Uncovering PPAR-γ agonists: An integrated computational approach driven by machine learning.

Area of Science:

• Biochemistry
• Computational Chemistry
• Pharmacology

Background:

• Peroxisome proliferator-activated receptor gamma (PPAR-γ) is crucial for metabolic regulation and insulin sensitivity in diabetes.
• Existing PPAR-γ agonists like Thiazolidinediones have significant side effects.
• Selective PPAR-γ modulators offer a potential alternative with improved safety profiles.

Purpose of the Study:

• To develop an integrated computational strategy using machine learning and in silico drug design to discover novel selective PPAR-γ modulators.
• To identify compounds with high affinity and favorable interactions within the PPAR-γ ligand-binding site.
• To assess the conformational stability of identified compounds in complex with PPAR-γ.

Main Methods:

• Constructed a machine learning classification model trained on chemical and physicochemical descriptors of known PPAR-γ modulators.
• Performed virtual screening of 31,750 compounds using the machine learning model.
• Utilized molecular docking to evaluate binding affinity and interactions of selected compounds with PPAR-γ.
• Analyzed molecular dynamics simulations to assess the stability and conformational changes of the PPAR-γ-ligand complexes.

Main Results:

• Identified 68 potential PPAR-γ modulators from virtual screening, with four compounds selected for further analysis.
• Docking scores for the top compounds ranged from -8.0 to -9.1 kcal/mol.
• Key hydrogen bond interactions were observed with conserved residues (His323, Leu330, Phe363, His449, Tyr473) in the PPAR-γ binding site.
• Molecular dynamics simulations indicated moderate conformational changes in the orthosteric site, with stability indices suggesting favorable interactions.

Conclusions:

• The computational strategy successfully identified novel potential selective PPAR-γ modulators.
• CHEMBL-3185642 and CHEMBL-3554847 demonstrated outstanding results, highlighting stable conformations within the PPAR-γ orthosteric site.
• These identified compounds represent promising leads for developing safer and more effective diabetes therapies targeting PPAR-γ.