Influence of the hapten conjugation site on the characteristics of antibodies generated against metabolites of clostebol acetate

Article abstract of DOI:10.1021/jf9906984

4-chloro-androst-4-ene-3,17-dione (CLAD) and 4-chlorotestosterone (clostebol, β-CLT or CLT) were made immunogenic by coupling to protein carriers via the 3 and 17 positions, respectively. These immunogens were used to elicit polyclonal and monoclonal antibodies to CLAD and to clostebol. The antibodies were characterized in an enzyme immunoassay for sensitivity and specificity. Polyclonal antisera generated through position 17 reacted preferentially with 4-chlorotestosterone-17-acetate (clostebol acetate, CLTA), 4-chloro-epitestosterone (epi-clostebol, 17α-clostebol, 17α-CLT), and clostebol, whereas polyclonal antisera generated through the 3 position almost did not react with these derivatives. Interestingly, the monoclonal antibody generated through the 3 position recognized (35%) epi-clostebol. These results suggest that polyclonal antisera generated through the 17 position have a broad specificity profile and can be used to analyze by immunoassay methods urinary metabolites of clostebol acetate and thereby detect the illegal use of clostebol acetate in livestock farming.

Full text of DOI:10.1021/jf9906984

J. Agric. Food Chem. 2000, 48, 3633−3638
3633
In flu en ce of th e Ha p ten Con ju ga tion Site on th e Ch a r a cter istics of
An tibod ies Gen er a ted a ga in st Meta bolites of Clostebol Aceta te
,
†
†
‡
§
Patricia Crabbe,* Carlos Van Peteghem, Martin Salden, and Fort u¨ ne Kohen
Laboratory of Food Analysis, Faculty of Pharmaceutical Sciences, University of Ghent, Harelbekestraat 72,
9
000 Ghent, Belgium, Euro-Diagnostica B.V., Arnhem, The Netherlands, and Department of Biological
Regulation, The Weizmann Institute of Science, Rehovot, Israel
4
-chloro-androst-4-ene-3,17-dione (CLAD) and 4-chlorotestosterone (clostebol, â-CLT or CLT) were
made immunogenic by coupling to protein carriers via the 3 and 17 positions, respectively. These
immunogens were used to elicit polyclonal and monoclonal antibodies to CLAD and to clostebol.
The antibodies were characterized in an enzyme immunoassay for sensitivity and specificity.
Polyclonal antisera generated through position 17 reacted preferentially with 4-chlorotestosterone-
1
7-acetate (clostebol acetate, CLTA), 4-chloro-epitestosterone (epi-clostebol, 17R-clostebol, 17R-CLT),
and clostebol, whereas polyclonal antisera generated through the 3 position almost did not react
with these derivatives. Interestingly, the monoclonal antibody generated through the 3 position
recognized (35%) epi-clostebol. These results suggest that polyclonal antisera generated through
the 17 position have a broad specificity profile and can be used to analyze by immunoassay methods
urinary metabolites of clostebol acetate and thereby detect the illegal use of clostebol acetate in
livestock farming.
Keyw or d s: Immunogen synthesis; conjugation site, metabolites, clostebol acetate, 4-chlorandrost-
4
-ene-3,17-dione, cross-reactivity
INTRODUCTION
The importance and abundance of certain metabolites
also depend on the method of administration of the
parent drug, the age, and the sex of the animals. In
general, it is possible to distinguish two groups of major
metabolites (Walshe et al., 1998): after intramuscular
administration the main metabolites are epi-clostebol,
Clostebol acetate is an anabolic steroid which is often
used for illegal fattening purposes in livestock farming.
To safeguard public health, the use of anabolic steroids
as growth-promoting agents in cattle has been prohib-
ited under Directive 86/469 of the European Commis-
sion (EC, 1986).
In Belgium, routine screening of illegal use is orga-
nized by the Institute of Veterinary Inspection, which
depends on the Ministry of Public Health. Notwith-
standing the total ban, illegal use in Belgium is rife.
Analyses in our laboratory (Van Oosthuyze et al., 1994)
showed that the contribution of clostebol acetate to the
positive injection sites increased from 52% in 1990 to
4
-chloro-4-androst-4-en-3R-ol-17one, and CLAD, while
after oral intake in addition to these there is also a
higher abundance of 4-chloro-4-androsten-3R,17â-diol
and 4-chloroandrostane-3â-ol-17-one.
About 95% of these appear in urine as sulfate conju-
gates. Only 4-chloro-4-androst-4-en-3R-ol-17-one has
been found for 20% to 25% as glucuronide conjugate.
On the basis of this knowledge, the abuse of clostebol
acetate can be evidenced by the presence of certain
unique metabolites in urine. The parent drug and its
metabolites have a typical 4-chlorine group which gives
them a higher anabolic and a less androgenic activity
compared to non-chlorinated steroids. Since there is no
knowledge of the existence of endogenous chlorinated
steroids in cattle, the presence of one of those chlori-
nated metabolites in urine can be used as an indication
of illegal use. This explains why the 4-chloro group is
an important feature for recognition by the antibodies.
