Quantitative structure-activity relationships of mosquito larvicidal chalcone derivatives

Article abstract of DOI:10.1021/jf203374r

The mosquito larvicidal activities of a series of chalcones and some derivatives were subjected to a quantitative structure-activity relationship (QSAR) study, using more than a thousand constitutional, topological, geometrical, and electronic molecular descriptors calculated with Dragon software. The larvicidal activity values for 28 active compounds of the series were predicted, showing in general a good approximation to the experimental values found in the literature. Chalcones having one or both electron-rich rings showed high toxicity. However, the activity of chalcones was reduced by electron-withdrawing groups, and this was roughly diminished by derivatization of the carbonyl group. A set of six chalcones being structurally similar to some of the active ones, with a still unknown larvicidal activity, were prepared. Their activity values were predicted by applying the developed QSAR models, showing that two chalcones of such set, both 32 and 34, were expected to be highly active.

Full text of DOI:10.1021/jf203374r

Article
pubs.acs.org/JAFC
Quantitative Structure−Activity Relationships of Mosquito Larvicidal
Chalcone Derivatives
†
†,‡
†
§
⊥
Gustavo Pasquale, Gustavo P. Romanelli, Juan C. Autino, Javier García, Erlinda V. Ortiz,
,
§
and Pablo R. Duchowicz*
†
Cat
́ ́
edra de Química Organica, Facultad de Ciencias Agrarias y Forestales, Universidad Nacional de La Plata (UNLP), 60 y 119
(
B1904AAN) La Plata, Buenos Aires, Argentina
‡
Centro de Investigacion
́
y Desarrollo en Ciencias Aplicadas “Dr. Jorge J. Ronco” (CINDECA), CONICET-CCT La Plata,
Universidad Nacional de La Plata (UNLP), 47 No. 257, (B1900AJK), La Plata, Argentina
§
Instituto de Investigaciones Fisicoquímicas Teor
́
icas y Aplicadas INIFTA (UNLP, CCT La Plata-CONICET), Diag. 113 y 64,
Sucursal 4, C.C. 16, 1900 La Plata, Argentina
⊥
Facultad de Tecnología y Ciencias Aplicadas, Universidad Nacional de Catamarca, Av. Maximio Victoria 55, (4700) Catamarca,
Argentina
*
S Supporting Information
ABSTRACT: The mosquito larvicidal activities of a series of chalcones and some derivatives were subjected to a quantitative
structure−activity relationship (QSAR) study, using more than a thousand constitutional, topological, geometrical, and electronic
molecular descriptors calculated with Dragon software. The larvicidal activity values for 28 active compounds of the series were
predicted, showing in general a good approximation to the experimental values found in the literature. Chalcones having one or
both electron-rich rings showed high toxicity. However, the activity of chalcones was reduced by electron-withdrawing groups,
and this was roughly diminished by derivatization of the carbonyl group. A set of six chalcones being structurally similar to some
of the active ones, with a still unknown larvicidal activity, were prepared. Their activity values were predicted by applying the
developed QSAR models, showing that two chalcones of such set, both 32 and 34, were expected to be highly active.
KEYWORDS: chalcone derivatives, mosquito larvicidal activity, QSAR theory, molecular descriptors
4
1
. INTRODUCTION
methodologies include the use of dry HCl, a Lewis acid such as
5
6
7
TiCl , p-toluenesulfonic acid, and more recently BF −Et O.
Chalcones or 1,3-diaryl-2-propen-1-ones are a family of aromatic
ketones with two aromatic groups bridged by an enone linkage
4
3
2
Chalcones have attracted increasing attention due to numerous
pharmacological applications. They are the main precursors for the
biosynthesis of flavonoids, which are common pharmacologically
1
(
ArCOCHCHAr′). The numbering of chalcones and the
designation of the aryl ring as A and B are indicated for the
8
active components of the human diet. For example,
licochalcone A isolated from the roots of Glycyrrhiza inflata
licorice) has in vitro and in vivo antimalarial and anti-
leishmanial activities, and 3-methoxy-4-hydroxyloncocarpin
unsubstituted compounds 1 (Figure 1).
9
(
10
isolated from the roots of Lonchocarpus utiliz inhibits
11
NADH:ubiquinone oxidoreductase activity (Figure 2).
Compounds with the backbone of chalcones, which possess
12
various biological activities such as antimicrobial, anti-inflamma-
Figure 1. Chalcone structure.
13
14
15
16
17
tory, antiplatelet, antimalarial, anticancer, antileishmanial,
18
Chalcone is commonly synthesized via the Claisen−Schmidt
aldol condensation between acetophenone and benzaldehyde
antioxidant, antifungal, and inhibition of leukotriene B4, have
been reported. Chalcones also find several application in
1
9
(
Scheme 1). This reaction is catalyzed by bases and acids under
agriculture; for example, chalcones exhibit allelopathic activity,
20
and many chalcones are used as insecticides and fungicides; in
18,21
this way chlorochalcones are potent insect antifeedants.
