Effects of α,β-methylene ATP on resistance and capacitance blood vessels of the cat intestinal circulation; a comparison with other vasoconstrictor agents and sympathetic nerve stimulation

~r~r~~u/~ lounlal of Phar~~z~~o~o~,205 ( 1991) 35 -4 f 0 1991 Elsevier Science Publishers B.V. All rights reserved OOl4-2999/91/S03.%

EJP 52142

et ci

s Edwin M. Taylor and Michael E. Parsons

SmithKline Beecham Phnrmaceuticds, The Frythe, Welwyn, Hertfordshire AL6 9AR U.K.

Received 30 May 1991. revised MS received 13 August 1991, accepted 27 August 199t

The autoperfused intestinal circulation of pentobarbifone anaesthetized cats was used to s.pldy the effects of a,@methylcne ATP (l-100 pg i.a.1 on pre-capillary resistance vessels and post-capillary capacitance (venous) blood vessels in comparison with other vasoconstrictor agents (also given i.a.) and the effects of sympathetic nerve stimulation (0.2516 Hz). All tits were treated with atropine and propranofol. cr.@-Methylene ATP, noradrenaline and sympathetic nerve stimufation a11 caused dose- or ~quen~-de~ndent constriction of both resistance and capacitance vessels. ~,~-~ethylene ATP was particuliarfy active on capacitance vessels causing a greater constriction than either noradrenaline or sympathetic nerve stimuiation. In comparison, angiotensin II and vasopressin caused a selective constriction of resistance vessels and prostaglandin F,, a selective constriction of capacitance vessels. The results demonstrate that functional Pz, purinoceptors are present on both arterial and venous blood vessels of the cat intestinal circulation. ~,~-~ethylene

ATP; Pr pu~noceptors;

1. Introduction

Recently Burnstock and Kennedy (1985) have proposed that Pz pu~nocepto~, which mediate the extracellular effects of ATP, can be divided into two subtypes, Pt, and Pzy, involved, respectively, in contraction and relaxation of smooth muscle. In blood vessels Pz,, purinoceptors are thought to be present on endothelial cells with a few exceptions (Kennedy and Burnstock, 1985; Mathieson and Burnstock, 1985; Brizzolara and Bumstock, 1991) and to act by the release of endothehum-derived relaxing factor (EDRF) and prostacyclin (Pearson et al., 1983; Hellewell and Pearson, 1984; Fieehvood and Gordon, 1987). Pz, purinoceptors are located on vascular smooth muscle perhaps concentrated around sympathetic nerve endings where a physiological role for ATP has been proposed as a cotransmitter with noradrenaline (Su, 1983; Burnstock, 1986). Two synthetic analogues of ATP, cY,P-methylene ATP and &y-methylene ATP, are claimed to be selective Pz, purin~eptor agonists at low concentrations

Correspondence to: E.M. Tayior, Smith Kline Beecham Pharmaceuticals, The Frythe, Weiwyn, Hertfordshire AL6 9AR, U.K. Tel. 44.438.782 000 ext. 3506, fax 44.438.782 570.

Intestinal circulation; Vas~onst~cto~

while some 2-substituted ATP analogues tend to have a relatively greater agonist potency at PzY purinoceptors (Burnstock and Kennedy, 1985 for review and references). In addition, high concentrations of a,& methylene ATP cause a selective desensitisation of Pz, purinoceptors in vitro and the compound can be used as an effective antagonist. Recently ru,/3-methylene ATP has also been shown to elicit a vasoconstrictor effect in vivo (DeIbro et al., 1985; Collis and Nowell, 1986; Defbro and Bumstock, 1987; Taylor et al., 1989) and to cause some desensitisation of Pa purinoceptors after repeated high doses (Bulloch and McGrath, 1988; Taylor and Parsons, 1989) although some doubt exists concerning the selectivity of the effect (Taylor and Parsons, 1989). In previous experiments we have shown that Eu,p-methylene ATP causes dose-dependent increases in vascular resistance of the cat intestinal circulation with no effect on resistance vessels of the hindquarters and renal circulations (Taylor et al., 1989), suggesting a very heterogeneous distribution of Pz, purinoceptors in this species. These studies have been extended to determine the vasoconstrictor effects of a,@nethylene ATP on both precapillary resistance vessels and post-capillary capacitance vessels of the cat intestinal circulation using a double isovolaemic perfusion technique (Fielden et al., 1974; Taylor et al., 1981).

