Glucocerebrosidase enhancers for selected gaucher disease genotypes by modification of α-1-C-substituted Imino- D -xylitols (DIXs) by click chemistry

Article abstract of DOI:10.1002/cmdc.201402023

A series of hybrid analogues was designed by combination of the iminoxylitol scaffold of parent 1C9-DIX with triazolylalkyl side chains. The resulting compounds were considered potential pharmacological chaperones in Gaucher disease. The DIX analogues reported here were synthesized by CuAAC click chemistry from scaffold 1 (α-1-C-propargyl-1,5-dideoxy-1,5-imino-D- xylitol) and screened as imiglucerase inhibitors. A set of selected compounds were tested as β-glucocerebrosidase (GBA1) enhancers in fibroblasts from Gaucher patients bearing different genotypes. A number of these DIX compounds were revealed as potent GBA1 enhancers in genotypes containing the G202R mutation, particularly compound DIX-28 (α-1-C-[(1-(3-trimethylsilyl) propyl)-1H-1,2,3-triazol-4-yl)methyl]-1,5-dideoxy-1,5-imino-D-xylitol), bearing the 3-trimethylsilylpropyl group as a new surrogate of a long alkyl chain, with approximately threefold activity enhancement at 10 nM. Despite their structural similarities with isofagomine and with our previously reported aminocyclitols, the present DIX compounds behaved as non-competitive inhibitors, with the exception of the mixed-type inhibitor DIX-28.

Full text of DOI:10.1002/cmdc.201402023

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201402023
Glucocerebrosidase Enhancers for Selected Gaucher
Disease Genotypes by Modification of a-1-C-Substituted
Imino-d-xylitols (DIXs) by Click Chemistry
Jenny Serra-Vinardell,[a] Lucꢀa Dꢀaz,[b] Josefina Casas,[b] Daniel Grinberg,[a] Llu¨ısa Vilageliu,[a]
Helen Michelakakis,[c] Irene Mavridou,[c] Johannes M. F. G. Aerts,[d] Camille Decroocq,[e]
Philippe Compain,*[e, f] and Antonio Delgado*[b, g]
A series of hybrid analogues was designed by combination of
the iminoxylitol scaffold of parent 1C9-DIX with triazolylalkyl
side chains. The resulting compounds were considered poten-
tial pharmacological chaperones in Gaucher disease. The DIX
analogues reported here were synthesized by CuAAC click
chemistry from scaffold 1 (a-1-C-propargyl-1,5-dideoxy-1,5-
imino-d-xylitol) and screened as imiglucerase inhibitors. A set
of selected compounds were tested as b-glucocerebrosidase
(GBA1) enhancers in fibroblasts from Gaucher patients bearing
different genotypes. A number of these DIX compounds were
revealed as potent GBA1 enhancers in genotypes containing
the G202R mutation, particularly compound DIX-28 (a-1-C-[(1-
(3-trimethylsilyl)propyl)-1H-1,2,3-triazol-4-yl)methyl]-1,5-di-
deoxy-1,5-imino-d-xylitol), bearing the 3-trimethylsilylpropyl
group as a new surrogate of a long alkyl chain, with approxi-
mately threefold activity enhancement at 10 nm. Despite their
structural similarities with isofagomine and with our previously
reported aminocyclitols, the present DIX compounds behaved
as non-competitive inhibitors, with the exception of the
mixed-type inhibitor DIX-28.
Introduction
Lysosomal storage disorders constitute a group of genetic dis-
eases characterized by the accumulation of non-degraded sub-
strates within the lysosomes.[1] These types of disorders are
usually classified according to the nature of the substrate accu-
mulated. In particular, sphingolipidoses are the result of the
abnormal lysosomal metabolism of glycosphingolipids. Among
the different sphingolipidoses, Gaucher disease (GD) is particu-
larly relevant, due to its prevalence. It is characterized by the
accumulation of glucosylceramide (GlcCer), as a result of the
deficiency of b-glucocerebrosidase (GCase or GBA1), a lysoso-
mal b-glucosidase that hydrolyses GlcCer into glucose and ce-
ramide.[2] The disease has classically been divided into three
types, based on neurological involvement: type 1 (non-neuro-
nopathic), type 2 (acute neuronopathic), and type 3 (subacute
neuronopathic).
Despite current therapeutic approaches for the treatment of
GD, such as enzyme replacement and substrate reduction
therapies, alternative strategies are desirable in order to avoid
the drawbacks associated with the above classical ap-
proaches.[3,4] The deficiency of GBA1, in most instances, results
from the presence of one or several enzyme mutations that
give rise to misfolded forms of the enzyme that are premature-
ly removed in the endoplasmic reticulum (ER) by the ER-associ-
ated degradation (ERAD) system before reaching the lyso-
some.[5] In this context, pharmacological chaperones have
become an active field of research.[6]
[a] J. Serra-Vinardell,+ Prof. Dr. D. Grinberg, Prof. Dr. L. Vilageliu
Departament de Genꢀtica, Universitat de Barcelona (UB)
IBUB; CIBER de Enfermedades Raras (CIBERER)
[e] Dr. C. Decroocq, Prof. Dr. P. Compain
Laboratoire de Synthꢀse Organique et Molꢂcules Bioactives
Universitꢂ de Strasbourg/CNRS (UMR 7509)
Ecole Europꢂenne de Chimie, Polymꢀres et Matꢂriaux
25 rue Becquerel, 67087 Strasbourg (France)
E-mail: philippe.compain@unistra.fr
Av. Diagonal 643, 08028, Barcelona (Spain)
[b] Dr. L. Dꢁaz,+ Dr. J. Casas, Prof. Dr. A. Delgado
Research Unit on BioActive Molecules (RUBAM)
Departament de Quꢁmica Biomꢀdica
[f] Prof. Dr. P. Compain
Institut de Quꢁmica AvanÅada de Catalunya (IQAC-CSIC)
Jordi Girona 18, 08034 Barcelona (Spain)
Institut Universitaire de France
103 Bd. Saint-Michel, 75005 Paris (France)
[c] Prof. Dr. H. Michelakakis, Dr. I. Mavridou
Department of Enzymology & Cellular Function
Institute of Child Health, Athens 11527 (Greece)
[g] Prof. Dr. A. Delgado
Facultat de Farmꢃcia, Unitat de Quꢁmica Farmacꢀutica (Unitat Associada
al CSIC), Universitat de Barcelona (UB)
Avda. Joan XXIII, s/n, 08028 Barcelona (Spain)
E-mail: delgado@rubam.net
[d] J. M. F. G. Aerts
Department of Medicinal Biochemistry, Academic Medical Center
Meibergdreef 15, 1105 AZ, Amsterdam (The Netherlands)
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201402023.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1744

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Application of this concept to the development of alterna-
tive strategies for GD has focused on the discovery of small
molecules able to target the mutated enzyme, thus preventing
its premature degradation, giving rise to an enzyme activity
enhancement. Although classical approaches for the develop-
ment of pharmacological chaperones have relied on the use of
ligand-related reversible enzyme inhibitors, other approaches
based on allosteric ligands have recently been disclosed.[7,8]
Since Sawkar et al.[9] demonstrated that N-nonyldeoxynojiri-
micin (NN-DNJ) produced a 1.5- to 2-fold increase in the activi-
ty of the N370S mutation, several molecules have been evalu-
ated, primarily for mutations N370S and L444P,[10,11] the two
most frequent GD mutant alleles. In particular, isofagomine
(IFG), which was shown to produce one of the highest increas-
es in the N370S mutant GBA1,[12,13] entered clinical trials. How-
ever, after the results of Phase I and II, the trials were discontin-
ued.[10,14]
Figure 1. Conceptual design of DIX-x compounds by molecular hybridization
of the iminoxylitol scaffold present in 1C9-DIX with the N-substituted triazol-
yl side chain of AC-x compounds.
potent GBA1 inhibitors[23,24] and also exhibited interesting in
vitro and in cellular enzyme enhancement activities toward
several GBA1 mutations (unpublished results).
In recent years, our laboratories have been actively working
on the discovery of new small GBA1 inhibitors as potential
pharmacological chaperones of diverse mutant forms of this
enzyme. Thus, a series of potent iminocyclitol derivatives[15,16]
culminated in the second generation iminoxylitol derivative
1C9-DIX (Figure 1), whose efficiency as pharmacological chap-
erone at low nanomolar concentration in N370S fibroblasts
from GD patients was reported.[17] Interestingly, this compound
was used by Overkleeft as a chemical tool to understand GC
metabolism and the basis of GD.[18,19]
Similarly, with a series of aliphatic N-alkyl aminocyclitols,[20,21]
interesting GBA1 enhancement in patient fibroblasts with dif-
ferent enzyme mutations were obtained.[22] Subsequent modifi-
cations at the nitrogen side chain using Cu-promoted alkyne–
azide cycloaddition (CuAAC) between a parent aminocyclitol
and a set of azides, carefully chosen to ensure a high degree
of diversity in the resulting library, led to aminocyclitols of the
general structure AC-x (see Figure 1). These were reported as
Results and Discussion
Collection design
The remarkable effects elicited by the N-substituted triazoly-
lalkyl side chain in aminocyclitols AC-x prompted us to use this
structural motif to explore the chemical diversity around the
iminoxylitol scaffold present in compound 1C9-DIX (Figure 1).
