Smooth muscle cell cycle and proliferation. Relationship between calcium influx and sarco-endoplasmic reticulum Ca2+ATPase regulation.

The role of Ca2+ influx in the regulation of the sarco-endoplasmic reticulum Ca2+ATPases (SERCA) associated with intracellular Ca2+ pools was investigated during smooth muscle cell (SMC) proliferation induced by platelet-derived growth factor (PDGF). We first defined that the previously described up-regulation of the SERCA2a isoform found in vascular SMC after a 24-h stimulation with PDGF (Magnier, C., Papp, B., Corvazier, E., Bredoux, R., Wuytack, F., Eggermont, F., Maclouf, J., and Enouf, J. (1992) J. Biol. Chem. 267, 15808-15815) was precisely associated with SMC entry into S phase as it appeared linked with [3H]thymidine incorporation. This was further confirmed by testing the effect of transforming growth factor-β1, which inhibited both aortic SMC proliferation associated with G1 cell cycle arrest and PDGF-induced SERCA2a up-stimulation. Then, we tested the role of Ca2+ influx by using SR 33805, a new Ca2+ channel blocker, which was characterized with regard to the voltage Ca2+ channel blocker nifedipine and the capacitative entry Ca2+ blocker SKF 96365. SR 33805 was found to be the most potent inhibitor of both PDGF-induced SMC proliferation and the associated rise in intracellular Ca2+ concentration with IC50 values of 0.2 ± 0.1 and 0.31 ± 0. 04 µM, respectively. Finally, by examining in parallel both SERCA2a and SERCA2b isoforms, in terms of activity and expression, we could determine that PDGF-induced stimulation of total SERCA activity (detected by formation of the phosphorylated intermediate, E∼P) and of SERCA2a expression (Western blotting) were abolished when extracellular Ca2+ entry was prevented by SR 33805. This study demonstrates that SERCA2a up-regulation is: 1) related to the G1/S transition step of cell cycle and 2) dependent on Ca2+ entry during PDGF-induced SMC proliferation.

Calcium is an essential regulator of living cells, and it controls numerous aspects of the cellular physiology, including cell proliferation (1). An intracellular rise in cytosolic Ca 2ϩ concentration occurs upon cell activation due to both Ca 2ϩ influx from the extracellular medium and Ca 2ϩ release from intracellular storage Ca 2ϩ pools (for reviews, see Refs. 2 and 3). Ca 2ϩ channels in the plasma membrane allow Ca 2ϩ influx following a stimulation due to membrane depolarization (voltage-operated channels) (3,4) or binding of a ligand to its receptor (receptoroperated channels) (5). Intracellular Ca 2ϩ pools are also strongly implicated in the increase in cytosolic Ca 2ϩ concentration by liberating their Ca 2ϩ content into the cytosol through intracellular Ca 2ϩ channels: the inositol 1,4,5-trisphosphate (IP 3 ) 1 receptor (6) and the ryanodine receptor (7).
This increase in cytosolic Ca 2ϩ concentration is retro-controlled by the activation of Ca 2ϩ transport ATPases which decrease the Ca 2ϩ concentration of the cytosol. Plasma membrane Ca 2ϩ ATPases (PMCA) (8) eliminate Ca 2ϩ from the cell by trans-plasma membrane Ca 2ϩ transport, and sarco-endoplasmic reticulum Ca 2ϩ ATPases (SERCA) (9) enzymes reaccumulate cytosolic Ca 2ϩ in intracellular Ca 2ϩ storage pools. A close relationship between Ca 2ϩ influx and efflux mechanisms involved in the regulation of cytosolic Ca 2ϩ concentration has been formulated: indeed, according to the capacitative Ca 2ϩ entry model, the gating of Ca 2ϩ entry across the plasma membrane may be controlled by the depletion of intracellular stores (10 -12).