~a!e or femsle cats (2.5~-t.O kg) were anaesthetized rbitone sodium (sagatal) 60 me/kg i.p. nnulated to ensure a clear airway ressure (BP) was recorded from a ~a! artery. Ii? some esperiments heart Fate was tored using an instantaneous rate meter triggered the BP p&se. A superficial vein in the foreleg was ~~~~~ated for intravenous (i.v.1 injection of drugs OF 3d~~t~~~a~doses of anaesthetic if required. The left c3rotid artery and left femoral vein were esposed and premred for c3nnulation. The abdomen was opened with a midline incision carefully ligating any points where bleeding occurred. The large intestine, caudal to tbe posterior mesenteric artery. and the duodenum, at the point where it lies alongside the single mesenteric vein. were tied with double ligatures. The posterior mesenteric artery was also ligated. The sheets of conneetive tissue benveen the intestines and the trunk of the animal were cut ligating any small blood vessels as necessary. leaking the intestines connected to the rest of tbe animal only by the anterior me:-nteric artery 3nd associated nerve plexus. the mesenteric vein and lymph ducts. The anterior mesenteric artery was freed from the nerve plexas and prepared for cannulation. Tbe nerve ple;ws was ligated. The mesenteric vein was also prepared for cannulatioa. As far as possible the &mph ducts were left intact or cut so that the contents drained into the abdominal cavity. After completion of the surgery the cats were left for about 1 h to allow seepage of blood to cease and clots to form. During this time the perfusion apparatus was prepared for use and primed with 0.9% saline. Heparin, 1000 U/kg i.v., atropine, 1 mg/kg i.v., and propranolol, 1 mg/kg i.v., were administered and the intestinal circulation autoperfused at constant flow (20-30 ml/min) as described in the next section. 22. Autope$usion

apparatus

The intestinal circulation was autoperfused via an extracorporeal circuit of silicone rubber tubing (internal diameter 2 mm. total volume 5 ml), the blood flow under the control of two pulsatile pumps (Taylor et al., 1981). Blood taken from the cannulated carotid artery was pumped at constant flow into the anterior mesenteric artery (arterial pump). Perfusion pressure
the cannulated femoral vein (venous pump). Venous outflow pressure (P,) was recorded with a second Bell and Howell transducer connected into the extracorporal circuit between the mesenteric vein and the venous pump. Venous constriction increases P, (decrease in capacitance) while venous dilatation decreases Pv. Previous experiments (Fielden et al., 1974; Taylor et al., 1981) have shown that changes in PA and P, reflect independent changes in arterial (pre-capillary resistance) blood vessels and post-capillary capacitance tessels respectively. During testing of the apparatus it was found that significant changes in systemic BP altered the output of the arterial pump despite a constant pump speed, presumably due to a slightly greater filling of the tubing with increased input pressure. To overcome this problem a third pump was placed in the extracorporeal circuit between the carotid artery and the arterial pump. This pump was servocontrolled to operate at a constant pressure output achieved by rapid and continual alteration of pump speed. Output pressure was monitored with a third Bell and Howell transducer connected to the extracorporeal circuit between the servocontrolled pump and the arterial pump. So a constant pressure input to the arterial pump was maintained despite considerable fluctuations in systemic BP. A second advantage was that the actual input pressure into the arterial pump could be set and maintained at any level between O-500 mm Hg thus adding a greater control over the accurate balance of the two constant flow pumps so that P, was stable and in the region of 2-5 mm Hg. Any inbalance of flows was quickly detected by progressive changes in P,. As the mesenteric nerve was destroyed PA was relatively low, around 70-100 mm Hg. 2.3. Sympathetic nerve stimulation and drug administration

In experiments designed to investigate the effects of sympathetic nerve stimulation an electrode, similar in shape to an electromagnetic flow probe, was fitted around the anterior mesenteric artery and nerve plexus distal to the point of cannulation and nerve destruction. Transmural stimulation (30 V, 0.5 ms pulse duration) was applied across the artery and nerve plexus at different frequencies from 0.25 to 16 Hz. Each stimulation period was continued for 45-60 s until a peak vasoconstrictor response, to the given frequency, was obtained and intervals of 5-10 min were left between stimulation periods to allow the tissue to recover. Thus a frequency-response curve was constructed for both arterial and venous responses and compared directly in each experiment to the responses elicited by noradrenaline.

