Characterisation of intact recombinant human erythropoietins applied in doping by means of planar gel electrophoretic techniques and matrix-assisted laser desorption/ionisation linear time-of-flight mass spectrometry

Proteoforms and their expanding role in laboratory medicine

The term “proteoforms” describes the range of different structures of a protein product of a single gene, including variations in amino acid sequence and post-translational modifications. This diversity in protein structure contributes to the biological complexity observed in living organisms. As the concentration of a particular proteoform may increase or decrease in abnormal physiological states, proteoforms have long been used in medicine as biomarkers of health and disease. Notably, the analytical approaches used to analyze proteoforms have evolved considerably over the years. While ligand binding methods continue to play a large role in proteoform measurement in the clinical laboratory, unanticipated or unknown post-translational modifications and sequence variants can upend even extensively tested and vetted assays that have successfully made it through the medical regulatory process. As an alternate approach, mass spectrometry—with its high molecular selectivity—has become an essential tool in detection, characterization, and quantification of proteoforms in biological fluids and tissues. This review explores the analytical techniques used for proteoform detection and quantification, with an emphasis on mass spectrometry and its various applications in clinical research and patient care including, revealing new biomarker targets, helping improve the design of contemporary ligand binding in vitro diagnostics, and as mass spectrometric laboratory developed tests used in routine patient care.

NEGATIVE ION MASS SPECTROMETRY FOR THE ANALYSIS OF N-LINKED GLYCANS

N-glycans from glycoproteins are complex, branched structures whose structural determination presents many analytical problems. Mass spectrometry, usually conducted in positive ion mode, often requires extensive sample manipulation, usually by derivatization such as permethylation, to provide the necessary structure-revealing fragment ions. The newer but, so far, lesser used negative ion techniques, on the contrary, provide a wealth of structural information not present in positive ion spectra that greatly simplify the analysis of these compounds and can usually be conducted without the need for derivatization. This review describes the use of negative ion mass spectrometry for the structural analysis of N-linked glycans and emphasises the many advantages that can be gained by this mode of operation. Biosynthesis and structures of the compounds are described followed by methods for release of the glycans from the protein. Methods for ionization are discussed with emphasis on matrix-assisted laser desorption/ionization (MALDI) and methods for producing negative ions from neutral compounds. Acidic glycans naturally give deprotonated species under most ionization conditions. Fragmentation of negative ions is discussed next with particular reference to those ions that are diagnostic for specific features such as the branching topology of the glycans and substitution positions of moieties such as fucose and sulfate, features that are often difficult to identify easily by conventional techniques such as positive ion fragmentation and exoglycosidase digestions. The advantages of negative over positive ions for this structural work are emphasised with an example of a series of glycans where all other methods failed to produce a structure. Fragmentation of derivatized glycans is discussed next, both with respect to derivatives at the reducing terminus of the molecules, and to methods for neutralization of the acidic groups on sialic acids to both stabilize them for MALDI analysis and to produce the diagnostic fragments seen with the neutral glycans. The use of ion mobility, combined with conventional mass spectrometry is described with emphasis on its use to extract clean glycan spectra both before and after fragmentation, to separate isomers and its use to extract additional information from separated fragment ions. A section on applications follows with examples of the identification of novel structures from lower organisms and tables listing the use of negative ions for structural identification of specific glycoproteins, glycans from viruses and uses in the biopharmaceutical industry and in medicine. The review concludes with a summary of the advantages and disadvantages of the technique. © 2020 John Wiley & Sons Ltd. Mass Spec Rev.

