While under certain nutrient-replete culture conditions, CHO cell lines can undergo a metabolic shift to lactic acid consumption, the molecular mechanisms governing this phenomenon are still elusive. Systems analysis of fed-batch cultures demonstrating a robust metabolic shift were performed to identify proteins that might be induced under such conditions.  One protein stood out and subsequent analysis revealed that conditions such as low pH, high levels of glucose, lactate, and other metabolic byproducts, and additionally the dynamics of accumulation of these byproducts, all controlled the induction of this protein.  An enhanced understanding of the molecular mechanisms underlying the shift to lactic acid uptake will help in developing novel strategies for enhanced performance of cell culture process in large-scale bioreactors.

We present a high-throughput LC-MS method that is capable of simultaneously monitoring 100+ cell culture metabolites. This method has high precision and accuracy and has been successfully applied to the daily profiling of bioreactors and raw material qualification. Information obtained in these studies has been used to identify limiting amino acids during production, which guided adjustments to the feed strategy that prevented the potential misincorporation of amino acids.

Monoclonal antibody perfusion production through continuous capture has been optimized by leveraging cell culture media design, Omics, and residence time distribution modeling. The cell culture media refinements addressed sieving efficiency and improved the process yield. Omics data was captured to understand the differences in metabolism between fed-batch and perfusion processes to improve specific productivity, allowing further yield improvements. Residence time distribution modeling was performed using both a small molecule glycan inhibitor and a large molecule (> 50 kDa) to ascertain product quality deviations and impact of sieving on the mean residence time of the mAb.

In this case study, we will demonstrate that a reduction of cysteine or cystine and tryptophan concentrations in the feed lead to reduce charge variants, as well as decrease of drug color intensity. Based on the results of the study, a model was developed to predict charge variants and was combined to an integrated model predicting cell growth, cell productivity, and cell metabolism. This modeling approach shed further light on the impact and control of the feeding strategy on product quality.

Cell line development focuses on optimizing two main processes: Maximizing single-cell clone isolation and ensuring top clones demonstrate high quality expression and/or production. This may be hindered by lab size, budget, limited technical expertise, finicky cells, or harsh instruments. Here we present case studies that address these issues.

The presentation will cover topics on the development and optimization of a CHO based expression system. The cell culture process and media are known to have a significant effect on the expression levels. With a robust production platform, we are addressing the needs of early hit triage (for thousands of candidates using high throughput expression) through candidate selection for Lead Optimization and final production of gram quantities of the recombinant therapeutic for advancement to the clinic.

Under bioprocess conditions, CHO cells can be exposed to oxygen microheterogeneity, free radicals generated by cell culture medium components and high oxidative metabolism, which can lead to oxidative stress and loss of productivity. To characterize the metabolic response to oxidative stress in industrial cell lines, we have modulated intracellular GSH levels and performed a genome-wide proteomic analysis. Oxidative stress-related proteins, and interestingly, cholesterol biosynthesis were mostly impacted.

GPEx® cell line development technology generates highly stable, high titer production cell lines. Recently, GPEx technology and a glutamine synthase knock-out CHO cell line were combined in a unique way to create GPEx Boost. The new technology results in higher specific productivities, higher titers, improved consistency, and improved cell growth characteristics as compared to GPEx for most protein products.

Metabolic network models provide mechanistic understandings of cell metabolism, and therefore, could guide the rational design of cell lines and culture processes. However, such models are liable to overfit due to their high degrees of freedom. Here, we systematically evaluate a wide range of model parameters important to describing CHO fed-batch culture performance. We evaluated thousands of combinations of parameter settings to identify five critical parameters, their optimal ranges of values, and their risk of overfitting. These insights will guide future model development and applications for industrial mammalian cell cultures.

Glycosylation of therapeutic proteins can have severe implications on activity, half-life and response. We have engineered a large panel of CHO cell lines for the production of therapeutic proteins with tailormade glycosylation. Using this panel we are able to quickly produce defined glycovariants of protein-based drug candidates to screen for the optimal variant to bring into development. Thereby improving the likelihood of bringing the optimal candidates into clinical trials.

There have been many studies for the lactate metabolism in CHO cell culture, as it is known to affect productivity and culture longevity significantly. The presentation shares Samsung’s strategy for the identification and improvement of lactate metabolism variations, describing solution to lactate re-accumulation, which leads to successful scale-up for production.

The time required to develop and produce a new molecule and put it in the hands of patients is an important consideration in the drug development process. Utilization of novel tools, such as high-throughput systems and modeling allow for increased and more targeted development, ultimately reducing timelines and costs. Here, we demonstrate the application and impact of these tools on major activities in upstream process development.

In this talk, single cell RNASeq of a model CHO cell line producing an anti IL-8 monoclonal antibody will be presented. Through this analysis, transcriptome profiles of > 3,500 individual CHO cells were acquired revealing heterogeneity in the expression of heavy and light chain mRNA and enabling the association of host cell mRNA levels with these variations in transcription of the recombinant protein.

This presentation will introduce how the Beacon platform enables users to generate cell lines secreting traditional and non-traditional antibody molecules with >99% monoclonality assurance in under 1 week.  Case studies will highlight how Beacon users are rapidly generating cell lines with titers superior to clones selected with alternative CLD methods.