Enzyme-Linked immunosorbent assay for the pyrethroid permethrin

Article abstract of DOI:10.1021/jf000351x

Permethrin is a predominant pyrethroid widely used in agriculture and public health. A competitive enzyme-linked immunosorbent assay (ELISA) for the detection of permethrin was developed. Two haptens, the trans- and cis-isomers of 3-(4-aminophenoxy)benzyl-3-(2,2-dichloroethenyl)-2,2-dimethylcyclopropanecarb oxylate, were synthesized and conjugated with thyroglobulin as immunogens. Four antisera were generated and screened against six different coating antigens. The resulting ELISA has an I₅₀ value of 2.50/μg/L and relatively low cross-reactivities with other major pyrethroids, such as esfenvalerate, cypermethrin, deltamethrin, and cyfluthrin. Methanol was found to be the best solvent for this ELISA, with optimal sensitivity observed at a concentration of 40% (v/v). The assay parameters are unchanged at pH values between 5.0 and 8.0, whereas higher ionic strengths (>0.2 M PBS) strongly suppress the absorbances. River water samples fortified with permethrin were analyzed according to this method and validated by GC-MS. Good recoveries and correlation with spike levels were observed, suggesting this immunoassay is valuable for environmental monitoring and toxicological studies at parts per trillion levels of permethrin.

Full text of DOI:10.1021/jf000351x

4032
J. Agric. Food Chem. 2000, 48, 4032−4040
En zym e-Lin k ed Im m u n osor ben t Assa y for th e P yr eth r oid
P er m eth r in
Guomin Shan,† Whitney R Leeman,‡ Donald W. Stoutamire,† Shirley J . Gee,†
Daniel P. Y. Chang,‡ and Bruce D. Hammock*,†
Department of Entomology and Department of Civil and Environmental Engineering,
University of California, Davis, California 95616
Permethrin is a predominant pyrethroid widely used in agriculture and public health. A competitive
enzyme-linked immunosorbent assay (ELISA) for the detection of permethrin was developed. Two
haptens, the trans- and cis-isomers of 3-(4-aminophenoxy)benzyl-3-(2,2-dichloroethenyl)-2,2-di-
methylcyclopropanecarboxylate, were synthesized and conjugated with thyroglobulin as immunogens.
Four antisera were generated and screened against six different coating antigens. The resulting
ELISA has an I50 value of 2.50 µg/L and relatively low cross-reactivities with other major pyrethroids,
such as esfenvalerate, cypermethrin, deltamethrin, and cyfluthrin. Methanol was found to be the
best solvent for this ELISA, with optimal sensitivity observed at a concentration of 40% (v/v). The
assay parameters are unchanged at pH values between 5.0 and 8.0, whereas higher ionic strengths
(>0.2 M PBS) strongly suppress the absorbances. River water samples fortified with permethrin
were analyzed according to this method and validated by GC-MS. Good recoveries and correlation
with spike levels were observed, suggesting this immunoassay is valuable for environmental
monitoring and toxicological studies at parts per trillion levels of permethrin.
Keyw or d s: Permethrin; ELISA; environmental monitoring; toxicological studies; pyrethroid
INTRODUCTION
Current analytical methods for the detection of per-
methrin involve multistep sample cleanup procedures
followed by gas chromatography and detection by elec-
tron capture (GC-EC) or high-performance liquid chro-
matography (HPLC) (Wells et al., 1994; Boccelli et al.,
1993; Hengel et al., 1997). Such methods are relatively
time-consuming and expensive and, particularly, are not
suitable for large sample screening. An immunoassay
would provide a sensitive, selective, and rapid method
for the detection of permethrin at trace levels (Hammock
and Mumma, 1980; Hammock et al., 1990). Several
immunoassays have been developed for the detection
of permethrin (Stanker et al., 1989; Skerritt et al., 1992;
Bonwick et al., 1994); however, these do not have the
desired sensitivity for trace level analysis in environ-
mental samples without significant sample cleanup and
concentration. In this paper, we describe the develop-
ment of a sensitive immunoassay for permethrin, the
evaluation of the assay’s performance in water matrix,
and method validation by GC-MS.
Pyrethroid insecticides are likely to become more
widely used, particularly because organophosphate
insecticides, which currently dominate the market, are
being phased out in the United States. In 1997, >600000
lb of pyrethroid active ingredient were applied to various
crops throughout California. Among those pyrethroids
used in California, permethrin contributes >53% of the
total amount (Pesticide Use Report, 1997). Permethrin
is highly toxic to some nontarget invertebrates and fish
(Tomlin, 1997) and has high bioconcentration factors in
some aquatic animals (Schimmel et al., 1983). This
compound has exhibited detrimental effects to aquatic
species at low parts per billion levels of exposure (Kidd
and J ames, 1995; Tomlin, 1997). In addition to ecosys-
tem health, humans consume contaminated water and
aquatic food, leading to potential exposure to pyre-
throids. Although pyrethroids are thought to be safe for
humans, reversible symptoms of poisoning and sup-
pressive effects on the immune system have been
reported after exposure (He et al., 1988, 1989; Repetto,
1996). Some pyrethroids may cause lymph node and
splenic damage as well as carcinogenesis (Hallenbeck
and Cunningham-Burns, 1985). Because pyrethroids
offer significant advantages to the agricultural ecosys-
tem if used carefully, but have a potential for environ-
mental damage, a sensitive, selective, and rapid method
for monitoring residue levels of pyrethroids in aquatic
ecosystems is desirable.
MATERIALS AND METHODS
Ch em ica ls a n d Im m u n or ea gen ts. All pyrethroid stand-
ards were obtained from Riedel de Haen (Seelze, Germany).
Organic starting materials for hapten synthesis were pur-
chased from Aldrich Chemical Co. (Milwaukee, WI) and Fisher
Scientific (Pittsburgh, PA). Thin-layer chromatography (TLC)
was performed using 0.2 mm precoated silica gel 60 F254 on
glass plates from E. Merck (Darmstadt, Germany), and detec-
tion was made by ultraviolet light or iodine vapor stain. Flash
chromatographic separations were carried out on Baker silica
gel (40 µm average particle size) using the indicated solvents
where the f notation denotes a stepwise concentration gradi-
ent.
* Author to whom correspondence should be addressed
[telephone (530) 752-7519; fax (530) 752-1537; e-mail
bdhammock@ucdavis.edu].
† Department of Entomology.
The coupling reagents were purchased from Aldrich. Goat
anti-rabbit (GAR) immunoglobulin conjugated to horseradish
‡ Department of Civil and Environmental Engineering.
10.1021/jf000351x CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/10/2000

Permethrin Immunoassay
J. Agric. Food Chem., Vol. 48, No. 9, 2000 4033
F igu r e 1. Scheme for hapten synthesis.
peroxidase (HRP), bovine serum albumin (BSA), thyroglobulin,
Tween 20, and 3,3′,5,5′-tetramethylbenzidine (TMB) were
purchased from Sigma Chemical Co. (St. Louis, MO).
methylcyclopropanecarboxylic acid (0.75 g, 3.59 mmol) in
chloroform (1 mL) containing 1 µL of dimethylformamide
(DMF) was treated with thionyl chloride (0.52 mL, 7.12 mmol)
and stirred under N2 in an oil bath at 65 °C for 2 h. The
mixture was stripped briefly of solvents, hexane (2 mL) was
added, and the mixture was restripped. The residue was
dissolved in chloroform (3 mL) and added all at once to an
ice-cooled solution of 1 (0.88 g, 3.59 mmol) in chloroform (3
mL) and pyridine (0.348 mL), and the mixture was allowed to
stand at ambient temperature for 3 h. The reaction mixture
was washed in order with water, diluted HCl solution,
saturated NaHCO3 solution, and water and dried over Na2SO4.
