Infections with enterohemorrhagic Escherichia coli (EHEC) cause disease ranging from mild diarrhea to hemolytic-uremic syndrome (HUS) and are the most common cause of renal failure in children in high-income countries. The severity of the disease derives from the release of Shiga toxins (Stx). The use of antibiotics to treat EHEC infections is generally avoided, as it can result in increased stx expression. Here, we systematically tested different classes of antibiotics and found that their influence on stx expression and release varies significantly. We assessed a selection of these antibiotics in vivo using the Citrobacter rodentium ϕstx2dact mouse model and show that stx2d-inducing antibiotics resulted in weight loss and kidney damage despite clearance of the infection. However, several non-Stx-inducing antibiotics cleared bacterial infection without causing Stx-mediated pathology. Our results suggest that these antibiotics might be useful in the treatment of EHEC-infected human patients and decrease the risk of HUS development.

Colicins are toxins that mediate interference competition in microbial ecosystems. They serve as a “common good” for the entire producer population but are synthesized by only few members which pay the costs of colicin production. We have previously shown that production of colicin Ib (cib), a group B colicin, confers a competitive advantage to Salmonella enterica serovar Typhimurium (S. Tm) over commensal E. coli strains. Here, we studied regulation of S. Tm cib expression at the single cell level. Comparative analysis of a single- and a multicopy gfp-reporter for the colicin Ib promoter (Pcib) revealed that the latter yielded optimal signal intensity for a diverse range of applications. We further validated this reporter and showed that gfp expression correlated well with colicin Ib (ColIb) protein levels in individual cells. Pcib is negatively controlled by two repressors, LexA and Fur. Only a small fraction of S. Tm expressed cib under non-inducing conditions. We studied Pcib activity in response to mitomycin C mediated DNA damage and iron limitation. Both conditions, if applied individually, lead to an increase in the fraction of GFP+ S. Tm, albeit an overall low fluorescence intensity. When both conditions were applied simultaneously, the majority of S. Tm turned GFP+ and displayed high fluorescence intensity. Thus, both repressors individually confine cib expression to a subset of the population. Taken together, we provide the first thorough characterization of a conventional gfp-reporter to study regulation of a group B colicin at the single cell level. This reporter will be useful to further investigate the costs and benefits of ColIb production in human pathogenic S. Tm and analyze cib expression under environmental conditions encountered in the mammalian gut.

Stx2 is the major virulence factor of EHEC and is associated with an increased risk for HUS in infected patients. The conditions influencing its expression in the intestinal tract are largely unknown. For optimal management and treatment of infected patients, the identification of environmental conditions modulating Stx2 levels in the human gut is of central importance. In this study, we established a set of chromosomal stx2 reporter assays. One system is based on superfolder GFP (sfGFP) using a T7 polymerase/T7 promoter-based amplification loop. This reporter can be used to analyze stx2 expression at the single-cell level using FACSs and fluorescence microscopy. The other system is based on the cytosolic release of the Gaussia princeps luciferase (gluc). This latter reporter proves to be a highly sensitive and scalable reporter assay that can be used to quantify reporter protein in the culture supernatant. We envision that this new set of reporter tools will be highly useful to comprehensively analyze the influence of environmental and host factors, including drugs, small metabolites and the microbiota, on Stx2 release and thereby serve the identification of risk factors and new therapies in Stx-mediated pathologies.