Iminosugar-based galactoside mimics as inhibitors of galactocerebrosidase: SAR studies and comparison with other lysosomal galactosidases

Article abstract of DOI:10.1002/cmdc.201402411

Several families of iminosugar-based galactoside mimics were designed, synthesized, and evaluated as galactocerebrosidase (GALC) inhibitors. They were also tested as inhibitors of lysosomal β- and α-galactosidases in order to find new potent and selective pharmacological chaperones for treatment of the lysosomal storage disorder, Krabbe disease. Whereas 1-C-alkyl imino-l-arabinitols are totally inactive toward the three enzymes, 1-C-alkyl imino-L-galactitols were found to be active only toward α-galactosidase A. Finally, 1-N-iminosugars provided the best results, as 4-epi-isofagomine was found to be a good inhibitor of both lysosomal β-galactosidase and GALC. Further elaboration of this structure is required to achieve selectivity between these two galactosidases.

Full text of DOI:10.1002/cmdc.201402411

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201402411
Iminosugar-Based Galactoside Mimics as Inhibitors of
Galactocerebrosidase: SAR Studies and Comparison with
Other Lysosomal Galactosidases
[a]
[a]
[a]
[a]
´
Anna Biela-Banas, Farah Oula¨ıdi, Sophie Front, Estelle Gallienne, Kyoko Ikeda-
Obatake,[b] Naoki Asano,[b] David A. Wenger,[c] and Olivier R. Martin*[a]
Several families of iminosugar-based galactoside mimics were
designed, synthesized, and evaluated as galactocerebrosidase
(GALC) inhibitors. They were also tested as inhibitors of lysoso-
mal b- and a-galactosidases in order to find new potent and
selective pharmacological chaperones for treatment of the ly-
sosomal storage disorder, Krabbe disease. Whereas 1-C-alkyl
imino-l-arabinitols are totally inactive toward the three en-
zymes, 1-C-alkyl imino-d-galactitols were found to be active
only toward a-galactosidase A. Finally, 1-N-iminosugars provid-
ed the best results, as 4-epi-isofagomine was found to be
a good inhibitor of both lysosomal b-galactosidase and GALC.
Further elaboration of this structure is required to achieve se-
lectivity between these two galactosidases.
Introduction
Deficiencies in the lysosomal enzyme b-galactocerebrosidase
(GALC, EC 3.2.1.46) are responsible for Krabbe disease, a neuro-
degenerative lysosomal storage disorder (LSD), also known as
globoid-cell leukodystrophy.[1] Mutations in the GALC gene
result in decreased enzyme activity and in the accumulation of
its substrates, galactosylceramide, a major component of
myelin, as well as psychosine (b-galactosylsphingosine). This
latter compound is cytotoxic and is responsible for apoptosis
of the myelin-forming cells, leading to demyelination.[2] Most
patients with Krabbe disease suffer from rapidly progressing
leukoencephalopathy before the age of six months; death gen-
erally occurs before two years. Krabbe disease is very rare (1 in
100000 births), but there are important variations among dif-
ferent countries.
chaperone therapy (PCT), which consists of using small mole-
cules to stabilize the mutant enzyme structure and to assist its
transport to the lysosome.[6] This approach generally relies on
the availability of potent and reversible inhibitors of the corre-
sponding enzyme; such compounds have strong interactions
with the enzyme, they can be used at low concentrations, and
can be displaced in the lysosome by high substrate concentra-
tions.[7] Iminosugars, which are known potent glycosidase in-
hibitors,[8] are a family of compounds with great promise as
potential pharmacological chaperones for LSDs.[9] One such
compound, 1-deoxygalactonojirimycin (DGJ), has reached clini-
cal trials for Fabry disease (an a-galactosidase-linked LSD).[10]
As regards GALC, recent pioneering studies by Eckman and co-
workers[11] as well as Luddi and colleagues[12] have shown that
several mutations of this enzyme are amenable to the chaper-
one effect by using a-lobeline or a flavonoid (3’,4’,7-trihydrox-
yisoflavone): a promising result in the perspective of using PCT
for Krabbe disease. A high-throughput assay using patient cells
has also been developed recently to screen small molecules as
chaperones.[13]
While a diversity of iminosugar[14] and aminocyclitol[15] deriv-
atives have been shown to be potent inhibitors of lysosomal
b-galactosidase (b-Gal), no information is yet available on the
activity of such compounds on GALC. The evaluation of galac-
tose-related iminosugars toward GALC and comparison of their
activity as human b-Gal inhibitors are very important in ad-
dressing the challenging question of selectivity of the new
compounds for either b-galactosidase. We have therefore un-
dertaken a study of several types of iminosugar-based galacto-
side mimics as inhibitors of GALC, and herein we report our in-
itial results. They allowed us to gather important structure–ac-
tivity relationships, which should help us design potent and se-
lective pharmacological chaperones for the treatment of
Krabbe disease.
There is currently no effective treatment for Krabbe disease.
Hematopoietic stem cell transplantation using bone marrow or
umbilical cord blood cells was shown to slow the clinical
course of the disease when performed on pre-symptomatic pa-
tients.[3] Direct intracerebral injection of the recombinant
enzyme[4] or virus-mediated gene transfer[5] in mouse models
of the disease have been shown to correct enzyme deficiency,
but these techniques have not yet led to clinical applications.
