Design, synthesis and biological evaluation of new potent 5-nitrofuryl derivatives as anti-Trypanosoma cruzi agents. Studies of trypanothione binding site of trypanothione reductase as target for rational design

Article abstract of DOI:10.1016/j.ejmech.2004.02.007

Design, using force-field calculations on the catalytic site of trypanothione reductase from Trypanosoma cruzi, has led to the development of new 5-nitrofuryl derivatives as potential anti-trypanosomal agents. The synthesized compounds were tested in vitro against T. cruzi and more than 75% of the prepared derivatives showed higher activity than nifurtimox. Compounds 5 and 11, hexyl 4-(5-nitrofurfurylidene)carbazate and N-hexyl 3-(5-nitrofuryl) propenamide, showed the highest in vitro trypanocidal effect reported to date for members of the nitrofuran family. Partition coefficients and energies for the single-electron reduction of compounds were theoretically determined. These properties could be not the major cause of the activities' differences. The physicochemical environment around E19, W22, C53 and Y111 residues within the trypanothione binding site of trypanothione reductase resulted a valuable target for the rational design of anti-trypanosomal drugs.

Full text of DOI:10.1016/j.ejmech.2004.02.007

European Journal of Medicinal Chemistry 39 (2004) 421–431
www.elsevier.com/locate/ejmech
Original article
Design, synthesis and biological evaluation of new potent 5-nitrofuryl
derivatives as anti-Trypanosoma cruzi agents. Studies of trypanothione
binding site of trypanothione reductase as target for rational design
a
a
a,
a
Gabriela Aguirre , Eliana Cabrera , Hugo Cerecetto *, Rossanna Di Maio ,
a
a
b
c
Mercedes González , Gustavo Seoane , Adelina Duffaut , Ana Denicola ,
d
d
María J. Gil , Victor Martínez-Merino
a
Departamento de Química Orgánica, Facultad de Química—Facultad de Ciencias, Iguá 4225, 11400 Montevideo, Uruguay
b
o
Liceo N 1 Río Negro, Consejo de Enseñanza Secundaria, Fray Bentos, Uruguay
Laboratorio de Fisicoquímica Biológica, Facultad de Ciencias, Iguá 4225, 11400 Montevideo, Uruguay
c
d
Departamento de Química Aplicada, Universidad Pública de Navarra, 31006 Pamplona, Spain
Received 1 September 2003; received in revised form 21 February 2004; accepted 25 February 2004
Abstract
Design, using force-field calculations on the catalytic site of trypanothione reductase from Trypanosoma cruzi, has led to the development
of new 5-nitrofuryl derivatives as potential anti-trypanosomal agents. The synthesized compounds were tested in vitro against T. cruzi and
more than 75% of the prepared derivatives showed higher activity than nifurtimox. Compounds 5 and 11, hexyl 4-(5-
nitrofurfurylidene)carbazate and N-hexyl 3-(5-nitrofuryl)propenamide, showed the highest in vitro trypanocidal effect reported to date for
members of the nitrofuran family. Partition coefficients and energies for the single-electron reduction of compounds were theoretically
determined. These properties could be not the major cause of the activities’differences. The physicochemical environment around E19, W22,
C53 and Y111 residues within the trypanothione binding site of trypanothione reductase resulted a valuable target for the rational design of
anti-trypanosomal drugs.
©
2004 Elsevier SAS. All rights reserved.
Keywords: 5-Nitrofuryl derivatives; Anti-trypanosomal compounds; Structure–activity relationship
1
. Introduction
dependent flavoenzyme responsible for many trypanosome
protections against free radicals. TR occurs exclusively in
trypanosomatids, which lack a glutathione reductase (hGR,
EC: 1.6.4.2, the human equivalent anti oxidative stress en-
zyme), and many authors have indicated TR to be one of the
most promising targets for research on trypanocidal drugs
[8–11]. Thus, numerous papers have been published on the
determination of the experimental inhibitory activity towards
TR of several families of compounds and the relationships
with their in vitro trypanocidal activity [10,12–16]. In gen-
eral, these studies did not reveal any tangible correlation
[10,12–14] or show major exceptions [16] between TR inhi-
bition and in vitro trypanocidal effect against T. cruzi. Rapid
transformations of the compounds in the biological medium
A great deal of attention has been focused on developing
efficient chemotherapeutic agents against Chagas’ disease,
which is produced by Trypanosoma cruzi [1–3]. Current
treatment is based on benznidazole (a nitroimidazole deriva-
tive) or nifurtimox (a nitrofuran derivative, Nfx, Scheme 1),
compounds that cause side effects and show poor clinical
efficacy [4–7]. These two drugs are effective against the
circulating form of the parasite (trypomastigotes) during the
acute phase of the disease, but not during the chronic stage.
There is, therefore, an urgent need for the development of
effective agents acting at key targets in T. cruzi. One of them
is trypanothione reductase (TR, EC 1.6.4.8) an NADPH-
[10,13,16] and drug delivery aspects, including cellular pen-
etration [14,15] amongst others, have been postulated to
explain the absence of the aforementioned straightforward
correlation. It seems that the crucial parameter for a high
*
Corresponding author.
E-mail addresses: hcerecet@fq.edu.uy (H. Cerecetto),
merino@unavarra.es (V. Martínez-Merino).
©
2004 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2004.02.007

422
G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
Me
O
N
N
X
O N
2
O
O N
2
O
6 Å
N
N
H
N
H
SO2
Nfx
1, X= Et
2
, X= OMe
1
4-17 Å
Scheme 1. Nifurtimox and lead compounds.
trypanocidal activity is the combination of both redox cy-
cling activity and inhibition of trypanothione reduction [12].
An alternative way to the experimental determination of
key parameters in the design of potential anti-chagasic com-
pounds could be the use of computational techniques to
obtain theoretical properties related with those parameters.
In this work, we present the design of anti-T. cruzi com-
pounds by means of computational chemistry calculations
and docking analysis on TR from T. cruzi (TcTR). Thus, the
new 5-nitrofuryl derivatives designed were evaluated against
epimastigote forms of T. cruzi, were also analyzed as poten-
tial ligands for TR and their lipophilicity and Gibbs free
energies for the single-electron reduction were theoretically
calculated.
vivo anti-parasitic activities against T. cruzi, this in vitro
model was found to be in agreement with the active site of
TcTR [27].
2
. Chemistry
2
.1. Design
Initially, based in our 3-D QSAR model, we designed
some 5-nitrofuryl derivatives with appropriate steric and
electronic properties (Scheme 1) [27]. So, we proposed car-
bazates 4–5, amides 10–11, ester 16–17 and aminoesters
2
1–22 (Scheme 2). Further, we analyzed their capacity to
interact with TR.
We have selected nitrofuran derivatives for this study,
because it is known [17] that they act as substrates for TR and
they also effectively inhibit enzymatic reduction of trypan-
othione disulphide (the enzyme’s physiological substrate).
Members of this family of compounds, as nifuroxazide, ni-
furoxime or nifurprazine, are subversive substrates (turncoat
inhibitors) of TR and they divert electrons from the physi-
ologically catalyzed reactions, while the respective reaction
with hGR is negligible [18]. Single-electron enzymatic re-
ductions of nitro-aromatic compounds to their anion radicals
are accompanied by futile redox cycling and oxidative stress,
since nitro radicals are rapidly reoxidized by oxygen with the
formation of superoxide. The trypanocidal activity of the
most potent TR inhibitors seems to be related with the spe-
cific release of superoxide anion radicals in the parasite, a
consequence of their redox cycling activity [12]. Indeed, this
was the suggested mode of action of the nitrofuran deriva-
tives as trypanocidals, exemplified by Nfx, which is reduced
to the corresponding nitro anion radical by T. cruzi cells
Reduction of 5-nitrofuryl derivatives may occur at the
active site of TR, or at different sites as already suggested for
chinifur [28]. Thus, we theoretically studied the docking
concerning nitrofurazone derivative 1 on different TcTR
sites. They showed that the active site (trypanothione site)
was the most energetically favorable point of interaction
between TcTR and nitrofurazone 1 (Fig. 1A) in detriment of
the other known studied sites, i.e. the ‘FAD site’ [29], the
‘NAD site’ [30] and the ‘third site’ [31] (for details, see
supplementary data). The most favorable disposition of 1 in
terms of the active site of TcTR, as shown in Fig. 1A, is that
in which the nitro group points towards the reactive hole
(S15-water-C53-V54-T335-H461′ residues). The semicar-
bazone group of 1 participates in two hydrogen bonds with
E19 and Y111 residues, and its N-tail (butyl group) fills a
hydrophobic hollow (W22-L18-M114). The distances (ob-
tained from X-ray data) [29] between significant points of
–
C53 (SH), Y111(OH), E19(COO ) and W22 (C H) residues
2
′
–
[19–22]. In agreement with the redox cycling behavior of
of the site are 8.2 Å (SH...OH), 11.7 Å (SH...O ), 16.5 Å
–
2′ 2′
(
(
SH...C H), 7.7 Å (OH...O ), and 10.0 Å (OH...C H)
nitrofuran derivatives, their electrochemical potentials were
directly correlated with the oxygen uptake under mitochon-
drial respiratory inhibition by cyanide, but were not corre-
lated with growth inhibition [23].
