Molecular detection of Theileria species and Babesia caballi from horses in Nigeria

Equine piroplasmosis (EP) is an infectious, tick-borne disease caused by the hemoprotozoan parasites, Theileria equi, Babesia caballi, and a recently reported new species, T. haneyi. Infections by these apicomplexan parasites limit performance and cause economic losses for the horse industry. Equine piroplasmosis is widespread in the northern regions of Nigeria, where an increasing portion of the animal population is composed of horses. This disease has remained epidemiologically challenging, especially as the movement of horses increases across Nigeria. In this study, blood samples from 300 horses were collected in three states of northwestern Nigeria. The presence of piroplasms was screened by nested PCR targeting 18S rDNA and positive samples were analyzed using species-specific-nested PCR-targeting genes including ema1 (T. equi), rap1 (B. caballi), and a gene coding a protein of unknown function (T. haneyi). Species-specific-nPCR results demonstrated that the prevalence of T. equi was 13.0% (39/300), B. caballi was 3.3% (10/300) and T. haneyi was 2.7% (8/300). Mixed infections with T. equi and B. caballi was 2.7% (8/300) while T. equi, B. caballi, and T. haneyi multiple infection prevalence was 0.6% (2/300). We used 18S rDNA sequences to determine close relationships between T. equi by phylogenetic analysis and demonstrated that among 57 sequences of Theileria parasites, 28 samples belonged to clade A (49%), 13 samples were found to be clade C (22%), and 16 were clade D (28%). These results demonstrate the genetic diversity of T. equi circulating in horses from Nigeria.


Introduction
Equine piroplasmosis (EP) is an infectious, tick-borne disease caused by the hemoprotozoan parasites Theileria equi, Babesia caballi, and a recently reported new species, T. haneyi (Hall et al. 2013;Scoles and Ueti 2015;Knowles et al. 2018). It has a global distribution, but a few countries are reported to be EP-free (Wise et al. 2013;Scoles and Ueti 2015; World Organisation for Animal Health 2019). Genera of Ixodid ticks, namely Rhipicephalus, Dermacentor, and Hyalomma, are known to transmit these protozoa (Scoles and Ueti 2015). These protozoan pathogens cause acute and chronic infections with a mortality rate of up to 50% (de Waal 1992). Horses infected with T. equi remain persistently infected while those affected by B. caballi are infected for an extended period before the immune system naturally eliminates the parasite without therapeutic intervention. Conversely, the parasitemia may fall below the level of detection (Friedhoff and Soule 1996). Chronically infected horses are reservoirs to spread the pathogens via tick or iatrogenic transmission (Ueti et al. 2008;Short et al. 2012). It was reported that poor management of equine populations might have exacerbated the impact of infection in the most endemic areas (Scoles and Ueti 2015). Timoney (2000) suggests that the international trade in horses and other equids could lead to an increase in the global spread of equine diseases.
Previous reports have suggested that global warming has resulted in changes in ecosystems as well as microbial ecology (Harvell et al. 2002). The recent EP outbreaks in the USA and the Netherlands, without visible premonitory signs and reduced detectability during long-term persistence (Short et al. 2012;Butler et al. 2012), have made EP a challenging disease. The discovery of T. haneyi, defined as a new species infective to equids that lacks ema-1, the current diagnostic test target, allows the new parasite to elude current regulatory tests (Knowles et al. 2018). To overcome this issue, a gene coding a protein of unknown function in the syntenic T. haneyi locus of the vacated ema1 gene was selected to detect T. haneyi by PCR assay (Knowles et al. 2018;Sears et al. 2019). T. haneyi was initially thought to belong to a new isolate of T. equi, but phylogenetic evidence supports that it diverged from T. equi (Knowles et al. 2018). Previous studies based on phylogenetic analysis showed genetic diversity among T. equi isolates across the world, including clades A, B, C, D, and E, with the recently identified novel species T. haneyi belonging to clade C (Knowles et al. 2018).
