Anti-allergic principles from Thai zedoary: Structural requirements of curcuminoids for inhibition of degranulation and effect on the release of TNF-α and IL-4 in RBL-2H3 cells

Article abstract of DOI:10.1016/j.bmc.2004.08.027

The 80% aqueous acetone extract of the rhizomes of Curcuma zedoaria cultivated in Thailand (Thai zedoary) was found to inhibit release of β-hexosaminidase, as a marker of antigen-IgE-mediated degranulation, in RBL-2H3 cells and passive cutaneous anaphylaxis reaction in mice. Effects of four curcuminoids from Thai zedoary and several related compounds on the degranulation were examined. Among them, curcumin showed the highest activity against β-hexosaminidase release with IC₅₀ of 5.3 μM, followed by bisdemethoxycurcumin (IC₅₀ = 11 μM). With regard to the structural requirements of curcuminoids for the activity, the conjugated olefins at the 1-7 positions and the 4′- or 4″-hydroxyl groups of curcuminoids were suggested to be essential for the strong activity, whereas the 3′- or 3″-methoxyl group only enhanced the activity. Furthermore, effects of curcumin and bisdemethoxycurcumin on calcium ionophores (A23187 and ionomycin)-induced degranulation and antigen-induced release of TNF-α and IL-4 were examined. The 80% aqueous acetone extract of the rhizomes of Curcuma zedoaria cultivated in Thailand (Thai zedoary) was found to inhibit release of β-hexosaminidase, as a marker of antigen-IgE-mediated degranulation, in RBL-2H3 cells and passive cutaneous anaphylaxis reaction in mice. From the active fraction, four curcuminoids (curcumin, dihydrocurcumin, tetrahydrodemethoxycurcumin, and tetrahydrobisdemethoxycurcumin) were isolated together with two bisabolane-type sesquiterpenes, and the effects of four curcuminoids from Thai zedoary and several related compounds on the degranulation were examined. Among them, curcumin showed the highest activity against β-hexosaminidase release with IC₅₀ of 5.3 μM, followed by bisdemethoxycurcumin (IC₅₀ = 11 μM). With regard to the structural requirements of curcuminoids for the activity, the conjugated olefins at the 1-7 positions and the 4′- or 4″-hydroxyl groups of curcuminoids were suggested to be essential for the strong activity, whereas the 3′- or 3″-methoxyl group only enhanced the activity. Furthermore, effects of curcumin and bisdemethoxycurcumin on calcium ionophores (A23187 and ionomycin)-induced degranulation and antigen-induced release of TNF-α and IL-4 were examined.

Full text of DOI:10.1016/j.bmc.2004.08.027

Bioorganic & Medicinal Chemistry 12 (2004) 5891–5898
Anti-allergic principles from Thai zedoary: structural
requirements of curcuminoids for inhibition of degranulation
and effect on the release of TNF-a and IL-4 in RBL-2H3 cells
Hisashi Matsuda,a Supinya Tewtrakul,a,b Toshio Morikawa,a Akihiko Nakamuraa
and Masayuki Yoshikawaa,*
aKyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto 607-8412, Japan
bFaculty of Pharmaceutical Sciences, Prince of Songkla University, Hat-Yai, Songkhla 90112, Thailand
Received 20 July 2004; revised 20 August 2004; accepted 20 August 2004
Available online 15 September 2004
Abstract—The 80% aqueous acetone extract of the rhizomes of Curcuma zedoaria cultivated in Thailand (Thai zedoary) was found
to inhibit release of b-hexosaminidase, as a marker of antigen-IgE-mediated degranulation, in RBL-2H3 cells and passive cutaneous
anaphylaxis reaction in mice. From the active fraction, four curcuminoids (curcumin, dihydrocurcumin, tetra-
hydrodemethoxycurcumin, and tetrahydrobisdemethoxycurcumin) were isolated together with two bisabolane-type sesquiterpenes,
and the effects of four curcuminoids from Thai zedoary and several related compounds on the degranulation were examined. Among
them, curcumin showed the highest activity against b-hexosaminidase release with IC50 of 5.3lM, followed by bisdemethoxycurcu-
min (IC50 = 11lM). With regard to the structural requirements of curcuminoids for the activity, the conjugated olefins at the 1–7
positions and the 40- or 400-hydroxyl groups of curcuminoids were suggested to be essential for the strong activity, whereas the 30- or
300-methoxyl group only enhanced the activity. Furthermore, effects of curcumin and bisdemethoxycurcumin on calcium ionophores
(A23187 and ionomycin)-induced degranulation and antigen-induced release of TNF-a and IL-4 were examined.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
pounds as alternative for passive cutaneous anaphylaxis
(PCA) reactions in laboratory animals.2 The late phase
reaction occurs within 4–6h after the early phase reac-
tion in type I allergy. The mediators such as cytokines
(TNF-a, IL-4, etc.) from the cells are involved in the late
phase. These mediators increase endothelial cell adhe-
sion and recruitment of inflammatory cells to the af-
fected site.3
Type I allergy is induced by certain types of antigens
such as foods, dust mites, medicines, cosmetics, mold
spores, and pollen. This class of antigens induces the
production of antigen-specific IgE antibodies that bind
to receptors on mast cells or basophils. Recently, the
biphasic reactions, the early phase and the late phase
reactions, have been reported in type I allergy. The early
phase reaction in type I allergy occurs within minutes
and then the mediators such as histamine and serotonin
are released from the cell. These mediators induce vaso-
dilation, mucous secretion, and bronchoconstriction. b-
Hexosaminidase is also stored in the secretory granules
of mast cells and basophils, and released along with hist-
amine when mast cells and basophils are activated.
