J Biol Chem, Vol. 275, Issue 15, 11270-11277, April 14, 2000

Proliferation of Intimal Smooth Muscle Cells
ATTENUATION OF BASIC FIBROBLAST GROWTH FACTOR 2-STIMULATED PROLIFERATION IS ASSOCIATED WITH INCREASED EXPRESSION OF CELL CYCLE INHIBITORS^*

N. Eric Olson, Jeff Kozlowski, and Michael A. Reidy

From the Department of Pathology, University of Washington, Seattle, Washington 98195

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Basic fibroblast growth factor (FGF2) is a potent mitogen for medial smooth muscle cells and is necessary for their proliferation after balloon catheter injury; however, intimal smooth muscle cells do not require FGF2 for their proliferation, and they respond only weakly to exogenous FGF2. The present study examined the activation of extracellular signal-regulated kinase (ERK) signaling as well as the expression and activity of cell cycle proteins in FGF2-stimulated intimal smooth muscle cells. FGF2 activates ERKs 1 and 2, and Western blot analysis showed that cyclin D, cyclin E, and cyclin-dependent kinase (CDKs) 2 and 4 were expressed in intimal smooth muscle cells after FGF2 infusion. FGF2 stimulation, however, did not lead to phosphorylation of the retinoblastoma protein (Rb), CDK 2 activation, or expression of cyclin A. Western blot analysis showed that intimal smooth muscle cells express elevated levels of the cell cycle inhibitors p15^INK4b and p27^Kip1, compared with medial smooth muscle cells, and that FGF2 stimulation does not reduce the level of these inhibitors. These studies suggest that despite activation of ERKs 1 and 2 and expression of the cell cycle activators, cyclin D and cyclin E, high levels of cell cycle inhibitors may inhibit cell cycle transit in FGF2-stimulated intimal smooth muscle cells.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Excessive growth of vascular smooth muscle cells is an important component in the development of atherosclerotic lesion and in restenosis. In order to study which factors control the growth of these cells, we and many others have used a model of smooth muscle cell proliferation induced by mechanical injury of the rat carotid artery (1, 2). In this model, an inflated Fogarty balloon catheter is passed into the lumen of the common carotid artery, denuding the artery of its endothelial cell lining and damaging the underlying medial smooth muscle cells. This injury results in a predictable response; within 2 days the medial smooth muscle cells begin proliferating, and after 4 days medial smooth muscle cells migrate into the intima, where they continue to proliferate for up to 2 weeks. This leads to the formation of a thickened neointima comprised primarily of smooth muscle cells and extracellular matrix and results in luminal narrowing. In this model, basic fibroblast factor (FGF2)¹ has been shown to be a critical mitogen for the proliferation of medial smooth muscle cells. The addition of FGF2 significantly increases medial smooth muscle cell proliferation when administered after injury of the rat carotid artery (3), and medial smooth muscle cell proliferation can be significantly inhibited by neutralizing antibodies to FGF2 (4). Unlike medial smooth muscle cells, FGF2 does not seem to be involved in regulating the proliferation of the smooth muscle cells that have migrated into the intima; neutralizing antibodies to FGF2 do not inhibit intimal smooth muscle cell proliferation (5) after balloon catheter injury, and the addition of FGF2 to arteries with existing intimal lesions causes only a small increase in proliferation (3). These data suggest that, in contrast to medial smooth muscle cells, FGF2 is not necessary for intimal smooth muscle cell proliferation, nor is it a potent mitogen for those cells. The purpose of this study was to determine whether differences in the activation of cytoplasmic signaling pathways and/or cell cycle regulation could be responsible for this apparent attenuation of FGF2-stimulated proliferation in intimal smooth muscle cells.

FGF2 signal transduction involves the activation of many different cytoplasmic signaling molecules, including the extracellular signal-regulated kinases 1 and 2 (ERKs 1 and 2) (6-9). Activation of the ERKs is required for FGF2-stimulated proliferation in several different cell types, and recently we have shown that the ERK signaling pathway is activated following balloon catheter denudation of the rat carotid artery and that ERK activity is required for smooth muscle cell proliferation following this injury (10). FGF2 stimulation also activates the PI 3-kinase pathway (11). Activation of this pathway is required for FGF2-stimulated proliferation in a variety of cell types including smooth muscle cells (12), and recent data suggest that this pathway is also activated following balloon catheter injury (10).

