Preparation and cytotoxic effect of ceanothic acid derivatives

Article abstract of DOI:10.1021/np9800856

Six ceanothane and 1-norceanothane derivatives (1, 2, 8-11) were prepared from ceanothic acid dibenzyl ester. These ring-A homologues of betulinic acid exhibited cytotoxic effects. Among these, 1-decarboxy-3-oxo- ceanothic acid (2) was found to be the most cytotoxic against OVCAR-3 and HeLa cancer cell lines, with an IC₅₀ of 2.8 and 6.6 μg/mL, respectively, and an IC₅₀ of 11.3 μg/mL against normal cell line FS-5.

Full text of DOI:10.1021/np9800856

J . Nat. Prod. 1998, 61, 1343-1347
1343
P r ep a r a tion a n d Cytotoxic Effect of Cea n oth ic Acid Der iva tives
Shoei-Sheng Lee,*,† Wen-Chuan Chen,† Chih-Fen Huang,† and Yeu Su‡
School of Pharmacy, College of Medicine, National Taiwan University, and Institute of Pharmacology, National Yang-Ming
University, Taipei, Taiwan, Republic of China
Received March 11, 1998
Six ceanothane and 1-norceanothane derivatives (1, 2, 8-11) were prepared from ceanothic acid dibenzyl
ester. These ring-A homologues of betulinic acid exhibited cytotoxic effects. Among these, 1-decarboxy-
3-oxo-ceanothic acid (2) was found to be the most cytotoxic against OVCAR-3 and HeLa cancer cell lines,
with an IC50 of 2.8 and 6.6 µg/mL, respectively, and an IC50 of 11.3 µg/mL against normal cell line FS-5.
Sch em e 1a
The lupane-type triterpene, betulinic acid, is widely
distributed in plants. This pentacyclic compound has
proved to be selectively cytotoxic to human melanoma1 and
is active against the HIV virus.2 Structure-activity rela-
tionship studies of the latter activity showed that the
3-hydroxyl, 17-carboxylic acid, and 19-isopropyl (or isopro-
pylidene) moieties are essential pharmacophores.2 A recent
study revealed that Paliurus ramossisimus (Rhamnaceae)
is rich in ceanothic acid (3), a ring-A homologue of betulinic
acid.3 As a result, we decided to investigate the biological
activity of analogues of this acid. Results of this work are
described in this report.
Resu lts a n d Discu ssion
1-Decarboxy-ceanothic acid (1) and 1-decarboxy-3-oxo-
ceanothic acid (2) were prepared as shown in Scheme 1.
Benzylation of ceanothic acid (3), by reacting with benzyl
chloride in alkaline conditions, gave ceanothic acid dibenzyl
ester (4), whose 1H NMR spectrum showed two AB systems
for benzylic proton signals, one at δ 4.95 (d) and 5.08 (d)
(J ) 11.9 Hz) and the other at δ 5.07 (d) and 5.12 (d) (J AB
) 12.3 Hz), in addition to the signals of 3.4 Oxidation of 4
with pyridinium chlorochromate (PCC)5 yielded 3-oxo-
ceanothic acid dibenzyl ester (5), whose 1H NMR spectrum
a
Key: (a) BnCl, K2CO3, MeOH, reflux; (b) PCC, CH2Cl2, room temper-
showed the absence of an H-3 signal and a downfield shift
ature; (c) KOH, EtOH, reflux; (d) NaBH4, EtOH-dioxane (1:1), room
temperature; (e) LiI, γ-collidine, reflux.
of H-1 (δ 3.01, s) relative to that in 4 (δ 2.60, s). Treatment
of 5 under alkaline conditions yielded 6, which had a
molecular formula of C36H50O3 based on HREIMS. The 1H
NMR spectrum of 6 exhibited only one AB system for
benzylic protons (δ 5.08 and 5.14, J AB ) 12.3 Hz) and one
additional AB system at δ 1.83 and 2.13, J AB ) 15.9 Hz),
as well as the absence of the corresponding H-1 singlet (δ
3.01) of 5. These data established 6 as 1-decarboxy-2-oxo-
ceanothic acid benzyl ester. This reaction apparently
saponified the C-2 ester selectively, accompanied by a six-
membered decarboxylation mechanism for â,γ-unsaturated
carboxylic acid. Reduction of 6 with sodium borohydride
yielded exclusively 1-decarboxy-ceanothic acid benzyl ester
(7), with an IR absorption at 3450 cm-1 (OH) and a
molecular formula of C36H52O3. The orientation of the
hydroxyl group of 7 was confirmed to be the same as 3 by
an NOE experiment in which enhancement of H-3 (δ 3.95,
d, J ) 7.7 Hz) and H-24 (δ 0.88) was observed upon
irradiation of H-23 (δ 0.98). The hindered benzyl group at
C-28 was removed by treating 7 with lithium iodide/γ-
collidine under reflux6 to give 1-decarboxy-ceanothic acid
(1). The IR spectrum of 1 showed absorption at 2500-3100
cm-1 and 1698 cm-1 for the carboxylic acid function. Its
1H NMR spectrum lacked signals of benzylic protons, and
the molecular formula C29H46O3, as deduced from the
HREIMS, supported the assigned structure. Treating 5
directly with lithium iodide/γ-collidine under reflux yielded
1-decarboxy-3-oxo-ceanothic acid (2). The IR spectrum
displayed the absorption at 2600-3200 cm-1 and 1690 cm-1
for a carboxylic function and 1738 cm-1 for a five-membered
1
ketone function. Its H NMR spectrum is similar to that
of 6 except for lack of benzyl proton signals. These data
and the molecular formula C29H44O3, deduced from the
HREIMS, established the structure of 2 as shown.
