Citation
Interaction of adenosine receptors in a smooth muscle cell line

Material Information

Title:
Interaction of adenosine receptors in a smooth muscle cell line
Creator:
Xie, Fan, 1965-
Publication Date:
Language:
English
Physical Description:
viii, 121 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Agonists ( jstor )
Cell growth ( jstor )
Cell membranes ( jstor )
Down regulation ( jstor )
Incubation ( jstor )
Inurement ( jstor )
Pneumothorax ( jstor )
Pretreatment ( jstor )
Rats ( jstor )
Receptors ( jstor )
Adenosine -- pharmacology ( mesh )
Department of Pharmacology and Therapeutics thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pharmacology and Therapeutics -- UF ( mesh )
Muscle, Smooth -- cytology ( mesh )
Muscle, Smooth -- physiology ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1992.
Bibliography:
Includes bibliographical references (leaves 113-120).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Fan Xie.

Record Information

Source Institution:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
26842050 ( OCLC )
026885827 ( ALEPH )

Downloads

This item has the following downloads:


Full Text











INTERACTION OF ADENOSINE RECEPTORS
IN A SMOOTH MUSCLE CELL LINE














BY


FAN XIE
















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY



UNIVERSITY OF FLORIDA 1992
































This dissertation is dedicated to my grandmother, father, mother and brother.















ACKNOWLEDGEMENTS


First and foremost, I wish to sincerely thank my mentor, Dr. Stephen Baker, for his guidance, both professional and personal, and for the contribution he has made to my education. I would also like to thank Dr. Luiz Belardinelli for his enthusiasm, valuable suggestions and constant supply of wonderful drugs. Many thanks are also expressed to my helpful and friendly committee: Mohan Raizada, Edwin Meyer and Thomas Rowe. I wish to thank all members of Dr. Baker's and Dr. Belardinelli's laboratory, especially Debbie Otero, Dr. John Shryock, Mary Anne Locksmith, Cheryl Spence and Xingmin Tang for their support and technical assistance. I also deeply thanks Drs. Allen Neims, David Silverman, Chingkuang Tu and Thomas Muther for their confidence in me. A special word of thanks goes to Dr. Sandra Rattray for her editorial help. My best wishes are also extended to all my fellow (and former fellow) graduate students especially Sukanya Kanthawatana, Nelida Sjak-Shie, Walter Folger, Daniel Danso, Jiahui Zhang and Magdalena Wozniak; thanks for their friendship and encouragement. I would also like to thank other faculty members who worked so hard to improve the graduate program, and the secretarial and administrative staff who keep the department running. Special thanks are




iii










given to Judy Adams and Barbara Reichert for being so nice to me no matter how many times I bothered them.



















































iv
















TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .........................................iii

ABSTRACT .... ............. ................................ vi

CHAPTERS

1 INTRODUCTION........................................ 1

Functions of Adenosine......................... 1
Synthesis and Metabolism of Adenosine .......... 5 Adenosine Receptors............................ 7
Regulation of Adenosine Receptors .............. 20
Goals................................................ 24

2 EXPERIMENTAL PROCEDURES

Source of Materials ............................ 27
Methods........................................ 28

3 INTERACTION OF ADENOSINE RECEPTORS

Introduction................................... 35
Results.............................................. 37
Discussion..................................... 49
Summary ........................................ 62

4 INHIBITORY EFFECT OF Al-ADENOSINE RECEPTOR ON
IRREVERSIBLE ACTIVATION OF THE 8-ADRENORECEPTOR

Introduction................................... 94
Results.............................................. 96
Discussion..................................... 99

LIST OF REFERENCES....................................... 113

BIOGRAPHICAL SKETCH..... ................................. 121










v















Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy


INTERACTION OF ADENOSINE RECEPTORS
IN A SMOOTH MUSCLE CELL LINE By

Fan Xie

May, 1992


Chairman: Stephen P. Baker, Ph.D. Major Department: Pharmacology and Therapeutics


DDT1 MF-2 cells have been shown to express inhibitory Al and stimulatory A2 adenosine receptors (AdoRs) coupled to 3', 5'-cyclic adenosine monophosphate (cAMP) accumulation. The objective of this study was to investigate the possible interaction between the two AdoRs. The AdoR agonists, adenosine and 5'-N-ethylcarboxamido-adenosine (NECA) attenuated isoproterenol (ISO)-stimulated cAMP accumulation in a dose-dependent manner with a maximal inhibition of 68% and 98%, respectively. No cAMP stimulation was observed with either compound. In contrast, the selective A2-AdoR agonist 2-[2-(2 naphthyl)ethoxy]adenosine (WRC-0018) produced a biphasic response. Stimulation of cAMP accumulation (8-fold) occurred at low concentrations (5 500 nM) followed by an attenuation at high concentrations (>500 nM). The attenuation component was prevented by 1) the selective A1-AdoR vi










antagonist () N6-endonorbornan-2-yl-9-methyladenine (N-0861, 10 M), 2) pretreatment of cells with pertussis toxin (PTX, 25 ng/ml, 18 hr) which uncoupled the inhibitory A1-AdoR response or 3) pretreatment of cells with the selective A1AdoR agonist 2-chloro-N6-cyclopentyladenosine (CCPA, 0.1 JIM, 16 hr). CCPA-pretreatment reduced by 13-fold the potency of the A1-AdoR agonist N6-cyclopentyladenosine (CPA) to inhibit ISO-stimulated cAMP formation, and decreased the A1-AdoR level by 48%. Stimulation of cAMP accumulation by adenosine and NECA was uncovered in the presence of N-0861 and by PTXpretreatment. However, no stimulation by either agonist was observed after CCPA-pretreatment. The data indicate that the inhibitory Al-AdoR response in DDT1 MF-2 cells is predominant and masks the A2-AdoR mediated stimulatory effect. The A2-AdoR response was expressed by a selective A2-AdoR agonist or under conditions where the function of the A1-AdoR is blocked.

The ability of the activated A1-AdoR to modulate agonist interaction with the beta-adrenoreceptor (BAR) was studied using the irreversible BAR agonist, 5-[2-[[3-[4(bromoacetamido)phenyl]-2-methylprop-2-yl]amino]-1hydroxyethyl]-8-hydroxycarbostyril (C-Br). Activation of the A1-AdoR attenuated cAMP accumulation of the permanently stimulated BAR, and did not alter the irreversible binding of C-Br. In addition, CPA decreased basal cAMP level and had no effect on the interaction of the reversible BAR agonist ISO with the BAR. These data indicate that Al-AdoR inhibitory


vii










effect is not mediated by alteration of agonist interaction with the BAR but rather occurs via a post-receptor mechanism.





















































viii















CHAPTER 1
INTRODUCTION


Functions of Adenosine


History


Drury and Szent-Gyorgyi (1929) were the first to report on the cardiovascular effects of adenosine and adenine nucleotides. They described the isolation of crystalline adenine from acid extracts of ox heart muscle. Adenosine was crystalized from yeast nucleic acid hydrolysate. Intravenous injection of either extract into different mamalian species after atropinization produced primarily sinus bradycardia, transient heart block and other physiological effects (Drury and Szent-Gyorgyi, 1929).

Interestingly, Drury and Szent-Gyorgyi did not comment on the physiological implications of their observations. Within two years, however, Lindner and Rigler (1931) had crystallized adenosine, obtained from the degradation of AMP, from heart muscle extracts. They showed that adenosine was a potent coronary vasodilator in a number of species. Based on the findings that adenosine was present in heart muscle extract and had potent vasoactive effects, Lindner and Rigler advanced the hypothesis that adenosine is a physiological





1






2



regulator of coronary blood flow. This hypothesis received little immediate attention and interest at the time.

It was not until 1963 that Berne (1963) and Gerlach et al. (1963) independently revived Lindner and Rigler's adenosine hypothesis by demonstrating the release of adenosine catabolites from hypoxic or ischemic heart muscle. The revived hypothesis states that an imbalance between oxygen supply and oxygen demand leads to alterations of the cellular release of adenosine, which in turn changes the contractile state of vascular smooth muscle of the resistance vessels.

In 1970, Sattin and Rall (1970) and Shimizu and Daly (1970) proposed the existence of extracellular adenosine receptors (AdoRs) based on the observation that adenosine and some adenine nucleotides elevate adenosine 3', 5'-cyclic monophosphate (cAMP) in cerebral cortical slices. This effect was competitively inhibited by methylxanthines, caffeine and theophylline. Within a decade, it became clear that there are AdoRs that inhibit as well as others that stimulate adenylyl cyclase (AC) (Londos and Wolff, 1977; Van Calker et al., 1979), constituting a system for the bidirectional control of the catalytic activity of this key enzyme. Physiological Effects


Adenosine is present in every cell of the human body and exerts a wide spectrum of effects on various tissues and





3



organs. For example, in the heart, adenosine is a potent coronary vasodilator (Berne, 1963; Gerlach et al., 1963). It has also been shown to depress cardiac activity, e.g., (1) depress sinoatrial and atrioventricular (AV) node activity,

(2) reduce atrial contractility, (3) attenuate the stimulatory actions of catecholamines primarily in ventricular myocardium, and (4) depress ventricular automaticity (Belardinelli et al., 1989). These actions characterize adenosine as an endogenous cardioprotective substance whose actions lead to an increase in oxygen supply and decrease in cardiac work. Together, these actions tend to restore the balance between oxygen supply and demand.

In most tissues including skeletal muscle, adenosine is a vasodilator. Thus, intravenous infusion of adenosine causes hypotension. However, the ultimate cardiovascular effects of adenosine in vivo depends on the dose, rate and mode of administration, and on the autonomic reflexes triggered as a result of adenosine's direct action (Pelleg and Porter, 1990).

In the kidney, adenosine produces vasoconstriction of the afferent glomerular artery, a decrease in glomerular filtration rate and inhibition of renin release (Pelleg and Porter, 1990). Adenosine is a depressant of the respiratory center and it causes bronchoconstriction (Pelleg and Porter, 1990). However, adenosine also stimulates arterial chemoreceptors (Biaggioni et al., 1991a). Hence, when given intravenously, adenosine causes hyperventilation (Biaggioni






4



et al., 1991a). In the nervous system, adenosine produces hyperpolarization of neurons resulting in decreased nerve firing. Adenosine also inhibits neurotransmitter release through putative presynaptic inhibitory receptors, both in the brain and in the periphery nervous system. Adenosine inhibits the release of practically all neurotransmitters studied, including norepinephrine, acetylcholine, dopamine, glutamate, aspartate, y-aminobutyric acid and serotonin. Adenosine also has a central depressor action and has been proposed as an endogenous anticonvulsant (see references in Biaggioni et al., 1991a). In fat cells, adenosine abolishes the breakdown of stored triglycerides to free fatty acids and glycerol (lipolysis) which is induced by adrenergic stimulation. Adenosine can also prevent platelet aggregation (Berne, 1986; Pelleg and Porter, 1990) Present and Future Therapeutic Uses


Based on its negative dromotropic effect on AV nodal

conduction, adenosine was recently approved by the U.S. Food and Drug Administration as an antiarrhythmic drug for the acute management of paroxysmal supraventricular tachycardia involving the AV node (Belardinelli and Lerman, 1990). In addition, the transient AV block caused by adenosine can also be used to unmask underlying atrial activity in other forms of atrial arrhythmias and hence help in the differential diagnosis of arrhythmias (Belardinelli and Lerman, 1990).






5



Experimental studies mainly in animal models have

indicated several potential uses of adenosine agonists and antagonists. The agonists could be used as antiepileptic, analgesic and sedative drugs due to their inhibitory effect on neurotransmission (Pelleg and Porter, 1990).

AdoR antagonists could be used for the relief of AV block associated with acute myocardial infarction. In addition, they accelerate recovery of myocardial contractility during cardioversion (Wesley and Belardinelli, 1989).


Synthesis and Metabolism of Adenosine


Adenosine is a local hormone (or autacoid) rather than a circulating hormone or neurotransmitter. It acts within the same organ(s), perhaps even on the cell(s), that is the site of its production. Unlike neurotransmitters, adenosine can be produced by virtually any cell. Adenosine per se does not appear to be stored in exocytotic vesicles, but rather is produced on demand, much like prostaglandins and leukotrienes. The primary mechanism for the production of adenosine in heart muscle, liver and leukocytes is the dephosphorylation of AMP by a 5'-nucleotidase located on the cell membrane or in the cytosol. Adenosine produced is then released into the interstitial space from the parenchymal cells for receptor interaction. Physiological stimuli that cause inadequate tissue oxygenation (e.g., hypoxia, ischemia,






6



exercise) greatly increase adenosine production (Olsson and Pearson, 1990).

Adenosine can also arise from ATP which is released and rapidly broken down by ectonucleotidases. ATP is released from nerve endings (where it is stored in vesicles along with biogenic amines or other classical neurotransmitters), from platelets (where it is stored in secretory granules along with ADP), and from cells that are undergoing lysis. These sources of adenosine probably are important under specific circumstances (i.e., at particular synapses or at sites of injury) (Olsson and Pearson, 1990; Bruns, 1990).

Another intracellular source of adenosine is S-adenosylhomocysteine (SAH), which arises from S-adenosylmethionine. SAH-hydrolase catalyzes the reversible reaction between SAH and adenosine plus homocysteine. Adenosine also tightly binds to SAH-hydrolase. Hence, under basal conditions, the intracellular concentration of free adenosine is probably very low (Delahaba and Cantoni, 1959; Olsson and Pearson, 1990).

Adenosine crosses cell membranes by simple diffusion and and more importantly by facilitated diffusion. Facilitated diffusion is carrier mediated, nonconcentrative and is inhibited by dipyridamole (DIP) (Kolassa et al., 1970), 6-S(p-nitrobenzyl-thio)inosine (Paterson and Oliver, 1971) and dilazep (Bruns, 1990). The carrier appears to transport other nucleosides, which are competitive inhibitors of adenosine transport. The carrier is also symmetrical, mediating both






7



the uptake and release of adenosine with identical kinetics (see references in Olsson and Pearson, 1990).

Adenosine is metabolized very rapidly in the blood with a half-life of 0.6 10 sec. The principal route of metabolism is deamination to inosine by adenosine deaminase and further degradation of inosine to hypoxanthine, xanthine and eventually to uric acid. Adenosine deaminase can be inhibited by erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA) and 2'-deoxycoformycin (see references in Bruns, 1990). In addition, adenosine can reenter the nucleotide pool by phosphorylation to adenine nucleotides (See references in Olsson and Pearson, 1990).


Adenosine Receptors


Classification and Charaterization


Adenosine receptors comprise a group of cell surface

receptors that mediate the physiological and pharmacological effects of the nucleoside adenosine. At least two distinct subtypes of cell surface AdoRs are responsible for these actions. These receptors have been classified as A1-AdoRs and A2-AdoRs based on biochemical and pharmacological criteria, i.e., modulation of adenylyl cyclase (AC) and differential selectivity for a series of adenosine analogs. The Al-AdoR that mediates the inhibition of AC has an agonist potency series of R-phenylisopropyladenosine ((R)-PIA) > 5'-Nethylcarboxamide adenosine (NECA) > (S)-PIA. The A2-AdoR that






8



mediates the stimulation of AC has an agonist potency series of NECA > (R)-PIA > (S)-PIA (Van Calker et al., 1979; Londos et al., 1980). The A2-AdoR in brain has been further subdivided into A2a and A2b subclass. Central A2a-AdoRs are localized primarily in the striatum, nucleus accumbens and olfactory tubercle, whereas central A2b-AdoRs are present in all brain regions. Adenosine and NECA have a higher affinity for the A2a-AdoRs than the A2b-AdoRs (Daly et al., 1983; Bruns et al., 1986).

In addition to modulating AC activity, AdoRs are coupled to other effector systems such as K+ and Ca2+ channels and to phospholipid hydrolysis. Electrophysiological studies indicate that adenosine activates an inwardly rectifying potassium current (IKAdo) in sinoatrial (Bellardinelli et al., 1988), atrial (Belardinelli and Isenberg, 1983) and neuronal (Trussel and Jackson, 1985) cells. The activation of IKAdo is mediated via a pertussis toxin-sensitive guanosine triphosphate (GTP)-binding protein (Kurachi et al., 1986). Adenosine also attenuates the activity of the voltagesensitive Ca2+ channels in hippocampal neurons via presynaptic A1-AdoRs (Schubert, 1985). Adenosine has also been found to indirectly stimulate inositol phosphate accumulation in guinea pig cortex, FRTL-5 thyroid cells and vas deferens. In these tissues, the effect of adenosine is to potentiate the responses of neurotransmitters such as histamine, norepinephrine and angiotensin II. However, neither adenosine nor its analogs alone increase inositol phopholipid






9



hydrolysis in these tissues (Hill and Kendall, 1987; Hollingsworth and Dally, 1985; Linden, 1991). In contrast to potentiating the stimulatory response of other neurotransmitters on phospholipid metabolism, in several other tissues (eg., mouse cortex, brown fat and GH3 pituitary cells), activation of A1-AdoR leads to inhibition of inositol phosphate accumulation (Kendall and Hill, 1988; Delahunty et al., 1988; Linden and Delahunty, 1989; Linden, 1991).

Both Al- and A2-AdoRs are widely distributed in the

central nervous system and peripheral tissues. For example, A1-AdoRs are present in the brain, heart, kidney, lung, pancreas and adipocytes and A2-AdoRs are present in the brain, coronary arteries, kidney and lung (Olsson and Pearson, 1990).

Analysis of structure-activity relationship indicates that certain N-6 substituents of adenosine enhance the potency of adenosine as an Al-AdoR agonist (Olsson and Pearson, 1990). For example, cyclopentyladenosine (CPA) and 2-chloro-N6-cyclopentyl-adenosine (CCPA) have Ki(A2)/Ki(A1) ratios of 2500 and 9750, respectively (Lohse et al., 1988). Several purines with C-2 substituents (e.g., 2aralkoxyadenosine, 2-alkoxyadenosine) have increased potency as A2-AdoR agonists. For instance, 2-[2-(2-naphthyl)ethoxy] adenosine (WRC-0018) is a highly selective A2-AdoR agonist (Ueeda et al., 1991). Examples of AdoR agonists and antagonists and their chemical structures are shown in Figures 1-1, 1-2.






10



Radioligand binding studies of AdoRs have been attempted within the past decade and some success has been achieved, particularly with A1-AdoR ligands. The first successful radioligand binding studies of AdoRs were reported in the early 1980s. Several groups used a variety of tritiated or iodinated radioligands including both agonists and antagonists (Linden et al., 1985; Trost and Schwabe, 1981; Bruns et al., 1980; Williams and Risley, 1980). Radioligand binding studies in membrane preparations from various tissues revealed all the appropriate characteristics; that is 1) saturability, 2) reversibility, 3) stereoselectivity and 4) the pharmacological specificity expected of the physiologically relevant receptor. A recent development in this field has been the synthesis of the high affinity AlAdoR selective radioligand [3H]8-cyclopentyl-1,3-dipropylxanthine ([3H]CPX) (Bruns et al., 1987). This Al-AdoR antagonist has a 740-fold Al-AdoR selectivity over A2-AdoR, the highest selectivity reported for an adenosine antagonist. CPX also has very high affinity for the Al-AdoR (KD = 0.4 nM) and extremely low non-specific binding (=3% of total binding) in rat brain membranes.

Although the availability of agonist and antagonist

radioligands has enabled detailed characterization of the A1AdoR in various tissues (Stiles et al., 1985; Jacobson et al., 1986; Ramkumar and Stiles, 1988; Martens et al. 1987), the lack of a highly selective A2-AdoR antagonist has hampered a similar characterization of the A2-AdoR. Because






11



[3H]NECA has high affinity for A2a-AdoR, it has been used as a radioligand for this receptor. However, NECA also binds to Al-AdoR with high affinity. Thus, when [3H]NECA is used, it is necessary to block the A1-AdoR by adding a highly selective AI-AdoR ligand such as the agonist CPA or the antagonist CPX (Hutchison et al., 1989; Linden, 1991). Another A2-AdoR agonist radioligand recently synthesized is 2-[4-(2-{[4aminophenyl]methylcarbonyl}ethyl)phenyl]ethylamino-5'-Nethylcarboxamido adenosine ([125I]PAPA-APEC) (Ramkumar et al., 1990). Likewise, 2-[p-(2-carboxyethyl)phenethylamino]-5'-Nethyl-carboxamido adenosine (CGS 21680), an agonist with high affinity and selectivity for A2- over Al-AdoR has been synthesized (Hutchison et al., 1989; Lupica et al., 1990). Specific binding of the newly synthesized [3HICGS 21680 to rat striatal membranes was saturable and reversible. Saturation studies revealed that [3H]CGS 21680 binds with high affinity (KD=16 nM) to a single class of binding sites. Adenosine agonists competed for the binding of [3H]CGS 21680 with the following potency order: CGS 21680 > NECA > (R)-PIA > (S)-PIA. The specific binding of [3H]CGS 21680 was greatest in rat striatal membranes but negligible in rat cortical membranes. These results indicate that [3H]CGS 21680 directly labels the high affinity A2a-AdoR in rat brain without the need to block binding to Al-AdoRs (Jarvis et al., 1989; Jarvis and Williams, 1989).

A1-AdoRs of brain, heart, or fat cells, when labeled with photoaffinity ligands or by means of photoaffinity






12



cross-linking, migrate on sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) with an molecular mass between 35 and 38 kDa (Ramkumar et al., 1990). In comparison with Al-AdoRs, photolabeled A2-AdoRs migrate on SDS-PAGE with a Mr of 42,000 (Ramkumar et al., 1990). Both receptors are known to be glycosylated (Ramkumar et al., 1990). Treatment of photoaffinity-radiolabeled Al-AdoR of brain with an endoglycosidase (Stiles, 1985) or with either trifluoromethanesulfonic acid or a-mannosidase (Klotz and Lohse, 1986) reduces the molecular mass of the ligand-binding peptide to 32 kDa. A role for protein glycosylation in the function and cellular processing of the AdoR is at present unknown. However, glycosylation has been reported to be essential in the synthesis of insulin receptors (Ronnett and Lane, 1981) but not B-adrenoreceptors (SAR) (Doss et al., 1985). In addition, glycosylation also appears to be required for the maintenance of cell surface muscarinic receptors (Liles and Nathanson, 1986).

A1-AdoRs from rat brain membranes were purified in 1989 (Nakata, 1989a,b; Munshi and Linden, 1989). Al-AdoR was solubilized with digitonin and purified approximately 50,000fold to apparent homogeneity by two cycles of affinity chromatography using an antagonist affinity column. The purified receptor migrated on SDS-PAGE with Mr=34,000 either in the absence or presence of 2-mercaptoethanol, suggesting that the receptor does not contain disulfide-linked subunits (Nakata, 1989a).






13



Recently, two previously cloned proteins (RDC7 and RDC8) with deduced seven transmembrane helices have been identified as canine A1- and A2a-AdoR subtypes, respectively (Libert et al., 1989; Maenhaut et al., 1990). The deduced molecular masses of RDC7 (36,356 Da) and RDC8 (45,008 Da) correspond closely to the apparent molecular masses of Al- and A2a-AdoRs estimated by photoaffinity labeling. Features shared by both proteins include small N-termini, conserved transmembrane domains and at least one cysteine in exofacial loops 1 and 2. Notable by their absence are consensus sequences for N-linked glycosylation in the N-terminal segments and an aspartate residue in the third transmembrane segment, the hallmark of cationic amine receptors. Clusters of serine and threonine residues in the C-terminal segments, commonly seen in guanine nucleotide-binding protein (G protein)-linked receptors, are absent in RDC7. Finally, at 326 amino acids, RDC7 is among the smallest members of the G-protein-coupled superfamily of receptors. RDC7 and RDC8 are similar to a variety of other superfamily members, but none of these is more than 30% identical to either of the AdoRs. The expression of RDC8 in adrenal cells, thyrocytes and Xenopus oocytes resulted in activation of AC in the absence of added AdoR agonist. Membranes from Cos 7 cells transfected with RDC8 cDNA exhibited binding characteristics of an A2-AdoR. Moreover, RDC8 mRNA and A2-AdoR displayed a very similar distribution in the brain (Maenhaut et al., 1990). These data all support that RDC8 is an A2-AdoR. The gene(s) for the low-affinity A2b-






14



AdoR has yet to be cloned. There is no published report about the binding and functional characteristics of RDC7 in an expression system.


Receptor-Guanine Nucleotide Regulatory Protein Coupling


Adenosine receptors modulate AC activity via G proteins, the stimulatory Gs and the inhibitory Gi for A2-AdoR and AlAdoR, respectively.

The activation of AC by receptors coupled to Gs can be described by the ternary complex model. This model has been developed for the BAR system and is likely to be applicable for other stimulatory receptors including the A2-AdoRs. In brief, an agonist (Ag) binds to the receptor (R) to form an Ag-R complex. The affinity of the agonist in the Ag-R complex is relatively low. The complex then undergoes a conformational change and interacts with a Gs protein to form the ternary complex of Ag-R-Gs. The affinity of the agonist in the Ag-R-Gs complex is relatively high. When GTP binds to Gs, Ag-R-Gs is rapidly converted to Gs-GTP and Ag-R. Once formed, Gs-GTP interacts with the catalytic subunit (C) of AC to form the active complex C-Gs-GTP resulting in a conversion of ATP to cAMP. The enzyme activity returns to basal level when guanosine 5'-triphosphatase (GTPase) activity in Gs hydrolyzes the bound GTP to guanosine diphosphate (GDP) with the resultant regeneration of inactive catalytic unit and GsGDP. The destabilization of the ternary complex decreases






15



agonist affinity and increases the dissociation rate (Gilman, 1987).

Several lines of evidence suggest that the AdoR belongs to the class of G-protein-coupled receptors. For A1-AdoR, GTP and stable GTP analogues decreased the apparent affinity of A1-AdoR agonists to the receptor in membranes (Goodman et al., 1982; Yeung and Green, 1983; Lohse et al. 1984), solubilized receptors (Gavish et al., 1982; Stiles, 1985; Stroher et al., 1989) and slices of brain tissue (Fastbom and Fredholm, 1990). Treatment with N-ethylmaleimide which uncouples receptors from G proteins inhibited the action of A1-AdoR agonists without affecting the binding of antagonists (Fredholm et al., 1985). In addition, detergent-solubilized A1-AdoR co-eluted with a G protein from an agonist affinity chromatography column where GTP or N-ethylmaleimide, agents known to uncouple receptors from G proteins and simultaneously lower the affinity of agonists for the receptor, were used (Munshi and Linden, 1989; Linden, 1991). It was suggested that Al-AdoR, unlike other G protein coupled receptors, tightly binds to G protein. The coelution did not occur when an antagonist affinity column was used (Nakata, 1989a, b).

For the A2-AdoR, experiments showed a GTP dependence for NECA stimulation of AC in purified hepatic plasma membranes (Cooper and Londos, 1979). AC activity from other tissues has been shown to be enhanced by or dependent on the presence of a guanine nucleotide (Londos et al., 1979; Fain and Malbon,






16



1979; Londos, et al., 1981; Wolff et al., 1981). Moreover, adenosine shortened the lag period for the onset of AC activation by Gpp(NH)p (Sevilla et al., 1977; Lad et al., 1980).

The interaction of an activated receptor with G protein is a key step in signal transduction. The regulatory G proteins are heterotrimers with subunits designated cl,8 and 7 in order of decreasing mass. The functional difference between Gs and Gi resides in the respective as and ti subunits. The structural features common to as and ai include a GTPase activity and regions that recognize and couple these proteins to a receptor, to the B+y complex, and to the effector systems. Common 3 and y subunits are functionally interchangable (Gilman, 1987). G proteins cycle between an inactive GDP state and an active GTP state. When GDP is bound, the a subunit associates with the 3 and y subunits to form a Gay complex (denoted by G-GDP) that is membrane-bound. When GTP is bound, the CL subunit (Ga-GTP) dissociates from

the 8 and y subunits (Gay). Ga-GTP released from Gay then alters the activity of the target, such as AC or an ion channel (Stryer and Bourne, 1986).

Another feature of G proteins is that the cysteine residue in their a subunits is a substrate for adenosine diphosphate (ADP)-ribosylation which transfers an ADP-ribose moiety from nicotinamide-adenine dinucleotide (NAD). Gs is selectively ADP-ribosylated by cholera toxin which in turn inhibits the receptor-stimulated GTPase activity of the G






17



protein causing GTP to be bound to as for a prolonged period. This results in a permanent activation of AC and a large increase in cAMP accumulation (Gilman, 1987; Nathanson, 1987). Pertussis toxin (PTX) selectively ADP-ribosylates Gi (and another guanine nucleotide binding protein, Go). This covalent modification of Gi inactivates the protein resulting in the loss of receptor mediated inhibition of AC activity (Gilman, 1987; Nathanson, 1987). Thus, these bacterial toxins have been used as an investigative tool to determine the role of G proteins in the action of biologic messengers. For example, studies showed that PTX blocked the ability of AlAdoR to inhibit AC (Hazeki and Ui, 1981) and prevented adenosine-induced changes in the rate of beating in rat atria (Endoh et al., 1983).


Mechanisms for Inhibitory Receptor Action


Receptors that mediate the attenuation of AC activity

include the Al-AdoR, muscarinic M2-acetylcholine receptor, a2adrenergic receptor, 8 opiate receptor and D2-dopamine receptor.

Compared to stimulatory receptors, much less is known about the mechanisms whereby inhibitory receptors attenuate the activation of AC by stimulatory receptors. Currently, there is experimental evidence supporting three different models. First, inhibition may occur by preventing the formation of the ternary complex which is composed of the






18



agonist, the stimulatory receptor and Gs. This results in the inability of the receptor to form a high affinity binding state. Experiments on rat ventricular myocyte membranes showed that PIA inhibited isoproterenol (ISO)-stimulated AC activity (Romano et al., 1988; Romano et al., 1989). This inhibition was antagonized by theophylline. PIA was much less effective at attenuating forskolin-stimulated AC activity and had no effect on 5'-guanyl-imidodiphosphate (Gpp(NH)p)induced stimulation. In [125I]cyanopindolol (CYP)/ISO competition binding experiments, ISO produced a concentration-dependent displacement of specific [125I]CYP binding with an IC50 of 48 nM and Hill slope of 0.6. About 38% of BARs were in the high affinity state. Gpp(NH)p shifted the competition curve to the right (IC50 = 520 nM) and steepened the slope (Hill slope = 1.2) indicating that all of the BARs were in low affinity state. PIA significantly increased the IC50 for ISO in the absence of Gpp(NH)p (IC50 = 140 nM) and steepened the slope (Hill slope = 0.9). These findings were interpreted to indicate that binding of ISO to the high affinity state of the BAR was decreased in the presence of PIA. PIA had no effect on the ISO competition curve in the presence of Gpp(NH)p (Romano et al., 1988; Romano et al., 1989).

The second is the subunit dissociation model. This model is based on the finding that both Gs and Gi share two common subunits, B and y, and that ai is present in excess relative to as in most cells. Activation of Gi leads to subunit






19



dissociation and the release of sufficient quantities of the 1+y complex. The large increase in the amount of free 8+y complex in the membrane would disturb the equilibrium which exists between undissociated and dissociated Gs under resting conditions. Thus, by mass action, 3+y complex would combine with the released stimulatory a subunits preventing its dissociation and subsequent activation of AC. This model, therefore, implies that Gi will only be effective under conditions where the activity of AC is stimulated, i.e., the effectiveness of Gi is related to the level of the dissociated Gsa (Morgan, 1989). Experiments supporting this model showed that when platelet membranes were treated for brief periods of time with GTPyS and an a2-adrenergic agonist in low Mg2+ conditions, AC was "irreversibly" inhibited. This inhibition was of the same magnitude as that produced by maximally effective concentrations of 1+y complex, and it was not additive with the effect of +7y. This inhibition is completely overcome by reconstitution of the membranes with physiological concentrations of Gia-GDP. The most likely explanation for this observation is the interaction between Gia-GDP and Ga+y to relieve the inhibition caused by free +7y complex in the membrane (Katada et al., 1984a).

The third is the direct interaction model. It involves inhibition of AC by the released ai subunits acting directly on the catalytic subunit of the enzyme (Gilman, 1987). Evidence favoring this model over the subunit dissociation model includes the observation that inhibitory agonists are






20



capable of reducing AC activity in the cyc- S49 cell mutant. These cells lack Gsa, and logically, 8+y is not inhibitory when reconstituted with cyc- membranes. It was also demonstrated that the isolated Gia from rat liver can inhibit AC activity in membranes from cyc- S49 cell. The inhibitory effect of Gia was therefore proposed to explain the ability of inhibitory agonists to decrease AC activity in the cycS49 cell mutant (Katada et al., 1984b; Jakobs and Schultz, 1983). This inhibitory effect of Gia-GTPYS has also been observed by Roof et al. (1986) in the bovine central nervous system.


Regulation of Adenosine Receptors


Similar to many other receptors, the AdoR appears to undergo desensitization and down-regulation during chronic exposure to an agonist. This effect prevents overstimulation of the receptor. The mechanism(s) of desensitization and down-regulation has/have been studied extensively in the BARAC system and may be applicable to the AdoR. Desensitization describes the phenomenon where an initial exposure of a cell to an agonist results in a reduced capacity of the cell to respond to a second challenge. Two main types of desensitization have been described. "Homologous" desensitization is referred to hormone-specific type where loss of response is only to the activated receptor, whereas other receptor mediated responses remain unaffected.






21



"Heterologous" desensitization is referred to hormonenonspecific type where activation of one receptor causes loss of response mediated by other receptors. Desensitization of BAR system appears to be initiated by receptor phosphorylation which results in functional uncoupling of the BAR from Gs. Two kinases have been implicated: the cAMPdependent protein kinase (PKA) which plays a major role in heterologous desensitization; and the cAMP-independent kinase, termed BAR kinase which specifically phosphorylates the agonist-occupied receptor leading to homologous desensitization. After uncoupling, the receptors appear to be sequestered within the cells. Removal of the agonist after sequestration leads to rapid resensitization of the system. During longer-term agonist treatment, there appears to be a loss of receptors (down-regulation) due to receptor degradation or loss of the recognition site for ligand binding (Harden, 1983).

Studies have shown that in vitro exposure of cultured rat adipocytes to (R)-PIA causes concentration- and timedependent loss of Al-AdoR and decrease in the content of Gi protein (Green, 1987). These changes were accompanied by attenuation of the antilipolytic effect of (R)-PIA (homologous desensitization) (Green, 1987). In another study, the number of cardiac Al-AdoRs in chick embryos was decreased by 63% after pretreatment with 1 AM (R)-PIA for 44 hrs (Shryock et al., 1989). Experiments also showed that the desensitization of Al-AdoR system in DDT1 MF-2 cells was






22



accompanied by a decrease in the number of A1-AdoRs which can form a high affinity agonist binding site and a 3-4 fold increase in the phosphorylation of Al-AdoR (Ramkumar et al., 1991). Thus, similar to the BAR system, Al-AdoR can also undergo desensitization and/or down-regulation after chronic exposure to an agonist. Uncoupling, down-regulation and phosphorylation of the Al-AdoR may contribute to the desensitization of this inhibitory receptor.

Desensitization of A2-AdoR-AC system has also been described. Using clonal neuronal cells (NG108-15), which express both A2-AdoR and prostaglandin E (PGE1) receptors, PGE pretreatment reduced the effects of both PGE1 and adenosine to activate AC (heterologous desensitization). In contrast, exposure of NG108-15 to 2-chloroadenosine resulted in a rapid loss of response to 2-chloroadenosine (homologous desensitization), but PGE1-stimulated AC activity decreased only slightly (Kenimer and Nirenberg, 1981).

Adenosine receptors are regulated during chronic drug treatment with AdoR antagonist or dexamethasone. A recent study showed that exposure of guinea pig myocardium to AdoR antagonist theophylline increased the the number of Al-AdoR (Wu et al., 1989). In humans, after 7 days of caffeine (AdoR antagonist) abstinence, NECA produced a concentrationdependent inhibition of thrombin-induced platelet aggregation with an EC50 value of 69 nM (Biaggioni et al., 1991b). Subjects were then given caffeine 250 mg p.o. 3 times a day for 7 days. Caffeine withdrawal significantly shifted the






23



concentration response of NECA to the left (EC50=49 nM, p<0.01 by ANOVA) indicating sensitization of AdoRs (Biaggioni et al., 1991b). Other examples include pretreatment of DDT1 MF-2 cells with dexamethasone (Gerwins and Fredholm, 1991). This glucocorticoid caused a concentration- and time-dependent increase in the number of Al-AdoRs, but did not affect the KD or the proportion of Al-AdoRs in high and low affinity states (Gerwins and Fredholm, 1991). (R)-PIA was more potent as an inhibitor of cAMP formation induced by ISO in dexamethasonetreated cells. Addition of glucocorticoid receptor antagonist RU 486 or protein synthesis inhibitor cycloheximide prevented the up-regulation of Al-AdoR (Gerwins and Fredholm, 1991). In contrast to sensitization of Al-AdoR-AC system, the A2-AdoR-AC system was desensitized as indicated by the decreased ability of NECA to increase cAMP formation in dexamethasone-treated cells (Gerwins and Fredholm, 1991).

AdoRs are also regulated under normal and

pathophysiological conditions. For example, in the hypothyroid state, (R)-PIA mediated inhibition of AC and its antilipolytic effect is enhanced (Ohisalo and Stouffer, 1979). In contrast to hypothyroidism, the hyperthyroid state is characterized by enhanced lipolytic activity and cAMP accumulation in adipocytes. These are likely related to the loss of inhibitory tone mediated by Al-AdoR due to a 35% decrease in Al-AdoR number (Malbon et al., 1978; Rapiejko and Malbon, 1987). Other conditions that may alter AdoRs include






24



pregnancy, lactation, starvation, obesity and aging (Ramkumar et al., 1988).


Goals


Over the past decade, a great deal has been learned

about the pharmacology, biochemistry and physiology of AdoRs. However, much remains unknown. In general, receptors that mediate inhibition of cAMP formation appear to dominate over stimulatory receptors. In the case of AdoRs, if both subtypes are present in a single cell and are simultaneously activated by adenosine, it becomes important to determine under what conditions the Al- and A2-AdoR mediated responses will be expressed. Thus, the major goals of this study were 1) to determine pharmacologically if an interaction between Al- and A2-AdoR occurs and 2) to define the conditions whereby the expression of the A2-AdoR mediated response can be demonstrated.

In addition, the hypothesis that the mechanism for AlAdoR inhibitory effects involves alteration in the ability of BAR agonists to interact with the BAR was investigated. By using an irreversible BAR agonist that permanently activates the BAR, it was determined if the resulting response can be modulated by the inhibitory Al-AdoR.




25







AGONIST ANTAGONIST
CPA N-0861 Al CCPA CPX 0 G NECA 8PST ADO


GA WRC-001 8 8PST A2 NECA ? cAMP ADO




A, -AdoR --- decrease cAMP

A2 -AdoR --- increase cAMP


Figure 1-1. Agonists and antagonists of AdoR subtypes.






26







NNH
NH NH N N


HHO O HO 0 HO- 0


HO OH
HO OH HO OH


CPA CCPA WRC-0018


NH2 NH2
N N N N
NN N
HO O O C2HsN H

HO OH HO OH


ADENOSINE NECA


o 0 HC3I H H3C H

NN N I CH CH3 CH3

N-0861 CPX 8PST


Figure 1-2. Chemical structures of AdoR agonists and antagonists. Abbreviations used: N6-cyclopentyladenosine (CPA), 2-chloro-N6-cyclopentyladenosine (CCPA), 2-[2-(2Naphthyl)ethoxy]adenosine (WRC-0018), 5'-N-ethylcarboxamidoadenosine (NECA), (+) N6-endonorbornan-2-yl-9-methyladenine (N-0861), 8-cyclopentyl-1,3-dipropylxanthine (CPX), 8(psulfophenyl)theophylline (8PST).















CHAPTER 2
EXPERIMENNTAL PROCEDURES


Source of Materials


The radioligands [2,8-3H]adenosine 3', 5'-cyclic

monophosphate ([3H]cAMP; 31.2 Ci/mmol), [3H]8-cyclopentyl-1,3dipropylxanthine ([3H]CPX; 99-107 Ci/mmol) and (-) [125I]iodocyanopindolol ([125I]CYP; 2,000-2,200 Ci/mmol) were purchased from New England Nuclear Corp. (Boston, MA, USA). Adenosine, N6-cyclopentyladenosine (CPA), (-)-N6-(2-phenyl-isopropyl) adenosine ((R)-PIA), dipyridamole (DIP), erythro-9-(2hydroxy-3-nonyl)adenine (EHNA), adenosine deaminase (ADA) type VI, 3-[(3-cholamidopropyl)dimethylammoniol]-1-propanesulfonate (CHAPS), benzamidine, (-)-isoproterenol (ISO), 5'guanylyl-imididodiphosphate (Gpp(NH)p), propranolol (PROP),

()-alprenolol, penicillin G, streptomycin sulfate, amphotericin B, theophylline, protein kinase, hydroxyapatite and bovine serum albumin were from Sigma Chemical Co. (St. Louis, MO, USA). 8-cyclopentyl-1,3-dipropylxanthine (CPX), 8(p-sulfophenyl)theophylline (8PST), 5'-N-ethylcarboxamidoadenosine (NECA) and 2-chloro-N6-cyclopentyladenosine (CCPA) were purchased from Research Biochemicals Inc. (Natick, MA, USA). The DDT1 MF-2 (DDT) cell line was obtained from American Type Culture Collection (Rockville, MD, USA).




27






28



Dulbecco's modified Eagle's media (DMEM) and fetal bovine serum were from Gibco (Grand Island, NY, USA). Liquiscint was purchased from National Diagnostics (Somerville, NY, USA). 2[2-(2-Naphthyl)ethoxy]adenosine (WRC-0018) was a kind gift of Dr. Ray. A. Olsson (Univ. of South Florida, Tampa, FL, USA).

()N6-endonorbornan-2-yl-9-methyladenine (N-0861) was a gift of Whitby Research, Inc. (Richmond, VA, USA). Pertussis toxin was a gift of Dr. Eric Hewlett (Univ. of Virginia, Charlottesville, VA, USA). Rolipram was a gift of Berlex Laboratories (Cedar Knolls, NJ, USA). 5-[2-[[3-[4(bromoacetamido)phenyl]-2-methylprop-2-yl]amino]-1hydroxyethyl]-8-hydroxycarbostyril (C-Br) was synthesized as described previously (Milecki et al., 1987). All other reagents were from Sigma Chemical Co. (St. Louis, MO, USA) or Fisher Scientific (Orlando, FL, USA).


Methods


Cell Culture


The DDT cell line was derived from a steroid-induced

leiomyosarcoma tumor of the vas deferens of an adult Syrian hamster (Norris and Kokler, 1974). These cells were obtained at low passage number and grown as a monolayer on 150 mm plastic culture dishes (Falcon) in DMEM supplemented with 5% fetal bovine serum, 100 U/ml penicillin G, 0.1 mg/ml streptomycin and 2.5 gg/ml amphotericin B in an atmosphere of 5% C02/95% air at 370C. Cells were seeded at 0.2-1 x 104






29



cells/cm2 and subcultured twice weekly after detachment using

1 mM ethylenediamine tetraacetic acid (EDTA) in phosphate buffered saline (PBS). DDT cells have a doubling time of about 20 hours and a confluent density of 1.3 x 105 cell/cm2. Experiments were performed on cells 1-day pre-confluent. Drug Preparation


Stock solutions of WRC-0018 (10 mM) and rolipram (50 mM) were prepared in dimethylsulfoxide (DMSO). CPA (1 mM), CCPA (1 mM), DIP (10 mM) and C-Br (1 mM) were prepared in ethanol. NECA (1 mM) was dissolved in 5 mM HC1 and N-0861 (1 mM) was dissolved in a mixture of ethanol (10%) and 50 mM Tris buffer containing 10 mM MgCl2 (90%). These stock solutions were diluted in Hank's Balanced Salt Solution (HBSS) without divalent cations to the desired concentrations just prior to use. HBSS contains 137 mM NaCl, 6 mM D-glucose, 5 mM KC1, 4 mM NaHCO3, 0.6 mM Na2HPO4, 0.4 mM KH2PO4, 0.5 mM MgCl2, 0.4 mM MgS04 and 1 mM CaCl2 at pH 7.4. All other drugs were dissolved in HBSS before use.


Drug Treatment


The growth medium in culture dishes was aspirated and

fresh medium (20 ml) was then added followed by the drug. The cells were then incubated at 370C for the various period of times as indicated in the text. At the end of the incubation period, the media was aspirated and the attached cells were






30



washed four times with 10 ml of ice-cold HBSS without divalent cations.


Membrane Preparation


DDT cells were harvested from culture dishes in 5 ml of 50 mM Tris-HC1 buffer at pH 7.4 containing 5 mM MgCl2 with a rubber policeman and were pelleted by centrifugation at 48,000g for 15 min.

For the determination of Al-AdoR, the pellet was

resuspended in ice-cold 50 mM Tris-HCI (pH 7.4) containing 1 mM MgC12, 1 mM EDTA and 0.1 mM benzamidine (trypsin inhibitor) and then homogenized with a Ten Broeck Tissue Grinder (glass-glass). The homogenate was centrifuged at 48,000g for 15 min to pellet the membranes. The membranes were homogenized a second time in 50 mM Tris-HC1 (pH 7.4) containing 1 mM MgCl2, then used for protein measurement and receptor binding assays.

For determination of the BAR, the pellet from the first centrifugation as described above was homogenized in ice-cold 50 mM Tris-HCI (pH 7.4) containing 5 mM MgC12 with a Tekmar SDT-100EN homogenizer (setting 5, 20 s). After centrifugation at 48,000g for 15 min and homogenization with the Tekmar as above, the membrane suspension was used for assays.






31


Protein Measurement


The protein content of cells and membranes was

determined by the method of Lowry et al.(1951) using bovine serum albumin as standard.


Radioligand Binding Assay


A1-AdoRs in DDT cells were determined by specific [3H]CPX binding. Membrane protein was initially incubated with 2 U/ml (1 U = 6.25 gg) ADA for 20 min at 40C to metabolize endogenous adenosine. Cell membranes (=0.1 mg protein) were then incubated in a total volume of 0.2 ml with 50 mM TrisHCI buffer at pH 7.4, 5 mM MgCl2 and 0.06-4 nM [3H]CPX, with or without 50 .M (R)-PIA, for 150 min at room temperature on an orbital shaker. The bound and free ligand were rapidly separated on GF/C glass fiber filters (Whatman Inc., Clifton, NJ, USA) using a Brandel Cell Harvester (Brandel Scientific, Gaithersburg, MD, USA). Filters were rinsed three times with 4 ml of ice-cold 50 mM Tris-HC1 buffer containing 10 mM MgCl2 and 0.1 % CHAPS (to reduce non-specific binding). The filters were placed in standard scintillation vials with 10 ml of Liquiscint and the radioactivity was determined in a liquid scintillation counter. Specific binding to A1-AdoR was calculated as the difference between the total binding in the absence of (R)-PIA and the nonspecific binding in the presence of 50 uM (R)-PIA. Specific binding was generally 90-






32



95% of total binding. All assays were performed in triplicate, and the determinations differed by less than 6%.

BARs were quantitated by specific [125I]CYP binding.

Membrane protein (30-50 gg) was incubated in a total volume of 0.25 ml with 50 mM Tris-HCI buffer at pH 7.4, 5 mM MgCl2 and 6-100 pM [125I]CYP, with or without 3 JIM ()-alprenolol, for 60 min at 360C. The bound and free ligand were then rapidly separated on GF/B glass fiber filters using a Brandel Cell Harvester. Filters were rinsed three times with 4 ml of ice-cold 50 mM Tris-HC1 buffer containing 5 mM MgC12, placed in omnivials and the radioactivity was determined in a gamma counter. Specific binding was typically 80-90% of total binding. All assays were performed in triplicate, and the determinations differed by less than 6%.


cAMP Assay


Cells were detached from culture dishes with a cell

lifter and centrifuged at 500g for 5 min. The cells (0.4 mg protein/tube = 2x106 cells) were gently resuspended in HBSS containing 50 gM rolipram (phosphodiesterase inhibitor), and incubated at 360C for 7 min in Beckman microcentrifuge tubes. Drugs were then added and the cells were incubated at 360C for the various period of times as indicated in the text. At the end of the incubation period, the tubes were immediately placed in a boiling water bath for 5 min. The protein was






33



pelleted by centrifugation at 9,000g for 2 min, and the supernatants were saved for cAMP assays.

The cAMP content of the supernatant was determined by a modification of a competitive protein binding assay described previously (Baker et al., 1985). An aliquot (usually 50 p1) of the supernatant was incubated in a total volume of 0.2 ml with 25 mM Tris-HCI buffer at pH 7.0, 8 mM theophylline, 0.8 pmol of [3H]cAMP and 24 gg of bovine heart cAMP dependent protein kinase at 40C for 60 min. At the end of the incubation, 70 p. of a 50% (v/v) hydroxyapatite suspension was added to each tube. The suspensions were then poured onto a Whatman GF/C glass fiber filter under reduced pressure. The filters were rinsed three times with 4 ml of ice-cold 10 mM Tris-HCI buffer and placed in minivials with 3 ml of Liquiscint. Radioactivity was determined in a liquid scintillation counter. The amount of cAMP present was calculated from a standard curve determined using known concentrations of unlabeled cAMP.


Data Analysis


Receptor density (Bmax) and dissociation constant (KD)

for the radiolabeled ligands were determined from regression analysis of Scatchard plots (1949). The concentrations of compounds which inhibited ligand binding by 50% (IC50) were obtained from Hill plots of the competition data (Hill, 1913). The effective concentrations of drugs which gave 50%






34



of a maximal response (EC50) were determined using a concentration-effect analysis with a non-linear regression algorithm (Marquardt-Levenberg). Statistical analysis of significance of difference was performed using the Student's t-test.















CHAPTER 3
INTERACTION OF ADENOSINE RECEPTORS Introduction


Cell surface adenosine receptors (AdoRs) have been

classified by pharmacological and biochemical criteria. Two subtypes of AdoRs i.e., Al- and A2-AdoR have been thus far clearly identified. In most cell types studied to date, the A1-AdoR mediates an inhibition of AC activity whereas the A2AdoR mediates a stimulation of the enzyme (Van Calker et al., 1979; Londos et al., 1980). In general, receptors that mediate inhibition of cAMP formation dominate over receptors that mediate stimulation. For example, in mouse atria, carbachol antagonized ISO-stimulated cAMP accumulation by direct activation of the muscarinic receptors. The interaction between carbachol and ISO was not competitive, since cholinergic inhibition could not be surmounted by increasing concentrations of ISO (Brown, 1979). In atria isolated from rats, carbachol decreased the ISO-induced elevation of cAMP levels and inhibited the positive chronotropic and inotropic responses to ISO (Endoh et al., 1985). In addition, after desensitization of the muscarinic system in AtT-20 cells by oxotremorine, cAMP accumulation stimulated by ISO was approximately doubled (Heisler et al.,




35





36



1985). In most if not all of these examples, the inhibitory effect cannot be overcome even when the receptor is maximally activated.