Horses from various regions in northern Nigeria infected with protozoa parasites that cause EP were reported without substantive information on molecular detection and genetic diversity (Oladosu and Olufemi 1992). The mechanisms of EP transmission in Nigeria are poorly characterized. It is unknown which tick species in Nigeria are competent vectors for the pathogens that cause EP. Several ticks have been implicated in transmission of EP to horses, including Rhipicephalus evertsi evertsi, Rhipicephalus decoloratus, Hyalomma impeltatum, Hyalomma. truncantum, Hyalomma dromederii, and Amblyomma variegatum (Onyiche et al. 2019;Scoles and Ueti 2015), and some of these are found in Nigeria (Oguntomole et al. 2018), though not on equids. However, horses are not the preferred host for some ticks identified, and their natural disease relationships are not well-known. Besides tick transmission, other cogent factors contribute to the mechanical of EP transmission. These include injections with contaminated needles and syringes, bloodletting, and "acupuncture-like" practices (Mshelia PW, personal observation).
The diagnosis of EP has been based mostly on microscopic examination of blood smears and serological assays (Oladosu and Olufemi 1992;Knowles Jr. et al. 1992). However, the limitations of these diagnostic tools made it difficult to identify and genetically characterize species of Babesia and Theileria present in Nigeria. Horses in Nigeria are used for leisure riding, polo games, horse racing, and traditional ceremonies resulting in extensive horse movement across Nigeria without strict control. This could impact the epidemiology of EP. Therefore, in this study, emphasis was given to the molecular detection of these parasites, the causative agents of equine piroplasmosis, to begin to define their importance in Nigeria.

Materials and methods
Sample collection from horses All procedures performed in this study involving animals were in strict compliance with the ethical standards of Ahmadu Bello University, Zaria, Nigeria, ABUCAUC/2020/34. Blood samples were collected from 300 horses ranging from 2.5 to 18 years old from three States in northwestern Nigeria (Fig. 1 began at the center of the circle and moved out spirally towards the edge to result in a uniform distribution and the opportunity to isolate DNA from repeated sample punches. After applying the blood, the card was stored at room temperature for 6 h to dry. The dried card was sealed in a re-sealable bag with a pack of desiccant and stored in a cool, dry, and dark place until analysis. DNA extraction A sample disc approximately 4 mm in diameter was taken from each sample spot using a disposable biopsy punch with a plunger (Integra Life Sciences Services, France). The sample discs were placed in 0.2-mL PCR tubes. Two hundred microliters of FTA Purification Reagent (Whatman TM GE Healthcare UK Limited, UK) was added to each PCR tube and incubated for 5 min at room temperature. The supernatant was discarded. This procedure was repeated for a total of 3 washes with FTA Purification Reagent. The discs were suspended in 200 μl of TE buffer (10 mM Tris-HCl, 0.1 mM EDTA, pH 8.0) and incubated for 5 min at room temperature. The same procedure was repeated for a total of 2 washes with TE buffer. The discs were then allowed to dry at room temperature for 1 h. The FTA disc was used for PCR amplification.
Nested PCR amplification of the V4 region of 18S rRNA gene A hemi-nested PCR (the reverse primer being used in each reaction) assay was used to amplify the hypervariable region of the 18S rRNA gene as previously described (Liu et al. 2016). Briefly, the external PCR reaction was set up in a 19-μl total volume containing 10 μl Sigma 2× Jumpstart Red Taq mix, 7 μl of nuclease-free water, 1 μl of 10 μM RLB-F2, and RLB-R2 primers (Table 2) and the processed DNA disc from the FTA card. The external PCR reaction consisted of an initial denaturation at 95°C for 3 min, followed by 35 cycles of 95°C for 60 s, 52°C for 50 s, and 72°C for 90 s. The final extension was performed with one step at 72°C for 5 min. The