OUP user menu

Adenosine A2 receptor function in rat ventricular myocytes

James G Dobson , Richard A Fenton
DOI: http://dx.doi.org/10.1016/S0008-6363(97)00023-0 337-347 First published online: 1 May 1997


Objective: This study was undertaken to investigate the functional significance of adenosine A2 receptor stimulation in a mammalian ventricular myocyte preparation. Methods: Isolated contracting rat ventricular myocytes were employed to assess the contractile, adenylyl cyclase and cyclic AMP responses to adenosine receptor stimulation. Results: In single myocytes the presence of A1 receptors was confirmed, as indicated by the A1 receptor agonist, phenylisopropyladenosine (PIA), reducing by 60 and 74% the inotropic response and activation of adenylyl cyclase, respectively, elicited by the β-adrenergic agonist, isoproterenol. An A1 receptor antagonist, dipropylcyclopentylxanthine (DPCPX), prevented the antiadrenergic action of PIA. The A2 receptor agonist, carboxyethylphenethyl-aminoethylcarboxamido-adenosine (CGS-21680; 0.01–10 μM) increased myocyte inotropy in a concentration-dependent manner, reaching a maximum of 41–45%. Ethylcarboxamidoadenosine (NECA), naphthyl-substituted aralkoxy-adenosine (SHA-082) and adenosine in the presence of DPCPX also increased myocyte inotropy, as evidenced by increases in myocyte shortening, duration of shortening, time-to-peak shortening, time-to-75% relaxation and rate of maximal shortening. The agonists, however, did not effect the maximal rate of relaxation. The A2 receptor antagonists, chlorofuranyldihydrotri-azoloquinazolinimine (CGS-15943) and chlorostyrylcaffeine (CSC), the latter selective for the A2a receptor, prevented the contractile responses elicited by the A2 agonists. Compared to the concentrations of A2 receptor agonists necessary to increase myocyte contractile variables, 3–12 times greater concentrations of the agonist were required to increase myocyte adenylyl cyclase activity and cAMP levels. Conclusion: The results suggest the presence of adenosine A2a receptors in the rat ventricular myocyte that appear to be responsible for an increase in inotropy via cAMP-dependent and -independent mechanisms. © 1997 Elsevier Science B.V.

  • Adenosine receptor
  • Adenosine receptor antagonists
  • Adenylyl cyclase
  • Adenosine
  • Isoproterenol
  • Adrenergic receptors
  • Contractility
  • c-AMP
  • Rat, ventricular myocyctes
  • CGS-21680, 2-p-(-2-carboxyethyl)phenethyl-amino-5′-N-ethylcarboxamido-adenosine
  • CGS-15943, 9-chloro-2-(2-furanyl)-5,6-dihydro-1,2,4-triazolo-(1,5-C)quinazolin-5-imine
  • CSC, 8-(3-chlorostyryl)caffeine
  • DMPX, 3,7-dimethyl-1-(2-propargyl)xanthine
  • DPCPX, 1,3-dipropyl-8-cyclopentylxanthine
  • NECA, N-ethylcarboxamidoadenosine
  • ISO, l-isoproterenol
  • SHA-082, naphthyl-substituted aralkoxyadenosine
  • PIA, phenylisopropyladenosine

Time for primary review 33 days.

1 Introduction

Adenosine has been known for years to exert profound effects in the heart. This nucleoside via adenosine A1 receptors displays antiadrenergic [1, 2]and antiarrthymogenic [3]actions. The antiadrenergic action of adenosine involves reductions in β-adrenergic catecholamine-induced increases in adenylyl cyclase activity [2, 4], cyclic AMP (cAMP) formation [2], intracellular Ca2+ transient magnitude [5], protein kinase A activation [2, 6], myocardial protein phosphorylation [7, 8], and cardiac atrial [9, 10]and ventricular [1, 2, 6]contractility.

Adenosine via A2 receptors elicits coronary vessel vasodilation [11]and increases contractile performance in neonatal avian [12]and adult rat [13]ventricular myocytes. Some reports utilizing mammalian ventricular myocyte preparations indicate that A2 receptor stimulation causes an increase in myocardial adenylyl cyclase activity and cAMP [4, 14, 15]while other reports do not [16, 17]. Despite an increase in cAMP, A2 agonists have been reported to be without effect on the mechanical performance of the mammalian cardiac myocyte [15]. Administration of adenosine at high concentrations (100–1000 μM) to ventricular muscle preparations has been reported to cause either an increase [18–20]or no change [21]in contractile force development. Adenosine has been reported to have a positive contractile effect with an EC50 of 10 nM on rat ventricular myocytes [22]. Moreover, increases in contractile force development were not associated with increases in adenylyl cyclase activities or cAMP concentrations [19]. Thus, there is considerable controversy concerning whether A2 receptor stimulation increases myocardial contractility and if cAMP is involved. Regarding contractility, it is not known how A2 receptor stimulation affects individual indices of contractile function: ventricular myocyte shortening, duration of shortening, time-to-peak shortening, time-to-75% relaxation and maximal rates of shortening and relaxation.