37

Vasoconstrictor drugs were given directly into the arterial supply to the intestines through a thick-walled section of the extracorporeal circuit (viton tubing, wall thickness 3 mm, length 2 cm) between the servo-controlled pump and the arterial pump. Noradrenaline was used as the standard and sequential dose-response curves were obtained in each experiment and compared directly with either a,&methylene ATP, angiotensin II, vasopressin (ADH) or prostaglandin F,, The volume of injection did not exceed 0.2 ml. All experiments were carried out in the presence of propranolol 1 mg/kg i.v. to prevent neuronally released or injected noradrenaline from stimulating fiadrenoceptors (Pate1 et al., 1981) and atropine 1 mg/kg i.v. to inhibit the parasympathetic nerves.

250

BP

1

mmHg

125 1 OJ

87 lmin

+I

2.4. Drug used

q&t lop

AAA

The following drugs were used: a$-methylene ATP (Sigma UK), atropine sulphate (BDH), noradrenaline HCl (Aldrich Chem. Co.), propranolol HCl (ICI), phenoxybenzamine (Smith Kline Beecham), angiotensin II (hypertensin; Ciba), vasopressin/ADH (pitressin; Parke Davis) and prostaglandin F,, (Upjohn). Phenoxybenzamine was dissolved in hot polyethylene glycol ‘400’ and diluted with an equal volume of 0.9% saline. All other drugs were dissolved in either distilled water or 0.9% saline. Dilutions were made in 0.9% saline. Ascorbic acid was added to the noradrenaline solutions to prevent oxidation. 2.5. Statjstical analysis

“q&3 qg

$3

NA

N

Fig. 1. Original tracing showing the effects of bolus injections of a$-methylene ATP (A) 2.5, 10 and 25 pg La. and noradrenaline (N) 2.5 and 5 fig i.a. on systemic blood pressure (BP) - upper record and on arterial perfusion pressure (P,) - middle record and venous outflow pressure (Pv) - lower record of the autoperfused intestinal circulation of an anaesthetised cat treated with atropine 1 mg/kg i.v. and propranolol 1 mg/kg i.v. Increases in PA and P, indicate arterial and venous constriction respectively of the intestinal circulation.

Increase in P,or P,, tmmHg> 115-17.5 15045

Results are expressed as mean the number of determinations n Statistical significance was tested and a probability level of P < statistically significant.

values + S.E.M. with given in parentheses. using Student’s t-test 0.05 was considered

125-42.5

100-40

3. Results 3.1. Resting levels of PA and P, Intestinal perfusion pressure (P,) was relatively low in these experiments probably because the efferent sympathetic nerves were destroyed. The mean value of PA from all the experiments was 83.1 f 3.9 mm Hg (n = 35). The equivalent value for the venous outflow pressure (Pv) was 2.9 + 0.5 mm Hg (n = 35). 3.2. Effects of a,@methylene

ATP and noradrenaline

Bolus injections of cu,P-methylene ATP (0.1-100 pg i.a.1 and noradrenaline (0.01-10 pg i.a.) caused dosedependent but relatively short-lasting increases in PA

25--2.5

o-o I 0.01

0.1

I 1

10

1 100

Dosecpg i.v.1

Fig. 2. Dose-dependent increases of arterial perfusion pressure PA (O 01, scale O-175 mm Hg and venous outflow pressure Pv (0 *I. scale O-17.5 mm Hg, in the autoperfused intestinal circulation of pentobarbitone anaesthetised cats (treated with atropine 1 mg/kg i-v. and propranolol 1 mg/kg i.v.) elicited by bolus injections of a$-methylene ATP (solid symbols) 0.1-100 pg La. (n = 8) and noradrenaline (open symbols) 0.01-10 fig i.a. (n = 35). Points represent mean values + S.E.M. The noradrenaline curve represents the mean values calculated from all the experiments (n = 35).

responses to noradrenaline were obtained before the addition of phenoxybenzamine. Angiotensin II caused a selective constriction of resistance vessels (increase in PAI with relatively little effect on capacitance vessels (figs. 3, 4). Vasopressin/ADH (OS-500 pg i.a.1 was a selective constrictor of resistance vessels having little effect on Pv even at high doses (fig. 31, while in contrast prostaglandin F,, (0.01-10 pg i.a.) caused a relatively seiective vasoconstrictor effect on capacitance vessels with only modest increases in PA (figs. 3, 4). 3.4. Effects of sympathetic newe stimulation

