α-1-C-Octyl-1-deoxynojirimycin as a pharmacological chaperone for Gaucher disease

Article abstract of DOI:10.1016/j.bmc.2006.08.003

The most common lysosomal storage disorder, Gaucher disease, is caused by inefficient folding and trafficking of certain variants of lysosomal β-glucosidase (β-Glu, also known as β-glucocerebrosidase). Recently, Sawker et al. reported that the addition of subinhibitory concentrations (10 μM) of the pharmacological chaperone N-nonyl-1-deoxynojirimycin (NN-DNJ) (10) to Gaucher patient-derived cells leads to a 2-fold increase in activity of mutant (N370S) enzyme [Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15428]. However, we found that the addition of NN-DNJ at 10 μM lowered the lysosomal α-glucosidase (α-Glu) activity by 50% throughout the assay period in spite of the excellent chaperoning activity in N370S fibroblasts. Hence, we prepared a series of DNJ derivatives with an alkyl chain at the C-1α position and evaluated their in vitro inhibitory activity and potential as pharmacological chaperones for Gaucher cell lines. Among them, α-1-C-octyl-DNJ (CO-DNJ) (15) showed 460-fold stronger in vitro inhibitory activity than DNJ toward β-Glu, while NN-DNJ enhanced in vitro inhibitory activity by 360-fold. Treatment with CO-DNJ (20 μM) for 4 days maximally increased intracellular β-Glu activity by 1.7-fold in Gaucher N370 cell line (GM0037) and by 2.0-fold in another N370 cell line (GM00852). The addition of 20 μM CO-DNJ to the N370S (GM00372) culture medium for 10 days led to 1.9-fold increase in the β-Glu activity without affecting the intracellular α-Glu activity for 10 days. Only CO-DNJ showed a weak β-Glu chaperoning activity in the L444P type 2 variant, with 1.2-fold increase at 5-20 μM, and furthermore maximally increased the α-Glu activity by 1.3-fold at 20 μM. These experimental results suggest that CO-DNJ is a significant pharmacological chaperone for N370S Gaucher variants, minimizing the potential for undesirable side effects such as α-Glu inhibition.

Full text of DOI:10.1016/j.bmc.2006.08.003

Bioorganic & Medicinal Chemistry 14 (2006) 7736–7744
a-1-C-Octyl-1-deoxynojirimycin as a pharmacological chaperone
for Gaucher disease
Liang Yu,a Kyoko Ikeda,a Atsushi Kato,b Isao Adachi,b Guillaume Godin,c
Philippe Compain,c Olivier Martinc and Naoki Asanoa,*
aFaculty of Pharmaceutical Sciences, Hokuriku University, Ho-3 Kanagawa-machi, Kanazawa 920-1181, Japan
bDepartment of Hospital Pharmacy, Toyama Medical and Pharmaceutical University, Toyama 930-0194, Japan
cInstitut de Chimie Organique et Analytique, Universite´ d’Orle´ans, CNRS UMR 6005, 45067 Orle´ans, France
Received 28 June 2006; revised 3 August 2006; accepted 4 August 2006
Available online 21 August 2006
Abstract—The most common lysosomal storage disorder, Gaucher disease, is caused by inefficient folding and trafficking of certain
variants of lysosomal b-glucosidase (b-Glu, also known as b-glucocerebrosidase). Recently, Sawker et al. reported that the addition
of subinhibitory concentrations (10 lM) of the pharmacological chaperone N-nonyl-1-deoxynojirimycin (NN-DNJ) (10) to Gaucher
patient-derived cells leads to a 2-fold increase in activity of mutant (N370S) enzyme [Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15428].
However, we found that the addition of NN-DNJ at 10 lM lowered the lysosomal a-glucosidase (a-Glu) activity by 50% throughout
the assay period in spite of the excellent chaperoning activity in N370S fibroblasts. Hence, we prepared a series of DNJ derivatives
with an alkyl chain at the C-1a position and evaluated their in vitro inhibitory activity and potential as pharmacological chaperones
for Gaucher cell lines. Among them, a-1-C-octyl-DNJ (CO-DNJ) (15) showed 460-fold stronger in vitro inhibitory activity than
DNJ toward b-Glu, while NN-DNJ enhanced in vitro inhibitory activity by 360-fold. Treatment with CO-DNJ (20 lM) for 4 days
maximally increased intracellular b-Glu activity by 1.7-fold in Gaucher N370 cell line (GM0037) and by 2.0-fold in another N370
cell line (GM00852). The addition of 20 lM CO-DNJ to the N370S (GM00372) culture medium for 10 days led to 1.9-fold increase
in the b-Glu activity without affecting the intracellular a-Glu activity for 10 days. Only CO-DNJ showed a weak b-Glu chaperoning
activity in the L444P type 2 variant, with 1.2-fold increase at 5–20 lM, and furthermore maximally increased the a-Glu activity by
1.3-fold at 20 lM. These experimental results suggest that CO-DNJ is a significant pharmacological chaperone for N370S Gaucher
variants, minimizing the potential for undesirable side effects such as a-Glu inhibition.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
enzyme replacement therapy (ERT) for patients with
type 1 Gaucher disease. Ceredase in 1991 and its recom-
binant successor Cerezyme in 1994 were introduced as
ERT of this type. However, ERT is not available for
types 2 and 3 of Gaucher disease since proteins do not
cross the blood–brain barrier, and another problem in
this therapy is the cost, which prevents many patients
from obtaining this treatment.
Gaucher disease is the most common lysosomal storage
disorder caused by a deficiency of lysosomal b-glucosi-
dase (b-Glu, known as b-glucocerebrosidase), resulting
in the progressive accumulation of glucosylceramide.
