A highly sensitive immunoassay for atrazine based on covalently linking the small molecule hapten to a urea–glutaraldehyde network on a polystyrene surface

Article abstract of DOI:10.1016/j.intimp.2016.10.003

A new enzyme-linked immunosorbent assay (ELISA) for atrazine was developed based on covalent bonding of the small molecule hapten, 2-mercaptopropionic acid-4-ethylamino-6-isopropylamino-1,3,5-triazine (MPA–atrazine), to urea-glutaraldehyde (UGA)-treated microtiter plates. In this assay, the microtiter plate surface was treated with the UGA network to both introduce amino groups, which were used to cross-link with the hapten carboxylate groups, and efficiently prevent non-specific adsorption of antibodies, which successfully eliminated the time-consuming routine blocking step. Compared with HNO₃-H₂SO₄-APTES-hapten coated ELISA (modified with a HNO₃-H₂SO₄-APTES mixture and covalent-linked hapten) and conventional ELISA (coated with hapten–carrier protein conjugates), the novel ELISA format increased the sensitivity by approximately 3.5-fold and 7.5-fold, respectively, and saved 2.5?h and 34?h of coating hapten time, respectively. The method's 50% inhibition concentration for atrazine was 5.54?ng?mL^??1, and the limit of detection was 0.16?ng?mL^??1 after optimization of reaction conditions. Furthermore, the ELISA was adapted for analysis of atrazine in corn, rice, and water samples, demonstrating recoveries of 90%–108%. Thus, the assay provides a convenient alternative to conventional, laborious immunoassays for routine supervision of residue detection in food and the environment.

Full text of DOI:10.1016/j.intimp.2016.10.003

International Immunopharmacology 40 (2016) 480–486
Contents lists available at ScienceDirect
International Immunopharmacology
journal homepage: www.elsevier.com/locate/intimp
A highly sensitive immunoassay for atrazine based on covalently linking
the small molecule hapten to a urea–glutaraldehyde network on a
polystyrene surface
Na Sai, Wenjing Sun, Yuntang Wu, Zhong Sun, Guanggui Yu, Guowei Huang ⁎
Department of Nutrition and Food Hygiene, School of Public Health, Tianjin Medical University, 300070 Tianjin, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 June 2016
Received in revised form 25 September 2016
Accepted 7 October 2016
Available online xxxx
A new enzyme-linked immunosorbent assay (ELISA) for atrazine was developed based on covalent bonding of
the small molecule hapten, 2-mercaptopropionic acid-4-ethylamino-6-isopropylamino-1,3,5-triazine (MPA–at-
razine), to urea-glutaraldehyde (UGA)-treated microtiter plates. In this assay, the microtiter plate surface was
treated with the UGA network to both introduce amino groups, which were used to cross-link with the hapten
carboxylate groups, and efficiently prevent non-specific adsorption of antibodies, which successfully eliminated
the time-consuming routine blocking step. Compared with HNO
3 2 4
-H SO -APTES-hapten coated ELISA (modified
Keywords:
with a HNO -H SO -APTES mixture and covalent-linked hapten) and conventional ELISA (coated with hapten–
3
2
4
Urea-glutaraldehyde
Immunoassay
Atrazine
carrier protein conjugates), the novel ELISA format increased the sensitivity by approximately 3.5-fold and 7.5-
fold, respectively, and saved 2.5 h and 34 h of coating hapten time, respectively. The method's 50% inhibition con-
centration for atrazine was 5.54 ng mL− , and the limit of detection was 0.16 ng mL after optimization of re-
action conditions. Furthermore, the ELISA was adapted for analysis of atrazine in corn, rice, and water samples,
demonstrating recoveries of 90%–108%. Thus, the assay provides a convenient alternative to conventional, labo-
rious immunoassays for routine supervision of residue detection in food and the environment.
1
−1
Small molecule hapten
©
2016 Elsevier B.V. All rights reserved.
1
. Introduction
and polyaniline-based sensors [18]. However, these methods require
expensive instrumentation and time-consuming sample pretreatments,
resulting in complex and laborious screening procedures. As an alterna-
tive, the enzyme-linked immunosorbent assay (ELISA) has gained ac-
ceptance as a suitable technique because it is inexpensive, rapid,
sensitive, and selective [19].
In traditional ELISAs, the small molecule hapten and carrier protein
conjugates as coating antigens were fixed on the surface of microtiter
plates through hydrophobic interactions [20–22]. However, such ELISAs
exhibit several disadvantages. First, the ratio of hapten molecules to car-
rier protein is inconsistent and non-reproducible during conjugate
preparation; thus, the assay standardization and evaluation of hapten–
protein stoichiometry is unfavorable. Second, immobilization of hap-
ten–protein conjugates on the surface relies on hydrophobic interac-
tions, which often modifies the antigenicity owing to unsuitable
presentation and orientation of the hapten molecules [23].