Fig. 1B). The four points indicated above belong to the same
plane (RMS distance of atoms from plane <0.01 Å). These
four interaction points in TcTR seem suitable for use in the
design of new trypanocidals because the active site of the
human equivalent anti oxidative stress enzyme (hGR) is
notably different. The disulphide holes in the TcTR and hGR
catalytic sites have the same kind of residues, but sites differ
in the fact that the anionic point in TcTR (E19 residue) is a
cationic point in hGR (R347) and, furthermore, the lipophilic
wall (W22) in TcTR is occupied by another cationic residue
in hGR (R37) [29,32]. In agreement with these distances,
4
In this work, N -butyl derivative of nitrofurazone (5-nitro-
2-furaldehyde semicarbazone) 1 was selected as the lead
compound (Scheme 1). We have recently explored nitrofura-
zone derivatives, where compounds 1 and 2 (Scheme 1)
afforded the best anti-chagasic profile [24,25]. Moreover,
nitrofurazone derivative 1 also proved to generate nitro anion
radicals [26]. On the other hand, derivative 1 was a good
example of the reported 3-D QSAR model on in vitro and in

G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
423
1
3
, R = Me
1
O
4, R = butyl
X= NO2, a
1
N
5, R = hexyl
OR1
O2N
O
N
H
1
6
, R = heptyl
1
7, R = octyl
X
O
CHO
X= H, b
c
OH
X
O2N
O
O
O2N
O
O
O
O
O
8
9
, X = -O-N
e
d
O
1
0, X= butylamino
1
1, X= hexylamino
OR3
3
O2N
12, X= -NH(CH2)2OMe
13, X= furfurylamino
14, X= -NHCH CO Et
1
5, R = Me
3
1
6, R = butyl
3
2
2
1
7, R = hexyl
O
f
1
8, R= α-naphtyl
19, R= β-naphtyl
0, R= H
N
R
O2N
O2N
O
N
H
N
H
X= O
2
O2N
X
CHO
NH2
g
2
1, X= O
OEt
X
22, X= S
X= O or S
O
1
Scheme 2. Preparation of derivatives 3–22: (a) H NHNCO R /toluene/p-TsOH/rt; (b) (1) malonic acid/Py/D, (2) Ac O/HNO /H SO (c)/0 °C; (c)
2
2
2
3
2
4
2
3
N-hydoxysuccinimide/DCC/CH Cl /0 °C to rt; (d) R NH /CH Cl /rt; (e) R OH(solvent)/p-TsOH/reflux; (f) H NHNCONHR/toluene/p-TsOH/rt; (g) ethylgly-
2
2
2
2
2
2
cinate hydrochloride/toluene/p-TsOH/rt.
carbazates 4–5 (Fig. 1C) or amides 10–11 (Fig. 1D) would
give rise to good interactions with the active site of the
enzyme on the same points as semicarbazone 1. Docking
study also showed that esters 16–17 despite losing a possible
hydrogen bond with E19 could also connect with the en-
zyme, as shown by the data in Table 1 (E, binding energy, see
Section 6). Even amino esters 21–22 (Table 1) could dock in
the active site of TcTR in a similar way to 1.
prepared according to procedures depicts in Scheme 2. Car-
bazate derivatives 3–7 were obtained in a very clean process
from the corresponding carbazate reactants (commercially
available in the case of R = Me or synthesized from the
corresponding alcohol [33]). On the other hand, attempts to
1
transform the acid 8 [34,35] into the amide derivatives 10–14
via the acyl chloride intermediate (using SOCl [35]) led to
2
complete decomposition of the starting material. One-pot
procedures [36] using different activating agents generated
the desired amides in very low yields or produced mixtures
that proved difficult to separate (see Table 2 for derivative
2.2. Synthesis
11). Furthermore, amides 10–14 were prepared by following
The designed compounds were synthesized along with
some derivatives in which variations in the lipophilicity
and/or in redox properties were done. Derivatives 3–22 were
a two-step procedure from the corresponding acid 8, which
was transformed into the intermediate 9 (Scheme 2). Treat-
ment of intermediate 9 with different amines at room tem-

424
G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
Fig. 1. (A) Dock of semicarbazone derivative 1, displayed in ball and stick format, into the trypanothione site of TR. Nearest residues are displayed in space
filling format. (B) Main interaction residues within the trypanothione site of TcTR used in the design. Distances between significant points are in Å. (C) Dock
of carbazate 5, displayed in ball and stick format, into the trypanothione site of TR. (D) Dock of amide 11, displayed in ball and stick format, into the
trypanothione site of TR.
perature permit to obtain the desired products in moderate
yields. The esters 15–17 were obtained in good yield by
stirring 8 with the corresponding alcohol as solvent in an acid
medium (Scheme 2). Semicarbazones 18–20 and amino-
esters 21 and 22 were obtained by standard procedures
3. Pharmacology
3
.1. Anti-trypanosomal studies
The compounds were tested in vitro against epimastigote
forms of T. cruzi (Tulahuen strain) at 5, 10 and 25 µM, as
indicated in Section 6 [38,39]. Table 3 shows the percentage
of inhibition for the evaluated derivatives at 5 µM. We used
the no infective epimastigote form, as has been done in
several other studies with other nitrofuran and nitroimidazole
(
Scheme 2) [24,25,37]. The structures of all derivatives were
1
13
confirmed by NMR ( H, C, and HETCOR experiments), IR
and MS, and their purities were established by TLC and
microanalysis.

G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
425
Table 1
Table 3
Docking energies in the trypanothione site of TcTR (E), maximum length
Anti-trypanosomal in vitro activity (% Inh.), binding energy in the active site
of TcTR including desolvation terms (BE) and octanol/water partition coef-
ficient (log P) of developed compounds
.
–
(L) and Gibbs free energy of single electron reduction [DG(L )] for de-
signed compounds
R
R
O2N
X
O2N
X
b
.–
c
d
X R
E
L (Å)
DG(L )
X R
%
Inh.
BE
LogP
–
1
a
–1
c
a,b
–1
(
kcal mol
)
(kcal mol
–46.81
)
(kcal mol ) [44]
Compounds
Compounds
(
5 µM)
1
2
4
5
1
1
1
1
2
2
O =N–NH–CO–
NH(CH ) CH
3
–34.06
–31.93
–29.11
–30.74
15.6
18.2
16.0
18.6
1
2
O =N–NH–CO–NH(CH ) CH
3
30.0
20.0
–41.89
–35.32
1.94
0.26
2
3
2
3
O =N–NH–CO–
NH(CH ) OCH
3
O =N–NH–CO–
NH(CH ) OCH
3
–47.18
–47.95
–49.15
2
2
2
2
3
4
5
6
7
8
9
O =N–NH–CO–OCH3
O =N–NH–CO–O(CH ) CH
32.0
84.2
96.2
92.7
83.7
20.0
6.0
–27.28
–40.02
–47.79
–52.94
–57.06
–13.27
–25.12
–0.37
1.22
2.27
2.80
3.33
O =N–NH–CO–
O(CH ) CH
2
3
3
3
3
3
2
3
3
O =N–NH–CO–O(CH ) CH
2
5
O =N–NH–CO–
O(CH ) CH
O =N–NH–CO–O(CH ) CH
2
6
2
5
3
O =N–NH–CO–O(CH ) CH
2
7
0 O =CH–CO–NH(CH ) CH –32.55
1 O =CH–CO–NH(CH ) CH –34.60
2 5 3
6 O =CH–CO–O(CH ) CH
7 O =CH–CO–O(CH ) CH
1 O =C(NH )-CO-OCH CH –23.26
2 S =C(NH )–CO–OCH CH –25.65
15.2
17.7
15.0
17.5
11.9
12.5
–51.62
–51.64
–54.53
–54.80
–48.74
–54.09
2
3
3
O =CH–CO–OH
O =CH–CO–O–(1-
succinimidyl)
1.16 (0.68)
0.00
–25.68
–23.34
2
3
3
2
5
3
1
0 O =CH–CO–NH(CH ) CH
1 O =CH–CO–NH(CH ) CH
2 O =CH–CO–NH(CH ) OCH
3
3 O =CH–CO–NHCH (2-furyl)
75.0
82.0
31.0
14.0
35.0
–34.00
–47.14
–29.45
–30.29
–27.73
2.31 (2.36)
3.37
2
3
3
2
2
3
1
1
1
1
2 5
3
2
2
3
0.74
2
2
a
Energy of the ligand–cavity binding from Tripos force field within the
1.67
2
SYBYL-DOCK program.