2,3,28-Trihydroxy-ceanoth-20(30)-ene (8), ceanotha-1(3),-
20(30)-diene-2,28-dioic acid (9), 2-norceanotha-1(3),20(30)-
dien-28-oic acid (10, 1-deformyl-zizyberenalic acid), and
2-norceanothan-28-oic acid (11) were prepared as shown
in Scheme 2. Reduction of 4 with lithium aluminum
hydride gave 8, with a molecular formula C30H50O3 as
1
deduced from the HREIMS. The H NMR spectrum of 8
* To whom correspondence should be addressed. Tel.: (886) 2-2391-6127.
Fax: (886) 2-2391-9098. E-mail: shoeilee@ha.mc.ntu.edu.tw.
† National Taiwan University.
(in C5D5N + D2O) exhibited signals for H-28 as an AX
system at δ 4.09 and 3.65, J AX ) 10.6 Hz, for H-2 as
multiplets (δ 4.14 and 3.81), and for H-3 as a doublet (δ
‡ National Yang-Ming University.
10.1021/np9800856 CCC: $15.00
© 1998 American Chemical Society and American Society of Pharmacognosy
Published on Web 10/10/1998

1344 J ournal of Natural Products, 1998, Vol. 61, No. 11
Lee et al.
Sch em e 2a
NOE experiment, which enhanced H-3 (δ 5.42, d) upon
irradiation at the frequency of H-23 (δ 1.00, s) or H-24 (δ
0.89, s). Another irradiation at the signal of H-25 (δ 0.92)
enhancing H-1 (δ 5.90, d) and H-24 also confirmed the
assignment of H-1 and H-24. HREIMS showed the mo-
lecular ion at m/z 424.3351, consistent with that calculated
for C29H44O2. Catalytic hydrogenation of 10 (Pd-H2)
yielded 2-norceanothan-28-oic acid (11). The 1H NMR
spectrum of 11 lacked the AX system for H-1 and H-3 and
1
two broad singlets for H-30’s in the H NMR spectrum of
10. Instead, two doublet methyls at δ 0.84 (J ) 7.0 Hz)
and 0.74 (J ) 6.8 Hz), assignable for both H-29 and H-30,
appeared in the spectrum of 11.
Compounds 8 and 9 are reduction and dehydration
derivatives of 3, respectively. Their 1H and 13C NMR
assignments, obtained by analysis of NOE, COSYDQF,
HSQC, and HMBC spectra, are shown in Tables 1 and 2.
The cytotoxicity assay of these prepared compounds was
undertaken using the tetrazolium salt (MTT) method.8 The
results indicate that 1-norceanothane derivatives such as
1, 2, and 10 are more potent than the other ceanothane
analogues tested. It appears that the exo-ring C-1 substi-
tution decreases cytotoxicity, especially polar groups such
as 3 and 9 (Table 3). Further study on the IC50 of the most
cytotoxic compounds (Table 4) revealed that 1-norcean-
thane derivatives 2 and 10 were about equally potent, but
10 was about two times less toxic to normal cells than was
2. This observation could be critical to design a lead
compound having selective toxicity. This result indicates
that the double bond between C-20 and C-30 and the 3-OH
group are not essential to the cytotoxic activity of 1-norcean-
othane derivatives. However, the carboxylic group at C-17
is required for this activity, as was reported in the betulinic
acid study.2
a
Key: (f) LiAlH4, THF, reflux; (g) MsCl, pyridine, room temperature;
(h) K2CO3, i-PrOH, reflux; (i) H2, 10% Pd-C, EtOAc, room temperature.
4.69). O-Mesylation of 4 (MsCl-pyridine, 0 °C)7 gave 3-O-
mesylceanothanic acid dibenzyl ester (12), whose H NMR
1
spectrum showed the characteristic signal, δ 2.92 (s, 3H),
for the mesyl group, and a downfield-shifted H-3 singlet
(δ 4.92) relative to that in the reactant (δ 4.16). Treatment
of 12 under alkaline conditions gave the elimination
product ceanotha-1(3),20(30)-diene-2,28-dioic acid dibenzyl
ester (13). The 1H NMR spectrum of 13 revealed an
additional olefinic proton at δ 6.20 (s) for H-3, but lacked
a signal for 3-OMs. O-Debenzylation of the ester 13 using
lithium iodide yielded ceanotha-1(3),20(30)-diene-2,28-dioic
acid (9). The HREIMS showed the molecular ion at m/z
468.3238, consistent with that calculated for the expected
formula C30H44O4. The 1H NMR spectrum of 9 was similar
to that of 13 except for the absence of signals for two benzyl
groups.
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. The physical data
of the prepared compounds were obtained from the following
instruments: Fisher-J ohns melting point apparatus (uncor-
rected); Perkin-Elmer 1760-X infrared FT spectrometer;
Hitachi 150-20 UV (MeOH); J EOL J MX-HX110 (HREIMS) and
Finnigan TSQ-700 (EIMS) mass spectrometers; Bruker AMX-
400 spectrometers using solvent peak as reference standard.
