Selective enzyme-linked immunosorbent assay for triclosan. Application to wastewater treatment plant effluents

Article abstract of DOI:10.1021/es702300s

A sensitive and highly selective enzyme-linked immunosorbent assay was developed for triclosan, one of the most common bactericides in personal care products, which is also considered as an emerging contaminant, given its presence in the effluents of sewage and wastewater treatment plants. From synthesized haptens, polyclonal rabbit antibodies against triclosan were raised. The best ELISA immunoassay was based on an antibody-coated format, yielding a detection limit of 0.03 μg L^-1, an I₅₀ of 3.85 μg L^-1, and a dynamic range from 0.22 to 42.16 μg L^-1, with little or no cross-reactivity (<10%) to similarly structured compounds, including its metabolite methyltriclosan (CR < 6%). The assay was applied as a screening method to quantify triclosan in surface water and in wastewaters. After C18 solid-phase extraction, nanogram per liter concentrations were determined in effluents of urban wastewater treatment plants. The satisfactory recoveries achieved as well as the agreement between immunochemical and chromatographic methods (GC-MS) indicate its potential for either screening or laboratory quantification.

Full text of DOI:10.1021/es702300s

Environ. Sci. Technol. 2008, 42, 1665–1672
Due to its low volatility (5.3 × 10-4 Pa at 20 °C), its stability
Selective Enzyme-Linked
against hydrolysis, and its relative water solubility (10 mg/L
at 20 °C), triclosan is often detected in the environment and
in raw foods (4). Triclosan has a log Kow )4.76, which indicates
a high potential for bioaccumulation. Its presence in surface
waters, wastewaters, and sediments, in concentrations rang-
ing from 2 ng L-1 in rivers to 2.3 µg L-1 in streams, has been
reported (5, 6). This can give rise to a critical environmental
impact resulting from its toxicity (7). Once in the environment,
triclosan can undergo photolysis to produce more toxic
compounds such as polychlorinated dibenzo-p-dioxins (8).
Immunosorbent Assay for Triclosan.
Application to Wastewater
Treatment Plant Effluents
E V A M . B R U N , † E M I L I O B O N E T , ‡
R O S A P U C H A D E S , * , † A N D
Á N G E L M A Q U I E I R A * , †
Departamento de Química, Instituto de Química Molecular
Aplicada, Universidad Politécnica de Valencia, Camino de
Vera s/n, 46071 Valencia, Spain, and Gamaser, S. L.,
Pedrapiquers, 4, 46014 Valencia, Spain
Recent research (9) has shown that triclosan readily reacts
with free chlorine under drinking water treatment conditions
to produce significant quantities of chloroform, considered
as a probable human carcinogen. These amounts of chlo-
roform can exceed EU Environmental Protection Agency
(EPA) recommended regulatory levels (30 µg L-1). This is
now of great concern, as water treatment plants do not
completely remove triclosansbeing present in effluents in
the range of 70-650 ng L-1 (5)sand free chlorine is
occasionally employed during wastewater treatment to
disinfect the final effluent (10).
Received September 14, 2007. Revised manuscript received
November 30, 2007. Accepted December 6, 2007.
A sensitive and highly selective enzyme-linked immunosorbent
assay was developed for triclosan, one of the most common
bactericides in personal care products, which is also considered
as an emerging contaminant, given its presence in the
effluents of sewage and wastewater treatment plants. From
synthesizedhaptens, polyclonalrabbitantibodiesagainsttriclosan
were raised. The best ELISA immunoassay was based on an
antibody-coated format, yielding a detection limit of 0.03 µg L-1,
an I50 of 3.85 µg L-1, and a dynamic range from 0.22 to 42.16
µg L-1, with little or no cross-reactivity (<10%) to similarly
structured compounds, including its metabolite methyltriclosan
(CR < 6%). The assay was applied as a screening method
to quantify triclosan in surface water and in wastewaters. After
C18 solid-phase extraction, nanogram per liter concentrations
were determined in effluents of urban wastewater treatment
plants. The satisfactory recoveries achieved as well as
the agreement between immunochemical and chromatographic
methods (GC-MS) indicate its potential for either screening
or laboratory quantification.
Australia’s industrial chemical regulatory agency declared
triclosan a priority existing chemical in 2003 (11). Besides,
the EU Water Framework Directive regards this compound
as an emerging contaminant.
Analytical methods generally applied for triclosan de-
termination in environmental samples are based on gas
chromatography–mass spectrometry (GC-MS) (12–14) and
liquid chromatography–mass spectrometry (LC-MS) (15–17).
High sensitivity and good recoveries (between 82 and 114%)
have been reported (3, 15) in wastewater and sediments,
with detection limits taken by both methods in the low
microgram per liter/(high nanogram per liter)range. How-
ever, extraction steps need to be taken, especially in the
analysis of complex matrices (18). Moreover, in the case of
gas chromatography, derivatization of triclosan improves the
sensitivity of the method, resulting in laborious multistep
procedures. Common derivatization techniques are meth-
ylation with diazomethane, acetylation, and silylation (19).
