Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Purine and pyrimidine nucleotides are released from a variety of sources and act through P2 receptors (ligand-gated P2X receptor cation channels and G protein-coupled P2Y receptors) to regulate arterial tone. ATP is costored and coreleased with noradrenaline from sympathetic nerves, mediates vasoconstriction through artery P2X receptors (Burnstock, 1997), and can account for up to 65 to 100% of the neurogenic response in resistance arteries (Ramme et al., 1987; Gitterman and Evans, 2001). ATP is also released from endothelial (Dubyak, 2002) and blood cells or because of local tissue damage (Burnstock, 1997) and can produce vasoconstriction through the stimulation of P2X and P2Y receptors (Ralevic and Burnstock, 1998). Platelets release vasoactive diadenosine polyphosphates that act through P2X receptors to mediate vasoconstriction (Schluter et al., 1994; Ralevic et al., 1995). The pyrimidines UTP and UDP are released from endothelial cells and platelets and can mediate sustained vasoconstriction through the activation of pyrimidine-sensitive P2Y receptors (Ralevic and Burnstock, 1998). Thus, nucleotides may provide local and systemic control of blood flow.

Seven P2X receptors subunits (P2X_1-7) have been identified, and the subunits form a variety of homo- and heterotrimeric channels with a range of phenotypes (North and Surprenant, 2000). The P2X₁ receptor subunit is expressed at high levels in artery smooth muscle (Vulchanova et al., 1996), and the properties of the native artery P2X channels (,-meATP-sensitive transient responses that are antagonized by suramin) correspond to those of homomeric P2X₁ receptors (Lewis and Evans, 2000). However, other P2X receptor subunits have also been detected in arteries, e.g., P2X_2, P2X₄, and P2X₅ (Nori et al., 1998; Phillips et al., 1998), and there is pharmacological evidence for the presence of novel diadenosine polyphosphate-sensitive (van der Geit et al., 1999) and suramin-insensitive P2X receptors (Gitterman and Evans, 2000). This raises the distinct possibility that arteries may express heteromeric P2X receptors with properties dominated by the P2X₁ receptor subunit. In addition, at rest, the blood pressure of P2X₁ receptor-deficient mice was normal or slightly elevated (Mulryan et al., 2000), suggesting that P2X₁ receptors may not be essential for the expression of artery P2X receptors. Because of the lack of effective subtype-selective P2X receptor antagonists that can be used in organ-bath studies, it has been difficult to examine directly the role of the P2X₁ receptor in physiological responses or to determine whether other P2X receptor subunits contribute to native artery P2X receptors; therefore, we studied arteries from P2X₁ receptor-deficient mice (Mulryan et al., 2000).

Seven mammalian P2Y receptor subtypes have been cloned: P2Y₁, P2Y₂, P2Y₄, P2Y₆, P2Y₁₁, P2Y₁₂, and P2Y₁₃ (Ralevic and Burnstock, 1998; Communi et al., 2001; Hollopeter et al., 2001). Because of the paucity of selective antagonists, the attribution of molecular correlates of native phenotypes is often taken from agonist potencies. For mouse isoforms, these potencies are the following: mP2Y₁ and mP2Y₁₂ are ADP-sensitive (Fabre et al., 1999; Leon et al., 1999; Foster et al., 2001), mP2Y₂ UTP  ATP (Homolya et al., 1999), mP2Y₄ UTP  ATP > ITP, and mP2Y₆ is a UDP receptor (Lazarowski et al., 2001). Mouse orthologs of P2Y₁₁ and P2Y₁₃ receptors have yet to be cloned, so their exact agonist sensitivities remain to be determined; the human hP2Y₁₁ receptor is ATP-sensitive and pyrimidine-insensitive (Communi et al., 1997), and the canine cP2Y₁₁ receptor is more sensitive to ADP than to ATP (Qi et al., 2001). The hP2Y₁₂ and hP2Y₁₃ receptors are ADP-sensitive and negatively coupled to adenylate cyclase (Communi et al., 2001; Hollopeter et al., 2001), and they are unlikely to contribute to vasoconstriction. P2Y₂, P2Y₄, and P2Y₆ receptors have been shown to mediate vasoconstriction in a variety of species (von Kugelgen and Starke, 1990; Erlinge et al., 1998; Hartley et al., 1998). However, the expression and properties of arterial P2Y receptors in mice is unclear and remains to be determined. Our preliminary studies on mouse mesenteric arteries indicated that there are marked differences in the complement of vasoconstrictor P2Y receptors on these arteries compared with those in the rat, and this may have important considerations for the use of transgenic mouse models for circulation studies.

