Infection, Genetics and Evolution 23 (2014) 115–120
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Molecular characterization of the Babesia caballi rap-1 gene and epidemiological survey in horses in Israel Adi Rapoport, Karin Aharonson-Raz, Dalia Berlin, Saar Tal, Yuval Gottlieb, Eyal Klement, Amir Steinman ⇑ Koret School of Veterinary Medicine, The Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, POB 12, Rehovot 76100, Israel
a r t i c l e
i n f o
Article history: Received 8 October 2013 Received in revised form 28 January 2014 Accepted 30 January 2014 Available online 10 February 2014 Keywords: Piroplasmosis Babesia caballi cELISA RAP-1 Horses
a b s t r a c t Equine piroplasmosis imposes great concerns for the equine industry regarding international horse movement, and therefore requires reliable diagnostic tools. Recent studies from South Africa and Jordan, including a preliminary study in Israel, reported extremely low seroprevalence to Babesia caballi (B. caballi) (0–1%) using the acceptable rhoptry-associated protein-1 (RAP-1) cELISA. In accordance with the study from South Africa demonstrating a significant heterogeneity in the rap-1 gene sequence of South African B. caballi isolates, the objectives of this study were to phylogenetically characterize the rap-1 gene of the Israeli isolates and determine the prevalence of B. caballi in horses in Israel. Out of 273 horses tested using the RAP-1 cELISA, only one was sero-positive, while 9.3% were positive on PCR performed on the rap-1 gene. Phylogenetic analysis of the rap-1 gene grouped the Israeli isolates in a cluster together with the South African strains (99% nt identity), but in a separate cluster from the American/ Caribbean strains (81–82% nt identity). These findings support the existence of heterogeneity in the RAP-1 amino-acid sequences of the Israeli and South African isolates as compared to that used in the cELISA commercial kit and raise doubts as to the ability of this assay to serve as a sole regulatory test for international horse movement. Risk factor analysis found management and age to significantly associate with prevalence of B. caballi, as higher prevalence was noted in horses held out on pasture and a negative association was recorded with age. In addition, B. caballi was not detected in horses in the steppe-arid and extreme-arid climatic regions as compared to the wetter regions. Findings of this study emphasize the need to combine several detection methods to ameliorate the control and spread of the disease. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Located at the junction of 3 continents: Europe, Asia and Africa, Israel has a Mediterranean climate ranging from temperate to tropical, and is known to be endemic to many arthropod borne diseases. Equine piroplasmosis (EP) is a tick-borne disease caused by two haemoprotozoan parasites, Babesia caballi (B. caballi) and Theileria equi (T. equi). Previous studies conducted in Israel, using serological and molecular assays, have shown that T. equi is highly prevalent in Israel with a 33.7% seroprevalence and 26.4% prevalence tested by PCR in 1998 and 2002, respectively (Shkap et al., 1998; Steinman et al., 2012). B. caballi is reported annually by the Israeli Veterinary Services and Animal Health, with few serologically positive diagnosed cases a year (Bellaiche, 2013). However, its prevalence in horses in Israel has not been adequately addressed.
⇑ Corresponding author. Tel.: +972 3 9688534; fax: +972 3 9604079. E-mail address: [email protected] (A. Steinman). http://dx.doi.org/10.1016/j.meegid.2014.01.033 1567-1348/Ó 2014 Elsevier B.V. All rights reserved.
Identification of the specific EP agent based on clinical signs alone is not probable due to the similar clinical signs manifested by horses infected with both parasites, including fever, anaemia, jaundice, edema, depression and even death (Knowles, 1996). Mixed infection with both parasites have been documented (RosGarcía et al., 2013) as well as a subclinical carrier state identified for T. equi (de Waal, 1992). Identification of the parasite by means of blood smear examination can provide, in some cases, differentiation between the parasites. Paired merozoites joined at their posterior ends are a diagnostic feature of B. caballi infection (Mehlhorn and Shein, 1984), while a typically smaller four pear-shaped merozoites forming a tetrad known as a ‘Maltese cross’ is a characteristic feature of T. equi (Brüning, 1996). However, the differentiation is difficult, particularly when parasitaemia is low and when mixed infections are encountered. Molecular methods for parasite detection in blood are both sensitive and specific in the clinical stage but may fail to detect the parasite in carrier animals (Ribeiro et al., 2013). Therefore, serological tests are currently the most recommended detection method by the World Organization for Animal Health (OIE) (OIE, 2008). Among those methods is the
116
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
competitive-inhibition enzyme-linked immunosorbent assay (cELISA) which uses T. equi and B. caballi recombinant merozoite proteins and monoclonal antibodies to immunogenic epitopes on these proteins in order to identify the agent (Kappmeyer et al., 1999; Terkawi et al., 2012). A commercial cELISA has been developed using the recombinant B. caballi rhoptry-associated protein 1 (RAP-1), a protein believed to play an essential role in parasite invasion, attachment and expansion of the host cell (Sam-Yellowe, 1996). The above cELISA method was found to be effective in detecting exposure to the parasite in horses from North and South America, Europe and parts of Africa (Kappmeyer et al., 1999) and therefore served as the main method of detection of EP in many studies around the globe, typically showing the existence of both parasites in endemic areas (García-Bocanegra et al., 2012; Kouam et al., 2010; Sevinc et al., 2008). In a preliminary study we conducted using cELISA commercial kit, we found approximately 50% seroprevalence for T. equi, but surprisingly more than 99% of the horses were found sero-negative for B. caballi, an unexpected result in an area endemic to T. equi (unpublished data). A recent study, conducted in South Africa in 2010, has shown that the commercial kit was unable to detect antibodies in horses infected with B. caballi. The authors had succeeded to demonstrate a significant heterogeneity in the rap-1 gene sequences from South African B. caballi isolates, which is reflected in the predicted amino acid sequences, particularly in the carboxy-terminal regions. It was suggested that the inability of the cELISA to detect B. caballi antibodies in positive sera was due to the differences of the rap-1 gene (Bhoora et al., 2010). Accordingly, the objectives of this study were to characterize the B. caballi rap-1 gene obtained from blood of horses in Israel in order to determine whether sequence variations have also led to lack of ability to identify serologically positive individuals by the RAP1-cELISA. Based on these results, the prevalence of B. caballi in Israel was determined together with risk factors analysis.
2. Materials and methods 2.1. Sampled horses A total of 273 blood samples were collected randomly from clinically healthy horses, from 37 farms (ranging from 1–25 horses in each farm) representing the distribution of horses in Israel during the years 2007–2008. Data collected for each horse was breed (Tennessee, Quarterhorse, Arabian, Crossbreeds, Pony, Thoroughbred, Warmbloods), gender, age, type of management (stall, yard and pasture), body condition score and climatic zone. Two more blood samples were collected in 2012 from two sick horses with cytological evidence of B. caballi, intended to serve as positive controls. Blood samples were collected from the jugular vein of each horse into sterile vacuum tubes without an anticoagulant agent for serum retrieval and with EDTA for whole blood.
2.2. Sera preparation and cELISA Sera were obtained from clotted blood samples by centrifugation (3000g for 8 min) and stored at 80 °C until used. Serum samples were tested for the presence of antibodies to B. caballi using a commercial cELISA test kit (VMRD Inc., Pullman, WA, USA) according to the manufacturer’s instructions. Samples associated with inhibition values P40% were considered positive. The optical density values were obtained using an automatic microplate reader (Bio-Tek, Winooski, USA).