In conceptually related approaches, the use of click chemistry
to explore the chemical diversity in N-substituted 1-deoxynojir-
imycin[25] and the thiol-ene click reaction in a-1-C-substituted
imino-d-xylitols[26] have also been reported. In this way, a small
library of iminoxylitols, DIX-x (Figure 1), arising from the substi-
tution of the parent 1,5-dideoxy-1,5-imino-d-xylitol scaffold
(DIX) with some of the most interesting triazolylalkyl side
chains used in compounds AC-x,[23] was assembled by means
of CuAAC chemistry from iminoxylitol 1 and the azides shown
in Figure 2 and Scheme 1. The required azides were selected
Figure 2. Azides used in the click chemistry reaction of iminoxylitol 1 under conventional CuAAC conditions. Azide numbering has been maintained for com-
parison with data reported in ref. [23].
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1745

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Scheme 1. General approach to DIX-x compounds. The azide partner is de-
noted by [x] (for azide structures, see Figure 2).
from a collection of commercial precursors (mainly alcohols
and bromides), which were filtered as previously described.[23]
In addition, a new silyl-containing side chain was found to
confer interesting properties to the resulting DIX adduct (see
below).
Scheme 2. Reagents and conditions: a) for details, see Ref. [27]; b) OsO4,
NaIO4, THF/H2O (1:1), RT, 3 h; c) (MeO)2P(O)C(N2)C(O)Me, K2CO3, MeOH, RT,
5 h, 68% over two steps; d) BCl3, pentamethylbenzene, CH2Cl2, ꢀ788C, 4 h.
Synthesis of iminoxylitol scaffold 1
Our approach was based on the protected advanced inter-
mediate 3,[27] recently synthesized in one of our groups in five
steps and 44% overall yield from 2,3,4-tri-O-benzyl-d-xylopyra-
nose (2)[28] (Scheme 2). Conversion of alkene 3 into the corre-
sponding alkyne 4 was performed in two steps (68% com-
bined yield) by means of Lemieux-Johnson oxidative cleavage
and reaction of the intermediate aldehyde with an excess of
the Bestmann–Ohira reagent.[29] At this stage, the chemoselec-
tive deprotection of the O-benzyl groups in 4 proved difficult,
and our first attempts using boron trichloride[30] or TMSI[31,32]
led to a complex mixture of products. In this context, the de-
protection of O-benzyl groups using boron trichloride in the
presence of an alkyne functionality were described as trouble-
some,[23] due to the reactivity of the CꢀC triple bond. Fortu-
nately, a combination of boron trichloride and pentamethyl-
benzene was found to be optimal, leading to the
out on a small scale (~8 mg of scaffold and a 1.2 molar ratio of
the required azide; Figure 2) in 1 mL of an H2O/THF (1:1) mix-
ture for 24 h at room temperature to ensure total consumption
of the starting scaffold. The identity of the resulting DIX ad-
ducts was confirmed by UPLC-MS analysis of the crude reac-
tion mixtures. As the components of the click reaction (copper
salts and ascorbic acid) do not interfere with the enzymatic
assay,[23] the crude mixtures were individually screened as
GBA1 inhibitors in recombinant imiglucerase to determine the
IC50 value at the optimal pH for enzyme activity (5.2) and at
the neutral pH of the ER (7.0) (Table 1).
On the basis of the results obtained from in situ screening
of the crude CuAAC reaction mixtures (see below), the five
most promising iminoxylitol derivatives were selected for fur-
ther biological evaluation. Compounds DIX-1, DIX-4, DIX-17,
concomitant deprotection of the O-benzyl and N-tert-
butoxycarbonyl groups. The interest in pentamethyl-
Table 1. IC50 values for the library of iminoxylitols DIX toward imiglucerase at pH 5.2
benzene as a cation scavenger for related boron tri-
chloride-mediated debenzylation of aryl benzyl
ethers was previously reported.[33] The main advant-
age of this scavenger is that it does not decrease the
Lewis acidity of boron trichloride. This one-pot pro-
cess afforded the desired iminoxylitol 1 in 67% yield
after chromatographic purification. It is noteworthy
that the corresponding monobenzylated analogue 5,
resulting from a partial debenzylation reaction, was
and pH 7.0.[a]
Iminoxylitol
IC50 [nm]
Iminoxylitol[b]
IC50 [nm]
pH 5.2
pH 5.2
pH 7.0
pH 7.0
DIX-1
DIX-3
DIX-4
DIX-5
DIX-6
DIX-7
DIX-8
DIX-9
DIX-10
DIX-11
DIX-12
DIX-13
7.9ꢁ0.2
6.1ꢁ0.31
DIX-14
DIX-15
DIX-17
DIX-18
DIX-21
DIX-24
DIX-25
DIX-26
DIX-27
DIX-28
DIX-29
1C9-DIX
12.5ꢁ0.5
6.8ꢁ0.3
166.0ꢁ8.3 116.0ꢁ5.7
355.0ꢁ16.5 314.8ꢁ15.1
8.7ꢁ0.3
17.2ꢁ0.6
160.0ꢁ7.5
11.6ꢁ0.5
8.1ꢁ0.3
4.6ꢁ0.2
7.5ꢁ0.4
57.5ꢁ3.1
5.9ꢁ0.3
5.1ꢁ0.2
10.4ꢁ0.4
7.6ꢁ0.5
7.6ꢁ0.4
5.5ꢁ0.2
10.0ꢁ0.4
50.5ꢁ2.3
44.5ꢁ2.1
12.1ꢁ0.5
5.4ꢁ0.3
7.2ꢁ0.3
6.4ꢁ0.2
27.3ꢁ1.2
3.9ꢁ0.2
112.0ꢁ5.1
83.0ꢁ3.9
18.9ꢁ0.7
5.5ꢁ0.3
1
also isolated in 7% yield. H and 13C-HMBC NMR ex-
29.4ꢁ1.2
periments showed that the benzyl group was at-
tached to the C2 position of the iminoxylitol system.
223.0ꢁ10.5 115.9ꢁ4.7
7.7ꢁ0.4
33.0ꢁ1.4
223.0ꢁ10.8 136.4ꢁ6.5
8.2ꢁ0.5 5.8ꢁ0.3
20.9ꢁ1.1
7.8ꢁ0.3
94.0ꢁ4.5
6.8ꢁ0.3[b]
Library synthesis and preliminary screening
[a] All compounds, except 1C9-DIX, were tested as crude mixtures from the CuAAC re-
action between 1 and the appropriate azide (Figure 2). Inhibitors were tested at five
different concentrations from a 47 mm click reaction mixture, assuming a quantitative
conversion of the starting iminoxylitol (for details, see Experimental Section). [b] Data
taken from Ref. [16].
The iminoxylitols used in this study were obtained by
the CuAAC of scaffold 1 with the azides shown in
Figure 2, following our previously reported protocol
(see Experimental Section).[23] Reactions were carried
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1746

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
DIX-27, and DIX-28 were thus resynthesized and fully charac-
terized (Scheme 3). With the aim of facilitating purification of
the iminosugars obtained by CuAAC, iminoxylitols were pre-
pared in two steps from the corresponding azides and protect-
ed alkyne 6, which was obtained, in turn, by treatment of 4
with trifluoroacetic acid (TFA). This strategy, which was first
As with the aminocyclitol analogues,[23,24] compounds with
an aliphatic chain (DIX-17, DIX-18, and DIX-26) were among
the most potent members of the series with IC50 values in the
low nanomolar range, similar to that of 1C9-DIX. In general,
the affinity of alkylated iminosugar[15–17] or aminocyclitol[24] de-
rivatives toward GBA1 increases with the length of the alkyl
chain. However, long alkyl chain derivatives may also
be cytotoxic, mainly due to membrane insertion and
pore formation.[34] To shorten the alkyl chain length
while maintaining an optimal level of lipophilicity,
a 3-trimethylsilylpropylgroup was introduced as a sub-
stituent by means of azide [28] (see Figure 2). Inter-
estingly, the resulting iminoxylitol, DIX-28, showed an
inhibitory potency similar to that of the above longer
alkyl chain analogues. In agreement with this obser-
vation, the short chain and polar DIX-15 was a less
potent member of the library, while other short chain
analogues bearing strongly lipophilic aryl substitu-
ents, such as DIX-1, DIX-4, and DIX-27 were among
the most potent inhibitors. The negative effect of
a terminal polar substituent is again evidenced in
DIX-3, DIX-6, DIX-10, and DIX-21, which were less
potent members of the series, albeit with IC50 values
still below the micromolar range.
Biochemical studies with purified compounds
From the above preliminary screening with crude
Scheme 3. Reagents and conditions: a) TFA/CH2Cl2 (1:4), RT, 1 h, 93%; b) CuSO4·5H2O, Na
ascorbate, azide, H2O/THF (1:1), RT, 15 h, 66% (DIX-28); 60% (DIX-17); 78% (DIX-1); 79%
(7); 86% (8); c) BCl3, CH2Cl2, ꢀ608C!RT, 16 h, 80%; d) BCl3, CH2Cl2, ꢀ608C!08C, 4 h,
40% (DIX-27); 59% (DIX-27B).
