Method of identifying partial agonists of the A2A receptor

Abstract

The present invention provides a method for identifying and using partial adenosine A2A receptor agonists that are useful as adjuncts in myocardiological perfusion imaging.

Description

FIELD OF THE INVENTION

[0002] This invention relates to a method of identifying compounds that are selective partial A2A-adenosine receptor agonists, preferably With a short duration of action. Such compounds provide coronary dilation in mammals without causing corresponding significant peripheral vasodilation. The invention also relates to a method of using such compounds as adjuncts in cardiac imaging.

BACKGROUND

[0003] Myocardial perfusion imaging (MPI) is a diagnostic technique useful for the detection and characterization of coronary artery disease. Perfusion imaging uses materials such as radionuclucides to identify areas of insufficient blood flow. In MPI, blood flow is measured at rest, and the result compared with the blood flow measured during exercise on a treadmill (cardiac stress testing), such exertion being necessary to stimulate blood flow. Unfortunately, many patients are unable to exercise at levels necessary to provide sufficient blood flow, due to medical conditions such as peripheral vascular disease, arthritis, and the like.

[0004] Therefore, a pharmacological agent that increases CBF for a short period of time would be of great benefit, particularly one that did not cause peripheral vasodilation. Vasodilators, for example dipyridamole, have been used for this purpose in patients prior to imaging with radionuclide. Dipyridamole is an effective vasodilator, but side effects such as pain and nausea limit the usefulness of treatment with this compound.

[0005] Adenosine, a naturally occurring nucleoside, also is useful as a vasodilator. Adenosine exerts its biological effects by interacting with a family of adenosine receptors characterized as subtypes A1, A2A, A2B, and A3. AdenoScan® (Fujisawa Healthcare Inc.) is a formulation of a naturally occurring adenosine. AdenoScan has been marketed as an adjuvant in perfusion studies using radioactive thallium-201. However, its use is limited due to side effects such as flushing, chest discomfort, the urge to breathe deeply, headache, throat, neck, and jaw pain. These adverse effects of adenosine are due to the activation of other adenosine receptor subtypes in addition to A2A, which mediates the vasodilatory effects of adenosine. Additionally, the short half-life of adenosine necessitates multiple treatments during the procedure, further limiting its use. AdenoScan is contraindicated in many patients including those with second-or third-degree block, sinus node disease, bronchoconstructive or bronchospastic lung disease, and in patients with known hypersensitivity to the drug.

[0006] Other potent and selective agonists for the A2A adenosine receptor are known. For example, MRE-0470 (Medco) is an adenosine A2A receptor agonist that is a potent and selective derivative of adenosine. VWRC-0470 (Medco) is an adenosine A2A agonist used as an adjuvant in imaging. These compounds, which have a high affinity for the A2A receptor, and, consequently, a long duration of action, which is undesirable in imaging.

[0007] Thus, there is still a need for a method of producing rapid and maximal coronary vasodilation in mammals without causing corresponding peripheral vasodilation, which would be useful for myocardial imaging with radionuclide agents.. Preferred compounds would be selective for the A2A adenosine receptor and have a short duration of action (although longer acting than compounds such as adenosine), thus obviating the need for multiple dosing.

[0008] Selective A2A receptor agonists are well known; for example, see Provisional Patent Applications Serial Nos. 60/1844296 and 60/21987. The compounds disclosed therein have a high specificity for the adenosine A2A receptor subtype but are not necessarily selective to heart. We have discovered a method for identifying A2A receptor agonists that produce the desired vasodilation in the heart but do not significantly affect the peripheral vasculature and have a short duration of action.

[0009] In addition to discovering a method to identify A2a agonists that are selective coronary vasodilators, we have discovered that compounds that meet our criteria that would be superior as adjuncts to MPI techniques.

[0010] Additionally, in the above-identified Provisional Patent Applications Serial Nos. 60/1844296 and 60/219876 the effective dose of the compounds of is disclosed to be in the range of 0.01-100 mg/kg/day. Surprisingly, we have discovered that the compounds are active at much lower doses (0.0002-0.009 mg/kg) than those disclosed as effective in Accordingly, a novel and effective method of using the compounds is provided, which is virtually free of undesirable side effects.

SUMMARY OF THE INVENTION

[0011] In a first aspect, the invention relates to a method for identifying compounds useful as adjuncts in MPI, comprising the steps;

[0012] a. measuring the intrinsic efficacy of test compounds in a cell line that stabily express adenosine A2A receptors.

[0013] b. measuring the intrinsic efficacy of a full agonist in said cell line.

[0014] c. selecting those compounds that have a lower intrinsic efficacy than said full agonist;

[0015] d. measuring the binding affinity (K,) of the selected compounds; and

[0016] e. selecting a compound with a Ki>1 uM.

[0017] Such compounds are selective partial A2A-adenosine receptor agonists, which produce coronary dilation in mammals without causing corresponding peripheral vasodilation at significant levels. They are low affinity compounds having a short duration of action.

[0018] In a, second aspect, the invention relates to a method of measuring coronary blood flow (CBF) in a mammal, comprising administering to a a mammal low doses of an A2A agonist referred to as CVT-3033, or CVT-3146.

BRIEF DESCRIPTION OF FIGURES

[0019] FIG. 1. Competititive radiolabeling binding assays of of adenosine receptor, agonists for A2A and A1 binding sites Membranes prepared from HEK-293 expressing human A2A adenosine receptors were incubated with [3H]ZM241385 (1.5-5 M) and from 10−9M-10−5 M of the various agonists (FIG. 1A). Membranes from CHO-K1 cells expressing human A1 adenosine receptors were incubated with [3H]CPX (2.5-3.0 M) and from 10−9M-10−5 M of the various agonists (FIG. 1B). The cells were incubated for two hours at room temperature in 50 mmol/L Tris-HCl buffer (pH 7.4) containing ADA (1U/mL) and 100 μM Gpp(NH)p. Non-specific binding of [3H]ZM241385 or [3H]CPX was determined in the presence of either 100 μmol/L NECA or 1 μmol/L CPX, respectively. Each point represents the mean±SEM of triplicates pooled from at least three experiments. Values of Ki and pKi are given in Table 2.

[0020] FIG. 2. Competititive radiolabeling binding assays of of adenosine receptor agonists for A2band A3 binding site. Membranes suspensions from HEK-293 cells expressing A2b adenosine receptors were incubated with [3H]DPCPX (1.5-5 M) and from 10−8M-10−5 M of the various agonists (FIG. 2A). Membranes from CHO-K1 cells expressing A3 adenosine receptors were incubated with [125I]ABMECA (2.5-3.0 M) and from 10−8M-10−5 M of the various agonists (FIG. 2B). The cells were incubated for two hours at room temperature in 50 mmol/L Tris-HCl buffer (pH 7.4) containing ADA (1U/mL). Each point represents the mean±SEM of triplicates pooled from three experiments.

[0021] FIG. 3. Effects of adenosine receptor agonists on cAMP content in intact PC12 cells. (FIG. 3A) PC12 cells were incubated for 10 minutes with various concentrations of adenosine receptor agonists in the presence of 50 μmol/L rolipram. Cyclic AMP levels were determined as described under “Methods”. Values represent mean±SEM of results of triplicate samples from three experiments. (FIG. 3B) Effect of CVT-3033 (1 μM) on CGS21680 stimulated increase cAMP accumulation in PC12 cells. PC12 cells were stimulated for 10 minutes with various concentrations of CGS21680 in the absence or presence of the partial agonist CVT-3033 (1 μM). Values represent mean±SEM of triplicate determinants from one representative experiment.

[0022] FIG. 4. Effect of CVT-3146 and CVT-3033 on cAMP content in HEK-293 cells. HEK-293 cells were incubated for 10 minutes with various concentrations of CVT-3146 or CVT-3033 in the presence of 50 μmol/L rolipram. Cyclic AMP levels were determined as described under “Methods”. Values represent mean±SEM of triplicate samples from three experiments.

[0023] FIG. 5. Time course of the decline in agonist stimulated cAMP following addition of an A2A adenosine receptor antagonist SCH58261. (FIG. 5A) PC12 cells were stimulated with various adenosine receptor agonists in the presence of 50 μmol/L rolipram at 37° C. After a 10 minute incubation, SCH58261 (20 μmol/L) was added and cAMP content was determined at the times indicated. (FIG. 5B) Linear regression analysis of the relationship between the T1/2 (min) and pKi for the agonists. Binding affinities were determined by measurement of the displacement of specific binding of [3H] ZM241385 from membranes prepared from PC12 cells using the data from Table 2

[0024] FIG. 6. Effects of adenosine receptor agonists on coronary conductance in isolated rat hearts. Concentration-dependent stimulation by CGS21680, CVT-2995, WRCO470, CVT-3146, CVT-3033, CVT-3032 and CVT-3100 of coronary conductance of isolated rat hearts. Symbols and error bars represent the mean±SEM of single determination from 4 to 6 hearts per agonist. Hearts were paced at a cycle length of 250 msec. Coronary conductance in the absence of drug was 0.17±0.01 ml/min/mmHg (mean±SEM, n=26).

