A Newly Developed LC-MS/MS Method for the Quantitation of Glucagon in Plasma Shows the Lack of Specificity of Former RIA Methods Jordi Farre-Segura (Presenter) University of Liège, CHU de Liège Presenter Bio: Jordi Farré obtained his degree and master-equivalent diploma in pharmacy in September 2017 by University of Barcelona. In 2016, he was firstly introduced to the clinical chemistry field in his hometown hospital’s laboratory (Althaia-Xarxa Hospitalària Universitària. Manresa, Spain) as a trainee. Later this same year he was granted with an Erasmus exchange to Liège, where he develops his research in proteomics since October 2017 as a PhD student at the Department of Clinical Chemistry, University Hospital of Liège (Belgium) under the supervision of Prof. Etienne Cavalier and Dr. Anne-Catherine Servais. During his first stay in Belgium he contributed in the development of a new LC-MS/MS method for quantitation of Vitamin D metabolites in serum, under the supervision of Dr. Neus Fabregat and Dr. Caroline Le Goff.

Authors: Jordi Farré-Segura(1), Neus Fabregat-Cabello(1), Laurent Nyssen(1), Stéphanie Peeters(1), Caroline Le Goff(1), Etienne Cavalier(1).
(1)Department of Clinical Chemistry, University Hospital of Liège, University of Liège, Belgium.

We have developed a sensitive method for intact glucagon quantitation by LC-MS/MS based on a simple solid phase extraction. The method showed good precision, accuracy and detection limits throughout its advancement. However, our preliminary comparison with our routine radioimmunoassay (RIA) showed a significant bias. Hence, RIA provides spurious results compared to the LC-MS/MS that could suppose a change in the perspective of this peptide’s measurement.

Introduction

Glucagon is a 29 amino acid peptide processed by the pancreatic α-cells from its precursor proglucagon. Its main function is to maintain correct glucose levels when fasting and to raise them in situations of hypoglycemia[1]. It is also a biomarker for pathologies such as diabetes, pancreatic cancer or certain neuroendocrine tumors [2].

Currently it is majorly measured in clinical laboratories by either ELISA (enzyme-linked immunosorbent assay) or RIA (radioimmunoassays). These techniques present good sensitivity although they sometimes lack specificity due to cross-reaction phenomena[3].

Glucagon is a relatively small peptide with a molecular weight of 3483 Da. As a consequence it is suitable for intact LC-MS/MS analysis, avoiding complicated proteolytic digestion steps [4]. Nevertheless its correct determination poses some challenges due to its very low expected concentration levels (in the picomolar range), together with a significant non-specific binding, poor solubility and stability issues in biological matrices. Consequently the preanalytical steps become a key factor for the correct quantitation of the hormone[5].

The aim of this work was to develop a method for LC-MS/MS glucagon quantitation based on a simple and straightforward sample preparation and to compare it with our routine method.

Methods

Proposed sample preparation.

Sample preparation was performed by means of a solid phase extraction (SPE) by using Oasis® MAX µElution 96-well plate (Waters, Milford, MA, USA). Before loading 400 µL of 1:1 diluted samples spiked with 20µL of internal standard solution, wells were conditioned with MeOH and NH4OH 2.5%(aq.). After two washing steps with ACN and NH4OH, samples were eluted with an acidic solution of acetonitrile/water (3:1).

Eight-point calibration curve levels were prepared in the range of 12.5 to 1000 pg/mL by spiking a compound standard (Genscript, NJ, USA) into commercially available Mass Spect Gold EDTA Plasma 7000 (Golden West Biologicals, CA, USA). A weighting of 1/x2 was applied for best curve fitness.

Liquid chromatography and tandem mass spectrometry.

A Nexera X2 ultraperformance liquid chromatography (UHPLC) system (Shimadzu Co.., Kyoto, Japan) coupled to a Sciex QTRAP® 6500 triple quadrupole MS/MS (Framingham, MA, USA) with an electrospray ion source operated in positive mode were used for identification and quantification of the compound. Three MRM transitions (one as quantifier and two as qualifiers) were selected both for the natural compound and for the amino acid-labelled internal standard.

Chromatographic separation was achieved using an ACQUITY® Peptide BEH C18 Column [130Å, 1.7 µm, 2.1 mm x 150 mm] (Waters, Milford, MA, USA) at 45ºC, applying a binary gradient of (A) 0.1% HCOOH(aq.) and (B) 0.1% HCOOH in ACN. The flow rate was set to 0.5 mL/min with a run time of 6.5 min. Injection volume was 40µL for extracted samples.

The newly developed method for LC-MS/MS was compared with our RIA routine assay by EURIA (Euro Diagnostica AB, Malmo, Sweden) in a Wallac Wizard 1470 Gamma Counter (PerkinElmer, MA, USA).