I

0

50

100

I

I

150

200

increase in 5 tmmHg1 Fig. 3. b~creases in arterial perfusion pressure PA (horizontal axis) @otted against concomitant increases in venous outflow pressure P, (vertical atis) elicited hy cr$ lene ATP. 0.1-100 pg i.a. ( n = Xl. noradrenaline. O.Oi-10 n = 35). angiotensin II. 0.001 pg La. (@. n = 10). vasopressin, 0.5-500 pg i.a. (0, n = 10). prostaglandin Fz,. 0.01-10 rug i.a. (4, n = 3-4) and sympathetic nerve stimulation, O.Z-16 Hz tc. n = 10). to show the relative effects of the drugs on intestinal resistance (P,) and capacitance (P,) blood vessels. Prosraglandin F,, caused a selective venous constriction while angiotensin II and vasopressin a selective arterial constriction. Points represent mean values-f:S.E.M.

and P, (figs. 1, 2) indicating vasoconstriction of both pre-capillary resistance vessels and venous blood vessels respectively. More complete dose-response curves for noradrenaline were not attempted as repeated high doses caused deterioration of the preparation. a$thylene ATP was comparatively more effective than adrenaline on venous blood vessels for a given increase in PA. In contrast, noradrenaline was relatively more effective on resistance blood vessels and was able to induce a greater vasoconstrictor response than a& methylene ATP. Indeed, the increase in PC elicited by 1 rg a,&methylene ATP was slightly less than the increase seen following the 30 pg dose (fig. 2). The relative effects of the two compounds on resistance and capacitance blood vessels can be shown more clearly from a plot of changes in PA against changes in Pv (fig. 3). Clearly over the dose-range studied a& methylene ATP was a more effective vasoconstrictor of venous blood vessels than noradrenaline (curve for cu,P-methylene ATP is to the left of the noradrenaline curve in fig. 3). 3.3. Eflects of other c’asoconstrictor agents Angiotensin II (0.001-1.0 bg i.a.) was found to release catecholamines and so the experiments were carried out in the presence of phenoxybenzamine 10 mg/kg i.v. Obviously the comparative vasoconstrictor

Stimulation of the sympathetic nerves caused frequency-dependent constriction of both resistance and capacitance blood vessels (fig. 5a). The reactivity of the capacitance vessels to sympathetic nerve stimulation appeared to be relatively greater, at frequencies from 0.25 to 8 Hz, than the reactivity of the resistance vessels, when calculated at a percentage of the respective maximal responses (fig. 5b). Similarly the maximal response of the capacitance vessels was achieved with a stimulation frequency of about 8 Hz whereas a frequency of 16 Hz was required to induce a maximal response of the resistance vessels (fig. 5b). Sympathetic nerve stimulation elicited a relatively greater constrictor effect on capacitance vessels than

1

250 BP mmHg

1 min m

OJ

rr 2+g N

44 +g P

o.H5)(g AN

12&r V

Fig. 4. Original tracings (from different experiments) showing the effects of bolus injections of noradrenaline (NJ 2.5 1.18 La., prostaglandin F,, (P) 10 pg i.a., angiotensin II (AN)0.25 ~g La. (in the presence of phenoxybenzamine (10 mg/kg i.v.) and vasopressin 6')125 pg La. on systemic blood pressure (BP) - upper record and on arterial perfusion pressure (P,) - middle record and venous outflow pressure (P,) - lower record of the autoperfused intestinal circulation of pentobarbitone anaesthetised cats treated with atropine 1 mg/kg iv. and propranolol 1 mg/kg i.v. Noradrenaline constricts arterial and venous blood vessels while prostaglandin F,, is a selective venous constrictor and vasopressin and angiotensin II selective arterial constrictors. Note the more prolonged response to vasopressin.

