Bioorganic & Medicinal Chemistry Letters 27 (2017) 40–44

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry Letters

journal homepage: www.elsevier.com/locate/bmcl

Anthraquinones from Morinda longissima and their insulin mimetic

activities via AMP-activated protein kinase (AMPK) activation
Phi-Hung Nguyen a,1, Hong Seok Choi a, Thi Kim Quy Ha b, Ji Yeon Seo b, Jun-Li Yang b, Da-Woon Jung c,
Darren R. Williams c, Won-Keun Oh b,⇑
a
College of Pharmacy, Chosun University, Gwangju 501-759, Republic of Korea
b
Korea Bioactive Natural Material Bank, Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University, Seoul 08826, Republic of Korea
c
New Drug Targets Laboratory, School of Life Sciences, Gwangju Institute of Science and Technology, 1 Oryong-Dong, Buk-Gu, Gwangju 500-712, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: AMP-activated protein kinase (AMPK) activators are known to increase energy metabolism and to reduce
Received 28 August 2016 body weight, as well as to improve glucose uptake. During for searching AMPK activators, a new anthra-
Revised 19 October 2016 quinone, modasima A (10), along with eighteen known analogues (1–9 and 11–19) were isolated from an
Accepted 14 November 2016
ethanol extract of the roots of Morinda longissima Y. Z. Ruan (Rubiaceae). Using the fluorescent tagged
Available online 15 November 2016
glucose analogues, 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxy-D-glucose (2-NBDG), insu-
lin mimetics were screened with compounds 1–19 in 3T3-L1 adipocytes. Among them, compounds 2,
Keywords:
8 and 10 enhanced significantly glucose uptake into adipocytes and up-regulated the phosphorylated
Morinda longissima
Anthraquinone
AMPK (Thr172) whereas the glucose uptake enhancing activities of compounds 2, 8 and 10 were abrogated
AMP-activated protein kinase (AMPK) by treatment of compound C, an AMPK inhibitor. Taken together, these anthraquinones showed the
Glucose uptake potential action as insulin mimetic to improve glucose uptake via activation of AMPK.
Ó 2016 Elsevier Ltd. All rights reserved.

The number of patients with type 2 diabetes (T2D) has been
increasing worldwide due to the upsurge in obesity that was
caused by sedentary lifestyles and high calorie foods.1 The AMP-
activated protein kinase (AMPK), a heterotrimeric functional
enzyme complex, plays a key role in metabolic control and regulat-
ing cellular energy homeostasis.2 The AMPK activation on the
phosphorylation of Thr172 at the catalytic a subunit3 can be regu-
lated by calmodulin-dependent protein kinase kinases (CaMKKs)
and serine/threonine kinase (LKB1), which are two upstream
kinases, and also can be positively influenced by the intracellular
AMP/ATP ratio.4 The activated AMPK is able to regulate lipid home-
ostasis, and this function has been regarded as a promising strat-
egy for the treatment of type 2 diabetes due to the strong
association between the deregulation of the fatty acid metabolism
and the development of insulin resistance.5 In addition, it has been
reported the potential of some downstream targets of AMPK in
glucose uptake regulation, such as reducing ectopic lipid accumu-
lation for improving insulin sensitivity, suppressing hepatic
glucose production in the liver, and stimulating glucose uptake in
the muscle.6 Insulin is the only available agent for both type 1
⇑ Corresponding author.
E-mail address: wkoh1@snu.ac.kr (W.-K. Oh).
1
Present address: Institute of Natural Products Chemistry, Vietnam Academy of
Science and Technology (VAST), 18-Hoang Quoc Viet, Cau Giay, Ha Noi, Viet Nam.
and type 2 diabetes treatment so far.7 Some synthetic small mole-
cules and natural products have been reported to mimic the action
of insulin and promote glucose uptake in cell culture and animal
models of diabetes.8 Therefore, it is a need to search for new
anti-diabetic agents that can mimic insulin.
The genus Morinda (Rubiaceae) comprising about 80 species
was found in tropical regions worldwide, including vines, shrubs,
or trees with aggregated fruits being fleshy or dry. M. citrifolia
(Noni) is regarded as an important herb plant in Hawaiian, and
has been utilized throughout the Pacific Polynesia for a long time.9
M. longissima among these species is closely related to M. citrifolia
in phytotaxonomy,10 but there are not many reports on phyto-
chemical and biological investigations of this plant. 2-NBDG assay
suggested that an ethanol extract of the roots of M. longissima
could improve glucose uptake in 3T3-L1 adipocytes. This observa-
tion and a recent discovery of four potential insulin mimetics from
M. citrifolia promoted us to search new insulin mimetics from
M. longissima.11 Herein we will describe the isolation and structure
elucidation of these constituents and their potential on glucose
uptake and their biological effects on AMPK activation.
Bioassay-guided fractionation of the active ethanol extract of
the roots of M. longissima by column chromatography (silica gel,
RP-C18, and prep-HPLC) resulted in the isolation of 19 anthraqui-
none derivatives (1–19), including a new that were named moda-
sima A (10) (Fig. 1).12

http://dx.doi.org/10.1016/j.bmcl.2016.11.034
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.
 P.-H. Nguyen et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 40–44
O R1
R2
R3
O O
7 R1 = R 2 = OCH3
1 R1 = OH, R2 = R 3 = H
13
2 R1 = H, R 2 = CH3, R3 = H
16
3 R1 = R 3 = OCH3 , R 2 = CH3
4 R1 = OH, R2 = CH3 , R 3 = H
5 R1 = OH, R2 = CH3 , R 3 = OCH3
6 R1 = R 2 = OH, R 3 = OCH 3
11 R1 = R 3 = OH, R 2 = CH 3
12 R1 = H, R 2 = CH 3, R3 = OH
14 R1 = H, R 2 = CH3, R3 = OCH 3
17 R1 = OCH 3, R2 = R 3 = OH
9 R 1 = R2 = OH, R3 = CH2 CH2 CH2CH 3
18 R1 = OCH 3, R2 = CH3 , R 3 = OH
15 R 1 = OCH3 , R 2 = OH, R 3 = CH2CH3
19 R1 = OCH 3, R2 = CH2 OH, R 3 = OH
Fig. 1. Chemical structures of compounds 1–19 from the roots of Morinda longissima.
Compound 10 was isolated as a yellowish amorphous powder.
Its molecular formula of C21H22O5 was determined from a HREIMS
peak at m/z 354.1465 [M]+ (calcd for C21H22O5, 354.1467). The IR
spectrum suggested the presence of hydroxy (3331 cm
1), unsatu-
rated carbonyl (1670 cm
1), and olefinic (1565 cm
1) functionali-
ties. The UV absorptions at kmax 242, 281, 336, and 352 nm and
the 13C NMR signals at dC 181.4 and 183.9 implied a 9,10-anthra-
quinone structure.13 The 1H NMR spectrum showed two aromatic
broad doublets at dH 8.17 (1H, br d, J = 7.6 Hz, H-5) and 8.23 (1H,
br d, J = 7.2 Hz, H-8), and two broad triplets at dH 7.86 (1H, br t,
J = 7.6 Hz, H-6) and 7.89 (1H, br t, J = 7.2 Hz, H-7), which indicated
a typical A2B2 aromatic system. The above observations together
with an aromatic singlet proton at dH 7.59 (1H, s, H-4) indicated
that 10 was an anthraquinone with a tri-substituted C-ring. A phe-
nolic hydroxyl group demonstrated a proton signal at dH 13.2 in the
1
H NMR spectrum measured in CDCl3, which was due to the forma-
tion of a hydrogen bond between this hydroxyl group and C(9)@O
group. This analysis indicated that a hydroxyl group was located at
the peri position to the C-9 carbonyl group.14 An oxygenated
methylene (dH 4.76, 2H, s, H2-10 ) was bond to C-2 based on HMBC
correlations from H-10 (dH 4.76) to C-1 (dC 163.9), C-2 (dC 126.8),
and C-3 (dC 163.5). In addition, a 1H-1H COSY experiment showed
an isopentyloxy group [dH 3.63 (2H, t, J = 6.8 Hz), 1.51 (1H, q,
J = 6.4 Hz), 1.72 (1H, m), and 0.90 (6H, d, J = 6.4 Hz)], and it was
connected to C-10 from HMBC correlation between H-10 (dH 4.76)
and C-20 (dC 70.4). One methoxy group at dH 3.94 (3H, s) was
located at C-3 due to a cross peak between this methoxy
(dH 3.94) and C-3 (dC 163.5) in HMBC experiment (Fig. 2).
