Extracellular calcium-dependent and -independent effects of adenosine in bovine coronary artery

MUDUMBI Vo RAMAGBPAE~ AND S. JAMAL MUSTAFA~ Department of Phan~acokogy, School of Medicine, East Caas/iraez kltaiversity, Grt?elk~il&?, hTC 2 7858-4354, Ua S. A.

Received July 18, 1989

RAMAGOPAL, M. V., and MUSTAFA, S. J. 1996. Extracellular calcium-dependent and -independent effects of adenosine in bovine coronary artery. Can. J. Physiol. Pharmacol. 68: 608 -613.

Adenosine relaxes the coronary arteries of various species through A, receptors. The aim of the present investigation was to evaluate the relaxing effects of adenosine in relation to the role of calcium in bovine coronary arteries by studying the vasdilatory effect s f adenosine in normal and calcium-free medium and on calcium-45 efflux into calcium-free medium. Acetylcholine (ACh) and norepinephrine (NE) were used to induce tone in coronary artery rings. Adenosine, 5'-(N- ethylcarboxamido)adenosine (NECA), and N6-(L-phenylisopropy1)adenosine (L-PHA) produced concentration-dependent relaxations of the coronary artery rings. Both in normal and calcium-free medium, the order of potency for adenosine analogs (NEGA > L-PIA > adenosine) was similar and 8-phenyltheophylline antagonized the relaxation responses to adenosine and its analogs. Removal of extracellular calcium shifted the concentration-response curves to the right in a parallel fashion, slowed the rate of relaxation, and in NE contracted rings reduced the maximum responses for adenosine and its analogs. In calcium-free medium, adenosine was without an effect on calcium-45 efflux in the presence of ACh. However, adenosine inhibited the stimulated calcium-45 efflux induced by NE. The data suggest that the vasodilatory action of adenosine in bovine coronary smooth muscle has both extracellular calcium-dependent and -independent components.

Key words: adenosine receptors, calcium, coronary circulation, vascular smooth muscle, acetylcholine, norepinephrine.

RAMAGOPAL, M. V., et MUSTAFA, S. J. 1990. Extracellular calcium-dependent and -independent effects of adenosine in bovine coronary artery. Can. J. Physiol. Pharmacol. 68 : 608 - 613.

L'adCnosine relaxe les artkres coronaires de diverses espkces par I'intermCdiaire des rCcepteurs A'. Le but de la prCsemte Cmde a kt6 d'Cvaluer les effets relaxants de l'addnosine par rapport au r61e du calcium b n s les artkes coronaires de bovins, en examinant l'effet vasodilatateur de lTadCnosine dans un milieu normal et b n s un milieu sans calcium, et SLIT l'efflux de "caIciurn dans un milieu sans calcium. L'acCtylcholine (ACh) et la norkpinkphrine (NE) ont CtC utilisCes pour induire une contraction dans les anneaux des artbres coronaires. L'adCnosine, la 5'-(N-Cthylcarboxa~~lide)adCnosirae (NEGA) et la N'-(L- phtny1isopropyI)adCnosine (L-PIA) ont provoquk des relaxations concentrations dCpendlantes des anneaux des artkres coro- naires. Dans le milieu sans calcium et dans le milieu normal, I'srdre de puissance des analogues de I'adCnosine (NECA > L-PIA > aclknosine) a CtC similaire et la 8-phCnyl~Cophylline a antagonis6 les rkponses de relaxation i? B'adCnosine et h ses analogues. L'extraction de calcium extracellulaire a dCpBacC les courbes de concentration -rCponse vers la droite de manikre parallkle, a ralemti Be taux de relaxation et, dans les anneaux contractCes par la NE, a rCduit les rkponses maimales lqadCno- sine et A ses analogues. Dans un milieu sans calcium, l'adCnosine a CtC sans effet sur l'effltax de 45calcium en prksence d'AGh. Toutefois, 19adCnosine a inhibC l'efflux de "calcium stimulk, induit par la NE. k s rksultats suggkrent que l'action vasdilatatrice de 19adCnosine dans le muscle lisse coronaire bovin a des composantes indkpendantes et dCpendantes du cal- cium extracellulaire.

[Traduit par la revue]

Int reduction The present investigation was am attempt to study the

Adenosine has been proposed as a mediator in the regulation mechanism of relaxing effects of adenosine and its analogs in

of coronary blood flow (Berne 1988). Adenosine causes relax- relation to the role of calcium in coromry arteries with special

ation of canine O<usachi et 1983), bovine (Mustafa and regards to an action on intracellular calcium utilization.