The DDT smooth muscle tumor cell line is derived from a steroid-induced leiomyosarcoma of the vas deferens of an adult Syrian hamster (Norris and Kokler, 1974). This cell line has been shown to express a relatively high density of 82-ARs which mediate a robust stimulation of cAMP formation (Norris et al., 1983). The DDT cells has been used therefore as a model to study this receptor system. These cells have also been shown to express functional histamine H1 (Mitsuhashi and Payan, 1988) and steroid receptors (Norris and Kohler, 1977). Recently, the presence of both Al- and A2AdoR have been demonstrated in DDT cells by radioligand binding, photoaffinity labeling and their ability to alter AC activity (Ramkumar et al., 1990). In keeping with the classical definition of these two subtypes, the Al-AdoR in DDT cells inhibits AC activity whereas the A2-AdoR stimulates the enzyme. The presence of both AdoR subtypes on a single cell coupled to the same second messenger system provides a unique opportunity to characterize pharmacologically if an interaction between the Al- and A2-AdoR occurs. Assuming that an Al-AdoR inhibitory response would mask any A2-AdoR mediated stimulation of cAMP accumulation, three experimental approaches were used to attempt to alter Al-AdoR responsiveness and allow the expression of the A2-AdoR response. These three approaches are shown diagrammatically






37



in Figure 3-1. First, selective blockade of the A1-AdoR; second, uncoupling of the A1-AdoR with PTX and third, downregulation and/or desensitization of the A1-AdoR using a selective agonist.


Results


Effects of Selective and Nonselective Adenosine Receptor Aaonists on cAMP Accumulation in DDT Cells


Experiments were designed to investigate the effects of the selective A1-AdoR agonist CPA, the putative nonselective agonists NECA and Adenosine (ADO), and the selective A2-AdoR agonist WRC-0018 on cAMP accumulation in DDT cells.

Figure 3-2 illustrates the concentration-dependent

inhibition of ISO (10 gM)-induced cAMP accumulation by CPA, NECA and ADO in DDT cells. The effect of ADO was studied in the presence of DIP (inhibitor of adenosine transporter) and ENHA (inhibitor of adenosine deaminase) to prevent the uptake and degradation of this nucleoside respectively during the 7 min incubation period. The rank order of potency of these AdoR agonists to inhibit ISO-induced cAMP accumulation was CPA > NECA > ADO with EC50 values of 1.5, 14 and 97 nM, respectively. The maximal inhibition of cAMP accumulation caused by the AdoR agonists was 89% for CPA, 97% for NECA and 79% for ADO. Figure 3-3 illustrates the concentration response of ISO to stimulate cAMP accumulation in the absence and presence of CPA (0.1 IM) or NECA (1 gM). These






38



concentrations of A1-AdoR agonists caused maximal inhibition of ISO-stimulated cAMP accumulation (Figure 3-2). ISO alone increased cAMP accumulation in a concentration-dependent manner with an EC50 of 2.4 nM and a maximal stimulation of 62fold above the basal level achieved at 0.1 M. In the presence of CPA or NECA, ISO still stimulated cAMP accumulation in a concentration-dependent manner, but the maximal response was decreased by 75% and 60% in the presence of CPA and NECA, respectively. In the presence of 1 gM CPX, a selective A1-AdoR antagonist, the inhibitory effect of CPA on ISO-stimulated cAMP accumulation was significantly attenuated (data not shown). Thus, our results indicated that the A1AdoR agonists had an inhibitory effect on cAMP accumulation and this effect was mediated by A1-AdoR.

The effects of the selective A2-AdoR agonist WRC-0018 (Ueeda et al., 1991) and the nonselective agonist NECA on cAMP accumulation are illustrated in Figure 3-4. WRC-0018 produced a biphasic response whereas NECA caused no effect on cellular cAMP level. WRC-0018 at low concentrations (5 500 nM) stimulated cAMP accumulation with a maximal response of 81 -fold above the basal level. The estimated EC50 was 8.6 nM. However, higher concentrations (>500 nM) of WRC-0018 decreased cAMP accumulation with an EC50 value of 5.1 pM. In contrast to WRC-0018, NECA over the entire concentration range of 0.1 nM 10 JM did not affect (increase or decrease) cellular cAMP content. Similar to NECA, ADO (0.1 nM 10 M)





39



also did not affect the cellular cAMP level above the basal level (data not shown).

Figure 3-5 depicts the effect of the non-selective AdoR antagonist 8PST on the stimulatory effect of WRC-0018. WRC0018 (0.1 JiM) induced a 3-fold increase in cAMP content above the basal level and 8PST (5 JiM) inhibited the WRC-0018induced cAMP accumulation by 80%. Figure 3-6 illustrates a similar effect of the Aj-AdoR agonist CPA on the stimulatory effect of WRC-0018. At a concentration of 0.1 gM, WRC-0018 stimulated cAMP accumulation 3-fold above the basal level and CPA (1 gM) inhibited the WRC-0018-induced cAMP accumulation by 78%.

Figure 3-7 illustrates the effects of WRC-0018 on cAMP accumulation in the absence and presence of 10 M ISO. The biphasic concentration response curve of WRC-0018 was replotted from Figure 3-4. In the presence of ISO, low concentrations of WRC-0018 (1 nM 100 nM) did not affect ISO-stimulated cellular cAMP accumulation. However, at higher concentrations (2500 nM), WRC-0018 attenuated the stimulatory effect of ISO in a concentration-dependent manner with an EC50 of 840 nM and a maximal inhibition of 73%.

These data show that nonselective AdoR agonists (NECA and ADO) themselves only inhibit cAMP accumulation. In contrast, the selective A2-AdoR agonist (WRC-0018) stimulated at lower and inhibited cAMP accumulation at higher concentrations, thereby resulting in a biphasic concentration response curve. To further investigate the expression of the





40



A2-AdoR mediated cAMP accumulation, three experimental approaches were used (Figure 3-1). These were 1) selective blockade of A1-AdoR, 2) uncoupling of Al inhibitory effects with PTX, and 3) selective desensitization and/or downregulation of Al-AdoR.


Effect of a Selective Aj-AdoR Antagonist on the A2-AdoR
Mediated Response


The first experimental approach used to uncover the A2AdoR mediated effect on cAMP accumulation was blockade of the A1-AdoR with a selective A1-AdoR antagonist. The ability of the highly selective A1-AdoR antagonist, N-0861 (Shryock et al., 1992), to compete with [3H]CPX for the Al-AdoR binding site is shown in Figure 3-8. N-0861 produced a concentrationdependent displacement of specific [3H]CPX binding with an IC50 of 0.8 M and a Hill slope of 1.0. This indicated that N0861 bound to a single class of binding sites. Figure 3-9 illustrates the effect of N-0861 on the Al-AdoR inhibitory effect of CPA. ISO (10 pM) produced a 12-fold increase in cAMP accumulation above the basal level and CPA (0.1 AM) inhibited the stimulatory effect of ISO by 70%. N-0861 (10 JiM) attenuated the inhibitory effect of CPA by 62%. A similar effect of N-0861 on NECA-induced inhibition of ISO-stimulated cAMP accumulation is shown in Figure 3-10. That is, ISO (10 pIM) stimulated cAMP accumulation 47-fold above the basal level and NECA (1 gM) inhibited this increase by 74%. N-0861 (10 gM) attenuated the inhibitory effect of NECA by 60%. The






41



effect of N-0861 on WRC-0018 stimulation of cAMP accumulation is shown in Figure 3-11. The 3-fold increase in cAMP accumulation caused by 0.1 JIM WRC-0018 was not affected by N0861 (0.1 nM 10 pM). Based on these results, in the remaining experiments of this series, 10 JIM N-0861 was used to selectively block the A1-AdoR mediated inhibition of cAMP accumulation.

The effects of N-0861 on the A2-AdoR mediated response of selective and nonselective agonists were investigated. Figure 3-12 illustrates the concentration response of WRC0018 in the absence and presence of N-0861. The biphasic concentration response curve of WRC-0018 in the absence of N0861 was replotted from Figure 3-4. N-0861 (10 gM) completely abolished the A1-AdoR mediated inhibition of cAMP accumulation caused by WRC-0018 and hence, the downward component of the biphasic concentration response curve of WRC-0018 was eliminated. The EC50 for the A2-AdoR mediated effect of WRC-0018 on cAMP accumulation was 93 nM. The maximal stimulation of cAMP accumulation by WRC-0018 was increased (p0.05) from 81 -fold in the absence to 156 fold in the presence of N-0861. The effects of NECA on cAMP accumulation in the absence and presence of N-0861 are shown in Figure 3-13. In the absence of N-0861, NECA did not stimulate cAMP accumulation. In contrast, in the presence of N-0861 (10 pM), NECA stimulated cAMP accumulation over the concentration range of 1 pM 100 gM. At concentrations of NECA >10 HM, there was a decrease in the cAMP level despite






42



the presence of N-0861 in the incubation medium. Similar results to those of NECA were observed with ADO (Figure 314). ADO (0.1 nM 10 pM) caused no effect on the cellular cAMP content in the absence of N-0861 whereas ADO (1 pM 5 JIM) stimulated cAMP accumulation in the presence of 10 pM N0861. At 10 JM ADO, there was a decrease in cAMP formation despite the presence of N-0861 in the incubation medium.

In summary, these data show that the selective A1-AdoR antagonist N-0861 abolishes the downward phase of the WRC0018 biphasic concentration response and uncovers a stimulatory (i.e., A2-AdoR mediated) response of NECA and ADO.


Effect of PTX on the A2-AdoR Mediated Response


The second experimental approach used to unmask the A2AdoR mediated effect on cAMP accumulation was uncoupling of Al inhibitory effects with PTX. Based on a previous study which showed that inhibition of AC activity by (R)-PIA was markedly attenuated after an 18 hr pretreatment of DDT cells with 100 ng/ml PTX (Ramkumar et al., 1990), I chose to incubate our DDT cells with PTX for 18 hr. The ability of CPA to inhibit ISO-stimulated cAMP accumulation in cells pretreated with various concentrations of PTX for 18 hr is shown in Figure 3-15. In control (untreated) DDT cells, 10 VLM ISO produced a 15-fold increase in cAMP content above the basal level. As expected, CPA (1 gM) attenuated the






43



stimulatory effect of ISO by 84%. Pretreatment of cells with 25 ng/ml PTX for 18 hr resulted in a complete loss of the CPA inhibitory effect on cAMP accumulation. Basal and 10 kM ISO-stimulated cAMP accumulation were not significantly affected after pretreatment of the cells with PTX. Based on these results, in the remaining experiments of this series, DDT cells were pretreated for 18 hr with 25 ng/ml of PTX.

Figure 3-16 depicts the effects of WRC-0018 on cAMP accumulation in control and PTX-pretreated cells. The biphasic WRC-0018 concentration response in control cells was replotted from Figure 3-4. After pretreatment of the cells with PTX, similar to using A1-AdoR antagonist, the Al-AdoR mediated inhibition of cAMP accumulation is blocked and hence, the downward component of the concentration response curve of WRC-0018 was abolished. The EC50 value of the A2-AdoR mediated effect of WRC-0018 on cAMP accumulation was 90 nM. The maximal stimulation of cAMP accumulation was 143 -fold above basal in PTX-pretreated cells (control cells, 81 -fold above basal). The effect of NECA on cAMP accumulation in control and PTX-pretreated cells is shown in Figure 3-17. The data for NECA in control cells were replotted from Figure 313. In PTX-pretreated cells, NECA (10 nM 10 gM) stimulated cAMP accumulation with an EC50 value of 180 nM. The maximal response was 5-fold above the basal level. Figure 3-18 illustrates the effect of ADO on cAMP accumulation in control and PTX-pretreated cells. The data for ADO in control cells were replotted from Figure 3-14. Over the concentration range






44



of 1 .M 10 AM, ADO increased the cAMP level in PTXpretreated cells. A maximal response was not achieved even at 10 AM ADO and hence, the EC50 value of the A2-AdoR mediated response of ADO could not be calculated.

In summary, these data show that following PTXpretreatment, the selective A2-AdoR agonist WRC-0018 causes a sustained A2-AdoR mediated stimulatory effect on cAMP accumulation. Likewise, the A2-AdoR mediated stimulatory effects of the nonselective agonists NECA and ADO were unmasked after pretreatment of DDT cells with PTX. The results also indicate that the Al inhibitory effect of AdoR agonists was Gi protein mediated.


Effect of Desensitization and/or Down-reaulation of Ai-AdoR
on the A-AdoR Mediated Response


The third experimental approach used to uncover the A2AdoR mediated effect on cAMP accumulation involves desensitization and/or down-regulation of A1-AdoR. Figure 319 illustrates a representative Scatchard plot of [3H]CPX binding to cell membranes from control (untreated) and CCPA (Al-AdoR agonist)-pretreated cells. The inhibition of ISO (10 AM)-stimulated cAMP accumulation by 1 pM CPA was investigated in cells pretreated with various concentrations of CCPA (0.1 nM 1 AM) for 16 hr (data not shown). Cells incubated with

0.1 AM CCPA for 16 hr at 370C followed by four wash cycles showed a 48% reduction in specific [3H]CPX binding (control, 0.4 pmol/mg protein) with no change in the KD value for the






45



remaining receptors labeled with [3H]CPX (control, 0.5 nM; CCPA-pretreated, 0.4 nM).

The concentration-response relationships of CPA, NECA and WRC-0018 induced attenuation on ISO-stimulated cAMP accumulation in control and CCPA-pretreated cells are shown in Figures 3-20, 3-21 and 3-22, respectively. In control cells, CPA decreased cAMP content in a concentrationdependent manner with an EC50 value of 1.2 nM and maximal inhibition of 78% (Figure 3-20). In comparison, after pretreatment of the cells with CCPA, the EC50 value for CPAinduced inhibition on ISO-stimulated cAMP accumulation was increased by 13-fold (EC50, 15 nM) without any significant change in the maximal inhibition. Pretreatment of the cells with 0.1 .M CCPA for a longer period (68 hr) did not cause a further shift to the right of the concentration response curve of CPA and had no effect on the maximal response of this agonist (data not shown). As illustrated in Figure 3-21, in control cells, NECA decreased cAMP content in a concentration-dependent manner with an EC50 value of 7.2 nM and maximal inhibition of 75%. After pretreatment of the cells with CCPA, the EC50 value for NECA-induced inhibition on ISO-stimulated cAMP accumulation was increased by 17-fold (EC50, 120 nM) without any significant change in the maximal inhibition. The effect of pretreatment of the cells with CCPA on WRC-0018 induced attenuation on ISO-stimulated cAMP accumulation is depicted in Figure 3-22. In control cells, WRC-0018 decreased the cAMP content with an EC50 value of 830






46



nM and caused a maximal inhibition of 77%. After pretreatment of the cells with CCPA, the inhibition of ISO-stimulated cAMP accumulation by WRC-0018 was markedly attenuated with a maximal inhibition of cAMP accumulation of about 20%.

The effects of pretreatment of cells with CCPA on the A2-AdoR mediated effect of WRC-0018 is shown in Figure 3-23. The control data from Figure 3-4 were replotted in Figure 323. In CCPA-pretreated cells, the WRC-0018 mediated attenuation of cAMP accumulation was abolished. The EC50 of the WRC-0018 mediated stimulation of cAMP accumulation was 17 nM. The maximal stimulation of cAMP accumulation was not significanly increased (control, 81 -fold above basal; CCPApretreated, 30 -fold above basal). The effects of pretreatment of the cells with CCPA on the A2-AdoR mediated response of NECA is shown in Figure 3-24. The data for NECA in control cells were replotted from Figure 3-13. In CCPApretreated cells, NECA did not stimulate cAMP accumulation, however, the basal level of cAMP was significantly increased (p<0.0005) from 32 in control cells to 111 pmol/mg protein/min in pretreated cells. A longer time of preincubation of the cells with CCPA (68 hr) also failed to unmask a NECA mediated A2-AdoR mediated increase in cAMP accumulation (data not shown). The effects of pretreatment of the cells with CCPA on the A2-AdoR mediated response of ADO is shown in Figure 3-25. The data for ADO in control cells were reploted from Figure 3-14. In CCPA-pretreated cells, ADO did not stimulate cAMP accumulation.






47



In summary, these data show that after CCPApretreatment, the selective A2-AdoR agonist WRC-0018 causes a sustained A2-AdoR mediated stimulatory effect on cAMP accumulation. However, the A2-AdoR mediated stimulatory effects of the nonselective agonists NECA and ADO were still not uncovered after pretreatment of DDT cells with CCPA.


Effect of Adenosine on the Desensitization of Al- and A.zAdoR
Systems


The endogenous agonist for the AdoR is ADO which has

been shown to be a nonselective agonist (Olsson and Pearson, 1990; Londos et al., 1980). In a separate series of experiments, I investigated whether ADO had differential effects on the desensitization of AdoR subtypes and hence, determined if the expression of the A2-AdoR mediated response could be uncovered. Cells were pretreated with 100 gM ADO to ensure that both receptor subtypes were stimulated with the agonist.

Desensitization of Ai-AdoR

The effect of ADO on the desensitization of Al-AdoR was investigated by studying the effect of CPA to inhibit ISOstimulated cAMP accumulation in control (untreated) and ADO (100 RM, 24 hr)-pretreated cells (Figure 3-26). DIP and EHNA were present during the period of pretreatment of the cells with ADO to prevent ADO metabolism and thereby maintain ADO concentration in the incubation medium relatively constant. As depicted in Figure 3-26, in control cells, CPA decreased






48



cAMP content in a concentration-dependent manner with an EC50 value of 2.0 nM and maximal inhibition of 77%. In comparison, after pretreatment of the cells with ADO, the EC50 value for CPA-induced inhibition on ISO-stimulated cAMP accumulation was increased by 15-fold (EC50, 30 nM) without any significant change in the maximal inhibition (pretreated, 73%). Desensitization of A-AdoR

The effect of ADO on the desensitization of A2-AdoR were investigated by studying the effect of WRC-0018 on cAMP accumulation (Figure 3-27). In control and DIP+EHNApretreated cells, WRC-0018 produced a biphasic concentration response curve. At low concentrations (1 500 nM), WRC-0018 stimulated cAMP accumulation with a maximal increase of 4fold above the basal level. At higher concentrations (>500 nM) of WRC-0018, the cAMP accumulation was attenuated. After pretreatment of the cells with ADO, the basal level of cAMP was significantly increased (p50.025) from 31 to 91 pmol cAMP/mg protein/min. However, WRC-0018 did not stimulate cAMP accumulation above the basal level.

The effect of ADO on cAMP accumulation after

pretreatment of the cells with ADO is shown in Figure 3-28. In both control or ADO-pretreated cells, ADO did not stimulate cAMP accumulation above the basal level.






49


Discussion


The interaction of the inhibitory A1-AdoR and the

stimulatory A2-AdoR was investigated using DDT cells. The coexpression of both AdoR subtypes was first shown in this cell line by Ramkumar et al. (1990). In addition, BARs which mediate the stimulation of cAMP accumulation are also expressed in DDT cells (Norris et al., 1983).

In the presence of a constant concentration of ISO which increased cAMP accumulation, several Al-AdoR agonists inhibited the ISO stimulatory effect in a concentrationdependent manner. Furthermore, in the presence of a fixed concentration of Al-AdoR agonists, the effect of ISO on cAMP accumulation was greatly attenuated. The inhibitory effect of CPA on ISO-stimulated cAMP formation was blocked by 1 gM CPX, a selective Al-AdoR antagonist (data not shown). These findings indicate that CPA exerts its inhibitory effect on BAR mediated cAMP accumulation by activating Al-AdoRs which are known to be negatively coupled to AC. This Al-AdoR mediated inhibition on cAMP formation confirms and extends the recent reports on AdoRs in DDT cells (Ramkumar et al., 1990; Gerwins et al., 1990; Gerwins and Fredholm, 1991). One difference between the present data and the previous reports is that NECA produced a 97% maximal inhibition of cAMP accumulation in intact DDT cells in our experiments. However, only a 30% maximal inhibition of AC activity in DDT cell membranes by NECA was reported by Ramkumar et al.(1990). The






50



lower maximal inhibition of AC activity in DDT cell membranes may be due to the homogenization process during membrane preparation affecting the function of A1-AdoR.

Our experiments also suggest that the stimulation of

cAMP by low concentrations of WRC-0018 was mediated by an A2AdoR. Evidence supporting this include the inhibitory effect of the AdoR antagonist 8PST and lack of effect of the selective Al-AdoR antagonist N-0861 on the stimulatory response of WRC-0018. Since a selective A2-AdoR antagonist has yet to be synthesized, the next best candidate, 8PST, which is only slightly more selective to A1-AdoR with a Ki(A2)/Ki(A1) ratio of 5.9 (Trivedi et al., 1990) was used.

Interestingly, low concentrations of WRC-0018 which increase cAMP accumulation failed to potentiate ISOstimulated cAMP accumulation (Figure 3-7). This may be because A2-AdoR and BAR share the same pool of AC for cAMP production and the system may be maximally stimulated at 10 RM ISO. Also observed was a greater stimulatory effect of ISO on cAMP accumulation (3 times greater) as compared with WRC0018. This finding could be explained by 1) a higher density of BAR on DDT cells, 2) a higher coupling efficiency of BAR for the Gs protein or 3) an easier access of BAR to the AC pool.

The Al-AdoR agonist CPA inhibited the A2-AdoR

stimulatory effect of WRC-0018 in our experiment suggesting that the inhibitory Al-AdoRs dominate over the stimulatory A2AdoRs in DDT cells. This dominant role of Al-AdoR may also






51



explain two observations. First, the assumed nonselective agonists NECA and ADO showed no stimulation of cAMP accumulation. At effective concentrations of either agonist, both AdoR subtypes would be activated, but the A1-AdoR mediated inhibition would predominate and hence inhibit the A2-AdoR mediated stimulation of AC and thereby accumulation of cAMP. However, this finding is in contrast to the observation reported by Ramkumar et al. (1990). NECA (10 JM), in absence of any Al-AdoR antagonist, produced a 2.1 fold stimulation of AC activity over the basal level in DDT cell membranes (Ramkumar et al., 1990). The discrepancy between this report and our data may be due to the homogenization process during membrane preparation affecting the function of A1-AdoR. The second observation which could be explained by the dominant role of Al-AdoR is the downward portion of the WRC-0018 biphasic concentration response may reflect the activation of the Al-AdoR. This was indicated by the inhibitory effect of WRC-0018 on ISO-stimulated cAMP accumulation over the same concentration range where the downward part of the biphasic response of WRC-0018 occurred (Figure 3-7).

Three approaches were used to study the expression of A2-AdoR mediated responses and further establish that the downward phase of the WRC biphasic response is Al-AdoR mediated. The first approach involved the use of the selective Al-AdoR antagonist N-0861. The selectivity of N0861 for the A1-AdoR was tested by receptor binding and cAMP






52



accumulation. Several lines of evidence support the high selectivity of N-0861 for A1-AdoR. These include 1) a concentration dependent displacement of specific [3H]CPX binding at the A1-AdoR site by N-0861 with a Hill slope of 1, indicating the interaction of this antagonist with a single class of binding sites, 2) the substantial attenuation of the CPA mediated inhibition of cAMP accumulation by 10 JM N-0861, and 3) the unaltered stimulatory effect of WRC-0018 by N-0861 at concentrations ranging from 0.1 nM 10 IM.

In the presence of N-0861, the stimulatory response of WRC-0018 was sustained, i.e., the downward portion of the biphasic response was abolished. This finding is consistent with the hypothesis that the downward portion of the WRC-0018 biphasic response was A1-AdoR mediated. Assuming that the interaction of N-0861 with the A1-AdoR and the interaction of WRC-0018 with A1- and A2-AdoR are reversible and competitive, then in the presence of N-0861, the downward portion of WRC0018 concentration response should be shifted to the right. This was not tested in the present study due to the insolubility of WRC-0018 to obtain and test concentrations above 100 JM. Similar to the results with WRC-0018, blockade of the Al-AdoR with N-0861 uncovered a cAMP stimulation response of the agonists NECA ( 100 gM) and ADO ( 5 LM). These results are in keeping with the observation on NECA mediated A2-AdoR stimulatory response on AC activity reported by Ramkumar et al. (1990). However, at 210 pM NECA (Figure 313) and 10 )IM ADO (Figure 3-14), there was a decrease in cAMP






53



accumulation. This attenuation may be due to the higher concentrations of NECA and ADO overcoming the N-0861 (10 pM) blockade of the A1-AdoR allowing its activation.

The second approach used to investigate the expression of A2-AdoR mediated response was uncoupling of the A1-AdoR inhibitory effect with PTX. It is well established that PTX selectively ADP-ribosylates the ai subunit. This covalent modification inactivates the Gi protein and uncouples inhibitory receptors including the A1-AdoR resulting in loss of receptor mediated inhibition of AC activity (Gilman, 1987; Nathanson, 1987; Hazeki and Ui, 1981). Experiments showed that the inhibition of AC activity by (R)-PIA in DDT cells was greatly attenuated after pretreatment of the cells with 100 ng/ml PTX for 18 hrs (Ramkumar et al., 1990). In the present study, the ability of CPA to inhibit ISO-stimulated cAMP accumulation was used as means to determine the effect of PTX-pretreatment on the Al-AdoR mediated inhibition of cAMP accumulation. PTX (25 ng/ml)-pretreatment for 18 hr was found to be sufficient to cause complete loss of the CPA inhibitory effect. In PTX-pretreated cells, the downward phase of the WRC-0018 biphasic response was completely eliminated. Furthermore, NECA and ADO which had no effect on cAMP accumulation in untreated cells stimulated cAMP accumulation in cells pretreated with PTX. The result with NECA, i.e., the stimulation of cAMP accumulation in PTXpretreated cells, is also consistent with other reports (Gerwins et al., 1990). That is, NECA (10 gM) was reported to






54



increase cAMP content 1.5 fold above the basal level in DDT cells pretreated with 200 ng/ml PTX for 4 hr (Gerwins et al., 1990). Our data strongly suggest that the downward phase for WRC-0018 concentration response is due to activation of an inhibitory receptor. The EC50 values of the A2-AdoR effect of WRC-0018 in the presence of N-0861 (93 nM) and after PTXpretreatment (90 nM) were very similar. In addition, similarity also exists between the maximal responses of the A2-AdoR effect of WRC-0018 in the presence of N-0861 (156 fold above basal) and after PTX-pretreatment (143 -fold above basal). These results suggest that N-0861 and PTX have a similar net blocking effect on the Al-AdoR mediated inhibitory action of WRC-0018 and are unlikely to affect the interaction of WRC-0018 with the A2-AdoR. As a consequence, the A2-AdoR mediated stimulatory action of WRC-0018 is sustained to the same extent.

Selective desensitization and/or down-regulation of A1AdoR was the third approach used. Chronic pretreatment (16 hr) of cells with the selective Al-AdoR agonist CCPA (0.1 9M) caused a 48% loss of Al-AdoR. This indicates that the Al-AdoR is down-regulated after chronic pretreatment with a selective A1-AdoR agonist. Down-regulation of receptors after long term exposure to an agonist is a widely reported phenomenon. However, the loss of Al-AdoR during down-regulation is significantly less in comparison with other receptor systems. For instance, after 1321N1 astrocytoma cells were incubated with ISO for 12-24 hr, greater than 90% of the BARs were lost






55



from the cells (Doss et al., 1981). In DDT cells, downregulation of BAR occurred rapidly with a tl/2 of about 3 hr and proceeded to 80-85% loss of receptors by 13 hr of incubation of the cells with 10 JM epinephrine (Toews, 1987).

Desensitization of Al-AdoR system was studied by

examining the concentration responses of CPA, NECA and WRC0018 to inhibit ISO-stimulated cAMP accumulation in control and CCPA-pretreated cells. For CPA (Figure 3-20) and NECA (Figure 3-21), the concentration response curves were shifted to the right (10-20 -fold) without any change in the maximal responses. Longer pretreatment periods did not increase the magnitude of shift (data not shown) indicating that maximal desensitization was achieved. The shift in the concentration response is not surprising because the sensitivity of the cells to the agonist should decrease as the receptor number decreases provided little or no change in the affinity of the agonist for the remaining receptors. The ability of CPA and NECA to produce the same maximal response in control (untreated) cells and CCPA-pretreated cells where the receptor number is decreased by almost 50% may be explained by the presence of spare Al-AdoRs. In this situation, the responsiveness is not directly proportional to receptor occupancy and the maximal response can be obtained when less than 100% of the receptors are activated. Many studies have shown the existence of spare receptors in other systems (Stephenson, 1956; Nickerson, 1956; Nelson et al., 1986; Gunst et al., 1989). For example, in guinea pig lung, after






56



=50% of SARs were inactivated by an irreversible antagonist, the maximal airway responsiveness to ISO was still maintained (Nelson et al., 1986). In canine trachealis muscle, the maximal contractile response for acetylcholine was achieved when only 4% of muscarinic receptors were occupied (Gunst et al., 1989).

Interestingly, in control cells, WRC-0018 produced the same maximal inhibitory response on ISO-stimulated cAMP accumulation as CPA. However in CCPA-pretreated cells, the maximal response produced by WRC-0018 was greatly reduced (Figure 3-22) whereas the maximal response achieved by CPA was not changed from that in control cells. This indicated that WRC-0018 acted as a full Al-AdoR agonist (as compared with CPA) in control cells whereas it acted as a partial agonist in pretreated cells. One explanation for this may be related to the intrinsic efficacy of the agonists. In control cells, WRC-0018 may need to occupy more receptors than CPA to achieve its maximal response. Thus, for WRC-0018, there may be little or no spare Al-AdoRs and the responsiveness may be more directly related to receptor occupancy. As the receptor number decreases, the maximal response for WRC-0018 will be reduced as was observed after chronic CCPA-pretreatment. A desensitization-induced change in agonist efficacy has also been reported for some BAR agonists. In membranes prepared from untreated L6 skeletal muscle cells, several compounds acted as full BAR agonists when compared with ISO, but after desensitization of the BAR system, these compounds acted as






57



partial agonists (Pittman et al., 1984). Alternative explanations may also be possible. For example, there may be subpopulations of Al-AdoRs in DDT cells where CPA is nonselective and WRC-0018 is selective for only one subtype which was selectively down-regulated and desensitized by chronic CCPA-pretreatment. Further experiments will be necessary to determine the reason for the differential changes between CPA and WRC-0018 responsiveness after a decrease in receptor number.

Because CCPA-pretreatment had little effect on the A2AdoR mediated stimulation component of the WRC-0018 concentration response (Figure 3-23), this suggests that CCPA has little or no desensitization effect on the A2-AdoR. On the other hand, chronic CCPA-pretreatment abolished the downward phase of the WRC-0018 biphasic concentration response curve (Figure 3-23). CCPA is a highly selective AlAdoR agonist (Lohse et al., 1988). The loss of the downward phase of the WRC-0018 response in conjunction with a desensitization and/or down-regulation of Al-AdoR is consistent with the downward phase being Al-AdoR mediated. Thus, the elimination of the downward phase may be due to 1) desensitization and/or down-regulation decreasing the sensitivity to Al-AdoR agonists thereby further increasing the A2-AdoR selectivity of WRC-0018, and 2) the loss of efficacy of WRC-0018 as an Al-AdoR agonist in the partially desensitized state.






58



In contrast to the effects of N-0861 and PTX, chronic

CCPA-pretreatment did not uncover any NECA or ADO stimulation of cAMP accumulation. This observation can be explained by the differences in the agonist's sensitivity to produce inhibition or stimulation of cAMP accumulation.


Summary of EC50 Values (nM)


Agonists A1-AdoR A2-AdoR
Control CCPA (in PTX-pretreated cells)


WRC-0018 830 90 NECA 7.2 120 180 ADO 97 >1000 CPA 1.2 15



In the case of NECA, the estimated EC50 values for the inhibition (Al-AdoR effect) and stimulation (A2-AdoR effect) of cAMP accumulation in untreated cells were 7.2 and 180 nM, respectively. The latter estimate was derived from the NECA concentration response curve after PTX-pretreatment and this value was based on the assumption that PTX-pretreatment had little or no effect on the potency of this agonist to stimulate cAMP accumulation. The higher potency of NECA to inhibit cAMP accumulation indicates that this agonist shows some degree of selectivity for the A1-AdoR over the A2-AdoR in DDT cells. In contrast, NECA has been reported to be a nonselective agonist in other systems (Hutchinson et al, 1990). After chronic CCPA-pretreatment to partially down-






59



regulate the A1-AdoR, the EC50 value for the NECA inhibitory response was increased from 7.2 nM in control cells to 120 nM in CCPA-pretreated cells which was similar to the EC50 value for stimulation (180 nM) (see the table above). That is, the EC50 value for NECA-induced cAMP accumulation (A2-AdoR effect) is still greater than NECA's EC50 value for inhibition of cAMP accumulation (A1-AdoR effect). Similar EC50 values for NECA to inhibit and stimulate cAMP accumulation in cells chronically pretreated with CCPA indicates that this agonist would occupy both receptor subtypes, but activation of the Al-AdoR would prevent expression of an A2-AdoR response. It appears that the expression of an A2-AdoR response in untreated cells requires a selective A2-AdoR agonist. This was evident with WRC-0018 which based upon its stimulatory and inhibitory concentration ranges, showed an approximately 500-600 -fold selectivity for the A2-AdoR (Figure 3-4). Although chronic CCPA-pretreatment reduced the sensitivity for the A1-AdoR response of NECA, this reduction was not sufficient to result in activation of the A2-AdoR without significant stimulation of the A1-AdoR. A similar explanation may account for the lack of an ADO mediated stimulation of cAMP accumulation in cells chronically pretreated with CCPA. The EC50 for ADO to inhibit and stimulate (in PTX-pretreated cells) cAMP accumulation was 97 nM and >1 gM, respectively. This A1-AdoR selectivity for ADO is in contrast to previous studies in other cell types where this agonist has been shown to be nonselective (Londos et al., 1980; Olsson and Pearson, 1990).






60



Assuming a 10-20 fold reduction in ADO sensitivity (like CPA) to inhibit cAMP accumulation in cells chronically pretreated with CCPA, this would not be sufficient to allow selective expression of the A2-AdoR response. However, it will be of interest to determine if selective down-regulation of the AlAdoR would reduce the sensitivity of a non-selective agonist sufficient to uncover expression of the A2-AdoR response.

Interestingly, CCPA-pretreatment was found to increase the basal cAMP level (Figure 3-24). Desensitization of inhibitory receptors resulting in an increase in basal cAMP levels or potentiation of stimulatory receptor effects has been widely reported. For example, a 3-fold increase in basal cAMP level was observed after pretreatment of NG108-15 cells with 10 gM carbachol for 19 hr (Nathanson et al., 1978). This increase in basal cAMP level may be due to the loss of an inhibitory tone on AC activity mediated by the Gi protein. The inhibitory tone may involve the basal dissociation of Gi (in the absence of an inhibitory agonist) with the ai subunit directly attenuating the activity of the catalytic subunit of AC (Gilman, 1987). Several mechanisms may be responsible for the loss of the inhibitory tone. First, chronic CCPApretreatment may decrease the cellular content of Gi protein. Evidence supporting this contention has been reported by Green (1987) who showed a decrease in Gi protein after desensitization of the Al-AdoR system in primary culture of rat adipocytes. Second, CCPA-pretreatment may induce an impairment in the function of Gi. For example, a CCPA-induced






61



phosphorylation of Gi may prevent the basal release of the activated ai subunit.

Because Al-AdoR inhibition of cAMP accumulation

dominated over A2-AdoR mediated stimulation, it was of interest to investigate whether the endogenous agonist ADO would have a differential desensitization effect on the receptor subtypes when both were chronically stimulated. After pretreatment of cells with 100 pM ADO for 24 hr, the concentration response of CPA to inhibit ISO-stimulated cAMP accumulation shifted to the right indicating that the Al-AdoR system was desensitized (Figure 3-26). This was similar to that observed with CCPA-pretreatment. However, WRC-0018 did not increase cAMP accumulation in ADO-pretreated cells, suggesting that the A2-AdoR system was also desensitized (Figure 3-27). In keeping with this conclusion, ADO also did not stimulate cAMP accumulation in ADO-pretreated cells. The observation that the A1-AdoR was still present (albeit at reduced agonist sensitivity) whereas the A2-AdoR response was abolished indicates that the extent of desensitization was different for each receptor subtype. This may be explained by one or more factors: 1) the density of Al-AdoR in DDT cells was higher than that of A2-AdoR. An Al-AdoR to A2-AdoR ratio of 4:1 in DDT cells has been reported by Ramkumar et al. (1990), 2) the coupling efficiency between the receptors and their second messenger system may be different, 3) the rate of desensitization of A2-AdoR was much faster than that of A1AdoR. Ramkumar et al. (1991) reported that the tl/2 for the






62



desensitization of Al- and A2-AdoR was 8 and 0.75 hr in DDT cells, respectively, 4) the efficacy of the agonist to induce desensitization of the receptor subtypes may be different, or 5) the mechanism for desensitization of the receptor subtypes may be different. Evidence for a differential desensitization mechanism has been reported recently (Ramkumar et al., 1991). During desensitization, the Al-AdoR was down-regulated (internalized), uncoupled from G proteins and phosphorylated whereas the A2-AdoR was not (Ramkumar et al., 1991).


Summary


Adenosine is an autocoid produced by the same cells on which and/or adjacent cells it exerts its effects. The results of the present study demonstrate that in DDT cells adenosine acts on at least two receptor subtypes (Al- and A2AdoR) whose actions result in opposing effects on the formation of cAMP. There may be circumstances in which adenosine production rapidly and transiently increases and thereby maximally activates AdoRs. When the concentration of adenosine rises far above the physiological range, the coexistence of two receptor subtypes on the same cell with opposing functional effects would dampen the responses to the transient extremes of adenosine concentration. That is, in the example illustrated in Figure 3-14 (ADO alone), activation of the Al-AdoR would attenuate the effect of A2-






63



AdoR activation of AC and thereby dampen a rise in cellular cAMP.

The data from the present study show that in DDT cells activation of the inhibitory AI-AdoR will predominate and thus mask the stimulatory A2-AdoR response. This observation may have broader implications because other cell types and tissues have been reported to express both Al- and A2-AdoRs. These include porcine coronary vascular smooth muscle cells (Mills and Gewirtz, 1990), FRTL-5 cells derived from normal rat thyroid (Nazarea et al., 1991), ventricular myocytes from chick embryo (Xu et al., 1992) and heart tissue (Olsson and Pearson, 1990).

Our data demonstrate that cells with both receptor

subtypes can be pharmacologically manipulated to uncover an A2-AdoR response with a highly selective A2-AdoR agonist or with the use of a selective Al-AdoR antagonist. There may be situations where these pharmacological approaches have therapeutic potential and suggest possible strategies for drug development. For example, in cells or tissues where both receptor subtypes are expressed, the A2-AdoR mediated responses may be suppressed under normal conditions. However, by selectively blocking the Al-AdoR system or activating the A2-AdoR with a highly selective agonist, A2-AdoR mediated responses (e.g., vasodilation to increase blood flow, inhibition of platelet aggregation during thrombosis, generation of superoxide radicals to prevent reperfusion injury) may be expressed for possible therapeutic effect.






64



Another implication of our study with potential

therapeutic value is the possibility to decrease the side effects caused by nonselective AdoR agonists. For instance, adenosine and its analogs may activate both receptor subtypes and hence, if one subtype can be selectively blocked, the side effects mediated by this receptor subtype should be attenuated.

The development of more selective agonists and

antagonists may be needed to increase the concentration range over which a response can be achieved. Thus, the ultimate objective would be to develop specific agonists and antagonists.

Finally, in cell types in which the Al- and A2-AdoR coexist, the question arises as to whether an A2-AdoR mediated response initiated by ADO would ever be expressed under physiological or pathological conditions? Although A1AdoR suppression of the A2-AdoR response may be normal under most physiological conditions (at least in cells that express both receptor subtypes), the A2-AdoR response may be needed under some pathological conditions. For example, under stress, the chronic release of adenosine or other cellular mechanisms may result in the loss of the Al-AdoR or other components of its signal transduction pathway (e.g., Gi protein) allowing A2-AdoR expression. Although our data indicated that chronic stimulation of Al- and A2-AdoR resulted in desensitization of both receptors and Ramkumar et al. (1991) found that the rate of desensitization for the A2-AdoR






65



was faster than that for the A1-AdoR in DDT cells, in other cells, the opposite may occur. Thus, additional studies are necessary to test this latter possibility, further characterize the dual modulation of AC by adenosine in the same cell and/or organ and define the implications for such role.





66




1. Selective A,-AdoR antagonist 2. Pertussis toxin (PTX)




9 Gi A" G1
AC AC
0 Gs A2 s A2
cAMP cAMP



3. Desensitization / Down-regulation of A,-AdoR




At
QG

Gs A2 Gs A2
cAMP cAMP





Figure 3-1. Experimental approaches to express the A2AdoR mediated response.






67






100



080



E 60


(G 40
1m6 O CPA


o 20 A NECA

o ADO

-11 -10 -9 -8 -7 -6 -5


Log [Drug], M





Figure 3-2. Inhibition of ISO-stimulated cAMP accumulation in DDT cells by CPA, NECA and ADO. Cells were incubated in HBSS containing 50 PM rolipram, 10 gM ISO or ISO plus the indicated concentrations of CPA, NECA or ADO + 1 gM DIP + 1 JIM EHNA for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The control cAMP accumulated in the presence of ISO alone was 5712, 879 and 684 pmol/mg protein/min for the CPA, NECA and ADO experiments, respectively. The basal level of cAMP accumulated was below the detection level for CPA and NECA and 61 pmol/mg protein/min for the ADO experiment. Each data point is the meanSD of quadruplicate determinations and is representative of 2 experiments.





68






o ISO
80
C__ 8 ISO+CPA E A ISO+NECA

60 E o



440
(mm


uE 20




-9 -8 -7 -6 -5


Log [Isoproterenol], M





Figure 3-3. The effect of CPA and NECA on ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 PIM rolipram, the indicated concentrations of ISO, without or with 0.1 PM CPA or 1 gM NECA for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The basal level of cAMP accumulated was 11 pmol/mg protein/min. Each data point is the meanSE, n=3 5.






69






20 O WRC-0018 *


no 0_ A NECA m 15




O1 *




<~ 5
0
0 _


-10 -9 -8 -7 -6 -5 -4


Log [Drug], M





Figure 3-4. The effect of WRC-0018 and NECA on cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 iM rolipram and the indicated concentrations of WRC-0018 or NECA for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The basal level of cAMP accumulated was 31, 32 pmol/mg protein/min and each data point is the meanSE, n=15 for WRC-0018 and n=3 for NECA. The
* indicates a p<0.01 for each WRC-0018 data point as compared to its basal level.





70





25 E BASAL
O WRC-0018
.I a WRC-0018+8PST
20
E E


CD 15



o. E
2O


40
Ef 5


0









Figure 3-5. The effect of 8PST on WRC-0018-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 pM rolipram, without or with 0.1 gM WRC-0018 or WRC-0018 plus 5 JIM 8PST for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determination. The indicates a p<0.01 for the WRC-0018 value compared to the basal value and a p<0.005 for the WRC-0018 plus 8PST value compared to the WRC-0018 value.






71






25 0 BASAL
25
0 WRC-0018
g E 1 WRC-0018+CPA




O 15



eE 10






0









Figure 3-6. The effect of CPA on WRC-0018-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 M rolipram, without or with 0.1 M WRC-0018 or WRC-0018 plus 1 .M CPA for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determination. The indicates a p<0.005 for the WRC-0018 value compared to the basal value and for the WRC-0018 plus CPA value compared to the WRC-0018 value.





72





80 WRC-0018 ALONE
80
S. U WRC-0018+ISO

60



O 40

a E 20
20
CL


0 L
-10 -9 -8 -7 -6 -5

Log [WRC-0018], M




Figure 3-7. Effect of WRC-0018 on ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 JIM rolipram, the indicated concentrations of WRC-0018, without or with 10 4M ISO for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The control cAMP accumulated in the presence of ISO alone was 629 pmol/mg protein/min. Each data point is the meanSE, n=4.





73






100


80


00

60
O O



O 40


20



-8 -7 -6 -5 -4


Log [N-0861], M





Figure 3-8. Inhibition of specific [3H]CPX binding to DDT cell membrane by N-0861. Membranes protein (320 gg) was incubated with 1 nM [3H]CPX and the indicated concentrations of N-0861 for 150 min at room temperature. Specific binding was determined as described in the "Methods" section. Data points are the mean of triplicate determinations which varied by less than 5%. The control specific [3H]CPX binding was 132 fmol/mg protein.





74





0 ISO

-, 50 o ISO+CPA

E ISO+CPA+N-0861
S40


03


L E 20o

<0 10 ZT CL


0 L
-10 -9 -8 -7 -6 -5


Log [N-0861], M






Figure 3-9. The effect of N-0861 on the ability of CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 gM rolipram, 10 M ISO, without or with 0.1 JIM CPA or CPA plus the indicated concentrations of N-0861 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determinations. The indicates a p<0.0005 for the ISO + CPA + N-0861 value compared to the ISO + CPA value.






75






0 ISO

,w 60 ~ ISO+NECA

E A ISO+NECA+N-0861


40





- 20




0 -I
-10 -9 -8 -7 -6 -5


Log [N-0861], M






Figure 3-10. The effect of N-0861 on the ability of NECA to inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 p.M rolipram, 10 pM ISO, without or with 1 M NECA or NECA plus the indicated concentrations of N-0861 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determination. The indicates a p<0.0005 for the ISO + NECA + N-0861 point from the ISO + NECA point.





76





25 3 WRC-0018
U WRC-0018+N-0861

20


** 15


mm
O

2 O


Ef 5
.


0 L
-10 -9 -8 -7 -6 -5


Log [N-0861], M





Figure 3-11. The effect of N-0861 on the ability of WRC-0018 to stimulate cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 p.M rolipram, 0.1 M WRC-0018, without or with the indicated concentrations of N-0861 for 7
min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The basal level of cAMP accumulated was 61 pmol/mg protein/min. Each data point is the meanSD of quadruplicate determinations.





77






O WRC-0018

n 30 WRC-0018+N-0861




E 1 20
O


o E
410
oE 0



0 I
-10 -9 -8 -7 -6 -5 -4


Log [WRC-0018], M





Figure 3-12. The effect of N-0861 on WRC-0018-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 JM rolipram, the indicated concentrations of WRC-0018, without or with 10 pM N-0861 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The dose response curve of WRC-0018 in the absence of N-0861 is taken from Figure 3-4. The basal level of cAMP accumulated in the presence of N-0861 was 21 pmol/mg protein/min. Each data point is the meanSE, n=3






78






25 A NECA

A NECA+N-0861
20



E 15
=o

t 10


0, 0




0 -- t I- --I
-9 -8 -7 -6 -5 -4


Log [NECA], M






Figure 3-13. The effect of N-0861 on NECA-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 pM rolipram, the indicated concentrations of NECA, without or with 10 pM N-0861 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The basal level of cAMP accumulated was 11 pmol/mg protein/min. Each data point is the meanSD of quadruplicate determinations and is representative of 2 experiments. The indicates a p<0.0005 for the NECA points from their respective control values.






79






15 O ADO

ADO+N-0861 *



10





oE




0


-10 -9 -8 -7 -6 -5


Log [Adenosine], M





Figure 3-14. The effect of N-0861 on ADO-stimulated cAMP accumulation in DDT cells. Cells were incubated in HBSS containing 50 JIM rolipram, 1 gM DIP, 1 JIM EHNA, the indicated concentrations of ADO, without or with 10 JM N-0861 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The basal level of cAMP accumulated was 60 pmol/mg protein/min. Each data point is the meanSE, n=4. The indicates a p<0.01 for the ADO points from their respective control values.






80






100



Z*E 80


*E 60
0
o A BASAL

CMG 40 0 ISO


ISO+CPA
oE 20

0

0 50 100 150 200


[Pertussis Toxin], ng/mi






Figure 3-15. The effect of PTX pretreatment on the ability of CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media with the indicated concentrations of PTX for 18 hr at 370C. At the end of the incubation period, the cells were washed 4 times with icecold HBSS and detached from the plate. The cells were then incubated in HBSS containing 50 pIM rolipram, without or with 10 gM ISO or ISO plus 1 pM CPA for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determination.






81






O CONTROL
25
.g, U PTX

20


= 0 15




0 1




I I I

-9 -8 -7 -6 -5


Log [WRC-0018], M





Figure 3-16. The effect of PTX pretreatment on WRC-0018stimulated cAMP accumulation in DDT cells. Cells were incubated in growth media without or with 25 ng/ml PTX for 18
hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. Cells were then incubated in HBSS containing 50 gM rolipram and the indicated concentrations of WRC-0018 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The dose response curve of WRC-0018 in control cells is taken from Figure 3-4. The basal level of cAMP accumulated was 31 and 21 pmol/mg protein/min for the control and PTX experiments, respectively. Each data point is the meanSE, n=6.






82






20 A CONTROL

--N A PTX


15
-I

= 0

m 10






eI I I I '


-9 -8 -7 -6 -5 -4


Log [NECA], M






Figure 3-17. The effect of PTX pretreatment on NECAstimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 25 ng/ml PTX
for 18 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. Cells were then incubated in HBSS containing 50 [IM rolipram and the indicated concentrations of NECA for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The data for NECA in control cells are taken from Figure 3-13. The basal level of cAMP accumulated was 11 and 42 pmol/mg protein/min for the control and PTX experiments, respectively. Each data point is the meanSD of quadruplicate determinations and is representative of 2 experiments.






83







o CONTROL
_._ 15
C *0 PTX




E




0


,0



-10 -9 -8 -7 -6 -5


Log [Adenosine], M






Figure 3-18. The effect of PTX pretreatment on ADO-stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 25 ng/ml PTX for 18 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. Cells were then incubated in HBSS containing 50 jM rolipram, 1 JIM DIP, 1 jM EHNA and the indicated concentrations of ADO for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The control data are taken from Figure 3-14. The basal level of cAMP accumulated was 60 and 41 pmol/mg protein/min for the control and PTX experiments, respectively. Each data point is the meanSE, n=3. The indicates a p<0.05 for the ADO point from their respective control value.






