Biocomputational Prioritization of Cucurbitacins as Candidate Inhibitors of Human Pancreatic Lipase

Aim/Background Obesity is a multifactorial metabolic disorder associated with substantial morbidity and mortality. Human pancreatic lipase is a validated therapeutic target because its inhibition can reduce the digestion and intestinal absorption of dietary triglycerides. This study aimed to computationally evaluate cucurbitacins as candidate pancreatic lipase inhibitors. Materials and Methods A library of 71 cucurbitacins was screened against human pancreatic lipase using AutoDock Vina. Orlistat, a clinically approved pancreatic lipase inhibitor, was included as the positive-control compound. The three highest-ranked cucurbitacins and orlistat were subjected to 100-ns molecular dynamics simulations using GROMACS 2022. Physicochemical, pharmacokinetic, and toxicity properties of the selected cucurbitacins were predicted using the absorption, distribution, metabolism, excretion, and toxicity–artificial intelligence (ADMET-AI) platform. Results Cucurbitacins F, H, and K produced docking scores of −8.8, −8.7, and −8.4 kcal/mol, respectively, compared with −6.7 kcal/mol for orlistat. The selected compounds occupied the pancreatic lipase catalytic pocket and maintained persistent protein-ligand interactions during the 100-ns simulations. Cucurbitacin K showed the lowest mean protein-backbone root mean square deviation (RMSD), whereas cucurbitacin F showed the lowest ligand RMSD and the highest average number of hydrogen bonds. Numerical ADMET predictions supported further prioritization of the compounds while identifying physicochemical and metabolic properties requiring experimental evaluation. Conclusion Cucurbitacins F, H, and K were computationally prioritized as candidate binders of human pancreatic lipase. These findings provide a basis for subsequent binding free-energy calculations, pancreatic lipase inhibition assays, and in vivo validation, but do not establish confirmed inhibitory or anti-obesity efficacy.