Absolute stereostructure of carabrane-type sesquiterpene and vasorelaxant-active sesquiterpenes from Zedoariae Rhizoma

Article abstract of DOI:10.1016/S0040-4020(01)00828-6

The aqueous acetone extract of Zedoariae Rhizoma was found to show inhibitory effects on contractions induced by high concentrations of potassium cation (K⁺) in isolated rat aortic strips. From the extract, two new carabrane-type sesquiterpenes

Full text of DOI:10.1016/S0040-4020(01)00828-6

TETRAHEDRON
Pergamon
Tetrahedron 57 +2001) 8443±8453
Absolute stereostructure of carabrane-type sesquiterpene and
vasorelaxant-active sesquiterpenes from Zedoariae Rhizoma
p
Hisashi Matsuda, Toshio Morikawa, Kiyofumi Ninomiya and Masayuki Yoshikawa
Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto 607-8412, Japan
Received 15 June 2001;accepted 8 August 2001
AbstractÐThe aqueous acetone extract of Zedoariae Rhizoma was found to show inhibitory effects on contractions induced by high
concentrations of potassium cation +K ) in isolated rat aortic strips. From the extract, two new carabrane-type sesquiterpenes, curcarabranols
1
A and B, were isolated together with a few carabrane-type sesquiterpenes, such as curcumenone and 4S-dihydrocurcumenone. Their absolute
stereostructures were determined on the basis of chemical and physicochemical evidence, which included the application of the modi®ed
Mosher's method and chemical conversion from curcumenone to curcarabranols A and B. In addition, several sesquiterpenes and diaryl-
heptanoids +e.g. germacrone, glechomanolide, isocurucmenol, b-eudesmol, and b-dictyopterol) showed potent inhibitory effects on con-
1
tractions induced by high concentrations of K in isolated rat aortic strips +inhibition .80% at 100 mM), while they did not inhibit
norepinephrine-induced contractions, so that the vasorelaxant activities of these sesquiterpenes were presumed to be dependent on their
calcium channel-blocking activity. q 2001 Elsevier Science Ltd. All rights reserved.
1
. Introduction
paper presents a full account of the absolute stereostructure
elucidation of carabrane-type sesquiterpenes, curcumenone
+1), 4S-dihydrocurcumenone +2), and curcarabranols A +3)
and B +4), which included the conversion of curcumenone
+1) into curcarbranols A +3) and B +4). In addition, we
describe vasorelaxant effects of the constituents on contrac-
tions induced by high concentrations of potassium cation
The Zingiberaceae plant Curcuma zedoaria Roscoe +the
common name, Zedoary) has been widely cultivated as a
vegetable or spice in South and Southeast Asian countries.
The rhizomes of this plant +Zedoariae Rhizoma) are used as
a stimulant, stomachic, carminative, diuretic, anti-diarrheal,
anti-emetic, anti-pyretic, depurator, and also to clean and
cure ulcers, wounds, and other kinds of skin disorders in
India and Southeast Asian countries. In Japanese and
Chinese traditional medicines, Zedoariae Rhizoma is
prescribed as a stomachic, emmenagogue, and for the treat-
ment of `Oketsu' syndrome caused by blood stagnation in
various traditional preparations. During the course of our
characterization studies on the bioactive constituents from
1
+high K ) and dl-norepinephrine +NE) in isolated rat aortic
strips.
2
2. Results and discussion
The isolation of the chemical constituents from Zedoariae
Rhizoma was carried out through the following procedure.
Zedoariae Rhizoma +cultivated in Szechwan province,
China, purchased from Tochimoto Tenkaido Co., Ltd,
Osaka) was extracted with 80% aqueous acetone at room
temperature. The relaxant effects of the 80% aqueous
1
Zingiberaceae plants, we found that the 80% acetone
extract of Zedoariae Rhizoma showed vasorelaxant,
2
3
hepatoprotective, and nitric oxide +NO) production
4
inhibitory activities. From the aqueous acetone extract,
we have isolated eleven sesquiterpenes called curcara-
branols A +3) and B +4), curcumenolactones A +5), B +6),
and C +7), 4-epicurcumenol +19), neocurcumenol +21),
gajutsulactones A +30) and B +31), and zedoarolides A
1
acetone extract on high K - and NE-induced contractions
in rat thoracic aorta were examined. Cumulative application
of the 80% aqueous acetone extract +6.25±50 mg/ml)
inhibited the sustained contractions induced by high K .
1
3
,4
+
32) and B +33). In the preceding papers, we reported
The IC50 value of the 80% aqueous acetone extract was
18 mg/ml as shown in Table 1, and that of nifedipine as a
reference drug was 6.4 nM. On the other hand, the 80%
aqueous acetone extract did not inhibit the contractions
induced by NE in isolated rat aortic strips.
the hepatoprotective and NO production inhibitory activities
of the sesquiterpene and diarylheptanoid constituents and
the structure elucidation of 5±7, 19, 21, and 30±33. This
Keywords: pharmacologically active compounds;terpenes and terpenoids;
cyclopropanes;stereochemistry.
Corresponding author. Tel.: 181-75-595-4633;fax: 181-75-595-4768;
e-mail: shoyaku@mb.kyoto-phu.ac.jp
The 80% aqueous acetone extract was partitioned in an ethyl
acetate and water mixture to give an ethyl acetate-soluble
portion and an aqueous phase. The aqueous phase was
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020+01)00828-6

8444
H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
1
Table 1. Inhibitory effects of 80% aqueous acetone extracts of Zedoariae Rhizoma on the contractions induced by high K concentration and norepinephrine in
isolated rat thoracic aorta
Inhibition +%)
1
High K +54 mM)
6
.25 mg/ml
12.5 mg/ml
25.0 mg/ml
50.0 mg/ml
pp
pp
pp
pp
80% Aqueous acetone extract
15.8^3.4
nM
32.2^2.7
69.8^6.2
98.8^6.5
3
10 nM
30 nM
100 nM
pp
pp
pp
pp
Nifedipine
28.4^4.2
64.1^4.4
80.6^3.4
84.0^3.7
Contraction +%)
dl-Norepinephrine
100 nM
Conc.
10 nM
1 mM
Control
0% Aqueous acetone extract
Prazosin
21.6^2.3
31.9^2.2
9.9^1.4
67.0^2.7
75.1^5.3
46.6^5.6
100.0^8.8
105.3^5.2
72.2^9.7
pp
8
50.0 mg/ml
210
10
pp
p
M
M
29
pp
pp
1
0
0.8^0.7
9.9^6.4
41.0^9.5
p
pp
Each value represents the mean^SEM +nˆ4±5). Signi®cantly different from the control: p,0.05, p,0.01.
