Exploring the interplay of physicochemical properties, membrane permeability and giardicidal activity of some benzimidazole derivatives

Article abstract of DOI:10.1016/j.ejmech.2012.03.014

This study evaluated the relationship between the physicochemical properties, membrane permeability and in vitro giardicidal activity of twenty nine benzimidazole derivatives (1-7n). The retention time data from reverse phase high performance chromatography (RP-HPLC) were used to estimate aqueous solubility and lipophilicity of these compounds. The apparent permeability was determined using Caco-2 cell monolayer. The calculation of some descriptors, such as Clog P, PSA, was performed using ACD labs software. For benzimidazole derivatives with NHCOOCH₃, CH₃, NH₂, SH and SCH₃ groups at the 2-position, a quadratic type of regression model was obtained with giardicidal activity and aqueous solubility or lipophilicity. On the other hand, giardicidal activity of 2-(trifluoromethyl)-1H-benzimidazole derivatives was influenced by lipophilicity, hydrogen bond donors and molecular volume but it was not determined by their apparent permeability in Caco-2 cell line.

Full text of DOI:10.1016/j.ejmech.2012.03.014

European Journal of Medicinal Chemistry 52 (2012) 193e204
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Exploring the interplay of physicochemical properties, membrane permeability
and giardicidal activity of some benzimidazole derivatives
,
Carlos Hernández-Covarrubias a b, Miguel A. Vilchis-Reyes a, Lilian Yépez-Mulia c,
,
Remedios Sánchez-Díaz a d, Gabriel Navarrete-Vázquez e, Alicia Hernández-Campos a, Rafael Castillo a,
,
Francisco Hernández-Luis a
*
a Facultad de Química, Departamento de Farmacia, Universidad Nacional Autónoma de México, México, DF 04510, Mexico
b Universidad Tecnológica de Tlaxcala, Huamantla, Tlax. 90500, Mexico
c Unidad de Investigación en Enfermedades Infecciosas y Parasitarias, Instituto Mexicano del Seguro Social, México, DF 06720, Mexico
d CISALUD, Valle de Las Palmas, Universidad Autónoma de Baja California, Tijuana, BC 22390, Mexico
e Facultad de Farmacia, Universidad Autónoma del Estado de Morelos, Cuernavaca, Mor. 62209, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
This study evaluated the relationship between the physicochemical properties, membrane permeability
and in vitro giardicidal activity of twenty nine benzimidazole derivatives (1e7n). The retention time data
from reverse phase high performance chromatography (RP-HPLC) were used to estimate aqueous
solubility and lipophilicity of these compounds. The apparent permeability was determined using Caco-2
cell monolayer. The calculation of some descriptors, such as Clog P, PSA, was performed using ACD labs
software. For benzimidazole derivatives with NHCOOCH3, CH3, NH2, SH and SCH3 groups at the 2-
position, a quadratic type of regression model was obtained with giardicidal activity and aqueous
solubility or lipophilicity. On the other hand, giardicidal activity of 2-(trifluoromethyl)-1H-benzimidazole
derivatives was influenced by lipophilicity, hydrogen bond donors and molecular volume but it was not
determined by their apparent permeability in Caco-2 cell line.
Received 31 October 2011
Received in revised form
6 March 2012
Accepted 8 March 2012
Available online 15 March 2012
Keywords:
Benzimidazole
Giardia duodenalis
Solubility
Ó 2012 Elsevier Masson SAS. All rights reserved.
Lipophilicity
Caco-2 permeability
1. Introduction
Consequently, the therapeutic control of this disease is worsening.
Due to these reasons, there exists a real necessity of discovering
Giardia duodenalis (syn. Giardia lamblia or Giardia intestinalis) is
a parasite that colonizes the duodenum and upper jejunum of
humans causing a disease known as giardiasis. During its life cycle
presents a parasitic stage represented by the vegetative trophozoite
and a resistant form responsible for transmission, denominated
cyst, which is eliminated in the host feces [1]. Giardiasis has been
recently included in the WHO Neglected Diseases Initiative [2]. The
major burden of infection occurs in children in developing coun-
tries, where infection is often ubiquitous [1,2]. Due to its role in
diarrheal disease outbreaks in industrialized countries, giardiasis is
considered as a re-emerging infectious disease [3]. The available
drugs for the giardiasis treatment are directed against the tropho-
zoite stage. Metronidazole and other nitroimidazoles have been
used since the late 1950s; however, side effects on patients and
clinical resistance to these drugs have been well documented [3].
alternative new drugs for this parasitic infection.
As a part of ongoing research programs to find new molecules
for the treatment of giardiasis, a series of benzimidazole derivatives
bearing different substituents at the positions C(1), C(2), C(5) and
C(6), have previously been synthesized and in vitro tested against
G. duodenalis [4e8]. The in vitro antiparasitic activity of a number of
them, sometimes close to or better than Metronidazole (MTZ), has
prompted to investigate the physicochemical properties and
membrane permeability to optimize the giardicidal activity of this
group of compounds.
Lead optimization based on biological activity and physico-
chemical profile plays a major role in modern drug discovery stage
[9]. Some of the properties included in this profile are the aqueous
solubility, lipophilicity and membrane permeability [10e13].
Aqueous solubility (S) [10] and lipophilicity, usually expressed by 1-
octanol/water distribution coefficient, log P [11,12], are among the
most important properties that influence the transport, uptake, and
distribution of chemicals in biological systems [13]. With respect to
membrane permeability, this property often ensures good oral
* Corresponding author. Tel./fax: þ52 56225287.
E-mail address: franher@unam.mx (F. Hernández-Luis).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2012.03.014

194
C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
absorption and sufficient target tissue uptake of the compounds.
Undoubtedly, the permeation of a molecule is determined by the
delicate balance of numerous parameters that are clearly interre-
lated such that changing one will affect the others [14]. Nowadays,
there is no scientific data on the influence of these parameters in
the giardicidal activity of benzimidazole derivatives.
In this paper, the physicochemical profile determination of
twenty-nine benzimidazole derivatives (Table 1) and its correlation
with membrane permeability and the in vitro antiparasitic activity
on G. duodenalis trophozoites are presented. Knowledge of optimal
physicochemical properties and membrane permeability of
a compound is crucial for giardicidal activity because the in vivo
contact time with G. duodenalis is no more than 2 h due to the
location of the parasite in the human duodenum and upper
jejunum.
SigmaeAldrich. Compounds 3ae7d were prepared by methods
described in the literature [4,15e18]. Synthesis of 7ee7n was
carried out by the sequence of reactions summarized in Scheme 1.
The required substituted 1,2-phenylenediamines (9a, 9b and 14)
were prepared by reduction of the corresponding 2-nitroanilines
(8a, 8b and 13) with hydrogen in the presence of 5% palladium
on carbon. 1,2-Phenylendiamine 12 was produced by basic hydro-
lysis of 13 as describe above. The synthesis of intermediate 13
started with the methylation from 3-Hydroxy-4-nitrobenzoic acid
(10) with Dimethylsulfate and NaOH. The compound obtained (11)
was treated, under pressure in a seal reactor, with an excess of
CH3NH2 to give the expected N-methylacetamide (13). The prod-
ucts of reduction without isolation were cyclocondensed with
CF3COOH and HCl as a catalyst to give the respective 2-(tri-
fluoromethyl)-1H-benzimidazole derivative (7ee7g and 7n).
Compound 7h was obtained when 7e reacted with 1,10-Carbon-
yldiimidazole (CDI) and MeOH; meanwhile, 7i and 7j were
obtained when 7f and 7g reacted with Dimethylsulfate and
NaHCO3. Reactions of 7f or 7g with 2-Chloro-N,N-diethylacetamide
afforded 7k and 7l, respectively. Finally, the preparation of 7m was
carried out when 7e became acylchloride with SOCl2 and then was
treated with CH3NH2. All compounds synthesized were obtained as
solids and their structures were established by infrared (IR) spectra,
proton nuclear magnetic resonance (1H NMR), mass spectra (MS)
and elementary analysis.
2. Chemistry
Benzimidazole derivatives studied were arranged according to
the substituent attached at the 2-position (Table 1). 1H-Benzimid-
azole (1), Methyl-1H-benzimidazol-2-ylcarbamate (Carbendazim,
2a),
Methyl
[5-(propylthio)-1H-benzimidazol-2-yl]carbamate
(Albendazole, 2b), Methyl [5-(phenylthio)-1H-benzimidazol-2-yl]
carbamate (Fenbendazole, 2c), Methyl (5-benzoyl-1H-benzimida-
zol-2-yl)carbamate (Mebendazole, 2d) were supplied by
Table 1
Structure and giardicidal activity of benzimidazole derivatives.
4
R
N
1
R
3
N
R
2
R
Compound
R1
R2
R3
R4
mp, ꢀC
MW
IC50 (mM), G. duodenalis
1
H
H
H
H
H
H
H
H
H
H
H
Cl
Cl
H
H
H
Cl
H
H
Cl
H
H
H
C3H7S
C6H5S
C6H5C(O)
H
H
Cl
Cl
H
Cl
H
H
Cl
H
H
Cl
Cl
H
COOH
COOH
H
COOCH3
COOCH3
172e173
302e307
208e210
233e234
288e289
106e108
157e159
197e200
210e211
189e191
230e231
273e275
53e54
118.14
191.19
265.34
299.35
295.30
146.19
180.63
215.08
181.62
164.23
198.67
198.67
178.25
212.70
212.70
186.11
255.02
234.61
234.61
230.14
244.17
244.17
244.16
258.19
258.19
357.33
357.33
243.15
257.21
136.538
0.057
0.037
0.056
0.055
0.089
0.044
0.055
0.242
0.018
0.045
0.020
0.033
0.028
0.122
0.107
0.078
0.042
0.127
0.172
2.770
0.330
1.190
2.450
3.820
0.133
0.470
0.320
0.211
2a
2b
2c
2d
3a
3b
3c
4
5a
5b
5c
6a
6b
6c
7a
7b
7c
7d
7e
7f
NHCOOCH3
NHCOOCH3
NHCOOCH3
NHCOOCH3
CH3
CH3
CH3
NH2
SH
SH
SH
SCH3
SCH3
SCH3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
H
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
92e93
80e81
209e211
236e238
108e109
158e160
280e281
252e254
270e272
165e166
121e122
107e108
97e98
H
Cl
H
Cl
H
H
COOH
H
H
COOCH3
H
COOCH2CON(C2H5)2
H
CH3
CH3
H
CH3
CH3
H
CH3
CH3
CH3
CH3
H
7g
7h
7i
7j
7k
7l
H
COOCH2CON(C2H5)2
H
CONHCH3
H
115e116
259e261
218e219
7m
7n
CF3
CH3
CONHCH3

