Insecticidal and neuroblocking potencies of variants of the imidazoline moiety of imidacloprid-related neonicotinoids and the relationship to partition coefficient and charge density on the pharmacophore

Article abstract of DOI:10.1021/jf0623440

The pharmacophore of the neonicotinoid insecticide imidacloprid, nitroiminoimidazolidine, was modified to heterocycles such as thiazolidine, pyrrolidine, dihydroimidazole, dihydrothiazole, and pyridone conjugated to nitroimine (=NNO₂) or nitromethylene (=CHNO₂). Their 6-chloro-3-pyridylmethyl or 5-chloro-3-thiazolylmethyl derivatives were examined for insecticidal activity against the American cockroach by injection and for neuroblocking activity using the cockroach ganglion. Most of the compounds having the neonicotinoidal pharmacophore exhibited insecticidal activity at the nanomolar level, which was enhanced in the presence of synergists, and high neuroblocking activity at the micromolar level. Quantitative analysis for the compounds showed that the neuroblocking potency is proportional both to the Mulliken charge on the nitro oxygen atom and to the partition coefficient log P value. The equation for the insecticidal versus neuroblocking potencies indicated that both potencies are related proportionally with each other when the other factors are the same.

Full text of DOI:10.1021/jf0623440

812 J. Agric. Food Chem. 2007, 55, 812−818
Insecticidal and Neuroblocking Potencies of Variants of the
Imidazolidine Moiety of Imidacloprid-Related Neonicotinoids
and the Relationship to Partition Coefficient and Charge
Density on the Pharmacophore
SHINZO KAGABU,*,† RIKA ISHIHARA,† YOSUKE HIEDA,†
#
KEIICHIRO NISHIMURA,§ AND YUJI NARUSE
Department of Chemistry, Faculty of Education, Gifu University, Gifu 501-1193, Japan; Research
Institute for Advanced Science and Technology, Osaka Prefecture University, Osaka 599-8570, Japan;
and Department of Chemistry, Faculty of Engineering, Gifu University, Gifu 501-1193, Japan
The pharmacophore of the neonicotinoid insecticide imidacloprid, nitroiminoimidazolidine, was modified
to heterocycles such as thiazolidine, pyrrolidine, dihydroimidazole, dihydrothiazole, and pyridone
conjugated to nitroimine (dNNO2) or nitromethylene (dCHNO2). Their 6-chloro-3-pyridylmethyl or
5-chloro-3-thiazolylmethyl derivatives were examined for insecticidal activity against the American
cockroach by injection and for neuroblocking activity using the cockroach ganglion. Most of the
compounds having the neonicotinoidal pharmacophore exhibited insecticidal activity at the nanomolar
level, which was enhanced in the presence of synergists, and high neuroblocking activity at the
micromolar level. Quantitative analysis for the compounds showed that the neuroblocking potency is
proportional both to the Mulliken charge on the nitro oxygen atom and to the partition coefficient log
P value. The equation for the insecticidal versus neuroblocking potencies indicated that both potencies
are related proportionally with each other when the other factors are the same.
KEYWORDS: Imidacloprid; neonicotinoid insecticide; pharmacophore; neuroblocking activity; Mulliken
atomic charge; log P
INTRODUCTION
Neonicotinoids, synthetic insecticides acting on the insect
nicotinic acetylcholine receptor (nAChR), have been increas-
ingly used to control various insects during recent decades, since
imidacloprid (1) was introduced to the market (1). The supreme
biological profile of imidacloprid (1-3) is giving impulse to
the development of a new product by modifying the structural
features of the prototype (4). Now nitenpyram (5), acetamiprid
(6), thiamethoxam (7), thiacloprid (8), clothianidin (9), and
dinotefuran (10) are on the market with their own prominence
(Figure 1).
The potent neonicotinoids share common structural fea-
tures: a chlorine-substituted heteroaromatic ring bearing a
nitrogen atom at the quasi meta position coupled through a
methylene juncture to an amidinyl or guanidinyl moiety that is
conjugated to N-nitroimine (dNNO2), N-cyanoimine (dNCN),
or nitromethylene (dCHNO2). Exceptionally, dinotefuran bears
a 3-tetrahydrofuran moiety in place of the chlorine-substituted
3-N-heteroaromatic ring.
* Author to whom correspondence should be addressed [telephone (+81)-
58-2932253; fax (+81)58-2932207; e-mail kagabus@gifu-u.ac.jp].
† Department of Chemistry, Faculty of Education, Gifu University.
§ Osaka Prefecture University.
Figure 1. Commercialized neonicotinoid insecticides.
# Department of Chemistry, Faculty of Engineering, Gifu University.
10.1021/jf0623440 CCC: $37.00
© 2007 American Chemical Society
Published on Web 01/17/2007

Quantitative Activity Analysis for Variants of Imidacloprid
J. Agric. Food Chem., Vol. 55, No. 3, 2007 813
Figure 2. Schematic model for binding of neonicotinoid to nAChR (Y
)
N, CH; X
) NH, S, CH2). The nitro oxygen atom interacts through
H-bonding, the electron-deficient moiety (δ+) of the central ring through
the electrostatic force, and the heteroaromatic nucleus (depicted as Het)
in an auxiliary way with the recognition site on the nAChR.
On the basis of the structural features of neonicotinoid
isotypes, we have proposed a binding model: the ligand
molecule interacts with the group on the receptor by a Coulomb
interaction through an electron-deficient (e-deficient) nitrogen
atom on the guanidine (or amidine) and by a hydrogen bond
(H-bond) through the H-accepting tip at the electron-withdraw-
ing group and the binding of the appending heterocyclic group
to the receptor in an auxiliary way (11) (Figure 2). Matsuda et
al. rectified this model on the PM3 calculations so that the
electronic flow cascade from the electron-donating nitrogen atom
to the electron-negative tip can be fortified by the H-bond (12).
