Genome of Ultra-Resistant E. coli Published

One of things I've done in my job is write letters of recommendation for various genome sequencing projects, particularly antibiotic resistance related projects, so it's always good to see that those letters might result in published work. So onto to an incredibly resistant Escherichia coli strain.

E. coli SMS-3-5 is a strain of environmental E. coli that is incredibly resistant to a broad range of antibiotics, to the point where the antibiotic does not go into solution. Of 33 antibiotics tested, only imipenem, meropenem, amikacin, gentamicin, and nitrofurantoin were effective against this bacterium. Before I go any further, I want to stress one key point: this is not a clinical isolate. It was isolated from Charleston Harbor, South Carolina. As I've beaten to death before, the resistance problem does not start at the hospital door. OK, back to the bug.

What's interesting is that most of the antibiotic resistance appears to be encoded in a 42 kb region on a single plasmid (mini-chromosome)--and, yes, the plasmid can be transferred between strains. Then there's the quinolone resistance, which includes ciprofloxacin ('Cipro') and moxifloxacin, both of which are medically critical frontline antibiotics. Ciprofloxacin resistance in this strain is the second highest ever recorded, while moxifloxacin resistance is the highest recorded. Not only are the usual point mutations in the polymerase observed, but there appear to be several efflux pump systems (protein complexes that pump stuff out of the cell) involved, as well as a couple of altered 'normal' uptake systems. The point mutations are not that much of a concern (because they're not likely to be transferred between strains), but the uptake systems mean that novel genetic elements are involved in quinolone resistance (which could be transferred).

The ecology and evolution of this strain is also odd. This strain can grow on sucrose and raffinose, and, unusually, these pathways are chromosomally encoded along with other genes that are usually found on plasmids. Given the high number of insertion elements in this region, I think this might be a recently integrated plasmid. This could be one way that "pathogenicity islands"--regions of the chromosome that usually encoded disease-related functions and often differ in sequence (e.g., GC content) from the rest of the genome--have evolved.

It also is a phylogenetic outlier to the other sequenced genomes. Currently, I'm leading a project in which we're sequencing over 100 commensal E. coli, including some E. coli that are 'lake bloom' E. coli which grow in aquatic habitats. I'll be curious to see how similar (and related) SMS-3-5 is to these E. coli. On the other hand, SMS-3-5 also has similarities to avian pathogenic E. coli--strains that cause disease in birds.

SMS-3-5: a very interesting and disturbing E. coli.

Cited article: Fricke, W.F., Wright, M.S., Lindell, A.H., Harkins, D.M., Baker-Austin, C., Ravel, J., and Stepanauskas, R. 2008. Insights into the environmental resistance gene pool from the genome sequence of the multidrug-resistant environmental isolate Escherichia coli SMS-3-5. J Bacteriol. 190:6779-94.