An enriched structural kinase database to enable kinome-wide structure-based analyses and drug discovery

Area of Science:

• Biochemistry
• Structural Biology
• Bioinformatics

Background:

• Kinases are crucial enzymes involved in cell signaling.
• Understanding kinase structure is vital for drug discovery.
• Existing structural data is fragmented and lacks a unified annotation system.

Purpose of the Study:

• To develop a comprehensive structural database for human kinases.
• To establish a common numbering scheme for kinase structures.
• To facilitate comparative analysis and identification of conserved features.

Main Methods:

• Compilation of all human kinase domain structures from the Protein Data Bank.
• Development and application of a common residue numbering scheme.
• Automated annotation of conserved residues, motifs, and conformational states (DFG-loop, Helix C).
• Analysis of residue conservation within the ATP binding site using sequence alignments.

Main Results:

• The Kinase Knowledge Base (KKB) now covers all deposited human kinase domain structures.
• A common numbering scheme allows for efficient cross-comparison and querying.
• Automated annotation identified conserved residues and motifs, aiding conformational classification.
• A conserved hydrogen bond between the hinge region backbone and glycine in the specificity surface was identified.
• Over 90% of kinases exhibit previously undescribed stabilizing interactions in the hinge region.

Conclusions:

• The KKB provides a valuable resource for kinase research.
• The common numbering scheme and automated annotations streamline structural analysis.
• Novel conserved interactions in the ATP binding site and hinge region were discovered.
• These findings have implications for understanding kinase function and developing targeted therapeutics.