Protein Kinase C-δ (PKCδ) Regulates Proinflammatory Chemokine Expression through Cytosolic Interaction with the NF-κB Subunit p65 in Vascular Smooth Muscle Cells*

Background: Proinflammatory chemokines released by vascular smooth muscle cells (VSMCs) play a critical role in vascular inflammation. Results: Promoting protein kinase C-δ (PKCδ) translocation or inhibition of NF-κB pathway diminishes proinflammatory chemokine production. Conclusion: PKCδ regulates proinflammatory chemokine expression through cytosolic interaction with the NF-κB subunit p65, thus modulating inflammation. Significance: Learning how PKCδ regulates proinflammatory chemokine expression is crucial for understanding vascular inflammation. Proinflammatory chemokines released by vascular smooth muscle cells (VSMCs) play a critical role in vascular inflammation. Protein kinase C-δ (PKCδ) has been shown to be up-regulated in VSMCs of injured arteries. PKCδ knock-out (Prkcd−/−) mice are resistant to inflammation as well as apoptosis in models of abdominal aortic aneurysm. However, the precise mechanism by which PKCδ modulates inflammation remains incompletely understood. In this study, we identified four inflammatory chemokines (Ccl2/Mcp-1, Ccl7, Cxcl16, and Cx3cl1) of over 45 PKCδ-regulated genes associated with inflammatory response by microarray analysis. Using CCL2 as a prototype, we demonstrated that PKCδ stimulated chemokine expression at the transcriptional level. Inhibition of the NF-κB pathway or siRNA knockdown of subunit p65, but not p50, eliminated the effect of PKCδ on Ccl2 expression. Overexpressing PKCδ followed by incubation with phorbol 12-myristate 13-acetate resulted in an increase in p65 Ser-536 phosphorylation and enhanced DNA binding affinity without affecting IκB degradation or p65 nuclear translocation. Prkcd gene deficiency impaired p65 Ser-536 phosphorylation and DNA binding affinity in response to TNFα. Results from in situ proximity ligation analysis and co-immunoprecipitation performed on cultured VSMCs and aneurysmal aorta demonstrated physical interaction between PKCδ and p65 that took place largely outside the nucleus. Promoting nuclear translocation of PKCδ with peptide ψδRACK diminished Ccl2 production, whereas inhibition of PKCδ translocation with peptide δV1-1 enhanced Ccl2 expression. Together, these results suggest that PKCδ modulates inflammation at least in part through the NF-κB-mediated chemokines. Mechanistically, PKCδ activates NF-κB through an IκB-independent cytosolic interaction, which subsequently leads to enhanced p65 phosphorylation and DNA binding affinity.

Vascular inflammation is a complex biological response trigged by chemical and mechanical injuries as well as by infectious stimuli. Inflammation is observed to various degrees in major cardiovascular diseases including atherosclerosis, myocardial infarction, congestive heart failure, and aortic aneurysm (1)(2)(3). A critical step of vascular inflammation is the recruitment of circulating leukocytes including monocytes and T lymphocytes into the vascular wall. The recruitment process is primarily the result of coordinated expression of vascular adhesion molecules as well as proinflammatory chemokines and cytokines (3). As a major component of the arterial wall, vascular smooth muscle cells (VSMCs) 2 are critical in maintaining normal physiological functions of blood vessels as well as in modulation of pathological processes taking place in the vascular wall (4). Numerous studies have shown that VSMCs can be an important source of cytokines in the vessel wall (5)(6)(7).
Protein kinase C-␦ (PKC␦), a member of the novel PKC isoforms of serine-threonine kinase, is expressed in many types of mammalian cells including cancer cells, leukocytes, and VSMCs (8). Like other members of the PKC family, PKC␦ exists normally in an inactive conformation and becomes activated upon binding to diacylglycerol or its mimetic phorbol ester such as phorbol 12-myristate 13-acetate (PMA) or through other molecular mechanisms such as proteolytic cleavage and/or tyrosine phosphorylation (9). PKC␦ has emerged as an important regulator of apoptosis. The proapoptotic function of PKC␦ is achieved by its interaction with and phosphorylation of several key apoptotic regulators (10). For example, PKC␦ associates with and phosphorylates caspase-3, promoting the apoptotic activity of the cysteine caspase during etoposide-induced apoptosis and in spontaneous apoptosis of monocytes (10). Diverse apoptosis-inducing agents including Fas ligation, etoposide, mitomycin, cytosine arabinoside, etoposide, UV light, and ionizing radiation are found to induce cleavage of PKC␦ in the linker region, freeing the catalytic domain from the regulatory domain (11,12). The free catalytic domain or fragment is thought to translocate to the nucleus and execute apoptosis (13). Consistent with this hypothesis, overexpression of the catalytic fragment in the absence of an apoptotic stimulus was sufficient to induce apoptosis in a variety of cell types (10). In contrast, several lines of studies showed a pivotal role for PKC␦ in antiapoptotic function in response to cytokines including tumor necrosis factor-␣ (TNF␣) (14). Silencing PKC␦ expression by siRNA resulted in inhibition of TNF-mediated extracellular signal-regulated kinase 1/2 activation (15). Another study showed that PKC␦ interacts with Smac, a mitochondrial protein. On exposure to apoptotic stimuli such as paclitaxel, this interaction is disrupted and results in the release of Smac into the cytosol and promotes apoptosis by activating caspases in the cytochrome c/Apaf-1/ caspase-9 pathway. Activation of PKC␦ rescues the interaction during paclitaxel exposure and suppresses paclitaxel-mediated cell death (16). What dictates whether PKC␦ should exert a proor antiapoptotic role in a given cell is unclear.
