Docking study and biological evaluation of pyrrolidine-based iminosugars as pharmacological chaperones for Gaucher disease

Article abstract of DOI:10.1039/c5ob02223a

We report on the synthesis and biological evaluation of a series of α-1-C-alkylated 1,4-dideoxy-1,4-imino-d-arabinitol (DAB) derivatives as pharmacological chaperones for Gaucher disease. The parent compound, DAB, did not show inhibition of human β-glucocerebrosidase but showed moderate intestinal α-glucosidase inhibition; in contrast, extension of α-1-C-alkyl chain length gave a series of highly potent and selective inhibitors of the β-glucocerebrosidase. Our design of α-1-C-tridecyl-DAB (5j) produced a potent inhibitor of the β-glucocerebrosidase, with IC₅₀ value of 0.77 μM. A molecular docking study revealed that the α-1-C-tridecyl group has a favorable interaction with the hydrophobic pocket and the sugar analogue part (DAB) interacted with essential hydrogen bonds formed to Asp127, Glu235 and Glu340. Furthermore, α-1-C-tridecyl-DAB (5j) displayed enhancement of activity at an effective concentration 10-times lower than isofagomine. α-1-C-Tridecyl-DAB therefore provides the first example of a pyrrolidine iminosugar as a new class of promising pharmacological chaperones with the potential for treatment of Gaucher disease. 2016 The Royal Society of Chemistry.

Full text of DOI:10.1039/c5ob02223a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Docking study and biological evaluation of
pyrrolidine-based iminosugars as pharmacological
Cite this: DOI: 10.1039/c5ob02223a
chaperones for Gaucher disease†
Atsushi Kato,*a Izumi Nakagome,b Kasumi Sato,a Arisa Yamamoto,a Isao Adachi,a
Robert J. Nash,c George W. J. Fleet,d Yoshihiro Natori,e Yasuka Watanabe,e
Tatsushi Imahori,e Yuichi Yoshimura,e Hiroki Takahatae and Shuichi Hirono*b
We report on the synthesis and biological evaluation of a series of α-1-C-alkylated 1,4-dideoxy-1,4-
imino-D-arabinitol (DAB) derivatives as pharmacological chaperones for Gaucher disease. The parent
compound, DAB, did not show inhibition of human β-glucocerebrosidase but showed moderate intestinal
α-glucosidase inhibition; in contrast, extension of α-1-C-alkyl chain length gave a series of highly potent
and selective inhibitors of the β-glucocerebrosidase. Our design of α-1-C-tridecyl-DAB (5j) produced a
potent inhibitor of the β-glucocerebrosidase, with IC50 value of 0.77 μM. A molecular docking study
revealed that the α-1-C-tridecyl group has a favorable interaction with the hydrophobic pocket and the
sugar analogue part (DAB) interacted with essential hydrogen bonds formed to Asp127, Glu235 and
Glu340. Furthermore, α-1-C-tridecyl-DAB (5j) displayed enhancement of activity at an effective concen-
tration 10-times lower than isofagomine. α-1-C-Tridecyl-DAB therefore provides the first example of a
pyrrolidine iminosugar as a new class of promising pharmacological chaperones with the potential for
treatment of Gaucher disease.
Received 28th October 2015,
Accepted 26th November 2015
DOI: 10.1039/c5ob02223a
www.rsc.org/obc
dysfunction.2 Gaucher disease has three clinical phenotypes
with differences in primary neurologic involvement and age of
Introduction
Gaucher disease is caused by mutations in the GBA1 gene onset. Type I presents clinical symptoms without neuro-
encoding β-glucocerebrosidase (acid β-glucosidase, EC nopathic involvement. In contrast, types II and III patients have
3.2.1.45), responsible for the hydrolysis of glucosylceramide to central nervous system impairment. Type I Gaucher disease is
ceramide and glucose.1 The mutations of GBA1 exhibit total or in the most common form, occurring in approximately 1 in
partial deficiency of β-glucocerebrosidase activity as a conse- 40 000–60 000 live births in the general population and 1 in
quence of protein misfolding and degradation by the endo- 500–1000 among the Ashkenazi Jewish population.3 GBA1 is
plasmic reticulum (ER) and this causes the progressive located in chromosome 1q21. There is an adjacent pseudogene
accumulation of glucosylceramide in macrophages. This with 96% exonic sequence homology that further complicates
abnormal metabolite eventually leads to various clinical mani- mutation detection strategies.4
festations that include hepatosplenomegaly, secondary hypers-
To date, more than 200 different point mutations have been
plenism, anaemia, skeletal abnormality, and neurologic identified within the GBA1 gene.5 Among them, the mutation
N370S is the most common genotype in the Ashkenazi Jewish
population which has a less severe form of type 1 Gaucher
disease. Previous studies suggested that N370S mutation
aDepartment of Hospital Pharmacy, University of Toyama, 2630 Sugitani,
Toyama 930-0194, Japan. E-mail: kato@med.u-toyama.ac.jp
affects the α-helix 7 at the interface between the catalytic
bSchool of Pharmaceutical Sciences, Kitasato University, Tokyo 108-8641, Japan.
