Investigators and Scientific Areas

1. Investigators and Scientific Areas • Discovery of Cancer Associated Glycans • Michael Pierce, Lance Wells, University of Georgia • Bill Hancock, Northeastern University • Milos Novotny, Indiana University • Antibodies to Tumor-Associated Glycans and Glycopeptideepitopes • Margaret Huflejt, New York University. • Denong Wang, Ten Feizi, Stanford University, Imperial College London • Tony Hollingsworth, Eppley Institute, University of Nebraska Medical Center, Henrik Clausen and Ola Blixt, University of Copenhagen • AjitVarki, Richard Schwab, University of California San Diego

2. The Problem – Diagnosis of Early Cancer and Determination of Cancer Progression

3. Glycomics Technologies applied to Cancer Glycotranscriptome analysis Glycan and glycoprotein analysis of tissues, serum, fluids Glycan and glycopeptide arrays for detecting unique autoantibodies

4. Collaborations • Clinical samples and technologies from several institutions • Unique resources • Pancreatic cancer rapid autopsy samples (Nebraska) • Prostate cancer samples (Stanford) • Discovery and Reference sets of sera and plasma from EDRN • Collaborations with Investigators from EDRN, SPORE program, and P01s

5. Targeted Glycoproteomics: exploiting glycan expression on specific glycoproteins to identify potential cancer biomarkersTumor Glycomics Laboratory, Univ. of Georgia, and UGA Cancer Center Cancer-specific glycan • Use lectins/antibodies to separate • glycoproteins with these glycan changes • Identify cancer-specific glycoproteins • expressing Targeted Glycans via proteomics • Verify and validate in tissue and serum • with a two-step lectin:antibody assay • Markers require 2-dimensional specificity: • Protein plus Glycan expression Identify glycan changes by comparison of direct glycan analyses, glycotranscriptome qRT-PCR analyses, or lectin/antibody binding to cancer and control tissues. Target the glycoproteins that express these glycans (Methods developed in NCRR Glycomics Center, UGA) Protein Space-filling model of a secreted glycoprotein. Its three glycans are depicted in blue; polypeptide in white/red. Courtesy of R. Woods, CCRC, UGA

6. Application of targeted glycoproteomics methodology to discovery of invasive ductal breast carcinoma markers • The lectin L-PHA binds glycans not present in breast epithelia, but which are expressed in • late adenoma and carcinoma. • Utilized Lectin L-PHA to target glycoproteins that expressed the glycan of interest in four cases • of breast carcinoma with matched, non-diseased tissue controls. • Identified 12 glycoproteins bound by L-PHA common to all four carcinomas but not in • any of the controls. Two glycoproteins have been verified in breast carcinoma tissue and serum. PERIOSTIN and MIMICAN are found bound by L-PHA only in cancer tissues and serum Both are EMT-expressed, secreted glycoproteins. They are now being validated as potential serum markers. Collaboration with Dr. Ruth O’Reagan Emory Winship Cancer Center Abbott KL et al. Targeted glycoproteomic identification of biomarkers for human breast carcinoma. J. Proteome Res. 2008 Apr;7(4):1470-80.

7. Platform used for the analysis of the breast cancer and control sera. • Four separation approaches – • Depletion, • Glycoprotein fractionation • IEF fractionation using dPC and • RP-LC/MS peptide separation. Human Serum (Breast Cancer or Control) High abundance protein depletion Multi-lectin affinity chromatography (M-LAC) Bound fraction Unbound fraction dPC (+) IEF fractionation (dPC) 10 fractions, 8 gel plugs/fraction, pH 4.20~8.00 Periostinwas identified with high confidence. The isoelectric focusing profile showed a shift to more acidic pI values in the disease samples, which indicates a greater sialic acid content in breast cancer. pH 5.15 pH 5.10 pH 4.20 pH 6.20 Gel plug # Fraction # # 1 # 2 # 3 # 4 # 5 pH 6.95 pH 6.90 pH 6.00 pH 8.00 Gel plug # Fraction # # 6 # 7 # 8 # 9 # 10 In gel digestion dPC (-) dPC (-) Protein identification by LC-MS analysis

8. Targeted Approach EGFR (A431 cells) 11 sites of N-linked glycosylation Specific polyclonal antibody pull down RP-HPLC purification Q. Lu

