The application and validation of high-throughput methods in Chinese hamster ovary cell line development

Chinese hamster ovary (CHO) cells are the main production platform for biotherapeutic proteins. In this thesis, two areas were identified and investigated with improving recombinant protein production from CHO cells and cell line development; transcriptomics and technologies for industrial cell line development (CLD).

Transcriptomics wise, publicly available data were identified and analysed in order to identify 'common' transcriptomic signatures of cell growth or productivity in order to devise novel cell line engineering strategies. From the literature, 19 different transcriptomic datasets were aggregated that explored the differences between high productivity and fast growth phenotypes. Here, we proceeded to analyse the data in terms of the two simplest dimensions - the frequencies of genes appearing across these data sets and the concordance (the arithmetic mean of expression values) with regard to cell growth (µ) and productivity (Qp). By mapping out the contributing genes it was possible to construct a transcriptomic 'fingerprint' of a high-performing cell line. After identifying the most common and concordant genes, those genes that had a frequency of two or more were analysed using a pathway enrichment algorithm. From this it was identified that the cell cycle and lysosome pathways are significant targets for cell line engineering. To our knowledge, this effort is the first of its kind within CHO transcriptomics.

CLD involves labour and resource intensive cloning out a genetically diverse pool of cells engineered to produce the protein of interest. We sought to analyse a new single cell analysis methodology (Berkley Lights Beacon, BLB) against an industrial ClonePix™ 2 CLD process adapted from FUJIFILM Diosynth Biotechnologies (FDB). We found that there were no statistically significant differences between cell groups generated from the BLB or ClonePix™ 2 processes. Using the Beacon® system, it was possible to predict 3 out of the top 5 producing clones for both Etanercept and Blosozumab. Within the standard ClonePix™ 2 CLD group of cell lines, predictions were most accurate from 24-well plate fed-batch and TubeSpin® batch culture ranks. Further, using the BLB, the time from recovery from transfection to cultures that were ambr® ready was reduced from 65 days to 42 days. Based on the findings of this research it is proposed that the Beacon® is an attractive and powerful new tool in industrial cell line development efforts. To the authors knowledge, this is the first in depth work validating a next-generation CLD process in such detail.