internal PCR reaction was set up in a 20-μl total volume containing 10 μl Sigma 2× Jumpstart Red Taq mix, 7.5 μl of nuclease-free water, 1 μl of 10 μM RLB-FINT, and RLB-R2 primers and 0.5 μl of the first PCR product. The reaction cycling in the second PCR was carried out with an initial denaturation at 95°C for 3 min, followed by 35 cycles Nested PCR targeting species-specific parasites All 18Snested PCR positive samples were analyzed using nPCR (hemi-nested for B. caballi rap1) to identify specific piroplasms present in the samples. Primer sets for T. equi (ema-1 gene), T. haneyi (gene coding protein of unknown function), and B. caballi (rap-1 gene) are shown in Table 2. The reaction conditions for T. equi and B. caballi nPCR are as follows: the PCR external reaction was prepared with 12.5 μl 2× Platinum SuperFi PCR master mix (Invitrogen, USA), 1 μl of 10 μM of each primer, the FTA disc, and 9.5 μl of nuclease-free water. The PCR cycling conditions for T. equi and B. caballi were a denaturation at 98°C for 5 min, followed by 35 cycles of 98°C for 20 s, 59.5°C for 20 s, and 72°C for 20 s, with a final extension at 72°C for 5 min. In the internal PCR reaction, 1.0 μl of each PCR product from the external PCR reaction was used as a template and added to 12.5 uL 2× Platinum Superfi master mix and 11.5 uL of water. The internal reaction consisted of denaturation at 98°C for 5 min followed by 35 cycles of 98°C for 5 s, 61°C for 5 s, and 72°C for 5 s, with a final extension at 72°C for 5 min (Baptista et al. 2013;Schwint et al. 2009). For T. haneyi, the PCR external reaction was prepared with 12.5 μl 2× Platinum SuperFi PCR master mix (Invitrogen, USA), 1 μl of 10 μM of each primer, the FTA disc, and 9.5 μl of nuclease-free water. The cycling condition for the external reaction of T. haneyi included an initial denaturation at 95°C for 4 min., followed by 35 cycles of 95°C for 20 s, 63.5°C for 30 s, and 72°C for 20 s and a final extension at 72°C for 7 min (Knowles et al. 2018;Sears et al. 2019). The PCR internal reaction was prepared in a final volume of 25 μl, with 12.5 μl 2× Platinum SuperFi PCR master mix (Invitrogen, USA), 1 μl of 10 μM of each set of primers, 1 μl of DNA template from the external reaction, and 9.5 μl of nuclease-free water. The external reaction cycle includes initial denaturation at 95°C for 4 min., followed by 35 cycles of 95°C for 20 s, 58.1°C for 30 s, and 72°C for 20 s and a final extension at 72°C for 7 min.
Finally, a Babesia bovis nPCR based on the rap-1 gene was applied to two equine DNA samples (Thatcher198 and Castora213 RC) for which 18S-PCR and sequencing unexpectedly resulted in B. bovis 18S rDNA identity. The nPCR primers targeting B. bovis rap-1 gene are reported in Table 2, and performed according to the procedure reported elsewhere (Figueroa et al. 1993).
Phylogenetic analysis The 18S rDNA nPCR products from samples that were positive for 18S nPCR but negative for a species-specific reaction were sequenced to identify other parasites infecting horses. The samples were amplified for the V4 region of the 18S rRNA gene. The PCR products were cleaned up using ExoSAP-IT reagent (Applied Biosystems, Lithuania) before being sent for sequencing using amplifying primers (Sanger sequencing method, Eurofins Genomics, SimpleSeq service, Louisville, KY, USA). Sequencher (Genecodes, Ann Arbor, MI, USA) was used to trim raw sequences, and the resulting sequences were analyzed at the NCBI website (https://blast.ncbi.nlm.nih.gov/Blast.cgi) using nucleotide Blast with default settings. Reference sequences for 18S rDNA representing T. equi genotypes A-E, T. haneyi, B. caballi, and B. bovis were retrieved from GenBank for inclusion in the phylogenetic analysis (Bhoora et al. 2009;Knowles et al. 2018;Hall et al. 2013). The reference sequences are for T. equi clade A (JX177671, Z15105), T. equi clade B (AB515310), T. equi clade C (JQ390047, EU888903), T. equi clade D (AB515307, AB515312), and T. equi clade E (HM229408). Reference sequences for B. bovis (AY150059) and B. caballi (EU642512) were also included in the analysis as outgroups.
