Rapid Detection of Equine Piroplasms Using Multiplex PCR and First Genetic Characterization of Theileria haneyi in Egypt

Equine Piroplasmosis (EP) is an infectious disease caused by the hemoprotozoan parasites Theileria equi, Babesia caballi, and the recently identified species T. haneyi. Hereby, we used a multiplex PCR (mPCR) targeting the 18S rRNA gene of T. equi and B. caballi for the simultaneous detection of EP in Egyptian equids and examined the presence of T. haneyi infections in Egypt. Blood samples from 155 equids (79 horses and 76 donkeys) collected from different governorates of Egypt were examined by mPCR and PCR targeting T. hayeni. The mPCR method revealed a prevalence of T. equi of 20.3% in horses and of 13.1% in donkeys and a prevalence of B. caballi of 1.2% in horses. B. caballi was not detected in donkeys in the current study. The mPCR method also detected coinfections with both species (2.5% and 1.3% in horses and donkeys, respectively). Additionally, we report the presence of T. haneyi in Egypt for the first time in 53.1% of the horse and 38.1% of the donkey tested samples. Coinfection with T. haneyi and T. equi was found in 13.5% of the samples, while infection with the three EP species was found in 1.9% of the samples.


Introduction
In rural areas of many developing countries, including Egypt, there is a huge reliance on working equids, which include horses, donkeys, mules, and ponies. These animals play important roles in sustaining the livelihoods of millions of people by providing support in industries that include agriculture, construction, tourism, mining, and public transport [1,2]. The health and welfare of domesticated equids are often overlooked in rural areas. Although equids can be affected by a myriad of diseases that include amongst others, African Horse Sickness, Epizootic Lymphangitis (EZL), Tetanus, Rabies, Trypanosomiasis, and Piroplasmosis, there is a general lack of knowledge regarding the identification, management, and prevention of infectious diseases [3].
followed by amplicon sequence comparison with South African and American T. haney isolates.

Comparative Analysis and Sequence Conservation of the 18S rRNA Amplicons among Different Isolates
The 360-bp fragment of the T. equi 18S rRNA gene was amplified and sequenced from nine selected positive samples. The identity percent among the different Egyptian amplicons from T. equi and B. caballi is shown in supplementary Tables S1 and S2. Blast analysis indicated that the amplicon derived from the Egyptian isolates showed between 95.7 and 99% identity to previously published T. equi 18S rRNA gene sequences. In addition, the amplified B. caballi amplicon (540 bp) from two selected positive sample was sequenced. Blast analysis indicated that the B. caballi Egyptian isolate showed an identity percent ranging from 98.1 to 99.3% to published B. caballi isolates. Comparative analysis showed that one T. equi Egyptian amplicon derived from one horse with accession number MW659071.1 and two amplicons from donkeys with accession numbers MW659072.1 and MW659079.1 clustered with sequences from Chile (MT463613.1) [31], Israel (MK932052.1) [13], China (MT093496.1) [31], Jordan (KX227 623.1) [32], and Nigeria (MN620483.1) [33], whereas only one Egyptian amplicon derived from one donkey (MW659078.1) clustered with sequences from the State of Palestine (KX227632.1) [32] and Nigeria (MN093917.1) [34]. In addition, three sequences derived from horses (MW659073.1, MW659074.1, and MW659075.1) and two from donkeys (MW659076.1 and MW659077.1) clustered together in a separate group from the other sequences obtained in the current study ( Figure 3).

Sequencing Analysis of a T. haneyi Hypothetical-Protein-Coding Gene
BLASTn analysis of the five T. haneyi Egyptian samples sequenced in this study showed 100% sequence identity to published T. haneyi sequences from South African isolates (MW591580-MW591586) [36] and to the published sequences of T. haneyi Eagle Pass strain gene for a hypothetical protein (MT896770.1) ( Figure S1). The comparative analysis, based on amplicons derived from infected Egyptian horses (n = 2) (MW591694.1, MW591695.1) and donkeys (n = 3) (MW591692.1, MW591693.1, MW591697.1), indicated that the Egyptian T. haneyi sequences all clustered together with the reference T. haneyi sequence and with sequences from South African isolates; T. equi genotype C (18S r RNA) was selected as an outgroup ( Figure 5)