39

a Increase

b

in

96 of maximal

~orPvCmmHgI

1

I

0.25 0.5

increases in p,orP,

I

I

I

I

I

1

2

4

8

18

I

I

0.25 0.5

I

I

I

I

I

1

2

4

8

18

Frequency (Hz)

Frequency CHz)

Fig. 5. (4 Frequency-dependent increases in arterial perfusion pressure (PAI W, scale O-200 mm Hg, and venous outflow pressure P, ( O-20 mm Hg in the autoperfused intestinal circulation of pentobarbitone anaesthetised cats (treated with atropine 1 mg/kg i.v. and propranolol 1 mg/kg iv.1 elicited by stimulation of the sympathetic nerves (30 V, 0.5 ms, 0.25-16 Hz). Points represent mean values (n = IO) + S.E.M. (b) The same values as in (a) but plotted as a percentage of the maximal increases in PA and Pv observed during the 16 Hz stimulation period. Points represent mean values f S.E.M.

noradrenaline for a given increase in resistance (fig. 3). The curves for QLmethylene ATP and sympathetic nerve stimulation were similar except for the relatively greater effects of high doses of a$-methylene ATP (lo-100 pg i.a.) on capacitance vessels (fig. 3).

4. -Discussion The present series of experiments have shown that cy,@methylene ATP is a powerful vasoconstrictor agent on both pre-capillary resistance vessels and postcapillary capacitance vessels. In fact, the effects on capacitance vessels are relatively more impressive than the previously reported effects on intestinal resistance vessels (Taylor et al., 1989). The smaller response of the resistance vessels to the 100 pg dose of cu,p-methylene ATP may indicate some tachyphylaxis or desensitisation of the receptors but it is rather surprising that no such effect was seen on the capacitance vessels. It is possible that high doses of ct$-methylene ATP may have weak, endothelial dependent, vasodilator activity. Rather surprisingly, u,&methylene ATP was able to elicit a greater increase in Pv (venoconstriction) than either injected noradrenaline or high frequency stimulation of the sympathetic nerves. The findings show that PzXpurinoceptors are present on both arterial and venous blood vessels of the cat intestinal circulation. The exact location of the receptors is not known but a postsynaptic site within the sympathetic neuromuscular junction is likely as it has been shown that purines may be taken up (as adenosine) and stored in sympathetic nerve endings and to function as a cotransmitter when released with noradrenaline (Su, 1983; Sneddon and Burnstock, 1984; Von Kugelgen and Starke, 1985; Kennedy et al., 1986; Burnstock, 1986; Burnstock and

Warland, 1987; Hirst and Lew, 1987; Muramatsu, 1987). It is not known if, or to what extent, extrajunctional Pt, purinoceptors are present on vascular smooth muscle. Negative feedback may also occur via presynaptic P, (adenosine) purinoceptors (Su, 1983). The present results are therefore consistent with the view that purines (particularly ATP) may play a part in sympathetic transmission by acting as a cotransmitter with noradrenaline. As cotransmitters, noradrenaline and ATP may have synergistic effects which are important in the control of the sphlanchnic circulation and the response to stress. Stimulation of the sympathetic nerves causes frequency-dependent constrictio=r of both resistance and capacitance blood vessels of the intestinal circulation with the latter showing slightly greater reactivity than the former which agrees with the resu!ts obtained by Mellander (1960) on the skeletal muscle circulation. The relatively greater effect of sympathetic nerve stimulation on capacitance blood vessels, in comparison with injected noradrenaline, is interesting and may have at least two possible explanations. Firstly, noradrenaline injected i.a. will result in a higher concentration at arterial cy-adrenoceptors than at venous (Yadrenoceptors (Vane, 1969). This effect will be more pronounced a: lower doses but will tend to disappear when higher aoses are given (see fig. 3 in the Results section). In contrast, neuronally released noradrenaline will result in a roughly equivalent concentration of transmitter at postsynaptic a-adrenoceptors in arterial and venous branches of the sympathetic nerves. Secondly, injected noradrenaline may stimulate receptors at different locations to neuronally released noradrenaline which will probably be largely restricted to intrajunctional a-adrenoceptors. Previous experiments (Taylor et al., 1989) using the autoperfused intestinal

