Two New Anti-Tumor Promoting Serratane-Type Triterpenoids from the Stem Bark of Picea jezoensis var. jezoensis

Article abstract of DOI:10.1055/s-2003-45153

Two new serratane-type triterpenoids, 1 and 2, were isolated from the stem bark of Picea jezoensis Carr. var. jezoensis (Pinaceae). Their structures were determined to be 3β-methoxyserrat-13-en-21β-ol (1) and 13/3, 14β-epoxy-3β-methoxyserratan-21β-ol (2) on the basis of spectroscopic methods and partial syntheses. Compounds 1 and 2 and their acetates were screened as potential anti-tumor promoters by using the in vitro short-term 12-O-tetradecanoylphorbol 13-acetate (TPA)-induced Epstein-Barr virus early antigen (EBV-EA) activation assay. IC₅₀ value evaluation showed that compound 1 was more effective than others. In addition, compounds 1 and 2 were examined for anti-tumor promoting activities in a two-stage carcinogenesis assay of mouse skin tumors induced by 7,12-dimethylbenz[a] anthracene (DMBA) as an initiator and TPA as a promoter. Compounds 1 and 2 exhibited significant anti-tumor promoting effects on mouse skin carcinogenesis.

Full text of DOI:10.1055/s-2003-45153

Reiko Tanaka1
Yohei Ishikawa1
Toshifumi Minami1
Katsuhiko Minoura1
Harukuni Tokuda2
Shunyo Matsunaga1
Two New Anti-Tumor Promoting Serratane-Type
Triterpenoids from the Stem Bark of Picea jezoensis
var. jezoensis
Abstract
others. In addition, compounds 1 and 2 were examined for anti-
tumor promoting activities in a two-stage carcinogenesis assay
Two new serratane-type triterpenoids, 1 and 2, were isolated of mouse skin tumors induced by 7,12-dimethylbenz[a]anthracene
from the stem bark of Picea jezoensis Carr. var. jezoensis Pina- DMBA) as an initiator and TPA as a promoter. Compounds 1 and
ceae). Their structures were determined to be 3b-methoxyser- 2 exhibited significant anti-tumor promoting effects on mouse
rat-13-en-21b-ol 1) and 13b, l4b-epoxy-3b-methoxyserratan- skin carcinogenesis.
21b-ol 2) on the basis of spectroscopic methods and partial
syntheses. Compounds 1 and 2 and their acetates were screened Key words
as potential anti-tumor promoters by using the in vitro short- 3b-Methoxyserrat-13-en-21b-ol
´ 13b,14b-epoxy-3b-methoxy-
term 12-O-tetradecanoylphorbol 13-acetate TPA)-induced Ep- serratan-21b-ol ´ Picea jezoensis var. jezoensis ´ Pinaceae ´ 2D
stein-Barr virus early antigen EBV-EA) activation assay. IC50 val- NMR ´ in vitro EBV-EA activation assay ´ in vivo two-stage mouse
ue evaluation showed that compound 1 was more effective than skin carcinogenesis test
Introduction
Recently we reported two novel compounds, jezananals A and B
from this bark [8]. Among them, 13a,14a-epoxy-3b-methoxyser-
1041
Chemoprevention is regarded as one of the efficient strategies for ratan-21b-ol 3) showed significant anti-tumor promoting activ-
cancer prevention [1]. Inhibition of the tumor promotion stage in ity in the in vivo two-stage mouse-skin carcinogenesis assay
multistage chemical carcinogenesis has been regarded as the using 7,12-dimethylbenz[a]anthracene DMBA) as an initiator
most promising method for cancer chemoprevention [2]. In the and TPA as a tumor promoter [9].
search for cancer chemopreventive agents, the inhibitory effects
on Epstein-Barr virus early antigen EBV-EA) induction by the tu- Therefore, serratane-type triterpenoids were considered to re-
mor promoter, 12-O-tetradecanoylphorbol 13-acetate TPA), present appropriate lead compounds to develop more potent
have been studied as a primary screening test [3]. In our search agents with anti-tumor promoting activity for clinical use. In
for naturally occurring cancer chemopreventive agents, the our further search for naturally occurring cancer chemopreven-
stem bark of Picea jezoensis Carr. var. jezoensis Japanese name: tive agents, a detailed investigation of this stem bark was carried
Ezomatsu) Pinaceae) was selected for detailed investigation. In out.
previous papers, we had reported a considerable number of ser-
ratane-type triterpenenoids from this stem bark [4], [5], [6], [7].
