The COVID-19 pandemic stressed the need for accelerating the development of novel vaccines. Over the past decades, the bottleneck in the biopharmaceutical process development shifted from optimizing fermentation processes to developing suitable purification strategies. Thereby, improving process understanding can significantly accelerate the development of purification processes. Unlike other biopharmaceutical products, vaccines are often more complex products containing molecules from different origins. The manufacturing process is therefore also more demanding. Consequently, no platform process is available for protein subunit vaccine purification. In the case of expressing a novel antigen in a host cell system, knowledge of the possible impurities – in this case host cell proteins (HCPs) – allows for rational and systematic process development. Therefore, this thesis focuses on developing characterization strategies of recurrent HCP impurities (from E. coli host cells) and on the integration of this information into modeling tools that advance removal strategies of these proteins, with a focus on protein-based antigen vaccines.

Firstly, the complete host cell proteome from antigen expressing E. coli host cells (BLR(DE3) and HMS174(DE3)) were characterized in chapter 2. Around 2000 HCPs were identified from the E. coli harvest sample using mass spectrometry based proteomics. Furthermore, an extensive HCP database including their expression levels, and physicochemical properties was constructed. Additionally, the profiles of an antigen expressing and null plasmid strain were compared. From a downstream processing perspective, the differences may be minor and the findings from the BLR(DE3) null strain can be applied to determine a purification strategy for the BLR(DE3) antigen-producing strain and HMS174(DE3) strain. The dataset of identified proteins was connected to databases describing the physicochemical properties of HCPs. Finally, protein property maps that help to identify a suitable downstream processing (DSP) strategy in comparison with the physicochemical properties of the target antigen, were generated.

Preparative chromatography based on differences in physicochemical properties is one of the main techniques for purification of vaccines. As follow-up to chapter 2, an experimental retention map of the host cell proteome during a salt gradient on hydrophobic interaction chromatography (HIC) and ion exchange chromatography (IEX) was constructed and reported and described in chapter 3. Furthermore, this study identified patterns in the retention behavior of HCPs based on their protein-protein interactions, molecular function, and cell location. To be able to predict the retention behavior of yet uncharacterized proteins, a quantitative structure-property relationship (QSPR) model was constructed using IEX retention data. Subsets of proteins, identified according to retention patterns, were used to build additional QSPR models, with monomer subsets yielding the most accurate predictions.

To achieve a higher level of process understanding, mechanistic models (MM) of chromatography columns are used in process development. These models primarily describe behavior of the target protein and selected process- or product-related impurities. However, it is beneficial to also include recurring HCP impurities in MMs. Hereby, critical HCPs causing issues when remaining in the product, are not necessarily abundant in the cell lysate and are often not individually described. A method for determining binding parameters of the entire host cell proteome including low abundant proteins to selected chromatography resins is still lacking.

Chapter 4 introduces a method to determine the above mentioned isotherm parameters of individual HCPs in a comprehensive manner. Fractions obtained from linear gradient elution experiments with different gradient lengths are analyzed by shotgun proteomics in order to extract the retention times of the individual HCPs. From the extracted retention volumes per gradient, isotherm parameters for all individual HCPs detected in the harvest were regressed. This method was exemplified using the BLR E. coli harvest, validated, and subsequently employed to optimize a capture step in silico.

Finally, chapter 5 gives an overview of the additionally investigated high-throughput sample preparation and analysis methods. This involved packing filter plates with resin for batch adsorption, which was explored to determine isotherm parameters instead of low gradient elution (LGE) experiments. Additionally, ion exchange high-performance liquid chromatography (IEX-HPLC) was investigated as an analytical technique instead of mass spectrometry (MS).

