Methods
Bacterial strains, phage, media and culture conditions. Experiments were performed with two different strains of Pseudomonas aeruginosa: strain PAO1 carrying a GFP-expressing plasmid, which was susceptible to a specific phage, and strains PA14 (PA14-WT) or carrying an mCherry-expressing plasmid (PA14-mCherry), which were both resistant to this same phage. Both plasmids contained a gentamicin resistance gene. All three strains were kindly provided by Kevin Foster. The phage used for this study was chosen based on a spot assay. This assay is used to investigate which phage produces phage plaques when they are in contact with our bacterial strains. Among 14 phages tested, we chose one that induces the lysis of PAO1-GFP but not the lysis of both PA14 strains, which were entirely resistant. To count the colony-forming units (CFU) of bacterial monocultures and plaque-forming units (PFU) of phages, Tryptic Soy Agar (TSA; Trypticase TM Soy Agar, Difco) plates were used. To distinguish both strains within a community, co-cultures were plated onto TSA plates to count PA14-WT CFUs and onto LB agar plates containing 10 μg/ml of gentamicin to count only PAO1-GFP CFU. Overnight cultures were grown in tryptic soy broth (TSB; Bacto TM, Detroit, MI, USA) at 37°C, shaken at 200 rpm. Before each experiment, the optical density (OD600) of the overnight cultures of PAO1-GFP, and either PA14-mCherry or PA14-WT strains (depending on the experiment) was measured with a spectrophotometer (Ultrospec 10, Amersham Biosciences). Bacterial overnight cultures were then inoculated into Erlenmeyer flasks (100 ml) containing 20 ml of TSB to obtain a standardized OD600 of 0.05. Bacterial cultures were grown in a shaking incubator at 200 rpm and 37°C for 3 hours to obtain bacteria in exponential phase with a final concentration of approximately 108 CFU/ml at the beginning of each experiment. These cultures were then diluted to the desired starting population size.
Phage treatment in liquid cultures. A 96-well plate was filled with 200 μl of TSB in each well, additionally containing 106 CFU/ml PAO1-GFP or 108 CFU/ml PA14-WT alone, or together with or without 106 PFU/ml of phages (MOIPAO1 = 1). Each condition (PAO1-GFP alone, PA14-WT alone, and the co-culture) was performed in triplicate. These initial population sizes were chosen since they allowed the two strains to co-exist and grow over 48 hours. The plate was then put in a Tecan i-control infinite M200 PRO plate reader at 37°C under agitation for 48 hours. After 6, 24 and 48 hours, the untreated samples were transferred into Eppendorf tubes and centrifuged at 4°C, 8000 rpm for 15 minutes, resuspended in 200 μl of PBS, serially diluted and plated on either TSA or LB Agar gentamicin 10 μg/ml plates. For the treated samples, the same protocol was followed except that after centrifugation, the supernatants containing phages were kept in the fridge at 4°C. They were further diluted in NaCl 0.9% to count the PFU/ml. The pellets containing the bacteria were then washed 3 times with 1 ml of fresh PBS at 8000 rpm, 4°C for 5 minutes to remove all the potential phages remaining in the pellet. Finally, the phage resistance rate of PAO1-GFP was calculated after 24 and 48 hours by plating them on TSA plates containing approximately 1010 PFU (plaque forming units) of pre-absorbed phages. If PAO1-GFP were growing in co-culture with PA14-WT, plates additionally containing 10 μg/ml gentamicin were used. To evaluate resistance rates, the CFU/ml of PAO1-GFP growing on plates saturated with phages was then compared to the CFU/ml growing on plates with no phage.
Quantifying phage resistance in liquid mono- and co-cultures. To understand the role of population size on resistance emergence, two experiments were performed (Fig. 1E, F). In the first, a 96-well plate was filled with 10 up to 108 PAO1-GFP, with 10-fold increases, together with phage to achieve an MOIPAO1 = 1 in 200 μl of TSB. We grew the bacteria for 21 hours at 37°C under agitation in the plate reader, and then assessed phage resistance rates and total population sizes. For the second experiment, a 96-well plate was filled with 106 CFU/ml of PAO1-GFP and phages at an MOIPAO1 = 1 in 200 μl of TSB, to which we added increasing amounts of PA14, starting at 102 up to 108 in 10-fold increments. Bacteria were again grown in the plate reader for 21 hours at 37°C under agitation, at the end of which we assessed phage resistance rates and total population sizes of both strains.
Phage treatment in biofilms. To grow bacteria in a biofilm, liquid cultures were prepared and a drop spotted onto a membrane filter (Isopore®Membran, GTTP04700, Merck) previously placed in the centre of agar plates containing 0.1x LB (1 g/L of tryptone (ThermoScientific TM Oxoid TM Tryptone), 0.5 g/L of yeast extract (ThermoScientific TM Oxoid TM Yeast Extract Powder), 10 g/L of NaCl (ACROS Organics TM , 99.5%) and 15 g/L of agar (Bacto TM Agar solidifying agent, BD Diagnostics). Liquid cultures of the two strains were prepared as described for the liquid experiments in PBS to obtain a final concentration of 106 CFU/ml of PAO1-GFP and 108 CFU/ml of PA14-WT in a drop of 2μl that was spotted onto the filter. Nine replicate plates were prepared for each condition (PAO1-GFP, PA14-WT and the mixture of both), and incubated at 37°C. After 12 hours of incubation, three replicates were removed in order to count the CFUs of both strains by removing the filters from the agar using sterile tweezers and placing them in tubes containing 3 ml of PBS. The tubes were well shaken to remove and resuspend the biofilms in the PBS, the filters removed and the bacteria plated on either TSA plates (monoculture, co-culture) or on gentamicin-containing (co-culture). Among the six remaining replicates, three were placed onto new 0.1x LB agar plates without phages and the three others were placed onto new plates pre-absorbed with a 50μl drop containing approximately 1010 phages (diameter similar to filter diameter) and incubated at 37°C. After 36 hours, the filters containing biofilms growing on agar were removed and placed in 3 ml of PBS to count the CFUs and PFUs as described above. Furthermore, for the treatments involving phage, the whole agar was also collected and put in 50 ml falcon tubes containing 10 ml of SM buffer, well-shaken, centrifuged for 20 minutes at 8000 rpm at 4°C and the supernatant containing phages further diluted in NaCl 0.9% to count the PFU/ml. Both phage concentrations (filter + agar) were added to have the final PFU/ml value.
Image analysis. Images of the different biofilms were acquired after 12 and 48 hours using a fluorescence microscope.
Statistical analysis. Each experiment was performed using three biological replicates per condition.