A s h r 1985), and human (Ramagopal et al. 1988) coronary arteries in vgrdtro through activation s f A2 subtype adenosine receptors. The mechanism(s) by which adenosine induces relaxation of the coronary artery is not well understood. It has been suggested that the relaxing effect of adenosine might involve inhibition of cdcium influx (Fentom et al. 1982; Ramagopal and Mustafa 1988). However, in calcium-free medium, adenosine inhibited norepinephrine-induced contrac- tions in dog saphenous vein (Verhaeghe 1977) and rabbit femoral artery (Young and Merrill 1983). indicating the involvement of am extracellular calcium-independent comp- nemt, i.e., lowering of the intracellular calcium levels, for the vasodilatony action of adenosine.

'Present address: Division of Pharmacology, Glaxo Research Laboratories, Five Moore Drive, Research Triangle Park, NC, U.S.A.

'Author for correspondence and reprint requests.

Materials and methods Tissue prepamtion

Bovine hearts were obtained from local slaughter houses within 20 min after the death of the animal and transported to the laboratory in ice-cold oxygenated physiological salt solution (PSS) with the fol- lowing composition (t9nha): NaC1, 119; MGl, 4.7; KH2P04, 1.2; MgSO,, 1 2; NaHGO,, 14.9; GaGl,, 1.6; gllucose, 5.5; sucrose, 50; and disodium EDTA, 6.026. Branches of left anterior descending coronary artery (approximately 2 mm o.d.) were carefully dissected, cleaned of fat and eonmective tissue, and cut into 2 - 3 mm wide rings.

Recording of rrausc'e tension The coronary artery rings were mounted via silk sutures and stain-

less steel wires in 10-mL organ baths containing oxygenated (95% 0, and 5% GO,) PSS at 37°C (pH 7.4) and equilibrated for 1 h. Constant reproducible contractions were obtained with potassium chloride (KC1, 50 mM). The csntraetion cycles were separated by 30-min intervals during which time the buffer was changed every 15 min. Later, the coronary artery rings were contracted either with

hinted in Canada / Imprim6 au Canada

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

YO

RK

UN

IV o

n 11

/10/

14Fo

r pe

rson

al u

se o

nly.

RAMAGOPAL AND MUSTAFA 609

acetylcholine (ACh, 1 x 1W6 M) or norepinephrine (NE, 1 x 10-' M) plus gropranolol (1 X 1 0 - w ) . Cumulative concentra- tion -response curves for adenosine, 5 '-(N-ethy lcarboxamido)adeno- sine (NECA), and N6-(L-phenylisopropyI)adenosine (L-BIA) were obtained in the presence and absence of 8-phenyltheophylline (3 x M). After washing the rings with normal PSS, repro- ducible contractions were once again obtained with 50 ndvf KCI. Later, n o m l PSS was replaced with calcium-free PSS. In calcium- free PSS, calcium was omitted from the solution and 2 mM ethylene glycol bis((3-aminoethyl ether)-N,N'-tetraacetic acid (EGTA) was added to remove the superficially and loosely bound (extra- and intra- cellular) calcium stores (Meisheri et al. 1980). After 1 h incubation in calcium-free PSS, the rings were contracted either with ACh or NE (plus propranolol) and cumulative concentration - response curves were obtained once again for adenosine, NECA, and L-PIA in the absence and presence of 8-phenyltheophylline. In both the presence and absence of endothelium, ACh and NE induced similar contrac- tions in bovine coronary arteries. Moreover, NECA also induced similar relaxations (Sabouni et al. 1989), indicating that the adeno- sine-induced relaxation is not dependent on the presence of an intact endothelium, thus confirming the report of Kennedy et al. (1985). The antagonist, 8-phenyltheophylline, was allowed to equilibrate with the coronary artery rings for 30 min. 8-Phenyltheophylline (3 x BW6 M) by itself had no effect on ACh- and NE-induced contrac- tions in calcium and calcium-free PSS.

The changes in tension were measured using a force displacement transducer (Grass FT.03) and recorded on a Beckman Dynograph (model R611). Percent relaxation was calculated by taking the magni- tude of the contraction of the contracting agent (NE or ACh) as 100%. EC,, values for adenosine and its analogs were calculated from individual concentration -response curves by linear regression analysis.

Calcim-45 e$m The calcium-45 efflux was measured by a modification ~f the

method of van Breemen et al. (1982). The coronary rings were mounted on a stainless steel wire and incubated in oxygenated cal- cium PSS at 37°C for l h. The rings were hrther incubated with 50 EM KC1 for 5 - 10 min and in normal PSS for 20-25 nin. These steps were repeated three to four times to equilibrate and simulate the conditions of the isometric tension studies. Later, the rings were loaded with calcium-45 (2 pCi/mL) (1 Ci = 37 GBq) for 2 h at 37°C in calcium PSS (containing 1.6 PnM nonradioactive calcium). After loading, the rings were briefly rinsed (about 2 s) and transferred through a series of vials containing oxygenated calcium-free (contain- ing 2 mM EGTA) PSS at 37°C. In the initial 30 min of the experi- ment, the rings were incubated for 18 f i n in each vial before transfer to the next vial. For the remainder (90 min) of the experiment, the rings were incubated for 5 min in each vial before transfer to the next vial. W e r e indicated, calcium-free PSS was replaced with calcium- free PSS containing either ACh (1 x M) or ME (1 X M) plus propranslol (1 x M) in the presence or absence of adenosine.

