Chemopreventive potential of cyclic diarylheptanoids.

Article abstract of DOI:10.1016/S0968-0896(02)00164-5

Eleven cyclic diarylheptanoids and seven related compounds were screened as potential antitumor promoters by using the in vitro short-term 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced Epstein-Barr virus early antigen (EBV-EA) activation assay. In addition, the cyclic diarylheptanoid myricanone (2) was examined for antitumor initiating activity in a two-stage carcinogenesis assay of mouse skin tumors induced by peroxynitrite as an initiator and TPA as a promoter. Myricanone (2) exhibited significant antitumor-initiating effect on mouse skin. These data suggest that cyclic, as well as linear, diarylheptanoids might be valuable chemopreventors.

Full text of DOI:10.1016/S0968-0896(02)00164-5

Bioorganic & Medicinal Chemistry 10 (2002) 3361–3365
Chemopreventive Potential of Cyclic Diarylheptanoids
Junko Ishida,a Mutsuo Kozuka,b Harukuni Tokuda,c Hoyoku Nishino,c Seiji Nagumo,a
Kuo-Hsiung Leeb,* and Masahiro Nagaia
aHoshi University, 2-4-41, Ebara, Shinagawa-ku, Tokyo 142-8501, Japan
bNatural Products Laboratory, School of Pharmacy, University of North Carolina, Chapel Hill, NC 27599-7360, USA
cDepartment of Biochemistry, Kyoto Prefectural University of Medicine, Kamigyo-ku, Kyoto 602-0841, Japan
Received 28 November 2001; accepted 30 January 2002
Abstract—Eleven cyclic diarylheptanoids and seven related compounds were screened as potential antitumor promoters by using
the in vitro short-term 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced Epstein–Barr virus early antigen (EBV-EA) activation
assay. In addition, the cyclic diarylheptanoid myricanone (2) was examined for antitumor initiating activity in a two-stage carci-
nogenesis assay of mouse skin tumors induced by peroxynitrite as an initiator and TPA as a promoter. Myricanone (2) exhibited
significant antitumor-initiating effect on mouse skin. These data suggest that cyclic, as well as linear, diarylheptanoids might be
valuable chemopreventors. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
Epstein–Barr virus early antigen (EBV-EA) activation
assay.12 We have also found that 2 strongly inhibited
tumor promotion induced by TPA following initiation
with 7,12-dimethylbenz [a] anthracene (DMBA) in
mouse skin.12 In this study, we have extended our
screening to clarify the structure–activity relationships
for antitumor promotion activity in this compound
class. In addition, we also assayed the antitumor-initi-
ating activity of 2 in a two-stage mouse skin carcino-
genesis test initiated by peroxynitrite. These results
should help in understanding the physiological func-
tions and biochemical mechanism of cancer chemo-
preventive activity of cyclic diarylheptanoids.
Diarylheptanoids belong to a class of natural products
based on 1,7-diphenylheptane. Their structures are
classified into linear and cyclic types, and the latter is
divided further into biphenyl ([7.0]-metacyclophanes)
and diphenyl ether (14-oxa-[7.1]-metaparacyclophanes)
types. Each cyclic type may be formed from the corre-
sponding linear type by phenolic oxidative coupling.1
Linear type diarylheptanoids exhibit a broad range of
potent biological activities including anti-inflammatory,
antihepatoxic, antifungal, antibacterial, and related
effects.2 For example, curcumin (1) (Fig. 1), a yellow
component present in turmeric (Curcuma longa Linn)
and that has been used as a coloring agent exerts anti-
carcinogenic or anti-mutagenic effects in diverse animal
models and also in cultured cells.3À9 It has also been
found to have antioxidant and oxygen radical scaven-
ging activities, including the inhibition of ROS (super-
oxide, peroxyl, and hydroxy radicals) and NO donor
effect.10,11 However, few studies have been reported on
cancer chemopreventive activity of the cyclic diaryl-
heptanoid class.