The above conclusion was also supported by the HMBC correla-
tions from the aromatic singlet H-4 (dH 7.59, 1H) to C-2 (dC 126.8),
C-3 (dC 163.5), C-10 (dC 183.9), C-13 (dC 119.8), and C-14 (dC 137.7)
(Fig. 2). Thus, compound 10 was elucidated as 1-hydroxy-2-
((isopentyloxy)methyl)-3-methoxy-9,10-anthraquinone, and
named modasima A.15
By detailed analysis and comparison of the spectroscopic and
optical rotation values with literature data, the known compounds
were characterized to be 1-hydroxy-anthraquinone (1),16 tecto-
quinone (2),17 rubiadin-dimethyl ether (3),17 1-hydroxy-2-
O OH
O microscopy. Compared to control group (treated with DMSO), com-
OCH 3
O these fluorescence intensities were as much as insulin (0.1 lM).
10
Fig. 2. Key HMBC ( 1 1
) and H– H COSY ( ) correlations for modasima A (10).
41
O R1 O OCH3
O O
R2 O
O
8
R1 = OCH 3, R2 = OH
R1 = R 2 = OH
5'
O OH
8 1' 2'
12 13 1
O R1 9 O
A C 6'
10
OR 3 5
11 14
4
OCH3
O
R2 10
O
methyl-9,10-anthraquinone (4),17 rubiadin-3-methyl ether (5),17
1,2-dihydroxy-3-methoxy-anthraquinone (6),17 1,3-dimethoxy-2-
methoxymethylanthraquinone (7),17 1-methoxy-20 ,20 -dimethyl-
dioxine-(50 ,60 :2,3)-anthraquinone (8),18 lucidin-x-butyl ether
(9),18 rubiadin (11),13 3-hydroxy-2-methylanthraquinone (12),13
1-methoxy-3-hydroxy-2-methoxymethylanthraquinone (13),13
2-methoxy-3-methyl-anthraquinone (14), 19
damnacanthol-x-
ethyl ether (15),20 lucidin-x-methyl ether (16),20 1-methoxy-2,3-
dihydroxy-anthraquinone (17),20 rubiadin-1-methyl ether (18),20
and damnacanthol (19).20
All of the isolates (1–19) were evaluated for their potential on
glucose metabolism and insulin mimetic action. 2-NBDG has been
used as a fluorescent glucose analog widely used for monitoring
glucose uptake in cells, and it was established to be a useful
reagent for discovering insulin mimetic compounds.21 Herein we
examined the stimulatory effects of compounds 1–19 on 2-NBDG
uptake in 3T3-L1 adipocyte cells by using an in vitro assay (2-NBDG
assay).22 The 3T3-L1 adipocytes were chosen for this assay as fat
represents one of the major body tissues that is sensitive to insulin.
3T3-L1 fibroblasts were induced to differentiate into adipocytes,23
and then isolated compounds were treated to the differentiated
3T3-L1 adipocytes with 2-NBDG.11,24 DMSO and insulin (01. lM)
were used as negative and positive controls in this assay,
respectively.
As shown in Fig. 3A, anthraquinones 2, 7–10, 12–14 and 17–18
significantly enhanced 2-NBDG uptake at a concentration of
10 lM. Compared with insulin, compounds 2 and 8 showed the
most potent stimulatory activities. Detailed analysis of structure–
activity relationships (SARs) of anthraquinones 1–19 indicated
the key role of the substitution at C-10 in their stimulatory effects
on glucose uptake. Attachment of ether bond at C-10 gave the most
important positive influence on the 2-NBDG uptake, such as com-
pounds 8–10. There are no functional groups attached to C-10 in
compounds 2–5, 11, 12, 14 and 18. Among them, compound 2
showed the strongest stimulatory effect on 2-NBDG uptake. This
observation indicated that more electronegative functionalities
(OH/OMe) in 9,10-anthraquinones may be responsible for decreas-
ing the stimulation on glucose uptake.
To confirm the transportation efficacy of 2-NBDG into the cells
by compounds 2, 8 and 10, we further measured fluorescent sig-
nals in adipocytes after compound treatment using fluorescence
pounds 2, 8 and 10 showed higher intensity of fluorescent signals
from the adipocytes at a concentration of 10 lM (Fig. 3B), and
The results indicated the potential of compounds 2, 8 and 10 to
improve glucose uptake.
42 P.-H. Nguyen et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 40–44

The translocation of glucose transporter GLUT4 is essential for
inducible glucose uptake into the plasma membrane at muscle
and adipocyte. This process mainly depends on the regulation of
two physiological pathways: the AMP-activated protein kinase
(AMPK) pathway and the insulin signaling pathway.25 In order to
investigate the underlining mechanism for the stimulation of
2-NBDG uptake into the cells by the anthraquinones from M.
longissima, we further evaluated the effects of compounds 2, 8
and 10 on the AMPK pathway due to the key role of this pathway
in glucose uptake regulation during exercise or in response to some
anti-diabetic agents such as metformin and AICAR.25 To investigate
the effects of compounds on AMPK activation, we measured the
phosphorylation of AMPK on differentiated mouse C2C12 skeletal
myoblasts.26 The results of Western blot analysis27 revealed that
compounds 2, 8 and 10 significantly stimulated the phosphoryla-
tion of AMPK at a concentration of 10 lM, whereas the total AMPK
expression was not changed (Fig. 4A). In addition, increased phos-
phorylation of AMPK induced by compounds 2, 8 and 10 was abro-
gated by pretreatment with compound C, an AMPK inhibitor
(Fig. 4B). The activated AMPK is responsible for metabolic control
via phosphorylation of acetyl-CoA carboxylase (ACC), a down-
stream substrate of AMPK, and this function is strongly associated
with fatty acid oxidation. Thus, the phosphorylated ACC was mea-
sured by binding with phosphorylated ACC antibody for under-
standing the AMPK action in ACC phosphorylation by Western
blot analysis. As shown in Fig. 4A, compounds 8 and 10 activated
phosphorylation of ACC, which were comparable to AICAR used
as a positive control in this assay. The results suggested that these
anthraquinones improved the glucose uptake by regulating the
AMPK signaling pathway. As activated AMPK is able to inhibit lipid
synthesis, increase fatty acid oxidation, and improve insulin action,
AMPK pathway was suggested as a likely mechanism for the
glucose uptake stimulation by anthraquinones from M. longsis-
sima.28 The MTT assay was performed to test the cytotoxicity of
active compounds 8 and 10 using the differentiated 3T3-L1 and
C2C12 cells, respectively. The results demonstrated that com-
pounds 8 and 10 had no cytotoxicity at a concentration of 10 lM
for not only 2 h but also 24 h in C2C12 cells. In the case of differen-
tiated 3T3-L1 adipocytes, the results were shown similar tendency
in C2C12 cells (see Supplementary Fig. S6). For the determination
whether anthraquinones themselves have influence on 2-NBDG
uptake, the fluorescence at excitation 450 nm and emission
535 nm on cells were examined with compounds 8 and 10 at the
same condition with or without 2-NBDG treatment. The fluores-
cence intensity was no significant changes in 3T3-L1 adipocytes
for indicating that anthraquinones themselves do not effect on
fluorescence excitation of 2-NBDG uptake assay (see Supplemen-
tary Fig. S7).
A number of 9,10-anthraquinones have been discovered from
nature and a variety of biological activities have reported for them,
including antioxidant, anticancer, anti-inflammatory, liver-protec-
tive, cytotoxic, and immunomodulatory effects.29 Emodin, a natu-
ral anthraquinone compound, increased AMPK activity at a much
lower concentration than metformin, and also showed
enhanced-GLUT4 translocation into the cells of myotube.
Danthron from rhubarb was also reported on its AMPK activation
in both HepG2 and C2C12 cells.30 However, there are not many
reports regarding AMPK activities of anthraquinone derivatives
and especially their SAR. In this study, some 9,10-anthraquinone
derivatives isolated from M. longissima showed promising biologi-
cal effects on the glucose uptake and stimulation of AMPK
enzyme. Further investigation and optimization of these deriva-
tives may enable the discovery of new drug candidates for the

Fig. 3. Effects of anthraquinones presented in M. longissima on glucose uptake in 3T3-L1 adipocytes. (A) Stimulatory effects of anthraquinones 1–19 on 2-NBDG uptake in
3T3-L1 adipocytes were measured by using a fluorescent glucose probe (2-NBDG). The test compounds (1–19) were treated at a concentration of 10 lM to the cells for
90 min. DMSO and insulin (0.1 lM) were treated to the cells as negative and positive controls, respectively. (B) Increment of 2-NBDG uptake in 3T3-L1 adipocytes by
compounds 2, 8 and 10 at a concentration of 10 lM for 90 min. Representative pictures were taken after compound treatment. Increased green fluorescent signals in the cells
were observed by the compound treatment indicating that 2-NBDG was transported into the cells.
 P.-H. Nguyen et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 40–44
Fig. 4. Regulatory effects of compounds 2, 8 and 10 on the expressions of pAMPK
(Thr172) and pACC (Ser79) in C2C12 cells. (A) Stimulatory effects of compounds 2, 8
and 10 on phosphorylated AMPK were determined by Western blot analysis in
C2C12 cells. C2C12 cells were exposed to test compounds for 2 h or an AMPK
activator (AICAR, 1 mM) for 1 h, and phosphorylation of AMPK at Thr172 and ACC at
Ser79 were analyzed by Western blotting. (B) C2C12 cells were exposed to test
compounds at 10 lM for 2 h and the expression of phosphorylated AMPK (Thr172)
was analyzed by Western blotting. The compound C, an AMPK inhibitor, was
treated to the cells at 10 lM for 10 min.
treatment of type-2 diabetes, obesity, and metabolic disorders by
targeting AMPK.
Acknowledgements
This work was supported in part by grants from the Marine
Biotechnology Program of the Ministry of Oceans and Fisheries
(PJT200669) and the Korea Bioactive Natural Material Bank (NRF-
2012M3A9B8021570) of the National Research Foundation of
Korea (NRF), which is funded by the Korean government.
A. Supplementary data
1D (1H and 13C) and 2D (1H-1H COSY and HMBC) NMR spectra
for modasima A (10). These materials are available free of charge
on the Internet at http://www.elsevier.com. Supplementary data
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.bmcl.2016.11.034.
43
References
1. Narkar VA, Downes M, Yu RT, et al. Cell. 2008;134:405.
2. Hardie DG, Carling D. Eur J Biochem. 1997;246:259.
3. Stein SC, Woods A, Jones NA, Davison MD, Carling D. Biochem. J.. 2000;345:437.
4. Misra P. Expert Opinion Ther. Targets. 2008;12:91.
5. Carling D. Trends Biochem Sci. 2004;29:18.
6. (a) Viollet B, Mounier R, Leclerc J, Yazigi A, Foretz M, Andreelli F. Diabetes Metab.