The main objective of this work was to generate a
suitable antiserum for immunochemical screening of the
metabolites of clostebol acetate (e.g., CLAD, epi-closte-
bol, etc.) in urine of food-producing animals.
Since the site of attachment of clostebol to protein
carriers elicits formation of antibodies with enhanced
specificity toward selected parts of the steroid molecule,
we utilized two different hapten-conjugation methods,
one through the 3 position and the other through the
17 position. In this report we describe the characteristics
of monoclonal and polyclonal antibodies (CLAD antise-
7
6% in 1993.
In previous studies, the excretion and biotransforma-
tion of clostebol acetate in humans (Debruyckere et al.,
992, 1994; Van Puymbroeck et al., 1996) and in cattle
Hendricks et al., 1994; Le Bizec et al., 1993, 1996, 1998;
1
(
Andr e´ et al., 1994; Leyssens et al., 1994) after intra-
muscular injection or oral intake were studied, and a
wide range of urinary metabolites (Figure 1) were
identified using HPTLC and GC-MS.
When analyzing urine samples of treated animals, in
general no traces of the parent drug clostebol acetate
were detected. Comparing the work of several authors,
there is no doubt about the identity of the different
metabolites, though their relative concentrations do not
always correspond.
*
To whom correspondence should be addressed (telephone
+
32-9-264.81.34; fax +32-9-264.81.99; e-mail Patricia.Crabbe@
rug.ac.be).
†
‡
§
University of Ghent.
Euro-Diagnostica B.V.
The Weizmann Institute of Science.
1
0.1021/jf9906984 CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/20/2000

3634 J. Agric. Food Chem., Vol. 48, No. 8, 2000
Crabbe et al.
reactivity with clostebol metabolites, one of our antisera
scores better than the commercial kit.
MATERIALS AND METHODS
Rea gen ts. Steroids, bovine serum albumin (BSA), goat anti-
rabbit IgG (whole molecule; containing 3.6 mg of specific
antibody per mL), horseradish peroxidase (HRP), ProClin 300,
Tween 20, casein, antifoam A emulsion, and tetramethylben-
zidine (TMB) were purchased from Sigma Chemical Co.
(Bornem, Belgium). Fluoxymesterone and 17R-19-nortestoster-
one were purchased from Steraloids (Wilton, NH). Organic
solvents were of analytical reagent grade. ‘HiPerSolv for
HPLC’ water, methanol, and acetonitrile were from BDH
Laboratory Supplies (Poole, England). Helix pomatia juice was
purchased from Boehringer Mannheim (Germany). Clostebol
was obtained from Alltech (Applied Science Laboratories, State
College, PA), and CLAD was a gift from Dr. L. Leyssens (Dr.
Willems-Instituut, Diepenbeek, Belgium). Epi-clostebol was
generously donated by RIVM (The Netherlands). The metabo-
lites 4-chloro-androstene-3R-ol-17-one and 4-chloro-andros-
tene-3R,17â-diol were kindly donated by Dr. G. Pieraccini
(CISM, Florence, Italy). Rabbit anti mouse IgG was obtained
from DAKO (Glostrup, Denmark).
Solu tion s. (a) Assay buffer: Phosphate-buffered saline
(
PBS), pH 7.5 (0.2 g of KH
of NaCl, 0.2 g of KCl/L of water) containing 0.05% Tween 20;
.1% BSA. (b) Coating buffer (Tijssen, 1985): 50 mM carbonate
buffer, pH 9.6 (1.59 g of Na CO ; 2.93 g of NaHCO /L of water).
c) Blocking solution: PBS containing 2% casein. (d) Wash
2
PO
4
, 1.44 g of Na
2
HPO
4
‚2H
2
O, 8 g
0
2
3
3
(
solution: PBS with 0.005% Tween 20, 0.1% casein, and 0.004%
Antifoam A Emulsion. (e) Substrate solutions A and B:
prepared as described by D u¨ rsch and Meyer (1992) with little
modification. Solution A (pH 5) contained 1.00 g of urea
hydrogen peroxide, 18.00 g of Na
acid‚H O per liter water. Solution B (pH 2.4) contained 500
mg of TMB, 40 mL of dimethyl sulfoxide, 10.30 g of citric acid‚
O, and 960 mL of water. Both solutions were stored
2
HPO
4
2
‚2H O, 10.30 g of citric
2
H
2
separately and protected from light at 4 °C. Before the assay,
equal volumes of A and B were mixed. To all buffers and
solutions, 0.05% ProClin was added as a preservative.
Disp osa bles a n d E q u ip m en t . Flat-bottomed microtiter
plates (96 wells, Nunc Maxisorp) were obtained from Life
Technologies (Merelbeke, Belgium). Optical activities were
measured with a Micro Plate Reader, model MPR-A4, from
Eurogenetics (Tessenderlo, Belgium).
PD-10 columns (Sephadex G-25 M) and the Mab-Trap GII
kit for total IgG isolation was purchased from Pharmacia
Biotech (Roosendaal, The Netherlands). The UltraFree-15
centrifugal filter device (Biomax-50 NMWL) was purchased
from Millipore (Brussels, Belgium). The commercial ELISA kit
was supplied by Laboratoire d’Hormonologie (Marloie, Bel-
gium).