2.3. DNA extraction and PCR amplification Out of the 273 samples only 257 were tested again using PCR method as well as the two additional samples collected during 2012. Genomic DNA was extracted from 250 ll of whole blood using a commercial kit (illustra blood genomicPrep Mini Spin Kit, GE Healthcare, UK) according to the manufacture’s instruction. The primers Bc9_RAPF (50 AGCAGTGCTGTATATGTCTGTGTC30 ) and Bc9_RAPR (50 GCTGATGCGATGTGTGTCGTAGG30 ) targeting an approximately 1500 bp fragment containing the rap-1 gene were used in a PCR program described elsewhere (Bhoora et al., 2010). A second pair of primers, Bc9_RAP2F (50 ACTAGCGACCCCAACGCTACTGAC30 ) and Bc9_RAP2R (50 TTGGAGCATGAAGTCCTTCAGC30 ) targeting an approximately 400 bp fragment of the rap-1 gene was used as an internal set for the purpose of sequencing (Bhoora et al., 2010) and as the main primer sets for the molecular survey. PCR cycling included an initial denaturation step at 94 °C for 5 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 62 °C for 30 s, extension at 72 °C for 30 s. This was followed by a final extension step at 72 °C for 7 min. All PCR reactions included a negative control, consisting of the reaction mix and 1 ll of DNase/RNase-free water (Sigma, St. Louis, MO, USA), and a positive control that consisted of a DNA sample from the blood of a horse with clinical EP, which was confirmed by nucleotide sequencing. All PCR reactions were performed in 20 ll of a mixture consisting of 10 ll Taq master mix (Lambda Biotech Inc.), 8 ll DNase/ RNase-free water (Sigma, St. Louis, MO, USA), 1 lM primers and 1 ll DNA template. PCR products were electrophoresed in 1.5% Agarose in Tris–Acetate–EDTA (TAE) buffer, and stained with ethidium Bromide to visualize the amplified DNA fragments under ultraviolet light. 2.4. DNA sequencing PCR products were purified using a PCR purification kit (ExoSAP-IT; USB, Cleveland, OH, USA). All positive PCR products (for both the 400 and the 1500 bp fragments) were sequenced using the above mentioned two sets of primers. DNA sequencing was performed at the Centre for Genomics Technologies, Hebrew University of Jerusalem. Obtained DNA sequences were compared for similarity to sequences in GenBank, using the BLAST program. Five rap-1 nucleotide sequences of the Israeli B. caballi isolates were deposited in the GenBank database under the following accession numbers: KF059875, KF059876, KF059877, KF059878, KF059879. 2.5. Phylogenetic analysis Phylogenetic analysis was performed using rap-1 gene sequences of previously published South African strains (accession numbers: GQ871778, GQ871779, GQ871780), and American/Caribbean strains, retrieved from the USDA (accession numbers: EU669865, AF092736, AB017700), which were compared to the 1500 bp fragment of the 5 new Israeli equine strains obtained in this study. The B. divergens rap-1 gene sequence (accession number: Z49818) was used as an out-group. In DNAbaser (Heracle Biosoft, Romania) sequencing chromatograms were quality-trimmed (trim percent = 65, window size = 16, QV threshold for good base = 21, QV threshold for gap = 22) and subsequently assembled to contigs (Word size = 15, identity percent = 70, Minimum overlap = 15). The contigs and other B. caballi rap-1 sequences were aligned using Muscle (Edgar, 2004) and were trimmed on both ends so all the sequences were of equal length (leaving a length of 981 bp). Phylogenetic analysis of the rap-1 gene in B. caballi was concluded in MEGA5 (Tamura et al., 2011) using a Maximum-likelihood tree based on the Tamura-Nei model. Bootstrap consensus was inferred from 100 replicates.
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
117
2.6. Statistical analysis Statistical analyses were performed using SPSS 17.0Ò software. Prevalence with 95% confidence interval was calculated for each category in each demographic variable (breed, gender, age, management, body condition score and climatic zone). In order to determine the association between bioclimatic zones and prevalence of B. caballi, Israel was divided into its 5 representative climatic regions (from the wettest to driest region): Mediterranean, semi-arid, steppe-arid, desert-arid and extreme-arid (Goldreich, 2003). Statistical significance of the differences in prevalence between these various groups was assessed by the two-sided Chisquare test. Map showing the different prevalence in the different farms was constructed using ArcGIS 9 (version 9.3.1) software. Only the farms comprising 5 or more horses were included in the maps. A logistic regression model was constructed with variables inclusion in the model performed by the stepwise likelihood ratio method. Variables were included in the model according to statistical significance, and p-value for change in the model likelihood ratio after insertion of each additional variable was tested. In each such iteration a p-value of 0.1 were excluded. The association between age and prevalence to B. caballi was also analysed in the different management types separately. 3. Result 3.1. Detection of Babesia caballi antibodies Out of the 273 serum samples collected during 2007–2008, only one horse was found sero-positive for B. Caballi. The positive sample was from a Quarterhorse mare kept on pasture in the north of Israel (Mediterranean region). The two suspected positive cases from 2012 were also found to be sero-negative for B. caballi using the c-ELISA detection kit.
Fig. 1. Phylogenetic tree of the Babesia caballi rap-1 gene. DNA sequences were obtained from GenBank database and compared to five new Israeli equine isolates (ISR01, ISR02, ISR53, ISR490, ISR539). The B. divergens rap-1 gene sequence (accession number Z49818) was used as an out group. Bootstrap consensus was inferred from 100 replicates (bootstrap values are shown next to the branches). Isolates are divided into 2 distinct clusters: A, B and cluster A is further sub-divided into A1 and A2.
3.4. Demographic and climatic risk factors for infection with Babesia caballi B. caballi was detected in blood of horses located in the Mediterranean (21/157, 13.4%) and semi-arid (3/41, 7.3%) climatic regions of Israel but was absent from blood of horses located in the steppearid and extreme arid climatic zones (0/59) (Fig. 2, Table 1). Both the univariate and the multivariate analyses showed a significant association between management type and B. caballi prevalence, with the parasite being significantly more prevalent in horses held
3.2. PCR detection of Babesia caballi Out of the 257 samples tested during 2007–2008, 24 were found to be positive for B. caballi (using primers directed against the 400 bp fragment), demonstrating a prevalence of 9.3%. The one sample that was found positive in the cELISA test was also PCR-positive. Both sampled horses with blood smears suggestive of B. caballi in 2012 were found to be PCR-positive. All 24 positive PCR results were sequenced and B. caballi infection was confirmed in all of them. DNA sequencing of the 400 bp fragments amplified by PCR showed 98–100% sequence similarity to the South African rap-1 gene sequences (GenBank GQ871779.1; GQ871778.1). 3.3. Phylogenetic analysis of the rap-1 gene All five samples that were positive on PCR, amplifying the 1500 bp fragment, and thereafter sequenced, were used to construct the phylogenetic tree of the rap-1 gene. The Israeli rap-1 B. caballi sequences grouped separately from the rap-1 sequences of the American/Caribbean B. caballi strains (cluster B), but in the same cluster with the South African isolates (cluster A). In addition, Israeli isolates could be further divided into two subgroups (A1, A2), consistent with the clades observed by phylogenetic analysis of the rap-1 gene of the South African isolates (Fig. 1). A 99% nucleotide identity was observed between the Israeli isolates and either one of the South African subgroups (B. caballi isolates 9, 13, 443), but only an 81–82% nucleotide identity with the American/Caribbean isolates of cluster B.
