Development and evaluation of an immunoassay for biological monitoring chlorophenols in urine as potential indicators of occupational exposure

Article abstract of DOI:10.1021/ac010419n

Trichlorophenols (TCP) eliminated by the urine can be considered as potential biomarkers of exposure of many chemicals (chlorophenols, chlorophenoxy acid herbicides, prochloraz, lindane, hexachlorobenzene, etc). Highthroughput screening methods are necessary to carry out efficient monitoring programs that may help to prevent certain occupational health diseases. For this purpose, an indirect enzyme-linked immunosorbent assay (ELISA) for 2,4,6-trichlorophenol detection has been developed using polyclonal antisera raised against 3-(3-hydroxy-2,4,6-trichlorophenyl)propanoic acid (hapten 5) covalently coupled by the mixed anhydride (MA) method to keyhole limpet hemocyanin (KLH). The indirect ELISA uses a heterologous coating antigen prepared by conjugation of 3-(2-hydroxy-3,6-dichlorophenyl)propanoic acid (hapten 4) to bovine serum albumin (BSA) using the active ester (AE) method. The optimum hapten density for the coating antigen was found to be 3 mol of hapten/mol of protein. The assay shows a limit of detection of 0.245 ± 0.116 μg L^-1, and it is performed on 96-well microtiter plates in about 1.5 h. The ELISA reported recognizes on a much less extent other chlorinated phenols, such as 2,3,4,6-tetrachlorophenol (2,3,4,6-TtCP, 21%), 2,4,5-TCP (12%) and 2,3,5-TCP (15%); however, brominated phenols (BP) are even more recognized than the corresponding chlorinated analogues (ex. 2,4,6-TBP, 710%; 2,4-DBP, 119%). With the aim of finding an explanation for this behavior, theoretical calculations have been performed for those and other halogenated phenols (2,4,6-triiodophenol and 2,4,6-trifluorophenol) to clarify which physicochemical parameter can explain better the recognition pattern observed. Finally, the assay has been adapted to the analysis of urine samples. The studies have shown that a limit of detection of 1 μg L^-1 can be accomplished on this biological matrix by combining the ELISA procedure with a C18 solid-phase extraction method.

Full text of DOI:10.1021/ac010419n

Anal. Chem. 2002, 74, 468-478
Development and Evaluation of an Immunoassay
for Biological Monitoring Chlorophenols in Urine as
Potential Indicators of Occupational Exposure
Roger Galve, Mikaela Nichkova, Francisco Camps, Francisco Sanchez-Baeza, and M.-Pilar Marco*
Department of Biological Organic Chemistry. IIQAB-CSIC. Jorge Girona, 18-26, 08034 Barcelona, Spain
compartments. They have been used as antistain agents for wood
products since the early 1930s, and they are also important
synthetic intermediates of many chemicals such as chlorophenoxy
herbicides (i.e., 2,4,5-T), dyes,1 fungicides (i.e., prochloraz), or
bleaching agents (i.e., chloranil). The importance of this industry
is evidenced by the great amount of chlorophenoxy herbicides
and chlorophenols produced.2 Chlorophenols also appear in the
environment when water resources containing organic materials,
such as phenolic derivatives (i.e., derived from lignine), or certain
aromatic acids are disinfected using chlorinating agents. Waste-
waters from wood pulp and paper mills contain important levels
of chlorinated organic compounds.3-5 Finally, chlorophenols are
also formed during the combustion of organic matter in the
presence of chlorine or chlorine-containing compounds.6,7
Chlorophenols have, thus, been identified as usual contami-
nants of surface waters. Particularly, 2,4,6-trichlorophenol (2,4,6-
TCP) has been reported to contaminate several rivers and lakes
in Europe, Canada, and South Africa.8,9 Contamination of the
groundwater near sawmills or wastesites at levels ranging from
0.03 to 91.3 ppb has also been reported.10 Similarly, trichlorophe-
nols have been noticed as contaminants in soils and sediments in
Sweden,11 Finland,12 British Columbia in Canada,13 and New
Zealand.14 As a consequence, significant levels of 2,4,6-TCP may
Trichlorophenols (TCP ) eliminated by the urine can be
considered as potential biomarkers of exposure of many
chemicals (chlorophenols, chlorophenoxy acid herbicides,
prochloraz, lindane, hexachlorobenzene, etc). High-
throughput screening methods are necessary to carry out
efficient monitoring programs that may help to prevent
certain occupational health diseases. For this purpose,
an indirect enzyme-linked immunosorbent assay (ELISA)
for 2 ,4 ,6 -trichlorophenol detection has been developed
using polyclonal antisera raised against 3 -(3 -hydroxy-
2,4,6-trichlorophenyl)propanoic acid (hapten 5) covalent-
ly coupled by the mixed anhydride (MA) method to
keyhole limpet hemocyanin (KLH). The indirect ELISA
uses a heterologous coating antigen prepared by conjuga-
tion of 3 -(2 -hydroxy-3 ,6 -dichlorophenyl)propanoic acid
(hapten 4 ) to bovine serum albumin (BSA) using the
active ester (AE) method. The optimum hapten density
for the coating antigen was found to be 3 mol of hapten/
mol of protein. The assay shows a limit of detection of
0 .2 4 5 ( 0 .1 1 6 µg L-1 , and it is performed on 9 6 -well
microtiter plates in about 1 .5 h. The ELISA reported
recognizes on a much less extent other chlorinated phe-
nols, such as 2 ,3 ,4 ,6 -tetrachlorophenol (2 ,3 ,4 ,6 -TtCP ,
2 1 %), 2 ,4 ,5 -TCP (1 2 %) and 2 ,3 ,5 -TCP (1 5 %); however,
brominated phenols (BP ) are even more recognized than
the corresponding chlorinated analogues (ex. 2 ,4 ,6 -TBP ,
7 1 0 %; 2 ,4 -DBP , 1 1 9 %). With the aim of finding an
explanation for this behavior, theoretical calculations have
been performed for those and other halogenated phenols
(2 ,4 ,6 -triiodophenol and 2 ,4 ,6 -trifluorophenol) to clarify
which physicochemical parameter can explain better the
recognition pattern observed. Finally, the assay has been
adapted to the analysis of urine samples. The studies have
shown that a limit of detection of 1 µg L-1 can be
accomplished on this biological matrix by combining the
ELISA procedure with a C1 8 solid-phase extraction
method.
(1) Kumagai, S.; Matsunaga, I. Occup. Environ. Med. 1 9 9 4 , 51, 120-124.
(2) Kauppinen, T.; Kogevinas, M.; Johnson, E.; Becher, H.; Bertazzi, P. A.; Bas
Bueno de Mesquita, H.; Coggon, D.; Green, L.; Littorin, M.; Lynge, E.;
Mathews, J.; Neuberger, M.; Osman, J.; Plannet, B.; Pearce, N.; Winkelmann,
R.; Saracci, R. Am. J. Ind. Med. 1 9 9 3 , 23, 903-920.
(3) Trapp, S.; Rantio, T.; Paasivirta, J. Environ. Sci. Pollut. Res. Int. 1 9 9 4 , 1,
247-252.
(4) Clapp, R. T.; Truemper, C. A.; Aziz, S.; Reschke, T. Tappi J. 1 9 9 6 , 79, 111-
113.
(5) Blaney, C. A.; Hossain, S. U. Chemtech 1 9 9 7 , 27, 48-51.
(6) Angerer, J.; Goen, T.; Lehnert, G. Organohalogen Compd. 1 9 9 3 , 14, 163-
167.
(7) Wrbitzky, R.; Goen, T.; Letzel, S.; Frank, F.; Angerer, J. Int. Arch. Occup.
Environ. Health 1 9 9 5 , 68, 13-21.
(8) Hodson, P. V.; McWhirter, M.; Ralph, K.; Gray, B.; Thivierge, D.; Carey, J.
H.; Van der Kraak, G.; Whittle, D. M.; Levesque, M. C. Environ. Toxicol.
Chem. 1 9 9 2 , 11, 1635-1651.
(9) Owens, J. W.; Swanson, S. M.; Birkholz, D. A. Chemosphere 1 9 9 4 , 29, 89-
109.
(10) Hidin, E. Report number OWRT-C-10108-V(1458), 1983.
(11) Heeb, N. V.; Dolezal, I. S.; Buhrer, T.; Mattrel, P.; Wolfensberger, M.
Chemosphere 1 9 9 5 , 31, 1407-1411.
Despite the fact that the use of chlorophenols as preservatives
has dropped in many developed countries and bleaching proce-
dures use methods other than chlorination processes, chlorophe-
nols are widespread in many environmental and biological
(12) Kukkonen, J. V. K.; Eadie, B. J.; Oikari, A.; Holmbom, B.; Lansing, M. B.
Sci. Total Environ. 1 9 9 6 , 188, 15-27.
(13) Wan, M. T. J. Environ. Qual. 1 9 9 2 , 21, 225-231.
