Some Thoughts About EAEC, Reservoirs, and Pathogenicity

Regarding the German outbreak strain of E. coli, the data are fairly clear: it is an enteroaggregative E. coli ('EAEC') which has acquired antibiotic resistance genes and a Shiga-like toxin from an Shiga-toxinogenic E. coli ('STEC'). EAEC are interesting--according to the European Food Safety Authority:

EAEC have been implicated as a cause of persistent diarrhea in children and acquired immunodeficiency syndrome-associated diarrhea, as well as acute diarrhea in travelers. Not all EAEC strains have been shown to cause diarrhea in humans. The EAEC are a heterogeneous group of bacteria that display a wide array of virulence factors. The relative contribution to disease for each one of these virulence factors is unclear. Furthermore, the interaction between the host immune response and heterogeneity of virulence factors of EAEC is complex. Outbreaks of diarrhoeal illness due to EAEC have been reported and linked to the ingestion of food which was contaminated by food handlers. In addition, it has been shown that EAEC carriage by humans is possible (Huang et al., 2003; Huang et al., 2006).

...EAEC have rarely been identified in animals, suggesting that they are not zoonotic, but exclusive to humans as a pathogen (Cassar et al., 2004).

Basically, EAEC are very good at sticking to various tissues, and thus are can cause persistent diarrhea. Add a nasty toxin, such as Shiga-like toxin, and suddenly you have a very dangerous infection. Unlike STEC, many EAEC strains which possess the genes involved in causing disease, especially the 'stickiness' genes, don't cause disease.

In fact, they seem to be infant specialists--strains that colonize newborns and small children*. Only a subset of EAEC appear to cause disease. In a study of Nicaraguan infants, 27.8% of children with diarrhea tested positive for EAEC (the reminder had other kinds of diarrhea-causing E. coli). ZOIKS! But 33.1% of infants in the control group (no diarrhea in the previous month) also carried EAEC. Using phenotypic data (let's not get started about E. coli typing and portability...), the EAEC isolates from both the healthy and sick children were very diverse, suggesting that there are many different types of EAEC and 'sorta-EAEC.' EAEC and 'sorta-EAEC' seem to be common--this isn't an oddball strain or two.

As more genomes of EAEC and asymptomatic EAEC are sequenced, we could discover if the ability to cause diarrhea is a trait that has been lost by some EAEC (aka 'the E. coli formerly known as EAEC'), or if the non-disease causing, 'sticky' form came first, and then occasionally evolved into a more virulent form.

Of course, this being biology, both scenarios are probably operative....

*Shigella, which are E. coli, are also thought to reside in some children asymptomatically.

Cited articles: Vilchez S, Reyes D, Paniagua M, Bucardo F, Möllby R, Weintraub A. 2009. Prevalence of diarrhoeagenic Escherichia coli in children from León, Nicaragua. J Med Microbiol. 58: 630-7. doi: 10.1099/jmm.0.007369-0

Reyes D, Vilchez S, Paniagua M, Colque-Navarro P, Weintraub A, Möllby R, Kühn I. 2010. Diarrheagenic Escherichia coli markers and phenotypes among fecal E. coli isolates collected from Nicaraguan infants. J Clin Microbiol. 48: 3395-6. doi:10.1128/JCM.00228-10

More like this

Has it been established that the strain is an EAEC that acquired EHEC virulence determinants rather than vice versa?

By Anonymous (not verified) on 14 Jun 2011 #permalink

Reply to anonymous: yes it has. Have a look at this phylogenetic tree of E. coli:

http://bacpathgenomics.wordpress.com/2011/06/15/snp-base-phylogeny-conf…

You can clearly see that this is a genome nearly identical to the EAEC strain Ec55989 from Africa, which has acquired the Shiga-toxin via phage which is what makes it dangerous. It also carries the classic agreggative adhesion clusters of genes (on a plasmid), which are typical of EAEC.

The outbreak strain is about as distinct as you can possibly be from E. coli O157:H7, the classic EHEC strains (see the tree), and it doesn't carry the LEE pathogeneticity island that is typical of EHEC.