We have shown previously that PKC␦ plays a proapoptotic role during the vascular injury response (17). PKC␦ expression is up-regulated in human aneurysmal aortic tissues and restenotic lesions as well as in animal models of vascular injury such as a mouse abdominal aortic aneurysm model as well as a rat carotid angioplasty model (18,19). In mouse models of abdominal aortic aneurysm, Prkcd Ϫ/Ϫ mice are resistant to apoptosis, which is coupled with diminished proinflammatory factor expression and inflammatory cell infiltration (18). Although this mouse study indicates a role for PKC␦ in modulating inflammatory signaling, the underlying mechanisms remain obscure.
NF-B is one of the major transcription factors that control the expression of inflammatory factors (20). Activation of NF-B is regulated by multiple distinct signaling cascades including the IB kinase signalosome (21). In response to a variety of stimuli, IB kinase phosphorylates IB␣ at Ser-32 and Ser-36, resulting in its ubiquitination and subsequent proteasomal degradation. The released NF-B is targeted to the nucleus and binds to a B site located in the promoter region, thereby inducing the expression of specific target genes (22). In addition to nuclear translocation of the NF-B complex, previous studies have shown that a subunit of NF-B, RelA/p65, is post-translationally modified by phosphorylation or acetylation, and those changes influence its DNA binding and transcriptional activity (23,24). Although the effect of PKC␦ on the NF-B pathway has been reported, the precise mechanisms underlying PKC␦-mediated NF-B activation remain largely unclear.
In this study, we tested the hypothesis that PKC␦ activation is a pivotal signal for proinflammatory chemokine expression in VSMCs. Through both in vitro and in vivo experiments, we systematically evaluated the role of PKC␦ in the regulation of chemokine expression. Using chemokine (CC motif) ligand 2 (CCL2) as a chemokine prototype, we further defined the molecular mechanism of PKC␦ regulation. Our data suggest that PKC␦ acts through the RelA/p65 subunit of NF-B in the cytosolic fraction of VSMCs that subsequently activates Ccl2 transcription.
Cell Culture-Primary VSMCs were isolated from arteries of mice or rats according to a method described previously (26). Cells were maintained in DMEM supplemented with 10% FBS, 100 units/ml penicillin, and 100 g/ml streptomycin in a 5% CO 2 , water-saturated incubator at 37°C. Adenoviral Vectors and Infection-Adenoviral vectors expressing PKC␦ (AdPKC␦) and empty vector (AdNull) were constructed and purified by CsCl gradient centrifugation as described previously (27,28). In vitro adenovirus infection was carried out as described previously (19). Briefly, VSMCs (1 ϫ 10 5 cells/well in 6-well plates) were infected with adenovirus (m.o.i. ϭ 1 ϫ 10 4 ) in DMEM containing 2% FBS overnight at 37°C followed by starvation in DMEM containing 0.5% FBS for 24 h. The cells were then treated with PMA (1 nM) or solvent (DMSO) for the indicated periods of time. Cells were harvested and used for mRNA extraction and analysis, and medium was collected and tested by ELISA.
Microarray and Biological Functional Analyses-Following PKC␦ activation by PMA, total RNA was isolated utilizing the RNeasy Plus Mini kit (Qiagen, Valencia, CA) according to the manufacturer's protocol. Genomic DNA was removed using the provided gDNA Eliminator columns. Microarray hybridization was carried out by the microarray core facility at the University of Wisconsin-Madison Biotech Center. Briefly, total RNA was quantified on a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE), and RNA quality was analyzed on an Agilent RNA Nano Chip (Agilent Technologies, Palo Alto, CA). Samples were labeled using an Ambion GeneChip WT Expression kit (Invitrogen), and labeled cRNA was fragmented and hybridized to the GeneChip Gene 1.0 ST Array (Affymetrix, Santa Clara, CA) according to the manufacturer's protocol. The arrays were washed and stained using a GeneChip Fluidic Station 450 and scanned using an Affymetrix GeneChip Scanner 3000. Data were extracted and processed using Affymetrix Command Console version 3.1. After correction and normalization of background using the Robust Multichip Array algorithm, differentially expressed genes were identified with a moderated t test implemented in ArrayStar software (DNAStar, Madison, WI). The list of differentially expressed genes was loaded into Ingenuity Pathway Analysis 9.0 software to perform biological functional and transcription factor analyses.