4 O =CH–CO–NHCH –CO–
1.19
2
b
Maximum length of the ligand in its extended conformation.
Gibbs free energy of single-electron reduction of ligand from AM1
OCH CH
2
3
c
1
1
1
1
5 O =CH–CO–OCH3
6 O =CH–CO–O(CH ) CH
7 O =CH–CO–O(CH ) CH
8 O =N–NH–CO–NH(a-naphthyl) 32.0
19 O =N–NH–CO–NH(b-naphthyl) 59.0
20 O =N–NH–CO–NH
1 O =C(NH )–CO–OCH CH
2 S =C(NH )–CO–OCH CH
39.0
42.0
17.0
–22.22
–33.90
–42.69
–48.99
–49.01
–21.82
–20.99
–30.37
1.53 (1.65)
3.12 (2.83)
4.18
method at 298 K in the gas phase.
2
3
3
2
5
3
derivatives as well as other natural and synthetic potential
trypanocidal agents, which have also proved to be efficient
against the trypomastigote and amastigote forms [40]. The
existence of the epimastigote form as an obligate mammalian
intracellular stage has been revisited [41,42] and confirmed
recently [43]. ED , 50% effective doses to inhibit the para-
3.43
3.43
67.0
0.20 (0.23)
1.30
2
e
2
2
n.a.
2
2
3
9.0
1.86
2
2
3
a
Values are means of two experiments.
%
at day 5.
5
0
b
Inh. = percentage of inhibition of T. cruzi growth at a dosage of 5 µM
site growth, values were determined for derivatives 1, 5, and
1 (Table 4 and Fig. 2).
1
c
Binding energy from the SYBYL-LEAPFROG program including the
desolvation of ligand and cavity.
d
Calculated logP (measured value in parenthesis) from ClogP 4.0.
n.a. = not active at 5 µM.
3.2. Theoretical determinations
e
The partition coefficients (log P) for all derivatives were
calculated using C log P 4.0 (Table 3) [44]. The Gibbs free
energies of the single-electron reductions of the designed
Table 2
One pot conditions assayed and yield obtained in the synthesis of derivative 11
Conditions
OH
NH(CH ) CH
3
2
5
O N
2
O
O N
2
O
O
O
8
11
Conditions
Yield (%)
a
b
(
(
(
(
1) DCC /CH Cl /rt; (2) hexylamine/CH Cl /rt
14
2
2
2
2
b
1) DCC/1-hydroxybenzotriazole/CH Cl /rt; (2) hexylamine/CH Cl /rt
20
2
2
2
2
c
1) CDI /CH Cl /rt; (2) hexylamine/CH Cl /rt
10
4
2
2
2
2
1) ClCO Et/Et N/THF/–8 °C; (2) hexylamine/THF/0 °C to rt
2
3
a
b
c
DCC: dicyclohexylcarbodiimide.
Compound 11 is contaminated with dicyclohexylurea.
CDI: carbonyldiimidazole.

426
G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
Table 4
ED50 values and calculated parameters for the most active alkyl derivatives of each tested family of 5-nitrofuryl derivatives
–
1
a
.–
–1
b
c
Compounds
1
5
Family
ED50 (µM)
BE (kcal mol
–41.89
)
DG(L )(kcal mol
)
LogP
1.94
2.27
3.37
3.12
1.30
d
Semicarbazone
Carbazate
Amide
7.4 ± 0.5
–46.8
–49.2
–51.6
–54.5
–48.7
d
3.2 ± 0.5
–47.79
d
1
1
2.0 ± 0.5
–47.14
e
1
2
6
1
Ester
> 5
–33.90
e
Amino-ester
> 10
–20.99
a
TcTR cavity–ligand binding energies from the LEAPFROG module of SYBYL.
Gibbs free energy of single-electron reduction of ligand from AM1 method at 298 K in the gas phase.
Calculated from C log P version 4.0.
Values are means of two experiments.
Values estimated.
b
c
d
e
–
ligands [DG(L )] were estimated using the AM1 method in
the gas phase at 298 K. This kind of calculation offers a good
correlation with corresponding experimental single-electron
reduction potentials of nitroaromatics [45]. Esters 16 and 17,
amides 10 and 11, carbazates 4 and 5 and amino-ester 21, in
decreasing order, present more favorable single-electron re-
ductions than the corresponding 5-nitrofurfurylidene-
semicarbazides 1 and 2, as shown by the data in Table 1.
Additional ligand–cavity binding energies, including
ligand and cavity desolvation terms (BE in Table 3), were
estimated by means of the LEAPFROG module of SYBYL
trypanosomal activity than Nfx at 5 µM (Nfx inhibition:
30.0%), i.e. compounds 4–7, 10, 11, 16, 19 and 20 (Table 3).
ED50 values were determined for the most active carbazate
and amide derivatives. Carbazate 5 and amide 11 proved to be
one of the most active novel compounds against epimastigote
form of T. cruzi strain Tulahuen, with ED50 values (3.2 and
2.0 µM, respectively) better than that observed for Nfx
(7.7 µM). These compounds could be more effective in the
infective and intracellular replicative forms since the mo-
lecular weight thiol content in the trypomastigote and
amastigote forms of T. cruzi is lower than in epimastigote
form [23,50].
[
46] from DOCK solutions. More negative BEs implicate
more favorable compounds’ binding energy in the trypan-
othione site of TcTR. In a previous test, LEAPFROG gave a
good reproduction of the SM2-AM1 [47] desolvation ener-
gies of small molecules (Section 6). Moreover, ligand–cavity
binding energies from LEAPFROG are a good approxima-
The new 5-nitrofuryl derivatives reported here comply
with some of the reported conditions to be good trypanocid-
als, they are potentially substrates for TcTR and show even
more favorable single-electron reduction than derivative 1
.–
(
cathodic peak potential E = –0.86 V [25], DG(L ) = –
pc
tion to reproduce K ligand–enzyme binding, as it has been
–1
.–
i
46.81 kcal mol ) and Nfx (E = –0.91 V [24], DG(L ) = –
pc
demonstrated for some NADPH-dependent enzymes
–1
4
6.08 kcal mol ), as can be deduced from the corresponding
DG(L ) (Table 1). However, only the Gibbs free energy of
the single-electron reduction (DG(L ) in Table 1) of the
-nitrofuryl derivatives does not explain their differences as
[48,49].
.–
.
–
5
4. Results and discussion
trypanocidals, at 5 µM. This could be in agreement with a
redox cycling behavior of the compounds, as has been re-
cently reported for the nitrofurazone 20 [23]. The experimen-
tal trypanocidal data shown in Table 3 confirm that the most
active trypanosome growth inhibitors, at 5 µM, have the
highest binding energies (BE) in the trypanothione site of
TcTR (see also Fig. 3). It is important to note that energies
from LEAPFROG (including the cavity and ligand desolva-
tion, BE in Table 3) showed some relation with the inhibitory
activity, at 5 µM, than energies from DOCK, where desolva-
tion was not taken into account (E in Table 1). It should be
noted that the metabolism of the compounds cannot be dis-
carded as an explanation of some inhibition differences be-
tween families with similar BE values (Fig. 3).
On the other hand, the passive transport for all compounds
does not seem to be related to biological response (see log P
and % Inh. in Table 3). However, within each assayed family
of derivatives appears a clear relation between log P and
activity at 5 µM. In the family of alkyl carbazates 3–7, it can
be observed that there is an optimum partition coefficient
value (at about 2.3) to obtain the highest inhibition. The most
New 5-nitrofuryl derivatives were developed as potential
anti-trypanosomal agents. The majority of these compounds
showed complete inhibition of the parasite growth at a dos-
age of 10 and 25 µM (data not shown). Some of the newly
developed 5-nitrofuryl derivatives displayed better anti-
100
80
60
40
20
0
Nfx
Compd1
Compd5
Compd11
0
5
10
15
20
Dose (µM)
Fig. 2. Curves dose–response of Nfx and derivatives 1, 5, and 11.