Sequencing Analysis of a T. haneyi Hypothetical-Protein-Coding Gene
BLASTn analysis of the five T. haneyi Egyptian samples sequenced in this study showed 100% sequence identity to published T. haneyi sequences from South African isolates (MW591580-MW591586) [36] and to the published sequences of T. haneyi Eagle Pass strain gene for a hypothetical protein (MT896770.1) ( Figure S1). The comparative analysis, based on amplicons derived from infected Egyptian horses (n = 2) (MW591694.1, MW591695.1) and donkeys (n = 3) (MW591692.1, MW591693.1, MW591697.1), indicated that the Egyptian T. haneyi sequences all clustered together with the reference T. haneyi sequence and with sequences from South African isolates; T. equi genotype C (18S r RNA) was selected as an outgroup ( Figure 5).

Sequencing Analysis of a T. haneyi Hypothetical-Protein-Coding Gene
BLASTn analysis of the five T. haneyi Egyptian samples sequenced in this study showed 100% sequence identity to published T. haneyi sequences from South African isolates (MW591580-MW591586) [36] and to the published sequences of T. haneyi Eagle Pass strain gene for a hypothetical protein (MT896770.1) ( Figure S1). The comparative analysis, based on amplicons derived from infected Egyptian horses (n = 2) (MW591694.1, MW591695.1) and donkeys (n = 3) (MW591692.1, MW591693.1, MW591697.1), indicated that the Egyptian T. haneyi sequences all clustered together with the reference T. haneyi sequence and with sequences from South African isolates; T. equi genotype C (18S r RNA) was selected as an outgroup ( Figure 5)

Discussion
Piroplasms are Apicomplexa tick-borne parasites distributed worldwide which are responsible for piroplasmosis (theileriosis and babesiosis) in vertebrates. The aim of the present study was to use molecular methods for the detection of the prevalence of EP in Egypt caused by T. equi and B. caballi. We also aimed at detecting the occurrence of T. haneyi in equids in Egypt, which was unknown. Importantly, the DNA sequence data generated in this study also allowed for some genetic characterization of T. equi, B. caballi, and T. haneyi Egyptian strains currently circulating in this country.
The prevalence of T. equi was higher than that of B. caballi, and this is consistent with previous reports [37,38]. This phenomenon may be due to the increased susceptibility of B. caballi to treatment compared to T. equi. In addition, the horse immune system may be more efficient in eliminating B. caballi-infected erythrocytes than T. equi-infected ones, the latter parasites having a long persistence [9,39].
The observed difference in the prevalence of EP compared to other countries may be due to the type of equids (race or working) examined, hygienic measures, differences in environmental conditions-which can have a significant impact on tick activity-tick control strategies, number of samples analyzed, and type of PCR used for molecular diagnosis [24].
Blast analysis of the amplified fragments from T. equi and B. caballi showed sequence identities between 96 and 99% to published sequences. While lower sequence similarities may indicate distinct parasite species, it is important to note that the analysis was based on small fragments of the 18S rRNA gene. However, initial epidemiological studies on South African T. equi and B. caballi 18S rRNA gene sequences reported identities between 96.1 and 99.9% to the previously published T. equi sequence from South Africa (accession number: Z15105) and between 96.9 and 99.9% to a published B. caballi sequence from South Africa (accession number: Z15104). Phylogenetic analysis of the South African sequences and subsequently of sequences from other parts of the world led to the identification of distinct parasite genotypes, which may even represent distinct parasite species [36]. Therefore, the sequences obtained in this study could represent Egyptian isolates that belong to theses distinct parasite genotypes. However, amplification and sequencing of the complete 18S rRNA gene would be necessary to confirm these identities.
Theileria haneyi was defined as a new species infective to equids [5] and has since been reported to occur in several countries in North and South America, Africa, and Asia [5,9,36,42]. In the current study, T. haneyi was identified in both horses and donkeys in Egypt, and the sequence of the hypothetical-protein-coding gene was identical to the published T. haneyi Eagle Pass reference sequence and to sequences from South African isolates, confirming the presence of T. haneyi in Egypt, as reported here for the first time.
The results of the current study are in agreement with Sears et al., [10] who reported that coinfection of T. haneyi and T. equi could be induced experimentally in horses, which can explain the presence of the three parasites in naturally infected animals in our study. That means there was no cross immunity induced by T. haneyi and other two equine piroplasm (T. equi and B. caballi) and the infection with these two parasites does not protect equines from the infection with T. haneyi and vice versa.
The prevalence of T. haneyi either as single or as a mixed infection with T. equi and B. caballi was higher than that recorded for imported Argentine horses in Nigeria (2.7% and 0.6%, respectively) [9], and this observation may be explained by the factors mentioned earlier that include environmental conditions, husbandry, and tick vectors. Differences in sampling size and time of sample collection could also be contributing factors.
The application of new technologies with higher sensitivities and specificities could better facilitate the diagnosis of EP in Egypt. A multiplex EP real-time PCR assay targeting the 18S rRNA gene was developed for the simultaneous, quantitative detection of T. equi and B. caballi in field animals. Quantitative molecular genotyping assays for T. equi were also developed and enable the rapid detection of distinct T. equi parasite genotypes. Future studies in Egypt should focus on further characterizing the T. equi and B. caballi genotypes that may be circulating within the different governorates, with a view to determining risk factors in disease control. It has been noted that T. haneyi species classification was based on differences in the equi merozoite antigen (EMA) multigene family, and the identification of T. haneyi in South African horses infected with T. equi genotype C indicated that T. haneyi may be a subgroup of T. equi Genotype C [5,36]. The identification of T. haneyi in Egyptian equids is not surprising but warrants further investigation. the 18S rRNA gene was developed for the simultaneous, quantitative detection of T. equi and B. caballi in field animals. Quantitative molecular genotyping assays for T. equi were also developed and enable the rapid detection of distinct T. equi parasite genotypes. Future studies in Egypt should focus on further characterizing the T. equi and B. caballi genotypes that may be circulating within the different governorates, with a view to determining risk factors in disease control. It has been noted that T. haneyi species classification was based on differences in the equi merozoite antigen (EMA) multigene family, and the identification of T. haneyi in South African horses infected with T. equi genotype C indicated that T. haneyi may be a subgroup of T. equi Genotype C [5,36]. The identification of T. haneyi in Egyptian equids is not surprising but warrants further investigation.

