Structure-guided Discovery of a Computationally Prioritized Candidate for MMP-13 Inhibition with Favorable Drug-likeness

Collagenases, a subgroup of matrix metalloproteinases (MMPs), contribute to pathological angiogenesis by degrading extracellular matrix components, and their dysregulation is closely linked to tumor invasion and metastasis. This study aimed to identify potential collagenase inhibitors through an integrated computational pipeline combining structure-based pharmacophore modeling, virtual screening, molecular docking, ADMET evaluation, density functional theory (DFT) calculations, and molecular dynamics (MD) simulations. A pharmacophore model derived from the MMP-13 crystal structure (PDB ID: 1YOU) was validated using receiver operating characteristic-based metrics and employed to screen multiple chemical databases, yielding 355 hits. A total of 119 records with docking scores ≤ −9.2 kcal/mol (docking score of co-crystallized ligand PFD) were identified, which yielded 112 unique compounds after duplicate removal. ADMET-based filtering identified seven drug-like and lead-like candidates. Cross-docking against related MMP isoforms (MMP-1, MMP-2, MMP-3, MMP-8, MMP-9 and MMP-14) revealed variable predicted off-target binding profiles among the screened compounds, while DFT-derived electronic descriptors provided complementary evidence supporting the selection of cmd6 for MD simulations. MD simulations indicated that the MMP-13–cmd6 complex maintained stable interactions during the 100-ns MD simulation under physiological ionic strength conditions, without inducing major structural fluctuations. Overall, this integrated computational workflow prioritized cmd6 as a lead candidate with a comparatively balanced docking profile across the evaluated MMP isoforms, supporting its progression to experimental validation.