INVESTIGATION OF THE ADENOSINE RECEPTORS

PRESENT ON THE THORACIC AORTA AND

CORONARY ARTERIES OF THE RAT

By

Cheryl LEWIS, B.Sc. (Hons)

A thesis submitted in accordance with the requirements of the University of Surrey for the Degree of Doctor of Philosophy.

Receptors and Cellular Regulation Research Group November 1995.School of Biological Sciences,University of Surrey,Guildford,Surrey GU2 5XH.

ProQuest Number: 27606619

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is d e p e n d e n t upon the quality of the copy subm itted.

In the unlikely e v e n t that the author did not send a c o m p le te manuscript and there are missing p a g e s , these will be n oted . Also, if material had to be rem oved,

a n o te will ind icate the deletion .

uestProQuest 27606619

Published by ProQuest LLO (2019). Copyright of the Dissertation is held by the Author.

All rights reserved.This work is protected against unauthorized copying under Title 17, United States C o d e

Microform Edition © ProQuest LLO.

ProQuest LLO.789 East Eisenhower Parkway

P.Q. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346

ACKNOWLEDGEMENTS

I would like to thank Dr. Susanna Hourani for her supervision of this research. Thank

you for your help not only in the practical side of the project but also with the difficult

task of writing my paper and this thesis. Things may not always have worked but I

have thoroughly enjoyed working on this project. It has been a wonderful experience

and I have learnt many things over the last three years.

I would also like to acknowledge Pfizer Central Research for the funding of this

project and thank Dr. Mike Collis and Dr. Clive Long for the help, advice and

guidance given over the last three years.

Many thanks go to Ian, Julie, Julia, Debbie, Mary, Surekha, Runi, Mike and lately

George and Jason who have made my time in the lab happy and enjoyable and who

have also provided some of the wonderfiil memories I have of Guildford.

My thanks to Debbie Prentice for reading and correcting the many pieces of paper that

have contained my thesis and answering all the questions I've posed in the last few

months.

Special thanks go to David Hauton whose fiiendship during my time in Guildford was

very much appreciated. Thank you for being there when help was needed and for

supplying some good times, an occasional shoulder to cry on and a hot meal when it

was needed!

Lastly, to my parents to whom I owe my greatest thanks. Thank you for your unfailing

support over the last seven years, for all the sacrifices made and all the love given.

LIST OF ABBREVIATIONS

8 -SPT 8 -p-(sulphophenyl)theophylline

APNEA N -2-(4-amino-3-phenyl)ethyladenosine

ATP Adenosine triphosphate

B W-A522 3-(3-iodo-4-aminobenzyl)-8-(4-oxyacetate)-1 -propylxanthine

CGS 21680 2-[p-(carbonylethyl)phenylethylamino)-5’-(N-carboxamido)adenosine

CHA N^-cyclohexyladenosine

CPA N^-cyclopentyladenosine

CSC (8-(3-chlorostryryl)caffeine)

CV 1808 2-phenylaminoadenosine

DPCPX 1,3 -dipropyl-8 -cyclopentylxanthine

EDRF Endothelium derived relaxing factor

L-NAME N^-nitro-L-arginine methyl ester

MTA 5 ’ -methylthioadenosine

NA Noradrenaline

NEC A 5’-(N-ethylcarboxamido)adenosine

NO Nitric oxide

PD 115,199 (N-[2-(dimethylamino)ehtyl]-N-methyl-4-(2,3,6,7-tetrahydro-2,6-

dioxo-1,3-dipropyl-1 H-purin-8 -yl)benzene sulphonamide

PIA

SNP

XAC

N -(phenylisopropyl)adenosine

Sodium nitroprusside

Xanthine amine congener

ABSTRACT

1. The classification of adenosine receptors has been summarised and the involvement

of these receptors in the cardiovascular system has been described.

2. Adenosine and its analogues induced relaxations of the rat isolated thoracic aorta

and the rat coronary arteries. The agonist potency orders indicate the presence of A2

receptors although the high potency of the A2 selective agonist CGS 21680 suggests

that these relaxations were mediated via A2 receptors. However, as the maximal

responses to CGS 21680 in the rat thoracic aorta was only 6 6 % of the NEC A maximal

response, it is possible that CGS 21680 was acting as a partial agonist on the A2

receptors present on this tissue.

3. On the rat thoracic aorta, the endogenous nucleoside MTA was a very weak agonist

but its effects were not inhibited by the non selective antagonist 8 -SPT. This suggests

that the relaxations induced by this agonist were not mediated by the A2 receptors on

the aorta but by another site which is resistant to blockade by 8 -SPT. However, MTA

also appears to act as an antagonist on the A2 receptors on the rat aorta inhibiting the

relaxations induced by the adenosine agonists with a potency similar to that of 8 -SPT.

4. In the rat thoracic aorta, MTA, the non selective antagonist 8 -SPT and the A2 a

selective antagonist PD 115,199 inhibited the relaxations induced by the adenosine

agonists. However, NECA was inhibited to a greater extent than the other agonists by

these antagonists. This suggests that all of these agonists, vvith the exception of MTA,

were acting on the A2 a receptors present on the aorta. It also suggests that CGS

21680, CPA, adenosine and APNEA may not be acting solely on this receptor. It is

possible that these agonists were also acting on another site on the aorta which

displays some resistance to blockade by these antagonists, hence the lesser degree of

antagonism. This site may be the same xanthine-resistant site on the aorta at which

MTA acts as a weak agonist.

5. Aj receptors are unlikely to be present on the rat thoracic aorta as an Aj selective

concentration of DPCPX (InM) did not inhibit the relaxations induced by APNEA,

adenosine and CPA. The dififering degrees of antagonism of the agonist responses by

8 -SPT and MTA and the xanthine-resistant relaxant effects of MTA do not appear to

involve A3 receptors as the agonist potency order produced in the presence of these

antagonists was different to the potency order which would be expected if A3

receptors were present.

6 . Initially, when studying the adenosine receptors in the rat coronary arteries using the

Langendorff isolated heart technique, the concentration response curves for the

adenosine agonists in the presence and absence of antagonists and L-NAME were

carried out in separate hearts. However, interpretation of the increases in flow induced

by the agonists became difiScult due to other effects, such as decreases in basal flow,

which were produced by the inhibitors. Therefore the effect of the inhibitors on the

basal flow and the increases in flow induced by SNP and NECA were then investigated

in a paired fashion in each heart. It was found that these inhibitors caused a decrease

in the basal flow through the heart indicating vasoconstriction of the coronary vessels.

This vasoconstriction lead to physiological/functional antagonism of the agonist

induced responses causing a reduction in the vasodilation obtained. The results

obtained with the agonist concentration response curves were subsequently

reinterpreted in the light of this complication.

7. In the rat isolated heart, 8 -SPT inhibited the increase in flow induced by the

adenosine agonists confirming the involvement of the Pj receptors in this response and

also suggesting that A3 receptors are not involved in mediating coronary vasodilation.

8 , In the rat coronary arteries, the A2 a selective antagonist PD 115,199 inhibited the

increase in flow induced by NECA and CGS 21680. Whilst the concentration response

curve produced by NECA was shifted to the right by PD 115,199 (30nM), a lower

concentration of PD 115,199 (lOnM) virtually abolished the concentration response

curve produced by CGS 21680. Initially this suggested that both of these agonists act

via the A2 g receptors on this tissue although the different degrees of antagonism by PD

115,199 could not be explained. However, it was later found that PD 115,199 caused

vasoconstriction of the coronary vessels which led to physiological/functional

antagonism of the agonist responses and therefore a reduction in the vasodilations

induced by the agonists. When taking this physiological antagonism into

consideration, the concentration response curves for NECA in the presence and

absence of PD 115,199 were very similar, suggesting that, in fact, PD 115,199 had no

effect on the NECA induced responses. Therefore NECA was probably not acting via

the A2 a receptors but possibly via A215 receptors to produce vasodilatation of the

coronary arteries.

9. Mechanical removal of the endothelium from the rat isolated aorta, or incubation of

the intact aorta with the nitric oxide synthase inhibitor L-NAME (30pM), virtually

abolished the relaxations induced by the adenosine agonists. This indicates that these

responses are endothelium-dependent and are mediated via the nitric oxide pathway.

Initially in the rat isolated heart, the presence of L-NAME inhibited the concentration

response curves for NECA and CGS 21680 shifting the curves to the right, whilst

abolishing the responses to CPA and adenosine, although this was due to their low

potency in the absence of the inhibitor. This would imply that in the coronary arteries

the agonist induced vasodilations are partially endothelium-dependent and are

mediated via the endothelium and the vascular smooth muscle. However, it was found

that L-NAME caused vasoconstriction of the coronary vessels, which led to

physiological antagonism of the agonist responses and therefore a reduction in the

vasodilations induced by the agonists. When this physiological antagonism was taken

4

into consideration the concentration response curves of the agonists in the presence

and absence of L-NAME were very similar indicating that L-NAME had no effect on

the increases in flow induced by the agonists. This suggests that, unlike the rat

thoracic aorta, the vasodilation induced by the adenosine agonists in the coronary

arteries is endothelium-independent and does not involve the nitric oxide pathway.

1 0 . As well as decreasing the basal flow through the heart and reducing the increases in

flow induced by the agonists, L-NAME also decreased the force of contraction of the

heart. SNP however, not only increased the force but also increased the rate of

contraction of the heart although it had no effect on the isolated atria. These effects of

L-NAME and SNP were probably not direct effects but consequences of the changes

in vessel tone and subsequent changes in the flow rate induced by L-NAME and SNP.

A decrease or increase in the flow rate would lead to an insufficient or greater

perfiision of the heart respectively and consequently this would alter the hearts' ability

to contract.

11. In conclusion, the relaxations of the rat thoracic aorta induced by the adenosine

agonists are mediated via A2 a receptors and by another site on the aorta which displays

resistance to blockade by 8 -SPT, MTA and PD 115,199. The agonist responses were

abolished by endothelium removal and inhibition by the nitric oxide synthase inhibitor

L-NAME and therefore they were endothelium-dependent and were mediated via the

nitric oxide pathway. In the rat coronary arteries, the adenosine agonists caused

increases in flow through the heart indicating vasodilation of the vessel. Inhibitors

such as L-NAME, 8 -SPT and PD 115,199 caused vasoconstriction of the coronary

vessels, indicated by a decrease in the basal flow through the heart, which led to

physiological antagonism of the agonist induced vasodilations. This physiological

antagonism and a time dependent reduction in the agonist responses were taken into

consideration when interpretation of the agonist responses in the isolated heart were

carried out. The vasodilations induced by the agonists were mediated by A2 receptors

5

although the high potency of the A2 selective agonist CGS 21680 would suggest that

these receptors are of the A2 a receptor subtype. However, the increase in flow

induced by NECA was not inhibited by PD 115,199 suggesting that NECA is acting

via A2b receptors to mediate coronary vasodilation. The increases in flow induced by

the agonists were not inhibited by L-NAME indicating that they were endothelium-

independent and therefore mediated via receptors present on the vascular smooth

muscle of the coronary arteries.

LIST OF CONTENTS

Page number

ACKNOWLEDGEMENTS ABSTRACT LIST OF FIGURES LIST OF TABLES

12911

CHAPTER 1. INTRODUCTION

1.1. HISTORICAL BACKGROUND

1.2. PRODUCTION AND METABOLISM OF ADENOSINE

1.3. CLASSIFICATION OF PURINOCEPTORS

1.3.1. Pj purinoceptors

1.3.1.1. AI receptors

1.3.1.2 . A 2 receptors

1.3.1.3. A2 receptor subtypes

1.3.1.4. As receptors

1.3.1.5. A4 receptors

1.3.2. The P site

1.3.3. Adenosine antagonists

1.3.3.1. AI receptor antagonists

1.3.3.2 . A2 receptor antagonists

1.3.3.3. As receptor antagonists

1.3.4. Molecular biology studies

1.3.4.1. AI receptors

1.3.4.2 . A 2a receptors

1.3.4.3. A 2b receptors

1.3.4.4. As receptors

12

13

15

16

17

18

20

22

25

25

26

27

27

30

30

31

32

32

33

Page number

1.4. ACTIONS OF ADENOSINE ON THE CARDIOVASCULAR 34

SYSTEM

1.4.1. Ai receptor 34

1.4.2. Â2 receptor 35

1.5. AIM OF THE THESIS 41

CHAPTER 2. METHODS

2.1. PHARMACOLOGICAL METHODS 42

2.1.1. Rat isolated thoracic aorta 42

2.1.2. LangendorfT isolated rat heart preparation 44

2.1.3. Rat isolated atria 50

2 .2 . MATERIALS 51

CHAPTER 3. RESULTS

3.1. RAT THORACIC AORTA 53

3.2. RAT ISOLATED HEART 63

3.3. RAT ISOLATED ATRIA 72

CHAPTER 4. DISCUSSION

4.1. RAT THORACIC AORTA 76

4.2. RAT ISOLATED HEART AND ATRIA 84

4.3. CONCLUSIONS 98

REFERENCES 107

LIST OF FIGURES

Page number

Figure 1. Schematic diagram of the production and

metabolism of adenosine

Figure 2. Strucures of some adenosine agonists

Figure 3. Structures of some xanthine and non xanthine

adenosine antagonists

Figure 4. Concentration response curve for NA in the rat

thoracic aorta

Figure 5. Schmematic diagram of the isolated heart equipment

Figure 6 . Increase in flow induced by NECA in the rat

isolated heart

Figure 7. Relaxation of the rat thoracic aorta by adenosine

analogues and acetylcholine

Figure 8. Effect of removal of the endothelium on the

responses of adenosine analogues in the rat thoracic

aorta

Figure 9. Effect of removal of the endothelium and L-NAME

on the NA induced contractions of the rat thoracic

aorta

Figure 10. Effect of L-NAME on the responses of adenosine

analogues in the rat thoracic aorta

Figure 11. Effect of L-NAME on the SNP induced relaxation

of the rat thoracic aorta

Figure 12. Effect of 8 -SPT and MTA on the relaxations of the

rat thoracic aorta induced by adenosine analogues

Figure 13. Effect of PD 115,199 on the responses to NECA

and CGS 21680 in the rat thoracic aorta

9

14

19

29

43

46

48

54

56

57

58

60

61

62

Page number

Figure 14. Effect of DPCPX on the relaxations induced by

APNEA, adenosine and CPA in the rat thoracic aorta

Figure 15. Representative traces of some of the responses of

adenosine analogues in the rat thoracic aorta

Figure 16. Increase in flow induced by the adenosine analogues

in the rat isolated heart

Figure 17. Effect of 8 -SPT on the responses of adenosine

analogues in the rat isolated heart

Figure 18. Effect of PD 115,199 on the responses to NECA

and CGS 21680 in the rat isolated heart

Figure 19. Effect of L-NAME on the responses of adenosine

analogues on the rat isolated heart

Figure 20. Effect of L-NAME on the responses to ATP and

SNP in the rat isolated heart

Figure 21. Representative traces of the responses to SNP and

isoprenaline in the rat isolated atria

64

65

66

68

69

70

73

75

10

LIST OF TABLES

Page number

Table 1.

Table 2.

Table 3.

Table 4.

Agonist and antagonist potency orders for the

adenosine receptors

The classification and endothelium dependency of

adenosine receptors in the peripheral arteries of a

variety of species

pECgo values and the dose-ratios for 8 -SPT, MTA

and PD 115,199 for adenosine analogues in the

rat thoracic aorta

Effect of L-NAME, PE and adenosine antagonists

on the basal flow, and the increases in flow induced

by SNP and NECA in the isolated rat heart

21

37

55

74

11

CHAPTER 1: INTRODUCTION

1.1. HISTORICAL BACKGROUND

In 1929, an extract of ox heart was shown to produce physiological effects on the

cardiovascular system when injected into intact animals (Drury and Szent-Gyorgi,

1929). It was demonstrated that adenosine and adenylic acid (AMP, adenosine 5'-

monophosphate) isolated from these extracts, could modify the cardiovascular system

by inducing bradycardia and heart block, reducing the contractility of the atria and

decreasing the arterial blood pressure, which was, in part, due to arterial dilation

(Drury and Szent-Gyorgi, 1929). This was the first indication that physiological

systems could be modulated by purines such as adenosine.

Little interest was shown in these findings until 1963 when Berne proposed that

adenosine was involved in the metabolic regulation of coronary blood flow (Berne,

1963). When myocardial oxygen levels are low, for example during hypoxia and

ischaemia, high concentrations of inosine and hypoxanthine, metabolites of adenosine,

are measured in the blood. Berne proposed that adenosine released from the

myocardium was involved in a feedback mechanism to restore the myocardial oxygen

levels by causing arteriolar dilation and an increase in coronary blood flow (Berne,

1963).

It was not until 1970 that evidence was produced to suggest that adenosine modulated

physiological functions by acting at specific receptors. Sattin and Rail (1970)

demonstrated adenosine induced accumulation of cAMP in brain slices and the

inhibition of this effect by the presence of a methylxanthine, theophylline. This was the

first evidence to suggest that this effect of methylxanthines was due to antagonism of

adenosine receptors rather than phosphodiesterase inhibition, as previously thought

(Sattin and Rail, 1970).

12

Following these initial findings, the actions of adenosine on physiological systems other

than the cardiovascular system has been investigated, and fiirther characterisation of

the adenosine receptors has been carried out.

1.2. PRODUCTION AND METABOLISM OF ADENOSINE

There are two main pathways for the production of adenosine: i) hydrolysis of adenine

nucleotides, ii) hydrolysis of S-adenosylhomocysteine.

Intracellular dephosphorylation of adenosine 5'-triphosphate (ATP) to adenosine 5-

diphosphate (ADP), ADP to AMP and AMP to adenosine are catalysed by the

cytosolic enzymes adenosine triphosphatase, myokinase and 5'-nucleotidase

respectively (Pearson and Slakey, 1990; Schütz et al., 1981, Figure 1). This process

can also occur extracellularly catalysed by ectonucleotidases present on the outer

surface of the plasma membrane of vascular endothelial cells (Fleetwood et ah, 1989;

Schrader et a l, 1990). The second source of adenosine is S-adenosylhomocysteine

(SAH) which is derived fi'om its parent compound S-adenosylhomomethionine (SAM)

when SAM is utilised in transmethylation reactions. SAH undergoes enzymatic

hydrolysis by SAH hydrolase to produce adenosine and homocysteine (De La Haba

and Cantoni, 1959; Schrader et al., 1981, Figure 1). It is now understood that due to

the relative activities of the enzymes SAH hydrolase and 5’-nucleotidase the

production of adenosine through these pathways is balanced. The SAH pathway is

more dominant during normoxia when the activity of 5’-nucleotidase is very low and

the hydrolysis of 5'AMP becomes more important during hypoxia when 5’-

nucleotidase activity increases greatly whilst the activity of SAH hydrolase only

increases by 1.5 fold (Lloyd and Schrader, 1987). Adenosine can be metabolised to

inosine and hypoxanthine by adenosine deaminase and xanthine oxidase which occur

13

ATP

Adenosinetriphosphatase

ADP

Myokinase

HYPOXANTHINE SAM5’-AMPXanthine

Oxidase Adenosine kinase ^

ATP

5-nucleotidaseINOSINE

>^SAHhyrolase

homocysteineAdenosine"deaminase ADENOSINE

INTRACELLULAR

EXTRACELLULAR

ADENOSINEAdenosinedeaminase. Ecto

jiucleotidase

INOSINE 5-AMPXanthineOxidase^ Jiucleotidase

HYPOXANTHINE ADP

lucleotidase

ATP

Figure 1. Schematic diagram of the production and metabolism of adenosine

14

both as intracellular and extracellular enzymes. Alternatively, phosphorylation of

adenosine by cytosolic adenosine kinase can regenerate 5'-AMP (Figure 1).

Adenosine can be transported across the cell membrane by carrier mediated facilitated

diffusion (Oliver and Paterson, 1971). The carrier is able to transport nucleosides

other than adenosine across the membrane which subsequently can become

competitive inhibitors of adenosine transport. Some inhibitors of adenosine transport /

uptake have been synthesised including dipyridamole, 6 -S-(p-nitrobenzyl-

thio)guanosine (NBTGR) and 6 -S-(p-nitrobenzylthio)inosine (NBTI) (Paterson and

Oliver, 1971; Cass et al., 1974). Clinically dipyridamole has been used as a coronary

vasodilator mainly based on its action of adenosine uptake inhibition. The production

and metabolism of adenosine are described in greater detail in reviews by Olsson and

Pearson (1990) and Pearson and Slakey (1990).

1.3. CLASSIFICATION OF PURINOCEPTORS

Purinoceptors are cell surface receptors which can associate with, and may have their

actions modified by purines such as adenosine and its related analogues. The

classification of receptors such as purinoceptors have been carried out in a variety of

ways including observation of biological effects, association with individual second

messenger systems, radioligand binding and molecular biology techniques, but mainly

through agonist and antagonist potency orders and structure activity relationships

established for each of the receptors (Kenakin et al., 1992).

In 1978, Bumstock proposed that purinoceptors could be subdivided into 2 main

groups based on the potency order of adenosine and adenine nucleotides, sensitivity to

xanthine antagonists, modification of adenylate cyclase activity and stimulation of

prostaglandin synthesis. Pi purinoceptors are more sensitive to adenosine than ATP

producing a potency order of adenosine > AMP > ADP > ATP. Agonists acting at

15

these receptors cause the inhibition or stimulation of adenylate cyclase and this effect

can be antagonised by methylxanthines. purinoceptors however, have the reverse

potency order, ATP > ADP > AMP > adenosine and the effects of agonists are not

antagonised by methylxanthines but may lead to stimulation of prostaglandin synthesis

(Bumstock, 1978).

1.3.1. Pj purinoceptors

Pi purinoceptors have been further classified into Ai and A% purinoceptors. Van

Calker et al. (1979) reported that in cultured brain cells adenosine acted by either

inhibiting or stimulating adenylate cyclase via A \ and A% receptors respectively. An

alternative nomenclature was used by Londos et al. (1980) to classify observations

seen in adipocytes and hepatocytes. Inhibition of adenylate cyclase was denoted by Ri

whilst activation of adenylate cyclase was denoted by Ra, the R signifying that an intact

ribose moiety was required for the modulation of adenylate cyclase activity. However,

it has since been shown that adenosine receptors can bind to effector systems other

than adenylate cyclase so for this reason the A1/A2 nomenclature has now been

adopted as it does not imply modulation of a specific second messenger system.

Further classification of the adenosine receptors has been carried out with an A3

purinoceptor in atrial tissue and on presynaptic sites of frog neurones being proposed

by Ribeiro and Sebastiao (1986). However, this receptor has not been widely

accepted. Another A3 receptor was proposed in 1992 by Zhou et a l based on the

cloning and functional characterisation of a receptor fi-om the rat striatum (Zhou et a l,

1992). This receptor has been cloned from other species such as the sheep and human

(Linden et a l, 1993b; Salvatore et a l, 1993) as well as functional characterisation of

A3 receptors in cultured rat mast cells (Ramkumar et a l, 1993).

16

1.3.1,1. A l receptors

In the central nervous system adenosine acts via pre synaptic Aj receptors to inhibit

the release of neurotransmitters such as noradrenaline, 5HT, dopamine, acetylcholine

and GABA (Stone, 1981; Harms et a l, 1978; Williams, 1984). In the periphery A%

receptors also occur post junctionally mediating cardiac depression (Collis, 1983),

smooth muscle contraction (Bailey et al., 1992), vasoconstriction (Stoggall and Shaw,

1990) and bronchoconstriction (Farmer et ah, 1988), inhibition of renin secretion

(Murray and Churchill, 1984) and inhibition of lipolysis (Londos et al., 1980). From

the original classification of the adenosine receptors, adenosine acting via Aj receptors

inhibit adenylate cyclase activity (Van Calker et a l, 1979; Londos et a l, 1980).

However, it is now known that A receptors can couple to second messenger systems

other than adenylate cyclase for example, activation of potassium channels (Belardinelli

et a l, 1988), inhibition of phospholipase A2 and calcium channels (Schimmel and

Elliot, 1988; Ribeiro and Sebastiao, 1986), stimulation of guanylate cyclase (Kurtz,

1987) and the activation and inhibition of phospholipase C (Hill and Kendall, 1987;

Petcoff and Cooper, 1987).

Currently the principal method of classification of Pi purinoceptors is by determination

of the potency orders of agonists and antagonists and possible structure activity

relationships (SAR). The adenosine structure can be modified in a variety of ways to

produce analogues of differing selectivity and potency for the adenosine receptors. Ai

receptors are preferentially activated by adenosine analogues substituted at the

position of the adenine moiety, for example N^-cyclopentyladenosine (CPA) and N^-

cyclohexyladenosine (CHA) (Bruns et a l, 1986). N^-(phenylisopropyl)adenosine

(PIA) can be created by substituting the phenylisopropyl stereoisomers of

amphetamine at the position of adenosine to produce R- and S- isomers of PIA,

both of which are useful in classification of the purinoceptors due to the high potency

of R-PIA and the 10 to 100 fold stereoselective preference for R-PIA over S-PIA at

17

the Aj receptors (Collis, 1983; Bumstock and Buckley, 1985). A potency order for

various adenosine analogues at the Aj receptor can be seen in Table 1, where CPA is

the most potent and selective Aj agonist (Kj = 0.59nM, 784 and 407 fold selectivity

for the Aj receptor versus A2 a and A3 receptors respectively, Bmns et ah, 1986; Van

Galen et al., 1994). The stmcture of some adenosine agonists can be seen in Figure 2.

L3.1.2. A2 receptors

A2 receptors occur post junctionally mediating relaxation of smooth muscleCvJivN R/)

(Bumstock et ah, 1984;|NichoUs et al., 1992)f vasorelaxation (Rose’meyer and Hope, / ^

/( 1990; Martin, 1992)i bronchorelaxation (Brown and Collis, 1982) and inhibition of

platelet aggregation (Dionisotti et al., 1992).

