Synthesis and biological activity of novel deoxynojirimycin derivatives as potent α-glucosidase inhibitors

Article abstract of DOI:10.5560/ZNB.2013-2318

Thirteen 1-deoxynojirimycin (DNJ) derivatives of five different skeletal structures were designed and synthesized. The newly synthesized compounds were evaluated using an in vitro α-glucosidase assay, and kinetic parameters (Ki, IC₅₀) were measured. Some DNJ derivatives showed weak α-glucosidase inhibitory activities, and the compounds 1-(3-benzyloxy-2-hydroxypropyl)- 2-hydroxymethyl-piperidine-3,4,5-triol (2a) and 1-{3-[1-(4-fluorophenyl)-1H-[1,2,3]triazol-4-ylmethoxy]- 2-hydroxypropyl}-2- hydroxymethyl-piperidine-3,4,5-triol (13d) showed activities comparable to that of DNJ. While 2a was found to be a reversible, non-competitive inhibitor of a-glucosidase with a Ki value of 1.56×10^-4 M and an IC ₅₀ value of 3.07×10^-4 M, 13d was a reversible, competitive inhibitor of α-glucosidase with a Ki value of 2.08×10^-4 M and an IC₅₀ value of 3.31×10 ^-4 M.

Full text of DOI:10.5560/ZNB.2013-2318

Synthesis and Biological Activity of Novel Deoxynojirimycin Derivatives
as Potent α-Glucosidase Inhibitors
Dan Yua, Fangfang Hub, Yu Zhanga, Xiaorui Zhengb, Chunxiang Kuangb, and Qing Yanga
a
School of Life Sciences, Fudan University, Handan Road 220, Shanghai 200433, P. R. China
Department of Chemistry, Tongji University, Siping Road 1239, Shanghai 200092, P. R. China
b
Reprint requests to Dr. Qing Yang. Fax: +86-21-6564-3446. E-mail: yangqing68@fudan.edu.cn
Z. Naturforsch. 2013, 68b, 383 – 390 / DOI: 10.5560/ZNB.2013-2318
Received December 1, 2012
Thirteen 1-deoxynojirimycin (DNJ) derivatives of five different skeletal structures were de-
signed and synthesized. The newly synthesized compounds were evaluated using an in vitro α-
glucosidase assay, and kinetic parameters (Ki, IC50) were measured. Some DNJ derivatives showed
weak α-glucosidase inhibitory activities, and the compounds 1-(3-benzyloxy-2-hydroxypropyl)-
2-hydroxymethyl-piperidine-3,4,5-triol (2a) and 1-{3-[1-(4-fluorophenyl)-1H-[1,2,3]triazol-4-ylme-
thoxy]-2-hydroxypropyl}-2-hydroxymethyl-piperidine-3,4,5-triol (13d) showed activities compara-
ble to that of DNJ. While 2a was found to be a reversible, non-competitive inhibitor of α-glucosidase
with a Ki value of 1.56 × 10−4 M and an IC50 value of 3.07 × 10−4 M, 13d was a reversible, competi-
tive inhibitor of α-glucosidase with a Ki value of 2.08 × 10−4 M and an IC50 value of 3.31 × 10−4 M.
Key words: 1-Deoxynojirimycin, α-Glucosidase Inhibitor, Diabetes Mellitus
Introduction
in structure as well as in the transition state of
glucosidase-catalyzed reactions, DNJ is able to inhibit
Type-II noninsulin-dependent diabetes mellitus α-glucosidase in a competitive manner [9, 10]. Given
(NIDDM), one of the most severe human metabolic that DNJ showed good inhibitory activity against α-
disorder diseases which accounts for 90 – 95% of all glucosidase [11], a number of DNJ derivatives have
diabetes, has reached epidemic proportions globally, been synthesized in order to achieve significant po-
and its incidence is continuing to rise [1 – 3]. NIDDM tency of suppressing post-prandial elevation of the glu-
is characterized by chronic hyperglycemia, insulin re- cose level in blood. Successful results were first ob-
sistance and impaired insulin secretion [4]. One ther- tained with miglitol and emiglitate. The ED50 was
apeutic approach to treat NIDDM is to retard the ab- evaluated respectively as 0.24 and 0.16 mg/kg body
sorption of glucose via the inhibition of enzymes, such weight [12, 13].
as α-glucosidase, in the digestive organs [5, 6].
The first marketed α-glucosidase inhibitor is acar-
α-Glucosidase, which is a membrane-bound en- bose, a pseudo-tetrasaccharide containing valien-
zyme located at the epithelium of small intestine, cat- amine, developed by Bayers group in the early
alyzes the cleavage of a glucose molecule from disac- 1990s [14]. Valienamine shows structural similarity
charides and plays an essential role in the proper diges- with DNJ except that the cyclohexane ring and the ring
tion of carbohydrates [7]. It is beneficial for NIDDM nitrogen atom are replaced with a cyclohexene ring
patients to delay glucose release and its absorption af- and an exocyclic amino group, respectively. At present
ter meals by inhibition of α-glucosidase in their ther- pseudosaccharide acarbose and azasugar miglitol are
apy treatments [8].
clinically used [15 – 17] for NIDDM patients to re-
1-Deoxynojirimycin (DNJ), a characteristic con- tard glucose absorption; however, these inhibitors are
stituent of mulberry leaves, is a D-glucose analog associated with severe side effects including adverse
with a secondary amine group instead of an oxy- gastrointestinal effects and abdominal discomfort [18].
gen atom in the pyranose ring. Resembling D-glucose In addition, Kiyotaka et al. found that the absorption
© 2013 Verlag der Zeitschrift fu¨r Naturforschung, Tu¨bingen · http://znaturforsch.com
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM

384
D. Yu et al. · Deoxynojirimycin Derivatives as Potent α-Glucosidase Inhibitors
H
O
ability of DNJ in rats was relatively low, and DNJ
had a short half-life in vivo [19]. Hence, it is still of
great interest to find new potent DNJ derivatives as α-
glucosidase inhibitors.
In continuation of our work on the development
of chemotherapeutic agents from DNJ, we designed
several kinds of structural skeletons to introduce aryl,
alkenyl, hydroxyl, and triazole groups into DNJ. We
present here the synthesis and biological activities of
thirteen DNJ derivatives, which are defined as 2a–c,
6a–c, 13a–e, 8, and 15.
HO
H
R
O
H
H
R
O
N
O
H
r. t.
H
3O
C
NH
O
+
H
O
O
HO
O
H
O
H
O
1
1a
2
2a
DNJ
R=
R
=
2b
2c
R=
R=
r
B
1b
1c
R
r
=
B
R
=
H
CO2C
H
CO2C
3
3
Scheme 1. General reaction pathway to compounds 2a–c.
HO
H
O
Results and Discussion
CH3OH
r. t.
HO
H
O
N
NH
+
Synthesis of DNJ derivatives
H
O
H
O
HO
O
H
O
H
O
DNJ derivatives 2, 6, 8, 13, and 15 were synthe-
sized as described in Schemes 1 – 5. Compounds 2, 8
and 15 were prepared by the direct reactions between
DNJ and epoxides 1a–c (Scheme 1), 2-phenyloxirane 7
7
8
Scheme 3. General reaction pathway to compound 8.
(Scheme 3) and the triazole compound 14 (Scheme 5), Inhibition study
respectively.
Using cinnamic acid derivative 3 as a starting mate-
As shown in Schemes 1 – 4, the synthesis of com-
rial, after esterification and reduction, cinnamyl alco- pounds 2, 6, 8, and 13 introduced a new stereogenic
hol derivative 4 was obtained, which subsequently re- center. In the present study, the biological tests were
acted with (chloromethyl)oxirane to yield epoxy com- done on the mixtures of diastereomers.
pound 5. Condensation of DNJ and 5 afforded the
designed compound 6 containing an alkene group microtiter plate as described previously with a minor
(Scheme 2). modification [20, 21]. The values of Ki and inhibition
The inhibition assays were performed in a 96-well
Click reaction between phenyl azide 9 and propar- types were determined by a Lineweaver-Burk plot. The
gyl alcohol 10 gave triazole 11, which subsequently concentration required to inhibit enzyme activity by
reacted with 2-(chloromethyl)oxirane to yield epoxy 50% (IC50) was calculated by regression analysis. The
compound 12. Condensation of DNJ and 12 afforded α-glucosidase inhibition constants Ki, inhibition types
the target compound 13 and resulted in the introduc- and IC50 values of DNJ, DNJ derivatives and acarbose
tion of a triazole group (Scheme 4).
are summarized in Table 1; the structure-activity rela-
1) ClCO2C2H5, Et3N, THF
2) NaBH4, CH3OH
Cl
COOH
, NaOH/H2O, TBAB
r. t.
OH
O
R
R
3
4
HO
HO
NH
HO
HO
OH
O
HO
HO
R
O
N
O
CH3OH, r. t.
R
OH
OH
5
6
R = H(a), Me(b), Cl(c)
Scheme 2. General reaction pathway to compounds 6a–c. TBAB = tetrabutylammonium bromide.
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM

D. Yu et al. · Deoxynojirimycin Derivatives as Potent α-Glucosidase Inhibitors
385
N
N
N
u
a ascor a e
e
M N H
C 2O
I N
,
b t
C
/
,
H
O
R
N3
+
R
H
O
10
9
11
l
C
e
M
H DNJ
,
O
N
N
N
a
O
N
H H
TBAB
O
/
O
,
,
O
2
O
R
r. t.
r. t.
12
H
O
H
H
O
N
N
R
O
N
N
H
O
O
H
O
13
=
R
a
b
Br(c),
F(d), Cl(e)
H ( ), Me ( ),
Scheme 4. General reaction pathway to compounds 13a–e.
HO
Table 1. Ki and IC50 values of DNJ derivatives obtained from
yeast α-glucosidase enzymatic assays.
N
N
N
K2CO3 DNJ
,
H
O
l
C
N
N
N
Cl
Cl
r. t.