Stripping of solvent gave a pale yellow oil, which was flash
chromatographed on silica gel (20 g) (5% f 100% CH2Cl2 in
hexane) to recover after vacuum stripping 1.36 g (87%) of 2:
In str u m en ts. NMR spectra were obtained using a General
Electric QE-300 spectrometer (Bruker NMR, Billerica, MA).
Chemical shift values are given in parts per million (ppm)
downfield from internal standard tetramethylsilane. Gas-
liquid chromatograms were determined on an HP 5890
(Hewlett-Packard Corp., Avondale, PA) with a 15 m, 0.32 mm
i.d. capillary column filled with a 0.25 µm film of dimethyl-
polysiloxane containing 5% of methyls substituted by phenyls
(J &W Scientific, Folsom, CA). Radial chromatographic separa-
tions were carried out on a Chromatotron apparatus (Harrison
Research, Inc., Palo Alto, CA), using 2-mm silica gel plates.
Fast atom bombardment mass spectra (FAB-MS) were ob-
tained on a ZAB-2SE mass spectrometer (VG Analytical,
Wythenshawe, U.K.), using high-energy cesium ions at a
density flux of 1-2 mA and 35-38 kV to generate secondary
[MH]+ ions. The liquid matrix was a 1:1 mixture of glycerol
and 3-nitrobenzyl alcohol, and cesium iodide was used for mass
calibration at a dynamic resolution of 5000:1. ELISA experi-
ments were performed in 96-well microplates (Nunc, Roskilde,
Denmark), and the absorbances were measured with a Vmax
microplate reader (Molecular Devices, Menlo Park, CA) in
dual-wavelength mode (450-650 nm).
1
TLC Rf 0.27 (hexane/CH2Cl2 ) 1:1); H NMR (CDCl3) δ 1.25
(s, 6 H, CH3), 1.90 (d, J ) 8.5 Hz, 1 H, CHCO2), 2.06 (dd, J )
8.7, 8.7 Hz, 1 H, CH), 5.11 (s, 2 H, CH2Ar), 6.24 (d, J ) 8.9
Hz, 1 H, CdCH), 7.03 (d, J ) 9.2 Hz, 2 H, NO2Ar), 7.03-7.46
(m, 4 H, Ar), 8.21 (d, J ) 9.2 Hz, 2 H, NO2Ar).
3-(4-Nitr oph en oxy)ben zyl (()-tr a n s-3-(2,2-Dich lor oeth -
en yl)-2,2-d im eth ylcyclop r op a n eca r boxyla te [(()-tr a n s-
4-Nit r op er m et h r in ] (3). Using (()-trans-2-(2,2-dichloro-
ethenyl)-3,3-dimethylcyclopropanecarboxylic acid in the same
reaction procedure as described for 2 above, the product was
obtained in 93% yield as a colorless oil: TLC Rf 0.30 (hexane/
Ha p ten Syn th esis a n d Ver ifica tion . Syntheses of the
haptens were carried out as outlined in Figure 1. All reactions
were straightforward and followed procedures employed in
preceding publications (Wengatz et al., 1998; Shan et al.,
1999a). NMR spectral data supported all structures, and mass
spectra further confirmed the structures of the final haptens.
3-(4-Nitr op h en oxy)ben zyl Alcoh ol (1). 3-(4-Nitrophe-
noxy)benzaldehyde (Loewe and Urbanietz, 1967) (3.0 g, 13.2
mmol) was added in portions over 1 h to a stirred solution of
sodium borohydride (280 mg., 7.4 mmol) in ethyl alcohol (15
mL). After 3 h, the mixture was diluted with water, and the
resulting solid was filtered, water washed, and dried. This was
dissolved in butyl chloride/methylene chloride and filtered
through silica gel (3 g) followed by methylene chloride to
remove polar impurities. The stripped residue was recrystal-
lized from butyl chloride by dilution with hexane to yield in
1
CH2Cl2 ) 1:1); H NMR (CDCl3) δ 1.19 (s, 3 H, CH3), 1.28 (s,
3 H, CH3), 1.66 (d, J ) 5.3 Hz, 1 H, CHCO2), 2.27 (dd, J ) 5.3,
8.3 Hz, 1 H, CdC-CH), 5.12 (d, J ) 12.8 Hz, 1 H, OHCH),
5.18 (d, J ) 12.8 Hz, 1 H, OHCH), 5.61 (d, J ) 8.3 Hz, 1 H,
CdCH), 7.00-8.23 (m, 8 H, Ar).
3-(4-Am in op h en oxy)ben zyl (()-cis-3-(2,2-Dich lor oeth -
en yl)-2,2-d im et h ylcyclop r op a n eca r b oxyla t e [(()-cis-4-
Am in op er m eth r in ] (4). A solution of 2 (218 mg, 0.5 mmol)
in ethanol (2 mL) was treated with stannous chloride dihydrate
(565 mg, 2.5 mmol), and the mixture was stirred in an oil bath
at 70 °C for 30 min, then cooled, diluted with ether and water,
and treated with Celite (0.45 g) and NaHCO3 (0.42 g, 5 mmol)
in portions (foaming). The mixture was filtered, and solids
were washed with ether. Stripping the organic phase gave a
heavy oil. TLC showed one primary product spot, which
darkened, on exposure to light. This was flash chromato-
graphed on silica gel (5 g) (50% CH2Cl2/hexane f 2-10% ether
in hexane) to recover 25 mg of 2 and 144 mg (80%) of 4 as a
1
two crops 2.86 g (95%) of a white solid: mp 73-75.5 °C; H
NMR (CDCl3) δ 1.80 (bs, 1 H, OH), 4.73 (s, 2 H, CH2), 7.02 (d,
J ) 9.2 Hz, 2 H, NO2Ar), 7.03-8.02 (m, 4 H, Ar), 8.20 (d, J )
9.2 Hz, NO2Ar).
1
yellow oil: TLC Rf 0.26 (CH2Cl2); H NMR (CDCl3) δ 1.23 (s,
3-(4-Nitr op h en oxy)ben zyl (()-cis-3-(2,2-Dich lor oeth en -
yl)-2,2-d im eth ylcyclop r op a n eca r boxyla te [(()-cis-4-Ni-
t r op er m et h r in ] (2). (()-cis-2-(2,2-Dichloroethenyl)-3,3-di-
6 H, CH3), 1.87 (d, J ) 8.5 Hz, 1 H, CHCO2), 2.03 (dd, J )
8.66, 8.66 Hz, 1 H, CdC-CH), 3.5 (bs, 2 H, NH2), 5.03 (s, 2 H,
CH2Ar), 6.26 (d, J ) 8.86 Hz, 1 H, CdCH), 6.67-7.29 (m, 8 H,

4034 J. Agric. Food Chem., Vol. 48, No. 9, 2000
Shan et al.
Ar); addition of D2O removed the NH2 peak; FAB-MS m/z calcd
for [M]+ ) C21H21Cl2NO3 405, obsd 405.
3-[(()-Cya n o[(()-cis-3(2,2-d ich lor oeth en yl)-2,2-d im eth -
ylcyclop r op a n eca r bon yloxy]m eth yl]p h en oxya cetic Acid
(9). A solution of 7 (200 mg, 0.409 mmol) in CH2Cl2 (0.5 mL)
was stirred under nitrogen, and iodotrimethylsilane (77 µL,
0.54 mmol) was added. After 3 h at ambient temperature and
1 h at 35 °C, the mixture was added to methanol (0.5 mL).
After 5 min, it was diluted with water and extracted with
CH2Cl2, and the extracts were washed with water and stripped.
