New aminocyclitols as modulators of glucosylceramide metabolism

Article abstract of DOI:10.1039/b411473f

A series of 13 aminocyclitol derivatives belonging to two different families is described. Their configuration is governed by the regio- and stereocontrolled epoxide opening of a suitably protected conduritol-B epoxide. Studies on several glycosyl processing enzymes indicate that some of them are good inhibitors of glucosylceramide hydrolase. A rationale to account for preliminary structure-activity relationships is provided. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b411473f

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
New aminocyclitols as modulators of glucosylceramide metabolism
Meritxell Egido-Gaba´s,a Pedro Serrano,a,b Josefina Casas,a Amadeu Llebariaa and
Antonio Delgado*a,b
a Research Unit on Bioactive Molecules (RUBAM), Departament de Qu´ımica Orga`nica
Biolo`gica, Institut d’Investigacions Qu´ımiques i Ambientals de Barcelona (IIQAB-C.S.I.C),
Jordi Girona 18-26, 08034, Barcelona, Spain
b Universitat de Barcelona, Facultat de Farma`cia, Unitad de Qu´ımica Farmace`utica (Unitat
Associada al CSIC), Avda. Joan XXIII, s/n, 08028, Barcelona, Spain.
E-mail: adelgado@cid.csic.es; Fax: int +34-932045904
Received 27th July 2004, Accepted 12th January 2005
First published as an Advance Article on the web 1st March 2005
A series of 13 aminocyclitol derivatives belonging to two different families is described. Their configuration is
governed by the regio- and stereocontrolled epoxide opening of a suitably protected conduritol-B epoxide. Studies on
several glycosyl processing enzymes indicate that some of them are good inhibitors of glucosylceramide hydrolase. A
rationale to account for preliminary structure–activity relationships is provided.
Introduction
Results
Aminocyclitols are a group of natural compounds that are
widely distributed throughout nature.1 Several families of
aminocyclitol analogues have been synthesized over the last
decades and their biological properties have been thoroughly
studied.2 Among them, glycosidase inhibitors have gained mo-
mentum as pharmacological tools, as well as for their potential
as new therapeutic agents.3
Despite epoxide opening of substituted 1,2-epoxycyclohexanes
being documented,10 examples of regio- and stereocontrolled
opening of cyclitol epoxides with nitrogen nucleophiles are
scarce.11 Although a trans-diaxial stereoselectivity is generally
found, based on stereoelectronic effects,12,13 the observed regio-
selectivity depends largely on conformational bias and can be
modulated by a judicious choice of reaction conditions.
Over the course of our current research on new chemical
entities as modulators of sphingolipid and glycosphingolipid
(GSL) metabolism,4–6 we have focussed on the regulation of
ceramide (Cer) and glucosylceramide (GlcCer) levels owing to
their relevance as biochemical targets for the development of new
pharmacological tools with potential therapeutic applications.7,8
Along this line, in this work we report on the synthesis and
glucosidase inhibitory activities of a series of new aminocyclitols
derived from the regio- and sterocontrolled epoxide opening of a
suitably protected (+)-conduritol-B epoxide 149 (see Scheme 1).
In a previous work, we accounted for an unexpected chelation-
controlled C1 regioselective opening of epoxide 14 with nitrogen
nucleophiles in the presence of Yb(OTf)3.14 Further to this, we
have shown that Li salts behave similarly to promote epoxide
opening through an intramolecular chelation process, which
would require the participation of an apparently highly energetic
“all-axial” conformation (Scheme 2).15
Thus, both aminolysis or azidolysis of epoxide 14 in the
presence 2 N LiClO4 cleanly afforded the corresponding C1
adducts 15–18, as shown in Scheme 2. The regiochemistry of the
Scheme 1 Aminocyclitols described in this work.
Scheme 2 Synthesis of cyclitol analogues by regioselective C1 opening of the starting epoxide.
T h i s j o u r n a l i s T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5 O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 9 5 – 1 2 0 1
1 1 9 5
©

View Article Online
Scheme 3 Synthesis of N-acylcyclitols 5–7, 9 and ureidocyclitol 8 from azide 18. Reagents and conditions: (a) BnCl, NaH, THF (95%); (b) LiAlH4,
THF (87%); (c) R1COCl, Et3N, CH2Cl2 (81–87%); (d) (i) COCl2, toluene; (ii) octylamine, CH2Cl2 (67% overall); (e) PhCH2CHO, NaBH3CN, AcOH,
MeOH (85%); (f) BCl3, CH2Cl2 (quantitative).
Scheme 4 Synthesis of aminocyclitols by regioselective C2 opening of epoxide 14. Reagents and conditions: (a) NaN3, NH4Cl, MeOH–H2O (80–81%);
(b) LiAlH4, THF (81%); (c) RCHO, NaBH3CN, AcOH, MeOH (83–91%); (d) BCl3, CH2Cl2 (quantitative).
epoxide opening under these conditions was corroborated by the
simplification of the NMR pattern observed for these adducts
due to symmetry considerations.
Debenzylation of 15–17 (Scheme 2) and 20 (obtained by O-
benzylation and azide reduction of 18, Scheme 3) afforded the
corresponding aminocyclitols 1–4 (Scheme 1). Amide analogues
5–7, 9 and ureido cyclitol 8 were obtained from 20, as outlined
in Scheme 3.
(from commercial crude almond), GlcCer synthase (from rat
liver microsomes), lysosomal glucocerbrosidase (from rat liver
membranes) and recombinant GlcCer hydrolase (Imiglucerase,
R
Cerezymeꢀ, from Genzyme). All compounds were inactive when
tested at 0.5 mM and 1 mM concentrations against GlcCer
synthase and almond b-glucosidase, respectively, whereas 12 and
13 exhibited significant a-glucosidase inhibitory activity (IC50
22 lM and 142 lM, respectively, Table 1).
=
Treatment of epoxide 14 with sodium azide under acidic
conditions (excess ammonium chloride in a methanol–water
mixture) afforded the corresponding C2 regioadduct 27 with
total regioselectivity. Standard azide reduction, reductive ami-
nation and O-benzyl deprotection led to aminocyclitols 11–13
in good overall yields (Scheme 4).
