Exometabolome analysis reveals hypoxia at the up-scaling of a Saccharomyces cerevisiae high-cell density fed-batch biopharmaceutical process.

Zhibiao Fu,Brante P Sampey,Thomas D Verderame,Pramatesh S Patel,Edward R Appelbaum,Julie M Leighton,Juan C Aon

doi:10.1186/1475-2859-13-32

Zhibiao Fu, Brante P Sampey + Show 5 more

Open Access

https://doi.org/10.1186/1475-2859-13-32

Copy DOI

Abstract

BackgroundScale-up to industrial production level of a fermentation process occurs after optimization at small scale, a critical transition for successful technology transfer and commercialization of a product of interest. At the large scale a number of important bioprocess engineering problems arise that should be taken into account to match the values obtained at the small scale and achieve the highest productivity and quality possible. However, the changes of the host strain’s physiological and metabolic behavior in response to the scale transition are still not clear.ResultsHeterogeneity in substrate and oxygen distribution is an inherent factor at industrial scale (10,000 L) which affects the success of process up-scaling. To counteract these detrimental effects, changes in dissolved oxygen and pressure set points and addition of diluents were applied to 10,000 L scale to enable a successful process scale-up. A comprehensive semi-quantitative and time-dependent analysis of the exometabolome was performed to understand the impact of the scale-up on the metabolic/physiological behavior of the host microorganism. Intermediates from central carbon catabolism and mevalonate/ergosterol synthesis pathways were found to accumulate in both the 10 L and 10,000 L scale cultures in a time-dependent manner. Moreover, excreted metabolites analysis revealed that hypoxic conditions prevailed at the 10,000 L scale. The specific product yield increased at the 10,000 L scale, in spite of metabolic stress and catabolic-anabolic uncoupling unveiled by the decrease in biomass yield on consumed oxygen.ConclusionsAn optimized S. cerevisiae fermentation process was successfully scaled-up to an industrial scale bioreactor. The oxygen uptake rate (OUR) and overall growth profiles were matched between scales. The major remaining differences between scales were wet cell weight and culture apparent viscosity. The metabolic and physiological behavior of the host microorganism at the 10,000 L scale was investigated with exometabolomics, indicating that reduced oxygen availability affected oxidative phosphorylation cascading into down- and up-stream pathways producing overflow metabolism. Our study revealed striking metabolic and physiological changes in response to hypoxia exerted by industrial bioprocess up-scaling.

Highlights

Scale-up to industrial production level of a fermentation process occurs after optimization at small scale, a critical transition for successful technology transfer and commercialization of a product of interest
Performance of the optimized fermentation process at the small scale In previous work, we reported the optimization of a S. cerevisiae high-cell density fed-batch process using a Bayesian approach [17]
The highest pure oxygen sparge demand was determined to be less than 50% of the total capacity of 5 standard liters per minute by the end of run (EOR) at elapsed fermentation time (EFT) 80 hr

Summary

Introduction

Scale-up to industrial production level of a fermentation process occurs after optimization at small scale, a critical transition for successful technology transfer and commercialization of a product of interest. The cultivation of microbial strains over-expressing recombinant or native proteins is an established technique in the biopharmaceutical industry [1] In this sector, fermentation is normally developed and optimized at small scale in the laboratory and scaled up for commercial production. Studies using computational fluid dynamics have shown that considerable glucose gradients are expected when a standard 500 g/L glucose solution is fed into a 22,000 L bioreactor [7] These findings have received experimental confirmation, and it has been further shown that cells were frequently exposed to peak glucose concentrations in the addition zone that were several times higher than the mean [5,8] causing changes in cell metabolism, productivity and product quality [9,10,11,12,13]. Small laboratory bioreactors exhibit low mixing times (≤ 5 s) without significant spatio-temporal variations [14]

Methods

Results

Discussion

Conclusion