In this study, we compared P2 receptor-mediated vasoconstriction in mouse mesenteric arteries from normal and P2X₁ receptor-deficient mice to 1) determine the contribution of P2X₁ receptor subunits to the native P2X receptor phenotype in arterial smooth muscle and whether, in the absence of P2X₁ receptors, there is a residual P2X receptor and 2) characterize P2Y receptor-mediated vasoconstrictions in mouse mesenteric arteries.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Adult wild-type (WT, +/+) or P2X₁ receptor-deficient (KO, /) mice (Mulryan et al., 2000) were killed by cervical dislocation and exsanguinated. A portion of the gut with attached mesenteric arcade was removed and placed in Ringer's solution containing 120 mM NaCl, 11 mM glucose, 25 mM NaHCO₃, 5 mM KCl, 1 mM NaH₂PO₄, 2.5 mM CaCl₂, and 2 mM MgCl₂, and then the solution was gassed with 95% O₂/5% CO₂. Mesenteric arteries of different diameters were dissected; large vessels were from the superior mesenteric artery, and medium-sized vessels were from the second- or third-order branches. Femoral, uterine, and tail arteries were also studied.

Immunohistochemical Studies. Mesenteric arteries were dissected as described above and processed for immunohistochemical detection of P2X₁ receptors (Vial and Evans, 2000). The primary antibody directed against the P2X₁ receptor subtype was obtained from Alomone Labs (Jerusalem, Israel).

RT-PCR Studies. Mesenteric arteries were dissected and disrupted using a sterile blade. Isolation of total RNA was processed with RNeasy Mini Kit (QIAGEN, Dorking, Surrey, UK). Total RNA was then treated with deoxyribonuclease I (amplification grade; Sigma Chemical, Poole, Dorset, UK), and cDNA was synthesized using Superscript II Rnase H Reverse Transcriptase (Invitrogen, Carlsbad, CA). Amplification of P2Y receptor subtypes was carried out using BIOTAQ DNA Polymerase (Bioline, London, UK) and the primer pairs shown in Table 1. Amplification of the murine -actin was used as a control of cDNA quality. The PCR thermal profile comprised 5 min at 94°C followed by 35 cycles of 30 s at 94°C, 30 s at 57°C, and 30 s at 72°C. The identity of PCR products was confirmed by sequencing.

Constriction Studies. Changes in arterial diameter were measured in vitro using video-imaging microscopy as described previously (Gitterman and Evans, 2000). Data were presented as changes in internal diameter. Agonists were added to the superfusate at 30-min intervals for purine compounds and 15-min intervals for noradrenaline or KCl. Agonists were washed out after a peak/sustained response was observed. Antagonists were superfused for 15 min before being applied concomitantly with the agonist. Trains of electrical-field stimulation (100 pulses at 10 Hz, 50V, 0.25-ms pulse width) were given at 5-min intervals as described previously (Gitterman and Evans, 2001). Electrically evoked constrictions were reversibly abolished by treatment with tetrodotoxin (0.3 µM), demonstrating that they resulted from nerve stimulation.

Patch-Clamp Recording. Medium mesenteric artery smooth muscle cells were dissociated, and patch-clamp recordings were made in response to rapid U-tube application of drugs, as described previously (Lewis and Evans, 2000). Experiments were performed at a holding potential of 60 mV at room temperature. Voltage-dependent potassium currents were evoked in voltage jumps to +20 mV.

