Dihydropyridine Ca2+ channel antagonists inhibit the salvage pathway for DNA synthesis in human vascular smooth muscle cells

European Journal of Pharmacology - Molecular Pharmacology Section, 244 (1993) 269-275

269

© 1993 Elsevier Science Publishers B.V. All rights reserved 0922-4106/93/$06.00

EJPMOL 90415

channel antagonists inhibit the salvage pathway for DNA synthesis in human vascular smooth muscle cells

D i h y d r o p y r i d i n e C a 2+

A l e x Agrotis, P e t e r J. Little, J o h n Saltis a n d Alex B o b i k Baker Medical Research Institute and Alfred Baker Medical Unit, Alfred Hospital, Prahran, Victoria, Australia

Received 6 May 1992,revised MS received28 October 1992, accepted 3 November 1992

We examined the mechanisms by which Ca 2+ channel antagonists inhibit the growth of smooth muscle cells by determining their effect on epidermal growth factor (EGF)-stimulated (i) induction of the early signalling gene, c-los, (ii) incorporation of [3H]thymidine into cells as a measure of DNA synthesis, and (iii) increase in cell number. Verapamil, diltiazem, and the dihydropyridines felodipine, MDL 72892 A-15 (MDL) and nisoldipine had no effect on EGF-stimulated c-los mRNA induction. Furthermore, only small inhibitory effects were observed on EGF-stimulated increases in cell number; felodipine, MDL, and nisoldipine at 0.3/xM inhibited EGF-stimulated cell proliferation by 9, ll, and 15%, respectively. In contrast, the dihydropyridine Ca 2+ channel antagonists were found to be potent inhibitors of [3H]thymidine incorporation suggesting that they inhibit DNA synthesis. However, further examination revealed that the potent effects of dihydropyridine Ca z+ channel antagonists on [3H]thymidine incorporation were due not to an effect on incorporation of [3H]thymidine into DNA, but to a marked inhibitory effect on the cellular uptake of [3H]thymidine. Thus, we conclude that the small antiproliferative effects of the dihydropyridine antagonists are predominantly due to their ability to inhibit the activity of the salvage pathway for thymidylate synthesis in human vascular smooth muscle cells. Smooth muscle (vascular); Ca e+ channel antagonists; Proliferation

I. Introduction The abnormal proliferation of vascular smooth muscle cells is a major underlying characteristic of a number of vascular diseases, including hypertension, atherosclerosis (Ross, 1986; Schwartz et al., 1986; Klagsbrun and D'Amore, 1991) and the vascular growth response to balloon angioplasty (Hanson et al., 1991). In regard to atherosclerosis, a number of investigations have focussed on the potential antiatherogenic effects exerted by Ca 2+ channel antagonists in animals with dietary hypercholesterolemia. In cultured vascular smooth muscle cells a number of processes potentially contributing to the antiatherogenic effects of Ca 2+ channel antagonists have been reported. These include the inhibition of smooth muscle cell migration (Nomoto et al., 1987, 1988), a reduced net uptake of cholesterol into vascular smooth muscle cells (Etingin and Hajjar, 1985), and the inhibition of smooth muscle cell proliferation (Nilsson et al., 1985; Stein et al., 1987; Sperti and Colucci, 1991).

Correspondence to: Dr. A. Agrotis, Baker Medical Research Institute, Alfred Hospital, Commercial Road, Prahran, Victoria 3181, Australia. Tel. 61-3-522 4333; Fax 61-3-521 1362.

Since the abnormal proliferation of vascular smooth muscle cells is a prime event in the etiology of atherosclerosis and vascular neointimal growth following balloon injury, it is of fundamental importance to understand how this proliferation can be regulated in vivo. A number of growth factors, in particular those with receptor tyrosine kinase activity, have been identified in atherosclerotic lesions and the neointima of damaged vessels. Following injury to the vessel wall, growth factors can be released from damaged endothelial cells, attached platelets, and even macrophages. One such growth factor is epidermal growth factor (EGF). This peptide can be released from platelets (Bowen-Pope and Ross, 1983); macrophages also appear to release EGF-like substances. With respect to the latter cell type, macrophage-like U-937 cells have recently been shown to secrete a peptide growth factor whose predicted amino acid sequence indicates that it is a new member of the epidermal growth factor family (Higashiyama et al., 1991). This heparin-binding EGFlike growth factor also binds to E G F receptors on A-431 epidermoid carcinoma cells and smooth muscle cells indicating a role for this growth factor in macrophage-mediated cellular proliferation (Higashiyama et al., 1991). Both platelets and macrophages have been implicated in the pathogenesis of atheroscle-

