Characterisation of Prostate-Specific Antigen Isolated from Patients’ Urine Alan Moran (Presenter) Leiden University Medical Centre Presenter Bio: Alan Moran, from Ireland, joined GlySign in 2017 under the supervision of Prof. Dr. Manfred Wuhrer, Guinevere Kammeijer (LUMC, NL), and Dr. Daniel Spencer (Ludger Ltd., UK). He will spend a significant proportion of his PhD at each institution where he will focus on the analysis of the glycosylation of prostate-specific antigen (PSA) in prostate cancer in order to improve patient stratification. For this purpose, Alan is exploring the usage of a variety of separation techniques (LC/CE) and the on-line coupling (ESI) to mass spectrometry (Q-TOF/FT-ICR), as well as the development and validation of a clinical assay. His scientific interests include the translation of proteomic and glycomic research to the clinics and industry.

Authors: A.B. Moran (1), G.S.M. Kammeijer (1), J. Nouta (1), T.M. de Reijke (2), E. Dominguez Vega (1), M. Wuhrer (1)
(1) Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, The Netherlands (2) Academic Medical Center, Department of Urology, University of Amsterdam, Amsterdam, The Netherlands

The glycoprotein prostate specific antigen (PSA) is a clinical biomarker of prostate cancer (PCa) but has a rather poor specificity and predictive value limiting its value for PCa diagnosis. It has been hypothesized that through examining its glycosylation, PSA could be used as a more specific biomarker of PCa. This study presents a novel approach to analyse the glycosylation variants as well as other proteoforms of PSA from patients’ urine using capillary electrophoresis coupled to a mass spectrometer with a sheathless interface (CE-ESI-MS). Analysis of intact PSA revealed a wide variety of glycoforms and proteoforms and results obtained from intact and glycopeptide analysis were compared.

Introduction

The risk of developing cancer during one’s lifetime is approximately 38% for men and women (1). However, many of the diagnoses are made when the cancer is already at an advanced stage which often prohibits curative treatment. This highlights the importance of effective screening and early detection using accurate biomarkers of the disease. The glycoprotein prostate-specific antigen (PSA) is a well-known biomarker for prostate cancer (PCa) screening, however it suffers specificity (2,3). Aberrant glycosylation of this protein has been reported during cancer development (4,5) and PSA production occurs at close proximity to the original tumor site. Therefore, several groups have studied PSA glycosylation profiles extensively and have shown that alterations in the glycosylation profile could be an attractive alternative as a biomarker of this disease. However, most studies focused on lectin binding assays (4,6), the digestion of the glycan (7) or peptide portion (8) and were not able to provide a more thorough insight on the existence of different proteoforms and other post-translational modifications of PSA. In this study, we established an analytical platform that enables the analysis of intact PSA captured from patient material (urine). In addition, the same samples were digested based on a previously published method (8) and the glycosylation profiles obtained with the two approaches (intact and bottom-up) were compared.

Methods

Patients with an elevated PSA serum concentration (>3 ng/mL) and suspected of PCa donated a urinary sample prior to digital rectal examination and prostate biopsy. PSA was isolated and immunopurified from patients’ urine using antibody specific beads (8). In addition, a urine pool of the patients urine (5 mL per sample) was used as a positive control and a female urine pool was used as a negative control. Notably, the samples were split into fractions for intact protein analysis (80%) or digestion with porcine trypsin (20%). Analysis was performed using sheathless capillary electrophoresis electrospray ionization mass spectrometer (CE-ESI-MS) on a CESI-8000 instrument (Sciex, Brea, CA) coupled to an Impact quadrupole time-of-flight (QToF)-MS (Bruker) employed with a nano-electrospray source. Separation was carried out on an in-house coated capillary using either polyethyleneimine (intact analysis) or Ultratrol LN (bottom-up analysis), in both cases the capillary was equipped with a porous tip. As a background electrolyte, solutions of 20% acetic acid with 0-10% methanol were used. Glycoform and proteoform assignments were assisted by results obtained from exoglycosidase-treated and reduced seminal plasma PSA.

Results

The analysis of intact PSA allowed the separation of a wide variety of PSA glycoforms and proteoforms. For example, mono-, di- and tri--sialylated glycoforms were found on intact PSA as well as high-mannose types (Man3 till Man5). The most abundant N-glycan in the pooled sample was found to be a diantennary disiaylated glycan with a core fucose (H5N4F1S2). Preliminary results already revealed a total of nine glycan structures on intact PSA. Furthermore, three proteolytic cleavage sites were found as is in agreement with literature (9). Cleavages can occur between positions Asn108/Arg109/Phe110, Lys169/Lys170, and Lys206/Ser207. The internal cleavage at a lysine position was the most prominent form in pooled urinary PSA samples. Literature has shown that internally cleaved PSA seems to be more elevated in benign prostate hyperplasia (BPH) in comparison with PCa (10). Therefore, the information obtained during this study could be of importance to study the differences (e.g. internal cleavages and the abundance of certain N-glycans) between BPH and PCa patients, to aid patient stratification. Additionally, non-glycosylated PSA was observed in the urinary samples while this was not observed in commercially available PSA (seminal plasma). Next to that, clear differences were found in the abundances of glycoforms and other proteoforms between urinary PSA and commercial PSA. This could be due to the source of the PSA (seminal vs urinary) and/or the standard is retrieved from healthy volunteers, while all urine samples originated from patients with elevated PSA levels. Next to the intact analysis, part of the sample underwent a bottom-up approach. In contrast to the intact analysis, no internal cleavages and non-glycosylated forms could be detected. This is most likely due to the enzymatic digestions which produces the same glycopeptide for internally cleaved and non-cleaved forms and no distinction can be made between these proteoforms. However, the bottom-up approach does provide additional information, for example this approach is able to distinguish α2,6- from α2,3-sialylated species which provides a more thorough characterization of the glycans present on all PSA forms. This could be of importance as previous studies have implicated that α2,3-sialylated PSA could be a potential biomarker of PCa (11). Furthermore, previous research was able to assign 67 N-glycan structures in a patient urinary pool (8) and a similar number is expected in this study.

