Contribution of cinnamic acid analogues in rosmarinic acid to inhibition of snake venom induced hemorrhage

Article abstract of DOI:10.1016/j.bmc.2011.02.013

In our previous paper, we reported that rosmarinic acid (1) of Argusia argentea could neutralize snake venom induced hemorrhagic action. Rosmarinic acid (1) consists of two phenylpropanoids: caffeic acid (2) and 3-(3,4-dihydroxyphenyl)lactic acid (3). In this study, we investigated the structural requirements necessary for inhibition of snake venom activity through the use of compounds, which are structurally related to rosmarinic acid (1). By examining anti-hemorrhagic activity of cinnamic acid analogs against Protobothrops flavoviridis (Habu) venom, it was revealed that the presence of the E-enoic acid moiety (-CHCH-COOH) was critical. Furthermore, among the compound tested, it was concluded that rosmarinic acid (1) (IC₅₀ 0.15 μM) was the most potent inhibitor against the venom.

Full text of DOI:10.1016/j.bmc.2011.02.013

Bioorganic & Medicinal Chemistry 19 (2011) 2392–2396
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Contribution of cinnamic acid analogues in rosmarinic acid to inhibition
of snake venom induced hemorrhage
⇑
Hnin Thanda Aung, Tadashi Furukawa, Toshiaki Nikai, Masatake Niwa, Yoshiaki Takaya
Faculty of Pharmacy, Meijo University, 150 Yagotoyama, Tempaku, Nagoya 468-8503, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
In our previous paper, we reported that rosmarinic acid (1) of Argusia argentea could neutralize snake
venom induced hemorrhagic action. Rosmarinic acid (1) consists of two phenylpropanoids: caffeic acid
(2) and 3-(3,4-dihydroxyphenyl)lactic acid (3). In this study, we investigated the structural requirements
necessary for inhibition of snake venom activity through the use of compounds, which are structurally
related to rosmarinic acid (1). By examining anti-hemorrhagic activity of cinnamic acid analogs against
Protobothrops flavoviridis (Habu) venom, it was revealed that the presence of the E-enoic acid moiety
(–CH@CH–COOH) was critical. Furthermore, among the compound tested, it was concluded that rosmari-
Received 7 December 2010
Revised 8 February 2011
Accepted 9 February 2011
Available online 13 February 2011
Keywords:
Rosmarinic acid
Cinnamic acid
nic acid (1) (IC50 0.15
lM) was the most potent inhibitor against the venom.
Ó 2011 Elsevier Ltd. All rights reserved.
Snake venom
Anti-hemorrhage
Protobothrops flavoviridis
1. Introduction
In our previous study, we reported on the anti-hemorrhagic
activity of rosmarinic acid (1) from the methanolic extract of Argu-
sia argentea against crude venoms as well as purified toxins.5 This
compound (1) consists of two similar phenylpropanoids: caffeic
acid (2) and 3-(3,4-dihydroxyphenyl)lactic acid (3). Both of these
compounds possess a 3,4-dihydroxyphenyl group in the molecule
which we speculated to be a key moiety in the molecule for the
exhibition of antivenom activity. In this report, we demonstrate
the structural requirements of rosmarinic acid (1) for the inhibition
of hemorrhage induced by the hemorrhagic snake venom of Proto-
bothrops flavoviridis (Habu) through the use of cinnamic acid
analogues.
Snakebites are a serious medical problem in many regions of
the world, especially in tropical and subtropical countries. Snake
venoms are complex mixtures of biologically active proteins, pep-
tides, metal ions and organic compounds1,2 possessing a variety of
biological activities. Envenomations due to snakebites are com-
monly treated by parenteral administration of horse or sheep-
derived polyclonal antivenom aimed at the neutralization of
toxins. However, antivenom does not provide sufficient protection
against venom induced hemorrhage, necrosis, nephrotoxicity and,
furthermore, can often cause hypersensitiviy reactions.3,4 It has
been known for many years that animal sera and many plants
can be used to neutralize snake venom, and, consequently, several
reports and reviews of their constituents with respect to antiven-
om activity have been published. However, to the best of our
knowledge, no report on the structural requirements to exhibit
antivenom activity has been published.
