Synthesis of imidacloprid derivatives with a chiral alkylated imidazolidine ring and evaluation of their insecticidal activity and affinity to the nicotinic acetylcholine receptor

Article abstract of DOI:10.1016/j.bmc.2012.09.007

A series of imidacloprid (IMI) derivatives with an alkylated imidazolidine ring were asymmetrically synthesized to evaluate their insecticidal activity against adult female housefly, Musca domestica, and affinity to the nicotinic acetylcholine receptor of the flies. The bulkier the alkyl group, the lower was the receptor affinity, but the derivatives methylated and ethylated at the R-5-position of the imidazolidine ring were equipotent to the unsubstituted compound. Quantitative structure-activity relationship (QSAR) analysis of the receptor affinity demonstrated that the introduction of a substituent into the imidazolidine ring was fundamentally disadvantageous, but the introduction of a substituent at the R-5-position was permissible in the case of its small size. The binding model of the synthesized derivatives with the receptor supported the QSAR analysis, indicating the existence of space for a short alkyl group around the R-5-position in the ligand-binding site. In addition, positive correlation was observed between the insecticidal activity and receptor affinity, suggesting that the receptor affinity was the primary factor in influencing the insecticidal activity even if the imidazolidine ring was modified.

Full text of DOI:10.1016/j.bmc.2012.09.007

Bioorganic & Medicinal Chemistry 20 (2012) 6305–6312
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis of imidacloprid derivatives with a chiral alkylated imidazolidine
ring and evaluation of their insecticidal activity and affinity to the
nicotinic acetylcholine receptor
Hisashi Nishiwaki a, , Mituhiro Kuriyama a, Hikaru Nagaoka a, Akira Kato a, Miki Akamatsu b,
⇑
Satoshi Yamauchi a, Yoshihiro Shuto a
a Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama, Ehime 790-8566, Japan
b Graduate School of Agriculture, Kyoto University, Kita-Shirakawa, Sakyo-Ku, Kyoto 606-8502, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of imidacloprid (IMI) derivatives with an alkylated imidazolidine ring were asymmetrically syn-
thesized to evaluate their insecticidal activity against adult female housefly, Musca domestica, and affinity
to the nicotinic acetylcholine receptor of the flies. The bulkier the alkyl group, the lower was the receptor
affinity, but the derivatives methylated and ethylated at the R-5-position of the imidazolidine ring were
equipotent to the unsubstituted compound. Quantitative structure–activity relationship (QSAR) analysis
of the receptor affinity demonstrated that the introduction of a substituent into the imidazolidine ring
was fundamentally disadvantageous, but the introduction of a substituent at the R-5-position was per-
missible in the case of its small size. The binding model of the synthesized derivatives with the receptor
supported the QSAR analysis, indicating the existence of space for a short alkyl group around the R-5-
position in the ligand-binding site. In addition, positive correlation was observed between the insecticidal
activity and receptor affinity, suggesting that the receptor affinity was the primary factor in influencing
the insecticidal activity even if the imidazolidine ring was modified.
Received 19 August 2012
Revised 2 September 2012
Accepted 5 September 2012
Available online 13 September 2012
Keywords:
Neonicotinoids
Nicotinic acetylcholine receptor
Comparative molecular field analysis
Musca domestica
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Our previous SAR studies demonstrated that some steric and elec-
trostatic factors around the pyridine and imidazolidine rings influ-
Neonicotinoids, including imidacloprid (1, IMI, Fig. 1), are one of
the widely used neuroactive insecticides. They specifically interact
with the nicotinic acetylcholine receptor (nAChR) of insects, which
results in superior selective toxicity to insects over vertebrates. A
significant number of researches have been performed to explore
the novel bioactive structure of neonicotinoids since the develop-
ment and launch of IMI.1–7 To clarify the structure–activity rela-
tionship (SAR) of neonicotinoids and to explore compounds with
high potency, we also synthesized various neonicotinoid deriva-
tives with the nitromethylene group. These include derivatives
having a substituted phenyl ring instead of a pyridine ring,8 various
alkyl groups attached at the N3 atom of the imidazolidine ring,9
and enlarged rings such as diazacyclohexane and diazacyclohep-
tane with some modifications instead of the imidazolidine ring.10
ence the receptor binding. However, most of the synthesized
compounds in our studies were less potent than the basic nitrom-
ethylene derivative 2 (CH-IMI, Fig. 1). Recently, we asymmetrically
synthesized four methylated imidacloprid derivatives to examine
whether substitutions to the ethylene moiety of the imidazolidine
ring influence bioactivity such as insecticidal activity and affinity
Abbreviations: CH-IMI, nitromethylene analogue of imidacloprid; IMI, imida-
cloprid; LBD, ligand binding domain; Ls, Lymnaea stagnalis; nAChR, nicotinic
acetylcholine receptor; NIA, propargyl propyl phenylphosphonate (Niagara 16388);
QSAR, quantitative structure–activity relationship; SAR, Structure–activity
relationship.
⇑
Corresponding author. Tel.: +81 89 946 9973; fax: +81 89 977 4364.
E-mail address: nishiwaki.hisashi.mg@ehime-u.ac.jp (H. Nishiwaki).
Figure 1. Chemical structures of neonicotinoid derivatives with an alkylated
imidazolidine ring.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.09.007

6306
H. Nishiwaki et al. / Bioorg. Med. Chem. 20 (2012) 6305–6312
Figure 2. Synthetic route of alkylated imidacloprid derivatives. (a) Lithium aluminum hydride/anhydrous THF; (Boc)2O/THF; (b) phthalimide, diethyl azodicarboxylate,
triphenylphosphine/anhydrous THF; (c) concentrated HCl/THF; 2-chloro-5- (chloromethyl) pyridine, Et3N/acetonitrile; (d) hydrazine hydrate/EtOH; 1,1-bis(methylthio)-2-
nitroethylene, K2CO3/EtOH; (e) hydrazine hydrate/EtOH; 2-chloro-5- (chloromethyl) pyridine, Et3N/acetonitrile; (f) concentrated HCl/THF; 1,1-bis(methylthio)-2-
nitroethylene, K2CO3/EtOH.
to the nAChR (Figs. 1, 3–6).11 As a result, the compound with a
methyl group at the 5-position of the imidazolidine ring (com-
pound 3) exhibited intrinsic activity comparable to the unsubsti-
tuted compound 2. These findings motivated us to clarify the size
of the substituent attached on the imidazolidine ring that would
be considered permissible for its interaction with the receptor
and to verify if these alkylated CH-IMI derivatives could exert high
insecticidal activity.
