Atypical λ/ιPKC Conveys 5-Lipoxygenase/Leukotriene B4-mediated Cross-talk between Phospholipase A2s Regulating NF-κB Activation in Response to Tumor Necrosis Factor-α and Interleukin-1β*

The transcription factor nuclear factor κB (NF-κB) plays crucial roles in a wide variety of biological functions such as inflammation, stress, and immune responses. We have shown previously that secretory nonpancreatic (snp) and cytosolic (c) phospholipase A2 (PLA2) regulate NF-κB activation in response to tumor necrosis factor (TNF)-α or interleukin (IL)-1β activation and that a functional coupling mediated by the 5-lipoxygenase (5-LO) metabolite leukotriene B4 (LTB4) exists between snpPLA2and cPLA2 in human keratinocytes. In this study, we have further investigated the mechanisms of PLA2-modulated NF-κB activation with respect to specific kinases involved in TNF-α/IL-1β-stimulated cPLA2phosphorylation and NF-κB activation. The protein kinase C (PKC) inhibitors RO 31-8220, Gö 6976, and a pseudosubstrate peptide inhibitor of atypical PKCs attenuated arachidonic acid release, cPLA2 phosphorylation, and NF-κB activation induced by TNF-α or IL-1β, thus indicating atypical PKCs in cPLA2regulation and transcription factor activation. Transfection of a kinase-inactive mutant of λ/ιPKC in NIH-3T3 fibroblasts completely abolished TNF-α/IL-1β-stimulated cellular arachidonic acid release and cPLA2 activation assayed in vitro, confirming the role of λ/ιPKC in cPLA2 regulation. Furthermore, λ/ιPKC and cPLA2 phosphorylation was attenuated by phosphatidyinositol 3-kinase (PI3-kinase) inhibitors, which also reduced NF-κB activation in response to TNF-α and IL-1β, indicating a role for PI3-kinase in these processes in human keratinocytes. TNF-α- and IL-1β-induced phosphorylation of λ/ιPKC was attenuated by inhibitors toward snpPLA2 and 5-LO and by an LTB4 receptor antagonist, suggesting λ/ιPKC as a downstream effector of snpPLA2 and 5-LO/LTB4 the LTB4 receptor. Hence, λ/ιPKC regulates snpPLA2/LTB4-mediated cPLA2 activation, cellular arachidonic acid release, and NF-κB activation induced by TNF-α and IL-1β. In addition, our results demonstrate that PI3-kinase and λ/ιPKC are involved in cytokine-induced cPLA2 and NF-κB activation, thus identifying λ/ιPKC as a novel regulator of cPLA2.

Phospholipase A 2 (PLA 2 ) 1 enzymes play a crucial role in liberating free fatty acids and lysophospholipids from membrane phospholipids, thereby initiating the production of biologically active lipids, which mediate inflammatory reactions. PLA 2 s cleave the fatty acid substituent from the sn-2 position of membrane phospholipids leading to production of plateletactivating factor, lysophosphatidic acid, and arachidonic acid (AA). AA is metabolized to the bioactive eicosanoids, including prostaglandins, hydroxy fatty acids, leukotrienes, and thromboxanes, whose action is in part mediated by cell surface Gprotein-coupled receptors (1,2). Numerous types of mammalian PLA 2 s having distinct enzymatic properties, including substrate specificity and Ca 2ϩ requirements, have been identified and classified into several groups (3,4). Secretory low molecular weight PLA 2 s (14 -16 kDa) are characterized by an absolute requirement for millimolar concentrations of Ca 2ϩ and a broad specificity for phospholipids with different polar head groups and fatty acyl chains (5). Several distinct secretory nonpancreatic PLA 2 s (snpPLA 2 ), including group IIA, IID, IIE, IIF, III, V, and X, have been identified, and the roles of different snpPLA 2 s in regulation of cellular AA release and lipid mediator generation are extensively studied (4,6,7). Increased levels of group IIA snpPLA 2 in circulation and affected tissues have been found in association with various pathological conditions, including rheumatoid arthritis, sepsis, acute pancreatitis, psoriasis, and in endotoxic shock (8 -11). Group IV cytosolic PLA 2 (cPLA 2 ) consists of three human paralogs (12), of which cPLA 2 ␣, herein denoted cPLA 2 , is ubiquitously expressed. CPLA 2 is essential for prostaglandin, leukotriene, and platelet-activating factor formation, as determined, e.g. in cPLA 2 knock-out mice (13). Activation of cPLA 2 upon cellular stimulation is mediated by Ca 2ϩ -dependent translocation to membranes and by phosphorylations catalyzed by MAP kinase cascades (14,15).
