Proteomic Applications in Cancer

over recent years, oncology has been a key focus of proteomic technologies as applied in both expression and functional studies. Expression proteo-mic studies are screening for differences in protein patterns between tumor and control tissues. Functional proteomic studies are defined as the combination of readouts with proteomics data from the same sample at different points of time (under different conditions), with the aim of detecting and prioritizing certain proteins or polypeptides of functional relevance.

5.1. Expression Proteomic Studies

As of today, expression proteomic studies are available for all common human cancers and some rare tumors.

5.1.1. Lung Cancer

Proteomic studies in lung cancer, the most common cancer worldwide, have been reviewed [37]. Initial studies on lung cancer proteomics were first published in the early 1990s. These early studies focused on the relationship between histopathological characteristics and 2D-PAGE reproducibility [38]. A few years later, the first differentially expressed proteins in lung cancer were identified in small cell lung cancer (SCLC), including ^-tubulin, heat-shock proteins 73 and 90, lamin B, and proliferating cell nuclear antigen (PCNA). This report demonstrated for the first time that 2D-PAGE combined with protein identification was an effective approach to identify biomarkers in cancer [39]. Later on, with improvements in MS technology, it was possible to identify about 20 potential biomarkers in lung cancer tissue [40]. Recently, a SELDI study in early lung cancer stages and premalignant bronchial lesions analyzed LCM specimens of normal lung, atypical adenomatous hyperplasia, and malignant tumors taken from patients participating in a screening program. Protein profiles were generated in each epithelial cell type and found to be highly reproducible in identifying populations at high risk for lung cancer [41].

When serum samples from lung cancer patients vs healthy controls were analyzed by SELDI technology, a decision classification tree yielded a diagnostic sensitivity and specificity of 93 and 97%, respectively [42]. Another study compared serum samples from lung cancer and healthy controls. Five protein peaks in a blinded test achieved a sensitivity of 87%, a specificity of 80%, and a positive predictive value of 92%. Sensitivity was even significantly better (91%) for detection of non-small cell lung cancers (NSCLC) [43]. Evaluation of circulating autoantibodies in lung cancer patients revealed antibodies against annexins I and II, recoverin, protein gene product 9.5, and a-enolase [44].

Lung cancers are traditionally classified into various subtypes on a his-tological basis. Squamous lung cancer, the most common histological subtype, is followed by adenocarcinoma, and other histologies. These various subtypes were classified on the basis of 2D-PAGE [45] and MS profiles [46].

  1. 1.1.1. Squamous Lung Carcinoma. Forty-three differentially expressed proteins were characterized by 2D-PAGE and MS in squamous lung cancer. Some were found to be related to oncogenes, whereas others were involved in cell cycle and signal transduction regulation [47, 48]. When 2D-PAGE profiles of normal, metaplasia, dysplasia, and carcinoma tissues of human bronchial epithelia were used to investigate successive steps of carcinogenesis in squamous carcinoma, 23 protein spots were identified in dysplasia and invasive carcinoma groups, including various kinases and kinase inhibitors, metalloproteinase receptors, and tumor antigens [49]. Using proteomic technologies, heterogeneous nuclear ribonucleoprotein (hnRNP) B1 was found to be specifically overexpressed in squamous cell carcinoma, stage I lung cancer, dysplasia, but not in normal bronchial epithelium [50]. It was suggested that overexpression of hnRNP B1 occurred in the early stage of carcinogenesis and inhibited DNA-PK activity. This interaction resulted in subsequent accumulation of erroneous rejoining of DNA double-strand breaks, thus causing tumor progression [51].
  2. 1.1.2. Lung Adenocarcinoma. 2D-PAGE and MS were used to identify 9 protein isoforms that were overexpressed in lung adenocarcinoma, including antioxidant enzyme AOE372, ATP synthase subunit d (ATP5D), ^1,4 galac-tosyltransferase, cytosolic inorganic pyrophosphatase, glucose-regulated 58-kDa protein, glutathione-S-transferase M4, prolyl 4-hydroxylase ^-subunit, triosephosphate isomerase, and ubiquitin thiolesterase [52]. Proteomic approaches also showed that napsin A, a member of the aspartic proteinase family, was expressed at various levels in primary lung adenocarcinomas and that this protein might be useful in degree of differentiation of primary lung adenocarcinoma as well as distinguishing primary and metastatic adeno-carcinoma [53]. In another study, systematic identification of lung adenocarcinoma proteins by 2D-PAGE and MS uncovered numerous cytokeratins isoforms. For example, 14 of 21 isoforms of cytokeratin 7, 8, 18, and 19 occurred at significantly higher levels in tumors compared to uninvolved adjacent tissue. Specific isoforms of four cytokeratins correlated with either clinical outcome or individual clinical-pathological parameters. Interestingly, all five of the CK7 isoforms associated with patient survival represented cleavage products [54]. Overexpression of truncated forms of cytokeratins 6 and 8 was also found by others, thus confirming proteolytic processing steps in tumor material [55]. Taken together, these studies suggest that specific isoforms/cleavage products of individual cytokeratins may have utility as diagnostic or predictive markers in lung adenocarcinomas.
  3. 1.2. Colorectal Cancer

