Sarcocystis neurona , seroprevalence of antibodies in equines and research of oocystis in opossum in Ilhéus Itabuna microregion, Bahia, Brazil

seroprevalence of in Abstract The aims of this study were to determine the seroprevalence of Sarcocystis neurona antibodies in equines in the Ilhéus-Itabuna microregion (BA), and identify possible factors associated with infection. The presence of sporocysts/oocysts of Sarcocystis spp. was also verified in Didelphis spp. A total of 669 serum samples were collected from equines in 56 properties located in 12 municipalities in the region. Indirect fluorescent antibody test was performed with slides containing merozoites of the S. neurona , using a cut-off titer of 1:80. Occurrence of 7.92% of anti- S. neurona antibodies was observed in the sampled equines. The purposes trade and work were significantly associated with the presence of antibodies (p<0.05), and being used for the purpose of work (21.6%) was considered a risk factor, while being used for the purpose of trade (3.6%) was a protective factor. A total of 25 Didelphis spp. was captured for research on sporocysts/oocysts in stool samples and intestinal scrapings, being all negative. Didelphis spp. were all negative for the presence of Sarcocystis spp. and this circumstance does not change the fact that seroprevalence of S. neurona has been observed in horses raised in the southern Bahia. from Federal University of Bahia -UFBA. The conjugate used was equine anti-IgG (Sigma, F7759) at a dilution of 1:600. The sera were considered positive with the 1:80 cut-off titer when they showed complete fluorescence of S. neurona merozoites (Pusterla et al., 2014). Positive samples were subjected to sequential dilutions (1:80, 1:160, 1:320...), until negative results.


Introduction
Brazil has a herd of more than 5 million horses (Lima & Cintra, 2016). It represents the third largest herd in the world and the largest in Latin America, moving approximately R$ 16.15 billion annually and generating around 610 thousand direct jobs and 2.4 million indirect jobs (Lima & Cintra, 2016). Bahia occupy the 3rd position on the national ranking, with 8.6% of the equine herd (IBGE, 2016).
The definitive hosts of S. neurona are marsupials of the Didelphis genus, Didelphis virginiana in North America (Dubey & Lindsay, 1998), and D. albiventris in South America (Dubey et al., 2001a). In South America, other species of opossum (D. aurita and D. marsupialis) may be involved in the transmission of the agent (Silva et al., 2003). A variety of mammals may act as intermediate hosts, including skunks, raccoons, armadillos, cats and sea otters (Dubey et al., 2015).
Horses accidentally become infected when they ingest water and food contaminated with sporocysts eliminated in the feces of the definitive host. After ingestion, sporozoites are released into the small intestine and asexual reproduction begins in the vascular endothelium, from where the parasites, in the form of merozoites, migrate into the bloodstream, cross the blood-brain barrier, and reach the Central Nervous System (MacKay, 2001).
The main risk factors for the emergence of EPM are related to age, geographic proximity to the occurrence of the definitive host, environmental factors (climate and hydrography), food management system, purpose of animal husbandry (sport, reproduction and work) being exposed to greater stress situations, that is, transportation, intense training, participation in races, postpartum and constant activities (Reed & Bayly, 2000;Dubey et al., 2015).
Anti-S. neurona antibodies have been found in equines in several countries with seroprevalence ranging from 0% in New Zealand (Vardeleon et al., 2001) to 89.2% in the United States (Bentz et al., 2003). In Brazil, records of equine exposure to the protozoa are scarce, with few studies having verified seroprevalence of S. neurona infection in the equine troop, with numbers ranging from 1.6% in Alagoas Northeast Region, (Valença et al., 2019) to 90,00% in Minas Gerais (Hoane et al. 2006).
Equine protozoan myeloencephalitis (EPM) had its first case reported in Brazil in Rio Grande do Sul, by Barros et al. (1986) in a 10-years-old horse. Clinical case reports or studies investigating EPM have been reported in several Brazilian states (Masri et al., 1992;Lins et al., 2008;Faria et al., 2017). Henker et al. (2020) concludes that EPM due to S. neurona infection represents an important neurologic disease of horses in Brazil and this disease should be considered as a main differential diagnosis in horses presenting neurologic signs.
Bahia is a state located in the Northeast region of the country. Although the data have not been published, some clinical cases characteristic of EPM have already been diagnosed in the southern mesoregion of the state. Therefore, the aim of this study was to determine seroprevalence and factors associated with S. neurona infection in equines in the Ilhéus-Itabuna microregion, Bahia, Brazil. To identify the dynamics of the disease in this population, Sarcocystis infection in the definitive hosts (opossums) of the region was also evaluated.