Therefore, this enzyme activity is used as a marker of
mast cell or basophil degranulation,1 and this assay
has been used for evaluation of the anti-allergic com-
The rhizomes of Curcuma zedoaria ROSCOE (so called
ÔzedoaryÕ) have been used as anti-flatulence and anti-
inflammatory in Thai traditional medicine. Moreover,
the rhizomes of this plant also have been used as fragrance
and spice in Asian countries. Regarding the pharmaco-
logical studies of this plant, anti-inflammatory effects of
the extract and sesquiterpenes were reported.4 In our pre-
vious studies, we reported the hepatoprotective, vasodila-
tive, and nitric oxide production inhibitory activities of
various sesquiterpenes and curcuminoids isolated from
the rhizomes of C. zedoaria cultivated in China.5
*
As a part of our studies on bioactive constituents
from natural medicines, we previously reported various
Corresponding author. Tel.: +81 75 595 4633; fax: +81 75 595
4768; e-mail: shoyaku@mb.kyoto-phu.ac.jp
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.08.027

5892
H. Matsuda et al. / Bioorg. Med. Chem. 12 (2004) 5891–5898
inhibitors against degranulation induced by antigen in
RBL-2H3 cells.6 In the present study, the 80% aqueous
acetone extract of the dried rhizomes of C. zedoaria culti-
vated in Thailand (Thai zedoary)7 was found to inhibit
release of b-hexosaminidase, as a marker of antigen-
IgE-mediated degranulation, in RBL-2H3 cells and ear
PCA reaction in mice. In the present study, four cur-
cuminoids and two bisabolane-type sesquiterpenes were
isolated from the active fraction, and we examined the
effects of curcuminoids and their derivatives on the anti-
gen-induced degranulations in RBL-2H3 cells to discuss
structural requirements of curcuminoids for the activity.
In addition, we describe effects of two active curcumi-
noids [curcumin (1) and bisdemethoxycurcumin (6)] on
calcium ionophores-induced degranulation and on anti-
gen-induced release of TNF-a and IL-4 in RBL-2H3
cells.
for the following bioassay-guided separation of active
compounds.
The extract was partitioned into an ethyl acetate
(EtOAc)–water mixture to furnish the EtOAc-soluble
fraction (9.8%) and water-soluble fraction (2.9%). The
EtOAc-soluble fraction, which showed potent activity
(IC50 = 45lg/mL), was subjected to normal-phase silica
gel
(SiO2)
[n-hexane–AcOEt
(10:1 ! 5:1 ! 2:1
! 1:1) ! AcOEt ! MeOH] and reversed-phase silica
gel (ODS) column chromatographies (MeOH–H2O)
and finally HPLC (YMC-Pack ODS-A, 250 · 20mm
i.d., CH3CN–H2O or MeOH–H2O) to give four
curcuminoids [curcumin (1,8 1.40% from the dried
rhizomes), dihydrocurcumin (2,8 0.20%), tetra-
hydrodemethoxycurcumin
(3,9
0.043%),
tetra-
hydrobisdemethoxycurcumin (4,9 0.055%)], and two
bisabolene-type sesquiterpenes [4,5-dihydrobisabola-
2,10-diene (5,10 0.021%) and bisacurone10 (0.046%)].
2. Results and discussion
Related curcuminoids, bisdemethoxycurcumin (6) iso-
lated from Chinese zedoary5 and tetrahydrocurcumin
(7),11 hexahydrocurcumin (8),12 monomethylcurcumin
(9),13 and dimethylcurcumin (10)13 derived from 1 in
the usual way were examined to clarify the structure–
activity relationships for curcuminoids (Fig. 1).
2.1. Isolation of active constituents from Thai zedoary
The dried rhizomes of C. zedoaria (1.45kg) were ex-
tracted with 80% aqueous acetone three times under
room temperature. The 80% aqueous acetone extract
(12.7% from this natural medicine) significantly inhib-
ited ear PCA reaction in mice (Table 1). The extract also
inhibited the release of b-hexosaminidase in RBL-2H3
cells (IC50 = 48lg/mL, Table 2) and this activity was
stronger than those of anti-allergic compounds trani-
last [282lM (=93lg/mL)] and ketotifen fumarate
[158lM (=67lg/mL)] (Table 2). Therefore, the inhibi-
tion against the release of b-hexosaminidase was used
2.2. Inhibitory effects of compounds 1–5 isolated from
Thai zedoary and related curcuminoids (6–10) on the
release of b-hexosaminidase from RBL-2H3 cells
As shown in Table 3, curcumin (1), bisdemethoxycurcu-
min (6), and monomethylcurcumin (9) exhibited potent
inhibitory activities with IC50 of 5.3, 11, and 15lM,
Table 1. Inhibitory effects of the 80% aq acetone extract from Thai zedoary and its fractions on ear PCA reactions in mice
Dose (mg/kg, p.o.)
n
Leakage of dye (O.D. at 620nm)
Inhibition (%)
Control (PBS)
Control (anti-DNP-IgE)
80% Aq acetone ext.
—
—
7
0.039 0.003**
0.398 0.055
0.299 0.033
0.285 0.032
0.276 0.034*
0.304 0.066
0.226 0.021**
0.365 0.053
—
—
12
7
100
200
400
100
200
400
27.6
31.5
34.0
26.2
47.9
9.2
7
7
EtOAc-soluble fraction
H2O-soluble fraction
5
8
6
Control (PBS)
Control (anti-DNP-IgE)
Tranilast
—
—
5
8
7
7
0.050 0.009**
0.310 0.054
0.177 0.019*
0.126 0.017**
—
—
100
200
51.2
70.8
Each value represents the mean SEM. Significantly different from the control, *p < 0.05, **p < 0.01.
Table 2. Inhibitory effects of the 80% aq acetone extract from Thai zedoary and its fractions on antigen-induced release of b-hexosaminidase in RBL-
2H3 cells
Inhibition (%)
IC50 (lg/mL)
Enzyme inhibition
at 100lg/mL
Concn:
0lg/mL
10lg/mL
30lg/mL
100lg/mL
80% Aq acetone extract
EtOAc-soluble fraction
H2O-soluble fraction
0.0 4.8
0.0 0.5
0.0 3.1
5.1 3.3
4.3 2.0
7.8 2.9
26.8 2.8**
35.4 3.2**
0.0 5.2
87.9 0.8**
97.6 1.9**
4.8 3.1
48
35
—
18%
27%
—
Each value represents the mean SEM. (n = 4). Significantly different from control, **p < 0.01.

