Protein Kinase Cα Is Involved in Interferon Regulatory Factor 3 Activation and Type I Interferon-β Synthesis*

Protein kinase C (PKC) isoforms are critically involved in the regulation of innate immune responses. Herein, we investigated the role of conventional PKCα in the regulation of IFN-β gene expression mediated by the Toll-like receptor 3 (TLR3) signaling pathway. Inhibition of conventional PKC (cPKC) activity in monocyte-derived dendritic cells or TLR3-expressing cells by an isoform-specific inhibitor, Gö6976, selectively inhibited IFN-β synthesis induced by double-stranded RNA polyinosine-polycytidylic acid. Furthermore, reporter gene assays confirmed that PKCα regulates IFN-β promoter activity, since overexpression of dominant negative PKCα but not PKCβI repressed interferon regulatory factor 3 (IRF-3)-dependent but not NF-κB-mediated promoter activity upon TLR3 engagement in HEK 293 cells. Dominant negative PKCα inhibited IRF-3 transcriptional activity mediated by overexpression of TIR domain-containing adapter inducing IFN-β and Tank-binding kinase-1. Additional biochemical analysis demonstrated that Gö6976-treated dendritic cells exhibited IRF-3 phosphorylation, dimerization, nuclear translocation, and DNA binding activity analogous to their control counterparts in response to polyinosine-polycytidylic acid. In contrast, co-immunoprecipitation experiments revealed that TLR3-induced cPKC activity is essential for mediating the interaction of IRF-3 but not p65/RelA with the co-activator CREB-binding protein. Furthermore, PKCα knock-down with specific small interfering RNA inhibited IFN-β expression and down-regulated IRF-3-dependent promoter activity, establishing PKCα as a component of TLR3 signaling that regulates IFN-β gene expression by targeting IRF-3-CREB-binding protein interaction. Finally, we analyzed the involvement of cPKCs in other signaling pathways leading to IFN-β synthesis. These experiments revealed that cPKCs play a role in the synthesis of IFN-β induced via both TLR-dependent and -independent pathways.

Type I interferons (IFNs), 4 comprising IFN-␤ and the IFN-␣ family are central to the innate immunity of mammals and the development of effective adaptive immune responses against viruses and tumors (1,2). Toll-like receptors (TLRs) expressed by innate immune cells, including dendritic cells (DCs), recognize distinct pathogen-associated molecular patterns, leading to the activation of innate immunity, which shapes the subsequent adaptive immune response (3,4). TLR3 triggering by double-stranded RNA generated during viral infection with synthetic double-stranded RNA analog poly(I:C), endogenous cellular mRNA structures, and small interfering RNAs results in DC activation to promote type I IFNs (5)(6)(7). Recent studies have demonstrated the involvement of TLR3 in antiviral defense and its prominent role in the cross-priming mechanism, since DCs from TLR3-deficient mice exhibit reduced inflammatory responses mediated by reovirus genomic doublestranded RNA and poly(I:C) and are compromised in CTL cross-priming (5, 8 -10).
The enhancer region of the IFN-␤ promoter has been extensively characterized as containing four overlapping regulatory elements that are designated as positive regulatory domain (PRD)-I, -II, -III, and -IV (17,19,20). Promoter region analysis of the Ifn␤ gene revealed that PRDII and PRDIV are bound by NF-B and ATF-2/c-Jun, respectively, and that these elements * The Institute for Medical Immunology is sponsored by the government of the Walloon Region and GlaxoSmithKline Biologicals. This study was also supported by the Fonds National de la Recherche Scientifique (Belgium) and an Interuniversity Attraction Pole of the Belgian Federal Science Policy. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. □ S The on-line version of this article (available at http://www.jbc.org) contains supplemental Fig. S1. 1 Supported by the Télé vie program of the Fonds National de la Recherche Scientifíque (FNRS). 2 These two authors share senior authorship. 3 To whom correspondence should be addressed. Tel.: 32-2-650-9584; Fax: 32-2-650-9563; E-mail: fwillems@ulb.ac.be. 4 The cooperate with PRDI and PRDIII in the induction of the IFN-␤ promoter (21). The PRDI and PRDIII elements were shown to be bound by distinct members of the IRF family with much attention focused on IRF-3, which binds to the PRDIII-I composite site on the IFN-␤ promoter and is required for IFN-␤ gene transcription (22)(23)(24). PKCs mediate an evolutionarily conserved function in host defense against fungal and bacterial infections from primitive organisms up to mammals (25,26). The PKC family subdivides into three main groups based on the presence or absence of distinct motifs determining co-factor requirements for their activity. Conventional PKCs (␣, ␤ I , ␤ II , and ␥) require Ca 2ϩ binding and are activated by diacylglycerol and phorbol esters, whereas novel PKCs (␦, ⑀, and ) do not require Ca 2ϩ but are also diacylglycerol/phorbol ester-sensitive. The last subgroup, atypical PKC ( and /) require neither Ca 2ϩ nor diacylglycerol/phorbol esters (27,28). Previous studies demonstrated that PKC isoforms regulate several signaling pathways, including innate immune responses induced by microbial products (26). Indeed, Gram-negative LPS was shown to activate distinct PKC isoforms in DCs and macrophages (M⌽s) (29 -33). Specifically, PKC⑀ is an essential integrator of LPS-mediated inflammatory cytokine production through activation of NF-B and mitogen-activated protein kinases (29,34). Importantly, PKC⑀-deficient M⌽s display defects in clearing both Gram-negative and positive bacterial infections (34). Although PKC involvement in poly(I:C)-mediated innate responses was previously observed using broad range pharmacological inhibitors (35,36), participation of PKC isoforms in poly(I:C)-mediated activation of TLR3 pathway still remains unknown.
