Predation of Chromobacterium piscinae by Bdellovibrio bacteriovorus HD100 was inhibited in dilute nutrient broth (DNB) but not in HEPES. Experiments showed that the effector responsible was present in the medium, as cell-free supernatants retained the ability to inhibit predation, and that the effector was not toxic to B. bacteriovorus. Violacein, a bisindole secondary metabolite produced by C. piscinae, was not responsible. Further characterization of C. piscinae found that this species produces sufficient concentrations of cyanide (202 µM) when grown in DNB to inhibit the predatory activity of B. bacteriovorus, but that in HEPES, the cyanide concentrations were negligible (19 µM). The antagonistic role of cyanide was further confirmed, as the addition of hydroxocobalamin, which chelates cyanide, allowed predation to proceed. The activity of cyanide against B. bacteriovorus was found to be twofold, depending on the life cycle stage of this predator. For the attack-phase predatory cells, cyanide caused the cells to lose motility and tumble, while for intraperiplasmic predators, development and lysis of the prey cell were halted. These findings suggest that cyanogenesis in nature may be employed by the bacterial strains that produce this compound to prevent and reduce their predation by B. bacteriovorus.

Summary Bdellovibrio bacteriovorus HD100 is a highly motile predatory bacterium that consumes other Gram‐negative bacteria for its sustenance. Here, we describe the impacts the media viscosity has both on the motility of predator and its attack rates. Experiments performed in polyethylene glycol (PEG) solutions, a linear polymer, found a viscosity of 10 mPa s (5% PEG) negatively impacted predation over a 24‐h period. When the viscosity was increased to 27 mPa s (10% PEG), predation was nearly abolished. Tests with three other B. bacteriovorus strains, i.e., 109J and two natural isolates, found identical results. Short‐term (2‐h) experiments, however, found attack rates were improved in 1% PEG, which had a viscosity of 5.4 mPa s, using bioluminescent prey and their viabilities. In contrast, when experiments were performed in dextran, a branched polymer, no increase in predation was seen even though the viscosity was a comparable 5.1 mPa s. The enhanced attack rates in this solution coincided with a 31% increase in B. bacteriovorus HD100 swimming speeds (62 μm s−1 in 1% PEG vs. 47.5 μm s−1 in HEPES‐salt).

No abstract