H. Matsuda et al. / Bioorg. Med. Chem. 12 (2004) 5891–5898
5893
4'
4''
OH
OH
HO
HO
6
2
4
5
3
3''
H3CO
1''
1'
OCH3
3'
1
2''
7
2'
O
OH
O
OH
curcumin (1)
bisdemethoxycurcumin (6)
OH
OH
HO
HO
H3CO
H3CO
OCH3
OCH3
O
OH
O
OH
dihydrocurcumin (2)
tetrahydrocurcumin (7)
OH
OH
HO
HO
H3CO
OCH3
OCH3
O
OH
O
OH
tetrahydrodemethoxycurcumin (3)
hexahydrocurcumin (8)
OH
OH
HO
H3CO
H3CO
OCH3
O
OH
O
OH
tetrahydrobisdemethoxycurcumin (4)
monomethylcurcumin (9)
OCH3
OCH3
H3CO
H3CO
OH
OH
O
OH
4,5-dihydrobisabola-2,10-diene (5)
dimethylcurcumin (10)
Figure 1. Chemical structures of compounds 1–5 isolated from the rhizomes of C. zedoaria and related compounds 6–10.
Table 3. Inhibitory effects of compounds 1–5 isolated from Thai zedoary and related compounds 6–10 on antigen-induced release of b-
hexosaminidase in RBL-2H3 cells
Compounds
Inhibition (%)
IC50 (lM)
Enzyme inhibition
at 100lM
Concn:
0lM
3lM
10lM
30lM
10 lM
0
1
2
3
4
5
6
7
8
9
10
0.0 0.5
0.0 8.6
0.0 1.8
0.0 0.5
0.0 5.6
0.0 7.5
0.0 2.3
28.5 5.9**
—
—
80.4 4.7**
9.1 1.1
100.1 1.1**
39.0 6.6**
27.1 2.6*
22.2 5.4*
25.1 3.2**
91.2 4.6**
102.3 0.8**
81.5 6.0**
78.0 9.7 **
41.6 7.1**
22.4 3.9**
92.5 0.8**
92.7 3.4**
30.5 3.3**
88.6 4.9**
8.5 0.6
5.3
39
27%
16%
2%
—
13.5 4.7
9.1 4.5
19.6 3.4**
41.0 9.3*
3.0 3.0
62
—
—
>100
>100
11
—
7.3 7.2
—
—
16%
2%
—
**
21.0 6.6
58
0.0 2.6
0.0 4.1
0.0 3.1
7.8 2.6
47.4 4.4**
3.1 2.5
11.4 6.1
67.5 3.6**
1.0 3.4
>100
15
>100
10.6 1.6
—
30%
—
0lM
30lM
10lM
30 lM
0
1000lM
**
75.0 0.6
100.7 1.1**
Tranilast
Ketotifen fumarate
0.0 1.7
0.0 1.8
8.2 1.8
7.4 1.5
22.4 2.5*
27.6 2.2*
56.9 3.4**
80.7 1.8**
282
158
—
—
Each value represents the mean SEM. (n = 4). Significantly different from control, *p < 0.05, **p < 0.01.
respectively. Whereas, compounds 2–5, 7, 8, and 10
showed moderate or weak activity. Effects of the active
compounds 1–3, 6, 7, and 9 were also examined on the
enzyme activity of b-hexosaminidase. As a result, these
active compounds showed weak or less inhibition against
b-hexosaminidase at 100lM. These results indicated that
these compounds inhibited the antigen-induced degranu-
lation but not enzyme activity of b-hexosaminidase.
Yano and co-workers reported that the extract of the
rhizomes of Curcuma longa and several curcuminoids
isolated from the active fraction inhibited compound
48/80, concanavalin A, and/or calcium ionophore
A23187-induced histamine release in rat peritoneal mast
cells.14 However, structure–activity relationships of cur-
cuminoids for anti-allergic activity have not been suffi-
ciently studied.

5894
H. Matsuda et al. / Bioorg. Med. Chem. 12 (2004) 5891–5898
With regard to the structure–activity relationships of
curcuminoids on anti-allergic activity, when the conju-
gated olefins at the 1–7 positions and the 30 and 300-meth-
oxyl groups decreased in number, the activities were
decreased. This indicated the importance of the conju-
gated olefins and methoxyl groups of curcuminoids.
However, the conjugated olefins seemed to affect the
activity more than the methoxyl groups, which can be
observed from the IC50 of compound 1, which was only
two times higher than that of compound 6 (lacks of two
methoxyl groups) [1 (IC50 = 5.3lM) > 6 (11lM)], but
eight times higher than that of compound 2 (lacks of
one olefinic group at the 1–2 positions), and 11 times
higher than that of compound 7 (lack of two olefinic
groups at the 1–2 and 6–7 positions) [1 (5.3lM) > 2
(39lM) > 7 (58lM) > 8 (100lM)]. Yano and co-work-
ers reported that 6 showed stronger activity than 1
against compound 48/80-induced histamine release in
rat peritoneal mast cells.14 However, in the present
study, the activity of 1 was stronger than that of 6
against antigen-IgE-mediated degranulation in RBL-
2H3 cells. The methylation of the 40-hydroxyl group of
1 reduced the activity ca. three times less than that of
1. Moreover, the methylation of both the 40- and 400-hyd-
roxyl groups of 1 dramatically decreased the activity
[1 (5.3lM) > 9 (15lM) > 10 (>100lM)]. These results
indicated that at least one p-hydroxyl group of aromatic
rings of 1 was essential to exhibit the activity.
