Protein Kinase C-δ Regulates Thrombin-Induced ICAM-1 Gene Expression in Endothelial Cells via Activation of p38 Mitogen-Activated Protein Kinase

Abstract

The procoagulant thrombin promotes the adhesion of polymorphonuclear leukocytes to endothelial cells by a mechanism involving expression of intercellular adhesion molecule 1 (ICAM-1) via an NF-κB-dependent pathway. We now provide evidence that protein kinase C-δ (PKC-δ) and the p38 mitogen-activated protein (MAP) kinase pathway play a critical role in the mechanism of thrombin-induced ICAM-1 gene expression in endothelial cells. We observed the phosphorylation of PKC-δ and p38 MAP kinase within 1 min after thrombin challenge of human umbilical vein endothelial cells. Pretreatment of these cells with the PKC-δ inhibitor rottlerin prevented the thrombin-induced phosphorylation of p38 MAP kinase, suggesting that p38 MAP kinase signals downstream of PKC-δ. Inhibition of PKC-δ or p38 MAP kinase by pharmacological and genetic approaches markedly decreased the thrombin-induced NF-κB activity and resultant ICAM-1 expression. The effects of PKC-δ inhibition were secondary to inhibition of IKKβ activation and of subsequent NF-κB binding to the ICAM-1 promoter. The effects of p38 MAP kinase inhibition occurred downstream of IκBα degradation without affecting the DNA binding function of nuclear NF-κB. Thus, PKC-δ signals thrombin-induced ICAM-1 gene transcription by a dual mechanism involving activation of IKKβ, which mediates NF-κB binding to the ICAM-1 promoter, and p38 MAP kinase, which enhances transactivation potential of the bound NF-κB p65 (RelA).

The proinflammatory mediator thrombin, released during intravascular coagulation and tissue injury, is an important regulator of polymorphonuclear leukocytes' (PMN) adhesion to the endothelium (45, 69). The basis of thrombin-induced endothelial adhesivity towards PMN involves endothelial cell surface expression of adhesive proteins, such as intercellular adhesion molecule 1 (ICAM-1; CD54) (45). ICAM-1, a ligand for the leukocyte β₂ integrins LFA-1 and Mac-1 (CD11a/CD18 and CD 11b/CD18) (15, 37), mediates the tight adhesive binding of PMN and thus facilitates PMN migration across the vascular endothelial barrier (55, 56). We have shown that the transcription factor NF-κB p65 (RelA) is the key regulator of endothelial ICAM-1 gene expression following thrombin activation of GTP-binding protein (G-protein)-coupled receptor, proteinase-activated receptor 1 (PAR-1) (45).

NF-κB, typically a heterodimer of 50-kDa (p50) and 65-kDa (RelA) subunits, is sequestered in the cytoplasm of most cells in association with IκB proteins that mask the nuclear localization sequence of NF-κB (3, 66). NF-κB activity is primarily regulated at the level of IκB degradation, which is accomplished through serine phosphorylation (Ser32 and Ser36) of IκBα (62), the principal inhibitor of NF-κB, by IκBβ kinase (IKKβ) (39, 68). Phosphorylation targets IκBα for ubiquitination and proteasome-mediated degradation (2, 9, 48). The released NF-κB then undergoes nuclear translocation and subsequent binding to NF-κB-responsive elements in genes including ICAM-1. Studies have shown the existence of an additional signaling pathway in which the transactivation potential of NF-κB is directly stimulated through phosphorylation (1, 4, 34, 65). Despite the requirement of NF-κB in mediating ICAM-1 expression (45), the exact mechanisms by which thrombin signals NF-κB activation in endothelial cells are unknown.

Protein kinase C (PKC) is a multigene family of serine/threonine kinases mediating intracellular signaling (38, 41, 43). PKC isoforms are classified into three groups based on their structure and activation mechanisms: phosphatidylserine-, diacylglycerol-, and Ca²⁺-dependent conventional PKC (cPKC-α, -βI, -βII, and -γ), Ca²⁺-independent novel PKC (nPKC-δ, -ɛ, -μ, -θ, and -η) isoforms, and diacylglycerol- and Ca²⁺-independent atypical PKC (aPKC-ζ and -λ/ι) isoforms. The tissue distribution of PKC isoforms varies considerably, with PKC-α, -δ, and -ζ being widespread, whereas others are localized in a tissue- or cell-type-specific manner. In addition to PKC-α, -δ and -ζ, endothelial cells also express the PKC-β, -ɛ, -η, and -θ isoforms (25, 47). Of these, PKC-α, -ɛ, -θ, and -ζ are known to activate NF-κB (27, 32, 47, 60, 61); however, the role of PKC-δ in this response is unclear.

The p38 mitogen-activated protein (MAP) kinases are another important family of serine/threonine kinases activated by a variety of stimuli, including thrombin (19, 21, 24, 29, 49, 50, 64). Activation of p38 MAP kinase has been implicated in the induction of multiple responses, including the regulation of NF-κB activity (7, 11, 65). In the present study, we determined whether activation of PKC-δ is a critical requirement for thrombin-induced NF-κB activity and ICAM-1 expression. Our data establish that thrombin induces PKC-δ activation in endothelial cells and that this event is required for IKKβ and p38 MAP kinase activation. We further demonstrate that PKC-δ activation of IKKβ and of p38 MAP kinase contributes to the mechanism of thrombin-induced ICAM-1 expression by activating NF-κB in the cytoplasm and increasing the transactivating potential of NF-κB p65 in the nucleus.

MATERIALS AND METHODS

Materials.

