Prevalence of Theileria and Babesia species in Tunisian sheep

In this study, the prevalence of Theileria and Babesia species in sheep was assessed with Giemsa-stained blood smear examination and polymerase chain reaction to identify the different piroplasms in 270 sheep from three Tunisian bioclimatic zones (north, centre, and south). The overall infection prevalence by Babesia spp. and Theileria spp. in Giemsa-stained blood smears was 2.9% (8/270) and 4.8% (13/270) respectively. The molecular results showed that sheep were more often infected by Theileria ovis than Babesia ovis with an overall prevalence of 16.3% (44/270) and 7.8% (21/270) respectively (p = 0.01). The molecular prevalence by Babesia ovis was significantly higher in females than in males (p < 0.05). According to localities B. ovis was found exclusively in sheep from the centre of Tunisia (Kairouan) whereas Theileria ovis was found in all regions. Infections with T. ovis and B. ovis were confirmed by sequencing. The sequence of T. ovis in this study (accession numbers KM924442) falls into the same clade as T. ovis deposited in GenBank. The T. ovis amplicons (KM924442) showed 99%–100% identities with GenBank sequences. Moreover, comparison of the partial sequences of 18S rRNA gene of B. ovis described in this study (KP670199) revealed 99.4% similarity with B. ovis recently reported in northern Tunisia from sheep and goats. Three nucleotides were different at positions 73 (A/T), 417 (A/T), and 420 (G/T). It also had 99% identity with B. ovis from Spain, Turkey and Iraq. The results suggest a high T. ovis prevalence in Tunisia with a decreasing north-south gradient. This could be correlated to the vector tick distribution.


Introduction
Small ruminant piroplasmosis is an important haemoprotozoan infection of sheep in tropical and subtropical regions . The piroplasms cause diseases that impair the development and productivity of the livestock industry and result in severe economic losses (Zhang et al. 2014). Piroplasmoses caused by Theileria and Babesia species lead to clinical infections in domestic and wild animals as well as in humans (Aydin, Aktas & Dumanli 2015). Theileriosis in small ruminants is caused by at least six species, namely Theileria ovis, Theileria separata, Theileria recondita, Theileria lestoquardi, Theileria uilenbergi and Theileria luwenshuni (Li et al. 2009;Schnittger et al. 2003;Zhang et al. 2014). Theileria ovis and T. separata are low-or non-pathogenic species (Friedhoff 1997), whereas T. lestoquardi is classified as malignant because it causes high mortality rates reaching 100% in some regions (Ahmed et al. 2003). In Tunisia, T. ovis was first reported in sheep using molecular techniques by M'ghirbi et al. (2013) and Rjeibi et al. (2014b). Recently, Rjeibi et al. (2014a) reported T. lestoquardi for the first time in sheep in southern Tunisia.
Babesiosis is a haemoparasitic disease belonging to a complex of several tick-borne diseases with different aetiological agents, such as protozoa, rickettsiae, and bacteria (Ranjbar-Bahadori et al. 2011) transmitted by ixodid ticks . The high lethality and morbidity caused by babesiosis explain its importance as a major constraint to livestock breeding development (Ahmed et al. 2002;Mehlhorn, Schein & Ahmed 1994). Several species of Babesia (Babesia ovis, Babesia motasi, Babesia crassa, and Babesia sp. Xinjiang) have been described in sheep; among them B. ovis and B. motasi are causative agents of sheep babesiosis (Liu et al. 2007;Ranjbar-Bahadori et al. 2011;Schnittger et al. 2003;Uilenberg 2001). In Tunisia, only B. ovis has been reported in small ruminants using molecular tools (Rjeibi et al. 2014b).
Babesia motasi is moderately virulent, whereas B. crassa appears to have little or no pathogenicity (Hashemi-Fesharki 1997). Haemaphysalis punctata is the vector of this species and is widespread in tropical Africa (Uilenberg et al. 1980). The most important Babesia species infecting small ruminants is B. ovis, which has been reported in Europe, Africa, Asia, and the Far East (Ahmed et al. 2006). Babesia ovis is highly pathogenic, especially in sheep; it causes severe infections characterised by In this study, the prevalence of Theileria and Babesia species in sheep was assessed with Giemsastained blood smear examination and polymerase chain reaction to identify the different piroplasms in 270 sheep from three Tunisian bioclimatic zones (north, centre, and south). The overall infection prevalence by Babesia spp. and Theileria spp. in Giemsa-stained blood smears was 2.9% (8/270) and 4.8% (13/270) respectively. The molecular results showed that sheep were more often infected by Theileria ovis than Babesia ovis with an overall prevalence of 16.3% (44/270) and 7.8% (21/270) respectively (p = 0.01). The molecular prevalence by Babesia ovis was significantly higher in females than in males (p < 0.05). According to localities B. ovis was found exclusively in sheep from the centre of Tunisia (Kairouan) whereas Theileria ovis was found in all regions. Infections with T. ovis and B. ovis were confirmed by sequencing. The sequence of T. ovis in this study (accession numbers KM924442) falls into the same clade as T. ovis deposited in GenBank. The T. ovis amplicons (KM924442) showed 99%-100% identities with GenBank sequences. Moreover, comparison of the partial sequences of 18S rRNA gene of B. ovis described in this study (KP670199) revealed 99.4% similarity with B. ovis recently reported in northern Tunisia from sheep and goats. Three nucleotides were different at positions 73 (A/T), 417 (A/T), and 420 (G/T). It also had 99% identity with B. ovis from Spain, Turkey and Iraq. The results suggest a high T. ovis prevalence in Tunisia with a decreasing north-south gradient. This could be correlated to the vector tick distribution.

