INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE SAPHENOUS VEIN IS USED to bypass arteriosclerotic occlusive disease of the coronary and other peripheral arteries; however, the long-term results of these interventions are not always satisfactory. The leading cause of failure of arterial bypass is neointimal hyperplasia, a process characterized by smooth muscle cell (SMC) migration, proliferation, and deposition of extracellular matrix protein within the arterial wall (5). After vascular reconstructions, SMCs are exposed to many factors that promote their migration and proliferation (38, 39). These factors stimulate a variety of intracellular signaling pathways that could potentially mediate both processes. Although a plethora of information is available about the growth factors and cytokines that stimulate vascular SMC migration and proliferation, less is known about the intracellular proteins that control these functions.

Protein kinase C (PKC) is a major serine-threonine kinase that regulates multiple intracellular events including secretion of various proteins as well as cellular proliferation. Eleven isotypes of PKC with varying tissue distribution and biochemical regulation have been reported (43). The discovery of these various isotypes and the postulation that these isotypes have distinct physiological functions provide an explanation for the plethora of cellular events that appear to be mediated by PKC. These isotypes differ with regard to their dependency on Ca²⁺ and phospholipids and their ability to be activated and downregulated by phorbol esters. The "conventional" PKC isoforms-, -I, -II, and - have Ca²⁺-binding domains, and thus their activation is dependent on physiological levels of Ca²⁺. The "novel" PKC isoforms, which include PKC-, -, -, -, and -µ, lack Ca²⁺-binding domains. Although Ca²⁺ is not required for their activation, diacylglycerol- and phorbol ester-binding domains are present in these novel isoforms. The "atypical" PKC isotypes, which include PKC-, -, and - ( and  are identical), lack both Ca²⁺- and phorbol ester-binding domains. The mechanism by which these atypical isotypes are activated is unclear; however, it has been proposed that the phospholipid phosphoinositide 3-kinase is upstream from PKC- (37).

The role of PKC activation in SMC proliferation has been studied in some detail, but a unified hypothesis has not emerged. Although many investigators (1, 14, 27, 29, 30) have found that activation of PKC leads to SMC proliferation, Sasaguri et al. (49) concluded that PKC activation inhibits SMC replication. In these latter studies of SMCs derived from the rat thoracic aorta, activation of PKC, specifically during the late G₁ phase of the cell cycle, resulted in inhibition of SMC proliferation. The relationship between PKC activation and SMC migration has been less well investigated. An integral role for PKC in cell locomotion is supported by the observations that 1) many of the cytoskeletal-related proteins such as talin, filamin, and vinculin are substrates for PKC (3, 32); 2) incubation of cells with phorbol esters, which are direct activators of PKC, results in changes in cell shape that are likely a consequence of an effect on cytoskeletal organization (36); and 3) incubation of cells with phorbol esters disrupts actin filaments and focal adhesions (47). The effect of PKC activation on cellular locomotion has been studied in multiple cell types other than SMC and appears to be cell specific. Activation of PKC enhances the migration of neutrophils, monocytes, and endothelial cells (ECs) (33, 42, 48, 56), whereas in human keratinocytes, glioma cells, and several epithelial cell lines, activation of PKC inhibits cell locomotion (2, 19, 21).

