INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

UNDERSTANDING THE MECHANISMS controlling proliferation of human vascular smooth muscle cells (VSMC) is of considerable importance due to the prevalence of vascular proliferative disease in which intimal smooth muscle cells migrate and proliferate, subsequently occluding the lumen of the blood vessel (26). In coronary artery bypass procedures two vessels are used: internal mammary artery (IMA) and saphenous vein (SV). In a significant number of patients, subsequent problems arise from occlusion of the grafted vessel by this intimal hyperplasia. However, IMA grafts show increased patency over longer periods of time than do SV grafts (4). Interestingly, it has been shown that the serum-stimulated proliferation of smooth muscle cells from SV explants was greater than that for IMA explants (36).

Intimal proliferation of VSMC is controlled by cell surface receptors, which regulate the cell cycle via mitogenic signaling pathways. These include growth factor receptors, G protein-coupled receptors, and integrins (11, 29). Platelet-derived growth factor (PDGF) has been widely studied as a promoter of VSMC proliferation (36). Yang et al. (36) reported greater proliferative responses to PDGF in VSMC derived from SV explants than in those from IMA explants. Rat VSMC in culture are stimulated to proliferate by both ATP and UTP acting via G protein-coupled P2Y receptors (9, 14, 19, 33). Both PDGF and nucleotides are released from activated platelets. In addition, ATP and UTP are known to be released from endothelial cells (17, 27). Nucleotides acting on P2Y receptors, either alone or in conjunction with PDGF, may play an important role in regulating the smooth muscle of blood vessels in both health and disease (3). Currently, there are five cloned and characterized mammalian P2Y receptors (P2Y₁, P2Y₂, P2Y₄, P2Y₆, and P2Y₁₁) that differ in their agonist profiles. These receptors are each coupled to phospholipase C (PLC) and increases in cytosolic free Ca²⁺ concentration ([Ca²⁺]_i). Additional signaling pathways have been studied (2), providing evidence that the ATP-stimulated proliferation of animal VSMC requires activation of p42/p44 mitogen-activated protein kinase (MAPK) pathways (14, 34). Little is known about the regulation of human VSMC by nucleotides. Studies on rings of human SV and coronary artery showed that ATP could regulate vasospasm (12, 25). Furthermore, the responses of human coronary VSMC to ATP and UTP suggested the presence of multiple P2Y receptors (31, 32). Here we investigate the hypothesis that nucleotides regulate proliferative responses in human VSMC, and we report the surprising observation that UTP has an antiproliferative effect.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation and culture of cells. Primary explant cultures were derived from IMA and SV fragments from patients undergoing aortocoronary or peripheral arterial bypass surgery, with the approval of the local Ethical Committee. All comparisons of IMA- and SV-derived cells were made on paired explants from the same patient. Tissue was dissected free of any fat and excess adventitial tissue and then opened along its longitudinal axis, and the endothelial lining was removed by scraping of the luminal surface. Small fragments (~1 mm³) were transferred to a flask containing culture medium (RPMI 1640 with 10% FCS, 2 mM L-glutamine, 50 U/ml penicillin, and 50 µg/ml streptomycin from GIBCO). After VSMC outgrowth, a confluent monolayer was obtained and passaged. Cells were used between passages 2 and 6 after 24 h serum free, when they were quiescent and positive for smooth muscle actin immunostaining with monoclonal mouse anti-human smooth muscle actin (Dako).

[Ca²⁺]_i measurements. Preconfluent cells on coverslips were made serum free for 24 h and loaded for 30 min with 4 µM fura 2-AM (Calbiochem) at room temperature in the dark. Cells were washed twice with PBS, and the coverslip was then placed in a multibarrel perfusion apparatus on the microscope stage (Zeiss microscope fitted with ×20 oil-immersion objective). Cells were perfused continuously with the addition of drugs to the perfusate as indicated. Images at 340 and 380 nm were collected, and ratios were calculated by using a microspectrofluorimetry system with Improvision software. Ratios from individual cells were pooled across fields of 8-20 cells.

