A novel 57-kDa merozoite protein of Babesia gibsoni is a
prospective antigen for diagnosis and serosurvey of canine
babesiosis by enzyme-linked immunosorbent assay
Gabriel Oluga Abogea, Honglin Jiaa, Mohamad Alaa Terkawia,
Younkyoung Gooa, Ken Kurikib, Yoshifumi Nishikawaa,
Ikuo Igarashia, Hiroshi Suzukia, Xuenan Xuana,*
aNational Research Center for Protozoan Diseases, Obihiro University of Agriculture and
Veterinary Medicine, Inada-cho, Obihiro, Hokkaido 080-8555, Japan
bKyoritsu Seiyaku Corporation, Chiyoda-ku, Tokyo 102-0073, Japan
Received 2 March 2007; received in revised form 9 May 2007; accepted 20 June 2007
thecDNAwas2387 bpwithanopenreadingframe(ORF)of1644 bpencodinga57-kDapredictedpolypeptidehaving547aminoacid
residues. The recombinant BgP57 (rBgP57) without a predicted signal peptide was expressed in Escherichia coli as a soluble
withmolecular weight ofpredicted maturepolypeptide. Anindirect enzyme-linkedimmunosorbentassay(ELISA) using the rBgP57
detected specific antibodies in the sequential sera from a dog experimentally infected with B. gibsoni. Moreover, the antigen did not
antigen for B. gibsoni antibodies. The diagnostic performance of ELISA based on rBgP57 using 107 sera from B. gibsoni-naturally
infected dogs was the same as the previously identified rBgP32 but performed better than the previously studied rBgP50. Although,
seminested PCR detected higher proportions (82%) of positive samples than the ELISAs, the Mcnemar’s chi-square test showed that
there was no significant difference in relative effectiveness of rBgP57-ELISA and seminested PCR (x2= 2.70; P = 0.1003) in
identifying positive samples. The rBgP57-ELISAwhen used in combination with rBgP32-ELISA and rBgP50-ELISA appeared to
improve sensitivity of the rBgP57-ELISA for detection of B. gibsoni antibodies. Overall, the rBgP57-ELISA and seminested PCR
when used in combination, could improve epidemiological surveys and clinical diagnosis of B. gibsoni infection.
# 2007 Elsevier B.V. All rights reserved.
Keywords: ELISA; 57-kDa merozoite protein; B. gibsoni
Babesia gibsoni is a tick-bone hemoprotozoan
parasite that causes canine babesiosis worldwide, in
Africa, Asia, the United States and Europe (Kjemtrup
et al., 2000). The acute form of B. gibsoni infection
usually clinically manifests itself as fever, anemia,
lethargy and splenomegaly (Conrad et al., 1991). In
contrast, the chronic form of the disease is clinically
characterized poorly (Conrad et al., 1991), and infected
animals may become chronic carriers without clinical
manifestations thus complicating the diagnosis. The
Veterinary Parasitology 149 (2007) 85–94
* Corresponding author. Tel.: +81 155 49 5648;
fax: +81 155 49 5643.
E-mail address: email@example.com (X. Xuan).
0304-4017/$ – see front matter # 2007 Elsevier B.V. All rights reserved.
classical diagnosis of animals acutely infected with B.
gibsoni is based on the light microscopic demonstration
of intraerythrocytic parasites in Giemsa-stained blood
smears (Conrad et al., 1991). However, in subclinical or
latent infections, this may be impractical due to low
levels of parasitemia.
Polymerase chain reaction (PCR) is an alternative
test, with good sensitivity and specificity (Ano et al.,
2001; Birkenheuer et al., 2003b), and is able to detect
early or carrier infections, but the test requires
specialized laboratory equipment and highly trained
personnel. Serological tests using the immunofluores-
cent antibody test (IFAT) and the enzyme-linked
immunosorbent assay (ELISA) have proved useful in
the detection of sub-clinical cases and field surveys
(Birkenheuer et al., 2003a). However, IFAT is not
is usually the case in epidemiological surveys. More-
over, the results of IFAT may be influenced by the
subjective judgment of the operator (Bose et al., 1995).
On the other hand, ELISA is quite sensitive and is
appropriate for testing large number of samples
especially in field surveys (Weiland and Reiter, 1988).
Previously, ELISAs have been evaluated for detection of
antibodies to Babesia parasites using native antigens.
