Search

Abstract Chinese hamster ovary (CHO) cells are widely utilized in the production of antibody drugs. To ensure the production of large quantities of antibodies that meet the required specifications, it is crucial to monitor and control the levels of metabolites comprehensively during CHO cell culture. In recent years, continuous analysis methods employing on-line/in-line techniques using Raman spectroscopy have attracted attention. While these analytical methods can nondestructively monitor culture data, constructing a highly accurate measurement model for numerous components is time-consuming, making it challenging to implement in the rapid research and development of pharmaceutical manufacturing processes. In this study, we developed a comprehensive, simple, and automated method for constructing a Raman model of various components measured by LC–MS and other techniques using machine learning with Python. Preprocessing and spectral-range optimization of data for model construction (partial least square (PLS) regression) were automated and accelerated using Bayes optimization. Subsequently, models were constructed for each component using various model construction techniques, including linear regression, ridge regression, XGBoost, and neural network. This enabled the model accuracy to be improved compared with PLS regression. This automated approach allows continuous monitoring of various parameters for over 100 components, facilitating process optimization and process monitoring of CHO cells.

Abstract Chinese hamster ovary (CHO) cells are widely used for manufacturing antibody drugs. We attempted to clone a novel high-expression promoter for producing monoclonal antibodies (mAbs) based on transcriptome analysis to enhance the transcriptional abundance of mAb genes. The efficacy of conventional promoters such as CMV and hEF1α decrease in the latter phase of fed-batch cell culture. To overcome this, we screened genes whose expression was maintained or increased throughout the culture period. Since CHO cells have diverse genetic expression depending on the selected clone and culture medium, transcriptome analysis was performed on multiple clones and culture media anticipated to be used in mAb manufacturing. We thus acquired the Hspa5 promoter as a novel high-expression promoter, which uniquely enables mAb productivity per cell to improve late in the culture period. Productivity also improved for various IgG subclasses under Hspa5 promoter control, indicating this promoter’s potential universal value for mAb production. Finally, it was suggested that mAb production with this promoter is correlated with the transcription levels of endoplasmic reticulum stress-related genes. Therefore, mAb production utilizing the Hspa5 promoter might be a new method for maintaining protein homeostasis and achieving stable expression of introduced mAb genes during fed-batch culture.

There is a continuous demand to improve monoclonal antibody production for medication supply and medical cost reduction. For over 20 years, recombinant Chinese hamster ovary cells have been used as a host in monoclonal antibody production due to robustness, high productivity and ability to produce proteins with ideal glycans. Chemical compounds, such as dimethyl sulfoxide, lithium chloride, and butyric acid, have been shown to improve monoclonal antibody production in mammalian cell cultures. In this study, we aimed to discover new chemical compounds that can improve cell-specific antibody production in recombinant Chinese hamster ovary cells. Out of the 23,227 chemicals screened in this study, 4-(2,5-dimethyl-1H-pyrrol-1-yl)-N-(2,5-dioxopyrrolidin-1-yl) benzamide was found to increase monoclonal antibody production. The compound suppressed cell growth and increased both cell-specific glucose uptake rate and the amount of intracellular adenosine triphosphate during monoclonal antibody production. In addition, the compound also suppressed the galactosylation on a monoclonal antibody, which is a critical quality attribute of therapeutic monoclonal antibodies. Therefore, the compound might also be used to control the level of the galactosylation for the N-linked glycans. Further, the structure-activity relationship study revealed that 2,5-dimethylpyrrole was the most effective partial structure of 4-(2,5-dimethyl-1H-pyrrol-1-yl)-N-(2,5-dioxopyrrolidin-1-yl) benzamide on monoclonal antibody production. Further structural optimization of 2,5-dimethylpyrrole derivatives could lead to improved production and quality control of monoclonal antibodies.

Two-way Euler-Lagrange simulations are performed to characterize the hydrodynamics in the single-use bioreactor Mobius® CellReady 3 L. The hydrodynamics in stirred tank bioreactors are frequently modeled with the Euler–Euler approach, which cannot capture the trajectories of single bubbles. The present study employs the two-way coupled Euler–Lagrange approach, which accounts for the individual bubble trajectories through Langrangian equations and considers their impact on the Eulerian liquid phase equations. Hydrodynamic process characteristics that are relevant for cell cultivation including the oxygen mass transfer coefficient, the mixing time, and the hydrodynamic stress are evaluated for different working volumes, sparger types, impeller speeds, and sparging rates. A microporous sparger and an open pipe sparger are considered where bubbles of different sizes are generated, which has a pronounced impact on the bubble dispersion and the volumetric oxygen mass transfer coefficient. It is found that only the microporous sparger provides sufficiently high oxygen transfer to support typical suspended mammalian cell lines. The simulated mixing time and the volumetric oxygen mass transfer coefficient are successfully validated with experimental results. Due to the small reactor size, mixing times are below 25 s across all tested conditions. For the highest sparging rate of 100 mL min−1, the mixing time is found to be two seconds shorter than for a sparging rate of 50 mL min−1, which again, is 0.1 s longer than for a sparging rate of 10 mL min−1 at the same impeller speed of 100 rpm and the working volume of 1.7 L. The hydrodynamic stress in this bioreactor is found to be below critical levels for all investigated impeller speeds of up to 150 rpm, where the maximum levels are found in the region where the bubbles pass behind the impeller blades.

Understanding the hydrodynamic conditions in bioreactors is of utmost importance for the selection of operating conditions during cell culture process development. In the present study, the two-phase flow in the lab-scale single-use bioreactor XcellerexTM XDR-10 is characterized for working volumes from 4.5 L to 10 L, impeller speeds from 40 rpm to 360 rpm, and sparging with two different microporous spargers at rates from 0.02 L min−1 to 0.5 L min−1. The numerical simulations are performed with the one-way coupled Euler–Lagrange and the Euler–Euler models. The results of the agitated liquid height, the mixing time, and the volumetric oxygen mass transfer coefficient are compared to experiments. For the unbaffled XDR-10, strong surface vortex formation is found for the maximum impeller speed. To support the selection of suitable impeller speeds for cell cultivation, the surface vortex formation, the average turbulence energy dissipation rate, the hydrodynamic stress, and the mixing time are analyzed and discussed. Surface vortex formation is observed for the maximum impeller speed. Mixing times are below 30 s across all conditions, and volumetric oxygen mass transfer coefficients of up to 22.1 h−1 are found. The XDR-10 provides hydrodynamic conditions which are well suited for the cultivation of animal cells, despite the unusual design of a single bottom-mounted impeller and an unbaffled cultivation bioreactor.