(155f) Altered Localization and Activity of Shp2 in Lung Cancer Cells with EGFR-Activating Mutations Contributes to Enhanced Cellular Sensitivity to EGFR Kinase Inhibitors
Multiple therapeutics have been designed to inhibit the activation of the epidermal growth factor receptor (EGFR) in cancer. For example, the EGFR kinase inhibitor gefitinib is used in the treatment of non-small cell lung cancers (NSCLC), 70-80% of which display elevated EGFR expression. Despite the frequency of elevated EGFR expression in NSCLC, gefitinib is typically only effective in the 8-10% of patients whose tumors bear EGFR-activating mutations. We previously reported that these mutations surprisingly result in an impairment of EGFR-mediated ERK activation and that this effect correlated with reduced phosphorylation of Shp2, a tyrosine phosphatase whose activity is required for complete activation of ERK downstream of many growth factor receptors. Here, we explore the mechanism underlying differential Shp2-mediated ERK activity and its implications in NSCLC cells expressing wild-type or mutant EGFR. Our previous results suggested that Shp2 activity might be impaired in NSCLC cells with EGFR-activating mutations. We explored this possibility by probing for the ability of Shp2 to co-immunoprecipitate with the adapter protein GAB1, whose binding to phosphorylated Shp2 in response to growth factors is one of the principal activators of Shp2 activity. Interestingly, we found that in the context of EGFR mutations, Shp2 and GAB1 were pre-associated in serum-starved cells. EGFR also co-immunoprecipitated with Shp2 in EGFR-mutant-expressing cells in the absence of stimulatory factors. The association between Shp2 and GAB1 was enhanced in all cell lines by the addition of exogenous EGF, but the fold increase in association was significantly lower in cells expressing EGFR mutants. To directly assess the activity of Shp2 in NSCLC cells, we performed biochemical activity assays with immunoprecipitated Shp2 from cells treated with EGF for up to one hour. We found that Shp2 was active in cells with EGFR mutations, but that the induction in response to ligand was not as significant as in cells expressing wild-type EGFR. Because we have previously demonstrated that EGFR mutants fail to undergo normal ligand-mediated endocytosis, we interpreted our findings as indicating that Shp2 is dephosphorylated and active, but sequestered at the cell periphery, in NSCLC cell lines with EGFR mutation. We found that such sequestration has important phenotypic implications for NSCLC cells with EGFR mutations. Specifically, we found that shRNA-mediated knockdown of Shp2 reduced basal and ligand-induced ERK phosphorylation in cell lines expressing wild-type, but not mutant, EGFR. This effect at the level of ERK in wild-type EGFR cells was accompanied by enhanced cellular sensitivity to the EGFR kinase inhibitor gefitinib. In total, our findings point to Shp2 and ERK as potential targets to be co-inhibited with EGFR in the treatment of NSCLC cells expressing wild-type EGFR. In a broader context, our findings highlight the possibility for activating mutations of kinases such as EGFR to shift the state of intracellular processes in ways that can have significant implications for cellular fates.