Clinically, Gaucher disease is classified into three major
types based on the absence (type 1) or the presence and
severity (types 2 and 3) of central nervous system (CNS)
involvement. Type 1 is nonneuronopathic and some-
times called the ‘adult’ form. Type 2 is the infantile form
of the disorder with severe CNS involvement, and type 3
is subacute neuronopathic with typically childhood or
early adult onset.1 The first successful treatment is the
In recent years, remarkable progress has been made in
developing a molecular therapy for the glycosphingolip-
id storage disorders.2–5 There are two novel approaches
in this field. One is substrate reduction therapy (SRT)
and another is pharmacological (or chemical) chaperone
therapy (PCT). The basic concept of SRT is to reduce
the substrate influx into the lysosomes by inhibitors of
glycosphingolipid synthesis. N-Butyl-1-deoxynojirimy-
cin (NB-DNJ, miglustat, ZavescaTM) (7) is an inhibitor
of ceramide-specific glucosyltransferase6 and has been
Keywords: Gaucher disease; Pharmacological chaperone; b-Glucocer-
ebrosidase inhibitor.
*
Corresponding author. Fax: +81 76 229 2781; e-mail: n-asano
@hokuriku-u.ac.jp
0968-0896/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.08.003

L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
7737
approved for the oral treatment of type 1 Gaucher dis-
ease. The concept of PCT is that an intracellular activity
of a misfolded mutant enzyme can be restored by
administering a competitive inhibitor that serves as a
pharmacological chaperone. Such inhibitor appears to
act as a template that stabilizes the native folding state
in the endoplasmic reticulum (ER) by occupying the ac-
tive site of the mutant enzyme, thus allowing its matura-
tion and trafficking to the lysosome.3 The concept of
PCT was originally introduced with Fabry disease.7
Residual a-galactosidase A (a-Gal A) activity in lym-
phoblasts derived from Fabry patients and in tissues
of R301Q a-Gal A transgenic mice was enhanced by
treatment with 1-deoxygalactonojirimycin (DGJ), a
competitive inhibitor of a-Gal A with a Ki value of
40 nM. Very recently, Yam et al. showed that DGJ
induces trafficking of ER-retained R301Q a-Gal A to
lysosomes of transgenic mouse fibroblasts and that
DGJ treatment results in efficient clearance of the sub-
strate, globotriaosylceramide (Gb3).8 Sawker et al.
reported that the addition of N-nonyl-DNJ (NN-DNJ)
(10) to the fibroblast culture medium of the fibroblasts
derived from N370S Gaucher patients led to a 2-fold in-
crease in the b-Glu activity.9
must be carefully considered in terms of side-effects since
they may inhibit folding, secretion, and trafficking of
other glycoproteins in patient’s cells or may inhibit
directly lysosomal a-glucosidase (a-Glu) after being tak-
en up into cells. Although NN-DNJ has been shown to
increase the trafficking and intracellular activity of
N370S mutant b-Glu,9 there is no statement on the
influence on other glycoproteins, especially for a-Glu.
Recently, examination of a series of DNJ analogues
on the residual activities of various b-Glu variants has
revealed that the nature of the alkyl moiety greatly influ-
ences their chaperoning activity: NB-DNJ is inactive,
DNJ derivatives with N-nonyl and N-decyl chains are
active, and N-dodecyl-DNJ is predominantly inhibito-
ry.20 Therefore, we prepared a series of N-alkyl and
a-1-C-alkyl-DNJ derivatives in order to assess their
b-Glu chaperoning activity and cellular a-Glu inhibitory
activity in Gaucher cell lines. Herein, we report that
a-1-C-octyl-DNJ (CO-DNJ) (15) is an excellent pharma-
cological chaperone increasing N370S b-Glu activity
without inhibiting intracellular a-Glu activity.
2. Results
Lysosomal glycosidases are glycoproteins with the
N-linked oligosaccharide chains. One of the functions
of N-glycans is to assist in the folding of glycoproteins
by mediating interactions of the lectin-like chaperone
calnexin with nascent glycoproteins.10,11 Glucose trim-
ming, catalyzed by ER a-glucosidases I and II, begins
on the nascent sugar chain and its inhibition by the
inhibitors, such as castanospermine (CST) and NB-
DNJ, can have significant biological consequences be-
cause of the blockage of the normal interaction with
calnexin.12 For example, treatment of cultured cells with
CST interferes with folding, secretion, and proper local-
ization of some glycoproteins, and stimulates their rapid
degradation.12–15 Tyrosinase, a key enzyme in melanin
biosynthesis, was catalytically inactive when synthesized
in the presence of NB-DNJ but was transported correct-
ly to the melanosome.16 However, inhibition of glucose
trimming by ER a-glucosidase inhibitors does not neces-
sarily lead to the impairment in glycoprotein expression.
When the murine macrophage cells, expressing signifi-
cant levels of functional class A scavenger receptors,
are cultured in the presence of NB-DNJ, the glucose
trimming inhibition does not result in gross misfolding
and degradation of this receptor or prevent its transport
to the cell surface, following significantly enhanced lev-
els of expression.17 Thus, it is not possible to predict the
precise effects of inhibiting N-glycan processing, and the
effects of ER processing a-glucosidase inhibitors on gly-
coproteins should be considered on a case-by-case.
2.1. In vitro inhibition of b-Glu and a-Glu
Naturally occurring iminosugars DNJ (1) and N-meth-
yl-DNJ (5) were isolated from Morus alba (Moraceae),21
a-homonojirimycin (a-HNJ, 2) and b-homonojirimycin
(b-HNJ, 3) from Stemona tuberosa (Stemonaceae),22
and a-1-C-methyl-DNJ (11, a-7-deoxyhomonojirimy-
cin) from Scilla sibirica (Hyacinthaceae)23 according to
the literature (Fig. 1). The N-alkyl derivatives of DNJ
were prepared by treatment with the appropriate alkyl
bromide and K2CO3 in DMF, and the a-1-C-alkyl-
DNJ derivatives were prepared according to the synthet-
ic route using a cross-metathesis reaction as the key step
reported previously.24–26 The b-Glu activity was mea-
sured with commercially available b-glucocerebrosidase
(Ceredase) as the enzyme source and 4-methylumbellife-
ryl (4-MU) b-D-glucoside as the substrate. The a-Glu
activity was assayed with the cell lysate of normal hu-
man fibroblasts (GM05659) as an enzyme source and
4-MU a-glucoside as substrate at pH 4.5. The IC50 val-
ues of iminosugars and their alkyl derivatives toward
b-Glu and a-Glu are shown in Table 1. Natural imi-
nosugars DNJ, N-methyl-DNJ, a-HNJ, b-HNJ, and a-
1-C-methyl-DNJ showed weak or null in vitro inhibition
of b-Glu. Introduction of the butyl group to the C-1b
position of DNJ abolished its inhibition toward b-Glu,
whereas introducing this group to the C-1a position
slightly improved its inhibitory activity. The DNJ deriv-
atives with a alkyl chain longer than the butyl group at
the C-1a or the imino group remarkably enhanced the
b-Glu inhibitory activity with increasing chain length,
whereas a-Glu was less sensitive to inhibition by the
DNJ derivatives with the elongated chain than b-Glu.