Although FGF2 stimulation requires activation of cytoplasmic signaling molecules such as the ERKs and PI 3-kinase to induce proliferation, the resultant signaling must ultimately lead to activation of the cyclin-dependent kinases (CDKs) in order for cells to progress through the G₁ phase of the cell cycle and into S phase. The activity of the CDKs is regulated in part by the controlled expression of the cyclins. The cyclins associate with and activate the CDKs. Recently, it has been shown that both ERK and PI 3-kinase signaling can regulate the expression of cyclin D (7, 13, 14), the activating partner for CDK 4. Cyclin D-CDK 4 activation is required for the phosphorylation and inactivation of Rb (15, 16). This frees the transcription factor E2F, which stimulates the expression of factors necessary for the initiation of S phase, including cyclin E (17-19). Cylin E activates CDK 2, and this activity along with that of cyclin D-CDK 4 leads to the increased expression of cyclin A (18, 20, 21), which is necessary for entry into S phase (22). Thus, for FGF2 signal transduction to result in proliferation, there must be activation of the ERK and/or PI 3-kinase signaling pathways, and this signaling must result in increased expression of cyclin D as well as activation of the cyclin D-CDK 4 complex.

Although the expression of the cyclins are necessary for activation of the CDKs, they are not sufficient; there are specific CDK inhibitors capable of inhibiting the CDKs even in the presence of the cyclins (23-26). The CIP/KIP family of inhibitors can inhibit both the cyclin D-CDK 4 and cyclin E-CDK 2 complexes (23), while the INK4 family of inhibitors is more specific, only inhibiting CDK 4 activity (27, 28). Overexpression of these inhibitors can attenuate the proliferative response (20, 29), while a reduction in their expression increases proliferation (30, 31). Therefore, the level of these proteins could be critical in determining whether growth factor stimulation results in proliferation.

Our data show that FGF2 stimulation of smooth muscle cells in established intimal lesions activates both the ERKs and PI 3-kinase and increases cyclin D expression but does not lead to phosphorylation of the retinoblastoma protein, activation of CDK 2, or increased expression of cyclin A. In these same arteries, high levels of the cyclin-dependent kinase inhibitors p27^Kip1 and p15^INK4b were noted, and we believe that their presence is responsible for the attenuation of FGF2-induced proliferation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Injury Models-- Male Harlan Sprague-Dawley rats (Tyler Laboratories, Bellevue, Washington), age 3-3.5 months, were used throughout these experiments. Rat carotid arteries were injured using either the gentle injury technique previously described (32) or balloon catheter denudation. The catheter was introduced into the left common carotid through the left external branch and passed up and down the common carotid three times to ensure complete denudation. To assess the effects of FGF2 on medial smooth muscle cell proliferation, FGF2 (generous gift of Scios Nova) was infused (60 µg/rat) into the rats via the tail vein immediately after gentle injury. To assess the effects of FGF2 in intimal smooth muscle cell proliferation, FGF2 was infused (60 µg/rat) into the rats via the tail vein 6 weeks after balloon catheter injury.

Quantification of Smooth Muscle Cell Proliferation-- Each animal received intraperitoneal injections of tritiated thymidine (specific activity 6.7 Ci/mmol; NEN Life Science Products) at 1, 9, and 17 h prior to being killed with an overdose of sodium pentobarbital (160 mg/kg of body weight; Anthony Products Co., Arcadia, CA) and perfusion-fixed for 5 min with 4% paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.3). Segments of carotid artery were embedded in paraffin, and 6-µm sections were cut (four per animal). Sections were dipped in photographic emulsion (NTB2; Eastman Kodak Co.) and exposed for 2 weeks at 4 °C before being developed. Smooth muscle cell proliferation was measured by counting the number of labeled nuclei, and the [³H]thymidine index ((labeled nuclei/total nuclei) × 100%) was calculated. At least four cross-sections per carotid were quantitated. Statistical analysis was performed using Student's t test (two-tailed, unpaired).