At the end of the experiment, the rings were blotted, weighed, and dried overnight at 85°C. Later, the tissues were wet asked in 30% H202 (85°C for 6 h) and dissolved in calcium-free PSS. The radio- activity in the vials and tissue aliquots were counted with a Bechan scintillation counter after adding 2 mL of scintillation cocktail (Ready Sslv, Bechan). Fractional release of calcium45 during the wash- out perid, which expresses the decline sf the tissue calcium-45 con- tent as percentage of the calcium45 present in the tissue during each time interval, was calculated. The data are plotted as fractional loss per minute (Casteels et al. 1981).

Dmgs The following drugs were used: NECA (Abhtt Laboratories,

Chicago, IL); adenosine, ACh, NE, D,L-prspranolol, and EGTA (Sigma, St. Louis, MO); L-PIA and adenosine deaminase (Boehringer Mannheim, Indianapolis, IN); 8-phenyltheophylline (CaHbiochem, San Diego, CA) ; 45CaC12 (specific activity, 24.6 1 mCi/

mg; NEN Research Products, Boston, MA). Adenosine, NE, ACh, and propranolol were dissolved in distilled water. NECA, L-PIA, and 8-phenyltheophylline were dissolved in 50% ethanol. Ethanol at the concentrations used (maximum 50 pL in 10 mL bath) was without an effect on the preparation. NE was prepared with ascorbic acid (1 x 10sQ ME) to prevent the oxidation. Dmgs were prepared fresh daily and all concentrations are expressed as final molar (M) concentrations.

Statistical analvsis The data are expressed as means + SEM. The concentration-

response curves were transformed to a Bog - logit plot and their slopes calculated. Ts test the parallelism of the concentration-response curves, the slopes were statistically analyzed using analysis of vari- ance (ANOVA) and Newmn-Keul's (NK) multlple range test. When ody two groups were compared, the data were analyzed by Student's t-test for paired and unpaired observations, as appropriate. A p value of less than 0.05 was considered to be significantly different. Approximately 16 coronary a e r y rings were obtained from each bovine heart. At least 10 coronary artery rings obtained from five to six bovine hearts were used for each treatment.

Results The contractions produced by KC1 (50 mM), ACh (1 x

10-%), and NE (1 x M) in normal and calcium-free PSS are shown in Fig. 1 . The steady-state contractions induced by ACh and NE were significantly attenuated (82%, p < 0.081 and 48 % , p < 0.801, respectively) in calcium-free PSS. Contractions evoked by KC1 in calcium PSS were abolished by the removal of extracellular calcium.

Adenosine-induced retaxation Extrwcetlulwr caicim-dependent component In normal medium, adenosine, NECA, and E-PEA produced

concentration-dependent relaxations of the bovine coronary arteries when precontracted either with ACh (Fig. 2A) or NE (Fig. 3A). The concentration -response curves were parallel to each other ( p > 0.05; ANOVA, NK; Figs. 2A and 3A). In the presence of 8-phenyltheophylline (3 x M), the concentration -response curves for adenosine and its andogs were shifted to the right in parallel ( p > 0.05; ANOVA, NK) (Figs. 2A and 3A).

In normal medium, when coronary arteries were precon- tracted with ACh, using EC50 values (Table 1, column a), the potency ratio of NECA and L-PIA to adenosine was found to be 27 and 4, respectively. In the presence of NE (Table 2, column a), the potency ratio of NECA and E-PIA to adenosine was found to be $2 and 13, respectively.

Extracellular calcium-independent component In calcium-free medium, adenosine, NECA, and E-PIA

produced concentration-dependent relaxations when coronary artery rings were precontracted either with ACh (Fig. 2B) or NE (Fig. 3B). The concentration - response curves were parallel to each other ( p > 0.05; ANOVA, NK). Compared with the concentration-response curves obtained in normal medium (Figs. 2A and 3A), the concentration -response curves in calcium-free medium (Figs. 2B and 3B) were shifted to the right in parallel (g > 0.05; ANOVA, NK) . Moreover, the maximum relaxation responses of adenosine and its analogs in NE-contracted rings, but not in ACh-contracted rings, were significantly reduced (g < 0.05) in calcium-free medium compared with the responses obtained in normal medium. Both in ACh-contracted and NE-contracted rings, in the presence of 8-phenyltheophylline (3 x 18-"), the

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

YO

RK

UN

IV o

n 11

/10/

14Fo

r pe

rson

al u

se o

nly.

