Nested polymerase chain reaction for detection of Theileria annulata and comparison with conventional diagnostic techniques: Its use in epidemiology studies

Abstract

In this work we studied the ability of a nested polymerase chain reaction (PCR) to detect Theileria annulata, the causative agent of Mediterranean theileriosis, in blood samples obtained from cattle on farms in different Spanish regions and its possible use in epidemiology studies. Of the 214 samples analyzed, 78.04%, 69.86%, and 62.26% were found to be positive by nested PCR, indirect immunofluorescent antibody test, and optical microscopy of Giemsa-stained smears, respectively. The three techniques were in agreement in 68.6% of the results. The observation that the prevalence of Mediterranean theileriosis estimated using nested PCR alone (70.3%) and that obtained using all three diagnostic techniques together (80.4%) did not significantly differ verifies the utility of this technique in epidemiology studies.

Full text

SHORT COMMUNICATION
J. Martõ
Â
n-Sa
Â
nchez á J. Viseras á F.J. Adroher
P. Garcõ
Â
a-Ferna
Â
ndez
Nested polymerase chain reaction for detection of
Theileria annulata
and comparison with conventional diagnostic techniques:
its use in epidemiology studies
Received: 30 June 1998 / Accepted: 15 August 1998
Abstract In this work we studied the ability of a nested
polymerase chain reaction (PCR) to detect Theileria
annulata, the causative agent of Mediterranean theile-
riosis, in blood samples obtained from cattle on farms in
dierent Spanish regions and its possible use in epide-
miology studies. Of the 214 samples analyzed, 78.04%,
69.86%, and 62.26% were found to be positive by nested
PCR, indirect immuno¯uorescent antibody test, and
optical microscopy of Giemsa-stained smears, respec-
tively. The three techniques were in agreement in 68.6%
of the results. The observation that the prevalence of
Mediterranean theileriosis estimated using nested PCR
alone (70.3%) and that obtained using all three diag-
nostic techniques together (80.4%) did not signi®cantly
dier veri®es the utility of this technique in epidemiol-
ogy studies.
Tropical or Mediterranean theileriosis (MT) is a disease
that aects bovine cattle within a wide geographic area
extending from the Mediterranean littoral regions of
Europe and Africa, the Near and Middle East, and part
of the former Soviet Union to India and China (Dolan
1989). In epidemiology studies, diagnosis is usually
made by direct visualization with an optical microscope
of the parasite in Giemsa-stained samples (OM) and by
serology tests such as the indirect immuno¯uorescent
antibody test (IFAT). Among the drawbacks of these
techniques is the low sensitivity of OM, especially in the
case of low levels of parasitemia, a common situation
when the previously infected animals later become car-
riers of the parasites. For the IFAT it has been reported
that the antibodies tend to disappear in long-term car-
riers in spite of the persistence of piroplasms, and the
infection can be con®rmed by xenodiagnosis (Bishop
et al. 1992; Dolan 1986).
Polymerase chain reaction (PCR) oers important
advantages such as the greater sensitivity and speci®city
over conventional techniques that has been veri®ed in a
number of studies performed on a wide range of para-
sites (Andersen et al. 1996; Bishop et al. 1992; Tahar
et al. 1997).
The aim of this work was to examine the feasibility of
using the PCR in epidemiology studies of MT and of its
application in Spain. Recently, D'Oliveira et al. (1995)
have described a PCR technique that speci®cally am-
pli®es DNA of Theileria annulata. These authors use
cDNA probe hybridization (Tams-1 gene encoding the
30-kDa major merozoite surface antigen of T. annulata)
to con®rm the ampli®ed product and increase the sen-
sitivity. Since we did not have the cDNA probe, we
carried out a second PCR using internal primers derived
from the nucleotide sequence of the Tams-1 gene (Shiels
et al. 1994, 1995).
A total of 214 blood samples obtained from cattle on
farms in the Alicante, Badajoz, Ca
Â
diz, Co
Â
rdoba, Gra-
nada, and Sevilla provinces in the South of Spain were
tested for the presence of T. annulata using three diag-
nostic techniques: nested PCR, IFAT, and OM. All 3
diagnostic techniques were used on 207 of the 214
samples, whereas 5 were processed by nested PCR and
IFAT and the 2 remaining samples were tested by nested
PCR and OM.
