Synthesis, anti-Trypanosoma cruzi activity and micelle interaction studies of bisguanylhydrazones analogous to pentamidine

Article abstract of DOI:10.1016/j.ejmech.2004.07.001

Three new bisguanylhydrazones analogous to pentamidine were synthesized, fully characterized and tested as anti-Trypanosoma cruzi candidates. Contrary to literature reports, that bicationic compounds are more active than monocationic compounds against Try

Full text of DOI:10.1016/j.ejmech.2004.07.001

European Journal of Medicinal Chemistry 39 (2004) 925–929
www.elsevier.com/locate/ejmech
Original article
Synthesis, anti-Trypanosoma cruzi activity and micelle interaction studies
of bisguanylhydrazones analogous to pentamidine
Márcia Narcizo Borges a,b, Jorge Cardoso Messeder a, José Daniel Figueroa-Villar a,*
a Departamento de Química, Instituto Militar de Engenharia, Praça General Tibúrcio 80, 22290-270 Rio de Janeiro, Brazil
b Departamento de Química Orgânica, Instituto de Química, Campus do Valonguinho, Universidade Federal Fluminense, Niterói, Rio de Janeiro, Brazil
Received 22 October 2003; received in revised form 13 May 2004; accepted 5 July 2004
Available online 12 August 2004
Abstract
Three new bisguanylhydrazones analogous to pentamidine were synthesized, fully characterized and tested as anti-Trypanosoma cruzi
candidates. Contrary to literature reports, that bicationic compounds are more active than monocationic compounds against Trypanosoma
brucei, it was found that these bisguanylhydrazones are much less effective against T. cruzi than simple aromatic monoguanylhydrazones, thus
suggesting different mechanism of action for both parasites. Spin-spin nuclear relaxation studies of the interaction of these compounds with
SDS and CTAB micelles showed that only the most trypanocidal compound displays significant discrimination between anionic and cationic
micelles.
© 2004 Elsevier SAS. All rights reserved.
Keywords: Bisguanylhydrazones; Micelle; Chagas disease; NMR relaxation
1. Introduction
binders [9–12]. On the other hand, using NMR, it has been
shown that the in vitro anti-T. cruzi activity of the aromatic
guanyl hydrazones is related to their selectivity to bind to
anionic SDS micelles in relation to cationic CTAB micelles,
which were used as a membrane models [8]. That work
showed that the differences in spin–spin relaxation times
between the conditions with the presence of SDS or CTAB
micelles are directly related to the IC50 values of the guanyl-
hydrazones [8], thus strongly suggesting that the mechanism
of anti-T. cruzi activity of these compounds involves interac-
tion with the parasite membrane. Also, guanylhydrazones,
being more active than gentian violet against the blood forms
of T. cruzi in contaminated blood, and not having deleterious
effects on the blood cells at the used concentrations [6], have
also been shown to interact weakly with plasmatic proteins
[12,13], therefore, making them promising agents for the
chemical prophylaxis of blood in blood banks. In this work,
we have prepared and tested against T. cruzi three novel
bisguanylhydrazones analogous to pentamidine and studied
their interaction with SDS and CTAB micelles.
Bicationic antibiotics, especially bisamidines [1,2] and
bisguanylhydrazones [3] have been shown to be active
against African trypanosomiasis. Among the bicationic anti-
biotics, until 2002 only the bisamidines have been tested
against Trypanosoma cruzi, the causative agent of American
trypanosomiasis [4,5]. It was only recently that biguanidines
and “reversed” diamidino compounds were also tested
against T. cruzi and L. donovani [5]. We have previously
shown that aromatic guanyl hydrazones exhibit in vitro
anti-T. cruzi activity [6]. The mechanism of action of these
compounds is not clear yet, but it has been suggested that it
may involve their interaction with the parasite DNA and/or
its cellular membrane [7,8]. A molecular modeling study
using docking methodologies showed that monoguanylhy-
drazones have great potential to interact with DNA, and that
they possibly do that at the DNA minor groove, with some
specificity forAT rich regions [7]. These molecular modeling
results agree with experimental DNA binding studies carried
out with African trypanocidal bicationic compounds berenil
and pentamidine, which have been shown to be minor groove
2. Chemistry
The compounds were prepared by alkylation of
4-hydroxybenzaldehyde with the appropriate dibromoalkane
* Corresponding author.
