Five New Cassane Diterpenes from Myrospermum frutescens with Activity
against Trypanosoma cruzi
Daniel Torres Mendoza,? Luis D. Uren?a Gonza?lez,? Eduardo Ortega-Barr??a,? Todd L. Capson,? and
Luis Cubilla Rios*,?
Laboratory of Natural Products, Faculty of Natural and Exact Sciences and Technology, Apartado 0824, University of
Panama, Panama City, Republic of Panama, Institute for Tropical Medicine and Health Science, Florida State
University-Panama, Panama City, Republic of Panama, and Smithsonian Tropical Research Institute,
Panama City, Republic of Panama
Received January 10, 2003
Five novel cassane diterpenes (1-5) with activity against Trypanosoma cruzi were isolated from leaves
of Myrospermum frutescens. The structures were determined as 18-hydroxycassan-13,15-diene (1), 6?,-
18-dihydroxycassan-13,15-diene (2), 6?-hydroxy-18-acetoxycassan-13,15-diene (3), 18-acetoxy-13,15-diene-
19-cassanoic acid (4), and 6?,13?-dihydroxy-18-acetoxycassan-14(17),15-diene (5). Structures were
elucidated by spectroscopic analysis (NMR and HRCIMS) and by the synthesis of derivatives 2a and 2b.
Compounds 3 and 5 were more active against the extracellular form of the parasite (11 and 16 ?M,
respectively) than the intracellular forms, while compounds 1 and 2 were more active against the more
clinically relevant intracellular forms of the parasite (17 ?M). Compounds 1 and 2 were approximately
9-fold more toxic toward T. cruzi than toward human fibroblasts, the cell type that serves as the parasite?s
mammalian host cell.
As part of the activities of the Panama International
Cooperative Biodiversity Groups Program,1 a program to
find treatments for parasitic diseases, cancer, and HIV, we
report the isolation of five novel cassane diterpenes (1-5)
from the leaves of Myrospermum frutescens Jacq. (Fa-
baceae, subfamily Papilionaceae). Compounds 1-5 dem-
onstrate activity against Trypanosoma cruzi, the causative
agent of Chagas? disease, a parasitic disease that outranks
the combined burden of malaria, schistosomiasis, and
leishmaniasis in terms of its economic impact in Latin
America.2 Current estimates from the World Health Or-
ganization indicate that 16-18 million people are infected,
with an additional 100 million at risk.3 There is an urgent
need for novel treatments for Chagas? disease, as there are
no treatments currently available for use on a large-scale
nor have any vaccines been developed.
There are no ethnopharmacologic uses for this plant
reported in the literature. Previous chemical investigations
of M. frutescens were limited to the isolation of oils from
the seeds.4,5 There are reports of diterpenes with activity
against T. cruzi, the majority of which possess kaurane
skeletons.6-8 Compounds with the cassane skeletons, such
as those described herein, have been isolated from the
genus Caesalpinia (Fabaceae, subfamily Caesalpiniaceae),
the majority of which are furanoditerpenoids.9-11
The structures of the cassane diterpenes 1-5 were
determined primarily by NMR techniques, in particular 1D
1H and 13C NMR (broad band and DEPT) and 2D NMR
(COSY-90, NOESY, HSQC, and HMBC), irradiation, and
NOE differential experiments. Herein we describe the
isolation, structural characterization, and anti-trypanoso-
mal activity of these diterpenes.
Results and Discussion
An MeOH/EtOAc extract of leaves of M. frutescens
showed activity against the extracellular form of the T.
cruzi parasite. The extract was suspended in H2O and
extracted successively with CH2Cl2 and EtOAc. The residue
obtained from the CH2Cl2 layer was then partitioned
between hexane and 10% aqueous MeOH. Activity against
T. cruzi was found in the hexane, EtOAc, and MeOH
fractions. The MeOH fraction was selected for bioassay-
guided fractionation due to the enhanced likelihood of
finding compounds of medium polarity and measurable
anti-trypanosomal activities, yielding the novel cassane
diterpenes 1-5.
Compound 1 was obtained as an amorphous solid. The
molecular formula was assigned as C20H32O on the basis
of the [M + H]+ ion that appeared at m/z 289.2537 in
HRCIMS experiments. The IR spectrum showed an absor-
bance at 3425 cm-1, consistent with a hydroxyl group. The
1H NMR spectrum showed three methyl singlets at ? 0.80,
0.89, and 1.73, which were assigned to positions C-19, C-20,
and C-17, respectively. Two doublets at ? 3.12 (1H, J ) 11
Hz) and 3.39 (1H, J ) 11 Hz) indicated the presence of
two diastereotopic protons on a hydroxyl-bearing carbon.
1H NMR spectra also showed the presence of olefinic
protons at ? 4.95 (1H, d, J ) 11 Hz), 5.03 (1H, d, J ) 17
Hz), and 6.80 (1H, dd, J ) 11, 17 Hz), consistent with a
terminal double bond. The 13C NMR experiments showed
signals for 20 carbon atoms, four of which were olefinic
* To whom correspondence should be addressed. Tel: (507) 681 5371.
