Materials and Methods

Patients

Hearts were explanted during the process of heart transplantation from eight patients who were between 17 and 64 years of age and had ishemic heart disease or dilated cardiomyopathy. No difference in contractile responses could be observed between the groups according to diagnosis.

Vasomotor Studies

Tissue Preparation.

The hearts were immediately examined in the operating room, and epicardial segments from the left anterior descending and the right coronary arteries and their branches were removed gently and immersed in cold oxygenated buffer solution (for composition, see later), transported to the laboratory, and dissected free of adhering tissue under a microscope. The vessels had a relaxed inner diameter of approximately 1 mm. The endothelium was removed by perfusion for 5 s with 0.1% Triton X-100, followed by an additional 5 s of perfusion with a physiologic buffer solution (for composition, see below) using a fine needle. The vessels were then cut into cylindrical segments (2–3 mm long) and were immediately used in the experiments. Each cylindrical segment was mounted onto two L-shaped metal prongs, one of which was connected to a force displacement transducer (FT03C) for continuous recording of the isometric tension, and the other was connected to a displacement device. The position of the holder could be changed by means of a movable unit that allowed fine adjustments of the vascular resting tension by varying the distance between the metal prongs. The mounted artery segments were immersed in temperature-controlled (37°C) tissue baths containing bicarbonate-based buffer solution composed of 119 mM NaCl, 15 mM NaHCO₃, 4.6 mM KCl, 1.2 mM MgCl₂, 1.2 mM NaH₂PO₄, 1.5 mM CaCl₂, and 5.5 mM glucose. The solution was continuously gassed with 5% CO₂ in O₂ resulting in pH 7.4. Twelve ring segments were studied at the same time in separate tissue baths. The segments were allowed to stabilize at a resting tension of 4 mN for 1 h before the experiments were started. The contractile capacity of each vessel segment was examined by exposure to a K⁺-rich (60 mM) buffer solution in which NaCl was exchanged for an equimolar concentration of KCl (for composition, see later). When two reproducible contractions had been achieved, the vessels were used for further studies.

Vasomotor Responses.

Endothelium removal was checked by monitoring responses to acetylcholine. Abolished dilatation indicated a properly removed endothelium. Unaffected K⁺-induced contractions indicated intact vascular smooth muscle cell function. Because the P2X receptors were quickly desensitized, each artery segment was exposed to a single concentration of αβ-MeATP or ATP, and the resultant responses of several segments exposed to different concentrations were grouped together. In this way, each artery segment was exposed to αβ-MeATP or ATP only once, and the problem of tachyphylaxis was avoided. These experiments are referred to as “single concentration”. To study the P2Y receptor-stimulated contractions without interference by the simultaneous activation of P2X receptors, UDP, UDPβS, UTP, UTPγS, ADPβS ATP, ATPγS, and Ap₄A were added after P2X receptor desensitization with 10 μM αβ-MeATP 8 min before each experiment. Because the P2Y receptors are only very slowly desensitized, these agonists could be added cumulatively to determine concentration-response relationships.

Reverse Transcription-Polymerase Chain Reaction (RT-PCR)

RNA Extraction.

The human coronary arteries were carefully dissected, and the endothelium was removed (see earlier). The arteries were snap-frozen in liquid nitrogen immediately after acquisition, and total cellular RNA was extracted using TRIzol reagent (Life Technologies, Grand Island, NY) according to the supplier's instructions. The resulting RNA pellet was finally washed with 70% ice-cold ethanol, air-dried, and redissolved in 10 μl of diethyl pyrocarbonate-treated water. The RNA concentration was determined spectrophotometrically considering anA₂₆₀:A₂₈₀ ratio of ≥1.6 as pure.

RT-PCR.