[³H]inositol phosphate assay. The procedure was essentially as described previously (35). Cells in 96-well plates were incubated in serum-free M199 (GIBCO) with 2 mM L-glutamine, 50 U/ml penicillin, 50 µg/ml streptomycin, and 3 µCi/ml myo-[³H]inositol (specific activity 86 Ci/mmol; Amersham, Amersham, UK). After the 24-h labeling was completed, cells were preincubated with 50 µl of balanced salt solution (BSS; in mM: 125 NaCl, 5.4 KCl, 16.2 NaHCO₃, 30 HEPES, 1 NaH₂PO₄, 0.8 MgSO₄, 1.8 CaCl₂, and 5.5 glucose, buffered to pH 7.4 with NaOH) containing 10 mM lithium chloride for 10 min. Cells were stimulated with 50 µl of PDGF-BB and/or nucleotides (Sigma) at twice the final concentration. The stimulation period was terminated by the addition of 100 µl of cold 1 M trichloroacetic acid, and then cells were neutralized with tri-n-octylamine/1,1,2-trichlorofluoroethane extraction and separated for total [³H]inositol phosphates ([³H]InsP_x) on small Dowex chloride AG1X8-400 mesh resin columns (Sigma).

Inositol 1,4,5-trisphosphate mass measurement. The levels of inositol 1,4,5-trisphosphate [Ins(1,4,5)P₃] were measured in cells cultured serum free for 24 h in 96-well plates. Cells were stimulated for the times indicated, and extracts were obtained as described in [³H]inositol phosphate assay but in the absence of lithium chloride. Samples (30 µl) were assayed (5) by the addition of ice-cold assay buffer (30 µl) at 4× concentration (final concentration in mM: 25 Tris · HCl and 1 EDTA, pH 8.0), 3.6 µCi in 30 µl of [³H]Ins(1,4,5)P₃ (specific activity 44 Ci/mmol; Amersham) and 30 µl of bovine adrenal cortical binding protein (~0.6 mg protein/tube), with an incubation period of 30 min. A standard curve of D-Ins(1,4,5)P₃ was generated, and nonspecific binding was defined with 10 µM Ins(1,4,5)P₃. The assay was terminated by rapid separation of bound and free [³H]Ins(1,4,5)P₃ using ice-cold washing buffer (in mM: 25 Tris · HCl, 1 EDTA, and 5 NaHCO₃, pH 8.0 at 4°C). The protein-bound [³H]Ins(1,4,5)P₃ was collected onto Whatman GF/B glass fiber filters with the use of a Brandel Cell Harvester and extracted overnight in scintillant before being counted.

[³H]thymidine incorporation. Incorporation of [³H]thymidine (Amersham) into DNA was measured essentially as described previously (35). Cells were cultured close to confluence in 96-well plates and used after they had been serum free for 24 h. Cells in 100 µl of serum-free medium were stimulated for 1 h by the addition of 100 µl of agonist at twice the final concentration. This was followed by incubation for 19 h in serum-free medium and a 4-h incorporation period with [³H]thymidine. DNA was extracted, and [³H]thymidine content was measured by scintillation counting.

Cell proliferation. Cultured cells were seeded at a density of 10⁵ cells/well in 24-well plates and left quiescent for 24 h in serum-free medium before pulse stimulation for 1 h daily with 1 nM PDGF, 300 µM UTP, or both. Cells were stimulated for 2, 4, or 7 days, at which time cell numbers were determined by using a hemacytometer.

Creatine phosphokinase treatment. Regeneration of UTP from UDP during incubation of cells with UTP was achieved by the inclusion of 10 mM creatine phosphate and 20 U/ml creatine phosphokinase (Sigma) for the duration of the 1-h incubation. The UTP was purified by high-performance liquid chromatography (HPLC) immediately before use. HPLC was by anion exchange chromatography (Whatman SAX column). We established that the creatine kinase system was capable of converting UDP to UTP by incubating 100 mM UDP with creatine phosphate and creatine phosphokinase and demonstrated by using HPLC that UTP was formed.

Phospho-MAPK Western blots. Cells were cultured close to confluence in six-well plates and were used after they had been made serum free for 24 h. Cells in 900 µl of serum-free medium were stimulated by the addition of 100 µl of agonist at 10-fold final concentration for 5 min. Stimulation was stopped by the addition of liquid N₂, and the cells were extracted into lysis buffer [in mM: 20 Tris · HCl, 250 NaCl, 3 EDTA, 3 EGTA, 1 phenylmethylsulfonyl fluoride, 2 sodium orthovanadate, and 1 -mercaptoethanol plus 0.5 % (vol/vol) Triton X-100, 20 µg/ml aprotinin, and 5 µg/ml leupeptin, pH 7.6]. The sonicated lysate was cleared by centrifugation at 14,000 g for 15 min at 4°C, samples were protein equalized, and Western blots were prepared with the use of an antibody specific for the phosphorylated (activated) forms of p42/p44 MAPK (Promega). The blots were developed with the enhanced chemiluminescence (ECL) procedure (Amersham).

