Acyclic analogues of deoxyadenosine 3',5'-bisphosphates as P2Y(1) receptor antagonists.

P2Y(1) receptors are activated by ADP and occur on endothelial cells, smooth muscle, epithelial cells, lungs, pancreas, platelets, and in the central nervous system. With the aid of molecular modeling, we have designed nucleotide analogues that act as selective antagonists at this subtype. The present study has tested the hypothesis that acyclic modifications of the ribose ring, proven highly successful for nucleoside antiviral agents such as gancyclovir, are generalizable to P2Y receptor ligands. Specifically, the binding site of the P2Y(1) receptor was found to be sufficiently accommodating to allow the substitution of the ribose group with acyclic aliphatic and aromatic chains attached to the 9-position of adenine. Three groups of adenine derivatives having diverse side-chain structures, each containing two symmetrical phosphate or phosphonate groups, were prepared. Biological activity was demonstrated by the ability of the acyclic derivatives to act as agonists or antagonists in the stimulation of phospholipase C in turkey erythrocyte membranes. An acyclic N(6)-methyladenine derivative, 2-[2-(6-methylamino-purin-9-yl)-ethyl]-propane-1, 3-bisoxy(diammoniumphosphate) (10), containing an isopentyl bisphosphate moiety, was a full antagonist at the P2Y(1) receptor with an IC(50) value of 1.60 micro¿. The corresponding 2-Cl derivative (11) was even more potent with an IC(50) value of 0.84 microM. Homologation of the ethylene group at the 9-position to 3-5 methylene units or inclusion of cis- or trans-olefinic groups greatly reduced antagonist potency at the P2Y(1) receptor. Analogues containing a diethanolamine amide group and an aryl di(methylphosphonate) were both less potent than 10 as antagonists, with IC(50) values of 14 and 16 microM, respectively, and no agonist activity was observed for these analogues. Thus, the ribose moiety is clearly not essential for recognition by the turkey P2Y(1) receptor, although a cyclic structure appears to be important for receptor activation, and the acyclic approach to the design of P2 receptor antagonists is valid.

Separation of Galfbeta1-->XGlcNAc and Galpbeta1-->XGlcNAc (X = 3, 4, and 6) as the alditols by high-pH anion-exchange chromatography and thin-layer chromatography: characterization of mucins from Trypanosoma cruzi.

The O-linked N-acetylglucosamine oligosaccharides in the mucins of Trypanosoma cruzi may contain galactofuranose or galactopyranose, depending on the strain, one of the components being the disaccharide. Since galactofuranose is a site for antibody recognition, it is desirable to have a sensitive method for the detection of the galactofuranosyl structures. In this paper, we present procedures for the separation of Galfbeta1-->XGlcNAc and Galpbeta1-->XGlcNAc (X = 3, 4, and 6) as the corresponding alditols by high-pH anion-exchange chromatography with pulse amperometric detection. All the isomeric disaccharides could be resolved on a CarboPac PA-10 column, the galactofuranose-containing disaccharides being more retained in the column. GlcNAcol and Galfbeta1-->4(Galpbeta1-->6)GlcNAcol could be analyzed in the same run. The compounds could also be separated by thin-layer chromatography on silica gel 60, a convenient method for analysis of the radiolabeled alditols obtained by reductive beta-elimination in the presence of NaB(3)H(4). Both methods were applied for the analysis of the O-linked sugars in the mucins of T. cruzi CL 14 and revealed that they contained only N-acetylglucosamine and the disaccharide Galpbeta1-->4GlcNAc.

Synthesis of beta-D-Galf-(1-3)-D-GlcNAc by the trichloroacetamidate method and of beta-D-Galf-(1-6)-D-GlcNAc by SnCl4-promoted glycosylation.

