From d-xylose to terminal polyols: a simple synthetic route

Article abstract of DOI:10.1016/j.carres.2009.12.006

A simple and efficient synthetic approach toward different terminal alkyl tetraols and triols, starting from d-xylose, is described. The opening of the oxetane ring in a suitably protected 3,5-anhydro-d-xylose derivative with Grignard reagents leads to d-xylose-derived 5-deoxy-5-C-alkyl derivatives, which are suitable for reduction to terminal polyols after protecting group hydrolysis.

Full text of DOI:10.1016/j.carres.2009.12.006

Carbohydrate Research 345 (2010) 543–546
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Note
From D-xylose to terminal polyols: a simple synthetic route
b
Pavle Hadzic a, , Mirjana Popsavin
*
a Institute for Security, Belgrade, Kraljice Ane bb., Serbia
b Faculty of Sciences, Department for Chemistry, Novi Sad, Trg D. Obradovica 3, Serbia
a r t i c l e i n f o
a b s t r a c t
Article history:
A simple and efficient synthetic approach toward different terminal alkyl tetraols and triols, starting from
Received 15 October 2009
Received in revised form 5 December 2009
Accepted 8 December 2009
Available online 8 January 2010
D
-xylose, is described. The opening of the oxetane ring in a suitably protected 3,5-anhydro-
ative with Grignard reagents leads to -xylose-derived 5-deoxy-5-C-alkyl derivatives, which are suitable
for reduction to terminal polyols after protecting group hydrolysis.
D-xylose deriv-
D
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Tetraols
Terminal polyols
Sugar oxetane
Ring opening
Grignard reagents
With the growing concern for environment protection, sugar-
based surfactants have received much attention as solubilizers
and emulsifying agents. Carbohydrates are generally considered
readily available, cheap, biodegradable, and nontoxic materials.1
Moreover, carbohydrate-based non-ionic detergents are consid-
ered non-denaturant and are widely used in the isolation of mem-
brane proteins in their biologically active form.2,3 The polar head
group in carbohydrate-based surfactants is a sugar moiety to
which a hydrophobic tail of choice is coupled.
generally between the oxygen and the least-substituted ring alpha
carbon.10 In contrast to the numerous synthetic applications of
oxetane ring opening with alkyl magnesium halogenides, the
opening of carbohydrate-based oxetane rings with Grignard re-
agents has not been studied much previously, mainly due to the
relatively limited number of known oxetane ring structures among
carbohydrates. In line with the general reactivity of oxetanes, the
Grignard reaction, when applied to 1, enables a one-step synthesis
of 5-C-alkyl-5-deoxy derivatives of
chains. Ready elongation of the -xylose skeleton with a deliber-
ately chosen lipophilic alkyl chain and with simultaneous preser-
vation of predefined -xylose configuration on optically active
D-xylose with a variety of alkyl
Carbohydrate surfactants are, as a rule, glycosides.4–6 The idea
of elongating a carbohydrate (head group) by forming a bond be-
tween a carbohydrate carbon and an alkyl carbon of a deliberately
chosen hydrophobic alkyl tail seems to be a straightforward solu-
tion for preparation of non-glycoside types of carbohydrate surfac-
tants. However, general synthetic methods for the preparation of
branched-chain carbohydrates do not exist; methods for the for-
mation of carbon–carbon bond are always adapted to the particu-
lar reactivity of the carbohydrate moiety.7 Herein, we report a
D
D
sites is a promising simple synthetic route to different terminal
tetraols.
According to our previous investigations, the opening of the
oxetane ring in 3,5-anhydro-1,2-O-cyclohexylidene-a-D-xylofura-
nose (1) with various nucleophilic reagents (hydrohalogen acids,9
acetyl haloganides,11 primary or secondary amines including alka-
loids12) always proceeds regioselectively and provides good yields
of the corresponding 5-deoxy-5-halo-, 3-O-acetyl-5-deoxy-5-halo-
simple method for the elongation of the D-xylose skeleton at C-5,
with almost any chosen alkyl group. The general synthetic strategy
consists in oxetane ring opening in easily obtainable 3,5-anhydro-
and 5-deoxy-5-amino-D-xylofuranose derivatives, respectively.
D
-xylose 18,9 with Grignard reagents (Scheme 1).
This regioselectivity of nucleophilic oxetane ring opening in 1 is
demonstrated this time by hydride reduction of the 3,5-anhydro
ring: the reaction proceeds almost quantitatively giving 2 in 97%
isolated yield. No reduction product resulting from hydride ap-
proach to the xylose C-3, as confirmed by NMR data (Tables 1
and 2), was formed.