Chemistry
The chemical structures are shown in Figure 1. The
isolation of test compounds, 12-dehydroporson (3),
porson (4), myricanone oxime (5), myricamine (6),
acerogenin K (7), isomyricanone (8), acerogenin C (10),
acerogenin C oxime (11), galeon (12), (À)-centrolobol
(13), (À)-di-O-methylcentrolobol (14), and (+)-rhodo-
dendrol (16), was described previously.13,14 Compound
15 was synthesized by refluxing 5-bromovanillin with
copper powder in anhydrous DMF. Compound 17 was
purchased from Tokyo Kasei Kogyo Co. Compound 18
was synthesized by methylation of 17 with CH3I/K2CO3
in refluxing dry acetone. The structures of compounds
15 and 18 were confirmed by NMRanalysis and MS
analysis, respectively. All compounds were recrys-
tallized and were homogeneous by TLC.
Recently, we described the significant inhibitory effect
of the cyclic diarylheptanoid myricanone (2) (Fig. 1),
derived from Myrica rubra, using the in vitro short-term
12-O-tetradecanoylphorbol-13-acetate (TPA)-induced
*Corresponding author. Tel.: +1-919-962-0066; fax: +1-919-966-
3893; e-mail: khlee@unc.edu
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00164-5

3362
J. Ishida et al. / Bioorg. Med. Chem. 10 (2002) 3361–3365
Results and Discussion
In contrast, replacing the C-11 carbonyl group of 2 with
other functional groups, including oxime (5) and amine
(6), decreased the activity. In addition, 7, which has a
hydroxy rather than a carbonyl group at C-11, showed
reduced activity. Also, isomerization of 2, which gave
the [7-ortho, meta-0] cyclophane structure 8, decreased
the activity. Thus, the 13-membered ring seems to be
important for the antitumor-promoting activity, per-
haps because of steric factors.
In vitro EBV-EA activation
The primary screening test was carried out utilizing a
short-term in vitro synergistic assay on EBV-EA acti-
vation. Table 1 lists inhibitory effects of 1–18 on the
EBV-EA activation induced by TPA and the associated
viability of Raji cells. Comparison of the data in Table
1, including statistical discussion, suggests the following
general trends.
All tested diphenyl ether-type cyclic diarylheptanoids
(9–12) were less active than the biphenyl-type com-
pounds (2, 3, and 13-oxomyricanol).12 In addition, 13,
which is the linear biosynthetic precursor to 9,12 and 14
exhibited moderate activity. Thus, cyclization to the
diphenyl ether-type cyclic diarylheptanoid diminished
the antitumor promoting activity. We also tested 15–18,
which constitute partial structures found in 2. These
four compounds showed reduced activity, thus empha-
sizing the importance of the cyclic structure.
At 100 mol ratio/TPA, all compounds tested in this
study, except 6, showed inhibitory effects on EBV-EA
activation without cytotoxicity on Raji cells as shown in
Table 1. Particularly, myricanone (2), 12-dehydro-
porson (3), and curcumin (1) showed strongly increased
inhibitory effects with increasing concentration, as
reflected by complete inhibition at 1000 mol ratio/TPA.
Compounds 120 and 212 have been reported previously
as potent antitumor promoters.
We previously reported that the extended quinol
structure of 2 is important in maintaining the anti-
tumor promoting activity of cyclic diarylheptanoids.12
Therefore, compounds 3 and 4, which are methylated at
the 5-hydroxy group, should be less potent than curcu-
min (1). However, although 4 showed reduced activity
as expected, 3 was essentially equally potent to 1. Com-
In summary, the following conclusions were drawn. (a)
The presence of an a-diketo moiety in the side chain is
important to activity. (b) Replacing the side-chain
ketone with other substituents (amine, oxime, hydroxyl)
reduced activity. (c) Cyclization as the meta, meta-
bridged biphenyl structure, which produces a 13-mem-
bered ring, rather than the diphenyl ether-type structure
is important for antitumor promotion activity.
paring the structures of
3 (a-diketo at C-11,12
positions), 4 (a-ketol), and monomethylmyricanone
(11-ketone), which was also less active than 2 as repor-
ted previously (data not shown),12 indicated that the
a-diketo moiety in the heptane chain is important for
activity.
