Synthetic secofriedelane and friedelane derivatives as inhibitors of human lymphocyte proliferation and growth of human cancer cell lines in vitro

Article abstract of DOI:10.1021/np010217m

Controlled silylation of friedelin (1) from cork smoker wash solids, a byproduct generated during processing of corkboard by steam baking, gave 3-trimethylsiloxyfriedel-2-ene (3) in high yields. Oxidation of 3 with OsO₄/NMMO produced 2α-hydroxyfriedelan-3-one (cerin) (5), from which the new 2,3-secofriedelan-2-al-3-oic acid (6) was obtained quantitatively by periodic acid oxidation. Oxidation of 3 with DDQ afforded friedel-1-en-3-one (8). Reductive ozonolysis of 3 gave 2α,3β-dihydroxyfriedelane, pachysandiol A (7). Compound 6 proved to be a potent inhibitor of human lymphocyte proliferation (IC₅₀ = 10.7 μM) and of the growth of a human cancer cell line (GI₅₀ = 5.4-17.2 μM). ¹³C NMR data for compounds (3, 4, 5, 6a, 7, and 8) are described for the first time.

Full text of DOI:10.1021/np010217m

J . Nat. Prod. 2001, 64, 1273-1277
1273
Syn th etic Secofr ied ela n e a n d F r ied ela n e Der iva tives a s In h ibitor s of Hu m a n
Lym p h ocyte P r olifer a tion a n d Gr ow th of Hu m a n Ca n cer Cell Lin es in Vitr o
Cristina Moiteiro,*,† Fa´tima J ustino,† Regina Tavares,*,† M. J . Marcelo-Curto,† M. Helena Floreˆncio,‡
Maria Sa˜o J ose´ Nascimento,§ Madalena Pedro,§ Fa´tima Cerqueira,§ and Madalena M. M. Pinto§
INETI-Instituto Nacional de Engenharia e Tecnologia Industrial, Departamento de Tecnologia de Indu´strias Quı´micas,
Servic¸o de Qu´ımica Fina, Estrada do Pac¸o do Lumiar, 1649-038 Lisboa, Portugal, Faculdade de Cieˆncias de Lisboa,
Departamento de Quı´mica e Bioquı´mica, Bloco C1, piso 5, Campo Grande, 1749-016 Lisboa, Portugal, and
Centro de Estudos de Quı´mica Orgaˆnica, Fitoquı´mica e Farmacologia da Universidade do Porto,
Faculdade de Farma´cia, 4050-047 Porto, Portugal
Received April 26, 2001
Controlled silylation of friedelin (1) from cork smoker wash solids, a byproduct generated during processing
of corkboard by steam baking, gave 3-trimethylsiloxyfriedel-2-ene (3) in high yields. Oxidation of 3 with
OsO4/NMMO produced 2R-hydroxyfriedelan-3-one (cerin) (5), from which the new 2,3-secofriedelan-2-
al-3-oic acid (6) was obtained quantitatively by periodic acid oxidation. Oxidation of 3 with DDQ afforded
friedel-1-en-3-one (8). Reductive ozonolysis of 3 gave 2R,3â-dihydroxyfriedelane, pachysandiol A (7).
Compound 6 proved to be a potent inhibitor of human lymphocyte proliferation (IC50 ) 10.7 µM) and of
the growth of a human cancer cell line (GI50 ) 5.4-17.2 µM). 13C NMR data for compounds (3, 4, 5, 6a ,
7, and 8) are described for the first time.
Sch em e 1
Processing of cork of Quercus suber L. (Fagaceae) is an
important industrial activity in Portugal, generating large
amounts of cork powder and cork smoker wash solids, a
black wax obtained in the process of making corkboard
from ground cork under pressure by treatment with
superheated steam.
Cork smoker wash solids are a convenient source of
friedelin (1) and 3-hydroxyfriedel-3-en-2-one (2).1-3 Many
triterpenoid compounds are reported to have antiinflam-
matory, antiviral, antibacterial, anticancer, or suppression
of tumor promotion activities.4,5 Friedelin (1) has reported
diuretic activity.6 There are no systematic studies of
structure-activity relationships based on chemical modi-
fication of friedelane triterpenoids. Compound 1 offers the
possibility of various structural modifications in ring-A;
thus we have focused on the preparation of derivatives
including seco acids and dihydroxylated and R,â-unsatur-
ated systems (Scheme 1). We report here the synthesis and
biological activity of several fridelane derivatives.
