The Atypical Protein Kinase C-interacting Protein p62 Is a Scaffold for NF-κB Activation by Nerve Growth Factor*

Nerve growth factor (NGF) binding to both p75 and TrkA neurotrophin receptors activates the transcription factor nuclear factor κB (NF-κB). Here we show that the atypical protein kinase C-interacting protein, p62, which binds TRAF6, selectively interacts with TrkA but not p75. In contrast, TRAF6 interacts with p75 but not TrkA. We demonstrate the formation of a TRAF6-p62 complex that serves as a bridge linking both p75 and TrkA signaling. Of functional relevance, transfection of antisense p62-enhanced p75-mediated cell death and diminished NGF-induced differentiation occur through a mechanism involving inhibition of IKK activity. These findings reveal a new function for p62 as a common platform for communication of both p75-TRAF6 and TrkA signals. Moreover, we demonstrated that p62 serves as a scaffold for activation of the NF-κB pathway, which mediates NGF survival and differentiation responses.

Nerve growth factor (NGF) binding to both p75 and TrkA neurotrophin receptors activates the transcription factor nuclear factor B (NF-B). Here we show that the atypical protein kinase C-interacting protein, p62, which binds TRAF6, selectively interacts with TrkA but not p75. In contrast, TRAF6 interacts with p75 but not TrkA. We demonstrate the formation of a TRAF6-p62 complex that serves as a bridge linking both p75 and TrkA signaling. Of functional relevance, transfection of antisense p62-enhanced p75-mediated cell death and diminished NGF-induced differentiation occur through a mechanism involving inhibition of IKK activity. These findings reveal a new function for p62 as a common platform for communication of both p75-TRAF6 and TrkA signals. Moreover, we demonstrated that p62 serves as a scaffold for activation of the NF-B pathway, which mediates NGF survival and differentiation responses.
The biological responses to neurotrophins such as NGF 1 include neuronal survival and differentiation (1). Two receptors, TrkA and p75, participate in the formation of the high affinity NGF binding site (2). TrkA enhances both NGF responsiveness and cell survival (3). The transcription factor nuclear factor B (NF-B) is activated by both TrkA and p75 receptor components (4). Moreover, p75 has been shown to interact with TRAF6 (5), a critical adapter in the activation of NF-B by interleukin-1 and other cytokines (6). In addition, inhibition of NF-B increases p75-mediated apoptosis in this system (7), demonstrating a prosurvival requirement for this transcription factor (8,9). Furthermore, mice deficient in IKK, the enzyme that phosphorylates and targets the inhibitory molecule IB leading to NFϪB activation, leads to a defect in neureulation (10). Similarly, TRAF6-deficient mice also display a failure of neural tube closure and exencephaly (11). Collectively, these findings underscore the importance of NFϪB in the nervous system. The activation of IKK and NF-B has been shown to require atypical protein kinase C (aPKC) in both neuronal and non-neuronal systems (reviewed in Ref. 12). Moreover, aPKC over-expression enhances NGF prosurvival signaling through up-regulation of NF-B (13). In contrast, proapoptotic signaling inhibits aPKC and blocks NF-B (14). Additionally, the selective aPKC-binding protein, p62 (15,16), has been shown to interact with TRAF6 and to be essential during interleukin-1 signaling to NF-B (17). Here we report that p62 plays a novel role as a scaffold for the activation of NF-B by nerve growth factor, linking both p75 and TrkA receptor components.

