Low levels of antibiotics cause multidrug resistance in 'superbugs'

In the Feb. 12 issue of Molecular Cell, research led by Boston University Professor James J. Collins details for the first time the biomolecular process that produces superbugs. When administered in lethal levels, antibiotics trigger a fatal chain reaction within the bacteria that shreds the cell's DNA. But, when the level of antibiotic is less than lethal the same reaction causes DNA mutations that are not only survivable, but actually protect the bacteria from numerous antibiotics beyond the one it was exposed to.

"In effect, what doesn't kill them makes them stronger," said Collins, who is also a Howard Hughes Medical Institute investigator. "These findings drive home the need for tighter regulations on the use of antibiotics, especially in agriculture; for doctors to be more disciplined in their prescription of antibiotics; and for patients to be more disciplined in following their prescriptions."

Two years ago, Collins -- together with graduate student Michael Kohanski and post-doctoral fellow Mark DePristo -- proved that when applied in lethal doses, antibiotics stimulate the production of reactive oxygen species (ROS) molecules, or free radicals that damage DNA, protein and lipids in bacterial cells, contributing to their demise. In the new study, the same co-authors demonstrated that the free radicals produced by a sub-lethal dose of an antibiotic accelerate mutations that protect against a variety of antibiotics other than the administered drug.

"We know free radicals damage DNA, and when that happens, DNA repair systems get called into play that are known to introduce mistakes, or mutations," said Collins. "We arrived at the hypothesis that sub-lethal levels of antibiotics could bump up the mutation rate via the production of free radicals, and lead to the dramatic emergence of multi-drug resistance."

Testing their hypothesis on strains of E. coli and Staphylococcus, the researchers administered sub-lethal levels of five kinds of antibiotics and showed that each boosted levels of ROS and mutations in the bacterial DNA. They next conducted a series of experiments to show that bacteria initially subjected to a sub-lethal dose of one of the antibiotics exhibited cross-resistance to a number of the other antibiotics. Finally, they sequenced the genes known to cause resistance to each antibiotic and pinpointed the mutations that protected the bacteria. Ironically, the researchers discovered that in some cases the bacteria were still be susceptible to the original antibiotic.

"The sub-lethal levels dramatically drove up the mutation levels, and produced a wide array of mutations," Collins observed. "Because you're not killing with the antibiotics, you're allowing many different types of mutants to survive. We discovered that in this zoo of mutants, you can actually have a mutant that could be killed by the antibiotic that produced the mutation but, as a result of its mutation, be resistant to other antibiotics."

The group's findings underscore the potentially serious consequences to public health of administering antibiotics in low or incomplete doses. This is common practice among farmers who apply low levels of antibiotics to livestock feed; doctors who prescribe low levels of antibiotics as placebos for people with viral infections; and patients who don't follow the full course of antibiotic treatment.

The study's findings may ultimately lead to the development of new antibiotic treatments enhanced with compounds designed to prevent the emergence of multi-drug resistance. For example, one potential treatment might inhibit the DNA damage repair systems that lead to the problematic mutations, while another might boost production of cell-destroying free radicals so that a low dose of antibiotic is sufficient to kill targeted bacterial cells.