Atrazine is one of the most prevalent herbicides for combating
weeds in corn, sugarcane, and sorghum crops due to its selectivity for
broadleaf weeds and annual grasses [1,2]. However, large-scale use of
atrazine has led to the contamination of foods and drinking water
through various pathways, causing serious health risks [3]. The United
States Environmental Protection Agency (USEPA) has reported that at-
razine overexposure may induce critical diseases, such as low blood
pressure, weight loss, muscle spasms, and adrenal gland damage [4–
5
]. Consequently, to limit human exposure, the European Food Safety
Authority (EFSA) has recommended setting the maximum residue
levels (MRLs) for all cereals at 0.05 mg kg . In accordance with the
−
1
WHO, atrazine concentrations in drinking water are limited to 2 μg
−
1
L
[6–8].
The analytical techniques used to determine atrazine include liquid
chromatography-tandem mass spectrometry [9–10], high-performance
liquid chromatography [11], ultra-fast liquid chromatography [12], mo-
lecularly imprinted photonic crystals [13], bacterial biosensors [14],
electrochemical immunosensors [15], molecular-imprinted quartz crys-
tal microbalance (QCM) sensors [16], enzyme-based biosensors [17],
The direct immobilization of hapten molecules on a polystyrene (PS)
support is a promising approach to avoid the drawbacks of traditional
ELISA. Based on this principle, Feng et al. and Kaur et al. utilized a mix-
ture of HNO
3 2 4
-H SO and 3-aminopropyltriethoxysilane (APTES) to gen-
erate amino groups on microtiter plate surface for covalent linkage of
hapten molecules, and corresponding immunoassays were successfully
developed for the detection of bisphenol and 2,4-dichlorophenoxyacetic
acid [24,25]. However, these methods are not suitable for practical
⁎
Corresponding author.
E-mail address: huangguowei@tmu.edu.cn (G. Huang).
http://dx.doi.org/10.1016/j.intimp.2016.10.003
567-5769/© 2016 Elsevier B.V. All rights reserved.
1

N. Sai et al. / International Immunopharmacology 40 (2016) 480–486
481
applications because the HNO
3
-H
2
SO
4
mixture presents a risk for oper-
China). Other chemicals were obtained from Tianjin Chemicals Inc.
(Tianjin, China). All chemicals, reagents, and solvents used in this
study were of high-purity analytical grade. Buffers were made using
Milli-Q water (Millipore, Bedford, MA, USA). By dissolving a calculated
amount of urea in a respective amount of aqueous glutaraldehyde
(GA) solution, the raw UGA mix was prepared. Mixing was kept at
room temperature until complete dissolution of urea, so the procedure
required no typical and laborious resin synthesis. The 96-well PS micro-
titer plates were obtained from Costar Inc. (Milpitas, CA, USA). A
Multiskan MK3 ELISA reader (Thermo, USA) was applied to measure
absorbance.
ators. In addition, the density of anchor groups on the treated surface is
often too high, which results in a high coupling density and decreased
sensitivity.
Herein, we described a facile and environmentally friendly process-
ing method for generating amino groups on PS microtiter wells by apply-
ing a urea-glutaraldehyde (UGA) mixture that is used as an adhesive
agent in the furniture industry. This method allowed us to covalently
link 2-mercaptopropionic acid-4-ethylamino-6-isopropylamino-1,3,5-
triazine (MPA-atrazine) to develop a highly sensitive immunoassay
(
UGA-hapten coated ELISA) for the detection of atrazine. In addition,
we developed two immunoassays for comparison, respectively
employing HNO -H SO and APTES as PS surface modifiers to covalently
link the MPA-atrazine (HNO -H SO -APTES-hapten coated ELISA) and
3
2
4
2.2. MPA-atrazine preparation
3
2
4
atrazine-ovalbumin (OVA) conjugates as the coating antigen (conven-
tional ELISA) (Fig. 1). The analytical performances of the developed as-
says in food/water matrices were further established by performing
spike and recovery studies with corn, rice, and drinking water samples.
MPA-atrazine was synthesized as described previously [26]. Atra-
zine (1.3 g) and ethanol (60 mL) were mixed in a 150-mL flask. Then,
KOH (2.4 g) and 3-mercaptopropionic acid (0.9 mL) were slowly
added, and the resulting reaction mixture was heated at reflux for 3 h.
After the reaction was completed, the solvent was removed by distilla-
tion under reduced pressure. The residue was completely dissolved in
2
. Materials and methods
NaHCO
3
solution (5%) and extracted with chloroform (10 mL) three
SO , the solvent
2
.1. Materials and instruments
times. After dehydrating the reaction mixture with Na
2
4
was concentrated under reduced pressure to yield the targeted product
(MPA-atrazine). The result was monitored by thin-layer chromatogra-
phy (TLC) and H nuclear magnetic resonance (NMR) spectroscopy.