EXPERIMENTAL PROCEDURES
Cell Culture and Reagents-Cultures of human embryonic kidney 293 (HEK 293) or NIH-3T3 cells were maintained in high glucose Dulbecco's modified Eagle's medium containing 10% fetal calf serum. Subconfluent cells were transfected by the calcium phosphate method. PC12 cells were grown on cultureware coated with rat tail collagen in RPMI containing 10% horse serum and 5% fetal calf serum and antibiotics (50 units/ml penicillin and 50 mg/ml streptomycin). PC12 or NIH-3T3 cells were routinely transfected using LipofectAMINE 2000 (Life Technologies, Inc.). 2.5 S NGF was purchased from Bioproducts for Science (Indianapolis, IN). The monoclonal 12CA5 anti-HA and anti-Flag antibodies were from Sigma. The rabbit anti-Myc, anti-TRAF6, anti-TrkA, and anti-IKK antibodies were from Santa Cruz Biotechnology. The monoclonal anti-p62 was obtained from BD Transduction Laboratories.
Immunoprecipitation and Western Blot Analysis-To detect endogenous proteins in PC12 cells or those cotransfected into HEK cells, lysates were prepared from subconfluent cultures of cells grown on 100-mm dishes. Typically cells were transfected for 36 -42 h with 5-10 g of construct and pcDNA3 plasmid to give 30 g of total DNA. After transfection, cells were stimulated or not with 50 ng/ml NGF. Cells were then harvested and lysed in PD buffer (40 mM Tris-HCl pH 8.0, 500 mM NaCl, 0.1% Nonidet P-40, 6 mM EDTA, 6 mM EGTA, 10 mM ␤-glycerophosphate, 10 mM NaF, 10 mM phenyl phosphate, 300 M Na 3 V0 4 , 1 mM benzamidine, 2 mM phenylmethysulfonyl fluoride, 10 g/ml aprotinin, 1 g/ml leupeptin, 1 g/ml pepstatin, 1 mM dithiothreitol). Extracts were centrifuged at 15,000 ϫ g for 15 min. Protein was determined, and equal amounts of whole-cell lysate were diluted in PD buffer and incubated with antibody as indicated for 2 h. Then protein A or G beads were added for an additional hour at 4°C. The immunoprecipitates were then washed five times with PD buffer. As control an aliquot of the cell lysate (1/10) volume was also analyzed by immunoblotting. Samples were fractionated on 7.5% SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and subjected to Western blot analysis with the corresponding antibodies. Proteins were detected with ECL reagents (Amersham Pharmacia Biotech).
Measurement of NF-B Activity-NF-B activation was measured using a reporter gene assay. HEK, 3T3, or PC12 cells were transfected with a B-luciferase reporter gene plasmid, 3EconA-Luc (17). After 24 or 48 h, the cells were stimulated with NGF and activity determined using a Promega luciferase assay kit. Individual constructs were transfected in duplicate and each assay measured in triplicate. Values are reported as the mean Ϯ S.E. of three individual experiments.

RESULTS AND DISCUSSION
Because aPKC is critically involved in the NGF prosurvival signaling pathway (13,18), we decided to investigate whether neurotrophin binding would stimulate the formation of a complex between NGF receptor components, TrkA and p75, and p62. PC12 cell lysates were prepared from NGF-treated cells followed by pull-down assays (19) employing GST-TrkA or GST-p75 (Fig. 1A). p62 associated with TrkA but not with p75. To confirm this finding in a more in vivo setting, p62 was coexpressed with either TrkA or p75 in HEK cells (17), and the coprecipitation of p62 with either receptor was examined. The results, shown in Fig. 1B, confirmed that p62 selectively associates with TrkA but not with p75. As in this experiment both TrkA and p62 were over-expressed, their interaction took place even in the absence of any stimulus. To demonstrate that endogenous p62 and TrkA interact in vivo and that this interaction may be induced in response to NGF, PC12 cells were stimulated with NGF for different times after which the coprecipitation of endogenous p62 with TrkA was examined. The results shown in Fig. 1C demonstrate that there is a significant portion of p62 bound to TrkA under basal conditions but that this interaction was reproducibly enhanced in the NGF-treated cells (Fig. 1C), indicating that this is an NGF-regulated process. To establish which region of p62 is required for interaction with TrkA, deletion mutants of p62 were transfected into HEK cells along with HA-tagged TrkA, and their association was determined by immunoprecipitation as above (17). The region encompassing amino acids 266 -446 of p62 binds TrkA, whereas the binding of p62 to TRAF6 has been mapped to amino acids 225-251 (Ref. 17 and Table I). This indicates that p62 may accommodate both TrkA and TRAF6 simultaneously. Deletion of the TRAF6 binding site did not effect TrkA binding to p62, thus further strengthening the notion that p62 interacts with TrkA and TRAF6 through two independent binding domains (Table I).