Isoforms analysis of recombinant human erythropoietin by polarity-reversed capillary isoelectric focusing

Recombinant human erythropoietin (rhuEPO) has been extensively used as a pharmaceutical product for treating anemia in the clinic. Glycosylation of rhuEPO was crucial for affecting biological activity, immunogenicity, and pharmacokinetics. Because of the heterogeneity of glycan, the structure of rhuEPO was complex with several isoforms. Characterization of isoforms was important for quality control of rhuEPO. Here, an improved cIEF method has been established and validated. A polarity-reversed focusing step was used by reversing both the polarity of the voltage and the catholyte and anolyte vials. A weak base (100 mM ammonium hydroxide solution) was used as a chemical mobilizer to make the acidic bands mobilize stably to the detection window. Compared with CZE method in European Pharmacopoeia, the numbers of isoforms and their peak area percentage were highly consistent. Better reproducibility and higher resolution have been obtained by the improved cIEF method. Moreover, in improved cIEF method, the isoelectric points (pI) of each isoform can be calculated and used for identification. It was also the first time that the cIEF method was fully validated for rhuEPO analysis according to the International Conference on Harmonization (ICH) guidelines.

Current landscape of protein glycosylation analysis and recent progress toward a novel paradigm of glycoscience research

This review covers the basics and some applications of methodologies for the analysis of glycoprotein glycans. Analytical techniques used for glycoprotein glycans, including liquid chromatography (LC), capillary electrophoresis (CE), mass spectrometry (MS), and high-throughput analytical methods based on microfluidics, were described to supply the essentials about biopharmaceutical and biomarker glycoproteins. We will also describe the MS analysis of glycoproteins and glycopeptides as well as the chemical and enzymatic releasing methods of glycans from glycoproteins and the chemical reactions used for the derivatization of glycans. We hope the techniques have accommodated most of the requests from glycoproteomics researchers.

Blood doping: risks to athletes' health and strategies for detection

Blood doping has been defined as the misuse of substances or certain techniques to optimize oxygen delivery to muscles with the aim to increase performance in sports activities. It includes blood transfusion, administration of erythropoiesis-stimulating agents or blood substitutes, and gene manipulations. The main reasons for the widespread use of blood doping include: its availability for athletes (erythropoiesis-stimulating agents and blood transfusions), its efficiency in improving performance, and its difficult detection. This article reviews and discusses the blood doping substances and methods used for in sports, the adverse effects related to this practice, and current strategies for its detection.

Differences in sialic acid O-acetylation between human urinary and recombinant erythropoietins: a possible mass spectrometric marker for doping control

Development of a mass spectrometric method for the unambiguous detection of doping with recombinant human erythropoietins (rhEPO) has been attempted for many years. Unfortunately, progress in this field was hampered by the unavailability of highly purified human endogenous EPOs (urinary[uhEPO], serum/plasma EPO)--a prerequisite for generating detailed mass spectrometric glycosylation data necessary for revealing significant differences between uhEPO and rhEPOs. The paper presents the worldwide first analytical data on purified human urinary EPO generated with a high resolution high accuracy mass spectrometer (LTQ-Orbitrap). The focus is on the tryptic O-glycopeptide (E117-R131) and its degree of sialic acid O-acetylation. Data are compared with results obtained from 40 rhEPO pharmaceuticals. It could be demonstrated that the O-glycopeptide of uhEPO (ca 100 IU) contains only trace amounts of mono-acetylated mono-and di-sialylated O-glycans but no other O-acetylated structures and in this respect significantly differs from all rhEPOs. Moreover, Dynepo--a rhEPO previously thought to be not O-acetylated--also contains small amounts of O-acetylations within the O-glycan structure. The results might be useful for anti-doping purposes as well as the development of EPO pharmaceuticals with closer structural similarity to the endogenous hormone.

The overlooked difference between human endogenous and recombinant erythropoietins and its implication for sports drug testing and pharmaceutical drug design