On the other hand, sensitive, high-throughput, and low-
cost analytical techniques such as immunoassay need to be
developed, especially for screening purposes. A magnetic
particle enzyme immunoassay for triclosan has been de-
veloped, showing high sensitivity (detection limit of 20 ng
L-1). The positive features of this ELISA were recently reported
(20). Nevertheless, this immunoassay provides a high cross-
reactivity with the metabolite methyltriclosan and other
related compounds such as polybrominated biphenyl ethers.
Introduction
Active ingredients in personal care products are continually
being introduced to the environment (1). Among these,
attention has focused on triclosan (5-chloro-2-(2,4-dichlo-
rophenoxy)phenol). This antimicrobial, antifungal, and
antiviral compound is commonly found in a variety of
household and industrial products (2). Triclosan is used
worldwide in an extensive field of applications, so it is difficult
to calculate the quantity utilized in products that in the end
go to wastewater, such as soup, shampoo, toothpaste,
disinfectants, household cleaning products, and the like, but
there are manufacturing facilities of triclosan, located in
different countries, whose output of pure triclosan can reach
to 1000 tons per year and plant. Thus, Singer estimated in
2002 that about 350 tons are used as antimicrobial substances
in personal care products at the European Union (EU) (3).
This study aims to develop a more selective triclosan
immunoassay as a screening procedure applied in aquatic
media. Thus, the synthesis of new haptens to elicit antibodies
against this germicide is described. Moreover, the determi-
nation of triclosan in natural and wastewater samples is
compared with a new solid-phase extraction protocol fol-
lowed by GC-MS quantification. Reported studies into
triclosan determination used solid-phase extraction (SPE)
and solid-phase microextraction (SPME) as extraction tech-
niques (14, 18). In this case, a simpler and cleaner stir bar
sorptive extraction (SBSE) mode was used for triclosan
extraction, avoiding the use of organic solvents and mini-
mizing the number of steps involved in the sample prepara-
tion previous to GS-MS analysis.
* Address correspondence to either author. Phone: +34-96-387
73 42. Fax: +34-96-387 93 49. E-mail: rpuchades@qim.upv.es (R.P.);
amaquieira@qim.upv.es (Á.M.).
† Universidad Politécnica de Valencia.
‡ Gamaser, S. L.
9
10.1021/es702300s CCC: $40.75
2008 American Chemical Society
VOL. 42, NO. 5, 2008
/
ENVIRONMENTAL SCIENCE & TECHNOLOGY 1665
Published on Web 01/31/2008

FIGURE 1. Chemical structures of triclosan and haptens.
bridge, MA) and an ELISA plate washer from Nunc MaxiSorp
(Roskilde, Denmark). Absorbances were measured in a
microtiter plate reader (Wallac, model Victor 1420 multilabel
counter, Turku, Finland).
Magnetic stir bars (20 mm length × 0.5 mm film thickness)
coated with poly(dimethylsiloxane) (PDMS), a twister de-
sorption unit (TDU) connected to a programmed-temper-
ature vaporization (PTV) injector (CIS-4), and a MPS2
autosampler were supplied by Gerstel (Mülheim, Ger-
many). The desorption-njection system was coupled to
the Agilent 6890 GC-5973 MS equipment (Agilent Technolo-
gies, Palo Alto, CA).
Immunoassay competitive curves were mathematically
analyzed by the Sigmaplot software package (Jandel Scientific,
Erkrath, Germany).
Hapten Synthesis. Eight haptens, named Trn (n ) 1–8)
were used in this study (Figure 1). Hapten Tr1 is triclosan
and haptens Tr3, Tr4, and Tr5 are commercially available
compounds. The synthetic routes for haptens Tr2, Tr6, Tr7,
and Tr8 are also illustrated in Figure 1. Synthesis procedures
and spectroscopic data of these compounds are specified in
the Supporting Information.
Experimental Section
Chemicals. Triclosan, standards used for cross-reactivity
studies, haptens Tr3 (2-(2,4-dichlorophenoxy)propionic acid),
Tr4 (4-(2,4-dichlorophenoxy)butyric acid), and Tr5 (4-chloro-
2-hydroxybenzoic acid), chemical reagents for hapten syn-
thesis and protein conjugation purposes, bovine serum
albumin (BSA), ovalbumin (OVA), complete and incomplete
Freund’s adjuvant, o-phenylenediamine (OPD), Tween 20,
horseradish peroxidase (HRP), and peroxidase labeled goat
antirabbit immunoglobulins (GAR-HRP) were purchased
from Sigma-Aldrich Química (Madrid, Spain) and Dr. Eh-
renstorfer (Augsburg, Germany). Keyhole limpet hemocyanin
(KLH) was provided by Pierce (Rockford, IL). The cartridges
used for solid-phase extraction were C18 high-performance
extraction disks, 47 mm, from Empore (Varian, Harbor City,
CA).
Instrumentation. 1H and 13C nuclear magnetic resonance
(NMR) spectra were obtained with a 300 Varian spectrometer
(300 MHz, Sunnyvale, CA).
