1 Introduction

Vast majority of Escherichia coli strains are harmless to humans, and they occur as a part of natural microbiota in the human gut. However, Shiga toxin-producing E. coli (STEC), and particularly a subset of these strains, called enterohemorrhagic E. coli (EHEC) are highly pathogenic (Karmali 2017). Usually STEC cause local outbreaks, which are often severe and dangerous, like that which took place in Germany in 2011 (Bloch et al. 2012).

STEC virulence depends on production of Shiga toxins. These strong toxins are encoded by stx genes, located in genomes of Shiga toxin-converting (Stx) prophages. Therefore, for expression of these genes, Shiga toxin production, and virulence of STEC, prophage induction (and subsequent expression of phage genes) is necessary (Łoś et al. 2011). Treatment of STEC infections is problematic, as many antibiotics cause prophage induction, thus enhancing pathogenicity even if bacterial cells are eventually killed; this is the reason of relatively high mortality among STEC-infected patients and a need for both development of new treatment options and prevention of Stx prophage induction (Hall et al. 2017).

In the light of the mechanism of STEC pathogenicity, and the fact that these bacteria are harmless if the Stx phage is kept in lysogenic bacteria in the form of the prophage, it appears important to learn about agents which might stimulate Stx prophage induction in human intestine. Although there are many physical and chemical factors able to cause this process, including UV irradiation, mitomycin C, high salt concentrations, EDTA, and hydrogen peroxide (Łoś et al. 2009; Harris et al. 2012; Imamovic and Muniesa 2012), it is not clear whether particular food products can either stimulate or prevent Stx prophage induction. Very recent report on food-related stress demonstrated that particular environmental factors (hydrogen peroxide and acids) could stimulate Stx prophage induction, however, tested stressors, rather than food products, were added to the experimental system (Fang et al. 2017). Therefore, we aimed to perform a pilot study to test whether commonly used drinks may influence efficiency of Stx prophage induction in E. coli cells.

2 Results and discussion

To test effects of various commonly used drinks on efficiency of Stx prophage induction in E. coli, MG1655 strain (Jensen 1993) and its derivative carrying a modified (for safety reasons) Stx 933W prophage, 933W Δstx2::cat-gfp (Nowicki et al. 2015), were employed. Following drinks were tested at indicated final concentrations in culture medium: Coca-Cola (final concentration 5%), Nestea (final concentration 5%), and vodka Żubrówka (final ethanol concentration 1%). These concentrations were chosen on the basis of previously reported volumes of human stomach and intestine (Schiller et al. 2005), and assuming one portion of each drink to be 100 ml. Thus, concentrations of the tested drinks in our experimental system could roughly correspond to their concentrations in human gastrointestinal tract shortly after consumption. Water and mitomycin C (1 μM), an antibiotic causing prophage induction, were used as negative and positive control, respectively.

When tested compounds were added to cultures of E. coli lysogenic with 933W Δstx2::cat-gfp, an inhibition of bacterial growth (no increase in absorbance of the culture after 120 min, contrary to earlier measurements) was observed in mitomycin C-treated cells. This may indicate prophage induction and subsequent lysis of host cells by bacteriophages (Fig. 1). Among tested drinks, there were only minor (not statistically significant) or negligible effects on bacterial culture growth (Fig. 1). This might be interpreted as either no induction of prophages or relatively low efficient induction. Nevertheless, because even small doses of Shiga toxins may be deleterious to humans, more detailed studies were performed.

Fig. 1
figure 1

Growth of E. coli MG1655 (933W Δstx2::cat-gfp) in LB medium at 37 °C following supplementation of the medium with different drinks. Following drinks were added to the medium at time 0: Coca-Cola (to final concentration of 5%), Nestea Lemon (to final concentration of 5%) and vodka Żurbówka (to final ethanol concentration of 1%). In control experiments, mitomycin C (positive control, added to final concentration of 1 μM) and water (negative control, added in the amount of 1:20 of the medium volume) were added at time 0. OD595 of bacterial cultures was measured at indicated times. Results of 3 independent experiments are shown with error bars indicating SD. Asterisks indicate statistically significant differences (p < 0.05) in the values (calculated as changes in OD595 during 30 min growth) determined in experiments with mitomycin C relative to the negative control (water). No statistically significant differences were detected between other values

