March 1, 2016 (Vol. 36, No. 5)

Rena Baktur Cell Culture Scientist GE Healthcare
Alicia Elwood Process Manager II-Cell Culture GE Healthcare
Jessica Anderson GE Healthcare Life Sciences
Mark Wight

Employing Design of Experiments and High-Throughput Medium Development Methods

Chinese hamster ovary (CHO) cells are extensively used in the production of recombinant proteins. Even when expressing similar types of biologicals, individual CHO cell clones have shown a large degree of variability when it comes to nutritional needs. Every transfected CHO cell line has unique nutritional requirements for optimal cell growth and recombinant protein production.

One of the most challenging aspects of culturing recombinant CHO cells is, in fact, the optimization of the diverse nutritional requirements unique to each clone. Cell culture optimization often involves the development of a customized, chemically defined medium along with process-specific feed supplements, all combined with process parameters individualized for each CHO cell clone.

Medium and feed development is a complex process. It involves the adjustment of numerous interacting components to their final concentrations, and it aims to achieve strong, consistent support of cell growth and productivity.

Traditional medium optimization strategies are labor intensive, costly, and time consuming. High-throughput screening technology along with statistical design of experiments (DoE) methodology provides a pathway to reducing costs and decreasing the development time for reaching optimal cell-culture conditions.

High-throughput screening and DoE approaches both figure in a cell culture medium and feed strategy that has been optimized for recombinant CHO cell cultures. This strategy has been shown to substantially increase peak cell density and productivity, all while reducing the development timeline by as much as 50%.1 It is not uncommon to realize improvements in peak cell density by as much as 35% and doubled overall cell productivity.

Following this streamlined approach, customization projects generally consist of two separate phases: Phase 1, basal medium optimization, and Phase 2, feed optimization. Each phase consists of two rounds: Round 1 employs high-throughput methods using 1 mL cultures in 96-well deep-well plates, and Round 2 uses DoE methodology in 35 mL shake-flask cultures. Taken together, these phases identify the prototype medium having the greatest effect on cell growth and productivity.

Conducting the Study

Phase 1, Round 1 begins with an initial screen of in-house formulations selected for their demonstrated performance with various monoclonal antibody (MAb)-producing CHO cell clones. After this initial screen, a second high-throughput screen is conducted using up to 48 conditions, in some cases representing different formulations and single-component variations. Thereafter, the top three formulations from this round are selected for further optimization in the next round.

In Phase 1, Round 2, the media are further optimized using 20 conditions and evaluated with a DoE mixture design and spent medium analyses of amino acids, D vitamins, and trace elements (Figure 1). The formulation improvements from these studies eliminate the need for additional rounds of optimization and result in a single, final basal medium formulation.

Phase 2, Round 1 feed development also uses high-throughput 1 mL cultures in 96-well deep-well plates. Round 1 includes up to 48 conditions that consist of various high-performing feed prototypes supplemented at differing feed intervals. The three conditions that emerge as the top performers with respect to productivity are advanced to the next round of optimization.

In Phase 2, Round 2, the DoE feed studies optimize specific components and evaluate lower-passage and higher-passage cells for any observed differences. The resulting productivity levels from this stage of the project commonly double a clone’s titer. In a recent study, a titer of 3.7 g/L was achieved with a low-producing clone, more than doubling the baseline conditions for this cell line and exceeding the initial target.

With the approach of coupling traditional shake-flask culture studies with high-throughput screening processes and DoE methods, efficient design and optimization of cell culture media and feeds can be achieved. This approach delivers significant time savings through the elimination of repetitive rounds for evaluation of basal medium formulations, feed formulations, and fed-batch process parameters, while still achieving optimal cell growth and productivity.

Another benefit realized by initiating the optimization protocol from in-house prototypes is the confirmed scalability of the final medium and feed formulations. By starting with known formulations, and then customizing those formulations using DoE methodology, this optimization approach can result in formulations that are fit for large-scale production and supply. Practical studies evaluating the scalability of conditions from the high-throughput screening format and shake flasks to bioreactor scale confirm the suitability of the final formulations for use in large-volume, stirred-tank bioreactor cultures.


Figure 1. Phase 1, Round 2: evaluation of peak viable cell density (VCD) and productivity (IgG) from 20 conditions using a DoE mixture design during medium optimization.

1 Poster: A rapid and efficient approach to medium and feed optimization increasing CHO antibody production. GE Healthcare, 29168696, Edition AB (2015).

Rena Baktur is a cell culture scientist, Alicia Elwood works as process manager II-cell culture, and Mark Wight ([email protected]) serves as manager, cell culture development at GE Healthcare. Jessica Anderson was formerly associated with the GE Healthcare Life Sciences business.

Previous articleIntratumor Heterogeneity Poses Challenge for Cancer “Moonshot” Program
Next articleOnxeo to Acquire DNA Therapeutics