Data Analysis. Data are presented throughout as mean ± S.E.M., with n representing the number of observations. Concentration-response relationships are expressed as the percentage of the maximum response and were fitted by the least-squares method using Origin software (Origin LabCorp, Northampton, MA) with the following equation: response = [A]^n_H/([A]^n_H + [EC₅₀]^n_H).  is the asymptote, n_H is the Hill coefficient, and [A] is the agonist concentration. EC₅₀ is the agonist concentration producing 50% of the maximum agonist response, and pEC₅₀ is log₁₀(EC₅₀). Differences between means were determined by the appropriate Student's t test and were considered significant when P < 0.05.

Drugs. ,-methylene ATP (,-meATP), ATP, cadmium chloride, collagenase, dithioerythritol, noradrenaline, hyaluronidase, papain, prazosin, suramin, UDP, UTP, P¹,P⁵-di(adenosine-5')pentaphosphate (AP₅A) (Sigma-Aldrich), and iso-pyridoxalphosphate-6-azophenyl-2'-5'-disulfonate (iso-PPADS) (Tocris Cookson Inc., Bristol, UK) were used in this study.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

P2X Receptor-Mediated Vasoconstriction in Medium Mesenteric Arteries. The metabolically stable ATP analog ,-meATP evoked concentration-dependent constrictions of medium (mean internal diameter, 90.0 ± 3.5 µm; n = 55) mouse mesenteric arteries (pEC₅₀ = 6.76 ± 0.09, n = 5) (Fig. 1). ATP evoked similar concentration-dependent vasoconstrictions (pEC₅₀ = 4.75 ± 0.12, n = 5), albeit with a reduction in potency compared with ,-meATP (Fig. 1). The low ATP potency probably results from the metabolic breakdown of ATP in the whole-tissue preparation (Benham and Tsien, 1987; Evans and Kennedy, 1994).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Concentration dependence of ,-meATP- and ATP-evoked vasoconstriction in mouse medium mesenteric arteries. a, ,-meATP- and ATP-evoked transient vasoconstriction. The application period is indicated by the bar (arteries shown; resting internal diameter of arteries was 79.3 and 120 µm, respectively). b, concentration-response relationships for ,-meATP- and ATP-evoked vasoconstrictions. The data are plotted as the mean percentage ± S.E. of the maximum response to ,-meATP (n = 6 and 5, respectively).

Effects of P2X₁ Receptor Deficiency on P2X Receptor-Mediated Vasoconstriction and Currents. P2X₁ receptor immunoreactivity was expressed at high levels in the smooth muscle layer of the mesenteric arterial wall of WT mice and was abolished in arteries from P2X₁ receptor-deficient mice or by use of the blocking peptide (Fig. 2a). ,-meATP (10 µM) or ATP (100 µM), which evoked maximal responses in normal arteries, had no effect on the diameter of mesenteric arteries from P2X₁ receptor-deficient mice (Fig. 2b). Diadenosine pentaphosphate (AP₅A, 100 µM) evoked constrictions of shape and amplitude that were similar to those of ,-meATP and ATP in WT mesenteric arteries but had no effect on artery diameter in the P2X₁ receptor / mouse (data not shown).

View larger version (39K): [in this window] [in a new window]   Fig. 2.   P2X1 receptor immunoreactivity and the effects of P2X1 receptor deficiency on P2X receptor-mediated contractions and currents in mouse arterial smooth muscle. a, P2X1 receptor immunoreactivity is localized to the smooth muscle layer of medium mesenteric arteries and reduced to background levels by incubation with control antigen-blocking peptide or in tissues from P2X1 receptor-deficient mice (the residual fluorescence is the autofluorescence of elastic lamina). b, 10 µM ,-meATP- and 100 µM ATP-evoked transient constrictions in WT (+/+) medium arteries were abolished in arteries from P2X1 receptor-deficient (/) mice (arteries shown; resting internal diameter was 105 and 153 µm, respectively). c, 10 µM ,-meATP- and 100 µM ATP-evoked transient inward currents from acutely dissociated smooth muscle cells from WT (+/+) medium arteries. There was no change in the holding current upon application of these agonists from cells from P2X1 receptor-deficient (/) mice. Holding potential, 60 mV; drugs were applied for the period indicated by the bar.

