And IRF3 Activation Enabling TRAF3/TBK1 Complex Formation

DOK3 Is Required for IFN-β Production by Enabling TRAF3/TBK1 Complex Formation and IRF3 Activation

This information is current as Susana Soo-Yeon Kim, Koon-Guan Lee, Ching-Siang Chin, of September 25, 2021. Say-Kong Ng, Natasha Ann Pereira, Shengli Xu and Kong-Peng Lam J Immunol 2014; 193:840-848; Prepublished online 13 June 2014;

doi: 10.4049/jimmunol.1301601 Downloaded from http://www.jimmunol.org/content/193/2/840

Supplementary http://www.jimmunol.org/content/suppl/2014/06/13/jimmunol.130160 Material 1.DCSupplemental http://www.jimmunol.org/ References This article cites 33 articles, 19 of which you can access for free at: http://www.jimmunol.org/content/193/2/840.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 25, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

DOK3 Is Required for IFN-b Production by Enabling TRAF3/TBK1 Complex Formation and IRF3 Activation

Susana Soo-Yeon Kim,*,† Koon-Guan Lee,† Ching-Siang Chin,† Say-Kong Ng,† Natasha Ann Pereira,† Shengli Xu,† and Kong-Peng Lam*,†,‡

The downstream of kinase (DOK) family of adaptors is generally involved in the negative regulation of signaling pathways. DOK1, 2, and 3 were shown to attenuate TLR4 signaling by inhibiting Ras-ERK activation. In this study, we elucidated a novel role for DOK3 in IFN-b production. Macrophages lacking DOK3 were impaired in IFN-b synthesis upon influenza virus infection or polyinosinic-polyribocytidylic acid stimulation. In the absence of DOK3, the transcription factor IFN regulatory factor 3 was not phosphorylated and could not translocate to the nucleus to activate ifn-b gene expression. Interestingly, polyinosinic- polyribocytidylic acid–induced formation of the upstream TNFR-associated factor (TRAF) 3/TANK-binding kinase (TBK) 1 2 2 complex was compromised in dok3 / macrophages. DOK3 was shown to bind TBK1 and was required for its activation. Downloaded from Furthermore, we demonstrated that overexpression of DOK3 and TBK1 could significantly enhance ifn-b promoter activity. DOK3 was also shown to bind TRAF3, and the binding of TRAF3 and TBK1 to DOK3 required the tyrosine-rich C-terminal domain of DOK3. We further revealed that DOK3 was phosphorylated by Bruton’s tyrosine kinase. Hence, DOK3 plays a critical and positive role in TLR3 signaling by enabling TRAF3/TBK1 complex formation and facilitating TBK1 and IFN regulatory factor 3 activation and the induction of IFN-b production. The Journal of Immunology, 2014, 193: 840–848. http://www.jimmunol.org/ he downstream of kinase (DOK) family of adaptors (6). DOK3 was also shown to have an inhibitory role as it restricts comprises seven structurally related proteins with each Ca2+ (7) and JNK (8) activation in B cell receptor signaling. T possessing an NH2-terminal pleckstrin homology (PH), Unlike B cells, which express DOK1 and 3, and T cells, which a central phosphotyrosine-binding (PTB), and a C-terminal express DOK1 and 2, myeloid cells express all three DOK adap- tyrosine-rich domain (1). DOK1–3 are preferentially expressed tors (1). The role of DOK proteins in myeloid cells is beginning in hematopoietic cells whereas DOK4–7 are found in neural and to be unraveled, and there is increasing evidence to indicate that other cell types (2). As these proteins have no catalytic activity, they are involved in TLR signaling and innate immunity. TLRs are they function mainly as adaptors to facilitate protein–protein pattern recognition receptors that bind pathogen-associated mo- by guest on September 25, 2021 interactions and possibly also as scaffolds to nucleate protein lecular patterns found on microbes (9). Examples of pathogen- complexes in signal transduction pathways. Recently, DOK1–3 associated molecular patterns include the bacterial cell wall com- were identified as tumor suppressors for lung cancer in human and ponent LPS that is recognized by TLR4 and viral dsRNA that is mice (3) and shown to play a role in preventing the development detected by TLR3 (10). DOK1 and 2 have been shown to participate of aggressive histiocytic sarcoma (4). in TLR4 signaling by negatively regulating ERK activation and Accumulating data suggest that DOK1–3 function to limit protein TNF-a production in LPS-stimulated macrophages (11). Recently, tyrosine kinase–mediated signaling in immune cells (5). In partic- DOK3 was also demonstrated to negatively regulate LPS-induced ular, DOK1 and 2 have been characterized as negative regulators of ERK activation and the production of inflammatory cytokines in Ras-ERK signaling downstream of Ag receptors in B and T cells macrophages (12). Thus, it seems that DOK1–3 act similarly to inhibit TLR4 signaling in macrophages. *Department of Pediatrics, Yong Loo Lin School of Medicine, National University of Whereas DOK1 and 2 were shown not to regulate TLR3 signaling Singapore, Singapore 119228; †Bioprocessing Technology Institute, Agency for Sci- ence, Technology and Research, Singapore 138668; and ‡Department of Microbiol- (11), the role of DOK3 in TLR3 signal transduction remains to be ogy, Yong Loo Lin School of Medicine, National University of Singapore, Singapore determined. TLR3 recognition of viral dsRNA or its synthetic an- 119228 alog, polyinosinic-polyribocytidylic acid [poly(I:C)] activates the Received for publication June 19, 2013. Accepted for publication May 13, 2014. TRIF-dependent signaling pathway, resulting in IFN-b production. This work was supported by the Agency for Science Technology and Research/ This signaling cascade leads to the formation of the TNFR- Biomedical Research Council. associated factor (TRAF) 3/TANK-binding kinase (TBK) 1 com- S.S.-Y.K., K.-G.L., C.-S.C., S.-K.N., N.A.P., and S.X. performed research; K.-P.L. conceptualized the project; and S.S.-Y.K., K.-G.L., and K.-P.L. analyzed data and plex and activation of TBK1 that subsequently leads to the phos- wrote the paper. phorylation and activation of the transcription factor IFN regulatory Address correspondence and reprint requests to Dr. Kong-Peng Lam, Bioprocessing Tech- factor (IRF) 3 (13). Activated IRF3 dimerizes and translocates from nology Institute, Agency for Science, Technology and Research, No. 06-01, Centros, 20 the cytoplasm to the nucleus to drive ifn-b mRNA synthesis (14). Biopolis Way, Singapore 138668. E-mail address: [email protected] Interestingly, dok3 mRNA was reported to be upregulated in The online version of this article contains supplemental material. virus-infected cells (15). This is in contrast to the demonstrated Abbreviations used in this article: BTK, Bruton’s tyrosine kinase; DOK, down- degradation of DOK1–3 proteins in LPS-stimulated macrophages stream of kinase; HA, hemagglutinin; HDAC1, histone deacetylase 1; IRF, IFN regulatory factor; NS, nonstructural; PH, pleckstrin homology; poly(I:C), polyinosinic- (12), suggesting that DOK3 could have a different role in TLR3 polyribocytidylic acid; PTB, phosphotyrosine-binding; TBK, TANK-binding kinase; activation in macrophages. In this study, we examined the involve- TRAF, TNFR-associated factor; WT, wild-type. ment of DOK3 in TLR3 signaling and showed that the adaptor was Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 phosphorylated upon poly(I:C) stimulation and played a critical www.jimmunol.org/cgi/doi/10.4049/jimmunol.1301601 The Journal of Immunology 841 role in TLR3-triggered IFN-b production. More importantly, we virus, 59-AAGGGCTTTCACCGAAGAGG-39 and 59-CCCATTCTCAT- demonstrate that DOK3 is needed for the induction of IRF3 by TACTGCTTC-39. facilitating the formation of the upstream TRAF3/TBK1 complex Transfection and reporter assays and the activation of TBK1 and IRF3. HEK293T cells (1 3 105) were seeded in 48-well dishes (Costar) and transfected the following day using Lipofectamine 2000 (Invitrogen). Empty control plasmid was added to ensure that each transfection received Materials and Methods the same amount of total DNA. To normalize for transfection efficiency, Mice, cells, and plasmids 100 ng pRL-TK (Renilla luciferase) reporter plasmid was added to each 2 2 transfection. Approximately 18 h after transfection, luciferase assays were Wild-type C57BL/6 and dok3 / mice were bred in our facilities 2 2 performed using a dual-specific luciferase assay kit (Promega). The IFN-b whereasBruton’styrosinekinase(btk) / mice were obtained from The 2 2 2 2 promoter luciferase reporter plasmid was obtained from Dr. Di Wang Jackson Laboratory (Bar Harbor, ME). tlr3 / and trif / mice were (Zhejiang University, Zhejiang, China). provided by Osamu Takeuchi and Shizuo Akira (Osaka University, Osaka, Japan). Experiments with mice were performed according to guide ELISA lines from the National Advisory Committee on Laboratory Animal Research. Macrophages were differentiated from bone marrow as pre- Untreated or naked poly(I:C)-stimulated or influenza virus–infected WT 2 2 viously described (16). HEK293T cells were transfected with recombi- and dok3 / macrophages were cultured in six-well plates at 5 3 106 nant vectors encoding murine TLR3, DOK3, TRAF, TBK1, TRIF, BTK,or cells/ml. Supernatants were harvested and assayed for cytokine produc- various truncated and site-directed mutants (Stratagene) with hemag- tion. The concentrations of IL-12p40, TNF-a (BD Pharmingen, San Diego, glutinin (HA), FLAG, or GFP tag using Lipofectamine 2000 (Invitrogen, CA), IFN-b (PBL InterferonSource), and RANTES (R&D Systems) were Grand Island, NY). Cells were stimulated with naked poly(I:C) or trans- determined using commercial ELISA kits. fected with poly(I:C) enclosed in liposomes [poly(I:C)/LyoVec] (Invivo- Downloaded from Gen, San Diego, CA). Determination of IRF3 activation by ELISA Activated mouse IRF3 was measured in cell nuclear extracts using Coimmunoprecipitations and immunoblotting ELISA according to the manufacturer’s instructions (TransAM; Active Cells were lysed on ice for 30 min in phospholysis buffer containing 1% Motif). Briefly, a 96-well plate of which oligonucleotides containing the Nonidet P-40, 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1 mM EDTA, 0.2 IRF3 consensus binding sites were immobilized were used to measure DNA binding activity of activated IRF3. The activated forms of IRF3 in mM Na3VO4, and a mixture of protease inhibitors (Roche Applied Science). Cell homogenates were centrifuged at 13,000 rpm for 15 min at 4˚C, and nuclear cell extracts bind to the oligonucleotides, and the amount bound http://www.jimmunol.org/ supernatants were recovered for protein quantification by a BCA protein is subsequently measured using Abs specific to IRF3. Protein concen- assay kit (Pierce). For immunoprecipitation studies, Abs were first coupled trations determined by a BCA protein assay kit (Pierce) were normal- to Protein A/G PLUS-Agarose (SC-2003; Santa Cruz Biotechnology, Santa ized across all samples prior to the ELISA and measurement was taken Cruz, CA) at 4˚C overnight. Beads were washed twice in lysis buffer and at 450 nm. incubated with precleared cell lysates for 2 h at 4˚C and subsequently boiled Virus infection in loading buffer for 5 min to release proteins. For Western blot analyses, 30 mg whole-cell lysates or 5 mg nuclear extracts isolated with NE-PER Influenza A virus, IVR-116 (A/New Caledonia/20/99 [H1N1] 3 IVR-6 nuclear and cytoplasmic extraction reagents kit (Pierce Biotechnology, [H3N2], NIBSC code 06/108), was obtained from the National Institute Rockford, IL) were electrophoresed in 10% SDS-polyacrylamide gels for Biological Standards and Control (Potters Bar, U.K.). The virus was and transferred onto immunoblot polyvinylidene difluoride membranes

