Identification and Visualization of Kinase-Specific Subpockets.

Area of Science:

• Computational drug discovery
• Structural bioinformatics
• Medicinal chemistry

Background:

• Selective compound design is vital for minimizing off-target effects and validating drug targets.
• Kinases are a therapeutically important protein family with conserved structures, posing challenges for selective inhibitor development.
• Existing methods often require extensive profiling data, limiting their application in early-stage design.

Purpose of the Study:

• To develop a novel, structure-based computational approach for identifying specificity-determining subpockets between closely related kinases.
• To provide an intuitive visualization tool for guiding the design of selective kinase inhibitors.
• To demonstrate the approach's utility in retrospective case studies without requiring prior profiling data.

Main Methods:

• Utilizing multiple X-ray structures to account for protein flexibility.
• Calculating and fusing binding pockets of target and off-target kinases.
• Applying fusion rules to identify shape differences and target-specific areas.
• Integrating with in silico alanine mutation studies to pinpoint selectivity-determining residues.

Main Results:

• The approach successfully identified specificity-determining subpockets in retrospective analyses of kinase pairs (p38α vs Erk2, PAK1 vs PAK4, ITK vs AurA, BRAF vs VEGFR2).
• It provides user-friendly visualizations highlighting regions for selective compound exploration.
• The method requires no training data and can visualize numerous protein structures simultaneously.

Conclusions:

• The novel structure-based approach effectively identifies kinase-specific subpockets for selective inhibitor design.
• It offers an intuitive and data-efficient method applicable to various kinase targets and potentially other protein families.
• This tool aids in reducing side effects and advancing target validation through improved compound selectivity.