Accumulation of rifampicin by Mycobacterium aurum, Mycobacterium smegmatisand Mycobacterium tuberculosis

The characteristics of the accumulation of 2 mg/L [C-14]rifampicin by wild- type strains of Mycobacterium aurum (A(+)), Mycobacterium smegmatis (mc(2)1 55) and Mycobacterium tuberculosis (H37Rv) were determined. After 10 min ex posure M. aurum had accumulated 220 ng rifampicin/mg cells, M. smegmatis ha d accumulated 120 ng rifampicin/mg cells and M. tuberculosis had accumulate d 154 ng rifampicin/mg cells. A steady-state concentration (SSC) of rifampi cin was accumulated rapidly by M. aurum and M. tuberculosis within minutes of drug exposure, unlike M. smegmatis, which accumulated rifampicin more sl owly. With an increase in the concentration of rifampicin from 0.12 mg/L to 2 mg/L there was an increase in the concentration of rifampicin accumulate d by M. tuberculosis, with no detectable loss of viability over the 20 min of the accumulation experiment. With an increase in temperature there was a lso an increase in the concentration of rifampicin accumulated by M. tuberc ulosis; between 15 and 30 degrees C the increase was linear. For all three species sub-inhibitory concentrations of ethambutol increased the concentra tion of rifampicin accumulated. However, both growth and accumulation of ri fampicin were lower in the presence of 0.05% Tween 80. Accumulation of rifa mpicin by M. smegmatis was unaffected by the presence of the proton motive force inhibitor, 2,4-dinitrophenol (1 mM), whether added before or after th e addition of rifampicin to the mycobacterial culture. For all three specie s, the Gram-positive bacterial efflux inhibitor reserpine (20 mg/L) slightl y increased the SSC of rifampicin, but the increase was not statistically s ignificant. Addition of glucose to energize a putative efflux pump had litt le effect on the accumulation of rifampicin in the presence or absence of r eserpine for M. tuberculosis; however, for M. aurum and M. smegmatis the re serpine effect was abolished by the addition of glucose. These data suggest that rifampicin may be removed from wild-type mycobacteria by efflux, but that the pump(s) is expressed at low level.