P2X₁ Receptor-Mediated Neurogenic Vasoconstriction. Sympathetic nerves corelease ATP and noradrenaline, and in the majority of peripheral arteries, nerve stimulation results in a vasoconstriction that comprises P2X and ₁-adrenoceptor-mediated components. Sympathetic nerve stimulation (100 pulses at 10 Hz) evoked arterial vasoconstriction; this consisted of an initial rapid peak that declined during the continuation of the train (63.6 ± 2.3% of the initial peak amplitude remains at the end of the 10-s train; n = 8). For WT mesenteric arteries, the P2 receptor antagonist PPADS (30 µM) reduced the amplitude of vasoconstriction by 47.9 ± 6.9% (n = 5). The residual nerve-evoked vasoconstriction was abolished by coapplication of PPADS and the ₁-adrenoceptor antagonist prazosin (0.1 µM). In contrast, in arteries taken from P2X₁ receptor-deficient mice, PPADS had no effect on the amplitude of vasoconstriction (potentiation of 7.6 ± 4.0%, n = 3), but the neurogenic response was abolished by prazosin (Fig. 3A). In addition, the amplitude of neurogenic vasoconstriction was significantly reduced for P2X₁ receptor-deficient arteries (+/+ 11.2 ± 2 µm and / 6.1 ± 1.6 µm, n = 7 and 5, respectively; P < 0.05). These results demonstrate that the P2X₁ receptor makes a substantial contribution to sympathetic nerve-evoked vasoconstriction in WT arteries.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   The contribution of P2X1 receptors to neurogenic vasoconstriction and sensitivity of P2X receptor-mediated vasoconstriction to the calcium-channel blocker cadmium. a, nerve stimulation (100 pulses, 10 Hz) evoked vasoconstriction in WT (+/+) and P2X1 receptor-deficient (/) arteries. The P2 receptor antagonist iso-PPADS (30 µM) reduced the amplitude of vasoconstriction for +/+ but not / arteries (arteries shown; resting internal diameter was 87 and 102 µm, respectively). Nerve-evoked responses in both +/+ and / mesenteric arteries were abolished by the coapplication of iso-PPADS and the 1-adrenoceptor antagonist prazosin (0.1 µM). b, the nonselective calcium-channel blocker cadmium (1 mM) had no effect on vasoconstrictions evoked by an EC90 concentration of ,-meATP (3 µM); in contrast, similar amplitude vasoconstrictions in response to depolarization by potassium chloride (60 mM) were abolished (artery shown; resting internal diameter was 112 µm).

Source of Calcium for P2X₁ Receptor-Mediated Vasoconstriction. P2X₁ receptor-mediated vasoconstrictions to applied agonists were abolished when the extracellular calcium was removed, demonstrating that calcium influx is essential for the contractile response (data not shown). Calcium could enter the cell either directly through the calcium-permeant P2X receptor and/or by the activation of voltage-dependent calcium channels as a result of P2X receptor-induced membrane depolarization. To determine the contribution of calcium influx through voltage-dependent calcium channels, we used the voltage-dependent calcium-channel blocker cadmium. Cadmium (1 mM) abolished responses to depolarization with 60 mM potassium chloride but had no effect on ,-meATP (3 µM)-evoked P2X₁ receptor constrictions (101 ± 8.4% of control response, n = 7) (Fig. 3b). These results indicate that calcium influx directly through the P2X₁ receptor mediates vasoconstriction.

Does the P2X₁ Receptor Deficiency Result in Compensatory Changes? To investigate possible compensatory changes in artery phenotype, we compared concentration-response relationships in WT and P2X₁ receptor-deficient arteries with the application of KCl and noradrenaline. Potassium chloride evoked concentration-dependent vasoconstriction in all arteries tested. Fifty percent of the maximal vasoconstriction was evoked by ~28 to 34 mM KCl for all arteries (Fig. 4a). Similarly, there was no difference in the sensitivity to noradrenaline in WT compared with P2X₁ receptor-deficient mice (pEC₅₀ = 5.27 ± 0.07 and 4.98 ± 0.13, respectively; n = 6 and 7) (Fig. 4B).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Lack of compensatory changes on potassium chloride and noradrenaline evoked vasoconstrictions in P2X1 receptor deficient arteries. The sensitivity to either potassium chloride (a) or noradrenaline (b) was unaffected by the P2X1 receptor deficiency. Data shown are mean ± S.E. mean of the maximum response to KCl (n = 4) or noradrenaline (n = 6).