Lineweaver–Burk plots of assays performed with a-
glucosidase at different substrate and inhibitor concentrations
showed that these compounds behave as non-competitive
inhibitors, indicating that the substrate and inhibitor bind
reversibly and independently at different enzyme sites. Replots
of slopes versus the inhibitor concentration indicate that the
Ki values for compounds 12 and 13 are 19 lM (Fig. 1) and
145 lM, respectively. Inhibitor 12 is the more active and also
Compounds 1–13 have been tested as inhibitors of a-
glucosidase (from commercial baker’s yeast), b-glucosidase
Table 1 Enzyme inhibition for compounds 1–13
a-Glucosidase (baker’s yeast)
GCH (lysosomal)
Imiglucerase
Compound
IC50/lM
Ki/lM
IC50/lM
Ki/lM
IC50/lM
Ki/lM
1
2
3
4
5
6
NI
NI
NI
254
47%a
24%a
NI
293
35%a
420
352
498
NI
15.0
16.0
15.7
30%a
NI
2.7
2.2
2.4
163
136
208
NI
7
8
9
NI
NI
NI
412
34%a
350
33%a
NI
910
275
ND
168
43
NI
1219
12.0
NI
285
0.9
10
11
12
13
DNJ
30%a
900
22
877
19
NI
940
64
ND
33%a
503
506c
142
ND
145
158
88
8.7b
NI: No significant inhibition at 1 mM; ND: not determined.a Inhibition at 1 mM. b See ref. 28. c See ref. 16.
1 1 9 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 9 5 – 1 2 0 1

View Article Online
binding of this class of compounds, including X-ray structures
of aminocyclitol derivatives bound to several glycosyl hydrolase
enzymes.18
Our compounds define two structural families arising from
either C1 or C2 opening of epoxycyclohexane 14 (Scheme 1).
Except where a long chain alkyl substituent was present, prod-
ucts resulting from C2 opening were inactive (see compounds
10–13, Table 1). In contrast, this effect is not observed for the C1
opening compounds. In this case, a similar inhibitory potency is
obtained with different alkyl chain lengths (compounds 2–4), the
primary amine 1 remaining inactive. In this family, replacement
of N-alkyl with N-acyl groups (compounds 5–8) resulted in less
active compounds, probably due to their minor proton acceptor
ability as a result of the lower basicity of the amide nitrogen. In
contrast, secondary amide 9 displays a strong inhibitory activity
that can be attributed to its particular nitrogen substitution
pattern. Interestingly, this observation indicates that, even in
the absence of an amine nitrogen atom, good active site affinity
can be achieved by a suitable arrangement of the lipidic moiety.
Based on the published X-ray structure of the enzyme19,20 and
according to the postulated binding model for GlcCer in the
enzyme active site,20 we have compared the three-dimensional
structure of GluCer with those of representative molecules of
each of the aminocyclitol families.21 In this qualitative analysis
we have found a close correlation between their inhibitory
potency and their ability to superimpose with the glucose moiety,
thus suggesting a sugar-mimicking role for the aminocyclitol
scaffold. The best fit is observed for C1 adducts (Fig. 3 A,
light blue color code) where close matches between, a) the CH–
OH groups of aminocyclitol and the glucose moiety of GlcCer
and b) the aminocyclitol nitrogen atom and the b-anomeric
oxygen are found. The only structural mismatches are imposed
by the presence of an aminocyclitol hydroxyl group in place of
the glucoside hydroxymethyl group and the replacement of the
pyranose oxygen by an aminocyclitol C(2)H–OH group.22
Fig. 1 Lineweaver–Burk plot for 12 with a-glucosidase (from baker’s
yeast).
highly specific, since its activity against GlcCer hydrolase is very
low (33% inhibition at 1 mM).
Inhibitory activity against GlcCer hydrolase (both lysosomal
and Imiglucerase) is shown in Table 1. IC50 values for the
lysosomal enzyme are higher than those calculated for the
purified enzyme, most likely due to the presence of other
proteins in the lysosomal preparation that may interact with
the compounds. Kinetic assays with active compounds were
performed and Ki values were calculated (see Table 1 and Fig. 2).
For comparative purposes, DNJ (deoxynojirimycin) was also
tested.16 Active compounds behaved as competitive inhibitors
of GlcCer hydrolase, irrespective of the enzyme source used.
Fig. 2 Lineweaver–Burk plot for 9 with Imiglucerase.
As observed from Table 1, the only active aminocyclitols as
GlcCer hydrolase inhibitors are those arising from regioselective
C1 opening of the corresponding starting epoxides (compounds
2–4 and 9, see Scheme 1). Nitrogen substitution also seems to be
important for enzyme inhibition. Thus, only N-alkyl substituted
aminocyclitols showed a remarkable inhibitory activity, whereas
free amino (compound 1), N-ureido (compound 8) or N-
acyl derivatives (compounds 5–7) showed only marginal or
no activity. Interestingly, the combination of both substitution
patterns (N-acyl and N-alkyl) into a single molecule (compound
9) afforded the most active compound in this study.
Fig. 3 Superimposition of the two regioisomeric aminocyclitol families
(A, B, coloured bonds) with glucosylceramide (grey). A: C1-adducts; B:
C2-adducts. Matching was done by first minimizing the structures on
Chem3D, followed by manual docking using the “Molecular Overlay”
tool provided with the DS Viewer Pro application from Accelrys.23
Discussion
The diverse range of competitive inhibition of GlcCer hydro-
lase displayed by the aminocyclitol derivatives described here
indicates a differential interaction with the amino acid residues
of the active site. To find a qualitative explanation for these
results, we decided to compare the structure of the synthesized
aminocyclitols with that of GlcCer, the enzyme substrate.17
The starting hypothesis was that aminocyclitols are recognized
as substrate mimics, the polyhydroxylated ring acting as a
surrogate for the glucose moiety present in GlcCer and the
exocyclic nitrogen playing the role of the anomeric oxygen.
In this prediction, the N-alkyl chain would mimic the lipidic
ceramide moiety present in the substrate.
The C2 opening families have in common the two trans-
diaxial substituents (the nitrogen group and the hydroxyl
substituent) arising from the stereocontrolled opening of the
starting epoxide. This generates an abrupt structural disparity
with GlcCer, where all substituents are equatorial. Therefore,
the ability of these compounds to mimic the substrate sugar
moiety is lower than that for the corresponding C1 open-
ing addcuts. Thus, if we maintain the initial restriction that
superimposes the aminocyclitol nitrogen atom to the GlcCer
anomeric oxygen, important mismatches are found for the rest
This hypothesis is supported by the competitive nature of
the inhibition and also by the extent of literature data on the
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 9 5 – 1 2 0 1
1 1 9 7

View Article Online
of hydroxyl substituents. In contrast, when the hydroxyl groups
are superimposed, the cyclitol amino group is “epimerized” with
respect to the glucose anomeric group (see Fig. 3, B). These
observations can explain why these compounds are, in general,
less potent inhibitors. However, as a general trend, we found an
increase in the inhibitory potency as the length (and hence the
liphophilic character) of the nitrogen alkyl substituent increases,
in agreement with the above mentioned contribution of the
lipidic moiety to the substrate binding. In contrast, this tendency
is not observed within the C1 adducts, which show similar Ki
values irrespective of the nature of the N-alkyl substituent.
In summary, the ability of the aminocyclitol scaffold to
superimpose the glucose moiety of the enzyme substrate appears
to control the inhibitory potency of these compounds against
GlcCer hydrolase. The aminocyclitol nitrogen substituent will be
capable of correcting, to some extent, the hydroxyl mismatches
when increasing its liphophilicity.