Fax: (507) 264 4450. E-mail: lucr@ancon.up.ac.pa.
? University of Panama.
? Florida State University-Panama.
? Smithsonian Tropical Research Institute.
928 J. Nat. Prod. 2003, 66, 928-932
10.1021/np030010o CCC: $25.00 ? 2003 American Chemical Society and American Society of Pharmacognosy
Published on Web 06/20/2003
carbons at ? 111.0, 128.9, 135.9, and 137.2 and one signal
for a hydroxyl-bearing carbon at ? 72.6. Carbon-bound
protons were assigned from HSQC 2D NMR spectral data.
Using HMBC 2D NMR data, the methyl group at ? 1.73
was assigned to C-17, on the basis of the correlation of its
protons to C-8, C-13, and C-14. The hydroxyl-bearing
methylene group was assigned to C-18 on the basis of the
correlation of the carbon-bound protons to C-3, C-4, and
C-5. The NOESY spectrum of 1 showed a correlation
between H-18a (? 3.12) and H-18b (? 3.39) with H-5 (? 1.20)
and H-19 (? 0.80). Correlations between H-15 (? 6.80) and
H-16 (? 4.95 and 5.03) and H-17 (? 1.73) were observed, in
addition to correlations between the protons of H-16 and
H-12 (? 2.02 and 2.31). These data correspond to a
diterpene with a cassane skeleton with no additional
substituents on the A, B, or C rings.12 The absence of any
NOE correlation between H-20 and the protons on C-18
was consistent with the placement of the hydroxymethyl-
ene group at C-4 in the R position, an observation consis-
tent with the NOEs between the H-18 protons and H-5.
Compound 2 was also obtained as an amorphous solid.
The molecular formula was assigned as C20H32O2 on the
basis of the [M + H]+ ion that appeared at m/z 305.2470
in the HRCIMS, 16 amu greater than 1. The IR spectrum
showed a broad band at 3433 cm-1 consistent with a
hydroxyl moiety. The 1H NMR spectrum of 2 was very
similar to that of compound 1, the major difference being
a resonance for a single proton at ? 4.38 in the spectrum
of compound 2, consistent with its attachment to a hy-
droxyl-bearing carbon. The 13C NMR spectrum showed
signals for 20 carbon atoms, including one resonance at ?
68.1 suggesting a hydroxyl-bound carbon. The HSQC
spectrum revealed that H-6 (? 4.38) was attached to C-6
(? 68.1), an observation consistent with the HMBC cor-
relations of ? 4.38 to C-4 and C-5. The NOESY spectrum
of 2 showed correlations between H-6 and H-7a, H-18a, and
H-18b. Irradiation of the signal at ? 4.38 in a differential
NOE experiment showed correlations with protons at H-5,
H-7a, and H-18a. The absence of any NOE correlation
between protons at H-6, H-19, and H-20 suggests that the
C-6 bound hydroxyl group is in the ? position, consistent
with the NOEs observed between the H-6 and H-18
protons.
To confirm the structure of 2, two derivatives were
synthesized. Oxidation of 2 with Jones reagent yielded a
mixture of products from which ketone 2a was isolated in
14% yield. The IR spectrum of derivative 2a showed the
presence of hydroxyl and carbonyl groups. The 1H NMR
signal at ? 4.38 in the 1H NMR of 2, assigned to position
6, was no longer evident in 2a. The 13C NMR spectrum of
2a showed 20 carbon signals with one signal at ? 211.8,
consistent with the carbonyl carbon of a cyclohexanone. The
assignment of the carbonyl group to position C-6 was based
upon correlations to protons H-5 (? 2.40), H-7a (? 2.67),
and H-8 (? 2.44) observed in HMBC experiments. The
NOESY spectrum showed correlation between H-18a and
H-18b with H-19 (? 1.17). NOE correlations between
protons H-18 with H-20 were not observed. Irradiation of
the H-18 protons in a differential NOE experiment showed
a correlation only with H-19, providing additional support
for the proposed relative configuration at C-4. Irradiation
of H-20 (? 0.84) showed correlations with C-19 methyl
protons (? 1.17) and H-8 (? 2.44).
Treatment of 2 with pyridine and acetic anhydride
yielded compound 3 (61%) and the diacetate 2b (11%). The
IR spectrum of 2b showed a broad band at 1730 cm-1,
consistent with the carbonyl groups of the two acetyl
residues. The 1H NMR spectrum indicated the presence of
two methyls, also assigned to the acetyl moieties. The 1H
NMR spectrum revealed a doublet at ? 5.39 (J ) 2.4 Hz),
which was assigned to the proton bound to C-6. Irradiation
at H-6 in a differential NOE experiment showed correla-
tions with H-18, H-7a, and H-5, while irradiation of the
H-18 proton showed correlations with the H-6, H-5, H-19,
and H-18 protons. These results confirmed the relative
configurations at C-4 and C-6. The 13C NMR indicated the
presence of 24 carbons, including resonances for carbonyl
carbons and methyl groups associated with the acetate
moieties.
