We use biochemical and structural methods (primarily X-ray crystallography) to study the structure and regulation of tyrosine kinases that are important in cancer. We are especially interested in understanding how cancer-causing mutations lead to loss of normal kinase regulation and in using structural approaches to develop new anticancer drugs. Active areas of investigation include: (1) lung cancer-derived mutations in the epidermal growth factor receptor (EGFR), (2) the structure and regulation of Jak-family kinases and their interactions with cytokine receptors, (3) formin proteins and their role in assembling the actin cytoskeleton, and (4) the structural biology of focal adhesion kinase (FAK).
Epidermal Growth Factor Receptor
Mutations in the tyrosine kinase domain of the epidermal growth factor receptor are a major cause of non-small cell lung cancer. We have elucidated the mechanisms by which many of these mutations lead to constitutive ligand-independent activation of the receptor, as well as mechanisms of differential inhibitor sensitivity and acquired resistance. Guided by our structural and enzyme kinetic findings, with our collaborators we developed the first mutant-selective irreversible EGFR inhibitors active against the T790M resistance mutation. Several such agents whose development in the pharmaceutical sector was directly inspired and informed by our work are now showing efficacy in clinical trials, and one (osimertinib) has recently received FDA approval. Most recently, we have developed a first-in-class mutant-selective allosteric inhibitor of EGFR L858R/T790M and L858R/T790M/C797S.
Jak Family Kinases
We have a long-standing interest in the regulation of Jak family kinases and their dysregulation in myeloproliferative neoplasms (MPNs). We reported the first structure of a Jak kinase domain (that of Jak3), which enabled inhibitor development. Our studies of the Jak1 pseudokinase domain revealed a concerted conformational switch involving the V658F activating mutation (equivalent to V617F in Jak2) and an SH2-pseudokinase linker segment that is also the site of mutations in MPNs. We are currently working to understand Jak auto-regulation and mechanisms of cytokine receptor recognition with a focus on Jak2.
Formin and Actin Nucleation Factors
Assembly of monomeric actin into filamentous cytoskeletal structures is highly regulated. In the early 2000’s diaphanous family formins were found to directly nucleate actin filaments upon activation by Rho GTPases. Unlike the Arp2/3 complex, the only known nucleator at the time, formins lacked any actin-like subunits and induced growth of linear filaments, unlike the highly reticular network of filaments produced by the Arp2/3 complex. We elucidated the first structure of an intact formin-homology 2 (FH2) domain. This work revealed a novel “tethered-dimer” architecture and conserved actin binding surfaces, providing a structural foundation for understanding the mechanism of actin nucleation by formins. In ongoing work with our collaborators, we have elucidated aspects of formin autoregulation and discovered and structurally characterized mechanisms of cooperation with partner proteins in both yeast and mammalian systems.
Focal Adhesion Kinase
Through structural and functional studies of diverse mutlidomain signaling proteins including Src-family kinases, focal adhesion kinase, and the phosphatase SHP2, our work has revealed common principles underlying specificity and catalytic control in cell signaling. In each of these signaling devices, modular protein interaction domains are assembled with catalytic domains in a manner that intrinsically couples cognate protein-protein interactions with release of intramolecular autoinhibitory interactions and subsequent catalytic activation. Our work in these systems has also provided a structural foundation for understanding the effect of disease-causing mutations and for drug discovery.