further extracted with 1-butanol to give a 1-butanol-soluble
portion and a water-soluble portion. The ethyl acetate-
soluble portion was subjected to silica gel, silver nitrate-
yield. The stereostructures of the 8-hydroxyl groups in 1b
and 1c and the 7-isopropyl group in 1d were clari®ed by the
NOESY experiment, which showed nuclear Overhauser
effect +NOE) correlations between the following proton
pairs [+1b: 1-H and 6a-H;5-H and 6 b-H, 9b-H;8-H and
9b-H, 14-H ; 9b-H and 14-H ), +1c: 1-H and 9a-H;5-H
treated silica gel, and ODS column chromatography and
®
natural medicine), 4S-dihydrocurcumenone +2, 0.0011%),
5
nally HPLC to furnish curcumenone +1, 0.041% from the
6
3
3
2
2
and curcarabranols A +3, 0.00030%) and B +4, 0.00030%),
together with 30 sesquiterpenes +5±26, 30, 31, 34, 35, 37±
and 6b-H;8-H and 9 a-H;9 b-H and 14-H ), +1d: 1-H and
3
7-H;5-H and 8-H, 14-H ;8-H and 9 b-H, 14-H ; 9b-H and
3
3
40) and diarylheptanoid, curcumin +41). The 1-butanol-
soluble portion was also separated by the above-mentioned
chromatography to give seven sesquiterpenes +26±29, 32,
14-H ), as shown in Fig. 1]. Finally, the absolute stereo-
3
structure of the 8-hydroxyl group in 1d was determined
1
0
by application of the modi®ed Mosher's method for the
+R)- and +S)-2-methoxy-2-tri¯uorophenylacetate +MTPA
esters, 1e and 1f), which were prepared by treatment of 1d
with +R)- and +S)-2-methoxy-2-tri¯uoromethylphenylacetic
acid in the presence of 1-ethyl-3-+3-dimethylaminopropyl)-
carbodiimide hydrochloride +EDC´HCl) and 4-dimethyl-
aminopyridine +4-DMAP). As shown in Fig. 1, the signals
due to protons on the 6, 7, 11, 12 and 13-carbons in the
+S)-MTPA ester +1f) were observed at a lower ®eld than
those of +R)-MTPA ester +1e) +Dd, positive), while the
signals due to protons on the 1, 9 and 14-carbons in 1f
were observed at higher ®elds than those of 1e +Dd,
negative). Thus, the absolute con®guration at the 8-position
in 1d has been shown to be S. Consequently, the absolute
stereostructure of 1 was determined.
33, 36) and bis+4-hydroxycinnamoyl)methane +42) +see
Scheme 1). The details of the isolation procedure were
described previously.
3
2
4
.1. Absolute stereostructures of curcumenone &1) and
S-dihydrocurcumenone &2)
Curcumenone +1) has been previously isolated as a principal
carabrane-type sesquiterpene from Zedoariae Rhizoma
and its relative stereostructure was clari®ed by nuclear5
Overhauser effect spectroscopy +NOESY) experiments.
The absolute stereostructure of 1 was deduced on the
6
±8
5
basis of biotransformation and hypothetical biogenesis
from germacrone +11) to 1 via +4S,5S)-germacrone 4,5-
epoxide. However, the absolute con®guration of the germa-
crone 4,5-epoxide was revised to the +4S,5R) con®guration
+
4S-Dihydrocurcumenone +2) was isolated as a colorless oil
9
14), so that the absolute stereostructure of 1 was concluded
27
with negative optical rotation +[a]
impact +EI)-MS of 2 showed a molecular ion +M ) peak at
ˆ25.18). The electron
D
1
as being left unclari®ed +Scheme 2). In order to determine
the absolute stereostructure of 1, we carried out the
chemical modi®cation from 1 to 1d and the application of
1
m/z 236 in addition to fragment ion peaks at m/z 218 +M ±
H O) and m/z 68 +base peak) and the high-resolution MS
2
1
0
the modi®ed Mosher's method.
analysis of the molecular ion peak revealed the molecular
formula of 2 to be C H O . The IR spectrum of 2 showed
1
5
24
2
2
absorption bands at 3436, 1678, and 1053 cm , which were
1
Thus, 1 was treated with 1,2-bis+trimethylsilyloxy)ethane
1
1
and trimethylsilyl tri¯uoromethane sulfonate +TMSOTf)
to give the 4-acetal derivative +1a) in 86% yield. The
-acetal derivative +1a) was subjected to NaBH reduction
assigned to hydroxyl, conjugated carbonyl, and cyclo-
1
propane functions, respectively. The H NMR +CDCl
)
3
1
3
4
and C NMR +Table 2) spectra of 2 showed signals assign-
able to a cyclopropane [d 0.46 +dt, Jˆ4.9, 6.4 Hz, 1-H), 0.65
4
in the presence of cerium chloride +CeCl ´6H O) to yield the
3
2
8a-hydroxyl +1b, 68% yield) and the 8b-hydroxyl +1c, 23%
yield) derivatives. Hydrogenation of 1b furnished 1d in 45%
+m, 5-H)], four methyls [d 1.12 +s, 14-H
3
), 1.18 +d, Jˆ
), 1.79, 2.09 +both s, 12 and 13-H )], and a
3
6.1 Hz, 15-H
3

H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
8445
Scheme 1.

Scheme 2. Chemical correlation of curcumenone +1) with curcarabranols A +3) and B +4). Reagents and conditions: +a) 1,2-bis+trimethylsilyloxy)ethane, TMSOTf, CH
8C, 1b 68%, 1c 23%, +c) Pd+OH) ±C, H , MeOH, r.t., 45%, +d) +R)- or +S)-MTPA, EDC´HCl, DMAP, CH Cl , r.t., 1e 57%, 1f 32%, +e) p-TsOH´H O, dioxane±H O [2:1 +v/v)], r.t., quant., +f) BH
, 3 M NaOH, r.t., +h) PCC, CH Cl , r.t., 3 41%, 4 14% from 1g.
2
2 4
Cl , 2788C, 86%, +b) NaBH ,CeCl
3 2
´H O, MeOH,
´THF, r.t., +g) 30%
0
2
2
2
2
2
2
3
H
2
O
2
2
2

H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
8447
Figure 1. NOE correlations of 1a±1d and application of modi®ed Mosher's method to 1e and 1f.
methine bearing an oxygen function [d 3.79 +m, 4-H)]
together with four methylenes +2, 3, 6, 9-H ) and four
quaternary carbons +7, 8, 10, 11-C). The proton and carbon
signals of 2 were shown to be very similar to those of a
dihydrocurcumenone 4-epimeric mixture, which was
obtained from the transformation products of germacrone
+10) by cultured cells and the stereostructure was only
2
6
,7
deduced on the basis of biogenetic considerations. The
13
Table 2. C NMR data for 1a±1d, 1g, 4S-dihydrocurcumenone +2), and curcarabranols A +3) and B +4)
1
a
1b
1c
1d
1g
2
3
4
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-8
24.1
23.7
39.3
109.8
24.7
28.1
128.3
202.0
49.1
19.9
147.1
23.4
23.5
19.0
23.7
30.4
24.1
39.3
110.1
24.8
26.3
129.6
66.0
40.4
16.7
131.3
20.5
20.5
19.4
23.7
28.7
23.8
39.4
110.0
25.9
23.5
127.4
66.0
41.9
15.3
129.7
15.3
20.5
22.0
23.7
29.2
24.2
39.4
110.2
25.4
22.2
43.3
69.7
44.2
18.3
26.0
16.2
20.5
21.5
23.7
21.6
23.5
44.3
209.3
25.0
24.4
140.7
115.9
32.6
18.1
72.9
28.7
28.9
20.2
30.0
24.2
25.2
39.5
67.8
24.6
28.1
128.3
202.0
49.1
19.9
147.1
23.4
23.4
19.1
23.7
33.6
23.4
43.6
208.5
23.8
29.3
55.8
218.8
50.4
22.2
71.6
28.5
25.2
19.9
30.1
23.1
23.4
43.8
208.4
24.0
23.7
54.0
217.2
49.2
18.6
72.9
28.3
25.0
19.0
30.1
C-9
C-10
C-11
C-12
C-13
C-14
C-15
a
a
a
a
a
a
a
a
a
a
a
a
ZOCH CH
2
2
OZ
64.6
64.6
64.6
64.6
Measured in CDCl
3
at 125 MHz.