propagated three times in Vero cells (ATCC CCL-81) in OptiPro serum- by guest on September 25, 2021 (Millipore, Billerica, MA). Membranes were subsequently probed with Abs free medium (Invitrogen, catalog no. 12309-019) supplemented with against various signaling molecules. Abs against DOK3, ERK2, phospho- 5 mg/ml porcine trypsin (Sigma-Aldrich, catalog no. T5266, 1500 N- 308 473 ERK2, phospho-AKT (Thr and Ser ), TRAF3, JNK1, p38, IRF3, IkBa, benzoyl-L-arginine ethyl ester units/mg) and subsequently stored at 280˚C. and histone deacetylase 1 (HDAC1) were from Santa Cruz Biotechnology. The virus was thawed at room temperature and used to infect WT and The anti-HA and anti-FLAG Abs were from Sigma-Aldrich (St. Louis, MO) dok32/2 primary macrophage cells in OptiMEM (Invitrogen) at a multi- Abs against phospho-p38, phospho-SAPK/JNK, phospho-AKT (Ser473), plicity of infection of 10 per cell for 3, 6, 12, or 24 h. The experiment was AKT, phospho-IRF3 (Ser396), phospho-TBK1 (Ser172), TBK1, and TRAF3 performed in triplicates. Total RNA was extracted from virus-infected cells were from Cell Signaling Technology. The anti-phosphotyrosine (4G10) Ab using TRIzol (Invitrogen), and viral DNA was detected with semiquanti- was from Upstate Biotechnology. tative RT-PCR using the following primers: forward, 59-AA- GGGCTTTCACCGAAGAGG-39,reverse,59-CCCATTCTCATTAC- Immunofluorescence and confocal microscopy TGCTTC-39. Densitometric analysis of bands obtained in semiquantitative RT-PCR analyses was carried out using a Bio-Rad imaging densi- Cells were washed twice with cold PBS containing 1% BSA and fixed 20 min tometer and multianalysis software (Bio-Rad Laboratories), and the pixel on ice with 4% paraformaldehyde in PBS. After permeabilization in 0.2% intensity (absorbance units/mm2) of the specific bands were normalized to saponin/0.03 M sucrose in 1% BSA/PBS at room temperature for 10 min, cells that corresponding to b-ACTIN, which acted as loading control, and this were washed twice with cold PBS and blocked with 5% normal goat serum in ratio was expressed as fold increase in intensity over the control non- 1% BSA/PBS at room temperature for 1 h before incubation with primary Abs stimulated sample. overnight at 4˚C. The slides were washed three times with 1% BSA/PBS, incubated 1 h at room temperature with anti-HA Alexa Fluor 594 anti-mouse Expression of recombinant proteins in bacteria and kinase IgG and chicken anti-rabbit Alexa Fluor 488 Abs (Invitrogen/Molecular activity assays Probes) to reveal the respective primary Abs. Slides were washed three times with 1% BSA/PBS, mounted, and viewed with an Olympus confocal The coding region of DOK3, BTK, and kinase-dead BTK (K430R) were laser scanning microscope under a 3100 oil objective. cloned into pGEX4T-1 (Pharmacia Biotech) and transformed into Escher- ichia coli BL21 (DE3) cells (Invitrogen). The cells were then cultured at Quantitative RT-PCR 37˚C until the OD600 value reached 0.6. Subsequently, isopropyl b-D-thi- 2/2 2/2 ogalactoside (Invitrogen) was added to the culture media to a final con- Total RNA was extracted from wild-type (WT), trif , and dok3 centration of 0.5 mM to induce protein expression for 2 h at 27˚C. Cells were macrophages using an RNeasy Mini kit (Qiagen). cDNA was synthesized harvested by centrifugation and lysed by sonication in PBS. GST-tagged with SuperScript II reverse transcriptase (Invitrogen) as per the manu- proteins were then purified by affinity chromatography with glutathione- facturer’s protocol. Quantitative PCR was performed with an Applied Sepharose beads (Amersham Biosciences) under native conditions. The Biosystems 7500 real-time PCR machine using the following primers: activity of the purified kinases and substrates was determined using the IFN-b,59-CAGCTCCAAGAAAGGACGAAC-39 and 59-GGCAGTGTAA- ADP-Glo kinase assay (Promega) as per the manufacturer’s instructions. CTCTT CTGCAT-39; RANTES, 59-GCTGCTTTGCCTACCTCTCC-39 and 59-TCGAGTGACAAACACGACTGC-39 ;IP10,59-CCAAGTGCTGCCG- Statistical analysis TCATTTTC-39 and 59-GGCTCGCAGGGATGATTCAA-39; b-actin, 59- AGATGACCCAGATCATGTTTGAGA-39 and 59-CACAGCCTGGATG- Statistical analysis was performed using an unpaired t test (Prism; GCTACGTA-39; DOK3, 59-TCCAAAAAGGGGCTTTGTTCC-39 and 59- GraphPad Software, San Diego, CA). A p value ,0.05 was considered GGAGGTAGGGTCCTTTCAGC-39; and nonstructural (NS) influenza statistically significant. 842 ROLE OF DOK3 IN IFN-b PRODUCTION