N
Compound
Ki value
Inhibition type
IC50
HO
(10−4 M)a
(10−4 M)
OH
2a
2b
2c
6a
6b
6c
8
13a
13b
13c
13d
13e
15
1.56
3.53
17
non-competitive
competitive
competitive
n. i.b
competitive
competitive
competitive
n. i.b
competitive
competitive
competitive
competitive
non-competitive
competitive
competitive
3.07
39
51
n. i.b
29
36
17
n. i.b
26
22
3.31
22
17
4.51
3.44
15
14
Scheme 5. General reaction pathway to compound 15.
n. i.b
5.94
8.55
7.41
n. i.b
9.17
4.18
2.08
8.85
6.77
4.40
4.25
tionship and the representative double-reciprocal plots
of DNJ, 2a and 13d are shown in Figs. 1 and 2, re-
spectively. Except 6a and 13a, the other DNJ deriva-
tives displayed α-glucosidase inhibitory activities in
vitro and were all found to be reversible inhibitors;
particularly, 2a and 13d were the strongest inhibitors
among them. The introduction of a COOCH3 group
in the benzene ring of 2a caused the reduction of
α-glucosidase inhibitory activity, as shown by com-
pound 2c in comparison with 2a. Series 6 was de-
DNJ
Acarbose
a
Substrate (PNP glycoside) concentration: 0, 62.75, 125, 250,
500, 1000 × 10−6 M, inhibitor concentration: 2 × 10−4
M
and
signed by adding an ethylene group in the N-linked 5 × 10−4 M, α-glucosidase concentration 0.2 U mL−1
;
not inhib-
b
ited.
side chain of 2, but 6a showed no inhibitory effect on
α-glucosidase, in spite of being an analog of 2a. There-
fore, the additional introduction of COOCH3 and ethy- a 1,2,3-1H-triazole structural moiety. 13a does not
lene in 2a did not increase its α-glucosidase inhibitory contain any other substituent in the benzene ring, while
activity. Two other compounds, 2b and 8, which pos- 13b–e in the same series contain a halogen or an alkyl
sess a shorter N-linked side chain than 2a, showed less group at the ring. Although 13a did not show any po-
α-glucosidase inhibitory activities than 2a. However, tency of inhibiting α-glucosidase, the introduction of
by replacing a hydrogen atom in the benzene ring of a halogen or an alkyl group at the benzene ring of 13a
6a with a methyl or chloro substituent, moderate α- resulted in significant inhibitory activities, in which the
glucosidase inhibitory activities were achieved.
Compounds 13a–e are another series of DNJ deriva- 13b–e. In this series of DNJ derivatives, 13d had po-
tives owing to their similar structures containing tency similar to that of 2a. Furthermore compounds 15
Ki values were in the range of 2.08 – 9.17 × 10−4 M for
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM

386
D. Yu et al. · Deoxynojirimycin Derivatives as Potent α-Glucosidase Inhibitors
Fig. 1. Structure-activity relationships of DNJ
derivatives as potent α-glucosidase inhibitors.
and 13e, containing both a chloro substituent at their
benzene rings, exhibited different activities owing to
the shorter side chain of 15.
Even though the studies of DNJ and acarbose
have been carried out widely for many years, the
sources reporting their IC50 and Ki values are lim-
ited. The IC50 values of acarbose in the literature
are quite different, ranging from 1.29 × 10−4 M to
4.025 × 10−3 M [22 – 26]. In our α-glucosidase in-
hibitory activity assay system, the observed IC50
of acarbose was 3.44 × 10−4 M, which was consis-
tent with the reference data of 3.36 × 10−4 M and
2.90 × 10−4 M [24, 25]. However, under the present
conditions, the IC50 of DNJ was determined to be
4.51 × 10−4 M, which was much higher than the sole
published data of 1.0 × 10−5 M [27]. This phenomenon
may be attributed to the complexity of the enzyme as-
say or the unidentified differences between the source
and purity of DNJ.
Conclusion
In this study, we have synthesized thirteen new
DNJ derivatives, evaluated their α-glucosidase in-
hibitory activities, investigated the structure-activity
relationship and discovered two potent α-glucosidase
inhibitors (compounds 2a and 13d) that showed α-
glucosidase inhibitory activity comparable to that of
DNJ. The inhibitor structures suggested that an N-
linked side chain containing an aryl ring which in-
creases the lipophility of DNJ would be beneficial for
the α-glucosidase inhibition. The length and the com-
ponent of the side chain as well as the substituent of the
Fig. 2. Double-reciprocal plots of the inhibition kinetics of
yeast α-glucosidase by DNJ (a), 2a (b) and 13d (c). Sub-
strate (PNP glycoside) concentration: 0, 62.75, 125, 250,
500, 1000 × 10−6 M, inhibitor concentration 2 × 10−4 M and
5 × 10−4 M, α-glucosidase concentration 0.2 U mL−1
.
aryl ring varied the inhibitory activities. Compounds Experimental Section
2a and 13d would be a lead for designing new com-
General
pounds, and further derivatives could be prepared with
modification of these particular moieties. Therefore,
our result provide very useful clues for the design and
development of more potent α-glucosidase inhibitors.
Commercially obtained reagents were used with-
out further purification. TLC: Huanghai GF254 silica
gel-coated plates. Column chromatography (CC): SiO2
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM

D. Yu et al. · Deoxynojirimycin Derivatives as Potent α-Glucosidase Inhibitors
387
(300 – 400 mesh), at medium pressure. The 1H and 13C
The starting chemicals 2-benzyloxymethyl-oxirane (1a)
NMR spectra were recorded on a Bruker DPX-300 spec- and 2-(4-bromo-phenoxymethyl)oxirane (1b) were used as
trometer in CDCl3 with TMS as an internal standard. The obtained commercially.
