Background

Bdellovibrio bacteriovorus is a Gram-negative predatory bacterium that "predates" upon other Gram-negative bacteria species, growing at their expense and eventually killing them. B. bacteriovorus as well as other predatory bacteria have the potential to be used as probiotics for disinfection of surfaces or for the treatment of bacterial infections (Tyson and Elizabeth Sockett, 2017; Madhusoodanan et al., 2019). The combined use of B. bacteriovorus with antimicrobials may enhance the effectiveness of these treatments. This necessarily entails B. bacteriovorus to resist the action of antimicrobials at their working concentration. Little is known on the antimicrobial susceptibility of B. bacteriovorus. However, the sequenced genomic DNA encodes membrane drug transporters (efflux pumps) and other antimicrobial resistance proteins (Rendulic et al., 2004). The sensitivity of B. bacteriovorus to antimicrobials cannot be assessed by a standard broth dilution assay, as it requires the presence of prey bacteria to grow.

In a 2014 study, Pasternak et al. made a first attempt to determine the antimicrobial sensitivity of predatory bacteria (Micavibrio aeruginosavorus EPB and Bdellovibrio exovorus JSS) by developing a modified antimicrobial disc diffusion method. However, this method provides a relative measure of antimicrobial resistance levels for the species analyzed, but does not allow to compare these values with those of non-predatory bacteria species as it does not use the same conditions of standardized assays (predatory bacteria require the presence of prey bacteria to grow).

In a study by Im et al. (2017), B. bacteriovorus was used in combination with violacein (an antimicrobial active against Gram-positive bacteria) to treat mixed populations of multi-drug resistant (MDR) pathogens. The aim of the study was to demonstrate that this combined treatment was more effective than treatments with individual agents (B. bacteriovorus or violacein). The efficacy of the treatments was evaluated in terms of reduction in the titer of MDR bacterial cells and, therefore, did not provide an indication of the degree of general antimicrobial resistance in B. bacteriovorus.

The method presented in this study allows to determine the MIC values of antimicrobials for B. bacteriovorus. Standard broth dilution assays (performed on non-predatory bacteria) are performed in liquid broth containing antimicrobial dilution series (typically two-fold dilutions) and are inoculated with a defined number of bacterial cells (5 x 10⁵ colony-forming units (CFU)/ml). In order to test antimicrobials on B. bacteriovorus, the initial numbers of both B. bacteriovorus and prey bacterial cells have to be considered (initial predator to prey ratio, PPR). The numbers of B. bacteriovorus cells can be estimated reliably only after conducting the test (by plating in double layer agar plates), as their viability is strongly affected by culture conditions. Higher initial PPRs (comparable to a higher inoculum in standard broth dilution assays) can lead to an increased estimate of the antimicrobial minimal inhibitory concentration (MIC; i.e., the lowest concentration (in mg/L) of the antimicrobial agent that prevents visible growth of a microorganism under defined conditions) (Wiegand et al., 2007). The presented method requires the determination of different MIC values of a single antimicrobial at different initial PPR values. For this, serial two-fold dilutions of B. bacteriovorus cells and a fixed number of prey bacteria are added to a master plate with serial two-fold antimicrobial dilutions. In order to quantify the degree of resistance to a specific antimicrobial, the MICs determined at different initial PPRs must be reduced to a single value. This MIC value is obtained by curve fitting the MIC values at each PPR for each antimicrobial and definition of the MIC at the arbitrary initial PPR of 0.02 (which constitutes a low ratio that allows B. bacteriovorus to grow well and lyse almost all prey cells within 24 h under the experimental conditions). In order to compare the MIC values of B. bacteriovorus with that of other bacterial species, such as the prey bacteria, it is necessary to determine the MIC values under conditions resembling experimental conditions employed for B. bacteriovorus. In this way, the ratio between predatory and prey bacteria MIC values provides a comparative measure of the sensitivity to the antimicrobials tested on both species.

This method may not allow the determination of absolute B. bacteriovorus MIC values for bactericidal antimicrobials at concentrations equal or higher than the minimal bactericidal concentration (MBC) for the prey bacteria (Marine et al., 2020). This is because B. bacteriovorus may not be able to grow on inactivated prey bacteria at concentrations equal or above their MBC, since this would result in a lower estimate of the MIC (this is true especially for lysed prey bacteria or those with a compromised structural integrity) (Marine et al., 2020). To overcome this limitation, prey bacteria may be transformed with resistance genes for the specific antimicrobial tested and used in the B. bacteriovorus MIC assay. However, these resistance genes should not encode for a resistance mechanism that inactivates the antimicrobial. One possibility might be the use of resistance genes encoding multidrug resistance pumps (e.g., AcrAB-TolC, MacAB-TolC, MdfA, etc.) that efflux antimicrobials out of the cell.

The protocol presented below may be adapted to test other predatory bacteria, or use different experimental conditions (i.e., culture broth, temperature, aeration, prey bacteria species), or can be used on different antimicrobials. The only limitation to our knowledge is the use of predatory bacteria that grow well within 24 h.