Human thrombin with an activity of 3,170 NIH U/mg was purchased from Enzyme Research Laboratories (South Bend, Ind.). Polyclonal antibodies against PKC-δ, p38 MAP kinase, IκBα, IκBβ, or NF-κB p65 and a monoclonal antibody against ICAM-1 were obtained from Santa Cruz Biotechnology (Santa Cruz, Calif.). A polyclonal antibody against MAP kinase-activated protein (MAPKAP) kinase 2 was purchased from Upstate Biotechnology (Lake Placid, N.Y.). Antibodies that detect PKC-δ only when activated by phosphorylation at Thr505 in a phosphoinositide 3-kinase-dependent fashion (30) or detect p38 MAP kinase when activated by dual phosphorylation at Thr180 and Tyr182 were obtained from New England Biolabs (Beverly, Mass.). The following items were purchased: polyvinylidene difluoride membrane from Millipore Corp. (Bradford, Mass.); phorbol myristate acetate (PMA), calphostin C, and staurosporine from Sigma Chemical Co. (St. Louis, Mo.); SB203580 and rottlerin from Calbiochem-Novabiochem Corp. (La Jolla, Calif.); a protein assay kit from Bio-Rad Laboratories (Hercules, Calif.); and a plasmid maxi kit from QIAGEN Inc. (Valencia, Calif.). LY379196 was kindly provided by Michael Jirousek (Lilly Research Laboratories, Indianpolis, Ind.). All other materials were from Fisher Scientific Co. (Pittsburgh, Pa.).

Cell culture.

Human umbilical vein endothelial cells (HUVEC; Clonetics, La Jolla, Calif.) were cultured as described (45) in gelatin-coated flasks using endothelial basal medium 2 (EBM2) with bullet kit additives (Clonetics). Confluent cells were incubated for 2 to 12 h in heat-inactivated 0.5 to 1% fetal bovine serum (FBS) containing EBM2 prior to thrombin challenge. All experiments, except where indicated, were made in cells under the eighth passage.

Cell lysis and immunoblotting.

Cells were challenged with the indicated concentrations of thrombin for the indicated periods of time. After treatment, the cells were lysed in a phosphorylation lysis buffer (50 mM HEPES, 150 mM NaCl, 200 μM sodium orthovanadate, 10 mM sodium pyrophosphate, 100 mM sodium fluoride, 1 mM EDTA, 1.5 mM magnesium chloride, 10% glycerol, 0.5 to 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride [PMSF], and 10 μg of aprotinin/ml). For ICAM-1 expression, cells were lysed with sodium dodecyl sulfate (SDS)-sample buffer (10 mM Tris-HCl [pH 6.8], 4% SDS, 20% glycerol, 0.4% dithiothreitol (DTT), and 1 mM sodium orthovanadate with bromophenol blue). Cell lysates were analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) and transferred onto nitrocellulose (Bio-Rad Laboratories) or polyvinylidene difluoride (Millipore Corp.) membranes, and the residual binding sites on the filters were blocked by incubating with 5% (wt/vol) nonfat dry milk in Tris-buffered saline–Tween solution (10 mM Tris-HCl [pH 8.0], 150 mM NaCl, and 0.05% Tween 20) for 1 h at room temperature or overnight at 4°C. The membranes were subsequently incubated with indicated antibodies and developed using an enhanced chemiluminescence method as described (47, 64).

PKC-δ and MAPKAP kinase 2 assay.

Cells were serum starved by overnight incubation in EBM2–1% FBS. The cells were subsequently challenged with thrombin (2.5 to 5 U/ml) for 5 min in the absence and presence of rottlerin, LY379196, or SB203580 (10 μM), which was added 30 to 60 min prior to thrombin treatment. The cells were then lysed with phosphorylation lysis buffer described above. Cell lysates were immunoprecipitated with an antibody against PKC-δ or MAPKAP kinase 2 using protein G-Sepharose (Amersham Pharmacia Biotech) as previously described (64). The immunocomplexes were washed three times with phosphorylation lysis buffer and two times with kinase buffer (25 mM Tris-HCl [pH 7.4], 5 mM MgCl₂, 0.5 mM EGTA, 1 mM DTT, 20 μg of phosphatidylserine, and 20 μM ATP [for PKC-δ] or 25 mM HEPES [pH 7.4], 25 mM MgCl₂, 25 mM β-glycerophosphate, 2 mM DTT, 0.1 mM sodium orthovanadate, and 20 μM ATP [for p38 MAP kinase]) and were resuspended in 30 μl of kinase buffer containing 5 μg of histone H1 (for PKC-δ) or heat shock protein 25 (Hsp-25) (for p38 MAP kinase), and 20 to 30 μCi of [γ-³²P]ATP was added. The reaction was incubated for 15 to 30 min at room temperature and was terminated by the addition of SDS-sample buffer. Proteins were analyzed by SDS-PAGE, and the phosphorylated form of histone H1 or Hsp-25 was detected by autoradiography.

Northern analysis.

Total RNA was isolated from HUVEC with the RNeasy kit (QIAGEN Inc.) according to the manufacturer's recommendations. Quantification and determination of the purity of RNA were performed by measuring A₂₆₀ and A₂₈₀, and an aliquot of RNA (20 μg) from samples with ratio above 1.6 was fractionated using a 1% agarose formaldehyde gel. The RNA was transferred to Duralose-UV nitrocellulose membrane (Stratagene, La Jolla, Calif.) and was covalently linked by UV irradiation using a Stratalinker UV cross-linker (Stratagene). Human ICAM-1 (0.96-kb SalI-to-PstI fragment) (58) and rat glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1.1-kb PstI fragment) were labeled with [α-³²P]dCTP using the random primer kit (Stratagene), and hybridization was carried out as described previously (46). Briefly, the blots were prehybridized for 30 min at 68°C in QuikHyb solution (Stratagene) and were hybridized for 2 h at 68°C with randomly primed ³²P-labeled probes. After hybridization, the blots were washed twice for 30 min at room temperature in 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) with 0.1% SDS followed by 2 washes for 15 min each at 60°C in 0.1× SSC with 0.1% SDS. Autoradiography was performed with an intensifying screen at −70°C for 12 to 24 h. The nitrocellulose membrane was soaked for stripping the probe with boiled water or 0.1× SSC with 0.1% SDS.

Reporter gene constructs, endothelial cell transfection, and luciferase assay.

