Introduction

Top Abstract Introduction Methods Results Discussion References

The A₁ and A₂ subtypes (A_2A and A_2B) of cell membrane AdoRs mediate the known cardiovascular effects of adenosine. Responses to activation of A₁AdoRs include (1) slowing of heart rate (negative chronotropy) and impulse propagation through the atrioventricular node (negative dromotropy), (2) reduction in atrial contractility (negative inotropy) and (3) inhibition of the stimulatory effects of catecholamines (anti-beta-adrenergic action) (Belardinelli et al., 1989; Olsson and Pearson, 1990). Activation of A₂AdoRs causes coronary dilation and increases coronary conductance (Olsson and Pearson, 1990; Mustafa et al., 1995).

The subtype of A₂AdoR (A_2A and A_2B) that mediates vasodilation in response to adenosine is uncertain (Mustafa et al., 1995). The nanomolar potency of the selective A_2AAdoR agonist CGS21680 [2-p-(2-carboxyethyl)phenethylamino-5-N-ethylcarboxamidoadenosine] to cause coronary vasodilation, and the similar rank orders of potency for adenosine analogs to bind to A_2AAdoRs in brain membranes and cause coronary vasodilation (Olsson and Pearson, 1990), suggest that A_2AAdoRs are mediators of coronary vasodilation. However, specific binding of [³H]CGS21680 to coronary artery membrane preparations could not be demonstrated (Mustafa et al., 1995), and thus the evidence for coronary A_2AAdoRs remains inconclusive.

The absence of radioligand binding data to confirm the functional pharmacological characterization of coronary adenosine receptors was recently overcome by the use of a new selective A_2AAdoR antagonist radioligand, [³H]SCH58261 (Belardinelli et al., 1996). SCH58261 [5-amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine] and its tritiated analog have been used in our laboratories to investigate A_2AAdoRs in several tissues (Belardinelli et al., 1996; Dionisotti et al., 1996; Zocchi et al., 1996a, 1996b). Specific high-affinity binding sites for [³H]SCH58261 were found to be present in coronary artery membranes in large numbers (900 fmol/mg of protein) (Belardinelli et al., 1996).

The present study uses the A_2AAdoR antagonist SCH58261 to test the hypothesis that the A_2AAdoR mediates coronary vasodilation caused by either exogenous or endogenous adenosine in the guinea pig. The interpretation of the results depends on the assumption that SCH58261 is a highly selective A_2AAdoR antagonist. Accordingly, we first present the results of pharmacological experiments that validate the use of SCH58261 as a selective A_2AAdoR antagonist in the guinea pig heart. Next, we use SCH58261 and CPX to distinguish the A₁- and A_2A-AdoR-mediated responses of the heart to CCPA and to CGS21680, which are relatively selective A₁AdoR and A_2AAdoR agonists, respectively. Last, we use SCH58261 to assess the role of the A_2AAdoR in mediating the coronary vasodilation caused by endogenous adenosine.

	Methods

Top Abstract Introduction Methods Results Discussion References

Chemicals. SCH58261 was synthesized by Prof. Dr. P.G. Baraldi according to methods described previously (Baraldi et al., 1994). The nucleoside uptake inhibitor draflazine (Van Belle and Janssen, 1991) was a gift from Dr. H. Van Belle of Janssen Pharmaceutics (Beerse, Belgium). The AdoR antagonist CPX, the AdoR agonists CGS21680, (R)-PIA) and CCPA, and the adenosine kinase inhibitor iodotubercidin were purchased from Research Biochemicals (Natick, MA). Adenosine, ADA type VII, DMSO and PEG (molecular weight, 400) were purchased from Sigma Chemical (St. Louis, MO). Stock solutions of drugs were prepared in DMSO or PEG, dissolved in perfusion medium and infused to achieve the desired perfusate concentration. The final concentrations of DMSO and PEG were determined, in separate studies, to have no effect on coronary perfusion pressure or stimulus-to-His bundle (S-H) interval.