Results and dok32/2 macrophages with influenza virus and monitored their DOK3 is required for influenza virus and poly(I:C)-triggered production of IFN-b over time. As depicted in Fig. 1B, WT mac- IFN-b production rophages rapidly synthesized ifn-b mRNAwithin6hofvirusin- 2/2 It was shown previously by deep sequencing that dok3 mRNA was fection. However, dok3 macrophages had severe impairment of b upregulated within 2 h of vaccinia virus infection of HeLa cells ifn- mRNA synthesis at 6 h and remained so up to 24 h of infection. (15). However, the relevance of this observation for host antiviral As a result, they could not secrete any substantial amount of IFN-b response remains unclear. To determine the significance of this find- when assayed by ELISA at the 24 h time point after infection ing, we first examined whether WT macrophages were also able compared with WT infected controls (Fig. 1C). Concomitant with the to upregulate dok3 mRNA upon influenza virus infection. As shown defect in ifn-b gene induction, RT-PCR analysis of influenza virus NS in Fig. 1A, dok3 mRNA,asmeasuredbystandardRT-PCR,was protein mRNA at 12 h postinfection revealed more pronounced significantly increased in virus-infected cells. Next we infected WT presence of the viruses in mutant compared with WT cells (Fig. 1D). Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 1. DOK3-deficient macrophages are impaired in IFN-b production and influenza virus clearance. (A) Semiquantitative RT-PCR analyses of dok3 mRNA in WT macrophages at 24 h postinfection with influenza virus. (B) Real-time RT-PCR analyses of ifn-b mRNA synthesis in WT and dok32/2 macrophages infected with influenza virus for various times. (C) ELISA measurement of IFN-b secretion by WT and dok32/2 macrophages after 24 h of influenza virus infection. (D) Semiquantitative RT-PCR analysis of influenza virus NS RNA in WT and dok32/2 macrophages at 12 h postinfection. Densitometry was performed on the band corresponding to the influenza NS gene and normalized to that of b-ACTIN and this was set as 1 for the respective uninfected samples. The numerical values after comparison with control samples are provided below each corresponding lane. (E) Quantitative real-time PCR analysis of influenza virus NS mRNA in WT and dok32/2 macrophages at 24 h postinfection. (F) Quantitative real-time PCR analyses of ifn-b mRNA synthesis in WT, dok32/2,and trif2/2 macrophages stimulated with naked poly(I:C) for 2 h. (G) ELISA measurement of IFN-b secretion by poly(I:C)-stimulated WT and dok32/2 macrophages at 6 h time point. Quantitative real-time PCR analyses of IFN-b mRNA synthesis in WT and dok32/2 macrophages stimulated with (H) liposome- enclosed poly(I:C), (I) poly(deoxyadenylic-thymidylic) acid [p(dA:dT)], and (J) CpG-ODN are shown. Control experiments represented uninfected or non- stimulated cells. Expression of ifn-b mRNA was normalized to that of b-actin mRNA. Triplicate experiments were performed. *p , 0.005. The Journal of Immunology 843

This was further confirmed by real-time PCR quantification of in- to engage only TLR3 (16), to study the role of DOK3 in the influenza virus NS mRNA at 24 h postinfection, which showed the duction of the IFN response. presence of greater amounts of viral NS mRNA in virus-infected DOK3 deficiency impairs the activation of IRF3 dok32/2 macrophages compared with WT cells (Fig. 1E). Taken together, these data suggest that DOK3 plays a positive role in me- Expression of ifn-b mRNA requires the coordinated actions of diating IFN-b production, as its absence impairs ifn-b mRNA syn- three major transcription factors, those of AP-1, NF-kB, and IRF3 thesis and IFN-b protein secretion during host antiviral response, (18). It remains to be determined whether the absence of DOK3 thereby allowing the accumulationofagreateramountofviruses would affect any of these signaling pathways downstream of inside the mutant cells. poly(I:C) stimulation in macrophages. The induction of AP-1 is Influenza virus replicates via a dsRNA intermediate, and this is dependent on MAPK signaling (19). Because DOK1 and 2 are recognized by TLR3 and RIG-I (17). To address whether DOK3 could play a role in the signaling of these two innate immune receptors, we stimulated WT and dok32/2 cells with naked poly(I:C) to engage only TLR3 or transfected them with liposome-enclosed poly(I:C) to engage RIG-I. Our results indicated that WT macrophages readily synthesize ifn-b mRNA when treated with naked poly(I:C). Alternatively, dok32/2 macrophages behaved similar to trif2/2 macrophages in that they could not respond to naked poly(I:C) stimulation (Fig. 1F). To corroborate this finding, we also mea- Downloaded from sured IFN-b protein production and showed that dok32/2 macrophages were impaired in IFN-b secretion when stimulated with naked poly(I:C) for 6 h (Fig. 1G). As it was possible that DOK3 deficiency could affect other cytokine responses during TLR3 signaling, we also examined IL-12 (Supplemental Fig. 1A) and TNF-a