P r ep a r a tion of 3-Ca r boxym eth yl-oxim e-CLAD. The
required 3-carboxylmethyl-oxime-CLAD (CMO-CLAD) was
prepared in three steps. In the first step, clostebol acetate (220
mg) was dissolved in absolute ethanol (2 mL). Carboxymethox-
ylamine hemihydrochloride (263.2 mg) and pyridine (144 µL)
were added, and the reaction was stirred overnight at room
temperature. Examination of the reaction mixture by TLC in
the solvent system chloroform:methanol:acetic acid (94.7:5:0.3)
F igu r e 1. Structures of clostebol acetate and its main
metabolites.
rum) generated through position 3 and of polyclonal
antibodies (clostebol antiserum) elicited through posi-
tion 17 of clostebol. Using CLAD as the reference
standard for cross-reaction studies, our data indicate
that the polyclonal antibodies elicited through the 17
position react preferentially with clostebol acetate, epi-
clostebol, and clostebol. On the other hand, the mono-
clonal and polyclonal antibodies generated through
position 3 do not recognize clostebol acetate; they cross-
react only to a low extent with clostebol, while only the
monoclonal antibodies do recognize epi-clostebol rela-
tively well and 4-chloro-androstene-3R-ol-17-one to a low
extent.
showed a single product of R ) 0.46.
f
In the second step, the reaction mixture containing the
product 3-carboxymethyl-oxime-clostebol acetate was hydro-
lyzed with acid (2 N HCl, 2 mL) for 3 h at 80 °C. The reaction
mixture was then cooled, and the ethanol was evaporated. The
formed precipitate was filtered and washed with water until
neutral.
In the third step of the synthesis, the crude products were
oxidized with J ones reagent. For this purpose, the crude
products (250 mg) were dissolved in acetone (20 mL) and cooled
Until now only one paper (Walshe et al., 1998) has
reported an enzyme-linked immunosorbent assay
(
ELISA) for the screening of clostebol metabolites, using
a commercial kit for the analysis of urine samples. This
paper is the first to describe the actual development of
antibodies, their characteristics, and their application
in ELISA. Our data indicate that with regard to cross-
3
to 0 °C. J ones reagent (prepared by adding 26.72 g of CrO to
23 mL of concentrated H SO4 and diluted to 100 mL with
2
double distilled water) (290 µL) was then added dropwise. The

Specificity of Antibodies Generated Against Different Derivatives of Clostebol Acetate
J. Agric. Food Chem., Vol. 48, No. 8, 2000 3635
reaction was continued for 5 min and stopped by the addition
of water (10 mL). The acetone in the reaction mixture was
evaporated under nitrogen, and the reaction mixture was
extracted into chloroform. The organic layer was washed,
dried, and evaporated. The residue was chromatographed on
silica gel 60. Elution with chloroform:methanol (95:5) gave the
desired product 3-carboxymethyl-oxime-CLAD in a very low
yield. This product was rechromatographed on silica gel 60
using chloroform:methanol (95:5) as a solvent system. 3-Car-
boxymethyl-oxime-CLAD (CMO-CLAD) showed one spot of
R ) 0.13 on TLC in the solvent system chloroform:methanol:
f
acetic acid (94.7:5:0.3).
1
The H NMR analysis of this compound in dimethyl sulfox-
ide showed the following signals: δ 0.994 (3-H, singlet, C-18
CH
O-CH
3
, 1.268 (3-H, singlet, C-19 CH
-COOH).
3
) and 4.76 (2-H, singlet,
2
Con ju ga tion of 3-Ca r boxym eth yl-oxim e-CLAD to Ma c-
r om olecu les. CMO-CLAD was conjugated to BSA and HRP
via a two-step reaction. In the first step of the synthesis, the
CMO-CLAD (3.15 mg) was dissolved in dry dioxane (200 µL),
and N-hydroxysuccinimide (1.7 mg) and carbodiimide (3.5 mg)
were added to the reaction mixture. After an overnight reaction
at room temperature, urea was formed and the supernatant
analyzed by TLC in a solvent system of chloroform:methanol:
acetic acid (74.5:25:0.5) for the presence of the active N-
hydroxysuccinimide ester derivative of CMO-CLAD. A com-
f
pound with an R of 0.95 was obtained. The N-hydroxysuccin-
imide active ester of CMO-CLAD in dioxane (200 µL) was
then used in the next step without further purification. BSA
3
(10 mg) was dissolved in 1 mL of 0.13 M NaHCO (pH 8.5)
followed by dropwise addition of the active ester (200 µL). The
reaction mixture was stirred for 2.5 h at room temperature,
dialyzed overnight against phosphate-buffered saline (PBS),
pH 7.4 at 4 °C, and stored at -20 °C until use.
The reaction scheme is shown in Figure 2. UV analysis
according to the method of Erlanger (Erlanger et al., 1957)
indicated that the conjugate contained 32.8 mol of CLAD per
mole of BSA.
F igu r e 2. Scheme of synthesis of 3-CMO-CLAD-BSA.