In this work, we investigated the role of cPKC isoforms in poly(I:C)-induced cytokines, particularly IFN-␤ and the molecular targets downstream of the TLR3 signaling pathway. Cellular inhibition assays, using a conventional PKC inhibitor, Gö6976, performed in monocyte-derived DCs, HEK 293, and HEK 293 cells stably expressing TLR3 demonstrated that conventional PKC activity is involved in IFN-␤ gene expression. In support of this observation, we showed that overexpression of dominant negative (DN) PKC␣ as well as Gö6976 inhibits IRF-3-dependent but not NF-B-mediated reporter gene activities, correlating with repression of IFN-␤ promoter activity. Further biochemical analysis revealed an unexpected role for cPKC activity that is essential for mediating IRF-3 association with CBP. Particularly, PKC␣ knockdown by siRNA in TLR3-293 cells established that PKC␣ participates in IRF-3-mediated IFN-␤ gene expression. Finally, our data show that cPKCs play a general role in IFN-␤ synthesis induced either by TLR-dependent or TLR-independent pathways.

EXPERIMENTAL PROCEDURES
Generation of Monocyte-derived DC-DCs were generated from peripheral blood mononuclear cells, as described by Romani et al. (37). Briefly, peripheral blood mononuclear cells were harvested from healthy volunteers by density centrifugation on Lymphoprep (Nycomed, Oslo, Norway), resuspended in culture medium, and allowed to adhere onto 75-cm 2 flasks. After 2 h at 37°C, nonadherent cells were removed, and adherent cells were cultured in 20 ml of medium containing granu-locyte-macrophage colony-stimulating factor (800 units/ml) and IL-4 (100 units/ml). Every 2 days, 800 units of granulocytemacrophage colony-stimulating factor and 100 units of IL-4 were added. After 6 days, nonadherent cells that corresponded to the DC-enriched fraction routinely contained more than 95% DC as assessed by morphologic and fluorescence-activated cell sorter analysis as described previously (38).
Transfection Assays-HEK 293 or TLR3-293 cells were seeded in 24-well plates at a density of 5 ϫ 10 5 /ml. The following day, cells were transfected with 1 g of the indicated luciferase reporter plasmids, using FuGENE TM -6 (Roche Applied Science) according to the manufacturer's specifications. All transfections included 40 ng of Renilla luciferase DNA in the pRL-TK vector (Promega, Leiden, The Netherlands) as an internal control. Where indicated, the cells were stimulated with poly(I:C) (50 g/ml) for 18 h, and subsequently the cells were harvested and the whole cell extracts (WCEs) were obtained, followed by analysis of promoter activities using the dual luciferase reporter assay system (Promega). Promoter activities were normalized to Renilla luciferase activities. Data are expressed as the mean relative stimulation Ϯ S.D.