Western Blot Analysis-- Carotid arteries were briefly flushed with Ringer's lactate, excised, stripped of adventitia, and frozen in liquid nitrogen. The frozen tissue was then ground with mortar and pestle under liquid nitrogen until reduced to a fine powder, which was suspended in a cell lysis solution containing 10 mM HEPES, pH 7.4, 50 mM Na₄P₂O₇, 50 mM NaF, 50 mM NaCl, 5 mM EDTA, 5 mM EGTA, 2 mM Na₃VO_4, 0.1% Triton X-100, 0.5 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin. Samples were sonicated for 5 s on ice, and insoluble matter was removed by centrifugation. A small sample of the resulting supernatant was reserved for protein determination; the rest was diluted 1:1 in sample buffer containing 5% -mercaptoethanol and boiled for 5 min. The protein concentration of each sample was determined using a bicinchoninic acid assay (Pierce) with BSA as a standard. Equal amounts of total protein were loaded onto polyacrylamide gels, electrophoresed, and transferred to nitrocellulose. The blot was then immunostained using antibodies specific for either phospho-PKB (New England Biolabs), cyclin D (Upstate Biotechnology, Inc., Lake Placid, NY), cyclin E (Upstate Biotechnology), cyclin A (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), Rb (Pharmingen), Cdc25A (Santa Cruz Biotechnology), p27^Kip1 (Transduction Laboratories), p21^Cip (Santa Cruz Biotechnology), or p15^INK4b (Santa Cruz Biotechnology). Briefly, the blot was first incubated in 5% nonfat dried milk in TBS-T (10 mM Tris, pH 7.6, 150 mM NaCl, and 0.1% Tween 20) for 1 h. The blot was then incubated with primary antibody diluted in TBS-T buffer for 1 h. After rinsing, the blot was exposed to a horseradish peroxidase-conjugated secondary antibody diluted in TBS-T. After rinsing, blots were incubated in ECL (Amersham Pharmacia Biotech) reagent for 1 min, blotted dry, and then exposed to Amersham Pharmacia Biotech ECL film until a signal was detected.

ERK Activity Assay-- Equal amounts of protein from carotid arteries prepared for Western blot analysis were electrophoresed on a 10% SDS-polyacrylamide gel containing 0.4 mg/ml myelin basic protein. SDS was removed from the gel by multiple washes with buffer A (50 mM HEPES, pH 7.4, 5 mM -mercaptoethanol) containing 20% isopropyl alcohol. Proteins were denatured with 6 M guanidine HCl in 50 mM HEPES, pH 7.4, 5 mM -mercaptoethanol and then renatured with buffer A containing 0.04% Tween 20. The gel was washed with buffer B (50 mM HEPES, pH 7.4, 100 µM Na₃VO₄, 10 mM MgCl₂, 5 mM -mercaptoethanol). The gels were then incubated with buffer B containing 50 µCi of [-³²P]ATP for 1 h at 37 °C. The reaction was stopped by washing the gel with 5.0% trichloroacetic acid with 10 mM Na₄P₂O₇. After several washes, the gel was dried and subjected to autoradiography. Autoradiographs were then scanned using a flat bed scanner. The images were then analyzed using NIH Image software.

CDK 2 Activity Assay-- 150 µg of protein was normalized in 1.0 ml of CDK 2 lysis buffer (50 mM HEPES, pH 7.4, 1 mM sodium fluoride, 150 mM sodium chloride, 10% glycerol, 2.5 mM EGTA, 1 mM EDTA, 10 mM -glycerophosphate, 0.1 mM sodium orthovanadate, 0.1% Tween 20, 1 mM dithiothreitol, 2.5 µg/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride). 2 µg of anti-CDK 2-agarose conjugate (Santa Cruz Biotechnology) was added to each sample and incubated overnight at 4 °C with constant rotation. Samples were washed three times with CDK 2 lysis buffer and then twice with CDK 2 kinase buffer (40 mM Tris, pH 7.6, 20 mM MgCl₂, 2 mM dithiothreitol). Washes were carried out at 4 °C with 1.0 ml of buffer, a gentle agitation, and then a 5-min spin at 3000 × g. Samples were then incubated with 30 µl of reaction buffer (40 mM Tris, pH 7.6, 20 mM MgCl₂, 2 mM dithiothreitol, 67 µg/ml histone H1, 7 µM ATP with 167 µCi/ml -ATP) for 30 min at 30 °C. The reaction was stopped with the addition of 6× Laemmli sample buffer followed by a 5-min boil. Samples were then run out on a 12% SDS-polyacrylamide gel. The gel was washed three times with gel fixative (5% trichloroacetic acid, 10 mM sodium pyrophosphate), dried, and subjected to autoradiography.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Response of Arterial Smooth Muscle Cell to Exogenous FGF2-- To evaluate the response of medial smooth muscle cells to FGF2, 60 µg of FGF2 was administered intravenously to rats immediately after their carotid artery was subjected to a denuding injury using a nylon filament loop. This procedure completely removes the endothelial cells lining the carotid artery and yet causes only a small increase (1.5%) in medial smooth muscle cell proliferation (32). Denudation is necessary because endothelialized arteries do not respond to systemic infusion of FGF2 (3). Intimal smooth muscle cells are not normally found in rat carotid arteries, so in order to evaluate the response of intimal smooth muscle cells to FGF2, it was necessary to induce the formation of a neointima by balloon catheter injury. 6 weeks after this injury, a well established intimal lesion composed primarily of intimal smooth muscle cells and extracellular matrix is present (1, 2). At this time, the lesion is not reendothelialized, and there is a low rate of spontaneous proliferation (~1%) that is comparable with that induced by the gentle injury. The smooth muscle cells in this established intimal lesion will be referred to as "intimal smooth muscle cells." This is in contrast to "medial smooth muscle cells," which refers to the smooth muscle cells in the media of a normal artery after deendothelialization by gentle injury.