In a continuation of our studies on the characterization of the glycoproteins of T. cruzi new galactofuranosyl disaccharides were synthesized. Beta-D-Galf-(1-3)-D-GlcNAc was prepared by employing the trichloroacetamidate procedure for the glycosylation step. The mild conditions of this reaction are appropriate for condensation of 2,3,5,6-tetra-O-benzoyl-beta-D-galactofuranosyl trichloroacetamidate with acid-labile benzyl 2-acetamido-4,6-O-benzylidene-2-deoxy-alpha-D-glucopyranoside. On the other hand, tin(IV) chloride promoted condensation of benzyl 2-acetamido-3-O-benzoyl-2-deoxy-alpha-D-glucopyranoside with penta-O-benzoyl-alpha-beta-D-galactofuranose gave the derivative of beta-D-Galf-(1-6)-D-GlcNAc in 78% yield.

One-pot synthesis of beta-D-Galf(1-->4)[beta-D-Galp(1-->6)]-D-GlcNAc, a 'core' trisaccharide linked O-glycosidically in glycoproteins of Trypanosoma cruzi.

Tin(IV) chloride-promoted condensation of benzyl 2-acetamido-3-O-benzoyl-2-deoxy-alpha-D-glucopyranoside (4) with penta-O-benzoyl-beta-D-galactopyranose (6) gave the derivative of beta-D-Galp-(1-->6)-alpha-D-GlcNAc 7 in 80% yield. This was glycosylated with penta-O-benzoyl-alpha, beta-D-galactofuranose (5), employing the same catalyst, to afford the protected benzyl per-O-benzoyl-beta-D-Galf(1-->4)[beta-D-Galp(1-->6)]D-GlcNAc 10 in 41% yield. Alternatively, compound 10 was obtained directly in a one-pot reaction from 4, by sequential addition of 6 and 5 (34% yield). beta-Glycosidic linkages were diastereoselectively formed. De-O-benzoylation of 10, followed by heterogeneous catalytic transfer hydrogenolysis of the benzyl group afforded the free trisaccharide beta-D-Galf(1-->4)[beta-D-Galp(1-->6)]-D-GlcNAc (14) in 98% yield from 10. Sodium borohydride reduction of 14 gave the corresponding alditol, whose spectral data were identical to those reported for the alditol obtained from the 38-43 kDa cell-surface glycoprotein of Trypanosoma cruzi.

Stimulation by alkylxanthines of chloride efflux in CFPAC-1 cells does not involve A1 adenosine receptors.

A series of 8-substituted derivatives of 1,3,7-alkylxanthines was synthesized as potential activators of chloride efflux from a human epithelial cell line (CFPAC) expressing the cystic fibrosis transmembrane regulator (CFTR) delta F508 mutation. Their interactions with rat brain A1 and A2a receptors were also studied in radioligand binding experiments. Substitution was varied at the xanthine 1-, 3-, 7- and 8-positions. 1,3-Dipropyl-8-cyclopentylxanthine (CPX) stimulated Cl- efflux in the 10(-8) M range, with a maximal effect reaching 200% of control and diminishing at higher concentrations. The potent adenosine antagonist 8-[4-[[[[(2-aminoethyl)amino]carbonyl]methyl]oxy]phenyl]- 1,3-dipropylxanthine, nonselective at human A1 and A2a receptors, was inactive in Cl- efflux. 1,3-Diallyl-8-cyclohexylxanthine (DAX) was highly efficacious in stimulating chloride efflux with levels reaching > 300% of control, although micromolar concentrations were required. 1,3,7-Trimethyl-8-(3-chlorostyryl)xanthine, an A2a-selective adenosine antagonist, was only weakly active. Caffeine, which acts as an nonselective adenosine antagonist in the range of 10(-5) M, was active in Cl- efflux in the low nanomolar range but with low efficacy. Thus, among the xanthine derivatives of diverse structure, there was no correlation between potency in Cl- efflux and adenosine antagonism. Poly(A)+ RNA isolated from CFPAC-1 cells showed no hybridization to a human A1 receptor cDNA probe, excluding this receptor as a mediator of CPX-elicited Cl- efflux. Thus, this action of xanthines in stimulating Cl- efflux in CFPAC cells, which express a defective CFTR, represents a novel site of action apparently unrelated to adenosine receptors.

A selective agonist affinity label for A3 adenosine receptors.