CRC is the second most common cancer worldwide. About 10 years ago, our group published the first 2D-PAGE map of purified colorectal epithelial cells [56]. At that time, we could identify about 50 polypeptides, most of them by N-terminal sequencing—since MS technology was only emerging. In the meantime, expression proteomic studies were carried out with cell lines, whole tissue biopsies, and purified epithelial cells of colorectal origin. 2D-PAGE reference maps, protein, and membrane protein databases are available on the internet (reviewed in [57]). It was possible to synthesize tran-slational research results obtained in CRC in a quasi-meta-analysis [58]: eight large-scale proteomic studies on CRC were retrieved. Out of 408 differentially expressed proteins, 83% were found to be differentially expressed only in a single study, 16 proteins in 3 studies, 10 in 4 studies, 3 in 5 studies, and only a single protein in 8 studies. Confirmation at proteome level using large-scale transcriptomic studies was possible in only 25%. This proportion was higher (67%) for confirming proteome results using transcriptomic technologies. Obviously, reproducibility and overlap between published gene expression results at proteome and transcriptome level are low in human CRC. Importantly, the total number of patients involved in the proteomic studies was only 11, a surprisingly low figure.

Using SELDI technology, a-defensin isoforms were found to be elevated in serum from colon cancer patients and in protein extracts from CRC [59]. This result was confirmed by expression analysis of microarray data obtained from 283 tumors and normal tissues followed by serum analysis of colon cancer patients and controls by ELISA. This study yielded a diagnostic sensitivity of 70% and specificity of 83% for a-defensin in colon cancer [60]. Although these figures appear too low for developing a screening test, this result is an interesting proof of concept for integrating tissue transcriptomic data with serum protein analysis as a means to discover serum biomarkers. Another study in CRC tissue combined 2D-PAGE with SELDI-MS. This study demonstrated that PACAP protein, hnRNP A1, flavin reductase, cal-gizzarin, NDK B (NM23-H2), cyclophilin A, and smooth muscle protein 22-a showed significantly different levels. Subsequent immunohistochemical analysis of tissue distribution and subcellular localization of some of the differentially expressed proteins demonstrated alterations in subcellular protein distribution [61].