The plasmid pNF-κB-LUC containing five copies of consensus NF-κB sequences linked to a minimal E1B promoter-luciferase gene was purchased from Stratagene. The expression vector pcDNA3 containing tagged kinase-defective PKC-α, -ɛ, and -δ isoforms was a gift from I. B. Weinstein (Columbia University, New York, N.Y.). The PKC-α, -ɛ, and -δ mutants were generated by replacing arginine with lysine 368, 437, and 376, respectively, and therefore lack a functional catalytic domain (57). The construct pCMVp38AGF containing a kinase-defective mutant of p38 MAP kinase (14) was kindly provided by R. J. Davis (Howard Hughes Medical Institute, University of Massachusetts, Worcester, Mass.). The construct encoding kinase-defective mutant of IKKβ was described elsewhere (67). Transfections were performed using the DEAE-dextran method (33) with slight modifications (47). Briefly, 5 μg of DNA was mixed with 50 μg of DEAE-dextran/ml in serum-free EBM2, and the mixture was added onto cells which were 70 to 80% confluent. We used 0.125 μg of pTKRLUC plasmid (Promega Corp., Madison, Wis.) containing Renilla luciferase gene driven by the constitutively active thymidine kinase promoter to normalize the transfection efficiencies. After 1 h, cells were incubated for 4 min with 10% dimethyl sulfoxide in serum-free EBM2. The cells were then washed twice with EBM2–10% FBS and grown to confluence. Using this protocol, we achieved a transient-transfection efficiency of 11 ± 2 (mean ± standard deviation; n = 3) for HUVEC.

In some experiments, we used Superfect (Qiagen) to transfect the cells as previously described (47). Briefly, reporter DNA (1 μg) was mixed with 5 μl of Superfect in 100 μl of serum-free EBM2 (Clonetics). We used 0.1 μg of pTKRLUC to normalize the transfection efficiencies. After a 5- to 10-min incubation at room temperature, 0.6 ml of EBM2–10% FBS was added and the mixture was applied to the cells that had been washed once with phosphate-buffered saline. Three hours later, the medium was changed to pure EBM2–10% FBS and the cells were grown to confluence. This protocol resulted in a transient-transfection efficiency of 20 ± 2 (mean ± standard deviation; n = 3). Cell extracts were prepared and assayed for luciferase activity using the Promega Biotech assay system (Promega). Firefly luciferase activity was normalized to Renilla luciferase activity and expressed as relative light units (RLU)/microgram of cell protein. Protein content was determined using a Bio-Rad protein determination kit (Bio-Rad Laboratories).

We used the trypan blue (Sigma Chemical Co.) exclusion assay to evaluate cell viability following transfection. The cells were washed gently with phosphate-buffered saline twice and trypsinized and were then resuspended and washed with EBM2–10% FBS. The cell suspension (10 μl) was mixed with 10 μl of 1× trypan blue solution, and 10 μl of the resulting mixture was loaded onto a hemocytometer. Results showed that >95% of the cells were viable.

Transfection of HUVEC with oligonucleotides.

Phosphorothioate oligonucleotides to PKC-ζ (sense [ATG CCC AGC AGG ACC] and antisense [GGT CCT GCT GGG CAT] have been described elsewhere (16); both are targeted to the translation initiation codon of PKC-ζ mRNA. Cells were grown in 100-mm-diameter dishes to 50% confluence. Transfection of oligonucleotides was performed with Lipofectin (Gibco-BRL, Grand Island, N.Y.) as described previously (46).

Cytoplasmic and nuclear extract preparation.

After treatments, cells were washed twice with ice-cold Tris-buffered saline and resuspended in 400 μl of buffer A (10 mM HEPES [pH 7.9], 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM [DTT], and 0.5 mM PMSF). After 15 min, NP-40 was added to a final concentration of 0.6%. Samples were centrifuged to collect the supernatants containing cytosolic proteins to determine IKBα degradation by Western blot analysis. The pelleted nuclei were resuspended in 50 μl of buffer B (20 mM HEPES [pH 7.9], 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, and 1 mM PMSF). After 30 min at 4°C, lysates were centrifuged and supernatants containing the nuclear proteins were transferred to new vials. The protein concentration of the extract was measured using a Bio-Rad protein determination kit (Bio-Rad Laboratories).

EMSA.

Electrophoretic mobility shift assays (EMSA) were performed as previously described (63). Briefly, 10 μg of nuclear extract was incubated with 1 μg of poly(dI-dC) in a binding buffer (10 mM Tris-HCl [pH 7.5], 50 mM NaCl, 0.5 mM DTT, and 10% glycerol [20-μl final volume]) for 15 min at room temperature. Then end-labeled double-stranded oligonucleotides containing an NF-κB site (30,000 cpm each) were added, and the reaction mixtures were incubated for 15 min at room temperature. The DNA-protein complexes were resolved by 5% native PAGE in low-ionic-strength buffer (0.25× Tris-borate-EDTA). The oligonucleotide used for the gel shift analysis was NF-κB 5′-AGTTGAGGGGACTTTCCCAGGC-3′ or ICAM-1 NF-κB 5′-AGCTTGGAAATTCCGGAGCTG-3′). The NF-κB oligonucleotide contains the consensus NF-κB binding site sequence (underlined) present in pNF-κB-LUC. The ICAM-1 NF-κB oligonucleotide represents a 21-bp sequence of ICAM-1 promoter encompassing the NF-κB binding site located 183 bp upstream of the transcription initiation site (22). The sequence motifs within the oligonucleotides are underlined.

PMN adhesion assay.

HUVEC were seeded at 50,000 cells/well in gelatin-coated 96-well plates and grown to confluence. PMN were isolated from whole blood of healthy donors using Polymorphprep (NYCOMED, Oslo, Norway). PMN were labeled with calcein (Molecular Probes, Eugene, Oreg.) for 35 min as described elsewhere (35). Following treatment, cells were washed twice and allowed to equilibrate in M2 buffer (150 mM NaCl, 20 mM HEPES, 10 mM glucose, 5 mM KCl, 1 mM CaCl₂, 1 mM MgCl₂, and 1 mg of bovine serum albumin/ml) for 15 min. PMN were added to the endothelial monolayer at a ratio of 5:1 and were incubated for 15 min at 37°C. PMN fluorescence readings were obtained using the Titertek Fluoroscan II (Titertek, Huntsville, Ala.). The plates were washed two or three times to remove nonadherent PMN, and PMN adhesion to endothelial cells was quantified by the ratio of final reading to initial reading.