Prevalence of Theileria and Babesia species in Tunisian sheep
Read online: Scan this QR code with your smart phone or mobile device to read online.
The advances in molecular biology enable genotypic characterisation and have proven to be useful for the identification and classification of several haemoparasites (Caccio et al. 2000). These molecular techniques are highly sensitive and specific compared to Giemsa-stained and serological techniques (Papadopoulos, Brossard & Perie 1996). The aim of the present study was to study the sheep piroplasms in Tunisia using Giemsa-stained blood smears, polymerase chain reaction (PCR), and PCRrestriction fragment length polymorphism (RFLP). Genetic characterisation was performed on the Tunisian isolates.

Study region and sample collection
The present study was carried out on 270 Barbarine sheep from three traditionally managed farms located in three Tunisian governorates (semi-arid, arid and Saharan) ( Figure 1). Data concerning altitude, annual rainfall, temperature, and moisture in the three study regions are reported in Table 1.
Based on their dentition, sheep were ranked into two age groups: less than and more than 2 years of age.
Blood samples were collected in EDTA tubes from each animal and stored at -20 °C until used. Giemsa-stained blood smears were examined microscopically with immersion oil at 1000x magnification for the presence of piroplasms. All the sheep included in the present survey were examined for ticks, which were collected and placed in tubes with 70% ethanol, then identified using the key of Walker et al. (2003). Three tick infestation indicators were determined (Margolis et al. 1982): • Infestation prevalence (%) = 100 × (number of infested sheep/total number of sheep). • Infestation intensity = number of ticks/number of infested sheep. • Abundance = number of ticks/total number of sheep.
Theileria spp. polymerase chain reaction and polymerase chain reaction-restriction fragment length polymorphism DNA was extracted from 300 μL of whole blood with the Wizard ® Genomic DNA purification kit (Promega, Madison, USA) according to the manufacturer's instructions. DNA was stored at -20 °C until used. A nested PCR detecting specific Theileria DNA of the 18S rRNA gene was performed using Thei F1 and Thei R2 primers for the primary PCR and Thei F2, Thei R2 primers for the secondary PCR (Heidarpour Bami et al. 2009) (Table 2). Distilled water and T. ovis DNA (Rjeibi et al. 2014b) were used as negative and positive controls respectively. PCR products were separated in 1% agarose gels to check the size of the amplicons.
The discrimination of three Theileria species (Theileria annulata, T. lestoquardi, and T. ovis) was done by RFLP analysis of amplicons with restriction enzyme: HpaII (Fermentas, Lithuania) (Heidarpour Bami et al. 2009). The enzymatic digestion mixture consisting of 10 μL PCR amplicons, 10x buffer, and restriction enzyme (10 U) was then incubated at 37 °C for 2 h. Restriction digests were separated by electrophoresis in 2% agarose and visualised in a UV transilluminator.