The current study investigated the role of PKC in the migration and proliferation of SMCs derived from the human saphenous vein. Although both migration and proliferation are essential processes for the development of intimal plaque, we found that the effect of PKC activation is paradoxical in that PKC activation enhances proliferation but inhibits migration of vascular SMCs. Furthermore, the inhibitory effect of PKC on SMC locomotion appears to be mediated by a pathway that involves cAMP and cAMP-dependent protein kinase (protein kinase A, PKA). Finally, our studies suggest that the effect of PKC on SMC migration and proliferation is mediated by one or more of the phorbol ester-responsive isotypes PKC-, -I, -, and -.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. Human recombinant basic fibroblast growth factor (bFGF), epidermal growth factor (EGF), and platelet-derived growth factor-AB (PDGF) were from Upstate Biotechnologies (Lake Placid, NY). The smooth muscle-specific actin immunostaining kit, phorbol-12,13-dibutyrate (PDBu), PKA catalytic subunit, ()-indolactam V [()-ILV], DMSO, phenylmethylsulfonyl fluoride (PMSF), SDS, and Triton X-100 were from Sigma (St. Louis, MO). DMEM, PBS, fetal bovine serum (FBS), EDTA, penicillin-streptomycin-amphotericin B solution, L-glutamine, HEPES, phorbol 12-myristate 13-acetate (PMA), and the cDNA preamplification system were from GIBCO-BRL (Gaithersburg, MD). Oligonucleotide primers were custom synthesized by GIBCO-BRL. Forskolin, the cAMP Rp-isomer (Rp-cAMPs), and the PKA-selective inhibitor myristoylated PKA inhibitory peptide (PKI) were purchased from Bio-Mol (Plymouth Meeting, PA). Protein A-Sepharose (PAS) was from Pharmacia Biotechnology (Uppsala, Sweden). Rabbit polyclonal anti-PKC-, -I, -II, -, -, -, -, -, -µ, -, and - were from Santa Cruz Biotech, (Santa Cruz, CA). Polycarbonate membranes (pore size 8 µm) were from Poretics (Livermore, CA). [Methyl-³H]thymidine was purchased from DuPont-NEN (Boston, MA). The PKA assay kit was purchased from MBL International (Watertown, MA). The ECL system was from Amersham Life Science (Little Chalfont, UK). The RNA isolation kit was purchased from Qiagen (Valencia, CA).

Cell cultures. Human SMC were harvested from explants of remnant portions of the saphenous vein intended for aortocoronary or peripheral arterial bypass grafting as previously described (18, 35). Briefly, sections of the saphenous vein were opened lengthwise, and the endothelial and adventitial layers were gently removed. Fragments of the medial layer were placed onto tissue culture plates, and outward growing SMC were harvested and subcultured. Cells were maintained at 37°C and 5% CO₂ in DMEM supplemented with 10% FBS, 25 mM HEPES, 40 U/ml penicillin G, 40 µg/ml streptomycin, 100 ng/ml amphotericin B, and 4.8 mM L-glutamine. SMC identity was verified by immunostaining with anti-human -actin antibody and by a characteristic "hill and valley" growth pattern. SMC isolated from human saphenous veins are proliferative in response to serum, PDGF, and other mitogens and remain so until their fifth passage. As such, we used cells in passages 2-5 for all experiments. To address variability related to the heterogeneity of patients from whom the saphenous veins were derived, we performed all experiments in cells isolated from the veins of three or more patients.

Downregulation of PKC. SMC were simultaneously starved and depleted of PKC by incubation for 72 h with 2 µM PMA in 0.5% FBS-DMEM. Control cells were made quiescent by incubation in 0.5% FBS containing DMEM with the solvent for PMA, DMSO, for 72 h.

Assays of proliferation. Confluent SMC in 100-mm dishes were detached with 0.05% trypsin-EDTA, seeded onto 24-well plates (10,000 cells/well) in 10% FBS-DMEM, and allowed to attach for 24 h. SMC were starved in 0.5% FBS-DMEM for 72 h and then stimulated for another 72 h with agonists as indicated. Cultures were washed with PBS, and, after removal with trypsin-EDTA, the number of cells was determined with a cell counter (Coulter Electronics; Hialeah, FL). For determination of DNA synthesis, quiescent SMC were stimulated with agonists for 24 h with 2 µCi of [methyl-³H]thymidine added to each well during the final 4 h of the assay. Cell lysate was precipitated with 10% trichloroacetic acid, and radioactivity of incorporated [³H]thymidine in the trichoroacetic acid-insoluble fraction was determined using a liquid scintillation counter.

Chemotaxis assay. SMCs were grown to confluence in 100-mm culture dishes and then made quiescent by incubation in 0.5% FBS-DMEM for 72 h. Cells were then washed in PBS, harvested using 0.05% trypsin-EDTA, and resuspended in serum-free DMEM. Assays were performed over 4 h at 37°C using a 48-well microchemotaxis chamber (Neuroprobe; Cabin John, MD), with the upper and lower wells separated by polycarbonate membrane (pore size 8 µm) filters. Agonists in DMEM were incubated in the lower wells, and SMCs suspended in serum-free DMEM were seeded at a density of 25,000 cells/well (3,000 cells/mm²) into the upper well of the chamber. For evaluation of the effects of protein kinase agonists and/or inhibitors, the indicated final concentrations of these agents were coincubated with the cell suspension in the upper well of the chamber. On completion of the assay, membranes were removed, fixed in 70% ethanol at 20°C for 20 min, and stained in hematoxylin overnight. The upper side of the membrane was scraped using a cotton swab to remove cells that had attached but not migrated, and the membrane was then mounted onto a microscope slide. Chemotaxis in each well was assessed by counting the number of cells in five independent high-power fields (×200 magnification).

