Naunyn-Schmiedeberg's

Archives of Pharmacology

Naunyn-Schmiedeberg's Arch Pharmacol (1989) 340:689-695

© Springer-Verlag1989

Effects of adenosine analogues on contractile response and cAMP content in guinea-pig isolated ventricular myocytes Joachim Neumann, Wilhelm Schmitz, Hasso Scholz, and Birgitt Stein Abteilung Allgemeine Pharmakologie, Universit~its-Krankenhaus Eppendorf, Universit/it Hamburg, Martinistrasse 52, D-2000 Hamburg 20, Federal Republic of Germany Summary. In the present study the effects of adenosine analogues were investigated on cAMP content and contractile response in guinea-pig ventricular myocytes. The adenosine analogues (-)-N6-phenylisopropyladenosine (R-PIA), 5'-Nethylcarboxamideadenosine (NECA) and (+)-N6-phenyl isopropyladenosine (S-PIA) in the presence of 0.01gmol/1 isoprenaline reduced contractile response concentrationdependently. R-PIA and N E C A were about equipotent (IC25:0.01 gmol/1 and 0.039 ~tmol/1 respectively), while SPIA was less potent (ICzs : 0.6 gmo1/1). Isoprenaline stimulated cAMP content was reduced by R-PIA (IC25: 0.004 ~tmol/1) and with lower potency by S - H A (IC2s: 0.15 ~tmol/1), but the extent of reduction of cAMP by R-PIA and S-PIA (to 55% and 64% respectively) was less than the reduction of contractile response (to 26% and 55% respectively). This suggests that the effects of R- and S-PIA on contractile response are only in part due to a reduction in cAMP content. In addition, N E C A did not decrease cAMP content but decreased contractile response to the same extent as R-PIA. Similar results were obtained in the presence of the phosphodiesterase inhibitor Ro 20-1724. Time course studies revealed that the effects of R-PIA (1 gmol/1) on cAMP content and contractile response coincided reaching steady state after 5 min and remained stable thereafter. The effects of N E C A (1 gmol/1) on contractility also reached steady state within 5 rain, whereas it did not change cAMP content. It is concluded that the reduction of contractility by adenosine analogues in the presence of isoprenaline can only in part be explained by a reduction of cAMP content. It is suggested that a cAMP-independent effect, possibly an activation of phosphatases, might be involved additionally. Key words: Adenosine analogues - cAMP - Isolated ventricular myocytes - Contractile response - Guinea-pig hearts - Isoprenaline - Phosphodiesterase inhibitor

Introduction Adenosine, an endogenous nucleoside, antagonizes the positive inotropic effects of catecholamines in ventricular myocardium (for review see Dobson and Fenton 1983), and there Abbreviations. R-PIA, (-)-N6-phenylisopropyladenosine; NECA,

5'-N-ethytcarboxamideadenosine; S-PIA, (+)-N6-phenylisopropyl adenosine; Ro 20-1724, 4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone Send offprint requests to J. Neumann at the above address

is evidence that the ionic basis of the antagonism between adenosine and isoprenaline consists finally in a reduction of isoprenaline-stimulated calcium inward current (B6hm et al. 1985). External adenosine receptors have been classified into two classes with respect to their inhibitory (A1) or stimulatory (A2) action on adenylate cyclase activity (van Calker et al. 1979; Londos et al. 1980). In atrial tissue an adenosine receptor couples via a GTP-binding regulatory protein directly to a potassium channel, giving rise to a cAMP-independent negative inotropic effect (B6hm et al. 1986; Schmitz et al. 1988). In ventricular tissue, however, there are conflicting reports. In initial studies on multicellular ventricular preparations some investigators observed a decrease in isoprenaline-stimulated cAMP content (Schrader et al. 1977; Dobson 1978, 1983), while others reported no change (B6hin et al. 1984; Schmitz et al. 1985; Scholz et al. 1987) or even an increase in cAMP content (Huang and Drummond 1978). These divergent results have been explained (Henrich et al. 1987) by cellular heterogeneity of cardiac tissue, which is comprised of cardiomyocytes and coronary endothelial cells, the latter presumably possessing Az-adenosine receptors (Schfitz et al. 1986; Des Rosier and Nees 1987). But even in studies on isolated ventricular myocytes conflicting results have been reported. Some investigations revealed a decrease of cAMP by adenosine analogues in the presence of isoprenaline (West et al. 1986; Henrich et al. 1987; Martens et al. 1987) whereas others presented evidence for stimulatory A2-adenosine receptors (Anand-Srivastava and Cantin 1983; Anand-Srivastava 1985). Inotropic responses were not determined in these studies. The present paper tries to correlate effects of adenosine analogues on cAMP content and contractile behaviour of isolated ventricular myocytes free of endothelial components. We used myocytes from guinea pigs in order to insure comparability of the present results to our earlier studies on cardiac preparations from guinea pigs. All experiments were performed in the presence of the enzyme adenosine deaminase to exclude interference from endogenous adenosine (LaMonica et al. 1984; Neumann et al. 1987). Parts of the results have been reported in abstract form (Neumann et al. 1988; Stein et al. 1988).

Methods Isolation o f myocytes. Myocytes were isolated as described

by Piper et al. (1982) with minor modifications. Guinea pigs (300-400 g) were stunned by cervical dislocation. Hearts

690 were rapidly excised, mounted on a modified Langendorff perfusion system and perfused (STA-multipurpose pump, Desaga, Heidelberg, FRG) retrogradely via cannulation of the aorta in a nonrecirculating manner at a constant rate of 10 ml/min for three rain with a calcium-free buffer (solution A) containing (in mmol/1) NaC1 100.0, KC1 10.0, KH2PO4 1.2, MgSO4 5.0, glucose 20.0, taurine 50.0, 3-(N-morpholino)propanesulfonic acid (MOPS) 10.0, gassed with O2, and maintained at 37°C; the pH was adjusted to 6.9. After washing to remove blood cells, the heart was perfused with the same medium containing 0.1% collagenase in a recirculating fashion at a rate of 30 ml/min for 27 min. Following enzyme perfusion atria were cut off and ventricles were minced and incubated for one hour in solution B ("power soup" according to Isenberg and K16ckner 1982) which consisted of (in mmol/1) KC1 70.0, KzHPO4 30.0, MgSO4 20.0, taurine 20.0, succinic acid 5.0, creatine 5.0, EGTA 1.0, flhydroxybutyric acid 7.3, pyruvic acid 5.0, NazATP 5.0, gassed with 02 at 37 ° C, pH adjusted to 7.4. The cell suspension was passed through 200 gm mesh nylon gauze (Heidland, Giitersloh, FRG) and centrifuged for 3 rain at 25 g (MSE 18, England). The resulting cell pellet was resuspended in solution A. Calcium was gradually increased to 0.38 retool/1. This cell suspension was centrifuged (MSE, SK 1, England) for 1 min through a gradient of solution A containing 4% albumin. The final cell pellet was resuspended in serum-free medium 199 (containing 0.25 retool/1 streptomycin and 0.25 retool/1 penicillin G) to adjust a density of 1 - 3 x 105 cells per ml. One ml of this cell suspension was plated on each petri dish (Falcon No. 3080, Becton Dickinson, Lincoln Park, New Jersey, USA) in a laminar air flow cabinet (Gelaire, TC 48, Flow Laboratories, Meckenheim, FRG). Dishes were kept in an incubator (Haereus, B 5060 EK, Hanau, FRG) providing an atmosphere with 5% CO2, 95% air and 95% humidity at 37°C. Based on protein the yield of cells per heart was about 30%. Final ventricular cell preparations contained 6 5 % - 9 0 % rod shaped myocytes. Preparations with less than 65% of viable myocytes in the presence of calcium were excluded from experiments. Cells were considered viable when, in millimolar calcium concentrations, they had a rod-shaped appearance, clear striation, sharp edges, and no evidence for granulation or blebs (Capogrossi et al. 1986). Histological examination of plated cells according to Burstone (1962) revealed less than 0.1% non-muscle cells. Passing the cells through the albumin gradient the activity of the endothelial marker angiotensin converting enzyme, detected according to Neels et al. (1983), was reduced from 7.56_+ 0.53 to 0.87 _+ 0.15 nmol/mg protein min (n = 5). Measurement of contractile response. Petri dishes were placed

in a perfusion chamber (Th. Zindt, Hamburg, FRG) on the stage of an inverted Labovert microscope (Leitz, Wetzlar, FRG). A specially designed cover glass was placed into the petri dish so that a continuous layer of cell suspension with a thickness of 3 mm was formed. This was necessary to obtain homogeneous stimulation of myocytes via platinum wire electrodes vertically protruding into the solution. Stimulation frequency was 1 Hz applied from a Grass SD 9 stimulator (Quincy, Mass., USA). According to Reithmann et al. (1987) settings were at 15 ms duration and output of 100 V direct current. The inlet to the perfusion chamber was connected to a roller pump (Minipuls HP 2, Abimed, Langenfeld, FRG) so that the ceils could be continuously

perfused (2.3 ml/min) with a variety of prewarmed (heat exchanger, Haake and Buchler, Berlin, FRG) test solutions from the reservoirs. The temperature in the dishes was maintained at 35°C. The basal solution was a modified Tyrode solution consisting of (in retool/l) NaC1 119.8, KC1 5.4, CaC12 1.8, NaHzPO4 0.42, MgCI2 1.05, glucose 5.0, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) 5.0, NazEDTA 0.05, ascorbic acid 0.28, adenosine deaminase 1 ~tg/ml, gassed with O2, pH adjusted to 7.4. Different drugs were applied by changing the reservoir. All drugs were freshly dissolved in prewarmed and preaerated Tyrode solution. Discharge of solution from the dish was achieved by a filter pump (Brand, Wertheim, FRG) so that the volume of solution in the dish was constant. The image of the myocytes observed through the inverse microscope was recorded by a TV camera (Grundig, FA 76, Ftirth, FRG), projected on a screen (Grundig, BG 431, Fiirth, FRG) and simultaneously fed into a video recorder (Sony, VO 5630, K61n, FRG). This technique enabled us to study simultaneously in one experiment the effect of drugs on up to 8 beating cells. Contractions were quantified by eye on a monitor in % change of diastolic cell length. The percental changes in cell length from 5 to 8 myocytes per dish and condition were averaged. The linear magnification of the whole system (microscope - camera - monitor) was 640 times. The extent of magnification was adequate to monitor changes in cell length. The contractions of myocytes were stable after 5 min. Up to 8 h after plating contractile response and cAMP content of stimulated myocytes were the same. The percental changes in cell length reported in this study are in agreement with other reports (Harding et al. 1988). Videotapes were analysed independently by two observers with high reproducibility. Reserpine pretreatment (5 mg/kg, 16 h) did not change the contractile response to isoprenaline. In addition, tyramine at 29 gmol/1 did not enhance contractile response. This indicates the lack of endogenous catecholamines. The effect of isoprenaline (0.01 lamol/1) on contractile response was completely reversible upon addition of propranolol 10 gmol/1. This indicates the presence of intact fl-adrenoceptors. Determination of cAMP content. At the end of the appropri-

ate perfusion periods 100 gl of 1 M HC1 was applied to each dish, which contained 1 ml myocyte suspension at this time. 1 ml from this cell solution was quickly transferred to microtubes, heated at 95°C for 10 min and kept at - 2 0 ° C until the next day to determine cAMP content. Microtubes were then heated at 56°C for 1 h, sonified (10 s; Sonorex RK 100, Bandelin, Berlin, FRG) and centrifuged for 15 min at 10000 x g (Eppendorf centrifuge 5414, Hamburg, FRG) as described by Hazeki and Ui (1980). Protein was determined according to Bradford (1976). cAMP was measured by radioimmunoassay as published previously (Brfickner et al. 1985). Recoveries of cAMP run with each experiment were 105.2 _+ 5.1% (n = 120). Drugs used were (-)-N6-phenylisopropyladenosine, (+)-N6-phenylisopropyladenosine, cAMP, adenosine deaminase from calf intestine (200 U/ml) (all from Boehringer, Mannheim, FRG), (+)-isoprenaline HC1 (Boehringer Ingelheim, FRG), Bio-Rad protein assay and gamma globulin standard were obtained from Bio-Rad (Miinchen, FRG). 2'-O-succinyladenosine 3',5'-monophosphate tyrosine methyl ester (Sigma, Chemical Co., Mfinchen, Materials.

691 Guinea-pig ventricu[ar

A ~Io

a

b

myocytes c

Guinea-pig ventricular A

myocytes

6-

d

5-

~2 13u

o

E

-+

"6

n

2-

~0


B gl0-

10-

B

H I , , , , , ,

I0"7

u

8-

g o

0

Ctr

IR-PIA

NECA

~_o/11 "~"~.~:

S-PIA I

11Jmol/[

Fig. 1. Cyclic AMP content (A; pmol/mg protein) and contractile response (B; % change in cell length) in electrically stimulated guinea-pig ventricular myocytes in drug-flee bathing solution (Ctr, a) and 5 rain after the application of (-)-N6-phenylisopropyladenosine (R-PIA, 1 ~tmol/1, b), 5'-N-ethylcarboxamideadenosine (NECA, 1 gmol/1, c) and (+)-N6-phenylisopropyladenosine (S-PIA, I gmol/1, d). Figures in columns denote number of petri dishes from 5 - 6 guinea-pig hearts. For the data in B the % change in cell length from five to eight different myocytes per petri dish were averaged. *p < 0.05 vs a

F R G ) were iodinated using Na125j (Amersham-Buchler, Braunschweig, F R G ) as described by Struck et al. (1977). 5'N-ethylcarboxamideadenosine was a gift from Byk Gulden Lomberg, Chemische Fabrik, Konstanz, F R G . Ro 20-1724 4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone was a gift from Hoffmann-LaRoche, Grenzach-Wyhlen, F R G . Collagenase and medium 199 were from Biochrom (Berlin, FRG). Streptomycin sulfate, Penicillin G, bovine serum albumin (A 9647) were from Sigma Chemical Co. (Mfinchen, FRG). Reserpine was from Ciba G m b H , Wehr, F R G . All other chemicals were of analytical or best commercial grade available. Deionized and twice distilled water was used throughout.

Statistics. The data shown are means _+ SEM. Statistical significance was estimated with Student's t-test for unpaired observations. A P-value of less than 0.05 was considered significant. The drug concentrations producing 25% of inhibition (IC25) were determined graphically.

Results Effects of adenosine analogues alone on contractile response and c A M P content At 1 ~tmol/1 R - P I A and N E C A reduced contractile response to about 50% (Fig. 1 B) while S-PIA was ineffective. The effects of R - P I A and N E C A were not accompanied by a reduction in c A M P content (Fig. 1 A). In initial experiments we, like others (Piper et al. 1982), employed fetal calf serum in our medium. This may be contaminated with catecholamines. Therefore, in the experiments reported here no serum was added. So the effects of adenosine analogues cannot be regarded as "antiadrenergic". In addition, iden-

.c o

0 Isoprenallne • Iso*R-PIA • Iso+NECA * P < 0.05 vs. S rnin n = 8 -12

2-

0-

~--II

0

~;

. . . .

,

,* [~JJl~

,1~ "~

10

115

Time (rnin)

Fig. 2. Time course for the effects of isoprenaline (Iso, 0.01 pmol/l), (--)-N6-phenylisopropyladenosine (R-PIA, 1 Ixmol/1) and 5'-Nethylcarboxamideadenosine (NECA, 1 gmol/1) in the presence of isoprenaline on cAMP content (A; pmol/mg protein) and contractile response (B; % change in cell length) in electrically stimulated guinea-pig ventricular myocytes. 5 min after the addition of Iso, PIA or NECA were applied for the indicated times, n = number of petri dishes from 5 guinea-pig hearts. For the data in B the % change in cell length from five to eight different myocytes per petri dish were averaged. © Isoprenaline; • Iso + R-PIA; • Iso + NECA; • p < 0.05 vs 5 min; n = 8--12

tical results were obtained when myocytes were prepared from guinea pigs pretreated with reserpine (data not shown).

Effects of adenosine analogues in the presence of isoprenaIine on contractile response and c A M P content Figure 2 shows the time course for the effects o f R - P I A and N E C A on c A M P content and contractile response in isoprenaline-stimulated myocytes. The time course for the effects of isoprenaline (0.01 lamol/1) alone is shown as control. Isoprenaline about doubled c A M P content and contractile response within 5 min. These effects were stable for the time of investigation. After 5 min perfusion with isoprenaline, additionally applied R - P I A (1 gmol/1) reduced both c A M P content and contractile response with a similar time course. These reductions were maximal and stable after 5 min. In contrast, additionally applied N E C A (1 I~mol/1) did not affect c A M P content during the entire time of investigation, whereas contractile response was reduced with the same time course and extent as by R-PIA. This N E C A induced decrease was also stable after 5 rain. Hence, the concentration response curves were obtained after 5 rain incubation of myocytes with isoprenaline (0.01 pmol/1) and subsequent 5 min in the presence of the adenosine analogue as indicated. In the presence of isoprenaline R - P I A concentrationdependently reduced contractile response (IC25:0.01 Ixmo1/1,

692 Guinea-pig ventricular myocytes A ~ 120

Guinea - pig ventricular myocy t e s

A

-

510-

o

b

e

Ctr

Ro

I R-PIA

d

•

No

I S-PIA

o

e~

=* 8c

~_ 5-

.i

fl,.

I[

B ,~10a~

r--[I,

II

,

,

,

,

,

,

,

B1

~I00

u

-

80-

//

",~i.._.._ T ~ m .z. ' - ~

\

8, c o

.el

6o

60-

£ _~ 40.~ 20

g

5"

.c_

• .EcA O S-PIA n :6-17

~,"'i

0 r----~l ,

lprno[/I

~ T --i_'~

o p<0.05 O v s . ° i p< 0.05• v s . 1 Ap<0.05 O v s . l

I~

N I

- log [agonist] mol/I Iso

10min

NECA

I

+ oganist

5min * 5min

Fig. 3. Cyclic AMP content (A; % of isoprenaline value) and contractile response (B; % of isoprenaline value) in electrically stimulated guinea-pig ventricular myocytes in drug-free bathing solution (Ctr) and 10rain after the application of isoprenaline (Iso, 0.01 gmol/1). 5 min after the addition of Iso, different concentrations of (-)-N6-phenylisopropyladenosine (R-PIA), 5'-N-ethylcarboxamideadenosine (NECA) or (+)-N6-phenylisopropyladenosine (SPIA) were applied additionally for 5 min. Iso stimulated cAMP content from 3.66 _+0.10 to 5.57 _+0.42 pmol/mg protein (n = 36) and contractility from 3.61 _+0.20 to 7.62 _ 0.33% change in cell length (n = 32-40). n = number ofpetri dishes from 5 - 6 guineapig hearts. For the data in B the % change in cell length from five to eight different myocytes per petri dish were averaged. • R-PIA; • NECA; O S-PIA; n = 6 - - 1 7 ; ~ p<0.05 © vs • ; I p < 0 . 0 5 • vs • ; i p < 0.05 © vs •

Fig. 3 B). The effect of R - P I A started at 0.001 gmol/1 and was maximal at 10 gmol/1. The c A M P content which had been enhanced by isoprenaline was also reduced by R - P I A starting at 0.01 gmol/1 and reached maximum at i ~tmol/1 (IC25: 0.004gmol/1, Fig. 3A). Likewise N E C A reduced contractile response concentration-dependently (IC25 : 0.039 gmol/1, Fig. 3B). However, this effect was not accompanied by a reduction in c A M P content (Fig. 3 A). SPIA, the stereoisomer o f R - P I A concentration-dependently reduced both contractile response (IC2s: 0.6~tmol/1, Fig. 3 B) and c A M P (IC25:0.15 gmol/1, Fig. 3 A). However, S-PIA was about 15 times less potent than R - P I A in reducing c A M P content and about 60 times less potent in reducing contractile response.

Effects of adenosine analogues in the presence of Ro 20-1724 on contractile response and cAMP content In order to amplify the effects of the adenosine analogues on c A M P content additional experiments were performed in the presence of the phosphodiesterase inhibitor Ro 201724. Ro 20-1724 (0.5 retool/l) alone approximately doubled

Fig. 4. Cyclic AMP content (A; pmol/mg protein) and contractile response (B; % change in cell length) in electrically stimulated guinea-pig ventricular myocytes after t 5 rain in drug-free bathing solution (Ctr, a), 15 min after addition of 0.5 mmol/1 Ro 20-1724 (Ro), 10 rain after addition of 0.5 mmol/1 Ro 20-1724 and 5 rain after additional application of (-)-N6-phenylisopropyladenosine (R-PIA, I gmol/1, e), 5'-N-ethylcarboxamideadenosine (NECA, 1 gmol/1, d) and (+)-N6-phenylisopropyladenosine (S-PIA, t gmol/1, e). Figures in columns denote number of petri dishes from 5 - 7 guinea-pig hearts. For the data in B the % change in cell length from five to eight different myocytes per petri dish were averaged. * p < 0.05 vs a; + p < 0.05 vs b

both c A M P content and contractile response (Fig. 4). The adenosine analogues, R-PIA, S-PIA and N E C A (1 gmol/1 each) reduced the effect of Ro 20-1724 on contractile response without affecting c A M P content (Fig. 4).

Effects of adenosine analogues in the presence of Ro 20-1724 and isoprenaline on contractile response and cAMP content The interaction between isoprenaline and adenosine analogues was also studied in the presence of R o 20-1724 (0.5 mmol/1). Ro 20-1724 enhanced the effect of isoprenaline on both contractile response and c A M P content. However, with the adenosine receptor agonists qualitatively the same results were obtained as in the absence of R o 20-1724. RP I A reduced contractile response (IC25: 0.015gmol/1, Fig. 5 B) and c A M P content (IC25:0.004 gmol/1, Fig. 5A). N E C A reduced contractile response ( I C 2 5 : 0 . 0 3 gmol/1, Fig. 5B) without reducing c A M P content (Fig. 5A). S-PIA was less potent than R - P I A in reducing both contractile response ( I C 2 5 : 1 . 5 gmol/1, Fig. 5B) and c A M P content (IC25:0.1 gmol/1, Fig. 5A).

Discussion

Similar to earlier work on whole hearts or papillary muscles (Schrader et al. 1977; D o b s o n 1978; B6hm et al. 1984) the present study shows that adenosine analogues reduce isoprenaline-stimulated contractile response. In guinea-pig ventricular myocytes similar results have been reported (Belardinelli and Isenberg 1983) for the interaction of isoprenaline and adenosine. However, no pharmacological

693 Guineo-pig ventriculor myocytes A

12o -

÷

,,-~,~y_

~,==80

II

1

i

i " i ~ " ; ~

Y,

I

I

I

- log [agonist] m o l / I

I

wZ.~

~o=4o

~/II/~

IE

I B

ii

I

I

I

I

I

120-

~,-I1~.~_.._ o

"7

g o ¢D

~> e®

,c n = 7-18 o p < 0.05 O vs. Q • p
.G

g

6

r-t~.,

Ctr

Ro

II Rol*lso II I

|5rain 5min*10min

Ro + Iso * a g o n i s t 5rnin+ Stain* 5rain

Fig. 5. Cyclic AMP content (A; % of Ro 20-1724 and isoprenaline) and contractile response (B; % of Ro 20-1724 and isoprenaline) in electrically stimulated guinea-pig ventricular myocytes after 15 rain in drug-free bathing solution (Ctr), t5 rain after addition of 0.5 mmol/1 Ro 20-1724 (Ro) and 5 rain after the application of 0.5 retool/1 Ro 20-1724 and additional application of isoprenaline for 10 rain (0.01 gmol/1, Ro + Iso). 5 rain after the addition of Ro 20-1724, isoprenaline was applied for 5 min and then different concentrations of (-)-N6-phenylisopropyladenosine (R-PIA), 5'-Nethylcarboxamideadenosine (NECA) or (+)-N6-phenylisopropyladenosine (S-PIA) were added subsequently for 5 rain. Iso in the presence of Ro stimulated cAMP content from 2.85 + 0.17 to 9.00_+0.32 (n=38-68) and contractility from 4.28 +0.22 to 9.37 _+0.23% change in cell length (n = 31-43). n = number of petri dishes from 5 - 7 guinea-pig hearts. For the data in B the % change in cell length from five to eight different myocytes per dish were averaged. • R-PIA; B NECA; © S-PIA; n = 7 - 1 8 ; p<0.05 O v s • ; l ~ p < 0 . 0 5 • v s • ; Ap<0.05 © v s • characterization with various adenosine analogues was attempted and cAMP content was not investigated by these authors. In the present study R-PIA was as potent as NECA in reducing contractile response, whereas S-PIA was less potent by about 1.5 orders of magnitude. This corresponds well to our previous findings in guinea-pig papillary muscles (B6hm et al. 1985). In the present study like in others (Martens et al. 1987; Henrich et al. 1987) PIA in the presence of isoprenaline reduced cAMP content. However, these authors only estimated cAMP content but did not simultaneously measure cAMP content and contractile response. In accord with the findings of Henrich et al. (1987) S-PIA like R-PIA reduced cAMP content in the presence of isoprenaline. In contrast, Martens et al. (1987) did not observe any effect of S-PIA up to 10 gmol/1 on cAMP content. Conceivably their divergent results were due to the fact that they employed higher concentrations of isoprenaline (1 ~tmol/1) than Henrich et al. (0.1 ~tmol/1) and in this study (0.01 gmol/1). In contrast to other studies (Henrich et al. 1987; Martens et al.

]987) NECA was without effect on cAMP content in the present study. This difference is not easily explained. It might be argued that our preparations were contaminated by endothelial cells. But angiotensin converting enzyme activity, a marker enzyme of endothelial cells (Ondetti and Cushman 1983), was practically not detectable in our preparations which were routinely passed through an albumine gradient. Moreover, on histological examination non-muscle cells which are reported to comprise up to 80% of cell number in the myocardium (Anversa et al. 1983) were reduced to less than 0.1%. Thus, endothelial contamination cannot account for the differences. It is also unlikely that our divergent results are due to non-equilibrium conditions in this study. As depicted in Fig. 2 the mechanical effects of R-PIA and NECA were in a steady state after 5 rain. Likewise the reduction of cAMP content by R-PIA had a similar time course as its mechanical effects. In addition, NECA exerted no effect on cAMP content over the time investigated. Hence, the divergent results can more likely be explained by species differences. This is not unreasonable since Henrich et al. (1987) in fact reported on species differences in the effects of NECA. While in dog ventricular myocytes NECA (1 ~tmol/1) greatly reduced isoprenaline-stimulated cAMP content it only slightly reduced cAMP content in rat ventricular myocytes. In order to amplify changes in cAMP content we performed experiments in the additional presence of the phosphodiesterase inhibitor Ro 20-1724 in the same concentration (0.5 mmol/l) as employed by others (Martens et al. 1987; Henrich et al. 1987). However, the pharmacological profiles for contractile response and cAMP content remained unchanged. The present study intended to investigate whether a close link exists between a reduction of cAMP content by adenosine agonists and a reduction of contractile response in a homogeneous preparation of ventricular myocytes. The closely related time courses of the R-PIA-induced reduction in cAMP content and contractility are in favour of an important role of cAMP. However, we observed qualitative and quantitative discrepancies between these parameters as we previously did in multicellular papillary muscle preparations (B6hm et al. 1984; Brfickner et al. 1985; B6hm et al. 1988). Quantitatively both R-PIA and S-PIA reduced isoprenaline-stimulated contractile response to a greater extent than they decreased stimulated cAMP content. In contrast, NECA which was as potent as R-PIA in reducing isoprenaline-stimulated contractile response did not reduce cAMP content. We, therefore, conclude that the reduction in isoprenaline-stimulated calcium current by adenosine derivatives which finally reduces contractile force (B6hm et al. 1985; Isenberg et al. 1987) can only in part be explained by a reduction in cAMP content in ventricular tissue. It is also conceivable that both A1- and A2-adenosine receptors are present on the same cell, both of them affecting cAMP content but in an opposite way. However, only the Al-adenosine receptors influence contractility. Interestingly, the coexistence of A1- and A2adenosine receptors on the same cell type have been demonstrated in adipocytes by Garcia-Sainz and Torner (1985). It is interesting to note that the effects of adenosine in the atrium which were previously regarded to be due to a decrease in cAMP content (Linden et al. 1985) have been shown to be independent of this second messenger (B6hm et al. 1986). In addition, there is increasing evidence that the negative inotropic effects of acetylcholine on isoprenaline-

694 stimulated force of contraction are independent of cAMP. They are accompanied by an increased activity of phosphatases (Ahmad et al. 1987). As both adenosine and acetylcholine reduce isoprenaline-stimulated force of contraction in ventricular tissue, it is not unreasonable to assume a similar mechanism of action. In summary, the effects of adenosine analogues on contractile response in the absence and presence of isoprenaline in guinea-pig ventricular myocytes can in part be explained by a reduction in c A M P content but c A M P - i n d e p e n d e n t mechanisms are also important. It is tempting to speculate that these mechanisms include, analogous to acetylcholine, an activation of phosphatases leading to a dephosphorylation of functional proteins and a decrease in ventricular contractility.

Acknowledgement. We are indebted to Prof. Dr. H. Schfifer, Institut ffir Pathologic, Universit/its-KrankenhausEppendorf, for histological examination of muscle cell cultures and we are grateful to Jutta Starbatty for her technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft. It represents part of the doctoral thesis of Birgitt Stein at the Faculty of Biology, University of Hamburg.

References Ahmad Z, Green FJ, Watanabe AM (1987) Autonomic regulation of type 1 phosphatase in cardiac muscle. Circulation 76 (Suppl IV): 342 Anand-Srivastava MB, Cantin M (1983) Regulation of adenylate cyclase in cultured cardiocytes from neonatal rats by adenosine and other agonists. Arch Biochem Biophys 223 : 468 - 476 Anand-Srivastava MB (1985) Regulation of adenylate cyclase by adenosine and other agonists in rat myocardial sarcolemma. Arch Biochem Biophys 243:439-446 Anversa P, Levicky V, Beghi C, McDonald SL, Kikkawa Y (1983) Morphometry of exercise-induced right ventricular hypertrophy in the rat. Circ Res 52: 5 7 - 64 Bellardinelli L, Isenberg G (1983) Action of adenosine and isoproterenol on isolated mammalian ventricular myocytes. Circ Res 53 : 2 8 7 - 297 B6hm M, Brfickner R, Hackbarth I, Haubitz B, Linhart R, Meyer W, Schmidt B, Schmitz W, Scholz H (1984) Adenosine inhibition of catecholamine-inducedincrease in force of contraction in guinea-pig atrial and ventricular heart preparations. Evidence against a cyclic AMP- and cyclic GMP-dependent effect. J Pharmacol Exp Ther 230 :483 - 492 B6hm M, Briickner R, Hackbarth I, Meyer W, Nose M, Schmitz W, Scholz H, Starbatty J (1985) Evidence for adenosine receptormediated isoprenaline-antagonistic effects of the adenosine analogs PIA and NECA on force of contraction in guinea-pig atrial and ventricular cardiac preparations. Naunyn-Schmiedeberg's Arch Pharmacol 331 : 131 - 139 B6hm M, Briickner R, Neumann J, Schmitz W, Scholz H, Starbatty J (1986) Role of guanine nucleotide-bindingprotein in the regulation by adenosine of cardiac potassium conductance and force of contraction. Evaluation with pertussis toxin. NaunynSchmiedeberg's Arch Pharmacol 332:403 - 405 Bradford MM (1976) A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248- 254 Brfickner R, Fenner A, Meyer W, Nobis T-M, Schmitz W, Scholz H (1985) Cardiac effects of adenosine and adenosine analogs in guinea-pig atrial and ventricular preparations: evidence against a role of cyclic AMP and cyclic GMP. J Pharmacol Exp Ther 234: 766- 773 Burstone MS (1962) Enzyme histochemistry and its application in the study of neoplasm. Academic Press, New York, London