When FGF2 was administered acutely after the gentle injury, medial smooth muscle cell proliferation was increased 20-fold (Fig. 1). In contrast, intimal smooth muscle cells in arteries with established lesions showed a small increase in proliferation after FGF2 stimulation (Fig. 1). These data show that intimal smooth muscle cells are significantly less responsive to FGF2 than are medial smooth muscle cells. FGF2 did not cause any increase in the proliferation of the medial smooth muscle cells in arteries with established intimal lesions (data not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Effects of FGF2 on medial and intimal smooth muscle cell proliferation. FGF2 (60 µg, gray bar) or saline (vehicle, black bar) was given intravenously to rats with either acutely injured carotid arteries (medial smooth muscle cells) or established intimal lesions (intimal smooth muscle cells). Thymidine index was determined 48 h after administration of FGF2 or saline.

FGF2 Stimulation of Cytoplasmic Signaling Pathways-- To verify that the infused FGF2 is able to bind to and activate its receptor in intimal smooth muscle cells, we measured the activation of the ERK signaling pathway. The extracellular signal-regulated kinase pathway is activated, via Ras/Raf/mitogen-activated protein kinase/extracellular signal-regulated kinase kinase, by many mitogenic stimuli, including growth factors such as platelet-derived growth factor, epidermal growth factor, and the FGFs (33-35). An in-gel kinase assay was used to measure ERK activity in medial and intimal smooth muscle cells before and after FGF2 stimulation. Low levels of activity were observed both in uninjured carotid arteries and in arteries with established intimal lesions in the absence of FGF2 stimulation. The administration of FGF2 significantly increased ERK activity, with maximal stimulation seen at 6 h after administration (Fig. 2) in both acutely injured arteries and arteries with established intimal lesions, demonstrating that FGF2 does bind to and activate its receptor in intimal smooth muscle cells.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   ERK activity after FGF2 stimulation. Autoradiographs from in-gel kinase assays were scanned, and the intensities of the bands representing ERK 1 and 2 activities were quantitated using image analysis software. A, ERK activation in arteries with established intimal lesions was measured at 5 min, 30 min, 1 h, 6 h, and 24 h after FGF2 stimulation. -Fold stimulation is relative to unstimulated artery with established intimal lesion (0 h). ERK activity data show that FGF2 stimulation increases ERK 1 and 2 activity in arteries with established intimal lesions and that this increase can be seen for up to 24 h after stimulation. B, ERK activation in acutely injured arteries, measured 5 min, 1 h, 6 h, and 24 h after FGF2 stimulation. -Fold stimulation is relative to normal uninjured artery (0 h). These data show that ERK activity in acutely injured carotid arteries is increased after FGF2 stimulation and that the level of ERK activity in these medial smooth muscle cells is similar to that in seen in the intimal smooth muscle cells of arteries with established intimal lesions.

Since smooth muscle cell proliferation may require activation of pathways other than the ERK pathway, we also examined activation of the PI 3-kinase pathway by measuring phosphorylation of protein kinase B (PKB), a downstream target of PI 3-kinase signaling (36). FGF2 administration also caused an increase in PKB phosphorylation in acutely injured arteries and in arteries with established intimal lesions (Fig. 3), suggesting that FGF2 stimulation also activates PI 3-kinase in both cell types.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   PKB phosphorylation following FGF2 stimulation. Quantitation of Western blot for phospho-PKB shows that FGF2 (black bars) stimulates an increase in PKB phosphorylation compared with control vessels (open bars) within 5 min. For medial SMC, the control is an uninjured carotid artery. For medial SMC, the control is an unstimulated artery with an established intimal lesion.