Statistical analysis. Data presented are the means ± SE for three separate experiments performed in triplicate or quadruplicate. Raw data were analyzed with GraphPad Prism (version 3.0) by a one-way analysis of variance (ANOVA) followed by Bonferroni's posttest to compare selected columns or Dunnett's posttest to compare all columns with controls. When two groups were compared, a two-tailed Student's t-test was used.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

[Ca²⁺]_i and PLC responses. Fura 2-loaded VSMC derived from both IMA and SV responded to ATP and PDGF with increased [Ca²⁺]_i, as shown in Fig. 1, A and B. UTP similarly stimulated an increase in [Ca²⁺]_i in both IMA- (Fig. 1C) and SV-derived cells (data not shown). No consistent difference was seen in the response of cells to UTP or ATP over a number of experiments or in the size of the response of IMA-derived cells compared with that of SV-derived cells. The P2X agonist ,-methylene-ATP (300 µM) did not raise [Ca²⁺]_i, and the removal of extracellular Ca²⁺ with EGTA did not significantly attenuate the response to ATP (data not shown). To determine whether the effects of ATP and UTP were consequent to PLC activation and Ins(1,4,5)P₃ generation, we measured nucleotide-induced accumulation of [³H]InsP_x was measured. As shown in Fig. 2A, PDGF stimulation of IMA-derived cells led to [³H]InsP_x accumulation. However, stimulation with ATP or UTP alone did not lead to an increase in [³H]InsP_x accumulation. In addition, in the presence of PDGF, the application of ATP and UTP did not lead to any further increase in [³H]InsP_x accumulation (Fig. 2A). These responses were the same for SV-derived cells (data not shown). To investigate the possibility that the nucleotides elicit a transient formation of Ins(1,4,5)P₃ that is not detected with the [³H]InsP_x procedure, we measured the mass of Ins(1,4,5)P₃ formed within seconds of stimulation. As shown in Fig. 2B, no increase in Ins(1,4,5)P₃ level was detected between 10 s and 5 min of stimulation in response to PDGF, despite the modest accumulation of [³H]InsP_x levels (Fig. 2A). As a positive control for the mass Ins(1,4,5)P₃ procedure, SH-SY5Y neuroblastoma cells were stimulated with carbachol and assayed concurrently with the VSMC; a clear increase in Ins(1,4,5)P₃ accumulation was seen (Fig. 2B, inset).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Intracellular Ca2+ responses to platelet-derived growth factor (PDGF), ATP, and UTP in human vascular smooth muscle cells (VSMC). Data are the averaged 340/380 ratios from 8-20 serum-starved cells in response to 1 nM PDGF, 300 µM ATP, or 300 µM UTP perfused over internal mammary artery (IMA) cells (A and C) or saphenous vein (SV) cells (B) as indicated. Traces are representative of at least 3 separate experiments on different cell preparations.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Phosphoinositide-directed phospholipase C responses in human VSMC. A: effect of ATP or UTP (300 µM) in the absence (hatched bars) or presence (solid bars) of PDGF (1 nM) on the accumulation of [3H]inositol phosphates ([3H]InsPx) in IMA-derived cells. Data are means ± SE; n = 3. Cont, control. **P < 0.01 compared with equivalent responses in the absence of PDGF, by ANOVA with Dunnett's posttest. B: level of Ins(1,4,5)P3 in response to stimulation with PDGF (1 nM), ATP, or UTP (300 µM) for the times indicated. Data are means ± SE; n = 3 separate experiments, each in quadruplicate. One-way ANOVA showed no significant effect of time of incubation with agonists. Inset: response of SH-SY5Y cells to 1 mM carbachol (or vehicle control) for the times indicated (n = 3).