RT-PCR was carried out using the GeneAmp RNA PCR kit (Perkin-Elmer, Foster City, CA) on a GeneAmp PCR system 2400 (Perkin Elmer Cetus, Norwalk, CT). Specific primers for the human P2X₁, P2Y₁, P2Y₂, P2Y₄, and P2Y₆ receptors were constructed based on published nucleotide sequences (Parr et al., 1994; Communi et al., 1995, 1996; Valera et al., 1995; Léon et al., 1996): P2X₁ forward (5′-GTTTGGATTCGCTTTGA-3′) and P2X₁ reverse (5′-GGCTGAGAGGGTAGGAGAC-3′) (383 bp); P2Y₁ forward (5′-ACGGGCTTCCAGTTTTAC-3′) and P2Y₁ reverse (5′-CCAAGGGGACACAGAACA-3′) (550 bp); P2Y₂ forward (5′-CTCTGCTTCCTGCCATTC-3′) and P2Y₂ reverse (5′-GCACAAGTCCTGGTCCTCT-3′) (432 bp); P2Y₄ forward (5′-CAGCACCAAAGGGACCAC-3′) and P2Y₄ reverse (5′-GCTTGCCACCACACAGA-3′) (530 bp); P2Y₆ forward (5′-ACAGGCATCCAGGGTAAC-3′) and P2Y₆ reverse (5′-CGGACACAATGGCAAATA-3′) (526 bp).

First-strand cDNA synthesis was carried out with the Amplitaq RNA-PCR kit (Perkin-Elmer Cetus) in a 20-μl volume using random hexamers. Amplification was performed using a modified profile (2 min at 95°C followed by 30 cycles of 1 min at 95°C, 1 min at 55–58°C, 30 s at 72°C, and a final extension step of 7 min at 72°C). The products were separated on a 2% agarose gel containing 1.0 μg/ml ethidium bromide and photographed. The DNA Ladder 100 bp (Promega, Madison, WI) was used as the molecular weight marker. Because these P2 receptors are intronless within their coding regions, PCR without the RT step was always used to exclude genomic DNA contamination.

Ethics

The project was approved by the Ethics Committee of Lund University in Sweden.

Drugs

Agonist purity and stability provide potential problems when delineating the pharmacological selectivity of P2 receptors, especially in intact tissues. To avoid phosphorylation of UDP, hexokinase and glucose were used to convert UTP to UDP (Nicholas et al., 1996). Stock solutions of UDP in a concentration of 0.1 M were preincubated for 1 h with 10 U/ml hexokinase and 22 mM glucose. Metabolism of nucleotides was also prevented by using more stable compounds, namely, UDPβS, UTPγS, ADPβS, and ATPγS. These include a thio substitution at the terminal phosphate, which provides stability to ectonucleotidase action (Jacobson et al., 1998).

Acetylcholine, ADPβS, Ap₄A, ATP, ATPγS, hexokinase, UDP, UTP, and αβ-MeATP were purchased from Sigma Chemical Co. (St. Louis, MO). UDPβS and UTPγS were gifts from Inspire Pharmaceuticals, Inc. (Durham, NC). All drugs were dissolved in 0.9% saline. If not stated otherwise, all reagents for the RT-PCR assay were purchased from Life Technologies.

Calculations and Statistics

The negative logarithm of the drug concentration that elicited 50% contraction (pEC₅₀) was determined by linear regression analysis using the values immediately above and below the half-maximum response. E_max refers to maximum contraction calculated as a percent of the contractile capacity of 60 mM K⁺. For UDP, UDP plus hexokinase, UTP, ATP, and Ap₄A, a plateau phase of the maximum contractile response was not reached within the agonist concentration interval. The real E_max value for respective agonist will therefore be equal to or higher than the obtained value. pEC₅₀ was in these cases calculated as the negative logarithm of 50% of the contraction reached at the highest concentration used and was marked as equal to or less than the obtained value.

Pharmacological experiments were repeated six to eight times (patients) for each substance, and statistical significance was accepted when P < .05, using Student's ttest. All differences referred to in the text have been statistically verified. Values are presented as mean ± S.E. RT-PCR experiments were repeated three times (patients) with similar results.

Results

Vasomotor Responses.

The contractile response of human coronary arteries to 60 mM K⁺ was 15.3 ± 1.3 mN.

Endothelium Removal.

After endothelium denudation, vascular relaxations to acetylcholine were abolished, indicating a properly removed endothelium. Vascular smooth muscle cell function was considered intact, as the contractile responses to 60 mM K⁺ were unaffected.

Contractile P2X Receptors.