Quantitative Real Time PCR (qPCR)-2 g of RNA was used for the first strand cDNA synthesis (Applied Biosystems, Carlsbad, CA). A no-reverse transcriptase control was included in the same PCR mixtures without reverse transcriptase to confirm the absence of DNA contamination in RNA samples. qPCR primers for CCL2, CCL7, CXCL16, CX3CL1, and GAPDH were purchased from Qiagen. Triplicate 20-l reactions were carried out in 96-well optical reaction plates using SYBR Green PCR Master Mix (Applied Biosystems) with gene-specific primers, and the qPCR was run in the 7500 Fast Real-Time PCR System (Applied Biosystems). Amplification of each sample was analyzed by melting curve analysis, and relative differences in each PCR sample were corrected using GAPDH mRNA as an endogenous control and normalized to the level of control by using the 2 Ϫ⌬⌬Ct method.
CCL2 ELISA-The BD OptEIA ELISA kit was purchased from BD Biosciences to measure CCL2 secreted by VSMCs according to the manufacturer's protocol.
Transfection-siRNAs to NF-B p50 and p65 were obtained from Invitrogen. siRNA to PKC␦ and its scrambled control were purchased from Qiagen. siRNA transfection was carried out as described previously (29). Briefly, VSMCs were plated onto 6-well plates in DMEM with 10% FBS. Cells were then transfected in Opti-MEM I medium with 10 nM siRNA for NF-B p50, NF-B p65, or control using Lipofectamine RNAiMAX transfection reagent as described by the manufacturer's protocol (Invitrogen). At 6 h post-transfection, Opti-MEM I medium was replaced with DMEM containing 2% FBS.
Immunoblotting-Cells were lysed in radioimmune precipitation assay buffer (Sigma-Aldrich), and total protein was extracted. Nuclear and cytoplasmic proteins were extracted using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific, Rockford, IL) according to the manufacturer's protocol. Equal amounts of protein extract were separated by SDS-PAGE and transferred to PVDF membranes. The membranes were then incubated with rabbit antibodies to phospho-NF-B p65 (Cell Signaling Technology, Danvers, MA), NF-B p65 (Cell Signaling Technology), PKC␦ (Santa Cruz Biotechnology, Dallas, TX), proliferating cell nuclear antigen (Santa Cruz Biotechnology), and tubulin (Cell Signaling Technology) and mouse antibodies to IB␣ (Cell Signaling Technology) and ␤-actin (Sigma-Aldrich) followed by horseradish peroxidase (HRP)-labeled goat anti-rabbit or anti-mouse immunoglobulin G (Bio-Rad). Labeled proteins were visualized with an enhanced chemiluminescence system (Thermo Fisher Scientific, Rockford, IL). For quantification, optical density of secreted proteins determined using NIH ImageJ (National Institutes of Health, Bethesda, MD) was normalized to the loading control density.
NF-B p65 DNA Binding Activity-An ELISA-based TransAM NF-B p65 assay kit (Active Motif, Carlsbad, CA) was used to measure the binding activity of NF-B to DNA in nuclear extracts according to the manufacturer's protocol. Briefly, 5 g of nuclear protein from each sample was added to different wells in which multiple copies of the consensus bind-ing site for NF-B had been immobilized and incubated for 1 h at room temperature for NF-B DNA binding. By using an antibody that is specific for NF-B p65, the activated NF-B subunit bound to the oligonucleotides can be detected. An HRPconjugated secondary antibody was added to provide a sensitive colorimetric readout. The signal was quantified by reading the absorbance at 450 nm on a FlexStation 3 Benchtop Multi-Mode Microplate Reader (Molecular Devices, Sunnyvale, CA).
In Situ Proximity Ligation Assay-An in situ proximity ligation assay (PLA) was performed to detect protein-protein interactions using a Duolink in situ fluorescence kit according to the manufacturer's protocol (Olink Bioscience, Uppsala, Sweden). Briefly, treated VSMCs were fixed with 4% paraformaldehyde at room temperature for 10 min followed by cell membrane permeabilization with 0.2% Triton X-100 in PBS for 10 min. Tissue sections were fixed for 10 min in cold acetone. The slides were washed three times with PBS, blocked for 1 h at room temperature with 5% BSA and normal donkey serum in PBS, and incubated with the indicated antibody pairs overnight at 4°C. Oligonucleotide-conjugated secondary antibodies (PLA probe MINUS and PLA probe PLUS) against each of the primary antibodies were applied, and ligation and amplification were carried out to produce rolling circle products. These products were detected with fluorescently labeled oligonucleotides, and the sections were counterstained using Duolink Mounting Medium with 4Ј,6-diamidino-2-phenylindole. Samples were examined using a Nikon microscope (Melville, NY).
Co-immunoprecipitation-Co-immunoprecipitation experiments were performed using the Pierce Classic IP kit (Pierce) according to the manufacturer's protocol. Briefly, whole cell extract was precleared with control agarose resin for 1 h at 4°C. The clarified supernatant was then incubated with anti-PKC␦ or -p65 antibody or its isotype control overnight at 4°C followed by a 1-h incubation with Protein A/G Plus-agarose beads. The beads were washed five times, and immunoprecipitated proteins were subjected to immunoblotting.
