Computational Screening of Persea americana Peel and Seed Phytochemicals as Potential Multi-Target Antidiabetic Agents

Abstract

Diabetes mellitus (DM) is a global health burden, and current oral antidiabetic drugs are associated with undesirable side effects. Avocado (Persea americana) peel and seed are underutilized by-products rich in bioactive phytochemicals. This study aimed to evaluate the antidiabetic potential of ten gas chromatography-mass spectrometry (GC-MS)-derived compounds (five from peel and five from seed) against five protein targets, α-glucosidase, α-amylase, dipeptidyl peptidase-4 (DPP-4), protein tyrosine phosphatase 1B (PTP1B), and AMP-activated protein kinase (AMPK) using PASS Online bioactivity prediction and molecular docking. PASS Online predicted high antidiabetic probabilities (probability active > 0.7) for epicatechin and chrysoeriol-4′-O-pentoside-7-O-rutinoside. Molecular docking with AutoDock Vina 1.2.0 showed that epicatechin bound strongest to α-glucosidase (−9.2 kcal/mol) and α-amylase (−8.7 kcal/mol), comparable to acarbose. Chrysoeriol-4′-O-pentoside-7-O-rutinoside exhibited high affinity for α-glucosidase (−8.8 kcal/mol), DPP-4 (−8.4 kcal/mol), PTP1B (−8.1 kcal/mol), and AMPK (−8.0 kcal/mol). ADMET predictions indicated that epicatechin has favorable drug-likeness properties (no Lipinski violations, high GI absorption). At the same time, chrysoeriol-4′-O-pentoside-7-O-rutinoside may face bioavailability challenges due to its large, glycosylated structure (molecular weight 742.68 g/mol). These results suggest that avocado seed flavonoids, particularly epicatechin, are promising multi-target antidiabetic candidates for nutraceutical development, though experimental validation is required to confirm these computational predictions.