Ligand-guided optimization of CXCR4 homology models for virtual screening using a multiple chemotype approach

Area of Science:

• Biochemistry
• Molecular Biology
• Pharmacology

Background:

• CXCR4, a G-protein coupled receptor, is crucial in HIV infection, cancer metastasis, and immune cell trafficking.
• The lack of an X-ray crystal structure for CXCR4 necessitates theoretical modeling for structure-function analysis.
• Understanding CXCR4 is vital for developing novel therapeutic antagonists.

Purpose of the Study:

• To develop optimized ligand-receptor models for CXCR4 using integrated experimental and computational approaches.
• To elucidate the molecular determinants governing small molecule binding and antagonism of CXCR4.
• To enhance the identification of effective CXCR4 antagonists through improved virtual screening.

Main Methods:

• Employed a ligand-guided homology modeling approach to refine the CXCR4 binding pocket.
• Integrated experimental data with molecular modeling techniques.
• Utilized multiple test sets of small compounds from single chemotypes for model refinement.

Main Results:

• Developed optimized CXCR4 ligand-receptor models capable of discriminating between known antagonists and decoys.
• The ligand-guided modeling approach effectively reshaped the CXCR4 binding pocket.
• Refined models demonstrated superior early enrichment performance in virtual screening.

Conclusions:

• The developed models serve as a valuable tool for structure-based drug design targeting CXCR4.
• This approach facilitates efficient virtual ligand screening for novel CXCR4 antagonists.
• The findings support the development of new therapeutics for CXCR4-related diseases.