Heterogenous effects of histamine on isolated rat coronary arteries

ejp ELSEVIER

European Journal of Pharmacology271 (1994) 17-23

Heterogenous effects of histamine on isolated rat coronary arteries Johan Van de Voorde *, Veerle Brochez, Bert Vanheel Department of Physiology and Physiopathology, University of Gent, De Pintelaan 185, B-9000 Gent, Belgium

Received 28 April 1994; revised MS received 22 August 1994; accepted 26 August 1994

Abstract

The influence of increasing concentrations of histamine (0.1/xM-1 mM) was studied on proximal and distal ring segments of left anterior descendens coronary arteries isolated from rats. Addition of histamine to prostaglandin F2~ (10/zM)-precontracted proximal segments elicited a further contraction. This effect was endothelium-independent and mediated by a histamine H 1 receptor mechanism since it was blocked by the histamine H 1 receptor antagonist, mepyramine (10 /zM), and not by the histamine H 2 receptor antagonist, cimetidine (100 /~M), and since it was mimicked by the histamine H1 receptor agonist, 2-pyridylethylamine, and not by the histamine H 2 receptor agonist, dimaprit. Addition of histamine to prostaglandin F2~-precontracted distal segments elicited concentration-dependent relaxation. This relaxation is histamine H 2 receptor-mediated since it was blocked by cimetidine (100 tzM) and not by mepyramine (10/zM) and since dimaprit but not 2-pyridylethylamine elicited relaxation. The relaxation was not due to the release of endothelial NO, prostaglandins or activation of ATP-regulated K + channels since it was not inhibited by N°-nitro-L-arginine methyl ester (100 /zM) or NG-nitro-L-arginine (100 /zM), indomethacin (10/zM) or glibenclamide (10 /~M). Our results show that the effects of histamine on rat left anterior descendens coronary arteries are heterogenous, depending on the relative location within the coronary vasculature and possibly, the relative preponderance of histamine H 1 or H 2 receptors on the smooth muscle cells. Keywords: Coronary artery, rat; Histamine; Artery, small; Histamine receptor; Nitric oxide (NO)

1. Introduction

The heart contains considerable amounts of histamine, either newly synthesized or stored in mast cells (Yoshitomi et al., 1989). W h e t h e r histamine has a physiological role in cardiac function or coronary circulation remains unclear. However, it is well established that large quantities of histamine are released by immunological stimuli. Non-immunological stimuli (i.e. ischemia, surgery, anesthesia, drugs, etc.) have also been reported to induce histamine release from the heart. It is believed that the released histamine contributes to the arrhythmogenic and coronary vasoconstricting effects associated with different cardiovascular pathologies (Wolff and Levi, 1986). The prophylactic use of histamine receptor antagonists has therefore

* Corresponding author. Tel. + 32 92 40 33 42, fax + 32 92 40 33 90. 0014-2999/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0014-2999(94)00529-X

been suggested in high-risk patients, e.g. under anesthesia (Tryba et al., 1986). In this respect it is noteworthy that the amount of histamine is significantly higher in coronary vessels of cardiac patients than in those of non-cardiac subjects (Kalsner and Richards, 1984). Although systemic administration of histamine consistently elicits a decrease in peripheral vascular resistance and a decrease in systemic blood pressure, the effect of histamine on isolated blood vessels is much less consistent. Even on a single vessel type such as coronary arteries, both relaxation and constriction have been reported depending on species, dose of histamine, caliber and initial vessel tone. These effects are the resultant of multiple actions of histamine on both smooth muscle cells and endothelium (Leusen and Van de Voorde, 1988; Levi et al., 1991). This p a p e r describes the influence of histamine on proximal and distal segments of left anterior descendens coronary arteries isolated from rats and reports on experiments designed to determine the receptor type and the mechanisms involved in these effects.