2. Results and discussion
2.1. Separation and anti-hemorrhagic activity of E- and
Z-isomers of rosmarinic acid (1) and contribution
of the double bond to the activity
Hemorrhage is one of the most conspicuous consequences of
snake envenoming, and it is sometimes lethal. Unfortunately,
the only clinical treatment is antiserum against snake venoms.
Snakebites often occur outdoors, far from medical institutions.
Because of this, drugs to treat snakebites must be transported to
remote locations ahead of time, requiring them to be stable against
light, oxygen, and other forms of decomposition without the aid of
refrigeration or a special container.
Prior to the investigation of rosmarinic acid analogue activities,
we tried to separate the E- and Z-isomers of rosmarinic acid (1).
Due to the precise nature of NMR analysis, the spectrum showed
signals correspond to (E)- and (Z)-rosmarinic acid (1a and 1b,
respectively). Based on the integration values of signals derived
from olefinic protons on the 1H NMR spectra, the ratio of each iso-
mer was revealed to be approximately 9:1. These isomers were
separated by using HPLC equipped with COSMOSIL Cholester col-
umn, and fractionated the isomers. Unfortunately, the Z-isomer
(1b) was quite unstable and thus difficult to isolate as a single iso-
⇑
Corresponding author. Tel./fax: +81 52 839 2740.
E-mail address: ytakaya@meijo-u.ac.jp (Y. Takaya).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.02.013

H. T. Aung et al. / Bioorg. Med. Chem. 19 (2011) 2392–2396
2393
mer. The Z-isomer (1b) was immediately converted to the E-isomer
(1a), even under dark or cold conditions. Conversely, the E-isomer
proved to be stable. Accordingly, only the E-isomer (1a) gave an ex-
act antivenom activity that was significantly stronger (IC50
we will go on to mention in the next section, the methoxyl group
on the benzene ring is not expected to affect the activity. As ex-
pected, compound 3 exhibited no activity. Moreover, in our previ-
ous report, rosmarinic acid methyl ester (5), which was also a
mixture of Z- and E-isomers (ca. 1:9), did not show anti-hemor-
rhagic activity, so the presence of a free carboxylic group might
be critical for anti-hemorrhagic activity. To test this hypothesis,
caffeic acid methyl ester (6) was subjected to anti-hemorrhage
tests, for which it showed no activity. The caffeic acid moiety in
rosmarinic acid (1) is esterified by the hydroxyl group of 3-(3,4-
dihydroxyphenyl)lactic acid (3), but the activity might be per-
formed by the double bond of caffeic acid moiety and the carbox-
ylic group.
0.15
(1) (IC50 0.60
l
M) than that of the mixture of (E)- and (Z)-rosmarinic acid
M). We further investigated the anti-hemorrhage
l
activity of saturated rosmarinic acid (4), but no activity was ob-
served. From the results, the existence of the E configuration of a
double bond seemed to be required to exhibit anti-hemorrhagic
activity.
2.2. Anti-hemorrhagic activity of caffeic acid (2), and 3-(3,4-
dihydroxyphenyl)lactic acid (3) as components of rosmarinic
acid (1) and the contribution of the carboxylic group
2.3. Contribution of phenolic hydroxyl groups to anti-
hemorrhagic activity
As mentioned above, rosmarinic acid (1) consists of caffeic acid
(2) and 3-(3,4-dihydroxyphenyl)lactic acid (3) moieties. We inter-
ested in the anti-hemorrhagic activities of these compounds which
might contribute to the activity. The activity of caffeic acid (2)
Caffeic acid (2) is suspected to play an important role to exhibit
anti-hemorrhagic activity. Because of this, the participation of phe-
nolic hydroxyl groups in relation to this activity should be dis-
cussed. Anti-hemorrhagic activities of various kinds of cinnamic
acid analogues were investigated as demonstrated in Table 2. As
a result, it was revealed that this activity does not depend on the
numbers or positions of the phenolic hydroxyl groups when the
configuration of a double bond is E. This is the result of cinnamic
acid (7), which possesses no phenolic hydroxyl group, does not
show a significant change of activity from that of caffeic acid.
