Antinociceptive activities of α-truxillic acid and β-truxinic acid derivatives

Article abstract of DOI:10.1248/bpb.29.580

Our recent study demonstrated that the dimeric structure of α-truxillic acid derivatives played an important role in the expression of their anti-inflammatory activities. In the present report, to investigate the correlation between the structure and anti

Full text of DOI:10.1248/bpb.29.580

580
Notes
Biol. Pharm. Bull. 29(3) 580—584 (2006)
Vol. 29, No. 3
a
b
Antinociceptive Activities of -Truxillic Acid and -Truxinic Acid
Derivatives
Yu-Ming CHI, ,a Motoyuki NAKAMURA,a Xi-Ying ZHAO,a Toyokichi YOSHIZAWA,a Wen-Mei YAN,b
*
c
c
d
Fumio HASHIMOTO, Junei KINJO, Toshihiro NOHARA, ,c and Shinobu SAKURADA
*
a Seiwa Pharmaceutical Ltd.; 187–11 Usuba, Hanakawa-machi, Kitaibaraki, Ibaraki 319–1535, Japan: b Beijing
University of Traditional Chinese Medicine and Pharmacy; Beijing 100029, China: c Faculty of Pharmaceutical Sciences,
Kumamoto University; 5–1 Oe-Honmachi, Kumamoto 862–0973, Japan: and d Department of Pharmacology, Tohoku
Pharmaceutical University; 4–4–1 Komatsushima, Aoba-ku, Sendai 981–8558, Japan.
Received August 12, 2005; accepted October 28, 2005
Our recent study demonstrated that the dimeric structure of a-truxillic acid derivatives played an impor-
tant role in the expression of their anti-inflammatory activities. In the present report, to investigate the correla-
tion between the structure and anti-inflammatory activity, a-truxillic acid (1) and its derivatives (2—6), b-trux-
inic acid (7) and its derivatives (8—10) were prepared, and their activities were evaluated in the formalin test. All
compounds showed only weak or no activities against the neurogenic pain response, but demonstrated significant
activities against the inflammatory pain response induced by formalin. The highest anti-inflammatory activities
were observed for a-truxillic acid (1) and its derivative 4,4ꢀ-dihydroxy-a-truxillic acid (2). In addition, a-truxillic
acid (1) and its derivative, a-truxillic acid bis(p-nitrophenyl)ester (5), showed higher anti-inflammatory activities
than b-truxinic acid (7) and the corresponding derivative (10). Furthermore, free carboxylic acids (1, 2) showed
higher activities than their dimethyl esters (3, 4) and bis(p-nitrophenyl)ester (5). These results confirmed that the
a-formation of dimeric structure and the free carboxylic acid were also important for the expression of anti-in-
flammatory activities. Otherwise, 4,4ꢀ-dichloro-b-truxinic acid (8) had higher activity than its parent compound
7; furthermore, 1,3-dibenzoyl-2,4-di(4-chlorophenyl)cyclobutane (6) also showed strong anti-inflammatory activ-
ity. These results suggested that substituents in the phenyl groups were also important for the expression of anti-
inflammatory activity. In order to gain information about their activity intensity, the anti-inflammatory activities
of 2 and 4,4ꢀ-dichlorolated derivatives (6, 8) were compared with that of indomethacin (a nonsteroidal anti-in-
flammatory drug) in the formalin test. As a result, compounds 2, 6 and 8 showed stronger anti-inflammatory ac-
tivities than indomethacin. These results suggested that a-truxillic acid and b-truxinic acid derivatives might be
developed into a new type of anti-inflammatory drug.
Key words a-truxillic acid; b-truxinic acid; anti-inflammatory activity; indomethacin; formalin test
In the previous report, we disclosed that a-truxillic acid
(1) and its derivative 4,4ꢀ-dihydroxy-a-truxillic acid (2)
showed significant anti-inflammatory activities in the forma-
lin test. Their monomer components (E)-cinnamic acid and
(E)-p-coumaric acid exhibited only weak or no activity.1)
This result suggested that the dimeric structure played an im-
portant role in the expression of anti-inflammatory activity.