Resu lts a n d Discu ssion
Extraction of cork smoker wash solids with petroleum
ether gave friedelin (1) (8.0%) and 3-hydroxyfriedel-3-en-
2-one (2) (5.4%). Small amounts of â-sitosterol (1%),
campesterol (0.3%), â-amyrin, and sitost-4-en-3-one (0.3%)
were also identified, the latter having not been previously
reported in this material.7 Compounds 1 and 2 and its
derivative, the 3-acetoxyfriedel-3-en-2-one (2a ), have 1H
and 13C NMR spectra identical to those described in the
literature.8-11
It has been reported that, unlike the 4,4-dimethyl-3-keto
triterpenoids, attempts to carry out the transformation of
1 into the respective seco acids with m-chloroperbenzoic
acid12,13 as with other peracids,14 in the presence of a strong
acid such as p-toluenesulfonic acid (p-TsOH), gave only the
ꢀ-lactone instead of the desired seco acids. For example,
the oxidation of 1 with hydrogen peroxide in the presence
of selenium dioxide gave a seco acid in only 2.1% yield,
besides the ꢀ-lactone (48%).15 Although R-acetoxylation of
1 has also been achieved by oxidation with lead tetraac-
* To whom correspondence should be addressed. Tel: 351 217165141.
Fax: 351 217168100. E-mail: regina.tavares@mail.ineti.pt or cristina.
moiteiro@mail.ineti.pt.
† INETI-Instituto Nacional de Engenharia e Tecnologia Industrial.
‡ Faculdade de Cieˆncias de Lisboa.
§ Faculdade de Farma´cia do Porto.
10.1021/np010217m CCC: $20.00
© 2001 American Chemical Society and American Society of Pharmacognosy
Published on Web 09/20/2001

1274 J ournal of Natural Products, 2001, Vol. 64, No. 10
Moiteiro et al.
Ta ble 1. 1H NMR Data (δ) for Compounds 3-7 in CDCl3 (J in Hz)
position
3
4
5a
6a
7b
1R
1.93 (m)
2.30 (m)
1.97 (m)
2.29 (ddd, J gem ) 18.6,
J 1,10 ) 3.9, J 1,2 ) 1.5)
2.45 (ddd, J gem ) 18.6,
J 1,10 ) 5.4, J 1,2 ) 1.5)
1.58 (m)
1â
1.83 (td, J gem ) J 1â,10R )
13.0, J 1â,2â ) 2.7)
2R
2â
4.79 (dd, J ) 1.5, 1.5)
1.80 (m)
2.04 (m)
4.08 (ddd, J ) 3.3, 3.3, 3.3)
3.99 (q, J 2â,1â ) J 2â,1R
) J 2â,3R ) 2.7)
1.58 (m)
1.78 (dd, J 10R,1R ) 3.3,
J 10R,1â ) 13.0)
0.94 (d, J ) 7.2)
0.94 (s)
0.85 (s)
1.01 (s)
0.99 (s)
1.17 (s)
4R
10R
1.95 (m)
1.45 (m)
2.82 (q, J ) 6.6)
1.75 (dd, J 10R,1R ) 3.0,
J 10R,1â ) 10.8)
0.89 (d, J ) 6.6)
0.72 (s)
0.89 (s)
1.05 (s)
1.00 (s)
1.00 (s)
2.30 (q, J ) 6.9)
2.02 (dd, J ) 3.9, 5.4)
1.45 (m)
23
24
25
26
27
28
29
30
0.84 (d, J ) 7.2)
0.90 (s)
0.75 (s)
1.00 (s)
1.00 (s)
1.17 (s)
1.00 (s)
0.94 (s)
0.17 (s)
1.47 (s)
0.83 (s)
0.95 (s)
1.00 (s)
1.00 (s)
1.17 (s)
1.00 (s)
0.96 (s)
0.16 (s)
1.09 (d, J ) 6.9)
0.91 (s)
0.89 (s)
1.03 (s)
0.98 (s)
1.17 (s)
0.98 (s)
0.94 (s)
1.18 (s)
0.95 (s)
0.99 (s)
0.94 (s)
OSi(Me)3
CO2Me
CHO
3.62 (s)
9.76 (br s)
a
b
δ 2.18 (d, J ) 3.3, OH). δ 3.54 (br t, H-3R).
