Figure 1. Phenotypic heterogeneity in the accumulation of the major classes of antibiotics.

(A) Illustration depicting the eight antibiotics employed in this study alongside their bacterial targets. (B) Accumulation of the fluorescent derivative of roxithromycin in 265 individual E. coli (continuous lines) after adding the probe at 46 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Fluorescence values were background subtracted and normalised first by cell size and then to the maximum value in the dataset (see Methods). The circles and shaded areas represent the mean and SD of the values from 265 bacteria collated from biological triplicate. The squares represent the fluorescent values of a representative bacterium that does not accumulate the fluorescent derivative of roxithromycin, whereas the triangles represent the fluorescent values of a representative bacterium that accumulates the drug. Insets: representative brightfield and fluorescence images after 7000 s incubation in the fluorescent derivative of roxithromycin, the symbols indicate the two representative bacteria above. Scale bar: 5 µm. (C) Population average (symbols) and SD (shaded areas) of the accumulation of the fluorescent derivatives of polymyxin B (triangles), trimethoprim (stars), roxithromycin (circles), and vancomycin (squares) probes added at 46 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Data are obtained by averaging at least one hundred single-cell values (i.e. N=103, 175, 265, and 236, respectively) collated from biological triplicate. Corresponding single-cell data along with data for the fluorescent derivatives of linezolid, tachyplesin, octapeptin, and ciprofloxacin probes are reported in Figure 1—figure supplement 2.

Figure 1—figure supplement 1. Measurement of single-cell elongation rates.

Average elongation rate during growth in lysogeny broth (LB) after injection of E. coli in the mother machine at t=0. The data are the mean and standard error of the mean calculated over measurements performed on 30 individual bacteria collated from biological triplicate.

Figure 1—figure supplement 2. Measurement of single-cell drug accumulation.

Accumulation of the fluorescent derivatives of (A) polymyxin B, (B) octapeptin, (C) tachyplesin, (D) linezolid, (E) ciprofloxacin, (F) trimethoprim, (G) roxithromycin, and (H) vancomycin in 103, 104, 128, 115, 122, 175, 265, and 236 individual E. coli, respectively (continuous lines), after adding each probe at 46 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Data were collated from biological triplicate. Fluorescence values were background subtracted and normalised by cell size (see Methods). The symbols and shaded areas represent the mean and SD of the corresponding single-cell values. Insets: representative fluorescence images showing the accumulation of each probe at the specified time point. Scale bar: 5 µm. The vertical dotted lines represent the time point at which the median of each dataset became larger than zero. The median remained zero throughout the entire experiments carried out with vancomycin-NBD, hence the dotted line has been arbitrarily set at 11,500 s in (H) for comparison purposes only.

Figure 1—figure supplement 3. Measurement of normalised single-cell drug accumulation.

Normalised accumulation of the fluorescent derivatives of (A) polymyxin B, (B) octapeptin, (C) tachyplesin, (D) linezolid, (E) ciprofloxacin, (F) trimethoprim, (G) roxithromycin, and (H) vancomycin. These data are reproduced from Figure 1—figure supplement 2 after normalising all fluorescent values to the maximum fluorescence value in each dataset.

Figure 1—figure supplement 4. Heterogeneity in the accumulation of different antibiotics.

Population averages (symbols) and SDs (shaded areas) of the accumulation of the fluorescent derivatives of tachyplesin (triangles), octapeptin (crosses), linezolid (hexagons), and ciprofloxacin (diamonds) added at 46 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Data were obtained by averaging N=128, 104, 115, and 122 single-cell values, respectively, collated from biological triplicate presented in Figure 1—figure supplement 2.

Figure 1—figure supplement 5. Measurement of single-cell doubling times.