TRAF6 has been reported to interact with p75 (5). To determine whether TRAF6 interacts with TrkA as well, both p75 and TrkA were coexpressed in HEK cells along with Flag-TRAF6. Whereas p75 associated with TRAF6 in an NGF-dependent manner ( Fig. 2A) as previously reported (5), TRAF6 failed to associate directly with TrkA. The interaction of p75 with TRAF6 was mapped to the C-proximal TRAF-C domain, a region that also accommodates p62 (17). Coexpression experiments in HEK cells revealed that TRAF6 can bind both p75 and FIG. 1. Interaction of p62 with TrkA. A, equivalent amounts of NGF-stimulated (50 ng/ml, 15 min) PC12 cell lysates were bound to GST-TrkA, GST-p75, or GST alone (18). The associated p62 protein was determined by Western blotting (WB) (15). B, Myc-p75 or HA-TrkA was coexpressed with either HA-p62 or Myc-p62 in HEK cells followed by treatment with NGF. Coassociation was determined by immunoprecipitation (IP)/Western blotting (21). A fraction of the lysate was blotted with anti-HA or anti-Myc antibody to check for expression of p75, TrkA, or p62 as indicated. C, lysates prepared from NGF-stimulated (50 ng/ml, 0 -60 min) PC12 cells were immunoprecipitated with either TrkA or p62 antibody followed by Western blotting with either p62 or TrkA antibody as indicated. A fraction of the cell lysate was analyzed by blotting with p62 or TrkA antibody. Similar results were obtained in four separate experiments.
FIG. 2. p62 links TrkA to TRAF6-p75. A, subconfluent cultures of HEK cells were cotransfected with expression plasmid for Flag-tagged TRAF6 and either Myc-p75 or HA-TrkA followed by stimulation with NGF (50 ng/ml) for 15 min. The interaction was determined by immunoprecipitation (IP) and Western blotting (WB). A fraction of the lysate was blotted with anti-Myc, HA, or FLAG antibody to check for expression of p75, TrkA, or TRAF6. B, TRAF6 enhances the interaction of p62 with p75. HEK cells were cotransfected with expression plasmid for Flag-tagged TRAF6 or HA-p75. Endogenous p62 was immunoprecipitated from 3 mg of cell lysate followed by Western blotting with either HA or Flag antibody. A fraction of the lysate was blotted with HA, Flag, or p62 antibody to check for expression of p75 or TRAF6. HEK cells were cotransfected with expression plasmid for Flag-tagged TRAF6, Myc-62, or HA-TrkA. Cell lysates (750 g) were immunoprecipitated with antibody to HA followed by Western blotting with either p62 or Flag antibody. A fraction of the lysate was blotted with HA, Flag, or Myc antibody to check for expression of TrkA, TRAF6, or p62. Similar results were obtained in three separate experiments. C, the interaction of endogenous TRAF6 with p62 was determined by stimulating PC12 cells with NGF (50 ng/ml) for up to 1 h followed by immunoprecipitation of equivalent cell lysates with either aPKC or p62 antibody as indicated. A fraction of the lysate was blotted with aPKC or p62 antibody. Similar results were obtained in five separate experiments.