Characterization of P2Y Receptor-Mediated Vasoconstriction. ATP-sensitive P2Y receptor-mediated vasoconstrictions have been reported widely in many rat arteries (Ralevic and Burnstock, 1998). Therefore, it was a surprise that ATP (an agonist at recombinant mP2Y₂, mP2Y₄, and hP2Y₁₁ receptors) had no effect on the tone of femoral, tail, uterine, and mesenteric arteries from P2X₁ receptor-deficient mice. We focused on the mouse mesenteric artery to characterize the P2Y receptors present. ADP (1 mM), an agonist at mP2Y₁, P2Y₁₂, and P2Y₁₃ receptors, had no effect on the tone of medium mesenteric arteries from P2X₁ receptor-deficient mice (n = 3), and ADP (100 µM) had no effect on arteries in which the tone had been increased with noradrenaline (10 µM) (n = 3). Similarly, the mP2Y₄ receptor agonist ITP (300 µM) (Lazarowski et al., 2001) was ineffective as a contractile agonist on arteries from P2X₁ receptor-deficient mice (n = 3). In contrast, the pyrimidines UTP and UDP evoked concentration-dependent sustained vasoconstriction of normal medium mesenteric arteries with similar potency (pEC₅₀ = 4.74 ± 0.15; pEC₅₀ = 4.97 ± 0.11; n = 5 and 4, respectively) (Fig. 5a). There was no compensatory change in the potency of UTP (pEC₅₀ = 5.05 ± 0.13) or in the amplitude of responses evoked by UDP in mesenteric arteries from P2X₁ receptor-deficient mice. The maximum responses to UTP and UDP were 72.9 ± 14.2% and 86.6 ± 8.6%, respectively, of the maximum response to ,-meATP. UTP and UDP responses persisted in nominally calcium-free extracellular solution (100 ± 12% of the peak response, n = 5) indicative of a G protein-coupled P2Y receptor and not a novel P2X receptor. The P2 receptor antagonist suramin (100 µM) was equally effective in antagonizing responses to an EC₉₀ concentration (300 µM) of UTP or UDP (32.3 ± 9.6% and 38.0 ± 0.1% inhibition, respectively; n = 6) (Fig. 5, b and c). Similarly, the P2 receptor antagonist iso-PPADS (30 µM) had an equivalent inhibitory effect on vasoconstrictions in response to an EC₉₀ concentration (300 µM) of UTP or UDP (34.0 ± 15.6% and 11.0 ± 3.2% inhibition, respectively; n = 4) (Fig. 5, b and c). These results indicate that there is a UTP- and UDP-sensitive P2Y receptor on mouse arteries.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Comparison of UDP- and UTP-evoked vasoconstriction of normal mouse medium mesenteric arteries. a, UDP evoked concentration-dependent vasoconstrictions of mouse medium mesenteric arteries of potency similar to those observed in response to UTP. b, the amplitude of 300 µM UTP-evoked vasoconstrictions are reduced by the P2 receptor antagonists suramin (100 µM) and iso-PPADS (30 µM) (arteries shown; resting internal diameter was 94 and 68 µm, respectively). c, suramin and iso-PPADS had a similar inhibitory effect on 300 µM UTP-evoked vasoconstriction (artery shown; resting internal diameter was 102 µm).