In summary, this thesis presents a comprehensive, large-scale characterization of HCPs from widely employed E. coli host cell strains for the production of protein vaccines. Moreover, a validated approach to determine isotherm parameters of all detectable HCPs in the harvest sample is presented.
@en

Mechanistic models mostly focus on the target protein and some selected process- or product-related impurities. For a better process understanding, however, it is advantageous to describe also reoccurring host cell protein impurities. Within the purification of biopharmaceuticals, the binding of host cell proteins to a chromatographic resin is far from being described comprehensively. For a broader coverage of the binding characteristics, large-scale proteomic data and systems level knowledge on protein interactions are key. However, a method for determining binding parameters of the entire host cell proteome to selected chromatography resins is still lacking. In this work, we have developed a method to determine binding parameters of all detected individual host cell proteins in an Escherichia coli harvest sample from large-scale proteomics experiments. The developed method was demonstrated to model abundant and problematic proteins, which are crucial impurities to be removed. For these 15 proteins covering varying concentration ranges, the model predicts the independently measured retention time during the validation gradient well. Finally, we optimized the anion exchange chromatography capture step in silico using the determined isotherm parameters of the persistent host cell protein contaminants. From these results, strategies can be developed to separate abundant and problematic impurities from the target antigen.@en

Optimizing a biopharmaceutical chromatographic purification process is currently the greatest challenge during process development. A lack of process understanding calls for extensive experimental efforts in pursuit of an optimal process. In silico techniques, such as mechanistic or data driven modeling, enhance the understanding, allowing more cost-effective and time efficient process optimization. This work presents a modeling strategy integrating quantitative structure property relationship (QSPR) models and chromatographic mechanistic models (MM) to optimize a cation exchange (CEX) capture step, limiting experiments. In QSPR, structural characteristics obtained from the protein structure are used to describe physicochemical behavior. This QSPR information can be applied in MM to predict the chromatogram and optimize the entire process. To validate this approach, retention profiles of six proteins were determined experimentally from mixtures, at different pH (3.5, 4.3, 5.0, and 7.0). Four proteins at different pH's were used to train QSPR models predicting the retention volumes and characteristic charge, subsequently the equilibrium constant was determined. For an unseen protein knowing only the protein structure, the retention peak difference between the modeled and experimental peaks was 0.2% relative to the gradient length (60 column volume). Next, the CEX capture step was optimized, demonstrating a consistent result in both the experimental and QSPR-based methods. The impact of model parameter confidence on the final optimization revealed two viable process conditions, one of which is similar to the optimization achieved using experimentally obtained parameters. The multiscale modeling approach reduces the required experimental effort by identification of initial process conditions, which can be optimized.@en

Purification of recombinantly produced biopharmaceuticals involves removal of host cell material, such as host cell proteins (HCPs). For lysates of the common expression host Escherichia coli (E. coli) over 1500 unique proteins can be identified. Currently, understanding the behavior of individual HCPs for purification operations, such as preparative chromatography, is limited. Therefore, we aim to elucidate the elution behavior of individual HCPs from E. coli strain BLR(DE3) during chromatography. Understanding this complex mixture and knowing the chromatographic behavior of each individual HCP improves the ability for rational purification process design. Specifically, linear gradient experiments were performed using ion exchange (IEX) and hydrophobic interaction chromatography, coupled with mass spectrometry-based proteomics to map the retention of individual HCPs. We combined knowledge of protein location, function, and interaction available in literature to identify trends in elution behavior. Additionally, quantitative structure–property relationship models were trained relating the protein 3D structure to elution behavior during IEX. For the complete data set a model with a cross-validated R2 of 0.55 was constructed, that could be improved to a R2 of 0.70 by considering only monomeric proteins. Ultimately this study is a significant step toward greater process understanding.@en

Mass-spectrometry-based proteomics is increasingly employed to monitor purification processes or to detect critical host cell proteins in the final drug substance. This approach is inherently unbiased and can be used to identify individual host cell proteins without prior knowledge. In process development for the purification of new biopharmaceuticals, such as protein subunit vaccines, a broader knowledge of the host cell proteome could promote a more rational process design. Proteomics can establish qualitative and quantitative information on the complete host cell proteome before purification (i.e., protein abundances and physicochemical properties). Such information allows for a more rational design of the purification strategy and accelerates purification process development. In this study, we present an extensive proteomic characterisation of two E. coli host cell strains widely employed in academia and industry to produce therapeutic proteins, BLR and HMS174. The established database contains the observed abundance of each identified protein, information relating to their hydrophobicity, the isoelectric point, molecular weight, and toxicity. These physicochemical properties were plotted on proteome property maps to showcase the selection of suitable purification strategies. Furthermore, sequence alignment allowed integration of subunit information and occurrences of post-translational modifications from the well-studied E. coli K12 strain.

@en