G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
427
1
00
obtained from vacuum-dried samples (over phosphorous
pentoxide at 3–4 mm Hg, 24 h at room temperature) and
performed on a Fisons EA 1108 CHNS-O analyzer. Infrared
spectra were recorder on a Perkin–Elmer 1310 apparatus,
using potassium bromide tablets; the frequencies are ex-
pressed in cm . H-NMR and C-NMR spectra were re-
corded on a Bruker DPX-400 (at 400 and 100 MHz) instru-
ment, with tetramethylsilane as the internal reference and in
the indicated solvent; the chemical shifts are reported in ppm.
J values are given in Hertz. Mass spectra were recorded on a
Shimadzu GC-MS QP 1100 EX instrument at 70 eV.
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
–
1
1
13
-60
-50
-40
-30
-20
-10
6
3
.1.1. General procedure for the synthesis of carbazates
–7
BE
Fig. 3. Percentage of trypanosome growth inhibition at a dosage of 5 µM (%
Inh.) vs. TcTR cavity–ligand binding energies from LEAPFROG calcula-
tions (BE). Semicarbazones are displayed as C, amides as ., carbazates as
A mixture of 5-nitro-2-furaldehyde (1 eq.), the corre-
sponding carbazate (1 eq.), p-TsOH (catalytic amounts) and
toluene as solvent was stirred at room temperature until the
_
and the rest of compounds as ".
carbonyl compound was not present (SiO , 1% MeOH in
2
CH Cl ). The resulting precipitate was collected by filtration
and was crystallized with the indicated solvent.
active examples of the assayed amides and esters also have
their log P near to the aforementioned value (see activity of
compound 11 and activity of compound 16).
2
2
6
.1.1.1. Methyl 4-(5-nitrofurfurylidene)carbazate (3).
Yellow-orange solid (60%); mp 205.0–207.0 °C (from petro-
leum ether: EtOAc) (Found: C, 39.4; H, 3.3; N, 19.7.
C H N O requires C, 39.4; H, 3.3; N, 19.7); d (400 MHz,
5. Conclusions
7
7
3
5
H
Thirteen of the 17 new derivatives developed in this work
DMSO-d
6
) 3.73 (3H, s, O–CH
3
), 7.14 (1H, d, J 3.9, furan-H),
showed better trypanocidal effect against epimastigote forms
of T. cruzi than nifurtimox at 5 µM. These compounds
showed the highest in vitro trypanocidal effect reported to
date for members of the nitrofuran family of compounds. The
method employed by us, which detects different micro-
orientations for the ligands on the TcTR site, renders a design
of trypanocidals that is as good as others based on the experi-
mental inhibition of TR. Compounds that bind in the trypan-
othione site of TcTR and also bring favorable energy of the
single-electron reduction and adequate log P showed good
trypanocidal activity.
7.74 (1H, d, J 3.9, furan-H), 7.97 (1H, s, –CH=N) and 11.54
(1H, br s, NH); d (100 MHz, DMSO-d ) (HMQC, HMBC)
54.50 (–CH ), 115.50 (furan-C), 116.50 (furan-C), 134.20
(H–C=N), 152.17 (furan-C), 153.48 (furan-C) and 154.78
(–C=O). m/z 213 (M , 100.0%), 197 (2.5), 182 (1.8) and 167
(12.6).
C
6
3
.
+
6
.1.1.2. Butyl 4-(5-nitrofurfurylidene)carbazate (4).
Yellow-orange needles (38%); mp 147.6–148.5 °C (from
EtOH:H O) (Found: C, 47.05; H, 5.1; N, 16.3. C H N O
10 13 3 5
2
requires C, 47.1; H, 5.1; N, 16.46); d (400 MHz, CDCl )
H
3
0
.96 (3H, t, J 7.4, –CH ), 1.42 (2H, sextet, J 7.4, –CH ), 1.69
Studies related with activity and toxicity in vivo with the
more active derivatives are currently in progress.
3
2
(
(
2H, quintet, J 6.8, –CH ), 4.26 (2H, t, J 6.7, O–CH ), 7.00
1H, d, J 3.8, furan-H), 7.36 (1H, d, J 3.8, furan-H), 8.05 (1H,
2 2
br s, –CH=N) and 8.62 (1H, br s, NH); d (100 MHz, CDCl )
C
3
6. Experimental protocols
(HMQC, HMBC) 14.04 (–CH ), 19.35 (–CH ), 31.20
3 2
(
1
1
(
–CH ), 66.83 (–CH ), 112.30 (furan-C), 113.69 (furan-C),
2 2
6
.1. Chemistry
32.00 (H–C=N), 152.25 (furan-C), 152.50 (furan-C) and
.
+
53.00 (–C=O); m/z 255 (M , 22.9%), 239 (0.5) and 182
All starting materials were commercially available
24.9).
research-grade chemicals and used without further purifica-
tion. The compounds 8, 20–22, 4-(1-naphthyl)
6.1.1.3. Hexyl 4-(5-nitrofurfurylidene) carbazate (5).
Yellow-orange needles (50%); mp 139.9–140.8 °C (form
EtOH:H O) (Found: C, 50.85; H, 6.2; N, 14.8. C H N O
semicarbazide,
4-(2-naphthyl)-semicarbazide,
butyl-,
hexyl-, heptyl-, and octylcarbazate were prepared according
to literature procedures [33–35,37,51]. All solvents were
dried and distilled prior to use. All the reactions were carried
out in a nitrogen atmosphere. The typical work-up included
washing with brine and drying the organic layer with sodium
sulphate before concentration. Melting points were deter-
mined using a Leitz Microscope Heating Stage Model
2
12 17 3 5
requires C, 50.9; H, 6.05; N, 14.8); d (400 MHz, CDCl )
H
3
0.92 (3H, t, J 6.8, –CH ), 1.39 (6H, m, –CH ), 1.71 (2H,
3
2
quintet, J 7.0, –CH ), 4.27 (2H, t, J 6.7, O–CH ), 7.01 (1H, d,
2
2
J 3.8, furan-H), 7.38 (1H, d, J 3.8, furan-H), 8.08 (1H, br s,
–CH=N) and 8.46 (1H, br s, NH); d (100 MHz, CDCl )
C
3
(HMQC, HMBC) 14.34 (–CH ), 22.90 (–CH ), 25.78
3
2
350 apparatus and are uncorrected. Elemental analyses were
(–CH ), 29.14 (–CH ), 31.77 (–CH ), 67.11 (–CH ), 112.28
2 2 2 2

428
G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
(
(
(
furan-C), 113.65 (furan-C), 132.77 (H–C=N), 152.22
br s, NH), 6.67 (1H, d, J 15.3, =CH–), 6.69 (1H, d, J 3.7,
furan-H), 7.34 (1H, d, J 3.7, furan-H) and 7.41 (1H, d, J
15.3, =CH–); d (100 MHz, CDCl ) (HMQC, HMBC) 14.07
furan-C), 153.00 (furan-C) and 153.50 (–C=O); m/z 283
.
+
M , 53.7%), 267 (0.8) and 182 (14.3).
C
3
(
(
(
–CH ), 20.45 (–CH ), 31.94 (–CH ), 40.14 (–CH ), 113.59
3 2 2 2
6
.1.1.4. Heptyl 4-(5-nitrofurfurylidene)carbazate (6). Yel-
low needles (45%); mp 121.9–123.0 °C (from petroleum
ether: EtOAc) (Found: C, 52.6; H, 6.3; N, 14.3. C H N O
furan-C), 115.11 (furan-C), 125.74 (–CH=), 125.93
–CH=), 152.44 (furan-C), 153.58 (furan-C) and 164.53
.