Isolated perfused hearts. Adult Hartley guinea pigs of either sex weighing between 250 and 350 g were anesthetized with methoxyflurane and killed by cervical dislocation or decapitation. The hearts were quickly removed, rinsed in ice-cold modified K-H solution and perfused retrogradely through the aorta with modified K-H solution at a rate of 10 ml/min. The modified K-H solution had the following composition (in mM): NaCl 117.9, KCl 4.5, CaCl₂ 2.5, MgSO₄ 1.18, KH₂PO₄ 1.18, pyruvate 2.0, glucose 5.5, Na₂EDTA 0.57, ascorbic acid 0.007 and NaHCO₃ 25.0. The pH of the solution was adjusted to pH 7.4 with sodium hydroxide, and the solution was continuously gassed with 95% oxygen/5% CO₂ and warmed to 35.0 ± 0.5°C. The hearts were electrically paced at a fixed cycle length of 290 to 300 msec (unless otherwise noted) by a bipolar electrode placed on either the left atrium or right ventricle. In experiments in which AV nodal conduction time was measured, parts of the right atrium, including the sinoatrial nodal region, were removed to facilitate the placement of a unipolar electrode in the region of the AV node for recording of the HBE and facilitate atrial pacing of the heart. The HBE was recorded and analyzed as described previously (Jenkins and Belardinelli, 1988). Prolongation of the S-H interval was used as a measure of the negative dromotropic effect of adenosine on AV nodal conduction. To measure the coronary perfusion pressure, a pressure transducer was connected to the aortic cannula via a T-connector positioned ~2 cm above the aorta. Coronary perfusion pressure (in mm Hg) was monitored throughout an experiment and recorded on a strip-chart recorder and/or with a personal computer. Coronary conductance (in ml/min/mm Hg) was calculated as the ratio between coronary perfusion rate (10 ml/min) and coronary perfusion pressure (in mm Hg). After dissection and instrumentation were completed, the hearts were allowed to equilibrate for at least 30 min before experiments were begun. Experimental interventions were preceded and followed by control periods, and if preintervention and postintervention values of measured parameters differed by >15%, the data were discarded. Unless otherwise noted, measurements of steady state responses are shown.

Antagonism by SCH58261 of the Increases in Coronary Conductance Caused by Adenosine and CGS21680

Antagonism of adenosine. This series of eight experiments determined the concentration-response relationship for SCH58261 (2-300 nM) to attenuate the increase of coronary conductance caused by 50 nM adenosine. This concentration of adenosine caused near-maximal vasodilation.

Antagonism of CGS21680. A series of seven experiments assessed the attenuations by 10, 30 and 100 nM SCH58261 of the increase in coronary conductance caused by CGS21680 (0.01 nM to 100 µM). The steady state decrease in coronary perfusion pressure (increase in conductance) caused by each concentration of CGS21680 was recorded. After the CGS21680 concentration-response curve was completed, hearts were perfused with drug-free K-H solution until the perfusion pressure returned to base line. Hearts were then perfused for 10 to 15 min with K-H solution containing 10 nM SCH58261. The CGS21680 concentration-response curve was repeated in the continued presence of 10 nM SCH58261 and then in the presence of 30 and 100 nM SCH58261 in a similar manner.

Specificity of Action of SCH58261

The effects of 60 nM SCH58261 to attenuate the increases of coronary conductance caused by SNP (5 nM to 20 µM) and diltiazem (1 nM to 50 µM) were assessed. The steady state increase in coronary conductance caused by each concentration of SNP or diltiazem was determined. After control responses to increasing concentrations of either SNP or diltiazem were recorded from a heart, the heart was perfused with drug-free K-H solution until coronary conductance returned to base line. Hearts were then perfused for 15 min with K-H solution containing 60 nM SCH58261. The SNP or diltiazem concentration-response curve was repeated in the continued presence of 60 nM SCH58261. Three experiments each were performed to determine the effect of SCH58261 on increases in coronary conductance caused by either SNP or diltiazem.

To further assess the specificity of SCH58261, the effects of 60 nM SCH58261 to attenuate the increases of coronary conductance caused by 1 nM CGS21680, 70 nM CCPA, 60 nM (R)-PIA, 1.2 µM SNP and 400 nM diltiazem were determined. Vasodilators were used at concentrations that caused equivalent increases of coronary conductance of ~0.3 to 0.35 ml/min/mm Hg (see fig. 3). Each heart (n = 28) was exposed to at least three of the five vasodilators tested. The vasodilators were infused in random order, and a 10- to 15-min washout period followed each intervention. An infusion of 60 nM SCH58261 was begun, and 15 to 20 min later, the interventions were repeated, again in random order, as the infusion of SCH58261 continued. The corresponding coronary perfusion pressure in response to each vasodilator was recorded.