The well-known basis of the quantitative structure−activity
Scheme 1. Chalcone Synthesis
22−25
relationships (QSAR) theory
is the hypothesis that the
biological activity exhibited by a chemical compound is mainly
Received: August 22, 2011
Revised: December 7, 2011
Accepted: December 7, 2011
Published: January 3, 2012
homogeneous conditions. The condensation reaction in basic
medium is usually carried out in the presence of sodium hydroxide,
potassium hydroxide, or alkali alcoholate, and the acid-catalyzed
2,3
©
2012 American Chemical Society
692
dx.doi.org/10.1021/jf203374r | J. Agric.Food Chem. 2012, 60, 692−697

Journal of Agricultural and Food Chemistry
Article
moment, and number of electrons. The total number of calculated
descriptors was 1502.
3
3
We resorted to the replacement method (RM) as the molecular
descriptor selection approach, an algorithm that was proposed by our
research group some years ago for generating multivariable linear
regression QSAR models with minimized standard deviation (S). The
quality of the results achieved with this technique is quite close to that
obtained by performing an exact (combinatorial) full search (FS) of
molecular descriptors, although, of course, it requires much less
3
4
computational work. In addition, the RM provides models with
statistical parameters better than those from the forward stepwise
regression procedure and similar to those from the more elaborated
Figure 2. Licochalcone A and 3-methoxy-4-hydroxyloncocarpin.
3
5,36
genetic algorithms approach.
All of the models presented here were properly validated by means
determined by its molecular structure. This fact does not offer
specific details on the usually complex mechanism/path of the
process that leads to the final biological effect. However, it is
possible to get some insight into the underlying mechanism of
action by means of the QSAR-based predicted activities.
Although chalcones have been widely studied in the past, there
are not many QSAR studies on the insecticide activity of these
compounds. Only a recent publication of Begum and co-
37
of the leave-one-out (loo) cross-validation technique and more
strictly by using an external test set of selected compounds (test). We
3
8
also practiced the Y-randomization technique, which consists of
scrambling the experimental activity values in such a way that they no
longer correspond to the respective compounds. The smallest standard
Rand
deviation S obtained after analyzing 1000 cases of Y randomization
for each developed QSAR turned out to be poorer (greater) than the
one found in the true calibration (S). This result supports the
assumption that the correlations derived here are not fortuitous but
the result of actual structure−activity relationships.
2
6
workers described a QSAR for mosquito larvicidal activity of
some chalcone derivatives.
In this work, we studied the larvicidal activity of some
chalcone-type compounds with various substitution patterns
along with some of their derived products against the larvae of
Culex quinquefasciatus. We collected experimental information
To determine the relative importance of each descriptor in the
linear regression model, we calculated standardized regression
3
9
coefficients as
sj × b
j
26
s
from the literature and established useful QSARs, with the
main purpose of applying this structure−activity relationship to
a set of six chalcone derivatives with unknown experimental
b =
j = 1, ..., d
j
s
(1)
Y
where b is the regression coefficient for the descriptor j and s and s
Y
j
j
27
larvicidal activities recently synthesized in our laboratory. We
employed multiparametric linear regression models, as this is
considered one of the most popular statistical techniques, and
are the standard deviations for that descriptor and for the experimental
s
j
activity, respectively. The larger the value of b , the greater is the
importance of the descriptor j.
28−30
explored more than a thousand theoretical descriptors.
3
. RESULTS AND DISCUSSION
2
. MATERIALS AND METHODS
The set of six chalcone derivatives 29−34 was prepared by the
Claisen−Schmidt condensation reaction of benzaldehydes and
acethophenone under solvent-free conditions catalyzed by amino-
2.1. Materials and Reagents. All solvents and chemicals were
commercially available and used without further purification unless
otherwise stated. Catalyst was prepared according to the literature.
.2. General Procedure for Chalcone Synthesis. Chalcone
synthesis was performed in a sealed tube in solvent-free conditions.
A mixture of the corresponding aldehyde (1.3 mmol), acetophenone
2
7
27
propylated silica sol−gel materials according to the literature.
2
The reactivity of different benzaldehydes and methyl aryl ketones
was tested under the same conditions (140 °C, benzaldehyde/
acetophenone ratio 1.3:1, catalyst 100 mg, and 240 min). Results
of the obtained yields are listed in Table 1. The results showed
(
1 mmol), and catalyst (100 mg) was warmed at 140 °C for 4 h. The
reaction mixture was diluted with hot toluene (10 mL), the catalyst
was filtered off, and then the solution was washed and dried with
anhydrous Na SO , the solvent was evaporated, and the residue was
2
4
Table 1. Chalcone Synthesis by Claisen−Schmidt Procedure
purified by column chromatography (silica gel, ethyl acetate/hexanes) to
afford pure chalcones. All yields were calculated from isolated products.