[0025] FIG. 7A. Effect of CPX and ZM241385 on CVT-3146 (10 nM) on CPP by isolated rat hearts. FIG. 7B. Effect of concentration of ZM241385 on CVT-3146 stimulation of CC by isolated rat hearts.

[0026] FIG. 8. Functional selectivity of CVT-3146 and CVT-3033 for adenosine receptor subtypes. (FIG. 8A) Effect of concentration on AV conduction time and coronary conductance in isolated rat hearts. Symbols and error bars represent means±SEM of single determinations from each of three hearts. (FIG. 8B)

[0027] FIG. 9. Effect of adenosine receptor agonists on coronary perfusion pressure (CCP) in isolated rat hearts. Decreases in CPP caused by infusion of CVT-3146 (10 nM, 4 min) or CGS21680 (100 nM, 4 min) (FIG. 9A and FIG. 9B). Decrease in CPP caused by the infusion of CVT-3146 (10 nM) with or without additional infusion of CGS21680 (100 nM).

[0028] FIG. 10. Reversal of effect of agonist stimulation on CC. CVT-3033, CGS21680 and adenosine were given as boluses by iv infusion to isolated rat hearts and then. CC was measured at 8, 16 and 24 minutes after administration (FIG. 10A). Also, CVT-3146, WRC 0470 and adenosine were given as boluses by iv infusion to isolated rat hearts and then CC was measured at 8, 16 and 24 minutes after administration (FIG. 10B). Linear regression analysis of the relationship between the pKi (data from Table 2) and the reversal time (t0.9) of coronary vasodilation are given in FIG. 10C and D. Each data point represents the mean±SEM of pKi and T0.9 values. R and N are the correlation coefficient and number of agonists, respectively.

[0029] FIG. 11. Increases in CBF caused by CVT-3146 and adenosine in conscious dogs. Each data point is mean±SEM of the peak effect in CBF from 6 dogs. *: p<0.05,†: p<0.01 and ‡: p<0.001. Adenosine: (6) and CVT-3146 (O).

[0030] FIG. 12. Time course of changes in average peak coronary (O) and peripheral (O) artery blood flow velocity after an IV bolus injection ofCVT-3146 (1 ug/kg) (A) and adenosine (200-300 μg/kg).(B). Each point represents the changes in average peak flow velocity (ΔAPV) in comparison with the baseline values and represent the mean±SEM of single determinations.

DETAILED DESCRIPTION

[0031] Definitions and General Parameters

[0032] This invention provides methods for identifying partial adenosine A2A receptor agonists, which are particularly useful in MPI.

[0033] The interaction of an adenosine A2A agonist with the extracellular domain of its receptor causes a cascade of intracellular responses that culminate with vasodilation. However, there are quantitative differences in the ability of various agonists to achieve this effect. With respect to drug design, the ideal agent for imaging would have a duration of action that is brief enough not to cause serious side effects, but long enough to obviate the necessity for multiple treatments. It would be selective for the A2A receptor, thus avoiding side effects due to interaction with other adenosine subtypes, and it would produce coronary vasodilation without causing corresponding peripheral vasodilation.

[0034] Compounds that act as A2A agonists produce a variety of effects that depend on both the characteristics of the agonist, its receptor, and the tissue bearing A2A receptors. Factors relate to agonist properties are the intrinsic efficacy (E) and the equilibrium dissociation constant of the agonist-receptor complex (Kd).

[0035] Intrinsic efficacy (maximal efficacy) is the maximum effect that an agonist can produce if the dose is taken to its maximum. Efficacy is determined mainly by the nature of the receptor and its associated effector system. By definition, partial agonists have a lower maximal efficacy than full agonists.

[0036] The Kd of a drug is obtained from data generated from a saturation experiment analyzed according to a Scatchard plot (B/F versus F), which leads to a linear curve. The Kd is estimated as the negative reciprocal of the slope of the line of best fit, and Bmax by the abscissa intercept of the line. The reciprocal of Kd measures the affinity constant (Ka) of the radioligand for the receptor. Thus, for a given ligand-receptor pair, the smaller the Kd (0.1-10 nM) the higher its affinity. Bmax is expressed as pmol or fmol per mg tissue or protein.

[0037] When the saturation experiment is performed in the presence of a displacer (competitor), the line of best fit of the Scatchard plot can be modified in a manner that depends on the type of receptor interaction exhibited by the displacer. Two main cases exist: (1) decreased slope and unchanged Bmax, the displacement is of the competitive type; (2) unchanged slope and unchanged displacement of the non-competitive type. Intermediate cases where both the slope and Bmax are modified also exist.

[0038] Data generated from a displacement experiment are generally fitted by a sigmoidal curve termed the displacement or inhibition curve, that is the percentage radiolabeled ligand specifically bound versus log [displacer] in M). The abscissa of the inflexion point of the curve gives the IC50 value, the concentration of dsplacer that displaces or inhibitor 50% of the radioactive ligand specifically bound. IC50 is a measure of the inhibitor or affinity constant (Ki) of the displacer for the receptor. IC50 and Ki are linked as follows if the displacement is of the competitive type then

K i =IC 50 /(1+[C*]/Kd*

[0039] This is the Cheng-Prusoff equation (Biochem. Pharmacol, 22:3099 (1973)). [C*] is the concentration of radioligand and Kd* is its dissociation constant. The duration of the biological effect of an agonist is directly related to the binding affinity of a compound. It is desirable that compounds that act as adjuncts in imaging have an effect that is long enough to produce a response without repeated administration but short enough to avoid adverse side effects. Consequently, the, preferred compounds of the invention will have a relatively low binding affinity and a relatively short duration of action.

[0040] The potency is the dose or concentration required to bring about some fraction of a compound's maximal effect (i.e., the amount of compound needed to produce a given effect). In graded dose-response measurements, the effect usually chosen is 50% of the maximum effect and the dose causing the effect is called the EC50. Dose-response ratios using EC50 values for an agonist for a given receptor in the absence and presence of various concentrations of an antagonist for the same receptor are determined and used to construct a Schild plot from which the Kb and PA2 (−log10Kb) values are determined.

[0041] The concentration of antagonist that causes 50% inhibition is known as the IC50. IC50 is used to determine the Kb, the equilibriumdissociation constant for the antagonist-receptor complex. Thus,

K b =[IC 50 ]/1+[A]/KA

[0042] Wherein KA=equilibrium dissociation constant for an agonist binding to a receptor (concentration of agonist that causes occupancy of 50% of the receptors) and [A] is the concentration of agonist.

[0043] A compound may be potent but have less intrinsic activity than another compound. Relatively potent therapeutic compounds are preferable to weak ones in that lower concentrations produce the desired effect while circumventing the effect of concentration dependent side effects.

[0044] The tissue specific factors that determine the effect of an agonist are the number of viable specific receptors in a particular tissue [RT] and the efficiency of the mechanisms that convert a stimulus (S) into an effector response. Thus, there exists for a given tissue, a complex function f(S) that determines the magnitude of the response: 1$\mathrm{Response}=f\left(S\right)=\frac{f\left(\left[A\right]E\left[\mathrm{RT}\right]\right)}{\left(\left[A\right]+K_d\right)}$

[0045] Therefore, a response to a drug is a function of both the stimulus produced by agonist interaction with the receptor and the efficiency of the transduction of that stimulus by the tissue. Stimulus is proportional to the intrinsic efficacy of the agonist and the number of receptors. Consequently, variation in receptor density in different tissues can affect the stimulus for response. Furthermore, some tissues have very efficiently coupled receptors and other relatively inefficient coupled receptors. This has been termed ‘receptor reserve’ (or spare receptor) since in the first case, a maximum effect can be achieved when a relatively small fraction of the receptor is apparently occupied and, further receptor occupancy can produce no additional effect. The magnitude of the response will thus depend on the intrinsic efficacy value so that, by classical definition, full agonists (E=1) produce the maximum response for a given tissue, partial agonists produce a maximum response that is below that induced by the full agonist (0≦E≦1), and antagonists produce no visible response and block the effect of agonists (E=0). These activities can be completely dependent upon the tissue, i.e., upon the efficiency coupling. Therefore, low-efficacy adenosine agonists may be partial agonists in a given tissue and yet full agonists in peripheral arteries with respect to a function such as vasodilation.

[0046] The presence of spare receptors in a tissue increases sensitivity to an agonist. An important biologic consequence of spare receptors is that they allow agonists with low efficacy for receptors to produce full responses at low concentrations and therefore elicit a selective tissue response. Thus, a drug may be designed to elicit a maximal effect in a desired tissue but elicit a less than maximal effect in other tissues when such action of a drug would lead to undesirable side effects.

