Acylated dolabellane-type diterpenes from Nigella sativa seeds with triglyceride metabolism-promoting activity in high glucose-pretreated HepG2 cells

Article abstract of DOI:10.1016/j.phytol.2013.01.004

Two new acylated dolabellane-type diterpenes, nigellamines B₃ (9) and D (10), were isolated from Nigella sativa (Ranunculaceae) seeds using column chromatography and preparative HPLC. Their structures were determined based on chemical and physi

Full text of DOI:10.1016/j.phytol.2013.01.004

Phytochemistry Letters 6 (2013) 198–204
Contents lists available at SciVerse ScienceDirect
Phytochemistry Letters
journal homepage: www.elsevier.com/locate/phytol
Acylated dolabellane-type diterpenes from Nigella sativa seeds with triglyceride
metabolism-promoting activity in high glucose-pretreated HepG2 cells
Toshio Morikawa a, Kiyofumi Ninomiya a, Fengming Xu b, Naomichi Okumura a, Hisashi Matsuda b,
Osamu Muraoka a, Takao Hayakawa a, Masayuki Yoshikawa b,
*
a Pharmaceutical Research and Technology Institute, Kinki University, 3-4-1 Kowakae, Higashi-osaka, Osaka 577-8502, Japan
b Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto 607-8412, Japan
A R T I C L E I N F O
A B S T R A C T
Article history:
Two new acylated dolabellane-type diterpenes, nigellamines B3 (9) and D (10), were isolated from Nigella
sativa (Ranunculaceae) seeds using column chromatography and preparative HPLC. Their structures were
determined based on chemical and physicochemical evidence, and confirmed using previously isolated
related compounds as reference. Of the seed constituents, nigellamines A2 (2), A3 (3), A5 (5), B1 (6), and B2
(7) had in vitro triglyceride metabolism-promoting activities in the high glucose-pretreated human liver
carcinoma cell line, HepG2.
Received 25 September 2012
Received in revised form 17 January 2013
Accepted 26 January 2013
Available online 20 February 2013
Keywords:
ß 2013 Phytochemical Society of Europe. Published by Elsevier B.V. All rights reserved.
Nigellamine
Nigella sativa
Triglyceride metabolism promotor
Ranunculaceae
Dollabellane-type diterpene
Black cumin
1. Introduction
2. Results and discussion
The Ranunculaceae annual plant Nigella sativa L. is widely
distributed and cultivated in Arabic and Mediterranean countries.
The seeds of N. sativa (commonly known as ‘‘black cumin’’) are
used in many food preparations, in addition to being prescribed by
Arabian folk medicine practitioners for the treatment of asthma,
flatulence, polio, kidney stones, abdominal pain, etc. (El Sayed et al.,
2000; Enomoto et al., 2001). Previously, we reported that a
methanol extract of the seeds had a lipid metabolism-promoting
activity in primary cultured mouse hepatocytes (Morikawa et al.,
2004a,b). A re-examination on N. sativa to identify new activities
detected its triglyceride (TG) metabolism-promoting activity in
high glucose-pretreated human liver carcinoma cell line, HepG2.
Further separation of the active constituents in the extract allowed
us to isolate two new acylated dolabellane-type diterpenes,
nigellamines B3 (9) and D (10), in addition to the previously
reported nigellamines A1–A5 (1–5), B1 (6), B2 (7), and C (8). This
paper describes the isolation and structure elucidation of the new
diterpenes and the TG metabolism-promoting activity of the
diterpene constituents in high glucose-pretreated HepG2 cells.
2.1. Effect of the MeOH extract on liver TG content in mice
The seeds of N. sativa were powdered and extracted with
methanol to give a methanol extract (17.4% from the seeds). As
shown in Tables 1 and 2, the methanol extract significantly
reduced liver TG content by oral administration at doses of 250 and
500 mg/kg/day for 3 days without affecting food intake, body and
liver weights, and with no detectable toxic effects including
plasma glucose, TG, and free fatty acid (FFA) levels.
2.2. Effects on TG content in high glucose-pretreated HepG2 cells
Fatty liver is recognized as a significant risk factor for serious
liver disease (Bellentani et al., 1994; El-Hassan et al., 1992). There
is
a strong causal linkage between fatty liver disease and
hyperinsulinemic insulin resistance (Marceau et al., 1999;
Marchesini et al., 1999). Thus, fatty liver is considered to be
highly associated with obesity and type 2 diabetes (Marchesini
et al., 1999). During an exploratory study of the novel bioactive
functions of the natural medicine in N. sativa, we examined the
inhibitory effects of its methanol extract on the TG levels in high
glucose-pretreated HepG2 cells (Morikawa et al., 2004a,b). The
extract significantly reduced the TG levels in hepatocytes (% of
control at 86.7 ꢀ 5.2, p < 0.01 at 10 mg/mL). The methanol extract
* Corresponding author. Tel.: +81 75 595 4633; fax: +81 75 595 4768.
E-mail address: myoshika@mb.kyoto-phu.ac.jp (M. Yoshikawa).
1874-3900/$ – see front matter ß 2013 Phytochemical Society of Europe. Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.phytol.2013.01.004

T. Morikawa et al. / Phytochemistry Letters 6 (2013) 198–204
199
Table 1
Effects of the methanol extract of N. sativa seeds on food intake, visceral fat, liver weight, and liver TG content in mice.
