Structural mechanisms of assembling, activating and inhibiting CDK4

The growth and division of cells is strictly controlled at the molecular level by a series of enzymes that include the cyclin dependent protein kinases (CDKs). These enzymes are switched on and off in an orderly sequence, to ensure that cell division starts and stops at the required time. Though closely related in sequence, biological studies of the different members of the CDK family have revealed that each has unique properties. In particular, their activities can be regulated both by association with different families of regulatory proteins and by modification of the protein sequence by addition of a phosphate group to specific amino acid side chains. Our work, using the technique of X-ray crystallography, allows us to see the structures of these molecules at atomic resolution and so to learn how they differ from each other. The first aim of our work is to understand how members of the family of cyclin-dependent kinase inhibitors (CKIs) can bind to one member of the CDK family called CDK4 and either promote its assembly with its regulator cyclin D (resulting in an active enzyme complex) or inhibit its activity. CDK4 is also distinguished from other members of the CDK family in that it has a very limited number of substrates.
The second aim of our work is to understand why this is the case. Aberrant CDK activity has been linked to cancer, neurological diseases, and rheumatoid arthritis. In particular, elevated levels of CDK4 activity, resulting from mutations in the genes for the proteins that regulate its activity, is commonly found in a number of cancers and is associated with a poor prognosis. In recent years, understanding a particular defect that leads to disease has led to exciting new medicines directed towards a particular target (e.g. the drugs Gleevec, Iressa and Herceptin for cancer treatment). A number of CDK-selective inhibitors are in clinical trials for the treatment of cancer. These agents all act by binding to the CDK active site to block CDK activity. Compounds that block other interactions made by CDK/cyclin complexes represent an alternative target for CDK-directed therapies. The work described in this project, to elaborate the interactions that mediate the binding of substrates to CDKs and CDK-regulatory proteins, will aid the further development of such compounds and may also reveal additional targets for inhibitor development.

Cyclin-dependent protein kinases (CDKs) are implicated in many cellular regulatory processes, notably in the control of cell cycle progression and transcription. A paradigm for CDK activation by cyclin binding and activation loop phosphorylation has been elaborated through structural studies of various CDK2/cyclinA complexes. Our recent determination of the structure of CDK4/cyclinD3 has revealed that CDK4 does not conform to this model: cyclin binding does not induce an active CDK4 conformation. CDK4 is also distinguished from CDK2 by its interactions with members of the Cip/Kip family. Cip/Kips were identified as cyclin-dependent kinase inhibitors (CKIs), but were subsequently shown to act as CDK4/cyclinD assembly factors and controversially the resulting CKI/CDK4/cyclinD complexes were found to retain catalytic activity. Recent studies have shown that CKIs can adopt alternative CDK4/cyclinD binding modes that are dependent on CKI phosphorylation status. The first aim of this project is to elaborate the structural basis for the twin roles of Cip/Kip family members as assembly factors and as inhibitors for CDK4/cyclinD complexes. In particular, we will carry out biochemical, biophysical and crystallographic studies to investigate whether these different roles result from different modes of Cip/Kip binding. The second aim of this project is to determine the structural basis for the exquisite substrate selection displayed by CDK4/cyclinD complexes, and whether this relates to the structural peculiarities of CDK4 activation. In particular, we will map rates of phosphorylation of different pRb sites to identify those that are most susceptible to phosphorylation. We will subsequently use fragments of pRb that span from these sites to the R/KXL motif in co-crystallization experiments to determine the structure of a Michaelis complex of CDK4/cyclinD. This structure will reveal if and how an appropriate substrate imposes an active conformation on the CDK4/cyclinD complex. The third aim of the project is to determine whether the structural peculiarities of a CDK4/cyclinD complex predispose it to the binding of inhibitors that select an inactive conformation. Specifically, we will carry out co-crystallization and inhibitor-soaking experiments to identify whether known CDK4 ATP-competitive inhibitors exploit the preferred inactive conformation of CDK4 that we have observed in CDK4/cyclinD complexes. We will also carry out fragment-based screening experiments to explore whether fragment-binding pockets that are characteristic of the inactive ?C-helix out? conformation can be exploited. In addition, we will evaluate the effect on CDK4/cyclinD of peptidic and non-peptidic substrate recruitment inhibitors that we have identified.