Attachment assay. Assays of SMC attachment were performed for 2 h at 37°C in a 48-well microchemotaxis chamber with upper and lower wells separated by a polycarbonate membrane (pore size 8 µm). For all assays, SMC were seeded at a density of 2,000 cells/well into the upper well of the chamber with and without the various agonists and inhibitors. After a 2-h period, the membranes were washed, fixed, and stained in the same manner as for the migration assays. Attachment was assessed by counting the number of cells in five independent high-power fields at ×200 magnification on the upper side of the membrane.

PKA activity assay. SMC were grown to confluence in six-well plates. After 72-h incubation with 0.5% FBS-DMEM, cells were stimulated with agonists as described and then washed and lysed in radioimmunoprecipitation assay (RIPA) buffer. Lysates were added to 96-well plates coated with the immobilized PKA substrate peptide (Arg-Phe-Ala-Arg-Lys-Gly-Ser-Asn-Val) for 15 min. This reaction was stopped by adding 20% H₃PO₄. Phosphorylation was quantified by adding a biotinylated anti-phosphorylated PKA substrate peptide, followed by peroxidase-conjugated streptavidin. Color development was achieved by o-phenylenediamine and hydrogen peroxide, and optical density was measured at 492 nm in a spectrophotometer (57).

RNA extraction and RT-PCR. Total RNA was extracted from cultured SMCs, and RNA (2 µg) was reversed transcribed using reverse transcriptase according to the manufacturer's instructions. The first strand cDNA encoding each PKC isoenzyme gene sequence was amplified by PCR using specific primers (Table 1) designed from the reported human sequences deposited with the GenBank database (53, 54). PCR amplification was conducted in a reaction volume of 50 µl using a thermal cycler and 0.5 units of Taq DNA polymerase set on a program consisting of an initial denaturation at 94°C for 3 min followed by 30 cycles of 1-min denaturation (94°C), 2-min annealing (55°C), and 2-min elongation (72°C), with a final extension period of 10 min at 72°C. PCR products were subsequently size fractionated on 2% agarose gels, stained with ethidium bromide, and visualized under ultraviolet light.

Immunoblotting and immunoprecipitation. For analysis of PKC isotypes, SMCs, made quiescent in 0.5% FBS-DMEM for 72 h, were lysed in RIPA buffer (PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 100 µg/ml PMSF, 30 µl/ml aprotinin, and 1 mM sodium orthovanadate). Total protein concentration was determined by a modification of the method of Lowry, and samples were subjected to SDS-PAGE. The electrophoretically separated proteins were transferred to a polyvinylidene difluoride membrane and then incubated with rabbit polyclonal antibodies to the various PKC isotypes, followed by biotinylated goat anti-rabbit IgG. Labeled proteins were visualized with an ECL system. Immunoprecipitation was necessary for detection of PKC- and was performed as follows: Confluent SMC in 100-mm dishes were rinsed and lysed in RIPA buffer. Samples were then centrifuged for 20 min at 4°C, and the resulting supernatant was incubated overnight at 4°C with antibody to PKC- bound to PAS. Immunoprecipitates were then washed three times with RIPA buffer. Samples were heated and denatured for 5 min and then subjected to SDS-PAGE with the buffer system of Laemmli. Western blotting was performed as described above.

Statistical analysis. All experiments were performed at least in triplicate. All values are provided as means ± SD. Significance of results was determined with Statview (BrainPower; Calabasas, CA) on a Apple Macintosh system. Comparisons were performed using an unpaired Student's t-test, and P < 0.05 was considered to indicate a significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

