Activity of a hydroxybibenzyl bryophyte constituent against Leishmania spp. and Trypanosoma cruzi: In silico, in vitro and in vivo activity studies

Article abstract of DOI:10.1016/j.ejmech.2007.11.007

The synthesis and potent antiprotozoal activity of 14-hydroxylunularin, a natural hydroxybibenzyl bryophyte constituent is reported. 14-Hydroxylunularin was highly active in vitro assays against culture and intracellular forms of Leishmania spp. and Trypanosoma. cruzi, in absence of cytotoxicity against mammalian cells. Preliminary structure-activity relationship studies showed that the reported bioactivity depends on hybridization at the carbon-carbon bridge, position and number of free hydroxy group on the aromatic rings. The reported results were also in agreement with the in silico prediction using Non-Stochastic Quadratic Fingerprints-based algorithms. The same compound also showed antiprotozoal activity in Leishmania spp. infected mice by oral and subcutaneous administration routes, with an optimal treatment of a daily subcutaneous administration of 10 mg/kg of body weight for 15 days. This study suggested that 14-hydroxylunularin may be chosen as a new candidate in the development of leishmanicidal therapy.

Full text of DOI:10.1016/j.ejmech.2007.11.007

Available online at www.sciencedirect.com
European Journal of Medicinal Chemistry 43 (2008) 1797e1807
http://www.elsevier.com/locate/ejmech
Original article
Activity of a hydroxybibenzyl bryophyte constituent against
Leishmania spp. and Trypanosoma cruzi: In silico, in vitro and in
vivo activity studies
a
b
c
d
´
Virginia Roldos , Hector Nakayama , Miriam Rolon , Alina Montero-Torres ,
Fernando Trucco a, Susana Torres b, Celeste Vega c, Yovanni Marrero-Ponce d,
a
b
c
b
´
Viviana Heguaburu , Gloria Yaluff , Alicia Gomez-Barrio , Luis Sanabria ,
a,**
Maria Elena Ferreira b, Antonieta Rojas de Arias b, , Enrique Pandolfi
*
a Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, Universidad de la Repu´blica, Av.Gral. Flores 2124, CC 1157,
Montevideo 11800, Uruguay
b Instituto de Investigaciones en Ciencias de la Salud, Universidad Nacional de Asuncio´n, CP 2511, Asuncio´n, Paraguay
c Department of Parasitology, Faculty of Pharmacy, UCM, 28 040 Madrid, Spain
d Department of Synthesis and Drug design, Chemical Bioactive Center, Central University of Las Villas, Santa Clara 54830, Villa Clara, Cuba
Received 22 December 2006; received in revised form 12 November 2007; accepted 13 November 2007
Available online 19 November 2007
Abstract
The synthesis and potent antiprotozoal activity of 14-hydroxylunularin, a natural hydroxybibenzyl bryophyte constituent is reported. 14-
Hydroxylunularin was highly active in vitro assays against culture and intracellular forms of Leishmania spp. and Trypanosoma. cruzi, in ab-
sence of cytotoxicity against mammalian cells. Preliminary structureeactivity relationship studies showed that the reported bioactivity depends
on hybridization at the carbonecarbon bridge, position and number of free hydroxy group on the aromatic rings. The reported results were also
in agreement with the in silico prediction using Non-Stochastic Quadratic Fingerprints-based algorithms. The same compound also showed
antiprotozoal activity in Leishmania spp. infected mice by oral and subcutaneous administration routes, with an optimal treatment of a daily
subcutaneous administration of 10 mg/kg of body weight for 15 days. This study suggested that 14-hydroxylunularin may be chosen as
a new candidate in the development of leishmanicidal therapy.
Ó 2007 Elsevier Masson SAS. All rights reserved.
Keywords: Leishmania; Trypanosoma cruzi; Bryophytes; Hydroxybibenzyl constituents; TOMOCOMD software
1. Introduction
leishmaniasis stress the deficiencies of the currently available
therapeutic agents and the urgent need for new candidates
[1,2]. Leishmaniasis is a group of tropical diseases caused
by parasites of about 20 species of the genus Leishmania,
which were transmitted by a group of 50 species and sub-
species of phebotomine insects [3,4]. Official data show that
there are 12 million infected people around the world, 350 mil-
lion at risk of acquiring the disease, and 1.5e2 million that
will be infected annually [5,6]. Leishmaniasis showed a com-
plex and diverse clinical manifestations and epidemiology [7].
Currently used drugs for treatment of human cutaneous
leishmaniasis are quite toxic and cause severe side effects
The discovery and development of essential drugs for ne-
glected diseases such as Chagas’ and Leishmaniasis is a major
concern in the pharmaceutical world. Recent reviews on the
chemotherapy of Chagas’ disease and American cutaneous
* Corresponding author. Tel.: þ595 21 450495; fax: þ595 21 214094.
** Corresponding author. Tel.: þ598 2 9244066; fax: þ598 2 9241906.
E-mail addresses: arias1@telesurf.com.py (A. Rojas de Arias), epandolf@
fq.edu.uy (E. Pandolfi).
0223-5234/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2007.11.007

1798
V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
such as pancreatitis and cardiac toxicity that have therefore
limited their use. Conventional therapy consists in the paren-
teral administration of pentavalent antimonium [sodium stibo-
gluconate or PentostamÒ (GSK) and meglumine antimoniate
or GlucantimeÒ (Aventis)] during 28 days, generally under
strict medical supervision, making the treatment unaffordable
for most patients.
The emergence of antimony-resistant strains of Leishmania
donovani is alarming. For example, up to 60% of the patients
in Bihar state (India) do not respond to antimonials [8,9].
The most significant recent advance in leishmaniasis ther-
apy came from the oral treatment with MiltefosineÒ (Zentaris),
although its use is restricted due to teratogenicity. This drug is
currently in Phase IV clinical trials for visceral and cutaneous
leishmaniasis [10].
On the other hand, the World Health Organization has iden-
tified Chagas’ disease as a major and increasing problem in
public health, particularly in Latin America [6,11], where
18e20 million people in 17 countries are infected, and almost
40 million are exposed to the infection of Trypanosoma cruzi
transmitted by triatomine vectors, blood transfusion or con-
genital. In the past the pattern of the disease was almost exclu-
sively rural, nevertheless, the rising levels of urbanization and
emigration extended the disease into an urban environment.
Medication for Chagas’ disease is usually effective only
when given during the acute stage of infection. Once the
disease has progressed to later stages, no medication has
consistently proven to be effective. As happens in Leishmania,
the current synthetic drugs such as nifurtimox (a nitrofuran
derivative) and benznidazole (a nitroimidazole derivative),
are associated to severe side effects, including cardiac and/or
renal toxicity, which accounts for the need to search new ef-
fective chemotherapeutic and chemo prophylactic agents
against T. cruzi [12,13].
