Regulation of interleukin receptor-associated kinase (IRAK) phosphorylation and signaling by iota protein kinase C.

We have previously shown that the activity of the interleukin-1 (IL-1) receptor-associated kinase (IRAK) is required for nerve growth factor (NGF)-induced activation of NF-kappaB and cell survival ((2002) J. Biol. Chem. 277, 28010-28018). Herein we demonstrate that NGF induces co-association of IRAK with atypical protein kinase C iota (PKC) and that the iota PKC.IRAK complex is recruited to the p75 neurotrophin receptor. Recruitment of IRAK to the receptor was dependent upon the activity of the iota PKC. Moreover, transfection of kinase-dead iota PKC blocked both NGF- and IL-1-induced IRAK activation and the activity of NF-kappaB. Hence, iota PKC lies upstream of IRAK in the kappaB pathway. Examining the primary structure of IRAK, we identified three putative PKC phosphorylation sites; iota PKC selectively phosphorylated peptide 1 (RTAS) within the death domain domain at Thr66, which is highly conserved among all IRAK family members. Mutation of Thr66 to Ala impaired the autokinase activity of IRAK and reduced its association with iota PKC but not TRAF6, resulting in impaired NGF- as well as IL-1-induced NF-kappaB activation. These findings provide insight into the underlying mechanism whereby IRAK regulates the kappaB pathway and reveal that IRAK is a substrate of iota PKC.

Like interleukin-1 (IL-1), 1 the neurotrophin receptor p75 NTR utilizes the interleukin-1 receptor-associated kinase (IRAK), for activation of the NF-B pathway (1). In neurons, the functional outcome of this pathway is trans-activation of numerous genes that play a role in central nervous system survival. The molecular events at the p75 receptor complex leading to induction of NF-B have been characterized recently. The adapter protein MyD88 recruits IRAK to the p75 receptor complex where IRAK subsequently interacts with TRAF6 and p62 scaffold, which bridges them to IKK␤ (NF-B kinase kinase). However, the mechanism whereby IRAK is activated in the B pathway has not yet been fully elucidated. The identification of a kinase responsible for the signal-induced phosphorylation of IRAK has been the subject of intense study (2). In this regard, mutation of the its catalytic domain reveals that IRAK is still phosphorylated upon IL-1 stimulation (3), suggesting that IRAK is likely to be the substrate of another kinase lying upstream proximal to the receptor. However, in the IL-1 pathway, the catalytic activity of IRAK is not required for mediating for activation of the B pathway (4). In contrast, in both the p75 NTR and TNF pathways, IRAK catalytic activity is required for mediation of the B response (5). Hence, in some systems, upstream phosphorylation by another kinase may regulate not only the catalytic activity of IRAK but also the enzyme's ability to interact with effectors of the B pathway; thus phosphorylation of IRAK may serve as a bifurcation point in the B pathway.
NGF, IL-1, and TNF are potent activators of the atypical PKCs. Moreover, the highly homologous atypical PKC (aPKC) isoforms zeta and lambda/iota have been shown to play a critical role during B activation (6,7). NGF binding to the p75 NTR receptor results in production of second messenger metabolites such as ceramide (8), which may lead to activation of iota aPKC (9). In addition, the p75 NTR receptor employs p62, the atypical protein kinase C-interacting protein, as a selective scaffold for activation of NF-B (1). Thus, we hypothesized that iota PKC may be recruited into the p75 receptor complex to serve as an IRAK. Herein, we report the discovery of a novel pathway revealing that iota PKC phosphorylates IRAK within the death domain (DD) at the conserved residue Thr 66 . Mutation of this site impairs IRAK autophosphorylation, interaction with iota PKC, and NF-B activation. Altogether these findings reveal that IRAK is a substrate of iota PKC and that the Thr 66 phosphorylation site serves to enhance interaction between kinase and substrate. Site Mutagenesis-The cDNA of His/Myc-mPLK/IRAK or His/Myc-cimPLK, a mutant of IRAK that has no autophosphorylation activity, was site-mutated Thr 66 to Ala using the QuikChange site-directed mutagenesis kit from Stratagene by GeneMed Synthesis (San Francisco). Two complimentary oligonucleotides containing the desired mutation and flanked by unmodified nucleotide sequences were synthesized (5Ј-CGCTCCGGGCAGCGCGCGGCCAGCGTC-3Ј and 5Ј-GAC-GCTGGCCGCGCGCTGCCCGGAGCG-3Ј). Following transformation, * This work was supported by NINDS, National Institutes of Health, NS-33661. 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.
individual colonies were screened for the ␤-galactosidase (␤-galϩ, blue), phenotype and the Thr 66 to Ala mutation was verified by sequencing.
