2-[3-[2-(4,5-Diphenyl-2-oxazolyl)ethyl]phenoxy]acetic acid, 1, has been described as a non-prostanoid PGI2 mimetic that demonstrates anti-thrombotic properties of long duration in animal models of thrombosis. The effects of substitution and modification of the carbon beta-to the oxazole heterocycle of 1 were examined and equated with the potency of the compounds as inhibitors of ADP-induced human platelet aggregation in vitro. Potency was sensitive to both the size of the substituent and the identity of the beta-atom. The carbamates 13c-e demonstrated IC50's of 0.28-0.36 microM and were significantly more potent than the progenitor 1, IC50 = 1.2 microM. The ethyl carbamate 13c displaced [3H]-iloprost from platelet membranes in a concentration-dependent fashion that was half maximal at 20 nM, which compares with IC50's of 171 nM for 1 and 39 nM or unlabelled iloprost. Carbamate 13c stimulated platelet adenylate cyclase but the maximal effect was less than that observed for PGI2, identifying 13c as a partial agonist at the platelet PGI2 receptor.

The oral activity and antithrombotic efficacy of BMY 42393 was examined in ex vivo platelet aggregation studies and arterial thrombosis animal models. In a heterologous ex vivo platelet aggregation assay, ADP-induced human platelet aggregation was inhibited when washed human platelets were combined with rat platelet-poor plasma, taken from rats previously orally-dosed with BMY 42393. The IC50 for platelet aggregation inhibition was approximately 10 mg/kg. In a laser-induced thrombosis model, thrombus formation in a revascularized rabbit ear chamber was prevented in a dose-dependent fashion with an ED50 of about 2 mg/kg. A relatively long duration of anti-thrombotic activity was observed in the rabbit ear laser-induced thrombus study and the ex vivo platelet studies. Inhibition of thrombus formation was also demonstrated in a canine model of electrically-induced coronary artery thrombosis. BMY 42393 also prevented cyclic flow reductions in a monkey stenotic renal artery model. These studies indicate that BMY 42393 is orally active and capable of preventing laser and electric current-induced thrombus formation in animal models of arterial thrombosis.

The 4,5-diphenyloxazole derivatives 2-4 were previously identified as nonprostanoid prostacyclin (PGI2) mimetics. A series of derivatives of 2-4 bearing substitutents at the carbon atom alpha to the oxazole ring were synthesized and evaluated as inhibitors of ADP-induced aggregation of human platelets in vitro. In the unsaturated series, the alpha-carbethoxy derivative 10a, evaluated as an equal mixture of geometrical isomers, inhibited platelet aggregation with an IC50 of 0.36 microM. Evaluation of the individual methyl ester derivatives (E)-9a and (Z)-9a revealed that (E)-9a was 10-fold more potent than (Z)-9a. In the saturated series, the alpha-carbomethoxy-substituted compound 12a inhibited platelet aggregation with an IC50 of 0.08 microM, 15-fold more potent than the unsubstituted prototype 2. The potency of 12a was found to be sensitive to variation of the methoxy moiety. The ethyl (12b) and isopropyl (12d) esters were less effective as were the acid 12e and a series of amides (12f-h). Other substituents introduced at this site of the pharmacophore included P(O)(OEt)2 (25), SCH3 (31a), S(O)CH3 (31b), SO2CH3 (31c), isopropyl (31d), phenyl (31f), and CH2OH (31i). However, none were significantly more potent inhibitors of platelet function than the parent compound 2. The results indicate the presence of a pocket in the PGI2 receptor protein that preferentially recognizes small, polar but uncharged substituents. The structure-activity correlates are suggestive of a hydrogen-bond interaction between a donor moiety on the PGI2 receptor and the methoxycarbonyl functionality of 12a that is sensitive to both the size of the substituent and its stereochemical presentation in this structural class of PGI2 mimetic. The ethyl ester 12b dose-dependently displaced [3H]iloprost from human platelet membranes and stimulated adenylate cyclase. However, the maximal stimulation was less than that recorded for iloprost, indicating that 12b functions as a partial agonist at the PGI2 receptor.

