Toxoplasma gondii infections in chickens - performance of various antibody detection techniques in serum and meat juice relative to bioassay and DNA detection methods.

Chickens, especially if free-range, are frequently exposed to Toxoplasma gondii, and may represent an important reservoir for T. gondii. Poultry products may pose a risk to humans, when consumed undercooked. In addition, chickens are regarded as sensitive indicators for environmental contamination with T. gondii oocysts and have been used as sentinels. The aim of the present study was to determine the suitability of commonly used antibody detection methods, i.e. the modified agglutination test (MAT), IFAT and ELISA to detect T. gondii-infected chickens. Samples of experimentally and naturally infected chickens were used. The infection state of all chickens was determined by Magnetic-Capture (MC-) real-time PCR (RT PCR). Naturally exposed chickens were additionally examined by mouse bioassay and conventional RT PCR on acidic pepsin digests (PD-RT PCR). Blood serum and meat juice of various sources were tested for antibodies to T. gondii. In naturally infected chickens, there was substantial agreement between the mouse bioassay and MC-RT PCR or the mouse bioassay and conventional PD-RT PCR. PD-RT PCR was slightly more sensitive than MC-RT PCR, as all (26/26) bioassay-positive chickens also tested positive in at least one of the tissues tested (heart, drumstick). By MC-RT PCR, 92.3% (24/26) of the naturally infected bioassay-positive chickens were positive. The diagnostic sensitivity of MC-RT PCR was clearly related to the organ examined. Based on a quantitative assessment of the MC-RT PCR results in experimentally infected chickens, brain and heart tissues harbored an at least 100 times higher parasite concentration than breast, thigh or drumstick musculature. In naturally infected chickens, only three out of 24 birds, which were MC-RT PCR-positive in heart samples, also tested positive in drumstick musculature. Under experimental conditions, the agreement between MC-RT PCR and the serological techniques revealed 100% diagnostic sensitivity and specificity. Under field conditions, examinations of sera by ELISA, IFAT and MAT showed good performance in identifying chickens that were positive in either a mouse bioassay, MC-RT PCR, or PD-RT PCR as illustrated by diagnostic sensitivities of 87.5%, 87.5% and 65.2%, respectively, and diagnostic specificities of 86.2%, 82.8% and 100%, respectively. The examination of meat juice samples from breast, drumstick or heart musculature revealed similar or even better results in the ELISA. The results in the MAT with meat juice from breast musculature were less consistent than those of ELISA and IFAT because a number of negative chickens tested false-positive in the MAT. The MAT performed similar to ELISA and IFAT when applied to test meat juice samples collected from heart, thigh or drumstick musculature.