Ligand Based Pharmacophore Modeling and Virtual Screening for Novel Antidiabetics Targeting PPAR-gamma

Abstract

A modern sedentary lifestyle with a more calorigenic fast-food diet increasing the prevalence of metabolic syndrome in middle and high-income countries. Peroxisome proliferator-activated receptors (PPARs) are a group of the nuclear receptor, which regulates the metabolic process in physiological systems via influencing gene expressions of cell proliferation, glucose, lipid metabolism, and inflammation. Later one PPAR-γ agonist is a well-established class of pharmacological agents for diabetic control with some promising molecule in the clinical stages. Herein, we have chosen a hybrid indole and azaindole class for developing an effective pharmacophore model. A series of compounds with indole carboxylic acid and hydroxyazaindole core along with their tested biological activity were selected for generating a valid pharmacophore model using Hip-Hop and HypoGen algorithm of Discovery Studio v3.1. A total of 38 numbers of ligand were considered for pharmacophore generation and mapping including test set and training set. Depending upon proper calculative measures the best-validated hypothesis with two hydrophobic, one hydrogen bond acceptor, and one ring aromatic features are set forth for further shortlisting of compounds. A similarity search tool in PubChem structure database with a 70% similarity of best active compounds yields more than four lakhs compounds. The screened drug-like compounds were further shortlisted using 'rules of five' and TOPKAT module. The best-validated HypoGen pharmacophore was utilized for further screening to get the best structures for future in-silico consideration and identifying potential hits for effective diabetes drug discovery research.