cis-[3-[2-(4,5-Diphenyl-2-oxazolyl)ethenyl]phenoxy]acetic acid (3) was previously identified as a nonprostanoid prostacyclin (PGI2) mimetic that potently inhibits ADP-induced aggregation of human platelets with an IC50 of 0.18 microM. As part of an effort to further explore structure-activity relationships for this class of platelet inhibitor and to provide additional insight into the nonprostanoid PGI2 mimetic pharmacophore, the effect of constraining the cis-olefin moiety of 3 into various ring systems was examined. Incorporation of the cis-olefin of 3 into either an oxazole (26) or an unsubstituted pyrazole (35) heterocycle provided compounds that are equipotent with progenitor 3. However, the oxazole 11f, which is isomeric with 26, inhibits ADP-induced human platelet aggregation in vitro with an IC50 of 0.027 microM, 6-fold more potent than 3, 26, or 35. These results suggest that the central oxazole ring of 11f is functioning as more than a simple scaffold that provides optimal stereodefinition for interaction with the PGI2 receptor. The nitrogen atom of the central heterocycle of 11f is postulated to engage in hydrogen-bond formation with a donor moiety in the PGI2 receptor protein, an interaction not available to 26 due to the markedly different topology. In support of this contention, the crystal structures of 11f and 26 contain strong intermolecular hydrogen bonds between the carboxylic acid hydrogen atom and the nitrogen atom of the central oxazole ring. Although 11f and 26 are exact isosteres and could, in principle, adopt the same molecular packing arrangement in the solid state, this is not the case, and the intermolecular hydrogen-bonding interactions in 11f and 26 are accommodated by entirely different molecular packing arrangements. Incorporation of the olefin moiety of 3 into a benzene ring provided a compound, 40, over 60-fold weaker with an IC50 of 11.1 microM. The affinities of 11f, 26, 31, 32, and 40 for the human platelet PGI2 receptor, determined by displacement of [3H]iloprost, correlated with inhibition of platelet function. The solid-state structures of 11f, 26, 31, 32, and 40 were determined and revealed that the more potent compounds 11f and 26 adopt a relatively planar overall topography. In contrast, the central phenyl ring and the phenoxy ring of the weakly active compound 40 are rotated by 53 degrees from planarity. The chemical shifts of the protons of the phenoxy rings of 3, 11f, 18, 26, 31, 32, and 40 suggest that in solution 3, 11f, 18, and 26 adopt a planar conformation while 40 does not.(ABSTRACT TRUNCATED AT 400 WORDS)

A series of 1,3-dihydro-2H-imidazo[4,5-b]quinolin-2-one derivatives, substituted at the 7-position with functionalized side chains, was synthesized and evaluated as inhibitors of human blood platelet cAMP phosphodiesterase (PDE) as well as ADP- and collagen-induced platelet aggregation, in vitro. Structural modifications focused on variation of the side-chain terminus, side-chain length, and side-chain connecting atom. Functionality incorporated at the side-chain terminus included carboxylic acid, ester and amide, alcohol, acetate, nitrile, tetrazole, and phenyl sulfone moieties. cAMP PDE inhibitory potency varied and was dependent upon the side-chain terminus and its relationship with the heterocyclic nucleus. Methylation at N-1 or N-3 of the heterocycle diminished cAMP PDE inhibitory potency. Several representatives of this structural class demonstrated potent inhibition of ADP- and collagen-induced blood platelet aggregation and were half-maximally effective at low nanomolar concentrations. Amides 13d, 13f, 13h, 13k, 13m, and 13w are substantially more potent than relatively simply substituted compounds. However, platelet inhibitory properties did not always correlate with cAMP PDE inhibition across the series, probably due to variations in membrane permeability. Several compounds inhibited platelet aggregation measured ex vivo following oral administration to rats. Ester 11b, acid 12b, amide 13d, and sulfone 29c protected against thrombus formation in two different animal models following oral dosing and were found to be superior to anagrelide (2) and BMY 20844 (5). However, ester 11b and acid 12b demonstrated a unique pharmacological profile since they did not significantly affect hemodynamic parameters in dogs at doses 100-fold higher than that required for complete prevention of experimentally induced vessel occlusion in a dog model of thrombosis.