[0047] Thus, the invention provides a method of identifying drugs by first determining their efficacy compared to a known full agonist. Then, the binding affinity of the compound is determined. Compounds identified by this method will demonstrate partial agonist effects in the cAMP assays and a low IC as determined by affinity binding assays.

[0048] One preferred compound of the invention that is a selective partial A2A-adenosine receptor agonist with a short duration of action is a compound of the formula: 1

[0049] CVT-3033 is particularly useful as an adjuvant in cardiological imaging. The preparation of CVT-3033 and related compounds is described in U.S. patent application Ser. No. 09/338185, filed on Jun. 22, 1999, the specification of which is incorprated herein by reference.

[0050] Another preferred compound of the invention that is a selective partial A2A-adenosine receptor agonist with a short duration of action is a compound of the formula: 2

[0051] The preparation of CVT-3146 and related compounds is set forth in U.S. patent application Ser. No. 09/338327 filed on Jun. 22, 1999, the specification of which is incorporated herein by reference.

[0052] Compounds identified by the method of the invention are partial A2A agonists that increase CBF but do not significantly increase peripheral blood flow. That is, the stimulation of blood flow in the periphery is less than 50% of the increase of that in the heart.

[0053] Preferred compounds identified by the method of the invention have a duration of less than 5 seconds but longer than the effect produced by adenosine.

[0054] The compounds identified by the method of the invention are useful A2A agonists that may be used as adjuncts in cardiac imaging when added: either prior to dosing with an imaging agent or simultaneously with an imaging agent.

[0055] Suitable imaging agents are 201Thallium or 99mTechnetium-Sestamibi, 99mTcteboroxime, and 99mtc(III). :

[0056] The term “optional” or “optionally” means that the subsequently described event or circumstance may or may not occur, and that the description includes instances in which said event or circumstance occurs and instances in which it does not.

[0057] The compositions may be administered orally, intravenously, through the epidermis or by any other means known in the art for administering therapeutic agents.

[0058] The method of treatment of comprises the administration of an effective quantity of the chosen compound, preferably dispersed in a pharmaceutical carrier. Dosage units of the active ingredient are generally selected from the range of 0.3 to 103 μg/kg, but will be readily determined by one skilled in the art depending upon the route of administration, age and condition of the patient. These dosage units may be administered one to ten times daily for acute or chronic disorders. No unacceptable toxicological effects are expected when compounds of the invention are administered in accordance with the present invention.

[0059] Pharmaceutical compositions including the compounds of this invention, and/or derivatives thereof, may be formulated as solutions or lyophilized powders for parenteral administration. Powders may be reconstituted by addition of a suitable diluent or other pharmaceutically acceptable carrier prior to use. If used in liquid form the compositions of this invention are preferably incorporated into a buffered, isotonic, aqueous solution. Examples of suitable diluents are normal isotonic saline solution, standard 5% dextrose in water and buffered sodium or ammonium acetate solution. Such liquid formulations are suitable for parenteral a administration, but may also be used for oral administration. It may be desirable to add excipients such as polyvinylpyrrolidinone, gelatin, hydroxy cellulose, acacia, polyethylene glycol, mannitol, sodium chloride, sodium citrate or any other excipient known to one of skill in the art to pharmaceutical compositions including compounds of this invention. Alteratively, the pharmaceutical compounds may be encapsulated, tableted or prepared in an emulsion or syrup for oral administration. Pharmaceutically acceptable solid or liquid carriers may be added to enhance or stabilize the composition,%or to facilitate preparation of the composition. Liquid carriers include syrup, peanut oil, olive oil, glycerin, saline, alcohols and water. Solid carriers include starch, lactose, calcium sulfate, dihydrate, teffa alba, magnesium stearate or stearic acid, talc, pectin, acacia, agar or gelatin. The carrier may also include a sustained release material such as glycerol monostearate or glycerol distearate, alone or with a wax. The amount of solid carrier varies but, preferably, will be between about 20 mg to about 1 gram per dosage unit. The pharmaceutical dosages are made using conventional techniques such as milling, mixing, granulation, and compressing, when necessary, for tablet forms; or milling, mixing and filling for hard gelatin capsule forms. When a liquid carrier is used, the preparation will be in the form of a syrup, elixir, emulsion or an aqueous or non-aqueous suspension. Such a liquid formulation may be administered directly or filled into a soft gelatin capsule.

[0060] The Examples that follow serve to illustrate this invention. The Examples are intended to in no way limit the scope of this invention, but are provided to show how to make and use the compounds of this invention.

EXAMPLES

[0061] The following abbreviations are used in the Examples:

TABLE 1
List of Abbreviations and Activities
of Experimental Compounds
Chemical Compound	Abbreviation	Receptor/Activity
4-aminobenzyl-5′-N-	ABMECA	A3 agonist
methylcarbox-
amidoadenosine
adenosine deaminase	ADA
2-p-(2-carboxy-ethyl)	CGS21680:	High affinity A2A agonist
phenethyl-
amino-5′-N-ethyl-
carboxamido-
adenosine
8-cyclopentyl-	CPX	A1 antagonist
1,3-dipropylxanthine
5′-guanylyl-imididodiphosphate	Gpp(NH)p	Stabilizes GPCR
2-hexynyladenosine-	HENECA
5′-N-ethyl-
N-ethylcarboxamido-adenosine	NECA	Non-selective adenosine
		receptor agonist
Phenylisopropyl-	R-PIA	A1 receptor agonist
adenosine
Rolipram		phosphodiesterase
		inhibitor
	SCH58261:	A2A antagonist
2-cyclohexylmethyliden-	WRC-0470:	High affinity 2A agonist
ehydrazinoadenosine
4-(2-[7-amino-2-	ZM241385:	2A antagonist
(2furyl)[1,2,4]-
triazolo[2,33-a][1,
3,5]triazin-5-yl
amino]ethyl)phenol

[0062] Other abbreviations include cAMP (cyclic adenosine monophosphate), APV (average peak velocity), CBF (coronary blood flow), CHO-Ki (Chinese hamster ovary cell line), HEK-293 (human cell line),CPP (coronary perfusion pressure), CR (coronary resistence), HR (heart rate), im (intramuscular), iv (intravenous) LVSP (left ventricle systolic pressure), MAP (mean arterial pressure). PBF (peripheral blood flow).

[0063] Adenosine deaminase was purchased from Boehringer Mannheim Biochemicals Indianapolis, Ind.). [3H] ZM241385 was purchased from Tocris Cookson Ltd (Langford, Bristol, UK). [3H] CPX was from New England Nuclear (Boston, Mass.). HENECA CGS21680, adenosine, NECA, R-PIA, phenylephrine, DMSO, rolipram and HEK-hA2AAR membranes were obtained from Sigma-RBI (Natick, Mass.). Nitroglycerin was obtained from Parke-Davis, Morris Plains, N.J. Aminophylline was obtained from Abbott Laboratories, Chicago, Ill.

[0064] HENECA was a gift from Professor Gloria Cristalli of the University of Camerino, Italy. Drug stock solutions (10 mmol/L) were prepared in DMSO. Sprague Dawley rats were purchased from Simonsen Laboratories (Gilroy, Calif.). Ketamine was purchased from Fort Dodge AnimalHealth (Fort Dodge, Iowa) and xylazine from Bayer (Shawnee Mission, Kans.). Succinyl cAMP-tyrosyl methyl ester (ScAMP-TME) was purchased from Sigma and iodinated in the presence of chloramine T. CVT-2995, CVT-3003-((4S,2R,3R,5R)-2-(6-amino-2-(2-thienyl)pruin-9-yl)-5-(hydroxymethyl)oxolane-3,4-diol), CVT-3006-((4S,2R,3R,5R)-2-(6-amino-2-{3-[2-benzylphenoxy]prop-1-ynyl} purin-9-yl)-5-(hydroxymethyl)oxolane-3,4-diol), CVT-3032, CVT-3033, CVT-3100-((4S,2R,3R,5R)-2-[6-amino-2,-(5-methyl(2-thienyl))purin-9-yl]-5-(hydroxymethyl)oxolane-3,4-dion), CVT-3101-(4-(3-{9-[(4S,2R,3R,5R)-3,4-dihydroxy-5-(hydroxymethyl)oxolan-2-yl]-6-aminopurin-2-yl} prop-2-ynyloxy)benzenecarbonitrile), CVT-3126-(4S,2R,3R,5R)-2-{6-amino-2-[1-(3-phenylpropyl)pyrazol-4-yl]purin-9-yl}-5-(hydroxymethyl)oxolane-3,4-diol), CVT-3127 -(ethyl 1-{9-[(4S,2R,3R,5R)-3,4-dihydroxy-5-(hydromethyl)oxolan-2-yl]-6-aminopurin-2-yl}pyrazole-4-carboxylate), CVT-3141-((4S,2R,3R,5R)-2-{6-amino-2-[4-(4-chlorophenyl)pyrazole]purin-9-yl}-5-(hydroxymethyl)oxolane-3,4-diol), CVT-3144-((4S,2R,3R,5R)-2-{6-amino-2-[4-(4-methylphenyl)pyrazolyl]purin-9-yl}-5-(hydroxymethyl)oxolane-3,4-diol), CVT-3 146, YT-146 and WRC0470 were synthesized by CV Therapeutics, Department of Medicinal and Bioorganic Chemistry, Palo Alto, Calif. The structures of several of these compounds are set forth on the following page. 3

EXAMPLES 1 AND 2

[0065] Examples 1 and 2 demonstrate the selectivity and binding affinity of the compounds of the invention to adenosine receptors on cells obtained as follows. Rat pheochromocytoma PC12 cells were obtained from the American Type Culture Collection and grown in DMEM with 5% fetal bovine serum, 10% horse serum, 0.5 mmol/L L-glutamine, 100U/mL penicillin, 0.1 mg/mL streptomycin, and 2.5 μg/mL amphotericin.