2007;33:395;
(b) Hardie DG. Annu Rev Pharmacol Toxicol. 2007;47:185.
7. Umpierrez GE, Palacio A, Smiley D. Am J Med. 2007;120:563.
8. Zhang B, Salituro G, Szalkowski D, et al. Science. 1999;284:974.
9. Deng S, Palu AK, West BJ, Su CX, Zhou BN, Jensen JC. J Nat Prod. 2007;70:859.
10. Razafimandimbison SG, McDowell TD, Halford DA, Bremer B. Mol Phylogenet
Evol. 2009;52:879.
11. Nguyen PH, Yang JL, Uddin MN, et al. J Nat Prod. 2013;76:2080.
12. Roots of M. longissima were collected in Cao Bang province, North of Vietnam.
The sample was authenticated by Professor Won Keun Oh and a voucher
specimen (No. 0012) was deposited at the College of Pharmacy, Seoul National
University, Seoul, Korea. The dried roots (1 kg) of M. longissima were extracted
with ethanol at 50 °C using sonication (1 day  3). After removing the solvent,
the residue (45 g) was chromatographed on a silica gel column (10  60 cm;
63–200 lm particle size) using a stepwise gradient of n-hexane:EtOAc:n-BuOH
(15:1:0.1 to 0:1:1) to yield five fractions (Fr.1–Fr.5) according to their TLC
profiles. Fraction 2 (Fr.2, 5.4 g) was further subjected to a silica gel open
column (6.0  60 cm; 40–63 lm particle size) eluted with n-hexane/acetone
(20:1 to 1:5) to afford five sub-fractions (Fr.2-1 to Fr.2-5) and compound 7
(55.3 mg). Purification of Fr.2-2 (53 mg), by a silica gel column (2.0  60 cm,
40–63 lm particle size) using CHCl3:MeOH (30:1), afforded compounds 1
(5.0 mg) and 2 (3.6 mg). Fr.2-3 (281 mg) was purified by prep-HPLC (mobile
phase, 70% MeCN in H2O containing 0.1% formic acid; flow rate, 2 mL/min; UV
detection at 205 and 254 nm), resulted in the isolation of compounds 5
(4.7 mg; tR = 26.8 min), 6 (32.5 mg; tR = 20.5 min), and 17 (2.8 mg;
tR = 18.1 min). Compounds 3 (8.2 mg) and 4 (23.6 mg) were purified from
Fr.2-4 (213 mg) using a silica gel column (3.0  80 cm, 40–63 lm particle size),
eluted with CHCl3:MeOH (20:1). Fraction 3 (Fr.3, 14 g) was also subjected to a
silica gel open column (6.0  60 cm; 40–63 lm particle size) eluted with n-
hexane/acetone (10:1 to 1:10) to afford compound 8 (15 mg) and five sub-
fractions (Fr.3-1 to Fr.3-5). Further purification of Fr.3-2 (324 mg) by prep-
HPLC (mobile phase 85% MeOH in H2O containing 0.1% formic acid; flow rate,
2 mL/min; UV detections at 205 and 254 nm) yielded compounds 12 (3.7 mg,
tR = 14.5 min), 9 (1.8 mg; tR = 17.8 min), 10 (10.5 mg, tR = 20.5 min), and 11
(12.1 mg, tR = 23.9 min). Similarly, Fr.3-3 (500 mg) was also purified by prep-
HPLC, using an isocratic solvent system of 70% MeOH in H2O containing 0.1%
formic acid, to give compounds 13 (4.2 mg, tR = 34.5 min), 14 (15.8 mg;
tR = 43.3 min), 15 (98.5 mg, tR = 47.5 min), and 16 (6.5 mg, tR = 55.1 min).
Compounds 18 (62.1 mg) and 19 (50 mg) were isolated from Fr.3-4 (430 mg)
by repeated column chromatography over RP-C18 (4.5  50 cm; 150 lm
particle size) using MeOH/H2O system (1.5:1 to 2:1).
