Negative Regulation of Nmi on Virus-Triggered Type I IFN Production by Targeting IRF7

This information is current as Jie Wang, Bo Yang, Yu Hu, Yuhan Zheng, Haiyan Zhou, of September 29, 2021. Yanming Wang, Yonglei Ma, Kairui Mao, Leilei Yang, Guomei Lin, Yongyong Ji, Xiaodong Wu and Bing Sun J Immunol published online 16 August 2013 http://www.jimmunol.org/content/early/2013/08/16/jimmun

ol.1300740 Downloaded from

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 29, 2021 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. Published August 16, 2013, doi:10.4049/jimmunol.1300740 The Journal of Immunology

Negative Regulation of Nmi on Virus-Triggered Type I IFN Production by Targeting IRF7

Jie Wang,*,1 Bo Yang,†,1 Yu Hu,* Yuhan Zheng,* Haiyan Zhou,* Yanming Wang,† Yonglei Ma,‡ Kairui Mao,* Leilei Yang,‡ Guomei Lin,* Yongyong Ji,* Xiaodong Wu,* and Bing Sun*,†,‡

Viral infection causes host cells to produce type I IFNs, which play a critical role in viral clearance. IFN regulatory factor (IRF) 7 is the master regulator of type I IFN-dependent immune responses. In this article, we report that N- and STATs interactor (Nmi), a Sendai virus–inducible protein, interacted with IRF7 and inhibited virus-triggered type I IFN production. The overexpression of Nmi inhibited the Sendai virus–triggered induction of type I IFNs, whereas the knockdown of Nmi promoted IFN production. Furthermore, the enhanced production of IFNs resulting from Nmi knockdown was sufficient to protect cells from infection by vesicular stomatitis virus. In addition, Nmi was found to promote the K48-linked ubiquitination of IRF7 and the proteasome- Downloaded from dependent degradation of this protein. Finally, an impairment of antiviral responses is also detectable in Nmi-transgenic mice. These findings suggest that Nmi is a negative regulator of the virus-triggered induction of type I IFNs that targets IRF7. The Journal of Immunology, 2013, 191: 000–000.

iral infection triggers innate immune responses, leading ruses, such as Newcastle disease virus, vesicular stomatitis virus http://www.jimmunol.org/ to the production of type I IFNs. Type I IFNs, including (VSV), influenza virus, and Sendai virus (SeV), and infection by V IFN-b and many IFN-a species, are essential in induc- positive-sense ssRNA viruses, such as Japanese encephalitis virus ing host antiviral responses (1–3). Host cells recognize viral in- (6). Activated RIG-I and MDA5 are predicted to interact with vasion by at least two distinct mechanisms. One mechanism is IFN-b promoter stimulator 1 (IPS-1) via a caspase recruitment through TLR 3, and the other is through retinoic acid–inducible domain–caspase recruitment domain interaction, which induces -I (RIG-I) and melanoma differentiation–associated gene-5 the recruitment of downstream TNFR-associated factor (TRAF) (MDA5). Both of these mechanisms involve host pathogen rec- family members (7–11). TRAF3 is essential for the IPS-1–medi- ognition receptors (4, 5). RIG-I and MDA5, two RNA helicase ated activation of type I IFNs, whereas TRAF6 is essential for the proteins, are cytoplasmic viral RNA sensors. RIG-I–deficient mice activation of NF-kB (12, 13). Signaling downstream of TRAF3 by guest on September 29, 2021 are defective for the induction of type I IFNs and proinflammatory involves the TRAF family member–associated NF-kB activator– in response to infection by negative-sense ssRNA vi- binding kinase 1 (TBK1) and inducible IkB kinase (IKK-i or IKK- ε). The activation of TBK1 and IKK-i results in the phosphory- *State Key Laboratory of Cell Biology, Institute of Biochemistry and Cell Biology, lation and subsequent dimerization and nuclear translocation of Shanghai Institutes of Biological Sciences, Chinese Academy of Sciences, Shanghai the transcription factors IFN regulatory factor (IRF) 3 and IRF7 † 200031, People’s Republic of China; School of Life Sciences, University of Science (14–16). IRF7 is a master regulator of type I IFN–dependent and Technology of China, Hefei, Anhui 230026, People’s Republic of China; and ‡Molecular Virus Unit, Key Laboratory of Molecular Virology and Immunology, immune responses. The viral induction of MyD88 (myeloid dif- Institute Pasteur of Shanghai, Chinese Academy of Sciences, Shanghai 200025, ferentiation primary response gene 88)–independent IFN-a/b People’s Republic of China is severely impaired in Irf72/2 fibroblasts. Irf72/2 mice are more 1 J.W. and B.Y. contributed equally to this work. vulnerable to viral infection and have marked lower serum IFN Received for publication March 18, 2013. Accepted for publication July 14, 2013. levels (17). This work was supported by grants from the National Key Project of 973 Although it is essential for activating the innate immune re- (2013CB530504), the National Natural Science Foundation of China (31230024, sponse and enhancing adaptive immunity against viruses, the host 31030029, 31100662, 91213303), the National Ministry of Science and Technology (2007DFC31700), the National 863 Project (2012AA020103), the National Science antiviral response must be tightly controlled to prevent harmful and Technology Major Project (2012ZX10002007-003, 2013ZX10004-101-005, effects resulting from excessive activation. Several molecules have 2011ZX10004-001), and the Sanofi-Aventis–Shanghai Institutes for Biological Sci- ences Discovery Innovation Grant. been reported to regulate the virus-triggered signaling pathway, including A20, Pin1, SIKE, RNF125, DUBA, DAK, TBK1s, MIP- Address correspondence and reprint requests to Dr. Bing Sun and Dr. Xiaodong Wu, State Key Laboratory of Cell Biology, Institute of Biochemistry and Cell Biology, T3, Trim28, NLRX1, ABIN1, ARL16, PIASy, and GBP4 (18–25). Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yue All of these molecules target distinct components using specific Yang Road, Shanghai 200031, China. E-mail addresses: [email protected] (B.S.) and [email protected] (X.W.) mechanisms. For example, Pin1 is associated with the ubiquitin- mediated degradation of IRF3, whereas TBK1s disrupts the in- Abbreviations used in this article: BALF, bronchoalveolar lavage fluid; BMDC, bone marrow–derived dendritic cell; HA, hemagglutinin; IFP, IFN-induced protein; IKK-i, teraction between RIG-I and IPS-1. IKK-ε, inducible IkB kinase; IPS-1, IFN-b promoter stimulator 1; IRF, IFN regula- N-Myc and STATs interactor (Nmi) is an IFN-a–inducible pro- tory factor; ISRE, IFN-stimulated regulatory element; MDA5, melanoma differenti- ation–associated gene 5; Nmi, N-Myc and STATs interactor; RIG-I, retinoic acid– tein (26) that interacts with C-Myc, N-Myc, Max, Sox-10, IFN- inducible gene I; SeV, Sendai virus; siRNA, small interfering RNA; TBK1, TRAF induced protein (IFP) 35, TIP60, CKIP-1, and all of the STAT family member–associated NF-kB activator–binding kinase 1; TRAF, TNFR- proteins except for STAT2 (27–35). Nmi shares 25% identity and associated factor; VSV, vesicular stomatitis virus. 46% homology over a region of ∼180 aa with its family member Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 IFN-induced protein (IFP) 35, which has been reported to play an