+
13
19
3
5
(–C=O); m/z 238 (M , 8.2%), 221 (4.8) and 166 (100.0).
requires C, 52.5; H, 6.4; N, 14.1); mmax 3221.5, 1743.9,
1
–
4
230.7 and 819.8; d (400 MHz, CDCl ) 0.89 (3H, t, J 7.1,
6.1.3.2. N-Hexyl 3-(5-nitrofuryl)propenamide (11). Chro-
matographic column (SiO2, petroleum ether:EtOAc
(0–20%)); brown-orange solid (48%); mp 102.5–104.0 °C
H
3
CH ), 1.30 (8H, m, –CH ), 1.69 (2H, quintet, J 6.9, –CH ),
3
2
2
.24 (2H, t, J 6.7, O–CH ), 7.00 (1H, d, J 3.8, furan-H), 7.36
2
(
1
1H, d, J 3.8, furan-H), 8.05 (1H, br s, –CH=N) and 8.50 (bs,
H, NH); m/z 297 (M , 8.8%), 267 (0.1) and 182 (3.6).
(Found: C, 53.75; H, 6.8; N, 10.4. C13H N O requires C,
18 2 4
.
+
53.8; H, 6.8; N, 10.5); mmax 3275.5, 1655.1 and 817.9;
d (400 MHz, CDCl ) 0.92 (3H, t, J 6.7, –CH ), 1.34 (4H, m,
H
3
3
6.1.1.5. Octyl 4-(5-nitrofurfurylidene)carbazate (7). Yellow
–CH ), 1.58 (4H, m, –CH ), 3.41 (2H, q, J 7.1, N–CH ), 5.79
2 2 2
needles (61%); mp 118.3–119.6 °C (from petroleum ether:
EtOAc) (Found: C, 53.9; H, 6.7; N, 13.3. C H N O re-
quires C, 54.0; H, 6.75; N, 13.5); mmax 3219.6, 1747.7,
(1H, br s, NH), 6.66 (1H, d, J 15.3, =CH–), 6.70 (1H, d, J 3.8,
furan-H), 7.34 (1H, d, J 3.7, furan-H) and 7.42 (1H, d J
15.3, =CH–). m/z 266 (M , 4.7%), 249 (7.8) and 166 (100.0).
14 21 3 5
.
+
1
7
–
226.8 and 810.0; d (400 MHz, acetone-d ) 0.88 (3H, t, J
H 6
.1, –CH ), 1.31 (10H, m, –CH ), 1.67 (2H, quintet, J 6.6,
6.1.3.3. N-(2-Methoxyethyl) 3-(5-nitrofuryl)propenamide
3
2
(
12). Chromatographic column (SiO , petroleum ether:E-
CH ), 4.18 (2H, t, J 6.6, O–CH ), 7.07 (1H, d, J 3.9,
2
2
2
tOAc (0–10%)); yellow-orange solid (40%); mp 125.0–
26.7 °C (Found: C, 49.9; H, 5.1; N, 11.6. C H N O
furan-H), 7.58 (1H, d, J 3.9, furan-H), 8.16 (1H, br s,
–
2
.
+
1
CH=N) and 10.43 (1H, br s, NH); m/z 311 (M , 10.7%),
10 12 2 5
requires C, 50.0; H, 5.0; N, 11.7); d (400 MHz, CDCl ) 3.42
82 (0.2) and 182 (4.8).
H
3
(
3H, s, O–CH ), 3.54 (2H, t, J 5.2, –CH ), 3.61 (2H, q, J 5.9,
3 2
6.1.2. N-[3-(5-Nitrofuryl)-2-propenyl]oxysuccinimide (9)
N–CH ), 6.16 (1H, br s, NH), 6.68 (1H, d, J 15.1, =CH–),
2
A mixture of acid 8 (0.50 g, 2.7 mmol) and
6.70 (1H, d, J 3.4, furan-H), 7.34 (1H, d, J 3.8, furan-H) and
7.43 (1H, d, J 15.3, =CH–); d (100 MHz, CDCl ) (HMQC,
N-hydroxysuccinimide (0.31 g, 2.7 mmol) in dry CH Cl
2
2
C
3
3
(
10.0 cm ) was stirred at 0 °C (ice/water), then DCC (0.56 g,
HMBC) 40.00 (–CH ), 59.19 (–CH ), 71.25 (–CH ), 113.50
3 2 2
2
.7 mmol) was added. The reaction mixture was allowed to
(furan-C), 115.13 (furan-C), 125.54 (–CH=), 126.15
(–CH=), 152.40 (furan-C), 153.49 (furan-C) and 164.53
(–C=O); m/z 240 (M , 7.2%), 208 (10.0) and 166 (100.0).
stir for a further 1 h at 0 °C and 24 h at room temperature. The
solid (dicyclohexylurea) was collected by filtration, washed
with CH Cl and the filtrate concentrated under reduced
.
+
2
2
6
.1.3.4. N-Furfuryl 3-(5-nitrofuryl)propenamide (13). Chro-
matographic column (SiO2, petroleum ether:EtOAc
0–40%)); oil (43 %) (Found: C, 54.75; H, 3.5; N, 10.3.
C H N O requires C, 55.0; H, 3.8; N, 10.7); d (400 MHz,
pressure. The resulting solid was washed with hot ethanol-
petroleum ether (1:1). Beige solid (0.34 g, 45%); mp
10.0 °C (d) (Found: C, 47.0; H, 2.5; N, 10.3. C H N O
:
2
(
11 8 2 7
requires C, 47.1; H, 2.9; N, 10.0); d (400 MHz, CDCl ) 2.88
12 10
2
5
H
H
3
CDCl ) 4.59 (2H, d, J 5.6, N–CH ), 6.29 (1H, d, J 3.1,
(4H, s, –CH ), 6.80 (1H, d, J 15.9, =CH–), 6.89 (1H, d, J 3.8,
3
2
2
furan-H), 6.34 (2H, m, NH + furan-H), 6.69 (1H, d, J
5.8, =CH–), 6.71 (1H, d, J 3.1, furan-H), 7.33 (1H, d, J 3.8,
furan-H), 7.38 (1H, dd, J 3.1, J 1.0, furan-H) and 7.43 (1H,
furan-H), 7.36 (1H, d, J 3.8, furan-H) and 7.62 (1H, d, J
.
+
1
15.9, =CH–); m/z 280 (M , 1.4%), 264 (0.1) and 166 (100.0).
1
2
6.1.3. General procedure for the synthesis of amides
d, J 15.3, =CH–); d (100 MHz, CDCl ) (HMQC, HMBC)
C 3
10–14
37.30 (–CH ), 108.22 (furan-C), 110.94 (furan-C), 113.53
2
A mixture of intermediate 9 (1 eq.), the corresponding
(furan-C), 115.40 (furan-C), 125.06 (–CH=), 126.60
(–CH=), 142.81 (furan-C), 151.01 (furan-C), 152.50 (furan-
C), 153.34 (furan-C) and 164.42 (–C=O); m/z 262 (M ,
amine (1.2 eq.) and dry CH Cl as solvent was stirred at room
2
2
.+
temperature until the compound 9 was not present (SiO2,
0% EtOAc in petroleum ether). The mixture was concen-
4
0.9%), 245 (2.1) and 166 (8.2).
trated in vacuo and treated with EtOAc. After the work-up
process the residue was purified as indicate.
6
.1.3.5.
N-(Ethyloxycarbonylmethyl)
3-(5-nitrofuryl)
propenamide (14). Chromatographic column (SiO , petro-
2
6
.1.3.1. N-Butyl 3-(5-nitrofuryl)propenamide (10). Chro-
matographic column (SiO2, petroleum ether:EtOAc
0–20%)); yellow oil (51%); (Found: C, 55.2; H, 5.6; N,
leum ether:EtOAc (0–40%)); beige solid (21%); mp 195.5–
196.7 °C (Found: C, 49.05; H, 4.15; N, 10.5. C H N O
12 2 6
11
(
1
requires C, 49.25; H, 4.5; N, 10.45); mmax 3358.5, 1720.7,
1672.5 and 831.4; d (400 MHz, CDCl ) 1.33 (3H, t, J 7.1,
1.55. C H N O requires C, 55.5; H, 5.9; N, 11.8); m
11 14 2 4 max
H
3
3
271.7, 1662.8 and 814.1; d (400 MHz, CDCl ) 0.96 (3H, t,
–CH ), 4.19 (2H, d, J 5.1, N–CH ), 4.25 (2H, q, J 7.1,
H
3
3
2
J 7.3, –CH ), 1.41 (2H, sextet, J 7.4, –CH ), 1.58 (2H,
O-CH ), 6.35 (1H, br s, NH), 6.72 (1H, d, J 3.6, furan-H),
3
2
2
quintet, J 7.1, –CH ), 3.42 (2H, q, J 7.0, N–CH ), 5.96 (1H,
6.74 (1H, d, J 15.4, =CH–), 7.34 (1H, d, J 3.7, furan-H) and
2
2

G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
429
7
.44 (1H, d, J 15.4, =CH–); d (100 MHz, CDCl ) (HMQC,
the carbonyl compound was not present (SiO , 1% MeOH in
C
3
2
HMBC) 14.52 (–CH ), 42.17 (–CH ), 62.19 (–CH ), 113.42
CH Cl ). The resulting precipitate was collected by filtration
3
2
2
2
2
(
(
furan-C), 115.47 (furan-C), 124.55 (–CH=), 126.85
–CH=), 151.50 (furan-C), 153.19 (furan–C), 164.49 (–C=O)
and was crystallized with the indicated solvent.