N-Octyl-DNJ (NO-DNJ) (9) and NN-DNJ were 290-
fold and 360-fold stronger inhibitors of b-Glu than
DNJ, respectively, and CO-DNJ and a-1-C-nonyl-
DNJ (CN-DNJ) (16) showed 460-fold and 890-fold
stronger inhibition toward b-Glu than DNJ.
NN-DNJ has been reported to be an effective pharmaco-
logical chaperone to increase the activity of mutant
b-Glu in Gaucher cells,9 while it is also known that this
compound, similarly to NB-DNJ, is a potent inhibitor
of ER processing a-glucosidases, and hence has poten-
tial as anti-viral agents to inhibit folding and trafficking
of viral envelope glycoproteins.18,19 Inhibitors targeting
a host function such as ER processing a-glucosidases

7738
L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
Figure 1. Structures of iminosugars and their alkyl derivatives. 1, 1-deoxynojirimycin (DNJ); 2, a-homonojirimycin (a-HNJ); 3, b-HNJ; 4, b-1-C-
butyl-DNJ; 5, N-methyl-DNJ; 6, N-propyl-DNJ; 7, N-butyl-DNJ (NB-DNJ); 8, N-hexyl-DNJ; 9, N-octyl-DNJ (NO-DNJ); 10, N-nonyl-DNJ
(NN-DNJ); 11, a-1-C-methyl-DNJ; 12, a-1-C-propyl-DNJ; 13, a-1-C-butyl-DNJ; 14, a-1-C-hexyl-DNJ; 15, a-1-C-octyl-DNJ (CO-DNJ); 16, a-1-C-
nonyl-DNJ (CN-DNJ).
comparison, those for NO-DNJ and NN-DNJ are 0.42
and 0.30 lM, respectively.
Table 1. Inhibition of lysosomal b-glucosidase (b-glucocerebrosidase)
and a-glucosidase by 1-deoxynojirimycin (DNJ) derivatives
2.2. Effects of DNJ derivatives on a-Glu and b-Glu
activities in Gaucher fibroblasts
Compound
IC50 (lM)
b-Glucosidase a-Glucosidase
DNJ (1)
240
—
1.0
1.0
With respect to the compounds showing potent in vitro
inhibitory activity toward b-Glu, we investigated the
influence of selected DNJ derivatives on b-Glu and
a-Glu activities in Gaucher fibroblasts cultured for 4
days. Treatment with NO-DNJ and NN-DNJ caused
2.3-fold maximal increase at 50 lM (Fig. 2C) and 2.4-
fold maximal increase at 10 lM (Fig. 2D), respectively,
in the b-Glu activity of an N370S cell line (GM00372),
while CO-DNJ and CN-DNJ maximally increased the
activity by 1.7-fold at 20 lM (Fig. 2A) and 1.7-fold at
2.5 lM (Fig. 2B), respectively. Although CN-DNJ
showed a chaperoning activity at low concentrations,
it was inhibitory at concentrations higher than 10 lM.
It should be noted that addition of NO-DNJ at only
5 lM led to 50% inhibition of the a-Glu activity, and
that NN-DNJ inhibited a-Glu in a dose-dependent man-
ner and decreased the activity by 70% at 50 lM. On the
other hand, the a-1-C-alkyl derivatives slightly inhibited
a-Glu activity at concentrations higher than 10 lM.
Another N370S cell line (GM00852) was less sensitive
to the N-alkyl-DNJ derivatives and CN-DNJ than
GM00372 cell line (Figs. 3B–D). CO-DNJ maximally in-
creased b-Glu activity in this cell line by 2.0-fold at
20 lM without inhibiting a-Glu activity, rather a slight
enhancement of a-Glu activity was observed over a con-
centration range of 10–20 lM (Fig. 3A). NO-DNJ and
NN-DNJ dose-dependently lowered a-Glu activity, with
30% and 50% inhibition at 50 lM, respectively.
a-Homonojirimycin (2)
b-Homonojirimycin (3)
b-1-C-butyl-DNJ (4)
N-Methyl-DNJ (5)
—
—
—
—
150
700
270
13
1.0
N-Propyl-DNJ (6)
N-Butyl-DNJ (7)
N-Hexyl-DNJ (8)
56
5.0
4.0
2.1
1.5
N-Octyl-DNJ (NO-DNJ, 9)
N-Nonyl-DNJ (NN-DNJ, 10)
a-1-C-Methyl-DNJ (11)
a-1-C-Propyl-DNJ (12)
a-1-C-Butyl-DNJ (13)
a-1-C-Hexyl-DNJ (14)
a-1-C-Octyl-DNJ (CO-DNJ, 15)
a-1-C-Nonyl-DNJ (CN-DNJ, 16)
0.82
0.66
900
28
400
100
270
31
4.2
12
0.50
0.27
5.0
4.8
‘—,’ less than 50% inhibition at 1000 lM.