An emerging therapeutic strategy for LSDs is pharmacological
´
¨
[a] Dr. A. Biela-Banas, Dr. F. Oulaıdi, S. Front, Dr. E. Gallienne, Prof. O. R. Martin
Institut de Chimie Organique et Analytique (ICOA), UMR 7311
Universitꢀ d’Orlꢀans & CNRS
Rue de Chartres, BP 6759, 45067 Orlꢀans cedex 2 (France)
E-mail: olivier.martin@univ-orleans.fr
[b] Dr. K. Ikeda-Obatake, Dr. N. Asano
Faculty of Pharmaceutical Sciences
Hokuriku University, Ho-3 Kanagawa-machi, Kanazawa 920-1181 (Japan)
[c] Dr. D. A. Wenger
Department of Neurology, Jefferson Medical College
1020 Locust Street, Room 346, Philadelphia, PA 19107 (USA)
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 2647 – 2652 2647

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
a 2,3-dihydroxypropyl aglycone were prepared by previously
reported methods.[17,19] The synthesis of DGJ derivatives carry-
ing a C-linked aglycone in the pseudo-anomeric position was
particularly challenging, but eventually provided the a- and b-
galactoside mimics 9–11.[19] The results of the biochemical
studies are reported in Table 2. In general, all compounds are
Results and Discussion
Several families of iminosugar-based galactoside mimics were
designed and synthesized. The new compounds were then
evaluated as GALC inhibitors, as well as inhibitors of lysosomal
b-Gal and a-galactosidase (a-Gal). Defects in these enzymes
are responsible for GM1-gangliosidosis and Morquio B disease,
and for Fabry disease, respectively.
Table 2. Inhibition of GALC, human lysosomal b-galactosidase, and
human a-galactosidase A by compounds 5–11 at 1 mm.
Imino-l-arabinitols: We have shown in previous studies that
1-C-alkyl iminoxylitol derivatives, designed as simplified b-glu-
coside mimics lacking the 5-CH2OH group, are very potent and
selective inhibitors of b-glucocerebrosidase (GCase), the
enzyme involved in Gaucher disease; in addition, these com-
pounds behaved as pharmacological chaperones for N370S
and L444P GCase, the most important mutant forms of the
enzyme responsible for Gaucher disease.[16] We first determined
whether 1-C-alkyl imino-l-arabinitols, which are galactoside
mimics lacking the 5-CH2OH group, would also be good galac-
topyranoside surrogates: compounds 1–4, prepared as previ-
ously reported,[17] are mimics of b- and a-galactosides carrying
a simple lipophilic chain (series 1 and 2, respectively), and of
galactosides carrying a more polar aglycone (compounds 3
and 4), as is present in the natural substrates. However, all of
these compounds were found to be devoid of activity as a- or
b-Gal inhibitors (Table 1), a result which showed that in all
Compd
Structure
Inhibition [%][a]
GALC
NI
b-Gal
a-Gal A
5
20
8
59
56
69
97
96
96
6
7
24
NI
NI
NI
19
20
35
26
28
37
NI
8
Table 1. Inhibition of GALC, human lysosomal b-galactosidase, and
human a-galactosidase A by compounds 1–4 at 1 mm.
Compd
Structure
GALC
b-Gal
a-Gal A
9
n=2 ND[a]
n=4 NI[b]
n=7 ND
n=2 NI
n=4 NI
n=7 NI
n=2 NI
n=4 NI
n=7 NI
1
10
11
n=2 ND
n=4 NI
n=7 ND
n=2 NI
n=4 NI
n=7 NI
n=2 NI
n=4 NI
n=7 NI
2
26
(R)-3
(S)-3
NI
NI
NI
NI
NI
NI
[a] Data are the mean of triplicate experiments; errors are in the range of
Æ2.5%. NI: no inhibition at 1 mm.
4
better inhibitors of a-galactosidase than of the b-galactosidas-
es. This is consistent with the analysis by Heightman and Vasel-
la[20] and the studies by Bols and colleagues[21] on the mecha-
nism of recognition of iminosugars by a- and b-glycosidases,
1-deoxynojirimycin-like structures being generally better inhibi-
tors of a-glycosidases, whereas so-called 1-N-iminosugars (re-
lated to isofagomine, IFG) behave as better b-glycosidase in-
hibitors. Furthermore, as expected, d-galacto compounds are
much better inhibitors than l-altro epimers, with compounds
9–11 being particularly potent inhibitors of a-galactosidase A.
Interestingly, the ‘anomeric’ configuration of these compounds
[a] ND: not determined [b] NI: no inhibition at 1 mm.
three cases, the presence of the CH2OH group is mandatory.
Such an observation had also been made by Ganem and co-
workers[18] in the evaluation of the activity of unsubstituted
1,5-imino-l-arabinitol toward coffee bean a-Gal and bovine b-
Gal.
Imino-l-altritols and -d-galactitols: We next investigated
hexopyranoside mimics in the l-altro and d-galacto series. The
1-C-alkyl imino-l-altritol derivatives 5–8 carrying an alkyl or
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 2647 – 2652 2648

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
and the alkyl group chain length do not appear to play an im-
portant role.
Table 3. Inhibition of GALC, human lysosomal b-galactosidase, and
human a-galactosidase A by compounds 12–15.