DNA Extraction
Genomic DNA was extracted from FTA Elute Micro Card [43,44], following the manufacturer's instructions.

DNA Extraction
Genomic DNA was extracted from FTA Elute Micro Card [43,44], following the manufacturer's instructions.
Positive control DNA samples extracted from T. equi and B. caballi in vitro cultures were provided by the OIE equine piroplasmosis reference lab located in Pullman, WA, USA. Field samples were tested for the presence of equine piroplasmosis using a published conventional mPCR assay designed for the simultaneous detection of T. equi and B. caballi infections [40]. The 18S rRNA gene was used, targeting the 943-1300-bp region for T. equi and the 562-1141-bp region for B. caballi [38,40]. Briefly, the universal forward primer Bec-UF2 and species-specific reverse primers (Cab-R, B. caballi; Equi-R, T. equi) were combined in reactions containing 3 µL of DNA sample, 12.5 µL of Sigma 2× JumpStart™ REDTaq ® ReadyMix™ (Foster City, California, USA), 5 µM of each primer, and 7.5 µL of nuclease-free water in a 25 µL total volume. Primers sequences are shown in Table 3. The amplification conditions were according to Abedi et al. [38], with minor modifications, which included an initial denaturation for 5 min at 94 • C, followed by 35 cycles each of 94 • C for 1 min as a denaturation period, an annealing period of 54 • C for 1 min, and an extension period at 72 • C for 1 min, with the addition of a final extension period of 7 min at 72 • C. The DNA extracted from T. equi and B. caballi in vitro cultures was used as a positive control, and the negative control was a no-template control (NTC). All amplicons were visualized by 2% agarose gel electrophoresis (Invitrogen, Waltham, USA).

Uniplex PCR (uPCR) for Confirmation of the mPCR Results for the Detection of T. equi and B. caballi
Samples that tested positive for piroplasmosis using the mPCR assay were confirmed by performing uPCR assays. For the amplification of T. equi parasite DNA, the primers TBM and Equi-R were used, while the amplification of B. caballi was done using the primers Bec-UF2 and Cab-R ( Table 3). The reactions were set up as previously described, and PCR amplification conditions were the same as those reported for the mPCR assay.

