One of the main virulence factors produced by Shiga toxin-producing Escherichia coli is the Shiga toxin (Stx), which is encoded on lambdoid phages (Stx phage). In Norway, an outbreak of hemorrhagic colitis and hemolytic uremic syndrome (HUS) caused by E. coli O103:H25 was reported during the winter of 2006, but stx(2)-positive isolates were only retrieved from two human samples.
Isolates of E. coli O103:H25 from patients with HUS in Norway, including sporadic cases and the outbreak cases, were investigated for the presence of phages encoding stx(2). The induced Stx phages were characterized morphologically and genetically, and the host susceptibility for these phages of various E. coli O103 isolates, including O103:H25 stx(2) negative isolates from the outbreak, was tested by a plaque assay.
The Stx2 phages in this study are very closely related in terms of morphology, sequence identity, and host infectivity. There may be a conserved phage within the E. coli O103:H25 population.
It is proposed that the Stx2 phage, present in the environment either as free phage particles or within a limited pool of Stx-producing E. coli O103 strains, have infected or integrated in the stx(2)-negative E. coli O103:H25 isolates from the Norwegian outbreak.
Shiga toxins (Stx) are key virulence factors of Shiga toxin-producing Escherichia coli (STEC) during development of haemolytic uremic syndrome (HUS). It has been suggested that not only specific stx2 subtypes, but also the amount of Stx2 expressed might be essential for STEC pathogenicity. We aimed to investigate if various anti-terminator (q) genes might influence the expression level of Stx2 in highly virulent STEC. A multiplex PCR detecting q933, q21, and qO111 was run on 20 stx2a-positive STEC strains, of which 18 were HUS associated serotypes (HAS) and two non-HAS. Relative expression of Stx2 mRNA was assessed for all strains, both in non-induced and induced (mitomycin C) state. The HAS STEC carried either q933 (n = 8), qO111 (n = 8), or both (n = 2). In basal state, no STEC strains showed higher expression of Stx2 mRNA than the calibrator EDL933 (non-sorbitol fermenting (NSF) O157:H7carrying q933). Variations among strains were not associated with different q genes present, but rather related to specific serogroups. In induced state, O104:H4 strains (q933) showed higher Stx2 mRNA level than EDL933, whereas sorbitol fermenting (SF) O157:H- (qO111) and O121:H? (q933) STEC showed levels comparable with EDL933. An association between the presence of q933 and higher Stx2 level was seen within some HAS, but not all. Interestingly, the O103:H25 STEC strains, responsible for a HUS outbreak in Norway, carried both q933 and qO111. However, the Stx2 mRNA level in these strains was significantly lower than EDL933 in both states, indicating that other factors than the level of Stx2 might explain the aggressiveness of these bacteria. The two non-HAS STEC did not carry any of the examined q genes. In induced state, these bacteria showed the lowest Stx2 mRNA level compared to EDL933. One of the non-HAS STEC was not induced by mitomycin C, suggesting that stx2a might be located on a defect bacteriophage. No association between specific q genes and Stx2 mRNA expression level was revealed in stx2a-positive HAS STEC. Our results suggest that other factor(s) than specific q genes might influence the level of Stx2 produced in highly virulent STEC.