Tuning selectivity in cation-exchange chromatography applied for monoclonal antibody separations, part 2: Evaluation of recent stationary phases

Leveraging bioanalytical characterization of fractionated monoclonal antibody pools to identify aggregation-prone and less filterable proteoforms during virus filtration

Monoclonal antibodies (mAbs) are an essential class of biotherapeutics. A platform process is used for mAb development to ensure clinically safe and stable molecules. Regulatory authorities ensure that mAb production processes include sufficient viral clearance steps to achieve less than one virus particle per million doses of product. Virus filtration is used for size-based removal of enveloped and nonenveloped viruses during downstream processing of mAbs. Process development in mAb purification relies on empirical approaches and often includes adsorptive prefiltration to mitigate virus filter fouling. Opportunities for molecular-level prediction of mAb filterability are needed to plug the existing knowledge gap in downstream processing. A molecular-level approach to understanding the factors influencing mAb filterability may reduce process development time, material loss, and processing costs due to oversized virus filters. In this work, pH step gradient fractionation was applied on polished bulk mAb feed to obtain concentrated pools of fractionated mAb variants. Biophysical properties and quality attributes of fractionated pools, including oligomeric state (size), isoelectric point profile, diffusion interaction parameters, and glycoform profile, were determined using bioanalytical methods. Filterability (loading and throughput) of fractionated pools were evaluated. Statistical methods were used to obtain correlations between quality attributes of mAb fractions and filterability on the Viresolve Pro virus filter.

Characterization of Complex Proteoform Mixtures by Online Nanoflow Ion-Exchange Chromatography-Native Mass Spectrometry

The characterization of proteins and complexes in biological systems is essential to establish their critical properties and to understand their unique functions in a plethora of bioprocesses. However, it is highly difficult to analyze low levels of intact proteins in their native states (especially those exceeding 30 kDa) with liquid chromatography (LC)-mass spectrometry (MS). Herein, we describe for the first time the use of nanoflow ion-exchange chromatography directly coupled with native MS to resolve mixtures of intact proteins. Reference proteins and protein complexes with molecular weights between 10 and 150 kDa and a model cell lysate were separated using a salt-mediated pH gradient method with volatile additives. The method allowed for low detection limits (0.22 pmol of monoclonal antibodies), while proteins presented nondenatured MS (low number of charges and limited charge state distributions), and the oligomeric state of the complexes analyzed was mostly kept. Excellent chromatographic separations including the resolution of different proteoforms of large proteins (>140 kDa) and a peak capacity of 82 in a 30 min gradient were obtained. The proposed setup and workflows show great potential for analyzing diverse proteoforms in native top-down proteomics, opening unprecedented opportunities for clinical studies and other sample-limited applications.

In-line buffer exchange in the coupling of Protein A chromatography with weak cation exchange chromatography for the determination of charge variants of immunoglobulin G derived from chinese hamster ovary cell cultures

Immunoglobulin G (IgG) is the most common monoclonal antibody (mAb) grown for therapeutic applications. While IgG is often selectively isolated from cell lines using protein A (ProA) chromatography, this is only a stepping stone for complete characterization. Further classification can be obtained from weak cation exchange chromatography (WCX) to determine IgG charge variant distributions. The charge variants of monoclonal antibodies can influence the stability and efficacy in vivo, and deviations in charge heterogeneity are often cell-specific and sensitive to upstream process variability. Current methods to characterize IgG charge variants are often performed off-line, meaning that the IgG eluate from the ProA separation is collected, diluted to adjust the pH, and then transferred to the WCX separation, adding time, complexity, and potential contamination to the sample analysis process. More recently, reports have appeared to streamline this separation using in-line two-dimensional liquid chromatography (2D-LC). Presented here is a novel, 2D-LC coupling of ProA in the first dimension (1D) and WCX in the second dimension (2D) chromatography. As anticipated, the initial direct column coupling proved to be challenging due to the pH incompatibility between the mobile phases for the two stages. To solve the solvent compatibility issue, a size exclusion column was placed in the switching valve loop of the 2D-LC instrument to act as a means for the on-line solvent exchange. The efficacy of the methodology presented was confirmed through a charge variant determination using the NIST monoclonal antibody standard (NIST mAb), yielding correct acidic, main, and basic variant compositions. The methodology was employed to determine the charge variant profile of IgG from an in-house cultured Chinese hamster ovary (CHO) cell supernatant. It is believed that this methodology can be easily implemented to provide higher-throughput assessment of IgG charge variants for process monitoring and cell line development.