84








8 0, CONTROL

o U CCPA

o 6

x

~E 4



2




0 0.1 0.2 0.3 0.4


[3H]CPX Bound
(pmol/mg protein)




Figure 3-19. Scatchard plot of specific [3H]CPX binding to DDT cell membranes after CCPA treatment. Cells were incubated in fresh growth media without or with 0.1 p.M CCPA for 16 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and the membranes prepared. Membrane protein (0.1 mg) was assayed with 0.06 4 nM [3H]CPX as described in the "Methods" section. The data are plotted as the ratio of the amount of specifically bound ligand (pmol/mg protein) to free ligand (pmol/L) versus the amount of specifically bound ligand/mg protein. Data points are the mean of triplicate determination and are representative of 3 experiments.






85







100


8




M. 60

S00
40


o 20 O CONTROL 20 CCPA

0 I I I
-11 -10 -9 -8 -7 -6


Log [CPA], M






Figure 3-20. The effect of CCPA pretreatment on the ability of CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 0.1 gM CCPA for 16 hr at 370C. At the end of the incubation period, the cells were washed 4 times with icecold HBSS and detached. The cells were then incubated in HBSS containing 50 M rolipram, 10 gM ISO or ISO plus the indicated concentrations of CPA for 7 min at 360C. The cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determinations and is representative of 2 experiments.






86







100


80


EO
-060



40

A CONTROL
0 20
A CCPA

0 I II-9 -8 -7 -6


Log [NECA], M






Figure 3-21. The effect of CCPA pretreatment on the ability of NECA to inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 0.1 JM CCPA for 16 hr at 370C. At the end of the incubation period, the cells were washed 4 times with icecold HBSS and detached. The cells were then incubated in HBSS containing 50 pM rolipram, 10 pM ISO or ISO plus the indicated concentrations of NECA for 7 min at 360C. The cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determination.





87








100


80

E2O
C 60


., 40

CONTROL
0 20
20 CCPA


0
-7 -6 -5


Log [WRC-0018], M






Figure 3-22. The effect of CCPA pretreatment on the ability of WRC-0018 to inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 0.1 gM CCPA for 16 hr at 370C. At the end of the incubation period, the cells were washed 4 times with icecold HBSS and detached. The cells were then incubated in HBSS containing 50 gM rolipram, 10 I.M ISO or ISO plus the indicated concentrations of WRC-0018 for 7 min at 360C. The cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSD of quadruplicate determinations and is representative of 2-4 experiments.






88






25
25 CONTROL


o CC CCPA
E" 20






0 10

O.E

40 0

0

oI I I I '

-10 -9 -8 -7 -6 -5 -4


Log [WRC-0018], M






Figure 3-23. The effect of CCPA pretreatment on WRC-0018stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 0.1 IM CCPA
for 16 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. Cells were then incubated in HBSS containing 50 M rolipram and the indicated concentrations of WRC-0018 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The dose response curve of WRC-0018 in control cells is taken from Figure 3-4. The basal level of cAMP accumulated was 31 and 61 pmol/mg protein/min for the control and CCPA experiments, respectively. Each data point is the meanSE,
n=3.






89






20
20 A CONTROL

A CCPA


15



0'0
L 10
cE E





0 II



-9 -8 -7 -6 -5 -4


Log [NECA], M





Figure 3-24. The effect of CCPA pretreatment on NECAstimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 0.1 M CCPA for 16 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. Cells were then incubated in HBSS containing 50 gM rolipram and the indicated concentrations of NECA for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The data for NECA in control cells are taken from Figure 3-13. The basal level of cAMP accumulated was 32 and 111 pmol/mg protein/min for the control and CCPA experiments, respectively. Each data point is the meanSD of quadruplicate determinations and is representative of 2 experiments.






90







15- 0 CONTROL


1*- 0 CCPA




=Eo




5






-10 -9 -8 -7 -6 -5


Log [Adenosine], M







Figure 3-25. The effect of CCPA pretreatment on ADOstimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 0.1 LM CCPA for 16 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. Cells were then incubated in HBSS containing 50 .LM rolipram, 1 lM DIP, 1 M EHNA and the indicated concentrations of ADO
for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The control data are taken from Figure 3-14. The basal level of cAMP accumulated was 60 and 73 pmol/mg protein/min for the control and CCPA experiments, respectively. Each data point is the meanSE, n=3.






91






100



0) 80



E 60


O 40



20 O DIP+EHNA
ADO+DIP+EHNA

0 I I I
-11 -10 -9 -8 -7 -6 -5


Log [CPA], M






Figure 3-26. The effect of ADO pretreatment on the ability of CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media with 5 RM DIP, 1 jIM EHNA, without or with 100 pM ADO for 24 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. The cells were then incubated in HBSS containing 50 gM rolipram and 10 gM ISO or ISO plus the indicated concentrations of CPA for 7 min at 360C. The cAMP accumulated was determined as described in the "Methods" section. Each data point is the meanSE, n=5.






92






20 0 CONTROL

E O0 DIP+EHNA

15 ADO+DIP+EHNA




0 10


oE

< 0 5,




0
-9 -8 -7 -6 -5


Log [WRC-0018], M






Figure 3-27. The effect of ADO pretreatment on WRC-0018stimulated cAMP accumulation in DDT cells. Cells were incubated in fresh growth media without or with 5 JIM DIP + 1 gM EHNA or 5 p.M DIP + 1 .M EHNA + 100 pM ADO for 24 hr at 370C. At the end of the incubation period, the cells were washed 4 times with ice-cold HBSS and detached. Cells were then incubated in HBSS containing 50 iM rolipram and the indicated concentrations of WRC-0018 for 7 min at 360C. At the end of the incubation period, the cAMP accumulated was determined as described in the "Methods" section. The basal level of cAMP accumulated was 32, 31, 91 pmol/mg protein/min for the control, DIP+ENHA and ADO+DIP+EHNA experiments, respectively. Each data point is the meanSE, n=3.




Full Text
36
1985). In most if not all of these examples, the inhibitory
effect cannot be overcome even when the receptor is maximally
activated.
The DDT smooth muscle tumor cell line is derived from a
steroid-induced leiomyosarcoma of the vas deferens of an
adult Syrian hamster (Norris and Kokler, 1974). This cell
line has been shown to express a relatively high density of
12~ARs which mediate a robust stimulation of cAMP formation
(Norris et al., 1983). The DDT cells has been used therefore
as a model to study this receptor system. These cells have
also been shown to express functional histamine Hi
(Mitsuhashi and Payan, 1988) and steroid receptors (Norris
and Kohler, 1977) Recently, the presence of both Ai~ and A2-
AdoR have been demonstrated in DDT cells by radioligand
binding, photoaffinity labeling and their ability to alter AC
activity (Ramkumar et al., 1990). In keeping with the
classical definition of these two subtypes, the Ai-AdoR in
DDT cells inhibits AC activity whereas the A2~AdoR stimulates
the enzyme. The presence of both AdoR subtypes on a single
cell coupled to the same second messenger system provides a
unique opportunity to characterize pharmacologically if an
interaction between the Ai~ and A2~AdoR occurs. Assuming that
an Ai-AdoR inhibitory response would mask any A2~AdoR mediated
stimulation of cAMP accumulation, three experimental
approaches were used to attempt to alter Ai-AdoR
responsiveness and allow the expression of the A2~AdoR
response. These three approaches are shown diagrammatically


59
regulate the Ai-AdoR, the EC50 value for the NECA inhibitory
response was increased from 7.2 nM in control cells to 120 nM
in CCPA-pretreated cells which was similar to the EC50 value
for stimulation (180 nM) (see the table above). That is, the
EC50 value for NECA-induced cAMP accumulation (A2~AdoR effect)
is still greater than NECA's EC50 value for inhibition of cAMP
accumulation (Ai-AdoR effect) Similar EC50 values for NECA to
inhibit and stimulate cAMP accumulation in cells chronically
pretreated with CCPA indicates that this agonist would occupy
both receptor subtypes, but activation of the Ai-AdoR would
prevent expression of an A2~AdoR response. It appears that
the expression of an A2~AdoR response in untreated cells
requires a selective A2~AdoR agonist. This was evident with
WRC-0018 which based upon its stimulatory and inhibitory
concentration ranges, showed an approximately 500-600 -fold
selectivity for the A2~AdoR (Figure 3-4) Although chronic
CCPA-pretreatment reduced the sensitivity for the Ai-AdoR
response of NECA, this reduction was not sufficient to result
in activation of the A2~AdoR without significant stimulation
of the Ai-AdoR. A similar explanation may account for the
lack of an ADO mediated stimulation of cAMP accumulation in
cells chronically pretreated with CCPA. The EC50 for ADO to
inhibit and stimulate (in PTX-pretreated cells) cAMP
accumulation was 97 nM and >1 |1M, respectively. This Ai-AdoR
selectivity for ADO is in contrast to previous studies in
other cell types where this agonist has been shown to be
nonselective (Londos et al., 1980; Olsson and Pearson, 1990).


96
al., 1989). In addition, C-Br bound to the receptor in an
irreversible manner and produced an antagonist-insensitive
activation of AC activity. This suggested that C-Br is an
irreversible agonist at the BAR.
DDT cells have been shown to express both BAR and Ai~
AdoR. As shown in Chapter 3 of this dissertation and other
reports (Ramkumar et al., 1990; Nordstedt and Fredholm, 1990;
Ramkumar et al., 1991), BAR mediated cAMP accumulation in DDT
cells is inhibited by Ai-AdoR agonists. To obtain additional
insight into the hypothesis that inhibition is mediated at
the stimulatory receptor level, the ability of the Ai-AdoR to
attenuate the cAMP response mediated by a permanently
activated BAR was investigated.
Results
Ai-AdoR Mediated Inhibitory Effects on BAR-stimulated cAMP
Accumulation by ISO and C-Br
Initial experiments were designed to determine the
effect of CPA on ISO- and C-Br-stimulated cAMP accumulation
in intact DDT cells. As shown in Figure 4-1, ISO and C-Br
stimulated cAMP accumulation in a concentration-dependent
manner with maximal effects of 29- and 32-fold above the
basal level, respectively. The maximal responses to ISO and
C-Br could not be readily compared because experiments were
performed on different batches of cells. The EC50 value for C-
Br was 0.2 nM and 2.7 nM for ISO. Both ISO and C-Br


88
Log [WRC-0018], M
Figure 3-23. The effect of CCPA pretreatment on WRC-0018-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 0.1 (1M CCPA
for 16 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 (1M rolipram
and the indicated concentrations of WRC-0018 for 7 min at
36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The dose response curve of WRC-0018 in control cells
is taken from Figure 3-4. The basal level of cAMP accumulated
was 31 and 61 pmol/mg protein/min for the control and CCPA
experiments, respectively. Each data point is the meanSE,
n=3.


7
the uptake and release of adenosine with identical kinetics
(see references in Olsson and Pearson, 1990) .
Adenosine is metabolized very rapidly in the blood with
a half-life of 0.6 10 sec. The principal route of
metabolism is deamination to inosine by adenosine deaminase
and further degradation of inosine to hypoxanthine, xanthine
and eventually to uric acid. Adenosine deaminase can be
inhibited by erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA) and
2'-deoxycoformycin (see references in Bruns, 1990) In
addition, adenosine can reenter the nucleotide pool by
phosphorylation to adenine nucleotides (See references in
Olsson and Pearson, 1990).
Adenosine Receptors
Glassification and Charaterization
Adenosine receptors comprise a group of cell surface
receptors that mediate the physiological and pharmacological
effects of the nucleoside adenosine. At least two distinct
subtypes of cell surface AdoRs are responsible for these
actions. These receptors have been classified as Ai-AdoRs and
A2-AdoRs based on biochemical and pharmacological criteria,
i.e., modulation of adenylyl cyclase (AC) and differential
selectivity for a series of adenosine analogs. The Ai~AdoR
that mediates the inhibition of AC has an agonist potency
series of R-phenylisopropyladenosine ((R)-PIA) > 5'-N-
ethylcarboxamide adenosine (NECA) > (S)-PIA. The A2~AdoR that


52
accumulation. Several lines of evidence support the high
selectivity of N-0861 for Ai-AdoR. These include 1) a
concentration dependent displacement of specific [3H]CPX
binding at the Ai-AdoR site by N-0861 with a Hill slope of 1,
indicating the interaction of this antagonist with a single
class of binding sites, 2) the substantial attenuation of the
CPA mediated inhibition of cAMP accumulation by 10 (1M N-0861,
and 3) the unaltered stimulatory effect of WRC-0018 by N-0861
at concentrations ranging from 0.1 nM 10 (1M.
In the presence of N-0861, the stimulatory response of
WRC-0018 was sustained, i.e., the downward portion of the
biphasic response was abolished. This finding is consistent
with the hypothesis that the downward portion of the WRC-0018
biphasic response was Ai-AdoR mediated. Assuming that the
interaction of N-0861 with the Ai-AdoR and the interaction of
WRC-0018 with Ai~ and A2~AdoR are reversible and competitive,
then in the presence of N-0861, the downward portion of WRC-
0018 concentration response should be shifted to the right.
This was not tested in the present study due to the
insolubility of WRC-0018 to obtain and test concentrations
above 100 |1M. Similar to the results with WRC-0018, blockade
of the Ai-AdoR with N-0861 uncovered a cAMP stimulation
response of the agonists ECA (<100 (1M) and ADO (<5 (1M) .
These results are in keeping with the observation on NECA
mediated A2~AdoR stimulatory response on AC activity reported
by Ramkumar et al. (1990). However, at >10 JIM NECA (Figure 3-
13) and 10 JIM ADO (Figure 3-14), there was a decrease in cAMP


22
accompanied by a decrease in the number of Ai-AdoRs which can
form a high affinity agonist binding site and a 3-4 fold
increase in the phosphorylation of Ai-AdoR (Ramkumar et al.,
1991) Thus, similar to the IbAR system, Ai-AdoR can also
undergo desensitization and/or down-regulation after chronic
exposure to an agonist. Uncoupling, down-regulation and
phosphorylation of the Ai-AdoR may contribute to the
desensitization of this inhibitory receptor.
Desensitization of A2~AdoR-AC system has also been
described. Using clonal neuronal cells (NG108-15), which
express both A2~AdoR and prostaglandin E (PGEi) receptors, PGE
pretreatment reduced the effects of both PGEi and adenosine
to activate AC (heterologous desensitization) In contrast,
exposure of NG108-15 to 2-chloroadenosine resulted in a rapid
loss of response to 2-chloroadenosine (homologous
desensitization), but PGEi-stimulated AC activity decreased
only slightly (Kenimer and Nirenberg, 1981) .
Adenosine receptors are regulated during chronic drug
treatment with AdoR antagonist or dexamethasone. A recent
study showed that exposure of guinea pig myocardium to AdoR
antagonist theophylline increased the the number of Ai-AdoR
(Wu et al., 1989) In humans, after 7 days of caffeine (AdoR
antagonist) abstinence, NECA produced a concentration-
dependent inhibition of thrombin-induced platelet aggregation
with an EC50 value of 69 nM (Biaggioni et al., 1991b).
Subjects were then given caffeine 250 mg p.o. 3 times a day
for 7 days. Caffeine withdrawal significantly shifted the


31
Protein Measurement
The protein content of cells and membranes was
determined by the method of Lowry et al.(1951) using bovine
serum albumin as standard.
Radioligand Binding Assay
Ai-AdoRs in DDT cells were determined by specific [3H]CPX
binding. Membrane protein was initially incubated with 2 U/ml
(1 U = 6.25 Jig) ADA for 20 min at 4C to metabolize
endogenous adenosine. Cell membranes (=0.1 mg protein) were
then incubated in a total volume of 0.2 ml with 50 mM Tris-
HC1 buffer at pH 7.4, 5 mM MgCl2 and 0.06-4 nM [3H]CPX, with
or without 50 |1M (R)-PIA, for 150 min at room temperature on
an orbital shaker. The bound and free ligand were rapidly
separated on GF/C glass fiber filters (Whatman Inc., Clifton,
NJ, USA) using a Brandel Cell Harvester (Brandel Scientific,
Gaithersburg, MD, USA). Filters were rinsed three times with
4 ml of ice-cold 50 mM Tris-HCl buffer containing 10 mM MgCl2
and 0.1 % CHAPS (to reduce non-specific binding). The filters
were placed in standard scintillation vials with 10 ml of
Liquiscint and the radioactivity was determined in a liquid
scintillation counter. Specific binding to Ai-AdoR was
calculated as the difference between the total binding in the
absence of (R)-PIA and the nonspecific binding in the
presence of 50 (1M (R)-PIA. Specific binding was generally 90-


BIOGRAPHICAL SKETCH
Fan Xie was born on June 2 6, 1965, in Changsha, Hunan,
People's Republic of China. She was accepted into the first
English Medical Class of Hunan Medical University in 1981 and
received a medical doctor degree in 1987. After graduation,
she came to the United States to further her education. In
the fall of 1987, she started her graduate studies under the
guidance of Dr. Stephen P. Baker in the Department of
Pharmacology and Therapeutics at the University of Florida.
After completing the requirements for the degree of Doctor of
Philosophy, she plans to do a postdoctoral fellowship at
Hoffmann-La Roche Inc.
121


51
explain two observations. First, the assumed nonselective
agonists NECA and ADO showed no stimulation of cAMP
accumulation. At effective concentrations of either agonist,
both AdoR subtypes would be activated, but the Ai-AdoR
mediated inhibition would predominate and hence inhibit the
A2~AdoR mediated stimulation of AC and thereby accumulation
of cAMP. However, this finding is in contrast to the
observation reported by Ramkumar et al. (1990) NECA (10 JIM) ,
in absence of any Ai-AdoR antagonist, produced a 2.1 fold
stimulation of AC activity over the basal level in DDT cell
membranes (Ramkumar et al., 1990). The discrepancy between
this report and our data may be due to the homogenization
process during membrane preparation affecting the function of
Ai-AdoR. The second observation which could be explained by
the dominant role of Ai-AdoR is the downward portion of the
WRC-0018 biphasic concentration response may reflect the
activation of the Ai-AdoR. This was indicated by the
inhibitory effect of WRC-0018 on ISO-stimulated cAMP
accumulation over the same concentration range where the
downward part of the biphasic response of WRC-0018 occurred
(Figure 3-7).
Three approaches were used to study the expression of
A2~AdoR mediated responses and further establish that the
downward phase of the WRC biphasic response is Ai-AdoR
mediated. The first approach involved the use of the
selective Ai-AdoR antagonist N-0861. The selectivity of N-
0861 for the Ai-AdoR was tested by receptor binding and cAMP


57
partial agonists (Pittman et al., 1984). Alternative
explanations may also be possible. For example, there may be
subpopulations of Ai-AdoRs in DDT cells where CPA is
nonselective and WRC-0018 is selective for only one subtype
which was selectively down-regulated and desensitized by
chronic CCPA-pretreatment. Further experiments will be
necessary to determine the reason for the differential
changes between CPA and WRC-0018 responsiveness after a
decrease in receptor number.
Because CCPA-pretreatment had little effect on the A2-
AdoR mediated stimulation component of the WRC-0018
concentration response (Figure 3-23), this suggests that CCPA
has little or no desensitization effect on the A2~AdoR. On
the other hand, chronic CCPA-pretreatment abolished the
downward phase of the WRC-0018 biphasic concentration
response curve (Figure 3-23). CCPA is a highly selective Ai~
AdoR agonist (Lohse et al., 1988). The loss of the downward
phase of the WRC-0018 response in conjunction with a
desensitization and/or down-regulation of Ai-AdoR is
consistent with the downward phase being Ai~AdoR mediated.
Thus, the elimination of the downward phase may be due to 1)
desensitization and/or down-regulation decreasing the
sensitivity to Ai-AdoR agonists thereby further increasing
the A2~AdoR selectivity of WRC-0018, and 2) the loss of
efficacy of WRC-0018 as an Ai-AdoR agonist in the partially
desensitized state.


15
agonist affinity and increases the dissociation rate (Gilman,
1987) .
Several lines of evidence suggest that the AdoR belongs
to the class of G-protein-coupled receptors. For Ai-AdoR, GTP
and stable GTP analogues decreased the apparent affinity of
Ai~AdoR agonists to the receptor in membranes (Goodman et
al., 1982; Yeung and Green, 1983; Lohse et al. 1984),
solubilized receptors (Gavish et al., 1982; Stiles, 1985;
Stroher et al., 1989) and slices of brain tissue (Fastbom and
Fredholm, 1990). Treatment with N-ethylmaleimide which
uncouples receptors from G proteins inhibited the action of
Ai-AdoR agonists without affecting the binding of antagonists
(Fredholm et al., 1985). In addition, detergent-solubilized
Ai_AdoR co-eluted with a G protein from an agonist affinity
chromatography column where GTP or N-ethylmaleimide, agents
known to uncouple receptors from G proteins and
simultaneously lower the affinity of agonists for the
receptor, were used (Munshi and Linden, 1989; Linden, 1991).
It was suggested that Ai-AdoR, unlike other G protein coupled
receptors, tightly binds to G protein. The coelution did not
occur when an antagonist affinity column was used (Nakata,
1989a, b) .
For the A2~AdoR, experiments showed a GTP dependence for
NECA stimulation of AC in purified hepatic plasma membranes
(Cooper and Londos, 1979). AC activity from other tissues has
been shown to be enhanced by or dependent on the presence of
a guanine nucleotide (Londos et al., 1979; Fain and Malbon,


4
et al., 1991a). In the nervous system, adenosine produces
hyperpolarization of neurons resulting in decreased nerve
firing. Adenosine also inhibits neurotransmitter release
through putative presynaptic inhibitory receptors, both in
the brain and in the periphery nervous system. Adenosine
inhibits the release of practically all neurotransmitters
studied, including norepinephrine, acetylcholine, dopamine,
glutamate, aspartate, y-aminobutyric acid and serotonin.
Adenosine also has a central depressor action and has been
proposed as an endogenous anticonvulsant (see references in
Biaggioni et al., 1991a). In fat cells, adenosine abolishes
the breakdown of stored triglycerides to free fatty acids and
glycerol (lipolysis) which is induced by adrenergic
stimulation. Adenosine can also prevent platelet aggregation
(Berne, 1986; Pelleg and Porter, 1990)
Present and Future Therapeutic Uses
Based on its negative dromotropic effect on AV nodal
conduction, adenosine was recently approved by the U.S. Food
and Drug Administration as an antiarrhythmic drug for the
acute management of paroxysmal supraventricular tachycardia
involving the AV node (Belardinelli and Lerman, 1990). In
addition, the transient AV block caused by adenosine can also
be used to unmask underlying atrial activity in other forms
of atrial arrhythmias and hence help in the differential
diagnosis of arrhythmias (Belardinelli and Lerman, 1990).


58
In contrast to the effects of N-0861 and PTX, chronic
CCPA-pretreatment did not uncover any NECA or ADO stimulation
of cAMP accumulation. This observation can be explained by
the differences in the agonist's sensitivity to produce
inhibition or
stimulation
of cAMP
accumulation.
Summary
of EC50
Values (nM)
Agonists
Ai-AdoR
A2~AdoR
Control
CCPA
(in PTX-pretreated cells)
WRC-0018
830
-
90
NECA
7.2
120
180
ADO
97
-
>1000
CPA
1.2
15
-
In the case of NECA, the estimated EC50 values for the
inhibition (Ai-AdoR effect) and stimulation (A2~AdoR effect)
of cAMP accumulation in untreated cells were 7.2 and 180 nM,
respectively. The latter estimate was derived from the NECA
concentration response curve after PTX-pretreatment and this
value was based on the assumption that PTX-pretreatment had
little or no effect on the potency of this agonist to
stimulate cAMP accumulation. The higher potency of NECA to
inhibit cAMP accumulation indicates that this agonist shows
some degree of selectivity for the Ai-AdoR over the A2~AdoR in
DDT cells. In contrast, NECA has been reported to be a
nonselective agonist in other systems (Hutchinson et al,
1990). After chronic CCPA-pretreatment to partially down-


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
./77liLil lLx.
Stephen P. Baker, Ph.D., Chairman
Professor of Pharmacology and
Therapeutics
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of poctor of Philosophy.
Les c c/ 1 i- C
4-
Luiz- Belrdinelli, M.EK
Professor of Pharmacology and
Therapeutics
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
rX-Tiy- "
Edwin Meyer^Jph.D.
Associate Professor of
Pharmacology and Therapeutics
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and guality, as
a dissertation for the degree of Doctor of' 'Philosophy^
/
Thomas F. Rowe, Ph.D.
Associate Professor of
Pharmacology and Therapeutics


120
Trivedi, B. K., Bridges, A. J. and Bruns, R. F. (1990)
Adenosine and Adenosine Receptors, ed. by Williams, M.,
Clifton, New Jersey: The Humana Press Inc. pp 57-103
Trost, T. and Schwabe, U. (1981) Mol. Pharmacol. 19: 228-235
Trussel, L. 0. and Jackson, H. B. (1985) Proc. Natl. Acad.
Sci. USA 82: 4857-4861
Ueeda, M., Thompson, R. D., Arroyo, L. H. and Olsson, R. A.
(1991) J. Med. Chem. 34(4): 1340-1344
Ullman, A. and Svedmyr, N. (1988) Thorax 43: 674-678
Van Calker, D., Miller, M. and Hamprecht, B. (1979) J.
Neurochem. 33: 999-1005
Watanabe, A. M., McConnaughey, M. M., Strawbridge, R. A.,
Fleming, J. W., Jones, L. R. and Besch, H. R. (1978) J. Biol.
Chem. 253: 4833-4836
Wesley, R. C. and Belardinelli, L. (1989) Circulation 80:
128-137
Williams, M. and Risley, E. A. (1980) Proc. Natl. Acad. Sci.
USA 77: 6892-6896
Wolff, J., Londos, C. and Cooper, D. M. F. (1981) Adv. Cyclic
Nucleotide 14: 199-214
Wu, S. N., Linden, J., Visentin, S., Boykin, M. and
Belardinelli, L. (1989) Circ. Res. 65(4): 1066-1076
Xu, D., Kong, H. and Liang, B. T. (1992) Circ. Res. 70(1):
56-65
Yeung, S. M. and Green, R. D. (1983) J. Biol. Chem. 258:
2334-2339


16
1979; Londos, et al., 1981; Wolff et al., 1981). Moreover,
adenosine shortened the lag period for the onset of AC
activation by Gpp(NH)p (Sevilla et al., 1977; Lad et al.,
1980) .
The interaction of an activated receptor with G protein
is a key step in signal transduction. The regulatory G
proteins are heterotrimers with subunits designated a, 5 and y
in order of decreasing mass. The functional difference
between Gs and Gi resides in the respective as and Cti
subunits. The structural features common to 0CS and Oti include
a GTPase activity and regions that recognize and couple these
proteins to a receptor, to the I5+y complex, and to the
effector systems. Common 15 and y subunits are functionally
interchangable (Gilman, 1987). G proteins cycle between an
inactive GDP state and an active GTP state. When GDP is
bound, the a subunit associates with the 15 and y subunits to
form a Gofiy complex (denoted by G-GDP) that is membrane-bound.
When GTP is bound, the a subunit (Ga-GTP) dissociates from
the 15 and y subunits (Guy) Ga-GTP released from G[jy then
alters the activity of the target, such as AC or an ion
channel (Stryer and Bourne, 1986).
Another feature of G proteins is that the cysteine
residue in their a subunits is a substrate for adenosine
diphosphate (ADP)-ribosylation which transfers an ADP-ribose
moiety from nicotinamide-adenine dinucleotide (NAD). Gs is
selectively ADP-ribosylated by cholera toxin which in turn
inhibits the receptor-stimulated GTPase activity of the G


75
T3.E
g E
3
E £
3 o
8a
< U)
CL E
<
o c
60
40
20
0
-10 -9 -8 -7 -6 -5
Log [N-0861], M
Figure 3-10. The effect of N-0861 on the ability of NECA to
inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells
were incubated in HBSS containing 50 p.M rolipram, 10 )1M ISO,
without or with 1 |1M NECA or NECA plus the indicated
concentrations of N-0861 for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSD of quadruplicate determination. The indicates a
p<0.0005 for the ISO + NECA + N0861 point from the ISO +
NECA point.


80
"o£
se
11
ES
3 O
8 o.
< O)
o. E
o E
o.
Figure 3-15. The effect of PTX pretreatment on the ability of
CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells.
Cells were incubated in fresh growth media with the indicated
concentrations of PTX for 18 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached from the plate. The cells were then
incubated in HBSS containing 50 (1M rolipram, without or with
10 |1M ISO or ISO plus 1 [IM CPA for 7 min at 36C. At the end
of the incubation period, the cAMP accumulated was determined
as described in the "Methods" section. Each data point is the
meanSD of quadruplicate determination.


18
agonist, the stimulatory receptor and Gs. This results in the
inability of the receptor to form a high affinity binding
state. Experiments on rat ventricular myocyte membranes
showed that PIA inhibited isoproterenol (ISO)-stimulated AC
activity (Romano et al., 1988; Romano et al., 1989). This
inhibition was antagonized by theophylline. PIA was much less
effective at attenuating forskolin-stimulated AC activity and
had no effect on 5'-guanyl-imidodiphosphate (Gpp(NH)p)-
induced stimulation. In [125I]cyanopindolol (CYP)/ISO
competition binding experiments, ISO produced a
concentration-dependent displacement of specific [125I]CYP
binding with an IC50 of 48 nM and Hill slope of 0.6. About 38%
of BARs were in the high affinity state. Gpp(NH)p shifted the
competition curve to the right (IC50 = 520 nM) and steepened
the slope (Hill slope = 1.2) indicating that all of the BARs
were in low affinity state. PIA significantly increased the
IC50 for ISO in the absence of Gpp(NH)p (IC50 = 140 nM) and
steepened the slope (Hill slope = 0.9). These findings were
interpreted to indicate that binding of ISO to the high
affinity state of the BAR was decreased in the presence of
PIA. PIA had no effect on the ISO competition curve in the
presence of Gpp(NH)p (Romano et al., 1988; Romano et al.,
1989).
The second is the subunit dissociation model. This model
is based on the finding that both Gs and Gi share two common
subunits, lb and y, and that (Xi is present in excess relative
to (X3 in most cells. Activation of Gi leads to subunit


74
-O £
i*
If
§1
< O
OL E
<1
o c
Q.
50
40
30
20
10
0
-10
-8
-6
Log [N-0861], M
Figure 3-9. The effect of N-0861 on the ability of CPA to
inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells
were incubated in HBSS containing 50 JIM rolipram, 10 }1M ISO,
without or with 0.1 |1M CPA or CPA plus the indicated
concentrations of N-0861 for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSD of quadruplicate determinations. The indicates a
p<0.0005 for the ISO + CPA + N-0861 value compared to the ISO
+ CPA value.


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as.
a dissertation for the degree of Doctor of PhilostSjkhy.
>han K. Raizada, Ph.D.
rofessor of Physiology
This dissertation was submitted to the Graduate Faculty
of the College of Medicine and to the Graduate School and was
accepted as partial fulfillment
degree of Doctor of Philosophy.
May, 1992
the requirements for the
4
Dean, Graduate School


102
indicates that even though C-Br produced an antagonist-
insensitive response once bound to the BAR, activation of the
inhibitory Ai-AdoR resulted in an attenuation of cAMP
accumulation mediated by the permanently activated BAR.
There is some evidence suggesting that one potential
mechanism for receptor mediated inhibition of AC activity may
involve signal transduction at the stimulatory receptor
level. In 1978, Watanabe et al. reported that activation of
the cardiac muscarinic inhibitory receptor decreased the
affinity of ISO for the BAR. Likewise, more recently, Romano
et al. (1988; 1989) showed the Ai-AdoR agonist PIA reduced
the affinity of ISO for the BAR in rat cardiac membranes.
This reduction in agonist affinity was similar to that
observed in the presence of Gpp(NH)p (Romano et al., 1988;
Romano et al., 1989). Because the guanine nucleotide
sensitive agonist high affinity binding state of the receptor
has been proposed to represent the ternary complex formation
(Gilman, 1987), it was suggested that PIA prevented the
stimulatory receptor from forming a ternary complex, a
necessary prerequisite for the activation of AC. However, in
the present study, CPA did not alter the interaction of ISO
with the BAR even though Gpp(NH)p did reduce agonist
affinity. This indicated that CPA had no effect on ternary
complex formation. The discrepancy between this observation
and that reported by Romano et al. (1988) may be due to
different cell types expressing different modes of
inhibition. The inhibition of ternary complex formation may


given to Judy Adams and Barbara Reichert for being so nice to
me no matter how many times I bothered them.
xv


49
Di .sens s ion
The interaction of the inhibitory Ai-AdoR and the
stimulatory A2~AdoR was investigated using DDT cells. The co
expression of both AdoR subtypes was first shown in this cell
line by Ramkumar et al. (1990). In addition, UARs which
mediate the stimulation of cAMP accumulation are also
expressed in DDT cells (Norris et al., 1983) .
In the presence of a constant concentration of ISO which
increased cAMP accumulation, several Ai-AdoR agonists
inhibited the ISO stimulatory effect in a concentration-
dependent manner. Furthermore, in the presence of a fixed
concentration of Ai-AdoR agonists, the effect of ISO on cAMP
accumulation was greatly attenuated. The inhibitory effect of
CPA on ISO-stimulated cAMP formation was blocked by 1 )iM CPX,
a selective Ai-AdoR antagonist (data not shown). These
findings indicate that CPA exerts its inhibitory effect on
BAR mediated cAMP accumulation by activating Ai~AdoRs which
are known to be negatively coupled to AC. This Ai-AdoR
mediated inhibition on cAMP formation confirms and extends
the recent reports on AdoRs in DDT cells (Ramkumar et al.,
1990; Gerwins et al., 1990; Gerwins and Fredholm, 1991). One
difference between the present data and the previous reports
is that NECA produced a 97% maximal inhibition of cAMP
accumulation in intact DDT cells in our experiments. However,
only a 30% maximal inhibition of AC activity in DDT cell
membranes by NECA was reported by Ramkumar et al.(1990). The


17
protein causing GTP to be bound to as for a prolonged period.
This results in a permanent activation of AC and a large
increase in cAMP accumulation (Gilman, 1987; Nathanson,
1987). Pertussis toxin (PTX) selectively ADP-ribosylates Gi
(and another guanine nucleotide binding protein, G0) This
covalent modification of Gi inactivates the protein resulting
in the loss of receptor mediated inhibition of AC activity
(Gilman, 1987; Nathanson, 1987). Thus, these bacterial toxins
have been used as an investigative tool to determine the role
of G proteins in the action of biologic messengers. For
example, studies showed that PTX blocked the ability of Ai~
AdoR to inhibit AC (Hazeki and Ui, 1981) and prevented
adenosine-induced changes in the rate of beating in rat atria
(Endoh et al., 1983).
Mechanisms for Inhibitory Receptor Action
Receptors that mediate the attenuation of AC activity
include the Ai-AdoR, muscarinic M2-acetylcholine receptor, 0C2-
adrenergic receptor, 5 opiate receptor and D2~dopamine
receptor.
Compared to stimulatory receptors, much less is known
about the mechanisms whereby inhibitory receptors attenuate
the activation of AC by stimulatory receptors. Currently,
there is experimental evidence supporting three different
models. First, inhibition may occur by preventing the
formation of the ternary complex which is composed of the


48
cAMP content in a concentration-dependent manner with an EC50
value of 2.0 nM and maximal inhibition of 77%. In comparison,
after pretreatment of the cells with ADO, the EC50 value for
CPA-induced inhibition on ISO-stimulated cAMP accumulation
was increased by 15-fold (EC50, 30 nM) without any significant
change in the maximal inhibition (pretreated, 73%).
Desensitization of A2~AdoR
The effect of ADO on the desensitization of A2~AdoR were
investigated by studying the effect of WRC-0018 on cAMP
accumulation (Figure 3-27). In control and DIP+EHNA-
pretreated cells, WRC-0018 produced a biphasic concentration
response curve. At low concentrations (1 500 nM), WRC-0018
stimulated cAMP accumulation with a maximal increase of 4-
fold above the basal level. At higher concentrations (>500
nM) of WRC-0018, the cAMP accumulation was attenuated. After
pretreatment of the cells with ADO, the basal level of cAMP
was significantly increased (p<0.025) from 31 to 91 pmol
cAMP/mg protein/min. However, WRC-0018 did not stimulate cAMP
accumulation above the basal level.
The effect of ADO on cAMP accumulation after
pretreatment of the cells with ADO is shown in Figure 3-28.
In both control or ADO-pretreated cells, ADO did not
stimulate cAMP accumulation above the basal level.


11
[3H]NECA has high affinity for A2a-AdoR, it has been used as a
radioligand for this receptor. However, NECA also binds to
Ai~AdoR with high affinity. Thus, when [3H]NECA is used, it is
necessary to block the Ai-AdoR by adding a highly selective
Ai-AdoR ligand such as the agonist CPA or the antagonist CPX
(Hutchison et al., 1989; Linden, 1991). Another A2~AdoR
agonist radioligand recently synthesized is 2[4(2{[4
aminophenyl]methylcarbonyl}ethyl)phenyl]ethylamino-5'-N-
ethylcarboxamido adenosine ( [125I]PAPA-APEC) (Ramkumar et al.,
1990). Likewise, 2-[p-(2-carboxyethyl)phenethylamino]-5'-N-
ethyl-carboxamido adenosine (CGS 21680), an agonist with high
affinity and selectivity for A2~ over Ai-AdoR has been
synthesized (Hutchison et al., 1989; Lupica et al., 1990).
Specific binding of the newly synthesized [3H]CGS 21680 to
rat striatal membranes was saturable and reversible.
Saturation studies revealed that [3H]CGS 21680 binds with
high affinity (Kd=16 nM) to a single class of binding sites.
Adenosine agonists competed for the binding of [3H]CGS 21680
with the following potency order: CGS 21680 > NECA > (R)-PIA
> (S)-PIA. The specific binding of [3H]CGS 21680 was greatest
in rat striatal membranes but negligible in rat cortical
membranes. These results indicate that [3H]CGS 21680 directly
labels the high affinity A2a-AdoR in rat brain without the
need to block binding to Ai-AdoRs (Jarvis et al., 1989;
Jarvis and Williams, 1989).
Ai-AdoRs of brain, heart, or fat cells, when labeled
with photoaffinity ligands or by means of photoaffinity


44
of 1 |1M 10 |1M, ADO increased the cAMP level in PTX-
pretreated cells. A maximal response was not achieved even at
10 |1M ADO and hence, the EC50 value of the A2~AdoR mediated
response of ADO could not be calculated.
In summary, these data show that following PTX-
pretreatment, the selective A2~AdoR agonist WRC-0018 causes a
sustained A2~AdoR mediated stimulatory effect on cAMP
accumulation. Likewise, the A2~AdoR mediated stimulatory
effects of the nonselective agonists NECA and ADO were
unmasked after pretreatment of DDT cells with PTX. The
results also indicate that the Ai inhibitory effect of AdoR
agonists was Gi protein mediated.
Effect of Desensitization and/or Down-regulation of Aj_-AdoR
on the A-AdoR Mediated Response
The third experimental approach used to uncover the A2-
AdoR mediated effect on cAMP accumulation involves
desensitization and/or down-regulation of Ai~AdoR. Figure 3-
19 illustrates a representative Scatchard plot of [3H]CPX
binding to cell membranes from control (untreated) and CCPA
(Ai-AdoR agonist)-pretreated cells. The inhibition of ISO (10
fiM) -stimulated cAMP accumulation by 1 (1M CPA was investigated
in cells pretreated with various concentrations of CCPA (0.1
nM 1 |j.M) for 16 hr (data not shown) Cells incubated with
0.1 HM CCPA for 16 hr at 37C followed by four wash cycles
showed a 48% reduction in specific [3H]CPX binding (control,
0.4 pmol/mg protein) with no change in the Kd value for the


INTERACTION OF ADENOSINE RECEPTORS
IN A SMOOTH MUSCLE CELL LINE
BY
FAN XIE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1992

This dissertation is dedicated to my grandmother, father,
mother and brother.

ACKNOWLEDGEMENTS
First and foremost, I wish to sincerely thank my mentor,
Dr. Stephen Baker, for his guidance, both professional and
personal, and for the contribution he has made to my
education. I would also like to thank Dr. Luiz Belardinelli
for his enthusiasm, valuable suggestions and constant supply
of wonderful drugs. Many thanks are also expressed to my
helpful and friendly committee: Mohan Raizada, Edwin Meyer
and Thomas Rowe. I wish to thank all members of Dr. Baker's
and Dr. Belardinelli's laboratory, especially Debbie Otero,
Dr. John Shryock, Mary Anne Locksmith, Cheryl Spence and
Xingmin Tang for their support and technical assistance. I
also deeply thanks Drs. Allen Neims, David Silverman,
Chingkuang Tu and Thomas Muther for their confidence in me. A
special word of thanks goes to Dr. Sandra Rattray for her
editorial help. My best wishes are also extended to all my
fellow (and former fellow) graduate students especially
Sukanya Kanthawatana, Nelida Sjak-Shie, Walter Folger, Daniel
Danso, Jiahui Zhang and Magdalena Wozniak; thanks for their
friendship and encouragement. I would also like to thank
other faculty members who worked so hard to improve the
graduate program, and the secretarial and administrative
staff who keep the department running. Special thanks are
iii

given to Judy Adams and Barbara Reichert for being so nice to
me no matter how many times I bothered them.
xv

TABLE OF CONTENTS
ACKNOWLEDGMENTS iii
ABSTRACT vi
CHAPTERS
1 INTRODUCTION 1
Functions of Adenosine 1
Synthesis and Metabolism of Adenosine 5
Adenosine Receptors 7
Regulation of Adenosine Receptors 20
Goals 24
2 EXPERIMENTAL PROCEDURES
Source of Materials 27
Methods 2 8
3 INTERACTION OF ADENOSINE RECEPTORS
Introduction 35
Results 37
Discussion 49
Summary 62
4 INHIBITORY EFFECT OF Ai-ADENOSINE RECEPTOR ON
IRREVERSIBLE ACTIVATION OF THE A-ADRENORECEPTOR
Introduction 94
Results 96
Discussion 99
LIST OF REFERENCES 113
BIOGRAPHICAL SKETCH 121
v

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
INTERACTION OF ADENOSINE RECEPTORS
IN A SMOOTH MUSCLE CELL LINE
By
Fan Xie
May, 1992
Chairman: Stephen P. Baker, Ph.D.
Major Department: Pharmacology and Therapeutics
DDTi MF-2 cells have been shown to express inhibitory Ai
and stimulatory A2 adenosine receptors (AdoRs) coupled to 3',
5'-cyclic adenosine monophosphate (cAMP) accumulation. The
objective of this study was to investigate the possible
interaction between the two AdoRs. The AdoR agonists,
adenosine and 51-N-ethylcarboxamido-adenosine (NECA)
attenuated isoproterenol (ISO)-stimulated cAMP accumulation
in a dose-dependent manner with a maximal inhibition of 68%
and 98%, respectively. No cAMP stimulation was observed with
either compound. In contrast, the selective A2~AdoR agonist
2-[2-(2 naphthyl)ethoxy]adenosine (WRC-0018) produced a
*
biphasic response. Stimulation of cAMP accumulation (8-fold)
occurred at low concentrations (5 500 nM) followed by an
attenuation at high concentrations (>500 nM). The attenuation
component was prevented by 1) the selective Ai-AdoR
vi

antagonist () N6-endonorbornan-2-yl-9-methyladenine (N-0861,
10 (1M) 2) pretreatment of cells with pertussis toxin (PTX,
25 ng/ml, 18 hr) which uncoupled the inhibitory Ai-AdoR
response or 3) pretreatment of cells with the selective Ai-
AdoR agonist 2-chloro-N6-cyclopentyladenosine (CCPA, 0.1 JIM,
16 hr). CCPA-pretreatment reduced by 13-fold the potency of
the Ai-AdoR agonist N6-cyclopentyladenosine (CPA) to inhibit
ISO-stimulated cAMP formation, and decreased the Ai-AdoR
level by 48%. Stimulation of cAMP accumulation by adenosine
and NECA was uncovered in the presence of N-0861 and by PTX-
pretreatment. However, no stimulation by either agonist was
observed after CCPA-pretreatment. The data indicate that the
inhibitory Ai-AdoR response in DDTi MF-2 cells is predominant
and masks the A2~AdoR mediated stimulatory effect. The A2~AdoR
response was expressed by a selective A2~AdoR agonist or
under conditions where the function of the Ai-AdoR is
blocked.
The ability of the activated Ai-AdoR to modulate agonist
interaction with the beta-adrenoreceptor (BAR) was studied
using the irreversible BAR agonist, 5 [2 [ [3[4
(bromoacetamido)phenyl]-2-methylprop-2-yl]amino]-1-
hydroxyethyl]-8-hydroxycarbostyril (C-Br). Activation of the
Ai-AdoR attenuated cAMP accumulation of the permanently
stimulated BAR, and did not alter the irreversible binding of
C-Br. In addition, CPA decreased basal cAMP level and had no
effect on the interaction of the reversible BAR agonist ISO
with the BAR. These data indicate that Ai-AdoR inhibitory
Vll

effect is not mediated by alteration of agonist interaction
with the BAR but rather occurs via a post-receptor mechanism.
Vlll

CHAPTER 1
INTRODUCTION
Functions of Adenosine
History
Drury and Szent-Gyorgyi (1929) were the first to report
on the cardiovascular effects of adenosine and adenine
nucleotides. They described the isolation of crystalline
adenine from acid extracts of ox heart muscle. Adenosine was
crystalized from yeast nucleic acid hydrolysate. Intravenous
injection of either extract into different mamalian species
after atropinization produced primarily sinus bradycardia,
transient heart block and other physiological effects (Drury
and Szent-Gyorgyi, 1929).
Interestingly, Drury and Szent-Gyorgyi did not comment
on the physiological implications of their observations.
Within two years, however, Lindner and Rigler (1931) had
crystallized adenosine, obtained from the degradation of AMP,
from heart muscle extracts. They showed that adenosine was a
potent coronary vasodilator in a number of species. Based on
the findings that adenosine was present in heart muscle
extract and had potent vasoactive effects, Lindner and Rigler
advanced the hypothesis that adenosine is a physiological
1

2
regulator of coronary blood flow. This hypothesis received
little immediate attention and interest at the time.
It was not until 1963 that Berne (1963) and Gerlach et
al. (1963) independently revived Lindner and Rigler's
adenosine hypothesis by demonstrating the release of
adenosine catabolites from hypoxic or ischemic heart muscle.
The revived hypothesis states that an imbalance between
oxygen supply and oxygen demand leads to alterations of the
cellular release of adenosine, which in turn changes the
contractile state of vascular smooth muscle of the resistance
vessels.
In 1970, Sattin and Rail (1970) and Shimizu and Daly
(1970) proposed the existence of extracellular adenosine
receptors (AdoRs) based on the observation that adenosine and
some adenine nucleotides elevate adenosine 3', 5'-cyclic
monophosphate (cAMP) in cerebral cortical slices. This effect
was competitively inhibited by methylxanthines, caffeine and
theophylline. Within a decade, it became clear that there are
AdoRs that inhibit as well as others that stimulate adenylyl
cyclase (AC) (Londos and Wolff, 1977; Van Calker et al.,
1979), constituting a system for the bidirectional control of
the catalytic activity of this key enzyme.
Physiological Effects
Adenosine is present in every cell of the human body and
exerts a wide spectrum of effects on various tissues and

3
organs. For example, in the heart, adenosine is a potent
coronary vasodilator (Berne, 1963; Gerlach et al., 1963). It
has also been shown to depress cardiac activity, e.g., (1)
depress sinoatrial and atrioventricular (AV) node activity,
(2) reduce atrial contractility, (3) attenuate the
stimulatory actions of catecholamines primarily in
ventricular myocardium, and (4) depress ventricular
automaticity (Belardinelli et al., 1989). These actions
characterize adenosine as an endogenous cardioprotective
substance whose actions lead to an increase in oxygen supply
and decrease in cardiac work. Together, these actions tend to
restore the balance between oxygen supply and demand.
In most tissues including skeletal muscle, adenosine is
a vasodilator. Thus, intravenous infusion of adenosine causes
hypotension. However, the ultimate cardiovascular effects of
adenosine in vivo depends on the dose, rate and mode of
administration, and on the autonomic reflexes triggered as a
result of adenosine's direct action (Pelleg and Porter,
1990) .
In the kidney, adenosine produces vasoconstriction of
the afferent glomerular artery, a decrease in glomerular
filtration rate and inhibition of renin release (Pelleg and
Porter, 1990). Adenosine is a depressant of the respiratory
center and it causes bronchoconstriction (Pelleg and Porter,
1990). However, adenosine also stimulates arterial
chemoreceptors (Biaggioni et al., 1991a). Hence, when given
intravenously, adenosine causes hyperventilation (Biaggioni

4
et al., 1991a). In the nervous system, adenosine produces
hyperpolarization of neurons resulting in decreased nerve
firing. Adenosine also inhibits neurotransmitter release
through putative presynaptic inhibitory receptors, both in
the brain and in the periphery nervous system. Adenosine
inhibits the release of practically all neurotransmitters
studied, including norepinephrine, acetylcholine, dopamine,
glutamate, aspartate, y-aminobutyric acid and serotonin.
Adenosine also has a central depressor action and has been
proposed as an endogenous anticonvulsant (see references in
Biaggioni et al., 1991a). In fat cells, adenosine abolishes
the breakdown of stored triglycerides to free fatty acids and
glycerol (lipolysis) which is induced by adrenergic
stimulation. Adenosine can also prevent platelet aggregation
(Berne, 1986; Pelleg and Porter, 1990)
Present and Future Therapeutic Uses
Based on its negative dromotropic effect on AV nodal
conduction, adenosine was recently approved by the U.S. Food
and Drug Administration as an antiarrhythmic drug for the
acute management of paroxysmal supraventricular tachycardia
involving the AV node (Belardinelli and Lerman, 1990). In
addition, the transient AV block caused by adenosine can also
be used to unmask underlying atrial activity in other forms
of atrial arrhythmias and hence help in the differential
diagnosis of arrhythmias (Belardinelli and Lerman, 1990).

