Biosensor-based enzyme engineering approach applied to psicose biosynthesis

Abstract

Bioproduction of chemical compounds is of great interest for modern industries, as it reduces their production costs and ecological impact. With the use of synthetic biology, metabolic engineering and enzyme engineering tools, the yield of production can be improved to reach mass production and cost-effectiveness expectations. In this study, we explore the bioproduction of D-psicose, also known as D-allulose, a rare non-toxic sugar and a sweetener present in nature in low amounts. D-psicose has interesting properties and seemingly the ability to fight against obesity and type 2 diabetes. We developed a biosensor-based enzyme screening approach as a tool for enzyme selection that we benchmarked with the Clostridium cellulolyticum D-psicose 3-epimerase for the production of D-psicose from D-fructose. For this purpose, we constructed and characterized seven psicose responsive biosensors based on previously uncharacterized transcription factors and either their predicted promoters or an engineered promoter. In order to standardize our system, we created the Universal Biosensor Chassis, a construct with a highly modular architecture that allows rapid engineering of any transcription factor-based biosensor. Among the seven biosensors, we chose the one displaying the most linear behavior and the highest increase in fluorescence fold change. Next, we generated a library of D-psicose 3-epimerase mutants by error-prone PCR and screened it using the biosensor to select gain of function enzyme mutants, thus demonstrating the framework's efficiency.

Keywords: Universal Biosensing Chassis; enzyme engineering; psicose; rare sugars; transcription factor-based biosensor.

Figure 1.

Design and characterization of six psicose biosensors. (A) Schematic representation of the UBC used as a platform to build the psicose biosensors (B). (C–H) In vivo characterization of mCherry expression by E. coli cells harboring (C) the psicose biosensor based on pPsiA promoter from A. tumefaciens and the PsiR transcription factor from A. tumefaciens (BBa_K2448025), (D) the psicose biosensor based on pPsiR promoter from A. tumefaciens and the PsiR transcription factor from A. tumefaciens (BBa_K2448026), (E) the psicose biosensor based on pPsiA promoter from S. fredii and the PsiR transcription factor from S. fredii (BBa_K2448028), (F) the psicose biosensor based on pPsiR promoter from S. fredii and the PsiR transcription factor from S. fredii (BBa_K2448029), (G) the psicose biosensor based on pPsiA promoter from S. meliloti and the PsiR transcription factor from S. meliloti (BBa_K2448030), (H) the psicose biosensor based on pPsiR promoter from S. meliloti and the PsiR transcription factor from S. meliloti (BBa_K2448031). Fluorescence values (background subtracted) were normalized by OD_600nm. The data and error bars are the mean and standard deviation of six measurements (three biological replicates, each measured as two technical duplicates).

Figure 2.

Design and characterization of a synthetic psicose biosensor. (A) Sequence comparison between the pTacI promoter (43) and the pPsiTacI synthetic promoter. (B) In vivo characterization of mCherry expression by E. coli cells harboring the psicose biosensor based on pPsiTacI synthetic promoter and the PsiR transcription factor from A. tumefaciens (BBa_K2448027). Fluorescence values (background subtracted) were normalized by OD_600nm. The data and error bars are the mean and standard deviation of six measurements (three biological replicates, each measured as two technical duplicates).

Figure 3.

DPEase mutant library screening. (A) Schematic representation of the psicose biosensor based on pPsiA promoter from A. tumefaciens and the PsiR transcription factor from A. tumefaciens with downstream the MDZ (BBa_K2448057). (B) The DPEase from C. cellulolyticum or a DPEase mutant library generated by error-prone PCR were inserted in the MDZ of (A) by Golden Gate cloning using the BsmBI restriction endonuclease (BBa_K2448058). (C) In vivo characterization of mEmerald expression as reporter gene of the psicose biosensor represented schematically in (A). Fluorescence values (background subtracted) were normalized by OD_600nm. The data and error bars are the mean and standard deviation of six measurements (three biological replicates, each measured as two technical duplicates). (D) FACS of E. coli cells harboring the psicose biosensor with a downstream DPEase library represented schematically in (B). Cells having the fluorescence/size ratio above average (dotted line) were isolated (regions R1, R2, R3). (E) In vivo characterization of mEmerald expression by E. coli cells harboring the psicose biosensor and ten DPEase mutants represented schematically in (B). All the data points are fluorescence values (background subtracted) normalized by OD_600nm of each mutant normalized by the same value from the control (wild-type DPEase). The data and error bars are the mean and standard deviation of three measurements.