Kiliani on ketoses: Branched carbohydrate building blocks from D-fructose and L-sorbose

Article abstract of DOI:10.1016/j.tetlet.2004.10.086

Protected branched sugar lactones are available via Kiliani-acetonation sequences on readily available ketoses such as d-fructose and l-sorbose. In both cases, the readily crystallized diacetonides have a 2,3-cis-diol relationship in the product lactone. An efficient double inversion of the configuration at C-4 and C-5 of the product from d-fructose gives access to the formal Kiliani product from l-psicose. Branched carbohydrate lactones are likely to be of significant value as chirons for homochiral targets with functionalized quaternary centres.

Full text of DOI:10.1016/j.tetlet.2004.10.086

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 9461–9464
Kiliani on ketoses: branched carbohydrate building blocks
from D-fructose and L-sorbose
David Hotchkiss, Raquel Soengas, Michela Iezzi Simone, Jeroen van Ameijde,
Stuart Hunter, Andrew R. Cowley and George W. J. Fleet*
Chemistry Research Laboratory, Department of Chemistry, University of Oxford, Mansfield Road, Oxford OX1 3TA, UK
Received 6 September 2004; revised 6 October 2004; accepted 14 October 2004
Abstract—Protected branched sugar lactones are available via Kiliani-acetonation sequences on readily available ketoses such as D-
fructose and L-sorbose. In both cases, the readily crystallized diacetonides have a 2,3-cis-diol relationship in the product lactone. An
efficient double inversion of the configuration at C-4 and C-5 of the product from D-fructose gives access to the formal Kiliani prod-
uct from L-psicose. Branched carbohydrate lactones are likely to be of significant value as chirons for homochiral targets with func-
tionalized quaternary centres.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
respectively, followed by acetonation of crude reaction
mixtures (Scheme 1).
Carbohydrate building blocks provide a wide range of
starting materials for the enantiospecific synthesis of
highly functionalized homochiral targets.1 However,
there are few, if any,2 branched sugar chirons3 that are
easily available on a reasonable scale from cheap unpro-
tected carbohydrates.
Although the yields of the two processes are moderate,
the low cost of the ketoses and the ease of crystallization
O
OH
OH
O
(i) NaCN, H2O
O
O
CH2OH
HOH2C
Although extensive studies on the Kiliani ascension4
from an aldose to a higher sugar with a linear carbon
chain have been reported,5 the reaction on unprotected
ketoses6 (to provide branched sugars) has been little
investigated. The initial work by Kiliani on the reaction
of D-fructose7 1 and L-sorbose8 5 showed that the
branched acids were not conveniently crystallized; fur-
ther studies have been limited9,10 and no simple protect-
ing group chemistry of the crude products has been
hitherto described. Access to simply prepared protected
derivatives of branched Kiliani products would provide
a new family of carbohydrate scaffolds from the chiral
pool.
+
CH2OH
(ii) Me2CO, H
OH
O
O
O
2
1 D-Fructose
double
inversion
OH
OH
OH
[Si]O
O
O
CH2OH
HOH2C
CH2O[Si]
OH
O
O
O
4
3 L-Psicose
O
OH
OH
This letter provides experimental details for the synthe-
sis of the two branched sugar building blocks 2 and 6 by
the Kiliani reaction on D-fructose 1 and L-sorbose 5,
O
(i) NaCN, H2O
O
O
CH2OH
HOH2C
+
CH2OH
(ii) Me2CO, H
OH
O
O
O
6
5 L-Sorbose
*
Scheme 1.
Corresponding author. E-mail: george.fleet@chem.ox.ac.uk
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.10.086

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
9462
D. Hotchkiss et al. / Tetrahedron Letters 45 (2004) 9461–9464
of two of the products make this a practical procedure
for the generation of a series of protected branched su-
gar chirons. A short sequence from 2 involving a double
inversion at C-4 and C-5 allows access to 4, the formal
product from a Kiliani reaction on the inaccessible sugar
L-psicose 3.