5
Experimental studies mainly in animal models have
indicated several potential uses of adenosine agonists and
antagonists. The agonists could be used as antiepileptic,
analgesic and sedative drugs due to their inhibitory effect
on neurotransmission (Pelleg and Porter, 1990).
AdoR antagonists could be used for the relief of AV
block associated with acute myocardial infarction. In
addition, they accelerate recovery of myocardial
contractility during cardioversion (Wesley and Belardinelli,
1989) .
Synthesis and Metabolism of Adenosine
Adenosine is a local hormone (or autacoid) rather than a
circulating hormone or neurotransmitter. It acts within the
same organ (s), perhaps even on the cell(s), that is the site
of its production. Unlike neurotransmitters, adenosine can be
produced by virtually any cell. Adenosine per se does not
appear to be stored in exocytotic vesicles, but rather is
produced on demand, much like prostaglandins and
leukotrienes. The primary mechanism for the production of
adenosine in heart muscle, liver and leukocytes is the
dephosphorylation of AMP by a 5'-nucleotidase located on the
cell membrane or in the cytosol. Adenosine produced is then
released into the interstitial space from the parenchymal
cells for receptor interaction. Physiological stimuli that
cause inadequate tissue oxygenation (e.g., hypoxia, ischemia,

6
exercise) greatly increase adenosine production (Olsson and
Pearson, 1990) .
Adenosine can also arise from ATP which is released and
rapidly broken down by ectonucleotidases. ATP is released
from nerve endings (where it is stored in vesicles along with
biogenic amines or other classical neurotransmitters), from
platelets (where it is stored in secretory granules along
with ADP), and from cells that are undergoing lysis. These
sources of adenosine probably are important under specific
circumstances (i.e., at particular synapses or at sites of
injury) (Olsson and Pearson, 1990; Bruns, 1990).
Another intracellular source of adenosine is S-adenosyl-
homocysteine (SAH), which arises from S-adenosylmethionine.
SAH-hydrolase catalyzes the reversible reaction between SAH
and adenosine plus homocysteine. Adenosine also tightly binds
to SAH-hydrolase. Hence, under basal conditions, the
intracellular concentration of free adenosine is probably
very low (Delahaba and Cantoni, 1959; Olsson and Pearson,
1990) .
Adenosine crosses cell membranes by simple diffusion and
and more importantly by facilitated diffusion. Facilitated
diffusion is carrier mediated, nonconcentrative and is
inhibited by dipyridamole (DIP) (Kolassa et al., 1970), 6-S-
(p-nitrobenzyl-thio)inosine (Paterson and Oliver, 1971) and
dilazep (Bruns, 1990). The carrier appears to transport other
nucleosides, which are competitive inhibitors of adenosine
transport. The carrier is also symmetrical, mediating both

7
the uptake and release of adenosine with identical kinetics
(see references in Olsson and Pearson, 1990) .
Adenosine is metabolized very rapidly in the blood with
a half-life of 0.6 10 sec. The principal route of
metabolism is deamination to inosine by adenosine deaminase
and further degradation of inosine to hypoxanthine, xanthine
and eventually to uric acid. Adenosine deaminase can be
inhibited by erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA) and
2'-deoxycoformycin (see references in Bruns, 1990) In
addition, adenosine can reenter the nucleotide pool by
phosphorylation to adenine nucleotides (See references in
Olsson and Pearson, 1990).
Adenosine Receptors
Glassification and Charaterization
Adenosine receptors comprise a group of cell surface
receptors that mediate the physiological and pharmacological
effects of the nucleoside adenosine. At least two distinct
subtypes of cell surface AdoRs are responsible for these
actions. These receptors have been classified as Ai-AdoRs and
A2-AdoRs based on biochemical and pharmacological criteria,
i.e., modulation of adenylyl cyclase (AC) and differential
selectivity for a series of adenosine analogs. The Ai~AdoR
that mediates the inhibition of AC has an agonist potency
series of R-phenylisopropyladenosine ((R)-PIA) > 5'-N-
ethylcarboxamide adenosine (NECA) > (S)-PIA. The A2~AdoR that

8
mediates the stimulation of AC has an agonist potency series
of ECA > (R)-PIA > (S)-PIA (Van Calker et al., 1979; Londos
et al., 1980). The A2~AdoR in brain has been further
subdivided into A2a and A2b subclass. Central A2a-AdoRs are
localized primarily in the striatum, nucleus accumbens and
olfactory tubercle, whereas central A2b-AdoRs are present in
all brain regions. Adenosine and NECA have a higher affinity
for the A2a-AdoRs than the A2b-AdoRs (Daly et al., 1983; Bruns
et al., 1986).
In addition to modulating AC activity, AdoRs are coupled
to other effector systems such as K+ and Ca2+ channels and to
phospholipid hydrolysis. Electrophysiological studies
indicate that adenosine activates an inwardly rectifying
potassium current dKAdo) i-n sinoatrial (Bellardinelli et al.,
1988), atrial (Belardinelli and Isenberg, 1983) and neuronal
(Trussel and Jackson, 1985) cells. The activation of IxAdo is
mediated via a pertussis toxin-sensitive guanosine
triphosphate (GTP)-binding protein (Kurachi et al., 1986).
Adenosine also attenuates the activity of the voltage-
sensitive Ca2+ channels in hippocampal neurons via presynaptic
Ai-AdoRs (Schubert, 1985) Adenosine has also been found to
indirectly stimulate inositol phosphate accumulation in
guinea pig cortex, FRTL-5 thyroid cells and vas deferens. In
these tissues, the effect of adenosine is to potentiate the
responses of neurotransmitters such as histamine,
norepinephrine and angiotensin II. However, neither adenosine
nor its analogs alone increase inositol phopholipid

9
hydrolysis in these tissues (Hill and Kendall, 1987;
Hollingsworth and Dally, 1985; Linden, 1991). In contrast to
potentiating the stimulatory response of other
neurotransmitters on phospholipid metabolism, in several
other tissues (eg., mouse cortex, brown fat and GH3 pituitary
cells), activation of Ai-AdoR leads to inhibition of inositol
phosphate accumulation (Kendall and Hill, 1988; Delahunty et
al., 1988; Linden and Delahunty, 1989; Linden, 1991).
Both A]_ and A2~AdoRs are widely distributed in the
central nervous system and peripheral tissues. For example,
Ai-AdoRs are present in the brain, heart, kidney, lung,
pancreas and adipocytes and A2~AdoRs are present in the
brain, coronary arteries, kidney and lung (Olsson and
Pearson, 1990).
Analysis of structure-activity relationship indicates
that certain N-6 substituents of adenosine enhance the
potency of adenosine as an Ai-AdoR agonist (Olsson and
Pearson, 1990). For example, cyclopentyladenosine (CPA) and
2-chloro-N^-cyclopentyl-adenosine (CCPA) have Kj. (A2)/Kj. (Ai)
ratios of 2500 and 9750, respectively (Lohse et al., 1988).
Several purines with C-2 substituents (e.g., 2-
aralkoxyadenosine, 2-alkoxyadenosine) have increased potency
as A2~AdoR agonists. For instance, 2-[2-(2-naphthyl)ethoxy]
adenosine (WRC-0018) is a highly selective A2~AdoR agonist
(Ueeda et al., 1991). Examples of AdoR agonists and
antagonists and their chemical structures are shown in
Figures 1-1, 1-2.

10
Radioligand binding studies of AdoRs have been attempted
within the past decade and some success has been achieved,
particularly with Ai-AdoR ligands. The first successful
radioligand binding studies of AdoRs were reported in the
early 1980s. Several groups used a variety of tritiated or
iodinated radioligands including both agonists and
antagonists (Linden et al., 1985; Trost and Schwabe, 1981;
Bruns et al., 1980; Williams and Risley, 1980). Radioligand
binding studies in membrane preparations from various tissues
revealed all the appropriate characteristics; that is 1)
saturability, 2) reversibility, 3) stereoselectivity and 4)
the pharmacological specificity expected of the
physiologically relevant receptor. A recent development in
this field has been the synthesis of the high affinity Ai-
AdoR selective radioligand [3H]8-cyclopentyl-l,3-dipropyl-
xanthine ([3H]CPX) (Bruns et al., 1987). This Ai-AdoR
antagonist has a 740-fold Ai-AdoR selectivity over A2~AdoR,
the highest selectivity reported for an adenosine antagonist.
CPX also has very high affinity for the Ai-AdoR (Kp = 0.4 nM)
and extremely low non-specific binding (=3% of total binding)
in rat brain membranes.
Although the availability of agonist and antagonist
radioligands has enabled detailed characterization of the Ai-
AdoR in various tissues (Stiles et al., 1985; Jacobson et
al., 1986; Ramkumar and Stiles, 1988; Martens et al. 1987),
the lack of a highly selective A2~AdoR antagonist has
hampered a similar characterization of the A2~AdoR. Because

11
[3H]NECA has high affinity for A2a-AdoR, it has been used as a
radioligand for this receptor. However, NECA also binds to
Ai~AdoR with high affinity. Thus, when [3H]NECA is used, it is
necessary to block the Ai-AdoR by adding a highly selective
Ai-AdoR ligand such as the agonist CPA or the antagonist CPX
(Hutchison et al., 1989; Linden, 1991). Another A2~AdoR
agonist radioligand recently synthesized is 2[4(2{[4
aminophenyl]methylcarbonyl}ethyl)phenyl]ethylamino-5'-N-
ethylcarboxamido adenosine ( [125I]PAPA-APEC) (Ramkumar et al.,
1990). Likewise, 2-[p-(2-carboxyethyl)phenethylamino]-5'-N-
ethyl-carboxamido adenosine (CGS 21680), an agonist with high
affinity and selectivity for A2~ over Ai-AdoR has been
synthesized (Hutchison et al., 1989; Lupica et al., 1990).
Specific binding of the newly synthesized [3H]CGS 21680 to
rat striatal membranes was saturable and reversible.
Saturation studies revealed that [3H]CGS 21680 binds with
high affinity (Kd=16 nM) to a single class of binding sites.
Adenosine agonists competed for the binding of [3H]CGS 21680
with the following potency order: CGS 21680 > NECA > (R)-PIA
> (S)-PIA. The specific binding of [3H]CGS 21680 was greatest
in rat striatal membranes but negligible in rat cortical
membranes. These results indicate that [3H]CGS 21680 directly
labels the high affinity A2a-AdoR in rat brain without the
need to block binding to Ai-AdoRs (Jarvis et al., 1989;
Jarvis and Williams, 1989).
Ai-AdoRs of brain, heart, or fat cells, when labeled
with photoaffinity ligands or by means of photoaffinity

12
cross-linking, migrate on sodium dodecyl sulfate-
polyacrylamide gel electrophoresis (SDS-PAGE) with an
molecular mass between 35 and 38 kDa (Ramkumar et al., 1990) .
In comparison with Ai-AdoRs, photolabeled A2~AdoRs migrate on
SDS-PAGE with a Mr of 42,000 (Ramkumar et al., 1990). Both
receptors are known to be glycosylated (Ramkumar et al.,
1990). Treatment of photoaffinity-radiolabeled Ai-AdoR of
brain with an endoglycosidase (Stiles, 1985) or with either
trifluoromethanesulfonic acid or a-mannosidase (Klotz and
Lohse, 1986) reduces the molecular mass of the ligand-binding
peptide to 32 kDa. A role for protein glycosylation in the
function and cellular processing of the AdoR is at present
unknown. However, glycosylation has been reported to be
essential in the synthesis of insulin receptors (Ronnett and
Lane, 1981) but not B-adrenoreceptors (BAR) (Doss et al.,
1985). In addition, glycosylation also appears to be required
for the maintenance of cell surface muscarinic receptors
(Liles and Nathanson, 1986).
A]_-AdoRs from rat brain membranes were purified in 1989
(Nakata, 1989a,b; Munshi and Linden, 1989). Ai~AdoR was
solubilized with digitonin and purified approximately 50,000-
fold to apparent homogeneity by two cycles of affinity
chromatography using an antagonist affinity column. The
purified receptor migrated on SDS-PAGE with Mr=34,000 either
in the absence or presence of 2-mercaptoethanol, suggesting
that the receptor does not contain disulfide-linked subunits
(Nakata, 1989a).

13
Recently, two previously cloned proteins (RDC7 and RDC8)
with deduced seven transmembrane helices have been identified
as canine Ai~ and A2a-AdoR subtypes, respectively (Libert et
al., 1989; Maenhaut et al., 1990). The deduced molecular
masses of RDC7 (36,356 Da) and RDC8 (45,008 Da) correspond
closely to the apparent molecular masses of Ai~ and A2a-AdoRs
estimated by photoaffinity labeling. Features shared by both
proteins include small N-termini, conserved transmembrane
domains and at least one cysteine in exofacial loops 1 and 2.
Notable by their absence are consensus sequences for N-linked
glycosylation in the N-terminal segments and an aspartate
residue in the third transmembrane segment, the hallmark of
cationic amine receptors. Clusters of serine and threonine
residues in the C-terminal segments, commonly seen in guanine
nucleotide-binding protein (G protein)-linked receptors, are
absent in RDC7. Finally, at 326 amino acids, RDC7 is among
the smallest members of the G-protein-coupled superfamily of
receptors. RDC7 and RDC8 are similar to a variety of other
superfamily members, but none of these is more than 30%
identical to either of the AdoRs. The expression of RDC8 in
adrenal cells, thyrocytes and Xenopus oocytes resulted in
activation of AC in the absence of added AdoR agonist.
Membranes from Cos 7 cells transfected with RDC8 cDNA
exhibited binding characteristics of an A2~AdoR. Moreover,
RDC8 mRNA and A2~AdoR displayed a very similar distribution
in the brain (Maenhaut et al., 1990). These data all support
that RDC8 is an A2~AdoR. The gene(s) for the low-affinity A2b-

14
AdoR has yet to be cloned. There is no published report about
the binding and functional characteristics of RDC7 in an
expression system.
Receptor-Guanine Nucleotide Regulatory Protein Coupling;
Adenosine receptors modulate AC activity via G proteins,
the stimulatory Gs and the inhibitory Gf for A2~AdoR and Ai~
AdoR, respectively.
The activation of AC by receptors coupled to Gs can be
described by the ternary complex model. This model has been
developed for the BAR system and is likely to be applicable
for other stimulatory receptors including the A2~AdoRs. In
brief, an agonist (Ag) binds to the receptor (R) to form an
Ag-R complex. The affinity of the agonist in the Ag-R complex
is relatively low. The complex then undergoes a
conformational change and interacts with a Gs protein to form
the ternary complex of Ag-R-Gs. The affinity of the agonist
in the Ag-R-Gs complex is relatively high. When GTP binds to
Gs, Ag-R-Gs is rapidly converted to GS-GTP and Ag-R. Once
formed, GS-GTP interacts with the catalytic subunit (C) of AC
to form the active complex C-GS-GTP resulting in a conversion
of ATP to cAMP. The enzyme activity returns to basal level
when guanosine 5'-triphosphatase (GTPase) activity in Gs
hydrolyzes the bound GTP to guanosine diphosphate (GDP) with
the resultant regeneration of inactive catalytic unit and Gs-
GDP. The destabilization of the ternary complex decreases

15
agonist affinity and increases the dissociation rate (Gilman,
1987) .
Several lines of evidence suggest that the AdoR belongs
to the class of G-protein-coupled receptors. For Ai-AdoR, GTP
and stable GTP analogues decreased the apparent affinity of
Ai~AdoR agonists to the receptor in membranes (Goodman et
al., 1982; Yeung and Green, 1983; Lohse et al. 1984),
solubilized receptors (Gavish et al., 1982; Stiles, 1985;
Stroher et al., 1989) and slices of brain tissue (Fastbom and
Fredholm, 1990). Treatment with N-ethylmaleimide which
uncouples receptors from G proteins inhibited the action of
Ai-AdoR agonists without affecting the binding of antagonists
(Fredholm et al., 1985). In addition, detergent-solubilized
Ai_AdoR co-eluted with a G protein from an agonist affinity
chromatography column where GTP or N-ethylmaleimide, agents
known to uncouple receptors from G proteins and
simultaneously lower the affinity of agonists for the
receptor, were used (Munshi and Linden, 1989; Linden, 1991).
It was suggested that Ai-AdoR, unlike other G protein coupled
receptors, tightly binds to G protein. The coelution did not
occur when an antagonist affinity column was used (Nakata,
1989a, b) .
For the A2~AdoR, experiments showed a GTP dependence for
NECA stimulation of AC in purified hepatic plasma membranes
(Cooper and Londos, 1979). AC activity from other tissues has
been shown to be enhanced by or dependent on the presence of
a guanine nucleotide (Londos et al., 1979; Fain and Malbon,

16
1979; Londos, et al., 1981; Wolff et al., 1981). Moreover,
adenosine shortened the lag period for the onset of AC
activation by Gpp(NH)p (Sevilla et al., 1977; Lad et al.,
1980) .
The interaction of an activated receptor with G protein
is a key step in signal transduction. The regulatory G
proteins are heterotrimers with subunits designated a, 5 and y
in order of decreasing mass. The functional difference
between Gs and Gi resides in the respective as and Cti
subunits. The structural features common to 0CS and Oti include
a GTPase activity and regions that recognize and couple these
proteins to a receptor, to the I5+y complex, and to the
effector systems. Common 15 and y subunits are functionally
interchangable (Gilman, 1987). G proteins cycle between an
inactive GDP state and an active GTP state. When GDP is
bound, the a subunit associates with the 15 and y subunits to
form a Gofiy complex (denoted by G-GDP) that is membrane-bound.
When GTP is bound, the a subunit (Ga-GTP) dissociates from
the 15 and y subunits (Guy) Ga-GTP released from G[jy then
alters the activity of the target, such as AC or an ion
channel (Stryer and Bourne, 1986).
Another feature of G proteins is that the cysteine
residue in their a subunits is a substrate for adenosine
diphosphate (ADP)-ribosylation which transfers an ADP-ribose
moiety from nicotinamide-adenine dinucleotide (NAD). Gs is
selectively ADP-ribosylated by cholera toxin which in turn
inhibits the receptor-stimulated GTPase activity of the G

17
protein causing GTP to be bound to as for a prolonged period.
This results in a permanent activation of AC and a large
increase in cAMP accumulation (Gilman, 1987; Nathanson,
1987). Pertussis toxin (PTX) selectively ADP-ribosylates Gi
(and another guanine nucleotide binding protein, G0) This
covalent modification of Gi inactivates the protein resulting
in the loss of receptor mediated inhibition of AC activity
(Gilman, 1987; Nathanson, 1987). Thus, these bacterial toxins
have been used as an investigative tool to determine the role
of G proteins in the action of biologic messengers. For
example, studies showed that PTX blocked the ability of Ai~
AdoR to inhibit AC (Hazeki and Ui, 1981) and prevented
adenosine-induced changes in the rate of beating in rat atria
(Endoh et al., 1983).
Mechanisms for Inhibitory Receptor Action
Receptors that mediate the attenuation of AC activity
include the Ai-AdoR, muscarinic M2-acetylcholine receptor, 0C2-
adrenergic receptor, 5 opiate receptor and D2~dopamine
receptor.
Compared to stimulatory receptors, much less is known
about the mechanisms whereby inhibitory receptors attenuate
the activation of AC by stimulatory receptors. Currently,
there is experimental evidence supporting three different
models. First, inhibition may occur by preventing the
formation of the ternary complex which is composed of the

18
agonist, the stimulatory receptor and Gs. This results in the
inability of the receptor to form a high affinity binding
state. Experiments on rat ventricular myocyte membranes
showed that PIA inhibited isoproterenol (ISO)-stimulated AC
activity (Romano et al., 1988; Romano et al., 1989). This
inhibition was antagonized by theophylline. PIA was much less
effective at attenuating forskolin-stimulated AC activity and
had no effect on 5'-guanyl-imidodiphosphate (Gpp(NH)p)-
induced stimulation. In [125I]cyanopindolol (CYP)/ISO
competition binding experiments, ISO produced a
concentration-dependent displacement of specific [125I]CYP
binding with an IC50 of 48 nM and Hill slope of 0.6. About 38%
of BARs were in the high affinity state. Gpp(NH)p shifted the
competition curve to the right (IC50 = 520 nM) and steepened
the slope (Hill slope = 1.2) indicating that all of the BARs
were in low affinity state. PIA significantly increased the
IC50 for ISO in the absence of Gpp(NH)p (IC50 = 140 nM) and
steepened the slope (Hill slope = 0.9). These findings were
interpreted to indicate that binding of ISO to the high
affinity state of the BAR was decreased in the presence of
PIA. PIA had no effect on the ISO competition curve in the
presence of Gpp(NH)p (Romano et al., 1988; Romano et al.,
1989).
The second is the subunit dissociation model. This model
is based on the finding that both Gs and Gi share two common
subunits, lb and y, and that (Xi is present in excess relative
to (X3 in most cells. Activation of Gi leads to subunit

19
dissociation and the release of sufficient quantities of the
B+Y complex. The large increase in the amount of free B+Y
complex in the membrane would disturb the equilibrium which
exists between undissociated and dissociated Gs under resting
conditions. Thus, by mass action, B+Y complex would combine
with the released stimulatory a subunits preventing its
dissociation and subsequent activation of AC. This model,
therefore, implies that Gi will only be effective under
conditions where the activity of AC is stimulated, i.e., the
effectiveness of Gi is related to the level of the
dissociated Gsa (Morgan, 1989). Experiments supporting this
model showed that when platelet membranes were treated for
brief periods of time with GTPyS and an 0C2-adrenergic agonist
in low Mg2+ conditions, AC was "irreversibly" inhibited. This
inhibition was of the same magnitude as that produced by
maximally effective concentrations of B+Y complex, and it was
not additive with the effect of B+Y- This inhibition is
completely overcome by reconstitution of the membranes with
physiological concentrations of Gia~GDP. The most likely
explanation for this observation is the interaction between
Gia-GDP and Gjj+y to relieve the inhibition caused by free B+Y
complex in the membrane (Katada et al., 1984a).
The third is the direct interaction model. It involves
inhibition of AC by the released OCi subunits acting directly
on the catalytic subunit of the enzyme (Gilman, 1987).
Evidence favoring this model over the subunit dissociation
model includes the observation that inhibitory agonists are

20
capable of reducing AC activity in the eye S49 cell mutant.
These cells lack Gsa, and logically, li+y is not inhibitory
when reconstituted with eye" membranes. It was also
demonstrated that the isolated Gitt from rat liver can inhibit
AC activity in membranes from eye- S49 cell. The inhibitory
effect of Gia was therefore proposed to explain the ability
of inhibitory agonists to decrease AC activity in the cyc-
S49 cell mutant (Katada et al., 1984b; Jakobs and Schultz,
1983). This inhibitory effect of Gia-GTPyS has also been
observed by Roof et al. (1986) in the bovine central nervous
system.
Regulation of Adenosine Receptors
Similar to many other receptors, the AdoR appears to
undergo desensitization and down-regulation during chronic
exposure to an agonist. This effect prevents overstimulation
of the receptor. The mechanism(s) of desensitization and
down-regulation has/have been studied extensively in the J5AR-
AC system and may be applicable to the AdoR. Desensitization
describes the phenomenon where an initial exposure of a cell
to an agonist results in a reduced capacity of the cell to
respond to a second challenge. Two main types of
desensitization have been described. "Homologous"
desensitization is referred to hormone-specific type where
loss of response is only to the activated receptor, whereas
other receptor mediated responses remain unaffected.

21
"Heterologous" desensitization is referred to hormone-
nonspecific type where activation of one receptor causes loss
of response mediated by other receptors. Desensitization of
15AR system appears to be initiated by receptor
phosphorylation which results in functional uncoupling of the
J5AR from Gs. Two kinases have been implicated: the cAMP-
dependent protein kinase (PKA) which plays a major role in
heterologous desensitization; and the cAMP-independent
kinase, termed J5AR kinase which specifically phosphorylates
the agonist-occupied receptor leading to homologous
desensitization. After uncoupling, the receptors appear to be
sequestered within the cells. Removal of the agonist after
sequestration leads to rapid resensitization of the system.
During longer-term agonist treatment, there appears to be a
loss of receptors (down-regulation) due to receptor
degradation or loss of the recognition site for ligand
binding (Harden, 1983) .
Studies have shown that in vitro exposure of cultured
rat adipocytes to (R)-PIA causes concentration- and time-
dependent loss of Ai-AdoR and decrease in the content of Gi
protein (Green, 1987). These changes were accompanied by
attenuation of the antilipolytic effect of (R)-PIA
(homologous desensitization) (Green, 1987). In another study,
the number of cardiac Ai~AdoRs in chick embryos was decreased
by 63% after pretreatment with 1 JIM (R)-PIA for 44 hrs
(Shryock et al., 1989). Experiments also showed that the
desensitization of Ai~AdoR system in DDTi MF-2 cells was

22
accompanied by a decrease in the number of Ai-AdoRs which can
form a high affinity agonist binding site and a 3-4 fold
increase in the phosphorylation of Ai-AdoR (Ramkumar et al.,
1991) Thus, similar to the IbAR system, Ai-AdoR can also
undergo desensitization and/or down-regulation after chronic
exposure to an agonist. Uncoupling, down-regulation and
phosphorylation of the Ai-AdoR may contribute to the
desensitization of this inhibitory receptor.
Desensitization of A2~AdoR-AC system has also been
described. Using clonal neuronal cells (NG108-15), which
express both A2~AdoR and prostaglandin E (PGEi) receptors, PGE
pretreatment reduced the effects of both PGEi and adenosine
to activate AC (heterologous desensitization) In contrast,
exposure of NG108-15 to 2-chloroadenosine resulted in a rapid
loss of response to 2-chloroadenosine (homologous
desensitization), but PGEi-stimulated AC activity decreased
only slightly (Kenimer and Nirenberg, 1981) .
Adenosine receptors are regulated during chronic drug
treatment with AdoR antagonist or dexamethasone. A recent
study showed that exposure of guinea pig myocardium to AdoR
antagonist theophylline increased the the number of Ai-AdoR
(Wu et al., 1989) In humans, after 7 days of caffeine (AdoR
antagonist) abstinence, NECA produced a concentration-
dependent inhibition of thrombin-induced platelet aggregation
with an EC50 value of 69 nM (Biaggioni et al., 1991b).
Subjects were then given caffeine 250 mg p.o. 3 times a day
for 7 days. Caffeine withdrawal significantly shifted the

23
concentration response of NECA to the left (EC5o=49 nM, p<0.01
by ANOVA) indicating sensitization of AdoRs (Biaggioni et
al., 1991b). Other examples include pretreatment of DDTi MF-2
cells with dexamethasone (Gerwins and Fredholm, 1991). This
glucocorticoid caused a concentration- and time-dependent
increase in the number of Ai-AdoRs, but did not affect the Kd
or the proportion of A]_-AdoRs in high and low affinity states
(Gerwins and Fredholm, 1991). (R)-PIA was more potent as an
inhibitor of cAMP formation induced by ISO in dexamethasone-
treated cells. Addition of glucocorticoid receptor antagonist
RU 486 or protein synthesis inhibitor cycloheximide prevented
the up-regulation of Ai-AdoR (Gerwins and Fredholm, 1991). In
contrast to sensitization of Ai-AdoR-AC system, the A2~AdoR-AC
system was desensitized as indicated by the decreased ability
of NECA to increase cAMP formation in dexamethasone-treated
cells (Gerwins and Fredholm, 1991) .
AdoRs are also regulated under normal and
pathophysiological conditions. For example, in the
hypothyroid state, (R)-PIA mediated inhibition of AC and its
antilipolytic effect is enhanced (Ohisalo and Stouffer,
1979). In contrast to hypothyroidism, the hyperthyroid state
is characterized by enhanced lipolytic activity and cAMP
accumulation in adipocytes. These are likely related to the
loss of inhibitory tone mediated by Ai-AdoR due to a 35%
decrease in Ai-AdoR number (Malbon et al., 1978; Rapiejko and
Malbon, 1987). Other conditions that may alter AdoRs include

24
pregnancy, lactation, starvation, obesity and aging (Ramkumar
et al., 1988).
Goals
Over the past decade, a great deal has been learned
about the pharmacology, biochemistry and physiology of AdoRs.
However, much remains unknown. In general, receptors that
mediate inhibition of cAMP formation appear to dominate over
stimulatory receptors. In the case of AdoRs, if both subtypes
are present in a single cell and are simultaneously activated
by adenosine, it becomes important to determine under what
conditions the Ai~ and A2~AdoR mediated responses will be
expressed. Thus, the major goals of this study were 1) to
determine pharmacologically if an interaction between Ai~ and
A2AdoR occurs and 2) to define the conditions whereby the
expression of the A2~AdoR mediated response can be
demonstrated.
In addition, the hypothesis that the mechanism for Ai~
AdoR inhibitory effects involves alteration in the ability of
BAR agonists to interact with the BAR was investigated. By
using an irreversible BAR agonist that permanently activates
the BAR, it was determined if the resulting response can be
modulated by the inhibitory Ai-AdoR.

25
AGONIST
ANTAGONIST
CPA
N-0861
CCPA
CPX
NECA
8PST
ADO
WRC-0018
8PST
NECA
?
ADO
A, -AdoR decrease cAMP
A2 -AdoR increase cAMP
Figure 1-1. Agonists and antagonists of AdoR subtypes.

26
Q
CPA
CCPA
WRC-0018
N-0861
CPX
8PST
Figure 1-2. Chemical structures of AdoR agonists and
antagonists. Abbreviations used: N6-cyclopentyladenosine
(CPA), 2-chloro-N6-cyclopentyladenosine (CCPA), 2[2(2
Naphthyl)ethoxy]adenosine (WRC-0018), 5'-N-ethylcarboxamido-
adenosine (NECA), ()N6-endonorbornan-2-yl-9-methyladenine
(N-0861), 8-cyclopentyl-l,3-dipropylxanthine (CPX), 8(p-
sulfophenyl)theophylline (8PST).

CHAPTER 2
EXPERIMENTAL PROCEDURES
Source of Materials
The radioligands [2,8-3H]adenosine 3', 5'-cyclic
monophosphate ([3H]cAMP; 31.2 Ci/mmol), [3H]8-cyclopentyl-l,3-
dipropylxanthine ([3H]CPX; 99-107 Ci/mmol) and (-) [125I]iodo-
cyanopindolol ([125I]CYP; 2,000-2,200 Ci/mmol) were purchased
from New England Nuclear Corp. (Boston, MA, USA) Adenosine,
N6-cyclopentyladenosine (CPA), (-)-N6-(2-phenyl-isopropyl)
adenosine ((R)-PIA), dipyridamole (DIP), erythro-9-(2-
hydroxy-3-nonyl)adenine (EHNA), adenosine deaminase (ADA)
type VI, 3-[(3-cholamidopropyl)dimethylammoniol]-1-propane-
sulfonate (CHAPS), benzamidine, (-)-isoproterenol (ISO), 5'-
guanylyl-imididodiphosphate (Gpp(NH)p), propranolol (PROP),
()-alprenolol, penicillin G, streptomycin sulfate,
amphotericin B, theophylline, protein kinase, hydroxyapatite
and bovine serum albumin were from Sigma Chemical Co. (St.
Louis, MO, USA). 8-cyclopentyl-l,3-dipropylxanthine (CPX),
8(p-sulfophenyl)theophylline (8PST), 5'-N-ethylcarboxamido-
adenosine (NECA) and 2-chloro-N^-cyclopentyladenosine (CCPA)
were purchased from Research Biochemicals Inc. (Natick, MA,
USA). The DDTi MF-2 (DDT) cell line was obtained from
American Type Culture Collection (Rockville, MD, USA).
27

28
Dulbecco's modified Eagle's media (DMEM) and fetal bovine
serum were from Gibco (Grand Island, NY, USA). Liquiscint was
purchased from National Diagnostics (Somerville, NY, USA). 2-
[2-(2-Naphthyl)ethoxy]adenosine (WRC-0018) was a kind gift of
Dr. Ray. A. Olsson (Univ. of South Florida, Tampa, FL, USA).
()N^-endonorbornan-2-yl-9-methyladenine (N-0861) was a gift
of Whitby Research, Inc. (Richmond, VA, USA). Pertussis toxin
was a gift of Dr. Eric Hewlett (Univ. of Virginia,
Charlottesville, VA, USA). Rolipram was a gift of Berlex
Laboratories (Cedar Knolls, NJ, USA). 5[2[ [3[4
(bromoacetamido)phenyl]-2-methylprop-2-yl]amino]-1-
hydroxyethyl]-8-hydroxycarbostyril (C-Br) was synthesized as
described previously (Milecki et al., 1987). All other
reagents were from Sigma Chemical Co. (St. Louis, MO, USA) or
Fisher Scientific (Orlando, FL, USA).
Methods
Cell Culture
The DDT cell line was derived from a steroid-induced
leiomyosarcoma tumor of the vas deferens of an adult Syrian
hamster (Norris and Kokler, 1974). These cells were obtained
at low passage number and grown as a monolayer on 150 mm
plastic culture dishes (Falcon) in DMEM supplemented with 5%
fetal bovine serum, 100 U/ml penicillin G, 0.1 mg/ml
streptomycin and 2.5 (Ig/ml amphotericin B in an atmosphere of
5% C02/95% air at 37C. Cells were seeded at 0.2-1 x 104

29
cells/cm2 and subcultured twice weekly after detachment using
1 mM ethylenediamine tetraacetic acid (EDTA) in phosphate
buffered saline (PBS). DDT cells have a doubling time of
about 20 hours and a confluent density of 1.3 x 105 cell/cm2.
Experiments were performed on cells 1-day pre-confluent.
Drug Preparation
Stock solutions of WRC-0018 (10 mM) and rolipram (50 mM)
were prepared in dimethylsulfoxide (DMSO). CPA (1 mM), CCPA
(1 mM), DIP (10 mM) and C-Br (1 mM) were prepared in ethanol.
ECA (1 mM) was dissolved in 5 mM HC1 and N-0861 (1 mM) was
dissolved in a mixture of ethanol (10%) and 50 mM Tris buffer
containing 10 mM MgCl2 (90%). These stock solutions were
diluted in Hank's Balanced Salt Solution (HBSS) without
divalent cations to the desired concentrations just prior to
use. HBSS contains 137 mM NaCl, 6 mM D-glucose, 5 mM KC1, 4
mM NaHCC>3, 0.6 mM Na2HP04, 0.4 mM KH2PO4, 0.5 mM MgCl2, 0.4 mM
MgS04 and 1 mM CaCl2 at pH 7.4. All other drugs were dissolved
in HBSS before use.
Drug Treatment
The growth medium in culture dishes was aspirated and
fresh medium (20 ml) was then added followed by the drug. The
cells were then incubated at 37C for the various period of
times as indicated in the text. At the end of the incubation
period, the media was aspirated and the attached cells were

30
washed four times with 10 ml of ice-cold HBSS without
divalent cations.
Membrane Preparation
DDT cells were harvested from culture dishes in 5 ml of
50 mM Tris-HCl buffer at pH 7.4 containing 5 mM MgCl2 with a
rubber policeman and were pelleted by centrifugation at
48,000g for 15 min.
For the determination of Ai-AdoR, the pellet was
resuspended in ice-cold 50 mM Tris-HCl (pH 7.4) containing 1
mM MgCl2, 1 mM EDTA and 0.1 mM benzamidine (trypsin
inhibitor) and then homogenized with a Ten Broeck Tissue
Grinder (glass-glass). The homogenate was centrifuged at
48,000g for 15 min to pellet the membranes. The membranes
were homogenized a second time in 50 mM Tris-HCl (pH 7.4)
containing 1 mM MgCl2/ then used for protein measurement and
receptor binding assays.
For determination of the JAR, the pellet from the first
centrifugation as described above was homogenized in ice-cold
50 mM Tris-HCl (pH 7.4) containing 5 mM MgCl2 with a Tekmar
SDT-100EN homogenizer (setting 5, 20 s). After centrifugation
at 48,000g for 15 min and homogenization with the Tekmar as
above, the membrane suspension was used for assays.

31
Protein Measurement
The protein content of cells and membranes was
determined by the method of Lowry et al.(1951) using bovine
serum albumin as standard.
Radioligand Binding Assay
Ai-AdoRs in DDT cells were determined by specific [3H]CPX
binding. Membrane protein was initially incubated with 2 U/ml
(1 U = 6.25 Jig) ADA for 20 min at 4C to metabolize
endogenous adenosine. Cell membranes (=0.1 mg protein) were
then incubated in a total volume of 0.2 ml with 50 mM Tris-
HC1 buffer at pH 7.4, 5 mM MgCl2 and 0.06-4 nM [3H]CPX, with
or without 50 |1M (R)-PIA, for 150 min at room temperature on
an orbital shaker. The bound and free ligand were rapidly
separated on GF/C glass fiber filters (Whatman Inc., Clifton,
NJ, USA) using a Brandel Cell Harvester (Brandel Scientific,
Gaithersburg, MD, USA). Filters were rinsed three times with
4 ml of ice-cold 50 mM Tris-HCl buffer containing 10 mM MgCl2
and 0.1 % CHAPS (to reduce non-specific binding). The filters
were placed in standard scintillation vials with 10 ml of
Liquiscint and the radioactivity was determined in a liquid
scintillation counter. Specific binding to Ai-AdoR was
calculated as the difference between the total binding in the
absence of (R)-PIA and the nonspecific binding in the
presence of 50 (1M (R)-PIA. Specific binding was generally 90-

32
95% of total binding. All assays were performed in
triplicate, and the determinations differed by less than 6%.
BARS were quantitated by specific [125I]CYP binding.
Membrane protein (30-50 |lg) was incubated in a total volume
of 0.25 ml with 50 mM Tris-HCl buffer at pH 7.4, 5 mM MgCl2
and 6-100 pM [125I]CYP, with or without 3 |1M ()-alprenolol,
for 60 min at 36C. The bound and free ligand were then
rapidly separated on GF/B glass fiber filters using a Brandel
Cell Harvester. Filters were rinsed three times with 4 ml of
ice-cold 50 mM Tris-HCl buffer containing 5 mM MgCl2, placed
in omnivials and the radioactivity was determined in a gamma
counter. Specific binding was typically 80-90% of total
binding. All assays were performed in triplicate, and the
determinations differed by less than 6%.
p.AMP Assay
Cells were detached from culture dishes with a cell
lifter and centrifuged at 500g for 5 min. The cells (0.4 mg
protein/tube = 2xl06 cells) were gently resuspended in HBSS
containing 50 )iM rolipram (phosphodiesterase inhibitor) and
incubated at 36C for 7 min in Beckman microcentrifuge tubes.
Drugs were then added and the cells were incubated at 36C
for the various period of times as indicated in the text. At
the end of the incubation period, the tubes were immediately
placed in a boiling water bath for 5 min. The protein was

33
pelleted by centrifugation at 9,OOOg for 2 min, and the
supernatants were saved for cAMP assays.
The cAMP content of the supernatant was determined by a
modification of a competitive protein binding assay described
previously (Baker et al., 1985). An aliquot (usually 50 |ll)
of the supernatant was incubated in a total volume of 0.2 ml
with 25 mM Tris-HCl buffer at pH 7.0, 8 mM theophylline, 0.8
pmol of [3H]cAMP and 24 |ig of bovine heart cAMP dependent
protein kinase at 4C for 60 min. At the end of the
incubation, 70 p.1 of a 50% (v/v) hydroxyapatite suspension
was added to each tube. The suspensions were then poured onto
a Whatman GF/C glass fiber filter under reduced pressure. The
filters were rinsed three times with 4 ml of ice-cold 10 mM
Tris-HCl buffer and placed in minivials with 3 ml of
Liquiscint. Radioactivity was determined in a liquid
scintillation counter. The amount of cAMP present was
calculated from a standard curve determined using known
concentrations of unlabeled cAMP.
Data Analysis
Receptor density (Bmax) and dissociation constant (Kd)
for the radiolabeled ligands were determined from regression
analysis of Scatchard plots (1949). The concentrations of
compounds which inhibited ligand binding by 50% (IC50) were
obtained from Hill plots of the competition data (Hill,
1913). The effective concentrations of drugs which gave 50%

34
of a maximal response (EC50) were determined using a
concentration-effect analysis with a non-linear regression
algorithm (Marquardt-Levenberg). Statistical analysis of
significance of difference was performed using the Student's
t-test.

CHAPTER 3
INTERACTION OF ADENOSINE RECEPTORS
Introduction
Cell surface adenosine receptors (AdoRs) have been
classified by pharmacological and biochemical criteria. Two
subtypes of AdoRs i.e., A^- and A2~AdoR have been thus far
clearly identified. In most cell types studied to date, the
Ai-AdoR mediates an inhibition of AC activity whereas the A2-
AdoR mediates a stimulation of the enzyme (Van Calker et al.,
1979; Londos et al., 1980). In general, receptors that
mediate inhibition of cAMP formation dominate over receptors
that mediate stimulation. For example, in mouse atria,
carbachol antagonized ISO-stimulated cAMP accumulation by
direct activation of the muscarinic receptors. The
interaction between carbachol and ISO was not competitive,
since cholinergic inhibition could not be surmounted by
increasing concentrations of ISO (Brown, 1979). In atria
isolated from rats, carbachol decreased the ISO-induced
elevation of cAMP levels and inhibited the positive
chronotropic and inotropic responses to ISO (Endoh et al.,
1985). In addition, after desensitization of the muscarinic
system in AtT-20 cells by oxotremorine, cAMP accumulation
stimulated by ISO was approximately doubled (Heisler et al.,
35

36
1985). In most if not all of these examples, the inhibitory
effect cannot be overcome even when the receptor is maximally
activated.
The DDT smooth muscle tumor cell line is derived from a
steroid-induced leiomyosarcoma of the vas deferens of an
adult Syrian hamster (Norris and Kokler, 1974). This cell
line has been shown to express a relatively high density of
12~ARs which mediate a robust stimulation of cAMP formation
(Norris et al., 1983). The DDT cells has been used therefore
as a model to study this receptor system. These cells have
also been shown to express functional histamine Hi
(Mitsuhashi and Payan, 1988) and steroid receptors (Norris
and Kohler, 1977) Recently, the presence of both Ai~ and A2-
AdoR have been demonstrated in DDT cells by radioligand
binding, photoaffinity labeling and their ability to alter AC
activity (Ramkumar et al., 1990). In keeping with the
classical definition of these two subtypes, the Ai-AdoR in
DDT cells inhibits AC activity whereas the A2~AdoR stimulates
the enzyme. The presence of both AdoR subtypes on a single
cell coupled to the same second messenger system provides a
unique opportunity to characterize pharmacologically if an
interaction between the Ai~ and A2~AdoR occurs. Assuming that
an Ai-AdoR inhibitory response would mask any A2~AdoR mediated
stimulation of cAMP accumulation, three experimental
approaches were used to attempt to alter Ai-AdoR
responsiveness and allow the expression of the A2~AdoR
response. These three approaches are shown diagrammatically

37
in Figure 3-1. First, selective blockade of the Ax-AdoR;
second, uncoupling of the Ax-AdoR with PTX and third, down-
regulation and/or desensitization of the Ax-AdoR using a
selective agonist.
Results
Effects of Selective and Nonselective Adenosine Receptor
Agonists on cAMP Accumulation in DDT Cells
Experiments were designed to investigate the effects of
the selective Ax-AdoR agonist CPA, the putative nonselective
agonists NECA and Adenosine (ADO), and the selective A2~AdoR
agonist WRC-0018 on cAMP accumulation in DDT cells.
Figure 3-2 illustrates the concentration-dependent
inhibition of ISO (10 (J.M)-induced cAMP accumulation by CPA,
NECA and ADO in DDT cells. The effect of ADO was studied in
the presence of DIP (inhibitor of adenosine transporter) and
ENHA (inhibitor of adenosine deaminase) to prevent the uptake
and degradation of this nucleoside respectively during the 7
min incubation period. The rank order of potency of these
AdoR agonists to inhibit ISO-induced cAMP accumulation was
CPA > NECA > ADO with EC50 values of 1.5, 14 and 97 nM,
respectively. The maximal inhibition of cAMP accumulation
caused by the AdoR agonists was 89% for CPA, 97% for NECA and
79% for ADO. Figure 3-3 illustrates the concentration
response of ISO to stimulate cAMP accumulation in the absence
and presence of CPA (0.1 JIM) or NECA (1 (IM) These

38
concentrations of Ai-AdoR agonists caused maximal inhibition
of ISO-stimulated cAMP accumulation (Figure 3-2). ISO alone
increased cAMP accumulation in a concentration-dependent
manner with an EC50 of 2.4 nM and a maximal stimulation of 62-
fold above the basal level achieved at 0.1 [1M. In the
presence of CPA or NECA, ISO still stimulated cAMP
accumulation in a concentration-dependent manner, but the
maximal response was decreased by 75% and 60% in the presence
of CPA and NECA, respectively. In the presence of 1 |1M CPX, a
selective Ai-AdoR antagonist, the inhibitory effect of CPA on
ISO-stimulated cAMP accumulation was significantly attenuated
(data not shown). Thus, our results indicated that the Ai-
AdoR agonists had an inhibitory effect on cAMP accumulation
and this effect was mediated by Ai-AdoR.
The effects of the selective A2~AdoR agonist WRC-0018
(Ueeda et al., 1991) and the nonselective agonist NECA on
cAMP accumulation are illustrated in Figure 3-4. WRC-0018
produced a biphasic response whereas NECA caused no effect on
cellular cAMP level. WRC-0018 at low concentrations (5 500
nM) stimulated cAMP accumulation with a maximal response of
81 -fold above the basal level. The estimated EC50 was 8.6
nM. However, higher concentrations (>500 nM) of WRC-0018
decreased cAMP accumulation with an EC50 value of 5.1 )1M. In
contrast to WRC-0018, NECA over the entire concentration
range of 0.1 nM 10 (1M did not affect (increase or decrease)
cellular cAMP content. Similar to NECA, ADO (0.1 nM 10 (1M)

39
also did not affect the cellular cAMP level above the basal
level (data not shown).
Figure 3-5 depicts the effect of the non-selective AdoR
antagonist 8PST on the stimulatory effect of WRC-0018. WRC-
0018 (0.1 |IM) induced a 3-fold increase in cAMP content above
the basal level and 8PST (5 |1M) inhibited the WRC-0018-
induced cAMP accumulation by 80%. Figure 3-6 illustrates a
similar effect of the Ai-AdoR agonist CPA on the stimulatory
effect of WRC-0018. At a concentration of 0.1 |1M, WRC-0018
stimulated cAMP accumulation 3-fold above the basal level and
CPA (1 |J.M) inhibited the WRC-0018-induced cAMP accumulation
by 78%.
Figure 3-7 illustrates the effects of WRC-0018 on cAMP
accumulation in the absence and presence of 10 [1M ISO. The
biphasic concentration response curve of WRC-0018 was
replotted from Figure 3-4. In the presence of ISO, low
concentrations of WRC-0018 (1 nM 100 nM) did not affect
ISO-stimulated cellular cAMP accumulation. However, at higher
concentrations (>500 nM), WRC-0018 attenuated the stimulatory
effect of ISO in a concentration-dependent manner with an EC50
of 840 nM and a maximal inhibition of 73%.
These data show that nonselective AdoR agonists (NECA
and ADO) themselves only inhibit cAMP accumulation. In
contrast, the selective A2~AdoR agonist (WRC-0018) stimulated
at lower and inhibited cAMP accumulation at higher
concentrations, thereby resulting in a biphasic concentration
response curve. To further investigate the expression of the

40
A2~AdoR mediated cAMP accumulation, three experimental
approaches were used (Figure 3-1). These were 1) selective
blockade of Ai-AdoR, 2) uncoupling of Ai inhibitory effects
with PTX, and 3) selective desensitization and/or down-
regulation of Ai-AdoR.
Effect of a Selective A^-AdoR Antagonist on the A?~AdoR
Mediated Response
The first experimental approach used to uncover the A2-
AdoR mediated effect on cAMP accumulation was blockade of the
Ai-AdoR with a selective Ai-AdoR antagonist. The ability of
the highly selective Ai-AdoR antagonist, N-0861 (Shryock et
al., 1992), to compete with [3H]CPX for the Ai-AdoR binding
site is shown in Figure 3-8. N-0861 produced a concentration-
dependent displacement of specific [3H]CPX binding with an
IC50 of 0.8 (IM and a Hill slope of 1.0. This indicated that N-
0861 bound to a single class of binding sites. Figure 3-9
illustrates the effect of N-0861 on the Ai-AdoR inhibitory
effect of CPA. ISO (10 |1M) produced a 12-fold increase in
cAMP accumulation above the basal level and CPA (0.1 fiM)
inhibited the stimulatory effect of ISO by 70%. N-0861 (10
JIM) attenuated the inhibitory effect of CPA by 62%. A similar
effect of N-0861 on NECA-induced inhibition of ISO-stimulated
cAMP accumulation is shown in Figure 3-10. That is, ISO (10
|1M) stimulated cAMP accumulation 47-fold above the basal
level and ECA (1 (1M) inhibited this increase by 74%. N-0861
(10 |1M) attenuated the inhibitory effect of NECA by 60%. The

41
effect of N-0861 on WRC-0018 stimulation of cAMP accumulation
is shown in Figure 3-11. The 3-fold increase in cAMP
accumulation caused by 0.1 |1M WRC-0018 was not affected by N-
0861 (0.1 nM 10 [1M) Based on these results, in the
remaining experiments of this series, 10 (1M N-0861 was used
to selectively block the Ai-AdoR mediated inhibition of cAMP
accumulation.
The effects of N-0861 on the A2~AdoR mediated response
of selective and nonselective agonists were investigated.
Figure 3-12 illustrates the concentration response of WRC-
0018 in the absence and presence of N-0861. The biphasic
concentration response curve of WRC-0018 in the absence of N-
0861 was replotted from Figure 3-4. N-0861 (10 H-M) completely
abolished the Ai~AdoR mediated inhibition of cAMP
accumulation caused by WRC-0018 and hence, the downward
component of the biphasic concentration response curve of
WRC-0018 was eliminated. The EC50 for the A2~AdoR mediated
effect of WRC-0018 on cAMP accumulation was 93 nM. The
maximal stimulation of cAMP accumulation by WRC-0018 was
increased (p<0.05) from 81 -fold in the absence to 156 -
fold in the presence of N-0861. The effects of NECA on cAMP
accumulation in the absence and presence of N-0861 are shown
in Figure 3-13. In the absence of N-0861, NECA did not
stimulate cAMP accumulation. In contrast, in the presence of
N-0861 (10 (1M) NECA stimulated cAMP accumulation over the
concentration range of 1 }1M 100 |1M. At concentrations of
NECA >10 HM, there was a decrease in the cAMP level despite

42
the presence of N-0861 in the incubation medium. Similar
results to those of NECA were observed with ADO (Figure 3-
14) ADO (0.1 nM 10 p.M) caused no effect on the cellular
cAMP content in the absence of N-08 61 whereas ADO (1 |1M 5
(1M) stimulated cAMP accumulation in the presence of 10 |1M N-
0861. At 10 |1M ADO, there was a decrease in cAMP formation
despite the presence of N-0861 in the incubation medium.
In summary, these data show that the selective Ai-AdoR
antagonist N-0861 abolishes the downward phase of the WRC-
0018 biphasic concentration response and uncovers a
stimulatory (i.e., A2~AdoR mediated) response of NECA and
ADO.
Effect of PTX on the A2~AdoR Mediated Response
The second experimental approach used to unmask, the A2-
AdoR mediated effect on cAMP accumulation was uncoupling of
Ai inhibitory effects with PTX. Based on a previous study
which showed that inhibition of AC activity by (R)-PIA was
markedly attenuated after an 18 hr pretreatment of DDT cells
with 100 ng/ml PTX (Ramkumar et al., 1990), I chose to
incubate our DDT cells with PTX for 18 hr. The ability of CPA
to inhibit ISO-stimulated cAMP accumulation in cells
pretreated with various concentrations of PTX for 18 hr is
shown in Figure 3-15. In control (untreated) DDT cells, 10 (1M
ISO produced a 15-fold increase in cAMP content above the
basal level. As expected, CPA (1 (1M) attenuated the

43
stimulatory effect of ISO by 84%. Pretreatment of cells with
> 25 ng/ml PTX for 18 hr resulted in a complete loss of the
CPA inhibitory effect on cAMP accumulation. Basal and 10 (IM
ISO-stimulated cAMP accumulation were not significantly
affected after pretreatment of the cells with PTX. Based on
these results, in the remaining experiments of this series,
DDT cells were pretreated for 18 hr with 25 ng/ml of PTX.
Figure 3-16 depicts the effects of WRC-0018 on cAMP
accumulation in control and PTX-pretreated cells. The
biphasic WRC-0018 concentration response in control cells was
replotted from Figure 3-4. After pretreatment of the cells
with PTX, similar to using Ai~AdoR antagonist, the A]_-AdoR
mediated inhibition of cAMP accumulation is blocked and
hence, the downward component of the concentration response
curve of WRC-0018 was abolished. The EC50 value of the A2~AdoR
mediated effect of WRC-0018 on cAMP accumulation was 90 nM.
The maximal stimulation of cAMP accumulation was 143 -fold
above basal in PTX-pretreated cells (control cells, 81 -fold
above basal). The effect of NECA on cAMP accumulation in
control and PTX-pretreated cells is shown in Figure 3-17. The
data for NECA in control cells were replotted from Figure 3-
13. In PTX-pretreated cells, NECA (10 nM 10 |IM) stimulated
cAMP accumulation with an EC50 value of 180 nM. The maximal
response was 5-fold above the basal level. Figure 3-18
illustrates the effect of ADO on cAMP accumulation in control
and PTX-pretreated cells. The data for ADO in control cells
were replotted from Figure 3-14. Over the concentration range

44
of 1 |1M 10 |1M, ADO increased the cAMP level in PTX-
pretreated cells. A maximal response was not achieved even at
10 |1M ADO and hence, the EC50 value of the A2~AdoR mediated
response of ADO could not be calculated.
In summary, these data show that following PTX-
pretreatment, the selective A2~AdoR agonist WRC-0018 causes a
sustained A2~AdoR mediated stimulatory effect on cAMP
accumulation. Likewise, the A2~AdoR mediated stimulatory
effects of the nonselective agonists NECA and ADO were
unmasked after pretreatment of DDT cells with PTX. The
results also indicate that the Ai inhibitory effect of AdoR
agonists was Gi protein mediated.
Effect of Desensitization and/or Down-regulation of Aj_-AdoR
on the A-AdoR Mediated Response
The third experimental approach used to uncover the A2-
AdoR mediated effect on cAMP accumulation involves
desensitization and/or down-regulation of Ai~AdoR. Figure 3-
19 illustrates a representative Scatchard plot of [3H]CPX
binding to cell membranes from control (untreated) and CCPA
(Ai-AdoR agonist)-pretreated cells. The inhibition of ISO (10
fiM) -stimulated cAMP accumulation by 1 (1M CPA was investigated
in cells pretreated with various concentrations of CCPA (0.1
nM 1 |j.M) for 16 hr (data not shown) Cells incubated with
0.1 HM CCPA for 16 hr at 37C followed by four wash cycles
showed a 48% reduction in specific [3H]CPX binding (control,
0.4 pmol/mg protein) with no change in the Kd value for the

45
remaining receptors labeled with [3H]CPX (control, 0.5 nM;
CCPA-pretreated, 0.4 nM).
The concentration-response relationships of CPA, NECA
and WRC-0018 induced attenuation on ISO-stimulated cAMP
accumulation in control and CCPA-pretreated cells are shown
in Figures 3-20, 3-21 and 3-22, respectively. In control
cells, CPA decreased cAMP content in a concentration-
dependent manner with an EC50 value of 1.2 nM and maximal
inhibition of 78% (Figure 3-20). In comparison, after
pretreatment of the cells with CCPA, the EC50 value for CPA-
induced inhibition on ISO-stimulated cAMP accumulation was
increased by 13-fold (EC50, 15 nM) without any significant
change in the maximal inhibition. Pretreatment of the cells
with 0.1 (IM CCPA for a longer period (68 hr) did not cause a
further shift to the right of the concentration response
curve of CPA and had no effect on the maximal response of
this agonist (data not shown). As illustrated in Figure 3-21,
in control cells, NECA decreased cAMP content in a
concentration-dependent manner with an EC50 value of 7.2 nM
and maximal inhibition of 75%. After pretreatment of the
cells with CCPA, the EC50 value for NECA-induced inhibition on
ISO-stimulated cAMP accumulation was increased by 17-fold
(EC50, 120 nM) without any significant change in the maximal
inhibition. The effect of pretreatment of the cells with CCPA
on WRC-0018 induced attenuation on ISO-stimulated cAMP
accumulation is depicted in Figure 3-22. In control cells,
WRC-0018 decreased the cAMP content with an EC50 value of 830

46
nM and caused a maximal inhibition of 77%. After pretreatment
of the cells with CCPA, the inhibition of ISO-stimulated cAMP
accumulation by WRC-0018 was markedly attenuated with a
maximal inhibition of cAMP accumulation of about 20%.
The effects of pretreatment of cells with CCPA on the
A2~AdoR mediated effect of WRC-0018 is shown in Figure 3-23.
The control data from Figure 3-4 were replotted in Figure 3-
23. In CCPA-pretreated cells, the WRC-0018 mediated
attenuation of cAMP accumulation was abolished. The EC50 of
the WRC-0018 mediated stimulation of cAMP accumulation was 17
nM. The maximal stimulation of cAMP accumulation was not
significanly increased (control, 81 -fold above basal; CCPA-
pretreated, 30 -fold above basal). The effects of
pretreatment of the cells with CCPA on the A2~AdoR mediated
response of NECA is shown in Figure 3-24. The data for NECA
in control cells were replotted from Figure 3-13. In CCPA-
pretreated cells, NECA did not stimulate cAMP accumulation,
however, the basal level of cAMP was significantly increased
(p<0.0005) from 32 in control cells to 111 pmol/mg
protein/min in pretreated cells. A longer time of
preincubation of the cells with CCPA (68 hr) also failed to
unmask a NECA mediated A2~AdoR mediated increase in cAMP
accumulation (data not shown). The effects of pretreatment of
the cells with CCPA on the A2~AdoR mediated response of ADO
is shown in Figure 3-25. The data for ADO in control cells
were reploted from Figure 3-14. In CCPA-pretreated cells, ADO
did not stimulate cAMP accumulation.

