Stress in recombinant protein producing yeasts.

It is well established today that heterologous overexpression of proteins is connected with different stress reactions. The expression of a foreign protein at a high level may either directly limit other cellular processes by competing for their substrates, or indirectly interfere with metabolism, if their manufacture is blocked, thus inducing a stress reaction of the cell. Especially the unfolded protein response (UPR) in Saccharomyces cerevisiae (as well as some other yeasts) is well documented, and its role for the limitation of expression levels is discussed. One potential consequence of endoplasmatic reticulum folding limitations is the ER associated protein degradation (ERAD) involving retrotranslocation and decay in the cytosol. High cell density fermentation, the typical process design for recombinant yeasts, exerts growth conditions that deviate far from the natural environment of the cells. Thus, different environmental stresses may be exerted on the host. High osmolarity, low pH and low temperature are typical stress factors. Whereas the molecular pathways of stress responses are well characterized, there is a lack of knowledge concerning the impact of stress responses on industrial production processes. Accordingly, most metabolic engineering approaches conducted so far target at the improvement of protein folding and secretion, whereas only few examples of cell engineering against general stress sensitivity were published. Apart from discussing well-documented stress reactions of yeasts in the context of heterologous protein production, some more speculative topics like quorum sensing and apoptosis are addressed.

Bacterial expression and refolding of human trypsinogen.

The expression of recombinant trypsinogens from different mammalian origins in Escherichia coli typically leads to the formation of insoluble aggregates. This work describes the high level expression of human trypsinogen 1 in E. coli using the T7 expression system. Direct expression of trypsinogen was not possible, but the N-terminal fusion of the first 11 amino acids of the T7 protein 10 resulted in an expression level of 200 mg g(-1) bacterial dry mass. A refolding procedure was optimized, and a method using continuous feed of denatured product was developed. Thus the working concentration of trypsinogen could be raised four-fold, while the yield of active protein could be maintained at 20-35%. The refolded trypsinogen was converted to trypsin by autocatalytic activation, and the utility for the detachment of mammalian cells in culture was proven.

Effects of gene dosage, promoters, and substrates on unfolded protein stress of recombinant Pichia pastoris.

The expression of heterologous proteins may exert severe stress on the host cells at different levels. Depending on the specific features of the product, different steps may be rate-limiting. For the secretion of recombinant proteins from yeast cells, folding and disulfide bond formation were identified as rate-limiting in several cases and the induction of the chaperone BiP (binding protein) is described. During the development of Pichia pastoris strains secreting human trypsinogen, a severe limitation of the amount of secreted product was identified. Strains using either the AOX1 or the GAP promoter were compared at different gene copy numbers. With the constitutive GAP promoter, no effect on the expression level was observed, whereas with the inducible AOX1 promoter an increase of the copy number above two resulted in a decrease of expression. To identify whether part of the product remained in the cells, lysates were fractionated and significant amounts of the product were identified in the insoluble fraction containing the endoplasmic reticulum, while the soluble cytosolic fraction contained product only in clones using the GAP promoter. An increase of BiP was observed upon induction of expression, indicating that the intracellular product fraction exerts an unfolded protein response in the host cells. A strain using the GAP promoter was grown both on glucose and methanol and trypsinogen was identified in the insoluble fractions of both cultures, but only in the soluble fraction of the glucose grown cultures, indicating that the amounts and distribution of intracellularly retained product depends on the culture conditions, especially the carbon source.

Assessing viability and cell-associated product of recombinant protein producing Pichia pastoris with flow cytometry.

This paper describes the establishment of flow cytometric methods for recombinant Pichia pastoris strains, and their application to a lab scale fed batch fermentation. Using a strain which secretes human trypsinogen, the viability and the product which remained associated to the cell were measured with propidium iodide and immunofluorescent staining, respectively. Viability decreases significantly below 70% during the methanol fed batch phase, indicating a stress situation triggered by the fermentation conditions. Cell associated product is accumulated earlier after methanol induction than secreted product. These data demonstrate that flow cytometry is a powerful tool for the analysis and optimization of recombinant protein production processes, and they indicate the need to further improve a widely used fermentation protocol for P. pastoris.