The resulting gum was immediately flash chromatographed
on silica gel (5 g) eluting with CH2Cl2, then 40% EtOAc in
hexane followed by 20 f 40% EtOAc in CH2Cl2 containing
1.5% acetic acid. The stripped product slowly crystallized on
standing to 128 mg (78%) of a white solid. The 1H NMR
(CDCl3) spectrum clearly indicated a mixture of two dia-
stereoisomer pairs in a ratio of about 3:2 as follows: δ 1.22 (s,
3H, CH3), 1.26 (s, 3H, CH3), 1.30 (s, 3H, CH3), 1.32 (s, 3H,
CH3), 1.92 (d, J ) 8.4 Hz, 2 × 1 H, CHCO2), 2.15 (dd, J ) 8.6,
11.2 Hz, 1 H, CdC-CH), 2.17 (dd, J ) 8.6, 11.2 Hz, 1 H, Cd
C-CH), 4.74 (s, 2 × 2 H, CH2), 6.19 (d, J ) 8.8 Hz, 1 H, Cd
CH), 6.20 (d, J ) 8.7 Hz, 1 H, CdCH), 6.36 (s, 1 H, CHCN),
6.39 (s, 1 H, CHCN), 7.00-7.44 (m, 2 × 4 H, Ar) (the
underlined values are the more intense of the pair): FAB-MS
m/z calcd for [M + H]+ ) C18H18Cl2NO5 398, obsd 398.
3-(4-Am in op h en oxy)ben zyl (()-tr a n s-3-(2,2-Dich lor o-
eth en yl)-2,2-dim eth ylcyclopr opan ecar boxylate [(()-tr a n s-
4-Am in op er m eth r in ] (5). A sample of 3 (218 mg, 0.5 mmol)
was reduced with stannous chloride dihydrate (620 mg, 3.0
mmol) according to the procedure described for 4 above to give
1
161 mg (80%) as a yellow oil: TLC Rf 0.2 (CH2Cl2); H NMR
(CDCl3) δ 1.18 (s, 3 H, CH3), 1.27 (s, 3 H, CH3), 1.65 (d, J )
5.4 Hz, 1 H, CHCO2), 2.25 (dd, J ) 5.3, 8.4 Hz, 1 H, CdC-
CH), 3.2 (bs, 2 H, NH2), 5.06 (s, 2 H, CH2), 5.60 (d, J ) 8.4 Hz,
1 H, CdCH), 6.67-7.29 (m, 8 H, Ar); FAB-MS m/z calcd for
[M]+ ) C21H21Cl2NO3 405, obsd 405.
Ben zyl 3-[(()-Cya n o[(()-cis-3(2,2-d ich lor oeth en yl)-
2,2-d im eth ylcyclop r op a n eca r bon yloxy]m eth yl]p h en oxy-
a ceta te (7). Benzyl 2-(3-formylphenoxy)acetate (6) (0.970 g,
3.59 mmol) was converted to the cyanohydrin as described in
a previous publication (Wengatz et al., 1998). Meanwhile,
(()-cis-3-(2,2-dichloroethenyl)-2,2-dimethylcyclopropanecar-
boxylic acid (0.75 g, 3.59 mmol) was converted to the acid
chloride as described above for 2. The acid chloride was
esterified with the cyanohydrin according to the procedure
described above. The reaction mixture was washed with 1 N
HCl solution (5 mL), saturated Na2CO3 solution, and water,
dried (MgSO4), and then filtered and stripped to give the crude
product as a yellow-orange oil. This was flash chromato-
graphed on silica gel (25 g) (10 f 100% CH2Cl2 in hexane) to
recover 0.92 g (52%) of pure 7 showing one spot on TLC: Rf
0.40 (CH2Cl2).
A 76 mg sample of 7 was separated into its two diastereo-
isomer pairs by radial chromatography. The sample was
applied to a 2 mm silica gel plate in 25% chlorobutane in
hexane and eluted with 5% 1,2-dimethoxyethane in hexane.
Separation of the isomer pairs was complete to give 40.7 mg
of the higher Rf isomer pair [1H NMR (CDCl3) δ 1.29 (s, 3 H,
CH3), 1.30 (s, 3 H, CH3), 1.89 (d, J ) 8.4 Hz, 1 H, CHCO2),
2.12,(dd, J ) 8.5, 8.7 Hz, 1 H, CdC-CH), 4.70 (s, 2 H, CH2),
5.25 (s, 2 H, CH2), 6.18 (d, J ) 8.8 Hz, 1 H, CdCH), 6.31 (s, 1
H, CHCN), 6.95-7.39 (m, 9 H, Ar); 13C NMR (CDCl3) δ 14.8,
28.1, 28.9, 31.0, 33.4, 62.3, 65.4, 67.1, 114.2, 116.0, 116.4 (2
C), 121.0, 121.8, 123.8, 128.5 (2 C), 128.6, (2 C), 130.5, 133.1,
135.0, 158.2, 168.2, 168.5] and 27.6 mg of the lower Rf isomer
pair [1H NMR (CDCl3) δ 1.21 (s, 3 H, CH3), 1.24 (s, 3 H, CH3),
1.89 (d, J ) 8.4 Hz, 1 H, CHCO2), 2.15 (dd, J ) 8.6 Hz, 1 H,
CdC-CH), 4.70, (s, 2 H, OCH2CO2), 5.25 (s, 2 H, PhCH2), 6.19
(d, J ) 8.7 Hz, 1 H, CdCH), 6.35 (s, 1 H, CHCN), 6.94-7.39
(m, 9 H, Ar); 13C NMR (CDCl3) δ 14.8, 28.1, 28.8, 31.0. 33.5,
62.3, 65.4, 67.2, 114.4, 115.9, 116.3 (2 C), 121.0, 122.0, 123.7,
128.5 (2 C), 128.6, (2 C), 130.5, 133.4, 134.9, 158.2, 168 2,
168.5].
3-[(()-Cya n o[(()-tr a n s-3-(2,2-d ich lor oeth en yl)-2,2-d i-
m eth ylcyclop r op a n eca r bon yloxy]m eth yl]p h en oxya cetic
Acid (10). A solution of 8 (200 mg, 0.409 mmol) in CH2Cl2
(0.5 mL) was cleaved with iodotrimethylsilane, and the reac-
tion mixture was worked up as described above for the
preparation of 9 to give a slightly gummy white solid. This
was triturated with a small amount of 50% butyl chloride/
hexane and filtered to yield 97 mg (60%) of 10 as a white solid.
1
The H NMR (CDCl3) spectrum clearly indicated a mixture of
two diastereoisomer pairs in a ratio of about 3:2: 1H NMR
(CDCl3) δ 1.19 (s, 3 H, CH3), 1.23 (s, 3 H, CH3), 1.26 (s, 3 H,
CH3), 1.34 (s, 3 H, CH3), 1.67 (d, J ) 5.3 Hz, 1 H, CHCO2),
1.69 (d, J ) 5.3 Hz, 1 H, CHCO2), 2.29 (dd, J ) 5.3, 8.2 Hz,
CdC-CH), 2.33 (dd, J ) 5.4, 8.3 Hz, 1 H, CdC-CH), 4.73 (s,
2 × 2 H, CH2), 5.60 (d, J ) 8.2 Hz, 1 H, CdCH), 5.63 (d, J )
8.1 Hz, 1 H, CdCH), 6.40 (s, 1 H, CHCN), 6.41 (s, 1 H, CHCN),
6.98-7.43 (m, 2 × 4 H, Ar) (the underlined values are the more
intense of the pair); FAB-MS m/z calcd for [M + H]+ ) C18H18
Cl2NO5 398, obsd 398.
-
Ha p ten Con ju ga tion . Conjugates were synthesized using
a water-soluble carbodiimide method for acids and a diazoti-
zation method for anilines (Tijssen, 1985; Erlanger, 1973). To
obtain the immunogens, haptens 4 and 5 were conjugated to
thyroglobulin. Coating antigens were made by coupling hap-
tens 4, 5, 9, and 10, and trans-and cis-permethrin acids (PA)
to BSA.