Co-immunoprecipitation and Western Blotting-To immunoprecipitate p75 (Promega) or His/Myc (Santa Cruz Biotechnology)-tagged IRAK, the cells were washed in phosphate-buffered saline, lysed with PD buffer (40 mM Tris, pH 7.6, 500 mM NaCl, 0.1% Nonidet P-40, 6 mM EDTA, 6 mM EGTA, 10 mM ␤-glycerophosphate, 10 mM NaF, 10 M p-nitrophenylphosphate, 300 M NaOV 3 , 1 mM DTT, 2 M PMSF, 10 g/ml aprotinin, and 1 g/ml leupeptin). 750 g of cell extract was incubated for 3 h with either 3 g of anti-p75 or anti-Myc antibody to capture the tagged IRAK construct followed by a 2-h incubation with 30 l of anti-rabbit IgG agarose. The samples were separated on 10% SDS-polyacrylamide gels, Western blotted with antibody to iota PKC, IRAK, anti-GST, or anti-His followed by the addition of horseradish peroxidase-conjugated secondary antibody, and developed employing ECL (Amersham Biosciences). Monoclonal antibody to IRAK was obtained from BD Biosciences. The immunogen for this antibody is the catalytic domain of IRAK.
Peptide Kinase Assay-To examine phosphorylation of IRAK peptides (synthesized by the Macromolecular Structure Facility, University of Kentucky) by aPKC, an in vitro kinase assay was employed (11). Each reaction contained 1 g of synthesized IRAK peptide control or mutant in 50 mM PIPES, pH 7.5, 15 l, 2.5 mM EGTA 17.5 l, 100 mM MgCl 2 10 l, 1 mg/ml phosphatidylserine 2 l, Ϯ purified aPKC, 100 ng (Calbiochem), and 1 Ci of [␥-32 P]ATP for 10 min at 30°C. The reaction was stopped by adding 280 mM H 3 PO 4 and spotted onto P81 paper followed by washing in 75 mM H 3 PO 4 three times. Radioactivity was counted using Cerenkov.
Endogenous Kinase Assay-To examine IRAK autophosphorylation and transphosphorylation ability by coexpressed aPKC, an endogenous kinase assay was employed. Transfected HEK cells were lysed in PD buffer, and IRAK was captured by immunoprecipitation of the cell lysates (750 g) with Myc antibody. The beads were washed three times in PD buffer followed by the addition of 30 l of kinase buffer (20 mM Tris, pH 7.6, 20 mM MgCl 2 , 20 mM ␤Ϫglycerophosphate, 20 mM p-nitrophenylphosphate, 1 mM NaOV 3 and 0.4 mM PMSF) and 5 Ci of [␥-32 P]ATP for 20 min at 30°C. 20 l of sample buffer was added to the samples to stop the reaction, and the reaction was separated by 10% SDS-PAGE and exposed to x-ray film. Aliquots of the whole cell lysates were blotted with anti-GST and anti-Myc to validate expression of iota PKC and IRAK. Monoclonal antibody to iota PKC or IRAK was obtained from BD Biosciences and polyclonal IRAK antibody from Santa Cruz Biotechnology.
Immune Complex Kinase Assay-To detect the activity of IRAK enzyme, anti-Myc immunoprecipitates (capturing expressed His/Myctagged IRAK) was incubated with 20 l of kinase buffer (20 mM Tris, pH 7.6, 1 mM DTT, 20 mM MgCl 2 , 20 mM ␤-glycerophosphate, 1 mM DTT, 20 mM p-nitrophenylphosphate, 1 mM EDTA, and 1 mM PMSF) with 5 Ci of [␥-32 P]ATP and 5 g of MBP for 10 min at 37°C (1). The reaction was stopped by adding 20 l of sample buffer and boiling for 2 min. Phosphorylated MBP was analyzed by 10% SDS-PAGE/autoradiography. Changes in IRAK activity were monitored as a function of the enzyme's ability to phosphorylate MBP as determined by phosphorimaging analysis.