UV–vis spectra were recorded on a Hewlett-Packard 8452
diode array spectrophotometer (Palo Alto, CA). Polystyrene
96-well microtiter plates were obtained from Costar (Cam-
9 ENVIRONMENTAL SCIENCE & TECHNOLOGY
/ VOL. 42, NO. 5, 2008
1666

Immunogen, Enzyme Tracer, and Coating Conjugate
Preparation. For immunization purposes, haptens were
covalently attached through their carboxylic acid moieties
to BSA and KLH by the active ester method (21). Additionally,
the set of haptens was covalently attached to HRP to prepare
enzyme tracers and to OVA to obtain the coating conjugates
(21).
Hapten Tr1 was also covalently linked to BSA, OVA, and
HRP through its hydroxyl group following the method
described by Beasley et al. (22).
Immunization Protocol and Screening of Sera. The
immunogens (0.20 mg in 0.5 mL of phosphate-buffered saline
(PBS)) were suspended in 0.5 mL of Freund’s adjuvant and
injected intramuscularly into two or four female New Zealand
rabbits (I-IV). After five boosts, whole blood was collected
and coagulated overnight at 4 °C. Then, serum was centri-
fuged and stored at -80 °C. Sixteen sera from immunogens
BSA-Trn (n ) 1–8) and twelve from KLH-Trn (n ) 6–8) were
obtained. Optimal concentrations of coating conjugates,
serum dilution, and enzyme tracers were selected by check-
board titration (23).
ELISA Optimization. ELISA optimization was performed
following Tijssen’s protocol (24). Borosilicate glass material
was used to minimize triclosan loss.
Effect of Organic Solvents. The influence of different
organic solvents used in extraction procedures was also
evaluated. Competitive curves were done for triclosan
containing percentages from 0 to 10% of 2-propanol and
methanol, solvents frequently used in the extraction of this
analyte from water samples.
Analysis of Water Samples. Ten bottled drinking water
samples with electrical conductivity between 25 and 874 µS
cm-1, at 20 °C, once adjusted to pH 9.6, were spiked with
triclosan at three levels: 2.5, 5, and 10 µg L-1 and analyzed
directly through the developed ELISA without previous
preconcentration or cleanup.
Further, a water composed sample prepared by mixing
equal volumes of the 10 drinking waters mentioned above
was spiked at 10, 25, 50, 75, and 100 ng L-1, preconcentrated
by solid-phase extraction using Empore disks C18 and
analyzed by ELISA and SBSE/GC-MS methods. Disks were
conditioned with methanol (2 × 10 mL) and distilled water
(2 × 10 mL) in a Büchner funnel connected to a vacuum
pump, and aliquots of 500 mL of the spiked samples (pH
adjusted to 3.5) were extracted. The elution step was
performed with 4 × 2 mL of methanol. The obtained eluates
were vacuum-concentrated and recomposed in carbonate
buffer to a final volume rendering a concentration of 2.5 µg
L-1 before ELISA analysis. Standards and samples were run
in three replicates while recording the absorbance values.
Analysis of Wastewater Samples. Ten urban wastewater
treatment plants located in the province of Valencia (Spain)
were analyzed. In all plants, the water is subjected to a primary
treatment for solid removal followed by an aerobic digestion.
A mean yield of 88% in removing the biological oxygen
demand of the influent water level is reached. In March 2007
water samples were collected from the treated effluent in
glass flasks (1 L) with Teflon caps. The samples were kept in
the dark at 4 °C.
The water samples (raw and spiked with 100 and 200 ng
L-1 triclosan) were preconcentrated by solid-phase extraction
using Empore disks C18 and analyzed by ELISA and SBSE/
GC-MS methods. Physicochemical parameters such as pH,
conductivity, and ion concentrations were determined by
the standard methods (25).
For GC-MS determinations, 100 mL water samples (pH
adjusted to 4.5 with hydrochloric acid) were extracted with
PDMS stir bars by agitation at 500 rpm during 2 h at room
temperature. Later, the stir bars were removed from samples,
dried with a paper tissue, and introduced into glass desorption
tubes. Desorption was carried out at 325 °C during 7 min
under helium (flow-rate of 75 mL min-1) in the splitless mode,
maintaining a cryofocusing temperature of 20 °C in the PTV
injector.
The analyses were conducted using a HP-5 MS column
(30 m × 0.25 mm i.d., 0.25 µm film thickness, 5:95 phenyl-
polydimethylsiloxane). The column was kept at 70 °C for 2
min. Later, a temperature ramp was applied, first until 150
°C at 25 °C min-1; when this temperature was reached a
second ramp of 200 at 3 °C min-1 was applied; the third
ramp was until 280 °C at 8 °C min-1 for 10 min. Detection
was achieved using electron impact ionization (EI+) mode
and full scan in the m/z range of 100–325. Characteristic
triclosan ions (m/z 290, 288, 218, and 146) were used to
identify triclosan, being quantified by the peak areas.
Results and Discussion
Hapten Design and Synthesis. Specific haptens for a given
compound can be synthesized, mimicking its structure,
electronic properties, and hydrophobicity. Taking this into
account, three types of haptens were used as antigens (Figure
1): Haptens type 1 link proteins directly through the hydroxyl
group (Tr1) or a spacer arm attached to the hydroxyl group
(Tr2), haptens type 2 (Tr3, Tr4, and Tr5) contain a small
fragment of the target structure,, and haptens type 3 have
a spacer arm joined to the aromatic ring (Tr6, Tr7, and Tr8).