chemical shifts are reported in parts per million (ppm)
4-Oxiranylmethoxymethyl-benzoic acid methyl ester (1c)
expressed in δ units; coupling constant (J) values are given
Epoxy chloropropane (0.462 g, 5 mmol), tetrabutylammo-
nium bromide (TBAB, 16 mg, 0.05 mmol) and 50% aque-
ous NaOH solution (0.6 mL) were added to a flask. The
mixture was violently stirred at r. t. until dissolution, then it
was chilled with an ice bath, and methyl 4-hydroxybenzoate
(166 mg, 1 mmol) was slowly added. The solution was stirred
for 18 h at 0 ◦C. After completion of the reaction, cold wa-
ter was added to the mixture, and then EtOAc for extrac-
tion. The organic layer was washed with salt water to neu-
tral, dried with anhydrous sodium sulfate, and EtOAc was re-
moved in vacuo. The crude product was purified by column
chromatography on silica gel to yield a pale-yellow liquid
(1c); yield 158 mg (71%). – 1H NMR (500 MHz, CDCl3):
δ = 8.02 (d, 2H), 7.42 (d, 2H), 4.65 (m, 2H), 3.82 (m, 1H),
3.45 (m, 1H), 3.21 (m, 1H), 2.82 (t, 1H), 2.63 (m, 1H), 1.59
(s, 3H). – MS (ESI): m/z = 222.1 [M]+. – HRMS ((+)-ESI):
m/z = 222.0890 (calcd. 222.0892 for C12H14O4, [M]+).
in Hertz (Hz). MS data were measured with a Varian-310
spectrometer. High-resolution mass spectra were determined
using a Finnigan-NAT GC/MS/DS 8430 spectrometer.
α-Glucosidase assay
p-Nitrophenyl-α-D-glucopyranoside (PNPG), yeast α-
glucosidase (EC 3.2.1.20), sodium phosphate and sodium
carbonate were purchased from Sigma-Aldrich (China).
α-Glucosidase inhibitory activity was performed follow-
ing the modified method of Pistia, Brueggeman and
Hollingsworth [20, 21]. A reaction mixture containing 50 µL
of phosphate buffer (5 × 10−2 M, pH = 6.8), 10 µL of yeast
α-glucosidase (1 U mL−1) and 20 µL of the DNJ derivative
at varying concentrations was pre-incubated for 10 min at
37 ◦C, and then 20 µL of 10−3 M PNPG was added to the
mixture as a substrate. After further incubation at 37 ◦C for
30 min, the reaction was terminated by adding one volume
of Na2CO3 (1 M). All the enzyme, inhibitor and substrate so-
lutions were made using the same buffer. Enzymatic activ-
ity was quantified by measuring the absorbance at 405 nm in
a microtiter plate reader (Thermo Multiskan MK3, China).
Each experiment was performed in triplicate. The concentra-
tion of an inhibitor required to inhibit 50% of enzyme ac-
tivity under the conditions is defined as the IC50 value. IC50
was calculated by regression analysis, using the following
equation, where v is the percentage of inhibition, A100 is the
maximum inhibition, I is the inhibitor concentration, IC50 is
the concentration required to inhibit activity of the enzyme
by 50%, and s is the cooperative degree:
1-(3-Benzyloxy-2-hydroxypropyl)-2-hydroxymethyl-
piperidine-3,4,5-triol (2a)
Colorless solid. Yield 87%. – 1H NMR (500 MHz, D2O):
δ = 7.37 (s, 5H), 4.54 (t, 2H), 3.98 (m, 1H), 3.82 (m, 2H),
3.43 – 3.52 (m, 3H), 3.26 (m, 2H), 3.00 (m, 1H), 2.76 (m,
1H), 2.53 (m, 1H), 2.33 (m, 2H). – 13C NMR (125 MHz,
D2O): δ = 54.1, 56.5, 57.9, 65.8, 67.4, 68.7, 70.1, 74.8, 78.3,
128.4, 128.6, 128.8, 137.5. – MS (ESI): m/z = 327.2 [M]+.
– HRMS ((+)-ESI): m/z = 327.1680 (calcd. 327.1682 for
C16H25NO6, [M]+).
1-[3-(4-Bromo-phenoxy)-2-hydroxypropyl]-
2-hydroxymethyl-piperidine-3,4,5-triol (2b)
A100
Colorless solid. Yield 81%. – 1H NMR (500 MHz, D2O):
δ = 7.38 (t, 2H), 6.83 (t, 2H), 4.17 (m, 1H), 3.78 – 3.98 (m,
4H), 3.56 (m, 1H), 3.12 (m, 2H), 3.13 (m, 1H), 2.96 (d,
1H, J = 13.5 Hz), 2.67 (m, 1H), 2.37 (m, 2H). – 13C NMR
(125 MHz, D2O): δ = 53.9, 57.2, 66.5, 67.5, 68.7, 70.0, 70.8,
78.3, 113.2, 116.9, 132.5, 157.6. – MS (ESI): m/z = 393.0
[M+2]+, 391.0 [M]+. – HRMS ((+)-ESI): m/z = 393.0608
(calcd. 393.0610 for C15H2281BrNO6, [M]+).
v =
(1)
S
1+(I/IC50
)
Kinetics of enzyme inhibition
The enzyme reaction was performed according to the
above reaction conditions with inhibitors at various concen-
trations (2 × 10−4 to 2 × 10−3 M). Inhibition strength for the
inhibitor was determined by a Dixon plot and its replot of
slope versus the reciprocal of the substrate concentration.
