AN ULTRASENSITIVE CAPTURE ELISA FOR DETECTION OF FASCIOLA

HEPATICA COPROANTIGENS IN SHEEP AND CATTLE USING A NEW

MONOCLONAL ANTIBODY (MM3)
Author(s): Mercedes Mezo, Marta Gonzlez-Warleta, Carmen Carro, and Florencio M. Ubeira
Source: Journal of Parasitology, 90(4):845-852. 2004.
Published By: American Society of Parasitologists
DOI: http://dx.doi.org/10.1645/GE-192R
URL: http://www.bioone.org/doi/full/10.1645/GE-192R

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and
environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published
by nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOnes Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.
 J. Parasitol., 90(4), 2004, pp. 845852
q American Society of Parasitologists 2004

AN ULTRASENSITIVE CAPTURE ELISA FOR DETECTION OF FASCIOLA HEPATICA

COPROANTIGENS IN SHEEP AND CATTLE USING A NEW MONOCLONAL
ANTIBODY (MM3)
Mercedes Mezo, Marta Gonzalez-Warleta, Carmen Carro, and Florencio M. Ubeira*
Centro de Investigaciones Agrarias, Mabegondo, P.O. Box 10, 15080 A Coruna, Spain. e-mail: mpubeira@usc.es

ABSTRACT: A capture enzyme-linked immunosorbent assay (ELISA) using a new monoclonal antibody (mAb MM3) is reported
for the detection of Fasciola hepatica excretorysecretory antigens (ESAs) in feces of infected hosts. The mAb MM3 was
produced by immunization of mice with a 7- to 40-kDa purified and O-deglycosylated fraction of F. hepatica ESAs, which has
previously been shown to be specific for the parasite. The specificity and sensitivity of the MM3 capture ELISA were assessed
using feces from sheep and cattle. Sheep feces were obtained from a fluke-free herd (with most animals harboring other nematodes
and cestodes), from lambs experimentally infected with 540 F. hepatica metacercariae and in some cases treated with tricla-
bendazole at 14 wk postinfection (PI), and from uninfected control lambs. Cattle feces were collected at the slaughterhouse from
adult cows naturally infected with known numbers of flukes (from 1 to 154) or free of F. hepatica infection (though in most
cases harboring other helminths). The MM3 capture ELISA assay had detection limits of 0.3 (sheep) and 0.6 (cattle) ng of F.
hepatica ESA per milliliter of fecal supernatant. The assay detected 100% of sheep with 1 fluke, 100% of cattle with 2 flukes,
and 2 of 7 cattle with 1 fluke. The false-negative animals (5/7) were probably not detected because the F. hepatica individuals
in these animals were immature (511 mm in length). As expected, coproantigen concentration correlated positively (r 5 0.889;
P , 0.001) with parasite burden and negatively (r 5 0.712; P , 0.01) with the time after infection at which coproantigen was
first detected. Nevertheless, even in animals with low fluke burdens (136 parasites), the first detection of F. hepaticaspecific
coproantigens by the MM3 capture ELISA preceded the first detection in egg count by 15 wk. In all sheep that were experi-
mentally infected and then untreated, coproantigen remained detectable until at least 18 wk PI, whereas in sheep that were
experimentally infected and then flukicide treated, coproantigen became undetectable from 1 to 3 wk after treatment. None of
the fecal samples from sheep or cattle negative for fascioliasis but naturally infected with other parasites including Dicroelium
dendriticum showed reactivity in the MM3 capture ELISA. These results indicate that this assay is a reliable and ultrasensitive
method for detecting subnanogram amounts of F. hepatica antigens in feces from sheep and cattle, facilitating early diagnosis.

Fascioliasis is a zoonotic disease caused by Fasciola spp.,
trematode parasites that infect a wide range of mammals
throughout the world. The prevalence of infection is often high
in sheep and cattle, probably because of close contact with in-
fective metacercariae on pasture and a high sensitivity to infec-
tion. Consequently, fascioliasis causes major economic losses
in many countries.
One of the main problems in the control of this disease is
the lack of sensitive and convenient tests for the diagnosis of
Fasciola spp. infection in large herds under field conditions and
for monitoring the efficacy of flukicide treatments. Diagnosis
of animal fascioliasis is largely based on microscopic demon-
stration of parasite ova in feces, but Fasciola spp. egg shedding
is intermittent and irregular and does not begin until 1012 wk
postinfection (PI) or later (Zimmerman et al., 1982). Besides
its low sensitivity, which can result in underdiagnosis of weak
infections (Conceicao et al., 2002), coproscopy is a laborious
technique requiring each sample to be examined individually
by trained personnel so that it is not suitable for analyzing large
herds.
Immunological techniques using selected excretorysecreto-
ry antigens (ESA) from Fasciola spp. allow the detection of
specific antiparasite circulating antibodies within 35 wk of the
infection (Mezo et al., 2003) and can readily be automated.
Nevertheless, obtaining sera is difficult with large herds, and
tests of this type are of limited diagnostic value in endemic
areas because antibody titers remain at high levels even when
animals have been successfully treated (Sanchez-Andrade et al.,
Received 23 July 2003; revised 4 November 2003; accepted 4 No-
vember 2003.
* Laboratorio de Parasitologa, Facultad de Farmacia, Universidad de
Santiago de Compostela, 15782 Santiago de Compostela, Spain.
To whom correspondence should be addressed.
2001). Alternatively, active infections by Fasciola spp. can be
diagnosed using immunological tests for the detection of met-
abolic products of flukes located in liver bile ducts, which are
released and passed into feces. Together with the fact that large-
scale fecal sampling is evidently easier than large-scale blood
sampling, this meant that research has focused on the devel-
opment of enzyme-linked immunosorbent assay (ELISA) tests
for the detection of parasite antigens in host stools, i.e., co-
proantigens.
Using capture ELISA tests based on polyclonal antibodies
raised against whole ESAs, it has been proved possible to detect
F. hepatica coproantigens in experimentally infected sheep (Al-
mazan et al., 2001) and rats (Paz-Silva et al., 2002). The spec-
ificity of these tests was not evaluated in depth, but it can be
expected to be low because some polypeptides present in whole
F. hepatica ESA preparations are shared by other helminths
commonly present in the digestive tract of grazing ruminants.
In addition, antisera produced in this way show considerable
among-batch variability so that results may vary substantially
from batch to batch. In contrast, monoclonal antibody technol-
ogy produces reagents with well-defined affinity and specificity
and enables production of large quantities with homogeneous
quality. Several monoclonal antibodies have been raised against
F. hepatica ESAs and somatic antigens (Hanna and Trudgett,
1983; Hanna et al., 1988; Solano et al., 1991), but to our knowl-
edge, only 2 have been reported to date as useful for the de-
tection of coproantigens in humans and animals with fascioli-
asis. One monoclonal antibody (mAb) named F10 (Abdel-Rah-
man et al., 1998, 1999) was reported to specifically identify
feces from calves with 10 or more flukes, whereas mAb ES78
(Espino and Finlay, 1994; Dumenigo et al., 1996, 2000; Espino
and Dumenigo, 2003) specifically identified 100% of feces from
calves with 5 or more flukes. However, ES78 was not tested in
animals with very low parasite burdens, i.e., less than 5 flukes.