The present investigation was undertaken to ascertain the contractile, adenylyl cyclase and cAMP responses to adenosine A2 receptor stimulation in isolated rat ventricular myocytes. The results indicate that both the antiadrenergic effect of A1 receptor stimulation and the positive inotropic effect of A2 receptor stimulation are observed in single contracting adult ventricular myocytes. Moreover, the contractile variables constituting the inotropic response to A2 receptor stimulation are described along with increases in adenylyl cyclase activity and cAMP levels.

2 Methods

2.1 Myocyte isolation

Isolated adult rat ventricular myocytes were prepared according to methods previously described [4]with several modifications. Briefly, Sprague-Dawley rats (Charles River, Wilmington, MA or Harlan, Indianapolis, IN) were decapitated and the hearts were rapidly excised and constant pressure (70 cmH2O and non-recirculated) perfused for 10 min through the aortas with filtered (0.45 μm membrane filter) perfusing solution (PS in mM: 118 NaCl, 10 glucose, 25 NaHCO3, 4.69 KCl, 1.18 MgSO4, 1.18 KH2PO4; pH 7.4, 37°C) to which 2.5 mM CaCl2 was added. After equilibration the hearts were constant-pressure-perfused with fresh PS containing no added Ca2+ until spontaneous contractions ceased (∼30 s). The hearts were then perfused for 4–10 min in a non-recirculating manner with PS containing 0.73 mg/ml collagenase, 0.16 mg/ml hyaluronidase, 1 mg/ml recrystallized bovine serum albumin (BSA) and 48.4 μM Ca2+ at a rate of 3–4 ml/min/heart. Ventricles, free of atria, were removed from the perfusion system, cut into 8 pieces and placed in a 50 ml Erlenmeyer flask with 5 ml of PS containing 0.73 mg/ml collagenase, 0.16 mg/ml hyaluronidase, 2.5 mg/ml BSA, and 50 μM Ca2+ (incubation solution). The flask was gently shaken (40 cycles/min) in a reciprocating water bath with continuous gassing (95% O2/5% CO2) for 7 min at 37°C. The incubation solution was aspirated and this shaking procedure was repeated with fresh incubation solution 3–5 times. After the final incubation period the solution was aspirated and replaced with 10 ml of fresh incubation solution and the flask was shaken rapidly (120 cycles/min) for 10 min with gassing to dissociate the myocytes. The contents of the flask were filtered through a 250 μm nylon mesh into a 50 ml polypropylene centrifuge tube to which 40 ml of PS containing 6 mg/ml BSA and 100 μM Ca2+ (wash solution) was gradually added.

The myocytes were allowed to settle for 15 min, and the upper two-thirds of the wash solution was aspirated. Upon the addition of 30–35 ml of wash solution this settling step was repeated. The wash solution was aspirated as above and the myocyte pellet was resuspended in 22 ml of minimum essential medium (MEM). A 2 ml volume of the myocyte suspension was seeded onto each of 60 mm culture dishes containing 2 ml of MEM. These myocytes were held in a 37°C incubator gassed with 5% CO2 in room air for 3–4 h before use. Because myocytes do not attach under these plating conditions, the myocytes were used for contractile experiments.

When myocytes were to be used for biochemical studies, prior to seeding each of the culture dishes was preincubated for 2 h with 1 ml of MEM containing 33 μg of laminin in a 37°C incubator gassed with 5% CO2 in room air. The laminin solution was removed prior to myocyte seeding. The dishes with myocytes were incubated for 3–4 h in the 37°C incubator. This settling and incubation procedure was performed to purify the myocytes so that >95% of the myocytes adhering to the dishes were rod-shaped. Each 60 mm culture dish contained 200–600 μg of adhering rod-shaped myocyte protein, indicating that a total of 3–6 mg of protein was obtained from a pair of hearts.

2.2 Mechanical measurements

The contractile function of individual myocytes was assessed by placing 50–100 cells in a 506 μl myocyte chamber (11×23×2 mm deep). The chamber was continuously suffused (850 μl/min) with fresh suffusion solution (SS, in mM: 136.4 NaCl, 4.7 KCl, 1.0 CaCl2, 10 hydroxyethylpiperazine-ethanesulfonic acid (HEPES), 1.0 NaHCO3, 1.2 MgSO4, 1.2 KH2PO4, 10 glucose, 0.6 ascorbate, 1.0 pyruvate) at 20°C. The chamber was mounted on an inverted microscope stage and contained platinum wire electrodes for initiating myocyte contraction at 0.5 Hz (voltage 10% above threshold for 5 ms duration).