In view of the central role intracellular Ca 2ϩ pools play in cell signaling, the activity of the associated SERCA pumps could conceivably control cell proliferation. Indeed, a study which made use of the SERCA inhibitor thapsigargin, suggested that SERCA proteins may act as regulators of cell growth, by controlling growth-and transformation-related genes (c-fos and c-jun) (13). Subsequent studies suggested that profound alterations of the proliferative system of the DDT 1 MF-2 smooth muscle cell line were caused by the emptying of the intracellular thapsigargin-sensitive Ca 2ϩ pools and demonstrated that SERCA expression and Ca 2ϩ pool function are closely associated with growth and proliferation of these cells (14 -16). In the meanwhile, we could demonstrate the implication of the SERCA in a physiopathological model of cell proliferation by showing the specific up-regulation of the SERCA2a isoform during platelet-derived growth factor (PDGF)-induced smooth muscle cell (SMC) proliferation (17).
The aim of the present study was first to further precise the time course of this up-regulation of the SERCA2a isoform. Since it has been shown earlier that the different phases of the cell cycle exhibit each a different sensitivity to Ca 2ϩ ions (18), we looked for a relationship between SMC cycle and SERCA expression during PDGF-induced aortic SMC proliferation. We further tested the effect of the transforming growth factor-␤ 1 (TGF-␤ 1 ), an inhibitor of SMC entry into S phase (19). In addition, we attempted to change the loading state of the SMC Ca 2ϩ pools during cell proliferation by interfering with the Ca 2ϩ influx. We used a new analog of fantofarone, SR 33805, a potent Ca 2ϩ antagonist as shown by its binding to the ␣1subunit of the L-type voltage channel, because it: (i) presented a vascular selectivity, (ii) exhibited a severalfold more efficiency than other Ca 2ϩ blockers in PDGF-induced SMC proliferation, and (iii) was previously shown to strongly reduce myointimal thickening following endothelial injury (20 -22). We characterized this Ca 2ϩ channel antagonist with regard to its effects on PDGF-induced SMC proliferation and Ca 2ϩ influx, compared it to the dihydropyridine Ca 2ϩ blocker, nifedipine as well as to the capacitative entry blocker, SKF 96365, and investigated the effect of SR 33805 on the PDGF-dependent regulation of SERCA expression in SMC. The results show that the expression of SERCA2a precedes the entry of SMC into the S phase of the cell cycle and that it is abolished by preventing Ca 2ϩ influx.

EXPERIMENTAL PROCEDURES
Cell Culture-SMC were obtained either from human or pig aortas. Media fragments from the human aorta were incubated for 16 h at 37°C in Dulbecco's modified Eagle's medium (DMEM) containing 0.15% collagenase, 5% fetal calf serum (FCS), 100 IU/ml penicillin, 100 g/ml streptomycin, and 4 mM L-glutamine (Life Technologies, Inc., Paisley, UK). After incubation, SMC were sedimented by gentle centrifugation (400 ϫ g for 10 min), resuspended in DMEM containing 10% FCS and grown at 37°C in a humidified atmosphere of 5% CO 2 in air. Culture medium (DMEM ϩ 10% FCS) was changed every 3 days, and a confluent SMC monolayer was obtained after approximately 7 days. Cells were routinely used from the third to the sixth passage. Porcine SMC were cultured according to Ross (23). Briefly, the media was cut into 1-mm sized explants, which were placed in culture flasks containing DMEM supplemented with 50 IU/ml penicillin, 50 g/ml streptomycin, 2 mM L-glutamine, and 10% FCS. The explants were incubated at 37°C with 5% CO 2 in a humidified atmosphere for 3 weeks. SMC were subcultured and all experiments were performed after less than five passages (17).
Measurement of SMC Growth-Human aortic SMC were plated sparsely (10 3 cells/well) in 96 well cluster plates in DMEM ϩ 0.5% FCS. After 3 days, fresh DMEM medium containing 50 ng/ml recombinant PDGF-BB (R and D Systems, Oxon, UK) was added together with different concentrations of nifedipine (Sigma, L'Isle d'Abeau, France), SKF 96365 (Tebu, Le Perray, France), and SR 33805. After 24 h or 3 days in culture, cells were detached from triplicate wells by trypsin treatment (0.05% trypsin and 0.02% EDTA) and counted in a Coulter counter (Coultronics, France).
