Elsevier

Analytica Chimica Acta

Volume 916, 15 April 2016, Pages 42-51
Analytica Chimica Acta

Quantitation of a PEGylated protein in monkey serum by UHPLC-HRMS using a surrogate disulfide-containing peptide: A new approach to bioanalysis and in vivo stability evaluation of disulfide-rich protein therapeutics

https://doi.org/10.1016/j.aca.2016.02.017Get rights and content

Highlights

  • A novel UHPLC-HRMS was developed to quantify a PEGylated protein in serum.

  • Two tryptic peptides were sensitively detected by HRMS with single ion monitoring.

  • Quantitation of one disulfide-containing peptide allows in vivo stability evaluation.

  • The assay was successfully applied to a non-clinical study for stability evaluation.

  • The method is applicable to disulfide rich-protein drugs not suited for SRM detection.

Abstract

To quantify a therapeutic PEGylated protein in monkey serum as well as to monitor its potential in vivo instability and methionine oxidation, a novel ultra high performance liquid chromatography-high resolution mass spectrometric (UHPLC-HRMS) assay was developed using a surrogate disulfide-containing peptide, DCP(SS), and a confirmatory peptide, CP, a disulfide-free peptide. DCP(SS) was obtained by eliminating the step of reduction/alkylation before trypsin digestion. It contains an intact disulfide linkage between two peptide sequences that are essential for drug function but susceptible to potential in vivo cleavages. HRMS-based single ion monitoring (SIM) on a Q Exactive™ mass spectrometer was employed to improve assay specificity and sensitivity for DCP(SS) due to its poor fragmentation and low sensitivity with SRM detection. The assay has been validated for the protein drug in monkey serum using both surrogate peptides with excellent accuracy (within ±4.4%Dev) and precision (within 7.5%CV) with a lower limit of quantitation (LLOQ) at 10 ng mL−1. The protein concentrations in monkey serum obtained from the DCP(SS)-based assay not only provided important pharmacokinetic parameters, but also confirmed in vivo stability of the peptide regions of interest by comparing drug concentrations with those obtained from the CP-based assay or from a ligand-binding assay (LBA). Furthermore, UHPLC-HRMS allowed simultaneous monitoring of the oxidized forms of both surrogate peptides to evaluate potential ex vivo/in vivo oxidation of one methionine present in each of both surrogate peptides. To the best of our knowledge, this is the first report of using a surrogate disulfide-containing peptide for LC-MS bioanalysis of a therapeutic protein.

Introduction

Recently, LC–MS technology has been increasingly used for protein and peptide quantitation in biological matrices to complement ligand-binding assays (LBA) in support of toxicokinetic (TK) or pharmacokinetic (PK) studies in drug discovery and development. An LC-MS based assay is capable of analyzing multiple drug related components within a single assay, as well as detecting in vivo modification of protein or peptides in specific regions of interest. A number of review papers have been published in recent years on protein and peptide quantification by LC-MS technology [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11].

Ideally, the concentration of protein or peptide drug would be measured by direct mass spectrometric detection of intact proteins or peptides [12], [13], [14]. However, the analysis of intact protein or peptide molecules is still very challenging due to the wide mass range requirement and low detection sensitivity caused by broad isotope distributions and multiply-charged ions [15]. Consequently, the concentrations of protein or peptide molecules have been predominantly measured by selected reaction monitoring (SRM) detection of one or more surrogate peptides using a triple quadrupole mass spectrometer [16], [17], [18]. The surrogate peptides are obtained via enzymatic digestion or chemical cleavage prior to LC-MS/MS analysis. The concentration of an intact drug molecule is obtained based on the stoichiometric relationship between the surrogate peptide and the intact protein or peptide molecule.

Disulfide-rich proteins are a group of small protein domains commonly found as independent (single-domain) proteins or as domains within larger polypeptides [19]. These proteins play a wide variety of roles, including growth factors (e.g. insulin-like growth factors (IGFs)), toxins, enzyme inhibitors, hormones, pheromones and allergens [19]. PEG-Protein-I, an investigational drug at Bristol-Myers Squibb (BMS), is a PEGylated small (<15 KDa) disulfide-rich protein. Structurally, it comprises multiple peptide chains linked by inter-chain disulfide bonds, with a 20 kDa PEG moiety attached to one of the peptide chains. Generally, LBA is the gold standard platform for protein quantitation in biological samples, and an LBA was previously validated and used to quantify PEG-Protein-I in serum samples in support of all toxicokinetic (TK) studies. However, the LBA only determined the total drug concentrations including the drug plus any modified forms of the drug. Previous reports suggested that the proteolytic degradation products of protein therapeutics may contribute to the systemic metabolic activity and possible side effects [20], [21]. In order to quantify and evaluate the in vivo stability of PEG-Protein-I, a site specific and sensitive LC-MS assay is needed. However, the presence of the PEGylated moiety in PEG-Protein-I makes it difficult to analyze it as the intact drug molecule by LC-MS due to several bioanalytical challenges, such as low mass spectrometric response, poor LC peak shape and potential contamination of the MS detector.

