Author + information
- Received May 10, 1999
- Revision received April 20, 2000
- Accepted June 26, 2000
- Published online November 1, 2000.
- Zhihong Yang, MD∗,†,
- Toshiyoki Kozai, MD∗,†,
- Bernd van de Loo, MD∗,†,
- Hema Viswambharan, MSc∗,†,
- Mario Lachat, MD‡,
- Marko I Turina, MD‡,
- Tadeusz Malinski, PhD§ and
- Thomas F Lüscher, MD, FRCP, FACC∗,†,*
- ↵*Reprint requests and correspondence: Dr. Thomas F. Lüscher, Cardiovascular Center, Cardiology, University Hospital, CH-8091 Zürich/Switzerland
This study examined effects of 3-hydroxy-3-methylglutaryl CoA (HMG-CoA) reductase inhibitor cerivastatin on human saphenous vein (SV), endothelial cells (EC) and smooth muscle cells (SMC).
Venous bypass graft failure involves EC dysfunction and SMC proliferation. Substances that improve EC function and inhibit SMC proliferation would be of clinical relevance.
Both EC and SMC were isolated from SV. Endothelial nitric oxide synthase (eNOS) expression and nitric oxide (NO) production were analyzed by immunoblotting and porphyrinic microsensor. The SMC proliferation was assayed by 3H-thymidine incorporation. Protein kinases and cell cycle regulators were analyzed by immunoblotting.
Cerivastatin (10−9 to 10−6 mol/liter) enhanced eNOS protein expression and NO release (about two-fold) in EC in response to Ca2+ ionophore (10−6 mol/liter). This was fully abrogated by the HMG-CoA product mevanolate (2 × 10−4 mol/liter). In SMC, platelet-derived growth factor (5 ng/ml) enhanced 3H-thymidine incorporation (298 ± 23%, n = 4), activated cyclin-dependent kinase (Cdk2), phosphorylated Rb and down-regulated p27Kip1 (but not p21Cip1). Cerivastatin reduced the 3H-thymidine incorporation (164 ± 11%, p < 0.01), inhibited Cdk2 activation and Rb phosphorylation, but did not prevent p27Kip1 down-regulation, nor p42mapk and p70S6K activation. Mevalonate abrogated the effects of cerivastatin on Cdk2 and Rb but only partially rescued the 3H-thymidine incorporation (from 164 ± 11% to 211 ± 13%, n = 4, p < 0.01).
In humans, SVEC inhibition of HMG-CoA/mevalonate pathway contributes to the enhanced eNOS expression and NO release by cerivastatin, whereas in SMC, inhibition of this pathway only partially explains cerivastatin-induced cell growth arrest. Inhibition of mechanisms other than p42mapk and p70S6K or Cdk2 are also involved. These effects of cerivastatin could be important in treating venous bypass graft disease.
Coronary bypass grafting using autologous internal mammary arteries and saphenous veins (SVs) is routinely used for treatment of patients with multivessel coronary artery disease. The patency rates of the arterial and venous grafts are, however, quite different. Occlusions are rare in the internal mammary artery, whereas they occur more frequently in the vein (1,2). Occlusion of the venous grafts is the major cause of recurrent ischemia in patients after coronary bypass surgery. Because of a shortage of arterial graft material and the relative ease of harvesting long vascular segments, SVs are still widely used. Considering the large number of patients who received venous bypass grafts in the last decades, the increasing incidence of venous bypass graft disease made the development of approaches to treat such patients an important therapeutic challenge.
Venous bypass graft failure is primarily related to biological properties of endothelial cells (ECs) and smooth muscle cells (SMCs) (3,4). The endothelium of the SV releases much less nitric oxide (NO) compared with that of the internal mammary artery (3,4). Nitric oxide is produced from l-arginine via endothelial nitric oxide synthase (eNOS) and is a potent vasodilator, platelet inhibitor and anti-proliferative agent (5). In the arterial circulation, the venous endothelium is unable to prevent platelet-vessel wall interaction, which favors thrombus formation and vascular occlusion (6). Furthermore, SMCs from SVs grow excessively in response to several growth factors such as platelet-derived growth factor (PDGF) and thrombin (4,7,8). This may contribute to the lower long-term patency rate of venous grafts.