In the course of our investigation of antinociceptive sub-
stances from natural sources, a novel monoterpene alkaloid,
incarvillateine, was characterized from a traditional Chinese
crude drug designed as ‘Tougucao’ (Incarvillea sinensis).2)
Incarvillateine possessed the same dimeric structure as that
of a-truxillic acid, and demonstrated more potent antinoci-
ceptive activity than morphine in the formalin test.3) The
mechanism of antinociception was regarded to be different
from that of morphine, and the activity was mainly evoked
by activation of m, k-opioid receptors and an adenosine re-
ceptor. Structure–activity relationship study of incarvillateine
revealed that the monoterpene alkaloid moiety, incarvilline,
and the dimeric structure played important roles in the ex-
pression of activities against the neurogenic and inflamma-
tory pain responses, respectively.4) In order to investigate the
Fig. 1. Structures of Compounds 1—10
structure–activity relationship between dimeric structure and
anti-inflammatory activity, a-truxillic acid and b-truxinic
acid, as well as their derivatives, were prepared (Fig. 1), and MATERIALS AND METHODS
their activities were evaluated in the formalin test.
Animals Male ddy mice were used. The mice (25ꢁ5 g)
∗ To whom correspondence should be addressed. e-mail: chi-59@unimatec.co.jp
© 2006 Pharmaceutical Society of Japan

March 2006
581
were allowed food and water ad libitum and they were COOMe), 4.20 (2H, m, b, bꢀ-H), 4.71 (2H, m, a, aꢀ-H),
housed in an air-conditioned room maintained at 22ꢁ2 °C 7.19 (4H, d, Jꢅ8.6 Hz, 3, 3ꢀ, 5, 5ꢀ-H), 7.44 (4H, d, Jꢅ8.6 Hz,
with a humidity of 55ꢁ5%. The experimental animals were 2, 2ꢀ, 6, 6ꢀ-H).8)
a
Synthesis of -Truxillic Acid Bis(p-nitrophenyl)ester
allocated randomly into groups.
Chemicals (E )-p-Coumaric acid, 4-chlorocinnamic acid, (5) A mixture of a-truxillic acid (435 mg, 1.47 mmol) and
3-(3-pyridyl)acrylic acid, p-nitrophenol, cinnamyl chloride, thionyl chloride (2.47 g) and N,N-dimethylformamide (one
N,N-dimethylformamide, trimethylsilyldiazomethane were drop) was refluxed for 3 h. After drying, the excess thionyl
purchased from Tokyo Kasei (Tokyo, Japan). Thionyl chlo- chloride was removed d: and a-truxillic acid chloride was
ride and lithium aluminium hydride (LiAlH4) were bought obtained. To a solution of p-nitrophenol (230 mg) in THF (3
from Kanto Kagaku (Tokyo, Japan). (E)-Cinnamic acid and ml) and pyridine (0.5 g), a-truxillic acid chloride (500 mg) in
Tween 80 (polyoxyethylene sorbitan monooleate) were pur- THF (1 ml) was added dropwise at 0—5 °C under stirring.
chased from Nacalai Tesque (Kyoto, Japan). Ringer solution After 2 h of stirring, the reaction mixture was poured into
was obtained from Fuso Pharmaceutical (Osaka, Japan).