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1300740 2 Nmi NEGATIVELY REGULATES TYPE I IFN PRODUCTION antiviral role through association with the bovine Tas regulatory RT and real-time PCR protein of bovine foamy virus (36). However, the biological func- Total RNA was extracted from cultured cells with TRIzol (Invitrogen) tion of Nmi has not been thoroughly investigated. One group according to the manufacturer’s instructions. Oligo(dT) primers and M- suggested that Nmi inhibits Wnt/b-catenin signaling by reducing MLV reverse transcriptase (Invitrogen) were used for the reverse tran- b-catenin levels through proteosome-dependent degradation (37). scription of purified RNA. All of the gene transcripts were quantified by In this article, we reported that Nmi, which could be induced by real-time PCR with SYBR Green QPCR Master Mix and a 7900HT Fast Real-Time PCR System (Applied Biosystems). The relative fold induction SeV infection, interacted with IRF7 and inhibited SeV-induced was calculated with the 22DD CT method. Primers for real-time PCR are type I IFN expression and cellular antiviral responses. Our findings listed as follows: revealed the existence of a negative feedback mechanism mediated Hypoxanthine phosphoribosyltransferase sp 59-GCCCTTGACTATAA- by Nmi that regulates cellular antiviral responses. TGAG-39, hypoxanthine phosphoribosyltransferase as 59-GATAAGCGA- CAATCTACC-39; IFN-a sp 59-ATGGCTAGRCTCTGTGCTTTCCT-39, IFN-a as 59-AGGGCTCTCCAGAYTTCTGCTCTG-39;IFN-b sp 59- Materials and Methods CGTTCCTGCTGTGCTTCT-39,IFN-b as 59-GCATCTTCTCCGTCA- TCT-39; IL-6 sp 59-GAGGATACCACTCCCAACAGACC-39, IL-6 as 59- Plasmid constructs and reagents AAGTGCATCATCGTTGTTCATACA-39; TNF-a sp 59-CGTAGGCGAT- The mouse Nmi sequence was amplified by PCR using cDNA from SeV- TACAGTCACGG-39, TNF-a as 59-GACCAGGCTGTCGCTACATCA-39. infected bone marrow–derived dendritic cells (BMDCs) and confirmed by sequencing. The PCR primers were based on the National Center ELISA for Biotechnology Information GenBank sequence (accession number The levels of IFN-a,IFN-b, IL-6, and TNF-a in the supernatants were NM_001141949). The open reading frame of Nmi was cloned into the determined using IFN-a (PBL), IFN-b (PBL), IL-6 (R&D Systems), TNF-a pcDNA3.0-hemagglutinin (HA) vector (Invitrogen) and the pcDNA3.0- (R&D Systems) ELISA kits according to the manufacturer’s instructions. Downloaded from Myc vector (Invitrogen). Nmi deletion mutants were cloned into the pcDNA3.0-Myc vector. HA-K48-ubi and HA-K63-ubi plasmids were Viral infections and plaque assays kindly provided by Hongbing Shu (Wuhan University, Wuhan, China). Other plasmids were generated or obtained as described before (20). HEK293T cells, RAW264.7 cells, and BMDCs were incubated with SeV Polyclonal Ab to Nmi (anti-Nmi) was generated by immunization of rabbits in serum-free medium for 2 h, and then the same volume of medium with full-length Nmi expressed by Escherichia coli. Polyclonal Ab to IRF7 containing 20% FBS was added. Post infection, the cells were collected (anti-IRF7) was generated as described previously (20). SeV was provided and used for various experiments. L929 cells were used in the plaque assay. by Wuhan Institutes of Virology, Chinese Academy of Sciences (Shanghai, Monolayers of L929 cells were infected with 10-fold dilutions of virus http://www.jimmunol.org/ China). stock for 2 h and then overlaid with 0.8% agarose in RPMI 1640 containing 1% FBS. After incubation for 48 h at 37˚C and 5% CO2 in a humidified Cell culture and transfections incubator, the cells were fixed with 4% paraformaldehyde and then stained with 1% crystal violet. The numbers of plaques were counted to calculate HEK293T and RAW264.7 cells were cultured in DMEM. L929 cells were the viral titer. grown in RPMI 1640. All of the media were supplemented with 10% FBS (Invitrogen), 4 mM L-glutamine, 100 U/ml penicillin, and 100 U/ml strep- IFN bioassay tomycin under humidified conditions with 5% CO at 37˚C. BMDCs were 2 The IFN bioassay was performed as described previously (20). In brief, generated as previously described (38). HEK293T and RAW264.7 cells were supernatants from SeV-infected RAW264.7 cells were overlaid onto L929 transfected with Lipofectamine (Invitrogen) or Lipofectamine 2000 (Invitro- cells seeded in a 48-well plate. After 24 h, the L929 cells were infected with by guest on September 29, 2021 gen) according to the manufacturer’s instructions. VSV at a multiplicity of infection of 0.1 for 2 h in serum-free medium and then washed twice to remove resident VSV before replacing the medium. RNA interference At 24 h post infection, the supernatants were subjected to plaque assays. Mouse Nmi ON-TARGET plus SMARTpool small interfering RNA (siRNA) Mice (catalog no. L-044766-01-0005) and the negative control siRNA (catalog no. D-001206-14-20) were obtained from Thermo Scientific. RAW264.7 C57BL/6 mice were obtained from The Jackson Laboratory. Nmi-transgenic cells were transfected with siRNA, using Lipofectamine 2000 according mice were obtained from Model Animal Research Center, Nanjing Uni- to the manufacturer’s instructions. At 24 h after transfection, the cells were versity (Nanjing, Jiangsu, China). The mice were initially created on the used for further experiments. 129 3 C57BL/6 background, and then extensively backcrossed to C57BL/6. The Nmi-transgenic effect was confirmed by real-time PCR and immu- Reporter gene assay noblot. All the mice were maintained under specific pathogen–free con- ditions at the Animal Care Facility of the Chinese Academy of Sciences HEK293T and RAW264.7 cells were transfected with reporter plasmids and were handled in compliance with the Institute of Biochemistry and and a Renilla luciferase plasmid as an internal control plus the indicated Cell Biology’s guidelines. expression plasmids. The total DNA concentration was kept constant by making up the difference with the empty pcDNA3.0 vector. At 24 h after Infection of mice transfection, the cells were lysed in passive lysis buffer or infected with SeV for a further 24 h and then lysed. The luciferase activity in the lysates A group of 6-wk-old Nmi-transgenic or wild-type mice were anesthetized was analyzed with a dual luciferase reporter assay system (Promega). with pentobarbital sodium salt (Genview) and infected intranasally with 5 3 105 PFU virus in 50 ml. Mice were sacrificed at the indicated days; then Immunoblotting and coimmunoprecipitation lung tissue and bronchoalveolar lavage fluid (BALF) were collected for further experiments. For BALF, the left lung was ligated, the trachea was Immunoprecipitation and immunoblot analysis were performed as de- cannulated, and the right lung was lavaged twice with 500 ul PBS. The scribed previously (39). Briefly, HEK293T cells were transfected with fluid was collected by gentle aspiration and then centrifuged for 10 min at various combinations of plasmids. At 24 h after transfection, cell lysates 4˚C, and the supernatants were subjected to ELISA. were prepared in lysis buffer and incubated with the indicated Ab together with protein A/G Plus-Agarose Immunoprecipitation Reagent (Santa Statistics Cruz Biotechnology) at 4˚C for 3 h or overnight. After three washes, the All of the data are presented as the mean 6 SD from at least three inde- immunoprecipitates were boiled in SDS sample buffer for 10 min and pendent experiments. Statistical comparisons between different treatments analyzed by immunoblotting. For ubiquitination, HEK293T cells were were made using the unpaired Student t test, and p , 0.05 was considered transfected with GFP-IRF7 together with Myc-Nmi, HA-ubiquitin, and the indicated plasmids. Then, the proteins were immunoprecipitated with statistically significant. anti-GFP and immunoblotted with anti-HA and anti-GFP. For endogenous Accession codes coimmunoprecipitation experiments, RAW264.7 cells (2 3 107)were infected with SeV for 12 h, and cell lysates were prepared in lysis buffer For the Omnibus microarray data, GSE32499 (series), and incubated with 1 ml preimmune serum or antiserum against Nmi. The and GSM804441 and GSM804442 (specific array files) are available at subsequent procedures were performed as described above. www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE32499. The Journal of Immunology 3

