What Has Been Achieved in Cancer Proteomics

6.1. Understanding Cancer

Large-scale proteomic studies have addressed on a molecular basis the complexity of the alterations associated with transition from benign to malignant tumor cells that lead to appearance and progression of cancer. For example, proteomic-based studies of the effects of transforming growth factor (TGF)-^ in early and late stages of carcinogenesis have widened our knowledge of TGF-^-dependent regulation of cell proliferation, apoptosis, DNA damage repair, and transcription (reviewed in [118]). Another example of the utility of global protein expression approaches in understanding carcinogenesis is demonstrated by an extensive comparative survey of alterations found in breast epithelium during malignant transformation [119]. These include differences in levels of key regulators of the cell cycle, signal transduction, apoptosis, transcriptional regulation, and cell metabolism. The impact of proteomic approach was also demonstrated in a study that compared the similarities and differences between inflammation and cancer [120]. This unique study examined the expression of intracellular proteins in a regressive cancer cell line and a progressive inflammatory cell-promoting cancer cell line.

Thus, simultaneous monitoring of thousands of proteins in cancer has uncovered novel signaling mechanisms and unravelled a series of complex processes involving multiple changes in protein expression, abundance, and function. In addition to isolated proteins, protein-protein interactions are fundamental to carcinogenetic processes. A comprehensive determination of all important protein-protein interactions that occur within a given tissue (as an integrated system) is critical to ultimately providing a fundamental understanding of cancer biology. The availability of genome-scale sets of cloned open-reading frames has facilitated systematic efforts at creating proteome-scale datasets of protein-protein interactions that are represented as complex networks or "interactome" maps in normal and cancer patients. Although far from complete, currently available maps provide insight into how biochemical properties of proteins and protein complexes are integrated into normal and cancerous tissues [121]. Taken together, theses studies document a significant role for proteomics in basic research and for identifying mechanisms of carcinogenesis. It should be noted, however, that this approach is viewed as being complementary to others.

6.2. Screening and Diagnosis

On the basis of the numerous studies described above, it is reasonable to claim that the proteomic approach, using 2D-PAGE, SELDI-TOF, or other emerging technologies, has the potential to identify novel biomarkers in human cancer. For example, peptide biomarkers were overwhelmingly (88%) viewed as likely to have diagnostic application in oncology in a survey of experts from the pharmaceutical industry [122]. However, the numerous expression research results described above have not yet been translated into in vitro diagnostics tests. So far, it has only been demonstrated that a limited panel of protein or peptide biomarkers can improve the specificity of available tests and thus has the potential of facilitating early diagnosis of cancer.

This gap between expectations and actual practice might have technical reasons:

  • There is a relative lack of experience with the technology of proteomics compared to transcriptomics and genomics,
  • Only increasing understanding of complexity, dynamic range, incomplete sampling, false positive matches, and integration of diverse datasets for plasma and serum proteins will lay the foundation for validation of novel in vitro diagnostics in cancer [123],
  • Concerns remain about SELDI-TOF technology because of the lack of sequences and consequently of a biological hypothesis, the discrepancy between the sensitivity of SELDI-MS compared to the concentration of known cancer biomarkers, and possible bioinformatics overfitting of MS data,
  • Protein microarray assay or multichannel ELISA is still not widely established.

Indeed, detection and diagnosis of cancer can only rely on an integrated approach using clinical history, physical examination, imaging, and his-topathology. Some bias in study design can also have prevented a large acceptance of diagnostic research results by the medical community:

  • Most studies have been performed by biochemists and the clinical aspect has been neglected,
  • For most biomarkers discovery studies, sample size was not large enough,
  • Choice of control groups was inadequate or incomplete,
  • No confirmatory studies were performed.

Moreover, regulatory requirements were not fulfilled by most research groups because they were not considered for study design. For regulatory agencies and pharmaceutical companies to be convinced that combining biomarkers performs better than the conventional, single-marker approach, it appears essential to fulfill requirements for in vitro diagnostics development such as:

  • Providing diagnostic tests that allow for definite and reliable diagnosis tied to a decision on interventions (prevention, treatment, or nontreatment),
  • Meeting stringent performance characteristics for each analyte (in particular test accuracy, including both precision of the measurement and trueness of the measurement), and
  • Providing adequate diagnostic accuracy, that is diagnostic sensitivity and diagnostic specificity, determined by the desired positive and negative predictive values which depend on disease frequency [124].