Fig. 1. Hydroxybibenzyl constituent of bryophytes.
compound 2 was prepared following a described methodology
[28] based on a Wittig olefination with a high to excellent
yield. The olefination reaction was carried out according to
Boden’s protocol [29,30] in order to improve the final yield
of compound 4, required for a multigram scale synthesis
(Scheme 1). The reaction proceeds straightforward using
K2CO3 (2 eq) with 18-crown-6 ether as catalyst in refluxing
toluene, giving 4 in 88% yield. Under these conditions, (2,5-
dimetoxybenzyl)triphenylphosphonium chloride (3) reacted
with p-anisaldehyde to give the stilbene 4 exclusively as the
E-isomer. Stilbene 4 was hydrogenated to 5 and further depro-
tected to obtain 2,5,40-trihydroxybibenzyl (2) at an overall
yield of 56% in three steps. Finally, all compounds were
characterized by 1H and 13C NMR studies, and mass
spectrometry.
Resveratrol (trans-3,5,40-trihydroxystilbene) and lunularin
(1), required as a control compounds for the in vitro test,
were synthesized according to Yus et al. [31] and Huneck
[32], respectively.
3. Computational analysis
The handling and screening of large databases to find re-
duced sets of potential new active compounds is possible
[14,15]. Thus, the development of computational approaches
based on discrimination functions plays an important role, al-
lowing the previous in silico identification from large chemical
libraries of structural subsystems responsible for a property or
biological activity with a considerable reduction of costs, time
and efforts to select a new drug candidate.
A novel strategy to generate molecular fingerprints-based
on the application of discrete mathematics and linear algebra
theory has been recently developed. The approach affords to
carry out rational in silico molecular design (selection/identi-
fication) and QSAR/QSPR studies. This algorithm has been
applied to the prediction of several physical, physicochemical,
chemical, pharmacokinetical parameters as well as biological
properties.
The purpose of this work is to demonstrate the in vitro and
in vivo activity of a hydroxybibenzyl bryophyte constituent
against Leishmania spp. and T. cruzi predicted through in sil-
ico studies.
In the current study, 14-hydroxylunularin (2) was in silico
evaluated by the TOMOCOMD strategy [33,34] in order to
predict antitrypanosomal and antileishmanial activities. Bio-
logical assays were also performed to confirm in vitro activi-
ties on both Leishmania spp. and T. cruzi proliferative forms
as well as cytotoxicity on mammalian cells. In addition, the
in vivo assays against Leishmania amazonensis are here
reported, resulting this compound as a new and promising
leishmanicidal drug candidate.
In silico predictions were conducted for antitrypanosomal
activity using the discriminant function previously reported
by some of the present authors [35]. A new model [Eq. (1)]
to predict antileishmanial activity of 14-hydroxylunularin (2)
was generated considering a data set of 168 organic-chemicals
(see Section 5).
2. Chemistry
Hydroxybibenzyl derivatives are endowed with a large va-
riety of biological effects including: antifungal, antimicrobial,
and cytotoxic activities, among others [16e20]. In particular,
liverworts are a rich source of hydroxybibenzyl compounds
[21,22]; such as lunularin (1) [23,24] and 14-hydroxylunularin
(2) [25] (Fig. 1).
As a part of our work in the synthesis and biological
evaluation of natural compounds from bryophytes [26,27],

V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
1799
Scheme 1. Preparation of compound 2.
À
Á
Class¼ꢀ4:49þ1:37ꢁ10ꢀ5 q7 ðxÞꢀ0:17q1L xE þ6:72
H
strains but a lower IC50 value (0.5 mM) when tested on L. do-
novani promastigotes. To note, compound 2 tested in a range
of (0.43e434 mM) showed a higher leishmanicidal activity
when compared with pentamidine (IC50 ¼ 9.8 mM), taken as
a reference drug. Regardless of the Leishmania strain tested,
no live parasites could observe beyond 13.03 mM (data not
shown).
Likewise, the in vitro results on T. cruzi epimastigote forms
evidenced an important trypanocidal activity (IC50 ¼ 5.8 mM),
higher than that obtained with the reference drug benznidazole
(54.7 mM). Nevertheless altogether the leishmanicidal activity
of compound 2 was consistently higher than that on T. cruzi
epimastigotes.
Multiplicative forms of the Leishmania and T. cruzi that
appears in humans are amastigotes. For this reason, more se-
lective methods are required to determine the activity of this
compound [38]. The antiamastigote assays were performed to
determine the activity of 2, pentamidine and benznidazole
against Leishmania amastigotes infecting peritoneal macro-
phages and T. cruzi amastigotes infecting NCTC-929 fibro-
blasts, respectively. Our results (Table 2) showed that
compound 2 (IC50 ¼ 1.5 mM) on intracellular amastigotes of
L. amazonensis is less potent than the reference drug pentam-
idine (IC50 ¼ 0.9 mM) with full disappearance of intracellular
amastigotes at concentration higher than 4.34 mM (data not
shown). On T. cruzi amastigotes, the ranking was reversed,
being the IC50 of compound 2 more active than the reference
drug benznidazol IC50 ¼ 17.3 mM and 192.1 mM, respectively.
In order to establish a preliminary structureeactivity rela-
tionship we tested three bibenzyl derivates of 2 against pro-
mastigotes form of Leishmania spp. as compiled in Table 3.
Compounds 1, 5 and resveratrol showed less activity than
14-hydroxylunularin (2) at 100 mg/mL. The results suggest
that a chemical structure with sp3 hybridization at the car-
bonecarbon bridge and 2,5,40-substitution with free hydroxy
groups on the aromatic rings increased the leishmanicidal
and trypanocidal activities in agreement with the recently re-
sults from Kedzierski et al. [39].
À
Á
À
Á
ꢀ2
ꢀ3
q
ꢁ10
q
xE þ1:44ꢁ10
xE
0L
5L
ꢀ8
H
ꢀ0:27ꢁ10
q
ðxEÞ
ð1Þ
15L
N ¼ 120,
l
¼ 0.316, D2 ¼ 8.49, F(5,115) ¼ 49.644,
p < 0.0001.