The mechanical function of a single contracting myocyte was assessed using instrumentation described below. The image of the myocyte was projected via an inverted microscope at 300× onto a line scan camera (Fairchild, Model 1600R) containing a linear array (1×3456) of photodiodes operating at 200 Hz. When aligned with the longitudinal axis of the cell, the camera detected the movement of the 2 ends of the myocyte upon contraction (shortening and relengthening). The signals from the line scan camera were displayed on an oscilloscope (Hitachi, Model V-660) which permitted optimal positioning of the camera over the myocyte. When the myocyte was transilluminated, both ends of the cell were easily discriminated. By determining the pixels in which the appropriate transitions between light and dark occurred, measurement of myocyte length and length change with respect to time for a single contraction was achieved. The line scan camera was calibrated with a stage micrometer scaled in 10 μm divisions. The signals from the camera were directed to a Hewlett Packard computer (Model Vectra RS/20C). Custom computer programming (MCS Computer Consulting, Keene, NH) permitted determination of maximum delta length (shortening) with contraction (ΔL), duration of shortening (DS), time-to-peak shortening (TPS), time-to-75% relaxation (relengthening, TR) and maximal rates of shortening (+dL/dtmax) and relaxation (−dL/dtmax). DS was the time in milliseconds from the onset of myocyte shortening to the point of full relaxation. TPS was the time in milliseconds from the onset of myocyte shortening to the point of maximum ΔL. TR was the time in milliseconds between the onset of myocyte shortening to the point of 75% relengthening.

2.3 Crude ventricular myocyte membranes

After a 3–4 h incubation period, attached myocytes (as described above) were harvested by placing the culture dishes on ice, aspirating and discarding the medium, scraping the attached cells into 1 ml of ice cold buffer (pH 7.4) containing 10 mM HEPES and 1 mM dithiothreitol (DTT), and transferring the mixture to a 40 ml centrifuge tube. The culture dishes were rinsed twice with buffer and the rinses added to the centrifuge tube. The myocytes were centrifuged at 45 000×g for 45 min and the supernatant discarded. The pellet was resuspended in 40 mM HEPES (pH 7.4) to yield 3.5–4.0 mg protein/ml and homogenized with a small clearance Dounce tissue grinder (8 strokes). The membranes were centrifuged as described above, resuspended (3.0–4.0 mg protein/ml) in 40 mM HEPES (pH 7.4), and assayed for adenylyl cyclase activity immediately or stored at −80°C and assayed within 10 days.

2.4 Crude ventricular myocardial membranes

Isolated rat hearts were perfused with 5 ml of ice-cold saline (0.9% NaCl) to wash out the blood, minced into 2–3 mm3 cubes, and placed in 10 ml of homogenization buffer (HB) containing 10 mM HEPES (pH 7.4), 1 mM ethylenediamine-tetraacetic acid (EDTA), 1 mM DTT and 10 μg/ml soybean trypsin inhibitor. The suspension was homogenized with a PT-10 Polytron generator at a speed of 6 for two 15-s periods separated by 15 s. The homogenate material was also treated with 2 strokes of a glass/teflon motor-driven Potter Elvehjen homogenizer operated at 1/2 full speed. Upon addition of 4.7 ml of HB containing 1.25 M sucrose, the homogenate was mixed and centrifuged at 1000×g for 15 min. The supernatant was filtered through 4 layers of cheese-cloth and 14.5 ml HB was added. The mixture was centrifuged at 45 000×g for 45 min and the pellet was suspended in 3–5 ml of 40 mM HEPES buffer (pH 7.4) with the small clearance Dounce tissue grinder (6 strokes) to yield 3–5 mg protein/ml. Membranes were assayed immediately for adenylyl cyclase activity. All preparative steps for the crude membranes were performed at 0–1°C.

2.5 Adenylyl cyclase assay

The assay system for measurement of adenylyl cyclase activity minimizes the formation of endogenous adenosine and has been previously described [4]. Myocyte membranes (15–25 μg protein) were incubated in 50 μl of a buffer containing 40 mM HEPES (pH 7.4), 5 mM MgCl2, 1 mM DTT, 5.5 mM KCl, 0.1 mM 2′-deoxy-cAMP (dcAMP), 0.1 mM 2′-deoxy ATP (dATP), 20 mM phosphoenolpyruvate, 2 U pyruvate kinase, 0.25 U adenosine deaminase, 1 mM ascorbic acid, 100 mM NaCl, 0.1 mM ethyleneglycol-bis(β-aminoethylether)-N,N,N′,N′-tetraacetic acid (EGTA), 10 μM guanosine 5′-triphosphate (GTP), and ∼2×106 counts/min of [α-32P]dATP for 20 min at 30°C. The reaction was stopped by adding 40 μl of a solution containing 2% sodium dodecyl sulfate (SDS), 45 mM ATP, 1.3 mM cAMP, and [3H]dcAMP (∼4000 counts/min) and by boiling for 2 min. The formed [α-32P]dcAMP was separated from the [α-32P]dATP by sequential chromatography using columns of cation exchange resin AG 50W-X4 (200–400 mesh) and neutral alumina AG 7 (100–200 mesh) after the methods of Salomon [23]. All results were corrected for column recovery of [3H]dcAMP, which ranged between 60 and 90%. The protein levels were assessed by a bicinchoninic acid technique (BCA, Pierce) using BSA as a standard. The activity of the adenylyl cyclase is expressed as pmol [α-32P]dcAMP formed/min/mg protein.

2.6 Cyclic AMP determination

Levels of cAMP were determined in ventricular myocytes. After the 3–4 h incubation period the MEM was aspirated and replaced with 2 ml of fresh MEM. Adenosine receptor agonists, isoproterenol (ISO), or adenosine receptor antagonists were added to the medium bathing the cells at the concentrations and times indicated. Experiments were terminated by removing the medium from the dish and adding 200 μl of 1 N HCl over the myocyte surface. The dishes were then frozen in liquid N2 and stored at −70°C or held on ice momentarily until extraction was initiated.

For assay the dishes were scraped with 1 ml of distilled/deionized H2O into microcentrifuge tubes. The extracts were heated for 1 h at 57°C and sonicated for 10 min. The extracts were then centrifuged at 14 000×g for 15 min. The supernatant was removed, evaporated, reconstituted in 500 μl of 50 mM sodium acetate buffer and assayed for cAMP using an 125I-cAMP RIA kit (Amersham). The pellet was solubilized with 1 N NaOH and protein determined. The cAMP values are reported in pmol cAMP/mg protein of the extract pellet (total cell protein). This cAMP assay procedure routinely provided recovery values in excess of 90%.

2.7 Statistical methods

All data are expressed as means±one standard error of the mean (s.e.). The concentration of agonist that produced 50% of the maximum stimulatory response (EC50) was determined from non-linear regression analysis using sigmoid curve fitting (GraphPAD InPlot). If the concentration of a ligand required to produce a maximal response could not be definitely determined, an apparent EC50 was calculated. Statistical analysis was performed on actual (not normalized) data. Statistical significance was determined using one-way independent analysis of variance. A probability (P value) of less than 0.05 was accepted as indicating a statistically significant difference.

2.8 Animals

The animals in this study were maintained and used in accordance with recommendations in the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources, National Research Council (DHEW Publ. NIH #85-23, Rev. 1996) and the guidelines of the Institutional Animal Care and Use Committee of the University of Massachusetts Medical School.

2.9 Materials

Buffer salts and acids were obtained from Fisher Scientific (Medford, MA). Phosphoenolpyruvate, pyruvate kinase, adenosine deaminase, adenosine, GTP, ATP, and dATP were purchased from Boehringer Mannheim (Indianapolis, IN). ISO, DTT, HEPES, tris(hydroxymethyl)-amino-methane (Tris) base, cAMP, dcAMP, EGTA, EDTA, dimethyl sulfoxide (DMSO), and BSA were obtained from Sigma Chemical (St. Louis, MO). SDS, AG 50W-X4, and AG 7 were purchased from Bio-Rad (Richmond, CA). MEM was obtained from GIBCO Laboratories (Grand Island, NY). Laminin was supplied by Collaborative Biomedical (Bedford, MA). PIA, NECA, DPCPX, CSC, CGS-21680, CGS-15943, and DMPX were purchased from Research Biochemicals (Natick, MA). For initial studies Dr. R.A. Lovell of Ciba-Geigy (Summit, NJ) generously supplied CGS-15943. SHA-082 was synthesized by Dr. R.A. Olsson of the University of South Florida (Tampa, FL). [α-32P]dATP (800 Ci/mmol) was obtained from Amersham (Arlington Heights, IL) and [3H]dcAMP (5.2 Ci/mmol) was purchased from ICN Pharmaceuticals (Irving, CA).

Stock solutions of PIA, CSC, DPCPX, CGS-21680, CGS-15943, and SHA-082 (10 mM) were prepared in DMSO. ISO (10 mM) was prepared in the appropriate buffer containing 1 mM ascorbic acid. NECA (1 mM), DMPX (1 mM) and adenosine (10 mM) were prepared in distilled/deionized water. Stock solutions were serially diluted with the appropriate buffers to the desired concentrations.

3 Results

3.1 Phenylisopropyladenosine reduction of isoproterenol-induced contractile responses and adenylyl cyclase activity

An adenosine A1 receptor agonist elicited a reduction of β-adrenoceptor-mediated increases in ventricular myocyte contractility that was antagonized by an A1 receptor antagonist. In contracting rat ventricular myocytes, ISO at 0.2 μM increased by 63% the maximum rate of shortening, +dL/dtmax (Fig. 1). An A1 receptor agonist, PIA, at 2 μM reduced the ISO induced increase in +dL/dtmax by 60%. An A1 receptor antagonist, DPCPX, at 0.2 μM reversed the PIA reduction of the ISO induced contractile responses.

Fig. 1

Effect of ISO, PIA, and DPCPX on the maximum rate of ventricular myocyte shortening (+dL/dtmax). Values are mean±s.e. (n=6). After the control values were obtained, the three combinations, ISO (0.2 μM), ISO+PIA (2 μM), and ISO+PIA+DPCPX (0.2 μM) were added sequentially every 8–10 min and the maximum responses were recorded. Asterisk denotes a significant difference from the control value (no additions). Dagger denotes a statistically significant difference from the corresponding ISO value. Double dagger denotes a statistically significant difference from the corresponding ISO+PIA value.

In membranes from ventricular myocyte preparations similar to those used for the above contractile studies, the PIA reduction of ISO stimulated adenylyl cyclase activity was prevented by DPCPX. ISO at 0.1 μM caused increases in adenylyl cyclase activity of 88% (3.72±0.54 to 6.98±0.83 pmol/min/mg protein) in the absence and 23% in the presence of 1 μM PIA. DPCPX at 0.1 μM prevented the PIA inhibition, thereby allowing an 80% ISO induced increase in adenylyl cyclase activity. The above results taken together indicate that the ventricular myocytes possessed functional A1 receptors.

3.2 CGS-21680 elicited contractile responses

The adenosine A2 receptor agonist, CGS-21680, caused a concentration-dependent increase in ventricular myocyte contractility. CGS-21680 at 0.1 μM caused 18–21% increases in delta length (ΔL), duration of shortening (DS), time-to-peak shortening (TPS), time-to-75% relaxation (TR) and +dL/dtmax (Fig. 2). The maximal observed increases in these contractile variables ranged from 41 to 45% with 10 μM CGS-21680. The agonist did not affect the maximum rate of relaxation (−dL/dtmax). The EC50 values for CGS-21680 were 121.0±26.8, 80.0±21.8, 97.6±30.9, 84.01±29.3 and 90.9±27.9 nM for ΔL, DS, TPS, TR and +dL/dtmax, respectively. There were no statistically significant differences between the EC50 values.

Fig. 2

Dose–response effect of CGS-21680 on ventricular myocyte contractile variables. CGS-21680 was administered at increasing concentrations as indicated every 10 min and the above contractile responses obtained. The upper panel illustrates length change (DELTA LENGTH), rate of maximum shortening (+dL/dtmax) and relaxation (−dL/dtmax). The lower panel illustrates duration of shortening, time-to-peak shortening and time-to-75% relaxation. Values are the mean±s.e. for 12 experiments. Asterisks denote a statistically significant difference from the appropriate zero CGS-21680 value.

3.3 Comparison of adenosine A2 receptor and β-adrenergic induced contractile responses

A comparison of typical responses to CGS-21680 and ISO with respect to contractility of ventricular myocytes is illustrated in Fig. 3. The ΔL and +dL/dtmax were increased by both agents (compare values in Table 1 and Fig. 3). However, DS, TPS and TR were increased with CGS-21680 but decreased with ISO. CGS-21680 did not affect −dL/dtmax, but ISO increased this contractile variable.

Fig. 3

Tracings depicting the comparison of CGS-21680 and ISO elicited contractile responses in a rat ventricular myocyte. Traces are length changes (DELTA LENGTH) associated with single contractile shortenings of a representative myocyte exposed to 20 μM CGS-21680 for 5 min, washed for 15 min in suffusion solution containing no CGS-21680, then exposed to 0.2 μM ISO for 3 min.

View this table:
Table 1

Antagonism of the CGS-21680-produced rat ventricular myocyte contractile responses by CGS-15943 and comparison of these responses with those elicited by isoproterenol

3.4 Adenosine receptor agonists and contractile responses in the presence of adenosine receptor antagonists

Adenosine A2 receptor antagonists inhibited the increases in the contractile variables elicited by CGS-21680. The A2 receptor antagonist, CGS-15943, at 10 μM prevented the increase in myocyte contractile variables produced by 10 μM CGS-21680 (Table 1). CGS-21680 at 1 μM in the presence of 0.2 μM DPCPX increased ΔL, DS, TPS, TR, and +dL/dtmax by 24–39% (Table 2). These increases in the contractile variables were similar to the increases observed in the absence of DPCPX (Fig. 2). Moreover, the A2a receptor antagonist, CSC, at 1 μM prevented the increase in the contractile variables elicited by 1 μM CGS-21680 in the presence of 0.2 μM DPCPX.

View this table:
Table 2

Antagonism of the CGS-21680-elicited rat ventricular myocyte contractile responses by CSC in the presence of DPCPX

3.5 SHA-082, NECA and adenosine elicited contractile responses

Two adenosine analogues, SHA-082 and NECA, as well as adenosine, increased contractility of ventricular myocytes (Fig. 4, Table 3). SHA-082 at 10 μM increased +dL/dtmax by 30%. NECA at 10 μM increased ΔL, DS, TPS and TR by 20–28%. The contractile responses elicited by SHA-082 and NECA were not observed when A2 receptors were blocked with 10 μM CGS-15943. While 100 μM adenosine or 0.2 μM DPCPX were without effect when administered alone, the two agents when administered together caused 20–37% increases in ΔL, DS, TPS, TR and +dL/dtmax. These adenosine-induced increases were inhibited by 10 μM CGS-15943.

Fig. 4

Effect of SHA-082 and CGS-15943 on the maximum rates of shortening (+dL/dtmax) and relaxation (−dL/dtmax) of isolated rat cardiomyocytes and adenylyl cyclase activity of cultured rat cardiomyocytes. SHA-082 and CGS-21680 were used at 10 μM. Values are means±s.e. (n=6). Asterisk denotes significance from corresponding control. Dagger denotes significance from the corresponding SHA-082 value.

View this table:
Table 3

Contractile responses of rat ventricular myocytes to NECA and adenosine in the absence and presence of adenosine receptor antagonists

3.6 Adenosine receptor agonists and adenylyl cyclase activity

In a membrane preparation from ventricular myocytes CGS-21680 caused an activation of adenylyl cyclase at concentrations of 1-100 μM (Fig. 5). However, the effect appears to be biphasic in that subsequent to a rise in adenylyl cyclase activity with 1.0 μM CGS-21680, the enzyme activity fell progressively until significance was absent with 1 mM CGS-21680. In the presence of 0.1 μM DPCPX, the activation curve of adenylyl cyclase caused by increasing concentrations of CGS-21680 was sigmoidal with an EC50 of 0.26±.08 μM. The latter increase in cyclase activity of 49% was prevented by 10 μM CGS-15943. It is interesting that when homogenates of rat heart ventricular myocardium were used, CGS-21680 increased adenylyl cyclase activity only when concentrations of this A2 agonist exceeded 100 μM (data not shown).

Fig. 5

Effect of CGS-21680 on ventricular myocyte membrane adenylyl cyclase activity. Adenylyl cyclase activity was assessed in the absence (s̄) or presence (c̄) of 0.1 μM DPCPX or DPCPX+10 μM CGS-15943 at 0–1 mM CGS-21680, as indicated, for 9 min. Each value represents the mean±s.e. for 16 individual membrane preparations. Asterisks denote significance from the appropriate zero CGS-21680 value. Daggers denote significance from the corresponding value in the presence of DPCPX.

NECA increased adenylyl cyclase activity at concentrations of 1 μM and higher, with 100 μM NECA increasing the activity approximately two-fold (Fig. 6). NECA is an adenosine receptor agonist with equal affinities for A1 and A2 subtypes. In the presence of 0.1 μM DPCPX the NECA adenylyl cyclase activity relationship became sigmoidal with an EC50 value of 0.30±.07 μM. With A1 receptor blockade, the adenylyl cyclase activity was stimulated by 1 μM NECA an additional 43%, with a commensurate increase occurring with 10 μM NECA. CGS-15943 at 10 μM prevented the NECA-induced increase in membrane cyclase activity. SHA-082 at 10 μM had no effect on the activity of adenylyl cyclase activity (Fig. 4). Adenosine at 0–1 mM in the absence or presence of 1 μM DPCPX did not increase adenylyl cyclase activity above basal levels in membranes obtained from ventricular myocytes or myocardium (data not shown).

Fig. 6

Effect of NECA on ventricular myocyte membrane adenylyl cyclase activity. Adenylyl cyclase activity was assessed in the absence (s̄) or presence (c̄) of 0.1 μM DPCPX or DPCPX+10 μM CGS-15943 at 0–0.1 mM NECA, as indicated, for 9 min. Each value represents the mean±s.e. for 11 individual membrane preparations. Asterisks denote significance from the appropriate zero NECA value. Daggers denote significance from the corresponding NECA values in the absence of CGS-15943. Double daggers denote significance from the corresponding NECA values in the absence of DPCPX.

3.7 Isoproterenol and adenosine A2 receptor agonists augment cAMP levels

β-Adrenergic receptor stimulation with ISO at 0.1 and 1 μM for 1 min increased basal ventricular myocyte cAMP (4.86 pmol/mg cell protein) by 90 and 138%, respectively (n=6). CGS-21680 at 1 and 10 μM for 10 min caused an increase in myocyte cAMP by 77 and 174%, respectively (Fig. 7). The apparent EC50 value for the A2 agonist was at least 1.15±0.37 μM. The A2 receptor antagonist, DMPX, prevented the increase in cAMP caused by 10 μM CGS-21680. SHA-082 at 0.1–10 μM had no effect on cAMP in isolated ventricular myocytes (data not shown).

Fig. 7

Effect of CGS-21680 on cAMP content of ventricular myocytes. Cultures were exposed for 10 min to 0 (B), 0.1 μM (C7), 1 μM (C6), or 10 μM (C5) CGS-21680, 1 μM DMPX (D) or 1 μM DMPX plus 10 μM CGS-21680 (DC5). Myocyte cAMP levels were then determined as described in Section 2. Each value represents the mean±s.e. of 5 individual culture plates obtained from 16 separate preparations. Asterisks denote significance from control or basal (B). The dagger denotes significance from the corresponding C5 and C6 value.