Tomizawa et al. evolved our model so that the guanidine (or
amidine) part would function as a π-acid toward the tryptophan
residue on the putative insect receptor (13). Nakayama and
Sukekawa (14) and Okazawa et al. (15) rationalized the necessity
of a partially positive electrostatic amidine/guanidine site and
the electronegative NO2 or CN group for the binding to the
receptor on the basis of similarity indices and comparative
molecular field analysis, respectively. All of these pharmacoph-
ore models commonly underscore the importance of the
negatively charged (n-charged) H-accepting tip conjugated to
the amidine moiety.
Figure 3. Tested compounds and general preparation scheme.
Not only the pharmacophore but also the pharmacokinetics
related to the physicochemical properties of the whole molecule
including the auxophore such as the water solubility or the
molecular size are other indispensable aspects for the drug action
(16, 17). In the neonicotinoid field, questions regarding how
the lipophilicity difference among the functional groups
(dNNO2, dCHNO2, dNCN) or the cyclic/acyclic bioisosteres
or the size difference among the alkyl substituents affects the
biological response have been addressed (4, 18).
on the basis of toxicity estimation by injection and nerve impulse
measurements using the American cockroach (AmC): the
substituents on the pyridine ring (19, 20), imidazolidine N3-
alkyl substituents (21, 22), acyclic compounds (23-25), and
the enantiomers at the juncture (26). We expect in the present
work that the approach based on the quantitative consideration
for the biological response using the quantum chemical and the
physiochemical parameters will deepen our insight into the
binding mode for neonicotinoid insecticides.
We examined in this paper the influence of the modification
of the imidazolidine ring of imidacloprid on the physicochemical
properties and the resultant biological trend. The modified ring
includes thiazolidine, pyrrolidine, unsaturated 1,3-dihydroimid-
azole and -thiazole, or a fully conjugated pyridone ring (hereafter
these are referred to as the central ring), and the central rings
of the tested molecules are conjugated to nitroimine or nitrom-
ethylene. As the appending heteroaromatic ring were chosen
2-chloro-5-pyridyl (ClPy) and 2-chloro-5-thiazolyl (ClTh) moi-
eties. Different atom alignments in the amidine (guanidine)
moiety would produce the electronic distribution and the
molecular lipophilicity differently, and the constructed molecules
will exhibit the biological behavior in the reflection upon the
newly attained physical properties. We further studied the
quantitative relationships between the biological potency and
the quantum chemical index, the Mulliken charge, and a
lipophilicity parameter, log P, the octanol/water partition
coefficient.
MATERIALS AND METHODS
Preparation of Compounds. Compounds 2 (27), 3, 10, and 11 (28),
6 (29), and 12 and 13 (30) were prepared according to the described
procedures. The other compounds disclosed on patents (8, 31-35)
without the specific preparative procedures were obtained according
to the scheme in Figure 3, and the structures were confirmed by the
analytical data (see Supporting Information).
Biological Tests. Chemicals. Biological data for compounds 1 and
11 (22), and 12 (36) were taken from our previous literature. Reagent-
grade piperonyl butoxide (PB) purchased from Tokyo Kasei Kogyo
Ltd. (Tokyo, Japan) was used as an inhibitor of oxidative metabolism.
Propargyl propyl benzenephosphonate (NIA 16388) was the same
sample used in our previous studies (27). NIA originally was an
inhibitor of the hydrolytic metabolism of pyrethroids (37), and recently
Nishiwaki et al. evidenced the interference of the enzymatic hydroxy-
lation at the imidazolidine ring of imidacloprid by NIA (38).
Insecticidal Test against AmC. The insecticidal assay against male
adult AmC, Periplaneta americana L., was conducted as described
previously (19-22, 27). Various volumes (1-10 µL) of each compound
dissolved in methanol containing some amount of dimethyl sulfoxide
We have thus far pursued the relationship of the insecticidal
activity of the neonicotinoids to the following structural parts

814 J. Agric. Food Chem., Vol. 55, No. 3, 2007
Kagabu et al.
Quantum Chemistry Calculations. The geometry optimization was
carried out at the B3LYP/6-31G(d) level using the Gaussian 98 program
(39). The optimized geometries are confirmed with no imaginary
frequencies by the vibrational analyses. We used the Mulliken charges
to evaluate the charges at C, X, Y, O1, and O2 in compounds I-XIV
(Figure 5), which are summarized in Table 2.
Correlation Analysis. Variations in the neuroblocking activity were
analyzed by using free energy related physicochemical parameters of
compounds according to eq 1 (40, 41)
Table 1. Biological Potencies of Tested and Reference Compounds
insecticidal potencyb
neuroblocking potencyc
obsdf calcdg
compda
alone/ratiod
+
(PB/NIA)/ratioe
1h
2
3
4
5
6
7
8
9
10
11j
12k
13
14
15
16
17
8.96/1
10.15(10.07)/15
10.12(9.77)/80
10.19(9.87)/10
10.44 (10.85)/3.9
10.37(10.62)/4.0
9.62(9.71)/16
10.24(9.99)/8.1
8.91(9.39)/32
10.06(9.95)/13
8.62(8.87)/1.3
8.07(7.99)/2.0
8.72(9.16)/10
10.25(10.11)/25
8.82(8.94)/5.1
8.85(8.84)/10
9.99(9.63)/13
9.74(9.42)/6.7
5.64 (5.60
5.35 (5.26
5.02 (4.95
6.43 (6.32
6.19 (6.12
5.97 (5.89
5.73 (5.58
6.00 (5.92
6.31 (6.17
3.81i
−
5.69)
5.44)
5.09)
6.59)
6.27)
6.02)
5.87)
6.09)
6.46)
5.38
5.56
5.17
6.13
6.20
5.54
5.58
5.69
6.50
4.22
3.76
5.57
5.54
5.45
6.39
5.31
5.63
8.22/5.5
9.19/0.59
9.84/0.13
9.76/0.15
8.42/3.5
9.34/0.42
7.42/35
8.96/1.0
8.53/2.7
7.77/15
7.72/17
8.85/1.3
8.12/6.9
7.85/13
8.88/1.2
8.92/1.1
−
−
−
−
−
−
−
−
log(1/BC) ) aQ + b(log P) - c(log P)2 + dS - eS2 + constant
(1)
where Q represents the Mulliken charges of compounds represented in
Figure 5. Log P is the hydrophobicity parameter, the values of which
are listed in Table 2. S is the steric parameter of the compounds. To
determine the existence of an optimum in the hydrophobicity and steric
dimensions, the squared parameter terms were added so that c and e g
0.