Model 1, generated by Eq. (1), allowed the correct classifi-
cation of 96.66% of chemicals included in the active training
set (58 from 60 compounds) and 90.16% of the inactive group
(55 from 61). An overall 93.39% of correct classification was
observed for the whole group. Both models (for antitrypanoso-
mal and antileishmanial activities, respectively) predicted the
activity of 2 showing a positive expectancy of being active
(DP ¼ 98.33% for antileishmanial and DP ¼ 28.06% for anti-
trypanosomal activity). Positive values of DP% indicate a pos-
itive probability of being active and vice versa. Compounds 1,
5 and resveratrol were also evaluated using Model 1 and the
results are compiled in Table 1.
4. Biological results and discussion
According to in silico results, accuracy of the prediction
was tested by assaying compound 2 in vitro against epimasti-
gote and amastigote forms of T. cruzi, and on promastigote
and amastigote stages of Leishmania spp. Promastigotes and
epimastigotes are the extracellular dividing forms inside the
insect vectors of Leishmania spp. and T. cruzi, respectively.
The easily culturable and drug sensitivity of these forms
make these models an excellent choice for preliminary in vitro
screening [34,36,37]. The results for inhibition of in vitro
Leishmania proliferation after 72 h of incubation with com-
pound 2 are shown in Table 2. Interestingly, compound 2
showed a similar IC50 (1.1 mM) when was tested against
both Leishmania braziliensis and L. amazonesis promastigote
Table 1
Classification of compounds 1,2,5 and resveratrol using Model 1 (prediction
for antileishmanial activity)
An important criterion in the search of active compounds
with antiprotozoal activity is their toxicity on mammalian
host cells [40]. For this purpose, the cytotoxicity was deter-
mined using two different cells lines and peritoneal macro-
phage concentrations ranging from 2.17 to 21.70 mM (Table
4). As observed in the resazurin tests [41], compound 2 showed
Compounds
DP%a
1
97.42
98.33
62.68
99.14
2
5
Resveratrol
a
DP% ¼ [P(active) ꢀ P(inactive)] ꢁ 100.

1800
V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
Table 2
In vitro activity of 14-hydroxylunularin (2) and reference drugs towards extracellular and intracellular forms of Leishmania spp. and T. cruzi
Extracellular forms
Compounds
Intracellular forms
Leishmanicidal activity on promastigotesa
Antiepimastigote
Antiamastigote
activity of T. cruzi activity of Leishmania spp. activity of T. cruzi
Antiamastigote
L. donovani L. braziliensis L. amazonensis Clone CL-B5
(PP75)
L. amazonensis (PH8)
Clone CL-B5
(M2903)
(PH8)
14-Hydroxylunularin (2) (IC50,mM) 0.5
Pentamidine (IC50,mM)
Benznidazole (IC50,mM)
1.1
9.8
e
1.1
9.8
e
5.8
1.5
0.9
e
17.3
e
192.1
9.8
e
e
54.7
a
Control test was included.
no toxicity at the highest concentration assayed, whereas in the
trypan blue exclusion assay, a low toxicity was observed on
peritoneal murine macrophages at 21.70 mM with 18.88% of
cell taking the dye, reduced to 2.11% at 4.34 mM. Noteworthy
the toxicity is very low on logarithmically growing mamma-
lian host cells, within the range of high antiparasitic activity.
Due this low value observed for compound 2 at these concen-
trations, we suggest that the activity described on parasites was
the result of a specific toxicity.
of this drug. The results showed that the evaluated compound
exhibited a strong antiproliferative activity on all developmen-
tal stages of the parasites, Leishmania spp.
In summary, the TOMOCOMD approaches allowed the
assignment of the antitrypanosomal and antileishmanial poten-
tialities of the novel compound 2 by a preliminary in silico
study. These predictions were also confirmed by in vitro and
in vivo assays. Based on this preliminary study of structuree
activity relationship we suggest that the antiparasitic activity
of bibenzyl phenolic compounds is strongly dependent on the
position and number of free hydroxy group on the aromatic
ring. Furthermore, the presence of a saturated carbon bridge
linking both aromatic groups is another important factor for
biological activity. In all, 14-hydroxylunularin bearing 2,5,40-
hydroxy substituent on a scaffold consisting of two phenyl
groups linked through carbonecarbon [CH2] bridge showed
the highest leishmanicidal and trypanocidal activity among
the set of hydroxybibenzyl derivatives assayed. To our knowl-
edge, this is the first study to show these activities for second-
ary metabolites from bryophytes. Although the mode of action
of compound 2 is still unknown, we concluded that this com-
pound provides a new perspective in the treatment against
leishmaniasis.
Taken into consideration the high potency of compound 2 on
promastigote and amastigote forms of Leishmania and the ab-
sence of cytotoxicity in the different models assayed, the in
vivo assays were performed as previously described [42e44].
The outcome of the treatment with intravenous Glucantime
and oral and subcutaneous 14-hydroxylunularin (2) on L. ama-
zonensis infected BALB/c mice are presented in Table 5.
The subcutaneous treatment with antimonial drug Glucan-
time at 100 mg/kg along 15 days reduced the lesion weight by
72.1% (P < 0.05) and the parasite load by 98.2% (P < 0.05)
compared with the untreated mice. When 2 was subcutane-
ously administrated on a daily basis at 25 mg/kg for 15 days
or orally in a single dose for 15 days, we observed a potent
antiparasite effect. In this case, the weight of the lesion de-
creased by 98.3% (P < 0.01) and 80.9% (P < 0.01), respec-
tively. Surprisingly, treatment with 2 at 10 mg/kg for 15
days was sufficient to decrease the lesion weight and the par-
asite load either by the subcutaneous or the oral route. This
specific treatment caused the lesion weight to decrease by
95.9% and the parasite load by 93% (subcutaneous route).
Also oral administrations at 10 mg/kg of 2 caused a decrease
of 90% and 68.6%. Particularly, the oral route of administra-
tion will be optimal regarding a potential commercialization
5. Experimental protocols
5.1. Experimental procedure for the chemical synthesis
5.1.1. E-2,5,40-Trimethoxystilbene (4)
p-Anisaldehyde (1.0 eq) and (2,5-dimethoxybenzyl) triphe-
nylphosphonium chloride (1.1 eq) were dissolved in the mini-
mum amount of toluene, and K2CO3 (2.0 eq) and a trace
Table 3
In vitro activity of bibenzyl derivatives against promastigotes of three Leish-
mania spp. strain
Table 4
Cytotoxicity effects of compound 2 on mammalian cells
Compound
14-Hydroxylunularin
(2) (mM)
NCTC-Clone 929
fibroblasta
(%C ꢂ SDc)
J 774 Murine
macrophagea
(%C ꢂ SD)
Peritoneal
macrophageb
Compound
(100 mg/mL)
Percentual lysis of
Leishmanial promastigotes
(%C ꢂ SD)
L. donovani
(PP75)
L. braziliensis
(M2903)
L. amazonensis
(PH8)
21.7
8.6
4.3
2.1
a
7.91 ꢂ 2.89
2.93 ꢂ 0.42
0.66 ꢂ 2.04
0
18.88 ꢂ 3.25
4.89 ꢂ 4.56
2.11 ꢂ 2.36
0.76 ꢂ 1.18
1.40 ꢂ 0.9
0
0
0
1
N/Ra
100
<70
100
<70
80
<70
100
2
5
Resveratrol
N/R
<70
<70
<70
According to resazurin test.
b
c
According to trypan blue exclusion method.