and then passed through a column of Amberlite IR-
120 H+ form and the solvent evaporated. Acetone
(150mL), concentrated sulfuric acid (2.5mL) and anhy-
drous copper sulfate were added to the residue, and the
resulting mixture was stirred at room temperature. After
6h, TLC(ethyl acetate/cyclohexane 1:1) indicated the
formation of a major product (Rf 0.28) and a minor
one (Rf 0.31). The mixture was then neutralized with
solid sodium carbonate, filtered and the solvent evapo-
rated. The residue was shaken with dichloromethane
(100mL) and water (100mL) and the aqueous layer
2. Synthesis
The Kiliani reaction of D-fructose 1 with aqueous so-
dium cyanide afforded a crude mixture of the two epi-
meric lactones 7 and 12 together with their open chain
acids (Scheme 2). After the solvent had been removed,
the residue was extracted with acetone in the presence
of sulfuric acid to give a mixture that mainly consisted
of the two diacetonides 2 and 10. Although 2 and 10
are difficult to separate cleanly by flash chromatogra-
phy, it is easy to crystallize the diacetonide 2, derived
from 7, as the major product in 51% yield from the mix-
ture. A smaller amount of the diacetonide 10, derived
from the minor isomer 12, was isolated in 9% yield.
was
further
extracted
with
dichloromethane
(2 · 50mL). The combined organic extracts were dried
(magnesium sulfate), filtered and evaporated to produce
a mixture which was purified by flash chromatography
(ethyl acetate/cyclohexane 1:1) and crystallization of
the impure fractions from ether/hexane, to yield 211
(8.19g, 51% yield) as a white crystalline solid. A minor
product 1012 (1.44g, 9%) was also recrystallized from
ether/hexane.
The structure of 10 was firmly established by X-ray cryst-
allographic analysis13 of the corresponding silyl ether
21
11, {mp 76–78ꢁC, ½aŠ +46.0 (c, 0.42)14} formed by
As
a
practical procedure, D-fructose
1
(10.00g,
D
treatment of 10 with tert-butyldimethyl (TBDMS) tri-
flate in dichloromethane in the presence of 2,6-lutidine.
55.51mmol) and sodium cyanide (3.54g, 72.24mmol)
were stirred together in water (100mL) at room temper-
ature for 24h. The reaction mixture was then refluxed
until evolution of ammonia had ceased (approx 12h).
The mixture was allowed to cool at room temperature
A beautiful series of papers by Ho exploited the crossed
aldol reactions of 2,3-O-isopropylidene protected sugars
OH
OH
OH
O
O
O
O
O
O
O
O
O
(iv)
O
CH2OH
(ii)
(i)
HOH2C
O
(iii)
OH
OH
HOH2C
O
CH2OH
CH2OH
OH
O
CH2OH
HO
OH
O
O
O
O
O
O
7
1 D-Fructose
2
8
9
+
(v)
OH
O
OH
OH
OH
[Si]O
O
O
[Si]O
O
O
O
HO
O
(ii)
O
O
(vii)
(viii)
O
HOH2C
O
OH
O
CH2O[Si]
O
CH2O[Si]
CH2OH
O
O
RO
O
O
HO
CH2OH
O
O
O
10 R = H
(vi)
12
13
14
4
11 R = [Si]
OH
OH
OH
OH
O
+
O
O
HOH2C
O
O
(i)
O
O
HOH2C
CH2OH
O
OH
OH
HOH2C
O
O
OH
CH2OH
CH2OH
OH
O
HO
CH2OH
(ii)
HOH2C
HO
OH
HO
O
OH
O
15
16
5 L-Sorbose
17
3 L-Psicose
O
O
O
O
O
O
O
O
O
O
O
(x)
(iv)
O
(ix)
OH
OH
O
CH2OH
CH2OH
O
O
[Si] = tert-BuMe2Si
O
O
O
O
O
O
18
19
20
6
Scheme 2. Reagents and conditions: (i) NaCN, H2O; (ii) Me2CO, H2SO4, CuSO4; (iii) Br2, BaCO3, H2O; (iv) CH2O, K2CO3, H2O; (v) CH3CO2H,
H2O; (vi) TBDMSOSO2CF3, 2,6-lutidine, CH2Cl2; (vii) TBDMSCl, pyridine, DMF; (viii) (CF3SO2)2O, pyridine, CH2Cl2, then KOH, H2O, dioxane;
(ix) DIBAL-H, THF; (x) Br2, BaCO3, H2O, dioxane.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
D. Hotchkiss et al. / Tetrahedron Letters 45 (2004) 9461–9464
9463
with formaldehyde.15 Thus, an identical sample of 2 was
also prepared from diacetone mannose 916 by an initial
aldol condensation with aqueous formaldehyde to give 8
tion and acetonation,24 and the microbial oxidation–en-
zyme catalyzed epimerizations reported by Izumori and
co-workers,25 may provide a powerful armory for the
synthesis of new densely functionalized homochiral
targets.
