1,5-Anhydro-d-fructose from d-fructose

Article abstract of DOI:10.1016/j.carres.2007.02.026

1,5-Anhydro-d-fructose was efficiently prepared from d-fructose via regiospecific 1,5-anhydro ring formation of 2,3-O-isopropylidene-1-O-methyl(tolyl)sulfonyl-d-fructopyranose and subsequent deprotection.

Full text of DOI:10.1016/j.carres.2007.02.026

Carbohydrate Research 342 (2007) 1249–1253
Note
1,5-Anhydro-D-fructose from D-fructose
Gyula Dekany,b Inge Lundt,c,* Fabian Niedermair,a Sabine Bichler,a Josef Spreitz,d
Friedrich K. (Fitz) Sprengerd and Arnold E. Stutza,*
¨
aGlycogroup, Institut fu¨r Organische Chemie der Technischen Universita¨t Graz, Stremayrgasse 16, A-8010 Graz, Austria
bGlycom ApS, Technical University of Denmark (DTU), Building 201, DK-2800 Kgs. Lyngby, Denmark
cDepartment of Chemistry, Technical University of Denmark (DTU), Building 201, DK-2800 Kgs. Lyngby, Denmark
dAglycon, Stremayrgasse 16, A-8010 Graz, Austria
Received 4 December 2006; received in revised form 21 February 2007; accepted 22 February 2007
Available online 28 February 2007
Abstract—1,5-Anhydro-D-fructose was efficiently prepared from D-fructose via regiospecific 1,5-anhydro ring formation of
2,3-O-isopropylidene-1-O-methyl(tolyl)sulfonyl-D-fructopyranose and subsequent deprotection.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Keywords: 1,5-Anhydro-D-fructose; Synthesis
1,5-Anhydro-D-fructose, AF, (1) is a versatile naturally
occurring monosaccharide and chiral building block
featuring interesting chemical as well as biological
properties.1 Due to its structural features, synthetic
approaches have been met with limited success. Com-
pound 1 has not been available until 1980 despite the
fact that its protected derivatives such as 2-acyloxy-
glycals 2 have been known (Scheme 1),2 but their
attempted direct conversion into compound 1 by base-
catalysed deacylation3–6 gave a complex mixture of
(degradation) products.6
yield by Lichtenthaler and co-workers from 1,5-
anhydro-2,3,4,6-tetra-O-benzoyl-D-arabino-hex-1-enitol
3 (Scheme 2) in 1980.7
No direct structural proof of the obtained product
was given, but dimeric forms were mentioned.7
In 1988, a different synthetic route to AF was pat-
ented, which was based upon the oxidation of the C-2
hydroxyl function in a partially protected 1,5-anhydro-
D-glucitol derivative 11, available from 9 by standard
procedures (Scheme 3).8
In 1988, a glucan lyase which degrades starch and
related oligosaccharides from the non-reducing end of
the oligosaccharides to give 1,5-anhydro-D-fructose
was identified.9 Since, several other lyases isolated from
different sources have also been found to catalyse this
process.1
In 1993, a pyranose 2-oxidase catalysed oxidation of
1,5-anhydro-D-glucitol 9 was reported10 and in 1998 a
process furnishing 1 on a preparative scale in 98% yield
based on this method was published.11
Employing various experimental detours, 1,5-anhy-
dro-D-fructose (1) was finally synthesised in low overall
OAc
O
OH
O
base
OAc
OH
X
AcO
HO
OAc
O
Exploiting Baute’s9 enzymatic method, AF was pre-
pared from starch in 40–50% yield in a process which
has attracted commercial interest.12–14 The enzyme has
also been cloned and expressed in Aspergillus niger pro-
viding a basis for a more efficient and industrially com-
patible production strategy.15,16
2
1
Scheme 1.
*
Corresponding authors. Fax: +43 316 873 8740 (A.E.S.); e-mail:
stuetz@tugraz.at
0008-6215/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2007.02.026

1250
G. Dekany et al. / Carbohydrate Research 342 (2007) 1249–1253
OBz
O
OBz
O
OH
O
OBz
OBz
OH
1
BzO
BzO
HO
OBz
NOH
NOH
3
4
5
OBz
OBz
O
OH
O
OBz
O
OBz
OH
SEt
BzO
SEt
BzO
HO
O
SEt
SEt
6
7
8
Scheme 2.