A newly synthesized, chemically reactive adenosine derivative, N6-(3-isothiocyanatobenzyl)adenosine-5'-N-methyluronamide, was found to bind selectively to A3 receptors. Ki values for this isothiocyanate derivative in competition binding at rat brain A1, A2a, and A3 receptors were 145, 272 and 10.0 nM, respectively. A preincubation with this derivative resulted in irreversible inhibition of radioligand binding at rat A3 receptors in membranes of transfected CHO cells or RBL-2H3 mast cells, but not at rat A1 or A2a receptors. The loss of binding sites for 0.1 nM [125I]N6-(4-aminobenzyl)adenosine-5'-N-methyluronamide, a high affinity A3 receptor radioligand, in transfected CHO cell membranes was concentration-dependent with an IC50 of 50 nM. No change was observed in the Kd value of the remaining A3 receptor sites. The inhibition was also insensitive to theophylline (1 mM), consistent with the pharmacology of rat A3 receptors. Structurally similar adenosine analogues lacking the chemically reactive isothiocyanate group failed to irreversibly inhibit A3-binding.

A binding site model and structure-activity relationships for the rat A3 adenosine receptor.

A novel adenosine receptor, the A3 receptor, has recently been cloned. We have systematically investigated the hitherto largely unexplored structure-activity relationships (SARs) for binding at A3 receptors, using 125I-N6-2-(4-aminophenyl)ethyladenosine as a radioligand and membranes from Chinese hamster ovary cells stably transfected with the rat A3-cDNA. As is the case for A1 and A2a receptors, substitutions at the N6 and 5' positions of adenosine, the prototypic agonist ligand, may yield fairly potent compounds. However, the highest affinity and A3 selectivity is found for N6,5'-disubstituted compounds, in contrast to A1 and A2a receptors. Thus, N6-benzyladenosine-5'-N-ethylcarboxamide is highly potent (Ki, 6.8 nM) and moderately selective (13- and 14-fold versus A1 and A2a). The N6 region of the A3 receptor also appears to tolerate hydrophilic substitutions, in sharp contrast to the other subtypes. Potencies of N6,5'-disubstituted compounds in inhibition of adenylate cyclase via A3 receptors parallel their high affinity in the binding assay. None of the typical xanthine or nonxanthine (A1/A2) antagonists tested show any appreciable affinity for rat A3 receptors. 1,3-Dialkylxanthines did not antagonize the A3 agonist-induced inhibition of adenylate cyclase. A His residue in helix 6 that is absent in A3 receptors but present in A1/A2 receptors may be causal in this respect. In a molecular model for the rat A3 receptor, this mutation, together with an increased bulkiness of residues surrounding the ligand, make antagonist binding unfavorable when compared with a previously developed A1 receptor model. Second, this A3 receptor model predicted similarities with A1 and A2 receptors in the binding requirements for the ribose moiety and that xanthine-7-ribosides would bind to rat A3 receptors. This hypothesis was supported experimentally by the moderate affinity (Ki 6 microM) of 7-riboside of 1,3-dibutylxanthine, which appears to be a partial agonist at rat A3 receptors. The model presented here, which is consistent with the detailed SAR found in this study, may serve to suggest future chemical modification, site-directed mutagenesis, and SAR studies to further define essential characteristics of the ligand-receptor interaction and to develop even more potent and selective A3 receptor ligands.

125I-4-aminobenzyl-5'-N-methylcarboxamidoadenosine, a high affinity radioligand for the rat A3 adenosine receptor.