845
846 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 4, AUGUST 2004
In this report, we describe a capture ELISA assay using a new
monoclonal antibody (MM3, IgG1/k) that detects specific F.
hepatica ESA coproantigens in feces from sheep with 1 fluke
and cattle with 12 flukes. This MM3 capture ELISA is much
more sensitive than previously described assays.
MATERIALS AND METHODS
Fasciola hepatica ESAs
Fasciola hepatica ESAs were obtained as described previously
(Mezo et al., 2003). In brief, live adult flukes collected from bile ducts
of naturally infected cows were washed, first in sterile saline solution
containing antibiotics (100 IU/ml penicillin and 100 mg streptomycin)
and glucose (2 mg/ml) at 38 C and then in Roswell Park Memorial
Institute (RPMI) cell culture medium (RPMI 1640 supplemented with
20 mM N-2-hydroxyethylpiperazine-N9-2-ethanesulfonic acid, 0.3 g/L
L-glutamine, 2 g/L sodium bicarbonate, and antibiotics) at 38 C under
5% CO2 in air. Flukes were then transferred to 75-cm2 tissue culture
flasks (Iwaki, Sciteck Div. Asahi Techno Glass, Funabashi City, Chiba,
Japan) and maintained in culture medium (3 ml/fluke) at 38 C under
5% CO2 in air. After 24 hr of incubation, the medium containing ESA
was removed and centrifuged at 10,000 g for 20 min at 4 C in the
presence of protease inhibitors. The supernatant was passed through a
0.45-mm pore filter disk and then concentrated using an Amicon 8050
ultrafiltration cell (Amicon, Inc., Beverly, Massachusetts) equipped with
a YM10 membrane (10,000 molecular weight cutoff), dialyzed against
phosphate-buffered saline (PBS) (10 mM sodium phosphate buffer, 150
mM NaCl, pH 7.4), sterilized by filtration, and stored at 280 C until
required. Protein concentration was measured by the bicinchonionic
acid method (Pierce Biotechnology Inc., Rockford, Illinois).
A fraction containing F. hepaticaspecific antigens was purified from
whole F. hepatica ESA by size exclusion chromatography using an fast
protein liquid chromatographyhigh-performance liquid chromatogra-
phy system (AKTA Basic 10, Amersham Biosciences Europe GmbH,
Barcelona, Spain) on a Superdex 75 HR 10/30 column (Amersham Bio-
sciences), as described previously (Mezo et al., 2003). The column was
calibrated with a mixture of knownmolecular weight proteins (Gel Fil-
tration LMW Calibration Kit, Amersham Biosciences). The fractions in
the fourth peak (peak IV), which eluted between chymotrypsinogen A
(25 kDa) and ribonuclease A (13 kDa), were pooled, concentrated by
ultrafiltration using Centricon YM-10 centrifugal filter devices (Milli-
pore Corporation, Bedford, Massachusetts), and stored at 220 C until
used. Some of these antigens were O-deglycosylated with 0.01 M
NaOH as described previously (Lorenzo et al., 2000) and used to im-
munize mice as indicated below.
Monoclonal antibody production
The IgG1/k monoclonal antibody MM3 was obtained by fusion with
P3-X63-Ag8.653 myeloma cells of spleen cells from BALB/c mice hy-
perimmunized with the F. hepatica O-deglycosylated antigens con-
tained in peak IV. The immunization protocol consisted of 3 intraperi-
toneal injections, 3 wk apart, of 50 mg of peak-IV proteins (for the first
injection emulsified in Freunds complete adjuvant) and a final intra-
venous booster of 75 mg of protein administered 3 days before killing
and spleen cell fusion. Positive hybridomas producing antibodies that
recognized F. hepatica but not Dicrocoelium dendriticum antigens were
obtained and selected according to the procedure described previously
(Estevez et al., 1994; Iglesias et al., 1997). MM3 hybridoma cells were
intraperitoneally injected into pristan-primed BALB/c mice, and the
antiF. hepatica IgG1/k antibodies were purified from the ascitic fluid
by affinity chromatography on a protein G column (5 ml HiTrap Protein
G, Amersham Biosciences), according to the manufacturers protocol.
Rabbit polyclonal antibodies
Polyclonal IgG antibodies were raised by hyperimmunization of 2
New Zealand White rabbits with 4 subcutaneous injections, at 3-wk
intervals, of 200 mg of F. hepatica ESAs emulsified in either Freunds
complete adjuvant (for the first injection) or Freunds incomplete ad-
juvant (for the 3 booster injections). A week after the last immunization,
the animals were bled, and the serum was obtained. The IgG fraction
was purified from serum on a protein G column, as above.
Affinity chromatography with the monoclonal antibody MM3
The F. hepatica ESAs recognized by the MM3 mAb were purified
by affinity chromatography with the mAb immobilized on a HiTrap
NSH-activated HP column (Amersham Biosciences), according to the
suppliers instructions. Fasciola hepatica ESAs (1.2 mg/ml) were ap-
plied to the affinity column previously equilibrated with PBS, and the
unbound fraction was washed through with 10 bed volumes of PBS.
Bound proteins were eluted with 0.1 M glycineHCl, pH 2.5, then neu-
tralized with Tris, dialyzed against PBS, and concentrated by ultrafil-
tration using Centricon YM-10 centrifugal filter devices (Millipore Cor-
poration).
Immunoblotting
The F. hepatica antigens obtained by MM3 affinity chromatography
were separated by polyacrylamide gel electrophoresis under reducing
conditions (sodium dodecyl sulfatepolyacrylamide gel electrophoresis
[SDS-PAGE], 520% linear gradient, as per Laemmli, 1970) and then
analyzed by immunoblotting. In brief, after blocking with 5% dry
skimmed milk in Tris-buffered saline (20 mM, pH 7.4), the nitrocellu-
lose membranes were first incubated with fluorescein isothiocyanate
(FITC)labeled mAb MM3 (as described by Liddell and Cryer, 1991),
then with peroxidase-conjugated rabbit anti-FITC Ig (DAKO Diagnos-
ticos SA, Barcelona, Spain; dilution 1:500), and finally with the sub-
strate (3,3-diaminobenzidine tetrahydrochloride tablets, SigmaAl-
drich). As a positive control, we used membranes incubated with the
rabbit antiF. hepatica polyclonal IgG antiserum and peroxidase-con-
jugated goat anti-rabbit IgG (SigmaAldrich; 1:2,000).
Two-site capture ELISA
A 2-site capture ELISA was used to evaluate whether the epitope
recognized by mAb MM3 is present at single or multiple sites on the
target antigen. Polystyrene microtiter plates (Nunc Immuno Plate Max-
isorp Surface, NUNC, Roskilde, Denmark) were coated at 37 C for 2
hr with 100 ml/well of a solution containing 10 mg/ml of MM3 in PBS.
After blocking uncoated sites with 5% skimmed dry milk in PBS, 100
ml of the MM3 affinity-purified antigen fraction (protein content 5 mg/
ml in PBS) was added to each well. After incubation for 2 hr at 37 C,
the plates were washed and then incubated (90 min at 37 C) with 100
ml/well of FITC-labeled MM3 diluted 1:100 in PBS with 0.2% Tween
20 and 1% skimmed dry milk (PBS-T). After washing 6 times, the
reaction was detected by incubation first with peroxidase-conjugated
rabbit anti-FITC Ig (DAKO Diagnosticos SA) diluted 1:1,000 in PBS-
T and then with the substrate (buffered H2O2 and o-phenylenediamine
[OPD] SigmaAldrich, Madrid, Spain). After incubation for 30 min at
room temperature, the optical density (OD) was measured at 492 nm.
As a positive control to detect the antigenMM3 complex, we used
rabbit antiF. hepatica polyclonal IgG antiserum detected in turn with
peroxidase-conjugated goat anti-rabbit IgG (SigmaAldrich; 1:20,000).
Detection of Fasciola hepatica coproantigens in ovine and bovine
feces
Sheep: Feces were obtained from an autochtonous Galician sheep
strain (raza gallega), from a fluke-free herd (group A), from lambs
experimentally infected with very small numbers of F. hepatica meta-
cercariae (group B), and from uninfected lambs (group C, control). The
group-A feces were obtained from fifty 4- to 6-mo-old lambs just before
slaughter. Specifically, fecal samples were taken from each animal just
before inspection to confirm the absence of fascioliasis. Other parasitic
infections were diagnosed by visual examination and microscopic anal-
ysis of feces: 4 of the 50 lambs did not have detectable parasite infec-
tion, whereas the remaining 46 had intestinal nematodes from 1 or more
genera of Trichostrongylidae, Molineidae (Nematodirus spp.), Ancylo-
stomatidae, Strongylidae, and Trichuridae (Trichuris spp.). Of the 46
parasite-infected animals, 21 had gastrointestinal nematodes alone,
whereas the remaining 25 had double or triple infections with lung
nematodes (1 or more of Cystocaulus ocreatus, Dyctiocaulus filariae,
or Neostrongylus linearis), Moniezia spp., or Cysticercus tenuicollis.
The group-B feces were obtained from twenty-one 4-mo-old lambs
reared under parasite-free conditions and subsequently infected once
with 5 (n 5 3), 10 (n 5 3), 20 (n 5 3), or 40 (n 5 12) F. hepatica
metacercariae obtained in our laboratory from experimentally infected
 MEZO ET AL.FASCIOLA HEPATICA COPROANTIGENS IN SHEEP AND CATTLE 847