Measurement of DNA Synthesis in SMC-Pig aortic SMC were grown to preconfluence in 12-well plates in DMEM ϩ 10% FCS. They were then switched to serum-free medium (DMEM ϩ 0.2% bovine serum albumin) for 96 h, in order to synchronize them into the G 0 phase. Cells were then incubated for time intervals ranging from 0 to 48 h, in the presence of the growth factors, versus in serum free medium, and with 0.5 Ci/ml [ 3 H]thymidine for a 1-h pulse, to assay the mitogenic activity. Cells were then washed twice in phosphate-buffered saline, pH 7.4. After precipitation with 10% ice-cold trichloroacetic acid for 10 min, cells were solubilized with 200 mM NaOH. Incorporated radioactivity was quantified using a scintillation ␤ counter (Beckman, Fullerton, CA).
Preparation of SMC Membranes-Smooth muscle cells in monolayers were lysed in an ice-cold buffer containing 10 mM Hepes, pH 7.0, 10 mM KCl, 0.05 mM EGTA, 0.05 mM DTT, and a mixture of protease inhibitors (0.1 mg/ml soybean trypsin inhibitor, 0.05 mg/ml aprotinin, and 0.01 mg/ml leupeptin, Sigma). Cell suspensions were homogenized in a Teflon-glass homogenizer at 4°C, sonicated on ice, and centrifuged for 10 min at 3,500 ϫ g and 4°C. The supernatants were centrifuged for 60 min at 100,000 ϫ g and 4°C, and the resulting pellets containing microsomes were resuspended in 17 mM Hepes, pH 7.0, 160 mM KCl, and 0.1 mM DTT, frozen, and stored at Ϫ80°C.
SDS-PAGE and Western Blotting-Proteins from lysates or microsomes of SMC were solubilized and reduced for 30 min at room temperature in 50 mM Tris, pH 6.8, containing 4% (w/v) SDS, 0.01% (w/v) bromphenol blue, 25% (v/v) glycerol, and 5% (v/v) ␤-mercaptoethanol. Samples were analyzed on 8% SDS-PAGE and blotted onto nitrocellulose membranes. For the immunodetection of SERCA2a and SERCA2b isoforms, the nitrocellulose membranes were blocked for 1 h at room temperature in TBS (10 mM Tris/HCl, pH 7.5, and 150 mM NaCl) containing 0.05% Tween-20 and then incubated for 1 h at room temperature with a 1:100 dilution of anti-SERCA2a or anti-SERCA2b antibodies. After three washes in TBS-Tween, the membranes were incubated for 1 h at room temperature with a 1:50,000 dilution of horseradish peroxidase-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch, West Grove, PA). Immunostaining was revealed with a chemiluminescence kit (Amersham, Little Chalfont, UK) and autoradiography on Kodak X-Omat AR films.
Fura-2 Loading and Measurement of Ca 2ϩ Fura-2 Fluorescence-Cells grown to confluence in 75-cm 2 flasks were detached with 0.05% trypsin, 0.02% EDTA. The cell suspension was washed twice with Hepes buffer, pH 7.4 (10 mM Hepes, 125 mM NaCl, 2.6 mM KCl, 1.2 mM KH 2 PO 4 , 1.2 mM MgSO 4 , 1 mM CaCl 2 , 5 mM glucose), and resuspended in this buffer to a final concentration of 2.5-3 ϫ 10 6 cells/ml. Nifedipine, SKF 96365, SR 33805, or the vehicle were preincubated with the cells for 1 h before the experiment and the suspension was incubated with 2 M Fura-2/AM and 0.02% pluronic acid at 37°C for 45 min under continuous shaking. After washing with Hepes buffer, the cells were stimulated with 50 ng/ml PDGF-BB. Cytosolic Ca 2ϩ concentrations were determined after 1.5 min of incubation. Ca 2ϩ concentrations were measured with a SLM 8000 C spectrofluorometer at 37°C (excitation, 340 and 380 nm; emission, 510 nm). The calculation of these cytosolic free Ca 2ϩ concentrations was performed as described by Grynkiewicz et al. (31).