18

J. Van de Voorde et al. / European Journal of Pharmacology 271 (1994) 17-23

2. Materials and methods

2.1. Dissection and mounting of the preparations The heart of a rat (male Wistar, 250-350 g) was excised after cervical dislocation and placed in ice-cold Krebs-Ringer bicarbonate solution (composition in mmol/l: NaCI 135; KC1 5; NaHCO 3 20; glucose 10; CaCI 2 2.5; MgSOa.7aq 1.3; KH2PO 4 1.2; EDTA 0.026). During dissection and mounting of the preparations, the Krebs-Ringer bicarbonate solution was continuously chilled and bubbled with 5% CO2 in 0 2. The left anterior descendens coronary artery and its branches were exposed by removing carefully the overlying cardiac tissue. Two stainless steel wires (40 ~m diameter) were guided into the lumen of 1-2 mm segments of proximal (above the first side-branch) or distal (below the first side-branch) parts of the artery. The ring segments were then isolated from the heart and mounted by fixing the wires on one side to a force-displacement transducer and on the other side to a micrometer in an automated dual myograph for small vessels (model 500 A, J.P. Trading, Aarhus, Denmark). This set-up allows direct determination of vessel wall tension while controlling the internal circumference of the vessels. After mounting, the vessels were equilibrated for 30 min in Krebs-Ringer bicarbonate solution bubbled with 95% 02-5% CO2 at 37°C. Then the preparations were normalized as described by Mulvany and Halpern (1977). In brief, the passive wall tension-internal circumference characteristics were determined. On the basis of this relationship, the circumference was set to a normalized internal circumference, Lo, corresponding to 90% of the internal circumference the vessels would have under a passive transmural pressure of 100 mm Hg, in order to obtain optimal conditions for active force development (Nyborg et al., 1987). The effective lumen diameter, lo, was calculated as Lo/~r. This entire procedure is automated, using a programmed microprocessor that controls the displacements of the motor-driven micrometers and performs the calculations. After normalization the preparation was allowed to equilibrate again for at least half an hour. The vessels were then contracted 3 times with Krebs-Ringer bicarbonate solution containing 120 mM of K +, prepared by adequate equimolar substitution of NaCI by KC1.

2. 2. Experimental protocols All concentration-relaxation curves were made with preparations precontracted with prostaglandin F2,~ (10 ~M). The response to aeetylcholine was assessed on every preparation to evaluate the presence of functional endothelial cells (Nyborg and Mikkelsen, 1990).

In experiments with mepyramine, cimetidine, N ~nitro-L-arginine methylester, NG-nitro-L-arginine, glibenclamide or indomethacin, the effect of cumulative concentrations of histamine was compared on the same preparation before and after addition of the antagonist to the experimental chamber. In the experiments with the histamine agonists, 2-pyridylethylamine and dimaprit, both agonists and histamine were tested on the same preparations in changing order. In some experiments the endothelium was removed by rubbing the intimal surface with a human hair. Endothelial removal was confirmed by the lack of relaxing influence of acetylcholine. Vessel responses are represented as active wall tension. This was calculated by dividing the active vessel wall force by twice the length of the vessel segment. Relaxations are expressed as percentages of the active wall tension (= tension developed under prostaglandin F2~ minus passive resting wall tension).

2.3. Drugs and statistics Acetylcholine chloride, cimetidine crystalline, glibenclamide, histamine dihydrochloride, indomethacin crystalline, mepyramine maleate, NG-nitro-L-arginine methylester hydrochloride, NC-nitro-L-arginine crystalline (Sigma, St. Louis, MO, USA); dimaprit dihydrochloride, 2-pyridylethylamine dihydrochloride (gift from Smith, Kline and French, Genval, Belgium); prostaglandin F2a (dinoprostum trometamolum; Dinolytic, Upjohn, Puurs, Belgium). All concentrations are expressed as final molar concentrations. Concentrationresponse curves were made by cumulative addition of a small volume (100 ~1) in the experimental chamber (10 ml). All solutions were freshly prepared from appropriate stock solutions. The results are expressed as means + S.E.M. Statistical significance was evaluated using Student's t-test for paired observations, n indicates the number of preparations tested.