Con ju ga t es of Ha p t en s 4 a n d 5 w it h Th yr oglob u lin
a n d BSA. Hapten 4 or 5 (0.10 mmol) was dissolved in 4 drops
of ethanol and treated with 1 mL of 1 N HCl. The resulting
solution was stirred in an ice bath as 0.5 mL of 0.20 M sodium
nitrite was added. DMF (0.4 mL) was then added dropwise to
give a homogeneous solution, which was divided into two equal
aliquots (each aliquot was used for one protein). Fifty mil-
ligrams of thyroglobulin or 45 mg of BSA was dissolved in 5
mL of 0.2 M borate buffer (pH 9.6) and 1.5 mL of DMF.
Aliquots (∼0.9 mL each) of the activated hapten solution were
added dropwise to the two stirred protein solutions. The
reaction mixture was stirred in an ice bath for 45 min and
then dialyzed against PBS over 72 h at 4 °C. The purified
conjugates were suspended in water and stored in aliquots at
-80, -20, and 4 °C.
Con ju ga tes of Ha p ten s 9, 10, (()-tr a n s-P A, a n d (()-cis-
P A w ith BSA. Hapten (0.025 mmol) was dissolved in 2 mL
of dry DMF, and then 6 mg (0.05 mmol) of N-hydroxysuccin-
imide (NHS) and 5.8 mg (0.03 mmol) of 1-ethyl-3-(3-dimethyl
aminopropyl) carbodiimide hydrochloride (EDC) were added.
The reaction mixture was stirred overnight at room temper-
ature. Forty-five milligrams of BSA was dissolved in 8 mL of
PBS and 2 mL of carbonate buffer (pH 9). The activated hapten
was added dropwise to the protein solution. The mixture was
stirred for 30 min at room temperature and for 6 h at 4 °C.
Ben zyl 3-[(()-Cya n o[(()-tr a n s-3-(2,2-d ich lor oeth en yl)-
2,2-d im e t h ylcyclop r op a n e ca r b on yloxy]m e t h yl]p h e n -
oxya cet a t e (8). By substituting (()-trans-3-(2,2-dichloro-
ethenyl)-2,2-dimethylcyclopropanecarboxylic acid in the same
reaction procedure as described for 7 above, pure 8 was
obtained in 55% yield: TLC Rf 0.35 (CH2Cl2).
A 76 mg sample of 8 was separated into its two diastereo-
isomer pairs by radial chromatography as described above to
give 40.5 mg of the higher Rf isomer pair [1H NMR (CDCl3)
δ 1.22 (s, 3 H, CH3), 1.33 (s, 3 H, CH3), 1.65 (d, J ) 5.3 Hz, 1
H, CHCO2), 2.29 (dd, J ) 5.3, 8.2 Hz, 1 H, CdC-CH), 4.69 (s,
2 H, CH2CO2), 5.24, (s, 2 H, OCH2Ph), 5.59 (d, J ) 8.2 Hz, 1
H, CdCH), 6.35 (s, 1 H, CHCN), 6.94-7.4 (m, 9 H, Ar); 13C
NMR (CDCl3) δ 19.9, 22.4, 30.4, 33.8 (2 H), 62.6, 65.3, 67.1,
114.3, 116.0, 116.4 (2 C), 121.0, 122.9, 126.0, 128.5 (2 C), 128.6
(2 C), 130.5, 133.2, 134.9, 158.2, 168 2, 169.2] and 25.5 mg of
the lower Rf isomer pair [1H NMR (CDCl3) δ 1.18 (s, 3 H, CH3),
1.26 (s, 3 H, CH3), 1.66 (d, J ) 5.3 Hz, 1 H, CH2CO2), 2.32 (dd,
J ) 5.3, 8.2 Hz, 1 H, CdC-CH), 4.69 (s, 2 H, OCH2CO2), 5.25
(s, 2 H, OCH2Ph), 5.62 (d, J ) 8.2 Hz, 1 H, CdCH), 6.37 (s, 1
H, CHCN), 6.94-7.41 (m 9 H, Ar);13C NMR (CDCl3) δ 20.0,
22.4, 30.2, 33.8, 33.9, 62.6, 65.4, 67.2, 114.3, 115.9, 116.3, 116.4,
121.0, 123.1, 126.0, 128.5 (2 C), 128.6 (2 C), 130.5, 133.4, 135.0,
158.2, 168.2, 169.3].

Permethrin Immunoassay
J. Agric. Food Chem., Vol. 48, No. 9, 2000 4035
The solution was then dialyzed against PBS over 72 h at 4 °C
and stored at -20 °C.
1999a). The final eluent in methanol was used directly for
immunoassay. An eluent aliquot was evaporated to dryness
and then redissolved in toluene for GC-MS measurement.
Assay validation was conducted with an HP 5890 GC equipped
with an HP 5973 mass selective detector and a 30 m × 0.25
mm i.d. (df ) 0.25 µm) HP-5MS 5% phenyl methyl siloxane
column. The injector was operated at 280 °C. An HP model
6890 series autoinjector was used to inject 1 µL of sample. The
oven temperature was programmed from 200 to 280 °C at 5
°C/min and held for 5 min. Helium was used as a carrier gas
(1 mL/min). The detection limit of quantitative analysis was
10 pg for permethrin.
Im m u n iza tion a n d An tiser u m P r ep a r a tion . Permethrin
antisera were obtained following the protocol reported earlier
(Shan et al., 1999b). Briefly, for each immunogen (4-Thyr,
5-Thyr), two New Zealand white rabbits were immunized
(rabbits 546 and 548 for 4-Thyr and rabbits 549 and 550 for
5-Thyr). The antigen solutions (100 µg in PBS) were emulsi-
fied with Freund’s complete adjuvant (1:1, v/v) and injected
intradermally. After 1 month, the animals were boosted with
an additional 100 µg of immunogen that was emulsified with
Freund’s incomplete adjuvant (1:1 v/v). Booster injections were
given at 4 week intervals. The rabbits were bled ∼10 days after
each boost. The serum was isolated by centrifugation for 10
min at 4 °C. The results of antibody characterization were
obtained from sera of terminal bleeds after four boosters.
En zym e Im m u n oa ssa y. The competitive inhibition ELISA
format in this study was based on methods described by Voller
et al. (1976). Microplates were coated overnight at 4 °C with
100 µL/well of the appropriate coating antigen concentration
in 0.1 M carbonate-bicarbonate buffer (pH 9.6). After the
plates had been washed with washing solution (0.05% Tween
in distilled water), the surface of the wells was blocked with
200 µL of a 0.5% BSA-PBS solution by incubation for 30 min
at room temperature to minimize nonspecific binding in the
plate. After another washing step, 100 µL of antiserum diluted
in PBS (PBS: 8 g/L NaCl, 1.15 g/L Na2HPO4, 0.2 g/L KCl) per
well (for titration experiment) or 50 µL/well of antiserum
diluted in PBS with 0.2% BSA (PBSB) and 50 µL/well of
standard analyte or sample solution were added and incubated
for 1 h. The standard analyte concentrations ranged from 0.05
µg/L to 5 mg/L. Following a washing step, GAR-HRP (diluted
1:3000 in PBS with 0.05% Tween 20, 100 µL/well) was added
and incubated for 1 h at room temperature. The plates were
washed again, and 100 µL/well of substrate solution (3.3 µL
of 30% H2O2, 400 µL of 0.6% TMB in DMSO per 25 mL of
acetate buffer, pH 5.5) was added. The color development was
stopped after 15 min with 50 µL/well of 2 M H2SO4. The
absorbance was measured using a dual wavelength mode
at 450 nm minus 650 nm. Standard curves were obtained by
plotting absorbance against the logarithm of analyte concen-
tration, which were fitted to a four-parameter logistic equa-
tion: y ) {(A - D)/[1 + (x/C)B]} + D, where A is the maximum
absorbance at no analyte, B is the curve slope at the inflection
point, C is the concentration of analyte giving 50% inhibition
(I50), and D is the minimum absorbance at infinite concentra-
tion.
