Molecular screening for rifampicin and fluoroquinolone resistance in a clinical population of Brucella melitensis.

OBJECTIVES: The aim of this study was to determine, using molecular methods, whether rifampicin and fluoroquinolone resistance was present in a clinical Brucella melitensis population.
METHODS: Sixty-two B. melitensis strains, isolated from humans-most experiencing their first brucellosis episode-over an 11 year period in Spain, were genotyped by multiple locus variable analysis (MLVA-16) for future studies. In the present work, molecular screening was undertaken to detect the presence of rpoB and gyrA/gyrB/parC/parE mutations (previously described in in vitro Brucella spp. mutants) related to resistance to rifampicin and fluoroquinolones, respectively.
RESULTS: Sixty-two MLVA-16 genotypes were identified among the B. melitensis population, with genetic similarity values ranging from 32% to 94%. rpoB mutations related to rifampicin resistance (positions 154, 526, 536, 539, 541, 574) were not detected. Neither were changes in GyrA described in in vitro mutants (67, 71, 87, 91 and an insertion at 340) detected in these strains. All showed identical GyrA, GyrB, ParC and ParE sequences with respect to B. melitensis 16M, except for one strain (ciprofloxacin and moxifloxacin MICs 0.25-0.50 mg/L) that harboured the Val264Ala replacement outside the GyrA quinolone resistance-determining region (QRDR); no differences were seen, however, in the NorMI/II efflux pump genes.
CONCLUSIONS: The absence of rpoB mutations clearly related to rifampicin resistance in clinical B. melitensis strains reinforces the first-choice status of this antibiotic in the treatment of first brucellosis episodes, and demonstrates the usefulness of molecular screening for resistant genotypes. The absence of topoisomerase II-IV mutations, however, cannot rule out fluoroquinolone resistance due to the interplay of different mechanisms.