Compound 3 (first isolated as a natural product and then
as a product of acetylation from 2) was isolated as an
amorphous solid. The molecular formula was assigned as
C22H34O3 on the basis of HRCIMS data. The IR, 1H NMR,
and 13C NMR were consistent with the presence of an acetyl
moiety. The HMBC experiment showed correlations of the
diastereotopic protons attached to C-18 with the carbonyl
carbon of the acetyl moiety and with C-19. The protons
attached to C-19 also showed correlations with the C-18
protons and to the quaternary carbon at C-4. These data
are consistent with the attachment of the acetyl group to
C-18.
Compound 4 showed an apparent molecular ion of m/z
360.2300, consistent with the formula C22H32O4. The IR
spectrum showed absorbances consistent with a carboxylic
acid moiety. The 1H NMR spectrum showed the presence
of two methyl groups. The 13C NMR spectrum showed two
signals for carbonyl carbons, one associated with a car-
boxylic acid moiety at ? 180.5 and the other assigned to
an acetyl group at ? 171.5. HMBC data showed correlations
between the carbonyl of the carboxylic acid group with
H-18a (? 4.45), H-18b (? 3.92), H-5 (? 1.25), and H-3a (?
1.01), consistent with the carboxyl group at C-4. The
absence of any detectable NOE between H-18 and H-20 was
consistent with the relative configuration shown.
Compound 5 was assigned a molecular formula of
C22H34O4 on the basis of HRCIMS data. The IR spectrum
showed absorbances consistent with a hydroxyl group and
the carbonyl carbon of an acetyl moiety. The 1H NMR
spectrum showed evidence for two terminal double bonds.
One terminal double bond showed an ABX system where
the X part of the system was at ? 6.09 (1H, dd, J ) 11 and
15 Hz) and the AB part of the system was at ? 5.37 (1H,
dd, J ) 1.3 and 11 Hz) and 5.38 (1H, dd, J ) 1.3 and 15
Hz). The other terminal double bond gave an AB system
at ? 4.92 (1H, d, J ) 1.7 Hz) and 5.08 (1H, d, J ) 1.7 Hz).
The 1H NMR spectrum showed signals for three methyl
groups. The 13C NMR showed 22 signals including four
olefinic carbons, comprising one quaternary carbon, two
exomethylene carbons, and one RdC(H)R? moiety. Com-
pound 5 also revealed one oxygen-bound quaternary carbon
at ? 86.0, subsequently assigned to C-13. The spectral data
were consistent with two double bonds, one between C-15
and C-16 as seen in compounds 1-4, and another between
C-14 and C-17. Correlations observed in HSQC and HMBC
experiments also supported the proposed structure for
compound 5. NOESY experiments showed correlations
between the H-17 proton at ? 4.92 and both C-7 protons (?
1.51, 1.90, dt, J ) 3 and 13 Hz), as well as correlations
between H-15 with H-12 (? 1.58), H-16 (? 5.37), and the
H-17 proton at ? 5.08. These data are consistent with a
hydroxyl group on C-13 in the ? position.
The anti-trypanosomal activities of cassane diterpenes
1-5 are presented in Table 3. Bioassay-guided fraction-
ation of M. frutescens employed the extracellular (epimas-
Cassane Diterpenes from Myrospermum Journal of Natural Products, 2003, Vol. 66, No. 7 929
tigote) stage of T. cruzi. However, T. cruzi are obligatory
intracellular organisms and require a host cell in order to
survive and multiply in the mammalian host, and the
intracellular forms (trypomastigote and amastigote) are
those that are responsible for disease manifestations in
humans.13 Compounds 3 and 5 were more active against
the extracellular form of the parasite (11 and 16 ?M,
respectively) than the intracellular form, while compounds
1 and 2 were more active against the more clinically
relevant intracellular form of the parasite (17 and 16 ?M,
respectively). Compound 4 showed no significant activity
against the extracellular form of the parasite (IC50 ) 104
?M) and was not isolated in sufficient quantity for testing
against the intracellular forms of the parasite.
