Search

The conjugation reaction is the central step in the manufacturing process of antibody-drug conjugates (ADCs). This reaction generates a heterogeneous and complex mixture of differently conjugated sub-species depending on the chosen conjugation chemistry. The parametrization of the conjugation reaction through mechanistic kinetic models offers a chance to enhance valuable reaction knowledge and ensure process robustness. This study introduces a versatile modeling framework for the conjugation reaction of cysteine-conjugated ADC modalities-site-specific and interchain disulfide conjugation. Various conjugation kinetics involving different maleimide-functionalized payloads were performed, while controlled gradual payload feeding was employed to decelerate the conjugation, facilitating a more detailed investigation of the reaction mechanism. The kinetic data were analyzed with a reducing reversed phase (RP) chromatography method, that can readily be implemented for the accurate characterization of ADCs with diverse drug-to-antibody ratios, providing the conjugation trajectories of the single chains of the monoclonal antibody (mAb). Possible kinetic models for the conjugation mechanism were then developed and selected based on multiple criteria. When calibrating the established model to kinetics involving different payloads, conjugation rates were determined to be payload-specific. Further conclusions regarding the kinetic comparability across the two modalities could also be derived. One calibrated model was used for an exemplary in silico screening of the initial concentrations offering valuable insights for profound understanding of the conjugation process in ADC development., Competing Interests: Authors MW, RB and KH are employed by AstraZeneca and report stock ownership and/or stock options or interest in AstraZeneca. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Weggen, Bean, Hui, Wendeler and Hubbuch.)

Virus-like particles (VLPs) are a promising class of biopharmaceuticals for vaccines and targeted delivery. Starting from clarified lysate, VLPs are typically captured by selective precipitation. While VLP precipitation is induced by step-wise or continuous precipitant addition, current monitoring approaches do not support the direct product quantification, and analytical methods usually require various, time-consuming processing and sample preparation steps. Here, the application of Raman spectroscopy combined with chemometric methods may allow the simultaneous quantification of the precipitated VLPs and precipitant owing to its demonstrated advantages in analyzing crude, complex mixtures. In this study, we present a Raman spectroscopy-based Process Analytical Technology (PAT) tool developed on batch and fed-batch precipitation experiments of Hepatitis B core Antigen VLPs. We conducted small-scale precipitation experiments providing a diversified data set with varying precipitation dynamics and backgrounds induced by initial dilution or spiking of clarified Escherichia coli -derived lysates. For the Raman spectroscopy data, various preprocessing operations were systematically combined allowing the identification of a preprocessing pipeline, which proved to effectively eliminate initial lysate composition variations as well as most interferences attributed to precipitates and the precipitant present in solution. The calibrated partial least squares models seamlessly predicted the precipitant concentration with R 2 of 0.98 and 0.97 in batch and fed-batch experiments, respectively, and captured the observed precipitation trends with R 2 of 0.74 and 0.64. Although the resolution of fine differences between experiments was limited due to the observed non-linear relationship between spectral data and the VLP concentration, this study provides a foundation for employing Raman spectroscopy as a PAT sensor for monitoring VLP precipitation processes with the potential to extend its applicability to other phase-behavior dependent processes or molecules., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Dietrich, Schiemer, Kurmann, Zhang and Hubbuch.)