Percentage of cytotoxicity ꢂ standard deviation.
a
N/R ¼ not reported.

V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
1801
Table 5
Effect of treatment with 14-hydroxylunularin (2) by oral and subcutaneous (SC) route on L. amazonensis-infected BALB/c mice
Compound 2
(mg/kg)
No. of
mice
Route of
adminst.
Lesion weight
(g) ꢂ SD
Supp. of weight
lesion (%)
Supp. of parasite
load in lesion (%)
Mean parasite quantitation
in lesion (ꢁ106)
1.34
0.29
0.29
0.12
0.76
4.28
10
10
11
12
10
10
13
13
Oral
SC
0.0350 ꢂ 0.0364
0.0144 ꢂ 0.0151
0.0067 ꢂ 0.0123
0.0058 ꢂ 0.0045
0.0980 ꢂ 0.0857
0.3520 ꢂ 0.1534
ꢀ90.0b
ꢀ95.9c
ꢀ80.9c
ꢀ98.3c
ꢀ72.1b
e
ꢀ68.6b
ꢀ93.0b
ꢀ93.1b
ꢀ97.0b
ꢀ98.2b
e
25
25
Glua
Oral
SC
SC
Oral
None (control)
Supp ¼ suppresion.
a
Glucantime 100 mg/kg.
P < 0.05 (treated mice vs control).
P < 0.01 (treated mice vs control).
b
c
amount of 18-crown-6 ether were added at r.t. The reaction
mixture was then heated to reflux for 18 h. The solvent was
evaporated under vacuum and the residue was purified by
chromatographic column on silica gel (88% yield). In particu-
lar, for this reaction the ratio of (E/Z ) isomers is heavily de-
pendent on the solvent and base, ranging from 70/30 (E/Z )
to exclusively the E-isomer in toluene and K2CO3 as base,
yielding compound 4 in 88%.
White solid, mp: 61.9e62.7 ꢃC. 1H NMR (CDCl3,
400 MHz) d: 7.46 (d, 2H, J ¼ 8.4 Hz, AreH), 7.31 (d, 1H,
J ¼ 16.2, CH]C), 7.10 (d, 1H, J ¼ 2.7 Hz, AreH), 7.04 (d,
1H, J ¼ 16.2 Hz, CH]C), 6.89 (d, 2H, J ¼ 8.7 Hz, AreH),
6.82 (d, 1H, J ¼ 8.7 Hz, AreH), 6.76 (dd, 1H, J1 ¼ 3 Hz,
J2 ¼ 9 Hz, AreH), 3.83 (s, 3H, OCH3), 3.82 (s, 3H, OCH3),
3.81 (s, 3H, OCH3). 13C NMR (CDCl3, 100.5 MHz) d: 55.2,
55.7, 56.2, 107.4, 111.4, 112.2, 113.2, 114.0, 121.1, 127.6,
127.7, 128.8, 130.1, 130.6, 151.2, 151.7, 153.7. EM: m/z
(%) 270 (100), 255 (11), 240(3), 227 (43).
was then heated at 160 ꢃC for 15 min under argon. The cooled
reaction mixture treated slowly with 20 mL of 10% aqueous
NH4Cl and extracted with three 20 mL portions of ether.
The combined organic extract was washed with 20 mL of wa-
ter and 20 mL of brine. The dried (MgSO4) organic phase was
concentrated under reduced pressure. The residue was purified
by flash chromatography on a silica gel column yielding com-
pound 2 in 75%.
White solid, mp: 145.7e146.1 ꢃC. 1H NMR (CDCl3,
400 MHz) d: 7.04 (d, 2H, J ¼ 8.4 Hz, AreH), 6.72 (d, 2H,
J ¼ 8.4 Hz, AreH), 6.65 (d, 1H, J ¼ 8.6 Hz, AreH), 6.58 (d,
1H, J ¼ 3 Hz, AreH), 6.48 (dd, 1H, J1 ¼ 3 Hz, J2 ¼ 8.7 Hz,
AreH), 2.77 (br s, 4H, AreCH2]CH2eAr). 13C NMR
(CDCl3, 100.5 MHz) d: 33.7, 35.9, 113.8, 115.8, 116.3,
117.4, 117.5, 129.9, 130.0 (ꢁ2), 134.0, 148.7, 151.0, 156.2.
EM: m/z (%) 230 (26.2), 136 (6), 107 (100). Anal.
C14H14O3 (230): calcd (C 73.04, H 6.08); found (C 72. 99,
H 5.98).
5.1.2. 2,5,40-Trimethoxybibenzyl (5)
5.2. Biological assays
Compound 4 (1 mmol) was dissolved in ethyl acetate, pal-
ladium on activated carbon (5% Pd) was added (50 mg/mmol)
and hydrogenation performed at 4 bar hydrogen pressure. The
catalyst was filtered off and the solvent removed in vacuum.
The crude material was purified by chromatographic column
(SiO2; CH2Cl2), yielding compound 5 in 85%.
White solid, mp: 164.9e166.2 ꢃC. 1H NMR (CDCl3,
400 MHz) d: 7.18 (d, 2H, J ¼ 8.3 Hz, AreH), 6.88 (d, 2H,
J ¼ 8.3 Hz, AreH), 6.83e6.85 (m, 1H, AreH), 6.75e6.77
(m, 2H, AreH), 3.83 (s, 3H, OCH3), 3.81 (s, 3H, OCH3),
3.79 (s, 3H, OCH3), 2.87e2.94 (m, 4H, CH2eCH2). 13C
NMR (CDCl3, 100.5 MHz) d: 33.2, 35.7, 55.6, 56.0, 56.3,
111.5, 111.7, 114.1, 116.7, 129.7, 132.0, 134.8, 152.2,
153.9, 158.2. EM: m/z (%) 272 (47.5), 151 (25.3), 121 (100).