domain III and domain II and it caused an overall confor-
E-mail: hironos@pharm.kitasato-u.ac.jp
mational change of β-glucocerebrosidase.6 Thus, the mutation
cInstitute of Biological, Environmental and Rural Sciences/Phytoquest Limited,
point is not close to the catalytic site: the mutated enzyme
keeps the kinetic properties, while its thermostability is lower
than the wild-type enzyme at neutral pH conditions.7,8 N370S
mutated enzyme exhibits residual hydrolytic activity (∼32%),
which is higher than the other mutations G202R (∼10%) and
L444P (∼12%).9
Plas Gogerddan, Aberystwyth, Ceredigion SY23 3EB, UK
dChemistry Research Laboratory, Department of Chemistry, University of Oxford,
Oxford, OX1 3TA, UK
eFaculty of Pharmaceutical Sciences, Tohoku Pharmaceutical University,
Sendai 981-8558, Japan
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5ob02223a
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Pharmacological chaperone therapy is based on the sufficient to significantly reduce symptoms of Gaucher
concept that reversible competitive inhibitors assist the stabil- disease. One of the reasons for this may be that while isofago-
ization of the three-dimensional conformation of unstable mine has favorable van der Waals interactions and hydrogen
mutated enzymes and facilitate the proper folding, thereby bonding networks with β-glucocerebrosidase, it has no inter-
accelerating the mature processing and transport to the final actions with the hydrophobic domain.15
cellular destination of the protein.10 In order to establish the
In this context, recent pharmacological chaperones for
concept of using competitive inhibitors as specific pharmaco- Gaucher disease belong mainly to three different pyranose
logical chaperones, a number of naturally occurring and sugar mimics (Fig. 1): (i) 1-deoxynojirimycin (DNJ)-based
chemically synthesized iminosugars were tested for intracellu- pharmacological chaperones, such as N-nonyl-DNJ,11 α-1-C-
lar enhancement of β-glucocerebrosidase activity.2
octyl-DNJ,16 6S-NAdB-NJ,17 (ii) isofagomine (IFG)-based
N-Nonyl-1-deoxynojirimycin (N-nonyl-DNJ: Fig. 1) is an pharmacological chaperones, such as N-butyl-IFG,18 α-7-C-
effective pharmacological chaperone for the treatment of nonyl-IFG,15
N-[5-{(adamantan-1-ylmethyl)amino}-5-oxopen-
Gaucher disease.11 The parent iminosugar, DNJ, closely tyl]-IFG,19 and (iii) 1,5-dideoxy-1,5-imino-D-xylitol (DIX)-based
mimics the D-glucose structure. This structural similarity to pharmacological chaperones, such as α-1-C-nonyl-DIX,20 2-O-
D-glucose can achieve tight affinity with β-glucocerebrosidase; hexyl-DIX,21 α-1-[2-(4-morpholino-4-oxobutylthio)ethyl]-DIX.22
it is also known that these DNJ-based chaperones are potent
These pharmacological chaperones are all limited to the
inhibitors of ER processing α-glucosidases and intestinal piperidine structures as pyranose analogues; new approaches
digestive α-glucosidases. This non-specific α-glucosidase inhi- and concepts need to be explored. Thus, we turned our atten-
bition in various organs causes side effects such as diarrhea tion to the design of pyrrolidine-based pharmacological cha-
and impairment of liver function. Similarly, N-butyl-1-deoxyno- perones. It is not usually easy to predict the glycosidase
jirimycin (Zavesca®) with a shorter N-alkyl chain was launched inhibition profiles of polyhydroxylated pyrrolidine alkaloids
as a potential drug for substrate reduction therapy (SRT). The based on the configuration of the hydroxyl groups.23 For the
basic concept of SRT is to reduce the rate of biosynthesis of rational drug design of clinically available pharmacological
glycosphingolipids to offset the catabolic defect, restoring the chaperones, we first reported the synthesis from the enantio-
balance between the rate of biosynthesis and the rate of cata- mers of glucuronolactone of all 10 stereoisomeric 2,5-dideoxy-
bolism. Despite this therapeutic breakthrough, several draw- 2,5-iminohexitols.24 A side-by-side comparison showed that 8
backs concerning this therapy have also been reported.12 The of the 10 stereoisomers gave significant inhibition of varying
most common adverse events due to Zavesca® included weight glycosidases. On the basis of these preliminary studies, we pre-
loss, diarrhea, and trembling of the hand in approximately pared 1,4-dideoxy-1,4-imino-D-arabinitol (DAB) with different
30% of patients; this may be because Zavesca® inhibits not alkyl chain lengths. In particular, we have investigated the
only glucosyltransferase but also intestinal digestive glycosi- effect of the configuration of the hydroxyl groups on the pyrro-
dases and ER processing α-glycosidase I and II.13
lidine ring and alkyl chain length on the stabilization
In phase II clinical trials it was shown that isofagomine of β-glucocerebrosidase under heat-shock conditions and
(Plicera®: Fig. 1), also a potent inhibitor of β-glucocerebrosi- pharmacological chaperone effects against N370S Gaucher
dase in vitro, increases patient β-glucocerebrosidase activity fibroblasts.
without unwanted side effects;14 however, the effects were not
Results and discussion
Chemistry
We previously reported the synthesis of α-1-C-alkyl-LABs25 and
-DABs 5a and 5e.26 By similar procedures, α-1-C-alkyl-DABs 5a–
j with various lengths of side chains from C4 to C13 at the
1-position were synthesized. A chiral common intermediate
1 25,26 was subjected to the Negishi coupling reaction with
alkylzinc bromides to afford coupled products 2a–j in 57–89%
yields. Epoxidation of 2a–j with trifluorodimethyldioxirane,
generated by the treatment of trifluoroacetone and Oxone™,
gave epoxides 3a–j as single diastereomers in good yields. As
we reported in previous studies, the steric bulk of a C1 substi-
tuent ensured that highly diastereoselective addition occurred
from the α-side of the dihydropyrrole ring.25,26 Cleavage of the
oxirane ring of 3a–j under acidic conditions afforded trans-
diols 4a–j which were further treated with sodium hydroxide in
aqueous ethanol to give the desired α-1-C-alkyl-DABs 5a–j in
Fig. 1 Structures of representative piperidine iminosugar-type GCase
pharmacological chaperon.
46–90% yields (Scheme 1).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
inhibitory potency against maltase and sucrase, whereas intro-
duction of an N-butyl substituent into LAB reduced effective-
ness as a glycosidase inhibitor.
Table 1 shows the 50% inhibitory concentrations (IC50) of
α-1-C-alkyl-DAB and LAB against rat intestinal maltase,
sucrase, and human β-glucocerebrosidase. The parent com-
pound, DAB was a much weaker inhibitor of rat intestinal
maltase and sucrase inhibitor than LAB. Furthermore, it is
noteworthy that both DAB and LAB itself did not show inhibi-
tory activity against human β-glucocerebrosidase. The intro-
duction of a butyl chain at C-1 position of DAB reduced their
effectiveness as rat intestinal α-glucosidase inhibitors, while it
was still inactive against human β-glucocerebrosidase. In
sharp contrast, α-1-C-octyl-DAB showed good inhibitory
potency against β-glucocerebrosidase, with an IC50 value of
Scheme 1 Synthesis of α-1-C-alkyl-DAB. Reagents: (a) alkylzinc
bromide (3.2 eq.), Ni(cod)2 (16 mol%), (R,R)-2,6-bis(4-isopropyl-2- 6.2 μM. This behaviour could not be observed in α-1-C-octyl-
oxazolin-2-yl)pyridine (32 mol%), N,N-dimethylacetamide, rt, 20
h
LAB. These results indicated that the addition of an alkyl
chain to the C-1 of DAB might lead to the production of a
highly potent and selective inhibitor of human β-glucocerebro-
sidase. Thus, we focused on the anomeric position and
whether extension of the alkyl chain at this position could
influence the inhibition characteristics of DAB. α-1-C-Butyl-
DAB (5a) had no effect against this enzyme, whereas α-1-C-
pentyl-DAB (5b) showed moderate inhibitory activity, with IC50
value of 34 μM (Table 2).
(b) Oxone™ (5.0 eq.), CF3COCH3 (10.0 eq.), NaHCO3 (7.5 eq.), 0 °C, 2 h
(c) CF3CO2H (12.0 eq.), THF/H2O (3 : 2), reflux, 72 h (d) NaOH (10 eq.),
EtOH/H2O (2 : 1), reflux, 1 h.
Influence of α-1-C-alkylation toward LAB and DAB as
glycosidase inhibitors
We first compared α-1-C-alkylated LAB and DAB analogues.
