Practical synthesis of (-)-1-amino-1-deoxy-myo-inositol from achiral precursors

Article abstract of DOI:10.1016/j.carres.2007.04.024

A new synthesis of enantiomerically pure 1-amino-1-deoxy-myo-inositol is reported. The route described employs p-benzoquinone, an achiral compound, as the starting material to give conduritol B tetraacetate in three steps. Kinetic resolution of this compound using a palladium catalyst with a chiral ligand allows access to a conduritol B tetraester in high enantiomeric excess. This compound is transformed into tetrabenzyl conduritol B epoxide, which is regioselectively opened with azide to give the key azidocyclitol. Final transformation into (-)-1-amino-1-deoxy-myo-inositol hydrochloride is achieved in four synthetic steps. This sequence allows the synthesis of this compound in high enantiomeric purity in a semi-preparative scale.

Full text of DOI:10.1016/j.carres.2007.04.024

Carbohydrate Research 342 (2007) 1947–1952
Note
Practical synthesis of (À)-1-amino-1-deoxy-myo-inositol
from achiral precursors
Patricia Gonzalez-Bulnes,a Josefina Casas,a Antonio Delgadoa,b and Amadeu Llebariaa,*
a
`
`
Research Unit on Bioactive Molecules (RUBAM), Departament de Quı´mica Organica Biologica,
Institut d’Investigacions Quı´miques, i Ambientals de Barcelona (IIQAB–CSIC), Jordi Girona 18–26, 08034 Barcelona, Spain
b
`
´
`
Universitat de Barcelona, Facultat de Farmacia, Unitad de Quımica Farmaceutica (Unitat Associada al CSIC),
Avda., Joan XXIII, s/n, 08028 Barcelona, Spain
Received 7 March 2007; received in revised form 13 April 2007; accepted 22 April 2007
Available online 5 May 2007
Abstract—A new synthesis of enantiomerically pure 1-amino-1-deoxy-myo-inositol is reported. The route described employs p-
benzoquinone, an achiral compound, as the starting material to give conduritol B tetraacetate in three steps. Kinetic resolution
of this compound using a palladium catalyst with a chiral ligand allows access to a conduritol B tetraester in high enantiomeric
excess. This compound is transformed into tetrabenzyl conduritol B epoxide, which is regioselectively opened with azide to give
the key azidocyclitol. Final transformation into (À)-1-amino-1-deoxy-myo-inositol hydrochloride is achieved in four synthetic steps.
This sequence allows the synthesis of this compound in high enantiomeric purity in a semi-preparative scale.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Keywords: Aminoinositol; Conduritol B; myo-Inositol; Aminoconduritol; Enantiomer; Kinetic resolution
Carbohydrates and their derivatives in which one or
more hydroxyl groups are replaced by amino functions
represent a remarkable class of compounds whose
chemistry and biology has received intense scrutiny.1
Phosphatidylinositol (PI) and phosphorylated inositol
derivatives, which are among carbohydrate-related com-
pounds, play a crucial role in biology and are involved
in important cell processes.2 To mention only a few
examples, they are involved in the release of Ca2+, and
can affect glycosidase activity or display antibiotic activ-
ity.3 The relevance of inositol and their derivatives in cell
function and regulation makes them very attractive
compounds for research.
large number of inositol analogues3,5–10 reported to date
clearly indicates the increasing importance of research
activities in this field. Moreover, recent work in our lab-
oratory on aminocyclitol derivatives as glycolipid
mimetics has resulted in the discovery of several new
glucocerebrosidase enzyme inhibitors.11,12 In this con-
text, the preparation of 1-amino-1-deoxy-myo-inositol
(À)-1 having the natural myo-inositol configuration
to obtain new analogues for biological assays was
necessary.