In this study, a series of imidacloprid derivatives with various
alkylated imidazolidine rings (Fig. 1, 7–22) were synthesized to
evaluate their receptor affinity. The SAR of the modified imidazoli-
dine ring was then quantitatively analyzed using the Hansch–Fuj-
ita method, conventional quantitative SAR (QSAR) method, and
comparative molecular field analysis (CoMFA) that is a three-
dimensional QSAR technique. In addition, a ligand-binding model
of the receptor combined with a test chemical was constructed
to discuss the validation of results from the QSAR analyses. The
insecticidal activities of the synthesized compounds were also
evaluated to elucidate the relationship between their bioactivities.
tion-modified imidazolidine ring, the t-butoxycarbonyl group was
deprotected using concd HCl, whereas hydrazine monohydrate
deprotected phthalimide to amino group for synthesizing the
compound with the 4-position-modified imidazolidine ring.
2-Chloro-5- (chloromethyl) pyridine was then attached to the
resultant amino group, followed by the deprotection of the other
protected amino group. The target compounds were finally synthe-
sized via cyclization using the resultant diamine and 1,1-
bis(methylthio)-2-nitroethylene. Reagents used for the syntheses
were purchased from Wako Pure Chemical Industries, Ltd (Osaka,
Japan), Nacalai Tesque, Inc (Kyoto, Japan), Tokyo Chemical Industry
Co, Ltd (Tokyo, Japan), and Aldrich Chemical Co (Milwaukee, WI,
USA). The metabolic inhibitor, NIA 16388 (NIA; propargyl propyl
phenylphosphonate), was our stock sample.11 The 1H and 13C
NMR analyses were performed using a JEOL ECS-400 NMR spec-
trometer in deuterochloroform (CDCl3) with tetramethylsilane as
the internal standard. The authenticity of the final compounds
was also confirmed by HRMS using Xevo Q-TOFMS (Waters, UK).
Melting points of the compounds were measured with a Yanaco
melting point apparatus (Kyoto, Japan) and were uncorrected.
Optical rotation values were evaluated by using a P-2100 polarim-
eter (Jasco, Tokyo, Japan).
2. Experimental
2.1. Insects
2.2.1. R-1-(6-Chloro-3-pyridylmethyl)-5-ethyl-2-nitromethylene-
An insecticide-susceptible strain of the housefly (Musca domes-
tica L., Takatsuki strain) was reared at 25 °C in our laboratory.
imidazolidine (7)
Mp 139–140. ½a 2D5
ꢀ
+56 (c 0.02, CHCl3). NMR dH (CDCl3): 0.92
(3H, t, J = 7 Hz), 1.58 (1H, m), 1.78 (1H, m), 3.49 (1H, dd,
J = 10 Hz, J = 8 Hz), 3.78 (1H, m), 3.89 (1H, t, J = 8 Hz), 4.27 (1H, d,
J = 16 Hz), 4.37 (1H, d, J = 16 Hz), 6.56 (1H, s), 7.37 (1H, d,
J = 8 Hz), 7.55 (1H, dd, J = 8 Hz, J = 2 Hz), 8.28 (1H, d, J = 2 Hz),
8.71 (1H, br). NMR dC (CDCl3): 8.7, 24.8, 44.2, 47.3, 61.1, 96.5,
124.9, 129.5, 137.5, 148.3, 151.8, 159.4. ESIMS m/z [M+H]+: calcd
for C12H16N4O2Cl, 283.0962; found, 283.0967.
2.2. Chemicals
Compounds 7–22 (Fig. 1) were newly synthesized through the
synthetic scheme reported earlier (Fig. 2).11 In brief, the amino
group of each 2-aminobutanol (for compounds 7–10) or an appro-
priate amino alcohol (for others), which was prepared by the
reduction of corresponding amino acid (norvaline for compounds
11–14, valine for compounds 15–18, and leucine for compounds
19–22) with lithium aluminium hydride, was protected with the
t-butoxycarbonyl group. The hydroxy group was changed to
phthalimide using triphenyl phosphine, diethyl azodicarboxylate,
and phthalimide. For synthesizing the compounds with the 5-posi-
2.2.2. S-1-(6-Chloro-3-pyridylmethyl)-5-ethyl-2-nitromethylene-
imidazolidine (8)
Mp 137–138. ½a 2D5
ꢁ62 (c 0.03, CHCl3). The NMR spectral data
ꢀ
agreed with those of its enantiomer 7. ESIMS m/z [M+H]+: calcd
for C12H16N4O2Cl, 283.0962; found, 283.0973.

H. Nishiwaki et al. / Bioorg. Med. Chem. 20 (2012) 6305–6312
6307
2.2.3. R-1-(6-Chloro-3-pyridylmethyl)-4-ethyl-2-nitromethylene-
2.2.11. R-1-(6-Chloro-3-pyridylmethyl)-4-isopropyl-2-
imidazolidine (9)
nitromethylene-imidazolidine (17)
Mp 138–139. ½a 2D5
ꢀ
ꢁ21 (c 0.04, CHCl3). NMR dH (CDCl3): 0.97
Mp 114–115. ½a 2D5
ꢁ47 (c 0.03, CHCl3). NMR dH (CDCl3): 0.91
ꢀ
(3H, t, J = 8 Hz), 1.67 (2H, m), 3.18 (1H, dd, J = 10 Hz, J = 7 Hz),
3.64 (1H, t, J = 10 Hz), 4.02 (1H, m), 4.30 (2H, s), 6.63 (1H, s), 7.38
(1H, d, J = 8 Hz), 7.55 (1H, dd, J = 8 Hz, J = 2 Hz), 8.29 (1H, d,
J = 2 Hz), 8.82 (1H, br). NMR dC (CDCl3): 9.5, 28.1, 46.5, 53.3, 56.2,
96.2, 124.9, 129.1, 137.9, 148.8, 152.0, 158.6. ESIMS m/z [M+H]+:
calcd for C12H16N4O2Cl, 283.0962; found, 283.0958.
(3H, d, J = 7 Hz), 0.97 (3H, d, J = 7 Hz), 1.82 (1H, m), 3.25 (1H, dd,
J = 10 Hz, J = 8 Hz), 3.62 (1H, t, J = 6 Hz), 3.85 (1H, m), 4.32 (2H,
s), 6.64 (1H, s), 7.38 (1H, d, J = 8 Hz), 7.57 (1H, dd, J = 8 Hz,
J = 2 Hz), 8.29 (1H, d, J = 2 Hz), 8.84 (1H, br). NMR dC (CDCl3):
17.8, 18.1, 32.5, 46.5, 51.7, 60.7, 96.1, 124.9, 129.1, 137.9, 148.8,
152.0, 158.7. ESIMS m/z [M+H]+: calcd for C13H18N4O2Cl,
297.1118; found, 297.1114.
