Events

Title	Abstract
"Warburg Revisited: Using Proteomic Techniques to Evaluate Oxidative Phosphorylation Enzyme Levels in Cancer Cells " Paul Herrmann, National Cancer Institute, Food and Drug Administration herrmanp@mail.nih.gov	Regulation of oxidative phosphorylation has long attracted interest in cancer research due to shifts in metabolism noted upon tumorigenesis by Otto Warburg. In the oxidative phosphorylation machinery, the enzyme complex cytochrome oxidase is a crucial link between oxygen and transported electrons. This complex consists of , among others, protein subunits I and II, which are, encoded in the mitochondrial genome and protein subunits IV, Vb and VIc, which are encoded in the nuclear genome. A signaling pathway for coordination of transcription at these disparate sites of subunit coding exists, but presently is not fully elucidated. Normal cells are expected to maintain subunit ratios very close to those of the functional complex since individual subunits have not been found to retain functionality useful to the cell. We hypothesize that during tumorigenesis the communication pathway for coordinated subunit production becomes altered, resulting in unequal subunit ratios. To test this hypothesis, the quantitative relationships of these proteins in normal and cancerous prostate tissue were examined. Lysates from normal and tumor epithelial cells were obtained by Laser Capture Microdissection of whole organ tissue mounts, permitting comparison of patient matched material. Subsequent Western blot analysis revealed that the concentrations of subunits IV, Vb and VIc varied independently of each other between cancerous and normal cells. Subunits Vb and VIc are of greater relative abundance in prostate cancer compared to normal tissue. Subunit IV shows only slight changes between tumor and nontumor cells. Subunits I and II are covariant with fixed ratio to one another in all dissected tissue. This study demonstrates that, within normal and cancerous cells from the same specimen the relative concentrations of the subunits encoded in the mitochondrial genome remain fixed, while the relative abundance of some of the subunits encoded in the nuclear genome change as a cause or consequence of disease. Evaluation to determine the mechanism responsible for these differences and what affect this has on oxidative phosphorylation is currently under way. Importantly, studies are also under way to assess potential changes in oxidative phosphorylation pathways that may arise in early stage cancer and relate to the clinically aggressive versus clinically indolent prostate cancer phenotype.
"Proteomics, Host-Parasite Interactions, and Rapid Identification of Microorganisms" Hercules Moura, Universidade do Estado do Rio de Janeiro & C.D.C. HXM3@cdc.gov	Parasitic diseases continue to be an important problem in the world, increasingly associated with travel and immigration, immunodeficiency, and emerging and re-emerging infectious agents. Proteomics applications to parasitic agents are in infancy but have already lead to new insights on molecular pathogenesis and microorganism identification. As it is rapidly occurring in other areas, the study of all proteins encoded for by the genome of parasites and hosts using proteomics is being used to investigate global protein synthesis and gene expression. We describe the use of the classical proteome approach (2-DE /MS) to study protein expression of two human microsporidian isolates, Encephalitozoon intestinalis (CDC V:307) and Brachiola algerae (CDC V:404) during their multiplication within monkey kidney (E6) cells. Microsporidial parasites survive and develop inside the host cells and during their merogonic multiplication and sporogonic differentiation they display different patterns of protein expression. In the western blots of the 2-DE gels, approximately 50 spots were detected and these may be considered as markers of spore maturation proteins. In addition, an immunoreactive 10 kDa band, present only in extracts of E-6 infected cells, is probably of merogonic or sporogonic origin and correspond to at least 8 spots in the pI range of 4.5–7. Identification of selected spots is under way by mass spectrometry. Rapid identification of whole microorganisms by matrix-assisted laser desorption-ionization mass spectrometry (MALDI-TOF MS) is an emerging technique with a wide range of applications. Pathogens belonging to different genera and species were analyzed and unique profiles were determined. This approach, now referred to as a proteomics application by serendipity, has demonstrated its feasibility and practical value to successfully characterize a wide range of cultivated microorganisms at genus, species, and strain level. The proteome approach is a powerful tool to investigate parasites and will provide new data for the development of innovative diagnostic, therapeutic and control strategies for parasitic diseases.
