Molecular and microscopic detection of Babesia caballi and Theileria equi among working horses and donkeys in Cairo and Giza provinces of Egypt

Equine piroplasmosis (EP) is an ixodid tick-borne disease caused by Theileria equi and/or Babesia caballi that can lead to severe health issues and economic losses among equine population. This study aimed to determine the prevalence of T. equi and B. caballi among Egyptian equines based on microscopy and conventional PCR. Also, to determine the effect of season, age, and sex of on their prevalence and determining the difference in sensitivity between microscopy and conventional PCR in the diagnosis of EP. This study was carried out on 432 blood samples randomly collected from 146 horses and 286 donkeys during a period from April 2016 to March 2018. Microscopic examination revealed that among horses, 13 (8.9%) and 4 (2.7%) were infected by T. equi and B. caballi respectively. While among donkeys, 22 (7.7%), 16 (5.6%) respectively. While mixed infections were detected in 4 (1.4%) donkeys. There was a statistically nonsignicant relation between prevalence of infection and season and sex of equines but the highest prevalence was recorded in age group less than 5 years old. By conventional PCR, among 64 horses, 15 (23.4%) and 8 (12.5%) were infected by T. equi and B. caballi, respectively. While among 76 donkeys, 36 (47.4%), 16 (21.1%), and 5 (6.6%) were infected by T. equi, B. caballi, and mixed infection, respectively. Our nding proved the existence of T. equi and B. caballi among equines. PCR at 392 bp fractionated on 1.5% agarose gel. Lane M: 50 bp DNA ladder, Lane 1: T. equi negative control, Lane 2: T. equi positive control, Lane 3–14 T. equi positive eld samples


Introduction
Members of family Equidae either wild or domestic ones are affected by a haemoprotozoic disease of widespread geographical distribution and international economic importance called Equine piroplasmosis (EP) that may progress as hyperacute, acute, subacute, or chronic illness (Mehlhorn and Schein, 1998). Ixodid ticks including the genera Dermacentor, Hyalomma, and Rhipicephalus are the main transmitters of the causative organisms (Babesia caballi (B. caballi) and Theileria equi (T. equi)) which are Intra-erythrocytic piroplasms belonging to phylum Apicomplexa that are endemic in tropical, subtropical, and temperate areas of the world (Rothschild, 2013). In addition to ectoparasitic way of transmission, piroplasms can also can be transmitted iatrogenically by blood transfusion or by using contaminated needles or surgical instruments (Homer et al., 2000;Zobba et al, 2008;Farkas et al., 2013). B. caballi and T. equi can cause variable and nonspeci c symptoms such as transient fever, jaundice, petechial hemorrhages, bilirubinuria, hemoglobinuria, distal limb edema, poor performance, loss of body conditions, inappetence, abortions, and even death (Zweygarth et al., 2002;Malekifard et al., 2014). Infected equines may remain carriers for B. caballi for several years while in case of T. equi, infected equines can remain carriers for their whole life. These carrier animals represent a huge threat as a source of infective stages of piroplasms for tick vectors (Rüegg et al., 2007). Mixed infections were documented in different countries where a common vector is present (Scoles and Ueti, 2015). Some factors have a great effect on the spreading of EP such as climatic conditions and transnational movement of equids. So, proper control policies must be directed toward importing equines. Serological examinations are commonly applied between borders for controlling equine movement (Sigg et al., 2010;Guidi et al., 2015).
Diagnosis of EP depends mainly on direct (microscopical examination and molecular techniques) and indirect (serological examinations) methods. The gold standard for diagnosis of EP is a microscopical examination of stained blood smears but it is highly signi cant only during the acute infection while it is of low sensitivity in the case of carrier animals due to low parasitemia (Quintão-Silva and Ribeiro, 2003). Several serological examinations were applied for the detection of antibodies against B. caballi and T. equi such as complement xation test (CFT), immuno uorescent antibody technique (IFAT), and enzymelinked immunosorbent assay (ELISA) (Kim et al., 2008). These serological tests are more sensitive but have some drawbacks related to cross-reactivity to other Babesia species and antibody detection limit. Molecular techniques were proved to have the highest sensitivity among the diagnostic procedures (BrÜning et al., 1997). With parasitemia as low as 0.000001%, the polymerase chain reaction (PCR) can detect parasitic DNA from a blood sample (Alhassan et al., 2005;El-Seify et al., 2018).
In Egypt, Studies on B. caballi and T. equi are infrequent and slight data had been provided. Therefore, this study aimed to determine the infection rate of T. equi and B. caballi among equines in Cairo and Giza governorates, Egypt based on microscopic examinations of Giemsa-stained blood lms and Polymerase Chain Reaction (PCR). Also, this study aimed to determine the effect of season, age, and sex of on the infection rate and determining the difference in sensitivity between microscopic examination and molecular techniques in the diagnosis of EP.