Lymnaea truncatula snails. Six of the 12 lambs infected with 40 meta-

cercariae were treated with a single oral dose of 10 mg/kg of triclaben-
dazole (Fasinex 5%, Novartis Animal Health, Barcelona, Spain) at 14
wk PI. Fecal samples were taken from each animal just before infection
and then weekly for 18 wk PI. At the end of the experiment (18 wk
PI), all animals were necropsied and all liver parasites collected. The
group-C feces were obtained from four 4-mo-old uninfected lambs
raised under the same conditions as for group B.
Cattle: One hundred and eighty livers, together with corresponding
samples of feces, were collected at the slaughterhouse from both nat-
urally F. hepaticainfected (n 5 80) and F. hepaticafree (n 5 100)
adult cattle (Friesian strain, 4- to 8-yr-old). Each liver was examined,
and all flukes were collected and counted. Visual examination and mi-
croscopic analysis of feces from the F. hepaticafree group revealed
that only 19 of the 100 animals did not have detectable parasite infec-
tion, whereas the remaining 81 had intestinal nematodes from 1 or more
genera of Trichostrongylidae, Trichuridae (Trichuris spp.), and Stron-
gylidae. Of these 81 animals, 73 had gastrointestinal nematodes alone,
whereas the remaining 8 also had infections with Moniezia spp. (2 an-
imals), Paramphistomun cervi (1), Dicrocoelium dendriticum (3), or
Echinococcus spp. metacestodes (2).