Sequential deglycosylation by exoglycosidase treatment (Reagent Array Analysis Method, RAAM) and subsequent sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) revealed a profound structural difference between human endogenous and recombinant erythropoietins. While both proteins behaved similarly upon digestion with Arthrobacter ureafaciens α-sialidase and Steptococcus pneumoniae β-D-galactosidase, the action of N-acetyl-β-D-glucosaminidase from Steptococcus pneumoniae was partly blocked by endogenous but not recombinant erythropoietins. Consequently, further treatment with Jack bean α-D-mannosidase and Helix pomatia β-D-mannosidase led to only very limited additional deglycosylation of endogenous EPO, while rhEPO glycans continued to be degraded. The behaviour was visualized by SDS-PAGE combined with Western blotting. While the apparent molecular masses of most endogenous glycoforms did not further decrease after treatment with the first three enzymes, masses of most rhEPO glycoforms continued to drop after digestion with the two mannosidases. Both, human urinary and serum EPO showed this blocking effect, and all of the tested 28 recombinant epoetins were accessible to further degradation by exo-mannosidases. The majority of EPO pharmaceuticals is produced in Chinese hamster ovary (CHO) cell lines, few in other ones (i.e. baby hamster kidney (BHK) or human fibrosarcoma (HT-1080) cells). Since human endogenous EPO is primarily produced by the kidneys, tissue specific glycosylation might explain the altered deglycosylation behaviour. This difference was overlooked since EPO was first isolated from human urine in 1977. The results might prove useful for anti-doping testing and future EPO drug development.

Optimization of erythropoietin production with controlled glycosylation-PEGylated erythropoietin produced in glycoengineered Pichia pastoris

Pichia pastoris is a methylotropic yeast that has gained great importance as an organism for protein expression in recent years. Here, we report the expression of recombinant human erythropoietin (rhEPO) in glycoengineered P. pastoris. We show that glycosylation fidelity is maintained in fermentation volumes spanning six orders of magnitude and that the protein can be purified to high homogeneity. In order to increase the half-life of rhEPO, the purified protein was coupled to polyethylene glycol (PEG) and then compared to the currently marketed erythropoiesis stimulating agent, Aranesp(®) (darbepoetin). In in vitro cell proliferation assays the PEGylated protein was slightly, and the non-PEGylated protein was significantly more active than comparator. Pharmacodynamics as well as pharmacokinetic activity of PEGylated rhEPO in animals was comparable to that of Aranesp(®). Taken together, our results show that glycoengineered P. pastoris is a suitable production host for rhEPO, yielding an active biologic that is comparable to those produced in current mammalian host systems.

Analysis of carbohydrates and glycoconjugates by matrix-assisted laser desorption/ionization mass spectrometry: an update for the period 2005-2006

This review is the fourth update of the original review, published in 1999, on the application of MALDI mass spectrometry to the analysis of carbohydrates and glycoconjugates and brings coverage of the literature to the end of 2006. The review covers fundamental studies, fragmentation of carbohydrate ions, method developments, and applications of the technique to the analysis of different types of carbohydrate. Specific compound classes that are covered include carbohydrate polymers from plants, N- and O-linked glycans from glycoproteins, glycated proteins, glycolipids from bacteria, glycosides, and various other natural products. There is a short section on the use of MALDI-TOF mass spectrometry for the study of enzymes involved in glycan processing, a section on industrial processes, particularly the development of biopharmaceuticals and a section on the use of MALDI-MS to monitor products of chemical synthesis of carbohydrates. Large carbohydrate-protein complexes and glycodendrimers are highlighted in this final section.

SDS-PAGE of recombinant and endogenous erythropoietins: benefits and limitations of the method for application in doping control

Doping of athletes with recombinant and genetically modified erythropoietins (EPO) is currently detected by isoelectric focusing (IEF). The application of these drugs leads to a significant change in the isoform profile of endogenous urinary erythropoietin (uhEPO). Dynepo, MIRCERA, biosimilars with variable IEF-profiles as well as active urines and effort urines have made additional testing strategies necessary. The new generation of small molecule EPO-receptor stimulating agents like Hematide will also challenge the analytical concept of detecting the abuse of erythropoiesis stimulating agents (ESA). By determining their apparent molecular masses with SDS-PAGE a clear differentiation between endogenous and exogenous substances also concerning new EPO modifications is possible. Due to the orthogonal character of IEF- and SDS-PAGE both methods complement each other. The additional benefits of SDS-PAGE especially in relation to active and effort urines as well as the detection of Dynepo were investigated. Due to significant differences between the apparent molecular masses of uhEPO/serum EPO (shEPO) and recombinant, genetically or chemically modified erythropoietins the presence of active or effort urines was easily revealed. The characteristic band shape and apparent molecular mass of Dynepo on SDS-PAGE additionally evidenced the presence of this substance in urine. A protocol for the detection of EPO-doping in serum and plasma by SDS-PAGE was developed. Blood appears to be the ideal matrix for detecting all forms ESA-doping in the future.