2
.3. Quantitative Structure−Activity Relationships Ana-
lyses. The initial conformations of the compounds were drawn by
3
1
means of the “model build” modulus available in HyperChem 7.
Each molecular structure was first preoptimized with the molecular
mechanics force field (MM+) procedure, and the resulting geometry
was further refined by means of the semiempirical method PM3
entry
product
R
R
1
yield (%)
29
30
31
29
30
31
32
33
34
−H
−H
−H
−H
−CH3
−OCH3
−OH
−Cl
−H
−Cl
83
83
84
88
90
84
(
parametric method 3). We chose a gradient norm limit of 0.01 kcal
−1
́
−1
mol Å . The numerical descriptors for each compound were
32
32
calculated with Dragon software and included several variable types
characterizing the multidimensional aspects of the chemical structure:
constitutional, topological, geometrical, charge, GETAWAY (geome-
try, topology, and atom-weighted assembly), WHIM (weighted holistic
invariant molecular descriptors), 3D-MoRSE (3D molecular repre-
sentation of structure based on electron diffraction), molecular walk
counts, BCUT descriptors, 2D autocorrelations, aromaticity indices,
Randic molecular profiles, radial distribution functions, functional
groups, and atom-centered fragments. We also added quantum
chemical descriptors to the pool such as the highest occupied
3
3
4
−CH3
3
−Cl
that, in general, the reactions were clean and products were
isolated by liquid column chromatography in pure form without
further purification ( H and C NMR).
As a first step, we partitioned the complete set of 28 chalcone
derivatives having experimental activities into 23 training (82%)
and 5 test set compounds (18%). The members of such sets
were chosen by exploring the structure−activity relationship
1
13
(
HOMO) and lowest unoccupied (LUMO) molecular orbital
energies, HOMO−LUMO gap (ΔHOMO−LUMO), total dipole
6
93
dx.doi.org/10.1021/jf203374r | J. Agric.Food Chem. 2012, 60, 692−697

Journal of Agricultural and Food Chemistry
Article
present in these observations. Afterward, we searched various
predictive linear QSAR models for the chalcone structures that
obey the “rule of thumb”. This rule states that at least five or six
data points should be present for each fitting parameter to
avoid overfitting problems.
coefficient, o > xS indicates the number of molecules with a
predicted residual being greater than x times S, R2 represents the
max
maximum squared correlation coefficient between two given
Rand
descriptors of the model, and S
stands for Y randomization.
Table 1S of the Supporting Information includes a brief
description for each molecular descriptor involved in the QSAR
models for the larvicidal activity of the chalcone-type compounds.
Table 2S of the Supporting Information includes the numerical
values of such molecular descriptors.
It is clearly appreciated from Table 2 that the 0D−3D QSAR
models tended to overfit the 23 training set compounds and, as
revealed by the test set results, had a lower predictive power on
structurally related molecules than the 0D−2D QSAR of the
same dimension (d) shown in Table 3. Therefore, we
concluded that topochemical descriptors worked better for
modeling the larvicidal activity of chalcone derivatives. This was
explained by considering that these variables did not depend
upon 3D geometry optimization artifacts that may affect the
calculation of the more sophisticated conformation-based
molecular descriptors.
The optimal linear regression equations were established by
means of the replacement method approach, which minimized
its standard deviation (S) and included the best “representative”
d = 1−3 molecular descriptors. We proceeded by using two
different approaches: (a) extracted d descriptors from the
complete pool containing D = 1502 variables (0D−3D QSAR);
(
b) extracted d descriptors from a pool containing D = 762
variables, which included only topochemical (conformation
independent) type of descriptors (0D−2D QSAR). Tables 2
Table 2. Statistical Characteristics for 0D−3D QSAR Models
on Log LC of Chalcone Derivatives
10
50
regression
coefficient
(
standard
error)
statistical quality
We decided to adopt the QSAR-5 model as the best one,
which included the following two molecular descriptors: the
QSAR-1
2
40
intercept
Mor22p
3.425 (0.2) N/d = 23, R = 0.59, S = 0.54, o ≥ 2.5S = 0,
Balaban U index, Uindex, and the highest eigenvalue 1 of the
Rand
= 0.58, R = 0.50, S = 0.60, R2
2
S
=
41
8.644 (2)
loo
loo
test
Burden matrix/weighted by atomic masses, BEHm1. The
0
.50, Stest = 0.60
Uindex is a molecular descriptor calculated as information
content of molecules, based on the calculation of equivalence
classes from the molecular graph. This kind of descriptor may
be characterized as not having degeneracy for alkanes with up
to 15 vertices. BEHm1 is the highest eigenvalue 1 of the Burden
matrix (B), weighted by atomic masses. As is well-known, B is a
modified adjacency matrix (A) with the main diagonal elements
being weighted with a given atomic property, in the present
case by using atomic masses.
