Poster Presentation HUPO 2019 - 18th Human Proteome Organization World Congress

Metastatic melanoma model system selection: implications for glyco-marker discovery (#401)

Jodie L Abrahams 1 , Matthew P Campbell 1 , Andrea Maggioni 1 , Daniel Kolarich 1 , Mark von Itzstein 1 , Nicolle H Packer 1 2
  1. Institute for Glycomics, Griffith University, Gold Coast, QLD, Australia
  2. Department of Molecular Sciences, Macquarie University, Sydney, NSW, Australia

There is a critical need to identify novel predictive and prognostic markers with improved sensitivity and specificity for the management of advanced stage melanoma. Despite a growing understanding of the role of protein glycosylation in melanoma biology, there has been little progress towards using glycosylation changes as melanoma biomarkers.

To date most glycosylation studies have relied on the availability of cell lines. Although these models are an extremely valuable resource for cancer biology research we must question whether the glycosylation patterns of continuously cultured cell lines are truly representative of complex tumour tissue samples. Here we present a comprehensive comparison of commonly used cell lines, patient derived primary cells and lymph node tumour tissue from metastatic melanoma patients to help answer this question.

N-glycans released from melanoma cell membrane proteins were characterised using a porous graphitized carbon (PGC) liquid chromatography mass spectrometry glycomics platform. Structures were fully assigned using MS/MS fragmentation patterns, PGC retention behaviours, and linkages confirmed using exoglycosidase enzymes. The glycosylation profile of cell line panels, patient derived primary cell lines and tumour tissue were compared to identify sample specific glycan features.

Global glycomics identified metastatic melanoma associated glycan features common to all sample sources including branching, sialylation and fucosylation. However, significant differences in the relative abundance of glycan classes were observed between the sample origins, most notably the ratio of mannose to complex type glycans. We show that patient tissue exhibited a more complex profile than cultured cells and that there is a high degree of variation in protein N-glycosylation between patient tumours.

This study contributes to our understanding of glycosylation alterations in melanoma and highlights that cell surface glycosylation not only varies between different cell lines, stored cell lines and primary cell lines, but also markedly varies between all cell lines and tissue.