Mouse Models of Abdominal Aortic Aneurysm-The generation of Prkcd target deletion in mice was described elsewhere (30). Prkcd Ϫ/Ϫ (PKC␦ KO) mice and their wild-type (WT) littermates were generated by mating heterozygous pairs. Male mice (12 weeks of age) underwent a CaCl 2 -or elastase-induced abdominal aortic aneurysm model as described previously (31,32). Briefly, animals were anesthetized using a continuous flow of 1-2% isoflurane. For the CaCl 2 model, the infrarenal region of the aorta was isolated and perivascularly treated with 0.5 M CaCl 2 or 0.5 M sodium chloride (NaCl control) via gauze for 15 min. For the elastase model, the infrarenal region of the aorta was isolated, and temporary silk ligatures were placed at proximal and distal portions of the aorta. An aortotomy was created near the distal ligature using a 30-gauge needle, and heat-tapered polyethylene tubing (Baxter Healthcare Corp.) was introduced through the aortotomy and secured with a silk tie. The aorta was filled with saline containing 0.295 unit/ml type I porcine pancreatic elastase (Sigma) or heat-inactivated elastase solution (control) at a constant pressure of 100 mm Hg. Buprenorphine was administered subcutaneously at a dose of 0.05 mg/kg immediately after surgery. Subsequently, a 2.5% Xylocaine topical ointment was applied to the suture site. The maximum external diameter of the infrarenal aorta was measured using a digital caliper (VWR Scientific, Radnor, PA) prior to treatment and at the time of tissue harvest. At selected time points, mice were sacrificed by an overdose of isoflurane, and tissues were harvested. Tissues meant for RNA isolation were stored in RNAlater RNA Stabilization Reagent (Qiagen). Tissues meant for in situ proximity ligation assay were freshly imbedded in O.C.T. Compound (Sakura Tissue Tek, Netherlands). All frozen sections were cut to 6 m thick using a Leica CM3050S cryostat. All animal experiments in this study were approved by the Institutional Animal Care and Use Committee at the University of Wisconsin-Madison (Protocol M02284) and performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health.
Statistical Analysis-Student's t test or one-way ANOVA with Bonferroni's post hoc test was used to evaluate the statistical differences. Differentially expressed genes in microarray analysis were identified by moderated t test implemented in ArrayStar software. All experiments were repeated at least three times. Data are presented as mean Ϯ S.E. Differences with p Ͻ 0.05 were considered significant.

PKC␦ Regulates Proinflammatory Chemokine Expression-
We have previously reported a 2.51 Ϯ 0.24-fold increase in PKC␦ protein in VSMCs of aneurysmal arteries (18,33). To study the role of PKC␦ in chemokine expression by VSMCs, we mimicked this pathological PKC␦ up-regulation by overexpressing PKC␦ in cultured arterial VSMCs to a similar level (2.88 Ϯ 0.15-fold induction) (Fig. 1A) followed by a brief activation with a low concentration of PMA (1 nM for 6 h). At this concentration, PMA alone had no effects on Ccl2 expression (Fig. 1B). Next, we determined transcriptome profiles using microarray analysis on total RNA isolated from PKC␦-overexpressing and control VSMCs. After normalization of hybridization intensities using the Robust Multichip Array algorithm in the ArrayStar software, the average expression level of each gene was calculated from biological triplicates. Differentially expressed genes were identified with a moderated t test implemented in ArrayStar software. To analyze the biological function of the transcripts differentially regulated by PKC␦, the entirety of differentially expressed transcripts was loaded into the pathway analysis program based on the Ingenuity Pathways Knowledge Base. Among the top five implicated disease conditions, two involve inflammation (Table 1). Indeed, 45 differentially expressed genes have known inflammatory roles. And a series of proinflammatory chemokines including Ccl2/Mcp-1, Ccl7, Cxcl16, and Cx3cl1 are among those genes.
Next, we validated expression of the chosen chemokines using qPCR. As shown in Fig. 1, C-F, PKC␦ significantly increased expression of Ccl2, Ccl7, Cxcl16, and Cx3cl1 to a similar extent as observed by microarray analysis. In contrast, Ccl4 and Cxcl1, which were not among the list of target genes, were not affected by PKC␦ (Fig. 1, G and H).
PKC␦ Is Required for Chemokine Expression in Experimental Aneurysm-Next, we addressed whether PKC␦ regulates chemokine expression in abdominal aortic aneurysm, a disease condition known to involve inflammation. PKC␦ knock-out mice and their WT littermates were subjected to aneurysm induction by perivascular administration of CaCl 2 . Consistent with our previous report, PKC␦ knock-outs were resistant to aneurysm induction, whereas their wild-type counterparts displayed visible aortic expansion ( Fig. 2A). qPCR analysis showed a 2-14-fold increase in arterial expression of Ccl2, Ccl7, Cxcl16, and Cx3cl1 by induction of aneurysm in the wild-type mice (Fig.  2, B-E). Prkcd gene deletion eliminated or markedly dimin- ished this chemokine up-regulation (Fig. 2, B-E), further supporting the stimulatory role of PKC␦ in the regulation of chemokine expression during aneurysm pathogenesis.
