Development of Saccharomyces cerevisiae as a recombinant antibody factory

The yeast Saccharomyces cerevisiae has been widely used as an expression host for the manufacturing of products like biofuels, small molecules, and of recombinant proteins. To increase the yields of economically interesting proteins, the secretory pathway has been engineered extensively, however, secretion titers have often remained low. One example for these short-comings are full-length IgG antibodies, which are currently mostly produced in CHO cells, although microbial and plant based production platforms are emerging. We believe that S.cerevisiae has the potential to become an industrially relevant antibody factory. In this thesis, using targeted and random screening approaches, we aimed to identify genetic factors that can beused to create strains with an increased IgG secretion efficiency.

First, we focused on genes that are regulated by the unfolded protein response and encode proteins affecting the ER folding environment. We enlarged the functional folding space in the ER through deletion of the OPI1 gene, a modification that increased IgG titers by up to 4.8-fold. Out of a screen of folding catalysts and molecular chaperones, the over expression of the peptidyl-prolyl isomerase Cpr5p provided the most beneficial effect, increasing IgG titers by up to3.26-fold. Finally, by combining the OPI1 deletion with CPR5 overexpression IgG secretion was increased over tenfold when compared with the wild type background. In contrast, in a set of strains with deletions of genes encoding proteins of the ER associated degradation pathway only deletion of HTM1 increased titers by 1.15-fold. Development of a clearance assay allowedus to distinguish differences in cellular IgG clearance among the ERAD deletion strains.

As targets for rational strain engineering are limited, we developed a high throughput method for screening a transposon mediated yeast deletion library and identified genes that influence IgG secretion. With this approach, we were able to identify the genes VPS30 and TAR1 that after deletion improved IgG secretion by up to 2.5-fold and up 1.13-fold, respectively, thus validating the applicability of the method. Finally, we aimed to gain insight into the changes that recombinant antibody production inflicts on selected intracellular metabolites. Metabolic footprints of strains expressing a scFv, a scFv-Fc fusion, and a full-length IgG were found to besignificantly different based on a semi-quantitative metabolomics method. The most apparent changes were found in metabolites involved in amino acid and redox metabolism.

In conclusion, we identified genes at several places along the secretory pathway that can beused to improve IgG secretion in S. cerevisiae.