Metabolite profiling and molecular dynamics-based in silico evaluation of methanolic leaf extracts of Stevia rebaudiana Bertoni against α-amylase and DPP-IV

Abstract

Stevia rebaudiana Bertoni is a natural sweetener with reported antidiabetic properties; however, the molecular basis underlying the interactions of its major phytochemicals with key glycaemic-regulating enzymes remains insufficiently understood. The present study employed an integrated in silico workflow to evaluate the antidiabetic potential of selected steviol-derived phytochemicals against human pancreatic α-amylase and dipeptidyl peptidase-IV (DPP-IV). Methanolic leaf extracts were profiled by GC–MS, and steviol, together with the principal steviol glycosides (stevioside, rebaudioside A, and rebaudioside D), was selected for computational analyses. Pharmacokinetic properties were predicted using SwissADME, followed by molecular docking and 100 ns molecular dynamics simulations incorporating principal component analysis (PCA), free-energy landscape (FEL), dynamic cross-correlation analysis (DCCM), and hydrogen-bond analysis, together with MM-PBSA binding free energy and per-residue energy decomposition. Among the evaluated phytochemicals, steviol exhibited the most favourable predicted docking profile against α-amylase (−8.46 kcal/mol), whereas rebaudioside A showed the most favourable predicted docking profile against DPP-IV (−8.34 kcal/mol). Molecular dynamics and complementary structural analyses indicated stable conformational behaviour and provided additional insights into the dynamic and energetic characteristics of the selected protein–ligand complexes. ADME analysis further demonstrated distinct pharmacokinetic characteristics among the evaluated phytochemicals, with steviol exhibiting comparatively more favourable predicted oral drug-likeness, whereas the steviol glycosides displayed physicochemical properties that may favour gut-localised α-amylase inhibition. These findings provide computational evidence supporting the potential interactions of steviol-derived phytochemicals with α-amylase and DPP-IV. However, the results are based entirely on computational predictions and require experimental validation through in vitro enzyme inhibition assays, pharmacokinetic studies, and in vivo investigations before their therapeutic potential can be established.