To accurately interpret the results of the intracellular
assay, it was necessary to determine whether the measured
toxicity of a compound resulted from the killing of the
parasite or from simply killing the host cells. Accordingly,
we determined the cytotoxicity of these compounds against
mammalian cell lines using primary human foreskin
fibroblasts (HFF), the cell line routinely employed to
culture the intracellular forms of T. cruzi.14 HFF cells were
grown in 96-well culture plates and incubated in the
presence and absence of the pure compounds for 6 days,
followed by the addition of MTT, a reagent that is actively
metabolized by living cells and routinely used to determine
cell viability.15 In the case of compounds 1 and 2, the
intracellular form of the parasite was approximately 9-fold
more sensitive to the compounds than were the HFF cell
lines (Table 3). The enhanced toxicity of compounds 1 and
2 toward the intracellular form of the parasite as compared
to the HFF cell line is particularly relevant to the develop-
ment of new drugs for treating Chagas? disease, as it
involves the more clinically and biologically relevant form
of the parasite for which there are currently no effective
treatments. Compounds 3 and 5 showed approximately 8-
and 10-fold greater effect, respectively, on the extracellular
form of the parasite than on the HFF cells. Efforts are
currently underway in our laboratories to better under-
stand how substituents on the cassane skeleton affect the
potency and specificity of their anti-trypanosomal activity.
Table 1. 1H NMR Data for Compounds 1-5a
proton 1 2 3 4 5
1 0.90 br 0.89 br 0.90 br 0.91 br 0.90 br
1.74 br 1.71 br 1.75 br 1.79 br 1.75 br
2 1.52 br 1.52 br 1.51 br 1.17 br 1.62 br
1.65 br 1.69 br 1.68 br 1.50 br
3 1.28 br 1.17 br 1.33 br 1.01 br 1.41 br
1.36 br 1.41 br 2.23 br 1.52 br
5 1.20 br 1.20 br 1.18 br 1.25 br 1.22 br
6 1.30 br 4.38 d (2.3) 4.27 d (1.7) 1.78 br 4.36 d (1.7)
1.55 br
7 0.89 br 1.19 br 1.19 br 2.16 br 1.51 br
2.13 br 2.13 dt (3,13) 2.13 dt (3,13) 1.90 dt (3,13)
8 2.02 br 2.40 br 2.42 br 1.93 br 2.71 br
9 0.96 br 0.92 br 0.92 br 0.84 br 0.92 br
11 0.96 br 1.70 br 1.18 br 1.78 br 1.52 br
1.70 br 2.02 br 1.85 br 1.69 br
12 2.02 br 1.99 br 2.03 br 2.25 br 1.58 br
2.31 br 2.31 br 2.34 br 2.22 dt (3,13)
15 6.80 dd (11,17) 6.78 dd (11,17) 6.77 dd (11,17) 6.79 dd (11,17) 6.09 dd (11,15)
16 4.95 d (11) 4.93 d (11) 4.94 d (11) 4.95 d (11) 5.37 dd (1.3,11)
5.03 d (17) 5.08 d (17) 5.10 d (17) 5.09 d (17) 5.38 dd (1.3,15)
17 1.73 s 1.72 s 1.73 s 1.66 s 4.92 d (1.7)
5.08 d (1.7)
18 3.12 d (11) 3.19 d (11) 3.69 d (11) 3.92 d (10) 3.73 d (11)
3.39 d (11) 3.51 d (11) 4.02 d (11) 4.45 d (10) 4.04 d (11)
19 0.80 s 1.16 s 1.25 s 1.30 s
20 0.89 s 1.15 s 1.19 s 0.72 s 1.25 s
CH3CO-18 2.02 s 1.97 s 2.06 s
a All spectra were recorded in CDCl3, 300 MHz at 27 ?C using TMS as internal reference (? in ppm, J are given in Hz).
Table 2. 13C NMR Data for Compounds 1-5a
carbon 1 2 3 4 5
1 38.6 40.4 41.2 38.8 41.1
2 18.7 18.3 18.7 19.2 19.3
3 35.5 38.4 38.3 33.1 37.6
4 38.0 36.0 37.6 48.2 37.8
5 48.9 51.2 51.6 51.7 50.4
6 22.5 68.1 68.2 24.5 67.2
7 31.8 40.4 40.5 32.0 39.9
8 41.7 37.6 36.4 41.6 32.0
9 54.1 54.4 55.0 53.8 56.6
10 36.9 36.9 37.6 37.4 37.3
11 21.6 21.1 21.6 27.0 18.3
12 26.9 26.3 26.8 26.9 33.1
13 128.9 129.1 129.7 129.0 86.0
14 137.2 136.2 136.6 136.8 151.7
15 135.9 135.4 135.8 135.8 139.4
16 111.0 110.5 111.2 111.3 111.5
17 16.1 15.8 16.3 16.1 115.7
18 72.6 72.3 73.2 72.7 73.9
19 17.7 19.7 20.1 180.5 19.9
20 15.0 17.0 17.5 13.0 18.0
Ac-18 171.6 171.5 171.8
21.3 21.2 21.0
a ? in ppm, 75 MHz in CDCl3.