CuAAC reaction mixtures, iminoxylitols DIX-1, DIX-4,
DIX-17, DIX-27, DIX-27B, and DIX-28 were selected for
further biochemical studies and resynthesized for
complete chemical characterization (Scheme 3). The
IC50 values (see Table 2) were within the range of
those obtained with the crude reaction mixtures. The new DIX
compounds behaved as non-competitive imiglucerase inhibi-
tors, with the exception of DIX-28, which showed mixed-type
inhibition kinetics (Figure 3). In all cases, the Ki values were in
the nanomolar range, with the exception of the weak inhibitor
DIX-27B.
performed to synthesize DIX-4, DIX-27, and DIX-27B gave un-
satisfactory results for DIX-1, DIX-17, and DIX-28, which were fi-
nally obtained in acceptable yields by click reactions from fully
deprotected 1. The boron trichloride-mediated removal of the
O-benzyl groups from 7 and 8 required longer reaction times
and higher temperatures to avoid partial debenzylation reac-
tions, as exemplified by the formation of DIX-27B. In fact, this
iminoxylitol was obtained as the major compound from de-
benzylation of 8 and, due to its structural singularity, was in-
cluded in the GBA1 enhancement assays described below.
Imiglucerase is the recombinant DNA-produced analogue of
human b-glucocerebrosidase used in enzyme replacement
therapy of GD. The results of the imiglucerase inhibitory assay
are shown in Table 1. In general, all compounds behaved as
potent imiglucerase inhibitors, with IC50 values in the nanomo-
lar range. It is worth mentioning the higher inhibitory potency
elicited by iminoxylitol derivatives in comparison with the cor-
responding aminocyclitol counterparts. Thus, for identical side
chain substitution, even in crude reaction mixtures, com-
pounds with the iminoxylitol scaffold (see Figure 1) gave rise
to significantly more potent inhibitors than the corresponding
aminocyclitols counterparts, which were active in the micro-
molar range.[23,24]
Table 2. Inhibition data for selected compounds toward imiglucerase.
Compd
IC50 [nm]
pH 5.2
Ki [nm][a]
pH 7.0
pH 5.2
pH 7.0
DIX-1
DIX-4
4.7ꢁ0.2
10.2ꢁ0.4
5.1ꢁ0.3
4.3ꢁ0.2
7.7ꢁ0.3
4.7ꢁ0.3
22.2ꢁ1.1
71.4ꢁ3.8
3.8ꢁ0.2
3.9ꢁ0.2
1.1ꢁ0.1
3.7ꢁ0.2
4.4ꢁ0.2
16.5ꢁ0.7
1.9ꢁ0.1
3.7ꢁ0.2
3.5ꢁ0.2
10.1ꢁ0.4
DIX-17
DIX-27
DIX-27B
DIX-28
1C9-DIX
26.7ꢁ1.2
83.8ꢁ4.7
6.5ꢁ0.2
>100
>100
2.6ꢁ0.1[b]
1.7ꢁ0.1[c]
6.8ꢁ0.3[e]
2.2ꢁ0.1[d]
2.2ꢁ0.1[d,e]
[a] Noncompetitive inhibitors (unless otherwise noted). [b] Mixed-type in-
hibitor (a=3.6). [c] Mixed-type inhibitor (a=2.3). [d] Data taken from
Ref. [16]. [e] Competitive inhibitor.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1747

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
trations (Figure 5). Considering
that a relatively small amount of
enzyme activity may be enough
to avoid substrate accumula-
tion,[35] the observed increase in
activity could be clinically signifi-
cant. Compound DIX-28 also
showed a remarkable effect on
the N370S/N370S genotype,
close to that of 1C9-DIX at
10 nm (Figure 4) and even great-
er than that of 1C9-DIX at
100 nm (Table S1). This is rele-
vant, considering the high preva-
lence of the N370S mutation,
which is related to the non-neu-
ronopathic GD type I.
In all cases, the selected DIX
compounds showed preferential
GBA1 enhancement towards
genotypes E and F (Figure 4; see
also Table S1 in the Supporting
Information). Compound DIX-
1 was the most potent member
of the series toward the latter,
with a twofold increase in activi-
ty at concentrations as low as
0.05 nm (Figure S2). This trend
was found even for the weak in-
hibitor DIX-27B, which showed
a modest activity enhancement
Figure 3. Kinetics for the inhibition of imiglucerase by DIX-17 and DIX-28. Double reciprocal plot of imiglucerase
incubated at different concentrations of substrate and compounds. A) DIX-17: 0 mm (&); 3 mm (*); 4.5 mm (~);
^
^
6 mm ( ). B) DIX-28: 0 mm (&); 3 mm (*); 6 mm (~); 10 mm ( ). Regression lines arise from data obtained in two
different experiments performed in triplicate.
at
100 nm
concentration
(Table S1 in the Supporting Infor-
mation). As the concentration that gave a maximal activity en-
hancement differed for different compounds, a wide range of
concentrations was analyzed for each product and genotype
(Figure 5; see also Figures S2 and S3 in the Supporting Infor-
mation).
Biological studies with purified compounds
The selected compounds were nontoxic in a wild-type (WT) fi-
broblasts MTT assay at concentrations up to 300 nm after
6 days incubation (Figure S1). The above compounds were
next evaluated for their ability to enhance GBA1 residual activi-
ty in several GD genotypes (Figures 4 and 5, and Table 1; see
also, Figures S2 and S3 in the Supporting Information). In par-
ticular, six different genotypes, together with WT fibroblasts,
were used in this study (see Experimental Section).
1C9-DIX,[17] with a reported enhancement of N370S enzyme
activity (1.6-fold at 10 nm), was used as a reference. In order to
determine whether this compound was active against other
mutant GBAs, we tested it on the fibroblasts used in this study.
The best results were found for those bearing the genotypes
G202R/[L444P; E326K] (genotype E) and G202R/G202R (F) (2–3-
fold increase, at 10 nm). For N370S/N370S (C), an enhancement
similar to that previously reported was observed (Figure 4).
DIX-28 exhibited the best activity enhancement, reaching
a 2.5–3-fold increase for genotypes E and F at 10 nm (Figure 4)
and up to 4–5-fold increase at 100 nm (Table S1 and Figure S2).
This compound behaved similarly to 1C9-DIX at low concentra-
tions (10 nm) but showed a wider enhancement window, as it
was significantly more efficient than 1C9-DIX at higher concen-
It is worth noting that genotypes E and F contain the G202R
mutation, which affects trafficking of the enzyme, precludes its
transport to the lysosome,[36] and is associated with the neuro-
nopathic phenotype of the disease. It has been reported that
both the N370S and G202R mutations are located in the cata-
lytic domain,[37] but the latter is located much farther from the
active site than N370S. We found that, in most cases, com-
pounds that increase the activity of N370S/N370S (genotype C)
also increase the activity of genotypes containing the G202R
mutation, as reported by others.[38] Moreover, the difference is
greater in genotypes containing the G202R mutation than in
the N370S/N370S genotype. Some authors suggest that, while
the N370S mutation affects substrate binding and catalytic ac-
tivity, the G202R mutation destabilizes GBA but does not dis-
rupt the catalytic activity of the folded protein.[38] This could
be the reason for the good results of several pharmacological
chaperones toward the G202R mutation. In this regard, several
compounds have been evaluated against this mutation with
successful results, such as a DNJ analogue described by Sawkar
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1748

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Figure 4. Enhancement of residual GBA1 activity of selected compounds in fibroblasts from GD patients. Fold increase: Increase of the residual GBA1 activity
in GD fibroblasts in comparison with untreated cells at different DIX concentrations (for complete data, see Table S1 in the Supporting Information). Geno-
types: WT: wild-type; A: [D409H;H255Q]/[D409H;H255Q]; B: [D409H;H255Q]/L444P; C: N370S/N370S; D: L444P/L444P; E: G202R/[L444P;E326 K]; F: G202R/
G202R. Values are the median ꢁconfidence interval (CI) of 3–6 separate experiments performed in triplicate.
of GBA and may be the reason why none of our compounds
produced a significant enhancement in activity. In fact, exam-
ples of pharmacological chaperones able to target this muta-
tion are scarce in the literature. Calystegine B2 was the first to
be reported on fibroblasts with the L444P/L444P genotype
(230% activity increase at 10 mm)[11] and, more recently, some
bicyclic iminocyclitols have shown promise as enzyme enhanc-
ers in induced pluripotent stem cell (iPSc) lines with the
L444P/G202R genotype[42] and in monkey kidney fibroblast
COS-7 cells expressing the L444P mutation.[43]
On the other hand, none of the DIX compounds showed ap-
preciable GBA2 inhibition (in mouse testes homogenates) at
concentrations up to 1 mm, which indicates an irrelevant effect
on this enzyme. With regard to other lysosomal enzymes, com-
Figure 5. Activity enhancements (fold increase relative to untreated cells) for
pounds were inactive against the two human a-glycosidases
tested, namely a-glucosidase and a-galactosidase. On the
other hand, human b-galactosidase was slightly inhibited by
compound DIX-17 (25% inhibition) at 10 mm, a concentration
four orders of magnitude higher than required for the ob-
served GBA1 activity enhancements (Table S2 in the Support-
ing Information). Finally, despite the fact that activity of human
b-hexosaminidase was somehow affected by several of the
tested compounds, no good concentration–inhibition correla-
tion was observed. In summary, these compounds exhibited an
excellent selectivity towards GBA1. With regard to other carbo-
hydrate-processing enzymes, 1C9-DIX was recently described
as an inhibitor of the cytosolic b-glucosidase GBA3.[18] However,
as this enzyme does not appear to modify GD manifestations,
no efforts along this line were carried out.
selected DIX compounds at different concentrations toward fibroblasts con-
taining the G202R/[L444P;E326K] genotype.
et al.[38] with a 270% increase in activity, bicyclic nojirimycin an-
alogues with sp2 iminosugar structure[39] and approximately
a 250% increase in activity, and an azepine analogue[40] ]with
a modest 20% increase in activity. The best results were ob-
tained for isofagomine derivatives,[41] with a 7.2-fold increase in
activity. However, it must be mentioned that this result was
obtained at a concentration of 150 mm, which is above the
threshold for clinical use.