Although mechanisms of cell growth regulation have not yet been fully understood, recent evidence suggests that signal transduction molecules such as MAPK, PI-3K and p70S6K, which are activated by tyrosine kinase receptors such as PDGF receptors, are important for cell cycle progression (9–11). Cell cycle progression is positively regulated by the orderly activation of cyclin-dependent kinases (Cdks) and negatively regulated by several Cdk inhibitors known as CKIs (12–14). p27Kip1 is most likely involved in cell cycle control in human SV SMC (8). Cyclin D-Cdk4/Cdk6 regulates G1 progression, and cyclin E-Cdk2 is essential for G1/S transition by phosphorylating and inactivating the tumor suppressor gene product, pRB (12–14), which causes release and activation of E2F transcription factor and in turn regulates several proteins required for cell proliferation (14).
Lipid-lowering therapy with 3-hydroxy-3-methylglutaryl CoA (HMG-CoA) reductase inhibitors strikingly decreases cardiovascular morbidity and mortality (15–18). The effects of those agents have been attributed to inhibition of cholesterol synthesis and lowering of low-density lipoprotein plasma levels. However, not all the clinical benefits of the agents can be explained by their lipid-lowering effects (19). Moreover, whether all the effects of statins could be explained by inhibition of HMG-CoA reductase is not clear. A recent study indicates that HMG-CoA reductase inhibitors could possess biological effects independent of HMG-CoA reductase (20).
In the clinical setting no optimal therapy for venous bypass graft disease is available (21). This may be related, at least in part, to the fact that antiproliferative drugs may not be able to improve endothelial function and at the same time inhibit SMC proliferation. In this study, we hypothesized that the new HMG-CoA reductase inhibitor cerivastatin could meet these criteria in human SVs and therefore may have the important clinical impact of improving venous bypass graft function. Moreover, we also investigated whether cerivastatin affects EC and SMC function in human SVs solely by the classic mechanism (i.e., inhibition of HMG-CoA reductase).
Chemicals and materials
Bovine serum albumin (BSA, 7.5%), monoclonal antibody against smooth muscle α-actin, and all chemicals for immunoblotting were from Sigma (Buchs, Switzerland); recombinant human PDGF-BB was from (R&D Systems GmbH, Wiesbaden, Germany); all tissue culture materials were from Gibco (Basel, Switzerland); 3H-methyl-thymidine was from Amersham (Zürich, Switzerland); trichloroacetic acid was from Fluka (Buchs, Switzerland); rabbit polyclonal anti-human p42mapk (C14), p27Kip1 (C19), and Cdk2 (M2) were from Santa Cruz Biotechnol (Basel, Switzerland); mouse monoclonal antibody against human p21Cip1 (Clone EA10) was from Calbiochem (JURO Supply AG, Lucerne, Switzerland); mouse monoclonal antibody against human eNOS (N30020) was from Transduction Laboratories (Basel, Switzerland); rabbit polyclonal anti-human phospho-pRB (Ser807/811) was from New England BioLabs (Schwalbach/Taunus, Germany); mevalonate was from Sigma (Buchs, Switzerland); and cerivastatin was kindly supplied by Bayer AG (Leverkusen, Germany).
Cultivation of ECs and SMCs
Endothelial cells were isolated from human SVs as described (4). Briefly, fresh blood vessels were harvested in a cold sterile RPMI-1640 medium with antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin). The vessels cleaned of connective tissue and adventitia were incubated with collagenase type II 75 U/ml for 15 min in phosphate-buffered saline (PBS). Cell pellets were then collected by centrifugation at 1,000 rpm for 10 min and seeded in culture dishes coated with 25 μg/ml human fibronectin and cultured in RPMI-1640 supplemented with 20 mmol/liter l-glutamine, HEPES buffer solution, 100 U/ml penicillin and 100 μg/ml streptomycin, 50 μg/ml endothelial cell growth supplement, 25 μg/ml heparin and 20% fetal calf serum (FCS). The day after, cells were washed with the medium to eliminate blood cell contamination. Endothelial cells were characterized by typical cobblestone and nonoverlapping appearance and indirect immunofluorescence staining using specific antibodies against von Willebrand factor. Cells of third and fourth passage were used.
