- Joint Appointment in Oncology
Techniques for biomolecular modeling of protein assembly and function
Dr. Gray develops techniques for biomolecular modeling of protein assembly and function, including protein-protein docking, allostery, and protein-surface interactions. Dr. Gray wrote the RosettaDock algorithm, which is a multiscale routine that can rapidly search docking conformational space and accurately identify near-native conformations. Current work is improving recognition for proteins with backbone flexibility and tailoring the algorithm for application to therapeutic antibodies, including those applicable in cancer studies. Functional application to allosteric proteins and bionanoengineering application to protein-surface interactions is are being developed.
Therapeutic antibodies provide relief from various human diseases and provide a strategy for several new and promising treatments for cancer. One such strategy targets the epidermal growth factor receptor (EGFR), since EGFR signaling is implicated in a wide variety of cancers (breast, ovarian, head and neck, lung, pancreatic, colorectal),. In particular, monoclonal antibody (mAb) 806 binds EGFR, but unlike similar anti-EGFR drugs, it binds cancer cells exclusively, thus avoiding side effects such as hair and skin loss. However, the mAb 806-EGFR structure is unknown, limiting both our understanding of the drug mechanism and our ability to rationally improve the drug. Our objectives are to predict the structure of the complex of mAb 806 bound to EGFR and to improve methodologies for docking of therapeutic antibodies in general such that predicted structures are of sufficient quality for rational drug design and engineering. We are appling RosettaDock, our multi-scale protein-protein docking algorithm, and we are combining it with novel homology modeling and free energy calculations to create accurate structural models. This research will enable rapid and inexpensive means to understand the molecular basis of new cancer therapeutics, and it will enable the rational design and improvement of mAb 806 and the entire class of antibody drugs.
In a new study relevant to cancer research, we are studying the molecular motions and underlying mechanisms of allosteric proteins. Allosteric proteins have two distinct biochemical states (e.g. active and inactive) dependent upon the presence of an effector molecule such as GTP or a hormone. Transcription factors, signaling proteins, and regulatory proteins are allosteric, and cancer cells often hijack these proteins to set them permanently in one particular state to enable cell proliferation. For example, the signaling protein /ras/ is permanently switched ‘on’ in 30 percent of all solid tumors. We have collected a large set of structures of allosteric proteins and we are mining this set to describe the structural mechanism of switching. We are also using network theory to determine the nature of the intramolecular signal transduction. Finally, we are using Rosetta to simulate the conformational transitions experienced by these proteins in response to a signal. These simulations and structural studies will help us to identify particular target sites for drug design to restore the normal functioning of these proteins. This work will ultimately speed rational design of small molecule cancer drugs.
For additional information on Dr. Gray's research, see http://graylab.jhu.edu