Evolutionary analysis by maximum likelihood method The evolutionary history was inferred by using the maximum likelihood method and Kimura 2-parameter model (Kimura 1980). The bootstrap consensus tree inferred from 1000 replicates is taken to represent the evolutionary history of the taxa analyzed (Felsenstein 1985). Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches (Felsenstein 1985). Initial tree(s) for the heuristic search were obtained by applying the neighbor-joining method to a matrix of pairwise distances estimated using the maximum composite likelihood (MCL) approach. There was a total of 319 positions in the final dataset and a tree is presented in Fig. 2. Evolutionary analyses were conducted in MEGA 6 (Kumar et al. 2018) and 22 randomly selected sequences from distinct areas were used for phylogenetic analysis. NCBI accession numbers were obtained for the 22 generated Nigerian 18S rDNAsequences used in the phylogenetic analysis (MN384237, MN384238, MN384239, MN384240, MN384241, MN384242, MN384243, MN384244, MN384245, MN384246, MN384247, MN384248, MN384249, MN384250, MN384251, MN384252, MN384253, MN384254, MN384255, MN384256, MN384257, MN384258). In the phylogenetic analysis, we used 32 nucleotide sequences, including 22 Nigerian 18S rDNA sequences and 10 references (Fig. 2).
In this study, we were able to sequence 61 products of 18S-nPCR and 57 were closely related to Theileria that infect horses, two were B. caballi and two were B. bovis. The familiar clade members that were recognized for Theileria with the highest percent similarity to the query sequence were 28 as clade A, 13 as clade C, and 16 as clade D. All the familiar clades were found in Nigeria with exception of clades B and E. To construct the phylogenetic tree, we used 22 randomly selected Nigerian samples of which 18 were closely related to Theileria spp. that infect horses, two were B. caballi and two were B. bovis (Fig. 2). For phylogenetic analysis, sequences were manually checked by alignment to verify that no duplicate sequences were used in the phylogram. No unique sequences were left out of the analysis.

Discussion
The emphasis of the current study is to understand genotypic diversity of protozoan parasites that cause EP in Nigeria using a molecular biology approach. In Nigeria, horses are an important commodity for recreation, sport, transportation, and work. Igabi LGA (Kaduna) has one of the highest numbers of polo ponies in the region, both exotic and indigenous, and annually hosts international polo tournaments, horse racing, and other horse-related activities. Zaria LGA hosts the most prestigious traditional horse procession (durbar), as well as polo and horse racing. This region has one of the largest populations of horses in Kaduna state. Hundreds of horse transits in and out of Zaria, particularly the ancient city, are coming mostly from different parts of northern Nigeria and the Niger Republic. A variety of events occur in this region including twice a year horse processions (durbar), polo events, and horse racing. These events have made Zaria an epidemiologically important niche for disease transmission, infection, and enabling possible adaptation of apicomplexans to other hosts, which could potentially result in various clinical manifestations that relate to EP that have not yet been fully recognized.
Though the number of horses sampled in Zaria was higher compared to other LGAs, it was evident that the result in this area represented the genotypes and parasites found in the other areas. Katsina and Jigawa both share a substantial border with the Niger Republic; the markets were part of the network of centuries-old markets when the region played an important role as the hub of trans-Saharan trade routes. These livestock markets linked into the road networks and settlement patterns, which makes it easier to transport animals across Nigeria. The movement of horses across the northern borders of Nigeria from various neighboring countries for various purposes could have been responsible for the pattern of results seen in the study areas. Therefore, these results could be an epidemiological reflection of EP status in West African Sub-region.
T. equi genotpes A C and D were identified in Kadunga and Kasinga, while genotypes A and D were only detected in Jigawa. Significantly, T. equi clades B and E were absent in all the samples analyzed. Consistent with other studies, genotype diversity of Theileria was reported in horses in South Africa, Israel, USA, and Spain (Criado-Fornelio et al. 2003;Bhoora et al. 2009;Hall et al. 2013;Bhoora et al. 2019;Coultous et al. 2020). Bhoora et al. (2019) found clades A, B, C, and D, but not E, in a survey of horses and zebras in South Africa. Coultous et al. (2020) having assayed only for clades A-D (but not E), finding clades A, C, and D in Gambia. Similar to our data, clade B sequences were not identified in Gambia, another west African country. In both studies, coinfections with 2 or more clade members were common, especially in the case of clades A and C.