5.2.1. In vitro activity on Leishmania spp. promastigote
Cultures of Leishmania spp. were obtained from IICS (Insti-
´
tuto de Investigaciones en Ciencias de la Salud, Asuncion, Par-
aguay) and identified by isoenzyme analysis. The maintenance,
cultivation and isolation of promastigote parasites were per-
formed as usual. Promastigotes inhibition studies were per-
formed on L. amazonensis (IFLA/BR/67/PH8), L. braziliensis
(MHOM/BR/75/M2903) and L. donovani (MHOM/BR/74/
PP75) grown at 22 ꢃC in Schneider’s Drosophila medium con-
taining 20% foetal bovine serum. Promastigotes cultures in the
logarithmic phase were transferred at a concentration of 106
cells per milliliter. Compounds (1 mg) were dissolved in
40 mL of dimethyl sulphoxide (DMSO) and added to 1 mL
of the medium from which aliquots are drawn. From this stock
solution 200 mL were dissolved in 800 mL of medium. 100
microlitres of this solution was mixed with 100 mL of parasite
culture reaching a concentration of the compound of 100 mg/
mL and 0.4% of DMSO, and transferred into microtitre plates
by triplicates. The activity of compounds was evaluated after
72 h by optical counting of the cells after appropriate dilutions
and referred to the untreated control cells. IC50 was determined
5.1.3. 2,5,40-Trihydrohybibenzyl (14-hydroxylunularin) (2)
Methyl iodide (87.8 mmol) and magnesium (23.4 mmol)
were stirred together in 24 mL of dry diethyl ether until the
initial exothermic reaction had subsided. A solution containing
1 mmol of 5 in 24 mL of dry diethyl ether was added dropwise
and the solution was concentrated under reduced pressure. The
residue was heated to 100 ꢃC while still under vacuum and

1802
V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
Fig. 2. Active compounds included in the training set.
in three independent assays. Pentamidine was used as a refer-
ence drug.
tryptose broth (LIT) with 10% foetal bovine serum (FBS), peni-
cillin and streptomycin and harvested at the exponential
growth phase. The screening assay was performed in 96-well
microplates (Sarstedt, Sarstedt, Inc.) with cultures that had
not reached the stationary phase. Briefly, epimastigotes were
seeded at 1 ꢁ 105 parasites/mL in 200 mL growth media. The
plates were then incubated with the respective drugs at 28 ꢃC
5.2.2. T. cruzi epimastigote susceptibility assay
For in vitro studies, the clone CL-B5 of T. cruzi stably trans-
fected with Escherichia coli b-galactosidase gene (lacZ), was
used. The epimastigotes were grown at 28 ꢃC in liver infusion

V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
1803
Fig. 2. (continued).
for 72 h, afterwards 50 mL of CPRG solution was added result-
ing in a final concentration of 200 mM. The plates were incu-
bated at 37 ꢃC for an additional 6 h and then read at 595 nm.
Each concentration was tested in triplicate. The efficacy of
each compound was estimated by calculating the IC50 values.
These values were calculated by the sigmoidal doseeresponse
curve adjustment using the statistical software program Graph-
Pad Prims 3.0. Benznidazole was used as reference drug.
microscope was used and the reading was taken with trypan
blue to 0.4%. One hundred macrophages were counted and
the frequency of live and dead, according to trypan blue exclu-
sion method, was determined. The assays with the evaluated
compound, its dilutions and controls were made in triplicate.
Pentamidine was used as reference drug.
5.2.4. In vitro cytotoxicity in cellular lines
The cell lines used were NCTC clone 929 and murine
J774 macrophages, which were grown in Minimal Essential
Medium (SIGMAÒ) and in RPMI 1640 medium (SIGMAÒ),
respectively. Both media were supplemented with 10% heat-
inactivated FBS, penicillin G (100 U/mL) and streptomycin
(100 mg/mL). Cell cultures were maintained at 37 ꢃC in a hu-
midified 5% CO2 atmosphere. The procedure for cell viability
measurement was evaluated with resazurin by a colorimetric
method.
Macrophages. J774 macrophages were seeded (5 ꢁ 104
cells/well) in 96-well flat-bottomed microplates with 100 mL
of RPMI 1640 medium. The cells were allowed to attach for
24 h at 37 ꢃC, 5% CO2, then the medium was replaced by dif-
ferent concentrations of the drugs in 200 mL of medium, and
exposed for another 24 h. Growth controls were also included.
Afterwards, a 20 mL of 1 mM resazurin solution was added
5.2.3. In vitro cytotoxicity on peritoneal macrophages
Healthy BALB/c mice (18e20 g) fasted for 12 h were used.
Macrotitration plates with 24 wells with round slides were im-
pregnated with nitric acid for better macrophage adherence.
The procedure was carried out in conditions of total sterility
using GELAIREÒ BSB4A Flow Laboratories Laminar Flow.
Mice peritoneum was open and 6 mL of cold PBS was inocu-
lated and mixed inside the animal cavity, the content was taken
with a syringe and centrifuged to 1200 rpm during 10 mm,
resuspended in RPMIÒ and counted in a Neubauer chamber;
finally was placed inside the wells (1.105 macrophages/mL).
14-Hydroxylunularin (2) was weighed and DMSO (MERCKÒ)
was used as solvent. Dilutions of the product at 10, 5 and 1 mg/
mL were tested. Macrowell was incubated at 37 ꢃC (5% CO2)
for 48 h and for the evaluation an OLYMPUSÒ inverted

1804
V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
Fig. 3. Active compounds included in the test set.
and plates were returned to incubator for another 3 h to eval-
uate cell viability. Fibroblasts. NCTC clone 929 cells were
plated in 96-microtitre plates at 3 ꢁ 104 cells per well in
100 mL growth medium. The cells were grown overnight at
37 ꢃC, 5% CO2. Thereafter the medium was removed and
the compounds were added in 200 mL medium for 24 h. After
incubation, 20 mL of 2 mM resazurin solution was added to
each well. The plates were incubated for 3 h to allow optimal
oxidationereduction. The reduction of resazurin was deter-
mined by dual wavelength absorbance measurement at 490
and 595 nm. Background was subtracted. Each concentration
was assayed three times. Medium and drug controls were
used in each test as blanks.
500 rpm for 10 min to remove the remaining tissues. The su-
pernatant, containing amastigotes, was washed with RPMIÒ,
and centrifuged again at 3000 rpm for 10 min; amastigotes
in the pellet were counted in a Neubauer chamber. Macro-
phages in the macrotitration plates allowed to adhere for 3 h
were infected with the hamster amastigotes of L. amazonensis
(PH8) at a 1:5 macrophage:amastigotes ratio and incubated at
37 ꢃC (5% CO2) for 2 h. A stock of 14-hydroxylunularin (2)
was weighed and diluted in DMSO (Merck) at 1% in the cul-
ture media to a concentration of 1 mg/mL (4.34 mM). Dilution
of 0.5, 0.3, 0.1 mg/mL and controls were prepared in triplicate.
