Back to Search Start Over

High affinity central benzodiazepine receptor ligands. Part 3: insights into the pharmacophore and pattern recognition study of intrinsic activities of pyrazolo[4,3-c]quinolin-3-ones

Authors :
Carotti, Andrea
Altomare, Cosimo
Savini, Luisa
Chiasserini, Luisa
Pellerano, Cesare
Mascia, Maria P.
Maciocco, Elisabetta
Busonero, Fabio
Mameli, Manuel
Biggio, Giovanni
Sanna, Enrico
Source :
Bioorganic & Medicinal Chemistry. Nov2003, Vol. 11 Issue 23, p5259. 14p.
Publication Year :
2003

Abstract

Novel 2-phenyl-2,5-dihydropyrazolo[4,3-c]quinolin-3-(3H)-ones (PQs) endowed with high affinity for central benzodiazepine receptor (BzR) were synthesized. In particular, 9-fluoro-2-(2-fluorophenyl)-2,5-dihydro-3H-pyrazolo[4,3-c]quinolin-3-one (22) showed binding affinity in the subnanomolar concentration range and proved to be in vitro a potent antagonist. This finding allowed the nature of the hydrogen bonding receptor site H2 to be established, as located between the N-1 nitrogen of the PQ nucleus and the ortho position of the N-2-aryl group. [35S]tert-Butylbicyclophosphorothionate ([35S]TBPS) binding assays and electrophysiological measurements of the effects on GABA-evoked Cl− currents at recombinant human α1β2γ2L GABAA receptors, expressed in Xenopus laevis oocytes, were used to assess the intrinsic activities of a large series of PQs. With the aim of extracting discriminant information and distinguishing BzR ligands with different profiles of efficacy, 51 PQ derivatives, including full and partial agonists, antagonists, and inverse agonists, were analyzed in a multidimensional chemical descriptor space, defined by the lipophilicity parameter CLOG P and 3-D molecular WHIM descriptors, by means of principal component analysis, k-nearest neighbors (k-NN) method, and linear discriminant analysis (LDA). The classification methods were applied to subsets of pairs of efficacy classes, and lipophilicity and 3-D size descriptors were detected as the discriminant variables by a stepwise linear discriminant analysis. LDA proved to be superior to k-NN, especially in classifying PQ ligands (60–84% of success in prediction ability) into categories of efficacies which were contiguous and quite overlapped in the hyperspace of variables. [Copyright &y& Elsevier]

Details

Language :
English
ISSN :
09680896
Volume :
11
Issue :
23
Database :
Academic Search Index
Journal :
Bioorganic & Medicinal Chemistry
Publication Type :
Academic Journal
Accession number :
11250442
Full Text :
https://doi.org/10.1016/S0968-0896(03)00527-3