3. Results

In this study experiments were performed on 23 proximal and 33 distal preparations with a mean normalized lumen diameter of respectively 437 + 10 and 278 + 9/~m. After the normalization procedure, adaptation to l o resulted in a passive resting wall tension (= passive wall force divided by twice the length of the vessel segment) of respectively 1.26 + 0.07 and 0.68 + 0.03 m N / m m . During stabilization of the preparations, a basal tension developed (0.89 + 0.16 and 0.51 + 0.07 mN/mm). In all experiments, concentration-response curves were made on preparations precontracted with prostaglandin F2,~ 10 ~M. In the presence

19

J. Van de Voorde et al. / European Journal of Pharmacology 271 (1994) 17-23

H I S T ~

ACh -9

-8 -?

-4

_

loo

l---.-i~,

.~_.,.

ol

t-

"~ 50 |

~-410)Jm ,,

~

v' -? o ! -.o

_~ -8 -V s !

en W

: / , ',/

V

~

'

"

~

. -4

~

0

¥ -3

2 pnn I PGF2~:(10aM>

I

I

1 PGF2odlOaM)

Fig. 1. Original tracing of an experiment on a proximal (prox) and a distal (dist) segment of a rat left anterior descendens coronary artery tested in parallel in a dual myograph. After precontraction with prostaglandin F2,~ 10 /xM, both relaxed in response to increasing concentrations (expressed in log molar) of acetylcholine. Histamine elicits further contraction of the proximal segment, but relaxation of the distal segment. ( 0 = effective lumen diameter; W = wash).

of prostaglandin F2~, active wall tensions (= tension under prostaglandin F2,, minus passive resting wall tension) were respectively 2.36 + 0.21 and 1.27 + 0.1 mN/mm. 3.1. Effect of histamine on proximal rat coronary arteries The effect of increasing concentrations (0.1 ~ M - 1 mM) of histamine was studied on prostaglandin F2,~(10 ~M)-contracted preparations isolated from the proximal part of the left anterior descendens coronary artery. This resulted in a progressive further contraction of the preparations (Figs. 1, 2, and 3). 3.1.1. Histamine receptors The receptor type involved in the contractile effect of histamine was studied by assessing the response to

®

® '- 100

....

"/t't"~

100

....

,.<**/;"

ol

e-

x

50

50

|

I

i

I

I

I

11)7 I()6 I()5 104 10-3 HIS TAMINE (M)

0

I

I

I

I

I

I

I

lo 7 106 lo ~ lo 4 1031M)

I

I

I

11~7 1()6 1(~5 1(~4 103 HISTAMINE (M)

Fig. 2. Effects (in percentage of tension under prostaglandin F2,~ 10 /zM) on proximal segments of rat left anterior descendens coronary arteries of increasing molar concentrations of histamine (A) before (_e) and after ( e - - - e ) adding cimetidine (100 # M ) or (B) before (e e) and after ( e - - - e ) adding mepyramine (10/zM) to the bath solution (n = 7) (* P < 0.05; *** P < 0.01).

Fig. 3. Effects (in percentage of tension under prostaglandin F2,~ 10 ~M) of increasing molar concentrations of histamine (A -), 2-pyridylethylamine (e e) and dimaprit ( e - - - e) on proximal segments of rat left anterior descendens coronary arteries (n = 4).

