Fig 1.
Phage-Antibiotic Synergy (PAS) of phage HK620 and filamentation-inducing antibiotics.
(a) HK620 lysis plaques from a double agar overlay assay on a lawn of E. coli TD2158 PL4, with and without filamentation-inducing antibiotics such as cephalexin and ciprofloxacin at ½ MIC. (b) Lysis plaque radii comparison between treatments applied in a. P values for a two-tailed test of less than 0.001 are denoted with three asterisks, less than 0.01 with two asterisks. (c) PAS in liquid culture. The filamentation-inducing antibiotics (ceftazidime, cephalexin and ciprofloxacin), as well as kanamycin, that does not induce filamentation, were present at inoculation time. The vertical dashed line represents the time of phage HK620 addition at a concentration of 106 PFU/mL. Error bars indicate standard error of mean (S.E.M.), N = 4 biological replicates per conditions.
Fig 2.
Synergistic profile of phage T4, T5 and T7 on E. coli MG1655 in the presence of various antibiotics.
(a,b,c) Lysis of a liquid culture of E. coli MG1655 by phage T4 (a), T5 (b), and T7 (c) at 37°C under agitation 180 rpm. The dotted vertical line represents the time of phage addition. Error bars represent the S.E.M.
Fig 3.
Distribution of HK620 adsorption on E. coli TD2158 PL4 cells.
(a) Epifluorescence microscopy of GFP-coated HK620 on the surface of untreated, ciprofloxacin, and cephalexin-treated E. coli cells. (b) Comparison of phage adsorption distribution between untreated, ciprofloxacin, and cephalexin-treated cultures; the dashed lanes and percentages indicate the fraction of cells with one or more phages per cell. (c) Comparison of frequencies of adsorption between regular-size and filamentous subpopulations within the ciprofloxacin and cephalexin-treated cultures. Bacterial cells were segmented and fluorescent-phage foci were detected and assigned to the corresponding cell using MicrobeJ, an ImageJ plug-in. Percentages represent the fraction of cells that have adsorbed at least one phage on their surface. (d) Mean number of phage adsorption for untreated, ciprofloxacin, and cephalexin-treated cultures. (e) Mean number of phages adsorbed to regular-size or filamentous subpopulations within the ciprofloxacin and cephalexin-treated cultures. (f) Mean number of phage adsorption per unit of surface for regular-size and filamentous subpopulations within the ciprofloxacin and cephalexin-treated cultures. Error bars indicate standard error of mean (S.E.M.). P values less than 0.001 for a two tailed test are summarised with three asterisks, non-significant differences are marked as “ns”.
Fig 4.
Distribution of phage T5 adsorption on E. coli MG1655.
(a) GFP channel of an epifluorescence microscopy image showing the appearance of fluorescent T5 phages (coated with Pb10-GFP fusions) adsorbing to E. coli MG1655. (b) Comparison of phage adsorption distribution between untreated, ciprofloxacin, and ceftazidime-treated cultures; percentages to the right of the dashed lanes indicate the fraction of cells with one or more phages per cell. (c) Comparison of frequencies of adsorption between regular-size and filamentous subpopulations within the ciprofloxacin and ceftazidime-treated cultures. Percentages represent the fraction of cells that have adsorbed at least one phage on their surface. (d) Mean number of phage adsorption for untreated, ciprofloxacin, and ceftazidime-treated cultures. (e) Mean number of phages adsorbed to regular-size or filamentous subpopulations within the ciprofloxacin and ceftazidime-treated cultures. (f) Mean number of phage adsorption per unit of surface for regular-size and filamentous subpopulations within the ciprofloxacin and ceftazidime-treated cultures. Error bars indicate standard error of mean (S.E.M.). P values less than 0.001 for a two tailed test are summarised with three asterisks, non-significant differences are marked as “ns”.
Fig 5.
Infection rates of regular-size and filamentous subpopulations within antibiotic-treated cultures.
(a) Phage-encoded GFP reporter allows to discriminate between naïve and infected cells at time = 20 minutes. Images show HK620 infection on E. coli TD2158 PL4 at MOI = 1. Antibiotics are present at ½ of the MIC. (b) Infection rate of each subpopulation in ciprofloxacin-treated (½ MIC) cultures. (c) Infection rate of each subpopulation in cephalexin-treated (½ MIC) cultures. Error bars indicate standard error of mean (S.E.M), N = 3 biological replicates per treatment.