270 rotic lesions and neointimal smooth muscle cell proliferation; significantly, mRNA for both transforming growth factor (TGF)-a, a polypeptide that interacts with the same receptor as EGF, and the EGF-receptor have been shown to be present in human atheromatous plaque tissue (Bauriedel et al., 1991). In order to understand how EGF-stimulated vascular smooth muscle cell proliferation may be regulated we assessed the effects of five Ca 2+ channel antagonists of diverse chemical structure on [3H]thymidine uptake into vascular smooth muscle cells, its incorporation into DNA, and their effects on c-fos proto-oncogene induction in EGF-stimulated vascular smooth muscle cells. These biochemical effects of the Ca 2÷ channel antagonists were related to their ability to attenuate vascular smooth muscle cell proliferation. We demonstrate that the ability of the dihydropyridine Ca 2÷ channel antagonists to inhibit vascular smooth muscle cell proliferation is related to their ability to inhibit the salvage pathway for thymidylate biosynthesis.

2. Materials and methods

2.1. Isolation and culture of human uascular smooth muscle cells Primary cultures of vascular smooth muscle cells were prepared from human internal mammary artery explants from two patients undergoing coronary bypass surgery as previously described by Neylon et al. (1990). In brief, the internal mammary artery was cut longitudinally and pieces of media were peeled from the lining of the vessel. Small pieces ( ~ 1-2 mm 2) were then placed at the bottom of a 90 mm tissue culture dish and covered with sterile coverslips which were anchored down with parraffin. 10 ml of Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum (FCS) and 60/zg/ml penicillin G (10% FCS/DMEM) was then added and the dish was incubated in a 1.5% CO 2 atmosphere at 37°C. The medium was changed every 2-4 days for 2-3 weeks when growth of vascular smooth muscle cells occurred under and beyond the coverslips. Cells were then routinely passaged every 5-7 days in 10% FCS/DMEM.

2.2. RNA dot-blot analysis Cytoplasmic RNA was prepared from confluent vascular smooth muscle cells as previously described (Gough, 1988). Equal amounts of RNA were denatured and transferred to nitrocellulose membranes using a BioRad dot-blot apparatus (Bobik et al., 1990). Membranes were baked under vacuum for 2 h at 80°C and prehybridised for 2 h at 42°C in 10-15 ml of 50%

formamide, 5 x SSC (750 mM NaCI, 75 mM sodium citrate, pH 7.0), 0.1% sodium dodecyl sulfate (SDS), 5 x Denhardt's solution, and 100 /zg/ml denatured salmon sperm DNA. Hybridisation was carried out at 42°C for ~ 15 h in the same buffer containing 32p_ labelled c-fos c D N A p r o b e (1-1.5 x 10 6 counts/min/ml). The probe was prepared using the random priming procedure as previously described (Feinberg and Vogelstein, 1982). Unhybridised probe was removed from the nitrocellulose membranes with four 10 min washes in 2 x SSC, 0.1% SDS at room temperature followed by two 30 min washes in 0.2 x SSC, 0.1% SDS at 55°C. Membranes were sealed in plastic and exposed at -70°C for 24-72 h to Kodak X-Omat AR film with intensifying screens and the resulting autoradiographs analysed by laser densitometry at a wavelength of 600 nm.

2.3. [ 3H]Thymidine incorporation into DNA The incorporation of methyl-[3H]thymidine into DNA was determined essentially as described in Bobik et al. (1991). Briefly, cells were grown to subconfluency in 1.0 ml of 10% FCS/DMEM on 24-well tissue culture dishes. The medium was then replaced with 1.0 ml of DMEM and the cells were further incubated for 48 h. EGF was added to quiescent cells (~ 1-2 x 105 cells/well) at a concentration of 30 ng/ml. Ca 2+ channel antagonists were added at a concentration of 3/zM in 0.5% ethanol (this concentration of ethanol was not toxic to cells). After incubation for 18-20 h the medium was aspirated; the cells were washed once with 1.0 ml DMEM and then incubated for 2 h in 1.0 ml of DMEM containing methyl-[3H]thymidine (1 /zCi/ml). At the end of this period the medium was removed, the cells were washed twice with 1.0 ml of ice-cold Dulbecco's phosphate-buffered saline (PBS) and incubated on ice in 0.5 ml of 10% trichloroacetic acid (TCA) for 15-30 min. After removing the TCA the cells were solubilised in 1 M NaOH prior to quantification of the radioactivity by liquid scintillation spectrometry.