[52% yield]; subsequent oxidation of the lactol 8 with
21
D
bromine water gave 2 mp 130–132ꢁC, ½aŠ +34.7 (c,
1.0) in 86% yield. However, the direct Kiliani procedure
appears to be easily scalable, requires less effort and is
completed at a small fraction of the cost of starting from
D-mannose.
Acknowledgements
Financial support (to R.S. and M.I.S.) provided through
the European CommunityÕs Human Potential Pro-
gramme under contract HPRN-CT-2002-00173 is grate-
fully acknowledged.
Some chemistry of the diacetonide 2 was explored. Mild
acid hydrolysis of 2 by aqueous acetic acid gave the
monoacetonide 13 in quantitative yield {mp 128–
24
132ꢁC, ½aŠ +64.4 (c, 0.93, MeOH)}. Treatment of 13
21
D
with TBDMS chloride in DMF in the presence of pyri-
References and notes
dine gave the disilyl ether 14 [73% yield, oil, ½aŠ +36.5
D
(c, 1.26)]. The disilyl ether 14 was treated with triflic
anhydride in dichloromethane in the presence of pyri-
dine to give the corresponding triflate which, when trea-
1. Bols, M. Carbohydrate Building Blocks; John Wiley &
Sons: New York, 1996; Lichtenthaler, F. W. Carbohy-
drate Synthons in Natural Product Chemistry. ACS Symp.
Ser. 2002, 841, 47–83; Lichtenthaler, F. W.; Peters, S. C.
R. Chim. 2004, 7, 65–90; Lichtenthaler, F. W. Mod. Synth.
Methods 1992, 6, 273–376.
2. Whistler, R. L.; BeMiller, J. N. Methods Carbohydr.
Chem. 1963, 2, 477–479; Whistler, R. L.; BeMiller, J. N.
Methods Carbohydr. Chem. 1963, 2, 484–485; Ireland, R.
E.; Andersen, R. C.; Baboud, R.; Fitzsimmons, B. J.;
McGarvey, G. J.; Thaisrivongs, S. A.; Wilcox, C. S. J. Am.
Chem. Soc. 1983, 105, 1988–2006; Peters, S.; Lichten-
thaler, F. W.; Lindner, H. J. Tetrahedron: Asymmetry
2003, 14, 2475–2479.
3. Hanessian, S. Total Synthesis of Natural Products: the
Chiron Approach; Pergamon: New York, 1983.
4. Hudson, C. S. Adv. Carbohydr. Chem. 1945, 1, 2–36;
Hudson, C. S. J. Am. Chem. Soc. 1951, 73, 4498–4499;
Pratt, J. W.; Richtmyer, N. K.; Hudson, C. S. J. Am.
Chem. Soc. 1953, 75, 4503–4507.
5. Varma, R.; French, D. Carbohydr. Res. 1972, 25, 71–79;
Blazer, R. M.; Whaley, T. W. J. Am. Chem. Soc. 1980,
102, 5082–5085; Beacham, A. R.; Bruce, I.; Choi, S.;
Doherty, O.; Fairbanks, A. J.; Fleet, G. W. J.; Skead, B.
M.; Peach, J. M.; Saunders, J.; Watkin, D. J. Tetrahedron:
Asymmetry 1991, 2, 883–900; Bell, A. A.; Nash, R. J.;
Fleet, G. W. J. Tetrahedron: Asymmetry 1996, 7, 593–604.
6. Lichtenthaler, F. W. Carbohydr. Res. 1998, 313, 69–90.
7. Kiliani, H. Ber. Dtsch. Chem. Ges. 1885, 18, 3066–3072;
Kiliani, H. Ber. Dtsch. Chem. Ges. 1886, 19, 221–227.
