Enzymatic procedures in the preparation of regioprotected D-fructose derivatives

Article abstract of DOI:10.1016/j.carres.2004.11.003

A combination of different lipases from Pseudomonas cepacia, Candida antarctica B, Candida rugosa and Mucor miehei, aided the regioesterification of the free fructose allowing the synthesis of 1,6-di-O-acetyl-d-fructofuranose, 1,4,6-tri-O-acetyl-d-fructofuranose, 1,6-di-O-acetyl-4-O-benzoyl-d- fructofuranose and 1,6-di-O-benzoyl-d-fructofuranose. Using C. antarctica B and C. rugosa lipases the alcoholysis of fructose peracetate (α, β-form) has furnished 1,2,3,4-tetra-O-acetyl-α-d-fructofuranose and 2,3,4,6-tetra-O-acetyl-β-d-fructofuranose. 1,4,6-Tri-O-acetyl-d- fructofuranose was successfully employed to produce a rare ketohexose, namely d-psicose.

Full text of DOI:10.1016/j.carres.2004.11.003

Carbohydrate
RESEARCH
Carbohydrate Research 340 (2005) 319–323
Note
Enzymatic procedures in the preparation of regioprotected
D-fructose derivativesq
Nicola DÕAntona,a Mostafa El-Idrissi,b Najim Ittobaneb and Giovanni Nicolosia,*
aCNR Istituto Chimica Biomolecolare, Sezione CT, via del santuario 110, 95028 Valverde CT, Italy
b
´ ´ `
Faculte des Sciences, Universite MoulayIsmaı¨l, BP 4010 Beni M’hamed, Meknes, Morocco
Received 20 May 2004; received in revised form 5 November 2004; accepted 6 November 2004
Available online 18 December 2004
Abstract—A combination of different lipases from Pseudomonas cepacia, Candida antarctica B, Candida rugosa and Mucor miehei,
aided the regioesterification of the free fructose allowing the synthesis of 1,6-di-O-acetyl-D-fructofuranose, 1,4,6-tri-O-acetyl-D-fruc-
tofuranose, 1,6-di-O-acetyl-4-O-benzoyl-D-fructofuranose and 1,6-di-O-benzoyl-D-fructofuranose. Using C. antarctica B and
C. rugosa lipases the alcoholysis of fructose peracetate (a, b-form) has furnished 1,2,3,4-tetra-O-acetyl-a-D-fructofuranose and
2,3,4,6-tetra-O-acetyl-b-D-fructofuranose. 1,4,6-Tri-O-acetyl-D-fructofuranose was successfully employed to produce a rare keto-
hexose, namely D-psicose.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Keywords: Lipase; Fructose; Regioprotection; Biocatalysis
Biocatalysis of hydrolases in organic solvents is today a
well-known area in organic synthesis for the preparation
of regioprotected polyfunctionalised molecules.1 Sugars
have received particular attention in this context because
of their structural features, stereochemistry and biologi-
cal properties. Using lipases (EC 3.1.1.3) different selec-
tive modifications have been achieved on free sugars,2
aminosugars,3 sugar component of natural compounds4
and the sugar moiety of nucleosides.5 Although there are
many examples of the use of hydrolases in the regiopro-
tection of compounds containing fructose as a structural
moiety,6 there are few studies on biocatalysed protec-
tion–deprotection of pure fructose. These studies report
the differentiation between the hydroxyl groups of fruc-
tose by esterification in the presence of porcine pancre-
atic lipase, as well as protease subtilisin, but only
mixtures of 1-O-acyl-D-fructose and 6-O-acyl-D-fructose
were obtained.7 Furthermore to prepare surfactants and
food additives, Pseudomonas sp. lipase and Candida ant-
arctica lipase have been exploited for the synthesis of
fatty acid fructose mono and diesters. In all the cases
only the primary hydroxyl functions were recognised,
as in previous studies.8
The aim of the present studies was to further develop
a route towards the synthesis of differently regiopro-
tected sugar derivatives, to establish the potential of
hydrolases operating in organic solvent. We report the
results obtained using four different lipases in esterifica-
tion and alcoholysis processes of D-fructose 1 and perac-
etate 6. One of the regioprotected derivatives obtained
has been used as a starting material to obtain the rare
fructose isomer D-psicose.