Moreover, compounds with a methoxyl group on phenyl group
(16 and 17) exhibited just slightly lower activities than that of
caffeic acid (2).
(0.19 lM) was as potent as that of (E)-rosmarinic acid (1a) (Ta-
ble 1). On the other hand, we tried to prepare compound 3 to
examine its activity, but were unsuccessful. As an alternative, we
tested 3-(4-hydroxy-3-methoxyphenyl)lactic acid (3b) since, as
Table 1
Anti-hemorrhagic activity of rosmarinic acid and its analogues
2.4. Contribution of the carboxyl group and side chain to anti-
hemorrhagic activity
Compounds 13–15 with three phenolic hydroxyl groups, whose
configurations of double bonds are all Z, did not show the activity
(Table 2). When the carboxyl group is reduced to aldehyde and
alcohol, the resultant compounds (18 and 19) were also inactive.
Moreover, methyl cinnamate 20 showed no activity. On the other
hand, compound 21, which possessed a conjugated E diene struc-
ture, showed similar activity (0.22 lM) to that of caffeic acid while
saturated caffeic acid (22, 3-(3,4-dihydroxyphenyl)propanoic acid)
did not. As a result, this evidence suggests that the configuration of
the double bond and the existence of the carboxyl group were
more effective to the activity than the positions of oxygen substit-
uents on the phenyl group.
2.5. Anti-hemorrhagic activity of chlorogenic acid
As shown in Table 3, chlorogenic acid (23), an ester of caffeic
acid and quinic acid, did not change the IC50 value (0.20 lM) signif-
icantly in comparison to that of E-caffeic acid (2). The carboxyl
group of chlorogenic acid (23) is esterified by quinic acid, but the
compound exhibited the activity by the combination of the double
bond of caffeic acid moiety and the carboxyl group of quinic acid,
as in the case with rosmarinic acid (1a).
2.6. Contribution of the phenyl group to anti-hemorrhagic
activity
Compounds
IC50 (lM)
Rosmarinic acid (1)
(E)-Rosmarinic acid (1a)
Caffeic acid (2)
3-(4-Hydroxy-3-methoxyphenyl)lactic acid (3b)
Saturated rosmarinic acid (4)
Rosmarinic acid methyl ester (5)
Methyl caffeate (6)
0.60
0.15
0.19
NAa
NA
As mentioned above, the hydroxyl group(s) on the phenyl group
is not critical to exhibit anti-hemorrhagic activity, but the E-enoic
acid moiety is. Then we tested the anti-hemorrhagic activity of
the simplest compounds, crotonic acid (24) and sorbic acid
(2,4-hexadienoic acid) (25). As a result, as shown in Table 4, these
two compounds also exhibited comparable activity as that of
NA
NA
a
NA denotes ‘not active’.