Then, they were incubated again at 37 ꢃC during 48 h and
reading was made in an OLYMPUSÒ inverted microscope.
Percentages of infected macrophages in each assay were deter-
mined microscopically at magnification of 1000ꢁ.
5.2.5. Activity in Leishmania amastigotes
Intracellular forms (amastigotes) of L. amazonensis (PH8)
maintained in Golden hamsters (Mesocricetus auratus) by
successive passages every two months were used. Briefly,
macrophages were obtained from mice as aforementioned in
cytotoxicity section. The granuloma, containing the parasites,
was removed from hamster feet, grinded and centrifuged to
5.2.6. T. cruzi amastigote susceptibility assay
The activity was evaluated by colorimetric method using
CPRG, NCTC-929 fibroblasts were seeded in 24-well tissue
culture plates at an optimal concentration of 2.5 ꢁ 103 cells/
well, previously determined. NCTC-929-derived trypomastigotes

V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
Table 6 (continued)
1805
Class
Table 6
Classification of active and inactive compounds included in the training set
using Model 1
Compound
DP%a
Class Compound
DP%a
Glisolamide
Glibutimine
Ag 307
ꢀ90.00
ꢀ96.50
ꢀ99.30
ꢀ99.90
ꢀ99.50
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Cefalexin
ꢀ43.10
63.60
ꢀ
þ
ꢀ
ꢀ
ꢀ
Compound
DP%a
Class Compound
DP%a
Class
Streptomycin
Azirinomycin
Gentamicin A1
(2-Hidroxypropyl)
trimetilamonium
hydroxide
Training active group
1e
ꢀ100
98.89
98.21
98.44
98.32
91.2
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
ꢀ
þ
þ
þ
þ
31e
32e
33e
34e
35e
36e
37e
38e
39e
40e
41e
42e
43e
44e
45e
46e
47e
48e
49e
50e
51e
52e
53e
54e
55e
56e
57e
58e
59e
60e
49.93
98
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
ꢀ
þ
þ
þ
Bromcholine
Mebetide
ꢀ95.60
ꢀ99.90
2e
3e
61.61
73.14
70.38
25.56
88.62
91.54
96
4e
5e
Minoxidil
ꢀ96.40
Results of the classification of compounds obtained from Eq. (1) (using
ꢀ
6e
7e
89.4
a
93.64
94.91
59.54
73.09
78.04
69.28
75.27
80.02
99.46
99.72
99.84
99.90
99.97
99.98
99.98
99.97
99.99
99.99
99.98
non-stochastic quadratic indices): DP% ¼ [P(active) ꢀ P(inactive)] ꢁ 100.
8e
9e
10e
11e
12e
13e
14e
15e
16e
17e
18e
19e
20e
21e
22e
23e
24e
25e
99.59
93.44
91.25
99.75
87.26
96.93
94.08
96.92
52.83
99.88
99.54
48.79
2.561
18.05
94.81
98.41
67.53
were added to the monolayers at a parasite:cell ratio of 5:1 and
incubated for 24 h at 33 ꢃC with 5% CO2. The infected cells
were then washed twice with PBS to remove the extracellular
trypomastigotes. The drugs were added in triplicate to give
a final volume of 900 mL/well. The plates were incubated
for 7 days at 33 ꢃC. At this time, 100 mL of CPRG solution (fi-
nal concentration 400 mM) in 0.3% Triton X-100 was added.
Following 4 h of incubation at 37 ꢃC, the colorimetric reaction
was quantified by measuring at optical densities (OD) of
595 nm. Background controls (only NCTC-929 cells) were
subtracted from all values. The results were expressed as
IC50 values.
26e
27e
28e
29e
30e
ꢀ33.3
100
ꢀ35
Table 7
Classification of active and inactive compounds included in test series using
Model 1
97.37
96.84
99.88
57.74
96.27
98
Compound
DP%a
Class Compound
DP%a
Class
Training inactive group
3-Episiostatin B
Thiacetazone
ꢀ99.70
ꢀ
ꢀ
Ganglefene
Metadiphenil
bromidum
ꢀ82.80
ꢀ
þ
Test active set
ꢀ98.00
22.93
1p
2p
3p
4p
5p
6p
7p
8p
9p
ꢀ60.50
92.94
67.00
83.82
99.68
99.80
99.99
99.81
ꢀ98.10
ꢀ
þ
þ
þ
þ
þ
þ
þ
ꢀ
10p
11p
12p
13p
14p
15p
16p
17p
95.58
þ
þ
þ
þ
þ
þ
þ
þ
33.37
84.37
84.02
92.52
99.76
97.91
10.43
TBHQ
Cloral betaine
Vernelan
55.71
ꢀ99.40
ꢀ99.00
ꢀ94.10
ꢀ96.60
ꢀ86.30
ꢀ95.60
ꢀ51.10
þ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Quateron
Pancuronium
Ethylene
ꢀ83.80
ꢀ91.30
ꢀ96.60
ꢀ99.90
ꢀ62.50
ꢀ98.30
ꢀ89.40
ꢀ98.50
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Cetohexazine
Carbavin
Dioxychlorane
Aliflurane
Phenacemide
Tetharbital
Brofoxine
Vinyl ether
Tiouracilo
Thiamazol
Test inactive set
Amantadine
Mizoribine
Triclofos
methyl iodide
Diclofutime
mesilate
ꢀ60.70
ꢀ90.20
ꢀ100
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
þ
ꢀ
ꢀ
ꢀ
ꢀ
Cyclopropane
Basedol
ꢀ95.90
ꢀ98.30
ꢀ99.80
ꢀ99.10
ꢀ99.90
73.42
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
þ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Norantoin
ꢀ73.10
ꢀ
ꢀ65.30
ꢀ
Mipimazole
Didym levulinate
Metriponate
Prasterone
Orotonsan Fe
Ferrocholinate
ꢀ96.20
ꢀ99.90
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
þ
Percloroetane
Lindane
Nitrodan
ꢀ99.20
ꢀ98.10
ꢀ99.90
ꢀ98.80
ꢀ91.40
50.09
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
þ
ꢀ
ꢀ
ꢀ
Nitroinosite
Methenamine
Cobalti glutamas
Cobalti besilas
Canrenone
ꢀ100
ꢀ100
Ferrosi ascorbas ꢀ64.20
ꢀ99.00
ꢀ94.20
91.98
Arecoline
ꢀ99.20
ꢀ97.80
ꢀ90.40
ꢀ99.40
ꢀ50.80
42.40
Ascaridole
Pyrantel tartrate
Fentanilo
Febensamin
Guanazole
ꢀ99.30
ꢀ100
Butanolum
Etamsylate
Sango-Stop
Besunide
Urea
Pallirad
ꢀ99.60
ꢀ99.80
ꢀ90.90
ꢀ87.50
Fluorembichin
Mitoguazone
Acetylcholine
Methacholine
chloride
ꢀ99.90
ꢀ100
Tenalidine tartrate
Dioxoprometazine
N-Hidroxymetil-
N-metilurea
2,4,5-Triclofenol
Norgamem
Furtrethonium
iodide
ꢀ97.60
ꢀ99.80
ꢀ99.90
Quimbosan
Glicondamide
ꢀ100
Spironolactone
ꢀ100
Glycerol
Propamin soviet ꢀ98.80
ꢀ99.20
ꢀ
ꢀ
ꢀ
28.40
ꢀ99.90
ꢀ99.80
þ
ꢀ
ꢀ
RMI 11894
Barbismetylii
iodidum
Frigen 113
a
ꢀ65.30
ꢀ100
ꢀ
ꢀ
Dopamine
Ampicillin
5.41
ꢀ8.03
þ
ꢀ
Cystamine
ꢀ99.90
ꢀ98.30
ꢀ
Kanamycin A
ꢀ99.7
ꢀ
Amifostine
Adeturon
ꢀ100
ꢀ100
ꢀ
ꢀ
Isofenefrine
Phenylethanolamine ꢀ49.20
ꢀ99.20
ꢀ
ꢀ
Results of the classification of compounds obtained from Eq. (1) (using
non-stochastic quadratic indices): DP% ¼ [P(active) ꢀ P(inactive)] ꢁ 100.

