Selection of CHO host cell subclones with increased specific antibody production rates by repeated cycles of transient transfection and cell sorting

Optimization of host cell lines both for transient and stable protein production is typically hampered by the inherent heterogeneity of cells within a population. This heterogeneity is caused not only by “hard fact” gene mutations, but also by subtle differences in the cellular network of regulation, which may include epigenetic variations. Taking advantage of this heterogeneity, we sorted for naturally occurring variants of CHO‐K1 and CHO‐S host cells that possess an improved cellular machinery for transient antibody production. The long‐term goal of this study was both to identify host cells that yield recombinant cell lines with on average higher productivity, but also to study the molecular differences that characterize such cells, independent of the site of gene integration or gene amplification. To identify such cells we optimized the procedure for transient transfection by electroporation to a degree that gave uniform transfer of plasmid DNA into nearly 100% of the cells and resulted in reproducible average productivities, with a standard deviation of 16% between independent experiments. Using this optimized protocol, the 1% of cells with the highest specific productivity was sorted and subcloned with a cold capture secretion assay. Upon re‐transfection, the resulting subclones showed the same specific productivity as their respective parental cell line. To enrich for cells with potentially stable improved properties, the 1% highest producers were sorted three times, 2 days after transient transfection each, and the enriched population was again sorted into microtiter plates for subcloning. For each of the two parental cell lines tested, three subclones were obtained that had a threefold higher specific productivity after transient transfection. This property was stable for approximately 3 months, indicating that the changes in productivity were regulatory and not mutational. Biotechnol. Bioeng. 2011;108: 386–394. © 2010 Wiley Periodicals, Inc.