Search

Glycosylation is considered a critical quality attribute of therapeutic proteins. Protein heterogeneity introduced by glycosylation includes differences in the nature, number and position of the glycans. Whereas analysis of released glycans and glycopeptides provides information about the composition and/or position of the glycan, intact glycoprotein analysis allows assignment of individual proteoforms and co-occurring modifications. Yet, resolving protein glycoforms at the intact level is challenging. We have explored the capacity of hydrophilic liquid chromatography-mass spectrometry (HILIC-MS) for assessing glycosylation patterns of intact pharmaceutical proteins by analyzing the complex glycoproteins interferon-beta-1a (rhIFN-β - 1a) and recombinant human erythropoietin (rhEPO). Efficient glycoform separation was achieved using a superficially-porous amide HILIC stationary phase and trifluoroacetic acid (TFA) as eluent additive. In-source collision-induced dissociation proved to be very useful to minimize protein-signal suppression effects by TFA. Direct injection of therapeutic proteins in aqueous formulation was possible without causing extra band dispersion, provided that the sample injection volume was not larger than 2 μL. HILIC-MS of rhIFN-β - 1a and rhEPO allowed the assignment of, respectively, 15 and 51 glycoform compositions, next to a variety of posttranslational modifications, such as succinimide, oxidation and N-terminal methionine-loss products. MS-based assignments showed that neutral glycan units significantly contributed to glycoform separation, whereas terminal sialic acids only had a marginal effect on HILIC retention. Comparisons of HILIC-MS with the selectivity provided by capillary electrophoresis-MS for the same glycoproteins, revealed a remarkable complementarity of the techniques. Finally it was demonstrated that by replacing TFA for difluoroacetic acid, peak resolution somewhat decreased, but rhEPO glycoforms with relative abundances below 1% could be detected by HILIC-MS, increasing the overall rhEPO glycoform coverage to 72., (Copyright © 2018 The Authors. Published by Elsevier B.V. All rights reserved.)

A capillary zone electrophoresis (CZE) method with UV detection was developed for the quantification of the E.colil-asparaginase (l-ASNase) and its acidic variants. During the initial method development, a variety of experimental conditions were screened. Subsequently, a Design of Experiments (DoE) was used to optimize the pH and concentration of the selected background electrolyte (BGE) containing both TRIS and boric acid. Optimization was performed taking into account both the separation efficiency of l-ASNase and its acidic variants as well as overall method robustness. A repeatable separation between E.colil-ASNase and its acidic variants was achieved on a bare fused silica capillary in combination with a BGE consisting of both 400 mM TRIS and boric acid. The method was validated for linearity, accuracy, precision, LOD, LOQ and robustness. The recovery for l-ASNase was 97.9-104.4% with a precision RSD of 1.5-3.2%, while the recovery of impurities was 92.1-109.8% with a RSD of 1.7-4.6%. The quantification limit was 1.9% (m/m). Moreover, the CZE-UV method was applied to determine the degradation rate in the presence of ammonium bicarbonate, confirming the suitability of the method. The degraded, partially charged l-ASNase was evaluated for its in-vitro enzymatic activity showing an insignificant different enzyme activity compared to the unmodified sample., (Copyright © 2018 Elsevier B.V. All rights reserved.)

We describe the development and evaluation of a GC-MS fractionation platform that combines high-resolution fraction collection of full chromatograms with parallel MS detection. A y-split at the column divides the effluent towards the MS detector and towards an inverted y-piece where vaporized trap solvent is infused. The latter flow is directed outside the GC oven allowing subsequent condensation and stepwise collection of liquid fractions with trapped analytes on a 384-well plate. For study and optimization of the effluent split ratio, restriction capillaries of different lengths and diameters were evaluated. For a wide range of settings, local pressures were monitored during fractionation to assess the influence of MS vacuum and trap solvent infusion on the GC system stability. The platform performance was evaluated by GC-MS analysis and continuous fractionation of an n-alkane mixture followed by GC analysis of each fraction. Comparison of the on-line recorded and fraction-reconstructed chromatogram showed the GC separation is maintained during fractionation. Multiple fractionation cycles of the n-alkane sample on the same 384-well plate yielded a reconstructed chromatogram which was highly similar to that of a single analysis, demonstrating the high repeatability. The applicability of the GC-MS-fractionation platform for bioactivity screening was investigated by applying the AR-Ecoscreen reporter gene bioassay on fractions obtained after analysis of standard solutions and dust samples containing the anti-androgenic compounds vinclozolin and p,p'-DDE., (Copyright © 2017 Elsevier B.V. All rights reserved.)

