1A: Identify mutant EGFR substrates using in vitro kinase assays on protein arrays (2.5% of budget) We are in the final phases of validating potential targets identified in this screen. We have a list of around 10 proteins that we have purified biochemically and currently performing in vitro kinase assays with Wild type and mutant EGFR kinases. We have performed new bioinformatics analysis to select these proteins for further validation. Some of them are described below. We sought to identify targets of mutant EGFR kinases using in vitro kinase assays on protein arrays. We have used human protein arrays developed by Invitrogen and more recently collaborated with the laboratory of Dr. Heng Zhu at Johns Hopkins Medical Institute to use the custom protein arrays developed by their group. We used WTEGFR, L858R EGFR, L858R/T790M EGFR in presence or absence of tyrosine kinase inhibitor, erlotinib to perform in vitro kinase assays on these protein arrays. We identified several proteins that are phosphorylated in vitro by mutant kinases, but not WTEGFR. For example, we found STK3, PKCtheta, PAK3, MST4 (MASK), TBK1, LynA to be phosphorylated by L585R EGFR, but not WT EGFR. The protein arrays from Dr. Zhu laboratory have around 20,000 full-length human proteins spotted on glass slides. We have completed the screening and have analyzed the kinase assay data this past year. We are currently validating some of the targets of mutant EGFR identified with this assay. We have undertaken both biochemical and genetic approaches to validate the targets. One such target is the protein DLG1 (disc large). We also identified this protein as a phosphorylation target of mutant EGFR from our mass spectrometry-based phosphoproteomics experiments (see below). We have shown that DLG1 indeed interacts with EGFR. Further characterization of this interaction and the significance of the phosphorylation is underway. 1B: Compare the degree of interaction of proteins that associate with WT EGFR, L858R EGFR and Del E746-A750 EGFR This part of the project is complete and is not further being done. We are currently validating individual interaction partners of mutant EGFR obtained in our kinase assay and mass spectrometry based interaction assay screens. Further description is below. The two most common TKI-sensitizing lung cancer-specific EGFR mutations are constitutively active. However two small clinical trials of patients treated with erlotinib have shown that the prognosis of patients harboring the Del EGFR is better than those with L858R EGFR. Patients with Del EGFR respond better and have more prolonged progression free survival than those harboring the L858R EGFR. We undertook an approach of identifying EGFR-interacting proteins by mass spectrometry. We expressed WT EGFR, L858R EGFR and Del E746-A750 EGFR in NR6 cells, a variant of mouse 3T3 fibroblasts that do not express endogenous EGFR. We used stable isotope labeling with amino acids in cell culture (SILAC) to differentially label cells expressing each of the EGFR variants. Immunoprecipitation of EGFR with a monoclonal antibody specific to the extracellular domain of EGFR was performed to isolate EGFR-interacting proteins that were further identified by mass spectrometry. SILAC labeling of cells enabled relative quantitation of degree of interaction of proteins that interacted with both WT EGFR and mutant EGFRs. We identified several proteins that interacted specifically with mutant EGFRs. Of particular interest is receptor tyrosine kinases that interacted more with L858R and Del EGFR compared to WT EGFR (e.g. EPHA2, DDR2, AXL), suggesting cross talk of receptor tyrosine kinases in cells expressing mutant EGFRs.