(Supplemental Fig. 1B) production in naked poly(I:C)-stimulated http://www.jimmunol.org/ WT and dok32/2 macrophages and found them to be unaffected in the absence of DOK3. Next we examined the response of WT and dok32/2 macrophages to RIG-I engagement by treating them with liposome-enclosed poly(I:C) or poly(I:C)/LyoVec. Mutant macrophages were also impaired in ifn-b mRNA synthesis when poly(I:C) was transfected into the cytosol of these cells using LyoVec (Fig. 1H). We further stimulated WT and dok32/2 macrophages with dsDNA/poly(deoxyadenylic-thymidylic) acid or CpG ODN, both of which were also known to induce IFN-b production in macro- by guest on September 25, 2021 phages, and found that ifn-b mRNA synthesis was unaffected by these stimuli in mutant cells (Fig. 1I , 1J). Hence, the data indicate that DOK3 is likely to be specifically involved in macrophage responses to RNA ligands through the engagement of TLR3 and RIG-I and leads to IFN-b production. DOK3 is phosphorylated in macrophages upon poly(I:C) stimulation DOK3 is phosphorylated upon B cell Ag receptor engagement (6). Hence, we examined whether DOK3 would also be phosphorylated when WT macrophages were treated with naked poly(I:C) or transfected with poly(I:C)/LyoVec as a further indication that the adaptor could be involved in TLR3 and RIG-I signaling. Western blot analysis of immunoprecipitated DOK3 revealed that the protein was indeed tyrosine phosphorylated in naked poly(I:C)- stimulated WT macrophages (Fig. 2A). This phosphorylation was dependent on TRIF signaling, as it was abrogated in naked poly(I:C)- 2/2 stimulated trif macrophages (Fig. 2B). As a further indication FIGURE 2. DOK3 is phosphorylated downstream of TRIF upon poly(I:C) that DOK3 was involved in TRIF signaling, gene induction of the stimulation. (A) WT macrophages were stimulated with naked poly(I:C) for TRIF-dependent chemokines Rantes (Fig. 2C) and IP10 (Fig. 2D) various times and DOK3 was immunoprecipitated and examined for phos- were also defective in naked poly(I:C)-stimulated dok32/2 macro- phorylation via Western blot analysis using anti-phosphotyrosine (p-Tyr) Ab. 2/2 phages. We also measured the secretion of RANTES protein and The anti-DOK3 blot was included as control. (B)WTandtrif macrophages confirmed that it was indeed defective in naked poly(I:C)-stimulated were untreated or stimulated with naked poly(I:C) for 30 min and examined for A C D dok32/2 macrophages (Fig. 2E). Furthermore, we showed that DOK3 phosphorylation as in ( ). ( and ) Semiquantitative real-time RT- PCR analyses of (C) Rantes and (D) ip10 mRNA in WT and dok32/2 mac- DOK3 was phosphorylated in poly(I:C)/LyoVec-transfected WT rophages at 2 h after stimulation with naked poly(I:C). (E) ELISA measure- macrophages, which signal via RIG-I (Fig. 2F). Taken together, the ment of RANTES secretion by WT and dok32/2 macrophages stimulated with data indicate that DOK3 is phosphorylated upon TLR3 and RIG-I naked poly(I:C). (F) DOK3 phosphorylation in WT macrophages transfected engagement and is required for these receptors to signal IFN-b with liposome-enclosed poly(I:C). Western blot analysis was performed as in production. For subsequent experiments, we mainly used naked (A). Data shown are representative of more than three independent experi- poly(I:C) stimulation of macrophages, which we showed previously ments. *p , 0.005. 844 ROLE OF DOK3 IN IFN-b PRODUCTION known to participate in Ras-ERK signaling downstream of TLR4 cytoplasms of poly(I:C)-treated dok32/2 macrophages, similar to (11) and growth receptors (20), and DOK3 is involved in JNK the situations found in poly(I:C)-treated trif2/2 and tlr32/2 mac- signaling downstream of immune receptors (21), we examined rophages. This was further confirmed by Western blot analysis whether ERK, JNK, and p38 MAPK signaling would be affected demonstrating IRF3 nuclear localization in poly(I:C)-stimulated in poly(I:C)-stimulated dok32/2 macrophages. Surprisingly, the WT but not dok32/2 macrophages (Fig. 3B). The lack of nu- induction of these MAPKs was largely normal in the mutant clear IRF3 in poly(I:C)-stimulated dok32/2 macrophages was cells (Supplemental Fig. 2A). We next examined the activation not due to altered expression of IRF3, as we showed comparable of NF-kB and again found this signaling pathway to be intact, as expression of cytosolic IRF3 (Supplemental Fig. 2C) and com- the kinetics of degradation of the inhibitory IkB subunit was parable presence of basal nuclear IRF3 (Supplemental Fig. 2D) comparable between poly(I:C)-stimulated WT and dok32/2 mac- in unstimulated WT and dok32/2 macrophages. It was also not rophages (Supplemental Fig. 2B). Thus, DOK3 deficiency did not due to improper isolation of cytosolic and nuclear fractions, as affect NF-kB or MAPK signaling in naked poly(I:C)-stimulated we had shown using anti-HDAC1 and anti-tubulin Abs that we had macrophages. obtained good separation of nuclear and cytosolic fractions, Finally, we investigated the activation of IRF3 in poly(I:C)- respectively (Supplemental Fig. 2E). stimulated dok32/2 macrophages. This transcription factor is ac- Other than nuclear translocation, we also examined IRF3 tivated by serine phosphorylation and translocates from the cy- activation by phosphorylation. Indeed, IRF3 activation as indi- toplasm to the nucleus of cells to bind the ifn-b gene promoter cated by Ser396 phosphorylation was also defective in poly(I:C)- (22). Our confocal microscopy analyses indicated that IRF3 was stimulated dok32/2 macrophages (Fig. 3C). Finally, our ELISA initially cytoplasmic-bound in nontreated cells but located pre- measuring IRF3 promoter/DNA binding activity demonstrated Downloaded from dominantly in the nucleus of poly(I:C)-stimulated WT macro- that IRF3 was indeed activated in poly(I:C)-stimulated WT but not phages (Fig. 3A). In contrast, IRF3 was largely retained in the dok32/2 macrophages (Fig. 3D). Taken together, our data indicate http://www.jimmunol.org/