Sampling and study design
This study was conducted on 432 blood samples from apparently healthy equids including 146 horses and 286 donkeys, which were randomly selected from different localities belonging to Cairo and Giza governorates, Egypt during a period from April 2016 to March 2018. The samples were collected by jugular venipuncture method using EDTA-tubes. All samples were labeled by necessary data (age, sex, and date of collection) then were transported in an icebox to the Parasitology Department, Faculty of Veterinary Medicine, Kafrelshiekh University for examination.

Microscopic examination
In the lab, thin and thick whole blood lms were done, stained by Giemsa stain according to (Levine, 1982), and examined for the presence of intra-erythrocytic merozoites of B. caballi and T. equi using oil emersion lens of the light microscope. The remaining blood was kept at -80°C until further molecular examinations.

DNA extraction
Genomic DNA was extracted from blood samples showed positive results of EP in addition to some randomly selected microscopically negative samples using Thermo Scienti c™ GeneJET Genomic DNA Puri cation Kit (Cat No #K0722) according to manufacturer's instructions. The obtained DNA was stored at -20°C until used in the downstream applications. Molecular analysis was done in Biotechnology Department, Animal Health Research Institute, Giza, Egypt.

PCR ampli cation
Conventional PCR technique was performed via primers established by (Alhassan et al., 2005). that speci cally amplify fragments of 540 bp and 392 bp from 18s rRNA gene of B. caballi and T. equi, respectively. We applied a universal forward primer (Bec-UF2) with a sequence 5-TCGAAGACGATCAGATACCGTCG-3, a B. caballi speci c reverse primer (Cab-R) with a sequence 5´-CTCGTT CATGATTTAGAATTGCT-3´and a T. equi speci c reverse primer (Equi-R) with a sequence 5-TGCCTTAAACTTCCTTGCGAT-3. PCR was performed using GoTaq® G2 Flexi PCR Kit (Promega, USA) with a total volume of 25 µl containing 5X Green GoTaq® Flexi Buffer (10 µl), 25mM MgCl2 Solution (2 µl), PCR Nucleotide Mix (dNTPs) 10 mM each (1µl), Primer mix 10 pmol (1µl), 1.25 u GoTaq®G2 Flexi DNA Polymerase template DNA (5 µl), DNase/RNase free water (6 µl) (two PCR tubes one for B. caballi and the other for T. equi). The thermal pro le was 95°C for 5 min, followed by 35 cycles of repeated denaturation at 96°C for 1 min, annealing at 60°C for 1 min, and extension at 72°C for 1 min. then a nal extension at 72°C for 5 min, after that holding stage at 4°C for an in nite time. 8 µl of the generated PCR products were migrated on 1.5 % ethidium bromide-stained agarose gel under a constant voltage of 80 V for 40 min. visualization of agarose gel by Gel documentation system (XR) (Bio-Rad.UK). B. caballi and T. equi positive samples showed a band of 540 pb 392 bp, respectively.

Statistical analysis
The statistical analysis was done by using the statistical package Microsoft Excel software. Data were computed and represented in gures. The relationship between equine infection rate and risk factors such as age and sex were determined by Chi-square analysis with Fisher's exact option (SPSS, Version 17.0, Chicago, IL). Statistical signi cance was declared at P < 0.05.

Microscopical ndings:
Results of microscopic examination revealed that B. caballi was found in horses at a rate of 4/146 (2.7%) and in donkeys at a rate of 16/286 (5.6%), while T. equi was prevalent in horses at a rate of 13/146 (8.9%) and donkeys at a rate of 22/286 (7.7%). Mixed infection was observed only in donkeys at a rate of 4/286 (1.4%). (Table 1). Microscopically, B. caballi appeared as large Babesia measuring 2.5-4 µm in length double pyriform or single pyriform while T. equi appeared as small Babesia measuring 2 µm in length single round, double round, single pyriform, and maltase cross shapes. (Table 2) showed the effect of Season, Age, and Sex on the prevalence of B. caballi among horses and donkeys by microscopic examination of Giemsa-stained blood smears as follows: seasonal prevalence of B. caballi among horses was the highest in winter (4.5%). While amongst donkeys the highest was in summer (10.2%). The highest prevalence of B. caballi in horses was in > 15 years old animals (15.8%) While in donkeys, the highest prevalence of B. caballi was in young donkeys under 5 years (13.2%). The prevalence of B. caballi in female horses was (2.9%) which is slightly higher than in males (2.6%) while in donkeys, the prevalence in males was (6.2%) higher than in females (4.8%). (Table 3) illustrated the effect of Season, Age, and Sex on the prevalence of T. equi among horses and donkeys by microscopic examination of Giemsa-stained blood smears as follows: seasonal prevalence of T. equi among horses was the highest in autumn (11.5%). While amongst donkeys the highest was in winter (15.4%). The highest prevalence of T. equi in horses was in the age group beneath 5 years old animals (16.7%). While in donkeys, the highest prevalence of T. equi was in young donkeys under 5 years (11.8%). The prevalence of T. equi in male horses was (11.8%) which is higher than females (5.7%) while in donkeys, the prevalence in males was (9.3%) higher than females (5.6%).
Chi-Square test was applied to detect the relation between prevalence of EP and equine species, season, sex, and age of equines. The obtained Chi2 values in the case of equine species, season, and sex showed insigni cant at p. value (α ≤ 0.05) which means that prevalence of infection and species, season, and sex of collected samples are independent factors. While for animals age, the obtained Chi2 value showed the highest prevalence of infection was among age group < 5 years compared with other age groups so, it was signi cant at p. value (α ≤ 0.05) which means that prevalence of infection and age of equines are two correlated factors.   1) and (Fig. 2), respectively.
Comparison between microscopical examination and conventional PCR for diagnosis of EP was statistically signi cant at p. value (α ≤ 0.001). So, our results showed that PCR is a more sensitive and accurate diagnostic method than microscopic examination of blood smears for the detection of equine piroplasms.

