Romas Skudas

Host cell protein quantification by fourier transform mid infrared spectroscopy (FT-MIR)†

Process development in up‐ and downstream processing requires enhanced, non‐time‐consuming, and non‐expensive monitoring techniques to track product purity, for example, the level of endotoxins, viral particles, and host cell proteins (HCPs). Currently, HCP amounts are measured by laborious and expensive HCP‐enzyme‐linked immunosorbent assay (ELISA) assays best suited for measuring HCP amounts in the low concentration regime. The measurement of higher HCP amounts using this method requires dilution steps, adding dilution errors to the measurement. In this work we evaluated the suitability of attenuated total reflection spectroscopy for HCP quantification in process development, using clarified cell culture fluid from monoclonal antibody producing Chinese hamster ovary‐cells after treatment with different polyelectrolytes for semi‐selective clarification. Forty undiluted samples were chosen for multivariate data analysis in the middle infrared range and predicted HCP‐values were in good agreement with results obtained by an ELISA‐assay, suggesting the suitability of this new method for HCP‐quantification. As this method is able to quantify HCP titers ranging from approximately at least 20,000–200,000 ng mL−1, it is suitable especially for monitoring of process development steps with higher HCP concentrations, omitting dilution errors associated with ELISA assays. Biotechnol. Bioeng. 2013; 110: 252–259.

Development of a Univariate Membrane-Based Mid-Infrared Method for Protein Quantitation and Total Lipid Content Analysis of Biological Samples

Biological samples present a range of complexities from homogeneous purified protein to multicomponent mixtures. Accurate qualification of such samples is paramount to downstream applications. We describe the development of an MIR spectroscopy-based analytical method offering simultaneous protein quantitation (0.25–5 mg/mL) and analysis of total lipid or detergent species, as well as the identification of other biomolecules present in biological samples. The method utilizes a hydrophilic PTFE membrane engineered for presentation of aqueous samples in a dried format compatible with fast infrared analysis. Unlike classical quantification techniques, the reported method is amino acid sequence independent and thus applicable to complex samples of unknown composition. By comparison to existing platforms, this MIR-based method enables direct quantification using minimal sample volume (2 µL); it is well-suited where repeat access and limited sample size are critical parameters. Further, accurate results can be derived without specialized training or knowledge of IR spectroscopy. Overall, the simplified application and analysis system provides a more cost-effective alternative to high-throughput IR systems for research laboratories with minimal throughput demands. In summary, the MIR-based system provides a viable alternative to current protein quantitation methods; it also uniquely offers simultaneous qualification of other components, notably lipids and detergents.

Mid-infrared spectroscopy-based antibody aggregate quantification in cell culture fluids

Therapeutic antibody purification involves several steps which potentially induce antibody aggregation. Currently, aggregate monitoring mainly employs chromatographic, SDS-PAGE and light scattering techniques. In this study, the feasibility of mid-infrared spectroscopy (MIR) for the quantification of soluble antibody aggregates was investigated. Several multivariate models were evaluated to quantify antibody aggregation in chromatography elution streams and in clarified CHO cell culture supernatants (a surrogate for bioreactor output). A general model was established that is applicable for aggregate quantification directly from different cell culture solutions. Real-process samples and process-sample mimics were used to verify the general aggregate quantification model using two different antibodies. Results showed good prediction ability down to 1% aggregate content. Together with recently published results using MIR for host cell protein and target protein quantification, the results presented here indicate that MIR could provide multi-parameter process information from a single, fast, cost-effective and straightforward measurement. In conclusion, our study demonstrates that MIR is suitable for aggregate quantification in therapeutic antibody purification processes.

Mid‐infrared spectroscopy‐based analysis of mammalian cell culture Parameters

Within the framework of process analytical technology, infrared spectroscopy (IR) has been used for characterization of biopharmaceutical production processes. Although noninvasive attenuated total reflection (ATR) spectroscopy can be regarded as gold standard within IR‐based process analytics, simpler and more cost‐effective mid‐infrared (MIR) instruments might improve acceptability of this technique for high‐level monitoring of small scale experiments as well as for academia where financial restraints impede the use of costly equipment. A simple and straightforward at‐line mid‐IR instrument was used to monitor cell viability parameters, activity of lactate dehydrogenase (LDH), amount of secreted antibody, and concentration of glutamate and lactate in a Chinese hamster ovary cell culture process, applying multivariate prediction models, including only 25–28 calibration samples per model. Glutamate amount could be predicted with high accuracy (R2 0.91 for independent test‐set) while antibody concentration achieved good prediction for concentrations >0.4 mg L−1. Prediction of LDH activity was accurate except for the low activity regime. The model for lactate monitoring was only moderately good and requires improvements. Relative cell viability between 20 and 95% could be predicted with low error (8.82%) in comparison to reference methods. An initial model for determining the number of nonviable cells displayed only acceptable accuracy and requires further improvement. In contrast, monitoring of viable cell number showed better accuracy than previously published ATR‐based results. These results prove the principal suitability of less sophisticated MIR instruments to monitor multiple parameters in biopharmaceutical production with relatively low investments and rather fast calibration procedures.

