(283f) Clinical Mutations in the Epidermal Growth Factor Receptor and Relevance to Oncogenic Transformations: from Molecular to Systems Level Modeling

Radhakrishnan, R., University of Pennsylvania
Shih, A., University of Pennsylvania
Liu, Y., University of Pennsylvania
Purvis, J., University of Pennsylvania
Chou, S., University of Pennsylvania
Lemmon, M. A., University of Pennsylvania

It is well established that EGFR and ErbB2 -- two (of 4) members of the ErbB family of receptor tyrosine kinases (RTKs) -- are over expressed in a large percentage of clinical cancers of various types. Therefore, small molecule receptor tyrosine kinase inhibitors for EGFRTK and ErbB2 RTK are of significant interest as cancer therapeutic drugs. The EGFRTK inhibitor, gefitinib (Iressa), was recently approved for the treatment of Non-Small-Cell Lung Cancer (NSCLC) in many countries. Despite the promise in a small (demographic) sub-population with significantly advanced disease, clinical responses to gefitinib varied among population samples. Recently, a major epidemiological breakthrough correlated the gefitinib response with somatic mutations in the EGFRTK domain. Most of the mutations (L858R, L861Q, G719S, del L747-P753 ins S) correlated with an increase to the gefitinib sensitivity (referred-to here as gefitinib-sensitizing mutations); Some mutations conferring resistance to the drug were also reported. Despite the growing number of studies addressing such correlations, the underlying biochemical basis involving the drug sensitivity to the mutations is still not clear.

The remarkable sensitivity of cell lines carrying gefitinib sensitizing mutants appears not to be centered around inhibitor or ATP binding affinities, but rather on other biochemical mechanisms regulating the EGFRTK activity and EGF-mediated signaling. Identifying and understanding the role and significance of drug sensitizing mutations of ErbB2 would also be of enormous value, and would help in targeting patients for treatment with another small molecule inhibitor lapatinib, a dual-specific agent that inhibits both EGFR and ErbB2 TKs, currently in clinical trials.

We have carried out an integrated modeling and experimental study of the EGFR mutant systems to delineate the molecular regulatory mechanisms governing receptor activation. The drug sensitizing mutations impact epidermal growth factor-mediated signaling by affecting the activation (conformational switching of the A-loop) mechanism and the catalytic activity of EGFRTK, but do not significantly affect the enzymes interaction with ligands (ATP and inhibitors). We have contrasted the A-loop conformational switching mechanisms for wildtype EGFRTK and mutated forms associated with gefitinib sensitivity (L858R, G719S, and del L747-P753 ins S) by analyzing the relevant EGFRTK dimers using molecular dynamics simulations. Using the wildtype studies as a reference, we have computationally quantified the effect of sensitizing mutations (L858R, G719S, and del L747-P753 ins S) on the specificity of binding different peptide sequences, ATP, and the drugs (gefitinib, erlotinib, and lapatinib) by employing free energy docking studies. Experimentally, we have measured the biochemical properties of EGFRTK and full-length EGFR to assess the influence of EGFR mutations (L858R, L861Q, G719S, and del L747-P753 ins S) on the biochemical properties of the kinase. We have also analyzed phosphorylation kinetics for monomeric and dimeric forms of the full length intracellular domains of the EGFRTK. Based on the activation mechanism, we have identified candidate mutations likely to upregulate receptor activation. We have also determined ab-initio, the parameters characterizing the interaction map of EGFR with its phosphorylation substrates for the mutant systems and determined the signaling response of cell lines carrying the mutations in relation to the wildtype system.

The computational approaches described here are also extended to identify which of the newly isolated mutations are likely to be associated with enhanced constitutive activation of the kinase domain, and thus EGFR/ErbB2 dependence (and sensitivity to gefinitib/lapatinib). We believe that our model driven approach will in the long-term significantly impact the optimization of future small molecule therapeutic inhibition strategies.