Detection of T. haneyi
For the detection of T. haneyi, instead of performing a nested PCR as done by Knowles et al. [5], a gradient annealing temperature in PCR using the internal nested primers described in Table 3 was used. The best annealing temperature was 56 • C, which was chosen to complete the amplification process. Amplicons were visualized by 1.5% agarose gel electrophoresis.

Sequencing and Sequence Analysis
Samples (T. equi n = 9; B. caballi n = 2 and T. haneyi n = 5) that gave strong positive amplification reactions were selected for further sequencing and comparative analyses. Briefly, amplicons were purified using the GeneDirex PCR clean-up and Gel Extraction kit (Taiwan) according to the manufacturer's instructions and sent for bi-directional sanger sequencing to Macrogen ( Seoul, South Korea ) using ABI3730XL DNA Sanger sequencer (ThermoFisher) (Waltham, MA, United States) All sequence data were edited using MEGA 7 software (https://www.megasoftware.net/download_form accessed on 2 January 2021). Query coverage and the percent of identity among the compared sequences were calculated by non-redundant National Centre for Biotechnology Information (NCBI) and Clustal Omega (https://blast.ncbi.nlm.nih.gov/Blast.cgi accessed on 2 January 2021) and (https://www.ebi.ac.uk/Tools/msa/clustalo/ accessed on 1 March 2021). In the present study, samples were aligned with the reference sequences for 18S rRNA representing T. equi (Z15105.1) [45] and for a gene coding a hypothetical protein of unknown function but specific for T. haneyi genome (MT896770.1 T. haneyi Eagle Pass strain) [5], available in the NCBI database. In addition, B. caballi gene sequence was kindly provided by Lowell S. Kappmeyer [Animal Diseases Research Unit, USDA-ARS, Pullman, WA 99164-6630, US]. Moreover, the T. equi and B. caballi sequences of the present study were compared with different 18S rRNA reference sequences collected from distinct geographical areas worldwide and available in GenBank (Tables S3 and S4) [46][47][48][49][50][51][52][53][54]. T. haneyi sequences were compared with the sequence of a hypothetical-protein-coding gene of T. haneyi Eagle Pass strain present in GenBank and with six T. haneyi South African (SA) isolate sequences [36]. All sequence data were edited using MEGA 7 software. Query cover and identity percentage among the compared sequences were calculated by NCBI and Clustal Omega (https://blast.ncbi.nlm.nih.gov/Blast.cgi accessed on 16 March 2021) and (https://www.ebi.ac.uk/Tools/msa/clustalo/ accessed on 23 February 2021). The resulted sequences data were submitted to GenBank to get accession numbers for T. equi, B. caballi, and T. haneyi Egyptian isolates.

Comparative Analysis
To assess the genetic diversity of hemoparasites within the study samples, speciesspecific dendrograms were constructed using a phylogenetic tree prediction generated by MEGA 7 (https://www.megasoftware.net/download_form accessed on 3 April 2021). This dendrogram was constructed using the Maximum Likelihood method based on the Kimura 2-parameter mode [55]. Egyptian T. equi and B. caballi isolates and the 18S rRNA gene of T. equi and B. caballi of different reference sequences in GenBank were used for comparative analyses, which were classified into genotypes A, B, C, D, and E for T. equi and genotypes A, B1, and B2 (C). The 18S rRNA gene sequences of B. bovis (AY150059.1) [56] were included in the dendrogram as outgroups for the T. equi dendrogram, while Eimeria sp. cytochrome oxidase subunit I (COI) gene (KT305929.1) [52] was used as the outgroup for the B. caballi dendrogram. Hypothetical-protein-coding gene of unknown function of T. haneyi Egyptian isolates, South African isolate (SA) [36], and T. haneyi Eagle Pass strain reference sequence [5] were used in T. haneyi's dendrogram construction. Theileria equi genotype C South Africa (EU888903.1) [12] was used as the outgroup.

Statistical Analysis
The chi-square (χ 2 ) test was applied at a probability of p < 0.05 to compare infection rates between equids determined by mPCR and cPCR. Significant associations were identified when a p value of less than 0.05 was observed [57].

Conclusions
The mPCR technique is a rapid diagnostic method for the simultaneous detection of both T. equi and B. caballi, especially in mixed-infected cases. This study represents a first report on the presence of T. haneyi in Egyptian equids and, specifically, in donkeys. Further