OH
O
O
O
O
O
O
O
OH
OH
OR
OR
HO
O
O
O
O
OH
OH
OH
9
10
11
12
OH
O
OHHO
1
+
HO
OH
1 hydrate
Scheme 3.
1,5-Anhydro-D-fructose has never been reported in a
pure monomeric form. Due to its presumably hygro-
scopic nature it has been isolated as mixtures consisting
of monomeric and dimeric forms17,18 or the monomeric
and the hydrated form (Scheme 4).10,17
Only very few derivatives of AF have become known.
Amongst these are fatty acid esters at C-6 which were
prepared by regioselective chemical as well as enzy-
matic methods by Lundt and co-workers,19 a-D-gluco-
pyranosides at this position20 as well as at C-321
obtained by enzymatic glycosyl transfer. Due to the
high base sensitivity of the unprotected parent com-
pound, other substituted derivatives have been difficult
to prepare.
food industry22,23,25–27 (preventing pigment discolour-
ation, enzymic browning, protecting flavour, aroma
and nutrient content, extending shelf life), cosmetics
industry28 (replacing the unstable L-ascorbic acid) and
pharmaceutical industry (stimulating hormone secretion
of insulin and glucagon-like peptide 1 (GLP 1)).29,30
Intriguingly, it has also been proven that in mammals
and humans, glycogen degrades by a-(1!4)-glucan
lyase to 1,5-anhydro-D-fructose31,32 which subsequently
is reduced to 1,5-anhydro-D-glucitol by a NADPH-
dependent 1,5-anhydro-D-fructose specific reductase.33
This constitutes a third glycogenolytic pathway, in addi-
tion to the phosphorolytic and hydrolytic degradation
cascades.31
It has been demonstrated that 1,5-anhydro-D-fructose
is a powerful antioxidant and antimicrobial agent.22,23
Moreover, an aqueous solution of AF is very stable
under neutral conditions and is, unlike L-ascorbic acid,
not oxidised by molecular oxygen.24 Thus, 1,5-anhy-
dro-D-fructose might find potential applications in the
In spite of these interesting biological properties and
of the impressive progress made in enzymatic prepara-
tion of AF, the product is neither available at bulk scale
nor has it been commercially utilised, presumably due to
the lack of an efficient and cost-efficient preparation of
this compound.

G. Dekany et al. / Carbohydrate Research 342 (2007) 1249–1253
1251
Intramolecular anhydro ring closure was found a
straightforward process employing sodium hydride in
N,N-dimethylformamide (75% from tosylate 16).
Improvements of the reaction conditions employing
aqueous sodium hydroxide at gentle reflux for 15 min
allowed access to 1,5-anhydro-D-fructose derivative 18
in 90% isolated yield.
HO
O
OH
O
HO
O
OH
OH
HO
HO
1
HO
O
O
O
O
O
Deprotection of acetal 18 with 80% aqueous acetic
acid at 65 ꢁC for 15 min gave compound 1 in an isolated
yield of 94%.
This sequence comprises of five simple steps employ-
ing inexpensive starting material and chemicals. Despite
our best efforts, the overall yield from D-fructose via
5-tosylate 14, was only 28%.
HO
HO
OH
dimer 1a
OH
dimer 1b
Gratifyingly, exploiting mesylate 15 together with
slightly modified procedures as given in Section 1, 1,5-
anhydro-D-fructose (1) is available from D-fructose in
50% overall yield at a scale of 10–100 g in each step,
demonstrating the potential for scaling up. Another
redeeming feature of the extremely simple sequence pre-
sented is the regioselective en route accessibility of O-4
in partially protected compound 18, allowing for a wide
variety of defined modifications at this position with a
view to the feasible applications of such products as
mentioned in the introduction.
1 hydrate
Scheme 4.
Consequently, 1,5-anhydro-D-fructose and its deriva-
tives are attractive products bearing substantial indus-
trial and commercial potential provided scaleable and
efficient approaches are at hand.
We considered D-fructose, a cheap and abundant
renewable sugar a suitable starting material for a new
approach towards 1,5-anhydro-D-fructose (Scheme 5).