Many genes that are implicated in the initiation of inflammatory processes are regulated by the transcription factor NF-B, e.g. inflammatory cytokines (IL-1␤, TNF-␣), chemokines , and the adhesion molecules ICAM-1 and VCAM-1 (16). In resting cells, NF-B is sequestered in an inactive form in cytosol bound to its inhibitor, IB. Following cellular activation * This work was supported by grants from the University of Trondheim (to M. W. A.), the Norwegian Research Council Grants 135802/310 (to M. W. A.) and 102264/310 (to B. J.), and the Norwegian Cancer Society Grants A00038/003 (to M. W. A.) and A98086/003 (to B. J.). 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.
We have recently reported the involvement of snpPLA 2 , LTB 4 , and cPLA 2 in TNF-␣ and IL-1␤-induced NF-B activation in human keratinocytes (28,29). We also found that a functional coupling between snpPLA 2 and cPLA 2 , in which cPLA 2 phosphorylation and activation was preceded by snpPLA 2 /5-LO/LTB 4 activity, was of importance for cytokineinduced NF-B activation (29). In this study, we have addressed the signaling mechanisms and kinases mediating cytokine-stimulated cPLA 2 and NF-B activation. We present herein novel results showing that atypical /PKC, in a PI3kinase-dependent manner, acts downstream of snpPLA 2 and 5-LO/LTB 4 and regulates cPLA 2 phosphorylation, leading to cellular AA mobilization and PLA 2 -mediated NF-B activation in response to TNF-␣ and IL-1␤.
Cell Culture-The spontaneously immortalized skin keratinocyte cell line HaCaT (kindly provided by Prof. N. Fusenig, Heidelberg, Deutsches Krebsforschungszentrum, Germany (30)) and NIH-3T3 fibroblasts were cultured in Dulbecco's modified Eagle's medium (DMEM) with 1 g of glucose/liter (HaCaT) or 4.5 g of glucose/liter (NIH-3T3), supplemented with 10% (v/v) fetal calf serum (FCS), 0.3 mg/ml gluta-mine, and 0.1 mg/ml gentamicin at 37°C with 5% CO 2 in a humidified atmosphere. The generation of HaCaT cells stably expressing the plasmid pBIIX with two copies of an HIV-NF-B sequence cloned upstream of the mouse fos promoter and the Photinus pyralis luciferase coding sequence (31) has been described previously (29). Isolation of NIH-3T3 fibroblasts stably transfected with v-ras or doubly transfected with v-ras-dn/PKC, which was grown in the presence of hygromycin (0.3 mg/ml), has been reported previously (32,33).
Luciferase Assay-HaCaT cells were seeded in 24-round multiwell plates (2.8 ϫ 10 5 cells/well). Treatment of the cells was carried out 2 days after reaching confluency. Treated cells were washed two times with phosphate-buffered saline and lysed, and luciferase activities were determined using the Luciferase Reporter Assay System (Promega) and Turner Luminometer model TD-20/20 (Turner Designs) as described by the manufacturer.