water (500 ml) and filtered off. The precipitate was recrystal-
General Methods Melting points were uncorrected. IR lized from methyl ethyl ketone gave a-truxillic acid bis(p-ni-
spectra were recorded on a Hitachi 270-30 spectrometer. 1H- trophenyl)ester (580 mg, 73.4%) as a needle crystal, mp
and 13C-NMR spectra were obtained with a JEOL a-500 230—231 °C. IR (KBr) cmꢂ1: 1763, 1734 (CꢅO), 1522,
1
spectrometer, and chemical shifts were given on a d (ppm) 1350 (NO2). Positive FAB-MS m/z: 539 [Mꢄ1]ꢄ (100). H-
scale with tetramethylsilane as an internal standard. The EI- NMR (CDCl3) d: 4.34 (2H, m, b, bꢀ-H), 4.75 (2H, m, a, aꢀ-
MS and FAB-MS were measured with a JOEL DX-303 HF H), 7.1—7.5 (10H, m, aromatic-H), 6.45, 8.08 (each 4H, dd,
spectrometer. TLC was performed on precoated Kieselgel 60 Jꢅ1.98, 6.92 Hz, p-nitrophenyl-H).9)
F254 plates (Merck). Column chromatography was carried out
Synthesis of 1,3-Dibenzoyl-2,4-di(4-chlorophenyl)cyclo-
on Kieselgel 60 (70—230 mesh and 230—400 mesh). All butane (6) p-Chlorobenzalacetophenone (1.0 g, 4.13 mmol)
materials and reagents purchased were used without further was treated in a manner similar to the conversion of (E)-cin-
purification.
namic acid to a-truxillic acid (1). After 2 d irridation, the
Synthesis of a-Truxillic Acid (1) Compound 1 was pre- product was filtered off and recrystallized twice from CHCl3
pared according to a previous report.1) mp 274—275 °C (lit. to give 1,3-dibenzoyl-2,4-di(4-chlorophenyl)cyclobutane as a
274—278 °C). Negative FAB-MS m/z (rel. int.): 295 needle crystal (200 mg, 10.1%), mp 255—258 °C (lit. 257—
1
[MꢂH]ꢂ. H-NMR (DMSO-d6) d: 3.81 (2H, m, b, bꢀ-H), 259 °C). IR (KBr) cmꢂ1: 1671 (C–O). Positive FAB-MS m/z:
4.28 (2H, m, a, aꢀ-H), 7.22—7.33 (10H, aromatic-H).5,6)
486 [Mꢄ1]ꢄ. H-NMR (DMSO-d6 : D2Oꢅ1 : 1) d: 4.80 (2H,
1
Synthesis of 4,4ꢀ-Dihydroxy-a-truxillic Acid (2) Com- m, b, bꢀ-H), 4.87 (2H, m, a, aꢀ-H), 7.25 (10H, m, benzoyl-
pound 2 was prepared according to a previous report.1) mp H), 7.34 (4H, d, Jꢅ7.8 Hz, 4-chlorophenyl-3, 3ꢀ, 5, 5ꢀ-H),
ꢃ300 °C (lit. 340 °C). Anal. Calcd for C18H16O6: C, 65.85%; 7.71 (4H, d, Jꢅ7.8 Hz, 4-chlorophenyl-2, 2ꢀ, 6, 6ꢀ-H).5)
H, 4.91%. Found: C, 65.75%; H, 4.92%. Positive FAB-MS
Synthesis of b-Truxinic Acid (7) A mixture of bis-(p-
1
m/z (rel. int.): 329 [MꢄH]ꢄ. H-NMR (DMSO-d6) d: 3.64 nitrophenyl) b-truxinic acid (690 mg, 1.28 mmol) and KOH
(2H, m, b, bꢀ-H), 4.13 (2H, m, a, aꢀ-H), 6.69 (4H, d, Jꢅ8.8 (360 mg) in MeOH (7 ml) was refluxed for 3 h. After being
Hz, 3, 3ꢀ, 5, 5ꢀ-H), 7.11 (4H, d, Jꢅ8.8 Hz, 2, 2ꢀ, 6, 6ꢀ-H).7)
adjusted to pH 3 with HCl, the reaction mixture was poured
Synthesis of a-Truxillic Acid Dimethylester (3) To a into water and filtered off. The precipitate was recrystallized
solution of a-truxillic acid (1.0 g, 3.38 mmol) in methyl ethyl from acetic acid to give b-truxinic acid as a needle crystal