At the beginning of our investigation, different lipases
were tested for the esterification of D-fructose in tetrahy-
drofuran (THF) using vinyl acetate as an acyl donor. In
the conditions tested, lipase from C. antarctica B (CAL-
B) was able to catalyse in 2h the esterification of 1 at pri-
mary hydroxyl groups giving 1,6-di-O-acetyl-D-fructo-
furanose (2) as the sole product in quantitative yields.
No further esterification was observed on prolonging
the reaction time, neither was any further reaction ob-
served after addition of a fresh amount of catalyst. Most
q Presented at the 6th International Symposium on Biocatalysis and
Biotransformation, Olomouc, Czech Republic (June 28–July 3,
2003).
*
Corresponding author. Tel.: +39 0957212136; fax: +39 0957212141;
e-mail: giovanni.nicolosi@icb.cnr.it
0008-6215/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.11.003

320
N. DÕAntona et al. / Carbohydrate Research 340 (2005) 319–323
likely the stereochemistry hindered the recognition of
CAL-B of hydroxyl groups at C-3 and C-4 such as at the
anomeric position. Esterification experiments in the
presence of lipase from Pseudomomas cepacia (PSL),
Mucor miehei (MML) and Candida rugosa (CRL) were
unsuccessful, obtaining no or poor reactivity.
Regiodiscrimination of secondary hydroxyl groups
has been attempted by exploiting 2 as starting material;
this compound has the advantage of being constrained
in furanosic forms, so consequently lowering the num-
ber of possible isomers present and giving a well-
behaved solubility in apolar solvents.
The four lipases were screened for the esterification of
2 in tert-butylmethylether (TBME), and in presence of
vinyl acetate as acyl donor. Under these conditions,
CRL gave excellent recognition of the hydroxyl group
at C-4, affording 1,4,6-tri-O-acetyl-D-fructofuranose (3)
as the sole product. An analogous catalytic action was
evidenced for PSL, but associated with a very low
conversion. MML and CAL left the substrate 2 un-
transformed. The real action, a transesterification at
primary hydroxyl groups exclusively, of these last two
lipases was revealed when esterification using vinyl ben-
zoate as acyl donor was tried: catalysis in the presence of
CRL or PSL gave 1,6-di-O-acetyl-4-O-benzoyl-D-fructo-
furanose 4 as expected; conversely ester 5 was the prod-
uct obtained under the catalysis of MML and CAL.
For derivatives bearing unprotected primary hydroxyl
functions, the alcoholysis of D-fructose peracetate 6 was
examined. The preparation of this pentacetate in a and b
furanose form, was obtained by subjecting 2, prepared
by the use of CAL-B as previously described, to treat-
ment with acetic anhydride and pyridine. 1H NMR analy-
sis of the obtained product 6 indicated a a/b ratio of
1:1. The anomeric mixture of compound 6 was submit-
ted to alcoholysis in the presence of CAL-B, with buta-
nol as the nucleophilic agent, and TBME as solvent. The
reaction proved highly regiospecific and stereoselective,
giving differential CAL-B recognition of a/b anomers
and affording 1,2,3,4-tetra-O-acetyl-a-D-fructofuranose
(7) and 2,3,4,6-tetra-O-acetyl-b-D-fructofuranose (8)
(Scheme 1).
A different outcome was observed on alcoholysis
employing CRL as catalyst, with recognition of the ester
group at C-1b exclusively, to give 2,3,4,6-tetra-O-acetyl-
b-D-fructofuranose (8): the a-anomer remained as an
unreacted substrate.