Measurement of [ 3 H]Arachidonic
Acid Release-NIH-3T3 and Ha-CaT cells were cultured in DMEM (with 4.5 or 1 g of glucose/liter for NIH-3T3 and HaCaT, respectively) and 10% FCS to 50% confluency (NIH-3T3) or to 2 days post-confluency (HaCaT). Cells were labeled for 16 h with [ 3 H]AA (0.4 Ci/ml) in DMEM culture medium containing 0.5% FCS. About 90% of the added [ 3 H]AA was incorporated by this procedure. After labeling, the cells were washed two times with phosphate-buffered saline containing fatty acid-free BSA (2 mg/ml) in order to remove unincorporated radioactivity. Cells were then allowed to equilibrate at 37°C before addition of indicated reagents. Treatments were carried out in the presence of fatty acid-free BSA (0.1 mg/ml). After the indicated times, the supernatants were cleared of detached cells by centrifugation (300 ϫ g, 10 min). The release of [ 3 H]AA from the cells was assessed by liquid scintillation counting. Adherent cells were dissolved in 1 M NaOH in order to determine incorporated [ 3 H]AA in the cells by liquid scintillation counting. The results are given as released [ 3
Statistical Analysis-All data are expressed as means Ϯ S.D. of separate experiments. Differences between means were determined by Student's t test for unpaired samples, and those at p Ͻ 0.05 were considered significant. In the figures the asterisks indicate that values are statistically significant from the given reference values.

PKC-and p38 MAP Kinase-dependent Pathways Are Involved in TNF-␣-and IL-1␤-induced NF-B Activation in Human Keratinocytes-PKC and MAP kinases have been implicated in cytokine-induced NF-B activation in several cell
types (37)(38)(39). To investigate this in human HaCaT keratinocytes, the effects of PKC and MAP kinase inhibitors on TNF-␣/IL-1␤-induced NF-B activation were examined. The PKC inhibitor RO 31-8220 and the p38 MAP kinase inhibitor SB203580 dose-dependently inhibited TNF-␣and IL-1␤-stimulated NF-B activation (Fig. 1A), indicating the participation of PKC and p38 MAP kinases in cytokine-elicited NF-B activation. In contrast, the p42/44 MAP kinase inhibitor PD 98059 did not affect cytokine-induced NF-B activation (data not shown). It has been demonstrated that RO 31-8220, in addition to its inhibitory effects on PKC isoforms, may inhibit the activity of p70 S6 kinase (p70 rsk ) (40). To investigate this possibility, we examined whether rapamycin, an inhibitor of the p70 rsk pathway (41,42), impaired NF-B activation induced by TNF-␣ or IL-1␤. Pretreatment of HaCaT cells with 1 nM to 1 M rapamycin had no significant effect on NF-B activation (data not shown). Thus, the observed inhibitory effects of RO 31-8220 on TNF-␣/IL-1␤-induced AA signaling and NF-B activation in HaCaT keratinocytes can be attributed to inhibition of PKC enzymes, independent of rapamycin-sensitive pathways. The PKC inhibitors RO 31-8220 and Gö 6976 preferentially inhibit conventional PKC isoenzymes (␣, ␤, and ␥) and novel PKCs (␦, ⑀, and ) with much greater efficacy (IC 50 values in the nM range) than atypical PKCs (IC 50 values in M range (43)). Both RO 31-8220 and Gö 6976 had significant inhibitory effects on TNF-␣/IL-1␤-elicited NF-B activation only at 1 and 5 M (Fig.  1, data not shown). Also, phorbol 12-myristate 13-acetate (PMA) in a wide range of concentrations failed to activate NF-B in HaCaT cells, indicating that PMA-insensitive atypical PKCs are involved in NF-B activation in keratinocytes. Therefore, to examine if atypical PKCs participate in TNF-␣/ IL-1␤-elicited NF-B activation, we investigated cytokine-induced NF-B activation in the presence of a cell-permeable myristoylated pseudosubstrate peptide inhibitor toward atypical PKCs (43,44). The pseudosubstrate peptide inhibitor, mimicking a conserved sequence in atypical PKCs (/PKC and PKC), profoundly inhibited TNF-␣-or IL-1␤-induced NF-B activation with more efficient inhibition at increasing concentrations (Fig. 1B), exerting similar inhibitory effects as observed for PKC-mediated insulin-stimulated glucose uptake in adipocytes (43). Inhibitors or vehicles alone had no effect on B-dependent transcription (data not shown). Thus, these results indicate that atypical PKCs and p38 MAP kinase are involved in TNF-␣/IL-1␤-induced NF-B activation in human HaCaT keratinocytes.