PKC and SMC proliferation. To evaluate whether activation of PKC resulted in proliferation of human vascular SMC, quiescent SMC were incubated with varying concentrations of the PKC activators PDBu and ()-ILV in a 3-day assay with cell number as the index of proliferation. PDBu (10 nM) produced a 27 ± 3% increase in SMC proliferation; this proliferative effect was sustained at PDBu concentrations of up to 100 nM (Fig. 1A). A similar response was observed with ()-ILV, which produced an ~40% increase in cell number at concentrations ranging from 0.5 to 5 µM. Although these increases were significant, they were less than the increase in cell number that was found in SMC stimulated by bFGF (79 ± 3%) and PDGF-AB (112 ± 4%) (Fig. 1B). We also evaluated the effect of PDBu and ()-ILV on DNA synthesis. PDBu and ()-ILV produced, respectively, a 200 ± 5 and 388 ± 8% increase in DNA synthesis in vascular SMC; however, the magnitude of this response was also significantly less than that observed when SMC were stimulated with PDGF-AB and bFGF (PDGF-AB: 845 ± 15% and bFGF: 495 ± 18%, means ± SD, n = 3-5).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   The effect of protein kinase C (PKC) activation on smooth muscle cell (SMC) proliferation. A: stimulation of SMCs with increasing concentrations of the PKC activators phorbol-12,13-dibutyrate (PDBu) and ()-indolactam V [()-ILV]. *P < 0.05 compared with unstimulated control. B: stimulation of downregulated (open bars) and control SMC (solid bars) with PDBu (100 nM), ()-ILV (1 µM), epidermal growth factor (EGF; 10 ng/ml), basic fibroblast growth factor (bFGF; 5 ng/ml), or platelet-derived growth factor (PDGF)-AB (5 ng/ml). *P < 0.05 compared with nondownregulated control. C: stimulation of SMCs coincubated with increasing concentrations of the selective PKC inhibitor chelerythrine with EGF (; 10 ng/ml), bFGF (; 5 ng/ml), or PDGF-AB (; 5 ng/ml). For all experiments, SMC proliferation was assayed over 72 h (cell counts) as described in MATERIALS AND METHODS. Results are expressed as the percent increase ± SD compared with unstimulated controls. Experiments were performed in triplicate and repeated in cells from 3 different patients. Data from representative experiments are shown.

PKC and SMC chemotaxis. To evaluate whether PKC is an intermediate in the signaling pathway for migration, chemotaxis was measured in the presence of increasing concentrations of the two PKC activators PDBu and ()-ILV. Our initial studies showed that activation of PKC inhibited, rather than enhanced, basal (unstimulated) migration of vascular SMCs (data not shown). We then studied the effect of PKC activation on SMC chemotaxis induced by PDGF-AB. Stimulation of PKC with PDBu inhibited SMC chemotaxis induced by PDGF in a concentration-dependent manner, with an IC₅₀ of ~100 nM (Fig. 2A). A similar inhibitory effect was seen when PDGF-AB-stimulated SMCs were coincubated with ()-ILV (IC₅₀ = 3 µM). Because activation of PKC inhibited SMC chemotaxis, we then postulated that inhibition of PKC might enhance chemotaxis. To test this hypothesis, cellular PKC was eliminated by downregulation as previously described. PKC depletion significantly increased the rate of basal (unstimulated) migration (Fig. 2A). Moreover, downregulation not only reversed the inhibitory effect of PDBu on SMC chemotaxis but also significantly enhanced the overall chemotactic effect of PDGF-AB (Fig. 2A). Treatment of SMC with the PKC inhibitor chelerythrine also accentuated the chemotactic effect of PDGF-AB (Fig. 2B). Thus PKC, in contrast to its stimulatory effect on SMC proliferation, appears to have an inhibitory effect on SMC migration.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   The effect of PKC on SMC chemotaxis. A: baseline chemotaxis and chemotaxis to PDGF-AB (5 ng/ml, lower chamber) were measured in control (solid bars) and PKC-depleted SMCs (open bars) in the presence of increasing concentrations of PDBu (upper chamber). *Comparison of downregulated and nondownregulated SMCs stimulated with various concentrations of PDBu and PDGF-AB. B: chemotaxis of SMCs to PDGF-AB (5 ng/ml, lower chamber) was measured in the presence of increasing concentrations of chelerythrine (upper chamber). *P < 0.05 compared with PDGF-AB-induced chemotaxis without chelerythrine. For all experiments, SMC chemotaxis was assayed using a microchemotaxis chamber as described in MATERIALS AND METHODS. Results are expressed as the degree of increase ± SD compared with unstimulated controls. Experiments were performed in triplicate and repeated in cells derived from 3 different patients. Data from representative experiments are shown.