For the preparation of a conjugate of HRP to CMO-CLAD,
HRP (4.8 mg) was dissolved in 1 mL of 0.13 M NaHCO
3
(pH
continuous stirring. Subsequently, the mixture was added
dropwise to a solution of BSA in 0.1 M phosphate buffer (pH
8
.5) and 50 µL of the N-hydroxysuccinimide ester derivative
of CMO-CLAD in dioxane was added. The reaction mixture
was stirred for 3 h at room temperature, dialyzed twice against
PBS, pH 8, followed by PBS, pH 7.2. UV-analysis (Erlanger
et al., 1957) indicated that the HRP conjugate contained 2 mol
of CLAD per mole of HRP. It was stored at -20 °C until use.
P r ep a r a tion of Clostebol-17-h em isu ccin a te. Clostebol
acetate (20 mg) was dissolved in a 20 mL mixture of acetone:1
N HCl (9:1, v/v) and left to react for 15 h at 75 °C under reflux
and continuous stirring. The reaction mixture was then made
8
) and left overnight stirring at room temperature. The pH
was checked at regular intervals and adjusted when necessary.
Finally, the solution was diluted with 0.05 M NaHCO (pH 8)
3
up to a final percentage of 10% DMF and dialyzed, concen-
trated, and lyophilized.
The hapten/carrier ratio was determined spectrophotometri-
cally according to the method of Erlanger (Erlanger et al.,
1
957); results showed a molar ratio of 8.59 mol of clostebol-
hemisuccinate per mole of BSA.
2 3
alkaline with 1 M Na CO and extracted with diethyl ether.
By examination on HPTLC (silica gel 60 with F254) in the
solvent system hexane:ethyl acetate (2:1; v/v), no clostebol
(ii) Clostebol-17-hemisuccinate-HRP. For conjugation of
clostebol-hemisuccinate to HRP, the same reaction procedure
was followed as that for conjugation with BSA, except that
the reaction mixture with HRP was left stirring at 4 °C. Finally
the solution was dialyzed, preconcentrated by ultracentrifu-
gation, and lyophilized. The molar hapten/carrier ratio was
determined by UV (Erlanger et al., 1957). Analysis showed a
molar ratio of CLT-HS/HRP of 1.8.
P r ep a r a tion of Mon oclon a l An tibod ies. The CLAD-
BSA conjugate was used as to immunize female CD2 mice (age:
2 months; 50 µg/animal). Subsequently, three booster injec-
tions were given using the CLAD-BSA conjugate in incom-
plete Freund’s adjuvant. After 2 months of immunization, the
antibody titer was checked using rabbit anti-mouse IgG coated
plates and CLAD-HRP as a label. Three months after the
initial immunization, the spleen cells of the mouse with high
titer of antibodies to CLAD were fused with a mouse myeloma
cell line (the non-Ig producer mouse myeloma cell line NSO/1
was kindly provided by Dr. C. Milstein, MRC Laboratory of
Molecular Biology, Cambridge, UK) using the hybridoma
technique of K o¨ hler and Milstein (1976) with some modifica-
tions introduced (Kohen and Lichter, 1986). The culture
supernatants of growing hybridomas were screened for anti-
body activity using rabbit anti-mouse IgG coated plates and
acetate (R
hydrolyzed into clostebol (R
f
) 0.57) was left; it had been almost completely
f
) 0.18).
Clostebol (100 mg) was redissolved in pyridine (3 mL) with
an excess of succinic acid anhydride (1 g) and stirred at room
temperature for 5 days. After extraction with diethyl ether, it
was purified over a silica gel G 60 column and eluted with
dichloromethane and 10% acetonitrile. Examination of the
reaction product by HPTLC (silica gel plate with F254 indica-
tor; mobile phase ) dichloromethane with 10% acetonitrile)
showed a spot with R ) 0.41. The structures of clostebol and
f
1
clostebol-17-hemisuccinate were confirmed by H NMR analy-
sis. Comparison of the spectra clearly revealed the presence
of the succinylic protons (CH
.55-2.75).
P r ep a r a tion of Ma cr om olecu la r Con ju ga tes of Closte-
2 2
CH ) as a complex multiplet (δ:
2
bol Hem isu ccin a te. (i) Clostebol-17-hemisuccinate-BSA. The
clostebol-17-hemisuccinate (CLT-HS, 9.8 mg) was dissolved in
0
.5 mL of dimethylformamide (DMF). A solution of N-hydrox-
ysuccinimide (6 mg) in 0.5 mL of DMF was added, followed
by a solution of dicyclohexylcarbodiimide (13.4 mg) in 1 mL of
DMF, while the mixture was cooled on ice. Afterward, the
reaction mixture was left overnight at room temperature under

3636 J. Agric. Food Chem., Vol. 48, No. 8, 2000
Crabbe et al.
CLAD-HRP as a label. Suitable hybridomas were selected,
cloned, and propagated in vitro as culture supernatant or in
vivo in pristane primed mice as ascites. Purified immunoglo-
bulins were isolated from ascites by chromatography on
Sepharose-Protein A as described previously (Strasburger and
Kohen, 1990).