RESULTS

Thrombin activates PKC-δ in endothelial cells.

We determined the phosphorylation and/or activation of PKC-δ following thrombin stimulation of HUVEC. We used phospho-PKC-δ (Thr505) antibody to determine the phosphorylation status of PKC-δ. Western blot analysis showed that thrombin induced the phosphorylation of PKC-δ in a time-dependent manner. The phosphorylated form of PKC-δ was detected as early as 1 min, and the peak phosphorylation occurred 5 min after thrombin challenge (Fig. 1A). Phosphorylation of PKC-δ declined after 20 min (Fig. 1A). We also determined whether phosphorylation resulted in increased kinase activity of PKC-δ. In an in vitro kinase assay in which histone H1 was used as a substrate, we found that PKC-δ immunoprecipitates from thrombin-treated cells showed increased phosphorylation of histone H1 compared to PKC-δ immunoprecipitates from control cells (Fig. 1B), suggesting the activation of PKC-δ by thrombin. Pretreatment of cells with rottlerin, an inhibitor of PKC-δ (50% inhibitory concentration = 3 to 6 μM) (20), prevented thrombin-induced PKC-δ activity in a dose-dependent manner (Fig. 1C). In control experiments, the PKC-β inhibitor LY379196 (50% inhibitory concentration = 5 nM) (54) failed to prevent thrombin-induced PKC-δ activity (Fig. 1C). These data demonstrate that the effect of rottlerin on PKC-δ activity is quite specific and are consistent with findings from previous studies (6, 10, 20).

FIG. 1.

(A) Thrombin induces phosphorylation of PKC-δ. Confluent HUVEC monolayers were challenged with thrombin (2.5 U/ml) for the indicated time periods. Total cell lysates (10 μg/lane) were separated by SDS-PAGE and immunoblotted with an antibody against the phosphorylated (Thr505) form of PKC-δ. The blots were subsequently stripped and reprobed with an antibody against PKC-δ. (B and C) Thrombin induces PKC-δ activity. Confluent HUVEC monolayers were pretreated without (B) or with (C) rottlerin (5 and 10 μM) and LY379196 (10 nM) for 30 min prior to challenge with thrombin (2.5 U/ml) for 5 min. −, absence of rottlerin or LY379196; +, presence of thrombin. Cell lysates were immunoprecipitated with an antibody against PKC-δ, and in vitro kinase assays were carried out on immunoprecipitates using histone H1 as an exogenous substrate. Proteins were analyzed by SDS-PAGE, and a phosphorylated form of histone H1 was detected by autoradiography.

Inhibition of PKC-δ reduces thrombin-induced ICAM-1 mRNA expression.

We used general and isoform-specific inhibitors to determine the involvement of PKC-δ in mediating thrombin-induced ICAM-1 gene transcription. Northern blot analysis showed that pretreatment of HUVEC monolayers with calphostin C, a broad-spectrum PKC inhibitor (28), or staurosporine, an inhibitor of cPKC and nPKC isoforms but not of aPKC isoforms (36, 52), prevented thrombin-induced ICAM-1 mRNA expression (Fig. 2A and B). We also found that depletion of cPKC and nPKC by prolonged exposure of HUVEC to phorbol esters (500 nM) for 24 h (47) prevented ICAM-1 mRNA expression induced by thrombin or by PAR-1-activating peptide (Fig. 2C). In control experiments, depletion of cPKC and nPKC isoforms prevented ICAM-1 mRNA expression in response to subsequent stimulation of cells with phorbol ester (100 nM; 3 h) (Fig. 2C). Thus, these data indicate a role for cPKC and nPKC isoforms but not for aPKC isoforms in the mechanism of thrombin-induced ICAM-1 gene transcription.

FIG. 2.

(A and B) Inhibitors of PKC prevent thrombin-induced ICAM-1 mRNA expression. Confluent HUVEC monolayers were pretreated with calphostin C (A) or with staurosporine (B) prior to challenge with thrombin for 3 h. Total RNA was isolated and analyzed by Northern hybridization with a human ICAM-1 cDNA, which hybridizes to a 3.3-kb transcript. Blots were stripped and reprobed to determine GAPDH mRNA expression as a measure of RNA loading. DMSO, dimethyl sulfoxide. (C) Phorbol ester-induced depletion of cPKC and nPKC isoforms prevents thrombin-induced ICAM-1 mRNA expression. Confluent HUVEC monolayers were treated without (−) or with (+) PMA (500 nM in 10% FBS–EBM2) for 24 h followed by stimulation with thrombin (2.5 U/ml), PAR-1-activating peptide (TRAP; 25 μM), or PMA (100 nM) for 3 h. ICAM-1 and GAPDH mRNA expression was determined by Northern blotting as described in Materials and Methods.

We used an antisense oligonucleotide that specifically inhibits the synthesis of PKC-ζ (47), an abundantly expressed aPKC isoform in endothelial cells, to address the involvement of this PKC isoform in thrombin response. Results showed that the antisense oligonucleotide to PKC-ζ failed to inhibit thrombin-induced ICAM-1 mRNA expression (Fig. 3).

FIG. 3.

Inhibition of PKC-ζ fails to prevent thrombin-induced ICAM-1 mRNA expression. HUVEC were transfected with sense (S) or antisense (AS) oligonucleotide to PKC-ζ as described in Materials and Methods. After 36 to 48 h, cells were stimulated for 3 h with thrombin (2.5 U/ml). ICAM-1 and GAPDH mRNA expression was determined by Northern blotting as described in Materials and Methods. (A), autoradiogram; (B), bar graph showing the relative intensities of ICAM-1 mRNA signals.

We used rottlerin to evaluate the role of the PKC-δ isoform in mediating ICAM-1 mRNA expression following thrombin challenge of HUVEC. Figure 4A shows that rottlerin inhibited thrombin-induced ICAM-1 mRNA expression in a dose-dependent manner. In contrast, inhibition of PKC-β by LY379196 failed to prevent the thrombin response (Fig. 4B). These data indicate that activation of PKC-δ is required, at least in part, in signaling thrombin-induced ICAM-1 transcription.

