5.1. Proteomic studies to uncover molecular mechanisms associated with malignancies
Most available MS technologies have been used to elucidate the proteomics of breast carcinoma. The degree of tissue heterogeneity of breast carcinomas, a serious problem obscuring quantitative comparative experiments, may be overcome by using LCM. Current emphasis is on infiltrating vs. in situ ductal carcinoma, aiming to uncover differential profiles for diagnosis as well as monitoring disease response to therapy . The complexity of the breast cancer proteome may be simplified by concentrating on specific subcellular compartments. For example, MS-based approaches have been explored to study lysosomes, such as the aspar-tic protease cathepsin that has been shown to be involved in disease progression . Despite the fact that removal of cells from their natural microenvironment may lead to gaining or losing certain characteristics, the in vitro study of cell cultures still has obvious advantages. Novel uses of MS include the measurement (GC/MS) of epithelial cell proliferation using 2H2O labeling for assessing the effects of antiproliferation chemopreventive and chemotherapeutic agents , and the 2D LC/MS analysis of similarities and differences between hundreds of membrane proteins in MCF7 and BT474 cell lines .
Several studies have been carried out to obtain proteomic profiles in human lung cancer cell lines. Proteomic signatures were obtained for different histological types of lung cancer. Hierarchial clustering analysis and principal component analysis of separated (2D-DIGE) proteins revealed 32 proteins that were used to categorize cancer cell into distinct histological groups . Investigations of the proteome of lung squamous carcinoma utilized MALDI-TOF-MS and several databases to identify some 76 differentially expressed protein spots obtained by electrophoresis .
Proteomic analyses of exosomes from malignant pleural effusions  and human mesothelioma cells  revealed several discrete sets of proteins involved in antigen presentation, signal transduction, migration, and adhesion, suggesting interactions between tumor cells and their environment. A large number of proteins were identified in a study of human pleural effusions including several that were suggested to play a role in the development and progression of the cancer phenotype .
A study of protein profiles in gastric adenocarcinoma revealed diverse alterations related to self-protection efforts of cells and changes during the malignant transformation. An 18 kDa antrum mucosa protein was significantly underex-pressed in progressing tumors. It was concluded that the global consideration of the expressed profile alterations will provide insights into the pathogenesis of the tumor .
5.2. Proteomic profiles to provide predictors of drug-modulated targets and responses
Individuals with inherited familial adenomatous polyposis (FAP) develop numerous polyps, the premalignant precursors to colorectal carcinoma. A remarkable heterogeneity in patient response was observed in a clinical trial with a cyclooxygenase-2-inhibitor, celocoxib, which is known to be efficacious in FAP. SELDI proteomic profile revealed that a putative marker at 16,961.4 Da was a strong discriminator between response and nonresponse .
5.3. Profiles to identify proteins associated with disease progression
In cell line studies, an 11 kDa protein was identified by MALDI-MS and database search as S100C (calgizzarin) which is significantly downregulated in bladder cancer and is associated with poor survival; loss of S100C which was also significantly associated with poor survival in patients . Two downreg-ulated proteins, identified as isocitrate cytoplasmic and peroxiredoxin-II, were identified in both bladder cancer cell lines and human biopsies. Loss of these proteins marked the progression of malignancy . In a study of tumor subsets in lung cancer, 15/1600 separated peaks provided a call-prediction model to distinguish primary tumors from metastasis and to distinguish between patients with resected nonsmall-cell lung cancer and poor prognosis from those with good prognosis .
5.4. Targeted biomarker detection via whole protein analysis
The developing technology of "top-down" protein identification does not rely on peptide ions for identification; thus, it avoids the dilution effect for small proteins associated with digestion. Targeted characterized proteins may be analyzed by readily available ion trap mass spectrometers. The concept was demonstrated by using top-down MS/MS for the identification of N-terminally acetylated thymosin p4, that is expressed in certain lung adenocarcinoma cells, and is considered as a putative biomarker. This work is of interest because of the methodological details .
5.5. Sphingolipids in cancer pathogenesis and treatment
Ceramide, a major component of sphingolipid metabolism, functions as a tumor suppressor lipid, inducing antiproliferative and apoptotic responses in various malignant neoplastic cells. Conversely, sphingosine-1-phosphate (S1P) has been shown to be a tumor promoting lipid. Various exogenously supplied ceramides are now known to induce antiproliferative and other important cell function-related responses and thus represent a target for cancer therapy [139,140]. The development of a series of protocols for the high-throughput, structure-specific, and quantitative analysis of sphingolipids by HPLC-tandem MS (both triple quadrupole and ion trap) permits the investigation of this large and chemically complex group of compounds .