Results vation, as shown in Fig. 1E. These results suggest that Nmi can Upregulation of Nmi expression by SeV infection downregulate SeV-triggered IFN-b activation. To investigate signaling during viral invasion, we performed a Knockdown of Nmi promotes type I IFN production cDNA microarray analysis using BMDCs infected with SeV and To determine whether Nmi is involved in the regulation of SeV- BMDCs that were not infected. We found that the expression of induced type I IFN production under physiological conditions, we Nmi mRNAwas higher at 4 h after SeV infection [GEO: microarray silenced endogenous Nmi expression in RAW264.7 cells, using data, GSE32499 (series), and GSM804441 and GSM804442]. siRNA. At 24 h post transfection, the cell lysates were collected To confirm the induction of Nmi by SeV, we generated a rabbit to analyze the protein level of endogenous Nmi. As shown in polyclonal Ab against recombinant Nmi. This Ab specifically Fig. 2A, the endogenous Nmi expression in RAW264.7 cells ∼ detected a band of 37 kDa in HEK293T cells that were trans- was suppressed by siRNA. After Nmi was silenced in the cells, fected with HA-tagged Nmi, but not in cells transfected with the SeV-induced IFN-a and IFN-b levels were increased at both the empty vector or HA-tagged IFP35, which is the only known mRNA and the protein levels (Fig. 2B, 2C). During viral in- family member of Nmi (Fig. 1A). We confirmed the upregulation fection, type I IFNs protect cells from viral invasion by inter- of Nmi expression by SeV infection in several cell types. BMDCs fering with the replication of viruses. To further confirm the and mouse macrophage RAW264.7 cells were infected with SeV, function of Nmi, the supernatants from SeV-infected RAW264.7 and the expression of the Nmi protein was detected by Western cells in which Nmi was silenced or were not subjected to an IFN blotting. As shown in Fig. 1B and Fig. 1C, the Nmi expression bioassay that measures the ability of IFN to block replication of in BMDCs (Fig. 1B) and RAW264.7 (Fig. 1C) increased as early VSV. The data showed that the supernatants from Nmi-silenced Downloaded from as 2 h after SeV infection. These results suggest that Nmi serves cells inhibited VSV replication more efficiently than did super- as an SeV-inducible protein and may play a role in viral infection natants collected from the control cells (Fig. 2D). The results in the cells. suggest that Nmi negatively regulates IFN production upon SeV Nmi overexpression suppresses IFN-b activation infection in cells. To determine the functions of Nmi, HEK293T cells overexpressing Nmi inhibits activation of type I IFN production by targeting Nmi were infected with SeV, and SeV-induced signaling, in- IRF7 http://www.jimmunol.org/ k cluding NF- B, IFN-stimulated regulatory element (ISRE), and To elucidate the molecular mechanisms responsible for the Nmi- b IFN- activity, was investigated in the cells. HEK293T cells were mediated inhibition of SeV-triggered IFN production, we first transfected with an Nmi-expressing plasmid (or vector control) used reporter gene assays to identify the target of Nmi. Because and different reporter genes, and after 24 h, the cells were infected SeV activates type I IFN production through a signaling pathway with SeV. Of interest, Nmi overexpression inhibited SeV-induced composed of RIG-I, IPS-1, TBK1,IRF3andIRF7,wedetermined b ISRE and IFN- reporter gene activation, whereas activation of whether Nmi affected this signaling pathway. Using a luciferase k the NF- B reporter gene was changed only slightly (Fig. 1D). We assay, we found that Nmi overexpression mainly inhibited RIG-I-, confirmed this result in RAW264.7 cells. Consistently, the over- IPS-1-, TBK1- and IRF7-induced ISRE reporter gene activa- by guest on September 29, 2021 expression of Nmi impaired SeV-induced ISRE reporter gene tion, but Nmi did not affect IRF3-induced activation (Fig. 3A), k activation, with no obvious effect on NF- B reporter gene acti- suggesting that Nmi functions specifically at or downstream of IRF7. The coexpression of Nmi suppressed IRF7-induced ISRE reporter activity in a dose-dependent manner (Fig. 3B). Fur-