4,5-Diphenyl-2-oxazolenonanoic acid (18b) was synthesized and found to inhibit ADP-induced aggregation of human platelets with an IC50 of 2.5 microM. Acid 18b displaced [3H]iloprost from human platelet membranes in a concentration-dependent fashion, consistent with 18b inhibiting platelet function by acting as a prostacyclin mimetic. By inserting a phenoxy ring into the side-chain moiety of 18b and systematically varying the pattern of substitution and length of the tethers, more potent inhibitors of platelet aggregation were identified. A phenoxy ring inserted centrally in the side chain proved to be the optimal arrangement but significant activity was observed when the aromatic ring was bound directly to the 2 position of the heterocycle. The meta-substituted cis-(ethenylphenoxy)acetic acid 37 is the most potent platelet aggregation inhibitor synthesized as part of this study with an IC50 of 0.18 microM. Acid 37 displaces [3H]iloprost from human platelet membranes with an IC50 of 6 nM. The trans-olefinic isomer of 37 (25p) is 72-fold weaker as an inhibitor of ADP-induced platelet aggregation, but the saturated derivative 25w (BMY 42393) is intermediate in potency. Structure-activity studies using 25w as a template focused on modification of the tethers intervening between the side-chain phenyl ring and the oxazole and carboxylate termini and substitution of the phenyl ring. These studies revealed that biological activity was sensitive to both the identity of the concatenating atoms and the pattern of ring substitution. The structure-activity relationships provide insight into the topographical relationship between the diphenylated oxazole ring and the carboxylic acid terminus that comprise the nonprostanoid prostacyclin mimetic pharmacophore.

4,5-Diphenyl-2-oxazolenonanoic acid (2) and 2-[3-[2-(4,5-diphenyl-2-oxazolyl)ethyl]phenoxy]acetic acid (3) were previously identified as nonprostanoid prostacyclin (PGI2) mimetics that inhibit ADP-induced aggregation of human platelets in vitro. The effects on biological activity of substitution and structural modification of the 4- and 5-phenyl rings of 3 was examined. Potency showed a marked sensitivity to the introduction of substituents to these aromatic rings and only the bis-4-methyl derivative 9j, IC50 = 0.34 microM, demonstrated enhanced potency compared to the parent structure 3, IC50 = 1.2 microM. Substitution at the ortho or meta positions of the phenyl rings, replacement by thiopheneyl or cyclohexyl moieties, or constraining in a planar phenanthrene system resulted in compounds that were less effective inhibitors of ADP-induced platelet aggregation. In contrast, variation of the heterocycle moiety revealed a much less stringent SAR and many 5- and 6-membered heterocycles were found to effectively substitute for the oxazole ring of 2 and 3. The diphenylmethyl moiety functioned as an effective isostere for 4,5-diphenylated heterocycles since 13aad showed similar platelet inhibitory activity to 3. With the exception of the 3,4,5-triphenylpyrazole derivative 13g, compounds presenting the (m-ethylphenoxy)acetic acid side chain discovered with 3 demonstrated enhanced potency compared to the analogously substituted alkanoic acid derivative. The structure-activity findings led to a refinement of a model of the nonprostanoid PGI2 mimetic pharmacophore.

A series of phenylated pyrazoloalkanoic acid derivatives were synthesized and evaluated as inhibitors of ADP-induced human platelet aggregation. 3,4,5-Triphenyl-1H-pyrazole-1-nonanoic acid (8d), with an IC50 of 0.4 microM, was the most potent inhibitor identified in this study. Biochemical studies determined that 8d increased intraplatelet cAMP accumulation and stimulated platelet membrane-bound adenylate cyclase in a concentration-dependent fashion. Displacement of [3H]iloprost by 8d from platelet membranes indicated that the platelet prostacyclin (PGI2) receptor is the locus of biological action. Structure-activity studies demonstrated that the minimum structural requirements for binding to the platelet PGI2 receptor and inhibition of ADP-induced platelet aggregation within this series are a vicinally diphenylated pyrazole substituted with an omega-alkanoic acid side chain eight or nine atoms long. Potency depended upon both side-chain length and its topological relationship with the two phenyl rings.

A series of 1,3-dihydro-2H-imidazo[4,5-b]quinolin-2-one derivatives was synthesized and evaluated as inhibitors of cAMP hydrolysis by a crude human platelet phosphodiesterase preparation and as inhibitors of ADP- and collagen-induced aggregation of rabbit blood platelets. The parent structure 7a, demonstrated potent inhibitory activity that was enhanced by the introduction of alkyl, alkoxy, or halogen substituents at the 5-, 6-, 7-, and 8-positions. Methylation at N-1 or N-3 produced weaker inhibitors of cAMP PDE and platelet aggregation. 1,3,9,9a-Tetrahydro-2H-imidazo[4,5-b]quinolin-2-ones (6) were found to be equipotent with their fully oxidized congeners (7). On the basis of platelet inhibitory properties in vitro, efficacy at preventing thrombus formation in animal models of thrombosis, and a favorable hemodynamic profile, 1,3-dihydro-7,8-dimethyl-2H- imidazo[4,5-b]quinolin-2-one (7o, BMY 20844) was selected for advancement into toxicological evaluation and clinical trial. An efficient synthesis of 7o is described.

