Our current understanding of carcinogenesis indicates that cancer cells incur increasing mutational burden, with each mutation defeating new protective mechanisms encountered during tumor etiology, so that there is a progression in morphology and behavior. In melanoma, this progression is reflected in morphological """"""""stages"""""""": atypical naevus, radial growth phase (RGP), vertical growth phase (VGP), and metastasis. Although these changes are driven by signaling pathways, the complexity of these pathways makes it impossible to directly or quantitatively connect them to specific signaling events. A more precise method of delineating the phenotype of the cell in a way that is quantifiable and directly related to the underlying signaling status would be valuable both for diagnostics and for basic research. The goal of this proposal is to show that the new proteomics methodologies can be used in this way. Using two-dimensional gel electrophoresis (2D), we have surveyed approximately1500 proteins in three cell lines derived from metastatic melanoma. We find 43, 47, and 58 proteins in the pI range 4-7 with >10 fold change in expression, compared to an immortalized, non-- tumorigenic melanocyte cell line. When the metastatic cell lines are compared to each other, we find that each line shows a progressively altered proteome, exactly as predicted by the progression model for carcinogenesis. We hypothesize that (a) other stages can be identified in a similar manner, using cell lines representing different stages in melanoma progression, (b) the proteins in common between the three lines are general markers for metastasis, and (c) successive mutations in protooncogenes, survival factors, or tumor suppressors which regulate tumor progression direct the expression of each marker subset. In this proposal, we will test these hypotheses by (i) comparing additional cell lines derived from different stages of melanoma and identifying the marker proteins by in-gel digestion and mass spectrometric sequencing of the peptides, (ii) confirming marker expression in melanoma biopsies, and (iii) determining the involvement of specific signaling pathways in regulating the expression of these markers, focusing particularly on the role of Ras in RGP to VGP transition, and PI3K in metastasis. The complexity of signaling pathways require that cancer research shift from an emphasis on linear signaling pathways to paradigms that consider the entire signaling pattern, and we believe that the methodology used in this proposal will be an important driving force for that change.