Conclusions & Discussion

This study demonstrates the complementarity between intact protein analysis and the bottom-up approach. The intact analysis provides information regarding the entire protein which could not be observed with a bottom-up approach (e.g. co-occurring modifications such as glycosylation and internal cleavages), while the bottom-up approach provides higher sensitivity, enabling the identification of more N-glycan species present on PSA. Finally, further analysis of the data is required in order to determine whether patients exhibit any notable variation in PSA glycoforms and proteoforms. This, in combination with the final diagnosis regarding the status of PCa, might indicate whether this type of analysis has potential for identifying glycomic and proteomic biomarkers and enabling differentiation between aggressive PCa, indolent PCa and BPH.

1. Das, V., Kalita, J. and Pal, M., 2017. Predictive and prognostic biomarkers in colorectal cancer: A systematic review of recent advances and challenges. Biomedicine & Pharmacotherapy, 87, pp.8-19.

2. Gilgunn, S., Conroy, P.J., Saldova, R., Rudd, P.M. and O'kennedy, R.J., 2013. Aberrant PSA glycosylation—a sweet predictor of prostate cancer. Nature Reviews Urology, 10(2), p.99.

3. Mechref, Y., Hu, Y., Garcia, A., Zhou, S., Desantos-Garcia, J.L. and Hussein, A., 2012. Defining putative glycan cancer biomarkers by MS. Bioanalysis, 4(20), pp.2457-2469.

4. Tabares, G., Radcliffe, C.M., Barrabés, S., Ramírez, M., Aleixandre, R.N., Hoesel, W., Dwek, R.A., Rudd, P.M., Peracaula, R. and de Llorens, R., 2005. Different glycan structures in prostate-specific antigen from prostate cancer sera in relation to seminal plasma PSA. Glycobiology, 16(2), pp.132-145. 4

5. Llop, E., Ferrer-Batallé, M., Barrabés, S., Guerrero, P.E., Ramírez, M., Saldova, R., Rudd, P.M., Aleixandre, R.N., Comet, J., de Llorens, R. and Peracaula, R., 2016. Improvement of prostate cancer diagnosis by detecting PSA glycosylation-specific changes. Theranostics, 6(8), p.1190.

6. White, K.Y., Rodemich, L., Nyalwidhe, J.O., Comunale, M.A., Clements, M.A., Lance, R.S., Schellhammer, P.F., Mehta, A.S., Semmes, O.J. and Drake, R.R., 2009. Glycomic characterization of prostate-specific antigen and prostatic acid phosphatase in prostate cancer and benign disease seminal plasma fluids. Journal of proteome research, 8(2), pp.620-630.

7. Hsiao, C.J., Tzai, T.S., Chen, C.H., Yang, W.H. and Chen, C.H., 2016. Analysis of urinary prostate-specific antigen glycoforms in samples of prostate cancer and benign prostate hyperplasia. Disease markers, 2016.

8. Kammeijer, G.S.M, Nouta, J., de la Rosette, J.J., de Reijke, T.M. and Wuhrer, M., 2018. An In-Depth Glycosylation Assay for Urinary Prostate-Specific Antigen. Analytical chemistry, 90(7), pp.4414-4421.

9. Mattsson, J.M., Valmu, L., Laakkonen, P., Stenman, U.H. and Koistinen, H., 2008. Structural characterization and anti‐angiogenic properties of prostate‐specific antigen isoforms in seminal fluid. The Prostate, 68(9), pp.945-954.

10. Linton, H.J., Marks, L.S., Millar, L.S., Knott, C.L., Rittenhouse, H.G. and Mikolajczyk, S.D., 2003. Benign prostate-specific antigen (BPSA) in serum is increased in benign prostate disease. Clinical chemistry, 49(2), pp.253-259.

11. Ishikawa, T., Yoneyama, T., Tobisawa, Y., Hatakeyama, S., Kurosawa, T., Nakamura, K., Narita, S., Mitsuzuka, K., Duivenvoorden, W., Pinthus, J.H. and Hashimoto, Y., 2017. An Automated Micro-Total Immunoassay System for Measuring Cancer-Associated α2, 3-linked Sialyl N-Glycan-Carrying Prostate-Specific Antigen May Improve the Accuracy of Prostate Cancer Diagnosis. International journal of molecular sciences, 18(2), p.470.