Ta ble 1. Ch a r a cter istics of P olyclon a l An tiser a in ELISA
Clostebol
CLAD
antiserum
antiserum
FD (PAb)
FD (IgG)
1:500 000
1:1 500 000
1:100 000
FD (HRP-conjugate) CLT-17-HS-HRP
:400 000
CLAD-3-CMO-HRP
1:25 000
2 h
30 min
0.36
Gen er a tion of P olyclon a l An tibod ies. Rabbits were
immunized by intradermic multisite injection. First a stock
1
incubation time
coloring time
ED 50 (ppb)
1 h
30 min
0.09
-
1
solution of the immunogen (1 mg mL ) in NaCl 0.9% was
prepared. For the first injection, 250 µL of stock solution was
mixed with 250 µL of NaCl (0.9%) and 500 µL of Freund’s
complete adjuvant (total injection volume: 1 mL). Boosts were
given every 28 days with a mixture of 150 µL of stock solution,
Bo (mean)
standard deviation
0.987 (n ) 16)
0.054
1.232 (n ) 14)
0.041
detection limit (ppb) 0.03
in assay buffer)
0.08
3
50 µL of NaCl (0.9%), and 500 µL of incomplete Freund’s
(
adjuvant. Every 10 days after injection, blood samples were
collected and the sera were stored at -20 °C until use. The
various sera were then tested for titer, sensitivity, and
specificity using ELISA techniques described below.
possible to elicit the formation of antibodies with
enhanced specificity toward selected parts of the steroid
molecule (Bauminger et al., 1974; Kohen et al., 1975).
Accordingly, in this work we investigated the influence
of the site of conjugation to clostebol and to CLAD on
the specificity of antibodies generated. Being the most
important metabolite, CLAD was chosen as hapten for
conjugation at position 3 This first mode of conjugation
was intended to preserve the substituents of ring D,
more specific the 17-side chain, as prime determinants
in immuno recognition. The second conjugation through
the 17-position, using clostebolswith its free hydroxyl
groupsas a hapten, is a convenient way of coupling
while leaving the A ring fully exposed for antibody
recognition.
Since clostebol was not immediately available as a
reference standard, clostebol acetate had to be hydro-
lyzed as a starting material for the required synthesis.
Basic hydrolysis of clostebol acetate led mainly to
decomposition products. No further attempts were made
using alkaline hydrolysis. Using previously described
methods for acid hydrolysis (Leyssens et al., 1994; Le
Bizec et al., 1996), unsatisfactory results were obtained.
On the other hand, modification of the method of Le
Bizec (Le Bizec et al., 1996) resulted in a more efficient
hydrolysis of clostebol acetate in acidic conditions.
Develop m en t of a Com p etitive ELISA. Differences
in characteristics (titer, sensitivity, and specificity) were
examined by use of a direct competitive ELISA with a
second-antibody technique.
Isola tion of Tota l IgG fr om th e An ti-Clostebol An ti-
ser u m . The immunoglobulin G (IgG) fraction of the antiserum
was isolated with the aid of recombinant protein G (Mab-Trap
GII kit) according to the procedure described by the manu-
facturer and then applied to a PD-10 column in order to
exchange the buffer into PBS and to remove remaining salts.
Before storage at -20 °C, the solution was preconcentrated
using an UltraFree device for ultrafiltration. The titer of the
IgG solution was determined with a checkerbord titration in
combination with clostebol-17-hemisuccinate-HRP.
Develop m en t of a Com p etitive ELISA w ith Dou ble-
An tibod y Tech n iqu e. A microtiter plate was coated with
-
1
GAR (100 µL per well) at a concentration of 5 µg mL in
coating buffer and left for overnight incubation at 4 °C.
After decanting the solution, the plate was blocked with 2%
casein in PBS (300 µL per well) for 30 min at room tempera-
ture on a shaker. The plate was washed four times, and
reagents were added in the following order: 50 µL of standard
-1
solutions (0.02-0.05-0.1-0.2-0.5-1-2-5 ng mL ) of CLAD
in assay buffer or sample solutions, 25 µL of HRP label
(
CLAD-HRP or CLT-HS-HRP), and 25 µL of antiserum
(respectively, CLAD antiserum or clostebol antiserum). After
incubation for 2 h at room temperature, the plate was washed
four times. Substrate solution was added (100 µL per well),
and the plate was incubated for 30 min at room temperature
in the dark. The reaction was stopped by adding 100 µL of 1
2 4
M H SO to each well. The absorbance was read at 450 nm.
An tiser a Titer Deter m in a tion . In the double-antibody,
solid-phase ELISA binding studies between the antisera and
hapten-HRP label were used to determine their optimal
dilution. The optimal dilutions were those that resulted in an
absorbance of 1.0 at 450 nm, following incubation of the
substrate for 30 min at room temperature.
An tiser a Sen sitivity Deter m in a tion . Calibration curves
for CLAD (standards in assay buffer) were determined with
the two polyclonal antisera according to the procedure de-
scribed above.
The corresponding B/Bo 50% value and the detection limit,
defined as the concentration of CLAD that corresponds with
the response of ‘mean Bo - 3 SD’, were determined.
An tiser a Sp ecificity Deter m in a tion . Cross-reactivity
studies were carried out with various structurally related
compounds (relative to CLAD at 100%) at different concentra-
tions.
For testing the monoclonal antibodies, a rabbit anti-mouse
IgG was used as second antibody in the ELISA instead of the
GAR used for the polyclonal antisera.
