Abstract 2489: Glycoproteomic analysis of breast cancer cell lines for biomarker discovery

Abstract Glycosylation is one of the most ubiquitous protein post-translational modifications observed in eukaryotic organisms. In cancer, key cellular pathways and interactions, related to invasion and metastasis, are disrupted by altered protein glycosylation. Many current clinical cancer biomarkers are glycoproteins (i.e. CEA, CA15-5, CA125, and PSA). Glycosylation is known to be aberrant in breast cancer. However, systematic studies to identify glycoproteins in cancer have been limited, and typically focus on single proteins of interest or glycans released from proteins. We have developed a novel protein fractionation and enrichment strategy for glycoproteomic analysis of the secreted proteins, incorporating lectin affinity capture for glycoprotein enrichment, intact protein separation, and enzymatic digestion followed by LC-MS/MS analysis. Here, we apply this workflow to capture and analyze glycoproteins in the cancer secretome, which includes proteins secreted and/or shed from cancer cells into the extracellular space. Four breast cancer cell lines (HCC70, HCC1806, BT474, and ZR75-1) and one normal mammary epithelial cell line (MCF10A) were subjected to glycoproteomic analysis. Proteins corresponding to 3360 unique gene names were identified. Subsequent data analysis yielded a variety of types of information about glycoproteins in the cell line secretomes. 1) Fucosylated and branched-sugar containing glycoforms of proteins were identified by lectin binding properties. For example, PHA-P/PHA-L lectins showed high affinity for aldolase, immunoglobulins, and inter-alpha-trypsin inhibitor, suggesting these proteins contain branched sugar structures. 2) Enrichment of some proteins was observed in breast cancer cell lines compared to the normal mammary epithelial cell line and vice versa. For example, high expression of fucosylated vimentin was observed in MCF10A, but not breast cancer cell lines. 3) Differences in glycosylation patterns of proteins were observed between cell lines. For example, the protease legumain was observed to be highly fucosylated in MCF10A, but had a higher affinity for PHA-P/PHA-L in HCC70. Additional substrates and inhibitors of legumain were also identified. Proteins of interest were further studied in additional follow-up experiments including western blotting and glycostaining. Additional proteins of interest were measured in plasma samples from women with (benign or malignant) breast lesions. Citation Format: Maria Arampatzidou, Majlinda Kullolli, Sharon J. Pitteri. Glycoproteomic analysis of breast cancer cell lines for biomarker discovery. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2489. doi:10.1158/1538-7445.AM2014-2489