C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
195
HOOC
HOOC
NH
N
N
NO
R
HOOC
2
2
b
a
CF
3
R
R
8a: R = NH2
9a: R = NH2
7e: R = H
7f: R = CH3
8b: R = NHCH3
9b: R = NHCH3
d
c
7i
7h
7e
7f
f
e
7k
7m
HOOC
HOOC
OH
NO
CH OOC
OCH
NHCH
3
g
3
3
NO
NH
2
2
2
12
11
13
10
i, a
h
b
CH
3
CH NHOC
3
CH NHOC
NHCH
NHCH
HOOC
3
3
N
3
a
CF
3
NO
N
NH
2
2
7g
14
b
d
f
7n
7j
7l
Scheme 1. Reagents and conditions: (a) H2, CePd 5%; (b) CF3CO2H, HCl (cat), heat; (c) DCI, CH3OH; (d) (CH3)2SO4, NaHCO3;(e) SOCl2, CH3NH2, (C2H5)3N; (f) ClCH2CON(C2H5)2,
(C2H5)3N, KI (cat); (g) (CH3)2SO4, NaOH; (h) CH3NH2; (i) NaOH/MeOH.
3. Results and discussion
As regards the 5- or 6-positions, 1-Methyl-2-(trifluoromethyl)-
1H-benzimidazole derivatives containing COOH (7f, 7g), COOCH3
(7i, 7j), COOCH2CON(C2H5)2 (7k, 7l) and CONHCH3 (7n) presented
a lower activity compared to analogues with Chlorine atom (7c, 7d)
3.1. Antiprotozoal activity
The biological assays for compounds 1e6c and 7ee7n were
carried out as reported before [4e8]. Antiparasitic activity of 7ae7d
was taken from reference [4]. The results displayed in Table 1
summarize the effect of substituents at the 1-, 2-, 5- and 6-
positions of benzimidazole nucleus on giardicidal activity. With
the exception of 1, 7f, 7i and 7j, all tested compounds were more
at the same positions (IC50 ¼ 0.042 and 0.127
mM, respectively).
These data pointed out this region with a significant role for the
improvement of the antiparasitic profile.
With respect to substitution at 1-position, the results reinforce
the information obtained in previous studies [4,7,8] that showed
that 1-Methyl-1H-benzimidazole should be considered a scaffold
for development of new giardicidal compounds.
active than MTZ (IC50 ¼ 1.22
mM) against G. duodenalis. Among the
six types of substituents at the 2-position (NHCOOCH3, CH3, NH2,
SH, SCH3, CF3), the better profile was observed with SH (5ae5c,
IC50 ¼ 18e45 nM); these compounds exhibited approximately
a 67-, 61- and 27-fold higher activity than MTZ, respectively. The
lower antiparasitic activity profile involved derivatives that contain
CF3 group (7a, 7de7n, IC50 > 100 nM), except for 7b and 7c which
presented IC50 values of 78 and 42 nM, respectively. It should be
noted that the incorporation of a CF3 group in a scaffold increases
its lipophilicity thereby enhancing membrane permeability as well
as favoring the docking with receptor(s); however in this study, 2-
(trifluoromethyl)-1H-benzimidazole derivatives presented a lower
activity compared to other benzimidazole derivatives.
3.2. Lipophilicity and aqueous solubility
Lipophilicity has been pointed out as a major physicochemical
property of benzimidazole anthelmintics, because it arise the drug
availability in a target parasite [13]. In fact, besides their numerical
value, lipophilicity data contain a variety of information about
inter- and intra-molecular forces affecting partitioning, binding,
and their related biological phenomena [10]. Nowadays, a number
of reliable calculation approaches are available for the prediction of
the lipophilicity of the neutral species (log P) [19]. Less efficient is
the estimation of distribution coefficient (log D) for acid or base

196
C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
compounds, which further involves the prediction of pKa values
and the effect of counter ions, as well [20]. In the case of neutral
compounds, log P and log D are identical. With regard to aqueous
solubility, this property is usually expressed as log 1/S, where S is
the saturated compound concentration in equilibrium with solid
under defined conditions such as physiological pH at room
temperature. This property is the most difficult to predict because is
a function of many molecular parameters (polarity, size, molecular
structure, stereochemistry, electronic structure) and hence remains
one of the most measured physical properties in a drug discovery
step [21]. Aqueous solubility is a significant property to achieve for
in vivo giardicidal activity because the molecules must be dissolved
at the anatomical site of action (duodenum and upper jejunum).
For experimental determination of lipophilicity and aqueous
solubility a number of chromatographic approaches have been
applied with accuracy and reliability using reversed-phase high
performance liquid chromatography (RP-HPLC) methods [22e29].
A general technique uses the retention factor (log k), based on the
chromatographic retention time, which permits to estimate lip-
ophilicity, and aqueous solubility with good correlations. Chro-
matographic methods are a promising alternative to the shake flask
method, with advantages such as a higher throughput, inattention
to impurities or degradation products and broader lipophilicity
range [30]. Moreover, the mobile/stationary phase models are
better for the biological partitioning process than the solute par-
titioning in the bulk 1-octanol/water phase [31]. Additionally, RP-
HPLC retention is more suitable for studying the interaction
between compounds and membrane lipids than bulk-phase
hydrocarbon/water partitioning because the alkyl chains of
stationary phases are structured and the penetration of compounds
into these membranous chains is based on an entropy-driven
process not associated with the transfer into bulk lipids [32,33].
In this work, log D7.4 of each compound were determined using
RP-HPLC. For this purpose, an appropriate set of control compounds
(1e2d, Thiabendazole and Metoprolol), with log D7.4 [13,34e36]
and molar aqueous solubility (25 ꢀC, pH ¼ 7.4) [35e40] known,
were injected at isocratic elution (90, 80, 75, 70 and 60% MeOH/
H2O) in order to calibrate the RP-HPLC system. The log k was
calculated using their retention times (tr) and dead time (t0),
(log k ¼ log [(tr ꢁ t0)/t0]) [29]. Since the partition of the chemicals
between the mobile phase and the hydrocarbon stationary phase is
closely related to their partition coefficients, only the retention
times are required in the RP-HPLC method while concentration
determination is not necessary. Obtained log k values were plot
against isocratic MeOH concentrations to establish the linear
regression equations for each control compound. Their intercepts
(log kw), which are the extrapolated retention factor to zero MeOH
concentration, were used to establish a regression analysis. The
calibration curves were constructed by plotting log kw versus
log D7.4 and log kw versus log 1/S. The linear correlations were
calculated according equations (1) and (2).
lipophilicity (Clog D7.4) and no correlation was found (r2 ¼ 0.14,
n ¼ 33, F ¼ 5) (Fig. 1). The difference between experimental and
calculated values seems to be due to several factors. First, the inter-
molecular effects that govern the benzimidazole derivatives lip-
ophilicity in a nonpolar stationary phase (column C18 AtlantisÔ)
may lead to higher values of lipid solubility. Second, Clog D7.4
included the tautomers forms 4amine/imine, 5athiol/thione, 5bthiol/thione
and 5cthiol/thione (Table 2) (Scheme 2); experimental data can not
establish this differentiation. Third, although most of benzimid-
azole derivatives were preferentially as neutral molecules at
pH ¼ 7.4 due to their pKa1 values (log D ¼ log P) (Table 3),
compounds 7e, 7f and 7g were ionized (COOH, pKa1 values
3.62e3.85) (Scheme 2); situation that RP-HPLC procedure did not
distinguish and yielded values close to log P (2.07, 2.39 and 2.39,
respectively) instead of Clog D7.4 (ꢁ0.96, ꢁ0.63 and ꢁ0.55,
respectively) (Table 2). In spite of that, our experimental approach
could discriminate between two values of these properties from
benzimidazole regioisomer (5b or 5c, 6b or 6c, 7c or 7d, 7f or 7g, 7i
or 7j and 7k or 7l).
Similarly, a study was carried out to establish the correlation
between S and calculated solubility (Scalc., Table 2). The results were
similar to those of lipophilicity. Scalc. was obtained from general
solubility equation (GSE) proposed by Yalkowsky and co-workers
[41,42], log S ¼ ꢁ0.01(mp ꢁ 25) ꢁ Clog P þ 0.5. This equation
provides a simple method of estimating the molar aqueous solu-
bility of a neutral organic compound as a function of its celsius
melting point (mp) and log P.
Since lipophilicity is an important factor to explain the biological
activityofbenzimidazole anthelmintics [13], first it was explored the
lipophilicityegiardicidal activity relationship of 1e7n. The results
showed no correlation with log D7.4 (r2 ¼ 0.20, n ¼ 29, F ¼ 6) or
Clog D7.4 (r2 ¼ 0.01, n ¼ 29, F ¼ 0.3). Second, the possible aqueous
solubilityegiardicidal activity relationship was investigated; the
results were similar to lipophilicity (r2 ¼ 0.19, n ¼ 29, F ¼ 6).
Under the above results, we decided to separate compounds
2ae7n into two categories. The first included compounds 2ae6c;
the second covered 2-(trifluoromethyl)-1H-benzimidazole deriva-
tives (7ae7n). Since compound 1 showed poor activity against
G. duodenalis (IC50 ¼ 136.538
mM), it was excluded.
For the first category (2ae6c), a quadratic type of regression
model was obtained with giardicidal activity and lipophilicity or
aqueous solubility. A significant (p ¼ 0) link between solubility and
antiparasitic activity was established in equation (3).
IC50 ¼ 0:198 ꢁ 0:079 log1=S þ 0:008 ð1=SÞ2
À
Á
(3)
r2 ¼ 0:823; n ¼ 14; s ¼ 0:02; F ¼ 25
Fig. 2 shows the parabolic curve obtained from IC50 versus log 1/
S. The minimum of the curve is centered on the compounds with
IC50 < 0.05 M and log 1/S between 3.5 and 5.5. Although the
m
biphasic dependence observed in Fig. 2 can be interpreted as
a consequence of a transport-controlled process, IC50 values
depend upon both rate of penetration into protozoa and affinity of
the benzimidazole derivative for its receptor.
For the second category (7ae7n), there were no significant
correlations (p ꢂ 0.36) between IC50 and log D7.4 or aqueous
solubility.
logD7:4 ¼ 1:15 kw þ 1:29
À
Á
Á
(1)
r2 ¼ 0:917; n ¼ 7; s ¼ 0:49; F ¼ 55
log1=S ¼ 2:79 logkw þ 0:228
À
(2)
r2 ¼ 0:953; n ¼ 6; s ¼ 0:55; F ¼ 81
In these and the following equations: r2, the squared correlation
coefficient; n, the number of the compounds; s, the standard
deviation and F the Fisher’s test.
3.3. Membrane permeability
After obtaining the calibration chart, benzimidazole derivatives
(3ae7n) were injected to RP-HPLC and their log kw were deter-
mined in order to estimate their experimental lipophilicity
(log D7.4) and aqueous solubility (S) value (Table 2). Experimental
lipophilicity data (1e7n) were compared with the calculated
One of the most frequently used in vitro systems for deter-
mining membrane permeability is the Caco-2 cell system. At their
critical density, Caco-2 cells form a polar monolayer with apical (A,
normally adjacent to the intestinal lumen) and basolateral (B,
normally adjacent to the blood supply) sides [43]. It is generally

C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
197
Table 2
Experimental and calculated solubility and lipophilicity of benzimidazole derivatives.
Scalc., Mf
Log D7.4
Clog D7.4
Clog Pg
g
Compound
Log kw
S,
m
g/mL (M)
Log 1/S
1
0.27
0.90
1.95
2.44
1.68
0.42
1.05
2.24
3.19
1.23
1.61
1.90
1.78
2.46
2.49
1.80
2.78
2.33
2.37
2.09
1.47
1.13
1.94
2.05
2.24
2.40
2.29
1.21
1.33
7604a (3.49 ꢃ 10ꢁ2
2441b (1.27 ꢃ 10ꢁ2
0.750c (2.80 ꢃ 10ꢁ6
)
)
)
)
)
)
)
)
1.45a
1.89b
5.55c
7.07d
5.43c
1.40
3.15
6.47
9.12
3.65
4.71
5.59
5.19
7.09
7.17
5.27
8.00
6.75
6.85
6.08
4.35
3.39
5.64
5.96
6.47
6.94
6.62
3.60
3.93
4.41 ꢃ 10ꢁ3
1.53 ꢃ 10ꢁ4
3.89 ꢃ 10ꢁ5
4.62 ꢃ 10ꢁ6
1.08 ꢃ 10ꢁ5
4.89 ꢃ 10ꢁ3
3.89 ꢃ 10ꢁ4
5.18 ꢃ 10ꢁ5
3.93 ꢃ 10ꢁ4
6.60 ꢃ 10ꢁ4
4.02 ꢃ 10ꢁ5
2.08 ꢃ 10ꢁ5
3.05 ꢃ 10ꢁ3
3.19 ꢃ 10ꢁ4
4.21 ꢃ 10ꢁ4
1.77 ꢃ 10ꢁ4
4.26 ꢃ 10ꢁ6
2.16 ꢃ 10ꢁ4
6.76 ꢃ 10ꢁ5
7.49 ꢃ 10ꢁ5
6.76 ꢃ 10ꢁ5
4.46 ꢃ 10ꢁ5
7.32 ꢃ 10ꢁ4
1.53 ꢃ 10ꢁ2
2.11 ꢃ 10ꢁ2
3.50 ꢃ 10ꢁ3
2.31 ꢃ 10ꢁ3
1.28 ꢃ 10ꢁ3
1.64 ꢃ 10ꢁ3
1.37a
1.49b
3.83e
3.93e
3.73e
1.77
2.49
3.86
4.95
2.70
3.14
3.47
3.33
4.11
4.15
3.36
4.49
3.98
4.02
2.59
2.98
3.70
3.52
3.65
3.86
4.06
2.81
2.68
3.92
1.37
1.49
3.06
3.75
2.83
1.96
2.58
3.05
1.38 ꢄ 0.20
1.52 ꢄ 0.40
3.07 ꢄ 0.40
3.75 ꢄ 0.85
2.83 ꢄ 0.87
1.99 ꢄ 0.52
2.58 ꢄ 0.57
3.05 ꢄ 0.57
2.05/1.65h
2.60/1.46i
2a
2b
2c
2d
3a
3b
3c
4
5a
5b
5c
6a
6b
6c
7a
7b
7c
7d
7e
7f
0.0250d (8.35 ꢃ 10ꢁ8
1.1100e (3.70 ꢃ 10ꢁ6
5785.3230(3.95 ꢃ 10ꢁ2
125.6868(6.95 ꢃ 10ꢁ4
0.0716(3.32 ꢃ 10ꢁ7
0.0001(7.44 ꢃ 10ꢁ10
)
2.02/0.72h
0.19/1.46i
1.02/2.49i
1.10/2.18i
2.72
3.32
3.32
2.40
3.09
3.30
3.30
ꢁ0.96
ꢁ0.63
ꢁ0.55
2.23
1.35
1.35
2.23
2.23
35.9541(2.10 ꢃ 10ꢁ4
3.7860(1.90 ꢃ 10ꢁ5
0.5876(2.95 ꢃ 10ꢁ6
1.1390(6.39 ꢃ 10ꢁ6
0.0172(8.10 ꢃ 10ꢁ8
0.0142(6.68 ꢃ 10ꢁ8
0.9922(5.33 ꢃ 10ꢁ6
0.0025(9.91 ꢃ 10ꢁ9
0.0412(1.75 ꢃ 10ꢁ7
0.0330(1.40 ꢃ 10ꢁ7
0.1886(8.19 ꢃ 10ꢁ7
10.9010(4.46 ꢃ 10ꢁ5
98.7273(4.04 ꢃ 10ꢁ4
0.5549(2.27 ꢃ 10ꢁ6
0.2807(1.08 ꢃ 10ꢁ6
0.0859(3.32 ꢃ 10ꢁ7
0.0401(1.12 ꢃ 10ꢁ7
0.0572(2.35 ꢃ 10ꢁ7
87.8192(2.45 ꢃ 10ꢁ4
29.6201(1.15 ꢃ 10ꢁ4
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
3.20/2.49i
3.20/2.18i
2.73 ꢄ 0.57
3.32 ꢄ 0.57
3.32 ꢄ 0.59
2.40 ꢄ 0.18
3.75 ꢄ 0.23
3.33 ꢄ 0.91
3.33 ꢄ 0.91
2.07 ꢄ 0.71
2.39 ꢄ 1.18
2.39 ꢄ 1.18
2.23 ꢄ 1.25
1.35 ꢄ 1.21
1.35 ꢄ 1.21
2.23 ꢄ 1.21
2.23 ꢄ 1.21
1.04 ꢄ 0.75
1.35 ꢄ 1.21
7g
7h
7i
7j
7k
7l
7m
7n
0.95
1.35
a
Reference [35].
Reference [36].
Reference [37].
Reference [38].
b
c
d
e
Reference [13].
f
Calculated from general solubility equation: log S ¼ ꢁ0.01(mp ꢁ 25) ꢁ Clog P þ 0.5.
Calculated from ACD/Labs software v.9.0 (Advanced Chemistry Development, Inc.).
Amine/imine tautomers.
g
h
i
Thiol/thione tautomers.
accepted that the apparent permeability coefficient (Papp) obtained
in Caco-2 cells reflects the ability of a compound to permeate
through the cell membrane (transcellular or paracellular route). In
this sense, the resultant Papp can be applied to predict Giardia
membrane permeability to the test molecules; although it should
be possible that differences between both membranes are found in
the size of the aqueous porous and the composition of lipoidal
membrane, the biophysical processes of transfer through any of
these biological barriers should be very similar. An additional
application of Caco-2 system for giardicidal compounds is that have
a functional expression of P-glycoprotein (P-gp) and other active
efflux transporters, such as BCRP and MRP, which reduce apical to
basolateral (AB) permeability by pumping substrates back out of
the cells [44]. Therefore, if a compound shows efflux behavior in
Caco-2 cells can be assumed that in vivo studies would return to the
intestinal lumen, in the duodenum and jejunum, where it could act
against G. duodenalis. This behavior would mean that the
compound was more time on these anatomical parts, thereby
increasing its giardicidal efficiency. Previous research has demon-
strated correlations between log P, molecular weight (MW), polar
surface area (PSA), H-bonding capabilities and the compounds
propensity to interact with P-gp [45]. It would provide valuable
insights into development of a successful giardicidal candidate.
In this study, permeability across differentiated monolayers of
Caco-2 was measured on fully differentiated cells grown for 28 days
on permeable filter supports. The bilateral permeation of twenty
one benzimidazole derivatives (2b, 2d, 3ae3c, 4, 5ae5c, 6ae6c,
7ee7l and 7n) increased linearly as a function of time up
100 min. Papp was calculated in the apical to basolateral (AB) and
basolateral to apical (BA) directions. The accumulated amount of
compound appearing in the receptor compartment over time, dC/
dt, was used to calculate Papp with the following equation [31]:
Vr
60ðC0AÞ
Papp
¼
dC=dt
where Vr is the volume of receptor compartment (cm3), C0 is the
initial concentration of compound in the donor compartment ( M),
m
Fig. 1. Clog D7.4 and log D7.4
.