Fig. 2. Geographical distribution of the sampled farms, demonstrating B. Caballi prevalence in the 5 different climatic zones: Mediterranean, semi-arid, steppe-arid, desert-arid and extreme arid. Only farms with P5 horses are presented.
118
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
Table 1 The association between B. caballi prevalence and demographic (Management type and age) and climatic factors using Pearson’s chi-square univariate analysis.
Management
Climatic zones
Age (years)
à
Category
Positive/total (%)
Odds ratio
95% CI
P
Pasture Stall Yard
14/43 (32.6) 10/129 (7.8) 0/85 (0)
REF 0.17à 0.00à
0.06–0.48 0.00–0.11
15 years (Table 1). When horses were divided according to the different management type (to control for this factor), horses held out on pasture were shown to be infected with B. caballi only in the younger age groups (0–9 years) while none of the older horses (age groups >9 years) were positive for the parasite. In comparison, in horses held in stalls, the prevalence was similar (7.9% in horses age 0–9 years and 7.8% in horses age >9 years) between these age groups. Breed, gender and body condition score (BCS) were not significantly associated with B. caballi prevalence.
4. Discussion Israel and surrounding countries are considered endemic to EP. Studies from Turkey, Greece and Jordan suggest the presence of both T. equi and B. caballi in these countries (Kouam et al., 2010; Qablan et al., 2013; Sevinc et al., 2008). Previous studies conducted in Israel demonstrated a wide spread exposure to equine theileriosis (Shkap et al., 1998; Steinman et al., 2012). Hence, using a cELISA commercial kit, designed to identify antibodies to B. caballi, we anticipated finding a high percentage of sero-positive horses. However, only one horse was identified as positive. These surprising differences between the sero-prevalence of both parasites in an endemic area, together with repeated reports by veterinarians suggesting the presence of B. caballi parasite in blood smears, as well as annual reports from the Israeli Veterinary Services and Animal Health on serological evidence for B. caballi, raised concerns regarding the accuracy of the current serological results. Following the recently published report regarding differences in the rap-1 gene sequences of South African isolates that might result in inability to recognize the agent by cELISA test (Bhoora et al., 2010), we turned to molecular characterization of the B. caballi rap-1 gene in order to determine presence of sequence variation.
Table 2 Logistic regression analysis for the association between B. caballi prevalence, age, management type and climatic zones.
Age Management – pasture Management – yard Management – stall
B
S.E.
Sig
95% CI
0.209 REF 19.961 1.994
0.083
0.012
0.689–0.955
4206.5 0.587
0.996 0.001
0.00 0.043–0.431
Indeed 24 horses were recognised as positive to B. caballi by PCR, targeting a 400 bp fragment of the South African reported rap-1 gene. A phylogenetic analysis of the rap-1 gene (981 bp) showed a high similarity (99%) between the Israeli isolates and the South Africans ones, and a lower similarity (81–82%) to the sequences reported so far in literature for the American/Caribbean strains. Similar connection between African and Israeli pathogens was also demonstrated with the equine encephalosis virus (EEV), another arthropod-borne disease of horses, which was isolated for the first time in 2009 outside of South Africa (Aharonson-Raz et al., 2011). These Israeli isolates were shown to have a 92% sequence identity to EEV-3 South African reference isolate. Interestingly, although all PCR products clustered with the South African strains and not with the previously reported American/Caribbean strains, the cELISA serology kit was still able to recognize one of the samples. Sequence analysis of the 400 bp fragment from this sample showed high similarity to isolate 443 (GQ871778) which subgrouped separately from the two other published rap-1 sequences of South Africa, supporting the occurrence of two antigenically distinct B. caballi strains in South Africa and Israel. However, except for this one sample, other samples that showed higher sequence similarities to the South Africa isolate 443 were not recognized by the cELISA kit. A similar contradictory result between molecular and serological detection was observed in Jordan, a country sharing a common border with Israel, where the same commercial cELISA kit yielded zero positivity to B. caballi (Abutarbush et al., 2012), while a more recent study using a Multiplex PCR test has indicated a 7.3% positivity (Qablan et al., 2013). Despite the above discrepancies, positive serological cases are reported annually from the Veterinary Services and Animal Health of the Israeli Ministry of Agriculture and Rural Development (Bellaiche, 2013). However, these reports are obtained using the Indirect Immunofluorescence Assay (IFA), in which the entire antigen spectrum of the whole pathogen is available, thus allowing the detection of a variety of polyclonal antibodies. These results may once again support the inability to identify the B. caballi antibodies by RAP1-cELISA in this region. The PCR-prevalence of B. caballi in horses in Israel was found to be 9.3% in the presented study. Only horses located in the Mediterranean and the semi-arid climatic regions were found to be positive. The lack of B. caballi positive horses in the drier regions of Israel (i.e. desert/steppe arid and the extreme arid regions) in our study is not surprising since piroplasmosis is generally known to occur in Mediterranean /tropical climatic regions. The prevalence of ticks on horses in Israel, their species and distribution, are currently not known. Such studies are needed and might better explain the differences in piroplasmosis prevalence in the different areas. Our findings are further sup-
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
ported by a recent study performed in Tunisia, a country located in the Mediterranean region as well, in which from the three bioclimatic zones investigated (Humid, Sub-humid and Semi-arid), neither horses nor ticks were found positive for B. caballi in the Semi-arid zone representing the drier zone (Ros-García et al., 2013). It should be taken into consideration that sampling of the horses was performed throughout a one year period and covered all seasons. Therefore, differences in prevalence between the bioclimatic regions might derive from differences in ticks’ activity throughout the different seasons. However, taken the notion that the B. caballi is likely to be cleared from blood of horses only after 1–4 years (de Waal, 1992), differences in sampling periods throughout one year might not have much influence on the PCR-prevalence results. Both the management type and the age of the horses were found to be significantly associated with prevalence of B. caballi both in the univariate and multivariate analyses. A significantly higher prevalence was found in horses with access to pasture in comparison to horses housed in stalls. This difference was probably due to the higher burden of ticks in pasture. Age was shown to be negatively associated with B. caballi prevalence as an increase in age was correlated with a decrease in prevalence of the parasite. When analysing this association in the two different management types separately, only horses between the ages 0–9 years were shown to be infected with the parasite in pasture while horses held in stalls demonstrated similar prevalence in ages 0–9 years and >9 years (7.9% and 7.8% respectively). Most of the horses that were held out on pasture at the time of the study have been on pasture since an early age. Consequently, they were more likely to be exposed to the parasite at a very early stage in life and thus developed an immune response that perhaps prevented later infection with the parasite. Further studies are needed in order to support this assumption. However, horses held in stalls are usually exposed to pasture later in life, reducing the probability for exposure