There are literally hundreds of MS techniques described for the quantification of antineoplastic drugs in body fluids . A representative publication describes the simultaneous determination of methotrexate and cyclophosphamide in urine by a validated LC-ESI-MS/MS method. The impressive lower limits of detection were 0.2 (xg/L for methotrexate and 0.04 (xg/L for cyclophosphamide . The advantages of multiple reaction monitoring may be appreciated by reviewing a method developed for the quantification of the farnesyl transferase inhibitor lonafarnib in human plasma using HPLC coupled with tandem MS . Attention is called to the increasing inclusion of validation experiments, using FDA guidelines, a common requirement for techniques used to obtain data for pharmacokinetic studies, particularly in Phase I clinical trials of new drugs (www.fda.gov/cder/guidance/4252fnl. html).
The stomachs of about half of the people in the world are colonized by H. pylori, a Gram-negative organism that is assigned as a Class 1 carcinogen. While most colonized individuals are asymptomatic, a subpopulation of 10-20% develops peptic ulcers that may in turn evolve into adenocarcinoma; 30-90% of gastric cancers (a major health problem worldwide) are tied to this microorganism . There are significant clinical and economic aspects of screening for, and diagnosis of, this infection .
The "gold" standard of diagnosis is the 13C-urea breath test which is based on the fact that while humans have no endogenous urease activity in the stomach, Helicobacter species have high urease activity. In the stomach H. pylori hydrolyzes 13C-enriched urea to 13C and NH3; thus, the determination of the area ratios of the 13CO2 to 12CO2 peaks in expired air is diagnostic. Being a stable, nonradioactive isotope, 13C can be administered safely to children and pregnant women. The excess 13C in exhaled breath can be determined accurately with dual-inlet gas isotope-ratio mass spectrometers [146-148]. Bench-top GC/MS instruments (SIM mode) have also been evaluated for the determination of the 13CO2 to 12CO2 peak area ratios; both sensitivity and specificity values were in the 96-98% range .
Lysates and extracts from six different H. pylori strains were analyzed by MALDI-TOF-MS. It was concluded that the strain-specific biomarkers identified might be used in a fingerprinting technique for strain typing . In another MALDI-TOF-MS investigation, a potential biomarker of 58,268 Da could distinguish H. pylori from H. mustalae and Campylobacter species. It was concluded that, together with three strain-nonspecific markers, the technique is adequate for the rapid detection of these organisms in foods, beverages, or manufactured products . Utilizing predictive information from the H. pylori genome, some 20 candidate proteins were identified by MALDI-TOF-MS in proteolytic digests of H. pylori lysates from blood samples of infected patients. It was concluded that this approach has potential for vaccine development . In another approach to recognize antigenic proteins as candidates for vaccines, hundreds of proteins were separated by 2D electrophoresis and analyzed by MALDI-TOF-MS, revealing some 960 mass spectra leading to the confirmation of the presence of 24 previously unidentified proteins . Use of similar methodology to study subproteomes of soluble and structure-bound H. pylori proteins led to the identification of several structure-bound proteins that may be candidates for diagnostic and/or vaccine investigations .
All definitions of cancer chemoprevention (and there are many) include the use of chemical means for the inhibition, retardation, or reversal of the carcinogenetic process. In contrast to tumor biomarkers which are associated with established neoplasia or metastasis, there are two types of biomarkers relevant to cancer chemoprevention: risk biomarkers, referring to genetic predisposition, medical history, lifestyle, exposure, and cellular abnormalities (no detectable premalignant or malignant disease), and biomarkers of chemopreventive intervention, referring to such biological alterations of early/intermediate carcinogenesis which may be effected by chemopreventive agents .
An important group of relevant biomarkers are DNA adducts which originate from the chemical modification of bases in DNA or amino acids in proteins by toxic chemicals. Advantages of LC-ESI-MS methods, compared to 35P-postlabeling (the "gold" standard) and GC-MS (with negative CI), include detection of highly polar compounds without derivatization, direct analyte identification, and accurate quantification using internal standards (preferably stable isotope labeling); current detection limit is 1/108 nucleotides. Relevant applications include aflatoxin B1 and hepatocellular carcinoma, oxidative DNA adducts of prostate carcinoma, and M1G adducts in a variety of malignancies and in the predisposition to gastric adenocar-cinoma resulting from infection by H. pylori .
Was this article helpful?