FIGURE 3. Defective IRF3 phosphorylation and nuclear localization in poly(I:C)-stimulated macrophages lacking DOK3. (A) Confocal microscopy ex- by guest on September 25, 2021 amining IRF3 nuclear localization in nontreated (media) and poly(I:C)-stimulated WT, dok32/2, trif2/2,and tlr32/2 macrophages. Cells were stained with FITC-la- beled anti-IRF3 Ab (green). Nuclei of cells were stained with DAPI (blue). Original magnification 3100. (B) Western blot analysis examining nuclear IRF3 in poly(I:C)-stimulated WT and dok32/2 macrophages. Nuclear extracts from nontreated and stimulated cells were first probed with anti-IRF3 Ab and subsequently with anti-HDAC1 Ab to serve as loading control. (C) Western blot analysis examining IRF3 phosphorylation in poly(I:C)-treated WT and dok32/2 macrophages. Cells were stimulated for various times and whole-cell lysates were probed with an Ab that recognized phospho-Ser396 residue of IRF3. The IRF3, tubulin, and b-actin blots were included as loading controls. Data shown are representative of more than three independent experiments. (D)IRF3promoter/ DNA-binding activity was measured using poly(I:C)- stimulated nuclear extract proteins obtained at 2 h time point from WT and dok32/2 macrophages and assayed using an IRF3 TransAM (Active Motif) ELISA kit. Data shown are representative of two independent experiments. The Journal of Immunology 845 that DOK3 deficiency specifically impairs IRF3 but not NF-kBor and other innate immune receptor signaling to activate IRF3 (26). MAPK signaling and that it likely affects IRF3 activation farther Furthermore, in TLR3 signaling, the TRAF3/TBK1 complex is upstream in the signaling cascade. recruited to the TRIF adaptor following activation (27). Hence, we first examined whether TRAF3/TBK1 complex formation would DOK3 is required for TBK1 activation and synergizes with be compromised in the absence of DOK3. As shown in Fig. 5A, TBK1 to induce ifn-b gene expression we were able to coimmunoprecipitate both TRAF3 with TBK1 The lack of IRF3 phosphorylation in poly(I:C)-stimulated (top panel) and TBK1 with TRAF3 (bottom panel) in poly(I:C)- 2/2 dok3 macrophages prompted us to examine whether the stimulated WT macrophages but were unable to do so in the mu- activation of kinases required for IRF3 phosphorylation would tant cells, indicating that TRAF3 and TBK1 interaction did not occur be compromised. Two kinases are important for the activation of when DOK3 was absent. Because DOK3 is an adaptor protein, we IRF3, namely TBK1 (23) and AKT (24), and the activation of directly assessed whether it could bind either TBK1 or TRAF3 or AKT is further dependent on TBK1 during TLR3 signaling (25). both. Our overexpression studies indicated that TBK1 could coim- Our Western blot analyses indicated that the activation of AKT, as munoprecipitate with DOK3 (Fig. 5B), suggesting that DOK3 308 473 indicated by phosphorylation of its Thr and Ser residues, was physically binds TBK1. We also performed endogenous coimmu- 2/2 defective in poly(I:C)-stimulated dok3 macrophages (Fig. 4A). noprecipitation experiments using WT macrophages and surprisingly More importantly, the activation of TBK1, as determined by ex- found DOK3 and TBK1 to be constitutively bound to each other 172 amining the phosphorylation of its Ser residue, was also im- (Fig. 5C, left and right panels). We also examined the interaction 2/2 paired in poly(I:C)-stimulated dok3 macrophages (Fig. 4B). between TRAF3 and DOK3 and demonstrated that the two mole-

These data together indicated that the impairment in IRF3 acti- cules bind each other (Fig. 5B). Furthermore, the binding of TRAF3 Downloaded from vation was mainly due to defective TBK1 activation in poly(I:C)- to DOK3 appeared to be more intense than the binding of TRAF2 or 2/2 stimulated dok3 macrophages. To better understand the rela- TRAF6 to DOK3, suggesting that the binding of TRAF3 to DOK3 is tionship between DOK3 and TBK1, we overexpressed DOK3 or specific (Supplemental Fig. 3A). We also tested the binding of TBK1 or both in HEK293T cells and examined their effect in DOK3 to TRIF and showed that the two molecules did not bind each inducing ifn-b promoter activity using a luciferase-based reporter other, unlike the interaction of TRAF3 with TRIF, which were b assay (Fig. 4C). DOK3 alone had no effect on the induction of ifn- known to interact with each other (28) and served as controls for our http://www.jimmunol.org/ promoter activity whereas TBK1 alone could induce a certain level studies (Supplemental Fig. 3B). These data together suggest that of activity. Interestingly, when DOK3 was expressed in increasing DOK3 likely acts as a bridge to facilitate the formation of the concentrations with TBK1 in HEK293T cells, there was substantial TRAF3/TBK1 complex that is necessary for TBK1 and IRF3 acti- induction of ifn-b promoter activity that was significantly higher vation, leading to IFN-b production. than that achieved with TBK1 alone. Hence, DOK3 acts in concert DOK3 is a multidomain adaptor containing a PH, PTB, and C- with TBK1 to further induce ifn-b promoter activation. terminal domain (29). To better understand the interaction between DOK3 binds TBK1 and TRAF3 and is required for TRAF3/ TBK1 and DOK3, we expressed various mutant forms of DOK3 TBK1 complex formation bearing specific or combinations of different domains (as illus- trated in Fig. 5D) and examined their abilities to further enhance by guest on September 25, 2021 Our data so far suggested that DOK3 acted at the level of TBK1 TBK1 activation of ifn-b promoter activity. It was apparent from activation in poly(I:C)-stimulated macrophages (Fig. 4). TBK1 is these studies that the C-terminal domain of DOK3 was required to knowntoformacomplexwithTRAF3downstreamofTLR3,RIG-I, act in concert with TBK1 to induce ifn-b promoter activity, as its deletion alone (as in the mutant form D2) substantially impaired the induction of the reporter luciferase gene expression (Fig. 5E). This finding together with the data obtained in Fig. 5B also implied that the C-terminal domain of DOK3 might be responsible for binding TBK1. To test this possibility, we overexpressed the various mutant forms of DOK3 and TBK1 in HEK293T cells and examined their physical associations via coimmunoprecipitation experiments (Fig. 5F). Indeed, TBK1 could coimmunoprecipitate with full-length DOK3 as well as the mutant forms bearing the PTB and C-terminal domains (D4) or the mutant form bearing only the C-terminal domain (D5), indicating that the C-terminal domain of DOK3 binds TBK1. Similar experiments also indicated that the C-terminal domain of DOK3 binds TRAF3 (Fig. 5G). Hence, DOK3 binds both TBK1 and TRAF3 via its C-terminal domain. DOK3 is phosphorylated by BTK in poly(I:C)-induced FIGURE 4. DOK3 deficiency impairs the activation of TBK1 upstream of IRF3. Western blot analyses are shown of poly(I:C)-stimulated WT and signaling pathway 2 2 dok3 / macrophages examining (A) AKT activation using an Ab that Because DOK3 is phosphorylated upon poly(I:C) stimulation 308 473 recognized phospho-Thr (top panel) and phospho-Ser (bottom panel) (Fig. 2A), we were also interested to identify the kinase responsi- residues of AKT and (B) TBK1 activation using an anti–phospho-TBK1 ble for this event. We recently demonstrated a role for BTK in A B Ab. Blots were also probed with anti-AKT ( ) and anti-TBK1 ( ) Abs as TLR3 signaling (16) and wondered whether BTK could phos- loading controls. (C) DOK3 synergizes with TBK1 to induce ifn-b pro- phorylate DOK3. To examine this possibility, we immunoprecipi- moter activity. HEK293T cells were transfected with vector alone (control) 2/2 or with vectors expressing DOK3 or TBK1 or both (with DOK3 in in- tated DOK3 from poly(I:C)-stimulated WT and btk macrophages creasing dosage of 20, 50 and 100 ng) and together with plasmids bearing and showed that the adaptor protein was not phosphorylated in IFN-b promoter luciferase reporter (IFN-b-luc) and Renilla luciferase for the mutant cells (Fig. 6A). To definitively demonstrate that normalization. Luciferase activity was measured at 24 h. Data shown are TLR3-induced DOK3 phosphorylation is BTK-dependent, we representative of more than three independent experiments. *p , 0.001. also immunoprecipitated DOK3 from poly(I:C)-stimulated but 846 ROLE OF DOK3 IN IFN-b PRODUCTION Downloaded from http://www.jimmunol.org/