Cea n oth ic Acid Diben zyl Ester (4). The ceanothic-acid
(3) -rich fraction was obtained from the partitioning of the
Treatment of 12 with lithium iodide/γ-collidine under
reflux yielded 2-norceanotha-1(3),20(30)-dien-28-oic acid
(10). The H NMR spectrum of 10 displayed two coupled
olefinic proton signals at δ 5.90 and 5.42, J AX ) 5.7 Hz,
assignable to H-1 and H-3, which were confirmed by an
1
Ta ble 1. 1H NMR and HMBC (J ) 8 Hz) Data for Compounds 8 and 9 (δ/ppm, J in Hz) in C5D5Na
HMBC of 8
HMBC of 9
correlations (C#)
proton
8b
correlations (C#)
9
1
2
3
5
9
13
18
19
23
24
25
26
27
28
2.26 dd (9.6, 4.3)
4.14 m/3.81 m
4.69 d (3.2)
1.55 m
1.87 dd (12.3, 2.5)
1.80 dt (3.2, 12.2)
1.69 t (11.6)
2.60 dt (5.7, 11.0)
1.21 s
1.26 s
1.34 s
1.06 s
1.09 s
2, 3, 4, 5, 10
2, 10
4, 10, 23, 24
8, 11, 25
6.52 s
1.57 m
1, 2, 4, 5, 10
4, 6, 10
2.17 dd (12.7, 3.1)
2.73 dt (3.5, 12.2)
1.65 t (11.3)
3.46 dt (5.5, 10.8)
1.08 s
8, 10, 11, 25, 26
12, 18, 27
13, 16, 17, 28
18, 20, 21, 29, 30
3, 4, 5, 24
13, 16, 17, 28
13, 18, 20, 21, 29, 30
3, 4, 5, 24
3, 4, 5, 23
1, 5, 9, 10
7, 8, 9, 14
8, 13, 14, 15
22
0.93 s
3, 4, 5, 23
1.44 s
1, 5, 9, 10
1.14 s
7, 8, 9, 14
0.97 s
8, 13, 14, 15
4.09 dd (4.2, 10.6)
3.65 dd (4.6, 10.6)
1.75 br s
4.87 br s (Z)
4.73 br s (E)
16
29
30
19, 20, 30
19, 29
1.71 br s
4.83 br s (Z)
4.65 br s (E)
19, 20, 30
19, 20, 29
a
Data overlapped or poorly resolved signals were assigned from COSY-45, HSQC, and HMBC spectra: 8 δ 1.49/1.37 (H-7), 1.40-1.46
(H-11), 1.68/1.22 (H-12), 1.93/1.05 (H-15), 2.42/1.30 (H-16), 2.14/1.49 (H-21), 2.40/1.15 (H-22); 9 δ 1.58 (H-6), 1.40 (H-7), 2.49/1.83 (H-11),
b
1.91/1.21 (H-12), 1.90/1.19 (H-15), 2.61/1.54 (H-16), 2.23/1.50 (H-21), 2.23/1.52 (H-22). δ2-OH 5.93 (br s), δ3-OH 6.21 (d, J ) 3.2 Hz),
δ28-OH 5.80 (dd, J ) 4.6, 4.2 Hz).

Ceanothic Acid Derivatives
J ournal of Natural Products, 1998, Vol. 61, No. 11 1345
Ta ble 2. 13C NMR Data for Compounds 8 and 9 (δ/ppm)a
11.9 Hz) (2-OCH2Ph), 5.07 (d) and 5.12 (d) (J ) 12.3 Hz) (28-
OCH2Ph); EIMS m/z [M]+ 666 (2), 575 (6), 91 (100), 338 (2),
203 (2), 327 (3), 311 (1), 175 (6); HREIMS (20 eV) m/z [M]+
666.4313 (calcd for C44H58O5, 666.4286).