.
+
6
.1.5.1. 4-(1-Naphthyl)-1-(5-nitrofurfurylidene)semicarba-
zide (18). Bright-orange needles (52%); mp 236.5–238.0 °C
from acetone) (Found: C, 59.2; H, 3.7; N, 17.0. C H N O
and 169.92 (–C=O); m/z 268 (M , 2.6%), 222 (71.2) and 166
100.0).
(
(
1
6 12 4 4
6.1.4. General procedure for the synthesis of esters 15–17
requires C, 59.3; H, 3.7; N, 17.3); d (400 MHz, acetone-d )
H 6
A mixture of intermediate 8, p-toluenesulphonic acid
7.29 (1H, d, J 3.9, furan-H), 7.52 (1H, t, J 7.9, naphthyl-H),
7.58 (1H, dt, J 8.0, J 1.0, naphthyl-H), 7.64 (1H, dt, J 8.3,
(
catalytic amounts) and the corresponding alcohol as solvent
was heated at reflux until the compound 8 was not present
SiO , 40% EtOAc in petroleum ether). The mixture was
1
2
1
J2 1.4, naphthyl-H), 7.65 (1H, d, J 3.9, furan-H), 7.74 (1H, d,
J 8.2, naphthyl-H), 7.96 (1H, d, J 8.0, naphthyl-H), 8.05 (1H,
d, J 7.8, naphthyl-H), 8.11 (1H, d, J 8.3, naphthyl-H), 8.12
(1H, s, –CH=N), 9.14 (1H, s, NH) and 10.43 (1H, s, NH);
d (100 MHz, acetone-d ) (HMQC, HMBC) 113.09 (furan-
(
2
concentrated in vacuo and treated with EtOAc. After the
work-up process the residue was purified as indicated.
C
6
6
.1.4.1. Ethyl 3-(5-nitrofuryl)propenoate (15). Chromato-
C), 114.12 (furan-C), 119.12, 119.33, 121.43, 124.78,
126.34, 126.52, 127.58, 128.87, 133.80, 134.67 (naphthyl-
C), 128.91 (H-C=N), 151.00 (furan-C), 152.82 (furan-C) and
graphic column (SiO , petroleum ether:EtOAc (0–10%));
beige solid (67%); mp 128.8–130.4 °C (Found: C, 51.1; H,
2
.
+
4.2; N, 6.6. C H NO requires C, 51.2; H, 4.3; N, 6.6);
152.93 (–C=O); m/z 324 (M , 59.2%), 169 (85.8), 155
9
9
5
d (400 MHz, CDCl ) 1.36 (3H, t, J 7.1, –CH ), 4.30 (2H, t, J
(100.0) and 143 (70.5).
H
3
3
7
.1, O–CH ), 6.66 (1H, d, J 15.9, =CH–), 6.77 (1H, d, J 3.7,
2
6
.1.5.2.
bazide (19). Bright-orange needles (56%); mp 129.0–
30.5 °C (from acetone:MeOH) (Found: C, 59.25; H, 3.7; N,
17.2. C H N O requires C, 59.3; H, 3.7; N, 17.3); d
H
4-(2-Naphthyl)-1-(5-nitrofurfurylidene)semicar-
furan-H), 7.34 (1H, d, J 3.8, furan-H) and 7.43 (1H, d, J
1
.
+
5.9, =CH–); m/z 211 (M , 50.9%), 195 (1.6) and 166 (89.7).
1
6
.1.4.2. Butyl 3-(5-nitrofuryl)propenoate (16). Chromato-
1
6
12
4
4
graphic column (SiO , petroleum ether:EtOAc (0–30%)); oil
(400 MHz, DMSO-d ) 7.40 (1H, dt, J 7.9, J 1.0, naphthyl-
2
6 1 2
(
32%) (Found: C, 55.0; H, 5.2; N, 5.4. C H NO requires
H), 7.43 (1H, d, J 4.0, furan-H), 7.48 (1H, dt, J 8.1, J 1.2,
1 2
11
13
5
C, 55.2; H, 5.4; N, 5.9); d (400 MHz, CDCl ) 0.98 (3H, t, J
naphthyl-H), 7.75 (1H, dd, J 8.8, J 2.0, naphthyl-H), 7.81–
H
3
1 2
7
.4, –CH ), 1.44 (2H, sextet, J 7.3, –CH ), 1.71 (2H, quintet,
7.85 (3H, m, furan-H + naphthyl-H + naphthyl-H), 7.87 (1H,
d, J 9.0, naphthyl-H), 7.96 (1H, s, –CH=N), 8.20 (1H, d, J
1.7, naphthyl-H), 9.14 (1H, s, NH) and 11.30 (1H, s, NH); dC
3
2
J 7.1, –CH ), 4.24 (2H, t, J 6.6, O–CH ), 6.66 (1H, d, J
1
2
2
5.9, =CH–), 6.77 (1H, d, J 3.8, furan-H), 7.35 (1H, d, J 3.8,
furan-H) and 7.42 (1H, d, J 15.9, =CH–); d (100 MHz,
(100 MHz, DMSO-d ) (HMQC, HMBC) 113.74 (furan-C),
C
6
CDCl ) (HMQC, HMBC) 14.05 (–CH ), 19.52 (–CH ),
116.02 (furan-C), 116.45, 121.77, 125.29, 127.20, 127.98,
128.31, 128.98, 130.49, 134.29, 137.20 (naphthyl-C), 129.88
(H-C=N), 152.00 (furan-C), 153.34 (furan-C) and 153.61
3
3
2
3
1.03 (–CH ), 65.45 (–CH ), 113.23 (furan-C), 115.52
2 2
(
furan-C), 123.31 (–CH=), 129.02 (–CH=), 151.00 (furan-
.
+
.+
C), 152.86 (furan-C) and 166.07 (–C=O); m/z 239 (M ,
(–C=O); m/z 324 (M , 35.0%), 169 (86.0), 155 (38.1) and
25.0%), 223 (0.8) and 166 (100.0).
143 (100.0).
6.1.4.3. Hexyl 3-(5-nitrofuryl)propenoate (17). Chromato-
6.2. Biology
graphic column (SiO , petroleum ether); oil (45%) (Found:
2
6
.2.1. Anti-trypanosomal bioassays
C, 58.3; H, 6.2; N, 5.0. C H NO requires C, 58.4; H, 6.4;
13
17
5
T. cruzi epimastigotes (Tulahuen 2 strain) were grown at
8 °C in an axenic medium (BHI-Tryptose) complemented
N, 5.2); d (400 MHz, CDCl ) 0.92 (3H, t, J 6.7, –CH ),
H
3
3
2
1
4
.31–1.43 (6H, m, –CH ), 1.69 (2H, quintet, J 6.9, –CH ),
2
2
with 10% fetal calf serum. Cells from a 10-day old culture
.23 (2H, t, J 6.7, O–CH ), 6.66 (1H, d, J 15.9, =CH–), 6.77
2
3
(
stationary phase) were inoculated into 50 cm of fresh cul-
(
(
1H, d, J 3.7, furan-H), 7.35 (1H, d, J 3.7, furan-H) and 7.42
1H, d, J 15.9, =CH–); d (100 MHz, CDCl ) (HMQC,
6
ture medium to give an initial concentration of 1 × 10
C
3
3
cells/cm . Cell growth was followed every day by measuring
HMBC) 14.34 (–CH ), 22.90 (–CH ), 25.96 (–CH ), 28.96
3
2
2
the absorbance of the culture at 600 nm. Before inoculation,
the medium was supplemented with the indicated amount of
the drug from a stock solution in DMSO. The final concen-
tration of DMSO in the culture media never exceeded 0.4%
and the control was run in the presence of 0.4% DMSO and in
the absence of any drug. No effect on epimastigote growth
was observed with the presence of up to 1% DMSO in the
culture media. The percentage of inhibition was calculated as
follows: % = {1 – [(A – A )/(A – A )]} × 100, where
(
1
(
(
–CH ), 31.79 (–CH ), 65.75 (–CH ), 113.25 (furan-C),
15.54 (furan-C), 123.30 (–CH=), 129.02 (–CH=), 152.00
2 2 2
furan-C), 152.87 (furan-C) and 166.07 (–C=O); m/z 267
.