Procedures for monitoring recombinant erythropoietin and analogs in doping

Hemoglobin concentration is one of the principal factors of aerobic power and, consequently, of performance in many types of physical activities. The use of recombinant human erythropoietin is, therefore, particularly powerful for improving the physical performances of patients, and, more generally, improving their quality of life. This article discusses procedures for monitoring recombinant erythropoietin and its analogues in doping for athletic performance.

Erythropoietin and analogs

Erythropoietin (EPO), a glycoprotein hormone, stimulates the growth of red blood cells and as a consequence it increases tissue oxygenation. This performance enhancing effect is responsible for the ban of erythropioetin in sports since 1990. Especially its recombinant synthesis led to the abuse of this hormone, predominatly in endurance sports. The analytical differentiation of endogenously produced erythropoietin from its recombinant counterpart by using isoelectric focusing and double blotting is a milestone in the detection of doping with recombinant erythropoietin. However, various analogous of the initial recombinant products, not always easily detectable by the standard IEF-method, necessitate the development of analytical alternatives for the detection of EPO doping. The following chapter summarizes its mode of action, the various forms of recombinant erythropoietin, the main analytical procedures and strategies for the detection of EPO doping as well as a typical case report.

The analytical chemistry of drug monitoring in athletes

The detection and deterrence of the abuse of performance-enhancing drugs in sport are important to maintaining a level playing field among athletes and to decreasing the risk to athletes' health. The World Anti-Doping Program consists of six documents, three of which play a role in analytical development: The World Anti-Doping Code, The List of Prohibited Substances and Methods, and The International Standard for Laboratories. Among the classes of prohibited substances, three have given rise to the most recent analytical developments in the field: anabolic agents; peptide and protein hormones; and methods to increase oxygen delivery to the tissues, including recombinant erythropoietin. Methods for anabolic agents, including designer steroids, have been enhanced through the use of liquid chromatography/tandem mass spectrometry and gas chromatography/combustion/isotope-ratio mass spectrometry. Protein and peptide identification and quantification have benefited from advances in liquid chromatography/tandem mass spectrometry. Incorporation of techniques such as flow cytometry and isoelectric focusing have supported the detection of blood doping.

Structural analysis of the glycosylation of gene-activated erythropoietin (epoetin delta, Dynepo)

Recently, a novel recombinant human erythropoietin (epoetin delta, Dynepo) has been marketed in the European Union for the treatment of chronic kidney disease, cancer patients receiving chemotherapy, and so forth. Epoetin delta is engineered in cultures of the human fibrosarcoma cell line HT-1080 by homologous recombination and "gene activation." Unlike recombinant erythropoietins produced in other mammalian cells, epoetin delta is supposed to have a human-type glycosylation profile. However, the isoelectric focusing profile of epoetin delta differs from that of endogenous erythropoietin (both urinary and plasmatic). In this work, structural and quantitative analysis of the O- and N-glycans of epoetin delta was performed and compared with glycosylation from recombinant erythropoietin produced in Chinese hamster ovary (CHO) cells. From the comparison, significant differences in the sialylation of O-glycans were found. Furthermore, the N-glycan analysis indicated a lower heterogeneity from epoetin delta when compared with its CHO homologue, being predominantly tetraantennary without N-acetyllactosamine repeats in the former. The sialic acid characterization revealed the absence of N-glycolylneuraminic acid. The overall sugar profiles of both glycoproteins appeared to be significantly different and could be useful for maintaining pharmaceutical quality control, detecting the misuse of erythropoietin in sports, and establishing new avenues to link glycosylation with biological activity of glycoproteins.

