By Professor Maxey Chung, Associate Professor, National University of Singapore, Singapore.
Dated: 1/1/2008
The current thinking is that a panel of multiple biomarkers, rather than a single biomarker, will be needed for complex, multi-gene diseases like cancer. Proteomics, being a high throughput technology that analyzes proteins at the global level, is well-suited to this task.
By Professor Maxey Chung, Associate Professor, Departments of Biochemistry, Yong Loo Lin School of Medicine & Biological Sciences, Faculty of Science, National University of Singapore, Singapore.
Cancer is a major public health problem in the United States as well as in many other developed countries worldwide. According to CA Cancer J Clin 2006, cancer killed 6.7 million people in 2002 worldwide and this number is estimated to rise to 10.3 million in 2010 despite advances in cancer treatment. The high morbidity of the disease is largely due to late diagnosis attributable to the lack of specific and sensitive biomarkers for early detection and monitoring of disease progression.
Among the important tools critical to detection, diagnosis, treatment, monitoring, and prognosis are biomarkers. Biomarkers are biological molecules produced in our bodies that are indicative of our physiological health and change during a disease process. They are essential for screening, diagnosis, prognosis, theranostics, monitoring response to treatment, and detection of disease recurrence.
Cancer biomarkers must have high specificity and sensitivity, and be detectable early in the disease process. A common biomarker currently in use for hepatocellular carcinoma is alpha feto-protein (AFP). However, 70% of the patients have AFP levels below the diagnostic range of 500ng/ml whilst patients with benign chronic liver disease can have elevated levels of AFP. This single marker thus scores low in terms of sensitivity (52%) and specificity (84%), respectively. There is therefore, a need for more sensitive and specific biomarker (or biomarkers) for cancer detection. The current thinking is that a panel of multiple biomarkers, rather than a single biomarker, will be needed for complex, multi-gene diseases like cancer. Proteomics, being a high throughput technology that analyzes proteins at the global level, is well-suited to this task.
Proteomics
Proteomics refers to the analysis of the entire PROTEin complement expressed by a genOME (PROTEOME). In contrast to the genome, which is a rather constant entity, the proteome is dynamic. It differs from cell to cell and is constantly changing through its interactions with the genome and the environment, in physiological health as well as in disease states. Moreover, proteins, not genes, are the functional workhorses of a cell. At the protein level, distinct changes occur during the transformation of a healthy cell into a neoplastic cell, including differential expression and post-translational protein modifications, changes in specific activity, and aberrant localization, all of which may participate in the carcinogenesis process. Identifying and understanding these changes is the underlying theme in cancer proteomics.
Proteomic technologies allow for identification of the protein changes caused by the carcinogenesis process in a relatively accurate manner. Whilst tissue biopsies are the ideal source of cancer biomarkers, they have limited utility for screening, diagnostic, and monitoring purposes due to their invasive nature. In addition, subtle, invisible changes in early cancers may be missed during sampling, leading to failure to detect early cancers using biomarkers found in tissue biopsies. On the other hand, the diagnostic utility of serum/plasma as a non-invasive test holds tremendous potential. However, the discovery of biomarkers from blood remains an enormous challenge with 99% of the serum proteome comprising of 22 high abundant proteins.
The biomarkers of interest, representing disease proteins secreted/shed by the cancer tissue, are in the minority 1%. Considering that early tumors are usually 1.5–2 cm in diameter (corresponding to volumes of 4–9 cm3), these would represent less than 0.006% of an individual’s total body volume. Thus any disease biomarkers secreted by the tumors would be severely diluted and “lost” when sampled from the circulatory system encompassing of a total circulating volume of 5 L of blood, according to experts in Mol. Cell. Proteomics 2005.
Biological fluids represent another potential reservoir of biomarkers for the early diagnosis of various cancers. It is the most proximal fluid bathing diseased cells and is believed to closely reflect the ensemble of proteins represented in the tissue. Cancer markers that are secreted/shed by the tissues into their proximal fluid would be in much higher concentration here than in the serum. Therefore, regional detection of cancer biomarkers from biofluids would have far greater sensitivity than detection of these biomarkers diluted in the circulation.
Proteomics strategies have largely been the central focus of cancer biomarker studies for biofluids. This stems from the fact biological fluids such as saliva, gastric juice, urine, nipple aspirate, amniotic and cerebrospinal fluids, etc., do not have corresponding transcriptomes in which cancer gene expression can be measured.
Proteomics platforms
The Oncoproteomics Centre (OPC) housed at the Department of Biological Sciences, National University of Singapore (NUS), has established proteomics platforms to perform high-throughput research towards the discovery of novel biomarkers and drug targets from tissues, serum and biological fluids in various cancers. The Centre has the aim to develop early diagnostic kits and novel anticancer drugs. Established platform technologies include:
1. Quantitative proteomics that is based on
(a) 2D-difference gel electrophoresis (2D-DIGE),
(b) isotope-coded affinity tags (ICAT), and
(c) isobaric tags for relative and absolute quantification (iTRAQ).
2. Shotgun proteomics that use a 2D-LC system to separate peptide mixtures by strong-cation-exchange followed by reverse-phase columns.
3. Protein identification by MALDI-TOF/TOF mass spectrometry. This is a state-of-the-art mass spectrometer that provides both MS and MS/MS data for protein identification.
4. Bioinformatics and data mining. A database system based on the PEDRo standard called SPLASH, has been developed in-house to provide proteomics researchers a common platform to store, manage, search, analyze, and exchange their data. This system contains three modules - data maintenance, data search, and data mining.
5. Other techniques. Depending on the sample selected and the method of preparation, we could also employ the following methods for the project:
(a) Sub-fractionation of cell lysates using either affinity, 1D- or 2D-liquid chromatography for glycoproteins, DNA binding proteins, etc,
(b) Subcellular proteomics (or subproteomics) for profiling the proteins present in the subcellular structures of a given cell type, e.g., plasma membrane and secretomes,
(c) Serological screening of tumor antigens using autoantibodies from autologous patients, and
(d) Serum peptidome patterns obtained by identifying distinctive cancer-specific peptide signatures in serum through mass spectrometry (MS).
Potential biomarkers identified will be further validated using tissue microarrays and functional assays.
Researcher 1
Researcher 2
ABI 4800 MALDI TOF/TOF Analyzer
IPGphors
Nano-flow 2D-LC system
|
