L. Chaix, J. C. Olivo-Marin, C. Bre´chot, and D. Kremsdorf. 2005. Human and J. T. Guo. 2009. Activation of pattern recognition receptor-mediated innate MxA protein participates to the interferon-related inhibition of hepatitis B virus immunity inhibits the replication of hepatitis B virus in human hepatocyte- replication in female transgenic mice. J. Hepatol. 43: 965–972. derived cells. J. Virol. 83: 847–858. 42. Gordien, E., O. Rosmorduc, C. Peltekian, F. Garreau, C. Bre´chot, and 16. Wang, X., Y. Li, A. Mao, C. Li, Y. Li, and P. Tien. 2010. Hepatitis B virus X D. Kremsdorf. 2001. Inhibition of hepatitis B virus replication by the interferon- protein suppresses virus-triggered IRF3 activation and IFN-beta induction by inducible MxA protein. J. Virol. 75: 2684–2691. disrupting the VISA-associated complex. Cell. Mol. Immunol. 7: 341–348. 43. Yoneyama, M., M. Kikuchi, T. Natsukawa, N. Shinobu, T. Imaizumi, 17. Wei, C., C. Ni, T. Song, Y. Liu, X. Yang, Z. Zheng, Y. Jia, Y. Yuan, K. Guan, M. Miyagishi, K. Taira, S. Akira, and T. Fujita. 2004. The RNA RIG-I Y. Xu, et al. 2010. The hepatitis B virus X protein disrupts innate immunity by has an essential function in double-stranded RNA-induced innate antiviral downregulating mitochondrial antiviral signaling protein. J. Immunol. 185: responses. Nat. Immunol. 5: 730–737. 1158–1168. 44. Berke, I. C., and Y. Modis. 2012. MDA5 cooperatively forms dimers and ATP- 18. Kumar, M., S. Y. Jung, A. J. Hodgson, C. R. Madden, J. Qin, and B. L. Slagle. sensitive filaments upon binding double-stranded RNA. EMBO J. 31: 1714– 2011. Hepatitis B virus regulatory HBx protein binds to adaptor protein IPS-1 1726. and inhibits the activation of beta interferon. J. Virol. 85: 987–995. 45. Peisley, A., C. Lin, B. Wu, M. Orme-Johnson, M. Liu, T. Walz, and S. Hur. 2011. 19. Jiang, J., and H. Tang. 2010. Mechanism of inhibiting type I interferon induction Cooperative assembly and dynamic disassembly of MDA5 filaments for viral by hepatitis B virus X protein. Protein Cell 1: 1106–1117. dsRNA recognition. Proc. Natl. Acad. Sci. USA 108: 21010–21015. 20. Yu, S., J. Chen, M. Wu, H. Chen, N. Kato, and Z. Yuan. 2010. Hepatitis B virus 46. Rehwinkel, J., C. P. Tan, D. Goubau, O. Schulz, A. Pichlmair, K. Bier, N. Robb, polymerase inhibits RIG-I- and Toll-like receptor 3-mediated beta interferon F. Vreede, W. Barclay, E. Fodor, and C. Reis e Sousa. 2010. RIG-I detects viral induction in human hepatocytes through interference with interferon regulatory genomic RNA during negative-strand RNA virus infection. Cell 140: 397–408. factor 3 activation and dampening of the interaction between TBK1/IKKepsilon 47. Hornung, V., J. Ellegast, S. Kim, K. Brzo´zka, A. Jung, H. Kato, H. Poeck, and DDX3. J. Gen. Virol. 91: 2080–2090. S. Akira, K. K. Conzelmann, M. Schlee, et al. 2006. 59-Triphosphate RNA is the 21. Wang, H., and W. S. Ryu. 2010. Hepatitis B virus polymerase blocks pattern ligand for RIG-I. Science 314: 994–997. recognition receptor signaling via interaction with DDX3: implications for im- 48. Plumet, S., F. Herschke, J. M. Bourhis, H. Valentin, S. Longhi, and D. Gerlier. mune evasion. PLoS Pathog. 