to the parasite. Those that are not exposed do not develop an immune response and remain susceptible to infection also at an older age. A lifelong carrier state is believed to occur following infection with T. equi (Butler, 2013; Ribeiro et al., 2013). Such a carrier state was not thoroughly investigated for B. caballi but it is believed to be cleared from the blood of horses after 1–4 years (de Waal, 1992). The decrease of prevalence with age that was demonstrated in this study together with the fact that none of the horses older than 9 years of age with access to pasture were found to be positive for the parasite, suggest that, in contrast to T. equi, B. caballi does not remain in its host for a long period of time. Nevertheless, it is possible that a carrier state does exist in horses but the amount of parasite in the blood at this stage is below the detection limits of the PCR test performed in this study. The international character of the equine industry with an increase in intercontinental horse travel, presents a serious risk for the introduction of new pathogens into countries that are free of them. To minimise the risk of importing infected equids from areas in which EP is endemic, it is required that blood from these equids be tested for the presence of antibodies to B. caballi and T. equi before importation. In the United States, a country believed to be free of B. caballi, the official testing method was changed in 2005 from complement fixation test (CFT) to cELISA, due to the low sensitivity of the CFT (USDA, 2009). However taking into account the significant heterogeneity in the rap-1 gene sequences found both in the South African and the Israeli B. caballi strains, as well as recent inconsistencies found between infection and cELISA serological results in Unites States horses, with high false sero-positive results (Awinda et al., 2013), it is possible that equids infected with B. caballi are imported into the United States or other free of disease countries. In addition, it is important to understand the
119
geographical worldwide distribution of the different B. caballi strains and understand whether the sequence variations also produce different clinical manifestation and severities of the disease. In conclusion, despite a 9.3% prevalence found for B. caballi in horses in Israel in this study, seroprevalence tested using RAP-1 cELISA was lower than 1%. Phylogenetic analysis performed on the rap-1 gene grouped the Israeli B. caballi isolates in cluster A together with the South African isolates but separated from the American/Caribbean B. caballi strains. These results further support the notion that RAP-1 cELISA relies on the recognition of epitopes that are not conserved across all B. caballi strains due to sequence heterogeneity and emphasizes the need to combine other detection methods to assure control of disease introduction into non-endemic regions. Conflict of interest The authors declare that no competing financial, or other, interests exist. Acknowledgements We thank Dr. Zohar Paternak for helping with the phylogenetic analysis of the rap-1 gene. References Abutarbush, S.M., Alqawasmeh, D.M., Mukbel, R.M., Al-Majali, A.M., 2012. Equine babesiosis: seroprevalence, risk factors and comparison of different diagnostic methods in Jordan. Transbound Emerg Dis 59, 72–78. Aharonson-Raz, K., Steinman, A., Bumbarov, V., Maan, S., Maan, N.S., Nomikou, K., Batten, C., Potgieter, C., Gottlieb, Y., Mertens, P., Klement, E., 2011. Isolation and phylogenetic grouping of equine encephalosis virus in Israel. Emerg Infect Dis 17, 1883–1886. Awinda, P.O., Mealey, R.H., Williams, L.B., Conrad, P.A., Packham, A.E., Reif, K.E., Grause, J.F., Pelzel-McCluskey, A.M., Chung, C., Bastos, R.G., Kappmeyer, L.S., Howe, D.K., Ness, S.L., Knowles, D.P., Ueti, M.W., 2013. Serum antibodies from a subset of horses positive for Babesia caballi by competitive ELISA demonstrate a protein recognition pattern not consistent with infection. Clin Vaccine Immunol 20, 1752–1757. Bellaiche, Michel. (2013). Annual Reports of the Veterinary Services and Animal Health. Bellaiche, (2013). Ministry of Agriculture and Rural Development (in Hebrew). Retrieved from: . Bhoora, R., Quan, M., Zweygarth, E., Guthrie, A.J., Prinsloo, S.A., Collins, N.E., 2010. Sequence heterogeneity in the gene encoding the rhoptry-associated protein-1 (RAP-1) of Babesia caballi isolates from South Africa. Vet. Parasitol. 169, 279– 288. Brüning, A., 1996. Equine piroplasmosis an update on diagnosis, treatment and prevention. Br. Vet. J. 152, 139–151. Butler, C., 2013. Can Theileria equi be eliminated from carrier horses? Vet. J. 196, 279. de Waal, D.T., 1992. Equine piroplasmosis: a review. Br. Vet. J. 148, 6–14. Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792–1797. García-Bocanegra, I., Arenas-Montes, A., Hernández, E., Adaszek, L., Carbonero, A., Almería, S., Jaén-Téllez, J.A., Gutiérrez-Palomino, P., Arenas, A., 2012. Seroprevalence and risk factors associated with Babesia caballi and Theileria equi infection in equids. Vet. J. 195, 172–178. Goldreich, Y., 2003. The Climate in Israel: observation, research and applications. Kluwer Academic/Plenum Press, New York. Kappmeyer, L.S., Perryman, L.E., Hines, S.A., Baszler, T.V., Katz, J.B., Hennager, S.G., Knowles, D.P., 1999. Detection of equine antibodies to Babesia caballi by recombinant B. caballi rhoptry-associated protein 1 in a competitive-inhibition enzyme-linked immunosorbent assay. J. Clin. Microbiol. 37, 2285–2290. Knowles, D.P., 1996. Control of Babesia equi parasitemia. Parasitol. Today 12, 195– 198. Kouam, M.K., Kantzoura, V., Gajadhar, A.A., Theis, J.H., Papadopoulos, E., Theodoropoulos, G., 2010. Seroprevalence of equine piroplasms and hostrelated factors associated with infection in Greece. Vet. Parasitol. 169, 273–278. Mehlhorn, H., Shein, E., 1984. The piroplasms: life cycle and sexual stages. Adv. Parasitol. 23, 37–103. OIE 2008. Manual of diagnostic tests and vaccines for terrestrial animals. Ch. 2.5.8. Equine piroplasmosis. Qablan, M.A., Oborník, M., Petrzˇelková, K.J., Sloboda, M., Shudiefat, M.F., Horˇín, P., Lukeš, J., Modry´, D., 2013. Infections by Babesia caballi and Theileria equi in
120
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
Jordanian equids: epidemiology and genetic diversity. Parasitology 140, 1096– 1103. Ribeiro, I.B., Câmara, A.C., Bittencourt, M.V., Marçola, T.G., Paludo, G.R., Soto-Blanco, B., 2013. Detection of Theileria equi in spleen and blood of asymptomatic piroplasm carrier horses. Acta Parasitol. 58, 218–222. Ros-García, A., M’ghirbi, Y., Hurtado, A., Bouattour, A., 2013. Prevalence and genetic diversity of piroplasm species in horses and ticks from Tunisia. Infect. Genet. Evol. 17, 33–37. Sam-Yellowe, T.Y., 1996. Rhoptry organelles of the apicomplexa: their role in host cell invasion and intracellular survival. Parasitol. Today 12, 308–316. Sevinc, F., Maden, M., Kumas, C., Sevinc, M., Ekici, O.D., 2008. A comparative study on the prevalence of Theileria equi and Babesia caballi infections in horse subpopulations in Turkey. Vet. Parasitol. 156, 173–177. Shkap, V., Cohen, I., Leibovitz, B., Pipano Savitsky, E., Avni, G., Shofer, S., Giger, U., et al., 1998. Seroprevalence of Babesia equi among horses in Israel using competitive inhibition ELISA and IFA assays. Vet. Parasitol. 76, 251–259.
Steinman, A., Zimmerman, T., Klement, E., Lensky, I.M., Berlin, D., Gottlieb, Y., Baneth, G., 2012. Demographic and environmental risk factors for infection by Theileria equi in 590 horses in Israel. Vet. Parasitol. 187, 558–562. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739. Terkawi, M.A., Alhasan, H., Ueno, A., Ratthanophart, J., Luo, Y., Cao, S., Kamyingkird, K., Aboulaila, M., Youn-Kyoung, G., Nishikawa, Y., Yokoyama, N., Xuan, X., Igarashi, I., 2012. C-Terminal region of 48-kDa rhoptry protein for serological detection of Babesia caballi antibodies in horses. Parasitol. Int. 61, 493–496. USDA. Seroprevalence of equine piroplasmosis disease agents in the United States, 2009. Available from: .