Low LC-MS/MS detection of glycopeptides released from pmol levels of recombinant erythropoietin using nanoflow HPLC-chip electrospray ionization

The test used by anti-doping laboratories to detect the misuse of recombinant erythropoietin (rhEPO) is based on its different migration pattern on isoelectric focusing (IEF) gel compared with the endogenous human erythropoietin (hEPO) that can possibly be explained by structural differences. While there is definitely a need to identify those differences by LC-MS/MS, the extensive characterization that was achieved for the rhEPO was never performed on human endogenous EPO because its standard is not available in sufficient amount. The goal of this study was to develop an analytical method to detect pmol amounts of N-linked and O-linked glycopeptides of the recombinant hormone as a model. Using a nanoflow HPLC-Chip electrospray ionization/ion trap mass spectrometer, the diagnostic ion at m/z 366 of oligosaccharides was monitored in the product ion spectra to identify the four theoretical glycosylation sites, Asn24, Asn38, Asn83 and Ser126, respectively, on glycopeptides 22-37, 38-55, 73-96 and 118-136. With 3 pmol of starting material applied on Chip, only the desialylated N-glycopeptides 22-37 and 38-55/38-43 could be observed, and of all the glycan isoforms, those with the smaller structures were predominantly detected. While the preservation of the sialic acid moieties decreased the detection of all the N-glycopeptides, it allowed a more extensive characterization of the O-linked glycopeptide 118-136. The technique described herein provides a mean to detect glycopeptides from commercially available pharmaceutical preparations of rhEPO with the sensitivity required to analyze pmol amounts of hEPO, which could ultimately lead to the identification of structural differences between the recombinant and the human forms of the hormone.

Evaluation of protein N-glycosylation in 2-DE: Erythropoietin as a study case

The structure, function, and physico-chemical properties of many proteins are determined by PTM, being glycosylation the most complex. This study describes how a combination of typical proteomics methods (2-DE) combines with glycomics strategies (HPLC, MALDI-TOF-MS, exoglycosidases sequencing) to yield comprehensive data about single spot-microheterogeneity, providing meaningful information for the detection of disease markers, pharmaceutical industry, antidoping control, etc. Recombinant erythropoietin and its hyperglycosylated analogue darbepoetin-alpha were chosen as showcases because of their relevance in these fields and the analytical challenge they represent. The combined approach yielded good results in terms of sample complexity (mixture glycoforms), reproducibility, sensitivity ( approximately 25 pmoles of glycoprotein/spot), and identification of the underlying protein. Heterogeneity was present in all spots but with a clear tendency; spots proximal to the anode contained the highest amount of tetra-antennary tetra-sialylated glycans, whereas the opposite occurred for spots proximal to the cathode with the majority of the structures being undersialylated. Spot microheterogeneity proved a consequence of the multiple glycosylation sites as they contributed directly to the number of possibilities to account for a discrete charge in a single spot. The interest of this combined glycoproteomics method resides in the efficiency for detecting and quantifying subtle dissimilarities originated from altered ratios of identical glycans including N-acetyl-lactosamine repeats, acetylation, or antigenic epitopes, that do not significantly contribute to the electrophoretic mobility, but affect the glycan microheterogeneity and the potential underlying related functionality.

LC−MS/MS Method for Confirmation of Recombinant Human Erythropoietin and Darbepoetin α in Equine Plasma