Vascular SMC were cultivated from SVs obtained from patients undergoing coronary bypass surgery using explant technique as previously described (4,8). Briefly, the cells were cultured in Dulbecco’s modified eagle medium containing 20% FCS supplemented with 20 mmol/liter l-glutamine and HEPES buffer solution, 100 U/ml penicillin and 100 μg/ml streptomycin in a humidified atmosphere (37°C; 95% air/5% CO2). Culture medium was replaced every three days. Cells were passaged by trypsination (0.01% EDTA). Experiments were performed between passages five to eight. The SMC were characterized by typical morphological pattern (multilayer sheets and “hills and valleys”) and indirect immunofluorescence staining using specific mouse monoclonal antibodies against human SMC α-actin (Sigma).
The eNOS expression
Confluent ECs were rendered quiescent for 24 h by changing the medium to RPMI-1640 with the same ingredients as described above except that EC growth supplement and heparin were avoided and only 0.5% FCS was added. The cells were then treated with cerivastatin at different concentrations in the presence or absence of mevalonate for 24 h and then washed twice with PBS and harvested in the extraction buffer (120 mmol/liter sodium chloride, 50 mmol/liter Tris, 20 mmol/liter sodium fluoride, 1 mmol/liter benzamidine, 1 mmol/liter DTT, 1 mmol/liter EDTA, 6 mmol/liter EGTA, 15 mmol/liter sodium pyrophosphate, 0.8 μg/ml leupeptin, 30 mmol/liter p-nitrophenyl phosphate, 0.1 mmol/liter PMSF, and 1% NP-40) for immunoblotting. All cell debris were removed by centrifugation at 12,000 g for 10 min at 4°C. The samples (20 μg) were treated with 5xLaemmli’s SDS-PAGE sample buffer (0.35 mol/liter Tris-Cl, pH 6.8, 15% SDS, 56.5% glycerol, 0.0075% bromophenol blue), followed by heating at 95°C for 3 min and then subjected to 8% SDS-PAGE gel for electrophoresis. The proteins were then transferred onto ImmobilonTM-P filter papers (Millipore AG, Volketswil, Switzerland) with a semi-dry transfer unit (Hoefer Scientific, Switzerland). The membranes were then blocked by using 5% skim milk in PBS-Tween buffer (0.1% Tween 20; pH = 7.5) for 1 h and incubated with the antibody against human eNOS (1:600). The immunoreactive bands were detected by an enhanced chemiluminescence system (Amersham, Zürich, Switzerland).
Measurement of NO release
In parallel experiments the cells treated with cerivastatin and mevalonate as described above were used for measurement of NO release by porphyrin microsensor (22). Amperometric mode detection was used at a constant potential equal to the peak potential for NO oxidation of the electrode. The NO concentration was determined from the measured current by means of a calibration curve with NO standards. Standard NO solutions (1.8 mmol/liter) were prepared from aequeous solutions saturated with pure gaseous NO. Immediately before NO measurements, the active tip of the L-shaped porphyrinic NO microsensor was placed directly on the surface of the EC monolayer. The monolayer had been washed three times with PBS, pre-warmed to 37°C, immediately before the measurements. For maximal stimulation of eNOS, Ca2+ ionophore A23187 was injected into the cell culture dish to yield a final concentration of 10−6 mmol/liter. In some experiments, EC monolayers were pre–incubated with l-NAME (10−4 mmol/liter) for 15 min before NO measurements. Only a very marginal signal could be recorded under this condition, and this was subtracted during the calculation.
The SV SMCs were seeded on 12-well plates at a density of 2 × 104/well and exposed to PDGF-BB in the presence or absence of cerivastatin or cerivastatin plus mevalonate. The DNA synthesis was measured by 3H-thymidine (1 μCi/ml; 70 to 85 Ci/mol) incorporation after 24 h as described (4,8).
Activation of protein kinase
Activation of MAPK as well as p70S6K were analyzed by mobility shift of the kinases on Western blots as described (4,8). The cells were pre–treated with different substances tested, then stimulated with PDGF-BB (10 ng/ml) for 10 min and harvested in the extraction buffer as previously described (8). Cell lysates above 20 μg were treated with 5xLaemmli’s SDS-PAGE sample buffer and subjected to 10% SDS-PAGE gel for electrophoresis. Western blot was performed by using specific antibodies.
