Greene/Feizi - U of Penn; Imperial College

Principal Investigator: Mark Greene, M.D., Ph.D. 
Institution: University of Pennsylvania, Philadelphia, PA

Subcontract Principal Investigator: Ten Feizi, M.D., FmedSci
Institution: Imperial College London

Former Principal Investigator: Minoru Fukuda, Ph.D. (retired)
Institution: Sanford Burnham Medical Research Institute, La Jolla, CA

Our project Carbohydrate Antigenic Biomarkers for Epithelial Cancers has as its main focus the characterization and application as serological biomarkers cancer-associated antigens of prostate and breast. We will explore the use of the antigens or autoantibodies to them as unique biomarkers in the diagnosis and prognosis of epithelial cancers. Current focus is the prostate cancer-associated F77 antigen and the more widely expressed epithelial cancer associated AE3 antigen.

Minoru Fukuda and Ten Feizi, together with Michiko Fukuda, have had a long record of successful collaborations and key contributions to discoveries and characterizations of glycan sequences of the poly-N-acetyllactosamine type. More recently Feizi and colleagues at the now Wellcome Trust-supported Biomedical Resource at Imperial College London have established glycan array technologies, including those for 'designer' arrays from targeted glycoconjugates, coupled with mass spectrometry to characterize novel carbohydrate antigens. On the first phase of the project the Fukudas at Sanford Burnham Medical Research Institute brought novel and highly complementary glycomics strategies encompassing advanced molecular biology expertise on glycosyltransferases, sulfotransferases and their applications to characterize biosynthetically the key elements of carbohydrate antigens.

The glycobiological strengths are fortified by expertise in immunobiology, cancer cell biology, oncology, proteomics, and computational biology. Mark Greene and Hongtao Zhang at the University of Pennsylvania have developed the monoclonal antibody F77 which targets a glycolipids and glycoproteins associated with a multiprotein ensemble on human prostate cells. F77 antibody inhibits growth of incipient tumors in animal models. There is strong oncology support for the tissue and serological studies from Donna Peehl at Stanford University School of Medicine, who has extensive and well characterized archival serum and tissue banks from men with prostate cancer and benign prostatic hyperplasia, and Joy Burchell and Joyce Taylor-Papadimitriou, at Kings College London, who have a large collection of paired sera taken pre and post breast cancer surgery. For the serological and tissue studies there will be expert bioinformatics advice and supervision from Tim Ebbels at Imperial College.

Synopsis of Research and Network Resources

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

The main focus of our project is the characterization of cancer-associated carbohydrate antigens and their application as biomarkers with special reference to cancers of the prostate and the breast. Minoru Fukuda and Ten Feizi, together with Michiko Fukuda, have a long record of successful collaborations and key contributions to discoveries and characterizations of glycan sequences of the poly-N-acetyllactosamine type. More recently, Feizi and colleagues, now at the Wellcome Trust-supported Biomedical Resource at Imperial College London, developed a unique glycan array platform (neoglycolipid-[NGL]-based microarray technology) using lipid-linked probes in conjunction with mass spectrometry. The technology can also be extended to 'designer' microarrays derived from ligand-bearing glycomes in order to reveal the glycan ligands/antigens they harbor, so that these can be isolated and characterized. The Fukudas brought to the project novel and highly complementary glycomics strategies encompassing advanced molecular biology expertise on glycosyltransferases, sulfotransferases and their applications to characterize biosynthetically the key elements of carbohydrate antigens.

The strengths in the above techniques are further fortified by the application of immunobiology, cancer cell biology, oncology, proteomics, and computational biology. Mark Greene and Hongtao Zhang have developed a monoclonal antibody against the carbohydrate antigen F77, which targets human prostate cancer cells and inhibits growth of incipient tumors in animal models. There is strong oncology support for tissue and serological studies from Donna Peehl, who has extensive and well-characterized archival serum and tissue banks from men with prostate cancer and benign prostatic hyperplasia, and from Joy Burchell and Joyce Taylor-Papadimitriou at King’s College London, who have a large collection of paired sera taken pre- and post-breast cancer surgery. For the serological and tissue studies there will be expert bioinformatics advice and supervision from Tim Ebbels at Imperial College.