E-mail address: figueroa@ime.eb.br (J.D. Figueroa-Villar).
0223-5234/$ - see front matter © 2004 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2004.07.001

926
M.N. Borges et al. / European Journal of Medicinal Chemistry 39 (2004) 925–929
are more active than the compounds with an even number of
CH2 groups [17]. Interestingly, the bisguanylhydrazones
tested here are less active against T. cruzi than most
monoguanylhydrazones tested in our previous work [6],
where some compounds displayed IC50 values as low as
17 µM. This is in opposition to literature reports on the
behavior of cationic antibiotics when tested against Trypano-
soma brucei, where an increase in the number of positively
charged groups leads to increased anti-parasite activity [3]. It
has been shown that bisamidines active against T. brucei
exert their action by interaction with the parasite DNA [18].
This difference in behavior regarding the biological activity
of bicationic compounds, when comparing the literature re-
sults obtained with African trypanosomiasis with our results
obtained for American trypanosomiasis, suggests that their
mechanism of action against T. cruzi is different, and possi-
bly related to the interaction of the compounds with the
parasite membrane, instead of interaction with the DNA [8].
In order to model the interaction of the cationic com-
pounds with membranes we carried out NMR longitudinal
relaxation studies [8] in the presence of cationic (CTAB) and
anionic (SDS) micelles as membrane models. The relaxation
studies were done using 2.5 mM samples of the products, to
make sure that the measurements were carried out with stable
solutions of all the compounds. Also, this concentration is
close to the determined IC50 values. For these studies, the
chosen tensoactive agent concentration was 20 mM, to en-
sure that we were working well above the critical micellar
concentration (CMC) of SDS and CTAB [19]. The longitu-
dinal relaxation (T1) values for all observed hydrogens of the
bisguanylhydrazones are summarized in Table 2 and Fig. 2.
It must be noticed that we were not able to measure the
relaxation time of the hydrogens of the junction alkyl chain
of the bicationic compounds because their signals are ob-
scured or superimposed with either the water signal or the
tensoactive agent signals. The data from Table 2 show that
the bisguanylhydrazones interact more strongly with the
SDS micelles than with the CTAB micelles. This is somehow
expected, as the interaction should be much favored by the
electrostatic attraction between the cationic compounds and
the anionic micelles. However, in our previous work we
showed that some monoguanylhydrazones, despite their
positive charge, do not interact preferentially with anionic
SDS micelles as compared to cationic CTAB micelles [8]. A
better way to compare the relaxation behavior of the different
compounds is using Fig. 2, which shows the behavior of all
the observed hydrogens for the three bisguanylhydrazones
under all the studied conditions.
Fig. 1. Synthesis of the bisguanilhydrazones.
followed by conversion of the resulting dialdehyde to the
respective bisguanylhydrazone by treatment with ami-
noguanidine hydrochloride, as shown in Fig. 1.
The alkylation of 4-hydroxybenzaldehyde with the dibro-
moalkanes, using K2CO3 in methanol, was appropriate for
the preparation of compounds 2 and 3 [14]. However, in
order to obtain 1 it was necessary to use sodium ethoxide in
ethanol as base. Dialdehydes 1 and 2 were easily converted to
the respective bisguanylhydrazones by treatment with ami-
noguanidine hydrochloride in 95% ethanol containing a few
drops of fuming hydrochloric acid [3,6]. On the other hand,
the conversion of dialdehyde 3 to the respective bisguanylhy-
drazone required the use of a Dean–Stark apparatus using
toluene as solvent and
a
catalytic amount of
p-toluenesulfonic acid [6]. The bisguanylhydrazones are in-
soluble in all common organic solvents and sparingly soluble
in water. Their purification was better accomplished by re-
peatedly washing with an appropriate solvent. The purity of
the samples obtained in this way was checked out by 1H and
13C NMR using concentrated samples (0.1 M) in DMSO-d6
and elemental analysis.