2.2.4. S-1-(6-Chloro-3-pyridylmethyl)-4-ethyl-2-nitromethylene-
imidazolidine (10)
2.2.12. S-1-(6-Chloro-3-pyridylmethyl)-4-isopropyl-2-
Mp 138–139. ½a 2D5
ꢀ
+20 (c 0.11, CHCl3). The NMR spectral data
nitromethylene-imidazolidine (18)
agreed with those of its enantiomer 9. ESIMS m/z [M+H]+: calcd
for C12H16N4O2Cl, 283.0962; found, 283.0968.
Mp 115–116. ½a 2D5
ꢀ
+51 (c 0.03, CHCl3). The NMR spectral data
agreed with those of its enantiomer 17. ESIMS m/z [M+H]+: calcd
for C13H18N4O2Cl, 297.1118; found, 297.1119.
2.2.5. R-1-(6-Chloro-3-pyridylmethyl)-5-n-propyl-2-nitrometh-
ylene-imidazolidine (11)
2.2.13. R-1-(6-Chloro-3-pyridylmethyl)-5-isobutyl-2-
Mp 123–125. ½a 2D5
ꢀ
+98 (c 0.04, CHCl3). NMR dH (CDCl3): 0.95
nitromethylene-imidazolidine (19)
(3H, d, J = 8 Hz), 1.30 (2H, m), 1.53 (1H, m), 1.71 (1H, m), 3.48
(1H, t, J = 9 Hz), 3.80 (1H, m), 3.89 (1H, t, J = 9 Hz), 4.28 (1H, d,
J = 16 Hz), 4.36 (1H, d, J = 16 Hz), 6.55 (1H, s), 7.36 (1H, d,
J = 8 Hz), 7.55 (1H, dd, J = 8 Hz, J = 2 Hz), 8.29 (1H, d, J = 2 Hz),
8.71 (1H, br). NMR dC (CDCl3): 13.9, 18.1, 34.0, 44.3, 47.8, 60.1,
96.6, 124.9, 129.6, 137.5, 148.3, 151.7, 159.4. ESIMS m/z [M+H]+:
calcd for C13H18N4O2Cl, 297.1118; found, 297.1117.
Mp 131–133. ½a 2D5
ꢀ
+75 (c 0.16, CHCl3). NMR dH (CDCl3): 0.87
(3H, d, J = 6 Hz), 0.96 (3H, d, J = 6 Hz), 1.55 (3H, m), 3.46 (1H, t,
J = 8 Hz), 3.85 (1H, m), 3.91 (1H, t, J = 8 Hz), 4.28 (1H, d,
J = 16 Hz), 4.35 (1H, d, J = 16 Hz), 6.53 (1H, s), 7.36 (1H, d,
J = 8 Hz), 7.55 (1H, dd, J = 8 Hz, J = 2 Hz), 8.29 (1H, d, J = 2 Hz),
8.71 (1H, br). NMR dC (CDCl3): 21.5, 23.8, 25.1, 41.1, 44.4, 48.3,
59.0, 96.7, 124.9, 129.5, 137.5, 148.3, 151.8, 159.3. ESIMS m/z
[M+H]+: calcd for C14H20N4O2Cl, 311.1275; found, 311.1286.
2.2.6. S-1-(6-Chloro-3-pyridylmethyl)-5-n-propyl-2-nitrometh-
ylene-imidazolidine (12)
2.2.14. S-1-(6-Chloro-3-pyridylmethyl)-5-isobutyl-2-
Mp 114–116. ½a 2D5
ꢀ
ꢁ94 (c 0.02, CHCl3). The NMR spectral data
nitromethylene-imidazolidine (20)
agreed with those of its enantiomer 11. ESIMS m/z [M+H]+: calcd
Mp 133–135. ½a 2D5
ꢁ71 (c 0.10, CHCl3). The NMR spectral data
ꢀ
for C13H18N4O2Cl, 297.1118; found, 297.1115.
agreed with those of its enantiomer 19. ESIMS m/z [M+H]+: calcd
for C14H20N4O2Cl, 311.1275; found, 311.1275.
2.2.7. R-1-(6-Chloro-3-pyridylmethyl)-4-n-propyl-2-nitrometh-
ylene-imidazolidine (13)
2.2.15. R-1-(6-Chloro-3-pyridylmethyl)-4-isobutyl-2-
Mp 111–113. ½a 2D5
ꢀ
ꢁ34 (c 0.04, CHCl3). NMR dH (CDCl3): 0.97
nitromethylene-imidazolidine (21)
(3H, d, J = 8 Hz), 1.38 (2H, m), 1.57 (1H, m), 1.66 (1H, m), 3.17
(1H, dd, J = 8 Hz, J = 6 Hz), 3.65 (1H, t, J = 10 Hz), 4.06 (1H, m),
4.30 (2H, s), 6.63 (1H, s), 7.38 (1H, d, J = 8 Hz), 7.56 (1H, dd,
J = 8 Hz, J = 2 Hz), 8.29 (1H, d, J = 2 Hz), 8.81 (1H, br). NMR dC
(CDCl3): 13.8, 18.7, 37.2, 46.5, 53.9, 54.9, 96.3, 124.9, 129.2,
137.9, 148.8, 151.9, 158.6. ESIMS m/z [M+H]+: calcd for
C13H18N4O2Cl, 297.1118; found, 297.1110.
Mp 100–103. ½a 2D5
ꢁ35 (c 0.20, CHCl3). NMR dH (CDCl3): 0.95
ꢀ
(3H, d, J = 2 Hz), 0.97 (3H, d, J = 2 Hz), 1.43 (1H, m), 1.65 (2H, m),
3.13 (1H, dd, J = 9 Hz, J = 8 Hz), 3.65 (1H, t, J = 9 Hz), 4.12 (1H, m),
4.26 (1H, d, J = 16 Hz), 4.32 (1H, d, J = 16 Hz), 6.62 (1H, s), 7.38
(1H, d, J = 8 Hz), 7.56 (1H, dd, J = 8 Hz, J = 2 Hz), 8.29 (1H, d,
J = 2 Hz), 8.81 (1H, br). NMR dC (CDCl3): 22.2, 22.7, 25.2, 44.2,
46.5, 53.4, 54.4, 96.3, 124.9, 129.1, 137.9, 148.8, 152.0, 158.5.
ESIMS m/z [M+H]+: calcd for C14H20N4O2Cl, 311.1275; found,
311.1281.
