Mutators and long-term molecular evolution of pathogenic Escherichia coli O157:H7.

It has been proposed that an increased mutation rate (indicated by the frequency of hypermutable isolates) has facilitated the emergence of Escherichia coli O157:H7. Analysis of the divergence of 12 genes shows no evidence that the pathogen has undergone an unusually high rate of mutation and molecular evolution.

It has been proposed that an increased mutation rate (indicated by the frequency of hypermutable isolates) has facilitated the emergence of Escherichia coli O157:H7. Analysis of the divergence of 12 genes shows no evidence that the pathogen has undergone an unusually high rate of mutation and molecular evolution.

Figure.
Evolutionary distance in terms of synonymous and nonsynonymous changes per 100 sites (4) for 12 genes sequenced from Escherichia coli O157:H7, E. coli K-12, and Salmonella enterica Typhimurium. The points for synonymous sites are (left to right): gap, crr, mdh, icd, fliC (conserved 5' and 3' ends), trpB, putP, aceK, mutS, trpC, tonB, and trpA. Under the mutator hypothesis, the genetic distance between the pathogenic O157:H7 strain (or the closely related strain ECOR37) and the outgroup (Typhimurium) is expected to exceed the distance between the commensal K-12 and the outgroup. Prolonged periods of enhanced mutation rate should drive the points above the dotted line marking equal rates of molecular evolution. Two loci (tonB and trpA) show departure from the equal rate line, but neither has evolved differently from that expected by the molecular clock. The sequences of 12 genes were obtained from GenBank or the original sources as follows: aceK (5), crr (6-8), fliC (9), gap (10), icd (11), mdh (12), mutS (13,14), putP (15), tonB (16,17), trp (17)(18)(19)(20). mutations per 100 sites) between Typhimurium and K-12 is shown against the distance between Typhimurium and O157:H7 for synonymous and nonsynonymous sites separately ( Figure). The line indicates equal rates of evolution in the two Dispatches lineages. An elevated mutation rate in O157:H7 over evolutionary time should result in greater divergence from Typhimurium than from K-12 and in the distribution of points above the equalrate line. For both synonymous and nonsynonymous sites, most genes fall below or very near the equal-rate line with only two exceptions: tonB and trpA deviate in the direction expected under the mutator hypothesis. To test the significance of these deviations, we compared the observed degrees of divergence of K-12 and O157:H7 from Typhimurium and the expectations of the molecular evolutionary clock hypothesis (21).
The basis of this test is that a constant rate of mutation results in equal numbers of substitutions in two sequences from an outgroup (21). Considering synonymous and nonsynonymous changes together with Typhimurium as an outgroup, we found that 11 of the 12 loci, including tonB (m 1 = 0, m 2 = 3, X 2 = 3.00, p > 0.05) and trpA (m 1 = 4, m 2 = 10, X 2 = 2.57, p > 0.05), did not deviate significantly from a uniform rate of evolution predicted by a molecular clock. Only mdh exhibited a significant departure from the molecular clock (m 1 = 14, m 2 = 5, X 2 = 4.26, p < 0.05); however, the direction was away from that predicted by the mutator hypothesis (the TyphimuriumK-12 distance exceeded the Typhimurium-O157:H7 distance). Our findings do not conflict with the observation that MMR defects occur in relatively high frequency in emerging pathogens; however, the findings indicate no evidence of a genomewide elevation of the mutation rate in pathogenic E. coli O157:H7. The uniform rate of divergence of O157:H7 and K-12 suggests several possibilities. One is that the mutator state is transient and so brief that the impact on long-term rates of evolution is undetectable. This possibility is consistent with the view that mutators may generate favorable mutations in periods of intense selection and then revert to a nonmutator phenotype (22,23). Another possibility is that all bacterial populations experience brief episodes of adaptive evolution driven by hypermutation. Matic and co-workers (24) found equivalent frequencies of mutators among strains of commensal bacteria and both emerging and classical pathogenic E. coli.
Finally, defects in MMR that produce the mutator phenotype also relax the normal barriers to recombinational exchange between bacterial species (25). The enhanced recombina-tion that accompanies the mutator phenotype may explain why E. coli O55:H7, the immediate ancestor of O157:H7 (26) that also carries the same defective MMR allele (3), harbors such an extraordinary variety of plasmid and chromosomal virulence factors (27). Together with our finding of clock-like divergence of E. coli O157:H7 housekeeping genes, these observations indicate that the main evolutionary benefit of the mutator phenotype is the enhanced ability to acquire useful foreign DNA (3), not an increased rate of point mutation over the long term.
Dr. Whittam is professor of biology at the Institute of Molecular Evolutionary Genetics, Pennsylvania State University. His research focuses on understanding how evolutionary forces operate to determine the amount and organization of genetic variation in natural populations of bacteria. Specific studies include the evolution of pathogenic forms of Escherichia coli associated with intestinal and extraintestinal infections, the evolution of virulence and resistance in a host-parasite interaction using an amoeba-Legionella system, and the ecologic determinants and evolution of host specificity in Rhizobium-legume associations.