We examined the relationship between the insecticidal activity
against AmC, which was measured with the synergists, and the nerve-
blocking activity with the AmC nerve cord using eq 2 (42).
3.84 (3.74
4.31 (4.26
5.68 (5.54
4.61 (4.56
5.79 (5.75
5.09 (5.02
5.75 (5.61
−
−
−
−
−
−
−
3.97)
4.39)
5.89)
4.66)
5.84)
5.14)
5.88)
a Structure, see Figure 3. b log(1/MLD) (mol). c log(1/BC) (M). d MLD (in mol)
ratio to that of imidacloprid (1). e Values in parenthesss are calculated from eq 6;
ratio of MLD(PB+
NIA) to MLD(alone) in mol. f Values in parentheses indicate the
range of error estimated from standard deviation. g Calculated from eq 4.
h Imidacloprid. Biological data taken from ref 22. i The value was predicted from
the datum in ref 21. j Biological data taken from ref 22. k Biological data taken
from ref 36.
log(1/MLD) ) a(log 1/BC) + b(log P) - c(log P)2 + constant
(2)
The squared log P term was added to identify the optimum hydrophobic
effect, so that c g 0. The coefficients a, b, c, d, and e, and the constants
in the equations were determined by the least-squares method. Unless
otherwise noted, statistical significance levels of the correlation
equations and the independent terms in each equation were above 95%
as examined by the t test.
(DMSO) were injected into the abdomen of an AmC. Organic solvents
alone in this range did not have a toxic effect. Details of the dosage
were fundamentally the same as described previously (19, 20, 27). The
doses were varied by every 1.25 times in moles. In some experiments,
a methanol solution (1 µL) containing PB (50 µg) and NIA (50 µg)
was injected 1 h before injection of the test compound. The metabolic
inhibitors in these amounts did not have a toxic effect. Three insects
were used to test each dose of each compound and were kept at 22-
25°C for 24 h after injection. The minimum dose at which two of three
insects were considered killed was taken as the minimum lethal dose
(MLD in moles). Paralyzed insects were also counted as having died.
The MLD values for the test compounds are listed as log(1/MLD)-
(mol) in Table 1. Each value is the mean of at least two experiments
with a deviation of (0.2 in log units.
Neurophysiological Assay. The neurophysiological test of the
compounds was conducted as described previously (21-27). In brief,
a nerve preparation containing the abdominal fifth and sixth ganglia
of a male adult AmC was excised and placed in a saline solution. One
of two bundles of the nerve cord was taken up from the thoracic side
with saline into a glass tube, in which a silver wire was set as an
electrode. As the reference electrode, another wire was set outside the
cut end of the tube. The silver wires were thinly coated with silver
chloride. The number of spontaneous discharges that were larger than
approximately 15 µV was consecutively counted with a pulse counter
(MET-1100, Nihon Kohden, Tokyo, Japan) for every 30-s period. The
frequency was usually quite high for a few minutes after setting and
then normally subsided. When the frequency decreased at around a
range of 30-400 counts per 30 s for about 2 min, the saline solution
was substituted to one with a test compound dissolved in methanol
containing some amount of DMSO. The final concentration of the
organic solvents was lower than 1% (v/v), which did not essentially
affect the nerve activity. Measurements were conducted at 22-25°C.
The neuroblocking concentration in terms of log(1/BC) (M) defined
below are listed in Table 1 along with their deviation ranges.
Hydrophobicity Parameter. Log P, where P is the partition
coefficient of compounds in the 1-octanol/water partitioning system,
was determined by the shaking-flask method (21). The concentration
of compounds in the water phase was measured by HPLC using an
ODS column (LiChrosorb RP-18, Merck, Darmstadt, Germany) with
a mixture of acetonitrile and water (3:7 to 1:1 v/v) as the mobile phase.
The log P values are listed in Table 2.
RESULTS AND DISCUSSION
The biological potencies are listed in Table 1. In the
insecticidal estimation without synergist (alone), nitromethylene
compounds (3-5, 7, and 9) surpassed significantly the related
nitroimines (1, 2, 6, and 8) irrespective of the kind of central
or heteroaromatic ring. Of them the activities of compounds
3-5 and 7 stood out. These compounds were 2-8 times as
potent as imidacloprid (1). The influence of the central ring on
the activity was not uniform; the imidazolidine, pyrrolidine, and
thiazolidine rings made no appreciable difference among
nitromethylene derivatives irrespectively substituted with the
ClPy (3-5) or the ClTh group (7 and 9). The effect of the
introduction of a CdC bond to the central five-membered rings
was not unified either; the ClPy-1,3-dihydroimidazole derivative
(12), apparently, and ClTh derivative (14) were somewhat lower
in activity than, respectively, imidacloprid (1) and compound
6, whereas ChTh thiazole derivatives 8 and 15 showed almost
equal potency. On the other hand, interestingly, the fully
conjugated pyridone systems afforded high activity (16 and 17),
which should be contrasted with the enlarged saturated ring
showing an appreciable drop (1 vs 2). Methylation at the N3-
position on the imidazolidine ring resulted in compounds of
low activity (10 and 11 vs 1 and 3). As for the appending
heteroaromatic nucleus, the activity of the ClPy residue was
great over the ClTh one.