The triclosan hydroxyl group seemed to be the most
accessible to derivatize the molecule. One possibility was to
couple triclosan (Tr1) directly to a carrier protein through
a carbamate linkage, by reaction with carbonyl diimidazole.
Another strategy was to obtain an ether, ended in a
carboxylic acid, following the Williamson ether synthesis
procedure (20). Due to phenol acidity, it is easy to generate
the corresponding alkoxide with potassium carbonate; the
subsequent reaction with methyl bromoacetate, followed by
ester hydrolysis, leads to the acid Tr2 in good yield
(Figure 1).
The use of fragmentary haptens provided the best coating
compounds in other studies (23). There are dichlorophenoxy
ethers and chlorophenols which contain triclosan aromatic
rings and a carboxylic acid function capable of protein
coupling. Thus, compounds 2-(2,4-dichlorophenoxy)propi-
onic acid, 4-(2,4-dichlorophenoxy)butyric acid, and 4-chloro-
2-hydroxybenzoic acid were used as haptens Tr3, Tr4, and
Tr5, respectively.
Finally, a new approach was based on obtaining azo
compounds, maintaining triclosan structure to a major
degree, and leaving the hydroxyl group free. These com-
pounds can be obtained by a coupling reaction between
triclosan phenoxide and a diazonium salt (Figure 1). There
are two possible positions (ortho and para to the hydroxyl
group) to introduce the spacer arm. Nevertheless, only
products with para substitution were isolated in all cases,
probably due to the steric hindrance provided by the chlorine
substituent in the meta position. Three different aryl amino
acids were to generate the diazonium salt in order to test the
influence of the spacer arm length on the immunogenicity
of the conjugate. This approach should produce an antibody
sensitive to the hydroxyl group of triclosan, which is expected
to differ from the others.
Serum Screening. All sera were titrated in homologous
and heterologous conjugate-coated format (Supporting
Information, Table S1). Sera from fragmentary haptens Tr4
and Tr5 showed high titers with homologous conjugates.
The highest titers were obtained from immunogens BSA-
Tr1 and Tr6 and haptens Tr6, Tr7, and Tr8 as coating
conjugates. All the combinations of serum/coating conjugates
that showed specific recognition were used to carry out
competitive assays. Results for the most sensitive combina-
9 1667
/ ENVIRONMENTAL SCIENCE & TECHNOLOGY
VOL. 42, NO. 5, 2008

TABLE 1. Influence of Different Variables on ELISA Perfor-
mance
phosphate buffer
a
variable
I50 (µg L-1
)
A0
Tween 20 (%)
0
0.005
0.01
0.05
0.1
5.2 ( 0.4
35.7 ( 1.2
62.3 ( 1.4
80.2 ( 1.5
135.0 ( 1.7
422.3 ( 2.3
1.15 ( 0.04
1.35 ( 0.05
1.25 ( 0.01
1.13 ( 0.08
1.10 ( 0.03
0.86 ( 0.07
0.5
pH
4
5
6
7.5
8
8.5
9
9.5
10
10.5
5.6 ( 0.3
1.4 ( 0.2
7.3 ( 0.3
5.2 ( 0.4
9.0 ( 0.5
12.6 ( 0.8
6.2 ( 0.4
5.0 ( 0.2
15.7 ( 0.9
27.1 ( 1.3
0.87 ( 0.03
0.60 ( 0.07
1.00 ( 0.01
1.15 ( 0.04
0.99 ( 0.01
1.00 ( 0.01
1.03 ( 0.01
0.99 ( 0.02
0.63 ( 0.05
0.31 ( 0.02
FIGURE 2. Calibration curve for the triclosan immunoassay.
Mean values ( standard deviation (n ) 3). Experimental
conditions: 50 mM CB, pH 9.6, and 50 min incubation time.
Table 2. As noted the interferences were negligible for most
of the tested compounds (<0.03%). Thus, the developed
immunoassay is specific for triclosan (maximum CR < 10%)
discerning this compound to the main triclosan metabolites
methyl triclosansand a battery of chlorophenols. Besides,
the structurally related herbicides nitrofen and oxyfluorfen
(the latter being a widely used phytosanitary product) did
not show cross-reactivity with triclosan. Only one of the
compounds (BDE congener 28) showed a high cross-reactivity
(39%) with triclosan. However, the low concentrations found
(26) in wastewater and sediments for this compound (<1.4
ng L-1 and <10 ng g-1, respectively) indicate that this would
not be a real interference. These results confirm that the
designed haptens are valuable for eliciting high-affinity
antibodies to triclosan, suitable for its specific detection at
low microgram per liter levels.
Organic Solvents Effect. The analysis of triclosan in waters
and sediments using chromatographic methods requires a
previous extraction with organic solvents. In sediments,
sample preparation is especially tedious, given the extraction
steps in which the use of solvents such as dichloromethane,
methanol, acetone, ethyl acetate, and acetonitrile is manda-
tory (5).
a Mean value ( standard deviation (n ) 3).
tions are given in Table S2 (Supporting Information).
Nevertheless, poor I50 values were obtained in this format.