48/80, and A23187-induced histamine release in rat peri-
toneal mast cells and concluded that curcumin (1) potently
suppressed the histamine release probably through the
inhibition of the degranulation process following a rise
in intracellular Ca2+ levels.14
In the present study, effects of curcumin (1) and bis-
demethoxycurcumin (6) on A23187 and ionomycin-
induced degranulations of RBL-2H3 cells were
examined. In agreement with the previous report,14
compounds 1 and 6 inhibited A23187 and ionomycin-in-
duced degranulation of RBL-2H3 cells, and their IC50
values (7.2 and 13lM for A23187, 6.0and 15 lM for
ionomycin) were similar to those by antigen (5.3 and
11lM) (Table 4).
Recently, ionomycin was reported to cause the activa-
tion of calcium release-activated calcium channels
(CRAC) by depleting intracellular calcium stores rather
than acting as an ionophore and may cause calcium in-
flux in a similar way, as does antigen.15 In the present
study, compounds 1 and 6 inhibited the degranulation
of RBL-2H3 cells by both the antigen and ionomycin
at similar IC50 values. These results suggested that both
the compounds might inhibit Ca2+ influx via CRAC
channels or mechanisms after Ca2+ influx.
2.4. Inhibitory effects of curcumin (1) and bisdemeth-
oxycurcumin (6) on antigen-induced TNF-a and IL-4
production in RBL-2H3 cells
The structure–activity relationships of curcuminoids on
anti-allergic activity can be concluded that the conju-
gated olefins at the 1–7 positions and the 40- and 400-hydr-
oxyl groups of curcuminoids are essential for the
activity, whereas 30- and 300-methoxyl groups only en-
hanced the activity.
Curcumin (1) was previously reported to inhibit the pro-
duction of TNF-a and IL-1 in a human macrophage cell
line by LPS and inhibit IL-12 production in mouse
spleen macrophages induced by LPS,16 but the effects
of 1 on the antigen-induced production of cytokines in
basophils was not investigated so far. To investigate
whether curcumin (1) and bisdemethoxycurcumin (6)
can inhibit the late phase reaction or not, inhibitory ef-
fects of 1 and 6 on the release of TNF-a and IL-4 were
examined. As a result, 1 inhibited the release of TNF-a
and IL-4 with IC50 of 20and 18 lM, respectively.
Whereas those of 6 were 38 and 34lM (Table 5). IC50
values of both compounds 1 and 6 for release of TNF-
a and IL-4 were ca. four times lower than those for re-
lease of b-hexosaminidase. These findings suggested that
1 and 6 might exhibit weaker effects on the late phase
reactions than the early phase reactions.
2.3. Inhibitory effects of curcumin (1) and bisdemeth-
oxycurcumin (6) on the release of b-hexosaminidase
induced by calcium ionophores
Calcium ionophores are compounds that enhance Ca2+
influx into the cell by increasing membrane permeability
of Ca2+. The increase in intracellular Ca2+ induces the
movement of granules to plasma membrane followed
by the degranulation of mast cells or basophils and acti-
vates the formation of inflammatory mediators such as
prostaglandins and leukotrienes.3 Previously, curcumin
(1) was reported to inhibit concanavalin A, compound
Table 4. Inhibitory effects of curcumin (1) and bisdemethoxycurcumin (6) on the release of b-hexosaminidase induced by calcium ionophores,
A23187 and ionomycin
Inhibition (%)
IC50
Concn:
0lM
3lM
10lM
20lM
30lM
(lM)
A23187 (1lM)
Curcumin (1)
6
0.0 2.4
0.0 2.3
12.6 0.4**
—
58.1 6.7
33.2 4.1**
98.1 0.9**
77.1 0.8**
110.0 1.6**
95.4 0.1**
7.2
13
Ionomycin (1lM)
Curcumin (1)
0.0 2.3
0.0 2.3
16.8 2.8**
—
79.5 1.6**
24.7 2.0**
100.9 0.5**
72.3 1.3**
101.0 0.3**
90.5 0.4**
6.0
15
6
Each value represents the mean SEM. (n = 4). Significantly different from control, **p < 0.01.

H. Matsuda et al. / Bioorg. Med. Chem. 12 (2004) 5891–5898
5895
Table 5. Inhibitory effects of curcumin (1) and bisdemethoxycurcumin (6) on antigen-induced TNF-a and IL-4 production from RBL-2H3
Inhibition (%)
10lM
IC50
(lM)
Concn:
0lM
3lM
20lM
30lM
10 lM
0
TNF-a Release
Curcumin (1)
6
0.0 4.4
0.0 8.8
0.0 3.1
—
—
12.7 3.8
7.0 7.0
89.0 1.3 **
37.8 2.6**
5.2 4.6
—
98.8 1.1**
47.0 5.3**
101.3 1.2**
110.0 1.6**
110.6 1.4**
—
20
38
5.86c
Luteolin
25.0 3.0**
IL-4 Release
Curcumin (1)
6
0.0 1.1
0.0 1.4
0.0 2.1
—
—
5.0 2.1
3.7 1.7
89.4 1.0**
56.0 1.9**
14.7 0.8**
—
97.6 1.5**
55.6 1.6**
99.2 0.2**
—
101.8 0.4**
—
18
34
3.76c
Luteolin
41.6 1.3**
Each value represents the mean SEM. (n = 4). Significantly different from the control, **p < 0.01.