Protein crystallization as opposed to well-established chromatography processes has the benefits to reduce production costs while reaching a comparable high purity. However, monitoring crystallization processes remains a challenge as the produced crystals may interfere with analytical measurements. Especially for capturing proteins from complex feedstock containing various impurities, establishing reliable process analytical technology (PAT) to monitor protein crystallization processes can be complicated. In heterogeneous mixtures, important product characteristics can be found by multivariate analysis and chemometrics, thus contributing to the development of a thorough process understanding. In this project, an analytical set-up is established combining offline analytics, on-line ultraviolet visible light (UV/Vis) spectroscopy, and in-line Raman spectroscopy to monitor a stirred-batch crystallization process with multiple phases and species being present. As an example process, the enzyme Lactobacillus kefir alcohol dehydrogenase (L k ADH) was crystallized from clarified Escherichia coli ( E. coli ) lysate on a 300 mL scale in five distinct experiments, with the experimental conditions changing in terms of the initial lysate solution preparation method and precipitant concentration. Since UV/Vis spectroscopy is sensitive to particles, a cross-flow filtration (cross-flow filtration)-based bypass enabled the on-line analysis of the liquid phase providing information on the lysate composition regarding the nucleic acid to protein ratio. A principal component analysis (PCA) of in situ Raman spectra supported the identification of spectra and wavenumber ranges associated with productspecific information and revealed that the experiments followed a comparable, spectral trend when crystals were present. Based on preprocessed Raman spectra, a partial least squares (PLS) regression model was optimized to monitor the target molecule concentration in real-time. The off-line sample analysis provided information on the crystal number and crystal geometry by automated image analysis as well as the concentration of Lk ADH and host cell proteins (HCPs) In spite of a complex lysate suspension containing scattering crystals and various impurities, it was possible to monitor the target molecule concentration in a heterogeneous, multi-phase process using spectroscopic methods. With the presented analytical set-up of off-line, particle-sensitive on-line, and in-line analyzers, a crystallization capture process can be characterized better in terms of the geometry, yield, and purity of the crystals., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Wegner, Eming, Walla, Bischoff, Weuster-Botz and Hubbuch.)

Chemometric modeling for spectral data is considered a key technology in biopharmaceutical processing to realize real-time process control and release testing. Machine learning (ML) models have been shown to increase the accuracy of various spectral regression and classification tasks, remove challenging preprocessing steps for spectral data, and promise to improve the transferability of models when compared to commonly applied, linear methods. The training and optimization of ML models require large data sets which are not available in the context of biopharmaceutical processing. Generative methods to extend data sets with realistic in silico samples, so-called data augmentation, may provide the means to alleviate this challenge. In this study, we develop and implement a novel data augmentation method for generating in silico spectral data based on local estimation of pure component profiles for training convolutional neural network (CNN) models using four data sets. We simultaneously tune hyperparameters associated with data augmentation and the neural network architecture using Bayesian optimization. Finally, we compare the optimized CNN models with partial least-squares regression models (PLS) in terms of accuracy, robustness, and interpretability. The proposed data augmentation method is shown to produce highly realistic spectral data by adapting the estimates of the pure component profiles to the sampled concentration regimes. Augmenting CNNs with the in silico spectral data is shown to improve the prediction accuracy for the quantification of monoclonal antibody (mAb) size variants by up to 50% in comparison to single-response PLS models. Bayesian structure optimization suggests that multiple convolutional blocks are beneficial for model accuracy and enable transfer across different data sets. Model-agnostic feature importance methods and synthetic noise perturbation are used to directly compare the optimized CNNs with PLS models. This enables the identification of wavelength regions critical for model performance and suggests increased robustness against Gaussian white noise and wavelength shifts of the CNNs compared to the PLS models., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Schiemer, Rüdt and Hubbuch.)

Non-enveloped virus-like particles (VLPs) are versatile protein nanoparticles with great potential for biopharmaceutical applications. However, conventional protein downstream processing (DSP) and platform processes are often not easily applicable due to the large size of VLPs and virus particles (VPs) in general. The application of size-selective separation techniques offers to exploit the size difference between VPs and common host-cell impurities. Moreover, size-selective separation techniques offer the potential for wide applicability across different VPs. In this work, basic principles and applications of size-selective separation techniques are reviewed to highlight their potential in DSP of VPs. Finally, specific DSP steps for non-enveloped VLPs and their subunits are reviewed as well as the potential applications and benefits of size-selective separation techniques are shown., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Hillebrandt and Hubbuch.)