Cellular AA Release and cPLA 2 Phosphorylation in Response to IL-1␤ Are Mediated by PKC-To examine if PKCs regulate PLA 2 enzymes in HaCaT cells, we examined the effects of RO 31-8220 and the atypical PKC pseudosubstrate peptide inhibitor on cellular AA mobilization. Both inhibitors dose-dependently impaired IL-1␤-elicited AA release ( Fig. 2A), indicating that atypical PKCs act upstream of AA releasing PLA 2 enzymes. We have previously shown that TNF-␣-and IL-1␤stimulated AA release from HaCaT cells is accompanied by cPLA 2 phosphorylation (29), and phosphorylation of cPLA 2 is important for its activation and AA releasing activity in response to cytokines (45)(46)(47). To determine if the inhibitory effect of RO 31-8220 on AA mobilization correlated with reduced phosphorylation of cPLA 2 , we analyzed cPLA 2 phosphorylation levels by in vivo phosphorylation experiments. HaCaT cells were metabolically labeled with [ 32 P]orthophosphate and pretreated with RO 31-8220 before addition of IL-1␤. Treat- ment with TNF-␣ or IL-1␤ for 1 h resulted in 2-4-fold increase in cPLA 2 phosphorylation as reported previously (29), and IL-1␤-stimulated phosphorylation of cPLA 2 was reduced in the presence of RO 31-8220 (1 M ; Fig. 2B). Similar results were obtained with TNF-␣ (data not shown). These results indicate that atypical PKCs (since the specific pseudosubstrate inhibitor attenuates AA liberation) are upstream mediators of TNF-␣/IL-1␤-induced cPLA 2 phosphorylation and AA mobilization in HaCaT keratinocytes.
Transfection of a Kinase-inactive /PKC Abrogates TNF-␣stimulated AA Mobilization and cPLA 2 Activation-To investigate further the participation of the atypical PKC isoform /PKC in cPLA 2 activation and AA release, we examined TNF-␣-stimulated AA release in NIH-3T3 fibroblasts transfected with v-ras or doubly-transfected with v-ras and a kinase-inactive mutant of /PKC previously shown to act in a dominant negative manner (48). Transfection of the kinase-inactive /PKC abolished TNF-␣-elicited AA release compared with v-ras-transfected cells, reducing AA liberation in v-ras-dn/ PKC-transfected cells to similar levels as in wild type NIH-3T3 cells (Fig. 3A). Also, TNF-␣-induced AA liberation in v-rastransfected cells was completely blocked by the PKC inhibitor RO 31-8220 (data not shown). To determine if impaired cellular AA mobilization in v-ras-dn/PKC-transfected (compared with v-ras-transfected cells) correlated with decreased cPLA 2 activity, TNF-␣-induced cPLA 2 activity was determined in vitro using sonicated vesicles of arachidonoylphosphatidylcholine as substrate. We did not observe any significant changes in cPLA 2 activity in lysates from v-ras-dn/PKC-transfected cells treated from 30 min to 10 h, whereas cPLA 2 activity in lysates from v-ras-transfected cells was reproducibly increased 1.5-2fold (Fig. 3B). The TNF-␣-induced cPLA 2 activity was inhibited by the c/iPLA 2 inhibitor MAFP (Fig. 3B). These results confirm results shown in Fig. 2, thus suggesting that /PKC is an upstream mediator of TNF-␣-elicited cPLA 2 activation.

TNF-␣-and IL-1␤-induced Phosphorylation of Atypical PKCs Is Impaired by Inhibitors of the 5-LO Pathway-
We have recently shown that 5-lipoxygenase (LO) metabolites, including LTB 4 , regulate cPLA 2 phosphorylation and AA release in cytokine-treated HaCaT keratinocytes (29). Hence, from our previous and present results, we propose that both atypical /PKC and 5LO/LTB 4 are involved in cytokine-induced NF-B activation, cellular AA release, and cPLA 2 phosphorylation. To examine the sequential involvement of 5-LO/LTB 4 and /PKC, we studied the effect of inhibitors of the 5-LO pathway on cytokine-induced /PKC phosphorylation. Generally, a 2.5-5fold increase in 32 P incorporation into /PKC was observed in keratinocytes that had been incubated in the presence of [ 32 P]orthophosphate prior to TNF-␣/IL-1␤ treatment . TNF-␣/IL-1␤-induced phosphorylation of /PKC was markedly reduced in the presence of the 5-LO inhibitors L-655,238 and MK-886 (Fig. 4). Importantly, the specific LTB 4 receptor antagonist LY255283 also attenuated cytokine-induced /PKC phosphorylation (Fig. 4A), showing specifically the role of LTB 4 and the LTB 4 receptor in /PKC regulation. Hence, these results suggest that /PKC is a novel downstream effector of LTB 4 and possibly of other 5-LO metabolites.

