Date of Award


Document Type

Campus Access Dissertation


College of Pharmacy


Pharmaceutical Science

First Advisor

Campbell McInnes


CDK2/cyclin A and CDK4/cyclin D1 are validated targets for cancer drug discovery with ATP competitive compounds being the most common inhibitors. However, issues arising from the selectivity of current inhibitors have prevented clinical validation of these important targets. To generate cell cycle specific CDK inhibitors with high selectivity, an alternative way of designing CDK inhibitors is afforded by targeting the cyclin binding groove (CBG). Peptidic inhibitors have been developed for CBGs which are shallow areas on the surface of cyclins responsible for recruiting substrates and natural inhibitors of CDKs and serve as the basis for the development of new cancer therapeutics. A structural comparison of the CBGs of cyclin D and cyclin A revealed the features of cyclin D including a less acidic arginine binding site, and a shallower primary hydrophobic pocket but which is further extended. A series of surrogates for the C-terminal phenylalanine in p21 and p107 derived peptides were used to probe their binding to the specific primary hydrophobic pockets of CDK2/A2 and CDK4/D1 due to a Leu/Val exchange in CBGs. A 3-thienylalanine replacement in SAKRRLFG (p107) context was found to be more potent than the parent peptide in the binding to CDK2/cyclin A2 while a cyclohexylalanine replacement of Phe in the HAKRRLIF (p21) sequence shows good potency against CDK4/cyclin D1. Further to this, a Fluorescence Polarization (FP) assay was developed for both CDK2/cyclin A2 and CDK4/cyclin D1 to study the binding of inhibitors. The REPLACE methodology was subsequently applied with the goal of generating a nonpeptidic inhibitor after combination of fragment capping groups and requirements for binding of FLIPs to CBGs of cyclin A and cyclin D were examined through structure-activity relationships of generated N-terminal and C-terminal capping groups. The conclusions of this study are as follows: 1) the secondary hydrophobic pocket of cyclin A and D contribute to the selectivity of inhibitors. 2) Basic groups presented appropriately make strong ionic interactions with acidic residues of cyclin A and D, and are therefore critical for potency. 3) Appropriate halogen substitution of a bisarylether C-capping group leads to increased binding. In addition to the above, a series of peptide analogs were synthesized with alternatives for Arg, Leu and Phe to expand our SAR study and provide more insightful ideas for inhibitor development. The potency increase observed for these analogs suggest their use as tools for cellular studies. Overall, all the SAR studies described above direct the optimization of inhibitors for CDK4/cyclin D and CDK2/cyclin A as well as providing essential information for the development of highly selective potential therapeutics with antitumor effects.


© 2012, Shu Liu