Immunosensors for estradiol and ethinylestradiol based on new synthetic estrogen derivatives: Application to wastewater analysis

Article abstract of DOI:10.1021/ac303406c

Novel electrochemical immunosensors for sensitive detection of 17-β estradiol (E2) and ethinylestradiol (EE2) are described on the basis of the use of magnetic beads (MBs) as solid support and screen-printed electrodes as sensing platforms. Four synthetic estrogen derivatives containing either a carboxylic group or an amine group at the C-3 position were synthesized and covalently bound to MBs functionalized with amine or carboxyl groups, respectively. The assay was based on competition between the free and immobilized estrogen for the binding sites of the primary antibody, with subsequent revelation using alkaline phosphatase-labeled secondary antibody. Preliminary colorimetric tests were performed in order to validate the applicability of the synthetic estrogens to immuno-recognition and to optimize different experimental parameters. In a second step, electrochemical detection was carried out by square wave voltammetry (SWV). Under the optimized working conditions, the electrochemical immunosensors showed a highly sensitive response to E2 and EE2, with respective detection limits of 1 and 10 ng/L. Cross-reactivity evaluated against other hormones demonstrated an excellent selectivity. The developed devices were successfully applied to analysis of spiked and natural water samples. These new immunosensors offer the advantages of being highly sensitive, easy, and rapid to prepare, with a short assay time.

Full text of DOI:10.1021/ac303406c

Article
pubs.acs.org/ac
Immunosensors for Estradiol and Ethinylestradiol Based on New
Synthetic Estrogen Derivatives: Application to Wastewater Analysis
Hussein Kanso,† Lise Barthelmebs,† Nicolas Inguimbert,‡ and Thierry Noguer*,†
†Univ. Perpignan Via Domitia, Institut de Model
́
isation et d’Analyse en Geo
́
-Environnement et Sante,
́
EA 4218, F-66860, Perpignan,
France
‡Univ. Perpignan Via Domitia, Laboratoire de Chimie des Biomolec
́
ules et de l′Environnement, EA 4215, F-66860, Perpignan, France
S
* Supporting Information
ABSTRACT: Novel electrochemical immunosensors for sensitive
detection of 17-β estradiol (E2) and ethinylestradiol (EE2) are
described on the basis of the use of magnetic beads (MBs) as solid
support and screen-printed electrodes as sensing platforms. Four
synthetic estrogen derivatives containing either a carboxylic group or an
amine group at the C-3 position were synthesized and covalently bound
to MBs functionalized with amine or carboxyl groups, respectively. The
assay was based on competition between the free and immobilized
estrogen for the binding sites of the primary antibody, with subsequent
revelation using alkaline phosphatase-labeled secondary antibody.
Preliminary colorimetric tests were performed in order to validate the applicability of the synthetic estrogens to immuno-
recognition and to optimize different experimental parameters. In a second step, electrochemical detection was carried out by
square wave voltammetry (SWV). Under the optimized working conditions, the electrochemical immunosensors showed a highly
sensitive response to E2 and EE2, with respective detection limits of 1 and 10 ng/L. Cross-reactivity evaluated against other
hormones demonstrated an excellent selectivity. The developed devices were successfully applied to analysis of spiked and natural
water samples. These new immunosensors offer the advantages of being highly sensitive, easy, and rapid to prepare, with a short
assay time.
with mass spectrometry10 or liquid chromatography with UV11
nvironmental water quality monitoring is an area of
Esustained scientific and technological interest. In the past or mass spectrometry detection.12−14 Other detection
techniques like fluorescence polarization and quenching
resonance energy transfer technique have also been described,
involving a preliminary derivatization of estrogens using
fluorescein and lanthanide (III) chelates, respectively.15
Although these methods are highly sensitive and specific, they
are rather tedious and expensive and are not adapted to on-site
measurement. Therefore, there is still a need of highly sensitive
and consistent methods to perform rapid monitoring of real
samples. For this purpose, a lot of research has been directed
toward the development of immunoassays like enzyme-linked
immunosorbent assays (ELISA),16,17 chemiluminescence en-
zyme immunoassays,18−20 and fluorescence immunoassays.21,22
Several ELISA kits are already available on the market for
estrogen monitoring (Abraxis kits, Biosense, Cayman) and have
also been applied for research use.23 However, these assays are
time-consuming, involve multistep processes, and display quite
elevated detection limits, close to 50 ng/L.
decade, increased attention has been given to evaluate adverse
effects of emerging contaminants like endocrine-disrupting
chemicals (EDCs),1,2 which interfere with the endocrine system
of organisms by mimicking or antagonizing natural hormones.3
Several substances are classified as EDCs, such as natural
substances (phytoestrogens) and synthetic chemicals (alkyl-
phenols, pesticides, phthalates, polychlorinated biphenyls, and
bisphenol A). However, natural estrogens (17-β estradiol (E2),
estrone, and estriol) as well as the synthetic estrogen
ethinylestradiol (EE2) have been shown to display the highest
affinity to nuclear estrogen receptors (ERs) and therefore
present the greatest estrogenic potency.4,5 Although these
molecules are found in very low concentrations in water (ng/L
range), they are suspected to induce adverse effects on aquatic,
terrestrial organisms, and humans.6,7 They are potentially
involved in fish feminization, decreasing fertility of males,8 and
increased incidence of obesity and breast and testis cancers in
humans.9 The need to detect estrogenic compounds has been
discussed in numerous publications, and environmental guide-
lines setting contamination limits have been developed by
several regulatory authorities (REACH (2006), US EPA
(2006)).
