Lessons learned from the optimization of natural product based kinase inhibitors through molecular calculations

The family of protein kinases includes some of the most promising targets aimed by the search for chemotherapeutic drugs[1]. Kinase deregulation is frequent not only in malignancies but also in inflammative and neurodegenerative diseases. Several highly active molecular scaffolds with kinase inhibiting potential have been evolved from natural product leads and in most cases optimization of the lead was assisted by molecular simulations. Alterations on the Indirubin molecule are a representative example of such optimization[2–3]. Through various computational chemistry methods in different levels of theory from molecular mechanics and free energy perturbation theory to high level ab initio calculations an attempt was made to explore and gain insight in the factors that govern affinity of a number of indirubin analogues ranging in potency from inactives to low nanomolar level. At the same time selectivity being an aspect of kinase inhibition equally important as potency was also addressed through a combination of computational methods and bioinformatics and results on a variety of serine threonine kinases including CDKs, the auroras and GSK3b homologs from different organisms imply that the molecular point of view offered by computational methods can be of great assistance in rational drug design

References: 1. Cohen, P. e al (2002) Nat Rev Drug Discovery 1:309–315

2. Polychronopoulos, P. et al (2004)J Med Chem 47: 935–946

3. Myrianhopoulos, V. et al (2007)J Med Chem 50: 4027–4037