ketone (40 ml), trimethylsilyldiazomethane (4 ml) in ether (205 mg, 53.9%), mp 208—209 °C (lit. 209—210 °C). IR
(10 ml) was added dropwise under stirring. After 1 h reac- (KBr) cmꢂ1: 3035 (OH), 1690 (CꢅO). EI-MS m/z: 296
1
tion, the reaction solution was concentrated under reduced [M]ꢄ. H-NMR (DMSO-d6) d: 4.01 (2H, m, b, bꢀ-H), 4.24
pressure, and the residue was purified by column chromatog- (2H, m, a, aꢀ-H), 7.15 (10H, m, aromatic-H).5,10)
raphy on silica gel with CHCl3–MeOH (50 : 1) and recrystal-
Synthesis of 4,4ꢀ-Dichloro-b-truxinic Acid (8) 4-
lized with CHCl3 to give a-truxillic acid dimethylester as a Chlorocinnamic acid crystals (500 mg, 2.75 mmol) were
needle crystal (308 mg, 28.1%), mp 172—174 °C (lit. treated in a manner similar to the conversion of (E)-cinnamic
174 °C). IR (KBr) cmꢂ1: 1695 (CꢅO). EI-MS m/z: 325 acid to a-truxillic acid (1). After 2 d irridation, the product
1
[Mꢄ1]ꢄ. H-NMR (CDCl3) d: 3.30 (6H, s, COOMe), 3.97 was filtered off and recrystallized twice from acetic acid to
(2H, m, b, bꢀ-H), 4.45 (2H, m, a, aꢀ-H), 7.30 (10H, m, aro- give 4,4ꢀ-dichloro-b-truxinic acid as a plate crystal (314 mg,
matic-H).6)
62.8%), mp 159—160 °C (lit. 160 °C). IR (KBr) cmꢂ1: 3102
Synthesis of 4,4ꢀ-Dihydroxy- -truxillic Acid Dimethyl- (OH), 1719 (CꢅO). Positive FAB-MS m/z: 365 [Mꢄ1]ꢄ. 1H-
ester (4) To a solution of 4,4ꢀ-dihydroxy-a-truxillic acid NMR (DMSO-d6) d: 3.59 (2H, m, b, bꢀ-H), 4.05 (2H, m, a,
(1.0 g, 3.05 mmol) in methyl ethyl ketone (40 ml) and MeOH aꢀ-H), 6.64, 7.06 (each 4H, d, Jꢅ8.25 Hz, aromatic-H).11)
a
(20 ml), trimethylsilyldiazomethane (4 ml) in ether (10 ml)
Synthesis of 1,2-Bis(3-pyridyl)cyclobutane-3,4-dicar-
was added dropwise under stirring. After 1 h reaction, the re- boxylic Acid (9) 3-(3-Pyridyl)acrylic acid (500 mg, 3.36
action solution was concentrated under reduced pressure, and mmol) was treated in a manner similar to the conversion of
the residue was purified by column chromatography on silica (E)-cinnamic acid to a-truxillic acid (1). After 4 d irridation,
gel with hexane–AcOEt (3 : 1→1 : 2), then recrystallized the product was filtered off and gave 1,2-bis(3-pyridyl)-
with CHCl3 to give 4,4ꢀ-dihydroxy-a-truxillic acid dimethyl- cyclobutane-3,4-dicarboxylic acid as
a white powder
ester as a needle crystal (892 mg, 82.2%), mp 178—181 °C (98.0%), mp 200 °C. IR (KBr) cmꢂ1: 3398 (OH), 1578
(lit. 174—177 °C). IR (KBr) cmꢂ1: 1690 (CꢅO). EI-MS m/z: (C–O). Negative FAB-MS m/z: 297 [Mꢂ1]ꢂ. 1H-NMR (pyri-
357 [Mꢄ1]ꢄ. 1H-NMR (pyridine-d5) d: 3.37 (6H, s, dine-d5) d: 4.45 (2H, m, b, bꢀ-H), 4.82 (2H, m, a, aꢀ-H),

582
Vol. 29, No. 3
7.25—7.92 (8H, m, pyridine-H).12)
ple range test for multiple comparisons were used. Differ-
b
Synthesis of -Truxinic Acid Bis(p-nitrophenyl)ester ences were considered significant at pꢆ0.01.