1806
V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
5.2.7. In vivo assays with L. amazonensis
order to test the predictive ability of the obtained model. In
this case the system was able to perform well in 89.36% of
the cases. Only two from 17 active compounds (88.23%)
and three from 30 inactive ones (90.00%) were misclassified.
Active compounds against Leishmania, which were consid-
ered in this study, are representatives of families with diverse
structural patterns. Figs. 2 and 3 show the whole active set (train-
ing and test groups) collected from the literature. The selected
inactive group included antivirals, sedative/hypnotics, diuretics,
anticonvulsivants, haemostatics, oral hypoglycaemics, antihy-
pertensives, antihelminthics, anticancer compounds as well as
some other kinds of drugs, ensure a large structural variability.
Classification of compounds in both, training and active
groups are shown in Tables 6 and 7.
5.2.7.1. Animals. Male and female BALB/c mice bred in the
´
animal colony at IICS (Asuncion, Paraguay) were used. Six
groups of 10e13 animals of 6e8 weeks (approx. 18e20 g)
were inoculated in the right hind footpad with 2 ꢁ 106 amas-
tigotes obtained from donor hamsters. The parasites were de-
livered in 100 mL of phosphate buffered saline (PBS). Lesion
development was monitored by serial measurements of foot-
pad thickness and expressed as the difference between the
infected and uninfected footpad. The six groups were distrib-
uted as follows: Group I: infected control. Group II: reference
drug; N-methylglucamine antimoniate (GlucantimeÒ Aventis
France) with pentavalent antimony Sbv was dissolved in 50 mL
of PBS and administered by subcutaneous route (100 mg/
kg/d). Group III: compound 2 was made up in 50 mL of PBS
and 5 mL of polysorbate (Tween 80, OSI France) and adminis-
tered by oral route (10 mg/kg/d). Group IV: compound 2 was
administered by oral route (25 mg/kg/d). Group V: compound
2 was administered by subcutaneous route (10 mg/kg/d). Group
VI: compound 2 was administered by subcutaneous route
(25 mg/kg/d).
Acknowledgements
This work was supported by grants from PDT (Project No.
29/194), PEDECIBA (Programa de Desarrollo de las Ciencias
´
Basicas, URU/06/004). We wish to express our gratitude to
Prof. Dr. Gustavo Seoane and Dr. Luis Rivas for helpful dis-
cussion and for critical review of the manuscript.
5.2.7.2. Treatment. The treatment started four weeks after in-
oculation once the infection was well-established and lesion
was apparent and carried pit for 15 days. The animals were
killed one week after the end of the treatment to assess para-
sitological burden in the infected footpad. Granulomas were
extracted, weighed and prepared as usual. The number of
amastigotes per host lesion cell nucleus was counted. Percent-
age of parasite suppression was calculated from the ratio of the
counts of drug-treated mice and the infected untreated control
mice multiplied by 100. A corrected parasite suppression in-
dex was calculated by the following formula: 100-mean para-
sites treated mice/mean parasites untreated mice ꢁ 100.
References
[1] S.L. Croft, M.P. Barrett, J.A. Urbina, Trends Parasitol. 21 (2005)
508e512.
[2] S.L. Croft, S. Sundar, A.H. Fairlamb, Clin. Microbiol. Rev. 19 (2006)
111e126.
[3] R.B. Tesh, Isr. J. Med. Sci. 25 (1989) 214e217.
[4] D. Young, M. Duncan, Mem. Amer. Ent. Inst. 54 (1994) 1e881.
[5] http://www.who.int/tdr/diseases/leish/diseaseinfo.htm.
[6] http://www.who.int/tdr/dw/leish2004.htm.
[7] P. Desjeux, Comp. Inmunol. Microbiol. Infect. Dis. 27 (2004) 305e318.
[8] R. Lira, S. Sundar, A. Makharia, R. Kenney, A. Gam, E. Saraiva,
D. Sacks, J. Infect. Dis. 180 (1999) 564e567.
[9] S.L. Croft, G.H. Coombs, Trends Parasitol. 19 (2003) 502e508.
´
[10] S. Azzouz, M. Maache, R. Garcıa, A. Osuna, Basic Clin. Pharmacol.
5.3. Statistical analysis
Toxicol. 96 (2005) 60e65.
[11] F.R. Opperdoes, J. Bioenerg. Biomembr. 26 (1994) 145e146.
´
[12] S. Muelas-Serrano, J.J. Nogal, R.A. Martinez-Dıaz, J.A. Escario,
For the statistical analysis in biological assays, means ꢂ
standard deviation (SD) were used and calculated by the
Microsoft Office Excel 2003 software, unless indicated other-
wise. Multiple comparisons between groups were made with
a one-way analysis of variance (ANOVA). Paired comparisons
between groups were carried out by Student’s t-test. P values
of <0.05 were considered significant (two tailed).