In the antibody-coated format (see Table S3, Supporting
Information), results indicated that only conjugates HRP-
Tr6, HRP-Tr7, and HRP-Tr8 were useful as tracers. In general,
titers were low, especially with sera from fragmentary haptens,
although high titers were obtained with haptens Tr1 and Tr6
conjugated to BSA. Medium to low titers were obtained with
KLH immunogens.
As stated earlier, the combinations of serum/tracers that
showed specific recognition were used to carry out competi-
tive assays. Results for the most sensitive combinations are
given in Table S4 (Supporting Information). The pair (KLH-
Tr8-I)/HRP-Tr6 ((1:4000)/1/32 mg L-1) provided the lowest
I50 value and was selected for the setup.
ELISA Optimization. The effect of different variables on
assay performances (A0 and I50) was studied by the direct
ELISA format (Table 1). First, surfactant concentration (0,
0.005, 0.01, 0.05, 0.1, and 0.5% Tween, v/v) using 10 mM PBS
and 1 h incubation time was tested. I50 values increased with
Tween 20 concentration, producing a negative influence on
the competition. Consequently, Tween 20 was not added to
the buffer.
Since the use of organic solvents is inevitable when
employing reference methods, it is desirable to assess the
effect of these media on ELISA performance. Different
percentages of 2-propanol and methanol in the assay buffer
were studied.
Next, the effect of pH, between 4.0 and 10.5, was studied.
Good I50 values were obtained at pH 4 and 5, but A0 values
were low at these pH levels. The best combinations I50/signal
were obtained at pH values of 7.5 and 9.6. A better I50 value
and curve fitting was obtained in carbonate buffer at pH 9.6,
but the signal was poor (A0 ) 0.71). To improve this, the
enzymatic reaction time was adjusted. A reaction time of 20
min provided the maximum A0 (1.30) with good I50 (4.2).
The influence of incubation time on the competition step,
from 10, 20, 30, 40, 50, and 60 min, was examined. A period
of 50 min was selected in order to obtain a sensitive assay
with an appropriate signal in the minimum time.
Methanol was tolerated up to 6% with I50 at the microgram
per liter level (12.42 µg L-1) and acceptable signal (A0 ) 0.61),
while 2-propanol was tolerated up to 4% (I50 ) 6.15 µg L-1
;
A0 ) 0.61). Even under high solvent percentages (10%) the
I50 values are good, maintaining the sensitivity at the
microgram per liter level. These are interesting results
considering that when SPE is applied, analyte elution with
an organic solvent is necessary.
GC–MS Optimization. The effects of different variables
on the extraction of triclosan in water with PDMS sorptive
bar, such as stirring speed, extraction time, pH, and salt effect
(sodium chloride addition), were studied. Five stirring speeds
were tested (300, 400, 500, 800, and 1000 rpm), concluding
that a high stirring speed does not mean a higher triclosan
recovery. The yield of extraction increased with longer
extraction times (15 min, 30 min, 45 min, 1 h, 1.5 h, 2 h, 3 h,
and 4 h were studied), but no significant effect was observed
for periods longer than 2 h. On the other hand, sample pH
significantly affected the efficiency of the extraction. Different
pH values were applied (between pH 3 and 8), obtaining the
best results at pH 4.5, with the compound in its neutral form.
Under the optimized conditions, a competition curve was
establihed (Figure 2). The assay exhibited I50 ) 3.85 µg L-1
,
LOD ) 0.03 µg L-1, and a working range between 0.22 and
42.16 µg L-1
.
Cross-Reactivity Studies. Assay selectivity was evaluated
using a set of chlorophenols, polybrominated biphenyl ethers
(BDEs), the metabolite methyltriclosan, nitrofen, and oxy-
fluorfen, as their structures are quite similar to that of
triclosan. I50 and CR data for each compound are given in
9 ENVIRONMENTAL SCIENCE & TECHNOLOGY
/ VOL. 42, NO. 5, 2008
1668

TABLE 2. Cross-Reactivity of Triclosan-Related Compounds and Metabolites
9 1669
/ ENVIRONMENTAL SCIENCE & TECHNOLOGY
VOL. 42, NO. 5, 2008

TABLE 3. Direct Triclosan Determination by ELISA in Spiked Mineral Water Samples at Microgram per Liter Levelsa
sample
origin and water composition (mg/L)
[triclosan] added
[triclosan] found
recovery (%)
1
Segovia (Spain)
EC, 25 µS cm-1; pH 6.5; dry residual, 26 mg/L;
[Cl-], 0.7; [HCO3-], 18.0; [Ca2+], 4.0; [Mg2+], 2.0;
[Na+], 1.0; [NO3-], 2.8
2.5
5
10
2.5 ( 0.3
4.1 ( 1.6
8.3 ( 1.0
100
83
83
2
3
4
5
6
7
8
9
Beteta, Cuenca (Spain)
EC 394 µS cm-1; pH 7.4; dry residual 261 mg/L;
2.5
5
10
2.5 ( 0.2
5.0 ( 0.4
10.3 ( 0.4
100
100
103
[Cl-] 7.9, [HCO3-] 286.0, [Ca2+] 58.3, [Mg2+
25.1, [Na+] 5.2, [NO3-] 2.1.