In vivo two-stage carcinogenesis test on mouse skin
papillomas initiated by peroxynitrite
On the basis of the above results, myricanone (2) was
the most active antitumor promoting agent among the
tested cyclic diarylheptanoids. To elucidate the bio-
chemical mechanism of cancer chemopreventive activity
of 2, we tested 2 for its effect on in vivo mouse skin
carcinogenesis, initiated by peroxynitrite and promoted
by TPA. The incidence (%) of papilloma-bearing mice
and the average numbers of papillomas per mouse are
presented in Figure 2A and B, respectively.
Figure 2A shows the time course of tumor formation in
the two groups. In group I (control, treated with per-
oxynitrite/TPA), the first tumor appeared after 6 weeks.
In group II (treated with peroxynitrite and myricanone/
TPA), the first tumor appeared after 8 weeks. The per-
centage of tumor-bearing mice in group I was 100%
after 10 weeks, whereas that in group II was approxi-
mately 26% at that same time and 86% after 20 weeks
of promotion. Figure 2B shows the average number of
tumors per mouse. Group I produced approximately
eight tumors per mouse after 20 weeks, whereas group
II had only approximately five tumors per mouse.
Therefore, treatment with 0.0025% of myricanone
(group II) caused 38% reduction in the average number
of tumors per mouse after 20 weeks of promotion.
From these results, myricanone (2) appears effective for
the inhibition of peroxynitrite-induced carcinogenesis
on mouse skin. Peroxynitrite, which is produced by the
reaction of nitric oxide (NO) with superoxides, is a
Figure 1. Structures of cyclic diarylheptanoids and derivatives.

J. Ishida et al. / Bioorg. Med. Chem. 10 (2002) 3361–3365
3363
potent oxidant as well as nitrating and hydroxylating
agent. We have previously reported that peroxynitrite
shows tumor initiation.17 Together with results from
our previous study,12 these data show that 2 plays a
useful physiological role, including both anti-initiating
and anti-promoting activity on two-stage carcinogen-
esis. Curcumin (1) is known to have potent antioxidant
activity, which suggests that it intercepts and neutralizes
potent chemical carcinogens, such as ROS (superoxide,
peroxyl, hydroxy radicals) and NO donors.10,11 Corre-
spondingly, possible antioxidative effects of the cyclized
curcumin derivative myricanone (2) might account for
the observed inhibition of peroxynitrite-induced carcino-
genesis with this compound.
(St. Louis, MO, USA). EBA-EA positive serum from a
patient with nasopharyngeal carcinoma (NPC), used for
an immunofluorescence test, was a gift from Prof. H.
Hattori, Department of Otorhinolaryngology, Kobe
University. Peroxynitrite was purchased from Dojindo
Laboratories Co (Kumamoto, Japan).
Cells
EBV genome carrying lymphoblastoid cells (Raji cells
derived from Burkitt’s lymphoma) were cultured in
RPMI-1640 medium (Nissui, Tokyo, Japan) under the
conditions described previously.15 Spontaneous activa-
tion of EBV-EA in our subline of Raji cells was less
than 0.1%.
Conclusions
Animals
In summary, not only linear-type (1) but also cyclic-type
(2) diarylheptanoids can have substantial chemopreven-
tive properties, and 2 could be a valuable chemo-
preventor. Additional study is needed to elucidate the
relationship of diarylheptanoid structure to chemo-
preventor effects. However, these derivatives offer new
insights into possible derivatives that may hold promise
for the development of potent new chemopreventive
agents.
