Oligonucleotide Purification by Ion-Exchange Chromatography: Predicting the Impact of Salt Type and Buffer pH with Mechanistic Modeling

Published in BioProcess International in collaboration with Roche

David Pfister, Lucrèce Nicoud, Ludivine Larue, Philipp Weber, Martin Olbrich

• by Ypso-Facto
• May 21, 2025

 

Check out Ypso-Facto's new publication on oligonuclotide purification by Ion Exchange Chromatography in collaboration with Roche.

 

ABSTRACT

Ion-exchange chromatography (IEC) is a common method for oligonucleotide purification (1, 2). Separation by IEC often uses strong–anion-exchange (AEX) resins at moderate to strongly basic pH values. A typical process involves loading of crude solution (potentially after composition adjustment), one or several wash steps, elution with a salt gradient and fractionation of material to collect purified product, and finally regeneration with a high–salt-concentration solution. Developing an IEC process requires determination of many operating parameters — e.g., quantity of crude solution to be injected, wash-step duration, elution-gradient slope and/or final salt concentration, buffer and salt types, buffer-solution pH and conductivity, flow rates, bed height, and (if applicable) organic-solvent content in process solutions. A common objective for IEC processes is to maximize yield while satisfying requirements in terms of global purity and concentrations of individual impurities. Minimizing raw-material consumption and waste generation is becoming essential, too, as the pressure to devise ecologically sustainable processes increases (3).

Predictive mechanistic simulation is a valuable tool for speeding up process development and characterization. It helps to reduce the number of required experiments while substantially increasing process understanding. But the power of mechanistic simulation can be unleashed only when using a reliable simulator and an efficient approach to determining model parameters.

Attempts have been made to derive mechanistic models describing biomolecule purification by IEC (4–9). However, previous models often do not account for fundamental differences between adsorption and ion exchange and therefore are limited in their predictive capabilities. Ion exchange, as the term suggests, is based on an exchange mechanism. In other words, “if one charge goes in, one charge goes out” (10, 11). That factor can have a dramatic impact on solution chemistry because pH is likely to change when ions are released from a resin into the liquid phase, in turn affecting the charge of the molecule to be purified and influencing its retention on the solid phase. Moreover, peculiar phenomena sometimes are observed with IEC processes, including pH overshoots, retention beyond the isoelectric point (12), and drastic shifts in elution behavior with changes in buffer type (13) or raw material (e.g., from different suppliers) (14). Such interesting phenomena can be explained only by using a sound ion-exchange model.

Here, we combine a detailed IEC mechanistic model (15) with novel experimental data generated at Roche. Our case study deals with purification of an N-acetylgalactosamine (GalNAc)-cluster–conjugated 20-mer oligonucleotide mixture produced by solid-phase synthesis. Purification was carried out on a strong anion exchanger using a phosphate buffer and a linear salt gradient. We focus on the impact of solution pH (8.0 to 12.0) and elution salt type (sodium chloride (NaCl) and sodium bromide (NaBr)).

Click here to read the full article in Bioprocess International

AUTHORS INFORMATION

David Pfister is head of research and development, Ludivine Larue is a project manager, and corresponding author Lucrèce Nicoud is cochief executive officer, all at Ypso-Facto, 19 avenue Foch 54000 Nancy, France. 

Philipp Weber is a junior scientist in the synthetic molecules department, and Martin Olbrich is section head of process chemistry, both at F. Hoffmann-La Roche, Grenzacherstrasse 124, 4070 Basel, Switzerland.

---