QSAR-2
N/d = 11.5, R = 0.69, S = 0.47, R = 0.14, o
2
2
intercept
PJI3
6.941 (1)
−4.239 (2)
7.134 (2)
max
Rand
2
≥
2.5S = 0, S
= 0.52, R = 0.58, Sloo =
loo
2
0
.56, R = 0.63, Stest = 1.03
test
Mor22p
QSAR-3
2
2
intercept
2.523 (0.4) N/d = 7.7, R = 0.79, S = 0.40, R = 0.55, o ≥
max
Rand
2
loo
2
R
.5S = 0, S
test
= 0.46, R = 0.73, Sloo = 0.46,
RDF030e
Mor22p
R5e
−0.319 (0.1)
9.523 (1)
2
= 0.59, Stest = 0.82
3.336 (0.8)
A comparison of our results with those found by Begum et al.
26
in their QSAR study revealed that the authors established a
model solely based on 3D molecular descriptors, characterizing
the shape and electronic structure of chalcone derivatives. This
set of descriptors were two Jurs descriptors (Jurs_PPSA_1 and
and 3 summarize the main statistical results found for the various
QSAR via these two alternative procedures. Here, N denotes the
2
number of training set molecules, R is the squared correlation
Table 3. Statistical Characteristics for 0D−2D QSAR Models on Log LC of Chalcone Derivatives
1
0
50
6
94
dx.doi.org/10.1021/jf203374r | J. Agric.Food Chem. 2012, 60, 692−697

Journal of Agricultural and Food Chemistry
Article
Jurs_RPCG) and the highest occupied molecular orbital
Table 4 includes the predicted log LC values achieved for
10
50
(
HOMO) energy. Such Jurs -based descriptors were calculated
both the training and test sets of chalcone analogues, according to
by mapping atomic partial charges on solvent-accessible surface
areas of individual atoms. Although that model reported its
statistical calibration results, it did not include any result on
validation. Therefore, we checked the predictive capability of
this model by applying these three descriptors to the training
and test sets of the present work. The results were as follows:
Table 4. Experimental and QSAR Predicted Log LC
50
1
0
Insecticidal Activities of Chalcones
entry
compound
exptl
QSAR- QSAR-
5
1
2
Rand
= 0.59, R2
1
(2E)-1,3-diphenylprop-2-en-1-one
1.954
1.740
2.255
2.291
2.223
2.465
N/d = 7.7, R = 0.58, S = 0.57, o ≥ 2.5S = 0, S
0
=
loo
a
2
test
2
(2E)-1-(2-hydroxyphenyl)-3-
phenylprop-2-en-1-one
.43, S = 0.68, R = 0.67, S = 0.73. The predictive
loo
test
performance of the model based on Jurs descriptors was lower
than that provided by QSAR-5 in Table 3.
3
4
(2E)-3-(2-hydroxyphenyl)-1-
2.017
3.168
2.836
2.998
2.721
0.699
2.994
2.377
0.699
2.982
1.959
2.945
1.908
1.949
1.279
2.334
2.674
2.826
2.225
2.447
1.721
1.867
2.370
1.337
2.281
2.538
2.504
1.566
2.483
1.955
2.327
1.929
2.093
2.137
2.405
1.013
2.699
2.145
1.635
2.197
2.310
2.569
2.102
2.457
1.921
phenylprop-2-en-1-one
Chalcones showing high larvicidal activity presented
enhanced electron densities on one or both rings, in general
having electron-releasing group(s), as 8, 11, and 17; in some
cases an o-hydroxyl group at the A ring enhanced the activity,
suggesting that the resonance-assisted H-bond at ring A could
raise the electron density at ring B (2 vs 1, 5 vs 4, etc.).
However, chalcone 11, having a p-Cl at ring B, was one of the
most active, and an o-HO group at the A ring lowered its
activity in the same manner as occurred with 8 and 9. Besides,
the presence of electron-withdrawing group(s) in general
decreased the activity of chalcones, only slightly in some cases
such as 13 and 16, where a p-HO group was at the A ring, or
more greatly for 6, 7, 12, and 14. Derivatization of the carbonyl
group roughly decreased the activity, and extension of the
branched chromophore of 1 to both vinylogous 19 and 20
neither gave a relevant activity.