p62 Is a Scaffold for NF-B Activation by NGF p62 simultaneously (not shown). Because p62 interacts with TRAF6 (17), it is conceivable that p62 may be brought into a p75 complex via TRAF6 serving as a bridge. If this model is correct we should be able to coimmunoprecipitate p62 with p75 only in the presence of TRAF6. The results shown in Fig. 2B strongly suggest that these predictions are correct. Thus, in HEK cells transfected with different expression vectors, a small amount of p75 was found to associate with p62, likely through endogenous TRAF6. Upon coexpression of TRAF6, recruitment of p75 into the p62 complex was significantly enhanced (ϳ2.5-fold). The ability of TrkA to coassociate with TRAF6 was dramatically and consistently enhanced by the presence of exogenous p62 (Fig. 2B). We next determined whether NGF could stimulate the formation of an endogenous TRAF6-p62 complex in PC12 cells (Fig. 2C). In the absence of NGF little or no association of TRAF6 with p62 or aPKC could be detected. However, the addition of NGF resulted in a rapid interaction of TRAF6 with p62 and consequently with aPKC. Close examination revealed that the kinetics of association between TRAF6 and p62 (maximum 1-5 min) occurs prior to the recruitment of p62 to the TrkA receptor (Fig. 1C, peaks at   FIG. 3. Role of p62 in NF-B activation. A, HEK cells were cotransfected with pcDNA3-p75 (150 ng) along with either TRAF6 (150 ng) or antisense p62 (ASp62, 1 g) in the presence of NF-B reporter. NF-B activation was determined by luciferase assay (24). B, HEK cells were cotransfected with pcDNA3-p75 along with TRAF6 and increasing concentrations of p62 (1 or 2 g). NF-B activity was determined by luciferase assay. RLU, relative luciferase unit. C, PC12 cells were cotransfected with LipofectAMINE 2000 with increasing concentrations of p62 (0.5, 1.25, or 2.5 g) and NF-B reporter followed by stimulation with NGF (50 ng/ml) for 3 h. D, PC12 cells or NIH-3T3 cells expressing either p75 or TrkA were transiently transfected with increasing concentrations of antisense p62 construct (0.5, 1.25, or 2.5 g) or antisense p62 alone (2.5 g) with LipofectAMINE along with NF-B reporter. After 48 h the cells were stimulated with NGF (50 ng/ml) for 3 h followed by assay of luciferase activity.

FIG. 4. p62 regulates both p75 and TrkA functional properties.
A, HEK cells were cotransfected with pcDNA3-p75 (150 ng) along with either the IB␣ (IB␣M) super-suppressor (2.5 g), TRAF6 (2.5 g), p62 (2.5 g), or antisense p62 (2.5 g) as indicated. Cell death was monitored 48 h after transfection using trypan blue exclusion. (n ϭ 3). Antisense p 62(ASp62) significantly enhanced p75-mediated cell death (p Ͻ 0.05). PC12 cells were cotransfected with vector, p62, or antisense p62 construct using LipofectAMINE 2000. The following day, the medium was removed and fresh medium containing NGF (50 ng/ml) was added for 48 h. The neurite-bearing cells were derived by scoring cells with a process greater than two cell bodies in length for at least 300 cells in three separate fields. Antisense p62 impaired NGF-induced neurite outgrowth compared with control, vector, or p62 with NGF (p Ͻ 0.001). A paired t test was used to compare antisense p62 with other controls. B, PC12 cells were transiently transfected with 2.5 g of antisense p62 for 36 h followed by stimulation with NGF (50 ng/ml) for 15 min or 3 h. IKK was immunoprecipitated from an equivalent amount of lysate and suspended in kinase assay buffer, and the activity was determined by immune complex kinase assay employing GST-I␤␣ as substrate (4,21). Samples were analyzed by SDS-polyacrylamide gel electrophoresis/autoradiography. The lysates were Western-blotted (WB) with either IKK or p62 antibodies. C, a model depicting the ability of p62 to serve as a bridge between both p75 and TrkA receptor components. Formation of the signal complex enables aPKC to phosphorylate IKK leading to activation of NF-B.