Revealing charge heterogeneity of stressed trastuzumab at the subunit level

Trastuzumab is known to be heterogeneous in terms of charge. Stressing trastuzumab under physiological conditions (pH 7.4 and 37 °C) increases charge heterogeneity further. Separation of charge variants of stressed trastuzumab at the intact protein level is challenging due to increasing complexity making it difficult to obtain pure charge variants for further characterization. Here we report an approach for revealing charge heterogeneity of stressed trastuzumab at the subunit level by pH gradient cation-exchange chromatography. Trastuzumab subunits were generated after limited proteolytic cleavage with papain, IdeS, and GingisKHAN®. The basic pI of Fab and F(ab)₂ fragments allowed to use the same pH gradient for intact protein and subunit level analysis. Baseline separation of Fab subunits was obtained after GingisKHAN® and papain digestion and the corresponding modifications were determined by LC-MS/MS peptide mapping and middle-down MALDI-ISD FT-ICR MS. The described approach allows a comprehensive charge variant analysis of therapeutic antibodies that have two or more modification sites in the Fab region.

Development of a sub-hour on-line comprehensive cation exchange chromatography x RPLC method hyphenated to HRMS for the characterization of lysine-linked antibody-drug conjugates

This study details the development of on-line two-dimensional liquid chromatography (2D-LC) methods combining cation-exchange chromatography (CEX) and reversed-phase liquid chromatography (RPLC) for the separation of the charge variants of a lysine-linked antibody-drug conjugate (ADC). This combination gives an excellent example of the potential benefits of 2D-LC approaches for the analysis of such complex protein formats. CEX is considered the reference technique for the separation of protein charge variants but its retention mechanism usually requires the use of a high concentration of non-volatile salts, which impedes its compatibility with MS detection. In this context, the use of an on-line 2D-LC-MS approach not only allows on-line desalting and indirect coupling of CEX with mass spectrometry (MS) detection but it also provides increased and complementary information within a single analysis. The first part of this study was devoted to the choice of stationary phases and the optimization of chromatographic conditions in both dimensions. Based on the results obtained in 1D-CEX with ultraviolet detection (UV) and 1D-RPLC with UV and high-resolution mass spectrometry (HRMS) detections, an on-line comprehensive two-dimensional liquid chromatography method combining CEX and RPLC was developed. The last part of this study was devoted to the identification of the separated species using HRMS detection and in the comparison of three ADC samples exposed to different durations of thermal stress.

Towards a simple on-line coupling of ion exchange chromatography and native mass spectrometry for the detailed characterization of monoclonal antibodies

This work describes the direct hyphenation of cation exchange chromatography (CEX) with a compact, easy-to-use benchtop Time of Flight mass spectrometer (ToF/MS) for the analytical characterization of monoclonal antibodies (mAbs). For this purpose, a wide range of commercial mAb products (including expired samples and mAb biosimilars) were selected to draw reliable conclusions. From a chromatographic point of view, various buffers and column dimensions were tested. When considering pH response, buffer stability over time and MS compatibility, the best compromise is represented by the following recipe: 50 mM ammonium acetate, titrated to pH 5.0 (mobile phase A) and 160 mM ammonium acetate, titrated to pH 8.5 (mobile phase B). Despite the broader peaks observed with the 2.1 mm i.d. CEX column, this was preferentially selected for CEX-MS operation, since the efficiency loss (caused by extra-column dispersion) was still acceptable while MS compatibility was strongly enhanced (thanks to low flow rate). In terms of MS, it was important to avoid the use of glass-bottled mobile phases, laboratory glassware and glass vials to minimize loss of MS resolution, sensitivity, and mass accuracy due to metal contaminants. With this new CEX-MS setup, straightforward and rapid analysis (in less than 10 min) of charge variants was possible, allowing the separation and identification of several charge variants.