6: e1000986. 2007. Cytosolic 59-triphosphate ended viral leader transcript of measles virus as 22. Johnson, C. L., and M. Gale, Jr. 2006. CARD games between virus and host get activator of the RIG I-mediated interferon response. PLoS ONE 2: e279. a new player. Trends Immunol. 27: 1–4. 49. Kato, H., O. Takeuchi, E. Mikamo-Satoh, R. Hirai, T. Kawai, K. Matsushita, 23. Chang, C. M., K. S. Jeng, C. P. Hu, S. J. Lo, T. S. Su, L. P. Ting, C. K. Chou, A. Hiiragi, T. S. Dermody, T. Fujita, and S. Akira. 2008. Length-dependent S. H. Han, E. Pfaff, J. Salfeld, et al. 1987. Production of hepatitis B virus in vitro recognition of double-stranded ribonucleic acids by retinoic acid-inducible by transient expression of cloned HBV DNA in a hepatoma cell line. EMBO J. 6: gene-I and melanoma differentiation-associated gene 5. J. Exp. Med. 205: 675–680. 1601–1610. 24. Sells, M. A., M. L. Chen, and G. Acs. 1987. Production of hepatitis B virus 50. Schlee, M., A. Roth, V. Hornung, C. A. Hagmann, V. Wimmenauer, W. Barchet, particles in Hep G2 cells transfected with cloned hepatitis B virus DNA. Proc. C. Coch, M. Janke, A. Mihailovic, G. Wardle, et al. 2009. Recognition of 59 Natl. Acad. Sci. USA 84: 1005–1009. triphosphate by RIG-I helicase requires short blunt double-stranded RNA as 25. Yaginuma, K., Y. Shirakata, M. Kobayashi, and K. Koike. 1987. Hepatitis B contained in panhandle of negative-strand virus. Immunity 31: 25–34. virus (HBV) particles are produced in a cell culture system by transient ex- 51. Pichlmair, A., O. Schulz, C. P. Tan, J. Rehwinkel, H. Kato, O. Takeuchi, pression of transfected HBV DNA. Proc. Natl. Acad. Sci. USA 84: 2678–2682. S. Akira, M. Way, G. Schiavo, and C. Reis e Sousa. 2009. Activation of MDA5 3276 THE MDA5 SIGNALING PATHWAY SUPPRESSES HBV REPLICATION

requires higher-order RNA structures generated during virus infection. J. Virol. hepatitis B virus: an intermediate of covalently closed circular DNA formation. 83: 10761–10769. J. Virol. 81: 12472–12484. 52. Samanta, M., D. Iwakiri, T. Kanda, T. Imaizumi, and K. Takada. 2006. EB virus- 61. Ikeda, T., A. M. Lever, and H. C. Thomas. 1986. Evidence for a deficiency of encoded RNAs are recognized by RIG-I and activate signaling to induce type I interferon production in patients with chronic hepatitis B virus infection acquired IFN. EMBO J. 25: 4207–4214. in adult life. Hepatology 6: 962–965. 53. Samanta, M., D. Iwakiri, and K. Takada. 2008. Epstein-Barr virus-encoded small 62. Zachoval, R., J. Abb, V. Zachoval, J. Eisenburg, G. R. Pape, G. Paumgartner, and RNA induces IL-10 through RIG-I-mediated IRF-3 signaling. Oncogene 27: F. Deinhardt. 1988. Interferon alpha in hepatitis type B and non-A, non-B. 4150–4160. Defective production by peripheral blood mononuclear cells in chronic infec- 54. Delaloye, J., T. Roger, Q. G. Steiner-Tardivel, D. Le Roy, M. Knaup Reymond, tion and development of serum interferon in acute disease. J. Hepatol. 6: 364– S. Akira, V. Petrilli, C. E. Gomez, B. Perdiguero, J. Tschopp, et al. 2009. Innate 368. immune sensing of modified vaccinia virus Ankara (MVA) is mediated by TLR2- 63. Dunn, C., D. Peppa, P. Khanna, G. Nebbia, M. Jones, N. Brendish, R. M. Lascar, TLR6, MDA-5 and the NALP3 inflammasome. PLoS Pathog. 