Triterpenoids as inhibitors of erythrocytic and liver stages of Plasmodium infections

Article abstract of DOI:10.1016/j.bmc.2011.10.044

Bioassay-guided fractionation of the methanol extract of Momordica balsamina led to the isolation of two new cucurbitane-type triterpenoids, balsaminol F (1) and balsaminoside B (2), along with the known glycosylated cucurbitacins, cucurbita-5,24-diene-3β

Full text of DOI:10.1016/j.bmc.2011.10.044

Bioorganic & Medicinal Chemistry 19 (2011) 7474–7481
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Triterpenoids as inhibitors of erythrocytic and liver stages
of Plasmodium infections
Cátia Ramalhete a, Filipa P. da Cruz b, Dinora Lopes c, Silva Mulhovo d, Virgílio E. Rosário c,
Miguel Prudêncio b, Maria-José U. Ferreira a,
⇑
a iMed.UL, Faculdade de Farmácia, Universidade de Lisboa, Av. Prof. Gama Pinto, 1649-003 Lisboa, Portugal
b Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa, 1649-028 Lisboa, Portugal
c CMDT.LA, Instituto de Higiene e Medicina Tropical, UNL, R. da Junqueira 96, 1349-008 Lisbon, Portugal
d Escola Superior Técnica, Departamento de Ciências Agro-Pecuárias, Universidade Pedagógica, Campus de Lhanguene, Mozambique
a r t i c l e i n f o
a b s t r a c t
Article history:
Bioassay-guided fractionation of the methanol extract of Momordica balsamina led to the isolation of
two new cucurbitane-type triterpenoids, balsaminol F (1) and balsaminoside B (2), along with the
known glycosylated cucurbitacins, cucurbita-5,24-diene-3b,23(R)-diol-7-O-b-D-glucopyranoside (3)
Received 2 August 2011
Revised 7 October 2011
Accepted 14 October 2011
Available online 20 October 2011
and kuguaglycoside A (4). Compound 1 was acylated yielding two new triesters, triacetylbalsaminol
F (5) and tribenzoylbalsaminol F (6). The structures were elucidated based on spectroscopic methods
including 2D-NMR experiments (COSY, HMQC, HMBC and NOESY). Compounds 1–6, were evaluated
for their antimalarial activity against the erythrocytic stages of the Plasmodium falciparum chloro-
quine-sensitive strain 3D7 and the chloroquine-resistant clone Dd2. Assessment of compounds (1–3
and 5, 6) activity against the liver stage of Plasmodium berghei was also performed, measuring the
luminescence intensity in Huh-7 cells infected with a firefly luciferase-expressing P. berghei line,
PbGFP-Luccon. Active compounds were shown to inhibit the parasite’s intracellular development rather
than its ability to invade hepatic cells. Toxicity of compounds (1–3 and 5, 6) was assessed on the same
cell line and on mouse primary hepatocytes through the fluorescence measurement of cell confluency.
Furthermore, toxicity of compounds 1–6 towards human cells was also investigated in the MCF-7
breast cancer cell line, showing that they were not toxic or exhibited weak toxicity. In blood stages
of P. falciparum, compounds 1–5 displayed antimalarial activity, revealing triacetylbalsaminol F (5)
Keywords:
Antimalarials
Plasmodium
Schizontocidal activity
Triterpenoids
Cucurbitacins
Momordica balsamina
the highest antiplasmodial effects (IC50 values: 0.4 lM, 3D7; 0.2 lM, Dd2). The highest antiplasmodial
activity against the liver stages of P. berghei was also displayed by compound 5, with high inhibitory
activity and no toxicity.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
also of great concern. To date, drug resistance has been reported
in three malaria species known to affect humans, P. falciparum,
Malaria is still a major global threat though some success exists
in control programmes in several locations. It causes approxi-
mately 225 million incident infections per year, resulting in nearly
one million deaths, 91% of which were in Africa. About 85% of the
deaths occur in children under 5 years of age.1 The disease is
caused by protozoan parasites of the genus Plasmodium, and is
transmitted by mosquitoes of the genus Anopheles. Among the five
Plasmodia species affecting humans, Plasmodium falciparum is the
most prevalent worldwide, particularly in Africa, causing the most
severe form of the disease.1
Plasmodium vivax and Plasmodium malariae.2 However, the increas-
ing prevalence of drug-resistant P. falciparum strains represents the
greatest challenge in malaria control. Therefore, in order to over-
come drug-resistance, there is broad consensus that new antima-
larial drugs, that exhibit antimalarial efficacy, are urgently
needed. The discovery of new antimalarials by exploring natural
products is considered of particular interest.3,4 Natural product-
derived compounds have played a major role in drug discovery
and development.5–7 In the case of malaria drug discovery, the
great significance of plant-derived drugs for the treatment of the
disease is highlighted by quinine, artemisinin and their derivatives,
which are currently the mainstay of the antimalarial therapy.7
Most of the available antimalarial agents target blood stage par-
asites and only a limited number of drugs act on liver stages, the
clinically silent stage that obligatorily precedes blood infection.
In fact, presently, the only drug clinically approved by FDA against
In the absence of an effective vaccine, treatment and control of
malaria is more complex due to the emergence of drug-resistant
parasites. The spread of insecticide-resistant vectors of malaria is
⇑
Corresponding author. Tel.: +351 21 7946475; fax: +351 21 7946470.