histamine in the absence and presence of either the histamine n 1 receptor antagonist, mepyramine (10 /zM), or the histamine H 2 receptor antagonist, cimetidine (100/~M). The proximal arteries were incubated for 10 min with the antagonists before addition of prostaglandin F2~ and making of the concentration-response curve to histamine. The results of these experiments are presented in Fig. 2 and show clearly that the contraction was not blocked by cimetidine. On the other hand, in the presence of mepyramine, the contraction induced by histamine was completely blocked and even a small relaxation was observed. Fig. 3 represents the effects of increasing concentrations of the histamine H 1 receptor agonist, 2-pyridylethylamine (0.1 ~ M - 1 mM), and the histamine H 2 receptor agonist, dimaprit (0.1 /xM-0.1 mM) (n = 4), on prostaglandin F2,,-induced contractions. While 2pyridylethylamine elicited a further contraction, dimaprit exerted a relaxing effect. The observations with histamine agonists and antagonists indicate that the contractile effect of histamine on proximal coronary arteries of the rat can be attributed to activation of a histamine H~ receptor mechanism. 3.1.2. Role of NO-synthase and endothelium In these experiments the contractile effect of histamine was investigated in proximal arteries before and after removal of the endothelium and before and after addition of the NO-synthase inhibitor, NC-nitro-Larginine (100 ~M), added 10 min before the concentration-response curves were made. Removal of the endothelium resulted in complete loss of the relaxing effect of acetylcholine, but the contractile effect of histamine was still present and even somewhat potentiated (although not significantly; n = 5; data not shown). The addition of NG-nitro-L-arginine elicited a prominent increase in basal tone (38 + 8% of the total tone under prostaglandin F2, 10/~M) in these preparations, indicating a substantial basal NO-synthase activ-

J. Van de Voorde et al. / European Journal of Pharmacology 271 (1994) 17-23

20

Table 1 Active force developed after consecutive applications (1 to 4) of prostaglandin F2,, (10 ~ M ) to proximal and distal rat left anterior descendens coronary arteries in the absence (control - application 1 and 2) and presence (application 3 and 4) of 100 /xM N~-nitro-L arginine methylester (L-NAME) or NG-nitro-L-arginine (L-NNA) 1

2

3

4

n

Distal arteries Control 4.86+0.77 Control 3.31+0.54

4.81+0.80 3.13+0.52

+ L-NAME 4.73+0.78 + L-NNA 3.40+0.71

3.29 + 0.82

8

6.37 + 0.78

\

~ " ~ l \ - - L. I ¢-

x

I

50

|

o

N

0

I

I

50

\/

I

I

I

0

I

ll) 7 11)6 l() 5 11).4 10-3

6

+ L-NNA 6.94 + 0.83

The results are expressed in m N + S.E.M. n = n u m b e r of observations.

i

,.

i

T

I

I

10-7 10-6 1()5 10-4 10-3

HISTAMINE (M)

Proximal arteries Control -

100

'- 100,

o. 4.67 + 0.74

®

®

HISTAMINE (M)

Fig. 4. Effects (in percentage of tension under prostaglandin F2a 10 /~M) on distal segments of rat left anterior descendens coronary arteries of increasing molar concentrations of histamine (A) before (~ o) and after ( o - - - o ) adding cimetidine (100/~M) or (B) before (o o) and after ( o - - - o ) adding mepyramine (10/~M) to the bath solution (n = 5) (** P < 0.02; *** P < 0.01; **** P <

0.001). ity in proximal rat coronary arteries. However, the tone elicited by prostaglandin F2~ 10 /zM was not influenced by the presence of NC-nitro-L-arginine (100 /zM) (Table 1). It was found that the contractile effect of histamine was not potentiated by blockade of NOsynthase activity (n = 5; data not shown). 3.2. Effect of histamine on distal rat coronary arteries The effect of increasing concentrations (0.1 /zM-1 mM) of histamine was also studied on prostaglandin F2,, (10 ~M)-contracted preparations isolated from more distal parts of the left anterior descendens coronary artery: segments of the main left anterior descendens, distal to the first bifurcation, and segments of more distal side branches of the artery. Normalized diameters of these segments ranged from 386 to 181 /xM. On all these segments, histamine consistently elicited relaxation (Fig. 1). 3.2.1. Histamine receptors To investigate the receptor type involved in the relaxing effect of histamine, the relaxations were compared before and after addition of mepyramine (10 /~M) or cimetidine (100 /.~M). The preparations were incubated for 10 min with the antagonists before precontraction with prostaglandin F2~ and the concentration-response curve to histamine. The results are presented in Fig. 4. The relaxation response was potently inhibited by cimetidine, but not by mepyramine. These results indicate that the histamine-induced relaxation was histamine H 2 receptor-mediated. This was confirmed in another series of experiments on the effect of increasing concentrations of the histamine H 1 receptor agonist, 2-pyridylethylamine (0.1 /zM-1 mM), and the histamine H 2 receptor agonist, dimaprit (0.1 /zM-0.1 mM), on precontracted preparations. These results are shown in Fig. 5. The addition of 2-pyridylethylamine