Cell cycle regulatory proteins
The cells were stimulated with PDGF-BB (10 ng/ml) in the presence or absence of cerivastatin or cerivastatin plus mevalonate for 24 h, and then harvested with extraction buffer as described (8). Regulation of CKIs (p21Cip1 and p27Kip1), Cdk2 and pRb phosphorylation by PDGF-BB was analyzed by immunoblotting as described above; 12% SDS-PAGE was used for analysis of p21Cip1 and p27kip1; Cdk2 and 8% SDS-PAGE were used for detection of phosphorylation of pRb.
Data are means ± SEM. The 3H-thymidine incorporation and MAPK activity were expressed as percent increase above control. In all experiments, n equals the number of patients from which vessels were obtained. The Student t test for paired observations and analysis of variance followed by the Scheffe test for repeated measurements were used. A two-tailed p value smaller than 0.05 was considered significant.
Effects of cerivastatin on eNOS
In cultured human SV-EC, treatment of the cells with cerivastatin (10−9 to 10−6 mol/liter; 24 h) up-regulated the eNOS protein level in a concentration-dependent manner (Fig. 1, top panel). Densitometry showed that the increase in eNOS protein level reached about two-fold (Fig. 1, lower panel, n = 3). This up-regulation of eNOS expression by cerivastatin (10−6 mol/liter) was fully reversed by mevalonate (2 × 104 mol/liter) (Fig. 2A, B). Accordingly, NO production stimulated by Ca2+ ionophore (10−6 mol/liter) in the cells treated with cerivastatin (10−6 mol/liter) for 24 h was also markedly increased (from 393 ± 55 nmol/liter to 840 ± 42% nmol/liter, n = 3, p < 0.01) as measured by the porphyrinic microsensor (Fig. 2C). Similar to eNOS protein expression, the enhanced NO release by cerivastatin was also fully reversed by treatment of the cells with mevalonate (2 × 104 mol/liter; Fig. 2C, 372 ± 78 nmol/liter, p < 0.01 vs. cerivastatin-treated cells), whereas mevalonate alone had no effects.
Effects of cerivastatin on SMC proliferation
In cultured human SV SMC, the growth factor PDGF-BB (5 ng/ml) markedly enhanced 3H-thymidine incorporation (298 ± 23% above control, n = 4), which was concentration-dependently inhibited by cerivastatin (data not shown). The inhibitory effects of cerivastatin (10−6 mol/liter; 164 ± 11%) was only partially abrogated by co-incubation of the cells with mevalonate (2 × 104 mol/liter; 211 ± 13%, Fig. 3; n = 4, p < 0.05 vs. cerivastatin plus PDGF). Mevalonate alone had no effects on 3H-thymidine incorporation.
Effects of cerivastatin on MAPK and p70S6K activation
After stimulation of SMC with PDGF-BB (10 ng/ml, 10 min) both p42mapk and p70S6K were activated, as demonstrated by a slower mobility of the activated (phosphorylated) kinase on Western blots (Fig. 4). Cerivastatin (10−6 mol/liter) did not exhibit inhibitory effects on either p42mapk or p70S6K activation in response to PDGF-BB (Fig. 4). By contrast, activation of p42mapk was specifically inhibited by PD98059 (5 × 10−5 mol/liter), a specific MAPK kinase (MEK) inhibitor. The p70S6K activation stimulated by PDGF-BB (10 ng/ml) was prevented by rapamycin (10−5 mol/liter), a specific inhibitor of p70S6K, and also by wortmannin (10−5 mol/liter), a specific inhibitor of PI3-K (Fig. 4).
Effects of cerivastatin on cell cycle regulators
Western blotting demonstrated that stimulation of the SV SMC with PDGF-BB (10 ng/ml) for 24 h induced activation of Cdk2 accompanied by an increase in its electrophoretic mobility on SDS-PAGE (Fig. 5) due to Cdk2 phosphorylation on Thr160 (23), down-regulation of p27Kip1 and phosphorylation of pRb. In contrast, p21Cip1 was not down-regulated by PDGF-BB (Fig. 5). Cerivastatin (10−6 mol/liter) inhibited PDGF–induced activation of Cdk2 and pRb phosphorylation but had no significant effects on the down-regulation of p27Kip1(Fig. 5). These effects of cerivastatin on Cdk2 and Rb were almost fully reversed by mevalonate (2 × 10−4 mol/liter).