3. Pharmacology and discussion
Once the final products were purified and fully character-
ized, we carried out in vitro bioassays using 7 days trypomas-
tigote forms of Y-strain T. cruzi [15]. These tests were con-
ducted using the procedure developed by Brener [16] and
afforded the inhibitory concentration as summarized in
Table 1.
The data in Table 1 show that the compound with three
CH2 at the middle chain, compound 5, is the most active. This
result with T. cruzi follows the same trend as those found in
the results obtained for pentamidine and analogues against
African trypanosomiasis, where the compounds with an odd
number of CH2 groups in the middle chain (n = 3 and n = 5)
Table 1
Trypanocidal activity for the bisguanylhydrazones against the trypomastigote forms of Y-strain T. cruzi
Compound
Concentration (mg/ml)
IC50 (mM)
5.0
3.5
100
95
2.5
0
0.50
0
0.25
0
4
5
6
>100
93
56
46
3.6
0
1.2 0.1
5.6 0.4
5.5

M.N. Borges et al. / European Journal of Medicinal Chemistry 39 (2004) 925–929
927
Table 2
1H T1 values (in seconds) for all the bisguanylhydrazones (2.5 mM) in D2O and in D2O containing SDS or CTAB (20 mM)
Compound
Solutions
T1 (s)
H-2
H-3
H-5
H-6
H-7
4
D2O
1.895
0.882
1.013
1.442
0.453
1.295
0.722
0.573
1.327
–0.032
1.313
0.599
0.714
1.096
0.217
1.311
0.632
0.679
1.343
–0.032
1.222
0.570
0.652
1.053
0.169
0.791
0.467
0.324
1.104
–0.313
1.311
0.632
0.679
1.343
–0.032
1.222
0.570
0.652
1.053
0.169
0.791
0.467
0.324
1.104
–0.313
1.895
0.882
1.013
1.442
0.453
1.295
0.722
0.573
1.327
–0.032
1.313
0.599
0.714
1.096
0.217
2.486
0.972
1.514
1.966
0.520
1.654
0.908
0.746
1.576
0.078
1.627
0.781
0.846
1.237
0.390
SDS + D2O
dT1 (SDS)
CTAB + D2O
dT1 (CTAB)
D2O
5
6
SDS + D2O
dT1 (SDS)
CTAB + D2O
dT1 (CTAB)
D2O
SDS + D2O
dT1 (SDS)
CTAB + D2O
dT1 (CTAB)
Found average error = 0.03%.
Considering the data from Table 2 and Fig. 2, it can be
observed that all the bisguanylhydrazones interact with the
SDS micelles. In fact, the relaxation time changes in SDS are
greater for 4, which is the less active compound. Using the
relaxation measurements for H-7, which is the non-
exchangeable hydrogen closest to the cationic terminal, the
relaxation changes induced by the SDS micelles follow the
order 4 > 6 > 5 and do not show any correlation to the IC50
values. However, it can be observed that compound 5 does
not interact significantly with the cationic micelles, as its
values for T1 are not affected by the presence of CTAB above
the CMC. Interestingly, this is not the case for compounds 4
and 6, which also show a reasonable degree of interaction
with the cationic micelles. Despite the reduced number of
examples used in this work, the interaction results suggest
that compounds that interact with the anionic micelles but do
not so with the cationic micelles, as it is the case of 5, are
more active that the compounds that, even interacting more
strongly with SDS micelles, do interact with both types of
micelles, as it is the case for compounds 4 and 6. This
observation seems to be in agreement with our previous
demonstration that simple guanylhydrazones are more active
when they are able to discriminate better between both types
of micelles [8].
interaction leads to more efficient relaxation pathways. How-
ever, for compound 6, it is observed that T1 actually increases
for hydrogens H-3 and H-5, suggesting a decrease on the
correlation time of those hydrogens.