D-Fructose, via easily available 2,3:4,5-di-O-isopropyl-
idene-b-D-fructopyranose (13, 85%) was converted into
known34 tosylate 14 (95%) by standard procedures.
Alternatively, the known35 corresponding 1-mesylate
15 was prepared (95%) allowing for considerably shorter
reaction times.
The regioselective removal of the 4,5-O-isopropylid-
ene moiety in the presence of the 2,3-acetal turned out
to be the crucial step of the synthesis and, after several
pitfalls employing a large variety of different acidic cat-
alysts, could be achieved best exploiting oxalic acid in
85% aqueous acetonitrile at reflux temperature furnish-
ing around 50% of the desired intermediate 16 from tos-
ylate 14. Under optimised conditions, the corresponding
1-mesylate 15 gave 73% of desired product 17 and most
of the remaining starting material could be recovered
allowing for ‘yields by conversion’ of well over 90% in
this particular step.
1. Experimental
1.1. General methods
Melting points were recorded on a Tottoli apparatus
and are uncorrected. Optical rotations were measured
on a JASCO Digital Polarimeter or with a Perkin Elmer
341 with a path length of 10 cm. NMR spectra were
recorded at 200 as well as 500 MHz (1H), and at 50
and 125 MHz (13C). Chemical shifts are listed in d
employing residual, not deuterated, solvent as the inter-
nal standard. The signals of the protecting groups and of
aromatic substituents were found in the expected regions
and are not listed explicitly. Structures of crucial inter-
mediates were unambiguously assigned by 1D-TOCSY
and HSQC experiments. Electrospray mass spectra were
recorded on an HP 1100 series MSD, Hewlett Packard.
O
O
O
O
O
O
O
O
O
D- fructose
1
O
OR
HO
OR
OR
O
OH
O
16
17
R = Ts
R = Mes
18
R = H
13
14
15
R = H
18a
R = Ac
R = Ts
R = Mes
Scheme 5.

1252
G. Dekany et al. / Carbohydrate Research 342 (2007) 1249–1253
Samples were dissolved in acetonitrile/MeOH mixtures.
The scan mode for negative ions (mass range 100–
1000 D) was employed varying the fragmentation volt-
age from 30 to 130 V. TLC was performed on precoated
aluminum sheets (E. Merck 5554). Compounds were
detected by staining with concd H2SO4 containing
5% vanillin.
leum ether and the mixture was stirred vigorously at
ambient temperature. The suspension thus obtained
was cooled to 0 ꢁC and stirred for another 2 h. Filtration
20
gave product 17 (19 g, 71%): mp 90–94 ꢁC, ½aꢀD +13 (c
0.7, CH2Cl2); 1H NMR (acetone-d6): d 4.35 (m, 2H,
H-1a, H-1b), 4.16 (m, 2H, H-3, H-4), 3.95 (ddd, 1H,
J4,5 3 Hz, J5,6a 2.8 Hz, J5,6b 2 Hz, H-5), 3.70 (dd, 1H,
J6a,6b 11.5 Hz, H-6a), 3.64 (dd, H-6b); 13C NMR: d
100.5 (C-2), 76.9, 67.2, 67.1, 63.7, 63.0 (C-1, C-3, C-4,
C-5, C-6). HRAPIESIMS (neg. mode): [MꢁH] calcd
for C10H18O8S 374.4061, found 373.410. Anal. Calcd
for C10H18O8S: C, 40.26; H, 6.08. Found: C, 40.21; H,
6.13.
1.2. 2,3:4,5-Di-O-isopropylidene-b-D-fructopyranose (13),
2,3:4,5-di-O-isopropylidene-1-O-p-tolylsulfonyl-b-D-
fructopyranose (14), and 2,3:4,5-di-O-isopropylidene-1-O-
methylsulfonyl-b-D-fructopyranose (15)
Compounds (13), (14) as well as (15) were prepared
A second crop (0.7 g, 2.6%) was obtained from pro-
cessing of the mother liquor. The CH2Cl2 phase was
concentrated under diminished pressure to give unre-
acted starting material 15 (10 g, 25%).
following known34,35 procedures.