αβ-MeATP induced contraction at a low concentration (pEC₅₀ = 5.2 ± 0.1,E_max = 68 ± 3%), suggesting the presence of contractile P2X receptors (Table1 and Fig. 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Concentration-dependent contractions of human coronary arteries to single concentrations of αβ-MeATP and cumulative concentrations of UDPβS, UDP, UTPγS, UTP, ATPγS, ATP, ADPβS, and Ap₄A after desensitization with 10 μM αβ-MeATP. Contractions are expressed as a percentage of the response to 60 mM K⁺. Data are shown as mean ± S.E. of six to eight experiments (patients).

P2X Receptor Desensitization.

The addition of 10 μM αβ-MeATP elicited a transient contraction. After 8 min, the tension returned to normal levels, and if αβ-MeATP was added a second time, no contraction could be observed, indicating desensitized P2X receptors.

Contractile P2Y Receptors.

The contractile P2Y receptor subtypes were examined after P2X receptor desensitization (as described earlier). The endogenous nucleotides UDP, UTP, ATP, and Ap₄A elicited similar contractions in a concentration-dependent manner, although high concentrations were required (1–10 mM). The more stable UTP and ATP analogs, UTPγS and ATPγS, were 1 to 2 log units more potent (pEC₅₀= 4.6 ± 0.1 and 3.8 ± 0.1, respectively). The efficacy of these P2Y-mediated responses (E_max = 125 ± 25 and 126 ± 11% for UTPγS and ATPγS) were approximately double that of the P2X agonist αβ-MeATP (see earlier). When UDP was preincubated with hexokinase to increase purity, the contractile response decreased considerably, whereas the selective P2Y₆ agonist UDPβS was ineffective. The selective P2Y₁ agonist ADPβS did not stimulate contraction. In conclusion, the rank order potency of extracellular nucleotides for the P2Y receptor-mediated contractions was UTPγS > ATPγS > Ap₄A = UTP = ATP = UDP = UDP plus hexokinase. UDPβS and ADPβS did not stimulate contraction (see Table 1 and Figs. 2 and 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Concentration-dependent contractions of human coronary arteries in response to UDP, UDP plus hexokinase, and UDPβS after P2X receptor desensitization with 10 μM αβ-MeATP. Contractions are expressed as a percentage of the response to 60 mM K⁺. Data are shown as mean ± S.E. of six to eight experiments (patients).

PCR.

Agarose gel electrophoresis of PCR products from human coronary arteries demonstrated products of the expected size for the corresponding mRNA encoding human P2X₁ (383 bp), P2Y₁ (550 bp), P2Y₂ (432 bp), P2Y₄ (530 bp), and P2Y₆ receptors (526 bp). Bands for P2X₁ and P2Y₂ receptor mRNAs were especially prominent. No bands were detected in controls without an RT step. An example is shown in Fig. 4.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Electrophoresis of RT-PCR products corresponding to mRNA encoding human P2X₁, P2Y₁, P2Y₂, P2Y₄, and P2Y₆ receptors in human coronary arteries. The amplified products were of the predicted size for P2X₁ (383 bp), P2Y₁ (550 bp), P2Y₂ (432 bp), P2Y₄ (530 bp), and P2Y₆ receptors (526 bp). Samples were analyzed on a 2% agarose/ethidium bromide gel at 5 V/cm and photographed.

Discussion

The effects of extracellular nucleotides have received increasing attention since the first report of a role for ATP as cotransmitter (Burnstock et al., 1978). The same group later demonstrated a biphasic response to ATP in isolated perfused rat hearts, with an increase followed by a decrease in perfusion pressure due to the activation of P2X and P2Y receptors, respectively (Hopwood and Burnstock, 1987). Indeed, pharmacological and molecular studies confirm that smooth muscle cells of vascular beds may simultaneously express several P2 receptor subtypes, and the observed responses to extracellular nucleotides should be considered a result of their multiple sites of action. Characterization of the P2 receptors in human coronary arteries here shows that extracellular nucleotides primarily stimulate P2Y₂ receptors and P2X receptors (less efficaciously), whereas a role for P2Y₁ and P2Y₆ receptors can be excluded. mRNA could be detected by RT-PCR for P2X₁, P2Y₁, P2Y₂, P2Y₄, and P2Y₆ receptors.