E-mail address: mjuferreira@ff.ul.pt (M.-J.U. Ferreira).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.10.044

C. Ramalhete et al. / Bioorg. Med. Chem. 19 (2011) 7474–7481
7475
Table 1
Plasmodium liver stages, including dormant forms of the parasite, is
1H (400 MHz), 13C (150 MHz) NMR data for compounds 1 and 2
the 8-aminoquinoline primaquine, which presents several adverse
effects.8 The study of Plasmodium liver stage development has been
hampered by limitations in the experimental approaches required
to quantify hepatocyte infection by the parasite. Therefore, the
development of new drugs targeting Plasmodium liver stages repre-
sents an important and under-exploited site of intervention.9,10
Momordica balsamina L. (Cucurbitaceae), also referred to as the
balsam apple, or African pumpkin, is an extensively cultivated veg-
etable consumed in many tropical and subtropical regions of the
world.11,12 It has also been widely used in traditional medicine in
Africa to treat various diseases, mostly diabetes, and malaria symp-
toms.13–15
Bioassay-guided fractionation of the methanol extract of the
aerial parts of M. balsamina led to the isolation of several cucurbi-
tane-type triterpenoids with in vitro antimalarial activity against
blood schizonts of chloroquine-sensitive and -resistant strains of
P. falciparum.16,17
In this study, we report the isolation and structure elucidation
of four cucurbitacins (1–4) from the same plant, two of which
are new, along with two new ester derivatives (5–6). Compounds
1–6 were evaluated for their antimalarial activity against erythro-
cytic (blood) stages of P. falciparum (strains 3D7 and Dd2). Evalua-
tion of the activity of compounds 1–3 and 5, 6 against liver stages
of P. berghei infections was also carried out. The preliminary toxic-
ity toward human cells of compounds 1–6 was also evaluated in
the MCF-7 breast cancer cell line as well as in mouse primary
hepatocytes.
Position
1
2
a
dC
dH (J in Hz)
dC
dH (J in Hz)b
1
2
3
4
5
6
7
8
22.4
30.1
77.5
42.3
148.3
122.5
68.8
54.1
35.0
40.1
33.9
31.5
47.2
49.5
35.7
28.9
52.1
15.9
29.8
33.8
19.3
45.6
66.6
130.5
133.4
18.1
26.0
28.8
26.1
18.7
1.57 m; 1.70 m
1.71 m; 1.96 m
3.50 br s
–
22.3
30.1
77.4
42.4
148.7
122.1
74.4
49.9
35.1
40.1
33.7
31.5
47.3
49.2
35.7
28.8
52.3
16.1
29.4
33.8
19.3
45.7
66.6
130.5
133.3
18.2
26.0
28.8
26.1
18.8
1.57 m; 1.70 m
1.67 m; 1.97 m
3.49 br s
–
–
–
5.74 d (5.1)
3.93 br d (5.2)
1.98 br s
–
2.32 br d (7.0)
1.51 m; 1.89 m
1.54 m; 1.88 m
–
5.76 d (4.8)
4.11 br d (4.5)
2.13 br s
–
2.34 br d (9.3)
1.48 m; 1.70 m
1.58 m; 1.69 m
–
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
–
–
1.31 m; 1.40 m
1.39 m; 1.70 m
1.46 m
0.96 s
1.04 s
1.36 m
1.40 m; 1.91 m
1.48 m
0.99 s
1.03 s
1.48 m
1.54 m
0.97 d (6.4)
0.95 m; 1.63 m
4.41 td (9.6, 3.2)
5.16 d (8.5)
–
1.66 s
1.69 s
1.03 s
1.18 s
0.98 d (7.2)
1.82 m; 2.21 m
4.41 td (9.4, 2.8)
5.16 d (8.4)
–
1.67 s
1.70 s
1.03 s
1.18 s
0.74 s
0.75 s
2. Results and Discussion
2.1. Chemistry
a
13C NMR signals of 7-allose unit: 99.0 (C-10), 72.3 (C-20), 73.0 (C-30), 69.1 (C-40),
75.4 (C-50), 63.2 (C-60).
b
1H NMR signals of 7-allose unit: 4.75 (d, J = 8.0 Hz, H-10), 3.32 (m, H-20), 4.05 (br
s, H-30), 3.51 (m, H-40), 3.66 (m, H-50), 3.66 (m, H-60), 3.84 (br d, J = 9.4 Hz, H-60).
The air-dried powdered aerial parts of M. balsamina were
exhaustively extracted with methanol. Repeated bioassay-guided
chromatographic fractionation and further HPLC purification of
the ethyl acetate soluble fraction of the methanol extract yielded
the new cucurbitane-type triterpenoids 1–2. Acylation of com-
pound 1 gave rise to the new triester derivatives, triacetylbalsami-
nol F, (5) and tribenzoylbalsaminol F (6). Known glycosylated
cucurbitacins were also isolated. Their structures were assigned
the two-dimensional NMR data (1H–1H COSY, HMQC and HMBC-
Fig. 1), which allowed the clear assignment of all the carbons and
the location of the functional groups.
The relative stereochemistry of the tetracyclic nucleus of com-
pound 1 was characterized by a NOESY experiment (Fig. 2), taking
into account the coupling constants pattern of similar compounds
as
cucurbita-5,24-diene-3b,23(R)-diol-7-O-b-D-glucopyranoside
and assuming an
a orientation for H-10, characteristic of cucurbi-
tacins.20 Concerning the side chain, the configuration at C-23 was
assigned as R, mainly by comparison of the 13C NMR data of the
side chain carbons of 1 with those reported for a tetracyclic
triterpene, which stereochemistry was established by X-ray
(3), and kuguaglycoside A (4) based on their spectroscopic data
which were in good agreement with those described in the litera-
ture for these compounds.18,19
Compound 1, named balsaminol F, was obtained as an amor-
phous powder with ½a D26
ꢀ
+122 (c 0.12, MeOH). Its molecular for-
mula was established as C30H50O3, based on the HR-ESITOFMS
data [M+Na]+ at m/z 481.3649 (calcd. for C30H50O3Na 481.3652),
indicating the presence of six degrees of unsaturation. The strong
absorption band at 3396 cmꢁ1 in the IR spectrum of 1 revealed
the presence of hydroxyl groups corroborated by NMR signals cor-
responding to oxygenated methine protons and carbons [dH 3.50
(br s), dC 77.5; dH 3.93 (br d, J = 5.2 Hz), dC 68.8; dH 4.41 (td,
J = 9.6, 3.2)] (Table 1). Furthermore, the 1H NMR spectrum of 1 dis-
played signals due to seven tertiary methyl groups as singlets at dH
0.74, 0.96, 1.03, 1.04, 1.18, 1.66, 1.69, and a secondary methyl as a
doublet at dH 0.97 (J = 6.4 Hz). Vinylic NMR signals of two double
bonds, at dH 5.16 (d, J = 8.5 Hz), and 5.74 (d, J = 5.1 Hz), were also
found. The 13C NMR spectrum (Table 1) of 1 revealed 30 carbon sig-
nals, which were assigned by a DEPT experiment as eight methyls,
seven methylenes, nine methines (including three oxygenated and
two vinylic), and six quaternary carbons (two sp2).