AACT technology (see above) was applied to identify differentially regulated protein isozymes in a CRC cell line (DLD-1). For this purpose, deuterium-labeled (heavy) amino acids were incorporated into the proteome of p53-induced DLD-1 cells, whereas DLD-1 vector cells (controls) were grown in the unlabeled medium. In high-throughput LC-ESI-MS/MS analyses, the AACT-containing peptides were paired with their unlabeled counterparts, and their relative spectral intensities, reflecting the differential protein expression, were quantified. The p53-regulated proteins were associated with several distinct functional categories: cell cycle arrest and p53 binding, protein chaperoning, plasma membrane dynamics, stress response, antioxidant enzymes, and anaerobic glycolysis. This result suggested that p53-induced apoptosis involves the systematic activation of multiple pathways that are glycolysis-relevant, energy-dependent, oxidative stress-mediated, and possibly mediated through interorganelle cross talks [62].

5.1.3. Gastric Cancer

Primary biomarker screening was performed in gastric cancer using 2D-PAGE on purified gastric epithelial cells obtained with an epithelial cell enrichment technique from gastrectomy specimens. One hundred and ninety-one differentially expressed protein spots were identified by MS. Overexpression of cathepsin B was observed in most cancer tissue samples, as was serum level of cathepsin B, compared to healthy controls. Elevated serum levels were associated with a reduced survival rate, enabling the classification of some gastric cancer patients into a subgroup that should undergo aggressive therapy [63]. Then, using SELDI-TOF-MS, our group identified a single best mass that could separate gastric cancer from patients without cancer, with a sensitivity of 89% and a specificity of 90%. This peak was identified as thrombin light chain A, a proteolytic fragment of prothrombin, indicating that disturbances in the coagulation system are early events in gastric cancer biology and that a decrease or loss of thrombin light chain A may contribute to the diagnosis of cancer patients [64]. Importantly, cancer diagnosis was possible by a loss of protein expression.

5.1.4. Pancreatic Cancer

Because differential diagnosis of pancreatic cancer and chronic pancreatitis is not possible without surgery, there is an important medical need for novel biomarkers in pancreatic diseases. Global proteomic approaches in pancreatic cancer have been reviewed [65]. For example, 2D-PAGE and MS were used to compare cancer vs normal and pancreatitis protein patterns. Differentially expressed proteins included antioxidant enzymes, chaperones, and/or chaperone-like proteins, calcium-binding proteins, proteases, signal transduction proteins, and extracellular matrix proteins. Among these proteins, annexin A4, cyclophilin A, cathepsin D, galectin-1,14-3-3 x, a-enolase, peroxiredoxin I, TM2, and S100-A8 were specifically overexpressed in tumors [66]. Using SELDI technology, serum samples from patients with resectable pancreatic adenocarcinoma were compared with samples from matched patients with nonmalignant pancreatic diseases as well as healthy controls. Two peaks of interest were identified in this study and were found to provide improved diagnostic sensitivity (78%) and specificity (97%) when compared to the traditional CA19-9 serum marker [67]. Another SELDI-based study compared serum obtained from pancreatic cancer patients. This study found a panel of six unidentified peptides that provided a diagnostic sensitivity of 89% and a specificity of 74% [68]. One additional pancreatic cancer study found a circulating antigen identified as DEAD-box protein 48. Interestingly, this protein is highly similar to eukaryotic initiation factor 4A that plays a role in pre-mRNA processing [69].

5.1.5. Liver Cancer

Expression proteomic analysis of human hepatocellular carcinoma (HCC) was conducted using 2D-PAGE coupled with MS. A panel of proteins of interest were identified from well-differentiated and poorly differentiated tumor samples, including methionine adenosyltransferase, glycine N-methyltransferase, and betaine-homocysteine S-methyltransferase. These enzymes, involved in the methylation cycle in the liver, influence the level of S-adenosylmethionine (AdoMet), and chronic deficiency in AdoMet in the liver results in spontaneous development of HCC in knockout mice deficient in methionine adenosyltransferase [70]. This study is providing a good example of successful functional confirmation in the animal model of an expression proteomic result obtained in the human being. Analysis of human liver samples with 2D-PAGE was also used to identify proteins that could be molecular targets for diagnosis and treatment of Hepatitis C virus-related he-patocellular carcinoma. One of the numerous spots that showed stronger intensity in tumorous samples was identified as a-enolase, a key enzyme in the glycolytic pathway. Expression of this protein increased with tumor dedifferentiation, suggesting that a-enolase is a biomarker for tumor progression [71]. a-Enolase has been reported as a protein associated with lung and pancreatic cancer (see above).