FIGURE 1. Nmi is induced by SeV and inhibited virus-induced IFN-b FIGURE 2. Nmi knockdown increases virus-induced type I IFN pro- activation. (A) Immunoblot of lysates of HEK293T cells transfected with duction. (A) Knockdown of Nmi by siRNA in RAW264.7 cells. After HA-tagged Nmi, IFP35, or empty vector (Vec) analyzed with polyclonal transfection with control siRNA (sc) or Nmi-specific siRNA (ni) for 24 h, anti-Nmi, anti-HA, or anti–b-actin. (B and C) Immunoblot of lysates of the cell lysates were subjected to immunoblot analysis with anti-Nmi BMDCs (B) or RAW264.7 cells (C) infected with SeV for 0–24 h analyzed or anti–b-actin to determine the knockdown efficiency. (B and C)After with polyclonal anti-Nmi or anti–b-actin. (D and E) HEK293T (D) and transfection with sc or ni, RAW264.7 cells were treated with medium or RAW264.7 (E) cells were transfected with the indicated plasmids. After SeV, and the cell lysates were subjected to real-time PCR (B) or ELISA (C) transfection, the cells were treated with medium or SeV for 24 h, and the analysis. (D) IFN bioassay of conditioned media from siRNA-transfected cell lysates were subjected to a luciferase assay. Data are representative of RAW264.7 cells infected with SeV for 24 h. Data are representative of three independent experiments. The data are presented as the mean 6 SD three independent experiments. The data are presented as the mean 6 SD of duplicates. *p , 0.05, **p , 0.01. of duplicates. *p , 0.05. 4 Nmi NEGATIVELY REGULATES TYPE I IFN PRODUCTION thermore, IFN-a6, which is thought to be transcribed only by Domain mapping of the interactions between Nmi and IRF7 IRF7, was inhibited in a dose-dependent manner by the coex- To determine which region of IRF7 facilitates its interaction with pression of Nmi (Fig. 3B). We next sought to identify the - Nmi, various Flag-tagged IRF7 mutants, as described in Fig. 4A, tionship between Nmi and IRF7. The coimmunoprecipitation were expressed in HEK293T cells with HA-tagged Nmi, and assays were performed and the results revealed that Nmi inter- coimmunoprecipitation experiments were performed. As shown acted with IRF7 but not with IRF3 (Fig. 3C). IFP35, which is in Fig. 4B, IRF7D2 (aa 1–410), IRF7D3 (aa 133–457), and considered to be the only family member of Nmi, did not interact IRF7D4 (aa 238–457) coimmunoprecipitated with Nmi, whereas with IRF7 (Fig. 3D). Next, we addressed whether this interaction IRF7D1 (aa 1–237) did not, suggesting that IRF7 aa 238–410 occurred under physiological conditions. To test this hypothesis, might be essential for the association of IRF7 with Nmi. Then, we performed endogenous coimmunoprecipitation experiments. we constructed IRF7D5 (deletion of aa 238–410), and this muta- The results indicated that endogenous Nmi was associated with tion of IRF7 did not coimmunoprecipitate with Nmi (Fig. 4B). endogenous IRF7 in untransfected RAW264.7 cells, and this in- Similarly, we made a series of Nmi deletion mutants (Fig. 4C) teraction was more obvious in SeV-infected cells, in which IRF7 and determined which part was essential for the interaction of Nmi was up-regulated (Fig. 3E). All of these results suggest Nmi with IRF7. In transient transfection and coimmunoprecipitation interacts with IRF7 and inhibited IRF7 mediated type I IFN experiments, as shown in Fig. 4D, wide-type Nmi, NID1 (aa 98– production. 200), DNID2 (aa 1–200), and DCC (aa 98–314) interacted with HA-tagged IRF7, whereas CC (aa 1–103) did not. We concluded that the region between aa 98 and 200, which contains the first NID, is sufficient for the interaction, and the region between aa 1 Downloaded from and 103, which contains only the CC domain, is not able to in- teract with IRF7. Of interest, the DCC mutant, which contains both the first NID and the second NID, showed reduced binding ability with IRF7, compared with the NID1 mutants, suggesting http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 3. Nmi interacts with IRF7. (A) Nmi or empty vector (Vec) was cotransfected with the indicated plasmids into HEK293T cells. Cell lysates were subjected to a luciferase assay. (B) HEK293T cells were FIGURE 4. Identification of domains that mediate the interaction between cotransfected with the ISRE promoter or the IFN-a6 promoter and empty Nmi and IRF7. (A) Schematic representation of IRF7. (B) HEK293T cells vector or increasing amounts of Nmi (0.1 mg, 0.2 mg, and 0.4 mg). (C) were cotransfected with HA-tagged Nmi and the indicated Flag-tagged HEK293T cells were transfected with HA-tagged Nmi and empty vector plasmids. The cell lysates were immunoprecipitated with anti-Flag and or the indicated Flag-tagged plasmids. The cell lysates were immunopre- immunoblotted with anti-HA or anti-Flag. (C) Schematic representation cipitated (IP) with anti-Flag and immunoblotted (IB) with anti-HA or of Nmi. (D) HEK293T cells were transfected with HA-tagged IRF7 and anti-Flag. (D) HEK293T cells were transfected with Flag-tagged IRF7 the indicated Myc-tagged plasmids. The cell lysates were immunopre- and empty vector or the indicated HA-tagged plasmids. The cell lysates cipitated with anti-Myc and immunoblotted with anti-HA or anti-Myc. were immunoprecipitated with anti-HA and immunoblotted with anti- WCL, whole-cell lysates. The asterisks indicate the L chains and the HA or anti-Flag. (E) Interaction between endogenous Nmi and IRF7 in nonspecific bands. (E) HEK293T cells were transfected with the ISRE RAW264.7 cells infected with SeV for 12 h. The cell lysates were immu- promoter and Nmi or its mutants. After transfection, the cells were treated noprecipitated with anti-Nmi and immunoblotted with anti-Nmi or anti- with medium or SeV for 24 h, and the cell lysates were subjected to a IRF7. WCL, whole-cell lysates. Data are representative of three independent luciferase assay. Data are representative of three independent experi- experiments. Data are presented as the mean 6 SD of duplicates. *p , 0.05, ments. Data are presented as the mean 6 SD of duplicates. *p , 0.05, **p , 0.01. **p , 0.01. The Journal of Immunology 5 that the two NID domains may have different roles in mediating interaction. Then, we determined whether the interaction between Nmi and IRF7 was important for the inhibitory function of Nmi. We transfected HEK293T cells with the Nmi truncation mutants, as mentioned above, along with the ISRE reporter gene and in- fected the cells with SeV. As shown in Fig. 4E, the expression of NID1, DNID2, or DCC inhibited ISRE reporter gene activation, whereas the expression of CC had a slight effect on the activation. Taken together, these data suggest that the region of IRF7 har- boring aa 238–410 and the first NID domain of Nmi are important for their interaction, and this interaction is required for the Nmi- mediated inhibition of IRF7 function. Nmi targets IRF7 for ubiquitination and degradation When the plasmid expressing IRF7 was cotransfected with the empty vector or Nmi, Western blot analysis indicated weak ex- pression in the presence of Nmi (Fig. 5A). As a control, we cotransfected IRF3 and Nmi into cells and found that Nmi had no effect on IRF3 expression (Fig. 5A). These results indicated that Downloaded from the overexpression of Nmi caused the downregulation of IRF7, but not IRF3. Next, we addressed whether the reduction in endoge- nous Nmi expression levels affected the expression of IRF7. As shown in Fig. 5B, the knockdown of Nmi in RAW264.7 cells increased the expression levels of IRF7 after SeV infection. Fur-