Babesia ovis polymerase chain reaction
Five microlitres of DNA of each sample were amplified by 35 PCR cycles using B. ovis specific primers (Bbo-F and Bbo-R) (Aktas et al. 2005). PCR products were separated in 1% agarose gels to check the size of amplicons (Table 2).

Semi-nested polymerase chain reaction detecting Babesia motasi
This PCR was performed in 100 μL total volume consisting of 1x PCR buffer, 2.5 U Taq polymerase (Biobasic, Canada), 2 μL of each primer P1/P2 (20 mM), 200 μM each of dNTP, and 1.5 mM MgCl 2 . Five microlitres of DNA of each sample were amplified by 38 PCR cycles in an automated thermocycler (ESCO Swift MaxPro) (Shayan et al. 2008). The PCR products were analysed on 1.8% agarose gel and visualised using ethidium bromide with a UV transilluminator. For specific detection of B. motasi DNA, semi-nested PCR was used using a species-specific primer (P4) designated within the V4 hypervariable region of 18S rRNA gene. Semi-nested PCR was performed with the purified PCR product. One microlitre of the purified PCR product was amplified with the primers P4/P2 (Shayan et al. 2008). Semi-nested PCR was separately performed directly with 1 μL PCR product as well. PCR products were separated in 1.8% agarose gels to check the size of the amplicons (Table 2).

DNA sequencing and phylogenetic analysis
Three selected T. ovis amplicons from each region and two PCR products of B. ovis from Kairouan (central Tunisia) were purified with the Wizard SV gel and PCR clean-up system (Promega, Madison, USA) according to the manufacturer's instructions. Purified DNA fragments were sequenced in both directions, using the same primers as for PCR. The reactions were performed using a conventional Big Dye Terminator cycle sequencing ready reaction kit (Applied Biosystems, Foster City, CA, USA) with an ABI3730XL automated DNA sequence.
The obtained 18S rRNA gene sequences were edited using the ChromasPro software (version 1.7.4). The pairwise nucleotide per cent identity of the new sequences was calculated using MEGA 5.1 software (Tamura et al. 2011). Phylogenetic trees were constructed by the neighbour-joining method (Saitou & Nei 1987). Distances were estimated by the Tamura-Nei method (Tamura & Nei 1993). The sequences of B. ovis and T. ovis 18S rRNA genes identified in the present survey were deposited in GenBank under accession numbers KP670199 and KM924442 respectively ( Figure 2).

Statistical analyses
The infection prevalence percentages were compared using EpiInfo6 (Dean et al. 2011). The concordance between PCR and blood smears was estimated with Kappa test (Toma et al. 2007). A probability of 0.05 was used as a threshold for statistical significance.

Ethical considerations
This study was conducted in accordance with relevant national and international guidelines on handling animals, taking care to respect animal welfare. Agreement was obtained from all farmers before their sheep were sampled.

Ticks population
In total 135 adult ticks were collected from 270 sheep (85 males and 50 females). The ticks' sex ratio (M: F) was 1.7. These ticks were identified as Hyalomma dromedarii (28.9%) and Rhipicephalus  Note: The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1100 replicates) was shown next to the branches. The evolutionary distances were computed using the Tamura-Nei method (Tamura & Nei 1993) and are in the units of the number of base differences per site. Evolutionary analyses were conducted in MEGA5.1 (Tamura et al. 2011). GenBank accession numbers are given in parentheses. Species described in this study are indicated with a black square.

