Dynamic Ranking of Clones with Process Conditions Using a High-Throughput Micro-bioreactor Platform

Abstract

The economic production of recombinant proteins in mammalian cells is highly dependent upon the selection of high-producing cell lines (Rathore and Winkle Nat Biotechnol 27(1):26–34, 2009). Traditional screening technologies remain limited in their ability to mimic the bioreactor environment where the production clone will ultimately need to perform. Because of its inherent throughput, the SimCell platform uniquely enables the application of a Dynamic Ranking of Clones selection strategy. This approach achieves a comprehensive ranking of multiple cell line performances as a function of both media and process variables in a single experiment. In this study, eight CHO clones producing a recombinant monoclonal antibody were screened across several process variations, including different feeding strategies, temperature shifts and pH control profiles. A total of 240 micro-bioreactors were run in a single experiment, for two weeks, to assess the scale-down model as a high-throughput tool for clone evaluation. Clones were ranked for growth and titer performance in the fed-batch experimental design. Best and worst performers were clearly identified. The second phase experiment utilized 180 micro-bioreactors in a full factorial design comprised of a subset of 12 clone/process combinations run in parallel in duplicate shake flasks. Good correlation between the micro-bioreactor predictions and those made in shake flasks was obtained (R ² = 0.90).