NEW YORK (GenomeWeb News) – In a study in Nature Genetics online yesterday, Harvard University researchers described how they used genome sequencing to characterize the genetic changes contributing to antibiotic resistance in Escherichia coli populations independently exposed to three different drugs.
Using continuous culture systems that allowed them to grow the bacteria while slowly ramping up the antibiotic concentrations, the investigators generated more than a dozen E. coli populations with enhanced resistance to chloramphenicol, doxycycline, or trimethoprim. By sequencing the genomes of antibiotic-resistant clones from each population, they then identified genes and pathways containing mutations in bacteria that had acquired resistance to each of the three drugs.
Results of the study suggest that the sorts of genes that are mutated as E. coli becomes resistant to antibiotics depends, in part, on the compound involved. For instance, the team found that response to one of the drugs always happened in a stepwise fashion and typically involved changes to the same gene. On the other hand, resistance to the other two drugs arose in a more continuous manner and included mutations to a more widespread and variable set of genes.
"Whole-genome sequencing of the evolved strains identified mutations both specific to resistance to a particular drug and shared in resistance to multiple drugs," corresponding author Roy Kishony, a systems biology researcher at Harvard Medical School, and colleagues wrote.
To explore the genetic basis of drug resistance in E. coli, the team grew bacterial populations in the presence of one of three antibiotics — chloramphenicol, doxycycline, or trimethoprim — in a morbidostat, a continuous culture system that allowed them to adjust antibiotic concentrations to keep the bacterial populations growing while still maintaining selective pressure on the bugs.
Using this approach, researchers established 15 antibiotic-resistant E. coli populations, five parallel populations for each of the antibiotics of interest.
Two of the three drugs used for the study — chloramphenicol and doxycycline — interfere with bacterial protein synthesis by hindering ribosomal function, while a third, trimethoprim, inhibits the biosynthesis of the B vitamin folic acid.
Based on the growth rates that the researchers observed for the E. coli populations, they concluded that resistance to the ribosomal-inhibiting compounds chloramphenicol and doxycycline evolved in a continuous manner. On the other hand, populations appeared to acquire trimethoprim resistance in a series of steps.
Moreover, bacteria that had become resistant to either chloramphenicol or doxycycline also had resistance to the other drug, though the team did not see overlap between resistance to those two drugs and trimethoprim-resistance.
When they sequenced the genomes of E. coli clones from drug-resistant populations generated over 20 days or so using the Illumina GAIIx, the team tracked down 47 SNPs that could be verified by targeted Sanger sequencing.
In the chloramphenicol and doxycycline-resistant E. coli, these included a mishmash of changes affecting various combinations of genes, many of them transcription, translation, and transport-related. Even so, researchers did not find mutations within ribosomal genes themselves.
Together, the team's findings hinted that there are "multiple alternative ways to circumvent chloramphenicol- and doxycycline-induced protein synthesis stress, with each requiring a small number of mutations in a diverse set of genes."
"The waiting time for mutations to appear in these populations is likely to be short due to the large target size for possible mutations," the authors added, "consistent with the smooth phenotypic changes in these populations."
A far different pattern was found in the trimethoprim-resistant bacteria. For that drug, researchers detected most resistance mutations within just one E. coli gene: the dihydrofolate reductase enzyme-coding gene DHFR that trimethoprim normally binds and inhibits.
Mutations in DHFR were reproducible in different populations of E. coli, they reported, and sequencing experiments done over the course of the study suggest mutations in the gene crop up in a more or less defined order. Consequently, the study authors concluded that "resistance to trimethoprim proceeds through the sequential fixation of mutations in a target enzyme through ordered pathways."
"Our study is one of the first to directly show ordered adaptive pathways leading to high levels of antibiotic resistance in bacteria, complementing previous observations of the parallel evolution of virus populations," they wrote, noting that additional work is needed to "identify additional paths to resistance and determine how such paths depend on the environment, population size, and strength of selection pressure."