May 15, 2016 (Vol. 36, No. 10)

Payal Khandelwal Ph.D. Global Product Manager, Protein Purification Business Bio-Rad Laboratories
Xuemei He Ph.D. R&D Manager, Chromatography Media Chemistry Bio-Rad Laboratories

Mixed-Mode Chromatography for Monoclonal Antibody Aggregate Removal

Advances in monoclonal antibody (mAb) technology have revolutionized the field of biological research. With applications in both research and therapeutics, mAb development is one of the fastest growing sectors in the pharmaceutical industry.

Purification of mAbs faces many hurdles, including process challenges related to aggregate formation and removal. Multiple factors can lead to aggregate formation during upstream cell culturing as well as downstream manufacturing. The presence of aggregates can hinder the efficacy of therapeutic mAbs due to their different bioactivity, storage stability, immunogenicity, and pharmacokinetic properties relative to the monomeric versions.

For these reasons, the removal of antibody aggregates has become a major focus of downstream processing. In many cases, antibody aggregates are the most challenging product-related impurities to remove. Extensive process development is necessary in order to obtain highly pure monomeric mAbs without compromising product yield.

Advantages of Mixed-Mode Chromatography Media

Mixed-mode (MM) chromatography has emerged as a powerful tool for overcoming many common mAb purification challenges. It serves as an attractive and effective alternate to unimodal chromatography by combining the properties of two or more types of resins—ion exchange (IEX), and/or affinity, and/or hydrophobic interaction chromatography (HIC)—presenting unique selectivity for the purification target.

MM media often demonstrate better performance over traditional resins in terms of product yields and activity as a result of their larger design space for optimal target protein binding and elution. It is beneficial to screen multiple resins to obtain an optimized protocol for process purification of specific mAbs, as these biomolecules exhibit diversified chemical and physical properties such as in vivo and in vitro stabilities against aggregation and precipitation.

Product Comparison and Optimization Studies

The monoclonal antibody of interest in this present study, mAb S, has an isoelectric point (pI) of 6.9. It was captured from an overproducing Chinese hamster ovary (CHO) cell culture by Protein A affinity chromatography.

A typical eluate from Protein A chromatography contains ~25% high molecular weight aggregates of mAb S. Previous attempts to remove these aggregates using IEX chromatography proved futile.

Three MM chromatography media, CHT™ Ceramic Hydroxyapatite Type I, 40 µM (Bio-Rad Laboratories), Product A, and Product B, were therefore screened for their effectiveness in mAb S aggregate clearance. CHT is a naturally occurring inorganic mixed-mode media with a chemical composition of (Ca5(PO4)3OH)2. It binds biomolecules via cation exchange with the phosphoryl groups (P sites), hydrogen bonding with the hydroxyl groups, and/or chelation with the calcium ions (C sites) in its crystal lattice (Figure 1A).


Figure 1. A. Mechanism of action of CHT Ceramic Hydroxyapatite media

For Product A, the N-benzyl-N-methyl ethanol amine ligand provides hydrophobic interactions with aromatic amino residues in proteins via its benzyl side chain, ionic interactions with the charged amino acids through its quaternary amine, and potential hydrogen bonding with its terminal hydroxyl group by design (Figure 1B).

In general, antibody aggregates tend to bind to CHT more strongly than their monomeric counterparts. A sodium chloride gradient from 1 to 1,000 mM at pH 7.0 was carried out to resolve mAb S monomers from the aggregates. The mAb S monomers were efficiently recovered at high yield (>75%) by elution at a modest concentration of sodium chloride without contaminating aggregates (Figure 2, green line).

Further process development was performed to convert this gradient elution into a step elution using 550 mM sodium chloride at pH 7.0 with a final monomer content of 99.5% and recovery of 82.7% in just five column volumes (CV) of eluate (Table 1).


Figure 1. B. Structure of ligand for Product A and B

For the Product A media, design of experiment (DoE) studies were initially employed to screen for mAb S aggregate-removal conditions. Binding of mAb S by these resins required a low sodium chloride concentration (≤50 mM) at pH>8, which indicates the electrostatic interaction between the negatively charged mAb S molecules and the Product A ligands at alkaline pH. The bound antibodies could be eluted by acidic buffers.

However, increasing the sodium chloride concentration in the buffer simultaneously with pH reduction hampered the recovery of this antibody, which is evidence of hydrophobic interactions between the Product A ligands and the bound antibody molecules at lower pH. Therefore, a pH gradient from 8 to 5 was used to investigate the correlation between mAb S monomer-aggregate resolution, monomer recovery, and elution buffer pH.

While lower buffer pH enhanced the recovery of mAb S monomers from Product A, aggregate contamination became worse when the pH of the elution buffer approached pH 5 (Figure 2, light blue line). Similarly, mAb S monomers were efficiently bound by Product B at pH>8, and the antibody monomers were readily eluted from the column in a pH 8 to 5 gradient. Again, significantly more mAb S aggregates were detected in the later elution fractions (Figure 2, dark blue line).

When converting gradient elution to step elution with the Product A and B media, special precautions were taken to minimize aggregate contamination. In order to obtain mAb S monomer content of ≥99.5%, the elution had to be precisely controlled at pH 5.5 and 5.4 with final recovery of 49% and 62% for Product A and Product B, respectively (Table 1).

The relatively large eluate volume from the Product A and B columns, as compared with the eluate from CHT, was probably due to the strong hydrophobic interaction between these mAb S molecules and the hydrophobic anion exchangers at pH below the pI of mAb S (pI = 6.9).


Figure 2. CHT Media provides the best monomer recovery for mAb S under the tested conditions. A bind-and-elute strategy was employed for the three media. Sodium chloride and pH gradients were performed based on DoE studies for optimized separation of aggregate and monomer. The highest total recovery, at low aggregate content (=0.5%), was achieved with CHT Media: CHT Column, 1 mL, 0–1,000 mM sodium chloride gradient (■); Product A Column, 1 mL, pH 8–5 gradient (▲); Product B, 1 mL, pH 8–5 gradient (♦).

Conclusion

Both target purity and yield are essential to product quality and process efficiency/economics in the commercial production of monoclonal antibodies. Mixed-mode chromatography has been widely used for the clearance of mAb aggregates.

The present study underscores the importance of thorough screening of multiple media and an in-depth understanding of the interactions between target molecules and chromatography resins during purification method development. For the hydrophobic and mildly acidic mAb S molecules, CHT offers a robust production process for highly purified monomers at maximum yield.

Payal Khandelwal, Ph.D. ([email protected]), is global product manager, protein purification business and Xuemei He, Ph.D. ([email protected]), is R&D manager, chromatography media chemistry at Bio-Rad Laboratories.

Previous articleHistone Mutation Deranges DNA Methylation to Cause Cancer
Next articleTop 10 European Biopharma Clusters