198
C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
H
4
4
4
+
R
R
R
N
N
N
N
pKa2
pKa1
1
1
1
R
R
R
3
3
N
_
3
N
R
R
R
2
R
2
R
(1-2d, 7a_7b, 7e, 7h)
(1-3c, 4,6a-7n)
(1-3c, 4,6a-7n)
H
N
4
4
R
R
N
S
SH
3
3
N
N
R
R
CH
3
CH
3
thione form
thiol form
(5a-5c)
(5a-5c)
pKa2
pKa1
_
N
H
4
4
R
+
R
N
S
SH
3
N
3
R
N
CH
R
CH
3
3
Scheme 2. Acidebase equilibrium of benzimidazole derivatives studied.
Table 3
Calculated physicochemical properties.
PSA,a Å2
MV,a cm3
HBAc
HBDd
a
a
a,b
Compound
pKa1
pKa2
pKa3
1
5.67
6.09
5.62
5.47
5.02
6.16
5.15
3.89
6.43
2.69
2.27
2.37
5.44
4.22
4.18
3.03
0.38
1.92
1.88
1.99
1.92
1.88
1.93
1.92
1.88
1.81
1.77
1.98
1.98
12.60
11.97
11.20
11.08
10.65
e
e
28.68
67.01
92.31
92.31
84.08
17.82
17.82
17.82
95.10
134.54
203.19
213.80
212.66
134.03
143.41
152.78
1
4
4
4
5
1
1
1
2
1
1
1
1
1
1
1
1
1
1
3
3
3
3
3
3
5
5
3
3
1
2
2
2
2
0
0
0
2
0
0
0
0
0
0
1
1
0
0
2
1
1
1
0
0
0
0
2
1
2a
2b
2c
2d
3a
3b
3c
17.40 (NHCOOCH3)
17.17 (NHCOOCH3)
17.13 (NHCOOCH3)
16.99 (NHCOOCH3)
e
e
e
e
e
43.84/39.12e
56.62/47.36f
56.62/47.36f
56.62/47.36f
43.12
43.12
43.12
28.68
28.68
17.82
17.82
65.98
55.12
55.12
64.43
46.92
46.92
129.97/124.97e
127.50/123.32f
136.88/134.47f
136.68/134.37f
148.61
157.98
157.98
128.61
152.51
158.49
158.49
141.14
159.89
159.89
271.94
182.55
182.55
e
4amine/imine
5athiol/thione
5bthiol/thione
5cthiol/thione
6a
6b
6c
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
e
e
f
11.19
10.58
10.39
e
e
f
e
f
e
e
e
e
e
e
9.68
6.83
e
e
e
e
e
e
10.07
e
3.62 (COOH)
3.66 (COOH)
3.85 (COOH)
e
e
7.80
e
e
e
e
e
e
64.43
64.43
57.78
46.92
271.94
271.94
168.95
182.55
e
e
7m
7n
e
14.10 (CONHCH3)
14.00 (CONHCH3)
e
a
Calculated from ACD/Labs software v.9.0 (Advanced Chemistry Development, Inc.).
From subtitutents at 2- or 5-positions.
Hydrogen bond acceptors groups.
Hydrogen bond donors groups.
Amine/imine tautomers.
b
c
d
e
f
Thiol/thione tautomers.

C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
199
having high permeability. Propanolol, a transcellular marker with
lipophilic character, showed a Papp value of 48.35 ꢃ 10ꢁ6 cm/s.
Compound 7e had
a
minor permeability to Atenolol
(1.77 ꢃ 10ꢁ6 cm/s), the paracellular transport marker. The magni-
tude of bidirectional flux of 7e was minor to Atenolol, suggesting
paracelullar transport for this compound. Nevertheless, in parallel,
part of the transport may occur via the transcellular route due to
molecular weight of this compound (MW ¼ 230.14) [49]. Inter-
esting, with exception of 7e, the rest of 2-(trifluoromethyl)-1H-
benzimidazol derivatives (7fe7k) presented AB/BA > 1, point to
a predominance of AB over BA directions. When transport rates
were monitored in the direction (BA), Papp higher values were
observed for 3ae3c, 4, 5ae5c, 6ae6c and 7e. These results indi-
cated a trend of lower transport in the absorptive direction than in
the reflux direction.
In order to determine whether active transport was occurring
for these compounds, efflux ratio (BA/AB) was determined. None of
the compounds evaluated had a BA/AB > 2 value, therefore the
involvement of a transporter or of efflux pumps, such as P-glyco-
protein, was discarded.
Since passive diffusion and carrier-mediated flux are known to
be the two major pathways for a molecule to permeate across the
intestinal epithelium, the above results indicate that the benz-
imidazole derivatives studied move through the Caco-2 mono-
layers by passive diffusion mechanism. With respect to Pt
(estimated transcellular permeability ¼ (AB þ BA)/2) (Table 4),
values for 5a, 5c, 7he7n indicate that these compounds have high
intrinsic membrane permeability while 3a, 3c and 6c have
moderate permeability. Interestingly, Pt values of 7f and 7g were
higher than 7e (8.54, 4.87 versus 0.65) even though all have one
carboxylic acid residue at same position. This result might imply
that factors other than the polarity and charge, for example,
physicochemical profile (log P, log 1/S, PSA and MV) of 7f and 7g
(Tables 2and 3) or spatial arrangement of the molecule, might also
influence membrane permeability.
Fig. 2. Giardicidal activity and log 1/S for 2ae6c.
A is the area of insert (1.13 cm2) and 60 is a conversion factor
(s minꢁ1). The bilateral Papp values for selected benzimidazole
derivatives are summarized in Table 4. The actual numerical Papp
values obtained in the Caco-2 cell permeability assay depended
greatly on various factors, such as passage number, age of the cells,
transepithelial electrical resistance (TEER) values, and the media
and growth conditions used. Therefore, the technique is best used
in a comparative manner with all the compounds assessed in the
same assay under the same conditions [46].
The integrity of Caco-2 monolayers at the end of experiments
was maintained, indicating confluence with well established tight
junctions. This result shows that the benzimidazole derivatives
were not toxic to the cells in any way and do not disrupt the tight
junctions in the monolayer.
According to previous study [47,48], compounds with
P
app < 1 ꢃ10ꢁ6, 1e10 ꢃ 10ꢁ6, and >10 ꢃ 10ꢁ6 cm/s were defined as
poorly, moderately, and highly permeable, respectively.
In order to explore the correlation between membrane
permeability parameters and experimental (log D7.4, log 1/S) or
calculated (PSA, MV, hydrogen-bond acceptor groups (HBA),
hydrogen-bond donor groups (HBD), (MW), pKa) descriptors of
The Papp for the AB flux of most benzimidazole derivatives tested
ranged from 2 ꢃ 10ꢁ6 to 10 ꢃ 10ꢁ6 cm/s, therefore they can cate-
gorized as a moderate permeability group. Compounds 5c, 7h, 7i,
7j, and 7k had Papp > 10 ꢃ 10ꢁ6 cm/s, so they can be classified as
Table 4
Papp of selected benzimidazole derivatives.
Compound
Papp ꢃ 10ꢁ6, cm/s ABa
Papp ꢃ 10ꢁ6, cm/s BAa
Influx ratio (AB/BA)
Efflux ratio (BA/AB)
Ptb ꢃ 10ꢁ6, cm/s
2b
2d
3a
3b
3c
4
5a
5b
5c
6a
6b
6c
7e
7f
7g
7h
3.14 ꢄ 0.03
5.15 ꢄ 0.26
9.03 ꢄ 0.16
7.80 ꢄ 0.23
7.98 ꢄ 0.72
5.21 ꢄ 0.79
8.14 ꢄ 1.01
6.92 ꢄ 0.69
14.10 ꢄ 1.36
7.12 ꢄ 0.95
7.59 ꢄ 1.54
7.77 ꢄ 0.87
0.57 ꢄ 0.03
9.83 ꢄ 1.09
5.09 ꢄ 0.62
32.69 ꢄ 2.15
26.66 ꢄ 1.23
39.25 ꢄ 1.12
21.94 ꢄ 1.09
15.37 ꢄ 1.21
14.58 ꢄ 0.13
48.35 ꢄ 2.11
1.77 ꢄ 0.03
1.99 ꢄ 0.07
4.34 ꢄ 0.13
10.20 ꢄ 0.28
8.82 ꢄ 0.19
11.20 ꢄ 0.75
9.57 ꢄ 1.38
14.50 ꢄ 1.19
9.57 ꢄ 1.38
27.20 ꢄ 1.18
9.57 ꢄ 1.38
14.20 ꢄ 1.09
9.57 ꢄ 1.38
0.74 ꢄ 0.08
7.25 ꢄ 0.94
4.65 ꢄ 0.25
22.99 ꢄ 2.17
23.87 ꢄ 2.03
21.76 ꢄ 1.11
11.01 ꢄ 0.66
13.04 ꢄ 1.11
8.98 ꢄ 0.03
37.89 ꢄ 2.10
1.01 ꢄ 0.02
1.35
1.18
0.88
0.88
0.71
0.54
0.56
0.72
0.51
0.74
0.53
0.81
0.77
1.35
1.09
1.42
1.11
1.80
1.99
1.17
1.62
1.27
1.75
0.74
0.84
1.13
1.13
1.40
1.83
1.78
1.38
1.93
1.34
1.87
1.23
1.30
0.74
0.91
0.70
0.89
0.55
0.50
0.84
0.61
0.78
0.57
2.56
4.74
9.61
8.31
9.59
7.39
11.32
8.24
20.65
8.34
10.89
8.67
0.65
8.54
4.87
27.84
25.26
30.50
16.47
14.20
11.78
43.12
1.39
7i
7j
7k
7l
7n
Propranolol
Atenolol
a
The incubation time was up to 90 min and all experiments were carried out in triplicate. Data are means ꢄ SD.
Pt [estimated transcellular permeability ¼ ((AB þ BA)/2)] [30].
b