[0066] HEK-293 cells stabily expressing recombinant human A2B adenosine receptors (HEK-hA2B adenosine receptors) were grown in DMEM supplemented with 10% fetal bovine serum and 0.5 mg/mL G-418.

[0067] CHO-K1 cells stabily expressing the recombinant human A1 adenosine receptors (CHO-hA1 adenosine receptors) or A3 adenosine receptors (CHO-hA3 adenosine receptors) were grown as monolayers on 150-mm plastic culture dishes in Ham's F-12 media supplemented with 10% fetal bovine serum in the presence of 0.5 mg/mL G-418. Cells were cultured in an atmosphere of 5% CO2/95% air and maintained at 37° C.

[0068] Cell membranes were harvested from the cell lines by detaching cells from the culture plates into ice-cold 50 mmol/L Tris-HCl buffer (pH 7.4). The cell suspensions were homogenized and centrifuged at 48,000 g for 15 minutes. The pellets were washed three times by re-suspension in ice-cold Tris-HCl buffer and centrifugation. The final pellet was re-suspended in Tris-HCl, aliquoted and frozen at −80° C. until used for receptor binding assays. The protein concentration of membrane suspensions was determined using the Bradford (Bradford, M. M. (1976. Anal. Biochem. 72, 248) with bovine serum albumin as standard.

[0069] Membranes were also obtained from porcine striatial cells as follows. Porcine striatum was obtained from Pel Freeze Inc. Striatum was minced and homogenized in 10 volumes of ice-cold 50 mmol/L Tris HCl buffer (pH7.4). The homogenate was filtered through cotton gauze and centrifuged at 48,000 g for 15 minutes at 4° C. The supernatant was discarded, and the membrane pellet was suspended in 10 volumes of 50 mmol/L Tris-HCl buffer (pH 7.4) and washed three times by centrifugation and resuspended in fresh buffer. The final pellet was frozen at −80° C. until used for receptor binding assays.

[0070] Competitive radiolabeling binding assays were performed to determine the binding affinities of the compounds of the invention for the adenosine receptor subtypes, A1, A2A, A2B and A3. Briefly, membrane suspensions, obtained from cells expressing either adenosine receptor subtypes A1, A2A, A2B or A3 were incubated for 2 hours at room temperature in 50 mmol/L Tris-HCl buffer (pH 7.4) containing ADA (1U/mL). [3H]-ZM241385 (˜1.5 to 5 mmol/L) was added to membranes from cells expressing A2A, [3H]-CPX (˜2.5 to 3.0 mmol/L) was added to membranes from cells expressing A1, [3H]-CPX (30 nM) was added to membranes from cells expressing A2B and [125I]ABMECA (1 nM) was added to membranes from cells expressing A3. The competing agents, that is the agonists (10−9-10−4M) were also added along Gpp (NH) p (100 μg) which stabilizes the receptor in the low affinity state thereby obviating the complication of multiple affinity states (Gao., Z. et al. (1999). Biochem J 338(3):729). At the end of the incubation, free radioligand was separated from membrane-bound radioligand by filtration through Whatman GF/C glass fiber filters using a Brandel tissue harvester (Gaithersburg, Md.). Triplicate determinations were performed for each concentration of unlabelled compound.

[0071] The results of radioligand binding assays, presented in Table 2, show that binding affinities of CVT-3033 (1417 nM) and CVT-3146 (1095 nM) to A2A receptors in pig striatum were lower (i.e. Ki higher) than a full A2A agonist, CGS21680 (157 nM).

[0072] The binding affinities of CVT-3033 (3623 nM) and CVT-3146 (1734 nM) to A2A were also lower than CGS21680 (210 nM) in PC12 cells. Binding affinities of CVT-3033 (2895 nM) and CVT-3146 (1269 nM) were also lower than CGS21680 (609 nM)in HEK cells expressing human A2A receptors. CGS-21680, CVT-3033, and CVT-3146 had relatively low affinities for the A1 receptor expressed by CHO-K1 cells. Binding to the A2A receptor was relatively greater than binding affinities to A1, A2b or A3 (FIGS. 1A and B and FIGS. 2A and B).

TABLE 2
Binding affinity of adenosine receptor agonists for A2A and A1 AdoRs
	K1, nmol/L (pK1 ± SEM)
	Pig Striatum(A2A)	PC12 cells(A2A)	HEK-hA2AAdoR	CHO-hA1AdoR
Compounds	Affinity	N	Affinity	N	Affinity	N	Affinity	N
NECA	ND	297 (6.54 ± 0.06)	4	360 (6.45 ± 0.06)	3	328 (6.49 ± 0.06)	3
R-PIA	1106 (5.96 ± 0.06)	4	3476 (5.48 ± 0.10)	3	1656 (5.78 ± 0.02)	3	477 (6.45 ± 0.11)	3
CGS21680	157 (6.85 ± 0.23)	5	210 (6.74 ± 0.09)	3	609 (6.22 ± 0.06)	3	3540 (5.47 ± 0.20)	3
CVT-2995	14.4 (7.86 ± 0.12)	4	53 (7.28 ± 0.05)	4	305 (6.52 ± 0.04)	6	866 (6.07 ± 0.05)	3
WRC0470	20.7 (7.68 ± 0.04)	3	18 (7.77 ± 0.04)	8	272 (6.55 ± 0.04)	6	7278 (5.16 ± 0.09)	3
CVT-3032	1594 (5.88 ± 0.10)	3	5516 (5.26 ± 0.05)	3	13651 (4.87 ± 0.02)	3	6350 (5.22 ± 0.11)	3
CVT-3033	1417 (5.67 ± 0.16)	3	3623 (5.45 ± 0.03)	5	2895 (5.54 ± 0.03)	3	5836 (5.24 ± 0.04)	3
CVT-3146	1095 (5.95 ± 0.11)	5	1734 (5.76 ± 0.01)	3	1269 (5.90 ± 0.03)	7	>16460 (4.59 ± 0.35)	3
YT-146	16.3 (7.86 ± 0.33)	4	ND	ND	ND
CVT-3003	1007 (6.00 ± 0.07)	3	ND	ND	ND
CVT-3006	64 (7.24 ± 0.26)	3	ND	ND	ND
CVT-3100	697 (6.16 ± 0.06)	5	ND	ND	ND
CVT-3101	94 (7.03 ± 0.06)	3	ND	ND	ND
CVT-3126	1667 (5.78 ± 0.09)	3	ND	ND	ND
CVT-3127	64 (7.22 ± 0.17)	3	ND	ND	ND
CVT-3141	1138 (5.95 ± 0.07)	3	ND	ND	ND
CVT-3144	502 (6.31 ± 0.12)	3	ND	ND	ND
HENECA	7.96 (8.10 ± 0.08)	10	ND	ND	ND
The binding affinities of adenosine receptor agonists for A2AAdoRS and A1AdoRs were determined by their effect to compete with specific binding of [3H]ZM241385 or [3H]CPX, respectively, to membranes from the indicated tissue/cells. Values are means ± SEM
# of results of at least three experiments (n) performed in triplicate. Numbers in parentheses are means of pK1 ± SEM. Values of the equilibrium dissociation constant (Kd) for [3H]ZM241385 binding that were used in calclations of K1 values were 0.5, 0.5 and 0.8 nM for pig
# striatum, PC12 and HEK-hA2AAdoR cells, respectively. A Kd value of 1 nM for [3H]CPX binding to CHO-hA1AdoR membrane was used in calculation of K1 values is 1.

EXAMPLE 3

[0073] Example 3 demonstrates the ability of the compounds of the invention to stimulate cAMP levels, a measure of the intrinsic efficacy of the agonists. Briefly, PC12 cells were rinsed three times with Hanks' balanced saline solution (HBSS), detached using a cell lifter, and pelleted by centrifugation at 500 g for 5 minutes. Aliquots of the cell suspension (0.1 to 0.2 mg protein) were placed in microfuge tubes with 250 μL of HBSS containing rolipram (50 μmol/L) to inhibit phosphodiesterases that degrade cAMP and warmed to 37° C. Appropriate drugs were added to the cell suspensions, and incubations were allowed to continue for 10 minutes. Tubes were placed in a boiling water bath for 5 minutes to terminate the incubation. The samples were then cooled to room temperature, diluted by the addition of 1 mL of 10 mmol/L Tris-HCl buffer at pH 7.4, and then centrifuged for 2 minutes at 13000 g.

