QUANTITATIVE INTERACTION NETWORKS FOR TYROSINE-PHOSPHORYLATED PROTEINS


Year of Award:
2007
Award Type:
R33
Project Number:
CA128726
RFA Number:
RFA-CA-07-016
Technology Track:
Molecular & Cellular Analysis Technologies
PI/Project Leader:
MACBEATH, GAVIN
Other PI or Project Leader:
N/A
Institution:
HARVARD UNIVERSITY
Intracellular signaling networks that involve protein tyrosine kinases are critical in the control of most cellular processes, including growth, adhesion, migration, differentiation, and apoptosis. Misregulation of these networks results in a variety of human diseases, including cancer, diabetes, and immune deficiency. Many of the proteins in these networks contain Src homology 2 (SH2) or phosphotyrosine binding (PTB) domains, which recognize tyrosine-phosphorylated proteins in a sequence-specific fashion. In this proposal, the molecular recognition properties of virtually every SH2 and PTB domain encoded in the human genome will be investigated with respect to physiologically-relevant ligands using protein microarray technology. Recombinant SH2/PTB domains will be arrayed in the wells of microtiter plates and subsequently probed with 258 fluorescently-labeled phosphopeptides representing experimentally-verified sites of tyrosine phosphorylation on human receptor tyrosine kinases, as well as with 604 peptides representing sites of tyrosine phosphorylation on downstream proteins (nonreceptor tyrosine kinases and SH2/PTB-containing proteins). By probing the arrays with eight different concentrations of each peptide, equilibrium dissociation constants will be determined for the binding of each peptide to each protein (~140 active SH2/PTB constructs). This effort will produce high quality, quantitative protein interaction networks which will reveal individual connections between signaling proteins, as well as how network connectivity changes with protein concentration. We have previously proposed that the extent to which a protein becomes more promiscuous when overexpressed contributes to its oncogenicity, and the study described here will generate the quantitative data needed to investigate this hypothesis further. In addition, the information revealed by our systematic efforts should prove invaluable to cell and cancer biologists who study tyrosine kinase-mediated signaling, to computational biologists who study molecular recognition, and to systems biologists who seek to model signal transduction networks. As such, we intend to make our data easily accessible though an interactive web site in formats suitable both for broad computational studies and for more focused hypothesis-driven inquiries. It is our hope that the studies described here will shed light on how signaling proteins are integrated into complex networks and how we can intervene most effectively when these networks go awry.