In principle the insecticidal potency improves when synergists
eliminate or minimize the enzymatic metabolism. We have thus
far observed the activity enhancement of neonicotinoid com-
pounds by the presence of PB and NIA (19-27). Also in the
present case, the potency improvement was observed in all
compounds (Table 1). The high synergistic effect for com-
pounds 2, 8, and 13 was noticeable; as a result their potencies
were as high as or even higher than that of imidacloprid (1)
under synergistic conditions. The metabolic pathways for
neonicotinoids include the oxidation at several molecular sites

Quantitative Activity Analysis for Variants of Imidacloprid
J. Agric. Food Chem., Vol. 55, No. 3, 2007 815
Table 2. Calculated Mulliken Charges and Partition Coefficients of Selected Nitro Compounds
structural features
W/(CH2)n/X,Y,Z
Mulliken chargesa
O1
log Pb
type
X−Me
C
Y
O2
Pyr
Thy
I
II
III
IV
V
VI
VII
VIII
IX
X
XI
XII
XIII
XIV
XV
0/N,NH,N
0/N,NMe,N
0/N,NH,CH
0/N,NMe,CH
1/N,NH,N
0/N,CH2,CH
0/N,S,N
−0.468
−0.461
−0.486
−0.472
−0.471
−0.441
−0.401
−0.402
−0.472
−0.414
−0.467
−0.007
−0.007
0.054
0.019
0.834
0.810
0.630
0.638
0.832
0.416
0.385
0.143
0.823
0.365
0.581
0.423
0.455
0.070
0.186
−
−
−
−
−
0.253
0.449
0.268
0.471
0.258
0.043
0.263
0.264
0.243
0.360
0.035
0.249
0.454
0.263
0.049
−0.470
−0.429
−0.490
−0.463
−0.484
−0.453
−0.416
−0.434
−0.493
−0.432
−0.441
−0.479
−0.410
−0.411
−0.416
−0.398
−0.399
−0.435
−0.432
−0.401
−0.426
−0.391
−0.427
−0.416
−0.403
−0.397
−0.424
−0.415
−0.409
−0.396
1 (0.53)
6 (0.61)
10 (
3 (
11 (
−
−
−
0.07)
0.18)
0.84)
7 (0.03)
2 (0.57)
9
8 (0.80)
4 (0.45)
0/N,S,CH
5 (0.47)
12 (0.33)
13 (0.53)
16 (0.51)
24
25
26
27
CH
CH
d
d
CH/N,NH,N
CH/N,S,N
−
14 (0.27)
15 (0.96)
17 (0.67)
pyridone/N,CH,N
0/CH,NH,CH
0/CH,O,CH
0/CH,S,CH
0/CH,CH2,CH
−
−
−
a Hydrogens are summed into the heavy atoms. b Log P in parentheses denotes the log of the partition between octanol and water; the bold figure stands for the number
of the corresponding compound; log P was not measured for compounds 24 27.
−
and the reduction at the nitro group (43). The present synergists,
PB and NIA, were proved to interfere with the former enzymatic
action (38). Inherently highly active compounds do not show a
drastic activity drop until a substantial amount of the active
structure is decomposed (43). This is the case of nitromethylene
derivatives of 3-5 and 9 displaying the modest synergistic
effect, judging from the higher binding affinity to insect nAChR
of the nitromethylene analogues than the corresponding ni-
troimines (44). The lower sensitivity of the nitromethylene
moiety to reductive enzymes (43) will be an additional contribu-
tor to the smaller synergistic effect. The high synergistic effect
for compounds 8, 13, and 15 is possibly due to the sulfide
moiety sensitive to the enzymatic oxidation in the absence of
synergists as observed in other neonicotinoidal sulfide deriva-
tives (27). The synergistic ratio at 80 for the diazacyclohexane
compound (2) compared to 15 for the five-membered homologue
(1) is conspicuous. The extra methylene at the sixth position is
assumed to be responsible for it, considering that the methylene
at the more remote position from the e-deficient nitroguanidine
part should be more oxidation sensitive than the conjunct
methylenes.
Figure 4. Time courses of the effect of compound 13 on spontaneous
discharges in the excised central nerve cords of American cockroaches.
After counting for 5 min, the nerve preparations were started to treat with
-
5
-
6
-6
the compound (b, 3.1 × 10 M; 0, 2.4 × 10 M; 9, 1.2 × 10 M).
The primary target of the neonicotinoids is the nAChR. The
cation-permeable ion channels of the nAChR desensitize in
response to full occupation of the ligand binding sites with
acetylcholine. Thus, prolonged exposure of the nAChR to
nonhydrolyzable agonists such as imidacloprid results in a block
of cholinergic neurotransmission. The excitation duration time
(the time until the blocking begins) and the frequency of nerve
impulses is in principle proportional to the binding capacity to
the receptor and the quantity of the ligand on it. One can
estimate the neuroexcitation or neuroblocking capacity of a
ligand by comparing the minimum concentration to induce the
two-phase electric episode (18, 22, 24).
Figure 4 exemplifies the time courses for compound 13, in
which each symbol indicates cumulative counts for every 30 s
in each nerve preparation at three concentrations. The two-phase
characteristics, first increasing and then decreasing, were clearly
observed, and generally the dose for the excitation of the nerve
is smaller than that for the following blockade. Between these
two effects, the nerve-blocking measurement seems to be related
to the insecticidal tendency more significantly than the nerve
excitation (22, 24), so we evaluated only the BC in the present
study. To evaluate the neuroblocking activity, the time when
the number of firings decreased to a level lower than 10 counts
per 30 s, which was defined as t in minutes, was determined.
Figure 4 shows how the effect on the frequency and the time
for subsidence depended on the concentration of the compound.
Similar measurements were conducted at least at two concentra-
tions for each compound. More than three nerve preparations
were used for each concentration. From a concentration-
response relationship for each compound (data not shown), the
concentration required to reach 1 in terms of log t was
determined and was defined as BC (in moles). Table 1 lists
the log(1/BC) values of tested compounds.