HO
23
21
18
26
25
B
27
12
19
28
1
4
9
2
10
A
30
7
HO
OH
COSY
HMBC
29
A
(13C 1H)
The above data indicated a tetracyclic triterpenic scaffold for 1
carrying two hydroxyl groups in addition to the characteristic hy-
droxyl at C-3. Detailed structural information was obtained from
Figure 1. Key 1H–1H COSY and HMBC correlations of compound 1.

7476
C. Ramalhete et al. / Bioorg. Med. Chem. 19 (2011) 7474–7481
H3C
OH
18
CH3
27
CH3
H3C
H
25
OH
21
H3C
23
19
24
H
20
9
12
17
1
H
4
8
5
H3C
H
29
OH
7
CH3
30
28 CH3
H
H
H
Figure 2. Key NOESY correlations of compound 1.
diffraction.21 Identical configuration was previously found at C-23
in other cucurbitane derivatives isolated from the same
plant.16,17,22 Thus compound 1 was elucidated as cucurbita-5,24-
diene-3b,7b,23(R)-triol.
Compound 2, named balsaminoside B, was obtained as white
crystals (mp 230–232 °C). Its molecular formula, C36H60O8, was
established based on the pseudomolecular ion peak at m/z
643.4180 [M+Na]+ in its HR-ESITOFMS data (calcd. for C36H60O8Na
643.4180). In the low resolution ESIMS spectrum of 2 was observed
an ion at m/z 423 [M+H–H2O–C6H12O6]+, which suggested the pres-
ence of a sugar unit. Comparison of the 1H and 13C NMR spectra of
2 with those of compound 1 (Table 1), revealed six additional sig-
nals in the former, due to the sugar moiety. In the 13C NMR spec-
trum of 2, the signal corresponding to C-7 was shifted downfield
relative stereochemistry of the nucleus of compound 2, deduced
from a NOESY experiment, was found to be identical to that of 1.
The configuration of the glycosidic linkage was determined as b
based on the coupling constant value of the anomeric proton
(J = 7.8 Hz). From the above data, the structure of 2 was deduced
as cucurbita-5,24-diene-3b,23(R)-diol-7-O-b-D-allopyranoside.
Acylation of balsaminol F (1) at C-3, C-7 and C-23, with acetic
anhydride and benzoyl chloride, yielded the new triester deriva-
tives, triacetylbalsaminol F, (5) and tribenzoylbalsaminol F (6),
respectively (Fig. 3).
The molecular formula of the triester 5 was determined as
C36H56NaO6 by HR-ESITOFMS which showed a pseudomolecular
ion at m/z: 607.3961 [M+Na]+ (calcd. for C36H56NaO6Na
607.3969). Besides a molecular ion at m/z 584 [M]+, the EIMS spec-
trum of compound 5 exhibited ions at m/z 524 [M–CH3COOH]+,
464 [M–2 Â CH3COOH]+, and 404 [M–3 Â CH3COOH]+, resulting
from the sequential loss of three acetyl groups, and its IR spectrum
displayed absorption bands for ester carbonyls at 1729 and
1240 cmꢁ1. These structural features were also evidenced by the
NMR data of compound 5, which revealed the presence of signals
for three additional acetyl groups [dH 1.98 (6H, s), 1.96 (3H, s),
Me-20, Me-200, Me-2000); dC 170.5 (CO), 170.4 (CO), 170.3 (CO), 21.4,
21.2, and 21.1]. Moreover, when comparing the 1H NMR spectrum
(DdC = +5.6), indicating that the sugar was located at this carbon.
3
The heteronuclear JC–H correlations observed in the HMBC spec-
trum confirmed this location. This structural feature was identified,
unambiguously, as allose, a C-3 epimer of glucose, based on the
equatorial configuration of H-30, revealed by a broad singlet at dH
4.05, due to small equatorial-axial couplings with the vicinal pro-
´
´
´
tons at C-2 and C-4. In glucose, H-3 is axially oriented and appears
as a triplet (J3ax, 2ax ffi J3ax,2ax ffi 8,8) as observed in compound 3, and
4. The 3C NMR spectrum of 2 corroborated the above data.23,24 The
O
2'''
O
H3C
1'''
23
H
H
O
O
O
O
7
3
H3C
1'
O
O
1''
2'
N
O
2''
CH3
5
HO
23
7''
O
5'
H
H
2'''1'''
O
3'
23
3
7
HO
OH
N
H
H
1
O
3'
7
3
1'
O
2'
7'
O
O
5'
O
1''
Cl
2''
3''
7''
6
5''
Figure 3. Acylation of Balsaminol F (1).

C. Ramalhete et al. / Bioorg. Med. Chem. 19 (2011) 7474–7481
7477
of compound 5 with that of compound 1, the chemical shifts of the
oxymethine protons H-3, H-7, and H-23 were shifted downfield by
1.22, 1.19 and 1.22 ppm, respectively. Significant paramagnetic ef-
fects were also observed in the signals of the 13C NMR spectrum, at
respectively). Conversely, it is interesting to stress that the activity
was lost when the acetyl groups at positions C-3, C-7, and C-23 of
compound 5 were replaced by benzoyl groups in the benzoylated
derivative, tribenzoylbalsaminol F (6), which was inactive against
the
a
-carbons C-3, C-7, and C-23 and at the
c
-carbons C-5 and C-25
both strains (IC50 = 57.3 and 68.4
respectively).
lM for 3D7 and Dd2,
and diamagnetic effects at the b-carbons C-2, C-4, C-6, C-8, C-22,
and C-24. Similarly, acylation of balsaminol F (1) with benzoyl
chloride afforded the new derivative tribenzoylbalsaminol F (6).