The opening of oxetane rings with nucleophiles, including Grig-
nard reagents, is a known and useful reaction in non-carbohydrate
oxetanes. Ring cleavage with Grignard reagents and ring reduction
with metal hydrides are regioselective reactions: cleavage occurs
Further experimental confirmation of this general similarity in
reactivity and regioselectivity among carbohydrate and non-car-
* Corresponding author at present address: GOSA Institute, Belgrade, Milana
Rakica 35, Serbia.
E-mail address: phadzic@yahoo.com (P. Hadzic).
0008-6215/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2009.12.006

544
P. Hadzic, M. Popsavin / Carbohydrate Research 345 (2010) 543–546
Me
RO
O
O
O
O
Ref. 8, 9
a
D-Xylose
O
O
O
2 R = H
2a R = MeSO2
1
d
b
c
O
O
Ph
g
C7H15
Ph
O
O
O
OH
(from 4)
RO
RO
O
O
HO
OH
4 R = H
4a R = PhSO2
3 R = H
3a R = PhCH2
e
f
5
Scheme 1. Oxetane ring opening of 3,5-anhydro-1,2-O-cyclohexylidene-a-D-xylofuranose (1) with Grignard reagents. Reagents and conditions: (a) LiAlH4,THF, reflux, 3 h; (b)
C7H15MgBr, THF, reflux, 3 h; (c) PhCH2CH2MgBr, THF, reflux, 5 h; (d) MsCl, Py, CHCl3, rt, 12 h; (e) BnCl, KOH, DMSO, 50 °C, 1 h; (f) PhSO2Cl, Py, CHCl3, +4 °C, 48 h; (g) 40% aq
AcOH, reflux, 3 h.
bohydrate oxetanes has been obtained in the reaction of 1 with
Grignard reagents: heptylmagnesium bromide and 2-phenylethyl
magnesium bromide in reaction with 1 gave 5-deoxy-5-C-heptyl-
recorded with a Termo Finnigan Polaris Q (EIMS) or Finnigan
MAT 8230BE spectrometer (CIMS). The first number when refer-
ring to the mass spectra denotes the m/z value, while the num-
bers in parenthesis correspond to the relative abundance of the
mass peak. TLC was performed on Silica gel DC Alufolien (E.
Merck, Darmstadt), with 4:1 toluene–EtOAc as the mobile phase.
The polarity of the mobile phase was augmented by the addition
of CH3OH when necessary. Visualization of the spots was
achieved by spraying 50% sulfuric acid followed by subsequent
heating at 150 °C. Organic extracts were dried with anhydrous
Na2SO4. Solutions were concentrated using a rotary evaporator
under diminished pressure. Active carbon was used as the decol-
orizing agent.
and 5-deoxy-5-C-(2-phenylethyl)-derivatives
3 and 4, respec-
tively, as the only isolated products. From the standpoint of
searching for terminal polyols, 2–4 represent promising starting
compounds for the synthesis both terminal 1,2,3,4-alkyl tetraols
and 1,2,4-alkyl triols. The alkylidene protecting group in C-5-elon-
gated D-xylose derivatives is easily hydrolyzed with acid, as illus-
trated herein by hydrolysis of 4 to give the hemiacetal 5. Because
the starting anhydrosugar 1 is an easily obtainable compound, the
proposed set of reactions consisting of oxetane ring opening in 1
with Grignard reagents represent a simple and general synthetic
route toward terminal 1,2,3,4-alkyl tetraols and terminal 1,2,4-al-
kyl triols.
1.1.1. 1,2-O-Cyclohexylidene-5-deoxy-a-D-xylofuranose (2)
Oxetane 1 (2.30 g, 10.8 mmol) was dissolved in THF (30 mL). At
rt, LiAlH4 (0.6 g, 15.7 mmol) was added to the solution in portions
and the reaction mixture was heated at reflux for 3 h. After cooling,
10% sulfuric acid was added until the separation of layers; the or-
ganic layer was dried and evaporated to leave an oil that solidified
upon standing. Crystallization from iso-octane gave 2 (2.25 g, 97%),
mp 89.6 °C; EIMS: 213.9 (M+, 28), 184.9 (32), 171 (58), 99.1 (58),
81.0 (62); Anal. Calcd for C11H18O4: C, 61.66; H, 8.46. Found: C,
61.53; H, 8.43.