Ta ble 2. 13C NMR Data (δ) for Compounds 3-8 in CDCl3
etate in the presence of boron trifluoride etherate, the
synthesis is not regioselective and the yields range from
22 to 28%.16
position
3
4
5
6a
7
8
1
20.8
102.5
151.3
49.3
36.7
41.8
18.4
53.0
36.8
55.9
35.4
30.4
39.7
38.2
32.9
36.1
30.1
42.9
35.4
28.2
32.5
39.3
9.9
31.8
39.7
142.0
128.5
37.8
18.3
18.3
53.0
37.1
57.1
35.4
30.7
39.9
38.5
33.0
36.2
30.1
43.1
35.4
28.2
32.4
39.3
9.7
32.4
76.9
212.0
52.7
42.7
40.8
17.5
52.7
38.0
52.7
35.6
30.2
39.4
38.9
32.7
36.5
29.8
42.4
35.0
27.8
32.0
40.8
6.3
29.7
201.6
175.9
48.3
47.7
41.3
18.4
51.0
38.6
52.7
36.1
30.2
39.8
38.4
32.4
36.2
30.2
43.1
35.4
28.2
33.1
39.3
12.2
18.2
17.8
18.7
20.2
32.2
31.9
34.9
23.8
71.3
76.4
43.6
37.7
41.3
18.1
53.2
36.5
52.2
35.4
30.6
39.6
38.3
32.2
36.0
29.9
42.7
35.2
28.1
32.7
39.3
10.8
15.8
17.4
18.7
20.1
32.0
31.7
35.0
130.3
148.7
201.4
57.9
43.7
40.0
18.6
52.5
36.8
62.4
35.5
30.3
40.1
38.4
32.3
36.1
30.1
43.1
35.0
28.2
33.0
39.3
6.6
The regiochemistry of silyl enol ether formation via
kinetically and thermodynamically enolate derivatives, the
ease with which these silyl enol ethers may be prepared
in nearly quantitative yields, and the enhanced reactivity
of silyloxyalkenes toward electrophilic species make such
substances attractive intermediates for R-hydroxylation
methodology.17,18 A route leading to seco friedelanes was
devised via friedelin enol silyl ethers. Conversion of ketones
to silyl enol ethers is a well-known process; however, the
formation of mixtures in the application of such methods
is often observed.19-21 Friedelin (1) reacts with trimethyl-
silyl chloride in the presence of LDA to give no more than
52% of 3-trimethylsiloxyfriedel-2-ene (3), while with Et3N
and trimethylsilyl trifluoromethanesulfonate a 1:1 mixture
of the isomers 3 and 4 in 84% yield is obtained. Neverthe-
less, regiospecific silylation of 1 was achieved by reaction
with N,O-bis(trimethylsilyl)acetamide in the presence of
sodium and HMPA under kinetic control to give 98% of
3-trimethylsiloxyfriedel-2-ene (3) alone. In contrast, heat-
ing 1 in the presence of the same reagents at 40-50 °C for
3 h gave the silyl enol ether of thermodynamic enolate 4
as the major product (62%).
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
OSiMe3
OMe
14.0
17.5
18.6
20.4
32.1
31.9
35.0
0.3
20.7
17.8
18.6
20.1
32.2
31.9
35.0
0.7
13.8
17.9
18.4
19.9
32.1
31.7
34.7
13.8
17.9
19.8
18.6
32.2
31.8
34.9
The 1H NMR spectrum of 3 clearly shows one vinylic
proton as a triplet at δ 4.79 (J ) 1.5 Hz) assigned to H-2
and a singlet signal of nine protons at δ 0.17. Although,
the mass spectra of compounds 3 and 4 are similar, these
1
51.0
compounds can be distinguished by comparison of their H
NMR spectra; thus, in compound 4, the absence of the
olefinic proton was noted and the Me-23 was revealed as a
singlet downfield (δ 1.47) in relation to Me-23 in 3 (δ 0.84,
1
in the H NMR spectrum of 5. The oxymethine proton at δ
4.08 (ddd, J ) 3.3, 3.3, 3.3 Hz) indicated an equatorial
orientation. Irradiation of H-1 at δ 1.97 collapsed the ddd
of the oxymethine proton into a doublet, due to coupling
with the OH group (δ 2.18, d, J ) 3.3 Hz). Cerin (5) was
also obtained from the reaction of compound 3 with
m-chloroperbenzoic acid. This alternative route is advanta-
geous in terms of yield (85%) and lower toxicity of the
oxidizing reagent. Cerin acetate (5a ) was obtained by
acetylation of compound 5.23
1
d), due to the double bond position. The H NMR and 13C
NMR data of both 3 and 4 are shown in Tables 1 and 2,
respectively.
Oxidation of 3 with a catalytic amount of osmium
tetroxide together with N-methylmorpholine N-oxide
(NMMO‚H2O) afforded 2R-hydroxyfriedelan-3-one (cerin)
(5) in 59% yield, a compound previously reported as a
constituent of cork and not occurring in cork smoker wash
solids.22 The IR spectrum of 5 showed the presence of
hydroxyl and carbonyl groups. The EIMS of 5 exhibited a
molecular ion at m/z 442. The configuration of the 2-hy-
droxy group was derived from the shape of the 2â-H signal
Oxidation of 5 with periodic acid at 0 °C yielded 2,3-
secofriedelan-2-al-3-oic acid (6), characterized as its methyl
ester 6a .24 The molecular ion at m/z 472 together with the
isotopic peaks at m/z 473 and 474 in the EIMS were in

Synthetic Secofriedelane and Friedelane Derivatives
J ournal of Natural Products, 2001, Vol. 64, No. 10 1275
Ta ble 3. Effect of Compounds on the Growth of Human Cancer Cell Lines (concentration causing 50% cell growth inhibition)
GI50 µM
compounds
MCF-7 (breast)
TK-10 (renal)
UACC-62 (melanoma)
NCI-H460 (lung)
SF-268 (CNS)
1, 2, 6a , 8-11
>100
>100
>100
>100
>100
5
6
7
59.5 ( 12.0
7.0 ( 0.4
66.4 ( 5.6
90.5 ( 13.3
10.1 ( 1.1
95.7 ( 16.9
86.0 ( 22.6
9.9 ( 1.1
83.3 ( 15.5
67.8 ( 3.2
5.4 ( 0.7
79.4 ( 2.7
95.8 ( 9.5
17.2( 0.9
15.5 ( 2.1
a
Doxorubicin, GI50 MCF-7 ) 5.5 × 10-2 µM; GI50 TK-10 ) 57.0 × 10-2 µM; GI50 UACC-62 ) 9.4 × 10-2 µM; GI50 NCI-H460 ) 0.81 ×
10-2 µM; GI50 SF-268 ) 9.3 × 10-2 µM.