1.3. 2,3-O-Isopropylidene-1-O-p-tolylsulfonyl-b-D-fructo-
pyranose (16)
1.5. 1,5-Anhydro-2,3-O-isopropylidene-b-D-fructo-
pyranose (18)
To a soln of compound 1434 (99 g, 239 mmol) in 85% aq
MeCN (1.2 L), oxalic acid (24 g, 192 mmol) was added
and the mixture was kept under reflux for 48 h. Solid
NaHCO3 was added until pH 7, the mixture was filtered
and the filtrate was concentrated under diminished pres-
sure. The remaining material was washed with CH2Cl2,
the organic layer was dried (Na2SO4), filtered and con-
centrated under diminished pressure. The remaining
crystals were dried to give compound 16 (40 g, 45%):
Method A: To a soln of tosylate 16 (66.4 g, 176 mmol) in
DMF (1.5 L), NaH (15 g) was added and the mixture
was stirred at ambient temperature for 2 h. Water was
added, solvents were removed under diminished pres-
sure and the remaining residue was treated with CH2Cl2.
The soln was filtered over a plug of celite and the solvent
was removed under diminished pressure.
20
mp 82–85 ꢁC, ½aꢀD +16 (c 1.2, CH2Cl2); 1H NMR
Crystallisation from ether furnished 1,5-anhydrosugar
20
(MeOH-d4): d 4.18 (d, 1H, J1a,1b 10.4 Hz, H-1a), 4.14
(d, 1H, H-1b), 4.07 (br s, 2H, H-3, H-4), 3.90 (m, 1H,
H-5), 3.54 (m, 2H, H-6a, H-6b); 13C NMR: d 100.3
(C-2), 77.3, 66.9, 66.7, 63.6, 62.7 (C-1, C-3, C-4, C-5,
C-6). Anal. Calcd for C16H22O8S: C, 51.33; H, 5.92.
Found: C, 51.27; H, 5.97.
18 (26.9 g, 75%): mp 110–113 ꢁC, ½aꢀD +18 (c 0.8,
MeOH); 1H NMR (MeOH-d4): d 4.28 (d, 1H, J1a,1b
9.0 Hz, H-1a), 4.19 (dd, 1H, J5,6a 1.5 Hz, J6a,6b
10.2 Hz, H-6a), 4.16 (d, 1H, J3,4 3.6 Hz, H-3), 3.99 (d,
1H, H-4), 3.97 (dd, 1H, J5,6b 1.5 Hz, H-6b), 3.86 (d,
1H, H-1b), 3.85 (m, 1H, H-5); 13C NMR (CDCl3): d
115.6 (C-10), 99.4 (C-2), 87.8, 77.0, 74.6, 73.9, 71.2,
66.9 (C-1, C-3, C-4, C-5, C-6), 28.0 (C-20a), 25.8
(C-20b). HRAPIESIMS (neg. mode): [MꢁH] calcd for
C9H14O5 202.2045, found 201.190. Anal. Calcd for
C9H14O5: C, 53.46; H, 6.98. Found: C, 53.40; H, 7.01.
Method B: Compound 17 (30 g, 100 mmol) was added
to a soln of NaOH (7.24 g) in water (83 mL). The reac-
tion mixture was stirred under gentle reflux for 15 min,
then cooled in an ice bath and slowly acidified to pH 6
by addition of 10% aq HCl. Aq ammonia (2 mL, 25%)
was added and the solvent was removed under dimin-
ished pressure. The residue was vigorously stirred with
CHCl3 (500 mL) at ambient temperature for 20 min,
then filtered. The residue was washed with CHCl3
(300 mL). Combined organic layers were concentrated
under diminished pressure and the resulting solid resi-
due was stirred with toluene (150 mL) under reflux for
30 min, then brought to ambient temperature. The clou-
dy suspension was stirred at 0 ꢁC for 2 h and the precip-
itate was collected by filtration and tried to give
anhydrofructose derivative 18 (18 g, 90%).
The starting material 14 (29.8 g, 30%) could be recov-
ered from the mother liquor.