The blocking potencies of nitromethylene compounds (4, 5,
and 9) surpassed that of imidacloprid (1). Of them the activities
of pyrrolidine derivatives 4 and 9 were outstanding. Extendedly
conjugated systems, 1,3-dihydrothiazole derivatives (13 and 15)
with BC values of, respectively, 2.1 and 1.6 µM, and pyridone
(17) with a BC value 1.8 µM showed excellent neuroblocking
activity. Most of the other molecules elicited a neuroblocking
effect at the micromolar level, and such high activity is
understandable from their common structure, that is, a conju-
gated system composed of an electron-donating amidine or
guanidine part and a powerful electron-withdrawing group. On
the other hand, however, N3-methyl imidazolidine derivatives
(10 and 11) showed extraordinarily low neuroblocking potencies,

816 J. Agric. Food Chem., Vol. 55, No. 3, 2007
Kagabu et al.
coefficient, and F is the value of the ratio between regression
and residual variances. The figures in parentheses following the
intercept and the regression coefficients are their 95% confidence
intervals.
From careful examinations of the results, the experimentally
measured log(1/BC) values for compounds 12 and 14 seemed
to be much lower than the values calculated by eq 3 (data not
shown). By excluding these compounds, eq 3 was much
improved to eq 4.
Figure 5. Molecular structures for calculation of the atomic charge in
Table 2 and reference compounds 18−21.
log(1/BC) ) -8.527((5.786) - 32.303((13.791)(QO2) +
1.983((0.495)(log P) (4)
and those of dihydroimidazole compounds (12 and 14) were
only modest despite their sharing the common pharmacophore,
which suggests that some other factors should be also involved
in the activity exhibition.
n ) 15, s ) 0.315, r ) 0.930, F2,12 ) 38.16
Addition of the (log P)2 term did not improve the correlation.
Moreover, addition of the steric terms for compounds such as
Vw and/or (Vw)2, where Vw is the van der Waals volume (47),
did not improve these equations either. Because the steric
dimension of the tested compounds was not much varied as a
whole, steric parameters seemed to be inexpedient to improve
the correlations. Equation 4 indicates that the higher the O2
charge of the nitro group and the greater the hydrophobicity of
the molecule, the higher the neuroblocking activity. The
activities calculated by eq 4 are listed in Table 1. From the
equation, we can understand, for example, that the larger log P
value is contributing to the superior neuroblocking ability of
imidacloprid (1) to the nitromethylene compound (3), and in
contrast the detrimental effect by N3-methyl substitution for
compounds 10 and 11 must be caused by the smaller log P
values [the unusual water solubility of 10 and 11 has been
ascribed to the distortion of the π-conjugation coplanarity by
N-methylation (28)]. For the deviation of compounds 12 and
14 from eq 4, we suspect that a different factor is needed to
determine the ligand-receptor interaction because of the
strongly acidic NH proton (36, 48).
To investigate the structural and physicochemical parameters
associated with the biological potencies, we calculated the
Mulliken charges of the model compounds (I-XV; Figure 5),
where the heteroarylmethyl residue is simplified as a methyl
group (Table 2). Table 2 not only outlines the tendency of the
charge distribution among the composed atoms in a structure
but also demonstrates that the charge of the individual atom
varies among the constituted structures. We have emphasized
in the proposed model the importance of the presence of the
nitro oxygen atom as the H-acceptor. First, we examined the
relationship of the nitro oxygen atom O2 to the neuroblocking
potency. The O2 atom of all of the structures is strongly
n-charged, and compounds of high neuroblocking potencies such
as compounds 4, 5, 8, and 9 have factually larger negative
charges on it. To certify the relationship of the charge to the
activity, we took up compounds 18-21 for referral (Figure 5).
The reported binding strengths (ki) of compounds 18, 19, and
20 to the Drosophila nAChR are 1.2 ( 0.3, 180 ( 15, and
36000 ( 2750 nM (45), respectively. The calculated charge
for XII-XV (Table 2) is actually highly related with the
reported binding strength order as well as the nil insecticidal
activity of 21 (46). As above, we could grasp the parallel
relationship between the neuroblocking activity and the nitro
oxygen charge. However, the log(1/BC) values of some
compounds such as 10 and 11 were apparently lower than
predicted from the charge magnitudes. Actually, we could not
derive an equation of significant quality relating solely with
the charge on the O2 for the listed compounds, even excepting
10 and 11.
We could not derive any significant quantitative equation by
using the Mulliken charge on O1. This atom is taken to be
involved in a six-membered intramolecular H-bonding with the
N3-H on the imidazolidine ring (11), and the resultant geo-
metrical or electronic situations may complicate the relationship
for the interaction with the amino acid residue on the receptor.
As for the charge on the amidinyl nitrogen atom (X in Figure
5), it has been controversial if the N atom is positive or negative
(1, 3, 12, 13). In our calculation at B3LYP/6-31G*, the nitrogen
atom in question was evidently negative. Even the nitrogen atom
of tetramethylammonium was as negative as -0.359, and
considering the atomic charges of 0.334 and 0.225, respectively,
for C and H, we can recognize that the nitrogen atom withdraws
electrons from the surrounding four methyl groups, and in turn
the hydrogen atoms are positively charged (p-charged) to a great
extent. We can figure the situation in a neonicotinoid molecule
in a similar fashion as the n-charged nitrogen atom is tacked
with the p-charged hydrogen atoms on the conjunct methylenes.
The activity strength is determined not only by the pharma-
codynamic but also by the pharmacokinetic factors related to
the bioavailability. It is well proved that hydrophilic or ionized
molecules elicit only a mediocre effect because of the poor
permeability through the ion barriers in the neurons despite
fulfilling the pharmacophoric requirements (16). To examine
how the hydrophobicity affects concurrently the biological trend
of the compounds, we quantitatively analyzed the relationship
between neuroblocking potency and the Mulliken charge on the
O2 (QO2) and the log P term using eq 1 and found eq 3 as the
most fitting one.