Selective Antagonism by SCH58261 of A₂AdoR-Mediated Coronary Vasodilation and by CPX of A₁AdoR-Mediated S-H Interval Prolongation

Hearts were instrumented for continuous recording of S-H interval (A₁ response) and coronary perfusion pressure (A₂ response). The actions of SCH58261 (eight experiments) and CPX (six experiments) to attenuate the A₁AdoR-mediated S-H interval prolongation and the A₂AdoR-mediated coronary vasodilation caused by adenosine were determined. In each experiment, both the S-H interval prolongation and the increase in coronary conductance caused by a 100-µl bolus of 200 µM adenosine injected into the perfusion line (at least twice to test for reproducibility) were recorded simultaneously (see figs. 4 and 5). After washout of adenosine for a period of 5 to 10 min, SCH58261 or CPX were infused for 15 to 20 min to achieve a perfusate concentration of 100 nM, and the bolus injections of adenosine were repeated. Infusion of antagonist was then discontinued, and after a washout period of 20 to 30 min, the bolus injections of adenosine were again administered and responses were recorded.

Attenuation by SCH58261 and CPX of the effects of CCPA. The actions of SCH58261 (six experiments) and CPX (four experiments) to attenuate the S-H interval prolongation and the coronary vasodilation caused by CCPA were determined (see fig. 9). Concentration-response relationships for CCPA-induced prolongation of S-H interval and increase of coronary conductance were determined in the absence and presence of either SCH58261 (60 nM) or CPX (20 nM). The protocol was similar to that described above for antagonism by SCH58261 of the coronary vasodilatory effect of CGS21680. In brief, after control measurements of S-H interval and coronary perfusion pressure were recorded, progressively higher concentrations of CCPA were administered until maximal coronary vasodilation was achieved, and the responses were recorded. CCPA was then washed out for at least 30 min to allow the S-H interval and coronary perfusion pressure to return to base line. Hearts were then perfused for at least 15 min with K-H solution containing either 60 nM SCH58261 or 20 nM CPX, and the concentration-response relationships for CCPA were determined in the continued presence of antagonist. Last, CCPA and the antagonists SCH58261 and CPX were washed out for 30 min or until the S-H interval and coronary perfusion pressure had returned to base line. Hearts were then perfused with K-H solution containing progressively higher concentrations of CCPA until maximal S-H interval prolongation and coronary vasodilation were achieved, and the responses were measured. Because CCPA induces second-degree AV block at concentrations several-fold lower than those needed to cause maximal coronary dilation (table 1), hearts were paced using an electrode placed on the right ventricle.

Attenuation by SCH58261 and CPX of the effects of CGS21680. The actions of 60 nM SCH58261 and 100 nM CPX to attenuate the increase of coronary conductance caused by 1 nM CGS21680 were determined in four experiments. When a steady state response to CGS21680 was achieved, the effects of SCH58261 and CPX in the presence of CGS21680 were determined in alternating order. Washout periods of 15 to 20 min separated successive interventions.

Antagonism by SCH58261 of Coronary Vasodilator Effects of Draflazine and Iodotubercidin

These experiments (n = 13) assessed whether SCH58261 attenuates increases of coronary conductance caused by the nucleoside uptake inhibitor draflazine (six experiments) and by the adenosine kinase inhibitor iodotubercidin (seven experiments). Preliminary experiments showed that draflazine (30 nM) and iodotubercidin (20 nM) caused significant increases of coronary conductance that were stable for at least 30 min, without causing prolongation of the S-H interval. After a control coronary perfusion pressure was recorded, either draflazine or iodotubercidin was infused into the perfusion line for the remainder of the experiment. After coronary conductance had reached a new steady state, an infusion of SCH58261 (60 nM) was begun and continued for at least 15 min. Infusion of SCH58261 was then stopped, whereas the infusions of either draflazine or iodotubercidin were continued. In two hearts of each group, the effect of ADA (7 U/ml) on coronary perfusion pressure was determined in the continued presence of either draflazine or iodotubercidin. In a separate series of hearts (n = 6), using the same experimental protocol described above, the effect of the A₁AdoR antagonist CPX (20 nM) on the increases of coronary conductance caused by either 30 nM draflazine (n = 3) or 20 nM iodotubercidin (n = 3) was determined. Infusion of CPX was then discontinued, whereas infusion of SCH58261 (60 nM) was begun in the continued presence of either draflazine or iodotubercidin, and the steady state changes of coronary conductances were recorded.