4. Experimental protocols
4.1. Chemistry
All the organic starting materials used in this work were
Aldrich, and were used as purchased, without previous puri-
fication. The solvents were distilled and dried previous to
their use. Melting points were determined on a Fisher–Johns
apparatus and are uncorrected. Infrared spectra were ob-
tained on Perkin-Elmer, Models 1710 and 1420, spectropho-
tometers. The bisguanylhydrazones were prepared by reac-
tion of the respective aldehyde with aminoguanidine
hydrochloride in the presence of an acid catalyst [3,6].
4.1.1. 1,2-Bis(4-carboxyaldehydephenoxy)ethane 1
Dialdehyde 1 was prepared by reaction of 4-hydroxy-
benzaldehyde (5.0 g, 40 mmol) with 1,2-dibromoethane
(2.0 ml, 18 mmol) and sodium ethoxide (0.9 g, 40 matg of
sodium) in refluxing ethanol (25 ml) for 3 h. The crude
product obtained after cooling of the reaction mixture was
filtered by suction and washed with water to eliminate the
formed sodium bromide and the remaining solid was recrys-
tallized from 95% ethanol to afford pure 1 in 45% yield: m.p.
110–112 °C; IR (KBr) mmax 2950, 1760, 1670, 1300,
1240 and 1150 cm–1; 1H NMR (DMSO-d6, 300 MHz) d 9.80
(s, 2H), 7.91 (d, 9.7 Hz, 4H), 7.20 (d, 9.7 Hz, 4H) and 4.51 (s,
4H); 13C NMR (DMSO-d6, 75 MHz) d 191.3, 163.1, 131.8,
129.8, 115.0 and 66.6.
The normal variation of T1 when there is an increase in the
degree of intermolecular interaction is to decrease, as the
4.1.2. 1,3-Bis(4-carboxyaldehydephenoxy)propane 2
Dialdehyde 2 was prepared by refluxing a mixture of
4-hydroxybenzaldehyde (4.0 g, 30 mmol), 1,3-dibromo-
Fig. 2
. Differences between the relaxation times (T1) of the bisguanylhydra-
zones hydrogens in D2O and D2O + CTAB solutions.

928
M.N. Borges et al. / European Journal of Medicinal Chemistry 39 (2004) 925–929
propane (1.5 ml, 15 mmol) and K2CO3 (4.1 g, 30 mmol) in
dry acetone (30 ml). After the completion of the reaction
(18 h), the acetone was evaporated under vacuum and the
resulting solid washed with water to extract the formed KBr.
The crude solid product was then recrystallized from 95%
ethanol to afford pure 2 in 56% yield: m.p. 125–129 °C; IR
114.7, 64.5 and 28.5; EILRMS z/m (int. %) 396 (M+, 15) and
338 (100).Anal. Calc. for C19H26N8O2Cl2: C, 48.70; H, 5.60;
N 23.93 %. Found: C, 48.54; H, 5.53; N 23.78%.
4.1.5. 1,4-Bis(4-guanylhydrazonephenoxy) hydrochloride 6
Compound 6 was prepared by reaction of dialdehyde 3
with aminoguanidine hydrochloride in refluxing toluene with
p-toluenesulfonic acid in a Dean–Stark apparatus during 1 h.