siRNA Transfection-Duplexed siRNAs targeting human PKC␣ and GFP, each labeled with 3Ј Alexafluor 647, were purchased from Qiagen (KJ Venlo, The Netherlands). HEK 293 cells stably expressing TLR3 were seeded in 24-well plates at a density of 1.25 ϫ 10 5 . Twenty-four hours later, cells were transfected with the indicated concentrations of PKC␣ siRNA or control, GFP siRNA using X-treme gene (Roche Applied Science) according to the manufacturer's protocol. After 48 h of incubation, the supernatant was removed and replaced with fresh medium and subsequently stimulated with poly(I:C) (10 g/ml). The cells were collected after 18 h of stimulation, and the efficiency of PKC␣ knockdown was assessed by Western blotting. The cell supernatants were analyzed for IFN-␤ and IL-8 production, as described below. For DNA/RNA co-transfection studies, TLR3-293 cells (1.25 ϫ 10 5 /ml) were transfected with 1 g of IRF-3-Gal4 reporter plasmid using X-treme gene (Roche Applied Science). 24 h later, PKC␣or GFP-siRNA was added. After 48 h of incubation, the supernatant was removed and replaced with fresh medium. Where indicated, the cells were stimulated with poly(I:C) (10 g/ml) for 18 h, and WCEs were harvested. Promoter activities were analyzed as described above.
Determination of Cytokine Levels-DCs or HEK 293 cells (5 ϫ 10 5 /ml) were stimulated with poly(I:C) (10 and 50 g/ml, respectively) or LPS (1 g/ml) for 20 h followed by quantification of cytokines in culture supernatants. HEK 293 cells were seeded in 24-well plates at a density of 5 ϫ 10 5 /ml. The following day, cells were transfected with 2 g of synthetic ds-␤-DNA, using FuGENE TM -6 (Roche Applied Science) according to the manufacturer's specification, for 20 h. IFN-␤ levels were measured with ELISA kits from BIOSOURCE (Nivelles, Belgium). Quantification of IL-8 and tumor necrosis factor-␣ were performed using antibodies from BIOSOURCE.
Determination of NF-B and IRF-3 DNA Binding Activities-Nuclear extracts were prepared as described earlier (41). IRF-3 or NF-B DNA-binding activities in nuclear extracts were measured by Trans-AM IRF-3 or p65 transcription factor assay kits (Active Motif Europe, Rixensart, Belgium) according to the manufacturer's protocols. Briefly, 3 g of each nuclear extract was incubated in plates coated with consensus IFN-stimulated response elements and NF-B oligonucleotides, respectively. Plates were washed, and anti-IRF-3 or p65 antibodies were added to the wells. Antibody binding was detected with a secondary horseradish peroxidase-conjugated antibody and developed with TMB substrate. The intensity of the reactions was measured at 450 nm.
Confocal Microscopy-DCs (2 ϫ 10 5 ) were seeded on glass coverslips and allowed to adhere for 1 h. Following cellular activation, DCs were fixed in 2% paraformaldehyde and permeabilized with 100% cold methanol and then were incubated in blocking buffer (phosphate-buffered saline, 2% bovine serum albumin) for 20 min. The cells were incubated with IRF-3 monoclonal antibody (5 g/ml) (BD Biosciences) followed by p65/RelA (4 g/ml) (Santa Cruz Biotechnology) and detected with ALEXA-488-and ALEXA-568-conjugated secondary antibodies (2 g/ml) from Invitrogen. Following RNase treatment, nuclei were stained with TOTO-3 iodide (5 M) (Invitrogen). The cells were mounted in Vectashield mounting medium (Vector Laboratories Inc., Brussels, Belgium) and visualized with a Leica TCS SP2 AOBS confocal microscope (Leica, Deerfield, IL). Images were acquired using immersion oil objective ϫ100 HCV PL APO CS with a ϫ1.40 numerical aperture. The scanned images were acquired with Leica confocal software.
Statistical Analysis-Data were compared using paired Wilcoxon's nonparametric test or Student's t test (where indicated).
Supplemental Material- Fig. S1 shows that inhibition of conventional PKC activity suppresses IFN-␤ production mediated by the engagement of TLR3.

Selective Inhibition of Poly(I:C)-mediated Type I IFN-␤ Synthesis by a Conventional PKC
Inhibitor-Engagement of distinct TLRs by their specific ligands induces activation of several PKC isoforms in various cell types (26). Therefore, we first assessed whether poly(I:C) induced the activation of PKC isoforms in monocyte-derived DCs. As shown in Fig. 1A, DCs exhibited phosphorylated forms of conventional PKC␣/␤ and novel PKC␦ isoforms within 30 min following the poly(I:C) encounter. Gö6976, a potent and selective conventional PKC inhibitor targeting Ca 2ϩ -dependent PKC isoforms (42), completely abolished the phosphorylation-dependent activation of conventional PKC isoforms without effecting PKC␦ activation.