Although, such tests proved to be powerful tools for
serological surveys, the poor quality of the antigens, and
sometimes cross-reactions between the Babesia species
in false positive results have limited their application
(Bose et al., 1995). In contrast, previous studies have
shown that recombinant antigens provide better options
and offer higher specific activity than corresponding
native antigens thus avoiding the problem of false
positive results experienced with native antigens (Tebele
et al., 2000; Boonchit et al., 2006). To date, the
recombinant proteins such as rBgP50, BgBSA1 and
Abogeetal., 2007). However, their sensitivities have not
achieved perfect result necessitating further research on
development of novel candidate diagnostic antigens.
In this regard, we immunoscreened cDNA expres-
sion library constructed from B. gibsoni merozoite and
isolated a novel cDNA encoding 57-kDa protein
(BgP57) that shared no homology with any of the
apicomplexan parasites. The successful expression of a
57-kDa recombinant merozoite protein (rBgP57) was
demonstrated in E. coli and the corresponding 57-kDa
native merozoite protein of the parasite was identified.
An indirect ELISA using the rBgP57 antigen detected
specific antibodies of B. gibsoni in both experimentally
and naturally infected dog sera and compared favorably
with the other previously identified recombinant
antigens. Moreover, the rBgP57-ELISA was shown to
be suitable for serosurvey using sera from B. gibsoni-
naturally infected dogs, and that the indirect ELISA as
well as a seminested PCR, when used in combination
could improve epidemiological surveys and clinical
diagnosis of B. gibsoni infection.
2. Materials and methods
2.1. Experimental animals
Two 1-year-old female beagle dogs, which are free
of natural B. gibsoni-infection and five 6-week-old
female ddY mice were used in the study. The
experimental animals were housed, fed and given clean
drinking water in accordance with the stipulated rules
for the Care and Use of Research Animals Promulgated
by the Obihiro University of Agriculture and Veterinary
Medicine, Hokkaido, Japan.
B. gibsoni NRCPD strain (Fukumoto et al., 2001a)
previously isolated in our laboratory was maintained in
splenectomized beagle dogs. B. gibsoni-parasitized
erythrocytes were processed as described in our
previous study (Aboge et al., 2007).
2.3. Immunoscreening of cDNA expression library
A cDNA expression library of B. gibsoni merozoites
was previously constructed as described by Fukumoto
et al. (2001a). Briefly, the cDNA was synthesized by
using a Zap-cDNA synthesis kit, as per manufacturer’s
instructions (Stratagene, San Diego, California). The
cDNA expression library (107PFU) was immu-
noscreened with the serum from a B. gibsoni-infected
dog and the cDNA inserts of positive clones were
sequenced using an automated sequencer (ABI PRISM
3100 Genetic Analyzer, USA). Complete nucleotide
sequences of two identical cDNAs were analyzed using
basic local alignment search tool (BLAST) accessed
through the National Center for Biotechnology Infor-
hydropathic plot of the protein was determined using
approach (Kyte and Doolittle, 1982). The presence
and location of a putative N-terminal signal peptide in
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–94 86
the BgP57 sequence was predicted using the SignalP
2.4. Preparation of B. gibsoni genomic DNA and
The genomic DNA was prepared from merozoite
extracts by standard methods of proteinase K digestion,
phenol–chloroform extraction, and ethanol precipitation
as described by Fukumoto et al. (2001b). For Southern
blot analysis, the B. gibsoni genomic DNAwas digested
with SacI, AccI, HindIII and NdeI restriction enzymes
and then electrophorosed in 0.7% agarose gel. The DNA
was transferred to a nylon membrane (Hybond-N+,
Amersham-Buchler, Munich, Germany) using a method
described by Sambrook and Russell (2001). Preparation
stringency washes, as well as signal generation and
detectionwere performed using AlkPhos Direct labeling
kit (Amersham Biosciences, USA).
2.5. Analysis of intron presence in the genomic DNA
The genomic DNA of B. gibsoni was amplified by
PCR using forward and reverse primers described
below. The amplicon was electrophorosed using 1.0%
agarose gel and purified by geneclean kit. The PCR
product was ligated into pGEM-T vector (Promega
Corporation Madison, USA) and sequenced to confirm
the presence of introns.
2.6. Cloning of BgP57 gene into pGEX-4T-3 vector
One pair of oligonucleotide primers including the
EcoRI and XhoI restriction enzyme sites was designed
and used to clone the truncated gene encoding rBgP57
CAC-30; reverse primer, 50-ATCTCGAGCTTAGGAT-
into the EcoRI and XhoI restriction enzyme sites of the
pGEX-4T-3 E. coli expression vector (Amersham
Pharmacia Biotech, Piscataway, NJ). The construct of
resulting plasmid was checked for accurate insertion by
restriction enzyme digestion and nucleotide sequencing.