]
Huerta del Marquesado, Cuenca (Spain)
EC, 350 µS cm-1; pH 7.6; dry residual, 240 mg/L;
[Cl-], 2.1; [HCO3-], 252.4; [Ca2+], 62.1; [Mg2+],
18.2; [Na+], 0.7; [NO3-], 2.0; [SO42-], 21.3
2.5
5
10
2.2 ( 0.3
4.1 ( 1.1
9.9 ( 0.9
90
89
99
Pyrenees (France)
EC, 50 µS/cm; pH 6.8; dry residual, 28 mg/L; [Cl-],
0.6; [HCO3-], 5.2; [Ca2+], 3.0; [Mg2+], 0.6;
[Na+], 1.5; [NO3-], <1; [SO42-], 8.7
2.5
5
10
2.2 ( 0.3
4.7 ( 0.3
9.9 ( 1.0
90
94
99
Granada (Spain)
EC, 204 µS cm-1; pH 6.5; dry residual, 131 mg/L;
[Cl-], 2.8; [HCO3-], 105.0; [Ca2+], 27.2; [Mg2+],
8.8; [Na+], 4.8; [NO3-], 5.8; [SO42-], 17.3
2.5
5
10
2.5 ( 0.4
4.5 ( 0.5
10.5 ( 1.4
100
90
104
Extremadura (Spain)
EC, 238 µS cm-1; pH 7.3; dry residual, 58 mg/L;
2.5
5
10
2.6 ( 0.3
5.7 ( 0.2
10.2 ( 1.5
103
113
102
-
[HCO3 ], 18.0; [Ca2+], 3.0; [Mg2+], 3.0; [Na+], 5.0;
[F-], 0.1
Sant Hilari Sacalm, Gerona (Spain)
EC, 237 µS cm-1; pH 7.6; dry residual, 185 mg/L;
[Cl-], 16.2; [HCO3-], 149.0; [Ca2+], 38.5; [Mg2+], 9.7;
[Na+], 13.2; [SO42-], 16.4
2.5
5
10
2.5 ( 0.3
4.5 ( 0.3
9.3 ( 0.6
101
90
93
Valencia (Spain)
EC, 224 µS cm-1; pH 7.4; dry residual, 385 mg/L;
[Cl-], 35.8; [HCO3-], 297.2; [Ca2+], 88.7; [Mg2+],
23.4; [Na+], 18.6; [SO42-], 43.9
2.5
5
10
2.6 ( 0.6
5.1 ( 0.9
10.8 ( 1.4
104
103
108
Amer, Gerona (Spain)
EC, 198 µS cm-1; pH 7.0; dry residual, 140 mg/L;
[Cl-], 6.2; [HCO3-], 127.0; [Ca2+], 32.0; [Mg2+],
7.3; [Na+], 7.4; [NO3-], 3.0; [SO42-], 8.9
2.5
5
10
2.7 ( 0.6
5.3 ( 0.9
10.3 ( 1.2
107
105
103
10
French Alps (France)
EC, 397 µS cm-1; pH 7.2; [Cl-], 2.2; [HCO3-], 357;
[Ca2+], 78.0; [Mg2+], 23.0; [Na+], 5.5; [NO3-], 3.8;
[SO42-], 10.0
2.5
5
10
2.5 ( 0.3
5.0 ( 0.9
7.5 ( 0.9
100
99
75
mean
97.7
a Concentration in micrograms per milliliter. Values expressed as mean ( standard deviation (n ) 3).
Finally, the addition of sodium chloride (0, 0.5, and 1 g/L)
produced a decrease in the extraction yield. The optimized
conditions found were stirring at 500 rpm for 2 h, with a pH
of 4.5, and without adding salt to the samples.
Analysis of Water Samples. Bottled drinking water
samples (triclosan < LOD) were fortified with triclosan within
the dynamic range of the assay and analyzed by ELISA as
blind samples. A mean recovery of 97.7% was obtained
(Table 3).
To assess the potential of the developed immunoassay,
10 wastewater urban treatment plants were sampled, ana-
lyzing the effluents in parallel for triclosan residues by
immunochemical and GC–MS methods. The physicochem-
ical parameters of the waters used are shown in Supporting
Information (Table S5). As specified in Table 4, triclosan was
present in all of the analyzed effluents, demonstrating that
the water treatment is not effective enough to remove this
bactericide.
Further, the composed sample (mix of equal volumes of
drinking water samples), spiked at five different levels at
nanograms per liter, were extracted and quantified by ELISA
and GC–MS techniques as blind samples in two different
laboratoriesbytrainedpersonnel. Goodrecoveries(meanvalue,
94.3%) were obtained by ELISA; also the results correlated well
with GC–MS values (r ) 0.92), indicating that the developed
assay is promising as a screening procedure for triclosan
residues, as well as a sensitive quantitative method.
Better sensitivity was reached using ELISA rather than
chromatography, as the achieved immunoassay can detect
triclosan concentrations lower than 30 ng L-1, but in all cases
chromatographic results are higher than immunoanalytical
results. However, there is agreement between methods,
except for samples 8 and 10 which showed much higher
triclosan concentrations by chromatographic analysis. In a
parallel study, the same effluent water samples were spiked
with 100 and 200 ng L-1 of triclosan and quantified as blind
9 ENVIRONMENTAL SCIENCE & TECHNOLOGY
/ VOL. 42, NO. 5, 2008
1670

ACS Symposium Series 791; American Chemical Society:
Washington, D.C., 2001; pp 2–38.