in C5D5N
carbon
8
9
carbon
8
9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
62.07 d
61.77 d
85.33 d
43.27 s
58.59 d
19.21 t
34.55 t
41.95 s
42.64 d
46.40 s
23.99 t
25.50 t
38.05 t
43.57 s
27.87 t
149.03 s
171.08 s
149.57 d
43.46 sb
63.17 d
17.28 t
35.75 t
42.91 s
48.58 d
53.27 s
24.02 t
25.83 t
38.43 d
43.39 sb
30.49 t
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
30.11 t
48.51 s
49.22 d
48.31 d
151.31 s
30.44 t
34.85 t
33.07 q
20.44 q
19.38 q
17.12 q
15.13 q
59.50 t
19.38 q
109.92 q
33.07 t
56.51 s
49.84 d
47.82 d
151.10 s
31.27 t
37.66 t
28.64 q
20.91 q
20.31 q
18.12 q
14.96 q
178.90 s
19.45 q
110.04 t
3-Oxo-cea n oth ic Acid Diben zyl Ester (5). PCC (1.00 g,
4.64 mmol) was added to compound 4 (4.00 g, 6.01 mmol)
dissolved in CH2Cl2 (80 mL). The resultant solution was
stirred at room temperature for 2 h, and then Et2O (30 mL)
was added. Removal of the brown solid residue via filtering
through a Celite pad and evaporation of the filtrate gave a
viscous residue, which was chromatographed over Si gel (160
g, 230-400 mesh) and eluted with toluene to give 5 (3.20 g,
80.2% yield): Rf 0.50 (CHCl3); mp 123-126 °C; UV λmax (log ꢀ)
257.5 (2.58); IR (KBr) νmax 2950 (s), 1745 (s), 1723 (s), 1710
(s), 1640 (m), 1378 (s), 1459 (s), 1320 (s), 1220 (s), 1159 (s),
1
880 (m), 738 (s), 700 (s) cm-1; H NMR (CDCl3) δ 3.01 (1H, s,
H-1), 2.13 (1H, ddd, J ) 3.2, 11.8, 11.8 Hz, H-13), 2.96 (1H,
ddd, J ) 4.5, 10.8, 10.8 Hz, H-19), 1.64 (3H, s, H-29), 4.70 (1H,
br s, H-30Z), 4.60 (1H, br s, H-30E), 5.03 (d) and 5.08 (d) (J )
12.0 Hz) (2-OCH2Ph), 5.07 (d) and 5.12 (d) (J ) 12.3 Hz) (28-
OCH2Ph), 7.33 (10H, m, CH2Ph); EIMS m/z [M]+ 664 (1), 574
(13), 91 (10), 338 (2), 203 (2), 325 (1), 311 (1), 175 (5); HREIMS
(20 eV) m/z [M]+ 664.4077 (calcd for C44H56O5, 664.4130).
1-Deca r boxy-3-oxo-cea n oth ic Acid Ben zyl Ester (6).
Powdered KOH (1.00 g, 17.86 mmol) was added to compound
5 (3.20 g, 4.82 mmol) dissolved in EtOH (60 mL). The
resultant suspension was stirred under reflux for 3 h. After
removal of the organic solvent, the residue was triturated with
H2O (100 mL), and the suspension was adjusted to pH 3 with
concentrated HCl, and partitioned against CHCl3 (100 mL ×
3). The combined CHCl3 layer, after drying over anhydrous
Na2SO4, was evaporated to give a residue that was purified
on a Si gel column (120 g, 230-400 mesh) eluted with toluene
to give 6 (1.80 g, 70.5% yield): Rf 0.35 [Me2CO-toluene
(1:99)]; mp 143-144 °C; UV λmax (log ꢀ) 257.5 (2.14); IR (KBr)
νmax 3060 (w), 2950 (s), 1738 (s), 1640 (m), 1475 (s), 1150 (s),
a
b
Multiplicities were obtained from DEPT experiments. Both
close signals were not resolved in HMBC spectrum and might be
interchangeable.
Ta ble 3. Cytotoxicity of Betulinic Acid and Ceanothic Acid
Derivatives (1, 2, 8-11) on OVCAR-3, HeLa, and FS-5 Cells
using Doxorubicin (10 µg/mL) as Positive Control
cell survival (%)a (n ) 2)
compound
OVCAR-3
HeLa
FS-5
betulinic acid
ceanothic acid (3)
1
2
0
68
0
25
65
18
4
39
81
67
0
16
8
9
10
11
25
51
0
55
21
100
22
50
17
48
19
100
43
87
33
85
56
100
1129 (s), 885 (m), 755 (s), 700 (s) cm-1 1H NMR (CDCl3) δ
;
doxorubicin
2% FCS
1.83 and 2.12 (each 1H, d, J ) 15.9 Hz, H-1), 2.17 (1H, ddd, J
) 3.3, 12.1, 12.1 Hz, H-13), 2.99 (1H, ddd, J ) 4.7, 10.9, 10.9
Hz, H-19), 1.66 (3H, s, H-29), 4.70 (1H, br s, H-30Z), 4.57 (1H,
br s, H-30E), 5.08 (d) and 5.14 (d) (J ) 12.3 Hz) (28-OCH2Ph),
7.34 (5H, m, CH2Ph); EIMS m/z [M]+ 530 (11), 441 (12), 91
(100), 338 (11), 203 (12), 193 (27), 311 (1), 175 (20); HREIMS
(20 eV) m/z [M]+ 530.3764 (calcd for C36H50O3, 530.3762).
1-Deca r boxy-cea n oth ic Acid Ben zyl Ester (7). NaBH4
(42.3 mg, 1.12 mmol) was added portionwise at room temper-
ature to compound 6 (150 mg, 0.28 mmol) dissolved in EtOH
(10 mL) and dioxane (10 mL) and then was stirred for 1 h.
After removal of the organic solvent, the residue was parti-
tioned between H2O (50 mL) and CHCl3 (50 mL × 3). The
combined CHCl3 layer, after drying over anhydrous Na2SO4,
was evaporated to give a residue that was purified on a Si gel
column (5 g, 230-400 mesh) eluted with CHCl3 to give 7 (130
mg, 86.3% yield): Rf 0.50 [MeOH-CHCl3 (3:97)]; mp 75-78
°C; UV λmax (log ꢀ) 257.5 (2.15); IR (KBr) νmax 3450 (br s), 2948
(s), 1722 (s), 1640 (m), 1155 (s), 1123 (s), 1060 (m), 880 (s),
a
The cell survival (%) is equal to the OD50 value of cells cultured
in the presence of tested compounds divided by the OD50 value of
cells cultured in the presence of 2% FCS.