These data proved the synthetic potential of ester 3 in
the preparation of D-psicose, 9. Triester (3) was sub-
jected to inversion of configuration at C-3 in Mitsunobu
conditions, furnishing the corresponding 1,4,6-tri-O-
acetyl-3-O-benzoyl-D-psicose (10) that gave the desired
free D-psicose (9) in 80% overall yield (Scheme 2).
In conclusion, we have developed a set of reactions
catalysed by different lipases that allow access to differ-
ent fructose derivatives. Lipase from C. antarctica B has
high recognition for primary hydroxyl groups in the
esterification. Conversely, in the reverse alcoholysis pro-
cess of peracetate D-fructose, the configuration at the
anomeric centre strongly influenced the regioselectivity
of this biocatalyst. Lipase from C. rugosa showed pref-
erence for esterification at C-4, whilst able to affect the
AcO
O
OAc
OH
AcO
OH
3
HO
O
OAc
CRL
AcV-TBME
OAc
OAc
AcO
AcO
HO
AcO
AcO
O
OH
O
OAc
O
OAc
α-7
CAL-B
CAL-B
Ac2O/Py
OH
OH
OAc
OAc
BuOH-TBME
AcV-THF
HO
HO
OH
OH
AcO
AcO
O
OH
1
2
α,β -6
OAc
OAc
CRL
β-8
CRL or PSL
CAL-B or MML
BuOH-TBME
BzV-TBME
AcO
AcO
BzO
O
OAc
O
OAc
O
OBz
β-8
+
OAc
OAc
OH
OH
AcO
BzO
HO
OH
OH
4
5
α-6
AcV = Vinyl Acetale BzV = Vinyl Benzoate
Scheme 1.

N. DÕAntona et al. / Carbohydrate Research 340 (2005) 319–323
321
AcO
AcO
HO
OAc
O
O
OAc
O
OH
MeONa/MeOH
OH
OH
Mitsunobu
Conditions
OH
AcO
AcO
HO
OH
OBz
OH
3
10
9
Scheme 2.
ester function at C-1 only in the alcoholysis of perace-
tate ins b-anomeric form. The obtained fructose deriva-
tives are suitable starting materials for further trans-
formation and synthesis of valuable rare ketohexose
sugars.
drawn at regular time intervals and monitored by TLC
(Et2O–MeOH, 99:1). After 2h the reaction was
quenched, filtering off the catalyst and the filtrate evap-
orated to dryness in vacuum to give compound 2 (54mg,
95%) as a slightly yellow syrup. No further purification
was necessary. Anomeric ratio 1:2; Rf = 0.4 (Et2O–
1
MeOH, 99:1); H NMR (400MHz, CDCl3) major ano-
1. Experimental
mer: 4.32 (d, 1H, J3,4 2.1Hz, H-3), 4.29 (br d, 1H, J4,5
3.3Hz, H-4), 4.23 (d, 1H, J1a,1b 7.2Hz, H-1a), 4.12 (d,
1H, H-1b), 4.03 (d, 2H, J6,5 7.2Hz, H-6), 3.98 (dt, 1H,
H-5), 2.13 (s, 3H, COOCH3), 2.11 (s, 3H, COOCH3);
minor anomer: 4.29 (br d, 1H, H-4) 4.21–4.13 (m, 3H,
H-3, H-1a and H-1b), 4.07 (br d, 1H, H-6), 3.98 (dt,
1H, H-5), 2.15 (s, 3H, COOCH3), 2.10 (s, 3H,
COOCH3). ESI-MS negative mode CV 5 eV (MeOH+
LiCl 0.001M) m/z: 299.2 (M þ Clꢀ35, 100%), 301.2
1.1. General methods
1H NMR spectra were recorded in CDCl3 or DMSO
solution at 400.13MHz, on a Bruker AMX-400 instru-
ment using TMS as internal reference. The coupling
constants (J) are reported in hertz and the chemical
shifts (d) in parts per million downfield from TMS.