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Molecular characterization of the Babesia caballi rap-1 gene and epidemiological survey in horses in Israel Adi Rapoport, Karin Aharonson-Raz, Dalia Berlin, Saar Tal, Yuval Gottlieb, Eyal Klement, Amir Steinman ⇑ Koret School of Veterinary Medicine, The Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, POB 12, Rehovot 76100, Israel
a r t i c l e
i n f o
Article history: Received 8 October 2013 Received in revised form 28 January 2014 Accepted 30 January 2014 Available online 10 February 2014 Keywords: Piroplasmosis Babesia caballi cELISA RAP-1 Horses
a b s t r a c t Equine piroplasmosis imposes great concerns for the equine industry regarding international horse movement, and therefore requires reliable diagnostic tools. Recent studies from South Africa and Jordan, including a preliminary study in Israel, reported extremely low seroprevalence to Babesia caballi (B. caballi) (0–1%) using the acceptable rhoptry-associated protein-1 (RAP-1) cELISA. In accordance with the study from South Africa demonstrating a significant heterogeneity in the rap-1 gene sequence of South African B. caballi isolates, the objectives of this study were to phylogenetically characterize the rap-1 gene of the Israeli isolates and determine the prevalence of B. caballi in horses in Israel. Out of 273 horses tested using the RAP-1 cELISA, only one was sero-positive, while 9.3% were positive on PCR performed on the rap-1 gene. Phylogenetic analysis of the rap-1 gene grouped the Israeli isolates in a cluster together with the South African strains (99% nt identity), but in a separate cluster from the American/ Caribbean strains (81–82% nt identity). These findings support the existence of heterogeneity in the RAP-1 amino-acid sequences of the Israeli and South African isolates as compared to that used in the cELISA commercial kit and raise doubts as to the ability of this assay to serve as a sole regulatory test for international horse movement. Risk factor analysis found management and age to significantly associate with prevalence of B. caballi, as higher prevalence was noted in horses held out on pasture and a negative association was recorded with age. In addition, B. caballi was not detected in horses in the steppe-arid and extreme-arid climatic regions as compared to the wetter regions. Findings of this study emphasize the need to combine several detection methods to ameliorate the control and spread of the disease. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Located at the junction of 3 continents: Europe, Asia and Africa, Israel has a Mediterranean climate ranging from temperate to tropical, and is known to be endemic to many arthropod borne diseases. Equine piroplasmosis (EP) is a tick-borne disease caused by two haemoprotozoan parasites, Babesia caballi (B. caballi) and Theileria equi (T. equi). Previous studies conducted in Israel, using serological and molecular assays, have shown that T. equi is highly prevalent in Israel with a 33.7% seroprevalence and 26.4% prevalence tested by PCR in 1998 and 2002, respectively (Shkap et al., 1998; Steinman et al., 2012). B. caballi is reported annually by the Israeli Veterinary Services and Animal Health, with few serologically positive diagnosed cases a year (Bellaiche, 2013). However, its prevalence in horses in Israel has not been adequately addressed.
⇑ Corresponding author. Tel.: +972 3 9688534; fax: +972 3 9604079. E-mail address: [email protected] (A. Steinman). http://dx.doi.org/10.1016/j.meegid.2014.01.033 1567-1348/Ó 2014 Elsevier B.V. All rights reserved.
Identification of the specific EP agent based on clinical signs alone is not probable due to the similar clinical signs manifested by horses infected with both parasites, including fever, anaemia, jaundice, edema, depression and even death (Knowles, 1996). Mixed infection with both parasites have been documented (RosGarcía et al., 2013) as well as a subclinical carrier state identified for T. equi (de Waal, 1992). Identification of the parasite by means of blood smear examination can provide, in some cases, differentiation between the parasites. Paired merozoites joined at their posterior ends are a diagnostic feature of B. caballi infection (Mehlhorn and Shein, 1984), while a typically smaller four pear-shaped merozoites forming a tetrad known as a ‘Maltese cross’ is a characteristic feature of T. equi (Brüning, 1996). However, the differentiation is difficult, particularly when parasitaemia is low and when mixed infections are encountered. Molecular methods for parasite detection in blood are both sensitive and specific in the clinical stage but may fail to detect the parasite in carrier animals (Ribeiro et al., 2013). Therefore, serological tests are currently the most recommended detection method by the World Organization for Animal Health (OIE) (OIE, 2008). Among those methods is the
116
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
competitive-inhibition enzyme-linked immunosorbent assay (cELISA) which uses T. equi and B. caballi recombinant merozoite proteins and monoclonal antibodies to immunogenic epitopes on these proteins in order to identify the agent (Kappmeyer et al., 1999; Terkawi et al., 2012). A commercial cELISA has been developed using the recombinant B. caballi rhoptry-associated protein 1 (RAP-1), a protein believed to play an essential role in parasite invasion, attachment and expansion of the host cell (Sam-Yellowe, 1996). The above cELISA method was found to be effective in detecting exposure to the parasite in horses from North and South America, Europe and parts of Africa (Kappmeyer et al., 1999) and therefore served as the main method of detection of EP in many studies around the globe, typically showing the existence of both parasites in endemic areas (García-Bocanegra et al., 2012; Kouam et al., 2010; Sevinc et al., 2008). In a preliminary study we conducted using cELISA commercial kit, we found approximately 50% seroprevalence for T. equi, but surprisingly more than 99% of the horses were found sero-negative for B. caballi, an unexpected result in an area endemic to T. equi (unpublished data). A recent study, conducted in South Africa in 2010, has shown that the commercial kit was unable to detect antibodies in horses infected with B. caballi. The authors had succeeded to demonstrate a significant heterogeneity in the rap-1 gene sequences from South African B. caballi isolates, which is reflected in the predicted amino acid sequences, particularly in the carboxy-terminal regions. It was suggested that the inability of the cELISA to detect B. caballi antibodies in positive sera was due to the differences of the rap-1 gene (Bhoora et al., 2010). Accordingly, the objectives of this study were to characterize the B. caballi rap-1 gene obtained from blood of horses in Israel in order to determine whether sequence variations have also led to lack of ability to identify serologically positive individuals by the RAP1-cELISA. Based on these results, the prevalence of B. caballi in Israel was determined together with risk factors analysis.
2. Materials and methods 2.1. Sampled horses A total of 273 blood samples were collected randomly from clinically healthy horses, from 37 farms (ranging from 1–25 horses in each farm) representing the distribution of horses in Israel during the years 2007–2008. Data collected for each horse was breed (Tennessee, Quarterhorse, Arabian, Crossbreeds, Pony, Thoroughbred, Warmbloods), gender, age, type of management (stall, yard and pasture), body condition score and climatic zone. Two more blood samples were collected in 2012 from two sick horses with cytological evidence of B. caballi, intended to serve as positive controls. Blood samples were collected from the jugular vein of each horse into sterile vacuum tubes without an anticoagulant agent for serum retrieval and with EDTA for whole blood.
2.2. Sera preparation and cELISA Sera were obtained from clotted blood samples by centrifugation (3000g for 8 min) and stored at 80 °C until used. Serum samples were tested for the presence of antibodies to B. caballi using a commercial cELISA test kit (VMRD Inc., Pullman, WA, USA) according to the manufacturer’s instructions. Samples associated with inhibition values P40% were considered positive. The optical density values were obtained using an automatic microplate reader (Bio-Tek, Winooski, USA).
