Phosphatase PP4 Negatively Regulates Type I IFN Production and Antiviral Innate Immunity by Dephosphorylating and Deactivating TBK1 This information is current as of October 4, 2021. Zhenzhen Zhan, Hao Cao, Xuefeng Xie, Linshan Yang, Peng Zhang, Yihan Chen, Huimin Fan, Zhongmin Liu and Xingguang Liu J Immunol 2015; 195:3849-3857; Prepublished online 11

September 2015; Downloaded from doi: 10.4049/jimmunol.1403083 http://www.jimmunol.org/content/195/8/3849

Supplementary http://www.jimmunol.org/content/suppl/2015/09/11/jimmunol.140308 http://www.jimmunol.org/ Material 3.DCSupplemental References This article cites 43 articles, 10 of which you can access for free at: http://www.jimmunol.org/content/195/8/3849.full#ref-list-1

Why The JI? Submit online.

by guest on October 4, 2021 • 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

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 © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Phosphatase PP4 Negatively Regulates Type I IFN Production and Antiviral Innate Immunity by Dephosphorylating and Deactivating TBK1

Zhenzhen Zhan,*,† Hao Cao,* Xuefeng Xie,‡ Linshan Yang,* Peng Zhang,x Yihan Chen,*,† Huimin Fan,* Zhongmin Liu,* and Xingguang Liux

The effective recognition of viral infection and subsequent type I IFN production is essential for the host antiviral innate immune responses. The phosphorylation and activation of TANK-binding kinase 1 (TBK1) plays crucial roles in the production of type I IFN mediated by TLR and retinoic acid–inducible I–like receptors. Type I IFN expression must be tightly regulated to prevent the development of immunopathological disorders. However, how the activated TBK1 is negatively regulated by phosphatases remains poorly understood. In this study, we identified a previously unknown role of (PP)4 by acting as a TBK1 phosphatase. PP4 expression was upregulated in macrophages infected with RNA virus, vesicular stomatitis virus, and Sendai virus in vitro and in vivo. Knockdown Downloaded from of PP4C, the catalytic subunit of PP4, significantly increased type I IFN production in macrophages and dentritic cells triggered by TLR3/4 ligands, vesicular stomatitis virus, and Sendai virus, and thus inhibited virus replication. Similar results were also found in peritoneal macrophages with PP4C silencing in vivo and i.p. infection of RNA virus. Accordingly, ectopic expression of PP4C inhibited virus-induced type I IFN production and promoted virus replication. However, overexpression of a phosphatase-dead PP4C mutant abolished the inhibitory effects of wild-type PP4C on type I IFN production. Mechanistically, PP4 directly bound TBK1 upon virus infection, then dephosphorylated TBK1 at Ser172 and inhibited TBK1 activation, and subsequently restrained IFN regulatory factor 3 http://www.jimmunol.org/ activation, resulting in suppressed production of type I IFN and IFN-stimulated . Thus, serine/threonine phosphatase PP4 functions as a novel feedback negative regulator of RNA virus-triggered innate immunity. The Journal of Immunology, 2015, 195: 3849–3857.

key feature of the antiviral innate immune response is RNA sensors such as retinoic acid–inducible gene I (RIG-I) and the rapid induction and extracellular secretion of type I melanoma differentiation–associated gene 5 (MDA5), two well- A IFN that restricts virus replication (1, 2). Cytoplasmic known retinoic acid–inducible gene I–like family pro- teins, can recognize viral RNA and trigger antiviral signaling