Although modification of the ribose moiety of adenosine is not well tolerated by the

purinoceptors, substitution at the 5' position appears to be an exception. For example,

one such analogue, 5'-(N-ethylcarboxamido)adenosine (NECA) has a high potency at

A2 receptors and moderate potency at Aj receptors, and although it is ofl:en used as an

A2 selective agonist it is actually non selective for all the Pj purinoceptor subtypes

(Bmns et ah, 1986; Van Galen et ah, 1994, for stmcture see Figure 2). N^ substituted

adenosine analogues such as CPA and R-PIA have little activity at this receptor,

although an exception is N^-[2-(3,5-dimethyloxyphenyl)-2-(2-methylphenyl)ethyl]

adenosine (DPMA) which is a potent A2 agonist, and the stereoisomer R-PIA is only

1-10 times more potent than S-PIA at the A2 receptor (Bridges et a l, 1988; Brown

and Collis, 1982; Bumstock and Buckley, 1985). Substitution at the C2 position of

the adenine moiety can increase the potency of adenosine analogues for the A2

receptor. Increasing the size of this substituent to include cyclic and longer chain

stmctures creates analogues such as 2-phenylaminoadenosine (CV 1808) and 2 -[p-

(carbonylethyl)phenylethylamino)-5'-(N-carboxamido)adenosine (CGS 21680) which

18

Figure 2. Structures of some adenosine receptor agonists

RiNH

N

N R2

L x N

HO OH

AGONIST

CHA

Ri

CPA

NECA

CGS 21680

APNEA

H

H

NH2

R2

CH2OH

CH2OH

CONHC2H5

CONHC2H5

CH2OH

H

H

H

HO2C

H

NH

19

have been shown to be potent and selective agonists (CGS 21680 Kj = 15nM, 140

and 39 fold selective for the A%a receptor versus A% and A3 receptors respectively,

Bruns et a l, 1986; Hutchinson et a l, 1989; Van Galen et a l, 1994). A potency order

for various adenosine analogues at the A2 receptor can be seen in Table 1, and the

structure of some of these analogues can be seen in Figure 2.

1.3.1.3. Â2 receptor suhtypes

The A2 receptor was subdivided by Daly et a l (1983) into A2 a and A2b receptors

according to a difference in affinity for adenosine receptor analogues. A high affinity

binding site (A2 a) for adenosine was reported in the rat striatal membrane (in the

presence of an A% selective agonist CPA to preclude binding) whilst a low affinity

site (A2b) was reported on intact brain cells and human fibroblasts (Daly et al., 1983;

Bruns et a l, 1986; Bruns, 1980a). Agonists acting through both receptor subtypes

stimulate adenylate cyclase, but it was proposed that A2 a receptors were responsible

for activation of adenylate cyclase in striatal homogenates whilst stimulation of the A2b

receptors may elevate general cAMP levels in the brain and brain slices. Using ligand

potency orders it appears that the A2 a receptor has a relatively higher affinity for

agonists and a relatively lower affinity for antagonists than the A2b receptor (Bruns et

a l, 1986). Although there are no selective agonists for the A2b receptor, NECA has a

high potency, 2-chloroadenosine (2-CADO) moderate potency, and the N^ substituted

analogues low potency at this receptor (Table 1). Agonists with bulky C2 position

substituents i.e. CV 1808 and CGS 21680 have a very low affinity for the A2b receptor

and a high affinity for the A2 a receptor, for example, in the guinea-pig aorta (which

contains A2b receptors) CGS 21680 has a p[A5o] of 3.61 whilst in the guinea-pig

langendorff heart preparation (containing A2a receptors) a p[A5o] of 9.13 was

produced (Martin, 1992; Ueeda et a l, 1991a). As CGS 21680 binding and activation

of adenylate cyclase occurs mainly in the striatum, CGS 21680 is

20

IJC/DI<D

■Ba

IIQit o

IIIt oIoPÛH

I

II

H

00AONON

irT

sA

gQ

oc / 3 0 0

A NO

<

gC/3O

% OAl A<

OOoOO

>A U< A

5

HPhc/3OO

gQAONON

in

g

OOo00

1:z;AlO00NO

CNC/3ÜU

0II

1A

(0\O n

in

gAIHPLc/3OOII

gQ

c/3A<

oOO NO

ÿ 8< Al tr' o®S I

r '

A

III

<

oOONO

(N

A

21

considered to be an A2 a selective agonist and is often used to distinguish between the

A%a and A2b receptor subtypes (Jarvis et al., 1989; Hutchinson et a l, 1989). The

physiological and functional relevance of this receptor division is not yet known but

may become clearer with the use of selective agonists and antagonists to distinguish

between the receptors on various tissues and correlate them to the functional effects

observed.

L 3.1.4. As receptors

A third receptor subtype has been proposed by Ribeiro and Sebastiao (1986), although

this subtype has not been widely accepted. Based on a review of the existing literature

they called this receptor, the A3 receptor. They observed that in atrial tissue and at

presynaptic sites in peripheral and central neurones, especially in the frog, purines

exhibit an agonist potency order which does not fit the general K\l A2 receptor

classification. At A% receptors, substituted analogues such as R-PIA have a

greater potency than 5' substituted analogues such as NECA, at A2 receptors NECA

has a greater potency than R-PIA, but at these putative A3 receptors, the analogues

appear to be equipotent. Ribeiro and Sebastiao (1986) also proposed that this receptor

may not be linked to adenylate cyclase but to calcium channels and may be involved in

the inhibition of calcium influx or mobilisation. Kennedy et aï. (1992a) measured the

relative potencies of several adenosine agonists mediating inhibition of lipolysis in the

rat adipocyte (A% mediated response), inhibition of neuronally mediated contraction of

the guinea-pig ilieum (proposed A3 mediated response) and negative chronotropic and

inotropic effects in the guinea-pig and rat atria (proposed A3 mediated response). It

was found that the rank order of agonists was the same in all of these preparations

(CPA > CHA = R-PIA > NECA > S-PIA) and that the agonists had similar potencies

relative to NECA over all the tissues, suggesting that the receptors in the ilieum and

atria, which were proposed to be A3 receptors, are identical to the Ai receptors in the

22

adipocytes. It is known that A% receptors can couple with second messenger systems

other than adenylate cyclase so different agonist potency orders and effector systems

does not preclude this proposed A3 receptor from being of the A% subtype.

Recently, Zhou et al. (1992) reported the cloning and functional characterisation of a

receptor from the rat striatum which has also been called the A3 receptor, although this ,

receptor is different to that proposed by Ribeiro and Sebastiao (1986). This receptor

is expressed in high levels in the testes and lower levels in the lung, kidneys, heart and

the cortex, striatum and olfactory bulb of the central nervous system (Zhou et al.,

1992). Agonists such as R-PIA, NECA and the A selective agonist N^-2-(4-amino-3-

phenyl)-ethyladenosine (APNEA) have high potencies at this receptor with a 6 -fold

enantiomeric preference for R-PIA versus S-PIA (Zhou et a l, 1992; Van Galen et a l,

1994). Further development of A3 agonists has produced N^-benzyl substituted

analogues such as N^-benzyl-5’-(N-ethylcarboxamido)adenosine (Benzyl-NECA)

which has a 13 and 14 fold selectivity for the A3 receptor versus the Aj and A2

receptors respectively, and N^-(iodobenzyl)-5’-(N-methylcarboxamido)adenosine (IB-

MECA) which is also a very selective (50 fold) A3 agonist (Gallo-Rodriguez et a l,

1994; Van Galen et a l, 1994). Substitution at the C2 position of this analogue

produces another analogue 2-chloro- N^-(iodobenzyl)-5’-(N-methylcarboxamido)

adenosine (2-Cl-IB-MECA) which is the most potent and selective A3 receptor agonist

to date displaying a Kj of 0.33nM in binding in transfected CHO cells using [l^^I]-AB-

MECA (N^-[1^^I]- 4-amino-3-(iodobenzyl)-5’-(N-methylcarboxamido) adenosine) and

is selective for the A3 receptor versus the A% and A2 receptors by 2500 and 1400 fold

respectively (Kim et a l, 1994a). The rat A3 receptor appears to be insensitive to the

classical xanthine antagonists such as 8 -p-(sulphophenyl)theophylline (8 -SPT), DPCPX

and xanthine amine congener (8-{4-[({[(2-aminoethyl)amino]carbonyl}methyl)oxy]-

phenyl}-1,3-dipropylxanthine, XAC) (Zhou et a l, 1992). This receptor was found to

be identical to a receptor clone called tgpcrl previously cloned from a rat testes cDNA

library (Meyerhof et a l, 1991). Further A3 receptors have been cloned from a sheep

23

pars tuberalis cDNA library and human striatum (Linden et a l, 1993b; Salvatore et a l,

1993). However, the tissue distribution of these receptors differs from that of the rat

A3 receptor in that they are more widespread. In the sheep high levels were expressed

in the lung, spleen, pars tuberalis and pineal gland, and moderate levels in the testes,

kidney and brain, whilst in the human high levels of the A3 receptor are expressed in

the lung and Hver, moderate levels in the brain and aorta, and low levels in the testes

and heart (Linden et a l, 1993b; Salvatore et a l, 1993). Whilst the agonist potencies

for the sheep and human A3 receptors remain the same as for the rat A3 receptor, both

of these receptors can bind DPCPX and XAC with low affinity, and also have the

ability to bind some acidic substituted 8 -phenykanthines such as 3-(-iodo-4-

aminobenzyl)-8-(4-oxyacetate)-1 -propylxanthine (BW-A522, I-ABOPX) with

nanomolar affinity (Linden et a l, 1993b). The rat, sheep and human A3 receptors have

been shown to inhibit forskolin stimulated cAMP production in Chinese Hamster ovary

(CHO) cells (Zhou e/ a l, 1992; Linden et a l, 1993b; Salvatore et a l, 1993).

Recently, Ramkumar et a l (1993) reported the presence of A3 receptors on cultured

mast cells (RBL-2HL). This receptor has the same binding profile (R-PIA = NECA >

S-PIA) as the cloned A3 receptor fi’om the rat striatum as well as the low sensitivity to

xanthine antagonists. Activation of this cloned receptor stimulates a transient

production of inositol 1,4,5-triphosphate leading to a transient increase in the

intracellular calcium levels. Functionally this results in potentiation of the secretory

response to antigens in the mast cells (Ramkumar et a l, 1993). Studies carried out in

the angiotensin-n supported circulation of the pithed rat indicated that a decrease in

blood pressure induced by adenosine agonists including APNEA are resistant to

blockade by xanthine antagonists such as 8 -SPT and DPCPX (Fozard and Carruthers,

1993a; Carruthers and Fozard, 1992; 1993). This was attributed to the activation of

A3 receptors within the cardiovascular system which mediate hypotension. However,

it was found that the hypotensive response induced by APNEA could be inhibited by

the mast cell stablising agent sodium cromoglycate and was decreased after repeated

doses of the mast call degranulating agent compound 48/80. Therefore it was

24

concluded that A3 receptor mediated, APNEA induced hypotension was due to the

activation of A3 receptors present on the mast cells (Hannon et al., 1995). Further

evidence to support the involvement of A3 receptors was found using BW-A522 (I-

ABOPX) to block the hypotensive response of APNEA in pithed rats, although the

potency of BW-A522 as an antagonist is lower at the rat A3 receptor than at the sheep

or human A3 receptors (Fozard and Hannon, 1994). It appears that the properties of

this cloned receptor are distinct from those of the A% and A2 subtypes. Continued

emergence of functional roles for the A3 receptor will further strengthen the

classification of this receptor.

i. 3,1.5. receptors

An A4 receptor has been recently proposed by Cornfield et al. (1992) based on a

unique agonist potency order of CV 1808 > CGS 22988 » NECA > CGS 21680 for

[^H] CV 1808 competition binding in the rat striatum at 4°C. However, it has since

been shown that if binding is carried out at 21°C then the agonist potency order

becomes characteristic of an A2 receptor (Luthin and Linden, 1995). It would appear

that the A4 receptor is a temperature sensitive form of the A2 receptor producing a

different potency order at 4°C because CGS 21680 loses its affinity to the A2 a receptor

at this temperature (Luthin and Linden, 1995).

1.3.2. The P site

In vitro, high concentrations of adenosine (mM) inhibit the activation of adenylate

cyclase via the P site (Londos and WolfiF, 1977). This appears to be an intracellular

site on the catalytic subunit of adenylate cyclase which requires a purine moiety for

modulation, hence the term P site. Agonists include adenosine, 2-CADO and 2', 5'-

dideoxyadenosine but the 5'- and N^- substituted analogues NECA, 5'-N-cyclopropyl-

carboxamidoadenosine (CCPA), R-PIA and CHA are not agonists (Daly, 1982). The

25

p site is not susceptible to blockade by xanthines such as 8 -SPT but its activation in

intact cells can be attenuated by adenosine transport inhibitors such as dipyridamole

(Collis and Brown, 1983). Since high concentrations of adenosine are required to

activate this site it is unlikely that it is physiologically significant and therefore is

referred to as a site rather than a receptor.

1.3.3. Adenosine receptor antagonists

In 1970 Sattin and Rail demonstrated that the naturally occurring xanthines caffeine

and theophylline blocked the adenosine stimulated accumulation of cAMP in brain

slices. This was the first evidence to show that although xanthines can act as

phosphodiesterase inhibitors their main action is considered to be antagonism of

adenosine effects (Choi et a l, 1988). Caffeine and theophylline are non selective for

the A \ and A2 receptor but attempts have been made to produce more selective and

potent antagonists based on the xanthine structure. Bruns (1980b) suggested that

xanthine antagonists act at the same site as the adenosine agonists but probably in a

different orientation. In 1986 Bruns et al. developed this suggestion further by noting

that the 5 membered ring of the xanthine lies in an equivalent position to the 6

membered ring of the adenosine structure in the purinoceptor (Bruns et al., 1986).

Based on the observation that at the Aj receptor the structure activity relationships for

substituted adenosine agonists were similar to those obtained for the 8 -substituted

xanthine antagonists, Peet et al. (1990) suggested the " N^-C8 " model where the

and C8 positions of the agonists and antagonists occupy the same binding pocket in the

A% receptor.

Substitution at the 8 -position of the xanthine structure with a wide range of chemical

groups increases the potency of the adenosine antagonists considerably, as shown by

the increase in potency of 8 -phenyltheophylline (8 -PT) from theophylline, however this

analogue has poor water solubility which has limited its use (Bruns and Fergus, 1989).

26

The addition of a sulphur group to produce 8 -SPT increases the water solubility of the

analogue but as with 8 -PT this compound appears to be non-selective for A \ and A2

receptors (Collis et a l, 1987).

1.3.3.1, AI receptor antagonists

Further development of 8 -substituted xanthines produced antagonists with some Ai

selectivity, for example DPCPX has been shown to have a consistent and marked

selectivity for the A \ receptor in a range of preparations and species. In the guinea-pig

atria, which contains A% receptors, and the guinea-pig aorta, which contains A2

receptors, the non-selective antagonist 8 -SPT produces pA2 values of 4.9 and 5.0

respectively (Collis et a l, 1989). These are significantly different to the pA2 values

produced by DPCPX in these tissues. In the guinea-pig atria, DPCPX produces a pA2

value of 7.9 whereas in the guinea-pig aorta a pA2 value of 6 . 6 has been shown, hence

DPCPX can be used to differentiate between A% and A2 receptors (Collis et a l, 1989).

The structure of some xanthine antagonists can be seen in Figure 3 a.

1.3.3.2. A2 receptor antagonists

There are several non-xanthine analogues such as the triazoloquinazoline CP 66713

and the triazoloquinoxaline CGS 15943 which show some selectivity for the A2 a

receptor (Nikodijevic et a l, 1991; Ghai et a l, 1987). However, due to limited

selectivity for the A i receptor versus the A \ receptor (approximately 7 fold for CGS

15943), poor water solubility and the possibility that CGS 15943 may act non

competitively at some A2 a receptors, these antagonists are not used routinely

(Williams et a l, 1987a). Another non-xanthine antagonist that has recently been

described is ZM241385 (4-(2-[7-amino-2-(2-furyl)[l,2,4]triazolo[2,3-a][l,3,5]triazin-

5-yl)amino]ethyl)phenol). This analogue displays a high affinity for the A2 a receptor

(pA2 of 8.57) and appears to be the most selective A2 a receptor antagonist with

27

selectivities of 400-1000 fold, 30-80 fold and 6700 fold for the A \, A2 b and A3

receptors respectively (Poucher et al., 1995). The structures of CGS 15943 and

ZM241385 can be seen in Figure 3b. The xanthine analogue PD 115,199 (N-[2-

(dimethylamino)ethyl]-N-methyl-4-(2,3,6,7-tetrahydro-2,6-dioxo-1,3 -dipropyl-1H-

purin-8 -yl)benzene sulphonamide) has been reported to have equal affinity at the A \

and A2 a receptors in binding studies in the rat striatum producing Kj values of 13.9

and 15.5 nM respectively (Bruns et al., 1987). It has been also reported that PD

15,199 binds to the high affinity (A2 a) sites in the striatum whilst no binding was

observed at the low affinity (A2 b) sites, suggesting that this compound is selective for

the A2 a receptor versus the A2 b receptor but can not distinguish between the A2 a and

A% receptors (Bruns et al., 1987). This is fiirther supported by the finding that in the

dog saphenous vein and guinea-pig aorta, which contain A2 b receptors, PD 115,199

has a low affinity (ED3 0 of 151 and 275 nM respectively) compared to the high affinity

at the A2 a receptors in binding to the rat striatal membranes (IC5 0 of 16nM)

(Hargreaves et al., 1991; Jarvis et al., 1989). Recently it has been reported that the

introduction of 8 -styryl groups in the xanthine structure increases the potency and

selectivity of the antagonists for the A2 a receptor (Shimada et a l, 1992; Jacobson et

a l, 1993). One such compound, KF 1737 ((E)-1,1 -dipropyl-7-methyl-8-(3,4-

dimethoxystyryl)xanthine) has been shown to be a highly selective A2 a antagonist both

in vitro and in vivo, although this antagonist may have limited use due to its rapid

photoisomerisation when diluted in dimethylsulphoxide (DMSG) and exposed to light

(Jackson et a l, 1993; Nonaka et a l, 1994). CSC (8-(3-chlorostyryl)caffeine) is

another 8 -styryl substituted compound which has a 520 fold selectivity for the A2 a

receptor versus the A \ receptor in binding studies in the rat brain but only 22 fold

selectivity for the A2 a receptor in functional studies in rat pheochromocytoma (PC 12)

cells (Jacobson et a l, 1993).

28

Figure 3. Structure of some xanthine and non-xanthine adenosine antagonists

a) XANTHINE

NH

O

R?

ANTAGONIST Ri

8 -SPT

DPCPX

CH.

C3 H7

PD 115199 C3 H7

Ro

CH.

C3 H7

C3 H7

Rq

SOH3

y ■S02N(CH3)(CH2)2N(CH3)2

I-ABOPX C3 H7 — ^ ^ N H 2 ---- ^ ^ OCH2COO-

b) NON-XANTHINE

CGS 15943

Cl

NH;

ZM 241385

29

1.3.3,3. receptor antagonists

The rat A3 receptor was originally classified by its insensitivity to classical xanthine

antagonists such as 8 -SPT, DPCPX and XAC but some acidic substituted 8 -

phenylxanthines such as BW-A522 (I-ABOPX) and BW-A1433 have been shown to

be more effective antagonists at the A3 receptor (Zhou et a l, 1992; Linden et a l,

1993b). Kim et a l (1994b) have recently produced afidnity values in the micromolar

range for some antagonists at the rat A3 receptor producing an antagonist order of

BW-A522 (I-ABOPX) > DPCPX > BW-A1433 > XAC » CSC, where BW-A522 is

the most effective antagonist with an afiinity of 1.17pM whilst CSC was inactive (Kim

et a l, 1994b). In the rat, BW-A522 was 32 and 2 fold more selective for the A \ and

A2 a receptors versus the A3 receptor whilst BW-A1433 was 1000 fold and 19 fold

more selective for the A% and A2 a receptors versus the A3 receptor, therefore both of

these analogues are not selective for the rat A3 receptor (Kim et ah, 1994b). BW-

A522 and BW-A1433 are more potent at the sheep and human A3 receptors than at

the rat A3 receptor producing affinities of 3nM and 21nM in the sheep and 18nM and

55nM in the human respectively (Linden et a l, 1993b; Salvatore et a l, 1993). The

antagonist potency order produced in the sheep and human was BW-A522 > BW-

A1433 > XAC » DPCPX (Linden et a l, 1993b; Salvatore et a l, 1993). The structure

of I-ABOPX (BW-A522) can be seen in Figure 3 a.

1.3.4. Molecular biology studies

In recent years molecular biology techniques have provided another approach to the

study of adenosine receptors and their structure and distribution, although by their

nature these studies focus mainly on the structural properties of the purinoceptors

rather than their functional aspects. The division of adenosine purinoceptors on a

30

pharmacological basis into Ai, A2 and A3 subtypes has now been supported by

evidence produced using molecular biology techniques.

1.3.4.1. AI receptors

In 1989, Libert et al. began the interest in the cloning of adenosine receptors by

cloning two orphan receptors, RDC7 and RDC8 from a canine cDNA library whilst

screening for sequences homologous to transmembrane sections of other G protein-

coupled receptors (Libert et a l, 1989). Libert et a/. (1991) went on to identify RDC7

as the A \ receptor based on the distribution of the RDC7 transcript in the cortex,

hippocampus, thyroid gland and testes using in situ hybridisation, and functional data

such as rank order of agonist affinities in radioligand binding and inhibition of forskolin

stimulated cAMP production (Libert et a l, 1991; 1992). The A] receptor has also

been cloned from several other species, for example, Reppert et al. (1991) and Mahan

et a/. (1991) have cloned a rat A \ receptor which has 91% homology to the canine A \

receptor. The bovine A \ receptor has greater than 90% homology to the rat and

canine A \ receptors but displays a different agonist potency order of R-PIA > S-PIA >

NECA compared to the "classical" A \ receptor potency order of R-PIA > NECA > S-

PIA found in the rat and dog (Tucker et a l, 1992; Olah et a l, 1992; Linden et al,

1993a). Libert et al. (1992) also described the cloning of a human A \ receptor from a

hippocampus cDNA hbrary, whilst Salvatore et al. (1992) cloned a human A \ receptor

from ventricle, brain and kidney cDNA libraries. This clone encodes a protein of 326

amino acids in length and is 94% homologous to the canine, rat and bovine A \

receptors (Salvatore e ta l, 1992; Jacobson, 1995).

31

1.3.4.2. A 2a receptors

The striatal distribution of RDC8 , binding of [3r]NECA and [3h]CGS 21680 in COS-

7 cells and the activation of adenylate cyclase when RDC8 is expressed in transfected

cells such as Xenopus oocytes led to RDC8 being identified as the A%a receptor

(SchifiBnann et al., 1990; Maenhaut et al., 1990). A rat A2 a receptor was identified

by Chem et al. (1992) and Fink et al. (1992) which was 82% homologous to the

canine A2 a receptor (RDC8 ). Using in situ hybridisation Fink et al. (1992) also

showed that the A2 a adenosine receptor was co-expressed in the same striatal neurons

as the dopamine D2 receptors, but not the D\ receptors indicating a possible reason

for the interaction between these two receptor systems in the control of some motor

functions (Jarvis and Williams, 1987). The human A2 a receptor has been cloned fi'om

a hippocampal cDNA library and a heart ventricle and striatal cDNA library (Furlong

et al., 1992; Salvatore et al., 1992). This clone encodes a protein of 412 amino acids

with 93% homology to the canine RDC8 receptor but only 84% homology to the rat

A2 a receptor (Jacobson, 1995).

1.3.4.3. A2J) receptor

A protein (RFL9) has been cloned from the rat brain which has 45% and 46%

homology with A \ and A2 a adenosine receptors respectively, although RFL9 has a

different distribution to the Aj and A2 a receptors as it is found in high levels in the

large intestine, caecum and bladder. (Stehle et a l, 1992). Transfection of C0S-6M

cells with RFL9 resulted in the binding of NECA with high affinity, but not the A] and

A2 a selective agonists CCPA and CGS 21680 and also NECA stimulated the

production of cAMP whereas CGS 21680 did not (Stehle et a l, 1992). These

properties would suggest that this protein is an A2 b receptor. Further evidence was

provided by the finding that in Chinese hamster ovary (CHO) cells transfected with

32

RFL9 the stimulation of cAMP production by adenosine agonists and their inhibition

by the adenosine antagonists correlated highly with the responses seen in the human

VA13 fibroblasts which have been shown to contain A2 b receptors (Rivkees and

Reppert, 1992; Bruns et al., 1986). The human A2 b receptor was cloned by Pierce et

al. (1992) from a hippocampal cDNA library. This clone encodes a protein of 328

amino acids in length with 8 6 % homology to the rat A2 t> receptor and 45% homology

to the human A% and A2 a receptors (Pierce et al., 1992). When the hippocampus

clone was expressed in CHO cells similar pharmacological properties to the rat A2 b

receptor were seen with respect to agonist binding and stimulation of cAMP

production (Pierce et al., 1992).

I.3.4.4. receptor

This receptor was originally identified by molecular biology cloning techniques rather

than by pharmacological fianctional studies. Meyerhof et al. (1991) cloned an orphan

receptor tgpcrl fi-om a rat testes cDNA library which had 47% and 42% homology to

the RDC7 and RDC8 receptors respectively. In 1992, Zhou et al. cloned an identical

protein (R226) fi'om the rat striatum with 58% homology to the A \ and A2 receptors

vrithin the transmembrane regions of the receptors, and this has subsequently been

called the A3 receptor (Zhou et al., 1992). The A3 receptor has also been cloned from

a sheep pars tuberalis cDNA library (SI7) obtaining 72 % homology to the rat A3

receptor and fi'om a human striatal cDNA library (HS-21a) with a sequence that has

72% homology to the rat A3 receptor and 85% homology to the sheep A3 receptor

(Linden et al., 1993b; Salvatore et al., 1993). As mentioned previously there is a

difference in the distribution and affinity of xanthine antagonists between the rat and

the sheep and human A3 receptors. This difference can also be seen in the receptor

sequences with a higher similarity (85%) between the sheep and human A3 receptors

compared to 74% between the rat A3 receptor and the sheep and human A3 receptors.