(10) To a solution of p-nitrophenol (6.0 g, 40.96 mmol) in
THF (60 ml) and pyridine (1.9 g), cinnamyl chloride (6.8 g) RESULTS AND DISCUSSION
in THF (2 ml) was added at 0—5 °C, and the reaction was
stirred at the same temperature for 2 h. The reaction mixture
The formalin-induced licking response has been used as a
was then poured into water (800 ml) and filtered off to give a model for evaluating new analgesics.14,16) The duration of the
precipitate (9.7 g, 88.0%). The dried precipitate (9.0 g, 33.45 nociceptive response induced by formalin can be divided into
mmol) was treated in a manner similar to the conversion of two phases. The early phase is from 0 to 10 min after forma-
(E)-cinnamic acid to a-truxillic acid (1). After 10 h irrida- lin injection, and the late phase is from 10 to 30 min after the
tion, the product was filtered off and recrystallized from injection. The pain of the early phase is evoked by direct
methyl ethyl ketone to give b-truxinic acid bis(p-nitro- stimulation of the nerve fibers, and that of the late phase is
phenyl)ester as a needle crystal (6.4 g, 62.4%), mp 195— due to the inflammatory reaction. Centrally acting drugs such
196 °C (lit. 192—193 °C). IR (KBr) cmꢂ1: 1763, 1734
(CꢅO), 1522, 1350 (NO2). Positive FAB-MS m/z: 539
1
[Mꢄ1]ꢄ (100). H-NMR (CDCl3) d: 4.28 (2H, m, b, bꢀ-H),
4.62 (2H, m, a, aꢀ-H), 7.01—7.22 (10H, m, aromatic-H),
7.26, 8.27 (each 4H, dd, Jꢅ1.98, 6.92 Hz, p-nitrophenyl-
H).10)
Formalin Test and Treatments This method repre-
sented a modification of that described by Dubuisson and
Dennis.13) Male ddy mice (25ꢁ5 g) were used. The tested
drugs (40 mg/kg) were prepared as suspensions with 0.5%
Tween 80/saline and intraperitoneally administered 10 min
prior to the injection of an inducer (1.0% formalin/saline,
20 ml). The mice were observed for 30 min, and the time the
mice spent licking the injected right hindpaw was recorded.
The nociceptive scores normally peaked at 0 to 10 min after
formalin injection (early phase) and 10 to 30 min (late phase)
after the injection, representing the neurogenic and inflam-
matory pain responses, respectively.14) The time spent licking
the injected paw was recorded and the data are expressed as
total licking time in the early phase and late phase.
Comparison with Indomethacin The tested drugs
(0.625, 2.5, 10, 40 mg/kg for 2; 2.5, 10, 40 mg/kg for 6 and
8; 10, 20, 40 mg/kg for indomethacin) were prepared as sus-
pensions with 0.5% Tween 80/saline, and were intraperi-
toneally administered 10 min prior to the injection of an in-
ducer (2.0% formalin/saline, 20 ml).
Fig. 2. Activities of Compounds 1—10 (40 mg/kg, i.p.) against Formalin
(1.0%, 20 ml)-Induced Neurogenic Pain Response (Early Phase) and Inflam-
matory Pain Response (Late Phase) in Mice
Statistical Analysis All values are expressed as meanꢁ
S.E. (nꢅ10). The ED50 was determined according to the
Method of Lichfield and Wilcoxon.15) For statistical analysis,
Each column represents meanꢁS.E. (nꢅ10). Significant differences between control
one-way analysis of variance combined with Dunnett’s multi- and drug-treated groups are indicated by ∗∗ pꢆ0.01.
