Risk factors for Toxoplasma gondii and Neospora caninum seropositivity in buffaloes in Paraiba State, Brazil

The aims of this survey were to determine the frequency of anti-Toxoplasma gondii and anti-Neospora caninum antibodies and to identify the risk factors associated with seropositivity among buffaloes in the state of Paraíba, Brazil. This survey included 136 buffaloes belonging to 14 herds. To detect anti-T. gondii and anti-N. caninum antibodies, the indirect fluorescent antibody test (IFAT) was used. Among the 136 samples analyzed, 17 (12.5%) were positive for anti-T. gondii antibodies with titers ranging from 64 to 1,024, and 26 (19.1%) for anti-N. caninum with titers from 200 to 1,600. Animals seropositive for both T. gondii and N. caninum were found in 10 of the 14 herds (71.4%). Semi-intensive management systems (odds ratio = 2.99) and presence of pigs (odds ratio = 4.33) were identified as risk factors for T. gondii and N. caninum, respectively. It can be suggested that T. gondii and N. caninum are widespread in buffaloes in Paraíba, and that additional surveys are needed in order to ascertain the importance of these agents for this species and for pigs, and the influence of the farming type on occurrences of seropositive animals.


Introduction
Buffaloes are known for their rusticity, tolerance and resistance to infectious and parasitic diseases, and are better adapted than cattle to adverse environmental conditions. Because they are versatile animals, they have become an excellent alternative for livestock farming in Brazil, since they have triple aptitude (milk production, meat production and traction), in addition to their easy adaptability to hostile environments, such as flood-prone lands, coastal areas, mountains and plains (NARDI, 2005).
Currently, Brazil has a prominent position within buffalo-rearing and has a headcount of more than a million animals, of which the Northeastern region accounts for around 122,000 animals. In the state of Paraiba, the buffalo headcount totals approximately 933 animals (IBGE, 2012).
Increasing productivity and technification of buffalo-rearing has favored introduction of infectious and parasitic diseases into the herds, as a result of changes in management, increased animal density and confinement. Toxoplasma gondii and Neospora caninum have not yet been shown to be important causative agents of abortions among buffaloes, but this possibility cannot be discarded. Investigations on the implications of these agents in relation to reproductive losses in this species are needed (GONDIM et al., 2007).
T. gondii and N. caninum are two morphologically related protozoa that present cosmopolitan geographical distribution. Both species may infect a broad range of animal species, including buffaloes. They have life cycles in which the definitive hosts are carnivore species and both of them cause abortions and congenital diseases in ruminants DUBEY, 2003). Recent studies on occurrences of anti-T. gondii antibodies in buffaloes in Brazil have revealed rates ranging from 1.1% to 49.9% (SILVA et al., 2010;SOUZA et al., 2001). For N. caninum, high frequencies of antibodies in buffaloes were found in studies in the north (40.9%) (SILVA et al., 2010), southeast (64%) (FUJII et al., 2001) and northeast (35.9%) (GONDIM et al., 2007).
Because of the lack of studies on bubaline toxoplasmosis and neosporosis in the state of Paraiba, the present study aimed to determine the frequency of anti-T. gondii and anti-N. caninum antibodies in buffaloes in Paraiba, as well as to identify the risk factors associated with seropositivity.

Sampling
This study was carried out on 14 farmed buffalo herds in the counties of Alagoa Nova, Areia, Campina Grande, Guarabira, Juripiranga, Santa Helena, Sapé, Rio Tinto, Santana dos Garrotes, Itatuba, Solânea and Cacimbas (Figure 1). The study population was formed of bubaline females with an aptitude for meat and milk, of mixed breed and Murrah breed, with ages ≥ 24 months. To calculate the number of animals to be sampled, the formula (Equation 1) for simple random sampling (THRUSFIELD, 2007) was used: where: n = number of animals to be sampled Z = normal distribution value for the 95% confidence level P = expected prevalence of 50% (for maximization of the sample) d = error of 10% In total, 136 bubaline females with ages ≥ 24 months were selected from these 14 herds. Blood samples were collected between November 2012 and July 2013, by puncturing the jugular vein, using a disposable needle and a vacuum tube (without anticoagulant) with capacity for 15 ml. At the time of the blood collection, an epidemiological questionnaire was applied in order to obtain data for use in the risk factor analysis. The variables and respective categories used were as follows: management system (intensive, semi-intensive or extensive); type of exploitation (meat, milk or mixed); type of milking (manual or mechanical); number of milkings per day (none, once a day or twice a day); presence of other animal species (cattle, horses, goats/sheep, pigs, poultry, dogs or cats); presence of wildlife (yes or no); occurrences of miscarriages during the last 12 months (yes or no); presence of rodents (yes or no); use of rodent control (yes or no); feeding on native pasture (yes or no); water source (drinking troughs or watering points); animal purchases (yes or no); pasture rental (yes or no); presence of flooded areas (yes or no), presence of maternity pens (yes or no); separation of young from adult animals (yes or no); and presence of veterinary assistance (yes or no).