2.3. DNA extraction and PCR amplification Out of the 273 samples only 257 were tested again using PCR method as well as the two additional samples collected during 2012. Genomic DNA was extracted from 250 ll of whole blood using a commercial kit (illustra blood genomicPrep Mini Spin Kit, GE Healthcare, UK) according to the manufacture’s instruction. The primers Bc9_RAPF (50 AGCAGTGCTGTATATGTCTGTGTC30 ) and Bc9_RAPR (50 GCTGATGCGATGTGTGTCGTAGG30 ) targeting an approximately 1500 bp fragment containing the rap-1 gene were used in a PCR program described elsewhere (Bhoora et al., 2010). A second pair of primers, Bc9_RAP2F (50 ACTAGCGACCCCAACGCTACTGAC30 ) and Bc9_RAP2R (50 TTGGAGCATGAAGTCCTTCAGC30 ) targeting an approximately 400 bp fragment of the rap-1 gene was used as an internal set for the purpose of sequencing (Bhoora et al., 2010) and as the main primer sets for the molecular survey. PCR cycling included an initial denaturation step at 94 °C for 5 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 62 °C for 30 s, extension at 72 °C for 30 s. This was followed by a final extension step at 72 °C for 7 min. All PCR reactions included a negative control, consisting of the reaction mix and 1 ll of DNase/RNase-free water (Sigma, St. Louis, MO, USA), and a positive control that consisted of a DNA sample from the blood of a horse with clinical EP, which was confirmed by nucleotide sequencing. All PCR reactions were performed in 20 ll of a mixture consisting of 10 ll Taq master mix (Lambda Biotech Inc.), 8 ll DNase/ RNase-free water (Sigma, St. Louis, MO, USA), 1 lM primers and 1 ll DNA template. PCR products were electrophoresed in 1.5% Agarose in Tris–Acetate–EDTA (TAE) buffer, and stained with ethidium Bromide to visualize the amplified DNA fragments under ultraviolet light. 2.4. DNA sequencing PCR products were purified using a PCR purification kit (ExoSAP-IT; USB, Cleveland, OH, USA). All positive PCR products (for both the 400 and the 1500 bp fragments) were sequenced using the above mentioned two sets of primers. DNA sequencing was performed at the Centre for Genomics Technologies, Hebrew University of Jerusalem. Obtained DNA sequences were compared for similarity to sequences in GenBank, using the BLAST program. Five rap-1 nucleotide sequences of the Israeli B. caballi isolates were deposited in the GenBank database under the following accession numbers: KF059875, KF059876, KF059877, KF059878, KF059879. 2.5. Phylogenetic analysis Phylogenetic analysis was performed using rap-1 gene sequences of previously published South African strains (accession numbers: GQ871778, GQ871779, GQ871780), and American/Caribbean strains, retrieved from the USDA (accession numbers: EU669865, AF092736, AB017700), which were compared to the 1500 bp fragment of the 5 new Israeli equine strains obtained in this study. The B. divergens rap-1 gene sequence (accession number: Z49818) was used as an out-group. In DNAbaser (Heracle Biosoft, Romania) sequencing chromatograms were quality-trimmed (trim percent = 65, window size = 16, QV threshold for good base = 21, QV threshold for gap = 22) and subsequently assembled to contigs (Word size = 15, identity percent = 70, Minimum overlap = 15). The contigs and other B. caballi rap-1 sequences were aligned using Muscle (Edgar, 2004) and were trimmed on both ends so all the sequences were of equal length (leaving a length of 981 bp). Phylogenetic analysis of the rap-1 gene in B. caballi was concluded in MEGA5 (Tamura et al., 2011) using a Maximum-likelihood tree based on the Tamura-Nei model. Bootstrap consensus was inferred from 100 replicates.
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
117
2.6. Statistical analysis Statistical analyses were performed using SPSS 17.0Ò software. Prevalence with 95% confidence interval was calculated for each category in each demographic variable (breed, gender, age, management, body condition score and climatic zone). In order to determine the association between bioclimatic zones and prevalence of B. caballi, Israel was divided into its 5 representative climatic regions (from the wettest to driest region): Mediterranean, semi-arid, steppe-arid, desert-arid and extreme-arid (Goldreich, 2003). Statistical significance of the differences in prevalence between these various groups was assessed by the two-sided Chisquare test. Map showing the different prevalence in the different farms was constructed using ArcGIS 9 (version 9.3.1) software. Only the farms comprising 5 or more horses were included in the maps. A logistic regression model was constructed with variables inclusion in the model performed by the stepwise likelihood ratio method. Variables were included in the model according to statistical significance, and p-value for change in the model likelihood ratio after insertion of each additional variable was tested. In each such iteration a p-value of 0.1 were excluded. The association between age and prevalence to B. caballi was also analysed in the different management types separately. 3. Result 3.1. Detection of Babesia caballi antibodies Out of the 273 serum samples collected during 2007–2008, only one horse was found sero-positive for B. Caballi. The positive sample was from a Quarterhorse mare kept on pasture in the north of Israel (Mediterranean region). The two suspected positive cases from 2012 were also found to be sero-negative for B. caballi using the c-ELISA detection kit.
Fig. 1. Phylogenetic tree of the Babesia caballi rap-1 gene. DNA sequences were obtained from GenBank database and compared to five new Israeli equine isolates (ISR01, ISR02, ISR53, ISR490, ISR539). The B. divergens rap-1 gene sequence (accession number Z49818) was used as an out group. Bootstrap consensus was inferred from 100 replicates (bootstrap values are shown next to the branches). Isolates are divided into 2 distinct clusters: A, B and cluster A is further sub-divided into A1 and A2.
3.4. Demographic and climatic risk factors for infection with Babesia caballi B. caballi was detected in blood of horses located in the Mediterranean (21/157, 13.4%) and semi-arid (3/41, 7.3%) climatic regions of Israel but was absent from blood of horses located in the steppearid and extreme arid climatic zones (0/59) (Fig. 2, Table 1). Both the univariate and the multivariate analyses showed a significant association between management type and B. caballi prevalence, with the parasite being significantly more prevalent in horses held
3.2. PCR detection of Babesia caballi Out of the 257 samples tested during 2007–2008, 24 were found to be positive for B. caballi (using primers directed against the 400 bp fragment), demonstrating a prevalence of 9.3%. The one sample that was found positive in the cELISA test was also PCR-positive. Both sampled horses with blood smears suggestive of B. caballi in 2012 were found to be PCR-positive. All 24 positive PCR results were sequenced and B. caballi infection was confirmed in all of them. DNA sequencing of the 400 bp fragments amplified by PCR showed 98–100% sequence similarity to the South African rap-1 gene sequences (GenBank GQ871779.1; GQ871778.1). 3.3. Phylogenetic analysis of the rap-1 gene All five samples that were positive on PCR, amplifying the 1500 bp fragment, and thereafter sequenced, were used to construct the phylogenetic tree of the rap-1 gene. The Israeli rap-1 B. caballi sequences grouped separately from the rap-1 sequences of the American/Caribbean B. caballi strains (cluster B), but in the same cluster with the South African isolates (cluster A). In addition, Israeli isolates could be further divided into two subgroups (A1, A2), consistent with the clades observed by phylogenetic analysis of the rap-1 gene of the South African isolates (Fig. 1). A 99% nucleotide identity was observed between the Israeli isolates and either one of the South African subgroups (B. caballi isolates 9, 13, 443), but only an 81–82% nucleotide identity with the American/Caribbean isolates of cluster B.
Fig. 2. Geographical distribution of the sampled farms, demonstrating B. Caballi prevalence in the 5 different climatic zones: Mediterranean, semi-arid, steppe-arid, desert-arid and extreme arid. Only farms with P5 horses are presented.