Specific pathogen-free (SPF) female SENCARmice (6
weeks old) were obtained from Japan SLC, Inc.
(Hamamatsu, Japan), and maintained under SPF con-
ditions in Kyoto Prefectural University of Medicine,
Animal Center. The mice were housed five per poly-
carbonate cage, in a temperature-controlled room at
24Æ2 ꢀC, and given food (Oriental MF, Oriental Yeast
Co., Tokyo, Japan) and water ad libitum during the
experiments.
In vitro EBV-EA activation experiments
Experimental
Chemicals
The inhibition of EBV-EA activation was assayed using
Raji cells (virus non-producer type), an EBV genome-
carrying human lymphoblastoid cell, which were culti-
vated in 10% fetal bovine serum (FBS) RPMI 1640
medium. The indicator cells (Raji, 1Â106/mL) were
incubated at 37 ꢀC for 48 h in 1 mL of medium contain-
Cell culture reagents, n-butyric acid, and other reagents
were purchased from Nacalai Tesque, Inc. (Kyoto,
Japan). TPA was obtained from Sigma Chemical Co.
Table 1. Relative ratioa of EBV-EA activation with respect to positive control (100%) in presence of myricanone and its derivatives (1–18)
Compound
Concentration (mol ratio/TPA)b
100
c
1000
500
10
IC50
Curcumin (1)d
0.0Æ0.5 (60)
0.0Æ0.3 (60)
0.0Æ0.6 (70)
10.5Æ0.8 (60)
21.8Æ1.1 (70)
40.6Æ1.3 (70)
7.8Æ1.2 (60)
5.7Æ0.5 (60)
18.2Æ0.6 (60)
20.3Æ0.6 (60)
15.5Æ0.4 (60)
18.6Æ0.3 (60)
5.0Æ0.5 (60)
8.3Æ0.5 (60)
9.5Æ0.3 (60)
8.7Æ0.5 (60)
10.2Æ0.6 (60)
14.8Æ0.9 (60)
22.8Æ1.8 (>80)
25.4Æ1.2 (>80)
28.5Æ1.2 (>80)
39.6Æ1.4 (>80)
51.7Æ1.9 (>80)
72.1Æ2.1 (>80)
37.4Æ1.1 (>80)
33.7Æ1.2 (>80)
47.0Æ1.5 (>80)
60.1Æ1.3 (>80)
51.7Æ1.9 (>80)
54.0Æ1.2 (>80)
36.6Æ1.2 (>80)
39.7Æ1.1 (>80)
43.2Æ1.9 (>80)
39.8Æ1.2 (>80)
40.4Æ1.5 (>80)
43.0Æ1.9 (>80)
81.7 +1.9 (>80)
76.4Æ2.2 (>80)
77.0Æ1.6 (>80)
75.4Æ1.7 (>80)
84.9Æ2.0 (>80)
93.7Æ1.9 (>80)
76.8Æ2.0 (>80)
75.9Æ2.0 (>80)
83.5Æ2.7 (>80)
82.0Æ1.9 (>80)
75.0Æ2.2 (>80)
78.9Æ1.9 (>80)
74.0Æ2.0 (>80)
77.2Æ1.9 (>80)
72.4Æ2.1 (>80)
75.1Æ1.9 (>80)
75.7Æ1.8 (>80)
76.9Æ2.1 (>80)
100.0Æ0.2 (>80)
90.9Æ0.6 (>80)
96.8Æ0.5 (>80)
100.0Æ0.6 (>80)
100.0Æ0.3 (>80)
100.0Æ0.0 (>80)
100.0Æ0.3 (>80)
100.0Æ0.7 (>80)
100.0Æ0.3 (>80)
100.0Æ0.2 (>80)
100.0+0.4 (>80)
100.0+0.2 (>80)
96.0Æ0.5 (>80)
100.0Æ0.6 (>80)
100.0Æ0.5 (>80)
100.0Æ0.3 (>80)
100.0Æ0.8 (>80)
100.0Æ0.4 (>80)
320
260
300
450
500
820
420
350
490
620
450
580
320
420
340
360
400
460
Myricanone (2)d
3
4
5
6
7
8
9d
10
11
12
13
14
15
16
17
18
aValues represent relative percentage to the positive control value (100%). Data are expressed as meanÆSD. Values in parentheses are viability
percentages of Raji cells.
