Four-Dimensional Quantitative Structure−Activity Relationship Analysis of a Series of Interphenylene 7-Oxabicycloheptane Oxazole Thromboxane A₂ Receptor Antagonists

Abstract

A series of 39 (a training set of 29 and a test set of 10) interphenylene 7-oxabicyclo [2.2.1]heptane oxazole thromboxane A2 (TXA2) receptor antagonists were studied using four-dimensional quantitative structure-activity relationship (4D-QSAR) analysis. Two thousand conformations of each analogue were sampled to generate a conformational energy profile (CEP) from a molecular dynamic simulation (MDS) of 100,000 trajectory states. Each conformation was placed in a grid cell lattice for each of six trial alignments. Cubic grid cell sizes of 1 and 2 A were considered. The frequency of occupation of each grid cell was computed for each of seven types of pharmacophoric group classes of atoms of each compound. These grid cell occupancy descriptors (GCODs) were then used as independent variables in constructing three-dimensional (3D)-QSAR models after data reduction. The types of data reduction included doing no reducing, reduction based on individual GCOD correlation with activity, and reduction from minimum variance constraints over the GCOD population. The 3D-QSAR models were generated and evaluated by a scheme that combines a genetic algorithm (GA) optimization with partial least squares (PLS) regression. The 3D-QSAR models were evaluated by cross-validation using the leave-one-out technique. The cross-validated correlation coefficient, Q2, ranged from 0.27 to 0.86. The models are not from chance correlation because a scrambled data set was generated and evaluated (Q2 = 0.25-0.37). A composite 3D-QSAR model was constructed using the best models derived from GCODs of both 1 and 2 A grid cell size lattices. The 3D-QSAR models provide detailed 3D pharmacophore requirements in terms of atom types and corresponding locations needed for high TXA2 inhibition activity. Specific sites in space that should not be occupied by an active inhibitor are also specified. The GCOD measures for the compounds in the training set permit reference points regarding which pharmacophore sites can provide the largest boosts in inhibition activity relative to the existing analogues.

4D-QSAR Models of HOE/BAY-793 Analogues as HIV-1 Protease Inhibitors

“…In order to gain a better understanding of the behavior of the data fitted to the models, the cross-correlation matrix among the different GCODs in Models 1, 2, and 7 are given in Table B independent variables occur in both the same model and in different models, which is not surprising because many models (Table A, Supplementary Material) are highly correlated with one another. This behavior was also observed in a different data set [13].…”

“…It is a mathematical model of correlation statistically validated between the chemical structure and their activity profile. With the advent of molecular modeling, three-dimensional (3D) descriptors replaced the didactical physicochemical and bidimensional descriptors [12,13].…”

The 4D-QSAR Paradigm: Application to a Novel Set of Non-peptidic HIV Protease Inhibitors

“…4D-QSAR analysis has been extensively validated, and has been successfully applied to a number of different training sets: benzylpyrimidine inhibitors of dihydrofolate reductase [4], prostaglandin PGF 2 a antinidatory analogs [4], dipyridodiazepinone inhibitors of HIV-1 reverse transcriptase (RT) [4], interphenylene 7-oxabicycloheptane oxazole thromboxane A 2 receptor antagonists [15], glucose analogue inhibitors of glycogen phosphorylase [16], and Plasmodium falciparum dihydrofolate reductase inhibitors [17]. All these applications are examples of receptor-independent (RI) 4D-QSAR studies.…”

Um paradigma da química medicinal: a flexibilidade dos ligantes e receptores

“…A etapa de análise conformacional é o primeiro passo nesses estudos, buscando identificar aquela mais estável e, idealmente, a conformação bioativa 23,24 . O conhecimento de que tanto ligantes quanto seus receptores apresentam-se como estruturas flexíveis em escalas de tempo relativamente amplas 25 nos leva a supor que o próprio tempo nos quais ocorrem as mudanças conformacionais pode ser uma propriedade importante no estudo da interação de ligantes e seus receptores, tendo sido inclusive proposta como uma quarta dimensão em análises de QSAR 5 . Algumas abordagens alternativas foram também propostas, tais como modelos farmacofóricos dinâmicos 26 .…”

“…Este quadro vem se modificando com a ampliação do número de trabalhos utilizando outros tipos de metodologias capazes de produzirem modelos "dinâmicos" do complexo entre o ligante e sua proteí-na-alvo. Dentre estas podemos citar a QSAR-4D 5 , o "docking" flexível 6 , a dinâmica molecular [7][8][9] e o método de Monte Carlo 6,10 . Esta mudança no perfil dos modelos gerados por modelagem molecular está sendo possibilitada principalmente pela redução do custo computacional necessário à realização destes estudos, de forma que computadores pessoais já substituem eficientemente custosas estações de trabalho.…”