2.4. [ 3H]Thymidine uptake For the determination of [3H]thymidine uptake the vascular smooth muscle cells were grown to subconfluency on 24-well tissue culture dishes and made quiescent in DMEM as described above. EGF (30 ng/ml) was then added to each well and the cells were incubated at 37°C for 18-20 h. The ceils were then washed once with DMEM prior to a 30 rain incubation in DMEM. Ca 2+ channel antagonists were then added and the cells further incubated for 15 min at 37°C. At the indicated times (see Results), [3H]thymidine (1 /zCi/ml) was added. The amount of [3H]thymidine taken up by the cells was determined over a 4 min

271

interval at 37°C. At the end of the incubation, the cells were placed on ice and rapidly washed with ice cold PBS (see 2.3) prior to release of the radioactivity into solution by 1 M NaOH. The amount of [3H]thymidine incorporated into DNA was determined as described in 2.3. 2.5. Cell proliferation The effects of the Ca 2+ channel antagonists on vascular smooth muscle cell proliferation were determined as previously described by Bobik et al. (1991). Cells (~ 5 × 103) were plated onto 24-well tissue culture dishes in 1.0 ml of 10% FCS/DMEM. 24-48 h later the medium was replaced with 1.0 ml of DMEM and the cells incubated for a further 48 h. EGF (30 ng/ml), and Ca 2+ channel antagonists (see Results) were then added to the cells cultured in serum-free media containing transferrin and insulin (4% Monomed A) in DMEM. Cell number was determined on trypsinised cultures 2 days later using a Coulter counter.

120 z o

m

I-- z

,,~

100

o

o r, l.aJ n'n," o

60 ~

_J

40

20 b.- I.a_

~0 "1"

DIL

VER

DIL

VER

1i NIS

MDL

NIS

MDL

l

FEL

160 140

no

2.6. Materials

u~o_ 100 U.I~, 1 <

Human internal mammary artery was provided by the Alfred Hospital Cardiothoracic Surgical Unit. Fetal calf serum, penicillin G, Monomed A, and Dulbecco's phosphate-buffered saline were from the Commonwealth Serum Laboratories, Melbourne, Australia. DMEM was from Flow Laboratories, Melbourne, Australia. Tissue culture dishes were purchased from ICN Australia. Epidermal growth factor, diltiazem, and verapamil were obtained from Sigma Chemical Company, St. Louis, MO, USA. Nisoldipine and BAY K 8644 were obtained from Bayer, Australia. MDL 72892 A-15 was kindly provided by the Merrell Dow Research Institute, Strasbourg, France. Felodipine was kindly provided by Astra, Australia. Methyl-[3H]thymidine was obtained from Amersham, Melbourne, Australia. A c-fos full length cDNA clone was kindly provided by Dr. I. Verma of the Salk Institute, San Diego, CA, USA. [a-32p]dATP was from Bresatec, Adelaide, Australia and the random priming kit was from Boehringer Mannheim, Australia.

3. Results

3.1. Ca 2 + channel antagonists and [3H]thymidine incorporation into DNA A number of previous studies have reported that Ca 2+ channel antagonists such as diltiazem, nifedipine, and verapamil inhibit the incorporation of [3H]thymidine into DNA of cultured porcine and rat aortic vascular smooth muscle cells (Tomita et al., 1987; Roe

n-

80

ZI.L

r,,, (O

E "'

60 40

~v

20 0

FEL

Fig. 1. A: Effect of Ca 2+ channel antagonists on EGF-stimulated [3H]thymidine incorporation. Quiescent cells ( ~ 1 - 2 × 105) were incubated with each antagonist (3 /~M) for 15 min at 37°C prior to addition of E G F (30 n g / m l ) . [3H]Thymidine incorporation was measured 18-20 h later. Hatched histograms represent the percentage of [3H]thymidine incorporation in the presence of each antagonist as compared to cells stimulated with E G F alone ( c o n t r o l = 100%). D I L = diltiazem; V E R = verapamil; NIS = nisoldipine; M D L = M D L 72892 A-15; F E L = felodipine. Results represent the m e a n + S.E.M. of six experiments. B: Effect of Ca 2+ channel antagonists on c-los m R N A stimulated by EGF. Smooth muscle cells were incubated with 3 / ~ M of each antagonist for 15 rain prior to stimulation with E G F (30 n g / m l ) or with E G F alone for 30 min. Total R N A was then isolated, blotted onto a nitrocellulose filter, and hybridised with a c-los c D N A probe. The percentage of the c-los response for each antagonist as compared to the response for E G F alone (control = 100%) is indicated by the hatched histograms. E G F induced a 15-fold elevation in c-los m R N A when compared to unstimulated cells. Results represent the mean _+S.E.M. of six experiments.