Finally, none of the DIX compounds proved efficient against
genotype D (Figure 4; see also Table S1 in the Supporting In-
formation), and only a negligible effect was observed for some
of the compounds against genotypes A and B. In the case of
genotypes containing the L444P mutation, it is important to
note that this mutation is not located at the catalytic domain
Immunofluorescence staining and confocal microscopy
imaging were used to determine whether DIX-28 increased
trafficking of the G202R mutant enzyme to the lysosome
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1749

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
allowed identification of the 3-trimethylsilylpropyl group as
a promising surrogate of long alkyl chains that are known to
induce cytotoxicity.[34] Interestingly, incorporation of this group
into DIX-28 gives rise to a compound with low cytotoxicity ex-
hibiting a remarkable GBA1 enhancement. Regardless of their
structural similarities with IFG and with our previously reported
aminocyclitol analogues,[23] the present DIX compounds be-
haved as non-competitive inhibitors, with the exception of the
mixed-type inhibitor DIX-28. However, even though our com-
pounds were inspired by the common “active site-directed
ligand strategy” that has classically guided the design of phar-
macological chaperones,[6] our results represent an interesting
breakthrough that deserves further attention.
Figure 6. Confocal laser microscope images of immunofluorescence staining
for lysosomes (red) and GBA1 (green) in cultured fibroblasts derived from
a healthy individual (WT) and a GD patient with the G202R/G202R genotype
untreated (middle) and after treatment with DIX-28 (bottom). Nuclei were
stained with DAPI (blue). Immunofluorescent labeling of GBA1 was per-
formed with the 8E4 monoclonal antibody, and LysoTracker was used to
detect lysosomes.
Experimental Section
Chemistry
(Figure 6). Fibroblasts from a Gaucher patient homozygous for
the G202R mutation were incubated for 6 days with or without
100 nm of DIX-28. In treated cells, GBA colocalization of the
mutant enzyme with a lysosomal marker (LysoTracker) clearly
increased, as shown by higher stain intensity compared to un-
treated cells (Figure 6). In fact, the degree of colocalization fol-
lowing treatment was similar to that observed in wild-type fi-
broblasts (Figure 6). These results are in agreement with the
idea that the G202R mutation does not disrupt the catalytic
site. Thus, compound DIX-28 induced the correct folding and
trafficking of the enzyme into the lysosome, as shown in
Figure 6 and, once there, the enzyme could perform its role, as
indicated by the activity results shown here (Figures 4 and 5;
see also Table S1 in the Supporting Information).
General: CH2Cl2 was distilled over CaH2 under argon. MeOH was
distilled over Mg/I2 under argon. Tetrahydrofuran (THF) was dried
by passage through an activated alumina column under argon.
Flash column chromatography was carried out using silica gel 60
(230-400 mesh, 0.040–0.063 mm) purchased from Merck. Automatic
flash chromatography was carried out in a Grace Reveleris flash
system equipped with UV/Vis and ELSD detectors. Thin-layer chro-
matography (TLC) was performed on aluminum sheets coated with
silica gel 60 F254 purchased from Merck. IR spectra (cmꢀ1) were re-
corded on a Perkin–Elmer SpectrumOne spectrophotometer. NMR
spectra were recorded on a Bruker AC 300 or Bruker 1C 400 spec-
trometer, with residual solvent peaks used as reference. Carbon
multiplicities were assigned by distortionless enhancement by po-
1
larization transfer (DEPT) experiments. H and 13C signals were as-
signed by COSY (correlation spectroscopy), HSQC (heteronuclear
single quantum correlation spectroscopy), and HMBC (heteronu-
clear multiple-bond correlation spectroscopy) experiments. Cou-
pling constants (J) are in Hertz (Hz), and multiplicity is described as
(s) singlet, (d) doublet, (t) triplet, (q) quadruplet, and (br) broad.
Electrospray ionization (ESI)-high-resolution mass spectrometry
(HRMS) mass spectrometry was carried out on a Bruker MicroTOF
spectrometer. Specific rotations were determined at room temper-
ature (208C) in a PerkinElmer 241 polarimeter for sodium (l=
589 nm). Purity of compounds used for enzyme assays was ꢂ95%,
as judged by HPLC analysis (Gemini C18 column, 4.6 mmꢂ
250 mm, 5 mm) under the following chromatographic conditions:
mobile phase A, water containing 0.1% v/v trifluoroacetic acid
(TFA); mobile phase B, CH3CN containing 0.1% v/v TFA; flow rate
of 1.0 mLminꢀ1; detection, SATIN-ELS (evaporative light scattering)
(l=254 nm); gradient elution, 0 min, from 80% A/20% B to 0% A/
100% B over 20 min. Each run was followed by a 3 min wash with
80% CH3CN and 20% water.
Conclusions
A new series of a-1-C-substituted DIXs were designed by com-
bination of the iminoxylitol scaffold present in parent 1C9-DIX
with the triazolylalkyl side chains present in a series of amino-
cyclitols previously reported by our groups as GBA1 enhancers
with application in GD. The resulting hybrid structures were
synthesized using standard CuAAC click chemistry from scaf-
fold 1 and were initially screened as imiglucerase inhibitors. In
general, for identical side chain substitution, the DIX scaffold
gives rise to appreciably more potent inhibitors than the corre-
sponding aminocyclitol counterparts. The most potent mem-
bers of the series were resynthesized and tested as GBA1 activ-
ity enhancers in fibroblasts from GD patients bearing different
genotypes. In general, the DIX compounds reported here were
shown to be potent GBA1 enhancers in genotypes containing
the G202R mutation, which is responsible for a neuronopathic
phenotype of the disease. In particular, 1C9-DIX and the silyl
derivative DIX-28 showed approximately threefold activity en-
hancements at 10 nm, and four- to fivefold increases in activity
at 100 nm (Table S1 in the Supporting Information). Moreover,
these two compounds are also among the most potent mem-
bers of the series toward the highly prevalent N370S mutation.
Combining click chemistry and an in situ screening approach
a-1-C-Propargyl-2,3,4-tri-O-benzyl-1,5-dideoxy-1,5-tert-butoxy-
carboxylimino-d-xylitol (4): OsO4 (2.5% w/w in tBuOH,
0.058 mmol, 590 mL, 10% mol) was added to a solution of 3
(313 mg, 0.576 mmol) in THF/water (1:1, 7.4 mL), followed by addi-
tion of NaIO4 (246 mg, 1.151 mmol, 2 equiv). The mixture was
stirred for 3 h at RT, then water (3 mL) was added to the reaction.
The aqueous phase was extracted with CH2Cl2 (3ꢂ10 mL). The ex-
tracts were combined and dried over Na2SO4. The solution was fil-
tered, and the solvent was removed in vacuo. The crude aldehyde
(320 mg) was used in the next step without purification. This resi-
due was dissolved in MeOH (9 mL), and K2CO3 (159 mg,
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1750

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
1.152 mmol, 2 equiv) was added. Then, Bestmann–Ohira reagent
(104 mL, 0.691 mmol, 1.2 equiv) was added, and the solution was
stirred for 5 h at RT. Et2O (20 mL) and a solution of NaHCO3 (5%,
15 mL) were added. The aqueous phase was extracted with Et2O
(3ꢂ15 mL), and the extracts were combined and dried over
Na2SO4. The solution was filtered and concentrated in vacuo. The
residue was purified by automatic flash column chromatography
(EtOAc/petroleum ether, 0:1 to 1:4) to afford a 4 as a colorless oil
2.20 (br s, 1H), 2.40–2.58 (m, 2H), 2.90 (dd, J=13.1 and 6.2 Hz,
1H), 3.05 (dd, J=13.2 and 4.0 Hz, 1H), 3.27 (td, J=7.7 and 3.4 Hz,
1H), 3.45 (br q, J=5.1 Hz, 1H), 3.61–3.72 (m, 2H), 4.57 (d, J=
11.7 Hz, 1H), 4.59 (s, 2H), 4.65 (d, J=11.5 Hz, 1H), 4.66 (s, 2H),
7.22–7.39 ppm (m, 15H).