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Identification of P2Y receptor isoforms expressed in mouse medium mesenteric arteries by RT-PCR. RT-PCR showed that P2Y1 (410 bp), P2Y2 (440 bp), and P2Y6 receptors (452 bp) mRNAs were expressed in the medium mesenteric arteries from mouse. However, no P2Y4 receptor (499 bp) mRNA was amplified. -Actin (199 bp) mRNA amplification was used as a positive control for RT-PCR.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we determined the effect of P2X₁ receptor deficiency on the properties of mouse mesenteric arteries and characterized the pharmacology of vasoconstrictor P2Y receptors. The lack of subtype-selective P2X receptor antagonists made it difficult to define conclusively the contribution of the P2X₁ receptor to the regulation of arteries. We show that the P2X₁ receptor underlies the native P2X receptor-mediated responses in arterial smooth muscle and contributes ~50% to sympathetic nerve-evoked vasoconstriction; in addition, a uridine nucleotide-sensitive but ATP-insensitive P2Y₆-like receptor mediates sustained vasoconstriction. Thus, arterial P2 receptors can provide a mechanism for both short- and long-term regulation of blood flow.

In mouse mesenteric arteries, ,-meATP and ATP evoked transient inward currents and concentration-dependent constrictions. These properties are essentially the same as those of P2X receptor-mediated responses in the majority of arteries studied (Kennedy et al., 1986; Benham and Tsien, 1987). In the P2X₁ receptor-deficient mouse ,-meATP- and ATP-evoked responses were abolished in mesenteric, femoral, uterine, and tail arteries. These results demonstrate for the first time that the P2X₁ receptor subunit is essential for the production of functional P2X receptors in a range of arterial smooth muscles. Previous studies have indicated the presence of additional P2X receptor subunits in rat arterial smooth muscle (Nori et al., 1998; Phillips et al., 1998). ATP (100 µM) is an effective agonist at all recombinant P2X receptors, with the possible exception of the P2X₆ receptor, which does not readily form functional channels in recombinant systems; when fully glycosylated, however, it can form functional channels (Torres et al., 1999; North and Surprenant, 2000; Jones et al., 2001). If the native arterial smooth muscle P2X receptor was a heteromeric receptor dominated by the properties of the P2X₁ receptor, one would predict that in the P2X₁ receptor-deficient mouse there would be a residual phenotype resulting from the expression of non-P2X₁ receptor subunits. The lack of residual ATP (100 µM) current or constriction in P2X₁ receptor-deficient mouse arteries demonstrates that the native P2X receptor phenotype in arterial smooth muscle is most likely caused by the expression of homomeric P2X₁ receptors.

A component of the sympathetic nerve-evoked vasoconstriction in peripheral arteries is resistant to the blockade of -adrenoreceptors and is mediated by neurally released ATP acting through ,-meATP-sensitive P2X receptors (Burnstock, 1997). In the present study, the purinergic component accounted for ~50% of the neurogenic response. These stimulation conditions, i.e., a long train of stimulation, have been shown to favor adrenergic transmission, and shorter bursts of stimulation correspond more closely to those recorded under physiological conditions; in resistance arteries, the purinergic component dominates under these conditions (Ramme et al., 1987, Gitterman and Evans, 2001). The characterization of the underlying P2X₁ receptor response to applied agonists and the abolition of P2X receptor-mediated vasoconstriction to agonist application or nerve stimulation in mesenteric arteries from P2X₁ receptor-deficient mice demonstrate that the P2X₁ receptor underlies a significant component of the neurogenic vasoconstriction. This is supported by rat in vivo studies after stimulation of the sympathetic outflow, showing an ,-meATP-sensitive component of the vasoconstriction (Bulloch and McGrath, 1988) and suggesting that P2X receptors may be important in autoregulation in the kidney (Inscho, 2001). Thus, sympathetic nerves releasing ATP and noradrenaline can mediate vasoconstriction through the activation of P2X₁ and ₁-adrenoreceptors. However, at rest, the blood pressure of P2X₁ receptor-deficient mice was normal or slightly elevated (Mulryan et al., 2000). Similarly, in mice lacking noradrenaline, the agonist at -adrenoreceptors and cotransmitter with ATP in sympathetic nerves have normal resting blood pressure (Cho et al., 1999). This suggests that under resting conditions, either P2X₁ receptor or -adrenoreceptor-mediated responses are sufficient to maintain sympathetic regulation of blood pressure. The contribution of P2X₁ receptors to blood pressure under conditions of increased sympathetic tone or in disease states remains to be determined. It is interesting in coronary heart failure that P2X₁ receptor expression is decreased on coronary arterioles (Malmsjo et al., 1999), suggesting that the removal of this endogenous vasoconstrictor may improve blood flow to the heart. In addition, P2X₁ receptor immunoreactivity has been detected in human cerebral arteries (Bo et al., 1998), and P2X₁-like receptors mediate vasoconstriction in the cerebral microvasculature (Lewis and Evans, 2000). Because P2X₁ receptor-mediated arterial constrictions are resistant to -adrenoreceptor and calcium-channel antagonists, they may provide a novel drug target for the treatment of cardiovascular disorders, including heart disease and stroke.