2394
H. T. Aung et al. / Bioorg. Med. Chem. 19 (2011) 2392–2396
Table 2
Anti-hemorrhagic activity of cinnamic acid and its analogues
Compounds
2EZ
R1
R2
R3
R4
R5
IC50 (lM)
Cinnamic acid (7)
E
E
E
E
E
E
Z
Z
Z
E
E
E
E
E
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
CHO
H
OH
H
H
OH
H
OH
OH
H
H
H
H
H
OH
H
H
OH
OH
H
OH
OCH3
OCH3
H
H
H
H
H
H
H
H
H
OH
H
OH
OH
H
0.18
0.19
0.18
0.19
0.21
0.18
NA
NA
NA
0.20
0.24
NA
2-Hydroxycinnamic acid (8)
3-Hydroxycinnamic acid (9)
4-Hydroxycinnamic acid (10)
2,4-Dihydroxycinnamic acid (11)
3,4,5-Trihydroxycinnamic acid (12)
2,3,4-Trihydroxycinnamic acid (13)
2,4,5-Trihydroxycinnamic acid (14)
3,4,5-Trihydroxycinnamic acid (15)
Ferulic acid (16)
OH
OH
OH
OH
OH
OH
OH
OCH3
H
3,4-Dimethoxycinnamic acid (17)
Cinnamaldehyde (18)
Cinnam alcohol (19)
H
H
H
H
H
H
H
CH2OH
COOCH3
H
H
H
H
NA
NA
Methyl cinnamate (20)
Table 3
Anti-hemorrhagic activity of cinnamic acid analogues
functional groups was not discussed. Lans et al. reported that caf-
feic acid (2) might strongly bind to proteins through hydrophobic
interactions and hydrogen bonds with subsequent conformational
changes in the proteins.7,8 However, according to our results, phe-
nolic hydroxyl groups on the phenyl group would not be critical.
We speculate that the anti-hemorrhagic compound should bind
the venom allosterically through the carboxylic group and the dou-
ble bond to cause conformational change so as to exhibit activity.
However, further studies are needed to investigate the mode of
action of these compounds. We have already reported on the
mechanism of action of rosmarinic acid (1) with respect to its
anti-hemorrhagic activity.9 The investigation of the methods for
its administration is currently ongoing with the goal of developing
a potent alternative antidote compound for snake envenomation.
4. Experimental
4.1. General
Compounds
IC50 (lM)
(2E,4E)-5-Phenylpenta-2,4-dienoic acid (21)
3-(3,4-Dihydroxyphenyl)propanoic acid (22)
Chlorogenic acid (23)
0.22
NA
0.20
IR spectra were recorded on FT-IR-410 spectrophotometer. 1H
and 13C NMR spectra were recorded on a JEOL ECA-500 (1H:
500 MHz and 13C: 125 MHz). Chemical shifts for 1H and 13C NMR
are given in parts per million (d) relative to solvent signal (metha-
nol-d4: dH 3.30 and dC 49.0) as internal standard. EI- and FAB-MS
were obtained with a JEOL JMS MS-700 and HX-110 and m-nitro-
benzyl alcohol was used for FAB-MS as a matrix. Analytical TLC
was performed on Silica gel 60 F254 (Merck). Column chromatogra-
phy was carried out on silica gel BW-820MH (Fuji Silysia Chemi-
cals, Co. Ltd, Seto, Japan). Develosil ODS UG-5 (/20 ꢀ 250 mm,
Nomura Chemical, Seto, Japan), and COSMOSIL Cholester (/
20 ꢀ 250 mm, Nacalai Tesque, Kyoto, Japan) columns were used
for preparative HPLC, and Develosil ODS UG-5 (/4.6 ꢀ 250 mm,
Nomura Chemical, Seto, Japan), and Cosmosil Cholester (/4.6 ꢀ
250 mm, Nacalai Tesque, Kyoto, Japan) columns were used for
the analytical HPLC. All experiments involving the use of animals
were carried out in compliance with the guidelines for animal
experiments of Faculty of Pharmacy, Meijo University.
Table 4
Anti-hemorrhagic activity of aliphatic enoic acids
Compounds
IC50 (lM)
Crotonic acid (24)
Sorbic acid (25)
0.22
0.21
cinnamic acid (7). From this result, it revealed that phenolic group
is not also so effective to the activity.
