Pathogenic potential of emergent sorbitol-fermenting Escherichia coli O157:NM

Tracy Rosser, Tracy Dransfield, Lesley Allison, Mary Hanson, Nicola Holden, Judith Evans, Stuart Naylor, Roberto La Ragione, J. Christopher Low, David L. Gally

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output

Non-sorbitol-fermenting (NSF) Escherichia coli O157:H7 is the primary Shiga toxin-producing E. coli (STEC) serotype associated with human infection. Since 1988, sorbitol-fermenting (SF) STEC O157:NM strains have emerged and have been associated with a higher incidence of progression to hemolytic-uremic syndrome (HUS) than NSF STEC O157:H7. This study investigated bacterial factors that may account for the increased pathogenic potential of SF STEC O157:NM. While no evidence of toxin or toxin expression differences between the two O157 groups was found, the SF STEC O157:NM strains adhered at significantly higher levels to a human colonic cell line. Under the conditions tested, curli were shown to be the main factor responsible for the increased adherence to Caco-2 cells. Notably, 52 of 66 (79%) European SF STEC O157:NM strains tested bound Congo red at 37 degrees C and this correlated with curli expression. In a subset of strains, curli expression was due to increased expression from the csgBAC promoter that was not always a consequence of increased csgD expression. The capacity of SF STEC O157:NM strains to express curli at 37 degrees C may have relevance to the epidemiology of human infections as curliated strains could promote higher levels of colonization and inflammation in the human intestine. In turn, this could lead to increased toxin exposure and an increased likelihood of progression to HUS.
Original languageEnglish
Pages (from-to)5598-5607
Number of pages10
JournalInfection and Immunity
Issue number12
Publication statusPublished - Dec 2008


Dive into the research topics of 'Pathogenic potential of emergent sorbitol-fermenting Escherichia coli O157:NM'. Together they form a unique fingerprint.

Cite this