did not elicit a relaxation effect, only a contraction at the highest concentration. On the other hand, dimaprit elicited a marked concentration-dependent relaxation. 3.2.2. Role of NO-synthase and endothelium Experiments were also performed to investigate whether the histamine-induced relaxation could be due to activation of NO-synthase. Therefore the effects of acetylcholine and histamine were investigated before and after addition of the NO-synthase inhibitor, N Cnitro-L-arginine methylester (100 /zM), added 10 min before making the concentration-response curves. The results are shown in Fig. 6A. While the relaxing influence of acetylcholine was completely blocked in the presence of NG-nitro-L-arginine methylester, the relaxing influence of histamine was not affected. Since NC-nitro-L-arginine methylester has been reported to have muscarine blocking effects in addition to its NOsynthase inhibiting effect (Buxton et al., 1993), we also investigated the influence of another NO-synthase inhibitor, N~-nitro-L-arginine (100/zM). The results (Fig.

100

"

'\

.N

= ×

50

I

o.

~ ~,~

0

i

l

i

I

I

11)7 1()6 11)5 104 1113(M) Fig. 5. Effects (in percentage of tension u n d e r prostaglandin F2a 10 /zM) of increasing molar concentrations of 2-pyridylethylamine (o o) and dimaprit ( o - - - o) on distal segments of rat left anterior descendens coronary arteries (n = 10).

J. Van de Voorde et al. / European Journal of Pharmacology 271 (1994) 17-23

®

Q,I

~ 100

,__.,,,.

.90 I00 I ol .c_

.c_

kij "

"; 50

21

5o !

o.

Q.

0

I

I

I

I

I

I

I

I

I

I

'~

®

0

I

i

,

I

lo 7 lo 6 165 lo4 io 3

Q

g 100

HISTAMINE(M)

ol

Fig. 8. Effects (in percentage of tension under prostaglandin F2<~ 10 p,M) on distal segments of rat left anterior descendens coronary arteries of increasing molar concentrations of acetylcholine and histamine before (o ll) and after ( i t - - - o) addition of glibenclamide (10/zM) (n = 7) (* P < 0.05; ** P < 0.02).

.E

~,| 50 O.

0

I

I

I

I

9 168

I

I

I

I

I

I

I

11)7 166 105 I04 163 HISTAMINE ( M )

ill s

ACETYLCHOLINE(M)

Fig. 6. Effects (in percentage of tension under prostaglandin Fz~ 10 /zM) on distal segments of rat left anterior descendens coronary arteries of increasing molar concentrations of acetylcholine and histamine (A) before (ll o) and after ( o - - - o ) adding N cnitro-L-arginine methylester (100 /~M) (n = 8) or (B) before (it i ) and after ( i - - - I ) adding NG-nitro-L-arginine (100 /J,M) (n = 8) to the bath solution (** P < 0.02; *** P < 0.01; **** P < 0.001).

6B) illustrate that N°-nitro-L-arginine has an influence similar to that of NC-nitro-L-arginine methylester. That the relaxing influence of histamine is not related to activation of NO-synthase is further supported by the observation that histamine elicits similar concentration-relaxation responses from preparations lacking functional endothelium as indicated by the absence of a relaxing influence of acetylcholine (data not shown). To ascertain that the lack of relaxation in response to acetylcholine after NO-synthase inhibition was not due to a time-related effect, some experiments (n = 3)

g lOO o'1 c x

50

o

T-"

L,.