610 CAN. J. BHYSHBL. PHAWMACBL. VOL. 68, 1990

Calcium Calcium-Free

FIG. 1. Comparison of the contractions induced by KC1 (50 mM), ACh (1 x 10-"1, and NE ( I x in the presence of propranolol 1 x M) in bovine coronary artery rings in normal and calcium-free PSS. * g < 0.05 from respective controls. Values represent means f SEM of 10 artery rings from 6 to 10 bovine hearts.

concentration - response curves for adenosine, NECA, and L-PIA were shifted to the right. The rightward shift of the curves (i .e., the slopes) approached the significant level of parallelism.

In cdcium-free medium, using EC50 vdues, the potency ratio of NECA and L-PIA to adenosine was found to be 50 and 1 1, respectively, with ACh-contracted rings (Table 1, column c) and 29 and 4, respectively, with NE-contracted rings (Table 2, column c).

The mean ECSo values obtained with adenosine, NECA, and L-PHA both in normal and cdcium-free medium are shown in Tables 1 and 2 with ACh- and NE-contracted rings, respec- tively. In the absence of extracellular calcium, the control ECSO vdues for adenosine and its andogs were significantly higher ( p < 8.05) as compared with those obtained in norrnd medium (column c as compared with a in Tables 1 and 2). En ACh-contracted rings, the EC50 ratios (@/a) for adenosine and its analogs were approximately 2. In NE-contracted rings, the ECS0 ratios (@/a) were approximately 5 .

and cdcium-free medium, addition of adenosine d e h n a s e (0.4 p/mL) reversed the relaxations induced by adenosine, but not those of NEGA and L-PIA, when coronary artery rings were precontracted either with ACh or NE (data not shown).

Calcium-4.5 eflux The effect of adenosine (1 x M) on calcium-45 efflux

in the presence of NE is shown in Fig. 4. In calcium-free medium, the slow intracellular component of cdciurn-45 efflux emerged around 30 min and continued for the remainder of the experiment. The slow component of the calcium-45 efflux curve has been attributed to the release of calcium from tightly bound extracellular calcium and from intracellular storage sites (Aaromon et a%. 1979). Agonist induced intra- cellular cdcium-45 release can be detected as increased slow component of the cdciurn-45 efflux curve in calcium-free medium (van Breemen and Siege1 1980). The effect of ACh and NE (in the absence or presence of adenosine) was tested on the slow (intracellular) component of the calcium efflux

A. Calcium Medium

D r w Concentration (lirB)

B. Calcium-Free Medim

Drug Concentration (MI

FIG. 2. Concentration - response curves for adenosine, NECA, and L-BIA in bovine coronary arteries in the presence and absence of 8-phenyltheophylline (3 X 10s6 M). Coronary artery rings were grecontracted with ACh (1 x M) in normal (A) and calcium- free (B) medium. Each point represents a mean of 10 artery rings from five to six bovine hearts $ %EM.

curve. Addition of NE (I x l W 4 M, in the presence of 1 x 1W4 M propranolol) caused a significant increase ( p < 0.0%; unpaired t-test) in the rate of cdcium-45 efflux (Fig. 4) around 60 and 80 min. Adenosine (1 x M) significantly inhibited ( p < 0.0%; unpaired t-test) the NE-induced transient increases in calcium efflux (Fig. 4). However, the efflux of cdcium-45 in the control rings was not affected by ACh (I X 10-$ M) in the presence or absence of adenosine (1 x

M; data not shown). Addition of adenosine after expos- ing the tissues to NE significantly inhibited NE-induced second transient increase in calcium efflux (data not shown).

Discussion The present study shows that both in the presence and

absence of extracellular calcium, adenosine and its analogs induced relaxations with the same order of potency (NECA > L-PIA > adenosine). Both in normal and calcium-free medium, 8-phenyltheophylline antagonized the relaxation responses of adenosine and its andogs. However, in calcium- free medium, the concentration-response curves for adeno- sine and its analogs were shifted to the right in parallel and adenosine inhibited the increase in rate of cdciurn-4% efflux induced by NE.

En the absence of extracellular calcium, KC1 was unable to induce any contraction, indicating that KCl-induced contrac- tions were due to the potential dependent calcium entry from

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

YO

RK

UN

IV o

n 11

/10/

14Fo

r pe

rson

al u

se o

nly.

A. Calcium Medium c' Adenosine)

I 1 I I

I 0-8 r oi7 t o " ~ I o-5 I c3 Drug Concentration (MI

B. Calcium-Free Medium

roo I I I I 1

t b a r om7 1 0 ' ~ t 0'5 I (r4 t om3

Drug Concentratton (MB

FIG. 3. Concentration - response curves for adenosine, NECA, and L-PHA in bovine coronary arteries in the presence and absence of $-gheraylthesphylline (3 x M). Coronary artery rings were precontracted with NE (1 X 10-%) in n s m l (A) and calcium- free (B) medium. Each p i n t represents a mean of 10 artery rings from five to six bovine hearts f SEM.

superficial and extracellular sources of calcium. In the absence of extracellular calcium, the contractions induced by ACh were significantly attenuated. ACh-induced contractions were biphasic in nature (data not shown), i.e., there was a rapid increase in force that declined with time to a lower steady-state level. Ratz and Flaim (198%) reported similar findings in bovine coronary arteries md suggested that ACh-induced con- traction is largely through the sustained activation of receptor- operated calcium channels.