Another human clone (pl2H) has been produced from a heart cDNA library using the

33

rat A3 sequence, although this receptor only has 71% homology to the rat A3 receptor

(Sajjadi and Firestein, 1993). The proteins that these clones encode vary slightly in

length, for example, the rat A3 receptor is 320 amino acids in length, the sheep A3

receptor is 317 amino acids in length and the human striatal and heart A3 receptors are

318 and 319 amino acids respectively in length (Zhou et al., 1992; Linden et al.,

1993b; Salvatore a/., 1993; Sajjadi and Firestein, 1993).

1.4. ACTIONS OF ADENOSINE ON THE CARDIOVASCULAR SYSTEM

Adenosine and its analogues have been shown to modulate the fimctions of several

physiological systems including the cardiovascular, respiratory, gastrointestinal, central

nervous system and immunological systems. The actions of adenosine in the

cardiovascular system will be reviewed here but its actions on other systems are

described elsewhere in reviews by Jacobson, (1990) and Collis and Hourani, (1993).

1.4.1. Al receptor

Adenosine has negative dromotropic, chronotropic and inotropic actions on the heart.

Adenosine acts on the atrioventricular (AY) and sinoatrial (SA) nodes to cause a

decrease in the impulse conduction time through the AY node (heart block) and a

decrease in the pacemaker firing rate (bradycardia). Based on the agonist potency

order of CPA > R-PIA > NECA > 2-CADO it would appear that both of these effects

are due to the activation of A% receptors and also probably involve an increase in

potassium conductance (Clemo and Belardinelli, 1986; Belardinelli etal., 1988; Evans

et al., 1982). Adenosine also acts on the atria to produce negative inotropic effects.

The decrease in force of contraction caused by adenosine is again mediated via the Aj

receptor and involves a decrease in the action potential and an increase in the

potassium conductance (Evans et al., 1982; Collis, 1983; Belardinelli and Isenberg,

1983a; Jahwel and Nawath, 1989). Adenosine has little direct effect on the

34

contractility of the ventricular myocardium, however it has an indirect effect by

antagonising the positive inotropic effects of catecholamines. This anti-adrenergic

effect is achieved by the inhibition of the increase in cAMP and the slow inward

calcium current induced by the catecholamines (Belardinelli and Isenberg, 1983b)

1.4.2. A% receptor

Adenosine and its analogues induce vasodilation in a variety of blood vessels within the

cardiovascular system (Table 2). Agonist potency orders such as NECA > R-PIA> 2-

CADO > adenosine suggests that A% purinoceptors mediate vasodilation in vascular

tissues (Kusachi et al., 1983; King et al., 1990). Recently with the use of the A2 a

selective agonist CGS 21680 some of the A2 receptors found on vascular tissues have

been further subdivided into A2 a or A2b receptors. For example, the A2 receptors on

the dog coronary artery have been classified as A2 a receptors (Gurden and Kennedy,

1992; Kennedy et al., 1992b), whilst those on the guinea-pig aorta have been classified

as A2b receptors (Losinski et al., 1993; Hargreaves et al., 1991) due to the respective

high and low potency of CGS 21680 on these tissues. Further studies with xanthine

antagonists have also supported the proposal that A2 receptors mediate vasodilation.

For example, a pA2 value of 6 . 6 was reported in the guinea-pig aorta using DPCPX

indicating that vasodilation in this tissue was mediated by A2 receptors (Collis et al.,

1989), whilst Losinski et al. (1993) reported that PD 115,199 (an A2 a selective

antagonist) had a low affinity on this tissue confirming that the purinoceptor present

was probably of the A2b subtype.

Until recently the location of the A2 receptor was assumed to be on the vascular

smooth muscle with vasodilation achieved by direct action on these purinoceptors, for

example, endothelium independent vasodilations were seen in tissues such as guinea-

pig aorta, rat femoral artery, rabbit mesenteric artery and pig coronary artery (Collis

and Brown, 1983; Kennedy et al., 1985; Mathieson and Burnstock, 1985; King et al.,

35

1990). However, there are examples where adenosine acts on A2 receptors on the

endothelium where its effects can be abolished or attenuated by the removal of the

endothelium by mechanical means (Gordon and Martin, 1983; Kennedy and Burnstock,

1985; Rubanyl and Vanhoutte, 1985). Liang (1992) put forward a working model

suggesting that under basal conditions when endogenous adenosine concentrations are

low, vasodilation is induced via the endothelial A2 receptors producing endothelium

dependent vasodilation. During hypoxia when large concentrations of adenosine are

produced, it is possible that adenosine can pass through the endothelium and act on the

vascular smooth muscle A2 receptors inducing vasodilation by both endothelium

dependent and independent means.

The mechanism of action of endothelium dependent relaxation is thought to involve the

release of Endothelium Derived Relaxing Factor (EDRF) from the endothelium. Due

to similar biological and chemical properties of EDRF and nitric oxide (NO), it has

been concluded that EDRF and NO are identical (Palmer et a l, 1987; 1988; Ignarro et

a l, 1987). NO is synthesised by the conversion of L-arginine to L-citrulline by an

enzyme called NO synthase and although many isoforms of this enzyme exist, the

isoform present in the endothelial cells is a soluble, constitutive form which is regulated

by calcium/calmodulin and requires NADPH as a cofactor (Palmer et a l, 1988; Palmer

and Moncada, 1989). Another isoform of nitric oxide synthase is also present on the

vascular smooth muscle cells. This is a calcium independent inducible form which is

induced by bacterial products and cytokines but also requires NADPH as a cofactor

(for review see Moncada et a l, 1991). Research into the involvement of NO in the

relaxation of vascular smooth muscle has benefited from the production of NO

synthase inhibitors. Analogues of arginine such as N^-monomethyl-L-arginine (L-

NMMA) and N^-L-arginine methyl ester (L-NAME) are competitive stereospecifrc

inhibitors of NO synthase, although the inhibition can be reversed by the addition of

arginine (Rees etal., 1989; 1990)

36

Table 2. The classification and endothelium dependency of adenosine receptors in theperipheral arteries of a variety of species

ARTERY and RECEPTOR ENDOTHELIUM REFERENCESPECIES DEPENDENCY

C o r o n a r y a r t e r y

Rat Pi - Fleetwood & Gordon (1987)^ 2 - Hamilton et al., (1987)^ 2 - Oei et al., (1988)^ 2 a - Hutchinson et al., (1989)

Guinea-pig Pi Independent Keef et al, (1992)A2 (A2 a) Partial Vials & Burnstock (1993)

A2 - Martin et al, (1993),Canine Pi Independent White & Angus (1987)

A2 ' Kusachi gr a/., (1983)A2 a - Gurden & Kennedy (1992)A2 a - Kennedy et a l, (1992b)A2 a - Croning et a l, (1992)

Human A2 Independent Sabouni e ta l , (1990)A2 - Ramagopal et a l, (1988)A2 a - Makujina et a l, (1992)

Porcine A2 Independent King e ta l , (1990)A2 Independent Makujina et a l, (1993)

A2 (A2 a) Partial Abebe et al, (1994)A2 (A2 a) - Balwierczak et a l,{ \99 \)

Bovine A2 - Cushing et a l, {\99\)A2 - Mustafa & Askar (1985)

A2 (A2 a) - Conti et al, (1993)Rabbit A2 - Odwara et a l, (1986)

A o r t a

Rat A2

A2 (A2 a)A2 (A2 a)

DependentDependent

Partial

Rose'meyer & Hope (1990) Moritoki et a l, (1990) Conti et a l, (1993)Yen et a l, (1988)

Guinea-pig A2

A2

A2 (A2 b)A2 bA2bA2 bA2 bA2bA2 bA2 b

Partial

Independent

Stoggall & Shaw (1990) Headrick & Berne (1990) Collis & Brown (1983) Martin (1992)Alexander et a l, (1994) Gurden & Kennedy (1992) Hargreaves et a l,{ \99 \) Kennedy et a l, (1992b) Losinski et al, (1993) Martin et a l, (1993)

Rabbit A2 - Ghia & Mustafa (1982)

37

Table 2. (Continued)

ARTERY and RECEPTOR ENDOTHELIUM REFERENCESPECIES DEPENDENCY

F e m o r a l a r t e r y

Rat Pi Independent Kennedy et al., (1985)Rabbit Al Independent Cassis et a l, (1987)Canine - Independent De Mey & Vanhoute (1981)

M e s e n t e r ic A r t e r y

Rat Pi Partial Vuorinen et a l, (1992)A2b Independent Rubino et al., (1995)

Rabbit - Independent Mathieson & Burnstock (1985)

H e p a t ic A r t e r y

Rabbit PiA2 (A2 a)

Independent Brizzolara & Burnstock (1991) Mathie et a l,{ \99 \)

C e n t r a l e a r A r t e r y

Rabbit Pi Partial Kennedy & Burnstock (1985)

R e n a l A r t e r y

Rat A2 b Dependent Martin & Potts (1994)

P u l m o n a r y A r t e r y

Rabbit A2 Partial Steinhom et a l, (1994)Human A2 Independent McCormack et al., (1989)Lamb Pi - Konduri et al., (1992)

The classification and endothelium dependency of adenosine receptors in a selection of peripheral arteries in a variety of species. Receptor subtype in parenthesis indicates the possible subtype of A2 receptor that could be implied by the potency of A2 a selective agonists such as CGS 21680 and CV 1808 although this receptor subtype was not stated explicitly by the authors.

38

EDRF is believed to relax vascular smooth muscle by activating soluble guanylate

cyclase, therefore increasing levels of cyclic guanosine monophosphate (cGMP) within

the vascular smooth muscle (Rapoport and Murad, 1983; Waldman and Murad 1987;

Ignarro, 1989). The mechanism for cGMP induced relaxation has not been fully

elucidated. However, one such possible mechanism is the activation of cGMP

dependent protein kinase, and a modification of the phosphorylation of various smooth

muscle proteins for example, the dephosphorylation of myosin light chain has been

associated with vasorelaxation (Rapoport et ah, 1993). Alternative suggestions for the

mechanism of cGMP induced relaxation include decreasing the fi-ee cytosolic calcium

concentration by the inhibition of phospholipase C and therefore inositol phosphate

production, stimulation of Ca^-ATPase and the opening of potassium channels and the

inhibition of calcium influx (Rapoport, 1986; for review see Lincoln, 1989).

Originally, Van Calker et al. (1979) classified A% receptors based on stimulation of

adenylate cyclase by purines. This would suggest that the stimulation of endothelial A%

receptors would increase cAMP. Indeed, Graier et al. (1992) reported that in pig

aortic endothelial cells adenosine amplified bradykinin and ATP induced biosynthesis

of EDRF via an increase in cAMP, cAMP dependent phosphorylation and an increase

in the intracellular calcium concentration. Also in cultured guinea-pig coronary

endothelial cells, purines enhanced the cAMP accumulation with a potency order of

NECA > R-PIA and adenosine indicating the presence of A2 receptors (Des Rosiers

and Nees, 1987). However there are examples where an increase in cAMP is not

associated with vasorelaxation due to the stimulation of receptors. For example, in

the porcine coronary arteries, Herlihy et al. (1976) reported that there was no increase

in cAMP at the adenosine agonist concentrations that induced vasorelaxation via the

stimulation of A2 receptors, whilst Cassis et al. (1987) reported that in the rabbit

femoral artery a transient increase in cAMP was seen prior to, but not during

vasorelaxation. Classification of purinoceptors as an A2 subtype however should not

preclude it from being associated with second messenger systems other than adenylate

39

cyclase. For example, Moritoki et al. (1990) proposed that release of EDRF,

stimulation of guanylate cyclase and an increase in cGMP, rather than cAMP, mediates

vasodilation induced by adenosine via endothelial receptors on the rat thoracic aorta.

GrifiBth et al. (1985) also indicated that cGMP was involved in the endothelium

dependent relaxation of the rabbit aorta. For this reason classification of A2 receptors

on vascular tissues is based on agonist and antagonist potency orders rather than

modulation of second messenger systems.

Although the main action of adenosine on blood vessels is vasodilation via

receptors, in the kidney adenosine not only induces vasodilation of the afferent and

efferent arterioles via the stimulation of A2 receptors, it also induces vasoconstriction

of the afferent arterioles via stimulation of A receptors (Churchill and Churchill,

1985; Murray and Churchill, 1984). This vasoconstriction leads to a decrease in renal

blood flow and glomerular filtration rate and hence a decrease in energy demand as less

solute needs to be reabsorbed into the kidney. This is fiirther evidence for the

hypothesis that adenosine acts as a 'retaliatory metabolite’ (Murray and Churchill,

1984; Berne, 1963; Newby, 1984). Stimulation of A2 receptors by adenosine induces

an increase in renin secretion, whilst the activation of Aj receptors causes an inhibition

of renin secretion (Churchill and Churchill, 1985). Further details of the effects of

adenosine in the kidney can be found in the review by Spielman and Arend (1991).

40

1.5. AIM OF THE THESIS

Adenosine has been shown to mediate vasodilation of many blood vessels in a variety

of species via the activation of adenosine receptors. The aim of the thesis was to study

the ?! purinoceptors present on blood vessels such as the thoracic aorta and the

coronary arteries of the rat to characterise the purinoceptor subtype present and

investigate other properties of the receptors, for example, to determine whether the

adenosine receptors were present on the endothelium or on the vascular smooth

muscle.

The isolated rat thoracic aorta was chosen as an example of a large diameter artery

whilst the rat coronary arteries, studied using the Langendorff isolated heart

preparation, were chosen as an example of small diameter arteries. In both of these

cases adenosine has been shown to produce vasodilation of the vessels indicating the

presence of adenosine receptors, so consequently further investigation of these

receptors can be undertaken (Rose’meyer and Hope, 1990; Fleetwood and Gordon,

1987). The identity of the adenosine receptor subtype mediating the vasodilation of

the thoracic aorta and the coronary arteries of the rat was investigated using a range of

adenosine agonists and antagonists such as NECA, CGS 21680, 8 -SPT, PD 115,199

and DPCPX.

The location of the purinoceptors within the blood vessels was also investigated by

ascertaining whether the presence of endothelium was required for receptor mediated

vasodilation. This could be achieved by inhibition of nitric oxide synthesis using nitric

oxide synthase inhibitors such as L-NAME, or in the case of the thoracic aorta, also by

mechanical denudation of the vessel.

41

CHAPTER 2: METHODS

2.1. PHARMACOLOGICAL METHODS

2.1.1. Rat isolated thoracic aorta

Male Wistar rats (200-250g) obtained from the University of Surrey breeding unit

were killed by cervical dislocation. The thoracic cavity was opened up and the

thoracic aorta dissected out. The aorta was cleaned of connective tissue and a 6 mm

ring was cut from the central portion of the aorta. One ring was obtained from each

aorta. The rings were mounted on 6 mm stainless steel hooks in 4ml organ baths

containing modified Krebs Henseleit solution (mM: NaCl, 118; KCl, 4.7; NaHCOg, 25;

glucose, 11; MgS0 4 .7 H2 0 , 0.45; KH2 PO4 , 1.2; EDTA, 0.03; CaCl2 .2 H2 0 , 2.5),

maintained at 37°C and aerated with 95% O2 : 5% CO2 The tissues were equilibrated

for 1 hour at a resting tension of 2.5g with isometric responses subsequently recorded

with a Grass FT03 force displacement transducer and displayed on a Grass polygraph

(model 79).

A noradrenaline (NA) concentration response curve was obtained to ascertain a

submaximal concentration of NA for contraction of the aortic rings (Figure 4).

Subsequently the aortae were contracted three times with noradrenaline (lOOnM) to

produce a stable contraction. The integrity of the aortic endothelium was assessed by

the addition of acetylcholine (ACh, IpM) and those tissues showing less than 65%

relaxation to ACh were discarded. The endothelium was deliberately removed from

some aortae by gently rubbing the luminal surface with a metal rod and absence of the

endothelium was confirmed by the failure of ACh (IpM) to induce relaxations. To

measure the relaxant effects of the agonists the tissues were incubated with

noradrenaline (lOOnM) for 3-4 minutes until a stable plateau was reached before the

addition of the agonists. Responses were expressed as % inhibition of the NA-induced

contraction. All the agonist concentration response curves were constructed non

cumulatively, with a random order of agonist concentrations and a 2 0 minute

42

Figure 4. Concentration response curve for NA in the rat thoracic aorta.

3 -1

2.5-

2 -

II0.5-

10 100 10000.1 1

NA Concentration (nM)

Contraction of the intact rat thoracic aorta induced by noradrenaline. Each point represents the mean of at least 4 determinations with s.e. mean shown as vertical bars.

43

ECgo values can be defined as the concentration of the agonist producing 30% inhibition of

the NA contraction.

PÀ2 can also be defined as the logarithm of the (dose-ratio -1) minus the logarithm of the

molar concentration of the antagonist i.e. PÂ2 = log (dr-1) - log[A].

wash-out time between doses of noradrenaline. To study the effects of 8 -SPT, PD

115,199, DPCPX, MTA and N^-nitro-L-arginine methyl ester (L-NAME) the tissues

were equilibrated with these compounds for a period of 2 0 minutes prior to the

addition of noradrenaline. Only one concentration response curve was carried out on

each tissue due to the length of time (approamately 7 hours) required to construct one

curve. y

pECgo values (the negative logarithm of the EC3 0 value) were calculated from the

linear portion of individual log. concentration response curves by linear regression

analysis. Potency of the agonists was expressed as EC3 0 values calculated from the

mean of the PEC3 0 values. For the antagonists, the dose-ratios were calculated as the

ratio of the EC3 0 values in the presence and absence of one concentration of the

antagonist and estimated pA% values were calculated as the negative logarithm of the )

(molar concentration of the antagonists divided by the dose-ratio

2.1.2. Langendorff isolated rat heart preparation

Initial attempts to use a Langendorff isolated heart preparation involved using a

‘constant-flow’ system where the perfusion flow through the heart was held constant

and a change in perfusion pressure, measured by a Statham P23ID pressure transducer

gave an indication of the vasodilation induced by the adenosine agonists. The

equipment used was similar to that described later in the ‘constant-pressure’ system,

except the Krebs Henseleit solution was supplied to the junction box above the heart

via a Harvard peristaltic pump and a heating coil. One reason why the ‘constant-flow’

system was discarded was the very small changes in pressure induced by the adenosine

agonists. The size of these changes made reliable measurements of responses very

difficult, and would have made the use of antagonists impractical. It is possible that

the pressure changes were small due to the dampening of the responses by a irbubbles

in the equipment and the overriding pulsatile flow produced by the peristaltic pump

44

which was detected by the pressure transducer. Due to the many problems that arose

with the ‘constant-flow’ system it was decided to convert this equipment to a

‘constant-pressure’ isolated heart system which was then used for the rest of this study

and has been described below.

Male Wistar rats (280-3 80g) obtained from the University of Surrey breeding unit

were injected with 500 Units heparin i.v. via the tail vein 5 minutes prior to being

killed by cervical dislocation. The heart was rapidly excised from the thoracic cavity

and plunged into ice cold Krebs Henseleit solution (mM; NaCl, 118; KCl, 4.7;

NaHCOs, 25; glucose, 11; MgS0 4 .7 H2 0 , 1.17; KH2PO4 , 1.2; CaCl2 .2 H2 0 , 2.5). The

heart was subsequently mounted on a cannula via the ascending aorta and the cannula

suspended from the junction box while excess lung and connective tissue were

removed (see Figure 5 for diagram of equipment). The heart was perfused with the

Krebs Henseleit solution at a temperature of 37°C (measured at the junction box at the

point of entry to the cannula) and aerated with 95% 0% : 5% CO2 from a reservoir

which was at a height of Im above the heart. The perfusion of the heart was

maintained at a constant pressure of 75mmHg which was continuously measured by a

Statham P23ID pressure transducer connected to the junction box immediately above

the cannulated aorta. A Harvard Minipuls pump was used to replenish the Krebs

Henseleit solution within the reservoir to maintain the required pressure. A heart clip

was attached to the apex of the heart and the heart was then left to equilibrate for 2 0

minutes under a tension of 3g with isometric responses subsequently recorded with a

Grass FT03 force displacement transducer to measure the force of contraction of the

heart. This signal was integrated by a Grass tachograph to produce a measurement of

the heart rate. To constrain the heart to a constant rate, silver plated needle electrodes

were placed on the apex of the heart close to the heart clip and on the right ventricle

near to the right atrium and the heart was stimulated using a Grass S48 stimulator with

parameters of 5 volts, 2ms delay and a frequency of 4Hz to produce a heart rate of 240

45

Figures. Schematic diagram of the isolated heart equipment

Im

Minipuls pump

95% O2 : 5% CO2 ( T

to Circulator <="

from Circulator

Krebs Henseleit solution

Pressure transducer

to Grass recorder

Agonist infusion pump

Junction box

Electrodes/—S/Heart fromstimulator

to Grass recorder

Force transducer

Q_0Pulley system

46

beats/minute. The pressure, force and rate measurements were displayed on a Grass

polygraph (model 79).

All the agonists were infused into the heart, using a Harvard variable speed infusion

pump via a rubber septum in the junction box, at a rate which was 1 / 1 0 0 th of the basal

perfusion flow (control flow) through the heart. The coronary flow was measured

volumetrically by collecting the coronary effluent in a tared beaker beneath the heart.

In initial experiments the basal perfiision flow (control flow) through the heart was

calculated as the mean of 3 consecutive 30 second measurements of the flow. NEC A,

over a range of concentrations, was infused into the heart at 1/ 1 0 0 th of this flow rate

and flow measurements were taken over 30 second intervals from 0 to 4 minutes

(Figure 6 a). To determine whether the increase in flow induced by the NECA had

reached a plateau during this time, flow measurements of three concentrations of

NECA (final concentrations, 0.3pM, 3pM, 30pM) were taken over 30 second

intervals from 0 to 12 minutes (Fig 6 b). From this it was decided to simplify the

protocol. Taking into account the different initial velocities of the flow increases and

the different plateaux reached by the curves, it was decided to measure the control

flow over 0 to 5 minutes followed by a 0 to 5 minute measurement of flow whilst the

agonists were being infiised (agonist flow), and a 5-10 minute washout time between

agonist concentrations. At the start of every experiment a dose of sodium nitroprusside

(SNP) (lOOpM) was infiised into the heart to check its viability. If the increase in flow

induced by SNP was less than Iml/min the heart was discarded. Where inhibitors or

antagonists were used the first agonist applied to the heart after their equilibration was

also SNP (lOOpM).

Initially, the results were expressed as the percentage increase in the flow induced by

the agonists relative to the control flow (% Control flow (mls/min)). However, due to

the decrease in the basal perfusion flow (control flow) throughout the experiment and

the finding that this flow was sometimes influenced by the previous agonist response,

47

Figure 6. Increase in flow induced bv NECA in the ratisolated heart

21 -,

18-

3 -

0-150 100 150 200 2500

Time (Seconds)

21 -,18-

3 -

0 J0 150 300 450 600 750

Time (Seconds)

Increase in Flow in the rat isolated heart induced by NECA (O) 0.1 jiM, (■) 0.3pM, (T) IpM, (# ) 3joM, (♦) lOjiM, (A) 30jiM, measured every 30 seconds for (a) 240 seconds, (b) 740 seconds, (a) Each point is a mean of 5 determinations with s.e. mean shown as vertical bars, (b) Each point is only 1 determination.

48

it was decided not to calculate the results by this method. As SNP was used as the

first agonist on each heart, the increase in flow induced by the agonists was expressed

as a percentage of the increase in the flow induced by the initial dose of SNP (lOOpM)

(% SNP flow (mls/min)). Where inhibitors or antagonists were used the increase in

flow induced by the agonists was compared to the increase in flow induced by SNP

(lOOpM) in the presence of the inhibitor or antagonist. However, due to the

degradation of the condition of the heart over the experiment it seemed unwise to

compare later agonist responses to the initial SNP response of a fl esh heart or not to

consider any changes in the SNP response over the equilibration time given for the

inhibitors and antagonists. Also this method of calculating the results seemed to

magnify the errors inherent in the variable results obtained. The total flow induced by

the agonist concentrations (agonist flow (mls/min)) also did not seem to be an ideal

measurement of the responses because the individual agonist concentrations did not

relax the vessels to a definitive degree in the same or different hearts as the amount of

relaxation obtained seemed to depend on the contractile state of the vessels at the time

the agonists were infused. Finally it was decided that the responses should be

expressed as absolute increase in flow (mls/min), i.e. flow induced by the agonists

(agonist flow) - the basal perfusion flow (control flow). This takes into account the

contractile state of the vessels immediately prior to the agonists being infused but also

the degradation of the condition of the heart over the experiment.

To study the effects ofL-NAME, 8 -SPT and PD 115,199 on the agonist concentration

response curves the heart was equilibrated with these inhibitors for 2 0 minutes before

the agonist concentration response curves were carried out. All the agonist

concentration response curves were constructed non cumulatively, with a random

order of agonists on unpaired hearts. Only one concentration response curve was

carried out on each heart due to length of time required to construct one curve and the

limited viability of the heart.

49

EGsmis/mm values can be defined as the concentration of the agonist producing an absolute increase in flow of 3mls/min

^ ? A 2 can also be defined as the logarithm of the (dose-ratio -1) minus the logarithm of the

molar concentration of the antagonist i.e. PA2 = log (dr-1) - log[A].

The results obtained from the isolated Langendorff perfused heart indicated that SNP

modified the rate and force of contraction of the heart. As a consequence of this the effects of

SNP on spontaneously beating isolated atria was ascertained and compared to the responses

produced by isoprenaline on this tissue.