a-1-C-((1-(3,3-Diphenylpropyl)-1H-1,2,3-triazol-4-yl)-methyl)-
2,3,4-tri-O-benzyl-1,5-dideoxy-1,5-imino-d-xylitol (7): CuSO4·5H2O
(2.0 mg, 0.008 mmol, 0.1 equiv) and sodium ascorbate (3.2 mg,
0.016 mmol, 0.2 equiv), dissolved in water (1 mL), was added to
a solution of 6 (35.5 mg, 0.080 mmol) and 1,1-diphenyl-3-azidopro-
pane (26.7 mg, 0.113 mmol, 1.4 equiv) in THF (1 mL). The mixture
was stirred overnight at RT. The mixture was diluted with EtOAc
(5 mL), and the phases were separated. The organic phase was
washed with a 10% aq solution of NH4OH (5 mL) and then dried
over Na2SO4, filtered and concentrated in vacuo. The residue was
purified by column chromatography (CH2Cl2/MeOH, 1:0 to 95:5) to
afford 7 as a pale yellow oil (43 mg, 79%): Rf =0.44 (CH2Cl2/MeOH,
95:5); [a]D20 =ꢀ5.0 (c=1, CHCl3); 1H NMR (300 MHz, CDCl3): d=
2.12 (br s, 1H), 2.78 (q, J=7.4 Hz, 2H), 2.98-3.20 (m, 4H), 3.48–3.55
(m, 1H), 3.55-3.64 (m, 1H), 3.68 (dd, J=6.3 and 3.8 Hz, 1H), 3.89 (t,
J=6.0 Hz, 1H), 4.02 (t, J=8 Hz, 1H), 4.36 (t, J=7.2 Hz, 2H), 4.68 (d,
J=11.5 Hz, 1H), 4.74 (s, 2H), 4.77–4.87 (m, 3H), 7.28–7.57 ppm (m,
26H); 13C NMR (75 MHz, CDCl3): d=24.7, 36.0, 44.5, 48.3, 48.7, 55.2,
72.1, 72.5, 74.1, 76.8, 77.4, 78.1, 122.0, 126.8, 127.68, 127.70, 127.73,
127.8, 127.86, 127.87, 127.93, 128.1, 128.4, 128.5, 128.9, 138.6,
138.7, 143.3, 145.8 ppm; IR (neat): n˜ =3260 cmꢀ1 (NꢀH); HRMS
(ESI): m/z [M+H+] calcd for C44H47N4O3: 679.364, found: 679.365.
as a mixture of rotamers, according to NMR data (212 mg, 68%
20
over two steps): Rf =0.52 (EtOAc/petroleum ether, 1:3); [a]D
=
1
ꢀ14.0 (c=1, CHCl3); H NMR (300 MHz, CDCl3): d=1.45 (s, 9H), 1.94
(br s, 1H), 2.50 (ddd, J=17.5, 11.0 and 2.4 Hz, 1H), 2.61–2.85 (m,
2H), 3.34–3.70 (m, 3H, H-2), 4.13 (br dd, J=13.5 and 5.3 Hz, 0.5H),
4.42–4.58 (m, 1H), 4.60–4.78 (m, 4H), 4.78–4.98 (m, 2.5H), 7.27–
7.40 ppm (m, 15H); 13C NMR (75.5 MHz, CDCl3): d=16.1, 28.4, 40.0;
41.5, 50.8, 52.4, 70.3, 72.9, 73.2, 73.3, 75.7, 75.8, 77.4, 78.3, 78.4,
79.0, 79.4, 80.5, 80.7, 81.9, 82.0, 127.7, 127.88, 128.97, 128.02, 128.4,
128.6, 138.1, 138.3, 138.9, 154.8 ppm; IR (neat): n˜ =1694 cmꢀ1 (C=
O); HRMS (ESI): m/z [M+Na+] calcd for C34H39NNaO5: 564.272,
found: 564.271.
a-1-C-Propargyl-1,5-dideoxy-1,5-imino-d-xylitol (1) and a-1-C-
propargyl-2-O-benzyl-1,5-dideoxy-1,5-imino-d-xylitol (5): Pen-
tamethylbenzene (1.790 g, 12.07 mmol, 15 equiv) was added to
a solution of 4 (436 mg, 0.805 mmol) in CH2Cl2 (16 mL). Then, BCl3
(7.24 mL, 7.24 mmol, 9 equiv) was added dropwise at ꢀ788C, and
the mixture was stirred for 4 h. MeOH/H2O (1:20, 2 mL) was added
at ꢀ788C, and the mixture was evaporated to dryness. This step
was repeated twice, and the residue was purified by column chro-
matography (CH3CN/H2O/NH4OH, 15:0.5:0.5 to 10:0.5:0.5). Mono-
benzylated compound 5 eluted first and was obtained as a color-
less oil (14.5 mg, 7%): Rf =0.45 (CH3CN/H2O/NH4OH, 10:0.5:0.5);
a-1-C-(1-Benzyl-1H-1,2,3-triazol-4-yl)-methyl-2,3,4-tri-O-benzyl-
1,5-dideoxy-1,5-imino-d-xylitol
(8):
CuSO4·5H2O
(2.3 mg,
0.010 mmol, 0.1 equiv) and sodium ascorbate (3.7 mg, 0.020 mmol,
0.2 equiv), dissolved in water (1 mL), were added to a solution of 6
(42 mg, 0.095 mmol) and benzylazide (16.5 mg, 0.124 mmol,
1.3 equiv) in THF (1 mL). The mixture was stirred overnight at RT.
The mixture was diluted with EtOAc (5 mL), and the phases were
separated. The organic phase was washed with a 10% aq solution
of NH4OH (5 mL) and then dried over Na2SO4, filtered and evapo-
rated. The residue was purified by column chromatography
(CH2Cl2/MeOH, 1:0 to 95:5) to afford 8 as a pale yellow oil (47 mg,
86%): Rf =0.39 (CH2Cl2/MeOH, 95:5); [a]D20 =ꢀ3.0 (c=1, CHCl3);
1H NMR (300 MHz, CDCl3): d=2.07 (br s, 1H), 2.96–3.15 (m, 4H),
3.43–3.59 (m, 2H), 3.62 (dd, J=6.3 and 4.0 Hz, 1H), 3.83 (t, J=
6.2 Hz, 1H), 4.60 (d, J=11.7 Hz, 1H), 4.69 (s, 2H,), 4.71–4.79 (m,
3H), 5.52 (d, J=15.0 Hz, 1H), 5.58 (d, J=14.8 Hz, 1H), 7.28–
7.54 ppm (m, 21H); 13C NMR (75 MHz, CDCl3): d=24.7, 44.4, 54.1,
55.2, 72.1, 72.4, 74.0, 76.8, 77.4, 78.1, 121.7, 127.6, 127.70, 127.73,
127.86, 127.92, 128.0, 128.10, 128.15, 128.39, 128.44, 128.7, 129.1,
135.0, 138.6, 138.7, 146.4 ppm; IR (neat): n˜ =3295 cmꢀ1 (NꢀH);
HRMS (ESI): m/z [M+H+] calcd for C36H39N4O3: 575.302, found:
575.301.
[a]D ꢀ22.0 (c=0.33, MeOH); 1H NMR (400 MHz, MeOD): d=2.36
20
(t, J=2.6 Hz, 1H), 2.42 (ddd, J=16.8 Hz, 7.8 and 2.6 Hz, 1H), 2.48
(ddd, J=16.8 Hz, 7.4 and 2.5 Hz, 1H), 2.78 (dd, J=13.2 and 5.1 Hz,
1H), 3.03 (dd, J=13.1 and 3.3 Hz, 1H), 3.22 (td, J=7.6 and 3.3 Hz,
1H), 3.52 (br q, J=5.0 Hz, 1H), 3.63 (dd, J=5.0 and 3.4 Hz, 1H),
3.85 (t, J=5.2 Hz, 1H), 4.62 (d, J=11.3 Hz, 1H), 4.74 (d, J=11.3 Hz,
1H), 7.25–7.38 (m, 3H), 7.38–7.45 ppm (m, 2H); 13C (100 MHz,
MeOD): d=19.8, 47.1, 54.8, 69.8, 70.7, 71.8, 74.1, 78.8, 81.7, 128.9,
129.39, 129.44, 139.4 ppm; IR (neat): n˜ =3291 cmꢀ1 (OꢀH); HRMS
(ESI): m/z [M+H+] calcd for C15H20NO3: 262.144, found: 262.145.
The second fraction was constituted of compound 1 obtained as
a
white solid (92.5 mg, 67%): Rf =0.29 (CH3CN/H2O/NH4OH,
10:0.5:0.5); [a]D20 =ꢀ9.0 (c=1, MeOH); H NMR (400 MHz, MeOD):
d=2.31 (t, J=2.6 Hz, 1H), 2.36 (ddd, J=16.6 Hz, 7.4 and 2.5 Hz,
1H), 2.46 (ddd, J=16.6 Hz, 7.8 and 2.5 Hz, 1H), 2.80 (dd, J=13.3
and 4.1 Hz, 1H), 3.05 (dd, J=13.3 and 2.7 Hz, 1H), 3.09 (td, J=7.5
and 2.6 Hz, 1H), 3.55 (br qd, J=4.1 and 1.0 Hz, 1H), 3.69-3.73 (m,
1H), 3.76 ppm (t, J=4.5 Hz, 1H); 13C NMR (100 MHz, MeOD): d=
20.5, 47.5, 55.4, 70.4, 71.2, 71.3, 71.3, 82.0 ppm; IR (neat): n˜ =
3282 cmꢀ1 (OꢀH); HRMS (ESI): m/z [M+H+] calcd for C8H14NO3:
172.097, found: 172.096.
