Detection of Babesia bovis in cattle by PCR-ELISA.

Abstract

We established a new highly sensitive method, PCR-ELISA, for the dectection of Babesia bovis in cattle for farms in Thailand. The detection of around 2.4 x 10(-8)% parasitemia (equivalent to 1 infected erythrocyte per 2 ml) was achieved by PCR amplification followed by the ELISA detection of a biotin tagged gene. When comparing the sensitivity of PCR-ELISA with the microscopic method, our PCR-ELISA method is more sensitive than thin blood smears by at least 1,000 times. The established PCR-ELISA also showed high specificity to B. bovis with no cross reaction to other endemic parasites except for A. marginale. Regarding the detection threshold for B. bovis, the PCR-ELISA method could detect parasites inoculated into splenectomized calves at least 1 week earlier than the thin blood microscopic method. The PCR-ELISA method is a valuable screening technique for B. bovis and applicable for the routine detection of carrier states and automated analysis.

Full text

Page 1

DETECTION OF BABESIA BOVIS IN CATTLE
Vol 34 No. 4 December 2003751
Correspondence: Dr Sompong Thammasirirak, Depart-
ment of Biochemistry, Faculty of Science, Khon Kaen
University, 123 Amphoe Muang, Khon Kaen 40002,
Thailand.
Tel/Fax : +66(0) 4334-2911
E-mail: somkly@kku.ac.th
INTRODUCTION
The bovine babesiosis is a tick-borne dis-
ease of cattle caused by an intra-erythrocytic pro-
tozoan parasite of the genus Babesia, namely Ba-
besia bovis (B. bovis) and Babesia bigemina (B.
bigermina). These parasites cause substantial
monetary loss to the cattle industry in Thailand,
as well as other developing countries in Asia,
Africa, South America and many islands of the
Caribbean and South Pacific (Thompson, 1979;
McCosker, 1981; Bram, 1983, Brockelman,
1989). B. bovis is more virulent than B. bigermina
and the babesiosis caused by B. bovis is less sen-
sitive to some babesiacidal compounds making
it difficult to cure the infected animals
(Brockelman and Tan-ariya, 1991; Tan-ariya and
Sarathaphan, 1991). The disease is clinically
manifested by anemia, fever, hemoglobinuria,
marked splenomegaly and hepatomegaly, and
sometimes death. In Thailand, B. bovis causes
mainly babesiosis in cattle or dairy cows, which
leads to heavy economic losses in animal pro-
duction. Animals that survive from B. bovis in-
fection generally become carriers of the parasite
and serve as reservoirs for transmission
(Mahoney, 1969). Vaccination of animals with
live attenuated vaccines has been used to control
babesiosis (Mahoney, 1967; De Vos and Bock,
2002). However, the disadvantages of vaccines
containing parasites derived from the blood of
animals are well known, including the risk of re-
actions or contamination, and sensitization against
blood groups. Because of this and other limita-
tions of vaccines, sensitive methods for the de-
tection of B. bovis carriers or subclinical cattle
are very important. Generally, the standard
method used to diagnose B. bovis is by the detec-
tion of Babesia parasites in a Giemsa-stained thin
blood smear film examined by microscopy. Us-
ing this method, it is very difficult to detect the
parasites in cases of low parasitemia. The current
methods for the diagnosis of bovine babesiosis
are techniques of molecular biology, particularly
polymerase chain reaction (PCR) and DNA hy-
bridization (McLaughlin et al, 1986; Jasmer et
al, 1990; Fahrimal et al, 1992; Petchpoo et al,
1992; Calder et al, 1996; Salem et al, 1999). The
detection of the PCR product by gel electrophore-
sis and Southern blot hybridization has limitations
DETECTION OF BABESIA BOVIS IN CATTLE BY PCR-ELISA
Sompong Thammasirirak1, Jaruwan Siriteptawee1, Nison Sattayasai1, Patchima Indrakamhang2
and Tomohiro Araki3
1Department of Biochemistry, Faculty of Science, Khon Kaen University, KhonKaen;
2National Institute of Animal Health, Bang Khen, Bangkok, Thailand; 3Department of Bioscience,
School of Agriculture, Kyushu Tokai University, Aso, Kumamoto, Japan
Abstract. We established a new highly sensitive method, PCR-ELISA, for the dectection of Babesia
bovis in cattle for farms in Thailand. The detection of around 2.4 x 10-8% parasitemia (equivalent to 1
infected erythrocyte per 2 ml) was achieved by PCR amplification followed by the ELISA detection
of a biotin tagged gene. When comparing the sensitivity of PCR-ELISA with the microscopic method,
our PCR-ELISA method is more sensitive than thin blood smears by at least 1,000 times. The estab-
lished PCR-ELISA also showed high specificity to B. bovis with no cross reaction to other endemic
parasites except for A. marginale. Regarding the detection threshold for B. bovis, the PCR-ELISA
method could detect parasites inoculated into splenectomized calves at least 1 week earlier than the
thin blood microscopic method. The PCR-ELISA method is a valuable screening technique for B.
bovis and applicable for the routine detection of carrier states and automated analysis.

