Development of competitive enzyme-linked immunosorbent assays for boscalid determination in fruit juices

Article abstract of DOI:10.1016/j.foodchem.2012.04.090

Boscalid is a modern, broad-spectrum carboxamide pesticide highly efficient against most fungal diseases affecting valuable crops. In this study, a boscalid-mimicking derivative with a six-carbon spacer arm replacing the chlorine atom at the pyridine ring of the target molecule was synthesized and coupled to carrier proteins. Following rabbit immunization, antibodies against this agrochemical were obtained for the first time, and they were characterised in terms of affinity and specificity, tolerance to solvents, and robustness to changes in buffer pH and ionic strength, using two assay formats. Both of the optimised immunoassays showed limits of detection below 0.1 μg/L. Moreover, matrix effects of grape, peach, apple, and tomato juices were evaluated. Finally, a simple and easy procedure was set up for boscalid determination with spiked samples, affording limits of quantification of 10 μg/L, a value well below the sensitivity levels required for monitoring campaigns of pesticide residue analysis in food.

Full text of DOI:10.1016/j.foodchem.2012.04.090

Food Chemistry 135 (2012) 276–284
Contents lists available at SciVerse ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
Development of competitive enzyme-linked immunosorbent assays for boscalid
determination in fruit juices
Antonio Abad-Fuentes a, Francesc A. Esteve-Turrillas a, Consuelo Agulló b, Antonio Abad-Somovilla b,
,
⇑
Josep V. Mercader a,
⇑
a Department of Biotechnology, IATA-CSIC, Agustí Escardino 7, 46980 Paterna, València, Spain
b Department of Organic Chemistry, Universitat de València, Doctor Moliner 50, 46100 Burjassot, València, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Boscalid is a modern, broad-spectrum carboxamide pesticide highly efficient against most fungal diseases
affecting valuable crops. In this study, a boscalid-mimicking derivative with a six-carbon spacer arm
replacing the chlorine atom at the pyridine ring of the target molecule was synthesized and coupled to
carrier proteins. Following rabbit immunization, antibodies against this agrochemical were obtained
for the first time, and they were characterised in terms of affinity and specificity, tolerance to solvents,
and robustness to changes in buffer pH and ionic strength, using two assay formats. Both of the optimised
Received 9 February 2011
Received in revised form 26 October 2011
Accepted 15 April 2012
Available online 24 April 2012
Keywords:
Pesticide
Fungicide
Hapten
immunoassays showed limits of detection below 0.1
apple, and tomato juices were evaluated. Finally, a simple and easy procedure was set up for boscalid
determination with spiked samples, affording limits of quantification of 10 g/L, a value well below
the sensitivity levels required for monitoring campaigns of pesticide residue analysis in food.
Ó 2012 Elsevier Ltd. All rights reserved.
lg/L. Moreover, matrix effects of grape, peach,
l
Antiserum
Immunoassay
Competitive ELISA
Agrochemical residues
Food safety
1. Introduction
commercial use and to avoid resistance, it is frequently formulated
together with an active principle of the strobilurin family such as
pyraclostrobin, dimoxystrobin, or kresoxim-methyl. In line with
the ‘‘Good Agricultural Practices’’ concept defined by the FAO/
WHO aimed at protecting consumers, most countries have ap-
proved different legislations stating the maximum residues levels
(MRLs) of pesticides in food and feed (http://www.mrldatabase).
As indicated by the European Food Safety Authority (EFSA Journal,
2010), the tolerated levels for BL in some relevant food commodi-
ties in the EU are: 10 mg/kg for lettuce, spinach, most leafy bras-
sica, and strawberries; 5 mg/kg for kiwifruits and table and wine
grapes; 3 mg/kg for stone fruits; 2 mg/kg for pome fruits and most
head brassica; and 1 mg/kg for flowering brassica and tomatoes
(http://ec.europa.eu/sanco_pesticides/public/index.cfm). Similar
values are reported by the Codex Alimentarius Commission
(www.codexalimentarius.net) as well as by the US Environmental
Boscalid (BL, see Fig. 1 for the chemical structure) is a broad
spectrum fungicide that is being intensively employed throughout
the world to fight against highly destructive plant pathogens, such
as Botrytis cinerea, Sclerotinia spp., Leveillula taurica, or Spherotheca
macularis. It is the only member of the pyridine carboxamide group
of pesticides and it shows a biological mode of action consisting in
the inhibition of the enzyme succinate ubiquinone reductase, also
known as complex II, in the mitochondrial electron transport chain
(Avenot & Michailides, 2007). BL is a blockbuster fungicide of
BASF’s crop protection pipeline. In 2003, it was approved in the
US and in 2008 it was included into Annex I of the Council Direc-
tive 91/414/EEC, that is, the EU positive list of authorised agro-
chemicals (http://www.pesticideinfo.org; EC Regulation, 2008.).
Since then, due to its excellent performance, the actual peak sales
potential has increased to more than €300 million. This fungicide is
currently used to treat over 100 crop varieties, including vegeta-
bles, fruits, and cereals, across more than 70 countries and has
more than 200 indications (http://www.agro.basf.com). For
Protection
Agency
(http://www.epa.gov/pesticides/PPISdata/
index.html).
Current analytical methodologies for the determination of BL in
food and feed are based on chromatographic separations by both
gas and liquid chromatography coupled to mass spectrometry
detectors. BL can be extracted from the sample by alternative
methodologies, though they are mostly based on the use of high
amounts of organic solvents. Acetone has been employed to
⇑
Corresponding authors. Tel.: +34 963544509; fax: +34 963544328
(A. Abad-Somovilla), tel.: +34 963900022; fax: +34 963636301 (J.V. Mercader).
E-mail addresses: antonio.abad@uv.es (A. Abad-Somovilla), jvmercader@iata.
csic.es (J.V. Mercader).
0308-8146/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2012.04.090

A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
277
Fig. 1. Synthetic pathway of hapten BLa5, the activation reaction with disuccinimidyl carbonate, and the structure of boscalid.
effectively extract BL and other pesticides from different fruit and
vegetable matrices, followed by a clean-up step based on liquid–li-
quid extraction with dichloromethane (Hiemstra & de Kok, 2007;
Liu, Dong, Qin, & Zheng, 2010). For multi-residue analysis, the QuE-
ChERS (Quick Easy, Cheap, Effective, Rugged, and Safe) methodol-
ogy, which employs acetonitrile to extract pesticides together
with a clean-up of the extract based on solid-phase extraction with
primary–secondary amine (Lehotay, 2007), or with C18 or graphi-
tized carbon black phases (Walorczyk, 2008; Walorczyk & Gnu-
sowski, 2009), is widely used and it has been broadly validated.