(14) Judd, M. C.; Stuthridge, T. R.; McFarlane, P. N.; Anderson, S. M.; Bergman,
I. Chemosphere 1 9 9 6 , 33, 2209-2220.
* To whom correspondence should be sent. Phone: 93 4006171. Fax: 93
2045904. E-mail: mpmqob@iiqab.csic.es.
468 Analytical Chemistry, Vol. 74, No. 2, January 15, 2002
10.1021/ac010419n CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/13/2001

accumulate in fish and seafood.9,15 Concentrations varying from
0.03 to 0.7 µg L-1 have also been detected in the drinking water
of Janakka (Finland), Ville Mercier (Quebec, Canada), and in
Utah.16-18
The NIOSH (National Institute of Occupational Safety and Health)
had statistically estimated that 851 workers were potentially
exposed to 2,4,6-TCP during a three-year NOES survey.31 Worker
exposure has been reported in plants producing chlorinated
pesticides or fungicides,2,32,33 in pesticide spray operators,2 in
Finnish saw mills,34 industrial incinerator waste plants,7,35 and
electrical utility linemen in contact with chlorophenol-treated poles
used in electrical line construction.36 As an example, levels of 2,4,6-
TCP ranged between 1 and 12 µM (199-2369 µg L-1) in the urine
of workers of a Finnish saw-mill factory where chlorophenols had
been used as preservatives.37 In a German waste incinerator, the
levels of chlorophenols were significantly higher in the urine of
2,4,6-TCP is one of the 5 chlorophenols out of 19 thought to
have significant toxicological effects and potential carcinogenicity.
Although epidemiological human studies have led to inconclusive
results, those carried out on animals have provided sufficient
evidence of the 2,4,6-TCP carcinogenic effects.19 An oral minimal
risk level (MRL) of 0.003 mg/ Kg/ day would be applicable to 2,4,6-
trichlorophenol for an intermediate exposure duration according
to the updated toxicological profile for chlorophenols of the
Agency for Toxic Substances and Disease Registry (ATSDR) of
the U.S. Department of Health and Human Services.20 Similarly,
in Finland, an ADI (average daily intake) of 50 µg/ day has been
established.21
workers in contact with the incinerator than in the urine of those
7
in the administrative section, ranging from 0.16 to 28.30 µg L-1
.
Another important fact of the widespread exposure to chlo-
rophenols is that polychlorinated dibenzodioxins (PCDD), includ-
ing the most toxic dioxin congener, 2,3,7,8-tetrachlorodibenzo-p-
dioxin (TCDD), and polychlorinated dibenzofurans (PCDFs) are
common impurities, because chlorophenols are the precursors
of these substances. The median concentration of TCDD in
workers in a German production plant of 2,4,5-T products was
about 0.1 ppm in 1981.2 Thus, TCDD has been detected in the
lipid fraction of the serum of trichlorophenol plant production
workers.2 Similarly, in an Austrian production plant, an average
level of 340 pg/ g blood lipid was determined, and all plant
production workers showed symptoms of chloracne and neuro-
logical diseases.38-40 Chlorophenols have been named predioxins28
and can be thought of as indicators of the formation of PCDDs.
Therefore, routine monitoring of chlorophenols in urine,
environmental, and food matrixes may be important not only
because of the risk effects of the parent compounds but also
because of its association with dioxins. Establishment of reliable
Exposure of the general population has been noted via con-
taminated environment,3 domestic preservatives, and edible prod-
ucts.22,23 Chlorophenols may also be detected in the urine as a
consequence of the exposure to other chlorinated substances,
such as the bleaching agent prochloraz, chlorophenoxy acid her-
bicides, hexachlorocyclohexanes, and chlorobenzenes.1,22,24-26
Thus, by analyzing the urine of adults in the U.S. who have no
known occupational contact with chlorophenols or related sub-
stances, it was found that 2,4,6-TCP was present in a significant
number of all of the samples analyzed, with a mean value of 10.1
nM (2 µg L-1).22,23 Several studies have been performed in
Germany that reveal that the 95th percentile concentration of 2,4,6-
TCP in the urine of children (10-12 years old) was 1.74 µg L-1 27
,
but for adults the mean concentration of 2,4,6-TCP in urine could
reach values around 4.7 µg L-1 28 in some regions. In a recent
study, significant levels have been found in the plasma of German
adults, despite the fact that the use of chlorophenols as preserva-
tives has been prohibited by law in that country.29,30
Occupational exposure may occur through inhalation and
dermal contact with this compound at workplaces where 2,4,6-
TCP or the above-mentioned substances are used or produced.
(27) Bartels, P.; Ebeling, E.; Kra¨mer, B.; Kruse, H.; Osius, N.; Vowinkel, K.;
Wassermann, O.; Witten, J.; Zorn, C. Fresenius’ J. Anal. Chem. 1 9 9 9 , 365,
458-464.
(28) Wrbitzky, R.; Angerer, J.; Lehnert, G. Gesundheitswesen 1 9 9 4 , 56, 629-
635.
(29) Heudorf, U.; Letzel, S.; Peters, M.; Angerer, J. Int. J. Hyg. Environ. Health
(15) Tavendale, M. H.; Hannus, I. M.; Wilkins, A. L.; Langdon, A. G.; McFarlane,
P. N. Chemosphere 1 9 9 6 , 33, 2273-2289.
(16) Pakdel, H.; Couture, G.; Roy, C.; Masson, A.; Locat, J.; Gelinas, P.; Lesage,
S. Environ. Sci. Pollut. Control Ser. 1 9 9 2 , 4, 381-421.
(17) Nieminski, E. C.; Chaudhuri, S.; Flint, T. I.; Paxman, S. W.; Carman, J. R. J.
Am. Water Works Assoc. 1 9 9 6 , 88, 50-63.
2 0 0 0 , 203, 135-139.
(30) Schmid, K.; Lederer, P.; Goen, T.; Schaller, K. H.; Strebl, H.; Weber, A.;
Angerer, J.; Lehnert, G. Int. Arch. Occup. Environ. Health 1 9 9 7 , 69, 399-
406.
(31) National Institute for Occupational Safety and Health. Industrial Hygiene
Surveys of Occupational Exposure to Wood Preservative Chemicals; Publication
no. 83-106; U.S. Government Printing Office: Washington, DC, 1983.
(32) Messerer, P.; Zober, A.; Becher, H. Environ. Health Perspect. Suppl. 1 9 9 8 ,
106, 733-735.
(18) Nieminski, E. C.; Chaudhuri, S.; Lamoreaux, T. J. Am. Water Works Assoc.
1 9 9 3 , 85, 98-105.
(19) International Agency for Research on Cancer of the World Health Organiza-
tion. Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to
Humans 1 9 8 6 , 41, 319-356.
(20) Agency for Toxic Substances and Disease Registry (ATSDR), U.S. Depart-
ment of Health and Human Services; Minimum Risk Levels for Hazardous
Substances; U.S. Government Printing Office: Washington, DC, April 2001;
Electronic document at http:/ / www.atsdr.cdc.gov/ mrls.html.
(21) Lampi, P.; Vartiainen, T.; Tuomisto, J.; Hesso, A. Chemosphere 1 9 9 0 , 20,
625-634.
(33) Hryhorczuk, D. O.; Wallace, W. H.; Persky, V.; Furner, S.; Webster, J. R.;
Oleske, D.; Haselhorst, B.; Ellefson, R.; Zugerman, C. Environ. Health
Perspect. 1 9 9 8 , 106, 401-408.
(34) Kontsas, H.; Rosenberg, C.; Tornaeus, J.; Mutanen, P.; Jappinen, P. Arch.
Environ. Health 1 9 9 8 , 53, 99-108.
(35) Lindroos, L.; Kosniken, H.; Mutanen, P.; Ja¨rvisalo, J. Int. Arch. Occup.
Environ. Health 1 9 8 7 , 59, 463.
(36) Thind, K. S.; Karmali, S.; House, R. A. Am. Ind. Hyg. Assoc. J. 1 9 9 1 , 52,
547-52.
(22) Hill, R. H., Jr.; Ashley, D. L.; Head, S. L.; Needham, L. L.; Pirkle, J. L. Arch.
Environ. Health 1 9 9 5 , 50, 277-280.
(37) Pekari, K.; Luotamo, M.; J¨arvisalo, J.; Lindroos, L.; Aitio, A. Int. Arch. Occup.
(23) Angerer, J.; Heinzow, B.; Schaller, K. H.; Weltle, D.; Lehnert, G. Fresenius’
J. Anal. Chem. 1 9 9 2 , 342, 433-438.
Environ. Health 1 9 9 1 , 63, 57-62.
(38) Neuberger, M.; Landvoigt, W.; Denrtl, F. Int. Arch. Occup. Environ. Health
1 9 9 1 , 6, 325-327.
(39) Vena, J.; Boffetta, P.; Becher, H.; Benn, T.; Bueno-De-Mesquita, H. B.;
Coggon, D.; Colin, D.; Flesch-Janys, D.; Green, L.; Kauppinen, T.; Littorin,
M.; Lynge, E.; Mathews, J. D.; Neuberger, M.; Pearce, N.; Pesatori, A. C.;
Saracci, R.; Steenland, K.; Kogevinas, M. Environ. Health Perspect. Suppl.
1 9 9 8 , 106, 645-653.
(24) Koransky, W.; Mu¨ nch, G.; Noack, G.; Portig, J.; Sodomann, S.; Wirsching,
M. Naunyn Schmiedebergs Arch. Pharmacol. 1 9 7 5 , 288, 65-78.
(25) Hill, R. H., Jr.; Head, S. L.; Baker, S.; Gregg, M.; Shealy, D. B.; Bailey, S. L.;
Williams, C. C.; Sampson, E. J.; Needham, L. L. Environ. Res. 1 9 9 5 , 71,
99-108.
(26) Guidotti, M.; Ravaioli, G.; Vitali, M. HRC J. High Resolut. Chromatogr. 1 9 9 9 ,
22, 427-428.
