Computational Drug Design and SAR Analysis of Newly Developed Heterocyclic Compounds as Prospective Antidiabetic Molecules

Abstract

Diabetes mellitus (DM) represents one of the fastest-growing metabolic disorders globally, with type 2 diabetes mellitus (T2DM) accounting for approximately 95% of all cases. The present study investigates the computational drug design approach and structure-activity relationship (SAR) analysis of newly developed heterocyclic compounds as prospective antidiabetic molecules targeting key enzymes, namely α-glucosidase, α-amylase, and dipeptidyl peptidase-IV (DPP-IV). The primary objectives were to evaluate the binding affinity, drug-likeness, and pharmacokinetic profiles of selected heterocyclic scaffolds including thiazolidinediones, oxadiazoles, benzimidazoles, and thiadiazoles through molecular docking and ADMET prediction. The methodology employed in silico approaches comprising molecular docking using AutoDock Vina, SwissADME for drug-likeness, and ProTox-II for toxicity prediction. It was hypothesized that heterocyclic compounds bearing electron-withdrawing substituents would demonstrate superior binding affinity against antidiabetic target enzymes. Results revealed that compounds with para-substituted halogen and nitro groups exhibited binding energies ranging from −7.8 to −10.6 kcal/mol, surpassing the standard drug acarbose (−6.7 kcal/mol). SAR analysis confirmed that the nature and position of substituents critically influence antidiabetic potency. In conclusion, the study validates computational drug design as an effective strategy for identifying potent heterocyclic antidiabetic leads.