Rewiring the Yeast Transcriptome to Evolve Networks with Enhanced Heterologous Expression Capability.

View all posters

Oliver Windram, Sangjin Lee, Matthew Haines and Travis S Bayer

United Kingdom

Pichia pastoris is a useful heterologous protein expression system capable of complex post-translational modifications like glycosylation, disulfide bond formation, phosphorylation and proteolytic processing. However, heterologous protein expression presents a highly unnatural cellular state that invokes a number of responses that adversely affect product quality and quantity. Miss-folded protein can invoke unfolded protein stress response reducing production, while hyperglycosylation gives unfavorable by-products. Induction of many of these deleterious processes is governed transcriptionally. Reprogramming the host transcriptome is a potentially useful way to circumvent or repurpose these events to improve productivity. We have rewired the yeast transcriptome combining promoters open reading frames (ORFs) in a library to assess the influence of ~2800 unique networks on protein production. These are created using 67 promoters and 43 ORFs from transcriptional regulators involved in stress, protein transport and translation. The ability of regulatory rewirings to influence protein productivity was assessed using different heterologous proteins, a membrane bound opoid receptor, human insulin, a large repetitive spider silk protein and an immunotoxin fusion protein. Thus, we were able to probe the library for solutions to problems that currently limit protein production in this system. Our results indicate network rewiring can improve yields up to 8-fold in some cases. Library screening revealed enrichment for artificial rewired networks across all and selected subsets of heterologous protein types. This suggests that rewired networks can lead to general improvement in protein production as well as production improvement for proteins with specific physical characteristics such as large size or highly repetitive sequence. We will present how this network creation technique can lead to elegant expression improvement solutions such as negative feed back loop enforcement, which are not easily achievable using classical network perturbation methods.