Published online 4 April 2007 | Nature | doi:10.1038/news070402-6


Weak drug combos find new use

Antibiotics that don't work could beat back resistant bacteria.

Mix and match: some drugs are stronger together, some weaker.Mix and match: some drugs are stronger together, some weaker.Corbis

You would think that a combination of antibiotics that is less effective than either drug on its own would be fairly useless. But researchers now say that such ineffectual mixes could be used in the campaign against resistance to the drugs.

The counterintuitive conclusion comes from the observation — so far seen only in the lab — that less-effective drug mixes allow bacteria that are sensitive to drugs to out-compete those that are resistant them. Exploiting the trade-off between an antibiotic's ability to kill bacteria and the pressure it exerts on them to become resistant might help to overcome the problem of resistance, say Roy Kishony of Harvard Medical School in Boston and his colleagues.

Because very few new antibiotics are being discovered, looking at how the existing ones work in combination could be a fruitful approach, says microbiologist Bruce Levin of Emory University in Atlanta, Georgia.

"We are losing the arms race, and something that at least delays the loss would be useful," he adds. "If we are just counting on new drugs, we could be screwed."

Together we're weaker

Kishony's team looked at the effects of doxycycline — an antibiotic used to treat infections such as plague and syphilis — on the bacterium Escherichia coli. They tested the antibiotic on bacteria sensitive to the drug, and on those engineered to be resistant by pumping the antibiotic out.

One set of trials combined doxycycline with erythromycin. This combination is synergistic — together, the drugs' effect is greater than the sum of their individual effects. Not surprisingly, resistant bacteria grew much better than the sensitive strain under this treatment.

Other trials used doxycycline and ciprofloxacin. This combination of drugs is 'suppressive' — the two together are weaker than either used individually, although no one knows why. In this instance both strains of bacteria grew more quickly than they would when attacked with either drug alone.

But, the researchers found, this growth varied depending on how the drugs were mixed. With a low proportion of doxycycline, the sensitive bacteria out-grew the resistant ones. In this case, natural selection should work against antibiotic resistance, they reasoned.

The team confirmed their theory by setting up competitions between the two types of bacteria in different drug environments. They found they could get sensitive strains to out-compete the resistant ones (almost, but not quite completely, wiping out the hardy strains) with a range of drug combinations. The scheme also worked against the bacterium Staphylococcus aureus, a widespread and often lethal source of drug-resistant infections, they report in Nature1.

Double whammy

The results hint that a patient with an antibiotic-resistant infection could benefit from being treated first with a poor mix of drugs, so that the sensitive bugs flourish and the resistant ones die out through competition. A second bout of effective drugs could then clear away the remaining infection, says Edward Feil, who studies bacterial evolution at the University of Bath, UK.


Hitting bacteria with a one-two punch might be particularly useful in small, localized outbreaks of resistance, he says, such as in individual patients or hospital wards. But, he adds, finding the right combination might be hard, and it might not work quickly enough to benefit the infected patient.

And resistant bacteria might find a way round this effect, as they have for so many other antibiotic assaults, the team adds.

Antibiotics are not going to become useless, says Feil; there will always be some bacteria that are susceptible to them. But we are going to have to live with resistant bugs too, he adds. "There is no obvious way we will get rid of them in the foreseeable future."

Visit our drugcombosfindnew_use.html">newsblog to read and post comments about this story.  

  • References

    1. Chait R., Craney A. & Kishony R.. Nature, 446 . 668 - 671 (2007). | Article |