Two syntheses of this compound have been previously
described in the literature. The approach of Iinuma
et al.13 starts with (À)-D-chiro-inositol, which is selec-
tively oxidized with O2 and catalytic platinum black to
L-myo-inosose,14 followed by a two-step transformation
into amine (À)-1 via the corresponding oxime; this route
gives the product in 15% overall yield from inositol. In
an approach published by the Altenbach group,10 p-
benzoquinone is transformed into 2,3-dibromocyclo-
hex-5-ene-1,4-diol diacetate, which is subjected to a
kinetic resolution in a multigram scale with porcine
pancreatic lipase to give a mixture of the corresponding
The synthesis of molecules in which one or more ino-
sitol hydroxyl groups are modified or removed would
make possible not only a deeper study of their struc-
ture–activity relationships but also more thorough re-
search in the field of aminoglycoside antibiotics.4 The
*
Corresponding author. Fax: +34 932045904; e-mail: alsqob@
cid.csic.es
0008-6215/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2007.04.024

1948
P. Gonzalez-Bulnes et al. / Carbohydrate Research 342 (2007) 1947–1952
1R,2S,3S,4R-diol and its enantiomeric 1S,2R,3R,4S
diacetate. This compound can be separated and purified
by crystallization in 26% yield from p-benzoquinone
over three steps.15,16 Enantiomerically pure (1S,2R,
3R,4S)-1,4-diacetoxy-2,3-dibromocyclohex-5-ene is next
converted into aminociclitol (À)-1 in eight steps in 19%
overall yield (4.9% yield from p-benzoquinone).10
Here, we report the synthesis of enantiomerically pure
(À)-1-amino-1-deoxy-myo-inositol ((À)-1) hydrochlo-
ride by a new 12-step synthetic sequence in 5% overall
yield starting from p-benzoquinone. The enantiodis-
criminating step of this synthesis involves the resolution
of racemic conduritol B tetracetate ( )-2 using the
palladium-catalyzed kinetic resolution previously devel-
oped by Trost and Hembre.17 This catalytic asymmetric
reaction is currently used in our group for the prepara-
tion of the enantiomers of conduritol B from ( )-2.
The proposed retrosynthetic analysis (Scheme 1)
involves the regioselective nucleophilic attack of azide
nucleophile on epoxide (+)-6 to give azidoalcohol (+)-
7, which can be transformed into the desired final com-
pound (À)-1 after O-benzyl deprotection. Preparation of
epoxide (+)-6 would be straightforward by introducing
the protecting O-benzyl groups in epoxide (+)-5, which,
in turn, could be obtained from conduritol B (+)-4
derived from tetraester (+)-3. This compound can be
obtained enantiomerically pure through the kinetic
resolution of conduritol B tetraacetate ( )-2 using a
chiral palladium catalyst, as described by Trost and
co-workers.17,18
The advantages of this approach are manifold, be-
cause (1) it makes possible the synthesis of enantiomeri-
cally pure products using achiral starting materials; (2)
the compound responsible for the enantiodiscriminating
step can be used in catalytic amounts; (3) both enantio-
mers are available through this approach, and (4) it
opens the way to myo-inositol analogues by coupling
the amino group with different functionalities to obtain
new molecules for biological studies, a goal which is cur-
rently underway in our research group.
The synthetic sequence started with p-benzoquinone,
which was converted in three steps into racemic condur-
itol B tetraacetate ( )-2 in 90% yield following a re-
ported procedure (Scheme 2).18,19 When tetraacetate
( )-2 was subjected to palladium-catalyzed kinetic reso-
lution18 using chiral diphosphine (S,S)-8, the (+)-enan-
tiomer of 2 was converted to pivalate (+)-3, whereas
the (À)-2 enantiomer of the tetraacetate remained unre-
acted. A slight modification of Trost’s original proce-
dure was introduced, which consisted of the in situ
preparation of the catalyst precursor, bis(p-allyl)-palla-
dium dichloride, by reacting allyltrimethylsilane and
PdCl2(CH3CN)2 in acetonitrile.20 This modification
gave, in our hands, more reproducible results by elimi-
nating the problems associated to the manipulation of
this sensitive compound.
According to Trost et al.,21 kinetic resolution using
this ligand results from the different rates at which the
enantiomers of tetraacetate ( )-2 form the chiral p-allyl-
palladium complex with chiral diphosphine (S,S)-8, the
key intermediate of the transformation, and also from
the different reactivity of the allylic pivalate ester com-
pared with the corresponding allylic acetates. In our
hands, the reaction worked nicely on a multigram scale
and the progress of the reaction could be monitored by
HPLC. After chromatographic separation of the prod-
uct pivalate, (+)-3 was isolated by column chromatogra-
phy in 43% yield from unreacted tetraacetate (À)-2. The
enantiomeric purity of the product was checked by chi-
ral HPLC. The enantiomeric excesses determined for
pivalate (+)-3 and confirmed at a later stage for epoxide
(+)-6 were found to be always higher than 96%.
The transformation of pivalate (+)-3 into amine (À)-
11 (Scheme 3) was achieved by introducing minor
changes to the procedure described by Serrano et al.,22
which was used to synthesize 11 in racemic form. The
sequence initially involved the deprotection of hydroxyl
groups in (+)-3 using Zemplen conditions to give (+)-4
in 99% isolated yield. Epoxidation of this tetraol with
MCPBA afforded (+)-5 in 88% yield, followed by pro-
tection of the hydroxyl groups as benzyl ethers furnish-
ing compound (+)-6 in 65% isolated yield. Epoxide
opening of (+)-6 with NaN3/LiClO4 gave the desired
regioisomer (+)-7 in 61% yield.
The free hydroxyl group in cyclitol azide (+)-7 was
transformed into its mesylate (+)-9 in 88% yield and this
compound was used to perform the stereochemical
inversion to azidocylitol (+)-10. Following our previous
procedure,22 the inversion was attempted by heating
(+)-9 in DMF in a sealed tube at 140 ꢁC for prolonged
times. Whereas this reaction performed well in some
cases, in others we observed a slow transformation,
recovering, in part, the unreacted starting mesylate.
After conducting different optimization experiments,
we discovered that the presence of traces of water in
the reaction media was essential for this transformation.
This is in contrast with our previous report,22 where it
was claimed that this reaction was independent of the
water content. The only explanation found for this
apparent contradiction is the working scale. Thus, be-
cause our former experiments were carried out at less
than 0.2 mmol scale, some adventitious water could
have been present in the reaction media, making the
This reaction was accompanied with minor amounts of the regio-
isomer (below), which could be removed by column chromatography.
OBn
BnO
BnO
OH
N3
OBn

P. Gonzalez-Bulnes et al. / Carbohydrate Research 342 (2007) 1947–1952
1949
OAc
OAc
OBn
OAc
O
O
OH
OH
OBn
AcO
AcO
BnO
BnO
AcO
AcO
HO
HO
HO
HO
NH2
OH
BnO
BnO
N3
O
O
OH
OBn
(+)-6
OPiv
OH
(+)-5
OH
(-)-1
OBn
(+)-7
( )-2
(+)-3
Scheme 1. Retrosynthetic analysis of (À)-1-amino-1-deoxy-myo-inositol.
material. However, we finally succeeded in using this
catalyst in a mixture of THF and aqueous concentrated
HCl (ꢀ99:1 v/v) as the solvent under hydrogen pressure
(3 atm). Under these conditions, pure hydrochloride
(À)-1 was obtained in 96% isolated yield after a simple
filtration of the heterogeneous palladium catalyst
(Scheme 3).
In summary, a new procedure for the synthesis of an
inositol analogue where the 1-hydroxyl group present in
natural myo-inositol is replaced with an amino group is
reported. This procedure is currently employed in our
group to obtain (À)-1-amino-1-deoxy-myo-inositol
hydrochloride starting from p-benzoquinone in semi-
preparative scale.
CH3CN
SiMe3
PdCl2(CH3CN)2
+
Pd
Cl
25ºC, 1h
2
O
O
NH HN
OAc
OAc
OAc
OAc
OAc
PPh2 Ph2P
AcO
AcO
AcO
AcO
AcO
AcO
(S,S)-8
+
[η3-allylPdCl]2
n-Bu4NBr
tC4H9CO2H, NaOH
H2O, CH2Cl2
OPiv
(-)-2
(+)-3
( )-2
Scheme 2. Kinetic resolution of conduritol B tetraacetate.
addition of extra amounts of water unnecessary to
accomplish this transformation. In any case, when (+)-
9 was heated in DMF containing 2% water, we were able
to isolate (+)-10 with consistent yields around 70%.
The next step was the reduction of the azido group
with LiAlH4 in THF, which proceeded to give the prod-
uct in 70% isolated yield. Once the tetra-O-benzylated
amine (À)-11 was obtained, deprotection of the hydro-
xyl groups to complete the synthesis of the desired
amino inositol analogue (À)-1 was attempted. Hydrog-
enolysis using Pd/C at 1 atm in different solvents gave
poor results due to the low reactivity of the starting
1. Experimental
1.1. General methods
Solvents were distilled prior to use and dried by stan-
dard methods. FT-IR spectra are reported in m, cmÀ1
.
1H and 13C NMR spectra were obtained in CDCl3,
1
CD3OD, or D2O solutions at 500 or 300 MHz (for H)
and 125, 100, or 75 MHz (for 13C). Signal assignment
was based on 2D NMR experiments (COSY, HMQC).
Melting points were uncorrected. Chemical shifts were
OAc
OH
OH
AcO
AcO
HO
HO
NaOMe/MeOH
m-CPBA
HO
HO
NaH/ BnBr
DMF
O
MeOH
OPiv
OH
OH
(+)-5
(+)-3
(+)-4
OBn
OBn
OBn
BnO
BnO
LiClO4/ NaN3
CH3CN
CH3SO2Cl
BnO
BnO
N3
O
DMF, 140ºC
BnO
BnO
N3
OH
O
O
S
O
THF/TEA
CH3
OBn
OBn
OBn
(+)-7
(+)-9
(+)-6
OBn
OH
OBn
Cl
BnO
BnO
NH2
OH
HO
HO
NH3
OH
BnO
BnO
N3
LiAlH4
THF
Pd/C H2
THF/ aeqHCl
OH
OBn
OH
OBn
(-)-1.HCl
(+)-10
(-)-11
Scheme 3. Synthesis of (À)-1-amino-1-deoxy-myo-inositol hydrochloride from conduritol B tetraester (+)-3.

1950
P. Gonzalez-Bulnes et al. / Carbohydrate Research 342 (2007) 1947–1952
reported in delta (d units, parts per million (ppm) rela-
tive to the singlet at 7.24 ppm of CDCl3 for H and in
ppm relative to the center line of a triplet at 77.0 ppm
of CDCl3 for 13C. The ESIMS spectra were recorded
on a Waters LCT Premier Mass spectrometer.
[M+Na]+. Found: 379.1367; HPLC: Chiralpakꢂ IA
(0.46 · 25 cm) eluting with heptane/iPrOH (90:10),
0.7 mL/min, k = 254;nm; tR (+)-3 = 7.9 min, tR (À)-
3 = 8.5 min.
1
1.3. (+)-(1S,2R,3R,4S)-5-Cyclohexene-1,2,3,4-tetraol (4)
1.2. (+)-(1S,2R,3R,4S)-1-Pivaloxy-2,3,4-triacetoxy-5-
cyclohexene (3)
Na (35 mg) was dissolved in anhydrous MeOH
(180 mL) in a three-necked round bottomed flask under
a nitrogen atmosphere. Pivalate (+)-3 (6.1 g, 17.8 mmol)
was added and the reaction mixture was stirred at room
temperature for 24 h. After 24 h the solvent was re-
moved in vacuo and the residue was dried on a vacuum
pump. The solid was then dissolved in MeOH (80 mL)
and neutralized with 1 N HCl. The solvent was removed
furnishing product 4 as a white solid (2.78 g; 17.8 mmol,
In a Schlenk flask under nitrogen atmosphere, 198 mg
(0.76 mmol) of PdCl2(CH3CN)2 was suspended in
10 mL of dry CH3CN. Allyltrimethylsilane in CH3CN
(0.5 M, 1.52 mL, 0.76 mmol) was added via a syringe
and the reaction mixture was stirred at 25 ꢁC for 2 h.
The solvent was removed in vacuo and the subsequent
kinetic resolution of conduritol B was made assuming
that the reaction of the synthesis of (g3-allyl)-palladium
dichloride had been quantitative.
99%).
Mp = 175–177 ꢁC
(lit.23
mp = 172–173);
[a]D +164.5 (c 1.1, MeOH); lit.23 [a]D +183 (c 1.1,
MeOH); IR(KBr, m): 3294, 2904, 2859, 1458, 1374,
1035; 1H NMR (d, 500 MHz): 3.42–3.48 (m, 2H),
4.14–4.20 (m, 2H), 5.61 (s, 2H); 13C NMR (d,
125 MHz): 70.0, 75.5, 129.2; ESIMS: m/z calcd for
C6H10O4Na: 169.0478 [M+Na]+. Found: 169.0478.
Independently, a mixture of ( )-1,2,3,4-tetra-O-acet-
ylconduritol B ( )-2 (12 g, 38.2 mmol), (S,S)-ligand 8
(1.57 g, 2.28 mmol), tetrahexylammonium bromide
(3.32 g, 7.64 mmol) and pivalic acid (3.12 g, 30.6 mmol)
was degassed by evacuating and purging with argon
(3·). Under nitrogen, dry CH2Cl2 (60 mL) was added
and the mixture was stirred until the solution was com-
plete. The liquid was transferred via a canuula to the
Schlenk flask containing the catalyst (0.76 mmol) under
argon. Finally 40 mL of 0.5 M NaOH, previously de-
gassed, was added via a syringe. The reaction mixture
was stirred at 30 ꢁC monitoring the disappearance of
(+)-2 tetraacetate by HPLC (OL244 chiral column,
0.46 · 12.5 cm eluting with 90:10 heptane/isopropanol,
1.4. (+)-(1R,2R,3S,4S,5R,6S)-2,3,4,5-Tetraol-7-oxabi-
cyclo[4.1.0]heptane (5)
(+)-(1S,2R,3R,4S)-5-Cyclohexene-1,2,3,4-tetraol 4 (1.6 g,
11.1 mmol) was dissolved in MeOH (150 mL), MCPBA
(10.0 g, 45 mmol) was added and the reaction mixture
was stirred at room temperature overnight. The solvent
was removed in vacuo and the residue was triturated
with CH2Cl2 to dissolve excess MCPBA and 3-chloro-
benzoic acid. The remaining white solid was washed
with CH2Cl2 and dried at vacuum to give essentially
pure (+)-5 (1.46 g, 9.8 mmol, 88%). Mp = 155–157 ꢁC
(lit.24 mp = 158–160 ꢁC); [a]D +57.3 (c 0.5, H2O); lit.24
[a]D +66 (c 0.6, H2O); IR (KBr, m): 3362, 1408, 1346,
0.7 mL/min);
(+)-(1S,2R,3R,4S)-1-pivaloxy-2,3,4-tri-
acetoxy-5-cyclohexene tR = 4.54 min, (À)-(1R,2S,3S,4R)-
1,2,3,4-tetraacetoxy-5-cyclohexene tR = 5.67 min, (+)-
(1S,2R,3R,4S)-1,2,3,4-tetraacetoxy-5-cyclohexene tR =
6.62 min). When the transformation of (+)-2 into (+)-
3 was finished (8–12 h are usually required), the reaction
mixture was washed with 20% aq K2CO3 (60 mL) and
the aqueous layer extracted with diethyl ether
(2 · 40 mL). The combined organic layers were dried
over MgSO4, filtered through a plug of silica to remove
the catalyst and concentrated in vacuo to give a white
solid (13.1 g). The crude product obtained was filtrated
through a plug of silica eluting with hexanes/EtOAc
(4:1) to afford pivalate (+)-3 (5.92 g, 17.3 mmol, 43%
yield, ee >96%) and unreacted tetracetate (À)-2 (5.75 g,
18 mmol, 48% yield, 90% ee). Compound (+)-3:
mp = 93 ꢁC; (lit.18 mp = 110 ꢁC); [a]D +132.0 (c 1.1,
CHCl3); IR (m, film): 2980, 1758, 1369, 1215, 1150,
1
1268, 1101, 1046, 1016, 939, 849, 812, 719; H NMR
(d, 500 MHz): 3.14–3.17 (m, 3H), 3.35–3.36 (m, 1H),
3.66–3.68 (m, 1H), 3.82–3.84 (m, 1H); 13C NMR (d,
125 MHz): 56.7, 57.4, 70.5, 70.7, 71.5, 74.9; ESIMS:
m/z calcd for C6H10O5Na: 185.0426 [M+Na]+. Found:
186.0429.
1.5. (+)-(1S,2R,3S,4S,5R,6R)-2,3,4,5-Tetrakisbenzyl-
oxy-7-oxabicyclo[4.1.0]heptane (6)
Sodium hydride (2.4 g, 60% dispersion in mineral oil,
99.9 mmol) was placed in a three-necked round bot-
tomed flask under nitrogen atmosphere and the hydride
was washed several times with pentane to remove the
mineral oil. In another flask under nitrogen atmosphere,
epoxide (+)-5 (1.8 g, 11.1 mmol) was dissolved in anhy-
drous DMF (60 mL) and was then transferred via a
canuula to the flask containing the hydride. The mixture
1
1032, 964; H NMR (d, 500 MHz): 1.18 (s, 9H), 2.02
(s, 3H), 2.04 (s, 3H), 2.07 (s, 3H), 5.53–5.41 (m, 2H),
5.55–5.52 (m, 2H), 5.70 (s, 2H); 13C NMR (d,
125 MHz): 20.5, 20.6, 20.8, 26.8, 26.8, 26.8, 38.7, 70.9,
71.2, 71.2, 71.6, 127.1, 127.6, 169.6, 169.9, 170.2,
177.7; ESIMS: m/z calcd for C17H24O8Na: 379.1369

P. Gonzalez-Bulnes et al. / Carbohydrate Research 342 (2007) 1947–1952
1951
was stirred at 30 ꢁC for 30 min, BnBr (13.2 mL,
111 mmol) was then added dropwise and the reaction
mixture was stirred at 30 ꢁC overnight. Then the solvent
was removed in vacuo, the residue was dissolved in
water and the aqueous layer was extracted with diethyl
oxide. The organic layer was washed with brine, dried
over Na2SO4 and concentrated to give a yellow oil
(11 g), which was purified by flash chromatography
using a mixture of hexane/EtOAc (5:1). Product (+)-6
was obtained as a white solid (4.0 g, 7.6 mmol, 68%).
Mp 103–105 ꢁC (lit.25 mp 103–105 ꢁC); [a]D +32.9 (c
1.2, CHCl3) (lit.25 [a]D +33 (c 1.2, CHCl3); IR (film, m):
after 26 h the solvent was removed in vacuo, the residue
was taken in Et2O (40 mL), filtered and the solid was
washed several times with Et2O. The filtrate and wash-
ings were concentrated to give the crude azidomesilate,
which was purified by chromatography on silica eluting
with hexane/EtOAc (5.5:1) to afford compound (+)-9 as
a colorless oil (2.2 g, 3.2 mmol, 81%). [a]D +27.3 (c 1,
CHCl3); IR (film, m, cmÀ1): 2928, 2856, 2111, 1456,
1358, 1177, 1063, 953, 821, 742, 698; 1H NMR (d,
500 MHz, CDCl3): 3.06 (s, 3H), 3.49 (t, J = 9.3 Hz,
1H), 3.57–3.66 (m, 4H), 4.45 (t, J = 9.7 Hz, 1H), 4.85–
4.94 (m, 8H), 7.28–7.38 (m, 5H); 13C NMR (d,
100 MHz, CDCl3): 39.2, 64.6, 75.8, 75.9, 76.0, 76.1,
80.2, 80.6, 80.7, 82.0, 82.9; ESIMS: m/z calcd for
C35H37O7NaS: 666.2250 [M+Na]+. Found: 666.2250.
1
3028, 2867, 1491, 1450, 1370, 1088, 851, 690; H NMR
(d, 500 MHz, D2O): 3.22 (d, J = 3.9 Hz, 1H), 3.34–
3.35 (m, 1H), 3.48–3.52 (m, 1H), 3.48–3.52 (m, 1H),
3.63–3.66 (m, 1H), 3.91–3.93 (m, 2H), 4.72–4.88 (m,
8H), 7.28–7.42 (m, 20H); 13C NMR (d, 75 MHz,
CDCl3): 53.9, 55.2, 73.0, 73.2, 75.5, 75.9, 79.0, 79.2,
79.3, 83.4, 127.5–128.5, 137.6, 138.2, 138.5, 138.5;
ESIMS: m/z calcd for C34H35O5: 523.2484 [M+H]+.
Found: 523.2489; HPLC: Chiralpakꢂ IA (0.46 ·
25 cm) eluting with heptane/iPrOH (90:10), 0.7 mL/
min, k = 254 nm; tR (+)-6 = 17.2 min, tR (À)-6 =
16.1 min.
1.8. (+)-(1S,2R,3S,4R,5R,6S)-2-Azido-3,4,5,6-tetrakis-
benzyloxycyclohexanol (10)
Mesylate (+)-9 (6.1 g, 9.5 mmol) was treated with DMF
(75 mL) and water (1.5 mL) in a sealed tube at 140 ꢁC
for 96 h. The solvent was then removed in vacuo to give
a brown oil that was purified by silica column chroma-
tography eluting with hexane/EtOAc (6:1). Azidoalco-
hol (+)-10 was obtained as a white solid (3.6 g, 6.5
mmol, 68%). Mp = 111–114 ꢁC; [a]D +19.5 (c 1,
CHCl3); IR (film, m): 3030, 3007, 2103, 1494, 1453,
1.6. (+)-(1R,2R,3S,4R,5R,6S)-2-Azido-3,4,5,6-tetrakis-
benzyloxycyclohexanol (7)
1
1358, 1067, 736, 697; H NMR (d, 500 MHz, CDCl3):
A 2 N solution of LiClO4 in anhydrous CH3CN
(150 mL) was added dropwise under nitrogen to the
starting epoxide (+)-6 (5.0 g, 9.57 mmol). NaN3 (6.2 g,
95.7 mmol) was added next and the reaction mixture
was stirred at reflux. After 24 h the mixture was cooled
to 0 ꢁC, quenched by the addition of water (150 mL) and
extracted with CH2Cl2 (3 · 100 mL). The combined or-
ganic layers were washed with brine, dried over Na2SO4,
and concentrated to give a brown oil. Azidoalcohol (+)-
7 (5.8 g, 5.8 mmol, 61%) was isolated as a white solid
after flash column chromatography on silica eluting
with hexane/EtOAc (6:1). Mp = 91–93 ꢁC; [a]D +15 (c
1.0, CHCl3); IR (film, m): 3344, 2907, 2108, 1455, 1359,
2.51 (br s, 1H), 3.35–3.37 (dd, 1H, J = 9.5, 2.4 Hz),
3.47–3.49 (dd, 1H, J = 9.5, 2.4 Hz), 3.54 (t, 1H, J =
9.5 Hz), 3.96 (t, 1H, J = 9.5 Hz), 4.03 (t, 1H, J =
9.8 Hz), 4.20 (t, 1H, J = 2.4 Hz), 4.70–4.94 (m, 8H),
7.31–7.37 (m, 5H); 13C NMR (d, 100 MHz, CDCl3):
61.7, 68.1, 72.8, 76.1, 76.2, 76.2, 78.7, 80.6, 81.4, 83.8,
127.9–128.6, 137.2, 137.7, 138.3, 138.4, 159.9; ESIMS:
m/z calcd for C34H35N3O5Na: 588.2474 [M+Na]+.
Found: 588.2499.
1.9. (À)-(1S,2R,3S,4R,5R,6S)-2-Amino-3,4,5,6-tetra-
kisbenzyloxycyclohexanol (11)
1
1133, 1058, 739, 697; H NMR (d, 500 MHz, CDCl3):
A solution of azidoalcohol (+)-10 (3.5 g, 6.2 mmol) in
anhydrous THF (80 mL) was added dropwise under
nitrogen over a solution of LiAlH4 (470 mg, 12 mmol)
in anhydrous THF (50 mL) at 0 ꢁC. After stirring for
90 min at this temperature, the reaction was quenched
by adding EtOAc (10 mL) and, 5 min later, water
(10 mL). The solvent was then removed in vacuo and
the residue was suspended in water. The aqueous layer
was extracted with CH2Cl2 (3 · 100 mL) and the com-
bined organic layers were washed with brine, dried over
Na2SO4, and concentrated to give the crude aminoalco-
hol that was purified by silica flash chromatography.
Elution with CH2Cl2–MeOH–TEA (99:1:0.7) afforded
pure aminoalcohol (À)-11 as a white solid (2.32 g,
4,3 mmol, 70%). Mp = 158–161 ꢁC; [a]D À21.5 (c 1,
2.61 (br s, 1H), 3.42–3.47 (m, 4H), 3.54–3.58 (m, 1H),
3.62–3.65 (m, 1H), 4.79–4.98 (m, 8H), 7.33–7.38 (m,
5H); 13C NMR (d, 100 MHz, CDCl3): 66.5, 72.8, 75.8,
76.0, 76.0, 76.0, 81.2, 82.6, 82.8, 83.6, 127.8–128.8,
137.8, 138.2, 138.2, 138.3; ESIMS: m/z calcd for
C34H35N3O5Na: 588.2474 [M+Na]+. Found: 588.2489.
1.7. (+)-(1R,2S,3S,4R,5S,6R)-(2-Azido-3,4,5,6-tetrakis-
benzyloxy)cyclohexyl methanesulfonate (9)
MsCl (0.43 mL, 4.29 mmol) was added over a solution
of azidoalcohol (+)-7 (2.2 g, 3.9 mmol) in anhydrous
THF (40 mL) and TEA (1.7 mL) under nitrogen atmo-
sphere. The mixture was stirred at room temperature,

1952
P. Gonzalez-Bulnes et al. / Carbohydrate Research 342 (2007) 1947–1952
CHCl3); IR (film, m): 3030, 2923, 1453, 1357, 1074, 734,
696; H NMR (d, 500 MHz, CDCl3): 2.14 (br s, 1H),
References
1
1. Casiraghi, G.; Rassu, G.; Zanardi, F. Furan-, Pyrrole-,
and Thiophene-Based Siloxydienes en route to Carbo-
hydrate Mimics and Alkaloids. In Carbohydrate Mimics;
Chapleur, Y., Ed.; Wiley-VCH, 1998.
2. Xu, Y.; Sculimbrene, B. R.; Miller, S. J. J. Org. Chem.
2006, 71, 4919–4928.
3. Legler, G. Adv. Carbohydr. Chem. Biochem. 1990, 48, 319–
389.
4. Baer, H. H.; Siemsen, L.; Astles, D. Carbohydr. Res. 1986,
156, 247–255.
2.62–2.66 (m, 1H), 3.48 (dd, 1H, J = 9.5, 2.5 Hz), 3.51
(t, 1H, J = 9.5 Hz), 3.63 (t, 1H, J = 9.5 Hz), 3.98 (t,
1H, J = 9.5 Hz), 4.11 (t, 1H, J = 2.5 Hz), 4.65–5.03
(m, 8H), 7.30–7.37 (m, 20H); 13C NMR (d, 75 MHz,
CDCl3): 53.8, 69.4, 72.6, 75.6, 75.7, 75.8, 81.4, 81.9,
82.1, 84.7, 127.5–128.5, 137.9, 138.5, 138.6, 138.7;
ESIMS: m/z calcd for C34H38NO5: 540.2750 [M+H]+.
Found: 540.2748.
5. Butt, A. H.; Kariuki, B. M.; Percy, J. M.; Spencer, N. S.
Chem. Commun. 2002, 7, 682–683.
6. McPhee, F.; Downes, C. P.; Lowe, G. Biochem. J. 1991,
277, 407–412.
1.10. (À)-1-Amino-1-deoxy-myo-inositol hydrochloride
(1)
7. Moris, M. A.; Caron, A. Z.; Guillemette, G.; Schlewer, G.
Bioorg. Med. Chem. Lett. 2004, 12, 3995–4001.
8. Ogawa, S.; Tezuka, Y. Bioorg. Med. Chem. Lett. 2006, 16,
5238–5243.
9. Powis, G.; Aksoy, I. A.; Melder, D. C.; Aksoy, S.;
Eichinger, H.; Fouq, A. H.; Kozikowski, A. P. Cancer
Chemoth. Pharm. 1991, 29, 95–104.