et al., 1989; Sperti and Colucci, 1991). In our studies EGF (30 ng/ml) induced a 2.4-fold increase in [3H]thymidine incorporation into the DNA of human vascular smooth muscle cells. In the presence of felodipine, MDL, or nisoldipine (3/xM), [3H]thymidine incorporation was inhibited by 82%, 53%, and 36% respectively (fig. 1A). Diltiazem had no effect on [3H]thymidine incorporation into the human vascular

272

smooth muscle cells whilst verapamil inhibited incorporation by only 6%. The ability of the dihydropyridine Ca 2+ channel antagonists to inhibit [3H]thymidine incorporation was not a consequence of any non-specific toxicity since vascular smooth muscle cell number remained constant during the 18-20 h period of drug exposure (data not shown). Because early signalling mechanisms such as Ca 2÷ mobilization have been implicated in DNA synthesis in the A7r 5 vascular smooth muscle cell line (Sperti and Colucci, 1991) we examined whether c-fos mRNA induction by EGF was affected by any of the Ca 2÷ channel antagonists. Since maximal elevation of c-fos mRNA occurred 30 min after exposing the quiescent vascular smooth muscle cells to EGF, the effects of the Ca 2+ channel antagonists on c-fos mRNA levels were examined at this time. No inhibitory effect of the Ca 2÷ channel antagonists on the elevation in c-los mRNA was observed (fig. 1B). Thus, early signalling mechanisms, in particular those involved in EGF mediated c-fos induction, do not appear to be implicated in the ability of the dihydropyridine Ca 2÷ channel antagonists to inhibit [3H]thymidine incorporation into DNA. To determine precisely where in the cell cycle (G~ or S phase) the dihydropyridine Ca 2+ channel antagonists exerted their inhibitory effect on [3H]thymidine incorporation, human vascular smooth muscle cells were first stimulated with EGF (time zero) and then at various time intervals up to 18 h later felodipine (3 /zM) was added. [3H]Thymidine incorporation into DNA was then determined 18 h after the addition of EGF. Felodipine was found to be equally effective when added at any stage of the G] phase of the mitotic cell cycle (fig. 2), inhibiting [3H]thymidine incorporation by approximately 80% (P < 0.01) at all times. When felodipine was added to cells in the S phase of the cell cycle, i.e. 16 h after the addition of EGF and 2 h prior to the addition of [3H]thymidine, or immediately prior to adding the [3H]thymidine, a similar marked reduction in [3H]thymidine incorporation was observed (fig. 2). Identical results were obtained with the same concentrations of MDL and nisoldipine (not shown).

3.2. Ca 2+ channel antagonists and the uptake of [ 3H] thymidine

Z

2 I-.<

100

o n no (.b

80

_z-5-

,,,~ z~--

60

~ -r N

40

I - - v

~'

2o

~ao

0

IRJmmmm C

1

2

3

4

5

6

7 16 18

Time (hr) of felodlplne addition after addition of EGF Fig. 2. Effect of felodipine on progression of vascular smooth muscle cells through the cell cycle. Quiescent cells ( ~ 1-2 × 105) were first stimulated with EGF (30 ng/ml) and at the indicated time periods felodipine (3 p.M) was added. [3H]Thymidine incorporation was determined 18 h after addition of EGF. The hatched histograms represent the percentage of [3H]thymidine incorporation in the presence of felodipine compared to EGF alone (control = 100%). Results represent the mean + S.E.M. of six experiments.

muscle cells was incorporated into DNA; the remainder was freely releasable on exposing the cells to trichloroacetic acid (fig. 3). Preincubation of the smooth muscle cells with felodipine inhibited by 86% the uptake of [3H]thymidine into the cells (fig. 3). Trichloroacetic acid-insoluble [3H]thymidine was also inhibited by a similar amount (82%) at this time. Other dihydropyridine Ca 2÷ channel antagonists also affected [3H]thymidine uptake in a similar manner over 2 min (fig. 4). Of these, felodipine was the most potent and 1O0

~

CONTROL

z 80 _o

ao 60 ~- z

40

~

20

Q.