47
In summary, these data show that after CCPA-
pretreatment, the selective A2~AdoR agonist WRC-0018 causes a
sustained A2~AdoR mediated stimulatory effect on cAMP
accumulation. However, the A2~AdoR mediated stimulatory
effects of the nonselective agonists NECA and ADO were still
not uncovered after pretreatment of DDT cells with CCPA.
Effect of Adenosine on the Desensitization o.f_^Ai~ and^-AdoK
Systems
The endogenous agonist for the AdoR is ADO which has
been shown to be a nonselective agonist (Olsson and Pearson,
1990; Londos et al., 1980). In a separate series of
experiments, I investigated whether ADO had differential
effects on the desensitization of AdoR subtypes and hence,
determined if the expression of the A2~AdoR mediated response
could be uncovered. Cells were pretreated with 100 (J.M ADO to
ensure that both receptor subtypes were stimulated with the
agonist.
Desensitization of Aj^-AdoR
The effect of ADO on the desensitization of Ai-AdoR was
investigated by studying the effect of CPA to inhibit ISO-
stimulated cAMP accumulation in control (untreated) and ADO
(100 |1M, 24 hr)-pretreated cells (Figure 3-26) DIP and EHNA
were present during the period of pretreatment of the cells
with ADO to prevent ADO metabolism and thereby maintain ADO
concentration in the incubation medium relatively constant.
As depicted in Figure 3-26, in control cells, CPA decreased

48
cAMP content in a concentration-dependent manner with an EC50
value of 2.0 nM and maximal inhibition of 77%. In comparison,
after pretreatment of the cells with ADO, the EC50 value for
CPA-induced inhibition on ISO-stimulated cAMP accumulation
was increased by 15-fold (EC50, 30 nM) without any significant
change in the maximal inhibition (pretreated, 73%).
Desensitization of A2~AdoR
The effect of ADO on the desensitization of A2~AdoR were
investigated by studying the effect of WRC-0018 on cAMP
accumulation (Figure 3-27). In control and DIP+EHNA-
pretreated cells, WRC-0018 produced a biphasic concentration
response curve. At low concentrations (1 500 nM), WRC-0018
stimulated cAMP accumulation with a maximal increase of 4-
fold above the basal level. At higher concentrations (>500
nM) of WRC-0018, the cAMP accumulation was attenuated. After
pretreatment of the cells with ADO, the basal level of cAMP
was significantly increased (p<0.025) from 31 to 91 pmol
cAMP/mg protein/min. However, WRC-0018 did not stimulate cAMP
accumulation above the basal level.
The effect of ADO on cAMP accumulation after
pretreatment of the cells with ADO is shown in Figure 3-28.
In both control or ADO-pretreated cells, ADO did not
stimulate cAMP accumulation above the basal level.

49
Di .sens s ion
The interaction of the inhibitory Ai-AdoR and the
stimulatory A2~AdoR was investigated using DDT cells. The co
expression of both AdoR subtypes was first shown in this cell
line by Ramkumar et al. (1990). In addition, UARs which
mediate the stimulation of cAMP accumulation are also
expressed in DDT cells (Norris et al., 1983) .
In the presence of a constant concentration of ISO which
increased cAMP accumulation, several Ai-AdoR agonists
inhibited the ISO stimulatory effect in a concentration-
dependent manner. Furthermore, in the presence of a fixed
concentration of Ai-AdoR agonists, the effect of ISO on cAMP
accumulation was greatly attenuated. The inhibitory effect of
CPA on ISO-stimulated cAMP formation was blocked by 1 )iM CPX,
a selective Ai-AdoR antagonist (data not shown). These
findings indicate that CPA exerts its inhibitory effect on
BAR mediated cAMP accumulation by activating Ai~AdoRs which
are known to be negatively coupled to AC. This Ai-AdoR
mediated inhibition on cAMP formation confirms and extends
the recent reports on AdoRs in DDT cells (Ramkumar et al.,
1990; Gerwins et al., 1990; Gerwins and Fredholm, 1991). One
difference between the present data and the previous reports
is that NECA produced a 97% maximal inhibition of cAMP
accumulation in intact DDT cells in our experiments. However,
only a 30% maximal inhibition of AC activity in DDT cell
membranes by NECA was reported by Ramkumar et al.(1990). The

50
lower maximal inhibition of AC activity in DDT cell membranes
may be due to the homogenization process during membrane
preparation affecting the function of Ai-AdoR.
Our experiments also suggest that the stimulation of
cAMP by low concentrations of WRC-0018 was mediated by an A2-
AdoR. Evidence supporting this include the inhibitory effect
of the AdoR antagonist 8PST and lack of effect of the
selective Ai-AdoR antagonist N-0861 on the stimulatory
response of WRC-0018. Since a selective A2~AdoR antagonist
has yet to be synthesized, the next best candidate, 8PST,
which is only slightly more selective to Ai-AdoR with a
K(A2)/K(Ai) ratio of 5.9 (Trivedi et al., 1990) was used.
Interestingly, low concentrations of WRC-0018 which
increase cAMP accumulation failed to potentiate ISO-
stimulated cAMP accumulation (Figure 3-7). This may be
because A2~AdoR and BAR share the same pool of AC for cAMP
production and the system may be maximally stimulated at 10
¡1M ISO. Also observed was a greater stimulatory effect of ISO
on cAMP accumulation (3 times greater) as compared with WRC-
0018. This finding could be explained by 1) a higher density
of BAR on DDT cells, 2) a higher coupling efficiency of BAR
for the Gs protein or 3) an easier access of BAR to the AC
pool.
The Ai-AdoR agonist CPA inhibited the A2~AdoR
stimulatory effect of WRC-0018 in our experiment suggesting
that the inhibitory Ai-AdoRs dominate over the stimulatory A2-
AdoRs in DDT cells. This dominant role of Ai-AdoR may also

51
explain two observations. First, the assumed nonselective
agonists NECA and ADO showed no stimulation of cAMP
accumulation. At effective concentrations of either agonist,
both AdoR subtypes would be activated, but the Ai-AdoR
mediated inhibition would predominate and hence inhibit the
A2~AdoR mediated stimulation of AC and thereby accumulation
of cAMP. However, this finding is in contrast to the
observation reported by Ramkumar et al. (1990) NECA (10 JIM) ,
in absence of any Ai-AdoR antagonist, produced a 2.1 fold
stimulation of AC activity over the basal level in DDT cell
membranes (Ramkumar et al., 1990). The discrepancy between
this report and our data may be due to the homogenization
process during membrane preparation affecting the function of
Ai-AdoR. The second observation which could be explained by
the dominant role of Ai-AdoR is the downward portion of the
WRC-0018 biphasic concentration response may reflect the
activation of the Ai-AdoR. This was indicated by the
inhibitory effect of WRC-0018 on ISO-stimulated cAMP
accumulation over the same concentration range where the
downward part of the biphasic response of WRC-0018 occurred
(Figure 3-7).
Three approaches were used to study the expression of
A2~AdoR mediated responses and further establish that the
downward phase of the WRC biphasic response is Ai-AdoR
mediated. The first approach involved the use of the
selective Ai-AdoR antagonist N-0861. The selectivity of N-
0861 for the Ai-AdoR was tested by receptor binding and cAMP

52
accumulation. Several lines of evidence support the high
selectivity of N-0861 for Ai-AdoR. These include 1) a
concentration dependent displacement of specific [3H]CPX
binding at the Ai-AdoR site by N-0861 with a Hill slope of 1,
indicating the interaction of this antagonist with a single
class of binding sites, 2) the substantial attenuation of the
CPA mediated inhibition of cAMP accumulation by 10 (1M N-0861,
and 3) the unaltered stimulatory effect of WRC-0018 by N-0861
at concentrations ranging from 0.1 nM 10 (1M.
In the presence of N-0861, the stimulatory response of
WRC-0018 was sustained, i.e., the downward portion of the
biphasic response was abolished. This finding is consistent
with the hypothesis that the downward portion of the WRC-0018
biphasic response was Ai-AdoR mediated. Assuming that the
interaction of N-0861 with the Ai-AdoR and the interaction of
WRC-0018 with Ai~ and A2~AdoR are reversible and competitive,
then in the presence of N-0861, the downward portion of WRC-
0018 concentration response should be shifted to the right.
This was not tested in the present study due to the
insolubility of WRC-0018 to obtain and test concentrations
above 100 |1M. Similar to the results with WRC-0018, blockade
of the Ai-AdoR with N-0861 uncovered a cAMP stimulation
response of the agonists ECA (<100 (1M) and ADO (<5 (1M) .
These results are in keeping with the observation on NECA
mediated A2~AdoR stimulatory response on AC activity reported
by Ramkumar et al. (1990). However, at >10 JIM NECA (Figure 3-
13) and 10 JIM ADO (Figure 3-14), there was a decrease in cAMP

53
accumulation. This attenuation may be due to the higher
concentrations of NECA and ADO overcoming the N-08 61 (10 (1M)
blockade of the Ai-AdoR allowing its activation.
The second approach used to investigate the expression
of A2~AdoR mediated response was uncoupling of the Ai-AdoR
inhibitory effect with PTX. It is well established that PTX
selectively ADP-ribosylates the subunit. This covalent
modification inactivates the Gi protein and uncouples
inhibitory receptors including the Ai-AdoR resulting in loss
of receptor mediated inhibition of AC activity (Gilman, 1987;
Nathanson, 1987; Hazeki and Ui, 1981). Experiments showed
that the inhibition of AC activity by (R)-PIA in DDT cells
was greatly attenuated after pretreatment of the cells with
100 ng/ml PTX for 18 hrs (Ramkumar et al., 1990). In the
present study, the ability of CPA to inhibit ISO-stimulated
cAMP accumulation was used as means to determine the effect
of PTX-pretreatment on the Ai-AdoR mediated inhibition of
cAMP accumulation. PTX (25 ng/ml)-pretreatment for 18 hr was
found to be sufficient to cause complete loss of the CPA
inhibitory effect. In PTX-pretreated cells, the downward
phase of the WRC-0018 biphasic response was completely
eliminated. Furthermore, NECA and ADO which had no effect on
cAMP accumulation in untreated cells stimulated cAMP
accumulation in cells pretreated with PTX. The result with
NECA, i.e., the stimulation of cAMP accumulation in PTX-
pretreated cells, is also consistent with other reports
(Gerwins et al., 1990) That is, NECA (10 (1M) was reported to

54
increase cAMP content 1.5 fold above the basal level in DDT
cells pretreated with 200 ng/ml PTX for 4 hr (Gerwins et al.,
1990). Our data strongly suggest that the downward phase for
WRC-0018 concentration response is due to activation of an
inhibitory receptor. The EC50 values of the A2~AdoR effect of
WRC-0018 in the presence of N-0861 (93 nM) and after PTX-
pretreatment (90 nM) were very similar. In addition,
similarity also exists between the maximal responses of the
A2~AdoR effect of WRC-0018 in the presence of N-0861 (156 -
fold above basal) and after PTX-pretreatment (143 -fold
above basal). These results suggest that N-0861 and PTX have
a similar net blocking effect on the Ai-AdoR mediated
inhibitory action of WRC-0018 and are unlikely to affect the
interaction of WRC-0018 with the A2~AdoR. As a consequence,
the A2~AdoR mediated stimulatory action of WRC-0018 is
sustained to the same extent.
Selective desensitization and/or down-regulation of Ai-
AdoR was the third approach used. Chronic pretreatment (16
hr) of cells with the selective Ai-AdoR agonist CCPA (0.1 |1M)
caused a 48% loss of Ai-AdoR. This indicates that the Ai-AdoR
is down-regulated after chronic pretreatment with a selective
Ai-AdoR agonist. Down-regulation of receptors after long term
exposure to an agonist is a widely reported phenomenon.
However, the loss of Ai-AdoR during down-regulation is
significantly less in comparison with other receptor systems.
For instance, after 1321N1 astrocytoma cells were incubated
with ISO for 12-24 hr, greater than 90% of the BARS were lost

55
from the cells (Doss et al., 1981). In DDT cells, down-
regulation of BAR occurred rapidly with a ti/2 of about 3 hr
and proceeded to 80-85% loss of receptors by 13 hr of
incubation of the cells with 10 |IM epinephrine (Toews, 1987) .
Desensitization of Ai-AdoR system was studied by
examining the concentration responses of CPA, NECA and WRC-
0018 to inhibit ISO-stimulated cAMP accumulation in control
and CCPA-pretreated cells. For CPA (Figure 3-20) and NECA
(Figure 3-21), the concentration response curves were shifted
to the right (10-20 -fold) without any change in the maximal
responses. Longer pretreatment periods did not increase the
magnitude of shift (data not shown) indicating that maximal
desensitization was achieved. The shift in the concentration
response is not surprising because the sensitivity of the
cells to the agonist should decrease as the receptor number
decreases provided little or no change in the affinity of the
agonist for the remaining receptors. The ability of CPA and
NECA to produce the same maximal response in control
(untreated) cells and CCPA-pretreated cells where the
receptor number is decreased by almost 50% may be explained
by the presence of spare Ai-AdoRs. In this situation, the
responsiveness is not directly proportional to receptor
occupancy and the maximal response can be obtained when less
than 100% of the receptors are activated. Many studies have
shown the existence of spare receptors in other systems
(Stephenson, 1956; Nickerson, 1956; Nelson et al., 1986;
Gunst et al., 1989). For example, in guinea pig lung, after

56
=50% of BARs were inactivated by an irreversible antagonist,
the maximal airway responsiveness to ISO was still maintained
(Nelson et al., 1986). In canine trachealis muscle, the
maximal contractile response for acetylcholine was achieved
when only 4% of muscarinic receptors were occupied (Gunst et
al., 1989) .
Interestingly, in control cells, WRC-0018 produced the
same maximal inhibitory response on ISO-stimulated cAMP
accumulation as CPA. However in CCPA-pretreated cells, the
maximal response produced by WRC-0018 was greatly reduced
(Figure 3-22) whereas the maximal response achieved by CPA
was not changed from that in control cells. This indicated
that WRC-0018 acted as a full Ai-AdoR agonist (as compared
with CPA) in control cells whereas it acted as a partial
agonist in pretreated cells. One explanation for this may be
related to the intrinsic efficacy of the agonists. In control
cells, WRC-0018 may need to occupy more receptors than CPA to
achieve its maximal response. Thus, for WRC-0018, there may
be little or no spare Ai-AdoRs and the responsiveness may be
more directly related to receptor occupancy. As the receptor
number decreases, the maximal response for WRC-0018 will be
reduced as was observed after chronic CCPA-pretreatment. A
desensitization-induced change in agonist efficacy has also
been reported for some BAR agonists. In membranes prepared
from untreated L6 skeletal muscle cells, several compounds
acted as full BAR agonists when compared with ISO, but after
desensitization of the BAR system, these compounds acted as

57
partial agonists (Pittman et al., 1984). Alternative
explanations may also be possible. For example, there may be
subpopulations of Ai-AdoRs in DDT cells where CPA is
nonselective and WRC-0018 is selective for only one subtype
which was selectively down-regulated and desensitized by
chronic CCPA-pretreatment. Further experiments will be
necessary to determine the reason for the differential
changes between CPA and WRC-0018 responsiveness after a
decrease in receptor number.
Because CCPA-pretreatment had little effect on the A2-
AdoR mediated stimulation component of the WRC-0018
concentration response (Figure 3-23), this suggests that CCPA
has little or no desensitization effect on the A2~AdoR. On
the other hand, chronic CCPA-pretreatment abolished the
downward phase of the WRC-0018 biphasic concentration
response curve (Figure 3-23). CCPA is a highly selective Ai~
AdoR agonist (Lohse et al., 1988). The loss of the downward
phase of the WRC-0018 response in conjunction with a
desensitization and/or down-regulation of Ai-AdoR is
consistent with the downward phase being Ai~AdoR mediated.
Thus, the elimination of the downward phase may be due to 1)
desensitization and/or down-regulation decreasing the
sensitivity to Ai-AdoR agonists thereby further increasing
the A2~AdoR selectivity of WRC-0018, and 2) the loss of
efficacy of WRC-0018 as an Ai-AdoR agonist in the partially
desensitized state.

58
In contrast to the effects of N-0861 and PTX, chronic
CCPA-pretreatment did not uncover any NECA or ADO stimulation
of cAMP accumulation. This observation can be explained by
the differences in the agonist's sensitivity to produce
inhibition or
stimulation
of cAMP
accumulation.
Summary
of EC50
Values (nM)
Agonists
Ai-AdoR
A2~AdoR
Control
CCPA
(in PTX-pretreated cells)
WRC-0018
830
-
90
NECA
7.2
120
180
ADO
97
-
>1000
CPA
1.2
15
-
In the case of NECA, the estimated EC50 values for the
inhibition (Ai-AdoR effect) and stimulation (A2~AdoR effect)
of cAMP accumulation in untreated cells were 7.2 and 180 nM,
respectively. The latter estimate was derived from the NECA
concentration response curve after PTX-pretreatment and this
value was based on the assumption that PTX-pretreatment had
little or no effect on the potency of this agonist to
stimulate cAMP accumulation. The higher potency of NECA to
inhibit cAMP accumulation indicates that this agonist shows
some degree of selectivity for the Ai-AdoR over the A2~AdoR in
DDT cells. In contrast, NECA has been reported to be a
nonselective agonist in other systems (Hutchinson et al,
1990). After chronic CCPA-pretreatment to partially down-

59
regulate the Ai-AdoR, the EC50 value for the NECA inhibitory
response was increased from 7.2 nM in control cells to 120 nM
in CCPA-pretreated cells which was similar to the EC50 value
for stimulation (180 nM) (see the table above). That is, the
EC50 value for NECA-induced cAMP accumulation (A2~AdoR effect)
is still greater than NECA's EC50 value for inhibition of cAMP
accumulation (Ai-AdoR effect) Similar EC50 values for NECA to
inhibit and stimulate cAMP accumulation in cells chronically
pretreated with CCPA indicates that this agonist would occupy
both receptor subtypes, but activation of the Ai-AdoR would
prevent expression of an A2~AdoR response. It appears that
the expression of an A2~AdoR response in untreated cells
requires a selective A2~AdoR agonist. This was evident with
WRC-0018 which based upon its stimulatory and inhibitory
concentration ranges, showed an approximately 500-600 -fold
selectivity for the A2~AdoR (Figure 3-4) Although chronic
CCPA-pretreatment reduced the sensitivity for the Ai-AdoR
response of NECA, this reduction was not sufficient to result
in activation of the A2~AdoR without significant stimulation
of the Ai-AdoR. A similar explanation may account for the
lack of an ADO mediated stimulation of cAMP accumulation in
cells chronically pretreated with CCPA. The EC50 for ADO to
inhibit and stimulate (in PTX-pretreated cells) cAMP
accumulation was 97 nM and >1 |1M, respectively. This Ai-AdoR
selectivity for ADO is in contrast to previous studies in
other cell types where this agonist has been shown to be
nonselective (Londos et al., 1980; Olsson and Pearson, 1990).

60
Assuming a 10-20 fold reduction in ADO sensitivity (like CPA)
to inhibit cAMP accumulation in cells chronically pretreated
with CCPA, this would not be sufficient to allow selective
expression of the A2~AdoR response. However, it will be of
interest to determine if selective down-regulation of the Ai~
AdoR would reduce the sensitivity of a non-selective agonist
sufficient to uncover expression of the A2~AdoR response.
Interestingly, CCPA-pretreatment was found to increase
the basal cAMP level (Figure 3-24). Desensitization of
inhibitory receptors resulting in an increase in basal cAMP
levels or potentiation of stimulatory receptor effects has
been widely reported. For example, a 3-fold increase in basal
cAMP level was observed after pretreatment of NG108-15 cells
with 10 (1M carbachol for 19 hr (Nathanson et al., 1978). This
increase in basal cAMP level may be due to the loss of an
inhibitory tone on AC activity mediated by the Gi protein.
The inhibitory tone may involve the basal dissociation of Gi
(in the absence of an inhibitory agonist) with the subunit
directly attenuating the activity of the catalytic subunit of
AC (Gilman, 1987) Several mechanisms may be responsible for
the loss of the inhibitory tone. First, chronic CCPA-
pretreatment may decrease the cellular content of Gi protein.
Evidence supporting this contention has been reported by
Green (1987) who showed a decrease in Gi protein after
desensitization of the Ai-AdoR system in primary culture of
rat adipocytes. Second, CCPA-pretreatment may induce an
impairment in the function of Gi. For example, a CCPA-induced

61
phosphorylation of Gi may prevent the basal release of the
activated 0Ci subunit.
Because Ai-AdoR inhibition of cAMP accumulation
dominated over A2~AdoR mediated stimulation, it was of
interest to investigate whether the endogenous agonist ADO
would have a differential desensitization effect on the
receptor subtypes when both were chronically stimulated.
After pretreatment of cells with 100 JIM ADO for 24 hr, the
concentration response of CPA to inhibit ISO-stimulated cAMP
accumulation shifted to the right indicating that the Ai-AdoR
system was desensitized (Figure 3-26). This was similar to
that observed with CCPA-pretreatment. However, WRC-0018 did
not increase cAMP accumulation in ADO-pretreated cells,
suggesting that the A2~AdoR system was also desensitized
(Figure 3-27). In keeping with this conclusion, ADO also did
not stimulate cAMP accumulation in ADO-pretreated cells. The
observation that the Ai-AdoR was still present (albeit at
reduced agonist sensitivity) whereas the A2~AdoR response was
abolished indicates that the extent of desensitization was
different for each receptor subtype. This may be explained by
one or more factors: 1) the density of Ai-AdoR in DDT cells
was higher than that of A2~AdoR. An Ai-AdoR to A2~AdoR ratio
of 4:1 in DDT cells has been reported by Ramkumar et al.
(1990), 2) the coupling efficiency between the receptors and
their second messenger system may be different, 3) the rate
of desensitization of A2~AdoR was much faster than that of Ai-
AdoR. Ramkumar et al. (1991) reported that the ti/2 for the

62
desensitization of Ai- and A2~AdoR was 8 and 0.75 hr in DDT
cells, respectively, 4) the efficacy of the agonist to induce
desensitization of the receptor subtypes may be different, or
5) the mechanism for desensitization of the receptor subtypes
may be different. Evidence for a differential desensitization
mechanism has been reported recently (Ramkumar et al., 1991).
During desensitization, the Ai-AdoR was down-regulated
(internalized), uncoupled from G proteins and phosphorylated
whereas the A2~AdoR was not (Ramkumar et al., 1991) .
Summary
Adenosine is an autocoid produced by the same cells on
which and/or adjacent cells it exerts its effects. The
results of the present study demonstrate that in DDT cells
adenosine acts on at least two receptor subtypes (Ai~ and A2-
AdoR) whose actions result in opposing effects on the
formation of cAMP. There may be circumstances in which
adenosine production rapidly and transiently increases and
thereby maximally activates AdoRs. When the concentration of
adenosine rises far above the physiological range, the
coexistence of two receptor subtypes on the same cell with
opposing functional effects would dampen the responses to the
transient extremes of adenosine concentration. That is, in
the example illustrated in Figure 3-14 (ADO alone),
activation of the Ai-AdoR would attenuate the effect of A2-

63
AdoR activation of AC and thereby dampen a rise in cellular
cAMP.
The data from the present study show that in DDT cells
activation of the inhibitory Ai-AdoR will predominate and
thus mask the stimulatory A2~AdoR response. This observation
may have broader implications because other cell types and
tissues have been reported to express both Ai~ and A2~AdoRs.
These include porcine coronary vascular smooth muscle cells
(Mills and Gewirtz, 1990), FRTL-5 cells derived from normal
rat thyroid (Nazarea et al., 1991), ventricular myocytes from
chick embryo (Xu et al., 1992) and heart tissue (Olsson and
Pearson, 1990) .
Our data demonstrate that cells with both receptor
subtypes can be pharmacologically manipulated to uncover an
A2~AdoR response with a highly selective A2~AdoR agonist or
with the use of a selective Ai-AdoR antagonist. There may be
situations where these pharmacological approaches have
therapeutic potential and suggest possible strategies for
drug development. For example, in cells or tissues where both
receptor subtypes are expressed, the A2~AdoR mediated
responses may be suppressed under normal conditions. However,
by selectively blocking the Ai-AdoR system or activating the
A2~AdoR with a highly selective agonist, A2~AdoR mediated
responses (e.g., vasodilation to increase blood flow,
inhibition of platelet aggregation during thrombosis,
generation of superoxide radicals to prevent reperfusion
injury) may be expressed for possible therapeutic effect.

64
Another implication of our study with potential
therapeutic value is the possibility to decrease the side
effects caused by nonselective AdoR agonists. For instance,
adenosine and its analogs may activate both receptor subtypes
and hence, if one subtype can be selectively blocked, the
side effects mediated by this receptor subtype should be
attenuated.
The development of more selective agonists and
antagonists may be needed to increase the concentration range
over which a response can be achieved. Thus, the ultimate
objective would be to develop specific agonists and
antagonists.
Finally, in cell types in which the Ai~ and A2~AdoR
coexist, the question arises as to whether an A2~AdoR
mediated response initiated by ADO would ever be expressed
under physiological or pathological conditions? Although Ai~
AdoR suppression of the A2~AdoR response may be normal under
most physiological conditions (at least in cells that express
both receptor subtypes), the A2~AdoR response may be needed
under some pathological conditions. For example, under
stress, the chronic release of adenosine or other cellular
mechanisms may result in the loss of the Ai-AdoR or other
components of its signal transduction pathway (e.g., Gi
protein) allowing A2~AdoR expression. Although our data
indicated that chronic stimulation of Ai~ and A2~AdoR resulted
in desensitization of both receptors and Ramkumar et al.
(1991) found that the rate of desensitization for the A2~AdoR

65
was faster than that for the Ai-AdoR in DDT cells, in other
cells, the opposite may occur. Thus, additional studies are
necessary to test this latter possibility, further
characterize the dual modulation of AC by adenosine in the
same cell and/or organ and define the implications for such
role.

66
1. Selective A^AdoR antagonist 2. Pertussis toxin (PTX)
3. Desensitization / Down-regulation of A^AdoR
Figure 3-1. Experimental approaches to express the A2-
AdoR mediated response.

67
<
O
Figure 3-2. Inhibition of ISO-stimulated cAMP accumulation in
DDT cells by CPA, NECA and ADO. Cells were incubated in HBSS
containing 50 (1M rolipram, 10 |1M ISO or ISO plus the
indicated concentrations of CPA, NECA or ADO + 1 JIM DIP + 1
JIM EHNA for 7 min at 36C. At the end of the incubation
period, the cAMP accumulated was determined as described in
the "Methods" section. The control cAMP accumulated in the
presence of ISO alone was 5712, 879 and 684 pmol/mg
protein/min for the CPA, NECA and ADO experiments,
respectively. The basal level of cAMP accumulated was below
the detection level for CPA and NECA and 61 pmol/mg
protein/min for the ADO experiment. Each data point is the
meanSD of quadruplicate determinations and is representative
of 2 experiments.

68
-a £
gE
1 =
E
3 O
8S-
< O)
CL E
o c
CL
Log [Isoproterenol], M
Figure 3-3. The effect of CPA and NECA on ISO-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 |IM rolipram, the indicated concentrations of
ISO, without or with 0.1 }1M CPA or 1 p.M NECA for 7 min at
36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The basal level of cAMP accumulated was 11 pmol/mg
protein/min. Each data point is the meanSE, n=3 5.

69
T¡£
lE
1.1
E£
3 O
O *
o a.
< U)
CL E
11
Log [Drug], M
Figure 3-4. The effect of WRC-0018 and NECA on cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram and the indicated concentrations of
WRC-0018 or NECA for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. The basal level of cAMP
accumulated was 31, 32 pmol/mg protein/min and each data
point is the meanSE, n=15 for WRC-0018 and n=3 for NECA. The
* indicates a p<0.01 for each WRC-0018 data point as compared
to its basal level.

70
o.S

3.S
ES
3 O
8S.
< O)
CL E
i!
o.
25
20
15
10
0
BASAL
WRC-0018
WRC-0018+8PST
Figure 3-5. The effect of 8PST on WRC-0018-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram, without or with 0.1 |1M WRC-0018 or
WRC-0018 plus 5 (1M 8PST for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSD of quadruplicate determination. The indicates a
p<0.01 for the WRC-0018 value compared to the basal value and
a p<0.005 for the WRC-0018 plus 8PST value compared to the
WRC-0018 value.

71
BASAL
WRC-0018
Figure 3-6. The effect of CPA on WRC-0018-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram, without or with 0.1 JIM WRC-0018 or
WRC-0018 plus 1 JIM CPA for 7 min at 3 6C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meaniSD of quadruplicate determination. The indicates a
p<0.005 for the WRC-0018 value compared to the basal value
and for the WRC-0018 plus CPA value compared to the WRC-0018
value.

72
O -
¡|
3 C
£'
32
< ^
S: E
< O
e
a
Log [WRC-0018], M
Figure 3-7. Effect of WRC-0018 on ISO-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 }IM rolipram, the indicated concentrations of
WRC-0018, without or with 10 |1M ISO for 7 min at 36C. At the
end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The control
cAMP accumulated in the presence of ISO alone was 629
pmol/mg protein/min. Each data point is the meantSE, n=4.

73
Figure 3-8. Inhibition of specific [3H]CPX binding to DDT
cell membrane by N-0861. Membranes protein (320 (ig) was
incubated with 1 nM [3H]CPX and the indicated concentrations
of N-0861 for 150 min at room temperature. Specific binding
was determined as described in the "Methods" section. Data
points are the mean of triplicate determinations which varied
by less than 5%. The control specific [3H]CPX binding was 132
fmol/mg protein.

74
-O £
i*
If
§1
< O
OL E
<1
o c
Q.
50
40
30
20
10
0
-10
-8
-6
Log [N-0861], M
Figure 3-9. The effect of N-0861 on the ability of CPA to
inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells
were incubated in HBSS containing 50 JIM rolipram, 10 }1M ISO,
without or with 0.1 |1M CPA or CPA plus the indicated
concentrations of N-0861 for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSD of quadruplicate determinations. The indicates a
p<0.0005 for the ISO + CPA + N-0861 value compared to the ISO
+ CPA value.

75
T3.E
g E
3
E £
3 o
8a
< U)
CL E
<
o c
60
40
20
0
-10 -9 -8 -7 -6 -5
Log [N-0861], M
Figure 3-10. The effect of N-0861 on the ability of NECA to
inhibit ISO-stimulated cAMP accumulation in DDT cells. Cells
were incubated in HBSS containing 50 p.M rolipram, 10 )1M ISO,
without or with 1 |1M NECA or NECA plus the indicated
concentrations of N-0861 for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSD of quadruplicate determination. The indicates a
p<0.0005 for the ISO + NECA + N0861 point from the ISO +
NECA point.

76
-a £
SE
1 =
E 2
3 O
8S-
< O)
CL E
il
Q.
Log [N-0861], M
Figure 3-11. The effect of N-0861 on the ability of WRC-0018
to stimulate cAMP accumulation in DDT cells. Cells were
incubated in HBSS containing 50 JIM rolipram, 0.1 p.M WRC-0018,
without or with the indicated concentrations of N-0861 for 7
min at 36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The basal level of cAMP accumulated was 61 pmol/mg
protein/min. Each data point is the meaniSD of quadruplicate
determinations.

77
3 O
8S-
< O)
CL E
CL
Log [WRC-0018], M
Figure 3-12. The effect of N-0861 on WRC-0018-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram, the indicated concentrations of
WRC-0018, without or with 10 |1M N-0861 for 7 min at 36C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The dose
response curve of WRC-0018 in the absence of N-0861 is taken
from Figure 3-4. The basal level of cAMP accumulated in the
presence of N-0861 was 21 pmol/mg protein/min. Each data
point is the meanSE, n=3

78
u.E
is
E 2
3 O
8*
< U)
CL E
o £
Q.
Figure 3-13. The effect of N-0861 on NECA-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 [1M rolipram, the indicated concentrations of
NECA, without or with 10 N-0861 for 7 min at 36C. At the
end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The basal
level of cAMP accumulated was 11 pmol/mg protein/min. Each
data point is the meanSD of quadruplicate determinations and
is representative of 2 experiments. The indicates a
p<0.0005 for the NECA points from their respective control
values.

79
"O £
2e
1 =
E£
3 O
8S.
< O)
CL E
il
o c
Log [Adenosine], M
Figure 3-14. The effect of N-0861 on ADO-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 ^IM rolipram, 1 (IM DIP, 1 (IM EHNA, the indicated
concentrations of ADO, without or with 10 |1M N-0861 for 7 min
at 36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The basal level of cAMP accumulated was 60 pmol/mg
protein/min. Each data point is the meanSE, n=4. The *
indicates a p<0.01 for the ADO points from their respective
control values.

80
"o£
se
11
ES
3 O
8 o.
< O)
o. E
o E
o.
Figure 3-15. The effect of PTX pretreatment on the ability of
CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells.
Cells were incubated in fresh growth media with the indicated
concentrations of PTX for 18 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached from the plate. The cells were then
incubated in HBSS containing 50 (1M rolipram, without or with
10 |1M ISO or ISO plus 1 [IM CPA for 7 min at 36C. At the end
of the incubation period, the cAMP accumulated was determined
as described in the "Methods" section. Each data point is the
meanSD of quadruplicate determination.

81
Log [WRC-0018], M
Figure 3-16. The effect of PTX pretreatment on WRC-0018-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in growth media without or with 25 ng/ml PTX for 18
hr at 37C. At the end of the incubation period, the cells
were washed 4 times with ice-cold HBSS and detached. Cells
were then incubated in HBSS containing 50 |1M rolipram and the
indicated concentrations of WRC-0018 for 7 min at 36C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The dose
response curve of WRC-0018 in control cells is taken from
Figure 3-4. The basal level of cAMP accumulated was 31 and
21 pmol/mg protein/min for the control and PTX experiments,
respectively. Each data point is the meanSE, n=6.

82
Log [ECA], M
Figure 3-17. The effect of PTX pretreatment on NECA-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 25 ng/ml PTX
for 18 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 (1M rolipram
and the indicated concentrations of NECA for 7 min at 36C.
At the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The data
for NECA in control cells are taken from Figure 3-13. The
basal level of cAMP accumulated was 11 and 42 pmol/mg
protein/min for the control and PTX experiments,
respectively. Each data point is the meanSD of quadruplicate
determinations and is representative of 2 experiments.

83
Log [Adenosine], M
Figure 3-18. The effect of PTX pretreatment on ADO-stimulated
cAMP accumulation in DDT cells. Cells were incubated in fresh
growth media without or with 25 ng/ml PTX for 18 hr at 37C.
At the end of the incubation period, the cells were washed 4
times with ice-cold HBSS and detached. Cells were then
incubated in HBSS containing 50 [iM rolipram, 1 [IM DIP, 1 (1M
EHNA and the indicated concentrations of ADO for 7 min at
36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The control data are taken from Figure 3-14. The
basal level of cAMP accumulated was 60 and 41 pmol/mg
protein/min for the control and PTX experiments,
respectively. Each data point is the meantSE, n=3. The *
indicates a p<0.05 for the ADO point from their respective
control value.

84
Figure 3-19. Scatchard plot of specific [3H]CPX binding to
DDT cell membranes after CCPA treatment. Cells were incubated
in fresh growth media without or with 0.1 (1M CCPA for 16 hr
at 37C. At the end of the incubation period, the cells were
washed 4 times with ice-cold HBSS and the membranes prepared.
Membrane protein (0.1 mg) was assayed with 0.06 4 nM
[3H]CPX as described in the "Methods" section. The data are
plotted as the ratio of the amount of specifically bound
ligand (pmol/mg protein) to free ligand (pmol/L) versus the
amount of specifically bound ligand/mg protein. Data points
are the mean of triplicate determination and are
representative of 3 experiments.

85
Log [CPA], M
Figure 3-20. The effect of CCPA pretreatment on the ability
of CPA to inhibit ISO-stimulated cAMP accumulation in DDT
cells. Cells were incubated in fresh growth media without or
with 0.1 JIM CCPA for 16 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached. The cells were then incubated in HBSS
containing 50 |IM rolipram, 10 flM ISO or ISO plus the
indicated concentrations of CPA for 7 min at 36C. The cAMP
accumulated was determined as described in the "Methods"
section. Each data point is the meanSD of quadruplicate
determinations and is representative of 2 experiments.

86
Log [ECA], M
Figure 3-21. The effect of CCPA pretreatment on the ability
of NECA to inhibit ISO-stimulated cAMP accumulation in DDT
cells. Cells were incubated in fresh growth media without or
with 0.1 (1M CCPA for 16 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached. The cells were then incubated in HBSS
containing 50 JIM rolipram, 10 JIM ISO or ISO plus the
indicated concentrations of NECA for 7 min at 36C. The cAMP
accumulated was determined as described in the "Methods"
section. Each data point is the meanSD of quadruplicate
determination.

87
Log [WRC-0018], M
Figure 3-22. The effect of CCPA pretreatment on the ability
of WRC-0018 to inhibit ISO-stimulated cAMP accumulation in
DDT cells. Cells were incubated in fresh growth media without
or with 0.1 (1M CCPA for 16 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached. The cells were then incubated in HBSS
containing 50 ^.M rolipram, 10 |1M ISO or ISO plus the
indicated concentrations of WRC-0018 for 7 min at 36C. The
cAMP accumulated was determined as described in the "Methods"
section. Each data point is the meanSD of quadruplicate
determinations and is representative of 2-4 experiments.

88
Log [WRC-0018], M
Figure 3-23. The effect of CCPA pretreatment on WRC-0018-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 0.1 (1M CCPA
for 16 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 (1M rolipram
and the indicated concentrations of WRC-0018 for 7 min at
36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The dose response curve of WRC-0018 in control cells
is taken from Figure 3-4. The basal level of cAMP accumulated
was 31 and 61 pmol/mg protein/min for the control and CCPA
experiments, respectively. Each data point is the meanSE,
n=3.

89
Log [ECA], M
Figure 3-24. The effect of CCPA pretreatment on NECA-
stimulated cAMP accumulation in DDT cells. Cells were ,
incubated in fresh growth media without or with 0.1 [1M CCPA
for 16 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 |1M rolipram
and the indicated concentrations of NECA for 7 min at 36C.
At the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The data
for NECA in control cells are taken from Figure 3-13. The
basal level of cAMP accumulated was 32 and 111 pmol/mg
protein/min for the control and CCPA experiments,
respectively. Each data point is the meanSD of quadruplicate
determinations and is representative of 2 experiments.

90
Log [Adenosine], M
Figure 3-25. The effect of CCPA pretreatment on ADO-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 0.1 |1M CCPA
for 16 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 |1M rolipram,
1 p.M DIP, 1 p.M EHNA and the indicated concentrations of ADO
for 7 min at 36C. At the end of the incubation period, the
cAMP accumulated was determined as described in the "Methods"
section. The control data are taken from Figure 3-14. The
basal level of cAMP accumulated was 60 and 73 pmol/mg
protein/min for the control and CCPA experiments,
respectively. Each data point is the meanSE, n=3.

91
Log [CPA], M
Figure 3-26. The effect of ADO pretreatment on the ability of
CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells.
Cells were incubated in fresh growth media with 5 |1M DIP, 1
(1M EHNA, without or with 100 (1M ADO for 24 hr at 37C. At the
end of the incubation period, the cells were washed 4 times
with ice-cold HBSS and detached. The cells were then
incubated in HBSS containing 50 |!M rolipram and 10 [1M ISO or
ISO plus the indicated concentrations of CPA for 7 min at
36C. The cAMP accumulated was determined as described in the
"Methods" section. Each data point is the meanSE, n=5.

92
T3.E
|e
1 =
ES
D O
8*
< O)
CL £
o E
Log [WRC-0018], M
Figure 3-27. The effect of ADO pretreatment on WRC-0018-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 5 (1M DIP + 1
M-M EHNA or 5 (XM DIP + 1 JiM EHNA + 100 }1M ADO for 24 hr at
37C. At the end of the incubation period, the cells were
washed 4 times with ice-cold HBSS and detached. Cells were
then incubated in HBSS containing 50 |1M rolipram and the
indicated concentrations of WRC-0018 for 7 min at 36C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The basal
level of cAMP accumulated was 32, 31, 91 pmol/mg
protein/min for the control, DIP+ENHA and ADO+DIP+EHNA
experiments, respectively. Each data point is the meanSE,
n=3.

93
15
o.E
1 =
ES
D O
ga
< O)
CL E
11
Q.
10
0
CONTROL
ADO+D1P+EHNA
10
-9
-8
-7
Log [Adenosine], M
Figure 3-28. The effect of ADO pretreatment on ADO-stimulated
cAMP accumulation in DDT cells. Cells were incubated in fresh
growth media without or with 5 p.M DIP + 1 |IM EHNA + 100 (1M
ADO for 24 hr at 37C. At the end of the incubation period,
the cells were washed 4 times with ice-cold HBSS and
detached. Cells were then incubated in HBSS containing 50 p.M
rolipram, 1 (1M DIP, 1 (1M EHNA and the indicated
concentrations of ADO for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. The basal level of cAMP
accumulated was 61, 61 pmol/mg protein/min for the control
and ADO+DIP+EHNA experiments, respectively. Each data point
is the meanSD of quadruplicate determination.

