Dr. Martin's research in the field of molecular diagnostics has focused on using tumor cell lines to generate candidate proteins for biomarkers. He has explored the "secretome" of prostate cancer cells using a mass spectrometry based quantitative proteomic analysis. In his published work, he sought to identify prostate tumor cell proteins that are either "shed" or secreted into the extracellular environment and that may be of diagnostic value. In this context, he used a well-characterized prostate cancer cell line, LNCaP, and exploited the biochemical secretory program in these cells that is stimulated by androgenic hormones. To quantify secreted/shed proteins, isotopically light and heavy ICAT reagents were used. Proteomic analysis of the LNCaP media identified in excess of 600 proteins, 524 of which could be quantified. Ten percent of the proteins identified were androgen regulated and a subset of these proteins appeared to be expressed in abundance. Selected mass spectrometry observations were confirmed by Western analysis. The findings indicated that androgen-mediated release of proteins may occur through the activation of proteolytic enzymes rather than exclusively through transcriptional, translational, or secretory control mechanisms. In collaboration with Dr. Knudsen at FHCRC, we demonstrated that hepatocyte growth factor activator inhibitor 1 (HAI-1) is over expressed in primary prostate carcinoma.

Following up on the results described above, Dr. Martin has begun a second generation of proteomic experiments to develop a more focused list of potential biomarker targets using a selective capture strategy that significantly improves the specificity of proteomic screening of cell lines. Methods developed at the Institute for Systems Biology permit the covalent capture of proteins through N-linked carbohydrate side chains. These methods targets proteins destined for secretion and the extracellular membrane, which are glycosylated during post translational processing in the endoplasmic reticulum and Golgi. Two such glycocapture methodologies are currently being applied to prostate cancer cell lines. In the first, extracellular membrane proteins are biotinylated on live cells through the carbohydrate groups for capture by streptavidin. The second method captures N-linked glycoproteins in solution from culture media using an immobilized hydrazide chemistry. In both of these methods, proteins are digested to peptides which are released from the resin by cleaving the bond between sugar and the asparagine residue. Liberated peptides are analyzed using mass spectrometry. The Martin group has combined the glycocapture methodology with a cutting edge mass spectrometry based validation strategy called QCAT. This strategy utilizes a concatamer of the tryptic glycopeptides identified in the glycoprotein analysis for expression as an artificial epitope tagged protein which is purified and digested into peptides. This protein is expressed in isotopically heavy form, thus each peptide is isotopically heavy. The collection of peptides can be spiked into human serum for absolute quantification of the light peptides derived from an endogenous tumor. The analysis is performed using an ultra-sensitive mass spectrometry mode focused on only the peptides of interest. This method circumvents the problems of ELISA development, which can be expensive and time consuming. Its strengths include the ability to probe for cancer derived proteins rather than identify global changes associated with disease as is the case with top down discovery methods. Success with this method could dramatically alter the field of biomarker discovery.

With this list in hand, we will employ a cutting edge mass spectrometry based validation strategy called QCAT. This strategy utilizes a concatamer of the tryptic glycopeptides identified in the glycoprotein analysis for expression as an artificial epitope tagged protein which is purified and digested into peptides. This protein is expressed in bacteria grown in isotopically heavy ammonium chloride supplemented minimal media, and thus each peptide is isotopically heavy. The collection of peptides can be spiked into human serum for absolute quantification of the light peptides derived from an endogenous tumor. The analysis is performed using an ultra-sensitive mass spectrometry mode focused on only the peptides of interest. This method circumvents the problems of ELISA development, which can be expensive and time consuming. Its strengths include the ability to probe for cancer derived proteins rather than identify global changes associated with disease as is the case with top down discovery methods. Success with this method could dramatically alter the field of biomarker discovery.