1
a-1-C-((1-(3,3-Diphenylpropyl)-1H-1,2,3-triazol-4-yl)methyl)-1,5-
dideoxy-1,5-imino-d-xylitol (DIX-4): BCl3 (1m in CH2Cl2, 0.4 mL,
0.4 mmol, 6 equiv) was added dropwise to a solution of 7 (44 mg,
0.065 mmol) in CH2Cl2 (2.5 mL) at ꢀ608C. The solution was allowed
to warm to RT and was stirred overnight. Then, MeOH/H2O (20:1,
3 mL) was added, and the solution was evaporated to dryness. This
step was repeated, and the residue was purified by column chro-
matography (CH3CN/NH4OH/H20, 15:0.5:0.5) to afford DIX-4 as
a white solid (21 mg, 80%): Rf =0.29 (CH3CN/NH4OH/H2O, 10:1:1);
a-1-C-Propargyl-2,3,4-tri-O-benzyl-1,5-dideoxy-1,5-imino-d-xyli-
tol (6): TFA (1 mL) was added to a solution of 4 (113 mg,
0.21 mmol) in CH2Cl2 (4 mL). After stirring for 1 h at RT, the reaction
mixture was evaporated to dryness. The residue was dissolved in
CH2Cl2 (10 mL) and washed with a 5% aq solution of NaHCO3
(10 mL). The aqueous phase was extracted with CH2Cl2 (3ꢂ10 mL),
and the extracts were combined and dried over Na2SO4. The solu-
tion was filtered and evaporated to afford 6 (86 mg, 93%) of suffi-
1
[a]D20 =ꢀ6.5 (c=1, MeOH); H NMR (300 MHz, MeOD): d=2.69 (q,
J=7.4 Hz, 2H), 3.11 (dd, J=14.8 and 7 Hz, 1H), 3.16–3.30 (m, 2H),
3.44 (dd, J=13.2 and 2.0 Hz, 1H), 3.70–3.82 (m, 2H), 3.88–4.01 (m,
3H), 4.33 (t, J=7.2 Hz, 2H), 7.09–7.23 (m, 3H), 7.23–7.37 (m, 7H),
1
cient purity, as judged by H NMR, to be used directly in the next
1
CuAAC step: H NMR (300 MHz, CDCl3): d=2.01 (t, J=2.7 Hz, 1H),
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1751

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
7.82 ppm (s, 1H); 13C NMR (75 MHz, MeOD): d=25.9, 36.8, 47.5,
49.6, 50.1, 56.3, 67.8, 68.0, 69.6, 125.0, 127.6, 128.8, 129.7, 142.9,
145.0 ppm; IR (neat): n˜ =3330, 3025 cmꢀ1 (OꢀH and NꢀH); HRMS
(ESI): m/z [M+H+] calcd for C23H29N4O3: 409.225, found: 409.223.
60%): Rf =0.22 (CH3CN/NH4OH/H20, 10:1:1); [a]D20 =ꢀ8.0 (c=1,
1
MeOH); H NMR (300 MHz, MeOD): d=0.90 (t, J=6.8 Hz, 2H), 1.17–
1.46 (m, 14H), 1.80–197 (m, 2H), 3.12 (dd, J=14.8 and 6.6 Hz, 1H),
3.17–3.29 (m, 2H), 3.44 (dd, J=13.1 and 2.0 Hz, 1H), 3.71–3.74 (m,
1H), 3.78 (td, J=7.3 and 1.7 Hz, 1H), 3.90–4.01 (m, 2H), 4.39 (t, J=
7.1 Hz, 2H), 7.91 ppm (s, 1H); 13C NMR (75 MHz, MeOD): d=14.4,
23.7, 25.9, 27.5, 30.1, 30.4, 30.5, 30.6, 31.3, 33.0, 47.5, 51.4, 56.3,
67.9, 68.0, 69.6, 124.8, 142.9 ppm; IR (neat): n˜ =3348 cmꢀ1 (OꢀH
and NꢀH); HRMS (ESI): m/z [M+H+] calcd for C18H35N4O3: 355.271,
found: 355.270.
a-1-C-((1-Benzyl-1H-1,2,3-triazol-4-yl)-methyl)-1,5-dideoxy-1,5-
imino-d-xylitol (DIX-27) and a-1-C-((1-benzyl-1H-1,2,3-triazol-4-
yl)-methyl)-2-O-benzyl-1,5-dideoxy-1,5-imino-d-xylitol (DIX-27B):
BCl3 (1m in CH2Cl2, 0.5 mL, 0.50 mmol, 6 equiv) was added drop-
wise to a solution of 8 (47 mg, 0.082 mmol) in CH2Cl2 (0.5 mL) at
ꢀ608C. The mixture was allowed to warm to 08C over 3 h. Then,
MeOH/H2O (20:1, 3 mL) was added and, after 10 min of stirring at
08C, the solvent was evaporated. This step was repeated, and the
residue was purified by column chromatography (CH3CN/NH4OH/
H2O, 15:0.5:0.5 to 10:0.5:0.5). Monobenzylated compound DIX-27B
a-1-C-((1-(3-(Trimethylsilyl)propyl)-1H-1,2,3-triazol-4-yl)methyl)-
1,5-dideoxy-1,5-imino-d-xylitol (DIX-28): Following the general
method, iminoxylitol 1 (17 mg, 0.099 mmol) was combined with 3-
azidopropyl)trimethylsilane[44] (20.3 mg, 0.129 mmol) to afford DIX-
28 as a white solid (21.6 mg, 66%): Rf =0.43 (CH3CN/NH4OH/H2O,
1
was eluted first (19 mg, 59%): [a]D20 =ꢀ2.0 (c=1, MeOH); H NMR
1
(300 MHz, MeOD): d=2.80-3.03 (m, 3H), 3.10 (dd, J=13.6 and
2.2 Hz, 1H), 3.26-3.37 (m, 1H), 3.55-3.60 (m, 1H), 3.64 (br t, J=
3.4 Hz, 1H), 3.74–3.79 (m, 1H), 4.52 (d, J=11.9 Hz, 1H), 4.57 (d, J=
12.2 Hz, 1H), 5.56 (s, 2H), 7.18–7.41 ppm (m, 10H), 7.76 (s, 1H);
13C NMR (75 MHz, MeOD): d=27.6, 48.3, 54.9, 56.2, 67.7, 68.6, 73.3,
77.2, 124.4, 128.8, 128.9, 129.1, 129.4, 129.5, 130.0, 136.8, 139.5,
145.8 ppm; IR (neat): n˜ =3276, 3032, 2918 cmꢀ1 (OꢀH, NꢀH); HRMS
(ESI): m/z [M+H+] calcd for C22H27N4O3: 395.208, found: 395.207.
The second fraction was constituted of DIX-27, obtained as a pale
yellow oil (10 mg, 40%): [a]D20 =ꢀ4.0 (c=0.55, MeOH); 1H NMR
(300 MHz, MeOD): d=3.11 (dd, J=15.1 and 6.9 Hz, 1H), 3.17–3.28
(m, 2H), 3.43 (dd, J=13.1 and 1.8 Hz, 1H), 3.70–3.74 (m, 1H), 3.78
(br td, J=8.4 and 1.8 Hz, 1H), 3.90–3.99 (m, 2H), 5.59 (s, 2H), 7.28–
7.43 (m, 5H), 7.92 ppm (s, 1H); 13C (75 MHz, MeOD): d=25.9, 47.5,
55.0, 56.3, 67.8, 68.0, 69.6, 124.9, 129.3, 129.6, 130.0, 136.7,
143.3 ppm (C-7); IR (neat): n˜ =3240, 3124, 3001 cmꢀ1 (OꢀH, NꢀH);
HRMS (ESI): m/z [M+H+] calcd for C15H21N4O3: 305.161, found:
305.160.
10:1:1); [a]D20 =ꢀ7.5 (c=1, MeOH); H NMR (300 MHz, MeOD): d=
0.0 (s, 9H), 0.44–0.56 (m, 2H), 1.83–1.98 (m, 2H), 3.12 (dd, J=14.9
and 6.7 Hz, 1H), 3.18–3.39 (m, 2H), 3.45 (dd, J=13.1 and 1.9 Hz,
1H), 3.71–3.76 (m, 1H), 3.79 (td, J=7.5 and 1.5 Hz, 1H), 3.90–4.01
(m, 2H), 4.37 (t, J=7.1 Hz, 2H), 7.92 ppm (s, 1H); 13C NMR (75 MHz,
MeOD): d=ꢀ1.9, 14.3, 25.9, 26.3, 47.5, 54.4, 56.4, 67.8, 68.0, 69.6,
124.9, 142.8 ppm; IR (neat): n˜ =3336, 3022 (OꢀH and NꢀH) cmꢀ1
;
HRMS (ESI): m/z [M+H+] calcd for C14H29N4O3Si: 329.201, found:
329.200.
General method for parallel click chemistry and direct screening: A
solution of iminoxylitol 1 (8 mg, 0.047 mmol) and the correspond-
ing azides [1]–[29] (0.055 mmol) in a mixture of H2O/THF (1:1,
1 mL) in a 5 mL screw cap vial was treated with a catalytic amount
of CuSO4·5H2O (around 250 mg), followed by sodium ascorbate
(5 mmol, around 1 mg). After stirring for 24 h at RT, an aliquot was
analyzed (UPLC-HRMS) to confirm click adduct formation. The
crude reaction mixtures containing compounds DIX were used di-
rectly as mother solutions (47 mm) for in vitro IC50 calculation of
imiglucerase inhibition (Table 1).
In vitro activity was determined as previously reported.[24] IC50
values were determined by plotting percent activity versus log[I],
using at least five different inhibitor concentrations. Type of inhibi-
tion and Ki values for the most active inhibitors were determined
by Lineweaver–Burk and Dixon plots of assays performed with dif-
ferent concentrations of inhibitor and substrate. IC50 data from
crude click chemistry mixtures are reported in Table 1. Data from
individually synthesized compounds are reported in Table 2.