During development of adenovirus-based vaccines, samples have to be analyzed in order to either monitor the production process or control the quality and safety of the product. An important quality attribute is the total concentration of intact adenoviruses, which currently is determined by quantitative polymerase chain reaction (qPCR) or anion exchange-HPLC. Capillary Electrophoresis (CE) was evaluated as alternative to the current methods with the aim to have one single method that allows reliable and fast quantification of adenovirus particles throughout the full process. Intact adenoviruses samples from downstream processing and upstream processing were analyzed directly by CE with UV-detection at 214nm. Only the samples with high amounts of DNA required a simple sample pretreatment by benzonase. Adenovirus particles were separated from matrix components such as cell debris, residual cell DNA, and/or proteins on a PVA-coated capillary using a BGE consisting of 125mM Tris, 338mM tricine and 0.2% v/v polysorbate-20 at pH 7.7. Full factorial design of experiments was used for method optimization as part of the analytical quality by design (AQbD) method development approach. The method was validated for the quantification of adenoviruses on five representative samples from the manufacturing process in the range of 0.5×10 11 -1.5×10 11 adenovirus particles per ml (~80 to 250pmol/l). The CE method showed intermediate precision of 7.8% RSD on concentration and an accuracy (spiked recovery) of 95-110%. CE proved highly useful for process development support and is being implemented for in-process control testing for adenovirus vaccine manufacturing., (Copyright © 2017 Elsevier B.V. All rights reserved.)

Current methods for the identification and/or quantification of viral proteins in influenza virus and virosome samples suffer from long analysis times, limited protein coverage and/or low accuracy and precision. We studied and optimized capillary gel electrophoresis (CGE) in order to achieve faster and enhanced characterization and quantification of viral proteins. Sample preparation as well the composition of the gel buffer was investigated in order to achieve adequate protein separation in relatively short times. The total sample preparation (reduction and deglycosylation) could be carried out efficiently within two hours. Hydrodynamic injection, separation voltage, and capillary temperature were optimized in full factorial design. The final method was validated and showed good performance for hemagglutinin fragment 1 (HA1), hemagglutinin fragment 2 (HA2), matrix protein (M) and nucleoprotein (NP). The CGE method allowed identification of different virus strains based on their specific protein profile. B/Brisbane inactivated virus and virosome samples could be analyzed within one day. The CGE results (titers) were comparable to single radial immune-diffusion (SRID), but the method has the advantage of a much faster time to results. CGE analysis of A/Christchurch from upstream process demonstrated the applicability of the method to samples of high complexity. The CGE method could be used in the same analyte concentration range as the RP-HPLC method, but showed better precision and accuracy. Overall, the total analysis time for the CGE method was much shorter, allowing analysis of 100 samples in 4 days instead of 10 days for SRID., (Copyright © 2015 Elsevier B.V. All rights reserved.)

In order to assess the utility of a recently developed capillary electrophoresis-mass spectrometry (CE-MS) method for the study of anionic metabolites in urine, a comparison was made with hydrophilic interaction chromatography-MS (HILIC-MS) using negative electrospray ionization. After optimization of the HILIC conditions, a gradient employing 10mM ammonium acetate (pH 6.8) in acetonitrile-water (5 min 90% acetonitrile followed by 90%-50% acetonitrile in 10 min) was selected, providing baseline separation of five representative anionic test metabolites. Relative standard deviations (RSDs) for HILIC retention times and peak areas were below 0.2% and 7.7%, respectively, and detection limits were in the range 0.04-2.21 μM. Metabolites in rat urine could also be analysed in a reproducible way with retention time and peak area RSDs below 0.6% and 13.6%, respectively. The CE-MS and HILIC-MS methods were compared in terms of reproducibility, sensitivity, selectivity and coverage of the anionic urinary metabolome. In general, peak area RSDs were similar whereas HILIC-MS yielded better retention-time repeatability and up to 80 times lower detection limits (expressed in injected concentration) for test metabolites as compared to CE-MS. Rat urine analysis by HILIC-MS provided detection of 1360 molecular features compared to 347 molecular features revealed with CE-MS. Of these, a number of 144 molecular features were found with both HILIC-MS and CE-MS, which showed on average 10 times higher peak areas in HILIC-MS. The HILIC retention and CE migration times of the common features were clearly not correlated. The HILIC and CE behavior of the test metabolites and 16 putatively identified common features were evaluated involving their physicochemical properties, indicating a markedly different separation selectivity, and thus significant degree of orthogonality of HILIC and CE., (Copyright © 2014 Elsevier B.V. All rights reserved.)

In this study, a CE-MS method using a monolithic sol-gel concentrator for in-line solid-phase extraction (SPE) is evaluated for the analysis of methionine enkephalin in biological samples. Operational SPE parameters such as sample pH, loading volume, elution volume and composition have been studied. After optimization of the in-line preconcentration methodology, a 40-fold preconcentration was demonstrated for a methionine enkephalin test solution using a loading volume of 3200 nL. The method was linear in the range from 62.5 to 1000 ng/mL (R(2)>0.99). R.S.D. values for migration times and peak areas were 1.2% and 8.4%, respectively. Finally, the analysis of cerebrospinal fluid samples spiked with methionine enkephalin and deproteinized with perchloric acid (1:1, v/v) showed a detection limit (S/N=3) of approximately 1 ng/mL (ca. 5 nM). The recoveries of methionine enkephalin for three concentration levels (100, 10 and 1 ng/mL) were in the range of 74-91%, demonstrating the promising potential of the methodology for the analysis of biological samples.