Matrix effects during monitoring of antibody and host cell proteins using attenuated total reflection spectroscopy

Production of recombinant proteins, e.g. antibodies, requires constant real‐time monitoring to optimize yield and quality attributes and to respond to changing production conditions, such as host cell protein (HCP) titers. To date, this monitoring of mammalian cell culture‐based processes is done using laborious and time consuming enzyme‐linked immunosorbent assays (ELISA), two‐dimensional sodium dodecylsulphate polyacrylamide gel electrophoresis, and chromatography‐based systems. Measurements are usually performed off‐line, requiring regular sample withdrawal associated with increased contamination risk. As information is obtained retrospectively, the reaction time to adapt to process changes is too long, leading to lower yield and higher costs. To address the resulting demand for continuous online‐monitoring systems, we present a feasibility study using attenuated total reflection spectroscopy (ATR) to monitor mAb and HCP levels of NS0 cell culture in situ, taking matrix effects into account. Fifty‐six NS0 cell culture samples were treated with polyelectrolytes for semi‐selective protein precipitation. Additionally, part of the samples was subjected to filtration prior to analysis, to change the background matrix and evaluate effects on chemometric quantification models. General models to quantify HCP and mAb in both filtered and unfiltered matrix showed lower prediction accuracy compared to models designed for a specific matrix. HCP quantification in the range of 2,000–55,000 ng mL−1 using specific models was accurate for most samples, with results within the accepted limit of an ELISA assay. In contrast, mAb prediction was less accurate, predicting mAb in the range of 0.2–1.7 g L−1. As some samples deviated substantially from reference values, further investigations elucidating the suitability of ATR for monitoring are required.

Required polymer lengths per precipitated protein molecule in protein-polymer interaction

Protein precipitation using non-charged and charged polymers is a common method for protein purification, gaining broader interest among manufacturers in downstream processing. While during polymer- surface interactions, the formation of loops, tails and trains has been known for quite a long time, details of polymer conformation and chain length, interacting with the protein during protein precipitation are not fully discovered. Our research presents a more profound understanding of polymer-protein interaction, combining fluorescence and infrared spectroscopic measurements of proteins and polymer standards with well defined chain length to confirm different models of protein-polymer interaction. Lysozyme, chymotrypsinogen A, myoglobin and a monoclonal antibody, all of different molecular weight, isoelectric point and charge distribution at the protein surface, were used for protein-polymer precipitation. The use of polymers of various charge density and chain length showed that the required polymer length per precipitated protein (Ldef) is up to 25-times larger than the diameter of the corresponding protein, depending on the surface charge distribution of the protein, and its isoelectric point, as well as the charge density of the polymer. Our results support proposed mechanisms of polymer wrapping and loop formation for optimal charge neutralization during complexation and imply interaction of several polymer chains per precipitated protein molecule. Electrophoretic light scattering showed a qualitative correlation of the zeta potential of analyzed polymers with their corresponding Ldef values. Comparing protein precipitation behavior of long and short polymer chains, the latter exhibited reduced precipitation efficiency, visible as elevated Ldef.

Customization of copolymers to optimize selectivity and yield in polymer-driven antibody purification processes.

This manuscript describes customization of copolymers to be used for polymer‐driven protein purification in bioprocessing. To understand how copolymer customization can be used for fine‐tuning, precipitation behavior was analyzed for five target antibodies (mAbs) and BSA as model impurity protein, at ionic strength similar to undiluted cell culture fluid. In contrast to the use of standardized homopolymers, customized copolymers, composed of 2‐acrylamido‐2‐methylpropane sulfonic acid (AMPS) and 4‐(acryloylamino)benzoic acid (ABZ), exhibited antibody precipitation yields exceeding 90%. Additionally, copolymer average molecular weight (Mw) was varied and its influence on precipitation yield and contaminant coprecipitation was investigated. Results revealed copolymer composition as the major driving force for precipitation selectivity, which was also dependent on protein hydrophobicity. By adjusting ABZ content and Mw of the precipitant for each of the mAbs, conditions were found that allowed for high precipitation yield and selectivity. These findings may open up new avenues for using polymers in antibody purification processes.

Feasibility of polyelectrolyte‐driven Fab fragment separation

The use of antigen-binding fragments (Fabs) as biotherapeutic agents is gaining interest and thus requires development of adequate purification strategies aimed at separating Fabs from other proteins. Thus, the feasibility of using a copolymer for separation of Fabs from monoclonal antibodies (mAbs) and fragment constant regions (Fcs) was evaluated, employing a blend of purified solutions of these proteins. The use of a copolymer exerting both hydrophobic as well as anionic properties resulted in high precipitation yields for both the mAb and Fc fragment, even at ionic strength of 150 mM NaCl. On the contrary, Fabs exhibited reduced precipitation yields upon copolymer addition. These observations are attributed to differences in protein physicochemical parameters, allowing mAbs and Fcs to be precipitated via conjoint electrostatic and hydrophobic interactions. In contrast, Fabs were mainly precipitated via electrostatic interactions, being reduced at higher ionic strength. This finding was corroborated by hydrophobicity analysis using 2-p-toluidinonaphthalene-6-sulfonate, showing enhanced hydrophobicity of Fcs compared to mAbs alone, while Fabs exhibited the lowest hydrophobicity. Within the context of increasing demand for Fabs as therapeutic proteins, these results may open up a simpler purification strategy for this protein class, potentially also to be implemented within the context of polymer-driven protein purification during fermentation.