The present finding
The major findings of the present study are that in human SV the HMG-CoA reductase inhibitor cerivastatin enhances eNOS protein level and NO release by the classic mechanism (i.e., inhibition of HMG-CoA reductase) in EC, whereas inhibition of HMG-CoA reductase only partially contributes to the inhibitory effects of cerivastatin on SMC growth in response to PDGF. Mechanisms other than inhibition of p42mapk, PI-3K/p70S6K or Cdk2 are involved.
Effect of cerivastatin on SV EC
Clinical trials demonstrated that HMG-CoA reductase inhibitors markedly reduce mortality and morbidity in patients with coronary artery disease or stroke (15–17,24). There is increasing evidence that HMG-CoA reductase inhibitors exhibit several beneficial effects via non-lipid-dependent mechanisms (19). In line with the experiments with simvastatin and lovastatin (18,25–27), the present study demonstrated that cerivastatin enhanced eNOS protein expression and NO release by Ca2+ ionophore in the same cells. These effects of cerivastatin were fully abrogated by the HMG-CoA product mevalonate, indicating that cerivastatin up-regulates eNOS protein level via inhibition of HMG-CoA reductase in the human SV EC. By contrast, inhibition of HMG-CoA reductase in bovine aortic ECs does not increase eNOS expression (28). The mechanism for the discrepancy is not clear. It might be due to the different cell types used. The molecular mechanism of increase in eNOS expression by statins in the human SV EC is due to inhibition of Rho GTPase, whose activation is dependent on prenylation induced by mevalonate metabolites and negatively regulates stability of eNOS mRNA. (26,27). Our results provide evidence that the HMG-CoA reductase inhibitors help to adapt endothelial function in the human SV to the arterial circulation where higher amounts of NO production are required than the venous side.
Effect of cerivastatin on SV SMCs
In addition to endothelial dysfunction, an enhanced vascular SMC growth also importantly contributes to human venous bypass graft failure (6–8). Our previous study demonstrated that the growth factor PDGF released from aggregating platelets at the site of vascular injury is an important mediator of SV SMC growth (8). Further evidence demonstrated that p42/44mapk, PI-3K and p70S6K are important signaling molecules for cell growth (9–11). The present study demonstrated that cerivastatin is a potent inhibitor of SMC growth of the SV. However, the inhibitory effect of cerivastatin on the SMC growth is only partially reversed by the HMG-CoA product mevalonate, indicating that, unlike in the ECs, inhibition of the HMG-CoA/mevalonate pathway only partially contributes to the inhibitory effect of the drug on SMC growth. Other studies showed that HMG-CoA reductase inhibitors prevent cell growth by blocking Rho GTPase via inhibition of the protein prenylation induced by mevalonate metabolites (29–31). The results from the present study with mevalonate suggest that this mechanism is only partially involved in cell growth. We further analyzed whether cerivastatin could interfere with other signal transduction mechanisms such as p42mapk and p70S6K. Interestingly, our results showed that cerivastatin did not influence p42mapk or p70S6K activation. This suggests that cerivastatin inhibits SMC growth via mechanisms other than p42mapk and p70S6K. The exact mechanisms remain to be explored.
The cell growth is positively regulated by the orderly activation of Cdks and negatively regulated by several Cdk inhibitors (12–14). In the present study we demonstrated that the growth factor PDGF activated Cdk2 and phosphorylated Rb, both of which were inhibited by cerivastatin. These effects of cerivastatin were almost fully reversed by mevalonate, indicating that cerivastatin prevented PDGF-induced Cdk2 activation and Rb phosphorylation via the classic mechanism (i.e., inhibition of HMG CoA reductase). Of note, the rescuing effects of mevalonate on Cdk2 and Rb were not fully correlated with the reversal of 3H-thymidine incorporation by the substance. Only about one-third of the inhibition of 3H-thymidine incorporation by cerivastatin was reversed by mevalonate. This indicates that inhibition of HMG-CoA reductase linked to Cdk2 blockade is only one mechanism of the drug-induced growth arrest of SMCs. It is possible that other cell-cycle regulatory proteins are still inhibited by cerivastatin independent on HMG-CoA/mevalonate pathway. This issue needs additional investigation.