4-[2-Hydroxy-3-(3,4,5-trihydroxy-2-hydroxymethyl-
piperidin-1-yl)propoxymethyl]benzoic acid methyl ester (2c)
General procedure for the synthesis of DNJ derivatives 2
DNJ (100 mg, 0.6 mmol) and oxirane 1 (0.7 mmol) were
Colorless solid. Yield 82%. – 1H NMR (400 MHz, D2O):
stirred in methanol-H2O (5 mL : 1 mL) at r. t., for 24 h, and δ = 7.96 (d, 2H), 7.49 (d, 2H), 4.65 (s, 1H), 4.34 (d, 1H),
the reaction was monitored with TLC (EtOAc-MeOH = 4 : 4.07 (s, 1H), 3.68 – 3.97 (m, 5H), 3.31 – 3.57 (m, 5H), 3.09
1). After completion of the reaction, the methanol was re- (m, 1H), 2.85 (m, 1H), 2.65 (m, 1H), 2.37 (m, 2H). –
moved in vacuo, and the residue was purified by column MS (ESI): m/z = 385.2 [M]+. – HRMS ((+)-ESI): m/z =
chromatography on silica gel to yield DNJ derivative 2.
385.1734 (calcd. 385.1736 for C18H27NO8, [M]+).
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM

388
D. Yu et al. · Deoxynojirimycin Derivatives as Potent α-Glucosidase Inhibitors
General procedure for the synthesis of DNJ derivatives 6
1-{3-[3-(4-Chlorophenyl)allyloxy]-2-hydroxypropyl}-
2-hydroxymethyl-piperidine-3,4,5-triol (6c)
The mixture of cinnamic acid derivative 3 (10 mmol),
ethyl chloroformate (1.2 mL, 12 mmol), triethylamine
(1.21 g, 12 mmol), and tetrahydrofuran (100 mL) was
stirred at 0 ◦C for 30 min. Sodium borohydride (756.6 mg,
20 mmol) was added, and the temperature was raised to
10 ◦C, then methanol (1 mL) was added in drops into
the mixture. TLC analysis showed that the reaction was
completed in 24 h. The methanol was removed in vacuo, and
the residue was partitioned between dichloromethane and
water. The combined organic phase was dried with anhy-
drous sodium sulfate and then concentrated. Purification by
column chromatography on silica gel afforded the desired
cinnamyl alcohol derivative 4a–c [28].
Compound 4 (1 mmol) was slowly added to the solu-
tion of TBAB (16 mg, 0.05 mmol), 2-(chloromethyl)oxirane
(462 mg, 5 mmol) and 50% aqueous NaOH solution
(0.6 mL) at 0 ◦C. TLC analysis showed that the reaction was
completed in 18 h. To the reaction mixture cold water was
added, and then EtOAc for extraction. The organic layer was
washed to neutral with saturated NaCl solution, dried with
anhydrous sodium sulfate, and concentrated in vacuo. Pu-
rification by column chromatography on silica gel afforded
epoxy compounds 5a–c [29, 30].
Colorless solid. Yield 86%. – 1H NMR (500 MHz, D2O):
δ = 7.43 (t, 2H), 7.37 (t, 2H), 6.66 (m, 1H), 6.36 (m, 1H),
4.21 (m, 2H), 4.07 (m, 1H), 3.82 (m, 2H), 3.65 (m, 2H),
3.38 (m, 2H), 3.28 (m, 1H), 3.12 (m, 1H), 2.86 (m, 1H),
3.12 (m, 1H), 2.40 (m, 2H). – MS (ESI): m/z = 387.1 [M]+.
– HRMS ((+)-ESI): m/z = 387.1446 (calcd. 387.1448 for
C18H2635ClNO6, [M]+).
General procedure for the synthesis of DNJ derivative 8
2-Phenyloxirane (7) reacts with DNJ in MeOH at r. t.
to give 2-hydroxymethyl-1-(2-hydroxy-2-phenylethyl)piper-
idine-3,4,5-triol (8); detailed data were the same as that re-
ported in the literature [31].
General procedure for the synthesis of DNJ derivatives 13
A flask containing a mixture of an aniline derivative
(5 mmol) and an aqueous concentrated HCl-water (10 mL :
1 mL) solution was placed into an ice bath, and to the mix-
ture an aqueous NaNO2 solution was added in drops until
KI-starch paper became blue. After stirring for 5 min, sodium
azide (0.488 g, 7.5 mmol) in 4 mL water was slowly added,
and the reaction was maintained for additional 1 – 2 h. After
completion of the reaction, KOAc was added to adjust pH to
neutral prior to extraction with EtOAc. The organic layer was
washed with saturated NaCl solution, dried with anhydrous
sodium sulfate, and then concentrated to give pale-yellow
liquid phenyl azides 9a–e.
DNJ (10 mg, 0.06 mmol) and epoxy compound
5
(0.07 mmol) were stirred in methanol-H2O (1 mL : 200 µL)
at r. t. for 24 h, and the reaction was monitored with TLC
(EtOAc-MeOH = 4 : 1). After removing of the solvent, the
crude product was purified by chromatography on silica gel
to yield DNJ derivatives 6a–c.
The mixture of 9 (4.5 mmol), propargyl alcohol (0.280 g,
5.0 mmol), cuprous iodide (0.382 g, 2 mmol), sodium ascor-
bate (0.396 g, 2 mmol), and CH3CN-H2O (30 mL : 3 mL)
was stirred at r. t. under nitrogen for 24 h. After comple-
tion of the reaction, the mixture was partitioned between
EtOAc and water. The organic layer was dried with an-
hydrous sodium sulfate and concentrated. Purification by
column chromatography on silica gel afforded the desired
products 11a–e [32].
Compound 11 (5 mmol) was slowly added to the mix-
ture of TBAB (0.078 g, 0.2 mmol), 2-(chloromethyl)oxirane
(2.312 g, 25 mmol) and a 50% aqueous NaOH solution
(2 mL) at 0 ◦C. After completion of the reaction (18 h), cold
water was added, and then ethyl ether for extraction. The
organic layer was washed to neutral with saturated NaCl
2-Hydroxymethyl-1-[2-hydroxy-3-(3-phenyl-
allyloxy)propyl]piperidine-3,4,5-triol (6a)
Colorless solid. Yield 89%. – 1H NMR (500 MHz, D2O):
δ = 7.42 (t, 2H), 7.31 (m, 2H), 7.25 (t, 1H), 6.62 (m, 1H),
4.14 (t, 2H), 3.99 (m, 1H), 3.78 (m, 2H), 3.44 (m, 4H), 3.23
(t, 1H), 3.08 (m, 1H), 2.78 (m, 1H), 2.52 (m, 1H), 2.29
(m, 2H). – 13C NMR (125 MHz, D2O): δ = 53.75, 55.96,
57.01, 65.21, 66.27, 67.12, 67.90, 69.33, 70.86, 77.68, 124.8,
125.82, 125.91, 127.46, 128.24, 132.61, 135.79. – MS (ESI):
m/z = 353.2 [M]+. – HRMS ((+)-ESI): m/z = 353.4102
(calcd. 353.4100 for C18H27NO6, [M]+).
2-Hydroxymethyl-1-[2-hydroxy-3-(3-p-tolyl-
allyloxy)propyl]piperidine-3,4,5-triol (6b)
Colorless solid. Yield 70%. – 1H NMR (500 MHz, D2O): solution, dried with anhydrous sodium sulfate and concen-
δ = 7.37 (m, 2H), 7.19 (m, 2H), 6.64 (m, 1H), 6.30 (m, 1H), trated. Purification by column chromatography on silica gel
4.20 (m, 2H), 4.08 (m, 1H), 3.85 (m, 2H), 3.59 (m, 2H), 3.49 afforded the desired products 12a–e [33].
(m, 2H), 3.40 (m, 1H), 3.32 (m, 1H), 3.13 (m, 1H), 2.86 (m,
The mixture of 12 (0.7 mmol) and DNJ (100 mg,
1H), 2.40 (m, 2H), 2.29 (s, 3H). – MS (ESI): m/z = 367.2 0.6 mmol) was stirred in methanol-H2O (5 mL : 1 mL) at
[M]+. – HRMS ((+)-ESI): m/z = 367.1992 (calcd. 367.1995 r. t. for 16 h. After the reaction was completed, the solvent
for C19H29NO6, [M]+).
was removed in vacuo. The residue was purified by column
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM

D. Yu et al. · Deoxynojirimycin Derivatives as Potent α-Glucosidase Inhibitors
389
chromatography on silica gel to afford the desired products 3.11 – 3.19 (m, 1H), 2.94 (m, 1H), 2.66 (m, 1H), 2.45 (m,
13a–e.
2H). – 13C NMR (125 MHz, D2O): δ = 54.0, 56.5, 58.0,
66.1, 67.8, 68.6, 71.5, 76.1, 78.2, 116.9, 123.5, 132.7, 144.8,
163.7. – MS (ESI): m/z = 412.1 [M]+. – HRMS ((+)-ESI):
m/z = 412.1756 (calcd. 412.1758 for C18H25FN4O6, [M]+).
2-Hydroxymethyl-1-[2-hydroxy-3-(1-phenyl-1H-[1,2,3]-
triazol-4-ylmethoxy)propyl]piperidine-3,4,5-triol (13a)
Colorless solid. Yield 85%. – 1H NMR (500 MHz, D2O):
δ = 8.35 (s, 1H), 7.61 (d, J = 8.0 Hz, 2H), 7.48 (m, 3H), 4.66
(s, 2H), 3.99 (m, 1H), 3.79 – 3.89 (m, 1H), 3.45 – 3.60 (m,
4H), 3.21 (t, 1H), 3.02 (m, 1H), 2.78 (m, 1H), 2.52 (m, 1H),
2.20 – 2.35 (m, 2H). – 13C NMR (125 MHz, D2O): δ = 53.4,
56.3, 57.9, 63.3, 65.9, 67.7, 68.5, 70.0, 72.6, 78.3, 120.9,
123.5, 129.6, 129.9, 136.1, 144.5. – MS (ESI): m/z = 394.1
[M]+. – HRMS ((+)-ESI): m/z = 394.1850 (calcd. 394.1852
for C18H26N4O6, [M]+).
1-{3-[1-(4-Chlorophenyl)-1H-[1,2,3]triazol-4-ylmethoxy]-
2-hydroxypropyl}-2-hydroxymethyl-piperidine-3,4,5-triol
(13e)
Pale-yellow solid. Yield 80%. – 1H NMR (500 MHz,
D2O): δ = 8.34 (s, 1H), 7.56 (s, 2H), 7.44 (s, 2H), 4.66 (s,
2H), 3.99 (d, J = 3.0 Hz, 1H), 3.78 (m, 2H), 3.38 – 3.57 (m,
4H), 3.19 (m, 1H), 3.01 (m, 1H), 2.77 (m, 1H), 2.46 – 2.56
(m, 1H), 2.31 (m, 2H). – 13C NMR (125 MHz, D2O): δ =
52.7, 53.6, 55.9, 57.3, 59.7, 66.2, 67.0, 67.8, 69.3, 70.2,
70.8, 71.9, 77.6, 121.5, 122.6, 129.1, 133.8, 134.1, 144.0.