Atrazine, simazine, melamine, chlorpyrifos, monocrotophos, and
1
parathion were obtained from Three New Science and Technology
Chemical Co., Ltd. (Shandong, China). Bovine serum albumin (BSA),
OVA, N-hydroxysuccinimide (NHS), dicyclohexylcarbodiimide (DCC),
N,N-Dimethylformamide (DMF) and APTES were purchased from
Sigma-Aldrich (St. Louis, MO, USA). The monoclonal antibody against
atrazine and goat anti-mouse IgG- horse radish peroxidase (HRP)
were obtained from Beijing LEYBOLD Cable Technology Co. Ltd. (Beijing,
2.3. Modification of microtiter plates using UGA and directly coating MPA-
atrazine
Microtiter plates were incubated with a UGA mix with the proper
ratio at 37 °C for 2 h. After washing the plates two times with deionized
Fig. 1. Reaction mechanism of three different ELISA formats: a) UGA-hapten coated ELISA. The UGA was used to treat the microtiter plate surface for 2 h to introduce amino groups that can
covalent link MPA–atrazine; b) HNO -H SO -APTES-hapten coated ELISA. The microtiter plate surface was treated with HNO -H SO mixture for 0.5 h to create nitro groups, and then was
3
2
4
3
2
4
reacted with 5% APTES for 4 h to generate amino groups that can covalently attach MPA-atrazine; c) Conventional ELISA. Atrazine-OVA conjugate was synthesized with MPA- atrazine and
OVA spending 36 h, and coated the microtiter plate surface by the physical adsorption capacity.

482
N. Sai et al. / International Immunopharmacology 40 (2016) 480–486
water, amino groups of microtiter plates were created. Activated MPA-
atrazine (100 μL, 175 nmol mL ) was added to each modified well
2.6. Competitive inhibition ELISA protocol
−
1
and incubated for 12 h at 4 °C.
The three kinds of hapten-coated plates prepared were each used in
the following steps. Optimized specific anti-atrazine antibody (50 μL)
and different concentrations of free atrazine (50 μL) were added to
each well. After incubating 1.5 h at 37 °C, the plates were washed thor-
oughly with PBST, and then 100 μL secondary antibody (goat anti-
mouse IgG-HRP) at proper dilution (1:3000) was added for 1 h at
37 °C. The tetramethylbenzidine (TMB) substrate solution (100 μL
2
.4. Modification of microtiter plates using the HNO
3 2 4
-H SO and APTES
mixture and directly coating MPA-atrazine
3 2 4
Microtiter plates were first a strong acid treated with a HNO -H SO
mixture (47:53, v/v) for 30 min at room temperature. After washing
thoroughly three times with deionized water, nitro groups were intro-
duced on the PS surface. The modified wells were then incubated with
−
1
well ) was added after a final washing step, and the enzymatic reac-
tion was incubated for 15 min at room temperature and stopped by
−
1
−1
5
% APTES aqueous solution (pH 5.5) at room temperature for 2 h and
the addition of 2 mol L
2 4
H SO (50 μL well ). The absorbance of
then reacted at 62 °C for 2 h to generate amino groups on the PS surface.
each well was measured at 450 nm using an ELISA reader.
2.7. Sensitivity and specificity
−
1
Activated MPA-atrazine (100 μL, 350 nmol mL ) was then added to
each modified well and incubated overnight at 4 °C. Blocking was
−
1
done with formaldehyde solution (100 μL, 0.2 mol L ) by incubating
the plates for 2 h at 37 °C to block the unbound sites of the PS surface.
The sensitivity of the immunoassays was evaluated using normal-
0
ized assay signals as %B/B based on the following formula:
2
.5. Atrazine-OVA conjugate preparation and coating in the microtiter
plates
B
B0
A−Aex
A0−Aex
¼
ꢀ 100
Atrazine-protein conjugates were synthesized using the DCC meth-
od. Briefly, MPA-atrazine (10 mg), NHS (6 mg), and DCC (12 mg)
were dissolved in DMF (2 mL) for 12 h at room temperature. Next, the
mixture was centrifuged and the supernatant collected. To this solution,
0
where A is the absorbance of hapten at the standard concentration, A is
the absorbance at zero hapten concentration, and Aex is the absorbance
at excessive hapten concentration.
The specificity of the immunoassay was characterized by cross-reac-
tivity (CR). A standard solution of each compound was dissolved in
methanol and diluted in PBS. Then, CR was determined for five different
compounds structurally related to atrazine by performing competitive
assays using the 50% inhibition concentration (IC50) value and calculat-
ed as
−
1
1
0 mg mL
OVA in borate buffer was added, and the mixture was
stirred at room temperature for 24 h. The atrazine-OVA conjugate was
then purified by dialysis in phosphate-buffered saline (PBS, 0.01 M,
pH 7.4). The atrazine-OVA conjugates were identified by UV–vis spec-
trophotometry. The microtiter plate wells were coated with 100 μL atra-
zine-protein conjugates in coating buffer for 12 h at 4 °C. To solve the
non-specific adsorption problem, plates were treated with blocking
buffer involving 1% BSA at 37 °C for 2 h and washed three times with
−
1
IC50 for atrazine
IC50 for analogues
PBST (0.01 mol L , pH 7.4, PBS with 0.05% tween 20) for further
reaction.
CR ð%Þ ¼
ꢀ 100
Fig. 2. The verification of MPA-atrazine: a) TLC of MPA-atrazine. Compared with atrazine original drug, there was new product (MPA-atrazine) generated; b) 1H HMR spectrum of MPA-
atrazine. The inset showed details of the fine peaks corresponding to some of MPA-atrazine.