Furthermore, results of the present study showed that p27Kip1, but not p21Cip1, was down-regulated by PDGF. This confirmed our previous study (8) showing that p27Kip1 is most likely involved in cell growth regulation of these cells. Several studies showed that inhibition of HMG-CoA reductase prevents p27Kip1 or p21Cip1 down-regulation in established cell lines (32,33). To our surprise, the present study demonstrated that cerivastatin was unable to prevent the down-regulation of p27Kip1, indicating that HMG-CoA/mevalonate pathway is not involved in the regulation of the cell-cycle inhibitor in these particular human cells.
In IIC9 cells, a subculture of Chinese hamster embryo fibroblasts and also in FRTL-5 cells, a strain of rat thyroid cells, p27Kip1 down-regulation or degradation is dependent on Cdk2 activation through RhoA (29,34). However, this mechanism seems not to contribute to the down-regulation of the Cdk inhibitor in our cells, because inhibition of Cdk2 activation by cerivastatin did not correlate with up-regulation of p27Kip1. Our data contrast with the observation by Laufs et al. (35), who did show prevention of p27Kip1 down-regulation by another statin simvastatin in cultured SMCs from human aorta and SV. Unfortunately, this study did not specify the cell types from which the results were raised. It is also possible that simvastatin may have different effects on p27Kip1 than cerivastatin. This issue will be further investigated.
A recent clinical post-CABG trial showed that the progression of atherosclerosis in SV grafts was delayed by aggressive lovastatin treatment but not by moderate treatment (36). Because NO release in the human SV is very low, and excessive SMC proliferation importantly contributes to the poor long-term patency rate of venous grafts, the effects of cerivastatin on human SV EC and SMCs as shown in this study may be beneficial for patients with venous bypass grafts.
☆ Original research reported in this article was supported by grants of the Swiss National Research Foundation (No. 32-51069.97/1), the Swiss Heart Foundation, Swiss 3R Research Foundation (61/97), Prof. Dr. Max Cloetta-Foundation, and a research grant of Bayer AG (T. Kozai), Leverkusen, Germany.
- cyclin-dependent kinases
- endothelial cell(s)
- endothelial nitric oxide synthase
- fetal calf serum
- 3-hydroxy-3-methylglutaryl CoA
- nitric oxide
- phosphate-buffered saline
- platelet-derived growth factor
- smooth muscle cell(s)
- saphenous vein
- Received May 10, 1999.
- Revision received April 20, 2000.
- Accepted June 26, 2000.
- American College of Cardiology
- Yang Z.,
- Ruschitzka F.,
- Rabelink T.J.,
- et al.
- Lüscher T.F.,
- Vanhoutte P.M.
- Angelini G.D.,
- Newby A.C.
- Yang Z.,
- Oemar B.S.,
- Carrel T.,
- Kipfer B.,
- Julmy F.,
- Lüscher T.F.
- Blenis J.
- Downward J.
- Endres M.,
- Laufs U.,
- Huang Z.,
- et al.
- Rao S.,
- Porter D.C.,
- Chen X.,
- Herliczek T.,
- Lowe M.,
- Keyomarsi K.
- Plehn J.F.,
- Davis B.R.,
- Sacks F.M.,
- et al.
- Laufs U.,
- Liao J.K.
- Laufs U.,
- Fata V.L.,
- Plutzky J.,
- Liao J.K.
- Hernandez-Perera O.,
- Pérez-Sala D.,
- Navarro-Antolin J.,
- et al.
- Hu W.,
- Bellone C.J.,
- Baldassare J.J.
- Guijarro C.,
- Blanco-Colio L.M.,
- Ortego M.,
- et al.
- Lee S.J.,
- Ha M.J.,
- Lee J.,
- et al.
- Hengst L.,
- Reed S.I.
- Hirai A.,
- Nakamura S.,
- Noguchi Y.,
- et al.
- Laufs U.,
- Marra D.,
- Node K.,
- Liao J.K.