Data Analysis

All values are reported as mean ± standard error. Significance of differences among values in experiments with multiple treatment groups was determined by analysis of variance followed by Student-Newman-Keuls testing (unequal group sizes are allowed). A value of P < .05 was considered to indicate a statistically significant difference.

The potency of SCH58261 to antagonize a CGS21680-induced coronary vasodilation was estimated using Schild analysis (Arunlakshana and Schild, 1959). Dose-response ratios using EC₅₀ values for CGS21680 in the absence and presence of three different concentrations of SCH58261 were calculated and used to construct a Schild plot from which the K_B and pA₂ (log₁₀K_B) values were determined. To determine the IC₅₀ (concentration of antagonist that causes 50% inhibition) value for inhibition by SCH58261 of an adenosine-induced increase in coronary conductance, concentration-response data shown in figure 1 were fitted to the following logistic equation: y = d + (a  d)/(1 + [x/c]^b), where y, x, a, b, c and d denote coronary conductance, concentration of antagonist, extrapolated maximal value of coronary conductance, apparent Hill coefficient, concentration of antagonist causing half-maximal attenuation of the effect of adenosine and extrapolated minimum value of coronary conductance, respectively. The EC₅₀ values for CCPA- and CGS21680-induced increases of coronary conductance and AV nodal conduction time (S-H interval) were determined by fitting of the experimental data with a multiparameter polynomial equation using a nonlinear regression algorithm (Marquardt-Leuvenberg). The increases in coronary conductance caused by bolus injections of adenosine (see figs. 4 and 5) were quantified by numerically integrating the AUC of the coronary conductance-time relationship over the period in which coronary conductance was above base line with the use of Table Curve version 2 (Jandel).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Concentration-response relationship for inhibition by SCH58261 of the vasodilatory effect (increase of coronary conductance) of 50 nM adenosine in guinea pig isolated hearts. Points indicate the mean ± S.E.M. of single measurements in each of eight hearts.

	Results

Top Abstract Introduction Methods Results Discussion References

SCH58261 attenuated the increases of coronary conductance caused by adenosine and by the selective A_2AAdoR agonist CGS21680. A near-maximal increase of coronary conductance was elicited by 50 nM adenosine (not shown). SCH58261 dose-dependently attenuated the adenosine-induced increase of conductance with an IC₅₀ value of 6.8 ± 0.6 nM (fig. 1). The effect of 50 nM adenosine was fully attenuated by 100 nM SCH58261 (fig. 1). To study the nature of the antagonism by SCH58261, increases of coronary conductance caused by CGS21680 in the absence and presence of SCH58261 were measured. In the presence of either 10, 30, or 100 nmol/L SCH58261, the concentration-response relationship for an increase of coronary conductance by CGS21680 was shifted to the right (fig. 2A). The slope of the Schild plot of these data was 1.00 (fig. 2B), and pA₂ and K_B values for SCH58261 were 8.30 and 5.01 nM, respectively (fig. 2B).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   A, Concentration-response relationships for CGS21680 to increase coronary conductance of isolated guinea pig hearts in the absence and presence of SCH58261 (3, 10 or 100 nM). Points indicate the mean of four to seven independent determinations. Hearts were paced at a cycle length of 240 msec. Coronary conductance in the absence of drug was 0.18 ± 0.01 ml/min/mm Hg (mean ± S.E.M., n = 7). B, Schild plot of results shown in A. Values in parentheses indicate 95% confidence limits.