1-N-Iminosugar derivatives: To identify more potent b-Gal
inhibitors, we turned our attention to the family of 1-N-imino-
sugars. The 1,5-iminoribitol 12, a simple compound with a 1-N-
iminosugar-like structure and a configuration resembling that
of galactose, was prepared by known procedures.[22] 4-Epi-iso-
fagomine derivatives 13 and 14, carrying a hydroxyl group at
C5, were also prepared according to published procedures:
compound 13 was reported by Ichikawa[23] in the course of
their investigations on 4-epi-isofagomine, and the 5-epimer of
13, compound 14, was recently described by Fleet and co-
workers.[24] Finally, we prepared 4-epi-isofagomine 15 (previous
syntheses by Ichikawa et al.[22] and others[25]) by way of a new
stereoselective synthesis, as outlined in Scheme 1. The synthe-
Compd
Structure
Inhibition [%] (IC50 [mm])[a]
GALC
b-Gal
a-Gal A
12[b]
NI
45
NI
NI
67
95
13[c]
(450)
(50)
14[c]
NI
7
13
NI
98
(1)
99.7
(1)
15[b]
[a] Data represent the mean of triplicate experiments; errors are in the
range of Æ2.5%. NI: no inhibition at 1 mm. [b] Determined at 1 mm.
[c] Determined at 1.2 mm.
sugar). Compound 13 had been shown by Ichikawa[22] to be
a fairly good inhibitor of Aspergillus oryzae b-Gal (IC50 =
17.5 mm), but it is not an inhibitor of bovine b-Gal, as has also
been observed for compound 14.[24]
4-Epi-isofagomine 15 is clearly the most interesting com-
pound of the series. It is already known to be a powerful inhib-
itor of b-Gal of A. oryzae[22] (IC50 =12 nm), of bovine liver (IC50 =
21 mm),[25b] and of rat intestinal lactase (IC50 =0.51 mm).[25b]
Herein it is shown for the first time to be a potent inhibitor of
human lysosomal b-Gal, as well as of GALC, with, however,
little discrimination between the two enzymes.1
Scheme 1. Reagents and conditions: a) BnOH, p-TsOH, 608C, 78 h, 92%;
b) (MeO)2CMe2, p-TsOH, acetone, RT, 16 h, 48%; c) p-TsCl, pyridine, RT, 40 h,
91%; d) AcOH, H2O, 1108C, 85 min, 97%; e) tBuOK, THF, RT, 45 min, 71%;
f) TIPS-OTf, 2,6-lutidine, DMF, RT, 16 h, 82%; g) Et2AlCN, Et2O, 458C, 3.5 h,
73%; h) Bu4NF, THF, RT, 3.5 h, 89%; i) H2, 20% Pd(OH)2/C, H2O/1,4-dioxane
(1:4), HCl, RT, 21 h, then H2, 10% Pd/C, H2O/1,4-dioxane (1:1), AcOH, RT, 30 h,
purification on H+ ion-exchange resin, quant.
Thus, further modifications of the iminosugar core structure
will be required to achieve selectivity between the two types
of b-galactosidases. The recent availability of the X-ray struc-
tures of the two enzymes[29,30] should facilitate the design of
selective inhibitors of each enzyme, in particular by involving
interactions with the aglycone, which is quite different in the
substrates of human bGal and of GALC.
sis proceeds by means of a branched nitrile (compound 22) of
unequivocal configuration and therefore can lead only to the
desired d-galacto-like isomer of the final product. The nitrile
22 was obtained in excellent yield from benzyl 2,3-O-isopropy-
lidene-a-d-lyxopyranoside 16[26] by way of a procedure adapt-
ed from Vasella and colleagues;[27] notably, the introduction of
a triisopropylsilyl (TIPS) group at O2 is required to ensure the
regioselective opening of the 3,4-anhydro ring at C4. The final
step proceeds from nitrile 22 into 4-epi-isofagomine 15 in high
yield under hydrogenolytic conditions, using Pearlman’s cata-
lyst then palladium on charcoal in succession in acidic
medium.
Conclusions
A series of iminosugar-based galactoside mimics have been
evaluated as GALC inhibitors, in comparison with human lyso-
somal b-galactosidase and a-galactosidase A, in order to assess
selectivity. These studies have shown that the most active
compounds are in the 1-N-iminosugar series, and that 4-epi-
isofagomine is a suitable lead compound in the search for effi-
cient inhibitors of both GALC and human b-galactosidase. Fur-
All compounds were tested as inhibitors of the three galac-
tosidases (Table 3). As anticipated, these 1-N-iminosugars are
not inhibitors of a-galactosidase A. Iminoribitol 12 exhibited
only a very modest inhibition of human b-Gal. Compound 13
was found to be a good inhibitor of the lysosomal enzyme,
and to have only a modest effect on GALC, whereas this activi-
ty is essentially lost for its C5 epimer 14 (a pseudo-l-1-N-imino-
1
While these SAR studies suggest that N-alkylated derivatives of 15 would be
of particular interest, preliminary investigations showed that N-alkylation of
15 leads to a significant decrease in its activity as a b-Gal inhibitor.[28]
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 2647 – 2652 2649

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
3.97 (dd, 1H, J=2.8, 3.5 Hz, H2), 4.02 (dd, 1H, J=3.5, 8.3 Hz, H3),
4.47 (d, 1H, J=11.8 Hz, CH2Ph), 4.63 (dt, 1H, J=4.8, 8.3, 8.3 Hz,
H4), 4.71 (d, 1H, J=11.8 Hz, CH2Ph), 4.81 (d, 1H, J=2.8 Hz, H1),
7.12–7.41 (m, 7H, HAr), 7.82 ppm (d, 2H, J=8.3 Hz, HAr); 13C NMR
(100 MHz, CDCl3): d=21.7 (CH3Ts), 60.0 (C5), 69.4 (CH2Ph), 68.9 (C3),
70.5 (C2), 77.3 (C4), 98.6 (C1), 128.0–130.0 (CHAr), 133.0–145.3 ppm
(CAr); HRMS (ESI): m/z calcd for C19H26NO7S: 412.142450 [M+NH4]+,
found: 412.142255, m/z calcd for C19H22NaO7S: 417.097845 [M+
Na]+, found: 417.097405.
ther elaboration of this structure may provide a means to ach-
ieve the required selectivity for such compounds to be consid-
ered as potential pharmacological chaperones for diseases as-
sociated with mutations of these enzymes.
