Evaluation of four DNA extraction protocols for Brucella abortus detection by PCR in tissues from experimentally infected cows with the 2308 strain.

This study compared 4 protocols for DNA extraction from homogenates of 6 different organs of cows infected with the Brucella abortus 2308 strain. The extraction protocols compared were as follows: GT (guanidine isothiocyanate lysis), Boom (GT lysis with the carrying suspension diatomaceous earth), PK (proteinase K lysis), and Santos (lysis by boiling and freezing with liquid nitrogen). Positive and negative gold standard reference groups were generated by classical bacteriological methods. All samples were processed with the 4 DNA extraction protocols and amplified with the B4 and B5 primers. The number of positive samples in the placental cotyledons was higher than that in the other organs. The cumulated results showed that the Santos protocol was more sensitive than the Boom (p=0.003) and GT (p=0.0506) methods and was similar to the PK method (p=0.2969). All of the DNA extraction protocols resulted in false-negative results for PCR. In conclusion, despite the disadvantages of classical bacteriological methods, the best approach for direct diagnosis of B. abortus in organs of infected cows includes the isolation associated with PCR of DNA extracted from the cotyledon by the Santos or PK methods.