Specificity of action of SCH58261. To determine the specificity of action of SCH58261 to antagonize coronary vasodilation mediated by AdoR agonists, we tested whether SCH58261 could attenuate the increases of coronary conductance caused by SNP, diltiazem, and by the AdoR agonists CGS21680, CCPA and (R)-PIA. Both SNP and diltiazem caused concentration-dependent increases in coronary conductance. SCH58261 (60 nM) did not attenuate the increases of coronary conductance caused by either SNP or diltiazem. The EC₅₀ values for the increases in coronary conductance caused by SNP in the absence and presence of 60 nM SCH58261 were 170 ± 30 and 199 ± 60 nM (P  0.05), respectively; the EC₅₀ values for diltiazem in the absence and presence of 60 nmol/L SCH58261 were 889 ± 83 and 757 ± 141 nM (P  .05), respectively. As illustrated in figure 3, although 60 nM SCH58261 did not attenuate the increases of coronary conductance caused by either SNP or diltiazem, it antagonized those caused by AdoR agonists.

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Specificity of action of SCH58261 to antagonize the vasodilatory effect (increase of coronary conductance) of AdoR agonists in guinea pig hearts. Top, analog records of the increases in coronary conductance caused by the AdoR agonist (R)-PIA (left) and by the calcium channel blocker diltiazem (right). SCH58261 (60 nM) completely reversed the increase in coronary conductance caused by (R)-PIA but did not attenuate the effect of diltiazem. Bottom, summary of data demonstrating the effect of SCH58261 (60 nM) on increases in coronary conductance caused by CGS21680 (CGS, 1 nM, n = 4), CCPA (70 nM, n = 5), (R)-PIA (60 nM), n = 9), SNP (1200 nM, n = 6) and diltiazem (400 nM, n = 4). Note that only the increases in coronary conductance caused by AdoR agonists were attenuated by SCH58261. Each bar represents the mean ± S.E.M. of four to nine independent determinations. *, Significantly different from agonist alone, P < .05.

SCH58261 and CPX distinguish actions of adoR agonists mediated by A₂AdoRs and A₁AdoRs in the isolated guinea pig heart. SCH58261 attenuated the increase of coronary conductance but not the prolongation of S-H interval caused by adenosine (fig. 4, A and B). A bolus infusion of 200 µM adenosine caused transient and submaximal increases in both coronary conductance and AV nodal conduction time (S-H interval), but only the increase of coronary conductance was attenuated by 100 nM SCH58261 (fig. 4, A and B).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Summary of data demonstrating selective antagonism by 100 nM SCH58261 of the A2AdoR-mediated increase in coronary conductance in guinea pig hearts (A and B). Antagonism by 100 nM CPX of the A1AdoR-mediated negative dromotropic effect (S-H interval prolongation) of adenosine (ADO) on guinea pig hearts is shown for comparison (A and B). Top (A2 response), effects of adenosine (ADO, bolus injection of 100 µl of a 200-µM stock solution) on coronary conductance in the absence and presence of either 100 nM SCH58261 (A) or 100 nM CPX (A). AUC of each digitized recording of an ADO-induced increase of coronary conductance was determined with a computer. Bottom (A1 response), effects of ADO (bolus injections, as in A) on S-H interval in the absence and presence of either 100 nM SCH58261 (B) or 100 nM CPX (B). Measurements of coronary conductance and S-H interval were made simultaneously. Bars indicate the mean ± S.E.M. of determinations from five of six guinea pig hearts. *, Value significantly different from ADO alone (P < .05).

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Attenuation by 60 nM SCH58261, but not by 100 nM CPX, of the action of CGS21680 to increase coronary conductance. A, Increase in coronary conductance caused by 1 nM CGS21680 was fully attenuated by 60 nM SCH58261. B, Increase in coronary conductance caused by 1 nM CGS21680 was not attenuated by PEG, the vehicle for SCH58261 or 100 nM CPX but was reversible on washout. At the end of the experiment, coronary conductance was at the control level, evidence that the preparation remained stable. C, Summary of data. Each heart was exposed to both SCH58261 and CPX; exposure to CPX preceded exposure to SCH58261 in two of four experiments. Bars indicate mean ± S.E.M. of single determinations in each of four hearts.

View larger version (10K): [in this window] [in a new window]   Fig. 6.   Chart recording of coronary conductance to show attenuation by 60 nM SCH58261 of the action of CCPA. The recording is from the same experiment as that shown in figure 8. The A2AAdoR antagonist fully reversed the increase of coronary conductance caused by 100 nM CCPA. At the end of the experiment, coronary conductance was at the control level, evidence that the preparation remained stable.