Measurement of NF-B Activity-NF-B activation was measured by reporter gene assay (1). IRAK Ϫ I1A HEK or PC12 cells were seeded into 6-well plates and transfected with the same amount of B luciferase reporter gene plasmid, pGL-3, and Renilla luciferase (used as internal control for transfection efficiency). Individual constructs were transfected in duplicate, and each assay was measured in triplicate with the Promega dual luciferase assay. Values are reported as the mean Ϯ S.E. of four individual experiments.

NGF Induces Association of IRAK with Iota PKC in a p75
Receptor Complex-In previous experiments we observed that IRAK migrated as a doublet on Western blots coincident with its activation and increase in enzyme activity (1), thus suggesting that the IRAK antibody may detect the phosphorylated/activated form of the enzyme. To test this possibility, cell lysates were incubated with phosphatase 2A for various times followed by Western blotting with monoclonal antibody to IRAK (Fig. 1A).
Inclusion of phosphatase 2A abolished immunoreactivity to IRAK, thus revealing that this antibody detects the phosphorylated/activated form of the enzyme. Because we previously observed that IRAK interacts with iota PKC-interacting protein/ p62 upon stimulation with NGF (1), we next asked whether IRAK could also directly interact with iota PKC (Fig. 1B). Treatment of PC12 cells with NGF followed by immunoprecipitation of iota PKC confirmed that NGF stimulated a rapid but transient co-association of iota PKC with IRAK. Moreover, both IRAK and iota PKC were recruited to the p75 receptor in a coincident time frame (Fig. 1B). Likewise, IRAK immunoprecipitates contained both p75 and iota PKC. The interaction of both IRAK and iota PKC with p75 occurred prior to the interaction of iota PKC and IRAK, thus suggesting that the iota-IRAK interaction took place once recruited to the receptor. As control the lysates were blotted with each antibody and revealed equal amounts of protein (Fig.  1C). In addition, at a time when iota PKC was recruited to the p75 receptor (2 min), an immune complex kinase assay using FIG. 1. NGF-induced co-association of IRAK with aPKC-iota. A, IRAK monoclonal antibody recognizes the phosphorylated form of IRAK. PC12 cells were stimulated by NGF (50 ng/ml) for 2 min. Cell lysates were treated with phosphatase 2A for 1 and 3 h respectively, followed by immunoprecipitation (IP) and Western blotting (WB) with IRAK monoclonal antibody. B, PC12 cells were treated with NGF (50 ng/ml) at the indicated time points. Cell lysates (750 g) were immunoprecipitated with aPKC-iota antibody and Western blotted with IRAK polyclonal antibody, immunoprecipitated with p75 and Western blotted with either IRAK or iota PKC antibody, or immunoprecipitated with polyclonal IRAK antibody and Western blotted with p75 or IRAK antibody. C, as control, whole cell lysates (40 g) were also Western blotted with IRAK polyclonal and iota PKC or with p75 antibodies as shown.
hnRNPA1 as a defined iota PKC substrate revealed that the enzyme was active (data not shown).