RESULTS AND DISCUSSION
Ha p ten Syn th esis. Permethrin insecticide is a mix-
ture of four chiral isomers, a consequence of the two
chiral centers in the cyclopropane ring of the molecule.
The insecticidal activities of the cis- and trans-isomers
differ depending on the target insect (Elliott et al., 1978);
however, trans-permethrin is metabolized much more
rapidly than the cis-isomer in mammals (Gaughan et
al., 1979; Casida and Ruzo, 1980). Consequently, the
trans-isomers are made to predominate (60-75%) in the
commercially produced product. The haptens were thus
designed to contain both the cis- and trans-isomers. The
primary goal of this study was to develop a permethrin-
specific antibody. Because the alcohol portion of per-
methrin, the phenoxybenzyl group, is common to many
of the synthetic pyrethroids, the strategy for designing
the immunizing hapten was to link the protein through
the aromatic end of the permethrin molecule and leave
the relatively unique acid portion distal from the protein
to improve antibody specificity (Goodrow et al., 1995).
In addition, due to the lipophilicity of pyrethroids, a long
side chain (handle) could allow the lipophilic hapten to
fold back into the hydrophobic interior of the protein
and decrease the affinity of the resulting antibodies. In
our experience, a rigid or shorter side chain in an
immunizing hapten is important for the generation of
sensitive and selective antibodies (Sanborn et al., 1998).
Thus, the hapten we have chosen for immunization
contains the whole permethrin molecule and has no side
chain (Figure 1). This approach differed from that
considered in previous papers (Stanker et al., 1989;
Skerritt et al., 1992; Bonwick et al., 1994), in which
either a long side chain or a modification of the acid
moiety of permethrin was chosen for coupling.
Thyroglobulin conjugates of haptens 4 and 5 were
used for antibody production, and their BSA conjugates
and conjugates of other haptens served as coating
antigens. To facilitate the synthesis, isolation, and
purification of the haptens, the cis- and trans-isomer
mixtures were prepared separately. The permethrin
acids were prepared according to literature procedures
(Nakada et al., 1979; Kleschick 1986). The cis- and
trans-acids were separated by fractional vacuum distil-
lation of the ethyl esters through an annular still.
Subsequent hydrolysis and crystallization gave the
racemic cis- and trans-permethrin acids. GLC of the
methyl esters prepared via trimethylsilyldiazomethane
gave isomeric purity of >99% for both acids.
An tibod y Ch a r a cter iza tion a n d Assa y Op tim iza tion .
Antibodies and antigens were screened in a two-dimensional
titration for the best dilution of coating antigen and antiserum.
Then the competitive inhibition curves were measured for
different antibody and antigen combinations, and the one with
the lowest I50 was selected for further assay development.
The effects of solvents were tested by dissolving the analyte
in PBS buffers containing various proportions of solvent (0,
10, 20, 40, 60, and 80% solvent) and incubating these with
antibody in PBSB on the coated plate. Methanol and DMSO
were tested in this way.
In the experiment to evaluate pH effects, both the analyte
and antiserum were dissolved in PBS buffer at the specified
pH and added to a coated plate. pH values of 5, 6, 7, and 8
were tested in this incubation step with all other parameters
of the assay fixed. Ionic strength effects were determined in
the same manner as previously mentioned except that, instead
of pH, PBS concentration was varied. PBS concentrations of
0.1, 0.2, 0.3, and 0.5 M were tested.
Cr oss-Rea ctivity (CR). The optimized assays were applied
to cross-reactivity studies by using the standard solution of
the analyte and other structurally related compounds (listed
in Table 3). The CR was determined by dividing the I50 of the
chemical assigned to be 100% by the I50 of another compound
and multiplying by 100 to obtain a percent figure.
SP E a n d GC-MS. The SPE method used in this study was
the same as a previously reported method for the extraction
of esfenvalerate from water using a C18 column (Shan et al.,
The synthesis routes for the haptens are summarized
in Figure 1. The intermediate nitrophenoxybenzyl al-
cohol 1 was prepared by sodium borohydride reduction
of the corresponding aldehyde (Loewe and Urbanietz,
1967). Esterification with the acid chlorides of the
racemic cis- and trans-permethrin acids gave the cor-
responding nitropermethrins 2 and 3. Subsequent re-

4036 J. Agric. Food Chem., Vol. 48, No. 9, 2000
Shan et al.
Ta ble 1. Su m m a r y of Op tica l Den sity of Titr a tion Testsa
Ta ble 2. Selected Com p etitive ELISA Scr een in g Da ta
a ga in st P er m eth r in a
trans- cis-
Ab/immuno-
gen
PA- PA-
immunogen
antiserum
coating antigen
I50 (µg/L)
4-BSA 5-BSA 9-BSA 10-BSA BSA BSA
4-Thyr
Ab549
4-BSA
5-BSA
9-BSA
10-BSA
2.60
289
4.60
470
546/4-Thyr
548/4-Thyr
549/5-Thyr
550/5-Thyr
++
+
-
+
+++
-
-
-
-
-
-
-
-
+++
+++
+++
+++
++
++++
++++
+++
+++
++++
++++
Ab550
4-BSA
5-BSA
9-BSA
10-BSA
4.00
420
6.84
486
a
The data shown are at an antibody dilution of 1:16000 and a
coating antigen concentration of 1 mg/L. Symbols refer to optical
density ranges after 15 min of color development: (-) absorbance
< 0.25; (+) absorbance ) 0.25-0.50; (++) absorbance ) 0.50-
0.75; (+++) absorbance ) 0.75-1.00; (++++) absorbance > 1.00.
5-Thyr
Ab548
4-BSA
5-BSA
10-BSA
400
12.9
526
duction with stannous chloride (Bellamy and Ou, 1984)
followed by flash chromatography gave the pure haptens
4 and 5. Both gave TLC spots that rapidly darkened on
exposure to light as is typical of anilines in general.
The coating haptens 9 and 10 were prepared by a
similar acylation of the cyanohydrin of the aldehyde 6
as described previously (Wengatz et al., 1998). This
introduces an additional chiral center alpha to the
nitrile group that results in two diastereoisomer pairs
for each of the intermediate esters 7 and 8. A small
sample of each was separated by radial chromatography
to obtain definitive confirming NMR spectra; however,
the respective mixtures were carried through to the final
haptens. Cleavage of the benzyl esters with iodotri-
methylsilane gave the haptens 9 and 10 as mixtures of
two diastereoisomer pairs. Flash chromatography gave
pure mixtures that crystallized to white solids. NMR
spectra indicated these pairs to be in a ratio of ∼3:2 in
both cases, allowing tentative separation of the spectra.
No attempt was made to separate isomer pairs or to
determine relative chiral assignments.