0

The above experiments indicated that the uptake of [3H]thymidine into cells, or its incorporation into newly synthesised DNA, was being markedly inhibited by the dihydropyridine Ca 2+ channel antagonists. To distinguish between these two alternatives, the cells were initially exposed to EGF for 18 h and [3H]thymidine uptake and incorporation into DNA were assessed over a 4 min interval. During this time less than 15% of the [3H]thymidine taken up by the vascular smooth

FELODIPINE

/ •

T A

~

0

1

'

2

3

NTR

4

TIME (mln) Fig. 3. Effect of felodipine on uptake and incorporation of [3H]thymidine into smooth muscle cells. Quiescent cells ( ~ l - 2 X 105) were stimulated with EGF (30 ng/ml) for 18 h, and then incubated for a further 15 min at 37°C in the absence or presence of felodipine (3/zM). [3HJThymidine uptake (solid lines) or incorporation (dashed lines) at 1 min intervals over 4 min was then determined as described in the Materials and methods section. Results represent the mean + S.E.M. of six experiments.

273

nisoldipine the least potent, inhibiting [3H]thymidine uptake by 86% and 61% respectively. Verapamil and diltiazem exerted a small effect, inhibiting uptake by between 9% and 16% (fig. 4). Identical results were obtained in similar experiments when human athrombin (8 units/ml) was used to stimulate [3H]thymidine incorporation into DNA (data not shown). The dihydropyridine Ca 2+ channel agonist BAY K 8644 (Franckowiak et al., 1985) did not significantly inhibit [3H]thymidine uptake or reverse the inhibitory effect of felodipine (data not shown). Halfmaximal inhibition of [3H]thymidine uptake by felodipine, MDL, and nisoldipine was observed at 5 x 10 -7 M , 2 X 10 - 6 M , and 10 - 6 M respectively. 3.3. Effect of Ca 2 + channel antagonists on cell proliferation

A number of studies have reported that some Ca 2+ channel antagonists can inhibit the proliferation of smooth muscle cells in culture (Nilsson et al., 1985; Stein et al., 1987; Sperti and Colucci, 1991). We examined the relationship between the inhibition of [3H]thymidine uptake by the Ca 2+ channel antagonists and their ability to inhibit vascular smooth muscle cell replication stimulated by EGF. Growth arrested-subconfluent cultures of cells were incubated in D M E M / M o n o m e d medium containing EGF (30 ng/ml) in the presence or absence of the various Ca 2+ channel antagonists and the increase in cell number over 2 days was measured. Felodipine, MDL, and nisoldipine (3 × 10 - 7 M) attenuated the EGF-induced increase in cell number by 15%, 11%, and 9% respec-

100

100 tY

80 Z

6O (Z la.

z ,.o,

4o

,o "1-

I--- It. ~ O

~

20

20

r

L) Z

0

FEL

MDL

NIS

Fig. 5. Relationship between the effect of Ca 2÷ channel antagonists on cellular uptake of [3H]thymidine and cellular proliferation stimulated by EGF. Smooth muscle cells were stimulated with EGF (30 ng/ml) with and without the Ca 2÷ channel antagonists (3/xM) for 2 days prior to determination of cell number using a Coulter counter. [3H]Thymidine uptake was determined in the absence or presence of a Ca 2÷ channel antagonist (3 x 10 -7 M) as described in the legend to fig. 4. Hatched histograms represent the increase in EGF-stimulated cell number after 2 days in the presence of the Ca 2+ channel antagonist relative to the EGF control. Closed histograms indicate the percentage [3H]thymidine uptake in the presence of each Ca 2+ channel antagonist relative to the EGF control. Results represent the mean _+S.E.M. of six experiments.

tively (fig. 5). In contrast, the same concentrations of these Ca ~+ channel antagonists exerted far greater effect on [3H]thymidine uptake and incorporation into DNA (P < 0.01). Reduction in [3H]thymidine uptake/ incorporation averaged 48%, 39%, and 20% after adding felodipine, MDL, or nisoldipine respectively to the vascular smooth muscle cell cultures (fig. 5).