As an alternative, electrochemical sensors appear as
promising tools for the monitoring of estrogenic substances
due to their numerous advantages like reduced analysis time,
Received: November 23, 2012
Accepted: January 16, 2013
Published: January 16, 2013
Conventional methods for the determination of estrogens in
aqueous matrixes are based on gas chromatography coupled
© 2013 American Chemical Society
2397
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404

Analytical Chemistry
Article
low cost, portability, and possible remote use.24 Although label-
free methods allowing direct and specific monitoring of the
interaction between antibody and antigen were recently
reported,25−27 most sensors use an indirect approach based
on competition of free antigen and enzyme-labeled antigen for
their binding to immobilized antibody. In this case, the
detection principle is based on oxidation or reduction of an
electroactive compound produced upon enzyme reaction.28−34
Besides this classical configuration, another possible approach
consists of a competition between free antigen and an
immobilized antigen derivative for free primary antibody, an
enzyme-labeled secondary antibody being used for the
revelation. This strategy has been reported to be suitable for
the detection of low molecular-weight targets, probably due to a
better accessibility of the target to antibody binding sites.35−38
Despite these advantages, the method has seldom been
described for estrogen monitoring.
This paper presents the synthesis of four new estrogenic
derivatives and their use for the development of immunoassays
based on a competitive detection method. E2 and EE2
derivatives were chemically modified and functionalized for
their selective immobilization on magnetic microbeads, that
were chosen due to their numerous advantages such as easy
fixation of the target, stability, large surface area, easy
purification, and separation.39−43 The systems were first
optimized on microtitration plates and transferred to electro-
chemical sensing technology using square wave voltammetry
(SWV) as the detection method. After testing their cross-
reactivity, the electrochemical immunosensors were applied to
the determination of E2 and EE2 in various water samples.
Thermo Fisher Scientific. Adem-Mag 96 (adapted for 96-well
microtiter plates) and Adem-MagSV (single magnet position
adapted for 1.5 mL microtubes) were from Ademtech S.A
(France). A horizontal shaker (IKA, Vibrax-VXR) was also
used. Colorimetric measurements were performed with a lab
systems Multiskan EX microtiter plate reader (Thermo Life
Sciences, France).
Immunosensors Measurements. Square wave voltammetry
(SWV) measurements were performed using an AUTOLAB
PGSTAT100 potentiostat (Eco Chemie, Netherlands). Screen-
printed electrode (SPE) systems, with graphite as working and
counter electrodes and Ag/AgCl as a reference electrode, were
fabricated using a DEK 248 screen-printing system.44 A small 4
mm-diameter magnet was fixed on the backside of the working
electrode to immobilize the MBs onto the electrode surface.
Synthesis of Modified Estrogens. Synthesis of Ethyl
Acetate Derivatives 1a and 1b. 507 mg (7.4 mmol) of sodium
ethoxide was added to a mixture of dry THF and ethanol (10:8
v/v) containing 1.68 mmol of E2 or EE2. After stirring for 30
min, 0.8 mL (4 equiv) of ethyl bromoacetate was added and the
solution was heated at 80 °C for 24 h. The reaction mixture was
then filtered, and the filtrate was concentrated under reduced
pressure. The residue was diluted in 15 mL of ethyl acetate and
evaporated under vacuum; this operation was repeated four
times. The crude compound was purified by column
chromatography with dichloromethane/ethyl acetate (8.5:1.5)
as eluent to afford 330 mg (0.92 mmol, 55%) of ethyl-2-[(17β-
hydroxy-estra-1,3,5(10)-trien-3-yl) oxy]-acetate (1a) or 350 mg
(0.91 mmol, 54%) of ethyl-2-[(17α-ethinyl-17β-hydroxy-estra-
1,3,5(10)-trien-3-yl) oxy]-acetate (1b).
Synthesis of Acidic Derivatives 2a and 2b. Sodium
hydroxide (3 mL, 2 M) was added dropwise to a 10 mL
stirred methanolic solution of 1a or 1b (0.78 mmol) at 0 °C.
The solution was stirred for 1 h at RT and monitored by thin
layer chromatography (TLC) (dichloromethane/ethyl acetate,
8.5:1.5), until full consumption of the starting material. The
reaction mixture was diluted with water (5 mL) and acidified
with KHSO4 to pH 2. The mixture was extracted with ethyl
acetate (2 × 20 mL), and the organic phase was washed with
brine, dried over anhydrous sulfate, and evaporated to dryness,
affording 220 mg (85%) of 2-[(17β-hydroxy-estra-1,3,5(10)-
trien-3-yl) oxy]-acetic acid (2a) or 250 mg (90%) of 2-[(17α-
ethinyl-17β-hydroxy-estra-1,3,5(10)-trien-3-yl) oxy]-acetic acid
(2b). Throughout this text, these two derivatives will be called
E2-COOH and EE2-COOH, respectively.