Scr een in g a n d Selection of An tiser a . The antisera
of four rabbits were screened against six different
coating antigens using a two-dimensional titration
method with the coated antigen format. The objective
was to find the coating antigen that has the highest
affinity with tested antibody but could still be replaced
by target analyte. The results of the titration experi-
ments using the final bleeds are shown in Table 1. In
all combinations, the titer of Ab546 and Ab548 was
much lower than those of Ab549 and Ab550. The
homologous assay, in which the same hapten was used
in the coating antigen and immunogen, had a higher
titer than the heterologous assay. For Ab549, which was
generated from the trans-permethrin hapten immuno-
gen, coating antigen 5-BSA had the highest affinity,
followed by 10-BSA, then 4-BSA and 9-BSA. No
affinity was measured for the trans- and cis-PA-BSA
coating antigens with all four antibodies. Combinations
of coating antigen and antiserum that resulted in high
optical densities (OD > 0.75) were selected for further
development.
a
The permethrin analyte standards were prepared in a 40%
methanol/PBS solution.
allethrin and esfenvalerate immunoassay studies (Wing
and Hammock, 1979; Shan et al., 1999a). These results
further suggested that antibody stereoselectivity could
play an important role in the assay development by the
design of coating antigen and immunogen haptens. After
screening, the combination of antiserum 549 with coat-
ing antigen 4-BSA (hereafter designated 4/549) and
coating antigen 9-BSA (hereafter designated 9/549)
gave the lowest I50 values (2.6 and 4.6 µg/L, respectively)
when using permethrin as the inhibitor. These systems
were selected for CR studies, and one of them (4/549)
was chosen for further assay development and optimi-
zation.
CR. The ELISA systems 4/549 and 9/549 expressed
different selectivity patterns to pyrethroid analogues
(Table 3). As expected, because the trans-isomer hapten
was used for immunization, trans-permethrin was more
strongly recognized by both systems than the other
compounds tested. System 4/549 had the highest CR
with phenothrin (72%) followed by cypermethin (28%),
cyfluthrin (11%), and resmethrin (4%), whereas system
9/549 strongly cross-reacted with cypermethrin (56%),
phenothrin (42%), and cyfluthrin (24%). Little or no CR
was measured for deltamethrin, esfenvalerate, fluvali-
nate, or permethrin metabolites in either system. Among
the pyrethroids displaying CR, cypermethrin and cy-
fluthrin are actively used in the agricultural field;
however, the use of phenothrin is negligible in the
United States (Pesticide Use Report, 1997). Therefore,
cypermethrin and cyfluthrin may be important sources
of interference when these assays are used for real world
sample measurement. Because our primary goal was to
develop a compound-specific immunoassay, low CR to
common pyrethroids was an important factor in assay
development, and only assay 4/549 was selected for
further optimization.
Solven t, p H, a n d Sa lt Effects. Finding a proper
cosolvent for an immunoassay is very important for the
analysis of hydrophobic chemicals such as permethrin.
The solvent (methanol or DMSO) effects on the ELISA
system (4/549) were evaluated by preparing permethrin
in PBS containing various amounts of solvent. DMSO
strongly interferes with the assay sensitivity and maxi-
mum absorbance (Table 4). Using methanol as a cosol-
vent, absorbance was enhanced with higher MeOH
concentration (Figure 2): a similar effect was noticed
during our previous esfenvalerate immunoassay devel-
opment (Shan et al., 1999a). The I50 value of the assay
varied depending upon the different concentrations of
the cosolvent MeOH. The lowest I50 was found at 40%
Competitive inhibition experiments were performed
to optimize antiserum and coating antigen concentra-
tions for a high sensitivity. The reagent concentrations
having optical densities of ∼0.8 and lowest I50 values
were chosen as the optimal conditions. The I50 values
for each combination ranged from 2.60 to 500 µg/L
(Table 2). The stereoselectivity is significant in assay
sensitivity. The I50 of the homologous assay (trans-
hapten 5 and 10) was ∼100 times higher than that of
the heterologous (cis-hapten 4 and 9) assay. The ste-
reoselectivity of antibodies has been reported in bio-

Permethrin Immunoassay
J. Agric. Food Chem., Vol. 48, No. 9, 2000 4037
Ta ble 3. Cr oss-Rea ctivites of P yr eth r oid s a n d Th eir Meta bolites
Ta ble 4. Effects of DMSO Con cen tr a tion on Assa y
P a r a m eter sa
MeOH (I50 ) 13.6 ppb), the assay might still be useful
for permethrin quantitationn; however, the high back-
ground makes it impractical (A/D ) 3.6). These data
suggest that the cosolvent could be an important factor
in assay performance, especially for hydrophobic com-
pounds. A high concentration of cosolvent will help the
solubility of hydrophobic permethrin in reaction solu-
tion, but it could also affect the interaction of antibody
and antigen, resulting in a high background. On the
basis of the I50 values and the ratios of maximum and
minimum absorbances for the permethrin standard
curves, a MeOH concentration of 40% was selected for
subsequent experiments.
DMSOb
(%)
ABSmax
(A)
slope
(B)
I50 (ppb)
(C)
ABSmin
(D)
A/D R2
0
10
20
40
0.86 ( 0.10c 0.89 18.4 ( 2c
0.09
0.06
0.06
0.07
9.5 0.99
8.6 1.00
7.5 0.99
5.0 0.99
0.52 ( 0.06
0.45 ( 0.03
0.35 ( 0.03
0.63
0.61
0.72
110 ( 12.0
190 ( 18.5
400 ( 25
a
ELISA conditions: coating antigen 4-BSA (0.4 µg/mL); anti-
b
serum 549 (1/5000); goat anti-rabbit IgG-HRP (1/3000). Con-
centration of DMSO in permethrin standard solution (PBS-
DMSO). c Mean value ( SD. Each set of data represents the
average of three experiments.
MeOH (2.40 ppb), which is ∼15 times lower than that
at 20% MeOH (32 ppb) and 3 times lower than that at
60% MeOH (8.60 ppb). At concentrations up to 80%
To address potential interferences from aqueous
environmental samples, the effects of pH and ionic

4038 J. Agric. Food Chem., Vol. 48, No. 9, 2000
Shan et al.
F igu r e 4. ELISA competition curves of permethrin prepared
in buffer at various ionic strengths. Reagent concentrations:
coating antigen (4-BSA), 0.4 µg/mL; antiserum (Ab549),
1/6000 (final concentration in wells); goat anti-rabbit IgG-
HRP, 1/3000. Each curve represents the average of three
replicates.
F igu r e 2. ELISA competition curves of permethrin prepared
in PBS buffer containing various concentrations of methanol.
Reagent concentrations: coating antigen (4-BSA), 1/8000;
antiserum (Ab549), 1/6000 (final concentration in wells); goat
anti-rabbit IgG-HRP, 1/3000. Error bars represent standard
deviation of three replicates.
F igu r e 5. ELISA inhibition curves for permethrin. Reagent
concentrations: antiserum 549, 1/6000 (final dilution in wells);
coating antigen (4-BSA), 0.4 µg/mL. Standard curves repre-
sent the average of 16 curves.
F igu r e 3. ELISA competition curves of permethrin prepared
at various pH values. Reagent concentrations: coating antigen
(4-BSA), 0.4 µg/mL; antiserum (Ab549), 1/6000 (final concen-
tration in wells); goat anti-rabbit IgG-HRP, 1/3000. Each
curve represents the average of three replicates.
Ta ble 5. Reten tion Tim es a n d Detected Ma sses of
P yr eth r oid s
strength on ELISA performance were evaluated in this
study. In system 4/549, when analyte was dissolved in
buffer at various pH values, no significant effect upon
the I50 was detected, indicating that the assay could
effectively detect permethrin at pH values ranging from
5.0 to 8.0 (Figure 3). Ionic strength strongly influenced
ELISA performance (Figure 4). A higher salt concentra-
tion in the assay system resulted in lower optical
densities (OD) and higher I50 values. The OD values at
salt concentrations of 0.3 and 0.5 M PBS decreased by
approximately 20 and 65%, respectively, from the OD
value at a salt concentration 0.1 M PBS. Thus, the
maintenance of a minimal ionic strength appears to be
important.