thermore, the proteasome inhibitor MG132 reversed the Nmi- http://www.jimmunol.org/ mediated degradation of IRF7 (Fig. 5C), suggesting that a pro- FIGURE 5. Nmi increased the ubiquitination and degradation of IRF7. teasome-dependent mechanism underlies the negative regulation (A) HEK293T cells were transfected with the indicated plasmids for 24 h, of IRF7. Because ubiquitination is an important process during and Western blot was performed. (B) After transfection with control proteasome-dependent degradation, we detected whether the siRNA (sc) or Nmi-specific siRNA (ni), RAW264.7 cells were treated with downregulation of IRF7 was due to ubiquitination. The results medium or SeV for the indicated times, and the cell lysates were subjected revealed that the polyubiquitination of IRF7 increased consider- to immunoblot analysis. (C) After being transfected with the indicated ably with increasing Nmi expression levels (Fig. 5D). When the plasmids, HEK293T cells were treated with DMSO or MG132 for 8 h, and CC truncation was substituted in the same experiments, the level the cell lysates were subjected to immunoblot analysis. (D) HEK293T cells of polyubiquitination of IRF7 was lower (Fig. 5E). In addition, were transfected with the indicated plasmids and increasing concentrations by guest on September 29, 2021 Nmi knockdown decreased the endogeous polyubiquitination of of Nmi (0.5, 1, 2 g). Immunoprecipitation and immunoblot analysis were then performed. (E) HEK293T cells were transfected with the indicated IRF7 induced by SeV infection (Fig. 5F). Then, we determined plasmids. Immunoprecipitation and immunoblot analysis were then per- whether the ubiquitination induced by Nmi was K48 or K63 de- formed. (F) RAW264.7 cells were transfected with sc or ni, followed by pendent. We transfected cells with Nmi and IRF7 along with wild- SeV infection for 8 h. Immunoprecipitation and immunoblot analysis were type ubiquitin or ubiquitin mutants retaining only one lysine then performed. (G) Immunoassay of HEK293T cells transfected with residue, K48 or K63. We found that Nmi increased the ubiq- various combinations of plasmids, including GFP-IRF7, Myc-Nmi, HA- uitination of IRF7 with wild-type ubiquitin and K48 ubiquitin, but K48-ubi, and HA-K63-ubi, followed by immunoprecipitation with anti- not with K63 ubiquitin (Fig. 5G). All of these data suggest that GFP and immunoblot with anti-HA. Data are representative of three in- Nmi increases the level of K48-dependent ubiquitination and dependent experiments. subsequently induces the degradation of IRF7. Inhibition of anti-SeV responses in Nmi transgenic mice. To in- that Nmi-transgenic mice had lower IRF7 expression after SeV vestigate the physiological role of Nmi, we created Nmi-transgenic infection (Fig. 6E). Then we tested type I IFN and IL-6 expression mice that constitutively express Nmi under the control of a b-actin in the lung tissue or BALF of infected mice by real-time PCR promoter. The expression of the transgene was confirmed by or ELISA, respectively. We found that Nmi-transgenic mice had Western blot assay showing that BMDCs from Nmi-transgenic impaired type I IFN expression and normal IL-6 expression after mice expressed more Nmi protein than did those of wild-type virus infection (Fig. 6F, 6G). These data suggest that Nmi-transgenic controls (Fig. 6A). In vitro cultured Nmi-transgenic BMDCs dis- mice have impaired antiviral response. played lower IRF7 protein expression after SeV infection than did wild-type controls (Fig. 6B). Real-time PCR experiments indi- Discussion cated that SeV-induced IFN-b was inhibited in BMDCs from In this study, we identified Nmi as an SeV-inducible protein that Nmi-transgenic mice, with no obvious effect on the production of acts as a negative regulator of type I IFN production downstream IL-6 and TNF-a (Fig. 6C). Consistently, in ELISA experiments, of the viral recognition receptors. Our results demonstrated that SeV-induced IFN-a and IFN-b levels were significantly impaired Nmi could specifically interact with IRF7 and promote its K48- in BMDCs from Nmi-transgenic mice, whereas IL-6 and TNF-a dependent ubiquitination and subsequent proteasome-dependent were not affected (Fig. 6D). degradation. Thus, Nmi limited the excessive production of type I To test the physiological importance of Nmi in vivo, we chal- IFNs in response to viral replication. lenged Nmi-transgenic and wild-type control mice with SeV, and Type I IFN production is important for the antiviral response, and antiviral responses were examined. We infected mice via the in- it must be tightly regulated. Several reports have focused on the tranasal route with SeV and examined the expression kinetics of negative regulation of the virus-triggered pathway. In this study, IRF7 in the lungs of mice at the indicated days. The data showed we identified Nmi as a new negative regulator of the SeV-induced 6 Nmi NEGATIVELY REGULATES TYPE I IFN PRODUCTION Downloaded from