Experimental Section
Materials and methods: All reactions requiring anhydrous condi-
tions were carried out using oven-dried glassware under an atmos-
phere of dry Ar. THF was dried through a column containing acti-
vated alumina under nitrogen pressure (Dry Solvent Station GT
S100, GlassTechnology, Geneva, Switzerland). Pyridine and DMF
were dried over activated 4 ꢁ molecular sieves. Et2O was distilled
from CaH. All reagent-grade chemicals were obtained from com-
mercial suppliers and were used as received. Low-resolution mass
spectra were recorded on a PerkinElmer Sciex API 3000. High-reso-
lution mass spectra were recorded on a Bruker Q-TOF MaXis spec-
Benzyl 3,4-anhydro-b-l-ribopyranoside (19): A cold solution of
compound 18 (6.38 g, 16.19 mmol) in anhydrous THF (40 mL) was
added at 08C under Ar to
a suspension of tBuOK (2.18 g,
19.40 mmol, 1.2 equiv) in anhydrous THF (160 mL). The reaction
was stirred for 45 min at room temperature, and a saturated NH4Cl
solution (60 mL) was then added. After 30 min of stirring, the mix-
ture was extracted with EtOAc (2ꢃ160 mL). The organic phase was
dried over MgSO4 and evaporated under vacuum. The residue was
purified by flash chromatography on silica gel (PE/EtOAc 1:1) to
give 19 (2.55 g, 71%) as a white solid; mp: 66–678C; [a]D = +157.0
(c=1.1, CHCl3); 1H NMR (400 MHz, CDCl3): d=2.73 (d, 1H, J=
10.0 Hz, OH), 3.34 (d, 1H, J=4.4 Hz, H4), 3.51 (t, 1H, J=4.4 Hz, H3),
3.82 (ddd, 1H, J=2.0, 4.4, 10.0 Hz, H2), 3.94 (d, 1H, J=13.4 Hz,
H5a), 4.01 (dd, 1H, J=1.6, 13.4 Hz, H5b), 4.50 (d, 1H, J=11.6 Hz,
CH2Ph), 4.58 (d, 1H, J=2.0 Hz, H1), 4.72 (d, 1H, J=11.6 Hz, CH2Ph),
7.26–7.40 ppm (m, 5H, HAr); 13C NMR (100 MHz, CDCl3): d=51.4
(C3), 51.7 (C4), 58.0 (C5), 64.7 (C2), 69.6 (CH2Ph), 97.9 (C1), 127.9–
128.5 (CHAr), 137.0 ppm (CAr); HRMS (ESI): m/z calcd for C12H18NO4:
240.123034 [M+NH4]+, found: 240.123181, m/z calcd for
C12H14NaO4: 245.078430 [M+Na]+, found: 245.078620.
1
trometer. H and 13C NMR spectra were recorded on Bruker Avance
DPX 250 or Bruker Avance 400 spectrometers. Chemical shifts are
given in ppm and are referenced to the residual solvent signal or
to TMS as internal standard. Carbon multiplicities were assigned by
distortionless enhancement by polarization transfer (DEPT) experi-
1
ments. H and 13C signals were attributed on the basis of H–H and
C–H correlations. Specific rotations were measured at 208C in a 1-
dm cell with a PerkinElmer 341 polarimeter. Melting points were
determined in capillary tubes with a Bꢂchi apparatus and are un-
corrected. Analytical thin-layer chromatography was performed
using silica gel 60 F254 pre-coated plates (Merck) with visualization
by UV light and phosphomolybdic acid solution (2% in H2SO4/
EtOH 1:7). Flash chromatography was performed with silica gel 60
(230–400 mesh).
Benzyl 3,4-anhydro-2-O-triisopropylsilyl-b-l-ribopyranoside (20):
Compound 19 (1.84 g, 8.28 mmol) was dissolved under Ar in dry
DMF (18 mL). 2,6-Lutidine (2.2 mL, 18.88 mmol, 2.3 equiv) was
added, the mixture was cooled to 08C, and triisopropylsilyltriflate
(TIPS-OTf; 2.23 mL, 8.28 mmol, 1 equiv) was added. The reaction
was stirred at room temperature overnight and diluted with EtOAc
(60 mL). The organic phase was washed with a 2n HCl solution
(2ꢃ10 mL), water (10 mL), brine (10 mL), dried over MgSO4 and
evaporated under high vacuum. The residue was purified by flash
chromatography on silica gel (PE/EtOAc 95:5) to give 20 (2.57 g,
Benzyl 2,3-O-isopropylidene-4-O-(4-methylphenylsulfonyl)-a-d-
lyxopyranoside (17): Compound 16 (5.33 g, 0.019 mol) was dis-
solved in anhydrous pyridine (20 mL) under Ar. Toluenesulfonyl (Ts)
chloride (5 g, 0.026 mol, 1.4 equiv) was added, and the reaction
mixture was stirred for 40 h at room temperature. The mixture was
diluted with Et2O (25 mL) and washed with a 3n HCl solution (2ꢃ
25 mL), H2O (2ꢃ25 mL) and brine (25 mL). The organic phase was
dried over MgSO4 and concentrated under vacuum. The residue
was purified by flash chromatography on silica gel (PE/EtOAc 9:1)
to give 17 (7.47 g, 91%) as a white solid; mp: 88–908C; 1H NMR
(400 MHz, CDCl3): d=1.14, 1.24 (s, 3H, CH3iPr), 2.44 (s, 3H, CH3Ts),
3.71 (dd, 1H, J=9.6, 11.6 Hz, H5a), 3.85 (dd, 1H, J=5.2, 11.6 Hz,
H5b), 4.11 (dd, 1H, J=1.2, 5.2 Hz, H2), 4.18 (dd, 1H, J=6.8, 5.2 Hz,
H3), 4.41 (ddd, 1H, J=5.2, 6.8, 9.6 Hz, H4), 4.49 (d, 1H, J=11.6 Hz,
CH2Ph), 4.72 (d, 1H, J=11.6 Hz, CH2Ph), 4.92 (d, 1H, J=1.2 Hz, H1),
7.27–7.39 (m, 7H, HAr), 7.85 ppm (d, 2H, J=8.3 Hz, HAr); 13C NMR
(100 MHz, CDCl3): d=21.6 (CH3Ts), 26.1, 27.3 (CH3iPr), 58.6 (C5), 69.4
(CH2Ph), 74.5 (C3), 75.6 (C2), 77.0 (C4), 96.3 (C1), 109.7 (CiPr), 128.1–
129.8 (CHAr), 133.1–145.0 ppm (CAr); MS (ESI) m/z=435.0 [M+H]+,
457.0 [M+Na]+.
1
82%) as a colorless oil. H NMR (250 MHz, CDCl3): d=1.06–1.12 (m,
21H, TIPS), 3.32–3.37 (m, 1H, H4), 3.42 (t, 1H, J=3.8 Hz, H3), 3.96
(d, 1H, J=13.4 Hz, H5a), 4.02 (t, 1H, J=3.8 Hz, H2), 4.08 (dd, 1H,
J=2.4, 13.4 Hz, H5b), 4.53 (d, 1H, J=11.7 Hz, CH2Ph), 4.53 (d, 1H,
J=3.8 Hz, H1), 4.75 (d, 1H, J=11.7 Hz, CH2Ph), 7.24–7.38 ppm (m,
5H, HAr); 13C NMR (100 MHz, CDCl3): d=12.4, 18.0 (TIPS), 52.5 (C4),
53.6 (C3), 60.0 (C5), 68.6 (C2), 70.1 (CH2Ph), 99.6 (C1), 127.8–128.3
(CHAr), 137.3 ppm (CAr); HRMS (ESI): m/z calcd for C21H35O4Si:
379.229913 [M+H]+, found: 379.229719, m/z calcd for
C21H38NO4Si: 396.256462 [M+NH4]+, found: 396.256342.
Benzyl 4-cyano-4-deoxy-2-O-triisopropylsilyl-a-d-lyxopyranoside
(21): A 1m solution of Et2AlCN in toluene (7.26 mL, 7.26 mmol,
1.1 equiv) was added dropwise under argon at 08C to a solution of
compound 20 (2.54 g, 6.71 mmol) in anhydrous Et2O (40 mL). The
reaction was stirred for 3.5 h at 458C, then 1 h at room tempera-
ture and then cooled to À408C. A saturated solution of NH4Cl
(14 mL) was added dropwise, and the mixture was stirred for 2 h at
room temperature. The solid was filtered and washed with EtOAc
(2ꢃ10 mL). The filtrate was evaporated under vacuum, and the res-
idue was purified by flash chromatography on silica gel (toluene/
acetone 99.5:0.5) to give 21 (1.951 g, 73%) as a colorless oil.
1H NMR (250 MHz, CDCl3): d=1.01–1.10 (m, 21H, TIPS), 2.42 (d, 1H,
J=10.0 Hz, OH), 3.06 (dt, 1H, J=5.5, 10.0, 10.0 Hz, H4), 3.81–3.92
Benzyl 4-O-(4-methylphenylsulfonyl)-a-d-lyxopyranoside (18):
Compound 17 (7.45 g, 0.017 mol) was dissolved in glacial acetic
acid (28 mL), and the solution was heated at 1108C for 5 min.
Water (7 mL) was then added, and the mixture was maintained at
110 8C for 80 min. The solvents were evaporated under vacuum,
and the residue was dissolved in Et2O (80 mL). The organic phase
was washed with H2O (10 mL), with a saturated NaHCO3 solution
(2ꢃ10 mL) and dried over MgSO4. The solvents were removed
under vacuum to give 18 (6.56 g, 97%), which was used in the
next step without further purification. 1H NMR (250 MHz, CDCl3):
d=2.45 (s, 3H, CH3Ts), 2.89 (brs, 1H, OH), 3.65–3.68 (m, 2H, H5),
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 2647 – 2652 2650

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
(m, 2H, H5), 3.99 (t, 1H, J=2.7 Hz, H2), 4.10 (dt, 1H, J=2.7, 10.0,
10.0 Hz, H3), 4.49 (d, 1H, J=11.8 Hz, CH2Ph), 4.74 (d, 1H, J=
11.8 Hz, CH2Ph), 4.77 (d, 1H, J=2.7 Hz, H1), 7.25–7.44 (m, 5H, HAr);
13C NMR (100 MHz, CDCl3): d=12.5, 17.9–18.0 (TIPS), 32.5 (C4), 59.3
(C5), 67.8 (C3), 69.5 (CH2Ph), 70.1 (C2), 98.6 (C1), 118.4 (CN), 128.4–
128.7 (CHAr), 136.6 ppm (CAr). HRMS (ESI): m/z calcd for C22H36NO4Si:
406.240812 [M+H]+, found: 406.240625.
sample used in the assay (as determined by the method of Lowry
et al.).[32] Activity is expressed as nmol substrate hydrolyzed per
hour per (mg protein).
