The use of predictive models to develop chromatography-based purification processes.

Chromatography is the workhorse of biopharmaceutical downstream processing because it can selectively enrich a target product while removing impurities from complex feed streams. This is achieved by exploiting differences in molecular properties, such as size, charge and hydrophobicity (alone or in different combinations). Accordingly, many parameters must be tested during process development in order to maximize product purity and recovery, including resin and ligand types, conductivity, pH, gradient profiles, and the sequence of separation operations. The number of possible experimental conditions quickly becomes unmanageable. Although the range of suitable conditions can be narrowed based on experience, the time and cost of the work remain high even when using high-throughput laboratory automation. In contrast, chromatography modeling using inexpensive, parallelized computer hardware can provide expert knowledge, predicting conditions that achieve high purity and efficient recovery. The prediction of suitable conditions in silico reduces the number of empirical tests required and provides in-depth process understanding, which is recommended by regulatory authorities. In this article, we discuss the benefits and specific challenges of chromatography modeling. We describe the experimental characterization of chromatography devices and settings prior to modeling, such as the determination of column porosity. We also consider the challenges that must be overcome when models are set up and calibrated, including the cross-validation and verification of data-driven and hybrid (combined data-driven and mechanistic) models. This review will therefore support researchers intending to establish a chromatography modeling workflow in their laboratory.

Spherical nanoparticles can be used as non-penetrating tracers to determine the extra-particle void volume in packed-bed chromatography columns.

The optimization of downstream processing in silico can accelerate bioprocess development by limiting experiments to the most promising separation conditions that have been identified using chromatography models. Such models describe protein binding and mass transport in packed-bed columns and thus require precise knowledge about the columns and the resins they contain. One important set of properties is the resin porosities, often determined using combinations of penetrating and non-penetrating tracers. However, the former can be disproportionately small, providing data of limited practical relevance, and the latter can undergo unwanted interactions with the resin, interfering with porosity calculations. Here we characterize and minimize the interactions of three novel hard-sphere non-penetrating tracers with the model resin Q Sepharose HP under various conditions and determine the corresponding inter-particle porosities. We found that conductivities > 100 mS cm-1 were necessary to suppress tracer-resin interactions despite them sharing the same surface charge. We combined these data with those from proteins studied under non-binding conditions, which can be used as authentic penetrating tracers, to determine both the intra-particle and total porosities. Furthermore, we found that the inter-particle porosity was below the theoretical limit of dense sphere packing (25.95%) and provide experimental data showing that the discrepancy is caused by resin particle deformation during the packing of columns under pressure.

Precision analysis for the determination of steric mass action parameters using eight tobacco host cell proteins.

Plants are advantageous as biopharmaceutical manufacturing platforms because they allow the economical and scalable upstream production of proteins, including those requiring post-translational modifications, but do not support the replication of human viruses. However, downstream processing can be more labor-intensive compared to fermenter-based systems because the product is often mixed with abundant host cell proteins (HCPs). Modeling chromatographic separation can minimize the number of process development experiments and thus reduce costs. An important part of such modeling is the sorption isotherm, such as the steric mass action (SMA) model, which describes the multicomponent protein-salt equilibria established in ion-exchange systems. Here we purified ten HCPs, including 2-Cys-peroxiredoxin, from tobacco (Nicotiana tabacum and N. benthamiana). For eight of these HCPs, we obtained sufficient quantities to determine the SMA binding parameters (KSMA and ν) under different production-relevant conditions. We studied the parameters for 2-Cys-peroxiredoxin on Q-Sepharose HP in detail, revealing that pH, resin batch and buffer batch had little influence on KSMA and ν, with coefficients of variation (COVs) less than 0.05 and 0.21, respectively. In contrast, the anion-exchange resins SuperQ-650S, Q-Sepharose FF and QAE-550C led to COVs of 0.69 for KSMA and 0.05 for ν, despite using the same quaternary amine functional group as Q-Sepharose HP. Plant cultivation in summer vs winter resulted in COVs of 0.09 for KSMA and 0.02 for ν, revealing a small impact compared to COVs of 17.15 for KSMA and 0.20 for ν when plants were grown in different settings (climate-controlled phytotron vs greenhouse). We conclude that plant cultivation can substantially affect protein properties and the resulting SMA parameters. Accordingly, plant growth but also protein purification and characterization for chromatography model building should be tightly controlled and well documented.