The different products obtained by chemical or enzy-
matic routes were characterised by 2D 1H NMR experi-
ments. Optical rotations were measured on a DIP 135
JASCO instrument. Mass spectra were recorded on a
ESI-MS spectrometer Waters-Micromass ZQ2000 dis-
solving compounds in a 0.001M methanolic solution
of LiCl: in positive mode adducts ions with Li+ were ob-
served, in negative mode adducts ions with Clꢀ were ob-
served. GC analysis was performed using a Shimatsu
17A gas chromatograph equipped with a Zebron ZB-5
column; temp progr.: 100–300ꢁC 10ꢁC/min, inj.
180ꢁC, det. 220ꢁC. Elemental analyses were carried
out on a Perkin–Elmer 240 elemental analyser. Lipase
from P. cepacia was obtained from Amano international
Enzyme. C. rugosa lipase was from Sigma (St. Louis,
MO). Novozym 435 (immobilised lipase from C. antarc-
tica B) and Lipozyme IM (immobilised lipase from
Mucor miehei) were a kind gift of Novo Nordisk. Flash
chromatography was performed on 60mm mesh silica
gel (Merck, Darmstadt, Germany); analytical TLC (sil-
ica gel 60-F254 precoated plates) and preparative TLC
(Kieselgel F254, 0.5 and 1.0mm thickness) materials
were furnished by Merck and compounds were visual-
ised by spraying with molybdophosphoric acid.
(M þ Clꢀ , 32.6%). Anal. Calcd for C10H16O8: C,
37
45.46; H, 6.10. Found: C, 45.66; H, 6.06.
1.3. General procedure for the synthesis of 1,4,6-tri-O-
acyl-D-fructofuranose (3–4)
Lipase of choice (from C. rugosa or P. cepacia, 40mg)
was added to a solution of 2 (40mg) in tert-butylmeth-
ylether (4mL) containing appropriate vinyl ester
(3equiv). The suspension was shaken (300rpm) at
45ꢁC, aliquots were drawn at regular time intervals
and monitored by TLC (100% Et2O). After 7h the reac-
tion was quenched, filtering off the catalyst and the fil-
trate evaporated to dryness in vacuum. The residue
was purified by flash chromatography on silica gel
(petroleum ether/diethyl ether) or by PTLC (100%
Et2O).
1.3.1. 1,4,6-Tri-O-acetyl-D-fructofuranose (3). Yield
41mg (90%) obtained as a slight yellow syrup from com-
pound 2 and vinyl acetate, as described in the general
procedure: anomeric ratio 1:4; Rf = 0.52 (Et2O); 1H
NMR (400MHz, CDCl3) major anomer: 5.01 (t, 1H,
J4,5 and J4,3 5.2Hz, H-4), 4.31 (d, 1H, H-3), 4.17 (d,
1H, J1a,1b 10.8Hz, H-1a), 4.11 (d, 2 H, J6,5 7.1Hz, H-
6), 4.16–4.09 (m, 2H, H-5 and H-1b), 2.14 (s, 3H,
COOCH3), 2.12 (s, 3H, COOCH3), 2.10 (s, 3H,
COOCH3); minor anomer: 4.77 (br dd, 1H, H-4),
4.30–4.20 (m, 2H, H-3 and H-1a), 4.11 (br d, 2H, H-
6), 4.16–4.09 (m, 1H, H-5), 4.01 (m, 1 H, H-1b), 2.14
(s, 3H, COOCH3), 2.12 (s, 3H, COOCH3), 2.10 (s, 3H,
1.2. 1,6-Di-O-acetyl-D-fructofuranose (2)
Lipase from C. antarctica B (40mg) was added to a solu-
tion of D-fructose (1) (40mg) in tetrahydrofuran (4mL)
containing vinyl acetate (3equiv) as acyl donor. The sus-
pension was shaken (300rpm) at 45ꢁC, aliquots were

322
N. DÕAntona et al. / Carbohydrate Research 340 (2005) 319–323
COOCH3). ESI-MS negative mode CV 5eV (MeOH+
LiCl 0.001M) m/z: 341.4 (M þ Clꢀ35, 100%), 343.4
7 and 8 (see below) was confirmed by NMR signals.
