High-Throughput DoE Of 52 Parallel Bioreactors Doubles Titer In 1 Week
Data from an ongoing collaboration with Cytovance Biologics to rapidly optimize and scale bioprocesses for therapeutic proteins.
Culture Biosciences empowers scientists to run dozens of bioreactors in parallel for rapid bioprocess development. In collaboration with Cytovance Biologics, we ran a high-throughput process development design-of-experiments (DoE) study to increase production of recombinant human interleukin-2 (rhIL-2), a therapeutic protein, in E. coli. By running 52 bioreactors in parallel to generate a comprehensive process response surface, we quickly doubled rhIL-2 production while also identifying opportunities for additional process development for further increases in process robustness and titer.
Figure 1: Graphs generated on Culture Bioscience’s website depicting important process parameters in the 52 bioreactor runs. Note there was a programmed cool-down at the end of the process around 53 hours EFT, and the dynamic zoom is being used to examine the relevant process data.
This study was part of an ongoing collaboration with Cytovance to rapidly develop, scale-up, and characterize bioprocesses for therapeutic proteins. After transferring and validating the scale-down of Cytovance’s 5-L process to Culture’s 250-ml bioreactors, we used a central composite design methodology to generate a comprehensive process response surface. The study explored several factors, including pH, production temperature, and glucose bolus feed rate. All conditions were run in triplicate to improve robustness in the resulting model.
Figure 2: Depiction of the central composite experimental design parameters: glucose bolus feed rate, production temperature, and pH setpoint.
The results were analyzed in JMP and indicate that pH setpoint and production temperature have the greatest effect on product titer. While the response surface (shown in Figure 3) identified a low pH and high production temperature condition that increased titer by 2x over the baseline, the model also identified an increase in acetate production at higher temperatures. These results suggest there may be further improvements gained from exploring even lower pH and/or even higher production temperature. This additional understanding will aid in further process development and characterization work for production of rhIL-2.
Figure 3: The comprehensive process response surface generated by the DoE. Note that while the lowest pH and highest temperature conditions that were tested led to the highest titer, the shape of the surface indicates that there could be further improvements from exploring conditions outside these ranges.
We look forward to further optimization of this process, and to sharing the results of Cytovance scaling it up at their site. Sign up for our newsletter and follow us on social media to stay tuned for additional results of this collaborative effort to shorten bioprocess R&D timelines.
“Culture Biosciences has built an impressive facility and even more impressive team. Their work is top quality and their collaborative attitudes ensured the project was wildly successful and fun. Their real-time data portal was impressive and user-friendly. We look forward to continuing our collaboration project with them. We are working toward a demonstration that our clients could get full upstream process development or process characterization performed on a shorter timeline. ” - April Stanley, Associate Director of Open Innovation at Cytovance