TNF-␣-or IL-1␤-induced Phosphorylation of /PKC and cPLA 2 Is PI3-Kinase-dependent-Atypical
PKCs have been reported to be phosphorylated in a PI3-kinase dependent manner in response to different stimuli (49 -51). The PI3-kinase inhibitor LY294002 dose-dependently inhibited TNF-␣and IL-1␤stimulated NF-B activation in HaCaT keratinocytes (Fig. 5A). Also, wortmannin, a PI3-kinase inhibitor structurally unrelated to LY294002, resulted in complete inhibition of cytokinestimulated NF-B activation at 100 nM (data not shown), indicating that PI3-kinase contributes to NF-B activation. To investigate if PI3-kinase is involved in cytokine-elicited /PKC and cPLA 2 phosphorylation in HaCaT keratinocytes, we pretreated [ 32 P]orthophosphate-labeled cells with the PI3-kinase inhibitor LY294002 before addition of TNF-␣ or IL-1␤. TNF-␣and IL-1␤-induced phosphorylation of /PKC or cPLA 2 was reproducibly reduced in the presence of LY294002 (Fig. 5, B and C), thus indicating PI3-kinase as an upstream effector of /PKC and cPLA 2 . Similar results were obtained with the structurally unrelated PI3-kinase inhibitor wortmannin (data not shown). To examine if PI3-kinase is activated upon cytokine addition, HaCaT cells were stimulated with TNF-␣ for various times, and PI3-kinase was immunoprecipitated using antibodies toward the p85 subunit of tyrosine-activated class I A or the p110␥ subunit of G-protein receptor-activated class I B PI3-kinase (52,53), before in vitro determination of PI3-kinase activity. Activity of both PI3-kinase isoforms was enhanced by TNF-␣, with an earlier response for p85-than for p110␥-dependent activity (maximal at 20 versus 60 min; Fig. 5D). Thus, PI3-kinase isoforms are activated in response to cytokines and participate(s) in cytokine-elicited /PKC/cPLA 2 phosphorylation and NF-B activation in HaCaT keratinocytes. snpPLA 2 Activity Precedes /PKC Phosphorylation Induced by TNF-␣ and IL-1␤-We have demonstrated recently that catalytic activity of snpPLA 2 and 5-LO is necessary for TNF-␣/IL-1␤-induced cPLA 2 phosphorylation and NF-B activation (29). Therefore, we next examined if snpPLA 2 activity also precedes /PKC phosphorylation, by preincubating [ 32 P]orthophosphate-labeled HaCaT keratinocytes with snpPLA 2 inhibitors before cytokine treatment. The group IIA snpPLA 2 -active site-directed inhibitor SB203347 (54, 55) and 12-epi-scalaradial (56) significantly reduced /PKC phosphorylation induced by IL-1␤ and TNF-␣ (Fig. 6). To elucidate whether the inhibitory effect of snpPLA 2 inhibitors on /PKC phosphorylation was due to reduced AA levels, we examined the influence of AA addition on /PKC phosphorylation in the presence of snpPLA 2 inhibitors. Exogenously added AA was able to relieve the inhibitory effect of 12-epi-scalaradial on IL-1␤-and TNF-␣-induced phosphorylation of /PKC (Fig. 6A; data not shown). Similar results with addition of AA were obtained for /PKC phosphorylation inhibited by the active site-directed snpPLA 2 inhibitor SB203347 (data not shown). Hence, we may suggest that snpPLA 2 activity and AA . /PKC (B) or cPLA 2 (C) were immunoprecipitated and subjected to SDS-PAGE and electroblotting. Incorporated 32 P radioactivity was quantified by 32 P digitalization using a PhosphorImager, normalized to /PKC/ cPLA 2 protein levels, and is expressed as fold induction. Similar results were obtained in three independent experiments. D, PI3-kinase was immunoprecipitated with anti-p85 (q) or anti-p110␥ (E) antibodies from HaCaT cells treated with TNF-␣ (10 ng/ml) for various times, and PI3-kinase activity was assayed as described under "Experimental Procedures." Similar results were obtained in two different experiments. . /PKC was immunoprecipitated and subjected to SDS-PAGE and electroblotting. Incorporated 32 P radioactivity was quantified by 32 P digitalization using a PhosphorImager, normalized to /PKC protein levels, and is expressed as fold