Obtaining Fasciola hepatica antigens from fecal samples

The presence of F. hepatica coproantigens was investigated in fecal
supernatants prepared by mixing individual ovine or bovine stools with
distilled water at a ratio of 1:4 (1 g 1 4 ml) or 1:1 (3 g 1 3 ml),
respectively. After centrifugation at 1,000 g for 10 min, the supernatants
were collected and stored at 220 C until analysis by ELISA. Fecal
samples were also examined for F. hepatica egg count by a quantitative
sedimentation technique (Anderson et al., 1999) with a sensitivity of 2
eggs per gram of feces (e.p.g.).

MM3 capture ELISA for the detection of Fasciola hepatica

coproantigen
An ELISA was performed using rabbit polyclonal and mouse mono-
clonal IgG1 antibodies for the capture and detection of F. hepatica co-
proantigens, respectively. Polystyrene microtiter plates (Nunc Immuno
Plate Maxisorp Surface, NUNC) were coated at 37 C for 2 hr with the
rabbit antiF. hepatica polyclonal IgG antiserum (100 ml/well of a so-
lution containing 10 mg/ml protein) in 0.05 M carbonate buffer, pH 9.6.
After blocking as described for the 2-site capture ELISA, each fecal
supernatant was added in quadruplicate (100 ml/well) and incubated
overnight at 4 C. After washing 6 times, 100 ml of PBS-T containing
0.3 mg of MM3 was added to 2 of the 4 wells per sample, whereas the
other 2 wells (blanks) received PBS-T only. Plates were incubated for
90 min at 37 C and then washed as above. Bound MM3 was detected
by incubation first with peroxidase-conjugated goat anti-mouse IgG
(BioRad, Richmond, California; 1:3,000 in PBS-T for 1 hr at 37 C) and
then with OPD (SigmaAldrich), with incubation OD measurement as FIGURE 1. Immunoblotting analyses of F. hepatica ESAs, obtained
that for the 2-site capture ELISA. The OD value for each sample was by affinity chromatography with mAb MM3 immobilized on the HiTrap
calculated as OD1OD2, where OD1 is the mean for the 2 wells to affinity column (Amersham Biosciences). The figure shows lanes
which MM3 was added and OD2 is the mean for the 2 wells without probed with MM3 (lane 1) and with a rabbit antiF. hepatica ESA
MM3. polyclonal antiserum (lane 2). Lane 3 shows the reactivity of the rabbit
antiF. hepatica ESA polyclonal antiserum with whole F. hepatica
Statistics ESAs. Before immunoblotting, samples were separated by 520% linear
ELISA OD cutoff values were determined on the basis of values gradient reducingdenaturing SDS-PAGE. Relative molecular masses
obtained for fecal supernatants from sheep or cattle not infected with (Mr) of standard markers are shown on the left.
F. hepatica. Specifically, the cutoff for each species (sheep or cow) was
calculated as the mean for the fluke-free group plus 4 standard devia-
tions; values more than this cutoff were considered as positive for F. SDS-PAGE under reducing conditions, and its composition was
hepatica. The calculated cutoff values were 0.065 for sheep (mean 5
analyzed by Western blotting. As can be seen in Figure 1, MM3
0.013; SD 5 0.013; n 5 50) and 0.114 for cattle (mean 5 0.026; SD
5 0.022; n 5 100). Relationships between time PI at which fecal sam- did not recognize the isolated polypeptides in immunoblotting
ples became positive by ELISA, coproantigen concentration, and fluke (lane 1), suggesting that the epitope is conformational. In con-
burden were investigated by Pearson productmoment correlation anal- trast, the rabbit antiF. hepatica polyclonal IgG antiserum rec-
ysis using the SAS system (SAS Institute, Cary, North Carolina). ognized a minimum of 17 bands (Fig. 1, lane 2) with maximal
reactivity against polypeptides of 15 kDa, 3334 kDa, and 56
RESULTS
kDa (arrowheads). However, note that we cannot be certain that
Recognition of the MM3 affinity-purified Fasciola hepatica all these polypeptides are MM3-recognized polypeptides or
antigens fragments thereof because it is possible that some nonMM3-
After purification of F. hepatica ESAs by MM3 affinity chro- recognized contaminants are present in the MM3 affinity-puri-
matography, the retained fraction was eluted and separated by fied fraction. Unlike that in immunoblotting, MM3 showed
848 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 4, AUGUST 2004
FIGURE 2. Calibration curve for the MM3 capture ELISA, obtained
by plotting OD at 492 nm against known concentration (0.15610 ng/
ml) of whole F. hepatica ESAs in fecal supernatant.
clear reactivity in an indirect ELISA (immobilized MM3 affin-
ity-purified fraction as target antigen, MM3 as probe), but the
reactivity disappeared when the target antigen was previously
heated at 100 C for 5 min or at 80 C for 2 hr (although not
when the antigen was heated at 65 C for 6 hr or subjected to
alkaline deglycosylation) (results not shown). In addition, it
should be stressed that MM3 recognized the antigen in capture
ELISA with immobilized rabbit antiF. hepatica ESA poly-
clonal antibody (see below), but it was not possible to use MM3
in capture ELISA for both the capture and detection of the
antigen; this suggests that there is a single epitope on each
molecule recognized by MM3.
Detection limit and specificity of the MM3 capture ELISA
for the detection of Fasciola hepatica antigens in ovine
and bovine stools
To assess the suitability of the ELISA for the detection of F.
hepatica antigens in ovine and bovine stools, we first generated
a standard curve by adding known concentrations (from 0.156
to 40 ng/ml) of F. hepatica ESAs to the fecal supernatant ob-
tained from a pool of negative feces (n 5 25) and then deter-
mined OD values in ELISA for each concentration. As shown
in Figure 2, the assay showed linear response to coproantigen
concentration (r2 5 0.9954; P , 0.0001) for the range 0.156
10 ng/ml. The curve was the same regardless of whether ovine
or bovine feces were used, so the same curve was used for both
species. The detection limit for sheep feces was 0.3 ng/ml, es-
timated as the lowest antigen concentration giving an OD read-
ing (0.068) higher than the estimated cutoff for sheep (0.065;
see above). The detection limit for cattle was 0.6 ng/ml (OD
0.120, cutoff 0.114).
To test the specificity of the assay, we used feces from 50
sheep (group-A sheep) and 100 adult cows; both groups were
negative for F. hepatica, although most animals (46/50 and 81/
100, respectively) harbored other parasites, mainly gastrointes-
tinal nematodes but also trematodes and cestodes. The presence
of these parasite infections did not influence the ELISA results
FIGURE 3. Mean OD values (6SD) obtained in MM3 capture ELISA
of fecal supernatants from lambs experimentally infected with 40 meta-
cercariae (closed inverted triangles), 20 metacercariae (closed triangles),
10 metacercariae (closed circles), 5 metacercariae (closed squares), or
vehicle only (uninfected controls, open circles) (n 5 3 lambs per group,
except for the lambs infected with 40 metacercariae [n 5 6]) at different
times after infection.
because the OD values obtained were in all cases extremely
low both in sheep (mean OD 0.013; range 0.0020.044 OD)
and cattle (mean OD 0.026; range 0.0000.109 OD). This clear-
ly demonstrates the absence of cross-reactions with the most
frequent helminths parasitizing domestic ruminants in our geo-
graphical area. Absence of reactivity with D. dendriticum an-
tigens was also confirmed by testing samples made up by add-
ing whole somatic antigen of D. dendriticum (8 mg/ml) to a F.
hepaticanegative fecal supernatant pool.
Kinetics of Fasciola hepatica coproantigen excretion in
experimentally infected lambs before and after flukicide
treatment
The kinetics of fecal excretion of the antigens specifically
recognized by mAb MM3 were studied for 18 wk in lambs
belonging to groups B (experimentally infected; n 5 21) and C
(uninfected controls). Six lambs from group B were treated with
triclabendazole on week 14 PI to investigate the usefulness of
the MM3 capture ELISA for monitoring treatment efficacy on
the basis of coproantigen levels. Necropsies performed at the
end of the study showed that all untreated experimentally in-
fected animals (n 5 15) had between 1 and 36 flukes, whereas
all treated experimentally infected animals (n 5 6) had no fluke.
Coproantigen levels in the control group remained low through-
out the 18-wk period (OD 0.0010.042), whereas levels in the
untreated infected group increased above the cutoff at weeks
78 PI (or earlier in lambs infected with the highest dose, 40
metacercariae) and remained high thereafter (Fig. 3). A signif-
icant negative correlation (r 5 20.712; P , 0.01; n 5 15,
group-B lambs) was observed between fluke number and the
time (week) at which samples tested positive in ELISA (Fig.
4). Comparison with fecal egg count data (Fig. 5) indicated that
 MEZO ET AL.FASCIOLA HEPATICA COPROANTIGENS IN SHEEP AND CATTLE 849