4. Discussion 100 b,ii

80

ILl V

.< Y/s

I--

D W

60 o r~

z ~-

40

~2

20

/////

-i- ~

T

;y//,

VER

mi i NIS

MDL

FEL

Fig. 4. Effect of Ca 2+ channel antagonists on [3H]thymidine uptake in smooth muscle cells. Quiescent ceils ( ~ l - 2 x 1 0 5 ) were first stimulated with EGF (30 ng/ml) for 18 h and then preincubated for 15 min with each Ca 2÷ channel antagonist (3 ~M). [3H]Thymidine uptake was determined over a 2 min interval as described in Materials and methods. The hatched histograms represent the percentage [3H]thymidine uptake in the presence of each antagonist as compared to the agonist control (EGF alone). Results represent the mean _+S.E.M. of six experiments.

A number of previous studies have utilized the technique of [3H]thymidine incorporation into DNA of cultured smooth muscle cells to assess the antiproliferative effects of Ca 2+ channel antagonists on vascular smooth muscle cells (Orekhov et al., 1986; Stein et al., 1987; Tomita et al., 1987; Roe et al., 1989; Sperti and Colucci, 1991). Similar to the present study these workers have found reductions in [3H]thymidine incorporation when the vascular smooth muscle cells were incubated with diltiazem, verapamil, and a number of dihydropyridine Ca 2+ channel antagonists. However, our findings indicate that the inhibition of [3H]thymidine incorporation into cells may overestimate the ability of these agents to inhibit vascular smooth muscle cell proliferation. Inhibition of [3H]thymidine incorporation appeared predominantly the consequence of a reduction in the activity of the salvage pathway for thymidylate synthesis and subsequent incorporation into DNA. Our finding that the antiproliferative effects of these agents was relatively small compared to their effects on [3H]thymidine incorporation indicates that

274 synthesis of thymidylate via de novo pathways involving the transfer of a methyl group from an activated form of tetrahydrofolate to uridylate is sufficient to ensure adequate thymidylate cellular levels for DNA biosynthesis and subsequent cell replication. In essence, our findings indicate that the observed effects of Ca 2÷ channel antagonists on reductions in [3H]thymidine incorporation are not truly indicative of decreased cellular DNA synthesis. A number of transport processes mediate the entry of purine and pyrimidine nucleosides and of various nucleoside analogues into mammalian cells (Plagemann et al., 1988). A family of non-concentrative facilitated diffusion systems accept as substrates both purine and pyrimidine nucleosides, including a number of cytotoxic nucleosides used in cancer chemotherapy (Plagemann et al., 1988). In addition, several ion (Na ÷, K+)-dependent concentrative transport systems for nucleosides have also been described (Vijayalakshmi and Belt, 1988; Williams and Jarvis, 1991). Subtypes of these nucleoside transporters can be distinguished by their sensitivities to inhibition by the S6-thiopurine derivatives nitrobenzyl-thioinosine and nitrobenzylthioguanosine. The nitrobenzyl-thioinosine-sensitive subtypes have been thoroughly studied in a variety of cell types, while the nitrobenzyl-thioinosine-resistant transporters remain relatively uncharacterized (Hammond, 1991). In erythrocytes the major nucleoside transporter, which is sensitive to inhibition by nitrobenzylmercaptopurine riboside, is a membrane associated glycoprotein with an apparent molecular mass of 55 kDa (Young et al., 1983). Our data on [3H]thymidine uptake suggest that attenuation of the salvage pathway by the dihydropyridine Ca 2÷ channel antagonists is due to inhibition of nucleoside transport through a specific interaction with the nucleoside transporter. For example, Striessnig et al. (1985) found that nucleosides and nitrobenzyl-thioguanosine inhibited the binding of [3H]nimodipine to red blood cell ghosts, suggesting that the Ca 2÷ channel antagonist receptor sites of the human erythrocyte are coupled to the nucleoside transporter. Furthermore, Verma and Marangos (1985) demonstrated that nimodipine inhibited the binding of [3H]nitrobenzyl-thioinosine to cerebral cortical membrane preparations from humans, dogs, guinea pigs, and mice; significantly, verapamil was markedly less potent in eliciting such an effect. It is well known that the normal proliferation of mammalian ceils has a requirement for precursors of nucleic acids which can be provided either by de novo synthesis of purine and pyrimidine nucleosides or by salvage of extracellular nucleobases and nucleosides. Indeed, the potential contribution the salvage pathway plays in regard to cellular proliferation is indicated from observations that a feature of T-acute lymphoblastic leukemia is the high density of membrane