GALC: Stock solutions of [3H]GalCer, sodium taurocholate
(10 mgmLÀ1), and oleic acid (1 mgmLÀ1) in 2:1 CHCl3/MeOH were
prepared. The specific radioactivity of the tritiated substrate was
500–800 cpmnmolÀ1. Equal volumes of the sodium taurocholate
and oleic acid solutions were mixed. These solutions were stored
at 48C. In addition, a ‘theoretical lower phase’ solution was pre-
pared by mixing a 100 mL solution of galactose (0.5 mgmLÀ1) and
a 2:1 mixture of CHCl3 and MeOH (500 mL) in a separatory funnel,
shaking the mixture, and separating the lower phase. For the
assay, 20–25 nmol of [3H]GalCer and 0.1 mL of taurocholate-oleic
acid solutions were mixed, and the organic solvents evaporated in
an assay tube. Then 0.1 mL of citrate-phosphate buffer (0.2m in
phosphate, pH 4.2) was added, followed by the enzyme source (~
50 mg protein from a leukocyte sonicate) and distilled water up to
0.1 mL (total incubation volume was 0.2 mL). The mixture was in-
cubated for 2 h. Then it was placed on ice, a solution (0.1 mL) of
galactose (0.5 mgmLÀ1 H2O) was added, followed by a 2:1 mixture
of CHCl3 and MeOH (1.5 mL), the mixture was vortexed and then
centrifuged at 1200 g for 3 min. The lower phase was removed
with a pipette or a syringe, and saved in a waste tube. The lower
phase was replaced by adding 1 mL of ’theoretical lower phase’ so-
lution to the assay tube, the mixture was vortexed and centrifuged
as above. The upper phase was carefully removed and transferred
to a scintillation vial; scintillation fluid (10 mL) was added, and the
tritiated galactose corresponding to hydrolyzed [3H]GalCer was
counted. Specificity of the enzyme was expressed in
Benzyl 4-cyano-4-deoxy-a-d-lyxopyranoside (22): A 1m solution
of TBAF in THF (7.5 mL, 7.5 mmol, 1.9 equiv) was added at 08C to
a solution of compound 21 (1.585 g, 3.91 mmol) in THF (50 mL).
The reaction was stirred for 3.5 h at room temperature, and the
solvent was then removed under vacuum. The residue was purified
by flash chromatography on silica gel (CH2Cl2/acetone 9:1) to give
22 (862 mg, 89%) as a white solid; mp: 144–1468C; [a]D = +72.6
1
(c=1.1, CHCl3); H NMR (400 MHz, CD3OD), d=3.15 (dt, 1H, J=4.8,
10.7, 10.7 Hz, H4), 3.74 (t, 1H, J=2.4 Hz, H2), 3.82 (t, 1H, J=
10.7 Hz, H5a), 3.90 (dd, 1H, J=4.8, 10.7 Hz, H5b), 3.97 (dd, 1H, J=
2.4, 10.7 Hz, H3), 4.52 (d, 1H, J=12.0 Hz, CH2Ph), 4.72 (d, 1H, J=
12.0 Hz, CH2Ph), 4.82 (d, 1H, J=2.4 Hz, H1), 7.26–7.46 ppm (m, 5H,
CHAr); 13C NMR (100 MHz, CD3OD): d=32.7 (C4), 60.4 (C5), 68.2 (C3),
69.8 (C2), 70.4 (CH2Ph), 101.0 (C1), 119.9 (CN), 128.9–129.5 (CHAr),
138.6 ppm (CAr). HRMS (ESI): m/z calcd for C13H15NNaO4: 272.089329
[M+Na]+, found: 272.089464.
4-Epi-isofagomine (15): To a solution of compound 22 (80 mg,
0.321 mmol) in a mixture of 1,4-dioxane/H2O 4:1 (5 mL), concen-
trated HCl was added (27 mL, 0.324 mmol, 1 equiv). Pd(OH)2/C
(20%, 70 mg) was then added, and the mixture was stirred for 21 h
at room temperature under an H2 atmosphere. The catalyst was re-
moved by filtration over a membrane (0.2 mm), and the solvent
was removed under vacuum to give the intermediate 4-amino-
methyl-4-deoxy-d-lyxopyranose, which was dissolved in a mixture
of 1,4-dioxane/H2O 1:1 (6 mL). AcOH (0.5 mL) and 10% Pd/C
(50 mg) were added, and the reaction was stirred for 30 h at room
temperature under an H2 atmosphere. The catalyst was removed
by filtration over a membrane (0.2 mm), and the solvent was evapo-
rated under vacuum. The residue was dissolved in MeOH (10 mL),
and this solution was loaded onto a column (Dowex 50W X8 resin,
H+ form, 50–100 mesh). The resin was washed with MeOH (40 mL)
and H2O (20 mL). The desired compound 22 (47 mg, quant.) was
eluted using an aqueous solution of NH4OH (0.5n, 100 mL). HRMS
(ESI): m/z calcd for C6H14NO3: 148.096820 [M+H]+, found:
148.096935. NMR spectra are in accordance with published data.[22]
nmolhÀ1 (mg protein)À1 [33]
.
