Pierce - University of Georgia

Principal Investigator: J. Michael Pierce, Ph.D.
Institution: University of Georgia, Athens, GA

Our project, Discovery and Development of Cancer Glycomarkers, is a joint collaboration between our laboratories at the CCRC, which include Karen Abbott, Lance Wells, Kevin Dobbin, and Mike Tiemeyer, those at TGen, in Phoenix, AZ, Daniel Von Hoff, Haiyong Han, and Mike Demeure, and Caerus Discovery in Manassas, VA, which includes Cohava Gelber and S?ren Mogelsvang.

We have identified a novel N-glycan biomarker for pancreatic cancer defined by a suite of antibodies developed by Caerus Discovery. Screening >150 tissue sections by immunohistochemistry showed a high sensitivity and specificity for this marker in pancreatic carcinoma, even at early stages. Initial studies have detected the biomarker in serum from pancreatic cancer patients, suggesting that assays may be refined to develop it into a diagnostic test. Moreover, we are studying the glycomics and proteomics of stellate cells that are found tightly associated with pancreatic carcinoma and in many cases block chemotherapy from penetrating these tumors. These studies are likely to point to markers that can be used to develop additional molecular targets to aid in the diagnosis and treatment of pancreatic cancer.

In addition, we have identified a specific glycoform of periostin in tissues and in serum as a marker for invasive breast carcinoma. These tests are being expanded to screen large numbers of test and control sera to determine the efficacy of this serum marker. We have also developed a means to separate many GPI-anchored glycoproteins from serum and tissue. Two glycoproteins in serum associate with the presence of breast cancer. These biomarkers are being tested in sera from larger populations of patients and controls.

Many of our glycoproteomics technologies have been developed in our NIGMS National Center for Biomedical Glycomics at the Complex Carbohydrate Research Center, University of Georgia. We have state-of-the-art instrumentation and expertise to analyze glycan and glycopeptide structures and the expression of transcripts that regulate human glycans expression. The National Glycomics Center is focused on understanding the regulation of the glycome during embryonic stem cell, iPS cell, and cancer stem cell differentiation, as well as identifying potential cell surface biomarkers of these various cell types.

Synopsis of Research and Network Resources

1. Introduction: A Brief Summary of the State of the Science and Research Needs

Glycan expression on cell surfaces and on secreted glycoproteins clearly changes during oncogenesis and tumor progression. These changes range from mutations that eliminate particular glycosyltransferase activities, such as the T synthase that transfers to short O-linked glycans, to increases in the transcription of particular glycosyltransferases regulated by the activation of oncogenic signaling pathways. The serum markers in common clinical use to monitor cancer progression are glycoproteins, and newly discovered glycoprotein glycoforms are showing promise to increase the specificity of cancer diagnosis; for example, hepatocellular carcinoma. Moreover, advances have been reported in the ability to detect glycoprotein glycoforms in sera with the goal of developing diagnostic assay platforms. We are focusing on developing glycomarkers for various cancers; in particular, pancreatic cancer. The average pancreatic tumor has been growing 10 years before diagnosis; after diagnosis, only 7% of patients survive more than five years. Identifying a serum marker for pancreatic cancer, therefore, is critically important.

Our collaborators at TGen in Phoenix and we have identified a unique N-glycan marker for pancreatic adenocarcinoma with very high specificity and sensitivity (>85%). This N-glycan is expressed only on one protein in pancreatic cancer—CEACAM6 or CD66c. This protein is not expressed in adult pancreas, but becomes expressed during early stages of the cancer. We discovered this marker because it is the epitope of a monoclonal antibody, MAb109, developed by a small biotech company, Caerus Therapeutics, located in Manassas, VA. We have developed several assays for this epitope in serum and are determining if it is useful as a diagnostic marker. We are also exploring its utility as a prognostic marker for those patients that are CA19-9 negative, about 20% of those diagnosed with pancreatic cancer. Our main efforts are now focused on determining the precise structure of the glycan epitope by MSn and NMR analyses and identifying the enzyme that synthesizes the epitope.

2. Laboratory specific studies to meet the research needs

We have studied and are continuing to study the changes in glycosylation that occur on glycoproteins in pancreatic ductal fluid from patients with pancreatic cancer and pre-cancerous lesions. These glycoproteins may yield additional markers cancers. The glycan change that appears to be most promising is the glycan that comprises the epitope of the MAb109 antibody. We have found that HEK cells produce the epitope only when a fragment of CEACAM6 is expressed and secreted from them. We have used these cells, therefore, to produce large amount of fragment that contains the epitope. Out of the 12 N-linked sites that are expressed (and occupied) on full-length CEACAM6, the epitope is expressed only on one site; therefore, this specificity is an extreme example of protein-specific glycosylation. Using MSn analyses of the glycans released from this site, we will utilize technologies and software developed in our National Center for Biomedical Glycomics to identify the precise chemical structure of the epitope. Ultimately, it is likely that we will need to utilize NMR spectroscopy or chemical synthesis to obtain a definitive structure.