[0074] The cAMP content of the supernatant was determined by modification of a radioimmunoassay method described by Harper and Brooker (1975. J. Cyclic nucleotide Res 1:207). Briefly, an aliquot of the supernatant (0.01 ML) was mixed with 0.04 mL of HBSS, 0.05 mL of 50 mmol/L sodium acetate buffer (pH 6.2) containing 10 mmol/L CaCl2, [125]ScAMP-TME (12500 dpm), and 0.05 mL of anti-cAMP antibody (1:2000 dilution with 0. 1% bovine serum albumin in distilled water). The samples were then incubated at 4° C. for 16 hours. At the end of the incubation, 70 μL of a 50% (wt/vol) hydroxyapatite suspension was added to each tube. The suspensions were gently agitated and then incubated for 10 minutes at 4° C. Antibody-bound radioactivity adsorbed to hydroxyapatite was collected onto glass fiber filters by vacuum filtration using a Brandel cell harvester. Radioactivity retained by the filter was counted in a gamma counter. Nonspecific binding of [125I]ScAMP-TME was defined as radioactivity bound in the presence of 3 μmol/L unlabeled cAMP and was subtracted from total binding. The: amount of cAMP present in samples was calculated based on a standard curve using known amounts of cAMP.

[0075] As illustrated in FIG. 3A, all compounds increased the cellular content of cAMP in a concentration-dependent manner. The low affinity A2A agonists CVT-3032, CVT-3033 and CVT-3146, were not only less potent (10-15 fold), but also less effective in stimulating cAMP accumulation compared to CGS21680. The maximal responses induced by CVT-3146, CVT-3033 and CVT-3032 were 85%, 63% and 65% of that induced by CGS21680, respectively. These data demonstrate that CVT-3146, CVT-3033 and CVT-3032 behave as partial A2A agonists in PC12 cells. Additionally, CVT-3033 (1 μM) inhibit the ability of CGS21680 to stimulate cAMP levels (FIG. 3B) causing an approximate 5-fold shift to the right of the CGS21680 concentration-response curve

[0076] It is notable that the stimulation of cAMP levels in PC12 cells was related to the binding affinity of the compounds to A2A receptors (Table 3).

TABLE 3
Rate of decline (t0.5) of cAMP accumulated after
exposure to the A2A agonists in PC12 cells
	Agonists	t0.5 (min)	n
	WRC0470	5.9 ± 0.8	6
	CGS21680	5.3 ± 0.5	5
	CVT-2995	3.9 ± 0.6	5
	CVT-3146	2.6 ± 0.2	6
	CVT-3033	1.9 ± 0.1	4
	R-PIA	1.6 ± 0.2	6
	Values are mean ± SEM of the least three experiments performed in triplicate. The pK1 (−logK1) values were obtained from the results of competition binding assays under [3H]ZM241385 as the radioligand for A2A AdoRs. The pEC50 (−logEC50) values were
	# determined from concentration response relationships for agonist-induced cAMP accumulation in PC12 cells. The t0.5 values were calculated from the rate of decline of cAMP accumulated during exposure to each agonist (1 μM) following the addition of the A2A AdoR antagonists
	# SCH58261 (20 μM)>

[0077] The selectivity of the effects of CVT-3033 and CVT-3146 on cAMP accumulation in HEK-293 cells expressing A2B adenosine receptors is also shown. NECA, a non-selective adenosine receptors agonist, caused a concentration-dependent increase of cellular cAMP content whereas neither CVT-3033 nor CVT-3146 had any detectable effects even at a high concentration of 100 μM (FIG. 4). These results indicate that CVT-3033 and CVT-3146 have very weak, if any, interaction with A2B receptors.

[0078] The effect of an A2a antagonist on agonist-mediated cAMP accumulation in PC12 was also demonstrated. PC12 cells cultured in DMEM at 37° C. were treated with WRC0470, CGS21680, CVT-2995, CVT-3146, CVT-3033 and R-PIA each at a concentration of 1 μM in the presence of rolipram (50 μM) for 10 minutes. Then, an A2A antagonist, SCH58261 (20 μM), was added and cAMP content was determined at various periods. The time for cAMP levels to decrease to half maximal (t0.5) was calculated and plotted against the affinity (pKi) of each agonist for the A2A adenosine receptor, as determined by competition radioligand binding assays (above).

[0079] FIG. 5A shows the time-course of the decline of agonist-stimulated cAMP accumulation following the addition of SCH58261when compared to the control cultures (CGS2160 incubated without SCH58261). The calculated values of t0.5 from these experiments are presented in Table 3. The apparent t0.5 values of agonists were inversely related to their affinities for A2A adenosine receptors, that is, the greater the agonist affinity, the lower the rate of decline of cAMP content upon application of the A2A adenosine receptors antagonist SCH58261 (FIG. 5A). As depicted in FIG. 5B, the relationship between the apparent t0.5 and pKi for the agonists was best fit by linear regression with a correlation coefficient (r value) of 0.84.

EXAMPLE 4

[0080] The effect of the compounds of the invention on coronary conductance (CC), an estimate of vasodilation, was demonstrated ex vivo using perfused rat hearts. Briefly, rats of either sex weighing 230-260 grams were anesthetized by intraperitoneal injection of a mixture of ketamine (100 mg/ml) and xylazine (20 mg/ml). The chest of each rat was opened and the heart removed, and rinsed in ice-cold modified Krebs-Henseleit (K-H) solution containing NaCl 117.9, KCl 4.5, CaCl2 2.5, MgSO4 1.18, KH2PO4 1.18, pyruvate 2.0 mmo/L. The aorta was cannulated and the heart was perfused at a flow rate of 10 ml/min with modified K-H solution. The K-H solution (pH 7.4) was gassed continuously with 95% O2 and 5% CO2 and warmed to 35±0.50 C. The heart was electrically paced at a fixed cycle length of 240 ms (250 beats/min) using a bipolar electrode placed in the left atrium. The electrical stimuli were generated by a Grass stimulator (Model S48, W. Warwick, R.I.) and delivered through a Stimuli Isolation Unit (Model SIU5, Astro-Med, Inc., NY) as square-wave pulses of 3-msec in duration and with an amplitude of at least twice the threshold intensity.

[0081] As shown in FIG. 6A, adenosine, CGS21680, WRC0470, and the CVT compounds caused concentration-dependent increases in coronary conductance.

[0082] The potencies (EC50 values) of adenosine, CGS21680, WRC0470 and the CVT compounds are summarized in Table 4. The low affinity agonist CVT-3146 was found to be approximately 1 0-fold more potent than adenosine but 1 0-fold less potent than the high affinity agonists CGS21680 and WRC0470 with respect to increasing coronary conductance. The results show that CVT 3146 was a potent agonist of coronary conductance in heart but only a weak agonist in PC 12 cells: (EC50=291 nM).

TABLE 4
Potency of adenosine and A2A Adenosine
receptor agonists to increase cAMP accumulation
in PC12 cells and coronary conductance in rat
isolated perfused heart
EC50 (pEC50 ± SEM), nM
	cAMP Accumulation	Coronary Conductance
Agonist	(PC12 cells)	(Rat Isolated Heart)
CGS21680	18 (7.75 ± 0.03)	N = 3	0.54 (9.27 ± 0.03)	N = 3
CVT-2995	6.6 (8.82 ± 0.25)	N = 3	0.68 (9.17 ± 0.03)	N = 5
CVT-3146	291 (6.54 ± 0.03)	N = 3	6.40 (8.19 ± 0.04)	N = 4
CVT-3032	613 (6.21 ± 0.02)	N = 3	66.50 (7.18 ± 0.07)	N = 4
CVT-3033	487 (6.31 ± 0.01)	N = 3	67.95 (7.19 ± 0.08)	N = 4
WRC0470	ND		0.62 (9.19 ± 0.6)	N = 5
Adenosine	ND		59.20 (7.24 ± 0.11)	N = 4
Values are the mean concentrations of agonists that caused 50% increase in cAMP accumulation or coronary conductance (EC50 and pEC50).
ND; Not determined.