Recombinant human erythropoietin (rhEPO) and darbepoetin alpha (DPO) are protein-based drugs for the treatment of anemia by stimulating red blood cell production. Consequently, they are abused in human and equine sports. To deter their abuse in the horse racing industry, a sensitive and reliable method for confirmation of these agents in equine plasma has been in urgent need. Such a method by LC-MS/MS is described in this paper. The method involved analyte enrichment by immunoaffinity separation using anti-rhEPO antibody linked to magnetic beads, digestion by trypsin, and analysis by LC-MS/MS. Two specific proteotypic peptides, 46VNFYAWK52 and 144VYSNFLR150 from rhEPO and DPO were employed for confirmation of the analytes based on chromatographic retention times and major product ions. The limit of confirmation of this method was 0.2 ng/mL, and the limit of detection was 0.1 ng/mL for rhEPO and DPO in equine plasma. This method was successful in confirming the presence of rhEPO and DPO in plasma samples collected from research horses to which rhEPO or DPO was administered and from racehorses following competition and in noncompetition samples in North America. To our knowledge, this is the first LC-MS method with adequate sensitivity and specificity in providing unequivocal confirmation of rhEPO and DPO in equine plasma samples. This method provides a powerful enforcement tool that was lacking in the fight against the abuse of rhEPO and DPO in the horse racing industry.

Towards a reliable molecular mass determination of intact glycoproteins by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

Different matrices and sample-matrix preparation procedures have been tested in order to study their influence on the matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectra of intact glycoproteins, which present different degrees of glycosylation (human transferrin; bovine fetuin; bovine alpha(1)-acid-glycoprotein; recombinant human erythropoietin; and the novel erythropoiesis stimulating protein). Using sinapinic acid (SA) and the fast evaporation method, the studied glycoproteins became susceptible to fragmentation at any laser intensity, suggesting that this 'hot' matrix is unsuitable for a reliable molecular mass determination of glycosylated compounds. In contrast, 2,5-dihydroxybenzoic acid (DHB) and 6-aza-2-thiothymine (ATT), with an adequate sample-matrix preparation, provided improved results. Samples containing DHB after crystallization by vacuum drying demonstrated the best performance because the labile functional groups from the glycoforms were apparently fragmented to a lower extent. The average molecular masses obtained using this methodology were in all cases a better estimation than those values reported in the literature. The results were reproducible, and sensitivity was similar to that obtained with SA and the fast evaporation method. These excellent results suggest that this MALDI-TOF-MS methodology could be useful for an improved determination of the average molecular mass values of microheterogeneous compounds such as glycoproteins, glycosylated compounds or, in general, molecular mass values of molecules with similar labile functional groups.

Assessing the instability of the isoelectric focusing patterns of erythropoietin in urine

IEF can be used to differentiate human urinary erythropoietin (uEPO), recombinant human erythropoietin or epoetin (rEPO) and darbepoetin (novel erythropoiesis stimulating protein (NESP)). This is the basis of the method currently used to detect misuse of rEPO and NESP by elite athletes. Recently, an unknown activity has been attributed to some urine samples (denominated 'unstable' urine by the World Anti-Doping Agency; WADA). This activity has shown to give rise to artefactual profiles for both rEPO and NESP when incubated with such urine and, thus, raised concerns with respect to doping control. We have evaluated which charges produce the characteristic IEF profiles of uEPO, rEPO and NESP and how these profiles respond to distinct enzymatic reactions. From sialidase digestions it became evident that only uEPO contains charges different from sialic acid, and a comparison of all substances after complete de-N-glycosylation localized these charges in the carbohydrate moiety. Partial desialylation, or digestion with arylsulfatase from Helix pomatia yielded profiles for recombinants species similar to those observed for unstable urine samples. The contributions from our studies to the anti-doping problem include: (i) protocols that may corroborate the potential misuse of rEPO or NESP based on the particular enzymatic activity of an arylsulfatase preparation, or a broad-specificity sialidase; (ii) assurance that the instability observed in some urine samples may only result from false-negatives, but not from false-positive testing; and (iii) a simple remedy to prevent an unstable urine from altering the IEF profile by adding selective competitive substrates.

