Skip Navigation

Cardiovascular Research 2003 60(3):673-683; doi:10.1016/j.cardiores.2003.09.018
© 2003 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by O'Sullivan, M.
Right arrow Articles by Bennett, M. R
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by O'Sullivan, M.
Right arrow Articles by Bennett, M. R
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2003, European Society of Cardiology

Differential cyclin E expression in human in-stent stenosis smooth muscle cells identifies targets for selective anti-restenosis therapy

Michael O'Sullivana, Stephen D Scotta, Nicola McCarthya, Nicola Figga, Leonard M Shapirob, Peter Kirkpatrickc and Martin R Bennett*,a

aDivision of Cardiovascular Medicine, Cambridge University, Box 110, Addenbrooke's Hospital, Cambridge CB2 2QQ, UK
bCardiac Unit, Papworth Hospital, Cambridge, UK
cDepartment of Neurosurgery, Addenbrooke's Hospital, UK

*Corresponding author. Tel.: +44-1223-331504; fax: +44-1223-331505. Email address: mrb{at}mole.bio.cam.ac.uk

Received 24 June 2003; revised 27 August 2003; accepted 15 September 2003


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objective: Cell cycle inhibitors are promising agents to prevent or treat human coronary in-stent stenosis (ISS). However, their lack of specificity for ISS vascular smooth muscle cells (VSMCs) may inhibit medial VSMC proliferation and suppress vessel healing. Methods: To identify inhibitor targets that differentially regulate proliferation of ISS vs. medial VSMCs, we examined cell cycle regulation in human VSMCs derived from (A) normal media, (B) ISS sites and (C) primary atherosclerotic plaques (P-VSMCs) using time-lapse videomicroscopy, flow cytometry, immunoblotting and immunohistochemistry. Results: ISS-VSMC proliferation was intermediate between P-VSMCs and medial VSMCs. Compared with medial cells, P-VSMCs expressed increased p16 and p21, reduced p27, reduced cyclins D1 and E, and reduced pRb phosphorylation. In contrast, ISS-VSMCs expressed high levels of cyclins E and A with pRb hyperphosphorylation, both in vitro and in vivo, associated with increased and chronic cell proliferation in vivo. Roscovitine, a selective CDK2 inhibitor, inhibited VSMC proliferation by both pRb-dependent and independent pathways and more potently in ISS-VSMCs than medial VSMCs. Conclusions: Human ISS-VSMCs have marked differences in the stable expression of multiple cell cycle regulators, suggesting that ISS-VSMCs derive from P-VSMCs driven to proliferate through cyclin E overexpression. The critical role for cyclin E–CDK2 enables the identification of the first agent that selectively inhibits ISS-VSMC proliferation.

KEYWORDS Restenosis; Smooth muscle; Atherosclerosis


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
Intracoronary stents reduce the incidence of restenosis following percutaneous coronary intervention, but in-stent stenosis (ISS) remains an important problem [1]. The main contributor to ISS is neointima formation, resulting from migration and proliferation of vascular smooth muscle cells (VSMCs), followed by deposition of extracellular matrix [2]. While the kinetics and regulation of VSMC proliferation in animal vascular injury models are well known [3], the regulation of VSMC proliferation at human coronary ISS sites is unknown, with no studies examining cells cultured from the human ISS lesion. Many strategies inhibit proliferation of normal human or animal VSMCs in vitro, or neointima formation following vascular injury, but few have translated into useful human therapies [4]. This suggests that there are fundamental differences between the mechanisms of cell cycle regulation in VSMCs that generate the ISS neointima and ‘normal’ medial VSMCs. Such differences could identify targets for selective inhibition of ISS-VSMCs while sparing other VSMCs, contributing both to improved efficacy and vessel healing.

Cell cycle transit is regulated at a number of checkpoints, the best characterised of which is the G1–S transition, inhibited by hypophosphorylated retinoblastoma protein (pRb) [5]. Phosphorylation of pRb by cyclin-dependent kinases (CDKs) 2, 4 and 6 overcomes this inhibitory influence allowing cells to enter S-phase. Cyclin E catalyses CDK2, while cyclins D1–3 catalyse CDK4/6. The INK4 family CDK inhibitors (CDKIs) (p15, p16, 18, p19) inhibit CDK4/6 activity, while Cip/Kip family members (p21, p27 and p57) inhibit CDK2 and CDK4/6. Cyclin A associates with CDK2 to allow negotiation of early S-phase, but also interacts with CDK1 to permit mitosis. Although pRb phosphorylation is important for G1–S transit, there is evidence in some cell types for an alternative mechanism by which CDK2 regulates the G1–S checkpoint [6,7]. In this paper we report that VSMCs from sites of human coronary ISS display a characteristic pattern of cell cycle regulator expression, distinct from primary plaque and normal medial VSMCs. The high expression of cyclins E and A provides insight into how ISS forms and identifies a CDK2 inhibitor that selectively inhibits ISS-VSMCs. We also show that CDK2 inhibition is effective downstream of pRB at multiple cell cycle points. The identification of selective inhibitors of ISS-VSMCs has important implications for the development of future anti-restenotic therapies.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1. Human tissue collection and cell isolation
Human tissue collection was approved by the Local Ethical Committee, and consent obtained from patients before tissue harvest. The investigation conforms with the principles outlined in the Declaration of Helsinki (Cardiovascular Research 1997;35:2–4). Normal coronary arteries were explanted from patients undergoing cardiac transplantation for nonischaemic cardiomyopathy, ISS specimens by directional coronary atherectomy from sites of angiographically confirmed ISS, and plaque specimens by carotid endarterectomy. Normal human coronary medial or primary plaque VSMCs were isolated by explant outgrowth [8,9]. ISS-VSMCs were obtained by collagenase/elastase digestion of specimens [9]. Cells were studied at passages 2–5; VSMC cultures from individual patients were not pooled. Cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 20% fetal calf serum (FCS).