2.7. Expression and purification of the rBgP57 in
The rBgP57 was expressed as a GST-fusion protein
in the E. coli BL21 (DE3) strain according to the
manufacturer’s instructions (Amersham Pharmacia
Biotech). Supernatants containing the soluble rBgP57
were purified with glutathione-Sepharose 4B beads
(Amersham Pharmacia Biotech) according to the
manufacturer’s instructions. The rBgP57 fused with
GSTwas cleaved with thrombin protease to remove the
GST as described by the manufacturer’s instructions
(Amersham Biosciences, Corp. USA).
2.8. Preparation of mouse anti-rBgP57 immune
Mouse anti-rBgP57 antibody was generated in five 6-
purified rBgP57 without GST tag was administered
followed by two additional boosters with 250 mg of the
same protein emulsified with 0.25 ml of incomplete
Freud’s adjuvant. Serum samples were collected 14 days
after the last booster and stored at ?30 8C until use.
2.9. Sodium dodecyl sulphate polyacrylamide gel
electrophoresis (SDS-PAGE) and Western blotting
To identify the native BgP57 on the parasite, the
extracts of B. gibsoni merozoite and normal dog
erythrocytes lysates were sonicated, and then precipi-
tated with acetone. Thereafter, mouse anti-rBgP57
serum was used to analyze the merozoite extract by
described (Zhou et al., 2006). In addition, the rBgP57
was analyzed by SDS-PAGE as previously described
(Zhou et al., 2006) and processed in the sameway as for
merozoite extract. The rBgp57 was probed with B.
gibsoni-infected dog serum and non-infected dog
serum, respectively. Subsequent procedures were
performed as described above for the native BgP57.
2.10. Indirect fluorescent antibody test (IFAT)
B. gibsoni-infected dog erythrocytes fixed on IFAT
slides were reacted with mouse anti-rBgP57 serum and
observed under fluorescent microscope as previously
described by Fukumoto et al. (2001b). Alexa-Fluor
4881-conjugated to goat anti-mouse immunoglobulin G
2.11. Preparation of rBgP32 and rBgP50 for
Cloning and expression of rBgP50 and rBgP32
antigens were previously performed in our laboratory
(Verdida et al., 2004; Aboge et al., 2007).
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–9487
2.12. Serum samples
Dog serum samples used in this study were as
follows: 30 sera from specific pathogen-free (SPF) dogs
(Nihonnosan, Japan); sequential serum samples (0–541
days post-infection) from a dog experimentally infected
with B. gibsoni NRCPD strain; 2 sera each from dogs
experimentally infected with B. canis canis, B. canis
vogeli, and B. canis rossi; 2 sera from dogs
experimentally infected with Leishmania infantum; 2
sera from dogs infected with Neospora caninum and 2
sera from Toxoplasma gondii infected mice.
2.13. Clinical examination and blood samples
One hundred and seven samples (blood and serum)
were collected from domestic dogs, comprising males
and females of different breeds and ages, admitted to
eight different veterinary hospitals in Japan. Each dog
was examined for the clinical signs consistent with B.
gibsoni-infection and questionnaires containing general
information about each dog (i.e., breed, gender, age and
origin) were completed. Sample collections from the
hospitals were done between April and October 2006.
The duration between the time clinical symptoms were
first detected and the time the samples were collected
ranged from 1 day to 9 years. Based on the
questionnaires information, 49 dogs that showed
clinical signs consistent with B. gibsoni-infection even
with previous treatment for the infection were classified
as ‘‘symptomatic dogs’’ while 54 dogs that did not
reveal the clinical signs following the treatments were
referred to as ‘‘asymptomatic dogs’’. However, four
dogs could not be classified into the two categories. In
addition, 30 sera and blood samples were collected at
the veterinary hospitals from dogs shown by clinical
examination, microscopic and PCR findings to be free
of B. gibsoni-infection.
The parameters of indirect ELISA development and
validation including diagnostic sensitivity (Dse) as well
as diagnostic specificity (Dsp) are described in detailed
in our previous paper (Aboge et al., 2007). Briefly, cut-
off values were selected for each of the recombinant
antigens used in this study (rBgP57, rBgP32 and
rBgP50). The results of known reference sera were
classified as true positive (TP) or true negative (TN) if
they are in agreement with those of reference sera.
Alternatively, false positive (FP) or false negative (FN)
included samples, which disagree with results of serum
samples with known infection status. Diagnostic
sensitivity was calculated as TP/(TP + FN) whereas
diagnostic specificity was TN/(TN + FP) and the results
of both calculations were expressed as percentages.
The respective recombinant antigens and GST were
diluted with a 0.05 M carbonate–bicarbonate buffer (pH
9.6) as ELISA antigens to a final concentration of 4 mg/
ml for rBgP57 and 5 mg/ml for rBgP32 and rBgP50.
Each well of 96-well plates (Nunc-Immuno Plate;
Nunc, Roskilde, Denmark) was coated with 50 ml of
each of the antigens overnight at 4 8C. The subsequent
protocols were performed as previously described by
Boonchit et al. (2002). Horseradish peroxidase-con-
jugated to goat anti-dog IgG (Cappel, Durham, N.C.)
antibody was used as secondary antibody.
2.15. PCR analysis and determination of its
One hundred and seven blood samples for PCR
analysis were collected from dogs as described in
Section 2.13 above. The samples were analyzed using a
seminested PCR in order to detect DNA of B. gibsoni
using a method previously described by Birkenheuer
et al. (2003b). The sensitivity of PCR was determined
using samples with known infection status as described
elsewhere (Aboge et al., 2007), using the same method
as for ELISA.
2.16. Statistical analysis
The data was statistically analyzed using S-plus 6
software for windows (Insightful Corporation, Seattle
Washington, USA). Summary statistics were obtained
by cross-tabulations of categorical data and statistically
significant differences in the proportions of positive
samples between each of the recombinant antigens as
well as seminested PCR were determined using the
McNemar’s chi-square tests. The test results were
considered significantly different when P-value < 0.05.
3.1. Cloning of a cDNA encoding the BgP57
The cDNA expression library (107PFU) of B.
gibsoni was screened with infected dog serum. The
partial cDNA sequences of two hundred positive clones
were isolated and then subjected to BLAST analysis.
Two identical cDNA sequences, which shared no
homology with any of the apicomplexan parasites, were
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–94 88
selected for molecular characterization. The length of
nucleotide sequence of the cDNAwas 2387 bp; with an
ORF of 1644 bp from bases, 89 to 1732 bp (Fig. 1). The
ORF encoded a mature predicted polypeptide consist-
ing of 547 amino acid residues with a predicted size of
57-kDa as determined by computer analysis.
Analysis of the putative N-terminal signal peptide in
the BgP57 sequence using SignalP-server revealed that
this part of the sequence had a high-predicted signal
peptide probability (0.998) and a maximum cleavage
site probability of 0.495 between amino acids in
positions 19 and 20. The nucleotide sequence data
reported in this paper are available in the GenBank
nucleotide sequence databases with the accession
number EF455059. BLAST analysis of the predicted
polypeptide sequence of rBgP57 against all non-
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–9489
Fig. 1. Complete nucleotide sequence, including the 50-and 30-untranslated regions of the cDNA encoding the BgP57. The predicted amino acid
sequence translated from the ORF is shown below each codon. The underlined amino acids at the 50orientation of the sequence show the N-terminal
predicted signal peptide.
redundant databases accessed through NCBI did not
reveal significant homology with closely related
apicomplexan parasites. Computer based Kyte–Doolit-
tle’splotofthe predicted polypeptide revealed thatmost
regions within the amino acid sequence had a predicted
good antigenic index of 1.7 (data not shown).
3.2. Characterization of BgP57 gene
The cDNA clone encoding BgP57 was used to probe
B. gibsoni genomic DNA fragments digested with
BamHI, AccI, HindIII and NdeI using Southern
blotting. The cDNA probe hybridized to the DNA
resulting in a single band with BamHI (lane 1) and AccI
(lane 2) that do not cut within the probe sequence. On
the other hand, the cDNA probe hybridized to the
digested DNA resulting in two bands with HindIII (lane
3) and NdeI (lane 4) that cut once within the probe
sequence (Fig. 2).
3.3. Characterization of BgP57 on the parasite
The rBgP57 lacking N-terminal signal peptide
sequence was expressed in E. coli as a soluble GST
fusion protein and the amount was about 8.0 mg/l of the
culture. Moreover, B. gibsoni-infected dog serum
reacted with the rBgP57 on Western blot analysis
resulting in a specific band of 57-kDa without GSTand
83-kDa with GST (data not shown). The merozoite
extract of B. gibsoni was analyzed by Western blotting
usingmouse antiserum against the rBgP57. As shownin
extracts of B. gibsoni and the molecular weight of the
native protein was consistent with the predicted
molecular weight of mature polypeptide as well as
that of the expressed rBgP57. The mouse anti-rBgP57
serum did not react with erythrocyte lysates of B.
gibsoni non-infected dog. On the other hand, the B.
gibsoni merozoites reacted strongly with mouse anti-
rBgP57 serum by IFAT (data not shown).