Acknowledgements
Financial support by a grant from the European Leukodystrophy
Association (ELA; grant number 2011-028I1) is gratefully acknowl-
edged. We also thank LABEX SynOrg (ANR-11-LABX-0029) for fi-
nancial support.
Keywords: galactocerebrosidase · iminosugars · inhibitors ·
Krabbe disease · pharmacological chaperones
[1] a) K. Suzuki, Y. Suzuki, Proc. Natl. Acad. Sci. USA 1970, 66, 302–309;
b) D. A. Wenger, M. A. Rafi, P. Luzi, J. Datto, E. Costantino-Ceccarini, Mol.
Genet. Metab. 2000, 70, 1–9; c) “Krabbe Disease (Globoid Cell Leukodys-
trophy)”: D. A. Wenger in The Molecular and Genetic Basis of Neurologi-
cal Disease, 2nd ed. (Eds.: R. N. Rosenberg, S. Prusiner, S. DiMauro, R. L.
Barchi), Butterworth-Heinemann, Boston, 1997, pp. 421–431.
[2] a) H. Igisu, K. Suzuki, Science 1984, 224, 753–755; b) K. Suzuki, Neuro-
chem. Res. 1998, 23, 251–259.
[3] P. K. Duffner, V. S. Caviness, R. W. Erbe, M. C. Patterson, K. R. Schultz,
D. A. Wenger, C. Whitley, Genet. Med. 2009, 11, 450–454.
[4] W. C. Lee, Y. K. Tsoi, F. J. Troendle, M. W. DeLucia, Z. Ahmed, C. A. Dicky,
D. W. Dickson, C. B. Eckman, FASEB J. 2007, 21, 2520–2527.
[5] M. A. Rafi, H. Z. Rao, P. Luzi, M. T. Curtis, D. A. Wenger, Mol. Ther. 2012,
20, 2031–2042.
[6] R. E. Boyd, G. Lee, P. Rybczynski, E. R. Benjamin, R. Khanna, B. A. Wust-
man, K. J. Valenzano, J. Med. Chem. 2013, 56, 2705–2725.
[7] J.-Q. Fan, Trends Pharmacol. Sci. 2003, 24, 355–360.
[8] Iminosugars: from Synthesis to Therapeutic Applications (Eds.: P. Com-
pain, O. R. Martin), Wiley-VCH, Weinheim, 2007.
[9] J.-Q. Fan in Iminosugars: from Synthesis to Therapeutic Applications
(Eds.: P. Compain, O. R. Martin), Wiley-VCH, Weinheim, 2007, pp. 225–
247.
Biological assays
Human b-galactosidase and a-galactosidase A: These enzymes were
assayed using 4-methylumbelliferyl(4MU)-b-d-galactopyranoside
and 4MU-a-d-galactopyranoside as substrates and a total leuko-
cyte sonicate as the enzyme source.[31] The substrates were pur-
chased from Sigma Chemical Co. (St. Louis, MO, USA). A 0.5 mm
stock solution of 4MU-b-galactopyranoside for assay was made in
0.2m acetate buffer (pH 4.0 containing 0.2m NaCl). A 10 mm stock
solution of 4MU-a-galactopyranoside was made in citrate-phos-
phate buffer (0.2m in phosphate, pH 4.4). For the assay, 0.1 mL of
substrate stock solution was added to 10–25 mg of protein from
the sonicated control leukocytes in distilled water. The total incu-
bation volume was 0.2 mL. Any inhibitors to be tested were includ-
ed in the 0.2 mL total volume. After incubation for 20 min, 3.8 mL
of glycine-carbonate buffer (0.085m, pH 10) was added, and the
fluorescence was read on a PerkinElmer spectrofluorimeter. Calcu-
lations were made based on a standard curve made with the agly-
cone, 4-methylumbelliferone, and the protein concentration of the
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 2647 – 2652 2651

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[10] a) S. Ishii, Proc. Jpn. Acad. Ser. B 2012, 88, 18–30; b) R. Giugliani, S.
Waldek, D. P. Germain, K. Nicholls, D. G. Bichet, J. K. Simosky, A. C.
Bragat, J. P. Castelli, E. R. Benjamin, P. F. Boudes, Mol. Genet. Metab.
2013, 109, 86–92.
[11] W. C. Lee, D. Kang, E. Causevic, A. R. Herdt, E. A. Eckman, C. B. Eckman,
J. Neurosci. 2010, 30, 5489–5497.
[21] A. Bꢂlow, I. W. Plesner, M. Bols, J. Am. Chem. Soc. 2000, 122, 8567–8568.
[22] Y. Ichikawa, Y. Igarashi, M. Ichikawa, Y. Suhara, J. Am. Chem. Soc. 1998,
120, 3007–3018.
[23] M. Ichikawa, Y. Ichikawa, Bioorg. Med. Chem. 1995, 3, 161–165.