ICAT and 2D LC-MS were used to investigate the qualitative and quantitative proteomes of HCC, following LCM. A total of 644 proteins were identified, out of which 261 proteins showed differential expression [72]. In another ICAT study, hepatocytes labeled with different reagents were stimulated with interferon-a or Hepatitis C virus and compared with corresponding controls. Samples were combined, trypsinized, and subject to cation exchange and avidin affinity chromatographies. The resulting cysteine-containing peptides were then analyzed by microcapillary LC-MS/MS. More than 1200 proteins or related protein groups could be identified [73]. Direct glycoproteins targeting allowed, in combination with 2D-PAGE, to discover higher levels of a-1,6-linked fucose in rat HCC and hyperfocu-sylation of a glycoprotein, Golgi Protein 73 (GP73) in the serum of patients suffering HCC [74]. SELDI-TOF-MS was used for protein fingerprint of 106 serum samples form subjects with liver cancer, cirrhosis, and healthy individuals, and patterns analyzed with an artificial neural network: the sensitivity and specificity were 88 and 95%, respectively, which represents an improvement compared with the traditional methods [75].

5.1.6. Breast Cancer

Several proteomic technologies have been used to uncover biomarkers and molecular mechanisms associated with breast carcinoma, the commonest cancer in women (reviewed in [76]). For example, 2D-PAGE combined with MS analyzed changes in the proteome of infiltrating ductal carcinoma compared to normal breast tissue. Twenty-five differentially expressed proteins could be identified, comprising cell defense proteins, enzymes involved in glycolytic energy metabolism and homeostasis, protein folding and structural proteins, and proteins involved in cytoskeleton and cell motility. Further proteins were also mapped to establish a 2D-PAGE reference map of human breast cancer [77]. Another proteomic study, combining 2D-PAGE, MS, immunoblotting, and antibody arrays analyzed the proteome from adipose cells and interstitial fluid collected from mastectomy specimens of high-risk breast cancer patients to detect factors present in the tumor microenvironment and responsible for tumor growth and progression. A total of 359 unique proteins were identified, including numerous signaling molecules, hormones, cytokines, and growth factors involved in a variety of biological processes such as signal transduction and cell communication; energy metabolism; protein metabolism; cell growth and/or maintenance; immune response; transport; regulation of nucleobase, nucleoside, and nucleic acid metabolism; and apoptosis [78]. This proteomics study provided a unique phenotypic overview of tumor microenvironment in human epithelial cancer.

Using SELDI-TOF, it was shown that combined measurement of serum complement component C3a(desArg) and a C-terminal-truncated form of C3a(desArg) significantly differentiates breast cancer patients from noncan-cer controls [79]. In a confirmatory study on independent samples, C3a (desArg) appeared to lack specificity among patients with benign diseases [80]. This work could be partially validated in an independent prospective study where some peaks of interest could be recovered, but the sensitivity for cancer detection was only between 33 and 45% [81]. SELDI-TOF was also applied to the analysis of breast ductal lavage and was found to enhance the potential of cytology [82]. In another study looking for circulating autoanti-bodies in breast cancer patients, 15 proteins were repeatedly immunodetected in breast cancer patients and controls, some isoforms being preferentially immunodetected by breast cancer sera [83].