FIGURE 4. Plot of the week PI at which MM3 capture ELISA of
fecal supernatants tested positive against the number of flukes recovered
at necropsy. With the exception of the point at 3 flukes, 8 wk, repre-
senting 3 lambs, all other points represent individual experimentally
infected lambs (15 lambs in total).
FIGURE 5. Percentages of experimentally infected lambs testing pos-
itive by MM3 capture ELISA and by fecal examination for eggs, at
different times after infection. Specific F. hepatica coproantigens were
considered to be present if OD values were higher than the cutoff
(0.065) calculated as mean plus 4 SD of the OD values for fecal su-
pernatants from 50 lambs from a fluke-free herd (group A).

depending on parasite burden, infected lambs had detectable

amounts of the coproantigen from 1 to 5 wk before patency
(egg shedding).
Coproantigen levels, egg counts, and parasite burdens are
listed for the 6 triclabendazole-treated animals and the 15 un-
treated animals in Table I. All untreated animals were clearly
positive for the coproantigen throughout the study, concordant
with the presence of flukes in the liver and eggs in feces. By
contrast, in most of the treated lambs (5/6), the coproantigen
was not detected from the first week after triclabendazole ad-
ministration. In the sixth treated lamb, the coproantigen was not
detected from the third week after administration. These data
contrast with the egg count data in that eggs continued to be
intermittently shed in feces until the fourth week after treat-
ment.
Fecal samples from the untreated experimentally infected
lambs (n 5 15) were also used to investigate possible correla-
tion between coproantigen levels and fluke burden. To do this,
we first diluted fecal samples until we obtained OD readings in
the linear range of the standard curve and then transformed OD
values to antigen concentrations using the standard curve. Co-
proantigen concentrations in lambs with 136 flukes ranged be-
tween 13 and 413 ng/ml. A significant positive correlation (r
5 0.889; P , 0.001) was observed between these 2 variables
(Fig. 6).
MM3 capture ELISA detection of Fasciola hepatica
coproantigens in adult cattle with natural fascioliasis
The sensitivity of the MM3 capture ELISA for the detection
of F. hepatica coproantigens was evaluated with fecal samples
from adult cattle with natural fascioliasis confirmed by necrop-
sy at the slaughterhouse. Eighty positive cows were examined
and their flukes collected, counted, and measured. Liver parasite
burdens ranged from 1 to 154 flukes, but most of the animals
(73/80) had 50 or fewer flukes. Using the predefined cutoff

TABLE I. Fecal egg counts (eggs per gram [e.p.g.]) and coproantigen levels (OD values) in experimentally Fasciola hepaticainfected lambs
before (n 5 15) and after flukicide treatment (n 5 6). Fluke burdens were recorded at necropsy. The number of lambs testing positive (by ELISA
or egg count) is given in parentheses. OD492 cutoff value 5 0.065. Flukicide treatment with triclabendazole was done at week 14 PI.