1. Experimental
1.1. General methods
Melting points were determined on a Büchi MP 50 apparatus
and are not corrected. The NMR spectra were recorded on a Bru-
ker AC250E instrument in CDCl3 using (CH3)4Si as an internal
standard and are presented in Tables 1 and 2. Mass spectra were
Table 1
1H NMR Spectral data for 2–5
Chemical shifts (ppm), signal shape, and integrals
of xylose protons
Coupling constants (Hz)
Other signals
H-1
H-2
H-3
H-4
H-5
J1,2 J3,4 J4,5 Other
2
5.88, d, 1H
4.49,
d, 1H
4.74,
d, 1H
4.49,
d, 1H
4.62,
d, 1H
4.48,
d, 1H
4.65,
d, 1H
3.99,
4.29,
1.28,
3.8 2.4 6.3 J3,OH 6.1
3.8 2.8 5.4
3.8
1.31–1.74, m, 10H, cya; 2.13, br d, OH;
dd, 1H qd, 1H d, 3H
4.88,
d, 1H
4.06–4.13, m,
2H
3.80,
d, 1H
4.03,
dd, 1H td, 1H
4.77,
2a 5.91, d, 1H
4.44,
qd, 1H d, 3H
1.32,
1.35–1.76, m, 10H, cya; 3.07, s, 3H, Me from Ms;
1.18–1.78, 24H, 12 Â CH2 alkyl and cya; 0.88, t, 3H, CH3;
3
5.89, d, 1H, H-1
3a 5.93, d, 1H
4.14,
td, 1H
4.12,
3.9 3.0 7.0 CH2Ph,
Jgem12.0
3.8 2.4 6.6 J3,OH 6.6;
0.91, t, 3H, CH3; 1.14–1.85, m, 24H, 12 Â CH2 alkyl and cya; 4.50
and 4.73, 2d, 2H, CH2Ph; 7.25–7.41, m, 5H, Ar-H;
1.32–1.84, m, 15H, 7 Â CH2 and OH; 2.68, t, CH2Ph; 7.15–7.38, m,
4
5.89, d, 1H
CH2Ph, J 6.8 5H, Ar-H;
4a 5.89, d, 1H
4.14,
td, 1H
3.8 2.7 6.7
4.1
1.29–1.78, m, 14H, 7 Â CH2; 2.55, t, 2H, CH2Ph; 7.08–7.97, m,
d, 1H
10H, Ar-H;
5
5.14, s, 1H, b anomer
5.41, d, 1H, anomer
4.01–4.29, m, 6H, both
anomers
1.50–1.83, m, CH2-6 and CH2-5 2.63–2.76, m, CH2-7, both
anomers
a
a
cy = cyclohexylidene.

P. Hadzic, M. Popsavin / Carbohydrate Research 345 (2010) 543–546
545
Table 2
13C NMR data for 2–5 in CDCl3.
Chemical shifts (ppm) of xylose carbons
Other signals
C-1
C-2
C-3
C-4
C-5
2
103.86
85.00 76.30 75.92 12.70 23.49, 23.82,23.85, 35.54 and 36.14 cyclohexylidene; 112.02 Cq
2a 103.89
103.75
3a 104.11
83.20 83.00 74.64 13.13 23.38, 23.71, 24.70, 35.55, 36.04, 5 Â CH2; 38.28, Me from Ms; 112.77 Cq
3
84.78 75.43 80.21
81.68 81.94 80.33
14.09 Me, 22.64, 23.55, 23.89, 24.93, 26.03, 27.60, 29.19, 29.43, 29.72, 31.85, 35.62, 36.19 (12 Â CH2); 112.03 Cq
14.04, CH3; 22.58, 23.51, 23.83, 24.89, 25.99, 27.76, 29.13, 29.39, 29.68, 31.80, 35.64, 36.24, 12 Â CH2; 71.59, CH2Ph;
127.53, 127.68, 128.30, 137.65 (Ar-C); 111.73 Cq
4
103.74
84.74 75.43 80.05
82.76 82.84 78.68
23.53, 23.86, 24.89, 27.14, 27.76, 35.60, 35.81, 36.18 (8 Â CH2, cyclohexylidene and CH2-5, CH2-6, CH2-7); 125.81,
128.39, 141.94 (Ar-C); 112.06 Cq
4a 103.70
23.34, 23.69, 24.70, 27.18, 27.24, 35.51, 35.53, 36.01, 8 Â CH2; 125.72, 127.71, 128.18, 128.26, 129.30, 134.09,
135.92, 141.58 (Ar-C); 112.62 Cq
29.95, 30.14, 30.65, 31.27, 37.71 (CH2-5, CH2-6, CH2-7, both anomers); 77.97, 78.75, 79.08, 81.92, 83.68, 84.73 (C-2,
C-3, C-4, both anomers); 124.63, 124.69, 128.77, 131.42, 145.65 Ar-C both anomers;
5
98.38a
104.44b
a
a
-Anomer.
b
b-Anomer.