FIGURE 5. DOK3 is required for TRAF3/TBK1 complex formation and binds TRAF3 and TBK1 via its C-terminal domain. (A) Defective formation of by guest on September 25, 2021 TRAF3/TBK1 complex in poly(I:C)-stimulated dok32/2 macrophages. WT and dok32/2 macrophages were nontreated or stimulated with poly(I:C) for 60 min and TRAF3 was immunoprecipitated (IP) to examine TBK1 association (top panel) and vice versa with TBK1 immunoprecipitated to examine TRAF3 association (bottom panel) via Western blot analyses. (B) DOK3 binds TRAF3 and TBK1. HEK293T cells were cotransfected with FLAG-tagged DOK3 and HA-tagged TBK1 or HA-tagged TRAF3 and cell lysates were immunoprecipitated with anti-FLAG Ab and immunoblotted (IB) with anti-HA Ab to examine possible association of DOK3 with TBK1 and TRAF3. The FLAG immunoblot was included to show equivalent immunoprecipitation of FLAG- tagged DOK3 whereas whole-cell lysates (WCL) were immunoblotted to demonstrate expression of HA-tagged TBK1 or TRAF3 proteins. (C) Western blot analyses of endogenous TBK1/DOK3 interaction in primary macrophages that were nontreated (0 min) or stimulated for 60 min with naked poly(I:C). Western blots were performed as in (A). (D) Pictogram depicting FLAG-tagged full-length DOK3 protein or mutants harboring various combinations of PH, PTB, or tyrosine-rich carboxyl terminal domains. (E) The C-terminal domain of DOK3 acts in concert with TBK1 to further induce ifn-b promoter activity. HEK293T cells were transfected with empty vectors (control) or vectors carrying TBK1 and/or DOK3 or various mutant forms of DOK3 and assayed for the induction of luciferase activity as in Fig. 4C. Data shown are representative of three independent experiments. *p , 0.001. (F and G) The C-terminal domain of DOK3 binds TBK1 and TRAF3. HEK293T cells were transfected with either HA-tagged TBK1 (F) or HA-tagged TRAF3 (G) together with FLAG-tagged full-length DOK3 or various DOK3 mutants. Cell lysates were immunoprecipitated with anti-HA Ab and immunoblotted with anti-FLAG Abs to examine for association and with anti-HA for immunoprecipitation control. Whole-cell lysates were also included to ensure expression of the various transfected constructs. *NS, nonspeciﬁc band. Data shown are representative of three independent experiments.

BTK-inhibited (LFM-A13) or control DMSO-treated macrophages lated in the presence of a kinase-dead BTK (K430R) (Fig. 6E). We and showed that the adaptor protein was also not phosphorylated in also transfected HEK293T cells with HA-tagged or DOK3 4YF LFM-A13–treated and therefore BTK-inhibited cells (Fig. 6B). mutant in which the four tyrosine residues in the carboxyl ter- Furthermore, coexpression of DOK3 with BTK or with a constitu- minal is mutated to phenylalanine, together with FLAG-tagged tively active form of BTK (E41K) led to the phosphorylation of BTK, constitutive-active BTK (E41K), kinase-dead BTK (K430R), DOK3 in HEK293T cells whereas the coexpression of DOK3 or control GFP-tagged c-ABL (Supplemental Fig. 4B). Our data with a kinase-dead form of BTK (K430R) did not lead to the indicated that BTK phosphorylates one or more of the tyrosine phosphorylation of DOK3 (Fig. 6C). Next, we proceeded to ex- residues in the C-terminal of DOK3. amine whether BTK could associate with DOK3 by coexpressing Finally, we examined the effect of BTK coexpression on the HA-tagged BTK and FLAG-tagged DOK3 in HEK293T cells and ability of TBK1 and DOK3 to induce ifn-b mRNA synthesis using demonstrated that the two proteins could physically bind each the luciferase reporter assay as described in Fig. 4C. As shown in other (Fig. 6D). Finally, we performed an in vitro kinase phosphory- Fig. 6F, the presence of BTK led to much greater induction of lation assay using GST-puriﬁed BTK and DOK3 as substrate ifn-b promoter activity than that achieved with TBK1 alone or with (Supplemental Fig. 4A) and demonstrated that BTK could directly TBK1 and DOK3 together. Taken together, the data suggested that phosphorylate DOK3 in vitro and that DOK3 was not phosphory- DOK3 is likely phosphorylated by BTK during TLR3 signaling. The Journal of Immunology 847

dok32/2 macrophages and demonstrated that the adaptor protein was required for the induction of IRF3 and production of IFN-b.Further biochemical characterizations indicated that DOK3 could bind both TRAF3 and TBK1 but not TRIF, and that in the absence of DOK3, the formation of the TRAF3/TBK1 complex and the activation of TBK1, as indicated by its Ser172 phosphorylation, was defective. These data suggest that DOK3 acts as a bridge to bring TRAF3 and TBK1 together and is also needed for TBK1 activation that is required for the subsequent induction of IRF3 and IFN-b production. It is currently not known how TBK1 is activated by phosphorylation. A recent protein crystallization study suggested that TBK1 was activated by transautophosphorylation(30).Accordingtothismodel, TBK1 is first maintained in an inactive state until recruited to a signaling platform where it can undergo autoactivation owing to greater concentration of the kinase present or activation by another effector kinase that is recruited to the same scaffold. Our current findings fit into this model in that we show DOK3 binds TBK1 and that DOK3 is needed for TBK1 activation by phosphorylation and

