Induction of Shiga Toxin-Encoding Prophage by Abiotic Environmental Stress in Food

Abstract

The prophage-encoded Shiga toxin is a major virulence factor in Stx-producing Escherichia coli (STEC). Toxin production and phage production are linked and occur after induction of the RecA-dependent SOS response. However, food-related stress and Stx-prophage induction have not been studied at the single-cell level. This study investigated the effects of abiotic environmental stress on stx expression by single-cell quantification of gene expression in STEC O104:H4 Δstx2::gfp::amp^r In addition, the effect of stress on production of phage particles was determined. The lethality of stressors, including heat, HCl, lactic acid, hydrogen peroxide, and high hydrostatic pressure, was selected to reduce cell counts by 1 to 2 log CFU/ml. The integrity of the bacterial membrane after exposure to stress was measured by propidium iodide (PI). The fluorescent signals of green fluorescent protein (GFP) and PI were quantified by flow cytometry. The mechanism of prophage induction by stress was evaluated by relative gene expression of recA and cell morphology. Acid (pH < 3.5) and H₂O₂ (2.5 mM) induced the expression of stx₂ in about 18% and 3% of the population, respectively. The mechanism of prophage induction by acid differs from that of induction by H₂O₂ H₂O₂ induction but not acid induction corresponded to production of infectious phage particles, upregulation of recA, and cell filamentation. Pressure (200 MPa) or heat did not induce the Stx2-encoding prophage (Stx2-prophage). Overall, the quantification method developed in this study allowed investigation of prophage induction and physiological properties at the single-cell level. H₂O₂ and acids mediate different pathways to induce Stx2-prophage.IMPORTANCE Induction of the Stx-prophage in STEC results in production of phage particles and Stx and thus relates to virulence as well as the transduction of virulence genes. This study developed a method for a detection of the induction of Stx-prophages at the single-cell level; membrane permeability and an indication of SOS response to environmental stress were additionally assessed. H₂O₂ and mitomycin C induced expression of the prophage and activated a SOS response. In contrast, HCl and lactic acid induced the Stx-prophage but not the SOS response. The lifestyle of STEC exposes the organism to intestinal and extraintestinal environments that impose oxidative and acid stress. A more thorough understanding of the influence of food processing-related stressors on Stx-prophage expression thus facilitates control of STEC in food systems by minimizing prophage induction during food production and storage.

Keywords: E. coli O104:H4; RecA-independent SOS response; STEC; Shiga toxin; Shiga toxin-producing E. coli; environmental stress; membrane permeability; prophage induction; single-cell detection.

FIG 1

Flow cytometric quantification of gfp fluorescence in E. coli O104:H4 Δstx₂::gfp::amp^r in LB or after induction by mitomycin C. Exponential-phase E. coli O104:H4 Δstx₂::gfp::amp^r cultures were incubated in LB (A) or incubated after addition of mitomycin C to a concentration of 0.5 mg/liter (B). At each time point, the bacterial culture was harvested and the proportion of GFP fluorescent and filamented cells was quantified by flow cytometry. The gating for forward light scatter and GFP expression was set to account more than 96% of the cells in the control as normal sized and GFP negative. ○ and ●, GFP-positive cells; △ and ▲, GFP-negative cells. Open symbols, normal-sized cells; closed symbols, filamented cells. Data represent means ± standard deviations from three independent experiments.

FIG 2

Reduction of cell counts of E. coli O104:H4 Δstx₂::gfp::amp^r after exposure to heat, acid, and oxidative stress or to 200 MPa. Data represent means ± standard deviations from three independent experiments.

FIG 3

Effect of heat and H₂O₂ on gfp expression and membrane permeability in E. coli O104:H4 Δstx₂::gfp::amp^r. Exponential-phase E. coli O104:H4 Δstx₂::gfp::amp^r cultures were incubated in LB broth at 37°C (not shown), at 50°C (A), or at 37°C with addition of 2.5 mM H₂O₂ (B) for 3 h. At each time point, the bacterial culture was diluted and stained with PI. GFP and PI fluorescence was quantified by flow cytometry. Control conditions did not alter the GFP expression or the membrane permeability of cells (Fig. 1 and data not shown). ○ and ●, PI permeable; △ and ▲, PI impermeable. Open symbols, GFP negative; closed symbols, GFP positive. An arrow indicates the time when treatment reduced cell counts by 1 to 2 log CFU/ml. Data represent means ± standard deviations from three independent experiments.