o

~

o

i

I

I

167

106

I1~5

I

I

104 11i3

HISTAMINE

were also performed in which the concentration-response curve for acetycholine was repeated without addition of a NO-synthase inhibitor. No difference was found between the first and the second application (data not shown). The addition of NC-nitro-L-arginine methylester or N°-nitro-L-arginine elicited an increase in tone (respectively 16-t-5% and 10 + 5% of the tone under prostaglandin F2<~ 10 /zM) in most (respectively n = 6/8; n = 4/6) preparations, indicating basal NO-synthase activity in rat coronary arteries. However, the tone elicited by prostaglandin F2~ 10 /zM was not influenced by the presence of N°-nitro-L-arginine methylester or NG-nitro-L-arginine (100 /zM) (Table 1).

3.2. 3. Role of prostaglandins To investigate whether the histamine-induced relaxation results from activation of prostaglandin synthesis, concentration-response curves were made before and after incubation of the preparations for 20 min with indomethacin (10 /zM). From the results (Fig. 7), it is clear that the relaxing influence of histamine was not blocked in the presence of indomethacin. 3.2.4. Role of ATP-regulated K ÷ channels To investigate the possible contribution of ATP-regulated K ÷ channels, we also investigated the relaxation effect of histamine before and after incubation of the preparations for 10 min with glibenclamide 10 /zM. The results are presented in Fig. 8. These data clearly indicate that the relaxing influence of histamine is certainly not inhibited in the presence of glibenclamide.

(M)

Fig. 7. Effects (in percentage of tension under prostaglandin F2a 10 /~M) on distal segments of rat left anterior descendens coronary arteries of increasing molar concentrations of acetylcholine and histamine before (o o) and after ( l l - - - l l ) addition of indomethacin (10/zM) (n = 8).

4. Discussion

As far as we know, the present study is the first one to describe the functional effects of histamine on iso-

22

J. Van de Voorde et al. / European Journal of Pharmacology 271 (1994) 17-23

lated rat coronaries. Coronary arteries of the rat seem to be less sensitive to histamine than those of other species. On the other hand our observations in rat confirm the heterogeneity of the vascular responses to histamine occurring in the coronary circulation of other species (Levi et al., 1991): contraction on proximal segments, but relaxation on more distal segments of the same left anterior descendens coronary artery. Similar divergent reactions to histamine have been reported in human coronary circulation. Keitoku et al. (1988) showed that human distal epicardial coronary artery rings respond to histamine by relaxation, while proximal segments respond by contraction. In vivo studies on human coronary circulation yielded similar results: histamine administration causes spasm in large epicardial vessels, while distal coronary resistance vessels dilate in response to histamine, with a resulting increase in coronary blood flow (Vigorito et al., 1986). Histamine-induced vasoconstriction is generally associated with activation of histamine H 1 receptors on the smooth muscle cells (Levi et al., 1991). This is also true for the contraction we observed on proximal segments of rat coronary artery: it is blocked by the histamine H I receptor antagonist, mepyramine, and not by the histamine H 2 receptor antagonist, cimetidine. Furthermore, the selective histamine H 1 agonist, 2-pyridylethylamine, but not the selective histamine H 2 agonist, dimaprit, elicits contraction. The contraction is not related to the release of an endothelial contractile factor since histamine still contracts preparations without functional endothelium. Histamine-induced relaxation can result from activation of histamine H I receptors a n d / o r histamine H 2 receptors (Leusen and Van de Voorde, 1988; Levi et al., 1991). The relaxation induced by histamine in distal rat coronary segments is histamine H 2 receptor-mediated. This can be concluded from the observations that the relaxing influence of histamine is not antagonized by the selective histamine H I receptor antagonist, mepyramine, but largely inhibited by the histamine H 2 receptor antagonist, cimetidine. This conclusion is further supported by the observation that dimaprit but not 2-pyridylethylamine elicits relaxation. It is well known that the endothelium can play an important role in the relaxation induced by histamine (Van de Voorde and Leusen, 1983; Leusen and Van de Voorde, 1988) as is also the case for acetylcholine. Endothelium-mediated relaxation is mainly due to activation of NO-synthase, converting L-arginine to NO, which diffuses to the smooth muscle cells and elicits relaxation. This mechanism is blocked by L-arginine analogues, e.g. NG-nitro-L-arginine methylester and NG-nitro-L-arginine. In previous (unpublished) experiments on rat aorta we found that the relaxation in response to histamine is almost completely blocked by these analogues. However, we found that, in distal