8. Kiliani, H. Ber. Dtsch. Chem. Ges. 1928, 61, 1155–1169.
9. Woods, R. J.; Neish, A. C. Can. J. Chem. 1953, 31, 471–
475; Woods, R. J.; Neish, A. C. Can. J. Chem. 1954, 32,
404–414.
ted with potassium hydroxide in aqueous dioxane,
21
D
afforded the alcohol 4 (½aŠ +13.5 (c, 0.73), oil)—formed
by double inversion at C-5 and C-4—in 58% yield. The
structure of the L-allono-lactone 4, the formal product
from a Kiliani sequence from L-psicose 3, was firmly
established by X-ray crystallographic analysis of a
derivative.17
The Kiliani reaction on L-sorbose 5 to give the lactones
15 and 16, with subsequent acetonation, was performed
identically to that described above for D-fructose 1 to
the stage of the processing of the residue of the crude
mixture of diacetonides. Whereas for fructose 1 an ini-
tial flash column was necessary in the work-up, direct
crystallization of the residue from L-sorbose allowed iso-
lation of the diacetonide 618 in 17% yield without chro-
matography; however, further work-up of the mixture
of the epimeric spiro-acetonides 1719 is difficult.20
The diacetonide 6 was also prepared from the diaceto-
nide of L-gulonolactone 18.21 Reduction of 18 with di-
isobutylaluminium hydride (DIBAL-H) in THF gave
the corresponding lactol22,23 19 which underwent a
crossed aldol reaction with aqueous formaldehyde to
26
D
the branched gulose 20 mp 83–85ꢁC, ½aŠ +5.0 (c 0.73)
in 70% yield. Oxidation of the lactol 20 with bromine
in aqueous dioxane in the presence of barium carbonate
afforded the branched lactone 6 in 96% yield, identical in
all respects with the sample of 6 made from sorbose.
While the four-step route from L-gulonolactone is effi-
cient, it competes neither in terms of cost nor time with
the procedure from L-sorbose.
10. Gorin, P. A. J.; Perlin, A. S. Can. J. Chem. 1956, 36, 480–
485.
11. Selected data for 2-C-hydroxymethyl-2,3:5,6-di-O-isopro-
21
D
pylidene-D-mannono-1,4-lactone 2: mp 128–130ꢁC; ½aŠ
+34.7 (c, 1.0, CHCl3); dH (d6-acetone): 1.35, 1.41, 1.42
0
00
(4 · s, 12H, 4 · –CH3), 3.89–4.03 (m, 3H, H6, H2 , H2 ,
0
0
0
4.15 (dd, 1H, J6,6 8.9Hz, J6 ,5 6.2Hz, H6 , 4.41–4.46 (m,
1H, H5), 4.50 (dd, 1H, J3,4 3.4Hz, J4,5 7.0Hz, H4), 4.69
3. Conclusion
0
00
(dd, 1H, J2 , OH 4.8Hz, J2 , OH 6.1Hz, –OH), 4.92 (d, 1H,
J3,4 3.4Hz, H3); dC (CDCl3): 26.99, 26.16, 26.76 (4 ·
–CH3), 61.43, 66.24 (2 · –CH2–), 72.42, 78.45, 78.70 (3 ·
–CH–), 86.00, 109.73, 113.79 (3 · –C–), 175.69 (–C@O).