However, sample extraction and clean-up demand important hand
labour, thus decreasing sample throughput and increasing costs. In
addition, the use of organic solvents has a negative impact on the
environment. During the past decade, many competitive enzyme-
linked immunosorbent assays (cELISA) have been published for
the analysis of residues of a broad variety of pesticides in different
kinds of food samples, and several immunochemical methodolo-
gies have been approved to tackle the analysis of chemical residues
in food, feed, and environmental samples (http://www.epa.gov/
pesticides/methods/index.htm). Competitive immunoassays are
required because of the small size of the target molecules. They
have been widely used due to their simplicity, high sample
throughput, low cost, and minimal environmental impact. More-
over, immunoassays constitute a highly versatile technology, so
they can be implemented in many different analytical platforms
and adapted to specific applications. Particularly, cELISAs in a
microtiter plate can be developed using different formats for the
analysis of small chemical molecules, yet the antibody-coated di-
rect competitive assay (d-cELISA) and the conjugate-coated indi-
rect competitive assay (i-cELISA) are the most common formats.
To our knowledge, neither the production of antibodies nor the
development of immunoassays for BL residue analysis in food has
been published so far. In this article, we report the original synthe-
sis of a functionalised derivative used for the generation of novel
antibodies against BL and the development and characterisation
of direct and indirect immunoassays for the analysis of this rele-
vant pesticide in foodstuffs. The cross-reactivity of the developed
immunoassays towards commonly used fungicides was evaluated
and, after optimization of assay parameters, the selected analytical
procedures were applied to fortified juices of relevant commodi-
ties, such as grapes, peaches, apples, and tomatoes.
2. Experimental
2.1. Reagents and instrumentation
BL (2-chloro-N-(40-chlorobiphenyl-2-yl)nicotinamide, CAS Reg-
istry No. 188425-85-6, MW 343.2 g/mol) and other pesticide stan-
dards were purchased from Fluka/Riedel-de-Haën (Seelze,
Germany) or Dr. Ehrenstorfer (Augsburg, Germany). Technical
grade BL was kindly provided by BASF. Horseradish peroxidase
(HRP), ovalbumin (OVA), and o-phenylenediamine were purchased
from Sigma–Aldrich (Madrid, Spain). Sephadex G-25 HiTrap
Desalting columns from GE Healthcare (Uppsala, Sweden) were
used for conjugate purification with an ÄKTA workstation. Poly-
clonal goat anti-rabbit immunoglobulin peroxidase conjugate
(GAR–HRP) was from BioRad (Hercules, CA, USA). Bovine serum
albumin (BSA) fraction V was purchased from Roche Applied Sci-
ence (Mannheim, Germany). Fetal bovine serum (FBS) and Freund’s
adjuvants were from Sigma–Aldrich (Madrid, Spain). Costar flat-
bottom high-binding polystyrene ELISA plates were from Corning
(Corning, NY, USA). Ultraviolet–visible spectra and ELISA absor-
bances were read with a PowerWave HT from BioTek Instruments

278
A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
(Winooski, VT, USA). ELISA plates were washed with an ELx405
microplate washer also from BioTek Instruments.
nearly pure diacid 3 (735.4 mg, 89%) as a white solid. Mp 145–
148 °C (from C6H6–methanol); IR vmax/cmꢁ1 (KBr) 3495, 3100–
2500, 3071, 2957, 1716, 1682, 1561, 1423, 1389, 1300, 1224,
1076, 763, 718, 558; 1H NMR (300 MHz, DMSO-d6) d: 8.60 (1H,
dd, J = 4.6, 1.8 Hz, H-6 Py), 8.18 (1H, dd, J = 7.6, 1.8 Hz, H-4 Py),
7.21 (1H, dd, J = 7.6, 4.6 Hz, H-5 Py), 3.09 (2H, t, J = 6.3 Hz, H-1),
2.26 (2H, t, J = 6.6 Hz, H-4), 1.62 (4H, m, H-2 and H-3); 13C NMR
(75 MHz, CDCl3) d 172.27 (COOH), 166.30 (Py-COOH), 160.79 (C-
2 Py), 151.82 (C-6 Py), 138.88 (C-4 Py), 123.53 (C-3 Py), 118.62
(C-5 Py), 33.18 (C-1), 28.74 (C-4), 27.98 (C-2), 23.91 (C-3); MS
(EI) m/z 255 (M+, 28), 222 (4), 211 (3), 210 (6), 209 (12), 207
(21), 182 (21), 169 (100); HRMS (EI), calculated for C11H13NO4S
255.05603, found 255.05661.
The composition, concentration, and pH of the employed buf-
fers were as follows: (i) PB, 100 mM sodium phosphate buffer,
pH 7.4; (ii) PBS, 10 mM sodium phosphate buffer, pH 7.4 with
140 mM NaCl; (iii) PBST, PBS containing 0.05% (v/v) Tween 20;
(iv) 2 ꢀ PBST, 20 mM sodium phosphate buffer, pH 7.4 with
280 mM NaCl and 0.05% (v/v) Tween 20; (v) CB, 50 mM sodium
carbonate–bicarbonate buffer, pH 9.6; (vi) Washing solution,
150 mM NaCl and 0.05% (v/v) Tween 20; and (vii) Developing buf-
fer, 25 mM sodium citrate and 62 mM sodium phosphate buffer,
pH 5.4.
2.2. Hapten synthesis
2.2.2. Preparation of 2-(5-methoxy-5-oxopentylthio)nicotinic acid (4)
Previously known 40-chloro-[1,10-biphenyl]-2-amine (12)
(Bradsher & Wissow, 1946; Felpin, Fouquet, & Zakri, 2009) was
readily prepared in high yield in two steps from 1-iodo-2-nitroben-
zene (9) and 4-chlorophenylboronic acid (10) as described in the
Supplementary Data (Fig. S1). 2-Mercaptonicotinic acid (1), 5-
bromopentanoic acid (2) and all the other reagents used for the
synthesis of the functionalised hapten were acquired from com-
mercial sources and used without purification. Reactions were per-
formed in oven-dried glassware under a nitrogen atmosphere
containing a Teflon-coated stirrer bar and a dry septum. For the
exclusion of atmospheric oxygen from the reaction media, three
freeze-pump thaw cycles were preformed before the reagents were
mixed. All solvents were purified by distillation and, if required,
they were dried according to standard methods. The reactions
were monitored with the aid of thin-layer chromatography using
0.25 mm pre-coated silica gel plates. Visualisation was carried
out with UV light and aqueous ceric ammonium molybdate solu-
tion. Chromatography refers to flash column chromatography
and it was carried out with the indicated solvents on silica gel 60
(particle size 0.040–0.063 mm). All melting points were deter-
mined using a Kofler hot-stage apparatus or a Büchi melting point
apparatus and are uncorrected. All NMR spectra were recorded in
CDCl3 or DMSO-d6 at room temperature (rt) on a Bruker AC-300
spectrometer (300.13 MHz for 1H and 75.47 MHz for 13C). The
spectra were referenced to residual solvent protons in the 1H
NMR spectra (7.26 and 2.50 ppm) and to solvent carbons in the
13C NMR spectra (77.0 and 39.43 ppm). Carbon substitution de-
grees were established by distortionless enhancement by polarisa-
Trimethylsilyl chloride (ca. 20 lL, ca. 0.158 mmol) was added to
a stirred solution of diacid 3 (376 mg, 1.58 mmol) in 2,2-dime-
thoxypropane (2.4 mL) and anhydrous methanol (1.6 mL). The
mixture was stirred at rt for a week, poured into water, and ex-
tracted with ethyl acetate. The combined organic layers were
washed with brine, dried over anhydrous MgSO4, and concentrated
to give pure methyl ester 4 (338 mg, 85%) as a white solid. Mp 102–
103 °C (from C6H6); IR vmax/cmꢁ1 (KBr) 3300–2500, 3033, 2964,
1728, 1671, 1574, 1554, 1421, 1299, 1388, 1187, 1071, 888, 764,
718, 558; 1H NMR (300 MHz, CDCl3) d: 8.59 (1H, dd, J = 4.6,
1.8 Hz, H-6), 8.31 (1H, dd, J = 7.8, 1.8 Hz, H-4), 7.08 (1H, dd,
J = 7.8, 4.6 Hz, H-5), 3.67 (3H, s, CO2Me), 3.21 (2H, t, J = 6.9 Hz, H-
1), 2.38 (2H, t, J = 7.2 Hz, H-4), 1.80 (4H, m, H-2 and H-3); 13C
NMR (75 MHz, CDCl3) d: 173.98 (CO2Me), 169.61 (COOH),
163.21(C-2 Py), 152.62(C-6 Py), 139.85 (C-4 Py), 121.75 (C-3 Py),
118.15 (C-5 Py), 51.55 (CO2Me), 33.69 (C-1), 29.51 (C-4), 28.44
(C-2), 24.38 (C-3); MS (EI) m/z 269 (M+, 28), 239 (2), 238 (21),
236 (4), 224 (6), 210 (4), 182 (18), 169 (100); HRMS (EI), calculated
for C12H15NO4S 269.07218, found 269.07342.