Proteomic studies could also invalidate putative results obtained with transcriptomic technologies. For example, using a proteomic approach complemented by immunohistochemical analysis, it was demonstrated that levels of expression of 14-3-3 sigma were similar in matched malignant and non-malignant breast epithelial tissue. Besides its biological significance, the methodological relevance of this finding should be emphasized, since tran-scriptional expression of the sigma isoform of 14-3-3 is frequently impaired in human cancers, including breast, which has led to the suggestion that this protein might be involved in the neoplastic transformation of breast epithelial cells [84].

5.1.7. Ovarian Cancer

Ovarian cancer continues to be the leading cause of death from gynecologic malignancies because of the inability to identify disease at a stage when surgical therapy could be curative. This has driven a search for newer methods of detection of early ovarian cancer. For example, a 2D-PAGE study determined protein patterns associated with a predisposition to develop ovarian cancer in ovarian surface epithelium obtained during prophylactic oophorec-tomy in high-risk female patients. Eight proteins altered in high-risk patients were identified: three were already known as ovarian tumor-associated proteins, providing a proof of concept, whereas five were novel findings, representing potential early markers for the evaluation of the risk of developing ovarian cancer [85]. Numerous groups are pursuing similar serum-based approaches to ovarian cancer diagnosis (reviewed in [86]): characteristic SELDI-TOF and MALDI-TOF mass spectral patterns could be identified in sera of patients that may yield a sensitive and specific signature for ovarian cancer [87]. With refining analysis by quadrupole tandem LC-MS/MS, the sensitivity and specificity results of models now being generated are consistently 100 and 100%, respectively. Over 1000 proteins and peptides that may be selective or specific to ovarian cancer have now been sequenced from serum samples from women with early and advanced stage ovarian cancer [88]. Reverse-phase microarrays were applied for proteomic mapping of phos-phorylation end points in the metastatic tissue of ovarian cancer patients, and a reference standard based on a mixture of phosphorylated peptides was developed [89].

5.1.8. Cervical Cancer

In cervical cancer of the uterus, a 2D-PAGE study focused on the viral carcinogenesis characterized tumor-related proteins that were identified as oncogenes products, cell cycle-regulating proteins, genome-stabilizing, telomerase-activating, and cell-immortalizing proteins. The data have been synthesized from a human cervix cancer proteome database [90].

5.1.9. Endometrial Cancer

SELDI technology and peptide mass fingerprinting have also been applied in endometrial cancer. Two proteins, EC1 and EC2, were altered during carcinogenesis of the human uterine endometrium [91]. By improving analysis by quadrupole TOF-MS, it was possible to resolve two further potential biomarkers for endometrial cancer, chaperonin 10 and calgranulin A, in tissue homogenates [92]. Nine potential markers for endometrial cancer have been similarly discovered using a combination of differentially labeled tags, iTRAQ and cICAT (see above) with multidimensional LC and tandem MS. These include chaperonin 10, pyruvate kinase M1 or M2 isozyme, calgizzarin, heterogeneous nuclear ribonucleoprotein D0, macrophage migratory inhibitory factor, and polymeric immunoglobulin receptor precursor; underexpressed proteins were a-1-antitrypsin precursor, creatine kinase B, and transgelin. All these markers are known to be associated with various forms of cancer [93].

5.1.10. Prostate Cancer

The use of proteomics in prostate cancer, the most common cancer in men, was reviewed [94]. In vitro studies on prostate cell lines provided positive proof-of-principle results in the identification of novel biomarker when proteomics was utilized to query prostate tissue specimens. 2D-PAGE followed by MS was used to investigate protein profiles in voided urine after prostatic massage in patients with prostate cancer or with benign prostatic hyperplasia [95]. In this study, a potential novel cancer biomarker, calgranulin B/MRP-14, was identified. In another study, agarose 2D-gel electrophoresis followed by LC-MS was applied to differentiate high molecular mass and alkaline protein patterns of androgen-dependent and -independent prostate cancers. A total of 295 proteins were identified representing 91% of excised spots. Eighteen proteins were regulated when the tumor became androgen-independent, including radical scavenger enzymes such as antioxidant protein 2, superoxide dismutase 1, thioredoxin peroxidase; GTP-binding protein ,3-chain homologue; and the ha1225 gene product [96].