NE in the presence of propranolol, a P-adrenoceptor blocker, produced a sustained contraction in the calcium- containing medium that was attenuated in cdcium-free medium. Ratz and Flaim (1984) reported that in bovine coro- nary arteries, NE during @-blockade produced a very we& tonic contraction in dcim-containing medium. In our study, to obtain a significant contraction with NE, higher concentra- tion of propranolol was used to block P-adrenoceptors, The significant contractions obtained with ACh md NE in calcium- free medium in the present study indicate that in the absence of extracellular calcium, ACh and NE induced calcium release into the cytosol from membrane-bound and intracellu- lar storage sites, which were not affected by the presence of 2 mM EGTA.

Addition of adenosine deaminase reversed the relaxing effects of adenosine both in normal and calcium-free medium (data not shown) suggesting the involvement s f extracellular

TABLE 1 . EC,, ( X los4 M) values for relaxations of adenosine a d its analogs in the presence and absence of 8-phenyltheophylline (8-PT, 3 X los%) in bovine coronary arteries precontracted with

acetylcholine

Cdcium medium Calcium-free medium

Control 8-PT Control 8-PT (4 (b) (4 (4

Adenosine 4.79f 1.2 22.6f 4.9t 8.92f 1.87" 48.80f 5.Wt NECA 0.18f0.82 1.16f 0 . l t 0.31 f 8.06" 5.38f 0.53t L-PIA 1.13f 0.23 4.91 f 0.86t 2.38f 0.42' 8.38f 0.81P

NOTE: Vdues are means f SEM of 18 coronary artery rings from five to six bovine hearts.

*Significantly different from the values in calcium medium ( p < 0.05, c from a). tsignificmtly different from the control values (p < 0.05, b from a, and d from c).

TABLE 2. ECgO ( X M) values for relaxations of adenosine and its andogs in the presence and absence of 8-phenyltheophyS1ine (8-PT, 3 X M) in bovine coronary arteries precontracted with

norepinephrine -

Calcium medium Calcium-free medium

Control 8-PT Control 8-PT (4 (4 (4 (4

Adenosine 5.54f0.94 30.10f 5-20? 25.60f 2.40' 72.88f 6.9t NECA 0.09f0.02 1.06f 0.15t 0.51 f 0.07' 2.58+0.16? E-PIQL 0.43f 0.08 2-71 f 0.40t 2.30f 0.53" 4.05 f0.6t

NOTE: Vdues are means f SEM of 10 coronary artery rings from five t~ six bovine hearts.

*Significibndy different from the d u e s in calcium medium ( p < 0.05, c from a). ?Significantly different from the cont~ol values ( p < 8.05, b from a, a d d from c).

adenosine receptors in the vasodilatory action of adenosine. Moreover, NECA was more potent than L-PIA in relaxing the bovine coronary artery rings both in the presence and absence of calcium. The A2 adenosine receptors show a preference for the 5' substituted analog, NECA over the N6-substituted ana- log, L-PIA (Londos et al. 1980). The relative order of potency for adenosine and its malogs (NECA > L-PIA > adenosine) both in normal and calcium-free medigam (either in ACh- contracted or NE-contracted coronary artery rings) in the present study were similar to the earlier reports in canine (Kusachi et al. H983), bovine (Mustafa and Askar 1985) and human (Ramagopal et alel. 1988) coronary arteries. This sug- gests that adenosine, whether in the presence or absence of cdcium, induced relaxation through A2 subtype adenosine receptors in bovine coronary arteries. In this regard, it has been reported earlier from this laboratory (Dutta and Mustafa 1988) that calcium is not required for the binding of adenosine to coronary smooth muscle membranes.

8-Phenyltheophylline was able to antagonize the relaxing effects of adenosine, NECA, and L-PIA and shifted the concentration -response curves to the right both in normal and calcium-free medium. Moreover, maximal responses s f adenosine and its analogs were not significantly altered by 8-phenyltheophylline. The shift of the concentration - response curves for adenosine and its andogs to the right (the rightward shift approached the significant level of parallelism) in the presence of 8-phenyltheophylline indicated that both in calcium and calcium-free medium adenosine and its malogs were competing for the same receptor site.

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

YO

RK

UN

IV o

n 11

/10/

14Fo

r pe

rson

al u

se o

nly.

CAN. J . PHYSIOE. PHARMACBL. VOL. 68, 1990

o- Control

~ . . . N E ( 1 x ICPMI Q.25 e N E $ A % ) ( ~ x I Q - ~ M ?

Norepinephrine = m m r m m r m ~ m - e m m a c z a w a

0.2 F T

Minutes

FIG. 4. Effect of NE (1 x M) in the absence and presence sf adenosine (1 x 1W4 M) on the rate sf efflux of calcium-45 from coronary arteries into calcium-free medium. * p < 0.05 from control to NE stimulated condition; .f. p < 0.05 between NE stimulated and treatment with adenosine. Each point represents the mean s f at least six vascdar rings from four to five bovine hearts f SEM. The calcium45 efflux rate curves of treatment (between 30 and 68 rnin), which were not different significantly from the control, have been omiaed for the sake s f clarity.

In the present study, the relaxation responses induced by adenosine and its andogs increased when the coronary artery rings were contracted with low concentrations of ACh or NE in normd PSS (dab now shown). Merlihy (1978) reported similar results in hog carotid artery, i.e., the degree of adenosine-induced relaxation depends on the concentration and type s f agent used to contract the vessels, noting that the higher the contraction produced by NE the lesser is the relaxa- tion with adenosine. However, in calcium-free medium, the contractions induced by ACh and NE were reduced, and at the same time the amount of relaxation induced by adenosine and its analogs was also reduced. This indicated that the differ- ences in the effects of adenosine and its andogs in the absence of cdcium were not due to differences in precontraction amplitudes obtained with ACh and NE.

Relaxations of the bovine coronary arteries induced by adenosine and its analogs in the absence of calcium suggested the involvement of an extracellular calcium-independent mechanism. The control ECSo ratios (c/a in Tables 1 and 2) indicated that the responses of adenosine and its analogs were inhibited to a similar magnitude in the absence of extracellular cdcium. Moreover, it has been observed during the cumula- tive addition of adenosine and its analogs that in calcium-free medium, the rate of relaxation of adenosine and its analogs was slower than that obtained in normd medium. The parallel shift of the concentration -response curves to the right and the reduction of maximum responses (in case of NE-contracted rings) in the absence of calcium further suggests the involve- ment of extracellular cdciumdependent (i.e., by inhibition of cdeium influx) and -independent (i . e . , by affecting intracellu-

lar cdcium levels) mechanisms in adenosine-induced relaxa- tion. Verhaeghe (19'97) noted that adenosine inhibited the contractions of canine saphenous vein induced by NE in calcium-free solution and in the presence of verapamil and suggested that adenosine might be interfering with the chain of events beyond calcium influx, including its effects on contrac- tile proteins. Young and Merrill (1983) dso reported that in rabbit femoral artery, adenosine relaxes NE-induced contrac- tions in calcium-free medium, suggesting that a part of adenosine-induced relaxation may be mediated intracellularly. Bradley and Morgan (1985), by studying the effects of adeno- sine on the aequorin signal in ferret portal vein, reported that adenosine at lower doses decreased cytosolic ionized calcium and at higher concentrations desensitized the smooth muscle myofilaments to cytosolic ionized calcium. Furthermore? Kai et aal. (198'9) reported that in cultured rat aortic smooth muscle cells, adenosine decreased cytosolic free calcium concentra- tion both in calcium and calcium-free medium. The results of the present study confirm these findings and suggest that adenosine-induced relaxation is mediated in part by an extra- cellular calcium-independent mechanism.

There was no change in the efflux of calcium-45 in cdciurn- free medium when bovine coronary artery rings were incubated with ACh either in the absence or presence of adenosine. It has been reported that in porcine coronary arter- ies, prostaglandin produced a tonic contraction without detectable change in calcium signals with aequorin in cdcium- free medium (Bradley and Morgan 198'7). Furthermore, Rooke et d . (1984) reported that in canine coronary arteries, prostaglandin F2,-induced contractions in calcium-free medium might be due to undetectable release of intracellular calcium or might occur independently of changes in cytoplas- mic concentration of calcium. Recently, Heaslip and Sickels (1989) reported that in the rat aorta, prostaglandin induced contractions in calcium-free medium might be due to direct activation of protein Enase C. In light of these reports, it is tempting to suggest that a similar mechanism might be involved in ACh-induced contractions in calcium-free medium in bovine coronary arteries.

NE significantly increased the calcium45 efflux into the cdcium-free medium. Liejten and van Breemen (1986) reported ha t in rat mesenteric artery, when NE was present continuously in calcium-free medium, both force development and calcium45 efflux stimulation consisted of a fast and a slow component, and these components were sensitive to caffeine and are probably related to calcium release from the intracellular cdcium store. In the present study, in the absence of calcium, NE produced a sustained contraction and two tran- sient increases in cdcim-45 efflux around 60 and 80 min. The second transient increase in cdcium-45 efflux could be due to the sustained contraction observed with NE or the release of calcium from a second intracellular storage site. Moreover, in calcium-free medium, compared with the inhibi- tion of ACh-induced contractions (82 % from normal medium), the inhibition of NE-induced contractions (48%

medium) was less. This indicated that in bovine coronary arteries, NE-indued contractions are most probably due to the release of calcium into the eytosol from intracellular storage sites.