200
C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
benzimidazole derivatives, a multiple linear regression (MLR)
analysis was performed. The results showed moderate correlations
between some of the mentioned descriptors and BA/AB or Pt,
equations (4) and (5).
In the case of 2-(trifluoromethyl)-1H-benzimidazole derivatives
(7ee7k), they showed the correlation presented in equation (11).
IC50 ¼ ꢁ1:16þ0:584logD7:4 þ0:364HBDꢁ0:0039MV
ꢀ
ꢁ
(11)
r2 ¼ 0:98; n ¼ 7; ra2dj ¼ 0:960; s ¼ 0:07; F ¼ 49
BA=AB ¼ 0:879 þ 0:221 logD7:4 ꢁ 0:003 MV
ꢁ 0:09 HBA þ 0:0787 pKa1
The results indicated that for this subgroup, giardicidal activity
ꢀ
ꢁ
depends of lipophilicity, HBD and MV and it is not determined by
Papp parameters.
At this point, it is important to mention that if Clog D7.4 is used
instead of log D7.4, the regression is no significant (p ¼ 0.268),
equation (12). Therefore, the chromatographic method to estimate
r2 ¼ 0:808; n ¼ 19; ra2dj ¼ 0:754; s ¼ 0:219; F ¼ 14
(4)
Pt ¼ 14:6 ꢁ 0:26 MW þ 9:75 HBA ꢁ 12 HBD
þ 2:57 log1=S þ 0:103 MV
r2 ¼ 0:621; n ¼ 19; ra2dj ¼ 0:476; s ¼ 5:083; F ¼ 7
lipophilicity should be better than Clog D7.4
.
ꢀ
ꢁ
(5)
IC50 ¼ 1:48þ0:397ClogD7:4 þ0:564logHBDꢁ0:0092MV
ꢀ
ꢁ
(12)
r2 ¼ 0:687; n ¼ 7; ra2dj ¼ 0:374; s ¼ 0:29; F ¼ 2
Because it was found no important correlations for the others
Papp parameters (AB, BA, AB/BA), we separated the benzimidazole
selected (2b, 2de6c, 7ee7k) in two subgroups. The first included
compounds 2b and 2de6c; the second covered 2-(trifluoromethyl)-
1H-benzimidazole derivatives (7ee7k). Good correlations were
obtained for AB/BA and BA/AB for first subgroup and were
expressed as equations (6) and (7).
4. Conclusions
In this study we intended to understand the relationship
between the physicochemical properties, membrane permeability
and giardicidal activity of a set of benzimidazole derivatives
(2ae7n). In a first stage, it was explored the relationship that could
be presented between giardicidal activity and lipophilicity or
aqueous solubility, estimated from retention factors by RP-HPLC
method, but the results showed no correlation with these proper-
ties. Under the above results, we decided to separate compounds
into two categories according to the substituent at the 2-position:
(2ae6c) and (7ae7n). For the first category a quadratic type of
regressionmodelwasobtainedwith giardicidal activityand aqueous
solubility or lipophilicity. With the second category (7ae7n) no
correlationwas obtained with log D7.4 or log 1/S. In a second stage, in
order to study the membrane permeability influence on giardicidal
activity, twenty one benzimidazole derivatives were selected (2b,
2d, 3ae3c, 4, 5ae5c, 6ae6c, 7ee7l and 7n). MLR analysisshowed not
correlation between the giardicidal activity and Papp, indicating that
it is possible that one needs to combine various physicochemical
properties and Papp parameters to achieve better membrane
permeabilityegiardicidal activity relationship. With this in mind,
we separated benzimidazole derivatives selected into two
subgroups using the same criteria of the categories above: (2b,
2de6c) and (7ee7k). For the first subgroup, influx ratio (AB/BA),
HBA and HBD are features that determine the giardicidal activity of
these compounds. Influx ratio can modify the amount of the
compound to reaches the specific receptor and its time residence in
the gastrointestinal tract, which are critical in giardiasis treatment.
In the case of 2-(trifluoromethyl)-1H-benzimidazole derivatives
(7ee7k), the results indicated that the giardicidal activity depends
of lipophilicity, HBD and MV but not of Papp parameters. The results
of this study showed important interrelated factors in determining
physicochemical properties, membrane permeability and giardici-
dal activity which must be taken in account in order to guide us to
next step in our searching to design better giardicidal benzimidazole
derivatives.
AB=BA ¼ 2:72 ꢁ 1:53 logD7:4 þ 0:218 HBA
ꢁ 0:049 pKa1 þ 0:568 log1=S
ꢀ
ꢁ
r2 ¼ 0:839; n ¼ 12; ra2dj ¼ 0:747; s ¼ 0:09; F ¼ 9
(6)
(7)
BA=AB ¼ 0:365 þ 0:25 logD7:4 ꢁ 0:139 HBA þ 0:09 pKa1
ꢀ
ꢁ
r2 ¼ 0:719; n ¼ 12; ra2dj ¼ 0:614; s ¼ 0:2; F ¼ 6
Strikingly, equations (6) and (7) show that log D7.4, HBA and
pKa1 contribute in opposite effect to AB/BA and BA/AB for these
benzimidazole derivatives. This is consistent with Petrauskas and
co-workers [50] who have proposed that for a compound to be
considered a substrate of P-glycoprotein (BA/AB > 2) depends on
HBA, MW and pKa values.
In contrast, for 7ee7k no significant correlation equation was
found for AB/BA and BA/AB. However, the AB or BA directions for
these derivatives presented the correlations expressed as equations
(8) and (9).
AB ¼ 44:2þ14logD7:4 þ19:5HBDꢁ2:86PSAþ0:385MV
ꢀ
ꢁ
(8)
r2 ¼ 0:964; n ¼ 7; ra2dj ¼ 0:891; s ¼ 4:86; F ¼ 13
BA ¼ ꢁ 160 þ 105 pKa1 ꢁ 1:21 PSA þ 0:216 MV
ꢀ
ꢁ
(9)
r2 ¼ 0:982; n ¼ 7; ra2dj ¼ 0:965; s ¼ 1:79; F ¼ 56
These results propose that PSA and MV are common factors for
AB and BA directions; nevertheless, log D7.4, HBD, and pKa1 influ-
enced one of two directions only.
All these properties could be important descriptors to establish
Papp of 2-(trifluoromethyl)-1H-benzimidazole derivatives.
Finally, we explored the possible correlation between Paap
parameters, physicochemical properties and giardicidal activity for
the two subgroups of benzimidazole derivatives mentioned above.
The giardicidal activity of 2b and 2de6c and its correlation with
that parameters is expressed in the equation (10).
5. Experimental protocols
5.1. Chemistry
IC50 ¼ 0:127þ0:197AB=BAꢁ0:108HBAþ0:161HBD
ꢀ
ꢁ
(10)
r2 ¼ 0:851; n ¼ 12; ra2dj ¼ 0:795; s ¼ 0:028; F ¼ 15
5.1.1. Reagents and instruments
Starting materials 5-Chloro-2-nitroaniline, 3-Hidroxi-4-
nitrobenzoic acid, 2-chloro-N,N-diethylacetamide and control
compounds: Metronidazole, Salicylic acid, Metoprolol, Atenolol
Influx ratio (AB/BA), HBA and HBD are features that determine
the giardicidal activity of these compounds.