pathways (3, 4). RIG-I and MDA5 contain CARD domains that by guest on October 4, 2021 *Research Center for Translational Medicine and Shanghai Heart Failure Research recruit IFN-b promoter stimulator 1 (IPS-1), which is an adaptor Center, East Hospital, Tongji University School of Medicine, Shanghai 200120, China; †Key Laboratory of Arrhythmias, Ministry of Education, East Hospital, Tongji molecule localized on the mitochondrial membrane (4, 5). IPS-1 University School of Medicine, Shanghai 200120, China; ‡School of Pharmacology, x associates with TNFR-associated factor (TRAF)3, leading to the Anhui Medical University, Hefei 230032, China; and National Key Laboratory of phosphorylation at Ser172 and activation of TANK-binding kinase Medical Immunology and Institute of Immunology, Second Military Medical Uni- versity, Shanghai 200433, China 1 (TBK1) (6, 7). The activated TBK1 phosphorylates and activates Received for publication December 11, 2014. Accepted for publication August 6, IFN regulatory factor (IRF)3 and IRF7 to induce the production of 2015. type I IFN and IFN-stimulated genes (ISGs), which is responsible This work was supported by National Natural Science Foundation of China Grants for inhibition of virus replication and clearance of virus-infected 81170084, 31270943, 81373146, and 81422021, Shanghai Rising-Star Program cells (8). Additionally, TLR3 also can recognize viral dsRNA to Grants 13QA1403100 and 14QA1404600, Basic Research Program of Science and Technology Commission of Shanghai Municipality Grant 14JC1405200, a grant from trigger type I IFN production through TBK1-mediated IRF3 ac- the Foundation for the Author of National Excellent Doctoral Dissertation of Peo- tivation (9). ple’s Republic of China, and by Doctor Science Research Foundation of the Educa- tion Ministry of China Grant 20130072120062. Although sufficient production of type I IFN is crucial for virus Address correspondence and reprint requests to Prof. Xingguang Liu or Dr. Zhenzhen Zhan, clearance, uncontrolled and excessive type I IFN expression causes National Key Laboratory of Medical Immunology and Institute of Immunology, pathological immune response to the host or autoimmune disorders, Second Military Medical University, 800 Xiangyin Road, Shanghai 200433, China and thus host cells have developed distinct strategies to tightly (X.L.) or Research Center for Translational Medicine and Shanghai Heart Failure Research Center, East Hospital, Tongji University School of Medicine, 150 Jimo regulate the activation of antiviral signaling and maintain the ho- Road, Shanghai 200120, China (Z.Z.). E-mail addresses: [email protected] meostasis of both the innate and adaptive immunity (10–12). As (X.L.) or [email protected] (Z.Z.) a critical kinase and central player involved in type I IFN signaling, The online version of this article contains supplemental material. TBK1 activation can be regulated by multiple molecules in various Abbreviations used in this article: BMDC, bone marrow–derived dendritic cell; HA, ways, such as phosphorylation, ubiquitination, and kinase activity hemagglutinin; IPS-1, IFN-b promoter stimulator 1; IRAK1, IL-1R–associated kinase 1; IRF, IFN regulatory factor; ISG, IFN-stimulated gene; MDA5, melanoma modulation (13, 14). However, how the kinase activity of TBK1 is differentiation–associated gene 5; MOI, multiplicity of infection; ODN, oligo- switched off by its phosphatases remains poorly understood. Until deoxynucleotide; poly (I:C), polyinosinic-polycytidylic acid; PP, protein phospha- tase; PP4C, catalytic unit of PP4; PPM1B, PP Mg2+/Mn2+-dependent 1B; RIG-I, now, there are three phosphatases that have been reported to neg- retinoic acid–inducible gene I; SeV, Sendai virus; SHP-2, Src homology region 2 atively regulate type I IFN production by targeting TBK1 (15–17). domain–containing phosphatase 2; siRNA, small interfering RNA; TBK1, TANK- Src homology 2 domain-containing protein tyrosine phosphatase 2 binding kinase 1; TRAF, TNFR-associated factor; VSV, vesicular stomatitis virus. (SHP-2) inhibits TBK1 activity through a phosphatase activity–in- Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 dependent mechanism (15). The absence of SHIP-1 results in the www.jimmunol.org/cgi/doi/10.4049/jimmunol.1403083 3850 PP4 INHIBITS ANTIVIRAL RESPONSE BY DEPHOSPHORYLATING TBK1 increased level of TBK1 phosphorylation (16). However, it is un- prepared and cultured as described previously (26). The plasmids were likely that SHP-2 and SHIP-1 function as TBK1 phosphatases to nucleofected into mouse peritoneal macrophages and RAW264.7 cells with directly dephosphorylate and deactivate TBK1 (15, 16). Only one corresponding Nucleofector kits using Amaxa Nucleofector II biosystems (Lonza). serine/threonine phosphatase, protein phosphatase (PP) Mg2+/Mn2+- dependent 1B (PPM1B), is found to bind TBK1 and dephos- RNA interference 172 phorylate TBK1 at Ser , and then inhibit TBK1 activation in The sequences of small interfering RNAs (siRNAs) targeting two different HEK293T cells and HeLa cells (17). However, whether other sites of PP4 catalytic subunit (Ppp4c) gene were 59-TGGACTCGC- phosphatases also function in the direct regulation of phosphory- CAGTCACAGT-39 (no. 1) and 59-CCGTGGTTTCTACAGTGTT-39 (no. lation and kinase activity of TBK1 in innate immune cells remains 2). The control small RNA sequence was 59-TTCTCCGAACGTGTCA- 9 unknown, which needs to be further elucidated. CGT-3 . siRNA duplexes were synthesized by GenePharma (Shanghai, China) and transfected into mouse peritoneal macrophages and BMDCs The PP2A family, a serine/threonine phosphatase family that using INTERFERin (Polyplus Transfection) according to the manu- comprises a common domain of PP2A-like phosphatase, includes facturer’s instructions. For in vivo silencing of PP4C in peritoneal cells, PP2A, PP4, PP5, and PP6 and is implicated in many biological siRNA duplexes were complexed with in vivo jetPEI reagent (Polyplus processes (18). PP4 is a holoenzyme that is comprised of the Transfection) at the N/P ratio of 7 according to the instructions, and then the transfection mixture was i.p. injected into mice. catalytic subunit of PPR (PP4C) in association with PP4R regulatory subunits (19). PP4 is highly conserved during evolution and is RNA quantification ubiquitously expressed in many tissues. Although PP4 has been Total RNA was extracted with TRIzol reagent (Invitrogen). Quantitative showntobeinvolvedinmultiplecellular processes, including reg- real-time PCR analysis was performed with LightCycler (Roche) and ulation of microtubule organization, DNA damage repair, thymocyte a SYBR RT-PCR (Takara Bio). The primers for Ppp4c mRNA analysis development, and regulatory T cell functions, its role in the regula- were: 59-CTTGGTAGAAGAGAGCAACGTG-39 (forward) and 59- CGC- Downloaded from 9 tion of type I IFN signaling and antiviral immunity remains unknown CACCTACTCTGAACAGC-3 (reverse). The primers for mouse Ifnb1, Ifna4, human IFNB1, ISG15, and CCL15 mRNA analyses have been de- (20–23). In the present study, we found that PP4, but not other scribed (25, 26). Data were normalized to b-actin expression in each members of PP2A family, most significantly inhibited RNA virus- sample. induced IFN-b reporter activity by utilizing IFN-b luciferase re- porter–based screen. Furthermore, PP4 directly interacted with TBK1 Detection of cytokines after virus infection, then dephosphorylated TBK1 at Ser172 and IFN-b and IFN-a levels in the supernatants were measured by an ELISA http://www.jimmunol.org/ inhibitedTBK1activation,andsubsequently inhibited type I IFN kit (PBL Assay Science) according to the manufacturer’s instructions. production and antiviral innate immune response. Our findings iden- Assay of luciferase reporter gene expression tify serine/threonine phosphatase PP4 as a phosphatase for TBK1 to maintain immune homeostasis during antiviral innate immunity. HEK293 cells were cotransfected with the indicated luciferase reporter plasmid, pRL-TK-Renilla-luciferase plasmid, and the indicated other constructs. Total amounts of plasmid DNA were equalized with empty Materials and Methods control vector. After 24 h, luciferase activities in cell lysates were mea- Mice and reagents sured using Dual-Luciferase reporter assay system (Promega) according to the manufacturer’s instructions. Data were normalized for transfection C57BL/6 mice 6–8 wk old were from Joint Ventures Sipper BK Experi- efficiency by dividing firefly luciferase activity with that of Renilla lucif- by guest on October 4, 2021 mental Animal (Shanghai, China). All animal experiments were performed erase. in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, with the approval of the Scientific Investiga- Immunoblot and immunoprecipitation analyses tion Board of Second Military Medical University (Shanghai, China). Cells were lysed with cell lysis buffer (Cell Signaling Technology) sup- LPS (0111:B4) and polyinosinic-polycytidylic acid [poly(I:C)] was from plemented with protease inhibitor mixture. Nuclear protein was prepared Sigma-Aldrich. CpG oligodeoxynucleotide (ODN) was synthesized and with NE-PER nuclear and cytoplasmic protein extraction reagent (Thermo repurified as described previously (24). Ab specific for the PP4 catalytic Fisher Scientific). Protein concentrations of the extracts were measured with subunit was from Bethyl Laboratories. Abs to TBK1 phosphorylated at BCA assay (Thermo Fisher Scientific). The immunoblot analysis and immu- Ser172, IRF3 phosphorylated at Ser396, total TBK1, IKKε, IRF3, lamin noprecipitation analysis were performed as described previously (25, 26). A/C, b-actin, and hemagglutinin (HA) tags were from Cell Signaling Technology. Ab to IL-1R–associated kinase 1 (IRAK1) phosphorylated at GST pull-down assay Thr209 was from LifeSpan BioSciences. Anti-TRAF6 and anti-ubiquitin were from Santa Cruz Biotechnology. Anti–K48-ubiquitin and anti–K63- GST fusion proteins containing the wild-type PP4 catalytic subunit (GST- ubiquitin were from Millipore. Anti-Flag and the agarose used for im- PP4C) were expressed in Escherichia coli (BL21) and purified by GST munoprecipitation were from Agilent Technologies. Mouse rIFN-b and affinity chromatography (Thermo Fisher Scientific). GST-PP4C protein anti–IFN-b neutralizing Ab was from PBL Assay Science. was incubated with recombinant TBK1 protein (Abcam) at 4˚C for 30 min and further incubated with glutathione–Sepharose 4B (Thermo Fisher Plasmid constructs Scientific) for 2 h in immunoprecipitated buffer containing Nonidet P-40. After washing three times, the precipitates were subjected to Western blot Mouse cDNA encoding PP4C, PP2A (PP2CA and PP2CB), PP5 (PP5C), analysis as previously described (24). or PP6 (PP6C) was amplified from macrophage cDNA and cloned in pcDNA3.1 plasmid (Invitrogen) to generate a wild-type construct. PP4C In vitro phosphatase assay point mutant (R236L) was constructed by PCR amplification from the wild- type PP4C construct. Mouse cDNA encoding wild-type PP4C or mutant Recombinant active TBK1 protein (Abcam) was preincubated with wild- (R236L) was cloned into pGEX-4T3 vector (GE Healthcare) to generate type or R236L mutant GST-PP4C protein in iced water for 15 min. GST-tagged PP4C construct. RIG-I, MDA-5, IPS-1, wild-type TBK1, Then recombinant IRF3 protein and a reaction buffer containing ATP TBK1 truncated mutants, IKKε, IRF3, and IFN-b luciferase reporter (Upstate Biotechnology) were added, followed by incubation for 20 min at plasmids were described previously (24, 25). All constructs were con- 30˚C. Phosphorylation of TBK1 and IRF3 was detected by immunoblot firmed by DNA sequencing. with Ab to TBK1 phosphorylated at Ser172 and Ab to IRF3 phosphorylated at Ser396 as previously described (15). Cell culture and transfection Kinase activity assay Bone marrow–derived dendritic cells (BMDCs) were generated as de- scribed previously (26). Mouse macrophage cell line RAW264.7 and the Endogenous TBK1 or IKKε protein was immunoprecipitated from cell HEK293 cell line were obtained from the American Type Culture Col- lysates with anti-TBK1 or IKKε Ab. Pellets were washed three times with lection. HEK293 cells were transfected with jetPEI reagents (Polyplus lysis buffer and then resuspended in kinase reaction buffer (40 mM Tris- Transfection). Thioglycolate-elicited mouse peritoneal macrophages were HCl [pH 7. 5], 20 mM MgCl2, 0.1 mg/ml BSA, and 0.05 mM DTT). The The Journal of Immunology 3851 kinase activity of TBK1 or IKKε was assayed with a kinase system and an ADP-Glo kinase assay kit, according to the manufacturer’s instructions (Promega). In vitro and in vivo infection of vesicular stomatitis virus and Sendai virus Vesicular stomatitis virus (VSV; Indiana strain) was propagated and am- plified by infection of a monolayer of Vero cells. Mouse peritoneal mac- rophages, BMDCs, and RAW264.7 cells were in vitro infected with VSV (multiplicity of infection [MOI] of 10) or Sendai virus (SeV; MOI of 1) for the indicated time. VSV RNA levels in the supernatant and cells as well as viral titers were quantified as described previously (27, 28). For in vivo virus infection, age- and sex-matched mice were i.p. infected with VSV (5 3 107 PFU/g) or SeV (1 3 108 PFU/g) for the indicated times. The peritoneal lavages were harvested, and then peritoneal fluids were used for viral titer analysis or concentrated using an Amicon ultra filter (Millipore) for ELISA of type I IFN; meanwhile, macrophages were sorted using MoFlo XDP (Beckman Coulter) from peritoneal exudate cells that were stained with F4/80 and CD11b Abs. The sorted peritoneal macrophages were used for the analysis of type I IFN mRNA and VSV RNA levels as well as for immunoblot and immunoprecipitation analysis (25). FIGURE 1. Overexpression of PP4 suppresses the activation of IFN-b reporter and gene expression of IFN-b and ISGs induced by SeV. (A) Statistical analysis Downloaded from Luciferase assay of IFN-b reporter activation in HEK293 cells cotrans- The statistical significance of comparisons between two groups was de- fected with IFN-b luciferase reporter plasmid and indicated expression termined with a two-tailed Student t test, whereas comparisons among plasmids encoded the catalytic subunits of PP2A family members. PP2CA , multiple groups were assessed by ANOVA. A p value of 0.05 was and PP2CB are the a and b isoforms of the PP2A catalytic subunit, re- considered statistically significant. spectively; PP4C, PP5C, and PP6C are the catalytic subunits of PP4, PP5, and PP6, respectively. The expression levels of the above phosphatase were B D Results detected by Western blot. ( – ) HEK293 cells were transfected with http://www.jimmunol.org/ IFN-b luciferase screen identifies PP4 as a regulator of anti- PP4C. After 24 h, cells were infected with SeV (MOI of 0.01) for 18 h and b B C D RNA virus innate immunity then the mRNA expression of IFN- ( ), ISG15 ( ), and CCL5 ( ) was detected by quantitative PCR. Data are shown as mean 6 SEM of three To systematically uncover phosphatases of the PP2A family that independent experiments. **p , 0.01, ***p , 0.001. are involved in the regulation of innate immunity against RNA virus, we performed an IFN-b luciferase reporter–based screen in various phosphatase-overexpressed HEK293 cells infected with (Supplemental Fig. 2A, 2B). The transfection of these two siRNAs SeV. Luciferase activity assay showed that overexpression of PP4C, specific for PP4C did not affect the protein expression of other phosphatases such as PPM1B, SHP-2, and SHIP-1 (Supplemental

