Epidemiological Study of Equine Piroplasmosis (Theileria equi and Babesia caballi) by Microscopic Examination and Competitive-ELISA in Erbil Province North-Iraq.

Background
Equine piroplasmosis is a major tick-borne disease that can lead to serious health problems and economic losses in horse industry. The aim of the study is to determine the prevalence of Theileria equi and Babesia caballi in different species of Equus namely (Horse, mule, donkey and pony) by Giemsa stained blood films and competitive ELISA.


Methods
This study was conducted at various geographic areas of Erbil governorate. A total of 349 blood samples from equine (209 horses, 62 mule, 57 donkey and 21 ponies) were collected to estimate the prevalence rate of Theileria equi and Babesia caballi by using Giemsa stained blood smear and competitive ELISA.


Results
The overall prevalence rates were (10.6%) consisting of (8.3%), (1.7%) and (0.6) for T. equi, B. caballi and both infection respectively by Giemsa stained blood smears, and the rate was 38.97%, consisting of 20.9% for T. equi, 11.2% for B. caballi and 6.9% for both infection by cELISA. Seroprevalence rate of T. equi was significantly higher (P < 0.001) than that B. caballi in equids. There was also a significant difference associated between age (P < 0.01), activity (P < 0.01), management (P < 0.001) and tick infestation (P < 0.001), but neither the type of equids nor the gender was significant differences associated with prevalence rate.


Conclusion
The present study concluded that the equine piroplasmosis is a vital infection distributed among the equine in Erbil province North of Iraq. Thus a better control programme should be implemented to contain and control the prevalence of the disease within the area.


Introduction
quine piroplasmosis is one of the most common tick-borne hemoprotozoan diseases that poses serious threats to the equids including horses, ponies, mules, donkeys and zebras, has important implications for the international movement of horses (1). The disease is caused by either Theileria equi or Babesia caballi or both protozoa (2). These blood protozoa are principally transmitted by the tick of ixodid family (3).
Equine piroplasmosis can cause significant economic losses that include the cost of treatment decreased equids production, still birth, lack of performance or death and inability to meet international requirements for exports or involvements in equestrian sports (4). Also, some countries do not allow the entry of subclinical seropositive horses (5). The clinical manifestations of the disease ranging from subclinical to acute potentially fatal infection depend on the equide immune status and the virulence of protozoa. The clinical signs characterised by fever, anaemia, icteric sclera, hemorrhagic mucus membrane, weakness, lethargy, dark urine, anorexia and mild colic with the reduced fecal output (6). However, most of the equine that recovers from infection are become carriers for several years and reservoirs for vector ticks (7,8). Identification of carrier equines is essential for the evaluation of infection risk as they serve as a pool of infection for ticks and cause natural transmission of the disease (9). In the epidemiological studies in different countries, there are many risk factors related with the high prevalence of equine piroplasmosis such as equids species, age, gender, breed, the presence of tick, activity and regions (10,11).
Several methods can diagnose equine piroplasmosis, direct microscopic identification of the parasite in stained Giemsa blood smears is confirmatory only during the acute stage of the infection, but it seems to be unsatisfactory to diagnoses in carrier animal, and it is also not practical on a large scale (7). Therefore, serological tests are recommended as a select method of diagnosis during latent stage of infection, especially when horses designed to be imported into regions that are free from the disease, while the tick vectors are present (12). In 2004, the OIE approved the competitive inhibition enzyme-linked immunosorbent assay (cELISA) for detection of antibodies against T. equi and B. caballi, and as a specified test for global horse activity (13). Later, direct detection of these haemoprtozoan infection can be carried out using molecular tools such as conventional PCR (14), multiplex PCR (15), nested PCR (16), real-time PCR (17) and reverse line bloat (RLB) hybridization assay (18) have been developed and are reliable diagnostic tools.
There is no previous serological study concerning the occurrence of T equi and B. caballi infection in equine in Erbil governorate, north of, Iraq. Therefore, the objective of this research is firstly to determine the prevalence of T. equi and B. caballi in equine by Giemsa stained blood smears and secondly to determine the presence of anti-Theileria equi and anti-Babesia caballi antibodies using cELISA.

Sample collection and study design
This study was carried out under the supervision and regulations of Ethic Committee at College of Veterinary medicine, University of Duhok, Iraq.
A total of 349 blood samples from equine (209 horses, 62 mule, 57 donkey and 21 ponies) were collected via jugular vein using anti-coagulant and without any anti-coagulant sterile vacutainers ® tubes. They were sampled in various geographic areas of Erbil E governorate. The sex, age and breed of the equine were also recorded during this study.