Page 2

SOUTHEAST ASIAN J TROP MED PUBLIC HEALTH
752
Vol 34 No. 4 December 2003
in the number of samples that can be analyzed at
one time. Polymerase chain reaction coupled with
enzyme linked immunosorbent assay (PCR-
ELISA) is a high potential method, which has
been used for the diagnosis of many diseases (He
et al, 1993; Gale et al, 1996; Luk et al, 1997;
Masake et al, 2002). This method has not been
reported to have been used for the diagnosis of B.
bovis infection.
As an initial step to apply the PCR-ELISA
method to the detection of B. bovis for large scale
screening and in an attempt to develop alterna-
tive diagnostic methods for carriers of babesio-
sis, we evaluated the reliability, sensitivity, and
specificity of PCR-ELISA and compared it with
microscopic techniques and gel electrophoresis.
MATERIALS AND METHODS
Parasites used in this study included B. bovis,
B. bigermina, Anaplasma centrale (A. centrale),
Anaplasma marginale (A. marginale), and Try-
panosoma evansi (T. evansi). B. bovis was
subinoculated into splenectomized calves as de-
scribed previously (Callow and Mellors, 1966).
B. bovis (7.5 x 107) in 1 ml of whole blood were
subcutaneously inoculated into the splenecto-
mized calf after it was shown to be healthy by
examination, body temperature 38.2ºC, red blood
cell count 7.34 x 106/mm3, and no antibody titer
to B. bovis detected by the indirect fluorescent
antibody technique (IFA). Routine checking for
body temperature, red blood cell count, packed
cell volume (PCV) and thin blood smear were
taken every other day after inoculation until the
first parasite was found on the thin blood smear.
Fourteen days after inoculation, before the calf
succumbed to death, the body temperature rose
to 40.6ºC, the PCV dropped to 8%, and the para-
sitemia was 1%. The other blood specimens were
collected from cattle naturally infected with four
types of parasites (B. bigermina, A. centrale, A.
marginale, T. evansi). Parasites in the whole blood
were stored in liquid nitrogen until used. The per-
centage of parasitemia was obtained from the
middle of the thin blood smear stained with
Giemsa’s stain and examined in a standard way,
which usually has RBC ≥ 1,000 cells per field at
high magnification. The number of parasites in
twenty microscope fields examined at high mag-
nification was counted. This number was ex-
pressed as a percentage of the total erythrocytes
in the twenties field.
DNA of B. bovis as well as other parasites
was isolated using a high pure PCR template
preparation kit (Boehringer Mannheim Co, Ger-
many). The purified B. bovis DNA from the in-
oculum of the splenectomized calf was used to
determine the sensitivity and specificity of the
method. The pure DNA of B. bovis isolated from
1.5x107 parasites / 200 µl of infectious blood was
diluted in sterile water and used for the PCR tem-
plate.
A pair of nucleotide primers from the Bo6
sequence of the B. bovis Mexican strain (5′-
GGGTTTATAGTCGGTTTTGT-3′ and 5′-
ACCATTCTGGTACTATATGC-3′) which re-
sponded to the apocytochrome b gene were syn-
thesized and the primers were biotinylated at their
5′ end (GIBCO BRL, USA). The PCR condition
was described before with minor modifications
(Fahrimal et al, 1992). Namely, 10 ng of B. bovis
DNA was added to 100 µl of a PCR mixture com-
posed of 1 µM of each biotinylated primer, 200
mM of dATP, dCTP, and dGTP, 130 mM dTTP,
70 mM Dig-11-dUTP and 4 U of Taq DNA poly-
merase (Boehringer Mannheim, Germany) in 10
mM Tris-HCl buffer, pH 8.3 containing 6 mM
MgCl2 and 50 mM KCl. The PCR was taken
through 30 cycles in a DNA thermal cycler
(Cyclogene Cambridge). Conditions for amplifi-
cation were: first cycle of 5 minutes at 94ºC,
(denature), 2 minutes at 55ºC, (annealing), 5
minutes at 72ºC (extension), followed by 28
cycles of 1 minute at 94ºC, 2 minutes at 55ºC, 5
minutes at 72ºC followed by 1 cycle of 1 minute
at 94ºC, 2 minutes at 55ºC, 7 minutes at 72ºC. A
buffer was used as control for checking the con-
tamination of the PCR product.
Detection of PCR products
ELISA. A microtiter plate (96 well plate;
maxisorpTN, Nunc, Kamstrup, Denmark) was
coated with 50 µl of streptavidin (10 µg/ml)
per well in 0.06 M carbonate buffer, pH 9.6 at
4ºC, overnight. Unsaturated binding sites were
blocked with 3% (w/v) skim milk in TST buffer
(40 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.05%