The following experiments were designed to evaluate the functional outcomes of blocking poly(I:C)-mediated conventional PKC activation on cytokine production in DCs. As shown in Fig. 1B, Gö6976 dose-dependently inhibited poly(I:C)-induced IFN-␤ secretion from DCs. Next, we analyzed whether the inhibitory effect of Gö6976, observed on IFN-␤ protein secretion, was due to an effect on IFN-␤ mRNA transcription. Using quantitative reverse transcription-PCR analysis, we observed that the accumulation of IFN-␤ mRNA transcripts occurred as early as 2 h, peaking maximally at 4 h following poly(I:C) stimulation (data not shown). At 4 h, Gö6976-treated poly(I:C)-activated DCs exhibited significantly reduced IFN-␤ mRNA levels in comparison with poly(I:C)-activated control counterparts (Fig. 1C). In contrast, poly(I:C) exposure of DCs resulted in IL-8 and tumor necrosis factor-␣ production that was comparable between the Gö6976-treated cells and their control counterparts (Fig. 1, D and E).
PKC␣ Is Involved in TLR3-mediated Induction of IRF-3-dependent Promoter Activities-In the next set of experiments, we investigated the effects of inhibiting poly(I:C)-induced conventional PKC activity on human IFN-␤ promoter activity in TLR3-expressing 293 cells. As shown in Fig. 2, poly(I:C) stimulation of TLR3-293 cells resulted in a strong induction of IFN-␤ promoter activity, which was repressed 3-fold in Gö6976treated counterparts.
Our data pointed out that cPKCs were involved in IFN-␤ gene expression through regulation of IFN-␤ promoter activity. Furthermore, in two distinct cell types, we found that cPKC signaling is selectively involved in the regulation of IFN-␤ but not inflammatory cytokines upon poly(I:C) encounter. Since IFN-␤ gene transcription requires the cooperation of NF-B (12,43,44) and IRF-3 transcription factors (22)(23)(24), we investigated the involvement of conventional PKC isoforms in the regulation of IRF-3-and/or NF-B-mediated transcriptional activities using either IRF-3-or NF-B-dependent promoters in TLR3-293 cells, which also express PKC␣ and PKC␤ I conventional isoforms endogenously (Fig. 3A). We transfected TLR3-293 cells with either Gal4-IRF-3, the PRDIII-I site from the IFN-␤ promoter, or a multimerized NF-B site from the IL-8 promoter in the presence of increasing concentrations of Gö6976 and subsequently stimulated it with poly(I:C). In Fig.   3B (top and middle), we demonstrated that the Gö6976 treatment resulted in a dose-dependent repression of poly(I:C)-mediated Gal-4-IRF-3 reporter activity as well as transcription mediated at the PRDIII-I site in TLR3-293 cells. In order to assess which of the conventional PKC isoforms are involved in TLR3 signaling, we overexpressed DN PKC␣ and PKC␤ I isoforms in TLR3-293 cells. As shown in Fig. 3C (top), IRF-3 reporter activity was dose-dependently inhibited (ϳ6-fold at the highest dose) by DN PKC␣ but not by DN PKC␤ I overexpression. In a similar setting, poly(I:C)induced reporter gene transcription at the PRDIII-I site from the IFN-␤ promoter was similarly inhibited (Fig. 3C, middle). Strikingly, poly-(I:C)-induced NF-B reporter gene activity from the IL-8 promoter was altered by neither DN PKC␣ nor PKC␤ I expression vector nor Gö6976 treatment (Fig. 3, B and C,  bottom). In the reporter assay settings we tested, DN PKC␤ I had no effect while being optimally expressed in HEK 293 cells (Fig. 3A,  right). Hence, we concluded that PKC␣ selectively regulates transcriptional activity of IRF-3-dependent promoters induced by poly(I:C) triggering of TLR3.
PKC␣ Is Involved in TRIF-mediated IRF-3 Activation Downstream of TBK1-Transcriptional activity at the PRDIII-I site from the IFN-␤ promoter requires TRIF interaction with TBK1, which channels phosphorylation-dependent activation of IRF-3 (12,45,46). Next we investigated whether TRIF-mediated and TBK1-mediated signaling to IRF-3 are targeted by conventional PKC isoforms. To assess this question, we analyzed the effects of DN PKC␣ and DN PKC␤ I isoforms on IRF-3-mediated reporter gene activity induced by TRIF or TBK1 overexpression in HEK 293 parental cells. As shown in Fig. 4A (top), TRIF overexpression resulted in the increase of IRF-3-dependent reporter gene activity. Co-expression of DN PKC␣ but not DN PKC␤ I dose-dependently inhibited (ϳ3-fold at the highest dose) TRIF-mediated IRF-3 transcriptional activity. In Fig. 4A (bottom), TRIF-mediated induction of reporter gene activity at the PRDIII-I site in 293 cells was inhibited (ϳ6-fold at the highest dose) by DN PKC␣, whereas DN PKC␤ I had no effect. Furthermore, DN PKC␣ (but not DN PKC␤ I ) overexpression dosedependently repressed IRF-3-mediated reporter gene activity as well as transcription at the PRDIII-I site mediated by TBK1 (Fig. 4B, top and bottom). Taken together, these results indicated that PKC␣ participates in the optimal induction of IRF-3 transcriptional activity downstream of TRIF and its immediate effector kinase TBK1.