[0083] Coronary vasodilatory effect of CVT-3146 in the absence and presence of adenosine receptor antagonists were also demonstrated. The identity of the adenosine receptor subtype. (A1 or A2A) mediating the coronary vasodilation was determined. Hearts (n=6) were exposed to CVT-3 146 (10 nM), and after the effect of this agonist reached steady-state, CPX (60 nM), an A1 antagonist and then ZM241385 (60 nM), an A2a antagonist were added to, the perfusate and the changes in CPP were recorded. As depicted in FIG. 7A, CVT-3146 significantly increased coronary conductance to 0.22±0.01 ml mm Hg−1 min−1 from a baseline value of 0.16±0.02 ml mm Hg−1 min−1. This increase in coronary conductance caused by CVT-3146 was not affected by 60 nM CPX but was completely reversed by 60 nM ZM241385. Furthermore, the inhibition by ZM241385 of an increase of coronary conductance caused by CVT-3146 was concentration-dependent (FIG. 7B).

[0084] A1 adenosine receptor-mediated depression of A-V nodal conduction time by CVT-3033 and CVT-3146 (negative dromotropic effect) was measured using atrial and ventricular surface electrograms as described by Jenkins and Belardinelli (Circ Res 63:97).

[0085] As shown in FIG. 8, CVT-3146 and CVT-3033 increased coronary conductance in a concentration-dependent manner, but did not prolong A-V nodal conduction time

[0086] Coronary perfusion pressure (CPP) was measured using a pressure transducer that was connected to the aortic cannula via a T-connector positioned approximately 3 cm above the heart. CPP was monitored throughout an experiment and recorded either on a chart recorder (Gould Recorder 2200S, Valley View, Ohio) or a computerized recording system (PowerLab/4S, ADInstruments Pty Ltd, Australia). Only hearts with CPP ranging from 60 to 85 mmHg (in the absence of drugs) were used in the study. CC conductance (in ml/min/mmHg) was calculated as the ratio between coronary perfusion rate (10 ml/min) and CPP.

[0087] As shown in FIG. 9A the extent of the decrease in coronary perfusion pressure (an index of the coronary vasodilation) caused by CVT-3146 was similar to that caused by a supramaximal concentration of CGS21680 (FIG. 9B). Both 10 nM CVT-3146 and 100 nM CGS21680 decreased coronary perfusion pressure by 23 mmHg. In addition, in the presence of 10 nM CVT-3146, CGS21680 (100 nM) did not cause a further decrease of the coronary perfusion pressure (FIG. 9C). Thus, CVT-3146 is a full agonist with respect to coronary artery conductance.

[0088] The relationship between the affinity of agonists for the A2A receptor and the rate of reversal of agonist-mediated responses (coronary vasodilation in heart and cAMP accumulation in PC12 cells) upon termination of drug administration were determined. All compounds were given as boluses into the perfusion line at their respective minimal concentrations that caused equally or near-equally maximal increases in coronary conductance. Likewise, the onset and time to peak effect: (i.e. maximal coronary vasodilation) were similar for all agonists. Although adenosine and the various agonists caused equal maximal increases in coronary conductance, the durations of their effects were markedly different. The duration of the effect of adenosine was the shortest followed by those of CVT-3033 and CVT-3146. The effects of CGS21680 and WRC0470 had the longest duration (FIG. 10). The duration of the coronary vasodilation in isolated rat hearts caused by adenosine, the CVT compounds, and other agonists measured as the time to 50% and 90% (t0.5 and t0.9, respectively) reversal of the increases in coronary conductance after termination of drug administration are summarized in Table 4. The reversal time of coronary vasodilation correlated with the affinity of the adenosine derivatives for the A2A receptors. As shown in FIG. 9C, there was a significant (p<0.05) inverse relationship (r=0.87) between the affinity (pKi) of the agonists for the A2A adenosine receptors (Table 2) and the reversal time (t 0.9) (Table 4) of the coronary vasodilation caused by the same agonists in rat isolated hearts.

EXAMPLE 5

[0089] The magnitude of the effect of A2A adenosine receptor agonists on coronary dilation and the duration of the effect was determined in pigs weighing 22-27 kg. All animals received humane care according to the guidelines set forth in “The Principles of Laboratory Animal Care” formulated by the National Society for Medical research and the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH Publication No. 86-23, revised 1996). In addition, animals were used in accordance with the guidelines of the University of Kentucky Institutional Animal Care and Use Protocol.

[0090] Animals were anesthetized with ketamine (20 mg/kg, i.m.) and sodium pentobarbital (15-18 mg/kg, i.v.). Anesthesia was maintained with additional sodium pentobarbital (1.5-2 mg/kg, i.v.) every 15-20 minutes. Animals were ventilated via a tracheotomy tube using a mixture of room air and 100 % O2 Tidal volume, respiratory rate and fraction of O2in inspired air were adjusted to maintain arterial blood gas (ABG) and pH values. Core body temperature was monitored with an esophageal temperature probe and maintained at 37.0-37.5° C. by use of a heating pad. Lactate Ringers solution was administered via an ear or femoral vein as an initial bolus of 300-400 ml followed by a continuous infusion at a rate of 5-7 ml/kg/hr. A catheter was inserted into the femoral artery to monitor arterial blood pressure.

[0091] The heart was exposed through a median sternotomy and suspended in a pericardial cradle. Left ventricular pressure (LVP) was measured with a 5F high fidelity pressure sensitive tip transducer (Millar Instruments, Houston, Tex.) placed in the left ventricular cavity via an apical incision and secured with a purse string suture. A segment of the left anterior descending coronary artery (LAD), proximal to the origin of the first diagonal branch, was dissected free of surrounding tissue. A transit time perivascular flow probe (Transonic Systems Inc., Ithaca, N.Y.) was placed around this segment to measure CBF. Proximal to the flow probe, a 24-gauge modified angiocatheter was inserted for intracoronary infusions. All hemodynamic data were continuously displayed on a computer monitor and fed through a 32-bit analog to digital converter into an online data acquisition computer with customized software (Augury, Coyote Bay Instruments, Manchester, N.H.). A2A adenosine receptors agonists were dissolved in DMSO to produce stock concentrations of 1-5 mM, which were diluted in 0.9% saline and infused at rates of 1-1.5 ml/min via the catheter. The A2A, adenosine receptors agonists were administered intracoronary.

[0092] Relationship between affinity of various agonists for A2A adenosine receptor and the reversal time of their effect to increase coronary conductance Was determined in pigs. Each experiment was preceded by a 30-minute stabilization period following the completion of all instrumentation of the animal. Baseline hemodynamic data were then recorded and an intracoronary infusion of an A2A Adenosine receptors agonist was initiated. Each infusion was maintained for 4-5 minutes to allow LAD° CBF to reach a steady-state, after which the infusion was terminated. The times to recovery of CBF by 50% (t0.5) and 90% (t0.9) of the difference from peak effect to baseline CBF were recorded. Ten to 15 minutes after CBF returned to pre-drug values a second infusion with a different agonist was started. In preliminary studies it was found that the intracoronary infusion of adenosine receptor agonists produced varying degrees of systemic hypotension, and hence, in all subsequent experiments, phenylephrine was administered intravenously at dose of ˜1 μg/kg/min. Hemodynamic measurements were made prior to and following the initiation of the phenylephrine infusion. The phenylephrine infusion rate was adjusted during and following the infusions of the adenosine receptor agonists to maintain arterial blood pressure within 5 mmHg of pre-infusion values. The effect of a maximum of three different agonists was determined in each experiment.

[0093] All CVT-compounds as well as CGS21680 and other A2a adenosine receptors agonists (i.e., WRC-0470 and YT-146) caused significant increases in CBF (Table 6). Selected doses of these compounds given as continuous (4 to 5 min) intracoronary infusions caused 3.1 to 3.8-fold increases in CBF. Once it was established that all agonists caused comparable increases of CBF (i.e., “fold increase”) and caused little or no changes in heart rate and mean arterial blood pressure (data not shown), the reversal time of their coronary vasodilatory effects was determined. As summarized in Table 5 the reversal times of the effect of the low affinity, partial agonists, CVT-3146, CVT-3032 and CVT-3033, were shorter than those of CGS21680, WRC-0470 or YT-146. More importantly, as depicted in FIG. 9D, there was a significant (p<0.05) inverse relationship (r=0.93) between the affinity (pKi) of the A2A Adenosine receptors agonists for pig brain striatum A2A receptors and the reversal time (t0.9) of coronary vasodilation in pig heart Table 2).