The use of second antibodies gives better reproduc-
ibility and sensitivity and a much smaller amount of
the precious antiserum is consumed. It has been shown
that pH 9.6 is optimal for the passive adsorption of
antibodies to microtiter plates (Butler et al., 1993).
According to Tijssen (1985), the optimal concentration
of antigens and antibodies for coating is usually between
-
1
1 and 10 µg mL
.
Two coating concentrations, 5 and 10 µg mL-1, were
compared. Both gave similar results in the ELISA.
Therefore, the microtiter plates were coated with sec-
-1
ondary antibodies at a concentration of 5 µg mL
overnight at 4 °C.
An tiser u m Titer a n d Sen sitivity. (i) Monoclonal
Antibodies to CLAD. From the fusion experiment, 2000
hybridomas were screened, and one clone (#14H2)
secreting monoclonal antibodies to CLAD with high
binding activity to CLAD-HRP, specificity (Table 2), and
sensitivity was selected. This clone belonged to the IgG1
class and was propagated in vitro and in vivo. Using
colorimetry as an end point, a sensitivity of 0.8 ng of
The percentage cross-reactivity was determined as the ratio
of the amount of cross-reacting material that yields 50%
inhibition of binding with the amount of standard giving the
same inhibition.
RESULTS AND DISCUSSION
-
1
CLAD mL was obtained.
Syn th esis of Im m u n ogen s. It has been reported
that by judicious choice of the site of conjugation it is
(ii) Polyclonal Antibodies. Usable polyclonal antisera
were obtained with both conjugates. In general, the

Specificity of Antibodies Generated Against Different Derivatives of Clostebol Acetate
J. Agric. Food Chem., Vol. 48, No. 8, 2000 3637
Ta ble 2. Cr oss-Rea ctivity w ith Str u ctu r a l An a logu es
%
Cross-reactivity
Clostebol
CLAD Monoclonal
antiserum antiserum antiserum
CLAD
100
65
50
100
0.01
0.4
2
100
<0.01
35
clostebol acetate
epi-clostebol
clostebol
39
1
4
4
-chloroandrostene-3R-ol-17-one
0.7
6
-chloro-androstene-3R,17â-diol
0.5
0.7
0.5
0.9
0.2
0.2
0.2
0.2
1
0.2
0.05
<0.01
0.2
0
0
testosterone
epi-testosterone
testosterone-17-sulfate
epi-testosterone-sulfate
testosterone-17-glucuronide
<0.01
0.3
0
1
9-nortestosterone
epi-19-nortestosterone
7R-Me-testosterone
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.001
0.9
1
0.9
0.8
<0.001
0.02
<0.01
1
0.1
0.1
<0.01
<0.01
Progesterone
R-estradiol
â-estradiol
Androsterone
Androstene-3,17-dione
1
9-norandrostene-3,17-dione
0.07
<0.001
<0.001
7
<0.03
<0.001
0
Dehydroepiandrosterone
Dehydroepiandrosterone-sulfate
F igu r e 3. Sigmo ¨ı dal calibration curves of CLAD with the
polyclonal antisera.
5
5
5
5
R-androstane-3,17-dione
â-androstane-3,17-dione
R-androstane-3â-ol-17-one
R-androstane-3R-ol-17-one
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
free 4-Cl function of the CLT-17-HS hapten and thus
show more cross-reactivity toward the metabolites of
clostebol.
Boldenone
Me-boldenone
Fluoxymesterone
Stanozolol
The fact that the clostebol antiserum cross-reacts to
a different extent with clostebolacetate, clostebol, and
epi-clostebol demonstrates that the 17 position used for
conjugation is not completely deprived of its antigenic
determinant. A possible explanation for why the cross-
reactivity with clostebol acetate is higher than for
clostebol could be the structural similarity between the
hemisuccinate bridge of the immunogen and the acetate
group on clostebol acetate.
The CLAD antiserum only reacts weakly with the
metabolites and shows a high specificity toward CLAD.
Using the 3-conjugation site, possible steric hindrance
could occur, so the 4-Cl function is not recognizable as
a strong antigenic determinant for the generated anti-
bodies.
If we finally compare the cross-reactivity of our
antibodies with those of the antibodies used in the
commercial ELISA kit, we can see some differences. The
principal cross-reactivities of these antibodies as men-
tioned in the information from the manufacturer are
CLAD (100%), clostebol (28%), clostebol acetate (120%).
The cross-reactivity for epi-clostebol, as determined at
our laboratory, was found to be 8%. The high result for
clostebol acetate is, however, of little importance since
no parent drug occurs in urine; the cross-reactivity for
epi-clostebol, on the other hand, is much more important
since it represents one of the major metabolites in urine.
In this view we can state that antibodies with the
highest cross-reactivities toward CLAD, clostebol, and
epi-clostebol are most promising when it comes to
screening of urinary metabolites and avoiding false
negative results.
titers of the antisera produced against the clostebol-17-
hemisuccinate immunogen (clostebol antiserum) were
higher than those obtained against the 3-CMO-CLAD
immunogen (CLAD antiserum).