nucleoside transport sites in erythrocytes, leukocytes, and leukemic blast cells (Cass et al., 1974; Wiley et al., 1982; White et al., 1987). We attempted to investigate the relative importance of the two pathways for DNA biosynthesis by examining the relationship between the magnitude of the inhibition of [3H]thymidine incorporation into DNA by the Ca 2+ channel antagonists and their effect on vascular smooth muscle cell proliferation. Although we demonstrated that a 50% reduction in the activity of the salvage pathway reduces the proliferative ability of vascular smooth muscle cells by no more than 10-15%, the inability of the proliferating cells to remain attached to plastic dishes in the presence of high (e.g. 3 tzM) concentrations of Ca 2÷ channel antagonists, an effect also seen by Nomoto et al. (1987), prevented us from examining the full relationship. Despite this limitation our experiments demonstrate that large reductions in the activity of the salvage pathway only have a small effect on the proliferative ability of vascular smooth muscle ceils. In summary, our findings suggest that the ability of the Ca 2÷ channel antagonists to prevent vascular lesions in atherosclerosis or vascular neointimal thickening following balloon injury is unlikely to be due to their antiproliferative effects. Despite the fact that the dihydropyridine Ca 2÷ channel antagonists are potent inhibitors of [3H]thymidine incorporation into the DNA of proliferating vascular smooth muscle cells, their antiproliferative effect is small. Rather our data suggests that their ability to inhibit vascular smooth muscle cell migration may be the major mechanism by which they reduce the severity of atherosclerotic lesions and vascular neointimal formation.

Acknowledgements This study was supported by grants-in-aid from the Medical Research Council of the Alfred Hospital, Melbourne, Australia, and the National Heart Foundation of Australia. We thank Ms. Julie Simpson for her expertise in typing the manuscript, and Mr. Peter Kanellakis for the preparation of diagrams.

References Bauriedel, G., K. Klingel, B. Hofling and R. Kandolf, 1991, In situ detection of TGF-o~and EGF receptor mRNA in human atheromatous plaque tissue, ArteriosclerosisThromb. 11, 1504. Bobik, A., A. Grooms, J.A. Millar, A. Mitchell and S. Grinpukel, 1990, Growth factor activity of endothelin on vascular smooth muscle, Am. J. Physiol. 258 (Cell Physiol. 27), C408. Bobik, A., A. Grooms, P.J. Little, E.J. Cragoe, Jr. and S. Grinpukel, 1991, Ethylisopropylamiloride-sensitivepH control mechanisms modulate vascular smooth muscle cell growth, Am. J. Physiol.260 (Cell Physiol. 29), C581. Bowen-Pope, D.F. and R. Ross, 1983, ls epidermal growth factor present in human blood? Interference with the radioreceptor assay for epidermal growth factor, Biochem. Biophys. Res. Commun. 114, 1036.

275 Cass, C.E., L.A. Gaudette and A.R.P. Paterson, 1974, Mediated transport of nucleosides in human erythrocytes. Specific binding of the inhibitor nitrobenzylthioinosine to nucleoside transport sites in the erythrocyte membrane, Biochim. Biophys. Acta 345, 1. Etingin, O.R. and D.P. Hajjar, 1985, Nifedipine increases cholesteryl ester hydrolytic activity in lipid-laden rabbit arterial smooth muscle cells, J. Clin. Invest. 75, 1554. Feinberg, A. and B. Vogelstein, 1982, A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity, Anal. Biochem. 132, 6. Franckowiak, G., M. Bechem, M. Schramm and G. Thomas, 1985, The optical isomers of the 1,4-dihydropyridine Bay K 8644 show opposite effects on Ca channels, Eur. J. Pharmacol. 114, 223. Gough, N.M., 1988, Rapid and quantitative preparation of cytoplasmic RNA from small numbers of cells, Anal. Biochem. 173, 93. Hammond, J.R., 1991, Comparative pharmacology of the nitrobenzylthioguanosine-sensitive and -resistant nucleoside transport mechanisms of Ehrlic ascites tumor cells, J. Pharmacol. Exp. Ther. 259, 799. Hanson, S.R., J.S. Powell, T. Dodson, A. Lumsden, A.B. Kelly, J.S. Anderson, A.W. Clowes and L.A. Harker, 1991, Effects of angiotensin converting enzyme inhibition with cilazapril on intimal hyperplasia in injured arteries and vascular grafts in the baboon, Hypertension 18 (II), 70. Higashiyama, S., J.A. Abraham, J. Miller, J.C. Fiddes and M. Klagsbrun, 1991, A heparin-binding growth factor secreted by macrophage-like cells that is related to EGF, Science 251,936. Klagsbrun, M. and P.A. D'Amore, 1991, Regulators of angiogenesis, Annu. Rev. Physiol. 53, 217. Neylon, C.B., P.J. Little, E.J. Cragoe, Jr. and A. Bobik, 1990, Intracellular pH in human arterial smooth muscle: regulation by N a + / H ÷ exchange and a novel 5-(N-ethyl-N-isopropyl) amiloride-sensitive Na ÷- and HCO3-dependent mechanism, Circ. Res. 67, 814. Nilsson, J., M. Sjolund, L. Palmberg, A.M. Von Euler, B. Jonzon and J. Thyberg, 1985, The calcium antagonist nifedipine inhibits arterial smooth muscle cell proliferation, Atherosclerosis 58, 109. Nomoto, A., J. Hirosumi, C. Sekiguchi, S. Mutoh, I. Yamaguchi and H. Aoki, 1987, Antiatherogenic activity of FR34235 (nilvadipine), a new potent calcium antagonist: effect on cuff induced intimal thickening of rabbit carotid artery, Atherosclerosis 64, 255. Nomoto, A., S. Mutoh, H. Hagihara and I. Yamaguchi, 1988, Smooth muscle cell migration induced by inflammatory cell products and its inhibition by a potent calcium antagonist, nilvadipine, Atherosclerosis 72. 213. Orekhov, A.N., V.V. Tertov, S.A. Kudryashov, Kh.A. Khashimov and V.N. Smirnov, 1986, Primary culture of human aortic intima cells