Compounds containing the imidazoquinoline nucleus are a new class of potent, broad-spectrum inhibitors of platelet aggregation. This report describes studies with a simply-substituted imidazoquinoline (BMY 20844) and several new ether-linked side chain derivatives (BMY 21638 and BMY 43351). These compounds are potent inhibitors of platelet cAMP phosphodiesterase (IC50 values: BMY 20844, 1.3 X 10(-8); BMY 21638, 2 X 10(-10); and BMY 43351, 1 X 10(-10) M, measured using 0.15 microM cAMP) but have little effect on platelet homogenate cGMP phosphodiesterase (IC50 greater than 10(-5) M). Inhibition of different cAMP phosphodiesterase isozymes was tested to determine if the compounds inhibited similar isozymes in other tissues. Rabbit heart cAMP phosphodiesterase isozymes were resolved by ion-exchange chromatography and three peaks of activity were obtained. BMY 20844 inhibited only fraction III (a "cGMP-inhibitable, low Km" cAMP-specific phosphodiesterase) with an IC50 value of 5 X 10(-8) M. These compounds also inhibited canine cardiac sarcoplasmic reticulum membrane-bound "cGMP-inhibitable, low Km" cAMP-specific phosphodiesterase with virtually the same potency as inhibition of cAMP phosphodiesterase in platelet homogenate. In washed platelets these compounds elevated cAMP levels and activated the platelet cAMP dependent protein kinase. Activation of cAMP-dependent protein kinase was determined by cAMP-dependent protein kinase ratio measurements and phosphorylation of intracellular proteins. These studies suggest that this potent new class of agents inhibits platelet phosphodiesterase activity in intact platelets causing an elevation in cAMP levels sufficient to activate the cAMP-dependent protein kinase and stimulate protein phosphorylation. This mechanism is, at least in part, responsible for the ability of these compounds to prevent platelet aggregation and thrombosis in experimental animal models.

BMY-43351 is a new broad-spectrum inhibitor of platelet aggregation with greater aqueous solubility than earlier analogs from the imidazoquinoline series. This report compares the antithrombotic activity of BMY-43351 to that of two other imidazoquinolines: BMY-20844, a simply-substituted compound, and BMY-21638, a more potent ether-linked side chain analog. All of these compounds act, at least in part, via inhibition of platelet low-Km cyclic AMP phosphodiesterase. Antithrombotic activity was assessed in the rabbit ear chamber-biolaser preparation, an animal model of small vessel thrombosis, and in the canine coronary artery stenosis-occlusion model of large vessel thrombosis. BMY-43351 was found to be remarkably potent in the biolaser model, with an EDso of 0.074 mg/kg p.o. In comparison, compounds such as aspirin, ticlopidine, sulfinpyrazone, and dipyridamole demonstrate little or no activity at much higher doses, (eg. 100 mg/kg p.o.). Other inhibitors of platelet low Km cyclic AMP phosphodiesterase are active but substantially weaker than BMY-43351. Similarly, in the coronary artery stenosis-occlusion model, BMY-43351 demonstrated impressive activity, significantly inhibiting arterial thrombosis at intraduodenal doses as low as 1 micrograms/kg. The potential use of BMY-43351 as adjunct therapy in thrombolysis was suggested in a series of experiments where this drug was used in combination with a thrombolytic regimen of stretokinase plus heparin. In this experimental setting, time to reperfusion was reduced from 42 +/- 5 minutes to 11 +/- 5 minutes, and reocclusion was totally inhibited.

Encainide is a class 1C antiarrhythmic agent that is indicated for the treatment of life-threatening arrhythmias, such as sustained ventricular tachycardia. Furthermore, encainide possesses a moderate degree of antiserotonin activity, which was quantitated in this present study by determining displacement of [3H]spiperone binding from rat cortical 5-HT2 binding sites. The Ki for encainide in this model was 66.1 nM, compared to 2.6 nM for ketanserin. Two encainide metabolites, ODE and MODE, were also active, but were weaker than encainide. Additionally, these agents were found to inhibit platelet aggregation induced in vitro in human platelet-rich plasma by the combination of ADP and serotonin. In view of the fact that serotonin is one of a variety of humoral factors capable of activating blood platelets and has been recently implicated as playing a role in certain thrombotic syndromes, encainide, along with its two principal human metabolites, ODE and MODE, and another class 1C antiarrhythmic, flecainide, were evaluated in an in vivo model of intravascular thrombosis. Intraduodenal doses of 1 mg/kg of either encainide, ODE, or MODE significantly inhibited thrombosis in a canine model of coronary artery stenosis-occlusion.