Examination of blood smears
Thin blood smears were prepared from blood samples in EDTA tubes. The smears were then air dried, fixed in absolute methanol for 5 minute and stained with 10% Giemsa stained for 30 minute to determine the parasite in erythrocytes. The slides were examined with an oil immersion lens at magnification of ×1000 (19).

Serological detection of T. equi and B. caballi antibodies
All sera collected from Equine were assessed for the existence of antibodies to T. equi and B. caballi using a commercial cELISA test kit (VMRD, Inc., Pullman, and WA99163 USA) methods following the manufacturer's instructions. This serological analysis identify serum antibodies against Equine Merozoite Antigen 1 (EMA-1) surface protein of T. equi (20), and rhoptry-associated protein (RAP-1) of B. caballi (21). The optical density (OD) values of the controls and samples were measured at 630 nm wave length using an automatic UV max kinetic microplate reader (BioTek ® Elx800, USA), and the percentage of inhibition (%) was calculated as follows: I (%)=100-(sample OD ×100)/(mean OD of three negative controls). Serum samples with ≥40% inhibition were considered positive and samples with <40% inhibition were considered negative as recommended by the manufacturer.

Statistical analysis
The x 2 and Fischer's exact test were used to differentiate the prevalence of equine piroplasmosis between various groups. Binomial logistic regression in GenStat 12 th Edition was used to determine odds ratio and the effect of risk factors (type of equine, gender, age of group, purpose of keeping and whether tick is found on equine or not) on the incidence of T. equi, B. caballi and both protozoa. All candidate variables were kept in the model with significant attributes at P <0.05.

Prevalence of equine pirolasmosis (EP)
Both protozoa morphological evident in stained smears, the polymorphic shapes of T. equi, oval, rod, single pear and double pear were observed in erythrocyte as in Fig. 1-A, and B. caballi appeared as single pyriform or double pear and an obtuse angle was observed as in Fig. 1-B. The rate of infection was 6 (1.7%), 29 (8.3%) of B. caballi and T. equi respectively, the mixed infection rate was 2 (0.6%) by Giemsa stained blood smears (Table  1). The overall seroprevalence of EP in this study by cELISA was about 38.9% (136/349) ( Table 1). Regarding the type of protozoa, this study found that the seroprevalence was 11.2% (39/349) and 20.9% (73/349) for B. caballi and T. equi, respectively. This work also found that the prevalence rate of mixed infec-tion was 6.9% (24/349); the seroprevalence rate of T. equi was proximately two times more than B. caballi ( Table 1). The cELISA has significant advantage than microscopic detection because of greater sensitivity and specificity for diagnosis of EP.

Multivariable analysis for the seroprevalence of EP
Four variable conditions were analysed in this seroprevalence study including the type of equids, gender, age and purpose of keeping (Table 2). In the current research, the prevalence of T. equi, B. caballi and both protozoa did not significantly differ between types of equids. The seroprevalence of B. caballi was significant higher (P=0.04) in female (15.3%) than male (7.7%) while no significant differences regarding the seroprevalence of T. equi and both infection were detected between genders. The odds infection of B. caballi in an age less than five years was 2.88 times greater than equids of other age (CI, 1.31-6.34) with significant differences (P=0.01). Despite that no significant differences respecting the seroprevalence of T. equi and both infection were observed in age groups 5-10 and >10 years. It also should be noted that recreation (P=0.09) and loading (P=0.01) appeared to be only significant influence for the prevalence of mixed infection in this study ( Table 2).

Association of seroprevalence of EP with management factors
The results of this study revealed that the seroprevalence analysis about the accurate characteristics of the equine population hypothesised to be related with the occurrence of equine piroplasmosis. The current research has showed that the prevalence of T. equi, B. caballi and both protozoa infection was higher among equids mixed with other animals in stable compared to equids kept alone, but there were no significantly differences (Table  3). Outdoor feeding management (grazing), was found to be a significant influence for the presence of T. equi; B. caballi and mixed infection (P<0.001) ( Table 3).
The odds of infection in equines with tick infestation was 14.87 and 5.03 times higher than equids without tick infestation (CI, 6.78-32.62) and (CI, 1.95-12.91) with significant differences (P=0.001) and (P=0.001) respectively for T.equi and B.caballi. On the other hand on nearby animals was significantly higher (P=0.09) and (P<0.001) for T.equi and B.caballi respectively compared to equines without infestation.