(2E)-3-(4-hydroxy-3-methoxyphenyl)-
1
-phenylprop-2-en-1-one
a
5
(2E)-3-(4-hydroxy-3-methoxyphenyl)-
-(2-hydroxyphenyl)prop-2-en-1-one
1
6
7
8
(2E)-3-(3-nitrophenyl)-1-phenylprop-
2-en-1-one
(2E)-1-(2-hydroxyphenyl)-3-(3-
nitrophenyl)prop-2-en-1-one
(2E)-3-(3,4-methylenedioxyphenyl)-1-
phenylprop-2-en-1-one
9
(2E)-3-(3,4-methylenedioxyphenyl)-1-
(
2-hydroxyphenyl)prop-2-en-1-one
a
1
0
(2E)-1-(4-hydroxyphenyl)-3-
phenylprop-2-en-1-one
11
(2E)-3-(4-chlorophenyl)-1-
phenylprop-2-en-1-one
1
1
2
3
(2E)-3-(4-nitrophenyl)-1-phenylprop-
2
-en-1-one
(2E)-1-(4-hydroxyphenyl)-3-(4-
nitrophenyl)prop-2-en-1-one
It was possible to perform a mathematical interpretation of
these structure−activity findings. A proper standardization of
the regression coefficients of the QSAR-5 model enabled the
assignment of greater importance to the molecular descriptors
that exhibited larger (absolute) standardized coefficients. The
resulting order was
a
14
(2E)-1-(2-hydroxyphenyl)-3-(4-
nitrophenyl)prop-2-en-1-one
1
1
1
5
6
7
(2E)-3-(4-chlorophenyl)-1-(2-
hydroxyphenyl)prop-2-en-1-one
(2E)-1-(4-hydroxyphenyl)-3-(3-
nitrophenyl)prop-2-en-1-one
(2E)-3-(furan-2-yl)-1-(2-
hydroxyphenyl)prop-2-en-1-one
Uindex (0.99) > BEHm1 (0.81)
where the standardized values of the regression coefficients (b ) are
(2)
s
j
18
19
(3E)-4-phenylbut-3-en-2-one
2.680
3.315
2.238
2.990
2.759
2.846
shown in parentheses. It was seen that the most important
descriptor for the larvicidal activity of this set of chalcone derivatives
was Uindex, the numerical values of which changed most in
(1E,4E)-1,5-diphenylpenta-1,4-dien-3-
one
2
2
0
1
(2E,4E)-1,5-diphenylpenta-2,4-dien-1-
one
2.640
3.050
2.940
2.985
2.949
2.863
accordance with the numerical variations of the experimental
(2E,4E)-1-(2-hydroxyphenyl)-5-
phenylpenta-2,4-dien-1-one
s
j
activity. However, the relative magnitudes of the b (0.99 and 0.81)
suggested that these numerical variables complemented each other
inside the linear equation, and that resulted in comparable relevance
for predicting the studied property.
According to Table 2S of the Supporting Information, both
descriptors took positive numerical values: the range of Uindex was
22
23
24
25
26
27
28
29
phenylhydrazone of 1
3.329
3.403
2.667
3.370
3.320
2.538
3.421
2.927
3.274
3.116
3.031
3.381
2.477
3.892
2.344
3.356
4.030
2.318
3.165
3.114
3.010
3.926
1.834
a
2,4-dinitrophenylhydrazone of 1
phenylhydrazone of 18
2,4-dinitrophenylhydrazone of 18
semicarbazone of 18
19.349−57.823, whereas the range of BEHm1 was 3.815−3.979. In
oxime of 18
addition, from Table 3 it was noted that Uindex had a positive
regression coefficient in the QSAR-5 model and BEHm1 a negative
one. Therefore, we concluded that a simultaneous decrease of the
numerical value for Uindex and an increase of the BEHm1
descriptor would lead to structures having a highly predicted activity.
Also, a poor predicted activity would be found for a simultaneous
increase of Uindex and a decrease of BEHm1. This result was in
line with the predictions found for the four most active structures: 8
2,4-dinitrophenylhydrazone of 19
(2E)-3-(4-methylphenyl)-1-
phenylprop-2-en-1-one
3
3
0
1
(2E)-3-(4-methoxyphenyl)-1-
phenylprop-2-en-1-one
2.608
2.385
1.279
2.314
1.294
1.307
2.154
1.817
1.886
1.350
(2E)-3-(4-hydroxyphenyl)-1-
phenylprop-2-en-1-one
32
(2E)-1-(4-chlorophenyl)-3-
phenylprop-2-en-1-one
3
3
3
4
(2E)-1-(4-methylphenyl)-3-
phenylprop-2-en-1-one
(
3
2
Uindex, 28.031; BEHm1, 3.915), 11 (Uindex, 28.654; BEHm1,
.947), 17 (Uindex, 26.070; BEHm1, 3.884), and 2 (Uindex,
8.551; BEHm1, 3.878), for which QSAR-5 tended to predict low
log LC values. The four least active structures were 28 (Uindex,
(2E)-1-(4-chlorophenyl)-3-(4-
chlorophenyl)-prop-2-en-1-one
10
50
a
Member of test set.