The effect of phenylephrine, L-NAME, 8 -SPT and PD 115,199 on the increase in flow

induced by one concentration of SNP (lOOpM) and NECA (3{xM) and also on the

basal perfusion flow through the heart was investigated. The basal perfusion flow and

the agonist induced flows were measured before and after perfusion of the heart with

these compounds and the % decrease in basal flow or % decrease in the agonist

induced flows were calculated. Each heart was perfused for 20 minutes with only one

inhibitor.

pECimk/min values (the negative logarithm of the EC^nik/min value) were calculated

from the linear portion of individual log. concentration response curves by linear

regression analysis. Potency of the agonists was expressed as ECgmis/min values

calculated from the mean of the pECgmis/mln values. For PD 115,199, the dose-ratio

for NECA was calculated as the ratio of the ECg^ ig/^^ values in the presence and

absence of one concentration of the antagonist and estimated pA% values were

calculated as the negative logarithm of the (molar concentration of the antagonist

divided by the dose-ratio -l}^

2.1.3. Rat isolated atria.

Male Wistar rats (300-400g) obtained fi*om the University of Surrey breeding unit

were injected with 500 Units heparin i.v. via the tail vein 5 minutes prior to being

killed by cervical dislocation. The heart was rapidly excised fi'om the thoracic cavity

and plunged into ice cold Krebs Henseleit solution (mM: NaCl, 118; KCl, 4.7;

NaHCOs, 25; Glucose, 11; MgS0 4 .7 H2 0 , 1.17; KH2PO4 , 1.2; CaCl2 .2 H2 0 , 2.5)

whilst the excess lung and connective tissue were removed. The right atria was

dissected fi'om the heart and mounted in 4ml organ baths containing Krebs Henseleit

solution, maintained at 37°C and aerated with 95% O2 : 5% CO2 The atria were

allowed to beat spontaneously under a resting tension of 2.5g and equilibrated for 1

hour with isometric responses subsequently recorded with a Grass FT03 force

50

displacement transducer and displayed on a Grass polygraph (model 79). To measure

the effects of SNP and isoprenaline on the rate and force of contraction of the atria, the

agonists were added non cumulatively and the atria were incubated with SNP for 5

minutes, or in the case of isoprenaline until a stable response was produced

(approximately 2-3 minutes). Changes in the rate of contractions were expressed as %

increase in rate as compared to the control in the absence of the agonists, and changes

in force of contraction were calculated as % increase in force as compared to the

control in the absence of the agonists.

2.2. MATERIALS

PD 115,199 was kindly provided by Dr. M.G. Collis, Pfizer Central Research,

Sandwich, U.K. whilst APNEA was provided by Professor R. A. Olsson, University of

South Florida, USA. CGS 21680, CPA, DPCPX and 8-SPT were obtained from

Research Biochemicals Inc., U.S.A. and heparin was supplied by Leo Laboratories

Ltd., U.K. whilst all other compounds used were obtained from Sigma Chemical Co.,

U.K. Buffer salts were obtained from BDH Ltd. and Fisons and 95% 0%: 5% CO2

was supplied by B.O.C.

CPA, MTA, and DPCPX were made up to lOmM in 20% ethanol, 0. IM HCl and 20%

ethanol respectively. APNEA was made up at a concentration of 8mM in 70%

dimethylsulphoxide (DMSO) whilst CGS 21680 was made up to lOmM in 7% DMSO

for use in the isolated aorta and in 10% DMSO for the Langendorff heart preparation.

All these compounds were then fijrther diluted in distilled water. PD 115,199 was

made up to lOmM in 100% DMSO and diluted as required in 100% DMSO. All other

compounds were made up in distilled water. The solvents used for dissolving the

compounds had no effect on the responses of the tissues at the final concentrations

used. Isoprenaline (lOmM) and phenylephrine (lOmM) were made up daily in distilled

water. Noradrenaline (lOmM) was made up daily by dissolving in ascorbic acid

51

solution (20|ig/ml) whilst heparin (25,000 Units/ml) was diluted to 5,000 Units/ml in

normal saline solution. Adenosine and adenosine triphosphate (ATP) were made up in

distilled water and their final concentrations checked using UV spectrophotometry.

The absorbance of the solution was measured at 259nm and compared to the

extinction coefficient for adenosine or ATP at this wavelength, 15.1 x 10-3M"lcm-l

and 15.4 X 10-3M-lcm"l respectively.

52

CHAPTER 3: RESULTS

4Using a one-way ANOVA analysis followed by Dunnetts multiple comparison tests, the

pECgo value for NECA was not significantly different to the pECgg value of CGS 21680 (p

> 0.05), the pECgo value for adenosine was not significantly different to that of CPA and

APNEA (p >0.05), whereas the pECgg values of NECA and CGS 21680 were significantly

different to the pECgo values of adenosine CPA and APNEA (p <0.01).

3.1. RAT THORACIC AORTA

Adenosine and several of its analogues induced relaxation of the noradrenaline

contracted rat thoracic aorta producing an agonist potency order of NECA > CGS

21680 > adenosine = APNEA = CPA > MTA (Figure values for the agonists

are given in Table 3. The maximal responses of these agonists varied, with the lowest

maximal response being produced by CGS 21680 (approximately 40% inhibition)

followed by NECA (approximately 60% inhibition). Although the concentration

response curves of APNEA and adenosine did not plateau even at the highest

concentration tested (300pM), their maximal values appear to be higher than that

produced by NECA. The concentration response curve for CPA also did not reach a

plateau at 300pM but its maximal response appears to be lower than that of APNEA

and adenosine. MTA only induced relaxations at the highest concentration tested,

therefore its maximal response could not be predicted. Acetylcholine was the most

potent relaxing agent used on the rat thoracic aorta producing almost 1 0 0 % relaxation

at 3|xM and an EC3 0 value of 0.04jxM (Figure 7).

The relaxations produced by the adenosine agonists were abolished by the mechanical

removal of the endothelium, with the exception of the small residual responses seen at

300pM for NECA, CPA and MTA (Figure 8 ). Removal of the endothelium did not

affect the contractions induced by NA in the rat aortic rings (Figure 9a). At the

submaximal concentration of lOOnM NA chosen to precontract the aortic rings, there

was no difference in the tension developed in intact and endothelium denuded aorta

(Figure 9a).

Equilibration of the aortic rings with L-NAME (30pM) also abolished the responses of

the adenosine agonists, although again small residual responses were seen for all the

agonists at the highest concentrations used (Figure 10). The relaxations induced by the

endothelium independent vasodilator SNP in the presence and absence of L-

53

i§

•S

II0

12

i'S.§

&

_ oo

oo

o

o

1 - 0

1 o

§

Iiuto

I

oo o00

oCO § o

CM

Ci

tioipBJjuoo VN JO uopiqiqui %

54

1

HPh

I0 0

COO

I0CO- §

1

1CO

I

IÜ

I2

CO

I,O

13§

.B*000

1o\On

U"â.

oZ Q03 PL|H

iia

iiQ

Ia

H

IOO<

(NO ^+1 rs r- S's6

O ^+1m w VO

o$NccÜU

IT)(N

roT|-

e

i

m00

CNc5 +1m p.

0\CN

g

i lPo COa s

CO§Ig

l l lo ^60 4-, § §

l i îJh

CmO§

_ O

IIi §

CD

SCDcô

1§i (S

Î P§ QCOi l i

H CO

mg . ë

"S Q

i l ?

liî

w ^Cm CD

§

55

Figures. Effect of removal of the endothelium on theresponses of adenosine analogues in the rat thoracic aorta

80-1

6 0 -

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000

O 1 0 0 -,

8 0 -

6 0 -

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000

60-,

5 0 -

4 0 -

3 0 -

2 0 -

1 0 -

0 -J0.01 0.1 1 10 100 1000

1 0 0 -,

8 0 -

6 0 -

4 0 -

2 0 -

0 J0.01 0.1 1 10 100 1000

50-1

4 0 -

3 0 -

2 0 -

1 0 -

0 - 1

0.01 0.1 1 10 100 1000

Agonist Concentration (|liM)

Relaxations of the rat thoracic aorta precontracted with NA (lOOnM) induced by adenosine analogues in the presence (# ) and absence (■) of endothehum. (a) NECA, (b) CGS 21680, (c) Adenosine, (d) CPA, (e) APNEA, (f) MTA. Each point represents the mean of at least 4 determinations with s.e. mean shown as vertical bars.

56

Figure 9. Effect of endothelium removal and L-NAME on theNA induced contractions of the rat thoracic aorta.

W)

I

3-,

2.5-

2 -

.5

1

0.5-

0 - 1

1 0 1 0 0 00 . 1 1 1 0 0

W)

I

2.5-

2 -

0.5-

T1

—T" 1 0 0

— I1 0 0 01 0

NA Concentration (nM)

Contraction of the rat thoracic aorta induced by noradrenaline (a) in the intact (# ) and denuded (■ ) aorta, (b) in the presence (■ ) and absence ( • ) of L-NAME (30jiM). Each point represents the mean of at least 4 determinations with s.e. mean shown as vertical bars

57

Figure 10. Effect of L-NAME on the responses of adenosineanalogues in the rat thoracic aorta

80-,

6 0 -

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000

8 0 -

6 0 -

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000

50-,

4 0 -

3 0 -

2 0 -

1 0 -

0 - 1

0.01 0.1 1 10 100 1000

50-1

4 0 -

3 0 -

2 0 -

1 0 -

0 - 1

0.01 0.1 1 10 100 1000

60-,

5 0 -

4 0 -

3 0 -

2 0 -

1 0 -

0 J0.01 0.1 1 10 100 1000

1 0 0 -,

8 0 -

6 0 -

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000

Agonist Concentration (|iM)Relaxations induced by adenosine analogues of the rat thoracic aorta precontracted with NA (lOOnM) in the presence (■ ) and absence (# ) of L-NAME (30|xM). (a) NECA, (b) CGS 21680, (c) Adenosine, (d) CPA, (e) APNEA, (f) MTA. Each point represents the mean of at least 4 determinations with s.e. mean shown as vertical bars.

58

NAME (30[xM) were not significantly different (p^ 0.05, Students 'f test. Figure 11).

L-NAME (SOjiM) enhanced the contractions induced by NA, shifting the

concentration response curve to the left by approximately ten fold (Figure 9b).

However, at the noradrenaline concentration (lOOnM) used to precontract the aorta

throughout the study, the responses induced by NA in the presence and absence of L-

NAME were not significantly different (p 0.05, Students't' test). L-NAME (30pM)

did not affect the relaxations induced by isoprenaline (25pM) (results not shown).

8 -SPT (50|xM) inhibited the relaxations produced by all the adenosine agonists with

the exception of MTA, whose response at 300pM was slightly enhanced, although this

was not significant (p> 0.05, Students't' test. Figure 12). However, the degree of

inhibition and the dose-ratios obtained for 8 -SPT varied with the agonist used, with

NECA being inhibited to the greatest extent (Table 3). In addition to shifting the

agonist concentration response curves to the right, 8 -SPT also increased the maximal

response of CGS 21680 from approximately 40% to 60% inhibition and increased the

response induced by CPA at the highest concentration used (lOOpM).

MTA (lOOpM) also acted as an antagonist on the rat thoracic aorta (Figure 12). As

with 8 -SPT, the degree of inhibition and the dose-ratios obtained for MTA varied with

the agonist used, again with NECA being inhibited to the greatest extent (Table 3).

Similar to 8 -SPT, MTA increased the maximal response of CGS 21680 from

approximately 40% to 60% inhibition, but MTA also decreased the maximal response

of NECA from approximately 60% to 40% inhibition (Figure 12).

The A2 a selective antagonist PD 115,199 (IpM) inhibited the relaxations induced by

NECA and CGS 21680 shifting the concentration response curves to the right (Figure

13). The dose-ratios produced for NECA and CGS 21680 in the presence of the

antagonist were 30.2 and 6.2 (estimated pA% values of 7.5 and 6.7) respectively with

NECA being inhibited to a slightly greater extent than CGS 21680 (Table 3). In

59

Figure 11. Effect of L-NAME on the SNP inducedrelaxation of the rat thoracic aorta

120-,

S 1 0 0 -

80

60

40

2 0 -

0 - 1

0 . 1 1 1 0 1 0 0

SNP Concentration (nM)

Relaxations of the rat thoracic aorta precontracted with NA (lOOnM) induced by SNP in the presence (■ ) and absence (# ) of L-NAME (30pM). Each point represents the mean of at 4 determinations with the s.e. mean shown as vertical bars.

60

Figure 12. Effect of 8-SPT and MTA on the relaxations of the rat thoracic aorta induced bv adenosine analogues

8 0 -,

6 0 -

4 0 -

2 0 -

o J0.01 0.1

1------ 1------ 110 100 1000

6 0 -

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000

Iu

I

§

1 0 0 -,

8 0 -

6 0 -

4 0 -

2 0 -

0 -0 . 0 1 0 . 1 10 100 1000

1 0 0 -n

8 0 -

6 0 -

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000

1 0 0 -,

8 0 -

6 0 -

4 0 -

2 0 -

0 J0.01 0.1 1 10 100 1000

60-,

4 0 -

2 0 -

0 - 1

0.01 0.1 1 10 100 1000

Agonist Concentration (jiM)

Relaxations induced by adenosine analogues of the rat thoracic aorta in the absence of antagonist ( • ) , or in the presence of 8-SPT (50jiM), (■ ), or MTA (lOOjiM), (O). (a) NECA, (b) CGS 21680, (c) Adenosine, (d) CPA, (e) APNEA, (f) MTA. Each point represents the mean of at least 4 determinations with s.e. mean shown as vertical bars.

61

Figure 13. Effect of PD 115.199 on the responsesto NECA and CGS 21680 in the rat thoracic aorta

60

y 5 0 -

5 4 0 -oCh 30 —

1 1 0 1 0 0 1 0 0 00 . 0 1 0 . 1

NECA Concentration (|xM)

80-,

7 0 -

O 5 0 -

4 0 -

2 3 0 -

2 0 -

1 0 -

^ 0 - 1

10 . 0 1 0 . 1 1 0 1 0 0 1 0 0 0

CGS 21680 Concentration (pM)

Relaxations of the rat thoracic aorta precontracted with NA (lOOnM) induced by (a) NECA and (b) CGS 21680 in the presence (■) and absence (# ) of PD 115,199 (IpM). Each point represents the mean of at least 4 determinations with s.e. mean shown as vertical bars.

62

addition, the maximal response induced by CGS 21680 was increased from

approximately 40% to 80% in the presence of PD 115,199 and the maximal response

to NECA also appears to be increased although a new maximum response was not

achieved (Figure 13).

DPCPX (InM) did not alter the concentration response curves produced by APNEA

and adenosine, but it potentiated the relaxations induced by CPA, shifting the

concentration response curve slightly to the left, although this was not a significant

effect (p> 0.05, Students't' test. Figure 14). No contractions to CPA (< IpM) were

observed in the endothelium intact aorta, and in the presence of L-NAME (30pM) no

contractions to CPA were seen in the presence or absence of a NA contraction

(lOOnM) and an enhancement of the NA contraction was not observed when CPA was

added prior to the NA (Figure 15).

3.2. RAT ISOLATED HEART

Adenosine, CGS 21680, NECA and CPA induced an increase in flow rate when

infiised into the rat isolated heart Langendorff* preparation. The agonist potency order

obtained was CGS 21680 > NECA > CPA = adenosine, with ECq^iR/mm values of 0.8

pM, l.OpM, 52.0pM and 59.1pM. respectively (Figure 16). ATP had a similar

potency to that of adenosine and CPA producing an ECc mk/min value of 38.2pM

(Figure 16). The maximal response obtained for CGS 21680 was only 70% of that

achieved by NECA. CPA and adenosine did not achieve a maximal response within

the concentration range tested (Figure 16). When L-NAME (30pM) was infused into

the heart over a 5 minute period using the same protocol as the adenosine agonists, it

produced a small decrease in flow (0.3 ± 0.18 mls/min).

63

Figure 14. Effect of DPCPX on the relaxations induced bvAPNEA. Adenosine and CPA in the rat thoracic aorta

1 0 0 -,

8 0 -

6 0 -

4 0 -

2 0 -

0-i1 0 0 01 1 0 1 0 0

Io§

80-,

6 0 -

4 0 -

2 0 -

1 0 1 0 01 1 0 0 0

1 0 0 -,

8 0 -

6 0 -

4 0 -

2 0 -

0 - 1

1 0 1 0 0 1 0 0 01

Agonist Concentration (|iM)Relaxations induced by adenosine analogues of the rat thoracic aorta precontracted with noradrenaline (lOOnM) in the presence (■ ) and absence (# ) of DPCPX (InM). (a) APNEA, (b) Adenosine, (c) CPA. Each point represents the mean of at least 4 determinations with s.e. mean shown as vertical bars.

64

Figure 15. Representative traces of some of the responses ofadenosine analogues in the rat thoracic aorta.

• NECA NA (10//M)

(lOOnM)• CGS 21680

NA .(IO//M) (lOGoM)

CPA#NA

(lOOnM)(IO//M)

• Ado NA (IO;/M)

(IOOqM)

L-NAME(30//M)

NA(lOOnM)

CPA(IO;/M)

i \ \ -L-NAME(30/vM)

A

CPA(lOO/yM)

«NA

(lOOnM)

A •NA

(lOOnM)CPA

4 n iin

(lOOoM)

Representative traces of the relaxations induced by the adenosine analogues, the absence of contractile activity o f CPA in the presence of L-NAME (30pM) and the absence of enhancement of the NA contraction by CPA in the rat thoracic aorta.

65

g

2

III

■sI3

I.a

IVO1-H2

&

OO

o

o

odCO CO00 CM o

§

1to

Î

g l"TO Cd

§ - s

s sII< %O <0

§1 ► aoooVO

( N§5

ê

IoA g

w 1

ÎI

ta.aI

i s

112

Ia

(Oa

(D(A§î

(mm/siui) Avo| m ssbsjout s^npsqy

66

8 -SPT (50|iM) abolished the increase in flow produced by adenosine and its analogues

with the exception of a small increase in flow (0.7 mls/min) produced by the highest

concentration of NEC A (30|xM) (Figure 17).

In the presence of the Â2 a antagonist PD 115,199 (30nM), the increase in flow induced

by NECA was inhibited, shifting the concentration response curve to the right (Figure

18). The EC3„jis/uiin value calculated for NECA in the presence of PD 115,199 was

3.9pM, producing a dose-ratio of 3.9 (estimated pA2 of 8 ). In the presence of a lower

concentration of PD 115,199 (lOnM) the responses to CGS 21680 were almost

abolished. As the increases in flow achieved by CGS 21680 in the presence of PD

115,199 were very low producing a maximal response of only Iml/min increase in flow

at lOpM CGS 21680, it was impossible to calculate a ECgmis/min value (Figure 18).

Perfusion of the heart with a nitric oxide synthase inhibitor L-NAME (30pM) for 30

minutes caused an inhibition of the increase in flow rate induced by ATP, adenosine

and its analogues, producing a rightward shift in the concentration response curves

(Figure 19). For CPA and adenosine the shift was such that only the highest

concentration of agonist (100p,M) produced a small increase in flow (1.02 mls/min and

0.64 mls/min respectively) (Figure 19). It was impossible to obtain dose-ratios for CPA

and adenosine in the presence of L-NAME due to the low potency of the compounds

and the lack of responses produced within the agonist concentration range used.

In the presence of L-NAME (30pM) the responses for NECA and CGS 21680 were

inhibited, shifting the concentration response curves to the right (Figure 19). As the

maximal responses for these agonists in the presence of the inhibitor only reached 3

mls/min it was impossible to calculate dose-ratio values for the agonists. However, the

concentration curve for CGS 21680 was shifted approximately 30 fold to the right

67

Figure 17. Effect of 8-SPT on the responses of adenosineanalogues in the rat isolated heart

8 -,

B

BI

4 -

oCO

0.01 0.1 1 10 100

8 -,7 -6 -

5 -

4 -3 -

2 -

0 -

1 0 0100.01 0.1 1

"S' 8-1•fH

s 7 -c/aI6 -

1 5 -

.s 4 -

3 -2 -

s 1 -ag 0 -

-1 —0 ,

I----------- 1----

01 0.1 1 10 100

Agonist Concentration (jiM)

8 -,

7 -6 -

5 -4 -

3 -2 -

0 -

1 0 0100.01 0.1 1

Agonist Concentration (|iM )

Increase in flow induced by (a) NECA, (b) CGS21680, (c) Adenosine and (d) CPA in the rat isolated heart in the presence (■ ) and absence (# ) of 8-SPT (50pM). Each point is a mean of at least 4 determinations with s.e. mean shown as vertical bars.

68

Figure 18. Effect of PD 115.199 on the responsesto NECA and CGS 21680 in the rat isolated heart.

7 -I

(Dx/iS 3 -

I 2-I

0.1 1 10 1 0 0

NECA Concentration ( \ iM )

5

4

3

2

1

0

0.01 0.1 1 10 100

CGS 21680 Concentration (|iM)

Increase in flow induced by (a) NECA in the presence (■ ) and absence ( • ) of PD 115,199 (30nM) and (b) CGS 21680 in the presence (■ ) and absence (# ) of PD 115,199 (lOnM) in the rat isolated heart. Each point is a mean of at least 6 determinations with s.e. mean shown as vertical bars.

69

Figure 19. Effect of L-NAME on the responses of adenosineanalogues on the rat isolated heart.

8-,76

54-

2 -

1

0

0.01 0.1 10 100

8 -,

7 -6 -

5 -4 -

3 -2 -

0 -

1 0 00.01 0.1 101

I 7 -l e -

g

0.01 0.1 1 10 100

8 -,

7 -

6 -

5 -4 -

3 -2 -

0 -

0.01 0.1 10 1 0 01

Agonist Concentration ()xM) Agonist Concentration (jiM)

Increase in flow induced by (a) NECA, (b) CGS21680, (c) Adenosine and (d) CPA in the rat isolated heart in the presence (■ ) and absence (# ) of L-NAME (30pM). Each point represents the mean of at least 9 determinations with s.e. mean shown as vertical bars.

70

I talthough this reduction did not quite reach significance (0.05 < p < 0.1)

while that for NECA was shifted 20 fold to the right in the presence of L-NAME when

compared at an increase in flow rate of 2 mls/min. In addition, L-NAME appears to

have decreased the maximal response achieved by NECA and CGS 21680. The

maximal response of NECA decreased from 7.2 mls/min to 3.1 mls/min whilst that of

CGS 21680 decreased from 5.1 mls/min to 2.9 mls/min (Figure 19).

L-NAME (30pM) inhibited the responses to ATP shifting the concentration response

curve for ATP to the right, although small residual responses were seen at the highest

concentrations of ATP used. This curve did not differ markedly from that produced by

adenosine and CPA in the presence of L-NAME (Figure 20). Although the responses

produced by ATP did not reach a plateau at lOOpM, it appears that ATP in the

presence of L-NAME would achieve a lower maximal response than in the absence of

L-NAME (Figure 20). The concentration response curves for SNP in the presence and

absence of L-NAME were not significantly different (p> 0.05, Students t' test. Figure

20).

Perftjsion of the heart with L-NAME (30pM), phenylephrine (lOOnM), or the

adenosine antagonists 8 -SPT (lOOpM) and PD 115,199 (IfxM) significantly decreased

the basal flow through the heart (p < 0.05, paired Students 't' test), whereas perfiision

with Krebs solution only decreased the basal flow by 1.18 ± 11.6% (Table 4). The

increase in flow induced by SNP (lOOpM) was significantly reduced after perfiision of

the heart with L-NAME, phenylephrine, 8 -SPT and PD 115,199, but was also

significantly decreased when the heart was perfused with Krebs solution (p < 0.05,

paired Students 'f test. Table 4). Similarly, the increase in flow induced by NECA (3 p.

M) was also reduced by perfiision of the heart with Krebs solution or PE

Students't' test), and also with L-NAME or 8 -SPT (p < 0.05, Students't' test. Table

4). However, perfiision of the heart with PD 115,199 reduced the increase in flow

produced by NECA, although this was not significant (Table 4). Using a one-way

ANOVA analysis followed by a Dunnetts post-hoc test a significant difference (p <

71

0.05) was observed between the reduction in the NECA induced response caused by 8 -

SPT and that caused by L-NAME and Krebs solution alone.

Perfusion of the heart with L-NAME (30pM) significantly decreased the force of

contraction of the heart by 28%, from 2.9 ± 0.2 to 2.0 ± 0.2 g tension (p < 0.05,

Students 't' test). Infusion of SNP (lOOpM) however, increased the force of

contraction both in the presence and absence of L-NAME (24.3% and 28.9%

respectively). The overall force generated by SNP in the presence of L-NAME (30p

M) (1.96 ± 0.17g to 2.45 ± 0.2g was lower than that generated in the absence of L-

NAME (30pM) (2.91 ± 0.15g to 3.62 ± 0.2g) although the percentage increase

obtained was the same.

In unstimulated, spontaneously beating hearts SNP increased the rate of contraction of

the heart from 183.8 ± 9.9 beats/min prior to the SNP infusion to 247.5 ± 9.2

beats/min during the SNP infusion producing a 25% increase in heart rate.

3.3. RAT ISOLATED ATRIA

In the rat isolated atria, SNP (lOOpM) induced a small increase in the rate of

contraction of the atria (4.6% ±1.3 increase from control) but had no effect on the

force of contraction (Figure 21). Isoprenaline (lOnM) however, caused an significant

increase in the rate and force of contraction of the atria (59.5% ± 5.0 and 137.3% ±

17.8 respectively, p < 0.05, Students't' test. Figure 2 1 ).