2.2.8. S-1-(6-Chloro-3-pyridylmethyl)-4-n-propyl-2-nitrometh-
ylene-imidazolidine (14)
Mp 108–109. ½a 2D5
ꢀ
+32 (c 0.15, CHCl3). The NMR spectral data
2.2.16. S-1-(6-Chloro-3-pyridylmethyl)-4-isobutyl-2-
agreed with those of its enantiomer 13. ESIMS m/z [M+H]+: calcd
nitromethylene-imidazolidine (22)
Mp 102–105. ½a 2D5
ꢀ
+30 (c 0.20, CHCl3). The NMR spectral data
for C13H18N4O2Cl, 297.1118; found, 297.1122.
agreed with those of its enantiomer 21. ESIMS m/z [M+H]+: calcd
2.2.9. R-1-(6-Chloro-3-pyridylmethyl)-5-isopropyl-2-nitrometh-
for C14H20N4O2Cl, 311.1275; found, 311.1273.
ylene-imidazolidine (15)
Mp 163–164. ½a 2D5
ꢀ
+129 (c 0.04, CHCl3). NMR dH (CDCl3): 0.87
2.3. Evaluation of receptor binding affinity
(3H, d, J = 7 Hz), 0.91 (3H, d, J = 7 Hz), 2.11 (1H, m), 3.58 (1H, dd,
J = 10 Hz, J = 7 Hz), 3.75 (2H, m), 4.23 (1H, d, J = 16 Hz), 4.40 (1H,
d, J = 16 Hz), 6.64 (1H, s), 7.38 (1H, d, J = 8 Hz), 7.57 (1H, dd,
J = 8 Hz, J = 2 Hz), 8.28 (1H, d, J = 2 Hz), 8.72 (1H, br). NMR dC
(CDCl3): 14.4, 18.3, 27.7, 42.7, 44.0, 64.1, 96.4, 124.9, 129.5,
137.8, 148.5, 151.7, 159.4. ESIMS m/z [M+H]+: calcd for
Assay method was essentially the same as in our previous re-
ports.11,12 In the competitive binding inhibition assay, the housefly
head membrane fraction prepared according to our previous re-
port8 (3 mg protein mlꢁ1, 200
taining
concentrations (10ꢁ4 to 10ꢁ13 M) of the test compound. After
standing for 30 min, sodium phosphate buffer (50 l, 10 mM, pH
ll) was added to a plastic tube con-
a
dimethyl sulfoxide solution (2 l) of various
l
C
13H18N4O2Cl, 297.1118; found, 297.1125.
l
7.4) containing sodium chloride (50 mM), Triton X-100 (1 g lꢁ1
)
2.2.10. S-1-(6-Chloro-3-pyridylmethyl)-5-isopropyl-2-
and [3H]imidacloprid (100 nM) was added to the reaction mixture
and incubated for 1 h at ambient temperature. The reaction was
stopped by filtering the mixture through a glass filter (GF/B; What-
man International Ltd, Maidstone, UK) that was pretreated with
nitromethylene-imidazolidine (16)
Mp 165–166. ½a 2D5
ꢁ108 (c 0.06, CHCl3). The NMR spectral data
ꢀ
agreed with those of its enantiomer 15. ESIMS m/z [M+H]+: calcd
for C13H18N4O2Cl, 297.1118; found, 297.1110.

6308
H. Nishiwaki et al. / Bioorg. Med. Chem. 20 (2012) 6305–6312
Table 1
Insecticidal activity and receptor affinity of CH-IMI 2 and its alkylated derivatives 3–
22a
Ki (nM)
ED50 (pmol/fly)
2b
3b
4b
5b
6b
7
H
0.0367
0.0428
0.313
2.07
0.117
0.0626
0.340
0.427
R-5-Me
S-5-Me
R-4-Me
S-4-Me
R-5-Et
S-5-Et
R-4-Et
S-4-Et
R-5-n-Pro
S-5-n-Pro
R-4-n-Pro
S-4-n-Pro
R-5-Isopropyl
S-5-Isopropyl
R-4-Isopropyl
S-4-Isopropyl
R-5-Isobutyl
S-5-Isobutyl
R-4-Isobutyl
S-4-Isobutyl
5.64
0.562
0.0597 0.0115
18.8 3.40
31.0 1.27
24.4 8.53
0.258 0.137
25.2 12.7
211 38.4
73.7 12.9
1.11 0.29
2340 561
590 34.4
430 117
4.54 1.49
52.6 3.57
2750 1330
396 70.3
0.914 0.155
93.1 23.1
68.9 31.0
24.2 2.85
0.253 0.0813
45.3 18.3
433 130
79.9 5.74
33.1 8.52
>741 (22%)c
>891 (31%)c
33.7 7.88
6.01 1.62
603 71.9
1129 238
356 32.7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Figure 3. Superposed chemical structures of compounds 1–37.
a
The bioactivities of the compound 7–22 were represented as mean standard
error of means (n = 3).
b
The values of bioactivities of the compound were cited from Ref. 11.
The value in the parentheses was mortality at the concentration presented.
were included in the data set. The calculations for the optimization
and superposition of the test chemicals were performed using the
molecular modeling software package SYBYL ver. 7.1 (Tripos Asso-
ciates, Inc., St Louis, MO). For constructing the initial conforma-
tions of the compounds, the x-crystallographic data of
imidacloprid was utilized, which binds to the acetylcholine binding
protein of Lymnaea stagnails (Ls-AChBP; PDB code, 2ZJU).15 For all
rotatable bonds, a systematic search in SYBYL module was per-
formed. Structures were fully optimized by the semi-empirical
molecular orbital method PM3 to give relatively stable conforma-
tions. For each optimized conformer, electrostatic potential
charges were calculated using MNDO method (electron surface po-
tential charge). The optimized conformer of IMI was used as the
reference standard on which the other compounds were super-
posed. For the superposition of these compounds, four atoms were
selected on the basis of our earlier studies (Fig. 3),16,17 that is, the
nitrogen atoms of the pyridine ring and at the 1-position of the
imidazolidine ring, the carbon atom at the 2-position of the imi-
dazolidine ring, and the bridge carbon atom between the aromatic
and imidazolidine rings. For benzyl derivatives, their substituents
were orientated to the opposite site of the nitrogen atom of the
pyridine ring of IMI. All superposed conformers were placed in
the lattice with 2 Å spaces, and the potential energy fields of each
conformer were calculated at the lattice intersections. The electro-
static (Coulombic potential) and steric (Lennard-Jones potential)
field energies were calculated at each lattice point using sp3-car-
bon atom with a charge of +1.0 as a probe. The data for the receptor
affinity of the compounds 1–37 were in correlation with these
parameters by the partial least-squares method.
c
0.1% polyethyleneimine and mounted on a glass microanalysis
holder (KGS-25; Advantec Toyo, Tokyo, Japan). Filtration was per-
formed with the help of an air pump. The tube was rinsed with
1 ml of cold sodium phosphate buffer (10 mM, pH 7.4) containing
sodium chloride (50 mM), and the washings were immediately fil-
tered. The glass filter was washed four times with 2 ml of the same
cold buffer, and each glass filter was then transferred to a scintilla-
tion vial. Radioactivity was measured using a liquid scintillation
counter (ALOKA-1000; Aloka Co., Ltd., Tokyo, Japan) in 3 ml of
Aquasol-2 (Packard Instrument Co, Meriden, CT). From the concen-
tration–response curve, the molar concentration for 50% inhibition
(IC50) of [3H]imidacloprid binding to the receptor was calculated.