Conventional PKCs Are Not Involved in Poly(I:C)-mediated IRF-3 Phosphorylation, Dimerization, Nuclear Translocation, and DNA Binding-Our data showing diminished IFN-␤ gene expression and IRF-3-dependent transcriptional activity led us to investigate the functional outcomes of inhibiting conventional PKCs on early signaling events important for IRF-3 activation. Poly(I:C)-mediated site-specific IRF-3 Ser 396 phospho-rylation is critical for IRF-3 nuclear translocation and DNA binding on IFN-stimulated response element sites on IRF-3responsive promoters (39,47). Therefore, we first examined the phosphorylation status of IRF-3 at Ser 396 in the presence or absence of Gö6976 in poly(I:C)-activated DCs. Interestingly, Western blotting experiments demonstrated that poly(I:C)mediated phosphorylation of IRF-3 at Ser 396 residue was comparable whether or not DCs were exposed to Gö6976, shown by a slower migrating band (Fig. 5A). Following phosphorylation, IRF-3 forms a homodimer, an important process required for its transcriptional activity (39,48,49). We therefore analyzed the effects of inhibiting conventional PKCs on IRF-3 homodimer formation in DC extracts by native PAGE analysis. Our data demonstrated that conventional PKC inhibitor Gö6976 does not influence the formation of IRF-3 homodimers that occur following poly(I:C) encounter by DCs (Fig. 5B).
Next, using confocal microscopy, we showed that inhibition of conventional PKC activity did not affect poly(I:C)-mediated nuclear localization of IRF-3 in DCs, since poly(I:C)-induced IRF-3 nuclear translocation was comparable between Gö6976treated DCs and the vehicle-treated control counterparts (Fig.  5C, top). As displayed in Fig. 5C (middle), Gö6976-treated DCs exhibited similar levels of poly(I:C)-induced p65/RelA translocation in comparison with their control counterparts. Finally, we assessed poly(I:C)-induced IRF-3 DNA binding activity in DCs exposed to Gö6976. IRF-3 DNA binding was increased in nuclear extracts from DCs, detected at 90 min following poly(I:C) stimulation (Fig. 5D, left). The level of IRF-3 DNA binding in Gö6976-treated DCs was not significantly altered. Likewise, poly(I:C)-induced NF-B DNA binding was also not affected by Gö6976 treatment, indicating that neither NF-B nuclear translocation nor its DNA binding is influenced by Gö6976 (Fig. 5D, right). Hence, our data indicated that the modifications rendered by inhibiting conventional PKC activity in TLR3 signaling do not involve previously documented events leading to IRF-3 activation, specifically dimerization, nuclear translocation, and DNA binding ability.
Inhibition of Conventional PKC Activity Hinders IRF-3 Binding to Co-activator CBP-Thus far, we have shown that PKC␣ participates in TLR3-mediated transcriptional activity of IRF-3-dependent promoters. Given that conventional PKC inhibitor, Gö6976, diminishes IFN-␤ expression without influencing the initial events important for IRF-3 phosphorylation, homodimerization, nuclear translocation, and DNA binding, we investigated whether conventional PKCs are important for IRF-3 association with the co-activator CBP. In co-immunoprecipitation experiments, we observed a robust physical interaction between the endogenous forms of IRF-3 and CBP in poly(I:C)-stimulated DCs that was severely decreased in Gö6976-treated counterparts (Fig. 6A, left). Similar results, showing strongly reduced IRF-3 interaction with CBP, were obtained from TLR3-293 cells exposed to Gö6976 (Fig. 6B, left). Previously, the co-activator CBP was shown to interact with other transcription factors, including the p65/RelA member of NF-B (50). To find out whether the effects of Gö6976 resulting in the hindrance of IRF-3 interaction with CBP prevent CBP from binding to NF-B members, we performed co-immunoprecipitation assays to visualize p65 interaction with CBP. Strikingly, poly(I:C)-mediated interaction between p65 and CBP remained intact whether or not DCs or TLR3-293 cells were exposed to Gö6976 (Fig. 6, A and B, right). Overall, these observations indicated that inhibition of conventional PKC activity abrogates IRF-3 but not NF-B interaction with the co-activator CBP.