3. Conclusion
In this study, the structural requirement for cinnamic acid ana-
logues to exhibit anti-hemorrhagic activity was investigated. As a
result, compounds possessing a carboxylic group and an E-double
bond showed high potency with respect to their activities. More-
over, even the compounds without the phenyl group exhibit the
activity. It does not seem that the carboxylic group and the double
bond need to be conjugated in order for this to occur. Rosmarinic
acid (1) along with caffeic acid (2) and chlorogenic acid (23) are
known to have antivenom activity.6 However, the contribution of
4.2. Venoms and chemicals
P. flavoviridis venom (Okinawa) was purchased from Japan
Snake Institute, Gunma. Chlorogenic acid was isolated from Chry-
santhemum coronarium L. (garland chrysanthemum, Compositae).

H. T. Aung et al. / Bioorg. Med. Chem. 19 (2011) 2392–2396
2395
Cinnamic acid, 2-hydroxy cinnamic acid, 4-hydroxy cinnamic acid,
caffeic acid, 3-(4-hydroxy-3-methoxyphenyl)lactic acid, 4-hydro-
xy-3-methoxycinnamic acid, 3,4-dimethoxycinnamic acid, 3,4-
dihydroxyphenyllactic acid, 3,4,5-trimethoxycinnamic acid, 2,4,5-
trimethoxycinnamic acid, 2,3,4-trimethoxycinnamic acid, crotonic
acid and sorbic acid were obtained from Tokyo Chemical Industry
Co., Ltd (Tokyo, Japan). 3-Hydroxy cinnamic acid was purchased
from Wako Pure Chemical Industries, Ltd (Osaka, Japan). 2,4-
Dihydroxycinnamic acid was purchased from Sigma–Aldrich Co.
(St. Louis, MO, U.S.A.). Other chemicals used were of analytical
grade. Rosmarinic acid (1, Z and E-mixture) and rosmarinic methyl
ester (5) were isolated from methanolic extract of A. argentea as
described in Ref. 5.
Saturated caffeic acid (= 3-(3,4-Dihydroxyphenyl)propanoic
acid) (22): IR mmax (film) 1715 cmꢁ1 1H NMR (CD3OD, 500 MHz):
;
d 6.62 (1H, d, J = 7.6 Hz; H-50), 6.59 (1H, d, J = 1.5 Hz; H-20), 6.48
(1H, dd, J = 7.6, 1.5 Hz; H-60), 2.71 (2H, t, J = 7.6 Hz; H-3), 2.47
(2H, t, J = 7.6 Hz; H-2); EIMS m/z 182 [M]+, 123 (base).
4.7. Methylation of caffeic acid (2)
Trimethylsilyldiazomethane (174 lL) was added to caffeic acid
(2, 215 mg, 1.19 mmol) in 10 mL of acetone, and the solution was
stirred at 0 °C for 20 min. The solution was evaporated, and the res-
idue was purified by SiO2 column chromatography to afford
methyl caffeate (6, 145 mg, 67.4%).
Methyl caffeate (6): IR m
max (film) 3363 (br), 1684 cmꢁ1; 1H NMR
4.3. Separation of Z- and E-isomers of rosmarinic acid
(CD3OD, 500 MHz): d 7.52 (1H, d, J = 15.9 Hz; H-3), 7.01 (1H, d,
J = 1.5 Hz; H-20), 6.92 (1H, dd, J = 8.3, 2.2 Hz; H-60), 6.75 (1H, d,
J = 8.3 Hz; H-50), 6.24 (1H, d, J = 15.9 Hz; H-2), 3.73 (3H, s;
COOCH3); EIMS m/z 194 [M]+.
A mixture of Z- and E-isomers of rosmarinic acid were separated
using HPLC equipped with COSMOSIL Cholester (/20 ꢀ 250 mm)
eluted by a mixed solvent of acetonitrile–water–trifluoroacetic
acid (2:8:0.05) at a flow rate of 5 mL/min. The column was main-
tained at a constant temperature of 30 °C. UV monitoring was done
at 220 nm. (Z)- and (E)-Rosmarinic acid (1a and 1b) was eluted at
68 min and 76 min, respectively.