DGJ and NN-DNJ are competitive inhibitors and ac-
tive-site-specific chaperones.7,9 We previously reported
that a potent inhibitor shows an effective chaperoning
activity and less potent inhibitors require higher concen-
trations to achieve the same effect.27 Hence, the mode of
inhibition and inhibition constant of the DNJ deriva-
tives were determined by the Lineweaver–Burk plots
(Table 2). All of DNJ and its alkyl derivatives inhibited
b-Glu in a competitive manner. The N-butylation and a-
1-C-butylation of DNJ decreased the affinity for b-Glu,
but the introduction of an alkyl chain longer than the
butyl group remarkably enhanced the affinity. Addition
of the alkyl chain to the C-1a position (pseudo-anomeric
position) showed a slightly better affinity than that to
the imino group. The Ki values for CO-DNJ and CN-
DNJ are 0.28 and 0.20 lM, respectively. For a
The b-Glu and a-Glu activities of an L444P variant,
which characterize a more severe disease phenotype than
N370S, were measured using Gaucher disease type 2
patient-derived cell line (GM00877) after incubation
with the N-alkyl- and a-1-C-alkyl-DNJ derivatives for

L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
7739
Table 2. Measured Ki values and inhibition mode of human lysosomal
b-glucosidase
whereas NO-DNJ and NN-DNJ dose-dependently
inhibited the cellular a-Glu activity, with 40% and
70% inhibition at 50 lM, respectively (Figs. 4C and D).
Compound
Kia (lM)
Inhibitionb
DNJ (1)
N-Butyl-DNJ (7)
N-Hexyl-DNJ (8)
79
116
5.5
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
We selected CO-DNJ as the most promising pharmaco-
logical chaperone that did not affect the cellular a-Glu
activity in Gaucher cell lines. Longer time-course study
using N370S cell line (GM00372) was conducted for
CO-DNJ and NN-DNJ. Each compound was added
to the culture medium at concentrations (20 lM for
CO-DNJ and 10 lM for NN-DNJ) causing the maximal
increase in the cellular b-Glu activity and incubated for
10 days. The b-Glu activity was increased 1.9-fold by
CO-DNJ and 2.3-fold by NN-DNJ in a time-dependent
manner up to day 10 (Fig. 5). The cellular a-Glu activity
remained quite unaffected throughout the assay period
in the presence of CO-DNJ (20 lM), whereas addition
of NN-DNJ (10 lM) lowered the activity by 50% on
day 2 and the inhibition levels remained constant for
the duration of this experiment (Fig. 5B).
N-Octyl-DNJ (NO-DNJ, 9)
N-Nonyl-DNJ (NN-DNJ, 10)
a-1-C-Butyl-DNJ (13)
a-1-C-Hexyl-DNJ (14)
a-1-C-Octyl-DNJ (CO-DNJ, 15)
a-1-C-Nonyl-DNJ (CN-DNJ, 16)
0.42
0.30
110
2.3
0.28
0.20
a,bInhibition constant (Ki) and inhibition mode were determined by the
Lineweaver–Burk plots.
4 days. None of candidate compounds tested to date
shows effective b-Glu chaperoning activity for L444P
mutation.9,20,28 It has been reported that L444P b-Glu
is not chaperoned by any of the DNJ analogues and is
at least 50% inhibited at concentrations higher than
25 lM.20 In the present study, none of the N-alkyl-
and a-1-C-alkyl-DNJ derivatives likewise showed
distinct chaperoning activity in L444P cell line. In par-
ticular, the nonyl-DNJ derivatives weakly inhibited the
cellular b-Glu activity at concentrations exceeding
5 lM (Figs. 4B and D). Interestingly, CO-DNJ showed
a weak chaperoning activity toward both a-Glu and
b-Glu over the concentration range assayed (Fig. 4A),
NN-DNJ shows a better b-Glu chaperoning activity
than CO-DNJ in Gaucher cell lines but significantly im-
pairs the cellular a-Glu activity. Further experimental
approaches are needed to ascertain whether the loss of
the cellular a-Glu activity is due to gross misfolding,
degradation, and trafficking impairment of a-Glu de-
rived from processing a-glucosidase inhibition by the
N-alkyl-DNJ derivatives, or due to direct inhibition of
3
3
A
B
1.7
1.7
1.7
2
1
0
2
1.6
1.6
1.3
1.6
1.0
1.5
1.2
1.0
1.1
1.0
1.0
1.0
1.0
0.8
0.9
1
0
0.6
0.7
1.1
2.5
1.0
1
0.9
20
0.8
50
0
5
10
0
5
10
20
50
CO-DNJ (μM)
CN-DNJ (μM)
3
2
1
0
3
2
1
0
C
D
2.4
2.3
2.1
2.1
1.9
1.8
1.8
1.6
1.6
1.5
1.0
1.0
1.0
1.0
0.9
1
0.7
2.5
0.6
5
0.6
0.6
20
0.5
5
0.5
10
0.5
50
0.4
20
0.3
50
0
10
0
NO-DNJ (μM)
NN-DNJ (μM)
Figure 2. The influence of the alkylated DNJ derivatives on lysosomal a-glucosidase (red line) and b-glucosidase (blue line) activities in N370S
Gaucher fibroblasts (GM00372). The fibroblasts were incubated in the presence of (A) CO-DNJ, (B) CN-DNJ, (C) NO-DNJ, or (D) NN-DNJ over a
concentration range of 0–50 lM for 4 days. Enzyme activity is normalized to untreated cells, assigned a relative activity of 1. Mean values SD are
shown for triplicate experiments.

7740
L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
3
2
1
0
3
A
B
2.0
1.8
1.7
2
1
0
1.4
1.4
1.3
1.0
5
1.3
1.0
1.1
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.8
0.8
1.2
1.3
20
1.0
50
1.0
1
0.9
2.5
0
10
0
5
10
20
50
CO-DNJ (μM)
CN-DNJ (μM)
3
2
1
0
3
2
1
0
C
D
2.1
2.0
2.0
1.8
1.6
1.4
1.5
1.4
1.3
1.0
1.0
1.0
1.0
1.0
0.9
0.8
0.8
0.8
1
0.7
50
0.8
2.5
0.8
5
0.7
0.6
20
0.5
50
0
5
10
20
0
10
NO-DNJ (μM)
NN-DNJ (μM)
Figure 3. The influence of the alkylated DNJ derivatives on lysosomal a-glucosidase (red line) and b-glucosidase (blue line) activities in N370S
Gaucher fibroblasts (GM00852). The fibroblasts were incubated in the presence of (A) CO-DNJ, (B) CN-DNJ, (C) NO-DNJ, or (D) NN-DNJ over a
concentration range of 0–50 lM for 4 days. Enzyme activity is normalized to untreated cells, assigned a relative activity of 1. Mean values SD are
shown for triplicate experiments.