View larger version (31K): [in this window] [in a new window]   Fig. 7.   Contrasting actions of SCH58261 (left) and CPX (right) on the effects of CCPA to prolong S-H interval (A1 response) and increase coronary conductance (A2 response) in guinea pig hearts. Shown are concentration-response curves for the negative dromotropic (S-H interval prolongation) and vasodilator (increase of coronary conductance) actions of CCPA in the absence (control) and presence of the A2AAdoR antagonist SCH58261 (60 nM) or the A1AdoR antagonist CPX (20 nM). Symbols and error bars indicate mean ± S.E.M. of single determinations from each of five (SCH58261) or four (CPX) hearts.

Coronary vasodilation caused by inhibitors of nucleoside transport and adenosine kinase. Draflazine, an adenosine uptake blocker, and iodotubercidin, an adenosine kinase inhibitor, significantly increased coronary conductance. As illustrated in figure 8, 30 nM draflazine and 20 nM iodotubercidin caused a 1.3- and 1.7-fold increase in coronary conductance, respectively, from control. The increases in coronary conductance caused by draflazine or iodotubercidin were completely reversed by 60 nM SCH58261 (fig. 8). The effect of SCH58261 was reversible on washout (fig. 8). ADA (7 U/ml) also completely attenuated the increases of coronary conductance caused by either 30 nM draflazine or 20 nM iodotubercidin (fig. 8). Not shown, the A₁AdoR antagonist CPX (20 nM) did not attenuate the increases in coronary conductance caused by either draflazine or iodotubercidin. In three experiments, 30 nM draflazine increased coronary conductance from a base line of 0.177 ± 0.004 to 0.23 ± 0.01 ml/min/mm Hg. The increases in coronary conductance caused by 30 nM draflazine in the absence and presence of 20 nM CPX (a concentration at which this antagonist is selective for A₁AdoRs) were not significantly different: 0.055 ± 0.005 and 0.054 ± 0.005 ml/min/mm Hg, respectively. In contrast, in the same hearts, 60 nM SCH58261 reversed the increase in coronary conductance caused by draflazine from 0.23 ± 0.01 to 0.19 ± 0.01 ml/min/mm Hg. In an additional three experiments, 20 nM iodotubercidin increased coronary conductance from 0.19 ± 0.01 to 0.32 ± 0.01 ml/min/mm Hg. The increases in coronary conductance caused by 20 nM iodotubercidin in the absence and presence of 20 nM CPX were not significantly different: 0.13 ± 0.01 and 0.13 ± 0.01 ml/min/mm Hg, respectively. In contrast, 60 nM SCH58261 reduced coronary conductance in the presence of iodotubercidin from 0.32 ± 0.01 to 0.21 ± 0.01 ml/min/mm Hg, a value not significantly different from the base-line coronary conductance of 0.19 ± 0.01 ml/min/mm Hg.

View larger version (19K): [in this window] [in a new window]   Fig. 8.   Summary of data demonstrating that both 60 nM SCH58261 and ADA (7 U/ml) reversed the increases in coronary conductance caused by 30 nM draflazine (top) or 20 nM iodotubercidin (ITU; bottom) in guinea pig hearts. Neither draflazine nor iodotubercidin prolonged the S-H interval (not shown). Bars represent mean ± S.E.M. of data collected from six or seven hearts, except for ADA, for which each bar represents data from two hearts. *Significantly different from control (P < .05). Significantly different from draflazine or iodotubercidin alone.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The results of this study provide strong evidence that in the isolated, perfused guinea pig heart, the A_2AAdoR mediates the coronary vasodilations caused by adenosine, CGS21680, CCPA and (R)-PIA. Similarly, the A_2AAdoR mediates the coronary vasodilation caused by agents that increase the interstitial concentration of adenosine by inhibiting either cellular uptake (e.g., draflazine) or cellular metabolism (e.g., iodotubercidin) of adenosine. These conclusions rely on evidence presented here and elsewhere (Belardinelli et al., 1996; Dionisotti et al., 1996; Zocchi et al., 1996a, 1996b) that SCH58261 is a competitive, specific and selective antagonist of the A_2AAdoR when used at concentrations 100 nM.