The low resolution EIMS spectrum of 6 showed a molecular ion
at m/z 770 [M]+ and the high resolution mass spectrum, HR-ESI-
As can be observed in Table 2, the isolated compounds (1–4)
showed low toxicity (IC50 14.2–27.3 M) against the MCF-7 human
l
breast cancer cell line. However, a low selectivity index (SI) was
found (SI <7.0). On the other hand, no cytotoxic activity (IC50
TOFMS, exhibited
a
pseudomolecular ion at m/z: 793.4438
>133.3 lM) was exhibited by both balsaminol F (1) derivatives
[M+Na]+ (calcd. for C51H62NaO6Na 793.4439). The NMR spectra of
compound 6 provided evidence for the presence of three benzoyl
ester residues. As above, when comparing the NMR data of com-
pound 6 with that of the parent compound 1, the most remarkable
differences were found for proton and carbon resonances of rings A
and B and side chain, namely for the carbons bearing the new ester
functions and the corresponding geminal protons, which were
(5, 6). More importantly, a high SI value was found for triacetylb-
alsaminol F (5) (SI >162.4 and 342.9 for 3D7 and Dd2 P. falciparum
strains, respectively).
Compounds 1–3, 5 and 6 were also evaluated for their activity
against liver stages of the rodent malaria parasite P. berghei, at 1,
5 and 15
method. This method employs a transgenic P. berghei parasite,
PbGFP-Luccon expressing the bioluminescent reporter protein
lM, using a recently described bioluminescence-based
shifted downfield, and also for the
a,b and
c
-carbons, as expected.
,
luciferase to visualize and quantify parasite development in Huh-
7 cells, a human hepatoma cell line.9 (Fig. 4-bars). Primaquine
was used as positive control. Moreover, toxicity of compounds
(1–3, 5 and 6) was also assessed on the same cell line (Fig. 4-line)
and on mouse primary hepatocytes (Fig. 5) through the fluores-
cence measurement of cell confluency. As can be observed, all
the compounds tested in this model exhibited significant in vitro
activity against P. berghei liver stages (Fig. 4). However, balsaminol
F (1) and, in particular, balsaminoside B (2) displayed significant
toxicity against Huh-7 cells and mouse primary hepatocytes at
2.2. Biological activities
Compounds 1–6 were evaluated for their antimalarial activity
against chloroquine-sensitive (3D7) and resistant (Dd2) P. falcipa-
rum strains, using a standardized SYBR Green I-based fluorescence
assay.25,26 Chloroquine was used as positive control and the effects
of compounds against both strains were expressed as IC50 values in
lM and lg/mL. These results, as well as the compounds’ cytotoxic
activity against breast cancer cells (MCF-7) and their selectivity in-
dex (SI) are summarized in Table 2. A great diversity of criteria has
been adopted for assessing in vitro antimalarial activity of pure
compounds. In accordance with literature data,27 the antimalarial
activity of the compounds described in this work was defined as:
15 lM (Fig. 4-line, Fig. 5). Interestingly, balsaminoside B (2), which
displays a rare allose (C-30 glucose epimer) sugar at C-7, showed
superior anti-Plasmodium activity when compared with the glu-
cose analogue (3) thus demonstrating that activity is sugar-depen-
dent. As observed for the relative activities of the compounds
IC50 61
ity; and 10 < IC50 6 30
IC50 >30 M were considered inactive. Therefore, compounds 1–5
l
M, excellent/potent activity; 1 < IC50 6 10
lM, good activ-
l
M, moderate activity. Compounds with an
against the parasite’s blood stages, triacetylbalsaminol F (5)
l
showed the most potent inhibitory activity against the liver stages
of Plasmodium, displaying higher activity than primaquine, with no
detectable toxicity towards the cells assayed , at the concentrations
employed (Fig. 4 and Fig. 5). Further investigation is warranted in
order to elucidate its mechanism of action.
To assess whether the liver stage activity exhibited by the com-
pounds is due to their action on parasite motility and invasion, or
on intracellular parasite development, the effect of adding the
compounds prior to parasite addition was compared with the ef-
fect of adding compounds after invasion of the cells by the para-
sites had been completed. In this infection system, invasion is
>95% complete within 2 h of sporozoite addition to the cells.10
Therefore, any effects of compounds added after that time point
on overall infection, as measured by the luciferase-based assay em-
ployed, are not due to the compounds acting on parasite motility or
invasion, but rather on intracellular liver stages. Our data indicate
displayed antimalarial activity (Table 2) and compound 6 was inac-
tive. Among the isolated compounds, the glycoside derivatives (2–
4) revealed a good antimalarial activity against both strains tested
(2, IC50 = 2.9 and 6.3
IC50 = 3.4 and 7.2 M, for 3D7 and Dd2, respectively; 4, IC50 = 3.9
and 4.7 M, for 3D7 and Dd2, respectively), suggesting that the su-
lM, for 3D7 and Dd2, respectively; 3,
l
l
gar unit plays a role in the antiplasmodial activity. However, as
shown in Table 2, the acylated derivative triacetylbalsaminol F dis-
played the strongest antiplasmodial activity, being 22.5-
(IC50 = 0.8
active than the parent compound, balsaminol F (1, IC50 = 18.0 and
20.0 M for 3D7 and Dd2, respectively), against the sensitive and
lM for 3D7) and 50-fold (IC50 = 0.4 lM, for Dd2) more
l
resistant strains, respectively. These IC50 values are comparable
with those obtained with chloroquine, particularly against the
resistant Dd2 strain (IC50 = 0.016 and 0.20 lM for 3D7 and Dd2,
Table 2
In vitro antimalarial activity against P. falciparum 3D7, and Dd2 strains, cytotoxicity, and selectivity index and resistance index of compounds 1–6
Compounds
IC50 SD
P. falciparum Dd2
g/mL
SI
P. falciparum 3D7
g/mL
8.2 0.8
MCF7 cells
MCF7/3D7
MCF7/Dd2
l
M
l
lM
l
lM
Balsaminol F (1)
18.0 1.7
2.9 0.1
3.4 0.6
3.9 0.4
0.8 0.1
57.3 3.1
0.016
20.0 4.3
6.3 0.7
7.2 0.4
4.7 0.4
0.4 0.04
68.4 8.6
0.2
9.2 2.0
3.9 0.4
4.5 0.3
2.9 0.2
0.2 0.02
52.7 6.7
27.3 1.2
14.2 0.8
23.9 6.6
13.5 4.9
>133.3
1.5
4.8
7.0
3.5
>162.4
>2.3
1.4
2.3
3.3
2.9
>342.9
>2.0
Balsaminoside B (2)
Balsaminoside C (3)
Kuguaglycoside A (4)
Triacetylbalsaminol F (5)
Tribenzoylbalsaminol F (6)
Chloroquine
1.8 0.1
2.1 0.3
2.5 0.2
0.5 0.1
44.2 2.4
>133.3
Values shown are the mean IC50 SD (lM) from three independent experiments. Selectivity index (SI) is defined as the ratio of cytotoxicity (IC50) to antiplasmodial activity
(IC50).