Inhibitory effect of PKC on SMC chemotaxis is mediated through a pathway involving cAMP and PKA. We then wished to identify the mechanism by which PKC produced this inhibitory effect on SMC chemotaxis. Prior studies in this laboratory (unpublished observations) have revealed that activators of cAMP inhibit chemotaxis of rat aortic vascular SMCs. Moreover, an association between PKC and the cAMP/PKA signaling pathways has been previously demonstrated (44, 46). Therefore, we postulated that the inhibitory effect of PKC on SMC chemotaxis might be mediated through the cAMP/PKA pathway. Stimulation of human venous SMC with forskolin, a direct activator of cAMP, strongly inhibited PDGF-induced migration of SMCs derived from the human saphenous vein (Fig. 3A). Forskolin (1 µM) produced a 50% inhibition of PDGF-induced SMC chemotaxis, and forskolin (25 µM) completely eliminated the chemotactic effect of PDGF-AB. To evaluate whether stimulation of PKC in human SMCs activates the cAMP/PKA pathway, we measured PKA activity in SMC stimulated with the phorbol ester PDBu. Activation of PKC with PDBu produced a concentration-dependent increase in PKA activity. In fact, PKA activity induced by 100 nM PDBu was similar to that induced by 1 µM of forskolin (Fig. 3B). Interestingly, 100 nM PDBu and 1 µM forskolin both produced a similar (~50%) inhibition of PDGF-AB-induced migration. These data raise the possibility that PKC might effect inhibition of SMC chemotaxis by activating PKA. To further substantiate a relationship between PKC and cAMP/PKA, we studied the effect of agents that block either cAMP or PKA on PKC-induced inhibition of SMC chemotaxis. A competitive inhibitor of cAMP, Rp-cAMPs (7, 50), at a concentration of 5 µM, completely reversed the inhibitory effect of PDBu on PDGF-induced SMC chemotaxis (Fig. 3C), whereas this inhibitor did not substantially increase the rate of basal migration of vascular SMCs (data not shown). In similar studies (6, 52), we found that a selective cell-permeable peptide inhibitor of PKA, myristoylated PKI (100 nM), also eliminated the inhibitory effect of PKC on SMC chemotaxis, whereas basal chemotaxis was unaffected (Fig. 3C). These studies provide convincing evidence that the effect of PKC on SMC chemotaxis is mediated through the cAMP/PKA pathway.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   The role of the cAMP/protein kinase A (PKA) pathway in PKC-mediated inhibition of SMC chemotaxis. A: effect of cAMP activation on SMC chemotaxis. Chemotaxis of SMCs to PDGF-AB (5 ng/ml, lower chamber) was measured in the presence of increasing concentrations of forskolin (upper chamber). *P < 0.05 compared with PDGF-AB-stimulated chemotaxis without forskolin. B: activation of PKA by PKC and forskolin. SMCs were grown to confluence in 6-well plates, starved for 72 h, and then stimulated with either the cAMP activator forskolin (1 µM) or PDBu (100 nM) for 5 min. Cells were lysed, and PKA activity was assayed as described in MATERIALS AND METHODS. *P < 0.05, comparison with unstimulated controls. C: effect of blocking cAMP or PKA on PKC-induced inhibition of SMC chemotaxis. Chemotaxis to PDGF (5 ng/ml, lower chamber) was assayed in the presence of PDBu (100nM) and increasing concentrations of cAMP Rp-isomer (Rp-cAMPs; left) or myristoylated PKA inhibitory peptide (PKI; right) (upper chamber). *P < 0.05 compared with SMCs stimulated with PDGF-AB alone. D: effect of cAMP activation on SMC proliferation. SMCs were stimulated for 24 h with PDGF (5 ng/ml) coincubated with increasing concentrations of forskolin. DNA synthesis was measured using [3H]thymidine incorporation as described in MATERIALS AND METHODS. *P < 0.05 compared with SMCs stimulated with PDGF-AB alone. Results of all experiments are expressed as the degree of increase ± SD compared with unstimulated controls. Experiments were performed in triplicate and repeated in cells derived from 3 different patients. Data from representative experiments are shown.

Identification of PKC isotypes. To identify the isoenzymes of PKC that are expressed in human SMCs, we performed both RT-PCR and Western blotting. PCR amplification of reverse-transcribed RNA derived from human saphenous vein SMCs revealed the conventional isotypes PKC-, -I, and -II, the novel isotypes PKC-, -, and -µ, and the atypical isotypes PKC- and - (Fig. 4A). The isoenzymes PKC-, -, and - were not found.