Further extension of the length of the α-1-C-alkyl chain
Our design strategy was based on experience gained in our pre- improved the inhibition potency accordingly (5c–5j). The
vious studies on α-1-C-butyl-LAB26 and L-isoDMDP.27 These most potent inhibitor α-1-C-tridecyl-DAB (5j) displayed a
pyrrolidine derivatives were designed as selective inhibitors of remarkable IC50 value of 0.77 μM, which is 44-fold more
intestinal α-glucosidases; they showed a strong and early- potent relative to α-1-C-pentyl-DAB (5b). These results clearly
phase suppression of postprandial hyperglycemia with an indicate that a much longer alkyl chain than butyl is
effective dose 10 times lower than that for miglitol, a drug pre- required for achieving strong inhibition of β-glucocerebrosi-
sently used in the treatment of late onset diabetes. Further- dase. However, it is noteworthy that their inhibitory potency
more, we also showed that the introduction of a butyl chain at did not show linear characteristics but increased stepwise as
C-1 of LAB to give α-1-C-butyl-LAB dramatically improved its shown in Fig. 2.
Table 1 Concentration of α-1-C-alkyl-LAB and DAB derivatives giving 50% inhibition of rat intestinal maltase, sucrase, and human
β-glucocerebrosidase
IC50 (μM)
LAB derivatives
DAB derivatives
Enzyme
H
C4H9
C8H17
H
C4H9
C8H17
α-Glucosidase
Rat intestinal maltase
Rat intestinal sucrase
0.93a
1.0a
0.13a
0.32a
0.45a
55
16
NI
393
500
590
0.032a
β-Glucosidase
Human β-glucocerebrosidase
NIb
NI
NI
NIb
NI
6.2
a Our previous data (Kato, et al., J. Med. Chem., 55, 10347–10362). b NI: no inhibition (less than 50% inhibition at 1000 μM).
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 2 IC50 values (μM) for α-1-C-alkyl-DAB (5a–5j) against human
β-glucocerebrosidase
IC50 (μM)
Compounds
β-Glucocerebrosidase
α-1-C-Butyl-DAB (5a)
α-1-C-Pentyl-DAB (5b)
α-1-C-Hexyl-DAB (5c)
α-1-C-Heptyl-DAB (5d)
α-1-C-Octyl-DAB (5e)
α-1-C-Nonyl-DAB (5f)
α-1-C-Decyl-DAB (5g)
α-1-C-Undecyl-DAB (5h)
α-1-C-Dodecyl-DAB (5i)
α-1-C-Tridecyl-DAB (5j)
NIa
34
30
38
6.2
3.3
1.9
2.3
1.5
0.77
a NI: less than 50% inhibition at 1000 μM.
Fig. 3 Effect of α-1-C-alkyl-DAB (10 μM) on the in vitro thermostability.
The enzyme β-glucocerebrosidase was incubated at 45 °C in 100 mM
McIlvaine buffer (pH 5.2) containing 0.25% sodium taurocholate and
0.1% Triton X-100 for 0, 20, 40, and 60 min. After incubation with isofa-
gomine (IFG), α-1-C-pentyl-DAB (5b), α-1-C-nonyl-DAB (5f), and α-1-C-
tridecyl-DAB (5j), and then the remaining β-glucocerebrosidase activity
was assayed immediately using 3 mM 4-methylumbelliferl-β-D-gluco-
pyranoside as substrate. Each value represents the mean SEM (n = 3).
Intracellular enhancement of residual enzyme activity in
human N370S fibroblasts by α-1-C-alkyl-DAB
We next evaluated the chaperone activity of α-1-C-alkyl-DAB in
vitro in fibroblasts from a Gaucher patient with the N307S
mutation. Cells were cultured in the presence of α-1-C-heptyl-
DAB (5b: 25–100 μM), α-1-C-nonyl-DAB (5f: 1.0–5.0 μM), and
α-1-C-tridecyl-DAB (5j: 0.25–1.0 μM) for 6 days, and the intra-
cellular enzyme activities were determined with 4-methyl-
umbelliferyl β-glucoside as substrate (Fig. 4). Isofagomine
(5 μM), which is under Phase II clinical trial toward
Gaucher disease, was used as a positive control. Relative to
non-treated (NT), α-1-C-heptyl-DAB (5b) improved intracellular
Fig. 2 Representation of pIC50 for α-1-C-alkyl-DAB (5b–5j) against
human β-glucocerebrosidase as a bar graph.
In vitro stabilization effect of α-1-C-alkyl-DAB against
heat-shock β-glucocerebrosidase
We next investigated the effect of these compounds on the
stabilization of this enzyme at 45 °C (Fig. 3). The enzyme
β-glucocerebrosidase was incubated in 100 mM McIlvaine
buffer (pH 5.2) containing 0.25% sodium taurocholate and 0.1%
Triton X-100 for 0, 20, 40 and 60 min and the remaining
enzyme activity was determined with 4-methylumbellifery-β-D-
glucopyranoside as substrate. Isofagomine was used as positive
control. The enzyme activity was lost within 60 min under
incubation without test samples (2.3% activity remained at
60 min), while the enzyme activity remained over 60% even
after incubation for 60 min in the presence of 10 μM isofago-
mine (Fig. 3). Whereas 10 μM α-1-C-nonyl-DAB (5f) tended to
slightly protect the enzyme from heat stress (6.5% activity
remained at 60 min), the stabilization effect was much weaker
than α-1-C-tridecyl-DAB (5j). α-1-C-Tridecyl-DAB (5j) was the
most efficient compound in this test, achieving 56 and 40%
β-glucocerebrosidase protection after 20 and 40 min at 10 μM,
respectively. These results suggested that extension of the α-1-
C-alkyl chain length contributed to the thermostabilizing
effects.
Fig. 4 Effect of α-1-C-alkyl-DAB (5b, 5f, and 5j) on N370S GCase in
Gaucher cells treated with increasing concentrations of 5a, 5f, or 5j.
Each value represents the mean SEM (n = 3).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
β-glucocerebrosidase activities in N370S fibroblasts, with increases
of 1.6–1.7-fold with optimal concentrations of 100 μM. α-1-C-
Nonyl-DAB (5f) was also assayed under the same conditions
and was effective as a pharmacological chaperone at a lower
dose than α-1-C-heptyl-DAB (5b). The concentration range of
achieving maximum activity enhancement was clearly lower
than that of α-1-C-heptyl-DAB (5b). It is noteworthy that, α-1-C-
tridecyl-DAB (5j) displayed comparable enhancement of
activity to that obtained with isofagomine but at an effective
concentration which was 10 times lower than isofagomine.
These apparent differences between the enzyme affinities and/
or enzyme stabilization effects by using naked enzyme and
mutant β-glucocerebrosidase activity enhancement assay in
N370S Gaucher fibroblasts could be attributed to differential
penetration of the cells or organelles by isofagomine and α-1-
C-tridecyl-DAB (5j). These results clearly suggested that hydro-
phobic side-chain derivative, such as α-1-C-tridecyl-, would
assist in the cell-membrane-penetration of the inhibitor and
contribute to its recognition in the hydrophobic pocket of
β-glucocerebrosidase. Therefore, the correlation between the
enzyme affinities and mutant enzyme enhancement activities
in mutant Gaucher’s cell could be used to estimate how well
the inhibitor is able to reach an intracellular therapeutic
target.