The Ki value was calculated according to the following equation
using PRISM software:
Ki ¼ IC50=1 þ ð½Lꢀ=KdÞ
in which [L] is the final concentration of the radioligand (20 nM)
and Kd (2.65 nM) is the dissociation constant of [3H]imidacloprid
for the receptor fraction. Ki values of the test compounds were ob-
tained from three separate assays performed in duplicate and are
listed in Table 1.
2.4. QSAR analyses using Hansch–Fujita method and CoMFA
The relationship between the structure of the test compounds
and the receptor affinity (presented as the reciprocal logarithm
of Ki values (pKi)) was quantitatively analyzed. For the Hansch–Fuj-
ita method, the steric parameter B5 was used, which is the maxi-
mum width parameter of the Verloop’s STERIMOL parameters,13
2.5. Construction of ligand-binding domain model of housefly
nAChR combined with IMI
and
1, from the intact B5 value of each substituent. The employed
B5 values were as follows: 0 (H), 1.04 (Me), 2.17 (Et), 2.49 (n-
Pr), 2.17 (isopropyl), and 3.45 (isobutyl). Superscript characters
on the B5 term indicate the substituted positions. The Iipr param-
eter was the indicator variable, being one for compounds 15–18
with an isopropyl group and zero for others. Linear regression anal-
ysis was performed using the QREG 2.05 software.14
DB5 was calculated by subtracting the B5 value of hydrogen,
On the basis of the crystallographic data of the Ls-AChBP bound
with IMI, the homology model of the ligand-binding domain (LBD)
in houseflies nAChR was constructed using the homology modeling
software PDFAMS Pro 2.0 (In-Silico Sciences, Inc., Tokyo, Japan)
according to a previous report.18 The amino acid sequence of the
D
D
subunit
6 (isoform II) of M. domestica AChR (GenBank ID
ABJ09669), which was most common among the six splicing vari-
ants in the subunit 6 of the housefly receptor,19 was aligned with
that of Ls-AChBP using PDFAMS, as shown in Figure 4. Using the
For CoMFA, not only compounds 1–22 but also compounds 23–
37 (Table 2), whose pKi values have been reported previously,12

H. Nishiwaki et al. / Bioorg. Med. Chem. 20 (2012) 6305–6312
6309
3.2. QSAR analyses
First, the relationship between the structure of the compounds
and their receptor affinity was quantitatively analyzed using Han-
sch–Fujita method, a conventional QSAR method, shown as the fol-
lowing equation:
pKi ¼ ꢁ1:22
D
BS54 ꢁ1:37
D
BR54 ꢁ1:08
D
BS55 ꢁ0:31 B5R5 ꢁ1:36Iipr þ 10:31 ð1Þ
D
ð0:66Þ
ð0:63Þ
ð0:33Þ
ð0:33Þ
ð0:33Þ
ð0:33Þ
n ¼ 21; SD ¼ 0:53; r2 ¼ 0:91; Fð515Þ ¼ 29:42 > 0:999
In this and following equations, n is the number of compounds,
SD is the standard deviation and r is the correlation coefficient. F is
the ratio between the regression and residual variances. The num-
ber in the parentheses was 95% confidence interval. In the multiple
regression analyses, the number of variables reportedly influences
the correlation efficient, known as ‘chance correlation’.23 Because
the number of test compounds for generating Eq. 1 was a little
smaller than that recommended by Topliss and Costello (approxi-
mately 30 compounds for 5 variables, whereas 21 compounds for
5 parameters in this study), Eq. 1 may be considered as the tenta-
tive result. However, the distinctive SAR characteristics possibly
become clear by correlating the activity of these derivatives with
the physicochemical parameters of their substituents. For example,
the steric parameter, B5, which is the maximum width parameter
Figure 4. Multiple sequence alignment of subunit
acetylcholine receptor of Musca domestica (MdS6II) with an acetylcholine binding
protein of Lymnaea stagnalis (Ls-AChBP).
6 (subtype II) of nicotinic
simulated annealing method,20 the three-dimensional structure of
the ligand-binding domain, which is formed with two adjacent
subunits extracted from the homopentamer subunits, was con-
structed. The coordinate of IMI was fixed during the simulated
annealing. The constructed model was then energy-minimized
for 5000 iteration of conjugated gradients using the force field
and partial charges of the molecular mechanics MMFF94.21,22
graphical image of the surface of the ligand-binding pocket was
created using the multichannel surface tool of the MOLCAD mod-
ule in the SYBYL software.
A
of the Verloop’s STERIMOL parameters, was employed (DB5 repre-
sents the value relative to that of the hydrogen atom). The equa-
tions formulated using other steric parameters such as molar
refractivity and Taft’s Es showed inferior correlation coefficients
2.6. Evaluation of insecticidal activity
to Eq. 1 (data not shown). All terms other than
tion were significant above the 99.9% level as examined by the t
test, whereas the significance level was 92.9% for
B5R5. The minus
sign of the coefficient of each B5 term in Eq. (1) suggests that the
D
B5R5 in this equa-
Assay method was essentially the same as in our previous re-
port.12 For evaluating the insecticidal activity, female houseflies
anesthetized using carbon dioxide were topically treated with
methanol containing synergists piperonyl butoxide and NIA16388
D
D
introduction of the substituents at each position is unfavorable for
[0.2% (w/v)]. After 1 h, 0.22 ll of 50% ethanol solutions containing
receptor affinity, particularly unfavorable at the R-4-position. The
a test chemical at various concentrations were injected in the dorsal
side of the thorax of reanesthetized flies. Insecticidal activity was
evaluated 1 h after injection. The ED50 values (effective dose for
inducing paralysis or death in 50% of the houseflies) were calculated
using probit transformation and are listed in Table 1.