3.4. Diagnosis of B. gibsoni infection by ELISA
We determined the optimal concentrations and
dilutions of the antigen and antibodies to be used in
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–94 90
Fig. 2. Southern blot analysis of the BgP57 gene of B. gibsoni. The
genomic DNA (15 mg per lane) from extract of B. gibsoni merozoite
was digested with, BamHI (lane 1), AccI (lane 2), HindIII (lane 3) and
NdeI (lane 4), and then probed with BgP57 cDNA. The molecular
sizes (in kbp) of the specific DNA bands that hybridized with the
cDNA are shown on the right-hand side by the arrowheads.
Fig. 3. SDS-PAGE and Western blot analysis of native protein of
rBgP57 in B. gibsoni merozoite. (A) Lanes M and 1, low molecular
weight marker and B. gibsoni merozoite extract, respectively, on CBB
staining; (B) lane 2 shows a 57-kDa specific band due to the reactions
of anti-rBgP57 mouse antiserum with extract of B. gibsoni merozoite,
and lane 3 reveals no reaction between the antiserum and normal dog
of rBgP57 antigen coated into each well of the 96-well
plate was 0.20 mg. The optimal dilutions of serum
samples and enzyme-antibody conjugate used in this
assay were 1/100 and 1/2500, respectively. The cut-off
OD value for ELISA using the rBgP57 antigen was
0.200. A dog experimentally infected with B. gibsoni
developed detectable level of antibody response to the
rBgP57 by 10 days post-infection. The antibody titer
was maintained until 541 days post-infection, even
when the dog attained a chronic stage of infection as
evidenced by a significantly low level of parasitemia
(data not shown). In contrast, the parasite DNA was
detected in blood as early as day 2 post infection using
In addition, the rBgP57 reacted specifically with
no cross-reaction with all sera from dogs experimentally
infected with N. caninum and L. infantum as well as
mouse anti-T. gondii serum suggesting that this
recombinant protein was a specific antigen for detection
of antibodies to B. gibsoni in dogs. Serum samples from
dogs confirmed to be free of B. gibsoni-infection by
microscopy and PCR did not react with rBgP57 on
ELISA (data not shown). The indirect ELISA based on
the rBgP57 had diagnostic sensitivity and specificity of
80% and 83%, respectively. The results of assay
development and optimization including diagnostic
sensitivities and specificities of rBgP50 as well as
sensitivity of the seminested PCR determined using
samples with known infection status was 97%.
One hundred and seven sera from dogs clinically
diagnosed as having B. gibsoni infection at eight
indirect ELISA based on rBgP57, rBgP32 and rBgP50.
Overall, the rBgP57-ELISA detected 78 (73%) positive
samples for B. gibsoni antibodies followed closely by
rBgP32-ELISA, which detected 77 (72%) samples and
then rBgP50-ELISA that detected 71 (66%) positive
samples. The seminested PCR detected B. gibsoni DNA
in 88 (82%) samples out of the 107 blood samples
analyzed. When the results of rBgP57-ELISA were
merged together with those of the rBgP32-ELISA and
rBgP50-ELISA, the proportions of positive samples
increasedfrom 73 to83%thus compared favorably with
seminested PCR results. As shown in Table 1, 11
(10.3%) samples, which were positive by rBgP57-
ELISA, were negative by rBgP32-ELISA while 10
(9.3%) samples, which were positive by rBgP32-
ELISA, were negative for rBgP57-ELISA. In the case
of rBgP50, 14 (13.1%) samples that tested negative
were positive for rBgP57, while 7 (6.5%) samples that
revealed positive results by rBgP50-ELISA were
negative when rBgP57 was used as antigen. Addition-
ally, 10 out of 19 samples that were negative by PCR
tested positive for B. gibsoni antibodies when analyzed
by rBgP57-ELISA. Generally, the PCR negative
samples revealed strong seropositive reactions, with
five samples having OD-values of 1.0 and above (data
not shown). On the other hand, some 20 samples, which
were seronegative by rBgP57-ELISA tested positive for
parasite DNA by seminested PCR analysis.
proportions of positive samples than those of the
ELISAs, the Mcnemar’s chi-square test showed that
there was no significant difference in relative effec-
tiveness of rBgP57-ELISA and seminested PCR in
identifying positive samples (x2= 2.7000; P = 0.1003).