In conclusion, the 80% aqueous acetone extract of Thai
zedoary was found to inhibit antigen-IgE-mediated
degranulation in RBL-2H3 cells and ear PCA reaction
in mice. By bioassay-guided separation, three curcumi-
noids (1–3) were isolated as active compounds. Regard-
ing to the structural requirements of curcuminoids for
antigen-induced degranulation, the conjugated olefins
at the 1–7 positions and the 40- or 400-hydroxyl groups
of curcuminoids were suggested to be essential for the
strong activity, whereas the 30- or 300-methoxyl group
only enhanced the activity. In addition, curcumin (1)
and bisdemethoxycurcumin (6) were suggested to inhibit
Ca2+ influx via CRAC channels or the mechanisms after
Ca2+ influx. Furthermore, curcumin (1) inhibited anti-
gen-induced release of TNF-a and IL-4, both of which
participate in the late phase of type I allergic reaction,
in RBL-2H3. These results can support that the rhi-
zomes of C. zedoaria have been used as anti-inflamma-
tory agents in Thai traditional medicine, and the detail
mechanisms of action of curcuminoids including experi-
mental animal models need to be studied further.
0.25mm) (reversed phase); reversed-phase HPTLC,
pre-coated TLC plates with Silica gel RP-18 60WF254S
(Merck, 0.25mm); detection was achieved by
spraying with 1% Ce(SO4)2–10% aqueous H2SO4 and
heating.
3.2. Plant material
The rhizomes of C. zedoaria were purchased in Thailand
in July 2000, and were identified by Dr. Yutana Pong-
piriyadacha (Faculty of Nakhon si thammarat, Raja-
mangala Institute of Technology, Thailand) and one
of the authors (Dr. Supinya Tewtrakul). A voucher
specimen (No. T-11) of this herbal medicine is on file
in our laboratory.
3.3. Extraction and isolation
The dried rhizomes of C. zedoaria (1.45kg) were ex-
tracted with 80% aqueous acetone three times under
room temperature to give the 80% aqueous acetone ex-
tract (184.4g, 12.7% from this natural medicine). The
aqueous acetone extract (129.9g) was partitioned into
an ethyl acetate (EtOAc)–water mixture to furnish the
EtOAc-soluble fraction (100.5g, 9.8%) and water-solu-
ble fraction (29.4g, 2.9%). The EtOAc-soluble fraction
(76g) was subjected to ordinary-phase silica gel
[2.3kg, n-hexane–AcOEt (10:1 ! 5:1 ! 2:1 ! 1:1) !
AcOEt ! MeOH] to give 13 fractions [Fr. 1 (0.066g),
Fr. 2 (6.419g), Fr. 3 (9.149g), Fr. 4 (0.746g), Fr. 5
(3.904g), Fr. 6 (0.956g), Fr. 7 (0.892g), Fr. 8 (2.093g),
Fr. 9 (3.411g), Fr. 10 (10.099g), Fr. 11 (19.188g), Fr.
12 (12.483g), and Fr. 13 (6.594g)]. Fr. 10 (10.099g)
was separated by reversed-phase silica gel (ODS) col-
umn chromatography [300g, MeOH–H2O (50:50 !
60:40 ! 85:15) ! MeOH] to furnish seven fractions
[Fr.10–1 (0.429g), Fr. 10–2 (4.555g), Fr. 10–3
(0.414g), Fr. 10–4 (2.345g), Fr. 10–5 (1.501g), Fr. 10–
6 (0.425g), and Fr. 10–7 (0.430g)]. Fr. 10–2 (1.00g)
and Fr. 10–5 (1.00g) were further separated by HPLC
[YMC-Pack ODS-A, 250 · 20mm i.d., CH3CN–H2O
(45:55 or 50:50)] to give tetrahydrodemethoxycurcumin
(3, 72.4mg, 0.043%) and tetrahydrobisdemethoxycurcu-
min (4, 93.2mg, 0.055%) from Fr. 10–2 and 4,5-dihyd-
robisabola-2,10-diene (5, 73.3mg, 0.021%). Fr. 11
(19.188g) was separated by ODS column chromatogra-
phy [510g, MeOH–H2O (70:30)] and then purified
by HPLC [MeOH–H2O (70:30)] to give curcumin
3. Experimental
3.1. Instruments
The following instruments were used to obtain physical
data: specific rotations, Horiba SEPA-300 digital pola-
rimeter (l = 5cm); UV spectra, Shimadzu UV-1600 spec-
trometer;
IR
spectra,
Shimadzu
FTIR-8100
spectrometer; EI-MS and high-resolution MS, JEOL
JMS-GCMATE mass spectrometer; FAB-MS and
high-resolution MS, JEOL JMS-SX 102A mass spec-
trometer; 1H NMR spectra, JNM-LA500 (500MHz)
spectrometer and JEOL EX-270(270MHz); 13C NMR
spectra, JNM-LA500 (125MHz) and JEOL EX-270
(68MHz) spectrometer with tetramethylsilane as an
internal standard.
The following experimental conditions were used for
chromatography: normal-phase silica gel column chro-
matography, Silica gel BW-200 (Fuji Silysia Chemical,
Ltd, 150–350 mesh); reversed-phase silica gel column
chromatography, Chromatorex ODS DM1020T (Fuji
Silysia Chemical, Ltd, 100–200 mesh); TLC, pre-coated
TLC plates with Silica gel 60F254 (Merck, 0.25mm)
(ordinary phase) and Silica gel RP-18 60F254 (Merck,

5896
H. Matsuda et al. / Bioorg. Med. Chem. 12 (2004) 5891–5898
(1, 9.567g, 1.40%) and dihydrocurcumin (2, 1.380g,
0.20%). Fr. 12 (12.30g) was separated by ODS column
chromatography [369g (40:60 ! 60:40 ! 80:20) !
MeOH ! acetone] and then purified by HPLC
[MeOH–H2O (50:50)] to give bisacurone (350mg,
0.046%). The known compounds were identified by
the solution was administered orally at 10mL/kg in each
experiment, while vehicle was given orally at 10mg/kg in
the corresponding control group.