13. Kitajima M, Fischer U, Nakamura M, et al. Phytochemistry. 1998;48:107.
14. Fraga BM, Quitana N, Diaz CE. Chem Biodivers. 2009;6:182.
15. Yellowish, amorphous powder; UV (MeOH) kmax (log e): 242 (4.59), 281 (4.50),
336 (3.73), 352 (3.58) nm; IR (KBr) mmax 3331, 2927, 1670, 1565, 1299,
1109 cm
1; HREIMS m/z 354.1465 [M]+, (calcd for C21H22O5, 354.1467); 1H
NMR (acetone-d6, 400 MHz): 7.59 (1H, s, H-4), 8.17 (1H, br d, J = 7.6 Hz, H-5),
7.86 (1H, br t, J = 7.6 Hz, H-6), 7.89 (1H, br t, J = 7.2 Hz, H-7), 8.23 (1H, br d,
J = 7.2 Hz, H-8), 4.76 (2H, s, H-10 ), 3.63 (2H, t, J = 6.8 Hz, H-20 ), 1.51 (2H, q,
J = 6.4 Hz, H-30 ), 1.72 (1H, m, H-40 ), 0.90 (6H, d, J = 6.4 Hz, H-50 , H-60 ), 3.94 (3H,
s, 3-OCH3); 13C NMR (acetone-d6, 100 MHz): 163.9 (C-1), 126.8 (C-2), 163.5 (C-
3), 111.6 (C-4), 128.1 (C-5), 135.7 (C-6), 134.5 (C-7), 127.5 (C-8), 181.4 (C-9),
183.9 (C-10), 136.5 (C-11), 133.9 (C-12), 119.8 (C-13), 137.7 (C-14), 64.0 (C-10 ),
70.48 (C-20 ), 39.8 (C-30 ), 26.2 (C-40 ), 23.3 (C-50 ), 23.3 (C-60 ), 63.5 (3-OCH3).
16. Sato T, Fukuyama T, Yamada M, Suzuki T. J Biochem. 1956;43:21.
17. Chiriboga X, Gilardoni G, Magnaghi I, Finzi PV, Zanoni G, Vidari G. J Nat Prod.
2003;66:905.
18. Thao PT, Leuyen BTT, Lee SH, Jang HD, Kim YH. Nat Prod Sci. 2016;22:13.
19. El-Hady S, Bukuru J, Kesteleyn B, Van Puyvelde L, Van TN, De Kimpe N. J Nat
Prod. 2002;65:1377.
20. Kamiyaa K, Hamabea W, Tokuyamaa S, et al. Food Chem. 2010;118:725.
21. Lee J, Jung DW, Kim WH, et al. ACS Chem Biol. 2013;8:1803.
22. (a) Jung DW, Ha HH, Zheng X, Chang Y-T, Williams DR. Mol BioSyst. 2011;7:346;
(b) Millon SR, Ostrander JH, Brown JQ, Raheja A, Seewaldt VL, Ramanujam N.
Breast Cancer Res Treat. 2011;126:55.
23. Differentiation of 3T3-L1 to adipocytes: 3T3-L1 pre-adipocytes were maintained
for 48 h in Dulbecco’s modified Eagle’s medium (DMEM, Hyclone, GE
Healthcare Life Sciences, Little Chalfont, UK) supplemented with 100 U/mL of
penicillin (Hyclone) and 100 lg/mL of streptomycin (Hyclone) and 10% calf
serum (Hyclone) in an atmosphere of 5% CO2 at 37 °C. To differentiate 3T3-L1
pre-adipocytes to adipocytes, the medium was replaced to fresh differentiating
medium [DMEM supplemented with 1 lg/mL insulin (Roche Life Science, Basel,
Switzerland), 0.5 mM 3-isobutyl-1-metylzanthine (Sigma-Aldrich, St. Louis,
Missouri, USA), 2 lg/mL dexamethasone (Sigma-Aldrich), 10% fetal bovine
serum (FBS, Hyclone), 1 lM rosiglitazone (Sigma-Aldrich), 100 U/mL of
penicillin (Hyclone), and 100 lg/mL of streptomycin (Hyclone)] every 2 days.
44 P.-H. Nguyen et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 40–44
After observation of the differentiated adipocytes at 7th day, the cells were
used for assays.
24. Measurement of glucose uptake: The 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)
amino)-2-deoxy-D-glucose (2-NBDG assay was performed following as
reported in a previous paper with slight modifications.11 3T3-L1 cells were
seeded into 24-well tissue culture plate (BD, Falcon, NJ, USA) at a density of
3  104 cells per each well. Cells were in glucose-free culture media supple-
mented with 10% FBS and 100 U/mL of penicillin and 100 lg/mL of strepto-
mycin for 2-NBDG in the presence or absence of test compounds. The media
was replaced to DMEM containing 50 lM 2-NBDG after 30 min. After 1 h,
adipocytes were washed twice using pre-warmed phosphate-buffered saline
(PBS) and lysed by treating with 70 lL of 0.1 M K3PO4, and 1% Triton X-100 in
PBS for 10 min in the dark condition. 30 lL of DMSO was added, and the thus
formed solution was mixed. The fluorescence intensity of NBDG was recorded
on a VICTOR X3 multi-label plate reader (PerkinElmer, MA, USA) at excitation
450 nm and emission 535 nm.
25. Huang S, Czech MP. Cell Metab. 2007;5:237.
26. (a) Nguyen PH, Nguyen TN, Dao TT, et al. J Nat Prod. 2010;73:598;
(b) Nguyen PH, Gauhar R, Hwang SL, et al. Bioorg Med Chem. 2011;19:6254.
27. Western blot analysis for pAMPK and pACC: Mouse C2C12 skeletal cells were
maintained in DMEM supplemented with 10% FBS under 5% CO2 at 37 °C. The
cells were seeded into 12-well plates (BD, Falcon, NJ, USA) at a density of 105
cells per a well diluted to 2 mL of growth medium (DMEM supplemented with
10% FBS. When the cell confluency was approximately 95%, the growth
medium was replaced to DMEM supplemented with 5% horse serum. The
medium was changed at interval of 48 h. After incubation with the test
compounds for 30 min, the C2C12 cells were lysed by using an EBC lysis buffer
containing 50 mM NaF, 0.5% nonidet-P40 (NP-40), 1 mM EDTA (pH 8.0),
120 mM NaCl, and 50 mM Tris-HCl (pH 7.6). By centrifugation for 15 min at
12,000 rpm at 4 °C, the cell debris was cleaned. A Bio-rad protein assay kit (Bio-
rad, CA, USA) was used to calculate the concentrations of proteins in the cell
lysates. Western blot analysis was carried out via anti-phosphospecific ACC
Ser79 and anti-phosphospecific AMPKa Thr172 (Cell signaling, MA, USA) by
using about 30 lg of proteins. By using an enhanced chemiluminescence
detection kit (Thermo Scientific, MA, USA) and a horseradish peroxidase-
labeled anti-rabbit IgG immunoreactive antigen (Thermo Scientific) was
determined.
28. Musi N. Curr Med Chem. 2006;13:583.
29. (a) Deng S, West BJ, Jensen CJ. Food Chem. 2010;122:267;
(b) Lv L, Chen H, Ho CT, Sang S. Fitoterapia. 2011;82:704;
(c) Wang MY, Nowichki D, Anderson G, Jensen J, West B. Plant Foods Hum Nutr.
2008;63:59.
30. (a) Song P, Kim JH, Ghim J, et al. J Biol Chem. 2013;288:5732;
(b) Liu J, Chen Z, Zhang Y, et al. Diabetes. 2013;62:3927;
(c) Yang Y, Yan YM, Wei W, et al. Bioorg Med Chem Lett. 2013;23:3905;
(d) Zhou R, Wang L, Xu S, Chen J, Hu LH, Shen X. Acta Pharmacol Sin.
2013;34:1061.