Protein tyrosine phosphorylation plays an important role in many of the biological processes involved in tumorigenesis, progression, and metastasis, and thus the global pattern of tyrosine phosphorylation of a tumor cell is highly relevant to its biological activity. It is likely, therefore, that molecular diagnostic methods based on the detection and characterization of tyrosine phosphorylation patterns in tumors will be useful for classification and prognosis. In the cell, most tyrosine phosphorylated sites on proteins bind tightly and specifically to small modular protein binding domains, in particular Src Homology 2 (SH2) domains. We have recently described a method, termed SH2 profiling, in which a battery of SH2 domain probes is used to profile the global state of tyrosine phosphorylation of a protein sample. In the current proposal, we will develop a novel multiplexed SH2 profiling format based on the labeling of SH2 domain probes with unique oligonucleotide tags. This novel approach will result in a quantitative profiling assay that is rapid, robust, reproducible, and sensitive enough for routine analysis of clinical specimens. In the R21 phase of the proposal we will demonstrate the feasibility of the oligonucleotide-tagged multiplexed (OTM) method and also develop two other quantitative SH2 profiling formats that can be used to validate the OTM method and to establish the utility of the approach for classification of tumor samples. In the R33 phase we will further develop and optimize the OTM method to incorporate the entire complement of approximately 150 phosphotyrosine-binding modules in the human genome and evaluate the performance and cost-effectiveness of different methods of quantitation. We will also fully develop bioinformatic tools to analyze quantitative SH2 binding data and cluster samples based on similarities in binding patterns and perform pilot studies on clinical samples to assess the usefulness of such data for molecular diagnostic classification of cancer. These studies will provide a novel tool for classifying tumors based on tyrosine phosphorylation patterns, which is likely to be useful in predicting the course of disease and response to therapy. They will also establish the feasibility of the OTM approach as a more general proteomic tool for the rapid and sensitive profiling of clinical samples.