Formation of the Phosphorylated Intermediate (EϳP)-The Ca 2ϩ ATPases of SMC membranes were autophosphorylated on ice, in a reaction buffer containing 17 mM Hepes, pH 7.0, 160 mM KCl, 1 mM DTT, and 0.05 mM CaCl 2 , in the presence of 0.1 mM EGTA or without EGTA. The reactions were started by adding 0.05 M [␥-32 P]ATP (110 TBq/mmol, Little Chalfont, UK). One min later, they were stopped by adding ice-cold 6% trichloroacetic acid solution containing 1 mM ATP and 10 mM phosphoric acid. The resulting precipitates were washed in trichloroacetic acid, dissolved in the electrophoresis sample buffer, and submitted to 7.5% acidic SDS-PAGE as in Papp et al. (32). After electrophoresis, proteins were transferred onto nitrocellulose membranes (Schleicher & Schuell, Dassel, Germany) and autoradiographed for 48 h at Ϫ80°C using Kodak X-Omat AR films.

SERCA2 Expression and the Cell Cycle
We previously established a specific up-regulation of SERCA2a isoform after a 24 h-stimulation of porcine aortic SMC with PDGF-BB both by measuring SERCA activity and expression studies (17). To further point out the possible role of this PDGF-induced SERCA2a expression in cell proliferation, we looked for a correlation between the SERCA expression and a specific step of cell cycle.

Effect of PDGF on the Cell Cycle of Porcine SMC
We first defined the SMC cycle by following the incorporation of

Effect of TGF-␤ 1 on PDGF-induced Cell Cycle in Porcine SMC
To further establish the S phase step, we tested the effect of TGF-␤ 1 , known to block SMC cycle at the G 1 /S transition (19). From preliminary studies, we determined that 1 ng/ml TGF-␤ 1 initiated a significant inhibition (80%) of PDGF-induced SMC mitogenic activity (data not shown). Then, we tested the effect of TGF-␤ 1 on PDGF-stimulated SMC cycle progression. Quiescent SMC were incubated from 0 to 48 h in the presence of 5 ng/ml PDGF-BB and 1 ng/ml TGF-␤ 1 followed by a 1-h pulse of [ 3 H]thymidine (Fig. 1). TGF-␤ 1 inhibited 80% of cell entry into S phase. Only a very slight increase in [ 3 H]thymidine incorporation occurred after 20 -40 h of PDGF stimulation in the presence of TGF-␤ 1 , probably corresponding to entry into S phase of very few SMC.

Expression of SERCA2a and SERCA2b throughout the Cell Cycle
To study the effect of PDGF-induced SMC proliferation on the expression of SERCA2a and SERCA2b isoforms, SMC were stimulated by 5 ng/ml PDGF for 0 -34 h and lysed, and proteins were separated by 8% SDS-PAGE and analyzed by Western blotting. We used isoform-specific anti-SERCA2a and anti-SERCA2b antibodies, previously shown to react specifically in porcine SMC with the 100 kDa SERCA2a and SERCA2b proteins, respectively (17,29,30). As shown in Fig. 2A, the SERCA2a expression level was very low for times up to 14 h of stimulation with PDGF, but an increase could be detected after 16 h of stimulation. This up-regulation still persisted after 34 h in the presence of PDGF, although at that time a decline began to appear, as expected for a cell cycle dependent regulation. In sharp contrast with the modulation of the expression of SERCA2a isoform, no variation of the expression of the SERCA2b isoform occurred all along the cell cycle, between 0 and 34 h (Fig. 2B).