Octimibate inhibited ADP- and collagen-induced platelet aggregation in human, rabbit and rat platelet-rich plasma. Washed human platelets treated with octimibate had elevated cyclic AMP (cAMP) levels and cAMP-dependent protein kinase activity. When whole platelets were incubated with radiolabeled phosphate, octimibate produced an increase in the phosphorylation of platelet proteins with relative molecular weights of 22, 26, 50 and 80 kilodaltons. This pattern of protein phosphorylation is identical to that observed when the platelets were treated with forskolin, phosphodiesterase inhibitors or other compounds that elevate platelet cAMP levels. Octimibate also inhibited the rise in intracellular Ca++ caused by thrombin, as measured using Fura-2-loaded platelets, which is consistent with octimibate's ability to elevate platelet cAMP levels. When isolated platelet plasma membranes were treated with octimibate, adenylate cyclase activity was stimulated, reaching maximal activation at 1 microM octimibate. (The maximal activation of adenylate cyclase observed with octimibate is 70-75% of that observed with 10 microM PGE1.) This stimulation of platelet adenylate cyclase activity was enhanced by GTP. Octimibate competed for radiolabeled prostaglandin E1 and lloprost binding to isolated platelet membranes at submicromolar concentrations, but did not compete with radiolabeled prostaglandin D2 binding. These studies suggest that octimibate inhibits platelet aggregation by activating platelet adenylate cyclase through stimulation of platelet prostacyclin receptors.

The effects of 1,3-dihydro-7,8-dimethyl-2H-imidazo[4,5-b]quinolin-2-one (BMY-20844) on platelet function and experimental thrombosis were evaluated in a series of in vitro, ex vivo and in vivo experiments. The compound inhibited platelet aggregation in vitro in platelet rich plasma obtained from humans, rats and rabbits with EC50s of less than 1 microgram/ml when aggregation was induced by ADP, collagen or thrombin. Supra-additive interaction against ADP aggregation was also observed when BMY-20844 was combined with prostacyclin. BMY-20844 was orally active with an ex vivo ED50 in the rat of 3.2 mg/kg vs ADP. Significant antithrombotic activity was observed in two animal models (laser induced thrombosis in the microcirculation of the rabbit ear and coronary artery thrombosis in the dog). Inhibitions of 52% at 3 mg/kg p.o. in the laser model and 100% at 1 mg/kg i.d. in the coronary artery thrombosis model were obtained. Modest inotropic and hemodynamic effects were observed in ferrets and dogs. BMY-20844 was found to be a potent, specific inhibitor of platelet low Km cyclic AMP phosphodiesterase.

The anesthetized ferret model is introduced as an in vivo acute model for evaluating inotropic effects of new cardiovascular compounds. The effects of anagrelide and milrinone were compared to vehicle. Validation was accomplished using intravenous isoproterenol, which elicited a dose-dependent inotropic response. This suggests that the anesthetized ferret model is both reproducible and dependable. Additionally, in vivo inotropic activity can be assessed on small amounts of compound in a relatively inexpensive species.

Anagrelide is a member of the imidazo (2,1-b) quinazolin-2-one series of compounds, with a powerful antiaggregating effect on platelets. During studies in humans, anagrelide in small doses has produced thrombocytopenia. We therefore evaluated it in the treatment of thrombocytosis, and to date, platelet levels in 15 of 17 patients with primary thrombocythemia, 2 patients with polycythemia vera and thrombocytosis, and 1 patient with chronic granulocytic leukemia and thrombocytosis have been well controlled with the use of this agent. Induction doses of 1.0 to 1.5 mg given orally every six hours have produced a decrease in the platelet count, starting on day 5 and reaching a normal level by day 12. Side effects of anagrelide have been minimal. Maintenance therapy with 1.5 to 4.0 mg a day has continued to control the platelet count in patients for up to 28 months. This new agent appears promising in the treatment of thrombocytosis in patients with chronic myeloproliferative disease.