FIGURE 2:
The tree was constructed using the neighbour-joining method (Saitou & Nei 1987); (a), Partial sequence 18S rRNA gene phylogenetic tree of the species identified in this survey and the main small ruminants' Theileria species; (b) Partial sequence 18S rRNA gene phylogenetic tree of the species identified in this survey and the main small ruminants' Babesia species. The overall prevalence of tick infestation was 23.3% (63/270), with an intensity of 2.14 and an abundance of 0.5. The prevalence of piroplasms was not significantly different in tick-infested and non-infested sheep (p > 0.05).

Theileria ovis infection
The overall infection prevalence by Theileria spp. in blood smears was 4.8% (13/270) and the overall mean parasitaemia was 0.016% (range: 0.01% -0.03%). The enzymatic digestion profile by HpaII restriction enzyme showed that all PCR positive amplicons belonged to T. ovis (44/270). There was no concordance between PCR and blood smears for Theileria spp. in sheep (k = 0.00).
The three T. ovis amplicons showed 100% identity between them (830 bp length); one of them was submitted to GenBank (accession number: KM924442). The present sequence showed identities of 99% -100% with GenBank published T. ovis sequences (Figure 2a). The sequence of T. ovis in this study falls into the same clade with all the T. ovis sequences from Africa, Europe, and Asia and is clearly distinct from other Theileria species such as T. uilenbergi (JF719835), T. lestoquardi (AF0811335), and T. annulata (AY524666).

Babesia ovis infection
The overall infection prevalence by Babesia spp. in blood smears was 2.9% (8/270) and the overall mean parasitaemia was 0.024% (range: 0.01% -0.03%). Eight sheep were exclusively infected by Babesia spp. in blood smears; 21 were positive for B. ovis by PCR. There was no concordance between PCR and blood smears for B. ovis infection (k = 0.00).
All positive blood smears were positive by B. ovis PCR while B. motasi DNA was not detected.
The overall B. ovis molecular prevalence was 7.8% (21/270); it was significantly higher in yearling sheep compared to adults (p < 0.001). The infection by B. ovis was found exclusively in central Tunisia (Kairouan) with a prevalence of 17.5% ± 0.06 (21/120) (p < 0.001). The infection rate by B. ovis was significantly higher in females than males (p = 0.01). Two