As PDGF-induced SERCA2a expression seemed to coincide with the entry into the S phase, we checked the effect of TGF-␤ 1 on SERCA expression. Quiescent SMC were incubated for increasing time intervals up to the entry into S phase, i.e. ranging from 0 to 20 h, in the presence of both 5 ng/ml PDGF and 1 ng/ml TGF-␤ 1 , lysed, and analyzed by Western blotting as described above. Fig. 3, A and B, shows the expression of SERCA2a and SERCA2b in these conditions. They clearly demonstrate that the PDGF-induced up-regulation of SERCA2a was totally inhibited by TGF-␤ 1 , as a weak and identical expression of this protein was detected for each incubation time (see Fig. 3A). The expression level of SERCA2b was no more affected with TGF-␤ 1 and PDGF than it was with PDGF alone (Fig. 3B). As an additional control, the effect of TGF-␤ 1 alone (1 ng/ml) was investigated. Results demonstrated that SERCA2a and SERCA2b proteins were expressed at the same level between 0 and 20 h of incubation with TGF-␤ 1 , showing the specificity of its effect in the presence of PDGF (data not shown).

Relationship between Ca 2ϩ Influx and SERCA Regulation in PDGF-induced SMC Proliferation
The first events following the binding of PDGF to its receptor at the surface of SMC membrane are changes in cytosolic Ca 2ϩ concentration. This includes an initial and acute transient increase followed by a lower steady-state elevation, due to both release from Ca 2ϩ pools and entry of Ca 2ϩ from the extracellular media (33,34). As SERCA2 undergoes an isoform switch and hence the associated Ca 2ϩ pool might be affected during PDGF stimulation in SMC, and because Ca 2ϩ store content can be related to Ca 2ϩ influx, we investigated the effect of Ca 2ϩ entry on this process. Thus, we blocked Ca 2ϩ influx with the potent smooth muscle cell Ca 2ϩ channel antagonist SR 33805 and looked for the involvement of this influx on the regulation of the SERCA expression in proliferating SMC. In addition, as we obtained the induction of SERCA 2a expression in human SMC as well (see Figs. 6 and 7), we proceeded further using human SMC.

Characterization of SR 33805
Comparative Effects of Nifedipine, SR 33805, and SKF 96365 on PDGF-induced [Ca 2ϩ ] i Increase in Human SMC-First, in preliminary experiments, we found that 50 ng/ml PDGF-BB maximally induced human SMC proliferation and that it was associated with a rapid rise in cytosolic free Ca 2ϩ . This elevation in [Ca 2ϩ ] i was characterized by an EC 50 value of 15.3 Ϯ 4.8 ng/ml, and the half-time of the Ca 2ϩ accumulation was 1.0 Ϯ 0.4 min (data not shown). To look for the effect of SR 33805 on cytosolic Ca 2ϩ concentration, quiescent cells were incubated with various concentrations of SR 33805 for 1 h, treated for cytosolic free Ca 2ϩ measurements using fluorescent probes as described under "Experimental Procedures," and then stimulated by PDGF-BB. Moreover, because SR 33805 was suggested to possess a different profile of activity than other Ca 2ϩ chan-nel antagonists in PDGF-induced SMC proliferation (20), we investigated the activity of SR 33805 as a capacitative entry blocker by comparing its effect on PDGF-induced Ca 2ϩ increase with those of the dihydropyridine Ca 2ϩ blocker nifedipine and capacitative entry blocker SKF 96365. Fig. 4 shows these comparative effects of nifedipine, SR 33805, and SKF 96365 on PDGF-BB-mediated rise of Ca 2ϩ in SMC which demonstrated a dose-dependent inhibition of Ca 2ϩ increase by the three Ca 2ϩ blockers. With regard to their relative potencies, the higher inhibitory effect was observed by SR 33805 with an IC 50 value of 0.31 Ϯ 0.04 M. Indeed, this value was 7-fold higher than the IC 50 of 2.3 Ϯ 0.8 M obtained using SKF 96365 and 30-fold higher than that of nifedipine which reached 10.1 Ϯ 1.5 M. This meant that the SR 33805 inhibitory effect appeared in agreement with its suggested different mechanism of action than that of typical Ca 2ϩ channel blocker in SMC.