FIGURE 6. Nmi negatively regulates virus-triggered type I IFN production by targeting IRF7 in Nmi-transgenic mice. (A) Immunoblot analysis of Nmi protein in transgenic mice. (B) Immunoblot analysis of IRF7 protein in BMDCs from Nmi-transgenic (TG) or wild-type (WT) mice infected by SeV at the indicated times. (C and D) Real-time PCR analysis (C) or ELISA analysis (D) in BMDCs from Nmi-transgenic or wild-type mice infected by SeV. (E–G) Approximately 5 3 105 virus particles were infected intranasally into 6-wk-old Nmi-transgenic or wild-type C57BL/6 mice (n = 2 mice at each time point). Lung tissue and BALF were collected on the indicated days, and immunoblot analysis (E), real-time analysis (F), or ELISA analysis (G) was then per- http://www.jimmunol.org/ formed. Data are representative of three independent experiments. Data are presented as the mean 6 SD of duplicates. *p , 0.05, **p , 0.01. signaling pathway. The overexpression of Nmi inhibited virus- results of our study suggested that Nmi promoted the K48-linked triggered type I IFN production, whereas knockdown had the ubiquitination of IRF7, confirming our hypothesis that Nmi neg- opposite effect. In addition, Nmi-transgenic mice had impaired atively regulates the type I IFN production triggered by viral type I IFN expression triggered by virus infection. Because Nmi infection by targeting IRF7 and promoting the K48-dependent was a type I IFN-induced protein, it functioned as a negative ubiquitination and subsequent proteasome-dependent degradation feedback regulator of the virus-induced signaling pathway. of IRF7. Previous studies have demonstrated that IRF7 plays an essential Although Nmi increased the level of ubiquitination and the by guest on September 29, 2021 role in the virus-triggered induction of type I IFNs and innate subsequent degradation of IRF7, no report has been made regarding antiviral immunity. Therefore, determining how the activities of whether Nmi has E3 ubiquitin ligase activity or whether Nmi can IRF7 are regulated is important in fully understanding virus- directly catalyze ubiquitination. The underlying molecular mech- induced type I IFN production. In this study, our results suggest anism was not determined. Another protein that can ubiquitinate that Nmi interacts with IRF7 in both a mammalian overexpression IRF7 and induce its proteasome-dependent degradation may be system and in untransfected RAW264.7 cells. However, IRF3, involved. Therefore, further studies are needed, and the identi- which is similar to IRF7 with respect to sequence and structure, fication of an IRF7-specific ubiquitin ligase or conjugating enzyme did not interact with Nmi under the same experimental conditions. would increase understanding of the regulation of IRF7 signaling. Therefore, the interaction between Nmi and IRF7 is specific. Collectively, our results demonstrate that Nmi targets IRF7 to Structural studies have demonstrated that Nmi contains three inhibit IRF7 transactivation. The overall function of this process domains: one coiled–coiled domain followed by two NIDs. In this is to turn off or limit the excessive production of type I IFNs. Thus, study, we found that the Nmi truncation without the NIDs could we report a novel role for Nmi in regulating cellular antiviral re- not interact with IRF7 and lost its ability to limit the production sponses. of virus-induced type I IFNs. Therefore, Nmi interacts with IRF7 through its NID domain, and this interaction is important for the Disclosures D inhibitory function of Nmi. The Nmi mutants NID1, NID2, and The authors have no financial conflicts of interest. DCC were still capable of interacting with IRF7, but their ability of regulating type I IFN was impaired, suggesting the CC domain has some activities in mediating the regulation of type I IFN References through unknown mechanisms. 1. Roberts, R. M., L. Liu, Q. Guo, D. Leaman, and J. Bixby. 1998. The evolution of the type I interferons. J. Interferon Res. 18: 805–816. In general, the degradation of transcriptional factors is one of 2. Theofilopoulos, A. N., R. Baccala, B. Beutler, and D. H. Kono. 2005. Type I the principal mechanisms that reduce or terminate transcriptional interferons (alpha/beta) in immunity and autoimmunity. Annu. Rev. Immunol. 23: activation. This type of mechanism is involved not only in immune 307–336. 3. Samuel, C. E. 2001. Antiviral actions of interferons. Clin. Microbiol. Rev. 14: responses but also in many other biological phenomena. In this 778–809 (table of contents.). study, we found that Nmi increased the degradation of IRF7. 4. Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. Further experiments clarified that the Nmi-mediated downreg- Immunol. 21: 335–376. 5. Takeuchi, O., and S. Akira. 2008. MDA5/RIG-I and virus recognition. Curr. ulation of the IRF7 level was reversed by the proteasome inhibitor Opin. Immunol. 20: 17–22. MG132, which indicated that Nmi increased IRF7 degradation 6. Kato, H., O. Takeuchi, S. Sato, M. Yoneyama, M. Yamamoto, K. Matsui, S. Uematsu, A. Jung, T. Kawai, K. J. Ishii, et al. 2006. Differential roles of through a proteasome-dependent process. K48-linked ubiquitination MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 441: 101– plays an important role in proteasome-dependent degradation. The 105. The Journal of Immunology 7