(40) Neuberger, M.; Kundi, M.; Jager, R. Toxicol. Lett. 1 9 9 8 , 96-7, 347-350.
Analytical Chemistry, Vol. 74, No. 2, January 15, 2002 469

biomarkers for human exposure assessment for toxic substances
requires specific analytical methods reaching very low limits of
detection (LOD)41,42 in order to carry out appropriate toxicological
studies. Additionally, conducting efficient biological monitoring
programs for exposure assessment requires rapid, quantitative,
and high-throughput screening techniques.43,44 At present, no
efficient screening method exists to analyze clorophenols in urine
as indicators of exposure of the general population or occupation-
ally exposed risk groups to the above-mentioned compounds.
Immunoassays may fulfill these requirements as a result of their
simplicity, specificity, and high detectability.43,45-47 Thus, the
objective of this work has been the development and evaluation
of a heterologous indirect immunoassay for the determination of
2,4,6-TCP in urine samples as a potential tool for the assessment
of human exposure to polluted environments. Furthermore, an
evaluation of the physicochemical parameters governing antibody
recognition has been performed using molecular modeling and
theoretic studies in order to find an explanation of the observed
specificity of the optimized assay.
indicated, purification of the reaction mixtures was accomplished
by “flash” chromatography using silica gel as the stationary phase.
1H and 13C NMR spectra were obtained using a Varian Unity-300
1
(Varian Inc., Palo Alto, CA) spectrometer (300 MHz for H and
1
75 MHz for 13C) or a Gemini 200 (199.975 MHz for H and 50.289
for 13C). Infrared spectra were measured on a Bomen MB120 FTIR
spectrophotometer (Hartmann & Braun, Que´bec, Canada). Gas
chromatography/ mass spectrometry (GC/ MS) was performed on
a MD-800 capillary gas chromatograph equipped with a MS
quadrupole detector (Fisons Instruments, VG, Manchester, U.K.)
with an electronic impact (EI) source of 70 eV provided by a
capillary column (HP-5, 25 mm). The positive ions were recorded
in SCAN mode. The spectroscopic and spectrometric data are
given as Supporting Information.
2-Hydroxy-3,5,6-trichlorobenzaldehyde 1. CHCl3 (6.32 mL, 79
mmol, 3.12 equiv) was added dropwise for about 1 h to a mixture
of 2,4,5-trichlorophenol (5 g, 25.3 mmol), Ca(OH)2 (8.3 g, 0.112
mol), and Na2CO3 (9.1 g, 85.85 mol) in H2O (63.25 mL, 3.5 mol)
previously heated to 65 °C. A white paste was initially formed that
later became yellow. The reaction was stirred for 16 h at 65 °C.
The mixture was then acidified with 1 N HCl to pH 1 and was
distilled under vapor current. The compound obtained was
dissolved in CH2Cl2, dried with anhydrous MgSO4, filtered, and
evaporated to dryness to obtain 3 g of a mixture containing the
starting trichlorophenol and the aldehyde 1 . Separation of these
two compounds was accomplished by column chromatography
using hexane:CH2Cl2 9:1 as the mobile phase. Aldehyde 1 was
obtained pure as a yellow solid (870 mg, 15% yield) and identified
by its spectroscopic and spectrometric data.
EXPERIMENTAL SECTION
Molecular Modeling and Theoretical Calculations. Molec-
ular modeling was performed using the Hyperchem 4.0 software
package (Hyperube Inc, Gainesville, FL). Theoretical geometries
and electronic distributions were evaluated for 2,4,6-trichlorophe-
nol (2,4,6-TCP), 2,4,6-tribromophenol (2,4,6-TBP), 2,4,6-triiodophe-
nol (2,4,6-TIP), and 2,4,6-trifluorophenol (2,4,6-TFP) using semiem-
pirical quantum mechanics MNDO48 and PM349 models. All of
the calculations were performed using standard computational
chemistry criteria. Theoretical calculations regarding pKa and log
P values were carried out using the ACD/ Log P and the ACD/
pKa 1.2 software package (Advanced Chemistry Development Inc.,
Toronto, ON, Canada) at the Department of Analytical Chemistry
(University of Lund, Sweden).
Synthesis of the Haptens. The preparation of hapten 5 , 3-(3-
hydroxy-2,4,6-trichlorophenyl)propanoic acid, has already been
reported.50 Hapten 7, 2-hydroxy-3,5,6-trichlorobenzoic acid, as well
as other chemical reagents, were obtained from Aldrich Chemical
Co. (Milwaukee, WI). The synthesis of haptens 2 -4 and 6 is
described below.
3-(2-Hydroxy-3,5,6-trichlorophenyl)-2-propenoic Acid 2. A mix-
ture of the aldehyde 1 (0.4 g, 1.77 mmol) and methyl diethylphos-
phonacetate (0.4 mL, 2.18 mmol, 1.2 equiv) was added to a
suspension of KOH powder (0.4 g, 7.13 mmol, 4 equiv) in THF
(10.7 mL) under argon atmosphere and magnetic stirring at room
temperature (RT). Solubilization of the KOH followed by the
appearance of a solid was observed. The reaction was followed
by TLC (hexane:CH2Cl2, 7:3), until disappearance of the starting
aldehyde 1 after 1.5 h. The mixture was acidified with 1 N HCl to
obtain a white solid that was separated by filtration. The solid
was dissolved in ethyl acetate, dried with anhydrous MgSO4,
filtered, and evaporated to dryness under reduced pressure to
obtain 0.43 g of a white solid of the acid 2 (90% yield) identified
by its spectroscopic and spectrometric data and elemental analysis.
3-(2-Hydroxy-3,5,6-trichlorophenyl)propanoic Acid 3. The car-
boxylic group was first esterified by dissolving the acid 2 (200
mg, 0.75 mmol) in MeOH (2.5 mL, 63.5 mmol) with few drops of
concentrated H2SO4. The mixture was stirred for 6 h until
complete disappearance of the starting material was observed by
TLC (hexane:ethyl ether, 1:1). The solvent was evaporated, and
the residue was treated with aqueous 5% NaHCO3 and extracted
with ethyl acetate. The organic layer was washed with saturated
NaCl, dried with MgSO4, filtered, and evaporated to obtain 210
mg of methyl 3-(2-hydroxy-3,5,6-trichlorophenol)-2-propenoate
(95% yield). For the reduction of the double bond, a solution of
99.99% CuBr (10.87 mmol, 15.3 equiv) in anhydrous THF (2 mL)
was placed in a three-neck flask with a completely inert Ar
atmosphere, a magnetic stirrer was added, and the solution was
cooled to an inner temperature between -5 and 0 °C. A portion
General Methods and Instruments. Thin layer chromatography
(TLC) was performed on 0.25-mm, precoated silica gel 60 F254
aluminum sheets (Merck, Darmstadt, Germany). Unless otherwise
(41) Grandjean, P.; Brown, S. S.; Reavey, P.; Young, D. S. Clin. Chem. 1 9 9 5 ,
41, 1902-1904.
(42) Timbrell, J. A. Toxicology 1 9 9 8 , 129, 1-12.
(43) Harris, A. S.; Lucas, A. D.; Kramer, P. M.; Marco, M.-P.; Gee, S. J.; Hammock,
B. D. In New Frontiers in Agrochemical Immunoassay; Kurtz, D. A., Skerritt,
J. H., Stanker, L., Eds.; AOAC International: Arlington, VA 1995, pp 217-
235.
(44) Szurdoki, F.; Jaeger, L.; Harris, A.; Kido, H.; Wengatz, I.; Goodrow, M. H.;
Szekacs, A.; Wortberg, M.; Zheng, J.; Stoutamire, D. W.; Sanborn, J. R.;
Gilman, S. D.; Jones, A. D.; Gee, S. J.; Choudrary, P. V.; Hammock, B. D.
J. Environ. Sci. Health. 1 9 9 6 , B31, 451-458.
(45) Jaegler, L. L.; Jones, A. D.; Hammock, A. D. Chem. Res. Toxicol. 1 9 9 8 , 11,
342-352.
(46) Drummer, O. H. J. Chromatogr. B 1 9 9 8 , 713, 201-225.
(47) MacKenzie, B. A.; Striley, C. A. F.; Biagini, R. E.; Stettler, L. E.; Hines, C.
J. Bull. Environ. Contam. Toxicol. 2 0 0 0 , 65, 1-7.
(48) Dewar, M. J. S.; Thiel, W. J. Am. Chem. Soc. 1 9 7 7 , 99, 4899-4917.
(49) Stewart, J. J. P. J. Comput. Chem. 1 9 8 9 , 10, 209-221.
(50) Galve, R.; Camps, F.; Sanchez-Baeza, F.; Marco, M.-P. Anal. Chem. 2 0 0 0 ,
72, 2237-2246.
470 Analytical Chemistry, Vol. 74, No. 2, January 15, 2002

of 2.75 M Red-Al (8 mL, 22 mmol, 31 equiv) was added slowly.
The suspension, which soon acquired a black color, was stirred
for 1 h at 0 °C. The temperature was then decreased to -78 °C,
and anhydrous 2-butanol (2.4 mL, 26.16, 36.8 equiv) was added,
followed (about 5 min later) by the solution of the methyl ester
of the acid 2 (190 mg, 0.67 mmol) in anhydrous THF (5 mL).