72

Figure 20. Effect of L-NAME on the responses toATP and SNP in the rat isolated heart

B

Ph3

g1 0 01 100.1

ATP Concentration (|iM)

a

I(D

g1 10 1 0 00.1

SNP Concentration (|xM)

Increase in flow induced by ATP and SNP in the rat isolated heart in the presence (■ ) and absence (# ) of L-NAME (30jiM). Each point represents the mean of 4 determinations with s.e. mean shown as vertical bars.

73

bi) 92 cd '

c/3 A ^

<DI

s

l î

I I

1

1

(UQ

( NO Ô ^< N + 1 ^ .

A 3C N t—1 ‘^ + 1

V OC \Os

v oo ô

* ! l

^ « n

î s

ci4— o l

M o

: + '

a v o C 4 d ^ - 1 - 1

J î m q

2 ÏÏm + |

V O + ,S

i n + i^ 0 m + |

5

II

I<

1

CDP

“ 2m + i

mOm + 1 S “^ + 1

c 4 q

Î 2 Ï Ï

o \Pm + \

f l

* o o « O o r n + 1

* T To o ( N + l

* r - ^ o P + l

* ( NO o P 4-1

* ( N0 0 dT— 1 - f 1

l . s

ÛH g « ^ + 1 3 | , ^ + I- 2 e n 4 . 1

a S^ + 1

1I

1

CDP

Ci2^ + 1 iïi 35

m 4-1

w m 0 0M 4-1

v o

# " °* m O O i X + l

* VOm o V0 4-I

* ro e n 0 ^+\

* ^ T f 0v d + l

l . fSS-B

a , g » ° S" + i

^ 3' ^ + 1

3<23 S

H i! II<ü1

p4

1 èIC / 5 0

o\a\r H

lî

I1Ph

ino0 V A td

1IÎ

(U

I“1c/3<U T) H o> I

C/3

Î - Io o 'cd V V g> A A pcd cd <U (D1 1çë çë& §)C/D C/DB aI §T3 T3

*

IAî

a

1I

> i

s %

= 1I îr. . T3

II â I §cd ü

« § «> «

I I

11|4<u . 2

ilII

74

Figure 21. Representative traces of the responses to SNPand isoprenaline in the rat isolated atria.

(a) Rate of contraction

300-

250

b/min 200

150

100

SNP (lOOfxM)

Imin

b/min

600t

500

400

300

200 Isoprenaline (lOnM)1 min

(b) Force of contraction

0.25g

1 minSNP (lOOjiM)

0.5g

1 minIsoprenaline (lOnM)

Representative traces of the changes in the (a) rate of contraction and (b) force of contration of the rat isolated atria induced by SNP (lOOpM) and isoprenaline (lOnM).

75

CHAPTER 4: DISCUSSION

4.1. RAT THORACIC AORTA

The presence of adenosine receptors in the rat thoracic aorta has been previously

reported by Moritoki et al. (1990), Rose'Meyer and Hope, (1990) and Conti et al.

(1993). In these studies the 5' substituted analogue NECA was found to be more

potent than the substituted A% selective agonist R-PIA, which according to the

agonist potency orders ascertained for the adenosine receptors (see Table 1 in the

introduction) would indicate the presence of A2 receptors on this tissue. In this study,

adenosine agonists induced relaxations of the rat thoracic aorta with a potency order of

NECA > CGS 21680 > adenosine = APNEA = CPA > MTA again indicating the

presence of A% receptors as NECA is more potent than the N^ substituted Ai selective

agonists CPA and APNEA. It was also found that the 2-substituted agonist CGS

21680 has a comparable potency to NECA producing EC3 0 values of 0.5pM and 0.2p

M respectively. Conti et al. (1993) and Moritoki et a l (1990) also found that CGS

21680 and NECA were comparable, potent agonists on the aorta and it has also been

reported in binding studies in the rat striatum that both of these compounds produced

similar affinity values (14nM and 12nM respectively, Jarvis et al., 1989). As CGS

21680 is considered to be selective for the A%a receptor subtype (Bruns et a l, 1986;

Hutchinson et a l, 1989), the high potency of CGS 21680 in this study and in those

studies by Conti et al. (1993) and Moritoki et a l (1990) suggests that the receptors

present on the aorta are actually of the A2a receptor subtype although the receptors

reported by Conti et al. (1993) and Moritoki et a l (1990) were only classified as A2

receptors. Similar potency orders have been seen in a variety of tissues including the

porcine and dog coronary arteries, rat pheochromocytoma (PC 12) cells, human

neutrophils and platelets where the receptor has been classified as the A2 a receptor

subtype due to the high potency of A2 a selective agonists such as CGS 21680 and CV

1808 (Balwierczak et a l, 1991; Kennedy et a l, 1992, Gurden et a l, 1992; Hide et a l,

1992).

76

In contrast to Conti et a l (1993) who showed 100% relaxations of the rat thoracic

aorta by all of the adenosine agonists, in this study the maximal responses obtained

varied depending on the agonist used. The maximal values for CPA, adenosine,

APNEA and MTA were hard to predict due to their low potencies, whilst NECA

produced a maximal response of 60% relaxation and CGS 21680 only produced 40%

relaxation. This discrepancy could be explained by the use of a different strain of rat

(Sprague-Dawley), or more likely the different contractile agonist (PGp2 a) used by

Conti et a l (1993). It has also been shown by Moritoki et a l (1990) that in 4 week

old rats, adenosine produced a maximal response of 80% relaxation when using NA as

the contractile agonist. However, in the thoracic aorta of 8 weeks old rats

(approximately the age of the rats used in our study) the maximal response induced by

adenosine was decreased slightly so it did not exceed 70-75% relaxation. Also in 8

week old rats, a 30 fold higher concentration of adenosine was required to produce a

relaxation comparable to that achieved at 4 weeks of age. Increasing the age of the

rats to 1 2 weeks again the adenosine response was decreased and although a plateau

was not achieved, adenosine at the highest concentration used (300pM) only induced

approximately 30% relaxation. It would appear that the dilator response to adenosine

decreases with age due to a decrease in the number of purinoceptors, or a decrease in

the ability of the endothelium to produce EDRF, or a decrease in guanylate cyclase

activity leading to a decrease in the formation of cGMP (Moritoki et a l, 1990).

Therefore it may not have been possible to achieve 1 0 0 % relaxations to the adenosine

agonists used in this study as the older age of the rats may have prevented maximum

relaxations being obtained. This may also explain why Conti et al. (1993) using

younger rats (150-200g, approximately 5 weeks old) obtained a greater degree of

relaxation to the adenosine agonists. The major difference between the studies of

Conti et al. (1993), Moritoki et al (1990) and this study is the method of agonist

dosing. Cumulative dosing was used in the other studies whereas single dosing was

77

used in our study due to the unstable nature of the noradrenaline contraction and the

agonist relaxations. Another suggestion for the different maximal responses achieved

by NECA and CGS 21680 is the possibility that NECA is acting as a fiill agonist on

this tissue whereas CGS 21680, only achieving 6 6 % of the NECA maximal response,

may be acting as a partial agonist. If this was the case, CGS 21680, acting at the same

receptor as NECA, would behave as antagonist inhibiting the response to NECA.

Indeed, in the rat thoracic aorta CGS 21680 has been shown to inhibit NECA induced

relaxations (Prentice and Hourani, 1995). Other examples where CGS 21680 has been

shown to have a reduced efficacy compared to NECA include the A2 a mediated

stimulation of cAMP production in both human platelets and cultured guinea-pig

coronary endothelial cells where the maximal response to CGS 21680 was 65% and

50% of the NECA response respectively (Cristalli et al., 1994; Schiele and Schwabe,

1994).

The 5’ endogenous nucleoside MTA induced relaxations of the thoracic aorta only at

the highest concentration (300|xM) used and therefore was acting as a very weak

agonist. This response was not inhibited by the non-selective antagonist 8 -SPT

suggesting that the MTA induced relaxations were not mediated via the A2 a receptors

present on the aorta but via another site which is resistant to blockade by 8 -SPT.

Brackett and Daly (1991) also reported that in the guinea-pig trachea (A2b receptors)

MTA acts as a very weak agonist (EC5 0 = 225|xM) whose responses were not

inhibited by 8 -SPT (90|xM) concluding that MTA acts via a xanthine-resistant

mechanism. These results are contrary to those reported by Munshi et al. (1988) in the

rabbit jejunum, (containing A2 receptors which were not fiirther classified), and

Nishida et al. (1985) and Daly and Padgett (1992) in the rabbit aorta and rat PC12

cells (which contain A2 a receptors) where MTA acts as a weak or partial agonist

whose effects can be antagonised by the presence of 8 -SPT (lOO iM) and theophylline

(lOOpM). The possibility that MTA may act as an antagonist on these tissues was not

78

considered (Nishida et a l, 1985; Daly and Padgett, 1992). MTA also has been shown

to act as an agonist on the A receptors of the rat cell membrane and the rat

cerebellum (Daly and Padgett, 1992; Munshi et a l, 1988). It has previously been

reported that at the A2b receptors in VA13 human fibroblasts and neuroblastoma cells

MTA appears to act as an antagonist producing Kj values of 8.2 and 4.4 iM

respectively (Bruns et a l, 1986; Munshi et a l, 1988). However, it has also been

reported that MTA can inhibit the A2a receptor mediated effects of adenosine on

platelets (Agarwal and Parks, 1980). MTA has been reported to act as a xanthine

sensitive, weak or partial agonist at A2a receptors, and an antagonist at A2b receptors

(Daly and Padgett, 1992). However, on the A2 a receptors in this study on the rat aorta

MTA appears to act as an antagonist producing similar dose-ratios to those of 8 -SPT

but MTA also appears to act as a xanthine-resistant, weak agonist on another site on

the aorta.

The A2 a selective antagonist PD 115,199 inhibited the relaxations of the rat thoracic

aorta induced by NECA and CGS 21680, although the responses to NECA were

antagonised to a slightly greater extent than the responses of CGS 21680. In this

study, a pA2 value of 7.5 was produced for NECA which is comparable to the

published affinity of PD 115,199 in binding at the A2 a receptors in the rat striatum (pKj

of 7.8, Bruns et a l, 1987). This suggests that the relaxations induced by NECA were

mediated via the A2a receptors present on the rat thoracic aorta. CGS 21680

however, was inhibited by PD 115,199 to a slightly lesser degree than NECA

producing a pA2 value of 6.7 suggesting that although it is likely that CGS 21680 is

acting at the A2a receptors, another receptor site may also be involved.

Both MTA, when used as an antagonist, and the non selective antagonist 8 -SPT

inhibited the responses to all the agonists on the rat aorta, with the exception of the

relaxant responses to MTA which were resistant to blockade by 8 -SPT. However, as

79

with PD 115,199, the dose-ratios in the presence of 8 -SPT and MTA showed that

these antagonists inhibited the effects of NECA to a greater extent than the other

agonists. At the A%a receptors present in the porcine coronary artery King et al.

(1990) reported a pA% value of 5.3 for NECA in the presence of 8 -SPT which is

comparable to the pA% value of 5.6 produced for NECA in the presence of 8 -SPT in

this study. Similarly MTA antagonised the responses to NECA producing a pA] of 6 .1

which is consistent with the published binding affinity at the A%a receptors in the rat

striatum (pKf of 6 , Daly and Padgett, 1992). These results suggest that the relaxations

induced by NECA are mediated via the A%a receptors present on the aorta. The other

agonists used in this study were not inhibited by 8 -SPT and MTA to the same extent as

NECA producing lower pA2 values of 4.2-5.0, suggesting that these agonists may not

be solely acting on A2a receptors but may also be acting on a different second site

which shows some resistance to blockade by 8 -SPT and MTA. In the guinea-pig

trachea (which contains A2b receptors) it has also been shown that NECA was

inhibited by 8 -SPT to a greater extent than 2-CADO, R-PIA and CHA (dose-ratios of

16 and 4.2-5.4 respectively) and also in the bovine coronary arteries where 8 -PT

produced a greater shift in the concentration response curve to NECA than to 2-

CADO (Brackett and Daly, 1991; Sabouni et al., 1989). The maximum responses

achieved by some of the adenosine agonists were altered in the presence of PD

115,199, MTA and 8 -SPT. For example, MTA decreased the maximal response

achieved by NECA and both 8 -SPT and PD 115,199 appeared to slightly increase the

maximal response to NECA although a new maximum value was not reached. The

maximal responses to CGS 21680 were increased by PD 115,199, MTA and 8 -SPT so

that CGS 21680 achieved a similar maximum response to that produced by NECA.

This suggests that although CGS 21680 may be acting as a partial agonist and NECA

acting as a full agonist at the A2 a receptors, in the presence of the adenosine

antagonists, CGS 21680 and NECA may be activating a second site at which CGS

21680 may be acting as a full agonist achieving a similar maximum response to NECA.

80

Enhancement of the maximal response of adenosine has also been seen, in the guinea-

pig trachea in the presence of antagonists such as 8 -PT and DPCPX (Collis et al.,

1987; 1989). Overall the results obtained with the antagonists PD 115,199, 8 -SPT and

MTA suggests that NECA is acting on the A%a receptors present on the rat thoracic

aorta. It is likely that the CGS 21680, CPA and adenosine are also acting on the A%a

receptors but it also appears that their responses are also mediated via a different

second site which displays resistance to blockade by 8 -SPT and MTA. Whereas MTA

acts as an antagonist on the A%a receptors it also appears to act as an agonist on this

second xanthine-resistant site present on the rat thoracic aorta.

These unusual findings could be explained by the presence of other receptors on the

thoracic aorta. One possible candidate is the A% receptor, however in this study little

evidence could be found for this since DPCPX, an Ai selective antagonist did not

inhibit the relaxations induced by the A% agonists APNEA, CPA and adenosine and

contractions to low concentrations of CPA were not seen. Likewise, Conti et al.

(1993) reported that in the rat aorta there was no indication of contractile Ai receptors

and the concentration response curves of both NECA and CGS 21680 were not

affected by DPCPX (lOOnM) so concluding that A% receptors were not present on this

tissue. The small potentiation of the relaxations induced by CPA in the presence of an

Ai selective concentration of DPCPX (InM) may indicate the presence of a small

population of contractile Ai receptors on the aorta. These results are in contrast to

those reported by Farmer et al. (1988) and Stoggall and Shaw (1990) in the guinea-pig

trachea and aorta where A% agonists R-PIA, 2-CADO and CHA produced biphasic

concentration response curves with contractions at low concentrations of the agonists

and relaxations at higher concentrations indicating the presence of both contractile A%

and relaxant A% receptors on the same tissue.

81

The differing degrees of antagonism by 8 -SPT and the weak xanthine-resistant relaxant

effects of MTA could also indicate the presence of A3 receptors in addition to A2a

receptors on the thoracic aorta. This receptor was originally cloned from the rat

striatum and was described as being insensitive to classical xanthine antagonists such as

8 -SPT and DPCPX (Zhou et al., 1992). Also, in the pithed rat the hypotensive

responses induced by some adenosine agonists such as APNEA, R- and S-PIA were

shown to have a component which was resistant to antagonism by 8 -SPT (Fozard and

Carruthers, 1993a; 1993b; Carruthers and Fozard, 1993). These responses were found

to be mediated via an A3 receptor present on mast cells (Hannon et al., 1995).

APNEA and CPA are Ai selective agonists but show some selectivity for the A3

receptors over the A2 a receptors (1.5 and 2 fold respectively), whereas CGS 21680 is

an A2 a selective agonist which shows 39 fold selectivity for the A2 a over the A3

receptor and NECA is a non selective agonist which has a 11 fold selectivity for the

A2a receptor versus the A3 receptor (Kim et a l, 1994a; Van Galen et ah, 1994).

Therefore if a combination of both A3 and A2 a receptors were present on the thoracic

aorta, CGS 21680 would act preferentially on the A2 a receptors, NECA would act on

both receptors, although NECA would have a slightly greater preference for the A2 a

receptors over the A3 receptors, and APNEA and CPA would act equally on both A2 a

and A3 receptors. In these circumstances, 8 -SPT would antagonise the effects of CGS

21680 to the greatest degree, followed by NECA and then APNEA and CPA to the

least extent due to the resistance of the A3 receptor to blockade by 8 -SPT. However,

it can be seen from the dose-ratios produced for 8 -SPT that in this study NECA was

inhibited to the greatest extent, whilst CGS 21680 was inhibited to a similar degree as

CPA and APNEA. Therefore it is unlikely that the different antagonism of the agonist

responses by 8 -SPT can be explained by the presence of A3 receptors.

The relaxations produced by the adenosine agonists on the rat thoracic aorta were

virtually abolished by the mechanical removal of the endothelium, suggesting that these

82

responses were endothelium-dependent. The finding that the adenosine agonist

responses were essentially abolished by the nitric oxide synthase inhibitor L-NAME

(30jaM), whilst the relaxation induced by the endothelium-independent agonist sodium

nitroprusside (SNP) was unaffected, implied that adenosine agonist induced relaxations

of the aorta occur through an endothelium-dependent mechanism involving the nitric

oxide pathway. However, Moritoki g/ al (1990), RoseMeyer & Hope (1990) and

Yen et a l (1988) all reported that mechanical removal of the endothelium attenuated

rather than abolished the relaxations induced by the adenosine agonists leading them to

conclude that on the rat aorta the adenosine agonist induced relaxations are only

partially endothelium-dependent. An explanation for these different results could be

due to the degree of mechanical rubbing used to denude the endothelium. It was found

in our study that gentle rubbing was sufficient to abolish the relaxations to

acetylcholine but more vigorous rubbing was required to abolish the responses to the

adenosine agonists. If the aorta was only gently rubbed, relaxations to the adenosine

agonists were produced giving the impression that the agonist responses had only been

attenuated and therefore were only partially endothelial-dependent. RoseMeyer and

Hope (1990) also found that haemoglobin (lOfxM), which inhibits the action of nitric

oxide by binding and reacting with nitric oxide to form NO-haem complexes, inhibited

the response to NECA by 102 fold, shifting its concentration response curve to the

right. Moritoki et al (1990) also reported that both haemoglobin (lOnM) and the

soluble guanylate cyclase inhibitor methylene blue (0.1-3fxM) inhibited the relaxations

induced by adenosine (30|iM) in endothehum intact aorta but not in endothelium

denuded aorta. These results provide fiirther evidence that the responses induced by

the adenosine agonists in the rat thoracic aorta are endothelium-dependent and occur

via the nitric oxide pathway.

In conclusion, these results suggest that adenosine analogues induce relaxations of the

rat thoracic aorta by an endothelium-dependent mechanism which involves the nitric

83

oxide pathway. However, the results obtained using the adenosine antagonists PD

115,199, 8 -SPT and MTA were not straightforward. NECA induced relaxations of the

aorta were mediated via an A2 a receptor. CGS 21680, CPA and adenosine were not

inhibited by the antagonists to the same extent as NECA suggesting that their

responses were mediated via A2 a receptors but also via a second site displaying some

resistance to blockade by 8 -SPT and MTA. The different degrees of antagonism and

the xanthine-resistant effects seen in this study do not appear to involve A% or

xanthine-resistant A3 receptors. Whereas MTA has previously been shown to act as an

antagonist on A2b receptors and a xanthine sensitive agonist on A2 a receptors (Bruns

et al., 1986; Daly and Padgett, 1992), it appears that in this study MTA acts as an

antagonist on the A2 a receptors but also as a xanthine-resistant agonist at the second

site. Although CGS 21680 appears to act as a partial agonist on the A2 a receptors of

the rat aorta, in the presence of the adenosine antagonists CGS 21680 appears to act as

a full agonist at the second site achieving a similar maximum response to NECA. The

differing degrees of antagonism by 8 -SPT and the xanthine-resistant effects of MTA

shown here in the rat thoracic aorta appear to be similar to the responses shown in the

guinea-pig trachea (Brackett and Daly, 1991), indicating the possibility that the

receptors or sites present on these two tissues may be identical.

4.2. RAT ISOLATED HEART AND ATRIA

To investigate the effects of adenosine analogues in the rat isolated heart a

Langendorff constant pressure system was employed. Using this method the heart is

perfused in a retrograde manner, which closes the aortic valves so the perfusate

circulates through the coronary vessels rather than through the heart chambers. The

perfusate solution, which will drain from the heart via the coronary sinus and the cut

vena cava or pulmonary artery, can be collected as effluent underneath the heart for

measurement of the coronary flow. When the pressure is maintained at a constant

84

level a change in the coronary flow through the heart gives a general indication of the

change in coronary vessel tone. For example, an increase in the coronary flow

indicates a decrease in vascular tone i.e. vasodilation and conversely a decrease in flow

suggests an increase in vessel tone i.e. vasoconstriction. However, the heart is a

complex system which makes measurement of the flow through the heart problematic.

Other factors can influence the vasodilations induced by the adenosine agonists, for

example, the vasodilation observed will depend on the basal tone of the coronary

vessels. It is possible that a greater basal tone will lead to physiological antagonism of

the agonist vasodilator responses and therefore a reduction in the vasodilation

observed. Therefore any compound which alters the vessel tone will consequently

alter the vasodilation achieved by the adenosine agonists. As the heart fimctions as a

whole a change in one parameter, such as a change in flow rate, may consequently

affect other heart parameters, for example force or rate of contraction. If these

changes are caused by a drug prior to the construction of agonist concentration

response curves this may complicate the interpretation of the changes in the flow rate

through the heart. In this study on the rat heart the experiments to determine the effect

of adenosine agonists on the flow through the heart have been carried out by two

different protocols. Initially concentration response curves for ATP, SNP and the

adenosine agonists were performed in the presence and absence of the nitric oxide

synthase inhibitor L-NAME and the adenosine antagonists 8 -SPT and PD 115,199 on

unpaired hearts. However, it became apparent that some of the inhibitors caused

vasoconstriction so that physiological antagonism of the agonist responses could not

be ruled out and hence interpretation of these results became more complicated. To

clarify the results the effect of these inhibitors and the vasoconstrictor phenylephrine

on the basal flow and the increases in flow induced by SNP and NECA were

investigated using a protocol involving paired results in individual hearts. The use of

the different protocols may make the interpretations of the results easier. Initially, the

effect of the adenosine antagonists and L-NAME on the full agonist concentration

85

response curves were investigated then the results were subsequently re-examined in

the light of the effect of phenylephrine on the coronary arteries and the involvement of

physiological antagonism in the agonist responses.

In the rat isolated heart the adenosine agonists produced an increase in coronary flow

through the heart indicating vasodilation of the coronary vessels. An agonist potency

order of CGS 21680 > NECA > CPA = adenosine was produced suggesting that, as

the 5' substituted agonist NECA is more potent than the N^ substituted agonist CPA,

the adenosine receptors present on the rat coronary arteries are A2 receptors. CGS

21680 has a comparable if slightly greater potency than NECA in this study producing

FCgmis/min values of 0.8pM and l.OpM respectively. Similar findings were shown in

the studies in the rat heart by Ueeda et al (1991b) and Hutchinson et al. (1989) where

the potency values of l-2nM and 5nM were produced by CGS 21680 and NECA

respectively. In the earlier study on the rat thoracic aorta both of these compounds

had similar potency values, although the order was reversed and NECA was slightly

more potent than CGS 21680. As CGS 21680 is considered to be selective for A2

receptors, the high potency of this agonist in the isolated heart would suggest that the

receptors present on the coronary arteries are of the A2 a subtype (Bruns et a l, 1986;

Hutchinson et a l, 1989). These results are consistent vrith the findings shown in other

studies in the rat heart, where a high correlation was produced between the order of

agonist affinities for binding to A2 a receptors in the rat striatum and the order of

agonist potencies mediating reduction in the coronary flow in the rat heart (Hamilton

et a l, 1987; Oei et a l, 1988). Agonist potency orders similar to those seen in this

study can be seen in isolated coronary arteries fi'om other species, for example,

porcine, bovine, guinea-pig and canine coronary arteries (Abebe et a l, 1994; Conti et

a l, 1993; Vials and Bumstock, 1993; Gurden and Kennedy, 1992) as well as other

blood vessels such as the rat aorta and rabbit hepatic arterial bed (Moritoki et a l,

1990; Mathie et a l, 1991). Using A2 selective agonists such as CGS 21680 and CV

86

1808 the vasodilation of these blood vessels was shown to be mediated via A2 a

receptors.

ATP is normally considered to be an endothelium-dependent ? 2 purinoceptor agonist

mediating vasodilation of blood vessels via ? 2y receptors (De Mey and Vanhoutte,

1981; Kennedy et a l, 1985). ATP was initially used in this study to provide an

endothelium-dependent vasodilation which would be abolished by the presence of a

nitric oxide synthase inhibitor L-NAME. The vasodilations of the adenosine agonists

in the presence and absence of L-NAME could then be compared to an endothelium-

dependent agonist such as ATP as well as an endothelium-independent agonist such as

SNP. However, it has been shown that during passage through the coronary bed ATP

can be sequentially dephosphorylated by ectonucleotidases to produce adenosine

(Fleetwood et a l, 1989). Subsequently, the adenosine can stimulate the Pj (A2 )

receptors to induce vasodilation of the coronary vessels. ATP caused an increase in

coronary flow through the heart with a potency similar to that produced by adenosine

and CPA. This suggests that ATP is being degraded to adenosine before acting via the

Pj receptors on the coronary arteries, although evidence of antagonism by the non

selective Pj antagonist 8 -SPT or fiirther work with ectonucleotidase inhibitors would

be required to confirm it.