The crude product was washed with boiling 95% ethanol and
filtered to afford pure 6 in 45% yield: m.p. 226–234 °C
(dec.); IR (KBr) mmax 3300, 1680, 1610, 1515, 1255 and
1170 cm–1; 1H NMR (DMSO-d6, 300 MHz) d 11.92 (bs, 2H),
8.08 (s, 2H), 7.85 (d, 7.6 Hz, 4H), 7.8 (bs, 8H), 7.05 (d,
7.5 Hz, 4H), 4.15 (bs, 4H) and 1.95 (bs, 4H); EILRMS z/m
(int. %) 410 (M+, 8) and 352 (100). Anal. Calc. for
C20H28N8O2Cl2: C, 49.77; H, 5.85; N 23.23 %. Found: C,
49.70; H, 5.79; N 23.15 %.
1
(KBr) mmax 2950, 1700, 1600, 1460, 1240 e 1155 cm–1; H
NMR (DMSO-d6, 300 MHz) d 9.88 (s, 2H), 7.95 (d, 8.9 Hz,
4H), 7.25 (d, 8.9 Hz, 4H), 4.44 (t, 6.5 Hz, 4H) and 2.3
(quintet, 6.4 Hz, 2H); 13C NMR (DMSO-d6, 75 MHz) d
190.8, 162.9, 131.3, 129.2, 114.5, 64.3 and 27.8.
4.1.3. 1,4-Bis(4-carboxyaldehydephenoxy)butane 3
Dialdehyde 3 was prepared in the same manner as 2, but
using 4-hydroxybenzaldehyde (4.0 g, 30 mmol), 1,4-
dibromobutane (2.4 ml, 17 mmol), K2CO3 (4.1 g, 30 mmol)
and a reaction time of 24 h. The crude product was purified
by recrystallization from carbon tetrachloride to afford pure
3 in 58% yield: m.p. 95–99 °C; IR (KBr) mmax 2950, 1700,
1675, 1600, 1440, 1240 and 1150 cm–1; 1H NMR (DMSO-
d6, 300 MHz) d 9.92 (s, 2H), 7.88 (d, 9.0 Hz, 4H), 7.15 (d,
9.0 Hz, 4H), 4.18 (bs, 4H) and 1.88 (bs, 4H); 13C NMR
(DMSO-d6, 75 MHz) d 190.8, 163.0, 131.3, 129.5, 114.5,
67.2 and 24.7.
4.2. NMR drug-micelle interaction
4.2.1. Solutions preparations
The relaxation measurements for the pure bisguanylhy-
drazones were carried using 2.5 mM solutions of the drugs in
D2O. The measurements with the micelles were carried out
using solutions with a drug concentration of 2.5 mM and a
tensoactive agent concentration of 20 mM, which is above
the CMC for either SDS (8.0 × 10–3 M at 25 °C) or CTAB
(9.2 × 10–4 M at 25 °C) [11].
4.1.4. 1,2-Bis(4-guanylhydrazonephenoxy)ethane
hydrochloride 4
Compound 4 was prepared by refluxing a mixture of
aldehyde 1 (0.38 g, 2.0 mmol) and aminoguanidine hydro-
chloride (0.46 g, 4.1 mmol) in 95% ethanol (20 ml) contain-
ing a few drops of fuming HCl during 4 h. The solid obtained
after eliminating the solvent under vacuum was washed with
boiling chloroform and filtered to afford pure 4 in 38% yield:
m.p. 300–305 °C (dec.); IR (KBr) mmax 3360, 1670, 1620,
1254, 1172 and 1047 cm–1, 1H NMR (DMSO-d6, 300 MHz)
d 11.82 (bs, 2H), 8.18 (s, 2H), 7.85 (d, 8.2 Hz, 4H), 7.72 (bs,
8H), 7.08 (d, 8.2 Hz, 4H) and 4.42 (s, 4H); 13C NMR
(DMSO-d6, 75 MHz) d 159.7, 154.7, 146.3, 128.9, 125.8,
114.3 and 66.1; EILRMS z/m (int. %) 381 (M+, 17) and 324
(100). Anal. Calc. for C18H24N8O2Cl2: C, 47.56; H, 5.33; N
24.67%. Found: C, 47.38; H, 5.24; N 24.46%.