View larger version (61K): [in this window] [in a new window]   Fig. 4.   Identification of PKC isoenzymes. A: identification of PKC isoenzymes by RT-PCR. RT-PCR was performed using total RNA from SMCs. The products were subjected to electrophoresis on 2% agarose gels and visualized with ethidium bromide. RT-PCR product sizes for each PKC isoenzyme are shown in Table 1. B: identification of PKC isoenzymes by Western blotting. Western blotting for PKC isotypes was performed using lysates from control (a) and SMCs depleted of PKC (b) by downregulation as described in MATERIALS AND METHODS. For PKC-, cell lysates were immunoprecipitated with anti-PKC-. cPKC, conventional PKC isotypes; nPKC, novel PKC isotypes; aPKC, atypical PKC isotypes.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Our findings imply an important role for the second messenger PKC in the pathophysiological events that mediate intimal hyperplasia. Activation of PKC is necessary for SMC proliferation. Conversely, PKC activation markedly inhibits SMC chemotaxis. We were surprised by these findings and in fact anticipated that PKC activation would stimulate both proliferation and migration because many of the agonists of intimal hyperplasia (e.g., PDGF-BB and angiotensin II) both activate PKC and stimulate these two physiological events (10). This disparity may be explained by the fact that, in these studies, total PKC activity was measured rather than the activity of specific isotypes. Because there are multiple isotypes of PKC and activation of individual isotypes may lead to differing physiological events, the growth factor PDGF may stimulate an isotype of PKC that enhances proliferation but not an isotype that inhibits migration. Understanding the diversity of signaling pathways that mediate SMC migration and proliferation may provide insight on how to control the processes of intimal hyperplasia.

The necessity of PKC for cellular proliferation has been previously evaluated using SMCs derived from a variety of animals (1, 14, 27, 29, 30). Our observation that PKC activation stimulates human saphenous vein SMC proliferation is compatible with similar observations in other cell types. However, it has also been reported that PKC activation of SMC synchronized in the late G₁ phase can inhibit serum-induced proliferation (49). With the use of similar conditions and with the use of SMCs derived from the human saphenous vein, we were unable to replicate the findings of these authors (unpublished observation). It is of interest that the stimulatory effect of PKC on SMC proliferation was relatively modest [~30-40% increase for PDBu and ()-ILV compared with an 80 and 100% increase, respectively, for the growth factors bFGF and PDGF]. Moreover, downregulation of PKC diminished the proliferative effect of PDGF and bFGF by only ~25%, suggesting that PKC is only partially necessary for SMC proliferation in response to these agonists. Surprisingly, coincubation of SMC with the selective inhibitor of PKC chelerythrine completely eliminated the proliferative effect of bFGF and PDGF. One possible explanation for this discrepancy is that the phorbol ester-dependent isotypes of PKC are responsible only in part for the effect of PKC on SMC proliferation, whereas an isotype of PKC that is not responsive to phorbol esters but is inhibited by chelerythrine, such as PKC-, mediates a much stronger proliferative effect. Supporting this hypothesis is the observation that overexpression of PKC- in fibroblasts results in a significant increase in agonist-induced proliferation of these cells (4). Moreover, Liao et al. (28) recently showed that depletion of PKC- from vascular SMCs using antisense oligonucleotides diminishes angiotensin II-induced activation of mitogen-activated protein kinase (MAPK), a signaling protein considered to be essential for cellular proliferation. An alternative explanation for the differential effect on proliferation provided by downregulation of SMCs of chelerythrine is that, in the concentrations used, chelerythrine is toxic to SMCs. This explanation is less likely because similar concentrations of chelerythrine have been found to inhibit PKC in other cell types without cytotoxic effects (22).

The role of phorbol ester-responsive PKC activity in the signaling pathway for proliferation is distinct with regard to the three growth factors tested. SMC proliferation induced by PDGF and bFGF was inhibited by downregulation of PKC, whereas proliferation in response to EGF was unaffected. We (34) previously demonstrated in SMCs that activation and nuclear translocation of MAPK by EGF is also not facilitated by PKC, whereas MAPK activation in response to PDGF and bFGF is dependent on PKC. Thus EGF can stimulate both MAPK activation and proliferation through a pathway that is independent of at least a phorbol ester-responsive isotype of PKC.

A major novel finding of this study is that PKC activation inhibits human SMC chemotaxis. The evidence for this hypothesis is substantial. Two distinct activators of PKC inhibit SMC chemotaxis. Moreover, both a selective inhibitor of PKC and elimination of cellular PKC by downregulation accentuate basal and PDGF induced SMC chemotaxis. The observation that migration is increased by the elimination of PKC in unstimulated cells suggests the possibility that a basal level of activated PKC in SMC acts to chronically suppress migration.