4 Discussion

The major finding of this study is that adenosine A2 receptors share a presence with A1 receptors in the rat ventricular myocyte. In single myocytes A1 receptor stimulation was associated with the attenuation of β-adrenergic-induced contractile responses. This observation verified the presence of functional A1 receptors. Adenosine A2 receptor stimulation of the myocytes was associated with positive inotropic responses that were markedly different from β-adrenoceptor-mediated contractile responses.

4.1 Adenosine A2 receptors and positive inotropic responses

Adenosine A2 receptor agonists produced positive inotropic responses in contracting ventricular myocytes. CGS-21680 is a selective A2 receptor agonist with an affinity of 10–20 nM [24]. CGS-21680 caused a concentration-dependent increase in ventricular myocyte contractility. This strongly suggests the presence of functional A2 receptors. The agonist increased the delta length (shortening), duration of shortening, time-to-peak shortening, time-to-75% relaxation and +dL/dtmax with an EC50 ranging from 80 to 121 nM for these contractile variables. SHA-082, NECA and adenosine also increased myocyte contractility. However, adenosine required the presence of DPCPX to enhance myocyte contractility. DPCPX is a highly selective A1 receptor antagonist with an affinity of 0.5–0.7 nM [25]. The A2 receptor antagonist, CGS-15943 [26], prevented the contractile responses elicited by the A2 receptor agonists. It is interesting that A2-adenosinergic agonists increased +dL/dtmax without affecting −dL/dtmax.

The above results indicate that adenosine A2 receptor stimulation increases contractility in rat ventricular myocytes. Further evidence supporting A2 receptor activity is that A1 receptor antagonist administration did not alter the contractile responses caused by A2 receptor agonists. Moreover, the increase in contractility caused by CGS-21680 was antagonized by a selective A2a receptor antagonist, CSC. This antagonist has nanomolar affinity for A2a receptors [27]. This suggests that the A2-adenosinergic-elicited positive inotropic responses observed were primarily mediated by A2a receptor stimulation. However, activation of A2b receptors cannot be ruled out with higher concentrations of CGS-21680. Further investigations concerning the action of A2b receptors awaits availability of selective A2b antagonists. An adenosinergic-elicited positive contractile response has been reported previously in neonatal avian and rat ventricular myocyte preparations to occur by both A2a and A2b receptor stimulation [12, 13, 28]. It may be suggested that A2b receptor stimulation may also be involved in the present study because higher concentrations of adenosine and NECA increased myocyte contractility. However, this conclusion is indirect and remains to be substantiated. The current results with adenosine do not agree with the report of Xu et al. [12]indicating that 1 μM adenosine increases rat myocyte contraction. The reason for this difference is not known except that the preparation of ventricular myocytes is different. The present findings do not agree with reports indicating the absence of a positive inotropic effect of A2 agonists in isolated ventricular myocytes from guinea-pig, rabbit and rat hearts [15, 17]and perfused hearts from guinea-pigs [29]. An explanation for the absence of a contractile effect in these reports is not readily apparent. However, negative contractile results do not conclusively prove that a receptor and/or its respective signal transduction proteins are absent. The absence of an A2-adenosinergic effect in perfused guinea-pig hearts [29]may suggest species differences. In the present study individual ventricular myocytes were continually suffused with fresh medium and were free of other myocardial cell types. Moreover, adenosine has been previously reported to increase contractile force development in ventricular muscle preparations [18–20]as well as rat ventricular myocytes [13, 22].

4.2 Comparison of β and A2 receptor contractile responses

There are several differences between the positive inotropic responses caused by β-adrenergic and A2-adenosinergic stimulation of ventricular myocytes. While β and A2 receptor stimulation each increased the delta length and +dL/dtmax, the duration of shortening, time-to-peak shortening and time-to-75% relaxation were decreased by β-adrenergic stimulation, but increased by A2-adenosinergic stimulation. β-Adrenergic stimulation increased the maximum rate of relaxation, −dL/dtmax, whereas A2-adenosinergic stimulation was without effect on this contractile variable. The differences between adrenergic- and adenosinergic-induced contractile responses suggest that the inotropic responses are mediated by different mechanisms. The A2-adenosinergic-induced inotropic response is not an abbreviated contraction as is observed with β-adrenergic stimulation. Thus, the positive inotropic response by A2-adenosinergic stimulation is reminiscent of the inotropic response observed with an increase in extracellular Ca2+ concentration [30–32]. Because of the differences between the β-adrenergic- and the A2-adenosinergic elicited inotropic responses, the adenylyl cyclase activities and cAMP formation in ventricular myocytes were examined with each type of stimulation.