Recruitment of Phosphorylated IRAK to the p75 Receptor Requires Iota PKC Activity-Within minutes after PC12 cells are exposed to NGF, IRAK is recruited to the p75 receptor where it becomes highly phosphorylated and activated (1). Because our results reveal that iota PKC associates with IRAK, we analyzed whether the activity of iota PKC modulated recruitment of IRAK to the p75 receptor by transfecting a kinasedead mutant of iota PKC followed by immunoprecipitation of p75 and Western blotting for IRAK ( Fig. 2A). Interestingly, transfection of kinase-dead iota PKC diminished recruitment of IRAK to the p75 receptor complex and also blocked immunoreactivity of IRAK toward the monoclonal IRAK antibody, suggesting that the phosphorylation/activity of IRAK may be regulated by iota PKC. To examine this possibility, IRAK was immunoprecipitated, and its activity was measured in an immune complex kinase assay with MBP as substrate (Fig. 2B). Co-transfection of constitutively active iota PKC enhanced the basal as well as NGF-stimulated activity of IRAK. By comparison, the kinase-inactive form of iota PKC blocked NGF-stimulated activity of the enzyme along with diminished recruitment of IRAK to the p75 receptor complex. However, phosphorylation of MBP did not return to basal levels with transfection of inactive iota PKC. We believe that this effect is due to the co-association of kinases such as Src with iota PKC, which are also capable of phosphorylating MBP and thus contribute to the higher background. To test the effect of iota PKC on the ability of IRAK to activate the B pathway, HEK293 cells were transfected with either catalytically active iota PKC or a kinase-dead iota PKC construct in the presence or absence of IRAK (Fig. 2C). Whereas overexpression of catalytically active iota PKC enhanced IRAK-induced B, the kinase-dead form of iota PKC abrogated IRAK-induced NF-B activation. We also tested the effects of iota PKC in two systems in which receptor interaction results in activation of NF-B, e.g. NGF and IL-1 (Fig. 2D). In both systems, transfection of kinaseactive iota PKC enhanced NF-B activity, whereas transfection of kinase inactive iota PKC abrogated ligand-induced NF-B activation. Collectively they reveal that iota PKC lies upstream of IRAK in the B signaling pathway, strongly suggesting that iota PKC may participate directly in the activation of IRAK.
Threonine 66 in the IRAK Death Domain Is Phosphorylated by Iota PKC-IRAK is a multidomain protein (2) containing an N-terminal DD (residues 1-103) followed by a domain of unknown function (UD) (residues 104 -198), a kinase domain (KD) (residues 199 -522), and a two-part C-terminal domain, also of unknown function (residues 523-618 for C1 and residues 619 -712 for C2) (Fig. 3A). Analysis of the IRAK sequence revealed the presence of three putative PKC phosphorylation sites (RxxG) in the DD, UD, and C1/C2 domains of IRAK. As a first step in determining whether iota PKC might phosphorylate IRAK, an in vitro kinase assay employing purified kinases was undertaken (Fig. 3B). Increasing concentration of aPKC directly stimulated the phosphorylation of IRAK. As a separate means to assess site-specific phosphorylation of IRAK by iota PKC, three peptides (RTAS, RPSS, RAHS) corresponding to the putative phosphorylation sites (RXXG) within IRAK were synthesized and employed in an in vitro kinase assay as substrate with purified iota PKC enzyme (Fig. 3C). Iota PKC preferentially phosphorylated peptide 1 from the DD (Fig. 3C), which possesses a unique threonine within the putative consensus phosphorylation site (RXXS). Because neither peptide 2 nor 3 was phosphorylated by iota PKC, we deduced that Thr 66 within peptide 1 may be the amino acid targeted by iota PKC. To examine whether this was the case, a similar peptide was FIG. 2. Iota PKC kinase activity is required for recruitment and activation of IRAK. PC12 cells were transfected with a construct of GST-tagged aPKC-iota (wild type or kinase-inactive mutant). 24 h post-transfection, cells were treated with NGF (50 ng/ml) for 2 min. Cell lysates (750 g) were immunoprecipitated (IP) with p75 antibody and Western blotted (WB) with IRAK monoclonal antibody. As control, whole cell lysates were also blotted for IRAK and GST (aPKC). B, lysates were immunoprecipitated with IRAK polyclonal antibody and subjected to an immune complex kinase assay with MBP as substrate. C, HEK cells were co-transfected with constructs as shown (250 ng) along with NF-B reporter pGL3 (25 ng/well). 24 h post-transfection, NF-B activity was measured by dual luciferase assay and reported as RLU/g. D, HEK or PC12 cells were transfected as shown. After 24 h transfection, the cells were treated with either IL-1 or NGF for 3 h. NF-B activity was measured by dual luciferase assay and reported as RLU/g. The means Ϯ S.E. from three independent experiments are shown (C and D). synthesized, except Thr 66 was replaced by Ala, and the ability of iota PKC to phosphorylate the wild type and mutant peptide was assessed (Fig. 3D). Mutation of Thr 66 to Ala abolished the iota PKC ability to phosphorylate the IRAK peptide, thus revealing that Thr 66 serves as a phospho-acceptor site for iota PKC. As a separate means of confirming iota PKC phosphorylation of IRAK in vivo, we employed a catalytically inactive mutant of IRAK (cim PLK), where Thr 66 was converted to Ala (Fig. 3E). If iota PKC were responsible for phosphorylation of IRAK at Thr 66 , we reasoned that we would be able to more clearly discriminate phosphorylation of IRAK over its autophosphorylation activity by employing the catalytically active mutant. Under these conditions, iota PKC was able to phosphorylate the kinase-dead mutant of IRAK, and mutation of Thr 66 to Ala abolished iota PKC induced phosphorylation of IRAK (Fig. 3E).