4.2.2. NMR measurements
All the NMR measurements were carried out in a Varian
UNITY-300 NMR spectrometer (300 MHz) using 5 mm
sample tubes. The chemical shifts were determined relative
to the HOD signal (d 4.80 ppm) [20] of the solvent (D2O 98%
D, Aldrich). The T1 relaxation experiments were carried out
using the inversion recovery pulse sequence with presatura-
tion of the water signal with a decoupling power of 0.75 W.
The width of the 90° pulse for hydrogen was calibrated
before every set of measurements and was always close to
15 µs. The other pertinent acquisition parameters were: ac-
quisition time 3.774 s, spectral width 4000 Hz and
29952 data points for every transient. All the experiments
were conducted in triplicate at 36.0 0.1 °C using a relax-
ation delay of 12 s and collecting 68 transients for each FID.
4.1.5. 1,3-Bis(4-guanylhydrazonephenoxy)propane
hydrochloride 5
Compound 5 was prepared in the same manner as 4 but
using a reaction time of 3 h and the following amounts of
reagents: aldehyde 2 (0.8 g, 2.95 mmol), aminoguanidine
hydrochloride (0.7 g, 6.4 mmol) and 95% ethanol (30 ml)
containing a few drops of fuming HCl. The crude product
was washed with boiling 95% ethanol and filtered to afford
pure 5 in 45% yield: m.p. 300–310 °C (dec.); IR (KBr) mmax
4.3. Pharmacology
4.3.1. Bioactivity tests
Bloodstream trypomastigotes forms of Y-strain T. cruzi
[12] were obtained from Swiss albino female mice, weighing
18–20 g, at the peak of parasite concentration, 7 days after
intra-peritoneal inoculation. To determine the in vitro activ-
ity, the desired amount of each compound was dissolved in
1000 µl of DMSO, and 10 µl of this solution was mixed with
390 µl of blood from the acutely infected mice. The maxi-
mum amount of DMSO used (2.5% in volume) did not have
1
3410, 1680, 1630, 1514, 1250 and 1170 cm–1; H NMR
(DMSO-d6, 300 MHz) d 11.96 (bs, 2H), 8.08 (s, 2H), 7.82 (d,
9.0 Hz, 4H), 7.55 (bs, 8H), 7.15 (d, 9.0 Hz, 4H), 4.22 (t,
6.5 Hz, 4H) and 2.20 (quintet, 6.5 Hz, 2H); 13C NMR
(DMSO-d6, 75 MHz) d 160.4, 155.3, 146.6, 129.3, 126.1,

M.N. Borges et al. / European Journal of Medicinal Chemistry 39 (2004) 925–929
929
any significant effect on trypomastigote growth. One hun-
dred and fifty microliter of the suspension was distributed
into a sterile multi-well plate, which was incubated for 24 h at
4 °C. After the incubation, 5 µl of the suspension were
examined with a microscope for the presence of motile or-
ganisms: 50 microscopic fields were examined at 400 mag-
nification according to the method of Brener [16]. The num-
ber of parasites still present and their morphology were noted
and scored against a negative control (parasite suspension
and parasite suspension plus 2.5% DMSO solution), and a
positive control (parasite suspension and 18 µM gentian
violet). The values of ID50, the drug concentration (µM)
necessary to kill 50% of the parasites, were obtained by
linear and polynomial regression analysis [21].
chemotherapy of Chagas disease than the development of
new monocationic antibiotics.
References
[1] P. Ulrich, A. Cerami, J. Med. Chem. 27 (1984) 35–40.
[2] M. Cory, R.R. Tidwell, T.A. Fairley, J. Med. Chem. 35 (1992) 431–
438.