coronary arteries, NG-nitro-L-arginine methylester blocked the relaxation induced by acetylcholine but not that induced by histamine. Since NG-nitro-L-arginine methylester might have anti-muscarinic effects (Buxton et al., 1993), we also investigated the influence of NG-nitro-L-arginine which lacks this aspecific effect. The results were similar to those with NG-nitro-Larginine methylester. It should however be mentioned that the L-arginine analogues increase the basal tone of both proximal and distal arteries, suggesting a substantial basal NO-synthase activity in rat coronaries. Our observations thus indicate that endothelial NO is involved in the relaxation induced by acetylcholine but not in that induced by histamine. This is in line with the fact that the formation of NO by the endothelium is histamine H I receptor mediated (Van de Voorde and Leusen, 1983). This conclusion is further substantiated by the observation that preparations without functional endothelium, evidenced by the lack of response to acetylcholine, show the same relaxing effects of histamine as preparations with a functionally intact endothelium. A direct muscular histamine HE-mediated relaxation has also been described in coronaries of other species, including humans. However, in human coronaries, the relaxation caused by histamine is also partly mediated by a histamine H 1 receptor-mediated endothelium-dependent mechanism (as in the rat aorta) (Keitoku et al., 1988; Toda, 1987). Histamine is a known stimulator of arachidonic acid metabolism, producing potent vasoactive substances such as prostacyclin (Toda et al., 1982). A possible involvement of cyclo-oxygenase metabolites in histamine-induced relaxation of rat coronary arteries is excluded by the lack of influence of the cyclo-oxygenase inhibitor, indomethacin. It is well documented that H E receptors on vascular smooth muscle cells are linked to the adenyl cyclase system with cyclic AMP as second messenger. It is therefore most probable that histamine elicits its relaxation effect on distal rat coronaries by this mechanism. In this respect, it can be noticed that 8-Br-cAMP hyperpolarizes guinea-pig coronary artery (Parkington et al., 1993). To exclude the possibility that histamine might induce relaxation of distal segments by hyperpolarization resulting from activation of ATP-regulated K+-channels, we also investigated the influence of glibenclamide, a potent blocker of these channels (Nelson et al., 1990). The effect of histamine was clearly not blocked, excluding this possibility. On the contrary, the relaxation effect was potentiated, an observation which remains to be explained. A study of the effects of histamine in isolated perfused rat hearts suggests that the histamine-mediated vasodilation was due instead to histamine H 1 receptor activation (Akar et al., 1984). It is not unlikely therefore that vessels of the coronary vasculature other than

J. Van de Voorde et al. / European Journal of Pharmacology 271 (1994) 17-23

those investigated in this study relax by a histamine Ht-mediated mechanism. Since a Hi-mediated vascular relaxation is usually associated with an endothelium-dependent mechanism, the possible involvement of nitric oxide in histamine-induced vasodilation in rat coronary circulation cannot be excluded. In summary we can conclude that the influence of histamine on rat coronary arteries is very heterogenous. This seems to rely, at least partly, on a heterogenous distribution of H 1 and H 2 receptors on the smooth muscle cells, the activation of which respectively induces contraction and relaxation. The question as to whether the release of histamine and the use of antagonists are beneficial or deleterious for the coronary circulation will therefore depend on the balance of the contractile and relaxing influences.

Acknowledgements This work was supported by grants (7.0038.91 and 33.0016.94) from the National Fund for Scientific Research (Belgium). B.V. is a senior research associate of the National Fund for Scientific Research. This work was presented in part at the meeting of the Belgian Society for Fundamental and Clinical Physiology and Pharmacology, Antwerp, February 19, 1994.