Threonine 66 Regulates IRAK Functional Properties and NF-B Activation-Comparative sequence analysis of IRAK family members revealed that Thr 66 is highly conserved among IRAK family members: IRAK, IRAK-2, IRAK-M, and IRAK-4 ( Fig. 4A). Site-directed mutagenesis of IRAK changing Thr 66 to Ala was undertaken to assess the consequences of iota PKC on IRAK phosphorylation and ability to couple with downstream effectors (Fig. 4B). Transfection of IRAK Ϫ HEK cells with His/ Myc-tagged IRAK or His/Myc-tagged mutant IRAK (Thr 66 to Ala), along with iota PKC-active/inactive and/or TRAF6 was undertaken. The autokinase activity exhibited by wild type IRAK was completely abolished upon mutating Thr 66 to Ala. The basal autokinase activity of wild type IRAK was significantly reduced upon inclusion of inactive iota PKC. Mutation of Thr 66 not only blocked the auto-kinase activity of IRAK but also diminished its ability to associate with iota PKC without any effect on the interaction with TRAF6. Thus, Thr 66 appears not only to serve as a phosphorylation site but also serves a novel role in anchoring enzyme and substrate (iota PKC-IRAK). In this regard, IRAK can be considered to belong to a family of proteins known as STICKs (substrates that interact with C-kinase) (13). It has been shown previously that PKC is able to interact with, phosphorylate, and modify the function of specific substrates; thus IRAK appears to be a novel STICK.
The effects of the IRAK Thr 66 mutation were also evaluated in a physiological setting. Because IRAK plays a critical role in p75 NTR activation of NF-B, we next evaluated the effect that the Thr 66 mutant had upon NGF-induced NF-B activation (Fig. 4C). PC12 cells were transfected with vector or mutant IRAK followed by treatment with NGF and the measurement of NF-B employing dual luciferase reporter assay. We previously reported that constitutively active IRAK increases NGF-induced NF-B and that the catalytic activity of IRAK is required for activation of this pathway (1). Transfection of the mutant IRAK abrogated the ability of NGF to activate the-B pathway (Fig. 4C) as well as IL-1-induced activation of the-B pathway (Fig. 4D). kinase buffer. The autoradiogram was scanned, and the relative intensity changes were expressed as net volume. The average of three independent experiments is shown Ϯ S.E. C, peptides 1, 2, and 3 containing potential phosphorylation sites (underlined) were synthesized with the indicated sequence. Peptides were incubated with [␥-32 P]ATP and recombinant aPKC enzyme (0.1 g (A) or 0.2 g (B)). Radioactivity was counted using Cerenkov. D, peptide 1 M was synthesized with Ala instead of Thr 66 (underlined). Peptides (wild type or mutant) were incubated with increasing amounts (0.025, 0.05, 0.1, 0.2 g, respectively) of recombinant iota PKC enzyme. The means Ϯ S.E. from three independent experiments are shown (C and D). E, IRAK Ϫ cells were co-transfected with His/Myc-tagged catalytically inactive IRAK (WT) or mutant Thr 66 -Ala-inactive IRAK (M) along with GST-tagged iota PKC. The phosphorylation of IRAK was examined by endogenous kinase assay. The whole cell lysates were blotted with anti-GST or Myc antibodies as control. WB, Western blot; KD, kinase dead.
FIG. 3. Thr 66 is the site in IRAK that is targeted by iota PKC. A, IRAK sequence contains the D, UD, KD, and C1 and C2 regions. There are three peptide sequences that potentially could serve as phosphoacceptor sites for iota PKC, indicated as #1, #2, and #3. B, recombinant iota PKC and IRAK enzymes were incubated with [␥-32 P]ATP and