1H NMR (400MHz, CDCl3) a anomer: 5.85 (d, 1H,
J3,4 4.1Hz, H-3), 5.14 (dd, 1H, J4,5 6.3Hz, H-4), 4.62
(d, 1H, J1a,1b 12.0Hz, H-1a), 4.51 (ddd, 1H, J5,6a 3.6,
J5,6b 5.7, H-5), 4.40 (dd, 1H, J6a,6b 12.2Hz, H-6a), 4.35
(d, 1H, H-1b), 4.17 (dd, 1H, H-6b), 2.16 (s, 3H,
COOCH3), 2.11 (s, 3H, COOCH3), 2.10 (s, 6H,
2 · COOCH3), 2.07 (s, 3H, COOCH3); b anomer: 5.69
(dd, 1H, J4,3 2.2, J4,5 9.0Hz, H-4), 5.50 (d, 1H, H-3),
5.22 (ddd, 1H, J5,6a 2.7, J5,6b 4.9Hz, H-5), 4.92 (d, 1H,
J1a,1b 17.3Hz, H-1a), 4.68 (d, 1H, H-1b), 4.29 (dd, 1H,
J6a,6b 12.6Hz, H-6a), 4.12 (dd, 1H, H-6b), 2.20 (s, 3H,
COOCH3), 2.11 (s, 3H, COOCH3), 2.10 (s, 6H,
2 · COOCH3), 2.08 (s, 3H, COOCH3). ESI-MS
positive mode CV 5eV (MeOH+LiCl 0.001M) m/z:
391.4 (M+H+, 8%), 331.4 (MꢀCH3COOꢀ, 100%),
271.4 (MꢀCH3COOꢀꢀCH3COOH, 3%), 211.4 (Mꢀ
CH3COOꢀꢀ2 · CH3COOH, 3%). Anal. Calcd for
C16H22O11: C, 49.23; H, 5.68. Found: C, 49.55; H, 5.71.
(M þ Clꢀ , 32.6%). Anal. Calcd for C12H18O9: C,
37
47.06; H, 5.92. Found: C, 47.25; H, 5.87.
1.3.2. 1,6-Di-O-acetyl-4-O-benzoyl-D-fructofuranose (4).
Yield 50mg (90%) obtained as a slightly yellow syrup
from compound 2 and vinyl benzoate, as described in
the general procedure: anomeric ratio 1:4; Rf = 0.60
(Et2O); 1H NMR (400MHz, CDCl3) major anomer:
8.04 (d, 2H, J2 ,3 7.2Hz, H-20), 7.60 (d, 1H, J4 ,3
0
0
0
0
7.5Hz, H-40), 7.47 (t, 2H, H-30), 5.24 (t, 1H, J4,5 and
J4,3 5.4Hz, H-4), 4.44–4.41 (m, 1H, H-5), 4.40 (d, 1H,
H-3), 4.33 (d, 2H, J6,5 5.6Hz, H-6), 4.25 (d, 1H, J1a,1b
11.8Hz, H-1a), 4.18 (d, 1H, H-1b), 2.12 (s, 3H,
COOCH3), 2.10 (s, 3H, COOCH3); minor anomer:
8.04 (d, 2H, J2 ,3 7.2Hz, H-20), 7.60 (d, 1H, J4 ,3
0
0
0
0
7.5Hz, H-40), 7.47 (t, 2H, H-30), 5.02 (br dd, 1H, H-4),
4.47-4.46 (m, 2H, H-3 and H-5), 4.33 (br d, 2H, H-6),
4.30 (br dd, 2H, H-1), 2.11 (s, 3H, COOCH3), 2.10 (s,
3H, COOCH3). ESI-MS negative mode CV 5eV
(MeOH+LiCl 0.001M) m/z: 403.3 (M þ Clꢀ35, 100%),
1.6. General procedure for the alcoholysis of D-fructose
pentacetate (6)
405.3 (M þ Clꢀ , 32.6%). Anal. Calcd for C17H20O9: C,
37
55.43; H, 5.47. Found: C, 55.15; H, 5.49.
