Network Pharmacological Analysis of Lentinan in Regulating Intervertebral Disc Degeneration: Combined with Machine Learning-Based Screening and Molecular Dynamics Validation

Abstract

Lentinan (LNT) is a polysaccharide with antioxidant and anti-inflammatory properties; however, its link to intervertebral disc degeneration (IVDD) remains unclear. Here, we integrated network pharmacology, machine learning, molecular docking, and molecular dynamics (MD) simulations to investigate the LNT–IVDD relationship. Using dataset GSE176205, 16,361 potential IVDD-related genes were identified, and 97 LNT-associated targets were collected from four databases, yielding 46 intersecting targets. GO and KEGG analyses revealed enrichment in pathways including calcium, PI3K-Akt, Notch, and Rap1 signaling. Protein–protein interaction network analysis combined with three ML methods prioritized two core targets: signal transducer and activator of transcription 3 (STAT3) and fibroblast growth factor 2 (FGF2). The diagnostic relevance of these genes was subsequently validated using receiver operating characteristic curves across three independent IVDD datasets. Molecular docking showed strong binding (energy < −5 kcal/mol) between LNT and these targets, further supported by stable complex formation in MD simulations. Immune-infiltration analysis delineated the IVDD microenvironment, and virtual knockout based on single-cell transcriptomics systematically assessed how perturbing key genes remodels cell states and signaling networks. Together, this work provides novel insights into the molecular mechanisms underlying LNT’s therapeutic effects in IVDD and establishes a foundation for its translational study.