The pivotal carbon atom on the amidinyl part in almost all
of the tested molecules is p-charged (Table 2), in contrast with
the n-charged methyl carbon in the ammonium ion. This
difference will provide a picture of a neonicotinoid molecule
delineating an electronic flow from the amidinyl part to the nitro
oxygen group by π-conjugation. The amidine part has been
assumed to interact with the aromatic π-nucleus of tryptophan
on the receptor (13). We attempted to derive an equation
including the Mulliken charge (Table 2) or the LUMO
log(1/BC) ) -7.651((8.609) - 29.883((20.499)(QO2) +
1.976((0.737)(log P) (3)
n ) 17, s ) 0.477, r ) 0.839, F2,14 ) 16.57
In this and the following equations, n is the number of
compounds, s is the standard deviation, r is the correlation

Quantitative Activity Analysis for Variants of Imidacloprid
J. Agric. Food Chem., Vol. 55, No. 3, 2007 817
coefficients (data not shown) on the C for the neuroblocking
potency, but we could not attain one with a sufficiently reliable
coefficient.
Supporting Information Available: Preparation of com-
pounds 4, 5, 8, 9, and 14-17. This material is available free of
charge via the Internet at http://pubs.acs.org.
Although the log(1/MLD) measured with the synergists did
not show any clear relationship with the Mulliken charges, we
could derive the following equation between the insecticidal
and the neuroblocking potencies by referring eq 2.
ACKNOWLEDGMENT
We thank Satoshi Inagaki for helpful comments.
log(1/MLD) ) 6.057((1.414) + 0.748((0.266)
(log 1/BC) - 1.568((0.845)(log P)2 (5)
LITERATURE CITED
(1) Kagabu, S. Chloronicotinyl insecticidessdiscovery, application
and future perspective. ReV. Toxicol. 1997, 1, 75-129.
(2) Tomizawa, M.; Casida, J. E. Neonicotinoid insecticide toxicol-
ogy: mechanism of safety action. Annu. ReV. Entomol. 2003,
48, 339-364.
(3) Matsuda, K.; Shimomura, M.; Ihara, M.; Akamatsu, M.; Sattelle,
D. B. Neonicotinoids show selective and diverse actions on their
nicotinic receptor targets: electrophysiology, molecular biology,
and receptor modeling studies. Biosci., Biotechnol., Biochem.
2005, 69, 1442-1452.
(4) Kagabu, S. Molecular design of neonicotinoids: past, present
and future. In Chemistry of Crop Protection, Progress and
Prospects in Science and Regulation; Voss, G., Ramos, G., Eds.;
Wiley-VCH: Weinheim, Germany, 2003; pp 193-212.
(5) Akayama, A.; Minamida, I. Discovery of a new systemic
insecticide, nitenpyram and its insecticidal properties. In Nico-
tinoid Insecticides and the Nicotinic Acetylcholine Receptor;
Yamamoto, I., Casida, J. E., Eds.; Springer: Tokyo, Japan, 1999;
pp 127-148.
(6) Yamada, T.; Takahashi, H.; Hatano, R. A novel insecticide,
acetamiprid. In Nicotinoid Insecticides and the Nicotinic Ace-
tylcholine Receptor; Yamamoto, I., Casida, J. E., Eds.;
Springer: Tokyo, Japan, 1999; pp 149-176.
(7) Maienfisch, P.; Brandl, F.; Kobel, W.; Rindisbacher, A.; Senn,
R. CGA 293343: a novel broad-spectrum neonicotinoid insec-
ticide. In Nicotinoid Insecticides and the Nicotinic Acetylcholine
Receptor; Yamamoto, I., Casida, J. E., Eds.; Springer: Tokyo,
Japan, 1999; pp 177-209.
(8) Shiokawa, K.; Tsuboi, S.; Kagabu, S.; Sasaki, S.; Moriya, K.;
Hattori, Y. Preparation of (heterocyclylmethyl)imidazolines,
-thiazolidines, -tetrahydropyrimidines, and -tetrahydrothiazines
as insecticides. Eur. Pat. Appl. EP 0235725 A1, 1987.
(9) Jeschke, P.; Uneme, H.; Benet-Buchholz, J.; Stoelting, J.; Sirges,
W.; Beck, M. E.; Etzel, W. Clothianidin (TI-435)sthe third
member of the chloronicotinyl insecticide (CNITM) family.
Pflanzens.-Nachr. Bayer. 2003, 56, 5-25.
(10) Wakita, T.; Kinoshita, K.; Yamada, E.; Yasui, N.; Kawahara,
N.; Naoi, A.; Nakaya, M.; Ebihara, K.; Matsuno, H.; Kodaka,
K. The discovery of dinotefuran: a novel neonicotinoid. Pest
Manag. Sci. 2003, 59, 1016-1022.
(11) Kagabu, S.; Matsuno, H. Chloronicotinyl insecticides. 8. Crystal
and molecular structures of imidacloprid and analogous com-
pounds. J. Agric. Food Chem. 1997, 45, 276-281.
n ) 17, s ) 0.397, r ) 0.871, F2,14 ) 22.09
Because the experimentally determined insecticidal activities
of ClTh compounds 6-9, 14, 15, and 17 tended lower than
those calculated by eq 5, an indicator variable term I was added
to improve the equation, giving eq 6, where the I value for the
seven compounds was unity.
log(1/MLD) ) 5.634((1.178) + 0.851((0.226)
(log 1/BC) - 1.296((0.708)(log P)2 -0.524((0.378)I (6)
n ) 17, s ) 0.317, r ) 0.926, F2,14 ) 26.08
Insecticidal activities calculated by eq 6 are listed in Table 1.
The coefficient of the I term means that the insecticidal activity
of the ClPy compounds was about 3 times higher than the ClTh
compounds when other factors were the same. Furthermore, eq
6 indicates that the high neuroblocking activity gives the potent
insecticidal activity. Addition of the log P term to eq 6 or
replacement of the (log P)2 term by the log P term did not
improve the quality of the equation. In each set of the ClPy
and the ClTh compounds with the same neuroblocking activity,
the insecticidal activity was parabolically related to the log P
value with an optimum of around zero. Similar parabolic
relations were observed in such cases as the insecticidal activity
against the AmC for other series of neonicotinoids with the
neuroblocking (22, 24) and with the neuroexcitatory (19, 21)
activities.