[³H]thymidine responses. To compare the mitogenic responses of SV- and IMA-derived cells to PDGF, we used [³H]thymidine incorporation into DNA as an index of proliferation. As shown in Fig. 3A, increasing concentrations of PDGF led to increased incorporation of [³H]thymidine. PDGF was significantly more potent in IMA-derived cells than in the paired SV-derived cells (pEC₅₀: 9.22 ± 0.11 for IMA-derived cells and 8.45 ± 0.08 for SV-derived cells; P < 0.001, n = 3 experiments each in quadruplicate). However, the maximal response was greater in SV-derived cells (SV: 4,828 ± 330 dpm/mg protein; IMA: 1,792 ± 121 dpm/mg protein, P < 0.05, n = 3). The effect of passage number on [³H]thymidine incorporation was studied by comparing cells from passages 2-4 in paired IMA- and SV-derived cells from three different patients, with three separate experiments in quadruplicate for each. There was no effect of passage number on the [³H]thymidine incorporation to increasing concentrations of PDGF (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Stimulated [3H]thymidine incorporation into DNA in human VSMC in response to PDGF (1 nM) and ATP (300 µM). A: concentration-response curves to PDGF for SV- and IMA-derived cells. Statistical analysis of data is presented in the text. B: [3H]thymidine incorporation (dpm) expressed as a percentage of control values in IMA (left) or SV (right) cells. Data are means ± SE; n = 3 separate experiments, each in quadruplicate. **P < 0.01 compared with unstimulated control. PDGF + ATP was also significantly different from PDGF alone for the SV cells (P < 0.01, by ANOVA with Bonferroni's posttest).

View larger version (32K): [in this window] [in a new window]   Fig. 4.   [3H]thymidine incorporation into DNA of human VSMC in response to PDGF and various nucleotides in IMA (A) or SV (B) cells, expressed as mean percentages above unstimulated (basal) control values (indicated by the solid line). The broken line indicates the level of [3H]thymidine incorporation evoked by 1 nM PDGF alone (PDGF control). Data are means ± SE; n = 3 separate experiments, each in quadruplicate. **P < 0.01; ***P < 0.001 compared with unstimulated control. In A, PDGF + UTP and PDGF + UDP were both different from PDGF alone (P < 0.01). In B, differences from PDGF alone were significant for PDGF + UTP (P < 0.01) and PDGF + UDP (P < 0.001). Statistics were determined by ANOVA followed by Bonferroni's multiple comparison test.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   A: effect of ATP or UTP (300 µM) on [3H]thymidine incorporation evoked by 1 nM PDGF, expressed as a percentage of the level of incorporation with PDGF alone, in IMA (left) or SV (right) cells. Data are means ± SE; n = 3. ***P < 0.001; *P < 0.05 compared with PDGF alone, by ANOVA and Dunnett's posttest. B: concentration-response curve for inhibition of PDGF response by UTP. PDGF, in absence of PDGF; +PDGF, in presence of 1 nM PDGF throughout; ns, not significant. Incubations were carried out in the presence of a creatine phosphokinase-regenerating system as described in METHODS.

Cell proliferation. To measure cell proliferation directly, we counted cells over a number of days in response to repeated stimulation with PDGF in the presence or absence of UTP. Stimulations were daily, over a period of 1 h, at which point the medium was changed for serum-free medium with no agonists present. Table 1 shows results with both SV- and IMA-derived cells pooled across experiments, with stimulation for 2, 4, and 7 days. PDGF stimulation for 2 days alone had no effect, but over 4 or 7 days there was a significant increase in cell number. This increase was greater for SV-derived cells than for IMA-derived cells. UTP alone had no effect on cell number at any time. However, when UTP was present, there was no increase in cell number in response to PDGF. This was true for both cell types. These results show that UTP exerted a powerful antiproliferative effect on both SV- and IMA-derived cells.

Activation of MAPK. Consistent with an earlier report (36), PDGF stimulated phosphorylation of p42/p44 MAPK in both IMA- and SV-derived cells (Fig. 6). Stimulation of SV-derived cells with ATP alone led to an increased phosphorylation of p42/p44 MAPK, but ATP did not stimulate MAPK phosphorylation in IMA-derived cells. When ATP was applied in combination with PDGF, there was no detectable enhancement of p42/p44 MAPK phosphorylation (Fig. 6) over the response to PDGF alone. UTP had no effect on p42/p44 MAPK phosphorylation when added alone to either cell type. In SV-derived cells, but not in IMA-derived cells, UTP enhanced the PDGF-stimulated MAPK phosphorylation.