Our focus is also to continue development of a sandwich ELISA assay for detection of the epitope in serum and pancreatic ductal fluid. Presently we can detect in the nanogram range in three microliters of serum using a capture antibody against CEACAM6 polypeptide and the MAb109 as a probe. After optimization, we plan to participate in an ongoing clinical trial at TGen with 25 patients who are CA19-9 positive and 25 who are 19-9 negative to determine if the MAb109 assay can have prognostic value.

3. Resources and Reagents for sharing.

Our studies are in collaboration with our NIGMS P41 National Center for Biomedical Glycomics, which is charged with developing technologies and resources for glycomics of all types of tissues and cells. Through this Center we are collaborating with many investigators, including those studying various cancers. Our technologies include PCR and deep sequencing of glycan-related genes, state-of-the-art mass spectrometry, including use of the new Thermo-Fisher Fusion instrument, and bioinformatics to analyze MS data of glycan analysis and search databases from around the world. These resources are available as fee-for-service, or as collaborative projects. We also have collaborations that we classify as “Drivers”, which means their nature requires intensive technology development that is likely to be useful to many laboratories.

4. Public Health Implication/Advancing the Field of Glycobiology

Our results show that glycosylation of proteins can be very specific and can be exploited to develop glycomarkers. Using our specialized analytical technologies, glycomarkers will likely be identified not only for other cancers, but for many diseases or disorders, thereby having an impact on disease diagnosis and prognosis. These glycomarkers may also provide targets for therapies. For example, the MAb109 is internalized by endocytosis of pancreatic cancer cells after it binds to its target glycan on the cell surface and it is cytotoxic. This antibody, therefore, may be useful as a therapeutic, which is being explored.

In terms of the field of Glycobiology, it appears that the pancreatic cancer-specific N-glycan epitope is a unique structure that is conserved from insects to humans, likely being expressed in embryonic development during formation of the digestive tract. The enzyme that synthesizes the epitope would then be conserved over millions of years of evolution. The glycan, expressed on a cell adhesion molecule whose function is not clear, is conserved and, therefore, must perform a critical function during development, perhaps regulating the invasiveness or migration of cells. Its re-expression during oncogenesis would be consistent with its function as an onco-fetal epitope to regulate tumor invasiveness. Understanding the structure, biosynthesis, regulation, and function of the MAb109 N-glycan epitope may open a new chapter of a family of unique glycan structures. It should be mentioned that another, cancer-specific MAb has been produced by Caerus Therapeutics, which binds to an N-glycan in ovarian carcinoma and multiple myeloma and is useful as a glycomarker. It will be interesting to determine the structure of this epitope to determine if it is in some way similar to the MAb109 epitope. The two antibodies do not cross-react, however.

5. Cancer-Specific Relevance: detection, prevention and treatment

In adults, the MAb109 N-glycan epitope appears to be very pancreatic cancer-specific, requiring the expression of a particular glycoprotein, CEACAM6, before it can be synthesized. As mentioned earlier, we are investigating its expression in serum and ductal fluid as a diagnostic or prognostic, in place of CA 19-9. Since it was antigenic in mice, which lead to the production of the MAb109, it is possible that the epitope, when expressed in a cancer, could be antigenic and produce autoantibodies in humans to very early stage cancer. When we are able to isolate sufficient quantities of the epitope, either by purification or chemical synthesis, we will test if sera from pancreatic cancer patients contain antibodies that bind to the epitope. Moreover, as mentioned above, Caerus and we are exploring the possibility that the MAb109 may be used as a therapeutic against pancreatic cancer. In both colorectal cancer and non-small cell lung cancer, MAb109 binds and recognizes CEACAM5, which contains a peptide fragment identical to the region of CEACAM6 that contains the glycan epitope. We are also exploring the possibility that we can develop a diagnostic assay for these cancers based on serum expression of the MAb109 epitope.

6. Opportunities for Collaboration

As mentioned, our technology resources are shared as part of the National Center for Biomedical Glycomics, and these resources are available to all investigators. Please contact Michael Pierce, hawkeye@uga.edu, or the Technical Director of the Center, Parastoo Azadi, azadi@ccrc.uga.edu, for information and consultation.

In the Media

• "Tracking a killer," Maximizing Research Opportunities, online feature, Aug 22, 2007.
• Survivor a beacon in cancer research . Ken Foskett, Atlanta Journal-Constitution, Oct. 22, 2007.

Patents

• Methods to Identify and Quantify Oligosaccharide Modifications of Glycoproteins (US 7,351,544).
• Provisional Application: Glycoprotein Cancer Biomarkers Filed: June 11, 2010. (serial # 12/814,184).
• Provisional Application: Cancer-specific Glycans and Glycopeptides Filed: July 13, 2010 (serial# 61/363,782).
• Provisional Application: Alpha Toxin Detection of GPI Anchored Proteins Filed: August 6, 2010 (serial# 61/371,326).