In addition, the results of rBgP32-ELISA revealed
similar findings when matched with those of the PCR
(x2= 2.857; P = 0.0910). In contrast, the relative
effectiveness of rBgP50-ELISA in identifying the
positive samples when compared with seminested
PCR were significantly different (x2= 6.918; P =
0.0085). As shown in Table 2, the rBgP57-ELISA
detected higher proportion (79.6%) of positive samples
than seminested PCR (72.2%) in samples collected from
dogs, which were asymptomatic and considered to have
difference was not statistically significant (x2= 0.8150:
P = 0.3680). However, the rBgP57-ELISA detected
lower proportion (69.4%) of positive samples than
seminested PCR (91.8%) in samples collected from 49
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–9491
The results of matched paired samples for rBgP57-ELISAs with those of rBgP32-ELISA, rBgP50-ELISA and seminested PCR. The number and
percentages of positive (+) and negative (?) samples for 107 sera are shown below
rBgP57-ELISArBgP32-ELISArBgP50-ELISA Seminested PCR
A novel cDNA encoding B. gibsoni 57-kDa
merozoite protein that shared no homology with any
of the apicomplexan parasites was isolated. Southern
blotting result suggested that the BgP57 gene exists as a
single copy in the genome of B. gibsoni. In addition,
sequence comparison of the genomic DNA fragment
with the cDNA encoding BgP57 revealed that the
genomic DNA consisted of two exons (124 bp at the 50
end and 1520 bp at the 30end) interrupted by a 38 bp
TTCAG) intron between 213th and 214th bases within
the ORF of this sequence.
Computer based Kyte–Doolittle’s hydropathy ana-
lysis of the predicted polypeptide of rBgP57 demon-
strated that the amino acid sequence had a hydrophilic
core region with a good antigenic index suggesting
that the antigen could be a good candidate for
detection of B. gibsoni antibodies. Moreover, analysis
of predicted polypeptide of rBgP57 using SignalP-3.0
server revealed that this protein had a predicted
signal peptide indicating that it might be a secretary
protein or surface membrane protein. Previous
studies have shown that similar polypeptide sequences
with a predicted signal peptide and good antigenic
index are promising diagnostic candidates (Fukumoto
et al., 2003; Aboge et al., 2007). Therefore, we
hypothesized that the rBgP57 could be a good
ELISA using the rBgP57 demonstrated that the
recombinant protein detected B. gibsoni-specific anti-
bodies in serial serum samples of a dog experimentally
infected with the parasite starting from 10 days up to
541 days post-infection. It has been reported that
antibodies usually take 8–10 days to develop in B.
gibsoni infection (Boozer and Macintire, 2003). The
observation appear to be in agreement with the finding
of the current study, in which specific antibodies were
first detected as from 10 days post-infection although
other antigens have shown early detection time like 8
days (Fukumoto et al., 2003; Jia et al., 2006; Aboge
et al., 2007). Cross-reactivity resulting in B. gibsoni-
infected animals showing low-levels of positive titers
for B. canis has been documented (Yamane et al., 1993,
is important to differentiate the more common closely
related B. canis infections and other closely related
apicomplexan parasite infections from B. gibsoni
infection. In the current study, the rBgP57 reacted
specifically with B. gibsoni antibodies but not with B.
canis sub-species, N. caninum and L. infantum
antibodies. Similar finding was reported in the case
of other recombinant antigens of B. gibsoni shown to be
promising diagnostic antigen (Fukumoto et al., 2003;
Jia et al., 2006), indicating that the novel recombinant
antigen could be a potential specific antigen for
detection of B. gibsoni antibodies.
The proportions of seropositive samples from B.
gibsoni-naturally infected dogs increased from 73 to
83% when the results of rBgP57-ELISA were merged
together with those of the rBgP32 and rBgP50-ELISAs
thus compared favorably with seminested PCR results
(82%), suggesting that the sensitivities of the ELISA
could be greatly improved if the antigens are used in
combination. In addition, the rBgP57-ELISA detected
antibodies against the parasite in nine samples that
were PCR negative (Table 1), with five samples having
OD values of 1.0 and above. Either this could imply
low level of parasite DNA below detection limit of the
PCR or the persistence of the parasite antibodies for a
period even after the living pathogen has been
eliminated from the host. In this respect, this ELISA
B. gibsoni infection by monitoring their antibody
new infections rather than in treatment of clinical
infection, which require specific methods of parasite
samples tested positive for parasite DNA by semi-
nested PCR suggesting that the samples could have
been from dogs with early parasitic infection char-
acterized by undetectable antibody titers. Alterna-
tively, this could have been due to the relatively higher
sensitivity of seminested PCR compared with the
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–94 92
The results of rBgP57-ELISA and seminested PCR of positive (+) and negative (?) sera collected from asymptomatic and symptomatic dogs after
Seminested PCR rBgP57-ELISA
The rBgP57-ELISA detected higher proportion of
positive samples than seminested PCR in samples
collected from apparently healthy dogs thought to have
been cured after treatment with the drugs against canine
babesiosis (Table 2). From epidemiological point of
view, screening for sub-clinical or latent infections is
important because such dogs could serve as potential
source of B. gibsoni infection (Stegeman et al., 2003);
moreover, cases of relapses of infection are possible.