Affiliation
1 Osaka University of Pharmaceutical Sciences, Osaka, Japan
2 Kyoto Prefectural University of Medicine, Kyoto, Japan
Correspondence
Dr. Reiko Tanaka ´ Department of Medicinal Chemistry ´ 4±20±1 Nasahara ´ Takatsuki ´ Osaka 569±1094 ´ Japan ´
Phone and Fax: +81-726-90-1084 ´ E-mail: tanakar@gly.oups.ac.jp
Funding
This study was supported by a Grant-in-Aid for High Technology from the Ministry of Education, Science,
Sports and Culture, Japan
Received May 15, 2003 ´ Accepted August 10, 2003
Bibliography
Planta Med 2003; 69: 1041±1047 ´ ꢀ Georg Thieme Verlag Stuttgart ´ New York ´ ISSN 0032-0943

fractions 75 ± 83, 84 ± 90 and 91 ± 105 each 2 L). Further elution
with EtOAc gave a residue J 83.1 g) from fractions 106 ± 111
each 2 L). Recrystallization of residue E gave 3b-methoxyser-
rat-14-en-21b-ol 4) 33.8 g), and the filtrate 17.8 g) of com-
pound 4 was subjected to column chromatography on silica gel
600 g) using n-hexane-CHCl3 1:1) and afforded a crystalline
solid fractions 48 ± 52, 2.032 g) each 300 mL), which was sub-
jected to MPLC with n-hexane-EtOAc 10:1) followed by recrys-
tallization from MeOH-CHCl3 to give compound 1 1.518 g). Re-
peated column chromatography of residue G on silica gel 2 kg)
eluting with CHCl3-EtOAc 10:1) afforded a crystalline solid
fractions 38 ± 55, 1.220 g) each 500 mL), which was subjected
to MPLC using n-hexane-EtOAc 10:1) to give crude compound 2
fractions 77 ± 83, 312 mg) and recrystallized from MeOH-CHCl3
to give compound 2 246 mg).
3b-Methoxyserrat-13-en-21b-o1 1): Colorless prisms; m.p. 257
± 2598C from MeOH-CHCl3); [a]D23.5: + 59.1 c 0.28, CHCl3); HR-
El-MS: m/z = 456.3965 [M]+ calcd. for C31H52O2 :456.3965); lR
KBr) nmax = 3497 OH), 2960, 2850, 1457, 1383, 1224, 1201,
1
1181, 1098, 991, 963 cm±1; H- and l3C-NMR, see Table 1. El-MS:
m/z rel. int.) = 456 100) [M+], 441 66), 438 42), 423 41), 409
7), 391 14), 285 8), 255 17), 221 73), 203 57), 189 69), 135
42), 121 31).
Material and Methods
3b-Methoxyserrat-13-en-21b-yl acetate la): Compound 1 25
General experimental procedures
mg) was acetylated as usual Ac2O-pyridine, 1:1, 2 mL) to yield
Melting points were measured with a Yanagimoto micro-melting a crystalline solid. Purification by PTLC CHCl3-MeOH, 30:1) af-
point apparatus without correction. Optical rotations were de- forded a corresponding monoacetate 1a), m.p. 206 ± 2098C
termined with a JASCO DIP-IOOO digital polarimeter. IR spectra from MeOH-CHCl3); [a]D23.5: ±9.7 c 0.11, CHCl3); HR-El-MS: m/z
were recorded using a Perkin-Elmer 1720X FTIR spectrophoto-
=
498.4072 [M+] calcd. for C33H54O3 :498.4070); lR KBr):
meter. 1H- and 13C-NMR spectra were obtained on a Varian INO- nmax = 1727, 1246 cm±1 OAc); 1H- and 13C-NMR: see Table 1. El-
VA 500 spectrometer with standard pulse sequences, operating MS: m/z rel. int.): = 498 42) [M]+, 438 36),423 75), 221 52),
at 500 MHz and 125 MHz, respectively. CDCl3 was used as the 203 100), 189 90), 121 35).
1042
solvent and Me4Si TMS) as the internal standard. EI-MS were re-
corded on a Hitachi 4000H double-focusing mass spectrometer Conversion of 3b-methoxyserrat-13-en-21b-ol 1) to 3b-
70 eV). Column chromatography was carried out over silica gel methoxyserrat-14-en-21b-ol 4): Compound 4 140.1 mg) was
70 ± 230 mesh, Merck) and medium-pressure liquid chromato- acetylated as usual to give a corresponding mono-acetate 4a)
graphy MPLC) was carried out with silica gel 230 ± 400 mesh, 147.0 mg). A solution of compound 4a 100 mg) in glacial HOAc
Merck). TLC and preparative TLC were carried out on Merck silica 30 mL) was added H2SO4 3.0 mL) under ice cooling, and the
gel F254 plates.
mixture was kept at room temperature for 16 h. Then, the mix-
ture was poured into ice/water, and the resulting precipitate
was extracted with CHCl3 100 mL”3). The CHCl3 extract was
Plant material
Cuticles of P. jezoensis Sieb. et Zucc.) Carr. var. jezoensis were col- neutralized with 5% NaOH solution, washed with H2O, and dried
lected at ca. 1000 m in the mountains of Hidaka town, Saryu dis- over Na2SO4. Evaporation of CHCl3 yielded a crystalline solid
trict, Hokkaido, Japan, in June 2001. A voucher specimen PJJ-01- 94.8 mg) which was subjected to MPLC 230 ± 400 mesh, silica
01) is deposited at the Herbarium of the Laboratory of Medicinal gel, Merck) using n-hexane:EtOAc 20:1) to give 3b-methoxyser-
Chemistry, Osaka University of Pharmaceutical Sciences.
rat-13-en-21b-yl acetate 4b) fractions 13 ± 25, 85.5 mg). Com-
pound 4b 70 mg) was hydrolyzed with KOH 200 mg) in MeOH
30 mL) at 808C in 8 h. Work-up at usual gave a residue 64.2 mg)
Extraction and isolation
The freshly chopped cuticles 12 kg) of P. jezoensis var. jezoensis which was purified by MPLC using n-hexane:EtOAc 10:1) to af-
were extracted with CHCl3 20 L) employing an automatic perco- ford 3b-methoxyserrat-13-en-21b-ol 4c) fractions 37 ± 47, 62.5
lator for 7 days at 508C. The CHCl3 solution was evaporated un- mg), and 4c was identified by direct comparison with an authen-
der reduced pressure and the resulting dark green residue tic sample of 1.
685.0 g) was subjected to silica gel 13 kg) column chromato-
graphy. Elution of the column with CHCl3 afforded residues A 13b,14b-Epoxy-3b-methoxyserratan-21b-ol 2): Colorless prisms;
39.3 g), B 67.2 g), C 28.9 g), D 71.0 g), E 51.6 g), and F 23.8 m.p. 264 ± 2678C from MeOH-CHCl3); [a]D23.5: + 4.7 c 0.11,
g), from fractions 1 ± 13, 14 ± 26, 27 ± 30, 31 ± 40, 41 ± 60 and CHCl3); HR-El-MS: m/z = 472.3913 [M+] calcd. for
61 ± 74 each 2 L). Elution was continued with CHCl3-EtOAc
C31H52O3 :472.3914); lR KBr): nmax = 3500 OH), 2963, 2937,
5:1) to give residues G 60.1 g), H 52.9 g) and I 12.6 g) from 2874, 1460, 1385 and 1365 !gem. dimethyl), 1221, 1182, 1103,
Tanaka R et al. Two New Anti-Tumor¼ Planta Med 2003; 69: 1041±1047

Table 1 NMR Data for compound 1 ꢀ125 and 500 MHz, CDCl3)a,b
1
1c
Position
dC
dH
NOESY
dC
dH
La
1b
38,2 t
0.85 m
38.2 t
0.86m
1.81 m
11b
1.80 m
2a
22.3 t
1.79 m
22.3 t
1.80 m
2b
1.41 m
1.40 m
3a
88.6 d
38.9 s
56.2 d
18.8 t
2.63 dd ꢀ11.9, 4.1)
5a, 23
88.5 d
37.0 s
56.2 d
18.8 t
2.62 dd ꢀ11.9, 3,9)
4
5a
0.78m
9 a
0.78m
6a
1.49 m
23
1.48 m
6b
1.44 m
1.43 m
7a
44.9 t
1.21 ddd ꢀ13,3, 13.3, 4.3)
1.41 m
5a
44.9 t
1.23 ddd ꢀ13.3, 13.3, 4.1)
1.42 m
7b
s
835.8
9a
35.7 s
64.8d
38.2 s
21.5 t
65.0 d
38.3 s
21.6 t
0.91 d ꢀ11.9)
la, 5a
1b, 9a
0.92 dd ꢀ11.9, 1.8)
10
11a
11b
12a
12b
13
1.64 m
1.66 m
1.01 m
1.03 m
28.2 t
1.73 m
.1 t9a, 27a, 28218.76 m
19a, 19b
2.26 dd ꢀ14.2, 7.8)
2.25 dd ꢀ14.4, 8.0)
143.1 s
129.6 s
36.1 t
142.9 s
129.6 s
36.2 t
14
15a
15b
16a
16b
17b
1838
19a
19b
20a
20b
21a
22
1.90 dd ꢀ17.8, 5.5)
2.07 m
1.90 m
2.07 m
1.52 m
1.34 m
1.52 m
26
19.2 t
45.3 d
29.5 t
25.8t
1.36 m
28, 29
30
19.0 t
1.52 m
1.52 m
46.4 d
38.3 s
30.1 t
.4 s
1.61 m
1.64 m
1.46 m
1.61 m
1043
1.98dddd ꢀ12.6, 12.6, 5.5, 2.3)
1.64 m
28, 29
29, 30
23.4 t
1.91 m
1.68m
75.8d
37.7 s
28.0 q
16.1 q
16.3 q
19.4 q
52.8t
3.44 t ꢀ2.8)
77.8d
36.8s
4,68t ꢀ2.7)
23
0.94 s
0.74 s
6a
28.0 q
16.1 q
16.3 q
19.2 q
52.8t
0.96 s
24
23
0.74 s
25
0.76 s
2b, 11b, 26
7b, 11b, 15b, 17b
7 a
0.79 s
26
0.84 s
0.85 s
27a
27b
2819.2 q9 s
29
2.16 d ꢀ14.2)
1.35 d ꢀ14.2)
12
2.16 d ꢀ14.2)
1.37 d ꢀ14.2)
0.90 s
0.8
22.2 q
28.0 q
57.5 q
a, 16a, 19a, 20a
16a, 16b
19.2 q
27.6 q
21.8q
0.85 s
0.88 s
30
0.97 s
16b, 29
0.70 s
OMe
OAc
3.35 s
57.5 q
21.4 q
171.0 s
3.35 s
2.06s
a Assignments confirmed by H/H COSY, NOESY, HMQC and HMBC spectra.