After 15 min, the temperature was allowed to reach -20 °C, and
the reaction mixture was stirred for 4 h, when it was added to a
solution of saturated NH4Cl and stirred for 2 h at RT until the
color changed from black to dark blue. The mixture was then
extracted with ethyl ether, and the organic layer was washed with
saturated NaCl, dried with anhydrous MgSO4, filtered, and
evaporated to dryness under reduced pressure to obtain 200 mg
of a yellow-brownish oil that was purified by column chromatog-
raphy using mixtures of hexane:ethyl ether with increased polarity
as the mobile phase to obtain 80 mg of the methyl ester of the
desired compound methyl 3-(2-hydroxy-3,5,6-trichlorophenol)-
propanoate (40% yield) that was identified according to its
spectroscopic and spectrometric data. The ester (45 mg, 0.16
mmol) was then dissolved in THF (3 mL), and a solution of 0.5 N
NaOH (2 mL, 1 mmol) was added, which turned the reaction
mixture yellow. After 4 h at RT, the hydrolysis of the methyl ester
was complete, as observed by TLC (hexane:ether, 1:1). The
solvent was evaporated, and the residue was treated with 0.5 N
NaOH, washed with ethyl acetate, and acidified with 1 N HCl to
produce a precipitate. The aqueous layer was then extracted with
ethyl acetate, dried with MgSO4,and evaporated to dryness to
obtain 35 mg of a pale yellow solid corresponding to the desired
acid 3 (82% yield), which was identified according to its spectro-
scopic and spectrometric data.
(4 mL) and by slowly adding a Na-Pb alloy (10% Na, 5 g; 21.75
mmol; 3.6 equiv). The mixture was stirred for 3 h at RT to ob-
tain 0.47 g (93% yield) of a white solid corresponding to 3-(4-
hydroxyphenyl)-2-propanoic acid, according to its spectroscopic
data. The reduced acid (0.4 g, 2.4 mmol) was then esterified as
described before in MeOH (4 mL, 0.1 mol) and few drops of
concentrated H2SO4 for 12 h at RT to obtain 400 mg of methyl
3-(4-hydroxyphenyl)-propanoate (92% yield). The introduction of
the chlorine atoms was then accomplished by dissolving the ester
(300 mg, 1.67 mmol) in anhydrous CH2Cl2 (1 mL). To this solution,
SO2Cl2 (300 µL, 3.73 mmol) was added, followed by the dropwise
addition of anhydrous ethyl ether (525 µL, 5.02 mmol, 3 equiv).
The reaction, which acquired an intense yellow color, was kept
at RT for 12 h under a nitrogen atmosphere. The mixture was
brought to basic pH with 5% NaHCO3 and extracted with ethyl
ether. The organic layer was then washed with saturated NaCl,
dried with MgSO4, filtered, and evaporated under reduced pres-
sure to obtain 350 mg of a yellow oil that was purified by column
chromatography using mixtures of hexane and ethyl acetate with
increasing polarity to isolate 170 mg (24% yield) of methyl 3-(4-
hydroxy-3,5-dichlorophenyl)-propanoate that was identified ac-
cording its spectroscopic and spectrometric data. The hydrolysis
of the ester was carried out as described above by adding a
solution of 0.5 N NaOH (3.5 mL, 1.75 mmol) to the ester (70 mg,
0.28 mmol) dissolved in THF (4.5 mL). The mixture was stirred
for 12 h at RT to obtain 50 mg (76% yield) of a yellowish solid
corresponding to the desired hapten 6 , identified according to
its spectroscopic data.
P reparation of the P rotein Conjugates. Active Ester (AE)
Method. Following described procedures,51 haptens 2 -4 and 6
(10 µmol) were reacted with NHS (N-hydroxysuccinimide, 5.75
mg, 50 µmol) and DCC (dicyclohexylcarbodiimide, 20.63 mg, 100
µmol) in DMF (dimethylformamide, 200 µL) for about 1 h at RT.
The urea that was formed as a secondary product was precipitated
and removed by centrifugation, and the supernatant was divided
into three equivalent fractions and added to three vials containing
bovine serum albumin (BSA), ovalbumin (OVA), and conalbumin
(CONA) (10 mg/ each) dissolved in 0.2 M borate buffer (1.8 mL).
For certain conjugates, the molar ratio hapten:lysine residues of
the protein were 2:1, 1:1, 1:2.5, and 1:5.
3-(2-Hydroxy-3,6-dichlorophenyl)propanoic Acid 4. A Na-Pb
(10% Na-Pb) alloy (0.6 g, 2.61 mmol, 3.7 equiv) was added to a
solution of the acid 2 (95 mg, 0.36 mmol) in 5% NaOH (600 µL).
The reaction was stirred for 5 h at RT with a slight N2 current to
eliminate the H2 that was formed. The mixture was acidified with
0.1 N HCl to pH 1 and extracted with ethyl acetate. The organic
layer was washed with a saturated NaCl solution, dried with
MgSO4, filtered, and evaporated to dryness to obtain 80 mg of a
white solid containing a mixture of compounds. The crude mixture
was, thus, esterified as described above by dissolving the solid in
MeOH (4 mL), adding a few drops of concentrated H2SO4, and
stirring the mixture for 12 h at RT. The isolation of the desired
compound was accomplished by column chromatography using
mixtures of hexane:ethyl ether with increased polarity as mobile
phase to obtain 50 mg (57% yield) of methyl 3-(2-hydroxy-3,6-
dichlorophenyl)-propanoate, which was identified according to its
spectroscopic and spectrometric data. As described above, the
ester (50 mg, 0.2 mmol) was hydrolyzed by dissolving the
compound in THF (3.25 mL), adding a solution of 0.5 N NaOH
(2.5 mL, 1.25 mmol), and stirring the mixture for 12 h until the
complete disappearance of the starting material was observed by
TLC (hexane:ethyl acetate, 1:1). Treatment of the reaction afforded
40 mg (85% yield) of a white solid of the acid 4 , according to its
spectroscopic data.
Mixed Anhydride (MA) Method. According to described pro-
cedures,52 the haptens 2 -4 and 6 (20 µmol) were reacted with
tributylamine (5.3 µL, 22 µmol) and isobutylchloroformate (3.2
µL, 24 µmol) in anhydrous DMF (200 µL). The solution containing
the activated hapten was then divided into three equivalent
fractions and added to the BSA, OVA and CONA (20 mg/ each)
dissolved in 0.2 M borate buffer (1.8 mL). In both methods, the
protein conjugates were purified by dialysis against 0.5 mM PBS
(four times, 5 L each) and Milli-Q water (1 time, 5 L) and stored
freeze-dried at -40 °C. Unless otherwise indicated, working
aliquots were stored at 4 °C in 10 mM PBS at 1 mg/ mL.
Hapten Density Analysis. Hapten densities of 2 -4 and
6 -BSA conjugates were determined by matrix assisted laser
desorption ionization time-of-flight mass spectrometry (MALDI-
3-(4-Hydroxy-3,5-dichlorophenyl)propanoic Acid 6. Following the
procedure described for hapten 4 , the reduction of the double
bond was carried out by dissolving the commercial acid, 3-(4-
hydroxyphenyl)-2-propanoic acid (0.5 g, 3.04 mmol) in 5% NaOH
(51) Gasco´n, J.; Oubin˜ a, A.; Ballesteros, B.; Barcelo´, D.; Camps, F.; Marco, M.-
P.; Gonza´lez-Martinez, M.-A.; Morais, S.; Puchades, R.; Maquiera, A. Anal.
Chim. Acta 1 9 9 7 , 347, 149-162.
(52) Marco, M.-P.; Hammock, B. D.; Kurth, M. J. J. Org. Chem. 1 9 9 3 , 58, 7548-
7556.
Analytical Chemistry, Vol. 74, No. 2, January 15, 2002 471

TOFMS) by comparing the molecular weight of the standard BSA
and that of the conjugates. The MALDI-MS (matrix assisted laser
desorption ionization mass spectrometer) used for analyzing the
protein conjugates was a Perspective Biosystems Time-of-Flight
(TOF) Mass Spectrometer Voyager-DE RP equipped with a laser
unit that operates with an intensity of 2800. The instrument is
controlled by a BioSpectrometry Workstation equipped with the
software Voyager-DE-RP (version 4.03) developed by Perspective
Biosystems Inc. (Framingham, MA) and GRAMS/ 386 (for Mi-
crosoft Windows, version 3.04, level III) developed by Galactic
Industries Corporation (Salem, NH). MALDI spectra were ob-
tained by mixing 1 µL of the matrix [(E)-3,5-dimethoxy-4-
hydroxycinnamic acid; 10 mg/ mL in CH3CN:H2O 70:30; 0.1% TFA]
with 1 µL of a solution of the conjugates or proteins (5 mg/ mL in
CH3CN:H2O, 70:30; 0.1% TFA).
coating antigens and As dilution were chosen to produce around
0.7-1 units of absorbance in 30 min. The microplates were coated
with the antigens dissolved in coating buffer (100 µL/ well)
overnight at 4 °C. The next day, the plates were washed four times
with PBST, and the antiserum appropriately diluted in PBST was
added to the wells (100 µL/ well) and incubated for 30 min at RT.
The plates were washed as described before, and a solution of
goat anti-rabbit IgG coupled to horseradish peroxidase (anti-IgG-
HRP) in PBST (1/ 6000) was added to the wells (100 µL/ well)
and incubated for 30 min at RT. The plates were washed again,
and the substrate solution was added (100 µL/ well). Color
development was stopped after 30 min at RT with 4 N H2SO4 (50
µL/ well), and the absorbances were read at 450 nm.