C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
201
and Propranolol were commercially available (SigmaeAldrich).
Methanol used during the elution was HPLC grade and Dimethyl
sulfoxide were purchased from Malinckrodt Baker (Mexico). Water
was double distilled and deionized using water purification system
(Millipore, Brazil).
Melting points were determined in open capillary tubes with
a Büchi B-540 melting point apparatus and are uncorrected. Reac-
tions were monitored by TLC on 0.2 mm precoated silica gel 60 F254
plates (E. Merck) with visualization by irradiation with a UV lamp.
Silica gel 60 (70e230 mesh) was used for column chromatography.
IR spectra were recorded with a FT Perkin Elmer 16 PC spectrometer
on KBr disks. 1H NMR and 13C NMR spectra were measured with
a Varian EM-390 (300 MHz) spectrometer. Chemical shifts are given
br, exchangeable with D2O, COOH). 13C NMR (DMSO-d6):
d
31.28
(CH3), 114.30 (C7), 114.10, 117.04, 120.58, 123.28 (CF3), 121.01 (C4),
124.66 (C5), 128.04 (C6), 136.26 (C8), 136.27 (C9), 141.94,142.32,
142.70, 143.68 (C2), 167.66 (C]O). EI-MS m/z (abundance, %): 244
(Mþ, 100). Anal. Calc for C10H7F3N2O2: C, 49.19; H, 2.89; F, 23.34; N,
11.47. Found C, 49.14; H, 2.92; F, 23.31; N, 11.49.
5.1.2.4. N,1-Dimethyl-2-(trifluoromethyl)-1H-benzimidazole-6-carbox-
amide (7n). Recrystallized from Ethyl acetate to yield a white crys-
talline solid (69% yield); mp 218e219 ꢀC; Rf ¼ 0.35 (CHCl3/MeOH
90:10, UV); IR (KBr): 3295, 2937, 1409, 1269, 1183 cmꢁ1 1H NMR
.
(DMSO-d6; 300 MHz): 2.83 (3H, d, J ¼ 4.5 Hz, NeCH3), 3.99 (3H, q,
J ¼ 0.6 Hz, NeCH3), 7.84 (1H, d, J ¼ 0.9 Hz, H), 8.26 (1H, t,
J ¼ 1.05 Hz, H-4), 8.55 (1H, q, J ¼ 4.5 Hz, CONH). 13C NMR (DMSO-
in ppm relative to tetramethylsilane (Me4Si,
d
¼ 0); J values are given
in Hz. Splitting patterns have been designated as follows: s, singlet;
d, doublet; q, quartet; dd, doublet of doublet; t, triplet; m, multiplet;
bs, broad singlet. MS were recorded on a JEOL JMS-SX102A spec-
trometer by electron impact (EI). Elemental analyses (C, H, F, N) for
new compounds were performed on Fisons EA 1108 instrument and
agreed with the proposed structures within ꢄ0.4% of the theoretical
values. Catalytic hydrogenations were carried out in a Parr shaker
hydrogenation apparatus.
d6): d 26.41 (CH3N), 31.28 (CH3), 111.17 (C7), 114.10, 117.11, 120.70,
123.28 (CF3), 120.32 (C4), 122.44 (C5), 131.53 (C6), 135.78 (C8),
142.03 (C9), 141.17, 141.90 (C2), 166.36 (C]O). EI-MS m/z (abun-
dance, %): 257 (Mþ, 75), 227 (Mþ ꢁ 30, 100). Anal. Calc for
C11H10F3N3O: C, 51.37; H, 3.92; F, 22.16; N, 16.34. Found C, 51.35; H,
3.90; F, 22.17; N, 16.36.
5.1.3. Methyl 2-(trifluoromethyl)-1H-benzimidazole-5-carboxylate
(7h)
5.1.2. General procedure for the synthesis of compounds 7ee7g, 7n
The appropriate 1,2-phenylenediamine (9a, 9b, 12, 14,
0.0313 mol), 1.6 eq of CF3CO2H and one drop of concentrated HCl
were heated under reflux in a N2 atmosphere for 24 h. After
completion of the reaction, the cooled mixture was neutralized
with saturated NaHCO3 solution and the crude product was
extracted with ethyl acetate. The combined organic extracts were
washed with water, dried with Na2SO4 and the solvent was
removed under vacuum. The residue was dissolved in chloroform,
filtered through Al2O3 neutral type and concentrated under
vacuum to obtain the corresponding benzimidazole.
A mixture of 7e or 11 (0.008 mol, 1 eq), 1,10-Carbonyldiimidazole
(8.14 g, 0.013 mol, 1.1 eq) and CH2Cl2 (20 mL) was heated under
reflux in a N2 atmosphere for 2 h. Methanol (0.009 mol) in was
added dropwise and the reaction mixture was stirred for 72 h at
room temperature. The solution was then extracted twice with
CH2Cl2. The organic layers were combined, washed with water,
dried over Na2SO4 and evaporated under reduced pressure. The
solid obtained was recrystallized from Toluene to yield a white solid
(48% yield); mp 165e166 ꢀC; Rf ¼ 0.43 (CHCl3/Acetone/Ligroin/
Acetic acid 10:6:15:0.5, UV); IR (KBr): 3240, 1697, 1321, 1290, 1146,
966, 750 cmꢁ1. 1H NMR (DMSO-d6; 300 MHz):
d 3.87 (3H, s, CH3),
7.77 (1H, d, J ¼ 8.7 Hz, H-7), 7.95 (1H, dd, J ¼ 8.4, 1.5 Hz, H-6), 8.33
5.1.2.1. 2-(Trifluoromethyl)-1H-benzimidazole-5-carboxylic
acid
(1H, d, J ¼ 1.5 Hz, H-4). 13C NMR (DMSO-d6):
d 52.21 (CH3O), 113.37,
(7e). Obtained as a white solid (94% yield); mp 280e281 ꢀC;
Rf ¼ 0.26 (CHCl3/Acetone/Ligroin/Acetic acid 10:6:15:0.5, UV); IR
116.96, 120.55, 124.14 (CF3), 116.11 (C7), 119.19 (C4), 124.87 (C5),
125.37 (C6), 138.23 (C8), 140.34 (C9), 141.63,142.15, 142.68, 143,20
(C2), 166.25 (C]O). EI-MS m/z (abundance, %): 244 (Mþ, 50), 213
(Mþ ꢁ 31, 100). Anal. Calc for C10H7F3N2O2: C, 49.19; H, 2.89; F,
23.34; N, 11.47. Found C, 49.17; H, 2.84; F, 23.33; N, 11.45.
(KBr): 3195, 1711, 1669, 1318, 1217, 1192, 1158, 985, 771 cmꢁ1 1H
.
NMR (DMSO-d6; 300 MHz):
(1H, dd, J ¼ 8.4, 1.5 Hz, H-6), 8.30 (1H, d, J ¼ 1.5 Hz, H-4), 12.29 (1H,
br, exchangeable with D2O, COOH). 13C NMR (DMSO-d6):
113.48,
d
7.76 (1H, dd, J ¼ 8.4, 0.3 Hz, H-7), 7.96
d
117.07, 120.65, 124.24 (CF3), 116.10 (C7), 119.20 (C4), 125.19 (C5),
126.72 (C6), 138.13 (C9), 140.36 (C8), 141.46,141.98, 142.51, 143,03
(C2), 167.43 (C]O). EI-MS m/z (abundance, %): 230 (Mþ, 100). Anal.
Calc for C9H5F3N2O2: C, 46.97; H, 2.19; F, 24.77; N, 12.17. Found C,
46.96; H, 2.17; F, 24.78; N, 12.18.
5.1.4. General procedure for the synthesis of compounds 7i and 7j
Into a stirred mixture of compound 7f or 7g (0.0243 mol) in
Dimethylformamide (4 mL) was added a suspension of Dime-
thylsulfate (1.1 eq) and NaHCO3 (2.2 eq) at room temperature. The
resultingsolutionwasstirredfor1hat32ꢀC. Themixturewaspoured
into ice bath, worked up by addition of cold water. The obtained
precipitate was filtered to obtain the corresponding methyl ester.
5.1.2.2. 1-Methyl-2-(trifluoromethyl)-1H-benzimidazole-5-carboxylic
acid (7f). Recrystallized from Nitromethane to yield a white solid
(60% yield); mp 152e254 ꢀC; Rf ¼ 0.53 (CHCl3/MeOH 90:10, UV); IR
(KBr): 2829, 2566, 1406, 1685, 1267, 1141 cmꢁ1. 1H NMR (DMSO-d6;
5.1.4.1. Methyl
1-methyl-2-(trifluoromethyl)-1H-benzimidazole-5-
carboxylate (7i). Recrystallized from Dimethyformamide/H2O to
yield a white crystalline solid (82% yield); mp 121e122 ꢀC; Rf ¼ 0.40
(Hexane/Ethyl acetate 70:30, UV); IR (KBr): 3018, 2962, 1750, 1726,
300 MHz):
d
3.98 (3H, s, NeCH3), 7.82 (1H, dd, J ¼ 8.7, 0.3 Hz, H-7),
8.02 (1H, dd, J ¼ 8.7, 1.5 Hz, H-6), 8.34 (1H, d, J ¼ 0.6 Hz, H-4). 13C
NMR (DMSO-d6):
d
31.29 (CH3), 111.75 (C7), 113.48, 117.04, 120.64,
1306, 1131 cmꢁ1 1H NMR (DMSO-d6; 300 MHz):
. d 3.88 (3H, s,
124.24 (CF3), 122.61 (C4), 125.94 (C5), 126.36 (C6), 138.96 (C8),
139.91 (C9), 141.46,140.98, 141.79, 143.03 (C2), 167.36 (C]O). EI-MS
m/z (abundance, %): 244 (Mþ, 100).
OeCH3), 3.99 (3H, s, NeCH3), 7.88 (1H, d, J ¼ 8.7 Hz, H-7), 8.04 (1H,
dd, J ¼ 8.7,1.5 Hz, H-6), 8.36 (1H, d, J ¼ 1.2 Hz, H-4). 13C NMR (DMSO-
d6):
d 31.67 (CH3), 52.61 (CH3O), 112.38 (C7), 114.87, 117.82, 120.52,
124.14 (CF3),122.90 (C4),125.50 (C5),126.01 (C6),139.53 (C8),140.26
(C9), 141.93, 142.31 (C2), 166.60 (C]O). EI-MS m/z (abundance, %):
258 (Mþ, 65), 227 (Mþ ꢁ 31,100). Anal. Calc for C11H9F3N2O2: C, 51.17;
H, 3.51; F, 22.07; N, 10.85. Found C, 51.19; H, 3.48; F, 22.09; N, 10.81.
5.1.2.3. 1-Methyl 2-(trifluoromethyl)-1H-benzimidazole-6-carboxylic
acid (7g). Recrystallized from Isopropanol to yield a white solid
(61% yield); mp 270e272 ꢀC; Rf ¼ 0.53 (CHCl3/MeOH 90:10, UV); IR
(KBr): 2963, 1679, 1410, 1260, 1142 cmꢁ1 1H NMR (DMSO-d6;
.
300 MHz):
d
4.00 (3H, s, NeCH3), 7.82 (1H, d, J ¼ 8.7 Hz, H-4), 7.92
5.1.4.2. Methyl 1-methyl-2-(trifluoromethyl)-1H-benzimidazole-6-
carboxylate (7j). Recrystallized from Dimethyformamide/H2O to
(1H, dd, J ¼ 8.4, 1.5 Hz, H-5), 8.33 (1H, d, J ¼ 0.6 Hz, H-7), 13.10 (1H,

