Research article

Characterisation of atypical enteropathogenic E. coli strains of clinical origin

Sharon M Tennant^1,3* , Marija Tauschek^1* , Kristy Azzopardi¹ , Andrea Bigham¹ , Vicki Bennett-Wood¹ , Elizabeth L Hartland¹ , Weihong Qi^2,4 , Thomas S Whittam^{2^} and Roy M Robins-Browne¹

¹  Department of Microbiology and Immunology, The University of Melbourne, and Murdoch Childrens Research Institute, Royal Children's Hospital, Victoria 3010, Australia

²  Microbial Evolution Laboratory, National Food Safety and Toxicology Center, Michigan State University, Michigan, USA

³  Center for Vaccine Development, University of Maryland School of Medicine, Baltimore, MD 21201, USA

⁴  Functional Genomics Center Zurich, Uni/ETH Zurich, Winterthurerstrasse 190/Y32 H66, CH-8057 Zurich, Switzerland

author email corresponding author email^Deceased * Contributed equally

BMC Microbiology 2009, 9:117doi:10.1186/1471-2180-9-117

Published:	3 June 2009

Abstract

Background

Enteropathogenic E. coli (EPEC) is a prominent cause of diarrhoea, and is characterised in part by its carriage of a pathogenicity island: the locus for enterocyte effacement (LEE). EPEC is divided into two subtypes according to the presence of bundle-forming pili (BFP), a fimbrial adhesin that is a virulence determinant of typical EPEC (tEPEC), but is absent from atypical EPEC (aEPEC). Because aEPEC lack BFP, their virulence has been questioned, as they may represent LEE-positive Shiga toxin-producing E. coli (STEC) that have lost the toxin-encoding prophage, or tEPEC that have lost the genes for BFP. To determine if aEPEC isolated from humans in Australia or New Zealand fall into either of these categories, we undertook phylogenetic analysis of 75 aEPEC strains, and compared them with reference strains of EPEC and STEC. We also used PCR and DNA hybridisation to determine if aEPEC carry virulence determinants that could compensate for their lack of BFP.

Results

The results showed that aEPEC are highly heterogeneous. Multilocus sequence typing revealed that 61 of 75 aEPEC strains did not belong to known tEPEC or STEC clades, and of those that did, none expressed an O:H serotype that is frequent in tEPEC or STEC strains associated with disease. PCR for each of 18 known virulence-associated determinants of E. coli was positive in less than 15% of strains, apart from NleB which was detected in 30%. Type I fimbriae were expressed by all aEPEC strains, and 12 strains hybridised with DNA probes prepared from either bfpA or bfpB despite being negative in the PCR for bfpA.

Conclusion

Our findings indicate that clinical isolates of aEPEC obtained from patients in Australia or New Zealand are not derived from tEPEC or STEC, and suggest that functional equivalents of BFP and possibly type I fimbriae may contribute to the virulence of some aEPEC strains.