3.6.3. Effects on ear PCA reaction in mice. The ear PCA
reaction was performed according to the method re-
ported previously6f,18 with slight modification. Briefly,
10lL of anti-DNP IgE diluted in PBS (20lg/mL), or
PBS alone (normal group) was injected intradermally
into both ears of male ddY mice. Forty-seven hours
later, test compounds suspended in 5% acacia solution
was administered orally. After 1h, 0.25mL of PBS,
which contains 2.0% Evans blue and 0.25mg of DNP-
BSA, was injected into the vein. Thirty minutes later,
mice were killed by cervical dislocation and both ears
were removed and incubated with 1M KOH solution
overnight at 37ꢁC to dissolve them. The solution was
then mixed with 4.5mL of a mixture of acetone–0.2M
H3PO4 (13:5). After centrifugation at 4000rpm for
10min, absorbance was measured at 620nm using a
spectrophotometer (Beckmann DU 530). Tranilast was
used as a reference compound.
1
comparison of their physical data ([a]D, IR, H NMR,
13C NMR, MS) with reported values,8–10 or authentic
samples.5a,b,f
3.4. Hydrogenation of 1
A solution of 1 (1.0g) in MeOH (20.0mL) was treated
with 10% palladium carbon (Pd–C, 100mg) and the
whole mixture was stirred at room temperature under
an H2 atmosphere for 13h. The reaction mixture was fil-
tered and then evaporation of the solvent under reduced
pressure furnished crude products. The products were
separated by HPLC [MeOH–H2O (65:35)] to give tetra-
hydrocurcumin (7, 385mg, 38.1%) and hexahydrocurcu-
min (8, 386mg, 38.0%). Hydrocurcumins 7 and 8 were
identified by comparison of the physical data with re-
ported values.11,12
3.5. Methylation of 1
3.6.4. Effects on the release of b-hexosaminidase from
RBL-2H3 cells and on enzyme activity of b-hexosamin-
idase. Inhibitory effects on the release of b-hexosamini-
dase in RBL-2H3 [Cell no JCRB0023, obtained from
Health Science Research Resources Bank (Osaka,
Japan)] were evaluated by a method reported previ-
ously.6 Briefly, RBL-2H3 cells in 24-well plates [2 · 105
cells/well in MEM containing 10% FCS, penicillin
(100units/mL), streptomycin (100lg/mL)] were sensi-
tized with anti-DNP IgE (0.45lg/mL). The cells were
washed with Siraganian buffer [119mM NaCl, 5mM
KCl, 0.4mM MgCl2, 25mM piperazine-N,N0-bis(2-
ethanesulfonic acid) (PIPES), and 40mM NaOH,
pH7.2] supplemented with 5.6mM glucose, 1mM
CaCl2, and 0.1% bovine serum albumin (BSA) (incuba-
tion buffer) and then incubated in 160lL of the incuba-
tion buffer for 10min at 37 ꢁC. After that, 20lL of test
sample solution was added to each well and incubated
for 10min, followed by an addition of 20 lL of antigen
(DNP-BSA, final concn 10lg/mL) at 37ꢁC for 10min
to stimulate the cells to degranulate. The reaction was
stopped by cooling in an ice bath for 10min. The sup-
ernatant (50lL) was transferred into 96-well plate and
incubated with 50lL of substrate (1mM p-nitrophe-
nyl-N-acetyl-b-D-glucosaminide) in 0.1M citrate buffer
(pH4.5) at 37ꢁC for 1h. The reaction was stopped by
adding 200lL of stop solution (0.1M Na2CO3/NaH-
CO3, pH10.0). The absorbance was measured with a
microplate reader at 405nm. The test sample was dis-
solved in dimethylsulfoxide (DMSO), and the solution
was added to incubation buffer (final DMSO concn
was 0.1%). The inhibition (%) of the release of b-hexos-
aminidase by the test samples was calculated by the fol-
lowing equation, and IC50 values were determined
graphically:
A solution of 1 (200mg) in N,N-dimethylformamide
(DMF, 5.0mL) was treated with methyl iodide (CH3I,
1.0mL) in the presence of sodium hydride (NaH,
40mg) and the mixture was stirred at room temperature
for 13h. The reaction mixture was poured into ice-water
and the whole was extracted with EtOAc. The EtOAc
extract was successively washed with saturated aqueous
NaHCO3 and brine, then dried over MgSO4 powder and
filtered. Removal of the solvent from the filtrate under
reduced pressure furnished a residue, which was purified
by HPLC [MeOH–H2O (75:25)] to give monome-
thylcurcumin (9, 107mg, 51.5%) and dimethylcurcumin
(10, 22mg, 10.2%). Methylcurcumins 9 and 10 were
identified by comparison of their physical data with re-
ported values.13
3.6. Bioassay methods
3.6.1. Reagents. Minimum essential medium eagle
(MEM) and anti-DNP IgE (Monoclonal Anti-DNP)
were purchased from Sigma; fetal calf serum (FCS)
was from Gibco; the ELISA kit for determination of
TNF-a (TNF-a, rat, code 3012) was from Biosource
International Co., Ltd; ELISA kit for determination of
IL-4 (IL-4, rat, code 2737) was from Amersham Phar-
macia Biotech Co., Ltd; the dinitrophenylated bovine
albumin (DNP-BSA) was prepared as described previ-
ously.17 Other chemicals were from Wako; 24-well and
96-well microplates were from Sumitomo Bakelite Co.,
Ltd.
3.6.2. Animals. Male ddY mice (25–30g) were purchased
from Kiwa Laboratory Animals (Wakayama, Japan).
The animals were maintained at a constant temperature
of 23 2ꢁC and were fed with standard laboratory
chow (MF, Oriental Yeast, Japan) for one week. Test
samples were suspended with 5% acacia solution, and
Inhibition ð%Þ ¼ ½1 ꢀ ðT ꢀ B ꢀ NÞ=ðC ꢀ NÞꢁ ꢂ 100
Control (C): DNP-BSA (+), test sample (ꢀ); test (T):
DNP-BSA (+), test sample (+); blank (B): DNP-BSA

H. Matsuda et al. / Bioorg. Med. Chem. 12 (2004) 5891–5898
5897
(ꢀ), test sample (+); normal (N): DNP-BSA (ꢀ), test
3.7. Statistics
sample (ꢀ).