Inhibition of Conventional PKCs Does Not Interfere with ATF-2 and c-Jun Activation-We then evaluated the role of cPKCs in TLR3-mediated activation of ATF-2/c-Jun, which is the third transcription factor required for IFN-␤ induction. As shown in Fig. 7A, TLR3-293 cells exhibited a phosphorylated form of ATF-2 within 30 min following poly(I:C) exposure, which persisted up to 60 min after stimulation and was not affected by Gö6976 treatment. Similarly, the poly(I:C)-induced phosphorylation of c-Jun occurred in an identical kinetic and was maintained even in the presence of Gö6976 (Fig. 7B). Next, using co-immunoprecipitation experiments, we examined the role of cPKCs on ATF-2-or c-Jun-CBP interactions in the same system. TLR3-293 cells exhibited basal ATF-2 or c-Jun-CBP interaction but we were not able to detect an increase of such interaction after poly(I:C) stimulation. Nevertheless, the basal interaction between ATF-2 or c-Jun and CBP was not disrupted by Gö6976 (Fig. 7C). We then performed additional experiments in human DCs; in this setting, the increase of ATF-2 or c-Jun and CBP interaction was still barely detectable even upon poly(I:C) stimulation. However, in a single experiment, where the induction was clearly apparent, the addition of the drug did not affect the ATF-2or c-Jun-CBP interaction (data not shown).
PKC␣ Is Essential for IFN-␤ Expression in Response to Poly(I:C)-To further confirm the role of PKC␣ in the regulation of IFN-␤ gene expression, we transiently transfected TLR3-293 cells with siRNA targeting PKC␣. TLR3-293 cells were transfected with PKC␣-siRNA ranging from 0 to 200 M. As shown in Fig. 8A (left), PKC␣ knockdown by its target-specific siRNA occurred in a dose-dependent manner, whereas the control GFP-siRNA had no inhibitory effects on PKC␣ protein expression (Fig. 8A,  right). Next we investigated the effects of PKC␣ knockdown on IRF-3-mediated transcriptional activity. We co-transfected TLR3-293 cells with a Gal4-IRF-3 promoter and the indicated concentrations of PKC␣or GFP-siRNA followed by poly(I:C) stimulation. The poly(I:C)-induced transcriptional activity of Gal4-IRF-3 was inhibited with increasing concentrations of PKC␣-siRNA, whereas the control GFP-siRNA had no effect on its activity (Fig. 8B). Then we investigated the functional FIGURE 5. Inhibition of conventional PKCs does not prevent IRF-3 activation. Immature DCs were treated with vehicle (Me 2 SO) or Gö 6976 (1 M) for 2 h and then were activated by poly(I:C) (10 g/ml) or were left unstimulated. A, Gö 6976 effects on IRF-3 phosphorylation. At the indicated time intervals, cells were harvested and lysed in Laemmli buffer, followed by direct Western blotting using anti-phospho-IRF-3 Ser 396 antibody. One representative experiment of five is shown. B, IRF-3 homodimerization in poly(I:C)-activated DCs. 90 min after poly(I:C) activation, cells were harvested, and WCEs were prepared and subjected to native PAGE followed by immunoblotting as described under "Experimental Procedures." One representative of four independent experiments is shown. C, nuclear translocation of IRF-3 and NF-B in DCs. 90 min after poly(I:C) activation, cells were collected, washed, and fixed on coverslips, followed by staining using antibodies specific for IRF-3 (top) or p65/RelA (middle). Nuclei were stained with TOTOா-3 iodide (bottom). Nuclear translocation was visualized using confocal microscopy. The images shown are representative of nonoverlapping fields. One representative of three independent experiments is shown. D, DCs were harvested 90 and 120 min after cellular activation, and nuclear extracts were prepared and analyzed for IRF-3 (left) or p65/RelA (right) DNA binding activities, respectively, using the TransAM transcription factor assay kits. Data are means Ϯ S.E. of 5-8 independent experiments. ns, statistically not significant.
outcomes of PKC␣ knockdown on IFN-␤ expression. TLR3-293 cells were transfected with either PKC␣ or GFP-siRNA and then stimulated with poly(I:C), followed by analysis of cell supernatants for cytokine production by ELISA. IFN-␤ secretion from poly(I:C)-activated TLR3-293 cells was strongly inhibited by the knockdown of PKC␣ whereas the control siRNA had no effect (Fig. 8C, left). However, poly(I:C)-induced IL-8 production was unaffected by either siRNA (Fig.  8C, right), mirroring data obtained with Gö6976 treatment (supplemental Fig. S1, A and B). These data demonstrated that PKC␣ knockdown selectively down-regulates poly(I:C)-mediated IFN-␤ gene expression but does not affect the production of IL-8.