There is strong evidence that contractile P2 receptors are located solely on vascular smooth muscle cells and that the activation of endothelial P2 receptors stimulates the release of dilatory mediators. The addition of platelets or platelet suspensions, which contain high concentrations of nucleotides, to coronary arteries induced direct constriction, whereas the dilatation was endothelium-dependent (Cohen et al., 1983; Houston et al., 1986). Furthermore, endothelium denudation inhibited coronary artery dilatation to ATP and UTP, whereas the constriction was unaffected (Matsumoto et al., 1997a). Therefore, in the present study, the endothelium was removed before the contractile responses to extracellular nucleotides were examined.

P2X receptor desensitization with αβ-MeATP has been shown to abolish the response to ATP in rabbit coronary arteries, whereas the UTP contraction was unaffected, suggesting that the contractile effect of extracellular nucleotides is mediated by both P2X and P2Y receptors (Corr and Burnstock, 1994; Matsumoto et al., 1997a). Our results demonstrate that αβ-MeATP induces contraction at a low concentration, indicating the presence of contractile P2X receptors. This contraction was relatively weak compared with the P2Y effects of the stable nucleotide analogs (see later; Table 1 and Figs.1 and 2). The P2Y receptor-induced responses were studied here after P2X receptor desensitization with αβ-MeATP, as previously described by Kasakov and Burnstock (1983).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Representative example of contractions induced by cumulative concentrations of UTPγS in human coronary arteries after P2X receptor desensitization with 10 μM αβ-MeATP. At the end of the experiment, UTPγS was removed by washing (w). αβ-MeATP was added in single concentrations to different vessel segments to avoid desensitization. The tension of the segments was measured in millinewtons.

Cross-desensitization experiments with UDP and UTP, after P2X receptor desensitization, have revealed responses by different populations of pyrimidine-sensitive P2Y receptors in canine coronary arteries (Matsumoto et el., 1997b). It was suggested that UDP induces vascular constriction via UDP-preferring P2Y receptors, whereas UTP stimulated UTP-preferring P2Y receptors. Our results show a similar pattern of UDP and UTP responses, although increasing the stability of UDP with hexokinase treatment reduced its effect, and the stable UDP analog UDPβS was shown to be ineffective (Table 1 and Fig. 3). We therefore concluded that P2Y₆ receptors do not mediate contraction of human coronary arteries. This lack of contractile P2Y₆ receptors could be organ-specific because UDPβS is a potent vasoconstrictor in the rat mesenteric artery (Malmsjö et al., unpublished data). Furthermore, the results demonstrate the importance of stable nucleotides in the characterization of receptors in tissue preparations with ectonucleotidase activity. The UTP response was further studied by use of the stable analog UTPγS, which showed to stimulate the most potent and efficacious P2 receptor-mediated constriction of human coronary arteries (Table 1 and Fig.2). The similarity between this response and that of ATPγS (after P2X receptor desensitization) suggests activation of P2Y₂ or P2Y₄ receptors. A greater potency of UTPγS (pEC₅₀ = 6.6) compared with ATPγS (5.8) on the human recombinant P2Y₂receptor was observed by Lazarowski et al. (1995, 1996). A similar difference in potency by the two P2Y receptor agonists was found in our experiments (pEC₅₀ = 4.6 for UTPγS and 3.8 for ATPγS), making activation of P2Y₂ receptors likely. Ap₄A also elicited vascular contraction, which suggests activation of P2Y₂ receptors (Lazarowski et al., 1995), although this effect could also be mediated by the suggested but so far uncloned P2Y_Ap4Areceptor (Pintor et al., 1993; Table 1 and Fig. 2).

The P2Y₁ receptor does not seem to mediate contraction of human coronary arteries, as the selective P2Y₁ agonist ADPβS had no effect (Table 1 and Fig. 2). This is in agreement with previous suggestions that the P2Y₁ receptor is located only on endothelial cells in the blood vessel (Korchazhkina et al., 1999) and that desensitization with αβ-MeATP has been shown to abolish the contractile responses to 2-MeSATP in the isolated rabbit coronary artery (Corr and Burnstock, 1994).

Only a few studies have examined the presence of different P2 receptor mRNAs in vascular smooth muscle cells (Malam-Souley et al., 1993;Valera et al., 1994; Chang et al., 1995; Erlinge et al., 1998). In the present study, RT-PCR analysis of human coronary arteries generated strong bands for P2X₁ and P2Y₂ receptor mRNA, although bands for P2Y₁, P2Y₄, and P2Y₆ receptors could also be detected. RT-PCR detection is not a quantitative method, it is only evidence of mRNA expression for the receptor studied and may not correspond with receptor-protein expression on the surface of the cell. Therefore, molecular and pharmacological data do not always correlate. Despite this, the strong band for the P2Y₂ receptor mRNA and the weak band for the P2Y₄ receptor, together with the pharmacological data, suggest an important role for P2Y₂ receptors in nucleotide-stimulated contractions of human coronary arteries.