7478
C. Ramalhete et al. / Bioorg. Med. Chem. 19 (2011) 7474–7481
Figure 4. Drug inhibition of liver stage infection, determined by measurement of luciferase activity (bars), and compound toxicity, assessed by fluorescence measurement of
cell confluency (line), in PbGFP-Luccon-infected Huh-7. PQ-primaquine, used as positive control. DMSO- solvent-treated control. Error bars represent the standard deviations
of three independent measurements. (⁄p <0.05; paired student’s test calculated relative to DMSO controls).
Figure 5. Cytotoxicity activity of compounds (1–3 and 5–6) in mouse primary hepatocyte cells assessed by fluorescence measurement of cell confluency. H2O2 (500 lM) and
cells incubated with an amount of DMSO equivalent to that present in the highest drug concentrations were used as positive and negative controls, respectively. (⁄p <0.05;
paired student’s test calculated relative to DMSO).
that balsaminol F (1), balsaminoside B (2) and triacetylbalsaminol
F (5), the most active compounds (Fig. 4), do act on the parasite’s
intracellular liver stages and have little or no influence on parasite
motility/invasion (Fig. 6).
ter (1H 400 MHz; 13C 100.61 MHz), using CD3OD and acetone as
solvent. ESIMS were taken on a Micromass Quattro micro API
and HR-ESITOFMS on a Bruker-Microtof ESITOF (Biotof II Model,
Brucker) and on a Bruker Daltonics maXis (ESI/NanoSpray-Qq-
TOF). EIMS, HR-EIMS, HR-CIMS were recorded on a Micromass
Autospec spectrometer. TLC was performed on precoated SiO2
F254 plates (Merck 5554 and 5744), with visualization under UV
light and by spraying with sulfuric acid-methanol (1:1), followed
by heating. Column chromatography (CC) was carried out on silica
gel (Merck 9385). TLC were performed on precoated silica gel F254
plates (Merck 5554 and 5744) and visualized under UV light and by
spraying sulfuric acid-methanol (1:1) followed by heating. HPLC
was carried out on a Merck-Hitachi instrument, with UV detection
3. Experimental Section
3.1. Chemistry
3.1.1. General Experimental Procedures
Optical rotations were obtained using a Perkin Elmer 241 polar-
imeter. IR spectra were determined on a FTIR Nicolet Impact 400,
and NMR spectra recorded on a Bruker ARX-400 NMR spectrome-

C. Ramalhete et al. / Bioorg. Med. Chem. 19 (2011) 7474–7481
7479
Figure 6. Comparison of the effect of adding the active compounds (1, 2 and 5) prior to parasite addition, with that of adding them after invasion of the cells by the parasites.
(210 and 220 nm), using a Merck LiChrospher 100 RP-18 (10
250 Â 10 mm) column.
l
m,
13C NMR data, see Table 1; ESIMS m/z: 939 [2M+Na]+ (20), 481
[M+Na]+ (21); HR-ESITOFMS m/z: 481.3649 [M+Na]+ (calcd. for
C30H50O3Na 481.3652).
3.1.2. Extraction and Isolation
Dried aerial parts of Momordica balsamina (1.2 kg) were pow-
dered and exhaustively extracted with methanol (11 Â 8 L) at
room temperature, as previously described.28 Briefly, the MeOH
extract was evaporated to afford a residue (280 g), which was sus-
pended in H2O (1 L) and extracted with EtOAc (9 Â 0.5 L). The
EtOAc residue (85 g) was suspended in MeOH/H2O (9:1; 1 L), and
extracted with n-hexane (5 Â 0.5 L) for removal of waxy material
that was not further studied. The remaining extract was evapo-
rated under vacuum (40 °C), yielding a residue (45 g) that was
chromatographed over silica gel (1 kg), using mixtures of n-hex-
ane/EtOAc (1:0 to 0:1) and EtOAc/MeOH (19:1 to 0:1) as eluents
to obtain six fractions (Fr 1–6), which were combined according
to TLC analysis. Fr 2 (2.1 g), eluted with a mixture of n-hexane/
EtOAc (1:1), was chromatographed with mixtures of n-hexane/
EtOAc. A sub-fraction (372 mg) was fractionated by repeated col-
umn chromatography with mixtures of CH2Cl2/acetone. A final
purification was carried out by HPLC to afford 50 mg of com-
pounds 1. The crude Fr 5 (2.8 g), eluted with mixtures of n-hex-
ane/EtOAc (1:1 to 0:1) and EtOAc/MeOH (1:0 to 1:1), was
repetitively fractionated by column chromatography yielding a
sub-fraction, which was further purified by HPLC to afford 7 mg
of compound 4. The residue (11.9 g) of Fr 6 (EtOAc/MeOH, 93:7
to 9:1) was recrystalized from EtOAc/MeOH to give 500 mg of a
mixture, which was fractionated by column chromatography
(SiO2, 50 g), eluted with mixtures of n-hexane/EtOAc (3:1 to 1:0)
and EtOAc/MeOH (1:0 to 1:1). The fractions eluted with EtOAc/
MeOH (97:3) were evaporated and recrystallized, from n-hexane/
EtOAc, to afford 160 mg of compound 2. The mother liquors were
associated to the remaining fractions and after evaporation, the
residue (245 mg) was submitted to repeated column chromatogra-
phy, using as eluents mixtures of EtOAc/MeOH, and CH2Cl2/MeOH
of increasing polarity, to afford more 110 mg of compound 2 and
11 mg of compound 3.