FIG. 4.

Inhibition of PKC-δ reduces thrombin-induced ICAM-1 mRNA expression. Confluent HUVEC monolayers were pretreated with rottlerin (A) or with LY379196 (B) prior to challenge with thrombin for 3 h. ICAM-1 and GAPDH mRNA expression was determined by Northern blotting as described in Materials and Methods. (A and B) Autoradiograms. Bottom of panel A contains a bar graph showing the effects of rottlerin on relative intensities of ICAM-1 mRNA signals.

Inhibition of PKC-δ prevents thrombin-induced ICAM-1 protein expression and endothelial adhesivity towards PMN.

We next determined the effects of inhibition of PKC-δ on thrombin-induced ICAM-1 protein expression and resultant endothelial adhesivity towards PMN. Western blot analysis showed that stimulation of HUVEC with thrombin resulted in increased ICAM-1 protein expression (Fig. 5A). Preincubation of cells with rottlerin inhibited thrombin-induced ICAM-1 protein expression (Fig. 5A), consistent with its effect on ICAM-1 mRNA expression (Fig. 4A). In a control experiment, we showed that a cell-permeable specific peptide antagonist of PKC-θ failed to prevent thrombin-induced ICAM-1 protein expression (Fig. 5B), suggesting that PKC-θ is not involved in thrombin-induced ICAM-1 expression.

FIG. 5.

(A and B) Inhibition of PKC-δ prevents thrombin-induced ICAM-1 protein expression. Confluent HUVEC monolayers were pretreated with rottlerin or LY379196 (A) and peptide antagonist of PKC-θ (B) at the indicated concentrations prior to challenge with thrombin for 8 h. Expression of ICAM-1 protein was determined by Western blotting as described in Materials and Methods. The blots were subsequently stripped and reprobed with an antibody against PKC-δ or IκBβ to indicate equal loading of the gel. (C) Inhibition of PKC-δ prevents thrombin-induced endothelial adhesivity towards PMN. Confluent HUVEC monolayers were pretreated with rottlerin or LY379196 at the indicated concentrations prior to challenge with thrombin. Expression of endothelial adhesivity was determined by PMN adhesion assays as described in Materials and Methods.

As ICAM-1 expression results in endothelial adhesivity towards PMN (45), we determined whether inhibition of ICAM-1 protein expression would lead to inhibition of endothelial adhesivity. We found that rottlerin prevented thrombin-induced endothelial adhesivity towards PMN in a dose-dependent manner (Fig. 5C). In control experiments, we determined the effects of LY379196 on thrombin-induced ICAM-1 protein expression and on resultant endothelial adhesivity to compare the effects with those of rottlerin. Pretreatment of cells with LY379196 failed to prevent ICAM-1 protein expression and endothelial adhesivity in response to thrombin challenge (Fig. 5A and C).

Inhibition of PKC-δ reduces thrombin-induced NF-κB activity.

As NF-κB activation is essential for thrombin-induced ICAM-1 gene transcription (45), we addressed the role of PKC-δ in mediating the transcriptional activity of NF-κB. HUVEC were cotransfected with pNF-κB-LUC containing five copies of consensus NF-κB sequence linked to a minimal adenovirus E1B promoter-luciferase reporter gene in combination with constructs encoding kinase-defective PKC-δ (PKC-δ^mut), PKC-α (PKC-α^mut), or PKC-ɛ (PKC-ɛ^mut). As shown in Fig. 6, coexpression of PKC-δ^mut reduced thrombin-induced NF-κB activity, whereas PKC-α^mut had no effect on the response. Expression of PKC-ɛ^mut also reduced thrombin-induced NF-κB activity, albeit to a lesser extent (Fig. 6). These data indicate the involvement of PKC-δ and to a lesser extent, of PKC-ɛ, in the mechanism of thrombin-induced NF-κB activity.

FIG. 6.

Inhibition of NF-κB activity by expression of a kinase-defective mutant of PKC-δ. HUVEC were cotransfected with plasmid pNF-κB-LUC and the constructs encoding kinase-defective mutants of PKC-δ (PKC-δ^mut), -ɛ (PKC-ɛ^mut), or -α (PKC-α^mut) isoform using the DEAE-dextran method as described previously (47). In some experiments pcDNA3 alone was used as the vector control. Cells were stimulated for 8 h with thrombin (2.5 U/ml) before being harvested. Cytoplasmic extracts were prepared, and luciferase activity was determined. Firefly luciferase activity normalized to Renilla luciferase activity is expressed in RLU per microgram of protein. Data are mean ± standard error (n = 3 for each condition).

PKC-δ signals thrombin-induced NF-κB activation via IKKβ.

We evaluated the function of PKC-δ in mediating thrombin-induced IκBα degradation, a requirement for NF-κB activation (5, 9, 62). As IκBα degradation requires its phosphorylation by IKKβ, we determined the involvement of IKKβ in thrombin-induced NF-κB activation. HUVEC were cotransfected with pNF-κB-LUC in combination with the kinase-defective IKKβ mutant (IKKβ^mut). Coexpression of IKKβ^mut prevented thrombin-induced NF-κB activity (Fig. 7A), indicating the requirement of IKKβ in the response. We also addressed the function of IKKβ in PKC-δ-mediated NF-κB activation. Expression of constitutively active PKC-δ mutant (PKC-δ^CAT) induced NF-κB activity in the absence of thrombin challenge (Fig. 7B). Coexpression of IKKβ^mut prevented PKC-δ^CAT-induced NF-κB activity (Fig. 7B), indicating that PKC-δ mediates thrombin-induced NF-κB activity via activation of IKKβ.

FIG. 7.