View larger version (41K): [in this window] [in a new window]   Fig. 6.   Phosphorylation of p42/p44 mitogen-activated protein kinase (MAPK) in human VSMC in response to ATP and UTP either alone or in the presence of PDGF in IMA (A) or SV (B) cells. Incubations were carried out for 5 min with agonists present as indicated. Western blots were specific for the phosphorylated forms of p42/p44 MAPK. Data are representative of 3 separate experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study describes the influence of nucleotides on control of proliferation of human VSMC derived from both veins and arteries. In a healthy blood vessel, the endothelium provides an antiproliferative influence through agents such as nitric oxide and prostacyclin. However, in vascular disease, when endothelial function is compromised, the influence of these agents is diminished. Additionally, proliferative stimuli in the blood may have direct access to exposed smooth muscle cells, leading to intimal proliferation. It has been proposed that nucleotides such as ATP act in this capacity through their cell surface receptors (3). In the diseased vessel, nucleotide release from platelets (together with PDGF) will be enhanced due to downregulation of the anti-platelet influence of endothelium-derived nitric oxide and prostacyclin. In this model the P2Y receptors on VSMC play a critical role in the pathology of intimal proliferation.

The importance of P2Y receptors in this respect has been suggested by previous studies on animal VSMC (9, 14, 19, 33). In intact rat arterial tissue, ionotropic P2X receptors play a role in the contractile response. In culture, these cells lose their P2X receptors, while a P2Y response is found in both intact tissue and cultured cells (21, 24). A recent quantitative PCR study has shown that the transition from the contractile to the synthetic phenotype, which is necessary for intimal proliferation (and also occurs when cells are cultured), corresponds to a loss of mRNA for P2X₁ and an upregulation of mRNA for P2Y₁ and P2Y₂ receptors (8). This is consistent with the hypothesis that P2Y receptors regulate the proliferative response in the synthetic phenotype.

The increases in [Ca²⁺]_i reported here in both human venous and arterial smooth muscle cells provide clear evidence for the presence of receptors that respond to ATP and UTP. The studies presented do not enable the identification of these receptors, yet the results indicate that the nucleotides are acting at P2Y receptors. The response to UTP and the lack of effect of ,-methylene-ATP, together with the response to ATP in the absence of extracellular Ca²⁺, all argue against the involvement of P2X receptors. The demonstration that UTP stimulated PLC and increased [Ca²⁺]_i in rat VSMC (16, 24) led to the proposal of a P2Y receptor sensitive to UTP, both in the original smooth muscle cells and in culture (24). Because all of the cloned P2Y receptors are coupled to the release of intracellular Ca²⁺ via Ins(1,4,5)P₃, the lack of a detectable PLC response reported here was unexpected. However, a similar situation has been reported in other cell types in response to activation of native P2Y receptors (1, 10) and other G protein-coupled receptors (20, 22, 30), in which the [Ca²⁺]_i increase is apparently independent of Ins(1,4,5)P₃.

Mitogenic responses to ATP and UTP have been reported in cultured rat VSMC (9, 14, 19, 33). In some cases nucleotides act as progression factors, stimulating the cell cycle in concert with growth factors such as PDGF. This growth factor has been shown to stimulate the progression of quiescent rat aorta cells from G₀ to G₁ but no further (23). ATP acting on P2Y receptors had no effect on the cell cycle of cells in G₀ but could lead to progression from G₁ to S and M phases. These observations indicate that concomitant stimulation of VSMC with growth factors such as PDGF and agonists acting at P2Y receptors leads to proliferation.

The results of our experiments in which IMA and SV cells were stimulated with PDGF are consistent with previous work (29). We report that the intracellular Ca²⁺ response was of a similar magnitude in SV- and IMA-derived cells, while [³H]thymidine incorporation was greater in the SV cells. The larger maximal response to PDGF seen in SV cells occurred with an EC₅₀ that was significantly higher than that occurring in the IMA cells; this may reflect a difference in receptor reserve between these two cell types.