The manufacturing of antibody-drug conjugates (ADCs) involves the addition of a cytotoxic small-molecule linker-drug (= payload) to a solution of functionalized antibodies. For the development of robust conjugation processes, initially small-scale reaction tubes are used which requires a lot of manual handling. Scale-up to larger reaction vessels is often knowledge-driven and scale-comparability is solely assessed based on final product quality which does not account for the dynamics of the reaction. In addition, information about the influence of process parameters, such as stirrer speed, temperature, or payload addition rates, is limited due to high material costs. Given these limitations, there is a need for a modeling-based approach to investigate conjugation scale-up. In this work, both experimental kinetic studies and computational fluid dynamics (CFD) conjugation simulations were performed to understand the influence of scale and mixing parameters. In the experimental part, conjugation kinetics in small-scale reaction tubes with different mixing types were investigated for two ADC systems and compared to larger bench-scale reactions. It was demonstrated that more robust kinetics can be achieved through internal stirrer mixing instead of external mixing devices, such as orbital shakers. In the simulation part, 3D-reactor models were created by coupling CFD-models for three large-scale reaction vessels with a kinetic model for a site-specific conjugation reaction. This enabled to study the kinetics in different vessels, as well as the effect of process parameter variations in silico . Overall, it was found that for this conjugation type sufficient mixing can be achieved at all scales and the studied parameters cause only deviations during the payload addition period. An additional time-scale analysis demonstrated to aid the assessment of mixing effects during ADC process scale-up when mixing times and kinetic rates are known. In summary, this work highlights the benefit of kinetic models for enhanced conjugation process understanding without the need for large-scale experiments., Competing Interests: Authors MW and RB are employed by AstraZeneca PLC. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Weggen, Seidel, Bean, Wendeler and Hubbuch.)

When developping selective crystallization or precipitation processes, biopharmaceutical modalities require empirical screenings and analytics tailored to the specific needs of the target molecule. The multi-way chemometric approach called parallel factor analysis (PARAFAC) coupled with ultraviolet visible light (UV/Vis) spectroscopy is able to predict specific concentrations and spectra from highly structured data sets without the need for calibration samples and reference analytics. These calculated models can provide exploratory information on pure species spectra and concentrations in all analyzed samples by representing one model component with one species. In this work, protein mixtures, monoclonal antibodies, and virus-like particles in chemically defined and complex solutions were investigated in three high-throughput crystallization or precipitation screenings with the aim to construct one PARAFAC model per case. Spectroscopic data sets of samples after the selective crystallization or precipitation, washing, and redissolution were recorded and arranged into a four-dimensional data set per case study. Different reference analytics and pure species spectra served as validation. Appropriate spectral preprocessing parameters were found for all case studies allowing even the application of this approach to the third case study in which quantitative concentration analytics are missing. Regardless of the modality or the number of species present in complex solutions, all models were able to estimate the specific concentration and find the optimal process condition regarding yield and product purity. It was shown that in complex solutions, species demonstrating similar phase behavior can be clustered as one component and described in the model. PARAFAC as a calibration-free approach coupled with UV/Vis spectroscopy provides a fast overview of species present in complex solution and of their concentration during selective crystallization or precipitation, washing, and redissolution., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Wegner and Hubbuch.)