(A) Involvement of IKKβ in thrombin-induced NF-κB activity. HUVEC were cotransfected with pNF-κB-LUC and a construct encoding a kinase-defective mutant of IKKβ (IKKβ^mut) using the DEAE-dextran method as previously described (47). In some experiments pcDNA3 alone was used as the vector control. Cells were stimulated for 8 h with thrombin (2.5 U/ml) before being harvested. Cytoplasmic extracts were prepared, and luciferase activity was determined. Firefly luciferase activity normalized to Renilla luciferase activity was expressed in RLU per microgram of protein. Data are mean ± standard error (n = 3 for each condition). (B) Inhibition of IKKβ prevents PKC-δ-mediated NF-κB activity. HUVEC were cotransfected with pNF-κB-LUC in combination with the constructs encoding a kinase-defective mutant of IKKβ (IKKβ^mut) and a constitutively active PKC-δ mutant (PKCδ^CAT) using Superfect as described previously (45). In some experiments pcDNA3 alone was used as the vector control. Twenty-four hours later, cytoplasmic extracts were prepared and luciferase activity was determined. Firefly luciferase activity normalized to Renilla luciferase activity was expressed in RLU per microgram of protein. Data are mean ± standard error (n = 3 for each condition). +, presence of pcDNA3 or IKKβ^mut; −, absence of either plasmid.

Figure 8A shows that thrombin challenge of endothelial cells resulted in IκBα degradation; however, thrombin failed to induce IκBβ degradation. Inhibition of PKC-δ by calphostin C or staurosporine prevented thrombin-induced IκBα degradation (Fig. 8A), consistent with the involvement of IKKβ in PKC-δ-mediated NF-κB activation in endothelial cells. We next determined whether the effects of PKC-δ inhibition on thrombin-induced IκBα degradation correlated with the DNA binding function of NF-κB. Pretreatment of cells with rottlerin inhibited thrombin-induced NF-κB DNA binding (Fig. 8B). Similarly, pretreatment of cells with calphostin C and staurosporine also inhibited NF-κB binding to the ICAM-1 promoter (data not shown).

FIG. 8.

Inhibition of PKC-δ prevents thrombin-induced IκBα degradation and NF-κB DNA binding activity. Confluent HUVEC monolayers were pretreated with calphostin C and staurosporine (A) or rottlerin (B) at the indicated concentrations prior to challenge with thrombin for 1 h. Cytoplasmic (A) and nuclear (B) extracts were prepared and assayed for IκBα degradation by Western blot analysis (A) and for NF-κB DNA binding activity by EMSA (B) as described in Materials and Methods.

Thrombin induces p38 MAP kinase activation in endothelial cells.

Since thrombin can activate p38 MAP kinase (24, 50), which in turn has been shown to regulate NF-κB activity (7, 65), we determined the phosphorylation and/or activation of p38 MAP kinase in response to thrombin challenge of endothelial cells. Results showed that thrombin induced p38 MAP kinase (Thr180/Tyr182) phosphorylation within 1 min and that its level remained elevated up to 60 min after thrombin challenge (Fig. 9A).

FIG. 9.

(A) Thrombin induces phosphorylation of p38 MAP kinase. Confluent HUVEC monolayers were challenged with thrombin (2.5 U/ml) for the indicated periods. Total cell lysates (10 μg/lane) were separated by SDS-PAGE and immunoblotted with an antibody against a phosphorylated (Thr180/Tyr182) form of p38 MAP kinase. The blots were subsequently stripped and reprobed with an antibody against p38 MAP kinase. (B) Thrombin induces MAPKAP kinase 2 activity. Confluent HUVEC monolayers were pretreated for 30 min with SB203580 prior to challenge with thrombin for 5 min. Cell lysates were immunoprecipitated with an antibody against MAPKAP kinase 2, and in vitro kinase assays were carried out on immunoprecipitates using Hsp-25 as an exogenous substrate. Proteins were analyzed by SDS-PAGE, and the phosphorylated form of Hsp-25 was detected by autoradiography.

As MAPKAP kinase 2 is an in vivo substrate for p38 MAP kinase (18, 51), we determined if it was capable of activating this downstream effector of p38 MAP kinase. Figure 9B demonstrates that thrombin induced MAPKAP kinase 2 activity, as evident by the phosphorylation of Hsp-25 used as an exogenous substrate. Pretreatment of cells with the specific p38 MAP kinase inhibitor SB203580 prevented thrombin-induced MAPKAP kinase 2 activity (Fig. 9B), indicating the activation of p38 MAP kinase by thrombin.

Inhibition of PKC-δ prevents thrombin-induced p38 MAP kinase activation.

We next determined whether PKC-δ activation was required for p38 MAP kinase activation by pretreating confluent HUVEC monolayers with rottlerin, followed by stimulation with thrombin. Control or rottlerin-treated cells showed no phosphorylation of p38 MAP kinase (Fig. 10). In contrast, cells treated with thrombin for 5 min markedly increased the phosphorylation of p38 MAP kinase (Fig. 10). Preincubation of cells with rottlerin (5 μM) prevented thrombin-induced phosphorylation of p38 MAP kinase (Fig. 10). In another experiment, we found that inhibition of p38 MAP kinase by SB203580 failed to prevent thrombin-induced phosphorylation of PKC-δ (data not shown). These data demonstrate that p38 MAP kinase signals downstream of PKC-δ following thrombin challenge of endothelial cells.

FIG. 10.

Inhibition of PKC-δ prevents thrombin-induced phosphorylation of p38 MAP kinase. Confluent HUVEC monolayers were pretreated for 30 min with rottlerin prior to challenge with thrombin for 5 min. Total cell lysates (10 μg/lane) were resolved in SDS-PAGE and were immunoblotted with an antibody against the phosphorylated form of p38 MAP kinase.

Inhibition of p38 MAP kinase reduces thrombin-induced ICAM-1 mRNA expression.

We addressed the role of p38 MAP kinase activation in mediating thrombin-induced ICAM-1 gene transcription. Inhibition of p38 MAP kinase activity by SB203580 resulted in a significant decrease in ICAM-1 mRNA expression in response to thrombin challenge (Fig. 11A). We also determined the involvement of extracellular signal-regulated kinase (ERK1/2) in the thrombin response. Pretreatment of cells with PD98059, an inhibitor of MEK, the upstream kinase of ERK1/2, failed to prevent thrombin-induced ICAM-1 mRNA expression (Fig. 11B). These data show a critical role for p38 MAP kinase in signaling thrombin-induced ICAM-1 transcription in endothelial cells.