Lipase of choice (from C. antarctica B or C. rugosa,
40mg) was added to a solution of 6 (40mg) in tert-butyl-
methylether (4mL) containing butanol (3equiv) as
nucleophilic agent. The suspension was shaken
(300rpm) at 45ꢁC, aliquots were drawn at regular time
intervals and monitored by TLC (100% Et2O). After
7h the reaction was quenched, filtering off the catalyst
and the filtrate evaporated to dryness in vacuum. The
residue was purified by flash chromatography on silica
gel (petroleum ether/diethyl ether) or by PTLC (100%
Et2O).
1.4. 1,6-Di-O-benzoyl-D-fructofuranose (5)
Lipase of choice (from C. antarctica B or M. miehei,
40mg) was added to a solution of 2 (40mg) in tert-butyl-
methylether (4mL) containing vinyl benzoate (3equiv).
The suspension was shaken (300rpm) at 45ꢁC, aliquots
were drawn at regular time intervals and monitored by
TLC (100% Et2O). After 7h the reaction was quenched,
filtering off the catalyst and the filtrate evaporated to
dryness in vacuum. The residue was purified by flash
chromatography on silica gel (petroleum ether/diethyl
ether) or by PTLC (100% Et2O) to afford compound 5
(46mg, 80%) as a slightly yellow syrup. Rf = 0.36
(Et2O); 1H NMR (400MHz, CDCl3) major anomer:
8.04 (br s, 4H, H-20), 7.55 (br t, 2H, H-40), 7.42 (br t,
4H, H-30), 4.51 (br d, 1H, J3,4 2.3Hz, H-3), 4.41 (br d,
1 H, H-4), 4.33–4.27 (m, 2H, H-1), 4.20–4.14 (m, 3H,
H-5 and H-6). ESI-MS negative mode CV 5eV
(MeOH+LiCl 0.001M) m/z: 423.3 (M þ Clꢀ35, 100%),
1.6.1. 1,2,3,4-Tetra-O-acetyl-a-D-fructofuranose (7).
Yield 14mg (41%) obtained as a slight yellow syrup
from compound 6 and butanol in presence of C. antarc-
22
tica B lipase, as described in the general procedure: ½aꢁ
D
+22.86 (c 0.385, CHCl3) Rf 0.59 (100% Et2O); 1H NMR
(400MHz, CDCl3): 5.90 (d, 1H, J3,4 4.6Hz, H-3), 5.30
(dd, 1H, J4,5 6.6Hz, H-4), 4.68 (d, 1H, J1a,1b 12.0Hz,
H-1a), 4.39 (m, 1H, H-5), 4.28 (d, 1H, H-1b), 3.89 (dd,
1H, J6a,5 2.8, J6a,6b 12.7Hz, H-6a), 3.71 (dd, 1H, J6b,5
2.8, H-6b), 2.17 (s, 3H, COOCH3), 2.10 (s, 6H,
2 · COOCH3), 2.07 (s, 3H, COOCH3). ESI-MS positive
mode CV 5eV (MeOH+LiCl 0.001M) m/z: 355.4
(M+Li+, 100%). Anal. Calcd for C14H20O10: C, 48.28;
H, 5.79. Found: C, 48.55; H, 5.75.
425.2 (M þ Clꢀ , 32.6%). Anal. Calcd for C20H20O8: C,
37
61.85; H, 5.19. Found: C, 62.11; H, 5.21.
1.5. D-Fructose pentacetate (6)
1,6-Di-O-acetyl-D-fructofuranose 2 (500mg) was dis-
solved in 10mL of pyridine and 10mL of Ac2O. The
solution was stirred at room temperature for 6h and
then evaporated to dryness in vacuo to give compound
6 (724mg, 98%) in a/b 1:1 mixture as a slightly yellow
syrup. Rf = 0.84 (100% Et2O). Chemical synthesis of
pure a- and b-6 using as starting material compounds
1.6.2. 2,3,4,6-Tetra-O-acetyl-b-D-fructofuranose (8).
