Genome scale modeling of the protein secretory pathway reveals novel targets for improved recombinant protein production in yeast

Abstract

Eukaryal cells are used for the production of many recombinant pharmaceutical proteins, including several of the current top-selling products. The protein secretory pathway in eukaryal cells is complex and involves many different processes such as post-translational modifications, translocation, and folding. Furthermore, recombinant protein production competes with native secretory proteins for the limited energy and proteome resources allocated to the protein secretory pathway. Due to the complexity of this pathway, improvement through metabolic engineering has traditionally been relatively ad-hoc; and considering the industrial importance of this pathway, there is a need for more systematic approaches for novel design principles. Here, we present the first proteome-constrained genome-scale protein secretory model of a eukaryal cell, namely for the yeast Saccharomyces cerevisiae (pcSecYeast). The model contains all key processes of this pathway, i.e., protein translation, modification, and degradation coupled with metabolism. The model can capture delicate phenotypic changes such as the switch in the use of specific glucose transporters in response to changing extracellular glucose concentration. Furthermore, the model can also simulate the effects of protein misfolding on cellular growth, suggesting that retro-translocation of misfolded proteins contributes to protein retention in the Endoplasmic reticulum (ER). We used pcSecYeast to simulate various recombinant proteins production and identified overexpression targets for different recombinant proteins overproduction. We experimentally validated many of the predicted targets for -amylase production in this study, and the results show that the secretory pathways have more limited capacity than metabolism in terms of protein secretion.