References Akar, F., Z.S. Ercan and R.K. Tiirker, 1984, A marked H~-receptormediated vasodilator effect of histamine in the isolated perfused rat heart, Eur. J. Pharmacol. 97, 265. Buxton, I.L.O, D.J. Cheek, D. Eckman, D.P. Westfall, K.M. Sanders and K.D. Keel 1993, N°-Nitro L-arginine methyl ester and other alkyl esters of arginine are muscarinic receptor antagonists, Circ. Res. 72, 387. Kalsner, S. and R. Richards, 1984, Coronary arteries of cardiac patients are hyperreactive and contain stores of amines: a mechanism for coronary spasm, Science 223, 1435. Keitoku, M., Y. Maruyama and T. Takishima, 1988, The receptor

23

mechanisms for histamine actions in proximal portion differ from those in distal portion in human coronary artery, Tohoku J. Exp. Med. 153, 415. Leusen, I. and J. Van de Voorde, 1988, Endothelium-dependent responses to histamine, in: Vasodilatation: Vascular Smooth Muscle, Peptides, Autonomic Nerves, and Endothelium, ed. P.M. Vanhoutte (Raven Press, New York) p. 469. Levi, R., L.E. Rubin and S.S. Gross, 1991, Histamine in cardiovascular function and dysfunction: recent developments, in: Histamine and Histamine Antagonists, Handbook Exptl. Pharmacol., Vol. 97, ed. B. Uvn~is (Springer-Verlag, Berlin, Heidelberg) p. 347. Nelson, M.T., J.B. Patlak, J.F. Worley and N.B. Standen, 1990, Calcium channels, potassium channels, and voltage dependence of arterial smooth muscle tone, Am. J. Physiol. 259, C3. Nyborg, N.C.B. and E.O. Mikkelsen, 1990, 5-Hydroxytryptamine does not release endothelium-derived relaxing factor in rat isolated coronary arteries, Eur. J. Pharmacol. 186, 295. Nyborg, N.C.B., U. Baandrup, E.O. Mikkelsen and M.J. Mulvany, 1987, Active, passive and myogenic characteristics of isolated rat intramural coronary resistance arteries, Pfliig. Arch/Eur. J. Physiol. 410, 664. Mulvany, M.J. and W. Halpern, 1977, Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats, Circ. Res. 41, 19. Parkington, H.C., M. Tare, M.A. Tonta and H.A. Coleman, 1993, Stretch revealed three components in the hyperpolarization of guinea-pig coronary artery in response to acetylcholine, J. Physiol (London) 465, 459. Toda, N., 1987, Mechanism of histamine actions in human coronary arteries, Circ. Res. 61, 280. Toda, N., M. Konishi and M. Miyazaki, 1982, Involvement of endogenous prostaglandin 12 in the vascular action of histamine in dogs, J. Pharmacol. Exp. Ther. 223, 257. Tryba, M., F. Zevounou and M. Zenz, 1986, Prevention of histamine-induced cardiovascular reactions during the induction of anaesthesia following premedication with H 1- + H2-antagonists I.M., Br. J. Anaesth. 58, 478. Van de Voorde, J. and I. Leusen, 1983, Role of the endothelium in the vasodilator response of rat thoracic aorta to histamine, Eur. J. Pharmacol. 87, 113. Vigorito, C., S. Poto, G.B. Picotti, M. Triggiani and G. Marone, 1986, Effect of activation of the Hi-receptor on coronary hemodynamics in man, Circulation 73, 1175. Wolff, A.A. and R. Levi, 1986, Histamine and cardiac arrhythmias, Circ. Res. 58, 1. Yoshitomi, I., R. Oishi, Y. Itoh, K. Saeki, Y. Senoo and S. Teramoto, 1989, a-Fluoromethylhistidine decreases the histamine content of rat right atrium under the influence of sympathetic activity, Naunyn-Schmied. Arch. Pharmacol. 339, 528.