Yield 12mg (35%) when using C. antarctica B lipase
and 14mg (39%) in presence of C. rugosa lipase, ob-
tained as a slightly yellow syrup from compound 6
and butanol, as described in the general procedure:
22
D
½aꢁ ꢀ220.37 (c 0.362, CHCl3); Rf 0.72 (100% Et2O);

N. DÕAntona et al. / Carbohydrate Research 340 (2005) 319–323
323
1H NMR (400MHz, CDCl3): 5.62 (dd, 1H, J4,3 2.1, J4,5
9.0Hz, H-4), 5.49 (d, 1H, H-3), 5.25 (m, 1H, H-5), 4.37
(d, 2H, J1a,1b 5.4Hz, H-1), 4.29 (dd, 1H, J6a,5 2.5, J6a,6b
12.2Hz, H-6a), 4.14 (dd, 1H, J6b,5 4.5, H-6b), 2.16 (s, 9H,
3 · COOCH3), 2.07 (s, 3 H, COOCH3). ESI-MS positive
mode CV 5eV (MeOH+LsiCl 0.001M) m/z: 355.4
(M+Li+, 100%). Anal. Calcd for C14H20O10: C, 48.28;
H, 5.79. Found: C, 48.49; H, 5.82.
gously silylated; retention times: D-fructose persilylated
15.17, D-psicose persilylated 15.40min.
Acknowledgements
This work has been co-funded by MIUR (Roma) within
the Project ÔMateriali Innovativi—Metodologie e Diag-
nostiche per Materiali ed AmbienteÕ.
1.7. Preparation of D-psicose (9)
References
1.7.1. 1,4,6-Tri-O-acetyl-3-O-benzoyl-D-psicose (10).
To a solution of 1,4,6-tri-O-acetyl-D-fructofuranose 3
(130mg, 0.42mmol), Ph3P (240g, 0.9mmol) and benzoic
acid (111.0mg, 0.9mmol) in toluene (4.5mL) diethyl
azodicarboxylate (0.150mL, 0.9mmol) was added drop-
wise. After being stirred at room temperature for 4h, the
mixture was chromatographed (EtOAc–hexane) to
afford compound 10 (148mg, 85%). Anomeric ratio
1:7.5; Rf 0.35 (Et2O–MeOH = 98:2); 1H NMR
1. Klibanov, A. M. Nature 2001, 409, 241–246; Zacks, A.;
Dodds, D. R. Drug DiscoveryToday 1997, 2(12), 513–531;
Schulze, B.; Wubbolts, M. G. Curr. Opin. Biotechnol. 1999,
10(6), 609–615.
2. Lay, L.; Panza, L.; Riva, S.; Khitri, M.; Tirendi, S.
Carbohydr. Res. 1996, 291, 197–204; Danieli, B.; Luisetti,
M.; Sampognaro, G.; Carrea, G.; Riva, S. J. Mol. Catal. B:
Enzyme 1997, 3, 193–201.
3. Nicolosi, G.; Spatafora, C.; Tringali, C. Tetrahedron:
Asymmetry 1999, 15, 2891–2897.
0
0
(400MHz, CDCl3) major anomer: 8.04 (d, 2H, J2 ,3
7.4Hz, H-20), 7.60 (d, 1H, J4 ,3 7.4Hz, H-40), 7.47 (t,
2H, H-30), 5.08 (dd, 1H, J4,3 2.58, J4,5 4.57Hz, H-4),
4.81 (d, 1H, J1a,1b 12.2Hz, H-1a), 4.73 (d, 1H, H-1b),
4.69 (d, 1H, H-3), 4.47 (m, 1H, H-5), 4.42 (dd, 1H,
J6a,5 3.7, J6a,b 15.6Hz, H-6a), 4.14 (dd, 1H, J6b, 5 5.5,
H-6b), 2.12 (s, 3H, COOCH3), 2.09 (s, 3H, COOCH3),
2.02 (s, 3H, COOCH3). ESI-MS positive mode CV
5eV (MeOH+LiCl 0.001M) m/z: 417.3 (M+Li+,
100%). Anal. Calcd for C19H22O10: C, 55.61; H, 5.40.