Study population and sample collection
The study was carried out in the South Bahia mesoregion, Ilhéus-Itabuna microregion, comprising 12 municipalities with a high number of horse farms, these being: Itabuna, Itapé, Itajú do Colônia, Santa Cruz da Vitória, Ibicaraí, Floresta Azul, Jussari, Buerarema, São José da Vitoria, Gongogi, Barra do Rocha and Ibirataia. The total number of properties and animals by municipality was established from the IBGE database, referring to the year 2015 (IBGE, 2015) to guide the collection effort. However, the sampling was done by convenience, due to the logistical issues, such as the needed of permission from the owner for collection, horse available and personality, ability of installation and displacement of the researchers, geographical extension range of the study (as 12 municipalities) and access facility of the farm. In that way, the maximum number of properties sampled reached 56, being at least three per municipality. In addition, the number of animals per properties or municipality was conditioned also according to the logistical issues already mentioned.
All the properties were georeferenced with a GPS device to assess the distribution of the premises and the distribution of the health status of the animals.
The horses were selected regardless of gender, breed, age and economic purpose. The number of equines sampled per property was defined for convenience (availability and behavior of the animals).
Blood samples (from 5 to 10 ml) were collected from the equines through puncture of the jugular vein and packed in siliconized tubes containing anticoagulant (EDTA), and maintaining at 5 °C for up to 5 hours until centrifugation, for 10 minutes, at 900 G. The serum was separated and stored in microtubes at -20 °C until serology was performed.
This project was developed after approval by the Ethics Committee for the Use of Animals (CEUA) of the State University of Santa Cruz -UESC, with Process no. 002/2013 and no. 09/2018, having respected and followed all the recommended ethical principles regarding the use of animals in experiments.

Gathering information using the interview form
To determine the factors associated with S. neurona seropositivity, a structured interview was conducted with the owners or those responsible for the farms. Objective questions included topics related to the main activity of the property; purpose of the equines; age; sex; interaction with other animals; sanitary, nutritional and reproductive management; observation of clinical neurological signs characteristic of EPM; and presence of forest area and opossums.

Detection of anti-S. neurona antibodies
The serological test for Indirect Fluorescent Antibody Test (IFAT) was performed with merozoites of the S. neurona (SN138 strain), produced in cell culture. Positive and negative controls for S. neurona were performed on each slide. These materials were provided by Dr. Luis Fernando Pita Gondim from Federal University of Bahia -UFBA. The conjugate used was equine anti-IgG (Sigma, F7759) at a dilution of 1:600.
Didelphis spp.: capture, collection of biological material and diagnosis of Sarcocystis spp.
The capture of Didelphis spp. was performed using live traps (Tomahawk -50 × 17 × 17 cm), distributed in two fragments of each sampled property. Fragment A was located in a built area, where three traps were allocated. Fragment B was located in the pasture area, where two traps were placed, therefore totaling 5 traps per property.
The traps were baited at 6:00 pm with a mixture of cornmeal, banana, oats, peanut candy, cod liver oil and sardine, and monitored daily 6:00 am for 5 consecutive days. Didelphis spp. found accidentally dead on the studied properties were also collected. We sampled 25 Didelphis spp. in which 22 animals were captured through the trap and 3 animals were found dead on the properties.
The traps with captured Didelphis spp. were packed in burlap bags to be transported to the physical facilities of the study areas for further handling and sample collection. Didelphis spp. were euthanized with intraperitoneal administration of thiopental sodium anesthetic (120 mg/kg).The anesthetic dose was calculated according to the animal weight. The anesthetic dose was calculated according to the animal weight. For each captured species, stool and intestines (small and large intestines) were collected from each animal, and stored in plastic bags, which were identified and packed and kept under refrigeration for later analysis. The animals were recorded with photographs, identified as to species and sex (Bonvicino et al., 2008). When puppies, pregnant females and lactating females were captured, it was evaluated: the presence of milk in the teats and the presence of puppies in the marsupial pouch. In view of the finding, these animals were released at the place of origin of the capture. The same release occurred with young animals. The stool samples were processed using a centrifuge-fluctuation technique modified from Menezes (1994). After processing the content was stored in 2.5% potassium dichromate solution for conservation.
The entire intestinal mucosa (small and large intestine) of each euthanized animal was scraped. Digestion was scraped and homogenized in distilled water with a mixer. Subsequently, 5% sodium hypochlorite was added to the mixture, which was placed in an agitator for 30 minutes at room temperature. The material was filtered in gauze and passed through serial centrifugations at 1957g for 10 minutes to remove the hypochlorite. The sediment was re-suspended in a "Hank's Balanced Salt Solution" for microscope observation.
Stool and intestinal scrapings were analyzed under an Olympus BX51 optical microscope to verify the presence of oocysts/sporocysts of Sarcocystis spp.

