RESEARCH
Signaling by Receptor Tyrosine Kinases from the erbB/HER family
We are interested in understanding how growth factor receptors from the epidermal growth factor (EGF) receptor family signal across the membrane. For the EGF receptor itself, X-ray crystal structures recently determined in our lab and elsewhere have shown that EGF binding induces conformational changes that promote receptor dimerization (which is responsible for receptor activation).
It is also known that the four members of the EGF receptor family, which includes EGF receptor, erbB2 (also known as HER2/Neu), erbB3, and erbB4 from hetero-oligomers. We are now trying to understand this hetero-oligomerization process using cellular, biochemical, and biophysical approaches. Above all, we are interested in understanding how erbB2/HER2/Neu is activated.
This member of the family has no known ligand, yet is activated in trans by ligands for other family members – through receptor heteromerization. ErbB2 is overexpressed in some 30% of human breast cancer cases, and the value of HerceptinTM as a breast cancer drug has shown it to be an important therapeutic target.
If we are able to understand the normal mechanism of erbB2 regulation, we hope that this will suggest new pharmacological approaches for targeting this process that will not bring with them the disadvantages of HerceptinTM. Our approach to this is multidisciplinary, and currently draws substantially from insights gained from our recent structural studies.
Signal-Dependent Membrane Recruitment by Small Domains
The second main focus of the laboratory is on small (100 aa or so) domains in signaling, cytoskeletal, and other proteins that recognize membrane components and target their host proteins to cellular membranes. To date we have worked primarily with pleckstrin homology (PH) domains, and have shown structurally how a subset of PH domains recognize lipid products of agonist-dependent phosphoinositide 3-kinases, and so can drive acute recruitment of their host proteins to the plasma membrane. The PH domain is the 11th most common domain the human proteome. We now know that, while several bind to specific phosphoinositides, many (most) PH domains do not. We have recently embarked on a genome-wide analysis of PH domains in S. cerevisiae in order to ascertain what other roles PH domains play.
In addition to PH domains, we are also interested in the roles of FYVE domains and phox homology (PX) domains, which bind to phosphatidylinositol-3-phosphate, a lipid found in endosomal compartments. We have analyzed all S. cerevisiae PX domains, and are currently assessing their physiological roles.
A current focus in work on these domains is to test the hypothesis, suggested by several observations, that PH and PX domains may act as ‘coincidence’ detectors, effectively checking for the coincidence of a particular protein target and lipid target in the same cellular compartment. Our approaches again draw from biochemical, biophysical, and cell biological studies.