Weeks postinfection
14 15 16 17 18

Untreated animals
Mean e.p.g. 6 SD 56 6 73 (15) 64 6 93 (15) 59 6 92 (15) 95 6 115 (15) 100 6 102 (15)
Mean OD 6 SD 2.042 6 0.189 (15) 1.981 6 0.298 (15) 2.135 6 0.105 (15) 2.005 6 0.275 (15) 2.139 6 0.224 (15)
Range fluke burden ND ND ND ND 136 (15)
Treated animals
Mean e.p.g.* 6 SD 67 6 65 (5) 0 (0) 5 6 3 (5) 12 6 13 (2) 45 6 59 (2)
Mean OD* 6 SD 1.873 6 0.294 (6) 0.530 6 0.06 (1) 0.130 6 0.025 (1)
Range fluke burden ND ND ND ND 0 (0)

* Mean OD values and mean e.p.g. counts, both in triplicate, were calculated using data from positive animals only.
850 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 4, AUGUST 2004
FIGURE 6. Plot of the antigen concentration in fecal supernatants as
determined by MM3 capture ELISA at 18 wk PI against the number of
flukes recovered at necropsy. Each point represents an individual ex-
perimentally infected lambs (15 lambs in total).
value (mean 1 4SD 5 0.114), the F. hepatica coproantigen
was detected in 2 of the 7 cattle with 1 fluke and in all animals
with 2 or more flukes (Fig. 7). The coproantigen concentration
in cattle feces ranged between 0.805 and 199.8 ng/ml. The
ELISA was clearly more sensitive than egg count because no
F. hepatica egg was detected in the feces of 38.8% (31/80) of
these naturally infected cattle (Table II); furthermore, 59.2%
(29/49) of the positive counts had less than 10 e.p.g. In fact,
the egg count method was 100% sensitive only for cows with
more than 50 flukes; but such high burdens were uncommon in
our sample (7/80).
We measured a total of 1,375 flukes from 52 cows. Although
most flukes were of intermediate size, lengths ranged from 5 to
38 mm (Fig. 8). The possible relevance of fluke size is dis-
cussed in the next section.
DISCUSSION
Active infections by F. hepatica in humans and animals can
be diagnosed by fecal egg count, by measurement of serum
antigen levels during a limited period after infection, i.e., 721
days (Langley and Hillyer, 1989; Espino et al., 1997), or by
determination of ESAs in feces (coproantigens) after weeks 5
6 PI, using polyclonal or monoclonal antibodies (Abdel-Rah-
man et al., 1998; Dumenigo et al., 2000; Almazan et al., 2001).
Coproantigen analyses are possible because the antigenicity of
TABLE II. Fecal egg counts (eggs per gram [e.p.g.]) in naturally Fasciola hepaticainfected cattle, classified (columns) by fluke burden at necropsy.
1 210
Egg count
Mean e.p.g.* 6 SD 3 6 1.1
Number of positive animals (%) 1 (14.3)
Total animals 7 34
* Mean e.p.g. counts were calculated using data from positive animals only (three independent determinations).
FIGURE 7. Fasciola hepatica coproantigen concentrations measured
by MM3 capture ELISA in fecal supernatants from 180 cows killed at
the slaughterhouse. Six fluke burden categories: no fascioliasis (n 5
100 cows), 1 fluke (n 5 7), 210 flukes (n 5 34), 1120 flukes (n 5
17), 2150 flukes (n 5 15), and more than 50 flukes (n 5 7). The
dashed line is the detection limit in the ELISA assay for cattle (0.6 ng/
ml, corresponding to the lowest concentration giving an OD reading
[0.120] higher than the cutoff for cattle [0.114]).
many ESAs from F. hepatica is retained after contact with acid
and digestive enzymes. This approach has the advantage that it
enables diagnosis during the prepatent period, i.e., before egg
production starts, thus reducing the risk of pasture contamina-
tion. To date, however, this approach has been less effective for
detecting low-intensity infections, particularly in cows, whose
abundant feces greatly dilute coproantigens. In this study a
large proportion of animals (32.5%) showed very low-intensity
parasitizations (14 flukes), and similar situations have been
reported from other countries, e.g., Anderson et al. (1999) re-
ported that 22% of cattle in Vietnam had only 110 flukes.
The MM3 capture ELISA described in this report had detec-
tion limits of 0.3 and 0.6 ng/ml of F. hepatica ESAs for sheep
and cattle, respectively, allowing detection of 100% of cases of
fascioliasis in sheep with only 1 fluke or cattle with only 2
flukes. Even in cattle with only 1 fluke the assay detected the
infection in 2 of 7 individuals. The reason why the remaining
5 individuals with 1 fluke tested negative is unclear. However,
because the MM3 capture ELISA was able to detect coproan-
tigen concentrations less than the 1.581.70 ng/ml produced by
a single fluke in infected cows, it seems clear that the problem
was not due to underdetection of the antigen. An alternative
Number of flukes in the liver
1120 2150 .50
6.1 6 4.4 14.4 6 22.2 9.4 6 4.3 30.7 6 28.8
16 (47.1) 13 (76.5) 12 (80) 7 (100)
17 15 7
 MEZO ET AL.FASCIOLA HEPATICA COPROANTIGENS IN SHEEP AND CATTLE
FIGURE 8. Frequency distribution of fluke length for the 1,375 flukes
obtained from 52 naturally infected cows at the slaughterhouse.
possibility is that these false negatives were due to the absence
or very low concentrations of ESAs in feces, as may occur
when the flukes are still immature, i.e., until established in the
bile ducts, by weeks 810 PI, with length about 1012 mm
(Behm and Sangster, 1999), or if the presence of the parasite
leads to temporary obstruction of the bile ductules or duct
(Wensvoort and Over, 1982). If this were the case, we would
expect feces from these false-negative animals to test positive
on subsequent days, as occurred in sheep (see Fig. 5). Although
this possibility could not be tested in cattle because the fecal
samples were obtained at the slaughterhouse immediately after
slaughter, the small size of flukes from these 5 negative animals
(length 511 mm) and their location (mostly in the hepatic pa-
renchyma or still in the ductules) suggest that the coproantigen
negativity may be due to the immature stage of these flukes.
The coexistence of mature and immature flukes in a single in-
