Transcriptional network analysis identifies key elements governing the recombinant protein production provoked reprogramming of carbon and energy metabolism in Escherichia coli BL21 (DE3)

• Main Authors: Zhaopeng Li, Robert Geffers, Garima Jain, Frank Klawonn, Öznur Kökpinar, Manfred Nimtz, Wolfgang Schmidt‐Heck, Ursula Rinas
• Format: Article
• Language: English
• Published: Wiley 2021-09-01
• Series: Engineering Reports
• Subjects: Escherichia coli; metabolic burden; recombinant protein production; regulatory network analysis
• Online Access: https://doi.org/10.1002/eng2.12393

Abstract The impact of recombinant protein production on carbon and energy metabolism in Escherichia coli BL21 (DE3) was studied through transcriptome and proteome analysis of cells induced in carbon‐limited fed‐batch cultures during either fast or slow growth. Production of human basic fibroblast growth factor (pET expression system, T7 promoter) during fast growth leads to a macroscopically observable response classifiable into two consecutive steps: i. apparently unperturbed growth and respiration with concomitant formation of pyruvate and acetate followed by ii. inhibition of growth, respiratory activity and glucose uptake. Down‐regulation of genes involved in sugar and acetate uptake, tricarboxylic acid (TCA) cycle, and respiratory energy generation started already during apparently unperturbed growth with the exceptions of up‐regulated genes encoding the less energy efficient NADH dehydrogenase and terminal oxidases. A transcription factor target gene network analysis revealed that observed changes are mainly attributable to the vanishing influence of the transcription factor CRP‐cAMP but also to a strong down‐regulation of AcrA‐P repressed genes. Moreover, down‐regulation of MalT activated and up‐regulation of PdhR repressed genes contribute among others to the reorganization of the transcriptome. The main drivers were identified as accumulating metabolites, for example, pyruvate, which affect transcription factor activity. The resulting restructured proteome leads to reduced glucose uptake, TCA cycle, and respiratory capacities this way decreasing catabolic carbon breakdown and metabolite accumulation. At slow growth, the production provoked transcriptome rearrangements are more subtle not leading to a macroscopically evident response. In summary, the transcriptomic response towards recombinant gene expression mimics a carbon or nutrient up‐shift response aiming to match catabolic carbon processing with compromised anabolic capacities of induced cells. It is not the reason for growth inhibition and the metabolic burden but the cellular attempt to attenuate the “toxic effect” of recombinant gene expression by reducing carbon catabolism.