A_2AAdoR Antagonism by SCH58261

Potency of SCH58261. SCH58261 antagonized the CGS21680-induced increase of coronary conductance of guinea pig hearts in a concentration-dependent and competitive manner (fig. 2). The pA₂ value of 8.30 for SCH58261 to attenuate CGS21680-induced vasodilation is similar to the pK_i value of 8.59 ± 0.17 for this antagonist to compete with [³H]SCH58261 for binding to pig coronary artery membranes (Belardinelli et al., 1996). The IC₅₀ value for SCH58261 to attenuate an increase in coronary conductance caused by adenosine was 6.8 ± 0.6 nM (fig. 1). Thus, our results and those of Zocchi et al. (1996b) indicate that in functional assays, the potency of SCH58261 is in the low nanomolar range.

Specificity of SCH58261. SCH58261 (60 nM) attenuated the increases of coronary conductance caused by the AdoR agonists CGS21680, CCPA and (R)-PIA but not those caused by either SNP or diltiazem (fig. 3). Further evidence for the specificity of action of SCH58261 is found in the recent report by Zocchi et al. 1996b; these authors noted that SCH58261 (10 µM) had no effect on the activity of phosphodiesterase type III in homogenates of bovine heart and that no measurable binding of SCH58261 to preparations containing alpha-1, alpha-2 and beta-1 adrenergic receptors; D₁ and D₂ dopamine receptors; 5-hydroxytryptamine₁ and 5-hydroxytryptamin₂ receptors; M₁ and M₂ muscarinic receptors; µ-opioid receptors; benzodiazepine receptors and N-methyl-D-aspartate receptors could be demonstrated by use of radioligand binding assays.

Selectivity of SCH58261. The selectivity of SCH58261 to antagonize A_2AAdoR- but not A₁AdoR-mediated responses was also demonstrated in the present study. SCH58261 at concentrations of 60 or 100 nM did not attenuate the A₁AdoR-mediated negative dromotropic effects of either adenosine or CCPA (figs. 4 and 7). This finding indicates that SCH58261 has a low affinity for the A₁ subtype of adenosine receptor in the heart. In contrast to SCH58261, the prototypical A₁AdoR antagonist CPX (100 nM) did not attenuate the coronary vasodilations caused by adenosine (fig. 4) or by the A_2AAdoR agonist CGS21680 (fig. 5) but antagonized the S-H interval prolongation caused by either adenosine (fig. 4) or CCPA (fig. 7). These complementary data for CPX and SCH58261 indicate that the lack of an effect of SCH58261 on the S-H interval is not due to an inability of the heart to respond to an A₁AdoR antagonist.

Selectivities of CGS21680 and CCPA. The findings that 100 nM SCH58261 and 100 nM CPX selectively antagonized the negative dromotropic and coronary vasodilator effects, respectively, of adenosine, CGS21680 and CCPA also suggest that all three AdoR agonists activate both A₁ and A_2AAdoRs in the guinea pig heart. Thus, as shown in table 1, the EC₅₀ value for CCPA to increase coronary conductance (an A_2AAdoR action) was only 14.7-fold higher than the EC₅₀ value for CCPA to prolong the S-H interval (an A₁AdoR action). In contrast, the ratio of EC₅₀ values for CGS21680 to prolong the S-H interval and to increase coronary conductance was 10,927. The selectivity ratio (EC₅₀ A₂ response/EC₅₀ A₁ response) of >10,000 for CGS21680 that was calculated from assay of guinea pig heart functional responses is similar to selectivity ratios reported by Ueeda et al. (1991) and Poucher et al. (1995) for CGS21680-induced functional responses of guinea pig hearts. On the other hand, the 14.7-fold selectivity of CCPA for A₁ vs. A_2AAdoRs in our experiments is 10- to 100-fold lower than has been reported in previous studies of CCPA using both radioligand binding (Lohse et al., 1988; Klotz et al., 1989; Conti et al., 1993) and functional (Lohse et al., 1988; Conti et al., 1993) assays. A possible explanation of our results is that A_2AAdoRs may be more numerous than are A₁AdoRs in the target cells in heart tissues (Kollias-Baker et al., 1995; Belardinelli et al., 1996) and more efficiently coupled to the measured response (coronary vasodilation for A_2AAdoRs, increase in AV nodal conduction time for A₁AdoRs) than are A₁AdoRs. That is, receptor reserve may be greater for CGS21680 to increase coronary conductance than for CCPA to prolong the S-H interval (Srinivas et al., 1997; Shryock el al., 1998). In other tissues, receptor reserves for A_2AAdoR and A₁AdoR-mediated responses may be different from those in the heart, and as a consequence, agonist selectivity ratios for A₂ vs. A₁ responses may also be different.