TABLE I
Mapping of p62-TrkA interaction domains Myc-tagged p62 constructs were prepared as described previously (17) and coexpressed in HEK cells with HA-tagged TrkA. Interaction between p62 and TrkA was determined by coimmunoprecipitation (21). For comparison, the binding of p62 to TrkA was mapped relative to aPKC and TRAF6 (17). AID, acidic sequence and atypical PKC interacting domain; ZZ, zinc-finger domain.
p62 Is a Scaffold for NF-B Activation by NGF 7711 15 min), suggesting that it is a two-step process. Collectively, these results reveal that p62 interacts with TRAF6 in response to NGF and may likely serve as a bridge between both p75 and TrkA receptor components. As TRAF6 interaction with p75 results in activation of NF-B (7), it was of interest to determine whether the downregulation of p62 with a p62 antisense construct (17) would block the induction of NF-B as measured by a luciferase reporter system. The results shown in Fig. 3A demonstrate that this is the case, because there was a dramatic reduction of NF-B activation by the expression of p75 and TRAF6 in cells transfected with the p62 antisense construct as compared with the nontransfected cells. On the other hand, the transfection of a p62 expression vector, that by itself does not activate NF-B in this system (17), dramatically enhanced p75-TRAF6 or TrkA-TRAF6 activation of NF-B (Fig. 3B). In contrast, p62/ ZIP2, which lacks the TRAF6 binding site (16), failed to activate NF-B (Fig. 3B). Altogether this indicates that the recruitment of p62 to the NGF receptor signaling complex is critical for the activation of NF-B. Consistent with this notion, overexpression of p62 in PC12 cells resulted in a dose-dependent enhancement of both basal as well as NGF-stimulated activation of NF-B (Fig. 3C). We next conducted experiments to investigate whether the antisense construct of p62 would block NGF-induced activation of NF-B in cells expressing either one or both NGF receptors. Interestingly, antisense p62 blocked NGF-induced activation of NF-B in PC12 cells expressing both receptors (Fig. 3D). Likewise, p62 down-regulation in NIH-3T3 cells expressing either p75 or TrkA receptor (20) (Fig.  3D) also abrogated NGF-induced activation of NF-B. Collectively these findings demonstrate that p62, like aPKC (13), is essential in the activation of NF-B by NGF and that it serves to scaffold proximal NGF receptor components in this pathway.
The functional relevance of the presence of p62 in these complexes was addressed further in the following series of experiments. Overexpression of p75 in HEK cells results in ligand-independent cell death that is prevented by TRAF6 (7,8). Consistent with the role of NF-B in cell survival signaling (4), the expression of antisense p62 enhanced p75 mediated-cell death (Fig. 4A), whereas expression of TRAF6 or p62 blocked cell death. The activation of NF-B is likewise required for neuronal differentiation (9) and TrkA responsiveness (4). Transfection of antisense p62 significantly impaired NGF-induced neurite outgrowth, whereas overexpression of p62 enhanced NGF responsiveness (Fig. 4A). The mechanism whereby p62 regulates activation of NFϪB likely involves recruitment of TRAF6 and aPKC onto the p62 scaffold, thus enabling aPKC-mediated phosphorylation of IKK (21). To provide evidence for the involvement of p62 in this process, we assessed the activity of endogenous IKK activity in an in vitro kinase assay using GST-I␤␣ as the substrate (4,21). Transfection of antisense p62 suppressed NGF-stimulated activation of IKK (Fig. 4B).
The findings reported here provide new insight into the proximal components of the NGF/NF-B pathway and demonstrate formation of a p62 bridge that scaffolds together both p75 and TrkA receptors for the activation of NF-B (Fig. 4C). Our results stress the role of p62 as a common and critical intermediary that channels different signaling pathways toward IKK activation. Understanding the mechanism whereby the TRAF6-p62 complex is regulated in vivo is an area of ongoing study. Together, these findings underscore the importance of p62 as a scaffold for NF-B and as a common platform for communication of both p75 and TrkA receptor signals.