Polishing applications in antibody purification are dominated by traditional anion exchange chromatography (AEX), which typically
is based on strong quaternary amine (Q) ligands. Because AEX is performed in flow-through mode, membrane chromatography is
steadily encroaching on the ground formerly occupied by traditional AEX columns, a change that has significantly increased
the overall speed and productivity of manufacturing. However, Q membranes have not addressed any intrinsic limitations of
the ligand chemistry. The binding capacity of Q ligands is reduced at high conductivities, so concentrated feed streams must
be diluted to remove contaminants efficiently. To address this limitation, we have developed a novel membrane concept based
on weak anion exchange chemistry that features a high charge density covalently linked to a second-generation macroporous
membrane. We have named the new method salt tolerant interaction chromatography (STIC), and it is based on pure ion exchange chromatography principles. A model virus (ΦX174) was used to represent weak
acidic contaminants, and showed that stable removal (LRV >5) was possible in the presence of 150-mM NaCl (16.8 mS/cm). This
result was confirmed in a virus clearance study where the quantitative removal of minute virus of mice (MVM) was demonstrated
under the same conditions (LRV >4.96). This new method could be incorporated into existing processes with no requirement for
eluate dilution from the capture column.
Anion exchange chromatography (AEX) is an established technology platform for removing process contaminants such as host cell
proteins (HCP), DNA, and adventitious and endogenous viruses. As part of the purification train for antibody-like molecules,
AEX is usually carried out in flow-through mode, which has facilitated the implementation of convective media such as membrane
chromatography (MC). Because flow-through applications are driven by volume rather than mass, MC modules leave a small footprint
and allow for the implementation of disposables. MC, therefore, is leading the list of innovative technologies in industry
surveys.1 As titers increase, the feed streams after initial recovery (capturing) become more concentrated, yet the final bulk product
is still subject to the same high purity and safety requirements. It is necessary, therefore, to ensure that the polishing
steps can accommodate the requirements of high-titer processes, including the removal of residual contaminants and compliance
with orthogonal virus safety concepts.
Sartorius Stedim Biotech GmbH
Modern processes for antibody purification include only one or two polishing steps after initial capture, a concept that allows
ton amounts of proteins to be processed and does not appear amenable to further streamlining. Although the process itself
has been trimmed back as much as possible, peripheral operations such as buffer preparation and hold still offer opportunities
for cost savings, particularly when retrofitting existing facilities for the purification of high titer batches. One limitation
of the Q chemistry that is prevalent in polishing operations is its sensitivity to high salt conditions. Although virus clearance
works well around neutral pH and at conductivities up to and even beyond 10 mS/cm, the binding capacity of Q adsorbents drops
progressively as conductivity increases, leading to inefficient binding of problematic viruses such as minute virus of mice
(MVM) and the early breakthrough of HCP. Feed streams eluting from the CEX step therefore must be diluted to reduce conductivity,
increasing processing times, buffer requirements, and the amount of stress to which proteins are exposed. AEX with Q ligands
is a robust platform that has been validated in hundreds of processes, whereas alternatives such as mixed-mode chromatography
lack a clear and robust mechanism that would allow straightforward implementation during process development.
We sought to address the challenge of polishing under high-salt conditions by developing a salt-resistant AEX chemistry that
meets all existing capabilities of Q ligands and can be adapted to a membrane chromatography format for polishing applications.
We chose target ionic strength of approximately 16.8 mS/cm to increase the design space in process development.