[3] P. Ulrich, R.W. Grady, A. Cerami, Drug Dev. Res. 2 (1982) 219–228.
[4] J.R. Baker, L.F. Selden, Trans. Roy. Soc. Trop. Med. Hyg. 75 (1981)
80–85
W.E. Gutteridge, Trop. Disease Bull. 73 (1976) 699.
[5] C.E. Stephens, R. Brun, M.M. Salem, K.A. Werbovetz, F. Tanious,
W.D. Wilson, D.W. Boykin, Bioorg. Med. Chem. Lett. 13 (2003)
2065–2069.
[6] J.C. Messeder, L.W. Tinoco, J.D. Figueroa-Villar, E.M. Souza,
R. Santa-Rita, S.L. de Castro, Bioorg. Med. Chem. Lett. 5 (1995)
3079–3084.
[7] O. Santos-Filho, M.T. Araujo, J.D. Figueroa-Villar, Bioorg. Med.
Chem. Lett. 7 (1997) 1797–1802.
5. Conclusions
Bisguanylhydrazones are much less active anti-T. cruzi
compounds than simple guanylhydrazones. This result being
is in opposition to what have been reported in the literature
for cationic antibiotics used against T. brucei, where an
increase in the number of positively charged groups increases
the activity [1]. This result strongly suggests that the mecha-
nism of action of this type of compound is different for both
parasites. We believe that in the case of American trypanoso-
miasis the most likely mechanism is the interaction of the
cationic compounds with the parasite membrane, disrupting
it or changing its permeation properties, rather than interac-
tion with the parasite DNA [8], as its have been shown to be
the case for bicationic compounds in African trypanosomia-
sis. Longitudinal relaxation studies indicate that bisguanyl-
hydrazones do interact with the SDS micelles used as models
for the parasite membrane, with the most active compound
being the one whose relaxation properties are less affected by
the presence of cationic CTAB micelles.
[8] M.N. Borges, J.D. Figueroa-Villar, Biopolymers (Biospectroscopy)
62 (2001) 9–14.
[9] T.C. Jenkins, A.N. Lane, S. Neidle, D.G. Brown, Eur. J. Biochem. 213
(1993) 1175–1184.
[10] K.J. Edwards, T.C. Jenkins, S. Neidle, Biochem. 31 (1992) 7104–
7109.
[11] M. Durand, E. Seche, J.C. Maurizot, Eur. Bioph. J. Bioph. Lett. 30
(2002) 625–630.
[12] L.W. Tinoco, J.D. Figueroa-Villar, J. Braz. Chem. Soc. 10 (1999)
281–286.
[13] L.W. Tinoco, J.D. Figueroa-Villar, Spectrosc. Lett. 32 (1999) 941–
951.
[14] C.F.H. Allen, J.W. Gates Jr., Org. Synt. Coll. 3 (1966) 140–143.
[15] L.H.P. Silva, V. Nussenzweig, Folia Clin. Biol. 20 (1953) 191–208.
[16] Z. Brener, Rev. Inst. Med. Trop. 4 (1962) 389–396.
[17] I.O. Donkor, S.K. Jones, R.R. Tidwell, Bioorg. Med. Chem. Lett. 3
(1993) 1137–1140.
[18] A. Kumar, D.W. Boykin, W.D. Wilson, S.K. Jones, B.C. Bender,
C.C. Dykstra, J.E. Hall, R.R. Tidwell, Eur. J. Med. Chem. 31 (1996)
767–773.
[19] J.H. Fendler, Membrane Mimetic Chemistry, John Wiley and Sons,
New York, 1982.
[20] B. Almanac (Ed.), Bruker Analytical Instruments 1993, Germany.
[21] S. Chaterjee, B. Price, Regression Analysis by Example, John Wiley
& Sons, New York, pp. 270–274.
One important conclusion of this paper is that the work on
the development of bicationic compounds active against T.
cruzi seems to be less likely to produce useful results for the