Furthermore, in bovine coronary arteries adenosine has been shown to activate particulate adewylate cyclase (Kukovetz et d. 1978) and CAMP-dependent protein kinase (Silver et aH. 1984), suggesting the involvement of CAMP during adenosine-

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

YO

RK

UN

IV o

n 11

/10/

14Fo

r pe

rson

al u

se o

nly.

RAMAGOPAL AND MUSTAFA 613

induced relaxations. It has also been demonstrated that in vascular smooth muscle, activation of @-adrenoceptors by iso- proterenol increases CAMP levels (Htoh et id. 1982), which in turn leads to the increased sequestration of calcium into the intraeellular storage sites (Itoh et al. 1982; Van Eldere et al. 1982). Inhibition by adenosine of the stimulated calcium-45 efflux induced by NE may suggest that adenosine-induced relaxation in calcium-free medium could be due to the inhibi- tion of calcium release from intracellular storage site and (or) m increased reuptake sf free calcium into the intraceuular storage sites. From this laboratory (Dutta et d. 1984), it has been reported that in canine coronary arteries, adenosine inhibited the increased rate of calcium-45 efflux caused by high K+ in the normal medium. The results of the present study and Dutta et al. (1984) suggest the involvement of an extracellular calcium-independent component in the vasodila- tory action of adenosine.

In conclusion, the results of the present study indicate that adenosine-induced relaxation in bovine coronary arteries precontracted either with ACh or NE depend in part on extra- cellular calcium. However, an extracellular cdcium-indepen- dent mechanism still plays a significant role in the vasodilatory action of adenosine.

This work was supported by a grant from the National Insti- tutes of Health (HL 27339).

AARONSON, P., VAN BREEMEN, C., LOUTZBNHHSER, R., and KOLBER, M. A. 1979. A new method of measuring the kinetics of transmem- brane calcium-45 efflux in the smooth muscle of the guinea pig tae- nia coli. Life Sci. 25: 1781-1790.

BERNE, R. M. 1980. The role of adenosine in the regulation of coro- nary blood flow. Circ. Res. 47: $87-813.

BRADLEY, A. B., and MORGAN, K. G. 1985. Cellular Ca2' moni- tored by aequorin in adenosine mediated smooth muscle relaxation. Am. J. Physiol. 248: H189-H117.

1987. Alterations in cytoplasmic calcium sensitivity during porcine coronary artery contractions as detected by aequorin. J. Physiol. (London), 385: 437 -448.

CASTEEW, W., RAEYMAEKERS, L., SUZUKH, H., and VAN ELDEWE, J. 1981. Tension response and 4"a'9 reaease in vascular smooth muscle incubated in Ca-free colution. Pfluegers Arch. 392: 139- 145.

DWTA, P-, and MUSAFA, S. J. 1988. Binding of adenosine to the crude plasma membrane fraction isolated from dog coronary and carotid arteries. J. Phamcsl . Exp. Ther. 214: 496-502.

DUTTA, Po, JONES, A. We, and MUSTAFA, S. J. 1984, Uptake and efflux of calcium by canine coronary arteries and the action of adenosine. Basic Res. Cardiol. 79: 5 19 - 530.

P E N ~ N , R. A., BRUTTHG, S. P., RUBHO, R., and BEWNE, R. M* 1982. Effect of adenosine on cdcium uptake by intact and cultured vascu- lar smooth muscle. Am. J. PhysioH. 242: H797-H8W.

HEASLHP, R. J., and SICKEU, B. 0. 1989. Evidence that prostaglandins can contract the rat aorta via a novel protein b a s e C-dependent mechanism. J. Phamacol. Exp. Ther. 258: 44 -51.

WERLHHY, J. T. 1979. Action of adenosine on isolated vascular smooth muscle. Bn Physiological and regulatory function of adenosine and adenine nucleotides. Edited by H. P. Baer and G. I. Dmmond. Raven Press, New York. pp. 95 - 102.

I ~ H , T., IZUMI, H., and KURIYAMA, H. 1982. Mechanisms of relam- tion induced by activation of beta-adrenoceptors in smooth muscle cells of the guinea-pig mesenteric artery. J. Physiol. (London), 326: 475 -493.

KAI, H., KANAHDE, H., MAT SUM^, T., SHOGAKIUCRI, T., and N A K A ~ A , M. 1987. Adenosine decreases intracellular free cd-

cium concentration in cultured vascular smooth muscle cells from rat aorta. FEBS Lett. 212: 119- 822.