TABLE 5
Reversal time of coronary vasodilation by adenosine
and adenosine receptor agonists in rate isolated perfused
	Agonist	t0.5 (min)	t0.9 (min)	n
	R-PIA	7.9 ± 0.1	12.6 ± 0.8	3
	CGS21680	14.5 ± 0.5	19.5 ± 09	3
	CVT-2995	17.5 ± 1.2	23.2 ± 2.1	4
	WRC0470	21.9 ± 0.9	27.9 ± 1.4	6
	CVT-3032	4.1 ± 0.3	9.8 ± 1.4	4
	CVT-3033	3.4 ± 0.5	8.4 ± 2.2	4
	CVT-3146	5.2 ± 0.2	11.3 ± 1.1	5
	YT-146	17.7 ± 1.0	25.8 ± 4.0	3
	CVT-3003	3.4 ± 0.1	9.2 ± 2.2	4
	CVT-3006	16.1 ± 0.1	21.8 ± 2.0	3
	CVT-3100	5.1 ± 0.6	10.1 ± 0.2	4
	CVT-3101	16.7 ± 0.5	25.6 ± 0.3	3
	CVT-3126	8.3 ± 0.4	12.6 ± 0.3	4
	CVT-3127	14.8 ± 2.1	15.0 ± 0.8	3
	CVT-3141	14.4 ± 1.9	21.3 ± 3.9	4
	CVT-3144	13.6 ± 1.3	18.9 ± 1.9	4
	HENECA	28.6 ± 1.1	32.8 ± 3.1	3
	Adenosine	1.6 ± 0.1	5.6 ± 0.8	11
	Times (in minutes) to 50% and 90% (t0.5 and t0.9, respectively) reversal of the increases in coronary conductance caused by adenosine and adenosine receptor agonists. Values are the mean ± SEM of single determination in each of n preparation.

[0094]

TABLE 6
Magnitude and reversal time of coronary dilation caused by various
adenosine receptor agonists in open-chest anesthetized pigs
		CBF (fold-
Agonist	Dose	increase)	t0.5 (min)	t0.9 (min)	n
CVT-3146	10 ug/kg/min	3.40 ± 0.04	1.9 ± 0.2	10.1 ± 0.7	3
CVT-3146	30 ug/kg/min	3.38 ± 0.39	2.6 ± 0.5	12.3 ± 1.1	6
CVT-3032	30 ug/kg/min	3.78 ± 0.70	2.3 ± 0.6	9.6 ± 1.0	3
CVT-3033	50 ug/kg/min	3.33 ± 0.58	3.1 ± 0.9	12.0 ± 1.0	3
WRC0470	1 ug/kg/min	3.14 ± 0.24	9.5 ± 0.8	22.5 ± 1.6	6
CGS21680	2 ug/kg/min	3.54 ± 0.09	9.7 ± 0.8	21.4 ± 0.8	3
YT-146	1 ug/kg/min	3.44 ± 0.47	17.8 ± 3.4	32.9 ± 5.6	3
Maximal “fold-increase” and time (in minutes) to 50% and 90% (t0.5 and t0.9, respectively) reversal of the increases in coronary blood flow caused by various adenosine receptor agonists. Data represent mean ± SEM of one or two measurements in each of n pigs.

EXAMPLE 6

[0095] The following example demonstrate the effects of CVT-3146 on the hemodynamic parameters in dogs.

[0096] Ten mongrel dogs (weighing 23-27 kg) were premedicated with Acepromazine (0.3 mg/kg im) and anesthetized with sodium pentobarbital (25 mg/kg) and then intubated and ventilated with room air. A thoracotomy was performed in the left fifth intercostal space using sterile surgical techniques. A Tygon catheter (Cardiovascular Instruments, Inc.) was placed in the descending thoracic aorta for the measurements of blood pressure. A solid-state pressure gauge (P6.5, Konisberg Instrument, Inc) was placed in the apex of the left ventricle for the measurement of the left ventricular systolic pressure (LVSP) and calculation of first derivative of left ventricular pressure (LV dP/dt). A Doppler transducer (Craig Hartley) was placed on the left circumflex coronary artery for measurement of CBF. An hydraulic coronary occluder (In: Vivo Metric, Inc) was implanted in 4 dogs around the left circumflex coronary artery. The chest was closed in layers and the catheters and wires were run subcutaneously and exited in the interscapular area. The dogs were allowed 10 to 14 days to recover fully from the surgery and were trained to lie on a laboratory table. The protocols were approved by the Institutional Animal Care and Use Committee of New York Medical College and conform to the “Guiding Principles for the use and Care of Laboratory Animals” of the National Institute of Health and the American Physiology Society.

[0097] Arterial pressure was measured by connecting the previously implanted catheter to a strain-gauge transducer (P23ID, Statham) and mean arterial pressure (MAP) was derived using 2-Hz low-pass filter. LV pressure was measured from the solid-state pressure gauge, and LV dP/dt was calculated using a microprocessor set as a differentiator and having a frequency response flat to 700 Hz (LM 324, National Semiconductor). Left circumflex CBF was measured using a pulsed Doppler flowmeter (System 6, Triton technology), and mean CBF was derived using a 2-Hz low-pass filter. Mean CR was calculated as the quotient of MAP and CBF. Heart rate was monitored from the pressure pulse interval using a cardiotachometer (Beckman Instruments). The lead-2 of the electrocardiogram was recorded during the experiments in order to examine the alterations in the AV nodal conduction (PR interval). All signals were recorded on a direct-writing oscillograph (Gould 2800).

[0098] To determine the effects of adenosine and CVT 3146 on CBF and CR in resting dogs baseline hemodynamics and CBF were recorded and then, increasing doses of adenosine: 13, 27, 67, 134 and 267 μg/kg and CVT-3146: 0.1, 0.175, 0.25, 0.5, 1.0, 2.5, 5 μg/kg in 10 ml volumes were administered iv for 10 minutes in 10 ml via a catheter inserted into a peripheral vein. Hemodynamics were measured before, during and after each dose. Following each dose hemodynamics were allowed to return to baseline before the administration of the next dose. Changes in heart rate, blood pressure, CBF; and ECG were recorded

[0099] The duration of coronary vasodilation was determined using two different injection protocols: 1) an iv infusion of 10 ml in 10 seconds and; 2) iv infusion of 10 ml in 30 seconds. The time to the peak effect in mean CBF increase and the duration during which CBF remained at least 2 fold above baseline.

[0100] To determine whether tachyphylaxis occurred, three consecutive injections of 1 μg/kg CVT-3146 were given as intravenous injections (10 ml in 30 seconds) via a catheter inserted into a peripheral vein. The hemodynamics were allowed to return to baseline between doses. Hemodynamics were measured before, during and after each dose.

[0101] Hemodynamic results are expressed as mean±SEM. Data were analyzed using one way repeated measures analysis of variance, with Student-Neuman-Keuls post hoc analysis to identify which means were significantly (p<0.05) different (Sigma Stat, Version 2.2, Jandel Scientific, San Rafael, Calif.). To determine the agonist potency from dose-response curves, doses producing 50% of maximum effect (ED50) were calculated by fitting curves using the Boltzmann equation. ED50 were compared using a Student's t-test (p<0.05 being considered as significant). Because no statistical differences were found among the baseline values, between each dose for each parameter measure, the first baseline registered during each experiment was used as the control value. A Student's t-test was used to compare the changes in CBF and hemodynamic parameters produced by 2.5 μg/kg CVT-3146, 267 μg/kg adenosine and 25 kg/kg nitroglycerin (p<0.05 being considered significant).

[0102] A dose-response curve of the effect of CVT-3146 on CBF is shown in FIG. 11. An IV bolus injection of CVT-3146 caused a dose-dependent increase in mean CBF, with a ED50 of 0.34±0.08 μg/kg and a maximal increase of 154±16 ml/min from baseline (45±3 ml/min). In comparison to CVT-3146, adenosine was less potent having an ED50=51±15 kg/kg (p<0.05). The maximal increase in mean CBF stimulated by CVT-3146 and adenosine were similar.

[0103] CVT-3146 produced a maximal decrease in CR of 73±2% and 75±2% at 2.5 μ/kg at 5 μg/kg, respectively. Adenosine produced a maximal decrease of 73±1% at 267 μg/kg (data not shown).

[0104] The effects of CVT-3146 and adenosine on left ventricular systolic pressure and dP/dt were compared. Increasing doses of CVT-3 146 did not cause significant changes in LVSP (data not shown). In comparison, adenosine increased LVSP at 67 μg/kg, 134 μg/kg and 267 μg/kg, by 12±3%, 12±3% and 18±6%, respectively. Both adenosine (267 μg/kg) and CVT-3146 (2.5 μg/kg) increased the dP/dt by 29±7% and 39±7% respectively.

EXAMPLE7

[0105] Example 7 demonstrates the differential effects of CVT-3146 on blood flow velocity in coronary and peripheral arteries, systemic arterial blood pressure and heart rate in anesthetized dogs.

[0106] Mongrel dogs (either sex, 17-21 kg; n=6) were obtained from a local vendor (Barton, Oxford, N.J.). Blood flow velocity in the coronary and cranial circumflex arteries was measured using Doppler transducer-tipped guide wires 0.014″ in diameter (FloWire®, model 1400J) purchased from Cardiometrics, Inc., Mountain View, Calif. For the positioning of the FloWire, a Judkins left coronary guiding catheter (JL3.5, 8F; Cordis) was used. Vascular arterial angiography was performed using Hypaque-76 contrast fluid (Bracco Diagnostics, Inc., Princeton, N.J.; Lot #9H28899) and a mobile fluoroscopic unit (Philips, BV 29). Systemic arterial blood pressure was measured using an electronic transducer-tipped catheter (Millar). The following pharmacologic agents were used: CVT-3 146 (CV Therapeutics; Lot #315-53), heparin (Solopak Laboratories, Inc., Elk Grove Village, Ill.; Lot # 960211) and sodium pentobarbital (lot # 9700), and acepromazine (lot # 3960960), obtained from J. A. Webster (Fort Dodge, IW).