Statistical analysis
To assess the factors associated with the occurrence of the disease in horses, we identified some variables both intrinsic to the animal and the environment in which it lives, respecting the dynamics of infection by the parasite. In this sense the associations between these characteristics with the diagnosis result were evaluated using the chi-square test (χ2) and Fisher's exact test. Variables with a significance level of p ≤ 0.2 were considered candidates for the multivariable model, including all biologically plausible bidirectional interactions. The step-by-step revision approach was used and the best fit model was defined as one that included significantly associated variables (p value <0.05) and minimum Akaike Information Criterion (AIC) value.
Variables not included in the best model were subjected to univariate analysis. All analyzes were performed on R software (R Core Team, 2017), version 3.4.2.

Results
Of the 669 horses sampled 259 were male and 410 were female; 145 were young (equines ≤ 3 years), 283 were adult (equines >3 <10 years) and 242 were senior (equines ≥ 11 years). The IFAT revealed that 53 animals (7.92%) presented IgG antibodies for S. neurona, with titers of 80 (n= 50), 160 (n=2) and 320 (n=1). Most farms had more than 50% of their animals sampled. Ten out of 56 farms had at least one seropositive equine ( Table 1). Table 1. Number of animals collected by municipality and by property and detection and distribution of anti-S. neurona antibodies in horses in municipalities in the Ilhéus-Itabuna micro-region, Bahia, Brazil.

Municipality
Number of animals (IBGE*) Of the seropositive equines, 9.70% (25) were male and 6.80% (28) female; 7.58% (11) were young, 6.77% (19) adult and 9.50% (23) senior. There was no significant difference between sex or age range (Table. 2). The purposes of trade and work were significant (p<0.05), such that being used for the purpose of work (21.6%), was a risk factor (OR>1) and being used for the purpose of trade (3.6%) was a protective factor. There was no association between seropositivity of horses as to their purpose for sport and commerce, type of feeding, management of the stall, management of pasture and presence of neurological signs ( Table 2). Among the analyzed municipalities, Gongogi presented the highest occurrence of seropositive horses (16.0%), whereas there were no positive animals for anti-S. neurona antibodies in the municipalities of Ibirataia and Itabuna ( Figure 1; Table 1). There was no statistical difference among them. A total of 25 Didelphis spp. was captured, being 10 females (40%) and 15 males (60%), 60% (15/25) belonging to the D. aurita species and 40% (10/25) to D. albiventris.
Of the 18 properties where Didelphis spp. were captured, 6 (33,3%) presented equines seropositive in the IFAT. Of the 38 properties that had no successful captures, 25 (65.70%) presented positive equines. However, the association between the presence of Didelphis spp. on the properties and positive equines was not statistically significant, also there was no statistical significance for those properties that border or possess forest or bush area with S. neurona positive equines.
Of the 25 Didelphis spp. captured, all were negative for the presence of oocysts and sporocysts in the analyzed stool samples and scraping of small and large intestines. It is worth noting that the low number of animals captured was due to restriction imposed by the Ethics Committee for the Use of Animals (CEUA) of the Santa Cruz State University-UESC.