4.8. Preparation of (E)-3,4,5-trihydroxycinnamic acid (12)
Aniline (50 lL) was added to a solution of 3,4,5-trihydroxybenz-
aldehyde (250 mg) and malonic acid (250 mg) in anhydrous pyri-
dine (1.25 mL) and the reaction mixture was stirred at 50 °C for
12 h. After cooling, diethylether (20 mL) was added to the solution.
The organic phase was washed with 2 M hydrochloric acid and
water, and dried over anhydrous magnesium sulfate. The product
was purified by HPLC, which afforded the desired product (12,
160 mg, 64%) as a brown crystal.
4.4. Anti-hemorrhagic activity assay
Anti-hemorrhagic activity was examined under the modified
method reported previously.5,10 Male ddY mice of 20 g average
weight were used for this experiment. Test solutions were pre-
pared as follows: Crude snake venom solution of P. flavoviridis
(E)-3,4,5-Trihydroxycinnamic acid (12 ): 1H NMR (500 MHz,
CD3OD): d 7.42 (1H, d, J = 15.9 Hz; H-2), 6.58 (2H, s; H-20 and H-
60), 6.15 (1H, d, J = 15.9 Hz; H-3); EIMS m/z 196 [M]+, 122 (base).
(0.14 mg/mL in saline, 50
ylsulfoxide (DMSO)–saline, 50
37 °C for 10 min. The test solution (100
l
L) and sample solution in 10% dimeth-
L) were mixed and incubated at
L) was injected subcuta-
l
l
neously into the abdomen of the mice. Mice injected the vehicle
solution only served as a control group. After 24 h, mice were
euthanized by inhalation of chloroform, the skin covering the
abdomen was removed, and hemorrhagic lesions were determined
as follows. We evaluated the lesions by calculating a value of major
axes x minor axes, just as that of an ellipse, since the shapes of the
lesions are always amorphous.
4.9. Preparation of Z-trihydroxycinnamic acid
E-Trimethoxycinnamic acid was dissolved to dry CH2Cl2 under
an argon atmosphere and cooled to ꢁ78 °C. 1 M BBr3 in CH2Cl2
solution (2.8 equiv) was added dropwise to the solution and was
stirred at ꢁ78 °C for 1 h. Then the reaction temperature was grad-
ually raised to 0 °C. When the starting material disappeared on TLC,
methanol was added to the solution, and the reaction mixture was
evaporated. Methanol was added again to the residue and the solu-
tion was evaporated to remove methylborate. The product was
purified by SiO2 chromatography, and the desired product was
obtained.
(Z)-2,3,4-Trihydroxycinnamic acid (13) from E-2,3,4-trimethoxy-
cinnamic acid (8.9% yield): 1H NMR (CD3OD, 500 MHz): d 7.4 (1H,
d, J = 9.8 Hz; H-3), 6.87 (1H, d, J = 8.3 Hz; H-60), 6.47 (1H, d,
J = 8.3; H-50), 6.15 (1H, d, J = 9.8 Hz; H-2); 13C NMR (CD3OD,
125 MHz): d 162.1, 149.8, 145.3, 143.6, 132.1, 118.8, 112.6,
112.4, 110.8; EIMS m/z 178 [M–H2O]+, 150 (base).
4.5. Preparation of saturated rosmarinic acid (4)
Rosmarinic acid (1, 105 mg) was dissolved in acetone (1 mL),
and 5% palladium on activated carbon (20 mg) was added to the
solution. The heterogeneous mixture was stirred under a hydrogen
atmosphere at room temperature for 3 h. The mixture was filtered
through celite, affording a saturated rosmarinic acid (4) as a yellow
crystal (98 mg, 93%).