Table 3. Activities of Compounds 1-5 against Trypanosoma
cruzia
extracellular intracellular cytotoxicityb
compound IC50 (?M) IC50 (?M) IC50 (?M)
1 48.6 ( 15.8 17.4 ( 4.1 149
2 56 ( 3.8 16.6 ( 1.5 148
3 11.5 ( 8.4 25.9 ( 1.7 118
4 104 ( 0.74 NDc ND
5 16.5 ( 1.2 35.8 ( 0.4 124
amphotericin B 1 ( 0.12 ND ND
nifurtimox ND 11 ( 1.92 ND
a Results show the IC50 value ( the SD (n ) 3). Compounds 2a
and 2b were assayed in the extracellular T. cruzi assay and
showed IC50 values of 36 and 59 ?M, respectively. b Experiments
performed with human foreskin fibroblasts. c ND ) not deter-
mined.
930 Journal of Natural Products, 2003, Vol. 66, No. 7 Mendoza et al.
Experimental Section
General Experimental Procedures. Melting points were
determined using an Electrothermal 9100 apparatus and are
uncorrected. Optical rotations were measured on either JASCO
DIP-370 or Perkin-Elmer 141 polarimeters. IR spectra were
recorded on a Shimadzu FTIR-8300 spectrophotometer. The
NMR spectra were recorded on a Bruker Avance 300 spec-
trometer (300 MHz for protons and 75 MHz for carbon) with
TMS as an internal standard. Irradiation experiments were
recorded on Bruker AMX-500 and Bruker Avance 300 instru-
ments. HRCIMS were recorded on UG-Autospec or Kratos
MS50TC instruments.
Plant Material. Mature leaves of M. frutescens were
collected from Barro Colorado Nature Monument in Gatu?n
Lake in the Republic of Panama in January 2001. The material
was identified by Professor Mireya Correa of the University
of Panama and the Smithsonian Tropical Research Institute.
Vouchers of the plants have been deposited in the herbarium
of University of Panama (PMA 50913 and PMA 48733).
Extraction and Isolation. Upon collection, mature leaves
were transferred to sealed plastic bags, kept on ice, and
processed within 6 h. After removal of the stems, 0.3 kg of
fresh leaves was homogenized in 30 g aliquots with 240 mL of
cold MeOH for 30 s in a Waring blender followed by treatment
with a Polytron homogenizer (Brinkmann Instruments) for at
least 2 min or until the suspension of leaf material was
homogeneous. The mixture was filtered under vacuum through
Whatman # 4 filter paper, and the marc was then washed with
150 mL of EtOAc. The MeOH and EtOAc fractions were
combined and filtered through Whatman # 1 filter paper. The
extract was concentrated by rotary evaporation, yielding a total
of 116 g of crude extract, which was stored at -80 ?C until
further use. In 15 g aliquots, the crude extract was dissolved
in 500 mL of H2O and extracted with CH2Cl2 (4  250 mL).
The CH2Cl2 was removed under reduced pressure, and the
extract was dissolved in 250 mL of 1:9 H2O/MeOH, which was
then extracted with hexane (4  250 mL). Removal of MeOH
under reduced pressure yielded a total of 21 g of extract, which
was chromatographed in 5 g aliquots on a column of silica gel
60 (37-75 ?m, Geduran) eluting sequentially with 500 mL of
60:40 CH2Cl2/EtOAc, 200 mL of EtOAc, 300 mL of 1:1 EtOAc/
MeOH, and 250 mL of MeOH. The fractions were combined
according to composition by TLC into four main fractions
(fractions 1-4). Fraction 1 (1.5 g) was subjected to column
chromatography with silica gel (7GF, VWR Scientific) and
eluted with 200 mL of CH2Cl2 and 200 mL of CH2Cl2/EtOAc,
4:1, to yield compounds 1 (6.0 mg), 3 (292.0 mg), and 5 (9.0
mg). Fraction 2 (5.6 g) was chromatographed on a column of
silica gel 60 (63-200 ?m) and eluted sequentially with hexane/
EtOAc in the following proportions: 90:10 (200 mL), 85:15 (500
mL), 80:20 (400 mL), 70:30 (500 mL), 60:40 (300 mL), and 40:
60 (300 mL), followed by 300 mL of EtOAc, 300 mL of 1:1
EtOAc/MeOH, and 300 mL of MeOH, which were pooled into
six fractions (fractions 1a-6a). Fraction 3a (2.04 g) was
subjected to silica gel column chromatography (7GF) and
eluted with 250 mL of 60:20:20 hexane/toluene/acetone to yield
compound 2 (392.0 mg). Fraction 3a was also subjected to
preparative TLC (Whatman PK5F silica gel 150 A plates)
followed by elution with a 40:40:20 mixture of hexane/CHCl3/
THF, yielding compounds 2 (14.0 mg) and 4 (17.0 mg).
18-Hydroxycassan-13,15-diene (1): amorphous powder;
[R]25D -49? (c 0.0015, MeOH); IR (CHCl3) ?max 3425, 2928,
2866, 1631, 1385 cm-1; for 1H NMR and 13C NMR data, see
Tables 1 and 2; HRCIMS (CH4) m/z 289.2537 [M + H]+ (calcd
for C20H33O, 289.2531) 288 [M+] (45), 271 (100), 259 (22), 203
(20), 163 (47), 107 (73).