FIG. 11.

Inhibition of p38 MAP kinase reduces thrombin-induced ICAM-1 mRNA expression. Confluent HUVEC monolayers were pretreated with SB203580 (10 μM) (A) or PD98059 (50 μM) (B) prior to challenge with thrombin for 3 h. ICAM-1 and GAPDH mRNA expression was determined by Northern blotting as described in Materials and Methods.

Inhibition of p38 MAP kinase reduces thrombin-induced NF-κB activity without affecting DNA binding.

We addressed the function of p38 MAP kinase in the mechanism of thrombin-induced NF-κB activity. HUVEC were cotransfected with pNF-κB-LUC and a construct encoding kinase-defective p38 MAP kinase (p38^mut). We found that expression of p38^mut inhibited thrombin-induced NF-κB activity (Fig. 12). Since p38 MAP kinase signals downstream of PKC-δ, we also studied the role of p38 MAP kinase in thrombin-induced IκBα degradation. However, in contrast to the effects of PKC-δ inhibition, pretreatment of cells with SB203580 failed to prevent thrombin-induced IκBα degradation (Fig. 13A).

FIG. 12.

Inhibition of NF-κB activity by expression of the dominant negative mutant of p38 MAP kinase. HUVEC were cotransfected with pNF-κB-LUC and a construct encoding the dominant negative mutant of p38 MAP kinase (p38^mut) using the DEAE-dextran method as described previously (47). Cells were stimulated for 6 h with thrombin (2.5 U/ml) before being harvested. Cytoplasmic extracts were prepared, and luciferase activity was determined. Firefly luciferase activity normalized to Renilla luciferase activity was expressed in RLU per microgram of protein. Data are mean ± standard error (n = 3 for each condition).

FIG. 13.

Inhibition of p38 MAP kinase fails to prevent thrombin-induced IκBα degradation, nuclear translocation, and NF-κB DNA binding activity. Confluent HUVEC monolayers were pretreated for 30 min with SB203580 prior to challenge with thrombin for 1 h. Cytoplasmic (A) and nuclear (B and C) extracts were prepared and assayed for IκBα degradation (A) and NF-κB nuclear translocation (B) by Western blot analysis and for NF-κB DNA binding activity (C) by EMSA as described in Materials and Methods.

As IκBα degradation results in nuclear translocation and DNA binding of NF-κB, we evaluated the effects of inhibition of p38 MAP kinase on thrombin-induced nuclear uptake and subsequently DNA binding function of nuclear NF-κB. Pretreatment of cells with SB203580 failed to prevent nuclear translocation of NF-κB p65 and the resultant DNA binding (Fig. 13B and C). These data indicate that p38 MAP kinase contributes to the thrombin response by increasing the transcriptional activity of NF-κB. The findings are consistent with a previously described role for p38 MAP kinase in phosphorylating and thereby inducing the transactivation potential of NF-κB p65 without affecting its DNA binding activity (65).

DISCUSSION

PKC isoforms represent a family of serine/threonine kinases with different cofactors and substrate specificities (38, 41). Studies have shown that PKC-α, -β, -δ, -ɛ, -θ, and -ζ isoforms are expressed in endothelial cells (25, 47, 53). In the present study, we have demonstrated a novel role for PKC-δ in the mechanism of thrombin-induced NF-κB activity and ICAM-1 gene transcription in endothelial cells. Our data establish that thrombin induces PKC-δ activation in endothelial cells, which in turn activates IKKβ and p38 MAP kinase. Activation of IKKβ contributes to thrombin-induced ICAM-1 gene transcription by promoting IκBα degradation and thereby promoting NF-κB binding to the ICAM-1 promoter, whereas induction of p38 MAP kinase activity contributes to the thrombin response by increasing the transactivation potential of bound NF-κB.

We used pharmacological and genetic approaches to address the specificity of function of the PKC-δ isoform in mediating thrombin-induced NF-κB activity and ICAM-1 gene transcription. Calphostin C, a relatively broad-spectrum PKC inhibitor (28), or staurosporine, which inhibits both cPKC and nPKC but not aPKC isoforms (36, 52), each prevented thrombin-induced NF-κB activation and ICAM-1 mRNA expression. To rule out the role of the atypical PKC-ζ isoform in the mechanism for the thrombin response, we depleted endothelial cells of cPKC and nPKC isoforms but not of aPKC isoforms by prolonged exposure of the cells to phorbol esters (47). We also addressed the effects of inhibition of PKC-ζ synthesis by specific antisense oligonucleotides (16, 47). We found that depletion of cPKC and nPKC isoforms prevented thrombin-induced ICAM-1 gene transcription in endothelial cells. In contrast, antisense oligonucleotide to PKC-ζ failed to prevent thrombin-induced ICAM-1 mRNA expression. These results suggest the involvement of cPKC and nPKC isoforms but exclude the participation of the aPKC isoform PKC-ζ in signaling ICAM-1 gene transcription induced by thrombin. We next used LY379196 and the kinase-defective PKC-α mutant to inhibit PKC-β and -α isoforms, respectively. Inhibition of these cPKC isoforms failed to prevent NF-κB activity and ICAM-1 transcription, suggesting that these isoforms are not important in the mechanism of the thrombin response. However, the present results do not exclude the possibility that PKC-α, PKC-β, or PKC-ζ plays a role in activating NF-κB in response to other agonists. Indeed, studies have shown that PKC-α, PKC-β, and PKC-ζ can contribute to NF-κB activation in response to lipopolysaccharide and tumor necrosis factor alpha in a variety of cell types, including endothelial cells (8, 17, 27, 47).