´
A.R. Martinez-Fernandez, A. Gomez-Barrio, J. Ethnopharmacol. 71
´
(2000) 101e107.
[13] S. Muelas-Serrano, J.J. Nogal, A. Gomez-Barrio, Parasitol. Res. 86
(2000) 999e1002.
[14] J.H.V. Drie, M.S. Lajiness, Drug Discov. Today 3 (1998) 274e283.
[15] W.P. Walters, M.T. Stahl, M. Murcko, Drug Discov. Today 3 (1998)
160e178.
[16] H. Zinsmeister, H. Becker, T. Eicher, Angew. Chem., Int. Ed. Engl. 30
(1991) 130e147.
5.4. Topological molecular computer design
(TOMOCOMD) approach
[17] Y. Hernandez-Romero, L. Acevedo, L. Sanchez Mde, W.T. Shier,
H.K. Abbas, R. Mata, J. Agric. Food Chem. 53 (2005) 6276e6280.
[18] C. Niu, J.B. Qu, H.X. Lou, Chem. Biodivers. 3 (2006) 34e40.
[19] M.M. Salem, K.A. Werbovetz, Curr. Med. Chem. 13 (2006) 2571e2598.
[20] X. Zhang, J.K. Xu, J. Wang, N.L. Wang, H. Kurihara, S. Kitanaka,
X.S. Yao, J. Nat. Prod. 70 (2007) 24e28.
A Non-Stochastic Quadratic Fingerprints-based approach is
introduced to classify and to identify, in a rational way, new
antileishmanial compounds. A data set of 168 organic-chemi-
cals [45], 77 with antileishmanial activity and 91 having other
clinical uses, was divided at random in training and test data
sets. Making use of the LDA technique implemented in the
STATISTICA software [46], the linear model (Eq. (1)) was
obtained. An external validation procedure was conducted in
[21] Y. Asakawa, Phytochemistry 56 (2001) 297e312.
´
[22] Y. Asakawa, A. Bardon, Rev. Latinoamer. Quim. 29 (2001) 5e22.
[23] J. Gorham, Phytochemistry 16 (1977) 249e253.
[24] A. Basile, S. Giodano, S. Sorbo, M.L. Vuotto, M.T.L. Yelpo, R. Castaldo
Cobianchi, Int. J. Pharm. 36 (1998) 25e28.
[25] S. Kunz, H. Becker, Phytochemistry 36 (1994) 675e677.

V. Roldos et al. / European Journal of Medicinal Chemistry 43 (2008) 1797e1807
1807
´
[26] V. Lopez, E. Pandolfi, G. Seoane, Synthesis 10 (2000) 1403e1408.
[38] E.C. Torres-Santos, D. Lima Moreira, M.A. Kaplan, M.N. Meirelles,
B. Rossi-Bergmann, Antimicrob. Agents Chemother. 43 (1999) 1234e1241.
[39] L. Kedzierski, J.M. Curtis, M. Kaminska, J. Jodynis-Liebert, M. Murias,
Parasitol. Res. (2007). doi:10.1007/s00436-007-0729-y.
´
´
~
[27] D. Gamenara, E. Pandolfi, J. Saldana, L. Domınguez, M. Martınez,
G. Seoane, Arzneim.-Forsh./Drug. Res. 51 (2001) 506e510.
[28] F. Cullmann, H. Becker, E. Pandolfi, E. Roeckner, T. Eicher, Phytochem-
istry 45 (1997) 1235e1247.
[29] R. Boden, Synthesis (1975) 784.
[30] V. Roldos, E. Pandolfi, A. Rojas de Arias, M.E. Ferreira, A. Inchaustti, H.
Nakayama, U.R. Patent, Act No. 27607, 2002.
[40] G. Schmeda-Hirshmann, L. Astudillo, J. Bastida, C. Codina, A. Rojas de
Arias, M.E. Ferreira, A. Inchausti, G. Yaluff, J. Pharm. Pharmacol. 53
(2001) 563e567.
´
´
[41] M. Rolon, C. Vega, J.A. Escario, A. Gomez Barrio, Parasitol. Res. 99
´
[31] E. Alonso, D.J. Ramon, M. Yus, J. Org. Chem. 62 (1997) 417e421.
[32] S. Huneck, Tetrahedron 32 (1976) 109e113.
(2006) 103e107.
[42] A. Fournet, M.E. Ferreira, A. Rojas de Arias, S. Torres de Ortız,
´
[33] Y. Marrero-Ponce, V. Romero, TOMOCOMD Software, Version 1.0,
TOMOCOMD (TOpological MOlecular COMputer Design) for Win-
dows, Central University of Las Villas, 2002.
A. Inchausti, G. Yaluff, W. Quilhot, E. Fernandez, M.E. Hidalgo,
Comp. Biochem. Physiol. 116 (1997) 51e54.
´
[43] M.E. Ferreira, A. Rojas de Arias, S. Torres de Ortız, A. Inchausti,
´
[34] M.C. Vega, M. Rolon, A.R. Martınez-Fernandez, J.A. Escario,
´
H. Nakayama, C. Thovenel, R. Hocquemiller, A. Fournet, J. Ethnophar-
macol. 80 (2002) 199e202.
´
A. Gomez-Barrio, Parasitol. Res. 95 (2005) 296e298.
[35] A. Montero-Torres, M.C. Vega, Y. Marrero-Ponce, M. Rolon, A. Gomez-
Barrio, J.A. Escario, V.J. Aran, A.R. Martinez-Fernandez, A. Meneses-
Marcel, Bioorg. Med. Chem. 13 (2005) 6264e6275.
[44] H. Nakayama, F. Loisseau, C. Bories, S. Torres de Ortiz, A. Schinini,
E. Serna, A. Rojas de Arias, A.S. Fakhfakh, X. Franck, B. Figadere,
R. Hocquemiller, A. Fournet, Antimicrob. Agents Chemother. 49 (2005)
4950e4956.
~
[36] A. Fournet, A. Angelo, V. Munoz, R. Hocquemiller, F. Roblot, A. Cave,
P. Richomme, P. Bruneton, Phytother. Res. 8 (1994) 174e178.
´
[45] J.A. Valderrama, C. Zamarano, M.F. Gonzalez, E. Prina, A. Fournet,
´
[37] A. Gomez-Barrio, R.A. Martinez-Diaz, J. Atienza, J.A. Escario,
Bioorg. Med. Chem. 13 (2005) 4153e4159.
[46] STADISTICA Software, Version 5.5 for Windows, Statsoft Inc., 1999.
~
C. Diego, C. Avendano, Res. Rev. Parasitol. 57 (1997) 25e31.