1.1.2. 1,2-O-Cyclohexylidene-5-deoxy-3-O–methanesulfonyl-
-xylofuranose (2a)
A solution of methanesulfonyl chloride (0.75 g, 6.5 mmol) in
a
-
vacuum distillation gave 3a (0.8 g, 62%) as an oil, CI MS:
M + 1 = 403 (100); Anal. Calcd for C25H38O4: C, 74.59; H, 9.51.
Found: C, 74.54; H, 9.29.
D
CHCl3 (5 mL) was added to a solution of 2 (1.07 g, 5 mmol) in a
mixture of CHCl3 (5 mL) and pyridine (5 mL) in one portion at rt.
After standing for 12 h, CHCl3 (100 mL) was added and the mixture
was washed with 10% aqueous hydrochloric acid and then with
H2O. The organic solution was dried and evaporated to a colored
oil that solidified on standing. Crystallization from isopropanol
gave 2a (0.92 g, 63%), as silky crystals, mp 94 °C; EIMS: 291.8
(M+, 24), 248.8 (54), 153.0 (30), 81.1 (25), 79.1 (22), 69.1 (20),
55.1 (100); Anal. Calcd for C12H20O6S: C, 49.30; H, 6.90. Found: C,
49.46; H, 6.92.
1.1.4. 1,2-O-Cyclohexylidene-5-deoxy-5-C-(2-phenylethyl)-a-D-
xylofuranose (4)
A crystal of iodine was added to magnesium turnings (2.83 g,
116 mmol) covered with dry Et2O. A solution of 2-bromoethyl
benzene (20.35 g, 110 mmol) in dry Et2O (80 mL) was added drop-
wise to the suspension. The reaction started easily and was com-
plete after 2 h at reflux. The addition of oxetane 1 (23.3 g,
110 mmol in 80 mL of Et2O) then commenced. During the addi-
tion, the reaction mixture tended to solidify and at this stage,
dry benzene (80 mL) was added and the reaction flask was heated
to distill off Et2O until the temperature of distillate reached 55 °C.
After 5 h at reflux, the reaction was assumed to be complete. Then
aqueous hydrochloric acid (10%) was added until the organic and
H2O layers separated. The organic layer was washed with H2O and
concentrated. The addition of petroleum ether, followed by cool-
ing, resulted in crystallization of 4 (21.5 g, 61%), mp 100–102 °C.
EIMS: 317.9 (M+, 53), 288.9 (26), 274.9 (29), 203.1 (36), 185.1
(100), 157.1 (58); Anal. Calcd for C19H26O4: C, 71.67; H, 8.23.
Found: C, 71.36; H, 8.30.
1.1.3. 1,2-O-Cyclohexylidene-3-O-benzyl-5-deoxy-5-C-(2-
heptyl)-a-D-xylofuranose (3a)
A crystal of iodine was added to magnesium turnings (1.46 g,
60 mmol) in dry THF (25 mL) and then heptyl bromide (10.74 g,
60 mmol) was added while stirring, in one portion at rt. Vigorous
reaction started spontaneously and then the reaction mixture
was cooled externally, and afterward was heated at reflux. The
complete dissolution of magnesium was evident in 40 min. Oxe-
tane 1 (8.5 g, 40 mmol) in THF (10 mL) was added dropwise. After
the reaction mixture was heated at reflux for another 3 h, the reac-
tion was assumed to be complete. The mixture was quenched by
the addition of aqueous hydrochloric acid (10%) and products were
extracted with hexane (3 Â 80 mL). Evaporation of the combined
extracts left a mass with a soapy-waxy appearance on standing.