Fig. 5 Immunocytochemistry of glucocerebrosidase (Gcase) in N370S
Gaucher fibroblasts treated with α-1-C-alkyl-DAB (5b, 5f, and 5j). N370S
Gaucher fibroblasts were cultured with or without 100 μM α-1-C-
pentyl-DAB (5b), 10 μM α-1-C-nonyl-DAB (5f), and 1.0 μM α-1-C-tride-
cyl-DAB (5j) for 6 days. Immunocytochemistry was performed in which
cells were double labeled with GCase antibody (green), and a lysosome
marker (LAMP-1, red).
In vitro normalization effect of α-1-C-alkyl-DAB against cellular
trafficking of the N370S mutant
C-tridecyl-DAB (5j) occupied the active-site of this enzyme in
On the basis of these findings, it is clear that α-1-C-alkyl-DAB much the same way as isofagomine. In order to investigate the
(5b, 5f, and 5j) led to the stabilization of mutant β-glucocere- interaction between β-glucocerebrosidase and α-1-C-tridecyl-
brosidase structure and consequently increased the intracellular DAB (5j), we constructed the three-dimensional structure of
activity. Thus, we next focused on the subcellular localization the β-glucocerebrosidase-α-1-C-tridecyl-DAB (5j) complex using
and whether these iminosugars could improve the transport Induced Fit Docking protocol (see Experimental section:
and subcellular distribution of the mutant β-glucocerebrosi- docking studies). As shown in Fig. 6, the pyrrolidine group of
dase. Gaucher patient fibroblasts were incubated in the pres- α-1-C-tridecyl-DAB (5j) was found to bind to the same pocket
ence or absence of 100 μM α-1-C-pentyl-DAB (5b: Fig. 5D–F) or as the piperidine group of isofagomine in the crystal structure
10 μM α-1-C-nonyl-DAB (5f: Fig. 5G–I) or 1.0 μM α-1-C-tridecyl- in complex with β-glucocerebrosidase (pdb_2NSX). The pyrroli-
DAB (5j: Fig. 5J–L) for 6 days. The subcellular location of β-glu- dine group of α-1-C-tridecyl-DAB (5j) had favorable van der
cocerebrosidase was then determined by double immunofluor- Waals interactions with Tyr313 and Trp381 of β-glucocerebrosi-
escence analysis. Compared with no treatment (Fig. 5A), dase. These van der Waals interactions were similar to that of
treatment with 1.0 μM α-1-C-tridecyl-DAB (5j) resulted in the piperidine group of isofagomine. The positive nitrogen
increase in the intensity of GCase fluorescence throughout the and the hydroxyl groups of α-1-C-tridecyl-DAB (5j) formed
cells (Fig. 5J). Furthermore, it is different from the non-treated hydrogen bonding with Glu235, Glu340, Asp127 and Trp179 in
cells (Fig. 5A–C) where the GCase staining pattern showed sub- β-glucocerebrosidase. We have previously reported that the
stantially overlap with that for lysosomal marker LAMP-1 ionic hydrogen bonds to Glu235 and Glu340 and non-ionic
(Fig. 5J–L). These results clearly indicated that α-1-C-tridecyl- hydrogen bond to Asp127 are essential for the ligand binding
DAB (5j) improved the subcellular trafficking and distribution with β-glucocerebrosidase.28
of mutant enzyme.
Thus, α-1-C-tridecyl-DAB (5j) has the favorable hydrogen
bonding network with β-glucocerebrosidase in the same as
with isofagomine. It is noteworthy that the tridecyl group of
α-1-C-tridecyl-DAB (5j) is located in the long hydrophobic
Docking study of α-1-C-tridecyl-DAB (5j) to human
β-glucocerebrosidase
To understand the structural basis of the interaction of α-1-C- pocket of Trp312, Tyr313, Phe316, Val343 and the alkyl side
tridecyl-DAB (5j) with human lysosomal β-glucocerebrosidase, chain of Lys346 and forming a favorable hydrophobic inter-
we first determined the mode of inhibition and the inhibition action (Fig. 7). The target enzyme β-glucocerebrosidase is
constant (Ki) from Lineweaver–Burk plots. As a result, α-1-C-tri- responsible for the cleavage of the β-glycosidic bond of
decyl-DAB (5j) inhibited β-glucocerebrosidase in a competitive glucosylceramide to release ceramide and glucose. The active
manner, with Ki value of 0.71 μM. This result suggests that α-1- site, therefore, is considered to be comprised of sites to
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 6 Panel A and C show the binding modes of α-1-C-tridecyl-DAB (5j) (A) and isofagomine (C) (pdb code 2NSX) with human β-glucocerebro-
sidase. The residues related to van der Waals interactions are drawn in space-filling representation. The hydrogen bonds are depicted by dashed lines.
Panel B and D show the schematic diagram of hydrogen bonding interactions and vdW interactions of α-1-C-tridecyl-DAB (5j) (B) and isofagomine
(D) with human β-glucocerebrosidase.
potency was improved by extension of the α-1-C-alkyl chain
length (5c–5j). These results suggested that the α-1-C-tridecyl
group of α-1-C-tridecyl-DAB (5j) might be recognized by the
hydrophobic pocket and thereby stabilized the favorable pyrro-
lidine-ring orientation, finally leading to an increase in inhibi-
tory activity.
Conclusion
It is clear from this study that pyrrolidine iminosugars provide
a new class of pharmacological chaperones; all previously
reported pharmacological chaperones have been based on
piperidine structures. The present work elucidated the follow-
ing features: (a) α-1-C-tridecyl-DAB showed the most potent
inhibitory activity toward human β-glucocerebrosidase, with an
IC50 value of 0.77 μM; (b) the parent compound, DAB, did not
show any inhibition, whereas addition of an alkyl chain at the
C-1 of DAB led to the production of a highly potent and selec-
tive inhibition. Extension of the α-1-C-alkyl chain length
improved the inhibition potency. A greater length chain than
Fig. 7 The binding of α-1-C-tridecyl-DAB (5j) in the active site of
β-glucocerebrosidase. The molecular surface of β-glucocerebrosidase
was rendered by lipophilic potential using SYBYL-X 2.0 software (Certara
USA, Inc.).
accommodate both the sugar part “glucose” and also the butyl is required for significant inhibition; (c) the pyrrolidine
hydrophobic part “ceramide”. In this study, we demonstrated group of α-1-C-tridecyl-DAB has favorable van der Waals inter-
that α-1-C-butyl-DAB (5a) had no effect against this enzyme, actions (Tyr313 and Trp381) and hydrogen bonding network
whereas α-1-C-pentyl-DAB (5b) showed moderate inhibition, (Glu235, Glu340, Asp127 and Trp179) with β-glucocerebro-
with IC50 value of 34 μM (Table 2). Furthermore, the inhibition sidase. The α-1-C-tridecyl group of α-1-C-tridecyl-DAB occupies
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
the hydrophobic pocket (Trp312, Tyr313, Phe316, Val343 and
(5R,7aS)-5-Dodecyl-1,7a-dihydropyrrolo[1,2-c]oxazol-3(5H)-
1
Lys346); (d) α-1-C-tridecyl-DAB improved the thermostability of one (2i). 60% yield; H-NMR (400 MHz, CDCl3) δ: 0.88 (3H, t,
β-glucocerebrosidase in vitro and displayed comparable J = 6.8 Hz,), 1.26–1.56 (22H, m), 4.22 (1H, dd, J = 5.8, 8.2 Hz),
enhancement of activity at a concentration 10 times lower than 4.53 (1H, brs), 4.60 (1H, t, J = 8.7 Hz), 4.74 (1H, brs), 5.83–5.85
isofagomine. α-1-C-Tridecyl-DAB therefore represents a new (1H, m), 6.03 (1H, ddd, J = 2.4, 2.4, 6.3 Hz). 13C-NMR
class of potential pharmacological chaperones with potential (100 MHz, CDCl3) δ: 14.08, 22.64, 26.19, 29.31, 29.41, 29.49,
for the treatment of Gaucher disease.