7. Kawai, T., K. Takahashi, S. Sato, C. Coban, H. Kumar, H. Kato, K. J. Ishii, 23. Gao, L., H. Coope, S. Grant, A. Ma, S. C. Ley, and E. W. Harhaj. 2011. ABIN1 O. Takeuchi, and S. Akira. 2005. IPS-1, an adaptor triggering RIG-I- and Mda5- protein cooperates with TAX1BP1 and A20 proteins to inhibit antiviral signal- mediated type I interferon induction. Nat. Immunol. 6: 981–988. ing. J. Biol. Chem. 286: 36592–36602. 8. Xu, L. G., Y. Y. Wang, K. J. Han, L. Y. Li, Z. Zhai, and H. B. Shu. 2005. VISA is 24. Yang, Y. K., H. Qu, D. Gao, W. Di, H. W. Chen, X. Guo, Z. H. Zhai, and an adapter protein required for virus-triggered IFN-beta signaling. Mol. Cell 19: D. Y. Chen. 2011. ARF-like protein 16 (ARL16) inhibits RIG-I by binding with 727–740. its C-terminal domain in a GTP-dependent manner. J. Biol. Chem. 286: 10568– 9. Seth, R. B., L. Sun, C. K. Ea, and Z. J. Chen. 2005. Identification and charac- 10580. terization of MAVS, a mitochondrial antiviral signaling protein that activates 25. Kubota, T., M. Matsuoka, S. Xu, N. Otsuki, M. Takeda, A. Kato, and K. Ozato. NF-kappaB and IRF 3. Cell 122: 669–682. 2011. PIASy inhibits virus-induced and interferon-stimulated transcription 10. Meylan, E., J. Curran, K. Hofmann, D. Moradpour, M. Binder, through distinct mechanisms. J. Biol. Chem. 286: 8165–8175. R. Bartenschlager, and J. Tschopp. 2005. Cardif is an adaptor protein in the RIG- 26. Lebrun, S. J., R. L. Shpall, and L. Naumovski. 1998. Interferon-induced up- I antiviral pathway and is targeted by hepatitis C virus. Nature 437: 1167–1172. regulation and cytoplasmic localization of Myc-interacting protein Nmi. J. In- 11. Saha, S. K., E. M. Pietras, J. Q. He, J. R. Kang, S. Y. Liu, G. Oganesyan, terferon Cytokine Res. 18: 767–771. A. Shahangian, B. Zarnegar, T. L. Shiba, Y. Wang, and G. Cheng. 2006. Reg- 27. Bannasch, D., I. Weis, and M. Schwab. 1999. Nmi protein interacts with regions ulation of antiviral responses by a direct and specific interaction between TRAF3 that differ between MycN and Myc and is localized in the cytoplasm of neu- and Cardif. EMBO J. 25: 3257–3263. roblastoma cells in contrast to nuclear MycN. Oncogene 18: 6810–6817. 12. Oganesyan, G., S. K. Saha, B. Guo, J. Q. He, A. Shahangian, B. Zarnegar, 28. Sakamuro, D., and G. C. Prendergast. 1999. New Myc-interacting proteins: A. Perry, and G. Cheng. 2006. Critical role of TRAF3 in the Toll-like - a second Myc network emerges. Oncogene 18: 2942–2954. dependent and -independent antiviral response. Nature 439: 208–211. 29. Schlierf, B., S. Lang, T. Kosian, T. Werner, and M. Wegner. 2005. The high- 13. Ha¨cker, H., V. Redecke, B. Blagoev, I. Kratchmarova, L. C. Hsu, G. G. Wang, mobility group Sox10 interacts with the N-myc-interacting M. P. Kamps, E. Raz, H. Wagner, G. Ha¨cker, et al. 2006. Specificity in Toll-like protein Nmi. J. Mol. Biol. 353: 1033–1042. receptor signalling through distinct effector functions of TRAF3 and TRAF6. 30. Zhou, X., J. Liao, A. Meyerdierks, L. Feng, L. Naumovski, E. C. Bottger, and Nature 439: 204–207. M. B. Omary. 2000. Interferon-alpha induces nmi-IFP35 heterodimeric complex 14. Sharma, S., B. R. tenOever, N. Grandvaux, G. P. Zhou, R. Lin, and J. Hiscott. formation that is affected by the phosphorylation of IFP35. J. Biol. Chem. 275: 2003. Triggering the interferon antiviral response through an IKK-related 21364–21371. pathway. Science 300: 1148–1151. 31. Lee, N. D., J. Chen, R. L. Shpall, and L. Naumovski. 1999. Subcellular locali- Downloaded from 15. Hiscott, J., P. Pitha, P. Genin, H. Nguyen, C. Heylbroeck, Y. Mamane, zation of interferon-inducible Myc/stat-interacting protein Nmi is regulated by M. Algarte, and R. Lin. 1999. Triggering the interferon response: the role of IRF- a novel IFP 35 homologous domain. J. Interferon Cytokine Res. 19: 1245–1252. 3 transcription factor. J. Interferon Cytokine Res. 19: 1–13. 32. Chen, J., R. L. Shpall, A. Meyerdierks, M. Hagemeier, E. C. Bo¨ttger, and 16. Lin, R., Y. Mamane, and J. Hiscott. 2000. Multiple regulatory domains control L. Naumovski. 2000. Interferon-inducible Myc/STAT-interacting protein Nmi IRF-7 activity in response to virus infection. J. Biol. Chem. 275: 34320–34327. associates with IFP 35 into a high molecular mass complex and inhibits 17. Honda, K., H. Yanai, H. Negishi, M. Asagiri, M. Sato, T. Mizutani, N. Shimada, proteasome-mediated degradation of IFP 35. J. Biol. Chem. 275: 36278–36284. Y. Ohba, A. Takaoka, N. Yoshida, and T. Taniguchi. 2005. IRF-7 is the master 33. Zhang, K., G. Zheng, and Y. C. Yang. 2007. Stability of Nmi protein is con-