202
C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
yield a white solid (84% yield); mp 107e108 ꢀC; Rf ¼ 0.40 (Hexane/
(3H, d, J ¼ 4.5 Hz, NeCH3), 7.74 (1H, d, J ¼ 8.4 Hz, H-7), 7.87 (1H, d,
J ¼ 8.4 Hz, H-6), 8.21 (1H, J ¼ 0.3 Hz, H-6), 8.54 (1H, q, J ¼ 4.5 Hz,
Ethyl acetate 70:30, UV); IR (KBr): 3036, 2961, 1748, 1710, 1483,
1268, 1126 cmꢁ1
.
1H NMR (DMSO-d6; 300 MHz):
d
3.90 (3H, s,
CONH). 13C NMR (DMSO-d6):
d 26.42 (CH3), 108.42, 115.55, 122.69,
OeCH3), 4.02 (3H, s, NeCH3), 7.88 (1H, dd, J ¼ 8.7, 0.6 Hz, H-4), 7.94
129.62 (CF3), 113.13 (C7), 118.85 (C4), 122.38 (C5), 124.01 (C6),
135.74 (C8), 141.84, 142.81, 143.79, 144.76 (C2), 143.55 (C9),
165.47(C]O). EI-MS m/z (abundance, %): 243 (Mþ, 45), 213
(Mþ ꢁ 30, 100). Anal. Calc for C10H8F3N3O: C, 49.39; H, 3.32; F,
23.44; N, 17.28. Found C, 49.36; H, 3.37; F, 23.42; N, 17.25.
(1H, dd, J ¼ 8.7, 1.5 Hz, H-5), 8.38 (1H, dd, J ¼ 1.5, 0.6 Hz, H-7). 13C
NMR (DMSO-d6):
d 31.66 (CH3), 52.73 (CH3O), 114.24 (C7), 115.12,
117.82, 120.53, 123.23 (CF3), 121.21 (C4), 124.37 (C6), 126.74 (C5),
136.23 (C8), 142.16, 142.54, 142.94 (C2), 143.85 (C9), 166.57 (C]O).
EI-MS m/z (abundance, %): 258 (Mþ, 100). Anal. Calc for
C11H9F3N2O2: C, 51.17; H, 3.51; F, 22.07; N, 10.85. Found C, 51.16; H,
3.52; F, 22.09; N, 10.86.
5.1.6.1. 1-Methyl-2-(methylthio)-1H-benzimidazole (6a). 13C NMR
(CDCl3):
d 14.47 (CH3eS), 29.64 (CH3eN), 108.13 (C7), 117.80 (C4),
121.58 (C5), 121.61 (C6), 136.63 (C8), 143.02 (C9), 152.94 (C2).
5.1.5. General procedure for the synthesis of compounds 7k and 7l
A mixture of corresponding carboxylic acid 7f or 7g (0.006 mol)
Triethylamine (0.013 mol, 2.2 eq), 2-Chloro-N,N-diethylacetamide
(0.0068 mol, 1.1 eq), Potassium iodide (0.0006 mol, 0.1 eq) and
Dimethylformamide (8 mL) was stirred for 24 h at room tempera-
ture. The mixture poured into ice bath to obtain the N,N-diethyl
glycolamide ester corresponding.
5.1.6.2. 5-Chloro-1-methyl-2-(methylthio)-1H-benzimidazole (6b).
13C NMR (CDCl3):
d 14.53 (CH3eS), 29.98 (CH3eN), 108.87 (C7),
117.60 (C4), 121.97 (C6), 127.38 (C8), 135.23 (C9), 143.49 (C5), 154.62
(C2).
5.1.6.3. 6-Chloro-1-methyl-2-(methylthio)-1H-benzimidazole (6c).
13C NMR (CDCl3):
d 14.67 (CH3eS), 30.07 (CH3eN), 108.62 (C7),
5.1.5.1. 2-(Diethylamino)-2-oxoethyl 1-methyl-2-(trifluoromethyl)-
1H-benzimidazole-5-carboxylate (7k). Recrystallized from Acetoni-
trile/H2O to yield a white solid (76% yield); mp 97e98 ꢀC; Rf ¼ 0.39
(CHCl3/MeOH 90:10, UV); IR (KBr): 2967, 2934, 1726, 1665, 1471,
118.41 (C4), 122.54 (C5), 127.82 (C8), 137.06 (C9), 140.06 (C6), 154.19
(C2).
5.1.6.4. 1-Methyl-1H-benzimidazole-2-thiol (5a). 13C NMR (CDCl3):
d 30.10 (CH3), 108.62 (C7), 110.05 (C4), 122.80 (C5), 122.85 (C6),
1179, 1130 cmꢁ1
.
1H NMR (DMSO-d6; 300 MHz):
d
1.02 (3H, t,
J ¼ 6.6 Hz, CH3), 1.67 (3H, t, J ¼ 6.9 Hz, CH3), 3.31 (4H, m, J ¼ 7.2 Hz,
CH2), 4.01 (3H, s, NeCH3), 5.04 (2H, s, COCH2O), 7.92 (1H, d,
J ¼ 8.7 Hz, H-7), 8.08 (1H, dd, J ¼ 8.7, 1.5 Hz, H-6), 8.41 (1H, d,
130.72.68 (C8), 133.06 (C9), 1676.52 (C2).
5.1.6.5. 5-Chloro-1-methyl-1H-benzimidazole-2-thiol (5b). 13C NMR
(CDCl3): d 30.03 (CH3), 108.96 (C7), 109.84 (C4), 122.03 (C6), 128.21
(C8), 131.36 (C9), 131.70 (C5), 169.34 (C2).
J ¼ 0.6 Hz, H-4). 13C NMR (DMSO-d6):
d
13.35, 14.41 (CH3CH2), 31.78
(CH3), 62.44 (CH2O), 112.56 (C7), 114.12, 117.84, 120.54, 123.24 (CF3),
123.14 (C4), 125.47 (C5), 126.24 (C6), 139.72 (C8), 140.30 (C9),
142.07, 142.45 (C2), 165.48, 166.82 (C]O). EI-MS m/z (abundance,
%): 357 (Mþ not appreciable), 227 (Mþ ꢁ 30, 100). Anal. Calc for
C16H18F3N3O3: C, 53.78; H, 5.08; F, 15.95; N,11.76. Found C, 53.80; H,
5.06; F, 15.96; N, 11.74.
5.1.6.6. 6-Chloro-1-methyl-1H-benzimidazole-2-thiol (5c). 13C NMR
(DMSO-d6):
d 30.18 (CH3), 109.51 (C7), 110.47 (C4), 122.57 (C5),
126.73 (C8), 129.56 (C9), 134.10 (C6), 169.62 (C2).
5.1.6.7. 5-Chloro-1-methyl-1H-benzimidazol-2-amine (4). 13C NMR
5.1.5.2. 2-(Diethylamino)-2-oxoethyl 1-methyl-2-(trifluoromethyl)-
1H-benzimidazole-6-carboxylate (7l). Recrystallized from Dime-
(DMSO-d6): d 28.55 (CH3), 108.26 (C7), 114.05 (C4), 117.56 (C6),
124.68 (C8), 133.85 (C9), 143.91 (C5), 156.58 (C2).
thyformamide/H2O to yield
a white solid (83% yield); mp
115e116 ꢀC; Rf ¼ 0.62 (CHCl3/MeOH 90:10, UV); IR (KBr): 2980,
5.1.6.8. 1,2-Dimethyl-1H-benzimidazole (3a). 13C NMR (CDCl3):
d 9.23 (CH3), 25.33 (CH3), 104 (C7), 115.11 (C4), 118.90 (C5, C6),
2936, 1722, 1648, 1482, 1266, 1122 cmꢁ1
.
1H NMR (DMSO-d6;
300 MHz): 1.025 (3H, t, J ¼ 6.9 Hz, CH3), 1.171 (3H, t, J ¼ 6.9 Hz, CH3),
3.28 (2H, q, J ¼ 6.9 Hz, CH2), 3.32 (2H, q, J ¼ 6.9 Hz, CH2), 4.03 (3H, q,
J ¼ 0.9 Hz, NeCH3), 5.06 (2H, s, COCH2O), 7.92 (1H, dd, J ¼ 8.4,
0.6 Hz, H-6), 7.98 (1H, dd, J ¼ 8.4, 1.5 Hz, H-7), 8.42 (1H, dd, J ¼ 1.5,
132.12 (C8), 148.01 (C2).
5.1.6.9. 5-Chloro-1,2-dimethyl-1H-benzimidazole (3b). 13C NMR
(CDCl3):
d 13.75 (CH3), 29.92 (CH3eN), 109.03 (C7), 119.66 (C4),
0.6 Hz, H-4). 13C NMR (DMSO-d6):
d
13.35, 14.42 (CH3CH2), 31.74
122.34 (C5), 127.69 (C8), 136.10 (C6), 140.97 (C9), 152.65 (C2).
(CH3), 62.51 (CH2O), 114.44 (C7), 115.14, 117.84, 120.55, 123.25 (CF3),
121.27 (C4), 124.58 (C5), 126.70 (C6), 136.27 (C8), 142.25, 142.63,
143.01 (C2), 143.99 (C9), 165.39, 165.78 (C]O). EI-MS m/z (abun-
dance, %): 357 (Mþ not appreciable), 227 (Mþ ꢁ 30, 100). Anal. Calc
for C16H18F3N3O3: C, 53.78; H, 5.08; F, 15.95; N, 11.76. Found C,
53.81; H, 5.05; F, 15.94; N, 11.78.
5.1.6.10. 5,6-Dichloro-1,2-dimethyl-1H-benzimidazole (3c). 13C NMR
(CDCl3, DMSO-d6):d11.87.75(CH3),28.36(CH3eN),109.38(C7),117.42
(C4), 122.40 (C5), 12.64 (C6), 133.53 (C8), 139.96 (C9), 152.87 (C2).
5.2. Lipophilicity and aqueous solubility determination
5.1.6. N-Methyl-2-(trifluoromethyl)-1H-benzimidazole-5-
carboxamide (7m)
A Perkin Elmer liquid chromatograph (Series 200) with refrac-
tive index detector, Peltier column oven and Isocratic Pump Model
1310 were used for lipophilicity and solubility determinations. RP-
HPLC system was controlled by Total Chrom Workstation (v.6.2.1)
software. Chromatographic retention times were determined on
A mixture of 7e (0.0043 mol) Thionyl chloride (0.254 mol) was
heated under reflux conditions for 40 min. After distillation of
excess Thionyl chloride and treating the mixture with Benzene, the
oil obtained was added dropwise Methylamine hydrochloride
(0.005 mol) and Triethylamine (5 mL). The mixture was stirred by
2 h and worked up by addition of cold water, filtered by suction and
the crude product was treated with Nitromethane to obtain a white
solid (69% yield); mp 258e260 ꢀC; Rf ¼ 0.20 (CHCl3/Acetone/
Ligroin/Acetic acid 10:6:15:0.5, UV); IR (KBr): 3306,2915, 1619,
C18 (AtlantiseWater); 3.9 mm ꢃ 150 mm, 5
mm column at flow rate
of 1 mL/min, 36 ꢀC and at isocratic elution in MeOH/H2O. Thia-
bendazole, Metoprolol and compounds 1e2d were used as control
compounds. Solutions of control and compounds 3ae7n were
prepared in MeOH/H2O (9:1) and 0.3% of glacial acetic acid. The
compounds were then injected at isocratic elution 90, 80, 75, 70
and 60% MeOH. The log kw was calculated for each compound from
1550, 1319, 1196, 1155 cmꢁ1 1H NMR (DMSO-d6; 300 MHz): 2.81
.