2.2. Time-lapse videomicroscopy
Proliferation rates were determined for 50% confluent, nonoverlapping cells using time-lapse videomicroscopy [10]. Briefly, cells were washed three times in medium containing 20% FCS and then cultured in this medium±roscovitine. Flasks were gassed with 95% air and 5% CO2 every 24 h and sealed. The microscope was enclosed in a plastic environment chamber and maintained at 37 °C by an external heater. Cell division was scored at the time at which septa appeared between two daughter cells using an observer blind to cell type and treatment conditions.

2.3. Flow cytometry
Cell cycle distribution was determined by propidium iodide (PI) staining and flow cytometry [11].

2.4. Western blotting
Protein isolation, electrophoresis and blotting were performed using the following antibodies: mouse anti-pRb 1/500, anti-cyclin A 1/1000, anti-cyclin D 1/1000, anti-cyclin E 1/1000, anti-p21 1/1000, anti-p27 1/1000 and rabbit anti-p16 1/1000 (all from Pharmingen). Horseradish peroxidase-linked anti-mouse or anti-rabbit antibodies (Amersham) were used at 1/1000.

2.5. Immunohistochemistry
Five-micrometre formalin-fixed, paraffin-embedded sections were de-waxed in xylene and rehydrated in graded ethanol. After microwaving with Antigen Retrieval Solution (Vector Labs, California) (p21, p27 and Ki-67) or 0.01 M citrate buffer pH 6.0 (cyclins A and E) for 10 min, slides were equilibrated in PBS for 5 min and incubated for 10 min in 0.3 % hydrogen peroxide, washed in water and PBS. After blocking in 5% horse serum in PBS (p21, p27 and Ki-67) or 2% goat serum in PBS (cyclins A and E) for 30 min, primary antibody was applied for 1–2 h at 1/100 in PBS with 0.1% horse serum (anti-p21 and anti-p27 -Pharmingen), or 1/50 in 0.1% goat serum in PBS (anti-Ki-67, anti-cyclin A, anti-cyclin E; Pharmingen). Slides were washed in PBS, and biotinylated rabbit anti-mouse antibody (Vector Labs, 1/400 in PBS) applied for 30 min. After washing in PBS, sections were incubated in avidin–biotin complex (Vector Labs) for 30 min, then washed in PBS. Diaminobenzidine was applied for 1–5 min, slides were washed in water and counterstained with Harris's Haematoxylin (Sigma) for 5 min. After passage through destain solution for 10 s, slides were left in Scott's solution for 5 min, washed in water and dehydrated in graded ethanol.

2.6. Production of retrovirus-infected VSMC lines
Rat VSMCs isolated from medial thoracic aortic explants [11] were cultured in DMEM containing 10% FCS. Infectious retroviruses were produced by transfection of BOSC cells with pBabepuro proviral DNA encoding wild-type pRb (RbWT), a nonphosphorylatable form of pRb (Rb34) [12] or vector control as described [10]. Semi-confluent VSMCs at passage 4 were incubated for 2 h in virus-containing medium supplemented with 8 mg/l hexadimethrine bromide, washed in DMEM, and 10 µg/l puromycin added 24 h later and maintained throughout. Resistant cell populations were studied at least 6 weeks after infection. The generation of rat VSMCs expressing the adenoviral E1A protein has been reported [13].

2.7. Preparation of roscovitine solutions
Stock solution of 5 mmol/l Roscovitine (AG Scientific) was prepared in DMSO and stored at –20 °C. The highest concentration used was 40 µmol/l, corresponding to 0.8% DMSO.

2.8. Statistical analysis
Results are expressed as mean±SEM. Analysis of significance was by Student's t test or Mann–Whitney U test.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1. Culture of human medial, plaque and ISS-VSMCs
Coronary in-stent stenosis (ISS) VSMCs were derived from atherectomy specimens obtained 6–9 months after stent implantation from patients undergoing revascularisation for stable angina. Normal medial cells were derived from coronary arteries of patients with nonischaemic cardiomyopathy undergoing heart transplantation. Plaque (P) VSMCs were derived from carotid endarterectomy specimens. Three normal VSMC cultures (LAD I– III), three ISS-VSMC cultures (HCR3-5) and three P-VSMC cultures (C20, C36, C41) were studied. Immunocytochemistry demonstrated that all cells stained positive for {alpha}-smooth muscle actin, smooth muscle myosin and calponin, indicating their VSMC origin (not shown).