Since there is no restriction on horse importation, the current horse importation strategy in Nigeria could lead to the introduction of these clades in the future. A large number of the Argentine polo ponies were negative for EP. These negative results could be related to the routine health care management practice which involves the administration of imidocarb which could have led to the persistently low parasitemia that usually follows a fluctuating parasite-load pattern in recovered horses (Ueti et al. 2005;Ueti et al. 2008). Following the serendipitous discovery of T. haneyi, there were questions raised on whether co-infection with T. equi and T. haneyi occurs naturally. The data obtained from this work prove that T. equi and T. haneyi can co-infect naturally and additionally with B. caballi. Experimental co-infection of T. equi and T. haneyi was reported (Sears et al. 2019). One of the six polo ponies from Argentina that arrived in Igabi LGA had T. haneyi. It was evident that T. haneyi was not detected at the time of arrival. The failure to detect T. haneyi is no surprise; Knowles et al. (2018) reported that T. haneyi lacks ema-1 which allows the parasite to elude the current regulatory c E L I S A t e s t w h i c h i s b a s e d o n E M A -1 o f t h e immunodominant EMA superfamily and a specific monoclonal antibody (Knowles et al. 2018). The 18S-rDNA nested PCR with sequencing detected larger numbers of T. haneyi as compared to the species-specific-nested PCR. This discrepancy is potentially associated with a reduced sensitivity to the gene coding for a protein of unknown function used to detect T. haneyi. The T. haneyi reference JQ390047 in Fig. 2 branches off earlier than clade C identified from Nigeria. The identity of the Nigerian clade C samples to references EU888903 and JQ390047 are 100% and 99.53%, respectively. The T. haneyi reference JQ390047 has only 2 nucleotide differences with the Nigerian clade C samples yet gives the appearance of greater distinctness in the phylogram. For comparison, T. equi clade A reference JX177671 has 24 nucleotide differences compared to clade C members in the phylogram, demonstrating the distinction of a separated clade. No samples reported here were 100% identical to T. haneyi.
Compared with equine theileriosis, B. caballi is not predominant in areas that are EP-prone as confirmed using molecular assays. The occurrence of B. caballi ranges from 0 to 19.3% depending on the area (Heim et al. 2007;Motloang et al. 2008;Mahmoud et al. 2016;Sumbria et al. 2016). However, a high prevalence of B. caballi was reported in Mongolia (Munkhjargal et al. 2013;Mans et al. 2015) and Italy (Laus et al. 2013). The low number of B. caballi detected in this study could be a result of the very low parasitemia that characterizes B. caballi infections, which rarely exceeds 1% and the removal of B. caballi over a period of time by the horse immune system (Hanafusa et al. 1998;Guclu and Karaer 2007).
An interesting finding was the detection of B. bovis DNA from horses that arrived in Nigeria from Argentina. Two animals were both 18S rDNA-PCR positive with sequences closely related to B. bovis, and those samples were also B. bovis species-specific nPCR positive. Herein, we have presented that imported horses could be infected with B. bovis, where different strains could be distributed to various areas of the world, considering the human-driven global horse movement. Whether B. bovis established infection in these horses is unknown, further investigation is needed to rule out the possibility of B. bovis, and other bovine Apicomplexan parasites, to establish infections in horses. Nonetheless, if B. bovis established infection, these horses could be sources of parasite dissemination. This could necessitate testing horses frequently whenever they are moving into disease-controlled areas. This is supported by the recent identification of T. annulata, B. ovis, and B. canis species in horse blood samples in Turkey (Ozubek and Aktas 2018), and a more recent detection of T. equi and T. velifera in guard dogs kept in horse stables in Saudi Arabia (Salim et al. 2019). Also, T. equi has been detected in clinically ill dogs in Croatia (Beck et al. 2009) and South Africa (Matjila et al. 2008). This seeming lack of host specificity has raised a question on the clinical implications and impact of atypical piroplasmosis (Ozubek and Aktas 2018). Equid and non-equids species are likely to share closely related parasites that could be transmitted by a wide range of tick vectors. With the seeming emergence of atypical piroplasmosis in a variety of species worldwide, there will be a need to elucidate the tick vector, parasite, and host relationships that lead to the development of clinical disease.

Conclusion
It is evident that T. equi, T. haneyi, and B. caballi are present in Nigeria and circulating on an endemic basis among the equine species supported by the current study. The presence of infected horses may play a critical role in the spread of these pathogens via tick vectors. However, it is unknown which ticks are vector competent in Nigeria. Determination of the various competent tick vectors responsible for parasite transmission warrants additional study. Therefore, surveillance and restriction of the international movement of equids are required to prevent the introduction of infected horses into Nigeria to avoid pathogen dissemination.
in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.