The design of selective competitive inhibitors of GlcCer
hydrolase can be exploited for the development of chemical
chaperones,24 competitive inhibitors that promote the phar-
macological rescue of a defective enzyme at subinhibitory
concentrations. In this context, this approach represents a
promising alternative for a more efficient therapy for Gaucher
disease, a kind of sphingolipidosis caused by a deficiency of
GluCer hydrolase, due, in part, to enzyme misfolding leading to
its degradation during the control processes in the endoplasmic
reticulum.
in McIlvaine buffer (pH 5.2). Enzyme solutions were incubated
at 37 ◦C without (control) or with inhibitor (1 mM) during
30 min and, after addition of corresponding substrate solution,
incubations were maintained at 37 ◦C for 30 min (lysosomal glu-
cocerebrosidase) or 10 min (Imiglucerase). Enzymatic reactions
were stopped by the addition of 100 mM glycine–NaOH buffer
(pH 10.6). The amount of 4-methylumbelliferone formed was
determined using a 1420 VICTOR2 Multilabel Counter (Wallac)
fluorometer at 355 nm (excitation) and 460 nm (emission).
Glucosylceramide synthase: liver microsomes were obtained
from Sprage-Dawley rats as described.26 Microsomal protein
was incubated with 0.12 mM of N-octanoyl-D-sphingosine–
BSA complex and 1.6 mM NADP in 50 mM Tris–HCl buffer
(pH 7◦.4) with or without (control) 0.5 mM inhibitor for 10 min
at 37 C. After addition of 0.04 mM [14C]UDP-glucose (2 mCi
mmol−1) incubations were prolonged for 35 min and stopped
by the addition of 0.5 ml CHCl3 : CH3OH (2 : 1). The [14C]-
glucosyl-N-octanoyl-D-sphingosine formed was extracted27 and
quantified using a Wallac 1410 (Pharmacia) liquid scintillation
counter.
Inhibition parameters. The IC50 values were determined by
plotting percent activity versus log [I], using at least five different
inhibitor concentrations. Type of inhibition and Ki values for
more active inhibitors were determined by Lineweaver–Burk or
Dixon plots of assays performed with different concentrations
of inhibitor and substrate.
This rationale is being used for the optimisation of these
families of inhibitors, currently underway in our group.
Synthesis
(1S,2S,3S,4S,5R,6R)-3-Azido-4,5,6-tris-benzyloxycyclohex-
ane-1,2-diol 27. To a solution of the epoxide 14 (95 mg,
0.22 mmol) in a 8 : 1 methanol–water mixture (4.5 cm3) was
added portionwise NaN3 (162 mg, 2.5 mmol) and NH4Cl
(59 mg, 14.1 mmol). After stirring at 80 ◦C for 20 h, the reaction
mixture was cooled to rt, quenched with water (10 cm3) and
extracted with diethyl ether (3 × 10 cm3). The organic layers
were washed with brine (10 cm3), dried and concentrated under
a reduced pressure. The crude product was purified by flash
chromatography using hexane–ethyl acetate (2 : 1) as eluent
to afford 27 in 81% yield as a white solid. [a]2D5 +16.7 (c 1.00,
CHCl3). mmax(film)/cm−1 3452, 3031, 2907, 2109, 1497, 1454,
1362. 13C NMR: 60.2, 70.7, 72.9, 73.1, 75.7, 77.9, 82.3, 82.9,
127.8, 127.9, 128.0, 128.1, 128.5, 128.6, 137.4, 138.0. 1H NMR:
2.75 (m, 1H), 3.4–3.6 (m, 2H), 3.71–3.85 (m, 2H), 3.87–3.92
(m, 1H), 4.50–5.0 (m, 6H), 7.2–7.4(m, 6H). [a]2D5 +15.5 (c 1.45,
CHCl3). EIHRMS, calc. for C27H29N3O5: 475.2107; found:
475.2123.
Experimental
General
Solvents were distilled prior to use and dried by standard
methods.25 Chemical shifts are reported in delta units (d),
parts per million (ppm) relative to the singlet at 7.24 ppm
1
of CDCl3 for H and in ppm relative to the centre line of a
triplet at 77.0 ppm of CDCl3 for 13C. [a]D25 Values are given in
10−1deg cm2 g−1. Compound 14 (4,5,6-tri-O-benzyl-1,2-anhydro-
myo-inositol) was obtained following literature procedures.9
Enzyme inhibition: enzyme assays
p-Nitrophenyl glucosides, 4-methylumbelliferyl-b-D-glucoside,
ceramide and UDP-glucose were obtained from Sigma. [glucose-
14C(U)] UDP-glucose (303 mCi mmol−1) was from New England
Nuclear. a-Glucosidase (from baker’s yeast, G5003) and b-
glucosidase (from almonds, G0395), were purchased from
Reaction of epoxide 14 with amines in the presence of lithium
perchlorate: synthesis of amino alcohols 15–17. A solution of
epoxide 14 (300 mg, 0.7 mmol) in CH3CN (10 cm3) was added
dropwise under an argon atmosphere over lithium perchlorate
(2 g, 19 mmol) at rt. A solution of the corresponding amine
(5 mmol) in CH3CN (10 cm3) was next added and the reaction
R
Sigma. Imiglucerase (Cerezymeꢀ) was kindly provided by Gen-
zyme.
Commercial glucosidase solutions were prepared with the
appropriate buffer and incubated in 96-well plates at 37 ◦C
without (control) or with inhibitor (1 mM) for 5 min. After
addition of the corresponding substrate solution, incubations
were maintained during 3 min and stopped by the addition of
1 M Tris solution. The amount of p-nitrophenol formed was
determined at 405 nm using a UV/VIS Lector Spectramax
Plus (Molecular Devices Corporation) spectrophotometer. a-
Glucosidase activity was determined with 1 mM p-nitrophenyl-
a-D-glucopyranoside in 100 mM sodium phosphate buffer
(pH 7.2). b-Glucosidase activity was determined with 1 mM
p-nitrophenyl-b-D-glucopyranoside in 100 mM sodium acetate
buffer (pH 5.0).
Glucocerebrosidase: lysosomal glucocerebrosidase activity in
rat liver membrane suspensions16 was determined with 1 mM
4-methylumbelliferyl-b-D-glucopyranoside in McIlvaine buffer
(100 mM sodium citrate and 200 mM sodium phosphate buffer,
pH 5.2). Imiglucerase activity was determined with 2.4 mM
4-methylumbelliferyl-b-D-glucopyranoside in the presence of
0.25% (w/v) sodium taurocholate and 0.1% (v/v) Triton X-100
◦
mixture stirred at 80 C under argon. After 18 h, the reaction
mixture was cooled to rt, quenched with water (10 cm3), ex-
tracted with dichloromethane (3 × 20 cm3) and dried. Filtration
and evaporation afforded crude compounds, which were purified
as indicated below.
