Lead Validation and SAR Development via Chemical Similarity Searching; Application to Compounds Targeting the pY+3 Site of the SH2 Domain of p56lck

• Published in: Journal of chemical information and modeling 2005-11, Vol.45 (6), p.1759-1766
• Main Authors: Macias, Alba T, Mia, Md. Younus, Xia, Guanjun, Hayashi, Jun, MacKerell, Alexander D
• Format: Article
• Language: English
• Subjects: Amino acids; Chemical compounds; Computer Simulation; Drug Evaluation, Preclinical; Enzyme Inhibitors - pharmacology; Hydrogen Bonding; Immunoblotting; Immunoenzyme Techniques; Lead; Lymphocyte Culture Test, Mixed; Lymphocyte Specific Protein Tyrosine Kinase p56(lck) - antagonists & inhibitors; Lymphocyte Specific Protein Tyrosine Kinase p56(lck) - chemistry; Models, Chemical; Models, Molecular; Protein Conformation; Proteins; src Homology Domains - drug effects; Structure-Activity Relationship
• ISSN: 1549-9596; 1549-960X
• DOI: 10.1021/ci050225z

Compound selection based on chemical similarity has been used to validate active “parent” compounds identified via database searching as viable lead compounds and to obtain initial structure−activity relationships for those leads. Twelve parent compounds that have inhibitory activity against the SH2 domain of the p56 T-cell tyrosine kinase (Lck) are the focus of this study. Lck is involved in the T-cell mediated immune response, and inhibitors of Lck protein−protein interactions could potentially be used to develop novel immunosuppressants. Similarity searches for each parent compound were performed using 2D structural fingerprints on a database containing 1 300 000 commercially available compounds. The inhibitory activity of the selected compounds was assessed using enzyme immunoassay (EIA). In general, the most active parent compounds yield the most high activity similar compounds; however, in two cases low activity parent compounds (i.e. inhibitory activity < 25% at 100 μM) yielded multiple similar compounds with activities > 60%. Such compounds may, therefore, be considered as viable lead compounds for optimization. Structure−activity relationships were explored by examining both ligand structures and their computed bound conformations to the protein. Functional groups common to the active compounds as well as key amino acid residues that form hydrogen bonds with the active compounds were identified. This information will act as the basis for the rational optimization of the lead compounds.