Molecular Detection of Theileria ovis and Theleiria equi in Livestock from Palestine

Theileria and Babesia are intracellular protozoan parasites infecting a wide range of animals. In Palestine, there is limited information on the prevalence of Theileria and Babesia spp. in livestock. We used PCR of the 18S ribosomal RNA gene followed by DNA sequencing to detect and identify parasite DNA in blood samples from sheep (n = 49), goats (n = 48), horses (n = 40), camels (n = 34), donkeys (n = 28) and mules (n = 2) from four districts of Palestine. DNA of T. ovis and T. equi was detected in 19 and 2 ovine blood samples, respectively. None of the camels, donkeys, and goats were positive for T. ovis. Sheep had a significantly higher rate of infection than other animals (P < 0.05). Theileria ovis is highly prevalent in sheep, while T. equi DNA was detected in a small proportion of the equids in Palestine.

www.nature.com/scientificreports www.nature.com/scientificreports/ DNA Extraction, PCR amplification, and sequencing. DNA was extracted from 300 uL of blood using a commercial kit (Master Pure TM DNA purification kit for blood version II, Epicenter, Madison, WI, USA), following the manufacturer's instructions as previously described 9 . The PCR reactions were performed using primers BJ1 (5′-GTC TTG TAA TTG GAA TGA TGG-3′) and BN2 (5′-TAG TTT ATG GTT AGG ACT ACG3′) which amplify a fragment of 460-540 bp of the 18S rRNA gene of piroplasmid infection including the genus Babesia and Theileria parasites as described previously 7,8 and followed by sequencing to identify piroplasm DNA in positive samples.
All positive PCR products detectable by gel electrophoresis were sequenced at Hylabs in Jerusalem, Israel. The chromatograms were checked, and the sequences were assembled by the BioEdit software. The 18S rRNA sequences were trimmed and aligned using the Multalin Multiple sequence alignment tool (http://multalin.toulouse.inra.fr/multalin/). DNA sequences were compared with the GenBank database by the nucleotide sequence homology search facilitated by the National Centre for Biotechnology Information (NCBI) using the BLAST analysis database (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The species' identity of sequences was determined according to the closest BLAST match with an identity of 97-100% to GenBank accessions. All samples that were positive for piroplasmids were confirmed by RFLP of the PCR product using the ApoI restriction enzyme as done by Azmi and colleagues 8 . To verify the source of the animal's blood samples, mainly the horse and sheep blood, primers targeting the 12S and 16S mitochondrial rRNA gene, which amplify a polymorphic region from among a large diversity of species: 12-16SF (5′-ACACCGCCCGTCACCCTCC-3) and 12-16SR (5′-AACCAGCTATCACCAGGCTCG-3), were used 10 , and these samples were sequenced. phylogenetic analysis. Phylogenetic analyses of the 18S rRNA sequences were performed by the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) applying the neighbour joining and maximum likelihood algorithms. Phylogenetic tree analysis was conducted by the MEGA 6 program using the UPGMA module. The reliability of internal branches was assessed by bootstrapping with 1000 pseudoreplicates. Nodes with bootstrap support less than 70% were collapsed. horses (20.2%), 34 camels (16.7%), 28 donkeys (13.8%) and 2 mules (1.0%) from four localities in Palestine were included in the study. Samples were collected from 154 females and 48 male animals of all host species studied. The distribution of animals according to location is presented in Table 1.
Based on PCR, followed by DNA sequencing, 22 domestic animals, i.e. 10.8% of the livestock was found to be positive for piroplasmid infection. Theileria spp. DNA was detected in the blood of 21 sheep (21/49; 42.9%), of which 19 were found positive for T. ovis and 2 were found positive for T. equi. None of the camels, donkeys, goats, and mules were positive for T. ovis. Successful identification of the horse blood was achieved using the 12S and 16S mitochondrial rRNA gene and DNA sequencing showed 100% identity with mitochondrial ribosomal operon (accession number: MG001413.1) from a Chinese horse. Sheep had a significantly higher rate of infection when compared with other studied animals (P < 0.05) ( Table 1). Of the sheep studied, 43 were females (87.8%) and 6 were males (12.2%); of the 43 females, 15 were positive for T. ovis and 4 of the 6 males were positive for T. ovis. No significant association of PCR positivity was found with gender or location. No Babesia spp. were detected in any of the studied animals.
Sequencing and phylogenetic analysis. All the PCR positive samples were sequenced. A neighbour joining phylogenetic analysis was used to reveal the relationship between the partially generated 18S rRNA gene sequence and other Theileria species' sequences. The observed sequences were phylogenetically analysed to confirm their similarities. The T. ovis sequences described herein form a well-supported clade with all the other studied T. ovis sequences, whereas other Theileria sequences clearly clustered with T. equi sequences. The phylogenetic analysis for the 19 DNA sequences amplified in this study formed a well-supported clade which showed 100% identity with a T. ovis isolate from Turkey (Accession number: KU714608.1). PCR-RFLP patterns for all T. ovis amplicons showed the following patterns: 244, 121, 115, and 26 bp, which is characteristic of T. ovis (Fig. 1).
One of the two T. equi sequences obtained in this study showed 100% identity with T. equi (accession number: KJ801931.1) from a Saudi Arabian horse, while the second clustered with T. equi from a Sudanese horse (accession  www.nature.com/scientificreports www.nature.com/scientificreports/ number: AB515312.1) (Fig. 2). Two sequences of the newly described T. haneyi (KU647704.1; Ku647709.1) clustered separately from T. equi and T. ovis 11 . The T. equi 18S rRNA gene of the two sequences fell into two genotypes, Clade A and Clade B (Fig. 3)