e cat have shown that the vasoconstricdoses of noradrenaline are mediated ~~-adren~epto~ (effects inhibited by yohimbine t not by prazosin). whereas the effects of sympaetic nerve stimulation are mediated by aI- and LYEadrenoceptors with an additional purinergic compoaylor and Parsons, 1989). ne en studying changes in intestinal venous function care must be taken to ensure that the drugs used do not cause a pronounced increase in motility or spasm of gut muscle as this could result in changes in venous pressure unrelated to vascular effects (Walus and Jacobson, 19Sl). Such changes have been observed with bradykinin, acetylcholine, histamine and 5-HT. However. the compounds used in the present study, with the possible exception of prostaglandin F,, (Walus et al.. 1980: Konturek et al., 19851, do not cause spasm of gJt muscle. Indeed, noradrenaline. sympathetic nerve stimulation and cY$-methylene ATP tend to reduce gut motility (Walus and Jacobson, 1981; Satchel1 and Maguire, 1975). Angiotensin II and vasopressin elicited only minimal changes in Pv suggesting little effect on either venous or gut smooth muscle. Angiotensin II may stimulate Auerbachs plexus but any effects were inhibited in the present series of experiments as the animals were treated with atropine. The particular properties of vasopressin have led to its use in the treatment of portal hypertension (Richardson and Withringto~, 1952; Hogestltt et al., 1986) because it can reduce portal venous pressure by decreasing intestinal blood flow. Walus and Jacobson (1980) have suggested that a number of drugs, including prostaglandin Flar may alter gut motility but the observed vascular changes largely reflect vasoactive properties rather than effects on gut muscle. Clearly some part of the measured increase in Pv elicited by prostaglandin F,, may be due to an effect on gut muscle but the drug has been shown to constrict the portal vein in vitro (Hogestgtt et a!., 1986) and in our hands to constrict venous blood vessels of the cat paw (using the Haddy (1958) technique; unpublished observations). Presumably the selective effects of some vasoconstrictors on resistance and capacitance blood vessels reflects either differences in receptor density or relative effectiveness of the transduction and intracellular mechanisms involved in mediating the responses. In conclusion, a$-metbylene ATP has been shown to be a powerful vasoconstrictor of both arterial and venous blood vessels, the effects presumably mediated by Pt, purinoceptors on vascular smooth muscle. The results are consistent with the view that purines may be involved in synaptic transmission at sympathetic nerve endings. However, it is possible that the P,, purinoceptars involved are largely extrajunctional and therefore not involved in neurotransmission.

eferences Brizzolara. A.L. and G. Burnstock, 1991, ?ndothelium-dependent and endothelium-independent vasodilatation of the hepatic artery of the rabbit, Br. J. Pharmacol. 103, 1206. Bulloch. J.M. and J.C. McGrath, 1988, Blockade of vasopressor and vas deferens responses by a$-methylene ATT’ in the pithed rat, Br. J. Pharmacol. 91, 457. Burnstock. G., 1986, The changing face of autonomic neurotransmission. Acta Physiol. Stand. 126, 67. Burnstock, G. and C. Kennedy, 198.5,Is there a basis for distinguishing two types of Pa-purinoceptor?, Gen. Pharmacol. 16,433. Burnstock. G. and J.J.I. Warland, 1987, A pharmacological study of the rabbit saphenous artery in vitro: a vessel with a large purinergic contractile response to sympathetic nerve stimulation, Br. J. Pharmacol. 90, 1I. Collis, M.G. and C.G. Nowell. 1986, Investigation of the receptors mediating in vivo vasodilator responses to ATP and adenosine, Br. J. Pharmacol. 88, 438P. Delbro, D. and G. Bumstock, 1987, Depressor and pressor actions of purine nucleosides and nucleotides in the anaesthetized rat, Acta Physiol. &and. 130, 373. Delbro, D.. H. Hedlund. C. Kennedy and G. Burnstock, 1985, Potent vasoconstrictor actions of a$-methylene ATP, a stable analogue of ATP, on the rat vasculature, in vivo, Acta Physiol. Stand. 123, 501. Fielden, R., D.A.A. Owen and E.M. Taylor, 19i4, Hypotensive and vasodilator actions of SK&F24260,a new dihydropyridine derivative. Br. J. Pharmacol. 52. 323. Fleetwood, G. and J.L. Gordon, 1987, Purinoceptors in the rat heart, Br. J. Pharmacol. 90, 219. Haddy. F.J.. 1958, Vasomotion in systemic arteries, small vessels and veins determined by direct resistance measurements, Minn. Med. 41, 162. Hellewell, P.G. and J.D. Pearson, 1984, Purinoceptor mediated stimulation of prostacyclin release in porcine pulmonary vasculature, Br. J. Pharmacol. 83, 457. Hirst, G.D.S. and M.J. Lew, 1987, Lack of involvement of o-adrenoceptors in sympathetic neural vasoconstriction in the hindquarters of the rabbit, Br. J. Pharmacol. 90, 51. HogestHtt, E.D., L.-E. Hammarstrom, K.-E. Andersson and T. Holmin, 1986, Contractile effects of various vasoactive agents in small rat portal veins and hepatic arteries and the influence of sympathetic denervation on the noradrenaline responses, Acta Physiol. Stand. 128, 309. Kennedy, C. and G. Burnstock, 1985, Evidence for two types of P,-purinoceptor in the longitudinal muscle of the rabbit portal vein, Eur. J. Pharmacol. 111, 49. Kennedy, C., V.L. Saville and G. Burnstock, 1986, The contributions of noradrenaline and ATP to the responses of the rabbit central ear artery to sympathetic nerve stimulation depend on the parameters of stimulation, Eur. J. Pharmacol. 122, 291. Konturek, S.J., P. Thor, J.N. Konturek and W. Pawlik, 1985, Role of prostaglandins in intestinal secretion, motility and circulation, in: Advances in Prostacyclin, Thromboxane and Leukotriene Research, Vol. 15. eds. 0. Hayaishi and S. Yamamoto (Raven Press, New York) p. 647. Mathieson, J.J.I. and G. Burnstock, 1985. Purine-mediated relaxation and contraction of isolated rabbit mesenteric artery are not endothelium dependent, Eur. J. Pharmacol. 118, 221. Mellander, S., 1960, Comparative studies on the adrenergic neurohormonal control of resistance and capacitance blood vessels in the cat, Acta Physiol. Stand. 50 (Suppl. 1761, 1. Muramatsu, I., 1987, The effect of reserpine on sympathetic purinergic neurotransmission in the isolated mesenteric artery of the dog: a pharmacological study, Br. J. Pharmacol. 91, 467.