Values are expressed as means SEM. One-way ana-
lysis of variance (ANOVA) followed by DunnettÕs test
was used for statistical analysis.
To clarify that the anti-allergic effects of samples are due
to the inhibition on b-hexosaminidase release, but not
the false positive from the inhibition of b-hexosamin-
idase activity. The cell suspension (5 · 107 cells) in
6mL of PBS was sonicated. The solution was then cen-
trifuged, and the supernatant was diluted with the incu-
bation buffer and adjusted to equal the enzyme activity
of the degranulation tested above. The enzyme solution
(45lL) and test sample solution (5lL) were transferred
into a 96-well microplate and enzyme activity was exam-
ined as described above. Under these conditions, it was
calculated that 40–70% of b-hexosaminidase was re-
leased from the cells in the control groups by determina-
tion of the total b-hexosaminidase activity after
sonication of the cell suspension.
Acknowledgements
A part of this work was supported by a grant from
Tokyo Biochemical Research Foundation (TBRF),
Japan, and by the Promotion and Mutual Aid Corpora-
tion for Private School of Japan.
References and notes
1. (a) Tanaka, Y.; Takagaki, Y.; Nishimune, T. Chem.
Pharm. Bull. 1991, 39, 2072–2076; (b) Cheong, H.; Choi,
E. J.; Yoo, G. S.; Kim, K. M.; Ryu, S. Y. Planta Med.
1998, 64, 577–578.
2. Kaul, S.; Hoffmann, A. ALTEX 2001, 18, 55–58.
3. Goldsby, R. A.; Kindt, T. J.; Osborne, B. A.; Kuby, J.
Immunology, 5th ed.; WH Freeman and Company: New
York, 2002.
3.6.5. Effects on the release of b-hexosaminidase from
RBL-2H3 induced by calcium ionophores, A 23187 and
ionomycin. The inhibitory effects on A23187 or ionomy-
cin-induced release of b-hexsosaminidase in RBL-2H3
were evaluated similarly to that of antigen-induced
degranulation.6i,19 However, in the case of A23187 and
ionomycin, anti-DNP IgE and DNP-BSA were not
added to the wells. Briefly, the cells were incubated in
160lL of the incubation buffer for 10min at 37 ꢁC. After
that, 20lL of test sample solution was added to each
well and incubated for 10min, followed by an addition
of 20lL of A23187 or ionomycin (final concn 1lM) at
37ꢁC for 10min to stimulate the cells to degranulate.
The reactions were stopped by cooling in an ice bath,
and b-hexsosaminidase activity in the medium was
determined as described above. Under these conditions,
it was calculated that 50–70% of b-hexosaminidase was
released from the cells.
4. (a) Yoshioka, T.; Fujii, E.; Endo, M.; Wada, K.; Toku-
naga, Y.; Shiba, N.; Hohsho, H.; Shibuya, H.; Muraki, T.
Inflamm. Res. 1998, 47, 476–481; (b) Jang, M. K.; Sohn,
D. H.; Ryu, J. H. Planta Med. 2001, 67, 550–552; (c) Lee,
S. K.; Hong, C. H.; Huh, S. K.; Kim, S. S.; Oh, O. J.; Min,
H. Y.; Park, K. K.; Chung, W. Y.; Hwang, J. K.
J. Environ. Pathol. Toxicol. Oncol. 2002, 21, 141–148; (d)
Hong, C. H.; Hur, S. K.; Oh, O. J.; Kim, S. S.; Nam, K.
A.; Lee, S. K. J. Ethnopharmacol. 2002, 83, 153–
159.
5. (a) Matsuda, H.; Ninomiya, K.; Morikawa, T.; Yos-
hikawa, M. Bioorg. Med. Chem. Lett. 1998, 8, 339–344; (b)
Matsuda, H.; Morikawa, T.; Ninomiya, K.; Yoshikawa,
M. Bioorg. Med. Chem. 2001, 9, 909–916; (c) Matsuda, H.;
Morikawa, T.; Toguchida, I.; Ninomiya, K.; Yoshikawa,
M. Heterocycles 2001, 55, 841–846; (d) Matsuda, H.;
Morikawa, T.; Ninomiya, K.; Yoshikawa, M. Tetrahedron
2001, 57, 8443–8453; (e) Matsuda, H.; Morikawa, T.;
Toguchida, I.; Ninomiya, K.; Yoshikawa, M. Chem.
Pharm. Bull. 2001, 49, 1558–1566; (f) Morikawa, T.;
Matsuda, H.; Ninomiya, K.; Yoshikawa, M. Biol. Pharm.
Bull. 2002, 25, 627–631.
6. (a) Matsuda, H.; Morikawa, T.; Tao, J.; Ueda, K.;
Yoshikawa, M. Chem. Pharm. Bull. 2002, 50, 208–215;
(b) Morikawa, T.; Matsuda, H.; Sakamoto, Y.; Ueda, K.;
Yoshikawa, M. Chem. Pharm. Bull. 2002, 50, 1045–1049;
(c) Matsuda, H.; Morikawa, T.; Ueda, K.; Managi, H.;
Yoshikawa, M. Bioorg. Med. Chem. 2002, 10, 3123–3128;
(d) Morikawa, T.; Tao, J.; Ueda, K.; Matsuda, H.;
Yoshikawa, M. Chem. Pharm. Bull. 2003, 51, 62–67; (e)
Tao, J.; Morikawa, T.; Ando, S.; Matsuda, H.; Yos-
hikawa, M. Chem. Pharm. Bull. 2003, 51, 654–662; (f)
Matsuda, H.; Morikawa, T.; Managi, H.; Yoshikawa, M.
