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
Population Genetics
In bacteria, reproduction is not tied to sexuality. Bacteria divide by binary fission to produce clones. Genetic variation comes about primarily by way of mutations passed along to clones. Horizontal transfer, which can be regarded as a parasexual process, serves as an additional source of variability in populations. In a population or species, the balance between these two processes is called the degree of clonality.
A highly clonal species is distinguished by a collection of independently evolved lineages. In these cases, it is difficult to speak of a species, since there is no pool of shared genes. It is much more difficult to apply classical population-genetics theory and concepts. Evolution is accomplished by substitutions of complete lineages, whether by selection or by genetic drift. If, however, bacterial species exhibit high levels of recombination, one obtains panmictic populations, and it is possible to apply approximately the same ideas about populations and species that we use with diploid organisms (organisms with double sets of chromosomes). Until the mid-1990s most evidence led scientists to describe E. coli as a clonal species, although its capability for gene transfer was well known.
The fundamental problem is that most bacteria exhibit a great number of possible mechanisms of recombination, but these are not used in every generation (that is, reproduction is uncoupled from sexuality). Also, it is hard to get direct estimates of the degree of sexuality of bacterial populations—the relative importance of processes other than fission. In order to study them one needs to use indirect methods derived from the theory of population genetics.
Classical studies of the genetic structure and clonality of E. coli revealed high levels of genetic variation within its populations; values of one common measure of variation reach from 0.47 to 0.52 on a scale of 0 (no variation) to 1 (each individual unique). But the number of multiloci genotypes, or complete genetic patterns, is small. In fact, initially it was estimated that the "linkage disequilibrium," a patterning of gene arrangements different from what random mixing of the gene pool (sexual reproduction) would produce, was around the maximum. This would indicate that few of all the possible genotypes were found, or that there was no recombination between strains. Studies completed from 1980 through 1992 almost without exception concluded that recombination is a rare phenomenon in E. coli asssociated with humans or domesticated animals and suggested (using population-genetics theory) that the effective size of the reproducing E. coli population was about 107 or 10 million genetically distinct organisms, a number relatively small in comparison to the expected total population of the organism, which has been estimated at 1020 or 100,000,000,000,000,000,000 cells. The effective population was sufficiently low to assure that random processes of the birth and death of strains were the dominant evolutionary force.
If there is little recombination, how does one explain the high genetic diversity we observe? Periodic selection could be one answer; a genotype might selectively displace others present in the population. In an asexual population, once a favored mutation spreads by natural selection, it replaces not just the gene involved, but a complete genotype. Such a story might make sense as a mechanism of adaptation to particular niches. In other words, E. coli would be, according to these ideas, a collection of very different strains, each adapted to a different environment.
» Post Comment