The rat A3 adenosine receptor (AR) is a recently characterized AR subtype cloned from testis and brain cDNA libraries. N6-2-(4-Amino-3-[125I]iodophenyl)ethyladenosine, a high affinity A1AR agonist, has served as the only radioligand available for study of the A3AR. The relatively low affinity of N6-2-(4-amino-3-[125I] iodophenyl)ethyladenosine for the A3AR and its greater A1AR selectivity necessitate the development of more appropriate radioligands for A3AR analysis. This report characterizes 125I-4-aminobenzyl-5'-N-methylcarboxamidoadenosine (125I-AB-MECA), a high affinity radioligand for the A3AR, in two cell lines that express this AR subtype. Membranes from Chinese hamster ovary (CHO) cells expressing the rat A3AR and from the rat mast cell line RBL-2H3 bound 125I-AB-MECA with Kd values of 1.48 +/- 0.33 nM and 3.61 +/- 0.30 nM, respectively. As determined by 125I-AB-MECA binding, levels of A3AR expression in the A3AR-CHO cell line and RBL-2H3 cells were 3.06 +/- 0.21 pmol/mg and 1.02 +/- 0.13 pmol/mg, respectively. Binding of 125I-AB-MECA was characterized in competition assays. In the A3AR-CHO cell line a potency order of cyclohexyl-5'-N-ethylcarboxamidoadenosine (cyclohexyl-NECA) = benzyl-NECA > (-)-N6-[(R)-phenylisopropyl]adenosine = NECA was observed, and in RBL-2H3 cells (-)-N6-[(R)-phenylisopropyl]adenosine and NECA were equipotent. Xanthine amine congener (XAC) and 8-cyclopentyl-1,3-dipropylxanthine did not significantly inhibit 125I-AB-MECA binding. The parent compound, AB-MECA, dose-dependently inhibited forskolin-stimulated adenylyl cyclase activity in A3AR-CHO cell membranes. 125I-AB-MECA bound to the rat A1AR and canine A2aAR expressed in COS-7 cells with Kd values of 3.42 +/- 0.43 nM and 25.1 +/- 12.6 nM, respectively. This binding was significantly reduced in the presence of 1 microM XAC. In RBL-2H3 cells, XAC had no effect on 125I-AB-MECA affinity and reduced the level of radioligand binding by approximately 5%.

Structure-activity relationships of N6-benzyladenosine-5'-uronamides as A3-selective adenosine agonists.

Adenosine analogues modified at the 5'-position as uronamides and/or as N6-benzyl derivatives were synthesized. These derivatives were examined for affinity in radioligand binding assays at the newly discovered rat brain A3 adenosine receptor and at rat brain A1 and A2a receptors. 5'-Uronamide substituents favored A3 selectivity in the order N-methyl > N-ethyl approximately unsubstituted carboxamide > N-cyclopropyl. 5'-(N-Methylcarboxamido)-N6-benzyladenosine was 37-56-fold more selective for A3 receptors. Potency at A3 receptors was enhanced upon substitution of the benzyl substituent with nitro and other groups. 5'-N-Methyluronamides and N6-(3-substituted-benzyl)adenosines are optimal for potency and selectivity at A3 receptors. A series of 3-(halobenzyl)-5'-N-ethyluronamide derivatives showed the order of potency at A1 and A2a receptors of I approximately Br > Cl > F. At A3 receptors the 3-F derivative was weaker than the other halo derivatives. 5'-N-Methyl-N6-(3-iodobenzyl)adenosine displayed a Ki value of 1.1 nM at A3 receptors and selectivity versus A1 and A2a receptors of 50-fold. A series of methoxybenzyl derivatives showed that a 4-methoxy group best favored A3 selectivity. A 4-sulfobenzyl derivative was a specific ligand at A3 receptors of moderate potency. An aryl amino derivative was prepared as a probe for radioiodination and receptor cross-linking.

A role for central A3-adenosine receptors. Mediation of behavioral depressant effects.

The behavioral effects of a selective A3 adenosine receptor agonist 3-IB-MECA (N6-(3-iodobenzyl)-5'-N-methylcarboxamidoadenosine) in mice and the localization of radioligand binding sites in mouse brain were examined. Low levels of A3 adenosine receptors were detected in various regions of the mouse brain (hippocampus, cortex, cerebellum, striatum), using a radioiodinated, high-affinity A3-agonist radioligand [125I]AB-MECA (N6-(3-iodo-4-aminobenzyl)-5'-N-methylcarboxamidoadenosine). Scatchard analysis in the cerebellum showed that the Kd value for binding to A3 receptors was 1.39 +/- 0.04 nM with a Bmax of 14.8 +/- 2.1 fmol/mg protein. 3-IB-MECA at 0.1 mg/kg i.p. was a locomotor depressant with > 50% reduction in activity. Although selective A1 or A2a antagonists reversed locomotor depression elicited by selective A1 or A2a agonists, respectively, the behavioral depressant effects of 3-IB-MECA were unaffected. 3-IB-MECA also caused scratching in mice, which was prevented by coadministration of the histamine antagonist cyproheptadine. The demonstration of a marked behavioral effect of A3 receptor activation suggests that the A3 receptor represents a potential new therapeutic target.