Ta ble 4. Inhibitory Effect of Betulinic Acid and Norceanothic
Acid Derivatives (1, 2, and 10) on OVCAR-3, HeLa, and FS-5
Cells
IC50 (µg/mL) (n ) 4)
compound
OVCAR-3
HeLa
FS-5
betulinic acid
1
2
10
0.9
4.7
2.8
3.5
0.8
9.1
6.6
7.1
20.7
29.6
11.3
22.7
EtOH extract of roots of P. ramosissimus, which had been
described earlier.4 Part of this fraction (57 g) was mixed with
MeOH (250 mL), benzyl chloride (13.8 mL, 0.36 mol), and K2-
CO3 (16.60 g, 0.12 mol). The suspension was refluxed for 24
h, and the cooled mixture was partitioned between H2O (350
mL) and toluene (350 mL × 3). The toluene layer was dried
over anhydrous Na2SO4 and was evaporated under reduced
pressure to give a viscous residue (11 g). This reaction was
repeated for the remaining insoluble residue and worked up
similarly to give an additional 32 g of the crude ceanothic acid
dibenzyl ester. These two fractions were combined and
subjected to column chromatography over Si gel (500 g, 230-
400 mesh) eluted with 0-1% MeOH in CHCl3 to give 4 (19.00
g): amorphous solid; Rf 0.62 [MeOH-CHCl3 (1:9)]; mp 143-
145 °C; UV λmax (log ꢀ) 257.0 (2.70); IR (KBr) νmax 3500 (br s),
2950 (s), 1720 (s), 1700 (m), 1640 (w), 1179 (s), 1159 (s), 1125
(s), 900 (m), 740 (s), and 700 (s) cm-1; 1H NMR (CDCl3) δ 2.60
(1H, s, H-1), 4.16 (1H, s, H-3), 2.13 (1H, ddd, J ) 3.3, 12.2,
12.2 Hz, H-13), 2.97 (1H, ddd, J ) 4.4, 11.0, 11.0 Hz, H-19),
1.09 (3H, s, H-23), 0.89 (3H, s, H-24), 1.01 (3H, s, H-25), 0.84
(3H, s, H-26), 0.72 (3H, s, H-27), 1.64 (3H, s, H-29), 4.70 (1H,
br s, H-30Z), 4.58 (1H, br s, H-30E), 4.95 (d) and 5.08 (d) (J )
1
745 (s), 700 (s) cm-1; H NMR (CDCl3) δ 3.95 (1H, d, J ) 7.7
Hz, H-3), 2.14 (1H, ddd, J ) 3.4, 12.3, 12.3 Hz, H-13), 2.99
(1H, ddd, J ) 4.6, 11.1, 11.1 Hz, H-19), 1.66 (3H, s, H-29),
4.70 (1H, br s, H-30Z), 4.57 (1H, br s, H-30E), 5.07 (d) and 5.13
(d) (J ) 12.1 Hz) (28-OCH2Ph), 7.33 (5H, m, CH2Ph); EIMS
m/z [M]+ 532 (3), 441 (12), 91 (100), 338 (11), 203 (12), 193
(27), 311 (1), 175 (20); HREIMS (20 eV) m/z [M]+ 532.3865
(calcd for C36H52O3, 532.3919).
1-Deca r boxy-cea n oth ic Acid (1). The mixture of com-
pound 7 (110 mg, 0.21 mmol), γ-collidine (0.7 mL), and LiI
(85 mg, 0.64 mmol) was heated under reflux (180 °C) for 1.5
h. The viscous reaction mixture was partitioned between
CHCl3 (50 mL) and H2O (pH 1, 50 mL × 3). The CHCl3 layer,
after drying over anhydrous Na2SO4, was evaporated in vacuo
to give a residue that was purified on a Si gel column (8 g,
230-400 mesh) eluted with 0.25% trifluoroacetic acid in CHCl3
to give 1 (80 mg, 87.5% yield): Rf 0.33 [MeOH-CHCl3 (3:97)];
mp 240-241 °C; IR (KBr) νmax 3450 (br s), 2950 (s), 1698 (s),
1
1640 (m), 1450 (s), 1377 (s), 1180 (s), 880 (m) cm-1; H NMR

1346 J ournal of Natural Products, 1998, Vol. 61, No. 11
Lee et al.
(CDCl3) δ 3.96 (1H, d, J ) 7.9 Hz, H-3), 2.10 (1H, ddd, J )
3.4, 12.1, 12.1 Hz, H-13), 2.90 (1H, ddd, J ) 4.0, 10.4, 10.4
Hz, H-19), 1.67 (3H, br s, H-29), 4.71 (1H, br s, H-30Z), 4.58
(1H, br s, H-30E); EIMS m/z [M]+ 442 (24), 424 (25), 248 (40),
203 (23), 193 (100), 220 (10), 175 (52); HREIMS (20 eV) m/z
[M]+ 442.3450 (calcd for C29H46O3, 442.3449).