Optimized Competitive Indirect ELISA. Microtiter plates were
coated with 4-BSA(1:5) in coating buffer (0.625 µg/ mL, 100 µL/
well), covered with adhesive plate sealers, and allowed to sit
overnight at 4°C. The following day, the plates were washed with
PBST (four times, 300 µL/ well), and 2,4,6-TCP standards (1000-
0.625 nM, prepared in PBS) or samples were added to the coated
plates (50 µL/ well), followed by the sera As43 (1/ 2000 in PBST,
50 µL/ well). The plates were washed again, and after 30 min of
incubation at RT, a solution of anti-IgG-HRP (1/ 6000 in PBST),
was added (100 µL/ well), and the mixture was incubated for 30
min more at RT. The plates were processed as described above.
ELISA Evaluation. The specificity of the immunoassay as well
as the effect of physicochemical features of the media, such as
ionic strength and pH value, were evaluated as previously
reported.50,53
Correlation Studies between ELISA and Gas Chromatography
with Electron-Capture Detector (GC-ECD). PBS solutions were
spiked with 2,4,6-TCP at different concentration levels and split
into two parts for ELISA and GC-ECD analysis. Correlation was
performed by linear regression analysis of the results obtained
by both methods. For ELISA analysis, the PBS solutions were
directly added to the microtiter wells and measured using the
protocol described above. For GC-ECD analysis, tetrachloroni-
trobenzene was added to a concentration of 5 µg L-1, and the
PBS solutions were extracted with toluene (PBS:toluene, 5:1) on
an shaker for 30 min at 300 rpm. The extraction efficiency is 77.78
( 8.35%. The extracts were derivatized by adding of BSTFA (bis-
(trimethylsilyl)trifluoroacetamide, 2 µL) and incubated for 2 h at
RT. Injections (1 µL) were made on a HP 5890A gas chromato-
graph equipped with a HP 7673A autosampler and with an ECD
detector (6238Ni), and HP 3396 Series II integrator (recorder). A
capillary column (25 m × 0.22-mm i.d. × 0.25-µm film thickness)
BPX35 (35% Phenyl(equiv)polysilphenylene-siloxane; SGE Europe
Ltd, U.K.) was used. GC conditions were set as follows: temper-
ature program, 100 °C (1 min) to 250 °C (at 5 °C/ min), 250 °C to
300 °C (at 15 °C/ min); injection temperature, 250 °C; detector
temperature, 300 °C; He, the carrier gas, employed at 130 kPa
(20 cm/ sec); make up gas, N2 5.0.
Immunochemistry. General Methods and Instruments. The
pH and the conductivity of all buffers and solutions were measured
using a pH meter 540 GLP and a conductimeter LF 340,
respectively (WTW, Weilheim, Germany). Polystyrene microtiter
plates were purchased from Nunc (Maxisorp, Roskilde, Denmark).
Washing steps were carried out using a SLY96 PW microplate
washer (SLT Labinstruments GmbH, Salzburg, Austria). Absor-
bances were read using a Multiskan Plus MK II microplate reader
(Labsystems, Helsinki, Findland) or on a SpectramaxPlus (Mo-
lecular Devices, Synnyvale, CA) at a single wavelength mode of
450 nm. The competitive curves were analyzed with a four-
parameter logistic equation using the software Genesys (Lab-
systems), SoftmaxPro v2.6 (Molecular Devices), and GraphPad
Prism (GraphPad Software Inc., San Diego, CA). Unless otherwise
indicated, data presented correspond to the average of at least
two well replicates.
Chemicals and Immunochemicals. Immunochemicals were
obtained from Sigma Chemical Co. (St. Louis, MO). Standards
for cross-reactivity studies were purchased from Aldrich Chemical
Co. (Milwaukee, WI). The preparation of the antisera (As43, As44,
and As45) against TCP by immunizing with hapten 5-KLH
conjugate (3(3-hydroxy-2,4,6-trichlorophenyl)propanoic acid coupled
to keyhole limpet hemocyanin by the mixed anhydride method)
has already been reported.50 The antisera (As) were used without
further purification and were stored frozen in the presence of 0.02%
NaN3. Working aliquots were stored at 4 °C. Protein conjugates
were prepared as described below.
Buffers. Unless otherwise indicated, PBS is 10 mM phosphate
buffer, 0.8% saline solution; and unless otherwise indicated, the
pH is 7.5. Borate buffer is 0.2 M boric acid-sodium borate pH
8.7. Coating buffer is 50 mM carbonate-bicarbonate buffer pH
9.6. PBST is PBS with 0.05% Tween 20. 2× PBST is PBST double
concentrated. Citrate buffer is a 40 mM solution of sodium citrate,
pH 5.5. The substrate solution contains 0.01% TMB (tetrameth-
ylbenzidine) and 0.004% H2O2 in citrate buffer.
Non-Competitive Indirect ELISA. Screening of the coating
antigens and the antisera was performed by a noncompetitive
indirect ELISA protocol as described below. For this purpose, the
avidity of the different As (As43, As44 and As45) was determined
by measuring the binding of serial dilutions (1/ 1000 to 1/ 64 000)
of each antisera to microtiter plates coated with twelve different
2-fold dilutions (from 10 µg/ mL to 9 ng/ mL) of the 2 -4 and
6 -BSA, -CONA,and -OVA conjugates. Optimal concentrations for
Matrix Effect Studies.. Urine samples of the same individual
were collected over 2 days, stored in a refrigerator, and mixed to
have a large and homogeneous urine sample. The urine was
alkalized to pH 8 with 5 N NaOH and centrifuged (3000 rpm, 10
min) to eliminate the particulate matter. The supernatant was then
(53) Oubin˜ a, A.; Ballesteros, B.; Galve, R.; Barcelo´, D.; Marco, M.-P. Anal. Chim.
Acta 1 9 9 9 , 387, 255-266.
472 Analytical Chemistry, Vol. 74, No. 2, January 15, 2002

aliquoted and stored frozen at -40 °C. The samples were thawed
and were diluted at least two times with Milli-Q water prior to
the ELISA studies.
Sample Cleanup. A C18 cartridge (Sep-Pak Plus C18 cartridge,
Waters) was conditioned by passing MeOH (5 mL) followed by
0.1% aqueous acetic acid (5 mL). Urine samples (10 mL) were
then loaded into the column. The cartridge was washed with 5
mL of 50:50 MeOH:H2O (0.1% acetic acid) and 2,4,6-TCP eluted
with 5 mL of 85:15 MeOH:H2O (0.1% acetic acid). About 1-mL
fractions were collected and analyzed. All of the steps were
performed at a flow rate of 1 mL/ min. Under these conditions,
98% of the TCP loaded was eluted in the first fraction. Prior to
the ELISA studies, the fractions were diluted 10 times (initial urine
volume) with PBS.
Analysis of the Urine by GC/ MS. Urine samples (100 mL)
were extracted with toluene (1 mL) and derivatized by adding
BSTFA as mentioned above. For the GC/ MS analysis, injections
(1 µL) were splitless (48 s) solvent delay (5 min). A HP-5MS
(cross-linked 5% Ph Me siloxane, 30 m × 0.25-mm i.d. × 0.25-µm
film thickness) column was used. The ion source temperature was
set at 200 °C, and He was the carrier gas employed at 1 mL/ min.
The ionization energy was 70 eV. GC conditions were as follows:
temperature program, 100-300 °C (7 °C/ min); injector temper-
ature, 250 °C; and ions 253/ 268 (2,4,6-TCP and 2,4,5-TCP), 309/
324 (2,4-DBP), 387/ 402 (2,4,6-TBP), 289/ 304 (2,3,4,6-TtCP) were
monitored in the SIM mode. For the SCAN mode, the mass range
explored was 45-550. All of the data are reported as m/ z (relative
intensity).
Figure 1. Chemical structures of the haptens 1-6 and synthetic
pathways used for their preparation. Hapten 7 was commercially
available. Preparation of hapten 5 has already been described (50).
(i),CH3Cl/Ca(OH)2/Na2CO3/H2O; (ii) methyl diethylphosphonacetate/
KOH in THF; (iii) a, H2SO4/MeOH; b, Red-Al/CuBr/2-BuOH; c, aq
NaOH/THF; (iv) Na-Pb (10% Na)/aq NaOH; (v) a, H2SO4/MeOH; b,
SO2Cl2/Et2O in CH3Cl; c, aq NaOH/THF.
ELISA Matrix Effect Studies. Standard curves of 2,4,6-TCP were
prepared in PBS and in PBS-diluted urine (1/ 2-1/ 256) and run
in the competitive ELISA using the protocol described above to
compare parallelism of the calibration curves.
RESULTS AND DISCUSSION
indirect ELISA format suitable for the analysis of chlorophenols
in urine samples.