R5
smallest coefficient of
DB5 term suggests that the introduction
of a substituent at the R-5-position could be less adverse for the
receptor affinity in the case of the introduction of a substituent
with a small size. The introduction of the Iipr term improved the
correlation coefficient, suggesting that the branched isopropyl
group decreased the affinity drastically.
3. Results and discussion
3.1. Receptor affinity
The statistical analysis of the receptor affinity for 37 compounds
using CoMFA was shown in the following Eq. (2).
pKi ¼ 6:47 þ ½CoMFA field termsꢀ
ð2Þ
The inhibition constant, Ki (nM), was evaluated as an indicator
of the affinity to the receptor (Table 1). Among the ethylated com-
pounds, R-5-Et compound 7 showed the equivalent potency to the
unsubstituted 2 and R-5-Me compound 3, whereas the potencies of
compounds 8–10, which were nearly similar, were 4- to 60-fold
lower than those of the corresponding Me compounds 4–6. The
extension from ethyl to n-propyl group lowered the receptor affin-
ity, but only to a little extent (1.3- to 6.8-fold lower than those of
the corresponding compounds). In contrast, the change from ethyl
to isopropyl group markedly decreased the affinity (compared to
isopropyl with Et; 17-, 130-, 45- and 18-fold less potent at R-5,
S-5, R-4 and S-4, respectively). By comparing the affinities of the
compounds with the largest substituent, isobutyl group, with those
of the compounds having a n-Pro group, more than 10-fold lower
affinities were observed for R-position-substituted compounds
(11, 13 vs 19, 21), whereas only 2- to 5-fold lower potencies were
observed for S-position-substituted compounds (12, 14 vs 20, 22).
These results demonstrated that the introduction of a substituent
into the imidazolidine ring was fundamentally disadvantageous,
but the introduction of a small substituent such as Me or Et groups
to the R-5-position was permissible. This suggests that the position
and type of substituent on the ethylene moiety of the imidazoli-
dine ring should be recognized accurately by the receptor.
CN ¼ 6; n ¼ 37; s ¼ 0:32; r2 ¼ 0:97;
Cross-validated½scv ¼ 1:14; q2 ¼ 0:61ꢀ;
RC½Steric ¼ 63:6%; Electrostatic ¼ 36:4%ꢀ
In this equation, CN indicates the number of latent variables,
and scv and q are the standard deviation and correlation coefficient
obtained from the leave-one-out cross-validation, respectively. RC
refers to the relative contribution of steric and electrostatic effects
to variations in receptor affinity. The pKi values calculated by Eq. 2
are shown in Table 2. In the preliminary analyses, the orientation
of the substituents of benzyl derivatives to the same site of the
nitrogen atom of the pyridine ring of IMI decreased the cross-val-
idated scv and q2 values (data not shown). The major steric and
electrostatic potential contour maps with IMI were drawn accord-
ing to Eq. 2 (Fig. 5). In Figure 5A, two large steric-favorable regions
appeared around the chlorine atom attached at the pyridine ring
and the R-5-positioned hydrogen atom of the imidazolidine ring,
whereas the sterically hindered regions appeared around the nitro-
gen atom of the pyridine ring and the 4-positioned hydrogen
atoms of the imidazolidine ring. Electrostatic interaction regions
(Fig. 5B) did not appear around the imidazolidine ring, probably
because such substituents were limited only to alkyl groups. CoM-

6310
H. Nishiwaki et al. / Bioorg. Med. Chem. 20 (2012) 6305–6312
A
B
Figure 5. Contour diagrams of steric (A) and electrostatic (B) fields with imidacloprid according to Eq. 2. In (A), the green and yellow areas indicate the steric-permissible and
unfavorable regions for the receptor affinity, respectively. In (B), the red and blue areas indicate the regions where the more negative and positive electrostatic interactions
with the receptor binding site increase the receptor affinity, respectively.
FA as well as the analysis using Hansch–Fujita method suggests
that the introduction of a substituent at the R-5-position could
be less adverse.
able in this pocket (Fig. 5A and B). Over the R-5-position of the imi-
dazolidine ring of IMI, a vacant space to accept a small substituent
was observed (Fig. 6A), and actually R-5-Me-CH-IMI was able to be
packed in this space with some room directed over methyl group
Previously, various research groups, including ours, have per-
formed 3D-QSAR analyses for neonicotinoids, particularly focusing
on pharmacophores such as the nitroimino group and the pyridine
ring.16,17,24–26 The results showed various important physicochem-
ical properties influencing the biological activities of the ligands;
that is, an electron-withdrawing group such as the nitro, cyano,
or trifluoromethyl group attached to the imino group and the chlo-
rine atom attached to the pyridine ring were important. Because
these analyses did not refer to the compounds with various asym-
metrically modified ethylene moiety of the imidazolidine ring, the
QSAR analysis for this part has not been elucidated in detail. Since
the launch of acyclic neonicotinoids such as nitenpyram and ace-
tamiprid as effective agrochemicals, the ethylene moiety has not
been considered as an important pharmacophore. Nonetheless,
our present study demonstrates that the ethylene moiety should
also play an important role in the receptor affinity.
(Fig. 6B). Employing the sequences of
a2
(GenBank ID
ABD37617) or 5 (GenBank ID ABY40460) subunit instead of that
a
of subunit 6 for construction of ligand binding domain model, al-
most the same vacant space appeared over the R-5-position (data
not shown). The ethyl group was also accepted in this region, but
n-propyl group was a little outside this region (data not shown).
As the substituent-acceptable space was surrounded by the aro-
matic amino acid residues such as Trp and Tyr, the hydrogen atoms
of the alkyl group could interact with the receptor surface via CH-
p
interaction. The binding model was consistent with the trend of
the in vitro activity and the present QSAR analyses.
Crystal structures of mollusk AChBP interacting with imidaclo-
prid have been resolved in 2008.15,27 On the basis of these struc-
tures, it was pointed out that the structurally important parts of
imidacloprid such as chlorine and nitrogen atoms of pyridine moi-
ety and nitro group bind the receptor. Matsuda et al. also suggested
that the hydrogen atoms of the imidazolidine ring (R3 in Fig. 1)
3.3. Construction of ligand binding domain model of housefly
nAChR combined with IMI
interact with the tryptophan residue of the receptor via CH–
p
interaction.18 The substantial decrease in the binding affinity of
R-4-substituted compounds is compatible with their suggestion.
The introduction of an alkyl group at the 4-position might separate
The surface of the ligand-binding site of the receptor was drawn
to examine whether the ligand used in this study could be accept-
Table 2
The reciprocal of the Ki values observed and calculated using the equation derived from the CoMFA analysis
No.