CHAPTER 4
INHIBITORY EFFECT OF Ai-ADENOSINE RECEPTOR
ON IRREVERSIBLE ACTIVATION OF THE 15-ADRENORECEPTOR
Introduction
The guanine nucleotide binding proteins Gs and Gi play
an important role in mediating receptor activation and
inhibition of AC activity, respectively. Gs and Gi contain
distinct a (as and ai) and common 5 and y subunits.
Activation of either receptor type results in the exchange of
GTP for bound GDP followed by dissociation of the G protein
subunits. Although the mechanism by which ots mediates
stimulation of AC is well understood, less clear is the
mechanism (s) involved in the inhibition of the enzyme by oci.
Several mechanisms may be involved in the inhibition. That
is, inhibition may include: 1) the direct interaction of
activated OCi with the catalytic subunit of the enzyme, or 2)
5+Y from Gi complexing with activated a3, thus favoring the
undissociated and inactive form of Gs. Evidence for both
pathways has been reported (Katada et al., 1984a, b; Jakobs
and Schultz, 1983; Roof et al., 1986). An alternative
mechanism may involve prevention of stimulatory signal
transduction at the receptor level. In this latter case,
inhibition may be mediated by preventing the stimulatory
receptor from forming a ternary complex composed of the
94

95
agonist, receptor and Gs protein. This complex appears to be
a prerequisite for receptor mediated activation of the enzyme
(DeLean et al., 1980; Kent et al., 1980; Gilman, 1987).
Evidence supporting this mechanism has been recently reported
by Romano et al. (1988; 1989). They showed that binding of
ISO to the high affinity binding state of the BAR was
eliminated in the presence of PIA, an Ai-AdoR agonist.
Although these data were interpreted to indicate that
activation of the inhibitory Ai-AdoR prevented the BAR from
forming an agonist high affinity binding state (ternary
complex), it can not be ascertained whether the inhibitory
effect is mediated at the receptor or the G protein level. At
the receptor level, inhibitory agonists could theoretically
initiate alterations in the stimulatory receptor affecting
agonist-receptor interaction or prevent the agonist-receptor
complex from becoming active and thereby impairing signal
transduction.
In 1987, Milecki et al. (1987) reported on the synthesis
and partial characterization of an alkylating carbostyril
derivative (C-Br). This compound and several of its congeners
were shown to be potent BAR agonists. In a subsequent study
using membranes from rat reticulocytes, Standifer et al.
(1989) showed that compared to ISO, C-Br was a full agonist
and its interaction with the receptor was resistant to the
effects of a guanine nucleotide. That is, the nonhydrolyzable
guanine nucleotide derivative Gpp(NH)p only slightly reduced
the ability of C-Br to inhibit [125I]CYP binding (Standifer et

96
al., 1989). In addition, C-Br bound to the receptor in an
irreversible manner and produced an antagonist-insensitive
activation of AC activity. This suggested that C-Br is an
irreversible agonist at the BAR.
DDT cells have been shown to express both BAR and Ai~
AdoR. As shown in Chapter 3 of this dissertation and other
reports (Ramkumar et al., 1990; Nordstedt and Fredholm, 1990;
Ramkumar et al., 1991), BAR mediated cAMP accumulation in DDT
cells is inhibited by Ai-AdoR agonists. To obtain additional
insight into the hypothesis that inhibition is mediated at
the stimulatory receptor level, the ability of the Ai-AdoR to
attenuate the cAMP response mediated by a permanently
activated BAR was investigated.
Results
Ai-AdoR Mediated Inhibitory Effects on BAR-stimulated cAMP
Accumulation by ISO and C-Br
Initial experiments were designed to determine the
effect of CPA on ISO- and C-Br-stimulated cAMP accumulation
in intact DDT cells. As shown in Figure 4-1, ISO and C-Br
stimulated cAMP accumulation in a concentration-dependent
manner with maximal effects of 29- and 32-fold above the
basal level, respectively. The maximal responses to ISO and
C-Br could not be readily compared because experiments were
performed on different batches of cells. The EC50 value for C-
Br was 0.2 nM and 2.7 nM for ISO. Both ISO and C-Br

97
stimulated cAMP accumulation in a concentration-dependent
manner, however, in the presence of 1 flM CPA, the maximal
response produced by ISO and C-Br was decreased by 86% and
75%, respectively. Figure 4-2 depicts the effect of CPA on
cAMP accumulation stimulated by 10 HM ISO and 1 (1M C-Br,
concentrations which gave a maximal response. CPA attenuated
the effect of ISO and C-Br in a concentration-dependent
manner with the same potency (EC50 = 2.3 nM) and the same
maximal response (89% inhibition).
Figure 4-3 illustrates the effect of the Ai-AdoR
antagonist CPX on CPA mediated inhibition of ISO- and C-Br-
stimulated cAMP accumulation. ISO (10 (1M) and C-Br (1 JIM)
(concentrations which caused maximal stimulation of cAMP
accumulation, see Figure 4-1) increased cAMP accumulation to
a similar extent (27 and 28-fold above the basal level,
respectively) and CPA (1 |IM) inhibited their stimulation (81%
for C-Br and 84% for ISO). CPX (5 M-M) an A]_-AdoR antagonist,
largely blocked the inhibitory effect of CPA, and the cAMP
stimulations caused by C-Br and ISO were restored by 96% and
84%, respectively. This indicates that the inhibitory effect
of CPA on C-Br and ISO-stimulated cAMP accumulation is Ai-
AdoR mediated.
The effect of CPA on the basal level of cAMP in DDT
cells is shown in Figure 4-4. CPA decreased the basal level
of cAMP with an EC50 of 6.1 nM and a maximal inhibition of
82% .

98
Effect of CPA on the Insurmountable Component of C-3r-induc.£ii
Stimulation
Figure 4-5 illustrates the time course of C-Br-induced
increase in cAMP content and the effects of PROP and CPA. C-
Br increased the cAMP content over a 10 min incubation
period. After 3 min of incubation with C-Br, the addition of
PROP (20 (IM) had no effect on the subsequent rate of cAMP
accumulation. In contrast, the addition of PROP and CPA (1
(1M) 3 min after incubation with C-Br resulted in complete
inhibition of further cAMP accumulation during the next 7 min
of incubation. In fact, after addition of CPA, cAMP levels
decreased.
The effect of CPA on the irreversible binding of C-Br to
the BAR is depicted in Figure 4-6. This is a representative
Scatchard plot of specific [125I]CYP binding after
pretreatment of cells with C-Br, CPA and C-Br + CPA for 10
min at 37C. CPA pretreatment did not alter the specific
[125I]CYP binding as compared to control (control, 54 pmol/ mg
protein; CPA, 59 pmol/mg protein). After incubation of cells
with 1 (IM C-Br followed by 6 cell wash cycles, there was a
42% decrease in specific [125I]CYP binding and this decrease
was not affected by the presence of CPA during the 10 min
preincubation period (C-Br, 28 pmol/mg protein; C-Br + CPA,
30 pmol/mg protein). There was also no change in the Kq value
for [125I]CYP among the 4 groups (control, 34 pM; C-Br, 32 pM;
CPA, 25 pM; C-Br + CPA, 26 pM).

99
Figure 4-7 illustrates the effects of Gpp(NH)p and CPA
on the concentration response of ISO to compete with [125I]CYP
for the BAR binding site. ISO alone displaced specific
[125I]CYP binding with an IC50 of 60 nM and a Hill slope of
0.58. In the presence of Gpp(NH)p (100 |1M) the displacement
curve shifted to the right with an IC50 value of 406 nM and
steepened with a Hill slope of 0.80. In contrast to Gpp(NH)p,
CPA had no effect on either the IC50 value (70 nM) or the Hill
slope (0.60) of the ISO displacement curve as compared with
ISO alone.
Discussion
Receptor mediated activation of AC activity is a process
with multiple steps that involve an agonist, receptor, a
stimulatory guanine nucleotide binding protein (Gs) and the
catalytic subunit of AC (Levitzki, 1987; Rodbell, 1980).
General agreement now exists that agonists promote the
formation of a ternary complex consisting of the agonist,
receptor and Gs and this interaction accelerates the exchange
of GDP for GTP at G3. The binding of the agonist in the
ternary complex is of high affinity. GTP destabilizes the
ternary complex resulting in a decrease in agonist affinity
and the release of an activated as subunit of Gs which
directly stimulates the catalytic subunit of AC. Hydrolysis
of GTP to GDP deactivates as and AC activity returns to the
basal level (Gilman, 1987). In contrast to AC activation,

100
much less is known about the mechanism whereby receptor
mediates inhibition of the enzyme. Similar to stimulatory
receptors, inhibitory receptors are coupled to a guanine
nucleotide binding protein (Gj.) which dissociates into an
and IJ+Y subunits in the presence of GTP. There is evidence
supporting several inhibitory mechanisms which include 1) a
direct interaction of OCi with the catalytic subunit of the AC
(Katada et al., 1984b; Jakobs and Schultz, 1983; Roof et al.,
1986), 2) fi+Y subunits released from Gi complexing with as to
favor the inactive Gs complex (Katada et al., 1984a) and 3)
prevention of stimulatory ternary complex formation (Romano
et al., 1988; Romano et al., 1989).
In the present study, the effect of the inhibitory Ai~
AdoR to attenuate the response of a permanently activated 1AR
was investigated. Initial experiments were performed to
partially characterize the actions of the irreversible BAR
agonist C-Br, using intact DDT cells. This compound was found
to be a potent stimulator of cAMP accumulation, with a
concentration required to produce half-maximal stimulation in
the subnanomolar range. In addition, C-Br was a full BAR
agonist since it produced the same maximal response as the
classical agonist, ISO. The potency and efficacy of C-Br
using intact cells is consistent with a previous report on
the effect of this compound to stimulate AC activity in
isolated membranes from rat reticulocytes (Standifer et al.,
1989).

101
Similar to ISO, C-Br stimulated cAMP accumulation in DDT
cells was markedly attenuated by the selective Ai-AdoR
agonist CPA. This attenuation was not reversed by high
concentrations of either BAR agonists but was largely blocked
by the selective Ai-AdoR antagonist CPX. These data indicated
that the inhibition of C-Br stimulated cAMP accumulation was
mediated by an Ai-AdoR. Furthermore, the concentration of CPA
to inhibit cAMP accumulation by equipotent and equieffeetive
concentrations of ISO and C-Br was similar suggesting that
activation of the BAR by C-Br did not alter the the
inhibitory effect of CPA.
In the experiments discussed above, the incubation time
used was relatively short (6 min). Because the irreversible
binding of C-Br to BAR was time-dependent (Standifer et al.,
1989), the responses observed at the end of short periods of
incubation were probably due to both reversible and
irreversible interactions of this agonist with the BAR. To
investigate the irreversible agonist effects of C-Br, the BAR
antagonist PROP was employed to compete off any C-Br which
was reversibly bound to the BAR. After a 3 min incubation
with C-Br alone, the addition of an excess of PROP had no
subsequent effect on the rate of cAMP formation as compared
to that in the presence of C-Br alone. In contrast, when PROP
and CPA were added after a 3 min incubation with C-Br alone,
further accumulation of cAMP was inhibited. Although CPA
attenuated C-Br stimulated cAMP accumulation, it did not
affect the irreversible binding of C-Br to the BAR. This

102
indicates that even though C-Br produced an antagonist-
insensitive response once bound to the BAR, activation of the
inhibitory Ai-AdoR resulted in an attenuation of cAMP
accumulation mediated by the permanently activated BAR.
There is some evidence suggesting that one potential
mechanism for receptor mediated inhibition of AC activity may
involve signal transduction at the stimulatory receptor
level. In 1978, Watanabe et al. reported that activation of
the cardiac muscarinic inhibitory receptor decreased the
affinity of ISO for the BAR. Likewise, more recently, Romano
et al. (1988; 1989) showed the Ai-AdoR agonist PIA reduced
the affinity of ISO for the BAR in rat cardiac membranes.
This reduction in agonist affinity was similar to that
observed in the presence of Gpp(NH)p (Romano et al., 1988;
Romano et al., 1989). Because the guanine nucleotide
sensitive agonist high affinity binding state of the receptor
has been proposed to represent the ternary complex formation
(Gilman, 1987), it was suggested that PIA prevented the
stimulatory receptor from forming a ternary complex, a
necessary prerequisite for the activation of AC. However, in
the present study, CPA did not alter the interaction of ISO
with the BAR even though Gpp(NH)p did reduce agonist
affinity. This indicated that CPA had no effect on ternary
complex formation. The discrepancy between this observation
and that reported by Romano et al. (1988) may be due to
different cell types expressing different modes of
inhibition. The inhibition of ternary complex formation may

103
be operative in cardiac cells whereas other inhibitory
mechanisms may exist in DDT cells. Alternatively, during the
process of DDT cell membrane preparation, the inhibitory
effects on agonist binding may have become uncoupled.
The pathway of receptor activation initially involves a
reversible interaction of the agonist (Ag) with the receptor
(R) to form an Ag-R complex. The Ag-R complex then undergoes
a conformational change to form an activated Ag-R complex
which interacts with Gs. Although difficult to establish
directly, there is some intriguing kinetic and structural
evidence to suggest that the BAR undergoes a conformational
change induced by agonist (Contreras et al., 1986; Pedersen
and Ross, 1985) An inhibitory mechanism could therefore
involve preventing the Ag-R complex from becoming
conformationally active. However, the data with the
irreversible agonist are not consistent with this proposed
mechanism. C-Br produced antagonist-insensitive activation of
the BAR suggesting that the Ag-R complex is in a permanently
activated state, but the stimulation of cAMP accumulation is
still inhibited by CPA. If CPA mediated inhibition involved
inactivation of the BAR, then it would be expected that CPA
would have no effect on the irreversible component of C-Br-
stimulated cAMP formation.
Data from the present study are also consistent with an
inhibitory action beyond the receptor level. The basal cAMP
accumulated in the absence of a BAR agonist was also
inhibited by CPA. Furthermore, in experiments not shown, CPA

104
inhibited forskolin-stimulated cAMP accumulation in DDT
cells. Forskolin activates AC in a receptor-independent
manner which may involve direct interaction with the
catalytic subunit of the enzyme (Seamon and Daly, 1986).
Although the mechanism for Ai-AdoR mediated inhibition of
basal and forskolin-stimulated cAMP accumulation in DDT cells
has not been fully delineated, evidence from other cell lines
suggest that the subunits of Gi may be involved. Thus, OCi
released by activation of the Ai-AdoR may directly attenuate
the activity of the catalytic subunit of the enzyme, or the
13+Y subunits of Gi may complex with free as to form the
inactive Gs complex (Katada et al., 1984a, b; Jakobs and
Schultz, 1983; Roof et al., 1986). It is conceivable that
inhibition may involve more than one mechanism acting in
concert or a single mechanism may dominate depending on the
mode of enzyme activation. Further studies will be necessary
to elucidate these possibilities.
Over the past few years, significant effort has been
expended in developing insurmountable receptor ligands as
investigative tools (Posner et al., 1984; Homburger et
al.,1984; Nelson et al., 1986; Baker et al., 1986) and more
recently for therapeutics (Ullman and Svedmyr, 1988). These
agents have the advantage of very slow dissociation rates
from their receptors and hence may produce a long duration of
action. In addition, once the drug-receptor complex is
formed, the response will be independent of the plasma
concentration which can be allowed to fall. This may

105
translate into a reduced incidence of side effects. In the
case of agonists, the recently introduced BAR agonist
salmeterol has been shown to produce a sustained airway
dilation probably due to an extremely tight binding to the
BAR (Brittain et al., 1981; Oilman and Svedmyr, 1988). On the
other hand, one potential problem with slow dissociating
agonists is modulating the response produced under conditions
where the receptor is activated in an antagonist-insensitive
manner. Therefore, administration of an antagonist may have
no effect on the response. The data from the present study
suggest that the sustained response mediated by C-Br can be
controlled by activation of inhibitory receptor. Further
development of agonists that produce sustained responses may
need to take into account the interaction of inhibitory and
stimulatory receptors to fully exploit their therapeutic
potential.

106
Figure 4-1. The effect of CPA on ISO- and C-Br-stimulated
cAMP accumulation in DDT cells. Cells were incubated in HBSS
containing 100 (1M rolipram, the indicated concentrations of
ISO or C-Br, without or with 1 JIM CPA for 6 min at 37C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The basal
level of cAMP accumulated was 226, 143 pmol/mg protein/6
min for the ISO and C-Br experiments, respectively. Each data
point is the meanSE, n=4.

107
Log [CPA], M
Figure 4-2. Inhibition of ISO- and C-Br-stimulated cAMP
accumulation in DDT cells by CPA. Cells were incubated in
HBSS containing 100 [1M rolipram, 10 ^M ISO or 1 |1M C-Br and
the indicated concentrations of CPA for 6 min at 37C. At the
end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The control
cAMP accumulated in the presence of ISO or C-Br alone was
64622, 66029 pmol/mg protein/6 min, respectively. The basal
level of cAMP accumulated was 194 pmol/mg protein/6 min.
Each data point is the meanSE, n=4.

108
a
"5 .E
a
i2
3a
O U)
< E
i!
< Q.
O <3
BASAL
C-Br
ISO
^ C-Br+CPA
m iso+cpa
C-Br+CPA+CPX
m iso+cpa+cpx
Figure 4-3. The effect of CPX on the inhibition of ISO- and
C-Br-stimulated cAMP accumulation in DDT cells by CPA. Cells
were incubated in HBSS containing 100 JIM rolipram, 1 p.M C-Br
10 ISO, C-Br + 1 JIM CPA, ISO + 1 |IM CPA, C-Br + CPA + 5 JIM
CPX, or ISO + CPA + 5 JIM CPX for 6 min at 37C. At the end of
the incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSE, n=4.

109
Figure 4-4. The effect of CPA on basal cAMP accumulation in
DDT cells. Cells were incubated in HBSS containing 100 |1M
rolipram, without or with the indicated concentrations of CPA
for 6 min at 37C. At the end of the incubation period, the
cAMP accumulated was determined as described in the "Methods"
section. The basal control level of cAMP accumulated was 262
pmol/mg protein/6 min. Each data point is the meanSE, n=4.

110
Time (min)
Figure 4-5. The effect of CPA on C-Br-stimulated cAMP
accumulation in DDT cells. Cells were preincubated in HBSS
containing 100 JIM rolipram and 1 (1M C-Br at 37C. After 3 min
of incubation, 20 p.M propranolol or propranolol plus 1 |1M CPA
was added. At the times indicated, the cAMP content was
determined as described in the "Methods" section. Each data
point is the meanSE, n=4.

Ill
Figure 4-6. Scatchard plot of specific [125I]CYP binding to
DDT cell membranes after pretreatment with C-Br, CPA or C-
Br+CPA. Cells were incubated in HBSS without or with 1 |IM C-
Br, 1 JIM CPA or l^lM C-Br + 1 JIM CPA for 10 min at 37C. At
the end of the incubation period, the cells were harvested
and membranes prepared. Membrane protein was assayed with 6 -
100 pM [125I]CYP as described in the "Methods" section. The
data are plotted as the ratio of the amount of specifically
bound ligand (pmol/mg protein) to free ligand (pmol/L) versus
the amount of specifically bound ligand/mg protein. Data
points are the mean of triplicate determination and are
representative of 3 experiments.

112
Log [Isoproterenol], M
Figure 4-7. Inhibition of specific [125I]CYP binding to DDT
cell membrane by ISO in the presence of Gpp(NH)p or CPA.
Membrane protein (50 (lg) was incubated with 30 pM [125I]CYP,
the indicated concentrations of ISO, without or with 100 |1M
Gpp(NH)p or 1 JIM CPA for 60 min. at 37C. Data points are the
mean of triplicate determination which varied by less than 6%
and are representative of 3 experiments. The control specific
[12^I]CYP binding was 53 fmol/mg protein.

LIST OF REFERENCES
Baker, S. P., Liptak, A. and Pitha, J. (1985) J. Biol. Chem.
260: 15820-15828
Baker, S. P., Sumners, C., Pitha, J. and Raizada, M. K.
(1986) J. Neurochem. 47: 1318-1326
Belardinelli, L., Giles, W. R. and West, G. A. (1988) J.
Physiol. (Lond.) 405: 615-633
Belardinelli, L., Isenberg, G. (1983) Am. J. Physiol. 224:
H734-H737
Belardinelli, L and Lerman, B. B. (1990) Br. Heart J. 63: 3-4
Belardinelli, L., Linden, J. and Berne, R. M. (1989) Prog.
Cardiovasc. Dis. 32(1): 73-97
Berne, R. M. (1963) Am. J. Physiol. 204: 317-322
Berne, R. M. (1986) News Physiol. Sci. 1: 163-167
Biaggioni, I., Killian, T. J., Mosqueda-Garcia, R.,
Robertson, R. M. and Robertson, D. (1991a) Circulation 83:
1668-1675
Biaggioni, I., Paul, S., Puckett, A. and Arzubiaga, C.
(1991b) J. Pharmacol. Exp. Ther. 258(2): 588-592
Brittain, R. T., Dean, C. M. and Jack, D. (1981) Int. Encycl.
Pharmacol. Ther. 104: 613-652
Brown, J. H. (1979) J. Cyclic Nucleotide Res. 5(6): 423-433
Bruns, R. F. (1990) Ann. N.Y. Acad. Sci. 603: 211-226
Bruns, R. F., Daly, J. W. and Snyder, S. H. (1980) Proc.
Natl. Acad. Sci. USA 77(9): 5547-5551
Bruns, R. F., Fergas, J. H., Badger, E. W., Bristol, J. A.,
Santay, L. A., Hartman, J., Hays, S. J. and Huang, C. C.
(1987) Naunyn-Schmiedebergs Arch. Pharmacol. 335: 59-63
Bruns, R. F., Lu, G. H. and Pugsley, T. A. (1986) Mol.
Pharmacol. 29: 331-346
113

114
Contreras, M. L., Wolfe, B. B. and Molinoff, P. B. (1986) J.
Pharmacol. Exp. Ther. 239: 136-143
Cooper, D. M. F. and Londos, C. (1979) J. Cyclic Nucleotide
Res. 5: 289-302
Daly, J. W., Butts-Lamb, P. and Padgett, W. (1983) Cell. Mol.
Neurobiol. 3: 69-80
Delahaba, G and Cantoni, G. L. (1959) J. Biol. Chem. 234:
603-608
Delahunty, T. M., Cronin, M. J. and Linden, J. (1988)
Biochem. J. 255: 69-77
DeLean, A., Stadel, J. M. and Lefkowitz, R. J. (1980) J.
Biol. Chem. 255: 7108-7117
Doss, R. C., Kramarcy, N. R., Harden, T. K. and Perkins, J.
P. (1985) Mol. Pharmacol. 27: 507-516
Doss, R. C., Perkins, J. P. and Harden, T. K (1981) J. Biol.
Chem. 256: 12281-12286
Drury, A. N. and Szent-Gyorgyi, A. (1929) J. Physiol. (Lond.)
68: 213-237
Endoh, M., Maruyama, M. and Iijima, T. (1985) Am. J. Physiol.
249: H309-H320
Endoh., M., Muruyama, M. and Taira, N. (1983) Physiology and
Pharmacology of Adenosine Derivatives, ed. by Daly, J. W.,
Kuroda, Y. and Philis, J. W., New York: Raven Press, pp 127-
141
Fain, J. N. and Malbon, C. C. (1979) Mol. Cell Biochem. 23:
1-27
Fastbom, J. and Fredholm, B. B. (1990) Neuroscience 34: 759-
769
Fredholm, B. B., Lindgren, E. and Lingstrom, K. (1985) Br. J.
Pharmacol. 86: 509-513
Gavish, M., Goodman, R. R. and Synder, S. H. (1982) Science
215: 1633-1635
Gerlach, E., Deuticke, B. and Dreisbach, R. H. (1963)
Naturwissenschaft 50: 228-229
Gerwins, P. and Fredholm, B. B. (1991) Mol. Pharmacol. 40:
149-155

115
Gerwins, P., Nordstedt, C. and Fredholm, B. B. (1990) Mol.
Pharmacol. 38: 660-666
Gilman, A. G. (1987) Ann. Rev. Biochem. 56: 615-649
Goodman, R. R., Cooper, M. J., Gavish, M. and Snyder, S. H.
(1982) Mol. Pharmacol. 21: 329-335
Green, A. (1987) J. Biol. Chem. 262: 15702-15707
Gunst, S. J., Stropp, J. Q. and Flavahan, N. A. (1989) J.
Appl. Physiol. 67(3): 1294-1298
Harden, T. K. (1983) Pharmacol. Rev. 35 (1): 5-32
Hazeki, O. and Ui, M. (1981) J. Biol. Chem. 256(6): 2856-2862
Heisler, S., Desjardins, D. and Nguyen, M. H. (1985) J.
Pharmacol. Exp. Ther. 232(1): 232-238
Hill, A. V. (1913) Biochem. J. 7: 471-480
Hill, S. J. and Kendall, D. A. (1987) Br. J. Pharmacol. 91:
661-669
Hollingsworth, E. B. and Daly, J. W. (1985) Biochim. Biophys.
Acta 847: 207-216
Homburger, V., Pantaloni, C., Lucas, M., Gozlan, H. and
Bockaert, J. (1984) J. Cell. Physiol. 121: 589-597
Hutchison, A. J., Webb, R. L., Oei, H. H., Ghai, G. R.,
Zimmerman, M. B. and William, M. (1989) J. Pharmacol. Exp.
Ther. 251: 47-55
Hutchinson, A. J., Williams, M., Jesus, R. D., Yokoyama, R.,
Oei, H. H., Ghai, G. R., Webb, R. L., Zoganas, H. C., Stone,
G. A. and Jarvis, M. F. (1990) J. Med. Chem. 33: 1919-1924
Jacobson, K. A., Ukena, D., Kirk, K. L. and Daly, J. W.
(1986) Proc. Natl. Acad. Sci. USA 83: 4089-4093
Jakobs, K. H. and Schultz, G. (1983) Proc. Natl. Acad. Sci.
USA 80: 3899-3902
Jarvis, F. M., Schulz, R., Hutchison, A. J., Do, U. H.,
Sills, M. A. and Williams, M. (1989) J. Pharmacol. Exp. Ther.
251(3): 888-893
Jarvis, F. M. and Williams, M. (1989) Eur. J. Pharmacol. 168:
243-246
Katada, T., Bokoch, G. M.,Smigel, M. D., Ui, M. and Gilman,
A. G. (1984a) J. Biol. Chem. 259: 3586-3595

116
Katada, T., Northup, J. K., Bokoch, G. M., Ui, M. and Gilman,
A. G. (1984b) J. Biol. Chem. 259: 3578-3585
Kendall, D. A. and Hill, S. J. (1988) J. Neurochem. 50: 497-
502
Kenimer, J. G. and Nirenberg, M. (1981) Mol. Pharmacol. 20:
585-591
Kent, R. S., DeLean, A. and Lefkowitz, R. J. (1980) Mol.
Pharmacol. 17: 14-23
Klotz, K. N. and Lohse, M. J. (1986) Biochem. Biophys. Res.
Commun. 140: 406-413
Kolassa, N., Pfleger, K. and Rummel, W. (1970) Eur. J.
Pharmacol. 9: 265-268
Kurachi, Y., Nakajima, T. and Sugimoto, T. (1986) Pflugers
Arch. 407: 264
Lad, P. M., Nielsen, T. B., Londos, C., Preston, M. S. and
Rodbell, M. (1980) J. Biol. Chem. 255: 10841-10846
Levitzki, A. (1987) Biochem. Pharmacol. 27: 2083-2088
Libert, F., Parmentier, M., Lefort, A., Dinsart, C., Van
Sande, J., Maenhaut, C., Simons, M. J., Dumont, J. E. and
Vassart, G. (1989) Science 244: 569-572
Liles, W. C. and Nathanson, N. (1986) J. Neurochem. 46: 89-95
Linden, J. (1991) FASEB J. 5: 2668-2676
Linden, J. and Delahunty, T. M. (1989) Trends Pharmacol. Sci.
10: 114-120
Linden, J., Patel, A. and Sadek, S. (1985) Circ. Res. 56:
279-284
Lindner, F. and Rigler, R. (1931) Pfluegers Arch. 226: 697-
708
Lohse, M. J., Klotz, K. N., Schwabe, U., Cristalli, G.,
Vittori, S. and Griffantini, M. (1988) Naunyn-Schmiedeberg's
Arch. Pharmacol. 337: 687-689
Lohse, M. J., Lenschow, V., Schwabe, U. (1984) Mol.
Pharmacol. 26: 1-9
Londos, C., Cooper, D. M. F. and Wolff, J. (1980) Proc. Natl.
Acad. Sci. USA 77: 2551-2554

117
Londos, C. and Wolff, J. (1977) Proc. Natl. Acad. Sci. USA
74: 5482-5486
Londos, C., Wolff, J. and Cooper, D. M. F. (1983)
Physiological and Regulatory Functions of Adenosine and
Adenine Nucleotides, ed by Baer, H. P. and Drummond, G. I.,
New York: Raven Press, pp 271-281
Londos, C., Wolff, J. and Cooper, D. M. F. (1981) Purinergic
Receptors, ed by Burnstock, G., London: Chapman and Hall, pp
287-323
Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, R.
J. (1951) J. Biol. Chem. 193: 265-275
Lupica, C. R., Cass, W. A., Zahniser, N. R. and Dunwiddie, T
V. (1990) J. Pharmacol. Exp. Ther. 252(3): 1134-1141
Maenhaut, C., Van Sande, J., Libert, F., Abramowicz, M.,
Parmentier, M., Vanderhaegen, J. J., Dumont, J. E., Vassart,
G. and Schiffmann, S. (1990) Biochem. Biophys. Res. Commun.
173: 1169-1178
Malbon, C. C., Moreno, F. J., Cabelli, R. J. and Fain, J. N.
(1978) J. Biol. Chem. 253: 671-678
Martens, D. M., Lohse, J., Rauch, B. and Schwabe, U. (1987)
Naunyn-Schmiedeberg's Arch. Pharmacol. 336: 342-348
Milecki, J., Baker, S. P., Standifer, K. M., Ishizu, T.,
Chida, Y., Kusiak, J. W. and Pitha, J. (1987) J. Med. Chem.
30: 1563-1566
Mills, I. and Gewirtz, H.
Commun. 168(3): 1297-1302
Mitsuhashi, M. and Payan,
134: 367-375
Morgan, N. G. (1989) Cell
New York: Guilford Press,
Munshi, R. and Linden, J.
14589
(1990) Biochem.
D. G. (1988) J.
Signalling, ed.
pp. 63-66
(1989) J. Biol.
Biophys. Res.
Cellular Physiol,
by Morgan, N. G.,
Chem. 264: 14853-
Nakata, H. (1989a) J. Biol. Chem. 264: 16545-16551
Nakata, H. (1989b) Mol. Pharmacol. 35: 780-786
Nathanson, N. M., Klein, W. L. and Nirenberg, M. (1978) Proc
Natl. Acad. Sci. USA 75(4): 1788-1791
Nazarea, M., Okajima, F. and Kondo, Y. (1991) Eur. J.
Pharmacol.- Mol. Pharmacol. Section 206: 47-52

118
Nelson, C. A., Muther, T. F., Pitha, J. and Baker, S. P.
(1986) J. Pharmacol. Exp. Ther. 237: 830-836
Nickerson, M. (1956) Nature (Lond.) 178: 697-698
Norris, J. S., Cramer, D. J., Brown, F., Popovich, K. and
Cornett, L. E. (1983) J. Recept. Res. 3: 623-645
Norris,
422-424
J. S.
and
Kohler, P.
0.
(1974) Nature (Lond.)
248:
Norris,
613-618
J. S.
and
Kohler, P.
0.
(1977) Endocrinology
100:
Ohisalo,
249-251
J. J
. and Stouffer,
J.
E. (1979) Biochem. J.
178:
Olsson,
R. A.
and
Pearson, J
. D
. (1990) Physiol. Rev.
70 (2
761-845
Paterson, A. R. P. and Oliver, J. M. (1971) Can. J. Biochem.
49: 271-274
Pedersen, S. E. and Ross, E. M. (1986) J. Biol. Chem.
260(26): 14150-14157
Pelleg, A. and Porter, S. R. (1990) Pharmacotherapy 10(3):
157-174
Pittman, R. N., Reynolds, E. E. and Molinoff, P. B. (1984) J.
Pharmacol. Exp. Ther. 230(3): 614-618
Posner, P., Peterson, C. F., Pitha, J. and Baker, S. P.
(1984) Eur. J. Pharmacol. 100: 373-376
Ramkumar, V., Barrington, W. W., Jacobson, K. A. and Stiles,
G. L. (1990) Mol. Pharmacol. 37: 149-156
Ramkumar, V., Olah, M. E., Jacobson, K. A. and Stiles, G. L.
(1991) Mol. Pharmacol. 40: 639-647
Ramkumar, V., Pierson, G. and Stiles, G. L. (1988) Prog. Drug
Res. 32: 195-247
Ramkumar, V.and Stiles, G. L.(1988) J. Pharmacol. Exp. Ther.
246: 1194-1200
Rapiejko, P. J. and Malbon, C. C. (1987) Biochem. J. 241:
765-771
Rodbell, M. (1980) Nature 284: 17-22

119
Romano, F. D., Fenton, R. A. and Dobson, J. G. (1988) Second
Messengers Phosphoproteins 12(1): 29-43
Romano, F. D., Macdonald, S. G. and Dobson, J. G. (1989) Am.
J. Physiol. 257: H1088-H1095
Ronnett, G. V. and Lane, M. D. (1981) J. Biol. Chem. 256:
4704-4707
Roof, D. J., Applebury, M. L., Sternweis, P. C. (1985) J.
Biol. Chem. 260: 16242-16249
Sattin, A. and Rail, T. W. (1970) Mol. Pharmacol. 6: 13-23
Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51: 660-670
Schubert, P. (1985) Adenosine: Receptor and Modulation of
Cell Function, ed. by Stefanovich, V., Rudolphi, K. and
Schubert, P., Oxford, England: IRL Press Limited, pp. 117-129
Seamon, K. B. and Daly, J. W. (1986) Adv. Cyclic Nucleotide
Protein Phosphorylation Res. 20: 1-50
Sevilla, N., Tolkovsky, A. M. and Levitzki, A. (1977) FEBS
Lett. 81: 339-341
Shimizu, H. and Daly, J. W. (1970) Biochem. Biophys. Acta.
22: 465-473
Shryock, J., Patel, A., Belardinelli, L. and Linden, J.
(1989) Am. J. Physiol. 256: H321-H327
Shryock, J., Travagli, H. C. and Belardinelli, L. (1992) J.
Pharmacol. Exp. Ther. (accepted for publication)
Standifer, K. M., Pitha, J. and Baker, S. P. (1989) Naunyn-
Schmiedeberg's Arch. Pharmacol. 339: 129-137
Stephenson, R. P. (1956) Br. J. Pharmacol. 11: 379-393
Stiles, G. L. (1985) J. Biol. Chem. 260: 6728-6732
Stiles, G. L., Daly, D. J. and Olsson, R. A. (1985) J. Biol.
Chem. 260: 10806-10811
Stroher, M., Nanoff, C. and Schtz, W. (1989) Naunyn-
Schmiedeberg's Arch. Pharmacol. 340: 87-92
Stryer, L. and Bourne, H. R. (1986) Ann. Rev. Cell Biol. 2:
391-419
Toews, M. L. (1987) Mol. Pharmacol. 31: 58-68

120
Trivedi, B. K., Bridges, A. J. and Bruns, R. F. (1990)
Adenosine and Adenosine Receptors, ed. by Williams, M.,
Clifton, New Jersey: The Humana Press Inc. pp 57-103
Trost, T. and Schwabe, U. (1981) Mol. Pharmacol. 19: 228-235
Trussel, L. 0. and Jackson, H. B. (1985) Proc. Natl. Acad.
Sci. USA 82: 4857-4861
Ueeda, M., Thompson, R. D., Arroyo, L. H. and Olsson, R. A.
(1991) J. Med. Chem. 34(4): 1340-1344
Ullman, A. and Svedmyr, N. (1988) Thorax 43: 674-678
Van Calker, D., Miller, M. and Hamprecht, B. (1979) J.
Neurochem. 33: 999-1005
Watanabe, A. M., McConnaughey, M. M., Strawbridge, R. A.,
Fleming, J. W., Jones, L. R. and Besch, H. R. (1978) J. Biol.
Chem. 253: 4833-4836
Wesley, R. C. and Belardinelli, L. (1989) Circulation 80:
128-137
Williams, M. and Risley, E. A. (1980) Proc. Natl. Acad. Sci.
USA 77: 6892-6896
Wolff, J., Londos, C. and Cooper, D. M. F. (1981) Adv. Cyclic
Nucleotide 14: 199-214
Wu, S. N., Linden, J., Visentin, S., Boykin, M. and
Belardinelli, L. (1989) Circ. Res. 65(4): 1066-1076
Xu, D., Kong, H. and Liang, B. T. (1992) Circ. Res. 70(1):
56-65
Yeung, S. M. and Green, R. D. (1983) J. Biol. Chem. 258:
2334-2339

BIOGRAPHICAL SKETCH
Fan Xie was born on June 2 6, 1965, in Changsha, Hunan,
People's Republic of China. She was accepted into the first
English Medical Class of Hunan Medical University in 1981 and
received a medical doctor degree in 1987. After graduation,
she came to the United States to further her education. In
the fall of 1987, she started her graduate studies under the
guidance of Dr. Stephen P. Baker in the Department of
Pharmacology and Therapeutics at the University of Florida.
After completing the requirements for the degree of Doctor of
Philosophy, she plans to do a postdoctoral fellowship at
Hoffmann-La Roche Inc.
121

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
./77liLil lLx.
Stephen P. Baker, Ph.D., Chairman
Professor of Pharmacology and
Therapeutics
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of poctor of Philosophy.
Les c c/ 1 i- C
4-
Luiz- Belrdinelli, M.EK
Professor of Pharmacology and
Therapeutics
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
rX-Tiy- "
Edwin Meyer^Jph.D.
Associate Professor of
Pharmacology and Therapeutics
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and guality, as
a dissertation for the degree of Doctor of' 'Philosophy^
/
Thomas F. Rowe, Ph.D.
Associate Professor of
Pharmacology and Therapeutics

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as.
a dissertation for the degree of Doctor of PhilostSjkhy.
>han K. Raizada, Ph.D.
rofessor of Physiology
This dissertation was submitted to the Graduate Faculty
of the College of Medicine and to the Graduate School and was
accepted as partial fulfillment
degree of Doctor of Philosophy.
May, 1992
the requirements for the
4
Dean, Graduate School

UNIVERSITY OF FLORIDA



95
agonist, receptor and Gs protein. This complex appears to be
a prerequisite for receptor mediated activation of the enzyme
(DeLean et al., 1980; Kent et al., 1980; Gilman, 1987).
Evidence supporting this mechanism has been recently reported
by Romano et al. (1988; 1989). They showed that binding of
ISO to the high affinity binding state of the BAR was
eliminated in the presence of PIA, an Ai-AdoR agonist.
Although these data were interpreted to indicate that
activation of the inhibitory Ai-AdoR prevented the BAR from
forming an agonist high affinity binding state (ternary
complex), it can not be ascertained whether the inhibitory
effect is mediated at the receptor or the G protein level. At
the receptor level, inhibitory agonists could theoretically
initiate alterations in the stimulatory receptor affecting
agonist-receptor interaction or prevent the agonist-receptor
complex from becoming active and thereby impairing signal
transduction.
In 1987, Milecki et al. (1987) reported on the synthesis
and partial characterization of an alkylating carbostyril
derivative (C-Br). This compound and several of its congeners
were shown to be potent BAR agonists. In a subsequent study
using membranes from rat reticulocytes, Standifer et al.
(1989) showed that compared to ISO, C-Br was a full agonist
and its interaction with the receptor was resistant to the
effects of a guanine nucleotide. That is, the nonhydrolyzable
guanine nucleotide derivative Gpp(NH)p only slightly reduced
the ability of C-Br to inhibit [125I]CYP binding (Standifer et


34
of a maximal response (EC50) were determined using a
concentration-effect analysis with a non-linear regression
algorithm (Marquardt-Levenberg). Statistical analysis of
significance of difference was performed using the Student's
t-test.


73
Figure 3-8. Inhibition of specific [3H]CPX binding to DDT
cell membrane by N-0861. Membranes protein (320 (ig) was
incubated with 1 nM [3H]CPX and the indicated concentrations
of N-0861 for 150 min at room temperature. Specific binding
was determined as described in the "Methods" section. Data
points are the mean of triplicate determinations which varied
by less than 5%. The control specific [3H]CPX binding was 132
fmol/mg protein.


70
o.S

3.S
ES
3 O
8S.
< O)
CL E
i!
o.
25
20
15
10
0
BASAL
WRC-0018
WRC-0018+8PST
Figure 3-5. The effect of 8PST on WRC-0018-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram, without or with 0.1 |1M WRC-0018 or
WRC-0018 plus 5 (1M 8PST for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSD of quadruplicate determination. The indicates a
p<0.01 for the WRC-0018 value compared to the basal value and
a p<0.005 for the WRC-0018 plus 8PST value compared to the
WRC-0018 value.


109
Figure 4-4. The effect of CPA on basal cAMP accumulation in
DDT cells. Cells were incubated in HBSS containing 100 |1M
rolipram, without or with the indicated concentrations of CPA
for 6 min at 37C. At the end of the incubation period, the
cAMP accumulated was determined as described in the "Methods"
section. The basal control level of cAMP accumulated was 262
pmol/mg protein/6 min. Each data point is the meanSE, n=4.


CHAPTER 4
INHIBITORY EFFECT OF Ai-ADENOSINE RECEPTOR
ON IRREVERSIBLE ACTIVATION OF THE 15-ADRENORECEPTOR
Introduction
The guanine nucleotide binding proteins Gs and Gi play
an important role in mediating receptor activation and
inhibition of AC activity, respectively. Gs and Gi contain
distinct a (as and ai) and common 5 and y subunits.
Activation of either receptor type results in the exchange of
GTP for bound GDP followed by dissociation of the G protein
subunits. Although the mechanism by which ots mediates
stimulation of AC is well understood, less clear is the
mechanism (s) involved in the inhibition of the enzyme by oci.
Several mechanisms may be involved in the inhibition. That
is, inhibition may include: 1) the direct interaction of
activated OCi with the catalytic subunit of the enzyme, or 2)
5+Y from Gi complexing with activated a3, thus favoring the
undissociated and inactive form of Gs. Evidence for both
pathways has been reported (Katada et al., 1984a, b; Jakobs
and Schultz, 1983; Roof et al., 1986). An alternative
mechanism may involve prevention of stimulatory signal
transduction at the receptor level. In this latter case,
inhibition may be mediated by preventing the stimulatory
receptor from forming a ternary complex composed of the
94


23
concentration response of NECA to the left (EC5o=49 nM, p<0.01
by ANOVA) indicating sensitization of AdoRs (Biaggioni et
al., 1991b). Other examples include pretreatment of DDTi MF-2
cells with dexamethasone (Gerwins and Fredholm, 1991). This
glucocorticoid caused a concentration- and time-dependent
increase in the number of Ai-AdoRs, but did not affect the Kd
or the proportion of A]_-AdoRs in high and low affinity states
(Gerwins and Fredholm, 1991). (R)-PIA was more potent as an
inhibitor of cAMP formation induced by ISO in dexamethasone-
treated cells. Addition of glucocorticoid receptor antagonist
RU 486 or protein synthesis inhibitor cycloheximide prevented
the up-regulation of Ai-AdoR (Gerwins and Fredholm, 1991). In
contrast to sensitization of Ai-AdoR-AC system, the A2~AdoR-AC
system was desensitized as indicated by the decreased ability
of NECA to increase cAMP formation in dexamethasone-treated
cells (Gerwins and Fredholm, 1991) .
AdoRs are also regulated under normal and
pathophysiological conditions. For example, in the
hypothyroid state, (R)-PIA mediated inhibition of AC and its
antilipolytic effect is enhanced (Ohisalo and Stouffer,
1979). In contrast to hypothyroidism, the hyperthyroid state
is characterized by enhanced lipolytic activity and cAMP
accumulation in adipocytes. These are likely related to the
loss of inhibitory tone mediated by Ai-AdoR due to a 35%
decrease in Ai-AdoR number (Malbon et al., 1978; Rapiejko and
Malbon, 1987). Other conditions that may alter AdoRs include


97
stimulated cAMP accumulation in a concentration-dependent
manner, however, in the presence of 1 flM CPA, the maximal
response produced by ISO and C-Br was decreased by 86% and
75%, respectively. Figure 4-2 depicts the effect of CPA on
cAMP accumulation stimulated by 10 HM ISO and 1 (1M C-Br,
concentrations which gave a maximal response. CPA attenuated
the effect of ISO and C-Br in a concentration-dependent
manner with the same potency (EC50 = 2.3 nM) and the same
maximal response (89% inhibition).
Figure 4-3 illustrates the effect of the Ai-AdoR
antagonist CPX on CPA mediated inhibition of ISO- and C-Br-
stimulated cAMP accumulation. ISO (10 (1M) and C-Br (1 JIM)
(concentrations which caused maximal stimulation of cAMP
accumulation, see Figure 4-1) increased cAMP accumulation to
a similar extent (27 and 28-fold above the basal level,
respectively) and CPA (1 |IM) inhibited their stimulation (81%
for C-Br and 84% for ISO). CPX (5 M-M) an A]_-AdoR antagonist,
largely blocked the inhibitory effect of CPA, and the cAMP
stimulations caused by C-Br and ISO were restored by 96% and
84%, respectively. This indicates that the inhibitory effect
of CPA on C-Br and ISO-stimulated cAMP accumulation is Ai-
AdoR mediated.
The effect of CPA on the basal level of cAMP in DDT
cells is shown in Figure 4-4. CPA decreased the basal level
of cAMP with an EC50 of 6.1 nM and a maximal inhibition of
82% .


21
"Heterologous" desensitization is referred to hormone-
nonspecific type where activation of one receptor causes loss
of response mediated by other receptors. Desensitization of
15AR system appears to be initiated by receptor
phosphorylation which results in functional uncoupling of the
J5AR from Gs. Two kinases have been implicated: the cAMP-
dependent protein kinase (PKA) which plays a major role in
heterologous desensitization; and the cAMP-independent
kinase, termed J5AR kinase which specifically phosphorylates
the agonist-occupied receptor leading to homologous
desensitization. After uncoupling, the receptors appear to be
sequestered within the cells. Removal of the agonist after
sequestration leads to rapid resensitization of the system.
During longer-term agonist treatment, there appears to be a
loss of receptors (down-regulation) due to receptor
degradation or loss of the recognition site for ligand
binding (Harden, 1983) .
Studies have shown that in vitro exposure of cultured
rat adipocytes to (R)-PIA causes concentration- and time-
dependent loss of Ai-AdoR and decrease in the content of Gi
protein (Green, 1987). These changes were accompanied by
attenuation of the antilipolytic effect of (R)-PIA
(homologous desensitization) (Green, 1987). In another study,
the number of cardiac Ai~AdoRs in chick embryos was decreased
by 63% after pretreatment with 1 JIM (R)-PIA for 44 hrs
(Shryock et al., 1989). Experiments also showed that the
desensitization of Ai~AdoR system in DDTi MF-2 cells was


71
BASAL
WRC-0018
Figure 3-6. The effect of CPA on WRC-0018-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram, without or with 0.1 JIM WRC-0018 or
WRC-0018 plus 1 JIM CPA for 7 min at 3 6C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meaniSD of quadruplicate determination. The indicates a
p<0.005 for the WRC-0018 value compared to the basal value
and for the WRC-0018 plus CPA value compared to the WRC-0018
value.


119
Romano, F. D., Fenton, R. A. and Dobson, J. G. (1988) Second
Messengers Phosphoproteins 12(1): 29-43
Romano, F. D., Macdonald, S. G. and Dobson, J. G. (1989) Am.
J. Physiol. 257: H1088-H1095
Ronnett, G. V. and Lane, M. D. (1981) J. Biol. Chem. 256:
4704-4707
Roof, D. J., Applebury, M. L., Sternweis, P. C. (1985) J.
Biol. Chem. 260: 16242-16249
Sattin, A. and Rail, T. W. (1970) Mol. Pharmacol. 6: 13-23
Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51: 660-670
Schubert, P. (1985) Adenosine: Receptor and Modulation of
Cell Function, ed. by Stefanovich, V., Rudolphi, K. and
Schubert, P., Oxford, England: IRL Press Limited, pp. 117-129
Seamon, K. B. and Daly, J. W. (1986) Adv. Cyclic Nucleotide
Protein Phosphorylation Res. 20: 1-50
Sevilla, N., Tolkovsky, A. M. and Levitzki, A. (1977) FEBS
Lett. 81: 339-341
Shimizu, H. and Daly, J. W. (1970) Biochem. Biophys. Acta.
22: 465-473
Shryock, J., Patel, A., Belardinelli, L. and Linden, J.
(1989) Am. J. Physiol. 256: H321-H327
Shryock, J., Travagli, H. C. and Belardinelli, L. (1992) J.
Pharmacol. Exp. Ther. (accepted for publication)
Standifer, K. M., Pitha, J. and Baker, S. P. (1989) Naunyn-
Schmiedeberg's Arch. Pharmacol. 339: 129-137
Stephenson, R. P. (1956) Br. J. Pharmacol. 11: 379-393
Stiles, G. L. (1985) J. Biol. Chem. 260: 6728-6732
Stiles, G. L., Daly, D. J. and Olsson, R. A. (1985) J. Biol.
Chem. 260: 10806-10811
Stroher, M., Nanoff, C. and Schtz, W. (1989) Naunyn-
Schmiedeberg's Arch. Pharmacol. 340: 87-92
Stryer, L. and Bourne, H. R. (1986) Ann. Rev. Cell Biol. 2:
391-419
Toews, M. L. (1987) Mol. Pharmacol. 31: 58-68


12
cross-linking, migrate on sodium dodecyl sulfate-
polyacrylamide gel electrophoresis (SDS-PAGE) with an
molecular mass between 35 and 38 kDa (Ramkumar et al., 1990) .
In comparison with Ai-AdoRs, photolabeled A2~AdoRs migrate on
SDS-PAGE with a Mr of 42,000 (Ramkumar et al., 1990). Both
receptors are known to be glycosylated (Ramkumar et al.,
1990). Treatment of photoaffinity-radiolabeled Ai-AdoR of
brain with an endoglycosidase (Stiles, 1985) or with either
trifluoromethanesulfonic acid or a-mannosidase (Klotz and
Lohse, 1986) reduces the molecular mass of the ligand-binding
peptide to 32 kDa. A role for protein glycosylation in the
function and cellular processing of the AdoR is at present
unknown. However, glycosylation has been reported to be
essential in the synthesis of insulin receptors (Ronnett and
Lane, 1981) but not B-adrenoreceptors (BAR) (Doss et al.,
1985). In addition, glycosylation also appears to be required
for the maintenance of cell surface muscarinic receptors
(Liles and Nathanson, 1986).
A]_-AdoRs from rat brain membranes were purified in 1989
(Nakata, 1989a,b; Munshi and Linden, 1989). Ai~AdoR was
solubilized with digitonin and purified approximately 50,000-
fold to apparent homogeneity by two cycles of affinity
chromatography using an antagonist affinity column. The
purified receptor migrated on SDS-PAGE with Mr=34,000 either
in the absence or presence of 2-mercaptoethanol, suggesting
that the receptor does not contain disulfide-linked subunits
(Nakata, 1989a).


89
Log [ECA], M
Figure 3-24. The effect of CCPA pretreatment on NECA-
stimulated cAMP accumulation in DDT cells. Cells were ,
incubated in fresh growth media without or with 0.1 [1M CCPA
for 16 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 |1M rolipram
and the indicated concentrations of NECA for 7 min at 36C.
At the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The data
for NECA in control cells are taken from Figure 3-13. The
basal level of cAMP accumulated was 32 and 111 pmol/mg
protein/min for the control and CCPA experiments,
respectively. Each data point is the meanSD of quadruplicate
determinations and is representative of 2 experiments.


67
<
O
Figure 3-2. Inhibition of ISO-stimulated cAMP accumulation in
DDT cells by CPA, NECA and ADO. Cells were incubated in HBSS
containing 50 (1M rolipram, 10 |1M ISO or ISO plus the
indicated concentrations of CPA, NECA or ADO + 1 JIM DIP + 1
JIM EHNA for 7 min at 36C. At the end of the incubation
period, the cAMP accumulated was determined as described in
the "Methods" section. The control cAMP accumulated in the
presence of ISO alone was 5712, 879 and 684 pmol/mg
protein/min for the CPA, NECA and ADO experiments,
respectively. The basal level of cAMP accumulated was below
the detection level for CPA and NECA and 61 pmol/mg
protein/min for the ADO experiment. Each data point is the
meanSD of quadruplicate determinations and is representative
of 2 experiments.