Conventional PKCs Participate in the Control of LPS and ds-␤-DNA-induced IFN-␤ Synthesis-Our data
showing the critical role of conventional PKCs in TLR3-mediated IFN-␤ expression led us to investigate the possible involvement of cPKCs on other signaling pathways leading to IFN-␤ synthesis. Therefore, we first assessed the effect of cPKC inhibition on TLR4-mediated IFN-␤ synthesis in DCs. As shown in Fig. 9A, LPS-induced IFN-␤ secretion is significantly inhibited in Gö6976-treated DCs. We next evaluated the effect of inhibiting cPKCs on IFN-␤ synthesis induced by a synthetic ds-␤-DNA, previously described for its ability to trigger a TLR-independent cytosolic pathogen recognition receptor (51). For that purpose, we used HEK 293 cells, which do not express any TLRs. HEK 293 cells were transfected with ds-␤-DNA in the presence of increasing concentrations of Gö6976. As shown Fig. 9B, ds-␤-DNA-induced IFN-␤ was strongly inhibited in a dose-dependent manner, 5-fold at the highest dose. These results revealed that cPKCs are involved in the synthesis of IFN-␤ induced via both TLR4-and TLR-independent pathways.

DISCUSSION
PKC isoforms are essential signaling components that contribute to innate immune defense against microbial and viral infections and regulate adaptive immune responses (25,26). LPS triggering of TLR4 activates PKC⑀, which is involved in inflammatory cytokine expression by modulating NF-B activation in human DCs and murine M⌽s downstream of TRAM. PKC⑀ was shown to regulate TLR4 signaling by phosphorylating TRAM, a key process required for its function (29,33,34). Furthermore, poly(I:C)-activated PKCs were shown to be involved in IFN-␤ expression, since PKC inhibitors were shown to effectively block type I IFN production in human, simian, and mouse fibroblasts (35,36).
In this study, we have shown that conventional PKCs, particularly the PKC␣ isoform, participates in poly(I:C)-mediated IFN-␤ gene expression through modulation of IRF-3-depend-  Collectively, our data obtained from reporter assays with DN PKC␣ overexpression as well as PKC␣-siRNA knockdown establish that TLR3-induced conventional PKC␣ isoform is involved in the regulation/activation of early signaling events that induce activation of the IFN-␤ promoter. Since the cooperative action of IRF-3 and NF-B transcription factors for poly(I:C)-mediated induction of IFN-␤ promoter activity has been well documented (21, 22, 24, 52), we investigated the involvement of conventional PKC␣ and␤ I isoformsinIRF-3-andNF-Bdependent reporter gene transcription. Our data confirm that DN PKC␣ (but not DN PKC␤ I) ) effectively repressed IRF-3-dependent but not NF-B-mediated reporter gene activity. Therefore, PKC␣ mediates IRF-3 transcriptional activity but not NF-B-dependent transcription downstream of the TLR3 signaling pathway. In addition, the results from reporter gene assays performed using HEK 293 parental cells demonstrate that DN PKC␣ down-regulates IRF-3 transcriptional activity induced by TRIF or TBK1 overexpression. These findings suggest that PKC␣ operates downstream of TLR3-TRIF-TBK1 and participates in IRF-3 transcriptional activity.
The signaling events controlling poly(I:C)-mediated IRF-3 transcriptional activity are complex and multifaceted. Shortly after poly(I:C) or virus exposure, IRF-3 undergoes a conformational change following phosphorylation of specific serine residues in the C-terminal serinerich region of IRF-3 that allows its homodimerization (53). Following homodimer formation, IRF-3 translocates to the nucleus, where it binds to the PRDIII-I site on IFN-␤ promoter as well as IFN-stimulated response element sites found on IFN-responsive genes (24,52). The minimal phosphoacceptor site at Ser 396 on IRF-3 was shown to be targeted by poly(I:C) signaling, important for IRF-3 translocation and DNA binding ability upon poly(I:C) or viral activation (39). In addition, Fujita and co-workers (16) have demonstrated that point mutations of Ala of Ser 385 or Ser 386 (referred to as the 2S site), which is part of the serine-rich region, abolishes the phosphorylation and dimerization of IRF-3 by TBK1. Through biochemical analysis, we have assessed the contribution of conventional PKCs in the early signaling events critical for IRF-3 activation, such as (a) IRF-3 phosphorylation, (b) homodimerization, (c) nuclear translocation, (d) DNA-binding ability, and (e) IRF-3 association with CBP/p300. Our data from Western blotting experiments show that poly(I:C)-activated DCs in the presence of Gö6976 exhibited comparable levels of IRF-3 Ser 396 and also Ser 385/386 phosphorylation (data not shown). In accordance with the lack of effect of conventional PKC inhibition on these critical phospho-Ser residues, native PAGE experiments and confocal microscopy analysis demonstrated that conventional PKCs are not involved in IRF-3 phosphorylation, homodimerization, and nuclear translocation. Our findings concur with previous findings demonstrating that Gö6976 does not affect IRF-3 dimerization or NF-B DNA-binding in poly(I:C)-stimulated HeLa cells (48) or inhibit IRF-3 phosphorylation in response to virus (39,47).