6.3. Prognosis

Although most follow-up tests in clinical oncology are based on the quantification of serum proteins, proteomic technologies have found only limited use for the development of novel prognostic biomarkers thus far. This is surprising since proteomic technologies appear particularly well suited for analyzing dynamic processes, for example, in clinical oncology where prognosis is changing all the time and particularly following surgery, but also after chemotherapy, depending on immune defenses, and so on.

Forecasting a particular patient's prognosis is best possible by defining tumor extension at the time of diagnosis, using the tumor node metastasis (TNM) staging system of the International Union against Cancer (UICC) [125]. This validated (but static) prognostic information is giving the framework for proteomic studies, and proteomics analysis (such as repeated measurement of serum biomarkers over the course of disease) is complementing this information. A proof of concept for the application of proteomics for prognostic purpose in oncology is available: patient survival has been shown to correlate with protein expression profiles in lung cancer [126]. Such dynamic information cannot be provided by transcriptomics studies in cancer tissue.

6.4. Therapy

About 40 small protein or peptide drugs are currently in clinical use with an estimated annual market of US$1 billion. These include insulin, parathyroid hormone (PTH), calcitonin, growth hormones, and so on [127]. In oncology, several antibodies directed against surface proteins have entered the market over the last years, including herceptin in breast cancer, cetux-imab in CRC, imatinib in gastrointestinal stromal tumors, and so on.

Proteomic technologies have been increasingly applied by pharmaceutical companies as the object of significant investments for discovering and validating small protein drugs and targets for monoclonal antibodies at various stages of development.

6.4.1. Target Discovery

The main application of proteomics in drug development is clearly the identification of drug targets. Pathological processes are reflected by characteristic alterations in the proteome. Thus, most pharmaceutical companies active in oncology have recognized the need to implement proteomic discovery strategies to initiate the drug discovery process. In the meantime, large datasets have been generated that now need to be interpreted and prioritized. For this purpose, pharmaceutical companies increasingly rely on the integration with bioinformatics for data interpretation, and on further validation of proteomic technologies for their application in various areas of drug development.

Indeed, there are still many technological challenges to meet the needs for high sensitivity, reproducibility, and throughput required for successful target discovery in cancer proteomics, including:

  • Current and partial view of the proteome,
  • Limitations for the analysis of hydrophobic proteins—significant for analysis of receptors and ion channels,
  • Large dynamic range of protein concentration in serum or plasma, thus preventing accurate separation of these complex biological mixtures,
  • Low reproducibility and overlapping between studies.

However, it is the hope that the rapid progress in sample preparation procedures and in proteomic technologies particularly improved mass spectrometer sensitivity will solve these problems.

6.4.2. Target Validation

Given a trend to evaluate patterns rather than single biomarkers, it is unclear how to define an acceptable validation strategy to date. This problem might represent a unique opportunity for proteomic technologies because they allow analysis of large sets of molecules at the functional level. There are two ways in applying proteomic technologies for target validation:

  • Functional studies in animal models for validation of proteins of interest in humans,
  • Expression studies in humans at different steps of carcinogenesis for validation of proteins in animal or in vitro models.

For validation purposes, proteomics can be associated with other functional tools. For example, proteomics combined with small interfering RNA

proteomics in cancer

(siRNA) elucidated the mechanism of action of paclitaxel, a potent drug of natural origin widely used in the treatment of ovarian, lung, and breast cancer. In a first step, 2D-PAGE and MS were used to investigate the mechanisms of action of paclitaxel on cervical carcinoma cell line carrying HPV. This approach demonstrated that treatment suppressed the expression of the mitotic checkpoint protein BUB3. In addition, this study identified several cellular proteins that were responsive to treatment in HeLa cells, namely apoptosis-related, immune response-related, and cell cycle checkpoint-related proteins. Paclitaxel treatment diminished growth factor/ oncogene-related proteins and transcription regulation-related proteins. In a second step, functional proteomic analysis by siRNA targeting was used to investigate the role of mitotic checkpoint protein BUB3 in cell cycle progression in herpes virus positive and negative cervix cancer cell lines. Paclitaxel showed antiproliferative activity through the membrane death receptor (DR)-mediated apoptotic pathway involving activation of caspase-8 with a TRAIL-dependent fashion as well as the mitochondrial-mediated pathway involving downregulation of Bcl-2 by cytochrome c release. This study showed the power of proteomic profiling combined with siRNA technology for better understanding of the actions of cancer drugs [128].