The Effects of FGF2 on the Cell Cycle-- The observation that intimal smooth muscle cell proliferation is only weakly stimulated by FGF2, despite significant activation of the ERKs and PI 3-kinase, suggests that either this signaling was not sufficient to initiate entry into the cell cycle or that a block in the cell cycle prevented transit through G₁ and into S phase in these cells. To determine whether FGF2-stimulated intimal smooth muscle cells enter the cell cycle, we examined the expression of cyclins D, E, and A, using Western blot analysis. Cyclin D was detected neither in normal uninjured vessels nor in established intimal lesions, but 24 h after the administration of FGF2, cyclin D1 was elevated in arteries subjected to the gentle injury as well as in arteries with established intimal lesions (Fig. 4A). Cyclin E was present in both normal arteries and in arteries with established intimal lesions, and FGF stimulation had no effect on cyclin E expression (Fig. 4B). Cyclin A was expressed neither in normal uninjured arteries nor in arteries with established intimal lesions (Fig. 4C). FGF2 stimulation, however, did cause a marked increase in cyclin A expression in medial smooth muscle cells in acutely injured arteries but not in intimal smooth muscle cells (Fig. 4C). Western blot analysis also showed that CDK 4 and CDK 2 were expressed in normal uninjured arteries and in unstimulated arteries with intimal lesions and that FGF2 stimulation had no effect on their expression (data not shown).

View larger version (45K): [in this window] [in a new window]   Fig. 4.   Expression of cyclins D, E, and A after FGF2 stimulation. A, Western blot analysis shows that cyclin D expression is increased after FGF2 stimulation of both medial and intimal smooth muscle cells. B, Western blot analysis shows that medial and intimal smooth muscle cells express cyclin E. C, Western blot analysis shows that FGF2 stimulation does not increase the expression of cyclin A in vessels with established intimal lesions. Cyclin A expression is increased in FGF2-stimulated medial smooth muscle cells.

To determine whether the cyclin D-CDK 4 complex was active in FGF2-stimulated intimal smooth muscle cells, we examined the phosphorylation state of Rb. Rb regulates cell cycle transit by binding to and inactivating the E2F family of transcription factors. Phosphorylation of Rb by cyclin D-CDK 4 releases E2F, thus permitting the transcription of genes necessary for replication (15). Western blot analysis showed that Rb is hyperphosphorylated in medial smooth muscle cells 24 and 48 h after stimulation with FGF2 but not in intimal smooth muscle cells stimulated with FGF2 (Fig. 5A). This finding suggests that the signaling initiated by FGF2 stimulation in intimal smooth muscle cells is sufficient to increase the expression of cyclin D, but it is not sufficient to activate the cyclin D-CDK 4 complex.

View larger version (43K): [in this window] [in a new window]   Fig. 5.   Rb phosphorylation and CDK 2 activity following FGF stimulation. A, Western blot analysis shows that FGF2 stimulation increases the phosphorylation of Rb in medial smooth muscle cells but not in intimal smooth muscle cells. B, FGF2 stimulation does not activate CDK 2 in intimal smooth muscle cells. An in vitro kinase assay, using histone H1 as a substrate, shows that CDK 2 activity is not significantly increased in FGF2-stimulated intimal smooth muscle cells but is stimulated in FGF2-stimulated medial smooth muscle cells.

We next asked whether CDK 2 was activated by FGF2 stimulation of smooth muscle cells. Cyclin-CDK 2 complexes were immunoprecipitated with an antibody specific for CDK 2. An in vitro kinase assay, using histone H1 as a substrate, showed that the cyclin-CDK 2 complex was strongly activated at 48 h after FGF2 stimulation of medial smooth muscle cells, but only minimal activity was observed in FGF2-stimulated intimal smooth muscle cells (Fig. 5B).

The observation that CDKs 4 and 2 were not activated despite the expression of cyclin D and cyclin E prompted us to examine the expression of Cdc25A and CDK inhibitors. The phosphatase Cdc25A is required for the activation of CDK 4 and CDK 2, while CDK inhibitors inhibit the activity of CDKs 4 and 2 in the presence of the cyclins. Fig. 6 shows that Cdc25A is expressed in normal carotid arteries and FGF stimulation of medial smooth muscle cells increases its expression. Intimal smooth muscle cells express lower levels of Cdc25A, and FGF2 stimulation did not increase the expression of this protein.

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Expression of Cdc25 after FGF2 stimulation. Medial smooth muscle cells express higher levels of Cdc25A than do intimal smooth muscle cells. FGF2 stimulation increases this expression in medial smooth muscle cells but has little effect on the expression of Cdc25A in intimal smooth muscle cells.

We next examined the expression of the cell cycle inhibitors p21^Cip and p27^Kip1. These inhibitors bind to and inhibit the activity of the cyclin D-CDK 4 and the cyclin E-CDK 2 complexes. Western blot analysis showed that p27^Kip1 was expressed in both normal vessels and vessels with intimal lesions, with the level of expression being higher in the intimal lesions than in normal vessels (Fig. 7). FGF2 stimulation did not decrease the expression of p27^Kip1 in arteries with established intimal lesions (Fig. 7). p21^Cip was not expressed in normal carotid arteries, but increased expression of p21^Cip was observed after FGF2 stimulation of medial smooth muscle cells. Likewise, FGF2 stimulation caused a similar increase in the expression of p21^Cip in intimal smooth muscle cells.