Of the 214 samples analyzed, 26 were obtained from
®ghting bulls taken to Granada to be used in bull®ghts
and the other 188 were obtained from dairy cattle of
several breeds. Of these , 79 samples consisted of thin
blood smears that had been maint ained at room tem-
perature for several years (from sampling in 1993, 1994,
and 1996). The remaining 135 samples consisted of
Parasitol Res (1999) 85: 243±245 Ó Springer-Verlag 1999
J. Martõ
Â
n-Sa
Â
nchez á J. Viseras á P. Garcõ
Â
a-Ferna
Â
ndez
Departamento de Produccio
Â
n Animal, CIFA,
E-18071 Granada, Spain
J. Martõ
Â
n-Sa
Â
nchez (&) á J. Viseras á F.J. Adroher
Departamento de Parasitologõ
Â
a, Facultad de Farmacia,
Campus Universitario de Cartuja,
E-18071 Granada, Spain
E-mail: fparasi@ucartuja.ugr.es,
Tel.: +34-958-243857, Fax: +34-958-243862

blood samples collected in test tubes with ethylene
diaminetetraacetic acid (EDTA)-K
3
that either were
immediately used for DNA extraction or had been
conserved at )20 °C until processing; all of these had
been collected during 1997 except for 12, which had been
conserved since 1991.
Four oligonucleotide PCR primers speci®c for
T. annulata were used for the ne sted PCR. Primers were
derived from the Tams1-1 gene encoding the 30-kDa
major T. annulata merozoite surface antigen (D'Oliveira
et al. 1995; Shiels et al. 1994, 1995). The primers N516
and N517 were used in the ®rst ampli®cation reaction
under the conditions described by D'Oliveira et al.
(1995). The primers Ta 300: 5¢CACCTCAACATA-
CCCC3¢ and Ta 750: 5¢TGACCCACTTATCGTCC3¢
(selected using the PRIME program of software from
the Genetic Computer Group) were used in the second
PCR. The ®nal reaction volume was 50 ll and contained
16 mM (NH
4
)
2
SO
4
,67mM TRIS-HCl (pH 8.8), 0.01%
Tween 20, 1 mM MgCl
2
, 200 lM of each deoxynu-
cleoside triphosphate, 20 ng of each primer, and 1 llof
ampli®ed DNA in the ®rst PCR. The ampli®cation
conditions included 40 cycles at 94 °C/50 s, 60 °C/30 s,
and 72 °C/50 s followed by ®nal extension at 72 °C for
5 min. The ®nal PCR products were resolved by 1%
agarose gel electrophoresis with TBE buer (0.9 M
TRIS, 0.9 M boric acid, 20 mM EDTA). A visible band
at 721 bp in the ®rst PCR or 453 bp in the second PCR
was considered a positive result (D'Oliveira et al. 1995; the
nucleotide sequence of Tams1-1 is available in the EMBL
and GenBank data bases under the accession number
U22887). Repeat ampli®cations with internal primer
pairs veri®ed that the ampli®ed product was derived from
the targeted T. annulata gene (results not shown).
Microscope examination of the Giemsa-stained
smears and the serology study with the IFAT were
performed by the usual methods (Burridge et al. 1974;
Callow and Pepper 1974; Garcõ
Â
a-Ferna
Â
ndez et al. 1996;
Viseras et al. 1997). Results were analyzed statistically
using the v
2
test.
The results obtained when all three diagnostic tech-
niques were used are recorded in Table 1. In all, 78.04%
of the samples analyzed were found to be positive using
the nested PCR, 68.86% were positive in the IFAT, and
62.26% were positive in OM. Signi®cant dierences
were found betwe en the positivities obtained by OM and
those obtained with the nested PCR (P < 0.01). In
contrast, the dierences observed between the positivi-
ties obtained by OM and the IFAT were not signi®cant
(P > 0.1). Moreover, there was no signi®cant dierence
between the number of positive results obtained with the
IFAT and the number obtained with the nested PCR
(P > 0.1).