6?,18-Dihydroxycassan-13,15-diene (2): white solid; mp
128-130 ?C; [R]25D -131? (c 0.0037, MeOH); IR (CHCl3) ?max
3433, 2928, 1645, 1254, 1042 cm-1; for 1H NMR and 13C NMR
data, see Tables 1 and 2; HRCIMS (CH4) m/z 305.2470 [M +
H]+ (calcd for C20H33O2, 305.2480), 305 (3), 287 (25), 269 (18),
257 (11), 231 (6), 116 (30), 78 (16), 60 (42), 28 (100).
Oxidation of 2. Compound 2 (176 mg) was dissolved in
acetone (previously distilled over KMnO4), and Jones reagent
was added dropwise and stirred at room temperature until
color persisted, at which time the reaction was quenched with
MeOH, extracted with CH2Cl2, and dried over anhydrous Na2-
SO4. The products were purified by column chromatography
on silica gel 60 (37-75 ?m) and eluted with CHCl3, yielding
ketone 2a (26 mg, 14%): amorphous solid, mp 96-98 ?C; [R]24D
-22.8 (c 0.86, CHCl3); IR (CHCl3) ?max 3433, 2930, 2868, 1705,
1231, 1041 cm-1; 1H NMR (CDCl3, 300 MHz) ? 6.77 (1H, dd, J
) 11, 17 Hz, H-15), 5.15 (1H, d, J ) 17 Hz, H-16a), 5.01 (1H,
d, J ) 11 Hz, H-16b), 3.51 (1H, d, J ) 11 Hz, H-18a), 3.17
(1H, d, J ) 11 Hz, H-18b), 2.67 (1H, d, J ) 4.2 Hz, H-7a), 2.40
(1H, br, H-5), 2.12 (1H, br, H-7b), 1.69 (3H, s, H-17), 1.45 (1H,
br, H-9); 13C NMR (CDCl3, 75 MHz) ? 211.8 (CdO, C-6), 134.0
(CH, C-15), 133.1 (C, C-14), 128.4 (C, C-13), 110.9 (CH2, C-16),
71.4 (CH2, C-18), 60.5 (CH, C-5), 52.3 (CH, C-9), 46.8 (CH2,
C-7), 41.6 (CH, C-8), 40.3 (C, C-10), 37.2 (CH2, C-1), 36.0 (C,
C-4), 35.3 (CH2, C-3), 25.4 (CH2, C-12), 20.5 (CH2, C-11), 16.9
(CH2, C-2), 16.7 (CH3, C-19), 15.0 (CH3, C-20), 14.4 (CH3, C-17);
HRCIMS (CH4) m/z 302.2241 [M+] (calcd for C20H30O2,
302.2245), 302 [M+] (91), 284 (64), 269 (94), 257 (28), 229 (32),
203 (18), 167 (39), 149 (34), 137 (44), 123 (100), 109 (83), 91
(81).
Acetylation of 2. To 30 mg of 2 were added 2 mL of acetic
anhydride and 2 mL of pyridine, which was stirred overnight
at room temperature. Ice was added to the mixture, which was
then extracted with ether. The ether layer was washed
successively with 2 N HCl, 6% Na2CO3, and water and then
dried over anhydrous Na2SO4. After removal of solvent by
rotary evaporation the residue (27 mg) was purified by column
chromatography on silica gel (7GF) which was eluted with
CHCl3 to yield the diacetate 2b (4 mg, 13%) and the monoac-
etate 3 (22 mg, 73%). Diacetate 2b: [R]24D -24.1 (c 0.59,
CHCl3); IR (CHCl3) ?max 3020, 2928, 2858, 1730, 1246, 1217,
1035 cm-1; 1H NMR (CDCl3, 300 MHz) ? 6.81 (1H, dd, J ) 11,
17 Hz, H-15), 5.39 (1H, d, J ) 2.4 Hz, H-6), 5.14 (1H, d, J )
17 Hz, H-16a), 4.99 (1H, d, J ) 11 Hz, H-16b), 3.86 (1H, d, J
) 11 Hz, H-18a), 3.76 (1H, d, J ) 11 Hz, H-18b), 2.29 (1H, br,
H-7a), 1.72 (3H, s, H-17), 1.35 (1H, br, H-5), 1.15 (1H, br, H-7b),
1.01 (1H, br, H-9); 13C NMR (CDCl3, 75 MHz) ? 171.4 (MeCO-
18), 170.6 (MeCO-6), 135.9 (C, C-14), 135.7 (CH, C-15), 129.9
(C, C-13), 111.5 (CH2, C-16), 73.2 (CH2, C-18), 70.8 (CH, C-6),
54.6 (CH, C-9), 51.4 (CH, C-5), 40.7 (CH2, C-1), 38.4 (CH2, C-3),
37.8 (C, C-4), 37.5 (C, C-10), 37.0 (CH, C-8), 36.9 (CH2, C-7),
26.8 (CH2, C-12), 22.2 (CH3CO-6), 21.6 (CH2, C-11), 21.3 (CH3-
CO-18), 19.4 (CH3, C-19), 18.5 (CH2, C-2), 17.1 (CH3, C-20),
16.3 (CH3, C-17); HRCIMS (CH4) m/z 387.2520 [M + H]+ (calcd
for C24H35O4, 387.2535), 386 [M+] (15), 362 (9), 344 (20), 328
(100), 301 (10), 270 (30), 253 (40), 229 (16), 185 (24), 159 (39),
145 (39), 135 (61), 107 (55), 95 (59).