May be interchangeable within the same column.
a
Figure 2. H±H COSY, HMBC, and NOE correlations of 2 and application of modi®ed Mosher's method to 2a and 2b.

8448
H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
Scheme 3. Reagents and conditions: +a) CrO
3
2 2
, pyridine, r.t., quant. +b) +R)- or +S)-MTPA, EDC´HCl, DMAP, CH Cl , r.t., 2a 74%, 2b 84%.
1
planar structure of 2 was con®rmed by the H± H and
C± H correlation spectroscopy +H±H and C±H COSY),
1
3-H )], three methyls [d 1.08, 1.20 +both s, 12, 13-H ),
2 3
1.13 +s, 14-H )], and a hydroxyl group [d 4.22 +br s,
3
1
3
1
and heteronuclear multiple bond correlation +HMBC)
experiments as shown in Fig. 2. The stereostructure of 2
was elucidated on the basis of the NOESY experiment.
Furthermore, oxidation of 2 with chromium trioxide
11-OH)] together with two methylene +6, 9-H ), a methine
2
+7-H), and three quaternary carbons +8, 10, 11-C).
The planar structure of 3 was constructed on the basis of
H±H and C±H COSY and HMBC experiments. Thus, the
H±H COSY experiment on 3 indicated the presence of
the partial structure +1-C±3-C and 1-C±7-C), represented
by the bold line in Fig. 3. In the HMBC experiment, long-
range correlations were observed between the following
protons and carbons of 3 +3-H , 15-H and 4-C;7-H, 9-H
2
+
CrO ) in pyridine furnished 1, so that the absolute stereo-
3
structure of 2 was characterized except for the 4-position.
Finally, the modi®ed Mosher's method was also applied to
both 4-MTPA esters 2a and 2b prepared from 2 +Scheme 3).
As shown in Fig. 2, the proton signals due to the 1, 2, 3, 5, 6,
9
, and 14-positions in the +R)-MTPA ester +2a) appeared in
2
3
higher ®elds than those of the +S)-MTPA ester +2b), while
the signals ascribable to 15-C of 2a was observed in lower
and 8-C;9-H , 14-H and 10-C;12-H , 13-H and 11-C), so
2 3 3 3
that the connectivities of the quaternary carbons +4, 8, 10,
11-C) in 3 were clari®ed. The above-mentioned evidence
led us to con®rm the skeleton of 3 to be 4,8-dioxo-11-
hydroxycarabrane.
®
elds as compared to that of 2b. This evidence indicated the
absolute con®guration at the 4-position in 2 to be S, and the
absolute stereostructure of 2 was elucidated.
2.2. Absolute stereostructures of curcarabranols A &3)
and B &4)
Furthermore, the relative stereostructure of 3 was charac-
terized by the difference nuclear Overhauser effect +dif.
NOE) experiment. Namely, the NOE correlations were
observed between the following proton pairs: 1-H and
Curcarabranol A +3) was isolated as a colorless oil with
2
6
negative optical rotation +[a]D ˆ2104.08). The molecular
6a-H, 9a-H;5-H and 6 b-H, 7-H, 14-H
9b-H and 14-H +Fig. 3).
;7-H and 14-H ;
3 3
formula C H O of 3 has been determined from the molec-
24
3
1
5
3
1
ular ion peak at m/z 252 +M ) in the EI-MS of 3 and by high-
resolution MS measurement. The IR spectrum of 3 showed
absorption bands ascribable to hydroxyl and carbonyl
Curcarabranol B +4) was also isolated as a colorless oil with
ˆ177.08). The positive and
2
6
positive optical rotation +[a]
negative-ion fast atom bombardment +FAB)-MS of 4
D
2
1
1
groups at 3494, 1752, and 1713 cm . The H NMR
+
1
CDCl ) and C NMR +Table 2) spectra of 3 showed signals
3
1
showed a quasimolecular ion peak at m/z 253 +M1H)
and m/z 251 +M±H) , respectively. High-resolution MS
3
2
assignable to a cyclopropane [d 0.44 +dt, Jˆ5.2, 6.0 Hz,
-H), 0.59 +ddd, Jˆ5.2, 8.2, 8.2 Hz, 5-H)], a 3-oxobutyl
function [d 1.62 +m, 2-H ), 2.16 +s, 15-H ), 2.52 +m,
1
analysis revealed the molecular formula of 4 to be
C H O , which was the same as that of 3. The IR spectrum
15 24 3
2
3
Figure 3. H±H COSY, HMBC, and NOE correlations of 3 and 4.

H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
8449
1
Table 3. Inhibitory effects of constituents from Zedoariae Rhizoma on the contractions induced by high K concentration and norepinephrine in isolated rat
thoracic aorta
Inhibition +%)
1
High K +54 mM)
a
NE
Conc.
3 mM
10 mM
30 mM
100 mM
1 mM +NE)
Sesquiterpenes
1) Carabrane type
Curcumenone +1)
S-Dihydrocurcumenone +2)
Curcarabranol A +3)
2.9^0.9
0.8^0.8
1.4^3.3
0.0^1.8
0.9^1.5
1.8^0.8
0.0^0.9
5.4^1.5
2.6^1.2
10.2^3.4
7.6^2.3
8.2^3.2
10.8^1.6
3.6^1.3
12.8^2.7
11.6^3.3
12.8^2.6
21.9^11.1
22.8^3.3
25.4^4.0
2.5^1.4
2.6^6.1
218.3^12.3
26.5^5.3
pp
4
44.9^3.8
39.5^4.2
36.8^0.8
37.7^11.7
51.6^9.3
16.6^1.9
p
pp
pp
20.9^2.9
20.9^1.2
25.4^7.8
31.7^5.3
8.8^1.4
p
Curcarabranol B +4)
pp
pp
pp
Curcumenolactone A +5)
Curcumenolactone B +6)
Curcumenolactone C +7)
pp
2
) Germacrane type
p
pp
pp
pp
p
pp
Furanodiene +8)
4.1^1.5
4.4^0.6
5.6^0.9
3.2^1.4
20.5^1.0
20.2^2.0
13.3^1.5
10.4^1.8
19.3^3.0
7.7^3.9
9.3^4.8
4.8^5.0
33.9^3.3
29.2^3.8
68.4^5.4
26.1^3.3
36.8^9.5
17.3^6.8
60.3^5.9
76.7^7.4
94.7^1.8
74.7^2.2
92.0^6.3
47.6^6.3
6.8^1.5
1.5^2.8
27.3^1.7
23.2^3.2
12.9^10.0
23.0^11.8
pp
pp
pp
pp
pp
Zederone +10)
Germacrone +11)
3-Hydroxygermacrone +12)
Glechomanolide +13)
1)-Germacrone 4,5-epoxide
p
pp
1
+
+
14)
p
pp
Curdione +15)
Neocurdione +16)
Dehydrocurdione +17)
9.8^4.3
20.3^1.8
2.7^0.8
14.8^4.8
12.5^6.0
6.1^1.2
23.7^5.4
30.4^10.4
13.9^2.0
44.2^7.2
3.6^6.5
217.7^8.9
210.3^4.2
pp
pp
69.7^12.9
pp
30.3^3.0
3
) Guaiane type
p
pp
Curcumenol +18)
-Epicurcumenol +19)
Isocurcumenol +20)
Neocurcumenol +21)
Procurcumenol +22)
0.6^0.4
2.9^1.9
4.1^0.5
0.1^0.9
4.7^1.8
22.4^0.9
3.8^1.6
1.7^1.1
1.5^0.6
0.3^1.2
13.8^3.0
2.1^1.5
11.8^3.7
7.5^3.9
11.0^3.1
8.1^0.6
18.5^2.7
6.9^3.9
54.5^5.9
52.1^5.7
88.5^2.5
58.7^5.6
31.9^9.3
29.8^5.8
40.0^2.7
43.3^3.8
24.2^1.5
19.6^5.5
27.8^3.5
2.9^3.4
13.7^0.3
25.6^8.2
6.5^6.9
pp
pp
pp
p
4
pp
pp
55.3^6.4
14.2^3.2
20.4^7.0
16.9^5.7
23.1^2.6
19.0^2.3
p
Isoprocurcumenol +23)