3.1.2.2. Balsaminoside B, cucurbita-5,24-diene-3b,23(R)-diol-7-
O-b-
D
-allopyranoside (2).
Prismatic white crystals mp: [230–
+ 173 (c 0.11, MeOH); IR (KBR) mmax
231 °C (n-hexane/EtOAc)]; ½a D26
ꢀ
3396, 2945, 1645, 1453, 1383, 1057, 978, 939 cmꢁ1 1H and 13C
;
NMR data, see Table 1; ESIMS m/z: 659 [M+K]+ (4), 643 [M+Na]+
(100), 423 [M+H–H2O–C6H12O6]+ (10); HR-ESITOFMS m/z:
643.4180 [M+Na]+ (calcd. for C36H60O8Na 643.4180).
3.1.3. Preparation of balsaminol F esters
3.1.3.1. Acetylation with acetic anhydride.
Compound 1
(10 mg) was suspended in acetic anhydride (1 mL) and pyridine
(1 mL). After stirring at room temperature overnight, the excess
of reagents was eliminated with N2 and the product obtained
was purified by column chromatography (n-hexane/EtOAc, 4:1)
to afford 10 mg of compound 5.
3.1.3.1.1. Triacetylbalsaminol F, 3b,7b,23(R)-triacetoxycucurbita-
5,24-diene (5). Colorless oil; IR (KBR) mmax 1729, 1528, 1462,
1370, 1240, 1015, 987, 827, 611 cmꢁ1; EIMS, m/z (rel. int.): 584
[M]+ (<1), 524 [M–CH3COOH]+ (27), 482 (100), 464 [M–
2 Â CH3COOH]+ (10), 421 (11), 404 [M–3 Â CH3COOH]+ (4); HR-
ESITOFMS m/z: 607.3961 [M+Na]+ (calcd. for C36H56NaO6Na
607.3969); 1H NMR (400 MHz, CD3COCD3):
d 5.67 (1H, d,
J = 4.8 Hz, H-6), 5.63 (1H, td, J = 8.8, 3.2 Hz, H-23), 5.12 (2H, m,
H-7/H-24), 4.72 (1H, br d, J = 12.0 Hz, H-10), 1.98 and 1.96 (6H
and 3H, s, Me-20/Me-200/Me-2000), 1.97 (3H, br s, H-8), 1.71 (3H, s,
Me-27), 1.68 (3H, s, Me-26), 1.14 (3H, s, Me-28), 1.09 (3H, s,
Me-29), 1.01 (3H, s, Me-19), 0.96 (3H, d, J = 6.0 Hz, Me-21), 0.88
(3H, s, Me-18), 0.82 (3H, s, Me-30); 13C NMR (101 MHz,
CD3COCD3): d 170.5, 170.4, 170.3 (C-10/C-100/C-1000), 150.0 (C-5),
135.7 (C-25), 125.9 (C-24), 118.5 (C-6), 79.0 (C-3), 70.8 (C-7),
69.4 (C-23), 51.4 (C-8), 51.1 (C-17), 49.0 (C-14), 46.8 (C-13),
42.7 (C-22), 40.8 (C-4), 39.3 (C-10), 35.3 (C-15), 34.6 (C-9), 33.6
(C-20), 33.0 (C-11), 30.8 (C-12), 29.0 (C-19), 28.6 (C-16), 28.4 (C
-28), 27.0 (C-2), 25.7 (C-27), 25.2 (C-29), 22.4 (C-1), 21.4, 21.2,
21.1 (C-10/C-100/C-1000), 19.3 (C-21), 18.5 (C-26), 18.4 (C-30), 15.7
(C-18).
3.1.2.1. Balsaminol F, cucurbita-5,24-diene-3b,7b,23(R)-triol
(1).
Amorphous, white powder; ½a D26
ꢀ
+ 122 (c 0.12, MeOH);
IR (KBR) mmax 3396, 2936, 1645, 1459, 1389, 939 cmꢁ1
;
1H and

7480
C. Ramalhete et al. / Bioorg. Med. Chem. 19 (2011) 7474–7481
3.1.3.2. Acylation with benzoyl chloride.
Compound
1
3.2.2. In vitro antimalarial activity in human red blood cells
Human malaria parasites were cultured as previously described
by Trager and Jensen (1976), with minor modifications.29 Briefly,
3D7 and Dd2 Plasmodium falciparum strains were cultivated in re-
cently collected erythrocytes as host cells in RPMI 1640 medium
(Gibco) containing 25 mM HEPES (Sigma) and 6.8 mM hypoxan-
thine (Sigma) supplemented with 10% AlbuMAX II (Invitrogen).
Cultures were maintained at 37 °C under an atmosphere of 5%
O2, 3–5% CO2, and N2. The antimalarial activity of the compounds
was determined by a fluorometric method using SYBR Green
I.24,25 In brief, stock solutions of the samples were prepared in
DMSO (10 mg/mL), and were diluted to give a series of concentra-
(17.6 mg) was suspended in benzoyl chloride (1 mL) and pyridine
(1 mL). After stirring at room temperature overnight, the excess
of pyridine was eliminated as above and the product obtained
was purified by column chromatography (n-hexane/EtOAc, 1:0 to
4:1). Further purification by preparative TLC (n-hexane/EtOAc,
9:1) afforded 17 mg of compound 6.