2.2.3. Preparation of methyl 5-((3-((40-chloro-[1,10-biphenyl]-2-
yl)carbamoyl)pyridin-2-yl)thio) pentanoate (6)
Oxalyl chloride (35.8
stirred solution of acid 4 (175 mg, 0.65 mmol) in anhydrous CH2Cl2
(4 mL) followed by a catalytic amount (16 L) of dry N,N-dimethyl-
lL, 0.715 mmol) was added drop wise to a
l
formamide (DMF). The mixture was stirred at rt for 24 h and then
the solvent and the excess of reagent were removed under vacuum.
Dry benzene (10 mL) was added and the distillation repeated to
give crude methyl 5-(3-(chlorocarbonyl)pyridin-2-ylthio)pentano-
ate (5) (186.8 mg) that was used in the next step without further
purification.
tion transfer pulse sequences.
A combination of correlation
spectroscopy and heteronuclear single quantum coherence exper-
iments was used in most cases for the assignment of 1H and 13C
chemical shifts. Infrared (IR) spectra were measured as thin films
between NaCl plates for liquid compounds and as KBr pellets for
solids using a Nicolet Avatar 320 spectrometer. Mass spectra
(MS) and high-resolution mass spectra (HRMS) were run by the
A mixture of the crude acid chloride obtained above, 40-chloro-
[1,10-biphenyl]-2-amine (12) (100 mg, 0.488 mmol), pyridine
(52.6 lL, 0.65 mmol), and a catalytic amount of 4-dimethylamino-
pyridine (DMAP) (1.61 mg) in anhydrous CH2Cl2 (8 mL) was stirred
at rt for 3 days. After this time, the reaction mixture was poured
into water and extracted with ethyl acetate. The organic extracts
were washed with brine and dried over anhydrous Na2SO4. The
residue left after evaporation of the solvent was purified by flash
chromatography eluting with hexane–ethyl acetate mixtures (from
9:1 to 6:4) to afford the amide 6 (202 mg, 91%) as a white solid. Mp
105–107 °C (from hexane–C6H6); IR vmax/cmꢁ1 (KBr) 3447, 3246,
2945, 1735, 1651, 1514, 1478, 1292, 1173, 1089, 762; 1H NMR
(300 MHz, CDCl3) d: 8.45 (1H, dd, J = 4.8, 1.8 Hz, H-6 Py), 8.41
(1H, br d, J = 8.1 Hz, H-3 PhPh), 8.12 (1H, br s, NH), 7.79 (1H, dd,
J = 7.8, 1.8 Hz, H-4 Py), 7.44–7.22 (7H, m, H-20/H-60, H-30/H-50, H-
4, H-5, and H-6 PhPh), 7.04 (1H, dd, J = 7.8, 4.8 Hz, H-5 Py), 3.66
(3H, s, CO2Me), 3.14 (2H, t, J = 6.9 Hz, H-5), 2.36 (2H, t, J = 7.2 Hz,
H-2), 1.68 (4H, m, H-3 and H-4); 13C NMR (75 MHz, CDCl3) d:
173.80 (CO2Me), 164.45 (CON), 156.51 (C-2 Py), 150.70 (C-6 Py),
136.73 (C-4 Py), 136.59 (C-2 PhPh), 134.49 and 134.09 (C-10 and
electron impact (EI) at 70 eV in
spectrometer.
a Micromass VG Autospec
Compounds used in this study present minor safety concerns.
However, it is advisable to work in a well-ventilated fume hood
during synthesis reactions. A schematic representation of the syn-
thetic procedures is depicted in Fig. 1.
2.2.1. Preparation of 2-(4-carboxybutylthio)nicotinic acid (3)
A solution of sodium 5-bromopentanoate in H2O, prepared from
5-bromopentanoic acid (2) (700.1 mg, 3.89 mmol) and NaHCO3
(326.7 mg, 3.89 mmol) in 3 mL of H2O, was added drop wise to a
solution of 2-mercaptonicotinic acid (1) (500 mg, 3.23 mmol) in
10% aqueous KOH (5 mL) at rt. The reaction mixture was stirred
at 60 °C for 4 h, cooled in an ice-water bath, and acidified with con-
centrated HCl to pH 2–3. The solid precipitated product was col-
lected, washed with water, and dried under vacuum to give

A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
279
C-40 PhPh), 132.01 (C-3 Py), 130.92 (C-20/C-60), 130.14 (C-4 PhPh),
2.3.1. Immunizing conjugate
129.61 (C-1 PhPh), 129.17 (C-30/C-50), 128.81 (C-6 PhPh), 124.99
(C-5 PhPh), 122.23 (C-3 PhPh), 119.05 (C-5 Py), 51.53 (CO2Me),
33.54 (C-5), 29.81 (C-2), 28.63 (C-4), 24.17 (C-3); MS (EI) m/z 454
(M+, 9), 425 (6), 123 (16), 342 (11), 341 (11), 340 (12), 339 (21),
253 (9), 258 (39), 252 (39), 251 (100); HRMS (EI), calculated for
A conjugate to BSA was prepared using 200
l
L of a 50 mM solu-
tion of the N-succinimidyl ester 8 in anhydrous DMF. Briefly, the
active ester solution was added drop wise over a 15 mg/mL BSA
solution (2 mL) in CB and the mixture was gently stirred at rt in
amber glass vials during 4 h. The conjugate was purified by gel fil-
tration using PB as eluent. The collected volume was brought to
30 mL with PB and the conjugate was stored frozen at ꢁ20 °C.
The calculated final MR of the immunogen was 11.
C
24H2335ClN2O3S 454.11179, found 454.11181.
2.2.4. Preparation of 5-((3-((40-chloro-[1,10-biphenyl]-2-
yl)carbamoyl)pyridin-2-yl)thio)pentanoic acid (7, hapten BLa5)
A solution of methyl ester 6 (112 mg, 0.246 mmol) in methanol
(5.1 mL) and aqueous 2 M NaOH (1.23 mL) was stirred at rt for 4 h.
Most of the solvent was evaporated under reduced pressure and
the resulting residue diluted with water (2 mL) and acidified with
formic acid. The white precipitate formed was filtered out, washed
with water, and dried under vacuum to give pure hapten BLa5 (7)
(97 mg, 89%) as a white solid. Mp 172–175 °C (from cold metha-
nol); IR vmax/cmꢁ1 (KBr) 3500–2500, 3435, 3305, 3035, 2928,
2853, 1691, 1653, 1513, 1473, 1444, 1388, 757; 1H NMR
(300 MHz, DMSO-d6) d: 12.05 (1H, br s, COOH), 10.03 (1H, br s,
NH), 8.50 (1H, dd, J = 4.8, 1.8 Hz, H-6 Py), 7.67 (1H, br d,
J = 6.9 Hz, H-3 PhPh), 3.39–7.35 (9H, m, H-20/H-60, H-30/H-50, H-4,
H-5 and H-6 PhPh and H-4), 7.19 (1H, dd, J = 7.5, 4.8 Hz, H-5 Py),
3.10 (2H, t, J = 6.0 Hz, H-5), 2.25 (2H, t, J = 6.3 Hz, H-2), 1.61 (4H,
m, H-3 and H-4); 13C NMR (75 MHz, DMSO-d6) d: 174.17(COOH),
165.29 (CON), 156.94 (C-2 Py), 149.88 (C-6 Py), 135.07 (C-4 Py),
137.83 (C-2 PhPh), 135.07 (C-10 PhPh), 134.22 (C-40 PhPh), 131.95
(C-3 Py), 130.62 (C-1 PhPh), 130.48 (C-20/C-60), 130.09 (C-4 PhPh),
128.15 (C-30/C-50 PhPh and C-6 PhPh), 127.77 and 126.84 (C-3 and
C-5 PhPh), 118.60 (C-5 Py), 33.13 (C-5), 28.66 (C-2), 28.32 (C-4),
23.74 (C-3); MS (EI) m/z 440 (M+, 7), 342 (3), 341 (11), 340 (15),
339 (25), 239 (6), 238 (35), 237 (100); HRMS (EI), calculated for
2.3.2. Coating conjugate
Several OVA conjugates were prepared with different starting
quantities of the N-succinimidyl ester 8 in order to afford a collec-
tion of coating antigens with a range of hapten-to-protein MRs. A
fixed volume (100 lL) of a solution of the N-succinimidyl ester 8
in DMF (10, 50, or 100 mM) was added drop wise to 2 mL of a
15 mg/mL OVA solution in CB under stirring. The reaction was
gently mixed at rt during 2.5 h in amber glass vials and the conju-
gate was purified as before. The pooled fractions containing the
conjugate were brought to 30 mL using PB and stored frozen at
ꢁ20 °C.
2.3.3. Enzyme tracer
For the preparation of a reporter conjugate, HRP was used as
carrier. One hundred microlitres of activated hapten solution in
DMF (10 mM) was added drop wise to a 2.2 mg/mL HRP solution
(1 mL) in CB under gentle stirring. The coupling reaction was al-
lowed to proceed during 4 h at rt with moderate stirring in amber
glass vials. The conjugate was separated from uncoupled hapten by
gel chromatography and it was brought to 1 mg/mL with PB, and
then diluted 1/2 with PBS containing 1% (w/v) BSA and 0.01% (w/
v) thimerosal. Tracer conjugate aliquots were stored at ꢁ20 °C in
amber vials except the working solution that was kept at 4 °C.
C
23H2135ClN2O3S 440.09614, found 440.09463; UV (PB),
(280 nm) = 5.42 mMꢁ1 cmꢁ1 (260 nm) = 10.68 mMꢁ1 cmꢁ1
e
,
e
.
2.4. Production of rabbit polyclonal antibodies
2.2.5. Preparation of N-succinimidyl 5-((3-((40-chloro-[1,10-biphenyl]-
2-yl)carbamoyl)pyridin-2-yl)thio)pentanoate (8)
Two antisera were generated, named rBLa5#1 and rBLa5#2,
from two female New Zealand white rabbits weighing 1–2 kg,
which had been immunised by subcutaneous injection with
0.3 mg of BSA–BLa5 conjugate in 1 mL of a 1:1 mixture of PB and
complete Freund’s adjuvant. Animals were boosted at 21-day
intervals with the same immunogen suspended in a mixture of
0.5 mL of PB and 0.5 mL of incomplete Freund’s adjuvant. Whole
blood was collected from the ear vein of the rabbits and by intra-
cardiac puncture 10 days after the fourth injection. Blood samples
were allowed to coagulate overnight at 4 °C. Then, the serum was
separated by centrifugation and precipitated with a solution of sat-
urated ammonium sulphate. This procedure was repeated again
and the precipitates were stored at 4 °C. Animal manipulation
was performed in compliance with the laws and guidelines of
the Spanish Ministry of Agriculture, Fisheries, and Food.
To activate hapten BLa5 (7) for protein conjugation, the N-succ-
inimidyl ester was prepared (Fig. 1) according to a previously de-
scribed procedure (Esteve-Turrillas et al., 2010). Briefly, acid 7
(30 mg, 0.068 mmol) and N,N0-disuccinimidyl carbonate (DSC)
(27.9 mg, 0.109 mmol, 1.6 equiv.) were dissolved in 0.9 mL of dry
acetonitrile under nitrogen atmosphere. Then, triethylamine
(36 lL, 0.258 mmol, 3.8 equiv.) was added and the reaction mix-
ture was stirred at rt for 5 h. The solution was diluted with chloro-
form, washed with an aqueous saturated solution of NaHCO3 and
brine, and then dried over anhydrous Na2SO4. The residue left after
evaporation of the solvent was purified by column chromatogra-
phy, using chloroform as eluent, to give the pure active N-succin-
imidyl ester of hapten BLa5 8 (29.3 mg, 80%) as a viscous oil. 1H
NMR (300 MHz, CDCl3) : d 8.47 (1H, dd, J = 4., 1.8 Hz, H-6 Py),
8.41 (1H, br d, J = 8.1 Hz, H-3 PhPh), 8.11 (1H, br s, NH), 7.79 (1H,
dd, J = 7.8, 1.8 Hz, H-4 Py), 7.46–7.21 (7H, m, H-20/H-60, H-30/H-50,
H-4, H-5 and H-6 PhPh), 7.05 (1H, dd, J = 7.8, 4.8 Hz, H-5 Py),
3.17 (2H, t, J = 6.9 Hz, H-5), 2.83 (4H, br s, COCH2CH2CO), 2.67
(2H, t, J = 7.2 Hz, H-2), 1.90–1.60 (4H, m, H-3 and H-4).
2.5. Antibody-coated direct competitive ELISA
Ninety-six-well polystyrene ELISA plates were coated with
100
incubated overnight at rt. Coated plates were washed four times
with washing solution and received, afterwards, 50 L per well of
analyte standard in PBS plus 50 L per well of HRP–BLa5 tracer
lL of a particular antiserum diluted in CB, and plates were
2.3. Preparation of protein–BLa5 conjugates
l
l
Protein conjugates were prepared by reaction of the purified N-
succinimidyl ester of BLa5 (8) with the free amine groups of the
carrier protein. The final hapten-to-protein molar ratio (MR) was
calculated using the absorbance values of the conjugate by assum-
ing that the molar absorption coefficients of the hapten and the
protein were the same for the free and the conjugated forms.
solution in PBST. The immunological reaction took place during
1 h at rt, and then plates were washed as described above. Finally,
retained peroxidase activity was determined by addition of 100 lL
per well of freshly prepared 2 mg/mL o-phenylenediamine solution
containing 0.012% (v/v) H2O2 in developing buffer. The enzymatic
reaction was stopped after 10 min at rt by addition of 100 lL per

280
A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
well of 2.5 M sulphuric acid. The absorbance was immediately read
at 492 nm with a reference wavelength at 650 nm.
3. Results and discussion
3.1. Synthesis of BLa5 and its N-succinimidyl ester
2.6. Conjugate-coated indirect competitive ELISA
Antigens should closely resemble the target ligand, though frag-
ments of the molecule have been successfully used as mimics
(Connolly, Fodey, Crooks, Delahaut, & Elliott, 2002; Fodey, Dela-
haut, & Elliott, 2007; Mercader, Primo, & Montoya, 1995). Our ap-
proach was to prepare a functionalised hapten (BLa5) containing
the three characteristic rings of BL. Moreover, this compound con-
tained a five-carbon atom spacer arm linked to the pyridine ring
through a sulphur atom which replaced a chlorine atom present
in the molecule. The sulphur atom behaves as a good mimetic of
the chlorine atom, both in electronic and steric terms. This strategy
has been successfully followed before by our group (Mercader,
Agulló, Abad-Somovilla, & Abad-Fuentes, 2011) and others (Manc-
lús, Primo, & Montoya, 1996; Matsuo, Nishi, Morimune, Okawa, &
Miyake 2005) in the design of haptens for the production of anti-
bodies against chlorine-containing molecules. First attempts to
carry out the direct nucleophilic replacement of the chlorine atom
ELISA plates were coated with 100 lL per well of OVA–BLa5
solution in CB by overnight incubation at rt. Coated plates were
washed four times with washing solution and then received
50
lL per well of analyte in PBS plus 50 lL per well of antiserum
dilution in PBST. The immunological reaction took place during
1 h at rt and plates were washed again as before. Next, 100 lL
per well of a 1/104 dilution of GAR–HRP conjugate in PBST contain-
ing 10% FBS was added, and plates were incubated 1 h at rt. After
washing the plates, the retained peroxidase activity was deter-
mined as aforementioned.
2.7. Standards and data treatment
From a stock solution of BL at 10 g/L in anhydrous DMF, eight-
of BL by the (4-carboxybutyl)mercaptide group failed, so
straightforward synthetic sequence was designed for the prepara-
tion of the desired hapten (Fig. 1).
a
point standard curves were prepared from 10ꢁ2 to 104
lg/L by a
10-fold serial dilution in PBS, plus a blank (no analyte). Experimen-
tal values were fitted to a four-parameter logistic equation using
the SigmaPlot software package from SPSS Inc. (Chicago, IL, USA).
Assay sensitivity was defined as the concentration of analyte at
the inflection point of the fitted curve, typically corresponding to
a 50% inhibition (IC50) of the maximum absorbance (Amax) if the
background signal approaches to zero.
The synthesis of hapten BLa5 was initiated with the preparation
of nicotinic acid derivative 3. This intermediate was obtained in
two steps from commercially available 2-mercaptonicotinic acid
(1). Thus, reaction of 1 with 5-bromopentanoic acid (2) in aqueous
potassium hydroxide solution gave de dicarboxylic acid 3 (Da Set-
timo et al., 2000). The aliphatic carboxylic group of 3 was chemose-
lectively protected as its methyl ester by treatment with 2,2-
dimethoxypropane in the presence of a catalytic amount of anhy-
drous HCl, which was generated in situ from trimethylchlorosilane
(Rodriguez, Nomen, & Spur, 1998), to afford the intermediate
methyl ester 4 in about 76% overall yield for the two steps. With
compound 4 at hand, the synthesis of the hapten BLa5 was readily
completed as follows. First, the aromatic carboxylate acid group
was transformed into the corresponding acid chloride (5), which
was subsequently coupled to 40-chlorobiphenyl-2-amine (12)
without purification to give the methyl ester of hapten BLa5 (6)
in an excellent 91% overall yield. Finally, basic-promoted hydroly-
sis of the methyl ester moiety of 6 afforded the hapten BLa5 (7) in
89% yield after its chromatographic purification.
2.8. Influence of buffer composition on assay performance
Two immunoassays were selected using each of the two most
common cELISA formats. The first assay used the direct format
and plates coated with a 1/6 ꢀ 104 dilution of antibody rBLa5#1.
The second ELISA employed the indirect format and plates coated
with a 0.1 lg/mL solution of OVA–BLa5 (MR = 2) and antibody
rBLa5#2. BL standard solutions were mixed 1:1 with a 20 ng/mL
tracer solution for the d-cELISA, or with a 1/2 ꢀ 105 antibody dilu-
tion for the i-cELISA. For solvent studies, analyte standard curves
were prepared in water containing between 0.5% and 10% of meth-
anol, ethanol, acetonitrile, or acetone, and immunoreagents were
prepared in 2 ꢀ PBST. The influence of the buffer ionic strength
and the pH was evaluated following a central composite design,
Hapten coupling to proteins is very often performed by cova-
lently linking a carboxylate group of the hapten to the e-amine res-
idues of the carrier polypeptide. A very common strategy to
achieve such conjugates consists in the activation of the carboxylic
acid by formation of an active ester using carbodiimide and cou-
pling to the protein, all performed in a one-pot reaction. During
that reaction, undesired by-products are formed which reduce
the reaction yields and make it difficult to purify the activated hap-
ten. Lately, we have applied in our lab an activation reaction,
uncommon for the preparation of hapten bioconjugates, using
DSC to obtain the N-succinimidyl ester of carboxy-functionalised
haptens. This reagent was originally described by Ogura, Kobay-
ashi, Shimizu, Kawabe, and Takeda (1979) for the synthesis of ac-
tive esters. It is a simple strategy that allows an easy purification
of the active ester by chromatography on silicagel, mostly due to
the absence of significant by-products (Kang et al., 2009; Parra,
Mercader, Agulló, Abad-Fuentes, & Abad-Somovilla, 2011). The ac-
tive N-succinimidyl ester of hapten BLa5 (8) was obtained by this
procedure in an 80% yield after purification (Fig. 1), and it was sta-
ble when stored in solution at ꢁ20 °C under dry conditions. The
availability of pure active ester 8 facilitated the preparation of
immunogen and assay conjugates with tunable and reproducible
hapten-to-protein MRs.
consisting of a two-level full factorial design (a = 1.414) with two
factors and three replicates that included 12 cube, 12 axial, and
15 centre points, and involving a total of 39 randomised buffer
studies (see Table S1 in the Supplementary Data). The ionic
strength values of the evaluated buffers were between 50 and
300 mM, and the studied pH values ranged from 5.5 to 9.5. Initially,
a 40 mM trisodium citrate, 40 mM disodium hydrogen phosphate,
and 40 mM Tris solution was prepared. To aliquots of that solution,
known volumes of 5 M HCl was added in order to reach the re-
quired pH in each case. Then, the ionic strength of each solution
was calculated considering the initial buffer concentrations and
the added HCl, and then an appropriate volume of 2 M NaCl was
used to achieve the required ionic strength for every buffer. At last,
Tween 20 was added to all buffers to a final concentration of 0.05%
(v/v). BL standard curves were prepared in water and they were
mixed as described above with tracer or antibody solutions pre-
pared in every of the studied buffers. The Amax and IC50 values of
the inhibition curves were employed as response values and fitted
to a multiple regression equation, including curvature and interac-
tion terms, using Minitab 14.1 software (Minitab Inc., State College,
PA, USA).

A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
281
Table 1
Assay parameters for the checkerboard competitive assay with antisera rBLa5#1 and rBLa5#2.
Direct format
1/[pAb]
HRP–BLa5 (
l
g/L)
rBLa5#1
rBLa5#2
Amax
Slope
IC50
(lg/L)
Amax
Slope
IC50 (lg/L)
6 ꢀ 104
3 ꢀ 104
10
3
10
3
1.42
0.79
2.30
1.03
0.57
0.52
0.77
0.71
0.9
0.4
2.7
1.5
2.00
1.00
3.37
1.32
0.82
0.61
0.61
0.65
3.3
1.6
5.6
4.8
Indirect format
OVA–BLa5
rBLa5#1
rBLa5#2
1/[pAb]
MR
8
Conc. (mg/L)
1/[pAb]
Amax
Slope
IC50
(lg/L)
Amax
Slope
IC50 (lg/L)
1.0
0.1
1.0
0.1
1.0
0.1
105
1.49
0.63
1.34
0.56
2.11
0.89
1.32
0.57
2.00
0.89
1.62
0.81
0.67
0.76
0.66
0.76
0.57
0.71
0.67
0.66
0.61
0.64
0.80
0.65
48.2
33.5
4.0
3 ꢀ 105
1.54
0.59
1.37
0.48
1.98
0.79
1.42
0.45
1.98
0.79
1.68
0.88
0.70
0.82
0.56
0.75
0.71
0.74
0.71
0.87
0.53
0.55
0.62
0.57
75.2
69.8
5.7
3 ꢀ 105
106
105
3 ꢀ 105
106
3 ꢀ 105
105
3.8
5.0
4
2
20.9
17.5
1.8
1.3
5.6
4.3
2.0
0.9
3 ꢀ 105
38.5
28.7
3.9
4.4
9.1
7.3
1.5
0.9
3 ꢀ 105
106
105
3 ꢀ 105
106
3 ꢀ 105
105
3 ꢀ 105
3 ꢀ 105
3 ꢀ 104
105
106
105
3 ꢀ 105
3.2. Immunoreagent evaluation
3.2.1. Affinity
OVA–BLa5 conjugate solutions (0.01, 0.10, or 1.00 lg/mL), and
the competitive step was carried out using a range of antibody
dilutions (from 1/3 ꢀ 103 to 1/106). As a result, a collection of inhi-
bition curves was obtained – one curve for each pair of immunore-
agent concentrations – in each assay format. Fig. S2 of the
Supplementary Data section shows one of those series of curves
for one antibody (rBLa5#2) in the i-cELISA format using coating
conjugates with different MRs. Irrespectively of hapten loading in
Two rabbits were immunised with conjugate BSA–BLa5, and the
cleared-out antisera, called rBLa5#1 and rBLa5#2, were obtained
and evaluated. In order to find the proper immunoreagent concen-
tration, a checkerboard competitive screening analysis for each
antibody was performed in two immunoassay formats (d- and i-
cELISA), as previously described (Mercader, Suárez-Pantaleón,
Agulló, Abad-Somovilla, & Abad-Fuentes, 2008). For direct assays,
plates were coated with several dilutions of the antiserum (from
1/104 to 1/105), and next day a range of tracer conjugate concentra-
tions (from 3 to 100 ng/mL) was evaluated under competitive con-
ditions. For indirect assays, plates were coated with different
the bioconjugate, coating plates with OVA–BLa5 at 0.1 lg/mL re-
sulted in the best inhibition curves. A summary of the results of
the competitive checkerboard study, containing the respective
curve parameters (Amax, slope, and IC50 values), is listed in Table
1 for direct and indirect assays (only results for the two immunore-
agent concentrations giving Amax values immediately above and
Fig. 2. Effect of organic solvents over the maximum signal and sensitivity of the d-cELISA using antibody rBLa5#1 (blue bars) and the i-cELISA using antibody rBLa5#2 (red
bars). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

282
A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
Fig. 3. Influence of buffer conditions over the assay curve parameters of the rBLa5#1-based d-cELISA and the rBLa5#2-based i-cELISA.
below 1.0 are listed; and in the case of the i-cELISAs, data with
0.01 g/mL conjugate are not included).
l
100
80
60
40
20
0
d-cELISA with rBLa5#1
i-cELISA with rBLa5#2
For selection of the best reagent combination, the following cri-
teria were used: Amax of the inhibition curve between 1.0 and 1.5,
inferior asymptote equivalent to the background signal, lowest IC50
value, and minimum immunoreagent consumption. The best
immunoassay in the d-cELISA format was achieved with antiserum
rBLa5#1, whereas in the i-cELISA format it was antiserum rBLa5#2
the finest performing one. For the latter format, the sensitivity
could be improved not only with low coating concentrations but
also by reducing the hapten molarity of the coating conjugate.
Thus, conjugate OVA–BLa5 with a MR of 2 at 0.10
lected for this cELISA.
lg/mL was se-
3.2.2. Selectivity
10-3 10-2 10-1 100
101
102
103
104
0
Cross-reactivity studies were performed in order to find other
compounds that could be recognised by the generated antibodies.
Twenty-two widely used fungicides were assayed; i.e., kresoxim-
methyl, trifloxystrobin, pyraclostrobin, azoxystrobin, dimoxystrob-
in, fluoxastrobin, metominostrobin, picoxystrobin, fenhexamid,
captan, mepanipyrim, pyrimethanil, procimidone, tolylfluanid,
cyazofamid, tebuconazole, fenamidone, fludioxonil, vinclozolin,
imidacloprid, cyprodinil, and benzanilide. Calibration curves were
[Boscalid] (µg/L)
Fig. 4. Standard curves in buffer for the two optimised immunoassays. For the d-
cELISA, plates were coated with a 1/6 ꢀ 104 dilution of antiserum rBLa5#1, and
assays were run with 10 ng/mL HRP–BLa5 tracer conjugate. For the i-cELISA,
microwells were coated with a 0.1 lg/mL OVA–BLa5 (MR = 2) solution, and the
dilution of antiserum rBLa5#2 in the immunological reaction was 1/2 ꢀ 105. The
Amax value was around 1.3. For normalisation, the absorbance value of the blank
control that was run in the same microplate was employed. Values are the mean of
three independent determinations.
prepared up to 10 lM for each fungicide in PBS and measured by
the selected direct and indirect ELISAs. From this study, it could
be concluded that both antisera were highly specific to BL because
no inhibition was exerted by any of the studied pesticides.
competitive step was better accepted by the BLa5#1-based direct
assay than by the BLa5#2-based indirect one. Variations of the Amax
value were below 10% when the solvent concentration in the sam-
ple was equal or lower than 10% in the direct assay and 2% in the
indirect assay. Regarding sensitivity, the IC50 values were also
essentially unaffected by a 10% ethanol or acetonitrile in the
rBLa5#1-based assay, whereas methanol and ethanol exerted the
lowest influence over the rBLa5#2-based ELISA. Thus, the selected
d-cELISA could be recommended for the analysis of BL in liquid
food samples with high concentrations of ethanol, such as wine
or beer, and for solid samples that had been extracted with aceto-
nitrile or ethanol. Finally, acetone extracts should be avoided.
3.3. Assay characterisation and optimization
3.3.1. Solvent tolerance
The presence of high amounts of solvents in the sample extract
may have an effect on some curve parameters of the cELISAs,
mostly over the Amax and the IC50 values. Thus, the influence of dif-
ferent percentages of ethanol, methanol, acetonitrile, and acetone
was assessed in order to evaluate the robustness of the proposed
immunoassays (Fig. 2). In general, ethanol was the best tolerated
solvent in both assays, though the presence of solvents in the

A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
283
Table 2
Recovery values obtained with the proposed cELISAs from spiked juice samples.
ELISA
Sample dilution
Boscalid (
l
g/L)
Recovery (% s, n = 3)
Grape
Peach
Apple
Tomatoa
rBLa5#1 (d-cELISA)
1/100
10
50
100
88
106
101
103
96
8
4
6
3
6
9
130 19
96
119
100
105
113 12
107
111 22
1
9
5
6
97 19
96 18
112 12
116
105 10
108
4
500
9
119
103
117
6
7
4
1000
5000
50
100
500
98 14
111 13
110
4
rBLa5#2 (i-cELISA)
1/500
112 22
106
113 21
104
5
115 27
114 23
119 18
8
123 13
96 31
117
108
5
8
1000
5000
10,000
101
93
91
2
7
6
118
92 12
92 13
7
112 13
106 13
93
116 16
107 31
4
99
9
a
A 1/500 dilution was performed in tomato samples for the rBLa5#1-based d-cELISA.
3.3.2. Buffer conditions
obtained. It could also be observed that a better precision was gen-
erally noticed with the d-cELISA than with the i-cELISA, and that
the tomato juice produced more interferences than other assayed
food matrices.
The inhibition curves obtained using buffers with varying pH
and ionic strength values were almost equivalent for the BLa5#1-
based d-cELISA (Fig. 3). In fact, there was not a significant regres-
sion (P > 0.05) between the studied factors (pH and ionic strength)
and each response value (Amax or IC50). In the case of the BLa5#2-
based indirect immunoassay, Amax values were affected by changes
in buffer ionic strength and pH values, whereas the ionic strength
had a stronger influence over the IC50 values than the pH (P < 0.05).
A decrease in the buffer ionic strength provided a continuous in-
crease in the Amax, whereas the IC50 value stayed almost constant
at ionic strengths from 150 to 300 mM and it increased quickly
at values lower than 150 mM. From this study, we could conclude
that the developed direct assay was a more robust immunoassay
than the indirect one.
4. Conclusions
The first reported functionalised derivative of the fungicide BL
has been synthesized and novel high-affinity and selective anti-
bodies against such a widely used pesticide have been produced.
Those antibodies were characterised in two cELISA formats and
two assays were selected considering their equivalent sensitivi-
ties (IC50 values were 0.9 lg/L), each of them using a different
antibody and format. The antibody-coated direct ELISA, based
on antibody rBLa5#1, was demonstrated to be faster and a more
robust assay, showing good tolerance to ethanol, methanol, and
acetonitrile, as well as to pH and ionic strength variations. The
3.4. Determination of boscalid in food samples
two optimised cELISAs displayed a limit of detection of 0.05 lg/
Standard curves of the optimised immunoassays in PBS are de-
picted in Fig. 4. The theoretical limit of detection of those immuno-
assays, estimated as the IC10 value of the inhibition curve (the
concentration of BL that provided a 10% inhibition of Amax), was
L for BL standards in buffer. This value is well below the MRLs
established for BL by the different international legislations in
most commodities, providing a comfortable dilution margin for
the analysis of residues of this fungicide at pertinent levels. Final-
ly, both of the proposed immunoassays were employed in the
determination of BL residues in foodstuffs, evaluating the matrix
effects of grape, apple, peach, and tomato juices and estimating
the BL recoveries in fortified samples, with good results. This
study provides new evidence about the expediency of the ELISA
methodology for the selective and sensitive analysis of pesticide
residues in food samples.
0.05 lg/L for both ELISAs. For the present study, different juices were
selected in order to cover a wide range of fruit categories in which
this pesticide may be present, such as: berries (grapes), stone fruits
(peaches), pome fruits (apples), and fruiting vegetables (tomatoes).
3.4.1. Matrix effects
Commercial unprocessed juices were diluted with deionised
water, and a BL standard curve was prepared with every sample
dilution and mixed with tracer or antibody solution in 2 ꢀ PBST.
Fig. S3 in the Supplementary Data shows the resultant inhibition
curves with the two developed immunoassays for the four studied
food samples. First, it was found that a 1/100 sample dilution was
enough to remove the matrix effects in the direct ELISA, except for
the tomato juice which required a 1/500 dilution. Second, analysis
of BL in fruit juice samples by the i-cELISA demanded a 1/500 dilu-
tion for any of the studied food samples.
Acknowledgements
This work was supported by the Spanish Ministerio de Ciencia e
Innovación (AGL2009-12940-C02-01-02/ALI) and cofinanced by
FEDER funds. F.A.E.-T. and J.V.M. were hired by the Consejo Superior
de Investigaciones Científicas (CSIC), the former under a JAE-doc con-
tract and the latter under a Ramón y Cajal contract, both of them
financed by Ministerio de Ciencia e Innovación and the European So-
cial Fund. We thank Laura López-Sánchez and Ana Izquierdo-Gil for
their excellent technical assistance.