Compound
pKi Obsd.
CoMFA
pKi Calcd
No.
Compound
pKi Obsd.
CoMFA
pKi Calcd
D
D
1
2
3
4
5
6
7
8
Imidacloprid
H
R-5-Me
S-5-Me
R-4-Me
S-4-Me
R-5-Et
S-5-Et
R-4-Et
8.43
10.44
10.37
9.50
8.70
8.25
10.22
7.73
7.51
7.61
9.59
7.60
6.68
7.13
8.95
5.63
6.23
6.37
8.34
8.46
10.60
10.31
9.24
8.12
7.78
10.07
7.87
7.28
7.41
9.82
7.54
6.83
7.19
8.98
5.68
6.31
6.45
8.26
ꢁ0.03
ꢁ0.16
0.06
0.26
0.58
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
S-5-Isobutyl
R-4-Isobutyl
S-4-Isobutyl
4,4-(CH3)2
5,5-(CH3)2
phenyl
o-F
o-Cl
o-CH3
m-F
m-Cl
m-CH3
m-CH3O
m-CN
p-F
p-Cl
7.28
5.56
6.40
5.81
8.98
5.24
5.43
5.26
5.05
6.52
6.30
4.54
5.48
6.75
4.30
6.76
6.38
8.57
6.90
5.60
6.43
6.27
9.59
5.54
5.39
5.26
5.14
6.65
5.50
5.29
5.33
6.79
4.20
6.79
6.34
8.67
0.38
ꢁ0.04
ꢁ0.03
ꢁ0.46
ꢁ0.61
ꢁ0.30
0.04
0.47
0.15
ꢁ0.14
0.00
9
0.23
ꢁ0.09
ꢁ0.13
0.80
10
11
12
13
14
15
16
17
18
19
S-4-Et
0.20
R-5-n-Pro
S-5-n-Pro
R-4-n-Pro
S-4-n-Pro
R-5-Isopropyl
S-5-Isopropyl
R-4-Isopropyl
S-4-Isopropyl
R-5-Isobutyl
ꢁ0.23
0.06
ꢁ0.75
ꢁ0.15
ꢁ0.06
ꢁ0.03
ꢁ0.05
ꢁ0.08
ꢁ0.08
0.08
0.15
ꢁ0.04
0.10
ꢁ0.03
p-CH3
Olefin
0.04
ꢁ0.10

H. Nishiwaki et al. / Bioorg. Med. Chem. 20 (2012) 6305–6312
6311
A
B
Figure 6. The molecular surface (green) of the ligand-binding site of the housefly receptor. (A) the binding site interacts with IMI (violet). Five amino acids (Y88, S143, W144,
Y186 and Y193) constructing the vacant space over the R-5-position of the imidazolidine ring were also drawn together. (B) the binding model of the receptor with the R-5-
methylated CH-IMI as a ligand.
the nitro group, one of the important moiety interacting with the
receptor, from the interactive amino acid of the receptor.
Our binding model is consistent with the present QSAR analy-
ses, but not for all compounds reported previously. For example,
we cannot explain the high bioactivities of novel neonicotinoids
with a bulky modification.28 In relation to this, Tomizawa et al. re-
cently demonstrated that two binding modes of imidacloprid are
considerable in Lymnaea AChBP.29,30 If the size and structure of li-
gands change, it should be fully considered that the binding mode
would change. Because the ligand-binding site is orientated near
the open entrance of the receptor, the bulkier moiety might be
out of the ligand-binding region. The introduction of the alkyl
group might influence the conformation of the receptor or the
binding modes of the compounds, and we presently have no indi-
cation whether the binding mode of alkylated analogues with a
nitromethylene group to the receptor would change. However,
the R-5-position-specific bioactivity should be explained by the va-
cant space of our receptor model in this study.
3.4. Insecticidal activity and correlation between the biological
activities
ED50 values of the derivatives are shown in Table 1. Compared
with the unsubstituted CH-IMI 2, the methylated compounds 4–6
showed 3 to 5-fold lower activity, whereas R-5-methylated com-
pound 3 had 2-fold higher activity. The insecticidal activity of com-
pounds with an ethyl or a n-propyl group was over 100-fold lower
than that of the corresponding methylated compounds, except for
the S-4-ethylated, R-5-ethylated and n-propylated compounds (6
vs 10; 3 vs 7 and 11), whose activity fell only to 1/43, 1/14 and
1/4, respectively. The introduction of a wide substituent at any
positions of the imidazolidine ring drastically decreased the activ-
ity (ethyl group vs isopropyl group, and n-propyl group vs isobutyl
group). In particular, the ED50 values of the R-4- and S-5-isopropy-
lated and isobutylated compounds 16, 17, 20 and 21 were more
than 600 pmol/fly. Among the four 4- and 5-positions of the imi-
dazolidine ring, the introduction of the substituent at the R-4-posi-
tion was the most unfavorable for the insecticidal activity.
Positive correlation was observed between their insecticidal
activity (presented as the reciprocal logarithm of ED50 values
(pED50)) and receptor affinity (pKi) (Fig. 7; n = 19, r = 0.89), sug-
gesting that the receptor affinity was the main important factor
influencing the insecticidal activity even if the imidazolidine ring
was modified. In this study, we did not consider the metabolic ef-
fects in flies by a pre-application of the synergists. The ethylene
moiety of the imidazolidine ring was reportedly metabolized oxi-
Figure 7. Correlation between the receptor affinity (pKi) of the neonicotinoid
derivatives and their insecticidal activity (pED50). Open circles indicates compounds
16 and 17, which were omitted from the correlation analysis.
datively in houseflies.31 Using methylated CH-IMI derivatives, we
reported that the degree of the synergistic ratio changes according
to the substituted position.11 The modification of this part should
prevent the moiety from being metabolized. If a compound with
a modified imidazolidine ring has high receptor affinity and is
not easily metabolized in insects, it is anticipated to exert the high
insecticidal activity. In addition, we injected the test compounds
into the flies to prevent the absorption of the compounds through
the cuticle layer. Because the hydrophobicities of IMI and CH-IMI
are reportedly low,32 it is difficult for these neonicotinoids to exert
the insecticidal activity by the topical application. The introduction
of a hydrophobic substituent to the ethylene moiety is expected to
improve the absorption potency of the compounds through the
cuticle, resulting in the development of more effective insecticides.