The analysis of native P2Y receptors in smooth muscle has been complicated previously by the presence of ATP-sensitive P2X receptors; for example, in rat arteries, ATP-sensitive P2Y receptor-mediated constriction of arteries has been described previously (Saiag et al., 1990). In the present study, ATP-mediated vasoconstrictions were abolished in a range of arteries from P2X₁ receptor-deficient mice. This was a surprise and indicates that there is marked species variation in P2Y receptor function. UTP and UDP were equipotent at mouse artery vasoconstrictor P2Y receptors, and the purines ADP and ATP were ineffective. RT-PCR studies indicated that P2Y₁, P2Y₂, and P2Y₆ receptors are expressed in mesenteric artery segments; however, whether the RNA transcript amplification corresponds to expression in vascular smooth muscle, endothelial, or blood cells remains to be determined. The lack of ADP- and ATP-evoked responses rules out the functional contribution of P2Y₁ (ADP-sensitive) and P2Y₂ (ATP-sensitive) receptor subtypes (Cressman et al., 1999; Leon et al., 1999). Three subtypes of molecularly identified P2Y receptors are sensitive to uridine nucleotides (P2Y_2, P2Y_4, and P2Y₆). The receptor in the mesenteric arteries cannot be a P2Y₂ receptor because it is insensitive to ATP. Similarly, it is unlikely to be a P2Y₄ receptor because this receptor is below the limit of detection by RT-PCR and because the mouse P2Y₄ receptor agonist ITP (Lazarowski et al., 2001) is ineffective. This leaves the P2Y₆ receptor as a candidate for mediating vasoconstriction.

At recombinant mP2Y₆ receptors, UDP is an order of magnitude more potent that UTP, although it has been suggested that the effects of UTP are actually the result of agonist breakdown to UDP, presumably by ectonucleotidases (Lazarowski et al., 2001). Nucleotidases are active in whole-tissue preparations of mesenteric arteries, and the breakdown of ATP in vasoconstriction studies reduced the apparent potency of ATP ~100-fold (ATP and ,-meATP are equipotent at recombinant P2X₁ receptors and when applied under concentration-clamp conditions in patch-clamp studies to dissociated smooth muscle cells). In the present study, UTP and UDP are equipotent; this suggests that it is unlikely that the agonist actions of UTP result solely from interconversion to UDP by ectonucleotidases or from low levels of UDP contamination of commercially available UTP. Also, the high potency of the pyrimidines compared with many other arterial preparations indicates that there is limited agonist breakdown. This suggests that the receptor most probably corresponds to a P2Y₆-like receptor with increased potency of UTP. Recently it was shown that P2Y and adenosine receptors can dimerize, resulting in a change in their pharmacological properties (Yoshioka et al., 2001). A similar dimerization of P2Y₆ receptors with other P2Y receptors (e.g., P2Y₁ or P2Y₂) could provide a possible explanation of the P2Y₆-like response in mouse mesenteric arteries.

These studies show that arterial vasoconstriction can be rapidly and transiently regulated by ATP released after sympathetic nerve stimulation. They also have firmly established the essential role of P2X₁ receptor ligand-gated cation channels and have shown that these receptors may be novel molecular targets for the regulation of blood flow. Pyrimidine nucleotides are released from endothelial cells and platelets and, after damage to the arterial wall, may act through P2Y₆-like metabotropic receptors, giving rise to sustained vasoconstriction. This work demonstrates that there are marked species differences in P2Y receptor function in arteries. Given the increased use of transgenic mice, the characterization of the P2Y receptors in mouse arteries may have important considerations for studies on circulation.