Differential induction of Shiga toxin in environmental Escherichia coli O145:H28 strains carrying the same genotype as the outbreak strains

Authors: Carter, Michelle; PHAM, ANTARES; DU, WEN-XIAN; He, Xiaohua

Submitted to: International Journal of Food Microbiology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/14/2020
Publication Date: 12/23/2020
Citation: Carter, M.Q., Pham, A.C., Du, W.N., He, X. 2020. Differential induction of Shiga toxin in environmental Escherichia coli O145:H28 strains carrying the same genotype as the outbreak strains. International Journal of Food Microbiology. 339. Article 109029. https://doi.org/10.1016/j.ijfoodmicro.2020.109029.
DOI: https://doi.org/10.1016/j.ijfoodmicro.2020.109029

Interpretive Summary: Shiga toxin-producing Escherichia coli (STEC) consists of a group of diverse strains differing greatly in genetic make-up and pathogenicity potential. Virulence of STEC strains depends on production of Shiga toxins (Stxs), however, factors governing production of Shiga toxins are not fully understood. In this study, production of Stxs under stresses STEC encounter in their natural habitats was examined. Because genes coding Stxs are located on prophage genomes, effective production of Shiga toxin requires induction of Stx-prophages. Among the 12 environmental STEC O145:H28 strains tested, only one cattle isolate, RM9154-C1, exhibited a similar pattern in Shiga toxin (Stx2) production, as strain RM13514, a clinical strain linked to the 2010 romaine lettuce-associated outbreak in U.S. RM9154-C1 produced the highest amount of Stx2a following treatment with mitomycin C, an antibiotic known to induce prophages, and treatment with enrofloxacin, a fluoroquinolone antibacterial drug that has been used in veterinary medicine including food-producing animals in the U.S. In contrast, RM9154-C1 produced the least amount of Stx2a following acid challenge and recovery, and exhibited the lowest fold of induction following treatment with hydrogen peroxide, a natural inducer of Stx-prophages. Examining the genome sequences of Stx-prophages revealed that only the Stx2a-prophage in strain RM9154-C1 was grouped together with the Stx2a-prophage in strain RM13514, in which both exhibited high sequence similarity with the Stx2a phages induced from STEC strains associated with high virulence, including those linked to the 2011 large outbreak of STEC infection in Germany. Variation in production of Stxs among the environmental strains was largely attributed to the type of Stx (Stx1 or Stx2), the Stx-prophage lineage, the chromosomal insertion sites, the genes encoding antiterminator Q that activates the expression of stx genes, and the late promoter PR’ region (Between the gene encoding antiterminator Q and the stx coding region). Thus, profiling of the Stx-prophage lineage, their insertion sites, and the late promoter PR’ region allows for a more accurate prediction of the virulence potential of STEC strains. Identification of a cattle isolate harboring a Stx2a-prophage associated with high virulence supports the premise that cattle, a natural reservoir of STEC, serves as a source of hypervirulent STEC strains.

Technical Abstract: Shiga toxin-producing Escherichia coli (STEC) O145 is a major serotype associated with severe human disease. Production of Shiga toxins (Stxs), especially Stx2a, is thought to be correlated with STEC virulence. Since stx genes are located in prophages genomes, induction of prophages is required for effective Stxs production. Here, we investigated the production of Stxs in 12 environmental STEC O145:H28 strains under stresses STEC encounter in natural habitats and performed comparative analysis with two O145:H28 clinical strains, one linked to a 2010 U.S. lettuce-associated outbreak (RM13514) and the other linked to a 2007 Belgium ice cream-associated outbreak (RM13516). Similar to the outbreak strains, all environmental strains belong to Sequence Type (ST)-78 using the EcMLST typing scheme. Although all Stx1a-prophages were grouped together, variations in Stx1a production were observed prior to or following the inductions. Among all stx2a positive environmental strains, only the Stx2a-prophage in cattle isolate RM9154-C1 was clustered with the Stx2a-prophages in RM13514, the Stx2a-phage induced from a STEC O104:H4 strain linked to the 2011 outbreak of enterohemorrhagic infection in Germany, and the Stx2a-prophage in STEC O157:H7 strain EDL933, a prototype of enterohemorrhagic E. coli. Furthermore, the Stx2a-prophage in RM9154-C1 shared the same chromosomal insertion site and carried the same antiterminator Q gene and the late promoter PR’ as the Stx2a-prophage in RM13514. Following mitomycin C or enrofloxacin treatment, the production of Stx2a in RM9154-C1 was the highest among all environmental strains tested. In contrast, following acid challenge and recovery, the production of Stx2a in RM9154-C1 was the lowest among all the environmental strains tested, at a level comparable to the clinical strains. A significant increase in Stx2a production was detected in all strains when exposed to H2O2, although the induction fold was much lower than those by other inducers. This low-efficiency induction of Stx-prophages by H2O2, a natural inducer of Stx-prophages, supports the hypothesis of bacterial altruism in controlling Stxs production, a strategy that assures the survival of the STEC population as a whole by sacrificing a small fraction of cells for Stxs production and release. Differential induction of Stxs among strains carrying nearly identical Stx-prophages suggests a role of host bacteria in regulating Stxs production. Our study revealed diverse Stx-prophages in STEC O145:H28 strains that were genotypically indistinguishable. Identification of a cattle isolate harboring a Stx2a-prophage associated with high virulence supports the premise that cattle, a natural reservoir of STEC, serve as a source of hypervirulent STEC strains.