Empirical correction of non-linear pH gradients and a tool for application to protein ion exchange chromatography

This concept article reports a practical solution to improve the linearity of effluent pH response as observed in pH gradient cation exchange chromatography (CEX). When performing pH gradient CEX, it is not easy to develop buffer systems that will universally provide pH response proportional with the mobile phase (buffer) composition. It is an especially challenging pursuit when exploring MS compatible buffers (e.g. ammonium-acetate, ammonium-carbonate). In addition to "non-proportional" behavior from the mobile phase composition, the chromatographic column itself will sometimes impose an unpredictable impact on the effluent pH. Here, we propose a simple approach based on the on-line measurement of effluent pH response, conversion of pH to mobile phase volume fraction (φ) and then generation of the inverse response function in the time domain. In the end, when setting the inverse function as the gradient program instead of a linear gradient, an improved - ideally linear - pH response can be produced. A simple Excel tool was developed to assist analysts with this correction procedure, and it has been made available by download for public use.

Impact of the column on effluent pH in cation exchange pH gradient chromatography, a practical study

In ionexchange chromatography, the pH gradient mode becomes more and more popular today for the analysis of therapeutic proteins as this mode can provide higher or alternative selectivity to the commonly used salt gradient mode. Ideally, a linear pH response is expected when performing linear gradients. However up to now, only a very few buffer systems have been developed and are commercially available which can perform nearly linear pH responses when flowing through a given column. It is also known that a selected buffer system (mobile phase) can work well on one column but can fail on other column. The goal of this study was to practically evaluate the effects that ionexchange columns (weak and strong exchangers) might have on effluent pH, when performing linear pH gradient separations of therapeutic monoclonal antibodies. To attain this objective, the pH was monitored on-line at the column outlet using a specific setup. To make comprehensive observations of the phenomenon, four different mobile phase conditions and five cation exchange columns (weak and strong exchangers) were employed. The obtained pH responses were systematically compared to responses measured in the absence of the columns. From this work, it has become clear that both the column and mobile phase can have significant effects on pH gradient chromatography and that their combination must be considered when developing a new method. Phase systems (column + mobile phase) providing linear pH responses are indeed the most suitable for separating mAbs with different isoelectric points and, with them, it is possible to elute mAbs across wide retention time ranges and with high selectivity.

Practical considerations on the particle size and permeability of ion-exchange columns applied to biopharmaceutical separations

The goal of this study was to better understand the possibilities and limitations of modern cation exchange chromatography (CEX) columns for the separation of protein biopharmaceuticals (typically mAbs and related products). Several commercial and research columns consisting of a non-porous polymeric core particle with a thin hydrophilic coating and grafted ion-exchanger sulfonate groups, were compared. The impact of particle size, porosity and packing pressure on the separation of therapeutic proteins was evaluated in a systematic way. First, it was shown that the porosity of modern CEX columns depends on the applied conditions, and lower apparent porosity as well as increased column pressures were observed when using low ionic strength mobile phase (less than 0.01 M NaCl), due to swelling. Column pressure seemed to be dependent on the 1/d_p³ to 1/d_p⁵ relationships with particle size, depending on whether 0.3 M NaCl or pure water was used as mobile phase, respectively. Using 5 cm long columns packed with 2 or 2.5 µm particles could easily result in higher than 1000 bar pressure drops when the mobile phase ionic strength is low. Therefore, it is recommended that particle size not be decreased to below 2.5 µm so that technologies can remain compatible with the current state of ultra-high pressure (UHPLC) instrumentation. This recommendation is underscored by the fact that a decrease in particle size does not produce improved separations, since the particles are non-porous (no intra-particle diffusion nor resistance to mass transfer) and that large solutes follow an on-off (bind and elute) type retention mechanism. The only advantage of CEX columns packed with small particles is that they can provide more specific surface area per unit length of column, and thus facilitate higher throughput methods. In conclusion, it appears that there is no need to further decrease the particle size in CEX since decreasing their particle size may result in more drawbacks than benefits.