View larger version (46K): [in this window] [in a new window]   Fig. 7.   Expression of CDK inhibitors after FGF2 stimulation. Western blot analysis shows that intimal smooth muscle cells express higher levels of p27Kip1 and p15INK4b than do normal uninjured carotid arteries or FGF-stimulated medial smooth muscle cells.

Another family of inhibitors, the INK4 family, bind to CDK 4 and prevent its association with cyclin D. Examination of the expression of p15^INK4b showed that intimal smooth muscle cells express higher levels of this inhibitor than do medial smooth muscle cells and that this expression is increased after FGF2 stimulation (Fig. 7). Western blot analysis failed to detect expression of p16^INK4a (data not shown). Thus, our data show that FGF2-stimulated intimal smooth muscle cells do express CDK 4, cyclin D, CDK 2, and cyclin E, but the activity of the cyclin-CDK complexes is very low. This lack of activity correlates with low levels of Cdc25A and high levels of the CDK inhibitors p27^Kip1 and p15^INK4b in smooth muscle cells in established intimal lesions.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

FGF2 is a potent mitogen for medial smooth muscle cells and is required for their proliferation following arterial injury. In contrast, FGF2 does not seem to be required for intimal cell proliferation, since blocking antibodies to FGF2 do not inhibit intimal smooth muscle cell proliferation after injury (5, 37). Further, previous studies have shown that smooth muscle cells in intimal lesions respond weakly to FGF2 stimulation (3). This study demonstrates that there are dramatic differences in the ability of FGF2 to function as a mitogen for smooth muscle cells in a normal artery as compared with smooth muscle cells of an established intimal lesion. Although intimal smooth muscle cells respond only weakly to FGF2, they have not lost their ability to replicate, since if subjected to another balloon injury, the resultant replication can exceed 25% (37). These observations suggested that intimal smooth muscle cells suffer from a specific defect in FGF2 responsiveness rather than a general inability to proliferate.

The observation that FGF2 stimulates ERK activation in intimal smooth muscle cells to a similar degree as in medial smooth muscle cells suggests that the ability of FGF2 to bind to its receptor and activate cytoplasmic signaling pathways is not compromised in these cells. The activation of ERK 1 and 2 is known to be an early event in the stimulation of FGF receptors by FGF (35, 38, 39), and ERK activation is required for many FGF2-mediated cellular responses including proliferation (35). ERK activity increases after balloon catheter injury of the rat carotid artery and is, in fact, required for medial smooth muscle cell proliferation after balloon injury (10). Our data demonstrate that FGF2 is able to activate a cytoplasmic signaling pathway associated with smooth muscle cell proliferation in intimal smooth muscle cells and that the magnitude and duration of the activation is similar in intimal and medial smooth muscle cells. This observation is important because, although ERK activation is associated with proliferation, ERK activation does not necessarily lead to proliferation, and in some cell types ERK activation is associated with differentiation and inhibition of proliferation (40). How activation of the ERK pathway can result in such diverse cellular responses is not entirely understood, but it has been suggested the duration of ERK signaling can determine the outcome of ERK activation. An example of this is that in CCL39 cells short term activation of ERKs is not sufficient to stimulate proliferation (35). Sustained activation of ERK for at least 6-8 h is necessary for stimulation of proliferation in these cells. We found no significant difference in the duration of ERK activation in intimal and medial smooth muscle cells after FGF2 stimulation, with ERK being activated for up to 24 h in both cases. Despite this prolonged activation of the ERK signaling pathway, FGF2 induced only a small increase in intimal smooth muscle cell proliferation compared with medial smooth muscle cells.

FGF2 is capable of activating other signaling pathways that may be necessary for cell proliferation, including the PI 3-kinase pathway (11). To evaluate the effects of FGF2 on PI 3-kinase activity, we measured the phosphorylation of PKB, a downstream target of PI 3-kinase signaling and found that FGF2 stimulated an increase in PKB phosphorylation in both medial and intimal smooth muscle cells. Thus, a difference in PI 3-kinase signaling cannot explain differences in responsiveness to FGF2 in intimal and medial smooth muscle cells. Collectively, these data would therefore suggest that the block in the mitogenic signal in FGF2-stimulated intimal smooth muscle cells is downstream of ERK and PI 3-kinase signaling.