Treatments
Dose
N
Body weight
1st day (g)
Food intake
Liver weight
(mg)
Liver TG
(mg/kg/day, p.o.)
(g/mouse/day)
(mg/g)
2nd day (g)
3rd day (g)
Control
–
7
7
7
31.2 ꢀ 0.4
30.9 ꢀ 0.2
30.9 ꢀ 0.4
32.1 ꢀ 0.6
32.1 ꢀ 0.4
31.9 ꢀ 0.3
32.5 ꢀ 0.7
32.2 ꢀ 0.3
32.0 ꢀ 0.3
4.8 ꢀ 0.2
5.0 ꢀ 0.2
5.3 ꢀ 0.3
1145 ꢀ 52
1225 ꢀ 28
1236 ꢀ 54
56 ꢀ 3
40 ꢀ 2*
42 ꢀ 1*
MeOH ext.
250
500
Each value represents the mean ꢀ S.E.M.
*
Significantly different from the control, p< 0.01.
was partitioned into a mixture of EtOAc and water to produce an
EtOAc-soluble fraction (10.1%) and an aqueous phase. The aqueous
phase was subjected to Diaion HP-20 column chromatography to
yield H2O- and MeOH-eluted fractions (2.6 and 4.7%, respectively).
The EtOAc-soluble fraction was active (% of control at 81.8 ꢀ 4.1,
p < 0.01 at 10 mg/mL), whereas the other fractions lacked activity
(H2O-eluted fraction: % of control at 95.6 ꢀ 2.9 at 10 mg/mL, MeOH-
eluted fraction: 93.3 ꢀ 5.5).
9a: dC 71.6) was observed upon the conversion. Treatment of
9a with 0.1% sodium methoxide (NaOMe)–MeOH at room
temperature furnished 9b, the spectroscopic properties of which
were completely in accord with those of an authentic specimen
(Morikawa et al., 2004a), together with methyl nicotinate and
methyl n-hexanonate (Fig. 2). The positions of the two nicotinic
acids and the n-hexanoic acid were determined on the basis of
HMBC experiments, in which long-range correlations were
observed between H-2 and C-70; H-10 and C-700; H2-15 and C-7000
(Figure S2). Thus the connected positions of these acyl groups in 9
2.3. Isolation and structure determination of nigellamines B3 (9) and
D (10)
were unambiguously clarified, and the structure of
elucidated.
9 was
The EtOAc-soluble fraction was subjected to SiO2 and ODS
column chromatographies and finally HPLC to furnish acylated
dolabellane-type diterpenes nigellamines B3 (9) and D (10)
together with nigellamines A1–A5 (1–5), B1 (6), B2 (7), and C (8)
(Fig. 1), and two monoterpenes, carvacrol and thymoquinol.
Carvacrol and thymoquinol were identified by comparison of their
spectroscopic properties with those of commercially obtained
Nigellamine D (10) was also obtained as a white powder with
23
negative optical rotation ([
a]
ꢁ20.5 in CHCl3). The molecular
D
weight (m/z 666 [M+]) determined by the EIMS was sixteen units
more than that of nigellamine A2 (2) (m/z 650 [M+]), and the
molecular formula was determined to be C39H42N2O8 by the
HREIMS. The proton and carbon signals in the 1H and 13C NMR
spectra of 10 were superimposable on those of 2, except for the
samples. From the MeOH-eluted fraction, two saponins, 3-O-
rhamnopyranosyl-(1!2)- -arabinopyranosylhederagenin [28-
O- -glucopyranosyl-(1!6)-
-rhamnopyranosyl-(1!4)-
glucopyranosyl] ester (Shimizu et al., 1978) and 3-O- -xylopyr-
anosyl-(1!3)- -rhamnopyranosyl-(1!2)- -arabinopyrano-
sylhederagenin [28-O-
-rhamnopyranosyl-(1!4)-
-glucopyranosyl] ester (Ansari et al.,
a
-
L
-
signals due to the 20-position [d 4.10 (2H, s)]. In the HMBC
a
-L
experiment on 10, long-range correlations were observed between
the 20-proton and the C-12/C-18/C-19 as shown in Figure S2. The
geometric structure of the olefinic parts and relative stereostruc-
ture of 10 was characterized by the NOESY experiment as shown in
Figure S2. The CD spectrum of 10, which showed a negative Cotton
effect at 261 nm (De ꢁ1.73 in MeOH), was similar to that of 2
[256 nm (De ꢁ1.42) in MeOH], so that the absolute stereochemis-
try was proved to be the same as that of 2.
a
-
L
b
-
D
b-D-
b
-D
a
-L
a-L
a-
L
b-D-
glucopyranosyl-(1!6)-
b-D
1988) were isolated by SiO2 and ODS column chromatographies
and finally HPLC (Figure S1).
Nigellamine B3 (9) was obtained as a white powder with
27
D
positive optical rotation ([
a]
+33.4 in CHCl3). The positive-ion
2.4. Effects on nigellamines (1–9) on TG content in high glucose-
FABMS showed the quasimolecular ion peak at m/z 677 [M+H]+,
and the molecular formula was determined as C38H48N2O9 by
HRFABMS measurement. The 1H and 13C NMR (Table 3) spectro-
scopic properties, which were assigned with the aid of DEPT, 1H–1H
COSY, HMQC, and HMBC experiments, were quite similar to those
of nigellamine B2 (7) (Morikawa et al., 2004a) except for signals
due to a n-hexanoyl group instead of a benzoyl group at the 15-
pretreated HepG2 cells
We examined the inhibitory effects of nigellamines (1–9) on the
TG levels in high glucose-pretreated HepG2 cells. As shown in
Table 4, 2 (% of control at 10
(81.0 ꢀ 2.5) significantly reduced the TG levels in hepatocytes and
these reductions were equivalent to those using the hypoglycemic
medicine, metformin (Bellentani et al., 1994; Marchesini et al., 1999)
(82.9 ꢀ 2.4). In addition, 6 (% of control at 30 mM: 82.2 ꢀ 7.4), 7
(75.5 ꢀ 6.1), carvacrol (86.5 ꢀ 1.9), and thymoquinol (84.6 ꢀ 3.8)
demonstrated weak activities. The subsequent structural require-
ments affected the activities of these types of diterpene: (1) those
m
M: 89.7 ꢀ 5.7), 3 (62.1 ꢀ 4.5), and 5
position:
d
0.86 (3H, t, J = 7.0 Hz, H3-6000), 1.27, 1.30, 1.68 (2H each,
all m, H2-5000, 4000, 3000), 2.45 (2H, t, J = 7.0 Hz, H2-2000). Treatment of 9
with triphenylphosphine (PPh3) gave a compound (9a) with
sixteen units less molecular weight than the reactant (9).
Characteristic downfield shift of a signal due to C-18 (9: dC 82.6,
Table 2
Effects of the methanol extract of N. sativa seeds on plasma biochemicals in mice.
Treatments
Dose
N
Triglyceride
(mg/dL)
Total cholesterol
(mg/dL)
Free fatty acid
(mequiv./L)
(mg/kg/day, p.o.)
Control
–
250
500
7
7
7
101.8 ꢀ 4.8
106.1 ꢀ 7.2
104.9 ꢀ 6.5
127.6 ꢀ 13.8
133.9 ꢀ 10.3
126.9 ꢀ 14.3
1.625 ꢀ 0.081
1.511 ꢀ 0.117
1.491 ꢀ 0.122
MeOH ext.
Each value represents the mean ꢀ S.E.M.

200
T. Morikawa et al. / Phytochemistry Letters 6 (2013) 198–204
Fig. 1. Structures of nigellamines (1–10).
with a 2-O-nicotinoyl group were more potent than those bearing a 2-
O-benzoyl group [nigellamine A2 (2) ꢂ nigellamine A1 (1); nigella-
mine B2 (7) > nigellamine B1 (6), respectively]; (2) the 15-O-n-
hexanoyl group enhanced activity [nigellamine A3 (3) > nigellamines
A2 (2) and A5 (5) ꢂ nigellamine A4 (4)]. Thus, these nigellamines (2, 3,
and 5–7) and two known monoterpenes were responsible for the fatty
liver preventive effects of the extract.
3. Experimental
3.1. General experimental procedures
The following instruments were used to obtain spectral and
physical data: specific rotations, Horiba SEPA-300 digital polarim-
eter (l = 5 cm); CD spectra, JASCO J-720WI spectrometer; UV
spectra, Shimadzu UV-1600 spectrometer; IR spectra, Shimadzu
FTIR-8100 spectrometer; 1H and 13C NMR spectra, JEOL JNM-LA500
(500 and 125 MHz) spectrometer with tetramethylsilane as an
internal standard; EIMS and high-resolution EIMS, JEOL JMS-
GCMATE mass spectrometer; FABMS and high-resolution FABMS,
JEOL JMS-SX 102A mass spectrometer; HPLC detector, Shimadzu
RID-6A refractive index and Shimadzu SPD-10A UV-VIS detectors.
HPLC column, YMC-Pack ODS-5-A (YMC Co., Ltd.) columns were
used for analytical (250 mm ꢃ 4.6 mm i.d.) and preparative
(250 mm ꢃ 20 mm i.d.) purposes, respectively.
2.5. Agonistic activity for PPAR
a
To clarify the mechanism of action of the active nigellamines,
the PPAR agonistic activity was examined using a nuclear
a
receptor cofactor assay system. As shown in Table 5, the
relative binding activity of the active nigellamines (2, 3, and
5–7) indicated
a weak activity. The detailed mechanisms
of action of these active nigellamines need to be studied
further.

T. Morikawa et al. / Phytochemistry Letters 6 (2013) 198–204
201
Table 3
1H and 13C NMR data on nigellamines B3 (9) and D (10) and related compounds (9a and 9b) in CDCl3.
9
10
9a
9b
d
(J in Hz)
dC
d
(J in Hz)
dC
d
(J in Hz)
dC
d
(J in Hz)
dC
1
2
3
4
5
5
6
6
7
8
9
9
54.5
56.1
54.4
56.3
5.49 (d, 10.4)
73.9
122.0
142.1
38.1
5.47 (d, 10.1)
73.5
123.1
140.5
37.8
5.43 (d, 10.5)
73.9
122.4
141.7
38.2
4.05 (d, 10.7)
72.3
127.6
137.3
38.1
5.61 (dd, 0.9, 10.4)
5.77 (dd, 1.2, 10.1)
5.61 (dd, 0.9, 10.5)
5.48 (dd, 0.9, 10.7)
a
b
a
b
2.36 (br d, ca. 13)
2.41 (ddd, 4.9, 12.8, 12.8)
1.66 (m)
2.38 (m)
2.36 (m)
2.28 (m)
2.50 (ddd, 5.2, 12.9, 12.9)
1.71 (m)
2.43 (ddd, 4.9, 12.9, 12.9)
1.67 (m)
2.37 (ddd, 4.6, 12.2, 12.2)
1.61 (m)
22.9
22.7
22.9
23.0
1.99 (m)
2.03 (m)
1.99 (m)
1.92 (m)
2.90 (br d, ca. 10)
66.7
58.8
40.9
3.08 (br d, ca. 10)
65.4
58.4
42.2
2.90 (br d, ca. 10)
66.7
59.1
41.3
2.83 (br d, ca. 10)
66.6
59.5
43.3
a
b
2.65 (dd, 5.8, 13.7)
1.52 (dd, 12.2, 13.7)
5.88 (br dd, ca. 6, 12)
2.79 (br s)
2.57 (dd, 5.5, 132.8)
1.67 (dd, 12.5, 13.8)
5.79 (br dd, ca. 6, 13)
2.75 (br s)
2.47 (dd, 5.8, 13.8)
1.64 (dd, 12.5, 13.8)
6.24 (br dd, ca. 6, 13)
2.82 (br s)
2.35 (dd, 5.5, 13.5)
1.49 (dd, 12.5, 13.5)
4.51 (br dd, ca. 6, 13)
2.50 (br s)
10
11
12
13
14
14
15
73.8
51.4
74.6
48.7
73.9
51.7
70.1
52.4
146.0
128.6
38.2
139.7
27.2
149.4
125.1
38.2
149.1
126.2
36.4
5.75 (br s)
2.43 (2H, m)
2.29 (m)
5.49 (br s)
5.75 (br s)
a
b
2.59 (br s)
31.4
2.40 (br s)
2.79 (dd, 3.4, 17.4)
2.65 (br d, ca. 17)
4.16 (d, 11.0)
4.20 (d, 11.0)
1.69 (3H, d, 0.9)
1.33 (3H, s)
2.59 (br s)
2.17 (m)
2.40 (br s)
4.79 (d, 11.0)
4.83 (d, 11.0)
1.93 (3H, d, 0.9)
1.50 (3H, s)
65.5
4.92 (d, 11.0)
5.32 (d, 11.0)
1.88 (3H, d, 1.2)
1.55 (3H, s)
66.8
4.79 (d, 11.0)
5.07 (d, 11.0)
1.95 (3H, d, 0.9)
1.49 (3H, s)
65.5
63.6
16
17
18
19
20
17.3
17.2
82.6
26.1
27.2
16.6
18.4
17.4
17.1
71.6
32.2
32.7
16.8
17.5
71.6
33.0
33.2
130.8
17.7
1.46 (3H, s)
1.49 (3H, s)
2.01 (3H, s)
4.10 (2H, s)
1.44 (3H, s)
1.48 (3H, s)
1.45 (3H, s)
1.52 (3H, s)
65.2
10
20
40
50
60
70
126.3
150.8
153.6
123.3
137.0
164.9
125.8
151.0
153.4
122.9
136.9
165.2
126.5
150.9
153.6
123.3
136.9
164.9
9.18 (br s)
9.09 (br s)
9.22 (br s)
8.80 (br s)
8.66 (br d, ca. 4)
6.98 (dd, 4.9, 8.0)
7.94 (ddd, 1.8, 1.9, 8.0)
8.80 (br s)
7.40 (dd, 4.9, 8.0)
8.32 (ddd, 1.8, 1.9, 8.0)
7.38 (dd, 4.9, 8.0)
8.32 (ddd, 1.8, 1.9, 8.0)
100
200
400
500
600
700
125.8
150.9
153.7
123.6
137.3
165.2
125.5
150.8
153.9
123.6
137.2
164.5
125.7
151.0
153.8
123.5
137.3
165.2
9.27 (br s)
9.23 (br s)
9.27 (br s)
8.79 (br s)
8.79 (br d, ca. 4)
7.42 (dd, 4.9, 8.0)
8.31 (ddd, 1.8, 1.9, 8.0)
8.79 (br s)
7.42 (dd, 4.9, 8.0)
8.30 (ddd, 1.8, 1.9, 8.0)
7.42 (dd, 4.9, 8.0)
8.29 (ddd, 1.8, 1.9, 8.0)
1000
2000
3000
4000
5000
6000
7000
173.8
34.5
24.8
31.4
22.3
13.9
130.0
129.9
128.5
133.3
128.5
129.9
166.4
173.8
34.5
24.8
31.4
22.3
13.9
2.45 (2H, t, 7.0)
1.68 (2H, m)
1.30 (2H, m)
1.27 (2H, m)
0.86 (3H, t, 7.0)
8.10 (dd, 1.3, 8.5)
7.39 (dd, 7.8, 8.5)
7.59 (tt, 1.3, 7.8)
7.39 (dd, 7.8, 8.5)
8.10 (dd, 1.3, 8.5)
2.44 (2H, t, 7.0)
1.67 (2H, m)
1.30 (2H, m)
1.26 (2H, m)
0.86 (3H, t, 7.0)
The following experimental conditions were used for chro-
matography: highly porous synthetic resin, Diaion HP-20
(Mitsubishi Chemical Co.); normal-phase silica gel column
chromatography (CC), silica gel BW-200 (Fuji Silysia Chemical,
Ltd., 150–350 mesh); ODS CC, Chromatorex ODS DM1020T
(Fuji Silysia Chemical, Ltd., 100–200 mesh); normal-phase TLC,
pre-coated TLC plates with silica gel 60F254 (Merck, 0.25 mm);
reversed-phase TLC, pre-coated TLC plates with silica gel RP-18
F254S (Merck, 0.25 mm); reversed-phase HPTLC, pre-coated TLC
plates with silica gel RP-18 WF254S (Merck, 0.25 mm), detection
was achieved by spraying with 1% Ce(SO4)2–10% aqueous H2SO4,
followed by heating.
Fig. 2. Absolute stereostructure of 9. Reagents and conditions: (a) PPh3/CH2Cl2, 0 8C, 30 min, 71%; (b) 0.1% NaOMe–MeOH, r.t., 8 h, 90%.

202
T. Morikawa et al. / Phytochemistry Letters 6 (2013) 198–204
Table 4
Effects of nigellamines (1–9) on TG contents in high glucose-pretreated HepG2 cells.
TG/protein (% of control)
Conc. (
mM)
0
3
10
30
Nigellamine A1 (1)
Nigellamine A2 (2)
Nigellamine A3 (3)
Nigellamine A4 (4)
Nigellamine A5 (5)
Nigellamine B1 (6)
Nigellamine B2 (7)
Nigellamine C (8)
Nigellamine B3 (9)
Carvacrol
100.0 ꢀ 5.6
100.0 ꢀ 4.4
100.0 ꢀ 2.1
100.0 ꢀ 4.7
100.0 ꢀ 4.5
100.0 ꢀ 1.7
100.0 ꢀ 5.4
100.0 ꢀ 5.2
100.0 ꢀ 5.2
100.0 ꢀ 4.9
100.0 ꢀ 4.4
94.3 ꢀ 4.8
96.0 ꢀ 6.1
91.9 ꢀ 5.4
103.0 ꢀ 2.4
99.1 ꢀ 3.1
100.4 ꢀ 3.4
95.6 ꢀ 6.0
97.2 ꢀ 9.3
106.3 ꢀ 3.5
99.8 ꢀ 2.7
99.3 ꢀ 2.7
91.7 ꢀ 8.9
91.2 ꢀ 2.3
89.7 ꢀ 5.7**
62.1 ꢀ 4.5**
104.6 ꢀ 2.1
81.0 ꢀ 2.5**
93.1 ꢀ 1.0
95.5 ꢀ 2.1
100.9 ꢀ 4.6
95.7 ꢀ 6.9
97.2 ꢀ 4.6
90.7 ꢀ 5.6
78.5 ꢀ 8.6**
51.2 ꢀ 3.1**
96.4 ꢀ 1.3
83.8 ꢀ 1.1**
82.2 ꢀ 7.4**
75.5 ꢀ 6.1**
113.7 ꢀ 2.5
95.4 ꢀ 2.5
86.5 ꢀ 1.9**
84.6 ꢀ 3.8**
Thymoquinol
TG/protein (% of control)
Conc. (
mM)
0
0.1
1
10
Metformin
100.0 ꢀ 2.0
90.0 ꢀ 2.7*
89.3 ꢀ 3.1**
82.9 ꢀ 2.4**
Each value represents the mean ꢀ S.E.M. (n = 4).
*
Significantly different from the control, p < 0.05.
**
Significantly different from the control, p < 0.01.
3.2. Plant material
ODS CC [30 g, MeOH–H2O (50:50 ! 70:30 ! 85:15, v/v) ! MeOH]
to give seven fractions [Fr. 7-1 (112.3 mg), Fr. 7-2 (105.0 mg), Fr. 7-3
(68.2 mg), Fr. 7-4 (315.4 mg), Fr. 7-5 (87.8 mg), Fr. 7-6 (152.0 mg),
and Fr. 7-7 (120.0 mg)]. The fraction 7-4 (315.4 mg) was further
separated by HPLC [MeOH–H2O (85:15, v/v)] to give nigellamine A1
(1, 270.9 mg, 0.0096%). The fraction 7-6 (152.0 mg) was purified by
HPLC [MeOH–H2O (80:20, v/v)] to give nigellamine B1 (6, 17.5 mg,
0.00062%). The fraction 8 (9.68 g) was subjected to ODS CC [300 g,
MeOH–H2O (50:50 ! 75:25, v/v) ! MeOH] to give six fractions [Fr.
8-1 (1.466 g), Fr. 8-2 (109.2 mg), Fr. 8-3 (1.920 g), Fr. 8-4 (2.573 g),
Fr. 8-5 (748.2 mg), and Fr. 8-6 (2.804 g)]. The fraction 8-2 (109.2 mg)
was further separated by HPLC [MeOH–H2O (80:20, v/v)] to give 6
(15.4 mg, 0.00055%). The fraction 9 (6.69 g) was subjected to ODS CC
[200 g, MeOH–H2O (40:60 ! 60:40 ! 80:20, v/v) ! MeOH] to give
10 fractions [Fr. 9-1 (1.150 g), Fr. 9-2 (943.2 mg), Fr. 9-3 (627.1 mg),
Fr. 9-4 (1.312 g), Fr. 9-5 (142.8 mg), Fr. 9-6 (750.5 mg), Fr. 9-7
(150.7 mg), Fr. 9-8 (241.0 mg), Fr. 9-9 (805.0 mg), and Fr. 9-10
(428.0 mg)]. The fraction 9-6 (750.5 mg) was purified by HPLC
[MeOH–H2O (75:25, v/v)] to give nigellamines B2 (7, 101.9 mg,
0.0036%) and D (10, 2.1 mg, 0.000075%). The fraction 9-9 (805.0 mg)
was purified by HPLC [MeOH–H2O (80:20, v/v)] to give nigellamines
A2 (2, 220.7 mg, 0.0078%) and A3 (3, 13.6 mg, 0.00048%). The fraction
9-10 (428.0 mg) was purified by HPLC [MeOH–H2O (80:20, v/v)] to
give nigellamines A5 (5, 4.3 mg, 0.00015%) and B3 (9, 6.4 mg,
0.00023%). The fraction 10 (8.13 g) was subjected to ODS CC [240 g,
MeOH–H2O (40:60 ! 60:40 ! 85:15, v/v) ! MeOH] to afford eight
fractions [Fr. 10-1 (1.422 g), Fr. 10-2 (638.2 mg), Fr. 10-3 (816.4 mg),
Fr. 10-4 (1.218 g), Fr. 10-5 (491.5 mg), Fr. 10-6 (921.5 mg), Fr. 10-7
(545.7 mg), and Fr. 10-8 (2.027 g)]. The fraction 10-5 (491.5 mg) was
purified by HPLC [MeOH–H2O (80:20, v/v)] to furnish nigellamines
A4 (4, 4.2 mg, 0.00015%) and C (8, 8.7 mg, 0.00031%). An aliquot of
the MeOH-eluted fraction (20.00 g) was subjected to silica gel CC
[600 g, CHCl3–MeOH–H2O (7:3:1, lower layer ! 6:4:1) ! MeOH] to
give seven fractions [Fr. 1 (300.0 mg), Fr. 2 (1.01 g), Fr. 3 (420.0 mg),
Fr. 4 (630.0 mg), Fr. 5 (4.00 g), Fr. 6 (11.60 g), and Fr. 7 (352.6 mg)].
The fraction 6 (11.60 g) was separated by HPLC [MeOH–H2O (65:35,
The seeds of N. sativa were purchased in Cairo, Egypt and were
identified by one of the authors (M.Y.). A voucher of the plant
material is on file in our laboratory (Morikawa et al., 2004a,b).
3.3. Extraction and isolation
The seeds of N. sativa (10.0 kg) were finely powdered and
extracted three times with MeOH under reflux for 3 h. Evaporation
of the solvent under reduced pressure provided a methanol extract
(1740 g, 17.4% from the dried material). The methanol extract
(1245 g) was partitioned into mixture of EtOAc and water to give an
EtOAc-soluble fraction (723 g, 10.1%) and an aqueous fraction
(522 g). An aliquot of the aqueous fraction (200 g) was subjected to
Diaion HP-20 CC (3.0 kg, H2O ! MeOH, twice) to give H2O- (71.0 g,
2.6%) and MeOH-eluted fractions (129.0 g, 4.7%). The EtOAc-soluble
fraction (285 g) was subjected to silica gel CC [3.0 kg, n-hexane–
EtOAc
(20:1 ! 10:1 ! 5:1 ! 2:1 ! 1:2) ! CHCl3–MeOH–H2O
(20:3:1, lower layer ! 6:4:1) ! MeOH] to give 12 fractions [Fr. 1
(12.36 g), Fr. 2 (55.91 g), Fr. 3 (74.90 g), Fr. 4 (18.51 g), Fr. 5 (61.82 g),
Fr. 6 (2.55 g), Fr. 7 (0.97 g), Fr. 8 (9.68 g), Fr. 9 (6.69 g), Fr. 10 (8.13 g),
Fr. 11 (15.82 g), and Fr. 12 (16.57 g)]. The fraction 3 (21.23 g) was
subjected to ODS CC [630 g, MeOH–H2O (80:20, v/v) ! MeOH] to
give carvacrol (136.0 mg, 0.017%). The fraction 6 (2.55 g) was
subjected to ODS CC [80 g, MeOH–H2O (55:45 ! 70:30, v/
v) ! MeOH] and HPLC [MeOH–H2O (45:55, v/v)] to give thymo-
quinol (722.4 mg, 0.026%). The fraction 7 (0.97 g) was subjected to
Table 5
Effects of nigellamines (2, 3, 5–7) on binding activity for PPAR
a.
Relative binding activity (%)a
Conc. 100 mM
Nigellamine A2 (2)
Nigellamine A3 (3)
Nigellamine A5 (5)
Nigellamine B1 (6)
Nigellamine B2 (7)
31.3 ꢀ 2.5
30.7 ꢀ 1.0
21.4 ꢀ 0.9
26.4 ꢀ 1.3
17.6 ꢀ 0.8
v/v)] to give 3-O-
sylhederagenin [28-O-
anosyl-(1!6)-
-xylopyranosyl-(1!3)-
nopyranosylhederagenin [28-O-
a
-
L
-rhamnopyranosyl-(1!2)-
a- -arabinopyrano-
a-L
-rhamnopyranosyl-(1!L4)-
b-D-glucopyr-
b-D-glucopyranosyl] ester (54.8 mg, 0.15%) and 3-O-
b-D
a-L
-rhamnopyranosyl-(1!2)-
-rhamnopyranosyl-(1!4)-
b-D-glucopyranosyl] ester (420.2 mg,
a-L-arabi-
a-L
b-D-
Each value represents the mean ꢀ S.E.M. (n = 4).
a
Relative binding activity was calculated with the activity of 500 nM GW4647 as
glucopyranosyl-(1!6)-
100%.
1.15%).

T. Morikawa et al. / Phytochemistry Letters 6 (2013) 198–204
203
3.3.1. Nigellamine B3 (9)
heparin. Mice were killed by cervical dislocation, and then the liver
was removed and weighed. Plasma TG, total cholesterol, and FFA
levels were determined using commercial kits (Triglyceride E-test
Wako, Cholesterol CII-test Wako, and NEFA C-test Wako,
respectively). After removing the liver, ca. 100 mg of liver tissue
was homogenized with H2O (5 mL), and the TG concentration in
the suspension was determined using Triglyceride E-test Wako.
27
A white powder, [
a]
+33.4 (c 0.30, CHCl3); UV [MeOH, nm
D
(log
e
)]: 219 (4.31), 264 (3.88); CD [MeOH, nm, (De)]: 222 (+5.14),
242 (ꢁ1.33); IR (KBr) nmax cmꢁ1: 1717, 1647, 1636, 1593, 1541,
1509, 1420, 1284, 1115, 1024, 941, 743, 702; 1H and 13C NMR data,
see Table 3; positive-ion FABMS m/z: 677 [M+H]+; HRFABMS m/z:
677.3449 [M+H]+ (calcd for C38H49N2O9, 677.3438).
3.3.2. Nigellamine D (10)
3.7. Effects on TG content in high glucose pre-treated HepG2 cells
23
A white powder, [
(log
IR (KBr)
a
]
ꢁ20.5 (c 0.20, CHCl3); UV [MeOH, nm
D
e
)]: 220 (4.50), 264 (3.85); CD [MeOH, nm, (De)]: 261 (ꢁ1.50);
HepG2 cells (RIKEN) were maintained in Minimum Essential
Medium Eagle (MEM, Sigma–Aldrich) containing 10% fetal bovine
serum, 1% MEM non-essential amino acids (Invitrogen), penicillin
n
max cmꢁ1: 3410, 1717, 1671, 1647, 1592, 1559, 1509,
1453, 1281, 1111, 1026, 949, 743, 704; 1H and 13C NMR data, see
Table 3; EIMS m/z: 666 ([M+], 2), 648 ([M+–H2O], 2), 542 (23), 403
(5), 280 (19), 151 (18), 124 (100), 106 (25), 105 (47); HREIMS m/z:
666.2944 [M+] (calcd for C39H42N2O8: 666.2941).
G (100 units/mL), and streptomycin (100
CO2 atmosphere. The cells were inoculated in 48-well tissue
culture plate [105 cells/well in 200
L/well in MEM]. After 20 h, the
medium was replaced with 200 L/well of Dulbecco’s Modified
mg/mL) at 37 8C under 5%
m
m
3.4. Triphenylphosphine reduction of 9
Eagle’s Medium (DMEM) containing high glucose (4500 mg/L) and
cultured for 6 days with replacing the medium by a fresh one every
2 days. After accumulation of the lipid, the medium was exchanged
To a solution of 9 (3.7 mg) in CH2Cl2 (1.0 mL) was added
triphenylphosphine (PPh3, 8.0 mg) and the mixture was stirred at
0 8C for 30 min. Removal of the solvent under reduced pressure
furnished a residue, which was purified by HPLC [MeOH–H2O
(75:25, v/v)] to give 9a (2.6 mg, 71%).
to 200
mL/well of DMEM containing low glucose (1000 mg/L) and a
test sample, and the cells were cultured. After 20 h, the TG and
protein contents in the cells were determined by the same manner
as described above. An antidiabetic agent, metformin was used as a
reference compound (Muraoka et al., 2009).
3.4.1. Compound 9a
25
D
A white powder, [
a]
+35.8 (c 0.20, CHCl3); UV [MeOH, nm
3.8. Agonistic activity for PPARa
(log
e
)]: 218 (4.38), 264 (3.87); CD [MeOH, nm, (De)]: 223 (+5.83),
249 (ꢁ0.92); IR (KBr) nmax cmꢁ1: 3410, 1725, 1651, 1636, 1592,
1507, 1455, 1285, 1117, 1024, 941, 743, 700; 1H and 13C NMR data,
see Table 3; EIMS m/z: 666 ([M+], 2), 554 (16), 537 (8), 421 (14), 280
(26), 124 (100), 119 (25), 106 (35); HREIMS m/z: 660.3413 [M+]
(calcd for C38H48N2O8: 660.3410).
Agonistic activity for PPAR
receptor cofactor assay system (EnBio RCAS for PPAR
a
was examined using a nuclear
, EnBioTec
a
Laboratories) according to the manufacturer’s instructions. This
system is a cell-free assay system using nuclear receptors and
cofactors to screen chemicals. The change in absorbance (450 nm)
caused by GW4647, a selective PPARa agonist, at 500 nM was
3.5. Deacylation of 9a with 0.1% sodium methoxide (NaOMe)–MeOH
calculated as 100%.
A solution of 9a (2.5 mg) in 0.1% sodium methoxide (NaOMe)–
MeOH (1.0 mL) was stirred at room temperature for 8 h. By HPLC
analysis of the reaction mixture, methyl nicotinate (i, tR 5.58 min)
and methyl hexanoate (ii, tR 27.19 min) were identified [detection:
RI, mobile phase: MeOH–H2O (60:40, v/v), flow rate 0.7 mL/min].
The standard samples of methyl nicotinate (i) and methyl
hexanoate (ii) were obtained by diazomethane methylation of
commercially available nicotinic acid and n-hexanonic acid. The
rest of reaction mixture was neutralized over Dowex HCR W2 resin
(H+ form), which was then removed by filtration. The filtrate
was concentrated under reduced pressure, and the resulting
product was purified by HPLC [MeOH–H2O (80:20, v/v)] to give 9b
(1.2 mg, 90%).
3.9. Statistics
Values were expressed as means ꢀ S.E.M. One-way analysis of
variance (ANOVA) followed by Dunnett’s test was used for statistical
analysis. Probability (p) values less than 0.05 were considered
significant.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Numbers
24590037, 24590153, 22510240, and 21590031.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.phytol.2013.01.004.
3.5.1. Compound 9b
24
D
A white powder, [
a]
+35.6 (c 0.10, MeOH); IR (KBr) nmax
cmꢁ1: 3346, 1653, 1559, 1507, 1387, 1262, 1120, 1044, 938, 756,
718; 1H and 13C NMR data, see Table 3; EIMS m/z: 352 ([M+], 1), 334
([M+–H2O], 1), 286 (6), 138 (17), 120 (100), 95 (35); HREIMS m/z:
352.2254 [M+] (calcd for C20H32O5: 352.2250).
References
Ansari, A.A., Hassan, S., Kenne, L., Atta-ur-Rahman, Wehler, T., 1988. Structural
studies on a saponin isolated Nigella sativa. Phytochemistry 27, 3977–3979.
Bellentani, S., Tiribelli, C., Saccoccio, G., Sodde, M., Fratti, N., De Martin, C., Cristia-
nini, G., 1994. Prevalence of chronic liver disease in the general population of
northern Italy: the dionysos study. Hepatrogy 20, 1442–1449.
3.6. Effect on food intake, visceral fat, liver weight, liver TG content,
and plasma biochemicals in mice
El-Hassan, A.Y., Ibrahim, E.M., Al-Mulhim, F.A., Nabhan, A.A., Chammas, M.Y., 1992.
Fatty infiltration of the liver: analysis of prevalence, radiological and clinical
features and influence on patient management. Br. J. Radiol. 65, 774–778.
El Sayed, K.A., Ross, S.A., El Sohly, M.A., Khalafalla, M.M., Abdel-Halim, O.B., Ikegami,
F., 2000. Effect of different fertilizers on the amino acid, fatty acid, and essential
oil composition of Nigella sativa seeds. Saudi Pharm. J. 8, 175–181.
Enomoto, S., Asano, R., Iwahori, Y., Narui, T., Okada, Y., Singab, A.N.B., Okuyama, T.,
2001. Hemotological studies on black cumin oil from the seeds of Nigella sativa
L. Biol. Pharm. Bull. 24, 307–310.
A test sample was administered to ddY male mice (6 week old)
once a day (10:00–12:00) for 3 days fed a standard laboratory
chow (MF, Oriental Yeast Co., Ltd., Japan). Body weight was
measured every day before administration of the test sample. After
fasting for 20 h, blood samples (ca. 0.2 mL) were collected from an
infraorbital venous plexus and put into tubes containing 5 units

204
T. Morikawa et al. / Phytochemistry Letters 6 (2013) 198–204
Marceau, P., Biron, S., Hould, F.S., Marceau, S., Simard, S., Thung, S.N., Kral, J.G., 1999.
Liver pathology and the metabolic syndrome X in severe obesity. J. Clin.
Endocrinol. Metab. 84, 1513–1517.
Marchesini, G., Brizi, M., Morselli-Labate, A.M., Bianchi, G., Bugianesi, E., McCul-
lough, A.J., Forlani, G., Melchionda, N., 1999. Association of nonalcoholic fatty
liver disease with insulin resistance. Am. J. Med. 107, 450–455.
Morikawa, T., Xu, F., Kashima, Y., Matsuda, H., Ninomiya, K., Yoshikawa,
M., 2004a. Novel dolabellane-type diterpene alkaloids with lipid metabolism
promoting activities from the seeds of Nigella sativa. Org. Lett. 6,
869–872.
Morikawa, T., Xu, F., Ninomiya, K., Matsuda, H., Yoshikawa, M., 2004b. Nigellamines
A3, A4, A5, and C, new dolabellane-type diterpene alkaloids, with lipid metabo-
lism-promoting activities from the Egyptian medicinal food black cumin. Chem.
Pharm. Bull. 52, 494–497.
Muraoka, O., Morikawa, T., Zhang, Y., Ninomiya, K., Nakamura, S., Matsuda, H.,
Yoshikawa, M., 2009. Novel megastigmanes with lipid accumulation inhibitory
and lipid metabolism-promoting activities in HepG2 cell from Sedum sarmen-
tosum. Tetrahedron 65, 4142–4148.
Shimizu, M., Shingyouchi, K., Morita, N., Kizu, H., Tomimori, T., 1978. Triterpenoid
saponins from Pulsatilla cernua Spreng. I. Chem. Pharm. Bull. 26, 1666–1671.