N. Sai et al. / International Immunopharmacology 40 (2016) 480–486
483
2
.8. Sample preparation
For real samples, matrix effects are a common challenge in immuno-
assays. Two different cereal samples (corn and rice) and drinking water
were chosen to evaluate the performance of the immunoassays. Rice
and corn were purchased from the local market in Tianjin. The crushed
corn and rice samples (20 g) were respectively mixed with 20 mL meth-
anol/water (1:1, v/v) and then shaken for 30 min. After centrifuging at
4
000 ×g for 30 min, the supernatants were transferred to a 25-mL vol-
−
1
umetric flask. After adding sodium chloride (5 mol L , 5 mL), the sam-
ple was extracted with a methylene chloride and petroleum ether
mixture (3.5:6.5, v/v). The supernatant was evaporated, and the residue
−
1
was dissolved in 0.5 mL methanol and 0.01 mol L
corn and rice samples were spiked with a series of atrazine standards
5, 10, and 50 μg kg−1) for immunoassay analysis. The drinking water
was also spiked with known concentrations of atrazine (5, 10, and
PBS. Finally, the
(
−
1
5
0 μg kg ) and then analyzed by the ELISAs.
0 0
Fig. 4. Effect of the concentration of GA (from 0.1% to 2%) on A , Aex and Abg values. A
2
.9. Statistical data analysis
value is the absorbance at zero hapten concentration; Aex value means the absorbance
at excess hapten concentration; Abg value is the absorbance at negative group. Each
point represents the mean ± SD (standard deviation, n = 3).
Competitive curves of ELISAs were obtained by plotting the inhibi-
tion rate against the logarithm of atrazine concentration. Sigmoidal
curves were fitted to four-parameter logistic equation, using the Origin
pro 7.5 software package. The obtained results were expressed as
mean ± standard deviations (SD) and coefficient of variation (CV).
Analysis of ELISA data and statistical analyses were carried out in
Microsoft Excel 2007.
spectra were observed at same concentration. This confirmed that the
atrazine-OVA conjugates were successfully prepared.
3.2. Optimization of the essential steps of the assay
3.2.1. Effect of urea concentration and dilution level of GA
3
. Results and discussion
To directly immobilize MPA-atrazine, microtiter plates were
pretreated with UGA. It has been suggested that a UGA network with
amino groups was created and bound to the surface during this pre-
coating step. Thereafter, the MPA-atrazine was linked to the surface-
immobilized UGA at pH 8.0 by formation of resonance-stabilized
imino bonds. Thus, the urea/GA molar ratio was optimized to obtain a
stable network with amino groups.
3
.1. MPA-atrazine and atrazine-OVA conjugates verification
The obtained MPA-atrazine was first verified by TLC. Fig. 2a showed
that a new product was generated; thus, MPA-atrazine preparation was
initially confirmed. The obtained MPA-atrazine was further identified
1
1
from its H HMR spectra (Fig. 2b). The H NMR peaks (DMSO + D
were observed at δ = 1.168–1.099 (m, CH , 9H), 2.514–2.510 (t,
SCH CH COOH, 2H), 2.721–2.687 (t, SCH CH
NHCH CH , 2H), 4.104–4.089 (t, NHCH(CH
NHCH(CH
2
O)
0
The values of A and Aex were close to the background value (Abg) for
3
plates that were not treated with UGA (negative control), indicating
that the MPA-atrazine was not immobilized. Fig. 4 showed the effect
of different GA dilutions on the immunoassay parameters. When GA
2
2
2
2
COOH, 2H), 3.338 (m,
2
3
3
3
)
2
, 1H), 8594–8.322 (m,
)
2
, 2H), and 11.5 (s, COOH, 1H). The results indicated that
0
was diluted to less than 1%, the Abg value decreased; the A value in-
MPA-atrazine was successfully synthesized.
creased with increasing GA concentration and started to decrease. The
phenomenon was explained as follows. From 0.1% to 1% GA, the GA con-
centration increase would make the UGA network complex and react
with more urea molecules to produce more amino groups. This complex
network structure could effectively prevent non-specific adsorption of
antibodies, thus causing the Abg decrease. In particular, when using 1%
GA, the Abg value was only 0.6, which would allow elimination of the
routine blocking step. On the other hand, at GA concentrations above
Fig. 3 displayed marked differences in the UV spectra of equal con-
centrations of atrazine, OVA, and the atrazine-OVA conjugate, probably
because the different substances (atrazine, atrazine-OVA conjugate, and
OVA) have different absorbance ε. Thus, different absorbances in UV
1
%, the excess GA further increased and reacted with not only more
urea but also the amino groups on the network itself. Thus, the number
of free amino groups decreased, resulting in a decrease of the positive
value (A ). As a result, the maximum difference A − Aex and lowest
0 0
Table 1
Influence of concentration of urea (from 0.2 to 1.6 mol L−1) on assay parameters of UGA-
hapten coated ELISA (n = 3).
Concentration
Immunoassay parameters
−
1
(mol L
)
Aa
Abex
Acbg
0
Measured ± SD
Measured ± SD
Measured ± SD
0
0
0
1
.2
.4
.8
.6
1.213 ± 0.043
1.285 ± 0.028
1.322 ± 0.026
1.241 ± 0.050
0.148 ± 0.012
0.151 ± 0.009
0.153 ± 0.011
0.193 ± 0.012
0.079 ± 0.006
0.089 ± 0.011
0.083 ± 0.007
0.133 ± 0.009
a
b
c
Absorbance at zero hapten concentration (0 μg ml−1).
Absorbance at excess hapten concentration (100 μg ml−1).
Absorbance at negative group.
Fig. 3. UV spectrograms of MPA-atrazine (maximum absorption peak wavelength in
258.3 nm), atrazine-OVA conjugate (maximum absorption peak wavelength in 273 nm),
and OVA (maximum absorption peak wavelength in 279.8 nm).

484
N. Sai et al. / International Immunopharmacology 40 (2016) 480–486
3 2 4
In the general process of immunoassays, including HNO -H SO -
APTES-hapten coated ELISA and conventional ELISA, the blocking step
is necessary to prevent the nonspecific adsorption of antibodies. Thus,
in UGA-hapten coated ELISA, we tested the blocking effectiveness of dif-
ferent blocking reagents such as formaldehyde, acetaldehyde, and
ethanedial. However, the non-specific adsorption showed no obvious
differences with or without the blocking step (data not shown). This
might be because the blocking effect of the UGA network already
prevented most non-specific adsorption (Fig. 4). Thus, using UGA as a
modifier can both create amino groups on the microtiter plate surface
and save at least 2 h by eliminating the blocking process.
3
3
5
.3. Immunoassay quality assessment
.3.1. Sensitivity
The standard curves of the three ELISA formats were showed in Fig.
. The IC50 values of UGA-hapten coated ELISA, HNO -H SO -APTES-
3 2 4
hapten coated ELISA, and conventional ELISA were about 5.54 ng
−
1
−1
−1
mL , 14.62 ng mL , 28.70 ng mL , respectively. Their limits of de-
tection (LOD, calculated as IC20 value) were approximately 0.16 ng
−
1
−1
−1
mL , 0.63 ng mL , and 1.20 ng mL , respectively. Based on the
LOD, the sensitivity of UGA-hapten coated ELISA was improved 7.5-
folds approximately compared with that of conventional ELISA. This
phenomenon was explained as follows. Because of the very low molec-
ular weight and single reactive site of atrazine hapten, the reactive sites
were easily screened by OVA in the atrazine-OVA conjugate. The coating
of the atrazine-OVA conjugate on the surface of microtiter plates de-
pends on hydrophobic interactions, which might involve significant
conformational changes that allowed large OVA molecules to easily
screen atrazine molecules during the coating process. In contrast, the at-
razine was directly covalently linked with the surface of microtiter
plates in the UGA-hapten coated ELISA. Thus, this format avoided the
screening effect of the carrier protein and generated a UGA network
on the surface to increase the distance between atrazine molecules
and surface, thereby exposing more reactive sites.
Fig. 5. The indirect competitive inhibition standard curves of the UGA-hapten coated
3 2 4
ELISA, HNO -H SO -APTES-hapten coated ELISA and conventional ELISA. The LOD was
calculated as IC20 value. Their LOD were 0.16 ng mL− , 0.63 ng mL , and 1.20 ng mL
1
−1
−1
,
respectively. Each point represents the mean ± SD (n = 3).
1
A
bg value were achieved at 1% GA (Fig. 4) and 0.8 mol L− urea (Table
1), these concentrations were used for subsequent experiments.
Although they both employed hapten covalently linked to the sur-
face, the UGA-hapten coated ELISA displayed about 3.9-fold higher sen-
3
.2.2. Effect of solid immobilization conditions
Assay sensitivity depends greatly on the solid phase. An ideal solid
3 2 4
sitivity than the HNO -H SO -APTES-hapten coated ELISA. This high
phase should be consistent, stable, and have a low background signal.
The special ELISA microtiter plates (medium affinity) were selected be-
cause they can prevent non-specific adsorption.
sensitivity would be caused by a decrease in non-specific adsorption.
In the UGA-hapten coated ELISA, the UGA network can efficiently pre-
vent non-specific adsorption of antibodies. However, in HNO -H SO -
3 2 4
Table 2
CRs of the inhibition assay with selected compounds included atrazine, simazin, melamine, chlorpyrifos, monocrotophos and parathion (n = 3).
Compounds
Atrazine
Structure
IC50 (μg mL−1)
%CR
0.006
100
Simazine
0.061
30.3
19
Melamine
0.043
Chlorpyrifos
1.12 × 104
0.00012
Monocrotophos
Parathion
1.78 × 106
165.30
7.29 × 10−7
0.0079

N. Sai et al. / International Immunopharmacology 40 (2016) 480–486
485
Table 3
Accuracy and precision of atrazine in spiked samples (water, rice and corn) by ELISAs based on intra-assays.
ELISA formats
Sample (μg kg−1)
Water
Rice
5
Corn
5
5
10
50
10
50
10
50
UGA-hapten coated ELISA
Measured (μg kg−1) 4.5 ± 0.15 9.3 ± 0.25 48.5 ± 0.51 5.4 ± 0.24 9.3 ± 0.30
49.0 ± 0.99 4.6 ± 0.16 9.3 ± 0.24 50.3 ± 1.28
Recovery (%)
CV (%)
90.1
3.26
92.5
2.65
97.1
1.06
108.1
4.47
92.8
3.21
98.2
2.03
92.0
3.58
93.2
2.6
100.5
2.54
HNO
3
-H
SO
2 4
-APTES-hapten Measured (μg kg−1) 5.4 ± 0.20 8.9 ± 0.14 53.1 ± 1.93 4.3 ± 0.11 11.6 ± 0.32 49.4 ± 0.74 5 ± 0.14
9.6 ± 0.32 49.2 ± 1.22
coated ELISA
Recovery (%)
CV (%)
108.0
3.71
89.8
1.58
106.2
3.64
86.4
2.59
116.2
2.78
98.8
1.5
100.5
2.77
96.4
3.29
98.5
2.48
Conventional ELISA
Measured (μg kg−1) 4.3 ± 0.16 9.7 ± 0.31 51.6 ± 1.52 4.3 ± 0.08 9.2 ± 0.19
50.3 ± 1.59 4.6 ± 0.16 9.2 ± 0.20 51.3 ± 1.55
Recovery (%)
CV (%)
86.3
3.8
97.8
3.21
103.2
2.95
86.6
1.88
92.4
2.04
100.6
3.17
92.3
3.53
92.7
2.12
102.6
3.03
Intra-assay variation was determined by five replicates on a single day.
APTES-hapten coated ELISA, the carbon chain and benzene ring substit-
uents cannot efficiently prevent non-specific adsorption can easily cou-
ple in high density, thus yielding lower sensitivity.
UGA-hapten coated ELISA exhibited higher stability and successfully de-
tected atrazine in the real samples tested.
The higher stability of UGA-hapten coated ELISA might be explained
as follows. In conventional ELISA, the covalent bond of the atrazine-OVA
conjugate can dissociate during storage, and the ratio of atrazine and
OVA molecules is inconsistent and non-reproducible during conjugate
formation. In the HNO -H SO -APTES-hapten coated ELISA, the PS sur-
3
.3.2. Specificity and CR of the anti-atrazine antibody
CR is used as an important parameter for evaluating the specificity of
3
2
4
immunoassays. Table 2 showed the sensitivity and CR values of anti-at-
razine antibodies with atrazine and its analogues. The UGA-hapten coat-
ed ELISA showed very high sensitivity and specificity against the target
face was easily burned and made uneven by treatment with strong
acid (HNO and H SO ), and the reaction was non-reproducible. How-
ever, UGA-hapten coated ELISA eliminated the complex preparation of
atrazine-OVA conjugate and the use of strong acid, thus avoiding the
previous drawbacks.
3
2
4
−
1
molecule atrazine (IC50: 0.006 μg mL ) versus similarly structured at-
−
1
razine analogues such as simazine (IC50: 0.061 μg mL ) and melamine
−
1
(
IC50: 30.3 μg mL ) and structurally unrelated compounds such as
In summary, the operating procedures and the detection perfor-
mances of the UGA-hapten coated ELISA were superior to those of
HNO -H SO -APTES-hapten coated ELISA and conventional ELISA. In
4
−1
chlorpyrifos (IC50: 1.12 × 10 μg mL ), monocrotophos (IC50: 1.78 ×
0 μg mL ), and parathion (IC50: 165.30 μg mL ). This assay had
6
−1
−1
1
3
2
4
CR of 100% to atrazine, 19% to simazine, and less than 0.1% to the
other compounds. The results were similar to those observed with the
other two ELISA formats. Thus, the anti-atrazine antibody was very spe-
cific for atrazine.
UGA-hapten coated ELISA, the hapten-coating process consumed only
2 h, which was shorter than the HNO -H SO -APTES-hapten coated
3
2
4
ELISA (4.5 h) and the conventional ELISA (36 h). Furthermore, although
there was no blocking step, the UGA-hapten coated ELISA exhibited im-
−
1
proved sensitivity (LOD = 0.16 ng mL ) and stability compared to the
other two ELISA formats.
3
.3.3. Recovery
Corn, rice, and water samples were spiked with atrazine at 5, 10,
−
1
5
HNO
0 μg kg and then analyzed with the UGA-hapten coated ELISA,
2 4
-H SO -APTES-hapten coated ELISA and conventional ELISA.
4. Conclusion
3
Each sample was evaluated five times in duplicate and on five different
days to verify the repeatability.
We have developed a new strategy for the highly sensitive detection
of atrazine by employing small molecules hapten covalently linked to
UGA-treated microtiter plates. Compared with the other two ELISA for-
mats, the novel method was simpler and shorter features while
exhibiting high sensitivity, high stability, and high selectivity. The trans-
ferability of the findings and the parameters defined and verified for the
immunoassays to other analytes needs to be confirmed. This is the focus
of ongoing research.
As shown in Table 3 and Table 4, the intra-assay recoveries ranged
from 90.1%–108.1% with CVs of 1.06%–3.58% and the inter-assay recov-
eries were in the range of 91.5%–114.3% with CVs of 3.52%–8.34% in
UGA-hapten coated ELISA. These recoveries were consistent with
those in the other ELISA formats, but the CVs were much lower than
those of the other ELISA formats. Thus, the results indicate that the
Table 4
Accuracy and precision of atrazine in spiked samples (water, rice and corn) by ELISAs based on inter-assays.
ELISA formats
Sample (μg kg−1)
Water
Rice
5
Corn
5
5
10
50
10
50
10
50
UGA-hapten coated ELISA
Measured (μg kg−1) 4.8 ± 0.25 9.1 ± 0.32 49.2 ± 3.09 5.7 ± 0.29 9 ± 0.75
52.5 ± 2.01 5.1 ± 0.42 9.2 ± 0.41 50.5 ± 3.41
Recovery (%)
CV (%)
96.2
5.21
91.5
3.52
98.4
6.29
114.3
5.13
90.7
8.34
105.4
3.82
102.2
8.24
92.4
4.41
101.7
6.76
HNO
3
-H
SO
2 4
-APTES-hapten Measured (μg kg−1) 5.5 ± 0.27 8.8 ± 0.81 54.6 ± 6.15 4.5 ± 0.34 11.2 ± 0.55 49.7 ± 1.85 5.2 ± 0.52 9.3 ± 0.66 49.8 ± 3.15
coated ELISA
Recovery (%)
CV (%)
110.2
4.87
88.7
9.21
109.2
11.26
90.3
7.62
113.1
4.89
99.4
3.73
104.4
10.02
93.9
7.11
99.6
6.32
Conventional ELISA
Measured(μg kg−1) 4.1 ± 0.31 9.1 ± 0.60 52.8 ± 1.84 4.1 ± 0.21 9.8 ± 0.94
52.6 ± 4.53 4.9 ± 0.23 8.8 ± 0.62 52.5 ± 3.74
Recovery (%)
CV (%)
82.5
7.57
91.4
6.55
105.6
3.48
82.8
5.16
98.5
9.63
105.2
8.61
98.1
4.66
88.9
7.07
105.3
7.14
Inter-assay variation was determined by five replicates in five different days.

486
N. Sai et al. / International Immunopharmacology 40 (2016) 480–486
liquid chromatography-tandem mass spectrometry, J. Anal. Toxicol. 31 (2007)
81–186.
Acknowledgements
1
[
11] X. Li, Y. Wang, Q. Sun, B. Xu, Z. Yang, et al., Molecularly imprinted dispersive solid-
phase extraction for the determination of triazine herbicides in grape seeds by
high-performance liquid chromatography, J. Chromatogr. Sci. 54 (2016) 871–877.
12] X. Li, Y. Wang, Q. Sun, B. Xu, Y. Yu, X. Wang, Molecularly imprinted solid phase ex-
traction in a syringe filter for determination of triazine herbicides in Radix Paeoniae
Alba by ultra-fast liquid chromatography, Talanta 148 (2016) 539–547.
The authors gratefully acknowledge the financial supports by Na-
tional Natural Science Foundation of China (Grant 81302430), China
Postdoctoral Science Foundation (Grant 2014M560192), and Tianjin
City High School Science & Technology Fund Planning Project (Grant
[
2
0110143).
[13] Z. Wu, C.A. Tao, C.X. Lin, D.Z. Shen, G.T. Li, Label-free colorimetric detection of trace
atrazine in aqueous solution by using molecularly imprinted photonic polymers,
Chem. A Eur. J. 14 (2008) 11358.
References
14] A.
Hua, H. Gueuné, M. Cregut, G. Thouand, M.J. Durand, Development of a bacterial
[
bioassay for atrazine and cyanuric acid detection, Front. Microbiol. 6 (2015) 1–6.
15] X. Liu, W.J. Li, L. Li, Y. Yang, L.G. Mao, et al., A label-free electrochemical
immunosensor based on gold nanoparticles for direct detection of atrazine, Sens.
Actuators B 191 (2014) 408–414.
[
[
[
[
[
1] Z. Wu, C.A. Tao, C.X. Lin, D.Z. Shen, G.T. Li, Label-free colorimetric detection of trace
atrazine in aqueous solution by using molecularly imprinted photonic polymers,
Chem. A Eur. J. 14 (2008) 11358–11368.
2] N.D. Jablonowski, J.L. Krutz, R. Martinazzo, P. Zajkoska, G. Hamacher, et al., Transfer
of atrazine degradation capability to mineralize aged 14C-labeled atrazine residues
in soils, J. Agric. Food Chem. 61 (2013) 6161–6166.
3] Y. Braham, H. Barhoumi, A. Maaref, Urease capacitive biosensors using functional-
ized magnetic nanoparticles for atrazine pesticide detection in environmental sam-
ples, Anal. Methods 5 (2013) 4898–4904.