(2) Bhargava, H. N.; Leonard, P. A. Triclosan: Applications and safety.
Am. J. Infect. Control 1996, 24, 209–218.
TABLE 4. Comparison of Triclosan Determination by ELISA and
GC-MS Methods in Wastewater Treatment Effluentsa
(3) Singer, H.; Müller, S.; Tixier, C.; Pillonel, L. Triclosan: Occurrence
and fate of a widely used biocide in the aquatic environment:
Field measurements in wastewater treatment plants, surface
waters, and lake sediments. Environ. Sci. Technol. 2002, 36,
4998–5004.
(4) Sanches-Silva, A.; Sendón-García, R.; López-Hernández, J.;
Paseiro-Losada, P. Determination of triclosan in foodstuffs. J.
Sep. Sci. 2005, 28, 65–72.
(5) Lindström, A.; Buerge, I. J.; Poiger, T.; Bergqvist, P.-A.; Müller,
M. D.; Buser, H. R. Occurrence and environmental behavior of
the bactericide triclosan and its methyl derivative in surface
waters and in wastewater. Environ. Sci. Technol. 2002, 36, 2322–
2329.
(6) Kolpin, D. W.; Furlong, E. T; Meyer, M. T.; Thurman, E. M.;
Zaugg, S. D.; Barber, L. B.; Buxton, H. T. Pharmaceuticals,
hormones, and other organic wastewater contaminants in U.S.
Streams, 1999–2000: A national reconnaissance. Environ. Sci.
Technol. 2002, 36, 1202–1211.
(7) Orvos, D. R.; Versteeg, D. J.; Inauen, J.; Capdevielle, M.;
Rothenstein, A.; Cunningham, V. Aquatic toxicity of triclosan.
Environ. Toxicol. Chem. 2002, 21, 1338–1349.
(8) Latch, D. E.; Packer, J. L.; Arnold, W. A.; McNeill, K. Photo-
chemical conversion of triclosan to 2,8-dichlorodibenzo-p-
dioxin in aqueous solution. J. Photochem. Photobiol., A 2003,
158, 63–66.
(9) Rule, K. L.; Ebbett, V. R.; Vikesland, P. J. Formation of chloroform
and chlorinated organics by free-chlorine-mediated oxidation
of triclosan. Environ. Sci. Technol. 2005, 39, 3176–3185.
(10) Veslind, P. A., Ed. Wastewater Treatment Plant Design; Water
Environmental Federation, IWA: London, 2003.
(11) Commonwealth of Australia Gazette, No. C 5; 2003; pp 43–46.
(12) Morales, S.; Canosa, P.; Rodríguez, I.; Rubí, E.; Cela, R. Microwave
assisted extraction followed by gas chromatography with tandem
mass spectrometry for the determination of triclosan and two
related chlorophenols in sludge and sediments. J. Chromatogr.
A 2005, 1082, 128–135.
(13) Adolfsson-Erici, M.; Petterson, M.; Parkkonen, J.; Sturve, J.
Triclosan, a commonly used bactericide found in human milk
and in the aquatic environment in Sweden. Chemosphere 2002,
46, 1485–1489.
(14) Lee, H.-B.; Peart, T. E.; Svoboda, M. L. Determination of
endocrine-disrupting phenols, acidic pharmaceuticals, and
personal-care products in sewage by solid-phase extraction and
gas chromatography-mass spectrometry. J. Chromatogr. A 2005,
1094, 122–129.
(15) Quintana, J. B.; Reemtsma, T. Sensitive determination of acidic
drugs and triclosan in surface and wastewater by ion-pair
reverse-phase liquid chromatography/tandem mass spectrom-
etry. Rapid Commun. Mass Spectrom. 2004, 18, 765–774.
(16) Thompson, A.; Griffin, P.; Stuetz, R.; Cartmell, E. The fate and
removal of triclosan during wastewater treatment. Water
Environ. Res. 2005, 77, 63–67.
(17) Hua, W.; Bennet, E. R.; Letcher, R. J. Triclosan in waste and
surface waters from the upper Detroit River by liquid chro-
matography-electrospray-tandem quadrupole mass spectrom-
etry. Environ. Int. 2005, 31, 621–630.
wastewater
plant no.
ELISA [triclosan]
GC–MS [triclosan]
( SD (ng L-1
)
( SD (ng L-1
)
1
2
3
4
5
6
7
8
9
18.8 ( 4.1
<30
317.3 ( 13.9
<30
<30
81.1 ( 5.4
<30
310.9 ( 29.7
12.5 ( 2.4
12.8 ( 4.6
60.2 ( 7.2
20.1 ( 4.2
70.8 ( 13.2
94.9 ( 4.0
32.7 ( 4.4
210.1 ( 17.2
96.6 ( 0.4
607.6 ( 49.4
<30
10
584.2 ( 79.0
a Values expressed as mean ( standard deviation (n ) 3).
samples. Mean recovery values of 91% for the immunoassay
and 103.8% for GC–MS have been obtained. Recovery results
confirm the potential of the developed assay for the analysis
of environmental waters.
