No abstract

Candidiasis is recognized worldwide as an opportunistic infection. The severities of the infection increase in immunosuppression conditions with possible occurrence of visceral mycoses and sometimes are widespread and systemic. Increased resistance in strains of Candida albicans is a major obstacle to antifungal therapy. The aim of this study was to correlate the chemical structure of compounds with experimental data from biological activity anti-Candida albicans. We performed classical QSAR for a series of twenty derivatives of ketone -β unsaturated against resistant strains of Candida albicans. Ninety-four descriptors were calculated and multiparameter model was obtained through Partial Least Squares (PLS) method. The results showed that thermodynamic, dimensional and steric parameters are important in elucidating of action mechanism compounds. Four descriptors (molar refractivity, ionization potential, molecular length and Verloop B4) were selected and good model (n= 20; R 2 = 0.776; SEC = 0.229; F(3,16) = 14.172; Q 2 LOO = 0.609; SEV = 0.295; Q 2 pred = 0.709; SEP = 0.091; k = 0.709; k' = 1.00; ‫|‬R 2 0 -R' 2 ‫|0‬ = 0.0009) was built with three latent variables describing 96.14% of the original information. Leave-N-out cross validation and Y-randomization analysis were performed in order to confirm the robustness of the model. The proposed model may provide a better understanding of the anti-Candida albicans activity of chalcones and can be used as guidance for proposition of new chemopreventive agents.

Cinnamic acid analogs are natural phenolic compounds that have a wide range of biological and therapeutic activities. The present work aimed to predict, through in silico methodologies, the oral bioavailability and pharmacokinetic and toxicological analyzes for four cinnamic acid analogues (caffeic acid, ferulic acid, p-coumaric acid and synaptic acid). The study revealed that the analogues have good oral bioavailability, favorable pharmacokinetic and toxicological parameters. The Virtual Screening performed to predict oral bioavailability indicated that all analogues do not violate Lipinski's Rule. The in silico ADME study of pharmacokinetic parameters showed that all derivatives have high intestinal absorption, are permeable by Caco-2 cells, do not cross the blood-brain barrier, do not inhibit P-glycoprotein. There will be no inhibition of the cytochrome P450 complex isoenzymes (CYP450). The in silico Toxicological study revealed that the analogues do not have toxicity by the AMES Test, are not carcinogenic and do not present acute oral toxicity.

Flavone analogs are natural compounds of the flavonoid class that have a wide range of biological activities. The present study aimed to predict, with the aid of in silico methodologies, the oral bioavailability and pharmacokinetic and toxicological analyzes for three flavone analogues (apigenin, chrysin and luteonlin). The study revealed that the analogues have good oral availability, favorable pharmacokinetic and toxicological parameters. The Virtual Screening performed to predict oral bioavailability revealed that all analogues did not violate Lipinski's Rule. The in silico pharmacokinetic study revealed that all analogues have high intestinal absorption, do not cross the blood-brain barrier, are permeable by Caco-2 cells and do not inhibit P-glycoprotein. The in silico ADME study showed that all analogues inhibit the enzymes of the cytochrome P450 complex (CYP4501A2, CYP4502C9, CYP4502C19, CYP4503A4) and not only the CYP4502D6 enzyme. The in silico Toxicology study indicated that the analogues do not show toxicity by the AMES Test and are not carcinogenic. Apigenin and chrysin have low toxicity, while luteolin has moderate toxicity.

Os herbicidas correspondem à classe dos agrotóxicos ou defensivos agrícolas utilizados para controlar e matar plantas daninhas. Os análogos de PSA correspondem à categoria de herbicidas inibidores da enzima acetolactato sintase (ALS), que interferem na biossíntese dos aminoácidos ramificados (leucina, isoleucina e valina), interrompendo o mecanismo de síntese proteica nas células vegetais. O presente trabalho tem como objetivo realizar o estudo QSAR clássico com intuito de obter um modelo multiparamétrico para predizer as respectivas atividades biológicas experimentais e que por intermédio dos descritores moleculares selecionados através da Análise Quimiométrica, proporcionar uma melhor compreensão das interações intermoleculares entre o ligante (herbicida) e o receptor biológico (enzima: ALS). Observou-se que todos os compostos não violaram a regra de Lipinski e que os descritores moleculares: número de Sítios Aceptores de Ligações de Hidrogênio (SALH), Log do coeficiente de partição (Log P), Área Superficial Polar (ASP), contribuição da forma iônica (aniônica = desprotonada), como também a ocorrência de efeitos eletrônicos do tipo mesomérico e indutivo podem estar correlacionados com a atividade biológica experimental. Importante ressaltar a possibilidade de interação eletrostática, interações intermoleculares do tipo ligações de hidrogênio e dipolo induzido do ligante com os resíduos de aminoácidos no sulco ativo da enzima ALS. O estudo também revelou a importância dos descritores moleculares: Log KOW (hidrofóbico/hidrofílico), HBN (termodinâmico), Gap (eletrônico) e Volume Molecular (estereoquímico). O modelo linear multidimensional proposto apresenta bom grau de ajuste linear, significância estatística e grau de previsibilidade considerável com relação à atividade biológica dos dados experimentais.