The contractile response to αβ-MeATP is most likely mediated by the P2X₁ receptor subtype because electrophysiological responses with structure-activity relationships similar to those of the cloned P2X₁ receptor have been seen in isolated vascular smooth muscle cells (Evans and Kennedy, 1994). However, because αβ-MeATP also activates and rapidly desensitizes P2X₃ receptors, these receptors might also contribute, although there is no evidence of vascular smooth muscle cell expression (Ralevic and Burnstock, 1998).

An interesting conclusion based on the difference in potency between the endogenous (UTP and ATP) and the stable (UTPγS and ATPγS) nucleotides is the importance of ectonucleotidases for nucleotide degradation, as suggested by Lazarowski et al., 1997. In cell systems where the influence of ectonucleotidases has been minimized, UTP and UTPγS are equally potent (Lazarowski et al., 1996). In this preparation of vascular tissue, UTP and ATP were approximately 2 log units less potent than UTPγS and ATPγS at inducing contraction, indicating that more than 99% might have been hydrolyzed or degraded by ectonucleotidases. Because the endothelium was removed, it seems likely that the ectonucleotidases are located on the smooth muscle cells. The potency of the UTPγS-elicited contraction is approximately 2 log units lower than that for inositol phosphate formation in astrocytoma cells with the human P2Y₂receptor stably expressed (pEC₅₀ = 4.5 and 6.6, respectively; Lazarowski et al., 1996). The reason for this discrepancy might be that the receptor density is lower in the vascular smooth muscle cells in our preparation or that UTPγS is degraded by ectonucleotidases, although to a much lesser degree than UTP.

Levels of circulating extracellular nucleotides are low under normal conditions but can increase on stimulation of vascular cells by physiological agonists or under pathological situations. Increased ATP release has been observed in isolated perfused hearts during increased flow or hypoxic conditions, as well as after injury induced by ischemia and reperfusion (Clemens and Forrester, 1981; Forrester, 1990; Yang et al., 1993; Vials and Burnstock, 1996). ATP and UTP are constituents of platelets and may be released on stimulation (Gordon, 1986). Aggregating human platelets relax canine coronary arteries with intact endothelium, whereas in the absence of a functional endothelium, platelets cause a marked contraction of blood vessels (Houston et al., 1986). Adhesion of platelets to the subendothelial layers in coronary arteries with atherosclerosis has been thought of as an initial event in the genesis of occlusive thromboses leading to acute myocardial infarction (Hirsh et al., 1981; Conti and Mehta, 1987). Similar platelet deposition occurs at the site of coronary angioplasty, and the release of nucleotides induces vasospasm that may be the basis of acute occlusion of the affected artery. Extracellular nucleotides are also potent mitogens and may stimulate neointima formation by activating P2Y receptors on vascular smooth muscle cells (Erlinge, 1998). The net effect of extracellular nucleotides on isolated vessels or on vascular beds will be the result of actions mediated by P2 receptors on the endothelium and smooth muscle cells, respectively. Changes in vascular tone and in the integrity of nerves and endothelial cells may alter the balance of the response and may have important implications for the involvement of P2 receptors in, for example, coronary vasospasm (Ralevic and Burnstock, 1991). The present P2 receptor characterization of human coronary arteries might therefore be of importance in the design of future cardiovascular therapeutic agents.

Conclusions.

The stable pyrimidines UTPγS and UDPβS are useful tools in the pharmacological P2 receptor characterization in intact tissues with ectonucleotidase activity. Extracellular nucleotides stimulated contractions of human coronary arteries primarily by activation of P2Y₂ and P2X receptors, whereas a role for P2Y₁ and P2Y₆ receptors can be excluded. In agreement with this, the RT-PCR bands for P2Y₂ and P2X₁ receptor mRNAs were prominent. These results indicate that antagonists of P2Y₂ and P2X receptors may be useful in the treatment of coronary vasospastic disorders such as angina pectoris.