bJ values are given in Hz.
1078, 1070, 995, 976 cm±1; 1H- and l3C-NMR: see Table 2. El-MS: 205 ± 2078C from MeOH-CHCl3); [a]D23.5: ±18.9 c 0.11, CHCI3);
m/z rel. int.) = 472 1) [M+], 454 1), 421 3), 367 1), 319 1), 287 HR-El-MS: m/z = 514.4015 [M+] calcd for C33H54O4 :514.4020);
4), 269 8), 221 6), 189 11) 154 19), 136 100), 121 45).
lR KBr): nmax = 1728, 1244 cm±1 OAc); 1H- and l3C-NMR: see
Table 2; El-MS: m/z rel. int.) = 514 6) [M+], 496 10), 454 22),
13b,14b-Epoxy-3b-methoxyserratan-21b-yl acetate 2a): Com- 436 9), 421 68), 221 6), 136 100), 121 47).
pound 2 15 mg) was acetylated as usual Ac2O-pyridine, 1:1, 2
mL) to yield a crystalline solid, which was recrystallized from Preparation of 13b,14b-epoxy-3b-methoxyserratan-21b-ol 2)
MeOH-CHCl3 to give a corresponding monoacetate 2a), m.p. from 3b-methoxyserrat-13-en-21b-ol 4c): A solution of m-CPBA
Tanaka R et al. Two New Anti-Tumor¼ Planta Med 2003; 69: 1041±1047

Table 2 NMR data for compound 2 ꢀ125 and 500 MHz, CDCl3)a,b
2
2a
Position
dC
dH
HMBC ꢀC ® H)
5a, 9a, 25
COSY
dC
dH
la
38.3 t
22.3 t
0.84m
38.3 t
22.3 t
0.84 m
1b
1.86 dt ꢀ16.2, 3.2)
1.78m
la,2b
1.87 m
2a
l a, 2a, 2b
1a, 1b, 2a, 3a
1.80 m
2b
1.42 m
1.42 m
3a
88.5 d
38.9 s
56.1 d
18.1 t
2.61 dd ꢀ11.9, 4.3)
23, 24, OMe
3a, 5a, 23, 24
23, 24, 25
5a
88.5 d
38.9 s
56.1 d
18.1 t
2.62 dd ꢀ11.7, 4.3)
4
5a
0.67 dd ꢀ10.1, 3.9)
1.44 m
6a, 6b
5a, 7a
5a, 7a
0.67 dd ꢀ10.5, 3.0)
1.45 m
6a
6b
1.44 m
1.45 m
7a
45.0 t
1.43 m
27b
44.9 t
1.44 m
7b
1.14 td ꢀ13.3, 5.3)
6a, 6b, 7a
11b
1.15 td ꢀ13.0, 3.0)
837.0 s
9a
7
b, 9a, 27a
12b, 25, 26
5a, 9a, 25
9a
36.9 s
61.4 d
38.3 s
19.4 t
61.7 d
38.3 s
19.4 t
0.55 d ꢀ10.5)
0.57 d ꢀ10.3)
10
11a
11b
12a
12b
13
1.50 m
1.32 m
1.56 m
11b
1.51 m
9a, 11a, 12a
11a, 11b, 12b
11a, 12a
1.32 m
27.9 t
9a
27.9 t
1.55 m
2.14 ddd ꢀ14.6, 7.6, 1.8)
2.12 ddd ꢀ15.1, 8.0, 2.5)
70.0s
63.9 s
31.2 t
12b, 15a, 27a, 27b, 2870.0 s
12b, 15a, 27a, 27b
17b
14
63.7 s
31.1 t
15a
15b
16a
16b
17b
1837.8
19a
19b
20a
20b
21a
22
1.91 m
15b, 16a, 16b
15a, 16a, 16b
15a, 15b, 17b
15 a, 15b, 17b
16a, 16b
1.74 m
17.3 t
36.5 d
27.4 t
25.3 t
1.28m
15 b, 17a
17.1 t
37.6 d
28.0 t
22.9 t
1.26 m
1.26 m
1.88 m
1.28m
1.81 dd ꢀ9.6, 5.0)
19
28,29,30
b, 2837.3 s
2819
s
1.54 m
b, 20a
1.55 m
1.68m
19 a, 20b
1.55 m
1044
1.65 m
19a, 19b, 20b
1.85 m
1.89 m
1.70 ddd ꢀ14.9, 6.9, 3.4)
4.62 t ꢀ2.7)
75.5 d
37.4 s
28.1 q
16.1 q
15.7 q
21.9 q
53.8t
3.36 t ꢀ2.7)
29, 30
77.4 d
36.5 s
28.1 q
16.1 q
15.7 q
22.0 q
53.7 t
29, 30
23
0.95 s
0.73 s
3a, 24
0.95 s
24
1a, 3a, 5a, 23
la, 9a
0.77 s
25
0.77 s
0.74 s
26
1.02 s
7a, 7b, 9a, 27a
9 a, 26
1.01 s
27a
27b
2816.7 q
29
1.62 d ꢀ15.1)
1.42 d ꢀ15.1)
19
27b
27a
1.63 d ꢀ15.1)
1.42 d ꢀ15.1)
0.81 s
1.00 s
b
30
16.7 q
21.7 q
28.0 q
57.5 q
21.4 q
171.0s
22.0 q
28.5 q
57.5 q
0.83 s
0.95 s
30
0.91 s
29
1.03 s
OMe
OAc
3.35 s
3a
3.35 s
2.06 s
a Assignments confirmed by decoupling, H/H COSY, NOESY, HMQC, and HMBC spectra.
b J values are given in Hz.
100 mg) in dry CHCl3 5 mL) was gradually added over a solution Inhibition of EBV -EA activation test
of 4c 100 mg) in dry CHCl3 10 mL) under stirring at room tem- EBV-EA positive serum from a patient with nasopharyngeal car-
perature and allowed to stand for 24 h. Work-up as usual yielded cinoma NPC) was a gift from Dr. Y. Zaizen, Department of Bio-
a residue 97 mg), which was subjected to MPLC with n-hexane- chemistry, Oita Medicinal University. The EBV genome-carrying
EtOAc 10:1) to give compounds 4e 4 mg) fractions 7 ± 8) and lymphoblastoid cells Raji cells derived from Burkitts lympho-
4d 89 mg) fractions 25 ± 40). The synthetic compounds 4d and ma) were cultured in 10% fetal bovine serum FBS) in RPMI-
4e were identified as natural compounds 3 and 2, respectively, 1640 medium Nissui). Spontaneous activation of EBV-EA in our
by direct comparison.
sub-line Raji cells was less than 0.1%. The inhibition of EBV-EA
Tanaka R et al. Two New Anti-Tumor¼ Planta Med 2003; 69: 1041±1047

activation was assayed using Raji cells virus non-producer type) 4), which is the most abundant triterpenoid of this stem bark.
as described previously [9]. The indicator cells Raji cells, 1”106/ Isomerization of 3b-methoxyserrat-14-en-21b-yl acetate 4a)
mL) were incubated at 378C for 48 h in 1 mL of a medium con- with conc. H2SO4/HOAc furnished the corresponding 3b-methox-
taining n-butyric acid 4 mmol), TPA [32 pmol = 20 ng in dime- yserrat-13-en-21b-yl acetate 4b) in almost quantitative yield,
thyl sulfoxide DMSO)], as inducer and various amounts of test and then hydrolysis gave the corresponding alcohol 4c. The
compound in 5 mL DMSO. Smears were made from the cell sus- above synthetic 4b and 4c were identical with la and 1 in all re-
pension, and the activated cells that were stained by EBV-EA po- spects. To the best of our knowledge, other serrat-13-ene deriva-
sitive serum from NPC patients were detected by an indirect im- tives isolated so far are 3a-methoxyserrat-13-en-21b-ol from the
munofluorescence technique [2]. In each assay, at least 500 cells bark of Picea sitchensis Sitka sparuce) [10] and 21a-methoxyser-
were counted, the number of stained cells were counted, and the rat-13-en-3-one from the stem bark of Picea jezoensis Carr. var.
number of stained cells positive cells) present recorded. Tripli- hondoensis [11] and P. jezoensis Carr. var. jezoensis [7].
cate assays were performed for each compound. The average
EBV-EA induction of the test compounds was expressed as a re- Compound 2 was assigned as C31H52O3 M+; m/z = 472.3913) by
lative ratio to the control experiment 100%) which was carried HR-EI-MS. The IR spectrum showed a hydroxy 1: nmax = 3500
out only with n-butyric acid 4 mmol) plus TPA 32 pmol). EBV- cm±1) absorption. The 1H- and 13C-NMR spectra of 1 Table 2) ex-
EA induction was ordinarily around 35%. The viability of treated hibited seven tertiary methyls, eleven methylenes, three me-
Raji cells was assayed by Trypan blue staining methods.
thines, five quaternary carbons, a secondary hydroxy group
[dH = 3.36 1H, t); dC = 75.5 d)], a secondary methoxy group
[dH = 2.61 dd), 3.35 3H, s); dC = 57.5 q), 88.5 d)] and two sp3
Two-stage mouse-skin carcinogenesis test
Specific pathogen-free female ICR mice 6 weeks old, body quaternary carbons combined with one oxygen atom [dC = 63.9
weight approx. 30 g) were obtained from Japan SLC Inc., Shizuok, s), 70.0 s)]. Acetylation of 2 furnished a monoacetate 2a),
Japan, and the animals were housed, 5 per polycarbonate cage, in
C
33H54O4 M+; m/z = 514.4015), in which the hydroxymethine
a temperature-controlled room at 24  28C and given food and proton signal was shifted to dH = 4.69 t). The DEPT and HMQC
water ad libitum throughout the experiment. Animals were spectra of 2 showed the same carbon composition as 13a,14a-
divided into three experimental groups containing 15 mice epoxy-3b-methoxyserratan-21b-ol 3) [7]. The difference be-
each. The back 2”8 cm2) of each mouse was shaved with surgi- tween 2 and 3 is assumed to come from the bonding of the epoxy
cal clippers, and the mice were topically treated with DMBA ring at C-13 and C-14, considering that these carbons in 2 ap-
100 mg, 390 nmol) in acetone 0.1 mL) as an initiating treatment. peared at d = 70.0 s, C-13) and 63.9 s, C-14) whereas they ap-
One week after the initiation, papilloma formation was promo- peared at d = 72.9 s, C-13) and 65.7 s, C-14) in 3, and H-12a
ted twice weekly by the application of TPA 1 mg, 1.7 nmol) in and H-12b appeared at d = 2.52 dd) and 0.98 m) in 2 whereas
acetone 0.1 mL) to the skin. Group 1 received the TPA treatment they appeared at d = 1.56 m) and 2.14 ddd) in 3 [7]. The HMBC
alone, and groups II and III received a topical application of com- spectrum of 2 Table 2) supported this assumption, accordingly,
pounds 1 and 2 85 nmol), in acetone 0.1 mL), respectively, 1 h the structure of 2 was suggested as the epoxy epimer of 3. In the
before the TP A treatment. The incidence and numbers of papil- NOESY spectrum of 2 Fig.1), significant NOEs were observed be-
1045
lomas were monitored weekly for 20 weeks.
tween H-12a dH = 1.56) with H-9a and Me-28, between H-27a
dH = 1.62) with H-7a and H-9a, between H-3a dH = 2.61) with
H-5a and Me-23, and between H-21a dH = 3.36) with Me-29
and Me-30. However, the cross peaks of H-12a, H-15a and H-
17b which decide the conformation of the C and D rings of 2
were not observed clearly. The configuration of an epoxy ring
Results and Discussion
Structure determination of 1 and 2
The molecular formula of compound 1 was assigned as C31H52O2 and the conformation of the C, D rings were determined by em-
M+; m/z = 456.3965) by HR-EI-MS. The IR spectrum showed a ploying the NOE difference experiment. Upon selective irradia-
hydroxy nmax = 3497 cm±1) absorption. The 1H- and 13C-NMR
spectra of 1 Table 1) exhibited seven tertiary methyls, eleven
methylenes, three methines, five quaternary carbons, a second-
ary hydroxy group [dH = 3.44 1H, t); dC = 75.8 d)], a secondary
methoxy group [dH = 2.63 dd), 3.35 3H, s); dC = 57.5 q), 88.6
d)] and a tetrasubstituted double bond [dC = 129.6 s), 143.1
s)]. Acetylation of 1 gave a monoacetate la), C33H54O3 M+; m/z
= 498.4072), in which the hydroxymethine proton signal was
shifted to d = 4.68 t). In the 1H-1H COSY spectra of 1, H-12a
dH = 1.73) was related only to H-11a, H-11b and H-12b, and H-
12b dH = 2.26) was related only to H-11a, H-11b and H-12a. In
the NOESY spectrum of 1, significant NOEs were observed be-
tween H-3a with H-5a and Me-23, and between H-21a with
Me-29 and Me-30, which suggested that the methoxy group
had C-3b and the hydroxy group had C-21b configurations. All
these data suggested that the structure of 1 should be 3b-meth-
oxyserrat-13-en-21b-ol. In order to confirm this structure, we
Fig. 1 NOESY correlations of 2.
tried the synthesis of 1 from 3b-methoxyserrat-14-en-21b-ol
Tanaka R et al. Two New Anti-Tumor¼ Planta Med 2003; 69: 1041±1047

tion of the signal of H-12b dH = 2.14), 1.71%, 4.53%, 3.45% NOEs
were observed for the signals of H-16a, H-16b and Me-28. Irra-
diation of the signal for H-15a dH = 1.91) showed 1.91%, 3.85%,
1.97%, 1.93% and 0.87% NOEs for H-12a, H-27a, H-27b, Me-28
and Me-29. Irradiation of the signal for H-17b dH = 1.81)
showed 1.40%, 0.96%, 2.56% and 3.95% NOEs for H-12a, H-19b,
H-27b and Me-30. Thus, the stereostructure of 2 was established
as shown in Fig.1.