In this paper, we describe a UHPLC-HRMS method to quantify PEG-Protein-I in monkey serum using two tryptic surrogate peptides. One surrogate peptide is a disulfide-containing peptide, DCP(SS), which contains amino acids that are essential for the PEG-Protein-I function but susceptible to potential in vivo cleavage, while the other is a confirmatory peptide, CP, which is a disulfide-free peptide (Table 1). DCP(SS) was generated from an attempt to produce larger peptides that would more closely represent the intact molecule by eliminating the steps of reduction/alkylation before trypsin digestion. DCP(SS) contains an intact disulfide bond connecting two peptide sequences that include the labile amino acids of interest. Each surrogate peptide contains one methionine and can also be useful for evaluating potential in vivo oxidation of PEG-Protein-I. This UHPLC-HRMS assay to quantify PEG-Protein-I has been validated in monkey serum, and applied to a non-clinical toxicokinetic (TK) study in monkeys. To the best of our knowledge, this is the first report of using an intact disulfide-containing peptide as a surrogate peptide for LC-MS bioanalysis of a therapeutic disulfide bond-rich protein.

Section snippets

Materials and reagents

The investigational PEGylated protein drug candidate, PEG-Protein-I, was obtained from Research and Development, Bristol-Myers Squibb (BMS). Its chemical structure cannot be disclosed for proprietary reasons. The sequence and accurate monoisotopic mass information of the tryptic surrogate peptides and their internal standards (IS) used for quantitation of PEG-Protein-I in monkey serum are shown in Table 1. Four pairs of peptides were chemically synthesized by Discovery Chemistry within BMS

UHPLC–MS/MS method development and optimization

It was previously reported that the PEGylation of a protein or peptide changes its physicochemical behavior, making it possible to extract it into selected water-miscible organic solvents [25], [26], [27]. In particular, isopropanol with 0.1% formic acid was shown to efficiently extract a PEGylated protein of molecular weight of 11 KDa [26]. The same solvent has been used for another PEGylated protein drug candidate with consistent recovery of ∼100% [25], [26]. The presence of 0.1% formic acid

Conclusions

A UHPLC-HRMS method was developed and validated to quantify a PEGylated protein in monkey serum using a surrogate peptide containing a disulfide bridge connecting two peptide chains, in addition to another disulfide bond-free peptide as the confirmatory peptide, CP. SRM detection on a triple quadrupole mass spectrometer was not sensitive enough to detect the disulfide-containing peptide, DCP(SS), due to poor fragmentation. The application of UHPLC-HRMS allowed the sensitive detection of this

Acknowledgments

The authors would like to thank Dr. Binodh Desilva (Analytical & Bioanalytical Development, Bristol-Myers Squibb Company, Princeton, NJ 08540, USA) for her review of this manuscript.

References (31)

  • W.L. Nowatzke et al.

    Unique challenges of providing bioanalytical support for biological therapeutic pharmacokinetic programs

    Bioanalysis

    (2011)
  • N.C. van de Merbel et al.

    Protein quantification using LC-MS: can it make a difference?

    Bioanalysis

    (2012)
  • Y.J. Zhang et al.

    The integration of ligand binding and LC-MS-based assays into bioanalytical strategies for protein analysis

    Bioanalysis

    (2014)
  • J. Zheng et al.

    Application and challenges in using LC-MS assays for absolute quantitative analysis of therapeutic proteins in drug discovery

    Bioanalysis

    (2014)
  • X. Zhu et al.

    Quantification of recombinant human relaxin-2 (B-29/A-24) in non-pregnant rat plasma using ultra performance liquid chromatography-mass spectrometry

    Anal. Methods

    (2014)
  • Cited by (10)

    • Development of a simple non-reduced peptide mapping method that prevents disulfide scrambling of mAbs without affecting tryptic enzyme activity

      2022, Journal of Pharmaceutical and Biomedical Analysis
      Citation Excerpt :

      Tris buffer performed well in this study, but solvent may not impact the performance of the cystamine-added non-reduced peptide mapping method much as long as it does not inhibit tryptic activity. For the PK studies of some disulfide-rich protein therapeutics like PEGylated protein, disulfide peptides have to be selected as surrogate peptides for target quantification to evaluate its stability in vivo [32]. The quantification accuracy and precision could be compromised in regular non-reduced sample preparation because of disulfide scrambles in basic pH digestion conditions.

    • Bioanalysis of biological matrix samples using liquid chromatography–tandem mass spectrometry detection

      2020, Specification of Drug Substances and Products: Development and Validation of Analytical Methods, Second Edition
    • New developments and possibilities of wide-pore superficially porous particle technology applied for the liquid chromatographic analysis of therapeutic proteins

      2018, Journal of Pharmaceutical and Biomedical Analysis
      Citation Excerpt :

      Interestingly, due to differences in diffusion properties, 1.3 μm particles seemed to be more favourable for the separation of peptides than for small molecules when using longer columns and longer gradient times [57]. Finally, the 1.6 μm SPPs have been used for the determination of PEGprotein from biological matrix following tryptic digestion [58]. For interested readers, numerous further applications can be found in brochures of manufacturers reflecting that widepore SPP technology has gained more and more interest in modern analytical laboratories.

    • Protein Stability: Enhancement and Measurement

      2023, Methods in Molecular Biology
    View all citing articles on Scopus
    1

    Current address: Office of Bioequivalence, Division II, Office of Generic Drugs, Center for Drug Evaluation and Research, US Food and Drug Administration, 10903 New Hampshire Ave, Silver Spring, MD 20993, USA.

    View full text