Extrusion-based 3D bioprinting enables the production of customized hydrogel structures that can be employed in flow reactors when printing with enzyme-containing inks. The present study compares inks based on either low-melt agarose or agar at different concentrations (3-6%) and loaded with the thermostable enzyme esterase 2 from the thermophilic organism Alicyclobacillus acidocaldarius (AaEst2) with regard to their suitability for the fabrication of such enzymatically active hydrogels. A customized printer setup including a heatable nozzle and a cooled substrate was established to allow for clean and reproducible prints. The inks and printed hydrogel samples were characterized using rheological measurements and compression tests. All inks were found to be sufficiently printable to create lattices without overhangs, but printing quality was strongly enhanced at 4.5% polymer or more. The produced hydrogels were characterized regarding mechanical strength and diffusibility. For both properties, a strong correlation with polymer concentration was observed with highly concentrated hydrogels being more stable and less diffusible. Agar hydrogels were found to be more stable and show higher diffusion rates than comparable agarose hydrogels. Enzyme leaching was identified as a major drawback of agar hydrogels, while hardly any leaching from agarose hydrogels was detected. The poor ability of agar hydrogels to permanently immobilize enzymes indicates their limited suitability for their employment in perfused biocatalytic reactors. Batch-based activity assays showed that the enzymatic activity of agar hydrogels was roughly twice as high as the activity of agarose hydrogels which was mostly attributed to the increased amount of enzyme leaching. Agarose bioinks with at least 4.5% polymer were identified as the most suitable of the investigated inks for the printing of biocatalytic reactors with AaEst2. Drawbacks of these inks are limited mechanical and thermal stability, not allowing the operation of a reactor at the optimum temperature of AaEst2 which is above the melting point of the employed low-melt agarose., (Copyright © 2022 Wenger, Radtke, Gerisch, Kollmann, Niemeyer, Rabe and Hubbuch.)

Hydrogels based on renewable resources are a promising class of materials for future applications in pharmaceutics, drug delivery and personalized medicine. Thus, optional adjustments of mechanical properties such as swelling behavior, elasticity and network strength are desired. In this context, hydrogels based on the biological raw materials bovine serum albumin and casein were prepared by dityrosine-crosslinking of their tyrosine residues through visible light-induced photopolymerization. Changing the tyrosine accessibility by urea addition before photopolymerization increased the storage modulus of the hydrogels by 650% while simultaneously being more elastic. Furthermore, contributions of the buffer system composition, variation of protein concentration and storage medium towards mechanical properties of the hydrogel such as storage moduli, elasticity, fracture strain, compressive strength and relative weight swelling ratio are discussed. It could be shown, that changes in precursor solution and storage medium characteristics are crucial parameters towards tuning the mechanical properties of protein-based hydrogels., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Haas, Körner, Zintel and Hubbuch.)

Virus-like particles (VLPs) are macromolecular structures with great potential as vehicles for the targeted administration of functional molecules. Loaded with nucleic acids, VLPs are a promising approach for nanocarriers needed for gene therapy. There is broad knowledge of the manufacturing of the truncated wild-type lacking a nucleic acid binding region, which is mainly being investigated for vaccine applications. Whereas for their potential application as a nanocarrier for gene therapy, hepatitis B core antigen (HBcAg) VLPs with a nucleic acid binding region for efficient cargo-loading are being investigated. VLP structure, loading, and phase behavior are of central importance to their therapeutic efficacy and thereby considerably affecting the production process. Therefore, HBcAg VLPs with different lengths of the nucleic acid binding region were produced in E. coli . VLP attributes such as size, zeta potential, and loading with host cell-derived nucleic acids were evaluated. Capsid's size and zeta potential of the VLP constructs did not differ remarkably, whereas the analysis of the loading with host cell-derived nucleic acids revealed strong differences in the binding of host cell-derived nucleic acids dependent on the length of the binding region of the constructs, with a non-linear correlation but a two-zone behavior. Moreover, the phase behavior and purification process of the HBcAg VLPs as a function of the liquid phase conditions and the presence of host cell-derived nucleic acids were investigated. Selective VLP precipitation using ammonium sulfate was scarcely affected by the encapsulated nucleic acids. However, the disassembly reaction, which is crucial for structure homogeneity, separation of encapsulated impurities, and effective loading of the VLPs with therapeutic nucleic acids, was affected both by the studied liquid phase conditions, varying pH and concentration of reducing agents, and the different VLP constructs and amount of bound nucleic acids, respectively. Thereby, capsid-stabilizing effects of the bound nucleic acids and capsid-destabilizing effects of the nucleic acid binding region were observed, following the two-zone behavior of the construct's loading, and a resulting correlation between the capsid stability and disassembly yields could be derived., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Valentic, Müller and Hubbuch.)