Table 1. Comparison of the Activities of 4,4ꢀ-Dihydroxy-a-truxillic Acid (2), 1,3-Dibenzoyl-2,4-di(4-chlorophenyl)cyclobutane (6) and 4,4ꢀ-Dichloro-b-
truxinic Acid (8) with That of Indomethacin (INDO) in the Formalin Test (2.0%, 20 ml)
Inhibition (%)
ED50 (mmol/kg)
Early phase
Dose
(mg/kg, i.p.)
Group
Early phase
Late phase
Late phase
2
0.625
2.5
10
40
2.5
10
40
2.5
10
40
10
4.12
10.98
13.04
15.66
11.89
20.69
42.41
15.56
15.49
51.07
—
32.48
59.80
82.12
96.12
37.13
57.73
87.72
19.87
57.61
81.23
22.07
26.51
54.08
—
4.93 (1.92—12.6)
6
8
158.0 (27.4—909.0)
150.0 (38.1—594.0)
—
11.2 (3.48—36.2)
24.9 (10.9—57.0)
117.0 (52.6—258.0)
INDO
20
40
13.70
9.40

March 2006
583
as morphine inhibited both early and late phases equally. On induced by formalin. The highest anti-inflammatory activi-
the other hand, peripheral acting drugs such as indomethacin ties were observed for a-truxillic acid (1) and its derivative
and aspirin inhibited only the late phase.
4,4ꢀ-dihydroxy-a-truxillic acid (2). In addition, 4,4ꢀ-dichloro-
In the present study, a-truxillic acid (1) and its derivatives lated derivatives 1,3-dibenzoyl-2,4-di(4-chlorophenyl)cyclo-
(2—6), as well as b-truxinic acid (7) and its derivatives (8— butane (6) and 4,4ꢀ-dichloro-b-truxinic acid (8) also showed
10) were prepared, and their activities were evaluated in the strong anti-inflammatory activities. a-Truxillic acid (1) and
formalin test. All compounds were tested intraperitoneally its derivative a-truxillic acid bis(p-nitrophenyl)ester (5) had
for their activities at the dose of 40 mg/kg body weight by higher anti-inflammatory activities than b-truxinic acid (7)
formalin test, and the results are summarized in Fig. 2. All and the corresponding derivative (10). Free carboxylic acids
compounds showed only weak or no activity against the neu- (1, 2) showed higher activities than their dimethyl esters (3,
rogenic pain response (early phase), but exhibited significant 4). The anti-inflammatory activity of a-truxillic acid bis(p-
activity against the inflammatory pain response (late phase) nitrophenyl)ester (5) was weaker than that of a-truxillic acid
(1), but the activity of b-truxinic acid bis(p-nitrophenyl)ester
(10) was equal to that of b-truxinic acid (7). 4,4ꢀ-Dichloro-b-
truxinic acid (8) had higher activity than its parent compound
b-truxinic acid (7).
We then compared the anti-inflammatory activities of 4,4ꢀ-
dihydroxy-a-truxillic acid (2), which showed the highest
anti-inflammatory activity, and two 4,4ꢀ-dichlorolated deriva-
tives (6, 8) which also showed strong activities, with that of
indomethacin (a nonsteroidal anti-inflammatory drug) in the
formalin test, with a higher concentration formalin (2.0%,
20 ml) as an inducer. All compounds exerted only weak or no
activity against the neurogenic pain response (early phase),
but produced marked and dose-related inhibition of the for-
malin-induced inflammatory pain response (late phase). The
anti-inflammatory activities of 2, 6 and 8 were stronger than
that of indomethacin. The ED50 values for 2, 6 and 8 were
4.93, 11.2 and 24.9 mmol/kg respectively, while the value for
indomethacin was 117.0 mmol/kg (Table 1, Fig. 3). The activ-
ities of 2, 6 and 8 were about 5- to 24-fold more potent than
that of indomethacin. These results suggested that 2, 6 and 8
possess very strong anti-inflammatory activity, and a-truxil-
lic acid and b-truxinic acid derivatives might be developed
into a new type of anti-inflammatory drug.