24
pregnancy, lactation, starvation, obesity and aging (Ramkumar
et al., 1988).
Goals
Over the past decade, a great deal has been learned
about the pharmacology, biochemistry and physiology of AdoRs.
However, much remains unknown. In general, receptors that
mediate inhibition of cAMP formation appear to dominate over
stimulatory receptors. In the case of AdoRs, if both subtypes
are present in a single cell and are simultaneously activated
by adenosine, it becomes important to determine under what
conditions the Ai~ and A2~AdoR mediated responses will be
expressed. Thus, the major goals of this study were 1) to
determine pharmacologically if an interaction between Ai~ and
A2AdoR occurs and 2) to define the conditions whereby the
expression of the A2~AdoR mediated response can be
demonstrated.
In addition, the hypothesis that the mechanism for Ai~
AdoR inhibitory effects involves alteration in the ability of
BAR agonists to interact with the BAR was investigated. By
using an irreversible BAR agonist that permanently activates
the BAR, it was determined if the resulting response can be
modulated by the inhibitory Ai-AdoR.


110
Time (min)
Figure 4-5. The effect of CPA on C-Br-stimulated cAMP
accumulation in DDT cells. Cells were preincubated in HBSS
containing 100 JIM rolipram and 1 (1M C-Br at 37C. After 3 min
of incubation, 20 p.M propranolol or propranolol plus 1 |1M CPA
was added. At the times indicated, the cAMP content was
determined as described in the "Methods" section. Each data
point is the meanSE, n=4.


112
Log [Isoproterenol], M
Figure 4-7. Inhibition of specific [125I]CYP binding to DDT
cell membrane by ISO in the presence of Gpp(NH)p or CPA.
Membrane protein (50 (lg) was incubated with 30 pM [125I]CYP,
the indicated concentrations of ISO, without or with 100 |1M
Gpp(NH)p or 1 JIM CPA for 60 min. at 37C. Data points are the
mean of triplicate determination which varied by less than 6%
and are representative of 3 experiments. The control specific
[12^I]CYP binding was 53 fmol/mg protein.


26
Q
CPA
CCPA
WRC-0018
N-0861
CPX
8PST
Figure 1-2. Chemical structures of AdoR agonists and
antagonists. Abbreviations used: N6-cyclopentyladenosine
(CPA), 2-chloro-N6-cyclopentyladenosine (CCPA), 2[2(2
Naphthyl)ethoxy]adenosine (WRC-0018), 5'-N-ethylcarboxamido-
adenosine (NECA), ()N6-endonorbornan-2-yl-9-methyladenine
(N-0861), 8-cyclopentyl-l,3-dipropylxanthine (CPX), 8(p-
sulfophenyl)theophylline (8PST).


13
Recently, two previously cloned proteins (RDC7 and RDC8)
with deduced seven transmembrane helices have been identified
as canine Ai~ and A2a-AdoR subtypes, respectively (Libert et
al., 1989; Maenhaut et al., 1990). The deduced molecular
masses of RDC7 (36,356 Da) and RDC8 (45,008 Da) correspond
closely to the apparent molecular masses of Ai~ and A2a-AdoRs
estimated by photoaffinity labeling. Features shared by both
proteins include small N-termini, conserved transmembrane
domains and at least one cysteine in exofacial loops 1 and 2.
Notable by their absence are consensus sequences for N-linked
glycosylation in the N-terminal segments and an aspartate
residue in the third transmembrane segment, the hallmark of
cationic amine receptors. Clusters of serine and threonine
residues in the C-terminal segments, commonly seen in guanine
nucleotide-binding protein (G protein)-linked receptors, are
absent in RDC7. Finally, at 326 amino acids, RDC7 is among
the smallest members of the G-protein-coupled superfamily of
receptors. RDC7 and RDC8 are similar to a variety of other
superfamily members, but none of these is more than 30%
identical to either of the AdoRs. The expression of RDC8 in
adrenal cells, thyrocytes and Xenopus oocytes resulted in
activation of AC in the absence of added AdoR agonist.
Membranes from Cos 7 cells transfected with RDC8 cDNA
exhibited binding characteristics of an A2~AdoR. Moreover,
RDC8 mRNA and A2~AdoR displayed a very similar distribution
in the brain (Maenhaut et al., 1990). These data all support
that RDC8 is an A2~AdoR. The gene(s) for the low-affinity A2b-


82
Log [ECA], M
Figure 3-17. The effect of PTX pretreatment on NECA-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 25 ng/ml PTX
for 18 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 (1M rolipram
and the indicated concentrations of NECA for 7 min at 36C.
At the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The data
for NECA in control cells are taken from Figure 3-13. The
basal level of cAMP accumulated was 11 and 42 pmol/mg
protein/min for the control and PTX experiments,
respectively. Each data point is the meanSD of quadruplicate
determinations and is representative of 2 experiments.


66
1. Selective A^AdoR antagonist 2. Pertussis toxin (PTX)
3. Desensitization / Down-regulation of A^AdoR
Figure 3-1. Experimental approaches to express the A2-
AdoR mediated response.


9
hydrolysis in these tissues (Hill and Kendall, 1987;
Hollingsworth and Dally, 1985; Linden, 1991). In contrast to
potentiating the stimulatory response of other
neurotransmitters on phospholipid metabolism, in several
other tissues (eg., mouse cortex, brown fat and GH3 pituitary
cells), activation of Ai-AdoR leads to inhibition of inositol
phosphate accumulation (Kendall and Hill, 1988; Delahunty et
al., 1988; Linden and Delahunty, 1989; Linden, 1991).
Both A]_ and A2~AdoRs are widely distributed in the
central nervous system and peripheral tissues. For example,
Ai-AdoRs are present in the brain, heart, kidney, lung,
pancreas and adipocytes and A2~AdoRs are present in the
brain, coronary arteries, kidney and lung (Olsson and
Pearson, 1990).
Analysis of structure-activity relationship indicates
that certain N-6 substituents of adenosine enhance the
potency of adenosine as an Ai-AdoR agonist (Olsson and
Pearson, 1990). For example, cyclopentyladenosine (CPA) and
2-chloro-N^-cyclopentyl-adenosine (CCPA) have Kj. (A2)/Kj. (Ai)
ratios of 2500 and 9750, respectively (Lohse et al., 1988).
Several purines with C-2 substituents (e.g., 2-
aralkoxyadenosine, 2-alkoxyadenosine) have increased potency
as A2~AdoR agonists. For instance, 2-[2-(2-naphthyl)ethoxy]
adenosine (WRC-0018) is a highly selective A2~AdoR agonist
(Ueeda et al., 1991). Examples of AdoR agonists and
antagonists and their chemical structures are shown in
Figures 1-1, 1-2.


92
T3.E
|e
1 =
ES
D O
8*
< O)
CL £
o E
Log [WRC-0018], M
Figure 3-27. The effect of ADO pretreatment on WRC-0018-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 5 (1M DIP + 1
M-M EHNA or 5 (XM DIP + 1 JiM EHNA + 100 }1M ADO for 24 hr at
37C. At the end of the incubation period, the cells were
washed 4 times with ice-cold HBSS and detached. Cells were
then incubated in HBSS containing 50 |1M rolipram and the
indicated concentrations of WRC-0018 for 7 min at 36C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The basal
level of cAMP accumulated was 32, 31, 91 pmol/mg
protein/min for the control, DIP+ENHA and ADO+DIP+EHNA
experiments, respectively. Each data point is the meanSE,
n=3.


25
AGONIST
ANTAGONIST
CPA
N-0861
CCPA
CPX
NECA
8PST
ADO
WRC-0018
8PST
NECA
?
ADO
A, -AdoR decrease cAMP
A2 -AdoR increase cAMP
Figure 1-1. Agonists and antagonists of AdoR subtypes.


65
was faster than that for the Ai-AdoR in DDT cells, in other
cells, the opposite may occur. Thus, additional studies are
necessary to test this latter possibility, further
characterize the dual modulation of AC by adenosine in the
same cell and/or organ and define the implications for such
role.


2
regulator of coronary blood flow. This hypothesis received
little immediate attention and interest at the time.
It was not until 1963 that Berne (1963) and Gerlach et
al. (1963) independently revived Lindner and Rigler's
adenosine hypothesis by demonstrating the release of
adenosine catabolites from hypoxic or ischemic heart muscle.
The revived hypothesis states that an imbalance between
oxygen supply and oxygen demand leads to alterations of the
cellular release of adenosine, which in turn changes the
contractile state of vascular smooth muscle of the resistance
vessels.
In 1970, Sattin and Rail (1970) and Shimizu and Daly
(1970) proposed the existence of extracellular adenosine
receptors (AdoRs) based on the observation that adenosine and
some adenine nucleotides elevate adenosine 3', 5'-cyclic
monophosphate (cAMP) in cerebral cortical slices. This effect
was competitively inhibited by methylxanthines, caffeine and
theophylline. Within a decade, it became clear that there are
AdoRs that inhibit as well as others that stimulate adenylyl
cyclase (AC) (Londos and Wolff, 1977; Van Calker et al.,
1979), constituting a system for the bidirectional control of
the catalytic activity of this key enzyme.
Physiological Effects
Adenosine is present in every cell of the human body and
exerts a wide spectrum of effects on various tissues and


99
Figure 4-7 illustrates the effects of Gpp(NH)p and CPA
on the concentration response of ISO to compete with [125I]CYP
for the BAR binding site. ISO alone displaced specific
[125I]CYP binding with an IC50 of 60 nM and a Hill slope of
0.58. In the presence of Gpp(NH)p (100 |1M) the displacement
curve shifted to the right with an IC50 value of 406 nM and
steepened with a Hill slope of 0.80. In contrast to Gpp(NH)p,
CPA had no effect on either the IC50 value (70 nM) or the Hill
slope (0.60) of the ISO displacement curve as compared with
ISO alone.
Discussion
Receptor mediated activation of AC activity is a process
with multiple steps that involve an agonist, receptor, a
stimulatory guanine nucleotide binding protein (Gs) and the
catalytic subunit of AC (Levitzki, 1987; Rodbell, 1980).
General agreement now exists that agonists promote the
formation of a ternary complex consisting of the agonist,
receptor and Gs and this interaction accelerates the exchange
of GDP for GTP at G3. The binding of the agonist in the
ternary complex is of high affinity. GTP destabilizes the
ternary complex resulting in a decrease in agonist affinity
and the release of an activated as subunit of Gs which
directly stimulates the catalytic subunit of AC. Hydrolysis
of GTP to GDP deactivates as and AC activity returns to the
basal level (Gilman, 1987). In contrast to AC activation,


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
INTERACTION OF ADENOSINE RECEPTORS
IN A SMOOTH MUSCLE CELL LINE
By
Fan Xie
May, 1992
Chairman: Stephen P. Baker, Ph.D.
Major Department: Pharmacology and Therapeutics
DDTi MF-2 cells have been shown to express inhibitory Ai
and stimulatory A2 adenosine receptors (AdoRs) coupled to 3',
5'-cyclic adenosine monophosphate (cAMP) accumulation. The
objective of this study was to investigate the possible
interaction between the two AdoRs. The AdoR agonists,
adenosine and 51-N-ethylcarboxamido-adenosine (NECA)
attenuated isoproterenol (ISO)-stimulated cAMP accumulation
in a dose-dependent manner with a maximal inhibition of 68%
and 98%, respectively. No cAMP stimulation was observed with
either compound. In contrast, the selective A2~AdoR agonist
2-[2-(2 naphthyl)ethoxy]adenosine (WRC-0018) produced a
*
biphasic response. Stimulation of cAMP accumulation (8-fold)
occurred at low concentrations (5 500 nM) followed by an
attenuation at high concentrations (>500 nM). The attenuation
component was prevented by 1) the selective Ai-AdoR
vi


Ill
Figure 4-6. Scatchard plot of specific [125I]CYP binding to
DDT cell membranes after pretreatment with C-Br, CPA or C-
Br+CPA. Cells were incubated in HBSS without or with 1 |IM C-
Br, 1 JIM CPA or l^lM C-Br + 1 JIM CPA for 10 min at 37C. At
the end of the incubation period, the cells were harvested
and membranes prepared. Membrane protein was assayed with 6 -
100 pM [125I]CYP as described in the "Methods" section. The
data are plotted as the ratio of the amount of specifically
bound ligand (pmol/mg protein) to free ligand (pmol/L) versus
the amount of specifically bound ligand/mg protein. Data
points are the mean of triplicate determination and are
representative of 3 experiments.


84
Figure 3-19. Scatchard plot of specific [3H]CPX binding to
DDT cell membranes after CCPA treatment. Cells were incubated
in fresh growth media without or with 0.1 (1M CCPA for 16 hr
at 37C. At the end of the incubation period, the cells were
washed 4 times with ice-cold HBSS and the membranes prepared.
Membrane protein (0.1 mg) was assayed with 0.06 4 nM
[3H]CPX as described in the "Methods" section. The data are
plotted as the ratio of the amount of specifically bound
ligand (pmol/mg protein) to free ligand (pmol/L) versus the
amount of specifically bound ligand/mg protein. Data points
are the mean of triplicate determination and are
representative of 3 experiments.


CHAPTER 2
EXPERIMENTAL PROCEDURES
Source of Materials
The radioligands [2,8-3H]adenosine 3', 5'-cyclic
monophosphate ([3H]cAMP; 31.2 Ci/mmol), [3H]8-cyclopentyl-l,3-
dipropylxanthine ([3H]CPX; 99-107 Ci/mmol) and (-) [125I]iodo-
cyanopindolol ([125I]CYP; 2,000-2,200 Ci/mmol) were purchased
from New England Nuclear Corp. (Boston, MA, USA) Adenosine,
N6-cyclopentyladenosine (CPA), (-)-N6-(2-phenyl-isopropyl)
adenosine ((R)-PIA), dipyridamole (DIP), erythro-9-(2-
hydroxy-3-nonyl)adenine (EHNA), adenosine deaminase (ADA)
type VI, 3-[(3-cholamidopropyl)dimethylammoniol]-1-propane-
sulfonate (CHAPS), benzamidine, (-)-isoproterenol (ISO), 5'-
guanylyl-imididodiphosphate (Gpp(NH)p), propranolol (PROP),
()-alprenolol, penicillin G, streptomycin sulfate,
amphotericin B, theophylline, protein kinase, hydroxyapatite
and bovine serum albumin were from Sigma Chemical Co. (St.
Louis, MO, USA). 8-cyclopentyl-l,3-dipropylxanthine (CPX),
8(p-sulfophenyl)theophylline (8PST), 5'-N-ethylcarboxamido-
adenosine (NECA) and 2-chloro-N^-cyclopentyladenosine (CCPA)
were purchased from Research Biochemicals Inc. (Natick, MA,
USA). The DDTi MF-2 (DDT) cell line was obtained from
American Type Culture Collection (Rockville, MD, USA).
27


8
mediates the stimulation of AC has an agonist potency series
of ECA > (R)-PIA > (S)-PIA (Van Calker et al., 1979; Londos
et al., 1980). The A2~AdoR in brain has been further
subdivided into A2a and A2b subclass. Central A2a-AdoRs are
localized primarily in the striatum, nucleus accumbens and
olfactory tubercle, whereas central A2b-AdoRs are present in
all brain regions. Adenosine and NECA have a higher affinity
for the A2a-AdoRs than the A2b-AdoRs (Daly et al., 1983; Bruns
et al., 1986).
In addition to modulating AC activity, AdoRs are coupled
to other effector systems such as K+ and Ca2+ channels and to
phospholipid hydrolysis. Electrophysiological studies
indicate that adenosine activates an inwardly rectifying
potassium current dKAdo) i-n sinoatrial (Bellardinelli et al.,
1988), atrial (Belardinelli and Isenberg, 1983) and neuronal
(Trussel and Jackson, 1985) cells. The activation of IxAdo is
mediated via a pertussis toxin-sensitive guanosine
triphosphate (GTP)-binding protein (Kurachi et al., 1986).
Adenosine also attenuates the activity of the voltage-
sensitive Ca2+ channels in hippocampal neurons via presynaptic
Ai-AdoRs (Schubert, 1985) Adenosine has also been found to
indirectly stimulate inositol phosphate accumulation in
guinea pig cortex, FRTL-5 thyroid cells and vas deferens. In
these tissues, the effect of adenosine is to potentiate the
responses of neurotransmitters such as histamine,
norepinephrine and angiotensin II. However, neither adenosine
nor its analogs alone increase inositol phopholipid


62
desensitization of Ai- and A2~AdoR was 8 and 0.75 hr in DDT
cells, respectively, 4) the efficacy of the agonist to induce
desensitization of the receptor subtypes may be different, or
5) the mechanism for desensitization of the receptor subtypes
may be different. Evidence for a differential desensitization
mechanism has been reported recently (Ramkumar et al., 1991).
During desensitization, the Ai-AdoR was down-regulated
(internalized), uncoupled from G proteins and phosphorylated
whereas the A2~AdoR was not (Ramkumar et al., 1991) .
Summary
Adenosine is an autocoid produced by the same cells on
which and/or adjacent cells it exerts its effects. The
results of the present study demonstrate that in DDT cells
adenosine acts on at least two receptor subtypes (Ai~ and A2-
AdoR) whose actions result in opposing effects on the
formation of cAMP. There may be circumstances in which
adenosine production rapidly and transiently increases and
thereby maximally activates AdoRs. When the concentration of
adenosine rises far above the physiological range, the
coexistence of two receptor subtypes on the same cell with
opposing functional effects would dampen the responses to the
transient extremes of adenosine concentration. That is, in
the example illustrated in Figure 3-14 (ADO alone),
activation of the Ai-AdoR would attenuate the effect of A2-


114
Contreras, M. L., Wolfe, B. B. and Molinoff, P. B. (1986) J.
Pharmacol. Exp. Ther. 239: 136-143
Cooper, D. M. F. and Londos, C. (1979) J. Cyclic Nucleotide
Res. 5: 289-302
Daly, J. W., Butts-Lamb, P. and Padgett, W. (1983) Cell. Mol.
Neurobiol. 3: 69-80
Delahaba, G and Cantoni, G. L. (1959) J. Biol. Chem. 234:
603-608
Delahunty, T. M., Cronin, M. J. and Linden, J. (1988)
Biochem. J. 255: 69-77
DeLean, A., Stadel, J. M. and Lefkowitz, R. J. (1980) J.
Biol. Chem. 255: 7108-7117
Doss, R. C., Kramarcy, N. R., Harden, T. K. and Perkins, J.
P. (1985) Mol. Pharmacol. 27: 507-516
Doss, R. C., Perkins, J. P. and Harden, T. K (1981) J. Biol.
Chem. 256: 12281-12286
Drury, A. N. and Szent-Gyorgyi, A. (1929) J. Physiol. (Lond.)
68: 213-237
Endoh, M., Maruyama, M. and Iijima, T. (1985) Am. J. Physiol.
249: H309-H320
Endoh., M., Muruyama, M. and Taira, N. (1983) Physiology and
Pharmacology of Adenosine Derivatives, ed. by Daly, J. W.,
Kuroda, Y. and Philis, J. W., New York: Raven Press, pp 127-
141
Fain, J. N. and Malbon, C. C. (1979) Mol. Cell Biochem. 23:
1-27
Fastbom, J. and Fredholm, B. B. (1990) Neuroscience 34: 759-
769
Fredholm, B. B., Lindgren, E. and Lingstrom, K. (1985) Br. J.
Pharmacol. 86: 509-513
Gavish, M., Goodman, R. R. and Synder, S. H. (1982) Science
215: 1633-1635
Gerlach, E., Deuticke, B. and Dreisbach, R. H. (1963)
Naturwissenschaft 50: 228-229
Gerwins, P. and Fredholm, B. B. (1991) Mol. Pharmacol. 40:
149-155


6
exercise) greatly increase adenosine production (Olsson and
Pearson, 1990) .
Adenosine can also arise from ATP which is released and
rapidly broken down by ectonucleotidases. ATP is released
from nerve endings (where it is stored in vesicles along with
biogenic amines or other classical neurotransmitters), from
platelets (where it is stored in secretory granules along
with ADP), and from cells that are undergoing lysis. These
sources of adenosine probably are important under specific
circumstances (i.e., at particular synapses or at sites of
injury) (Olsson and Pearson, 1990; Bruns, 1990).
Another intracellular source of adenosine is S-adenosyl-
homocysteine (SAH), which arises from S-adenosylmethionine.
SAH-hydrolase catalyzes the reversible reaction between SAH
and adenosine plus homocysteine. Adenosine also tightly binds
to SAH-hydrolase. Hence, under basal conditions, the
intracellular concentration of free adenosine is probably
very low (Delahaba and Cantoni, 1959; Olsson and Pearson,
1990) .
Adenosine crosses cell membranes by simple diffusion and
and more importantly by facilitated diffusion. Facilitated
diffusion is carrier mediated, nonconcentrative and is
inhibited by dipyridamole (DIP) (Kolassa et al., 1970), 6-S-
(p-nitrobenzyl-thio)inosine (Paterson and Oliver, 1971) and
dilazep (Bruns, 1990). The carrier appears to transport other
nucleosides, which are competitive inhibitors of adenosine
transport. The carrier is also symmetrical, mediating both


INTERACTION OF ADENOSINE RECEPTORS
IN A SMOOTH MUSCLE CELL LINE
BY
FAN XIE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1992


61
phosphorylation of Gi may prevent the basal release of the
activated 0Ci subunit.
Because Ai-AdoR inhibition of cAMP accumulation
dominated over A2~AdoR mediated stimulation, it was of
interest to investigate whether the endogenous agonist ADO
would have a differential desensitization effect on the
receptor subtypes when both were chronically stimulated.
After pretreatment of cells with 100 JIM ADO for 24 hr, the
concentration response of CPA to inhibit ISO-stimulated cAMP
accumulation shifted to the right indicating that the Ai-AdoR
system was desensitized (Figure 3-26). This was similar to
that observed with CCPA-pretreatment. However, WRC-0018 did
not increase cAMP accumulation in ADO-pretreated cells,
suggesting that the A2~AdoR system was also desensitized
(Figure 3-27). In keeping with this conclusion, ADO also did
not stimulate cAMP accumulation in ADO-pretreated cells. The
observation that the Ai-AdoR was still present (albeit at
reduced agonist sensitivity) whereas the A2~AdoR response was
abolished indicates that the extent of desensitization was
different for each receptor subtype. This may be explained by
one or more factors: 1) the density of Ai-AdoR in DDT cells
was higher than that of A2~AdoR. An Ai-AdoR to A2~AdoR ratio
of 4:1 in DDT cells has been reported by Ramkumar et al.
(1990), 2) the coupling efficiency between the receptors and
their second messenger system may be different, 3) the rate
of desensitization of A2~AdoR was much faster than that of Ai-
AdoR. Ramkumar et al. (1991) reported that the ti/2 for the


CHAPTER 3
INTERACTION OF ADENOSINE RECEPTORS
Introduction
Cell surface adenosine receptors (AdoRs) have been
classified by pharmacological and biochemical criteria. Two
subtypes of AdoRs i.e., A^- and A2~AdoR have been thus far
clearly identified. In most cell types studied to date, the
Ai-AdoR mediates an inhibition of AC activity whereas the A2-
AdoR mediates a stimulation of the enzyme (Van Calker et al.,
1979; Londos et al., 1980). In general, receptors that
mediate inhibition of cAMP formation dominate over receptors
that mediate stimulation. For example, in mouse atria,
carbachol antagonized ISO-stimulated cAMP accumulation by
direct activation of the muscarinic receptors. The
interaction between carbachol and ISO was not competitive,
since cholinergic inhibition could not be surmounted by
increasing concentrations of ISO (Brown, 1979). In atria
isolated from rats, carbachol decreased the ISO-induced
elevation of cAMP levels and inhibited the positive
chronotropic and inotropic responses to ISO (Endoh et al.,
1985). In addition, after desensitization of the muscarinic
system in AtT-20 cells by oxotremorine, cAMP accumulation
stimulated by ISO was approximately doubled (Heisler et al.,
35


101
Similar to ISO, C-Br stimulated cAMP accumulation in DDT
cells was markedly attenuated by the selective Ai-AdoR
agonist CPA. This attenuation was not reversed by high
concentrations of either BAR agonists but was largely blocked
by the selective Ai-AdoR antagonist CPX. These data indicated
that the inhibition of C-Br stimulated cAMP accumulation was
mediated by an Ai-AdoR. Furthermore, the concentration of CPA
to inhibit cAMP accumulation by equipotent and equieffeetive
concentrations of ISO and C-Br was similar suggesting that
activation of the BAR by C-Br did not alter the the
inhibitory effect of CPA.
In the experiments discussed above, the incubation time
used was relatively short (6 min). Because the irreversible
binding of C-Br to BAR was time-dependent (Standifer et al.,
1989), the responses observed at the end of short periods of
incubation were probably due to both reversible and
irreversible interactions of this agonist with the BAR. To
investigate the irreversible agonist effects of C-Br, the BAR
antagonist PROP was employed to compete off any C-Br which
was reversibly bound to the BAR. After a 3 min incubation
with C-Br alone, the addition of an excess of PROP had no
subsequent effect on the rate of cAMP formation as compared
to that in the presence of C-Br alone. In contrast, when PROP
and CPA were added after a 3 min incubation with C-Br alone,
further accumulation of cAMP was inhibited. Although CPA
attenuated C-Br stimulated cAMP accumulation, it did not
affect the irreversible binding of C-Br to the BAR. This


45
remaining receptors labeled with [3H]CPX (control, 0.5 nM;
CCPA-pretreated, 0.4 nM).
The concentration-response relationships of CPA, NECA
and WRC-0018 induced attenuation on ISO-stimulated cAMP
accumulation in control and CCPA-pretreated cells are shown
in Figures 3-20, 3-21 and 3-22, respectively. In control
cells, CPA decreased cAMP content in a concentration-
dependent manner with an EC50 value of 1.2 nM and maximal
inhibition of 78% (Figure 3-20). In comparison, after
pretreatment of the cells with CCPA, the EC50 value for CPA-
induced inhibition on ISO-stimulated cAMP accumulation was
increased by 13-fold (EC50, 15 nM) without any significant
change in the maximal inhibition. Pretreatment of the cells
with 0.1 (IM CCPA for a longer period (68 hr) did not cause a
further shift to the right of the concentration response
curve of CPA and had no effect on the maximal response of
this agonist (data not shown). As illustrated in Figure 3-21,
in control cells, NECA decreased cAMP content in a
concentration-dependent manner with an EC50 value of 7.2 nM
and maximal inhibition of 75%. After pretreatment of the
cells with CCPA, the EC50 value for NECA-induced inhibition on
ISO-stimulated cAMP accumulation was increased by 17-fold
(EC50, 120 nM) without any significant change in the maximal
inhibition. The effect of pretreatment of the cells with CCPA
on WRC-0018 induced attenuation on ISO-stimulated cAMP
accumulation is depicted in Figure 3-22. In control cells,
WRC-0018 decreased the cAMP content with an EC50 value of 830


118
Nelson, C. A., Muther, T. F., Pitha, J. and Baker, S. P.
(1986) J. Pharmacol. Exp. Ther. 237: 830-836
Nickerson, M. (1956) Nature (Lond.) 178: 697-698
Norris, J. S., Cramer, D. J., Brown, F., Popovich, K. and
Cornett, L. E. (1983) J. Recept. Res. 3: 623-645
Norris,
422-424
J. S.
and
Kohler, P.
0.
(1974) Nature (Lond.)
248:
Norris,
613-618
J. S.
and
Kohler, P.
0.
(1977) Endocrinology
100:
Ohisalo,
249-251
J. J
. and Stouffer,
J.
E. (1979) Biochem. J.
178:
Olsson,
R. A.
and
Pearson, J
. D
. (1990) Physiol. Rev.
70 (2
761-845
Paterson, A. R. P. and Oliver, J. M. (1971) Can. J. Biochem.
49: 271-274
Pedersen, S. E. and Ross, E. M. (1986) J. Biol. Chem.
260(26): 14150-14157
Pelleg, A. and Porter, S. R. (1990) Pharmacotherapy 10(3):
157-174
Pittman, R. N., Reynolds, E. E. and Molinoff, P. B. (1984) J.
Pharmacol. Exp. Ther. 230(3): 614-618
Posner, P., Peterson, C. F., Pitha, J. and Baker, S. P.
(1984) Eur. J. Pharmacol. 100: 373-376
Ramkumar, V., Barrington, W. W., Jacobson, K. A. and Stiles,
G. L. (1990) Mol. Pharmacol. 37: 149-156
Ramkumar, V., Olah, M. E., Jacobson, K. A. and Stiles, G. L.
(1991) Mol. Pharmacol. 40: 639-647
Ramkumar, V., Pierson, G. and Stiles, G. L. (1988) Prog. Drug
Res. 32: 195-247
Ramkumar, V.and Stiles, G. L.(1988) J. Pharmacol. Exp. Ther.
246: 1194-1200
Rapiejko, P. J. and Malbon, C. C. (1987) Biochem. J. 241:
765-771
Rodbell, M. (1980) Nature 284: 17-22


41
effect of N-0861 on WRC-0018 stimulation of cAMP accumulation
is shown in Figure 3-11. The 3-fold increase in cAMP
accumulation caused by 0.1 |1M WRC-0018 was not affected by N-
0861 (0.1 nM 10 [1M) Based on these results, in the
remaining experiments of this series, 10 (1M N-0861 was used
to selectively block the Ai-AdoR mediated inhibition of cAMP
accumulation.
The effects of N-0861 on the A2~AdoR mediated response
of selective and nonselective agonists were investigated.
Figure 3-12 illustrates the concentration response of WRC-
0018 in the absence and presence of N-0861. The biphasic
concentration response curve of WRC-0018 in the absence of N-
0861 was replotted from Figure 3-4. N-0861 (10 H-M) completely
abolished the Ai~AdoR mediated inhibition of cAMP
accumulation caused by WRC-0018 and hence, the downward
component of the biphasic concentration response curve of
WRC-0018 was eliminated. The EC50 for the A2~AdoR mediated
effect of WRC-0018 on cAMP accumulation was 93 nM. The
maximal stimulation of cAMP accumulation by WRC-0018 was
increased (p<0.05) from 81 -fold in the absence to 156 -
fold in the presence of N-0861. The effects of NECA on cAMP
accumulation in the absence and presence of N-0861 are shown
in Figure 3-13. In the absence of N-0861, NECA did not
stimulate cAMP accumulation. In contrast, in the presence of
N-0861 (10 (1M) NECA stimulated cAMP accumulation over the
concentration range of 1 }1M 100 |1M. At concentrations of
NECA >10 HM, there was a decrease in the cAMP level despite


55
from the cells (Doss et al., 1981). In DDT cells, down-
regulation of BAR occurred rapidly with a ti/2 of about 3 hr
and proceeded to 80-85% loss of receptors by 13 hr of
incubation of the cells with 10 |IM epinephrine (Toews, 1987) .
Desensitization of Ai-AdoR system was studied by
examining the concentration responses of CPA, NECA and WRC-
0018 to inhibit ISO-stimulated cAMP accumulation in control
and CCPA-pretreated cells. For CPA (Figure 3-20) and NECA
(Figure 3-21), the concentration response curves were shifted
to the right (10-20 -fold) without any change in the maximal
responses. Longer pretreatment periods did not increase the
magnitude of shift (data not shown) indicating that maximal
desensitization was achieved. The shift in the concentration
response is not surprising because the sensitivity of the
cells to the agonist should decrease as the receptor number
decreases provided little or no change in the affinity of the
agonist for the remaining receptors. The ability of CPA and
NECA to produce the same maximal response in control
(untreated) cells and CCPA-pretreated cells where the
receptor number is decreased by almost 50% may be explained
by the presence of spare Ai-AdoRs. In this situation, the
responsiveness is not directly proportional to receptor
occupancy and the maximal response can be obtained when less
than 100% of the receptors are activated. Many studies have
shown the existence of spare receptors in other systems
(Stephenson, 1956; Nickerson, 1956; Nelson et al., 1986;
Gunst et al., 1989). For example, in guinea pig lung, after


37
in Figure 3-1. First, selective blockade of the Ax-AdoR;
second, uncoupling of the Ax-AdoR with PTX and third, down-
regulation and/or desensitization of the Ax-AdoR using a
selective agonist.
Results
Effects of Selective and Nonselective Adenosine Receptor
Agonists on cAMP Accumulation in DDT Cells
Experiments were designed to investigate the effects of
the selective Ax-AdoR agonist CPA, the putative nonselective
agonists NECA and Adenosine (ADO), and the selective A2~AdoR
agonist WRC-0018 on cAMP accumulation in DDT cells.
Figure 3-2 illustrates the concentration-dependent
inhibition of ISO (10 (J.M)-induced cAMP accumulation by CPA,
NECA and ADO in DDT cells. The effect of ADO was studied in
the presence of DIP (inhibitor of adenosine transporter) and
ENHA (inhibitor of adenosine deaminase) to prevent the uptake
and degradation of this nucleoside respectively during the 7
min incubation period. The rank order of potency of these
AdoR agonists to inhibit ISO-induced cAMP accumulation was
CPA > NECA > ADO with EC50 values of 1.5, 14 and 97 nM,
respectively. The maximal inhibition of cAMP accumulation
caused by the AdoR agonists was 89% for CPA, 97% for NECA and
79% for ADO. Figure 3-3 illustrates the concentration
response of ISO to stimulate cAMP accumulation in the absence
and presence of CPA (0.1 JIM) or NECA (1 (IM) These


78
u.E
is
E 2
3 O
8*
< U)
CL E
o £
Q.
Figure 3-13. The effect of N-0861 on NECA-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 [1M rolipram, the indicated concentrations of
NECA, without or with 10 N-0861 for 7 min at 36C. At the
end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The basal
level of cAMP accumulated was 11 pmol/mg protein/min. Each
data point is the meanSD of quadruplicate determinations and
is representative of 2 experiments. The indicates a
p<0.0005 for the NECA points from their respective control
values.


69
T¡£
lE
1.1
E£
3 O
O *
o a.
< U)
CL E
11
Log [Drug], M
Figure 3-4. The effect of WRC-0018 and NECA on cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram and the indicated concentrations of
WRC-0018 or NECA for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. The basal level of cAMP
accumulated was 31, 32 pmol/mg protein/min and each data
point is the meanSE, n=15 for WRC-0018 and n=3 for NECA. The
* indicates a p<0.01 for each WRC-0018 data point as compared
to its basal level.


104
inhibited forskolin-stimulated cAMP accumulation in DDT
cells. Forskolin activates AC in a receptor-independent
manner which may involve direct interaction with the
catalytic subunit of the enzyme (Seamon and Daly, 1986).
Although the mechanism for Ai-AdoR mediated inhibition of
basal and forskolin-stimulated cAMP accumulation in DDT cells
has not been fully delineated, evidence from other cell lines
suggest that the subunits of Gi may be involved. Thus, OCi
released by activation of the Ai-AdoR may directly attenuate
the activity of the catalytic subunit of the enzyme, or the
13+Y subunits of Gi may complex with free as to form the
inactive Gs complex (Katada et al., 1984a, b; Jakobs and
Schultz, 1983; Roof et al., 1986). It is conceivable that
inhibition may involve more than one mechanism acting in
concert or a single mechanism may dominate depending on the
mode of enzyme activation. Further studies will be necessary
to elucidate these possibilities.
Over the past few years, significant effort has been
expended in developing insurmountable receptor ligands as
investigative tools (Posner et al., 1984; Homburger et
al.,1984; Nelson et al., 1986; Baker et al., 1986) and more
recently for therapeutics (Ullman and Svedmyr, 1988). These
agents have the advantage of very slow dissociation rates
from their receptors and hence may produce a long duration of
action. In addition, once the drug-receptor complex is
formed, the response will be independent of the plasma
concentration which can be allowed to fall. This may


54
increase cAMP content 1.5 fold above the basal level in DDT
cells pretreated with 200 ng/ml PTX for 4 hr (Gerwins et al.,
1990). Our data strongly suggest that the downward phase for
WRC-0018 concentration response is due to activation of an
inhibitory receptor. The EC50 values of the A2~AdoR effect of
WRC-0018 in the presence of N-0861 (93 nM) and after PTX-
pretreatment (90 nM) were very similar. In addition,
similarity also exists between the maximal responses of the
A2~AdoR effect of WRC-0018 in the presence of N-0861 (156 -
fold above basal) and after PTX-pretreatment (143 -fold
above basal). These results suggest that N-0861 and PTX have
a similar net blocking effect on the Ai-AdoR mediated
inhibitory action of WRC-0018 and are unlikely to affect the
interaction of WRC-0018 with the A2~AdoR. As a consequence,
the A2~AdoR mediated stimulatory action of WRC-0018 is
sustained to the same extent.
Selective desensitization and/or down-regulation of Ai-
AdoR was the third approach used. Chronic pretreatment (16
hr) of cells with the selective Ai-AdoR agonist CCPA (0.1 |1M)
caused a 48% loss of Ai-AdoR. This indicates that the Ai-AdoR
is down-regulated after chronic pretreatment with a selective
Ai-AdoR agonist. Down-regulation of receptors after long term
exposure to an agonist is a widely reported phenomenon.
However, the loss of Ai-AdoR during down-regulation is
significantly less in comparison with other receptor systems.
For instance, after 1321N1 astrocytoma cells were incubated
with ISO for 12-24 hr, greater than 90% of the BARS were lost


43
stimulatory effect of ISO by 84%. Pretreatment of cells with
> 25 ng/ml PTX for 18 hr resulted in a complete loss of the
CPA inhibitory effect on cAMP accumulation. Basal and 10 (IM
ISO-stimulated cAMP accumulation were not significantly
affected after pretreatment of the cells with PTX. Based on
these results, in the remaining experiments of this series,
DDT cells were pretreated for 18 hr with 25 ng/ml of PTX.
Figure 3-16 depicts the effects of WRC-0018 on cAMP
accumulation in control and PTX-pretreated cells. The
biphasic WRC-0018 concentration response in control cells was
replotted from Figure 3-4. After pretreatment of the cells
with PTX, similar to using Ai~AdoR antagonist, the A]_-AdoR
mediated inhibition of cAMP accumulation is blocked and
hence, the downward component of the concentration response
curve of WRC-0018 was abolished. The EC50 value of the A2~AdoR
mediated effect of WRC-0018 on cAMP accumulation was 90 nM.
The maximal stimulation of cAMP accumulation was 143 -fold
above basal in PTX-pretreated cells (control cells, 81 -fold
above basal). The effect of NECA on cAMP accumulation in
control and PTX-pretreated cells is shown in Figure 3-17. The
data for NECA in control cells were replotted from Figure 3-
13. In PTX-pretreated cells, NECA (10 nM 10 |IM) stimulated
cAMP accumulation with an EC50 value of 180 nM. The maximal
response was 5-fold above the basal level. Figure 3-18
illustrates the effect of ADO on cAMP accumulation in control
and PTX-pretreated cells. The data for ADO in control cells
were replotted from Figure 3-14. Over the concentration range


ACKNOWLEDGEMENTS
First and foremost, I wish to sincerely thank my mentor,
Dr. Stephen Baker, for his guidance, both professional and
personal, and for the contribution he has made to my
education. I would also like to thank Dr. Luiz Belardinelli
for his enthusiasm, valuable suggestions and constant supply
of wonderful drugs. Many thanks are also expressed to my
helpful and friendly committee: Mohan Raizada, Edwin Meyer
and Thomas Rowe. I wish to thank all members of Dr. Baker's
and Dr. Belardinelli's laboratory, especially Debbie Otero,
Dr. John Shryock, Mary Anne Locksmith, Cheryl Spence and
Xingmin Tang for their support and technical assistance. I
also deeply thanks Drs. Allen Neims, David Silverman,
Chingkuang Tu and Thomas Muther for their confidence in me. A
special word of thanks goes to Dr. Sandra Rattray for her
editorial help. My best wishes are also extended to all my
fellow (and former fellow) graduate students especially
Sukanya Kanthawatana, Nelida Sjak-Shie, Walter Folger, Daniel
Danso, Jiahui Zhang and Magdalena Wozniak; thanks for their
friendship and encouragement. I would also like to thank
other faculty members who worked so hard to improve the
graduate program, and the secretarial and administrative
staff who keep the department running. Special thanks are
iii


28
Dulbecco's modified Eagle's media (DMEM) and fetal bovine
serum were from Gibco (Grand Island, NY, USA). Liquiscint was
purchased from National Diagnostics (Somerville, NY, USA). 2-
[2-(2-Naphthyl)ethoxy]adenosine (WRC-0018) was a kind gift of
Dr. Ray. A. Olsson (Univ. of South Florida, Tampa, FL, USA).
()N^-endonorbornan-2-yl-9-methyladenine (N-0861) was a gift
of Whitby Research, Inc. (Richmond, VA, USA). Pertussis toxin
was a gift of Dr. Eric Hewlett (Univ. of Virginia,
Charlottesville, VA, USA). Rolipram was a gift of Berlex
Laboratories (Cedar Knolls, NJ, USA). 5[2[ [3[4
(bromoacetamido)phenyl]-2-methylprop-2-yl]amino]-1-
hydroxyethyl]-8-hydroxycarbostyril (C-Br) was synthesized as
described previously (Milecki et al., 1987). All other
reagents were from Sigma Chemical Co. (St. Louis, MO, USA) or
Fisher Scientific (Orlando, FL, USA).
Methods
Cell Culture
The DDT cell line was derived from a steroid-induced
leiomyosarcoma tumor of the vas deferens of an adult Syrian
hamster (Norris and Kokler, 1974). These cells were obtained
at low passage number and grown as a monolayer on 150 mm
plastic culture dishes (Falcon) in DMEM supplemented with 5%
fetal bovine serum, 100 U/ml penicillin G, 0.1 mg/ml
streptomycin and 2.5 (Ig/ml amphotericin B in an atmosphere of
5% C02/95% air at 37C. Cells were seeded at 0.2-1 x 104


117
Londos, C. and Wolff, J. (1977) Proc. Natl. Acad. Sci. USA
74: 5482-5486
Londos, C., Wolff, J. and Cooper, D. M. F. (1983)
Physiological and Regulatory Functions of Adenosine and
Adenine Nucleotides, ed by Baer, H. P. and Drummond, G. I.,
New York: Raven Press, pp 271-281
Londos, C., Wolff, J. and Cooper, D. M. F. (1981) Purinergic
Receptors, ed by Burnstock, G., London: Chapman and Hall, pp
287-323
Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, R.
J. (1951) J. Biol. Chem. 193: 265-275
Lupica, C. R., Cass, W. A., Zahniser, N. R. and Dunwiddie, T
V. (1990) J. Pharmacol. Exp. Ther. 252(3): 1134-1141
Maenhaut, C., Van Sande, J., Libert, F., Abramowicz, M.,
Parmentier, M., Vanderhaegen, J. J., Dumont, J. E., Vassart,
G. and Schiffmann, S. (1990) Biochem. Biophys. Res. Commun.
173: 1169-1178
Malbon, C. C., Moreno, F. J., Cabelli, R. J. and Fain, J. N.
(1978) J. Biol. Chem. 253: 671-678
Martens, D. M., Lohse, J., Rauch, B. and Schwabe, U. (1987)
Naunyn-Schmiedeberg's Arch. Pharmacol. 336: 342-348
Milecki, J., Baker, S. P., Standifer, K. M., Ishizu, T.,
Chida, Y., Kusiak, J. W. and Pitha, J. (1987) J. Med. Chem.
30: 1563-1566
Mills, I. and Gewirtz, H.
Commun. 168(3): 1297-1302
Mitsuhashi, M. and Payan,
134: 367-375
Morgan, N. G. (1989) Cell
New York: Guilford Press,
Munshi, R. and Linden, J.
14589
(1990) Biochem.
D. G. (1988) J.
Signalling, ed.
pp. 63-66
(1989) J. Biol.
Biophys. Res.
Cellular Physiol,
by Morgan, N. G.,
Chem. 264: 14853-
Nakata, H. (1989a) J. Biol. Chem. 264: 16545-16551
Nakata, H. (1989b) Mol. Pharmacol. 35: 780-786
Nathanson, N. M., Klein, W. L. and Nirenberg, M. (1978) Proc
Natl. Acad. Sci. USA 75(4): 1788-1791
Nazarea, M., Okajima, F. and Kondo, Y. (1991) Eur. J.
Pharmacol.- Mol. Pharmacol. Section 206: 47-52


53
accumulation. This attenuation may be due to the higher
concentrations of NECA and ADO overcoming the N-08 61 (10 (1M)
blockade of the Ai-AdoR allowing its activation.
The second approach used to investigate the expression
of A2~AdoR mediated response was uncoupling of the Ai-AdoR
inhibitory effect with PTX. It is well established that PTX
selectively ADP-ribosylates the subunit. This covalent
modification inactivates the Gi protein and uncouples
inhibitory receptors including the Ai-AdoR resulting in loss
of receptor mediated inhibition of AC activity (Gilman, 1987;
Nathanson, 1987; Hazeki and Ui, 1981). Experiments showed
that the inhibition of AC activity by (R)-PIA in DDT cells
was greatly attenuated after pretreatment of the cells with
100 ng/ml PTX for 18 hrs (Ramkumar et al., 1990). In the
present study, the ability of CPA to inhibit ISO-stimulated
cAMP accumulation was used as means to determine the effect
of PTX-pretreatment on the Ai-AdoR mediated inhibition of
cAMP accumulation. PTX (25 ng/ml)-pretreatment for 18 hr was
found to be sufficient to cause complete loss of the CPA
inhibitory effect. In PTX-pretreated cells, the downward
phase of the WRC-0018 biphasic response was completely
eliminated. Furthermore, NECA and ADO which had no effect on
cAMP accumulation in untreated cells stimulated cAMP
accumulation in cells pretreated with PTX. The result with
NECA, i.e., the stimulation of cAMP accumulation in PTX-
pretreated cells, is also consistent with other reports
(Gerwins et al., 1990) That is, NECA (10 (1M) was reported to


LIST OF REFERENCES
Baker, S. P., Liptak, A. and Pitha, J. (1985) J. Biol. Chem.
260: 15820-15828
Baker, S. P., Sumners, C., Pitha, J. and Raizada, M. K.
(1986) J. Neurochem. 47: 1318-1326
Belardinelli, L., Giles, W. R. and West, G. A. (1988) J.
Physiol. (Lond.) 405: 615-633
Belardinelli, L., Isenberg, G. (1983) Am. J. Physiol. 224:
H734-H737
Belardinelli, L and Lerman, B. B. (1990) Br. Heart J. 63: 3-4
Belardinelli, L., Linden, J. and Berne, R. M. (1989) Prog.
Cardiovasc. Dis. 32(1): 73-97
Berne, R. M. (1963) Am. J. Physiol. 204: 317-322
Berne, R. M. (1986) News Physiol. Sci. 1: 163-167
Biaggioni, I., Killian, T. J., Mosqueda-Garcia, R.,
Robertson, R. M. and Robertson, D. (1991a) Circulation 83:
1668-1675
Biaggioni, I., Paul, S., Puckett, A. and Arzubiaga, C.
(1991b) J. Pharmacol. Exp. Ther. 258(2): 588-592
Brittain, R. T., Dean, C. M. and Jack, D. (1981) Int. Encycl.
Pharmacol. Ther. 104: 613-652
Brown, J. H. (1979) J. Cyclic Nucleotide Res. 5(6): 423-433
Bruns, R. F. (1990) Ann. N.Y. Acad. Sci. 603: 211-226
Bruns, R. F., Daly, J. W. and Snyder, S. H. (1980) Proc.
Natl. Acad. Sci. USA 77(9): 5547-5551
Bruns, R. F., Fergas, J. H., Badger, E. W., Bristol, J. A.,
Santay, L. A., Hartman, J., Hays, S. J. and Huang, C. C.
(1987) Naunyn-Schmiedebergs Arch. Pharmacol. 335: 59-63
Bruns, R. F., Lu, G. H. and Pugsley, T. A. (1986) Mol.
Pharmacol. 29: 331-346
113


76
-a £
SE
1 =
E 2
3 O
8S-
< O)
CL E
il
Q.
Log [N-0861], M
Figure 3-11. The effect of N-0861 on the ability of WRC-0018
to stimulate cAMP accumulation in DDT cells. Cells were
incubated in HBSS containing 50 JIM rolipram, 0.1 p.M WRC-0018,
without or with the indicated concentrations of N-0861 for 7
min at 36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The basal level of cAMP accumulated was 61 pmol/mg
protein/min. Each data point is the meaniSD of quadruplicate
determinations.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EBCIX22Y9_PU0O72 INGEST_TIME 2015-03-27T19:37:47Z PACKAGE AA00029753_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


86
Log [ECA], M
Figure 3-21. The effect of CCPA pretreatment on the ability
of NECA to inhibit ISO-stimulated cAMP accumulation in DDT
cells. Cells were incubated in fresh growth media without or
with 0.1 (1M CCPA for 16 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached. The cells were then incubated in HBSS
containing 50 JIM rolipram, 10 JIM ISO or ISO plus the
indicated concentrations of NECA for 7 min at 36C. The cAMP
accumulated was determined as described in the "Methods"
section. Each data point is the meanSD of quadruplicate
determination.


This dissertation is dedicated to my grandmother, father,
mother and brother.


83
Log [Adenosine], M
Figure 3-18. The effect of PTX pretreatment on ADO-stimulated
cAMP accumulation in DDT cells. Cells were incubated in fresh
growth media without or with 25 ng/ml PTX for 18 hr at 37C.
At the end of the incubation period, the cells were washed 4
times with ice-cold HBSS and detached. Cells were then
incubated in HBSS containing 50 [iM rolipram, 1 [IM DIP, 1 (1M
EHNA and the indicated concentrations of ADO for 7 min at
36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The control data are taken from Figure 3-14. The
basal level of cAMP accumulated was 60 and 41 pmol/mg
protein/min for the control and PTX experiments,
respectively. Each data point is the meantSE, n=3. The *
indicates a p<0.05 for the ADO point from their respective
control value.


85
Log [CPA], M
Figure 3-20. The effect of CCPA pretreatment on the ability
of CPA to inhibit ISO-stimulated cAMP accumulation in DDT
cells. Cells were incubated in fresh growth media without or
with 0.1 JIM CCPA for 16 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached. The cells were then incubated in HBSS
containing 50 |IM rolipram, 10 flM ISO or ISO plus the
indicated concentrations of CPA for 7 min at 36C. The cAMP
accumulated was determined as described in the "Methods"
section. Each data point is the meanSD of quadruplicate
determinations and is representative of 2 experiments.