29.54, 29.60, 29.62, 31.87, 34.38, 63.84, 67.47, 68.71, 127.96,
135.27, 162.97. IR (neat) cm−1: 1743, 2851, 2917. EI-MS (m/z):
293 (M+). HRMS Calcd for C18H31NO2: 293.2355, Found:
293.2354. [α]2D6 = −124.8 (c 0.15, CHCl3).
Experimental section
General experimental procedures
(5R,7aS)-5-Tridecyl-1,7a-dihydropyrrolo[1,2-c]oxazol-3(5H)-
one (2j). 60% yield; H-NMR (400 MHz, CDCl3) δ: 0.88 (3H, t,
1
Infrared (IR) spectra were recorded on a Perkin-Elmer 1600
series FT-IR spectrometer. Mass spectra (MS) were recorded on
a JEOL JMN-DX 303/JMA-DA 5000 spectrometer. Microanalyses
were performed on a Perkin-Elmer CHN 2400 Elemental Analy-
zer. Optical rotations were measured with a JASCO DIP-360 or
JASCO P-1020 digital polarimeter. Proton nuclear magnetic res-
onance (1H NMR) spectra were recorded on JEOL JNM-AL 400
(400 MHz) spectrometer, using tetramethylsilane as an
internal standard. The following abbreviations are used: s =
singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br =
broad. Column chromatography was carried out on Merck
Silica gel 60 (230–400 mesh) or KANTO Silica Gel 60N
(40–50 mm) for flash chromatography.
J = 6.3 Hz), 1.26–1.54 (24H, m), 4.22 (1H, dd, J = 5.3, 8.7 Hz),
4.53 (1H, brs), 4.60 (1H, t, J = 8.7 Hz), 4.74 (1H, m), 5.85(1H,
dd, J = 1.4, 5.8 Hz), 6.03 (1H, ddd, J = 2.4, 2.4, 6.2 Hz).
13C-NMR (100 MHz, CDCl3) δ: 14.08, 22.64, 26.19, 29.32, 29.41,
29.50, 29.54, 29.60, 29.62, 29.64, 31.87, 34.39, 63.84, 67.47,
68.71, 127.97, 135.27, 162.98. IR (neat) cm−1: 1745, 2849, 2917.
EI-MS (m/z): 307 (M+). HRMS Calcd for C19H33NO2: 307.2511,
Found: 307.2519, [α]1D7 = −107.9 (c 0.52, CHCl3).
Typical procedure for epoxidation
To a vial with alkene 2 (1.18 mmol) were added 11 mL of
CH3CN and 7.6 mL of 4 × 10−4 M ethylenediaminetetraacetic
acid in H2O. The solution was cooled to 0 °C, and 1,1,1-tri-
fluoroacetone (1.33 g, 11.9 mmol) was added. A mixture of
solid Oxone™ (3.63 g, 4.36 mmol) and NaHCO3 (743 mg,
8.85 mmol) was added in four portions over 45 min. The reac-
tion was stirred for 2 h at 0 °C, then diluted with 5 mL of H2O,
and extracted with CH2Cl2 (3 × 25 mL). The organic layers were
combined and dried over anhydrous Na2SO4. Filtration and
evaporation in vacuo furnished the crude product, which was
purified by column chromatography (silica gel eluting with
hexane/EtOAc) to provide epoxides 3a–j. Compounds 3a–h
were identical in 1H-NMR with ent-3a–h which were reported
previously.26
(5S,7aS)-5-(Iodomethyl)-1,7a-dihydropyrrolo[1,2-c]oxazol-3-
(5H)-one (1). The iodide 1 was prepared according to reported
procedure.25,26 [α]D24 = −131.7 (c 1.04, CHCl3). Lit. ent-1: [α]D23
+141.6 (c 1.0, CHCl3).25
=
Typical procedure for the preparation of alkylzinc bromide
Iodine chips (127 mg, 0.50 mmol) were added to a suspension
of Zn powder (981 mg, 15.0 mmol) in N,N-dimethylacetamide
(10 mL) at room temperature. After stirring for 10 min, the
resulting mixture changed from a dark-brown suspension to a
colorless suspension. Alkylbromide (10 mmol) was added by
syringe, and the reaction mixture was stirred at 80 °C for 3 h.
The concentration in N,N-dimethylacetamide was ∼0.75 M.
(1aS,1bR,6R,6aR)-6-Dodecyltetrahydrooxireno[2′,3′:3,4]pyrrolo-
[1,2-c]oxazol-4(1aH)-one (3i). 80% yield; 1H-NMR (400 MHz,
CDCl3) δ: 0.88 (3H, t, J = 6.8 Hz), 1.26–1.51 (22H, m), 3.55
Typical procedure for Negishi coupling
Yellow bis(1,5-cyclooctadiene)nickel(0) (83.1 mg, 0.302 mmol, (1H, d, J = 2.9 Hz), 3.61 (1H, d, J = 2.9 Hz), 3.98 (1H, dd, J = 3.7,
16 mol%) and (R,R)-2,6-bis(4-isopropyl-2-oxazolin-2-yl)pyridine 8.7 Hz), 4.04 (1H, dd, J = 4.8, 9.2 Hz), 4.47 (1H, t, J = 4.3 Hz),
(182 mg, 0.604 mmol, 32 mol%) were added to N,N-dimethyl- 4.55 (1H, dd, J = 9.2, 17.9 Hz). 13C-NMR (100 MHz, CDCl3) δ:
acetamide (12.0 mL) under an argon atmosphere, and the 14.08, 22.64, 26.06, 29.29, 29.31, 29.41, 29.50, 29.59, 29.61,
resulting mixture was stirred for 30 min at room temperature. 30.01, 31.87, 55.26, 57.17, 58.70, 60.45, 64.63, 162.54. IR (neat)
The resulting deep-blue solution was added to a solution of cm−1: 1747, 2850, 2919. EI-MS (m/z): 309 (M+). HRMS Calcd for
the alkylzinc bromide (in N,N-dimethylacetamide, 8.7 mL, C18H31NO3: 309.2304, Found: 309.2309, [α]2D5 = −34.3 (c 0.35,
6.05 mmol) and 1 (500 mg, 1.89 mmol). After being stirred for CHCl3).
20 h, the reaction was quenched with iodine chips (440 mg).