To try to identify this downstream inhibition, we examined the expression of cyclin D. Cyclin D expression is regulated by growth factors, increasing early in the G₁ phase, and is critical for most proliferative signals (41-43). Recently, PI 3-kinase (14) and the ERKs (7) have been found to regulate the expression of cyclin D, thus providing a direct link between growth factor-mediated signaling and initiation of the cell cycle. Our data show that cyclin D expression increases following FGF2 stimulation of both intimal and medial smooth muscle cells, suggesting that the FGF2-stimulated signaling is sufficient to increase the expression of cyclin D. This observation also demonstrates that intimal smooth muscle cells enter the cell cycle following FGF2 stimulation. We also noted that both cell types express cyclin E, CDK 4, and CDK 2; however, this appears to be constitutive, since FGF2 did not increase the expression of any of these proteins. Although intimal smooth muscle cells express cyclin D-CDK 4 and cyclin E-CDK 2 following FGF2 stimulation, the absence of cyclin A expression, Rb phosphorylation, and CDK 2 activation suggests that the cyclin-CDK complexes are not active in these cells. Active cyclin D-CDK 4 phosphorylates Rb, freeing active E2F. This, along with activation of cyclin E-CDK 2, leads to increased expression of cyclin A. These events are necessary for cells to progress through the G₁ phase and into S phase. Intimal smooth muscle cells showed neither Rb phosphorylation, CDK 2 activation, nor expression of cyclin A following FGF2 stimulation. We believe that these data support the hypothesis that intimal smooth muscle cells did enter the cell cycle following FGF2 stimulation but that progression through the G₁ phase into S phase was blocked due to an inability to activate CDK 4 and CDK 2.

In this report, we have identified several factors that could account for the inhibition of cyclin D-CDK 4 and cyclin E-CDK 2 activity in the smooth muscle cells from arteries with established intimal lesions. In addition to association with the appropriate cyclin, CDKs require dephosphorylation of inhibitory phosphorylation sites for activation (44). The phosphatase Cdc25A is thought to be responsible for this dephosphorylation, and its activity is necessary for proliferation (45). Our results show that while the expression of Cdc25A increases after FGF2 stimulation of medial smooth muscle cells, it does not increase after stimulation of intimal smooth muscle cells. Further, this increased expression of Cdc25A in the medial smooth muscle cells correlates with increased activity of CDK 4 and CDK 2. One possibility, therefore, is that the low level of Cdc25A expression in FGF2-stimulated intimal smooth muscle cells contributed to the lack of CDK 4 and CDK 2 activity in these cells and hence attenuated proliferation.

Specific inhibitors can also regulate the activity of CDK 4 and CDK 2. The p15^INK4b family of inhibitors binds to CDK 4, preventing its association with cyclin D and thus inhibiting activation of CDK 4 (27, 28). Interestingly, we found high levels of expression of p15^INK4b in intimal smooth muscle cells but not in medial smooth muscle cells, which may explain the apparent lack of CDK 4 activity in intimal smooth muscle cells expressing a high level of cyclin D expression. Another family of CDK inhibitors includes p27^Kip1, p21^Cip, and p57^Kip1. Members of this family are more promiscuous and inhibit the activity of cyclin D-CDK 4, cyclin E-CDK 2, and cyclin A-CDK 2 (23-26). Surprisingly, the expression of p21^Cip increased in intimal smooth muscle cells as well as medial smooth muscle cells after FGF2 stimulation. This result is at first puzzling, since higher expression of this cell cycle inhibitor correlates with higher CDK activity and proliferation in medial smooth muscle cells. Although expression of p21^Cip can result in inhibition of proliferation, there is now evidence that low level expression of p21^Cip actually promotes CDK 4 activity. We cannot say whether the level of p21^Cip expression in smooth muscle cells is inhibitory or stimulatory, but what is clear is that the expression of p21^Cip in FGF2-stimulated medial smooth muscle cells correlates with increased activity of CDK 4 and CDK 2 and increased proliferation. In contrast, the level of expression of the inhibitor p27^Kip1 did correlate with reduced CDK activity, with intimal smooth muscle cells expressing higher levels of p27^Kip1 than medial smooth muscle cells. The inhibitor p27^Kip1 is normally expressed in quiescent cells, and its expression is down-regulated upon mitogenic stimulation (31, 41). Overexpression of p27^Kip1 has been shown to inhibit intimal lesion formation (20), and blocking the expression of p27^Kip1 increased the sensitivity of cultured cells to mitogenic stimulation, including FGF (31). Our finding that arteries with established intimal lesions express high levels of two CDK inhibitors, p15^INK4b and p27^Kip1, might well explain the lack of CDK activity and the weak proliferative effects of FGF2 in these cells.