Addition of Hexanediamine Spacer. (Benzotriazol-1-
yloxy)tris(dimethylamino) phosphonium hexafluorophosphate
(BOP) (187 mg, 0.423 mmol) was added to a stirred solution
of 2a or 2b (0.282 mmol) and triethylamine (78 μL, 0.564
mmol) in 5 mL of DCM and DMF mixture (4:1 v/v) at room
temperature. After 10 min, N-Boc-1,6-hexanediamine hydro-
chloride (78.4 mg, 0.31 mmol) was added, followed by
triethylamine (78 μL, 0.564 mmol). The resulting mixture was
stirred for 4 h at room temperature. The reaction mixture was
supplemented with ethyl acetate (10 mL) and then washed
successively with HCl (1 M, 3 × 5 mL), H2O (1 × 5 mL),
NaHCO3 (1 M, 3 × 5 mL), and H2O (2 × 5 mL). The organic
phase was dried over anhydrous sulfate and concentrated to
provide the protected compounds N-Boc-(6-aminohexyl)-2-
[(17β-hydroxy-estra-1,3,5(10)-trien-3-yl)oxy]-acetamide (E2-
NH-Boc, 3a) (122 mg, 77%) and N-Boc-(6-aminohexyl)-2-
[(17α-ethinyl-17β-hydroxy-estra-1,3,5(10)-trien-3-yl)oxy]-acet-
amide (EE2-NH-Boc, 3b) (110 mg, 70%) as colorless powders.
EXPERIMENTAL SECTION
■
Reagents and Materials. Chemical Syntheses and
Characterizations. Estradiol (E2), ethinylestradiol (EE2), p-
nitrophenyl phosphate (p-NPP), diethanolamine (DEA), 1-
naphthyl phosphate (1-NP), N-hydroxysuccinimide (NHS), N-
(3-dimethylaminopropyl)-N-ethyl-carbodiimide hydrochloride
(EDC), N-Boc-1,6-hexanediamine hydrochloride, ethyl bro-
moacetate, sodium ethoxide, potassium bisulfate, acetyl
chloride, and (benzotriazol-1-yloxy)tris(dimethylamino) phos-
phonium hexafluorophosphate (BOP) were purchased from
Sigma (France). All the reactions were monitored using TLC
analysis, DC-Fertigfolien Alugram Xtra SIL G/UV254 (Macher-
ey-Nagel Germany). The chemicals and organic solvents
required for syntheses were purchased from Aldrich (France).
1
All the H and 13C NMR spectra were recorded using a JEOL
400 MHz spectrometer. Mass spectral analysis was performed
using electrospray ionization mass spectrometry (ESI-MS)
(Thermo Scientific).
Immunoassays. E2 and EE2 was first dissolved in ethanol (1
g/L) and then diluted in phosphate buffer saline (PBS 1×).
Buffer components, MES, Trizma base (tri[hydroxymethyl]-
amino-methane), Tween 20, bovine serum albumin (BSA),
casein blocking buffer, alkaline phosphatase (ALP)-labeled goat
antimouse IgG antibody, and alkaline phosphatase (ALP)-
labeled donkey antisheep IgG antibody were purchased from
Sigma (France). Monoclonal antibody (anti-E2, developed in
mouse) against E2 and polyclonal antibody (anti-EE2,
developed in sheep) against EE2 was obtained from Thermo
scientific. Magnetic beads (MBs) Dynabeads MyOne carboxylic
acid (Invitrogen, USA) and Amino-Adembeads (Ademtech,
France) were used as supports to immobilize the estrogens.
U96 PolySorp Nunc-MicroWell Plates were obtained from
2398
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404

Analytical Chemistry
Article
Deprotection of Boc Group. Deprotection of 3a derivative:
100 mg (0.189 mmol) of E2-NH-Boc was dissolved in 2 mL of
DCM and TFA (1:1 v/v). The reaction mixture was stirred at
room temperature for 1 h and monitored by TLC (ethyl
acetate/DCM, 7:3). The reaction mixture was then evaporated
under vacuum, and residual TFA was eliminated by
coevaporation with cyclohexane. The residue was triturated
with anhydrous ether (2 × 20 mL) to provide 89 mg (86%) of
N-(6-aminohexyl)-2-[(17β-hydroxy-estra-1,3,5(10)-trien-3-yl)
oxy] acetamide (E2-NH2, 4a). Deprotection of 3b derivative:
To a solution of EE2-NH-Boc (100 mg, 0.18 mmol) in ethyl
acetate (5 mL), 4 mL of 4 M solution of hydrochloride in ethyl
acetate was added dropwise. The reaction mixture was stirred at
room temperature for 60 min. The reaction mixture was
evaporated under vacuum, and the residue was dissolved in 10
mL of ethyl acetate. The solution was again evaporated, and the
residue was triturated with anhydrous ether, leading to 83 mg
(94%) of N-(6-aminohexyl)-2-[(17α-ethinyl-17β-hydroxy-
estra-1,3,5(10)-trien-3-yl) oxy] acetamide (EE2-NH2, 4b).
Immobilization of Estrogens Amine Derivatives on
Magnetic Beads. Fixation of E2-NH2 or EE2-NH2 to
Carboxyl-Functionalized Magnetic Beads. MBs coated with
carboxylic acid (carboxyl-MBs) were used as support for
immobilization of E2-NH2 or EE2-NH2. 200 μL of carboxyl-
MBs was collected and washed twice with 200 μL of binding
buffer (MES, 25 mM, pH 6). Estrogen immobilization was
carried out by mixing the MBs with 45 μL of MES and 5 μL of
estrogen solution (10 g/L) in ethanol, followed by 30 min of
gentle stirring. 30 μL of EDC (10 mg/mL) in cold MES buffer
(25 mM) was then added, and the solution was incubated on a
roller overnight at 4 °C. To eliminate unbound estrogens, the
activated MBs were washed twice with 200 μL of Tris 50 mM
containing BSA 0.5% and Tween 0.1%. The estrogen-coated
MBs (E2-hexa-MBs or EE2-hexa-MBs) were then resuspended
in the storage buffer (Tris 50 mM) and stored at 4 °C until use.
Fixation of E2-COOH or EE2-COOH to Amine-Function-
alized Magnetic Beads. 200 μL of amine-MBs (10 mg/mL)
was prepared and washed 3 times with 200 μL of MES. 160 μL
of solution of EDC (4 mg/mL) in MES and 5 μL of solution of
E2-COOH or EE2-COOH (10 g/L) in ethanol were added to
the beads suspension and incubated on a roller for 4 h at room
temperature. The supernatant was then removed, and 200 μL
of solution of BSA 0.05% in PBS (1×) was added and
incubated on a roller for 30 min at room temperature.
Unbound estrogens were eliminated by washing the activated
MBs twice with 200 μL of PBS 1× Tween 0.1%. Activated MBs
were resuspended in the PBS 1× and stored at 4 °C until use.
Colorimetric Immunoassay Protocol. Colorimetric
Assays Using EE2-MBs or EE2-Hexa-MBs. To avoid non-
specific interactions, a preliminary blocking step was performed
by introducing in each microwell 200 μL of PBS (1×)
containing 0.5% BSA. After 1 h incubation, the blocking buffer
was discarded and either 30 μL of EE2-MBs (at dilution 1/10)
or 20 μL of EE2-Hexa-MBs (at dilution 1/10) were deposited
in the wells. The solution was then removed, the magnetic
beads being maintained using the Adem-Mag 96 magnetic
support. The competition step was then performed for 50 min
using 50 μL of EE2 standard solutions at different
concentrations and 50 μL of anti-EE2 solution in PBS (1×).
It was shown that the optimum dilutions of anti-EE2 were 1/
1340 and 1/2000, using, respectively, EE2-MBs and EE2-Hexa-
MBs. 100 μL of ALP-labeled donkey antisheep secondary
antibody solution at 1/30 000 in PBS (1×) was then added in
each well. After 40 min of incubation, the bound enzyme
activity was revealed by addition of 100 μL of a 4 mg/mL p-
NPP solution in 10% DEA buffer (pH 9.5). After 30 min of
reaction, the absorbance of the yellow product was measured at
405 nm. All the experiments were carried out at room
temperature with constant shaking. Washing steps were
performed between each step using 200 μL of PBS (1×)
containing 0.05% of Tween 20.
Colorimetric Assays Using E2-MBs or E2-Hexa-MBs. In this
case, the blocking step was carried out by adding 200 μL of PBS
(1×) containing 1% casein for 1 h. 10 μL of E2-MBs at dilution
1/10 or E2-Hexa-MBs at dilution 1/50 were deposited in the
microwells. The buffer was then removed, and the competition
step was performed for 50 min using 50 μL of E2 standard
solutions at different concentrations and 50 μL of anti-E2
solution at dilution of 1/3750 in PBS (1×). Then, 100 μL of
ALP-labeled goat antimouse secondary antibody solution was
added at dilution of 1/2500 in PBS (1×) for 40 min. Washing
and revelation steps were performed as described for EE2.
Electrochemical Immunosensor Protocol. Electro-
chemical Immunosensor for EE2. 10 μL of the EE2-Hexa-
MBs at dilution 1/10 was added on the surface of working
electrode, and a blocking step was performed by adding 150 μL
of PBS (1×) containing 0.5% BSA for 1 h. The competition
step was performed using 50 μL of EE2 standard solutions at
different concentrations and 50 μL of anti-EE2 solution at
dilution 1/1340 in PBS (1×) for 30 min. Then, 100 μL of ALP-
labeled donkey antisheep secondary antibody solution was
incubated at dilution 1/30 000 in PBS (1×) for 30 min. ALP
activity was revealed by adding 100 μL of a 0.5 mg/mL 1-NP
solution in 10% DEA buffer (pH 9.5). After 2 min of reaction,
the electrochemical detection was performed by scanning the
potential between 0 and 0.5 V (frequency 8 Hz, step potential
0.01 V, modulation amplitude 0.05 V, standby potential 0 V).
The height of the resulting peak was recorded and plotted
against target concentration to give a calibration curve. Washing
steps were performed by adding 100 μL of PBS (1×) between
each step, a small magnet being used to retain the MBs on the
surface of working electrode. Assays were performed in
duplicate.
Electrochemical Immunosensor for E2. 10 μL of the E2-
Hexa-MBs at dilution of 1/50 was added on the surface of
working electrode, and a blocking step was performed by
adding 150 μL of PBS (1×) containing 1% casein for 1 h. The
competition step was performed using 50 μL of E2 standard
solutions at different concentrations and 50 μL of anti-E2
solution at dilution of 1/3750 in PBS (1×) for 30 min. Then,
100 μL of ALP-labeled goat antimouse IgG secondary antibody
solution was added at dilution of 1/2500 in PBS (1×) for 30
min. Washing and measurement steps were performed as
described for EE2.
Calibration of Immunoassays and Immunosensors.
Standard calibration curves were obtained using E2 or EE2
standard solutions prepared in PBS (1×). The matrix effects
were assayed using blank samples without the target. Recovery
was calculated by spiking the blank samples with a known
amount of target standard solution.
The absorbance values were converted into their correspond-
ing test inhibition values (A/A0) as follows: % (A/A0) = (A/A0)
× 100 where A is the absorbance or the intensity values of
competitive assay and A0 is the absorbance or the intensity
value of the noncompetitive assay. The results were calculated
2399
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404

Analytical Chemistry
Article
Figure 1. Synthesis of carboxyl (E2-COOH and EE2-COOH) and amine (E2-NH2 and EE2-NH2) estrogen derivatives.
Figure 2. Strategy for indirect competitive assay for both colorimetric and electrochemical detection based on magnetic nanoparticles.
as the relative response compared to the response in the
absence of free target, expressed in percentage.
hexanediamine present in compounds 4a and 4b. Furthermore,
all synthesized compounds were analyzed by HPLC, showing a
purity of at least 95%. Structural confirmation was assessed by
ESI+ mass spectroscopy (Supporting Information).
RESULTS AND DISCUSSION
■
Grafting of E2 or EE2 to MBs was performed either using
carboxyl or amine estrogenic derivatives. The carboxylic group
of E2-COOH or EE2-COOH was activated using EDC and
linked to an amine group of MBs, leading to, respectively, E2-
MBs or EE2-MBs. Using a comparable method, the amine
group of E2-NH2 or EE2-NH2 was conjugated to the carboxyl
group of MBs via a hexamethyl spacer arm, providing,
respectively, E2-hexa-MBs or EE2-hexa-MBs.
Colorimetric Detection of E2 or EE2 in Competitive
Assays. Colorimetric immunoassays were carried out as a rapid
method to confirm the covalent attachment of MBs to synthetic
estrogens (E2 or EE2), to assess the affinity of bound estrogens
(E2-MB or EE2-MB) to their respective antibodies, and to
optimize experimental factors. As described before, the assay
was based on the competition of free and immobilized
estrogens for their binding to antibodies used in solution. As
these primary antibodies were not labeled, a secondary anti-IgG
antibody labeled with ALP was used for performing the
colorimetric detection via hydrolysis of p-NPP into the yellow-
colored p-nitrophenol (p-NP) (Figure 2). In order to optimize
the different experimental parameters, first, assays were
performed in the absence of free estrogen. Estrogen-coated
MBs were magnetically immobilized in microtiter plate wells
Synthesis of Estrogens Amine Derivatives E2-NH2 (4a)
and EE2-NH2 (4b). As C17 seems to be the more likely binding
epitope of the antibody, the modification of the estrogen was
performed on a remote position by selective alkylation of E2 or
EE2 at position-3, leading to the title compounds 4a and 4b
(Figure 1). The 3-hydroxyl group of E2 or EE2 was first
alkylated by ethyl bromoacetate to provide ethylestradiol-3-
oxylacetate (1a) or ethylethinylestradiol-3-oxylacetate (1b) in
55% and 54% yield, respectively.45 Subsequent saponification of
1a and 1b allowed their conversion into the corresponding
acids 2a (85% yield) or 2b (90% yield). The carboxylic acid
group was then coupled to the N-Boc-1,6-hexanediamine
hydrochloride using the BOP reagent, yielding 77% and 70% of
3a and 3b, respectively. Deprotection of 3a N-Boc amine was
rapidly accomplished using TFA in methylene chloride, yielding
86% of product 4a. The deprotection of 3b was performed
using hydrochloride in ethyl acetate, yielding 94% of product
4b.
H NMR of compounds 2a and 2b are characterized by the
presence of peaks at δ= 4.5 ppm that correspond to the
methylene groups of the introduced acetyl moieties. On the
other hand, signals of amide protons at δ = 7.6−8.0 ppm
confirm the conjugation of the estrogen moiety to the
2400
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404

Analytical Chemistry
Article
Table 1. Optimal Experimental Parameters Applied in Colorimetric and Electrochemical Detection
MBs volume (μL),
primary antibody
dilution
incubation
time, min
secondary antibody
dilution
incubation
time, min
methods
MBs
dilution
blocking buffer
colorimetric
immunoassay
E2-MBs
10, 1/10
10, 1/50
1/7500
1/7500
50
1/2500
40
casein 1% 1 h
E2-hexa-
MBs
EE2-MBs
30, 1/10
20, 1/10
1/2680
1/4000
1/30000
BSA 0.5% 1 h
EE2-hexa-
MBs
electrochemical
immunosensor
E2-hexa-
MBs
10, 1/50
10, 1/10
1/7500
1/2680
30
1/2500
30
casein 1% 1 h
BSA 0.5% 1 h
EE2-hexa-
MBs
1/30000
■
●
○
Figure 3. Calibration curves of E2 (A) and EE2 (B) obtained by colorimetric assays using estrogen-hexa-MBs ( or ) and estrogen-MBs ( or
□
).
using Adem-Mag 96 magnetic support; the optimized contact
time was 5 min. The blocking step was carefully studied, since
this is a key point in immunoassays to avoid nonspecific
adsorption of antibodies. Casein blocking buffer (1%) and BSA
(0.5%) were selected as blocking solutions for E2 and EE2,
respectively. Experiments with different dilutions of primary
antibody and labeled secondary antibodies were also
performed. According to the theory, high antibody concen-
trations increased the signal intensity but decreased the
sensitivity of immunoassays and generated elevated background
signal.27 The best working dilutions obtained for each estrogen
derivative are presented in Table 1.
On the basis of the optimized parameters, calibration curves
were established of E2 and EE2 using each estrogen derivative
(Figure 3). As expected, the intensity of the response was
inversely proportional to the concentration of free estrogens
(E2 or EE2) in each assay. Due to the low standard deviation of
the calibration curve (between 1% and 4%), the LOD was
defined as the free estrogen concentration inducing a 15%
decrease of the maximum signal. The calibration curves were
fitted by the sigmoidal logistic four parameter-equations
(Origin Pro 8.6). Table 2 summarizes the main characteristics
of the colorimetric assays with the two different types of MBs
charged with each estrogen (E2 and EE2). The correlation
coefficient R, the B50, and LOD values derived from the
regression equations are summarized in Table 2. The B50 was
defined as the estrogen concentration inducing a 50% decrease
of the signal. Results showed that LOD values were 2-fold
lower using estrogen-hexa-MBs (E2-hexa-MBs or EE2-hexa-
MBs) when compared to estrogen-MBs (E2-MBs or EE2-
MBs). This difference can be explained by the presence of the
Table 2. Curve Parameters Derived from the Nonlinear
Four-Parameter Logistic Regression Fitting of the Plots
Obtained for Both Colorimetric and Electrochemical
Detections, Using the Two Types of Magnetic Beads (MBs)
LOD (ng/ B50 (ng/
methods
MBs
L)
L)
R
colorimetric immunoassay E2-MBs
EE2-MBs
80
100
36
4000
2500
1970
9640
0.999
0.99
E2-hexa-MBs
0.996
0.997
EE2-hexa-
MBs
50
electrochemical
immunosensor
E2-hexa-MBs
1
237
220
0.977
0.99
EE2-hexa-
MBs
10
hexamethyldiamine spacer arm, which allows minimizing steric
hindrance and favors accessibility of immobilized antigen to
antibody binding sites.
Electrochemical Immunosensors for E2 or EE2
Detection. On the basis of the higher sensitivity of estrogen-
hexa-MBs in colorimetric assays, this method was transferred to
immunosensor format, using screen-printed electrodes (SPEs)
as electrochemical transducers. SWV was used as detection
method, based on oxidation (+210 mV vs Ag/AgCl/Cl−) of 1-
naphthol produced upon hydrolysis of 1-naphthyl phosphate by
ALP. When compared to other electrochemical techniques like
cyclic voltammetry and differential pulse voltammetry, this
method was shown to be more rapid, efficient, and suited to the
development of electrochemical immunosensors.27 The main
parameters of E2-hexa-MBs based immunoassays were thus
transferred to immunosensors (Table 1); the incubation time of
2401
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404

Analytical Chemistry
Article
Figure 4. Calibration curves of E2 (A) and EE2 (B) obtained using electrochemical biosensor.
a
Table 3. Determination of E2 and EE2 in Spiked and Unspiked Water Samples
estradiol (ng/L)
ethinylestradiol (ng/L)
found
sample
added
found
recovery (%)
added
recovery (%)
mineral water (Evian)
0
0
1
10
20
1.3 0.6
130
98
10
20
30
10.3 2.7
103
9.8
2
21
31
3
4
105
20.5 2.5
102.5
103.3
lake water (La Raho)
0.7 0.15
0
1
10
20
1.5 0.3
80
98
10
15
30
11
3
110
10.5 0.5
14.5
33.5
4
6
96.7
111.7
22
74
3
5
7
106.5
WWTP raw water
18.8 1.2
1
20
20
102
140
115
20
20
42
116
WWTP treated water
42 4.5
15.5
41
2
65
6
127.5
a
WWTP = wastewater treatment plant.
antibodies was reduced to 30 min, leading to a total preparation
time of 120 min, compared to 280 min using classical
immunoassays.
The stability of the system was studied using electrodes
coated with estrogen-MBs and treated using the blocking
buffer. The prepared electrodes stored at 4 °C were tested after
1 week of storage with no detectable loss of activity and 80% of
activity remained after 3 weeks of storage. This good stability
was related to the covalent binding of estrogen amino derivative
on carboxylated MBs.
Competition assays were performed using free targets (E2 or
EE2). The calibration curves obtained for estrogen-hexa-MBs
are presented in Figure 4. The wider linear range was obtained
for the detection of E2, for concentrations ranging from 0.1 to
100 ng/L (Figure 4A). Due to the low experimental error (5%)
associated to electrochemical detection, LOD was calculated
considering the estrogen concentration inducing a signal
decrease of 15%. LOD values of 1 and 10 ng/L and B50 values
of 237 and 220 ng/L were obtained using E2-hexa-MBs and
EE2-hexa-MBs, respectively. The achieved LODs were in the
same order of magnitude of those obtained in other works28
and much lower than those of commercially available detection
kits, whose detection limits for estradiol were between 19 and
50 ng/L. When compared to colorimetric assays, it was shown
that electrochemical detection allowed one to increase the
sensitivity of detection for both types of modified MBs (Table
2).
Tests of Cross-Reactivity and Real Sample Analysis.
Various natural and synthetic hormones (E2, EE2, estrone,
stigmasterol, and ergosterol) were tested as potential interfering
substances for the immunosensor response toward E2 and EE2.
The cross-reactivity was tested using a maximum of 250 ng/L
of each compound, and all the tested species gave very low
percent of cross-reactivity, with values lower than 0.01%. These
results demonstrated the high selectivity of the developed
immunosensors. The developed immunosensors were then
tested for analyzing various waters, spiked with known
concentrations of estrogens. Three real samples were tested:
a mineral water (Evian), a lake water (lac de la Raho, France),
and water from a local wastewater treatment plant (Pia,
France). Each sample was filtered through a 0.45 μm
membrane and tested under the same procedure as described
before. The concentration of estrogen (E2 or EE2) was
calculated by correlation with the corresponding calibration
curve. The comparison of the results obtained using spiked and
nonspiked mineral and lake waters clearly demonstrated the
absence of matrix effect in estrogen determination (Table 3).
The values of recovery tests were acceptable and adequate.
Results were slightly different with the samples obtained from
the wastewater treatment plant; a weak matrix effect was
observed inducing a small overestimation of estrogen content
(increase of the recovery rate). As expected, raw waters
contained elevated concentrations of the natural hormone E2
and lower but significant amounts of the synthetic EE2. After
2402
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404

Analytical Chemistry
Article
(7) Ternes, T.; Kreckel, P.; Mueller, J. Sci. Total Environ. 1999, 225,
91−99.
the wastewater treatment process, the concentration of E2
decreased from 74 to 42 ng/L. This diminution may be likely
explained by the oxidation of E2 to estrone,7 even if adsorption
on clay and precipitation may also occur. The first hypothesis
was consolidated by the results observed for EE2, which is not
easily oxidized and eliminated during the wastewater treatment
process (Table 3).
(8) Flores-Valverde, A. M.; Horwood, J.; Hill, E. M. Environ. Sci.
Technol. 2010, 44, 3552−3558.
(9) Solomon, G. M.; Schettler, T. CMAJ 2000, 163, 1471−1476.
(10) Durant, A. A.; Fente, C. A.; Franco, C. M.; Vazquez, B. I.;
́
Cepeda, A. J. Agric. Food Chem. 2002, 50, 436−440.
(11) Stafiej, A.; Pyrzynska, K.; Regan, F. J. Sep. Sci. 2007, 30, 985−
991.
(12) Salvia, M.-V.; Vulliet, E.; Wiest, L.; Baudot, R.; Cren-Olive,
Chromatogr., A 2012, 1245, 122−133.
(13) Miege, C.; Gabet, V.; Coquery, M.; Karolak, S.; Jugan, M.-L.;
CONCLUSIONS
́
C. J.
■
In this paper, we have synthesized new estrogens derivatives:
carboxyl estrogens (E2-COOH or EE2-COOH) and amine
estrogens with hexamethylamine spacer arm (E2-NH2 or EE2-
NH2). These new derivatives were covalently immobilized to
MBs. Results demonstrate that chemical modification on the
C3 does not change the interaction with their respective
antibodies. A lower LOD was obtained using amine estrogens
with hexamethylamine spacer arm, probably due to the increase
of the accessibility of antibody to their target.
̀
Oziol, L.; Levi, Y.; Chevreuil, M. TrAC, Trends Anal. Chem. 2009, 28,
186−195.
(14) Mieg
Anal. Chem. 2009, 28, 237−244.
(15) Harma, H.; Sarrail, G.; Kirjavainen, J.; Martikkala, E.; Hemmila,
̀
e, C.; Bados, P.; Brosse, C.; Coquery, M. TrAC, Trends
̈
̈
̈
I.; Hanninen, P. Anal. Chem. 2010, 82, 892−897.
̈
(16) Schneider, C.; Scholer, H. F.; Schneider, R. J. Steroids 2004, 69,
̈
245−253.
(17) Hintemann, T.; Schneider, C.; Scholer, H. F.; Schneider, R. J.
The combination of MBs as immobilization platform and
SPE as transducer allowed numerous advantages over other
analytical techniques in terms of analysis time, sensitivity, and
cost effectiveness. Moreover, the use of SPE allows one to
miniaturize the system and make the device portable. These
new immunosensors were successfully applied for surface water
and wastewater analysis, demonstrating their potential
application for estrogen surveillance monitoring. In the future,
the synthesized estrogen derivatives could be easily immobi-
lized on other surfaces for the development of electrochemical
label-free immunosensors based on optical (surface plasmon
resonance), gravimetric (quartz crystal microbalance), or
electrochemical (electrochemical impedance spectroscopy)
methods.
̈
Water Res. 2006, 40, 2287−2294.
(18) Zhao, L.; Lin, J.-M.; Li, Z.; Ying, X. Anal. Chim. Acta 2006, 558,
290−295.
(19) Xin, T.-B.; Wang, X.; Jin, H.; Liang, S.-X.; Lin, J.-M.; Li, Z.-J.
Appl. Biochem. Biotechnol. 2008, 158, 582−594.
(20) Schneider, C.; Scholer, H. F.; Schneider, R. J. Anal. Chim. Acta
̈
2005, 551, 92−97.
(21) Wang, Y.-C.; Su, P.; Zhang, X.-X.; Chang, W.-B. Anal. Chem.
2001, 73, 5616−5619.
(22) Kuningas, K.; Ukonaho, T.; Pakkila, H.; Rantanen, T.;
̈
̈
Rosenberg, J.; Lovgren, T.; Soukka, T. Anal. Chem. 2006, 78, 4690−
̈
4696.
(23) Farre,
́
M.; Kuster, M.; Brix, R.; Rubio, F.; de Alda, M.-J. L.;
Barcelo,
́
D. J. Chromatogr., A 2007, 1160, 166−175.
(24) Kimmel, D. W.; LeBlanc, G.; Meschievitz, M. E.; Cliffel, D. E.
Anal. Chem. 2012, 84, 685−707.
ASSOCIATED CONTENT
* Supporting Information
Additional information as noted in the text. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
(25) Sun, N.; McMullan, M.; Papakonstantinou, P.; Gao, H.; Zhang,
X.; Mihailovic, D.; Li, M. Anal. Chem. 2008, 80, 3593−3597.
(26) Martínez, N. A.; Pereira, S. V.; Bertolino, F. A.; Schneider, R. J.;
Messina, G. A.; Raba, J. Anal. Chim. Acta 2012, 723, 27−32.
(27) Liu, X.; Duckworth, P. A.; Wong, D. K. Y. Biosens. Bioelectron.
2010, 25, 1467−1473.
(28) Liu, X.; Wong, D. K. Y. Talanta 2009, 77, 1437−1443.
(29) Pemberton, R. M.; Mottram, T. T.; Hart, J. P. J. Biochem.
Biophys. Methods 2005, 63, 201−212.
(30) Pemberton, R. M.; Hart, J. P. Methods Mol. Biol. 2009, 504, 85−
98.
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: noguer@univ-perp.fr.
■
Notes
The authors declare no competing financial interest.
(31) Volpe, G.; Fares, G.; delli Quadri, F.; Draisci, R.; Ferretti, G.;
Marchiafava, C.; Moscone, D.; Palleschi, G. Anal. Chim. Acta 2006,
572, 11−16.
ACKNOWLEDGMENTS
The authors thank M. Richard Bourdil for kindly providing raw
and treated water samples from Pia wastewater treatment plant.
■
(32) Fitzpatick, J.; Manning, B. M.; O’Kennedy, R. Food Agric.
Immunol. 2003, 15, 55−64.
(33) Butler, D.; Guilbault, G. G. Sens. Actuators, B 2006, 113, 692−
699.
(34) Martínez, N. A.; Schneider, R. J.; Messina, G. A.; Raba, J. Biosens.
Bioelectron. 2010, 25, 1376−1381.
(35) Hayat, A.; Barthelmebs, L.; Sassolas, A.; Marty, J.-L. Talanta
2011, 85, 513−518.
REFERENCES
■
(1) Segner, H.; Caroll, K.; Fenske, M.; Janssen, C.; Maack, G.;
Pascoe, D.; Schafers, C.; Vandenbergh, G.; Watts, M.; Wenzel, A.
̈
Ecotoxicol. Environ. Saf. 2003, 54, 302−314.
(2) Porte, C.; Janer, G.; Lorusso, L. C.; Ortiz-Zarragoitia, M.;
Cajaraville, M. P.; Fossi, M. C.; Canesi, L. Comp. Biochem. Physiol., Part
C: Toxicol. Pharmacol. 2006, 143, 303−315.
(36) Prieto-Simon
Bioelectron. 2008, 23, 995−1002.
́ ̀
, B.; Campas, M.; Marty, J.-L.; Noguer, T. Biosens.
(37) Radoi, A.; Dumitru, L.; Barthelmebs, L.; Marty, J.-L. Anal. Lett.
2009, 42, 1187−1202.
(3) Sumpter, J. P. Toxicol. Lett. 1998, 102−103, 337−342.
(4) Pillon, A.; Servant, N.; Vignon, F.; Balaguer, P.; Nicolas, J.-C.
Anal. Biochem. 2005, 340, 295−302.
(38) Mitchell, J. Sensors 2010, 10, 7323−7346.
(39) Shlyapnikov, Y. M.; Shlyapnikova, E. A.; Simonova, M. A.;
Shepelyakovskaya, A. O.; Brovko, F. A.; Komaleva, R. L.; Grishin, E.
V.; Morozov, V. N. Anal. Chem. 2012, 84, 5596−5603.
(40) Zacco, E.; Pividori, M. I.; Alegret, S.; Galve, R.; Marco, M.-P.
Anal. Chem. 2006, 78, 1780−1788.
(5) Balaguer, P.; Franco̧ is, F.; Comunale, F.; Fenet, H.; Boussioux, A.-
M.; Pons, M.; Nicolas, J.-C.; Casellas, C. Sci. Total Environ. 1999, 233,
47−56.
(6) Larsson, D. G.; Adolfsson-Erici, M.; Parkkonen, J.; Pettersson,
M.; Berg, A.; Olsson, P.-E.; Forlin, L. Aquat. Toxicol. 1999, 45, 91−97.
̈
2403
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404

Analytical Chemistry
Article
(41) Liao, M.-H.; Chen, D.-H. Biotechnol. Lett. 2002, 24, 1913−1917.
(42) Barthelmebs, L.; Hayat, A.; Limiadi, A. W.; Marty, J.-L.; Noguer,
T. Sens. Actuators, B 2011, 156, 932−937.
(43) Hayat, A.; Barthelmebs, L.; Marty, J.-L. Anal. Chim. Acta 2011,
690, 248−252.
(44) Galezowska, A.; Sikora, T.; Istamboulie, G.; Trojanowicz, M.;
Polec, I.; Nunes, G. S.; Noguer, T.; Marty, J.-L. Sens. Mater. 2008, 20,
299−308.
(45) Liu, J.; Zhang, X.; Zhao, M.; Peng, S. Eur. J. Med. Chem. 2009,
44, 1689−1704.
2404
dx.doi.org/10.1021/ac303406c | Anal. Chem. 2013, 85, 2397−2404