Mass Spectrometry-Based Glycoproteomic Approach Involving Lysine Derivatization for Structural Characterization of Recombinant Human Erythropoietin

Lysine-containing peptides comprising glycosylation sites derived from recombinant human erythropoietin (rHuEPO) by trypsin or Lys-C and PNGase F dual digestion were derivatized with 2-methoxy-4,5-dihydro-1H-imidazole and its deuterated analogues. In the same reaction, under reducing conditions (beta-mercaptoethanol), cysteines were converted into methyl-cysteines and lysines into Lys-4,5-dihydro-1H-imidazole. Both modifications on cysteines and lysines simplified the CID-MS/MS spectra, while preserving the structural information by yielding y-series ions and improved the mass spectral signal intensity up to 25 times. Moreover, by this approach, the N-glycan occupation sites were unambiguously determined. O-Glycosylation sites as well as O-glycan structures were determined by a LC-MS/MS experiment carried out on dually digested rHuEPO. N-Glycan mixture purified on a graphitized carbon column using a newly developed method that extracted only sialylated carbohydrates was analyzed first using MALDI-TOF in negative linear ion mode with low mass accuracy but without interferences and metastabile ions and then a reflectron with high mass accuracy. After defining the precursor ions, we performed the nanoESI QTOF MS/MS analysis on N-glycans, mainly targeting the distinction between carbohydrates with sialylated antennae and those lacking sialic acid moieties.

Glycoform characterization of erythropoietin combining glycan and intact protein analysis by capillary electrophoresis – electrospray – time-of-flight mass spectrometry

Glycosylation of recombinant human erythropoietin (rHuEPO) is a post-translational process that alters biological activity, solubility and lifetime of the glycoprotein in blood, and strongly depends on the type of cell and the cell culture conditions. A fast and simple method providing extensive carbohydrate information about the glycans present in rHuEPO and other glycoproteins is needed in order to improve current methods in drug development or product quality control. Here, an improved method for intact rHuEPO glycoform characterization by CZE-ESI-TOF MS has been developed using a novel capillary coating and compared to a previous study. Both methods allow a fast separation in combination with accurate mass characterization of the single protein isoforms. The novel dynamic coating provides a separation at an EOF close to zero, enabling better separation. This results in an improved mass spectrometric resolution and the detection of minor isoforms. In order to assign an unequivocal carbohydrate composition to every intact glycoform, a CZE-ESI-MS separation method for enzymatically released underivatized N-glycans has been developed. The TOF MS allows the correct identification of the glycans due to its high mass accuracy and resolution. Therefore, glycan modifications such as acetylation, oxidation, sulfation and even the exchange of OH by NH(2) are successfully characterized. Information of the protein-backbone molecular mass has been combined with results from peptide analysis (revealing information about O-glycosylation) and from the glycan analysis, including the detection of as yet undescribed glycans containing four antennae and five sialic acids. This allows an unequivocal assignment of an overall glycosylation composition to the molecular masses obtained for the intact rHuEPO glycoforms.

Characterization ofN- andO-glycopeptides of recombinant human erythropoietins as potential biomarkers for doping analysis by means of microscale sample purification combined with MALDI-TOF and quadrupole IT/RTOF mass spectrometry

The structural characterization of the O- and N-glycan structures of three different commercially available recombinant human erythropoietins (rhEPOs) is represented by means of a microscale sample purification using ZipTip technology and MALDI-TOF and MALDI low-energy CID MS. Glycopeptides were released from rhEPO samples by a differential endoproteolytic digestion to obtain site-specific glycosylation patterns. Mass accuracies in the range of +/- 0.04% obtained by the high-resolution TOF instrument allowed an unambiguous assignment of N-glycan structures via glycan database software. Furthermore, the O-glycan structures were directly analyzed on the glycopeptide level by MS/MS experiments. Principally, site-specific glycosylation was found to be very similar for the three different rhEPOs (EPO-alpha, EPO-beta, and novel erythropoiesis stimulating protein (NESP)) but exhibiting quantitative differences in distinct O- and N-glycan moieties. Significant differences were found in the degree of sialylation and acetylation. Especially, a considerable degree of variation of the O-acetylation of sialic acid residues could be realized on the glycan structures of O- and N-glycopeptides, whereas EPO-alpha and EPO-beta could be clearly differentiated from NESP solely on the O-glycopeptide level.