Discussion
Equine piroplasmosis is a significant disease that can lead to serious health problems and economic losses. Several studies have been previously done in Kurdistan region, Iraq, but were focused on the domestic animals including cattle, sheep and goat (22,23). Therefore, the objective of this study is firstly done to evaluate the seroprevalence rate of T. equi; B. caballi in equine using microscopic examination and cELISA. This study is also determining the several risk factors related to piroplasmosis in equine in Erbil Government.
In the current study, the rate of infection was 6 (1.7%), 29 (8.3%) and 2 (0.6%) of B. caballi, T. equi and mixed infection respectively by Giemsa stained blood smears. This low rate of infection occurs due to this method has limited sensitivity and specificity detection especially during latent or carrier stage of infection with low level of parasitemia (24). However this method is still commonly used as a cheapest and fastest method for identification of parasites in acute infected animals (25). The overall seroprevalence of EP was 38.9% for all piroplasms. The rate of infection was 11.2% for B. caballi and 20.9% for T. equi, while the percentage of mixed infection was 6.9% by cELISA. Our study showed that T. equi infections were more commonly observed than B. caballi, in agreement with most previous studies from another province of Iraq (26) who found that 81.11% for T. equi and 18.88% for B. caballi of infection in Mosul Governorate. Furthermore, the seroprevalence rate of infection in Basrah Province, Iraq was 86.58% and 54.39% for B. caballi and T. equi respectively (27). Similar studies conducted in Iraqi neighboring countries were observed that T.equi and B. caballi seroprevalence between horses was 51% and 2% respectively in Iran (28), 16.21% and 9.6% respectively in Turkey (29) , 97.7% and 40.9% respectively in Oman (30), 10.4% and 7.5% respectively in Saudi Arabia (31) , 77.1% and 11.4% respectively in Kuwait (32). The higher prevalence of T. equi as illustrated in this study could be described due to that T. equi is usually life-long than B. caballi which generally remain in 4-5 years (33,34). Moreover, T. equi were more pathogenic than B. caballi and more public in the endemic area (7). These differences in the prevalence in different climatic place within a country and among nations may be due to variation in sensitivity of the diagnostic tests employed, differences in number and incidence of tick vectors, activity of equines and the presence and useful of control programs (10).
A study of the risk factors for EP identified in this study is presented in Table 2. In our study, there were no significant differences among the type of equids, gender and breeds these indicate that EP is widespread in Erbil governorate due to that EP are affected all type of equines as same as. Mixed infection was significantly higher in recreation and loading. These results are in agreement with another study (35). This may be due to that a physical stressor that may temporarily the immune system and immune-compromised animals have been shown to be more susceptible to infection compared with their immunocompetent animals (29). However, the odds of female gender being infected with B. caballi was approximately two times greater than male gender, which was widely consistent with other findings (36,37). The explanation is that the immune-suppression caused by stress during the third stage of pregnancy and parturition might be the result of higher protozoan infections in female equids, especially if persistently infected and so increase their chance of exposure to disease (38). The prevalence of T. equi was significantly higher among equids infested with ticks compared to equids not infested with ticks. This higher seropositivity of protozoa may be due to the ticks that are known to be their main vectors. These results agree with other studies (10,11), where the presence of ticks was the risk factor associated with equine piroplasms.
The results showed that the seroprevalence of T, equi and B. caballi and both protozoa infection was higher two times among equids kept with other animals compared to equids isolated in the stable and away from other animals. The same result was obtained already (39). This may be due to these animals acting as reservoirs for T. equi and B. caballi, as well as tick vectors (40).
This work revealed that the seroprevalence of all type of protozoa infection was significantly higher among equids that grazed than those kept in a stable. The equids that are grazed are more exposed to external environmental conditions that consistent with other findings (10,41) who found that equids in grazing lands were a high-risk factor for seropositivity of EP.

Conclusion
Piroplasmosis in equine seems to be widely distributed in Erbil Province, North of Iraq. Our results revealed that the prevalence of EP was considerably higher among equids in the stable when ticks were found on equids, and nearby animals than equids were no ticks infestation found on equine and nearby animals only. The relatively high seroprevalence rate of equine piroplasmosis observed in our study may be the cause of high tick vector population in Erbil province, which is greatly responsible for the transmission of the disease among animals. These findings may help in planning prevention and control strategies for EP in Erbil north of Iraq. However more studies on molecular characterization of EP with large-scale sampling of both equine and vector tick population are needed to investigate for the whole country to allow better control of EP in Iraq.