Taking into consideration the high selectivity and the good
recovery percentages obtained by both methods, probably
the differences between results are due to, in the ELISA,
samples are filtered previous to the analysis and the
chromatographic method runs raw samples. As it has been
described (20), filtered and nonfiltered wastewater samples
display different results; up to 63% triclosan may be in the
suspended matter. Filtering greatly reduced the amount of
triclosan in the samples, due to removal of the bounded
triclosan. So as stir bars in our extraction procedure are in
contact with the raw sample, extraction yield increases and
the chromatographic method gives higher values than the
immunoanalytical procedure.
Recently, triclosan has been listed as an emerging
pollutant by the EU Water Framework Directive (27), whose
mandatory goal is to develop new and easy analytical
methodologies. To this end, immunoassay offers numerous
advantages with this aim: high throughput, applicability at
the point of need, huge information volume with in situ, and
quick responses. Moreover, the methodology is simple and
direct and does not require sophisticated equipment. In this
study it has been demonstrated that immunoassay for
triclosan results in a high selectivity and sensitivity, com-
parable to that obtained by SBSE/GC–MS. Finally, the low
cost of this assay compared with that of chromatographic
analysis as well as its speed makes this technique appropriate
for the determination of triclosan in environmental samples.
Acknowledgments
The authors thank Agustín Pastor for generously providing
polybrominated biphenyl ethers as well as Gamaser, S.L. for
helping to perform this study and to Debra Westall for revising
the manuscript.
(18) Canosa, P.; Rodriguez, I.; Rubí, E.; Cela, R. Optimization of solid-
phase microextraction conditions for the determination of
triclosan and possible related compounds in water samples.
J. Chromatogr. A 2005, 1072, 107–115.
(19) McAvoy, D. C.; Schatowitz, B.; Jacob, M.; Hauk, A.; Eckhoff,
W. S. Measurement of triclosan in wastewater treatment systems.
Environ. Toxicol. Chem. 2002, 21, 1323–1329.
(20) Shelver, W. L.; Kamp, L. M.; Church, J. L.; Rubio, F. M.
Measurement of triclosan in water using a magnetic particle
enzyme immunoassay. J. Agric. Food Chem. 2007, 55, 3758–
3763.
(21) Langone, J. J.; van Vunakis, H. Radioimmunoassay of nicotine,
cotidine, and γ-(3-pyridyl)-γ-oxo-N-methylbutyramide. Meth-
ods Enzymol. 1982, 84, 628–640.
(22) Beasley, H. L.; McAdam, D. P.; Larkin, K. A.; Ferguson, B. S.;
Bushway, R. J.; Skerritt, J. H. Laboratory and field enzyme-
immunoassays for diazinon and their application to residue
analysis in lanolin, water and fruit juice. Bull. Environ. Contam.
Toxicol. 1997, 59, 375–382.
Supporting Information Available
Hapten synthesis and characterization, ELISA antibody-
coated format procedure, sera titration results in direct and
indirect formats (Table S1 and Table S3), I50 values for the
best serum-coating conjugate and serum-tracer combina-
tions (Table S2 and Table S4), and the physicochemical
parameters of the studied wastewater samples (Table S5).
This material is available free of charge via the Internet at
http://pubs.acs.org.
Literature Cited
(1) Daughton, C. G. In Pharmaceuticals and Personal Care Products
in the Environment; Daughton, C. G., Jones-Lepp, T. L., Eds.;
(23) Brun, E. M.; Garcés-García, M.; Puchades, R.; Maquieira, A.
Enzyme-linked immunosorbent assay for the organophosphorus
9 1671
/ ENVIRONMENTAL SCIENCE & TECHNOLOGY
VOL. 42, NO. 5, 2008

insecticide fenthion. Influence of hapten structure. J. Immunol.
Methods 2004, 295, 21–35.
(26) Song, M.; Chu, S.; Letcher, R. J.; Seth, R. Fate, partitioning and
mass loading of polybrominated diphenyl ethers (PBDEs) during
the treatment processing of municipal sewage. Environ. Sci.
Technol. 2006, 40, 6241–6246.
(27) Roig, B., Allan, I. J., Greenwood, R., Eds. SWIFT-WFT. Report: A
Tool Box of Existing and Emerging Methods for Water Monito-
ring under the WFD, 2005. http://www.ecologic-events.de/swift-
germany/documents/Emerging_tools_manual.pdf.
(24) Tijssen, P. Practice and Theory of Enzyme Immunoassays. In
Laboratory Techniques in Biochemistry and Molecular Biology;
Burdon, R. H., Knippenberg, P. H., Eds.; Elsevier: Amsterdam,
1987; Vol. 15.
(25) Clesceri, L. S., Greenberg, A. E., Trussell, R. R., Eds. Standard
Methods for the Examination of Water and Wastewater, 17th
ed.; American Public Health Association: Washington, D.C.,
1989.
ES702300S
9 ENVIRONMENTAL SCIENCE & TECHNOLOGY
/ VOL. 42, NO. 5, 2008
1672