Metabolomics Analysis of the Toxic Effects of the Production of Lycopene and Its Precursors

Abstract

Using cells as microbial factories enables highly specific production of chemicals with many advantages over chemical syntheses. A number of exciting new applications of this approach are in the area of precision metabolic engineering, which focuses on improving the specificity of target production. In recent work, we have used precision metabolic engineering to design lycopene-producing Escherichia coli for use as a low-cost diagnostic biosensor. To increase precursor availability and thus the rate of lycopene production, we heterologously expressed the mevalonate pathway. We found that simultaneous induction of these pathways increases lycopene production, but induction of the mevalonate pathway before induction of the lycopene pathway decreases both lycopene production and growth rate. Here, we aim to characterize the metabolic changes the cells may be undergoing during expression of either or both of these heterologous pathways. After establishing an improved method for quenching E. coli for metabolomics analysis, we used two-dimensional gas chromatography coupled to mass spectrometry (GCxGC-MS) to characterize the metabolomic profile of our lycopene-producing strains in growth conditions characteristic of our biosensor application. We found that the metabolic impacts of producing low, non-toxic levels of lycopene are of much smaller magnitude than the typical metabolic changes inherent to batch growth. We then used metabolomics to study differences in metabolism caused by the time of mevalonate pathway induction and the presence of the lycopene biosynthesis genes. We found that overnight induction of the mevalonate pathway was toxic to cells, but that the cells could recover if the lycopene pathway was not also heterologously expressed. The two pathways appeared to have an antagonistic metabolic effect that was clearly reflected in the cells’ metabolic profiles. The metabolites homocysteine and homoserine exhibited particularly interesting behaviors and may be linked to the growth inhibition seen when the mevalonate pathway is induced overnight, suggesting potential future work that may be useful in engineering increased lycopene biosynthesis.

Rewiring of Metabolic Network in Mycobacterium tuberculosis During Adaptation to Different Stresses

“…The data matrix file obtained from LC-MRM/MS was used as input to perform statistical analysis by using online tool MetaboAnalyst 1 (Xia et al, 2012). The original data obtained were normalized to OD 600 equivalent of the number of cells at the time of harvesting post stress (OD s ) (Miguez et al, 2018) Normalization for paired comparison was performed using the following calculation [Normalized metabolite peak area for Control Experiment = (P c /OD c ) X OD s ], where P is peak area. The normalization parameters are given in Supplementary Table S11.…”

The role of sample preparation in multidimensional gas chromatographic separations for non‐targeted analysis with the focus on recent biomedical, food, and plant applications

“…As an example, Miguez et al. used silylation agents to study small polar metabolites from the lycopene and mevalonate pathways of E. coli [70]. Among these, homocysteine and homoserine have been identified as possible growth inhibitors associated with the overnight induction of the mevalonate pathway of E. coli .…”

Metabolic Engineering Escherichia coli for the Production of Lycopene

“…Further, they validated that the intermediate homocysteine could contribute to the growth inhibition and the antagonistic effect between the mevalonate and lycopene pathways, resulting in the homocysteine-induced toxicity in lycopene production. This work indicated that metabolomics would be beneficial to reveal the mechanisms of the metabolite toxicity, and subsequently help to improve the metabolic engineering for the biosynthesis of carotenoid [37].…”