Therefore, we suggest that the rBgP57-ELISA could
prove useful in screening surveys and in identification
of infected dogs that appear apparently healthy. Such
screening surveys might prove valuable in preventing
blood transfusion associated transmission of B. gibsoni
infection by infected dogs, which appear apparently
healthy but still harbor the parasite (Stegeman et al.,
2003). In contrast, seminested PCR detected higher
proportion of positive samples than rBgP57-ELISA in
samples collected from symptomatic dogs (Table 2). In
veterinary clinics, confirmatory diagnosis of canine
babesiosis should be as specific as possible to achieve
good response to treatment. In this respect, seminested
PCR could be useful especially in cases where
microscopic findings are inconclusive. Nevertheless,
the ELISA based on rBgP57 and seminested PCR are
two sides of the same coin, in that each test has its own
advantages and disadvantages depending on the
intended use. Therefore, the findings suggest that while
the rBgP57-ELISA appeared to be suitable for
serosurveys, the seminested PCR could be useful in
confirmatory diagnosis including detection of early
parasitic infections. Overall, the observations from this
study imply that the rBgP57-ELISA and seminested
PCR, when used in combination could improve
epidemiological surveys and clinical diagnosis of B.
The study describes isolation and characterization of
kDa recombinant merozoite protein as GST fusion
protein, and identification of the corresponding 57-kDa
native protein of the parasite was demonstrated. The
diagnostic performance of ELISA based on rBgP57
using sera from B. gibsoni-naturally infected dogs was
the same as the previously identified rBgP32 but
performed better than the other previously identified
rBgP50. Overall, epidemiological surveys and clinical
diagnosis of B. gibsoni infection could be improved
when the rBgP57-ELISA and seminested PCR are used
in combination. We concluded that the 57-kDa antigen
is a candidate antigen for epidemiological survey and
probably diagnosis of B. gibsoni infection.
This work was funded by a grant from the Ministry
of Education, Culture, Sports, Science, and Technology
ofJapan.Theauthors thankDrs.Y.Kitano, K.Ishimaru,
N. Ide, N. Suzaki, C. Sato, H. Sakai, S. Hayashi and K.
Tuchihashi for providing canine serum samples.
Aboge, G.O., Jia, H., Kuriki, K., Zhou, J., Nishikawa, Y., Igarashi, I.,
Fujizaki, K., Suzuki, H., Xuan, X., 2007. Molecular characteriza-
better diagnostic performance by enzyme linked immunosorbent
assay. Parasitology 23, 1–10.
Ano, H., Makimura, S., Harasawa, R., 2001. Detection of Babesia
species from infected dog blood by polymerase chain reaction. J.
Vet. Med. Sci. 63, 111–113.
Birkenheuer, A.J., Levy, M.G., Breitschwerdt, E.B., 2003b. Develop-
ment and evaluation of a seminested PCR for detection and
differentiation of Babesia gibsoni (Asian genotype) and B. canis
DNA in canine blood samples. J. Clin. Microbiol. 41, 4172–4177.
Birkenheuer, A.J., Levy, M.G., Stebbins, M., Poore, M., Breitsch-
werdt, E., 2003a. Serosurvey of anti-Babesia antibodies in stray
dogs and American pit bull terriers and American Staffordshire
terriers from North Carolina. J. Am. Anim. Hosp. Assoc. 39, 551–
Boonchit, S., Alhassan, A., Chan, B., Xuan, X., Yokoyama, N.,
Ooshiro, M., Goff, W.L., Waghela, S.D., Wagner, G., Igarashi,
I., 2006. Expression of C-terminal truncated and full-length
Babesia bigemina rhoptry-associated protein 1 and their potential
use in enzyme-linked immunosorbent assay. Vet. Parasitol. 137,
Boonchit, S., Xuan, X., Yokoyama, N., Goff, W.L., Wagner, G.,
assay with recombinant rhoptry-associated protein 1 antigen
against Babesia bovis for the detection of specific antibodies in
cattle. J. Clin. Microbiol. 40, 3771–3775.
Boozer, L.A., Macintire, D.K., 2003. Canine babesiosis. Vet. Clin.
North Am. Small Anim. Pract. 33, 885–904.
Bose, A., Jorgensen, W.K., Dalgliesh, Friedhoff, K.T., de Vos, A.J.,
Vet. Parasitol. 57, 61–74.