Collectively, these data suggest that the cytoplasmic signaling elicited by FGF2, while sufficient to up-regulate cyclin D expression in intimal smooth muscle cells, is not sufficient to overcome the elevated levels of CDK inhibitors. Pertinent to these findings are data linking PI 3-kinase (46, 47) and ERK (48, 49) to p27^Kip1 regulation in some cell types. Our data, however, show that elevated p27 levels are maintained despite activation of both ERK and PI 3-kinase in intimal smooth muscle cells, suggesting that neither PI 3-kinase nor ERK signaling regulates CDK inhibitor levels in smooth muscle cells.

There are several possible explanations for the increased expression of p27^Kip1 in intimal smooth muscle cells. Interactions with different extracellular matrices have been shown to affect the levels of cell cycle inhibitors, and interestingly, Koyama et al. found that cultured smooth muscle cells grown on polymerized type 1 collagen were less responsive to the mitogenic effects of platelet-derived growth factor (50). This reduced responsiveness was attributed to increased levels of cell cycle inhibitors in cells grown on polymerized collagen (50). We cannot confirm that changes in extracellular matrix are responsible for the increased expression of p27^Kip1, but increased expression of several extracellular matrix components, including type 1 collagen, has been demonstrated after arterial injury (51). Related to this observation are data showing that integrins can also affect p27^Kip1 expression; Murgia et al. have found proliferation defects associated with increased expression of p27^Kip1 in mice carrying a targeted deletion of the ₄-integrin cytoplasmic domain (52). Differences in the extracellular matrix and/or integrin expression may be responsible for changes in the level of p27^Kip1 expression and thus may provide a mechanism by which smooth muscle cells can modulate their response to growth factors.

Although the expression of p15^INK4b, p27^Kip1, and Cdc25A in intimal smooth muscle cells may be regulated by different factors, one factor, TGF-, could be responsible for regulating the expression of all of these proteins. TGF- has been shown to increase the expression of p27^Kip1 (31, 53) and is well known as an inhibitor of proliferation in many cell types, including smooth muscle cells (54). In addition to p27^Kip1, TGF- has been shown to also regulate the expression of p15^INK4b and Cdc25A (55-57). TGF- increases the expression of p15^INK4b in many different cell types, and repression of Cdc25A expression has recently been shown to be a part of TGF- inhibition in a human mammary epithelial cell line (57). It is tempting to speculate that TGF-1 may be responsible for decreased responsiveness to FGF2 found in arteries with established intimal lesions, since these cells express higher levels of TGF- than do cells in normal uninjured arteries (58, 59). Studies to examine this question are currently in progress.

This work shows that FGF2 is weakly mitogenic for smooth muscle cells in intimal lesions. Our data would support the conclusion that FGF2 stimulates prolonged activation of ERKs 1/2 as well as activation of PI 3-kinase and that this activation is similar to that seen in acutely injured medial smooth muscle cell stimulated with FGF2. Additionally, the signaling elicited by FGF2 in intimal smooth muscle cells is sufficient to stimulate an increase in the expression of cyclin D, indicating that these cells do enter the cell cycle. Further, intimal smooth muscle cells also express cyclin E, CDK 4, and CDK 2. The expression of cyclins D and E, however, was not sufficient to induce a high level of proliferation, and we hypothesize that high levels of p15^INK4b and p27^Kip1 inhibited the activity of cyclin D-CDK 4 and cyclin E-CDK 2, thus attenuating the proliferation of these intimal smooth muscle cells. Others have observed increased expression of cell cycle inhibitors after arterial injury (20, 29, 60) and suggested that this increased expression contributes to the cessation of the proliferative response after injury. This study supports those observations and provides evidence that the increased expression of cell cycle inhibitors after injury inhibits the ability of those cells to respond to subsequent mitogenic stimuli.

	FOOTNOTES

^* Supported by National Institutes of Health (NIH) Grants HL03174 and HL41103 and NIH Training Grant HL07312 (to N. E. O.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Dept. of Pathology, Box 357335, University of Washington, Seattle, WA 98195. Tel.: 206-616-5948; Fax: 206-685-3662; E-mail: olsonne@u.washington.edu.

	ABBREVIATIONS

The abbreviations used are: FGF, fibroblast growth factor; PI, phosphatidylinositol; CDK, cyclin-dependent kinase; ERK, extracellular signal-regulated kinase; Rb, retinoblastoma protein; PKB, protein kinase B; SMC, smooth muscle cell; INK4, inhibitor of CDK4.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES