FEATURE ARTICLE
The Evolutionary Ecology of Escherichia coli
Abundantly studied and much feared, E. coli has more genomic plasticity than once believed and may have followed various routes to become a pathogen
Valeria Souza, Amanda Castillo, Luis Eguiarte
Where Pathogenesis Comes From
Two of the most dangerous types of E. coli are the EHEC, or enterohemorrhagic, strains such as O157:H7 often implicated in food poisoning, and the EPEC, or enteropathogenic, strains, which are an important cause of infantile diarrhea in the developing world. These strains can adhere to the wall of the intestine and rub off the epithelial villi, creating a lesion that disrupts the function of the intestinal wall and causes severe diarrhea that is often fatal, particularly in children and the elderly. The lesion is called an attachment-and-effacement lesion, abbreviated A/E.
Genetic and genomic studies have identified the genetic underpinnings of this process. The chromosomes of these E. coli (and of some other enteric pathogens, such as Shigella and Salmonella) include the LEE. In addition, they have a plasmid called EAF (for "EPEC adherence factor"). Within the LEE we find genes for a type III secretion system, a system used to deliver toxins and other proteins to an infected host cell. The cassette also includes signal-transduction proteins and a gene called eae, which codes for intimin, a protein that tightly binds the bacterium to the host cell. The LEE package also includes tir, a gene coding for the intimin receptor Tir; genes for some secreted proteins (espA, espD, espB); and the regulator Ler. In the EAF plasmid there is a group of genes called bfp, responsible for producing a filament to attach the bacteria to the host cells, and the regulator Per, which controls the expression of the LEE genes through interaction with the chromosomal regulator Ler.
Within the past year the relations of various genes associated with the A/E lesion have been mapped in our laboratory with a branching diagram, or dendrogram, obtained by examining 155 Mexican strains. The genes cesT/eae and espB can be found together, forming part of the LEE, or existing independently in the strains of healthy animals. These genes appear to have other roles in addition to pathogenesis, since the gene espB predominates in animals with specialized diets, such as the ungulates, manatees, whales and bats. In contrast, the gene cesT/eae is most frequent in strains associated with rabbits, armadillos and anteaters. Both independent genes are found distributed throughout the dendrogram.
This suggests that cesT/eae and espB are ancient genes in E. coli, and that they have a role in the normal, nonpathogenic association with their host. In a sequence analysis of 50 strains that do and do not possess the LEE, we found that the genes eae and tir are the most variable in their sequences, and that the gene espB is less variable. There is an association between the types of sequences in eae and tir, although this is not clear with espB. From the same sample of the collection we also sequenced five genes outside the island (gapA, fimA, mdh, muts and putP). Additional statistical analysis suggests that pathogenic and nonpathogenic genes have different types of selection acting on them. Some genes show diversifying selection (an increase of diversity to escape the host response), whereas others show low levels of diversity owing to purifying selection (low diversity being selected in order to conserve a specific function). More interesting, LEE genes are almost twice as diverse as nonpathogenic genes, and they have very different substitution rates.
In 2000 Sean D. Reid and his colleagues at Pennsylvania State University proposed that pathogenic islands including the LEE are evolutionary units whose genes have integrated at the same time and evolved coordinately. If this is true, the origin of new pathogens could be explained by successive events of horizontal transfer, for example from EPEC to EHEC. But when we compared the genealogies for individual genes, we were surprised to see that each has a very different history to tell. In evolutionary terms, the LEE appears to be not a unit but rather a dynamic assemblage of genes that act together to make a lesion.
On the other hand, the plasmid genes per and bfp have never been detected in the strains associated with wild animals and are found only in EPEC and EHEC strains associated with human patients. This is evidence that horizontal-transfer events are important in the history of plasmid-carried pathogenesis.
It seems that the origin of new pathogens is a complex phenomenon produced by the dynamic between the chromosomal component (LEE, for example) and the plasmids within the bacterial populations. From these results we conclude that the evolution and emergence of new pathogens is still not well understood. We need to combine more work on population genetics, population ecology and molecular evolution in order to have a wider view and a better idea of the dynamics of E. coli natural populations; this is the only way to understand the evolutionary biology of this organism.
» Post Comment