TBK1 complex formation with TRAF3. Thus, DOK3 could be part Downloaded from of the signaling scaffold needed for the recruitment and activation of TBK1. We further identiﬁed the SH2-target motif of DOK3 to be important for its interaction with TBK1, and consistent with this ﬁnding, coexpression of DOK3 with TBK1 synergizes IFN-b promoter activity whereas a truncated DOK3 lacking the C terminus did

not manifest this synergy. Hence, these data strongly supported the http://www.jimmunol.org/ hypothetical model in which DOK3 is the scaffold that served to concentrate TBK1 for activation, either through autophosphorylation or by a different effector kinase. Additionally, given that the TRAF3/ FIGURE 6. BTK phosphorylates DOK3. (A) DOK3 is not phosphorylated TBK1 complex is formed downstream of many immune signaling 2/2 in poly(I:C)-stimulated btk macrophages. Western blot analysis of DOK3 pathways (31), our data further suggest that DOK3 could be involved 2/2 phosphorylation in WT and btk macrophages upon poly(I:C) stimulation is in the signaling of innate immune receptors other than TLR3 and shown. Cells were stimulated with poly(I:C) and cell lysates were immuno- RIG-I. This notion could be explored in future experiments. Alter- precipitated (IP) with an anti-DOK3 Ab or control IgG Abs and immuno- natively, our stimulation with dsDNA and CpG suggested that blotted (IB) with an anti-phosphotyrosine Ab (4G10). Membranes were by guest on September 25, 2021 subsequently stripped and immunoblotted with DOK3 Ab for loading control. DOK3 might be specific for the innate sensing of RNA ligands. (B) DOK3 is phosphorylated in poly(I:C)-stimulated but not BTK-inhibited We recently demonstrate a role for BTK in phosphorylating TLR3 macrophages. Cells were treated with DMSO as control or with the BTK- during an antiviral response (16). In the present study, we identify inhibitor LFM-A13, and DOK3 was immunoprecipitated and probed with another target of BTK in TLR3 signaling by showing that it could also anti-phosphotyrosine Ab and reprobed with anti-DOK3 Ab as control. (C) phosphorylate DOK3 upon poly(I:C) stimulation. Hence, there seems BTK kinase activity is required for DOK3 phosphorylation. HEK293T cells to be a broader and more coordinated role for BTK in the innate an- were nontransfected or transfected with FLAG-tagged DOK3 together with tiviral response. A global proteome study of BTK’s binding partners HA-tagged BTK, constitutively active BTK (E41K), or kinase-dead BTK could help to identify new targets of BTK in the host antiviral response. (K430R). Cell lysates were immunoprecipitated with anti-FLAG Ab and Myeloid cells express DOK1–3 (1). DOK1 and 2 had been shown immunoblotted with anti-phosphotyrosine (4G10) or anti-FLAG Abs. Whole- to play a role in TLR4 signaling where they negatively regulated cell lysates (WCL) were included to show expression of the transfected gene constructs. (D) BTK physically binds DOK3. HEK293T cells were trans- ERK activation and TNF-a production in macrophages (11). Previ- fected with FLAG-tagged DOK3 and with or without HA-tagged BTK. Cell ously, DOK3 has largely been characterized in the context of B cell lysates were immunoprecipitated with anti-FLAG Ab and immunoblotted receptor (32) and FcgRIIb (33) signaling where it serves as a negative with anti-HA Ab. Whole-cell lysates were included to indicate expression of regulator by interacting with Grb2 and the inhibitor SHIP to restrict the transfected constructs. (E) In vitro kinase phosphorylation assay exam- the intensity of Ca2+ mobilization (7) and limit JNK activation (8). ining the activities of purified WT and kinase-dead BTK (K430R) kinases on While this manuscript was being prepared, a report was published DOK3 as substrate as measured with the ADP-Glo kinase assay. Data are describing DOK3 as a negative regulator of TLR4 signaling in the plotted as nanomoles phosphate transferred from ATP to a specific substrate. same manner as DOK1 and 2 by inhibiting ERK as it was degraded F b ( ) BTK acts in concert with DOK3 and TBK1 to further increase ifn- gene upon LPS but not poly(I:C) stimulation (12). In contrast, we dem- expression. HEK293T cells were transfected with various combinations of onstrate here a positive role for DOK3 in the induction of IFN-b vectors bearing TBK1, DOK3, and BTK together with plasmids encoding IFN-b promoter luciferase (IFN-b-luc) reporter and Renilla luciferase and production during virus infection and poly(I:C) stimulation of mac- assayed at 24 h after transfection for luciferase activity. Data shown are rophages. We observed that DOK3 did not undergo degradation representative of two independent experiments. * p , 0.05. upon poly(I:C) stimulation, consistent with previous findings (12). Its presence facilitates TBK1 and TRAF3 interaction during virus infection and likely enables maximal transduction of downstream signaling for optimal IFN-b production. Discussion Our present study arises as a follow-up of a published observation Acknowledgments indicating that dok3 mRNA was induced during virus infection Wethank members ofthe laboratoryfor discussions and Dr. Wang Di (Zhejiang (15). Thus, we explored the role of DOK3 in innate immunity by University, Zhejiang, China) for the gift of the IFN-b promoter luciferase re- employing poly(I:C)-stimulation and influenza virus infection of porter plasmids. We thank Swan-Li Poh for the propagation of influenza virus. 848 ROLE OF DOK3 IN IFN-b PRODUCTION