There appears to be intercellular variation in the role of PKC in cell locomotion. Activation of PKC enhances migration of neutrophils and monocytes (33, 42). Recently, Ng and co-workers (41) reported that overexpression of PKC- by transient transfection of mammary epithelial cells stimulates migration on ₁-integrin substrates. In several other cell types, including human keratinocytes, glioma cells, and several epithelial cell lines, activation of PKC inhibits migration (2, 19, 21). A relationship between PKC and cellular locomotion is not surprising because stimulation of smooth muscle and other cells with phorbol esters leads to marked changes in the organization of actin filaments and cell shape.

Our data suggest that PKC inhibits SMC migration through activation of the cAMP pathway. cAMP is an intracellular second messenger formed from ATP by the membrane-bound enzyme adenylyl cyclase. cAMP then activates PKA by dissociating its regulatory subunit. The free catalytic subunit of PKA thereby initiates a series of enzymatic reactions, leading to phosphorylation and activation of multiple downstream proteins. We found that forskolin, a direct activator of adenylyl cyclase, strongly inhibited SMC migration. We then postulated that PKC might produce its inhibitory effect on SMC chemotaxis by directly activating adenylyl cyclase and increasing levels of cAMP. In support of this hypothesis, we found that activation of PKC in human SMCs resulted in an increase in PKA activity roughly equivalent to that produced when SMC were stimulated with concentrations of forskolin that were capable of inhibiting SMC chemotaxis. To confirm an association between PKC, cAMP, and chemotaxis, we employed the competitive cAMP inhibitor Rp-cAMPS (7, 50). Inactivation of cAMP by Rp-cAMPs reversed the inhibitory effect of PKC on SMC migration, verifying that cAMP is the mechanism through which PKC exerts this effect. To further implicate the cAMP/PKA pathway as the mechanism through which PKC inhibits SMC chemotaxis, we performed similar experiments with the PKA-specific inhibitor myristoylated PKI (6, 52). Inhibition of PKA with myristoylated PKI also substantially reversed the effect of PKC on SMC chemotaxis.

It has been previously shown that several other factors that inhibit SMC migration act through a pathway that involves cAMP/PKA. Stimulation of rat aortic SMC with adrenomedullin increases levels of cAMP, and the cAMP analog 8-bromo-cAMP inhibits SMC chemotaxis (15). Moreover, it has been postulated that the inhibitory effect of PDGF-AA on SMC chemotaxis is also mediated through activation of cAMP (26). Our data now reveal an interaction between the second messenger PKC and the cAMP/PKA pathway. Cross-talk between these two signaling systems has been previously reported; in cardiac muscle cells, PKC- and - were both found to phosphorylate and activate type V adenylyl cyclase (17, 20). Moreover, in rat tail artery SMCs, PKC activation induced by PMA results in enhanced production of cAMP (46).

The mechanism through which PKA inhibits SMC chemotaxis is not clear; however, elevation of cAMP has been found to disrupt actin filaments, and there are several downstream substrates of PKA that are associated with the cytoskeleton. We (55) have previously shown that elevation of intracellular calcium is necessary for SMC migration, and others (51) have shown that activation of PKA reduces levels of intracellular calcium. Another mechanism by which PKA might suppress migration is by inhibiting MAPK. We (40) recently found that activation of MAPK stimulates SMC migration, and it has been reported that PKA inhibits PDGF-BB induced MAPK activity in arterial SMCs (8).

Although our data suggest that the inhibitory effect of PKC on SMC chemotaxis is mediated through the cAMP/PKA pathway, the stimulatory effect of PKC on SMC proliferation does not appear to be regulated by this same pathway. Activation of cAMP in human SMCs inhibits proliferation. In fact, the concentrations of the cAMP activator forskolin that are required to produce this effect are ~25 times greater than the concentration required to inhibit SMC migration. The downstream pathway by which PKC stimulates SMC proliferation is not clear, but previous studies (34) in our laboratory and others suggest that MAPK activation may be integrally involved.