We employed multiple approaches to establish the role of PKC-δ in the mechanism of thrombin-induced NF-κB activity and ICAM-1 gene transcription in endothelial cells. Expression of a kinase-defective form of PKC-δ (PKC-δ^mut) or pretreatment of cells with rottlerin, a PKC-δ activation inhibitor (20), reduced thrombin-induced NF-κB activity and ICAM-1 gene transcription. We also found that coexpression of constitutively active PKC-δ mutant (PKC-δ^CAT) induced NF-κB activity in the absence of thrombin challenge. The partial inhibition by rottlerin and the kinase-defective PKC-δ mutant suggests the possible involvement of other nPKC isoforms, such as PKC-ɛ and -θ, in signaling the response. Our finding that pretreatment of cells with calphostin C, staurosporine, and PMA abrogated the thrombin response on ICAM-1 mRNA expression lends support to this contention, especially since PKC-ɛ and -θ can activate NF-κB (32, 60, 61). However, it is unlikely that PKC-θ was involved in the thrombin response in the present study, since inhibition of PKC-θ by a specific peptide antagonist failed to prevent thrombin-induced ICAM-1 expression. We also found that expression of constitutively active PKC-θ failed to induce NF-κB activity (data not shown). In contrast, expression of the kinase-defective PKC-ɛ mutant reduced thrombin-induced NF-κB activity, suggesting that PKC-ɛ can contribute to the response.

The finding that PKC-δ was critical in the activation of ICAM-1 transcription led us to address its role in mediating thrombin-induced expression of ICAM-1 protein and the resultant endothelial adhesivity towards PMN. We showed by using rottlerin that inhibition of PKC-δ prevented both ICAM-1 protein expression and PMN adhesion to endothelial cells. Interestingly, ICAM-1 protein expression was inhibited to a greater extent than was mRNA expression, raising the possibility that PKC-δ may also regulate ICAM-1 expression at the posttranscriptional and translational levels. This is consistent with the recent evidence that PKC-δ-induced mRNA stabilization is a key posttranscriptional mechanism by which interleukin 1β regulates the expression of inducible nitric oxide synthase (6).

Clues to the mechanisms by which PKC-δ induced ICAM-1 gene transcription were provided by the observations that thrombin increased p38 MAP kinase activity in endothelial cells and that rottlerin prevented this effect. Inhibition of p38 MAP kinase activity, however, failed to prevent PKC-δ phosphorylation induced by thrombin (data not shown). These data indicate that p38 MAP kinase signals downstream of PKC-δ in response to thrombin challenge of endothelial cells. Although the possibility that PKC can activate p38 MAP kinase has been explored in a variety of cells, the relationship between these kinases in response to proinflammatory cytokines remains uncertain (12, 40, 44, 59). The present data are consistent with other studies (23, 26, 42) showing that PKC can activate the p38 MAP kinase cascade in vascular endothelial cells. We demonstrated that inhibition of p38 MAP kinase activity by expression of the kinase-inactive mutant of p38 MAP kinase and by pretreatment with SB203580 reduced thrombin-induced NF-κB activity and ICAM-1 mRNA expression, respectively. Thus, these data implicate the role of p38 MAP kinase signaling downstream of PKC-δ in the mechanism of thrombin-induced NF-κB activity and ICAM-1 expression in endothelial cells. In contrast, we found that inhibition of MEK by PD98059 failed to prevent ICAM-1 transcription, suggesting that ERK1/2 is not required for thrombin-induced ICAM-1 transcription.

We also determined the mechanisms by which PKC-δ and its downstream effector, p38 MAP kinase, contributed to NF-κB activity and ICAM-1 expression. Studies have shown that NF-κB activity is primarily regulated at the level of IκB degradation, which is accomplished through serine phosphorylation of IκBα (Ser32 and Ser36) by IκBβ kinase (IKKβ) (13, 31, 68). We observed that expression of a kinase-defective mutant of IKKβ (IKKβ^mut) prevented NF-κB activity induced by thrombin as well as by the expression of PKC-δ^CAT. Consistent with these data, we showed that PKC-δ-mediated IκBα degradation resulted in the migration of NF-κB to the nucleus, where its binding to the promoter activated ICAM-1 gene transcription.

Evidence suggests the existence of an additional regulatory pathway that can be activated in parallel to the cascade inducing IκBα degradation and that can thus control the transactivation potential of NF-κB by targeting the p65 subunit. A variety of kinases, including p38 MAP kinase, may contribute to the transactivation potential of NF-κB by phosphorylating the p65 subunit of NF-κB (1, 4, 34, 65). Another mechanism by which p38 MAP kinase can regulate NF-κB activity may involve phosphorylation of the TATA binding protein (TBP), one of the subunits of transcription factor IID. Phosphorylation of TBP by p38 MAP kinase is necessary for TBP binding to the TATA box (7). Inhibition of phosphorylation of TBP reduced its binding to the TATA box and its interaction with the NF-κB p65 subunit (7). In the present study, we showed that p38 MAP kinase exerts its effect in mediating the thrombin-induced NF-κB activity downstream of IκBα degradation; therefore, it is possible that p38 MAP kinase contributes to thrombin-induced ICAM-1 transcription by phosphorylating NF-κB p65 and TBP.

In summary, the present study implicates PKC-δ as a critical kinase that signals ICAM-1 gene transcription by inducing IKKβ activation to promote NF-κB binding to the ICAM-1 promoter and activating p38 MAP kinase to increase the transcriptional activity of NF-κB (Fig. 14). Thus, the thrombin-induced expression of ICAM-1 and endothelial adhesivity may be regulated by PKC-δ through a dual mechanism involving the activation of NF-κB and the p38 MAP kinase-induced phosphorylation of the NF-κB p65 subunit and of TBP. These two mechanisms may operate in a synergistic fashion to sustain ICAM-1 expression in endothelial cells and thereby promote stable endothelial adhesivity.

FIG. 14.

Signaling events regulating thrombin-induced NF-κB activation and ICAM-1 transcription in endothelial cells. Thrombin challenge of endothelial cells results in PKC-δ activation, which in turn activates IKKβ and p38 MAP kinase. Activation of IKKβ contributes to thrombin-induced ICAM-1 gene transcription by promoting IκBα degradation and subsequently NF-κB binding to the ICAM-1 promoter. Activation of p38 MAP kinase contributes to ICAM-1 transcription, possibly by increasing the transactivation potential of bound NF-κB p65 through its phosphorylation. Alternatively, p38 MAP kinase may contribute to the response by promoting the interaction of NF-κB with the basal transcription machinery through phosphorylation of TBP.