"Proteomic analysis of human breast ductal carcinoma in situ (DCIS)" Julia Wulfkuhle, National Cancer Institute jdwulf@mail.nih.gov	The diagnosis of premalignant breast lesions and ductal carcinoma in situ (DCIS) confers an increased risk for the eventual development of invasive breast cancer. Despite the clinical evidence for the significance of these early breast lesions and their importance to the development of prevention efforts, we have an incomplete knowledge about their molecular differences. We report the proteomic analysis of matched DCIS and normal breast ductal epithelium. Approximately 75,000 cells from frozen sections of DCIS and matched normal ductal structures were obtained by laser capture microdissection (LCM). Each sample was subjected to two-dimensional gel electrophoresis, followed by image analysis and protein sequencing of selected spots by mass spectrometry. To date, six surgical specimens have been analyzed and approximately 400 spots have been sequenced to date. Sequencing has identified 140 different proteins and they fall into a number of unexpected categories such as subcellular trafficking, cytoskeletal structure and signaling, chaperone proteins, general signaling molecules, transcription and genomic instablility, invasion and metastasis, oxidative damage and protein turnover. Selected proteins overexpressed in DCIS compared to matched normal epithelial cells include: annexin V, annexin VII, S100A11, S100A7, Rab-11a, prohibitin, thioredoxin peroxidase II, Hsp 27 and chloride intracellular channel protein 1. Among the proteins with higher expression in normal tissue are trangelin, voltage-dependent ion channel 1, annexin II, and fatty acid binding protein 1. We are currently validating the differential expression of these potential targets in larger cohorts by immunohistochemistry and immunoblotting. This analysis has identified a number of novel proteins with potential involvement in breast tumorigenesis and has provided a unique set of differences from those identified in microarrays and SAGE analyses of breast cancer progression. Such leads may be useful in imaging and diagnostic studies. If causally involved in breast oncogenesis, differentially expressed proteins identified herein may enable the development of prevention schemes.
"A Proteomic Approach for Human Laryngeal Muscles: Implications for Future Strategies in Gene Therapy" Zhao Bo Li, Johns Hopkins University, School of Medicine zbli2001@yahoo.com Paul Flint presenting for Zhao Bo Li pflint@jhmi.edu	In order to characterize protein expression of human laryngeal muscles and to elucidate their functional role in airway protection, respiration and phonation, we compared the global protein expressions of 5 human laryngeal muscles by means of two-dimensional electrophoresis and mass spectrometry. PDQuest software (BIO-RAD) was used for analysis of the digitized gel images including spot detection, quantification, comparison and statistical analysis. The thyroarytenoid muscle (TA), lateral cricoarytenoid muscle (LCA), posterior cricoarytenoid muacle (PCA), vocalis muscle (VOC) and cricothyroid muscle (CT) were analyzed. An average of 897 was detected among the five muscle types; and, database search provided identification of 80 individual proteins. The densities of 159 matched spots were significantly different among paired groups in the pH 5-8 range and molecular weight150-10Kd (P<0.05). Twenty-five of the 159 were identified using mass spectrometry after in gel trypin digestion. The 25 protein identified can be classified into four broad functional groups: 1) membrane receptor and signal transduction proteins, 2) metabolic proteins associated with energy production, 3) stress responses proteins, and 4) serum protein expressed in muscle. Twelve protein spots with good mass spectrometry data were not identified in the database search and may represent novel proteins. Those results will provide useful dates for characterizing human laryngeal muscle and elucidating their functional role and relationship to fast versus slow muscle fibers.
"Comparative Proteomic Studies to Elucidate Mechanisms of Acquired Drug Resistance " Kristy J. Reynolds, University of Maryland, Department of Chemistry & Biochemistry kjr17@erols.com	The major cause of failure in breast cancer chemotherapy is the development of multi-drug resistance. In order to contribute to understanding the mechanisms of resistance, comparative studies are underway to detect differences in regulation or modification of proteins in the cytosolic fraction of drug-susceptible and drug-resistant human breast cancer MCF-7 cells. A desirable comparative proteomic method should allow for relative quantitation, high-throughput capabilities and allow for the study of difficult proteins. A proteolytic ¹⁸O labeling method combined with online capillary LC-MS/MS analysis is offered as a solution to these requirements. In this method, pools of proteins are first enzymatically digested in parallel in H₂¹⁶O. The counterpart peptide mixtures are dried and then isotopically labeled by incubation with additional enzyme in H₂¹⁶O and H₂¹⁸O, respectively. The separation of the protein digestion and ¹⁸O labeling steps allows for the protein mixtures to remain in solution, and therefore minimize losses. In one peptide pool, two atoms of ¹⁸O are incorporated into the carboxyl-terminus of each new peptide. Comparative proteomic studies can be performed by mixing the unlabeled peptide pool (generated in H₂¹⁶O) and the isotope labeled peptide pool (from H₂¹⁸O) and analyzing the peptide pairs by mass spectrometry. Relative quantitation information is derived by mass spectral measurements of the ratios of labeled and unlabeled isotope pairs. Tandem mass spectrometry experiments provide sequence information to identify the proteins. Kinetic studies have shown that endoprotease Glu-C, trypsin, Lys-C and chymotrypsin. reliably incorporate two ¹⁸O into the carboxyl-terminus of each new peptide and validate this as a universal method. Labeled and unlabeled peptide pairs co-elute chromatographically, and isotope ratios correlate well with spectroscopic determinations. Phosphorylation and glycosylation of proteins were found not to interfere with labeling. The strategy is being applied to compare levels and identify proteins in cytosol fractions from MCF-7 cell lines.