bMol ratio/TPA (20 ng=32 pmol/mL). The values of all compounds at 500 mol ratio/TPA and the values of 3 at 100 mol ratio/TPA were sig-
nificantly different from the control value (p <0.05) without cytotoxicity.
cThe concentration (Mol ratio/TPA) needed to inhibit 50% of positive control activated with TPA.
dIshida et al. reported previously.12

3364
J. Ishida et al. / Bioorg. Med. Chem. 10 (2002) 3361–3365
ing n-butyric acid (4 mM, inducer), TPA (32 nM), and
various amounts of the test compounds dissolved in
5 mL of DMSO. Smears were made from the cell sus-
pension. The EBV-EA inducing cells were stained with
high titer EBV-EA positive serum from NPC patients
and detected by an indirect immunofluorescence tech-
nique.16 In each assay, at least 500 cells were counted,
and the number of stained cells (positive cells) was
recorded. Triplicate assays were performed for each
data point. The EBV-EA inhibitory activity of the test
compound was expressed by comparison with that of
the positive control experiment (100%), which was car-
ried out with n-butyric acid (4 mM) plus TPA (32 nM).
In the experiments, the EBV-EA induction was ordina-
rily around 35%, and this value was taken as the posi-
tive control (100%). n-Butyric acid (4 mM) alone
induced 0.1% EA-positive cells. The viability of treated
Raji cells was assayed by the trypan-blue staining
method. Student’s t-test was used for all statistical ana-
lyses.
In vivo two-stage carcinogenesis test on mouse skin
papillomas initiated by peroxynitrite17
The animals were divided into two experimental groups
of 15 mice each. The back of each mouse was shaved
with surgical clippers, and the mice were treated topi-
cally with peroxynitrite (390 nmol) in 1 mM NaOH
(0.1 mL). One week after the initiation, papilloma for-
mation was promoted by the twice-a-week application
of TPA (1.7 nmol) in acetone (0.1 mL) on the skin (no
papilloma formation was seen with topical application
of the acetone solvent alone). Test samples (0.0025%
myricanone) were orally fed with drinking water for
only 2 weeks before the promotion treatment (1 week
both before and after initiation). Two groups received
initiation/promotion treatments as follows: (I) peroxy-
nitrite/TPA, n=15 (II) peroxynitrite and myricanone/
TPA, n=15. The incidence and numbers of papillomas
were detected weekly for 20 weeks, as described.18,19
Student’s t-test was used for all statistical analyses.
Acknowledgements
This investigation was supported in part by a grant
from the National Cancer Institute (CA17625) awarded
to K. H. Lee and from the Ministry of Education,
Science, Sports and Culture, and Ministry of Health
and Welfare, Japan.
References and Notes
1. Henley-Smith, P.; Whiting, D. A.; Wood, A. F. J. Chem.
Soc., Perkin Trans. 1 1980, 614.
2. Keseru, G. M.; Nogradi, M. In Studies in Natural Products
Chemistry; Atta-ur-Pahman, Ed.; Elsevier Science: New York,
1995; Vol. 17, p 357.
3. Soudamini, K. K.; Kuttan, R. J. Ethnopharmacol. 1989, 27,
227.
4. Nagabhushan, M.; Bhide, S. V. J. Am. Coll. Nutr. 1992, 11,
192.
5. Conney, A. H.; Lou, Y. R.; Xie, J. G.; Osawa, T.; New-
mark, H. L.; Liu, Y.; Chang, R. L.; Huang, M. T. Proc. Soc.
Exp. Biol. Med. 1997, 216, 234.
6. Rao, C. V.; Rivenson, A.; Simi, B.; Reddy, B. S. Cancer
Res. 1995, 55, 259.
Figure 2. Inhibition of peroxynitrite/TPA-induced tumor promotion
by multiple application of myricanone (2). (A) Percentage of mice
bearing papillomas. (B) Average number of papillomas per mouse. *,
peroxynitrite (390 nmol) + TPA (1.7 nmol) (Group I); *, peroxyni-
trite (390 nmol) + 0.0025% of 2 (2 weeks) + TPA (1.7 nmol) (Group
II). At 20 weeks of promotion, the number of papillomas per mouse
differed significantly (p <0.05) between Group I and Group II. The
number (standard deviations are shown in parentheses) of papillomas
per mouse for each group was 8.1 (1.9) and 5.0 (2.0) for Groups I and
II, respectively.
7. Abraham, S. K.; Sarma, L.; Kesavan, P. C. Mutat. Res.
1993, 303, 109.
8. Kuo, M. L.; Huang, J. T. S.; Lin, K. Biochim. Biophys.
Acta 1996, 1317, 95.
9. Mehta, K.; Pantazis, P.; McQeen, T.; Aggarwal, B. B.
Anticancer Drugs 1997, 8, 470.
10. Masuda, T.; Hidaka, K.; Shinohara, A.; Maekawa, T.;
Takeda, Y.; Yamaguchi, H. J. Agric. Food Chem. 1999, 47, 71.
11. Onoda, M.; Inano, H. Nitric Oxide 2000, 4, 505.

J. Ishida et al. / Bioorg. Med. Chem. 10 (2002) 3361–3365
3365
12. Ishida, J.; Kozuka, M.; Wang, H. K.; Konoshima, T.;
Tokuda, H.; Okuda, M.; Mou, X. Y.; Nishino, H.; Sakurai,
N.; Lee, K. H.; Nagai, M. Cancer Lett. 2000, 159, 135.
13. Nagai, M.; Dohi, J.; Morihara, M.; Sakurai, N. Chem.
Pharm. Bull. 1995, 43, 1674.
Biochem. Biophys. Part B 2000, 3, 256. (b) Tokuda, H.;
Okuda, M.; Mou, X. Y.; Ichiishi, E.; Mukainaka, T.; Nishino,
H.; Takasaki, M.; Konoshima, T. Arch. Biochem. Biophys.
Part B 2000, 3, 284.
18. Weeks, C. W.; Slaga, T. J.; Hennings, H.; Gleason, G. L.;
Bracken, W. M. J. Natl. Cancer Inst. 1979, 63, 401.
19. Tokuda, H.; Ohigashi, H.; Koshimizu, K.; Ito, Y. Cancer
Lett. 1986, 33, 279.
14. Inoue, T. Yakugaku Zasshi 1993, 113, 181.
15. Takasaki, M.; Konoshima, T.; Fujitani, K.; Yoshida, S.;
Nishimura, H.; Tokuda, H.; Nishino, H.; Iwashima, A.;
Kozuka, M. Chem. Pharm. Bull. 1990, 38, 2737.
16. Henle, G.; Henle, W. J. Bacteriol. 1966, 91, 1248.
17. (a) Tokuda, H.; Okuda, M.; Mou, X. Y.; Mukainaka, T.;
Ichiishi, E.; Nishino, H.; Takasaki, M.; Konoshima, T. Arch.
20. Huang, M. T.; Ma, W.; Lu, Y. P.; Chang, R. L.; Fisher,
C.; Manchand, P. S.; Newmark, H. L.; Conney, H. Carcino-
genesis 1995, 16, 2493.