Background

Bacteria have evolved several different strategies to avoid or resist protozoan predation (Rønn et al., 2012). Mechanisms that reduce chance of ingestion such as increased cell size, aggregation, biofilm formation and increased swimming speed are well documented (Matz and Kjelleberg, 2005; Rønn et al., 2012) whereas mechanisms that allow bacteria to resist digestion in protozoan food vacuoles after ingestion are less studied (Gong et al., 2016). Here we describe a method to investigate the ability of bacterial strains to survive after ingestion by the social amoeba Dictyostelium discoideum. We used this assay to investigate if copper resistant E. coli have higher chance of survival after ingestion than bacteria without copper resistance (Hao et al., 2016). We studied copper resistance because it is known that macrophages, which are phagocytotic cells with an important role in the immune system of vertebrates, use copper to kill bacteria in the phagosome (Hodgkinson and Petris, 2012). We hypothesized that this killing mechanism originally evolved in free-living protozoa long before multicellular life arose and hence copper resistance could be a factor that protects bacteria against protozoan predation.

The assay is a modification of the Gentamicin protection assay used for evaluation of the ability of macrophages to kill bacteria (see e.g., Kaneko et al., 2016) and it utilizes that gentamicin is not able to penetrate into eukaryotic cells. In the original assay the number of internalized bacterial cells in the macrophages is determined at two time points. First, macrophages are incubated with bacteria for a given time period and after this co-incubation the uningested bacteria are removed and the number of internalized bacteria is estimated. Second measurement is taken after macrophages are incubated without extracellular bacteria to allow them to digest the internalized bacteria. The ratio between these two estimates is a measure of the bacterial survival rate.

Rapid digestion of some of our strains led to reduced reliability of estimates of the disappearance rate and made the original procedure unsuitable for use with Dictyostelium. We therefore modified the procedure so that we only estimated the number of surviving internalized bacteria at one time point after co-incubation of bacteria and amoebae for 6 h. This estimate reflects an equilibrium between ingestion and digestion rate of bacteria. It should be noted that this will only allow comparison of the digestion rate of different bacterial strains if it can be assumed that the bacteria are consumed with similar rates. In our case we had evidence that the different E. coli strains were taken up by the amoebae to the same extent and therefore we used the assay as an indicator of digestion rate. The assay will still be useful even if it cannot be assumed that bacteria are ingested at the same rate but of course it should be interpreted accordingly. It must also be noted that the procedure can only be used with bacteria that are sensitive to gentamicin.