– MS (ESI): m/z = 428.1 [M]+. – HRMS ((+)-ESI): m/z =
428.1460 (calcd. 428.1462 for C18H2535ClN4O6, [M]+).
2-Hydroxymethyl-1-[2-hydroxy-3-(1-p-tolyl-1H-[1,2,3]-
triazol-4-ylmethoxy)propyl]piperidine-3,4,5-triol (13b)
Colorless solid. Yield 89%. – 1H NMR(500 MHz, D2O):
δ = 8.30 (s, 1H), 7.47 (d, J = 8.1 Hz, 2H), 7.29 (d, J =
8.1 Hz, 2H), 4.06 (m, 2H), 3.90 (m, 1H), 3.20 – 3.83 (m, 7H),
3.04 (m, 1H), 2.80 (m, 1H), 2.54 (m, 1H), 2.21 – 2.27 (m,
5H). – 13C NMR (125 MHz, D2O): δ = 20.3, 53.5, 54.4,
56.5, 58.1, 66.0, 67.0, 67.8, 68.7, 70.1, 72.7, 78.3, 120.9,
123.5, 130.4, 133.9, 140.3, 144.6. – MS (ESI): m/z = 408.2
[M]+. – HRMS ((+)-ESI): m/z = 408.2006 (calcd. 408.2008
for C19H28N4O6, [M]+).
1-[1-(4-Chlorophenyl)-1H-[1,2,3]triazol-4-ylmethyl]-
2-hydroxymethyl-piperidine-3,4,5-triol (15)
4-Chloromethyl-1-(4-chlorophenyl)-1H-[1,2,3]triazole
(14) was prepared according to the procedure reported in
the literature [34]. Compound 14 (150 mg, 0.13 mmol) was
added to a mixture of 1-DNJ (50 mg, 0.30 mmol), K2CO3
(35 mg, 0.25 mmol) and CH3CN (5 mL), and the reaction
was left to continue at 80 ◦C for 24 h. After the reaction
was finished, the solvent was removed in vacuo. The residue
was purified by column chromatography on silica gel to af-
ford the desired product 15. Colorless solid, 75 mg (71%).
1-{3-[1-(4-Bromophenyl)-1H-[1,2,3]triazol-4-ylmethoxy]-
2-hydroxypropyl}-2-hydroxymethyl-piperidine-3,4,5-triol
(13c)
1H NMR (500 MHz, D2O): δ = 8.38 (s, 1H), 7.65 (m,
2H), 7.56 (m, 2H), 4.65 (m, 2H), 4.00 (s, 1H), 3.76 (m,
2H), 3.43 (m, 4H), 3.04 (m, 2H), 2.78 (m, 1H), 2.57 (m,
1H), 2.35 (m, 2H). – MS (ESI): m/z = 472.1 [M]+. –
HRMS ((+)-ESI): m/z = 472.0956 (calcd. 472.0957 for
C18H2579BrN4O, [M]+).
–
1H NMR (500 MHz, D2O): δ = 8.46 (s, 1H), 7.75 (d,
2H), 7.57 (d, 2H), 4.14(d, 1H), 4.06 (m, 2H), 3.91(d, 1H),
3.50 (m, 1H), 3.40 (t, 1H), 3.08 (m, 1H), 2.98 (m, 1H),
2.14 (t, 1H), 2.07 (d, 1H). – MS (ESI): m/z = 354.1 [M]+.
– HRMS ((+)-ESI): m/z = 354.1092 (calcd. 354.1094 for
C15H1935ClN4O4, [M]+).
1-{3-[1-(4-Fluorophenyl)-1H-[1,2,3]triazol-4-ylmethoxy]-
2-hydroxypropyl}-2-hydroxymethyl-piperidine-3,4,5-triol
(13d)
Acknowledgement
This work was supported by the Key Projects of Shanghai
1H NMR (500 MHz, D2O): δ = 8.41 (s, 1H), 7.70 (dd, in Biomedical Sciences (no. 08431902700). We would like
J = 4.2 Hz, J = 13.3 Hz, 2H), 7.30 (t, 2H), 4.09 (m, 1H), to thank the Center for Instrumental Analysis, Tongji Uni-
3.79 – 3.90 (m, 2H), 3.55 – 3.65 (m, 4H), 3.28 – 3.44 (m, 3H), versity, China.
[1] K. C. Wong, Z. Q. Wang, Diabetes Res. Clin. Pr. 2006,
73, 126 – 134.
[3] J. W. M. Hoppener, B. Ahren, C. J. M. Lips, NEJM
2000, 343, 411 – 419.
[2] S. K. Das, R. Chakrabarti, Mini. Rev. Med. Chem. 2005,
5, 1019 – 1034.
[4] E. N. Liberopoulos, S. Tsouli, D. P. Mikhailidis, M. S.
Elisaf, Curr. Drug Targets 2006, 7, 211 – 228.
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM

390
D. Yu et al. · Deoxynojirimycin Derivatives as Potent α-Glucosidase Inhibitors
[5] K. Y. Kim, K. A. Nam, H. Kurihara, S. M. Kim, Phy- [22] B. Dineshkumar, M. Analava, M. Manjunatha, Int. J.
tochem. 2008, 69, 2820 – 2825. Adv. Pharm. Sci. 2010, 1, 75 – 85.