We have above discussed the neuroblocking potency using
the Mulliken charges of the ligands. However, consideration
involving the atoms of the ligand placing a protonated basic
residue on the receptor segment should be necessary for the
precise evaluation of the ligand receptor. Further investigation
on this point will be performed in the future.
In conclusion, most of the variations of the pharmacophore
structure in neonicotinoid afforded insecticidal activity against
the AmC at 0.1-30 nM concentrations under synergistic
conditions. The potency was dependent on the heterocyclic ring
conjugated with NNO2 or CHNO2. The neuroblocking potency
was proportional both to log P and the Mulliken charge on the
nitro oxygen atom. For the insecticidal potency, the high
neuroblocking effect gave the potent insecticidal effect when
the hydrophobicity and structure in the ClPy or ClTh moiety
were separately considered.
(12) Matsuda, K.; Buckingham, S. D.; Kleier, D.; Rauh, J. J.; Grauso,
M.; Sattelle, D. B. Neonicotinoids: insecticides acting on insect
nicotinic acetylcholine receptors. Trends Pharmacol. Sci. 2001,
22, 573-580.
(13) Tomizawa, M.; Zhang, N.; Durkin, A.; Olmstead, M. M.; Casida,
J. E. The neonicotinoid electronegative pharmacophore plays the
crucial role in the high affinity and selectivity for the drosophila
nicotinic receptor: an anomaly for the nicotinoid cation-π
interaction model. Biochemistry 2003, 42, 7819-7827.
(14) Nakayama, A.; Sukekawa, M. Quantitative correlation between
molecular similarity and receptor-binding activity of neonicoti-
noid insecticides. Pestic. Sci. 1998, 52, 104-110.
(15) Okazawa, A.; Akamatsu, M.; Ohoka, A.; Nishiwaki, H.; Cho,
W. J.; Nakagawa, Y.; Nishimura, K.; Ueno, T. Prediction of the
binding mode of imidacloprid and related compounds to house-
fly head acetylcholine receptors using three-dimensional QSAR
analysis. Pestic. Sci. 1998, 54, 134-144.
ABBREVIATIONS USED
nAChR, nicotinic acetylcholine receptor; H-bond, hydrogen
bond; n-charged, negatively charged; p-charged, positively
charged; e-dificient, electron-deficient; ClPy, 2-chloro-5-pyridyl;
ClTh, 2-chloro-5-thiazolyl; AmC, American cockroach; PB,
piperonyl butoxide; (PB), NIA 16388, propargyl propyl ben-
zenephosphonate; MLD, minimum lethal dose; BC, neurob-
locking concentration.

818 J. Agric. Food Chem., Vol. 55, No. 3, 2007
Kagabu et al.
(16) Silverman, R. B. The Organic Chemistry of Drug Design and
Drug Action, 2nd ed.; Elsevier: Amsterdam, The Netherlands,
2004; Chapter 2.
(35) Erdelen, C.; Kraemer, W.; Brueggen, K. U. Synergistic insec-
ticidal, nematocidal and acaricidal compositions. Ger. Offen DE
19519007 A1, 1996.
(17) Leo, A.; Hansch, C.; Hoekman, D. Exploring QSAR, Vol. 2,
Hydrophobicity, Electronic, and Steric Constants; American
Chemical Society: Washington, DC, 1995.
(36) Kagabu, S.; Kato, C.; Nishimura, K. Insecticidal and neurob-
locking activities toward American cockroach (Periplaneta
americana L.) of imidacloprid metabolites, 5-hydroxy-, 4,5-
dihydroxy- and 4,5-dehydroimidacloprid. J. Pestic. Sci. 2004,
29, 376-379.
(37) Suzuki, T.; Miyamoto, J. Metabolism of tetramethrin in houseflies
and rats in vitro. Pestic. Biochem. Physiol. 1974, 4, 86-97.
(38) Nishiwaki, H.; Sato, K.; Nakagawa, Y.; Miyashita, M.; Miya-
gawa, H. Metabolism of imidacloprid in houseflies. J. Pestic
Sci. 2004, 29, 110-116.
(39) Frisch, J. R.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, M. J.; Zakrzewski, V. G.; Montgom-
ery, J. A.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam,
J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.;
Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;
Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson,
G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski,
J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin,
R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.;
Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A.
GAUSSIAN98, revision A.7; Gaussian, Inc.: Pittsburgh, PA,
1998.
(18) Nishimura, K.; Kiriyama, K.; Kagabu, S. Quantitative structure-
activity relatuinships of imidacloprid and its analogs with
substituents at the C5 position on the pyridine ring in the
neuroblocking activity. J. Pestic. Sci. 2006, 31, 110-115.
(19) Nishimura, K.; Kanda, Y.; Okazawa, A.; Ueno, T. Relationship
between insecticidal and neurophysiological sctivities of imida-
cloprid and related compounds. Pestic. Biochem. Physiol. 1994,
50, 51-59.
(20) Kagabu, S.; Ito, N.; Imai, R.; Hieda, Y.; Nishimura, K. Effects
of halogen introduction at the 5th position on the imidacloprid
pyridine ring upon insecticidal activity. J. Pestic. Sci. 2005, 30,
409-413.
(21) Nishimura, K.; Tanaka, M.; Iwaya, K.; Kagabu, S. Relationship
between insecticidal and nerve-excitatory activities of imidaclo-
prid and its alkylated congeners at the imidazolidine NH site.
Pestic. Biochem. Physiol. 1998, 62, 172-178.
(22) Kiriyama, K.; Iwaya, K.; Kagabu, S.; Nishimura, K. Relationship
between insecticidal and nerve activities of an imidacloprid-
related nitromethylene compound and its N-alkyl congeners.