agreement with the molecular formula C31H52O3 , and
elimination of the aldehyde carbonyl and ester moieties was
indicated by a fragment ion at m/z 385. The 1H NMR
spectrum of 6a (Table 1) showed signals for seven tertiary
methyl groups in addition to the secondary Me-23 group
at δ 1.09 as a doublet (J ) 6.9 Hz). The double doublet at
δ 2.02 was assigned to H-10, coupled to the H-1 protons
(J 10,1 ) 5.4 and 3.9 Hz). One of the H-1 protons appeared
at δ 2.45, ddd, coupled to the H-10, gem-proton H-1, and
aldehyde proton (J ) 5.4, 18.6, and 1.5 Hz), respectively.
The methyl ester group and the aldehyde proton appear
as a singlet (δ 3.62) and a broad singlet at δ 9.76,
respectively. The 13C NMR spectrum of 6a showed carbonyl
signals at δ 201.6 and 175.9 assignable to the aldehyde
and the ester groups, respectively, and with C-1 downfield
at δ 29.7 due to the proximity of the aldehyde group. The
IR absorption bands at 2735 and 1732 cm-1 further
supported the presence of aldehydic and ester carbonyl
functionalities in the molecule.
Ozonolysis of reactive enol silyl ethers to produce car-
boxylic acids is also a well-established procedure for
regiospecific cleavage.25 Attempts to cleave ring-A of 3 by
reductive ozonolysis did not give the expected 2,3-seco-
friedelane derivative, giving instead 2R,3â-dihydroxy-
friedelane (7), pachysandiol-A, previously isolated from
Pachysandra terminalis Sieb et Zucc (Buxaceae)26 and from
leaves of Lithocarpus sp. (Fagaceae).27 An anomalous
oxidation had already been observed in the ozonization of
a hindered enol silyl ether.28 Compound 7 was character-
ized by 1H and 13C NMR data (Tables 1 and 2) and by
comparison with an authentic sample obtained by reduc-
tion of cerin (5) with sodium borohydride. Acetylation of 7
gave 2R,3â-diacetoxyfriedelane (7a ), with spectral data in
accordance with those published for pachysandiol-A diac-
etate 7a .26
R,â-Unsaturation of ring-A of 1 was achieved by treat-
ment of 3 with DDQ to afford friedel-1-en-3-one (8) in 76%
yield. This pathway leading to 8 is an improvement over
methods previously reported, such as reaction of 1 with
DDQ in dioxane (9%),14 H2O2, and SeO2 (2.5%)15 and of 2R-
bromofriedel-3-one with LiBr and Li2CO3 (20%).14 Com-
pound 8 was identified by comparing the physical and
spectral data with literature values.14,15 3â-Hydroxy-
friedelane (9) and 3R-hydroxyfriedelane (10) were obtained
by stereoselective reduction of friedeline (1) in quantitative
yields.
The reactive silyl enol ether 3 is a convenient intermedi-
ate for the synthesis of a variety of friedelane derivatives.
In particular, 2R,3â-dihydroxyfriedelane (7) is made avail-
able as a starting material for the production of known
secofriedelane derivatives.5,27
liferation. No inhibitory effect was observed with triter-
penes 1, 2, 6a , 8, 9, 10, 10a , and 11 when tested at 100
µM.
Compounds 5 and 7 exhibited a moderate suppression
of T-cell proliferation (IC50 ) 82.3 ( 4.0 and 79.1 ( 2.2
µM, respectively), while 6 was a potent inhibitor (IC50
)
10.7 ( 1.6 µM) of the mitogenic response of human
lymphocytes to PHA. No lymphocytotoxicity was observed
when these compounds were tested at their IC50 concentra-
tions (cell viability of 92.7%, 113.0%, and 89.0%, respec-
tively), which leads to the conclusion that the inhibitory
activity of these compounds was associated with cell
proliferation rather than to a toxic effect.