Similar to the earlier study on the isolated rat aorta the maximal response produced by

CGS 21680 was only 70% of the NECA maximal response (5 and 7mls/min increase in

flow respectively). This suggests that NECA is acting as a full agonist and CGS 21680

is acting as a partial agonist on the A2a receptors of the coronary vessels. CGS 21680

has also been reported to act as a partial agonist on A2 a receptors present on human

platelets and cultured guinea-pig coronary endothelial cells where the maximum

response to CGS 21680 was 65% and 50% of the NECA maximum response

respectively (Cristalli et a l, 1994; Schiele and Schwabe, 1994). The different maximal

87

responses produced by NECA and CGS 21680 could also be due to the presence of

two receptors on the coronary arteries with different maximum values, for example,

the presence of A2b receptors as well as A2 a receptors. As CGS 21680 is an A2 a

selective agonist it would preferentially stimulate the A2a receptors whereas NECA is a

potent agonist at both A2 and A2y receptors and therefore would stimulate both

receptors. In this instance NECA would have a greater maximal response than CGS

21680 due to the combination of the responses produced via both A2 a and A2y

receptors as long as maximum value achieved by the stimulation of A2y receptors is

greater than that achieved by the stimulation of A2 receptors. However, the presence

of A2b receptors cannot be determined directly as selective antagonists for this

receptor are not available, although they can be inferred by the lack of antagonism of

adenosine agonist responses by an A2 selective antagonist such as PD 115,199.

The non-selective antagonist 8 -SPT abolished the increases in flow induced by all of

the adenosine agonists confirming that Pj purinoceptors mediate this response. This

would also suggest that xanthine-resistant responses and A3 receptors are probably not

involved in mediating coronary vasodilation. This is contrary to the rat aorta where, in

comparison to NECA, the other adenosine agonists displayed some resistance to

blockade by 8 -SPT as the relaxations induced by NECA were inhibited by 8 -SPT to a

greater extent than the relaxations induced by the other agonists.

The A2a selective antagonist PD 115,199 (30nM) inhibited the responses induced by

NECA in the isolated heart shifting the concentration response curve to the right. A

dose-ratio of 3.8 was produced yielding an estimated pA2 value of 8 which is

comparable to the affinity of PD 115,199 in binding at A2 receptors in the rat striatum

(pKi of 7.8). This suggests that the receptors present on the rat coronary arteries are

of the A2 a subtype (Bruns et al., 1987). A lower concentration of PD 115,199 (lOnM)

virtually abolished the responses induced by CGS 21680 in the rat heart producing a

88

maximal response of Iml/min increase in flow at lOpM CGS 21680. As PD 115,199

not only shifted the concentration response curve for CGS 21680 to the right in a non­

parallel manner but also depressed the maximum, it is possible that PD 115,199 may

not be acting as a simple competitive antagonist at the A2 a receptors present on this

tissue. It is possible that PD 115,199 is acting as a non-competitive or irreversible

antagonist to produce the changes in the CGS 21680 concentration response curve

(Kenakin et a l, 1987). It has been previously suggested that NECA is acting as a full

agonist and CGS 21680 is acting as a partial agonist at the A2 receptors present on

the coronary arteries. In the presence of a non-competitive/ irreversible antagonist the

concentration response curve of a full agonist with a receptor reserve would be shifted

to the right in a parallel manner until the reserve has been depleted. The concentration

response curve of a partial agonist with no receptor reserve would be shifted to the

right but in a non-parallel manner as the maximum will be decreased. By definition, a

partial agonist must occupy all the receptors to elicit its full response. In the presence

of a non-competitive antagonist the number of receptors available to the agonist is

reduced due to the presence of the antagonist. Therefore, the maximum response

achievable by the partial agonist has been reduced, indicated by the decrease in the

maximum value of the agonist concentration response curve. If PD 115,199 was a

non-competitive/ irreversible antagonist the results obtained in this study would be

consistent with NECA acting as a full agonist and CGS 21680 acting as a partial

agonist on the A2 a receptors present in the coronary vessels. Non-competitive effects

such as a decrease in B^ax values have been seen in binding at A2 a receptors in the rat

striatum and at A receptors in the rat brain and guinea-pig atria using adenosine

antagonists such as CGS 15943 and PACPX (Williams et al., 1987b; Bumstock and

Hoyle, 1985). However, it is unlikely that PD 115,199 is acting as a non-competitive/

irreversible antagonist as changes in B ax values were not seen in binding at A2 a

receptors in the rat striatum (Bmns et a l, 1987).

89

Perfusion of the heart with the nitric oxide inhibitor L-NAME (30|xM) abolished the

responses to CPA and adenosine whilst shifting the concentration response curves to

CGS 21680 and NEC A to the right by approximately 30 fold and 20 fold respectively

(measured at 2ml/min increase in flow). The CPA and adenosine responses were

abolished due to the low potency of these agonists in the absence of the inhibitor. This

could suggest that the increases in flow induced by these adenosine agonists are

mediated partly via an endothelium dependent mechanism involving the nitric oxide

pathway and partly via receptors on the vascular smooth muscle. Alternatively the

concentration of L-NAME (30pM) used in the rat heart may not be high enough to

abolish the agonist induced vasodilations, although this is unlikely as the relaxations

induced by the adenosine agonists in the rat aorta were virtually abolished by the same

concentration of L-NAME. L-NAME also attenuated the responses induced by ATP

shifting the concentration response curve to the right, also suggesting that ATP

(possibly through conversion to adenosine) may be acting through a partially

endothelium-dependent mechanism. The effect of L-NAME on other endothelium

dependent vasodilators such as acetylcholine, bradykinin and the calcium ionophore

A23187 was investigated but it was found that the responses produced by these

agonists were not very consistent or reproducible both in the presence and absence of

L-NAME so the use of these agonists were not pursued. The concentration response

curve produced by the endothelium-independent vasodilator SNP was shifted slightly

to the right in the presence of L-NAME, although there was no significant difference

between the two curves (p> 0.05, Students't' test). Nitric oxide inhibitors such as L-

NAME and L-NNA have also been reported to inhibit the responses induced by

adenosine agonists in other blood vessels including the guinea-pig coronary artery, rat

renal artery and the rabbit pulmonary artery (Vials and Bumstock, 1993; Martin and

Potts, 1994; Steinhom et al., 1994). However, the majority of the blood vessels

studied have not been examined for endothelium dependency or appear to be

independent of the endothelial nitric oxide pathway (see Table 2).

90

Whilst carrying out the experiments with L-NAME it became apparent that other

factors were influencing the vasodilator responses produced by the adenosine agonists.

For example, whilst constructing the agonist concentration response curves the

vasodilator responses appeared to depend on the initial flow rate through the heart and

also perfusion of L-NAME appeared to decrease the basal flow and force of

contraction of the heart. Using the protocol above to produce agonist concentration

response curves on separate hearts, it was impossible to assess what effect these

changes were having on the adenosine vasodilator responses. Construction of full

agonist concentration response curves in the presence and absence of an antagonist on

one heart in a paired fashion was impossible due to limited viability of the heart,

therefore the effect of L-NAME, 8 -SPT and PD 115,199 on one concentration of SNP

(lOOpM) and NEC A (3pM) was investigated in a paired fashion on each heart (Table

4). This concentration of NEC A was chosen as it produced a similar increase in flow

to that achieved by SNP (lOOpM). The hearts were perfused with the inhibitors for 20

minutes to investigate their effects on the basal flow and the agonist induced increases

in flow in the isolated heart. L-NAME caused a significant decrease in the basal flow

through the heart (p < 0.05, Students't' test). Basal production of nitric oxide causes

a basal relaxant tone of the coronary vessels. Removal of this tone by inhibition by L-

NAME would effectively cause vasoconstriction and consequently a decrease in flow.

The adenosine antagonists 8 -SPT and PD 115,199 also caused a significant decrease in

the basal flow through the heart. It is likely that adenosine is being continually released

into the lumen of the coronary vessels and is stimulating the A2 a adenosine receptors

present on the coronary vessels to produce vasodilation and a relaxant tone. Perfusion

of the heart with the adenosine antagonists would effectively cause vasoconstriction

and therefore a decrease in the basal flow as the effects of the endogenous adenosine

on the A2 a receptors would be antagonised.

91

As well as decreasing the basal flow through the heart, perfusion of L-NAME, 8 -SPT

and PD 115,199 also significantly reduced the increase in flow induced by SNP (lOOji

M, p < 0.05, Students't' test). These results were rather unexpected as SNP is

considered to be an endothelium-independent vasodilator and therefore its responses

should not be affected by the presence of L-NAME. Indeed, the concentration

response curves for SNP in the presence and absence of L-NAME in unpaired hearts

were not significantly different, although a slight rightward shift in the SNP

concentration response curve was produced by L-NAME. The effects of SNP are not

mediated by adenosine receptors but by conversion to nitric oxide and direct activation

of guanylate cyclase in the vascular smooth muscle, so the adenosine antagonists

should also have no effect on the SNP response. The agonist, phenylephrine

(lOOnM) was used to investigate the effect of vasoconstriction on the basal flow

through the heart and the increases in flow induced by the agonists. PE not only

produced vasoconstriction of the coronary arteries indicated by a 31% decrease in the

basal flow through the heart, but also caused a 64% reduction in the SNP induced

increase in flow. It is possible that the vasoconstriction caused by PE may be

responsible for physiological antagonism of the SNP response and hence the decrease

in the vasodilation induced by this agonist. If this is the case, as L-NAME, 8 -SPT and

PD 115,199 effectively caused vasoconstriction by removing the basal relaxant tone of

the coronary vessels, the reduction in the SNP induced increase in flow produced by

these inhibitors could also be explained by physiological antagonism. The hearts were

also perfused with Krebs solution for 20 minutes to ascertain whether the basal and

agonist induced flows were affected by the time taken to perfuse the hearts with the

inhibitors, i.e. whether there was a time dependent effect on the vasodilator responses.

Over the 20 minute perfusion time the basal flow through the heart was not affected,

but the increase in flow induced by SNP was reduced by 33%. This decrease in the

SNP response can not be due to physiological antagonism as there was no change in

the basal flow and therefore no vasoconstriction of the coronary vessels. Therefore it

92

is likely that this 33% decrease in the SNP induced increase in flow is a time dependent

reduction. Consequently, it is possible that the decrease in the SNP response by L-

NAME, 8 -SPT and PD 115,199 could also be partly due to a time dependent

(approximately 33%) decrease as well as the influence of physiological antagonism.

A time dependent reduction in vasodilator responses would also explain the signiflcant

reduction (28%) in the NECA induced increase in flow seen over a 20 minute

perfusion of the heart with Krebs solution. Physiological antagonism due to

vasoconstriction could also explain the signiflcant reduction in the increase in flow

induced by NECA produced by L-NAME, 8 -SPT and PE (p < 0.05 and p < 0.1,

Students t ' test). The concentration response curves of NECA and CGS 21680 on

separate hearts were shifted to the right by the presence of L-NAME suggesting these

agonist responses were only partially endothelium-dependent. However when using

paired responses in the same heart, L-NAME caused a 3.5mls/min (6 8 %) reduction in

the increase in flow induced by NECA (3pM) which can be attributed to physiological

antagonism (Table 4). The true response of NECA in the presence of L-NAME would

therefore be approximately 3.5mls/min (6 8 %) higher than shown by the concentration

response curve. If the position of the NECA concentration response curve in the

presence of L-NAME was increased at each concentration by 3.5mls/min it would

become very similar to the concentration response curve produced in the absence of L-

NAME. This would imply that the vasodilation of the coronary arteries induced by

NECA is endothelium-independent. If this argument is repeated with the other

adenosine agonists it can be seen that in each case the concentration response curves in

the presence and absence of L-NAME become remarkably similar suggesting that all of

these adenosine agonist responses are probably endothelium-independent and are

mediated by adenosine receptors present on the vascular smooth muscle of the

coronary arteries. The apparent reduction in the agonist responses by L-NAME can in

each case be entirely explained by the effects of time and physiological antagonism.

93

The reduction of the NECA response caused by 8 -SPT (99.6%) was much greater than

the reduction in NECA response in the presence of L-NAME and Krebs solution (6 8 %

and 28% decrease in flow respectively). Using a one-way ANOVA analysis followed

by a Dunnetts multiple comparisons post-hoc test the reduction in the NECA induced

vasodilation caused by 8 -SPT was compared to the reduction seen with perfusion of L-

NAME and Krebs solution. A signiflcant difference (p < 0.05) was seen between these

values indicating that the reduction in the NECA induced increase in flow caused by 8 -

SPT was probably due to a combination of a time dependent reduction in response and

physiological antagonism, but also due to antagonism of the adenosine receptors. This

would imply that the inhibition of the adenosine agonist concentration response curves

by 8 -SPT seen earlier in this study was probably a true indication of the effect of 8 -

SPT. However, 8 -SPT and L-NAME caused approximately the same decrease in the

basal flow (28% and 24% respectively) and therefore a similar degree of physiological

antagonism would be expected. As the NECA induced increase in flow was decreased

by 6 8 % (3.5mls/min) by L-NAME due to physiological antagonism, a similar reduction

in the NECA induced response by 8 -SPT would also be expected due to the

physiological antagonism. If this is the case, to compensate for the physiological

antagonism an addition of approximately 3.5mls/min (6 8 %) to each concentration of

the response curves of the adenosine agonists in the presence of 8 -SPT would alter the

position of these curves. Hence, 8 -SPT would cause an inhibition of the adenosine

agonist responses rather than the abolition of the agonist responses that was observed.

Perfusion of the heart with PD 115,199 (30nM) only decreased the NECA induced

increase in flow by Iml/min (19%, Table 4). Again if this decrease is attributed to

physiological antagonism of the NECA response due to vasoconstriction of the

coronary vessels by PD 115,199 the true response of NECA in the presence of PD

115,199 is probably Iml/min higher than shown by the concentration response curve.

94

Increasing each concentration of the response curve for NECA in the presence of PD

115,199 by Iml/min makes this curve very similar the concentration response curve in

the absence of the antagonist. This suggests that PD 115,119 actually had no effect on

the vasodilation induced by NECA. As PD 115,199 is considered to be selective for

the A2a receptor this suggests that NECA induced vasodilation is not mediated via the

A2 a receptor. In contrast a lower concentration of PD 115,199 (lOnM) virtually

abolished the increase in flow induced by CGS 21680. Taking into consideration the

apparent vasoconstrictor effect of PD 115,199 shown by the decrease in basal flow,

this abolition of the CGS 21680 response cannot be explained entirely by physiological

antagonism and a time dependent reduction in response, but is probably also due to

antagonism of the A2 receptors. It was suggested earlier in this study that the greater

maximal response achieved by NECA compared to CGS 21680 could be explained by

a combination of both A2 a and A2y receptors in the coronary vessels. As the A2 a

selective antagonist PD 115,199 inhibits the CGS 21680 induced increase in flow but

does not inhibit the NECA induced increase in flow, this supports the suggestion that

CGS 21680 is acting via A2 receptors whilst NECA is acting via A213 receptors to

produce vasodilation of the coronary arteries.

As well as reducing the basal flow and the increases in flow induced by the agonists,

perfusion of the heart with L-NAME also decreased the force of contraction of the

heart. This action is probably a consequence of the decrease in the basal flow through

the heart induced by L-NAME. The decrease in flow would lead to an insufficient

perfiision and oxygenation of the heart and consequently a decrease in the ability of the

heart muscle to contract due to hypoxia and ischaemia. To establish whether the

converse was also true, the effect of SNP on the heart was also investigated. Perfusion

of the heart with SNP not only caused an increase in flow through the heart but also

caused an increase in the force and rate of contraction of the heart. However, SNP did

not have any effect on the isolated atria whereas isoprenaline significantly increased

95

both the rate and force of contraction of the atria, (p < 0.05, Students't' test). As

SNP did not have a positive chronotropic or inotropic effect on the atria it is unlikely

that the increase in force and rate of contraction of the isolated heart induced by SNP

was due to a direct positive chronotropic or inotropic effect. Instead the effect of SNP

on the rate and force of the heart was probably an indirect consequence of the

increases in flow induced by SNP. This would result in the opposite effect to L-

NAME, allowing greater perfiision and oxygenation of the heart enabling it to perform

better so an increase in the force and rate of contraction could be achieved. Nitric

oxide has recently been shown to act as a negative inotrope in cardiac muscle (Shah

and Lewis, 1993; Brady et al., 1993; Schülz and Triggle, 1994). Therefore SNP, a

nitric oxide donor would be expected to have a negative inotropic effect on the heart

whereas L-NAME an nitric oxide synthase inhibitor would be expected to have the

opposite effect and act as a positive inotrope. However it is clear from our study, that

L-NAME and SNP have negative and positive inotropic actions respectively.

Therefore it is unlikely that L-NAME and SNP are producing direct effects on the

heart as the changes in the force of contraction of the heart produced by these

compounds in this study are opposite to the effects which would be expected.

Therefore in this study it appears that L-NAME or SNP produced changes in the force

and rate of contraction of the heart through indirect effects via changes in vessel tone

and subsequent changes in flow rate through the heart.

In conclusion, the adenosine agonists produced an increase in flow through the isolated

heart indicating vasodilation of the coronary arteries. The agonist potency order

produced in the isolated heart would suggest that this response was mediated by A2

receptors, although the high potency of the A2 a selective agonist CGS 21680 would

suggest that these receptors are probably of the A2 a receptor subtype. Whilst carrying

out experiments with the nitric oxide synthase inhibitor L-NAME and the adenosine

antagonists 8 -SPT and PD 115,199 it became apparent that other factors such as

96

decreases in basal flow and changes in force of contraction of the heart were

influencing the vasodilator responses produced by the adenosine agonists. L-NAME,

8 -SPT and PD 115,199 produced decreases in the basal flow through the heart

indicating that these inhibitors induced vasoconstriction, which led to physiological

antagonism of agonist responses and hence a reduction in the vasodilation achieved by

these agonists. In addition to the physiological antagonism a time dependent reduction

in the agonist responses also had to be considered when interpreting the agonist

induced increases in flow in the presence and absence of the inhibitors. In the isolated

heart the non-selective antagonist 8 -SPT inhibited the increases in flow induced by the

adenosine agonists conflrming the presence of P^ receptors in the coronary vessels and

also ruling out the involvement of xanthine-resistant responses and A3 receptors in

coronary vasodilation. After taking into consideration the physiological antagonism

caused by PD 115,199, it was found that although PD 115,199 inhibited the increases

in flow induced by CGS 21680 it had no effect on the increases in flow induced by

NECA. This suggests that CGS 21680 is acting via A2 receptors to mediate coronary

vasodilation whereas NECA is not acting via A2 g receptors but is possibly via A21,

receptors to mediate this response. L-NAME also caused physiological antagonism of

the increases in flow induced by the agonists by causing vasoconstriction of the

coronary vessels. When this was taken into consideration it was found that L-NAME

had no effect on the increases in flow induced by the adenosine agonists implying that

these responses were endothelium-independent and were mediated via the vascular

smooth muscle. In addition to causing decreases in the basal flow and the agonist

induced increases in flow, L-NAME also caused a decrease in the force of contraction

of the heart. In contrast, SNP caused an increase in the force and also the rate of

contraction of the heart although it did not have any effect on the isolated atria. It is

likely that the responses induced by L-NAME and SNP are not direct effects but are

due to the changes in the vessel tone and subsequent changes in the flow rate through

the heart caused by these compounds.

97

4.3. CONCLUSION

In this study the presence of adenosine receptors on the isolated rat thoracic aorta and

the rat coronary arteries, studied using the Langendorff isolated heart preparation,

have been confirmed. Adenosine and some of its analogues such as NECA, CGS

21680, CPA and APNEA induce vasodilations of these vessels producing agonist

potency orders which are indicative of the presence of A2 receptors on these tissues.

Due to the high potency of the A2 a selective agonist CGS 21680 A2 a receptors are

likely to be present on both the rat aorta and coronary arteries. However, the maximal

response achieved by CGS 21680 in the thoracic aorta was only 6 6 % of the NECA

maximal response, which led to the conclusion that although NECA is acting as a full

agonist on the A2 receptors present on the aorta CGS 21680 is only acting as a partial

agonist achieving a smaller maximum value than that obtained by NECA. The

endogenous nucleoside MTA acted as very weak agonist inducing relaxations of the

thoracic aorta only at very high concentrations. This response was not inhibited by the

non selective antagonist 8 -SPT suggesting that MTA does not act via the A2 a

receptors present on the aorta but on another site on the aorta which is resistant to

blockade by 8 -SPT. However, MTA also appears to act as an antagonist on the A2 a

receptors on the rat thoracic aorta inhibiting the responses induced by the adenosine

agonists with a potency similar to that of 8 -SPT.

MTA, when used as an antagonist, the non selective antagonist 8 -SPT and the A2

selective antagonist PD 115,199 all inhibited the relaxations of the rat thoracic aorta

induced by the adenosine agonists. However in the presence of these antagonists

NECA was inhibited to a greater extent than the other agonists. It is likely that all of

the adenosine agonists, with the exception of MTA, are acting on the A2 a receptors

present on the rat aorta. However, as CGS 21680, CPA and adenosine were inhibited

98

to a lesser extent they may also be acting on another site on the aorta to mediate

vasodilation. This may also be the site which mediates the weak reponses induced by

MTA. The smaller dose-ratios produced by 8 -SPT and PD 115,199 using these

agonists could be explained by agonist stimulation of this second site which displays

some resistance to blockade by the adenosine antagonists. In the presence of 8 -SPT,

MTA or PD 115,199 the maximal response produced by CGS 21680 increased so that

CGS 21680 achieved a similar maximal response to that produced by NECA. This

suggests that although CGS 21680 acts as a partial agonist on the A2 a receptors of the

aorta, in the presence of these antagonists, CGS 21680 appears to be a full agonist

activating a second site at which it is able to achieve a higher maximum value.

These unusual findings could be explained by the presence of other adenosine

receptors present on the thoracic aorta. However, using DPCPX at an A% selective

concentration (InM) no evidence could be found for the presence of relaxant Aj

receptors on this vessel. Also the different degrees of antagonism and xanthine-

resistant effects of MTA could not be explained by the presence of A3 receptors as the

agonist potency order produced in the presence of 8 -SPT and MTA in this study is

different to the agonist potency order which would be expected in the presence of A3

receptors.

Initially in the isolated rat heart the effects of the adenosine antagonists and other

inhibitors such as the nitric oxide synthase inhibitor L-NAME on the agonist

concentration response curves were carried out in a similar fashion to the rat aorta,

using unpaired isolated hearts. However, it emerged that several problems arose with

this protocol as the interpretation of the agonist induced changes in flow became

difficult due to other effects such as a decrease in basal flow caused by the antagonists

and L-NAME. Experiments were therefore carried out so that the effect of the

inhibitors on the basal flow and the increases in flow induced by SNP and NECA could

99

be established on individual hearts in a paired manner. This made the interpretation of

the changes in flow rate easier and in some cases altered the conclusions made about

these results. It was found that L-NAME, phenylephrine and the adenosine

antagonists 8 -SPT and PD 115,199 all decreased the basal flow through the heart

indicating vasoconstriction of the coronary vessels. As all of these compounds also

reduced the increase in flow induced by NECA and SNP, it was proposed that the

reductions in vasodilation induced by these agonists were due to physiological

antagonism caused by the vasoconstriction produced by the inhibitors. Also perfusion

of the heart with Krebs solution for the same amount of time required for perfusion of

the inhibitors (20 minutes) caused a decrease in the SNP and NECA induced

responses. This could not be attributed to physiological antagonism as no change in

basal flow was observed, hence this was due to a time dependent reduction in the

agonist induced responses. Therefore it became apparent that the interpretation of the

changes in flow produced by the adenosine agonists and the effects of the antagonists

and inhibitors on these changes in flow also had to take into account a time dependent

reduction in responses as well as physiological antagonism which could oppose the

vasodilator responses produced by the agonists.

In the isolated rat heart the non selective antagonist 8 -SPT abolished the increases in

flow induced by all of the adenosine agonists. This confirmed that the vasodilation of

the coronary vessels is mediated by Pj purinoceptors, and also suggested that

xanthine-resistant responses and A3 receptors are probably not involved in this

response. As mentioned previously, when considering the effect of 8 -SPT on the

agonist induced responses the physiological antagonism due to the decrease in the

basal flow caused by 8 -SPT should also be taken into account. If the possible effect of

physiological antagonism is added onto the concentration responses curves of the

adenosine agonists in the presence of 8 -SPT it can be seen that this antagonist probably

inhibits the effects of the agonists rather than abolishing them.

100

In the isolated heart PD 115,199 (30nM) inhibited the NECA response shifting the

concentration response curve slightly to the right whilst a lower concentration of PD

115.199 (lOnM) virtually abolished the CGS 21680 induced increase in flow. This

suggests that both of these agonists act via the A2 a receptors present on the coronary

arteries, although this would not explain the different antagonistic effects of PD

115.199 on NECA and CGS 21680. The initial explanation for the different effects of

PD 115,199 was that it was acting as a non-competitive /irreversible antagonist. It has

been suggested that NECA is a full agonist and CGS 21680 is a partial agonist on the

A2 a receptors on the coronary arteries. In this case, the effect of such an antagonist on

their concentration response curves would be different, the concentration response

curve for NECA being shifted to the right in a parallel manner whilst the concentration

response curve for CGS 21680 would be shifted rightward in a non parallel manner

with a reduced maximum value. However this explanation is unlikely as no evidence

for non-competitive effects were seen for PD 115,199 in binding at A2 receptors in

the rat striatum. Moreover, because PD 115,199 caused vasoconstriction of the

coronary vessels indicated by the decrease in basal flow, some of the inhibitory effect

of PD 115,199 on the agonist induced responses was probably due to physiological

antagonism. When the effect of physiological antagonism was added to the NECA

concentration response curve in the presence of PD 115,199 it was found that this

curve became very similar to the curve in the absence of the antagonist. Unlike the

previous interpretation this suggests that NECA is not acting on A2 a receptors on the

coronary arteries. The different maximal responses produced by NECA and CGS

21680 in the isolated heart could instead have been due to the presence of two

receptors with different maximal values, for example the A2 and A2y receptors.

Therefore it is possible that the NECA induced response is mediated via A2y receptors

which may be present on the coronary arteries. In contrast, the virtual abolition of the

CGS 21680 induced response by PD 115,199 was probably due to a combination of a

101

time dependent reduction in response, physiological antagonism as well as antagonism

of the A2 a receptor mediated vasodilation induced by CGS 21680.