the catalytic subunit of PP4, most significantly inhibited SeV-induced by guest on October 4, 2021 IFN-b reporter activity, as compared with other PP2A family mem- Fig. 2C), further indicating that PP4C knockdown by these two bers (Fig. 1A). Furthermore, PP4C overexpression also markedly different siRNAs is specific. Additionally, the production of type I suppressed gene transcription of SeV-induced IFNB1 and ISGs, in- IFN induced by LPS or poly(I:C) was also markedly increased in cluding ISG15 and CCL5 in HEK293 cells (Fig. 1B–D). Additionally, PP4C-silenced macrophages than that in control macrophages, VSV or SeV in vitro infection upregulated the mRNA and protein suggesting that PP4 also inhibits TLR3- and TLR4-induced type I expression of PP4C in a time-dependent manner in macrophages IFN production in macrophages (Fig. 2E). (Supplemental Fig. 1A, 1B). The mRNA and protein expression Furthermore, we observed the effects of PP4C overexpression levels of PP4C were also markedly increased in peritoneal macro- on RNA virus-induced type I IFN production in macrophages. Over- phages harvested from mice i.p. infected with VSV or SeV for the expression of PP4C significantly suppressed the production of IFN-b indicated time (Supplemental Fig. 1C, 1D). Collectively, these results and IFN-a protein, as well as the expression of IFN-b and IFN-a4 suggest that PP4 may be a negative feedback regulator of RNA virus- mRNA in macrophages infected with VSV or SeV (Fig. 3A, induced innate immune response. Supplemental Fig. 3A). Considering that PP4 is a cytosolic protein phosphatase, we further explored whether the phosphatase activity PP4 inhibits type I IFN and inflammatory cytokine production of PP4C was responsible for its inhibitory role in RNA virus- triggered by RNA virus and TLR in macrophages and BMDCs triggered type I IFN production. It has been reported that the re- To further identify whether PP4 could affect RNA virus-triggered placement of arginine 236 with leucine in PP4C (PP4C-R236L) innate immune response in macrophages and BMDCs, we first results in the loss of its phosphatase activity, which is still capable investigated the role of PP4 in type I IFN production upon RNA of binding substrate proteins (21). The production of type I IFN virus challenge. siRNA (siRNA no. 1) specifically targeting PP4 induced by VSVor SeV was not inhibited in macrophages transfected catalytic subunit Ppp4c could significantly knock down PP4C with the PP4C-R236L mutant construct (Fig. 3A, Supplemental expression (Fig. 2A). Silencing of PP4C significantly increased Fig. 3A). Similar results were also found in RAW264.7 cells IFN-b and IFN-a protein production, as well as IFN-b and IFN- overexpressed with wild-type or R236L mutant PP4C (Fig. 3B, a4 mRNA levels in macrophages upon infection with VSV or SeV Supplemental Fig. 3B). These data indicate that PP4 suppresses (Fig. 2B, 2C). The same phenomenon was also observed in PP4C- type I IFN production in macrophages and BMDCs upon RNA silenced BMDCs infected with VSV or SeV (Fig. 2D). To further virus infection, which depends on the phosphatase activity of PP4. confirm the inhibitory role of PP4 on RNA virus-induced type I The effects of PP4 on inflammatory cytokine production trig- IFN production, we used another siRNA (siRNA no. 2) targeting geredbyTLRandRNAviruswere also investigated. PP4C another site of the Ppp4c gene. Consistently, PP4C knockdown knockdown significantly increased the production of TNF-a and by this siRNA (siRNA no. 2) also significantly increased IFN-b IL-6 in macrophages infected with VSV or SeV (Supplemental and IFN-a production triggered by VSV or SeV in macrophages Fig. 3C). Similar results were also found in PP4C-silencing 3852 PP4 INHIBITS ANTIVIRAL RESPONSE BY DEPHOSPHORYLATING TBK1

infected macrophages, we found that PP4C silencing could sig- nificantly inhibit VSV replication, whereas anti–IFN-b-neutralizing Ab could reverse the suppression of VSV replication mediated by PP4C silencing, indicating that the increase of type I IFN production mediated by PP4C silencing inhibits VSV replication (Fig. 4A). Furthermore, overexpression of wild-type PP4C markedly promoted VSV replication, whereas overexpression of the PP4C-R236L mutant or addition of recombinant murine IFN-b could reverse the en- hanced VSV replication mediated by wild-type PP4C overexpression (Fig. 4A). Correspondingly, the intracellular and supernatant VSV RNA levels were more significantly decreased in PP4C-silencing macrophages, and the increased VSV RNA levels were observed in wild-type PP4C-overexpressing macrophages but not in PP4C-R236L mutant–overexpressing macrophages (Fig. 4B, 4C). These data sug- gest that PP4 suppresses the production of type I IFN, thus promoting RNA virus replication. PP4 suppresses innate immunity against RNA virus infection in vivo