The team has selected for study two hitherto uncharacterized cancer-associated carbohydrate antigens. The first is the prostate cancer-associated F77 antigen [1], and the second, a widely distributed epithelial cancer-associated antigen recognized by a hybridoma antibody, AE3, considered a promising serum biomarker [2].

Laboratory-specific Studies to Meet the Research Needs

We have determined the structure of F77 epitope by a powerful combination of two technologies. The first involves carbohydrate microarray analyses using sequence-defined glycolipids and neoglycolipids, mucin-O-glycan designer arrays and mass spectrometry [3]. The second is by cell transfections and siRNA knockdown of specific glycosyltransferases [4]. We have thus demonstrated that the F77 epitope consists of blood group H on 6-linked branches of poly-N-acetyllactosamine backbones [4]. This type of branched arrangement has not been previously reported and likely represents a non-abundant structural element of glycans.

The question then arises as to why F77 antigen that includes a branched backbone sequence is strongly expressed predominantly in prostate cancer rather than in other cancers. We propose that the answers lie in the finding that the type 2 blood group H antigen is expressed in the normal prostate irrespective of the ABO blood group status, and is increased in prostate cancer tissues [5-8]. This contrasts with many other cancers in which the branching enzyme genes are also overexpressed, but the type 2 H antigen expression is decreased due to alpha1,3-fucosylation and the formation of Le^y antigen sequence, which we have shown not to be bound by mAb F77; alternatively, there may occur competition by sialylation on the terminal galactose instead of a1,2-fucosylation. With knowledge of the structure and prevalence of F77 antigen in prostate cancer, the road is open to exploring rationally its applications as a biomarker.

To elucidate the AE3 epitope, we are taking similar approaches, also encompassing binding analyses with MUC-1 specific glycoforms and chemo-enzymatic synthesis of the putative glyco-amino acid target in collaborations with Kelley Moremen at the Complex Carbohydrate Research Center, Georgia

Resources and Reagents for Sharing

Resources available:

• Screening analyses of antibodies and other carbohydrate-recognizing proteins using the NGL-based sequence-defined and designer arrays (Feizi)
• Expertise in molecular and cellular biology of the normal prostate and cancer (Peehl)
• Developing antibodies and scFvs to tumour-associated antigens for tissue targeting and structural studies of the antigen-antobody complexes. (Greene)

Reagents for sharing:

Prostate cancer

• Well-annotated archival prostate tissues for the discovery and validation of diagnostic and prognostic markers of prostate cancer (Peehl)
• Banked serum from men with prostate cancer or benign disease to evaluate circulating biomarkers relevant to diagnosis or prognosis (Peehl)

Breast cancer

• Cancer-associated glycoforms of MUC1 including MUC1-Tn, -T and –ST (Burchell)
• Serum samples from breast cancer patients and well women subject to ethical approval (Burchell)
• Breast tissue bank, subject to approval of the local Access Committee (Burchell)

Public Health Implication/Advancing the Field of Glycobiology

This research highlights the potential of glycobiology to offer specific biomarkers for cancer detection because some of the glycan structures produced by tumor cells are distinct from those in normal counterparts. Proteins normally do not exhibit this level of exclusive expression. The discovery of molecular components with great selectivity for different cancers will be invaluable in developing diagnostic tests with other possible applications towards imaging of tumors or therapeutics.

Cancer-Specific Relevance: Detection, Prevention and Treatment

We advocate studies of the regulation of the expression of glycosyltransferase and sulfotransferasegenes concomitant with glyco-antigen expression in conjunction with proteomics of the carrier proteins as a rational approach to identifying carbohydrate biomarkers of malignancies

Opportunities for Collaboration

Members of the team are happy to integrate with investigators interested in the elucidation and exploitation of cancer-associated carbohydrate antigens.