All the compounds were tested on the growth of five
human cancer cell lines, MCF-7 (breast), TK-10 (renal),
UACC-62 (melanoma), NCI-H460 (lung), and SF-268 (CNS),
and the results are summarized in Table 3. Only com-
pounds 5, 6, and 7 inhibited the growth of the cancer cell
lines tested in a dose-dependent manner. Compounds 1,
2, 8, 9, 10, 10a , 11, and 6a , the corresponding methyl ester
of 6, were inactive on all the cell lines even when tested at
100 µM, and 5 and 7 showed slight inhibition (GI50 > 50
µM). The seco compound 6 was a potent inhibitor of MCF-7
(breast), TK-10 (renal), UACC-62 (melanoma), and NCI-
H460 (lung) cell lines, with GI50 values ranging from 5.4
to 10.1 µM (Table 3). Curiously, diol 7 was more inhibitory
against SF-268 (GI50 ) 15.5 ( 2.1 µM) than against the
other four cell lines tested.
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. Melting point
determinations were performed on a Reichert Thermovar
apparatus and are uncorrected. UV and IR spectra were
recorded on a Hitachi 100-60 or 150-20 and Perkin-Elmer
298 spectrophotometers, respectively. 1H, 13C NMR, DEPT,
and 2D NMR (300 MHz) spectra (all in CDCl3) were
obtained on a GE-300 spectrometer, and chemical shifts
are given in parts per million (δ) using Me4Si as internal
standard. EIMS were measured on a Kratos MS-25 RF
spectrometer at 70 eV attached to a Mach 3 system. HRMS
were performed on a J EOL J MS-SX/SX 102A mass spec-
trometer. TLC was carried out on precoated silica gel 60
F
254 sheets (Merck), and detection was achieved by spraying
with 20% ethanolic H2SO4 followed by heating, or 10%
ethanolic 2,4-DNP; silica gel (Merck, 230-400 mesh) was
used for column chromatography. All solvents were dried
and distilled before use. Ozone was produced from O2 (20
L/h) in a Wallace & Tiernan ozonizer BA.023 at 2000 V.
LDA was freshly prepared from diisopropylamine and
n-BuLi in THF.
Fetal bovine serum and RPMI-1640 medium were ob-
tained from Gibco BRL. Cyclosporin A, 3-(4,5-dimethylthi-
azol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), N,N-
dimethylformamide (DMF), dimethyl sulfoxide (DMSO),
gentamicin, L-glutamine, Histopaque-1077, phytohemag-
glutinin (PHA), doxorubicin, sodium lauryl sulfate (SDS),
and sulforhodamine B (SRB) were purchased from Sigma
The inhibitory activity of the 11 compounds against the
in vitro proliferation of human lymphocytes was examined.
Phytohemaglutinin was used as a T-mitogen. Cyclosporin
A, tested in the same system, served as the internal control
(IC50 ) 0.34 µM). Only compounds 5, 6, and 7 showed a
dose-dependent inhibitory effect against lymphocyte pro-

1276 J ournal of Natural Products, 2001, Vol. 64, No. 10
Moiteiro et al.
Chemical Co. Stock solutions of compounds were prepared
in DMSO and stored at -20 °C, providing uniform samples
for retests. These frozen concentrates were then diluted to
the desired final concentrations prior to the assays. Final
concentrations of DMSO showed no interference with the
biological activities tested.
mmol) at 0 °C with stirring for 1 h. After 3 h at room
temperature, the reaction mixture was filtered and washed
with Et2O, and the filtrate concentrated to give 2,3-
secofriedelan-2-al-3-oic acid (6) (7.3 mg, 0.015 mmol, 94%),
which upon methylation with diazomethane afforded the
corresponding methyl ester 6a : mp 156-161 °C; IR (KBr)
νmax 2735 (C-H aldehyde), 1732 (CdO ester and aldehyde)
Ma ter ia l. Cork smoker wash solids were a gift from
Corticeira Amorim Algarve, Portugal.
1
cm-1; H and 13C NMR data in Tables 1 and 2; EIMS m/z
473 (9), 472 [M]+ (20), 457 (15), 441 (5), 385 (14), 273 (40),
218 (48), 205 (68), 123 (66), 95 (99), 69 (100), 88 (39);
HREIMS m/z 472.3937 (calcd for C31H52O3, 472.3916).
Extr a ction a n d Isola tion . Powdered cork smoker wash
solids (1.2 kg) were extracted using a Soxhlet apparatus
with petroleum ether for 2 h, the extract was evaporated
to dryness in vacuo to give a residue (230 g, 23%). The
crude extract was applied to a silica gel column eluted with
petroleum ether and petroleum ether-CH2Cl2 mixtures of
increasing polarity to give the following: waxes (0.6 g,
4.3%), friedelin (1) (19.0 g, 8.0%), 3-hydroxyfriedel-3-en-2-
one (2) (12.5 g, 5.4%), â-amirin, â-sitosterol (1%), campes-
terol (0.3%), and sitost-4-en-3-one (0.3%). These compounds
were isolated and identified by comparison with literature
data and/or authentic samples.