4] Z. Khoshood, S. Jamili, S. Khodabandeh, Histopathological effects of atrazine on gills
of caspian kutum Rutilus frisisii kutum fingerlings, Diseases of aquatic organisms dis
aquat org. 113 (2015) 227–234.
5] J. Liu, L. Tong, D. Li, W. Meng, W. Sun, et al., Comparison of two extraction methods
for the determination of 135 pesticides in Corydalis Rhizoma, Chuanxiong Rhizoma
and Angelicae sinensis radix by liquid chromatography-triple quadrupole-mass spec-
trometry. Application to the roots and rhizomes of Chinese herbal medicines, J.
Chromatogr. B Analyt. Technol. Biomed. Life Sci. 1017-1018 (2016) 233–240.
6] G. Mendaš, M. Vuletić, N. Galić, V. Drevenkar, Urinary metabolites as biomarkers of
human exposure to atrazine: atrazine mercapturate in agricultural workers, Toxicol.
Lett. 210 (2011) 174–181.
[
[16] V.K. Gupta, M.L. Yola, T. Eren, N. Atar, Selective QCM sensor based on atrazine
imprinted polymer: Its application to wastewater sample, Sens. Actuators B 218
(2015) 215–221.
[17] F.A. Mcardle, K.C. Persaud, Development of an enzyme-based biosensor for atrazine
detection, Analyst 118 (1993) 419–423.
[18] F.H. Ben, H.S. Sassi, Z. Affrezic-Renault, N. Abdelghani, Polyaniline based
immunosensor for atrazine sensing, Sensor letters. 7 (2009) 661–666.
[19] Z.W. Qie, J.L. Bai, B. Xie, L. Yuan, N. Song, et al., Sensitive detection of atrazine in tap
water using TELISA, Analyst 140 (2015) 5220–5226.
20] N.A. Byzova, A.V. Zherdev, E.A. Zvereva, B.B. Dzantiev, Immunochromatographic
assay with photometric detection for rapid determination of the herbicide atrazine
and other triazines in foodstuffs, J. AOAC Int. 93 (2010) 36–43.
[21] D.L. Lima, R.J. Schneider, V.I. Esteves, Development of an enzyme-linked immuno-
sorbent assay for atrazine monitoring in water samples, Environ. Sci. Pollut. Res.
Int. 20 (2013) 3157–3164.
[22] K.S. El-Gendy, N.M. Aly, E.M. Mosallam, A.K. Salama, Preparation of antibodies and
development of an enzyme immunoassay for determination of atrazine in environ-
mental samples, J. Environ. Sci. Health B 46 (2011) 321–327.
[
[
[
7] H. Barchanska, B. Babilas, K. Gluzicka, D. Zralek, I. Baranowska, Rapid determination
of mesotrione, atrazine and its main degradation products in selected plants by
MSPD – HPLC and indirect estimation of herbicides phytotoxicity by chlorophyll
quantification, Int. J. Environ. Anal. Chem. 94 (2014) 99–114.
[
23] O. Hofstetter, H. Hofstetter, D. Then, V. Schurig, B.S. Green, Direct binding of low mo-
lecular weight haptens to ELISA plates, J. Immunol. Methods 210 (1997) 89–92.
[24] Y. Feng, B.A. Ning, S. Pu, H.Y. Wang, C.H. Wang, et al., An immunoassay for bisphenol
A based on direct hapten conjugation to the polystyrene surface of microtiter plates,
Talanta 80 (2009) 803–808.
25] J. Kaur, R.C. Boro, N. Wangoo, K.R. Singh, C.R. Suri, Direct hapten coated immunoas-
say format for the detection of atrazine and 2,4-dichlorophenoxyacetic acid herbi-
cides, Anal. Chim. Acta 607 (2008) 92–99.
26] J. Kaur, K.V. Singh, R. Boro, K.R. Thampi, M. Raje, et al., Immunochromatographic dip-
stick assay format using gold nanoparticles labeled protein-hapten conjugate for the
detection of atrazine, Environ. Sci. Technol. 41 (2007) 5028–5036.
[
[
8] A. Ghanem, P. Bados, F. Perreau, R. Benabdallah, C. Plagellat, et al., Multiresidue anal-
ysis of atrazine, diuron and their degradation products in sewage sludge by liquid
chromatography tandem mass spectrometry, Anal. Bioanal. Chem. 391 (2008)
[
345–352.
9] P. Panuwet, J.V. Nguyen, P. Kuklenyik, S.O. Udunka, L.L. Needham, D.B. Barr, Quanti-
fication of atrazine and its metabolites in urine by on-line solid-phase extraction-
high-performance liquid chromatography-tandem mass spectrometry, Anal.
Bioanal. Chem. 391 (2008) 1931–1939.
[
[
10] J.V. Nguyen, A.O. Olsson, R. Bravo, L.L. Needham, D.B. Barr, Quantification of atrazine,
phenylurea, and sulfonylurea herbicide metabolites in urine by high-performance