Fig 6.
Tracking of phage T7::parS DNA in ciprofloxacin-treated E. coli MG1655.
(a) Merged phase contrast/epifluorescence microscopy images showing phage T7 DNA replication foci in ciprofloxacin-treated cells over 30 minutes under incubation at 30° C in minimal media. Fluorescence is due to the binding of a host encoded fusion protein ParB-GFP to the sequence parS present in T7. (b) Quantification of replication foci detected in ciprofloxacin treated samples at 10 min and 30 min post-infection. The detected foci converge in larger viral factories whose number positively correlates with cell length.
Fig 7.
One-Step Growth Curves of phage HK620 on E. coli TD2158 PL4 at 37°C.
HK620 displays a latent period of 40 min and a burst-size of approximately 400 PFU per infected TD2158 PL4 cell. Ciprofloxacin and cephalexin did not significantly modify the latent period, but produced an increased burst size of 28% and 36%, respectively.
Fig 8.
Cytograms of E. coli populations with or without phage and antibiotic treatment.
(a,b,c,d,e) E. coli TD2158 infected by phage HK620 in the presence or absence of cephalexin or ciprofloxacin. (a) After 2 h of exponential growth, and before antibiotic treatment. (b,c) After 2 h of exponential growth in the presence of cephalexin or ciprofloxacin at half of the MIC concentrations. (d,e) After one of round of HK620 lysis at MOI = 0.5. Percentages in red represents the population within the gating (red ellipse). (f,g,h) E. coli MG1655 infected by phage T7 in the presence or absence of ciprofloxacin. (f) Cytogram of an E. coli MG1655 culture after 2 hours of exponential growth. (g) Same bacterial strain in the presence of Ciprofloxacin at half the MIC. (h) Ciprofloxacin-treated E. coli MG1655 after 25 min post-infection with phage T7 (approximately one round of lysis), at MOI ~ 1. Percentages in red represents the population within the gating (red ellipse).
Fig 9.
A mathematical model to understand the synergy between antibiotics and phages.
(a) Schematic representation of the interactions between bacteria and phages in the presence of filamentation-inducing antibiotics (S1 Text, S1 Table). (b) Dynamics of the biomass of bacteria in the presence of sublethal doses of antibiotics and/or in the presence of phages. (c) Dynamics of the frequency of mutant bacteria in the presence of sublethal doses of antibiotics and/or in the presence of phages. The bottom black dashed line indicates the dynamics of the frequency of mutant bacteria in the absence of antibiotics. Note that we obtain the same dynamics if phages are present because phage only affect this dynamic if some filamenting bacteria are present.
Fig 10.
Phage HK620 effect on bacterial mutagenesis and SOS activation in regular-size cells and filaments.
(a) Fold-increase in mutagenesis rate following ciprofloxacin treatment, and its reduction after the addition of HK620 30 PFU/well at 11 hours post-inoculation. P values of less than 0.05 for a one-tailed test are summarised with a single asterisk. (b) The pcda’-gfp fusion expression was monitored at increasing concentrations of ciprofloxacin in filamentous as well as in non-filamentous subpopulations. The figure shows the percentage of each subpopulation above a threshold defined by the levels of fluorescence found in a ciprofloxacin-free bacterial population. (c) Fluorescence fold-change in cells carrying the pcda’-mcherry reporter versus cell length treated with ciprofloxacin (½ MIC). Percentages in the figure represent the abundance of each subpopulation in the culture. As expected from (b), a strong correlation between cell-length and SOS activation is observed. The dashed vertical line represents the size threshold for filamentation (5.2 μm). Percentages of HK620 hkcEF::PrrnB-gfp infection on each subpopulation (at MOI = 1) were of 41% and 82% for regular-sized cells and filaments, respectively. The filamentous, hypermutagenic subpopulation was infected more frequently than cells retaining a regular shape that had, contrarily, lower SOS activation rates.
Fig 11.
A graphical model for ciprofloxacin-enhanced killing by phage HK620, which prevents bacterial mutagenesis.