PKC␦ Stimulates Ccl2 Transcription-We used CCL2, one of the critical chemokines that mediate the pathogenesis of vascular diseases (34), as a prototype to elucidate how PKC␦ regulates chemokine expression. As shown in Fig. 3, A and B, overexpression/activation of PKC␦ led to a rapid increase in levels of CCL2 mRNA and protein, peaking around 6 and 24 h, respectively. We also mimicked different PKC␦ expression intensities by treating arterial VSMCs with increasing concentrations of adenoviruses carrying the Prkcd gene. Both CCL2 mRNA and protein responded to PKC␦ in a dose-dependent manner (Fig.   3, C and D). Next, we knocked down endogenous PKC␦ using a specific siRNA in VSMCs stimulated with TNF␣, a critical inflammatory mediator implicated in the pathogenesis of aneurysm (35). PKC␦-specific siRNA reduced the PKC␦ protein level by 70% compared with control (Fig. 3E). Although TNF␣ up-regulated Ccl2 expression in scrambled siRNA-transfected VSMCs, this proinflammatory function was significantly impaired by PKC␦ knockdown (Fig. 3F). Pretreatment of VSMCs with rottlerin, a PKC␦ chemical inhibitor, also significantly attenuated TNF␣-induced Ccl2 production in VSMCs (Fig. 3G).
Because introns are rapidly removed from heterogeneous nuclear RNA during splicing, levels of unspliced pre-mRNA Differentially expressed genes in PKC␦-overexpressing and control VSMCs were loaded into Ingenuity Pathway Analysis 9.0 software to perform biological functional analyses. The table indicates names of biological functions and differentially expressed genes (first column) and the number of genes in each disease condition (second column). Inflammatory chemokines are shown in bold.
To determine whether PKC␦ affects the stability of Ccl2 mRNA, we measured the rate of Ccl2 mRNA degradation in VSMCs with various PKC␦ activities with or without up-regulated PKC␦. As shown in Fig. 4C, overexpression of PKC␦ did not alter the rate of Ccl2 mRNA degradation. Similarly, the rate of Ccl2 mRNA decay was nearly identical in cells with or without the presence of the PKC␦ inhibitor rottlerin (Fig. 4D). These results collectively indicate that PKC␦ up-regulates Ccl2 through increasing its transcription without affecting the stability of Ccl2 mRNA in VSMCs.
NF-B Subunit p65, but Not p50, Is Required for the Up-regulation of Ccl2 by PKC␦-Because PKC␦ does not have a known DNA binding capacity, we postulated that it regulates chemokine transcription by directly or indirectly phosphorylating transcription factors. To aid identification of such downstream factors, we analyzed the PKC␦-regulated genes for potential common regulatory motifs. Results of Ingenuity Pathway Analysis 9.0 identified CDKN2A, PDX1, IRF1, STAT1, IRF3, STAT2, RB1, NF-B, and TP53 as common transcription factors shared by multiple PKC␦-regulated genes. Among these transcription factors, NF-B stood out as it is known to regulate each of the four PKC␦-dependent proinflammatory chemokines (Ccl2, Ccl7, Cxcl16, and Cx3cl1) ( Table 2).
Once again, we used CCL2 as a prototype to delineate how PKC␦ and NF-B may interact in the context of chemokine regulation. Sequence analysis of the Ccl2 gene promoter region confirmed the presence of NF-B binding motifs along with several other cis-elements including those for CCAAT/enhancer-binding protein, AP-1, Sp-1, and tonicity-response element/osmotic response element (data not shown). As shown in Fig. 5A, andrographolide (15 M), an NF-B inhibitor (37), completely blocked Ccl2 induction by PKC␦. In contrast, SR11302 (1 M), which is known to inhibit AP-1 activity (38), had no significant effect on Ccl2 induction (Fig. 5A). To determine which subunit of NF-B is involved in the proinflammatory function of PKC␦, siRNA was used to silence NF-B subunits. siRNAs against p65 and p50 efficiently and specifically knocked down p65 and p50 compared to control, respectively ( Fig. 5B). Although knockdown of p65 completely eliminated the effect of PKC␦ on Ccl2 mRNA (Fig. 5C), we were surprised to find that knockdown of p50 significantly increased, rather than suppressed, PKC␦-mediated Ccl2 production. Furthermore, p65 knockdown abolished the stimulatory effect of p50 knockdown on Ccl2 production (Fig. 5C). Taken together, these results indicate that NF-B subunit p65, but not p50, is critical for the up-regulation of Ccl2 by PKC␦.
PKC␦ Enhances p65 Phosphorylation and DNA Binding-Because IB degradation is a major signaling step leading to NF-B activation, we examined IB␣ levels in VSMCs that were infected by AdPKC␦ or AdNull and then treated with or without PMA. As shown in Fig. 6A, PKC␦ overexpression/activation did not produce any significant alteration of IB␣ levels.
We then sought to examine whether PKC␦ enhances nuclear translocation of p65 by analyzing p65 protein levels in cytoplasmic and nuclear fractions. PKC␦ did not cause any significant change in the ability of p65 to accumulate in the nucleus (Fig. 6,  C and D).