12. Selected data for 2-C-hydroxymethyl-2,20:5,6-di-O-isoprop-
This letter provides an indication of the potential ease of
access to protected branched sugar chirons; both from
fructose and from sorbose easily crystallized branched
diacetonides can be isolated. Such carbohydrate build-
ing blocks are not restricted to those available from
the Kiliani-acetonation procedure on D-fructose and L-
sorbose. The Ferrier–Kiliani combination of microbial
oxidation of alditols combined with cyanohydrin forma-
ylidene-D-glucono-1,4-lactone 10: mp 102–104ꢁC;
21
D
½aŠ +67.7 (c, 1.56, CHCl3); dH (CDCl3): 1.37, 1.45, 1.46
(3 · s, 12H, 4 · –CH3), 2.65 (d, 1H, JOH,3 3.1Hz, –OH),
4.05 (dd, 1H, J6,6 8.9Hz, J6 ,5 4.1Hz, H6 ), 4.20–4.25 (m,
0
0
0

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
9464
D. Hotchkiss et al. / Tetrahedron Letters 45 (2004) 9461–9464
0
2H, H6, H3), 4.30 (d, 1H, J2 ,2 9.6Hz, H2 , 4.30–4.35 (m,
1H, H5), 4.44 (d, 1H, J3,4 3.4Hz, H4), 4.47 (d, 1H,
00
00
18. Selected data for 2-C-hydroxymethyl-2,3:5,6-di-O-isoprop-
ylidene-L-gulono-1,4-lactone 6: mp 128–129ꢁC{lit.
23:5
D
0
0
00
J2 ,2 9.6Hz, H2 ); dC (CDCl3): 24.96, 25.28, 26.34, 26.81
(4 · –CH3), 65.39, 67.37 (2 · –CH2–), 72.69, 72.96, 81.35
(3 · –CH–), 82.66, 110.14, 112.73 (3 · –C–), 173.94
(–C@O).
[19] 132–133ꢁC}; ½aŠ
+57.9 (c 1.14, CHCl3); dH (CDCl3)
1.39, 1.39, 1.46, 1.47 (4 · s, 12H, 4 · CH3), 3.81–3.86 (m,
1H, H-6), 3.90 (d, 1H, H-20, J2 ,2 11.4Hz), 4.01 (d, 1H, H-
0
00
200, J2 ,2 11.4Hz), 4.19–4.23 (m, 1H, H-60), 4.41–4.48 (m,
00
0
13. Cowley, A. R.; Fleet, G. W. J.; Simone, M. I.; Soengas, R.
Acta Crystallogr. 2004, E60, o2142–o2143.
14. Specific rotations [a]D in this paper were determined in
CHCl3 as solvent, except where otherwise stated. 1H
NMR spectra were determined at 400MHz and 13CNMR
were determined at 100.6MHz.
15. Ho, P.-T. Tetrahedron Lett. 1978, 19, 1623–1626; Ho,
P.-T. Can. J. Chem. 1979, 57, 381–386; Ho, P.-T. Can.
J. Chem. 1985, 63, 2221–2224.
16. Schmidt, O. T. Methods Carbohydr. Chem. 1963, 2, 318–
325.
17. The disilyl ether 4 was converted into 5-amino-5-deoxy-2-
C-hydroxymethyl-2,3-O-isopropylidene-D-talono-1,5-lac-
tam, the structure of which was confirmed by X-ray
crystallography: see van Ameijde, J.; Cowley, A. R.; Fleet,
G. W. J.; Nash, R. J.; Simone, M. I.; Soengas, R. Acta
Crystallogr. 2004, E60, o2140–o2141.
2H, H-4, H-5), 4.70 (d, 1H, H-3, J3,4 2.9Hz); dC (CDCl3)
25.52, 26.32, 26.63, 26.85 (4 · CH3), 61.57 (C-20), 65.11
(C-6), 74.98 (C-5), 78.72 (C-3), 81.15 (C-4), 86.14 (C-2),
110.46, 114.18 (2 · C(CH3)2), 175.49 (C-1).
19. Anderson, R. A.; Einstein, F. W. B.; Hoge, R.; Slessor, K.
N. Acta Crystallogr. 1977, B33, 2780–2783.
20. Further studies on the acetonation reactions are in
progress.
21. Fleet, G. W. J.; Ramsden, N. G.; Witty, D. R. Tetrahedron
1989, 45, 319–326.
22. Garcia-Moreno, M. I.; Diaz-Perez, P.; Ortiz Mellet, C.;
Garcia Fernandez, J. M. J. Org. Chem. 2003, 68, 8890–
8901.
23. Hricoviniova, Z. Synthesis 2001, 751–754.
24. Ferrier, R. J. J. Chem. Soc. 1962, 3544–3549.
25. Granstrom, T. B.; Takata, G.; Tokuda, M.; Izumori, K.
J. Biosci. Bioeng. 2004, 97, 89–94.