Alismoxide +24)
a,11a-Epoxy-5b-hydroxy-
pp
pp
pp
p
7
29.9^5.7
9
-guaiaen-8-one +25)
Aerugidiol +26)
Zedoarondiol +27)
0.4^0.4
4.8^0.7
2.5^1.5
6.9^1.9
9.2^1.7
23.4^1.5
1.1^1.6
9.2^1.7
4.7^2.7
1.4^0.8
11.5^2.9
6.0^3.5
17.1^3.4
29.1^5.2
4.5^3.8
2.7^0.9
16.3^3.9
8.4^3.9
21.6^3.8
42.0^5.8
13.5^2.7
2.3^4.3
5.6^4.3
21.6^3.0
22.8^1.3
1.7^1.3
p
Isozedoarondiol +28)
Zedoalactone B +29)
Zedoarolide A +32)
Zedoarolide B +33)
p
p
pp
pp
12.7^3.0
19.7^3.0
21.6^3.3
p
p
pp
23.5^5.3
4
5
) Bisaborane type
+
p
pp
pp
pp
pp
1)-ar-Turumerone +34)
Bisacumol +35)
Bisacurone +36)
3.7^0.7
20.8^0.5
24.7^1.1
11.0^2.1
6.5^2.1
25.4^2.5
37.2^6.3
56.5^6.2
0.4^2.0
78.3^6.4
75.5^8.1
5.0^1.9
9.8^6.5
212.2^3.6
28.7^6.7
) Eudesmane type
b-Eudesmol +37)
b-Dictyopterol +38)
) Elemane type
Curzerenone +39)
) Xanthane type
Curcumadione +40)
Diarylheptanoids
Curcumin +41)
pp
pp
pp
pp
pp
pp
5.1^1.4
10.8^3.4
29.0^4.1
53.1^5.5
90.9^3.7
93.1^1.4
95.0^2.5
88.1^3.3
20.2^12.5
9.3^7.2
pp
6
7
p
pp
pp
pp
3.5^1.1
1.9^1.0
14.9^1.9
38.0^5.5
14.3^1.5
79.0^8.7
27.5^2.1
210.3^4.2
215.2^7.5
37.7^12.5
7.3^1.5
pp
pp
pp
pp
pp
pp
6.8^2.8
9.3^3.5
23.3^7.5
22.9^3.4
46.1^8.8
62.7^7.6
75.7^8.9
87.9^3.8
p
pp
Bis+4-hydroxycinnamoyl)-
methane +42)
38.4^2.3
p
pp
Each value represents the mean^SEM +nˆ4±6). Signi®cantly different from the control: p,0.05, p,0.01.
a
24
Sample concentrations were 10 M.
2
1
of 4 showed absorption bands at 3494, 1762, and 1709 cm
due to hydroxyl and carbonyl functions. The proton and
carbon signals in the H NMR +CDCl ) and C NMR
the protons and quaternary carbons were determined on the
basis of H±H, C±H COSY and HMBC experiments, and the
relative structure with the 7b-isopropyl group was also
elucidated by NOE correlations +1-H and 6a-H, 7-H,
1
13
3
spectra +Table 2) of 4 were shown to be very similar to
those of 3;a cyclopropane [ d 0.55 +dt, Jˆ6.9, 7.3 Hz,
9a-H;5-H and 6 b-H, 14-H ; 6a-H and 7-H;9 b-H and
3
14-H );thus 4 was characterized to be the 7-position isomer
3
1
-H), 0.72 +m, 5-H)], a 3-oxobutyl function [d 1.63 +dt,
Jˆ7.3, 7.3 Hz, 2-H ), 2.15 +s, 15-H ), 2.48 +m, 3-H )],
of 3.
2
3
2
three methyls [d 1.11 +s, 14-H ), 1.12, 1.17 +both s, 12,
3
1
correlation and connectivities of 4 were observed between
3-H )], and a hydroxyl group [d 4.49 +br s, 11-OH)]. The
Next, in order to clarify the absolute stereostructures of
curcarabranols A +3) and B +4), we carried out the
3

8450
H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
conversion of curcumenone +1) into 3 and 4. At ®rst, acid
treatment of 1b or 1c with p-toluenesulfonic acid +p-
TsOH´H O) furnished 1g, in which the hydroxyl group
migrated from the 8-position to the 11-position and the
ketal group was removed. Then, 1g was treated with
merone +34), bisacumol +35), b-eudesmol +37), b-dicytop-
terol +38), and curzerenone +39) relaxed high K -induced
1
contractions by their calcium channel-blocking activities
like nifedipine. The vasorelaxant effect of these active
sesquiterpenes may be related to the traditional medicinal
value of Zedoariae Rhizoma as the treatment effect of the
`Oketsu' syndrome caused by blood stagnation.
2
borane±tetrahydrofuran complex +BH ´THF), followed by
3
H O oxidation to yield a complex mixture of the triol +1h),
2 2
which was subsequently subjected to oxidation with pyridi-
nium chlorochromate +PCC) to give 3 +41% from 1g) and 4
+
14% from 1g) in an approximate 3:1 ratio. On the basis of
3. Experimental
the chemical transformation from 1 to 3 and 4, the absolute
stereostructures of curcarabranols A +3) and B +4) were
determined as shown.
The following instruments were used to obtain physical
data: speci®c rotations, Horiba SEPA-300 digital polari-
meter +lˆ5 cm);UV spectra, Shimadzu UV-1200 spectro-
1
2.3. Vasorelaxant effects of isolated constituents from
Zedoariae Rhizoma
meter;IR spectra, Shimadzu FTIR-8100 spectrometer;
H
1
3
NMR spectra, JNM-LA500 +500 MHz); C NMR spectra,
JNM-LA500 +125 MHz) spectrometer with tetramethyl-
silane as an internal standard;MS and high-resolution
MS, JEOL JMS-GCMATE mass spectrometer, JEOL
JMS-SX 102A mass spectrometer.
As shown in Table 3, 16 sesquiterpenes [curcumenolactone
B +6, IC ˆ100 mM), furanodiene +8, 67 mM), zederone
5
0
+
10, 46 mM), germacrone +11, 19 mM), 13-hydroxygerma-
crone +12, 52 mM), glechomanolide +13, 35 mM), neo-
curdione +16, 54 mM), curcumenol +18, 92 mM), 4-
epicurcumenol +19, 90 mM), isocurcumenol +20, 26 mM),
neocurcumenol +21, 85 mM), +1)-ar-turumerone +34,
The following experimental conditions were used for
chromatography: ordinary-phase column chromatography;
Silica gel BW-200 +Fuji Silysia Chemical, Ltd., 150±
350 mesh), reversed-phase column chromatography;
Chromatorex ODS DM1020T +Fuji Silysia Chemical,
Ltd., 100±200 mesh);TLC, pre-coated TLC plates with
Silica gel 60F254 +Merck, normal phase) and Silica gel
RP-18 F254S +Merck, reversed phase);HPTLC, pre-coated
TLC plates with Silica gel 60F254 +Merck, normal phase),
Silica gel RP-18 WF254S +Merck, reversed phase). Detection
3
8 mM), bisacumol +35, 37 mM), b-eudesmol +37, 16mM),
b-dictyopterol +38, 9 mM), and curzerenone +39, 38 mM)]
and two diarylheptanoids [curcumin +41, 32 mM) and bis-
+4-hydroxycinnamoyl)methane +42, 22 mM)] relaxed the
sustained contractions induced by high concentrations of
1
K . Especially, ®ve sesquiterpenes +11, 13, 20, 37, 38)
showed potent relaxation, and these results suggested that
activities of germacrane- and eudesmane-type sesquiter-
penes were stronger than those of the other types of ses-
quiterpene. Furthermore, compounds having an exo-methyl-
ene moiety, such as 20, and 37±39, showed potent activity.