In a previous article, we reported the merit of theoretical
calculations and molecular modeling studies for assessing the
selection of hapten 5 as the most appropriate immunizing hapten
to raise antibodies against 2,4,6-TCP.50 A homologous direct ELISA
was then developed and validated for the analysis of water
samples. Urine samples have often been quantitated by ELISA;
however, accurate quantitation of small organic analytes requires
the use of cleanup procedures or the application of high dilution
factors to overcome the effect of the matrix45,54,55. Direct ELISA
formats are more likely to suffer nonspecific interferences from
the sample matrix as a result of the fact that the enzyme label
coexists with the sample, whereas in the indirect ELISA format,
the label is added in a second step. On the other hand, although
it cannot be considered as a general behavior, an improvement
of the immunoassay detectability, by using a competitive hapten
showing a certain degree of heterology from the immunizing
hapten, has often been observed.56,57 In light of these consider-
ations, our aim has been the development of a heterologous
Competitive Hapten Design and Synthesis. Several chemi-
cal structures differing in the position of the spacer arm and the
number and location of the chlorine atoms were initially proposed
(see Figure 1). The haptens possessing the linker at the ortho
position were prepared from the aldehyde 1 . Initial attempts to
obtain this aldehyde moved toward the reduction of the carboxylic
acid of the corresponding commercially available salicylic acid
using DIBAH. However, we did not succeed in controlling the
reduction only to the aldehyde, but most of the time, the reaction
proceeded until the corresponding benzylic alcohol. Reexamining
classical organic chemistry procedures draws to our attention the
direct introduction of the aldehyde at the ortho position of a
phenolic group. Thus, both Duff and Reimer-Tieman reactions
had been used to prepare the aldehyde 1 , although with very low
yields.58-61 Following the procedure described by Stokker et al.,58
we were able to prepare the desired aldehyde at a 15% yield in a
pure form. The introduction of the spacer arm was accomplished
using the Wittig-Horner reaction following the procedure de-
(54) Linder, M. W.; Bosse, G. M.; Henderson, M. T.; Midkiff, G.; Valdes, R. Clin.
Chim. Acta 2 0 0 0 , 295, 179-185.
(55) Shackelford, D. D.; Young, D. L.; Mihaliak, C. A.; Shurdut, B. A.; Itak, J. A.
(58) Stokker, G. E.; Deana, A. A.; de Solms, S. J.; Schultz, E. M.; Smith, R. L.;
Cragoe, E. J., Jr. J. Med. Chem. 1 9 8 1 , 24, 1063-1067.
(59) Hyllar, T. L.; Failla, D. L. J. Org. Chem. 1 9 6 9 , 12, 420-424.
(60) Smith, W. E. J. Org. Chem. 1 9 7 2 , 37, 3972-3973.
(61) Blazevic, N.; Kolbah, D.; Belin, B.; Sunjic, V.; Kajfez, F. Synthesis 1 9 7 9 ,
161.
J. Agric. Food Chem. 1 9 9 9 , 47, 177-182.
(56) Schneider, P.; Goodrow, M. H.; Gee, S. J.; Hammock, B. D. J. Agric. Food
Chem. 1 9 9 4 , 42, 413-422.
(57) Abad, A.; Moreno, M. J.; Montoya, A. J. Agric. Food Chem. 1 9 9 8 , 46, 2417-
2426.
Analytical Chemistry, Vol. 74, No. 2, January 15, 2002 473

scribed by Texier-Bouller and Foucaud,62 affording hapten 2 at a
95% yield. Preparation of hapten 3 by reduction of the double bond
of 2 was initially approached by using a Na-Pb alloy.63 This
method that had afforded excellent results during the preparation
of the immunizing hapten 5 50 produced the hydrogenolysis of the
chlorine atom at the para position, leading to the saturated acid
4 (≈60% yield), instead of the expected acid 3 . All attempts using
described procedures50,64,65 to chlorinate the para position of
hapten 4 to obtain 3 were unsuccessful, leading to complex
reaction mixtures. In light of these results, we decided to explore
alternative reduction methods for the double bond present in the
spacer arm of the acid 2 . From all methods explored (H2/ Pd,
Na2S2O4, etc) only Red-Al (sodium bis(2-methoxyethoxy aluminum
hydride)66 in the presence of CuBr and 2-butanol accomplished
the reduction of the double bond of the methyl ester of 2 (40%
yield). The ester hydrolysis afforded the desired hapten 3 .
Synthesis of hapten 6 used the commercially available 4-hy-
droxycinnamic acid as the starting material. Reduction of the
double bond was accomplished using a Na-Pb alloy in NaOH
solution at a very high yield. The introduction of the chlorine
atoms was performed in the corresponding methyl ester with
SO2Cl2 and Et2O in CH2Cl2, as previously reported.50,65 However,
the expected methyl 3-(4-hydroxy-3,5-dichlorophenyl)propanoate
was obtained with only a 24% yield isolated from a complex
mixture of compounds that could not be identified. We suspect
that the reason for this low yield was the chlorination of the
activated benzylic position in para versus the hydroxy group.
Previously, we had used this method with a very good yield to
chlorinate an aromatic ring that also had a benzylic group, but in
a less activated meta position.50 The desired hapten 6 was obtained
after the hydrolysis of the ester.
Screening of the Antiserum Avidity for the Im m u-
noreagents and Their Influence on the Immunoassay De-
tectability. Haptens 2-4 and 6, initially conjugated to BSA, OVA,
and CONA using the active ester (AE) method, afforded assays
with acceptable detectability but showed low solubility in aqueous
buffers, causing stability and reproducibility problems. Analyses
made by MALDI-MS revealed a high, but not unusual, degree of
conjugation (see Table 1). The lipophilic character conferred by
the chlorine atoms of the analytes to the conjugates or to the
undesired side conjugation reactions produced by the carbodi-
imide67 could be some of the reasons for the lack of stability of
the immunoreagents. In contrast, the conjugates prepared by the
MA method were water-soluble and had lower conjugation yields
(see Table 1); however, only those prepared with hapten 4 and
the homologous antigen 5BSA afforded competitive assays,
although with a low detectability and background noise (see Table
2), which can be attributed to the use of a homologous coupling
method.68
Table 1. Hapten Densities Measured for the 1-6BSA
Conjugates
conjugation
methoda
hapten
conjugate
hapten:proteinb
densityc
2 BSA
AE
MA
AE
MA
AE
MA
MA
AE
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
14
8
3 BSA
4 BSA
10
10
17
18
7
5 BSA
6 BSA
25
12
MA
a
Method of conjugation: AE active ester, MA, mixed anhydride.
b Hapten:protein molar ratio employed during the conjugation reaction.
The ratio is calculated in terms of lysine residues of the protein.
c Hapten density was measured by MALDI-TOFMS.
Table 2. Immunoassay Features of the Competitive
ELISAa
Asa
CAb
conj. method Amax Amin slope IC50c
r2
43 4 BSA
AE
MA
AE
MA
AE
MA
MA
AE
1.021 0.001 -0.54
1.292 0.09 -1.08
1.509 0.011 -1.13
0.946 0.136 -1.67
1.233 0.009 -1.45
0.697 0.035 -1.36
0.940 0.335 -1.33 237.7
0.971 0.047 -1.46
1.334 0.068 -1.42
0.96 0.984
10.08 0.982
3.57 0.991
21.31 0.988
1.72 0.979
5.29 0.996
0.894
9.47 0.964
14.58 0.987
4 CONA
4 OVA
5 BSA
6 BSA
6 OVA
AE
44 4 BSA
4 CONA
5 BSA
AE
AE
MA
AE
1.747 0.001 -0.78
1.469 0.001 -0.70
24.21 0.988
23.31 0.984
1.305 0.01
-0.7
186.6
0.971
6BSA
0.951 0.051 -0.81
21.94 0.967
45 4 BSA
4 CONA
5 BSA
AE
AE
MA
AE
AE
1.848 0.017 -1.23
1.075 0.001 -0.99
1.244 0.244 -1.17
1.155 0.213 -1.28
0.849 0.001 -0.89
39.37 0.975
19.19 0.981
29.48 0.964
76.43 0.864
64.46 0.980
6 BSA
6 OVA
a
As, antisera. b CA, coating antigen. c IC50 values are expressed in
µg L-1
.
All of these considerations encouraged us to rationally prepare
new coating antigens with hapten 4 that had afforded the best
competitive immunoassays, using again the AE method by
controlling very well the hapten density on one side to avoid the
solubility problems that were observed but also to improve
immunoassay detectability. Table 3, shows the features of the
competitive assays obtained. A higher detectability can be
observed when using a coating antigen with the lower hapten
density, which supports the hypothesis of the existence of a certain
hapten density value for optimum immunoassay performance.69,70
Thus, the combination As43/ 4BSA(1/ 5) was selected for further
evaluation of an immunoassay to analyze chlorophenols in urine
samples (see Figure 2 and Table 4 for immunoassay parameters).
Immunoassay Evaluation. Physicochemical Properties of the
Media. Considering the particular properties of the urine samples
(pH e 6 and conductivity between 25 and 30 mS cm-1), it was
(62) Texier-Boullet, F.; Foucaud, A. Synthesis 1 9 7 9 , 884-885.
(63) Tabei, K.; Hiranuma, H.; Amemiya, N. J. Org. Chem. 1 9 6 5 , 39, 1085-1086.
(64) Goldberg, Y.; Alper, H. J. Org. Chem. 1 9 9 3 , 58, 3072-3075.
(65) Corey, E. J.; Letavic, M. A.; Sarshar, S. Tetrahedron Lett. 1 9 9 4 , 35, 7553-
7556.
(66) Semmelhack, M. F.; Stauffer, R. D.; Yamashita, A. J. Org. Chem. 1 9 7 7 , 42,
3180-3188.
(67) Goodfriend, T. L.; Levine, L.; Fasman, G. D. Science 1 9 6 4 , 144, 1344-
1346.
(68) Gendloff, E. H.; Casale, W. L.; Ram, B. P.; Tai, J. H.; Pestka, J. J.; Hart, L.
