Tuberculosis due to Mycobacterium bovis in the Australian population: DNA typing of isolates, 1970-1994.

SETTING: Bacteriologically confirmed cases of Mycobacterium bovis in the Australian population.
OBJECTIVE: To evaluate the DNA fingerprinting techniques commonly used for M. bovis on isolates from humans and determine whether they were useful for determining the origin of human infection.
DESIGN: M. bovis strains isolated between 1970 and 1994 were obtained from five Australian Reference Laboratories. Four DNA fingerprinting techniques, comprising Southern hybridisation with three different probes (the insertion sequence [IS]6110, the polymorphic guanine-cytosine-rich sequence [PGRS] and the direct repeat [DR]) and a PCR-based method (spoligotyping) were used.
RESULTS: The PGRS, DR and IS6110 RFLP methods identified 32, 22 and 14 different types respectively from the 45 isolates available. Spoligotyping identified 18 different types. When all methods were combined 41 different strains were identified. Clear differences were found between many isolates from Australian-born patients and those from patients born overseas.
CONCLUSIONS: The PGRS RFLP method was the most effective method for typing the human strains, but a combination of methods is recommended for maximum sensitivity. Most Australian-born patients that had worked in the meat and livestock industries were infected with strains similar to those that are commonly found in Australian cattle, confirming the occupational risk in these industries. Patients born overseas were typically infected with strains genetically different from those of patients born in Australia. This suggests that patients born overseas identified with M. bovis were presenting with reactivation of infection acquired outside Australia.