4. Conclusion
We found the steric permissive region of the receptor over the
ethylene moiety of imidacloprid, which was surrounded by

6312
H. Nishiwaki et al. / Bioorg. Med. Chem. 20 (2012) 6305–6312
7. Sun, C.; Zhu, J.; Wang, H.; Jin, J.; Xing, J.; Yang, D. Eur. J. Med. Chem. 2011, 46, 11.
8. Nishiwaki, H.; Nakagawa, Y.; Takeda, D. Y.; Okazawa, A.; Akamatsu, M.;
Miyagawa, H.; Ueno, T.; Nishimura, K. Pest Manag. Sci. 2000, 56, 875.
9. Nishiwaki, H.; Nakagawa, Y.; Ueno, T.; Kagabu, S.; Nishimura, K. Pest Manag. Sci.
2001, 57, 810.
10. Kagabu, S.; Nishiwaki, H.; Sato, K.; Hibi, M.; Yamaoka, N.; Nakagawa, Y. Pest
Manag. Sci. 2002, 58, 483.
11. Nishiwaki, H.; Nagaoka, H.; Kuriyama, M.; Yamauchi, S.; Shuto, Y. Biosci.
Biotechnol. Biochem. 2011, 75, 780.
12. Nishiwaki, H.; Nakagawa, Y.; Kuwamura, M.; Sato, K.; Akamatsu, M.; Matsuda,
K.; Komai, K.; Miyagawa, H. Pest Manag. Sci. 2003, 59, 1023.
13. Hansch, C.; Leo, A. In Exploring QSAR; Heller, S. R., Ed.; ACS: Washington, DC,
1995. Chapter 3.
aromatic amino acid residues. Both electrostatic substituents and
aromatic and unsaturated substituents could interact with this re-
gion to exert higher receptor affinity through electrostatic, van der
Waals, p–p, and CH–p interactions. A more detailed SAR study of
this moiety will be conducted in the near future. The present anal-
yses are consistent with experimental results and may provide
useful information for designing new insecticides.
Acknowledgements
14. Asao, M.; Shimizu, R.; Nakao, K.; Fujita, T. QREG 2.05. Japan Chemistry Program
Exchange; Society of Computer Chemistry: Japan, 1997.
15. Ihara, M.; Okajima, T.; Yamashita, A.; Oda, T.; Hirata, K.; Nishiwaki, H.;
Morimoto, T.; Akamatsu, M.; Ashikawa, Y.; Kuroda, S.; Mega, R.; Kuramitsu, S.;
Sattelle, D. B.; Matsuda, K. Invert. Neurosci. 2008, 8, 71.
16. Okazawa, A.; Akamatsu, M.; Ohoka, A.; Nishiwaki, H.; Cho, W.-J.; Nakagawa, Y.;
Nishimura, K.; Ueno, T. Pestic. Sci. 1998, 54, 134.
17. Okazawa, A.; Akamatsu, M.; Nishiwaki, H.; Nakagawa, Y.; Miyagawa, H.;
Nishimura, K.; Ueno, T. Pest Manag. Sci. 2000, 56, 509.
We are grateful to Professors Hisashi Miyagawa and Yoshiaki
Nakagawa of Kyoto University for gifting radiolabeled imidaclo-
prid. This study, in part, was supported by Japan Society of Promot-
ing Science for Young Scientists (B) (HNi) and Inamori foundation
(HNi). A part of this study was performed at the INCS (Johoku
and Tarumi stations) of Ehime University.
18. Matsuda, K.; Kanaoka, S.; Akamatsu, M.; Sattelle, D. B. Mol. Pharmacol. 2009, 76,
1.
Supplementary data
19. Gao, J.-R.; Deacutis, J. M.; Scott, J. G. Insect Mol. Biol. 2007, 16, 325.
20. Kirkpatrick, S.; Gelatt, C. D.; Vecchi, M. P. Science 1983, 220, 671.
21. Halgren, T. A. J. Comput. Chem. 1999, 20, 720.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.09.007.
These data include MOL files and InChiKeys of the most important
compounds described in this article.
22. Halgren, T. A. J. Comput. Chem. 1999, 20, 730.
23. Topliss, J. G.; Costello, R. J. J. Med. Chem. 1972, 15, 1066.
24. Sung, N.-D.; Jang, S.-C.; Choi, K.-S. Bull. Korean Chem. Soc. 2006, 27, 1741.
25. Akamatsu, M. J. Agric. Food Chem. 2011, 59, 2909.
26. Li, Q.; Kong, X.; Xiao, Z.; Zhang, L.; Wang, F.; Zhang, H.; Li, Y.; Wang, Y. J. Mol.
Model. 2012, 18, 2279.
References and notes
27. Talley, T. T.; Harel, M.; Hibbs, R. E.; Radic, Z.; Tomizawa, M.; Casida, J. E.; Taylor,
P. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 7606.
28. Shao, X.; Lee, P. W.; Liu, Z.; Xu, X.; Li, Z.; Qian, X. J. Agric. Food Chem. 2011, 59,
2943.
29. Tomizawa, M.; Maltby, D.; Talley, T. T.; Durkin, K. A.; Medzihradszky, K. F.;
Burlingame, A. L.; Casida, J. E. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 1728.
30. Tomizawa, M.; Casida, J. E. J. Agric. Food Chem. 2011, 59, 2825.
31. Nishiwaki, H.; Sato, K.; Nakagawa, Y.; Miyashita, M.; Miyagawa, H. J. Pestic. Sci.
2004, 29, 110.
1. Kagabu, S. J. Agric. Food Chem. 2011, 59, 2887.
2. Ohno, I.; Tomizawa, M.; Miyazu, N.; Kushibiki, G.; Noda, K.; Hasebe, Y.; Durkin,
K. A.; Miyake, T.; Kagabu, S. Bioorg. Med. Chem. 2010, 20, 5933.
3. Tomizawa, M.; Kagabu, S.; Casida, J. E. J. Agric. Food Chem. 2011, 59, 2918.
4. Ye, Z.; Xia, S.; Shao, X.; Cheng, J.; Xu, X.; Xu, Z.; Li, Z.; Qian, X. J. Agric. Food Chem.
2011, 59, 10615.
5. Lu, S.; Shao, X.; Li, Z.; Xu, Z.; Zhao, S.; Wu, Y.; Xu, X. J. Agric. Food Chem. 2012, 60,
322.
6. Tomizawa, M.; Durkin, K. A.; Ohno, I.; Nagura, K.; Manabe, M.; Kumazawa, S.;
Kagabu, S. Bioorg. Med. Chem. Lett. 2011, 21, 3583.
32. Kagabu, S.; Yokoyama, K.; Iwaya, K.; Tanaka, M. Biosci. Biotechnol. Biochem.
1998, 62, 1216.