(1R,2s,3S,4R,5r,6S)-4,5,6-Trisbenzyloxy-2-butylaminocyclo-
hexane-1,3-diol 15. (92%) The compound was purified by
flash chromatography using a 9 : 1 mixture of dichloromethane–
methanol. Oil; mmax(film)/cm−1 3432, 3027, 2924, 2912, 1495
1456. dH (200 MHz, CDCl3, Me4Si) 0.92 (3H, t, J = 7.2 Hz,
CH3), 1.19–1.43 (4H, m, 2 × CH2), 2.86 (1H, t, J2−1 = J2−3 10.8,
2-H), 3.18 (2H, m, NCH2), 3.47–3.50 (3H, m, 4-H, 5-H and
6-H), 3.81 (2H, t, J1–2 = J6−1 10.8, 1-H and 3-H), 4.6–5.0 (6H, m,
3 × OCH2Ph), 7.2–7.4 (15H, m, 3 × Ph). dC (50.3 MHz, CDCl3,
Me4Si) 13.9 (q), 20.3 (t), 32.4 (t) (CH3CH2CH2), 44.1 (t, NCH2),
61.2 (d, C2), 71.0 (d, C1, C3), 75.1 (OCH2Ph), 75.7 (OCH2Ph),
1 1 9 8
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 9 5 – 1 2 0 1

View Article Online
81.2 (d, C5), 84.3 (d, C4, C6), 127.5–128.4 (CHAr), 138.3–138.8
(CAr). Found: C, 73.81; H, 8.02; N, 2.67%. C31H39NO5 requires
C, 73.63; H, 7.77; N, 2.77%.
4.68–5.01 (m, 10H), 7.24–7.40 (20H, m). dC (75 MHz, CDCl3,
Me4Si) 66.8 (d), 75.8 (2 × d), 81.0 (2 × t), 82.4 (t), 83.2 (2 × t),
127.7–128.5 (25 × d), 138.2–138.3 (5 × s). Found: C, 78.35; H,
7.02; N, 2.37%. C41H43NO5 requires C, 78.19; H, 6.88; N, 2.22%.
(1R,2s,3S,4R,5r,6S)-4,5,6-Tris-benzyloxy-2-phenethylamino-
cyclohexane-1,3-diol 16. (91%) Purified by flash chromato-
graphy (silica gel treated with 3% triethylamine) using a 1 :
1 mixture of hexanes and ethyl acetate. Oil mmax(film)/cm−1
3575, 3421, 3087, 2923, 2856, 1953, 1879, 1741, 1439, 1454. dH
(200 MHz, CDCl3, Me4Si) 2.59 (1H, t, J2–1 = J2–3 10.8, 2-H),
2.86 (2H, m, NCH2), 3.23–3.43 (5H, m,CH2Ph, 4-H, 5-H and
6-H), 3.83 (2H, t, J1–2 = J6–1 10.8, 1-H and 3-H), 4.71–4.79 (6H,
m, 3 × OCH2Ph), 7.2–7.4 (20H, m, 3 × Ph). dC (50.3 MHz,
CDCl3, Me4Si) 36.9 (t, CH2Ph), 46.1 (t, NCH2), 61.4 (d, C2),
71.5 (d, C1, C3), 75.3 (OCH2Ph), 75.7 (OCH2Ph), 82.8 (d,
C5), 84.2 (d, C4, C6), 126.1–128.6 (CHAr), 138.3, 138.5, 139.7
(CAr). Found: C, 76.21; H, 7.42; N, 2.57%. C36H41NO5 requires
C, 76.16; H, 7.28; N, 2.47%.
(1S,2S,3S,4S,5R,6R)-3-Amino-4,5,6-trisbenzyloxycyclohex-
ane-1,2-diol 28. (81% from 27); oil. [a]2D5 +10.7 (c 1.00, CHCl3).
mmax(film)/cm−1 3423, 3316, 3153, 3102, 3084, 2925, 2886, 2856,
1499, 1469. dH (200 MHz, CDCl3, Me4Si) 3.53 (t, J = 3.7, 1H,
H3), 3.74 (t, J = 9.9, 1H, H5), 3.92 (m, 3H, H4, H5, H1),
4.06 (dd, J = 3.7, 10.5, 1H, H2), 4.6–5 (m, 6H, 3 × OCH2Ph),
7.23–7.41 (m, 15 H, 3 × Ph). dC (50.3 MHz, CDCl3, Me4Si)
51.3 (d, C3), 71.0 (d, C4), 72.1 (OCH2Ph), 72.7 (OCH2Ph), 74.9
(OCH2Ph), 75.1 (OCH2Ph), 80.3 (d, C1), 80.8 (d, C5), 81.2 (d,
C6), 127.7–128.6 (CHAr), 138.3–138.5 (CAr). Found: C, 72.25;
H, 7.12; N, 3.37%. C27H31NO5 requires C, 72.14; H, 6.95; N,
3.12%.
Reductive amination: synthesis of 22, 29–31. A solution of
the starting amino alcohol 20 or 28 (0.05 mmol) in methanol
(2 cm3) under an atmosphere of argon was treated successively
with sodium cyanoborohydride (4.2 mg, 0.108 mmol), acetic acid
(3.5 lL) and the aldehyde (0.05 mmol). After stirring for 18 h at
rt, the mixture was quenched with water (2 cm3) and extracted
with diethyl ether (3 × 20 cm3). The combined organic layers
were washed with brine, dried and concentrated under a reduced
pressure to afford crude amino alcohols, which were purified
by flash chromatography using a mixture of dichloromethane–
methanol (12 : 1).
(1R,2s,3S,4R,5r,6S)-4,5,6-Tris-benzyloxy-2-octylaminocyclo-
hexane-1,3-diol 17. (90%) Purified by flash chromatography
using a 9 : 1 mixture of dichloromethane–methanol. Oil;
mmax(film)/cm−1 3414, 3316, 3105, 3067, 3032, 2964, 2929, 2856,
1490, 1456. dH (200 MHz, CDCl3, Me4Si) 0.92–1.43 (15H, m,
CH3, 7 × CH2), 2.48 (2H, m, NCH2), 2.87, (1H, t, J2–1 = J2–3
10.5, 2-H), 3.17 (2H, m, NCH2), 3.47–3.50 (3H, 4-H, 5-H and
6-H), 3.78 (2H, t, J1–2 = J6–1 10.4, 1-H and 3-H), 4.6–5.0 (6H,
m, 3 × OCH2Ph), 7.2–7.4 (15H, m, 3 × Ph). dC (50.3 MHz,
CDCl3, Me4Si) 14.0 (q), 22.6 (t), 26.3 (t), 26.5 (t), 28.9 (t), 29.0
(t), 31.6 (t) (CH3CH2CH2CH2CH2CH2CH2), 49.7 (t, N CH2),
61.2 (d, C2), 69.6 (d, C1, C3), 75.3–75.7 (OCH2Ph), 82.4 (d,
C5), 84.3 (d, C4, C6), 127.7–128.5 (CHAr), 138.3–138.5(CAr).