In conclusion, these results clearly indicated that a-truxil-
lic acid and b-truxinic acid derivatives are another group of
leading compounds important for developing a new type of
anti-inflammatory drug. The result also supported that the
dimeric structure played an important role in the expression
of anti-inflammatory activity. Further investigation is re-
quired to develop more potent anti-inflammatory derivatives
and to elucidate the precise mechanism underlying these ac-
tivities.
REFERENCES
1) Chi Y. M., Nakamura M., Yoshizawa T., Zhao X. Y., Yan W. M.,
Hashimoto F., Kinjo J., Nohara T., Sakurada S., Biol. Pharm. Bull., 28,
1776—1778 (2005).
2) Chi Y. M., Yan W. M., Li J. S., Phytochemistry, 29, 2376—2378
(1990).
3) Nakamura M., Chi Y. M., Yan W. M., Nagasugi Y., Yoshizawa T., Irino
N., Hashimoto F., Kinjo J., Nohara T., Sakurada S., J. Nat. Prod., 62,
1293—1294 (1999).
4) Nakamura M., Chi Y. M., Yan W. M., Nagasugi Y., Yoshizawa T., Irino
N., Hashimoto F., Kinjo J., Nohara T., Sakurada S., Planta Medica, 67,
Fig. 3. Activities of 4,4ꢀ-Dihydroxy-a-truxillic Acid (2), 1,3-Dibenzoyl-
2,4-di(4-chlorophenyl)cyclobutane (6), 4,4ꢀ-Dichloro-b-truxinic Acid (8)
and Indomethacin (INDO) against Formalin (2.0%, 20 ml)-Induced Neuro-
genic Pain Response (Early Phase) and Inflammatory Pain Response (Late
Phase) in Mice
114—117 (2001).
5) Montaudo G., Caccamese S., J. Org. Chem., 38, 710—716 (1973).
6) Baracchi A., Chimichi S., Sio F. D., Donati D., Nesi R., Sarti-Fantoni
P., Heterocycles, 24, 2863—2870 (1986).
7) Morrison W. H., Hartley R. D., Himmelsbach D. S., J. Agric. Food
Chem., 40, 768—771 (1992).
Each column represents meanꢁS.E. (nꢅ10). Significant differences between control
and drug-treated groups are indicated by ∗∗ pꢆ0.01.
8) Koshino H., Terada S., Yoshihara T., Shimanuki T., Sato T., Tajimi A.,

584
Vol. 29, No. 3
Phytochemistry, 27, 1333—1338 (1988).
12) Lahav M., Schmidt Q. M., J. Chem. Soc. (B), 1967, 239—243 (1967).
9) Nishibuto T., Iizawa T., Murada M., Koshida M., Harita Y., JP 13) Dubuisson D., Dennis S. G., Pain, 4, 161—174 (1977).
H02/28208 (1990).
14) Hunskaar S., Hole K., Pain, 30, 103—114 (1987).
10) Nishikubo T., Takahashi E., Miyaji T., Iizawa T., Bull. Chem. Soc.
Jpn., 58, 3399—3400 (1985).
15) Lichfield J. T., Wilcoxon G. L., Jr., J. Pharm. Exp. Ther., 96, 99—105
(1949).
11) Cohen M. D., Schmidt G. M. J., Sonntag F. I., J. Chem. Soc., 1964,
2000—2013 (1964).
16) Shibata M., Ohkubo T., Takahashi H., Inoki R., Pain, 38, 347—352
(1989).