6?-Hydroxy-18-acetoxycassan-13,15-diene (3): white solid,
mp 127-130 ?C; [R]25D -75 (c 0.00325, MeOH); IR (CHCl3) ?max
3516, 2860, 1724, 1628, 1038 cm-1; for 1H NMR and 13C NMR
data, see Tables 1 and 2; HRCIMS (CH4) m/z 347.2583 [M +
H]+ (calcd for C22H35O3, 347.2586), 347 [M + H]+ (2), 329 (64),
287 (32), 269 (100), 146 (11), 107 (17), 61 (78), 45 (52), 29 (57),
19 (84).
18-Acetoxy-13,15-diene-19-cassanoic acid (4): white solid;
85-88 ?C; [R]24D -11.1 (CHCl3, c 0.45); IR (CHCl3) ?max 3680,
2937, 1730, 1244, 1037 cm-1; for 1H NMR and 13C NMR, see
Tables 1 and 2; HRCIMS (CH4) m/z 360.2300 [M]+ (calcd for
C22H32O4, 360.2300) 360 [M]+ (89), 346 (13), 300 (89), 255 (35),
243 (87), 227 (28), 187 (38), 159 (49), 135 (79), 105 (98), 91
(100).
6?,13?-Dihydroxy-18-acetoxycassan-14(17),15-diene (5):
amorphous powder; [R]25D 19? (MeOH, c 0.000733); IR (CHCl3)
?max 3423, 2868, 1720, 1645, 1036 cm-1; for 1H NMR and 13C
NMR, see Tables 1 and 2; HRCIMS (CH4) m/z 363.2507 [M +
H]+ (calcd for C22H35O4, 363.2490), 363 [M + H]+ (42), 301 (41),
284 (49), 259 (72), 241 (70), 229 (18), 146 (65), 129 (100), 113
(32), 83 (24), 71 (35).
Intracellular T. cruzi Bioassay. The recombinant Tu-
lahuen clone C4 of T. cruzi that expresses ?-galactosidase
(?Gal) as a reporter enzyme was used in the assay.12 Trypo-
mastigotes were grown on monolayers of human foreskin
fibroblasts (HFF) as previously described.13 Both the parasite
Cassane Diterpenes from Myrospermum Journal of Natural Products, 2003, Vol. 66, No. 7 931
and the HFF were grown in RPMI-1640 medium without
phenol red (Gibco BRL) plus 10% fetal calf serum. Briefly, 96-
well tissue culture plates (Costar) were seeded with HFF at 1
 103 per well in 100 ?L volumes and incubated overnight.
?Gal-expressing trypomastigotes were then added at 1  103
per well in 50 ?L volumes. After incubation overnight, the drug
or test compounds were added in serial dilutions in 50 ?L
volumes. The final concentrations of the test samples after
parasite addition are 50, 10, and 2 ?g/mL in 200 ?L of media.
At 7 days after infection, the assays were developed by addition
of the chlorophenolred-?-D-galactopyranoside (CPRG) to a final
concentration of 100 ?M and Nonidet P-40 to a final concen-
tration of 0.1%. Plates were incubated for 4-6 h at 37 ?C, and
the colorimetric reaction was quantified with a Benchmark
microplate reader (BIORAD) at an optical density of 570 nm.
Negative controls were identical to the conditions described
above but with DMSO alone at final concentrations identical
to those employed with test substances. The final DMSO
concentration did not exceed 0.1%, which has no measurable
affect on parasite growth (data not shown). The absorbance
observed at 570 nm from the negative control was subtracted
from the IC50 value determined in the presence of test
substance. Each test substance as well as control experiment
was tested in duplicate, and the results represent a minimum
of three separate experiments which were used to calculate
the standard deviation (SD). Results are expressed as IC50,
the concentration of compound that inhibited growth of the
parasites by 50% ( SD. Nifurtimox was chosen as a positive
control for the intracellular form of T. cruzi, as it is the drug
routinely used for the treatment of acute Chagas? disease in
humans.13
Extracellular T. cruzi Bioassay. Recombinant epimas-
tigotes were grown in liver infusion tryptone (LIT) medium
supplemented with 10% calf serum at 28 ?C as previously
described.13 Cultures were initiated with a cell density of 1 
105 epimastigotes per mL and were incubated in the presence
of test compounds. Three days after seeding of the culture with
parasite, the epimastigotes were treated with CPRG and
Nonidet P-40 as described above for the intracellular T. cruzi
bioassay, followed by quantification by measuring the optical
density at 570 nm. Amphotericin B was employed as a positive
control for the extracellular form of T. cruzi.13 Inhibition of
the parasites by test compounds or controls is presented as
IC50 as described above for the intracellular assay.