2r,3â-Dih yd r oxyfr ied ela n e (7). 3-Trimethylsiloxy-
friedel-2-ene (3) (32.5 mg, 0.065 mmol) was dissolved in
CH2Cl2-MeOH (4:1, 10 mL) and cooled to -70 °C. It was
then treated with ozone (30 L/min) for 1 h, followed by
addition of sodium borohydride (5.3 mg). The solution was
stirred overnight at room temperature. Excess hydride was
destroyed with diluted HCl (2%, 2 mL) and the reaction
mixture extracted with CH2Cl2 to give a crude solid (25.8
mg, 79.4%) and purified by silica gel column chromatog-
raphy (CH2Cl2-MeOH, 1%) to give 7 (16.0 mg, 0.036
Silyla tion of (1). Compound 1 (300 mg, 0.7 mmol) was
added, under nitrogen, to a mixture of N,O-bis(trimethyl-
silyl)acetamide (1.2 mL, 4.9 mmol), HMPA (0.6 mL, 3.5
mmol), and an excess of sodium. The mixture was stirred
at room temperature for 1 h, and pentane and iced water
were subsequently added. The organic layer was separated
and dried over Na2SO4. Evaporation of the solvent under
reduced pressure afforded 3-trimethylsiloxyfriedel-2-ene (3)
(344 mg, 0.69 mmol, 98.1%): white crystals (from petrol-
eum ether-CH2Cl2); mp 188-190 °C; IR (KBr) νmax 2920,
2860 (C-H), 1670 (CdC), 1050 (Si-O-C), 900, 840 (Si-
mmol, 55.4%): mp 270-273 °C (lit.26 291-292 °C); IR (KBr)
1
ν
max3440, 2940 cm-1; H and 13C NMR data in Tables 1
and 2; EIMS m/z 444 [M]+ (5), 429 (10), 410 (3), 291 (12),
149 (65), 123 (32), 109 (38), 95 (48), 69 (100).
Acetyla tion of 7. Compound 7 (1.5 mg) was dissolved
in pyridine (1 mL) and treated with Ac2O (1 mL) overnight
at room temperature. Evaporation yielded 2R,3â-diacetoxy-
friedelane (7a ) (1 mg, 56%) as a white solid. Physical and
spectroscopic data are in agreement with the literature.26
F r ied el-1-en -3-on e (8). 3-Trimethylsiloxyfriedel-2-ene
3 (17.0 mg, 0.034 mmol) was dissolved in C6H6 (1 mL) and
added dropwise to a solution of DDQ (9.0 mg, 0.04 mmol)
in C6H6 (3 mL). After stirring overnight at room temper-
ature, saturated aqueous NaHCO3 was added (3 mL). The
organic layer was separated, and the aqueous layer was
washed with Et2O (2 × 25 mL). The organic layers were
evaporated under reduced pressure, and the product was
purified by preparative chromatography (petroleum ether-
CH2Cl2, 2:1) to afford 3 (1.6 mg, 9.4%) and friedel-1-en-3-
one (8) (11.0 mg, 0.026 mmol, 76.0%): white needles (from
CH2Cl2-Me2CO); 13C NMR (see Table 2). Physical and
spectroscopic data are in agreement with the literature.14
1
C) cm-1; H and 13C NMR data in Tables 1 and 2; EIMS
m/z 498 [M]+ (30), 483 (72), 273 (9), 205 (24), 195 (22), 123
(50), 109 (62), 95 (95), 75 (96), 73 (100); HREIMS m/z
498.4270 (calcd for C33H58OSi, 498.4257).
Similar reaction at 40-50 °C for 3 h gave a mixture of 3
(19.6%) and 3-trimethylsiloxyfriedel-3-ene (4) (62.0%): mp
195-197 °C; 1H and 13C data in Tables 1 and 2; EIMS m/z
498 [M]+ (4), 483 (45), 207 (41), 205 (8), 209 (16), 135 (12),
123 (18), 119 (21), 95 (42), 75 (100), 73 (45); HREIMS m/z
498.4228 (calcd for C33H58OSi, 498.4257).
Oxidation of compound 1 with fresh LDA and chloro-
trimethylsilane in THF at -78 °C gave 3 (52%) and un-
reacted starting material. Compound 1 in CCl4 with Et3N
and trimethylsilyl trifluoromethanesulfonate in CH2Cl2 at
room temperature gave 84% of a 1:1 mixture of 3 and 4.