Bioorg. Med. Chem. Lett. 2003, 13, 3197–3202; (g) Sun, B.;
Morikawa, T.; Matsuda, H.; Tewtrakul, S.; Harima, S.;
Yoshikawa M. J. Nat. Prod. 2004, 67, in press; (h)
Morikawa, T.; Sun, B.; Matsuda, H.; Wu, L. J.; Harima,
S.; Yoshikawa, M. Chem. Pharm. Bull. 2004, 52(10), in
press; (i) Matsuda, H.; Tewtrakul, S.; Morikawa, T.;
Yoshikawa, M. Bioorg. Med. Chem. 2004, 12, 4871–4876;
3.6.6. Effects on antigen-induced TNF-a and IL-4 pro-
duction from RBL-2H3 cells. Inhibitory effects on the re-
lease of TNF-a and IL-4 from RBL-2H3 were evaluated
by the method reported previously.6c,f RBL-2H3 cells
(2 · 105 cells/well) were sensitized with anti-DNP IgE
as described above. The cells were washed with MEM
containing 10% FCS, penicillin (100units/mL), and
streptomycin (100lg/mL), and exchanged with 320lL
of fresh medium. Then, 40lL of test sample solution
and 40lL of antigen (DNP-BSA, final concentration
was 10lg/mL) were added to each well and incubated
at 37ꢁC for 4h. The supernatant (50lL) was transferred
into 96-well ELISA plates and then TNF-a and IL-4
concentrations were determined using commercial ELI-
SA kits. The test samples were dissolved in DMSO, and
the solution was added to MEM (final DMSO concn
was 0.1%). The inhibition of production of TNF-a and
IL-4 from cells was calculated by the following equa-
tion, and IC50 values were determined graphically:
Inhibition % ¼ ½1 ꢀ ðT ꢀ NÞ=ðC ꢀ NÞꢁ ꢂ 100
Control (C): DNP-BSA (+), test sample (ꢀ); test (T):
DNP-BSA (+), test sample (+); normal (N): DNP-BSA
(ꢀ), test sample (ꢀ).

5898
H. Matsuda et al. / Bioorg. Med. Chem. 12 (2004) 5891–5898
(j) Matsuda, H.; Morikawa, T.; Xie, H.; Yoshikawa, M.
Planta Med., in press.
12. Itokawa, H.; Morita, H.; Midorikawa, I.; Aiyama, R.;
Morita, M. Chem. Pharm. Bull. 1985, 33, 4889–4893.
13. Ohtsu, H.; Xiao, Z.; Ishida, J.; Nagai, M.; Wang, H. K.;
Itokawa, H.; Su, C. Y.; Shih, C.; Chiang, T.; Chang, E.;
Lee, Y. F.; Tsai, M. Y.; Chang, C.; Lee, K. H. J. Med.
Chem. 2002, 45, 5037–5042.
14. (a) Futagami, Y.; Yano, S.; Horie, S.; Tsuchiya, S.;
Ikegami, F.; Sekine, T.; Yamamoto, Y.; Fujimori, H.;
Takamoto, K.; Watanabe, K. J. Trad. Med. 1996, 13, 430–
431 (in Japanese); (b) Yano, S.; Terai, M.; Shimizu, K. L.;
Futagami, Y.; Horie, S.; Tsuchiya, S.; Ikegami, F.; Sekine,
T.; Takamoto, K.; Saito, K.; Ueno, K.; Watanabe, K.
Natural Medicines 2000, 54, 325–329.
15. Nakata, Y.; Hide, I. Life Sci. 1998, 62, 1653–1657.
16. (a) Chan, M. M. Biochem. Pharmacol. 1995, 49, 1551–
1556; (b) Abe, Y.; Hashimoto, S.; Horie, T. Pharmacol.
Res. 1999, 39, 41–47; (c) Kang, B. Y.; Song, Y. J.; Kim, K.
M.; Choe, Y. K.; Hwang, S. Y.; Kim, T. S. Br. J.
Pharmacol. 1999, 128, 380–384.
7. (a) The rhizomes of C. zedoaria cultivated in China Japan
Taiwan and Korea were known to be containing many
sesquiterpenes as the principal constituents.5,7b However
from Thai zedoary, which was commonly cultivated in
Thailand and identified to be C. zedoaria by Thai
scientists,7c many curcuminoids (1–4) were isolated as
the principal constituents. The variety of chemical con-
stituents in different localities is of interest in regard to
chemotaxonomy; (b) Shibuya, H.; Yoshihara, M.; Kitano,
E.; Nagasawa, M.; Kitagawa, I. Yakugaku Zasshi 1986,
106, 212–216 (in Japanese) and literatures cited therein; (c)
Saralamp, P.; Chuakul, W.; Temsiririrkkul, R.; Clayton,
T. In Medicinal Plants in Thailand; Amarin Printing and
Publishing Public Co., Ltd: Bangkok, 1996; Vol. I.
8. Uehara, S.; Ysuda, I.; Akiyama, K.; Morita, H.; Takeya,
K.; Itokawa, H. Chem. Pharm. Bull. 1987, 35, 3298–
3304.
9. Osawa, T.; Sugiyama, Y.; Inayoshi, M.; Kawakishi, S.
Biosci. Biotechnol. Biochem. 1995, 59, 1609–1612.
10. Ohshiro, M.; Kuroyanagi, M.; Ueno, A. Phytochemistry
1990, 29, 2201–2205.
11. Sui, Z.; Salto, R.; Li, J.; Craik, C.; de Montellano, P. R. O.
Bioorg. Med. Chem. 1993, 1, 415–422.
17. Tada, T.; Okumura, K. J. Immunol. 1971, 106, 1002–1011.
18. Inagaki, N.; Goto, S.; Nagai, H.; Koda, A. Int. Arch.
Allergy Appl. Immunol. 1984, 74, 91–92.
19. Vincent-Schneider, H.; Stumptner-Cuvelette, P.; Lankar,
D.; Pain, S.; Raposo, G.; Benaroch, P.; Bonnerot, C. Int.
Immunol. 2002, 14, 713–722.