A novel artificial intelligence protocol to investigate potential leads for Parkinson's disease

Area of Science:

• Pharmacology
• Computational Chemistry
• Neuroscience

Background:

• Parkinson's disease (PD) pathogenesis involves key proteins like NACHT, LRR and PYD domains-containing protein 3 (NLRP3), heat shock protein HSP 90-beta (HSP90AB1), caspase-1 (CASP1), and cellular tumor antigen p53 (TP53).
• Small molecule inhibitors targeting NLRP3 present a potential therapeutic strategy for PD.

Purpose of the Study:

• To identify novel traditional Chinese medicine (TCM) compounds with high affinity for NLRP3, HSP90AB1, CASP1, and TP53.
• To investigate potential drug-target interactions for PD treatment using computational methods.

Main Methods:

• Molecular docking was employed to screen the TCM Database@Taiwan for compounds targeting the identified PD-associated proteins.
• A drug-target interaction network was constructed to analyze potential therapeutic candidates.
• Artificial intelligence (AI) models, specifically 3D-Quantitative Structure-Activity Relationship (3D-QSAR), were developed to predict inhibitory activities (pIC50) against the target proteins.

Main Results:

• The screening identified three promising TCM compounds: 2007_22057 (indole derivative), 2007_22325 (valine anhydride), and 2007_15317 (indole derivative).
• These compounds demonstrated high affinities and potential as therapeutic agents targeting NLRP3, HSP90AB1, CASP1, and TP53.
• Molecular dynamics simulations further validated three candidate compounds for PD treatment.

Conclusions:

• The identified TCM compounds, particularly 2007_22057, 2007_22325, and 2007_15317, represent potential novel therapeutic candidates for Parkinson's disease.
• This study highlights the utility of integrating molecular docking, AI, and molecular dynamics for discovering new drug leads from traditional medicine databases.