3.1.3.1.2. Tribenzoylbalsaminol F, 3b,7b,23(R)-tribenzoyloxycucur-
bita-5,24-diene (6). Colorless oil; IR (KBR) mmax 1718, 1528, 1462,
1370, 1240, 1015, 836, 870, 719 cmꢁ1; EIMS, m/z (rel. int.): 770
[M]+ (<1), 526 [M–2 Â C6H5COOH]+ (26.7); HR-ESITOFMS m/z:
793.4438 [M+Na]+ (calcd. for C51H62NaO6Na 793.4439); 1H NMR
(400 MHz, CD3COCD3): d 8.10–7.99 (6H, m, H-30/H-300/H-300/H-70/
H-700/H-7000), 7.70–7.37 (9H, m, H-40/H-400/H-4000/H-50/H-500/H-5000/H-
60/H-600/H-6000), 5.92 (2H, m, m, H-6/H-23), 5.50 (1H, brd,
J> = 5.6 Hz, H-7), 5.29 (1H, d, J = 8.8 Hz, H-24), 5.02 (1H, br s, H-
3), 2.69 (1H, br d, J = 12.4 Hz, H-10), 2.16 (1H, br s, H-8), 1.83
(3H, s, Me-26), 1.73 (3H, s, Me-27), 1.27 (3H, s, Me-28), 1.25 (3H,
s, Me-19), 1.21 (3H, s, Me-29), 1.06 (3H, d, J = 5.2 Hz, Me-21),
0.97 (3H, s, Me-18), 0.88 (3H, s, Me-30); 13C NMR (101 MHz,
CD3COCD3): d 166.0, 165.8, 165.5 (C-10/C-100/C-1000), 150.5 (C-5),
136.2 (C-25), 133.6, 133.5 (C-50/C-500/C-5000), 131.9, 131.6, 131.5
(C-20/C-200/C-2000), 129.9, 129.8 (C-30/C-300/C-3000/C-70/C-700/C-7000),
129.3, 129.2 (C-40/C-400/C-4000/C-60/C-600/C-6000), 125.5 (C-24), 118.6
(C-6), 79.9 (C-3), 71.3 (C-7), 70.2 (C-23), 51.2 (C-8), 50:9 (C-17),
48.9 (C-14), 46.7 (C-13), 42.6 (C-22), 41.3 (C-4), 38.9 (C-10), 35.2
(C-15), 34.5 (C-9), 33.6 (C-20), 32.9 (C-11), 30.6 (C-12), 29.7 (C-
19), 28.6 (C-16), 27.2 (C-28), 26.4 (C-2), 25.6 (C-27), 25.5 (C-28),
22.4 (C-1), 19.2 (C-21), 18.4 (C-30), 18.5 (C-26), 15.7 (C-18).
tions ranging from 0.156 to 100
lg/mL. 50 lL of each testing con-
centration, together with 50 L of a 1% red blood parasitized cell
l
suspension with ring stages and 2% haematocrit were distributed
in duplicate, into each of the 96-well plates. Plates were incubated
for 48 h at 37 °C. After, 100
20 mM; pH 7.5, EDTA-5 mM, saponin-0.008%; wt/vol, Triton X-
100-0.08%; vol/vol, and 0.2 l of SYBR Green I/ml of lysis buffer)
lL of SYBR Green I in lysis buffer (Tris
l
was added to each well. Plates were covered, mixed and incubated
in the dark at room temperature for 1 h. Fluorescence intensity was
measured on a fluorescence multiwell plate reader, Anthos venyth
3100 (Alfagene) excitation and emission wavelengths of 485 and
535 nm, respectively. Values were expressed in relative fluores-
cence units. Analysis of the results obtained and IC50 determination
were performed with HN-NonLineV1.1 (H. Noedl, 2001) software.
3.2.3. In vitro anti-Plasmodium liver stage activity
Huh-7 cells, a human hepatoma cell line, were cultured in RPMI
1640 medium supplemented with 10% v/v fetal calf serum (FCS), 1%
v/v non-essential amino acids, 1% v/v penicillin/streptomycin, 1% v/
v glutamine and 10 mM 4-(2-hydroxyethyl)-1-piperazineethane-
sulphonic acid (HEPES), pH 7, and maintained at 37 °C with 5%
CO2. Huh-7 cells (12 Â 103 per well) were seeded in 96-well plates
the day before drug treatment and infection. Inhibition of liver
stage infection was determined by measuring the luminescence
intensity in Huh-7 cells infected with a firefly luciferase-expressing
P. berghei line, PbGFP-Luccon, as previously described.9 Compounds
were added 1 h prior to addition of sporozoites freshly obtained
through disruption of salivary glands of infected female Anopheles
stephensi mosquitos. Sporozoite addition was followed by centrifu-
gation at 1700g for 5 min. Inhibition of parasite development was
measured 48 h after infection. The effect of the compounds on the
viability of Huh-7 cells was assessed by measuring AlamarBlue
fluorescence (Invitrogen, UK), using the manufacturer’s protocol.
3.2. Biological assays
3.2.1. In vitro cytotoxicity assay
Human breast cancer MCF-7 cell line was cultured in RPMI 1640
medium supplemented with 10% heat inactivated horse serum,
L-glutamine (2 mM), and antibiotics, in a humidified atmosphere
of 5% CO2 at 37 °C. The effects of increasing concentrations of com-
pounds on cell growth were tested in 96-well flat-bottomed micro-
titer plates. The compounds were diluted in a volume of 50 lL
medium. Then, 2 Â 104 cells in 0.1 mL of medium were added to
each well, with the exception of the medium control wells. The cul-
ture plates were further incubated at 37 °C for 24 h. At the end of
the incubation period, 15
MO, USA) solution (from a 5 mg/mL stock) was added to each well.
After incubation at 37 °C for 4 h, 100 L of sodium dodecyl sulfate
lL of MTT (thiazolyl blue, Sigma, St Louis,
l
(SDS) (Sigma, St Louis, MO, USA) solution (10%) was measured into
each well and the plates were further incubated at 37 °C overnight.
The cell growth was determined by measuring the optical density
(OD) at 550 nm (ref. 630 nm) with a Dynatech MRX vertical beam
ELISA reader. Inhibition of cell growth (as a percentage) was deter-
mined according to the formula:
Acknowledgment
This study was supported by FCT, Portugal (BD/22321/2005 to
Cátia Ramalhete and SFRH/BPD/64539/2009 to Filipa P. da Cruz).