Elastin-like proteins (ELPs) are polypeptides with potential applications as renewable bio-based high-performance polymers, which undergo a stimulus-responsive reversible phase transition. The ELP investigated in this manuscript-ELP[V2Y-45]-promises fascinating mechanical properties in biomaterial applications. Purification process scalability and purification performance are important factors for the evaluation of potential industrial-scale production of ELPs. Salt-induced precipitation, inverse transition cycling (ITC), and immobilized metal ion affinity chromatography (IMAC) were assessed as purification protocols for a polyhistidine-tagged hydrophobic ELP showing low-temperature transition behavior. IMAC achieved a purity of 86% and the lowest nucleic acid contamination of all processes. Metal ion leakage did not propagate chemical modifications and could be successfully removed through size-exclusion chromatography. The simplest approach using a high-salt precipitation resulted in a 60% higher target molecule yield compared to both other approaches, with the drawback of a lower purity of 60% and higher nucleic acid contamination. An additional ITC purification led to the highest purity of 88% and high nucleic acid removal. However, expensive temperature-dependent centrifugation steps are required and aggregation effects even at low temperatures have to be considered for the investigated ELP. Therefore, ITC and IMAC are promising downstream processes for biomedical applications with scale-dependent economical costs to be considered, while salt-induced precipitation may be a fast and simple alternative for large-scale bio-based polymer production., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Haas, Desombre, Kirschhöfer, Huber, Schiller and Hubbuch.)

As virtual reality (VR) has drastically evolved over the past few years, the field of applications of VR flourished way beyond the gaming industry. While commercial VR solutions might be available, there is a need to develop a workflow for specific applications. Bioprinting represents such an example. Here, complex 3D data is generated and needs to be visualized in the context of quality control. We demonstrate that the transfer to a commercially available VR software is possible by introducing an optimized workflow. In the present work, we developed a workflow for the visualization of the critical quality attribute (cQA) cell distribution in bioprinted (extrusion-based) samples in VR. The cQA cell distribution is directly influenced by the pre-processing step mixing of cell material in the bioink. Magnetic Resonance Imaging (MRI) was used as an analytical tool to generate spatially resolved 2.5 and 3D data of the bioprinted objects. A sample with poor quality in respect of the cQA cell distribution was identified as its inhomogeneous cell distribution could be displayed spatially resolved in VR. The described workflow facilitates the usage of VR as a tool for quality inspection in the field of bioprinting and represents a powerful tool for visualization of complex 3D MRI data., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Gretzinger, Schmieg, Guthausen and Hubbuch.)

Hydrogels are polymer-based materials with a high water content. Due to their biocompatible and cell-friendly nature, they play a major role in a variety of biotechnological applications. For many of these applications, diffusibility is an essential property influencing the choice of material. We present an approach to estimate diffusion coefficients in hydrogels based on absorbance measurements of a UV area imaging system. A microfluidic chip with a y-junction was employed to generate a fluid-hydrogel interface and the diffusion of lysozyme from the fluid into the hydrogel phase was monitored. Employing automated image and data processing, analyte concentration profiles were generated from the absorbance measurements and fits with an analytical solution of Fick's second law of diffusion were applied to estimate diffusion coefficients. As a case study, the diffusion of lysozyme in hydrogels made from different concentrations (0.5-1.5% (w/w)) of an unmodified and a low-melt agarose was investigated. The estimated diffusion coefficients for lysozyme were between 0.80 ± 0.04×10 -10  m 2  s -1 for 1.5% (w/w) low-melt agarose and 1.14 ± 0.02×10 -10  m 2  s -1 for 0.5% (w/w) unmodified agarose. The method proved sensitive enough to resolve significant differences between the diffusion coefficients in different concentrations and types of agarose. The microfluidic approach offers low consumption of analyte and hydrogel and requires only relatively simple instrumentation., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Wenger and Hubbuch.)