TLC showed a spot corresponding to a new compound and another
faint spot that could be attributed to heptanol by comparison with
an analytical sample. Attempts to crystallize the product were
unsuccessful. The crude product was dissolved in hot CH3OH
(100 mL), and H2O (150 mL) was added. After cooling and standing
overnight a white soapy deposit (9 g) was formed. The mass was
again dissolved in hot isopropanol (100 mL), and H2O (150 mL)
was added. Cooling resulted in white soapy-waxy deposit (8.2 g,
66% calculated on 1), which was a single component according to
TLC and melted at 40–45 °C, after drying. However, satisfactory
elemental analysis could not be obtained. IR: 3450 sharp, hydroxyl
group; 1H NMR and 13C NMR data were in accordance with pro-
posed structure of 1,2-O-cyclohexylidene-5-deoxy-5-C-(2-hep-
1.1.5. Attempted sulfonylation of 4: 3-O-Benzenesulfonyl-1,2-
O-cyclohexylidene-5-deoxy-5-C-(2-phenylethyl)-a-D-
xylofuranose (4a)
To a solution of 4 (2.0 g, 6.2 mmol) in a mixture of CHCl3 (5 mL)
and pyridine (3 mL) was added a solution of benzenesulfonyl chlo-
ride (1.2 g, 6.7 mmol) in CHCl3 (5 mL) in one portion at rt. The reac-
tion mixture was left for 48 h at 4 °C and then the mixture was
diluted with CHCl3 (100 mL), washed with hydrochloric acid
(10%), and then with H2O. Drying and evaporation of the solvent
left 4a as an oil (2.2 g, 77%). Satisfactory proof of 4a purity (ele-
mental analysis) could not be obtained, but MS and NMR analysis
clearly support the proposed structure. EIMS C25H30O6S: 458.0 (M+,
85), 414.9 (48), 309.9 (92), 257.0 (43), 203.1 (64), 185.1 (100).
1.1.6. 5-Deoxy-5-C-(2-phenylethyl)-D-xylofuranose (5)
A solution of compound 4 (5.0 g, 16 mmol) was hydrolyzed in a
mixture of acetic acid (40 mL) and H2O (60 mL) at reflux for 3 h.
Then, H2O (150 mL) was added and the reflux condenser was chan-
ged for distillation. Part of the reaction mixture was distilled off
(ꢀ100 mL) to diminish the quantity of acetic acid. The remaining
was neutralized (aq sodium carbonate) and extracted with EtOAc
(4 Â 50 mL). Concentration of combined extracts and addition of
petroleum ether resulted in crystallization of pure 5 (2.83 g,
74%), mp 108 °C; CI MS: 239 (M++1, 42), 221 (45), 203 (40), 185
tyl)-a-D-xylofuranose (3), (see Tables 1 and 2).
To the mixture of crude 3 (1 g, 3.2 mmol), dimethylsulfoxide
(5 mL) and benzyl chloride (0.6 g, 48 mmol) was added finely
grounded sodium hydroxide (0.3 g) in one portion. The reaction
mixture was stirred at 50 °C for 1 h, then poured into H2O and ex-
tracted with CH2Cl2 (3 Â 50 mL). Combined extracts were washed
with H2O (2 Â 50 mL), dried, and evaporated to an oil. High

546
P. Hadzic, M. Popsavin / Carbohydrate Research 345 (2010) 543–546
4. Bell, P. C.; Bergsma, M.; Dolbnya, I. P.; Bras, W.; Stuart, M. C. A.; Rowan, A. E.;
Feiters, M. C.; Engberts, J. B. F. N. J. Am. Chem. Soc. 2003, 125, 1551–1558.
5. Johnsson, M.; Wagenaar, A.; Stuart, M. C. A.; Engberts, J. B. F. N. Langmuir 2003,
19, 4609–4618.
(100), 161 (65); Anal. Calcd for C13H18O4: C, 65.53; H, 7.61. Found:
C, 65.45; H, 7.56
6. Pilakowska-Pietras, D.; Lunkenheimer, K.; Piasecki, A. J. Colloid Interface Sci.
2004, 271, 192–200.
References
7. Inch, T. D. Tetrahedron 1984, 40, 3161–3213.
8. Kawana, M.; Kuzuhara, H.; Emoto, S. Bull. Chem. Soc. Jpn. 1981, 54, 1492–1504.
1. (a) Holmberg, K. Curr. Opin. Colloid Interface Sci. 2001, 6, 148–159; (b)
Stubenrauch, C. Curr. Opin. Colloid Interface Sci. 2001, 6, 160–170.
2. Von Jagov, G.; Link, T. A.; Schägger, H. Purification Strategies for Membrane
Proteins. In Membrane Protein Purification and Crystallization, A Practical Guide;
Hunte, C., von Jagow, G., Schägger, H., Eds., 2nd ed.; Academic Press: New York,
1994; pp 10–16.
´
´
9. Hadzic, P.; Vukojevic, N.; Popsavin, M.; Canadi, J. J. Serb. Chem. Soc. 2001, 66, 1–
8.
10. Ruotsalainen, V.; Palosari, V.; Virtanen, P. O. I. Suom. Kem. B 1972, 45, 40–42.
11. Hadzic, P.; Vukojevic, N. J. Serb. Chem. Soc. 2001, 66, 289–295.
12. Hadzic, P.; Vukojevic, N. Carbohydr. Res. 2003, 338, 1243–1249.
3. Privé, G. G. Struct. Biol. Membrane Prot. 2007, 41, 388–397.