2D-DIGE technology (see above) was applied to analyze the phosphopro-tein fractions in prostate cancer cells transfected with prostate collagen triple helix (PCOTH), a growth-promoting gene. The phosphorylation level of oncoprotein TAF-I^/SET was significantly elevated in prostate cancer cells transfected with PCOTH. In combination with siRNA findings, this observation suggested that PCOTH is involved in growth and survival of prostate cancer cells and may, in part, be related to the TAF-I^ pathway. This study implied that PCOTH might also be a drug target in prostate cancer [97], thus emphasizing importance of combining proteomic and siRNA technologies.

In prostate cancer, ICAT was used for assessing quantitative profile changes in the proteome of cultured cells in response to androgens. Changes in levels of 77 proteins in response to androgens were detected including spermine synthase, fatty acid synthase, and calreticulin precursor. A large number of proteins that had not been previously reported to be expressed in prostate cells were also quantitatively identified including members of the dual specificity protein phosphatase subfamily, "similar" to hypothetical protein DKFZp434B0328.1, to 14-3-3 protein to hypothetical protein 458 and actin components [98].

MS-based mass profiling combined with multivariate analysis identified platelet factor 4, a chemokine with prothrombolytic and antiangiogenic activities, as a diagnostically predictive protein in depleted serum of prostate cancer patients [99]. SELDI-TOF-MS was applied to the discovery of serum markers of bone metastasis in prostate cancer. Unique isoforms of serum amyloid A were identified in these patients. Machine-learning algorithms were used to identify these patients with a sensitivity and specificity of 89% [100].

Differences in the expression of cell surface proteins between "normal" prostate epithelial and prostate cancer cells were investigated using combined affinity chromatography of biotin-tagged surface proteins with MS. This analysis identified 26 integral membrane and 14 peripheral surface proteins, including ALCAM/CD166, ephrin type A receptor, EGF-R and prostaglan-din F2 receptor regulatory protein, the voltage-dependent anion selective channel proteins porins 1 and 2, ecto-5'-nucleotidase (CD73), and scavenger receptor B1. Costimulation with type I and II interferons had additive or synergistic effects on the membrane density of several of the peripheral surface proteins [101].

5.1.11. Bladder Cancer

Bladder cancer is the fifth most common malignancy in the world and represents the second most common cause of death among genitourinary tumors. Adipocyte fatty acid-binding protein (A-FABP) has been identified as a marker of progression in bladder cancer in a large-scale proteomics-based study. In this study, tumor profiling showed a substantial downregula-tion of A-FABP in invasive lesions. Results were confirmed by using a tissue microarray containing over 2000 samples. This work provided strong evidence that deregulation of A-FABP may play a role in bladder cancer progression and suggested a significant prognostic value for this marker [102]. In another comparative proteomic analysis in bladder cancer, a significant downregulation of S100C was found in invasive tumors (vs superficial tumors) and was associated with poorer survival [103].

5.1.12. Leukemia

2D-PAGE combined with advanced MS was used to profile proteins in acute leukemic cells. In this study, heat-shock 27-kDa protein 1 as well as other proteins were found to be highly expressed and may play a role in distinguishing this cancer from acute myloid leukemia. Another set of upre-gulated proteins was restricted to granulocytic lineage leukemia. High-level expression of NM23-H1 was associated with favorable prognosis. Thus, the protein patterns obtained could be used as an analytical tool for facilitating molecular definition of human classification of acute leukemia [104].