FIG 4

Effect of HCl and lactic acid on gfp expression and membrane permeability in E. coli O104:H4 Δstx₂::gfp::amp^r. Exponential-phase E. coli O104:H4 Δstx₂:::gfp::amp^r cultures were incubated in LB broth (not shown), in LB acidified with HCl to pH 2.5 (A), or in LB acidified with lactic acid to pH 3.5 (B) for 3 h. At each time point, the bacterial culture was diluted and stained with PI. GFP and PI fluorescence was quantified by flow cytometry. Control conditions did not alter the GFP expression or the membrane permeability of cells (Fig. 1 and data not shown). ○ and ●, PI permeable; △ and ▲, PI impermeable. Open symbols, GFP negative; closed symbols, GFP positive. An arrow indicates the time when treatment reduced cell counts by 1 to 2 log CFU/ml. Data represent means ± standard deviations from three independent experiments.

FIG 5

Effect of pressure on gfp expression and membrane permeability in E. coli O104:H4 Δstx₂::gfp::amp^r. Exponential-phase E. coli O104:H4 Δstx₂::gfp::amp^r cultures in LB broth were heat sealed in tubes and incubated at 37°C (A) or heat sealed in tubes and incubated at 200 MPa at 20°C for 7 min, followed by incubation at 37°C and ambient pressure for 3 h (B). At each time point, the bacterial culture was diluted and stained with PI. GFP and PI fluorescence was quantified by flow cytometry. ○ and ●, PI permeable; △ and ▲, PI impermeable. Open symbols, GFP negative; closed symbols, GFP positive. Pressure treatment reduced cell counts by 1 to 2 log CFU/ml. Data represent means ± standard deviations from three independent experiments.

FIG 6

Effect of stress on cell morphology of E. coli O104:H4 Δstx₂::gfp::amp^r as observed by phase-contrast microscopy. Shown are untreated exponential-phase cells (A) and exponential-phase cells treated with 200 MPa for 40 min (B), with HCl for 1 h (C), with lactic acid for 20 min (D), with 2.5 mM H₂O₂ for 1 h (E), and with mitomycin (0.5 mg/liter) for 3 h (F). Five pictures per sample were analyzed, corresponding to 500 to 1,000 cells per sample. Three biological replicates were done for each treatment. Scale bars represent 10 μm.

FIG 7

Expression of recA in E. coli O104:H4 Δstx₂::gfp::amp^r after stress treatments. Relative gene expression was quantified by RT-qPCR with gapA as a housekeeping gene and exponential cultures in LB broth as reference conditions. Exponential-phase cultures were treated with LB broth containing mitomycin C (0.5 mg/liter) or H₂O₂ (2.5 mM) for 40 min or with acidified LB broth with HCl or lactic acid for 20 min. Values for different treatments that do not share lowercase letters are significantly different (P < 0.05). Data represent means ± standard errors of the means from four independent experiments.

FIG 8

Dot plot of forward scatter and GFP fluorescence for E. coli O104:H4 Δstx₂::gfp::amp^r grown in LB broth (left panel) or treated with mitomycin C (0.5 mg/liter) for 180 min (right panel). The population was divided into four subpopulations by FSC and GFP reference lines. The reference lines were determined from untreated samples, where at least 96% of the population was GFP negative as shown in Fig. 1B. Percentage values of four populations (Q1, FSC⁻ GFP⁺; Q2, FSC⁺ GFP⁺; Q3, FSC⁺ GFP⁻; Q4, FSC⁻ GFP⁻) were automatically calculated and are presented in the corners of the graphs.

FIG 9

Example of a dot plot of GFP fluorescence and PI fluorescence for E. coli O104:H4 Δstx₂::gfp::amp^r grown in LB broth (left panel) or treated with HCl (pH 2.5) for 20 min (right panel). The population was divided into four subpopulations by GFP and PI reference lines. The reference lines were determined from untreated samples, where at least 97% of the population is GFP and PI negative. Percentage values of four populations (Q1, GFP⁻ PI⁺; Q2, GFP⁺ PI⁺; Q3, GFP⁺ PI⁻; Q4, GFP⁻ PI⁻) were automatically calculated and are presented in the corners of the graphs.