5
7.823; BEHm1, 3.976), 23 (Uindex, 52.152; BEHm1, 3.979),
5 (Uindex, 48.648; BEHm1: 3.971), and 22 (Uindex, 36.607;
BEHm1, 3.891).
2
the most predictive models appearing in Tables 2 and 3: QSAR-5
(0D−2D) and QSAR-1 (0D−3D). We included the QSAR-1
6
95
dx.doi.org/10.1021/jf203374r | J. Agric.Food Chem. 2012, 60, 692−697

Journal of Agricultural and Food Chemistry
Article
predictions in the table just for comparison. Figures 3A and 4A
plot the predicted activities as functions of the observed ones for
Figure 4. (A) Predicted log LC according to QSAR-1 as a function
10
50
of experimental values. (B) Residuals and predicted log LC .
1
0
50
Figure 3. (A) Predicted log LC according to QSAR-5 as a function
10
50
of experimental values. (B) Residuals and predicted log LC .
ASSOCIATED CONTENT
Supporting Information
10
50
■
*
S
Table giving details of the molecular descriptors employed in
the QSAR models established in this work and NMR spectra of
representatives chalcones 29−34. This material is available free
of charge via the Internet at http://pubs.acs.org.
such models, respectively; Figures 3B and 4B reveal that the
residuals were randomly distributed and did not follow any kind of
strange pattern, which would indicate the presence of nonmodeled
factors.
Finally, the established QSAR enabled us to predict the set of
six chalcone derivatives 29−34 synthesized in our group, the
experimental biological data of which were still not obtained.
Such compounds were structurally similar to some of the
training compounds (29 to 1; 30 and 31 to 4 or 8; 32 to 10,
AUTHOR INFORMATION
■
*
Corresponding Author
Phone: +54 221 425 7430, +54 221 425 7291. Fax: +54 221
425 4642. E-mail: pabloducho@gmail.com.
1
1, or 12; 33 to 2 or 10; 34 to 1, 10, 11, or 12). Their activity
Funding
values were predicted by the application of the novel QSAR
models (QSAR-5 and -1), showing that the expected activity of
We gratefully acknowledge financial support by the Research
Council of Argentina (Consejo Nacional de Investigaciones
2
9 fitted with 1 (active), that of 30 and 31 fitted with 4 (low
activity), 32 fitted well with 11 (highly active), 33 was between
and 10 (active), and 34 also showed a good fitting with 11
high activity). Table 4 includes the results for such predictions.
́
́
Cientificas y Tecnicas (CONICET)), PIP11220100100151
project. We also thank Universidad Nacional de La Plata.
2
(
REFERENCES
■
It was seen that both QSAR-5 and QSAR-1, although being
based on different types of descriptors, tended to achieve
comparable predictions for most structures of this unknown set,
with the exception of 30 ((2E)-3-(4-methoxyphenyl)-1-phenyl-
prop-2-en-1-one), having predicted log LC values of 2.608
for QSAR-5 and 1.307 for QSAR-1. Obviously, we adopted the
result of QSAR-5 (i.e., 2.608) due to the higher statistical quality
of this model. Now, such findings deserve to be experimentally
analyzed in forthcoming bioassays.
(
1) Dimmock, J. R.; Elias, D. W.; Beazely, M. A.; Kandepu, N. M.
Bioactivities of chalcones. Curr. Org. Chem. 1999, 6, 1125−1149.
2) Cave, G. W.; Raston, C. L. Toward benign syntheses of pyridines
(
involving sequential solvent free aldol and Michael addition reactions.
Chem. Commun. 2000, 22, 2199−2200.
1
0
50
(3) Mitsutani, A. Future possibilities of recently commercialized acid/
base-catalyzed chemical processes. Catal. Today 2002, 73, 57−63.
(4) Calloway, N. O.; Green, L. D. Reactions in the presence of
metallic halides. α,β-Unsaturated ketone formation as a side reaction in
Friedel-Crafts acylations. J. Am. Chem. Soc. 1937, 59, 809−811.
6
96
dx.doi.org/10.1021/jf203374r | J. Agric.Food Chem. 2012, 60, 692−697

Journal of Agricultural and Food Chemistry
Article
(
5) Sipos, G.; Sirokman, F. Chalcon formation of different
substituted acetophenones and p-hydroxy-benzaldehyde. Nature
964, 202, 489.