a-Glu after the uptake of an inhibitor into the cells. If
this is due to inhibition of the early glucose-trimming
steps in N-glycan processing and ensuing prevention of
the interaction of glycoproteins with calnexin, addition
of N-alkyl-DNJ to the culture medium may influence
the activity of other lysosomal glycosidases. The DNJ
derivatives with the octyl and nonyl chains are inhibitors
of both a-Glu and b-Glu (Table 1) but not of lysosomal
a-mannosidase (a-Man) (data not shown). Hence, the
change in the a-Man activity was investigated using an
N370S cell line (GM00372) when inhibitors were added
to the culture medium over a concentration range of 0–
50 lM and incubated for 4 days. CO-DNJ and CN-DNJ
gave no effect on the a-Man activity over a concentra-
tion range assayed (Figs. 6A and B), whereas the
N-alkyl-DNJ derivatives lowered the activity in a dose-
dependent manner (Figs. 6C and D). This result may
suggest that the N-alkyl-DNJ derivatives inhibit the ear-
ly glucose-trimming steps in N-glycan synthesis and lead
to misfolding, degradation, and trafficking impairment
of normal lysosomal glycosidases, but the a-1-C-alkyl
DNJ derivatives do not.
to a 2-fold increase in activity of the mutant (N370S)
enzyme in the fibroblasts derived from Gaucher disease
patients.9 Furthermore, it has been revealed that the
nature of the N-alkyl moiety of the DNJ derivatives
and analogues influences the pharmacological chaperon-
ing activity in mutant b-Glu, that is, the short
N-alkylation (4 carbons) of DNJ produces inactive
compounds, the long N-alkyl chains (12 carbons)
proved to be toxic, and the DNJ derivatives with 8–10
carbon side chains are well tolerated and enhance b-
Glu activity in some Gaucher cell lines other than
L444P.9 We prepared the DNJ derivatives with a variety
of the alkyl chain length at the pseudo-anomeric posi-
tion and measured their inhibitory activities toward
a-Glu and b-Glu. With respect to selected potent inhib-
itors of b-Glu, we evaluated their b-Glu chaperoning
activity in Gaucher cell lines and the influence on the cel-
lular a-Glu activity. Introduction of the alkyl side chains
to the C-1b position and the short chains to C-1a result-
ed in weak or null inhibition of b-Glu, whereas elonga-
tion of the alkyl chain from hexyl to nonyl at C-1a
markedly enhanced the b-Glu activity with increasing
chain length. On the other hand, a-Glu was less sensitive
to inhibition by the DNJ derivative with elongated chain
length than b-Glu. N-Alkyl-DNJs showed a slightly bet-
ter b-Glu chaperoning activity than a-1-C-alkyl-DNJs in
Gaucher cell lines with N370S mutation but significantly
impaired the a-Glu activity in the cells. There was seen a
difference in chaperoning activity of pharmacological
chaperones in the cells derived from different Gaucher
3. Discussion
It was reported that NN-DNJ is a potent inhibitor of
lysosomal b-glucocerebrosidase (b-Glu) with an IC50
value of 1 lM, and addition of subinhibitory concentra-
tion (10 lM) of this inhibitor to a culture medium leads

L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
7741
2
1
0
2
1
0
2
A
B
1.2
1.2
1.2
1.1
1.1
1.1
1.2
1.1
1.1
1.0
1.0
1.0
1.0
1.0
0.9
1.0
0.7
1.0
0.8
1.2
1.3
1
0
2
1
0
0.7
0.7
1.1
0
5
10
20
50
0
1
2.5
5
10
20
50
CO-DNJ (μM)
CN-DNJ (μM)
C
D
1.2
1.1
0.8
1.1
0.7
1.0
0.8
1.1
1.0
1.0
1.0
1.0
1.0
0.9
0.9
0.8
0.7
0.9
0.7
0.6
5
0.6
0.6
50
0.5
20
0.3
50
0
5
10
20
0
1
2.5
10
NO-DNJ (μM)
NN-DNJ (μM)
Figure 4. The influence of the alkylated DNJ derivatives on lysosomal a-glucosidase (red line) and b-glucosidase (blue line) activities in L444P
Gaucher fibroblasts (GM00877). The fibroblasts were incubated in the presence of (A) CO-DNJ, (B) CN-DNJ, (C) NO-DNJ, or (D) NN-DNJ over a
concentration range of 0–50 lM for 4 days. Enzyme activity is normalized to untreated cells, assigned a relative activity of 1. Mean values SD are
shown for triplicate experiments.
3
2
1
0
3
2
1
0
A
B
CO-DNJ (20
μM)
NN-DNJ (10
μM)
2.3
2.2
2.0
1.7
1.9
1.1
1.8
1.7
1.7
1.4
1.0
1.3
1.0
1.0
1.0
0.5
2
0.5
8
0.5
10
0.5
0.4
0
2
4
6
8
10
0
4
6
Days
Figure 5. Time course of the lysosomal a-glucosidase (red line) and b-glucosidase (blue line) activities in N370S Gaucher fibroblasts (GM00372) by
treatment with a-1-C-octyl-DNJ (CO-DNJ) or N-nonyl-DNJ (NN-DNJ). The fibroblasts were incubated in the presence of (A) CO-DNJ (20 lM) or
(B) NN-DNJ (10 lM) for 10 days. Enzyme activity is normalized to untreated cells, assigned a relative activity of 1. Mean values SD are shown for
triplicate experiments.
patients with the same N370S mutation. CO-DNJ
showed 1.7-fold (GM00372) and 2.0-fold (GM00852)
maximal increase in the b-Glu activity at 20 lM, while
NN-DNJ maximally enhanced its activity by 2.4-fold
at 10 lM in GM00372 and by 1.6-fold at 5 lM in
GM00852. Addition of CO-DNJ at 20 lM to the

7742
L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
A
B
200
150
100
50
250
200
150
100
50
0
0
0
5
10
20
50
0
5
10
20
50
CO-DNJ (μM)
C
-DNJ (μM)
C
D
200
150
100
50
250
200
150
100
50
0
0
0
5
10
20
50
0
5
10
20
50
NO-DNJ (μM)
NN-DNJ (μM)
Figure 6. The influence of the alkylated DNJ derivatives on lysosomal a-mannosidase activity in N370S Gaucher fibroblasts (GM00372). The
fibroblasts were incubated in the presence of (A) CO-DNJ, (B) CN-DNJ, (C) NO-DNJ, or (D) NN-DNJ over a concentration range of 0–50 lM for 4
days. Mean values SD are shown for triplicate experiments.