On the other hand, polyoxygenated sesquiterpenes such as
was completed by spraying with 1% Ce+SO
aqueous H SO followed by heating.
) ±10%
4 2
2
4
3.1. Extraction and isolation
2
6±29, 33, and 36 showed weak activity. These results led
Curcumenone +1), 4S-dihydrocurcumenone +2), and cur-
carabranols A +3) and B +4) were isolated from the dried
Zedoariae Rhizoma cultivated in Szechwan, China as
us to presume the following structural requirements of the
isolated sesquiterpenes for activity:
3
described earlier.
1
. germacrane- and eudesmane-type sesquiterpenes were
active structures;
2
7
3.1.1. 4S-Dihydrocurcumenone &2). Colorless oil, [a]D
25.18 +cˆ0.1, CHCl ). High-resolution EI-MS: Calcd for
ˆ
2. the exo-methylene moiety enhanced the activity;and
. polyoxygenated sesquiterpenes exhibited weak activity.
3
1
C H O +M ): 236.1776. Found: 236.1767. UV [EtOH,
3
1
5
24
2
nm +log e)]: 254 +2.90). IR +®lm): 3436, 2924, 1678,
1599, 1053 cm . H NMR +500 MHz, CDCl ) d: 0.46
2
1
1
In addition, inhibitory effects of the isolated constituents on
NE-induced contractions were examined in isolated thoracic
aorta of rat. As a result, diarylheptanoids +41, 42) slightly
inhibited the contractions. However, other sesquiterpene
constituents did not show the signi®cant inhibition as
shown in Table 3.
3
+1H, dt, Jˆ4.9, 6.4 Hz, 1-H), 0.65 +1H, m, 5-H), 1.12 +3H,
s, 14-H ), 1.18 +3H, d, Jˆ6.1 Hz, 15-H ), 1.33 +1H, m, 2-H),
3
3
1.46±1.50 +3H, m, 2-H and 3-H ), 1.79, 2.09 +3H each, both
2
s, 12 and 13-H ), 2.52, 2.56 +2H, ABq, Jˆ15.6 Hz, 9-H ),
3
2
1
3
2.82 +2H, br s, 6-H ), 3.79 +1H, m, 4-H). C NMR +125
2
1
MHz, CDCl ) d : given in Table 2. EI-MS: m/z 236 +M ,
15), 218 +M ±H O, 20), 68 +100).
2
3
C
1
It is well known that high K -induced contractions in
smooth muscles are the result of an increase in intracellular
1
21
26
3.1.2. Curcarabranol A &3). Colorless oil, [a]D ˆ2104.08
Ca , and calcium antagonists such as nifedipine inhibit the
voltage-dependent calcium channel, thereby inhibiting the
contractions in the depolarized aortic strips, but they show
+cˆ0.1, CHCl ). High-resolution EI-MS: Calcd for
3
1
C H O +M ): 252.1726. Found: 252.1741. IR +®lm):
1
5
24
3
1
2
21 1
3494, 2930, 1752, 1713, 1020 cm . H NMR +500 MHz,
weak inhibitory effects on the NE-induced contractions.
The results in the present experiments suggested that curcu-
menolactone B +6), furanodiene +8), zederone +10), germa-
crone +11), 13-hydroxygermacrone +12), glechomanolide
CDCl ) d: 0.44 +1H, dt, Jˆ5.2, 6.0 Hz, 1-H), 0.59 +1H, ddd,
3
Jˆ5.2, 8.2, 8.2 Hz, 5-H), 1.08, 1.20 +3H each, both s, 12 and
13-H ), 1.13 +3H, s, 14-H ), 1.46 +1H, m, 6a-H), 1.62 +2H,
3
3
+
+
13), neocurdione +16), curcumenol +18), 4-epicurcumenol
19), isocurcumenol +20), neocurcumenol +21), +1)-ar-turu-
m, 2-H ), 2.16 +3H, s, 15-H ), 2.22, 2.52 +2H, ABq, Jˆ
2
3
15.0 Hz, 9-H ), 2.37 +1H, dd, Jˆ5.5, 7.0 Hz, 7-H), 2.52
2

H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
8451
+
2H, m, 3-H ), 2.54 +1H, m, 6b-H), 4.22 +1H, br s, 11-OH).
2
6.4 Hz, 1-H), 1.03, 1.31 +3H each, both s, 14 and 15-H3),
1.39 +2H, m, 2-H ), 1.55 +1H, dd, Jˆ6.1, 14.8 Hz, 9a-H),
1
3
C NMR +125 MHz, CDCl ) d : given in Table 2. EI-MS:
3
C
2
1
m/z 252 +M , 5), 234 +M ±H O, 25), 43 +100).
1
1.68, 1.79 +3H each, both s, 12 and 13-H ), 1.73 +2H, t, Jˆ
2
3
8
.8 Hz, 3-H ), 1.99 +1H, dd, Jˆ6.1, 14.8 Hz, 9b-H), 2.03
2
2
6
3
.1.3. Curcarabranol B &4). Colorless oil, [a]D ˆ177.08
+1H, m, 6a-H), 2.69 +1H, dd, Jˆ6.4, 10.2 Hz, 6b-H), 3.92
13
+
Calcd for C H O +M1H) : 253.1804. Found: 253.1818.
cˆ0.1, CHCl ). High-resolution positive-ion FAB-MS:
+4H, m, ZOZCH CH ZOZ), 4.61 +1H, dd, Jˆ6.1, 6.1 Hz,
3
2
2
1
1
5
25
3
8-H). C NMR +125 MHz, CDCl ) d : given in Table 2.
3 C
2
IR +®lm): 3494, 2973, 1762, 1709, 1025 cm . H NMR
1
1
1
1
EI-MS: m/z 280 +M , 2), 262 +M ±H O, 10), 87 +100).
2
+
0
500 MHz, CDCl ) d: 0.55 +1H, dt, Jˆ6.9, 7.3 Hz, 1-H),
3
2
4
.72 +1H, m, 5-H), 1.11 +3H, s, 14-H ), 1.12, 1.17 +3H
3
1c: Colorless oil, [a]D ˆ251.88 +cˆ0.1, CHCl ). High-
3
1
resolution EI-MS: Calcd for C H O +M ): 280.2065.