Coronary Vasodilation Is Mediated by A_2AAdoRs

The selective A_2AAdoR agonist CGS21680 and the selective A_2AAdoR antagonist SCH58261 were shown in the present study to cause an increase or a decrease, respectively, in coronary conductance. These results strongly implicate A_2AAdoRs as mediators of an increase in coronary conductance. However, the A₁AdoR-selective agonists (R)-PIA and CCPA also increased coronary conductance (figs. 3, 6 and 7). SCH58261 (60 nM) reversed the vasodilations caused by both (R)PIA and CCPA (figs. 3 and 6). Unlike SCH58261, the A₁AdoR antagonist CPX (20 nM) did not attenuate the coronary dilations caused by CCPA (fig. 7) and (R)-PIA (not shown), but it antagonized the A₁AdoR-mediated S-H interval prolongation caused by CCPA (fig. 7). These findings demonstrate that the coronary dilations elicited by either CCPA or (R)-PIA are not mediated by A₁AdoRs but instead are mediated by A_2AAdoRs. The implication of these findings raises a major concern as to the interpretation often given in functional studies that the effects of CCPA and (R)-PIA are due to activation of A₁AdoRs. Although this interpretation is generally correct, it is not necessarily so because both CCPA and (R)-PIA have submicromolar affinities for A_2AAdoRs (Conti et al., 1993). Therefore, responses that are efficiently coupled to activation of A₂AdoRs, such as coronary vasodilation, can be elicited by concentrations of CCPA and/or (R)-PIA that are considerably lower than those that are needed to cause 50% occupancy of A_2AAdoRs. Consistent with this argument, 100 nM CCPA caused a significant increase in coronary conductance (fig. 7). In summary, the results of the present study, as well as those of Poucher et al. (1995) using ZM241385, revealed that A_2AAdoRs can mediate coronary vasodilation caused by adenosine and by AdoR agonists.

Vasodilator Effects of Draflazine and Iodotubercidin

The myocardial interstitial concentration of adenosine can be increased by agents that inhibit degradation of adenosine, such as adenosine uptake blockers and adenosine kinase inhibitors (Tietjan et al., 1990; Ely et al., 1992; Wang et al., 1992; Kroll et al., 1993; Kollias-Baker et al., 1994). An objective of this study was to investigate the role of A_2AAdoRs as mediators of coronary vasodilation caused by draflazine, a nucleoside uptake blocker (Van Belle and Janssen, 1991), and by iodotubercidin, an adenosine kinase inhibitor (Newby, 1981). Both draflazine and iodotubercidin have been previously shown to increase the S-H interval (Dennis et al., 1996) and coronary conductance (Kroll et al., 1993; Kollias-Baker et al., 1994), actions that are associated with a significant increase in the myocardial interstitial concentration of adenosine (Ely et al., 1992; Kollias-Baker et al., 1994; Dennis et al., 1996). Concentrations of draflazine and iodotubercidin were selected in our experiments to cause coronary vasodilation without significant prolongation of the S-H interval. The finding that SCH58261 completely reversed the vasodilator effects of draflazine and iodotubercidin (fig. 8) suggests that the vasodilation was mediated by A_2AAdoRs. Because direct activation of A_2AAdoR by draflazine and iodotubercidin is unlikely, the increase in coronary conductance caused by these agents was probably due to an increase in the interstitial concentration of adenosine. In keeping with this interpretation, the complete reversal by adenosine deaminase of coronary vasodilations caused by draflazine and iodotubercidin (fig. 8) indicated that the effects of draflazine and iodotubercidin were mediated by endogenous adenosine accumulated as a consequence of inhibitions of adenosine uptake and adenosine kinase, respectively, and by activation of A_2AAdoRs.

In summary, the results here reported provide compelling evidence that activation of A_2AAdoRs of the coronary artery contributes in a major way to the coronary vasodilations elicited by adenosine and adenosine receptor agonists. The results also demonstrate that SCH58261 is a suitable antagonist to study A_2AAdoR-mediated coronary vasodilation.