as a model for testing antiatherosclerotic drugs: effects of cyclic AMP, prostaglandins, calcium antagonists, antioxidants, and lipid-lowering agents, Atherosclerosis 60, 101. Plagemann, P.G.W., R.M. Wohlhueter and C. Woffendin, 1988, Nucleoside and nucleobase transport in animal cells, Biochim. Biophys. Acta 947, 405. Roe, M.W., J.R. Hepler, T.K. Harden and B. Herman, 1989, Platelet-derived growth factor and angiotensin II cause increases in cytosolic free calcium by different mechanisms in vascular smooth muscle cells, J. Cell. Physiol. 139, 100. Ross, R., 1986, The pathogenesis of atherosclerosis: an update, New Engl. J. Med. 314, 488. Schwartz, S.M., G.R. Campbell and J.H. Campbell, 1986, Replication of smooth muscle cells in vascular disease, Circ. Res. 58, 427. Sperti, G. and W.S. Colucci, 1991, Calcium influx modulates DNA synthesis and proliferation in A7r 5 vascular smooth muscle cells, Eur. J. Pharmacol. 206, 279. Stein, O., G. Halperin and Y. Stein, 1987, Long-term effects of verapamil on aortic smooth muscle cells cultured in the presence of hypercholesterolemic serum, Arteriosclerosis 7, 585. Striessnig, J.. G. Zernig and H. Glossmann, 1985, Human red bloodcell Ca2+-antagonist binding sites: evidence for an unusual receptor coupled to the nucleoside transporter, Eur. J. Biochem. 150, 67. Tomita, M., Y. Hirata, S. Takata and T. Fujita, 1987, Effects of calcium-antagonists and calmodulin inhibitors on DNA synthesis in cultured rat vascular smooth muscle cells, Endocrinol. Jpn. 34, 313. Verma, A. and P.J. Marangos, 1985, Nitrobenzylthioinosine binding in brain: an interspecies study, Life Sci. 36, 283. Vijayalakshmi, D. and J.A. Belt, 1988, Sodium-dependent nucleoside transport in mouse intestinal epithelial cells, J. Biol. Chem. 263, 19419. White. J.C., J.P. Rathmell and R.L. Capizzi, 1987, Membrane transport influences the rate of accumulation of cytosine arabinoside in human leukemia cells, J. Clin. Invest. 79, 380. Wiley, J.S., S.P. Jones, W.H. Sawyer and A.R.P. Paterson, 1982, Cytosine arabinoside influx and nucleoside transport sites in acute leukemia, J. Clin. Invest. 69, 479. Williams, T.C. and S.M. Jarvis, 1991, Multiple sodium-dependent nucleoside transport systems in bovine renal brush-border membrane vesicles, Biochem. J. 274, 27. Young, J.D., S.M. Jarvis, M.J. Robins and A.R.P. Paterson, 1983, Photoaffinity labelling of the human erythrocyte nucleoside transporter by N 6- (p-azidobenzyl) adenosine and nitrobenzylthioinosine, J. Biol. Chem. 258, 2202.