3.2. ISS-VSMC proliferation is intermediate between normal and plaque VSMCs
VSMC proliferation was examined using time-lapse videomicroscopy. Fig. 1 demonstrates that ISS-VSMCs proliferated more slowly than normal medial VSMCs (p<0.002), but more rapidly than primary plaque VSMCs (p<0.02). ISS-VSMCs failed to proliferate beyond P9 and displayed typical senescence characteristics (stellate, flat, vesiculated appearance) at this point or earlier. Medial cells survived culture beyond P10 without evidence of senescence, whereas primary plaque VSMCs senesced by P5. Thus, both proliferation and senescence of ISS-VSMCs were intermediate between medial and P-VSMCs.


Figure 1
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Proliferation of normal, P-VSMCs and ISS-VSMCs. Growth curves for normal medial VSMCs (LAD I, II, III), ISS-VSMCs (HCR 3, 4, 5) and P-VSMCs (C20, C36, C41). Data for medial VSMCs are shown by filled symbols connected by solid lines, for ISS-VSMCs by filled symbols connected by broken lines, for P-VSMCs by open symbols connected by solid lines. Number of divisions is expressed as % of original cell count. (Points represent mean±SEM, n = 3.)

 
3.3 ISS-, plaque and medial VSMCs have characteristic patterns of expression of G1–S regulators
To examine whether differences in cell cycle regulation underlie these differences in cell proliferation, we examined the expression of the G1–S cell cycle regulators pRb, cyclins D, E and A, and p16, p21 and p27 in 10% FCS medium or after 24 h serum starvation (Fig. 2). G1–S regulator levels were strikingly consistent between same cell type VSMC cultures, but differed markedly between ISS and normal medial VSMCs. Compared with medial VSMCs, ISS-VSMCs showed no Cyclin D1 expression, but increased expression of cyclins E and A. p16 expression was slightly increased and p21 expression markedly increased in ISS vs. medial VSMCs, whereas p27 expression was reduced. A greater proportion of pRb was hyperphosphorylated in ISS vs. medial cells, in which the bulk of pRb was phosphorylated or hypophosphorylated. Serum deprivation increased p27 expression in medial VSMCs although not in ISS-VSMCs, which expressed p27 at very low levels. Thus, cultured ISS-VSMCs express a characteristic pattern of G1–S regulators that differs markedly from normal medial cells.


Figure 2
View larger version (66K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Expression of cell cycle regulators in cultured coronary medial VSMCs (LAD I–III) and ISS-VSMCs (HCR 3–5). Lysates were prepared when cells were growing in serum (+) or after serum deprivation for 24 h (–). The positions of hyperphosphorylated, phosphorylated and hypophosphorylated pRb are shown; the position of cyclin D1 is indicated by an open arrowhead.

 
Expression of G1–S proteins was also analysed in two cultures of P-VSMCs. Fig. 3 shows that P-VSMCs, like ISS-VSMCs, express increased p16 and p21, reduced p27 and no detectable cyclin D1. However, P-VSMCs differ from ISS-VSMCs by expressing low levels of cyclins E and A, hyperphosphorylated pRb is not detected, and a greater proportion of pRb is hypophosphorylated. This pattern of expression strongly suggests that ISS-VSMCs are derived from primary plaque VSMCs driven to proliferate by cyclin E/CDK2.


Figure 3
View larger version (79K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Expression of cell cycle regulators in medial VSMCs and P-VSMCs. Lysates were prepared from two populations of normal medial VSMCs and two populations of P-VSMCs growing in serum-containing medium. The positions of hyperphosphorylated, phosphorylated and hypophosphorylated pRb are shown; the position of cyclin D1 is indicated by an open arrowhead.

 
3.4. In vivo expression of cell cycle regulators within ISS specimens and normal coronary arteries
To determine whether the differences in cyclins A, E, p21 and p27 expression observed in cultured VSMCs were also seen in vivo, expression was determined in ISS atherectomy specimens from six patients [median stent to atherectomy interval 11 months (range 5–15 months)] and media and intima of normal coronary arteries (Fig. 4). 3200–7150 nuclei for each atherectomy specimen and 1600–2200 nuclei for normal vessel intima were quantified within the whole atherectomy specimen or within the internal elastic lamina of normal coronary arteries. Over 98% of cells within ISS atherectomy samples were VSMCs, as indicated by {alpha}-smooth muscle actin staining (not shown). Compared with normal vessel intima, there was significantly increased expression of cyclins E and A (cyclin E 20.3±4.4% vs. 0%, cyclin A 1.9±0.9% vs. 0.03±0.01%), and reduced expression of p27 (0.47±0.19% vs. 13.9±1.9%) in ISS intima. Although p21 was also increased in ISS neointima (0.44±0.15% vs. 0%), expression levels were low and this difference did not attain significance. Cyclin A, cyclin E and p21 expression was not detectable in normal media, whereas p27 was detectable in most medial cells. 1.73% (±0.61%) of VSMCs within ISS lesions were Ki-67 positive, a marker of cell proliferation, whereas only endothelial cells stained positive in normal vessels (Fig. 4M). Thus, ISS sites showed evidence of ongoing or recent VSMC proliferation many months after stenting, a feature not seen within the normal artery, associated with increased expression of positive regulators (cyclins E and A) and reduced expression of negative regulators (p27) compared with normal vessels.