6) Le Gall, E.; Texier-Boullet, F.; Hamelien, J. ChemInform abstract:
products: structure-activity relationship analysis. Med. Chem. Res. 2011,
20, 184−191.
1
(
(27) Romanelli, G. P.; Pasquale, G.; Sathicq, A.; Thomas, H.; Autino,
́
J. C.; Vazquez, P. Synthesis of chalcones catalyzed by aminopropylated
silica sol-gel under solvent-free conditions. J. Mol. Catal. A: Chem.
2011, 340, 24−32.
(28) Duchowicz, P. R.; Goodarzi, M.; Ocsachoque, M. A.; Romanelli,
G. P.; Ortiz, E. V.; Autino, J. C.; Bennardi, D. O.; Ruiz, D.; Castro, E.
A. QSAR analysis on Spodoptera litura antifeedant activities for flavone
derivatives. Sci. Total Environ. 2009, 408, 277−285.
simple access to α,β-unsaturated ketones by acid-catalyzed solvent-free
reactions. Synth. Commun. 1999, 29, 3651−3657.
(
7) Narender, T.; Papi Reddy, K. A simple and highly efficient
method for the synthesis of chalcones by using borontrifluoride-
etherate. Tetrahedron Lett. 2007, 48, 3177−3180.
(
synthesized 2′-oxygenated chalcone derivatives: modulation of human
cell cycle phase distribution. Bioorg. Med. Chem. 2004, 12, 2679−2686.
(
8) Rao, Y. K.; Fang, S.-H.; Tzeng, Y.-M. Differential effects of
(
́
29) Goodarzi, M.; Duchowicz, P. R.; Wu, C. H.; Fernandez, F. M.;
Castro, E. A. New hybrid genetic based support vector regression as
QSAR approach for analyzing flavonoids-GABA(A) complexes.
J. Chem. Inf. Model. 2009, 49, 1475−1485.
9) Chen, M.; Theander, T. G.; Christensen, S. B.; Hviid, L.; Zhai, L.;
Kaharazmi, A. A new antimalarial agent, inhibits in vitro growth of the
human malaria parasite Plasmodium falciparum and protects mice from
P. yoelii infection. Antimicrob. Agents Chemother. 1994, 38, 1470−1475.
(30) Romanelli, G. P.; Virla, E. G.; Duchowicz, P. R.; Gaddi, A. L.;
Ruiz, D. M.; Bennardi, D. O.; Ortiz, E. V.; Autino, J. C. Sustainable
synthesis of flavonoid derivatives, QSAR study and insecticidal activity
against the fall armyworm, Spodoptera frugiperda (Lep.: Noctuidae).
J. Agric. Food Chem. 2010, 58, 6290−6295.
(
10) Fang, N.; Casida, J. E. New bioactive flavonoids and stilbenes in
resin insecticide. J. Nat. Prod. 1999, 62, 205−210.
11) Chen, M.; Zhai, L.; Christensen, S. B.; Theander, T. G.;
cube
(
́
(
31) Hyperchem 7 (Hypercube, Inc.), http://www.hyper.com (last
accessed on Oct 31, 2011).
32) Dragon Milano Chemometrics and QSAR Research Group, http://
michem.disat.unimib.it/chm (last accessed on Oct 31st, 2011).
33) Duchowicz, P. R.; Castro, E. A.; Fernandez, F. M.; Gonzalez, M. P.
Kharazmi, A. Inhibition of fumarate reductase in Leishmania major and
L. donovani by chalcones. Antimicrob. Agents Chemother. 2001, 45,
(
2
(
023−2029.
12) Bandgar, B. P.; Patil, S. A.; Korbad, B. L.; Nile, S. H.;
(
́
́
Khobragade, C. N. Synthesis and biological evaluation of β-chloro
vinyl chalcones as inhibitors of TNF-α and IL-6 with antimicrobial
activity. Eur. J. Med. Chem. 2010, 45, 2629−2633.
A new search algorithm of QSPR/QSAR theories: normal boiling points
of some organic molecules. Chem. Phys. Lett. 2005, 412, 376−380.
(
́
34) Duchowicz, P. R.; Castro, E. A.; Fernandez, F. M. Alternative
(
13) Hsieh, H. K.; Tsao, L. T.; Wang, J. P.; Lin, C. N. Synthesis and
algorithm for the search of an optimal set of descriptors in QSAR-
anti-inflammatory effect of chalcones. J. Pharm. Pharmacol. 2000, 52,
63−171.
14) Zhao, L.-M.; Jin, H.-S.; Sun, L.-P.; Piao, H.-R.; Quan, Z.-S.
