A promising new antibiotic is generating both excitement and despondency. It is the first new chemical class of antibiotic to be found in more than two decades, but experts fear that the hurdles to turning the compound into an effective commercial drug could mean that it ends up collecting dust on a shelf.

The drug, called platensimycin and reported on page 358, kills several of the major drug-resistant bacteria that plague hospitals. Among them are methicillin-resistant Staphyloccocus aureus (MRSA) and bacteria resistant to vancomycin, one of the last lines of antibiotic defence.

Only two other new chemical classes of antibiotic have been discovered and approved for use since the early 1960s: daptomycin and linezolid. Most antibiotics used today were discovered in the 1940s and 1950s, and newer versions have mainly been made by chemically nipping and tucking these compounds. They generally kill bacteria by blocking the production of proteins, DNA or the bacterial cell wall.

An antibiotic has been discovered that kills the deadly methicillin-resistant Staphylococcus aureus. Credit: K. LOUNATMAA/SPL

Platensimycin has a novel chemical structure and works differently from other commercially available antibiotics by crippling FabF, a bacterial enzyme involved in manufacturing fatty acids. It thus stops bacteria from making the fatty cell membranes they need to grow. Two commercially available antibiotics, triclosan and isoniazid, target another enzyme involved in fatty-acid synthesis, but these do not kill the same array of bacteria.

The researchers at Merck Research Laboratories in Rahway, New Jersey, who discovered platensimycin, did so by reviving and tuning an established strategy. They searched for naturally existing compounds that microbes typically make to kill neighbouring bacteria. This technique originally yielded many potent antibiotics, but was abandoned by many drug companies as the rate of return diminished and has been replaced by chemical methods to generate novel synthetic molecules.

Traditionally, researchers have tested whether a tiny amount of chemical can kill off a circle of bacteria growing on a plate. But by first making the bacteria more vulnerable, the Merck researchers searched for natural products that might have been missed in such conventional assays. They engineered S. aureus bacteria to make less of the FabF enzyme than normal, using a snippet of RNA to block the production of this protein (K. Young et al. Antimicrob. Agents Chemother. 50, 519–526; 2006).

The researchers then tested whether any of 250,000 natural-product extracts could block the growth of the disabled strain — and pulled out platensimycin, a small molecule made by a strain of Streptomyces platensis bacteria in a South African soil sample. “It allowed us to find a needle in a haystack, and something we might have missed with another type of screen,” says Stephen Soisson, one of the lead researchers.

Other scientists are delighted with the new compound, and say it could provoke renewed interest in natural compounds and in attacking the fatty-acid synthesis pathway. “From a scientific point of view it is what you want — a new class against a new target,” says Steven Projan of Wyeth Research in Pearl River, New York.

But “the next steps are fraught with danger”, warns microbiologist Carl Nathan of Weill Medical College of Cornell University in New York. “The obstacles are truly formidable.”

Platensimycin could stumble at one of many scientific, regulatory or financial hurdles. One concern is that it may be unstable in the body, because the research team had to infuse it continuously into mice to rid them of a S. aureus infection. The chemical might need extensive modification to make it more stable, and could prove useless if it has toxic side effects.

Platensimycin's discovery highlights the fact that compounds are still waiting to be found.

Testing antibiotics in clinical trials is also tough and expensive. This is partly because the US Food and Drug Administration (FDA) does not have clear guidelines for approving new antibiotics, says Frank Tally, chief scientific officer of Cubist Pharmaceuticals in Lexington, Massachusetts, which developed daptomycin. Even if platensimycin is approved for human use, it will probably meet the fate of its predecessors, as bacteria rapidly acquire resistance to it. “It might make it to the clinic and then fade because of drug resistance,” Nathan says.

With such a daunting task ahead, some microbiologists fear there is little incentive for Merck to develop the drug. Not only is the process hugely expensive, but the potential market for a new antibiotic — which is often used sparingly and administered for only a week or two — is small. This has prompted many pharmaceutical giants to cut back their antibiotic research programmes in recent years.

The Merck researchers refuse to spell out their plans for platensimycin, although they hint that they are working on it. “We are still in the business of doing antibiotic research, and we wouldn't be if we didn't want to develop our own compounds,” says Sheo Singh, another of Merck's lead researchers.

Several solutions have been proposed to keep new drugs flowing into the clinic as micro-organisms overcome old ones. The Infectious Diseases Society of America proposed an array of regulatory measures in 2004 that might entice drug companies to develop antibiotics, such as revised FDA standards for the approval of antibiotics and tax breaks for work on such drugs (see Nature 431, 892–893; 2004). Some researchers are turning to alternative measures to fight life-threatening bacteria, such as microbe-killing bacteriophage and vaccines.

But many workers still think that novel antibiotics remain the best tried and tested way to combat microbes. They say platensimycin's discovery highlights the fact that compounds are still waiting to be found. “With a paper like this, people might start to reinvest in this area,” Tally says, as it shows that new molecules can be found in well-known sources.