5.1.13. Lymphoma

An application of ICAT technology in cancer was to study the activation of p38 mitogen-activated protein kinase (MAPK) in follicular lymphomas. MAPK is a key mediator of stress, extracellular-, growth factor-, and cytokine-induced signaling, and has been implicated in the development of cancer. Inhibition of MAPK results in dose- and time-dependent caspase-3-mediated apoptosis [105]. Differential proteomic analysis using ICAT-LC-MS/MS of inhibited MAPK cells identified about 300 proteins belonging to diverse biochemical pathways. These included IL-6/phosphatidylinositol 3-kinase, insulin-like growth factor 2/Ras/Raf, WNT8d/Frizzled, MAPK-activated protein kinase 2, and nuclear factor kB. The differential phosphorylation status of selected kinase active proteins could be validated by Western blot analysis.

5.1.14. Melanoma

The plasma peptide component from 10 melanoma and healthy individuals was examined by a combination of RP-HPLC, SELDI-TOF-MS, and tandem MS. Fibrinogen a and inter-a-trypsin inhibitor heavy chain H4 fragments were absent in tumor samples [106].

5.1.15. Pheochromocytoma

The hypothesis that pheochromocytoma stages can be reflected by low molecular weight biomarkers was tested using peptide-profiling pattern recognition algorithms. It was possible to identify combinations of molecules that could distinguish all metastatic from all benign pheochromocytomas in a separate blinded validation set [107].

5.2. Functional Proteomic Studies

Most functional proteomic studies have been performed in cancer cell lines, that is, after exposition to toxicants, RNA inhibition, differentiation agents, viral transfection, and so on. These studies covered several aspects of mutagenesis, tumor promotion, and progression. In the recent years, it has been shown that repeated analysis of the proteome at different tumor stages also deliver distinct patterns and thus a functional picture of disease progression at the molecular level.

5.2.1. Signaling Pathways

Reverse-phase protein array technology has been applied to analyze the status of key points in cell signaling involved in prosurvival, mitogenic, apoptotic, and growth regulation pathways in the progression from normal prostate epithelium to invasive prostate cancer. using multiplexed reversephase protein arrays coupled with LCM, the states of signaling changes during disease progression from prostate cancer study sets were analyzed [108]. After differentiation with dimethylsulfoxide (DMSO), 2D-PAGE followed by MS was used to identify signaling proteins in differentiated vs undifferentiated cell lines. Differences in expression of GTP-binding/Ras-related proteins, kinases, growth factors, calcium-binding proteins, and phosphatase-related proteins were observed. Most signaling proteins were upregulated in differentiated cells, whereas only eight such proteins were observed in undifferentiated cells. The on/off switching profiles of several individual signaling proteins from different signaling cascades is likely the key to understanding biochemical mechanisms involved during the differentiation process [109].

proteomics in cancer

5.2.2. Cell Cycle

Using 2D-PAGE, cell cycle entry has been shown to be associated with a significant increase in p27(kip1) phosphorylation in human primary B lymphocytes. Detailed analysis revealed that different cyclins and cyclin-dependent kinases interact with distinct posttranslationally modified isoforms of p27(kip1) in vivo. These results have to be interpretated in the context of overexpression of cyclin D3 in the presence of high levels of p27(kip1) in human B-cell lymphomas with adverse clinical outcome [110].

5.2.3. Oxidative Stress

A cellular prooxidant state promotes cells to neoplastic growth in part because of modifications to proteins and their functions. Reactive nitrogen species formed from nitric oxide (No) or its metabolites can lead to protein tyrosine nitration that is elevated in some cancers. Using 2D-PAGE and MS in a lung cancer cell line exposed to No, more than 25 nitrated proteins were identified, including metabolic enzymes, structural proteins, and proteins involved in prevention of oxidative damage. These alterations may contribute to the mutagenic processes and promote carcinogenesis [111].