1278 (s), 1156 (s), 1122 (s), 1023 (s), 885 (s), 743 (s), 700 (s)
cm-1; 1H NMR (CDCl3) δ 6.20 (1H, s, H-3), 2.10 (1H, ddd, J )
2.8, 12.3, 12.3 Hz, H-13), 2.97 (1H, ddd, J ) 4.8, 11.0, 11.0
Hz, H-19), 1.65 (3H, s, H-29), 4.69 (1H, br s, H-30Z), 4.59 (1H,
br s, H-30E), 5.09 (d) and 5.12 (d) (J ) 12.3 Hz) (2-OCH2Ph),
5.08 (d) and 5.17 (d) (J ) 12.4 Hz) (28-OCH2Ph), 7.34 (10H,
m, CH2Ph); EIMS m/z [M]+ 648 (1), 557 (5), 91 (100), 338 (1),
203 (2), 309 (2), 311 (2), 175 (4); HREIMS (20 eV) m/z [M]+
648.4370 (calcd for C44H56O4, 648.4181).
Cea n oth a -1(3),20(30)-d ien e-2,28-d ioic Acid (9). When
reaction conditions and a workup process similar to those for
the preparation of 1 were used, 13 (180 mg, 0.28 mmol) gave
9 (120 mg, 92.3%): Rf 0.42 [MeOH-CHCl3 (1:19)]; mp 270-
271 °C; IR (KBr) νmax 3300 (br s), 2950 (s), 1960 (s), 1640 (m),
1450 (s), 1378 (s), 1230 (s), 1210 (s), 1192 (s), 880 (s) cm-1; 1H
and 13C NMR (C5D5N) data, see Tables 1 and 2, respectively;
NOE data (C5D5N), H-23 (δ 1.08) to H-3 (δ 6.52) (10.6%) and
H-24 (δ 0.93) (7.5%); H-24 to H-3 (6.7%), H-23 (5.8%), and H-25
(δ 1.44) (10.4%); H-25 to H-24 (7.8%) and H-26 (δ 1.14) (7.8%);
H-26 to H-13 (δ 2.73) (10.5%) and H-25 (13.6%); H-27 (δ 0.97)
to H-9 (δ 2.17) (9.9%) and H-18 (δ 1.65) (8.0%); H-29 (δ 1.71)
to H-19 (δ 3.46) (2.0%), H-30E (δ 4.65) (6.1%), and H-30Z (δ
4.83) (-2.3%); EIMS m/z [M]+ 468 (57), 248 (62), 220 (32), 219
(100), 203 (62), 175 (95); HREIMS (20 eV) m/z [M]+ 468.3238
(calcd for C30H44O4, 468.3241).
1-Deca r boxy-3-oxo-cea n oth ic Acid (2). Under similar
reaction conditions and workup process for the preparation of
1, 5 (300 mg, 0.45 mmol) gave 2 (170 mg, 85.5%): Rf 0.25
[MeOH-CHCl3 (3:97)]; mp 235-236 °C; IR (KBr) νmax 3300
(br s), 2950 (s), 1738 (s), 1690 (s), 1640 (m), 1450 (s), 1375 (s),
1
880 (s), 755 (s) cm-1; H NMR (CDCl3) δ 1.85 and 2.13 (each
1H, d, J ) 15.8 Hz, H-1), 2.19 (1H, ddd, J ) 3.2, 12.3, 12.3
Hz, H-13), 2.98 (1H, ddd, J ) 4.8, 10.8, 10.8 Hz, H-19), 1.67
(3H, br s, H-29), 4.72 (1H, br s, H-30Z), 4.59 (1H, br s, H-30E);
EIMS m/z [M]+ 440 (42), 425 (100), 248 (62), 203 (50), 191 (62),
220 (9), 175 (49); HREIMS (20 eV) m/z [M]+ 440.3292 (calcd
for C29H44O3, 440.3292).
2,3,28-Tr ih yd r oxy-cea n ot h -20(30)-en e (8). Anhydrous
THF (4 mL) was injected to LiAlH4 (40 mg, 0.53 mmol) under
nitrogen. The suspension was stirred at room temperature
for 10 min, then 4 (40 mg, 0.06 mmol), dissolved in anhydrous
THF (1 mL), was added dropwise. The mixture was heated
under reflux for 2.5 h. After cooling, hydrated Na2SO4 was
added to the suspension to destroy excess LiAlH4. The
suspension was then filtered through a Celite pad and the
filtered residue washed with MeOH. The filtrate and wash-
ings were combined and evaporated to give a residue that was
purified on a Si gel column (3 g, 230-400 mesh) eluted with
2% MeOH in CHCl3 to give 8 (20 mg, 72.7% yield): Rf 0.30
[MeOH-CHCl3 (1:19)]; mp 219-220 °C; IR (KBr) νmax 3380
(br s), 2947 (s), 1640 (m), 1020 (s), 1458 (s), 1377 (s), 883 (s)
1-Nor cea n oth a -1(3),20(30)-d ien -28-oic Acid (10). When
reaction conditions and a workup process similar to those for
the preparation of 1 were used, 12 (350 mg, 0.47 mmol) gave
10 (86 mg, 43.1%): Rf 0.55 [MeOH-CHCl3 (3:97)]; mp 203-
204 °C; IR (KBr) νmax 3300 (br s), 2950 (s), 1690 (s), 1640 (m),
1
and 880 (s) cm-1; H NMR (CDCl3) δ 5.90 (1H, d, J ) 5.7 Hz,
H-1), 5.42 (1H, d, J ) 5.7 Hz, H-3), 4.72 (1H, br s, H-30Z-),
4.59 (1H, br s, H-30E-), 2.98 (1H, dt, J ) 4.8, 12.5 Hz, H-19),
2.15 (1H, dt, J ) 3.6, 12.5 Hz, H-13), 1.72 (1H, dd, J ) 3.2,
12.5 Hz, H-9), 1.67 (3H, s, H-29), 1.58 (1H, dt, J ) 3.1, 13.7
Hz, H-15â), 1.58 (1H, t, J ) 11.3 Hz, H-18), 1.51 (1H, dq, J )
