Stereoisomers of N-substituted soft anticholinergics and their zwitterionic metabolite based on glycopyrrolate--syntheses and pharmacological evaluations.

Abstract

In this study, isomers of two N-substituted soft anticholinergics based on glycopyrrolate, SGM (PcPOAGP_NA.Me) and SGE (PcPOAGP_NA.Et) [3'-(2-cyclopentyl-2-phenyl-2-hydroxyacetoxy)-1'-methyl-1'-alkoxycarbonylpyrrolidinium bromide] and their zwitterionic metabolite, SGa (PcPOAGP_NA.H) [3'-(2-cyclopentyl-2-phenyl-2-hydroxyacetoxy)-1'-methyl-1'-carboxymethylpyrrolidinium inner salt] were synthesized and their pharmacological activities were evaluated in vitro and in vivo. The isomers of SGM and SGE were synthesized with both optically pure methyl-cyclopentylmandelate and 3-hydroxy-N-methylpyrrolidine. Trans-esterification followed by quarternization with alkyl bromoacetate gave four isomers of SGM or SGE with the nitrogen chiral center unresolved (2R3'S-SGM, 2R3'R-SGM, 2S3'S-SGM, 2S3'R-SGM or 2R3'S-SGE, 2R3'R-SGE, 2S3'S-SGE, 2S3'R-SGE). The hydrolysis of these four isomers followed by HPLC separation resulted in eight fully resolved isomers of SGa (2R3'R1'R, 2R3'S1'R, 2R3'R1'S, 2R3'S1'S, 2S3'R1'R, 2S3'S1'R, 2S3'R1'S, and 2S3'S1'S). Pharmacological activities were assessed by using in vitro receptor-binding assay and guinea pig ileum pA2-assay, and by evaluating the in vivo rabbit mydriatic effects. Results were compared to those obtained with conventional anticholinergic agents, such as glycopyrrolate, N-meythylscopolamine, and tropicamide, as well as those obtained with previously prepared racemic mixtures and 2R isomers. Receptor binding pKi values at cloned human muscarinic receptors (M1-M4 subtypes) were in the 6.0-9.5 range for the newly synthesized SGM and SGE isomers, and in the 5.0-8.6 range for the SGa isomers. In all cases, 2R isomers were significantly more active than 2S isomers (27 to 447 times for SGM isomers, and 6 to 4467 times for SGa isomers). Among the four SGM isomers with unresolved 1' (N) chiral center, the 3'R isomers were more active than the corresponding 3'S isomers (1.5-12.9 times), whereas, among the SGa isomers, the 3'S isomers were not always more active than the corresponding 3'R isomers indicating that activity determined based on configuration at chiral center 3' is significantly affected by the configuration of the other two chiral centers, 2 and 1'. Among the completely resolved eight SGa isomers (all three chiral centers resolved), 1'S isomers were always more active than the corresponding 1'R isomers (1.8-22.4 times). Results also indicate that some isomers showed good M3/M2 muscarinic-receptor subtype-selectivity (about 3-5 times), and 2R and 3'S were the determining configurations for this property. Guinea pig ileum assays and rabbit mydriasis tests on SGa isomers further confirmed the stereospecificity. In rabbit eyes, some 2R-SGa isomers showed mydriatic potencies similar to glycopyrrolate and exceeded tropicamide, but their mydriatic effects lasted considerably shorter, and they did not induce dilation of the pupil in the contralateral, water-treated eye. These results indicate that these compounds are locally active, but safe and have a low potential to cause systemic side effects. The pharmacological potency of the eight SGa isomers was estimated as 2R3'S1'S approximately equal to 2R3'R1'S approximately equal to 2R3'S1'R > 2R3'R1'R > 2S3'R1'S > 2S3'S1'S approximately equal to 2S3'R1'R > 2S3'S1'R (p < 0.05). The stereospecificity and M3/M2 muscarinic-receptor subtype-selectivity of soft anticholinergics, SGM, SGE, and SGa have been demonstrated. In agreement with previous results, the potential for their effective and safe use has been confirmed.