Psicose is a rare, non-toxic sugar and a natural sweetener present in nature in low amounts, with incredible properties and seemingly with the ability to prevent diabetes.
In the USA and Japan, Psicose is produced industrially using chemistry. We initially produced it chemically in order to assess the disadvantages of this type of production and to determine the advantages of production by micro-organisms. We tried synthesis by epimerisation of D-Fructose, an epimere in C3 of D-Psicose. NMR analysis allowed us to conclude that the chemical way is not optimal and was too expensive for large scale production.
The first part of our project is the bioproduction of D-Psicose by epimerization of D-Fructose in E. coli. After a literature search we found D-Psicose 3-epimerases (Dpe) of different strains that are able to convert fructose into psicose : two from Agrobacterium tumefaciens (UniProt ID Q7CVR7 and A9CH28), one from Clostrium cellulolyticum (UniProt ID B8I944), one from Flavonifractor plautii (UniProt ID G9YVF8) and one from Pseudomonas cichorii (UniProt ID O50580). The next step was the cloning of the epimerases in pSB1C3 in order to set up the optimal conditions for bioproduction. We readily produced clones of Dpe from C. cellulolyticum with RBS and pTacI promoter (BBa_K2448033) in pSB1C3. We chose to use batch cultures for a whole cell production of D-psicose and we selected BL21-AI as the best strain of E. coli to ensure high yields. We found that the carbon source had no impact on the fitness of the strain. We tested increasing concentrations of fructose in our culture medium and we saw a proportional relationship between the fructose concentration and the psicose measured by HPLC. We achieved a production level of 9 g/L of D-psicose from 50 g/L of D-fructose. But if we exceed a concentration of 100 g/L of D-fructose we created an osmotic stress on the culture that lowered production. Finally we saw that the intensity of the induction had no impact on psicose production.
To improve the bioproduction process, we decided to construct a biosensor to screen the best epimerase for the conversion of Fructose into Psicose and to be able to quantify production with a proportional emission of fluorescence. For this purpose, the biosensor has to be able to detect Psicose in the media, but not the Fructose. The most appropriate type of biosensor for our situation is transcription factor based. These types of biosensors are widely used in screening systems. After research in literature and databases, one transcription factor emerged: PsiR which is a predicted transcription factor of the LacI family with affinity for D-Psicose.
In order to standardize our system we created a Universal Biosensor Chassis (BBa_K2448023 and BBa_K2448024), this system allows us to build several biosensors and to test large numbers of different epimerases. The UBC is composed of a promoter inducible by IPTG, pTacI, two insertion markers: mEmerald for the transcription factor and LacZ-alpha for the promoter regulated by the transcription factor. The final assembly is done by Golden Gate, an efficient and fast method which allowed us to quickly build eight biosensors. After that we proceeded to characterization, we determined the specific dynamic range and linearity of each biosensor and then identified the most efficient biosensor for our screening process.
For the screening process, we first added, using Golden Gate, a Mutant Drop Zone (MDZ) to each biosensor for the insertion of the epimerase mutants. The first step in this process was the creation of the bank of mutants, perfomed by Error Prone PCR on Dpe of C. cellulolyticum. Then each of the mutants were inserted into the MDZ by Golden Gate Assembly. For the screening of the bank, we measured the fluorescence in the presence of 50 g/L of the substrate, fructose. We have been able to screen about 400 mutants and thanks to the fluorescence data we identified two enzymes with a potential increased activity.