Page 3

DETECTION OF BABESIA BOVIS IN CATTLE
Vol 34 No. 4 December 2003 753
Tween 20) (He et al, 1993). The plates were
then washed with TST buffer followed by the
addition of samples (4 µl of PCR product di-
luted in 50 µl of PBS, which contained 2 µg/
ml of sheared salmon sperm DNA) to each well.
The plate was incubated at 37ºC, for 1 hour.
After the reaction, the plate was subjected to
wash with TST buffer five times. Then 50 µl
of a 1:2,500 dilution of anti-Dig (Fab) conju-
gated with alkaline phosphatase (Boehringer
Mannheim, Germany) was added to each well
and incubated at 37ºC for 30 minutes. The plate
was washed three times with TST buffer, twice
with TS buffer (40 mM Tris-HCl, pH 7.4, 150
mM NaCl) and then equilibrated with detec-
tion buffer for 30 seconds. Substrate solution
(1 mg/ml of p-nitrophenyl phosphate in 0.05M
carbonate buffer, pH 9.8, 0.005 M MgCl2) was
added and the color was developed by adding
100 µl of 1 M NaOH per well. The absorbance
was measured at 405 nm using an ELISA reader
(Labsystems Muttiskan MS).
Agarose gel electrophoresis. Four microliters
of each PCR-amplified solution was loaded onto
each lane of 1% agarose gel. Electrophoresis
of amplified DNA was carried out in TBE buffer
(90 mM Tris-borate, 2 mM EDTA) at a con-
stant voltage of 100 V for 2 hours. Then, the
gel was stained by ethidium bromide and pho-
tographed (Sambrook et al, 1989). Hind III-
digested Lamda DNA was used as a marker for
determining the sizes of the amplified DNA
bands.
Southern blot hybridization. The electro-
phoretically separated DNA was blotted from
gel by capillary transfer onto a nylon mem-
brane (Boehringer Mannheim, Germany). The
DNA was denatured by 0.5 M NaOH, 1.5 M
NaCl. DNA was then crosslinked onto the
membrane by exposure to UV light. The im-
mobilized DNA membrane was placed into a
prehybridization solution (5xSSC, 0.1%
lauroylsarcosine, 0.02% SDS, 1% blocking re-
agent) and incubated at 68ºC for 1 hour. The
hybridization reaction was then carried out in
the fresh hybridization solution containing the
denature probe at 68ºC, overnight. The mem-
brane was washed twice in 2xSSC containing
0.1% SDS at room temperature for 15 minutes
and twice in 0.5xSSC containing 0.1% SDS at
68ºC for 15 minutes. The membrane was then
blocked by blocking solution (Boehringer
Mannheim, Germany) for 30 minutes. The di-
luted solution of 1:5,000 of anti-Dig conjugated
with alkaline phosphatase was added and incu-
bated for 30 minutes at room temperature. Excess
antibody was removed from the membrane by
washing twice with maleic acid buffer (0.1 M
maleic acid, 0.15 M NaCl, pH 7.5), and then
the membrane was equilibrated in detection
buffer (0.1 M Tris-HCl, 0.1 M NaCl, 50 mM
MgCl2, pH 9.5) for 2 minutes. The substrate
solution (NBT and BCIP, Boehringer Mannhein,
Germany) was added and allowed to form the
blue color for 16 hours. The reaction was stopped
by washing the membrane with sterile water
for 5 minutes.
Light microscopic examination. A thin blood
smear of blood samples was fixed with metha-
nol for 1 minute and stained with Giemsa so-
lution for 30 minutes, then the parasites were
determined by light microscopic examination.
RESULTS
Specificity of PCR reaction
There are many parameters in PCR that in-
fluence the specificity, fidelity and yield of the
desired product (Innis and Gelfand, 1990). In this
study the optimal concentrations of those PCR