General method for CuAAC from scaffold 1
Iminoxylitol 1 and the corresponding azide (1.3 equiv) were solubi-
lized in THF (1 mL). Next, CuSO4·5H2O (2.5 mg, 0.010 mmol,
0.1 equiv) and sodium ascorbate (0.2 equiv) in water (1 mL) were
successively added. The mixture was stirred overnight at RT, then
the solvents were evaporated to dryness. The crude product was
filtered through a plug of Celite and purified by flash chromatogra-
phy on silica gel (CH3CN/NH4OH/H2O, 15:0.5:0.5).
a-1-C-((1-(3,5-bis(Benzyloxy)benzyl)-1H-1,2,3-triazol-4-yl)methyl)-
1,5-dideoxy-1,5-imino-d-xylitol (DIX-1): Following the general
method, iminoxylitol 1 (16 mg, 0.093 mmol) was combined with 1-
azidomethyl-3,5-bis(benzyloxy)benzene[23] (42 mg, 0.122 mmol) to
afford DIX-1 as a white solid (37.5 mg, 78%): Rf =0.12 (CH3CN/
NH4OH/H2O, 10:0.5:0.5): [a]D20 =ꢀ3.0 (c=0.17, MeOH); 1H NMR
(300 MHz, MeOD): d=2.99 (dd, J=14.7 and 7.2 Hz, 1H), 3.03–3.15
(m, 2H), 3.22–3.30 (m, 1H), 3.57 (td, J=7.4 and 2.2 Hz, 1H), 3.61–
3.65 (m, 1H), 3.73–4.81 (m, 1H), 3.87 (t, J=3.8 Hz, 1H), 5.03 (s, 4H),
5.47 (s, 2H), 6.53–6.58 (m, 2H), 6.59–6.62 (m, 1H), 7.23–7.44 (m,
10H), 7.78 ppm (s, 1H); 13C NMR (75 MHz, MeOD): d=26.4, 47.5,
54.9, 56.2, 69.0, 69.3, 70.4, 71.2, 103.1, 108.4, 124.7, 128.6, 128.9,
129.5, 138.4, 138.8, 144.6, 161.8 ppm; IR (neat): n˜ =3253 cmꢀ1 (OꢀH
and NꢀH); HRMS (ESI): m/z [M+H+] calcd for C29H33N4O3: 517.245,
found: 517.244.
Biological evaluation
Fibroblast culture assay: Skin fibroblasts were obtained from non-
neurological (type 1) and neurological (types 2 and 3) GD patients
with distinct genotypes, which were diagnosed, in most cases
(genotypes:
N370S/N370S,
L444P/L444P,
[D409H;H255Q]/
[D409H;H255Q], G202R/G202R, [D409H;H255Q]/L444P), at the Insti-
tute of Child Health, Athens. In one case (G202R/[L444P;E326K]), di-
agnosis was performed at the Institut de Bioquꢀmica Clꢀnica, Barce-
lona. Fibroblasts from healthy individuals were used as controls. Fi-
broblast cultures were established following routine procedures in
Dulbecco’s modified Eagle’s medium (DMEM) with 10% inactivated
fetal bovine serum (FBS).
Lysosomal glucocerebrosidase assay (GBA1): For the assay of GBA1
in intact cells, 10000 cells were plated into 24-well assay plates
over 6 days in DMEM with 10% FBS serum at 378C under 5% CO2,
either with or without DIX compounds at different concentrations.
Culture media was replaced at day 3 with fresh media supplement-
a-1-C-((1-Decyl-1H-1,2,3-triazol-4-yl)methyl)-1,5-dideoxy-1,5-
imino-d-xylitol (DIX-17): Following the general method, iminoxyli-
tol 1 (18 mg, 0.105 mmol) was combined with 1-azidodecane[23]
(25 mg, 0.137 mmol) to afford DIX-17 as a white solid (22.2 mg,
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1752

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
ed with the corresponding compound dissolved in DMSO. Cells
were washed, and the enzyme assay was performed as follows:
substrate (100 mL, 5 mm 4MU-b-glucopyranoside) in 0.1 m acetate
buffer (pH 5.2) was added to each well, up to a total volume of
260 mL, then the plates were incubated for 1 h at 378C. The
enzyme reaction was stopped with 2 mL of 100 mm glycine/NaOH,
pH 10.7, and the released fluorescence was measured (lex: 355 nm;
lem:460 nm). For each experiment, untreated (no compound
added) and treated cells were plated in quadruplicate. Nonspecific
GBA1 activity was evaluated by addition of CBE (2 h 0.5 mm) to
control wells and was shown to account for about 1% of the activ-
ity in control fibroblasts.
Statistical analysis
For all measures, the hypothesis of normality was rejected, and the
nonparametric Mann-Whitney U test was used. Normal distribution
was assessed by Kolmogorov-Smirnov test. The SPSS statistical pro-
gram was used for statistical analysis.
Supporting Information
Toxicity of DIX compounds (MTT assay) and GBA1 enhancements
by DIX compounds on G202R/G202R, N370S/N370S, and GBA1 en-
hancements by DIX compounds on different genotypes. Copies of
1
13C and H NMR spectra for DIX compounds (19 pages).
Non-lysosomal glucocerebrosidase assay (GBA2): The effect of DIX
compounds on GBA2 activity was determined in mouse testes ho-
mogenates, following a procedure described by Walden et al.[45]
with modifications. Homogenates were prepared in 50 mm potassi-
um phosphate buffer, pH 5.8 (1:3; w/v) using a manual glass ho-
mogenizer and were centrifuged for 15 min at 13500 rpm at 48C
(Beckman J2-21). The pellets were washed three times in phos-
phate buffer and resuspended in the same buffer (1:1; w/v). The
concentration was adjusted to 1 mgmLꢀ1 of protein, and the ali-
quots were stored at ꢀ808C. For the GBA2 assay, aliquots of
mouse testes homogenates were pre-incubated at room tempera-
ture with conduritol b-epoxide (CBE) (Toronto Research Chemicals,
Downsview, ON, Canada) at a final concentration of 2.5 mm for
30 min. Then, 20 mL of homogenates and 5 mL of 50 mm potassium
phosphate buffer (pH 5.8), supplemented with products to the de-
sired concentration, were incubated at 378C for 15 min. Then,
15 mL of 4-metylumbelliferyl-b-d-glucoside (Sigma) was added to
a final concentration of 3 mm, and the mixture was incubated at
378C for 120 min. The reaction was stopped by adding 100 mL of
100 mm glycine/NaOH, pH 10.7, and the released fluorescence was
measured (lex: 355 nm; lem:460 nm).
Abbreviations
CuAAC: Cu-promoted alkyne-azide cycloaddition; DAPI: 4’,6-diami-
dino-2-phenylindole; DIX: 1,5-dideoxy-1,5-imino-d-xylitol; ER: en-
doplasmic reticulum; GBA1: b-glucocerebrosidase; GD: Gaucher
disease; GCase: b-glucocerebrosidase.
Acknowledgements
This work was supported by the Institut Universitaire de France
(IUF), the French National Center for Scientific Research (CNRS),
the University of Strasbourg (France), the French National Agency
for Research (ANR) (grant no. 11-BS07-003-02), the Spanish Minis-
try of Science and Innovation (MICINN) (grant SAF 2011-25431),
and the Generalitat de Catalunya, Spain (grants 2009SGR-1072
and 2009SGR-971). The following fellowships are also gratefully
acknowledged: a doctoral fellowship from the French Depart-
ment of Research to C.D, a FI predoctoral fellowship from the
Generalitat de Catalunya to J.S.-V., and a JAE-predoctoral fellow-
ship from the Consejo Superior de Investigaciones Cientꢁficas
(CSIC), Spain to L.D. The “CIBER de Enfermedades Raras” is an ini-
tiative of the Instituto de Salud Carlos III (Spain). The authors
thank the Institut de Bioquꢁmica Clꢁnica (Barcelona, Spain) for
collaboration, and Genzyme Corporation (Cambridge, MA, USA)
for a generous supply of imiglucerase (Cerezyme).
Cytotoxicity assay: The cytotoxicity of the selected compounds and
the cell viability over a period of 6 days were tested by the MTT (3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay[23]
(see Supporting Information).
Immunofluorescence staining and confocal microscopy imaging: For
immunofluorescence, 5000 cells per well were seeded overnight
on 10 mm glass coverslips (Marienfeld) in 24-well plates. The cells
were then incubated for 6 days in fresh medium (DMEM, 10% FBS)
at 378C under 5% CO2 in the absence or presence of 100 nm of
DIX-28. Culture media was replaced every 3 days with fresh media
supplemented with 100 nm of DIX-28. The untreated cells were in-
cubated with 0.1% DMSO. On day 6 of treatment, the cells were in-
cubated for 1 h with 75 nm LysoTracker Red DND-99 (L7528; Invi-
trogen) at 378C. Then, the cells were washed twice with phos-
phate-buffered saline (PBS) and fixed for 10 min with 4% parafor-
maldehyde at room temperature. The cover slips were then
washed 4 times for 5 min each with 0.3 m PBS, and the cells were
permeabilized with 0.1% Tween in PBS. The cells were then incu-
bated with primary antibody (mouse monoclonal anti-GBA 8E4, di-
luted 1:100)[46] in a solution of PBS with 10% NDS (normal donkey
serum, Millipore) and 0.1% Tween at 48C overnight. The coverslips
were then washed three times with 0.3 m PBS and incubated for
1 h with secondary antibody anti-mouse (Cy2-conjugated AffinPure
donkey anti-mouse IgG (H+L), diluted 1:100, Jackson Immuno Re-
search Laboratories, Inc.) followed by DAPI staining (1:10000 dilu-
tion in a solution of PBS with 10% NDS and 0.1% Tween for
10 min; Invitrogen). Staining was viewed with a Leica TCS-SP2, and
the images were analyzed using Fiji-Image J software.
