The purpose of our study was to evaluate the prevalence of Toxoplasma gondii infection in autochthonous Carpathian buffaloes from northwestern Romania by serology, PCR techniques, and mouse bioassay. Agreement between MAT and ELISA, correlation between indirect and direct detection methods, and risk factors were evaluated. The apparent overall seroprevalence of T. gondii was 8.1% by MAT and 6.6% by ELISA. The agreement between ELISA and MAT was fair. The apparent seroprevalence was significantly higher in adult buffaloes (12.5%) compared to calves (0.0%) and juveniles (1.9%) by MAT. Most of the positive adult buffaloes detected by MAT had antibodies at a low sera dilution and the highest dilution was 1:768 in a juvenile female (30 months). No viable T. gondii was detected by mouse bioassay, as no T. gondii cyst or DNA was found in the brain of mice and they did not seroconvert. However, T. gondii DNA was detected in two buffaloes: in a 30-month-old male buffalo by qPCR on the diaphragm digest and in a 252-month-old female buffalo by RE nPCR on the mesenteric lymph node. Both animals were negative in MAT and ELISA. The total prevalence of T. gondii by direct detection methods was 2.7%. There was no correlation between indirect and direct detection methods. Since no viable T. gondii was detected in buffaloes, the risk of human infection from buffalo meat is minimal. Buffaloes' biological response to a T. gondii infection appears to be very similar to the response of cattle.
Keywords: Buffaloes; ELISA; MAT; PCR; Seroprevalence; Toxoplasma gondii.