5.2.4. Chaperone Proteins

Although chaperones play an important role in tumor biology, no systematic work on their expression patterns had been reported thus far. 2D-PAGE combined with MS was used for the concomitant determination of several chaperones in 10 human tumor cell lines. Human tumor cell lines of neuro-blastoma, colorectal cancer, adenocarcinoma of the ovary, osteosarcoma, rhabdomyosarcoma, malignant melanoma, lung, cervical and breast cancer, promyelocytic leukemia were analyzed. The main chaperone groups included HSP90/HATPasC, HSP70, Cpn60_TCP1, DnaJ, Thioredoxin, TPR, Pro_ isomerase, HSP20, ERP29_C, KE2, Prefoldin, DUF704, BAG, GrpE, and DcpS. The 10 individual tumor cell lines showed different expression patterns. These important results have served as a reference map of chaperones and isoforms in human cancers [112].

5.2.5. Viral Oncogenesis

Comparison of global patterns of protein expression can be achieved in cell lines through the use of cDNA microarray and proteomic techniques. This approach has provided in-depth information on the impact of HPV-16 E6 on gene expression and served as a valuable resource for investigation of the biochemical basis of uterine cervical carcinogenesis. In three cases (CDK5, Bak, and I-TRAF), expression was matched in both analyses of cDNA microarrays and proteomics [113].

5.2.6. P53 Function

The p53 gene is a transcription factor essential for DNA damage checkpoints during cellular response to stress. Mutations within this gene are the most common genetic alterations found in human cancer. Although most pathogenetic modifications are missense mutations that abolish the p53 DNA-binding function, some p53 mutations may determine different phenotypes in distinct cell types. 2D-PAGE analyses in a thyroid cancer cell line indicated that expression of a significant portion (up to 25%) of protein species were modified by p53 mutants. Several of these proteins were identified by MS procedures, including HSP90 and T-complex proteins. Interestingly, these proteins were already known to be related to p53 function [114].

5.2.7. Mitochrondrial Dysfunction in Cancer

Bioenergetic dysfunction of mitochondria has been reported as a hallmark of many types of cancers (i.e., downregulation of ATP synthase ^-subunit expression in liver, kidney, colon, squamous esophageal, and lung carcinomas, as well as in breast and gastric adenocarcinomas). ATP synthase d-subunit was found to be associated with chemoresistance to 5-fluorouracil (5-FU) in CRC using 2D-PAGE. In a functional assay, suppressed ATP synthase d-subunit expression by siRNA transfection increased cell viability in the presence of 5-FU [115].

5.2.8. Chromatin-Associated Proteins

High-mobility group A (HMGA) proteins are nonhistone architectural nuclear factors that play a general role in chromatin dynamics. These proteins were analyzed by a combination of affinity chromatography, 2D-PAGE, and MS. About 20 putative HMGA proteins were identified and assigned to three different classes: mRNA-processing proteins, chromatin-remodeling related factors, and structural proteins. These experimental data indicated that HMGA proteins were highly connected nodes in the chroma-tin protein network. Because these proteins are strongly implicated with cancer development, the identification of molecules able to perturb the HMGA molecular network may provide an attractive tool to interfere with oncogenic activity [116].

5.2.9. Cell Invasion

Malignant gliomas (astrocytomas) are lethal invasive brain tumors. Invasive cell migration is initiated by extension of pseudopodia into interstitial spaces. In a DIGE technology study (see above), pseudopodia of glioma cells were harvested and their protein profile compared with the profile of whole cells. Increased pseudopodial constituents were identified as actin, hepatocyte growth factor, hepatocyte growth factor receptor, isoforms of annexins I and II, as well as several glycolytic enzymes [117]. These proteins may represent potential therapeutic targets to suppress tumor invasion.

Smoking Solutions

Smoking Solutions

How To Maintain The Stop Smoking Pledge From Your New Year’s Resolution. Get All The Support And Guidance You Need To Be A Success At Quitting Smoking. This Book Is One Of The Most Valuable Resources In The World When It Comes To How To Maintain The Stop Smoking Pledge From Your New Year’s Resolution.

Get My Free Ebook

Post a comment