For the CLAD antiserum, all of the following ELISA
tests were performed with the antiserum in a final
dilution of 1:100 000 in combination with a CLAD-HRP
dilution of 1:25 000.
The clostebol antiserum with the highest titer could
be used in a final dilution of 1:500 000 in combination
with the CLT-17-HS-HRP conjugate in a final dilution
of 1:400 000. The IgG fraction could be used in a final
dilution of 1:1 500 000 in combination with the HRP
conjugate in a final dilution of 1:400 000, resulting in a
similar calibration curve as with the crude antiserum.
Figure 3 represents the dose response curves obtained
for CLAD using the polyclonal CLAD and clostebol
antisera. They are characterized by an ED50 value of,
-
1
respectively, 0.36 and 0.09 ng mL , and the detection
limits, defined as ‘Bo - 3SD’, are, respectively, 0.08 and
0
-
1
.03 ng mL (Table 1).
Cr oss-Rea ctivity of th e An tiser a . The specificity
of the three antisera is shown in Table 2.
The monoclonal antibody generated through the 3
position of CLAD recognized epi-clostebol relatively well
(35%) and 4-chloro-androstene-3R-ol-17-one only to a
small extent. Little cross-reactivity with clostebol and
clostebol acetate was observed. There was also no cross-
reaction with endogenous hormones as testosterone,
estradiol, and progesterone, neither with other struc-
tural analogues nor anabolics (e.g., boldenone, flu-
oxymesterone, stanozolol). Moreover, the polyclonal
antiserum generated through the same position gave
rise to antibodies that were specifically directed toward
the 17 position and did not recognize epi-clostebol,
clostebol, or clostebol acetate well. On the other hand,
polyclonal antibodies generated through the 17 position
of clostebol are directed more specifically against the
CONCLUSIONS
The results presented in this paper confirm previous
studies (Bauminger et al., 1974; Kohen et al., 1975) that
the site of attachment of the steroid molecule to the
macromolecular carrier molecule considerably affects
the specificity of the generated antibodies. When the
steroid molecule was attached to the carrier through
position 17 and the 4-Cl function was left free for

3638 J. Agric. Food Chem., Vol. 48, No. 8, 2000
Crabbe et al.
recognition, we obtained antibodies that discriminate
more efficiently between clostebol metabolites and en-
dogenous hormones than do those produced through
conjugation at position 3, leaving a 17-keto function free.
This latter conjugation gave rise to highly specific
antibodies for metabolites with a 17-keto function (or
R-OH in case of the monoclonals) but not for metabolites
with a â-OH function or an acetate group.
The relative merit in ELISA of the two types of
antisera will also depend on the relative concentration
of potentially cross-reacting (endogenous) steroids in the
biological fluid of interest and the relative ease with
which these can be eliminated during the initial extrac-
tion and purification steps. However, in general, it can
be concluded that the polyclonal antibodies of the
clostebol antiserum have a great potential for prelimi-
nary screening of clostebol acetate and its metabolites
in urine.
Erlanger, B. F.; Borek, F.; Beiser, S. M.; Lieberman, S.
Preparation and characterization of conjugates of BSA with
testosterone and with cortisone. J . Biol. Chem. 1957, 228,
7
13-727.
Hendricks, L.; Gielen, B.; Leyssens, L.; Raus, J . Screening for
the illegal use of clostebol acetate in cattle by identification
of its urinary metabolites. Vet. Rec. 1994, 134, 192-193.
Kohen, F.; Bauminger, S.; Lindner, H. R. Preparation of
antigenic steroid-protein conjugates. In Steroid Immunoas-
say; Cameron, E. H. D., Hillier, S. G., Griffiths, K., Eds.;
Alpha Omega Publishing: Cardiff, Wales, U.K., 1975; pp
1
1-31.
Kohen, F.; Lichter, S. Monoclonal antibodies to steroid hor-
mones. In Monoclonal antibodies: Basic Principles, experi-
mental and Clinical Applications in endocrinology; Forti,
G., Lipsett, M. B., Serio, M., Eds.; Raven Press: New York,
1
986; pp 87-95.
K o¨ hler, G.; Milstein, C. Derivation of specific antibody-produc-
ing tissue culture and tumor lines by cell fusion. Eur. J .
Immunol. 1976, 6, 511-522.
Le Bizec, B.; Montrade, M.-P.; Monteau, F.; Andr e´ , F. Detection
and identification of anabolic steroids in bovine urine by gas
chromatography- mass spectrometry. Anal. Chim. Acta
Further studies of their application for analysis of
urine samples of food-producing animals are being
conducted at the moment.
1
993, 275, 123-133.
Le Bizec, B. Th e` se de doctorat: Utilisation de la spectrometrie
de masse pour le controle de l’usage des stero ı¨ des anab-
olisants en e´ levage, Nantes, France, 1996.