Discussion
The objectives of this study were to determine the infection rate of B. caballi and T. equi among the equine population in Cairo and Giza provinces of Egypt based on microscopic examinations and molecular techniques (PCR) in addition to studying the effect of season, age, and sex of infected animals on the infection rate and differentiation between microscopic examination and molecular techniques in the diagnosis of EP. For many years, the standard diagnostic and highly speci c method for EP was the microscopical examination of stained blood smears. But, due to low parasitemia in carrier animals, it is signi cant only during the acute phase of infection. So, we applied a conventional PCR technique that proved to be more sensitive than microscopical examination in such cases.
Equine piroplasms can be recognized based on biometrical and morphometric data. In the current study, B. caballi appeared as large Babesia measuring 2.5 to 4 μm (> 2.5 μm), double pyriform, or single pyriform. while T. equi appeared as small Babesia measuring 2 μm in length (< 2.5 μm), single round, double round, single pyriform, and maltase cross shapes. The morphological characteristics observed in B. caballi and T. equi in the current study were in agreement with the ndings of (Selim and Abd El-Gawad, 1982;Levine, 1982).
According to the current study, T. equi was more prevalent than B. caballi among equines. This was in agreement with (Sigg et al., 2010;Abedi et al., 2012). A possible reason for the low prevalence of B. caballi could be associated with the earlier removal of the parasites after a short term of infection (Salim et al., 2008).
In the present study, the prevalence of T. equi among donkeys was (7.7%) which was in agreement with (Inci, 2002) in Turkey (7.9%). while higher than (El-Kelesh et al., 2012) in Egypt who recorded (5.4%), and (Abedi et al., 2015) in Iran (3.8%). and lower than that recorded in Egypt (17.89%) by (Radwan, 2009), in Ethiopia (12.2%) by (Gizachew et al., 2013), and in India (35%) by (Sumbria et al., 2016). Our result for the prevalence of B. caballi among donkeys was (5.6%) which was in agreement with (Inci, 2002) in Turkey (5.3%). Higher than that recorded in Egypt by (Selim et al., 1983) who recorded B. caballi infection sporadically in Egypt and Ethiopia (1.8%) by (Gizachew et al., 2013). And lower than that recorded in Egypt (6.6%) by (Khalifa et al., 1988). The current study showed (1.4%) mixed infection among donkeys which was in agreement with (Khalifa et al., 1988) in Egypt who recorded (1.5%) Differences in prevalence among horses and donkeys either in Egypt or in other countries for T. equi and B. caballi may be attributed to differences in the various geographic areas in Egypt and high or low vector tick activity in the area of sampling, where climatic conditions such as temperature, humidity, and rainfall in uence the habitat for ticks. Also, the time of sampling can make a difference where samples were collected at the acute or chronic stage of the disease, more aggressive treatment of infected animals, and increasing efforts of tick control (Hosseini, et al., 2016;Mahdy et al., 2016).
In the current study, there was a statistically nonsigni cant relationship between the prevalence of EP and the season of the year either among horses or donkeys. This is in agreement with (Abedi et al., 2015;Malekifard et al., 2014).
Previous studies in different governorates in Egypt reported various seasonality for equine piroplasm's among horses, for the highest prevalence of T. equi that reported in October according to (Rashad, 1981), in January, June, and October according to (Selim et al., 1983), in summer according to (El-Fayoumy and Iman, 2006), in winter according to (Radwan, 2009) and in July according to (Salib et al., 2012). B. caballi was recorded sporadically among horses according to (Selim et al., 1983). While, among donkeys, the highest prevalence of T. equi that reported in January, June, and October according to (Selim et al., 1983), in autumn according to (Radwan, 2009). For B. caballi: it was recorded sporadically among donkeys according to (Selim et al., 1983).
Concerning the effect of sex on the prevalence of equine piroplasms among horses and donkeys, the current study showed a statistically non-signi cant effect of equine's sex on the incidence of T. equi and/or B. caballi among equines which mean, horses and donkeys of both sexes could be affected equally by T. equi and/or B. caballi. This is in agreement with (Abedi et al., 2015;Malekifard et al., 2014).