Bioprinting is increasingly regarded as a suitable additive manufacturing method in biopharmaceutical process development and formulation. In order to manage the leap from research to industrial application, higher levels of reproducibility and a standardized bioprinting process are prerequisites. This said, the concept of process analytical technologies, standard in the biopharmaceutical industry, is still at its very early steps. To date most extrusion-based printing processes are controlled over penumatic pressure and thus not adaptive to environmental or system related changes over several experimental runs. A constant set pressure applied over a number of runs, might lead to variations in flow rate and thus to unreliable printed constructs. With this in mind, the simple question arises whether a printing process based on a set flow rate could improve reproduciblity and transfer to different printing systems. The control and monitoring of flow rate aim to introduce the concept of PAT in the field of bioprinting. This study investigates the effect of different processing modes (set pressure vs. set flow rate) on printing reproducibility occurring during an extrusion-based printing process consisting of 6 experimental runs consisting of 3 printed samples each. Additionally, the influence of different filling levels of the ink containing cartridge during a printing process was determined. Different solutions based on a varying amount of alginate polymer and Kolliphor hydrogels in varying concentrations showed the need for individual setting of printing parameter. To investigate parameter transferability among different devices two different printers were used and the flow was monitored using a flow sensor attached to the printing unit. It could be demonstrated that a set flow rate controlled printing process improved accuracy and the filling level also affects the accuracy of printing, the magnitude of this effects varies as the cartridge level declined. The transferability between printed devices was eased by setting the printing parameters according to a set flow rate of each bioink disregarding the value of the set pressure. Finally, by a bioprinting porcess control based on a set flow rate, the coefficient of variance for printed objects could be reduced from 0.2 to 0.02 for 10% (w/v) alginate polymer solutions., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Strauß, Schroth and Hubbuch.)

Increasing the shelf life of enzymes and making them reusable is a prominent topic in biotechnology. The encapsulation inside hydrogel microparticles (HMPs) can enhance the enzyme's stability by preserving its native conformation and facilitating continuous biocatalytic processes and enzyme recovery. In this study, we present a method to immobilize β-galactosidase by, first, conjugating the enzyme onto the surface of polymer nanoparticles, and then encapsulating these enzyme-conjugated nanoparticles (ENPs) inside HMPs using microfluidic device paired with UV-LEDs. Polymer nanoparticles act as anchors for enzyme molecules, potentially preventing their leaching through the hydrogel network especially during swelling. The affinity binding (through streptavidin-biotin interaction) was used as an immobilization technique of β-galactosidase on the surface of polymer nanoparticles. The hydrogel microparticles of roughly 400 μm in size (swollen state) containing unbound enzyme and ENPs were produced. The effects of encapsulation and storage in different conditions were evaluated. It was discovered that the encapsulation in acrylamide (AcAm) microparticles caused an almost complete loss of enzymatic activity. Encapsulation in poly(ethylene glycol) (PEG)-diacrylate microparticles, on the other hand, showed a residual activity of 15-25%, presumably due to a protective effect of PEG during polymerization. One of the major factors that affected the enzyme activity was presence of photoinitiator exposed to UV-irradiation. Storage studies were carried out at room temperature, in the fridge and in the freezer throughout 1, 7 and 28 days. The polymer nanoparticles showcased excellent immobilization properties and preserved the activity of the conjugated enzyme at room temperature (115% residual activity after 28 days), while a slight decrease was observed for the unbound enzyme (94% after 28 days). Similar trends were observed for encapsulated ENPs and unbound enzyme. Nevertheless, storage at -26°C resulted in an almost complete loss of enzymatic activity for all samples., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Suvarli, Wenger, Serra, Perner-Nochta, Hubbuch and Wörner.)

