IGEM Evry Paris-Saclay

Chemistry

Production of Psicose by Chemical Epimerization of Fructose

Before any biological attempt, we tried to synthesize the rare sugar Psicose by chemical epimerization. This epimerization only works with ketose sugars. We have identified one sugar that can be used in the synthesis, D-Fructose, which is an epimere in C3 of D-Psicose. This sugar is abundant and cheap. It is found in honey, tree and vine fruits, flowers, berries, and most root vegetables. Fructose and Psicose exist naturally in 5 forms (anomers) (Figure 1).

After researching publications we found some that described the epimerization process (Figure 2) but nothing about chemical production and we decided to try the conditions described by Doner [1].

Figure 2. Reaction scheme of epimerization of Fructose in Psicose. — **Figure 2.** Reaction scheme of epimerization of Fructose in Psicose.

In this publication, the authors screened some bases to find the best experimental procedure (Table1).

Table 1. Yields (%) from isomerization of D-Fructose with various bases [1].

Base	Psicose	Fructose	Mannose	Glucose	Total
Sodium hydroxide	6.0	26.2	3.6	21.4	57.2
Calcium hydroxide	5.7	26.2	2.4	17.1	51.4
N,N-Dicyclohexylcarbodiimide	8.4	14.4	2.0	8.8	33.6
Pyridine	12.4	25.8	5.1	6.4	49.7
Methanolic triethylamine	10.6	60.5	2.2	12.4	85.7
Aluminate resin	3.7	10.1	1.0	9.0	23.8

We applied the conditions described in the literature [1] for the epimersation of D-Fructose into D-Psicose:

The first condition: with a Triethylamine like base, D-Fructose (5.00 g) was stirred in both 5% aqueous and methanolic triethylamine (20 mL) at 60°C for 18 h and evaporated to a syrup. The crude product was analyzed by NMR spectroscopy 360 MHz in D₂O.
The second condition: with a Pyridine like base, after early recognition that this is the most effective and convenient base, it was examined under various reaction-conditions. The same isolation procedure was used, regardless of the experiment. At the completion of a given reaction, pyridine was removed from the concentration with the aid of toluene.

Preliminary Experiments

We decided firstly to take the NMR spectra of Mannose, Glucose, Fructose and Psicose to have some references. The results are presented in Figures 3 to 7.

**Figure 3.** ¹³C NMR spectrum of commercial D-Psicose.

**Figure 4.** ¹H NMR spectrum of commercial D-Psicose.

**Figure 5.** ¹H NMR spectrum of commercial D-Fructose.

**Figure 6.** ¹H NMR spectrum of commercial D-Glucose.

**Figure 7.** ¹H NMR spectrum of commercial D-Mannose.

The NMR spectrum showed characteristic peaks for each of the sugars (Table 2). They will be necessary to analyze the NMR spectrum of the crude product of our synthesis experiments.

Table 2. Characteristic peaks of NMR spectra of Psicose, Fructose, Mannose and Glucose.

	δ₁ (ppm)	δ₂ (ppm)	δ₃ (ppm)	δ₄ (ppm)
D-Psicose	4.01	3.49	-	-
D-Fructose	3.97	-	-	-
D-Mannose	5.03	-	-	-
D-Glucose	4.50	3.10	5.01	-

Results

After reactions in the above described conditions and analysis by NMR (Figure 8) we observed the formation of Psicose, in trace amounts, and also the production another sugar similar to Glucose and Mannose. We were unable to estimate the quantity of Psicose in our batch because its peak is overlapped by other peaks.

**Figure 8.** NMR spectrum of the crude product of our chemical epimerisation experiment.

Conclusion

D-Psicose can be produced by chemical epimerization of D-Fructose in trace amounts that cannot be correctly estimated by RMN analysis because many characteristic peaks of different sugar are superimposed.

Our conclusion mets the one of Doner [1]: the bio-production is the solution to synthesize Psicose.

References

[1] Doner WL. Isomerization of d-fructose by base: Liquid-chromatographic evaluation and the isolation of d-psicose. Carbohydrate Res (1979) 70, 209-216.

Acknowledgements

Karen Becherel¹, Dr. Aurélien Alix², Dr. Geneviève Estenne-Bouhtou¹, Dr. Pierre Alexandre Driguez¹, Eric Nicolaï¹, Dr. Alain Fournier¹ and Prof. David Bonnaffé²

Team:Evry Paris-Saclay/Chemistry