We next examined p65 phosphorylation status, which is known to influence p65 dimerization (39), DNA binding (40), and transactivation capacity (40,41). As shown in Fig. 6B, PKC␦ overexpression/activation significantly increased p65 phosphorylation at serine 536. This induction was prohibited by andrographolide (15 M) (Fig. 6B). Next, we examined p65 in the nuclear and cytosolic fractions and found that PKC␦ overexpression/activation significantly increased p65 phosphorylation at serine 536 in both nuclear and cytosolic fractions. This Actinomycin D (5 g/ml) was added to shut down transcription. Cells were harvested at the indicated times, total RNA was isolated, and levels of Ccl2 mRNA were analyzed by qPCR. D, VSMCs were treated with 20 ng/ml TNF␣ for 6 h, and actinomycin D (5 g/ml) was then added with 2 M rottlerin or DMSO. Cells were harvested at the indicated times, total RNAs were isolated, and levels of Ccl2 mRNA were analyzed by qPCR. Results are presented as a percentage of Ccl2 mRNA remaining over time compared with the amount before the addition of actinomycin D. Data show the mean of independent experiments. Error bars represent S.E. n ϭ 3; *, p Ͻ 0.05, two-tailed Student's t test.

TABLE 2 Common transcription regulators shared by PKC␦-regulated genes in VSMCs
Differentially expressed genes in PKC␦-overexpressing and control VSMCs were loaded into Ingenuity Pathway Analysis 9.0 software to perform transcription factor analyses. The table indicates names of transcription regulators (first column) and names of genes sharing a common transcription regulator (second column). Inflammatory chemokines are shown in bold. induction was also found to be sensitive to andrographolide (Fig. 6, C and D). In contrast, PKC␦ did not alter the phosphorylation status of p65 at Ser-276 (data not shown). Moreover, despite the unaltered nuclear accumulation of p65, the levels of p65 that bound to DNA were significantly increased by PKC␦ overexpression/activation (Fig. 6E). Gene deficiency of Prkcd had no significant effect on the basal level of NF-B subunit p65. However, the lack of PKC␦ significantly attenuated the effect of TNF␣ on p65 including Ser-536 phosphorylation and DNA binding (Fig. 6, F and G). In addition, knockdown of PKC␦ did not produce any significant alteration in protein levels of IB␣ (Fig. 6F), suggesting that PKC␦ activates p65 through an IB-independent mechanism. PKC␦ Forms a Complex with p65 in VSMCs-Using the in situ proximity ligation assay that detects two proteins within 40 nm, we discovered that PKC␦ and p65 might be physically associated with one another. The number of PKC␦-p65 complexes was found to be more abundant in PKC␦-overexpressing cells (Fig. 7A). Co-immunoprecipitation analysis showed that p65 immunoprecipitated with PKC␦, and overexpression of PKC␦ by adenoviral vector increased the amount of co-immunoprecipitated p65 (Fig. 7C). In the reciprocal experiment, PKC␦ coimmunoprecipitated with p65, confirming the PKC␦-p65 interaction (Fig. 7D). To detect PKC␦-p65 interaction in vivo, we used an elastase-induced model of abdominal aortic aneurysm (42). Interestingly, PKC␦-p65 complexes were visible in aortas treated with elastase or heat-inactivated elastase (control).
However, the abundance of PKC␦-p65 complexes, particularly those detected in medial VSMCs (marked by positivity for smooth muscle-specific ␣-actin), was much more pronounced in elastase-treated or aneurysmal arteries than that detected in control arteries (Fig. 7B).
PKC␦ is believed to shuttle between the cytosol and nucleus and other membrane-associated subcellular compartments. Translocation of PKC␦ requires interaction of its N terminus with its specific anchoring protein called receptor for activated C-kinase (RACK) (25,43). Subcellular analyses of PKC␦ showed that the same principle applies to VSMCs with or without PKC␦ overexpression. Activation of PKC␦ with PMA increased nuclear accumulation of this kinase but with a slow dynamics (Fig. 8A). Because the PKC␦-p65 complexes were detected largely outside the nucleus, we postulated that manipulation of PKC␦ subcellular translocation would impact PKC␦-p65 interaction and thus chemokine expression. To this end, we utilized two TAT-linked peptides developed by Mochly-Rosen and co-workers (25). Similar to what this group reported in cardiomyocytes, VSMCs responded to PKC␦-specific translocation inhibitor ␦V1-1 and activator ␦RACK with reduced and enhanced PKC␦ nuclear translocation, respectively (Fig.  8B). Interestingly, ␦RACK nearly eliminated the ability of PKC␦ to induce Ccl2 in AdPKC␦-infected VSMCs and significantly impaired Ccl2 expression in cells treated with TNF␣ (Fig.  8, C and D), suggesting that PKC␦ nuclear translocation might impede its regulation of Ccl2 expression. Conversely, ␦V1-1 enhanced the Ccl2 induction in AdNull-infected VSMCs and in cells in response to TNF␣ (Fig. 8, C and D). However, under PKC␦-overexpressing conditions, ␦V1-1 did not cause an additional increase in Ccl2 expression (Fig. 8C).