Full size image

To assess efficiency of prophage induction more directly, samples of bacterial cultures were treated with chloroform, and lysates were titrated (using the MG1655 strain) for the presence of virions. As expected, treatment with mitomycin C caused appearance of about 100-times more pfu/cell (plaque forming units per cell) in the culture than in negative control experiments (with water; in this control, liberated phages represented effects of spontaneous prophage induction). Addition of Coca-Cola or vodka did not affect the efficiency of prophage induction significantly (Fig. 2a). However, in the presence of 5% Nestea, the number of pfu/cell in the culture was about ten times higher than in the negative control (Fig. 2a). These results indicate that Nestea may stimulate induction of Stx prophages in E. coli cells. Expression of stx2::cat-gfp has been monitored by Western blotting with anti-GFP antibodies. No signal could be detected at time 0 in any sample. Induction of the prophage by mitomycin C resulted in efficient expression of the fusion gene, as indicated by a strong signal on Western blot (Fig. 2b). Such a signal, though significantly less strong, could be observed in bacteria cultured in the presence of 5% Nestea for 180 min, while signals from cells treated with Coca-Cola, vodka or water were below the detection limit (Fig. 2b).

Fig. 2
figure 2

Induction of the 933W Δstx2::cat-gfp prophage and its lytic propagation in E. coli MG1655 (933W Δstx2::cat-gfp) (a), and expression of the stx2::cat-gfp fusion gene as assessed by Western blotting with anti-GFP antibodies (b), after supplementation of the medium with different drinks. Following drinks were added to the medium at time 0: Coca-Cola (to final concentration of 5%), Nestea Lemon (to final concentration of 5%) and vodka Żurbówka (to final ethanol concentration of 1%). In control experiments, mitomycin C (positive control, added to final concentration of 1 μM) and water (negative control, added in the amount of 1:20 of the medium volume) were added at time 0. Number of liberated phage progeny (which reflect frequency of prophage induction and efficiency of phage lytic development) per 1 lysogenic bacterial cell was determined at indicated times by phage titration on E. coli MG1655 host strain and calculation of the relative burst size taking into account the number of bacterial cells in the culture. At time 0, relative phage burst size was below 0.05 in all tested samples. Results of 3 independent experiments are shown with error bars indicating SD. Asterisk indicates statistically significant difference (p < 0.05) relative to the negative control (water). For Western blotting experiments, 1 × 109 cells were harvested by centrifugation, lysed, and the proteins were separated by SDS-PAGE, followed by Western blotting with anti-GFP antibodies. A representative blot is shown. Positions of molecular weight markers are indicated. For loading control, SDS-PAGE gel was stained with Coomassie Brilliant Blue (shown below the blot)

Full size image

The results presented in this report indicate that some commonly used drinks, like Nestea, may stimulate induction of Stx prophages in E. coli lysogenic cells. On the basis of these experiments, it is not possible to distinguish what stage of bacteriophage propagation is more efficient in the presence of Nestea: frequency of prophage induction or efficiency of intracellular lytic development of the phage after induction, or both. Nevertheless, irrespective of which option is true, significantly higher number of phage progeny, relative to control experiments, means more efficient production of new virions that must be preceded by more efficient expression of phage genes, including stx genes, implicating significantly more effective production of Shiga toxins. Such increased expression of stx after 933W prophage induction has already been demonstrated experimentally (Łoś et al. 2009), and is also confirmed here (Fig. 2b) by monitoring levels of GFP as a product of expression of the fusion gene stx2::cat-gfp. This suggests that consumption of Stx prophage induction-stimulating drinks by persons infected with STEC might cause either onset of the disease or more severe symptoms. One might speculate what components of Nestea Lemon, tested in this work, could cause stimulation of prophage induction or phage lytic development. However, apart from known chemicals included in this drink (sugar, citric acid, sodium citrate, and ascorbic acid), it also contains black tea extract and lemon juice which composition is not strictly defined. Recent studies suggested that sodium citrate may increase production of Shiga toxins by STEC (Lenzi et al. 2016), however, it was reported previously that citrate inhibits Stx phage lytic development (Nejman-Faleńczyk et al. 2012). Therefore, further extensive studies are necessary to learn about agents present in Nestea and influencing prophage induction and/or phage lytic development. The concentrations of the drinks in the medium (5%), tested in this work, may likely occur in the human digestive tract, particularly shortly after their consumption. Thus, we speculate that their effects on prophage induction in STEC infecting human intestine might be similar to those observed in the artificial system employed in this study. However, one should take into consideration that conditions in human intestine differ significantly from those in our experiments. Nevertheless, if our assumption is true, such drinks should be avoided if STEC infection is confirmed or even suspected.

This report provides results of the pilot study suggesting that consumption of particular food products may affect Stx prophage induction in STEC strains significantly. Therefore, we conclude that further complex studies on effects of a large set of food products (not only drinks but also other nutrients) on Stx prophage induction should be important to protect consumers infected with STEC strains and to assess safety of food to be used by patients suffering from such infections.