1.4. 2,3-O-Isopropylidene-1-O-methylsulfonyl-b-D-fructo-
pyranose (17)
A mixture of compound 1535 (40 g, 118 mmol) and oxa-
lic acid (8.5 g, 94 mmol) was kept under reflux in
MeCN/water [600 mL, 5:1 (v/v)] for 28 h. The reaction
was allowed to cool and 25% aq ammonia (16 mL)
was added slowly over 5 min. The resulting suspension
was stirred at 0 ꢁC for 30 min, then filtered. The preci-
pitate was washed with MeCN (100 mL). The filtrate
and the washing were combined and the solvent was
removed under diminished pressure. The remaining
residue was taken up into CH2Cl2 (200 mL) and the
resulting soln was filtered and washed with distilled
water (20 mL). The aq phase was extracted with CH2Cl2
(3 · 200 mL). The combined aq layers were concen-
trated under diminished pressure to give crude product
17 as a syrup. This was dissolved in 1:1 EtOAc–petro-

G. Dekany et al. / Carbohydrate Research 342 (2007) 1249–1253
1253
8. Nakamura, T.; Takahashi, A.; Kato, K. Japanese Patent
63072696, 1988; AN 1988, 510835.
9. Baute, M.-A.; Baute, R.; Deffieux, G. Phytochemistry
1988, 27, 3401–3403.
10. Taguchi, T.; Haruna, M.; Okuda, J. Biotechnol. Appl.
Biochem. 1993, 18, 275–283.
11. Freimund, S.; Huwig, A.; Giffhorn, F.; Ko¨pper, S. Chem.
Eur. J. 1998, 4, 2442–2455.
For further characterisation, 18a was prepared by
conventional O-acetylation employing acetic anhydride
20
in pyridine to give crystalline acetate: ½aꢀD +5.3 (c
0.85, CH2Cl2), mp 98–102 ꢁC; 1H NMR (CDCl3): d
5.16 (d, 1H, J3,4 4.2 Hz, H-4), 4.36 (d, 1H, J1a,1b
9.0 Hz, H-1a), 4.27 (dd, 1H, J5,6a 1.5 Hz, J6a,6b
10.2 Hz, H-6a), 4.22 (d, 1H, H-3), 4.13 (dd, 1H, J5,6b
1.5 Hz, H-6b), 4.05 (br s, 1H, H-5), 3.95 (d, 1H, H-
1b); 13C NMR: d 170.6 (C@O), 99.2 (C-2), 83.7, 77.0,
71.3, 71.1, 66.9 (C-1, C-3, C-4, C-5, C-6), 21.2 (COMe).
Anal. Calcd for C11H16O6: C, 54.09; H, 6.60. Found: C,
54.03; H, 6.66.
´
12. Yu, S.; Pedersen, M.; Kenne L. PCT Intl. Appl. WO
9409122, 1994; AN 1994, 503075.
13. Baute, M.-A.; Baute, R.; Deffieux, G. French Demande
FR 2617502, 1987; AN 1990, 97028.
14. Yu, S.; Bojsen, K.; Marcussen, J. J. PCT Int. Appl. WO
9612026, 1996; AN 1996, 386030.
15. Bojsen, K.; Yu, S.; Kragh, K. M.; Christensen, T. M. I. E.;
Marcussen, J. PCT Int. Appl. WO 9510617, 1995; AN
1995, 721196.
1.6. 1,5-Anhydro-D-fructopyranose (1)
16. Yu, S.; Bojsen, K.; Kragh, K. M.; Bojko, M.; Nielsen, J.;
Marcussen, J. PCT Int. Appl. WO 9510618,1995; AN
1995, 721197.
17. Andersen, S. M.; Lundt, I.; Marcussen, J.; Søtofte, I.; Yu,
S. J. Carbohydr. Chem. 1998, 17, 1027–1035.
18. Freimund, S.; Ko¨pper, S. Carbohydr. Res. 1998, 308, 195–
200.
A soln of compound 18 (10 g, 49.5 mmol) in aq acetic
acid (80%, 30 mL) was stirred at 65 ꢁC for 15 min and
the solvent was removed under diminished pressure.
The residue was dissolved in dry MeOH (50 mL) and
the soln was slowly added to cold ether (1000 mL)
under vigorous stirring. The resulting suspension was
stirred at 0 ꢁC for 2 h and filtered. The precipitate was
washed with dry ether (300 mL) and dried under dimin-
ished pressure to give known final product to give 1 (8 g,
94%) as a hygroscopic white powder. The product
showed a 13C NMR spectrum identical with that of an
authentical sample of 1 provided by Danisco (Copen-
hagen) and prepared according to Refs. 12 and 14: 13C
NMR (1 hydrate, D2O): d 92.8 (C-2), 80.8, 77.1, 71.9,
69.1, 61.4 (C-1, C-3, C-4, C-5, C-6); for comparative
data, see Refs. 17 and 18.