(1aS,1bR,6R,6aR)-6-Tridecyltetrahydrooxireno[2′,3′:3,4]pyrrolo-
After being stirred for 10 min, the dark-brown mixture was [1,2-c]oxazol-4(1aH)-one (3j). 80% yield; 1H-NMR (400 MHz,
passed through a short pad of silica gel (eluting with AcOEt/ CDCl3) δ: 0.88 (3H, t, J = 7.2 Hz), 1.26–1.55 (24H, m), 3.55–3.61
hexane = 1/1) (to remove N,N-dimethylacetamide, inorganic (2H, m), 3.97–4.02 (2H, m), 4.48–4.53 (2H, m), 13C-NMR
salts, and iodine). The filtrate was then concentrated, and the (100 MHz, CDCl3) δ: 14.10, 22.67, 26.08, 29.32, 29.33, 29.43,
residue was purified by flash chromatography (silica gel 29.51, 29.61, 29.66, 30.04, 31.89, 55.26, 57.19, 58.74, 60.46,
eluting with hexane/EtOAc) to provide coupled products 2a–j. 64.64, 162.54. IR (neat) cm−1: 1757, 2850, 2918. EI-MS (m/z):
Compounds 2a–h were identical in 1H-NMR with ent-2a–h 323 (M+). HRMS Calcd for C19H33NO3: 323.2460, Found:
which were reported previously.26
323.2449, [α]1D9 = −26.2 (c 0.21, CHCl3).
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Typical procedure for the preparation of a diol
α-1-C-Dodecyl-DAB (5i). 68% yield, 1H-NMR (400 MHz,
CD3OD) d: 0.92 (3H, t, J = 7.2 Hz), 1.29–1.45 (23H, m), 1.70
(1H, brs), 2.91 (1H, dd, J = 7.7, 12.6 Hz), 3.04 (1H, dd, J = 6.3,
10.6 Hz), 3.58–3.63 (2H, m), 3.71 (1H, dd, J = 3.9, 11.6 Hz), 3.79
(1H, t, J = 6.8 Hz). 13C-NMR (100 MHz, CD3OD) d: 14.44, 23.73,
27.70, 30.48, 30.68, 30.73, 30.74, 30.76, 30.78, 30.80, 33.08,
34.97, 62.87, 64.55, 79.30, 83.29. IR (KBr) cm−1: 2850, 2918,
3292. FAB-MS (m/z): 302 (M+ + 1). HRMS Calcd for C17H36NO3:
302.2695, Found: 302.2693, [α]2D3 = +13.4 (c 0.25, CH3OH).
α-1-C-Tridecyl-DAB (5j). 74% yield; 1H-NMR (400 MHz,
CD3OD) δ: 0.92 (3H, t, J = 7.2 Hz), 1.29–1.59 (23H, m), 1.73–1.83
(1H, m), 3.03–3.08 (1H, m), 3.16–3.20 (1H, m), 3.64–3.77 (3H, m),
3.86 (1H, t, J = 6.3 Hz). 13C-NMR (100 MHz, CD3OD) δ: 14.45,
23.74, 27.60, 30.48, 30.65, 30.72, 30.75, 30.77, 30.80, 30.80, 33.08,
34.38, 62.20, 63.16, 64.78, 78.77, 82.70. IR (KBr) cm−1: 2851,
2919, 3294. EI-MS (m/z): 315 (M+). HRMS Calcd for C18H37NO3:
315.2773, Found: 315.2773, [α]2D2 = +59.7 (c 0.12, CH3OH).
Trifluoroacetic acid (0.70 mL, 9.12 mmol) was added to a solu-
tion of the epoxide 3 (0.773 mmol) in THF/H2O (3/2, 6.8 mL).
After being stirred at 80 °C for 72 h, the reaction mixture was
cooled to room temperature and concentrated. The product
was purified by flash chromatography (silica gel, eluting with
CH2Cl2/CH3OH or Et2O/CH3OH) to yield trans-diols 4a–j. Com-
pounds 4a–h were identical in 1H-NMR with ent-4a–h which
were reported previously.26
(5R,6R,7R,7aR)-6,7-Dihydroxy-5-dodecyltetrahydropyrrolo-
[1,2-c]oxazol-3(1H)-one (4i). 82% yield, 1H-NMR (400 MHz,
CDCl3) δ: 0.88 (3H, t, J = 7.2 Hz), 1.26–1.67 (22H, m), 3.63–3.68
(1H, m), 3.83–3.97 (3H, m), 4.36 (1H, dd, J = 3.4, 9.2 Hz), 4.54
(1H, t, J = 8.7 Hz). 13C-NMR (100 MHz, CDCl3) δ: 14.10, 22.67,
26.19, 29.36, 29.50, 29.63, 29.68, 29.71, 29.73, 29.78, 29.91,
31.92, 33.02, 62.16, 65.35, 67.05, 80.80, 83.49, 162.73. IR (neat)
cm−1: 1717, 2850, 2919, 3379. EI-MS (m/z): 327 (M+). HRMS
Calcd for C18H33NO4: 327.2410, Found: 327.2396, [α]2D5 = −23.4
(c 0.29, CHCl3).
Enzyme assay
Brush border membranes were prepared from the rat small
intestine according to the method of Kessler et al.,29 and were
assayed at pH 6.8 for rat intestinal maltase and sucrase using
the appropriate disaccharide as substrate. The reaction
mixture contained 25 mM substrate and the appropriate
amount of enzyme, and the incubations were performed for
10–30 min at 37 °C. The reaction was stopped by heating at
100 °C for 3 min. After centrifugation (600g; 10 min), 0.05 mL
of the resulting reaction mixture were added to 3 mL of the
Glucose CII-test Wako (Wako Pure Chemical Ind., Osaka,
Japan). The absorbance at 505 nm was measured to determine
the amount of the released D-glucose. The inhibitory activity
toward β-glucocerebrosidase was measured with Cerezyme
(5R,6R,7R,7aR)-6,7-Dihydroxy-5-tridecyltetrahydropyrrolo[1,2-c]-
oxazol-3(1H)-one (4j). 99% yield, H-NMR (400 MHz, CDCl3) δ:
1
0.88 (3H, t, J = 6.8 Hz), 1.26–1.64 (24H, m), 3.49 (1H, brs),
3.85–3.93 (3H, m), 4.31–4.38 (1H, m), 4.46–4.54 (1H, m).
13C-NMR (100 MHz, CDCl3) δ: 14.09, 22.67, 26.13, 29.36,
29.46, 29.51, 29.62, 29.68, 29.73, 29.79, 29.91, 31.92, 32.90,
62.10, 65.20, 67.14, 80.72, 83.36, 162.86. IR (neat) cm−1: 1727,
2851, 2919, 3447. EI-MS (m/z): 341 (M+). HRMS Calcd for
C19H35NO4: 341.2566, Found: 341.2573, [α]2D3 = −8.3 (c 0.33,
CHCl3).