The optimized permethrin ELISA used coating anti-
gen 4-BSA at 0.4 µg/mL, antibody 549 at a dilution of
1/5000, and permethrin in 40% methanol-PBS buffer.
The I50 value of this assay was 2.50 ( 0.6 µg/L (Figure
5) with a limit of quantitation (LOQ) of 0.30 ( 0.08 µg/L
in buffer. The LOQ was estimated as the concentration
that corresponded to the absorbance of the control (zero
concentration of analyte) minus 3 times the standard
deviation of the control (Grotjan and Keel, 1996).
retention
detected masses,
pyrethroid
time (min)
m/z (% base peak)
permethrin
cyfluthrin
cypermethrin
esfenvaleate
deltamethrin
13.50
14.80
16.01
17.10
19.20
163 (32), 183 (100)
163 (100), 226 (52)
163 (100), 181 (51)
125 (100), 167 (81), 225 (60)
181 (81), 253 (100)
GC-MS An a lysis Assessm en t. A sensitive GC-MS
method was developed for the assay validation. Under
the GC conditions described, permethrin was completely
separated from other pyrethroids such as esfenvalerate,
cyfluthrin, and cypermethrin (Table 5). After GC sepa-
ration, the substances were detected using MS in the
selected ion monitoring (SIM) mode. The mass selective
detector provided characteristic information about each
of the separated substances, which resulted in high
specificity regarding the determination of a single
analyte. Due to the lower noise level, better sensitivity
was achieved using SIM. In the case of permethrin, the
most important peak of fragmentation (m/z 183) was
used for quantitation (Figure 6). The detection limit was
calculated from a signal-to-noise ratio of 3:1. The

Permethrin Immunoassay
J. Agric. Food Chem., Vol. 48, No. 9, 2000 4039
F igu r e 7. Relationship between permethrin levels as meas-
ured by GS-MS and ELISA. Water samples were from the
Sacramento River (9), Putah Creek (Davis) (b), Lake Ber-
ryessa (2), and tap water (Davis) (f). Y ) 1.211x - 0.068, R2
) 0.900, n ) 24.
Ta ble 6. Sp ik e Recover y in Wa ter Ma tr ixa
spiked
concn
(µg/L)
ELISA
measured % recovery
<0.002
GC-MS
measured % recovery
<0.002
0
F igu r e 6. GC-MS chromatogram and the corresponding mass
spectrum of permethrin.
0.01
0.05
0.5
0.011 ( 0.002 110 ( 20
0.008 ( 0.002 80 ( 20
0.036 ( 0.006 72 ( 12
0.038 ( 0.005
0.41 ( 0.042
76 ( 10
82 ( 8.4
0.38 ( 0.035
76 ( 7.0
resulting detection limit for permethrin in water ranged
from 0.005 to 0.01 µg/L.
a
Water samples are from the Sacramento River, Sacramento,
CA. Different amounts of permethrin (in methanol) were added
to water samples to final concentrations in water of 0.01, 0.05,
and 0.5 µg/L. After thorough mixing and sitting for at least 2 h,
these samples were extracted by SPE before immunoassay and
GC-MS measurement.
An a lysis of Sp ik ed Wa ter Sa m p les. To study the
spike recovery, water samples from the Sacramento
River (Sacramento, CA), Putah Creek (Davis, CA), Lake
Berryessa (CA), and local tap water (Davis, CA) were
spiked with permethrin (with concentrations ranging
from 0 to 10 ppb) in a blind fashion. The samples were
extracted using SPE and measured by ELISA and GC-
MS. No matrix effects were measured when 200 mL of
water sample was extracted with SPE and eluted with
3.5 mL of methanol. The volume of methanol was
finalized at 1 mL and diluted by a factor 5 before ELISA
analysis, resulting in a concentration factor of 40.
Because permethrin is highly hydrophobic, it can be
easily sorbed to glass or plastic surfaces. In one of our
studies, after leaving the spiked sample in the bottle
overnight, we found that >85% of the permethrin was
trapped to the glass surface (data not shown). Therefore,
a small amount of methanol was used for rinsing the
container, and this methanol was combined with the
SPE eluent. A good correlation between permethrin
concentration measured by the GC-MS and ELISA (R2
) 0.900, a ) 1.21, and b ) -0.068) was obtained from
linear regression analysis (Figure 7). In a subsequent
study, water samples from the Sacramento River were
spiked with four different concentrations of permethrin
(0, 0.01, 0.1, and 0.5 ppb) and analyzed (Table 6). Spike
recoveries in both ELISA and GC-MS methods were
compared. All recoveries were >72% of the spiked
values, and again, a good agreement between these two
detection methods was obtained. The results demon-
strate that these assays are suitable for the quantitative
detection of permethrin at trace levels in water samples.
Con clu sion s. A sensitive and selective immunoassay
for permethrin has been developed by using 5 [(()-trans-
isomer] as the immunizing hapten and 4 [(()-cis-isomer]
as the coating antigen hapten. ELISA 4/549 has a very
low I50 value and also exhibits good performance char-
acteristics at various pH values and high solvent (40%
methanol) levels. Even though the assay is sensitive to
high ionic strength, with the implementation of dilution
or SPE, interferences can be minimized for field sample
measurement. With a simple SPE step, this ELISA was
successfully applied to quantitate low parts per trillion
(ppt) of permethrin in water. A good correlation between
ELISA and GC-MS results was achieved in the valida-
tion study, suggesting that this permethrin immuno-
assay is useful for environmental monitoring and toxi-
cological studies.
ABBREVIATIONS USED
Ab, antibody; BSA, bovine serum albumin; cAg, coat-
ing antigen; CR, cross-reactivity; DMF, N,N-dimethyl-
formamide; DMSO, dimethyl sulfoxide; EDC, 1-ethyl-
3-(3-dimethylaminopropyl)carbodiimide hydrochloride;
ELISA, enzyme-linked immunosorbent assay; FAB-MS,
fast atom bombardment mass spectrum; GAR-HRP,
goat anti-rabbit immunoglobulin conjugated to horse-
radish peroxidase; GC-MS, gas chromatography with
mass spectral detection; I50, concentration of analyte
giving 50% inhibition; LOQ, limit of quantitation; NHS,
N-hydroxysuccinimide; NMR, nuclear magnetic reso-
nance; OD, optical density; PA, permethrin acid; PBS,
phosphate-buffered saline; PBST, phosphate-buffered
saline with 0.05% of Tween 20; SD, standard deviation;
SIM, selected ion monitoring; SPE, solid-phase extrac-
tion; TLC, thin-layer chromatography; TMB, tetra-
methylbenzidine.

4040 J. Agric. Food Chem., Vol. 48, No. 9, 2000
Shan et al.
Loewe, H.; Urbanietz, J . Ger. Patent 1 241 832, 1967.
ACKNOWLEDGMENT
Nakada, Y.; Eudo, R.; Muvamatsu, S.; Ide, J .; Yura, Y. Studies
on chrysanthemate derivatives, VI. A stereoselective syn-
thesis of trans-3-(2,2-dichlorovinyl) 2,2-dimethyl-1-cyclopro-
panecaboxylic acid and related compounds. Bull. Chem. Soc.
J pn. 1979, 52, 1511-1514.
We thank Axel Ehmann (Mass Search, Inc., Modesto,
CA) and A. Daniel J ones (Pennsylvania State Univer-
sity) for mass spectral determination and the Depart-
ment of Chemistry (UC Davis) for the use of their NMR
instrument.
Pesticide Use Report 1997, State of California EPA.
Repetto, R. C. Pesticides and the Immune System: The Public
Health Risks; Repetto, R., Baliga, S. S., Eds.; WRI (World
Resources Institute), National Center for Food and Agri-
cultural Policy: Washington, DC, 1996.