DISCUSSION
Since being cloned in 1987, PKC␦ has been studied in diverse cellular processes including cell growth, apoptosis, mitogenesis, differentiation, tumor progression, and tissue remodeling (12,44,45). Data presented here support that PKC␦ regulates vascular inflammation in part through stimulating expression of proinflammatory chemokines by VSMCs. We identified Ccl2, Ccl7, Cxcl16, and Cx3cl1 as PKC␦-regulated early response genes. All these chemokines were up-regulated in a PKC␦-dependent manner by experimental induction of abdominal aortic aneurysm, explaining at least in part the diminished inflammatory response in PKC␦-null mice we reported previously (18). . NF-B p65 DNA binding assays were carried out using nuclear protein (E). F, VSMCs were transfected with PKC␦-specific or scrambled siRNA for 24 h followed by incubation with TNF␣ (10 ng/ml) or PBS for the indicated time, and whole-cell lysates were subjected to immunoblot analysis with the indicated antibodies. G, VSMCs isolated from WT and PKC␦ KO mice were incubated with TNF␣ (10 ng/ml) or PBS for 30 min. Nuclear proteins were isolated, and NF-B p65 DNA binding assays were carried out. Data show the mean of independent experiments. Error bars represent S.E. n ϭ 3-6; *, p Ͻ 0.05, one-way ANOVA. PCNA, proliferating cell nuclear antigen; P-p65, phosphorylated p65; Andro, andrographolide.
CCL2, also known as MCP-1, is a member of the small inducible gene family. Increased CCL2 expression has been reported in mechanically injured arteries (46), aortic aneurysm (47), and vulnerable atherosclerotic plaques (48). Blocking CCL2 signaling through the use of neutralizing antibodies or gene deletion of the Ccl2 gene or its receptor, Ccr2, attenuates intimal hyperplasia, aneurysm formation, and vascular inflammation in experimental models (46,49). CCL2 is mostly known for its role in the recruitment of monocytes and other types of inflammatory cells. CCL2 can be produced by macrophages and other types of inflammatory cells. VSMCs are another important source of CCL2 when stimulated by cytokines, oxidized LDL, and bacterial lipopolysaccharide.
Prior studies including those from our own group have implicated PKC␦ in the regulatory pathway of CCL2 expression (18,19,33). Liu et al. (50) suggested that PKC␦ mediates the stability of Ccl2 mRNA when VSMCs are treated with PDGF-BB or angiotensin II. However, an early study from this group showed that TNF␣ has no obvious effect on Ccl2 mRNA stability (51). In our study, levels of Ccl2 mRNA declined at a similar rate regardless of the status of PKC␦ expression or activity. The fact that inhibition of NF-B by andrographolide or p65 gene silencing nearly eliminated the effect of PKC␦ supports the notion that transcriptional regulation might be a predominant mode of action by elevation of PKC␦ in the context of regulating chemokines.
The promoter regions of both the mouse and rat Ccl2 genes contain potential binding motifs for CCAAT/enhancer-binding protein, NF-B, AP-1, Sp-1, and tonicity-response element/ osmotic response element. Prior studies have confirmed the role of NF-B and AP-1 as important transcription factors for Ccl2 expression (52,53). In keeping with this notion, blocking the NF-B pathway either with a pharmacologic inhibitor or siRNA to p65 eliminated PKC␦-mediated Ccl2 production. Although AP-1 and mitogen-activated protein kinases are known to be critical for Ccl2 expression in a wide range of cell types (52,54,55), inhibition of AP-1 did not significantly impact Ccl2 expression in PKC␦-overexpressing cells. This result suggests that PKC␦ may not signal through the mitogen-activated protein kinase/AP-1 pathway in regulation of Ccl2; however, we cannot rule out the possibility that SR11302 at 1 M, a common concentration reported in the literature (56), might not produce sufficient inhibition of AP-1 in VSMCs.