parameters are 5 ng/ml of B. bovis Thai isolated
DNA, 2.5 mM of MgCl2, 0.5 µM of each primer,
and 1 U of Taq DNA polymerase in a mixture of
50 µl (data not shown). Several optimization pa-
rameters for the ELISA were also studied. We
found that 1 µg of the biotinylated PCR products
labeled-DIG (diluted in 50 µl of dilution buffer)
was sufficient for this ELISA and the optimum
incubation time was at 37ºC for 60 minutes. A
1:2,500 dilution of anti-Dig conjugated with al-
kaline phosphatase was enough to bind to
digoxigenin in the PCR products. The suitable
time for signal development color was 60 min-
utes with 100 µl of 1 mg/ml of pNPP (data not
shown).
The specificity of the PCR was examined

Page 4

SOUTHEAST ASIAN J TROP MED PUBLIC HEALTH
754
Vol 34 No. 4 December 2003
with the DNA extracted from B. bovis inoculated
splenectomized calf erythrocyte. The results (Fig
1) showed that the bands of the expected size of
apocytochrome b gene (740 bp for unincorporated
digoxyginin and 1,040 bp when digoxyginin was
incorporated into the DNA chain) were obtained.
These results coincided with previous results
(Fahrimal et al, 1992).
Determination of the sensitivity of PCR-ELISA
Ten-fold serial dilutions of B. bovis infected
erythrocytes from the inoculated splenectomized
calf (107 to 10-3 parasites) were added to 0.2 ml of
normal bovine blood. The DNA was then ex-
tracted from each diluted sample for testing the
sensitivity of the PCR method. As shown in Fig
2, agarose gel electrophoresis was sensitive in
detecting parasite DNA with equivalence to 1
parasite. The PCR-ELISA and Southern blot hy-
bridization could detect parasite DNA at about
10-2 parasites with equivalence to a parasitemia
of 2.4 X 10–8%.
The species specificity of the PCR
Other hemoparasites can be endemic with B.
bovis, including B. bigermina, A. marginale, and T.
evansi. Their DNA and leukocyte DNA were ex-
amined. Five nanograms of DNA from each spe-
cies was amplified by PCR and then the PCR prod-
10
7
10
6
10
510
4
10
Number of B. bovis
3
10
2
10
1
10
-110
-210
-3cut off
0
0.5
1
1.5
2
2.5
Absorbance 405 nm
A
B
C
ucts were analyzed by agarose gel electrophoresis
and the ELISA method. While the amplified prod-
uct detected was B. bovis DNA, no other PCR prod-
uct was generated when the DNA from other
hemoparasites or leukocytes was tested (Fig 3).
The species specificity of PCR-ELISA was
also analyzed. The results showed a few signals
in ELISA for A. marginale, however, the absor-
bance from B. bovis showed a different intensity.
When the incubation time for the color reaction
was changed to 30 minutes the intensity of color
was eliminated. However, cross-reaction with A.
marginale in PCR-ELISA was observed.
Detection threshold of PCR-ELISA
The blood sample from the experimental
splenectomized calf was used to check the detec-
Fig 2–Detection of sensitivity. The detection of sensi-
tivity of PCR, followed by agarose gel electro-
phoresis (A), Southern blot hybridization (B),
PCR-ELISA (C). Lane M is lambda Hind III
DNA digested marker. Lanes 1-11 are the num-
ber of parasites diluted tenfold from 107-10-3 in
bovine blood 200 µl. The cut-off of ELISA is
0.189 at 405 nm.
Fig 1–Specificity of the PCR method. Ethidium bro-
mide stained agarose gel electrophoresis of PCR
products from B. bovis with non-incorporated
digoxeginin (lane 1), and incorporated
digoxeginin (lane 2). Lane M lambda Hind III
DNA digested marker.

