Abstract 2483: Signaling phenotypes in non-small cell lung cancer and normal lung defined by kinase activity profiling and phosphoproteomics

Introduction: Novel treatments are an unmet clinical need for lung adenocarcinoma patients, particularly for those who have K-RAS mutations or completely lack known oncogenic driver mutations. This study seeks to improve our understanding of signaling processes in lung tumors and normal lung. Complementary proteomic approaches have been applied to study both kinase activity level and phosphorylation of substrates to reveal landscapes of signaling in lung cancer and adjacent normal lung tissue as well as patient-specific portraits. The ultimate goal is to use these datasets to suggest novel therapeutic interventions along with the appropriate companion biomarkers for clinical implementation. Experimental Procedures: Frozen specimens from lung adenocarcinomas and adjacent normal tissue resected from the same lobe of the lung were selected and analyzed for comparative phenotyping. An aliquot of each tissue sample was processed for phosphorylation analysis using phosphotyrosine immunoprecipitation (Cell Signaling Technologies) followed by SCX fractionation of the flowthrough and IMAC enrichment of serine/threonine phosphopeptides and LC-MS/MS peptide cataloging. Differentially phosphorylated peptides were discovered by MaxQuant label-free analysis, and selection based on 4-fold differences in ion signal and significant p-values for t-tests comparing tumor and normal tissue. Another aliquot was used for activity-based protein profiling (ActivX probes, Thermo) using ATP-mimics to target active kinases. These samples were analyzed by LC-MRM to enable targeted kinase detection. Data Summary: These experiments enabled quantitative comparison of ∼300 tyrosine phosphorylation sites, >2,000 serine/threonine phosphorylation sites, and ∼160 ABPP-labeled peptides that report the activity levels of specific kinases. Using this information, the overall changes in signaling in these adenocarcinomas were determined and mapped to pathways; then, the signaling in tumors from each patient was also examined using the same techniques. These data indicate common signaling features that can be disrupted with targeted therapy. As an example, phosphotyrosine signatures identify conserved changes from tumor to normal (IRS, DDR1, etc.). In addition, unique phosphosignatures are identified for each patient that yield insights into their unique driver pathways and explore signaling networks to explain drug sensitivity, synergy, and resistance. Conclusions: The combination of phosphoprote