GM00372 culture medium for 10 days enhanced the
b-Glu activity by 1.9-fold without inhibiting a-Glu,
whereas that of NN-DNJ at 10 lM lowered the cellular
a-Glu activity by 50% throughout the assay period (10
days) in spite of the excellent chaperoning activity.
The significant decrease of the a-Glu activity by the
N-alkyl-DNJ derivatives may be due to the inhibition
of the early glucose-trimming steps in N-glycan synthesis
since they are potent inhibitors of ER processing a-glu-
cosidases I and II, resulting in gross misfolding and deg-
radation, and subsequent mistrafficking to the
lysosomes. It has been reported that an L444P variant,
which characterizes a more severe disease phenotype
than N370S, is not chaperoned by any compounds test-
ed so far.9,20,28 However, only CO-DNJ showed a weak
b-Glu chaperoning activity in L444P and further en-
hanced the cellular a-Glu activity by 1.3-fold in the same
cell line.
4. Materials
4.1. Preparation of iminosugars and their alkyl derivatives
Natural DNJ (1) and N-methyl-DNJ (5) were isolated
from the root bark of Morus alba,21a-HNJ (2) and
b-HNJ (3) from Stemona tuberosa,22 and a-1-C-meth-
yl-DNJ (11) from Scilla sibirica23 according to the lit-
erature. The N-alkyl-DNJ derivatives N-propyl-DNJ
(6), N-butyl-DNJ (NB-DNJ, 7), N-hexyl-DNJ (8),
N-octyl-DNJ (NO-DNJ, 9), and N-nonyl-DNJ (NN-
DNJ, 10) were prepared as follows. Appropriate alkyl
bromide (20 lL) and K2CO3 (20 mg) were added to a
solution of DNJ (10 mg) in DMF. The reaction mix-
ture was heated at 60 °C for 24 h and filtered. The
filtrate was diluted with MeOH and applied to a
short column of Amberlyst 15 (5 mL), washed with
MeOH, and eluted with MeOH/28% aqueous ammo-
nia (4:1). The concentrated eluate was further chro-
matographed with Dowex 1-X2 (OHÀ form) with
water for 6, 7, and 8 or 70% aqueous MeOH for 9
and 10 as eluent. The 1-C-alkyl-DNJ derivatives,
b-1-C-butyl-DNJ (4), a-1-C-propyl-DNJ (12), a-1-C-
butyl-DNJ (13), a-1-C-hexyl-DNJ (14), a-1-C-octyl-
DNJ (CO-DNJ, 15), and a-1-C-nonyl-DNJ (CN-
DNJ, 16), were prepared according to the method
reported previously.24–26 The spectroscopic data of
4, 13, and 15 were superimposable with those report-
ed in the literature.24–26 The structures of 12, 14, and
16 were confirmed by 1H NMR, 13C NMR, and
COSY spectroscopic data.
Considering the side-effect profile, we selected CO-DNJ
as the most promising pharmacological chaperone to
bind the active site of the mutant b-Glu, stabilize it in
the ER, and enable proper trafficking to the lysosomes.
Enzyme replacement therapy (ERP) has been the most
successful therapeutic approach for lysosomal storage
disorders but does not effectively treat the neurological
complications because enzymes do not cross the blood-
brain barrier. Nonneuronopathic type 1 Gaucher patients
have ‘Cerezyme’ as ERP, whereas type 2 or 3 Gaucher pa-
tients have no effective therapy. The increase in mutant b-
Glu activity by CO-DNJ may be only double. However,
this increase would be very hopeful for type 2 or 3 Gauch-
er patients. We expect that a number of pharmacological
chaperones would be excellent candidates for a new
molecular therapy of human genetic disorders.
4.1.1. 1-Deoxynojirimycin (DNJ) (1). [a]D +40.3° (c 1.47,
H2O). HRMS (FAB): m/z 164.0923 [M+H]+ (C6H14NO4
requires 164.0923).

L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
7743
4.1.2. a-Homonojirimycin (a-HNJ) (2). [a]D +77.2° (c
0.57, H2O). HRMS (FAB): m/z 194.1026 [M+H]+
(C7H16NO5 requires 194.1028).
4.1.15. a-1-C-Octyl-DNJ (CO-DNJ, 15). [a]D +48.6° (c
0.52, MeOH). HRMS (FAB): m/z 276.2173 [M+H]+
(C14H30NO4 requires 276.2175).
4.1.3. b-Homonojirimycin (b-HNJ) (3). HRMS (FAB):
m/z 194.1022 [M+H]+ (C7H16NO5 requires 194.1028).
4.1.16. a-1-C-Nonyl-DNJ (CN-DNJ, 16). [a]D +68.2° (c
1
0.41, MeOH). H NMR (D2O): d 0.90 (t, 3 H), 1.28–
1.36 (13H), 1.43–1.54 (2H), 1.64 (m, 1H), 2.70 (ddd,
1H, J = 3.2, 7.8, 9.6 Hz, H-5), 3.00 (ddd, 1H, J = 3.2,
5.5, 11.0 Hz, H-1), 3.08 (dd, 1H, J = 8.7, 9.6 Hz, H-4),
3.42 (dd, 1H, J = 8.7, 9.6 Hz, H-3), 3.46 (dd, 1H,
J = 7.8, 11.0 Hz, H-6a), 3.58 (dd, 1H, J = 5.5, 9.6 Hz,
H-2), 3.88 (dd, 1H, J = 3.2, 11.0 Hz, H-6b). 13C NMR
(D2O): d 14.5, 23.8, 25.8, 27.4, 30.5, 30.7, 30.8, 30.9,
33.1, 56.2, 57.6, 63.9, 74.3, 74.4, 76.1. HRMS (FAB):
m/z 290.2335 [M+H]+ (C15H32NO4 requires 290.2331).