Saturated rosmarinic acid (4): IR mmax (film) 3364 (br) cmꢁ1
;
1H NMR (CD3OD, 500 MHz): d 6.71 (1H, d, J = 1.5 Hz; H-2), 6.61
(1H, d, J = 7.6 Hz; H-5), 6.55 (1H, dd, J = 7.6, 1.5 Hz; H-6), 4.97
(1H, dd, J = 9.8, 3.0 Hz; H-8), 2.84 (1H, dd, J = 14.4, 9.8 Hz; H-7),
3.04 (1H, dd, J = 14.4, 3.0 Hz; H-7), 6.58 (1H, d, J = 1.5 Hz; H-20),
6.66 (1H, d, J = 8.3 Hz; H-50), 6.36 (1H, dd, J = 8.3, 1.5 Hz; H-60),
2.65 (2H, t, J = 7.6 Hz; H-70), 2.52 (2H, t, J = 7.6 Hz; H-80); FAB-MS
m/z 385 [M+Na]+.
(Z)-2,4,5-Trihydroxycinnamic acid (14) from E-2,4,5-trimethoxy-
cinnamic acid (5.7% yield): 1H NMR (CD3OD, 500 MHz): d 7.75 (1H,
d, J = 8.3 Hz; H-3), 6.91 (1H, s, H-60), 6.73 (1H, d, J = 8.3, H-50),
6.15 (1H, d, J = 8.3 Hz; H-2); 13C NMR (CD3OD, 125 MHz): d 162.9,
150.7, 149.1, 144.7, 143.2, 111.7, 111.5, 111.1, 102.3; EIMS m/z
178 [M–H2O]+, 150 (base).
(Z)-3,4,5-Trihydroxycinnamic acid (15) from E-3,4,5-trimethoxy-
cinnamic acid (0.5% yield): 1H NMR (CD3OD, 500 MHz): d 7.6 (1H,
d, J = 9.1 Hz; H-3), 6.91 (1H, s; H-60), 6.72 (1H, s; H-20), 6.14 (1H,
d, J = 9.1 Hz; H-2); EIMS m/z 178 [M–H2O]+, 150 (base).
4.6. Preparation of saturated caffeic acid (22)
Five percentage palladium on activated carbon (20 mg) was
added to caffeic acid (2, 100 mg) in acetone (l mL). The heteroge-
neous mixture was stirred under a hydrogen atmosphere at room
temperature for 3 h. The mixture was filtered through celite which
afforded a saturated caffeic acid (22, 95 mg, 95%) as a brown
crystal.
4.10. Preparation of methyl cinnamate (20)
To a solution of (E)-cinnamic acid (7, 5.1 g) in 50 mL of acetone,
23.7 g of K2CO3 and 3.25 mL of dimethyl sulfate were added. The

2396
H. T. Aung et al. / Bioorg. Med. Chem. 19 (2011) 2392–2396
mixture was stirred at 80 °C for 6 h. The mixture was then filtered
to remove K2CO3. The filtrate was acidified with 1 M of hydrochlo-
ric acid and was extracted with ethyl acetate. The organic layer
was washed with water and brine, dried with anhydrous magne-
sium sulfate, filtered and evaporated to afford methyl (E)-cinna-
mate (20, 4.6 g, 90%).
10.6 Hz; H-4), 7.03 (1H, d, J = 15.9 Hz; H-5), 6.03 (1H, d,
J = 15.9 Hz; H-2); EIMS m/z 174 [M]+, 129 (base).
Acknowledgments
This work was supported by the bounty for academic research
from the Research Institute of Meijo University. One of the authors
(H.T.A.) is supported by the Japanese Government (MEXT) Scholar-
ship 2007.
