Abstract
We are using a candidate gene approach to identify genes contributing to cancer through somatic mutation. Somatic mutations were found in breast cancer samples in the human casein kinase I epsilon (CKIepsilon) gene, a homolog of the Drosophila gene dco in which certain point mutations lead to imaginal disc overgrowth. We therefore created fly genotypes in which the dco gene carried point mutations homologous to those discovered in CKIepsilon, and tested them in vivo. The results show that the most frequent mutation discovered in breast cancer, L39Q, causes a striking overgrowth phenotype in flies. Further experiments show that this mutation affects the newly recognized Fat/Warts signaling pathway, which controls organ size and shape in both flies and mammals. Another mutation, S101R, modifies the mutant phenotype so that the affected tissue disintegrates, mimicking more aggressive forms of breast cancer. Our results thus strongly support the conclusion that CKIepsilon mutations play important roles in breast carcinogenesis.