induction. Similar results were obtained in four independent experiments. tion, activation of polymorphonuclear neutrophils, superoxide anion release, and inflammatory gene expression (54,(57)(58)(59)(60)(61). We have recently shown that both snpPLA 2 and cPLA 2 , through AA-derived 5-LO metabolites as LTB 4 acting through its G-protein-coupled receptor in an autocrine feedback mechanism, contribute to activation of the immunomodulatory transcription factor NF-B in response to TNF-␣ and IL-1␤ (29). Other authors (62,63) have also described that snpPLA 2 and LTB 4 (acting in an autocrine fashion) increase AA release, cPLA 2 phosphorylation, and cPLA 2 activation. However, the mechanisms explaining how snpPLA 2 /LTB 4 regulate cPLA 2 activation is not understood in detail. In the present study we show that atypical /PKC in a PI3-kinase-dependent manner conveys snpPLA 2 /LTB 4 -mediated cPLA 2 phosphorylation and cellular AA mobilization in response to TNF-␣ and IL-1␤, promoting NF-B activation. Several lines of evidence support the sequential involvement of snpPLA 2 /LTB4, /PKC, and cPLA 2 as follows. 1) TNF-␣/IL-1␤-stimulated AA liberation and cPLA 2 phosphorylation are reduced by PKC inhibitors. 2) TNF-␣induced AA release and cPLA 2 activation are abrogated by transfection of a kinase-inactive mutant of /PKC kinase. 3) PI3-kinase is activated by cytokines and /PKC, and cPLA 2 phosphorylation is inhibited in a PI3-kinase-dependent manner. 4) Inhibitors of 5-LO/LTB 4 functionality attenuate cytokine-elicited /PKC phosphorylation. 5) TNF-␣/IL-1␤-stimulated /PKC phosphorylation is reduced by snpPLA 2 inhibitors and relieved by exogenously added AA. Additional information on the sequential involvement of PLA 2 s and LTB 4 is given by our previous results showing that snpPLA 2 /5-LO/LTB 4 regulates AA release and cPLA 2 phosphorylation (29), also illustrating the positive feedback mechanism exerted by LTB 4 in Ha-CaT keratinocytes, as suggested in other cell types.
Previously, only phorbol ester-sensitive PKCs have been implicated in cPLA 2 phosphorylation and activation (64 -66). In contrast, our results demonstrate for the first time that PMAinsensitive atypical /PKC regulates cPLA 2 activity and AA liberation. It has been shown previously that PMA induces cPLA 2 phosphorylation in human platelets, but PKC inhibitors failed to reduce cPLA 2 phosphorylation and AA release, thus illustrating that PMA-stimulated cPLA 2 phosphorylation is non-physiological (67). Although Ser-727 in cPLA 2 is located in a consensus motif that can be phosphorylated by PKC or PKA (14), and the enzyme is phosphorylated by PKC in vitro (68,69), it is uncertain if /PKC directly phosphorylates cPLA 2 in our model. Perhaps more probable is that cPLA 2 phosphorylation conveyed by /PKC is brought about in a p38 MAP kinasedependent manner causing phosphorylation on Ser-505 as reported (70,71) and possibly on Ser-727, recently shown to be phosphorylated by the p38 MAP kinase-activated MAP kinase interacting kinase 1 or a closely related kinase (15). MAP kinases are known to be activated downstream of atypical PKCs (33,(72)(73)(74)(75), and the involvement of p38 or p42/44 MAP kinases in TNF-␣/IL-1␤/sPLA 2 /LTB 4 -mediated cPLA 2 activation has recently been suggested (46,63,76,77). Also, p38 MAP kinase and /PKC have been implicated in TNF-␣/IL-1␤-stimulated NF-B activation (37-39, 79 -81). We find that TNF-␣ and IL-1␤ activates p38 and p42/44 MAP kinases in HaCaT keratinocytes, 2 but only p38 MAP kinase inhibition abrogates NF-B activation (Fig. 1). Furthermore, inhibition of p38 MAP kinase reduces TNF-␣-induced cPLA 2 phosphorylation, 2 suggesting that p38 MAP kinase participates in cPLA 2 and NF-B regulation. However, it is uncertain whether there is a direct sequential relationship between p38 MAP kinase and /PKC in HaCaT keratinocytes or if they are parts of parallel signaling pathways.