[0107] Dogs were sedated with acepromazine (0.25 mg/kg), anesthetized with sodium pentobarbital (30 mg/kg+additional doses (1 mg/kg) given as necessary to maintain the level of anesthesia), intubated with endotracheal tube and artificially ventilated with room air using a respirator. Following the administration of heparin (20 U/kg+100 U/hr), the pressure transducer-tipped catheter was introduced through the left femoral artery and positioned in the descending aorta. The Doppler FloWire was introduced through the right femoral artery. A peripheral vein was cannulated for the administration of all drugs. Doses of 1 μg/kg of CVT-3146 and 300 or 200 μg/kg adenosine (Adenosine) were given multiple times, once or twice when the Doppler catheter was positioned in a coronary artery and again when the catheter was positioned in the cranial circumflex humeral artery. The sequence of positioning of the Doppler catheter was reversed in consecutive experiments. In each dog, baseline values of measured parameters were recorded following a stabilization period of 20 minutes, and CVT-3146 and Adenosine were given as intravenous bolus injections (<0.5 ml) followed by a physiologic saline solution flush (10 ml); the time required for both injections of each drug was <15 sec. All parameters were allowed to recover (>30 min) to their respective baseline values between two consecutive drug administrations. In all six dogs studied, the effect of CVT-3146 on coronary artery APV was determined. The effect of CVT-3146 on peripheral artery APV was determined in five of the six dogs. The effect of adenosine on both coronary and peripheral artery APV was studied in five of the six dogs.

[0108] Systemic arterial blood pressure (BP) and electrocardiograms (ECG) were monitored and recorded using Gould Data Acquisition System (model 13-4615-65A), a video cassette recorder (Teac, XR 5000) and a chart recorder (Astromed, 9600). The following parameters were monitored and recorded: Average peak coronary and peripheral artery blood flow velocity (APV), mean arterial blood pressure (MAP) (mmHg), and sinus cycle length (SCL; msec). Differences in measure parameters were tested for statistical significance using ANOVA and Student's t test corrected for multiple measurements. Data are expressed as the mean±SEM.

[0109] CVT-3146 increased APV in the coronary vasculature by 2.6±0.2-fold while its increase of APV in the peripheral arteries was only 1.1±0.1 -fold (Table 7). In contrast the vasodilatory action of adenosine was similar in the two vascular beds: specifically, adenosine increased APV in the coronary vasculature by 2.5±0.3-fold while its increase of APV in the periopheral arteries was 2.0±0.4-fold.

TABLE 7
Magnitude of Increases in Coronary and
Peripheral Blood Flow Velocity of CVT-3146
and Adenosine in Anesthetized Close-chest Dogs
	Agonist	Coronary	N	Peripheral	N
	CVT-3146	2.6 ± 0.2	6	1.1 ± 0.1	5
	Adenosine	2.5 ± 0.3	5	2.0 ± 0.4	5
	Data are the maximal “folds increase” in average peak velocity (APV) above baseline (APVmax/APVbaseline) after intravenous injection of CVT-3146 (1 μg/kg) and adenosine (200 or 300 μg/kg)

[0110] The time course of the changes in CBF, PBF, HR and MAP and heart rate caused by CVT-3146 and adenosine are depicted in FIG. 12 and summarized in Table 8. The duration of >2-fold increase in coronary APV caused by CVT-3146 and adenosine was <120 sec and <20 sec, respectively; all parameters returned to baseline within 10 min and 5 min post injection of adenosine and CVT-3 146, respectively. Based on the differential selectivity of the vasodilatory effects of CVT-3146 and adenosine in the coronary and the peripheral arteries, and the dosage used, CVT-3146 is approximately 600 times more selective than adenosine in vasodilating the coronary vs. the peripheral arterial vasculature.

TABLE 8

Differential effects of CVT-3146 and Ado on blood flow velocity, heart rate and

mean arterial blood pressure (MAP)

CVT-3146 (1 ug/kg)

Ado (200 or 300 ug/kg)

Coronary (n = 6)

Peripheral (n = 5)

Coronary (n = 5)

Peripheral (n = 5)

Time

HR

MAP

APV

HR

MAP

APV

HR

MAP

APV

HR

MAP

APV

Baseline

171 ± 8

128 ± 7

49 ± 7

162 ± 9

116 ± 8

34 ± 8

170 ± 10

126 ± 6

42 ± 5

165 ± 10

118 ± 8

33 ± 3

10″

177 ± 8

126 ± 9

84 ± 4

167 ± 6

114 ± 9

34 ± 4

174 ± 7

126 ± 8

90 ± 10

171 ± 6

107 ± 9

62 ± 9

20″

186 ± 9

116 ± 9

106 ± 8

177 ± 7

107 ± 8

38 ± 4

166 ± 11

90 ± 9

85 ± 10

162 ± 15

85 ± 9

76 ± 10

30″

188 ± 7

117 ± 7

109 ± 10

183 ± 8

111 ± 6

41 ± 4

173 ± 6

107 ± 9

75 ± 9

168 ± 9

87 ± 9

59 ± 11

1′

186 ± 7

112 ± 7

100 ± 12

182 ± 7

102 ± 6

42 ± 4

175 ± 5

113 ± 6

39 ± 8

171 ± 6

105 ± 6

44 ± 7

1.5′

186 ± 7

110 ± 7

97 ± 11

184 ± 7

99 ± 7

42 ± 4

171 ± 5

117 ± 6

55 ± 8

171 ± 6

106 ± 7

38 ± 4

2′

185 ± 7

110 ± 7

88 ± 11

187 ± 8

98 ± 7

40 ± 4

170 ± 5

120 ± 6

52 ± 7

169 ± 6

110 ± 6

35 ± 3

3′

185 ± 7

110 ± 6

81 ± 11

187 ± 8

97 ± 7

40 ± 4

168 ± 8

125 ± 6

49 ± 6

165 ± 9

115 ± 7

33 ± 3

4′

183 ± 8

110 ± 6

75 ± 11

186 ± 8

100 ± 6

39 ± 4

164 ± 10

126 ± 6

47 ± 6

160 ± 9

119 ± 8

31 ± 3

5′

183 ± 8

112 ± 6

72 ± 11

185 ± 8

101 ± 6

37 ± 5

163 ± 10

128 ± 6

47 ± 6

158 ± 10

121 ± 7

30 ± 2

6′

182 ± 8

113 ± 6

65 ± 11

7′

181 ± 8

115 ± 6

63 ± 12

8′

180 ± 8

117 ± 6

55 ± 11

9′

177 ± 9

119 ± 6

55 ± 11

10′

176 ± 9

120 ± 6

53 ± 11

Values are the mean ± SEM of single determinations in each of the preparations (n).

Claims

1. A method for identifying compounds that are selective partial A2A adenosine receptor agonists with a short duration of action, comprising: a. measuring the intrinsic efficacy of a test compound in a cell line that express adenosine A2A receptors. b. measuring the intrinsic efficacy of a full agonist in said cell line; and c. selecting those compounds that have a lower intrinsic efficacy than said fall agonist;

2. The method of claim 1, comprising the additional steps of measuring the binding affinity (Ki) of the selected compounds; and selecting a compound with a Ki>1 μM.

3. The method of claim 1, wherein said test compound is a selective A2A adenosine receptor agonist.

4. The method of claim 1, wherein said cell line is rat pheochromocytoma PC12 cells, HEK-293 cells or porcine striatal cells.

5. The method of claim 4, wherein said cell line is rat pheochromocytoma PC12 cells.

6. The method of claim 1, wherein said full agonist is CGS21680 or WRC 0470.

7. The method of claim 6, wherein said full agonist is CGS21680.

8. The method of claim 7, wherein the intrinsic efficacy of the compound is 10%-95% of the intrinsic efficacy of CGS21680.

9. The method of claim 7, wherein the intrinsic efficacy of the compound is 50%-85% of the intrinsic efficacy of CGS21680.

10. The method of claim 1, wherein said binding affinity of the selected, compound is greater than or equal to 1 μM.

11. A method of myocardial perfusion imaging of a mammal, comprising administering a radionuclide and a compound identified by the method of claim 1 to a mammal in need thereof, and determining areas of insufficient blood flow.

12. The method of claim 11, wherein the compound is CVT-3033 or CVT-3146.

13. The method of claim 12, wherein the dose of CVT-3033 or CVT-3146 is from 0.20 μg/kg to 9 μg/kg.

14. The method of claim 12, wherein the dose of CVT-3033 or CVT-3146 is 0.25 to 5.0 μg/kg.