In biopharmaceutical production processes, freeze-thaw operations are used to ensure product integrity during long hold times, but they also introduce additional stresses such as freeze concentration gradients that might lead to a loss of protein activity. Process characterization of freeze-thaw operations at different scales should be conducted with attention to freezing time and boundary effects to ensure the product stability throughout the process and process development. Currently, process characterization often relies on one or very few temperature probes that detect freezing times based on raw temperature, which is largely influenced by freezing-point depression in case of concentrated solutions. A method to detect freezing based on the second derivative of temperature measurements from Fiber-Bragg-Grating sensors is presented to overcome this issue. The applicability of the method is demonstrated by process characterization of a novel small-scale freeze-thaw device with minimized boundary effects using freezing times of purified water and concentrated formulations. Freezing times varied from 35 to 81 min for temperatures between -60 and -20°C and impacted freeze concentration profiles. Furthermore, freezing time estimations based on the Plank equation revealed model limitations due to start-up temperature gradients, that can be corrected by an empirically extended Plank model. As a hypothesis, we conclude that freezing temperature, from a freeze concentration view, is less important in containers with small characteristic freezing distances such as freeze bags. Using a 2D-resolved temperature profile, a shift of the last point to freeze position from top to bottom of a container was observed when freezing above -30°C., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Weber and Hubbuch.)

Chimeric virus-like particles (cVLPs) are protein-based nanostructures applied as investigational vaccines against infectious diseases, cancer, and immunological disorders. Low solubility of cVLP vaccine candidates is a challenge that can prevent development of these very substances. Solubility of cVLPs is typically assessed empirically, leading to high time and material requirements. Prediction of cVLP solubility in silico can aid in reducing this effort. Protein aggregation by hydrophobic interaction is an important factor driving protein insolubility. In this article, a recently developed soft ensemble vote classifier (sEVC) for the prediction of cVLP solubility was used based on 91 literature amino acid hydrophobicity scales. Optimization algorithms were developed to boost model performance, and the model was redesigned as a regression tool for ammonium sulfate concentration required for cVLP precipitation. The present dataset consists of 568 cVLPs, created by insertion of 71 different peptide sequences using eight different insertion strategies. Two optimization algorithms were developed that (I) modified the sEVC with regard to systematic misclassification based on the different insertion strategies, and (II) modified the amino acid hydrophobicity scale tables to improve classification. The second algorithm was additionally used to synthesize scales from random vectors. Compared to the unmodified model, Matthew's Correlation Coefficient (MCC), and accuracy of the test set predictions could be elevated from 0.63 and 0.81 to 0.77 and 0.88, respectively, for the best models. This improved performance compared to literature scales was suggested to be due to a decreased correlation between synthesized scales. In these, tryptophan was identified as the most hydrophobic amino acid, i.e., the amino acid most problematic for cVLP solubility, supported by previous literature findings. As a case study, the sEVC was redesigned as a regression tool and applied to determine ammonium sulfate concentrations for the precipitation of cVLPs. This was evaluated with a small dataset of ten cVLPs resulting in an R 2 of 0.69. In summary, we propose optimization algorithms that improve sEVC model performance for the prediction of cVLP solubility, allow for the synthesis of amino acid scale tables, and further evaluate the sEVC as regression tool to predict cVLP-precipitating ammonium sulfate concentrations., (Copyright © 2020 Vormittag, Klamp and Hubbuch.)

