Multi-strategy investigation identifies amentoflavone as a novel D3R antagonist for Parkinson's disease treatment.

BACKGROUND: Parkinson's disease (PD) still lacks therapies that can simultaneously improve symptoms and slow neurodegenerative progression, making the screening of natural products with well‑defined mechanisms of action a promising strategy. The biflavonoid amentoflavone (AMF), which is present in plants including Ginkgo biloba L., has been shown to exert neuroprotective effects in PD models. However, its direct protein target remains unidentified. PURPOSE: This study aimed to investigate the anti‑parkinsonian activity of AMF and to elucidate its potential direct target and molecular mechanisms. METHODS: A multidisciplinary integrated strategy was employed. The protective effects of AMF on motor function and dopaminergic neurons were evaluated in MPTP-induced zebrafish and mouse models of PD. Potential target identification and signaling pathway analysis were performed by integrating multi-source target prediction, machine-learning-based reverse virtual screening, and transcriptomic analysis of the substantia nigra (SN) from AMF-treated mice. Compound-target interactions were validated using molecular docking, molecular dynamics simulations, and biophysical techniques including cellular thermal shift assay (CETSA), solvent-induced precipitation (SIP), and fluorescence correlation spectroscopy (FCS). Furthermore, immunofluorescence staining was conducted on mouse substantia nigra brain sections, signaling pathway validation was carried out at the cellular level, and siRNA interference was employed for in vitro target validation. RESULTS: AMF significantly improved motor deficits, alleviated neuroinflammation, and protected dopaminergic neurons in the substantia nigra in both zebrafish (100, 200, and 400 µM) and mouse models (15, 30, and 60 mg/kg, i.p.). Integrated machine‑learning screening and transcriptomic correlation analysis identified the dopamine D3 receptor (D3R) as a key target of AMF, with the model showing excellent classification performance (Voting Classifier AUC: 0.99). Molecular docking and dynamics simulations revealed that AMF stably binds to the orthosteric pocket of D3R. CETSA, SIP, and FCS experiments confirmed the stable binding and specific antagonistic activity of AMF toward D3R, with Asp110 and Tyr373 identified as critical binding residues. Furthermore, transcriptomic analysis combined with GO and KEGG enrichment suggested that beyond direct D3R antagonism, AMF may exert synergistic neuroprotective effects by targeting the PI3K/Akt signaling axis. Critically, D3R knockdown abolished the protective effects of AMF against MPP⁺-induced injury in vitro. CONCLUSIONS: This study systematically demonstrates for the first time that the natural biflavonoid AMF is a structurally novel D3R antagonist with anti‑parkinsonian potential. The work establishes and validates an integrated "computational prediction - experimental verification" framework for natural product target discovery, providing a new lead compound and a methodological paradigm for PD drug development.