A Bacteroides fragilis strain produces a low-molecular-weight (13,500 to 18,700), protemaceous bacteriocin during the stationary growth phase. The extracellular bacteriocin is not inducible by ultraviolet light or mitomycin C and is stable between pH 7.5 and 8.2. The majority of the bacteriocin is thermolabile, but a small proportion (3%) of the bacteriocin is stable after autoclaving at 121°C for 15 min. Killing of sensitive bacteroides cells follows single-hit kinetics, and the interaction of a single molecule of bacteriocin with a target cell occurs in two stages. The killing of susceptible cells is affected by temperature and the growth state of the susceptible cells. The bacteriocin is unusual in that the primary event in its mode of action is the inhibition of RNA synthesis. The bacteriocin inhibits RNA synthesis immediately but has no effect on DNA synthesis or intracellular ATP levels. Protein synthesis is inhibited after a delay of 20 min, presumably as a result of the initial inhibition of RNA synthesis.Since the discovery of bacteriocins by Gratia (10), much work has been done on the bacteriocins produced by aerobic bacteria (11,19,22,23). The colicins produced by Escherichia coli have been particularly well studied, and their modes of action have been identified (11,19,23). However, little work has been carried out on the bacteriocins produced by Bacteroides fragilis, which is the most frequently isolated anaerobe species from clinical specimens (13, 18). Booth et al. (3) investigated bacteriocin production by Bacteroides strains and the role of these strains in the ecology of the colon. Although Booth et al. (3) characterized and partially purified one of the bacteriocins, the mechanism of killing was not determined. We describe the characterization, purification, and unusual mode of action of a B. fragilis bacteriocin.
MATERIALS AND METHODSMedia and anaerobic techniques. Brain heart infusion (BHI) broth and agar (17) were used for bacterial growth and bacteriocin production. The glucose minimal medium was similar to that described by Varel and Bryant (28), except that 0.1 M phosphate buffer was used instead of 0.4% Na2CO3. The anaerobic glove box and techniques described by Moodie and Woods (17) (AU) were expressed as the reciprocal of the highest doubling dilution that gave a zone of inhibition surrounding the well.Bacteriocin production. The producer strain was inoculated into BHI broth, and samples of the culture supernatant were assayed for bacteriocin at different time intervals.Induction experiments. Mitomycin C (Calbiochem, La Jolla, Calif.) and ultraviolet irradiation were used to test for induction of the bacteriocin. Exponential-phase cells were added to BHI media containing 0.1, 0.2, 0.3, 0.4, 0.8 and 1.0,g of mitomycin C per ml and incubated for 6 h. Cultures were either assayed for bacteriocin directly or harvested by centrifugation, washed, and suspended in BHI broth for a further 6 h before assaying. Induction by ultraviolet irradiation was carried out on exponential-phase cel...