J. Pestic. Sci. 2001, 26, 55-59.
(23) Kagabu, S.; Azuma, A.; Nishimura, K. Insecticidal and neurob-
locking activities of thiacloprid and its acyclic analogues and
related cyanoguanidine derivatives. J. Pestic. Sci. 2002, 27, 267-
271.
(24) Kiriyama, K.; Nishimura, K. Structural effects of dinotefuran
and analogues in insecticidal and neural activities. Pest Manag.
Sci. 2002, 58, 669-676.
(25) Kiriyama, K.; Itazu, Y.; Kagabu, S.; Nishimura, K. Insecticidal
and neuroblocking activities of acetamiprid and related com-
pounds. J. Pestic. Sci. 2003, 28, 8-17.
(26) Kagabu, S.; Kiriyama, K.; Nishiwaki, H.; Kumamoto, Y.; Tada,
T.; Nishimura, K. Asymmetric chloronicotinyl insecticide, 1-[1-
(6-chloro-3-pyridyl)ethyl]-2-nitroiminoimidazolidine: prepara-
tion, resolution and biological activities toward insects and their
nerve preparations. Biosci., Biotechnol., Biochem. 2003, 67,
980-988.
(40) Hansch, C.; Fujita, T. F-σ analysis. A method for the correlation
of biological activity and chemical structure. J. Am. Chem. Soc.
1964, 86, 1616-1626.
(41) Nakagawa, S.; Okajima, N.; Nishimura, K.; Fujita, T.; Nakajima,
M. Quantitative structure-activity studies of substituted benzyl
crysanthemates. 2. Physicochemical substituent effects and
neurophysiological and symptomatic activities against American
cockroaches. Pestic. Biochem. Physiol. 1982, 17, 259-270.
(42) Nakagawa, S.; Okajima, N.; Kitahaba, T.; Nishimura, K.; Fujita,
T.; Nakajima, M. Quantitative structure-activity studies of
substituted benzyl crysanthemates. 1. Correlations between
symptomatic and neurophysiological activities against American
cockroaches. Pestic. Biochem. Physiol. 1982, 17, 243-258.
(43) Honda, H.; Tomizawa, M.; Casida, J. E. Neonicotinoid metabolic
activation and inactivation established with coupled nicotinic
receptor-CYP3A4 and -aldehyde oxidase systems. Toxicol. Lett.
2006, 161, 108-114.
(27) Kagabu, S.; Murata, N.; Hibino, R.; Hanzawa, M.; Nishimura,
K. Insecticidal and neuroblocking activities of thiamethoxam-
type compounds in the American cockroach (Periplaneta ameri-
cana L.). J. Pestic. Sci. 2005, 30, 111-115.
(44) Chao, S. L.; Casida, J. E. Interaction of imidacloprid metabolites
and analogs with the nicotinic acetylcholine receptors or mouse
brain in relation to toxicity. Pestic. Biochem. Physiol. 1997, 58,
77-88.
(28) Kagabu, S.; Yokoyama, K.; Iwaya, K.; Tanaka, M. Imidacloprid
and related compounds: structure and water solubility of N-alkyl
derivatives of imidacloprid. Biosci., Biotechnol., Biochem. 1998,
62, 1216-1224.
(45) Zhang, N.; Tomizawa, M.; Casida, J. E. R-Nitro ketone as an
electrophile and nucleophile: synthesis of 3-substituted 2-ni-
tromethylenetetrahydrothiophene and tetrahydrofuran as Droso-
phila nicotinic receptor probes. J. Org. Chem. 2004, 69, 876-
881.
(46) Boelle, J.; Schneider, R.; Gerardin, P.; Loubinoux, B.; Maien-
fisch, P.; Rindlisbacher, A. Synthesis and insecticidal evaluation
of imidacloprid analogs. Pestic. Sci. 1998, 54, 304-307.
(47) Bondi, A. van der Waals volumes and radii. J. Phys. Chem. 1976,
80, 1799-1806.
(48) Nauen, R.; Reckmann, U.; Armborst, S.; Stupp, H. P.; Elbert,
A. Whitefly-active metabolites of imidacloprid: biological
efficacy and translocation in cotton plants. Pestic. Sci. 1999, 55,
265-271.
(29) Zhang, A.; Kayser, H.; Maienfisch, P.; Casida, J. E. Insect
nicotinic acetylcholine receptor: conserved neonicotinoid speci-
ficity of [3H]imidacloprid binding site. J. Neurochem. 2000, 75,
1294-1303.
(30) Zhang, N.; Tomizawa, M.; Casida, J. E. Structural features of
azidopyridinyl neonicotinoid probes conferring high affinity and
selectivity for mammalian R4â2 and Drosophila nicotinic
receptors. J. Med. Chem. 2002, 45, 2832-2840.
(31) Shiokawa, K.; Tsuboi, S.; Kagabu, S.; Moriya, K. Nitromethylene
derivatives, intermediates, and their preparation as insecticides.
Eur. Pat. Appl. EP 163855 A1, 1985.
(32) Shiokawa, K.; Tsuboi, S.; Kagabu, S.; Moriya, K. Heterocyclic
compounds. Eur. Pat. Appl. EP 192060 A1, 1986.
(33) Shiokawa, K.; Tsuboi, S.; Sasaki, S.; Moriya, K.; Hattori, Y.;
Shibuya, K. Preparation of (cyano- and nitromethynyl)- and
(cyano- and nitroimino)heterocyclic compounds as insecticides.
Eur. Pat. Appl. EP 259738 A2, 1988.
Received for review August 14, 2006. Revised manuscript received
December 4, 2006. Accepted December 5, 2006. This work was partly
supported by a Grant-in Aid to S.K. for Scientific Research (No.
17580094) from the Japan Society for the Promotion of Science.
(34) Maienfisch, P.; Gonda, J.; Jacob, O.; Gsell, L. Novel 2-nitro-
methylene/2-cyanoimino/2-nitroimino-pyrrolidines and -piper-
idines, intermediates, and their use as pesticides. WO 94-EP963,
1994.
JF0623440