Toxoplasma gondii and Neospora caninum infections and factors associated in goats in the Parana state, Southern Brazil.

The aim of this study was to determine the occurrence of anti-Toxoplasma gondii and anti-Neospora caninum antibodies and their associated factors among goats in farms from the Paraná state, South Brazil. The serological analysis was performed by indirect ELISA on 629 goat serum samples collected from 32 farms distributed in five mesorregions from the state. Seropositivity was observed in 30.7% of the animals for T. gondii, 6.3% for N. caninum, and 3.0% for both agents. Final multiple regression models showed that the use of assisted reproduction decreased the chance of seropositivity for T. gondii (PR=0.70) and the slaughtering site on the farm (PR=2.03) increased it. To N. caninum, the supplying concentrated feed to animals (OR=4.80) and animal pre-weaning death (OR=9.96) increased the chance of seropositivity. The variables associated with these seropositivities suggest deficiencies in sanitation and reproductive management. In addition, it recognizes critical points of infection for animal production management.


Introduction
The national goat herd was estimated at 9,614,722 million heads (IBGE, 2014). According to the register of herds of the Secretariat of Agriculture and Supply, Paraná accounts for 1.6% of the national goat population, ranking eighth in the nation, with approximately 115,718 thousand heads (Paraná, 2019), demonstrating the importance of goat farming in the state. Among the diseases that affect goats is the toxoplasmosis, caused by Toxoplasma gondii, which is one of the main agents responsible for reproductive dysfunction. The prevalence of T. gondii in Brazil is quite variable, with reported prevalence rates of 14.4% in São Paulo (Mainardi et al., 2003), 17.1% in Rio Grande do Norte (Lima et al., 2008), 49.4% in Piauí (Rêgo et al., 2016), and 39.4% in Paraná (Garcia et al., 2012). In addition, to the direct damage caused by T. gondii, such as miscarriages and stillbirths, they can be major sources of infection to humans, through the consumption of raw or under cooked meat, raw milk, and its by-products (Dubey et al., 2014).
Toxoplasmosis and neosporosis represent important reproductive and economic losses in the country's goat herds (Dubey, 2010;Dubey et al., 2017). In addition, these diseases have an important role in public health, since this herds can be a source of T. gondii infection for men (Dubey & Beattie, 1988;Mesquita et al., 2013). Thus, the identification of factors associated with infection is essential to support programs for the development and promotion of goat farming worldwide (Topazio et al., 2014). The present study aimed to evaluate the seropositivity of T. gondii and N. caninum and the factors associated with their positivity in Paraná state, Brazil.

Materials and Methods
The study was approved by the Ethics Committee for Animal Use -EMBRAPA Goats and Sheep, Sobral -CE (number 010/2014). These activities were part of the project EMBRAPA (Zoosanitary characterization of goat and sheep farming in Brazil: epidemiology, associated factors and economic impact of diseases, number 02.12.01.032.00). To calculate the sample size (n), an infinite population was used as a reference using EpiInfo 7.2.3.1. The estimated prevalence were based on the study by Fortes et al. (2017), in which T. gondii was estimated to have a prevalence of 33.3%, and an error of 4%, resulting in a sample size of 533 animals, and the study by Figliuolo et al. (2004), who estimated a prevalence of 6.4% and an error of 2% for N. caninum, resulting in a sample size of 575 animals.
A non-probabilistic sampling was carried out to choose the farms because of a lack of a list of all goat producers in the state of Paraná. Farms were chosen in the municipalities with the greatest representation in goat farming from the mesoregions: Central North, Metropolitan of Curitiba, West, Southwest and Center-South. Within each farm, the animals used were stratified according to age (approximately 60% of adult females, 35% from 6 to 12 months old animals, and all adult males The blood was sampled by jugular puncture in a tube without anticoagulant and kept refrigerated until the centrifugation for obtaining the serum. The detection of anti-T gondii IgG antibodies was performed by an indirect ELISA technique as described by Garcia et al. (2007). The standardized T. gondii antigen concentration was 2.5 μg mL -1 , the sera were titrated at a ratio of 1:100 and the conjugate (Sigma Aldrich  , USA) was diluted to a ratio of 1:4,000. All the samples were tested in duplicate. The optimal conditions concerning antigen concentration, serum titration and conjugate dilution were established by the highest ratio of mean absorbance of positive samples to mean absorbance of negative samples. The cutoff point for each plate was obtained according to Garcia et al. (2006) using the mean negative sera plus three standard deviations. For the detection of anti-N. caninum antibodies, commercial indirect Elisa kits (Imunodot  , Jaboticabal, Brazil) were used. The cutoff index (CFI) was calculated by multiplying the mean optical density of the negative controls by 2.5; the positive samples exhibited a bright yellow color and had an optical density equal to or greater than the CFI, as described by the manufacturer.
A questionnaire was applied to each property with information such as: infrastructure, characteristics of production and commercialization, technological profile, sanitation and presence of other species on the property, domestic and wild. Of each animal whose blood was collected, sex, age and breed information were obtained. The EpiInfo 7.2.3.1 program was used to tabulate the variables from the epidemiological questionnaire along with the serological results and for bivariate analysis. The variables from the epidemiological questionnaire were screened by Yates-corrected chi-square (χ 2 ) and Fisher's exact tests, and those with P <0.20 were selected for inclusion in the regression analysis. The existence of associated factors was estimated by multiple logistic regression or multiple log-binomial regression, and the strength of the association was estimated by odds ratio (OR) or prevalence ratio (PR) and their respective 95% confidence intervals using R environment 3.6.2 version, packages epitools (Aragon, 2020) and epiDisplay (Chongsuvivatwong, 2018).

