Target Family Landscapes of Protein Kinases

Naumann and Matter used a set of 26 X-ray structures of eukaryotic protein kinases, which were classified into subfamilies with similar protein-ligand interactions in the ATP binding site. As can be seen in Fig. 3.13, which shows the GPGA score plot, PC 1 separates CDK and MAP/receptor kinases on the left from the family of PKA kinases. The CDK family is represented by two distinct clusters in the target family landscape, formed by two different ATP binding site conformations. They correspond to the activated and inactivated kinase conformations 

Interpretation of the structural features responsible for this landscape reveals that the main differences along PC 1 (i.e. between PKA and MAP/CDK kinases) are found in the purine and hinge binding region, whereas the discrimination in PC 2 is mainly driven by structural differences in the phosphate binding area. 

To illustrate the use of the target family landscape for understanding kinase selectivity profiles, Naumann and Matter used a series of 86 2,6,9-substituted purines. These selective CDK inhibitors bind to the kinase ATP binding site [46]. A detailed comparison with experimental selectivity profiles showed good agreement with the chemometric analysis. 

An important observation by Naumann and Matter is the identification of additional opportunities to achieve selective interactions in the kinases phosphate binding region, whereas most work so far has focussed on the purine binding regions. 

---