ATP did not act as a full mitogen in either SV or IMA cells; no stimulation of [³H]thymidine incorporation was observed in response to this nucleotide alone. However, in SV cells, ATP acted as a progression factor to enhance the PDGF-stimulated [³H]thymidine incorporation into DNA, consistent with the effects described in rat VSMC (23). However, this comitogenic action of ATP was not seen in IMA cells, despite the observation that the intracellular Ca²⁺ response of IMA cells to ATP is similar to that of SV cells. Differences were also seen in the effects of ATP on p42/p44 MAPK phosphorylation in the two cell types. In SV-derived cells ATP stimulated this index of MAPK activation, but this did not occur in IMA-derived cells; interestingly, this correlates with the ability of ATP to act as a comitogen. However, the addition of ATP in the presence of PDGF failed to further elevate MAPK phosphorylation above that of PDGF alone in either cell type. These results are consistent with the suggestion that activation of MAPK is necessary but not sufficient for the mitogenic response (14, 31).

A salient observation here is that ATP exerts a proliferative influence only in SV cells, and this may contribute to the increased intimal proliferation following vascular grafts with SV compared with grafts with IMA. This is consistent with the hypothesis that P2Y receptors on VSMC play a role in the progression of vascular proliferative disorders. 2-MeS-ADP had no proliferative effect, suggesting that P2Y₁ receptors (for which this is the most potent agonist) are not involved. ATP and UTP are equally effective at the P2Y₂ receptor, and the lack of proliferation in response to UTP suggests that the response is not mediated by this receptor subtype.

The reduction in the PDGF-stimulated [³H]thymidine incorporation and proliferation of both SV and IMA cells by the application of UTP was an unexpected finding. The conclusions we derive from the thymidine incorporation studies are powerfully confirmed by direct measurement of cell numbers. It takes several days for the cell numbers to be reliably increased by PDGF, but wherever this occurs it is essentially ablated by the presence of UTP. Indeed, the antiproliferative effect of UTP is apparently greater when counting cell number than when looking at [³H]thymidine incorporation.

There was one previous report of an antiproliferative response to UTP in endothelium-derived cells (18), although it was also reported that UTP is mitogenic in cardiac vasculature endothelium (28). Here we have shown that UTP can act directly; its antiproliferative action is not dependent on breakdown by ectonucleotidases. This is demonstrated by experiments in which HPLC-purified UTP was used in the presence of a creatine phosphokinase-regenerating system, which will convert any UDP formed during the incubation back into UTP. This system is usually used as an ATP-regenerating system (15), but we have shown directly that UDP can act as a substrate, although the rate of the kinase reaction is much slower than that with ADP as substrate. These experiments confirm that the antiproliferative P2Y receptor is responsive to UTP. However, the antiproliferative response to UDP observed here may be indirect and dependent on extracellular conversion to UTP. Harden and colleagues (13) showed that the addition of UDP can result in accumulation of extracellular UTP by the action of ectonucleoside diphosphokinase.

Candidates for the receptor responsible for the antiproliferative response to UTP include P2Y₂ and P2Y₄. Of these, the P2Y₄ receptor seems most likely since ATP, which does not elicit an antiproliferative response, is an agonist at P2Y₂ receptors but not at human P2Y₄ receptors (7). The suramin and PPADS results are consistent with this. Neither compound attenuated the effect of UTP on PDGF-stimulated [³H]thymidine incorporation. Both are effective antagonists at P2X and P2Y₁ receptors, suramin is a weak antagonist at P2Y₂ receptors, and neither are antagonists at P2Y₄ receptors (3, 6).

PDGF-stimulated proliferation was inhibited to a similar degree by UTP in both IMA- and SV-derived cells. This effect was apparently independent of Ca²⁺, since [Ca²⁺]_i was elevated by both ATP and UTP, but only UTP had an antiproliferative effect. Similarly, the antiproliferative effect showed no correlation with activation of p42/p44 MAPK, since UTP had no effect on PDGF-mediated MAPK phosphorylation in IMA-derived cells. These results rule out changes in [Ca²⁺]_i and p42/p44 MAPK phosphorylation state as the mechanism underlying this antiproliferative effect.

Our initial hypothesis was that nucleotides act on vascular smooth muscle P2Y receptors to enhance proliferation of these cells, thus contributing to intimal proliferation in diseased vessels. The results presented here are consistent with this hypothesis with respect to the action of ATP and indicate that differential activation of SV-derived cells by ATP may contribute to greater intimal proliferation seen with these vessels compared with that seen with IMA. However, our results indicate that the effects of nucleotides are more complex than we initially suggested, since they show that UTP is an antiproliferative regulator of human vascular smooth muscle of both arterial and venous origin.