The authors thank Dr. Teresa Vasconcelos, Instituto Superior de
Agronomia, Universidade de Lisboa, Portugal for the taxonomic
work on the plant material, and also Dr. Catarina Arruda from
the Portuguese Embassy in Mozambique, as well as the Portuguese
Office of International Affairs for plant transport.
ꢀ
ꢁ
OD sample ꢁ OD medium control
100 ꢁ
 100
OD cell control ꢁ OD medium control
Where IC50 is defined as the inhibitory dose that reduces the
growth of the compound-exposed cells by 50%. The IC50 values
are expressed as means SD from three experiments.
Supplementary data
Compound toxicity on mouse primary hepatocytes was evalu-
ated by determining cell confluence following incubation with
the compounds. Briefly, cells were incubated with the compounds
for 24 h and further incubated for 1 h with AlamarBlue-containing
medium prior to fluorescence measurement. Fluorescence was
Supplementary data (1D and 2D NMR spectra of compounds 1–
12) associated with this article can be found, in the online version,
at doi:10.1016/j.bmc.2011.10.044.
References and notes
normalized to that of cells incubated with 500 lM H2O2, used as
a positive control in the assay, and compared with that of cells
incubated with an amount equivalent to that present in the highest
drug concentrations.
1. WHO, World Malaria Report; Geneve, Switzerland, 2010.
2. WHO, Global report on antimalarial drug efficacy and drug resistance: 2000–
2010; Geneve, Switzerland, 2008.

C. Ramalhete et al. / Bioorg. Med. Chem. 19 (2011) 7474–7481
7481
3. Grimberg, B. T.; Mehlotra, R. K. Pharmaceuticals 2011, 4, 681.
4. Guantai, E.; Chibale, K. Malar. J. 2011, 10, S2.
17. Ramalhete, C.; Lopes, D.; Mulhovo, S.; Molnar, J.; Rosario, V. E.; Ferreira, M. J. U.
Bioorg. Med. Chem. 2011, 19, 330.
5. Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2007, 70, 461.
6. Newman, D. J. J. Med. Chem. 2008, 51, 2589.
18. Fatope, M. O.; Takeda, Y.; Yamashita, H.; Okabe, H.; Yamauchi, T. J. Nat. Prod.
1990, 53, 1491.
7. Butler, M. S.; Newman, D. J. Prog. Drug Res. 2008, 65, 3.
8. Wells, T. N.; Burrows, J. N.; Baird, J. K. Trends Parasitol. 2010, 26, 145.
9. Ploemen, I. H.; Prudêncio, M.; Douradinha, B. G.; Ramesar, J.; Fonager, J.; van
Gemert, G. J.; Luty, A. J.; Hermsen, C. C.; Sauerwein, R. W.; Baptista, F. G.; Mota,
M. M.; Waters, A. P.; Que, I.; Lowik, C. W.; Khan, S. M.; Janse, C. J.; Franke-
Fayard, B. M. PLoS One. 2009, 4, e7881.
19. Chen, J. C.; Lu, L.; Zhang, X. M.; Zhou, L.; Li, Z. R.; Qiu, M. H. Helv. Chim. Acta
2008, 91, 920.
20. Xu, R.; Fazio, G. C.; Matsuda, S. P. T. Phytochemistry 2004, 65, 261.
21. Cantrell, C. L.; Lu, T.; Fronczek, F. R.; Fischer, H. J. Nat. Prod. 1996, 59, 1131.
22. Ramalhete, C.; Mansoor, T. A.; Mulhovo, S.; Molnar, J.; Ferreira, M. J. U. J. Nat.
Prod. 2009, 72, 2009.
10. Prudencio, M.; Rodrigues, C. D.; Ataide, R.; Mota, M. M. Cell Microbiol. 2008, 10,
218.
23. Qing-Yan, L.; Hu-Biao, C.; Zhen Ming, L.; Bin, W.; Yu-Ying, Z. Magn. Reson. Chem.
2007, 45, 451.
11. Flyman, M. V.; Afolayan, A. J. Int. J. Food Sci. Nutr. 2007, 58, 419.
12. Thakur, G. S.; Bag, M.; Sanodiya, B. S.; Bhadauriya, P.; Debnath, M.; Prasad, G.;
Bisen, P. S. Cur. Pharm. Bio. 2009, 10, 667.
13. van de Venter, M.; Roux, S.; Bungu, L. C.; Louw, J.; Crouch, N. R.; Grace, O. M.;
Maharaj, V.; Pillay, P.; Sewnarian, P.; Bhagwandin, N.; Folb, P. J.
Ethnopharmacol. 2008, 119, 81.
14. Bandeira, S. O.; Gaspar, F.; Pagula, F. P. Pharm. Biol. 2001, 39, 70.
15. Geidam, M. A. D. E.; Hamza, H. G. Pakistan J. Biol. Sci. 2004, 7, 397.
16. Ramalhete, C.; Lopes, D.; Mulhovo, S.; Molnar, J.; Rosario, V. E.; Ferreira, M. J. U.
Bioorg. Med. Chem. 2010, 18, 5254.
24. Liu, X.; Cui, Y.; Yu, Q.; Yu, B. Phytochemistry 2005, 66, 1671.
25. Johnson, J. D.; Dennull, R. A.; Gerena, L.; Lopez-Sanchez, M.; Roncal, N. E.;
Waters, N. C. Antimicrob. Agents Chemother. 2007, 51, 1926.
26. Smilkstein, M.; Sriwilaijaroen, N.; Kelly, J. X.; Wilairat, P.; Riscoe, M. Antimicrob.
Agents Chemother. 2004, 48, 1803.
27. Batista, R.; Silva, A. D.; de Oliveira, A. B. Molecules 2009, 14, 3037.
28. Ramalhete, C.; Molnar, J.; Mulhovo, S.; Rosario, V. E.; Ferreira, M. J. U. Bioorg.
Med. Chem. 2009, 17, 6942.
29. Trager, W.; Jensen, J. B. Science 1976, 193, 673.