41 Patel. P.. D. Bose and C. Greenway, 1981, Effects of prazosin and phenoxybenzamine on (Y-and P-receptor mediated responses in intestinal resistance and capacitance vessels, J. Cardiovasc. Pharmacol. 3, 1050. Pearson, J.D., L.L. Slakey and J.L. Gordon, 1983, Stimulation of prostaglandin production through purinoceptors in cultivated porcine endothelial cells, Biochem. J. 214, 273. Richardson, P.D.I. and P.D. Withrington, 1982, Physiological regulation of the hepatic circulation, Ann. Rev. Physiol. 44, 57. Satchell, D.G. and M.H. Maguire, 1975, inhibitory effects of adenine nucleotide analogues on the isolated guinea-pig taenia coli, J. Pharmacol. Exp. Ther. 195. 540. Sneddon, P. and G. Burnstock, 1984, ATP as a neurotransmitter in rat tail artery, Eur. J. Pharmacol. 106, 149. Su. C., 1983, Purinergic neurotransmission and neuromodulation. Ann. Rev. Pharmacol. Toxicol. 23, 397. Taylor, E.M. and M.E. Parsons, 1989, Adrenergic and purinergic neurotransmission in arterial resistance vessels of the cat intestinal circulation, Eur. J. Pharmacol. 164. 23.

Taylor, E.M., D. Cameron, R.J. Eden, R. Fielden and D.A.A. Owen. 1981, Hemodynamic profile of a new antihypertensive agent D,L-3-[2-(3-t-butylamino-2-hydroxy-propo~)phenyl]-6-hydrazinopyridazine (SK&F 92657). J. Cardiovasc. Pharmacol. 3,337. Taylor, E.M., M.E. Parsons, P.W. Wright, M.A. Pipkin and W. Howson, 1989, The effects of adenosine triphosphate and related purines on arterial resistance vessels in vitro and in vivo, Eur. J. Pharmacol. 161, 121. Vane, J.R., 1969, The release and fate of vasoactive hormones in the circulation, Br. J. Pharmacol. 35, 209. Von Kiigelgen, I. and K. Starke, 1985, Noradrenaline and adenosine triphosphate as co-transmitters of neurogenic vasoconstriction in rabbit mesenteric artery, J. Physiol. 367, 435. Walus, K.M. and E.D. Jacobson, 1981, Relation between small intestinal motility and secretion, Am. J. Physiol. 241, G-l. Walus, K.M., J.D. Fondacaro and E.D. Jacobson, 1980, Hemodynamic and metabolic responses to increased intestinal motor activity, Gastroenterology, 78. 1287 (Abstract).