Discussion
The seroprevalence for S. neurona antibodies observed in the equines sampled in this study (7.92%) was similar to that described in Alagoas state, northeast Brazil (2.8%) (Valença et al., 2019), but considerably low compared to the numbers observed in Minas Gerais state, southeast Brazil (26.0%) (Ribeiro et al., 2016), and in the United States (27,6%) (Pusterla et al., 2014). All the above studies used the same serological test and cut-off values as the present study. The IFAT is an excellent choice, with a sensitivity of 94% and specificity of 88.9%. The test allows quantifying the infection through the final title of it (Duarte et al., 2003;Johnson et al., 2013).
The prevalence of S. neurona in horses is related to several intrinsic factors (physical, immunological condition and recent stress history) and extrinsic factors (environmental conditions and proximity to areas of occurrence of the definitive positive host), but which will determine high or low prevalence is the direct exposure of the horse to the S. neurona through feces of positive Didelphis spp.
In the present study, the 25 Didelphis spp. captured in the properties were negative for Sarcocystis spp. No sporocysts were found in the intestinal scrapings of all 39 examined opossums in Bahia, however Sarcocystis spp. DNA was identified in tissues of 16/39 (41%) animals (Gondim et al., 2017). S. neurona, that had been reported in Brazil infecting South American opossum D. albiventris (Dubey et al., 2001b), but has never been genetically confirmed in the country. S. lindsayi have been reported in D. aurita, but most isolates of Sarcocystis spp. derived from opossums reported in Brazil are S. falcatula-like organisms, as they are infective to birds (Stabenow et al., 2012;Valadas et al., 2016;Cesar et al., 2018;Gallo et al., 2018).
The relationship of Didelphis spp. and Sarcocystis spp. may be associated with some behavioral characteristics of skunks, as it is an omnivorous animal with general eating habits, since their diet is composed of a wide range of items including invertebrates, small vertebrates, eggs, brans, birds, fruits, carcasses and occasionally carrion, flowers, nectar and tree gum (Aléssio et al., 2005;Santori & Moraes, 2006). Thus, changes in diet lead to the possibility of negative animals since contamination occurs through the ingestion of Sarcocystis present in the muscle tissues of intermediate hosts.
The Atlantic Forest region has a great diversity of food for opossums, with a large quantity and variety of fruits and insects. Such availability diminishes the need to search for other types of food, such as intermediary hosts that are part of the biological cycle, since this species is omnivorous. This fact contributes to a lower probability of Didelphis spp. being infected with S. neurona, and, therefore, corroborating the low prevalence of infected species in the present study.
Another fact that corroborates the negativity of Didelphis spp. is its presence at some time on the studied properties, being observed on 92.86% (52/56) of the properties. This fact may be associated with the synanthropic characteristics of Didelphis spp. Opossums adapt easily to human coexistence, colonizing houses and their surroundings, taking advantage of shelter because they adapt to house linings and hollow trees, have access to food (household food waste, bird breeding, animal feed) and water availability (Andrade et al., 2002). This evidence of the migration of these animals onto the properties may correlate with the ease and unrestricted supply of food other than intermediate hosts. Although no opossums were found positive in this study, as a reduced number of animals was used, other opossums may be positive for S.neurona and contaminate the environment with feces containing sporocysts.
According to Dubey et al. (2015), an animal breeding servant can influence the possibility of the animal becoming infected. This corroborates the data found in this study. The use for work purposes (21.6%) was a risk factor. It is worth considering that within this topic "work" included in this variable, it considers horses that make trips transporting cattle to other properties, perform routine work inside the property, are used to go in the city as a means of transport. This makes it possible for horses to be exposed to different environments on the property, as well as other rural environments, at different times of the year, increasing the scope for interaction with the environment, other farms, having access to various foods such as chopped grass, hay, pastures, as well as access as pasture areas to the forest margin, and these environments may be contaminated with S. neurona sporocysts. Another important fact is that the equines remain on the property for many years, with a life expectancy of 20 years. This enables long-term interaction with Didelphis spp., which predisposes the equines to S. neurona infection (Morley et al., 2008).
Being used for the purpose of trade (3.6%) was considered a protective factor. This fact may be associated with the short time these animals spend on the property, since commercialization is rotational, as these animals present characteristics desired by the market such as breed, coat color, age, height, temperament and potential usability. Another relevant aspect is the management of the animals destined for trade; on most of the properties, the hostlers provided special care regarding the cleaning of the stalls, individual pickets per animal and directed food management.
Although Bentz et al. (2003) and Duarte et al. (2004) reported a direct association between age and seropositivity for S. neurona, the present study did not prove this relationship, corroborating the results of Pivoto et al. (2014) and Ribeiro et al. (2016). There was no association between seropositivity of horses and sex, which is in agreement with the findings of Blythe et al. (1997).

Conclusion
Horses raised in the south of Bahia mesoregion, in the Ilhéus-Itabuna microregion, have a low anti-S. neurona. The fact that these animals are destined for work is considered a risk factor as well as equine care for trade is a protective factor. Didelphis spp. sampled in the present study were negative for Sarcocystis spp.