Distribution of single cell doubling times during treatment with roxithromycin dissolved in (A) miminal medium M9 and (B) lysogeny broth (LB) as a drug milieu against the parental and (C) a ΔtolC knockout strain. The fluorescent derivative of roxithromycin was delivered at an extracellular concentration of 46 µg mL^–1 at t=0 to N=40, 38, and 46 individual bacteria, respectively. The average doubling time was (75±28), (29±9), and (80±30) min, respectively. The distributions of doubling times in (A) and (C) were not statistically different (p-value =0.51), whereas the distributions in (A) and (B) were statistically different (****). None of the bacteria reported in (A) and (C) underwent a full cell cycle from birth to division during treatment, since all the bacteria that divided during treatment were born before treatment started. 65% of the bacteria reported in (B) underwent a full cell cycle.

Figure 1—figure supplement 6. Interdependence between cell size and drug accumulation.

Absence of correlation between the area of each single bacterium before antibiotic treatment and the kinetic parameters (A) t₀, (B) k₁ and (C) F_max describing the onset, uptake rate and level of saturation of the fluorescent derivative of roxithromycin in N=104 E. coli after adding the probe at 192 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Data were collated from biological triplicate. We also found no correlation between cell area and the three kinetic accumulation parameters above for the other seven antibiotic probes investigated.

Figure 1—figure supplement 7. Staining of bacteria with different antibiotics.

Temporal dependence of the fraction of E. coli stained by fluorescent derivatives of polymyxin B (downwards triangles), tachyplesin (upwards triangles), octapeptin (crosses), linezolid (hexagons), trimethoprim (stars), ciprofloxacin (diamonds), roxithromycin (circles), or vancomycin (squares). The stained fraction at each time point is defined as the ratio of the number of bacteria displaying a fluorescence distinguishable from the background over the total number of bacteria at that time point. Symbols and error bars are the mean and standard error of the mean values calculated by averaging the N=103, 128, 104, 115, 175, 122, 265, and 236 individual bacteria, respectively, from biological triplicate presented in Figure 1—figure supplement 2.

Figure 1—figure supplement 8. Comparison of single-cell roxithromycin accumulation in E. coli and S. aureus.

Accumulation of the fluorescent derivative of roxithromycin in (A) N=265 individual E. coli and (B) N=195 individual S. aureus (continuous lines), after adding the probe at 46 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Data were collated from biological triplicate. Fluorescence values were background subtracted and normalised by cell size. The symbols and shaded areas are the mean and SD of the corresponding single-cell values. The measured minimum inhibitory concentration (MIC) of roxithromycin-NBD against E. coli and S. aureus was 192 and 1 µg mL^–1, respectively. Insets: representative fluorescence images showing the accumulation of the fluorescent derivative of roxithromycin 3600 s post addition to the bacteria hosting channels. Scale bar: 5 µm. The vertical dotted lines represent the time points at which the median of each dataset became larger than zero.

Figure 1—figure supplement 9. Comparison of single-cell vancomycin accumulation in E. coli and S. aureus.

Accumulation of the fluorescent derivative of vancomycin in (A) N=236 individual E. coli and (B) N=63 individual S. aureus (continuous lines) cells, after adding the probe at 46 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Data were collated from biological triplicate. Fluorescence values were background subtracted and normalised by cell size. The symbols and shaded areas are the mean and SD of the corresponding single-cell values. The measured minimum inhibitory concentration (MIC) of vancomycin-NBD against E. coli and S. aureus was >192 and 0.5 µg mL^–1, respectively. I Insets: representative fluorescence images showing the accumulation of the fluorescent derivative of roxithromycin 1800 s post addition to the bacteria hosting channels. Scale bar: 5 µm. The vertical dotted lines represent the time points at which the median of each dataset became larger than zero.

Figure 1—figure supplement 10. Comparison of single-cell ciprofloxacin accumulation in E. coli, P. aeruginosa, and B. cenocepacia.

Accumulation of the fluorescent derivative of ciprofloxacin in (A) N=122 individual E. coli, (B) N=84 individual P. aeruginosa and (C) N=86 individual B. cenocepacia (continuous lines) cells, after adding the probe at 46 µg mL^–1 extracellular concentration in M9 minimal medium from t=0 onwards. Data were collated from biological triplicate. Fluorescence values were background subtracted and normalised by cell size. The symbols and shaded areas are the mean and SD of the corresponding single-cell values. The measured minimum inhibitory concentration (MIC) of ciprofloxacin-NBD against E. coli, P. aeruginosa, and S. aureus was 8, 32, and 32 µg mL^–1, respectively. The vertical dotted lines represent the time points at which the median of each dataset became larger than zero. As expected ciprofloxacin-NBD accumulated to a significantly lower extent in P. aeruginosa since it lacks general porins, thus displaying a lower permeability compared to E. coli and B. cenocepacia (Rybenkov et al., 2021).

Figure 1—figure supplement 11. Impact of drug milieu and concentration on drug accumulation.

(A) Accumulation of the fluorescent derivative of roxithromycin in lysogeny broth (LB) (circles) or M9 medium (squares) drug milieu delivered to N=46 and 265 individual E. coli, respectively, at an extracellular concentration of 46 µg mL^–1. (B) Accumulation of the fluorescent derivative of roxithromycin delivered at a concentration of 192 (upward triangles) and 46 (squares) µg mL^–1 in a M9 medium drug milieu to N=110 and 265 individual E. coli, respectively. (C) Accumulation of the fluorescent derivative of polymyxin B delivered at a concentration of 46 (diamonds) and 1 (downward triangles) µg mL^–1 in a M9 medium drug milieu to N=101 and 97 individual E. coli, respectively. Data were collated from biological triplicate and fluorescence values were background subtracted and normalised by cell size. The symbols and shaded areas represent the mean and SD of the corresponding single-cell values.

Figure 1—figure supplement 12. Impact of drug labelling on drug accumulation.

(A) Accumulation of unlabelled ciprofloxacin (triangles) and of the fluorescent derivative ciprofloxacin-NBD (diamonds) delivered to N=48 and 122 individual E. coli, respectively, at an extracellular concentration of 200 and 46 µg mL^–1 in M9 medium, respectively. It is worth noting that unlabelled ciprofloxacin was not detectable neither extracellularly nor intracellularly at concentrations below 200 µg mL^–1. (B) Accumulation of the fluorescent derivatives roxithromycin-NBD (squares) and roxithromycin-DMACA (hexagons) at an extracellular concentration of 46 µg mL^–1 in a M9 medium drug milieu delivered to N=265 and 77 individual E. coli, respectively. In both figures data were collated from biological triplicate and fluorescence values were background subtracted and normalised by cell size. The symbols and shaded areas are the mean and SD of the corresponding single-cell values normalised to the maximum mean fluorescence value in each dataset.

Figure 1—figure supplement 13. Main single-cell kinetic parameters inferred using our mathematical model.

Distributions of t₀, k₁ and F_max kinetic parameters describing the accumulation of the fluorescent derivatives of polymyxin B, octapeptin, tachyplesin, linezolid, ciprofloxacin, trimethoprim, and roxithromycin (from top to bottom, respectively). These parameters were inferred by fitting the single-cell data reported in Figure 1—figure supplement 2 using our mathematical model (see Methods). Data for which the fitting algorithm returned divergent transitions were not reported and typically represented less than 1% of the data (compare N here and in Figure 1—figure supplement 2). t₀ is the inferred accumulation onset, i.e., the time at which each bacterium fluorescence became distinguishable from background fluorescence, k₁ is the inferred rate of uptake, F_max is the inferred fluorescence saturation level at steady-state. CV is the coefficient of variation of the single-cell values in each dataset.

Figure 1—figure supplement 14. Second order single-cell kinetic parameters inferred using our mathematical model.

Distributions of k₂, d_r and d_c kinetic parameters describing the accumulation of fluorescent antibiotic derivatives of polymyxin B, octapeptin, tachyplesin, linezolid, ciprofloxacin, trimethoprim, and roxithromycin (from top to bottom, respectively). These parameters were inferred by fitting the single-cell data reported in Figure 1—figure supplement 2 using our mathematical model (see Methods). Data for which the fitting algorithm returned divergent transitions were not reported and typically represented less than 1% of the data (compare N here and in Figure 1—figure supplement 2). k₂ is the inferred adaptive inhibitory rate constant that describes the dip we observed in some single-cell trajectories in Figure 1—figure supplement 2, d_r is the drug loss rate constant, d_c is the dampening rate constant. CV is the coefficient of variation of the single-cell values in each dataset. Membrane targeting antibiotic probes displayed, on average, a higher adaptive inhibitory rate constant (k₂=0.006, 0.007, and 0.006 a.u. s^–2 for tachyplesin, polymyxin B, and octapeptin, respectively) compared to antibiotics with intracellular targets (k₂=0.0001, 0.00005, 0.0003, and 0.0001 s for linezolid, trimethoprim, ciprofloxacin, and roxithromycin, respectively). Remarkably, we found notable cell-to-cell differences in k₂ across all investigated drugs with a maximum CV of 251% for roxithromycin and a minimum CV of 67% for trimethoprim. Membrane targeting antibiotic probes also displayed, on average, a higher drug loss rate constant (d_r =0.09, 0.09, and 0.03 s^–1 for tachyplesin, polymyxin B, and octapeptin, respectively) compared to antibiotics with intracellular targets (d_r =0.0003, 0.001, 0.0005, and 0.001 s for linezolid, trimethoprim, ciprofloxacin, and roxithromycin, respectively). Remarkably, we found notable cell-to-cell differences in d_r across all investigated drugs with a maximum CV of 208% for roxithromycin and a minimum CV of 44% for trimethoprim. Membrane targeting antibiotic probes also displayed, on average, a higher dampening rate constant (d_c =0.009, 0.01, and 0.009 s^–1 for tachyplesin, polymyxin B, and octapeptin, respectively) compared to antibiotics with intracellular targets (d_c =0.0006, 0.0005, 0.002, and 0.0003 s for linezolid, trimethoprim, ciprofloxacin, and roxithromycin, respectively). Remarkably, we found notable cell-to-cell differences in d_c across all investigated drugs with a maximum CV of 187% for tachyplesin and a minimum CV of 28% for linezolid.

Figure 1—figure supplement 15. Coupling between accumulation parameters.

Correlation between (A) t₀ and k₁, (B) t₀ and F_max, (C) k₁ and F_max describing the accumulation of the fluorescent derivatives of polymyxin B (downward triangles), tachyplesin (upward triangles), octapeptin (crosses), linezolid (hexagons), trimethoprim (stars), ciprofloxacin (diamonds), or roxithromycin (circles) in N=103, 106, 88, 178, 175, 122, and 265 individual E. coli, respectively. Each data point represents the values of two kinetic parameters inferred for an individual bacterium from the data in Figure 1—figure supplement 2 using our mathematical model. Statistical classification of the accumulation of (D) membrane-targeting antibiotics (i.e. polymyxin B, tachyplesin, and octapeptin) vs intracellular-targeting antibiotics (i.e. linezolid, trimethoprim, ciprofloxacin, and roxithromycin), (E) polymyxin B, (F) tachyplesin, or (G) octapeptin vs the remaining membrane-targeting antibiotics, (H) linezolid, (I) trimethoprim, (J) ciprofloxacin, or (K) roxithromycin vs remaining antibiotics with an intracellular target. These confusion tables are predictions generated using only the two kinetic parameters that can be rapidly measured experimentally, namely t₀ and k₁. Similar statistical classifications were obtained when using the full set of kinetic parameters, i.e., k₂, d_r, and d_c in addition to t₀ and k₁.