Distinct urinary glycoprotein signatures in prostate cancer patients

Abstract

// Rebeca Kawahara 1 , Fabio Ortega 2 , Livia Rosa-Fernandes 3 , Vanessa Guimarães 2 , Daniel Quina 1 , Willian Nahas 4 , Veit Schwämmle 3 , Miguel Srougi 2 , Katia R.M. Leite 2 , Morten Thaysen-Andersen 5 , Martin R. Larsen 3 and Giuseppe Palmisano 1 1 Instituto de Ciências Biomédicas, Departamento de Parasitologia, Universidade de São Paulo, USP, São Paulo, Brazil 2 Laboratório de Investigação Médica da Disciplina de Urologia da Faculdade de Medicina da USP, LIM55, São Paulo, Brazil 3 Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark 4 Instituto do Câncer do Estado de São Paulo, ICESP, São Paulo, Brazil 5 Department of Molecular Sciences, Macquarie University, Sydney, NSW, Australia Correspondence to: Giuseppe Palmisano, email: palmisano.gp@usp.br Keywords: prostate cancer; urine; glycopeptide; TMT-labeling; glycoproteomics Received: April 04, 2018      Accepted: July 31, 2018      Published: September 04, 2018 ABSTRACT Novel biomarkers are needed to complement prostate specific antigen (PSA) in prostate cancer (PCa) diagnostic screening programs. Glycoproteins represent a hitherto largely untapped resource with a great potential as specific and sensitive tumor biomarkers due to their abundance in bodily fluids and their dynamic and cancer-associated glycosylation. However, quantitative glycoproteomics strategies to detect potential glycoprotein cancer markers from complex biospecimen are only just emerging. Here, we describe a glycoproteomics strategy for deep quantitative mapping of N - and O -glycoproteins in urine with a view to investigate the diagnostic value of the glycoproteome to discriminate PCa from benign prostatic hyperplasia (BPH), two conditions that remain difficult to clinically stratify. Total protein extracts were obtained, concentrated and digested from urine of six PCa patients (Gleason score 7) and six BPH patients. The resulting peptide mixtures were TMT-labeled and mixed prior to a multi-faceted sample processing including hydrophilic interaction liquid chromatography (HILIC) and titanium dioxide SPE based enrichment, endo-/exoglycosidase treatment and HILIC-HPLC pre-fractionation. The isolated N - and O -glycopeptides were detected and quantified using high resolution mass spectrometry. We accurately quantified 729 N -glycoproteins spanning 1,310 unique N-glycosylation sites and observed 954 and 965 unique intact N - and O -glycopeptides, respectively, across the two disease conditions. Importantly, a panel of 56 intact N-glycopeptides perfectly discriminated PCa and BPH (ROC: AUC = 1). This study has generated a panel of intact glycopeptides that has a potential for PCa detection.