Keywords: activity enhancement
chemistry · Gaucher disease · glucosyl ceramide · iminoxylitol
·
chaperones
·
click
[1] F. M. Platt, B. Boland, A. C. van der Spoel, J. Cell Biol. 2012, 199, 723–
734.
[2] E. Beutler, G. A. Grabowski in The Metabolic and Molecular Bases of In-
herited Diseases (Eds.: C. R. Scriver, A. L. Beaudet, W. S. Sly, D. Valle),
McGraw-Hill, New York, 2001, pp. 3635–3668.
[3] F. M. Platt, R. H. Lachmann, Biochim. Biophys. Acta - Mol. Cell Res. 2009,
1793, 737–745.
[4] M. Beck, IUBMB Life 2010, 62, 33–40.
[5] S. S. Vembar, J. L. Brodsky, Nat. Rev. Mol. Cell Biol. 2008, 9, 944–957.
[6] D. Ringe, G. A. Petsko, J. Biol. 2009, 8, 80.
[7] R. A. Denny, L. K. Gavrin, E. Saiah, Bioorg. Med. Chem. Lett. 2013, 23,
1935–1944.
[8] I. Bendikov-Bar, G. Maor, M. Filocamo, M. Horowitz, Blood Cells Mol. Dis.
2013, 50, 141–145.
[9] A. R. Sawkar, W. C. Cheng, E. Beutler, C. H. Wong, W. E. Balch, J. W. Kelly,
Proc. Natl. Acad. Sci. USA 2002, 99, 15428–15433.
[10] 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.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1753

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[11] J. M. Benito, F. J. M. Garcꢀa, O. C. Mellet, Expert Opin. Ther. Pat. 2011, 21,
885–903.
[12] H. H. Chang, N. Asano, S. Ishii, Y. Ichikawa, J. Q. Fan, FEBS J. 2006, 273,
4082–4092.
[13] R. Steet, S. Chung, W. S. Lee, C. W. Pine, H. Do, S. Kornfeld, Biochem.
Pharmacol. 2007, 73, 1376–1383.
[14] K. J. Valenzano, R. Khanna, A. C. Powe, R. Boyd, G. Lee, J. J. Flanagan,
E. R. Benjamin, Assay Drug Dev. Technol. 2011, 9, 213–235.
[15] L. Yu, K. Ikeda, A. Kato, I. Adachi, G. Godin, P. Compain, O. Martin, N.
Asano, Bioorg. Med. Chem. 2006, 14, 7736–7744.
[30] S. Desvergnes, Y. Vallee, S. Py, Org Lett. 2008, 10, 2967–2970.
[31] O. R. Martin, O. M. Saavedra, F. Xie, L. Liu, S. Picasso, P. Vogel, H. Kizu, N.
Asano, Bioorg. Med. Chem. 2001, 9, 1269–1278.
[32] O. Gaurat, J. Xie, J. Valery, J. Carbohydr. Chem. 2003, 22, 645–656.
[33] K. Okano, H. Tokuyama, T. Fukuyama, Chem. Asian J. 2008, 3, 296–309.
[34] H. R. Mellor, F. M. Platt, R. A. Dwek, T. D. Butters, Biochem. J. 2003, 374,
307–314.
[35] U. H. Schueler, T. Kolter, C. R. Kaneski, G. C. Zirzow, K. Sandhoff, R. O.
Brady, J. Inherited Metab. Dis. 2004, 27, 649–658.
[36] M. Schmitz, M. Alfalah, J. M. Aerts, H. Y. Naim, K. P. Zimmer, Int. J. Bio-
chem. Cell Biol. 2005, 37, 2310–2320.
[37] H. Dvir, M. Harel, A. A. McCarthy, L. Toker, I. Silman, A. H. Futerman, J. L.
Sussman, EMBO Rep. 2003, 4, 704–709.
[38] A. R. Sawkar, S. L. Adamski-Werner, W. C. Cheng, C. H. Wong, E. Beutler,
K. P. Zimmer, J. W. Kelly, Chem. Biol. 2005, 12, 1235–1244.
[39] Z. Luan, K. Higaki, M. Aguilar-Moncayo, H. Ninomiya, K. Ohno, M. I.
Garcia-Moreno, C. O. Mellet, J. M. G. Fernandez, Y. Suzuki, ChemBioChem
2009, 10, 2780–2792.
[16] P. Compain, O. R. Martin, C. Boucheron, G. Godin, L. Yu, K. Ikeda, N.
Asano, ChemBioChem 2006, 7, 1356–1359.
[17] F. Oulaidi, S. Front-Deschamps, E. Gallienne, E. Lesellier, K. Ikeda, N.
Asano, P. Compain, O. R. Martin, ChemMedChem 2011, 6, 353–361.
[18] N. Dekker, T. Voorn-Brouwer, M. Verhoek, T. Wennekes, R. S. Narayan, D.
Speijer, C. E. Hollak, H. S. Overkleeft, R. G. Boot, J. M. Aerts, Blood Cells
Mol. Dis. 2011, 46, 19–26.
[19] T. Wennekes, R. J. B. H. N. van den Berg, T. J. Boltje, W. E. Donker-Koop-
man, B. Kuijper, G. A. van der Marel, A. Strijland, C. P. Verhagen, J. Aerts,
H. S. Overkleeft, Eur. J. Org. Chem. 2010, 1258–1283.
[40] S. D. Orwig, Y. L. Tan, N. P. Grimster, Z. Yu, E. T. Powers, J. W. Kelly, R. L.
Lieberman, Biochemistry 2011, 50, 10647–10657.
[20] P. Serrano, J. Casas, M. Zucco, G. Emeric, M. Egido-Gabas, A. Llebaria, A.
Delgado, J. Comb. Chem. 2007, 9, 43–52.
[41] Z. Yu, A. R. Sawkar, L. J. Whalen, C. H. Wong, J. W. Kelly, J. Med. Chem.
2007, 50, 94–100.
[21] M. Egido-Gabꢃs, D. Canals, J. Casas, A. Llebaria, A. Delgado, ChemMed-
Chem 2007, 2, 992–994.
[22] G. Sꢃnchez-Ollꢄ, J. Duque, M. Egido-Gabꢃs, J. Casas, M. Lluch, A.
Chabꢃs, D. Grinberg, L. Vilageliu, Blood Cells Mol. Dis. 2009, 42, 159–
166.
[23] L. Dꢀaz, J. Casas, J. Bujons, A. Llebaria, A. Delgado, L. Dꢀaz, J. Med. Chem.
2011, 54, 2069–2079.
[24] L. Dꢀaz, J. Bujons, J. Casas, A. Llebaria, A. Delgado, J. Med. Chem. 2010,
53, 5248–5255.
[25] J. D. Diot, I. G. Moreno, G. Twigg, C. O. Mellet, K. Haupt, T. D. Butters, J.
Kovensky, S. G. Gouin, J. Org. Chem. 2011, 76, 7757–7768.
[26] E. D. Goddard-Borger, M. B. Tropak, S. Yonekawa, C. Tysoe, D. J. Mahur-
an, S. G. Withers, J. Med. Chem. 2012, 55, 2737–2745.
[27] C. Decroocq, L. M. Laparra, D. Rodriguez-Lucena, P. Compain, J. Carbo-
hydr. Chem. 2011, 30, 559–574.
[42] G. Tiscornia, E. L. Vivas, L. Matalonga, I. Berniakovich, M. B. Monasterio,
C. Eguizꢃbal, L. Gort, F. Gonzꢃlez, C. O. Mellet, J. M. G. Fernꢃndez, A.
Ribes, A. Veiga, I. J. C. Belmonte, Hum. Mol. Genet. 2013, 22, 633–645.
[43] P. Alfonso, V. Andreu, A. Pino-Angeles, A. A. Moya-Garcia, M. I. Garcia-
Moreno, J. C. Rodriguez-Rey, F. Sanchez-Jimenez, M. Pocovi, C. O. Mellet,
J. M. G. Fernandez, P. Giraldo, ChemBioChem 2013, 14, 943–949.
[44] A. Amore, R. van Heerbeek, N. Zeep, J. van Esch, J. N. Reek, H. Hiemstra,
J. H. van Maarseveen, J. Org. Chem. 2006, 71, 1851–1860.
[45] C. M. Walden, R. Sandhoff, C. C. Chuang, Y. Yildiz, T. D. Butters, R. A.
Dwek, F. M. Platt, A. C. van der Spoel, J. Biol. Chem. 2007, 282, 32655–
32664.
[46] J. M. Aerts, W. E. Donker-Koopman, G. J. Murray, J. A. Barranger, J. M.
Tager, A. W. Schram, Anal. Biochem. 1986, 154, 655–663.
Received: February 11, 2014
Published online on June 27, 2014
[28] O. N. Nadein, A. Kornienko, Org. Lett. 2004, 6, 831–834.
[29] S. Mꢅller, B. Liepold, G. J. Roth, H. J. Bestmann, Synlett 1996, 521–522.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1744 – 1754 1754