PKC consists of a family of at least 11 isoenzymes, and it has become increasingly clear that many of the functions of PKC reflect the action of specific isoenzymes. The distribution of these isoenzymes varies among cell types, and even the function of a particular isoenzyme may differ from one cell to another. To establish which isoenzymes are present in human saphenous vein SMCs, we evaluated both RNA and protein using RT-PCR and Western blotting. We identified the PKC-, -I, -II, -, -, -µ, -, and - isotypes in human SMCs. There are a plethora of studies (29, 49, 53, 54) in which the isotypes of PKC in SMCs from a variety of vascular sources have been evaluated. Surprisingly, the isotypes identified have varied greatly. These discrepant findings may be related to differences between cell sources and passages. Also, the variability in the specificity and cross-reactivity of the antibodies used in these studies may also account for the disparate findings. It is likely that the eight proteins that we have identified are indeed the available isotypes of PKC in human saphenous vein SMCs because identical isoforms were identified by both PCR and Western blotting.

The existence of multiple isotypes provides an explanation for the many and diverse physiological actions that follow activation of PKC. In vascular SMCs, proliferation and migration are only two of many such cellular functions stimulated by PKC. Others include induction of SMC contraction, secretion of anticoagulant substances such as PGI₂ and tissue plasminogen activator, and production of extracellular matrix (25, 31, 45). It is conceivable that each of these functions is mediated by a different isotype of PKC. There is evidence in other cell types as well as SMCs that the various isotypes of PKC have distinct physiological effects (16, 23).

Downregulation of vascular SMCs leads to depletion or elimination of the isotypes PKC-, -I, -, and -. Ali et al. (1) also found in cultured vascular SMCs from the rat aorta that PKC-, -, and - were downregulated by prolonged treatment by PKC activators; however, PKC-I was not identified in their SMCs (9). We surmise that the phorbol ester-dependent effect of PKC on SMC proliferation and the inhibitory effect of PKC on SMC migration (which is completely phorbol ester dependent) are mediated by one or more of these isotypes. Prior studies provided additional support for a role for these four enzymes in SMC proliferation and migration, although there is great variability in the findings. Herbert et al. (13) found that the inhibitory effect of heparin on SMC proliferation is mediated through inhibition of PKC-. Although one might conclude from these findings that activation of PKC- is necessary for SMC proliferation, surprisingly, these same authors found that inhibition of PKC- with antisense oligonucleotides did not inhibit serum-induced proliferation. Haller et al. (9) found that levels of PKC- decreased in proliferating SMCs and increased during SMC differentiation, suggesting an inverse relationship between activation of PKC- and proliferation. Thus the role of PKC- in SMC proliferation remains unclear. There are additional data that suggest a possible association between the isotypes PKC-, -, and - and migration. Overexpression of PKC- in SMCs results in an increase in the quantity of actin and the number of actin filaments, and, also, PKC- can be activated by contractile stimuli (9, 10). PKC- has been associated with calcium-independent contraction (16), and PKC- has been shown to translocate to the cytoskeleton (10). The precise role of each of these isotypes in both migration and proliferation of vascular SMCs will require further study; however, our findings suggest that PKC-, -I, -, and - are the proteins that are most likely involved in both processes.

Although a variety of cells contribute to the abnormal response that can follow vascular injury, the behavior of ECs and SMCs is critically important in determining whether a vessel will return to its uninjured state or develop a hyperplastic response. In designing agonists that might control the response to injury, the ideal agent would inhibit SMC proliferation and migration but simultaneously stimulate (at least temporarily) these same two processes in ECs. In previous studies (22, 56), we showed that activation of PKC stimulates both EC proliferation as well as migration. Furthermore, we (11) found that overexpression of the isotype PKC- in ECs facilitates migration. Thus there is some potential that the same isotype or isotypes of PKC might inhibit migration and proliferation of SMCs and enhance these same processes in ECs. Theoretically, increased activation of this particular isotype might then produce an adaptive and healing response rather than a hyperplastic response after arterial injury.

In summary, we found that activation of PKC results in a differential effect with stimulation of SMC proliferation and inhibition of SMC chemotaxis. The inhibitory effect of PKC on SMC migration appears to be mediated through increased production of cAMP, which in turn activates PKA. Eight of eleven currently known isotypes of PKC were identified in human vascular SMCs. Our data suggest that the specific isotypes PKC-, -I, -, and/or - may be responsible for the effects of PKC on migration and proliferation. Understanding the signaling pathways that mediate the intimal hyperplastic response to vascular injury is a necessary prerequisite before inhibitors of this process can be successfully designed.