+
M , 13.1%), 251 (2.0) and 166 (100.0).
6
.1.5. General procedure for the synthesis
of semicarbazones 18 and 19
A mixture of 5-nitro-2-furaldehyde (1 eq.), the corre-
sponding semicarbazide (1 eq.), p-TsOH (catalytic amounts)
and toluene as solvent was stirred at room temperature until
p
0p
c
0c
Ap = A600 of the culture containing the drug at day 5;

430
G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
A0p = A600 of the culture containing the drug just after
obtained by subtracting the energies of the cavity and the
absolute minimum of the ligand from the energy of the
ligand–cavity complex from the DOCK program (the former
value was calculated by the corresponding conformational
analysis of the compound excluding conformations with in-
tramolecular hydrogen bonds as these are improbable in
aqueous media). The loss of translational and rotational en-
tropy for the ligand was not considered because semi-
empirical quantum-mechanical AM1 [55] calculations re-
flected the fact that entropies of the studied ligands were
essentially the same (42.4 ± 0.2 and 34.3 ± 0.8 cal mol K ,
respectively) and therefore, they should not contribute sig-
nificantly to the discrimination between ligands in this series.
The cavity–ligand binding energy including desolvation
energy (BE) was estimated by means of the LEAPFROG
module of SYBYL [46] in the same region as used in the
docking analysis, with default values used for the keywords.
In LEAPFROG the binding energy is calculated from three
major components: the direct steric, electrostatic, and im-
plicit hydrogen bonding enthalpies of ligand–cavity binding,
all of which were calculated using the Tripos force field [53].
The program also allows the use of optional cavity and ligand
desolvation energies, estimated from a very simple model. In
general, the method used to calculate binding energies in
LEAPFROG is similar to that in Goodford’s GRID program
addition of the inocula (day 0); A = A600 of the culture in the
c
absence of any drug (control) at day 5; A0c = A600 in the
absence of the drug at day 0.
6.2.2. ED50 determination
ED50 values were determined by following 50% effective
doses to inhibit the parasite growth in the absence (control)
and presence of increasing concentrations of the correspond-
ing drug. At day 5, the absorbance of the culture was mea-
sured and the percentage of inhibition was calculated. The
ED50 value was taken as the concentration of drug needed to
reduce the growth ratio to 50%.
–
1
–1
6.3. Molecular modeling
The docking analysis was carried out on the
trypanothione-binding site of T. cruzi trypanothione reduc-
tase (Structure 1BZL, EC: 1.6.4.8, from RCSB Protein Data
Bank, http://www.rcsb.org/pdb) where the residues were
bonded more closely to the natural substrate. The natural
substrate was extracted from the complex, and during the
calculations only the residues and water molecules in contact
with trypanothione (at a distance of 5 Å) were included in the
active site. Those residues were G14, S15, G16, L18, E19,
W22, N23, C53, V54, V59, K62, I107, S110, Y111, M114,
T335, P336, I339, F396′, T397′, P398′, L399′, M400′,
K402′, T457′, I458′, G459′, V460′, H461′, P462′, T463′,
[56]. However, in LEAPFROG, the ligand atom coordinates
are binned to increase speed and the binding energy of each
ligand atom is calculated as though the atom were actually
located in the center of a cube containing that atom. A simple
linear expression correlates the energy of interaction be-
tween the site and that particular ligand atom. Summing over
all ligand atoms yields the overall site–ligand interaction
energy. In a previous test carried out by us, LEAPFROG
1
2
3
S464′, E466′, E467′, C469′, S470′, R472′. The N , C and C
atoms of the indolyl group of the W22 residue in the binding
site were positioned at (27.86, 4.88, 1.80 Å), (27.64, 5.06,
3.15 Å) and (28.41, 4.19, 3.87 Å), respectively.
The analysis of nitrofurazone derivative 1 as a ligand onto
the active site of the enzyme was carried out by means of the
FLEXIDOCK [46,52] program from the BIOPOLYMER
module of SYBYL and using the default parameters. The
maximum number of generations allowed was 3000. The
active site was treated as a rigid molecule whereas the ligand
was treated as being flexible.
desolvation energies (E ) of formaldehyde, methanol, water,
nifuroxim, nitrofurazone 20 and its butyl derivative 1 showed
D
SM2-AM1
LEAPFROG
a good correlation (ED
= 2.790 + 0.783 ED
;
2
2
r = 0.953, S.E. = 1.46, p = 0.001, q = 0.871, S.E. = 2.42)
with the corresponding desolvation energies from a semi-
empirical quantum-mechanical method 1SCF SM2-
AM1//AM1 [47] (wavefunction//geometry) included in the
AMSOL package [57].
The ligand–cavity binding energy (E) was obtained from
the DOCK program by minimization of the ligand into the
rigid TR cavity using the Tripos force field [53], a gradient
The Gibbs free energies of the single-electron reduction
for the ligands were calculated in the gas phase at 298 K
–1
–1
termination of 0.05 kcal mol
Å and following the Powell
method. Compounds 2, 4, 5, 10, 11, 16, 17, 21, and 22 were
initially positioned within the active site of TcTR in a similar
disposition to the best orientation of nitrofurazone 1. These
arrangements were then optimized using the DOCK com-
mand. Charges for the ligand were obtained from Gasteiger–
Marsilii and Hückel methods whereas the site charges were
obtained using the Kollman force field [54] as implemented
within SYBYL. During minimization, a low dielectric model
using the AM1 [55] method (keywords
=
force
.–
thermo(298,298) rot = 1 doublet). DG(L ) values in
Tables 1 and 4 are the result of subtracting the Gibbs free
energy of the neutral ligand from that in corresponding anion
radical. Geometries were fully optimized in all calculations.
Molecular coordinates and charges (SYBYL mol2 files)
are available upon request (merino@unavarra.es).
(e = 1) with a distance-dependent dielectric function was
used because we assumed hydrogen bonds to be crucial for
the binding. The region where the minimization was carried
out has lower and higher corners at (4.00, –12.00, –6.00 Å)
and (32.00, 12.00, 12.00 Å), respectively. The term E was
Acknowledgements
Financial support from Fondo Clemente Estable and
C.S.I.C.-UDELAR (Uruguay) is acknowledged. H.C. thanks

G. Aguirre et al. / European Journal of Medicinal Chemistry 39 (2004) 421–431
431
the C.S.I.C.-UDELAR (Uruguay) and A. Duffaut thanks
PEDECIBA-QUÍMICA (Uruguay) for fellowships.
[26] C. Olea-Azar, A.M. Atria, F. Mendizabal, R. Di Maio, G. Seoane,
H. Cerecetto, Spectrosc. Lett. 31 (1998) 99–109.
[
27] V. Martínez-Merino, H. Cerecetto, Bioorg. Med. Chem. 9 (2001)
025–1030.
1
[
28] N. Cenas, D. Bironaite, E. Dickancaite, Z. Anusevicius, J. Sarlauskas,
J.S. Blanchard, Biochem. Biophys. Res. Commun. 204 (1994) 224–
References
2
29.
[
[
1] S.L. Croft, Mem. Inst. Oswaldo Cruz 94 (1999) 215–220.
2] 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)
[29] C.S. Bond, Y. Zhang, M. Berriman, M.L. Cunningham, A.H. Fair-
lamb, W.N. Hunter, Struct. Fold Des. 7 (1999) 81–89.
[30] S. Bailey, K. Smith, A.H. Fairlamb, W.N. Hunter, Eur. J. Biochem.
213 (1993) 67–75.
3
079–3080.
[
[
3] J.A. Urbina, Curr. Pharm. Des. 8 (2002) 287–295.
4] J.A. Castro, E.G. Díaz de Toranzo, Biomed. Environ. Sci. 1 (1988)
[31] S. Bonse, C. Santelli-Rouvier, J. Barbe, R.L. Krauth-Siegel, J. Med.
Chem. 42 (1999) 5448–5454.
1
9–33.
5] N.B. Gorla, O.S. Ledesma, G.P. Barbieri, I.B. Larripa, Mutat. Res.
24 (1989) 263–267.
[32] P.A. Karplus, G.E. Schulz, J. Mol. Biol. 210 (1989) 163–180.