[24] M. I. Simone, R. G. Soengas, S. F. Jenkinson, E. L. Evinson, R. J. Nash,
G. W. J. Fleet, Tetrahedron: Asymmetry 2012, 23, 401–408.
[25] a) Y. S. Reddy, P. V. Kancharla, R. Roy, Y. D. Vankar, Org. Biomol. Chem.
2012, 10, 2760–2773; b) A. Kato, S. Miyauchi, N. Kato, R. J. Nash, Y.
Yoshimura, I. Nagakome, S. Hirono, H. Takahata, I. Adachi, Bioorg. Med.
Chem. 2011, 19, 3558–3568; c) P. Spanu, C. de Candia, F. Ulgheri, Tetra-
hedron Lett. 2010, 51, 2400–2402; d) Y. Mihara, H. Ojima, T. Imahori, Y.
Yoshimura, H. Ouchi, H. Takahata, Heterocycles 2007, 72, 633–645; e) X.
Liang, A. Lohse, M. Bols, J. Org. Chem. 2000, 65, 7432–7437.
[26] G. E. Keck, D. F. Kachensky, E. J. Enholm, J. Org. Chem. 1985, 50, 4317–
4325.
[27] N. Mohal, B. Bernet, A. Vasella, Helv. Chim. Acta 2005, 88, 3232–3247.
[28] S. Front, E. Gallienne, O. R. Martin, A. Kato, unpublished results.
[29] J. E. Deane, S. C. Graham, N. N. Kim, P. E. Stein, R. McNair, M. B. Cachon-
Gonzales, T. M. Cox, R. J. Read, Proc. Natl. Acad. Sci. USA 2011, 108,
15169–15173.
[12] A. S. Berardi, G. Pannuzzo, A. Graziano, E. Costantino-Ceccarini, P. Piom-
boni, A. Luddi, Mol. Genet. Metab. 2014, 112, 294–301.
[13] J. Ribbens, G. Whiteley, H. Furuya, N. Southall, X. Hu, J. Marugan, M.
Ferrer, G. H. B. Maegawa, Anal. Biochem. 2013, 434, 15–25.
[14] a) B. A. Rigat, M. B. Tropak, J. Buttner, E. Crushell, D. Benedict, J. W. Calla-
han, D. R. Martin, D. J. Mahuran, Mol. Genet. Metab. 2012, 107, 203–212;
b) K. Fantur, D. Hofer, G. Schitter, A. J. steiner, B. M. Pabst, T. M. Wrod-
nigg, A. E. Stuetz, E. Paschke, Mol. Genet. Metab. 2010, 100, 262–268;
c) M. Aguilar-Moncayo, T. Takai, K. Higaki, T. Mena-Barragan, Y. Hirano, K.
Yura, L. Li, Y. Yu, H. Ninomiya, M. I. Garcia-Moreno, S. Ishii, Y. Sakakibara,
K. Ohno, E. Nanba, C. O. Mellet, J. M. G. Fernandez, Y. Suzuki, Chem.
Commun. 2012, 48, 6514–6516.
[15] J. Matsuda, O. Suzuki, A. Oshima, Y. Yamamoto, A. Noguchi, K. Takimoto,
M. Itoh, Y. Matsuzaki, Y. Yasuda, S. Ogawa, Y. Sakata, E. Nanba, K. Higaki,
Y. Ogawa, L. Tominaga, K. Ohno, H. Iwasaki, H. Watanabe, R. O. Brady, Y.
Suzuki, Proc. Natl. Acad. Sci. USA 2003, 100, 15912–15917.
[30] U. Ohto, K. Usui, T. Ochi, K. Yuki, Y. Satow, T. Shimizu, J. Biol. Chem.
2012, 287, 1801–1312.
[16] a) P. Compain, O. R. Martin, C. Boucheron, G. Godin, L. Yu, K. Ikeda, N.
Asano, ChemBioChem 2006, 7, 1356–1359; b) W. Schçnemann, E. Galli-
enne, K. Ikeda-Obatake, N. Asano, S. Nakagawa, A. Kato, I. Adachi, M.
Gꢄrecki, J. Frelek, O. R. Martin, ChemMedChem 2013, 8, 1805–1817.
[17] a) F. Oula¨ıdi, E. Gallienne, P. Compain, O. R. Martin, Tetrahedron: Asym-
metry 2011, 22, 609–612; b) A. Biela, F. Oula¨ıdi, E. Gallienne, M. Gorecki,
J. Frelek, O. R. Martin, Tetrahedron 2013, 69, 3348–3354.
[31] “Screening for Lysosomal Disorders”: D. A. Wenger, C. Williams in Tech-
niques in Diagnostic Human Biochemical Genetics: A Laboratory Manual
(Ed.: F. A. Hommes), Wiley-Liss, New York, 1991, pp. 587–617.
[32] O. H. Lowry, N. J. Rosebrough, A. L. Farr, R. J. Randall, J. Biol. Chem.
1951, 193, 265–275.
[33] D. A. Wenger, M. Sattler, C. Clark, H. McKelvey, Clin. Chim. Acta 1974, 56,
199–206.
[18] R. C. Bernotas, G. Papandreou, J. Urbach, B. Ganem, Tetrahedron Lett.
1990, 31, 3393–3396.
[19] A. Biela-Banas, E. Gallienne, S. Front, O. R. Martin, Tetrahedron Lett.
2014, 55, 838–841.
Received: September 23, 2014
[20] T. D. Heightman, A. Vasella, Angew. Chem. Int. Ed. 1999, 38, 750–770;
Angew. Chem. 1999, 111, 794–815.
Published online on November 5, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 2647 – 2652 2652