Concerning the effect of age on the prevalence of EP among equines either horses or donkeys, the current study showed that the highest prevalence was in the age group less than 5 years old which is statistically signi cant at P. value (α ≤ 0.05). Some previous studies revealed a non-signi cant effect of age on the prevalence of infection which means, horses and donkeys of all ages could be affected by T. equi and /or B. caballi according to (Abedi et al., 2015;Malekifard et al., 2014) in Iran. Others recorded the highest prevalence among different age groups such as (El-Fayoumy and Iman, 2006) in Egypt, reported that T. equi was the highest in 6-10years old horses, (Radwan, 2009) in Egypt, recorded that T. equi was more prevalent among horses in less than 2 years and 2-10 years old equally, (Ibrahim et al., 2011) in Egypt, recorded that T. equi was more prevalent among horses 2-5 years old, (Turaki et al., 2014) in Nigeria reported that T. equi was the highest in more than 4 years old horses and (Salib et al., 2012) in Egypt reported that T. equi was the highest in 5-10 years old horses. For B. caballi among horses (El-Fayoumy and Iman, 2006) recorded the highest prevalence was among 1-5 years old. While among donkeys, (Radwan, 2009) recorded the highest prevalence among the age group 2-10 years old donkeys.
DNA ampli cation for the detection of equine piroplasmosis is known as a powerful tool in both the early phase of infection and in carrier animals (Rampersad et al., 2003). In the current study, molecular techniques were proved to have a higher sensitivity than microscopic examination in the detection of subclinical and carrier animals that similar to (Mahdy et al., 2016).
Concerning the molecular prevalence of T. equi among donkeys in the current study, it was (47.4%), which was in agreement with  in Egypt (43.1%). While lower than (Hawkins et al., 2015) in Kenya (72%). And higher than (Machado et al., 2011) in Brazil (31.81%). While, the molecular prevalence of B. caballi among donkeys in the current study, was (21.1%), which was in agreement with (Machado et al., 2011) in Brazil (20.45%). While higher than  in Egypt (15.7%). The current study recorded dual infection among donkeys (6.6%) which, was higher than (Machado et al., 2011) in Brazil (4.5%).
The results of the current study were in agreement with many studies in Egypt and other countries in that PCR technique is more sensitive than microscopic examination in the diagnosis of equine piroplasms, so it is most useful in studying the epidemiological status of those parasites (Ibrahim et al., 2011;Mahmoud et al., 2016) in Egypt, (Malekifard et al., 2014;Sumbria et al., 2015) in India and (Guven et al., 2017) in Turkey.
Our results declared that T. equi is more prevalent than B. caballi, this agreed with (Abedi et al., 2014) this is due to earlier removal of B. caballi from blood (Qablan et al., 2013).
The results of molecular and microscopic examinations con rmed the simultaneous infection of equines in the study region with both equine piroplasms which were consistent with the nding of .

Conclusion
In conclusion, T. equi, B. caballi are endemic among Egyptian equines. Also, PCR has higher sensitivity than microscopical examination in diagnosis of EP especially in the subclinical and chronic phases of infection. PCR techniques must be incorporated in studying EP epidemiology. There was a statistically nonsigni cant relation between prevalence of infection and season and sex of equines but the highest prevalence was recorded in age group less than 5 years old.

Declarations
Author contribution: Ahmed M. Soliman conceived the study, performed eld work and collection of samples and accomplished laboratory work and analyzed data. Ahmed M. Soliman wrote the draft of the manuscript. Ahmed M. Soliman, Nagwa M. Elhawary, Nashwa M. Helmy and Sahar M. Gadelhaq reviewed and revised the nal draft of the manuscript. All authors read and approved the nal version of the manuscript.
Con ict of interest: The authors declare that there is no con ict of interest regarding the publication of this article Funding: No funding was received to assist with the preparation of this manuscript.
Financial and Non-nancial interests: The authors have no relevant nancial or non-nancial interests to disclose.
All experimental conditions for animals were performed according to the guidelines approved by the Animal Care and Use Committee of Cairo University, Giza, Egypt.