19. Andersen, S. M.; Lundt, I.; Marcussen, J.; Yu, S.
Carbohydr. Res. 1999, 320, 250–256.
20. Richard, G.; Yu, S.; Monsan, P.; Remaud-Simeon, M.;
Morel, S. Carbohydr. Res. 2005, 340, 395–401.
21. Yoshinaga, K.; Abe, J.; Tanimoto, T.; Koizumi, K.;
Hizukuri, S. Carbohydr. Res. 2003, 338, 2221–2225.
22. Elsser, D.; Morgan, J. A.; Thomas, L. V.; Yu, S. PCT, Int.
Appl. WO 02/26060, 2002; AN 2002, 256006.
23. Elsser, D.; Morgan, J. A.; Thomas, L. V.; Yu, S. PCT, Int.
Appl. WO 02/26061, 2002; AN 2002, 256007.
24. Kurata, T.; Miyake, N.; Otsuka, Y. Biosci. Biotech.
Biochem. 1996, 60, 1212–1214.
25. Yajima, S.; Furuhashi, T.; Muroya, M.; Yoshinaga, K.;
Fujisue M. Japanese Patent 2003047416, 2003; AN 2003,
123064.
26. Muroya, M.; Fujisue, M.; Matsuda, T. Japanese Patent
2002125626, 2002; AN 2002, 341296.
Acknowledgment
27. Hizukuri, S.; Abe, J.; Junichi, Y.; Muroya, K.; Yoshinaga,
K.; Fujisue, M.; Ishiba, H. PTC Int. Appl. WO
2001072124, 2001; AN 2001, 730498.
A.E.S. thanks the Technical University of Denmark
(DTU) for a guest professorship in March 2005.
28. Hisaku, S.; Abe, J.; Muroya, M.; Yoshinaga, K. Japanese
Patent 2002058477, 2002; AN 2002, 148565.
29. Ahren, B.; Yu, S. PCT Int. Appl. WO 2001051058, 2001;
AN 2001, 525926.
References
30. Ahren, B.; Holst, J. J.; Yu, S. Eur. J. Pharm. 2000, 397,
219–225.
1. Andersen, S. M.; Lundt, I.; Marcussen, J.; Yu, S.
Carbohydr. Res. 2002, 337, 873–890.
31. Kametani, S.; Shiga, Y.; Akanuma, H. Eur. J. Biochem.
1996, 242, 832–838.
32. Suzuki, M.; Kametani, S.; Uchida, K.; Akanuma, H. Eur.
J. Biochem. 1996, 240, 23–29.
33. Yamanouchi, T.; Tachibana, Y.; Akanuma, H.; Minoda,
S.; Shinohara, T.; Moromizato, H.; Miyashita, H.;
Akaoka, I. Am. J. Phys. 1992, 263, E268–E273.
34. Levene, P. A.; Tipson, R. S. J. Biol. Chem. 1937, 120, 607–
619.
2. Blair, M. G. Adv. Carbohydr. Chem. 1954, 9, 97–129;
Ferrier, R. J. Adv. Carbohydr. Chem. 1965, 20, 67–137;
Ferrier, R. J. Adv. Carbohydr. Chem. 1969, 24, 199–266.
3. Maurer, K.; Mann, H. Ber. Dtsch. Chem. Ges. 1927, 60,
1316–1320.
4. Maurer, K. Ber. Dtsch. Chem. Ges. 1929, 62, 332–338.
5. Maurer, K.; Muller, A. Ber. Dtsch. Chem. Ges. 1930, 63,
¨
2069–2073.
6. Corbett, W. M. J. Chem. Soc. C 1959, 3213–3216.
7. Lichtenthaler, F. W.; El Ashry, E. S. H.; Go¨ckel, V. H.
Tetrahedron Lett. 1980, 21, 1429–1432.
35. Barnett, J. E. G.; Atkins, G. R. S. Carbohydr. Res. 1972,
25, 511–515.