Found: C, 55.83; H, 5.38.
4. Riva, S. J. Mol. Catal. B: Enzyme 2002, 19–20, 43–54.
5. Ferrero, M.; Gotor, V. Chem. Rev. 2000, 100, 4319–4347.
6. Potier, P.; Bouchu, A.; Gagnaire, J.; Queneau, Y. Tetra-
hedron: Asymmetry 2001, 12, 2409–2419; Potier, P.; Bou-
chu, A.; Descotes, G.; Queneau, Y. Synthesis 2001, 3, 458;
Khaled, N.; Montet, D.; Pina, M.; Graille, J. Biotechnol.
Lett. 1991, 13, 167; Schlotterbeck, A.; Lang, S.; Wray, V.;
Wagner, F. Biotechnol. Lett. 1993, 15, 61.
7. Therisod, M.; Klibanov, A. M. J. Am. Chem. Soc. 1986,
108, 5638–5640; Riva, S.; Chopineau, J.; Kieboom, A. P.
G.; Klibanov, A. M. J. Am. Chem. Soc. 1988, 110, 584–589;
Carrea, G.; Riva, S.; Secundo, F.; Danieli, B. J. Chem. Soc.,
Perkin Trans. 1 1989, 1057–1061; Sin, Y. M.; Cho, K. W.;
Lee, T. H. Biotechnol. Lett. 1998, 20, 91–94.
0
0
1.7.2. Hydrolysis of 1,4,6-tri-O-acetyl-3-O-benzoyl-D-psi-
cose (10). To a solution of compound 10 (110mg) in
MeOH (3mL) 25wt% sodium methoxide (in MeOH,
three drops) was added. After stirring at room tempera-
ture for 2h, Amberlite IRC 50 was added to the mixture
till neutralisation, filtered and evaporated under reduced
8. Coulon, D.; Girardin, M.; Rovel, B.; Ghoul, M. Biotechnol.
Lett. 1995, 17, 183–186; Scheckermann, C.; Schlotterbeck,
A.; Schmidt, M.; Wray, V.; Lang, S. Enzyme Microb.
Technol. 1995, 17, 157–162; Tsitsimpikou, C.; Daflos, H.;
Kolisis, F. N. J. Mol. Catal. B: Enzyme 1997, 3, 189–192;
`
Arcos, J. A.; Bernabe, M.; Otero, C. Enzyme Microb.
Technol. 1998, 22, 27–35; Coulon, D.; Girardin, M.; Ghoul,
M. Process Biochem. 1999, 34, 913–918; Tsuzuki, W.;
Kitamura, Y.; Suzuki, T.; Kobayashi, S. Biotechnol. Bio-
eng. 1999, 64, 267–271; Chamouleau, F.; Coulon, D.;
Girardin, M.; Ghoul, M. J. Mol. Catal. B: Enzyme 2001,
11, 949–954; Degn, P.; Zimmermann, W. Biotechnol.
Bioeng. 2001, 74, 483–491; Sekeroglu, G.; Fadiloglu, S.;
Ibanoglu, E. J. Sci. Food Agric. 2002, 82, 1516–1522.
9. Kopper, S.; Freimund, S. Helv. Chim. Acta 2003, 86, 827–
843.
1
pressure to obtain pure D-psicose 9 (45.8mg, 95%). H
NMR data results indistinguishable by data obtained
with a commercial sample of D-psicose and from data
reported in literature.9 To obtain another proof of its
nature, compound 9 (10mg) was dissolved in 1mL of
a silylating mixture (HMDS–TMCS–pyridine = 3:1:9)
and compared by gas chromatography to samples of
commercially available D-fructose and D-psicose analo-