We further demonstrated the in vivo biological effect of PP4 on Downloaded from virus infection by knocking down PP4 in mice. The siRNA duplexes complexed with in vivo jetPEI reagent were i.p. injected into mice and could significantly reduce the endogenous protein level of PP4C in macrophages from the peritoneal cavity (Fig. 5A). Mice injected with siRNA/jetPEI complex were i.p. infected with

VSV or SeV. Silencing of PP4C significantly increased virus- http://www.jimmunol.org/ induced IFN-b and IFN-a production in peritoneal fluids, as well as IFN-b and IFN-a4 mRNA levels in peritoneal macro- phages (Fig. 5B, 5C). PP4C silencing also markedly inhibited VSV replication in peritoneal fluids and decreased the intracel- lular VSV RNA levels in peritoneal macrophages (Fig. 5D). These in vivo data, which are in line with the observation in vitro, further

FIGURE 2. Silencing of PP4 increases type I IFN production triggered by guest on October 4, 2021 by RNA virus and TLR3/4 ligands in macrophages and BMDCs. (A) Mouse macrophages were transfected with control siRNA (Ctrl) or PP4C siRNA (siRNA). Forty-eight hours later, the efficiency of silencing was detected by quantitative PCR and Western blot. (B–D) Mouse macrophages and BMDCs were transfected with PP4C siRNA or control siRNA as in (A). After 48 h, cells were infected with VSV (MOI of 10) or SeV (MOI of 1), then IFN-b and IFN-a levels in the supernatants after virus infection for 24 h were measured by ELISA (B and D), and the mRNA expression of IFN-b and IFN-a4 after virus infection for 18 h were detected by quan- titative PCR (C). (E) Mouse macrophages were transfected as in (A). After 48 h, cells were stimulated with LPS or poly(I:C) for 6 h, then IFN-b and IFN-a levels in the supernatants were measured by ELISA. Data are representative of three independent experiments with similar results (A, right) or are shown as mean 6 SEM of three independent experiments using five mice per experiment (total 15 mice for three independent experiments) (A, left, and B–E). **p , 0.01. macrophages stimulated with LPS, poly(I:C), or CpG ODN (Supplemental Fig. 3D). PP4C overexpression suppressed TNF-a and IL-6 production triggered by VSV, SeV, LPS, poly(I:C), or FIGURE 3. Overexpression of PP4 inhibits RNA virus-induced pro- CpG ODN in macrophages (Supplemental Fig. 3E, 3F). These duction of type I IFN in a phosphatase-dependent manner. (A and B) A B data suggest that PP4 also inhibits TLR- and RIG-I–triggered Mouse macrophages ( ) and RAW264.7 cells ( ) were transiently trans- fected with wild-type PP4C (PP4C) or PP4C-R236L mutant (PP4C mut) inflammatory cytokine production in macrophages. plasmid. After 48 h, cells were infected with VSV (MOI of 10) or SeV PP4 promotes VSV replication through impairment of type I (MOI of 1), then IFN-b and IFN-a levels in the supernatants after virus IFN production infection for 24 h were measured by ELISA (A, upper, and B), and the mRNA expression of IFN-b and IFN-a4 after virus infection for 18 h were To further investigate the biological significance of PP4 in RNA detected by quantitative PCR (A, lower). Data are shown as mean 6 SEM virus-induced immune response, we went on to investigate the of three independent experiments using 5 mice per experiment (total 15 effects of PP4 on VSV replication and the role of IFN-b in these mice for three independent experiments) (A), and three independent effects. By measuring VSV titers in the culture supernatants of the experiments (B). **p , 0.01. The Journal of Immunology 3853

TBK1-mediated IFN-b reporter activity as wild-type PP4C (Fig. 6G). Additionally, wild-type PP4C but not PP4C-R236L mutant also inhibited the activity of IRF3 reporter induced by TBK1 (Fig. 6H). These results indicate that PP4 negatively reg- ulates the activation of RNA virus-triggered type I IFN pathway by functioning at TBK1. PP4 directly interacts with TBK1 Next we investigated whether PP4 could interact with TBK1 to negative regulate RIG-I pathway activation. Immunoprecipitation assays with specific PP4C or TBK1 Ab showed that PP4 could interact with TBK1 in macrophages after VSV infection (Fig. 7A, 7B). The interaction between PP4C and TBK1 was also found in peritoneal macrophages from mice i.p. infected with VSV (Fig. 7C). Furthermore, GST pull-down assay using GST-tagged PP4C protein showed that TBK1 could be pulled down by GST- tagged PP4C, suggesting that these two proteins can interact di- rectly (Fig. 7D). However, PP4C did not bind IKKε, another ki- nase in RIG-I signaling, and IRF3 in immunoprecipitation assays, indicating that PP4 specifically interacts with TBK1 (Supplemental Downloaded from Fig. 4A). TBK1 contains the N-terminal kinase domain, the FIGURE 4. PP4 inhibits VSV replication in macrophages. (A) Mouse ubiquitin-like domain, and two coiled-coil domains at the C macrophages were transfected with control siRNA (Ctrl) or PP4C siRNA (siRNA) (left) and wild-type PP4C (WT) or PP4C-R236L mutant (Mut) plasmid (right). Forty-eight hours later, cells were infected with VSV at an

MOI of 10 for 1 h and washed, then fresh medium or fresh medium http://www.jimmunol.org/ containing anti–IFN-b-neutralizing Ab (100 neutralizing units/ml) (left)or recombinant mouse IFN-b (100 U/ml) (right) were added as indicated.