Conrad, P., Thomford, J., Yamane, I., Whiting, J., Bosma, L., Uno, T.,
Holshuh, H.J., Shelly, S., 1991. Hemolytic anemia caused by
Babesia gibsoni infection in dogs. J. Am. Vet. Med. Assoc.
Fukumoto, S., Xuan, X., Kadota, K., Igarashi, I., Sugimoto, C.,
Fujisaki, K., Nagasawa, H., Mikami, T., Suzuki, H., 2003.
High-level expression of truncated surface antigen P50 of Babesia
gibsoni in insect cells by baculovirus and evaluation of its immu-
nogenicity and antigenicity. Clin. Diagn. Lab. Immunol. 10, 596–
Fukumoto, S., Xuan, X., Nishikawa, Y., Inoue, N., Igarashi, I.,
Nagasawa, H., Fujisaki, K., Mikami, T., 2001a. Identification
and expression of a 50-kilodalton surface antigen of Babesia
gibsoni and evaluation of its diagnostic potential in an enzyme-
linked immunosorbent assay. J. Clin. Microbiol. 39, 2603–2609.
Fukumoto, S., Xuan, X., Shigeno, S., Kimbita, E., Igarashi, I.,
Nagasawa, H., Fujisaki, K., Mikami, T., 2001b. Development
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–9493
of a polymerase chain reaction method for diagnosing Babesia
infection in dogs. J. Vet. Med. Sci. 63, 97–981.
Jia, H., Zhou, J., Ikada, H., Matsuu, A., Suzuki, H., Garashi, I.,
Fujusaki, K., Xuan, X., 2006. Identification of a novel gene
encoding a secreted antigen of Babesia gibsoni and evaluation
of its use in serodiagnosis. Am. J. Trop. Med. Hyg. 75, 843–
Kjemtrup, A.M., Kocan, A.A., Whitworth, L., Meinkoth, J., Birken-
heuer, A.J., Cummings, J., Boudreaux, M.K., Stockman, S.L.,
Irizarry-Rovira, A., Conrad, P.A., 2000. There are at least three
genetically distinct small piroplasms from dogs. Int. J. Parasitol.
Kyte, J., Doolittle, R.F., 1982. A simple method for displaying
the hydropathic character of a protein. J. Mol. Biol. 157,
Sambrook, J., Russell, D.W., 2001. Molecular Cloning—A Labora-
harbor, NY, pp. 6.33–6.58.
Stegeman, J.R., Birkenheuer, A.J., Kruger, J.M., Breitschwerdt, E.B.,
Am. Vet. Med. Assoc. 222, 959–963.
Tebele, N., Skilton, R.A., Katende, J., Wells, C.W., Nene, V., McEl-
wain, T., Morzaria, S.P., Musoke, J.A., 2000. Cloning, character-
ization, and expression of a 200-kilodalton diagnostic antigen of
Babesia bigemina. J. Clin. Microbiol. 38, 2240–2247.
Terkawi, M.A., Jia, H., Zhou, J., Lee, E.G., Igarashi, I., Fujisaki, K.,
Nishikawa, Y., Xuan, X., 2006. Babesia gibsoni ribosomal phos-
phoprotein Po induces cross-protective immunity against B.
microti infection in mice. Vaccine 25, 2027–2035.
Verdida, R.A., Hara, O.A., Xuan, X., Fukumoto, S., Igarashi, I.,
Zhang, S., Dong,J., Inokuma, H., Kabeya, H., Sato, Y., Moritomo,
of Babesia gibsoni infection in dogs by an improved enzyme-
linked immunosorbent assay with recombinant truncated P50. J.
Vet. Med. Sci. 66, 1517–1521.
Weiland, G., Reiter, I., 1988. Methods for serological response to
Babesia. In: Ristic, M. (Ed.), Babesiosis of Domestic Animals and
Man. CRC Press, Boca Raton, Florida, pp. 143–158.
Yamane, I., Gardner, I.A., Ryan, C.P., Levy, M., Urrico, J., Conrad,
P.A., 1994. Serosurvey of Babesia canis, Babesia gibsoni and
Erhlichia canis in pound dogs in California, USA. Prev. Vet. Med.
Yamane, I., Thomford, J.W., Gardner, I.A., Dubey, J.P., Levy, M.,
Res. 54, 1579–1584.
Lin, J., Xuan, X., 2006. Characterization of the Babesia gibsoni
P18 as homologue of thrombospondin related adhesive protein.
Mol. Biochem. Parasitol. 148, 190–198.
G.O. Aboge et al./Veterinary Parasitology 149 (2007) 85–94 94