Results and Discussion
This study included 19 municipalities from five of the ten mesoregions in the state of Paraná, and 629 animals from 32 farms were evaluated. The distribution of the agents on the farms indicated that 90.6% (29/32) of the farms were positive for T. gondii and 53.1% (17/32) were positive for N. caninum, demonstrating an important distribution of agents on farms in Paraná ( The bivariate data analysis revealed an association between the T. gondii positivity and the following variables: use of assisted reproduction methods on the farm, the presence of cats on the farm, the presence of wild felines on the farm, the presence of animals who drink water directly from the source, the existence of a slaughtering site on the farm, supplying placental remains to cats, supplying concentrated feed to animals, and the presence of animals over eight months ( Table 2). In the final multiple regression model, the following variables were adjusted for: the use of assisted reproduction methods on the farm (P = 0.0042; PR = 0.70, 95% CI = 0.55-0.89) and the existence of a slaughtering site on the farm (P <0.0001; PR = 2.03, 95% CI = 1.59-2.60) ( Table 3).
The assisted reproduction variable (controlled breeding and artificial insemination), compared to uncontrolled natural breeding, was a protective factor for the positivity of T. gondii in the present study. Pereira et al. (2012) found a significant association for natural breeding, which increased the chance of toxoplasmosis occurrence by 6.09, in the multivariate analysis. Farms in which assisted reproductive techniques are used may represent the adoption of good production practices and management and improved selection of breeders, indicating proper sanitary care as well as beneficial management techniques. In addition, it was not a significant association between reproductive impairment and toxoplasmosis and/or neosporosis. Similar results are observed with N. caninum prevalence in goats of Paraíba state (Braz et al., 2018); however, it was contrary with other studies about T. gondii.
The slaughter of animals on rural farms, even if transported to a specific site for this purpose, does not always occur under satisfactory hygienic-sanitary conditions. In addition, blood and viscera may collect on the floor, which is highly attractive to cats. The presence of rodents in supplement stores is another condition that is attractive to cats that end up defecating and eliminating oocysts that are later ingested by the livestock via food (Romanelli et al., 2007). Weigel et al. (1999) suggested that the presence of cats on rural farms is the main risk factor for environmental contamination and human infection by T. gondii. Cavalcante et al. (2008) reported that the risk of T. gondii infection  The totals of the frequencies varied according to the number of questionnaires answered by the owners of the animals. P: p value; PR: prevalence ratio; CI: confidence interval.
in goat was increased by 1.73 times in farms from Ceará state with more than 10 cats This finding is explained by the free breeding of these animals which had access to various zootechnical facilities, pastures and natural water supply and storage sources. Although wild felids do not inhabit the human peridomiciliary, they have free access to pastures, where they often feed on animal carcasses and placental remains, become infected and subsequently spread T. gondii oocysts throughout the environment through their feces.
Regarding age, a significant association was observed for animals older than eight months, which increased the chance of toxoplasmosis by 2.53 times. Anderlini et al. (2011) found a positive association for the prevalence of T. gondii in adult animals compared to younger goats in herds from Alagoas state. Jittapalapong et al. (2005) observed that older animals (older than two years) were 2.70 times more likely to be T. gondii-seropositive. This is explained by the fact that there is an increased chance of animals becoming infected over time due to oocyst ingestion from the contaminated environment.
In relation to water source, Cavalcante et al. (2008) found that contamination was associated with the poor sanitation of drinking fountains and facilities used for capturing and storing large volumes of water, which are practically impossible to sanitize, thus becoming sources of infection to animals. This result was corroborated by Villari et al. (2009), who described a correlation between untreated water supplied to animals and the presence of anti-T. gondii antibodies. The epidemiological analysis results of water contamination by T. gondii oocysts can be extrapolated to water contamination by N. caninum oocysts from dog feces. The implementation of good hygiene practices on the farms is essential to avoid environmental contamination of food and water by T. gondii and N. caninum oocysts (Machacova et al., 2015).
The variables supplying concentrated feed to animals and animal pre-weaning death were factors associated with the positivity of neosporosis in goats in Paraná state according to the bivariate analysis ( Table 4). In the final multiple logistic regression model, two variables were associated factors, supplying concentrated feed to animals, which increased the odds of N. caninum seropositivity by 4.80 times (P = 0.0356; 95% CI = 1.32-74.70), and animal pre-weaning death, which increased the odds by 9.96 times (P = 0.0253; 95% CI = 1.11-20.75) ( Table 5). Table 5 shows that the odds ratio confidence intervals calculated by the logistic regression method are wide, and this maybe is due to the small number of seropositive animals for N. caninum. Feed supplementation with concentrates is a necessary practice in ruminant animal breeding during grazing shortages. However, the storage of this food product requires appropriate conditions. Although the presence of dogs was not a significant variable in this study, dogs are commonly found sheltering in these places, where they end up defecating, eliminating N. caninum oocysts in the feces and contaminating stored food, as described by Dubey & Schares (2011). In the present study, a mortality rate of 8.6% pre-weaning period was observed, but no etiological diagnosis was made. The observation of neonatal deaths or miscarriages may not be a direct result of neosporosis; however, there is a likelihood of spread among herds when dogs eventually feed on placental or miscarriage material or the remains of neonates who were infected Table 3. Final models of multiple regression analysis from variables statistically associated with the positivity of anti-Toxoplasma gondii antibodies in goats from Paraná state, Brazil, (P<0.05).

Toxoplasma gondii
Multiple log-binomial regression