Disclosures 17. Schro¨der, M., and A. G. Bowie. 2005. TLR3 in antiviral immunity: key player or bystander? Trends Immunol. 26: 462–468. The authors have no financial conflicts of interest. 18. Kim, T., T. Y. Kim, W. G. Lee, J. Yim, and T. K. Kim. 2000. Signaling pathways to the assembly of an interferon-b enhanceosome. Chemical genetic studies with a small molecule. J. Biol. Chem. 275: 16910–16917. References 19. Ameyar, M., M. Wisniewska, and J. B. Weitzman. 2003. A role for AP-1 in 1. Mashima, R., Y. Hishida, T. Tezuka, and Y. Yamanashi. 2009. The roles of Dok apoptosis: the case for and against. Biochimie 85: 747–752. family adapters in immunoreceptor signaling. Immunol. Rev. 232: 273–285. 20. Murakami, H., Y. Yamamura, Y. Shimono, K. Kawai, K. Kurokawa, and 2. Favre, C., A. Ge´rard, E. Clauzier, P. Pontarotti, D. Olive, and J. A. Nune`s. 2003. M. Takahashi. 2002. Role of Dok1 in cell signaling mediated by RET tyrosine DOK4 and DOK5: new Dok-related genes expressed in human T cells. Genes kinase. J. Biol. Chem. 277: 32781–32790. Immun. 4: 40–45. 21. Ng, C. H., S. Xu, and K. P. Lam. 2007. Dok-3 plays a nonredundant role in 3. Berger, A. H., M. Niki, A. Morotti, B. S. Taylor, N. D. Socci, A. Viale, negative regulation of B-cell activation. Blood 110: 259–266. C. Brennan, J. Szoke, N. Motoi, P. B. Rothman, et al. 2010. Identification of 22. Hiscott, J. 2007. Triggering the innate antiviral response through IRF-3 activa- DOK genes as lung tumor suppressors. Nat. Genet. 42: 216–223. tion. J. Biol. Chem. 282: 15325–15329. 4. Mashima, R., K. Honda, Y. Yang, Y. Morita, A. Inoue, S. Arimura, H. Nishina, 23. McWhirter, S. M., K. A. Fitzgerald, J. Rosains, D. C. Rowe, D. T. Golenbock, H. Ema, H. Nakauchi, B. Seed, et al. 2010. Mice lacking Dok-1, Dok-2, and and T. Maniatis. 2004. IFN-regulatory factor 3-dependent gene expression is Dok-3 succumb to aggressive histiocytic sarcoma. Lab. Invest. 90: 1357–1364. defective in Tbk1-deficient mouse embryonic fibroblasts. Proc. Natl. Acad. Sci. 5. Veillette, A., S. Latour, and D. Davidson. 2002. Negative regulation of immu- USA 101: 233–238. noreceptor signaling. Annu. Rev. Immunol. 20: 669–707. 24. Sarkar, S. N., K. L. Peters, C. P. Elco, S. Sakamoto, S. Pal, and G. C. Sen. 2004. 6. Yamanashi, Y., T. Tamura, T. Kanamori, H. Yamane, H. Nariuchi, T. Yamamoto, Novel roles of TLR3 tyrosine phosphorylation and PI3 kinase in double-stranded and D. Baltimore. 2000. Role of the rasGAP-associated docking protein p62dok RNA signaling. Nat. Struct. Mol. Biol. 11: 1060–1067. in negative regulation of B cell receptor-mediated signaling. Genes Dev. 14: 11– 25. Joung, S. M., Z. Y. Park, S. Rani, O. Takeuchi, S. Akira, and J. Y. Lee. 2011. Akt 16. contributes to activation of the TRIF-dependent signaling pathways of TLRs by 7. Stork, B., K. Neumann, I. Goldbeck, S. Alers, T. Ka¨hne, M. Naumann, interacting with TANK-binding kinase 1. J. Immunol. 186: 499–507. M. Engelke, and J. Wienands. 2007. Subcellular localization of Grb2 by the 26. Gatot, J. S., R. Gioia, T. L. Chau, F. Patrascu, M. Warnier, P. Close, J. P. Chapelle, adaptor protein Dok-3 restricts the intensity of Ca2+ signaling in B cells. EMBO E. Muraille, K. Brown, U. Siebenlist, et al. 2007. Lipopolysaccharide-mediated Downloaded from J. 26: 1140–1149. interferon regulatory factor activation involves TBK1-IKKε-dependent Lys63- 8. Robson, J. D., D. Davidson, and A. Veillette. 2004. Inhibition of the Jun linked polyubiquitination and phosphorylation of TANK/I-TRAF. J. Biol. Chem. N-terminal protein kinase pathway by SHIP-1, a lipid phosphatase that interacts 282: 31131–31146. with the adaptor molecule Dok-3. Mol. Cell. Biol. 24: 2332–2343. 27. Guo, B., and G. Cheng. 2007. Modulation of the interferon antiviral response by 9. Sandor, F., and M. Buc. 2005. Toll-like receptors. II. Distribution and pathways the TBK1/IKKi adaptor protein TANK. J. Biol. Chem. 282: 11817–11826. involved in TLR signalling. Folia Biol. (Praha) 51: 188–197. 28. Ha¨cker, H., V. Redecke, B. Blagoev, I. Kratchmarova, L. C. Hsu, G. G. Wang, 10. Takeda, K., and S. Akira. 2005. Toll-like receptors in innate immunity. Int. M. P. Kamps, E. Raz, H. Wagner, G. Ha¨cker, et al. 2006. Specificity in Toll-like

Immunol. 17: 1–14. receptor signalling through distinct effector functions of TRAF3 and TRAF6. http://www.jimmunol.org/ 11. Shinohara, H., A. Inoue, N. Toyama-Sorimachi, Y. Nagai, T. Yasuda, H. Suzuki, Nature 439: 204–207. R. Horai, Y. Iwakura, T. Yamamoto, H. Karasuyama, et al. 2005. Dok-1 and 29. Cong, F., B. Yuan, and S. P. Goff. 1999. Characterization of a novel member of Dok-2 are negative regulators of lipopolysaccharide-induced signaling. J. Exp. the DOK family that binds and modulates Abl signaling. Mol. Cell. Biol. 19: Med. 201: 333–339. 8314–8325. 12. Peng, Q., J. L. O’Loughlin, and M. B. Humphrey. 2012. DOK3 negatively 30. Ma, X., E. Helgason, Q. T. Phung, C. L. Quan, R. S. Iyer, M. W. Lee, regulates LPS responses and endotoxin tolerance. PLoS ONE 7: e39967. K. K. Bowman, M. A. Starovasnik, and E. C. Dueber. 2012. Molecular basis of 13. Vercammen, E., J. Staal, and R. Beyaert. 2008. Sensing of viral infection and Tank-binding kinase 1 activation by transautophosphorylation. Proc. Natl. Acad. activation of innate immunity by Toll-like receptor 3. Clin. Microbiol. Rev. 21: Sci. USA 109: 9378–9383. 13–25. 31. Siu, K. L., K. H. Kok, M. H. Ng, V. K. Poon, K. Y. Yuen, B. J. Zheng, and 14. Hiscott, J., and R. Lin. 2005. IRF-3 releases its inhibitions. Structure 13: 1235– D. Y. Jin. 2009. Severe acute respiratory syndrome coronavirus M protein 1236. inhibits type I interferon production by impeding the formation of TRAF3. ε 15. Yang, Z., D. P. Bruno, C. A. Martens, S. F. Porcella, and B. Moss. 2010. Si- TANK.TBK1/IKK complex. J. Biol. Chem. 284: 16202–16209. by guest on September 25, 2021 multaneous high-resolution analysis of vaccinia virus and host cell tran- 32. Lemay, S., D. Davidson, S. Latour, and A. Veillette. 2000. Dok-3, a novel scriptomes by deep RNA sequencing. Proc. Natl. Acad. Sci. USA 107: 11513– adapter molecule involved in the negative regulation of immunoreceptor sig- 11518. naling. Mol. Cell. Biol. 20: 2743–2754. 16. Lee, K. G., S. Xu, Z. H. Kang, J. Huo, M. Huang, D. Liu, O. Takeuchi, S. Akira, 33. Neumann, K., T. Oellerich, I. Heine, H. Urlaub, and M. Engelke. 2011. Fcg and K. P. Lam. 2012. Bruton’s tyrosine kinase phosphorylates Toll-like receptor receptor IIb modulates the molecular Grb2 interaction network in activated 3 to initiate antiviral response. Proc. Natl. Acad. Sci. USA 109: 5791–5796. B cells. Cell. Signal. 23: 893–900.