The development of a CDK2-docking site peptide that inhibits p53 and sensitizes cells to death

Cyclin-dependent protein kinases play important roles in cell cycle progression and are attractive targets for the design of anti-proliferative drugs. Two distinct synthetic CDK1/2 inhibitors, Roscovitine and NU2058, are pharmacologically distinct in their ability to modify p53-dependent transcription and perturb cell cycle progression. Although such active-site CDK1/2 inhibitors comprise the most standard type of enzyme inhibitor, many protein kinases are proving to harbour high affinity docking sites that may provide a potentially novel interface for the design of kinase-inhibitors. We examined whether CDK2 has a docking site for its oligomeric substrate p53, whether small-peptide leads can be developed that inhibit CDK2 function, and whether such peptide-inhibitors are pharmacologically distinct from Roscovitine or NU2058. A docking site for CDK2 was identified in the tetramerization domain of p53 at a site that is distinct from the phospho-acceptor site. Peptides derived from the tetramerization domain of p53 block CDK2 phosphorylation and identification of critical CDK2 contacts in the tetramerization domain of p53 suggest that kinase docking does not require tetramerization of the substrate. Transient transfection assays were developed to show that the GFP-CDK2 docking site fusion protein (GFP-CIP) attenuates p53 activity in vivo and suppresses p21^WAF1 induction which is similar to NU2058 but distinct from Roscovitine. A stable cell line with an inducible GFP-CIP gene attenuates p53 activity and induces significant cell death in a drug-resistant melanoma cell line, sensitizes cells to death induced by Doxorubicin, and suppresses cell growth in a colony formation assay. These data indicate that CDK2, in addition to cyclin A, can have a high affinity docking site for a substrate and highlights the possibility that CDK2 docking sites may represent effective targets for inhibitor design.