4.3 Adenosine A1 and A2 receptors, adenylyl cyclase and cAMP

Adenosine A1 receptor stimulation with PIA reduced the adrenergic-induced increase in adenylyl cyclase of ventricular myocyte membranes. DPCPX prevented the attenuation by PIA of the ISO induced increase in adenylyl cyclase activity. These results support previous studies [4]suggesting that the antiadrenergic action of adenosine is mediated by A1 receptors on the ventricular myocytes. The effect of CGS-21680 on adenylyl cyclase appeared to be biphasic, with the large increase in activity observed with 1 μM of agonist reversing as the agonist concentration became higher. However, this biphasic effect was not present with DPCPX, indicating that high concentrations of CGS-21680 may interact with myocyte A1 receptors [24], resulting in an attenuation of the A2 receptor induced increase in enzyme activity. CGS-15943 prevented the increase in adenylyl cyclase activity produced by CGS-21680, thereby confirming the action of A2 receptor activity on the enzyme activity. Only very high concentrations (mM) of CGS-21680 significantly increased adenylyl cyclase in membranes prepared from homogenates of ventricular myocardium, possibly because this membrane preparation contained membranes from a variety of cell types which originate from nervous, vascular and connective tissues.

NECA and adenosine were less effective in enhancing myocyte membrane adenylyl cyclase activity in the absence of DPCPX. Once again this may relate to the relative affinities of the agents for A1 and A2 receptors. At 1 μM, NECA, which has equal affinities for the two adenosine receptors, had a small, albeit significant, enhancing effect on adenylyl cyclase activity. In that inhibition of A1 receptors by DPCPX resulted in a greatly enhanced effect of NECA, it is suggested that the attenuating effect of A1 receptor activity masked the action of A2 receptors.

β-Adrenergic and A2-adenosinergic stimulation of ventricular myocytes increased cellular cAMP levels. This is in agreement with numerous reports [8, 12, 15, 28]. Moreover, the A2 receptor antagonist, DMPX, prevented the cAMP response caused by CGS-21680. DMPX is a selective A2 receptor antagonist with an affinity of ∼10 μM [33].

4.4 Integrated action of adenosine A1 and A2 receptors

The results of this study support the notion that A2a receptor stimulation fosters a positive inotropic response, activates adenylyl cyclase activity and increases cAMP levels in isolated rat ventricular myocytes which also exhibit A1 receptor activity. These findings agree with previous results reported for neonatal avian ventricular myocytes [12, 28]. The present mechanical results further describe increases in delta length (shortening), duration of shortening, time-to-peak shortening, time-to-75% relaxation and maximum rate of shortening with no change in the maximum rate of relengthening of contracting rat ventricular myocytes stimulated with A2 receptor agonists. In mammalian ventricular preparations A2 receptor stimulation has been reported to activate adenylyl cyclase and increase cAMP [4, 14, 15]. However, other reports suggest that A2 receptor stimulation is without effect on cAMP [16, 17]or contractility [17]. While A2 receptor stimulation of guinea-pig ventricular myocytes increased cAMP, it did not enhance contractile force [14, 15]. Exposure of ventricular muscle preparations to 100–1000 μM levels of adenosine has been reported either to increase [18–20]or not to change [21]contractile force development. In one report the increase in contractile force was not associated with an increase in adenylyl cyclase or cAMP [19]. The reasons for the discrepancies in these reports is not apparent.

The present study suggests that the positive inotropic responses elicited by A2a-adenosinergic stimulation may not be solely mediated by adenylyl cyclase/cAMP mechanisms. The EC50 values for the CGS-21680 elicited contractile responses were 3-12 times less than those for the induced adenylyl cyclase/cAMP responses. While SHA-082 at 10 μM increased myocyte contractility, it did not affect the levels of cAMP. Moreover, the absence of an increase in the rate of myocyte relaxation with A2 agonist stimulation, an event not characteristic of β-adrenergic stimulation, indicates that even when the higher levels of an A2 agonist are employed and cAMP levels increase, there is still no increase in the rate of myocyte relaxation. This implies that either there is an uncoupling of the adenylyl cyclase/cAMP response or a different contractile-mediated process. If any adenylyl cyclase/cAMP-independent components of the A2a-adenosinergic contractile responses exist, the present studies do not indicate what those processes might involve. Recently a cAMP-independent mechanism has been described, whereby the A2a receptor may enhance the inotropy in chick embryo ventricular cells [34]. The underlying action of A2a activity was suggested to include Gs-sensitive stimulation of cellular Ca2+ influx, perhaps providing an explanation for the presently reported A2 contractile action reminiscent of an increase in extracellular Ca2+, as discussed above. While the evidence presented herein suggests indirectly that the A2 receptor may not solely utilize the adenylyl cyclase/cAMP mechanism, this possibility deserves further study.

The importance of A2-adenosinergic stimulation in the normal oxygenated myocardium in either the absence or presence of A1 receptor blockade requires further investigation. Additionally, adenosine A2 receptor activation in the ischemic, hypoxic or failing myocardium [35]where the interstitial levels of adenosine are elevated [36]should be explored.


The authors thank Lynne M. Shea and Dan L. Macumber for their excellent technical assistance. This study was supported by Public Health Service Grants (primarily HL-22828 and in part by AG-11491 and HL-36964). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the awarding agencies.


  • 1 A preliminary report of this work has been presented in abstract form (Drug Dev Res 1994;1:265).


View Abstract