118
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
Table 1 The association between B. caballi prevalence and demographic (Management type and age) and climatic factors using Pearson’s chi-square univariate analysis.
Management
Climatic zones
Age (years)
à
Category
Positive/total (%)
Odds ratio
95% CI
P
Pasture Stall Yard
14/43 (32.6) 10/129 (7.8) 0/85 (0)
REF 0.17à 0.00à
0.06–0.48 0.00–0.11
15 years (Table 1). When horses were divided according to the different management type (to control for this factor), horses held out on pasture were shown to be infected with B. caballi only in the younger age groups (0–9 years) while none of the older horses (age groups >9 years) were positive for the parasite. In comparison, in horses held in stalls, the prevalence was similar (7.9% in horses age 0–9 years and 7.8% in horses age >9 years) between these age groups. Breed, gender and body condition score (BCS) were not significantly associated with B. caballi prevalence.
4. Discussion Israel and surrounding countries are considered endemic to EP. Studies from Turkey, Greece and Jordan suggest the presence of both T. equi and B. caballi in these countries (Kouam et al., 2010; Qablan et al., 2013; Sevinc et al., 2008). Previous studies conducted in Israel demonstrated a wide spread exposure to equine theileriosis (Shkap et al., 1998; Steinman et al., 2012). Hence, using a cELISA commercial kit, designed to identify antibodies to B. caballi, we anticipated finding a high percentage of sero-positive horses. However, only one horse was identified as positive. These surprising differences between the sero-prevalence of both parasites in an endemic area, together with repeated reports by veterinarians suggesting the presence of B. caballi parasite in blood smears, as well as annual reports from the Israeli Veterinary Services and Animal Health on serological evidence for B. caballi, raised concerns regarding the accuracy of the current serological results. Following the recently published report regarding differences in the rap-1 gene sequences of South African isolates that might result in inability to recognize the agent by cELISA test (Bhoora et al., 2010), we turned to molecular characterization of the B. caballi rap-1 gene in order to determine presence of sequence variation.
Table 2 Logistic regression analysis for the association between B. caballi prevalence, age, management type and climatic zones.
Age Management – pasture Management – yard Management – stall
B
S.E.
Sig
95% CI
0.209 REF 19.961 1.994
0.083
0.012
0.689–0.955
4206.5 0.587
0.996 0.001
0.00 0.043–0.431
Indeed 24 horses were recognised as positive to B. caballi by PCR, targeting a 400 bp fragment of the South African reported rap-1 gene. A phylogenetic analysis of the rap-1 gene (981 bp) showed a high similarity (99%) between the Israeli isolates and the South Africans ones, and a lower similarity (81–82%) to the sequences reported so far in literature for the American/Caribbean strains. Similar connection between African and Israeli pathogens was also demonstrated with the equine encephalosis virus (EEV), another arthropod-borne disease of horses, which was isolated for the first time in 2009 outside of South Africa (Aharonson-Raz et al., 2011). These Israeli isolates were shown to have a 92% sequence identity to EEV-3 South African reference isolate. Interestingly, although all PCR products clustered with the South African strains and not with the previously reported American/Caribbean strains, the cELISA serology kit was still able to recognize one of the samples. Sequence analysis of the 400 bp fragment from this sample showed high similarity to isolate 443 (GQ871778) which subgrouped separately from the two other published rap-1 sequences of South Africa, supporting the occurrence of two antigenically distinct B. caballi strains in South Africa and Israel. However, except for this one sample, other samples that showed higher sequence similarities to the South Africa isolate 443 were not recognized by the cELISA kit. A similar contradictory result between molecular and serological detection was observed in Jordan, a country sharing a common border with Israel, where the same commercial cELISA kit yielded zero positivity to B. caballi (Abutarbush et al., 2012), while a more recent study using a Multiplex PCR test has indicated a 7.3% positivity (Qablan et al., 2013). Despite the above discrepancies, positive serological cases are reported annually from the Veterinary Services and Animal Health of the Israeli Ministry of Agriculture and Rural Development (Bellaiche, 2013). However, these reports are obtained using the Indirect Immunofluorescence Assay (IFA), in which the entire antigen spectrum of the whole pathogen is available, thus allowing the detection of a variety of polyclonal antibodies. These results may once again support the inability to identify the B. caballi antibodies by RAP1-cELISA in this region. The PCR-prevalence of B. caballi in horses in Israel was found to be 9.3% in the presented study. Only horses located in the Mediterranean and the semi-arid climatic regions were found to be positive. The lack of B. caballi positive horses in the drier regions of Israel (i.e. desert/steppe arid and the extreme arid regions) in our study is not surprising since piroplasmosis is generally known to occur in Mediterranean /tropical climatic regions. The prevalence of ticks on horses in Israel, their species and distribution, are currently not known. Such studies are needed and might better explain the differences in piroplasmosis prevalence in the different areas. Our findings are further sup-
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
ported by a recent study performed in Tunisia, a country located in the Mediterranean region as well, in which from the three bioclimatic zones investigated (Humid, Sub-humid and Semi-arid), neither horses nor ticks were found positive for B. caballi in the Semi-arid zone representing the drier zone (Ros-García et al., 2013). It should be taken into consideration that sampling of the horses was performed throughout a one year period and covered all seasons. Therefore, differences in prevalence between the bioclimatic regions might derive from differences in ticks’ activity throughout the different seasons. However, taken the notion that the B. caballi is likely to be cleared from blood of horses only after 1–4 years (de Waal, 1992), differences in sampling periods throughout one year might not have much influence on the PCR-prevalence results. Both the management type and the age of the horses were found to be significantly associated with prevalence of B. caballi both in the univariate and multivariate analyses. A significantly higher prevalence was found in horses with access to pasture in comparison to horses housed in stalls. This difference was probably due to the higher burden of ticks in pasture. Age was shown to be negatively associated with B. caballi prevalence as an increase in age was correlated with a decrease in prevalence of the parasite. When analysing this association in the two different management types separately, only horses between the ages 0–9 years were shown to be infected with the parasite in pasture while horses held in stalls demonstrated similar prevalence in ages 0–9 years and >9 years (7.9% and 7.8% respectively). Most of the horses that were held out on pasture at the time of the study have been on pasture since an early age. Consequently, they were more likely to be exposed to the parasite at a very early stage in life and thus developed an immune response that perhaps prevented later infection with the parasite. Further studies are needed in order to support this assumption. However, horses held in stalls are usually exposed to pasture later in life, reducing the probability for exposure to the parasite. Those that are not exposed do not develop an immune response and remain susceptible to infection also at an