dividual is a frequent finding in natural infections of cattle and
may explain why 2 individuals with similar fluke burden have
different antigen concentrations in feces; as seen in Figure 7, 6
(17.6%) of the 34 cows with 210 flukes and 1 (5.8%) of the
17 cows with 1120 flukes showed coproantigen concentrations
similar to or lower than the 2 positive cows with single flukes.
Interestingly, feces from 53 (71%) of the 75 cows testing pos-
itive in MM3 capture ELISA had OD values corresponding to
antigen concentrations less than 10 ng/ml (range 0.8049.873
ng/ml); most (32) of these 53 cows were from the 210 fluke
group (n 5 34), but 13 were from the 1120 fluke group (n 5
17) and 6 from the 2150 fluke group (n 5 15). These OD
values are markedly lower than the lower limit of 25 ng/ml
reported by Dumenigo et al. (1996) in infected cattle using mAb
ES78 and highlight the need for sensitive methods for coproan-
tigen determination in cattle, both to reduce the number of false
negatives and to enable effective monitoring of coproantigen
levels after flukicide treatment. It might be argued that losses
due to fluke infection in sheep and cattle are minimal for ani-
mals with low parasite burden; however, diagnosis of these an-
imals is essential to avoid pasture contamination and when in-
troducing new animals into a Fasciola-free area.
851
In addition to its low detection limit, the MM3 capture
ELISA tested in this study seems to be highly specific because
none of the feces samples from F. hepaticanegative sheep or
cattle tested positive, despite natural infection with parasites of
diverse nematode, cestode, and trematode genera (including D.
dendriticum). This result is not surprising because mAb MM3
was produced by immunization of mice with a purified and O-
deglycosylated fraction (740 kDa) of F. hepatica ESAs that
we have recently shown to be specific to this parasite (Mezo et
al., 2003).
Compared with other mAb-based capture ELISA methods,
i.e., mAbs F10 and ES78, used to detect coproantigens in F.
hepaticainfected sheep and cattle, the MM3 capture ELISA
seems to be much more sensitive. F10-based assay was reported
to have a detection limit of 0.3 ng/ml, similar to that of our
MM3-based assay; however, the detection limit with F10 was
calculated using a calibration curve with affinity-purified anti-
gen (2628 kDa, ESA fraction), whereas the detection limit
with MM3 was determined with respect to complete F. hepatica
ESAs, of which the antigens recognized by MM3 make up only
a small part. This may explain at least in part why F10 did not
detect infection in cattle with fewer than 10 flukes. For ES78-
based assay, Dumenigo et al. (1996) reported a detection limit
of 15 ng/ml, similar to that obtained by Langley and Hillyer
(1989) using a 2-site ELISA with polyclonal antibodies but 25
times higher than that of our MM3-based assay. Thus, neither
F10- nor ES78-based assay is likely to be effective for detecting
the low antigen concentrations seen in feces of many of our
naturally infected cattle or for detecting failure of flukicide
treatment if, for example, less than 5 flukes survive. Indepen-
dent of this, the rather high detection limit of the ES78-based
assay is surprising, given that the target epitopes recognized by
mAbs ES78 and MM3 have many similarities. Namely, both
mAbs recognize single conformational epitopes present on sev-
eral F. hepatica polypeptides whose antigenicity is abrogated
by heating the antigen to more than 65 C (ES78; Daz et al.,
1998; Marcet et al., 2002) or 80 C (MM3) or by reducing con-
ditions in SDS-PAGE and immunoblotting. On the other hand,
MM3 binding is not abrogated by O-deglycosylation of the an-
tigen, whereas ES78 binding is; however, note that ES78 was
tested against F. hepatica antigens that were O-deglycosylated
in the presence of sodium borohydride, which has been reported
to destroy the protein moiety of glycoproteins during the de-
glycosylation process (Montreuil et al., 1994), so that the result
of Diaz et al. may be inconclusive. Under these circumstances,
it is difficult to establish whether the difference in detection
capacity between ES78- and MM3-based assay is due to dif-
ferences in the antigen recognized by each mAb or alternatively
to differences in affinity or immunoassay design. In this con-
nection, in ES78-based assay the antigen is first captured by
the immobilized mAb and then detected with a peroxidase-con-
jugated antiF. hepatica polyclonal antibody (Espino and Du-
menigo, 2003), whereas in MM3-based assay, coproantigens are
first captured by an immobilized polyclonal antibody, and spe-
cific antigens are then detected with MM3, which is detected
in turn with polyclonal peroxidase-conjugated goat anti-mouse
IgG antibody. Future studies comparing the 2 mAbs in com-
petitive ELISA may clarify this question.
In conclusion, the MM3 capture ELISA described in this
study is a reliable, robust, and ultrasensitive method for de-
852 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 4, AUGUST 2004
tecting F. hepatica infections in sheep and cattle, capable of
detecting subnanogram amounts of specific ESAs in feces. Pre-
liminary work in our laboratory seems to indicate that MM3,
like other antiF. hepatica mAbs, may also be useful for de-
tecting circulating F. hepatica antigens in infected animals, as
well as for the determination of antiFasciola serum antibodies
in animals and human infections.
ACKNOWLEDGMENTS
We thank Antonio Lagares, chief veterinary officer of the M. F. Mon-
tellos abattoir (Betanzos, A Coruna, Spain), for freely facilitating sam-
pling for this study. This work was supported in part by grants SC00-
085 from Instituto Nacional de Investigaciones Agrarias (Spain) and
SAF2002-04057-02 from Ministerio de Ciencia y Tecnologa (Spain).
LITERATURE CITED
ABDEL-RAHMAN, S., K. L. OREILY, AND J. B. MALONE. 1998. Evaluation