Methyl cinnamate (20): IR mmax (film) 3062, 2950, 1716 cmꢁ1
;
1H NMR (CD3OD, 500 MHz): d 7.65 (1H, d, J = 15.9 Hz; H-3),
7.56–7.35 (5H, m; aromatic), 6.48 (1H, d, J = 15.9 Hz; H-2), 3.75
(3H, s; COOCH3); EIMS m/z 162 [M]+, 131 (base).
Supplementary data
4.11. Preparation of (2E,4E)-5-phenylpenta-2,4-dienoic acid (21)
Supplementary data (profiles of HPLC anlyses of the mixture of
Z- and E-rosmarinic acid (1) and purified E-rosmarinic acid (1a).
1H NMR spectra of these compounds are also obtained) associated
with this article can be found, in the online version, at doi:10.1016/
j.bmc.2011.02.013.
A
mixture of methyl chloroacetate (4.8 g, 44 mmol) and
triethylphosphite (7.3 g, 44 mmol) was heated at 135 °C for 10 h.
The mixture was allowed to cool to room temperature to afford
the desired product, methyl 2-(diethylphosphoryl)acetate, in 98%
yield (6.2 g) as a mixture of rotamers.
To a suspension of sodium hydride (7.8 mmol) in dry THF
(10 mL), methyl 2-(diethyl phosphoryl)acetate (420 mg) in 5 mL
of THF was added dropwise at ꢁ60 °C. The reaction mixture was
then stirred at 0 °C for 15 min and cooled to ꢁ60 °C again. Next cin-
namaldehyde (18, 300 mg) in 2 mL of THF was added to the reac-
tion solution and stirred for 90 min. The reaction mixture was
stirred at 0 °C for 16 h, then at room temperature for 2 h. Next
the reaction was acidified with 2 M hydrochloric acid and ex-
tracted with ethyl acetate. The organic fractions were collected,
washed with brine, dried with anhydrous magnesium sulfate and
filtered. The crude product was purified using a SiO2 column to af-
ford 240 mg of (2E,4E)-5-phenylpenta-2,4-dienoic acid (21, 61%
yield).
References and notes
1. Stahel, E.; Wellader, R.; Frayvogel, T. A. Schweiz. Med. Wochenschr. 1985, 115,
890.
2. Bieber, A. L. In Snake Venoms: Metal and nonprotein constituents in snake venoms;
Springer: Berlin, 1979; Vol. 52, pp. 295–306.
3. Sutherland, S. K. Med. J. Aust. 1992, 157, 734.
4. Corrigan, P.; Russel, F. E.; Wainschal, J. In Animal; Rosenberg, P., Ed.; Plant and
Microbial Toxins; Elsevier: Amsterdam, 1978; pp 457–467.
5. Aung, H. T.; Nikai, T.; Niwa, M.; Takaya, Y. J. Nat. Med. 2010, 64, 482.
6. Panfoli, I.; Calzia, D.; Ravera, S.; Morelli, A. Toxins 2010, 2, 417.
7. Mors, W. B.; Nascimento, M. C.; Pereira, B. M. R.; Pereria, N. A. Phytochemistry
2000, 55, 627.
8. Lans, C.; Harper, T.; Georges, K.; Bridgewater, E. BMC Complement. Altern. Med.
2001, 1, 10.
9. Aung, H. T.; Nikai, T.; Komori, Y.; Nonogaki, T.; Niwa, M.; Takaya, Y. Toxins 2010,
2, 2478.
(2E,4E)-5-Phenylpenta-2,4-dienoic acid (21): IR mmax (film) 3022,
1679 cmꢁ1 1H NMR (CD3OD, 500 MHz): d 7.55–7.29 (5H, m; aro-
;
10. Min, H. M.; Aye, M.; Taniguchi, T.; Miura, N.; Monde, K.; Ohzawa, K.; Nikai, T.;
Niwa, M.; Takaya, Y. Tetrahedron Lett. 2007, 48, 6155.
matic), 7.44 (1H, dd, J = 15.9, 10.6 Hz; H-3), 7.11 (1H, dd, J = 15.9,