4.7, 12.5 Hz, H-11â), 1.39 (1H, m, H-6â), 1.42 (1H, dt, J ) 3.6,
13.2 Hz, H-7R or H-12R), 1.23 (1H, br d, J ) 11.8 Hz, H-5),
1.15 (1H, br d, J ) 13.0 Hz, H-15), 1.00 (3H, s, H-23), 0.98
(3H, s, H-27), 0.93 (3H, s, H-26), 0.92 (3H, s, H-25), 0.89 (3H,
s, H-24); NOE data, H-23 to H-24 (6.3%), H-3 (5.2%), and H-5
(7.2%); H-24 to H-3 (3.5%), H-23 (3.9%), and H-25 (2.4%); H-25
(δ 0.92) to H-1 (1.1%) and H-11â (5.5%); H-26 to H-11â (4.0%),
H-13 (8.9%), H-6â (9.1%), and H-15â (7.9%); H-27 (δ 0.98) to
H-9 (9.3%), H-18 (9.4%), H-15 (5.2%), and H-7R (or H-12R)
(10.2%); EIMS m/z [M]+ 424 (14), 409 (100), 248 (1), 220 (1),
203 (5), 175 (41); HREIMS (20 eV) m/z [M]+ 424.3351 (calcd
for C29H44O2, 424.3343).
cm-1 1H and 13C NMR (C5D5N) data, see Tables 1 and 2,
;
respectively; NOE data (C5D5N), H-23 (δ 1.21) to H-3 (δ 4.69)
(12.1%), H-5 (δ 1.55) (5.6%), and H-24 (δ 1.26) (5.9%); H-24 to
H-23 (3.7%) and H-25 (δ 1.34) (4.0%); H-25 to H-1 (δ 2.26)
(9.6%), H-24 (8.4%), and H-26 (δ 1.06) (10.8%); H-26 to H-13
(δ 1.80) (7.6%) and H-25 (8.1%); H-27 (δ 1.09) to H-9 (δ 1.87)
(9.4%) and H-18 (δ 1.69) (8.3%); H-29 (δ 1.75) to H-19 (δ 2.60)
(1.4%), H-30E (δ 4.73) (6.2%), and H-30Z (δ 4.87) (-1.0%); EIMS
m/z [M]+ 458 (41), 440 (42), 427 (49), 234 (17), 203 (45), 223
(45), 206 (16), 175 (45); HREIMS (20 eV) m/z [M]+ 458.3751
(calcd for C30H50O3, 458.3762).
3-O-Mesylcea n oth ic Acid Diben zyl Ester (12). Meth-
anesulfonyl chloride (MsCl, 0.5 mL) was added slowly to
compound 4 (200 mg, 0.30 mmol) dissolved in anhydrous
pyridine (1 mL) in an ice bath. The reaction was carried at 0
°C for 0.5 h and at room temperature for 1 h. Then pyridine-
H2O (2:1, 0.15 mL) was added and stirred for 15 min to destroy
excess MsCl. The resultant mixture was partitioned between
H2O (50 mL) and CHCl3 (50 mL × 3). The CHCl3 layer, after
drying over anhydrous Na2SO4, was evaporated to give a
residue that was purified on a Si gel column (10 g, 230-400
mesh) eluted with toluene to give 12 (160 mg, 71.6% yield):
Rf 0.68 [MeOH-CHCl3 (3:97)]; mp 163-164 °C; UV λmax (log
ꢀ) 257.5 (2.44); IR (KBr) νmax 2948 (s), 1724 (s), 1715 (s), 1641
(m), 1340 (s), 1180 (s), 1152 (s), 1130 (s), 1030 (m), 870 (s),
1-Nor cea n oth a n -28-oic Acid (11). Compound 10 (60 mg)
in EtOAc (5 mL) was catalytically hydrogenated (H2, 1 atm;
10% Pd-C, 30 mg) in the usual manner at room temperature
overnight. The resultant suspension was filtered through a
Celite pad and the residue washed with CHCl3. The combined
filtrate and washings were evaporated to give a residue that
was purified on a Si gel column (4 g, 230-400 mesh) eluted
with 1% MeOH in CHCl3 to give 11 (40 mg, 66.0% yield): Rf
0.38 [MeOH-CHCl3 (1:19)]; mp 200-202 °C; IR (KBr) νmax
3300 (br s), 2950 (s), 1680 (m), 1640 (m), 1460 (w), 1380 (w),
1
760 (s), 737 (s), 700 (s) cm-1; H NMR (CDCl3) δ 2.58 (1H, s,
1
H-1), 4.92 (1H, s, H-3), 2.92 (3H, s, 3-OMs), 2.13 (1H, ddd, J
) 3.2, 12.2, 12.2 Hz, H-13), 2.96 (1H, m, H-19), 1.63 (3H, s,
H-29), 4.70 (1H, br s, H-30Z), 4.58 (1H, br s, H-30E), 4.96 (d)
and 5.15 (d) (J ) 12.1 Hz) (2-OCH2Ph), 5.06 (d) and 5.12 (d)
(J ) 12.4 Hz) (28-OCH2Ph), 7.34 (5H, m, CH2Ph); EIMS m/z
[M]+ 744 (3), 653 (25), 91 (100), 338 (1), 203 (6), 405 (7), 311
(1), 175 (4); HREIMS (20 eV) m/z [M]+ 744.4072 (calcd for
and 1240 (w) cm-1; H NMR (CDCl3) δ 2.18 (1H, dt, J ) 3.4,
11.7 Hz, H-13), 0.95 (3H, s, H-27), 0.94 (3H, s, H-26), 0.88 (3H,
s), 0.85 (3H, s) and 0.76 (3H, s) (H-23, H-24, and H-25), 0.84
(3H, d, J ) 7.0 Hz) and 0.74 (3H, d, J ) 6.8 Hz) (H-29 and
H-30); EIMS m/z [M]+ 428 (9), 409 (100), 250 (1), 222 (1), 205
(3), 177 (100); HREIMS (20 eV) m/z [M]+ 428.3638 (calcd for
C
29H48O2, 428.3656).