The immobilization of enzymes in biocatalytic flow reactors is a common strategy to increase enzyme reusability and improve biocatalytic performance. Extrusion-based 3D bioprinting has recently emerged as a versatile tool for the fabrication of perfusable hydrogel grids containing entrapped enzymes for the use in such reactors. This study demonstrates the suitability of water-in-oil high internal phase emulsions (HIPEs) as 3D-printable bioinks for the fabrication of composite materials with a porous polymeric scaffold (polyHIPE) filled with enzyme-laden hydrogel. The prepared HIPEs exhibited excellent printability and are shown to be suitable for the printing of complex three-dimensional structures without the need for sacrificial support material. An automated activity assay method for the systematic screening of different material compositions in small-scale batch experiments is presented. The monomer mass fraction in the aqueous phase and the thickness of printed objects were found to be the most important parameters determining the apparent activity of the immobilized enzyme. Mass transfer limitations and enzyme inactivation were identified as probable factors reducing the apparent activity. The presented HIPE-based bioinks enable the fabrication of flow-optimized and more efficient biocatalytic reactors while the automated activity assay method allows the rapid screening of materials to optimize the biocatalytic efficiency further without time-consuming flow-through experiments involving whole printed reactors., (Copyright © 2020 Wenger, Radtke, Göpper, Wörner and Hubbuch.)

Virus-like particles (VLPs) are emerging nanoscale protein assemblies applied as prophylactic vaccines and in development as therapeutic vaccines or cargo delivery systems. Downstream processing (DSP) of VLPs comes both with challenges and opportunities, depending on the complexity and size of the structures. Filtration, precipitation/re-dissolution and size-exclusion chromatography (SEC) are potent technologies exploiting the size difference between product and impurities. In this study, we therefore investigated the integration of these technologies within a single unit operation, resulting in three different processes, one of which integrates all three technologies. VLPs, contained in clarified lysate from Escherichia coli , were precipitated by ammonium sulfate, washed, and re-dissolved in a commercial cross-flow filtration (CFF) unit. Processes were analyzed for yield, purity, as well as productivity and were found to be largely superior to a reference centrifugation process. Productivity was increased 2.6-fold by transfer of the wash and re-dissolution process to the CFF unit. Installation of a multimodal SEC column in the permeate line increased purity to 96% while maintaining a high productivity and high yield of 86%. In addition to these advantages, CFF-based capture and purification allows for scalable and disposable DSP. In summary, the developed set-up resulted in high yields and purities, bearing the potential to be applied as an integrated process step for capture and purification of in vivo -assembled VLPs and other protein nanoparticles., (Copyright © 2020 Hillebrandt, Vormittag, Bluthardt, Dietrich and Hubbuch.)

Virus-like particles (VLPs) are protein-based nanoscale structures that show high potential as immunotherapeutics or cargo delivery vehicles. Chimeric VLPs are decorated with foreign peptides resulting in structures that confer immune responses against the displayed epitope. However, insertion of foreign sequences often results in insoluble proteins, calling for methods capable of assessing a VLP candidate's solubility in silico . The prediction of VLP solubility requires a model that can identify critical hydrophobicity-related parameters, distinguishing between VLP-forming aggregation and aggregation leading to insoluble virus protein clusters. Therefore, we developed and implemented a soft ensemble vote classifier (sEVC) framework based on chimeric hepatitis B core antigen (HBcAg) amino acid sequences and 91 publicly available hydrophobicity scales. Based on each hydrophobicity scale, an individual decision tree was induced as classifier in the sEVC. An embedded feature selection algorithm and stratified sampling proved beneficial for model construction. With a learning experiment, model performance in the space of model training set size and number of included classifiers in the sEVC was explored. Additionally, seven models were created from training data of 24-384 chimeric HBcAg constructs, which were validated by 100-fold Monte Carlo cross-validation. The models predicted external test sets of 184-544 chimeric HBcAg constructs. Best models showed a Matthew's correlation coefficient of >0.6 on the validation and the external test set. Feature selection was evaluated for classifiers with best and worst performance in the chimeric HBcAg VLP solubility scenario. Analysis of the associated hydrophobicity scales allowed for retrieval of biological information related to the mechanistic backgrounds of VLP solubility, suggesting a special role of arginine for VLP assembly and solubility. In the future, the developed sEVC could further be applied to hydrophobicity-related problems in other domains, such as monoclonal antibodies., (Copyright © 2020 Vormittag, Klamp and Hubbuch.)