The relaxations induced by the adenosine agonists in the isolated rat aorta were

virtually abolished by the mechanical removal of the endothelium suggesting these

responses were endothelium-dependent. As the agonist responses were also virtually

abolished by the presence of the nitric oxide synthase inhibitor L-NAME this also

implies that these responses occur through an endothelium-dependent mechanism

involving the nitric oxide pathway. In the rat isolated heart the situation was slightly

more complex. Initially it was found that L-NAME abolished the response induced by

CPA and adenosine, probably due to their low potency in the absence of the inhibitor,

and also inhibited the responses induced by NECA and CGS 21680 shifting their

concentration response curves to the right. This would imply that the agonist induced

responses were only partially endothelium-dependent and were mediated via adenosine

receptors present on both the endothelium and the vascular smooth muscle. However,

L-NAME was found to cause vasoconstriction indicated by the reduction in the basal

flow through the heart, and this vasoconstriction caused a physiological antagonism

and therefore a reduction of the vasodilator responses produced by the adenosine

agonists. When the reduction in response due to the physiological antagonism was

added on to the agonist concentration response curves in the presence of L-NAME it

was found that the concentration response curves in the presence and absence of the

inhibitor were remarkably similar. This would imply that L-NAME has no effect on

the responses induced by the adenosine agonists, and that, unlike the rat thoracic aorta,

the vasodilations induced by the adenosine agonists are endothelium independent and

do not involve the nitric oxide pathway.

As well as decreasing the basal flow and the agonist induced flows through the heart.

L-NAME also decreased the force of contraction of the heart. SNP produced the

102

opposite effect and increased not only the force but also the rate of contraction of the

heart although it did not affect the force or rate of contraction of isolated atria. These

actions of L-NAME and SNP were therefore probably not direct effects but

consequences of the changes in the flow rate induced by these compounds. For

example, the decrease in basal flow caused by L-NAME would lead to an insufficient

perfusion and oxygenation of the heart and consequently a decrease in the hearts ability

to contract to its fiill potential due to hypoxia and ischaemia. Conversely, SNP caused

an increase in flow through the heart and hence better perfusion and oxygenation of the

heart and therefore the heart was able to perform better so an increase in the force and

rate of contraction of the heart was observed.

In conclusion, the characterisation of adenosine receptors present on the isolated rat

thoracic aorta and the rat coronary arteries, studied using the Langendorff isolated

heart technique has been carried out. On the rat thoracic aorta the relaxations induced

by the adenosine agonists were mediated via A2 receptors, with the exception of the

relaxations induced by MTA which were not inhibited by 8 -SPT. However, NECA

was inhibitied to a greater extent than the other agonists by 8 -SPT, MTA and PD

115,199 indicating that that the relaxations induced by the other agonists were

mediated by A2 receptors but also by another site which displays some resistance to

blockade by xanthines. The differing degrees of antagonism and xanthine-resistant

responses do not appear to involve Aj or A3 receptors. The relaxations induced by the

agonists in the rat aorta were abolished by removal of the endothelium and inhibition

by L-NAME therefore these responses are endothelium-dependent and involve the

nitric oxide pathway. In the rat coronary arteries the adenosine agonists induced

increases in flow through the heart indicating vasodilation of these vessels. L-NAME,

8 -SPT and PD 115,199 caused vasoconstriction of the coronary arteries which led to

physiological antagonism of the increases in flow induced by the agonists and hence a

reduction in the vasodilation induced by these agonists. This physiological antagonism

103

and also a time dependent reduction in the agonist responses had to be taken into

consideration when interpretation of the results was carried out. It appears that the

adenosine agonists mediate vasodilation of the coronary arteries via A2 receptors,

although due to its high potency it is likely that CGS 21680 is acting via A2 a receptors.

However as the increase in flow induced by NECA was not inhibited by the A2

selective antagonist PD 115,199 it appears that NECA is not acting via A2 a receptors

but possibly via A213 receptors to induce coronary vasodilation. When the effect of

physiological antagonism was taken into consideration, L-NAME did not have any

effect on the increases in flow induced by the adenosine agonists, so therefore these

responses were endothelium-independent and are mediated by the vascular smooth

muscle of the coronary arteries.

Characterisation of the purinoceptors present on the rat thoracic aorta and the

coronary arteries is based on agonist potency orders, the potency of the A2 a selective

agonist CGS 21680 and the use of selective antagonists such as PD 115,199.

Classification of the receptors on these and other tissues would become easier and

more reliable by development of more selective agonists and antagonists, especially for

the A2b and A3 receptors. The use of selective A3 antagonists in the rat thoracic aorta

would confirm that A3 receptors were not involved in the relaxation of the rat aorta.

A3 receptors antagonists such as BW-A522 and BW-A1433 have recently been

developed and could be used in the rat aorta to eliminate the presence of A3 receptors,

although they are not selective for the A3 receptor versus the Aj and A2 a receptors

(Kim et a l, 1994b). As the increases in flow induced by NECA in the isolated heart

were not mediated via A2 receptors it was suggested that this response was mediated

via A215 receptors. With selective A2y receptor agonists and antagonists it would be

possible to establish whether NECA does indeed mediate coronary vasodilation via

these receptors. It was found in the study on the isolated heart that interpretation of

the increases in flow induced by the agonists were complicated by physiological

104

antagonism due to vasoconstriction produced by the adenosine antagonists. Therefore

when using adenosine antagonists in the isolated heart the effect of physiological

antagonism on the agonist responses must be considered. The contribution of the

physiological antagonism produced by PD 115,199 on the increase in flow induced by

CGS 21680 should be established to ascertain the true vasodilator response produced

by CGS 21680 in the presence of this antagonist. PD 115,199 could also be used to

investigate whether the responses induced by the other agonists used in this study,

CPA and adenosine were also mediated via A2 receptors, like CGS 21680, or possibly

via A213 receptors, like NECA. The unusual results in the rat thoracic aorta using

MTA prompt further investigation of the effects of this endogenous nucleoside. It has

been reported that MTA acts as a xanthine-sensitive weak or partial agonist on A2

receptors on rat PC 12 cells and as an antagonist on A215 receptors on human fibroblast

and neuroblastoma cells (Daly and Padgett, 1992; Bruns et ah, 1986; Munshi et al.,

1988). However, in this study on the rat aorta MTA acted as a weak agonist on a

xanthine-resistant site and an antagonist on A2 a receptors present on the aorta. It

would be interesting to ascertain the actions of MTA on a wider range of tissues and

especially other isolated blood vessels to establish whether MTA has similar actions to

those seen on the rat aorta. It was noted during the discussion that the guinea-pig

trachea and the rat aorta appear to be very similar with respect to the differing degrees

of antagonism of the agonist responses and the weak xanthine-resistant effects of

MTA. It is possible that the xanthine-resistant site on the aorta and the guinea-pig

trachea may be identical, although in both cases further investigation of this site would

be needed.

105

REFERENCES

Abebe, W., Makujina, S.R. and Mustafa, SJ. (1994). Adenosine receptor-mediated

relaxation of porcine coronary artery in presence and absence of endothelium. Am. J.

Physiol, 266, H2018-H2025.

Agerwal, K.C. and Parks, RE. (1980). 5'- methylthioadenosine and 2',5'-

dideoxyadenosine blockade of the inhibitory effects of adenosine on ADP-induced

platelet aggregation by different mechanisms. Biochem. Pharmacol, 29, 2529-2532.

Alexander, S.P.H., Losinski, A., Kendall, D A. and Hill, S.J. (1994). A comparison of

A2 adenosine receptor-induced cyclic AMP generation in cerebral cortex and

relaxation of pre-contracted aorta. Br. J. Pharmacol, 1 1 1 , 185-190.

Bailey, S.J., Hickman, D. and Hourani, S.M.O. (1992). Characterisation of the P%

purinoceptors mediating contraction of the rat colon muscularis mucosae. Br. J.

Pharmacol, 105, 400-404.

Balwierczak, J.L., Sharif, R , Krulan, C.M., Field, F.P., Weiss, G.B. and Miller, M.J.S.

(1991). Comparative effects of a selective adenosine A2 receptor agonist, CGS 21680,

and nitroprusside in vascular smooth muscle. Eur. J. Pharmacol, 196, 117-123.

Belardinelli, L and Isenberg, G (1983a). Isolated atrial myocytes: adenosine and

acetylcholine increase potassium conductance. Am. J. Physiol, 244, H734-H737.

Belardinelli, L. and Isenberg. G. (1983b). Actions of adenosine and isoproterenol on

isolated mammalian ventricular myocytes. Circ. Res., 53, 287-297.

Belardinelli, L., Giles, W.R. and West, A. (1988). Ionic mechanisms of adenosine

actions in pacemaker cells from rabbit heart. J. Physiol, 405, 615-633.

107

Berne, R.M. (1963). Cardiac nucleotides in hypoxia; Possible role in regulation of

coronary blood flow. Am. J. Physiol, 204, 317-322.

Brackett, L.E. and Daly, J.W. (1991). Relaxant effects of adenosine analogs on guinea-

pig trachea in vitro: xanthine-sensitive and xanthine-insensitive mechanisms. J.

Pharmacol Exp. Ther., 257, 205-213.

Brady, A.J.B., Warren, J.B., Poole-Wilson, P.A., Williams, T.J. and Harding, S.E.

(1993). Nitric oxide attenuates cardiac myocyte contraction. Am. J. Physiol., 265,

H176-H182.

Bridges, A.J., Bruns, R.F., Ortwine, D.G., Priebe, S.R, Szolek, D.L. and Trivedi, B.K.

(1988). N^-[2(3,5-dimethoxyphenyl)-2- (2-methylphenyl)-ethyl] adenosine and its

uronamide derivatives. New adenosine agonist with both high affinity and high

selectivity for adenosine A2 receptor. J. Med. Chem., 31, 1282-1285.

Brizzolara, A.L. and Burnstock, G. (1991). Endothelium-dependent and endothelium-

independent vasodilatation of the hepatic artery of the rabbit. Br. J. Pharmacol, 103,

1206-1212.

Brown, C.M. and Collis, M.G. (1982). Evidence for an A2/Ra adenosine receptor in

the guinea-pig trachea. Br. J. Pharmacol, 76, 381-387.

Bruns, R.F. (1980a). Adenosine receptor activation in human fibroblasts: nucleoside

agonists and antagonists. Can. J. Physiol Pharmacol, 58, 673-691.

Bruns R.F. (1980b). Adenosine antagonism by purines, pteridines and benzopteridines

in human fibroblasts. Biochem. Pharmacol, 30, 325-333.

108

Bruns, R.F., Lu, G.H. and Pugsley, T A. (1986). Characterisation of the A2 adenosine

receptor labelled by [^H]-NECA in the rat striatal membrane. M ol Pharmacol, 29,

331-346.

Bruns, R.F., Fergus, J.H., Badger, E.W., Bristol, J.A., Santay, L A. and Hays, S.J.

(1987). PD 115,199; An antagonist ligand for adenosine A2 receptors. Naunyn-

Schmeidebergs'Arch. Pharmacol, 335, 64-69.

Bruns, R.F. and Fergus, J.H. (1989). Adenosine receptors in brain membranes: Binding

of N^-cyclohexyl[^H]adenosine and l,3-diethyl-8-[^H]phenylxanthine. Proc. Natl

Acad. Sci., 11, 5547-5551.

Burnstock, G. (1978). A basis for distinguishing two types of purinergic receptors. In

Cell membrane receptors for drugs and hormones: A multidiscipline approach. Ed. by

Bolis, L. and Straub, R.N. pp. 107-118. Raven Press: New York

Burnstock, G , Hills, J.M. and Hoyle, H.V. (1984). Evidence that the Pj purinoceptor

in the guinea pig taenia coli is an A2 subtype. Br. J. Pharmacol, 81, 533-542.

Bumstock, G. and Buckley, N. (1985). The classification of receptors for adenosine

and adenine nucleotides. In Methods used in adenosine research: Methods in

pharmacology. Ed. by Patton, D M. pp. 193-212. Plenium Press : New York.

Bumstock, G. and Hoyle, C.H.V. (1985). PAPCX- A substituted xanthine antagonises

both the Aj and A2 subclasses of the Pj purinoceptor: Antagonism of the A2 subclass

is competitive but antagonism of the Aj subclass is not. Br. J. Pharmacol, 85, 291-

296.

109

Carmthers, A.M. and Fozard, J R. (1992). Mechanisms of the hypotensive response to

selective adenosine Aj receptor agonists in the pithed rat. Br. J. Pharmacol., 106, 94P.

Carmthers, A.M. and Fozard, J R. (1993). Hypotensive responses to the putative

adenosine A3 receptor agonist N^-2-(4 aminophenyl)ethyl adenosine (APNEA) in the

rat. Br. J. Pharmacol., 106, 6 8 P.

Cass, C.E., Gaudette, L A. and Peterson, A.R.P. (1974). Mediated transport of

nucleosides in human erythrocytes. Specific binding of the inhibitor

nitrobenzylthioinosine to nucleoside transport sites in the erythrocyte membrane.

Biochim. Biophys. Acta., 354, 1-10.

Cassis, L.A., Loeb, A.L. and Peach, M.J. (1987). Mechanisms of adenosine- and ATP-

induced relaxation in rabbit femoral artery: Role of the endothelium and cyclic

nucleotides. In Topics and perspectives in adenosine research. Ed. by Gerlach, E. and

Becker, B.F. pp486-496. Springer Verlag: Berlin.

Chem, Y., King, K., Lai, H-L. and Lai, H-T. (1992). Molecular cloning of a novel

adenosine receptor gene firom rat brain. Biochem. Biophys. Res. Comm., 185, 304-309.

Choi, O.H., Shamin, M.T., Padgett, W.L. and Daly, J.W. (1988). Caffeine and

theophylline analogues: Correlation of behavioural effects with activity as adenosine

receptor antagonists and phosphodiesterase inhibitors. Life sci., 43, 387-398.

Churchill, P C. and Churchill, M.C. (1985). A and A2 adenosine receptor activation

inhibits and stimulates renin secretion of rat renal cortical slices. J. Pharmacol. Exp.

Ther., 232, 589-594.

110

Clemo, H.F. and Belardinelli, L. (1986). Effect of adenosine on atrioventricular

conduction. I. Site and characterization of adenosine action in the guinea pig

atrioventricular node. Circ. Res., 59, 427-436.

Collis, M.G. (1983). Evidence for an adenosine receptor in the guinea pig atrium.

Br. J. Pharmacol, 78, 207-212.

Collis, M.G. and Brown, C.M. (1983). Adenosine relaxes the aorta by interacting with

an A2 receptor and an intracellular site. Eur, J. Pharmacol., 96, 61-69.

Collis, M.G., Jacobson, K.A. and Tomkins, D M. (1987). Apparent affinity of some 8 -

phenyl-substituted xanthines at adenosine receptors in guinea-pig aorta and atria. Br. J.

Pharmacol, 92, 69-75.

Collis, M.G, Stoggall, S.M. and Martin, F.M. (1989). Apparent affinity of 1,3-

dipropyl-8 -cyclopentylxanthine for adenosine Aj and A2 receptors in isolated tissues

from guinea-pigs. Br. J. Pharmacol, 97, 1274-1278.

Collis, M.G. and Hourani, S.M.O. (1993). Adenosine receptor subtypes. Trends in

Pharmacol Sci., 14, 360-366.

Conti, A., Monopoli, A., Gamba, M., Borea, P.A. and Ongini, E. (1993). Effects of

selective Aj and A2 adenosine receptor agonists on cardiovascular tissues. Naunyn-

Schiedeberg's Arch Pharmacol, 348, 108-112.

Cornfield, L.J., Hu, S., Hurt, S.D. and Sills, M.A. (1992). pH]2-

phenylaminoadenosine (pH]CV 1808) labels a novel adenosine receptor in rat brain. J.

Pharmacol Exp. Ther., 263, 552-561.

I l l

Cristalli, G., Vittori, S., Thompson, R.D., Padgett, W.L., Shi, D., Daly, J.W. and

Olsson, R.A. (1994). Inhibition of platelet aggregation by adenosine receptor agonists.

Naunyn-Schmiedehergs' Arch. Pharmacol., 349, 644-650.

Croning, M.D.R., Gurden, M.F. and Kennedy, I. (1992). Potencies of adenosine

receptor-blocking drugs on dog isolated coronary artery and guinea-pig isolated aorta.

Br. J. Pharmacol., 105, 303P.

Cushing, D.J., Brown, G.L., Sabouni, M.H. and Mustafa, S.J. (1991). Adenosine

receptor-mediated coronary artery relaxation and cyclic nucleotide production. Am. J.

Physiol, 261, H343-H348.

Daly, J.W. (1982). Adenosine receptor subtypes: Target sites for drugs. J. Med.

Chem.,lZ, 197-207.

Daly, J.W., Butts-Lamb, P. and Padgett, W. (1983). Subclasses of adenosine receptors

in the central nervous system: Interactions with caffeine and related methylxanthines.

Cell M ol Neurobiol, 3, 69-80.

Daly, J.W. and Padgett, W.L. (1992). Agonist activity of 2 and 5' substituted

adenosine analogues and their N^- cycloalkyl derivatives at A and A2 adenosine

receptors coupled to adenylate cyclase. Biochem. Pharmacol, 43, 1089-1093.

De La Haba, G. and Cantoni, G.L. (1959). The enzymatic synthesis of S-adenosyl-L-

homocysteine from adenosine and homocysteine. J. Biol Chem., 234, 603-608.

De Mey, J.G. and Vanhoutte, P.M. (1981). Role of the intima in cholinergic and

purinergic relaxation of isolated canine femoral arteries. J. Physiol, 316, 347-355.

112

Des Rosiers, C. and Nees, S. (1987). Functional evidence for the presence of

adenosine A2 receptors in cultured coronary endothelial cells. Naunyn-Schmiedehergs'

Arch. Pharmacol, 336, 94-98.

Dionisotti, S., Zocchi, C , Varane, K., Borea, P.A. and Ongini, E. (1992). Effects of

adenosine derivatives on human and rabbit platelet aggregation correlation of

adenosine receptor affinities and antiaggregatory activity. Naunyn-Schmiedehergs’

Arch. Pharmacol, 346, 673-676.

Drury, A.N. and Szent-Gyorgi, A. (1929). The physiological activation of adenine

compounds with especial reference to their activity on mammalian heart. J. Physiol,

6 8 , 213-237.

Evans, D., Schenden, J. and Bristol, J.A. (1982). Adenosine receptors mediating

cardiac depression. Life Sci., 31, 2425-2432.

Farmer, S.G., Canning, B.J. and Wilkins, D.E. (1988). Adenosine receptor-mediated

contraction and relaxation of guinea-pig isolated tracheal smooth muscle: effects of

adenosine antagonists. Br. J. Pharmacol, 95, 371-378.

Fink, J.S., Weaver, D.R, Rivkees, S.A., Peterffeund, R.A., Pollack, A.E., Adler, E.M.

and Reppert, S.M. (1992). Molecular cloning of the rat A2 adenosine receptor:

Selective co-expression with D2 dopamine receptors in rat striatum. Mol Brain Res.,

14, 186-195.

Fleetwood, G , Coade, S B., Gordon, J.L. and Pearson, J.D. (1989). Kinetics of

adenine nucleotide catabolism in coronary circulation of rats. Am. J. Physiol, 256,

H1565-H1572.

113

Fleetwood, G. and Gordon, J.L. (1987). Purinoceptors in the heart. Br. J. Pharmacol.,

90, 210-227.

Fozard, J.R. and Carmthers, A.M. (1993a). Adenosine A3 receptors mediate

hypotension in the angiotension-II supported circulation of the pithed rat. Br. J.

Pharmacol, 109, 3-5.

Fozard, J.R. and Carmthers, A.M. (1993b). The cardiovascular effects of selective

adenosine and A2 receptor agonists in the pithed rat: No role for glibenclamide-

sensitive potassium channels. Naunyn-Schmiedebergs’ Arch. Pharmacol, 347, 192-

196.

Fozard, J.R. and Hannon, J.P. (1994). BW-A522 blocks adenosine A3 receptor-

mediated hypotensive responses in the rat. Eur. J. Pharmacol, 252, R5-R6.

Furlong, T.J., Pierce, K.D., Selbie, LA. and Shine, J. (1992). Molecular

characterization of a human brain adenosine A2 receptor. M ol Brain R es., 15, 62-66.

Gallo-Rodriguez, C , Ji, X-D., Melman, N., Siegman, B.D., Sanders, L.H., Orlina, J.,

Fischer, B., Pu, Q , Olah, M.E., Van Galen, P.J.M., Stiles, G.L. and Jacobson, K.A.

(1994). Stmcture-activity relationships of N^-benzyladenosine-5'-uronamises as A3 -

selective adenosine agonists. J. Med. Chem., 37, 636-646.

Ghai, G , Francis, J.E., Williams, M., Dotson, R.A., Hopkins, M.F., Cote, D.T.,

Goodman, F.R. and Zimmerman, M B. (1987). Pharmacological characterization of

CGS 15943A: CGS 15943A: A novel nonxanthine adenosine antagonist. J.

Pharmacol Exp. Ther., 242, 784-790.

114

Ghia, G. and Mustafa, S.J. (1982). Demonstration of a putative adenosine receptor in

rabbit aorta. Bloodvessels, 19, 117-125.

Gordon, J.L. and Martin, W. (1983). Endothelium dependent relaxation of the pig

aorta: relationship to the stimulation of ^^Rb efflux from endothelial cells. Br. J.

Pharmacol, 79, 531-541.

Graier, W.F., Groschner, K., Schmidt, F. and Kukovetz, W.R. (1992). Increases in

endothelial cyclic AMP levels amplify agonist-induced formation of endothelium-

derived relaxing factor (EDRF). Biochem. J., 288, 345-349.

Griffith, T.M., Edwards, D. H., Lewis, M.J. and Henderson, A.H. (1985). Evidence

that cyclic guanosine monophosphate (cGMP) mediates endothelium-dependent

relaxation. Eur. J. Pharmacol, 112, 195-202.

Gurden, M.F, and Kennedy, I. (1992). Characterisation of the adenosine receptors in

dog coronary artery, guinea-pig aorta and guinea-pig fundus. Br. J. Pharmacol, 105,

115P.

Gurden, M.F., Coates, J., Ellis, F., Evans, B., Foster, M., Hornby, E., Kennedy, I.,

Martin, D.P., Strong, P., Vardey, C.J. and Wheeldon, A. (1992). Functional

characterization of three adenosine receptor types. Br. J. Pharmacol, 109, 693-698.

Hamilton, H.W., Taylor, M.D., Steffen, R.P., Haleen, S.J. and Bruns, R.F. (1987).

Correlation of adenosine receptor affinities and cardiovascular activity. Life Set., 41,

2295-2302.

115

Hannon, J.P., Pfannkuche, H.J. and Fozard, J.R. (1995). A role for mast cells in

adenosine A3 receptor-mediated hypotension in the rat. Br. J. Pharmacol., 115, 945-

952.

Hargreaves, M.B., Stoggall, S.M. and Colhs, M.G. (1991). Evidence that the

adenosine receptor mediating relaxation in dog lateral saphenous vein and guinea-pig

aorta is of the A2y subtype. Br. J Pharmacol, 102,198P.

Harms, H.H., Wardeh, G. and Mulder, AH. (1978). Adenosine modulates

depolarization-induced release of ^H-noradrenaline from slices of rat brain neocortex.

Eur. J. Pharmacol, 49, 305-308.

Headrick, J.P. and Berne, R.M. (1990). Endothelium-dependent and -independent

relaxations to adenosine in guinea pig aorta. Am. J. Physiol, 259, H62-H67.

Herlihy, J.T., Bockman, E.L., Berne, R.M. and Rubio, R. (1976). Adenosine relaxation

of isolated vascular smooth muscle. Am. J. Physiol, 230, 1239-1243.

Hide, I., Padgett, W.L., Jacobson, K.A. Daly, J.W. (1992). A^^ adenosine receptors

from rat striatum and rat pheochromocytoma PC 12 cells: Characterization with

radiohgand binding and by activation of adenylate cyclase. M ol Pharmacol, 41, 352-

359.

Hill, S.J. and Kendall, D A. (1987). Studies on the adenosine-receptor mediating the

augmentation of histamine-induced inositol phospholipid hydrolysis in guinea-pig

cerebral cortex. Br. J. Pharmacol, 91, 661-669.

116

Hutchinson, A.J., Webb, R.L., Oei, H.H., Ghai, G.R., Zimmerman, M B. and Williams,

M. (1989). CGS 21680, an A2 selective adenosine receptor agonist with preferential

hypotensive activity. Br. J. Pharmacol., 251, 47-55

Ignarro, L.J., Buga, G.M., Wood, K.S., Byms, R.E. and Chaudhuri, G. (1987).

Endothelium-derived relaxing factor produced and released from artery and vein is

nitric oxide. Proc. Natl. Acad. Sci., 84, 9265-9269.

Ignarro, L.J. (1989). Endothelium-derived nitric oxide: Actions and properties. FASEB

y., 3, 31-36.

Jackson, K.K., Herzer, W.A. and Suzuki, F. (1993). KF 17837 is an A2 adenosine

receptor antagonist in vivo. J. Pharmacol. Exp.Ther., 267, 1304-1310.

Jacobson, M.A. (1995). Molecular biology of adenosine receptors. In Adenosine and

adenine nucleotides. From molecular biology to integrative phvsiologv. Ed by

Belardinelli, L. and Pelleg, A. pp5-13. Kluwer Academic Press, Boston.

Jacobson, K.A. (1990). Adenosine (P%) and ATP (P2 ) receptors. In Comprehensive

Medicinal Chemistrv. Vol. 3 Membranes and Receptors. Ed. by Emmett, J.C. pp 601-

642. Perrgamon Press, Oxford.

Jacobson, K.A., Nikodijevic, O , Padgett, W.L., Gallo-Rodriguez, C , Maillard, M. and

Daly, J.W. (1993). 8-(3-chlorostyryl)caffeine (CSC) is a selective A2 adenosine

antagonist in vitro and in vivo. FebsLett., 323, 141-144.