Discussion
This is the first preliminary study in which molecular diagnostic techniques were used to screen for the presence of Theileria and Babesia spp. in livestock in Palestine. Theileria ovis was found to be the most prevalent species. A similar result with a high rate of T. ovis infection had been reported previously 6 from ticks in Palestine (14.9% of the ticks from sheep), with T. ovis detected in Rhipicephalus bursa, Rhipicephalus sanguineus s.l. and R. turanicus that were collected while feeding on sheep. T. ovis is considered to be widely distributed in Asia, Europe and Africa 4 . The high prevalence of T. ovis in sheep (42.9%) in the present investigation was not surprising, since a high prevalence of this species was detected from sheep in Sudan (88.6%) 12 , Spain (18.9%) 13,14 and Turkey (54.0 to 67.9%) 3,4 . Interestingly, T. ovis is considered as causing sub-clinical infection in small ruminants in contrast to the virulent T. lestoquardi 8,15,16 . No infection was detected in goats, whereas most of the sheep were infected. The higher frequency of infection in sheep compared to goats is in agreement with studies from other countries, such as Ethiopia, where Theileria spp. infection in sheep was also more common than in goats 17 and also in a study from Turkey where 34.6% of the sheep and 10% of the goats surveyed were positive for Theileria spp. 18 . This variation in infection rates could be related to several factors such as the genetic variation among animals and the presence of tick species that act as vectors. No Babesia ovis infection was detected in the studied livestock. This suggests that B. ovis infection is rare in the surveyed area. This is in agreement with our previous findings 6 that demonstrated the existence of a low level of B. ovis (0.6%) infection in R. bursa ticks collected from sheep in Palestine. R. bursa plays an important role as a vector of B. ovis and has been reported as the only vector for B. ovis 4 . The presence of R. bursa was relatively low among the ticks collected, 2.9% 6 , and almost all specimens of this tick were collected from sheep. This might explain why B. ovis was not as common as previously thought.
BLAST searches for DNA sequences of T. ovis from sheep, performed tor this study, indicated that they all clustered together. Comparison of the T. ovis 18S rRNA gene sequences obtained in this study indicates that T. ovis sequences are closely related to T. ovis sequences from Turkey (GenBank sequence: KU714608.1, KT851432.1) and from hard tick from Palestine (GenBank sequence: KT587795.1), with 100% query coverage. Furthermore, T. ovis from sheep grouped in different clades from B. ovis (GenBank sequence: KT587793.1 & KT587794.1) that were detected in ticks collected from our previous study 8 . In a previous study, T. ovis was significantly associated with ticks from sheep and with R. turanicus ticks 9 , and was the only species of Theileria found in ticks from Palestinian sheep. Interestingly, T. equi DNA was detected in the blood of two sheep. This is not surprising because our results agree with other studies in which T. equi DNA was identified in cattle, goats and sheep 19 and it has also been found in dogs in Spain 20 . Equine piroplasmosis is caused by two intra erythrocytic protozoans, T. equi and Babesia caballi. T. equi is considered to cause a more virulent infection [21][22][23][24][25][26] . Our findings and other reports which indicate that T. equi also occurs in domestic ruminants further expand the host range of this organism 19,27 . T. equi is a major cause of disease in horses. In this study, none of the horses and donkeys were infected with T. equi. This may be because the collection was done in apparently healthy animals. The significance, extent, and consequences of infections with T. equi in domestic ruminants require further investigation. www.nature.com/scientificreports www.nature.com/scientificreports/ Sequencing of a 18S rRNA PCR amplicon from a sheep in the current study was compatible with T. equi with 100% identity to a sequence from a horse from Saudi Arabia (KJ801931.1). Another sequence showed 95% identity with T. equi from a horse from Sudan (AB515312.1) and phylogenetically distinct from the novel species T. haneyi n. sp (KU647704.1 & KU647709.1), which is infective to equids, with an exceptional genomic diversity within the genus of Theileria. The two T. haneyi sequences (KU647704.1; Ku647709.1) were located in separate clusters of the dendogram 11 data not shown. When comparing these sequences that were published from Saudi Arabian and Sudanese horses 28 with sequences from Palestinian samples (KX227633, KX227631, KX227632) and from neighbouring countries such as Israel (Kx227639), there is a 96% identity by BLAST, and all these sequences cluster in clade A. In addition, all these sequences clustered with T. equi from Israel and Mexico (KX227630.1, JQ390047.1) with a high bootstrap value (99) as shown in Fig. 3. Although the small sample size may have affected these results, further molecular studies covering larger geographic areas targeting only Theileria and Babesia spp. are required to estimate the prevalence and economic importance of these infections in Palestine and to ascertain whether other piroplasmid species are present in the region. This preliminary survey is based on one molecular method only, which is reliable but cannot by itself indicate the presence of the actual parasite species, only of gene sequences similar to those of parasite species reported in previous studies. Accurate comparisons between the various regions of Palestine were not possible because the livestock animals sampled in each region were very different.

conclusion
This study demonstrated that ovine theileriosis is present in Palestine and suggested that T. ovis is the dominant piroplasmid agent in this region. Furthermore, evidence of T. equi infection in sheep is reported herein.