After 72 h, 50% tissue culture–infective dose (TCID50) of virus in supernatants was measured. (B and C) Mouse macrophages were treated as in (A), and intracellular VSV RNA replicates were detected by quantitative PCR (B). RNAs from equal volume of cultural supernatants were extracted and supernatant VSV RNA replicates were measured by quantitative PCR (C). Data are shown as mean 6 SEM of three independent experiments using 5 mice per experiment (total 15 mice for three independent experiments). **p , 0.01. N.D., not detected. by guest on October 4, 2021 demonstrate that PP4 inhibits innate immunity against RNA virus infection by impairing type I IFN production. PP4 inhibits the activation of type I IFN signaling upon RNA virus infection Next, we examined the underlying signaling mechanism by which PP4 inhibited RNA virus-triggered innate immune responses. We found that VSV infection–induced phosphorylations of TBK1 (Ser172) and IRF3 (Ser396) were markedly enhanced in macro- phages with PP4C silencing compared with nonsilencing macro- phages (Fig. 6A). Accordingly, PP4C overexpression inhibited the FIGURE 5. PP4 suppresses innate immunity against RNA virus infec- A phosphorylation of TBK1 and IRF3 in macrophages after VSV tion in vivo. ( ) The siRNA duplexes targeting PP4C (siRNA) or control siRNA (Ctrl) were complexed with in vivo jetPEI reagent and then i.p. infection (Fig. 6B). Furthermore, knockdown of PP4C resulted in injected into mice. After 48 h, the efficiency of silencing in macrophages more VSV-induced nuclear translocation of IRF3, whereas over- separated from peritoneal lavages was detected by quantitative PCR and expression of PP4C induced less IRF3 translocation into nucleus Western blot. (B and C) Mice were injected with siRNA complex as de- (Fig. 6C, 6D), suggesting that PP4 inhibits the activation of the scribed in (A). Forty-eight hours later, mice were i.p. infected with VSV or TBK1–IRF3 signaling pathway upon RNA virus infection. SeV. The levels of IFN-b and IFN-a in the concentrated peritoneal fluids To investigate how PP4 functions in the inhibition of type I IFN after virus infection for 24 h were measured by ELISA (B), and the mRNA signaling activation, we performed reporter assays by cotrans- expression of IFN-b and IFN-a4 in macrophages separated from perito- fecting plasmids expressing PP4C and IFN-b reporter with RIG-I, neal lavages after virus infection for 18 h was detected by quantitative PCR MDA5, IPS-1, TBK1, IKKε, or IRF3 constructs. Overexpression (C). (D) Mice were treated as in (B), and 50% tissue culture-infective dose of PP4C inhibited the activation of IFN-b reporter induced by (TCID50) of virus in the concentrated peritoneal fluids after VSV infection for 24 h was measured and VSV RNA levels in macrophages separated RIG-I, MDA5, IPS-1, and TBK1 but not IKKε or IRF3 (Fig. 6E), from peritoneal lavages were detected by quantitative PCR. Data are suggesting that PP4 may act on TBK1 to suppress RNA virus- representative of three independent experiments with similar results (A, induced activation of type I IFN signaling. Furthermore, over- right), data are shown as mean 6 SEM of three independent experiments expression of PP4C dose-dependently inhibited TBK1-mediated using six mice for every virus infection group and three mice for PBS IFN-b reporter activity (Fig. 6F). However, overexpression of phos- group of one experiment (total 90 mice for six groups and three inde- phatase activity inactive PP4C-R236L mutant could not suppress pendent experiments) (A, left, and B–D). **p , 0.01. N.D., not detected. 3854 PP4 INHIBITS ANTIVIRAL RESPONSE BY DEPHOSPHORYLATING TBK1

FIGURE 6. PP4 suppresses the activation of IRF3 pathway induced by VSV infection in macrophages. (A and B) RAW264.7 cells were transfected with PP4C siRNA or control siRNA (A), and wild-type PP4C (PP4C) or mock plasmid (B). Forty-eight hours later, cells were infected with VSV for the indicated time. Phospho-TBK1 (S172) and phospho-IRF3 (S396) in lysates were detected by Western blot. (C and D) RAW264.7 cells were transfected and infected as in (A) and (B), and total IRF3 among nuclear proteins in macrophages was detected by Western blot. Numbers below lanes and graphs indicate den- Downloaded from sitometry of the protein presented relative to b-actin or lamin A/C expression in that same lane. (E) HEK293 cells were cotransfected with or without RIG- I–, MDA5-, IPS-1–, TBK1, IKKε-, or IRF3-expressing plasmid, IFN-b luciferase reporter plasmid, together with 100 ng wild-type PP4C-expressing plasmid. After 24 h, IFN-b luciferase activity was measured and normalized by Renilla luciferase activity. (F) Luciferase assay of IFN-b activation in HEK293 cells cotransfected with or without TBK1-expressing plasmid, IFN-b luciferase reporter plasmid, together with the indicated amount wild-type PP4C plasmid. (G and H) Luciferase assay of IFN-b (G) or IRF3 (H) activation in HEK293 cells cotransfected with or without TBK1-expressing plasmid, IFN-b reporter plasmid (G), or IRF3 reporter plasmids (80 ng Gal4 luciferase reporter plasmid, 20 ng Gal4-IRF3–expressing plasmid) (H), together with 100 ng wild-type PP4C (WT) or PP4C-R236L mutant (Mut) plasmid. Data are representative of three independent experiments with similar results (A–D), http://www.jimmunol.org/ or are shown as mean 6 SEM of three independent experiments (E–H). **p , 0.01, ***p , 0.001. terminus (29). Considering that PP4C has only one phosphoprotein wild-type PP4 protein (Fig. 8A), further proving that PP4 sup- phosphatase domain in the middle of its protein sequence that is presses the phosphorylation and kinase activity of TBK1 de- conserved in PP2A phosphatase family, we wanted to map the pending on its phosphatase activity. Furthermore, TBK1 kinase region in TBK1 responsible for interaction with PP4. HA-tagged activity assay showed that the kinase activity of TBK1 protein TBK1 mutants were constructed and transfected into HEK293 immunoprecipitated from cell lysates of PP4C-silenced macro-