[33] L.A. Carpino, B.A. Carpino, A.G. Chester, R.W. Murray, A.A. San-
tilli, P.H. Terry. Organic Synthesis Collective, Vol 5, 1973, pp. 168–
170, John Wiley & Sons.
[34] S. Rajagoplan, P. Raman. Organic Synthesis Collective, 1995, Vol 3,
pp. 425–427, John Wiley & Sons.
[
2
[
[
6] Available from http//www.who.int/tdr/.
7] H. Cerecetto, M. González, Curr. Topics Med. Chem. 2 (2002) 1185–
1
211.
[
[
8] J. Tovar, M.L. Cunningham, A.C. Smith, S.L. Croft, A.H. Fairlamb,
Proc. Natl. Acad. Sci. USA 95 (1998) 5311–5316.
9] L. Salmon-Chemin, E. Buisine, V. Yardley, S. Kohler, M.A. Debreu,
V. Landry, C. Sergheraert, S.L. Croft, R.L. Krauth-Siegel, E. Davioud-
Charvet, J. Med. Chem. 44 (2001) 548–565.
[35] M.O. Abdel-Rahman, M.N. Aboul-Enein, W.M. Tadros, J. Chem.
UAR 12 (1969) 69–75.
[36] M. Bodanzky, A. Bodanzky, The Practice of Peptide Synthesis, sec-
ond ed, Longman Group Limited, London, 1978.
[37] H. Cerecetto, R. Di Maio, M. González, G. Seoane, Heterocycles 45
(1997) 2023–2028.
[
10] S. Girault, T.E. Davioud-Charvet, L. Maes, J.F. Dubremetz,
M.A. Debreu, V. Landry, C. Sergheraert, Bioorg. Med. Chem. 9
[38] H. Cerecetto, R. Di Maio, M. González, M. Risso, P. Saenz,
G. Seoane,A. Denicola, G. Peluffo, C. Quijano, C. Olea-Azar, J. Med.
Chem. 42 (1999) 1941–1950.
[39] G. Aguirre, H. Cerecetto, R. Di Maio, M. González, G. Seoane,
A. Denicola, M.A. Ortega, I. Aldana, A. Monge, Arch. Pharm. 335
(2002) 15–21.
(
2001) 837–846.
11] A. Schmidt, R.L. Krauth-Siegel, Curr. Topics Med. Chem. 2 (2002)
239–1259.
[
[
[
1
12] R.H. Schirmer, J.G. Müller, R.L. Krauth-Siegel, Angew. Chem., Int.
Ed. Engl. 34 (1995) 141–154.
13] C. Chan, H. Yin, J. Garforth, J.H. McKie, R. Jaouhari, P. Speers,
K.T. Douglas, P.J. Rock,V.Yardley, S.L. Croft,A.H. Fairlamb, J. Med.
Chem. 41 (1998) 148–156.
[40] J. Rodríguez Coura, S.L. De Castro, Mem. Inst. Oswaldo Cruz 97
(2002) 3–24.
[41] J.F. Faucher, T. Baltz, K.G. Petry, Parasitol. Res. 81 (1995) 441–443.
[42] M. Almeida-de-Faria, E. Freymuller, W. Colli, M.J. Alves, Exp. Para-
sitol. 92 (1999) 263–274.
[
14] M.O.F. Khan, S.E. Austin, C. Chan, H.Yin, D. Marks, S.N. Vaghjiani,
H. Kendrick, V. Yardley, S.L. Croft, K.T. Douglas, J. Med. Chem. 43
(
2000) 3148–3156.
[43] K.M. Tyler, D.M. Engman, Int. J. Parasitol. 31 (2001) 472–481.
[44] ClogP version 4.0 (1999), BioByte Corp., 201 W. Fourth St.,
Suite#204, Claremont, CA 91711, USA. Available from
http//www.biobyte.com.
[45] N. Cenas, A. Nemeikaite-Ceniene, E. Sergediene, H. Nivinskas,
Z. Anusevicius, J. Sarlauskas, Biochem. Biophys. Acta 1528 (2001)
31–38.
[
[
15] B. Bonnet, D. Soullez, S. Girault, L. Maes, V. Landry, T.E. Davioud-
Charvet, C. Sergheraert, Bioorg. Med. Chem. 8 (2000) 95–103.
16] Z. Li, M.W. Fennie, B. Ganem, M.T. Hancock, M. Kobaslija, D. Rat-
tendi, C.J. Bacchi, M.C. O’Sullivan, Bioorg. Med. Chem. Lett. 11
(
2001) 251–254.
[
[
[
[
[
17] G.B. Henderson, P. Ulrich, A.H. Fairlamb, I. Rosenberg, M. Pereira,
M. Sela,A. Cerami, Proc. Natl.Acad. Sci. USA 85 (1988) 5374–5378.
18] K. Blumenstiel, R. Schoneck, V. Yardley, S.L. Croft, R.L. Krauth-
Siegel, Biochem. Pharmacol. 58 (1999) 1791–1799.
19] R. Docampo, S.N.J. Moreno, A.O.M. Stoppani, W. Leon, F.S. Cruz,
F. Villalta, R.P.A. Muniz, Biol. Pharmacol. 30 (1981) 1947–1951.
20] R. Docampo, S.N.J. Moreno, in: W.A. Pryor (Ed.), Free Radicals in
Biology, Academic Press, New York, 1984, pp. 243–288.
[46] SYBYL v. 6.8 (2001), Tripos Associates, 1699 South Hanley Road,
Suite 303, St. Louis, Missouri 63144.
[47] C.J. Cramer, D.G. Truhlar, Science 256 (1992) 213–217.
[48] D.B. Jordan, G.S. Basarab, D.I. Liao, W.M. Johnson, K.N. Winzen-
berg, D.A. Winkler, J. Mol. Graph. Model. 19 (2001) 434–447.
[49] N.V. Belkina, V.S. Skvortsov, A.S. Ivanov, A.I. Archakov, Vopr. Med.
Khim. 44 (1998) 464–473.
21] L.J. Vergara, J.A. Squella, J. Aldunate, M.E. Letelier, S. Bollo,
Y. Repetto, A. Morello, P.L. Spencer, Bioelectrochem. Bioenerg. 43
[50] J.D. Maya, Y. Repetto, M. Agosin, J.M. Ojeda, R. Tellez, C. Gaule,
A. Morello, Mol. Biochem. Parasitol. 86 (1997) 101–106.
[51] H. Cerecetto, M. González, M. Risso, G. Seoane, A. López de Ceráin,
O. Ezpeleta, A. Monge, L. Suescun, A. Mombrú, A.M. Bruno, Arch.
Pharm. 333 (2000) 387–393.
[52] M. Bertelli, E. El-Bastawissy, M.H. Knaggs, M.P. Barrett, S. Hanau,
I.H. Gilbert, J. Comput. Aided Mol. Des. 15 (2001) 465–475.
[53] M. Clark, R.D. Cramer III, N. Van Opdenbosch, J. Comp. Chem. 10
(1989) 982–1012.
[54] D. Hall, N. Pavitt, J. Comp. Chem. 5 (1984) 441–450.
[55] J.J.P. Stewart, J. Comput. Aided Mol. Des. 4 (1990) 1–105.
[56] P.J. Goodford, J. Med. Chem. 28 (1985) 849–857.
[57] AMSOL, QCPE Program No. 606 version 4.5, Quantum Chemistry
Program Exchange, Indiana University, Bloomington, Indiana USA.
(
1997) 151–156.
[
[
22] C. Olea-Azar, A.M. Atria, R. Di Maio, G. Seoane, H. Cerecetto,
Spectrosc. Lett. 31 (1998) 849–857.
23] J.D. Maya, S. Bollo, L.J. Nunez-Vergara, J.A. Squella, Y. Repetto,
A. Morello, J. Perie, G. Chauviere, Biochem. Pharmacol. 65 (2003)
9
99–1006.
[
[
24] H. Cerecetto, R. Di Maio, G. Ibarruri, G. Seoane, A. Denicola,
G. Peluffo, C. Quijano, M. Paulino, Farmaco 53 (1998) 89–94.
25] H. Cerecetto, R. Di Maio, M. González, M. Risso, G. Sagrera,
G. Seoane, A. Denicola, G. Peluffo, C. Quijano, M.A. Basombrío,
A.O.M. Stoppani, M. Paulino, C. Olea-Azar, Eur. J. Med. Chem. 35
(
2000) 343–350.