KENNEDY. C., DELBRO, D., and BURNSTOCK, G. 1985. P2-hrinocep- tors mediate both vasodilation (via the endothelium) and vasscon- striction of isolated rat femoral artery. Eur. J. Pharmacol. 107: 161 - 168.

KUKOVETZ, Wa R., PBcH, G. , HOLZMANN, S., WURM, A., and RNNER, I. 1978. Role sf cyclic nucleotides in adenosine-mediated regulation of coronary flow. Adv. Cyclic Nucleotide Res. 9: 397 - 489.

KUSACHI, S., THOMPSON, R. D., and OMSON, R. A. 1983. kigand selectivity of dog coronary adenosine receptor resembles that of adeqlate cyclase stimulatory (Ra) receptors. J . Pharmacol. Exp. Ther. 227: 316-321.

LEIJTEN, Pa A. A., and VAN BREEMEN, C. 1986. The relationship between noradrenaline-induced contraction and 45Ca efflux stimu- lation in rabbit mesenteric artery. Br. J. Phmacol. 89: 739-747.

LONWS, C., COOPER, D. M. F., and WLFF, J. 1980. Subclasses of adenosine receptors. Proc. Natl. Acad. Sci. U.S.A. 77: 2551 - 2554.

MEHSHERI, K. D. , PALMER, R. F., and VAN BREEMEN, C. 1980. The effects of amrinone on contractility, Ca2' uptake and CAMP in smooth n~uscle. Eur. J. PhamacoB. 61: 159 - 165.

MUSTAFA, S. J., and ASKAR, A. 0. 1985. Evidence suggesting an Ra-type adenosine receptor in bovine coronary arteries. J. Pbmacol. Exp. Ther. 232: 49 -56.

RAMAGOPAL, M. Q., and MUSTAFA, S. J. 1988. Effect of adenosine and its andogs on calcium influx in coromq artery. Am. J. Physiol. 255: HI492 -H1498.

RAMAGOPAL, EM. BI., CHITWOOD, R. W., Ja., and MUSTAFA, S . J. 1988. Evidence for an A, adenosine receptor in human coronary arteries. Eur. J. Phamacol. 151: 483-486.

WATZ, P. H. , and FLAIM, S. 1984. Mechanism of 5-HT contraction in isolated bovine ventricular coronary arteries: evidence for tran- sient receptor-operated calcium influx channels. Circ. Res. 54: 135- 143.

1985. Acetylcholine- and 5-hydroxytqphmine-stimulated contraction and calcium uptake in bovine coronary arteries: evi- dence for two populations of receptor-operated calcium channels. J. Phamacol. Exp. Ther. 234: 641-647.

~ O K E , T. We, RIMELE, T. J., and QANHOUTTE, P. M. 1984. Contrac- tions of the canine coronary artery in calcium-free solution. Am. J. Physiol. 247: M259 - H263.

SABOUNI, M. H., RAMAGOPAL, M. V., and MUSTAFA, S. 1. 1989. Roles of calcium and the endothelium in the relaxations produced by 5 "N-ethylcarboxamislo-adenosine (NECA) . Eur. J. Phamacol. 166: 311-314.

SILVER, P. J., WALWS, K., and DHSALVO, J. 1984. Ademsine-mediated relaxation and activation of cyclic AMP-dependent kinase in csro- nary arterial smooth muscle. J. Pharmacol. Exp. Ther. 228: 342 - 347.

VAN BWEEMEN, C., and SIEGEL, B. 1988. The mechanisms of alpha- adrenergic activation of the dog coronary artery. Circ. Res. &: 426 -429.

VAN BEEMEN, C., AARONSSON, I., CAUVIN, C. A., LOUTZEN- HHSER, R., MANGEL, A. W., and SAHBA, K. 1982. The calcium cycle in arterial smooth muscle. In Calcium blockers: mechanisms of action and clinical applications. Edited by S. F. Flaim and R. Zelis. Urban and Schwarzenberg, Bdtimore, MD. pp. 53-63.

VAN ELDEWE, J., ~ABYMAEKEWS, L., and CAsTaaU, R. 1982. Effect of isoprenaline on intracellular Ca2+ uptake and on Ca2' influx in arterial smooth muscle. Pfluegers Arch. 395: 81 - $3.

VERHAEGHB, R. 1977. Action of adenosine and adenine nucleotides on dog's isolated veins. Am. J. Physiol. 233: H1 14-H121.

YOUNG, M. A., and MEWWILL, G. E 1983. Comparative effects of adenosine and nifedipine in rabbit vascular smooth muscle. Can. J. Physiol. Phamacol. 61: 1057 - 1062.

Can

. J. P

hysi

ol. P

harm

acol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

YO

RK

UN

IV o

n 11

/10/

14Fo

r pe

rson

al u

se o

nly.