QSPR studies. MATCH Commun. Math. Comput. Chem. 2006, 55,
1
(
1
(
79−192.
35) Mercader, A. G.; Duchowicz, P. R.; Fernandez, F. M.; Castro, E. A.
́
Synthesis and evaluation of antiplatelet activity of trihydroxychalcone
QSPR study of solvent quenching of the 5D0→7F2 emission of
Eu(6,6,7,7,8,8,8-heptafluoro-2,2-dimethyl-3,5-octanedionate)3. Chem.
Phys. Lett. 2008, 462, 352−357.
derivatives. Bioorg. Med. Chem. Lett. 2005, 15, 5027−5029.
(
15) Liu, M.; Wilairat, P.; Go, L. M. Antimalarial alkoxylated and
hydroxylated chalcones: structure-activity relationship analysis. J. Med.
Chem. 2001, 44, 4443−4452.
(
Christensen, S. B. Synthesis of antiparasitic licorice chalcones. Bioorg.
Med. Chem. Lett. 1995, 5, 449−452.
(
(
36) Mercader, A. G.; Duchowicz, P. R.; Fernandez, F. M.; Castro, E.
A. Replacement method and enhanced replacement method versus the
genetic algorithm approach for the selection of molecular descriptors
in QSPR/QSAR theories. J. Chem. Inf. Model. 2010, 50, 1542−1548.
(37) Hawkins, D. M.; Basak, S. C.; Mills, D. Assessing model fit by
cross validation. J. Chem. Inf. Comput. Sci. 2003, 43, 579−586.
(38) Wold, S.; Eriksson, L. Statistical validation of QSAR results. In
Chemometrics Methods in Molecular Design; van de Waterbeemd, H.,
Ed.; VCH: Weinheim, Germany, 1995; pp 309−318.
16) Nielsen, S. F.; Chen, M.; Theander, T. G.; Kharazmi, A.;
17) Miranda, C. L.; Aponso, G. L.; Stevens, J. F.; Deinzer, M. L.;
Buhler, D. R. Antioxidant and prooxidant actions of prenylated and
nonprenylated chalcones and flavonoids in vitro. J. Agric. Food Chem.
2
(
000, 48, 3876−3884.
18) Svetaz, L.; Tapia, A.; Lop
́
ez, S. N.; Furlan
́
, R. L. E.; Petenatti, E.;
(39) Draper, N. R.; Smith, H. Applied Regression Analysis; Wiley:
New York, 1981.
Pioli, R.; Schmeda-Hirschmann, G.; Zacchino, S. A. Antifungal
chalcones and new caffeic acid esters from Zuccagnia punctata acting
against soybean infecting fungi. J. Agric. Food Chem. 2004, 52, 3297−
(40) Balaban, A. T.; Balaban, T. S. New vertex invariants and
topological indices of chemical graphs. Based on information on
distances. J. Math. Chem. 1991, 8, 383−397.
(41) Burden, F. R. Molecular identification number for substructure
searches. J. Chem. Inf. Comput. Sci. 1989, 29, 225−227.
3
(
300.
19) Thirunarayanan, G.; Vanangamudi, G. Insect antifeedant potent
chalcones. J. Indian Chem. Soc. 2008, 85, 447−451.
20) Jin, H.; Geng, Y.; Yu, Z.; Tao, K.; Hou, T. Lead optimization
(
and anti-plant pathogenic fungi activities of daphneolone analogues
from Stellera chamaejasme L. Pest Biochem. Physiol. 2009, 93, 133−137.
(
21) Tyrunarayanan, G.; Surya, S.; Srinivasan, S.; Vanangamudi, G.;
Sathiyendiran, V. Synthesis and insect antifeedant activities of some
substituted styryl 3,4-dichlorophenyl ketones. Spectrochim. Acta Part A
2
(
010, 75, 152−156.
22) Hansch, C.; Leo, A. Exploring QSAR. Fundamentals and
Applications in Chemistry and Biology; American Chemical Society:
Washington, DC, 1995.
(
23) Hansch, C.; Fujita, T. Classical and three-dimensional QSAR in
agrochemistry. J. Am. Chem. Soc. 1995, 606, 318−329.
24) Kubinyi, H. QSAR: Hansch Analysis and Related Approaches;
Wiley-Interscience: New York, 2008.
25) Puzyn, T.; Leszczynski, J.; Cronin, M. T. Recent Advances in
QSAR Studies: Methods and Applications, 1st ed.; Springer: New York,
009.
26) Begum, N. A.; Roy, N.; Laskar, R. A.; Roy, K. Mosquito
larvicidal studies of some chalcone analogues and their derived
(
(
2
(
6
97
dx.doi.org/10.1021/jf203374r | J. Agric.Food Chem. 2012, 60, 692−697