4.1.4. b-1-C-Butyl-DNJ (4). [a]D +2° (c 1.4, H2O).
HRMS (FAB): m/z 220.1549 [M+H]+ (C10H22NO4 re-
quires 220.1549).
4.1.5. N-Methyl-DNJ (5). [a]D +12.6° (c 1, H2O). HRMS
(FAB): m/z 178.1077 [M+H]+ (C7H16NO4 requires
178.1079).
4.1.6. N-Propyl-DNJ (6). [a]D À15.1° (c 1.57, H2O).
HRMS (FAB): m/z 206.1393 [M+H]+ (C9H20NO4 re-
quires 206.1392).
5. Experimental
4.1.7. N-Butyl-DNJ (NB-DNJ,7). [a]D À15.9° (c 0.77,
H2O). HRMS (FAB): m/z 220.1548 [M+H]+
(C10H22NO4 requires 220.1549).
5.1. General experimental procedures
The purity of samples was checked by HPTLC on Silica
Gel 60F254 (E. Merck) using the solvent systems PrOH/
AcOH/H2O (4:1:1) for free base compounds and BuOH/
AcOH/H2O (4:1:1) for their alkyl derivatives, and a
chlorine–o-tolidine reagent was used for detection. Opti-
cal rotations were measured with a Jasco DIP-370 digi-
4.1.8. N-Hexyl-DNJ (8). [a]D À15.5° (c 0.89, H2O).
HRMS (FAB): m/z 248.1860 [M+H]+ (C12H26NO4 re-
quires 248.1862).
1
4.1.9. N-Octyl-DNJ (NO-DNJ, 9). [a]D À17.4° (c 0.18,
MeOH). HRMS (FAB): m/z 276.2174 [M+H]+
(C14H30NO4 requires 276.2175).
tal polarimeter (Tokyo, Japan). H NMR (500 MHz)
and 13C NMR (125 MHz) spectra were recorded on a
JEOL ECP-500 spectrometer (Tokyo, Japan). Chemical
shifts are expressed in ppm downfield from sodium
3-(trimethylsilyl)propionate (TSP) in D2O or tetrameth-
ylsilane in CD3OD as internal standard. FABMS were
measured using glycerol as a matrix on a JEOL JMS-
700 spectrometer.
4.1.10. N-Nonyl-DNJ (NN-DNJ, 10). [a]D À14.2° (c
1.24, MeOH). HRMS (FAB): m/z 290.2334 [M+H]+
(C15H32NO4 requires 290.2335).
4.1.11. a-1-C-Methyl-DNJ (11). [a]D +78.9° (c 0.46,
H2O). HRMS (FAB): m/z 178.1081 [M+H]+
(C7H16NO4 requires 178.1079).
5.2. Fibroblast culture
Three Gaucher cell lines with b-Glu mutations of
GM00372 (N370S), GM00852 (N370S), and GM00877
(L444P), and a normal fibroblast culture (GM05659)
were obtained from the Coriell Cell Repositories, Cam-
den, NJ.
4.1.12. a-1-C-Propyl-DNJ (12). [a]D +60.7° (c 0.24,
1
H2O). H NMR (D2O): d 0.93 (t, 3 H), 1.24 (m, 1H),
1.40–1.58 (3H), 2.78 (m, 1H), 3.12 (m, 1H, H-1), 3.18
(dd, 1H, J = 9.2, 9.6 Hz, H-4), 3.53–3.56 (2H, H-3,
H-6a), 3.64 (dd, 1H, J = 5.5, 10.1 Hz, H-2), 3.87 (dd,
1H, J = 2.8, 11.5 Hz, H-6b). 13C NMR (D2O): d 16.0,
21.6, 29.0, 56.7, 57.9, 64.7, 75.1, 75.5, 77.0. HRMS
(FAB): m/z 206.1392 [M+H]+ (C9H20NO4 requires
206.1392).
5.3. Biological assays
Human b-glucocerebrosidase (Ceredase) was purchased
from Genzyme (Boston, MA) and assayed at pH 5.5.
The cell lysate of normal human fibroblasts
(GM00498B), which were cultured in MEM (Gibco) sup-
plemented with 15% fetal bovine serum and antibiotics at
37 °C under 5% CO2, was used as the source of lysosomal
a-glucosidase (a-Glu) and a-mannosidase (a-Man). The
reaction mixture consists of 50 lL of 0.15 M sodium
phosphate–citrate buffer (pH 4.5), 50 lL of 2% Triton
X-100 (Sigma Chemical Co.), 30 lL of the enzyme solu-
tion, and 20 lL of an inhibitor solution or H2O. The
reaction mixture was pre-incubated at 0 °C for 10 min
and the reaction was started by the addition of 50 lL
of 6 mM 4-methylumbelliferyl glycoside (Sigma Chemi-
cal Co.), followed by incubation at 37 °C. The reaction
was stopped by the addition of 2 mL of 0.1 M glycine
buffer (pH 10.6). Liberated 4-methylumbelliferone was
4.1.13. a-1-C-Butyl-DNJ (13). [a]D +50° (c 0.13, H2O).
HRMS (FAB): m/z 220.1550 [M+H]+ (C10H22NO4 re-
quires 220.1550).
4.1.14. a-1-C-Hexyl-DNJ (14). [a]D +61.5° (c 0.62,
1
H2O). H NMR (D2O): d 0.89 (t, 3H), 1.23–1.47 (8H),
1.50–1.60 (2H), 2.79 (ddd, 1H, J = 2.8, 7.3, 9.2 Hz,
H-5), 3.12 (m, 1H, H-1), 3.19 (dd, 1H, J = 9.2,
10.0 Hz, H-4), 3.56 (dd, 1H, J = 7.3, 11.5 Hz, H-6a),
3.56 (t, 1H, J = 10.0 Hz, H-3), 3.67 (dd, 1H, J = 5.5,
10.0 Hz, H-2), 3.92 (dd, 1H, J = 2.8, 11.5 Hz, H-6b).
13C NMR (D2O): d 16.2, 24.7, 26.5, 28.1, 31.1, 33.8,
56.7, 58.2, 64.8, 75.2, 75.5, 77.0. HRMS (FAB): m/z
248.1863 [M+H]+ (C12H26NO4 requires 248.1862).