14
AdoR has yet to be cloned. There is no published report about
the binding and functional characteristics of RDC7 in an
expression system.
Receptor-Guanine Nucleotide Regulatory Protein Coupling;
Adenosine receptors modulate AC activity via G proteins,
the stimulatory Gs and the inhibitory Gf for A2~AdoR and Ai~
AdoR, respectively.
The activation of AC by receptors coupled to Gs can be
described by the ternary complex model. This model has been
developed for the BAR system and is likely to be applicable
for other stimulatory receptors including the A2~AdoRs. In
brief, an agonist (Ag) binds to the receptor (R) to form an
Ag-R complex. The affinity of the agonist in the Ag-R complex
is relatively low. The complex then undergoes a
conformational change and interacts with a Gs protein to form
the ternary complex of Ag-R-Gs. The affinity of the agonist
in the Ag-R-Gs complex is relatively high. When GTP binds to
Gs, Ag-R-Gs is rapidly converted to GS-GTP and Ag-R. Once
formed, GS-GTP interacts with the catalytic subunit (C) of AC
to form the active complex C-GS-GTP resulting in a conversion
of ATP to cAMP. The enzyme activity returns to basal level
when guanosine 5'-triphosphatase (GTPase) activity in Gs
hydrolyzes the bound GTP to guanosine diphosphate (GDP) with
the resultant regeneration of inactive catalytic unit and Gs-
GDP. The destabilization of the ternary complex decreases


64
Another implication of our study with potential
therapeutic value is the possibility to decrease the side
effects caused by nonselective AdoR agonists. For instance,
adenosine and its analogs may activate both receptor subtypes
and hence, if one subtype can be selectively blocked, the
side effects mediated by this receptor subtype should be
attenuated.
The development of more selective agonists and
antagonists may be needed to increase the concentration range
over which a response can be achieved. Thus, the ultimate
objective would be to develop specific agonists and
antagonists.
Finally, in cell types in which the Ai~ and A2~AdoR
coexist, the question arises as to whether an A2~AdoR
mediated response initiated by ADO would ever be expressed
under physiological or pathological conditions? Although Ai~
AdoR suppression of the A2~AdoR response may be normal under
most physiological conditions (at least in cells that express
both receptor subtypes), the A2~AdoR response may be needed
under some pathological conditions. For example, under
stress, the chronic release of adenosine or other cellular
mechanisms may result in the loss of the Ai-AdoR or other
components of its signal transduction pathway (e.g., Gi
protein) allowing A2~AdoR expression. Although our data
indicated that chronic stimulation of Ai~ and A2~AdoR resulted
in desensitization of both receptors and Ramkumar et al.
(1991) found that the rate of desensitization for the A2~AdoR


60
Assuming a 10-20 fold reduction in ADO sensitivity (like CPA)
to inhibit cAMP accumulation in cells chronically pretreated
with CCPA, this would not be sufficient to allow selective
expression of the A2~AdoR response. However, it will be of
interest to determine if selective down-regulation of the Ai~
AdoR would reduce the sensitivity of a non-selective agonist
sufficient to uncover expression of the A2~AdoR response.
Interestingly, CCPA-pretreatment was found to increase
the basal cAMP level (Figure 3-24). Desensitization of
inhibitory receptors resulting in an increase in basal cAMP
levels or potentiation of stimulatory receptor effects has
been widely reported. For example, a 3-fold increase in basal
cAMP level was observed after pretreatment of NG108-15 cells
with 10 (1M carbachol for 19 hr (Nathanson et al., 1978). This
increase in basal cAMP level may be due to the loss of an
inhibitory tone on AC activity mediated by the Gi protein.
The inhibitory tone may involve the basal dissociation of Gi
(in the absence of an inhibitory agonist) with the subunit
directly attenuating the activity of the catalytic subunit of
AC (Gilman, 1987) Several mechanisms may be responsible for
the loss of the inhibitory tone. First, chronic CCPA-
pretreatment may decrease the cellular content of Gi protein.
Evidence supporting this contention has been reported by
Green (1987) who showed a decrease in Gi protein after
desensitization of the Ai-AdoR system in primary culture of
rat adipocytes. Second, CCPA-pretreatment may induce an
impairment in the function of Gi. For example, a CCPA-induced


46
nM and caused a maximal inhibition of 77%. After pretreatment
of the cells with CCPA, the inhibition of ISO-stimulated cAMP
accumulation by WRC-0018 was markedly attenuated with a
maximal inhibition of cAMP accumulation of about 20%.
The effects of pretreatment of cells with CCPA on the
A2~AdoR mediated effect of WRC-0018 is shown in Figure 3-23.
The control data from Figure 3-4 were replotted in Figure 3-
23. In CCPA-pretreated cells, the WRC-0018 mediated
attenuation of cAMP accumulation was abolished. The EC50 of
the WRC-0018 mediated stimulation of cAMP accumulation was 17
nM. The maximal stimulation of cAMP accumulation was not
significanly increased (control, 81 -fold above basal; CCPA-
pretreated, 30 -fold above basal). The effects of
pretreatment of the cells with CCPA on the A2~AdoR mediated
response of NECA is shown in Figure 3-24. The data for NECA
in control cells were replotted from Figure 3-13. In CCPA-
pretreated cells, NECA did not stimulate cAMP accumulation,
however, the basal level of cAMP was significantly increased
(p<0.0005) from 32 in control cells to 111 pmol/mg
protein/min in pretreated cells. A longer time of
preincubation of the cells with CCPA (68 hr) also failed to
unmask a NECA mediated A2~AdoR mediated increase in cAMP
accumulation (data not shown). The effects of pretreatment of
the cells with CCPA on the A2~AdoR mediated response of ADO
is shown in Figure 3-25. The data for ADO in control cells
were reploted from Figure 3-14. In CCPA-pretreated cells, ADO
did not stimulate cAMP accumulation.


115
Gerwins, P., Nordstedt, C. and Fredholm, B. B. (1990) Mol.
Pharmacol. 38: 660-666
Gilman, A. G. (1987) Ann. Rev. Biochem. 56: 615-649
Goodman, R. R., Cooper, M. J., Gavish, M. and Snyder, S. H.
(1982) Mol. Pharmacol. 21: 329-335
Green, A. (1987) J. Biol. Chem. 262: 15702-15707
Gunst, S. J., Stropp, J. Q. and Flavahan, N. A. (1989) J.
Appl. Physiol. 67(3): 1294-1298
Harden, T. K. (1983) Pharmacol. Rev. 35 (1): 5-32
Hazeki, O. and Ui, M. (1981) J. Biol. Chem. 256(6): 2856-2862
Heisler, S., Desjardins, D. and Nguyen, M. H. (1985) J.
Pharmacol. Exp. Ther. 232(1): 232-238
Hill, A. V. (1913) Biochem. J. 7: 471-480
Hill, S. J. and Kendall, D. A. (1987) Br. J. Pharmacol. 91:
661-669
Hollingsworth, E. B. and Daly, J. W. (1985) Biochim. Biophys.
Acta 847: 207-216
Homburger, V., Pantaloni, C., Lucas, M., Gozlan, H. and
Bockaert, J. (1984) J. Cell. Physiol. 121: 589-597
Hutchison, A. J., Webb, R. L., Oei, H. H., Ghai, G. R.,
Zimmerman, M. B. and William, M. (1989) J. Pharmacol. Exp.
Ther. 251: 47-55
Hutchinson, A. J., Williams, M., Jesus, R. D., Yokoyama, R.,
Oei, H. H., Ghai, G. R., Webb, R. L., Zoganas, H. C., Stone,
G. A. and Jarvis, M. F. (1990) J. Med. Chem. 33: 1919-1924
Jacobson, K. A., Ukena, D., Kirk, K. L. and Daly, J. W.
(1986) Proc. Natl. Acad. Sci. USA 83: 4089-4093
Jakobs, K. H. and Schultz, G. (1983) Proc. Natl. Acad. Sci.
USA 80: 3899-3902
Jarvis, F. M., Schulz, R., Hutchison, A. J., Do, U. H.,
Sills, M. A. and Williams, M. (1989) J. Pharmacol. Exp. Ther.
251(3): 888-893
Jarvis, F. M. and Williams, M. (1989) Eur. J. Pharmacol. 168:
243-246
Katada, T., Bokoch, G. M.,Smigel, M. D., Ui, M. and Gilman,
A. G. (1984a) J. Biol. Chem. 259: 3586-3595


38
concentrations of Ai-AdoR agonists caused maximal inhibition
of ISO-stimulated cAMP accumulation (Figure 3-2). ISO alone
increased cAMP accumulation in a concentration-dependent
manner with an EC50 of 2.4 nM and a maximal stimulation of 62-
fold above the basal level achieved at 0.1 [1M. In the
presence of CPA or NECA, ISO still stimulated cAMP
accumulation in a concentration-dependent manner, but the
maximal response was decreased by 75% and 60% in the presence
of CPA and NECA, respectively. In the presence of 1 |1M CPX, a
selective Ai-AdoR antagonist, the inhibitory effect of CPA on
ISO-stimulated cAMP accumulation was significantly attenuated
(data not shown). Thus, our results indicated that the Ai-
AdoR agonists had an inhibitory effect on cAMP accumulation
and this effect was mediated by Ai-AdoR.
The effects of the selective A2~AdoR agonist WRC-0018
(Ueeda et al., 1991) and the nonselective agonist NECA on
cAMP accumulation are illustrated in Figure 3-4. WRC-0018
produced a biphasic response whereas NECA caused no effect on
cellular cAMP level. WRC-0018 at low concentrations (5 500
nM) stimulated cAMP accumulation with a maximal response of
81 -fold above the basal level. The estimated EC50 was 8.6
nM. However, higher concentrations (>500 nM) of WRC-0018
decreased cAMP accumulation with an EC50 value of 5.1 )1M. In
contrast to WRC-0018, NECA over the entire concentration
range of 0.1 nM 10 (1M did not affect (increase or decrease)
cellular cAMP content. Similar to NECA, ADO (0.1 nM 10 (1M)


20
capable of reducing AC activity in the eye S49 cell mutant.
These cells lack Gsa, and logically, li+y is not inhibitory
when reconstituted with eye" membranes. It was also
demonstrated that the isolated Gitt from rat liver can inhibit
AC activity in membranes from eye- S49 cell. The inhibitory
effect of Gia was therefore proposed to explain the ability
of inhibitory agonists to decrease AC activity in the cyc-
S49 cell mutant (Katada et al., 1984b; Jakobs and Schultz,
1983). This inhibitory effect of Gia-GTPyS has also been
observed by Roof et al. (1986) in the bovine central nervous
system.
Regulation of Adenosine Receptors
Similar to many other receptors, the AdoR appears to
undergo desensitization and down-regulation during chronic
exposure to an agonist. This effect prevents overstimulation
of the receptor. The mechanism(s) of desensitization and
down-regulation has/have been studied extensively in the J5AR-
AC system and may be applicable to the AdoR. Desensitization
describes the phenomenon where an initial exposure of a cell
to an agonist results in a reduced capacity of the cell to
respond to a second challenge. Two main types of
desensitization have been described. "Homologous"
desensitization is referred to hormone-specific type where
loss of response is only to the activated receptor, whereas
other receptor mediated responses remain unaffected.


effect is not mediated by alteration of agonist interaction
with the BAR but rather occurs via a post-receptor mechanism.
Vlll


91
Log [CPA], M
Figure 3-26. The effect of ADO pretreatment on the ability of
CPA to inhibit ISO-stimulated cAMP accumulation in DDT cells.
Cells were incubated in fresh growth media with 5 |1M DIP, 1
(1M EHNA, without or with 100 (1M ADO for 24 hr at 37C. At the
end of the incubation period, the cells were washed 4 times
with ice-cold HBSS and detached. The cells were then
incubated in HBSS containing 50 |!M rolipram and 10 [1M ISO or
ISO plus the indicated concentrations of CPA for 7 min at
36C. The cAMP accumulated was determined as described in the
"Methods" section. Each data point is the meanSE, n=5.


50
lower maximal inhibition of AC activity in DDT cell membranes
may be due to the homogenization process during membrane
preparation affecting the function of Ai-AdoR.
Our experiments also suggest that the stimulation of
cAMP by low concentrations of WRC-0018 was mediated by an A2-
AdoR. Evidence supporting this include the inhibitory effect
of the AdoR antagonist 8PST and lack of effect of the
selective Ai-AdoR antagonist N-0861 on the stimulatory
response of WRC-0018. Since a selective A2~AdoR antagonist
has yet to be synthesized, the next best candidate, 8PST,
which is only slightly more selective to Ai-AdoR with a
K(A2)/K(Ai) ratio of 5.9 (Trivedi et al., 1990) was used.
Interestingly, low concentrations of WRC-0018 which
increase cAMP accumulation failed to potentiate ISO-
stimulated cAMP accumulation (Figure 3-7). This may be
because A2~AdoR and BAR share the same pool of AC for cAMP
production and the system may be maximally stimulated at 10
¡1M ISO. Also observed was a greater stimulatory effect of ISO
on cAMP accumulation (3 times greater) as compared with WRC-
0018. This finding could be explained by 1) a higher density
of BAR on DDT cells, 2) a higher coupling efficiency of BAR
for the Gs protein or 3) an easier access of BAR to the AC
pool.
The Ai-AdoR agonist CPA inhibited the A2~AdoR
stimulatory effect of WRC-0018 in our experiment suggesting
that the inhibitory Ai-AdoRs dominate over the stimulatory A2-
AdoRs in DDT cells. This dominant role of Ai-AdoR may also


3
organs. For example, in the heart, adenosine is a potent
coronary vasodilator (Berne, 1963; Gerlach et al., 1963). It
has also been shown to depress cardiac activity, e.g., (1)
depress sinoatrial and atrioventricular (AV) node activity,
(2) reduce atrial contractility, (3) attenuate the
stimulatory actions of catecholamines primarily in
ventricular myocardium, and (4) depress ventricular
automaticity (Belardinelli et al., 1989). These actions
characterize adenosine as an endogenous cardioprotective
substance whose actions lead to an increase in oxygen supply
and decrease in cardiac work. Together, these actions tend to
restore the balance between oxygen supply and demand.
In most tissues including skeletal muscle, adenosine is
a vasodilator. Thus, intravenous infusion of adenosine causes
hypotension. However, the ultimate cardiovascular effects of
adenosine in vivo depends on the dose, rate and mode of
administration, and on the autonomic reflexes triggered as a
result of adenosine's direct action (Pelleg and Porter,
1990) .
In the kidney, adenosine produces vasoconstriction of
the afferent glomerular artery, a decrease in glomerular
filtration rate and inhibition of renin release (Pelleg and
Porter, 1990). Adenosine is a depressant of the respiratory
center and it causes bronchoconstriction (Pelleg and Porter,
1990). However, adenosine also stimulates arterial
chemoreceptors (Biaggioni et al., 1991a). Hence, when given
intravenously, adenosine causes hyperventilation (Biaggioni


116
Katada, T., Northup, J. K., Bokoch, G. M., Ui, M. and Gilman,
A. G. (1984b) J. Biol. Chem. 259: 3578-3585
Kendall, D. A. and Hill, S. J. (1988) J. Neurochem. 50: 497-
502
Kenimer, J. G. and Nirenberg, M. (1981) Mol. Pharmacol. 20:
585-591
Kent, R. S., DeLean, A. and Lefkowitz, R. J. (1980) Mol.
Pharmacol. 17: 14-23
Klotz, K. N. and Lohse, M. J. (1986) Biochem. Biophys. Res.
Commun. 140: 406-413
Kolassa, N., Pfleger, K. and Rummel, W. (1970) Eur. J.
Pharmacol. 9: 265-268
Kurachi, Y., Nakajima, T. and Sugimoto, T. (1986) Pflugers
Arch. 407: 264
Lad, P. M., Nielsen, T. B., Londos, C., Preston, M. S. and
Rodbell, M. (1980) J. Biol. Chem. 255: 10841-10846
Levitzki, A. (1987) Biochem. Pharmacol. 27: 2083-2088
Libert, F., Parmentier, M., Lefort, A., Dinsart, C., Van
Sande, J., Maenhaut, C., Simons, M. J., Dumont, J. E. and
Vassart, G. (1989) Science 244: 569-572
Liles, W. C. and Nathanson, N. (1986) J. Neurochem. 46: 89-95
Linden, J. (1991) FASEB J. 5: 2668-2676
Linden, J. and Delahunty, T. M. (1989) Trends Pharmacol. Sci.
10: 114-120
Linden, J., Patel, A. and Sadek, S. (1985) Circ. Res. 56:
279-284
Lindner, F. and Rigler, R. (1931) Pfluegers Arch. 226: 697-
708
Lohse, M. J., Klotz, K. N., Schwabe, U., Cristalli, G.,
Vittori, S. and Griffantini, M. (1988) Naunyn-Schmiedeberg's
Arch. Pharmacol. 337: 687-689
Lohse, M. J., Lenschow, V., Schwabe, U. (1984) Mol.
Pharmacol. 26: 1-9
Londos, C., Cooper, D. M. F. and Wolff, J. (1980) Proc. Natl.
Acad. Sci. USA 77: 2551-2554


39
also did not affect the cellular cAMP level above the basal
level (data not shown).
Figure 3-5 depicts the effect of the non-selective AdoR
antagonist 8PST on the stimulatory effect of WRC-0018. WRC-
0018 (0.1 |IM) induced a 3-fold increase in cAMP content above
the basal level and 8PST (5 |1M) inhibited the WRC-0018-
induced cAMP accumulation by 80%. Figure 3-6 illustrates a
similar effect of the Ai-AdoR agonist CPA on the stimulatory
effect of WRC-0018. At a concentration of 0.1 |1M, WRC-0018
stimulated cAMP accumulation 3-fold above the basal level and
CPA (1 |J.M) inhibited the WRC-0018-induced cAMP accumulation
by 78%.
Figure 3-7 illustrates the effects of WRC-0018 on cAMP
accumulation in the absence and presence of 10 [1M ISO. The
biphasic concentration response curve of WRC-0018 was
replotted from Figure 3-4. In the presence of ISO, low
concentrations of WRC-0018 (1 nM 100 nM) did not affect
ISO-stimulated cellular cAMP accumulation. However, at higher
concentrations (>500 nM), WRC-0018 attenuated the stimulatory
effect of ISO in a concentration-dependent manner with an EC50
of 840 nM and a maximal inhibition of 73%.
These data show that nonselective AdoR agonists (NECA
and ADO) themselves only inhibit cAMP accumulation. In
contrast, the selective A2~AdoR agonist (WRC-0018) stimulated
at lower and inhibited cAMP accumulation at higher
concentrations, thereby resulting in a biphasic concentration
response curve. To further investigate the expression of the


UNIVERSITY OF FLORIDA


5
Experimental studies mainly in animal models have
indicated several potential uses of adenosine agonists and
antagonists. The agonists could be used as antiepileptic,
analgesic and sedative drugs due to their inhibitory effect
on neurotransmission (Pelleg and Porter, 1990).
AdoR antagonists could be used for the relief of AV
block associated with acute myocardial infarction. In
addition, they accelerate recovery of myocardial
contractility during cardioversion (Wesley and Belardinelli,
1989) .
Synthesis and Metabolism of Adenosine
Adenosine is a local hormone (or autacoid) rather than a
circulating hormone or neurotransmitter. It acts within the
same organ (s), perhaps even on the cell(s), that is the site
of its production. Unlike neurotransmitters, adenosine can be
produced by virtually any cell. Adenosine per se does not
appear to be stored in exocytotic vesicles, but rather is
produced on demand, much like prostaglandins and
leukotrienes. The primary mechanism for the production of
adenosine in heart muscle, liver and leukocytes is the
dephosphorylation of AMP by a 5'-nucleotidase located on the
cell membrane or in the cytosol. Adenosine produced is then
released into the interstitial space from the parenchymal
cells for receptor interaction. Physiological stimuli that
cause inadequate tissue oxygenation (e.g., hypoxia, ischemia,


87
Log [WRC-0018], M
Figure 3-22. The effect of CCPA pretreatment on the ability
of WRC-0018 to inhibit ISO-stimulated cAMP accumulation in
DDT cells. Cells were incubated in fresh growth media without
or with 0.1 (1M CCPA for 16 hr at 37C. At the end of the
incubation period, the cells were washed 4 times with ice-
cold HBSS and detached. The cells were then incubated in HBSS
containing 50 ^.M rolipram, 10 |1M ISO or ISO plus the
indicated concentrations of WRC-0018 for 7 min at 36C. The
cAMP accumulated was determined as described in the "Methods"
section. Each data point is the meanSD of quadruplicate
determinations and is representative of 2-4 experiments.


19
dissociation and the release of sufficient quantities of the
B+Y complex. The large increase in the amount of free B+Y
complex in the membrane would disturb the equilibrium which
exists between undissociated and dissociated Gs under resting
conditions. Thus, by mass action, B+Y complex would combine
with the released stimulatory a subunits preventing its
dissociation and subsequent activation of AC. This model,
therefore, implies that Gi will only be effective under
conditions where the activity of AC is stimulated, i.e., the
effectiveness of Gi is related to the level of the
dissociated Gsa (Morgan, 1989). Experiments supporting this
model showed that when platelet membranes were treated for
brief periods of time with GTPyS and an 0C2-adrenergic agonist
in low Mg2+ conditions, AC was "irreversibly" inhibited. This
inhibition was of the same magnitude as that produced by
maximally effective concentrations of B+Y complex, and it was
not additive with the effect of B+Y- This inhibition is
completely overcome by reconstitution of the membranes with
physiological concentrations of Gia~GDP. The most likely
explanation for this observation is the interaction between
Gia-GDP and Gjj+y to relieve the inhibition caused by free B+Y
complex in the membrane (Katada et al., 1984a).
The third is the direct interaction model. It involves
inhibition of AC by the released OCi subunits acting directly
on the catalytic subunit of the enzyme (Gilman, 1987).
Evidence favoring this model over the subunit dissociation
model includes the observation that inhibitory agonists are


TABLE OF CONTENTS
ACKNOWLEDGMENTS iii
ABSTRACT vi
CHAPTERS
1 INTRODUCTION 1
Functions of Adenosine 1
Synthesis and Metabolism of Adenosine 5
Adenosine Receptors 7
Regulation of Adenosine Receptors 20
Goals 24
2 EXPERIMENTAL PROCEDURES
Source of Materials 27
Methods 2 8
3 INTERACTION OF ADENOSINE RECEPTORS
Introduction 35
Results 37
Discussion 49
Summary 62
4 INHIBITORY EFFECT OF Ai-ADENOSINE RECEPTOR ON
IRREVERSIBLE ACTIVATION OF THE A-ADRENORECEPTOR
Introduction 94
Results 96
Discussion 99
LIST OF REFERENCES 113
BIOGRAPHICAL SKETCH 121
v


antagonist () N6-endonorbornan-2-yl-9-methyladenine (N-0861,
10 (1M) 2) pretreatment of cells with pertussis toxin (PTX,
25 ng/ml, 18 hr) which uncoupled the inhibitory Ai-AdoR
response or 3) pretreatment of cells with the selective Ai-
AdoR agonist 2-chloro-N6-cyclopentyladenosine (CCPA, 0.1 JIM,
16 hr). CCPA-pretreatment reduced by 13-fold the potency of
the Ai-AdoR agonist N6-cyclopentyladenosine (CPA) to inhibit
ISO-stimulated cAMP formation, and decreased the Ai-AdoR
level by 48%. Stimulation of cAMP accumulation by adenosine
and NECA was uncovered in the presence of N-0861 and by PTX-
pretreatment. However, no stimulation by either agonist was
observed after CCPA-pretreatment. The data indicate that the
inhibitory Ai-AdoR response in DDTi MF-2 cells is predominant
and masks the A2~AdoR mediated stimulatory effect. The A2~AdoR
response was expressed by a selective A2~AdoR agonist or
under conditions where the function of the Ai-AdoR is
blocked.
The ability of the activated Ai-AdoR to modulate agonist
interaction with the beta-adrenoreceptor (BAR) was studied
using the irreversible BAR agonist, 5 [2 [ [3[4
(bromoacetamido)phenyl]-2-methylprop-2-yl]amino]-1-
hydroxyethyl]-8-hydroxycarbostyril (C-Br). Activation of the
Ai-AdoR attenuated cAMP accumulation of the permanently
stimulated BAR, and did not alter the irreversible binding of
C-Br. In addition, CPA decreased basal cAMP level and had no
effect on the interaction of the reversible BAR agonist ISO
with the BAR. These data indicate that Ai-AdoR inhibitory
Vll


90
Log [Adenosine], M
Figure 3-25. The effect of CCPA pretreatment on ADO-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in fresh growth media without or with 0.1 |1M CCPA
for 16 hr at 37C. At the end of the incubation period, the
cells were washed 4 times with ice-cold HBSS and detached.
Cells were then incubated in HBSS containing 50 |1M rolipram,
1 p.M DIP, 1 p.M EHNA and the indicated concentrations of ADO
for 7 min at 36C. At the end of the incubation period, the
cAMP accumulated was determined as described in the "Methods"
section. The control data are taken from Figure 3-14. The
basal level of cAMP accumulated was 60 and 73 pmol/mg
protein/min for the control and CCPA experiments,
respectively. Each data point is the meanSE, n=3.


30
washed four times with 10 ml of ice-cold HBSS without
divalent cations.
Membrane Preparation
DDT cells were harvested from culture dishes in 5 ml of
50 mM Tris-HCl buffer at pH 7.4 containing 5 mM MgCl2 with a
rubber policeman and were pelleted by centrifugation at
48,000g for 15 min.
For the determination of Ai-AdoR, the pellet was
resuspended in ice-cold 50 mM Tris-HCl (pH 7.4) containing 1
mM MgCl2, 1 mM EDTA and 0.1 mM benzamidine (trypsin
inhibitor) and then homogenized with a Ten Broeck Tissue
Grinder (glass-glass). The homogenate was centrifuged at
48,000g for 15 min to pellet the membranes. The membranes
were homogenized a second time in 50 mM Tris-HCl (pH 7.4)
containing 1 mM MgCl2/ then used for protein measurement and
receptor binding assays.
For determination of the JAR, the pellet from the first
centrifugation as described above was homogenized in ice-cold
50 mM Tris-HCl (pH 7.4) containing 5 mM MgCl2 with a Tekmar
SDT-100EN homogenizer (setting 5, 20 s). After centrifugation
at 48,000g for 15 min and homogenization with the Tekmar as
above, the membrane suspension was used for assays.


42
the presence of N-0861 in the incubation medium. Similar
results to those of NECA were observed with ADO (Figure 3-
14) ADO (0.1 nM 10 p.M) caused no effect on the cellular
cAMP content in the absence of N-08 61 whereas ADO (1 |1M 5
(1M) stimulated cAMP accumulation in the presence of 10 |1M N-
0861. At 10 |1M ADO, there was a decrease in cAMP formation
despite the presence of N-0861 in the incubation medium.
In summary, these data show that the selective Ai-AdoR
antagonist N-0861 abolishes the downward phase of the WRC-
0018 biphasic concentration response and uncovers a
stimulatory (i.e., A2~AdoR mediated) response of NECA and
ADO.
Effect of PTX on the A2~AdoR Mediated Response
The second experimental approach used to unmask, the A2-
AdoR mediated effect on cAMP accumulation was uncoupling of
Ai inhibitory effects with PTX. Based on a previous study
which showed that inhibition of AC activity by (R)-PIA was
markedly attenuated after an 18 hr pretreatment of DDT cells
with 100 ng/ml PTX (Ramkumar et al., 1990), I chose to
incubate our DDT cells with PTX for 18 hr. The ability of CPA
to inhibit ISO-stimulated cAMP accumulation in cells
pretreated with various concentrations of PTX for 18 hr is
shown in Figure 3-15. In control (untreated) DDT cells, 10 (1M
ISO produced a 15-fold increase in cAMP content above the
basal level. As expected, CPA (1 (1M) attenuated the


105
translate into a reduced incidence of side effects. In the
case of agonists, the recently introduced BAR agonist
salmeterol has been shown to produce a sustained airway
dilation probably due to an extremely tight binding to the
BAR (Brittain et al., 1981; Oilman and Svedmyr, 1988). On the
other hand, one potential problem with slow dissociating
agonists is modulating the response produced under conditions
where the receptor is activated in an antagonist-insensitive
manner. Therefore, administration of an antagonist may have
no effect on the response. The data from the present study
suggest that the sustained response mediated by C-Br can be
controlled by activation of inhibitory receptor. Further
development of agonists that produce sustained responses may
need to take into account the interaction of inhibitory and
stimulatory receptors to fully exploit their therapeutic
potential.


68
-a £
gE
1 =
E
3 O
8S-
< O)
CL E
o c
CL
Log [Isoproterenol], M
Figure 3-3. The effect of CPA and NECA on ISO-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 |IM rolipram, the indicated concentrations of
ISO, without or with 0.1 }1M CPA or 1 p.M NECA for 7 min at
36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The basal level of cAMP accumulated was 11 pmol/mg
protein/min. Each data point is the meanSE, n=3 5.


29
cells/cm2 and subcultured twice weekly after detachment using
1 mM ethylenediamine tetraacetic acid (EDTA) in phosphate
buffered saline (PBS). DDT cells have a doubling time of
about 20 hours and a confluent density of 1.3 x 105 cell/cm2.
Experiments were performed on cells 1-day pre-confluent.
Drug Preparation
Stock solutions of WRC-0018 (10 mM) and rolipram (50 mM)
were prepared in dimethylsulfoxide (DMSO). CPA (1 mM), CCPA
(1 mM), DIP (10 mM) and C-Br (1 mM) were prepared in ethanol.
ECA (1 mM) was dissolved in 5 mM HC1 and N-0861 (1 mM) was
dissolved in a mixture of ethanol (10%) and 50 mM Tris buffer
containing 10 mM MgCl2 (90%). These stock solutions were
diluted in Hank's Balanced Salt Solution (HBSS) without
divalent cations to the desired concentrations just prior to
use. HBSS contains 137 mM NaCl, 6 mM D-glucose, 5 mM KC1, 4
mM NaHCC>3, 0.6 mM Na2HP04, 0.4 mM KH2PO4, 0.5 mM MgCl2, 0.4 mM
MgS04 and 1 mM CaCl2 at pH 7.4. All other drugs were dissolved
in HBSS before use.
Drug Treatment
The growth medium in culture dishes was aspirated and
fresh medium (20 ml) was then added followed by the drug. The
cells were then incubated at 37C for the various period of
times as indicated in the text. At the end of the incubation
period, the media was aspirated and the attached cells were


103
be operative in cardiac cells whereas other inhibitory
mechanisms may exist in DDT cells. Alternatively, during the
process of DDT cell membrane preparation, the inhibitory
effects on agonist binding may have become uncoupled.
The pathway of receptor activation initially involves a
reversible interaction of the agonist (Ag) with the receptor
(R) to form an Ag-R complex. The Ag-R complex then undergoes
a conformational change to form an activated Ag-R complex
which interacts with Gs. Although difficult to establish
directly, there is some intriguing kinetic and structural
evidence to suggest that the BAR undergoes a conformational
change induced by agonist (Contreras et al., 1986; Pedersen
and Ross, 1985) An inhibitory mechanism could therefore
involve preventing the Ag-R complex from becoming
conformationally active. However, the data with the
irreversible agonist are not consistent with this proposed
mechanism. C-Br produced antagonist-insensitive activation of
the BAR suggesting that the Ag-R complex is in a permanently
activated state, but the stimulation of cAMP accumulation is
still inhibited by CPA. If CPA mediated inhibition involved
inactivation of the BAR, then it would be expected that CPA
would have no effect on the irreversible component of C-Br-
stimulated cAMP formation.
Data from the present study are also consistent with an
inhibitory action beyond the receptor level. The basal cAMP
accumulated in the absence of a BAR agonist was also
inhibited by CPA. Furthermore, in experiments not shown, CPA


10
Radioligand binding studies of AdoRs have been attempted
within the past decade and some success has been achieved,
particularly with Ai-AdoR ligands. The first successful
radioligand binding studies of AdoRs were reported in the
early 1980s. Several groups used a variety of tritiated or
iodinated radioligands including both agonists and
antagonists (Linden et al., 1985; Trost and Schwabe, 1981;
Bruns et al., 1980; Williams and Risley, 1980). Radioligand
binding studies in membrane preparations from various tissues
revealed all the appropriate characteristics; that is 1)
saturability, 2) reversibility, 3) stereoselectivity and 4)
the pharmacological specificity expected of the
physiologically relevant receptor. A recent development in
this field has been the synthesis of the high affinity Ai-
AdoR selective radioligand [3H]8-cyclopentyl-l,3-dipropyl-
xanthine ([3H]CPX) (Bruns et al., 1987). This Ai-AdoR
antagonist has a 740-fold Ai-AdoR selectivity over A2~AdoR,
the highest selectivity reported for an adenosine antagonist.
CPX also has very high affinity for the Ai-AdoR (Kp = 0.4 nM)
and extremely low non-specific binding (=3% of total binding)
in rat brain membranes.
Although the availability of agonist and antagonist
radioligands has enabled detailed characterization of the Ai-
AdoR in various tissues (Stiles et al., 1985; Jacobson et
al., 1986; Ramkumar and Stiles, 1988; Martens et al. 1987),
the lack of a highly selective A2~AdoR antagonist has
hampered a similar characterization of the A2~AdoR. Because


56
=50% of BARs were inactivated by an irreversible antagonist,
the maximal airway responsiveness to ISO was still maintained
(Nelson et al., 1986). In canine trachealis muscle, the
maximal contractile response for acetylcholine was achieved
when only 4% of muscarinic receptors were occupied (Gunst et
al., 1989) .
Interestingly, in control cells, WRC-0018 produced the
same maximal inhibitory response on ISO-stimulated cAMP
accumulation as CPA. However in CCPA-pretreated cells, the
maximal response produced by WRC-0018 was greatly reduced
(Figure 3-22) whereas the maximal response achieved by CPA
was not changed from that in control cells. This indicated
that WRC-0018 acted as a full Ai-AdoR agonist (as compared
with CPA) in control cells whereas it acted as a partial
agonist in pretreated cells. One explanation for this may be
related to the intrinsic efficacy of the agonists. In control
cells, WRC-0018 may need to occupy more receptors than CPA to
achieve its maximal response. Thus, for WRC-0018, there may
be little or no spare Ai-AdoRs and the responsiveness may be
more directly related to receptor occupancy. As the receptor
number decreases, the maximal response for WRC-0018 will be
reduced as was observed after chronic CCPA-pretreatment. A
desensitization-induced change in agonist efficacy has also
been reported for some BAR agonists. In membranes prepared
from untreated L6 skeletal muscle cells, several compounds
acted as full BAR agonists when compared with ISO, but after
desensitization of the BAR system, these compounds acted as


63
AdoR activation of AC and thereby dampen a rise in cellular
cAMP.
The data from the present study show that in DDT cells
activation of the inhibitory Ai-AdoR will predominate and
thus mask the stimulatory A2~AdoR response. This observation
may have broader implications because other cell types and
tissues have been reported to express both Ai~ and A2~AdoRs.
These include porcine coronary vascular smooth muscle cells
(Mills and Gewirtz, 1990), FRTL-5 cells derived from normal
rat thyroid (Nazarea et al., 1991), ventricular myocytes from
chick embryo (Xu et al., 1992) and heart tissue (Olsson and
Pearson, 1990) .
Our data demonstrate that cells with both receptor
subtypes can be pharmacologically manipulated to uncover an
A2~AdoR response with a highly selective A2~AdoR agonist or
with the use of a selective Ai-AdoR antagonist. There may be
situations where these pharmacological approaches have
therapeutic potential and suggest possible strategies for
drug development. For example, in cells or tissues where both
receptor subtypes are expressed, the A2~AdoR mediated
responses may be suppressed under normal conditions. However,
by selectively blocking the Ai-AdoR system or activating the
A2~AdoR with a highly selective agonist, A2~AdoR mediated
responses (e.g., vasodilation to increase blood flow,
inhibition of platelet aggregation during thrombosis,
generation of superoxide radicals to prevent reperfusion
injury) may be expressed for possible therapeutic effect.


32
95% of total binding. All assays were performed in
triplicate, and the determinations differed by less than 6%.
BARS were quantitated by specific [125I]CYP binding.
Membrane protein (30-50 |lg) was incubated in a total volume
of 0.25 ml with 50 mM Tris-HCl buffer at pH 7.4, 5 mM MgCl2
and 6-100 pM [125I]CYP, with or without 3 |1M ()-alprenolol,
for 60 min at 36C. The bound and free ligand were then
rapidly separated on GF/B glass fiber filters using a Brandel
Cell Harvester. Filters were rinsed three times with 4 ml of
ice-cold 50 mM Tris-HCl buffer containing 5 mM MgCl2, placed
in omnivials and the radioactivity was determined in a gamma
counter. Specific binding was typically 80-90% of total
binding. All assays were performed in triplicate, and the
determinations differed by less than 6%.
p.AMP Assay
Cells were detached from culture dishes with a cell
lifter and centrifuged at 500g for 5 min. The cells (0.4 mg
protein/tube = 2xl06 cells) were gently resuspended in HBSS
containing 50 )iM rolipram (phosphodiesterase inhibitor) and
incubated at 36C for 7 min in Beckman microcentrifuge tubes.
Drugs were then added and the cells were incubated at 36C
for the various period of times as indicated in the text. At
the end of the incubation period, the tubes were immediately
placed in a boiling water bath for 5 min. The protein was


108
a
"5 .E
a
i2
3a
O U)
< E
i!
< Q.
O <3
BASAL
C-Br
ISO
^ C-Br+CPA
m iso+cpa
C-Br+CPA+CPX
m iso+cpa+cpx
Figure 4-3. The effect of CPX on the inhibition of ISO- and
C-Br-stimulated cAMP accumulation in DDT cells by CPA. Cells
were incubated in HBSS containing 100 JIM rolipram, 1 p.M C-Br
10 ISO, C-Br + 1 JIM CPA, ISO + 1 |IM CPA, C-Br + CPA + 5 JIM
CPX, or ISO + CPA + 5 JIM CPX for 6 min at 37C. At the end of
the incubation period, the cAMP accumulated was determined as
described in the "Methods" section. Each data point is the
meanSE, n=4.


47
In summary, these data show that after CCPA-
pretreatment, the selective A2~AdoR agonist WRC-0018 causes a
sustained A2~AdoR mediated stimulatory effect on cAMP
accumulation. However, the A2~AdoR mediated stimulatory
effects of the nonselective agonists NECA and ADO were still
not uncovered after pretreatment of DDT cells with CCPA.
Effect of Adenosine on the Desensitization o.f_^Ai~ and^-AdoK
Systems
The endogenous agonist for the AdoR is ADO which has
been shown to be a nonselective agonist (Olsson and Pearson,
1990; Londos et al., 1980). In a separate series of
experiments, I investigated whether ADO had differential
effects on the desensitization of AdoR subtypes and hence,
determined if the expression of the A2~AdoR mediated response
could be uncovered. Cells were pretreated with 100 (J.M ADO to
ensure that both receptor subtypes were stimulated with the
agonist.
Desensitization of Aj^-AdoR
The effect of ADO on the desensitization of Ai-AdoR was
investigated by studying the effect of CPA to inhibit ISO-
stimulated cAMP accumulation in control (untreated) and ADO
(100 |1M, 24 hr)-pretreated cells (Figure 3-26) DIP and EHNA
were present during the period of pretreatment of the cells
with ADO to prevent ADO metabolism and thereby maintain ADO
concentration in the incubation medium relatively constant.
As depicted in Figure 3-26, in control cells, CPA decreased


33
pelleted by centrifugation at 9,OOOg for 2 min, and the
supernatants were saved for cAMP assays.
The cAMP content of the supernatant was determined by a
modification of a competitive protein binding assay described
previously (Baker et al., 1985). An aliquot (usually 50 |ll)
of the supernatant was incubated in a total volume of 0.2 ml
with 25 mM Tris-HCl buffer at pH 7.0, 8 mM theophylline, 0.8
pmol of [3H]cAMP and 24 |ig of bovine heart cAMP dependent
protein kinase at 4C for 60 min. At the end of the
incubation, 70 p.1 of a 50% (v/v) hydroxyapatite suspension
was added to each tube. The suspensions were then poured onto
a Whatman GF/C glass fiber filter under reduced pressure. The
filters were rinsed three times with 4 ml of ice-cold 10 mM
Tris-HCl buffer and placed in minivials with 3 ml of
Liquiscint. Radioactivity was determined in a liquid
scintillation counter. The amount of cAMP present was
calculated from a standard curve determined using known
concentrations of unlabeled cAMP.
Data Analysis
Receptor density (Bmax) and dissociation constant (Kd)
for the radiolabeled ligands were determined from regression
analysis of Scatchard plots (1949). The concentrations of
compounds which inhibited ligand binding by 50% (IC50) were
obtained from Hill plots of the competition data (Hill,
1913). The effective concentrations of drugs which gave 50%


98
Effect of CPA on the Insurmountable Component of C-3r-induc.£ii
Stimulation
Figure 4-5 illustrates the time course of C-Br-induced
increase in cAMP content and the effects of PROP and CPA. C-
Br increased the cAMP content over a 10 min incubation
period. After 3 min of incubation with C-Br, the addition of
PROP (20 (IM) had no effect on the subsequent rate of cAMP
accumulation. In contrast, the addition of PROP and CPA (1
(1M) 3 min after incubation with C-Br resulted in complete
inhibition of further cAMP accumulation during the next 7 min
of incubation. In fact, after addition of CPA, cAMP levels
decreased.
The effect of CPA on the irreversible binding of C-Br to
the BAR is depicted in Figure 4-6. This is a representative
Scatchard plot of specific [125I]CYP binding after
pretreatment of cells with C-Br, CPA and C-Br + CPA for 10
min at 37C. CPA pretreatment did not alter the specific
[125I]CYP binding as compared to control (control, 54 pmol/ mg
protein; CPA, 59 pmol/mg protein). After incubation of cells
with 1 (IM C-Br followed by 6 cell wash cycles, there was a
42% decrease in specific [125I]CYP binding and this decrease
was not affected by the presence of CPA during the 10 min
preincubation period (C-Br, 28 pmol/mg protein; C-Br + CPA,
30 pmol/mg protein). There was also no change in the Kq value
for [125I]CYP among the 4 groups (control, 34 pM; C-Br, 32 pM;
CPA, 25 pM; C-Br + CPA, 26 pM).


72
O -
¡|
3 C
£'
32
< ^
S: E
< O
e
a
Log [WRC-0018], M
Figure 3-7. Effect of WRC-0018 on ISO-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 }IM rolipram, the indicated concentrations of
WRC-0018, without or with 10 |1M ISO for 7 min at 36C. At the
end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The control
cAMP accumulated in the presence of ISO alone was 629
pmol/mg protein/min. Each data point is the meantSE, n=4.


40
A2~AdoR mediated cAMP accumulation, three experimental
approaches were used (Figure 3-1). These were 1) selective
blockade of Ai-AdoR, 2) uncoupling of Ai inhibitory effects
with PTX, and 3) selective desensitization and/or down-
regulation of Ai-AdoR.
Effect of a Selective A^-AdoR Antagonist on the A?~AdoR
Mediated Response
The first experimental approach used to uncover the A2-
AdoR mediated effect on cAMP accumulation was blockade of the
Ai-AdoR with a selective Ai-AdoR antagonist. The ability of
the highly selective Ai-AdoR antagonist, N-0861 (Shryock et
al., 1992), to compete with [3H]CPX for the Ai-AdoR binding
site is shown in Figure 3-8. N-0861 produced a concentration-
dependent displacement of specific [3H]CPX binding with an
IC50 of 0.8 (IM and a Hill slope of 1.0. This indicated that N-
0861 bound to a single class of binding sites. Figure 3-9
illustrates the effect of N-0861 on the Ai-AdoR inhibitory
effect of CPA. ISO (10 |1M) produced a 12-fold increase in
cAMP accumulation above the basal level and CPA (0.1 fiM)
inhibited the stimulatory effect of ISO by 70%. N-0861 (10
JIM) attenuated the inhibitory effect of CPA by 62%. A similar
effect of N-0861 on NECA-induced inhibition of ISO-stimulated
cAMP accumulation is shown in Figure 3-10. That is, ISO (10
|1M) stimulated cAMP accumulation 47-fold above the basal
level and ECA (1 (1M) inhibited this increase by 74%. N-0861
(10 |1M) attenuated the inhibitory effect of NECA by 60%. The


106
Figure 4-1. The effect of CPA on ISO- and C-Br-stimulated
cAMP accumulation in DDT cells. Cells were incubated in HBSS
containing 100 (1M rolipram, the indicated concentrations of
ISO or C-Br, without or with 1 JIM CPA for 6 min at 37C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The basal
level of cAMP accumulated was 226, 143 pmol/mg protein/6
min for the ISO and C-Br experiments, respectively. Each data
point is the meanSE, n=4.


81
Log [WRC-0018], M
Figure 3-16. The effect of PTX pretreatment on WRC-0018-
stimulated cAMP accumulation in DDT cells. Cells were
incubated in growth media without or with 25 ng/ml PTX for 18
hr at 37C. At the end of the incubation period, the cells
were washed 4 times with ice-cold HBSS and detached. Cells
were then incubated in HBSS containing 50 |1M rolipram and the
indicated concentrations of WRC-0018 for 7 min at 36C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The dose
response curve of WRC-0018 in control cells is taken from
Figure 3-4. The basal level of cAMP accumulated was 31 and
21 pmol/mg protein/min for the control and PTX experiments,
respectively. Each data point is the meanSE, n=6.


100
much less is known about the mechanism whereby receptor
mediates inhibition of the enzyme. Similar to stimulatory
receptors, inhibitory receptors are coupled to a guanine
nucleotide binding protein (Gj.) which dissociates into an
and IJ+Y subunits in the presence of GTP. There is evidence
supporting several inhibitory mechanisms which include 1) a
direct interaction of OCi with the catalytic subunit of the AC
(Katada et al., 1984b; Jakobs and Schultz, 1983; Roof et al.,
1986), 2) fi+Y subunits released from Gi complexing with as to
favor the inactive Gs complex (Katada et al., 1984a) and 3)
prevention of stimulatory ternary complex formation (Romano
et al., 1988; Romano et al., 1989).
In the present study, the effect of the inhibitory Ai~
AdoR to attenuate the response of a permanently activated 1AR
was investigated. Initial experiments were performed to
partially characterize the actions of the irreversible BAR
agonist C-Br, using intact DDT cells. This compound was found
to be a potent stimulator of cAMP accumulation, with a
concentration required to produce half-maximal stimulation in
the subnanomolar range. In addition, C-Br was a full BAR
agonist since it produced the same maximal response as the
classical agonist, ISO. The potency and efficacy of C-Br
using intact cells is consistent with a previous report on
the effect of this compound to stimulate AC activity in
isolated membranes from rat reticulocytes (Standifer et al.,
1989).


79
"O £
2e
1 =
E£
3 O
8S.
< O)
CL E
il
o c
Log [Adenosine], M
Figure 3-14. The effect of N-0861 on ADO-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 ^IM rolipram, 1 (IM DIP, 1 (IM EHNA, the indicated
concentrations of ADO, without or with 10 |1M N-0861 for 7 min
at 36C. At the end of the incubation period, the cAMP
accumulated was determined as described in the "Methods"
section. The basal level of cAMP accumulated was 60 pmol/mg
protein/min. Each data point is the meanSE, n=4. The *
indicates a p<0.01 for the ADO points from their respective
control values.


107
Log [CPA], M
Figure 4-2. Inhibition of ISO- and C-Br-stimulated cAMP
accumulation in DDT cells by CPA. Cells were incubated in
HBSS containing 100 [1M rolipram, 10 ^M ISO or 1 |1M C-Br and
the indicated concentrations of CPA for 6 min at 37C. At the
end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The control
cAMP accumulated in the presence of ISO or C-Br alone was
64622, 66029 pmol/mg protein/6 min, respectively. The basal
level of cAMP accumulated was 194 pmol/mg protein/6 min.
Each data point is the meanSE, n=4.


93
15
o.E
1 =
ES
D O
ga
< O)
CL E
11
Q.
10
0
CONTROL
ADO+D1P+EHNA
10
-9
-8
-7
Log [Adenosine], M
Figure 3-28. The effect of ADO pretreatment on ADO-stimulated
cAMP accumulation in DDT cells. Cells were incubated in fresh
growth media without or with 5 p.M DIP + 1 |IM EHNA + 100 (1M
ADO for 24 hr at 37C. At the end of the incubation period,
the cells were washed 4 times with ice-cold HBSS and
detached. Cells were then incubated in HBSS containing 50 p.M
rolipram, 1 (1M DIP, 1 (1M EHNA and the indicated
concentrations of ADO for 7 min at 36C. At the end of the
incubation period, the cAMP accumulated was determined as
described in the "Methods" section. The basal level of cAMP
accumulated was 61, 61 pmol/mg protein/min for the control
and ADO+DIP+EHNA experiments, respectively. Each data point
is the meanSD of quadruplicate determination.


77
3 O
8S-
< O)
CL E
CL
Log [WRC-0018], M
Figure 3-12. The effect of N-0861 on WRC-0018-stimulated cAMP
accumulation in DDT cells. Cells were incubated in HBSS
containing 50 (1M rolipram, the indicated concentrations of
WRC-0018, without or with 10 |1M N-0861 for 7 min at 36C. At
the end of the incubation period, the cAMP accumulated was
determined as described in the "Methods" section. The dose
response curve of WRC-0018 in the absence of N-0861 is taken
from Figure 3-4. The basal level of cAMP accumulated in the
presence of N-0861 was 21 pmol/mg protein/min. Each data
point is the meanSE, n=3


CHAPTER 1
INTRODUCTION
Functions of Adenosine
History
Drury and Szent-Gyorgyi (1929) were the first to report
on the cardiovascular effects of adenosine and adenine
nucleotides. They described the isolation of crystalline
adenine from acid extracts of ox heart muscle. Adenosine was
crystalized from yeast nucleic acid hydrolysate. Intravenous
injection of either extract into different mamalian species
after atropinization produced primarily sinus bradycardia,
transient heart block and other physiological effects (Drury
and Szent-Gyorgyi, 1929).
Interestingly, Drury and Szent-Gyorgyi did not comment
on the physiological implications of their observations.
Within two years, however, Lindner and Rigler (1931) had
crystallized adenosine, obtained from the degradation of AMP,
from heart muscle extracts. They showed that adenosine was a
potent coronary vasodilator in a number of species. Based on
the findings that adenosine was present in heart muscle
extract and had potent vasoactive effects, Lindner and Rigler
advanced the hypothesis that adenosine is a physiological
1