regulator of type-I interferon-dependent immune responses. Nature 434: 772– trolled by its association with Tip60. Mol. Cell. Biochem. 303: 1–8. http://www.jimmunol.org/ 777. 34. Zhang, L., Y. Tang, Y. Tie, C. Tian, J. Wang, Y. Dong, Z. Sun, and F. He. 2007. 18. Komuro, A., D. Bamming, and C. M. Horvath. 2008. Negative regulation of The PH domain containing protein CKIP-1 binds to IFP35 and Nmi and is in- cytoplasmic RNA-mediated antiviral signaling. Cytokine 43: 350–358. volved in cytokine signaling. Cell. Signal. 19: 932–944. 19. Deng, W., M. Shi, M. Han, J. Zhong, Z. Li, W. Li, Y. Hu, L. Yan, J. Wang, Y. He, 35. Zhu, M., S. John, M. Berg, and W. J. Leonard. 1999. Functional association of et al. 2008. Negative regulation of virus-triggered IFN-beta signaling pathway by Nmi with Stat5 and Stat1 in IL-2- and IFNgamma-mediated signaling. Cell 96: alternative splicing of TBK1. J. Biol. Chem. 283: 35590–35597. 121–130. 20. Hu, Y., J. Wang, B. Yang, N. Zheng, M. Qin, Y. Ji, G. Lin, L. Tian, X. Wu, 36. Tan, J., W. Qiao, J. Wang, F. Xu, Y. Li, J. Zhou, Q. Chen, and Y. Geng. 2008. L. Wu, and B. Sun. 2011. Guanylate binding protein 4 negatively regulates virus- IFP35 is involved in the antiviral function of interferon by association with the induced type I IFN and antiviral response by targeting IFN regulatory factor 7. J. viral tas transactivator of bovine foamy virus. J. Virol. 82: 4275–4283. Immunol. 187: 6456–6462. 37. Fillmore, R. A., A. Mitra, Y. Xi, J. Ju, J. Scammell, L. A. Shevde, and 21. Liang, Q., H. Deng, X. Li, X. Wu, Q. Tang, T. H. Chang, H. Peng, F. J. Rauscher, R. S. Samant. 2009. Nmi (N-Myc interactor) inhibits Wnt/beta-catenin signaling

III, K. Ozato, and F. Zhu. 2011. Tripartite motif-containing protein 28 is a small and retards tumor growth. Int. J. Cancer 125: 556–564. by guest on September 29, 2021 ubiquitin-related modifier E3 ligase and negative regulator of IFN regulatory 38. Hou, W., Y. Wu, S. Sun, M. Shi, Y. Sun, C. Yang, G. Pei, Y. Gu, C. Zhong, and factor 7. J. Immunol. 187: 4754–4763. B. Sun. 2003. Pertussis toxin enhances Th1 responses by stimulation of dendritic 22. Allen, I. C., C. B. Moore, M. Schneider, Y. Lei, B. K. Davis, M. A. Scull, cells. J. Immunol. 170: 1728–1736. D. Gris, K. E. Roney, A. G. Zimmermann, J. B. Bowzard, et al. 2011. NLRX1 39. Huang, C., E. Bi, Y. Hu, W. Deng, Z. Tian, C. Dong, Y. Hu, and B. Sun. 2006. A protein attenuates inflammatory responses to infection by interfering with the novel NF-kappaB binding site controls human granzyme B gene transcription. J. RIG-I-MAVS and TRAF6-NF-kB signaling pathways. Immunity 34: 854–865. Immunol. 176: 4173–4181.