A 3D QSAR Study on a Set of Dopamine D₄ Receptor Antagonists

Abstract

The molecular alignments obtained from a previously reported pharmacophore model have been employed in a three-dimensional quantitative structure-activity relationship (3D QSAR) study, to obtain a more detailed insight into the structure-activity relationships for D(2) and D(4) receptor antagonists. The frequently applied CoMFA method and the related CoMSIA method were used. Statistically significant models have been derived with these two methods, based on a set of 32 structurally diverse D(2) and D(4) receptor antagonists. The CoMSIA and the CoMFA methods produced equally good models expressed in terms of q(2) values. The predictive power of the derived models were demonstrated to be high. Graphical interpretation of the results, provided by the CoMSIA method, brings to light important structural features of the compounds related to either low- or high-affinity D(2) or D(4) antagonism. The results of the 3D QSAR studies indicate that bulky N-substituents decrease D(2) binding, whereas D(4) binding is enhanced. Electrostatically favorable and unfavorable regions exclusive to D(2) receptor binding were identified. Likewise, certain hydrogen-bond acceptors can be used to lower D(2) affinity. These observations may be exploited for the design of novel dopamine D(4) selective antagonists.

The Effect of Molecular Fields, Lattice Spacing and Analysis Options on CoMFA Predictive Ability

“…Both approaches resulted in equivalent statistical significance for every test and thus only the results of the LOO analyses are detailed below. Inhibitors of angiotensin converting enzyme 114 ACHE [11] Inhibitors of acetyl-cholinesterase 111 AI [25,29] Steroid aromatase inhibitors 78 ARB [30] Nonpeptide angiotensin II receptor antagonists 28 ATA [31] Anti-tuberculosis agents 94 BZR [11] Inhibitors of benzodiazepine receptor 163 CBRA [32] Cannabinoid CB1 receptor agonists 32 COMT [33] Inhibitors of catechol-O-methyltransferase 92 COX2 [11] Inhibitors of cyclooxygenase-2 322 DAT [34] Piperidine analogues for dopamine transporter 42 DHFR [11] Inhibitors of rat dihydrofolate reductase 397 DR [35,36] Antagonists of dopamine receptor 38 ECR [37] Binding of diacylhydrazine to ecdysone receptor 50 EDC [38] Estrogen disrupting chemicals 123 GHS [39] Growth hormone secretagogue mimics 31 GPB [11] Inhibitors of glycogen phosporylase b 66 GSK3B [40] Inhibition of Glycogen synthase kinase 3 42 HIVPR [41] Inhibitors of human immunodeficiency virus protease 113 HIVRT [42,43] Inhibition of HIV-1 reverse transcriptase 101 KOA [44] Kappa opioid antagonists 39 MX [45] Mutagenicity of mutagen X analogues 29 PDE [46] Inhibition of phosphodiesterase-IV 29 PTC [47] Phase-transfer asymmetric catalysts 40 RYR [25,48] Binding of ryanoids to the ryanodine receptor 18 STEROIDS [3,25] Binding of steroids to carrier proteins 21 TCHK [49] Inhibition of trypanosoma cruzi hexokinase 42 THERM [11] Inhibitors of thermolysin 76 THR [11] Inhibitors of thrombin 88 TP2A [50] Inhibition of topoisomerase-IIa 25 YOPH [51] Inhibitors of yersinia protein tyros...…”

Fingal: A Novel Approach to Geometric Fingerprinting and a Comparative Study of Its Application to 3D-QSAR Modelling

“…For the first dataset, CoMSIA and CoMFA results are provided; while for the second dataset, the CoMFA, HQSAR, and FRED/ SKEYS results are given. The two datasets used in this study are a set of 38 D 2 receptor antagonists [4] and a set of 58 estrogen receptor agonists [5]. Fingal and Dragon descriptors were applied in modelling both datasets with topological (Fingal2D and Dragon2D) and geometric (Fingal3D and Dragon3D) descriptors separately; the geometries were calculated using Corina [7].…”

“…Fingal and Dragon descriptors were applied in modelling both datasets with topological (Fingal2D and Dragon2D) and geometric (Fingal3D and Dragon3D) descriptors separately; the geometries were calculated using Corina [7]. The D 2 receptor antagonists were additionally modelled using geometric descriptors and the semi-empirical geometries provided in [4] (FingalSE and DragonSE). The 38 D 2 receptor antagonists were partitioned, as reported in [4], into a training set of 32 structures and an unseen test set of 6 structures; although 41 structures were reported in [4], 3 of those structures were reported without the required binding affinities.…”

“…The D 2 receptor antagonists were additionally modelled using geometric descriptors and the semi-empirical geometries provided in [4] (FingalSE and DragonSE). The 38 D 2 receptor antagonists were partitioned, as reported in [4], into a training set of 32 structures and an unseen test set of 6 structures; although 41 structures were reported in [4], 3 of those structures were reported without the required binding affinities. The 58 estrogen receptor agonists were used to generate 8 separate models with each of the structural classes defined in [5] being applied as the unseen test set.…”

CoMFA/CoMSIA and Pharmacophore Modelling as a Powerful Tools for Efficient Virtual Screening: Application to Anti-Leishmanial Betulin Derivatives

“…(Perkins et al, 2003) The CoMFA Lennard-Jones and Coulomb potentials are sharp and may introduce errors in scaling, alignment sensitivity, and interpretation of contours. (Bostrom et al, 2003) In order to improve these shortcomings, the comparative molecular similarity indices (CoMSIA) methods have been developed that make usage in addition to the electrostatic fields of hydrophobic fields, supposed to account better for differences in the entropic contribution to binding free energy, hydrogen bonding fields, as well as use smoother potentials based on Gaussian functions, which are less sensible to variation in alignment and lead to more interpretable contours. (Buolamwini & Assefa, 2003) CoMFA/CoMSIA alignment rules…”