Barriers to the analysis of proteins in tumor samples by traditional methods include the difficulty in accurate quantitation of the relative amounts of individual proteins in multiple samples and the inhomogeneity of most solid tumor specimens. This proposal overcomes these barriers by combining several new technologies in a cross-specialty collaborative effort. This strategy draws upon expertise from clinicians, pathologists, basic scientists, mass spectroscopists, and bioinformaticists. In order to obtain relatively homogeneous populations of tumors, cancer cells will be isolated by laser capture microscopy. While this technique permits isolation of specific subpopulations of cells within tumors, the total number of cells is often low. The proposal will therefore draw on a newly developed approach called 2-D DIGE (2-Dimensional Differential In-Gel Electrophoresis). In this method, protein extracts are labeled in vitro with reactive cyanine dyes that fluoresce at one of several chosen wavelengths. Up to three extracts labeled with different dyes are mixed and analyzed in the same large-format two-dimensional gel. The gel is imaged at multiple wavelengths and analyzed to determine the precise ratio of proteins migrating in various spots. The unique aspect of the technology is that it allows independent quantitation of proteins derived from two or three biological samples in the same gel, eliminating issues of gel-to-gel reproducibility and thus providing an exceptionally accurate proteome map. A new generation of dyes allows detection and quantitation of individual polypeptides present in picogram amounts. By combining this sensitive proteomic detection method with laser capture microdissection technology, it should be possible to derive a useful proteome map derived from cytologically homogeneous specimens containing as few as 1,000 to 10,000 cells. In this phase, we will focus on three goals for working with cancer samples obtained by laser capture microdissection (LCM): to directly compare the new generation of cysteine-reactive dyes to the lysine-reactive dyes for the labeling of LCM samples, to identify strategies that permit identification of protein species of interest by mass spectrometry, and to perform pilot experiments of LCM and 2D-DICE using the optimized techniques. In the next phase, broader clinical utility will be demonstrated using a larger sample set. Changes in the pattern of protein expression will be identified. These changes can serve as clinically useful markers of tumor progression.