3â-Hyd r oxyfr ied ela n e (9). Compound 1 (24.9 mg,
0.058 mmol) was dissolved in CH2Cl2-MeOH (1:1 v/v, 4
mL) cooled to -78 °C and treated with sodium borohydride
(25 mg, 0.66 mmol). The solution was stirred for 1 h at -78
°C and overnight at room temperature, sodium hydroxide
(1 M, 15 mL) was added, and the reaction mixture was
extracted with dichloromethane to give 9 (25.0 mg, 0.058
mmol, 100%) as white crystals (from CH2Cl2): mp 283-
2r-Hyd r oxyfr ied ela n -3-on e (Cer in ) (5). Compound 3
(37.5 mg, 0.08 mmol) was dissolved in acetone (30 mL) and
cooled to -5 °C, then OsO4 (4.0 mg, 0.016 mmol) in tert-
butyl alcohol (0.3 mL) and NMMO (11.4 mg, 0.097 mmol)
in H2O (0.16 mL) and acetone (0.32 mL) were added
dropwise. The mixture was stirred for 3 h at 0 °C and then
overnight at room temperature. Sodium bisulfite (14.0 mg)
and Florisil (54.0 mg) were added with stirring for 30 min,
the mixture was filtered, and the solvent was evaporated
under reduced pressure. The mixture was acidified to pH
2 with 1 M H2SO4, saturated with salt, and extracted with
AcOEt (3 × 4 mL). The organic layer separated, dried, and
concentrated. The white residue (33 mg) was purified by
silica gel column chromatography (CH2Cl2-MeOH, 1%) to
yield 2R-hydroxyfriedelan-3-one (5) (21.0 mg, 0.047 mmol,
285 °C (lit.1 287-288 °C); [R]25 +17.26 (c 3.65, CHCl3).
D
Physical and spectroscopic data were compared with an
authentic sample.1,9
3r-Hyd r oxyfr ied ela n e (10). Compound 1 (100.0 mg,
0.234 mmol) was dissolved in butanol (25 mL) and treated
with an excess of sodium and HMPT. The mixture was
refluxed for 1.5 h, and H2O was subsequently added. The
organic layer was separated, and the aqueous layer was
washed with butanol (50 mL). The organic layers were
dried and concentrated to afforded 10 (100.1 mg, 0.234
mmol, 100%): mp 296-300 °C (lit.1 302-304 °C).
Lym p h ocyte Assa ys. The effects of compounds on the
mitogenic response of human lymphocytes to PHA (10 µg/
mL) were evaluated using a colorimetric MTT assay as
previously described.29 Mononuclear cells were isolated
from heparinized peripheral venous blood of healthy vol-
1
62.7%): white powder; mp 230 °C (lit.23 246-250 °C); H
and 13C NMR data in Tables 1 and 2; IR and EIMS are in
agreement with literature.23
2,3-Secofr ied ela n -2-a l-3-oic a cid (6). 2R-Hydroxy-
friedelan-3-one (5) (7.0 mg, 0.016 mmol) was dissolved in
Et2O (1.5 mL) and treated with periodic acid (4.5 mg, 0.019

Synthetic Secofriedelane and Friedelane Derivatives
J ournal of Natural Products, 2001, Vol. 64, No. 10 1277
(Unidade de I&D No. 226/94), POCTI (QCA III) and FEDER
for financial support. Ph.D. grants from FCT to M.P. (SFRH/
BD/1456/00) and F.C. (PRAXIS XXI/DB/21801/99) are acknowl-
edged.
unteers by Histopaque-1077 density centrifugation. Cy-
closporin A was used as positive control. The toxicity of
the compounds against the human lymphocytes was de-
termined by an assay based on the ability of metabolically
active cells to reduce the colorless tetrazolium salt MTT
to a colored formazan product.30 Briefly, in flat-bottom 96-
well plates, (2-3) × 106 cells/mL in RPMI-1640 medium
containing 10% heat-inactivated fetal bovine serum, 2 mM
glutamine, and 50 µg/mL of gentamicin were exposed for
24 h to the various concentrations of each sample. Follow-
ing this incubation period, the MTT solution (1 mg/mL) was
added. After incubation for 4 h the MTT formazan products
were solubilized with SDS/DMF solution (20% SDS in a
50% solution of DMF, pH 4.7) overnight at 37 °C. Absor-
bance (OD 550 nm) of the colored solution was measured
with a plate reader. Lymphocytoxicity, defined in terms of
the percentage of viable cells, was present when the
viability of the exposed cells was less than 70% of the
nonexposed control cells.
Cell Gr ow th In h ibitor y Assa y. Human tumor cell
lines MCF-7 (breast cancer), TK-10 (renal cancer), UACC-
62 (melanoma), NCI-H460 (lung), and SF-268 (CNS) were
kindly provided by the National Cancer Institute (NCI,
Bethesda, MD). Cells were routinely maintained as adher-
ent cell cultures in RPMI-1640 medium containing 5%
heat-inactivated fetal bovine serum, 2 mM glutamine, and
50 µg/mL gentamicin at 37 °C in a humidified air incubator
containing 5% CO2. They were subcultured weekly, and the
culture medium was changed twice a week. According to
their growth profiles, the optimal plating densities of each
cell line were determined to ensure exponential growth
throughout the experimental period. Cells were tested for
lack of mycoplasma contamination every 3-5 months. The
effects of the compounds on the growth of human cancer
cell lines were evaluated according to the procedure
adopted in the NCI’s in vitro anticancer drug screening that
uses the SRB assay to assess growth inhibition.31 Growth
inhibition of 50% (GI50) was calculated as described else-
where.31 IC50 and GI50 values, expressed as means ( SEM
were obtained from 3-6 independent experiments per-
formed in duplicate.
