High-throughput analysis of genome-wide receptor tyrosine kinase expression in human cancers identifies potential novel drug targets.

Abstract

Novel high-throughput analyses in molecular biology allow sensitive and rapid identification of disease-related genes and drug targets. We have used quantitative real-time reverse transcription-PCR reactions (n = 23000) to analyze expression of all human receptor tyrosine kinases (n = 56) in malignant tumors (n = 313) of different origins and normal control samples (n = 58). The different tumor types expressed very different numbers of receptor tyrosine kinases: whereas brain tumors and testicular cancer expressed 50 receptor tyrosine kinases, acute myeloid leukemia (AML) samples expressed only 20 different ones. Specimens of similar tumor origin exhibited characteristic receptor tyrosine kinase expression patterns and were grouped together in hierarchical cluster analyses. When we focused on specific tumor entities, receptor tyrosine kinases were identified that were disease and/or stage specific. Leukemic blasts from AML bone marrow samples differed significantly in receptor tyrosine kinase expression compared with normal bone marrow and purified CD34+ cells. Among the differentially expressed receptor tyrosine kinases, we found FLT3, c-kit, CSF1 receptor, EPHB6, leukocyte tyrosine kinase, and ptk7 to be highly overexpressed in AML samples. Whereas expression changes of some of these were associated with altered differentiation patterns (e.g., CSF1 receptor), others, such as FLT3, were genuinely overexpressed in leukemic blasts. These data and the associated database (http://medweb.uni-muenster.de/institute/meda/research/) provide a comprehensive view of receptor tyrosine kinase expression in human cancer. This information can assist in the definition of novel drug targets.