Found: 280.2039. IR +®lm): 3475, 2921, 1762, 1651,
1093, 1063 cm . H NMR +500 MHz, CDCl ) d: 0.10
each, both s, 12 and 13-H ), 1.63 +2H, dt, Jˆ7.3, 7.3 Hz,
3
17 28
3
2
1
1
1
2
-H ), 2.00 +1H, m, 6b-H), 2.07 +1H, m, 7-H), 2.15 +3H, s,
2
2
1
1
5-H ), 2.20 +1H, m, 6a-H), 2.46, 2.61 +2H, ABq, Jˆ
3
3
7.1 Hz, 9-H ), 2.48 +2H, m, 3-H ), 4.49 +1H, br s,
2
+1H, dt, Jˆ5.2, 7.3 Hz, 1-H), 0.54 +1H, br dd, Jˆca. 5,
2
1
1-OH). C NMR +125 MHz, CDCl ) d : given in Table
3
3
C
8 Hz, 5-H), 1.13, 1.31 +3H each, both s, 14 and 15-H ),
3
1
. Positive-ion FAB-MS: m/z 253 +M1H) . Negative-ion
1.39 +2H, m, 2-H ), 1.55, 1.64 +3H each, both s, 12 and
2
2
FAB-MS: m/z 251 +M±H) .
.2. Ketalization of curcumenone &1)
A solution of 1 +30.0 mg, 128 mmol) in CH Cl +2.0 ml) was
13-H ), 1.69 +2H, t, Jˆ6.7 Hz, 3-H ), 1.71 +1H, dd, Jˆ4.0,
3
2
1
4.7 Hz, 9a-H), 1.95 +1H, dd, Jˆ4.0, 14.7 Hz, 9b-H), 2.50
3
+1H, dd-like, 6b-H), 2.59 +1H, m, 6a-H), 3.93 +4H, m,
ZOZCH CH ZOZ), 4.57 +1H, dd, Jˆ4.0, 4.0 Hz, 8-H).
2
2
1
3
C NMR +125 MHz, CDCl ) d : given in Table 2.
3
2
2
C
1
1
treated with 1,2-bis+trimethylsilyloxy)ethane +38 ml) in the
presence of trimethylsilyl tri¯uoromethanesulfonate
EI-MS: m/z 280 +M , 5), 262 +M ±H O, 10), 87 +100).
2
+TMSOTf) as a catalyst and the whole mixture was stirred
at 2788C for 14 h. The reaction mixture was quenched in
3.4. Hydrogenation of 1b
pyridine, then poured into saturated aqueous NaHCO and
3
A solution of 1b +9.0 mg, 32 mmol) in MeOH +0.5 ml) was
treated with 10% palladium hydroxide carbon [Pd+OH) ±C,
the whole was extracted with CH Cl . The CH Cl extract
2
2
2
2
2
was washed with saturated aqueous NaCl and dried over a
:1 mixture of K CO and MgSO powder, then the mixture
10.0 mg] and the whole mixture was stirred at room
temperature under a H atmosphere for 3 h. The catalyst
1
2
3
4
2
was ®ltered. Removal of the solvent from the ®ltrate under
reduced pressure furnished a residue, which was puri®ed
by silica gel column chromatography [2.0 g, n-hexane±
AcOEtˆ10:1] to give 1a +30.6 mg, 86%).
was ®ltered off, and the solvent from the ®ltrate was evapo-
rated under reduced pressure to give a residue, which was
puri®ed by HPLC [YMC-Pack ODS-A, MeOH±H O
2
+80:20, v/v)] to give 1d +4.0 mg, 45%).
1a: Colorless oil. High-resolution EI-MS: calcd for
C H O +M ): 278.1882. Found: 278.1882. IR +®lm):
2
23
1
1d: Colorless oil, [a]
D
ˆ148.18 +cˆ0.1, CHCl
resolution positive-ion FAB-MS: Calcd for C17H O
30 3
3
). High-
Na
M1Na) : 305.2093. Found: 305.2121. IR +®lm): 3450,
1
7
26
3
2
932, 1752, 1680, 1061, 1097, 1061 cm . H NMR +500
1
1
1
+
MHz, CDCl ) d: 0.47 +1H, dt, Jˆ4.6, 7.0 Hz, 1-H), 0.65
21 1
3
2
955, 2872, 1743, 1464, 1219, 1063 cm . H NMR +500
+
s, 14 and 15-H ), 1.41 +2H, m, 2-H ), 1.68 +2H, t-like, 3-H ),
1H, br dd, Jˆca. 5, 7 Hz, 5-H), 1.12, 1.29 +3H each, both
3 2 2
MHz, CDCl ) d: 0.32 +1H, ddd, Jˆ5.4, 6.1, 7.0 Hz, 5-H),
3
0
.40 +1H, dt, Jˆ5.4, 7.0 Hz, 1-H), 0.82, 0.88 +3H each, both
1
.79, 2.09 +3H each, both s, 12 and 13-H ), 2.51, 2.56 +2H,
3
d, Jˆ7.0 Hz, 12 and 13-H ), 1.01 +1H, m, 7-H), 1.06, 1.31
3
ABq, Jˆ15.8 Hz, 9-H ), 2.82 +2H, br s, 6-H ), 3.92 +4H, m,
2
2
1
ZOZCH CH ZOZ). C NMR +125 MHz, CDCl ) d :
3
+3H each, both s, 14 and 15-H
3
), 1.39 +1H, dd, Jˆ10.0,
), 1.68, 1.71 +1H each,
2
2
2
3
C
1
given in Table 2. EI-MS: m/z 278 +M , 5), 87 +100).
11.3 Hz, 9a-H), 1.40 +2H, m, 2-H
both m, 3-H ), 1.75 +2H, m, 6-H ), 2.06 +1H, dd, Jˆ4.8,
2
2
1
+
1.3 Hz, 9b-H), 2.11 +1H, dq, Jˆ3.6, 7.0 Hz, 11-H), 3.40
3
.3. NaBH ±CeCl reduction of 1a
4
3
1H, ddd, Jˆ4.8, 5.5, 10.0 Hz, 8-H), 3.94 +4H, m,
1
ZOZCH CH ZOZ). C NMR +125 MHz, CDCl ) d :
3
2
2
3
C
A solution of 1a +24.2 mg, 87 mmol) in MeOH +1.0 ml) was
treated with NaBH +13.2 mg, 349 mmol) in the presence of
1
given in Table 2. Positive-ion FAB-MS: m/z 305 +M1Na) .
4
CeCl ´H O +10.0 mg) and the mixture was stirred at 08C for
3
2
2
0 min. The reaction mixture was quenched in acetone, then
3.5. Preparation of the &R)-MTPA ester &1e) and the &S)-
MTPA ester &1f) from 1d
poured into ice-water and the whole was extracted with
AcOEt. The AcOEt extract was washed with brine and
then dried over MgSO powder. Removal of the solvent
A solution of 1d +2.0 mg, 7 mmol) in CH Cl +1.0 ml) was
2
4
2
under reduced pressure yielded a residue, which was
puri®ed by HPLC [YMC-Pack ODS-A, MeOH±H Oˆ
treated with +R)-MTPA +16.6 mg, 70 mmol) in the presence
of EDC´HCl +14.6 mg, 70 mmol) and 4-DMAP +5.2 mg,
42 mmol) and the mixture was stirred at room temperature
for 18 h. The reaction mixture was poured into ice-water
and the whole was extracted with AcOEt. The AcOEt
extract was successively washed with 5% aqueous HCl,
2
8
0:20 v/v] to give 1b +16.5 mg, 68%) and 1c +5.6 mg, 23%).
2
4
1
resolution EI-MS: Calcd for C H O +M ): 280.2065.
b: Colorless oil, [a]D ˆ159.18 +cˆ0.2, CHCl ). High-
3
1
1
7
28
3
Found: 280.2070. IR +®lm): 3475, 2932, 1762, 1651,
1
saturated aqueous NaHCO , and brine, then dried over
3
MgSO powder. Removal of the solvent under reduced
4
2
100, 1069 cm . H NMR +500 MHz, CDCl ) d: 0.32
1
1
3
+
1H, br dd, Jˆca. 5, 6 Hz, 5-H), 0.84 +1H, dt, Jˆ4.9,
pressure furnished a residue, which was puri®ed by silica

8452
H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
gel column chromatography +300 mg, n-hexane±AcOEtˆ
:1) to give 1e +2.0 mg, 57%).