cells together with Flag-tagged PP4C. TBK1 without its by guest on October 4, 2021 N-terminal kinase domain did not bind PP4C (Fig. 7E), suggesting that PP4 binds the N-terminal kinase domain of TBK1. Given that a previous study showed that PP4 can bind TRAF6 and inhibit TLR4-induced NF-kBactivation(30),wewantedtoknow whether PP4 interacted with TRAF6 upon VSV infection. Im- munoprecipitation assays showed the interaction between PP4C and TRAF6 in VSV-infected macrophages (Supplemental Fig. 4A). Additionally, PP4C bound TRAF6 but not IRAK1 in LPS- stimulated macrophages (Supplemental Fig. 4B). These data suggest that PP4 also interacts with TRAF6, which is involved in TLR- and RIG-I–triggered NF-kB activation. PP4 dephosphorylates TBK1 at Ser172 and inhibits TBK1 activation Considering that the activation of TBK1 depends on phosphory- 172 lation at Ser within its activation loop (31), and PP4 over- A B 172 FIGURE 7. PP4 directly binds TBK1. ( and ) Mouse macrophages expression could inhibit Ser phosphorylation of TBK1 in were infected with VSV for the indicated time. Equal amounts of cell macrophages upon VSV infection (Fig. 6B), we wondered whether lysates were immunoprecipitated with PP4C or TBK1 Ab and then de- serine/threonine phosphatase PP4 could directly dephosphorylate tected with TBK1 and PP4C Abs, respectively. (C) Mice were i.p. infected TBK1 at Ser172. Incubation of recombinant activated TBK1 pro- with VSV for the indicated time. Equal amounts of cell lysates of mac- tein with purified PP4C protein resulted in substantially impaired rophages separated from peritoneal lavages were immunoprecipitated with Ser172 phosphorylation of TBK1 (Fig. 8A). We further investi- PP4C Ab and then detected with TBK1 and PP4C Abs. (D) GST pull-down gated the effect of PP4 on TBK1 kinase activity by evaluating assays were performed with GST-tagged PP4C and purified TBK1 protein. E recombinant TBK1-mediated phosphorylation of IRF3 as a sub- ( ) Flag-tagged PP4C plasmid together with various HA-tagged TBK1 truncates were transfected into HEK293 cells. After 24 h, cell extracts strate. We found that the activate TBK1 could phosphorylate IRF3 were immunoprecipitated with anti-HA Ab and then immunoblotted with and that, when incubated with PP4, its capacity to phosphorylate anti-Flag and anti-HA Abs. CC, coiled-coil domain; KD, kinase domain; IRF3 was substantially impaired (Fig. 8A). However, when TBK1 ULD, ubiquitin-like domain; WT, wild-type. Similar results were obtained protein was incubated with catalytically inactive PP4C-R236L in three independent experiments using three mice per experiment (total mutant protein, PP4C-R236L protein did not affect the phos- nine mice for three independent experiments) (A–C) or in three indepen- phorylation of TBK1 at Ser172 or IRF3 phosphorylation as did dent experiments (D and E). The Journal of Immunology 3855 phages infected with VSV was markedly higher than that of control macrophages, whereas the kinase activity of IKKε protein was not affected in PP4C-silenced macrophages upon VSV infection (Fig. 8B, 8C). Additionally, wild-type PP4C overexpression mark- edly inhibited TBK1 phosphorylation at Ser172 andIRF3phos- phorylation in macrophages upon SeV infection, but PP4C-R236L mutant overexpression did not (Fig. 8D, 8E). Furthermore, the phosphorylation of TBK1 at Ser172 was markedly enhanced in macrophages with PP4C silencing in vivo as compared with non- silencing macrophages from mice i.p. infected with VSV (Fig. 8F). Considering that K48-linked polyubiquitination of TBK1 promotes TBK1 degradation and K63-linked polyubiquitination enhances TBK1 activation (24, 32, 33), we investigated whether PP4 affected TBK1 polyubiquitination upon virus infection. However, we failed to find that PP4 promoted or inhibited TBK1 polyubiquitination (Fig. 8G), which indicates that PP4 terminates TBK1 activation by FIGURE 8. PP4 dephosphorylates TBK1 at Ser172 and inhibits TBK1 dephosphorylating TBK1 but not by affecting TBK1 ubiquitination. activation. (A) Wild-type PP4C protein or catalytically inactive PP4C- We also found that PP4C silencing enhanced K63-linked poly- R236L protein was preincubated with purified active TBK1 protein for ubiquitination of TRAF6 in macrophages infected with VSV, and 15 min, followed by incubation with purified IRF3 protein for 20 min. Downloaded from PP4C overexpression significantly inhibited TRAF6-driven NF-kB Phosphorylation of TBK1 (Ser172) and IRF3 (Ser396) was detected by reporter activity (Supplemental Fig. 4C, 4D), which is in line with Western blot. (B and C) Mouse macrophages were transfected with PP4C the previous study that PP4 can inhibit TRAF6 ubiquitination and siRNA (siRNA) or control siRNA (Ctrl). After 48 h, cells were infected NF-kB activation induced by TLR4 (30). Additionally, the knock- with VSV for 4 h or left untreated. TBK1 or IKKε protein in total cell ε down of PP4C did not affect LPS-induced phosphorylation of lysates was immunoprecipitated with anti-TBK1 or IKK Ab, respectively, and then kinase activity of TBK1 (B) or IKKε (C) was measured. (D and E) IRAK1 in macrophages (Supplemental Fig. 4E), suggesting that D k RAW264.7 cells were transfected with wild-type PP4C (PP4C) ( )or http://www.jimmunol.org/ PP4 inhibits TLR- and RIG-I–induced NF- B activation by tar- PP4C-R236L mutant (PP4C mut) (E) plasmid. Forty-eight hours later, cells geting TRAF6 but does not affect IRAK1 activation. were infected with SeV for the indicated time. Phospho-TBK1 (Ser172) and Collectively, our results demonstrate that PP4 can directly bind phospho-IRF3 (Ser396) in lysates were detected by Western blot. (F) The 172 TBK1, then dephosphorylate TBK1 at Ser , thus inhibiting siRNA duplexes targeting PP4C (siRNA) or control siRNA (Ctrl) were TBK1 kinase activation and downstream IRF3 activation, result- complexed with in vivo jetPEI reagent and then i.p. injected into mice. ing in the impaired type I IFN production. After 48 h, mice were i.p. infected with VSV for the indicated time. Phospho-TBK1 (Ser172) in lysates of macrophages separated from peri- Discussion toneal lavages was detected by Western blot. Numbers below lanes and b Viral infection initiates a series of signaling cascades that lead to graphs indicate densitometry of the protein presented relative to -actin expression in that same lane. (G) Mouse macrophages were transfected as by guest on October 4, 2021 the production of type I IFN, finally inducing ISGs to eliminate in (B). After 48 h, cells were infected with VSV for the indicated time. viruses (34). The phosphorylation and activation of TBK1 is es- Equal amounts of cell lysates were immunoprecipitated with TBK1 Ab sential for IRF3 activation and type I IFN production downstream and then detected with K48 or K63 ubiquitination Ab and TBK1 Ab. of RIG-I and MDA-5 (11, 12). Regulation of TBK1 activity is Ubiquitin in whole-cell lysates (WCL) of macrophages was detected by a key step in antiviral innate immunity, because excessive acti- Western blot. Data are representative of three independent experiments vation of type I IFN signaling results in immune disorder (13, 14). with similar results (A, D, and E) or representative of three independent However, the mechanisms underlying deactivation of TBK1 remain experiments with similar results using three mice per group of one ex- elusive. It is widely thought that phosphatases play major roles in periment (total 54 mice for six groups and three independent experiments) the control or termination of activity (18). However, (F), or are shown as mean 6 SEM of three independent experiments using B the phosphatases responsible for dephosphorylation and activity 3 mice per experiment (total 9 mice for three independent experiments) ( , C, and G). **p , 0.01. control of TBK1 are not well defined. In this study, we identify PP4 as a novel TBK1 phosphatase to dephosphorylate TBK1 at Ser172 and inhibit TBK1 activation, which suppresses the production of a member of PP2C family, can interact with TBK1 and dephos- type I IFN induced by RNA virus and TLR and restrains the innate phorylate TBK1 at Ser172 in HEK293T cells and HeLa cells (17). immune response against RNA virus in vitro and in vivo. In the present study, we provide the convincing evidence that PP4 Although the roles of phosphatases, including tyrosine phos- functions as a novel TBK1 phosphatase to dephosphorylate TBK1 phatases and serine/threonine phosphatases, in the regulation of at Ser172 and restrain the activation of TBK1 and downstream type innate immune response draw more attention (35), only three I IFN signaling in macrophages and dendritic cells. We further phosphatases, SHP-2, SHIP-1, and PPM1B, have been found to explored the relationship between PP4 and these above phospha- control the activation of type I IFN signaling by acting on TBK1 tases in regulation of TBK1 and found that PP4 did not bind until now (15–17). SHP-2 does not directly dephosphorylate PPM1B, SHP-2, or SHIP-1 upon VSV infection in macrophages TBK1 but suppresses TBK1 activity through a tyrosine phospha- (data not shown). Furthermore, overexpression of PP4 and PPM1B tase activity–independent mechanism to inhibit -dependent together more markedly suppressed transcription of SeV-induced type I IFN and proinflammatory cytokine production (15). SHIP-1 IFNB1 gene than that did PP4 or PPM1B overexpression alone in promotes TBK1 dissociation from endosomally located TLR3 HEK293 cells (data not shown). These results indicate that PP4 signaling complexes and hence inhibits TBK1 phosphorylation, does not associate with other phosphatases implicated in regula- whereas SHIP-1 is not found to bind and dephosphorylate TBK1 tion of TBK1 to form a complex, and PP4 may cooperate with (16). Thus, these studies indicate that SHP-2 and SHIP-1 inhibit PPM1B to dephosphorylate and deactivate TBK1 upon RNA virus TBK1 activation through other different mechanisms but not the infection. The interrelationship of the phosphatases in the regu- direct dephosphorylation of TBK1. However, only PPM1B, lation of TBK1 activation needs further investigation. 3856 PP4 INHIBITS ANTIVIRAL RESPONSE BY DEPHOSPHORYLATING TBK1