9. Glycan analysis by LC with LTQ-FTMS of EGFR glycopeptide eluted at 25 to 25.5 minutes from a reversed phase column 100 90 80 70 60 50 Relative Abundance 40 30 20 10 0 M – (SA+Gal+NGlc) CID → m/z 1169.73 (5+) M –(♦-●-■) 4+ QHGQFSLAVVSLNITSLGLRSLK (Asn 420) 1297.50 Tetra-antennary with 3 sialic acids Asn 420 M - SA [M -♦]4+ 1388.63 M – (SA+Gal+NGlc+Man) M –(♦-●-■-●) 4+ 1256.90 M - SA M –(♦-●-■-●-■) 3+ [M -♦]5+ M –(♦-●-■) 3+ ●-■ ♦-●-■ 1110.99 [M -●]3+ 1607.51 ●-■-● 997.36 366.14 1220.20 657.11 1729.21 1498.92 1894.87 528.11 400 600 800 1000 1200 1400 1600 1800 2000 Q. Lu, Billy Wu m/z

10. EGFR Glycosylation (A431 cell line) Major glycoform at Asn 420 site Membrane bound Soluble form (secreted) tetra-antennary branches with 3 terminal sialic acids bi-antennary structure with 1 terminal sialic acid

11. Healthy Breast Cancer Stage IV 1500 2200 2900 3600 4300 5000 Mass (m/z) Z. Kyselova, Y. Mechref, P. Kang, J. A. Goetz, L. E. Dobrolecki, G. Sledge, L. Schnaper, R. J. Hickey, L. H. Malkas, M. V. Novotny “Breast Cancer Diagnosis/Prognosis through Quantitative Measurements of Serum Glycan Profiles” Clin. Chem., in press.

12. 16 Healthy (N=19) 15 Cancer Stage I (N=8) 14 Cancer Stage II (N=7) 13 Cancer Stage III (N=4) 12 Cancer Stage IV (N=50) 11 10 9 8 7 6 5 4 3 2 PC2 Score 1 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 -14 -26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 PC1 Score PCA of MALDI/MS Profiling of Glycans Derived from Sera of Healthy Individuals and Breast Cancer Patients Z. Kyselova, Y. Mechref, P. Kang, J. A. Goetz, L. E. Dobrolecki, G. Sledge, L. Schnaper, R. J. Hickey, L. H. Malkas, M. V. Novotny “Breast Cancer Diagnosis/Prognosis through Quantitative Measurements of Serum Glycan Profiles” Clin. Chem., in press.

13. Microarrays for detecting antibodies with unique oligosaccharide and glycopeptide specificities • Huflejt - ~ 300 unique oligosaccharides on microarrays • Wang – Clustered oligosaccharides representative of prostate cancer • Feizi – Arrays of oligosaccharides from mucins of ovarian cystadenomas • Varki - Neu5Gc (N-Glycolylneuraminic acid) • Clausen, Blixt, Hollingsworth – Glycopeptides containing specific tumor associated oligosaccharides (Tn, T, sialylTn, sialylT) attached to mucin peptides

14. Mucins 21 genes Common features: Secretion into mucus layer High MW glycoprotein TANDEM REPEAT Heavily glycosylated Different tissues/ organs express differing sets of mucins Tumors express different core proteins and differential glycosylation of these Hollingsworth and Swanson, 2004

15. The Glycopeptide array Figure courtesy M. Tarp, E. Bennett and H. Clausen • 1st prototype • MUC-1 (60mer) • 6Tn-MUC-1 (60mer) • 9Tn-MUC-1 (60mer) • 15Tn-MUC-1 (60mer) • 9STn-MUC-1 (60mer) • 15STn-MUC-1 (60mer) • 15T-MUC-1 (60mer) • 9Core3-MUC-1 (60mer) • 15Core3-MUC-1 (60mer) • MUC-2 (33mer) • 6Tn-MUC-2 (33mer) • 12Tn-MUC-2 (33mer)

16. Immunization of cancer patients with STn on MUC1 yields glycopeptide specific antibodies

17. Core3-, STn- and Tn-MUC1 auto-antibodies are present in cancer patients with breast, ovarian, and prostate cancers.

18. Summary • Application of state of the art technologies to define unique glycan structures associated with cancer progression • Definition of antibody responses to unique oligosaccharide and glycopeptideepitopes during cancer progression • Investigators interface with EDRN, SPORE program, PO1s, CFG