Figure 4
View larger version (73K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Expression of cell cycle regulators in VSMCs within normal coronary intima and ISS specimens. Expression of cyclin A (A, B, C), cyclin E (D, E, F), p21 (G, H, I), p27 (J, K, L) and Ki-67 (M, N, O) in normal coronary arteries (A, D, G, J, M) and ISS neointima (B, E, H, K, N). Expression within intimal VSMCs was quantified by counting positive nuclei (C, F, I, L, O). Six samples were analysed per specimen type per antibody. (Values are mean±SEM.)

 
3.5. ISS-VSMCs are more sensitive to roscovitine than medial VSMCs
The expression patterns of G1–S regulatory proteins in plaque, normal and ISS-VSMCs suggest that cyclin E/CDK2 may be particularly important in driving ISS-VSMC proliferation, and that ISS-VSMCs may be especially sensitive to cyclin E–CDK2 inhibition. We therefore studied the effect of roscovitine, a potent and selective inhibitor of CDK2 [14] upon proliferation of normal and ISS-VSMCs. Cells were cultured in full medium supplemented with roscovitine (10–40 µmol/l) or DMSO (0.2–0.8% v/v). Roscovitine (30–40 µmol/l) induced apoptosis in ISS-VSMCs, but cell death was not observed at 10–20 µmol/l. Roscovitine caused a dose-dependent and statistically significant anti-proliferative effect in both normal and ISS cells (p<0.05) (Fig. 5A and B). Importantly, ISS-VSMCs were significantly more sensitive than medial cells to its anti-proliferative effect (IC50 of 6.8 µM vs. 25.0 µmol/l).


Figure 5
View larger version (33K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Effect of roscovitine upon proliferation of medial and ISS-VSMCs and cell cycle transit in ISS-VSMCs. (A,B) Normal (LAD III) and ISS (HCR 4) VSMCs were exposed to roscovitine at various concentrations or equivalent concentrations of vehicle (DMSO) and percentage inhibition of proliferation after 24 h (as compared to vehicle control) at each roscovitine concentration was determined by MTT assay (A) or time-lapse videomicroscopy (B). (C–E) ISS (HCR4) VSMCs were growth arrested by serum deprivation for 72 h and were then stimulated with 20% FCS in the presence of DMSO 0.4% (v/v), roscovitine 10 µmol/l or roscovitine 20 µmol/l and harvested at the time-points shown. Cell cycle distribution was determined by PI staining and flow cytometry. (F) Time-course of passage from G0/G1 to S/G2M, as indicated by % of cells in S+G2+M, is plotted for each condition. (Values are mean±SEM, n = 3.)

 
3.6 Roscovitine causes G1 and G2/M arrest of ISS-VSMCs
To determine the effect of roscovitine upon cell cycle transit in ISS-VSMCs, ISS cells were growth-arrested in 0.2% FCS for 72 h. Cells were then serum-stimulated and DMSO 0.4%, roscovitine 10 µmol/l or 20 µmol/l added. Cells were harvested at intervals to 60 h after serum-stimulation and cell cycle distribution was studied by flow cytometry. Fig. 5C–F shows that roscovitine delayed the passage of cells from G0/G1 to S/G2/M. A reduction in return of cells from G2/M to G1 was also observed indicating a G2 block—these effects were dose-dependent. Thus, roscovitine inhibits transit through both the G1–S and G2/M checkpoints in ISS-VSMCs.

3.7. The anti-proliferative effect of roscovitine is in part independent of pRb
The dual action of roscovitine on G1–S and S–G2 suggests both pRB-dependent and -independent effects. To examine the requirement of pRB for action of roscovitine, we generated rat VSMCs that expressed wild-type pRb (RbWT cells), constitutively active pRb (Rb34 cells), or adenoviral E1A 12S protein, which binds and inactivates pRb (VSM-E1A cells) and promotes a pRB- and CDK2-independent cell cycle progression [15,16] or vector control by retroviral infection. Enhanced expression of these proteins was demonstrated by Western blotting (not shown) and incorporation of the proviral DNA confirmed by PCR using vector-specific primers (not shown).

Time-lapse videomicroscopy demonstrated that roscovitine caused a dose-dependent inhibition of proliferation in all cell types (Fig. 6A–C), although RbWT and Rb34 cells were more sensitive than control cells (IC50 for RbWT 8.5 µmol/l, Rb34 6.7 µmol/l, control 26.2 µmol/l). Flow cytometric analysis showed that in all cell types, 20 µmol/l roscovitine caused a significant increase in G0/G1-phase and a reduction in S-phase fraction (Fig. 6D), indicating a G0/G1 arrest. Although overall roscovitine treatment had no significant effect on the G2/M fraction, the lack of diminution of G2/M fraction after 24 h of treatment with roscovitine also suggests a block to G2/M exit.


Figure 6
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Effect of roscovitine upon proliferation and cell cycle transit of Rb-infected VSMCs. Growth curves of (A) control, (B) RbWT and (C) Rb34 cells exposed to various concentrations of roscovitine (µmol/l) or DMSO 0.8% v/v (vehicle). (D) Unsynchronised control cells, RbWT cells and Rb34 cells were cultured in full medium containing DMSO 0.4% (v/v) or roscovitine 20 µmol/l for 24 h prior to determination of cell cycle profile by PI staining and flow cytometry. (*p<0.05, {dagger}p<0.001, {ddagger}p<0.0005, as compared to DMSO control).

 
3.8. The anti-proliferative effect of roscovitine does not require functional pRb
The increased sensitivity of VSMCs expressing high levels of WT or constitutively active RB to roscovitine suggests that CDK2 inhibition by roscovitine acts additively and in part independent to pRB. We therefore examined the effect of roscovitine upon VSM-E1A cells whose pRB function is disrupted. Time-lapse videomicroscopy demonstrated that VSM-E1A proliferation was similarly inhibited by roscovitine (Fig. 7A). This effect was dose-dependent with IC50 17.5 µmol/l. Flow cytometry showed that exposure of VSM-E1A cells to roscovitine caused a significant and dose-dependent diminution of S-phase fraction with a significant increase in G0/G1 fraction but no significant change in G2/M fraction (Fig. 7B). Thus, roscovitine blocks both S-phase entry and G2/M exit in VSM-E1A cells, indicating that CDK2 inhibition may block both G1–S and G2–M transit independently of pRb function.


Figure 7
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Effect of roscovitine upon proliferation and cell cycle distribution of VSM-E1A cells. (A) Cumulative divisions of VSM-E1A cells exposed to roscovitine at various concentrations or equivalent concentration of vehicle (DMSO). (B) Cell cycle distribution (% of cells in each cell cycle phase), as determined by PI-staining and flow cytometry, of VSM-E1A cells exposed to roscovitine or DMSO for 24 h. Roscovitine concentrations are expressed in µmol/l, DMSO concentration was 0.8% v/v. (*p<0.05, {dagger}p<0.005, {ddagger}p<0.0001.)

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
Despite the fact that VSMC proliferation is critically important in the animal response to vascular injury [3] and human coronary in-stent stenosis (ISS), the regulation of VSMC proliferation in ISS is largely unknown. In addition, prior to the advent of drug-eluting stents using rapamycin or paclitaxel, many strategies that limited the proliferation of human and animal medial VSMCs in culture, or inhibited neointima formation in animal models, had been unsuccessful in man [2,4]. Although deficiencies within animal models and technical difficulties with safe delivery of therapies in humans may partly underlie this failure, it may also reflect differences in the regulation of proliferation between ISS and normal medial VSMCs. Such differences could be exploited therapeutically and also used to limit the inhibition of medial VSMCs responsible for vessel healing. We therefore studied cell cycle regulation in VSMCs cultured from human coronary ISS lesions, compared with cells derived from primary carotid plaques and normal aorta. Although carotid and coronary VSMCs may arise from different embryological origins, we have previously found that coronary and other large human artery VSMCs proliferate at similar rates in culture [17].

We found that ISS-VSMCs proliferate less rapidly and senesce earlier than normal medial VSMCs but proliferate more rapidly and senesce later than P-VSMCs. This intermediate proliferative phenotype correlates with a distinct pattern of expression of G1–S checkpoint regulators, such that when compared with medial VSMCs, ISS-VSMCs express increased p16, p21, cyclins E and A, but reduced p27, associated with enhanced pRb phosphorylation. P-VSMCs exhibit a similar pattern of CDKI expression to ISS-VSMCs, but express lower cyclins E and A and a higher proportion of hypophosphorylated pRb.

Importantly, the differences in expression pattern of these proteins encountered in vitro paralleled the differences found in vivo using immunohistochemistry. Thus, significantly more VSMCs within ISS tissue expressed cyclins E and A and fewer p27 compared to normal coronary media or intima, with a trend towards increased p21 expression within ISS neointima. These results are consistent with previous reports that p27 is expressed abundantly within normal human coronary arteries, while p21, cyclin E and cyclin A are barely detectable [18–20]. Furthermore, pockets of VSMCs that express cyclin A and cyclin E have been demonstrated within atherectomy specimens from sites of nonstent restenosis [19]. These findings argue strongly that the high cyclin E, cyclin A and p21 expression and low p27 expression in ISS-VSMCs in vitro reflect their in vivo phenotype and are inherent characteristics of the ISS-VSMC.

The simplest explanation for the emergence of the ISS-VSMC phenotype, and the similarity of CDKI expression in ISS-VSMCs and P-VSMCs, is that intracoronary stenting induces cyclin E expression and pRb phosphorylation in P-VSMCs, triggering proliferation. According to this model, the ISS-VSMC is derived from the P-VSMC that has been driven to proliferate by cyclin E overexpression.