The El-MS of 2 showed the fragment ion peaks due to cleavage of
the D and C rings at m/z = 287.2366 [C20H31O]+, 221.1902
[C15H25O]+, and 189.1654 [C14H21]+, 154.1358 [C10H18O]+, 136.1245
[C10H16]+, 121.1021 [C9H13]+, and the fragment pattern was close
to that of 3 [7]. All these data indicated that the structure of 2 as
13b,14b-epoxy-3b-methoxyserratan-21b-ol, which was con-
firmed by synthesis. Oxidation of synthetic 3b-methoxyserrat-
13-en-21b-ol 4c) 100 mg) with m-CPBA gave 13a,14a-epoxy-
3b-methoxyserratan-21b-ol 4d) 89 mg) as a major product
and the 13b,14b-epoxy epimer 4e) 6 mg) of 4d as a minor prod-
uct, which was identical in all respects with 2. It is interesting to
note that Picea jezoensis Carr. var. jezoensis produces both
13a,14a-epoxy-3b-methoxyserratan-21b-ol and its 13b,14b-
epoxy epimer in the plant organ.
Fig. 2 Inhibition of TP A-induced tumor promotion by multiple appli-
cation of 3b-methoxyserrat-13-en-21b-ol ꢀ1) and 13b,14b- epoxy-3b-
methoxyserratan-21b-ol ꢀ2). All mice were initiated with DMBA ꢀ390
nmol) and promoted with 1.7 nmol of TP A, given twice weekly start-
ing 1 week after initiation. Percentage of mice bearing papillomas. l
control ꢀTPA alone); n TPA + 85 nmol of oleanolic acid; ~ TPA + 85
nmol of 1; & TPA + 85 nmol of 2.
In vitro EBV-EA activation
The primary screening test was carried out utilizing a short-term
in vitro assay on EBV-EA activation. Table 3 lists inhibitory effects
of compounds 1, la, 2 and 2a on the EBV-EA activation induced by In vivo two-stage carcinogenesis test on mouse skin
TPA and the associated viability of Raji cells. The inhibitory ef- papillomas initiated by DMBA
fects of all compounds 1, 1a, 2, 2a) were stronger at every con- The results of the in vitro experiments and structure-activity
centration than that of oleanolic acid [12] known as a represent- relationship of serratane-type triterpenoids prompted us to ex-
ative anti-tumor promoting agent. All compounds exhibited amine the effects of compounds 1 and 2 on the in vivo two-stage
dose-dependent inhibitory activities, and the viability percenta- carcinogenesis bioassay on mouse skin using DMBA as an initia-
ges of Raji cells treated with the test compounds 1, la, 2, 2a) tor and TPA as a promotor. The incidence %) of papilloma-bear-
were 70% at the highest concentration of 1000 mol ratio/TPA, ing mice and the average numbers of papillomas per mouse are
suggesting that the cytotoxicities of all compounds were rather presented in Figs. 2 and 3, respectively. No significant toxic ef-
moderate against in vitro cell lines Table 3). Among them, com- fects, such as inflammation and lesional damages, on the areas
pound 1 exhibited a stronger inhibitory activity than others la, of mouse skin topically treated with the test compounds were
2, 2a). The relative ratios of compound 1 with respect to TPA observed at the end of treatment except for the formation of pa-
100%) were 0, 20.3, 74.1 and 89.7% at the concentrations of pillomas, and also the body weight gains were not influenced
1000, 500, 100 and 10 mol ratio/TPA, respectively Table 3); during the treatment. As demonstrated in Fig. 2, the percentage
meaning 100, 79.7, 25.9 and 10.3% inhibition of the EBV-EA acti- of papilloma bearers in the control group DMBA and TPA only)
1046
vation by TPA, respectively.
increased rapidly from week 6 and reached 100% after week 9,
whereas the treatment with compound 1 85 nmol) along with
Table 3 Relative ratioa of EBV-EA activation with respect to positive control ꢀ100%) in the presence of compounds 1, 1a, 2 and 2a
Compounds
EBV-EA positive cells ꢀ% viability)
Compound concentration ꢀmol ratio/32 pmol TPA)
IC50
1000
500
100
10
ꢀmol ratio/32 pmol TPA)
1
0 ꢀ70)b
0 ꢀ70)
20.3
27.5
25.7
27.0
30.0
74.1
78.0
74.9
78.4
80.0
89.7
271
290
288
291
360
1a
91.5
92.6
94.7
2
0 ꢀ70)
2a
0 ꢀ70)
Oleanolic acid
12.7 ꢀ70)
100
a Values represent percentages relative to the positive control value ꢀ100%).
b Values in parentheses are the viability percentages of Raji cells.
