Fig. 2

Identification of targetable glycoproteins in bladder cancer supported by bioinformatics analysis. A) Venn Diagram showing glycoproteins identified by targeted glycoproteomics focusing on sialylated T antigens in different cell models. A total of 903 glycoproteins were identified by mass spectrometry; however, only 28% were common to both cell lines, suggesting considerable distinct glycoproteomes. B) Venn diagram for glycoproteins overexpressed in MIBC according to the Oncomine (RNAseq) database. Glycoproteins were comprehensively matched with transcriptomic data in Oncomine to sort species potentially upregulated in MIBC in relation to non-pathological urothelium. Glycoproteins upregulated in both NMIBC and MIBC were also included for downstream analysis. 11% (95/903 assignments) were upregulated in MIBC, with the cell lines showing again very distinct and unique glycoproteomes. These glycoproteins were elected for specificity evaluation. C) Gene expression (Oncomine, RNASeq) vs protein levels (Human Protein Atlas; IHQ) for glycoproteins showing gene upregulation in MIBC. This diagram shows that 95 out of the 903 sorted glycoproteins were detected in MIBC, 81% of which were also present in NMIBC. The majority (63%) was present in low or moderate levels. A restricted group of 18 proteins, 3 of which highly expressed, were exclusively found in MIBC and may play a critical role in advanced disease. In addition, 13% (12/95) of the identified glycoproteins have been previously detected by targeted glycoproteomics for the STn antigen in bladder tumours showing resistance to chemotherapy (identified in red). D) Target Score for 29 glycoproteins showing some degree of expression in bladder tumours. Briefly, glycoproteins were ranked according to their potential for targeted therapeutics considering its plasma membrane expression in relation to other subcellular locations, absence from healthy urothelium and other human tissues, and overexpression in bladder cancer, while severely penalizing lymphoid system and gametes expression. The overall goal was to maximize tumour specificity while minimizing off-target effects. Maximum possible score was 15 (arbitrary units). GLUT1 (SLC2A1) and HOMER3 were top-ranked proteins scoring 13