Effect of SR 33805 on Human SMC Proliferation-Quiescent human SMC were treated for 24 h or 3 days with 50 ng/ml PDGF-BB in the absence or in the presence of 10 M SR 33805, trypsinized, and counted. No cytotoxicity could be detected under these experimental conditions and up to 100 M SR 33805 as shown either by lactate dehydrogenase release or trypan blue exclusion test. Fig. 5A shows these results of PDGF-induced SMC proliferation after 24 h (lanes 1 and 2) or 3 days (lanes 3 and 4) of culture in the absence (lanes 1 and 3) or in the presence (lanes 2 and 4) of SR 33805. The control experiments (lanes 1 and 3)  Comparative Effects of Nifedipine, SR 33805, and SKF 96365 on SMC Proliferation-Quiescent human SMC were allowed to grow for 3 days with 50 ng/ml PDGF-BB in the absence or in the presence of various concentrations of the different Ca 2ϩ channel blockers. Fig. 5B shows their dose-response inhibitory effects on PDGF-induced SMC proliferation. No cytotoxicity was detected using either 10 M SKF 96365 or nifedipine as checked by lactate dehydrogenase release (data not shown). An inhibition was detected using the different Ca 2ϩ blockers, al-

Role of Ca 2ϩ Influx on SERCA2 Proteins
Effect of SR 33805 on Total SMC SERCA Activity-The effect of SR 33805 on the combined activity of SERCA2a and SERCA2b was followed on isolated membranes by means of their phosphorylated intermediate (EϳP) complex (Fig. 6). This EϳP corresponds to a transient step in the catalytic cycle of the two isoforms. We also verified that in the presence of 0. The phosphorylated bands in lanes 6 and 7 are representing the SERCA activity after 24 h or 3 days, respectively, of incubation with the SR 33805 alone as a control. As in PDGF-induced porcine SMC proliferation, these results demonstrated the increase in global SERCA activity in PDGF-induced human SMC. When SMC were incubated in the presence of both PDGF and SR 33805, the SERCA activity was very low. Thus, the up-regulation of SERCA activity due to PDGF was abolished by SR 33805. Furthermore, we confirmed that SR 33805 had no effect on the SERCA activity of nonstimulated SMC as a basal EϳP formation was detected.
Effect of SR 33805 on SERCA2a and SERCA2b Protein Expressions-To check whether this effect of SR 33805 on PDGFinduced stimulation of SERCA activity could be explained by changes in the expression level of SERCA2a and/or SERCA2b, or to modifications in the catalytic cycle of these enzymes, we performed Western blottings on isolated membranes from SMC using the same isoform-specific anti-SERCA2a and anti-SERCA2b antibodies as those used in Figs. 2 and 3. Fig. 7 shows the results of Western blottings using the anti-SERCA2a antibody (Fig. 7A) or the anti-SERCA2b antibody (Fig. 7B). On both blots, only one band at 100 kDa was detected corresponding to the recognition of the respective SERCA isoforms. Again,  6 and 7). Thus, neither PDGF nor SR 33805 had a significant effect on SERCA2b in SMC. In conclusion, SR 33805 was clearly shown to block the Ca 2ϩ influx in PDGF-treated SMC, and this was correlated with both an inhibition of SMC proliferation and of the upregulation of SERCA2a. DISCUSSION This work demonstrates the involvement of specific Ca 2ϩ pools, which accumulate Ca 2ϩ through the action of the SERCA2 isoforms of Ca 2ϩ pumps, and shows the importance of Ca 2ϩ influx in the regulation of proliferation of arterial SMC stimulated by PDGF. First, by exploring the expression of the two smooth muscle SERCA isoforms, SERCA2a and SERCA2b, throughout the cell cycle, we could determine that the previously observed PDGF-induced SERCA2a up-regulation coincided precisely with the entry into S phase of the SMC cycle ( Figs. 1 and 2A). This result was confirmed by the action of TGF-␤ 1 , which is associated with the arrest of SMC at the G 1 /S boundary of the cycle and with the concomitant abolishment of SERCA2a expression (Fig. 3A). Second, in an attempt to investigate the role of Ca 2ϩ entry in both PDGF-stimulated SMC proliferation and SERCA expression, using the Ca 2ϩ channel blocker SR 33805, we demonstrated the abolishment of (i) the PDGF-induced SMC proliferation (Fig. 5, A and B); (ii) the associated increase in cytosolic Ca 2ϩ concentration (Fig. 4); and (iii) the SERCA2a up-regulation both in terms of activity and protein levels (Figs. 6 and 7).
This work stresses the role of Ca 2ϩ ions in mitogenesis and suggests a specific function of the SERCA2a isoform, during the cell cycle, and consequently of its associated Ca 2ϩ pool in the regulation of intracellular Ca 2ϩ concentration during the critical period of S phase of the cell cycle. Although an increase in cytosolic Ca 2ϩ acts as a physiological trigger in cell proliferation, the different phases of the cell cycle present different requirements for Ca 2ϩ . Kobayashi et al. (35) showed that pretreatment of aorta in primary culture with ryanodine and NiCl 2 abolished cytosolic Ca 2ϩ concentration transients but did not prevent PDGF-stimulated entry of G 0 into G 1 phase. Conversely, the progression of cells into the S phase and the further mitosis were abolished, suggesting that these latter steps depend on Ca 2ϩ (36) and that growth factors action is governed by Ca 2ϩ -regulated events. Accordingly, cells at G 1 /S boundary enter quiescence in the absence of Ca 2ϩ (37). Besides, these data confirmed several previous studies showing that when Ca 2ϩ concentration could not be raised in the cytosol, the mitogenesis was prevented (38,39). However, on the other hand, SERCA inhibitors such as 2,5-di-(t-butyl)-1,4-benzohydroquinone and thapsigargin, which induce an irreversible increase in cytosolic Ca 2ϩ concentration by emptying Ca 2ϩ pools (40,41), cause a cell growth arrest (14 -16). This depletion of Ca 2ϩ pools would affect G 0 /G 1 and S, but not the subsequent G 2 and M cell cycle phases (18). A possible explanation for this apparent discrepancy might be that although an increase in cytosolic Ca 2ϩ concentration is required for a proliferative state, PDGF-induced smooth muscle cell proliferation also requires the action of SERCA pumps which refill Ca 2ϩ pools at the moment of entry into the S phase, as suggested for proliferation of DTT 1 MF-2 cells (15,16). We effectively demonstrated here a close relationship between the entry into S phase and an increase in functional SERCA2a-associated Ca 2ϩ pool, which can result in an oscillatory Ca 2ϩ response or in a locally lower Ca 2ϩ concentration as a specific cell cycle event necessary for the G 1 phase to S phase transition. The SERCA2a Ca 2ϩ pool may be close to, or included in the nuclear envelop (42,43) and thereby be proximal to the genetic machinery and nuclear events likely to be important for cell cycle control.
Another essential feature of this work concerns a differentiation between the SERCA2a and SERCA2b associated Ca 2ϩ pools in PDGF-induced SMC proliferation. Indeed, because the up-regulation of SERCA2a was strictly correlated with the S phase of the SMC cycle and was found to be specific, whereas the SERCA2b expression was found to be identical along the cell cycle, one can formulate the hypothesis according to which the two SERCA2a and SERCA2b associated Ca 2ϩ pools play distinct roles in the regulation of cytosolic Ca 2ϩ concentration, both in temporal and functional terms. The first events following the binding of PDGF to its receptor at the surface of SMC membrane are changes in cytosolic Ca 2ϩ concentration including an initial and acute transient increase in Ca 2ϩ concentration, due to Ca 2ϩ release from Ca 2ϩ pools FIG. 7. Effect of SR 33805 on SERCA2a and SERCA2b protein expression. 100 g of SMC membranes were reduced, analyzed on 8% SDS-PAGE, and electroblotted onto nitrocellulose sheets. Immunodetection of SERCA2a (A) and SERCA2b (B) was performed by Western blotting using the two isoform-specific anti-SERCA2a and anti-SERCA2b antibodies, as described under "Experimental Procedures." Lanes 1-7 were, respectively, loaded with membranes isolated control SMC, 24-h PDGF-stimulated SMC, 3-day PDGF-stimulated SMC, 24-h PDGF ϩ SR 33805-stimulated SMC, 3-day PDGF ϩ SR 33805-stimulated SMC, 24-h SR 33805-stimulated SMC, and 3-day SR 33805stimulated SMC. Number on the left side of the figures gives the molecular mass of the proteins detected, estimated with standard markers. Results (A and B) are typical of six experiments. (33,34). Based on the fact that normal untreated SMC mainly express the SERCA2b isoform (90%), one can suggest that the first burst of Ca 2ϩ comes from the recruitment of the corresponding Ca 2ϩ pool. Also, as the regulation of the SERCA2a expression seems to depend on Ca 2ϩ influx and is delayed by a few hours with respect to the onset of the PDGF treatment, the most likely interpretation remains that the SERCA 2a Ca 2ϩ pool plays a secondary role to protect the cells from a cytosolic Ca 2ϩ overload.
Finally, from this work, some further evidence for a store depletion-induced Ca 2ϩ influx in SMC (44 -47) can be obtained, based on the comparison of PDGF-induced SMC proliferation and SERCA 2a up-regulation in the presence of TGF-␤ 1 and SR 33805, respectively. First, although the mechanism of the inhibitory effect of TGF-␤ 1 on PDGF-induced proliferation is unknown, it was found abolishing polyphosphoinositide formation (48) and mitogen-induced Ca 2ϩ mobilization (49). So, an alternative is that TGF-␤ 1 presumably inhibits the first burst in Ca 2ϩ increase due to the activation of phospholipase C and the subsequent IP 3 -induced Ca 2ϩ release, from the SERCA2b associated Ca 2ϩ pool. The inhibition of SMC proliferation and SERCA regulation by TGF-␤ 1 would fit in the prime nature of store depletion event in this process. Second, by considering the results obtained in the presence of SR 33805, one can conclude that, although the IP 3 -induced elevation in intracellular Ca 2ϩ concentration following PDGF-BB stimulation can occur, Ca 2ϩ store depletion must also be linked to Ca 2ϩ influx in order to lead to SMC proliferation. Now, such a relationship between Ca 2ϩ influx and intracellular Ca 2ϩ stores constitutes the basis of the concept of the capacitative model of Putney (10) and Putney and Bird (11). Indeed, according to this model, the Ca 2ϩ influx is thought to be regulated by the Ca 2ϩ content of intracellular pools. The importance of SERCAs in this capacitative entry was suggested, since this theory comes from the observation that inhibitors of SERCA, such as thapsigargin, cause depletion of intracellular Ca 2ϩ pools (without IP 3 formation and as a result, mimic the ability of surface membrane, IP 3 agonists, to activate Ca 2ϩ entry) (50). Now, we might suggest the physiological role of SERCAs in the Ca 2ϩ signaling cascade by their function of refilling Ca 2ϩ stores following store depletion and Ca 2ϩ influx, as the further essential step in Ca 2ϩ movements associated with PDGF-induced SMC proliferation.
To conclude, this study brings new basic data regarding the involvement of SERCA proteins in Ca 2ϩ homeostasis with their regulatory role as a crucial event through the SMC cycle and proliferation. Their target step has been defined as well as a way to get the regulation of their expression by drugs interacting with Ca 2ϩ channels. Whether this restored basal SERCA expression is a direct or indirect consequence of this treatment remains to be established, it is of significant importance with regard to the numerous cardiovascular pathologies with abnormal SMC proliferation.