Sulfate limitation increases specific plasmid DNA yield and productivity in E. coli fed-batch processes

Abstract

Plasmid DNA (pDNA) is a key biotechnological product whose importance became apparent in the last years due to its role as a raw material in the messenger ribonucleic acid (mRNA) vaccine manufacturing process. In pharmaceutical production processes, cells need to grow in the defined medium in order to guarantee the highest standards of quality and repeatability. However, often these requirements result in low product titer, productivity, and yield. In this study, we used constraint-based metabolic modeling to optimize the average volumetric productivity of pDNA production in a fed-batch process. We identified a set of 13 nutrients in the growth medium that are essential for cell growth but not for pDNA replication. When these nutrients are depleted in the medium, cell growth is stalled and pDNA production is increased, raising the specific and volumetric yield and productivity. To exploit this effect we designed a three-stage process (1. batch, 2. fed-batch with cell growth, 3. fed-batch without cell growth). The transition between stage 2 and 3 is induced by sulfate starvation. Its onset can be easily controlled via the initial concentration of sulfate in the medium. We validated the decoupling behavior of sulfate and assessed pDNA quality attributes (supercoiled pDNA content) in E. coli with lab-scale bioreactor cultivations. The results showed an increase in supercoiled pDNA to biomass yield by 33% and an increase of supercoiled pDNA volumetric productivity by 13 % upon limitation of sulfate. In conclusion, even for routinely manufactured biotechnological products such as pDNA, simple changes in the growth medium can significantly improve the yield and quality. Graphical Abstract

Logistic PCA explains differences between genome-scale metabolic models in terms of metabolic pathways

“…A multinomial logistic regression (MLR) model, using the multinom function ("nnet", v7. [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19], was applied to calculate the contribution of specific reactions to pre-defined clusters of GSMMs (i.e. phylogenetic clades/tissue types).…”

“…GSMMs are mathematical representations of species-or context-specific metabolic reaction networks [1] and have been successfully applied to study diseases [2][3][4], optimize bioprocesses [5][6][7][8], or investigate metabolic differences across species [9][10][11]. To compare different GSMMs, dimensionality reduction techniques can be applied to the results of simulations, such as flux balance analysis [12].…”