7744
L. Yu et al. / Bioorg. Med. Chem. 14 (2006) 7736–7744
11. Helenius, A.; Aebi, M. Science 2001, 291, 2364.
12. Spiro, R. G. J. Biol. Chem. 2000, 275, 35657.
13. Rabouille, C.; Spiro, R. G. J. Biol. Chem. 1992, 267,
11573.
14. Moor, S. E. H.; Spiro, R. G. J. Biol. Chem. 1993, 268,
3809.
measured (excitation 362 nm, emission 450 nm) with a
F-4500 fluorescence spectrophotometer (Hitachi, Tokyo,
Japan). Kinetic parameters were determined by the dou-
ble-reciprocal-plot method of Lineweaver–Burk at
increasing concentrations of 4-MU-b-D-glucoside.
15. Ogier-Denis, E.; Bauvy, C.; Cluzeaud, F.; Vandewalle, A.;
Codogno, P. Biochem. J. 2000, 345, 459.
16. Petrescu, S. M.; Petrescu, A. J.; Titu, H. N.; Dwek, R. A.;
Platt, F. M. J. Biol. Chem. 1997, 272, 15796.
17. Tian, G.; Wilcockson, D.; Perry, V. H.; Rudd, P. M.;
Dwek, R. A.; Platt, F. M.; Platt, N. J. Biol. Chem. 2004,
279, 39303.
18. Block, T. M.; Lu, X.; Mehta, A. S.; Blumberg, B.
S.; Tennant, B.; Ebling, M.; Korba, B.; Lansky, D.
M.; Jacob, G. S.; Dwek, R. A. Nat. Med. 1998, 4,
610.
Fibroblasts established from Gaucher patients were cul-
tured in the presence of pharmacological chaperones in
a similar manner as described above in normal fibro-
blasts. After washing with a phosphate buffer twice,
the cell pellets were homogenized in a citrate buffer
(pH 5.2) containing 0.25% sodium taurocholate and
0.1% Triton X-100 using a microhomogenizer (Physco-
tron, Niti-on Inc., Chiba, Japan). The supernatant ob-
tained from the homogenate after centrifugation at
10,000g for 5 min was subjected to enzyme assays and
protein determination.
´
19. Zitzmann, N.; Mehta, A. N.; Carrouee, S.; Butters, T. D.;
Platt, F. M.; McCauley, J.; Blumberg, B. S.; Dwek, R. A.;
Block, T. M. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
11878.
20. Sawker, A. R.; Adamski-Werner, S. L.; Cheng, W.-C.;
Wong, C.-H.; Beutler, E.; Zimmer, K.-P.; Kelly, J. W.
Chem. Biol. 2005, 12, 1235.
21. Asano, N.; Yamashita, T.; Yasuda, K.; Ikeda, K.; Kizu,
H.; Kameda, Y.; Kato, A.; Nash, R. J.; Lee, H.-S.; Ryu,
K.-S. J. Agric. Food Chem. 2001, 49, 4208.
22. Asano, N.; Yamauchi, T.; Kagamifuchi, K.; Shimizu, N.;
Takahashi, S.; Takatsuka, H.; Ikeda, K.; Kizu, H.;
Chuakul, W.; Kettawan, A.; Okamoto, T. J. Nat. Prod.
2005, 68, 1238.
Acknowledgment
This work was supported in part by the Special
Research Fund of Hokuriku University.
References and notes
1. Beutler, E.; Grabowski, G. A. In The Metabolic and
Molecular Bases of Inherited Disease; Scriver, C. R.,
Beaudet, A. L., Sly, W. S., Valle, D., Eds.; McGraw-Hill:
New York, 2001; pp 3635–3668.
23. Yamashita, T.; Yasuda, K.; Kizu, H.; Kameda, Y.;
Watson, A. A.; Nash, R. J.; Fleet, G. W. J.; Asano, N.
J. Nat. Prod. 2002, 65, 1875.
24. Godin, G.; Compain, P.; Masson, G.; Martin, O. R.
J. Org. Chem. 2002, 67, 6960.
25. Godin, G.; Compain, P.; Martin, O. R. Org. Lett. 2003, 5,
3269.
26. Godin, G.; Compain, P.; Martin, O. R.; Ikeda, K.;
Yu, L.; Asano, N. Bioorg. Med. Chem. Lett. 2004, 14,
5991.
2. Asano, N. Glycobiology 2003, 13, 93R.
3. Fan, J.-Q. A. Trends Pharmacol. Sci. 2003, 24, 355.
4. Cohen, F. E.; Kelly, J. W. Nature 2003, 426, 905.
5. Butters, T. D.; Dwek, R. A.; Platt, F. M. Glycobiology
2005, 15, 43R.
6. Platt, F. M.; Neises, G. R.; Dwek, R. A.; Butters, T. D.
J. Biol. Chem. 1994, 269, 8362.
27. Asano, N.; Ishii, S.; Kizu, H.; Ikeda, K.; Yasuda, K.;
Kato, A.; Martin, O. R.; Fan, J.-Q. Eur. J. Biochem. 2000,
267, 4179.
28. Lin, H.; Sugimoto, Y.; Ohsaki, Y.; Ninomiya, H.; Oka,
A.; Taniguchi, M.; Ida, H.; Eto, Y.; Ogawa, S.; Matsu-
zaki, Y.; Sawa, M.; Inoue, T.; Higaki, K.; Nanba, E.;
Ohno, K.; Suzuki, Y. Biochim. Biophys. Acta. 2004, 1689,
219.
7. Fan, J.-Q.; Ishii, S.; Asano, N.; Suzuki, Y. Nat. Med.
1999, 5, 112.
8. Yam, G. H.-F.; Zuber, C.; Roth, J. FASEB J. 2005, 19, 12.
9. Sawker, A. R.; Cheng, W.-C.; Beautler, E.; Wong, C.-H.;
Baich, W. E.; Kelly, J. W. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 15428.
10. Ellgaard, L.; Molinari, M.; Helenius, A. Science 1999, 286,
1882.