Typical procedure for hydrolysis
NaOH (280 mg, 7.00 mmol) was added to a solution of diol 4 (Genzyme; Boston, MA) as the enzyme source and 4-methyl-
(230 mg, 0.674 mmol) in EtOH/H2O (2 : 1, 8.4 mL), and the umbelliferyl-β-D-glucopyranoside (Sigma-Aldrich Co; St Louis,
mixture was refluxed for 1 h. The reaction mixture was cooled Mo, USA) as substrate. The reaction mixture consisted 100 mM
to room temperature and concentrated. The product was puri- McIlvaine buffer (pH 5.2), 0.25% sodium taurocholate and
fied by flash chromatography (silica gel, eluting with Et2O/ 0.1% Triton X-100 (Nacalai Tesque Inc; Kyoto, Japan), and the
MeOH/25% NH3(aq) = 80/20/1 → 0/100/2) to yield iminosugars appropriate amount of enzyme. The reaction mixture was pre-
1
5a–j. Compounds 5a–h were identical in H-NMR with ent-5a– incubated at 0 °C for 45 min, and the reaction was started by
h which were reported previously.26
the using 3 mM substrate solution, followed by incubation at
α-1-C-Butyl-DAB (5a). 90% yield; [α]2D4 = +47.2 (c 0.80, 37 °C for 30 min. The reaction was stopped by the addition of
CH3OH). Lit. ent-5a26: [α]D27 = −47.7 (c 1.00, CH3OH).
1.6 mL of the solution of 400 mM glycine–NaOH solution (pH
α-1-C-Pentyl-DAB (5b). 47% yield; [α]2D7 = +54.3 (c 0.82, 10.6). The released 4-methylumbelliferone was measured (exci-
CH3OH). Lit. ent-5b26: [α]D24 = −53.5 (c 1.44, CH3OH).
tation 362 nm, emission 450 nm) with a F-4500 fluorescence
α-1-C-Hexyl-DAB (5c). 58% yield, [α]2D6 = +66.6 (c 0.32, spectrophotometer (Hitachi, Tokyo, Japan). Kinetic parameters
CH3OH). Lit. ent-5c26: [α]D22 = −51.1 (c 1.18, CH3OH).
were determined by the double-reciprocal plot method of Line-
α-1-C-Heptyl-DAB (5d). 67% yield, [α]2D7 = +42.9 (c 1.33, weaver–Burk plots at increasing substrate concentrations.
CH3OH). Lit. ent-5d26: [α]D24 = −46.1 (c 1.00, CH3OH).
Thermostability of β-glucocerebrosidase
α-1-C-Octyl-DAB (5e). 46% yield, [α]2D6 = +34.6 (c 0.17,
CH3OH). Lit. ent-5e26: [α]D23 = −44.2 (c 0.70, CH3OH).
The enzyme β-glucocerebrosidase was incubated at 45 °C in
α-1-C-Nonyl-DAB (5f). 47% yield, [α]2D7 = +41.2 (c 0.50, 100 mM McIlvaine buffer (pH 5.2) containing 0.25% sodium
CH3OH). Lit. ent-5f26: [α]D23 = −45.5 (c 1.33, CH3OH).
α-1-C-Decyl-DAB (5g). 68% yield, [α]2D6 = +36.1 (c 0.99, After incubation, the remaining β-glucocerebrosidase activity
CH3OH). Lit. ent-5g26: [α]D22 = −36.5 (c 0.41, CH3OH).
was assayed immediately using 3 mM 4-methylumbelliferyl-
taurocholate and 0.1% Triton X-100 for 0, 20, 40, and 60 min.
α-1-C-Undecyl-DAB (5h). 65% yield, [α]2D7 = +35.7 (c 0.72, β-D-glucopyranoside as substrate. The values were expressed as
CH3OH). Lit. ent-5h26: [α]D23 = −37.8 (c 1.00, CH3OH).
relative to the activity before the incubation (100%). Time zero
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
points correspond to the conditions that reaction mixtures the binding site around a ligand suggested that the side chain
were not heat-treated and kept on ice for the duration of the of Asn396 is flexible. We therefore employed Induced Fit proto-
heat inactivation experiment.
col (Schrödinger Suite 2009, Schrödinger, LLC) that allows
both structures of a ligand and a protein to move.31 Water
molecules were removed from the crystal structure and hydro-
Pharmacological chaperone activity
The Gaucher’s cell line with β-glucocerebrosidase mutation of gen atoms were added. The 3D structures of α-1-C-tridecyl-DAB
N307S (GM00372) was obtained from the Coriell Cell Reposi- (5j) were generated using LigPrep and ConfGen (Schrodinger
tories (Camden, NJ). The Gaucher N307S fibroblasts were cul- suite 2009). Standard-precision mode was used in the docking
tured in EMEM (Eagle’s Minimum Essential Medium; Sigma- analysis. For a first docking step of the Induced Fit protocol, the
Aldrich Co) supplemented with 15% FCS. Cells were cultured van der Waals radii scaling for the ligand and the protein were
in a water-jacket incubator at 37 °C under 5% CO2 in the pres- reduced to 0.5 and 0.7 respectively. Asn396 was mutated to
ence or absence of samples for 6 days. After washing twice alanine temporarily. In a second step, Ala396 was restored to
with a medium, the cell pellet were homogenized in a citrate Asn396 and residues within 5 Å around a docked ligand were
buffer (pH 5.2) containing 0.25% sodium taurochole and 0.1% refined so that their side chains became optimal. In a final step,
Triton X-100. The supernatant obtained from the homogenate the ligand was docked again into induced-fit receptor structures
after centrifugation at 1000g for 5 min was subject to enzyme and the output poses were ranked based on IFD score. The top1
assays and protein determination. Intercellular β-glucocerebro- pose ranked by IFDscore was selected as the complex structure
sidase activities were determined with 4-methylumbelliferyl- of α-1-C-tridecyl-DAB with β-glucocerebrosidase.
β-D-glucoside as substrate at pH 5.2.
Immunocytochemistry
Acknowledgements
The Gaucher’s cell line with β-glucocerebrosidase mutation of
This work was supported in part by a Grant-in-Aid for Scienti-
N307S (GM00372) were treated with rabbit polyclonal anti-
fic Research (C) (no: 26460143) (AK) from the Japanese Society
GCase serum as a primary antibody for the detection of GCase
for the Promotion of Science (JSPS) and a Leverhulme research
(GeneTex, San Francisco, CA), and monoclonal anti-LAMP-1
fellowship (GWJF).
serum (abcam plc., Cambridge, UK) for the detection of lyso-
somes. Hilyte Fluor™ 488-conjugated goat anti-rabbit IgG
serum and Fluor™ 594-conjugated goat anti-mouse IgG serum
(both from AnaSpec, Inc., Fremont, CA) were used as second-
References
ary antibodies. Gaucher fibroblasts (2.5 × 104 cells) were grown
on sterile coverslips with presence or absence of iminosugars
and cultured with 6 days in EMEM. All immunocytochemistry
procedures were performed at room temperature. After cells
had been washed three times with PBS, they were fixed with
1 mL of 4% paraformaldehyde in PBS for 15 min, and then
washed three times with PBS. The cells were permeated with
1 mL of 0.01% Triton X-100 in PBS for 15 min, and then
washed three times with PBS. After a 30 min treatment with
1 mL of blocking solution (5% skim milk in PBS), the cells
were incubated with the primary antibody diluted in 1% skim
milk in PBS for 30 min, and this was followed by three washes
with 1 mL of PBS. Coverslips were then incubated in the dark
with the secondary antibody diluted in 1% skim milk in PBS
for 30 min, and this was followed by three washes with 1 mL
of PBS. The coverslips were mounted with a drop of Vecta-
shield mounting medium (Vector Laboratories, Inc., Burlin-
game, CA), and fluorescence was visualized using an Olympus
fluorescence microscope (Olympus, Tokyo, Japan).