[6] R. R. Holman, C. A. Cull, R. C. Turner, Diabetes Care [23] E. Berna, B. Katrin, M. Abdul, W. Yuliastuti, A. Ban-
1999, 22, 960 – 964.
[7] E. B. de Melo, A. D. Gomes, I. Carvalho, Tetrahedron
gun, E. K. Septiana, J. Biomed. Biotechnol. 2012, ID
281078.
2006, 62, 10277 – 10302.
[8] H. Bischoff, Eur. J. Clin. Invest. 1994, 24, 3 – 10.
[9] M. Yoshimizu, Y. Tajima, F. Matsuzawa, S. Aikawa,
[24] E. M. Hye, N. I. Md, R. A. Bo, S. S. Chowdhury,
H. S. Sohn, H. A. Jung, J. S. Choi, Biosci. Biotechnol.
Biochem. 2011, 75, 1472 – 1480.
K. Iwamoto, T. Kobayashi, T. Edmunds, K. Fujishima, [25] C. M. Wu, J. Y. Shen, P. P. He, Y. Chen, L. Li, L. Zhang,
D. Tsuji, K. Itoh, M. Ikekita, I. Kawashima, K. Suga-
wara, N. Ohyanagi, T. Suzuki, T. Togawa, K. Ohno,
Y. Li, Y. J. Fu, R. J. Dai, W. W. Meng, Y. L. Deng, Rec.
Nat. Prod. 2012, 6, 110 – 120.
H. Sakuraba, Clinica Chimica Acta 2008, 391, 68 – 73. [26] P. C. Hsieh, G. J. Huang, Y. L. Ho, Y. H. Lin, S. S.
[10] T. Kimura, K. Nakagawa, H. Kubota, Y. Kojima,
Y. Goto, K. Yamagishi, S. Oita, S. Oikawa, T. Miya-
zawa, J. Agric. Food Chem. 2007, 55, 5869 – 5874.
[11] N. Asano, K. Oseki, H. Kizu, K. Matsui, J. Med. Chem.
1994, 37, 3701 – 3706.
Huang, Y. C. Chiang, M. C. Tseng, Y. S. Chang, Botan-
ical Studies 2010, 51, 293 – 302.
[27] T. Kajimoto, K. C. Liu, R. L. Pederson, Z. Y. Zhong,
Y. Ichikawa, J. A. Porco, C. H. Wong, J. Am. Chem.
Soc. 1991, 113, 6187 – 6196.
[12] K. M. Robinson, M. E. Begovic, B. L. Rhineheart, [28] C. Bruno, G. Lentini, A. Catalano, A. Carocci, A. Lo-
E. W. Heineke, J. B. Ducep, P. R. Kastner, F. N. Mar-
shall, C. Danzin, Diabetes 1991, 40, 825 – 830.
[13] W. Plus, U. Keup, H. P. Krause, G. Thomas, F. Hoff-
meister, Naturwissenschaften 1977, 64, 536 – 537.
[14] P. Hollander, Drugs 1992, 44, 47 – 53.
vece, A. Di Mola, M. M. Cavalluzzi, P. Tortorella,
F. Loiodice, G. Iaccarino, P. Campiglia, E. Novellino,
C. Franchini, Synthesis 2010, 24, 4193 – 4198.
[29] D. Mauleon, P. M. Dolors, G. Rosell, Eur. J. Med.
Chem 1988, 23, 421 – 426.
[15] A. J. Krentz, C. J. Bailey, Drugs 2005, 65, 385 – 411.
[16] L. J. Scott, C. M. Spencer, Drugs 2000, 59, 521 – 549.
[17] L. K. Campbell, D. E. Beker, R. K. Campbell, Annals of
Pharmacotherapy 2000, 34, 1291 – 1301.
[18] F. A. van de Laar, Vasc. Health Risk Manag. 2008, 4,
1189 – 1195.
[19] N. Kiyotaka, K. Hiroyuki, K. Toshiyuki, S. Yamashita,
T. Tsuzuki, S. Oikawa, T. Miyazawa, J. Agric. Food
Chem. 2007, 55, 8928 – 8933.
[30] C. X. Kuang, F. F. Hu, CN 102225907, 2011.
[31] S. Matsumura, H. Enomoto, Y. Aoyagi, JP 54106477,
1979.
[32] N. Boechat, V. F. Ferreira, S. B. Ferreira, M. de Lour-
des G. Ferreira, F. de C. da Silva, M. M. Bastos, M. dos
S. Costa, M. C. Lourenc¸o, A. C. Pinto, A. U. Krettli,
A. C. Aguiar, B. M. Teixeira, N. V. da Silva, P. R. Mar-
tins, F. A. Bezerra, A. L. Camilo, G. P. da Silva, C. C.
Costa, J. Med. Chem. 2011, 54, 5988 – 5999.
[20] B. G. Pistia, R. I. Hollingsworth, Tetrahedron 2001, 57, [33] Q. Yang, Y. Zhang, D. Yu, F. F. Hu, C. X. Kuang, CN
8773 – 8778. 102166211, 2011.
[21] J. Shinde, T. Taldone, M. Barletta, N. Kunaparaju, [34] N. Almirante, S. Cicardi, C. Napoletano, M. Serravalle,
B. Hu, S. Kumar, J. Placido, S. W. Zito, Carbohydr.
Res. 2008, 343, 1278 – 1281.
Tetrahedron 1987, 43, 625 – 630.
Brought to you by | Florida State University Library
Authenticated
Download Date | 9/9/15 3:48 PM