An important event required for IRF-3 transcriptional activity involves its physical association with the co-activator proteins CBP/p300 (16,18,53). Crystallographic analysis of IRF-3, NF-B, and ATF-2/c-Jun revealed that CBP/p300 is recruited to the IFN-␤ enhanceosome through distinct interactions with these transcription factors that mediate the maximal induction of transcription at the IFN-␤ promoter (54,55). It is well established that IRF-3 interaction with CBP occurs between the C terminus of IRF-3 and the IFN-binding domain of CBP (23,54). Our data from co-immunoprecipitation assays revealed that TLR3-induced conventional PKCs mediate IRF-3 interaction with CBP. Importantly, we showed that inhibition of conventional PKC activity severely hinders the physical interaction between IRF-3 and CBP, which occurs following poly(I:C) exposure. Furthermore, in reporter gene assays, we did not observe any effects of inhibiting conventional PKCs on Gal4-CBP DNA binding activity, indicating that the observed hindrance by Gö6976 of IRF-3-CBP interaction possibly involves an effect on IRF-3 (data not shown). Whether PKC␣ may phosphorylate other unidentified sites involved in IRF-3/CBP interaction or acts indirectly via an unidentified kinase(s) remains to be determined. Overall, our data show evidence that conventional PKCs, particularly PKC␣, is involved in the interaction of IRF-3 with CBP, which is essential for TLR3-mediated IFN-␤ promoter induction and expression.
Similar to TLR3, TLR4 also mediates type I IFN, using MyD88-independent and TRIF-dependent pathways to induce IRF-3 activation (13, 56). In our system, LPS-induced IFN-␤ production by DCs was likewise sensitive to cPKC inhibition. In conjunction with our data on the TLR3 pathway, this demonstrates a broader role of cPKCs in TLR-dependent signaling. Another important pathway leading to type I IFN synthesis involves the activation of the cytosolic pathway by ds-␤-DNA. Viruses as well as intracellular bacteria that have cytosolic phases in their life cycles are recognized by TLR-independent cytosolic receptors, such as the helicase retinoic acid-inducible gene 1 and melanoma differentiation-associated gene 5, that recognize viral RNAs and synthetic double-stranded RNA (57,58). These two receptors use the adaptor molecule interferon-␤ promoter stimulator 1 to induce IFN-␤ and cytokines. Subsequently, we investigated the possible role of cPKCs in the induction of IFN-␤ through this TLR-independent pathway. These results revealed that cPKCs are involved in IFN-␤ synthesis induced by both TLR-dependent and TLR-independent pathways.
Although the role of TLR3 in direct priming of antiviral response is controversial, viral evasion of TLR-dependent and -independent pathways argues for the contribution of these signaling pathways in optimal host defense against viruses (9,59,60). It is proposed that TLR signaling and the resulting cytokine network plays an active role in the pathogenesis of autoimmunity (2,61,62). Type I IFNs, induced by viral infections and/or by ligand-dependent TLR activation, are principal factors that contribute to the breakdown of peripheral tolerance and autoimmune disease generation or progression (2,62). In this perspective, our study and others (29,34) demonstrating that distinct PKC isoforms take diverging roles to control TLR signaling pathways highlight the significance of isoform-selective PKC inhibition not only as potential therapy to control inflammation-induced pathologies but also as a target for viral evasion of host immunity. Hence, control of PKC activity may represent a potential strategy to treat autoimmune pathologies originating from TLR-dependent or -independent type I IFN action. Finally, since PKC inhibitors are part of current regimens used for anti-tumor therapy, we suggest that the inhibitory effects of conventional PKC inhibitors on IRF-3 transcriptional activity and IFN-␤ expression should be more cautiously addressed in vivo.