Effect of trifluoromethyl and other substituents on activity of xanthines at adenosine receptors.

An aryl p-(trifluoromethyl) substituent increases the affinity of 1,3-disubstituted 8-phenylxanthines at A2a-adenosine receptors, while having little effect on affinity at A1-adenosine receptors. In contrast, an aryl p-(trifluoromethyl) substituent has little effect on affinity of 3,7-disubstituted and 1,3,7-trisubstituted 8-phenylxanthines. An aryl p-sulfo substituent reduces affinity of all 8-phenylxanthines at A1- and A2a-adenosine receptors. An 8-(trifluoromethyl) substituent markedly reduces affinity of 1,3-dialkylxanthines at both A1- and A2a-adenosine receptors. In contrast, 8-(trifluoromethyl)caffeine retains affinity for A2a-adenosine receptors, but does lose affinity for A1-adenosine receptors. 8-Bromo-, 8-acryl-, and 8-pent-1-enylcaffeines are also selective for A2-adenosine receptors, while 8-cyclobutylcaffeine is nonselective. 8-[trans-2-(tert-butyloxycarbonyl)vinylcaffeine is 20-fold selective for Aza vs A1 receptors.

8-(3-Chlorostyryl)caffeine (CSC) is a selective A2-adenosine antagonist in vitro and in vivo.

An adenosine antagonist, 8-(3-chlorostyryl)caffeine (CSC), was shown previously to be 520-fold selective for A2a-adenosine receptors in radioligand binding assays in the rat brain. In reversing agonist effects on adenylate cyclase, CSC was 22-fold selective for A2a receptors in rat phenochromocytoma cells (Kb 60 nM) vs. A1 receptors in rat adipocytes (Kb 1.3 microM). Administered i.p. in NIH mice at a dose of 1 mg/kg, CSC shifted the curve for locomotor depression elicited by the A2a-selective agonist APEC to the right (ED50 value for APEC shifted from 20 micrograms/kg i.p. to 190 micrograms/kg). CSC had no effect on locomotor depression elicited by an ED50 dose of the A1-selective agonist CHA. CSC alone at a dose of 5 mg/kg stimulated locomotor activity by 22% over control values. Coadministration of CSC and the A1-selective antagonist CPX, both at non-stimulatory doses, increased activity by 37% (P < 0.001) over CSC alone, suggesting a behavioral synergism of A1- and A2-antagonist effects in the CNS.

Structure-activity relationships of 8-styrylxanthines as A2-selective adenosine antagonists.

A series of substituted 8-styryl derivatives of 1,3,7-alkylxanthines was synthesized as potential A2-selective adenosine receptor antagonists, and the potency at rat brain A1- and A2-receptors was studied in radioligand binding experiments. At the xanthine 7-position, only small hydrophobic substituents were tolerated in receptor binding. 7-Methyl analogues were roughly 1 order of magnitude more selective for A2 versus A1 receptors than the corresponding 7-H analogues. 1,3-Dimethylxanthine derivatives tended to be more selective for A2-receptors than the corresponding 1,3-diallyl, diethyl, or dipropyl derivatives. Substitutions of the phenyl ring at the 3-(monosubstituted) and 3,5-(disubstituted) positions were favored. 1,3, 7-Trimethyl-8-(3-chlorostyryl)xanthine was a moderately potent (Ki vs [3H]CGS 21680 was 54 nM) and highly A2-selective (520-fold) adenosine antagonist. 1,3,7-Trimethyl-8-[(3-carboxy-1-oxopropyl)amino] styryl]xanthine was highly A2-selective (250-fold) and of enhanced water solubility (max 19 mM). 1,3-Dipropyl-7-methyl-8-(3,5-dimethoxystyryl) xanthine was a potent (Ki = 24 nM) and very A2-selective (110-fold) adenosine antagonist.