In a glass pressure flask, aminoalcohol (À)-11 (800 mg,
1.5 mmol) was dissolved in a mixture of THF (45 mL)
and concentrated HCl (0.5 mL). Pd/C (600 mg, 5–10%
Pd on activated C, water-wet) was then added. The flask
was repeatedly filled and evacuated with hydrogen and
vigorously stirred at room temperature for 24 h under
H2 (3 atm). After this period, the reaction mixture was
filtered through a plug of Celite to separate the catalyst,
and the filter was washed three times with THF/MeOH
(1:1). The filtrate and combined washings were concen-
trated to give compound (À)-1 as hydrochloride salt
(301 mg, 1.4 mmol, 94%). Mp = 188–190 (dec);
[a]D À4.3 (c 0.4, H2O); IR (KBr, m): 3380, 3061, 2906,
2470, 1570, 1495, 1123, 1034, 709; 1H NMR (d,
500 MHz, D2O): 3.27 (dd, 1H, J = 10.7 Hz, J =
2.4 Hz), 3.32 (t, 1H, J = 9.3 Hz), 3.55 (d, 1H,
J = 9.8 Hz, J = 2.9 Hz), 3.60 (t, 1H, J = 9.5 Hz), 3.71
(t, 1H, J = 10.0 Hz), 4.16 (t, 1H, J = 2.6 Hz); 13C
NMR (125 MHz, D2O): 53.7, 68.5, 69.3, 71.6, 72.0,
74.7; Calcd elemental analysis for C6H14ClNO5: C,
33.42; H, 6.54; Cl, 16.44; N, 6.50. Found: C, 33.21; H,
6.79; Cl, 16.20; N, 6.26; ESIMS: m/z calcd for
C6H14NO5: 180.0872 [M+H]+. Found: 180.0866.
10. Podeschwa, M. A. L.; Plettenburg, O.; Altenbach, H.-J.
Org. Biomol. Chem. 2003, 1, 1919–1929.
11. Serrano, P.; Casas, J.; Zucco, M. G. E.; Egido-Gabas, M.;
Llebaria, A.; Delgado, A. J. Comb. Chem. 2007, 9, 43–52.
12. Egido-Gabas, M.; Serrano, P.; Casas, J.; Llebaria, A.;
Delgado, A. Org. Biomol. Chem. 2005, 3, 1195–1201.
13. Iinuma, K.; Kondo, S.; Maeda, K.; Umezawa, H. Bull.
Chem. Soc. Jpn. 1977, 50, 1850–1854.
14. Post, G. G.; Anderson, L. J. Am. Chem. Soc. 1962, 84,
471–478.
15. Block, O.; Klein, G.; Altenbach, H.-J.; Brauer, D. J. J.
Org. Chem. 2000, 65, 716–721.
16. Adelt, S.; Plettenburg, O.; Stricker, R.; Reiser, G.;
Altenbach, H. J.; Vogel, G. J. Med. Chem. 1999, 42,
1262–1273.
17. Trost, B. M.; Hembre, E. J. Tetrahedron Lett. 1999, 40,
219–222.
18. Trost, B. M.; Patterson, D. E.; Hembre, E. J. Chem. Eur.
J. 2001, 7, 3768–3775.
19. Guo, Z.-X.; Haines, A. H.; Pyke, S. G.; Taylor, R. J. K.
Carbohydr. Res. 1994, 264, 147–153.
20. Ogoshi, S.; Yoshida, K. O.; Murai, S. Organometallics
1993, 12, 578–579.
21. Trost, B. M.; Machacek, M. R.; Aponick, A. Acc. Chem.
Res. 2006, 39, 747–760.
Acknowledgements
22. Serrano, P.; Llebaria, A.; Delgado, A. J. Org. Chem. 2005,
70, 7829–7840.
´
Financial support from Ministerio de Educacion y
23. Saito, S.; Shimazawa, R.; Shirai, R. Chem. Pharm. Bull.
2004, 52, 727–732.
Ciencia (Spain) (CTQ2005-00175/BQU), Ministerio de
Sanidad y Consumo (FIS PI040767), and Fondos Feder
(EU), DURSI (Generalitat de Catalunya: Project
2005SGR01063) is acknowledged. P.G.B. thanks Minis-
´
´
24. Jaramillo, C.; Fernandez de la Pradilla, R.; Martın-
Lomas, M. Carbohydr. Res. 1991, 209, 296–298.
25. Falshaw, A.; Hart, J. B.; Tyler, P. C. Carbohydr. Res.
2000, 329, 301–308.
´
terio de Educacion y Cultura (Spain) for a predoctoral
fellowship.