Phosphatase PP2A, which also belongs to the PP2A family as In conclusion, our data provide evidence that PP4 is a novel PP4, has been found to interact with IRF3, then dephosphorylate phosphatase known to dephosphorylate and deactivate TBK1. PP4 and deactivate IRF3, leading to limited type I IFN production upon negatively regulates innate immunity against RNA virus to protect SeV infection (36). In the present study, we failed to detect the the host from excessive immune responses. Better understanding of interaction between PP4 and IRF3; moreover, IRF3-induced IFN-b the deactivation process of TBK1 will benefit the design of im- reporter activity is not inhibited by coexpressed PP4, suggesting munotherapy strategies for type I IFN dysregulation diseases. that PP4 cannot dephosphorylate and inactivate IRF3 as does PP2A. Thus, we propose that PP4 may collaborate with PP2A to Acknowledgments control the activation of virus-induced type I IFN signaling We thank Panpan Ma, Bo Wang, and Junqi Yang for technical assistance. through targeting different components of RIG-I signaling. A previous study has shown that PP4 is a negative feedback Disclosures k regulator of TLR4-induced NF- B activation, although the data on The authors have no financial conflicts of interest. the effects of PP4 on the production of inflammatory cytokines and type I IFN induced by LPS are not provided (30). PP4 in- teracts with TRAF6 and inhibits its ubiquitination in RAW264.7 References cells upon TLR4 activation through an unknown mechanism, thus 1. Takeuchi, O., and S. Akira. 2010. Pattern recognition receptors and inflamma- k tion. Cell 140: 805–820. suppressing TRAF6 activation and the downstream NF- B acti- 2. Stetson, D. B., and R. Medzhitov. 2006. Type I interferons in host defense. vation (30). We also found that PP4 binds TRAF6 and PP4 inhibits Immunity 25: 373–381. K63-linked polyubiquitination of TRAF6 in macrophages infected 3. Barbalat, R., S. E. Ewald, M. L. Mouchess, and G. M. Barton. 2011. Nucleic acid recognition by the innate immune system. Annu. Rev. Immunol. 29: 185–214. Downloaded from with VSV. Moreover, PP4 overexpression also significantly 4. Kato, H., O. Takeuchi, S. Sato, M. Yoneyama, M. Yamamoto, K. Matsui, inhibits TRAF6-driven NF-kB reporter activity. Because TRAF6 S. Uematsu, A. Jung, T. Kawai, K. J. Ishii, et al. 2006. Differential roles of is a critical adapter protein involved in TLR- or RIG-I–triggered MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 441: 101–105. 5. Kowalinski, E., T. Lunardi, A. A. McCarthy, J. Louber, J. Brunel, B. Grigorov, proinflammatory cytokine production (1, 7, 11), together with our D. Gerlier, and S. Cusack. 2011. Structural basis for the activation of innate data that PP4 suppresses inflammatory cytokine production in- immune pattern-recognition receptor RIG-I by viral RNA. Cell 147: 423–435. 6. Kawai, T., K. Takahashi, S. Sato, C. Coban, H. Kumar, H. Kato, K. J. Ishii, duced by TLR or RNA virus, our data demonstrate that PP4 also O. Takeuchi, and S. Akira. 2005. IPS-1, an adaptor triggering RIG-I- and Mda5- can bind TRAF6 and hence inhibit the ubiquitination and activa- mediated type I interferon induction. Nat. Immunol. 6: 981–988. http://www.jimmunol.org/ tion of TRAF6, resulting in the impaired inflammatory cytokine 7. Xu, L. G., Y. Y. Wang, K. J. Han, L. Y. Li, Z. Zhai, and H. B. Shu. 2005. VISA is an adapter protein required for virus-triggered IFN-b signaling. Mol. Cell 19: production triggered by TLR- or RIG-I. It is possible that PP4 may 727–740. recruit other proteins to suppress TRAF6 ubiquitination, but the 8. Sharma, S., B. R. tenOever, N. Grandvaux, G. P. Zhou, R. Lin, and J. Hiscott. exact mechanism remains unclear and needs to be further ex- 2003. Triggering the interferon antiviral response through an IKK-related pathway. Science 300: 1148–1151. plored. Although IRAK1 is also a crucial kinase protein involved 9. O’Neill, L. A. 2008. The interleukin-1 receptor/Toll-like receptor superfamily: in TLR-induced proinflammatory cytokine production (1), we 10 years of progress. Immunol. Rev. 226: 10–18. 10. Theofilopoulos, A. N., R. Baccala, B. Beutler, and D. H. Kono. 2005. Type I found that PP4 does not bind IRAK1 and affect IRAK1 phos- interferons (a/b) in immunity and autoimmunity. Annu. Rev. Immunol. 23: 307– phorylation in LPS-activated macrophages, suggesting that PP4 336. by guest on October 4, 2021 inhibits TLR-triggered NF-kB activation and proinflammatory 11. Loo, Y. M., and M. Gale, Jr. 2011. Immune signaling by RIG-I-like receptors. Immunity 34: 680–692. cytokine production by targeting TRAF6 but not IRAK1. Con- 12. Patel, J. R., and A. Garcı´a-Sastre. 2014. Activation and regulation of pathogen sidering that TRAF6 does not play a major role in RIG-I–triggered sensor RIG-I. Cytokine Growth Factor Rev. 25: 513–523. IRF3 activation and type I IFN production, whereas TRAF6 is 13. Moore, C. B., and J. P. Ting. 2008. Regulation of mitochondrial antiviral sig- naling pathways. Immunity 28: 735–739. primarily involved in RIG-I–triggered NF-kB activation and 14. Zhao, W. 2013. Negative regulation of TBK1-mediated antiviral immunity. proinflammatory cytokine production (7, 11), we propose that PP4 FEBS Lett. 587: 542–548. primarily targets TBK1 but not TRAF6 and hence inhibits TBK1- 15. An, H., W. Zhao, J. Hou, Y. Zhang, Y. Xie, Y. Zheng, H. Xu, C. Qian, J. Zhou, Y. Yu, et al. 2006. SHP-2 phosphatase negatively regulates the TRIF adaptor driven type I IFN production upon virus infection. protein-dependent type I interferon and proinflammatory cytokine production. Viruses have evolved elaborate strategies, such as disruption of Immunity 25: 919–928. 16. Gabhann, J. N., R. Higgs, K. Brennan, W. Thomas, J. E. Damen, N. Ben Larbi, host recognition and impairment of type I IFN signaling, to evade G. Krystal, and C. A. Jefferies. 2010. Absence of SHIP-1 results in constitutive and subvert the innate immune responses (37, 38). Because TBK1 phosphorylation of Tank-binding kinase 1 and enhanced TLR3-dependent IFN-b functions as a key kinase to initiate type I IFN signaling upon viral production. J. Immunol. 184: 2314–2320. 17. Zhao, Y., L. Liang, Y. Fan, S. Sun, L. An, Z. Shi, J. Cheng, W. Jia, W. Sun, infection, viruses are more likely to evolve strategies to counteract Y. Mori-Akiyama, et al. 2012. PPM1B negatively regulates antiviral response via innate immunity by targeting TBK1. For example, Borna disease dephosphorylating TBK1. Cell. Signal. 24: 2197–2204. virus P protein, poxvirus protein N1L, and HCV protease NS2 can 18. Shi, Y. 2009. Serine/threonine phosphatases: mechanism through structure. Cell 139: 468–484. directly interact with TBK1 and inhibit its kinase activity, al- 19. Brewis, N. D., A. J. Street, A. R. Prescott, and P. T. Cohen. 1993. PPX, a novel though the mechanisms for suppression of TBK1 kinase activity protein serine/threonine phosphatase localized to centrosomes. EMBO J. 12: 987–996. remain unclear (39–41). Alternatively, viruses also can upregulate 20. Helps, N. R., N. D. Brewis, K. Lineruth, T. Davis, K. Kaiser, and P. T. Cohen. the expression levels or inhibitory functions of some negative 1998. Protein phosphatase 4 is an essential enzyme required for organisation of regulatory proteins, which suppresses the activation of type I IFN microtubules at centrosomes in Drosophila embryos. J. Cell Sci. 111: 1331–1340. 21. Zhou, G., K. A. Mihindukulasuriya, R. A. MacCorkle-Chosnek, A. Van Hooser, signaling to impair and evade the antiviral innate immunity (37, 38). M. C. Hu, B. R. Brinkley, and T. H. Tan. 2002. Protein phosphatase 4 is involved It has been reported that PP4 expression level is increased in HIV- in tumor necrosis factor-a-induced activation of c-Jun N-terminal kinase. J. Biol. infected monocyte-derived dendritic cells (Gene Expression Omni- Chem. 277: 6391–6398. 22. Shui, J. W., M. C. Hu, and T. H. Tan. 2007. Conditional knockout mice reveal an bus dataset GSE22589) and pandemic H1N1 influenza KY/136- essential role of protein phosphatase 4 in thymocyte development and pre-T-cell infected primary lung bronchial epithelial cells (Gene Expression receptor signaling. Mol. Cell. Biol. 27: 79–91. 23. Liao, F. H., J. W. Shui, E. W. Hsing, W. Y. Hsiao, Y. C. Lin, Y. C. Chan, Omnibus dataset GSE48466) (42, 43). Taken together with our T. H. Tan, and C. Y. Huang. 2014. Protein phosphatase 4 is an essential positive results that the expression level of PP4 is upregulated after RNA regulator for Treg development, function, and protective gut immunity. Cell virus infection and PP4 inhibited type I IFN production, it is rea- Biosci. 4: 25. 24. Wang, C., T. Chen, J. Zhang, M. Yang, N. Li, X. Xu, and X. Cao. 2009. The E3 sonable that some viruses may use PP4 to restrain and escape anti- ubiquitin Nrdp1 “preferentially” promotes TLR-mediated production of viral immunity, whereas which ones needs to be further investigated. type I interferon. Nat. Immunol. 10: 744–752. The Journal of Immunology 3857