6.1 Hz, 15-H ), 1.56, 1.64 +1H each, both m, 3-H ), 1.80,
3 2
2.09 +3H each, both s, 12 and 13-H ), 2.47, 2.52 +2H, ABq,
3
5
Jˆ15.2 Hz, 9-H ), 2.81 +2H, br s, 6-H ), 3.56 +3H, s,
2
2
By the same procedure as used for 1e, the +S)-MTPA ester
ZOCH ), 5.12 +1H, m, 4-H), [7.39 +2H, m), 7.53 +3H, m),
3
+
+
1f, 1.1 mg, 32%) was prepared from 1d +2.0 mg) using
S)-MTPA +16.6 mg, 70 mmol), EDC´HCl +14.6 mg, 70
ph-H].
1
2b: Colorless oil. H NMR +500 MHz, CDCl ) d: 0.42 +1H,
mmol) and 4-DMAP +5.2 mg, 42 mmol).
3
dt, Jˆ4.6, 7.3 Hz, 1-H), 0.59 +1H, br dd, Jˆca. 5, 8 Hz,
1
e: Colorless oil. H NMR +500 MHz, CDCl ) d: 0.38 +1H,
1
5-H), 1.06 +3H, s, 14-H ), 1.25 +3H, d, Jˆ6.1 Hz, 15-H ),
3
3
3
ddd, Jˆ4.8, 5.2, 6.1 Hz, 5-H), 0.43 +1H, dt, Jˆ6.1, 6.7 Hz,
1.28, 1.43 +1H each, both m, 2-H ), 1.61, 1.73 +2H, m, 3-H ),
2 2
1.80, 2.09 +3H each, both s, 12 and 13-H ), 2.50, 2.55 +2H,
3
1
1
7
9
-H), 0.64, 0.73 +3H each, both d, Jˆ7.0 Hz, 12 and 13-H ),
3
.09, 1.31 +3H each, both s, 14 and 15-H ), 1.26 +1H, m,
3
ABq, Jˆ15.5 Hz, 9-H ), 2.82 +2H, br s, 6-H ), 3.56 +3H, s,
2
2
-H), 1.38 +2H, m, 2-H ), 1.55, 2.18 +1H each, both m,
2
ZOCH ), 5.12 +1H, m, 4-H), [7.40 +2H, m), 7.47 +3H, m),
3
-H ), 1.64 +2H, m, 6-H ), 1.74 +2H, m, 3-H ), 1.75
2
ph-H].
2
2
+
1H, m, 11-H), 3.57 +3H, s, ZOCH ), 3.92 +4H, m,
3
ZOZCH CH ZOZ), 4.88 +1H, ddd, Jˆ4.3, 5.8, 10.1 Hz,
2
2
3.8. Acid treatment with 1b and 1c
8
-H), 7.32±7.55 +5H, m, ph-H).
1
f: Colorless oil. H NMR +500 MHz, CDCl ) d: 0.38 +1H,
A solution of 1b +13.6 mg, 48 mmol) in 1,4-dioxane±H
+2: 1 v/v, 1.5 ml) was treated with p-TsOH´H O +10 mg) and
the whole mixture was stirred at room temperature for 3 h.
The reaction mixture was poured into ice-water and the
whole was extracted with AcOEt. The AcOEt extract was
O
2
1
m, 5-H), 0.40 +1H, m, 1-H), 0.76, 0.84 +3H each, both d,
3
2
Jˆ7.0 Hz, 12 and 13-H ), 1.07, 1.31 +3H each, both s, 14
3
and 15-H ), 1.33 +1H, m, 7-H), 1.37 +2H, m, 2-H ), [1.42
3
2
+
1H, dd, Jˆ9.5, 13.4 Hz), 2.12 +1H, dd, Jˆ6.1, 13.4 Hz),
successively washed with saturated aqueous NaHCO and
3
brine and dried over MgSO powder. Removal of the
4
9
-H ], 1.65 +2H, m, 6-H ), 1.74 +2H, m, 3-H ), 1.83
2
2
2
+
1H, m, 11-H), 3.56 +3H, s, ZOCH ), 3.92 +4H, m,
3
solvent under reduced pressure furnished a residue, which
was puri®ed by silica gel column chromatography +500 mg,
n-hexane±AcOEtˆ5:1) to give 1g +11.5 mg, quant.).
Through a similar procedure, 1g +4.3 mg, quant.) was also
prepared from 1c +5.0 mg, 18 mmol).
ZOZCH CH ZOZ), 4.88 +1H, ddd, Jˆ4.4, 6.1, 9.5 Hz,
2
2
8
-H), 7.33±7.54 +5H, m, ph-H).
3.6. CrO ±pyridine oxidation of 4S-dihydrocurcume-
3
none &2)
2
5
1
g: Colorless oil, [a]D ˆ24.68 +cˆ0.1, CHCl ). High-
3
A solution of 2 +2.4 mg, 10 mmol) in pyridine +0.5 mg) was
treated with CrO +3.0 mg)±pyridine +0.5 ml) mixture, and
the whole mixture was stirred at room temperature for 1 h.
The reaction mixture was poured into saturated aqueous
NaCl and the whole was extracted with AcOEt. The
AcOEt extract was washed with brine, then dried over
1
resolution EI-MS: Calcd for C H O +M ±H O):
1
5
22
2
3
2
1
18.1654. Found: 218.1671. IR +®lm): 3650, 2939, 1717,
700, 1031 cm . H NMR +500 MHz, CDCl ) d: 0.53
21
1
3
+
1H, dt, Jˆ4.9, 7.0 Hz, 1-H), 0.56 +1H, ddd, Jˆ4.9, 4.9,
6.7 Hz, 5-H), 1.09 +3H, s, 14-H ), 1.27, 1.27 +3H each,
both s, 12 and 13-H ), 1.62 +2H, m, 2-H ), 2.12, 2.29 +1H
3
3
2
MgSO . Removal of the solvent under reduced pressure
4
each, both m, 9-H ), 2.14 +3H, s, 15-H ), 2.32, 2.38 +1H,
2
3
gave a crude product which was puri®ed by silica gel
column chromatography +200 mg, n-hexane±AcOEtˆ5:1)
to furnish 1 +2.3 mg, quant.).
each, both m, 6-H ), 2.46 +2H, t, Jˆ7.6 Hz, 3-H ), 5.51 +1H,
2
2
13
dd, Jˆ2.5, 5.2 Hz, 8-H). C NMR +125 MHz, CDCl ) d :
3
C
1
given in Table 2. EI-MS: m/z 218 +M ±H O, 10), 160 +100).
2
3
&
.7. Preparation of the &R)-MTPA ester &2a) and the
S)-MTPA ester &2a) from 4S-dihydrocurcumenone &2)
3.9. Conversion from 1g to 3 and 4
A solution of 2 +2.6 mg, 11 mmol) in CH Cl +1.0 ml) was
2
treated with +R)-MTPA +16.6 mg, 70 mmol) in the presence
of EDC´HCl +14.6 mg, 70 mmol) and 4-DMAP +5.2 mg,
A solution of 1g +15.0 mg, 64 mmol) in THF +1.0 ml) was
treated with 1.0 M BH ´THF +0.5 ml) and the whole mixture
2
3
was stirred at room temperature for 2 h. The reaction
mixture was quenched in ice-water, then the solution was
treated with 3.0 M NaOH +1.0 ml) and 30% aqueous H O
4
2 mmol) and the mixture was stirred at room temperature
for 5 h. The reaction mixture was poured into ice-water and
the whole was extracted with AcOEt. Work-up of the
AcOEt extract as described above gave a product, which
was puri®ed by silica gel column chromatography +300
mg, n-hexane±AcOEtˆ10:1) to give 2a +3.7 mg, 74%).