Page 5

DETECTION OF BABESIA BOVIS IN CATTLE
Vol 34 No. 4 December 2003 755
cut off
B. bovis
B. bigemina
T. evansi
bovine
WBC 1
A. centrale
A. marginale
0
0.5
1
1.5
2
2.5
Absorbance 405 nm
A
B
Fig 3–Species specificity of PCR-ELISA. PCR fol-
lowed by agarose gel electrophoresis (A), ELISA
(B). Lane M: lambda Hind III DNA digested
marker; lane 1 B. bovis; lane 2: B. bigermina;
lane 3: T. evansi; lane 4: bovine WBC DNA;
lane 5: A. centrale; lane 6: A. marginale.
A
cut off
pre
0
3
5
10
12
14
0
0.5
1
1.5
2
2.5
Absorbance 405 nm
Days post-infection
B
Fig 4–Detection of the parasite DNA of B. bovis in
splenectomized calf. PCR followed by agarose
gel electrophoresis (A), ELISA (B). Lane M: lam
bda Hind III DNA digested marker; lane 1: two
day prior to inoculation; lane 2: the day of inocu-
lation; lanes 3-7: 3-14 day after inoculation to
splenectomized calf respectively.
tion threshold of the PCR-ELISA. Blood samples
were collected from the calf at two days prior to
subinoculation, and at the days 0, 3, 5, 10, 12,
and 14 after inoculation. From our results, the
PCR-ELISA method detected parasites day 3 af-
ter inoculation of the splenectomized calf. On the
other hand, the thin blood smear technique de-
tected the parasite on day 12 after inoculation (Fig
4).
Applicability of the PCR-ELISA to the de-
tection of B. bovis in blood samples
The PCR-ELISA was used to detect blood
samples of naturally infected cattle from farm and
splenectomized calves, which were maintained
under tick free conditions. The results of PCR-
ELISA were compared with the immunology
method, thin blood smear method, and PCR-gel
electrophoresis. The results are shown in Table
1. PCR-ELISA had good sensitivity and speci-
ficity in detecting two samples of carrier cattle
from six naturally infected samples for which the
blood smear method and PCR-gel electrophore-
sis showed negative results. The splenectomized
calf showed all negative for PCR-ELISA.
DISCUSSION
We have developed a new method, PCR-
ELISA, for the detection of a carrier status of Thai
isolated B. bovis infection. The PCR product of
B. bovis Thai isolated is slightly different from
the PCR product (711 bp) amplified from B. bovis
Mexican, Australia L and S and Texas strains
DNA (Fahrimal et al, 1992). This difference may
be due to the polymorphism of the apocytochrome
b gene. The incorporation of digoxigenin moi-
eties into the PCR products caused the size of the