Jahnel, U. and Nawrath, H. (1989). Characterisation of adenosine receptors in guinea-

pig isolated left atria. Br. J. Pharmacol, 97, 1182-1190.

117

Jarvis, M.F. and Williams, M. (1987). Adenosine and dopamine function in the CNS.

Trends in Pharmacol Sci., 8,330-331.

Jarvis, M.F., Schultz, R., Hutchinson, A.J., Un Hoi Sills, M.A. and Williams, M.

(1989). [^H]-CGS 21680, A selective A2 adenosine receptor agonist directly labels A2

receptors in the rat brain. J. Pharmacol. Exp. Ther., 251, 888-893.

Keef, K.D., Pasco, J.S. and Eckman, DM. (1992). Purinergic relaxation and

hyperpolarization in guinea-pig and rabbit coronary artery: Role of the endothelium. J.

Pharmacol. Exp. Ther., 260, 592-600.

Kenakin, T.P. (1987). In Pharmacological analvsis of drug-receptor interaction. Raven

Press: New York.

Kenakin, T.P., Bond, R.A. and Bonner, T.I. (1992). Definition of pharmacological

receptors. Pharmacol. Rev., 44, 351-362.

Kennedy, C. and Bumstock, G. (1985). ATP produces vasodilation via Pj

purinoceptors and vasoconstriction via P2 purinoceptors in the rabbit central ear

artery. Blood Vessels, 22, 145-155.

Kennedy, C , Delbro, D. and Bumstock, G. (1985). P2 purinoceptors mediate both

vasodilation {via endothelium) and vasoconstriction of the isolated rat femoral artery.

Eur. J. Pharmacol., 107, 161-168.

Kennedy, I., Gurden, M.F., Strong, P. (1992a). Do adenosine A3 receptors e x is tGen.

Pharmacol., 23, 303-307.

118

Kennedy, I., Gurden, MF., Strong, P., Hornby, E., Foster, M. and Vardey, C.J.

(1992b). Pharmacological classification of adenosine receptors in isolated tissues. Int.

J. Purine and Pyrimidine Res., 3, 073.

Kim, H.O., Ji, X-D., Siddiqi, S.M., Olah, ME., Stiles, G.L. and Jacobson, K.A.

(1994a). 2-substitution of N^-benzyladenosine-5’-uronamides enhances selectivity for

A3 adenosine receptors. J. Med. Chem., 37, 3614-3621.

Kim, H.O., Ji, X-D., Melman, N., Olah, ME., Stiles, G.L. and Jacobson, K.A.

(1994b). Structure-activity relationships of 1,3-dialkylxanthine derivatives at rat A3

adenosine receptors. J. Med. Chem., 37, 3373-3382.

King, A., Milavec-Krizman, M. and Müller-Schweinitzer, E. (1990). Characterisation

of the adenosine receptor in porcine coronary arteries. Br. J. Pharmacol., 100, 483-

486.

Konduri, G G , Woodward, L.L., Mukhopadhyay, A. and Deshmukh, D R (1992).

Adenosine is a pulmonary vasodilator in newborn lambs. Am. Rev. Respir. Disease.

146, 670-676.

Kurtz, A. (1987). Adenosine stimulates guanylate cyclase activity in vascular smooth

muscle cells. J. Biol. Chem., 262, 6296-6300.

Kusachi, S., Thompson, R.D. and Olsson, R.A. (1983). Ligand selectivity of dog

coronary adenosine receptor resembles that of adenylate cyclase stimulatory (R^)

receptors. J. Pharmacol. Exp. Ther., 227, 316-321.

Liang, B.T. (1992). Adenosine receptors and cardiovascular fimction. Trends

Cardiovasc. Med., 2, 100-108.

119

Libert, F., Parmentier, M., Lefort, A., Dinsart, C , Van-Sande, J., Maenhaut, C ,

Simons, M-J., Dumont, LE. and Vassart, G. (1989). Selective amplification and

cloning of four new members of the G Protein-coupled receptor family. Science, 244,

569-572.

Libert, F., Schiffinann, S.N., Lefort, A., Parmentier, M., Gerard, C , Dumont, LE.,

Vanderhaeghen, J-J. and Vassart, G. (1991). The orphan receptor cDNA RDC7

encodes an Aj adenosine receptor. EMBO., 1 0 , 1677-1682.

Libert, F., Van Sande, L, Lefort, A., Czemilofsky, A., Dumont, I.E., Vassart, G.,

Ensinger, H.A. and Mendia, K.D. (1992). Cloning and functional characterisation of a

human Aj adenosine receptor. Biochem. Biophys. Res. Comm., 187, 919-926.

Lincoln, T.M. (1989). Cyclic GMP and mechanisms of vasodilation. Pharmacol,

Ther., 41, 479-502.

Linden, L, Tucker, A.L., Robeva, A.S., Graber, S G and Munshi, R (1993a).

Properties of recombinant adenosine receptors. Drug Development Res., 28, 232-236.

Linden, L, Taylor, H E., Robeva, A.S., Tucker, A.L., Stehle, J.H., Rivkees, S.A., Fink,

J.S. and Reppert, S.M. (1993b). Molecular cloning and functional expression of a

sheep A3 adenosine receptor with widespread tissue distribution. M ol Pharmacol, 44,

524-532.

Lloyd, H.G.E. and Schrader, J. (1987). The importance of the transmethylation

pathway for adenosine metabolism in the heart. In Topics and perspectives in

adenosine research. Ed. by Gerlach, E. and Becker, B.F. pp 199-207. Springer Verlag:

Berlin

120

Londos, C , Cooper, D M. and Wolff, J. (1980). Subclasses of external adenosine

receptors. Proc. Natl. Acad. Sci., 11, 2551-2554.

Londos, C. and Wolff, J. (1977). Two distinct adenosine-sensitive sites on adenylate

cyclase. Proc. Natl. Acad. Sci., 74, 5482-5486.

Losinski, A., Hill, S.J. and Alexander, S.P.H. (1993). A2 adenosine receptor relaxation

of guinea-pig aorta. Br. J. Pharmacol., 108, 147P.

Luthin, D.R. and Linden, J. (1995). Comparison of A4 and A2 a binding sites in

striatum and COS cells transzfected with adenosine A2 a receptors. J. Pharmacol. Exp.

2 7 2 , 5 1 1 - 5 1 8 .

Maenhaut, C , Van Sande, J. Libert, F., Abramowicz, M., Parmentier, M.,

Vanderhaegen, J-J., Dumont, J.E., Vassart, G. and Schiffinan, S. (1990). RDC8 codes

for an adenosine A2 receptor with physiological constitutive activity. Biochem.

Biophys. Res. Comm., 173, 1169-1178.

Mahan, L.C., McVittie, L.D., Smyk-Randall, E.M., Nakata, H., Monsma, F.J., Gerfen,

C.R. and Sibley, D.R. (1991). Cloning and expression of an A adenosine receptor

from rat brain. Mol. Pharmacol., 40,1-7.

Makujina, S R., Sabouni, M.H., Bhatia, S., Douglas, F.L. and Mustafa, S.J. (1992).

Vasodilatory effects of adenosine A2 receptor agonists CGS 21680 and CGS 22492 in

human vasculature. Eur. J. Pharmacol., 221, 243-247.

121

Makujina, S.R., Olsson, R.A., Esinhart, J.D. and Mustafa, S.J. (1993). Structure-

activity relationships of 2 -(ar) alkoxyadenosines at the adenosine A2 receptor in the

coronary artery. Eur. J. Pharmacol., 243, 35-38.

Martin, P.L. (1992). Relative agonist potencies of C2-substituted analogues of

adenosine: evidence for adenosine A2B receptors in the guinea pig aorta. Eur. J.

Pharmacol., 216, 235-242.

Martin, P.L., Ueeda, M. and Olsson, R A. (1993). 2-phenylethoxy-9-methyladenine:

An adenosine receptor antagonist that discriminates between A2 adenosine receptors in

the aorta and the coronary vessels from the guinea-pig. J. Pharmacol. Exp. Ther.,

265, 248-253.

Martin, P.L. and Potts, A. A. (1994). The endothelium of the rat renal atery plays an

obligatory role in A2 adenosine receptor-mediated relaxation induced by 5’-N-

ethylcarboxamidoadenosine and N^-cyclopentyladenosine. J. Pharmacol. Exp. Ther.,

270, 893-899.

Mathie, R.T., Alexander, B., Ralevic, R. and Bumstock, G. (1991). Adenosine-

induced dilatation of the rabbit hepatic arterial bed is mediated by A2 -purinoceptors.

Br. J. Pharmacol., 103, 1103-1107.

Mathieson, J.J.I. and Bumstock, G. (1985). Purine mediated relaxation and

constriction of isolated rabbit mesenteric artery are not endothelium dependent. Eur. J.

Pharmacol, 118, 221-229.

McCormack, D.G., Clarke, B. and Barnes, P.J. (1989). Characterization of adenosine

receptors in human pulmonary arteries. Am. J. Physiol, 256, H41-H46.

122

Meyerhof, W., Müller-Brechlin, R. and Richter, D. (1991). Molecular cloning of a

novel putative G-protein coupled receptor expressed during rat spermiogenesis. Fehs

Lett., 284, 155-160.

Moncada, S., Palmer, R.M.J. and Higgs, E.A. (1991). Nitric oxide: Physiology,

pathophysiology and pharmacology. Pharmacol Rev., 43,109-142.

Moritoki, H., Matsugi, T., Takase, H., Ueda, H, and Tanioka, A. (1990). Evidence for

the involvement of cyclic GMP in adenosine induced, age dependent vasodilation., Br.

J. Pharmacol, 1 0 0 , 569-575.

Munshi, R , Clanachan, A.S. and Baer, HP. (1988). 5'-deoxy 5' methylthioadenosine: a

nucleoside which differentiates between adenosine receptors sites. Biochem.

Pharmacol, 37, 2085-2089.

Murray, R.D. and Churchill, P C (1984). Effects of adenosine receptor agonists in the

isolated, perfused rat kidney. Am. J. Physiol, 247, H343-H348.

Mustafa, S.J. and Askar, AO (1985). Evidence suggesting an Ra-type adenosine

receptor in bovine coronary arteries., J. Pharmacol. Exp. Ther., 232,49-56.

Newby, A C (1984). Adenosine and the concept of retaliatory metabohtes. Trends in

Biochemical Sci., 9, 42-44.

Nichofis, J., Hourani, S.M.O. and Kitchen, I. (1992). Characterisation of the Pj

purinoceptors on rat duodenum and urinary bladder. Br. J. Pharmacol, 105, 639-642.

123

Nikodyevic, O , Sarges, R., Daly, J.W. and Jacobson, K.A. (1991). Behavioral effects

of Aj- and A2 -selective adenosine agonists and antagonists: Evidence for synergism

and antagonism. J. Pharmacol. Exp. Ther., 259, 286-294.

Nishida, Y., Suzuki, S. and Miyamoto, T. (1985). Pharmacological action of 5’-

methylthioadenosine on isolated rabbit aorta strips. Blood Vessels, 22, 229-233.

Nonaka, H., Ichimura, M., Takeda, M., Nonaka, Y , Shimada, J., Suzuki, F.,

Yamaguchi, K. and Kase., H. (1994). KF 17837 ((E)-8-(3,4-dimethylxanthine), a

potent and selective adenosine A2 receptor antagonist. Eur. J. Pharmacol., 267, 335-

341.

Odawara, S., Kurahashi, K., Usiu, H., Taniguchi, T. and Fujiwara, M. (1986).

Relaxations of isolated rabbit coronary artery by purine derivatives: A2 -adenosine

receptors. J. Cardiovasc. Pharmacol., 8 , 567-573.

Oei, H.H., Ghai, G.R., Zoganas, H.C., Stone, G.A., Zimmerman, MB., Field, F.P. and

Williams, M. (1988). Correlation between binding affinities for brain Aj and A2

receptors of adenosine agonists and antagonists and their effects on heart rate and

cardiovascular tone. J. Pharmacol, Exp, Ther., 247, 882-888.

Olah, M E., Ren, H., Ostrowski, J., Jacobson, K.A. and Stiles, G.L. (1992). Cloning,

expression and characterization of the unique bovine Aj adenosine receptors. J. Biol.

Chem., 267, 10764-10770.

Oliver, J.M. and Paterson, A.R.P. (1971). Nucleoside transport. I. Mediated process in

human erthyrocytes. Can. J. Biochem., 49, 262-270.

124

Olsson, R.A. and Pearson, J.D. (1990). Cardiovascular purinoceptors. Pharmacol

Rev., 70, 761-845.

Palmer, R.M.J., Ferrige, A.G. and Moncada, S. (1987). Nitric oxide release accounts

for the biological activity of endothelium-derived relaxing factor. Nature, 317, 524-

525.

Palmer, R M.J.and Moncada, S. (1989). A novel citrulline-forming enzyme implicated

in the formation of nitric oxide by vascular endothelial cells. Biochem. Biophys. Res.

Comm., 158, 348-352.

Palmer, R.M.J., Ashton, D.S. and Moncada, S. (1988). Vascular endothehal cells

synthèse nitric oxide from L-arginine. Nature, 333, 664-666.

Paterson, A.R.P. and Oliver, J.M. (1971). Nucleoside transport. II. Inhibition by p-

nitrobenzylthioguanosine and related compounds. Can. J. Biochem., 255, 6535-6538.

Pearson, J.D. and Slakey, L.L. (1990). Intracellular and extracellular metabolism of

adenosine and adenine nucleotides. In Purines in Cellular Signalling. Ed. by Jacobson,

K.A., Daly, J.W. and Manganiello, V. ppl3-19. Springer-Verlag: New York.

Peet, N.P., Lentz, N.L., Meng, E C., Dudley, M.W., Ogden, A-M.L., Demeter, D.A.,

Weintraub, H.J.R. and Bey, R (1990). A novel synthesis of xanthines: Support for a

novel binding model for xanthines vdth respect to adenosine at adenosine receptors. J.

Med. Chem., 33, 3127-3130.

Petcoff, D.W. and Cooper, D.M.F. (1987). Adenosine receptor agonists inhibit inositol

phosphate accumulation in rat striatal slices. Eur. J. Pharmacol, 137, 269-271.

125

Pierce, K.D., Furlong, T.J., Selbie, L.A. and Shine, J. (1992). Molecular cloning and

expression of an adenosine A215 receptor from human brain. Biochem. Biophys. Res.

Comm., 187, 86-93.

Poucher, S.M., Keddie, J.R., Singh, P., Stoggall, S.M., Caulkett, P.W.R., Jones, G.

and Collis, M.G. (1995). The in vitro pharmacology of ZM 241385, a potent, non­

xanthine, A2 a selective adenosine receptor antagonist. Br. J. Pharmacol., 115, 1096-

1102 .

Prentice, D.J. and Hourani, S.M.O. (1995). NEC A and R-PIA activate distinct sites to

cause relaxation of the isolated rat aorta. Br. J. Pharmacol., 115, 64P.

Ramagopal, M.V., Chitwood, J.R.W. and Mustafa, S.J. (1988). Evidence for an A2

adenosine receptor in human coronary arteries. Eur. J. Pharmacol., 151, 483-486.

Ramkumar, V., Stiles, G.L., Beaven, M.A. and Ali, H. (1993). The A3 adenosine

receptor is the unique adenosine receptor which facilitates release of allergic mediators

in mast cells. J. Biol. Chem., 268, 16887-16890.

Rapoport, R.M. and Murad, F. (1983). Endothelium-dependent and nitrovasodilator-

induced relaxation of vascular smooth muscle: Role of cyclic GMP. J. Cycl. Nucl.

Prot. Phosph. Res., 9, 281-296.

Rapoport, R.M. (1986). Cyclic guanosine monophosphate inhibition of contraction

may be mediated through inhibition of phosphotidylinositol hydrolysis in rat aorta.

Circulation Res., 58, 407-410.

126

Rapoport, R.M., Draznin, MB. and Murad, F. (1993). Endothelium dependent

relaxation in rat aorta may be mediated through cyclic GMP dependent protein

phosphorylation. Nature, 306, 174-176.

Rees, D.D, Palmer, R.M.J., Hodson, H.F. and Moncada, S. (1989). A specific inhibitor

of nitric oxide formation from L-arginine attenuates endothelium-dependent relaxation.

Br. J. Pharmacol., 96, 418-424.

Rees, D.D, Palmer, R.M.J., Schulz, R , Hodson, H.F. and Moncada, S. (1990).

Characterisation of 3 inhibitors of endothelial nitric oxide synthase in vitro and in vivo.

Br. J. Pharmacol., 101, 746-752.

Reppert, S.M., Weaver, D.R , Stehle, J.H. and Rivkees. (1991). Molecular cloning

and characterization of a rat K y adenosine receptor that is widely expressed in brain

and spinal cord. Mol. Endocrinol., 5, 1037-1048.

Ribeiro, J.A. and Sebastiao, A. (1986). Adenosine receptors and calcium: A basis for

proposing a third (A3) adenosine receptor. Progr. Neurobiol, 26, 179-210.

Rivkees, S.A. and Reppert, S.M. (1992). RFL9 encodes an A2y-adenosine receptor.

Mol. Endocrinol., 6, 1598-1604.

RoseMeyer, R.B. and Hope, W. (1990). Evidence that A2 purinoceptors are involved

in endothelium-dependent relaxation of the rat thoracic aorta. Br. J. Pharmacol, 100,

576-580.

Rubanyi, G. and Vanhoutte, P.M. (1985). Endothelium removal decreases relaxations

of canine coronary arteries caused by 13-adrenergic agonists and adenosine. J.

Cardiovasc. Pharmacol, 7, 139-144.

127

Rubino, A., Ralevic, V. and Bumstock, G. (1995). Contribution of Pj-(A2b subtype)

and P2 ~purinoceptors to the control of vascular tone in the rat isolated mesenteric

arterial bed. Br. J. Pharmacol., 115, 648-652.

Sabouni, M.H., Brown, G.L., Kotake, A.N., Douglas, F.L. and Mustafa, S.J. (1990).

Effects of CGS-15943A on the relaxations produced by adenosine analogs in human

blood vessels. Eur. J. Pharmacol., 187, 525-530.

Sabouni, M.H., Cushing, D.J. and Mustafa, S.J. (1989). Adenosine receptor-mediated

relaxation in coronary artery: Evidence for a guanyl nucleotide-binding regulatory

potein involvement. J. Pharmacol, Exp. Ther., 251, 943-948.

Sajjadi, F.G. and Firestein, G.S. (1993). cDNA cloning and sequence analysis of the

human A3 adenosine receptor. Biochim. Biophys. Acta, 1179, 105-107.

Salvatore, C.A., Luneau, C.J., Johnson, R.G. and Jacobson, M.A. (1992). Molecular

cloning and characterization of human A1 and A2 adenosine receptors. Int. J. Purine

and Pyrimidine Res., 3, 82(abstr)

Salvatore, C.A., Jacobson, M.A., Taylor, HE., Linden, J. and Johnson, R.G. (1993).

Molecular cloning and characterization of the human A3 adenosine receptor. Proc.

Natl Acad. Sci., 90, 10365-10369.

Sattin, A. and Rail, T.W. (1970). The effect of adenosine and adenine nucleotides on

cyclic AMP content of guinea-pig cerebral cortex slices. M ol Pharmacol, 6 , 13-23.

Schiele, J.O. and Schwabe, U. (1994). Characterization of the adenosine receptor in

microvascular coronary endothelial cells. Eur J. Pharmacol, 269, 51-58.

128

Schiffinan, S.N., Libert, F,, Vassart, G , Dumont, LE. and Vanderhaegen, J-J. (1990).

A cloned G protein-coupled protein with a distribution restricted to striatal medium­

sized neurons. Possible relationships with Dj dopamine receptors. Brain Res., 519,

333-337.

Schimmel, R.J. and Elliot, M E. (1988). Adenosine inhibits phenylephrine activation of

phospholipase A2 in hamster brown adipocytes. Biochem. Biophys. Res. Comm., 152,

886-892.

Schrader, J., Schütz, W. and Bardenheuer, H. (1981). Role of S-adenosylhomocysteine

hydrolase in adenosine metabolism in mammalian heart. Biochem. J., 196, 65-70.

Schrader, J., Borst, M.M., Kelm, M., Bading, B. and Bürrig, K.F. (1990). Formation

of adenosine in the heart from extracellular adenine nucleotides. In Purines in Cellular

Signalling. Ed. by Jacobson, K.A., Daly, J.W. and Manganiello, V. pp33-40. Springer-

Verlag: New York.

Schultz, R. and Triggle, C.R. (1994). Role of NO in vascular smooth muscle and

cardiac muscle fimction. TIPS, 15, 255-259.

Schütz, W., Schrader, J. and Gerlach, E. (1981). Different sites of adenosine formation

in the heart. Am. J. Physiol., 240, H963-H970.

Shah, M.A. and Lewis, M.J. (1993). Modulation of myocardial contraction by

endocardial and coronary vascular endothelium. Trends Cardiovasc. Med., 3, 98-103.

129

Shimada, J., Suzuki, F., Nonaka, H., Ishii, A. and Ichikawa, S. (1992). (E)-l,3-dialkyl-

7-methyl-8-(3,4,5-dimethoxy-styryi)xanthines: Potent and selective adenosine A2

antagonists. J. Med. Chem., 35, 2342-2345.

Spielman, W.S. and Arend, L.J. (1991). Adenosine receptors and signalling in the

kidney. Hypertension, 17,117-130.

Stehle, J.H., Rivkees, S.A., Lee, J.J., Weaver, D R., Deeds, J.D. and Reppert, S.M.

(1992). Molecular cloning and expression of the cDNA for a novel A2 -adenosine

receptor suntype. Mol. Endocrinol., 6, 384-393.

Steinhom, R.H., Morin, F.C., Van Wylen, D.G.L., GuGino, S.F., Giese, E.C. and

Russell, J.A. (1994). Endothelium-dependent relaxations to adenosine in juvenile rabbit

pulmonary arteries and veins. Am. J. Physiol., 266, H2001-H2006.

Stoggall, S.M. and Shaw, J.S. (1990). The coexistence of adenosine Aj and A2

receptors in guinea-pig aorta. Eur. J. Pharmacol., 190, 329-335.

Stone, T.W. (1981). Physiological roles for adenosine and adenosine 5'-triphosphate in

the nervous system. Neurosci., 6, 523-545.

Tucker, A.L., Linden, J., Robeva, A.S., D’angelo, D.D. and Lynch, K.K. (1992).

Cloning and expression of a bovine adenosine Aj receptor cDNA. Febs Lett., 297,

107-111.

Ueeda, M., Thompson, R.D., Arroyo, L.H. and Olsson, R.A. (1991a). 2-

alkoxyadenosine: Potent and selective agonists at the coronary artery A2 adenosine

receptor. J. Med. Chem., 34, 1334-1339.

130

Ueeda, M., Thompson, R.D., Padgett, W.L., Secunda, S., Daly, J.W. and Olsson, R.A.

(1991b). Cardiovasular actions of adenosines, but not adenosine receptors, differ in rat

and guinea-pig. Life Sci., 49, 1351-1358.

Van Calker, D., Müller, M. and Hamprecht, B. (1979). Adenosine regulates via two

different types of receptor, the accumulation of cyclic AMP in cultured brain cells. J.

Neurochem., 33, 999-1005.

Van Galen, P.J.M., Van Bergen, A.H., Gallo-Rodriguez, C , Melman, N., Olah, M.E.,

Ijerman, P., Stiles, G.L. and Jacobson, K.A. (1994). A binding site model and structure

activity relationships for the rat A3 receptor. Mol Pharmacol., 45, 1101-1111.

Vials, A. and Bumstock, G. (1993). Aj-purinoceptor-mediated relaxation in the

guinea-pig coronary vasculature: a role for nitric oxide. Br. J. Pharmacol, 109, 424-

429.

Vuorinen, P., Porsti, I., Metsa-Ketela, T., Manninen, V., Vapaatalo, H. and Laustiola,

K.E. (1992). Endothelium-dependent and -independent effects of exogenous ATP,

adenosine, GTP and guanosine on vascular tone and cyclic nucleotide accumulation of

rat mesenteric artery. Br. J. Pharmacol, 105, 279-284.

(!

Waldman, S. A. and Murad, F. (1987). Cyclic GMP synthesis and function. Pharmacol

i^ev., 39, 163-196.

White, T.D. and Angus, J.A. (1987). Relaxant effect of ATP and adenosine on canine

large and small coronary arteries in vitro. Eur. J. Pharmacol, 143,119-126.

131

Williams, M., Francis, J., Ghai, G., Braunwalder, A., Psychoyos, S., Stone, G.A. and

Cash, W.D. (1987a). Biochemical characterization of the triazoloquinazoline CGS

15943, a novel non-xanthine adenosine antagonist. J. Pharmacol Exp. Ther., 241,

415-420.

Williams, M., Jarvis. M.F., Sills, M.A., Ferkany, J.W. and Braunwalder, A. (1987b).

Biochemical characterization of the antagonists actions of the xanthines, PACPX (1,3-

dipropyl-8(2-amino-4-chloro)phenyl xanthine) and 8 -PT (8 -phenyltheophylline) at

adenosine Aj and A2 receptors in rat brain tissue. Biochem. Pharmacol, 36, 4024-

4027.

Williams, M. (1989). Adenosine: the prototypic neuromodulator. Neurochem. Int., 14,

249-264.

Yen, M-H., Wu, C-C. and Chiou, W-F. (1988). Partially endothelium-dependent

vasodilator effect of adenosine in rat aorta. Hypertension, 11, 514-518.

Zhou, Q-Y , Li, C , Olah, M.E., Johnson, R.A., Stiles, G.L. and Civelli, O. (1992).

Molecular cloning and characterization of an adenosine receptor: The A3 adenosine

receptor. Proc. Natl Acad. Sci., 89, 7432-7436.

UNVERSIW OF SURREY UBRÂRV

132