older age. A lifelong carrier state is believed to occur following infection with T. equi (Butler, 2013; Ribeiro et al., 2013). Such a carrier state was not thoroughly investigated for B. caballi but it is believed to be cleared from the blood of horses after 1–4 years (de Waal, 1992). The decrease of prevalence with age that was demonstrated in this study together with the fact that none of the horses older than 9 years of age with access to pasture were found to be positive for the parasite, suggest that, in contrast to T. equi, B. caballi does not remain in its host for a long period of time. Nevertheless, it is possible that a carrier state does exist in horses but the amount of parasite in the blood at this stage is below the detection limits of the PCR test performed in this study. The international character of the equine industry with an increase in intercontinental horse travel, presents a serious risk for the introduction of new pathogens into countries that are free of them. To minimise the risk of importing infected equids from areas in which EP is endemic, it is required that blood from these equids be tested for the presence of antibodies to B. caballi and T. equi before importation. In the United States, a country believed to be free of B. caballi, the official testing method was changed in 2005 from complement fixation test (CFT) to cELISA, due to the low sensitivity of the CFT (USDA, 2009). However taking into account the significant heterogeneity in the rap-1 gene sequences found both in the South African and the Israeli B. caballi strains, as well as recent inconsistencies found between infection and cELISA serological results in Unites States horses, with high false sero-positive results (Awinda et al., 2013), it is possible that equids infected with B. caballi are imported into the United States or other free of disease countries. In addition, it is important to understand the
119
geographical worldwide distribution of the different B. caballi strains and understand whether the sequence variations also produce different clinical manifestation and severities of the disease. In conclusion, despite a 9.3% prevalence found for B. caballi in horses in Israel in this study, seroprevalence tested using RAP-1 cELISA was lower than 1%. Phylogenetic analysis performed on the rap-1 gene grouped the Israeli B. caballi isolates in cluster A together with the South African isolates but separated from the American/Caribbean B. caballi strains. These results further support the notion that RAP-1 cELISA relies on the recognition of epitopes that are not conserved across all B. caballi strains due to sequence heterogeneity and emphasizes the need to combine other detection methods to assure control of disease introduction into non-endemic regions. Conflict of interest The authors declare that no competing financial, or other, interests exist. Acknowledgements We thank Dr. Zohar Paternak for helping with the phylogenetic analysis of the rap-1 gene. References Abutarbush, S.M., Alqawasmeh, D.M., Mukbel, R.M., Al-Majali, A.M., 2012. Equine babesiosis: seroprevalence, risk factors and comparison of different diagnostic methods in Jordan. Transbound Emerg Dis 59, 72–78. Aharonson-Raz, K., Steinman, A., Bumbarov, V., Maan, S., Maan, N.S., Nomikou, K., Batten, C., Potgieter, C., Gottlieb, Y., Mertens, P., Klement, E., 2011. Isolation and phylogenetic grouping of equine encephalosis virus in Israel. Emerg Infect Dis 17, 1883–1886. Awinda, P.O., Mealey, R.H., Williams, L.B., Conrad, P.A., Packham, A.E., Reif, K.E., Grause, J.F., Pelzel-McCluskey, A.M., Chung, C., Bastos, R.G., Kappmeyer, L.S., Howe, D.K., Ness, S.L., Knowles, D.P., Ueti, M.W., 2013. Serum antibodies from a subset of horses positive for Babesia caballi by competitive ELISA demonstrate a protein recognition pattern not consistent with infection. Clin Vaccine Immunol 20, 1752–1757. Bellaiche, Michel. (2013). Annual Reports of the Veterinary Services and Animal Health. Bellaiche, (2013). Ministry of Agriculture and Rural Development (in Hebrew). Retrieved from: . Bhoora, R., Quan, M., Zweygarth, E., Guthrie, A.J., Prinsloo, S.A., Collins, N.E., 2010. Sequence heterogeneity in the gene encoding the rhoptry-associated protein-1 (RAP-1) of Babesia caballi isolates from South Africa. Vet. Parasitol. 169, 279– 288. Brüning, A., 1996. Equine piroplasmosis an update on diagnosis, treatment and prevention. Br. Vet. J. 152, 139–151. Butler, C., 2013. Can Theileria equi be eliminated from carrier horses? Vet. J. 196, 279. de Waal, D.T., 1992. Equine piroplasmosis: a review. Br. Vet. J. 148, 6–14. Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792–1797. García-Bocanegra, I., Arenas-Montes, A., Hernández, E., Adaszek, L., Carbonero, A., Almería, S., Jaén-Téllez, J.A., Gutiérrez-Palomino, P., Arenas, A., 2012. Seroprevalence and risk factors associated with Babesia caballi and Theileria equi infection in equids. Vet. J. 195, 172–178. Goldreich, Y., 2003. The Climate in Israel: observation, research and applications. Kluwer Academic/Plenum Press, New York. Kappmeyer, L.S., Perryman, L.E., Hines, S.A., Baszler, T.V., Katz, J.B., Hennager, S.G., Knowles, D.P., 1999. Detection of equine antibodies to Babesia caballi by recombinant B. caballi rhoptry-associated protein 1 in a competitive-inhibition enzyme-linked immunosorbent assay. J. Clin. Microbiol. 37, 2285–2290. Knowles, D.P., 1996. Control of Babesia equi parasitemia. Parasitol. Today 12, 195– 198. Kouam, M.K., Kantzoura, V., Gajadhar, A.A., Theis, J.H., Papadopoulos, E., Theodoropoulos, G., 2010. Seroprevalence of equine piroplasms and hostrelated factors associated with infection in Greece. Vet. Parasitol. 169, 273–278. Mehlhorn, H., Shein, E., 1984. The piroplasms: life cycle and sexual stages. Adv. Parasitol. 23, 37–103. OIE 2008. Manual of diagnostic tests and vaccines for terrestrial animals. Ch. 2.5.8. Equine piroplasmosis. Qablan, M.A., Oborník, M., Petrzˇelková, K.J., Sloboda, M., Shudiefat, M.F., Horˇín, P., Lukeš, J., Modry´, D., 2013. Infections by Babesia caballi and Theileria equi in
120
A. Rapoport et al. / Infection, Genetics and Evolution 23 (2014) 115–120
Jordanian equids: epidemiology and genetic diversity. Parasitology 140, 1096– 1103. Ribeiro, I.B., Câmara, A.C., Bittencourt, M.V., Marçola, T.G., Paludo, G.R., Soto-Blanco, B., 2013. Detection of Theileria equi in spleen and blood of asymptomatic piroplasm carrier horses. Acta Parasitol. 58, 218–222. Ros-García, A., M’ghirbi, Y., Hurtado, A., Bouattour, A., 2013. Prevalence and genetic diversity of piroplasm species in horses and ticks from Tunisia. Infect. Genet. Evol. 17, 33–37. Sam-Yellowe, T.Y., 1996. Rhoptry organelles of the apicomplexa: their role in host cell invasion and intracellular survival. Parasitol. Today 12, 308–316. Sevinc, F., Maden, M., Kumas, C., Sevinc, M., Ekici, O.D., 2008. A comparative study on the prevalence of Theileria equi and Babesia caballi infections in horse subpopulations in Turkey. Vet. Parasitol. 156, 173–177. Shkap, V., Cohen, I., Leibovitz, B., Pipano Savitsky, E., Avni, G., Shofer, S., Giger, U., et al., 1998. Seroprevalence of Babesia equi among horses in Israel using competitive inhibition ELISA and IFA assays. Vet. Parasitol. 76, 251–259.
Steinman, A., Zimmerman, T., Klement, E., Lensky, I.M., Berlin, D., Gottlieb, Y., Baneth, G., 2012. Demographic and environmental risk factors for infection by Theileria equi in 590 horses in Israel. Vet. Parasitol. 187, 558–562. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739. Terkawi, M.A., Alhasan, H., Ueno, A., Ratthanophart, J., Luo, Y., Cao, S., Kamyingkird, K., Aboulaila, M., Youn-Kyoung, G., Nishikawa, Y., Yokoyama, N., Xuan, X., Igarashi, I., 2012. C-Terminal region of 48-kDa rhoptry protein for serological detection of Babesia caballi antibodies in horses. Parasitol. Int. 61, 493–496. USDA. Seroprevalence of equine piroplasmosis disease agents in the United States, 2009. Available from: .