Discussion
Piroplasmoses are caused by the tick-borne haemoprotozoans Theileria and Babesia and have a major impact on livestock production in tropical and subtropical regions ). In the present study, we estimated the prevalence and distribution of Theileria and Babesia species in Tunisian sheep in three regions (north, centre, and south). We detected Theileria spp. and Babesia spp. in blood smears of 4.8% and 2.9% sheep respectively, whereas the molecular prevalence of T. ovis and B. ovis was significantly higher (16.3% and 7.8% respectively) (p < 0.05). This difference shows that the PCR is a reliable screening method in carrier sheep. Indeed, the parasitaemia in the present study was very low (range: 0.01% -0.03%). This is in agreement with other studies in Greece (Papadopoulos et al. 1996), Turkey (Razmi et al. 2003) (0.01% -0.1%), and recently in northern Tunisia (Rjeibi et al. 2014b) (0.01% -0.05%). Our research showed that sheep were more frequently infected by T. ovis (16.3%) than B. ovis (7.8%) (p < 0.05). This in agreement with the findings of Nagore et al. (2004) who found that sheep were more often infected by T. ovis (18%) than B. ovis (2.5%), contrary to Rjeibi et al. (2014b) who showed that sheep were more infected by B. ovis (17.4%) than T. ovis (5.8%). Concerning Theileria species, our survey showed that only T. ovis was present in Tunisian sheep, this in agreement with M'ghirbi et al. (2013) and Rjeibi et al. (2014b). The same results were reported in Turkey ). In the present study, no T. lestoquardi was isolated, but Rjeibi et al. (2014a) detected two positive animals, raising the question whether this species is established in Tunisia or not.
There is a decreasing north-south prevalence gradient; indeed, T. ovis was highly prevalent in Ariana (north) (42.5%) compared to Tataouine (south) (1.43%) (p < 0.001). This gradient was also in accordance with R. turanicus distribution, which was more prevalent in Ariana (46.25%) followed by Kairouan (6.66%), whereas this tick species was absent in the south (p < 0.001). These observations strongly suggest that in Tunisia R. turanicus is the vector of T. ovis.
Regarding Babesia species, no B. motasi was detected in this study; sheep were exclusively infected by B. ovis with an overall infection prevalence of 7.8% (21/270). Similar results were reported in Turkey and Iran where small ruminants were only infected by B. ovis Esmaeilnejad et al. 2014). In northern Spain, B. ovis and B. motasi were present in sheep with an overall prevalence of 2.5% and 2% respectively (Nagore et al. 2004). Babesia motasi is transmitted by Haemaphysalis spp. (Uilenberg 2006); this species and B. ovis are the primary agents of ovine babesiosis. Geographic B. motasi isolates have different virulence. For example, in northern Europe it has low pathogenicity whereas in the Mediterranean Basin this species is highly pathogenic (Uilenberg 2006).
The highest B. ovis infection rate was observed in female sheep (10.8%) compared to males (2.1%) (p < 0.05). This is contradictory to the recent survey in northern Tunisia that reported no difference in B. ovis prevalence between males and females (Rjeibi et al. 2014b). Iqbal et al. (2011) in Pakistan showed that males were more infected than females. This could be explained by the fact that the majority of female sheep graze whereas males are kept indoors.
Yearling sheep were exclusively infected by B. ovis with a prevalence of (10.8%) (p < 0.001); our results are consistent with those reported in Pakistan where the prevalence in animals aged less than 1 year was higher (Iqbal et al. 2011). These results suggest that the visited farms are in a state of enzootic stability for B. ovis infection, contrary to the findings in Turkey and Iran Razmi et al. 2002Razmi et al. , 2003 where no significant difference between animals' ages and B. ovis infection prevalence was reported.
Overall prevalence rates of B. ovis differed statistically among localities (p ≤ 0.001); it was only observed in Kairouan (central Tunisia), with a prevalence of 17.5%, where sheep were only infested by R. turanicus, which is the potential vector of B. ovis (Table 3). Rhipicephalus turanicus was reported by Bouattour (1987) in the northern and central regions of Tunisia but no specimens were collected in the south. On the contrary, Rjeibi et al. (2014b) reported that sheep were infected by B. ovis in northern Tunisia with a prevalence of 17.4% (30/172).
There was no association between tick burdens and piroplasm prevalence (p > 0.05). This indicator has little value because this study was trans-sectional and carried out on healthy animals. Other findings reported the presence of a positive correlation between tick burdens and infection prevalence (Aktas et al. 2005;Durrani et al. 2012;Iqbal et al. 2013).
The B. ovis sequence described in this study from central Tunisia (KP670199) had 99.4% similarity with the recently reported sequence in northern Tunisia (KF723612) (Rjeibi et al. 2014b).This could be explained by the absence of genetic mixing between the two parasite populations.
In the phylogenetic tree, our samples clustered with all B. ovis in a single clade. Phylogenetic analyses provided evidence that B. ovis is distinct from other Babesia species, namely, B. motasi, B. crassa and Babesia sp. Xinjiang-2005.
These results suggest a high T. ovis prevalence in Tunisia with a decreasing north-south gradient corresponding to geographic distribution R. turanicus.

Conclusion
Further studies on representative tick samples are needed to establish the list of tick vectors of sheep piroplasms in Tunisia. M'ghirbi et al. (2013) found neither B. ovis nor T. ovis DNA in Rhipicephalus bursa (N = 10) and R. turanicus (N = 215). The pathogenicity of T. ovis either alone or in association with other pathogens is to be investigated.