Le Bizec, B.; Montrade, M.-P.; Monteau F.; Andr e´ , F. Metabo-
lites of some anabolic steroids in cattle after oral adminis-
tration. In Proceedings of the EuroResidue III Conference;
Haagsma, N., Ruiter, A., Eds.; University of Utrecht,
Faculty of Veterinary Medicine, Utrecht, The Netherlands,
ABBREVIATIONS USED
CLAD, 4-chloro-androstenedione; CLTA, clostebol
acetate; CLT, clostebol; BSA, bovine serum albumin;
ELISA, enzyme-linked immunosorbent assay; FD, final
dilution; GAR, goat anti-rabbit; HRP, horseradish per-
oxidase; OD, optical density; PBS, phosphate-buffered
saline; Bo, the value of the OD (in the ELISA) obtained
when the tracer is incubated alone with the antibodies;
B, the value of the OD (in the ELISA) obtained when
the tracer is incubated with the antibodies and the
antigen (CLAD); SD, standard deviation; DMF, di-
methylformamide; HS, hemisuccinate; CMO, carboxy-
methyl-oxime.
1
996; pp 248-252.
Le Bizec, B.; Montrade M.-P.; Monteau, F.; Gaudin, I.; Andr e´ ,
F. 4-Chlorotestosterone acetate metabolites in cattle after
intramuscular and oral administration. Clin. Chem. 1998,
4
4 (5), 973-984.
Leyssens, L.; Royackers, E.; Gielen, B.; Missotten, M.; Schoofs,
J .; Czech, J .; Noben, J . P.; Hendricks, L.; Raus, J . Metabo-
lites of 4-chlorotestosterone acetate in cattle urine as
diagnostic markers for its illegal use. J . Chromatogr. B
1
994, 654, 43-54.
ACKNOWLEDGMENT
Strasburger, C. J .; Kohen, F. Biotinylated probes in immu-
noassay. In Methods in Enzymology: Avidin-Biotin Tech-
nology; Wilchek, M., Bayer, E., Eds.; Academic Press: New
York, 1990; pp 481-496.
We thank Mrs. Batya Gayer and Mrs. Shoshana
Lichter for their technical assistance.
Tijssen, P. The immobilization of immunoreactants on solid
phases. In Practice and theory of enzyme immunoassays;
Burdon, R. H., Van Knippenberg, P. H., Eds.; Elsevier
Science Publishers: Amsterdam, The Netherlands, 1985.
Vanoosthuyze, K.; Daeseleire, E.; Van Overbeke, A.; Van
Peteghem, C.; Ermens, A. Survey of the Hormones Used in
Cattle Fattening Based on the Analysis of Belgian Injection
Sites. Analyst 1994, 119, 2655-2658.
Van Puymbroeck, M.; Royackers, E.; Witkamp, R. F.; Leyssens,
L.; Van Miert, A. S.; Gelan, J .; Vanderzande D.; Raus, J .
Microsomal preparations as an aid for the characterisation
of metabolites of illegal growth promoters. In Proceedings
of the EuroResidue III Conference; Haagsma, N., Ruiter, A.,
Eds.; University of Utrecht, Faculty of Veterinary Medicine,
Utrecht, The Netherlands, May 8-16, 1996; pp 808-813.
Walshe, M.; O’Keeffe, M.; Le Bizec, B. Studies on the deter-
mination of chlorotestosterone and its metabolites in bovine
urine. Analyst 1998, 123, 2687-2691.
LITERATURE CITED
Andr e´ , F.; Le Bizec, B.; Montrade, M.-P.; Maume, D.; Monteau,
F.; Marchand, P. Developments in residue assay and
metabolism study of growth-promoters by mass spectromet-
ric analysis. Analyst 1994, 119, 2529-2535.
Bauminger, S.; Kohen, F.; Lindner, H. R. Steroids as haptens:
optimal design of antigens for the formation of antibodies
to steroid hormones. J . Steroid Biochem. 1974, 5, 739-747.
Butler, J . E.; Ni, Li; Brown, W. R.; J oshi, K. S.; Chang, J .;
Rosenberg, B.; Voss, E. W., J r. The immunochemistry of
sandwich ELISAssVI. Greater than 90% of monoclonal and
7
5% of polyclonal anti-fluorescyl capture antibodies (Cabs)
are denatured by passive adsorption. Mol. Immunol. 1993,
0 (13), 1165-1175.
3
Debruyckere, G.; De Sagher, R.; Van Peteghem, C. Clostebol-
positive urine after consumption of contaminated meat.
Clin. Chem. 1992, 38 (9), 1869-1873.
Debruyckere, G.; De Sagher, R.; Van Peteghem, C.; Van
Vyncht, G.; Maghuin-Rogister, G.; De Pauw, E. Gas chro-
matographic-mass spectrometric confirmation of a clostebol
metabolite in urine. Anal. Chim. Acta 1994, 291, 155-160.
D u¨ rsch, I.; Meyer, H. H. D. A multi-residue enzyme immu-
noassay for screening illegally used â-agonists. Food Agric.
Immunol. 1992, 4, 211-218.
Received for review J une 25, 1999. Revised manuscript
received May 19, 2000. Accepted May 24, 2000. This work was
supported by the Commission of the European Communities
in the framework of the European Community Research
Program (EU project SMT4-CT96-2092) with coordinator Dr.
M. Salden (Euro-Diagnostica, The Netherlands); the authors
acknowledge financial support from this program.
EEC Directive 86/469, Off. J . Eur. Commun. 26/09/86 1986,
No. L275/36.
J F9906984