6.4.3. Analysis of Binding Sites

Proteomic technologies have been applied for fine-tuning drug development studies in cancer. This approach has been used in analyzing the binding sites of alkylating agents (i.e., microtubule disrupters) and thus powerful anticancer drugs. ,3-Tubulin was separated using 2D-PAGE from B16 cells incubated with an alkylating agent. Alkylated ,-tubulin had a lower apparent molecular weight and a more basic pi than the unmodified protein. Using MALDI-TOF-MS, it was possible to demonstrate that none of the cysteine residues of ,-tubulin was linked to the alkylating agent, as previously supposed. In contrast, a glutamic acid at amino acid position 198 was identified as target for alkylation via an ester bond with ICEU. This site should play an essential role in the conformational structure necessary for the interaction in the microtubule [129].

6.4.4. Response Prediction

Chemoresistance remains an unresolved problem in clinical oncology. This issue highlights the need for identifying molecular factors that lead to drug resistance in cancer cells. The hope is that a tumor biopsy can be analyzed to generate a molecular description of the tumor in addition to standard histopathology, allowing the selection of a therapy targeting specific molecular defects. Proteomic technologies have been used for studying global protein expression in chemosensitive and chemoresistant cancer cells to find candidate proteins that are associated with the drug-resistant phenotype. For example, molecular profiling of individual patient's tumors is currently being evaluated in clinical trials at the National Institutes of Health, National Cancer Institute for monitoring epidermal growth factor (EGF) cell signaling events for patients with breast and ovarian cancer [130], or at other institutions for patients with CRC [131]. Proteomic approaches have been proposed for selecting chemotherapy in neuroblastoma [132, 133], glioblastoma [134], for detecting proteins related to chemoresistance in cervix carcinoma [135], for discovering mechanisms of chemoresistance to 5-FU in CRC [136], or radioresistance-associated proteins in rectal cancer [137].

6.4.5. Therapy Monitoring

Following initiation of treatment, serum probes can be obtained to determine whether the tumor is responding to therapy, as well as determining whether the tumor has developed resistance mechanisms that may require modification of therapy, that is so-called responder profiling. Protein bio-markers such as CEA, CA 15-3, AFP, and PSA are already in clinical use for therapy monitoring, and it is reasonable that these biomarkers will be complemented by others in the future.

6.4.6. Prevention ofSide Effects

The need for predicting side effects in drug development and the limitations of conventional tests in animal models have been recently highlighted by life-threatening complications that occurred in healthy human volunteers during a stage-I oncology drug trial in Great Britain. Proteomic technologies are expected to play an important role for identifying patients at risk for experiencing side effects. For example, 88% of pharmaceutical experts considered that proteomic technologies are likely or very likely to play an important role in this respect [138]. In one study, SELDI technology was used to evaluate serum protein patterns in breast cancer patients after docetaxel infusion. The relative expression levels of target proteins were monitored following docetaxel injection. Two proteins, kininogen and apolipoprotein A-II, were correlated with adverse effects [139].

6.4.7. Toxicoproteomics

Toxicoproteomics is a new scientific discipline that combines proteomic technologies with bioinformatics. This approach provides a means to identify and characterize mechanisms of action of toxicants in carcinogenesis. In contrast to toxicogenomics, a discipline that determines genetic suspectibility of a particular individual following exposure to a carcinogenetic agent, toxicoproteomics allow the monitoring of the body's response to a particular toxicant. The current regulatory toxicological approach usually includes investigation of carcinogenicity, in generally lengthy (2 years) studies in rodents. This is especially true for detection of early protein biomarker signatures that precede neoplastic appearance [140]. Several examples demonstrate the potential of proteomic approaches to reduce time and expense of traditional carcinogenicity testing. For example, the liver carcinogen N-nitrosomorpholine (NNM) was investigated in rats to identify potential early protein biomarker signatures indicative of the carcinogenic processes. Analysis was performed 18 weeks following treatment revealed significant upregu-lation of stress proteins, including caspase-8 precursor, vimentin, and Rho GDP dissociation inhibitor. Interestingly, the proteins annexin A5 and fructose-1,6-bisphosphatase were found to be deregulated early after exposure. This finding may indicate their potential use as predictive biomarkers for early liver carcinogenicity. Another toxicoproteomics study was performed in municipal incinerator plant workers exposed to 2,3,7,8-tetrachlorodibenzo-^-dioxin (TCDD). TCDD is a chemical compound known to induce severe reproductive and developmental problems. Seven overexpressed proteins were identified in this study. Interestingly, the most overexpressed protein was identified as AFP, the classical serum marker for HCC. In cultured HepG2 cells, TCDD treatment resulted in increased mRNA and protein expression of AFP, but reduced albumin expression. This toxiproteomics study provided strong evidence that TCDD may induce liver cancer [141].

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment