The formation of persister cells in stationary-phase cultures of Escherichia coli is associated with the aggregation of endogenous proteins

Abstract

Persister cells (persisters) are transiently tolerant to antibiotics and usually constitute a small part of bacterial populations. Persisters remain dormant but are able to re-grow after antibiotic treatment. In this study we found that the frequency of persisters correlated to the level of protein aggregates accumulated in E. coli stationary-phase cultures. When 3-(N-morpholino) propanesulfonic acid or an osmolyte (trehalose, betaine, glycerol or glucose) were added to the growth medium at low concentrations, proteins were prevented from aggregation and persister formation was inhibited. On the other hand, acetate or high concentrations of osmolytes enhanced protein aggregation and the generation of persisters. We demonstrated that in the E. coli stationary-phase cultures supplemented with MOPS or a selected osmolyte, the level of protein aggregates and persister frequency were not correlated with such physiological parameters as the extent of protein oxidation, culturability, ATP level or membrane integrity. The results described here may help to understand the mechanisms underlying persister formation.

Figure 1. Stationary-phase E. coli cultures that accumulate protein aggregates produce increased levels of persisters.

(A) Growth curves of E. coli MC4100. The bacteria were grown aerobically (circles) or anaerobically (triangles) at 37°C in LB medium supplemented (black symbols) or not (white symbols) with 0.2% glucose. (B) Protein aggregates were isolated from the stationary-phase cultures as described in Materials and Methods. Equal volumes of insoluble fractions isolated in the sucrose gradient were resolved by SDS-PAGE and visualized by Coomassie staining. (C) After 24 h the cultures were diluted to an OD₅₉₅ = 0.1 and exposed to ampicillin (200 µg/ml) for 10 h at 37°C. Antibiotic- tolerant bacteria were plated for colony counting at the times indicated in the graph. The error bars indicate the standard deviations of three independent experiments.

Figure 2. Aggregation of proteins, formation of persisters and oxidation of proteins in the presence of acetate, MOPS and osmolytes.

Bacteria were grown aerobically for 24 h at 37°C in LB medium supplemented with 0.2% acetate, 40 mM MOPS pH 7.4, 0.2% glucose, 0.2% glycerol, 0.2% trehalose or 0.2% betaine, unless otherwise stated. The control culture was grown in LB without supplements. Protein aggregates were isolated from 24 h cultures, resolved by SDS-PAGE and visualized by Coomassie staining. Oxidized, DNPH-derivatized proteins were resolved by SDS-PAGE and immunodetected using anti-dinitrophenol antibodies (E). The amount of aggregated proteins in relation to total protein in cell extracts (A) and the relative level of oxidized proteins (C) were calculated on the basis of densitometry. (B) After 24 h the cultures were diluted to an OD₅₉₅ = 0.1 and exposed to ampicillin (200 µg/ml) for 10 h at 37°C. Antibiotic- tolerant bacteria were plated for colony counting at the times indicated in the graph. 100% corresponds to the number of colonies before antibiotic treatment. Means and standard deviation of three independent experiments are shown. (D) Representative growth curves of selected cultures are shown. (E) Samples corresponding to the same OD₅₉₅ were loaded on the gels. Lane 1- the control culture, lane 2- LB+acetate, lane 3 - LB +MOPS, lane 4 - LB+trehalose, lane 5- LB+glucose. The asterics indicates EF-Tu, the most abundant protein in the aggregates.

Figure 3. The effect of acetate, MOPS and osmolytes on the frequency of persister cells tolerant to kanamycin and ofloxacin.

Bacteria were grown aerobically for 24 h at 37°C in LB medium supplemented with 0.2% acetate, 40 mM MOPS pH 7.4, 0.2% glucose, 0.2% glycerol, 0.2% trehalose or 0.2% betaine. The control culture was grown in LB without supplements. To isolate persisters the cultures were diluted to an OD₅₉₅ = 0.1 and exposed to kanamycin (20 µg/ml) for 4 h or to ofloxacin (5 µg/ml) for 6 h at 37°C. The frequency of persisters was estimated as described in the legend to Figure 2. Means and standard deviation of three independent experiments are shown.

Figure 4. The influence of acetate, MOPS and osmolytes on the levels of non-dividing and persister cells.

Bacteria were grown aerobically for 24 h at 37°C in LB medium supplemented with 0.2% acetate, 40 mM MOPS pH 7.4, 0.2% glucose, 0.2% glycerol, 0.2% trehalose or 0.2% betaine. The control culture was grown in LB without supplements. (A) The total number of cells per ml was obtained using a hemocytometer. Dividing cell counts were estimated by plating serial dilutions on LA. The number of non-dividing cells was calculated by subtracting colony-forming units counts from the total number of cells. (B) E. coli stationary- phase cells grown in the presence of glucose are larger than control cells. 1000×magnification. (C) To isolate and estimate persisters the stationary cultures were diluted to an OD₅₉₅ = 0.1 and exposed to ampicillin (200 µg/ml) for 10 h at 37°C. Antibiotic- tolerant bacteria were plated for colony counting. 100% corresponds to the total number of cells (dividing and non-dividing) before antibiotic treatment. Means and standard deviation of three independent experiments are shown.

Figure 5. The level of ATP and membrane stability in E. coli stationary-phase cultures.

The cultures were grown as described in the legend to Figue 4. (A) ATP level was determined using BacTiter Glo chemicals (Promega). (B) Membrane stability was assessed after staining cells with the LIVE/DEAD assay solution (Invitrogen). Error bars represent standard deviation of three independent experiments. RFU, relative fluorescence units.

Figure 6. Persisters killing by antibiotics is more efficient in the presence of osmolytes.

To isolate persisters, 24 h stationary culture (LB) was diluted to an OD₅₉₅ = 0.1 in a fresh LB medium supplemented with amp (200 µg/ml). After 10 h of incubation at 37°C, persisters were pelleted, washed with 0.9% NaCl, resuspended in either M9 medium with 0.2% glucose or PBS. Persisters were then exposed to antibiotics (200 µg/ml amp, 20 µg/ml kan, 3 µg/ml oflox) at 37°C for 1.5 h in the absence or in the presence of 40 mM MOPS pH 7.4, 0.2% trehalose, 0.2% betaine or 0.2% glycerol. Means and standard deviation of three independent experiments are shown.