Unfortunately, it is not currently possible to demonstrate directly that human ISS-VSMCs derive from P-VSMCs in vivo—while experiments examining the time-course of expression of cell cycle regulators could address this issue, it is not feasible to obtain sufficient human ISS samples at earlier time points, as earlier lesions do not usually merit atherectomy. The age of the ISS samples studied may, however, have resulted in the underestimation of the differences between normal and ISS-VSMCs as differences in the expression of cell cycle regulators detected several months after stenting are likely to be less than those detected at earlier time points. The persistence of a distinct pattern of expression of cell cycle regulators in ISS-VSMCs at later time points argues strongly for an important role for this cell type in the pathogenesis of this condition.

Importantly, the increased cyclin E and A expression seen in ISS lesions correlated with chronic, low-level cell proliferation in these lesions, seen months after the stent implantation. The apparent low frequency of proliferating VSMCs within the in-stent neointima, consistent with previous reports [21], does not undermine the importance of VSMC proliferation in ISS for three reasons. Firstly, coronary atherectomy samples the ISS neointima but the majority of proliferating cells lie deep within the vessel wall, next to stent struts [21] Secondly, the frequency of proliferating cells declines with time after coronary stenting in animal models [6]. The tissue samples examined in this series were obtained 5–15 months after stent implantation, by which stage rates of proliferation might have fallen. Finally, chronic low-grade proliferation might be adequate to sustain a population of VSMCs sufficient to generate the in-stent lesion. Indeed, the finding that proliferating cells are more frequent in ISS vs. normal intima or media underscores the importance of proliferation in in-stent stenosis. It is noteworthy that ISS-VSMCs proliferate less rapidly than normal VSMCs despite increased cyclin E expression and pRb hyperphosphorylation, and normal VSMCs proliferate more rapidly than both ISS-VSMCs and P-VSMCs despite increased p27 expression. These data underline the importance of the interactions between cyclins and CDKIs in controlling proliferation.

It is well established in other cell types that cyclin E–CDK2 may operate via mechanisms other than pRb phosphorylation [7]. Given the proposed central role of cyclin E in stimulating ISS-VSMC proliferation, and the observation that enhanced pRb phosphorylation in ISS-VSMCs vs. normal cells does not correlate with enhanced proliferation, we determined whether CDK2 may affect G1–S transit via mechanisms separate from pRb phosphorylation in VSMCs. We generated rat VSMCs expressing increased wild-type pRb, constitutively active pRb (pRb34) [12], or adenoviral E1A protein, which inactivates pRb [15]. Infection with a retrovirus vector enables the analysis of prolonged low-level overexpression of an ectopic gene product. Importantly, this system avoids the effects of adenovirus vectors that may induce growth arrest independent of their inserts [22], or nonphysiological levels of proteins. The effects of CDK2 inhibition were analysed in these cell types using roscovitine, a C2,N6,N9-substituted adenine, which inhibits purified CDK2 with IC50 0.70 µmol/l. Roscovitine also inhibits CDK1 (IC50 0.65 µmol/l), CDK5 (IC50 0.16 µmol/l) and CDK7 (IC50 1.4 µmol/l), which may account for its identified actions on G2/M, and basal transcription via RNA polymerase II, but does not significantly inhibit other kinases tested [14].

Whilst cells expressing ectopic pRb (either pRbWT or pRb34) proliferated relatively normally, most likely because of the lower level of expression of these proteins compared with other studies [23], cells were particularly sensitive to the anti-proliferative action of CDK2 inhibition with roscovitine. This indicates that overexpression of active pRb acts additively to roscovitine, suggesting that roscovitine also acts in part independently of pRb. Indeed, flow cytometry demonstrated roscovitine-induced diminution of S-phase percentage with an increase in G0/G1 percentage, with no diminution of G2/M proportion. Thus, as in human fibroblasts [24], roscovitine causes both G1 and G2/M arrest in control, pRbWT and pRb34 VSMCs. While G1 arrest is attributable to inhibition of cyclin E–CDK2, arrest in G2/M may be secondary to inhibition of cyclin B–CDK1 and cyclin B–CDK5 [14,24]. Roscovitine also inhibited the proliferation of VSM-E1A cells, an effect largely due to G1 arrest. Taken together these data indicate that CDK2 inhibition may inhibit VSMC proliferation through a pathway separate from the classical pRb pathway (Fig. 8). This is the first report of a CDK2-dependent pathway that operates separately from pRb at the VSMC G1–S checkpoint.


Figure 8
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8 Model for dual pathway regulation of G1–S checkpoint by CDK2. G1–S transit depends upon phosphorylation of retinoblastoma protein (Rb) by CDK2 and an alternative downstream action of CDK2. Roscovitine (Rosc) inhibits proliferation by inhibition of both CDK2 functions. Low-level overexpression of unphosphorylated Rb, as in Rb34 cells, is insufficient to cause G1 arrest but reduces the degree of CDK2 inhibition required to achieve this effect. Roscovitine inhibits passage through the restriction point in VSM-E1A cells, which lack functional Rb, through inhibition of the Rb-independent CDK2-driven pathway. (P=phosphate group.)