Refer en ces a n d Notes
(1) Corey, E. J .; Ursprung, J . J . J . Am. Chem. Soc. 1956, 78, 5041-5051.
(2) Kane, V. V.; Stevenson, R. J . Org. Chem. 1960, 25, 1394-1396.
(3) Ribas-Marques, I.; Fernandez-Salgado, J . Anal. Quı´m. 1974, 70, 363-
366.
(4) Mahato, S. B.; Sucharita, S. Phytochemistry 1997, 44, 1185-1236.
(5) Baas, W. J . Phytochemistry 1985, 24, 1875-1889.
(6) Rizvi S. H.; Shoeb A.; Kapil R. S.; Popli S. P. Experientia 1980, 36,
146-150.
(7) Moiteiro, C. Mc. Dissertation, Universidade Nova de Lisboa, Portugal,
1992.
(8) Patra, A.; Chaudhuri, S. K. Magn. Reson. Chem. 1987, 25, 95-100.
(9) Patra, A.; Chaudhuri, S. K.; Acharyya, A. K. Magn. Reson. Chem.
1990, 28, 85-92.
(10) Patra, A.; Chaudhuri, S. K.; Ru¨bgger, H. J . Indian Chem. Soc. 1990,
67, 394-397.
(11) Talapatra, S. K.; Pradhan, D. K.; Talapatra, B. Indian J . Chem. 1978,
16B, 361-365.
(12) Patra, A.; Chaudhuri, S. K. Indian J . Chem. 1989, 28B, 376-380.
(13) Pradhan, B. P.; Chakraborty, S.; Weyerstahl, P. Tetrahedron Lett.
1989, 30, 5463-5466.
(14) Talapatra, B.; Lahiri, B.; Basak, A.; Pradhan, D. K.; Talapatra, S. K.
Indian J . Chem. 1983, 22B, 741-745.
(15) Anjaneyulu, V.; Rao, G. S. Indian J . Chem. 1984, 23B, 663-664.
(16) Dutta, G.; Bose, S. N. Indian J . Chem. 1989, 28B, 975-977.
(17) Rubotton, G. M.; Vazquez, M. A.; Pelegrina, D. R. Tetrahedron Lett.
1974, 49-50, 4319-4322.
(18) Prakash, O.; Saini, N.; Tanwar, M. P.; Moriarty, R. M. Contemp. Org.
Synth. 1995, 121-131.
(19) Brownbridge, P. Synthesis 1983, 1-28, and references therein.
(20) Corey, E. J .; Gross, A. W. Tetrahedron Lett. 1984, 25, 495-498.
(21) Moiteiro C.; Mata M. L.; Tavares R.; Curto M. M. J . Presented at the
7th European Conference on Biomass for Energy Industry and
Environment Agriculture and Industry Florence, Italy, October 5-9,
P.14.112, 1992.
(22) Patra, A.; Chaudhuri, S. K. Indian J . Chem. 1988, 27B, 1152-1153.
(23) Zhong, S.; Waterman, P. G.; J effreys, J . A. D. Phytochemistry 1984,
23, 1067-1072.
(24) Moiteiro, C.; Tavares, M. R.; Marcelo Curto, M. J . Eur. Pat. 651,760,
28 Apr 1994; Chem. Abstr. 1995, 122, 187820.
(25) Clark, R. D.; Heathcock, C. H. J . Org. Chem. 1976, 41, 1396-1403.
(26) Kikuchi, T.; Toyoda, T. Chem. Pharm. Bull. 1971, 19, 753-758.
(27) Hui, W.; Li, M. M.; Lee Y. J . Chem Soc., Perkin Trans. 1 1975, 7,
617-619.
(28) Clark, R. D.; Heathcock, C. H. Tetrahedron Lett. 1974, 23, 2027-
2030.
(29) Gonzalez, M. J .; Nascimento, M. S. J .; Cidade, H.; Pinto, M. M. M.;
Kijjoa, A.; Anantachoke, C.; Silva, A. M. S.; Herz, W. Planta Med.
1999, 65, 368-370.
(30) Mosmann, T. J . Immunol. Methods 1983, 65, 55-63.
(31) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J .; Vistica,
D.; Warren, J . T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J . Natl Cancer
Inst. 1990, 82, 1107-1112.
Ack n ow led gm en t. The authors thank Mr. C. Versluis
(Dept. Biomolecular Mass Spectrometry, University of Utrecht,
The Netherlands) for performing the high-resolution mass
measurements and Fundac¸a˜o para a Cieˆncia e Tecnologia
NP010217M