Cytotoxicity Tests with Human Foreskin Fibroblasts.
Primary human foreskin fibroblasts (HFF) adhered to 96-well
plates were used to evaluate the toxicity of the compounds
purified from M. frutescens on the basis of the reduction of
(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide14
(MTT, Sigma). After treatment with the test compounds, cell
viability was evaluated after a 3 day incubation at 37 ?C in
CO2. Cytotoxicity to the cell lines is presented as IC50 as
described above for the intracellular T. cruzi assay.
Acknowledgment. The authors extend special thanks to
Drs. Phyllis D. Coley and Thomas A. Kursar, who were
instrumental in establishing the present bioprospecting pro-
gram in Panama. Funding comes from the International
Cooperative Biodiversity Groups Program (grant number
#IUOI TWO1021-01 to PC) from the NIH, National Science
Foundation, and U.S. Department of Agriculture. We thank
F. Buckner from the University of Washington, Seattle, for
the recombinant Tulahuen clone C4 of T. cruzi. We gratefully
acknowledge the Smithsonian Tropical Research Institute
Equipment Fund for partial funding for the Bruker Avance
NMR. We thank Dr. Kerry McPhail from the Department of
Pharmacy of Oregon State University for expert assistance
with MS and NMR experiments. We also thank Dr. Ricardo
Riguera from the Department of Organic Chemistry of the
Universidad de Santiago de Compostela for assistance with
NMR experiments.
References and Notes
(1) Kursar, T. A.; Capson, T. L.; Coley, P. D.; Corley, D. G.; Gupta, M.
P.; Harrison, L. A.; Ortega-Barr??a, E.; Windsor, D. M. Pharm. Biol.
1999, 37, 114-126.
(2) World Bank. World Development Report 1993. Investing in Health;
Oxford University Press: New York, 1993; p 329.
(3) Ortega-Barr??a, E. Trypanosoma species. In Principles and Practice
of Pediatric Infectious Disease, 2nd ed.; Long, S., Prober, C., Pickering,
L., Eds.; Churchill Livingstone: London, 2003; Chapter 288, pp 1324-
1330.
(4) Vieira, H. S.; Takahashi, J.; De Oliveira, A. B.; Chiari, E.; Boaventura,
M. D. J. Brazilian Chem. Soc. 2002, 13, 151-157.
(5) Da Costa, F. B.; Albuquerque, S.; Vichnewski, W. Planta Med. 1996,
62, 557-559.
(6) Alves, T. M. A.; Chaves, P. G.; Santos, L. M. S. T.; Nagem, T. J.;
Murta, S. M.; Ceravolo, I. P.; Romanha, A. J.; Zani, C. L. Planta Med.
1995, 61, 85-87.
(7) Ragasa, C. Y.; Hofilen?a, J. G.; Rideaout, J. A. J. Nat. Prod. 2002, 65,
1107-1110.
(8) Lider, D. L.; Peter, S. R.; Tinto, W. F.; Visada, S. M.; McLean, S.;
Reynolds, W. F. J. Nat. Prod. 1998, 61, 1462-1465.
(9) Roengsumran, S.; Limsuwankesorn, S.; Ngamrojnavanich, N.; Pet-
som, A.; Chaichantipyuth, C.; Ishikawa, T. Phytochemistry 2000, 53,
841-844.
(10) Popoff, S.; Siino, M. R.; de Lima, H. Rev. Soc. Venezolana Qu??m. 1960,
31, 31-36.
(11) Gunstone, F. D.; Steward, S. R.; Cornelius, J. A.; Hammonds, T. W.
J. Sci. Food Agr. 1972, 23, 53-60.
(12) Jiang, R. W.; Ma, S.; But, P. P.; Mak, T. C. W. J. Nat. Prod. 2001,
64, 1266-1272.
(13) Buckner, F. S.; Verlinde, C. L. M. J.; La Flamme, A. C.; Van Voorhis,
W. C. Antimicrob. Agents Chemother. 1996, 40, 2592-2597.
(14) Chiari, E.; Camargo, E. P. In Genes and Antigens of Parasites. A
Laboratory Manual; Morel, C. M., Ed.; Fundaca?o Oswaldo Cruz: R??o
de Janeiro, Brazil, 1984; pp 23-28.
(15) Kilani, R. T.; Tamimi, Y.; Karmali, S.; Mackey, J.; Hanel, E. G.; Wong,
K. K.; Moore, R. B. Anticancer Drugs 2002, 13, 557-566.
NP030010O
932 Journal of Natural Products, 2003, Vol. 66, No. 7 Mendoza et al.