Study of glucagon stability in sample collection tubes.

A blood sample from 6 volunteers was drawn into BD™ P800 Blood Collection System and BD™ Vacutainer EDTA-K2 tubes (Becton Dickinson, NJ, USA). Immediately after collection, 1.8 mL of the sample in the latter tube were transferred to a Falcon tube containing 0.2 mL of an aprotinin (Trasylol®) solution. Therefore, three sets of plasma were obtained from each volunteer after centrifugation, plasma with and without aprotinin and plasma in P800 tubes.

Results

Preliminary validation results.

The present method is still under development and validation. Preliminary results show that the linearity of the calibration curve was successfully achieved between 12.5 and 1000 pg/mL, obtaining correlation coefficient (r) above 0.99 in all measurements.

Estimated limit of detection (LOD) and limit of quantification (LOQ) were calculated with the lowest concentration tested. LOD and LOQ were defined as 3:1 and 10:1 signal/noise ratio, these being 3.44 pg/mL and 11.47 pg/mL respectively.

Comparison of calibration curves by RIA and LC-MS/MS.

In a first approach we tested RIA calibration standards by LC-MS/MS, obtaining satisfactory recoveries ranging from 88% to 92%.

On the contrary, when analyzing LC-MS/MS calibration standards by RIA, a blank sample without any spiked glucagon returned a value of 175.2 pg/mL. This possible systematic bias was studied by subtracting this blank value (175.2 pg/mL) to every level in the calibration curve. After subtraction recoveries ranging from 75-135% were obtained.

Study of glucagon stability in sample collecting tubes.

Additionally, to assess glucagon stability in the preanalytical steps, samples collected in three different tubes, containing: EDTA-K2, EDTA-K2+Trasylol® and EDTA-K2+Proprietary Cocktail of Protease, Esterase and DPP-IV Inhibitors respectively were analyzed by RIA and LC-MS/MS simultaneously. Samples in the latter tubes presented higher concentrations compared to the other two in LC-MS/MS. In both techniques, samples in tubes containing Trasylol® presented lower concentration values, and those just containing EDTA-K2 gave similar results compared to the ones with the cocktail of inhibitors.

Conclusions & Discussion

In this project we have developed a LC-MS/MS method for glucagon quantitation in plasma with a rapid sample preparation based on a solid-phase extraction with an overall short run time. The selected analyte measuring interval from 12.5 pg/mL to 1000 pg/mL should be suitable for clinical applications.

The RIA method to which we compared our method with was cross-reacting with some other compounds in a glucagon-free plasma matrix. These results show the lack of specificity of the selected RIA method. Future works will be in line with the identification of these interferences.

BD™ P800 Blood Collection System tubes seemed to better stabilize glucagon and therefore higher concentration values were found when using LC-MS/MS, while no differences were seen in RIA. According to these results, the identification of other fragments rather than the entire glucagon remains a possibility in RIA determinations. Trasylol® had no positive impact in any of the performed studies.

[1] K.M. Habegger, K.M. Heppner, N. Geary, T.J. Bartness, R. DiMarchi, M.H. Tschöp, The metabolic actions of glucagon revisited, Nat. Rev. Endocrinol. 6 (2010) 689–697. doi:10.1038/nrendo.2010.187.

[2] A. Kolb, S. Rieder, D. Born, N. a Giese, T. Giese, G. Rudofsky, J. Werner, M.W. Büchler, H. Friess, I. Esposito, J. Kleeff, Glucagon/insulin ratio as a potential biomarker for pancreatic cancer in patients with new-onset diabetes mellitus., Cancer Biol. Ther. 8 (2009) 1527–33. doi:10.4161/cbt.8.16.9006.

[3] Y.-X. Li, M. Hackman, C.-Y. Wang, Quantitation of polypeptides (glucagon and salmon calcitonin) in plasma samples by ‘high resolution’ on a triple quadrupole mass spectrometer, Bioanalysis. 4 (2012) 685–691. doi:10.4155/bio.12.12.

[4] Y. V Karpievitch, A.D. Polpitiya, G.A. Anderson, R.D. Smith, A.R. Dabney, Liquid Chromatography Mass Spectrometry-Based Proteomics: Biological and Technological Aspects, Ann Appl Stat. 4 (2011) 1797–1823. doi:10.1214/10-AOAS341.Liquid.

[5] M.E. Lame, E.E. Chambers, S.S. Bangar, K.J. Fountain, Development of a High Sensitivity SPE-LC-MS / MS Assay for the Quantification of Glucagon in Human Plasma Using the ionKey / MS System, Waters Appl. Note. (2015) 1–11.