Combining Antibiotics May Backfire

Fatal weakness. Pathogens such as methicillin-resistant Staphylococcus aureus (above) develop vulnerabilities as they evolve resistance to antibiotics. Choosing antibiotics that exploit these weaknesses can thwart resist

Pasieka/Science Photo Library

You might think that combining two antibiotics would be a great strategy to take down a nasty disease fast. Think again. A new study suggests that such a two-pronged attack can backfire badly by giving super-resistant bacteria the opportunity they need to come out on top in the struggle for resources.

One way to fight a particularly stubborn infection is to prescribe two drugs at once that attack it in alternate ways—for example, two antibiotics can disrupt two different parts of the bacteria's protein-building machinery. Drugs that cooperate this way are called "synergistic," and doctors use such powerful multipronged attacks to battle HIV, tuberculosis, and even cancer. The strategy is thought not only to kill pathogens more effectively, but also to delay the emergence of resistance.

Some doctors also prescribe paired-up antibiotics to fight nasty infections such as the notoriously resilient staph infection methicillin-resistant Staphylococcus aureus (MRSA). But while laboratory experiments have shown that some antibiotics work well together in the short term, little is known about how well the combos perform over many days.

A team of evolutionary biologists, including Robert Beardmore of the University of Exeter in the United Kingdom, put long-term antibiotic synergy to the test. The researchers chose two antibiotics known to cooperate well in their first 24 hours, doxycycline and erythromycin, which are commonly given separately for infections such as Escherichia coli. The researchers put different amounts of each antibiotic—alone or combined—into tubes full of E. coli, which is easier to study than bacteria such as MRSA. They then tracked how well the microbes were able to multiply by shining a laser into the tubes and measuring how much the bacterial cells obscured the light.

Within 1 day, the researchers saw a huge dive in bacterial growth, up to 95%, in the bacteria subjected to the synergistic attack, just as they anticipated. But the following day, the same E. coli had experienced a population explosion—for example, bacteria that had dropped by 90% after half a day of drug exposure later skyrocketed by 500% after a day and a half. Beardmore recalls that "just looked wrong." Convinced they had accidentally contaminated the experiment, the researchers started over. But the bacterial explosion happened again the second time. Different strains of E. coli thrived under the combined regimen as well. "We kept seeing this idea that where you have the biggest attack you see the biggest number of bacteria," Beardmore says.

Searching for an explanation, the researchers sequenced the resistant E. coli's genomes, which contain just 5000 genes. The bacteria had widely duplicated genes that provided four different ways to withstand the drugs, the researchers report today in PLOS Biology. These included "efflux pumps" that thrust antibiotics and other intruding chemicals back out of the bacteria.

The team then used the genetic information to put together a mathematical simulation of the behavior of the cells and their genes to see if its results would match what the experiments showed. It did—soaking the bacteria in the synergistic drugs consistently created more bacteria equipped with pumps. The scientists deduced that the powerful attack had given bacteria that were exceptionally equipped to handle both antibiotics at once the upper hand by removing almost all of the competition. This gave them easy access to resources and the chance to reproduce like crazy, so much so that the overall number of E. coli became higher with increasingly stronger attacks.

Beardmore hopes his team's research will provide a warning that two-pronged attacks don't always work, which may help doctors keep antibiotic resistance at bay. "You want to slow down evolution," he says. "[Bacterial evolution] can happen fast and it can kill you." His team is hoping to come up with better antibiotic strategies against microbes such as MRSA and E. coli. One possibility might be, instead of taking both antibiotics at once, to alternate them each day, which could keep multiresistant bacteria from becoming dominant. "On day one, you take the blue pill," he says. "On day two, you take the red pill."

The study is one of a growing number that questions whether the hardest-hitting treatment strategies are always the best, says mathematical biologist Troy Day of Queen's University in Canada, who was not involved in the work. "In the case of HIV, combinations of drugs do seem to work quite well," Day says. "You could ask, 'Why aren't we using those kinds of cocktail type therapies for bacterial infections? Would they not work just like they seem to work for HIV?' And apparently, they don't always."

Follow News from Science