Found: C, 75.23; H, 8.65; N, 2.47%. C36H49NO5 requires C,
75.10; H, 8.58; N, 2.43%.
(1s,2R,3S,4r,5R,6S)-2,3,4,5,6-Pentakis-benzyloxy-N -phen-
ethylcyclohexylamine 22. (85% from 20), oil. mmax(film)/cm−1
3057, 3027, 2957, 1480, 1456. dH (500 MHz, CDCl3, Me4Si)
2.93 (3H, m), 3.15 (3H, m), 3.55–3.68 (5H, m), 4.81–5.00 (m,
10H), 7.12 (2H, d, J = 5.5), 7.24–7.40 (28H, m). dC (125 MHz,
CDCl3, Me4Si) 37.0 (t), 65.9 (d), 75.2 (2 × d), 76.0 (d), 76.1
(2 × d), 79.9 (2 × t), 82.8 (t), 84.8 (2 × t), 85.4 (t), 126.2 (2 ×
d), 127.7–128.9 (28 × d), 138.6–138.9 (5 × s), 140.6 (s).
(1S,2S,3R,4R,5S,6S)-3,4,5-Tris-benzyloxy-6-butylaminocyclo-
hexane-1,2-diol 29. (90% from 28), oil. [a]2D5 +34.3 (c 1.00,
CHCl3). mmax(film)/cm−1 3356, 3031, 2909, 1496, 1454, 1359. dH
(300 MHz, CDCl3, Me4Si) 0.88 (3H, t, 7.2, J CH3), 1.25–1.44
(4H, m, 2 × CH2), 2.40–2.60 (2H, m, CH2), 3.19 (t, 1H, J =
4.2, CHNHR), 3.69 (t, 1H, J = 8.7, CHOH), 3.85–4.05 (4H,
m, CH), 4.50–5.00 (6H, m, OCH2Ph), 7.2–7.4 (15H, m, Ar).
dC (75 MHz, CDCl3, Me4Si) 13.9 (q), 20.3 (t), 32.4 (t), 48.1 (t),
58.2 (d), 69.1 (2 × d), 71.0 (2 × d), 72.4 (2 × d), 74.9 (d), 75.1
(d), 79.6 (t), 81.2 (t), 127.5–128.6 (12 × d), 138.4 (s), 138.7 (s).
Found: C, 73.81; H, 7.98; N, 2.59%. C31H39NO5 requires C,
73.63; H, 7.77; N, 2.77%.
(1R,2s,3S,4R,5r,6S)-2-Azido-4,5,6-trisbenzyloxy-1,3-cyclo-
hexanol 18. A solution of sodium azide (1.0 g, 15.4 mmol) and
lithium perchlorate (6.9 g, 64.8 mmol) was added to a solution
of epoxide 14 (650 mg, 1.5 mmol) in anhydrous acetonitrile
(35 cm3) and stirred at 80 ◦C under an atmosphere of argon.
After 18 h, the reaction mixture was cooled to rt, quenched with
water (15 cm3) and extracted with diethyl ether (3 × 30 cm3).
The combined organic layers were washed with brine (25 cm3),
dried and concentrated under a reduced pressure. The crude
azidoalcohol was purified by filtration through a plug of silica
and elution with a 1 : 1 mixture of hexane and ethyl acetate to
afford 18 in 92% yield as an oil. mmax(film)/cm−1 3350 (br s),
2109. dH (200 MHz, CDCl3, Me4Si) 2.78 (1H, br s, 2-H), 3.46
(5H, m, 1-H, 3-H, 4-H, 5-H and 6-H), 4.67–4.99 (6H, m, 3 ×
OCH2Ph), 7.20–7.35 (15H, m, 3 × Ph). dC (50.3 MHz, CDCl3,
Me4Si) 53.5 (d, C2), 65.9 (d, C1 and C3), 73.0 (OCH2Ph), 75.7
(OCH2Ph), 82.4 (d, C5), 82.9 (d, C4 and C6), 127.8–128.7
(CHAr), 138.1–138.2 (CAr).
(1S,2S,3R,4R,5S,6S)-3,4,5-Tris-benzyloxy-6-phenethylamino-
cyclohexane-1,2-diol 30. (91% from 28). [a]2D5 +21.4 (c 0.8,
CHCl3). mmax(film)/cm−1 3356, 3042, 2919, 1506, 1437, 1423. dH
(300 MHz, CDCl3, Me4Si) 0.88 (3H, t, J = 7.2, CH3), 1.25–1.44
(4H, m, 2 × CH2), 2.40–2.60 (2H, m, CH2), 3.19 (t, 1H, J =
4.2, CHNHR), 3.69 (t, 1H, J = 8.7, CHOH), 3.85–4.05 (4H,
m, CH), 4.50–5.00 (6H, m, OCH2Ph), 7.2–7.4 (15H, m, Ar).
dC (75 MHz, CDCl3, Me4Si) 13.9 (q), 20.3 (t), 32.4 (t), 48.1 (t),
58.2 (d), 69.1 (2 × d), 71.0 (2 × d), 72.4 (2 × d), 74.9 (d), 75.1
(d), 79.6 (t), 81.2 (t), 127.5–128.6 (12 × d), 138.4 (s), 138.7 (s).
Found: C, 76.11; H, 7.28; N, 2.65%. C35H39NO5 requires C,
75.92; H, 7.10; N, 2.53%.
Reduction of azides 19 and 27: synthesis of 20 and 28. A solu-
tion of the starting azide (2.0 mmol) in anhydrous tetrahy-
drofuran (30 cm3) was added to a cooled solution (0 ◦C) of
lithium aluminium hydride (45 mg, 1.2 mmol) in anhydrous
tetrahydrofuran (30 cm3) under argon. The reaction mixture
was stirred for 30 min at rt. Ethyl acetate (20 cm3) and water
(20 cm3) were slowly added and the mixture was extracted
with diethyl ether (3 × 50 cm3). The combined organic layers
were washed with brine (50 cm3), dried and concentrated under
a reduced pressure. The crude product was purified by flash
chromatography on deactivated silica gel (3% triethylamine)
using dichloromethane–methanol (12 : 1) as eluent to afford
the final compounds.
(1S,2S,3R,4R,5S,6S)-3,4,5-Tris-benzyloxy-6-octylamino-
cyclohexane-1,2-diol 31. (83% from 28); oil. [a]2D5 +24.8 (c
1.00, CHCl3). mmax(film)/cm−1 3306, 3029, 2909, 1485, 1444,
1359, 1345. dH (300 MHz, CDCl3, Me4Si) 0.88 (3H, t, J = 7.2,
CH3), 1.25–1.44 (8H, m, 2 × CH2), 2.38–2.60 (2H, m, CH2),
3.23 (t, 1H, J = 4.6, CHNHR), 3.67 (t, 1H, J = 8.6, CHOH),
3.78–4.10 (4H, m, CH), 4.50–4.98 (6H, m, OCH2Ph), 7.2–7.4
(15H, m, Ar). dC (75 MHz, CDCl3, Me4Si) 14.1 (q), 20.4 (t),
(1s,2R,3S,4r,5R,6S)-2,3,4,5,6-Pentakis-benzyloxycyclohexyl-
amine 20. (87% from 19); oil. mmax(film)/cm−1 3331, 3062,
3031, 2907, 1580, 1456, 1359. dH (300 MHz, CDCl3, Me4Si)
2.97 (1H, t, J1–2 = J2–3 = 9.9, 2-H), 3.35 (2H, m), 3.62 (3H, m),
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 9 5 – 1 2 0 1
1 1 9 9

View Article Online
25.9 (t), 28.6 (t), 29.4 (t), 29.5 (t), 32.3 (t), 48.1 (t), 58.2 (d), 68.9
(2 × d), 71.1 (2 × d), 72.4 (2 × d), 75.0 (d), 75.2 (d), 79.6 (t),
81.3 (t), 127.5–128.6 (12 × d), 138.3 (s), 138.7 (s). Found: C,
75.01; H, 8.64; N, 2.77%. C35H47NO5 requires C, 74.83; H, 8.43;
N, 2.49%.
anhydrous THF (20 cm3) was added dropwise over an ice-cooled
suspension of sodium hydride (60% dispersion in mineral oil,
270 mg, 11.25 mmol), previously washed with hexane (3 ×
10 cm3) in tetrahydrofurane (25 cm3) under an atmosphere of
argon. Stirring was continued until complete gas evolution
and benzyl chloride (0.45 cm3, 3.79 mmol) was next added.
The reaction mixture was stirred at rt for 24 h, quenched with
water (10 cm3) and extracted with diethyl ether (3 × 25 cm3).
The combined organic layers were washed with brine (20 cm3),
dried and concentrated. The resulting oil was purified by flash
chromatography using a mixture of hexane–ethyl acetate (4 :
1) as eluent to afford azide 19 (95%); oil. mmax(film)/cm−1 3031,
2909, 2107, 1496, 1454, 1359. dH (300 MHz, CDCl3, Me4Si)
3.34–3.38 (2 H, m, 2 × CH), 3.46–3.58 (4 H, m, 4 × CH),
4.84–4.90 (10H, m, 5 × OCH2Ph), 7.2–7.4 (25 H, m, Ar). dC
(75 MHz, CDCl3, Me4Si) 66.9 (d), 75.9 (3 × d), 81.0 (2 × t),
82.4 (t), 83.1 (2× t), 127.7–128.4 (20 × d), 137.7 (2× s), 138.1
(3× s).
(1s,2S,3R,4r,5S,6R)-1-Octyl-3-(2,3,4,5,6-pentakis-benzyloxy-
cyclohexyl)urea 21. To an ice-cooled solution of 20 (50 mg,
0.077 mmol) in anhydrous dichloromethane (3 cm3) were added
60 ll (0.144 mmol) of a 20% solution of phosgene in toluene.
After 1 h stirring at rt, the reaction mixture was quenched
with aqueous NaOH 0.1 N (2 cm3) and the organic layer
washed with brine and dried. Removal of the solvent under
a reduced pressure afforded a residue which was treated with
octylamine (15 ll, 0.09 mmol) in anhydrous dichloromethane
(3 cm3). After stirring for 36 h at rt, the reaction mixture was
quenched with water (2 cm3). The organic layer was washed
with brine, dried and the solvent was removed under a reduced
pressure. The resulting crude mixture was purified by column
chromatography (hexane–ethyl acetate, 1 : 1) to give urea 21
(40 mg, 67%) as an oil. mmax(film)/cm−1 3028, 2980, 2910, 2865,
1638, 1456, 1407, 1356. dH (300 MHz, CDCl3, Me4Si) 0.83 (3H,
t, J = 7.1), 1.14–1.40 (12H, m, CH2), 2.95 (2H, q, J = 6.6,
NHCH2), 3.38 (1H, m. CHNHCO), 3.50–3–60 (m, 5H, CH),
4.4–5.00 (10H, m, 5 × OCH2Ph), 7.2 × 7.4 (25H, m, Ar). dC
(75 MHz, CDCl3, Me4Si) 14.1 (q), 22.7 (t), 26.8 (t), 29.1 (t),
29.2 (t), 29.6 (t), 29.8 (t), 31.1 (t), 40.4 (d), 75.3 (d), 75.5 (2 ×
d), 75.8 (d), 77.19 (2 × d), 80.4 (2 × t), 82.6 (t), 83.7 (2 × t),
127.6–128.4 (20 × d), 137.8 (2 × s), 138.48 (3 × s), 148.3 (s).
Found: C, 76.21; H, 7.62; N, 3.57%. C49H58N2O6 requires C,
76.33; H, 7.58; N, 3.63%.
N-Acylation: synthesis of 23–26. To a solution of the cor-
responding amine (0.060 mmol) and triethylamine (25 ll) in
anhydrous dichloromethane (1 cm3) were added 1.2 eq. per mol
of the corresponding acyl chloride. After stirring for 18 h at rt the
solution was washed with water (1 cm3). The organic layer was
dried and the solvent removed under a reduced pressure. The
residue was purified by column chromatography (hexane–ethyl
acetate, 10 : 1).
(1ꢀs,2ꢀR,3ꢀS,4ꢀr,5ꢀR,6ꢀS)-N-Phenethyl-N-(2,3,4,5,6-pentakis-
benzyloxycyclohexyl)octanamide
23. (82%
from
22).
mmax(film)/cm−1 3039, 3020, 2957, 2918, 1684, 1521, 1477,
1389, 1332. dH (500 MHz, CDCl3, Me4Si) 0.88 (3H, t, J = 6.5,
CH3), 1.20–1.40 (10H, m, CH2), 1.6–1.7 (m, 2H), 2.19 (2H, t,
J = 8.0), 2.76 (2H, t, J = 9.5, NHCH2CH2Ar), 2.97 (2H, m),
3.42 (2H, m), 3.51 (2H, t, J = 10.5), 3.69 (1H, t, J = 9.5), 4.60
(1H, d, J = 10.0), 4.67 (1H, t, J = 8.0), 4.81–5.02 (9H, m), 7.11
(2H, d, J = 5.0), 7.24–7.40 (28 H, m). dC (125 MHz, CDCl3,
Me4Si) 14.3 (q), 22.8 (t), 24.5 (t), 24.9 (t), 29.1 (t), 29.9 (t), 34.2
(t), 35.6 (t), 37.0 (t), 43.5 (t), 54.5 (d), 69.5 (d), 70.4 (d), 75.8 (d),
76.2 (d), 78.6 (d), 82.9 (t), 84.7 (t), 126.2 (d), 127.2–129.8 (29 ×
d), 138.4–138.7 (5 × s), 140.6 (d), 179.9 (s). Found: C, 76.01;
H, 7.39; N, 3.47%. C49H58N2O6 requires C, 76.33; H, 7.58; N,
3.63%.
(1ꢀs,2ꢀR,3ꢀS,4ꢀr,5ꢀR,6ꢀS)-N -(2,3,4,5,6-Pentakis-benzyloxy-
cyclohexyl)octanamide 24. (85% from 20); oil. mmax(film)/cm−1
3269, 3089, 3065, 3030, 2925, 2854, 1644, 1456, 1497, 1365,
1213. dH (300 MHz, CDCl3, Me4Si) 0.88 (3H, t, J = 7.2, CH3),
1.15–1.35 (10H, m, CH2), 1.99 (2H, t, J = 7.2, CH2CONHR),
3.60–3.68 (4H, m, CH), 3.88 (2H, m, CH), 4.62–4.95 (10H, m,
OCH2Ph), 5.43 (1H, d, J = 6.9, NH), 7.2–7.4 (25H, m, Ar). dC
(75 MHz, CDCl3, Me4Si) 14.1 (q), 22.6 (t), 25.56 (t), 29.3 (t),
29.4 (t), 29.7 (t), 31.1 (t), 55.5 (d), 74.9 (2 × d), 75.6 (2 × d),
75.7 (d), 78.9 (2 × t), 83.7 (t), 84.0 (2 × t), 127.6–128.4 (20 × d),
138.48 (5 × s), 173.49 (s). Found: C, 77.71; H, 7.82; N, 2.07%.
C49H57NO6 requires C, 77.85; H, 7.60; N, 1.85%.
(1ꢀs,2ꢀR,3ꢀS,4ꢀr,5ꢀR,6ꢀS)-N -(2,3,4,5,6-Pentakis-benzyloxy-
cyclohexyl)tetradecanamide 25. (87% from 20); oil.
mmax(film)/cm−1 3209, 3078, 3029, 2917, 2854, 1654, 1457,
1483, 1345. dH (300 MHz, CDCl3, Me4Si) 0.88 (3H, t, J =
7.2, CH3), 1.15–1.35 (26H, m, CH2), 1.99 (2H, t, J = 6.9,
CH2CONHR), 3.61–3.68 (4H, m, CH), 3.88 (2H, m, CH),
4.57–5.00 (10H, m, OCH2Ph), 5.69 (1H, d, J = 6.3, NH),
7.2–7.4 (25H, m, Ar). dC (75 MHz, CDCl3, Me4Si) 14.3 (q), 23.1
(t), 25.6 (t), 28.9 (t), 29.1 (2 × t), 29.3 (2 × t), 29.4 (2 × t), 29.5
(t), 29.6 (2 × t), 29.9 (t), 31.1 (t), 55.4 (d), 74.9 (2 × d), 75.6
(2 × d), 75.8 (d), 78.9 (2 × t), 83.6 (t), 84.2 (2 × t), 127.6–128.4
(20 × d), 138.7 (5 × s), 173.49 (s). Found: C, 78.51; H, 8.02; N,
1.75%. C55H69NO6 requires C, 78.63; H, 8.28; N, 1.67%.
(1ꢀs,2ꢀR,3ꢀS,4ꢀr,5ꢀR,6ꢀS)-N -(2,3,4,5,6-Pentakis-benzyloxy-
cyclohexyl)phenoxyacetamide 26. (81% from 20); oil.
mmax(film)/cm−1 3250, 3029, 2927, 1654, 1459, 1501, 1365.
dH (300 MHz, CDCl3, Me4Si) 3.63–3.68 (3H, m, CH), 3.82–3.87
(2H, m, CH), 4.62–4.95 (11H, m, CHN, OCH2Ph), 6.59 (1H, d,
J = 6.6, NH), 6.80 (2H, dd, J = 1.0, J = 7.8, CH), 7.01 (1H, t,
J = 7.6, CH), 7.2–7.4 (25H, m, Ar). dC (75 MHz, CDCl3, Me4Si)
55.2 (d), 67.0 (t), 75.1 (2 × d), 75.6 (2 × d), 75.7 (d), 78.9 (2 × t),
82.6 (t), 84.0 (2 × t), 114.5 (2 × d), 122.0 (s), 127.6–128.4 (20 ×
d), 129.7 (2 × d), 138.2 (5 × s), 156.8 (s), 168.4 (s). Found: C,
77.21; H, 6.62; N, 1.97%. C49H49NO7 requires C, 77.04; H, 6.47;
N, 1.83%.
Synthesis of amino alcohols 1–13 by debenzylation with boron
trichloride29. A solution of 0.05 mmol of the benzylated amino
alcohol in dichloromethane (2 cm3) under an atmosphere of
◦
nitrogen at −78 C, was treated with 2.5 eq. per benzyl group
of a 1 M solution of boron trichloride in heptane. After
stirring for 2 h at −78 ◦C, the reaction mixture was allowed
to warm to 0 ◦C and◦stirred for an additional 24 h. The mixture
was cooled to −78 C, quenched with methanol (1 cm3) and
evaporated under a reduced pressure. The resulting residue was
taken up in a 1 : 1 MeOH–H2O mixture, filtered through a
small pad of charcoal and lyophilised to afford the required
amino alcohols in quantitative yield. HRMS: 1: m/z calcd. for
C6H13NO5 (M + 1): 180.0872; found: 180.0866. 2: m/z calcd. for
C10H21NO5 (M + 1): 236.1498; found: 236.1507. 3: m/z calcd.
for C14H21NO5 (M + 1): 284.1498; found: 284.1503. 4: m/z
calcd. for C14H29NO5 (M + 1): 292.2124; found: 292.2119. 5:
m/z calcd. for C14H27NO6 (M + 1): 306.1916; found: 306.1905.
6: m/z calcd. for C20H39NO6 (M + 1): 390.2855; found: 390.2867.
7: m/z calcd. for C14H19NO7 (M + 1): 314.1239; found: 314.1233.
8: m/z calcd. for C15H30N2O6 (M + 1): 335.2182; found:
335.2193. 9: m/z calcd. for C22H35NO6 (M + 1): 410.2542; found:
410.2553; 10: m/z calcd. for C6H13NO5 (M + 1): 180.0872;
found: 180.0867. 11: m/z calcd. for C10H21NO5 (M + 1):
236.1498; found: 236.1504. 12: m/z calcd. for C14H21NO5 (M +
1): 284.1498; found: 284.1487. 13: m/z calcd. for C14H29NO5
(M + 1): 292.2124; found: 292.2137.
(1s,2S,3R,4r,5S,6R)-1-Azido-2,3,4,5,6-pentakis-benzyolxy-
cyclohexane 19. A solution of alcohol 18 (1.4 g, 3,0 mmol) in
1 2 0 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 9 5 – 1 2 0 1

View Article Online
12 B. Rickborn and D. K. Murphy, J. Org. Chem., 1969, 34, 3209.
Acknowledgements
13 E. L. Eliel, N. L. Allinger, S. J. Angyal and G. A. Morrison,
Conformational Analysis, Wiley, New York, 1965, p. 352.
14 P. Serrano, A. Llebaria and A. Delgado, J. Org. Chem., 2002, 67,
7165–7167.
Financial support from Ministerio de Ciencia y Tecnolog´ıa
(Spain), Project BQU2002-03737 and DURSI (Generalitat de
Catalunya), Projects 2001SGR00085 and 2001SGR00342 is
acknowledged. PS thanks Ministerio de Educacio´n y Cultura
(Spain) for a predoctoral fellowship. The authors also thank
Prof. Futerman for X-ray structural data and Genzyme for a
15 Manuscript in preparation.
16 H. S. Overkleeft, G. H. Renkema, J. Neele, P. Vianello, I. O. Hung,
A. Strijland, A. M. van der Burg, G. J. Koomen, U. K. Pandit and
J. M. Aerts, J. Biol. Chem., 1998, 273, 26522–26527.
17 In the case of a-glucosidase, the non-competitive inhibition shown
by the aminocyclitols described in this work precludes any rational-
ization of their structure–activity relationships based on comparison
with the enzyme substrate.
R
generous gift of Imiglucerase (Cerezymeꢀ).
References
1 Carbohydrate Mimics, ed. Y. Chapleur, Wiley-VCH, Weinheim, 1998.
2 See inter-alia: (a) S. Ogawa, S. Fujieda, Y. Sakata, M. Ishizaki, S.
Hisamatsu and K. Okazaki, Bioorg. Med. Chem. Lett., 2003, 13,
3461–3463; (b) K. S. Tanaka, G. C. Winters, R. J. Batchelor, F. W.
Einstein and A. J. Bennet, J. Am. Chem. Soc., 2001, 123, 998–999;
(c) C. Gravier-Pelletier, W. Maton, T. Dintinger, C. Tellier and Y. Le
Merrer, Tetrahedron, 2003, 59, 8705–8720, and references therein.
3 R. A. Dwek, T. D. Butters, F. M. Platt and N. Zitzmann, Nat. Rev.
Drug Discovery, 2002, 1, 65–75.
18 There are more than 60 complexes of the inhibitor acarbose (a
tetrasaccharide containing the unsaturated aminocyclitol valien-
amine) with different a-glucosidase proteins. For a recent example
see; G. D. Brayer, G. Sidhu, R. Maurus, E. H. Rydberg, C. Braun, Y.
Wang, N. T. Nguyen, C. M. Overall and S. G. Withers, Biochemistry,
2000, 39, 4778–4791.
19 A. H. Futerman, J. L. Sussman, M. Horowitz, I. Silman and A.
Zimran, Trends Pharmacol. Sci., 2004, 25, 147–151.
20 A. H. Futerman, private communication. See also; H. Dvir, M. Harel,
A. A. McCarthy, L. Toker, I. Silman, A. H. Futerman and J. L.
Sussman, EMBO Rep., 2003, 4, 704–709.
4 A. Gonzalez-Roura, I. Navarro, A. Delgado, A. Llebaria and J.
Casas, Angew. Chem., Int. Ed., 2004, 43, 862–865.
5 G. Triola, G. Fabrias, J. Casas and A. Llebaria, J. Org. Chem., 2003,
68, 9924–9932.
6 G. Triola, G. Fabrias and A. Llebaria, Angew. Chem., Int. Ed., 2001,
40, 1960–1962.
7 S. Brodesser, P. Sawatzki and T. Kolter, Eur. J. Org. Chem., 2003,
2021–2034.
8 T. Kolter and K. Sandhoff, Angew. Chem., Int. Ed., 1999, 38, 1532–
1568.
21 For the sake of comparison, we have used in this study the
corresponding N-butyl derivatives 2 (C1 adduct) and 11 (C2 adduct).
22 This structural feature is unprecedented in amino sugar glycosidase
inhibitors.
23 DSViewerPro; v 5.0, Accelrys Inc., San Diego, CA, 2002.
24 For the potential of chemical chaperones as new drugs for the
treatment of sphingolipidoses see ref. 19 and also; (a) J.-Q. Fan,
Trends Pharmacol. Sci., 2003, 24, 355–360; (b) A. R. Sawkar, W. C.
Cheng, E. Beutler, C. H. Wong, W. E. Balch and J. W. Kelly, Proc.
Natl. Acad. Sci. U S A, 2002, 99, 15428–15433.
9 C. Jaramillo, J. Chiara and M. Mart´ın-Lomas, J. Org. Chem., 1994,
59, 3135–3141.
10 M. Bartok and K. L. Lang, in Small Ring Heterocycles, ed.
A. Hassner, Wiley, New York, 1985, pp. 1–196.
25 D. D. Perrin and W. L. F. Armarego, Purification of Laboratory
Chemicals, Pergamon Press, Oxford, 3rd edn., 1988.
26 P. Paul, Y. Kamisaka, D. L. Marks and R. E. Pagano, J. Biol. Chem.,
1996, 271, 2287–2293.
27 F. M. Platt, G. R. Neises, G. B. Karlsson, R. A. Dwek and T. D.
Butters, J. Biol. Chem., 1994, 269, 27108–27114.
28 T. Kajimoto, K. Liu, R. Pederson, Z. Zhong, Y. Ichikawa, J. J. Porco
and C.-H. Wong, J. Am. Chem. Soc., 1991, 113, 6187–6196.
29 A. Arcelli, V. Cere, F. Peri, S. Pollicino and A. Ricci, Tetrahedron,
2001, 57, 3439–3444.
11 For C2 regioselective opening of related O-protected cyclophellitol
derivatives with NaN3 see; (a) K. Tatsuta and S. Miura, Tetrahedron
Lett., 1995, 36, 6721–6724; (b) S. Ogawa, N. Chida and T. Suami,
J. Org. Chem., 1983, 48, 1203–1207. For reactions of cyclitol epoxides
with amines see; (c) R. A. Cadenas, M. Y. Grass, J. Mosettig and
M. E. Gelpi, Nucleosides Nucleotides, 1990, 9, 21–34; (d) S. Leicach,
M. E. Gelpi and R. A. Cadenas, Nucleosides Nucleotides, 1994, 13,
2051–2058.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 9 5 – 1 2 0 1
1 2 0 1