Fig 1.
Genetic and/or environmental factors can lead to antibiotic tolerance in laboratory medium.
(A) Growth of WT and ΔhisG 14028 and WT and hisGP69L SL1344 Salmonella. Ten-fold dilution series were spotted on minimal medium plates in the absence (left) or presence (right) of histidine. (B) Illustration of the in vitro antibiotic survival assay where growers (pink) are killed by cefotaxime and non-growers (red) survive. (C) CFU enumeration of survival of WT and hisGP69L SL1344 Salmonella as well as WT and ΔhisG 14028 Salmonella exposed to cefotaxime treatment in M9GG in absence (left) or presence (right) of histidine. Data from the 24 h timepoint were compared using one way ANOVA with Tukey’s multiple-comparison test, ***p<0.001; NS, not significant. (D) CFU enumeration of survival of WT 14028 exposed to cefotaxime in minimal medium supplemented with carbon sources (Glucose/Glycerol) or not (carbon starvation). (E) CFU enumeration of survival of WT 14028 exposed to cefotaxime in M9GG medium in the presence or absence of chloramphenicol (CAM). (D-E) Statistical significant differences by two-sided t-test between the 24 h timepoint are indicated as **p<0.01; NS, not significant. Panels C-E, data represent the mean and standard deviation (SD) of at least three biological repeats. Of note, WT data are the same on panel D and E as all experiments were conducted in parallel. Auxotrophic strains are depicted in orange, prototrophic in turquoise.
Fig 2.
Growth dynamics at the single cell level discriminates antibiotic tolerance from persistence in cellulo.
(A) Illustration of the in cellulo antibiotic survival assay in primary Bone Marrow Derived Macrophages (BMDM) infected with Salmonella. (B) CFU enumeration of WT or hisGP69L SL1344 Salmonella as well as WT or ΔhisG 14028 Salmonella in BMDMs treated with cefotaxime in the absence or presence of histidine in the infection medium. Data represent the mean and SD of at least three biological repeats. Data from the 48 h timepoint were compared using one way ANOVA with Tukey’s multiple-comparison test, ***p<0.001; NS, not significant. (C) Illustration of the fluorescence dilution assay to assess growth dynamics of a population at the single cell level. Growing cells dilute a pre-formed pool of mCherry at each cell division whereas the non-growing bacteria maintain high intensity fluorescence. (D) Representative flow cytometry profile of Salmonella 14028 grown in minimal medium for 4 h in the absence or presence of histidine for WT and ΔhisG strains. (E) Representative flow cytometry contour (left) and histogram (right) plots of bacteria extracted from BMDM after 16 h of gentamicin (- antibiotic condition) or 16 h of cefotaxime (+ antibiotic condition). G, growers and NG, non-growers.
Fig 3.
Tolerance can mask persister determinants in laboratory medium.
(A-B) Survival of Salmonella 14028 exposed to cefotaxime in M9GG in the absence or presence of histidine for WT, ΔhisG, shpAB1 and shpAB1ΔhisG or WT, ΔhisG, ΔrecA and ΔrecAΔhisG. (C-D) Representative flow cytometry profile (C) and quantification of the non-grower fraction (D) of Salmonella 14028 grown in minimal medium in presence or absence of histidine for shpAB1, ΔrecA, shpAB1ΔhisG and ΔrecAΔhisG. (A-D) Data represent the mean and SD of at least three biological repeats. Data from the 24 h timepoint were compared using one way ANOVA with Tukey’s multiple-comparison test, *p<0.05, ***p<0.001; NS, not significant. Of note, WT and ΔhisG data are the same on panel A and B as all experiments were conducted in parallel. Auxotrophic strains are depicted in orange, prototrophic in turquoise.
Fig 4.
Tolerance can mask persister determinants during macrophage infection.
(A-B) CFU enumeration of survival of the WT, ΔhisG, shpAB1, shpAB1ΔhisG, ΔrecA and ΔrecAΔhisG Salmonella strains in BMDMs treated with cefotaxime and in absence of histidine. (C) Representative flow cytometry contour (left) and histogram plots (right) of the shpAB1, shpAB1ΔhisG, ΔrecA and ΔrecAΔhisG strains extracted from infected BMDM after 16 h of gentamicin. G, growers and NG, non-growers. (D) Quantification of the non-grower fraction for each strain represented in panel C. Data represent the mean and SD of three biological repeats. Data were compared using one way ANOVA with Tukey’s multiple-comparison test, ***p<0.001; NS, not significant. Of note, WT and ΔhisG data are the same on panel A and B as all experiments were conducted in parallel. Auxotrophic strains are depicted in orange, prototrophic in turquoise.
Fig 5.
Persistence is a more balanced strategy than tolerance.
(A) Illustration of the in cellulo competition assay where BMDMs were co-infected with the WT/shpAB1 or WT/ΔhisG pair and treated with cefotaxime to assess antibiotic survival (+AB) or gentamicin to assess proliferation (-AB). WT expressed GFP constitutively whereas the shpAB1 and ΔhisG mutants expressed mCherry. (B) Left panel, representative images of the bacterial population after extraction of 24 h gentamicin (-AB) or cefotaxime (+AB) treated BMDM. For the proliferation competitive assay (-AB), ten-fold dilution series were spotted on rich medium plate. For the antibiotic survival competitive assay (+AB), the entire bacterial population was plated on rich medium. Right panel, competitive index ratios. Data represent the mean and SD of at least three biological repeats. Statistically significant differences of each pair by two-sided t-test are indicated as ***p<0.001. (C) Illustration of the in vivo mice co-infection assay with the WT/shpAB1 or WT/ΔhisG pair used on panel A and B. Two days after oral gavage, mice were either sacrificed or treated with cefotaxime for 4 or 6 days. (D-E) Bacterial survival in the spleen of the WT/ ΔhisG (D) and WT /shpAB1 (E) mix. Results are expressed in percentage of the total population recovered on plate. Day 0 indicates the inoculum ratio. Statistical significant differences at each timepoint by two-sided t-test are indicated as *p<0.05 and ***p<0.001. In vivo mice experiments were carried out with at least 4 animals per time point.
Fig 6.
Trade-off between proliferation and survival in antibiotic tolerance and persistence.
In the case of antibiotic persistence, the majority of cells grow in the population (light blue), allowing bacteria to efficiently colonize their environment. Persisters (turquoise) represent a small reservoir of recalcitrant cells that outlive their growing counterpart under antibiotic treatment. The frequency of persisters in the population can be increased in some conditions (e.g. during macrophage infection) or in specific genetic backgrounds (shpAB1), at the detriment of the pool of growing bacteria. On the other hand, antibiotic tolerance (orange) strongly reduces the proliferation of the entire population which, in turn, limits the efficacy of the antibiotic treatment on all cells.