 
Previous studies of the role of CDK inhibitors in VSMCs support but did not prove the existence of such a pathway. Adenoviral transfection of porcine VSMCs with p16, p21 or p27 inhibits cell cycle transit, but this effect is much more marked with p21 and p27 than p16 [25]. Furthermore, transfection with p27 adenovirus at the time of porcine vascular injury inhibits neointima formation whereas p16 adenovirus has no such effect [25]. These data indicate that, of those CDK inhibitors studied in VSMCs, Cip/Kip family members (p21 and p27), which inhibit cyclin E–CDK2 and cyclin D–CDK4, have more important anti-proliferative effects than those of the INK4 family (p16), which inhibit cyclin D–CDK4 alone. These findings support our model in that inhibition of pRb phosphorylation alone, through inhibition of CDK4 activity, is insufficient to prevent VSMC proliferation. Rather, inhibition of CDK2, which may operate through both pRb phosphorylation and alternative pathways, is required. These studies indicate that therapies targeting CDK2 are more likely to be successful in vivo than those targeting pRb alone. Furthermore, these findings suggest that cyclin E overexpression, as seen in human ISS-VSMCs, could drive proliferation both through promoting pRb phosphorylation, as observed, and through separate pathways.

The anti-proliferative effect of roscovitine in rat VSMCs and the proposed dependence of human ISS-VSMC proliferation upon cyclin E–CDK2 activity prompted the study of the effects of this agent upon ISS-VSMC proliferation. We found that roscovitine inhibits the proliferation of ISS-VSMCs more potently than medial VSMCs, again resulting from G1 and G2/M arrest. Although anti-proliferative strategies are attractive for the prevention of stent restenosis, recent clinical studies suggest that they may impair medial healing and re-endothelialisation [2]. Recent trials of drug-eluting stents using anti-proliferative drugs have been very promising [26,27]. However, such stents have not been widely tested for treatment rather than prevention of the ISS lesion. Our demonstration that the ISS-VSMC has enhanced proliferative capacity relative to the P-VSMC may explain the particularly high incidence of recurrent in-stent restenosis following repeat intervention and suggests that alternative treatments may be required in this context. Roscovitine is the first agent shown to selectively inhibit the proliferation of ISS-VSMCs and may be particularly useful in the prevention and treatment of ISS whilst sparing the normal coronary media.

We conclude that human in-stent stenosis VSMCs exhibit a characteristic proliferative phenotype that differs from that of medial and P-VSMCs. This phenotype correlates with abnormalities in cell cycle regulation that enable the identification of targets for selective inhibition of proliferation of in-stent VSMCs. In particular, cyclin E overexpression may underlie the emergence of the ISS-VSMC and may operate through both pRb-dependent and -independent pathways. Roscovitine is the first agent shown to inhibit the proliferation of ISS-VSMCs more potently than that of medial VSMCs, and CDK inhibitors merit further assessment as an anti-restenotic therapy.


   Acknowledgements
 
MO'S is supported by a Wellcome Trust Training Fellowship, SDS by a British Heart Foundation Studentship; MRB is a British Heart Foundation Chairholder.