Serological diagnosis
To detect anti-T. gondii and anti-N. caninum antibodies, the indirect fluorescent antibody test (IFAT) was performed at the Laboratory of Parasitic Diseases, Federal University of Campina Grande, Patos, PB, Brazil. For T. gondii, the technique was carried out in accordance with the methodology described by Camargo (1974), using tachyzoites of the RH strain as antigens and a cutoff point of 1:64 (SILVA et al., 2010). For N. caninum, the cutoff point was 1:200 (GONDIM et al., 2007;SILVA et al., 2010) and the technique was carried out in accordance with Paré et al. (1995), using an NC-1 sample of N. caninum cultured in bovine monocytes as the antigen.

Statistical analysis
Risk factor analysis was performed in two steps: univariable and multivariable analysis. Univariable analysis was performed using the chi-square test or Fisher's exact test, and variables that presented P ≤ 0.20 were used for multivariable logistic regression. The multivariable analysis was then performed, using the stepwise forward method (HOSMER & LEMESHOW, 2000). Collinearity between independent variables was verified by a correlation analysis; for those variables with a strong collinearity (correlation coefficient > 0.9), one of the two variables was excluded from the multiple analysis according to the biological plausibility (DOHOO et al., 1997). The significance level in the multivariable analysis was 5%, and the SPSS 20.0 software for Windows was used.
Tables 1 and 2 show the variables associated with seropositivity for T. gondii and N. caninum (P ≤ 0.2), from the univariable analysis. In the final logistic regression model (Table 3), the semi-intensive management system was identified as a risk factor for toxoplasmosis (odds ratio = 2.99) and the presence of pigs was a risk factor for neosporosis (odds ratio = 4.33).
It is known that buffaloes are considerably resistant to T. gondii infection (DUBEY, 1988). However, surveys carried out in Brazil such as those of Souza et al. (2001) and Silva et al. (2013), in which the frequencies of antibodies were 49.9% and 36%, respectively, have suggested that there is a need for further studies on this parasitosis in buffaloes. In the present survey, the presence of antibodies was indicative of T. gondii circulation in buffalo herds in the state of Paraíba. This is a public health concern, given that in Brazil there has been an increase in the consumption of raw or undercooked beef, especially in areas with no federal inspection service (GUERRA et al., 2014).
The frequency of seropositivity for N. caninum among these buffaloes in the state of Paraíba was 19.1%, and serological surveys for N. caninum in buffaloes in Brazil have shown frequencies ranging from 40.9% (SILVA et al., 2010) to 49.9% (SOUZA et al., 2001). In cattle, neosporosis is considered to be the greatest cause of abortions, but its impact on buffaloes' reproductive sphere remains unclear. Given that there have been  RODRIGUES et al., 2004;CHRYSSAFIDIS et al., 2011), neosporosis needs to be investigated as a possible cause of abortions in this species. It is believed that the low frequency of seropositive animals for both T. gondii and N. caninum compared with other surveys conducted in Brazil could be explained by the low occurrence of definitive hosts in the herds since in 60% of the positive herds for T. gondii there wasn't report of cats, and in 40% of the positive herds for N. caninum the presence of dogs was not reported (data not shown).
Surveys on risk factors for toxoplasmosis among buffaloes are scarce in Brazil. In the present work, the semi-intensive management system was identified as a risk factor, as also found by Albuquerque et al. (2011) for cattle. This can be explained by the greater chances of contact with oocysts in the environment when this system is used, probably due to the fact that such contacts are favored both in the confinement and in the extensive management system. In the semi-intensive system, the animals remain part of the day free in pastures and part of the day confined, and are thus exposed to both environments.
The presence of pigs was identified as a risk factor for occurrences of anti-N caninum antibodies in buffaloes in the state of Paraíba. Even though the presence of pigs does not has direct importance with regard to transmission of this agent to buffaloes, Azevedo et al. (2010) reported findings of pigs positive for N caninum in Paraiba, using IFAT and immunoblotting with total and purified antigen (p38 -NcSRS2) of N. caninum tachyzoites. This suggests that complementarity of life cycles in the herds may exist when definite hosts are present. Therefore, studies aiming to isolate the agent in pigs and buffaloes are necessary in order to elucidate the real importance of neosporosis in these species.
The frequencies of buffaloes seropositive for T. gondii and N. caninum in the state of Paraíba, northeastern Brazil suggest that both of these agents are spreading in this region. From our risk factor analysis, it is suggested that further surveys on the importance of these agents in this species and in pigs are needed, as well as evaluation of the influence of the type of management system on occurrences of seropositive animals.