Inhibition of PI3-kinase attenuated cytokine-induced phosphorylation of /PKC/cPLA 2 and NF-B activation (Fig. 5), thus suggesting that PI3-kinase acts upstream of /PKC and cPLA 2 in NF-B activation. Indeed, PI3-kinase is known as an upstream regulator of atypical PKCs, which are subject to regulation by both lipid binding and phosphorylation in vivo (49 -51). Also, our results corroborate recent findings showing that PI3-kinase mediates cytokine-elicited NF-B activation (82,83). Interestingly, we observed that both p85-dependent class I A and G-protein activated class I B PI3-kinases were activated in response to TNF-␣ (with slower activation of the G-protein activated PI3-kinase isoform; Fig. 5). However, the precise location in the signaling pathway and the contribution of either PI3-kinase isoform to /PKC/cPLA 2 phosphorylation and NF-B activation in HaCaT cells is unknown and should be determined in future experiments. Class I A PI3-kinase has been reported to bind to IL1␤-induced tyrosine-phosphorylated IL-1 receptor 1 through binding of SH2 domains in the p85 subunit, whereas class I B PI3-kinase is activated downstream of G-protein-coupled receptors, coupling those to c-Jun N-terminal kinase/stress-activated protein kinase and MAP kinase cascades (84 -86). Of note, lysophosphatidic acid and LTB 4 acting via G-protein-coupled receptors induce PI3-kinase activity (75,87), of relevance to our results.
Atypical PKCs and PLA 2 enzymes/AA signaling contribute to signaling pathways controlling cell growth, differentiation, and survival (65, 78, 88 -91). Hence, our results linking atypical PKCs and PLA 2 s in NF-B signaling may provide a molecular explanation to the participation of both atypical PKCs and PLA 2 s to similar cellular functions. A mechanistic explanation to the roles of atypical PKCs in NF-B activation has been provided by the findings that atypical PKCs bind to the scaffold protein p62 via protein complex formation with receptor interacting protein or TRAF6 in TNF-␣-or IL-1␤-signaling, respectively (39,81). Importantly, these protein interactions serve to link atypical PKCs to IKK␤ and NF-B activation. The presence or identities of intermediary steps between the PLA 2 / LTB 4 -mediated autocrine feedback loop and the known NF-B signaling components are at present unknown. However, the PLA 2 -mediated activity contributing to NF-B activation could reside in a signaling pathway parallel to the classical MAP3K-IKK-IB␣ pathway. Alternatively, PLA 2 -promoted activity could couple upstream the NF-B-inducing kinase or IB kinases (IKKs). The coupling to IKKs could possibly be mediated through atypical PKC, acting in the PLA 2 -mediated positive feedback loop, e.g. by interaction of atypical PKC with IKK␤ through the p62 adapter protein as described (39,81). Thus, PLA 2 -mediated activity may contribute to NF-B activation by enhancing phosphorylation-induced events increasing kinase activities, e.g. of atypical PKC, resulting in enhanced NF-Bdependent transcription.
In conclusion, our results elucidate the mechanisms underlying the functional coupling and positive feedback mechanisms between snpPLA 2 /LTB 4 and cPLA 2 , identifying novel signaling pathways involving atypical /PKC as a downstream effector of 5-LO/LTB 4 promoting cPLA 2 phosphorylation, cellular AA mobilization, and NF-B activation in TNF-␣/IL-1␤treated human keratinocytes.