C
45H60O7S, 744.4079).
Cytotoxicity Assa y.8 Human ovarian adenocarcinoma
Cea n oth a -1(3),20(30)-d ien e-2,28-d ioic Acid Diben zyl
(OVCAR-3) cells were maintained in RPMI 1640 medium
supplemented with 10% heat-inactivated fetal calf serum
(FCS), 100 µg/mL streptomycin, 100 IU/mL penicillin, and 2
mM glutamine. Human foreskin fibroblast (FS-5) and cervical
carcinoma (HeLa) cells were grown in DMEM supplemented
with similar components. All cultures were incubated at 37
°C in a humidified atmosphere of 5% CO2. The cytotoxic
activity against OVCAR-3, HeLa, and FS-5 cells was deter-
mined after 48 h of incubation in the presence and absence of
Ester (13). The mixture of compound 12 (400 mg, 0.54 mmol),
i-PrOH (40 mL), and K2CO3 (10.00 g) were heated under reflux
for 48 h. After cooling, the precipitate was filtered and the
filtrate was evaporated to give a residue that was purified on
a Si gel column (20 g, 230-400 mesh) eluted with n-hexane-
CHCl3 (1:1) to give 13 (210 mg, 60.3% yield): Rf 0.45 [Me2-
CO-toluene (1:99)]; mp 129-131 °C; UV λmax (log ꢀ) 258.0
(2.22); IR (KBr) νmax 2950 (s), 1725 (s), 1717 (s), 1640 (m),

Ceanothic Acid Derivatives
J ournal of Natural Products, 1998, Vol. 61, No. 11 1347
Refer en ces a n d Notes
the test compounds. Briefly, cells were seeded at a density of
2 × 104 cells/well in 96-well microtiter plates and incubated
at 37 °C for 24 h before treatment. The growth medium was
then replaced with one that contained the tested compounds
and 2% FCS, and cells were incubated for 48 h. After
incubation with the compounds, the medium was withdrawn,
the cells were washed, 0.1 mL of fresh serum-free medium and
25 µL of MTT solution [5 mg/mL in phosphate-buffered saline
(PBS)] were added to each well and incubated at 37 °C for 4
h. Lysis buffer (100 µL, 20% SDS in 50% dimethylformamide)
was then added, and, after overnight incubation, the absor-
bance of each culture was read at 570 nm using a microplate
reader (Labsystems, Finland).
(1) Pisha, E.; Chai, H.; Lee, I. S.; Chagwedera, T. E.; Farnsworth, N. R.;
Cordell, G. A.; Beecher, C. W.; Fong, H. H.; Kinghorn, A. D.; Brown,
D. M. Nature Med. 1995, 1, 1046-1051.
(2) Fujioka, T.; Kashiwada, Y.; Kilkuskie, R. E.; Cosentino, L. M.; Ballas,
L. M.; J iang, J . B.; J anzen, P.; Chen, I. S.; Lee, K. H. J . Nat. Prod.
1994, 57, 243-247.
(3) Lee, S. S.; Su, W. C.; Liu, K. C. J . Nat. Prod. 1991, 54, 615-618.
(4) Lee, S. S.; Lin, C. J .; Liu, K. C. J . Nat. Prod. 1992, 55, 602-606.
(5) Lee, S. S.; You, B. J .; Wang, K. C. J . Chin. Chem. Soc. 1992, 40, 213-
216.
(6) Elsinger, F.; Schreiber, J .; Eschenmoser, A. Helv. Chim. Acta 1960,
43, 113-118.
(7) Chandrasekaran, S.; Turner, J . V. Synth. Commun. 1982, 12, 727-
731.
(8) Hansen, M. B.; Nielsen, S. E.; Berg, K. J . Immunol. Methods 1989,
119, 203-210.
Ack n ow led gm en t. This research was supported by Na-
tional Science Council, Republic of China.
NP9800856