2
2
+1.0 ml) at room temperature. After 2 h, the whole mixture
was extracted with AcOEt. The AcOEt extract was washed
with saturated aqueous NaHCO and brine and dried over
3
MgSO powder, then the mixture was ®ltered. Removal
4
of the solvent from the ®ltrate under reduced pressure
furnished the triol mixture +1h). A solution of the triol
mixture +1h) in CH Cl +1.0 ml) was successively treated
By the same procedure as used for 2a, the compound 2b
+4.5 mg, 84%) was prepared from 2 +2.8 mg, 11 mmol)
using +S)-MTPA +16.6 mg, 70 mmol), EDC´HCl +14.6 mg,
2
2
with PCC +20.0 mg) and the whole mixture was stirred at
room temperature for 1 h. The reaction mixture was poured
into ice-water and the whole was extracted with AcOEt. The
AcOEt extract was washed with saturated aqueous NaHCO3
7
0 mmol) and 4-DMAP +5.2 mg, 42 mmol).
1
a: Colorless oil. H NMR +500 MHz, CDCl ) d: 0.35 +1H,
2
3
dt, Jˆ4.9, 6.8 Hz, 1-H), 0.51 +1H, m, 5-H), 0.97 +3H, s, 14-
and brine, then dried over MgSO . Removal of the solvent
4
under reduced pressure gave a crude product, which was
H ), 1.15, 1.29 +1H each, both m, 2-H ), 1.32 +3H, d, Jˆ
3
2

H. Matsuda et al. / Tetrahedron 57 ꢀ2001) 8443±8453
8453
puri®ed by HPLC [YMC-Pack ODS-A, MeOH±H O +50:
2
3.10.4. Statistical analysis. Values are expressed as the
means^SEM. For statistical analysis, one-way analysis of
variance following Dunnett's test was used. IC50 values
were determined graphically.
5
0, v/v)] to furnish 3 +6.6 mg, 41%) and 4 +2.2 mg, 14%).
3
.10. Bioassay
3
2
.10.1. Tissue preparation. Male Wistar rats weighing
50±350 g were sacri®ced by severing both carotid arteries
References
under anesthesia, and the thoracic aorta was isolated and cut
into helical strips +2±3 mm£15±20 mm). Physiological salt
solution contained NaCl +118.0 mM), KCl +4.7 mM),
1
. +a) Yamahara, J.;Hatakeyama, S.;Taniguchi, K.;Kawamura,
M.;Yoshikawa, M. Yakugaku Zasshi 1992, 112, 645±655.
+
b) Yoshikawa, M.;Hatakeyama, S.;Taniguchi, K.;Matsuda,
H.;Yamahara, J. Chem. Pharm. Bull. 1992, 40, 2239±2241.
c) Yoshikawa, M.;Chatani, N.;Hatakeyama, S.;Nishino, Y.;
Yamahara, J.;Murakami, N. Yakugaku Zasshi 1993, 113,
12±717. +d) Yoshikawa, M.;Yamaguchi, S.;Kunimi, K.;
KH PO4 +1.2 mM), MgSO4 +1.2 mM), CaCl2 +2.5 mM),
2
NaHCO3 +25.0 mM), and d-glucose +10.0 mM). The
solution was aerated with a 95% O ±5% CO gas mixture
+
2
2
and kept at 378C. To investigate the mechanical response,
each preparation was suspended in an organ bath +6 ml)
and subjected to an initial load of about 1 g. One hour
equilibration period was allowed before initiation of the
experiments. Contractions were measured isometrically
via a force-displacement transducer +Nihon Denki Sanei,
Tokyo, Japan) and recorded on a polygraph.
7
Matsuda, H.;Okuno, Y.;Yamahara, J.;Murakami, N. Chem.
Pharm. Bull. 1994, 42, 1226±1230. +e) Yamahara, J.;
Matsuda, H.;Yamaguchi, S.;Shimoda, H.;Murakami, N.;
Yoshikawa, M. Natural Medicines 1995, 49, 76±83.
+f) Matsuda, H.;Morikawa, M.;Sakamoto, Y.;Toguchida,
I.;Yoshikawa, M. Heterocycles, in press.
1
.10.2. High K -induced contraction. After equilibration,
M KCl +0.1 ml) was added to the bath +®nal concentration
2
. This study was partly reported in our preliminary communica-
tion: Yoshikawa, M.;Murakami, T.;Morikawa, T.;Matsuda,
H. Chem. Pharm. Bull. 1998, 46, 1186±1188.
3
3
1
of K : 54 mM). The tissues were washed three times and re-
equilibrated after the contraction had reached the maximum
level. Sustained contraction was induced again by the
addition of KCl, and then test compound was cumulatively
applied at 3±100 mM. The contractile response prior to the
application of the test sample was taken to be 100%.
Nifedipine was used as a reference compound.
3
. +a) Matsuda, H.;Ninomiya, K.;Morikawa, T.;Yoshikawa, 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.
4
5
6
7
8
9
. Matsuda, H.;Morikawa, T.;Toguchida, I.;Ninomiya, K.;
Yoshikawa, M. Heterocycles 2001, 55, 841±846.
. Shiobara, Y.;Asakawa, Y.;Kodama, M.;Yasuda, K.;
Takemoto, T. Phytochemistry 1985, 24, 2629±2633.
. Sakui, N.;Kuroyanagi, M.;Ishitobi, Y.;Sato, M.;Ueno, A.
Phytochemistry 1992, 31, 143±147.
3
.10.3. NE-induced contraction. After equilibration, NE
was added to the bath +®nal concentration of NE: 1 mM)
in the presence of nifedipine +1 mM, a voltage-dependent
2
1
. Sakamoto, S.;Tsuchiya, N.;Kuroyanagi, M.;Ueno, A.
Phytochemistry 1994, 35, 1215±1219.
Ca -channel blocker). The tissues were washed three times
and re-equilibrated after the contraction had reached the
maximum level. This procedure was repeated, and a second
contraction was obtained. Tissues were incubated with each
test compound for 10 min and then NE was cumulatively
applied +1 nM ±1 mM). In order to minimize variability
between tissues, the contraction ratio of the third response
to the second response at 1 mM NE was used. The mean
contraction ratio in the control was taken to be 100%. Prazo-
sin hydrochloride was used as a reference compound. Table
. Piet, D. P.;Schrijvers, R.;Franssen, M. C. R.;Groot, A.
Tetrahedron 1995, 51, 6303±6314.
. Shibuya, H.;Hamamoto, Y.;Cai, Y.;Kitagawa, I.
Pharm. Bull. 1987, 35, 924±927.
Chem.
1
1
1
0. Ohtani, I.;Kusumi, T.;Kashman, Y.;Kakisawa, H.
Chem. Soc. 1991, 113, 4092±4096.
J. Am.
1. Hwu, J. R.;Leu, L.-C.;Robl, J. A.;Anderson, D. A.;Wetzel,
J. M. J. Org. Chem. 1987, 52, 188±191.
2. Hagiwara, S.;Mitsui, M.;Karaki, H. Eur. J. Pharmacol. 1993,
3
1
shows the inhibition +%) of each constituent +100 mM) at
mM NE.
2
34, 1±7.