Sanborn, J . R.; Gee, S. J .; Gilman, S. D.; Sugawara, Y.; J ones,
A. D.; Rogers, J .; Szurdoki, F.; Stanker, L. H.; Stoutamire,
D. W.; Hammock, B. D. Hapten synthesis and antibody
development for polychlorinated dibenzo-p-dioxin immuno-
assays. J . Agric. Food Chem. 1998, 46, 2407-2416.
Schimmel, S. C.; Garnas, R. L.; Patrick, J . M., J r.; Moore, J .
C. Acute toxicity, bioconcentration, and persistence of AC
222,705, benthiocarb, chlorpyrifos, fenvalerate, methyl para-
thion, and permethrin in the estuarine environment. J .
Agric. Food Chem. 1983, 31, 104-113.
LITERATURE CITED
Bellamy, F. D.; Ou, K. Tetrahedron Lett. 1984, 25, 839-842.
Boccelli, R.; Capri, E.; Delre, A. A. M. Residues of esfenvalerate
in irrigation water, stubble, husked and unhusked rice
samples of a paddy-field. Toxicol. Environ. Chem. 1993, 37,
185-189.
Bonwick, C. A.; Abdul-Latif, P.; Sun, C.; Baugh, P. J .; Smith,
C. J .; Armitage, R.; Davies, D. H. Comparison of chemical
methods and immunoassay for the detection of pesticide
residues in various matrices. Food Agric. Immunol. 1994,
6, 267-276.
Casida, J . E.; Ruzo, L. O. Metabolic chemistry of pyrethroid
insecticides. Pestic. Sci. 1980, 11, 257-269.
Elliott, M.; Farnham, A. W.; J anes, N. F.; Soderlund, D. M.
Insecticidal activity of pyrethrins and related compounds.
Part XI. Relative potencies of isomeric cyano-substituted
3-phenoxybenzyl esters. Pestic. Sci. 1978, 9, 112-116.
Erlanger, B. F. Principles and methods for the preparation of
drug protein conjugates for immunological studies. Phar-
macol. Rev. 1973, 25, 271-280.
Shan, G.-M.; Stoutamire, D. W.; Wengatz, I.; Gee, S. J .;
Hammock, B. D. Development of an immunoassay for the
pyrethroid insecticide esfenvalerate. J . Agric. Food Chem.
1999a , 47, 2145-2155.
Shan, G.-M.; Wengatz, I.; Stoutamire, D. W.; Gee, S. J .;
Hammock, B. D. An enzyme-linked immunoassay for the
detection of esfenvalerate metabolites in human urine.
Chem. Res. Toxicol. 1999b, 12, 1033-1041.
Skerritt, J . H.; Hill, A. S.; McAdam, D. P.; Stanker, L. H.
Analysis of the synthetic pyrethroids, permethrin and 1(R)-
phenothrin, in grain using a monoclonal antibody-based test.
J . Agric. Food Chem. 1992, 40, 1287-1292.
Stanker, L. H.; Bigbee, C.; Van Emon, J .; Watkins, B.; J ensen,
R. H.; Morris, C.; Vanderlaan, M. An immunoassay for
pyrethroids: detection of permethrin in meat. J . Agric. Food
Chem. 1989, 37, 834-839.
Gaughan, L. C.; Unai, T.; Casida, J . E. Permethrin metabolism
in rats. J . Agric. Food Chem. 1977, 25, 9-17.
Goodrow, M. H.; Sanborn, J . R.; Stoutamire, D. W.; Gee, S. J .;
Hammock, B. D. Strategies for immunoassay hapten design.
In Immunoanalysis of Agrochemicals: Emerging Technolo-
gies; Nelson, J . O., Karu, A. E., Wong, R. B., Eds.; American
Chemical Society Symposium 586; American Chemical
Society: Washington, DC, 1995; pp 1-19.
Grotjan, H. E.; Keel, B. A. Data interpretation and quality
control. In Immunoassay; Diamandis, E., Christopoulos, T.,
Eds.; Academic Press: San Diego, CA, 1996; pp 51-93.
Hallenbeck, W. H.; Cunningham-Burns, K. M. Pesticides and
Human Health; Springer-Verlag: New York, 1985; p 118.
Hammock, B. D.; Mumma, R. O. Potential of immunochemical
technology for pesticide analysis. In Recent Advances in
Pesticide Analytical Methodology; Harvey, J ., Zweig, G.,
Eds.; American Chemical Society: Washington, DC, 1980;
pp 321-352.
Hammock, B. D.; Gee, S. J .; Harrison, R. O.; J ung, F.; Goodrow,
M. H.; Li, Q. X.; Lucas, A.; Szekacs, A.; Sundaram, K. M. S.
Immunochemical technology in environmental analysis:
addressing critical problems. In Immunochemical Methods
for Environmental Analysis; Van Emon, J ., Mumma, R.,
Eds.; ACS Symposium Series 442; American Chemical
Society: Washington, DC, 1990; pp 112-139.
He, F.; Sun, J .; Han, K.; Wu, Y.; Yao, P. Effects of pyrethroid
insecticides on subjects engaged in packaging pyrethroids.
Br. J . Ind. Med. 1988, 45, 548-551.
Tijssen, P. In Practice and Theory of Enzyme Immunoassay;
Elsevier: Amsterdam, The Netherlands, 1985.
Tomlin, C. D. S. The Pesticide Manual, 11th ed.; Tomlin, C.
D. S., Ed.; British Crop Protection Council: Bracknell, U.K.,
1997; pp 470-471.
Voller, A.; Bidwell, D. E.; Bartlett, A. Enzyme immunoassays
in diagnostic medicine: Theory and practice. Bull. W.H.O.
1976, 53, 55-64.
Wells, M. J . M.; Riemer, D. D.; Wellsknecht, M. C. Develop-
ment and optimization of a solid-phase extraction scheme
for detection of a pesticide metribuzin, atrazine, metolachlor
and esfenvalerate in agricultural runoff water. J . Chro-
matogr. A 1994, 659, 337-348.
Wengatz, I.; Stoutmire, D. W.; Gee, S. J .; Hammock, B. D.
Development of an enzyme-linked immunosorbent assay for
detection of the pyrethroid insecticide fenpropathrin. J .
Agric. Food Chem. 1998, 46, 2211-2221.
He, F.; Wang, S.; Liu, L.; Chen, S.; Zhang, Z.; Sun, J . Clinical
manifestations and diagnosis of acute pyrethroid poisoning.
Arch. Toxicol. 1989, 63, 54-58.
Wing, K. D.;Hammock, B. D. Stereoselectivity ofa radioimmuno-
assay for the insecticide S-bioallethrin. Experientia 1979,
35, 1619-1620.
Hengel, M. J .; Mourer, C. R.; Shibamoto, T. New method for
analysis of pyrethroid insecticides: Esfenvalerate, cis-
permethrin, and trans-permethrin, in surface waters using
solid-phase extraction and gas chromatography. Bull. En-
viron. Contam. Toxicol. 1997, 59, 171-178.
Kidd, H.; J ames, D. R. The Agrochemical Handbook; Royal
Society of Chemistry: Thomas Graham House, Cambridge,
U.K., 1995.
Received for review March 17, 2000. Revised manuscript
received J une 20, 2000. Accepted J une 26, 2000. This project
was funded by National Institute of Environmental Health
Sciences (NIEHS) Superfund Basic Research program, 5 P42
ES04699-08, the NIEHS Center for Environmental Health
Sciences, and the Environmental Protection Agency Center for
Ecological Health Research, CR819658.
Kleschick, W. A. Stereoselective synthesis of cis-3-(2,2-di-
chlorovinyl)-2,2-dimethylcyclopropanecarboxylic acid. J . Org.
Chem. 1986, 51, 5429-5433.
J F000351X