1 R. O. Brady, J. N. Kanfer and D. Shapiro, Biochem. Biophys.
Res. Commun., 1965, 18, 221–225.
2 T. D. Butters, Curr. Opin. Chem. Biol., 2007, 11, 412–418.
3 M. Horowitz and A. Zimran, Hum. Mutat., 1994, 3, 1–11.
4 M. Horowitz, S. Wilder, Z. Horowitz, O. Reiner, T. Gelbart
and E. Beutler, Genomics, 1989, 4, 87.
5 E. Sidransky, Mol. Genet. Metab., 2004, 83, 6–15.
6 R. R. Wei, H. Hughes, S. Boucher, J. J. Bird, N. Guziewicz,
S. M. Van Patten, H. Qiu, C. Q. Pan and T. Edmunds,
J. Biol. Chem., 2011, 286, 299–308.
7 S. Ishii, R. Kase, H. Sakuraba and Y. Suzuki, Biochem.
Biophys. Res. Commun., 1993, 197, 1585–1589.
8 R. Kase, U. Bierfreund, A. Klein, T. Kolter, K. Utsumi,
K. Itoh, K. Sandhoff and H. Sakuraba, Biochim. Biopys. Acta,
2000, 1501, 227–235.
9 A. R. Sawkar, S. L. Adamski-Werner, W.-C. Cheng,
C.-H. Wong, E. Beutler, K.-P. Zimmer and J. W. Kelly, Chem.
Biol., 2005, 12, 1235–1244.
10 J.-Q. Fan, S. Ishii, N. Asano and Y. Suzuki, Nat. Med., 1999,
5, 112–115.
11 A. R. Sawker, W.-C. Cheng, E. Beautler, C.-H. Wong,
W. E. Baich and J. W. Kelly, Proc. Natl. Acad. Sci. U. S. A.,
2002, 99, 15428–15433.
Docking studies
The crystal structure of human β-glucocerebrosidase complex
with D-isofagomine (PDB code: 2NSX) was used for the
docking analysis of α-1-C-tridecyl-DAB (5j); this crystal struc- 12 G. M. Pastores, N. L. Barnett and E. H. Kolodny, Clin. Ther.,
ture was determined under acidic conditions (pH 4.5) reflect- 2005, 27, 1215–1227.
ing the lysosome acidic environment.30 The comparison 13 H. R. Mellor, F. M. Platt, R. A. Dwek and T. D. Butters,
between the B chain and D chain in the crystal structure about
Biochem. J., 2003, 374, 307–314.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
14 R. E. Boyd, G. Lee, P. Rybczynski, E. R. Benjamin, 24 B. J. Ayers, N. Ngo, S. F. Jenkinson, R. F. Martines,
R. Khanna, B. A. Wustman and K. J. Valenzano, J. Med.
Chem., 2013, 56, 2705–2725.
Y. Shimada, I. Adachi, A. C. Weymouth-Wilson, A. Kato and
G. W. J. Fleet, J. Org. Chem., 2012, 77, 7777–7792.
15 X. Zhu, K. A. Sheth, S. Li, H. H. Chang and J.-Q. Fan, 25 Y. Natori, T. Imahori, K. Murakami, Y. Yoshimura,
Angew. Chem., Int. Ed., 2005, 44, 7450–7453.
16 L. Yu, K. Ikeda, A. Kato, I. Adachi, G. Godin, P. Compain,
S. Nakagawa, A. Kato, I. Adachi and H. Takahata, Bioorg.
Med. Chem. Lett., 2011, 21, 738–741.
O. Maryin and N. Asano, Bioorg. Med. Chem., 2006, 14, 26 A. Kato, E. Hayashi, S. Miyauchi, I. Adachi, T. Imahori,
7736–7744.
Y. Natori, Y. Yoshimura, R. J. Nash, H. Shimaoka,
I. Nakagome, J. Koseki, S. Hirono and H. Takahata, J. Med.
Chem., 2012, 55, 10347–10362.
17 M. Aguilar-Moncayo, M. I. García-Moreno, A. Trapero,
M. Egido-Gabás, A. Llebaria, J. M. García Fernández and
C. Ortiz Mellet, Org. Biomol. Chem., 2011, 9, 3698–3713.
18 H.-H. Chang, N. Asano, S. Ishii, Y. Ichikawa and J.-Q. Fan,
FEBS J., 2006, 273, 4082–4092.
27 S. F. Jenkinson, D. Best, A. W. Saville, R. F. Martinez,
S. Nakagawa, T. Kunimatsu, D. S. Alonzi, T. D. Butters,
C. Norez, F. Becq, Y. Blériot, F. X. Wilson, A. C. Weymouth-
Wilson, A. Kato and G. W. J. Fleet, J. Org. Chem., 2013, 78,
7380–7397.
19 Z. Yu, A. R. Sawkar, L. J. Whalen, C. H. Wong and
J. W. Kelly, J. Med. Chem., 2007, 50, 94–100.
20 P. Compain, O. R. Martin, C. Boucheron, G. Godin, L. Yu, 28 A. Kato, I. Nakagome, S. Nakagawa, Y. Koike, R. J. Nash,
K. Ikeda and N. Asano, ChemBioChem, 2006, 7, 1356–1359.
21 F. Oulaidi, S. Front-Deschamps, E. Gallienne, E. Lesellier,
I. Adachi and S. Hirono, Bioorg. Med. Chem., 2014, 22,
2435–2441.
K. Ikeda, N. Asano, P. Compain and O. R. Maryin, ChemBio- 29 M. Kessler, O. Acuto, C. Strelli, H. Murer and
Chem, 2011, 6, 353–361.
G. A. A. Semenza, Biochem. Biophys. Acta, 1978, 506, 136–
22 E. D. Goddard-Borger, M. B. Tropak, S. Yonekawa, C. Tysoe,
154.
D. J. Mahuran and S. G. Withers, J. Med. Chem., 2012, 55, 30 R. L. Lieberman, B. A. Wustman, P. Huertas, A. C. Powe Jr.,
2737–2745.
C. W. Pine, R. Khanna, M. G. Schlossmacher, D. Ringe and
23 Y. Minami, C. Kuriyama, K. Ikeda, A. Kato, K. Takebayashi,
G. A. Petsko, Nat. Chem. Biol., 2007, 3, 101–107.
I. Adachi, G. W. J. Fleet, A. Kettawan, T. Okamoto and 31 W. Sherman, T. Day, M. P. Jacobson, R. A. Friesner and
N. Asano, Bioorg. Med. Chem., 2008, 16, 2734–2740.
R. Farid, J. Med. Chem., 2006, 49, 534–553.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015