Capogrossi MC, Kort AA, Spurgeon HA, Lakatta EG (1986) Single adult rabbit and rat cardiac myocytes retain the C a 2 +- and species-dependent systolic and diastolic contractile properties of intact muscle. J Gen Physiol 88 : 589- 613 Des Rosier C, Nees S (1987) Functional evidence for the presence of adenosine A2-receptors in cultured coronary endothelial cells. Naunyn-Schmiedeberg'sArch Pharmacol 336:94-98 Dobson JG Jr (1978) Reduction by adenosine of the isoproterenolinduced increase in cyclic adenosine 3',5'-monophosphate formation and glycogen phosphorylase activity in rat heart muscle. Circ Res 43:785-792 Dobson JG Jr (1983) Mechanism of adenosine inhibition of catecholamine-induced responses in heart. Circ Res 52 : 151 - 160 Dobson JG Jr, Fenton RA (1983) Antiadrenergic effects of adenosine in the heart. In: Berne RM, Rall TW, Rubio R (eds) Regulatory function of adenosine. Martinus-Nijhoff, Boston, pp 363 - 376 Garcia-Sainz JA, Torner ML (1985) Rat fat-cells have three types of adenosine receptors 8 (Rs, Ri and P). Differential effects of pertussis toxin. Biochem J 232:439-443 Harding SE, Vescovo G, Kirby M, Jones SM, Gurden J, PooleWilson PA (1988) Contractile response of isolated adult rat and rabbit cardiac myocytes to isoproterenol and calcium. J Mol Cell Cardiolog 20: 6 3 5 - 647 Hazeki O, Ui M (1980) Beta1- and betaz-adrenergic receptors reponsible for cyclic AMP accumulation in isolated heart and lung cells. MoI Pharmacol 17:8-13 Henrich M, Piper HM, Schrader J (1987) Evidence for adenylate cyclase-coupled Al-adenosine receptors on ventricular cardiomyocytes from adult rat and dog heart. Life Sci 41: 2381-2388 Huang M, Drummond GI (1978) Interaction between adenosine and catecholamines on cyclic AMP accumulation in guinea-pig ventricular myocardium. Biochem Pharmacol 27 : 187 - 191 Isenberg G, K16ckner U (1982) Calcium tolerant ventricular myocytes prepared by preincubation in a "KB medium". Pfliigers Arch 395:6-18 Isenberg G, Cerbai E, Kt6ckner U (1987) Ionic channels and adenosine in isolated heart cells. In: Gerlach E, Becker BF (eds) Topics and perspectives in adenosine research. Springer, Berlin Heidelberg New York Tokyo, pp 323- 335 LaMonica DA, Frohloff N, Dobson JG Jr (1985) Adenosine inhibition of catecholamine-stimulated cardiac membrane adenylate cyclase. Am J Physiol 248: H 737 - H 744 Leung E, Kwatra MM, Hosey MM, Green RD (1988) Characterization of cardiac AI adenosine receptors by ligand binding and photoaffinity labeling. J Pharmacol Exp Ther 244:1150-1156 Linden J, Hollen CE, Patel A (1985) The mechanism by which adenosine and cholinergic agents reduce contractility in rat myocardium. Correlation with cyclic adenosine monophosphate and receptor densities. Circ Res 56: 728- 735 Lohse M J, Ukena D, Schwabe U (1985) Demonstration of Ri type adenosine receptors in bovine myocardium by radioligand binding. Naunyn-Schmiedeberg'sArch Pharmacol 328 : 310 - 316 Londos C, Cooper DMF, Wolff J (1980) Subclasses of external adenosine receptors. Proc Natl Acad Sci (USA) 77:2551 -2554 Martens D, Lohse MJ, Rauch B, Schwabe U (1987) Pharmacological characterization of A1 adenosine receptors in isolated rat ventricular myocytes. Naunyn-Schmiedeberg'sArch Pharmacol 336: 342- 348 Neels HM, Van Sande ME, Scharpe SL (1983) Sensitive colorimetric assay for angiotensin converting enzyme in serum. Clin Chem 29:1399-1403 Neumann J, Schmitz W, Stein B (1988) Effects of(-)-N6-phenylisopropyladenosine on contractile response and cAMP content in electrically stimulated ventricular myocytes. NaunynSchmiedeberg's Arch Pharmacol 337 (Suppl):R62 Ondetti M, Cushman D (1982) Enzymes of the renin-angiotensin system and their inhibitors. Ann Rev Biochem 51:515- 529 Piper HM, Probst I, Schwartz P, Hiitter FJ, Spieckermann PG (1982) Culturing of calcium stable adult cardiac myocytes. J Mol Cell Cardiol 14:397--412

695 Reithmann C, Thomschke A, Werdan K (1987) The role of endogenous noradrenaline in beta-blocker withdrawal phenomenonstudies with cultured heart cells. Klin Wochenschr 65 : 308 - 316 Schmitz W, B6hm M, Brfickner R, Burmann H, Meyer W, Neumann J, Nose M, Scholz H (1985) Relationship between the effects of cAMP-increasing agents and adenosine in the mammalian heart. In: Stefanovich V, Rudolphi K, Schubert P (eds) Adenosine: Receptors and modulation of cell function. IRL Press, Oxford, pp 295-305 Schmitz W, Heller T, K6cher M, Neumann J, Scholz H, Scholz J, Stemmildt V, St6rtzel K (1988) On the mechanism of action of adenosine on myocardial contractility. In: Paton DM (ed) Adenosine and adenine nucleotides: Physiology and pharmacology. Taylor and Francis, London Philadelphia New York, pp I67--176 Scholz H, B6hm H, Brfickner R, Neumann J, Schmitz W (1987) Mechanism of the "antiadrenergic" effects of adenosine on myocardial force of contraction. In: Gerlach E, Becker BF (eds) Topics and perspectives in adenosine research. Springer, Berlin Heidelberg New York Tokyo, pp 369-382 Schrader J, Baumann G, Gerlach E (1977) Adenosine as inhibitor of myocardial effects of catecholamines. Pfliigers Arch 372: 2 9 - 35

Schfitz W, Freissmuth M, Hausleithner V, Tuisl E (1986) Cardiac sarcolemmal purity is essential for the verification of adenylate cyclase inhibition via Al-adenosine receptors. NaunynSchmiedeberg's Arch Pharmacol 333:156-162 Stein B, Neumann J, Schmitz W, Scholz H (1988) Evidence for cAMP independent effects of adenosine derivatives in ventricular myocytes. Naunyn-Schmiedeberg's Arch Pharmacol 338 (Suppl) : R39 Struck CJ, Ahnert G, Glossmann H, Schaeg W (1977) Solid phase radioimmunoassayfor cyclic AMP using staphylococcal protein as antibody adsorbent. Naunyn-Schmiedeberg's Arch Pharmacol 298 : 6 7 - 73 Van Calker D, Muller N, Hamprecht B (1979) Adenosine regulates by two different types of receptors the accumulation of cyclic AMP in cultured brain cells. J Neurochem 33:999-1005 West GA, Isenberg G, Belardinelli L (1986) Antagonism of forskolin effects by adenosine in isolated hearts and ventricular myocytes. Am J Physiot 250 : H 769 - H 777

Received July 10, 1989/Accepted August 18, 1989