5: e1000480. D. Brown, R. J. Gilson, R. J. Tedder, et al. 2009. Temporal analysis of early 55. Choi, M. K., Z. Wang, T. Ban, H. Yanai, Y. Lu, R. Koshiba, Y. Nakaima, immune responses in patients with acute hepatitis B virus infection. Gastroen- S. Hangai, D. Savitsky, M. Nakasato, et al. 2009. A selective contribution of the terology 137: 1289–1300. RIG-I-like receptor pathway to type I interferon responses activated by cytosolic 64. Dixit, E., S. Boulant, Y. Zhang, A. S. Lee, C. Odendall, B. Shum, N. Hacohen, DNA. Proc. Natl. Acad. Sci. USA 106: 17870–17875. Z. J. Chen, S. P. Whelan, M. Fransen, et al. 2010. Peroxisomes are signaling 56. Melchjorsen, J., J. Rintahaka, S. Søby, K. A. Horan, A. Poltajainen, platforms for antiviral innate immunity. Cell 141: 668–681. L. Østergaard, S. R. Paludan, and S. Matikainen. 2010. Early innate recognition 65. Moriyama, K., K. Hayashida, M. Shimada, S. Nakano, Y. Nakashima, and of herpes simplex virus in human primary macrophages is mediated via the Y. Fukumaki. 2003. Antisense RNAs transcribed from the upstream region of the MDA5/MAVS-dependent and MDA5/MAVS/RNA polymerase III-independent precore/core promoter of hepatitis B virus. J. Gen. Virol. 84: 1907–1913. pathways. J. Virol. 84: 11350–11358. 66. Standring, D. N., L. B. Rall, O. Laub, and W. J. Rutter. 1983. Hepatitis B virus 57. Ablasser, A., F. Bauernfeind, G. Hartmann, E. Latz, K. A. Fitzgerald, and encodes an RNA polymerase III transcript. Mol. Cell. Biol. 3: 1774–1782. V. Hornung. 2009. RIG-I-dependent sensing of poly(dA:dT) through the in- 67. Wang, Y., M. Swiecki, M. Cella, G. Alber, R. D. Schreiber, S. Gilfillan, and duction of an RNA polymerase III-transcribed RNA intermediate. Nat. Immunol. M. Colonna. 2012. Timing and magnitude of type I interferon responses by 10: 1065–1072. distinct sensors impact CD8 T cell exhaustion and chronic viral infection. Cell 58. Chiu, Y. H., J. B. Macmillan, and Z. J. Chen. 2009. RNA polymerase III detects Host Microbe 11: 631–642. Downloaded from cytosolic DNA and induces type I interferons through the RIG-I pathway. Cell 68. Liu, S., H. Wang, Y. Jin, R. Podolsky, M. V. Reddy, J. Pedersen, B. Bode, 138: 576–591. J. Reed, D. Steed, S. Anderson, et al. 2009. IFIH1 polymorphisms are signifi- 59. Kidd-Ljunggren, K., M. Zuker, I. L. Hofacker, and A. H. Kidd. 2000. The cantly associated with type 1 diabetes and IFIH1 gene expression in peripheral hepatitis B virus pregenome: prediction of RNA structure and implications for blood mononuclear cells. Hum. Mol. Genet. 18: 358–365. the emergence of deletions. Intervirology 43: 154–164. 69. Nejentsev, S., N. Walker, D. Riches, M. Egholm, and J. A. Todd. 2009. Rare 60. Guo, H., D. Jiang, T. Zhou, A. Cuconati, T. M. Block, and J. T. Guo. 2007. variants of IFIH1, a gene implicated in antiviral responses, protect against type 1 Characterization of the intracellular deproteinized relaxed circular DNA of diabetes. Science 324: 387–389. http://www.jimmunol.org/ by guest on September 23, 2021