of a diagnostic monoclonal antibody-based capture enzyme-linked
immunosorbent assay for detection of a 26- to 28-kd Fasciola he-
patica coproantigen in cattle. American Journal of Veterinary Re-
search 59: 533537.
, , AND . 1999. Biochemical characterization and LIDDELL, J. E., AND A. CRYER. 1991. Characterization, purification and
localization of Fasciola hepatica 2628 kDa diagnostic coproan-
tigen. Parasite Immunology 21: 279286.
ALMAZAN, C., G. AVILA, H. QUIROZ, F. IBARRA, AND P. OCHOA. 2001.
Effect of parasite burden on the detection of Fasciola hepatica
antigen in sera and feces of experimentally infected sheep. Veter-
inary Parasitology 97: 101112.
ANDERSON, N., T. T. LUONG, N. G. VO, K. L. BUI, P. M. SMOOKER, AND
T. W. SPITHILL. 1999. The sensitivity and specificity of two methods
for detecting Fasciola infections in cattle. Veterinary Parasitology
83: 1524.
BEHM, C. A., AND N. C. SANGSTER. 1999. Pathology, pathophysiology
and clinical aspects. In Fascioliasis, J. P. Dalton (ed.). CABI Pub-
lishing, Oxon, U.K., p. 185224.
CONCEICAO, M. A., R. M. DURAO, I. H. COSTA, AND J. M. CORREIA DA
COSTA. 2002. Evaluation of a simple sedimentation method (mod-
ified McMaster) for diagnosis of bovine fascioliasis. Veterinary
Parasitology 105: 337343.
DAZ, A., A. M. ESPINO, R. MARCET, O. OTERO, D. TORRES, C. M.
FINLAY, AND J. SARRACENT. 1998. Partial characterization of the
epitope on the excretory-secretory products of Fasciola hepatica
recognized by monoclonal antibody ES78. Journal of Parasitology
84: 5561.
DUMENIGO, B. E., A. M. ESPINO, AND C. M. FINLAY. 1996. Detection of
Fasciola hepatica antigen in cattle feces by a monoclonal antibody-
based sandwich immunoassay. Research in Veterinary Science 60:
278279.
, , , AND M. MEZO. 2000. Kinetics of antibody- SOLANO, M., R. K. RIDLEY, AND H. C. MINOCHA. 1991. Production and
based antigen detection in serum and feces of sheep experimentally
infected with Fasciola hepatica. Veterinary Parasitology 89: 153
161.
ESPINO, A. M., AND B. E. DUMENIGO. 2003. Fasciola hepatica. In In-
ternational handbook of foodborne pathogens, M. D. Milliotis and
J. W. Bier (eds.). Marcel Dekker, Inc., Basel, Switzerland, p. 539
562.
, AND C. M. FINLAY. 1994. Sandwich enzyme-linked immuno-
sorbent assay for detection of excretory secretory antigens in hu-
mans with fascioliasis. Journal of Clinical Microbiology 32: 190
193.
, M. MARCET, AND C. M. FINLAY. 1997. Fasciola hepatica: De-
tection of antigenemia and coproantigens in experimentally infect-
ed rats. Experimental Parasitology 85: 117120.
ESTEVEZ, J., J. LEIRO, M. T. SANTAMARINA, J. DOMNGUEZ, AND F. M.
UBEIRA. 1994. Monoclonal antibodies to turbot (Scophthalmus
maximus) immunoglobulins: Characterization and applicability in
immunoassays. Veterinary Immunology and Immunopathology 41:
353366.
HANNA, R. E. B., AND A. G. TRUDGETT. 1983. Fasciola hepatica: De-
velopment of monoclonal antibodies and their use to characterize
a glycocalyx antigen in migrating flukes. Parasite Immunology 5:
409425.
, , AND A. ANDERSON. 1988. Fasciola hepatica: Devel-
opment of monoclonal antibodies and their characterization by ul-
trastructural localization of antibody binding. Journal of Helmin-
thology 62: 1528.
IGLESIAS, R., J. LEIRO, M. T. SANTAMARINA, M. L. SANMARTN, AND F.
M. UBEIRA. 1997. Monoclonal antibodies against diagnostic Ani-
sakis simplex antigens. Parasitology Research 83: 755761.
LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assem-
bly of the head of bacteriophage T4. Nature 277: 680685.
LANGLEY, R. J., AND G. V. HILLYER. 1989. Detection of circulating par-
asite antigen in murine fascioliasis by two-site enzyme-linked im-
munosorbent assays. American Journal of Tropical Medicine and
Hygiene 41: 472478.
labelling. In A practical guide to monoclonal antibodies, J. W. God-
ing (ed.). Academic Press Inc., London, U.K., p. 105138.
LORENZO, S., F. ROMARS, R. IGLESIAS, M. T. AUDCANA, J. M. ALONSO,
J. LEIRO, AND F. M. UBEIRA. 2000. O-Glycans as a source of cross-
reactivity in determinations of human serum antibodies to Anisakis
simplex antigens. Clinical and Experimental Allergy 30: 551559.
MARCET, R., A. DAZ, E. ARTEAGA, C. M. FINLAY, AND J. SARRACET.
2002. Passive protection against fascioliasis in mice by immuni-
zation with a monoclonal antibody (ES-78 MoAb). Parasite Im-
munology 24: 103108.
MEZO, M., M. GONZALEZ-WARLETA, AND F. M. UBEIRA. 2003. Optimized
serodiagnosis of sheep fascioliasis by FPLC fractionation of Fas-
ciola hepatica excretory-secretory antigens. Journal of Parasitology
89: 843849.
MONTREUIL, J., S. BOUQUELET, H. DEBRAY, J. LEMOINE, J. C. MICHALSKI,
G. SPIK, AND G. STRECKER. 1994. Glycoproteins. In Carbohydrate
analysis, a practical approach, M. F. Chaplin and J. F. Kennedy
(eds.). IRL Press, New York, p. 181293.
PAZ-SILVA, A., J. PEDREIRA, R. SANCHEZ-ANDRADE, J. L. SUAREZ, P.
DAZ, R. PANADERO, P. DEZ-BANOS, AND P. MORRONDO. 2002. Time-
course analysis of coproantigens in rats infected and challenged
with Fasciola hepatica. Parasitology Research 88: 568573.
SANCHEZ-ANDRADE, R., A. PAZ-SILVA, J. L. SUAREZ, R. PANADERO, J.
PEDREIRA, P. DEZ-BANOS, AND P. MORRONDO. 2001. Effect of fas-
ciolicides on the antigenemia in sheep naturally infected with Fas-
ciola hepatica. Parasitology Research 87: 609614.
characterization of monoclonal antibodies against excretory-secre-
tory products of Fasciola hepatica. Veterinary Parasitology 40:
227239.
WENSVOORT, P., AND H. J. OVER. 1982. Cellular proliferation on bile
ductules and gamma-glutamyl transpeptidase in livers and sera of
young cattle following a single infection with Fasciola hepatica.
Veterinary Quarterly 4: 161172.
ZIMMERMAN, G. L., L. W. JEN, J. E. CERRO, K. L. FARNSWORTH, AND R.
B. WESCOTT. 1982. Diagnosis of Fasciola hepatica infections in
sheep by an enzyme-linked immunosorbent assay. American Jour-
nal of Veterinary Research 43: 20972100.