P. J. Immunol. Methods 1 9 8 6 , 92, 15-20.
(69) Garden, S. W.; Sporns, P. J. Agric. Food Chem. 1 9 9 4 , 42, 1379-1391.
(70) Carlson, R. E. In Immunoanalysis for Agrochemicals; Nelson, J. O., Karu, A.
E., Wong, R. B., Eds.; American Chemical Society: Washington, DC, 1995;
Vol. 586, pp 141-152.
474 Analytical Chemistry, Vol. 74, No. 2, January 15, 2002

Table 3. Effect of the Hapten Density on the Immunoassay Features
CA
hapten:proteina
hapten densityb
Amax
Amin
slope
IC50
r2
4 BSA
2:1
1:1
1:2.5
1:5
13
10
7
0.771 ( 0.214c
0.820 ( 0.263
1.209 ( 0.355
1.018 ( 0.301
0.075 ( 0.097
0.066 ( 0.096
0.056 ( 0.106
0.046 ( 0.092
-1.16 ( 0.13
-1.07 ( 0.11
-1.19 ( 0.07
-1.15 ( 0.06
2.25 ( 0.04
3.05 ( 0.13
3.12 ( 0.13
1.67 ( 0.16
0.991 ( 0.006
0.991 ( 0.009
0.939 ( 0.003
0.977 ( 0.007
3
a
Hapten:protein molar ratio used for the conjugation reaction. The molar ratio of the protein has been calculated in terms of the lysine residues
available. b Hapten density calculated by MALDI-TOFMS. c Average and standard deviation of two experiments performed on different days. The
IC50 is expressed in µg L-1
.
Table 4. Features of the Optimized Immunoassay
Antiserum 43/4BSAa
Amax
0.893 ( 0.177
0.041 ( 0.042
1.53 ( 0.35
-1.24 ( 0.19
4.49 ( 0.79 to 0.45 ( 0.24
0.24 ( 0.12
0.991 ( 0.006
18
Amin
IC50, µg L-1
slope
dynamic range, µg L-1
LOD, µg L-1
r2
N
a
The parameters are extracted from the four-parameter equation
used to fit the standard curve. The data presented correspond to the
average of 18 calibration curves run in different plates through a
1-month period. Each curve was built using three-well replicates.
Figure 2. Calibration curve of the 2,4,6-TCP heterologous indirect
immunoassay. The data presented correspond to the average and
the standard deviation of nine assays run on three different plates.
The curves were run in triplicate. See Table 4 for the features of the
optimized immunoassay.
mental and theoretical studies that a free phenol group was also
essential for antibody recognition.50 As in the homologous direct
format, brominated phenols were highly recognized in this assay.
Hence, 2,4,6-tribromophenol (TBP) crossreacted 710%, and 2,4-
dibromophenol (DBP), 119%. When the bromine atom in the para
position was replaced by a chlorine atom, the crossreactivity
dropped to 31%. As before, the bromine at position 4 was
necessary; if only the ortho positions had bromine atoms, the
analyte was not recognized. In every case, the recognition of the
brominated analytes was greater than that of the chlorinated
analyte analogues.
Influence of the P hysicochemical Features of the Analytes
in the Antibody Recognition. We have been discussing several
hypotheses to explain this high recognition of the brominated
analogues. One reason could be the size of the bromine atoms,
that being greater than the chlorine atoms (van der Waals radii
of 1.95 and 1.77 Å, respectively), they could fit better into the
antibody binding cavities. A second explanation could be the
differences in the pKa or the acidity of the phenol group. Lower
pKa values would induce the preferential existence of the depro-
tonated form of the analyte in assay media, favoring its electrostatic
interaction with the antibody. As it has been demonstrated in this
and other previously reported immunoassays for phenolic com-
pounds,50,53,71 these kinds of interactions may have a large
participation in the formation of the complex. A third explanation
could be found in the differences of the electronic distribution
between chlorinated and brominated analogues, which could
create differences in dipole-dipole interactions in the antigen-
antibody complex. Finally, we have also been considering the
hydrophobic interactions that, according to published works, may
also play an important role in the stabilization of the immuno-
complex.72,73
necessary to know the behavior of the immunoassay when varying
these parameters. As reported in other immunoassays for phenolic
compounds,50,53,71 a greater tolerance to basic conditions was also
observed in this case. The assay performs well between pH 7.5
and 9.5, although the detectability slowly decreased from pH 6.5
to pH 10.5. At pH values lower than 6, the assay is inhibited.
Regarding the effect of the ionic strength of the media, detect-
ability of the assay did not change significantly from 12 to 25 mS
cm-1, although the maximum signal slowly decreased in this range
from 1.3 to 0.7 units of absorbance. These experiments demon-
strated the necessity of buffering and lowering the ionic strength
of the urine sample before its measurement by ELISA.
ELISA Specificity. The degree of recognition was directly
related to the presence of the three chlorine atoms at the ortho
and para position, as in the target analyte as previously reported,50
but now the specificity of this assay was even greater (see Table
5). The best-recognized analyte, 2,3,4,6-tetrachlorophenol (TtCP),
possessing the three necessary chlorine atoms at the right
positions, was only recognized 20% of the time, but in the
50
previously developed assay, it crossreacted 86% of the time,
which means that now the fourth chlorine atom at the meta
position disfavors competition with hapten 4 for the antibody
binding. Other trichlorophenols, such as 2,4,5- and 2,3,5-TCP
crossreacted only 12 and 13%, respectively. 2,4-DCP was the only
dichlorophenol that was recognized, demonstrating again that the
presence of the chlorine atom in the para position was necessary.
Thus 2,6-DCP did not crossreact at concentrations higher than
2000 nM. Pentachlorophenol (PCP) and 2,3,5,6-TtCP were not
recognized, either. Previously, we had demonstrated by experi-
(71) Oubin˜ a, A.; Barcelo´, D.; Marco, M.-P. Anal. Chim. Acta 1 9 9 9 , 387, 266-
279.
(72) Van Oss, C. J. J. Mol. Recognit. 1 9 9 7 , 10, 203-216.
Analytical Chemistry, Vol. 74, No. 2, January 15, 2002 475

Table 5. Interference Caused by Structurally Related Chemicals, Expressed by Their IC50 and the Percentage of
Crossreactivitya
no. halogensb
phenolic compds
PCP
IC50 (nM)
>2000
32.5
% CR
brominated analogues
PBP
IC50 (nM)
% CR
5
4
<0.05
>2000
<0.05
2,3,4,6-TtCP
2,3,5,6-TtCP
21
<0.05
>2000
3
2
1
2,4,6-TCP
2,4,5-TCP
2,3,5-TCP
8.54 ( 1.44
100
12
13
2,4,6-TBP
1.5
710
56.5
52.1
2,4-DCP
410.1
2
2-B-4-CP
2,4-DBP
2,6-DBP
25.1
6.7
>2000
32
119
<0.05
2,6-DCP
>2000
<0.05
4-BP
232
4
a
Crossreactivity is expressed as % of the IC50 of 2,4,6-TCP/ IC50 phenolic compound. b B, bromo; C, chloro; DCP, dichlorophenol; TCP,
trichlorophenol; TtCP, tetrachlorophenol; PCP, pentachlorophenol; BP, bromophenol; DBP, dibromophenol; TBP, tribromophenol; PBP,
pentabromophenol.
Table 6. Crossreactivity Data of Trihalogenated
Phenolic Compounds Compared to Some of Their
Physicochemical Features
A
analytea
%CR
710
205
100
pKa
R (PhO-)b
log P
2,4,6-TBP
2,4,6-TIP
2,4,6-TCP
2,4,6-TFP
6.34 ( 0.23
6.47 ( 0.23
6.59 ( 0.23
7.47 ( 0.23
0.98
0.97
0.96
0.77
4.33 ( 0.49
3.88 ( 0.50
3.58 ( 0.33
2.24 ( 0.50
<0.05
B
analytea
%CR
sizec
surfaced
volumee
Figure 3. Accuracy of the ELISA method compared to the GC-
ECD method. Spiked samples were prepared in PBS and split into
two parts to be analyzed by both methods. Samples were directly
measured for ELISA, but for GC-ECD, samples were extracted in
toluene in the presence of an internal standard, derivatized, and then
injected into the chromatograph. ELISA shows better accuracy than
the GC-ECD method. Correlation between both techniques agrees
to the following equation y ) 1.26x - 4.27 and has a regression
coefficient of r2 ) 0.993.
2,4,6-TBP
2,4,6-TIP
2,4,6-TCP
2,4,6-TFP
710
205
100
<0.05
1.95
2.10
1.77
1.30
342
364
315
265
338
422
280
158
C
analytea
%CR
ACSf
HACg
HCDh
2,4,6-TBP
2,4,6-TIP
2,4,6-TCP
2,4,6-TFP
710
205
100
<0.05
-0.449
-0.430
-0.609
-0.321
-0.100
-0.111
-0.026
-0.125
-3.21
-2.86
-1.12
-13.5
expressed as the dissociated fraction at the working pH, we could
see that almost all of the compounds are in the phenolate form,
with the exception of the trifluoroderivative in which only 77%
exists as phenolate. This factor could have some influence in the
poor recognition of the last compound but could not justify the
differences among the other three. On the other hand, it indicates
clearly that the recognized compound is actually the phenolate
and not the phenol structure. The log P (n-octanol/ water partition
coefficient) calculated values for the four studied phenols (see
Table 6, section A) show a good correlation with the observed
recognition values. This parameter is related, at the microscopic
level, with the differential (solute:water, solute:octanol) hydrogen
bond interactions, the hydrophobic-hydrophilic interactions with
the solvents, and the energy needed to open a cavity by the solute
molecule in the solvent. All of these interactions are similar to
those involved in the antigen/ antibody recognition event73 except
for the fact that in the last case, they are spatially oriented. Thus
the n-octanol/ water partition could be a good model for the
antibody recognition.
a
2,4,6-TBP, 2,4,6-tribromophenol; 2,4,6-TIP, 2,4,6-triiodophenol; 2,4,6-
TCP, 2,4,6-trichlorophenol; 2,4,6-TFP, 2,4,6-trifluorophenol. b Relative
fraction of the phenoxide at pH ) 8.0. c van der Waals radii of the
halogen atom expressed in angstroms. d Accessible molecular surface
for phenoxide ion in Å2. e Volume of the minimal box containing the
molecule expressed in Å3. f Sum of the aromatic carbons point charges
expressed in e- units. g Average of the halogen atoms point charge in
e- units. h Halogen atom charge density in (e-/ Å3) × 103.