3.4.2. Recovery studies
Limited amounts of the immunoreagents described in this pa-
per are available upon request for evaluation.
Several commercial grape juices from different sources were
analysed by the described cELISAs but no boscalid-containing sam-
ples were found. Thus, grape, peach, apple, and tomato juice sam-
ples were spiked with BL (analytical standard grade) and they were
homogenised by vortex mixing. Fortified samples were measured
by the developed cELISAs and BL recoveries were calculated. As
seen in Table 2, satisfactory recoveries were mostly found and
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.foodchem.2012.
04.090.
experimental limits of quantification for BL as low as 10 lg/L were

284
A. Abad-Fuentes et al. / Food Chemistry 135 (2012) 276–284
Liu, X., Dong, F., Qin, D., & Zheng, Y. (2010). Residue analysis of kresoxim-methyl in
References
fruits, vegetables, and soil using liquid–liquid extraction and gas
chromatography–mass spectrometry. Biomedical Chromatography, 24(4),
367–373.
Avenot, H. F.,
& Michailides, T. J. (2007). Resistance to boscalid fungicide in
Alternaria alternata isolates from pistachio in California. Plant Disease, 91(10),
1345–1350.
Manclús, J. J., Primo, J., & Montoya, A. (1996). Development of enzyme-linked
immunosorbent assays for the insecticide chlorpyrifos. 1. Monoclonal antibody
production and immunoassay design. Journal of Agricultural and Food Chemistry,
44(12), 4052–4062.
Bradsher, C. K., & Wissow, L. J. (1946). 2-Iodo-40-chlorobiphenyl. Journal of the
American Chemical Society, 68(3), 404–405.
Connolly, L., Fodey, T. L., Crooks, S. R. H., Delahaut, P., & Elliott, C. T. (2002). The
production and characteristics of dinitrocarbanilide antibodies raised using
antigen mimics. Journal of Immunological Methods, 264, 45–51.
Da Settimo, A., Marini, A. M., Primofiore, G., Da Settimo, F., Salerno, S., La Motta, C.,
et al. (2000). Synthesis of novel 5H,11H-pyrido[20,30:2,3]thiopyrano[4,3-b]-
indoles by Fisher-indole cyclization. Journal of Heterocyclic Chemistry, 37(2),
379–382.
Esteve-Turrillas, F. A., Parra, J., Abad-Fuentes, A., Agulló, C., Abad-Somovilla, A., &
Mercader, J. V. (2010). Hapten synthesis, monoclonal antibody generation, and
development of competitive immunoassays for the analysis of picoxystrobin in
beer. Analytica Chimica Acta, 682(1–2), 93–103.
European Commission Regulation. (2008). Commission Directive 2008/44/EC of 4
April 2008 amending Council Directive 91/414/EEC to include benthiavalicarb,
boscalid, carvone, fluoxastrobin, Paecilomyces lilacinus and prothioconazole as
active substances. Official Journal of the European Union, L 94, 13–20.
European Food Safety Authority (2010). EFSA Journal, 8(9), 1780. doi:http://
dx.doi.org/10.2903/j.efsa.2010.1780. Available online: www.efsa.europa.eu/
efsajournal.htm.
Felpin, F. X., Fouquet, E., & Zakri, C. (2009). Improved Suzuki–Miyaura reactions of
aryldiazonium salts with boronic acids by tuning palladium on charcoal catalyst
properties. Advanced Synthesis & Catalysis, 351(4), 649–655.
Fodey, T. L., Delahaut, P., & Elliott, C. T. (2007). Use of antigen mimics to produce
specific antibodies to anti-coccidial drugs. Journal of Immunological Methods,
323, 31–38.
Hiemstra, M., & de Kok, A. (2007). Comprehensive multi-residue method for the
target analysis of pesticides in crops using liquid chromatography–tandem
mass spectrometry. Journal of Chromatography A, 1154(1–2), 3–25.
Kang, S., Mou, L., Lanman, J., Velu, S., Brouillette, W. J., & Prevelige, P. E. Jr., (2009).
Synthesis of biotin-tagged chemical cross-linkers and their applications for
mass spectrometry. Rapid Communications in Mass Spectrometry, 23(11),
1719–1726.
Matsuo, R., Nishi, K., Morimune, K., Okawa, H., & Miyake, S. (2005). Monoclonal anti-
alachlor antibody for immunoassay of herbicidal alachlor analogs in
environment, food and body. Patent No. JP2005035893.
Mercader, J. V., Primo, J.,
& Montoya, A. (1995). Production of high-affinity
monoclonal antibodies for azinphos-methyl from a hapten containing only
the aromatic moiety of the pesticide. Journal of Agricultural and Food Chemistry,
43(11), 2789–2793.
Mercader, J. V., Suárez-Pantaleón, C., Agulló, C., Abad-Somovilla, A., & Abad-Fuentes,
A. (2008). Production and characterization of monoclonal antibodies specific to
the strobilurin pesticide pyraclostrobin. Journal of Agricultural and Food
Chemistry, 56(17), 7682–7690.
Mercader, J. V., Agulló, C., Abad-Somovilla, A., & Abad-Fuentes, A. (2011). Synthesis
of site-heterologous haptens for high-affinity anti-pyraclostrobin antibody
generation. Organic & Biomolecular Chemistry, 9, 1443–1453.
Ogura, H., Kobayashi, T., Shimizu, K., Kawabe, K., & Takeda, K. (1979). Novel active
ester synthesis reagent (N, N0-disuccinimidyl carbonate). Tetrahedron Letters,
20(49), 4745–4746.
Parra, J., Mercader, J. V., Agulló, C., Abad-Fuentes, A., & Abad-Somovilla, A. (2011).
Concise and modular synthesis of regioisomeric haptens for the production of
high-affinity and stereoselective antibodies to the strobilurin azoxystrobin.
Tetrahedron, 67(3), 624–635.
Rodriguez, A., Nomen, M.,
& Spur, B. W. (1998). A selective method for the
preparation of aliphatic methyl esters in the presence of aromatic carboxylic
acids. Tetrahedron Letters, 39(47), 8563–8566.
Walorczyk, S. (2008). Development of
a multi-residue method for the
determination of pesticides in cereals and dry animal feed using gas
chromatography–tandem quadrupole mass spectrometry: II. Improvement
and extension to new analytes. Journal of Chromatography A, 1208(1–2),
202–214.
Walorczyk, S., & Gnusowski, B. (2009). Development and validation of a multi-
residue method for the determination of pesticides in honeybees using
acetonitrile-based extraction and gas chromatography–tandem quadrupole
mass spectrometry. Journal of Chromatography A, 1216(37), 6522–6531.
Lehotay, S. J. (2007). Determination of pesticide residues in foods by acetonitrile
extraction and partitioning with magnesium sulfate: Collaborative study.
Journal of AOAC International, 90(2), 485–520.