The nested PCR was more sensitive for the detection
of the parasite than was either OM or the serology
technique of indirect immuno¯uorescence; when the
nested PCR served as the de®nition of infection the
sensitivity of the IFAT was 80.1% and that of OM was
78.3%. When microscopy served as the de®nition of
infection the sensitivity of the nested PCR was 98.4%
and that of the IFAT wa s 89.6%. The three techniques
were in agreement in 68.6% of the results. All the
samples in which piroplasms were observed by OM (132)
also gave a positive res ult in the nested PCR, except for
3 blood-smear samples; in one case, repetition of the
DNA preparation process using a smear prepared with a
larger blood volume followed by repetition of the double
ampli®cation yielded positive results and the false neg-
ative was attributed to insucient sample. Repetition of
this procedure was not possible in the other two cases.
The IFAT results were not in agreement with those
obtained with the other 2 techniques in 28 samples (see
Table 1). Of these, the 13 animals with a positive IF AT
and negative PCR and smear presented antibody titers
of 40 (10 animals), 80 (1), and 640 (2). One animal with
an antibody titer of 640 was in clinical remission after
having received speci®c treatment for MT and could be
considered cured of the parasite. Animals with a nega-
tive IFAT and positive ne sted PCR and OM presented
levels of parasitemia lower than 0.3%.
The use of blood smears to obtain DNA for PCR
ampli®cation has been shown to be valid, provided that
a sucient amount of blood is used for the smear. Ad-
vantages of this method include easy transport and du-
rability of the sample, since this technique can be used
on samples that have been conserved for several years.
In 46 of the blood samples analyzed, Anaplas ma spp.
were detected by OM, and 34 samples were diagnosed as
being coinfected with T. annulata and Anaplasma spp.
Of the remai ning 12, considered as negative for T. an-
nulata by OM, 7 had a positive nested PCR (speci®c for
T. annulata), and the latter test was negative in the re-
maining 5 cases. The 7 samples with a positive nested
PCR also tested as positive in the IFAT, which once
again con®rms the speci®city of the technique.
For calculation of the prevalence of MT, only sam-
ples randomly collected in the sampling period 1996±
1997 were used. These consisted of 148 samples from the
regions of Ca
Â
diz (26), Co
Â
rdoba (78), and Granada (44).
When we considered a result to be positive when at least
one of the techniques gave a positive result, the mean
value of the prevalence in the study area was 80.4%. In
contrast, the prevalence values obtained when each
Table 1 Results obtained using all three diagnostic techniques:
nested PCR, IFAT, and OM of Giemsa-stained smears
Nested
+
Nested
)
Total
OM
+
111 2 113
IFAT
+
OM
+
15 0 15
IFAT
)
OM
)
18 13 31
IFAT
+
OM
)
17 31 48
IFAT
)
Total 161 46 207
244

technique was considered separately were 56.2% for
OM, 66.9% for the IFAT, and 70.3% for the nested
PCR. There was no signi®cant dierence between the
prevalence values obtained when the nested PCR
(P > 0.1) and the IFAT (P > 0.05) were used inde-
pendently and the value obt ained when the three tech-
niques were used together. However, when the sole
diagnostic technique used was microscopic observation
of the smear, signi®cant dierences were obtained
(P < 0.001). The percentages of positivity reported by
other authors using the IFAT and OM on samples from
dierent Spanish regions reach values of 76.9% and even
90% (Brandau et al. 1989; Garcõ
Â
a-Ferna
Â
ndez et al.
1996), although in other studies using only one technique
these values are considerably lower (Garcõ
Â
a-Ferna
Â
ndez
et al. 1985, 1987; Habela et al. 1993). In accordance with
these authors and taking into account that we used an
additional and more sensitive diagnostic technique, we
would have expected to obtain higher percentages of
positivity. We consider this discrepancy to be due to the
dierent epidemiological characteristics of the provinces
included in our study. The Granada province has been
considered by several authors as a hypoendemic zone
with a seroprevalence of 13.71% (Habela et al. 1993),
whereas the Ca
Â
diz province is considered to be a
hyperendemic zone (Garcõ
Â
a-Ferna
Â
ndez et al. 1985,
1987). In our case the prevalence values obtained in the
provinces using the nested PCR technique ranged from
100% in Ca
Â
diz to 31.8% in Granada.
References
Andersen K, Gaafar A, El Hassan AM, Ismail A, Dafalla M,
Theander TG, Kharazmi A (1996) Evaluation of the polymer-
ase chain reaction in the diagnosis of cutaneous leishmaniasis
due to Leishmania major: a comparison with direct microscopy
of smears and sections from lesions. Trans R Soc Trop Med
Hyg 90: 133±135
Bishop R, Sohanpal B, Kariuki DP, Young AS, Nene V, Baylis H,
Allsopp BA, Spooner P, Dolan TT, Morzaria SP (1992) De-
tection of a carrier state in Theileria parva infected cattle by the
polymerase chain reaction. Parasitology 104: 215±232
Brandau C, Monge A, Fermõ
Â
n ML, Jime
Â
nez-Mazzucchelli F,
Tesouro MA (1989) Tratamiento de la theileriosis bovina
con parvacuona y buparvacuona; control y seguimiento. Vet
Rec [Spanish edn]: 279±281
Burridge MJ, Brown CGD, Kimber CD (1974) Theileria annulata:
cross reactions between a cell culture schizont antigen and an-
tigens of East African Theileria species in the indirect ¯uores-
cent antibody test. Exp Parasitol 35: 374±379
Callow LL, Pepper PM (1974) Measurement of and correlations
between fever, changes in the packed cell volume and para-
sitaemia infection with Babesia argentina. Aust Vet J 50: 1±5
Dolan TT (1986) Chemotherapy of East Coast fever: the long term
weight changes, carrier state and disease manifestations of
parvaquone treated cattle. J Comp Pathol 96: 137±146
Dolan TT (1989) Theileriosis: a comprehensive review. Rev Sci
Tech O Int Epizoot 8: 11±36
D'Oliveira C, Weide M, Habela MA, Jacquiet P, Jongejan F (1995)
Detection of Theileria annulata in blood samples of carrier
cattle by PCR. J Clin Microbiol 33: 2665±2669
Garcõ
Â
a-Ferna
Â
ndez P, Romero-Rodrõ
Â
guez J, Hueli LE (1985) Piro-
plasmosis bovinas en Andalucõ
Â
a. I. Estudio en reses procedentes
de mataderos. Rev Iber Parasitol 45: 49±58
Garcõ
Â
a-Ferna
Â
ndez P, Romero-Rodrõ
Â
guez J, Hueli LE (1987) Piro-
plasmosis bovinas en Andalucõ
Â
a. II. Estudio en explotaciones
agropecuarias. Rev Iber Parasitol 47: 7±13
Garcõ
Â
a-Ferna
Â
ndez P, Hueli LE, Viseras-Alarco
Â
n J (1996) Encuesta
epizootiolo
Â
gica de la theileriosis bovina por Theileria annulata
en el toro de lidia. Med Vet 13: 538±542
Habela M, Aranda C, Rol JA, Navarrete I (1993) Seroprevalencia
de la theileriosis Mediterra
Â
nea en ganado bovino de aptitud
lechera en la provincia de Granada (Espan
Ä
a). Acta Parasitol
Port 1: 247
Shiels BR, Smyth A, Dickson J, McKellar S, Tetley L, Fujisaki D,
Hutchinson B, Hinnaird JH (1994) A stoichiometric model of
stage dierentiation in the protozoan parasite Theileria annul-
ata. Mol Cell Dier 2: 101±125
Shiels BR, D'Oliveira C, McKellar S, Ben-Miled L, Kawazu S,
Hide G (1995) Selection of diversity at putative glycosylation
sites in the immunodominant merozoite/piroplasm surface
antigen of Theileria parasites. Mol Biochem Parasitol 72: 149±
162
Tahar R, Ringwald P, Basco LK (1997) Diagnosis of Plasmodium
malariae infection by the polymerase chain reaction. Trans R
Soc Trop Med Hyg 91: 410±411
Viseras J, Garcõ
Â
a-Ferna
Â
ndez P, Adroher FJ (1997) Isolation and
establishment in in vitro culture of a Theileria annulata-infected
cell line from Spain. Parasitol Res 83: 394±396
245