25. Chen, W., C. Han, B. Xie, X. Hu, Q. Yu, L. Shi, Q. Wang, D. Li, J. Wang, 34. Goubau, D., S. Deddouche, and C. Reis e Sousa. 2013. Cytosolic sensing of P. Zheng, et al. 2013. Induction of Siglec-G by RNA viruses inhibits the innate viruses. Immunity 38: 855–869. immune response by promoting RIG-I degradation. Cell 152: 467–478. 35. Mustelin, T., T. Vang, and N. Bottini. 2005. Protein tyrosine phosphatases and 26. Liu, X., Z. Zhan, D. Li, L. Xu, F. Ma, P. Zhang, H. Yao, and X. Cao. 2011. the immune response. Nat. Rev. Immunol. 5: 43–57. Intracellular MHC class II molecules promote TLR-triggered innate immune 36. Long, L., Y. Deng, F. Yao, D. Guan, Y. Feng, H. Jiang, X. Li, P. Hu, X. Lu, responses by maintaining activation of the kinase Btk. Nat. Immunol. 12: 416–424. H. Wang, et al. 2014. Recruitment of phosphatase PP2A by RACK1 adaptor 27. An, H., J. Hou, J. Zhou, W. Zhao, H. Xu, Y. Zheng, Y. Yu, S. Liu, and X. Cao. protein deactivates transcription factor IRF3 and limits type I interferon sig- 2008. Phosphatase SHP-1 promotes TLR- and RIG-I-activated production of naling. Immunity 40: 515–529. type I interferon by inhibiting the kinase IRAK1. Nat. Immunol. 9: 542–550. 37. Bowie, A. G., and L. Unterholzner. 2008. Viral evasion and subversion of 28. Hou, J., P. Wang, L. Lin, X. Liu, F. Ma, H. An, Z. Wang, and X. Cao. 2009. pattern-recognition receptor signalling. Nat. Rev. Immunol. 8: 911–922. MicroRNA-146a feedback inhibits RIG-I-dependent type I IFN production in 38. Taylor, K. E., and K. L. Mossman. 2013. Recent advances in understanding viral macrophages by targeting TRAF6, IRAK1, and IRAK2. J. Immunol. 183: 2150– evasion of type I interferon. Immunology 138: 190–197. 2158. 39. Unterstab, G., S. Ludwig, A. Anton, O. Planz, B. Dauber, D. Krappmann, 29. Pomerantz, J. L., and D. Baltimore. 1999. NF-kB activation by a signaling G. Heins, C. Ehrhardt, and T. Wolff. 2005. Viral targeting of the interferon- complex containing TRAF2, TANK and TBK1, a novel IKK-related kinase. b-inducing Traf family member-associated NF-kB activator (TANK)-binding EMBO J. 18: 6694–6704. kinase-1. Proc. Natl. Acad. Sci. USA 102: 13640–13645. 30. Chen, L., W. Dong, T. Zou, L. Ouyang, G. He, Y. Liu, and Y. Qi. 2008. Protein 40. DiPerna, G., J. Stack, A. G. Bowie, A. Boyd, G. Kotwal, Z. Zhang, S. Arvikar, phosphatase 4 negatively regulates LPS cascade by inhibiting ubiquitination of E. Latz, K. A. Fitzgerald, and W. L. Marshall. 2004. Poxvirus protein N1L TRAF6. FEBS Lett. 582: 2843–2849. targets the I-kB kinase complex, inhibits signaling to NF-kB by the tumor ne- 31. Fitzgerald, K. A., S. M. McWhirter, K. L. Faia, D. C. Rowe, E. Latz, crosis factor superfamily of receptors, and inhibits NF-kB and IRF3 signaling by D. T. Golenbock, A. J. Coyle, S. M. Liao, and T. Maniatis. 2003. IKKε and Toll-like receptors. J. Biol. Chem. 279: 36570–36578. TBK1 are essential components of the IRF3 signaling pathway. Nat. Immunol. 4: 41. Kaukinen, P., M. Sillanpa¨a¨, L. Nousiainen, K. Mele´n, and I. Julkunen. 2013. 491–496. Hepatitis C virus NS2 protease inhibits host cell antiviral response by inhibiting 32. Cui, J., Y. Li, L. Zhu, D. Liu, Z. Songyang, H. Y. Wang, and R. F. Wang. 2012. IKKε and TBK1 functions. J. Med. Virol. 85: 71–82. NLRP4 negatively regulates type I interferon signaling by targeting the kinase 42. Manel, N., B. Hogstad, Y. Wang, D. E. Levy, D. Unutmaz, and D. R. Littman. TBK1 for degradation via the ubiquitin ligase DTX4. Nat. Immunol. 13: 387–395. 2010. A cryptic sensor for HIV-1 activates antiviral innate immunity in dendritic 33. Zhang, M., L. Wang, X. Zhao, K. Zhao, H. Meng, W. Zhao, and C. Gao. 2012. cells. Nature 467: 214–217. Downloaded from TRAF-interacting protein (TRIP) negatively regulates IFN-b production and 43. Gerlach, R. L., J. V. Camp, Y. K. Chu, and C. B. Jonsson. 2013. Early host antiviral response by promoting proteasomal degradation of TANK-binding ki- responses of seasonal and pandemic influenza A viruses in primary well- nase 1. J. Exp. Med. 209: 1703–1711. differentiated human lung epithelial cells. PLoS One 8: e78912. http://www.jimmunol.org/ by guest on October 4, 2021