Another interesting and unexpected finding is the lack of involvement of the p50 NF-B subunit. Because RelA/p65-p50 heterodimer is the major NF-B complex detected in most cell types, one would expect that knocking down either subunit would attenuate PKC␦-induced Ccl2. Our data clearly showed that knocking down p50 significantly increased, rather than suppressed, PKC␦-mediated Ccl2 production in the absence or presence of p65 siRNA. The p50 subunit of NF-B is synthesized as a precursor molecule of 105 kDa (p105) that functions as an inhibitor of p65. Although future studies are required to test whether the particular p50 siRNA used in the current study increases levels of p65 by reducing p105, our findings may represent the first report on differential interactions between PKC␦ and NF-B subunits. In resting cells, NF-B dimers are sequestered in the cytosol by association with IB proteins. Upon stimulation, IB proteins are phosphorylated, ubiquitinated, and degraded. This event constitutes a major regulatory step for NF-B activation. Although the mechanism by which PKC␦ regulates the NF-B pathway, particularly in VSMCs, has not been well studied, studies carried out in other cell types have indicated several potential mechanisms. In neutrophils, PKC␦ or its downstream effector, PKC, act similarly to IB kinases by phosphorylating IB, promoting IB degradation, and subsequently liberating NF-B to the nucleus (57). In endothelial cells, PKC␦ is reported to increase the transactivation potential of NF-B via an IB kinase/IB-independent pathway involving p38 MAPKor Akt-mediated phosphorylation of NF-B (58 -60). In HEK293 and U2OS cells, PKC␦ forms complexes with p65 in the nucleus following TNF␣ exposure. The PKC␦-p65 complex is believed to occupy NF-B target gene promoters and orchestrate RelA/p65 transactivation (14). Our study revealed another mechanism that involves a cytosol-specific PKC␦-p65 interaction that is independent of the IB pathway. In support of this novel mechanism, the PKC␦-p65 complexes were largely detected in the cytosol of VSMCs. We also observed an increased abundance of PKC␦-p65 complexes in the vascular wall of aneurysmal tissues that likely resulted from up-regulated PKC␦ in diseased arteries. We do not currently know whether PKC␦ activates p65 through direct or indirect phosphorylation, although this is a plausible scenario based on our observation that PKC␦ enhanced phosphorylation of p65 at Ser-536 and increased DNA binding of this transcription factor. Numerous studies have shown that inducible post-translational modifications of NF-B subunits, especially phosphorylation, are important for NF-B to efficiently induce transcrip- tion of target genes (24). The p65 subunit contains 12 potential phosphoacceptor sites. Among these sites, phosphorylation at Ser-276 by protein kinase A has been thought to mediate dimerization and DNA binding of p65 (39), whereas the phosphorylation of Ser-536 increases p65 transcriptional activity (41) likely through more efficient DNA binding and recruitment of p300 (40). Furthermore, sufficient phosphorylation of p65 at Ser-536 is critical for NF-B activation (40). Although NF-B mediates most of the chemokine expression, we observed that PKC␦ selectively up-regulates Ccl2, Ccl7, Cxcl16, and Cx3cl1 in VSMCs. The selective effect on gene expression might be due to site-specific phosphorylation of p65, which targets NF-B activity to particular gene subsets by influencing p65 and phosphorylated RNA polymerase II promoter recruitment (61).
Subcellular localization is believed to be an important regulatory mechanism in determining activation and substrate specificity of the PKC family of kinases. For instance, PKC␦ activates the apoptotic program through multiple inter-related events. The initiating event involves the transduction of a "death" signal to PKC␦ from various apoptotic stimuli via diacylglycerol or tyrosine phosphorylation (62). Another event involves caspase-mediated cleavage, which produces the proapoptotic catalytic fragment (␦CF) (12). Nuclear translocation is thought to be a final event through which PKC␦ commits a cell to an apoptotic fate (62). A series of in vitro and in vivo experiments conducted by Mochly-Rosen and co-workers (25) has convincingly illustrated the importance of subcellular localization of PKCs in cardiomyocytes. They showed that a PKC␦specific translocation inhibitory peptide (␦V1-1) protects cells in ischemic hearts, whereas a translocation activator peptide (␦RACK) increases myocyte damage after an ischemic insult (25). Using the same peptides, we confirmed their effect on PKC␦ nuclear translocation in VSMCs. Surprisingly, our data suggest that nuclear translocation of PKC␦ is not necessary for its action on Ccl2. In fact, the observation that ␦RACK completely disabled the ability of PKC␦ to induce Ccl2 production suggests that its key substrate(s) in regulation of the Ccl2 gene is located in the cytosol. In support of this idea, our proximity ligation analysis showed that the majority of PKC␦-p65 complexes existed in the cytosol. Furthermore, blocking PKC␦ translocation with ␦V1-1 favored Ccl2 expression. Of note, ␦V1-1 enhanced the Ccl2 induction in PKC␦-non-overexpressing VSMCs but did not cause additional stimulation of Ccl2 expression in PKC␦-overexpressing cells likely due to the large cytosolic presence of PKC␦ under overexpression conditions. The trivial role of nuclear PKC␦ in the regulation of Ccl2 gene expression is also demonstrated by our time course studies of Ccl2 mRNA induction as well as PKC␦ nuclear translocation. The relatively slow dynamics of PKC␦ translocation in VSMCs makes it an unlikely mechanism to be responsible for the rapid turning on of the Ccl2 gene.
It is important to point out that the PKC␦/p65 pathway is only one of several pathways involved in the regulation of proinflammatory chemokines. Aside from NF-B (63), TNF␣ also utilizes AP-1 (54), Sp-1 (64), and Akt/PKB (65) pathways to regulate Ccl2. In line with this notion, altering the PKC␦ pathway either through siRNA or translocation peptides caused a significant but moderate change in Ccl2 production triggered by TNF␣. Despite its moderate role in the regulation of chemokine expression, mice lacking PKC␦ are protected from developing vascular inflammation when subjected to aneurysm induction (18), underscoring the significance of the PKC␦ pathway in the inflammatory response.
Taken together, our results suggest that PKC␦ regulates CCL2 and likely other proinflammatory chemokines in VSMCs mainly at the transcriptional level. PKC␦ physically interacted with the NF-B subunit p65 in the cytosol and enhanced its DNA binding affinity likely through phosphorylation. The translocation activator peptide ␦RACK attenuated Ccl2 production, providing a way to specifically block PKC␦-regulated proinflammatory chemokines. Our evidence suggests that PKC␦ has a unique role in the regulation of Ccl2 in VSMCs, and we believe these data may aid the development of drugs for future treatment of inflammatory diseases.