Random epigenetic modulation of CHO cells by repeated knock-down of DNA-methyltransferases increases population diversity and enables sorting of cells with higher production capacities. Epigenetic Modulation of CHO by RNAi

Chinese hamster ovary (CHO) cells produce a large share of today’s biopharmaceuticals. Although, CHO cells bring the basic requirements for the production of human-like therapeutic proteins, the generation of a satisfactory producer cell line for an economically viable production process is still a tedious and elaborate undertaking. Epigenetics, including the methylation of cytosine bases in the DNA and various histone tail modifications, are a mechanism of eukaryotic cells to regulate their genome wide gene expression pattern without changing their genetic sequence. Hence, whole transcriptome patterns are controlled by epigenetic marks. Recently, it was found that CHO cells, when exposed or adapted to new environmental conditions, modify their epigenome, or more specifically their DNA methylation pattern, suggesting that cells thus adapt their gene expression pattern to handle new challenges. The major aim of the present study was to employ artificially induced, random changes in the DNA methylation pattern of CHO cells to diversify a given cell population and consequently isolate new cell lines with improved cellular characteristics, e.g. productivity. To achieve this, DNA methyltransferases (DNMT) 1 and 3a and/or the Ten-eleven Translocation (TET) enzymes 2 and 3 were selectively down-regulated by RNA interference (RNAi) by repeated transfection over a time span of ~16 days. Whole genome bisulfite sequencing of the resulting cell pools revealed that the knock down of DNMTs was highly effective in demethylating the genome by ~50% and in generating a diverse pattern of DNA methylation across the population. The same approach when applied to stable CHO producer cells resulted in (i) an increased diversity in the cell population as shown by flow cytometric analysis of secreted product, as i.e.: the number of outlying cells was increased upon treatment, and (ii) improved production characteristics in the sorted best producers. Intriguingly, this behavior was observed for two, initially lower producing cell lines, but was not seen in a cell line producing already at industrially relevant levels. This suggests that this “engineering” strategy is potentially most beneficial in non-optimized, early production clones while high producers already have an optimal transcriptome pattern. In summary, this study provides proof-of-concept that mammalian cell populations can be diversified upon inducing changes in their DNA methylation pattern by targeted knock-down of DNMTs and that this diversity can be exploited during cell line development to isolate new cell lines with improved production capacities