Tanaka R et al. Two New Anti-Tumor¼ Planta Med 2003; 69: 1041±1047

Acknowledgements
The authors are indebted to Mr. Kiyoshi Matsubara Green Ace
co. Ltd, Hidaka, Hokkaido) for collection of the plant materials.
Our thanks are also due to Mrs. Mihoyo Fujitake of this Universi-
ty for MS measurements.
References
1
Berenblum I. The mechanism of carcinogenesis. Cancer Res 1941; 1:
807±14
2 Murakami A, Ohigashi H, Koshimizu K. Anti-tumor promotion with
food phytochemicals: A strategy for cancer chemoprevention. Biosci
Biotech Biochem 1996; 60: 1±8
3 Tokuda H, Ohigashi H, Koshimizu K, Ito Y. Inhibitory effects of ursolic
acid and oleanolic acid on skin tumor promotion 12-O-tetradecanoyl-
phorbol 13-acetate. Cancer Letters 1986; 33: 279±85
4 Tanaka R, Senba H, Minematsu T, Muraoka O, Matsunaga S. 21a-Hy-
droxy-3b-methoxyserrat-14-en-30-al and other triterpenoids from
the cuticle of Picea jezoensis. Phytochemistry 1995; 38: 1467±71
5 Tanaka R, Ohmori K, Minoura K, Matsunaga S. Two new epoxyserra-
tanes from the cuticle of Picea jezoensis. J Nat Prod 1996; 59: 237±41
6 Tanaka R, Tsujimoto K, In Y, Matsunaga S. New methoxytriterpene
dione from the cuticle of Picea jezoensis var. jezoensis. J Nat Prod
1997; 60: 319±22
Fig. 3 Inhibition of TP A-induced tumor promotion by multiple appli-
cation of 3b-methoxyserrat-13-en-21b-ol ꢀ1) and 13b, 14b-epoxy-3b-
methoxyserratan-21b-ol ꢀ2). All mice were initiated with DMBA ꢀ390
nmol) and promoted with 1.7 nmol of TP A, given twice weekly start-
ing 1 week after initiation. Average number of papillomas per mouse.
l control ꢀTPA alone); n TPA + 85 nmol of oleanolic acid; ~ TPA + 85
nmol of 1; & TPA + 85 nmol of 2.
7 Tanaka R, Tsujimoto K, In Y, Ishida T, Matsunaga S, Terada Y. Structure
and stereochemistry of epoxyserratanes from the cuticle of
Piceajezoensis var.jezoensis. J Nat Prod 2001; 64: 1044±7
8 Tanaka R, Tsujimoto K, In Y, Ishida T, Matsunaga S, Tokuda H, Muraoka
O. Jezananals A and B: two novel skeletal triterpene aldehydes from
the stem bark of Picea jezoensis var. jezoensis. Tetrahedron 2002; 58:
2505±12
DMBA/TPA inhibited the formation of papillomas until week 8
and reduced the percentage of papilloma-bearing mice to ap-
proximately 47% during weeks 12 and 13 and thereafter 80%
over the period of week 20. On the other hand, the treatment
with compound 2 85 nmo1) along with DMBA/TPA inhibited
the formation of papillomas until week 7 and reduced the per-
centage of papilloma-bearing mice to approximately 47% during
weeks 14 and 15 and thereafter 80% over the period of week 20.
As shown in Fig. 3, in the control group, the number of papillo-
mas formed per mouse increased rapidly after week 6 and
reached 10.0 papillomas/mouse at week 20. On the other hand,
the mice treated with compound 1 or compound 2 bore 5.0 or
5.8 papillomas over the period of week 20, although the antitu-
mor-promoting activities of compounds 1 and 2 seem to be
weaker than that of 3. It is interesting to note that the 13a,14a-
epoxyserratane framework is important to enhance the activity
expression.
9 Tanaka R, Ohtsu H, Iwamoto M, Minami T, Tokuda H, Nishino H, Mat-
sunaga S, Yoshitake A. Cancer chemopreventive agents, labdane diter-
penoids from the stem bark of Thuja standishii Gord.) Carr. Cancer
Letters 2000; 161: 165±70
10
Kutney JP, Rogers IH, Rowe JW. The neutral triterpenes of the bark of
Picea sitchensis Sitka spruce). Tetrahedron 1969; 25: 3731±51
Tanaka R, Mun C, Usami Y, Matsunaga S. 3±0xo-serratene triterpe-
1047
11
noids from the stem bark of Picea jezoensis Carr. hondoensis. Phyto-
chemistry 1994; 35: 1517±22
Konoshima T, Takasaki M, Kozuka M, Tokuda H. Studies on inhibitors
12
on skin-tumor promotion. 1. Inhibitory effects of triterpenes from
Euphorbia polyandra on Epstein-Barr virus activation. J Nat Prod
1987; 50: 1167±70
Tanaka R et al. Two New Anti-Tumor¼ Planta Med 2003; 69: 1041±1047