   Notes
 
time for primary review 21 days


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Gershlick A.H. Role of stenting in coronary revascularisation. Heart (2001) 86:104–112.[Free Full Text]
  2. Bennett M.R., O'Sullivan M. Mechanisms of angioplasty and stent restenosis: implications for design of rational therapy. Pharmacol. Ther. (2001) 91:149–166.[CrossRef][Web of Science][Medline]
  3. Braun-Dullaeus R., Mann M., Dzau V. Cell cycle progression. New therapeutic target for vascular proliferative disease. Circulation (1998) 98:82–89.[Abstract/Free Full Text]
  4. O'Sullivan M., Bennett M.R. Gene therapy for coronary restenosis: is the enthusiasm justified? Heart (2001) 86:491–493.[Free Full Text]
  5. Weinberg R. The retinoblastoma gene protein and cell cycle control. Cell (1995) 81:323–330.[CrossRef][Web of Science][Medline]
  6. Carter A.J., Laird J.R., Farb A., Kufs W., Wortham D.C., Virmani R. Morphologic characteristics of lesion formation and time course of smooth muscle cell proliferation in a porcine proliferative restenosis model. J. Am. Coll. Cardiol. (1994) 24:1398–1405.[Abstract]
  7. Ewen M.E. Where the cell cycle and histones meet. Genes Dev. (2000) 14:2265–2270.[Free Full Text]
  8. Hafizi S., Chester A., Allen S., Morgan K., Yacoub M. Growth response of human coronary artery smooth muscle cells to angiotensin II and influence of angiotensin AT1 receptor blockade. Coron. Artery Dis. (1998) 9:167–175.[Web of Science][Medline]
  9. Proudfoot D., Skepper J.N., Hegyi L., Bennett M.R., Shanahan C.M., Weissberg P.L. Apoptosis regulates human vascular calcification in vitro: evidence for initiation of vascular calcification by apoptotic bodies. Circ. Res. (2000) 87:1055–1062.[Abstract/Free Full Text]
  10. Macdonald K., Bennett M.R. cdc25A is necessary but not sufficient for optimal c-myc-induced apoptosis and cell proliferation of vascular smooth muscle cells. Circ. Res. (1999) 84:820–830.[Abstract/Free Full Text]
  11. Bennett M.R., Littlewood T.D., Hancock D.C., Evan G.I., Newby A.C. Down-regulation of the c-myc proto-oncogene in inhibition of vascular smooth-muscle cell proliferation: a signal for growth arrest? Biochem. J. (1994) 302:701–709.[Web of Science][Medline]
  12. Hamel P.A., Cohen B.L., Sorce L.M., Gallie B.L., Phillips R.A. Hyperphosphorylation of the retinoblastoma gene product is determined by domains outside the simian virus-40 large T antigen-binding regions. Mol. Cell Biol. (1990) 10:6586–6595.[Abstract/Free Full Text]
  13. Bennett M.R., Evan G.I., Schwartz S.M. Apoptosis of rat vascular smooth muscle cells is regulated by p53-dependent and -independent pathways. Circ. Res. (1995) 77:266–273.[Abstract/Free Full Text]
  14. Meijer L., Borgne A., Mulner O., et al. Biochemical and cellular effects of roscovitine, a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2 and cdk5. Eur. J. Biochem. (1997) 243:527–536.[Web of Science][Medline]
  15. Whyte P., Williamson N.M., Harlow E. Cellular targets for transformation by the adenovirus E1A proteins. Cell (1989) 56:67–75.[CrossRef][Web of Science][Medline]
  16. Alevizopoulos K., Catarin B., Vlach J., Amati B. A novel function of adenovirus E1A is required to overcome growth arrest by the CDK2 inhibitor p27(Kip1). EMBO J. (1998) 17:5987–5997.[CrossRef][Web of Science][Medline]
  17. Bennett M.R., Evan G.I., Schwartz S.M. Apoptosis of human vascular smooth muscle cells derived from normal vessels and coronary atherosclerotic plaques. J. Clin. Invest. (1995) 95:2266–2274.[Web of Science][Medline]
  18. Ihling C., Menzel G., Wellens E., Monting J.S., Schaefer H.E., Zeiher A.M. Topographical association between the cyclin-dependent kinases inhibitor P21, p53 accumulation, and cellular proliferation in human atherosclerotic tissue. Arterioscler. Thromb. Vasc. Biol. (1997) 17:2218–2224.[Abstract/Free Full Text]
  19. Wei G.L., Krasinski K., Kearney M., Isner J.M., Walsh K., Andres V. Temporally and spatially coordinated expression of cell cycle regulatory factors after angioplasty. Circ. Res. (1997) 80:418–426.[Web of Science][Medline]
  20. Tanner F.C., Yang Z.Y., Duckers E., Gordon D., Nabel G.J., Nabel E.G. Expression of cyclin-dependent kinase inhibitors in vascular disease. Circ. Res. (1998) 82:396–403.[Abstract/Free Full Text]
  21. Yutani C., Ishibashi-Ueda H., Suzuki T., Kojima A. Histologic evidence of foreign body granulation tissue and de novo lesions in patients with coronary stent restenosis. Cardiology (1999) 92:171–177.[CrossRef][Web of Science][Medline]
  22. Brand K., Klocke R., Possling A., Paul D., Strauss M. Induction of apoptosis and G2/M arrest by infection with replication-deficient adenovirus at high multiplicity of infection. Gene Ther. (1999) 6:1054–1063.[CrossRef][Web of Science][Medline]
  23. Chang M., Barr E., Seltzer J., et al. Cytostatic gene therapy for vascular proliferative disorders with a constitutively active form of the retinoblastoma gene product. Science (1995) 267:518–522.[Abstract/Free Full Text]
  24. Alessi F., Quarta S., Savio M., et al. The cyclin-dependent kinase inhibitors olomoucine and roscovitine arrest human fibroblasts in G1 phase by specific inhibition of CDK2 kinase activity. Exp. Cell Res. (1998) 245:8–18.[CrossRef][Web of Science][Medline]
  25. Tanner F., Boehm M., Akyürek L., et al. Differential effects of the cyclin-dependent kinase inhibitors p27Kip1, p21Cip1 and p16Ink4 on vascular smooth muscle cell proliferation. Circulation (2000) 101:2022–2025.[Abstract/Free Full Text]
  26. Grube E., Silber S., Hauptmann K.E., et al. TAXUS I: six- and twelve-month results from a randomized, double-blind trial on a slow-release paclitaxel-eluting stent for de novo coronary lesions. Circulation (2003) 107:38–42.[Abstract/Free Full Text]
  27. Morice M.C., Serruys P.W., Sousa J.E., et al. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N. Engl. J. Med. (2002) 346:1773–1780.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Circ. Res.Home page
P. A. Calvert and M. R. Bennett
Restenosis Revisited
Circ. Res., April 10, 2009; 104(7): 823 - 825.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by O'Sullivan, M.
Right arrow Articles by Bennett, M. R
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by O'Sullivan, M.
Right arrow Articles by Bennett, M. R
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?