C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
203
three injections of 20
by injecting DMSO together with the sample [26].
m
L sample. The dead time (t0) was considered
statistics (F) were provided for significance
was utilized as evaluation criteria of the regression models.
a
¼ 0.05; Value of r2-adj
5.3. Membrane permeability
5.6. Calculated molecular descriptors
5.3.1. Preparation of Caco-2 monolayer
Calculated descriptors such as Clog P, Clog D7.4, PSA, MV, MW
and pKa were determined with ACD Labs v.9Ô software.
Caco-2 cells were donated by Dr. Ismael Hidalgo from Absorp-
tion System. Cells were maintained at 37 ꢀC in DMEM/high glucose
supplemented with 10% FBS, 1%, Glutamine, 1% NEAA, 1% Sodium
pyruvate and 100 U/mL Penicillin and 100 mg/mL Streptomycin in
atmosphere of 5% CO2 and 90% relative humidity. Cells were grown
and routinely seeded in 75 cm2-flasks and harvested by regular
trypsinization. For transport experiments, 60,000 cells/cm2 were
seeded onto each 12-mm collagen-coated (rat tail collagen type I).
Transwell polycarbonate filters (area of 1.13 cm2 and mean pore size
Role of the funding source
This study was supported by grants from PAPIIT-UNAM
IN210809 and CONACyT 80093.
Acknowledgements
of 0.4 mm). The medium was changed every 48 h during 6 days and
We are grateful to Rosa Isela del Villar, Nuria Esturau, Georgina
Duarte, Margarita Guzmán, Nayeli López and Marisela Gutiérrez
from Facultad de Química, USAI, UNAM, for the determination of all
spectra.
every 24 h thereafter. The monolayers were used between passages
58 and 65, after days 28 days of growth. Before starting the
experiment, monolayer integrity was verified by transepithelial
electrical resistance (TEER) measurements showing values
between 250 and 300
U
cm2, indicating confluent monolayers with
well established tight junctions.
References
[1] R.C.A. Thompson, Int. J. Parasitol. 30 (2000) 1259e1267.
[2] L. Savioli, H. Smith, A. Thompson, Trends Parasitol. 22 (2006) 203e208.
[3] L.J. Robertson, K. Hanevik, A.A. Escobedo, K. Mørch, N. Langeland, Trends
Parasitol. 26 (2010) 75e82.
5.3.2. Bidirectional transport studies
Atenolol, Propranolol and benzimidazole derivatives were dis-
solved in EtOH and diluted with water (10 mM, stock solution). Test
[4] G. Navarrete-Vázquez, R. Cedillo, A. Hernández-Campos, Lilián Yépez,
F. Hernández-Luis, J. Valdez, R. Morales, R. Cortés, M. Hernández, R. Castillo,
Bioorg. Med. Chem. Lett. 11 (2001) 187e190.
[5] J. Valdez, R. Cedillo, A. Hernández-Campos, L. Yépez, F. Hernández-Luis,
G. Navarrete-Vázquez, A. Tapia, R. Cortés, M. Hernández, R. Castillo, Bioorg.
Med. Chem. Lett. 12 (2002) 2221e2224.
[6] G. Navarrete-Vázquez, M.M. Rojano-Vilchis, L. Yépez-Mulia, V. Meléndez,
L. Gerena, A. Hernández-Campos, R. Castillo, F. Hernández-Luis, Eur. J. Med.
Chem. 41 (2006) 135e141.
[7] D. Valdez-Padilla, S. Rodríguez-Morales, A. Hernández-Campos, F. Hernández-
Luis, L. Yépez-Mulia, A. Tapia-Contreras, R. Castillo, Bioorg. Med. Chem. 17
(2009) 1724e1730.
[8] F. Hernández-Luis, A. Hernández-Campos, R. Castillo, G. Navarrete-Vázquez,
O. Soria-Arteche, M. Hernández-Hernández, L. Yépez-Mulia, Eur. J. Med. Chem.
45 (2010) 3135e3141.
[9] S. Bhattachar, L. Deschenes, J. Wesley, Drug Discov. Today 11 (2006)
1012e1018.
solutions 100
mM of each compound were prepared from stock
solution dissolved in the transport medium (Hank’s Balanced salt
solution, HBSS at pH 7.4). Transport medium containing the
compound was added either to the A or B compartment. The cells
monolayers were incubated at 37 ꢀC with 500
and 1500
each insert was changed at a new well with fresh transport
medium. A 1000 L of medium from the B side was removed at 15,
30, 45, 60 and 90 min, whereas for BA transport, 200 L of medium
from the A side was removed. The receiver medium was totally
replaced with fresh medium. Samples (100 L) were injected
mL in A compartment
mL in the B compartment. For AB transport experiments,
m
m
m
directly into a high performance liquid chromatograph system
equipped with a variable wavelength detector at 217 nm using an
[10] C. Hansch, A. Leo, S. Mekapati, A. Kurup, Bioorg. Med. Chem. 12 (2004)
3391e3400.
AtlantisÔ C18 column (250 mm ꢃ 4.6 mm; particle size 5
mm). Flow
was maintained at 1.0 mL/min. The monolayer integrity was eval-
uated both at the beginning and the end of the experiment by
monitoring TEER values and the determination of the permeability
of Lucifer Yellow. Experiments were carried out in triplicate.
[11] R.S. DeWitte, Drug Discov. Today 11 (2006) 855e859.
[12] M.J. Waring, Bioorg. Med. Chem. Lett. 19 (2009) 2844e2851.
[13] M. Mottier, L. Alvarez, M. Pis, C. Lanusse, Exp. Parasitol. 103 (2003) 1e7.
[14] A. Mälkiä, L. Murtomäki, A. Urtti, K. Kontturi, Eur. J. Pharm. Sci. 23 (2004)
13e47.
[15] F. Bertha, G. Hornyak, K. Zauer, A. Feller, K. Lempert, E. Pjeczka, G. Toth,
Tetrahedron 41 (1985) 2855e2860.
[16] Wei, H.L. Peter, U.S. Patent 4,520,196, 1985.
5.4. Activity assays against G. duodenalis
[17] T. Sato, S. Akimoto, S. Ponda, Japanese Patent 95e29028, 1996.
[18] S. Wang, Y.T. Chang, J. Am. Chem. Soc. 128 (2006) 10380e10381.
[19] V.K. Agrawal, M. Gupta, J. Singha, P.V. Khadikar, Bioorg. Med. Chem. 13 (2005)
2109e2120.
[20] M. Kah, C.D. Brown, Chemosphere 72 (2008) 1401e1408.
[21] P.R. Duchowicz, A. Talevi, L.E. Bruno-Blanch, E.A. Castro, Bioorg. Med. Chem.
16 (2008) 7944e7955.
[22] L. Zhou, L. Yang, S. Tilton, J. Wang, J. Pharm. Sci. 96 (2007) 3052e3071.
[23] Z. Hernández-Gallegos, P.A. Lehmann, J. Pharm. Sci. 79 (1990) 1032e1033.
[24] S. Poole, C. Poole, J. Chromatogr. B 797 (2003) 3e19.
[25] T. Hartmann, J. Schmitt, Drug Discov. Today Technol. 1 (2004) 431e439.
[26] A. Beerthod, S. Carda-Broch, J. Chromatogr. A 1037 (2004) 3e14.
[27] L. Escuder-Gilabert, J.M. Bermúdez-Saldaña, R.M. Villanueva-Camañas,
M.J. Medina-Hernández, S. Sagrado, J. Chromatogr. A 1033 (2004) 247e255.
[28] X. Liu, H. Tanaka, A. Yamauchi, B. Testa, H. Chuman, J. Chromatogr. A 1091
(2005) 51e59.
G. duodenalis strain IMSS: 0989; was cultured in TYI-S-33
modified medium, supplemented with 10% calf serum and bovine
bile. In vitro susceptibility assays were performed using a method
earlier described [4]. Briefly, 5 ꢃ 104 trophozoites of G. duodenalis
were incubated for 48 h at 37 ꢀC with increasing concentrations of
experimental compounds. As the negative control, trophozoites
were incubated with DMSO used in the experiments. After the
incubation, an aliquot (5
mL) of the treated trophozoites were
subcultured for another 48 h in fresh medium alone. At the end of
this period, trophozoites were counted and the 50% inhibitory
concentration (IC50) was calculated by Probit analysis. Experiments
were carried out in triplicate and repeated at least twice.
[29] D. Vrakras, C. Gianginis, A. Tsantili-Kakoulidou, J. Chromatogr. A 1116 (2006)
158e164.
[30] K. Valkó, J. Chromatogr. A 1037 (2004) 299e310.
5.5. Statistic analysis
[31] M.M. Hsieh, J.G. Dorsey, Anal. Chem. 67 (1995) 48e57.
[32] W.J. Lambert, J. Chromatogr. A 656 (1993) 469e484.
[33] J.G. Dorsey, K.A. Dill, Chem. Rev. 89 (1989) 331e346.
[34] A. Avdeef, Absorption and Drug Development. Solubility, Permeability, and
Charge State, John Wiley & Sons, New Jersey, 2003, p. 346.
The regression analysis was carried out using Minitab 14Ô
software package. Square correlation coefficient (r2), adjusted
regression coefficient (r2-adj), standard deviation (s) and Fischer

204
C. Hernández-Covarrubias et al. / European Journal of Medicinal Chemistry 52 (2012) 193e204
[35] http://www.chemspider.com/Chemical-Structure.5593.html?rid¼1d7811f9-
998a-40ec-9206-968442009913, (accessed 27 08 11).
[43] R.T. Borchardt, in: B. Testa, H. van de Waterbeemd, G. Folkers, R. Guy (Eds.),
Pharmacokinetic Optimization in Drug Research, Wiley-VCH, Germany, 2001,
pp. 117e126.
[44] Q. Wang, R. Strab, P. Kardos, C. Ferguson, J. Li, A. Owen, I.J. Hidalgo, Int. J.
Pharm. 356 (2008) 12e18.
[45] L. Zhang, P.V. Balimane, S.R. Johnson, S. Chong, Int. J. Pharm. 343 (2007)
98e105.
[46] I. Behrens, W. Kamm, A.H. Dantzig, T. Kissel, J. Pharm. Sci. 93 (2004)
1743e1747.
[47] M.I. Kaldas, U.K. Walle, T. Walle, J. Pharm. Pharmacol. 55 (2003) 307e312.
[48] T. Fichert, M. Yazdanianb, J.R. Proudfoot, Bioorg. Med. Chem. Lett. 13 (2003)
719e722.
[49] G. Camenisch, J. Alsenz, H. van de Waterbeemd, G. Folkers, Eur. J. Pharm. Sci. 6
(1998) 313e319.
[36] http://www.chemspider.com/Chemical-Structure.23741.html?rid¼47d0c940-
7971-4920-bb5e-be9bd2b57b93, (accessed 12 09 11).
[37] K. Daniel-Mwambete, S. Torrado, C. Cuesta-Bandera, F. Ponce-Gordo,
J. Torrado, Int. J. Pharm. 272 (2004) 29e36.
[38] http://www.fda.gov/downloads/AnimalVeterinary/DevelopmentApproval
Process/EnvironmentalAssessments/UCM072159.pdf, (accessed 20 08 10).
[39] http://www.chemspider.com/Chemical-Structure.5237.html?rid¼96fdf738-
c6e8-4939-b71d-4537422793c1, (accessed 20 09 11).
[40] S. Yalkowsky, S. Banerjee, Aqueous Solubility. Methods of Estimation for
Organic Compounds, Marcel Dekker, Inc., USA, 1992, pp. 51e54.
[41] Y. Ran, N. Jain, S.H. Yalkowsky, J. Chem. Inf. Comput. Sci. 41 (2001)
1208e1217.
[42] T. Sanghvi, N. Jain, G. Yang, S.H. Yalkowsky, QSAR Comb. Sci. 22 (2003)
258e263.
[50] R. Didziapetris, P. Japertas, A. Avdeef, A. Petrauskas, J. Drug Targeting 11
(2003) 391e406.