To assess all of these hypotheses, the recognition of several
2,4,6-halogenated phenolics in the assay was investigated. This
set of compounds has comparable directional interactions with
the receptor pocket in the antibody. Thus, the observed changes
in the affinity could be attributed to the variations in the strength
of the molecular interactions and the molecular volume and
surface. Table 6 shows some of these molecular parameters,
macroscopic and microscopic, that could explain the observed
relative affinities.
The acidity of these compounds is similar, with the exception
of the fluorinated derivative (see Table 6, section A). If it is
The size of the halogen atoms, their van der Waals radii, moves
in a short range (only a 16% from chlorine to iodide) and could
have only a little influence, considering the dimensions (molecular
(73) Van Oss, C. J. Int. J. Bio-Chromatogr. 1 9 9 7 , 3, 1-8.
476 Analytical Chemistry, Vol. 74, No. 2, January 15, 2002

Figure 4. Calibration graphs showing the interference of the urine in the ELISA method that was developed. Under the present immunoassay
conditions, the urine has to be diluted at least 50 times to ensure accurate results, which results on a LOD of ∼10 µg L-1. A cleanup step
consisting of a C18-SPE procedure allows direct measurement of the urine samples by just diluting the sample five times with the assay buffer.
A, calibration curves obtained with untreated urine; B, calibration graphs obtained after C18-SPE cleanup step.
the urine used in this study was first characterized by GC/ MS in
order to detect the presence of the most important phenolic
compounds interfering with the ELISA (2,4,6-TCP, 2,4,5-TCP,
Table 7. Concentration Levels of Chlorophenols Found
in Urine by GC/MS and Their Equivalents in 2,4,6-TCP
Immunoreactivity (IR)a
2,3,4,6-TtCP, 2,4,6-TBP, and 2,4-DBP; see above results from the
analyte
GC/ MS, µg L-1
% CR
2,4,6-TCP IR, µg L-1
crossreactivity studies). As is shown in Table 7, some of these
compounds were detected at the trace level. Considering the
crossreactivity values of those compounds in the ELISA, we
estimated that the total interference of this “blank urine” could
be equivalent to 0.18 µg L-1 of TCP immunoreactivity (TCP-IR).
2,4,6-TCP
2,4,5-TCP
2,3,4,6-TtCP
2,4,6-TBP
2,4-DBP
0.06
ndb
nd
nd
0.1
100
12
21
710
119
0.06
-
-
-
0.119
total IR
0.179
Because the LOD of the present ELISA is around 0.24 µg L-1
,
the urine sample used in this study could be considered to be a
real “blank matrix”.
a
2,4,6-TCP immunoreactivity equivalents was calculated considering
the percentage of cross-reactivity in the assay. b nd, not detected.
To assess the effect of the urine on the immunoassay, the urine
was collected and alkalized to enhance the solubility of the
chlorophenols, and the pH and conductivity were adjusted to place
the sample within the working conditions of the immunoassay.
Standard curves were prepared in raw urine and PBS-diluted urine.
Figure 4A shows a significant effect of the matrix interfering with
the assay. The urine has to be diluted over 50 times to obtain
reliable results when measuring urine samples. As was mentioned
in the Introduction, levels of chlorophenols of occupational
exposed workers can be high enough to directly analyze their
urine after diluting the sample with the assay buffer; however,
many samples containing significant levels of chlorophenols would
be considered as negative samples, because by diluting the sample
50 times, the LOD of the assay would have diminished to 10 µg
L-1, instead of the 0.2 µg L-1 shown when analyzing buffered
samples.
surface and volume) of the analytes (see Table 6, section B) in
respect to those for the binding pocket in the antibody. Further-
more, we take into account that the protein side chains in this
domain have some flexibility.74 Finally, partial atomic charges
could be related with dipole-dipole interactions (permanent or
induced), which are the main forces under the hydrophobic-
hydrophilic interactions. We have considered the sum of the
charges over the aromatic carbons and the average of the halogen
atoms charge (see Table 6, section C). These two parameters
hardly correlate with the recognition observed, but point charges
are poor descriptors for dipole interactions, whereas atomic charge
densities (expressed as the relation between point charges and
van der Waals atomic volume) are more realistic descriptors and
could be better related to the observed affinities.
In light of these results, a C18-based solid-phase extraction
(SPE) method was applied to extract 2,4,6-TCP from the urine in
order to diminish its interference in the ELISA. Figure 4B shows
that an important improvement can be achieved by introducing
this cleanup method, because urine analysis can be accomplished
after just 5-fold dilution with PBS. Thus, by combining C18-SPE
and ELISA, urine samples containing TCP-IR levels over 1 µg L-1
may now be detected. The LOD that is reached is sufficient for
occupational exposure assessment if we consider, as mentioned
in the Introduction, that the general population can have back-
ground levels of TCP-IR up to this value or even higher.
Correlation between GC-ECD and ELISA. Spiked PBS samples
(n ) 8) were analyzed by ELISA and by GC-ECD. The correlation
between both techniques is good enough and conforms to the
equation y ) 1.26x - 4.27 with a regression coefficient of r2
)
0.993. With respect to the spiked values, the results are shown in
Figure 3. It is possible to observe that the ELISA results (slope
) 0.89) are closer to the spiked values than the GC-ECD results
(slope ) 0.70), which can be explained by the null sample
treatment necessary to analyze these aqueous samples by ELISA.
Matrix Effect Studies. A blank urine sample would be desirable
to assess matrix effect in the ELISA; however, as was mentioned
in the Introduction, the general population is exposed to xenobi-
otics that are eliminated as halophenols in urine. For this reason,
CONCLUSIONS
Biological monitoring requires rapid and efficient high-
throughput screening methods. Analysis of xenobiotics or their
metabolites in urine can be used as biomarkers of exposure to
(74) Van Regenmortel, M. H. V. Biomer. Pept. Prot. Nucleic Acids 1 9 9 5 , 1, 109-
116.
Analytical Chemistry, Vol. 74, No. 2, January 15, 2002 477

certain toxic compounds of industrial origin. Chlorophenols are
widespread in the environment as a result of their extensive use
as wood and textile preservatives and also because of their
industrial use as intermediates for many other chemicals (pesti-
cides, bleaching agents, etc.). Biological monitoring of trichloro-
phenols as indicators of exposure is important not only because
of their own toxicological risk effects as the parent compounds
(trichlorophenols, lindane, hexachlorobenzene, prochloraz, etc)
but also because of their association with dioxins. Occupationally
exposed people constitute important risk groups because of their
contact with high concentrations of these compounds. Therefore,
it would be desirable to provide occupational practitioners and
laboratories with screening tools to assess levels of exposure in
order to provide appropriate health surveillance to the exposed
population. Because of its simplicity and its capacity to process
many samples simultaneously, the immunochemical technique
described in this paper could provide a useful tool to carry out
biological monitoring studies in target industrial sectors where
exposure to the mentioned chlorinated substances may occur.
Despite the complexity of the urine samples, the protocol
described in this paper allows measurement of chlorophenols in
this matrix with a LOD of around 1 µg L-1. To reach this detection
level in urine, the immunochemical method reported here needs
a cleanup step. A C18-SPE procedure has been shown to eliminate
most of the interferences from the matrix. Despite this, the ELISA
reported here allows the simultaneous measurement of many
samples. Furthermore, many commercial suppliers provide SPE
methods on a 96-well configuration to allow high-throughput
sample treatment on a format easily adjustable to the microplate-
based ELISA methods. We must also consider that chlorophenols
may appear in urine not only as free phenols but also as
glucuronides and sulfate conjugates; therefore, further studies will
be addressed to validate the applicability of the method presented
here to the analysis of hydrolyzed urine from individuals.
ACKNOWLEDGMENT
This work has been supported by the EC Program (Contract
IC15-CT98-0910) and by CICYT (AMB98-1048-C04-01 and ALI99-
1379-CE). Mikaela Nichkova thanks the Spanish Ministry for
Science and Technology for its fellowship of the FPU program.
SUPPORTING INFORMATION AVAILABLE
Spectroscopic data (1H and 13C NMR, IR), spectrometric data
(MS), and elementary analysis of the compounds described in
the paper are available as Supporting Information. This material
is available free of charge via the Internet at http:/ / pubs.acs.org.
Received for review April 11, 2001. Accepted October 9,
2001.
AC010419N
478 Analytical Chemistry, Vol. 74, No. 2, January 15, 2002