"Vertex thinks of antibiotics as the eternal drug category," says Joshua Boger, chief executive of Cambridge, Mass.-based Vertex Pharmaceuticals (nasdaq: VRTX - news - people ). "After all of mankind's other diseases have been cured or treated with breakthrough drugs of the future, we will still need new antibiotics because bacteria are so adept at evading and escaping control."

In 1977, Michael Jacobs, now the director of clinical microbiology at Case Western Reserve University, documented the first cases of resistance of Streptococcus pneumoniae, a common cause of pneumonia, ear and sinus infections, to several classes of antibiotics. Twenty-six years later, S. pneumoniae can resist a whole host of drugs. Amoxicillin, an antibiotic developed by GlaxoSmithKline (nyse: GSK - news - people ) but available generically, is still the first choice for treating it. But Jacobs notes that in a quarter of a century, the standard dosage of amoxicillin has increased fourfold as bacteria have developed ways to resist it.

Doctors give antibiotics 50% more often than they should, allowing bacteria to mutate even faster. A less toxic, more convenient drug can actually make the problem worse. Jacobs has been a vocal opponent of the use of Pfizer's (nyse: PFE - news - people ) Zithromax, an antibiotic that can be given over fewer days--or even in a single big, convenient dose. "It's a very safe drug," says Jacobs. "It doesn't have many side effects. If you have a patient who doesn't need antibiotics, that's a good drug to prescribe. The problem is, those patients don't need antibiotics in the first place."

Some studies have shown that as much as 31% of S. pneumoniae is now resistant to Zithromax. Since bacteria are persistent, it's good news that companies like Vertex are developing new antibacterial drugs in the lab. It means that even after a decade or so of development, there will still be a market for new antibiotics.

 

	At Vertex Pharmaceuticals, Paul Charifson is pursuing a new way to knock out gyrase, an enzyme bacteria.

Vertex decided to get into the antibacterial field four years ago, says Paul S. Charifson, a principal investigator at Vertex who heads a group of chemists designing new antibiotic drug candidates. Resistance rates were rising, he notes, and there didn't seem to be much competition.

"Big pharma seemed to be less interested in doing antibiotic drug development," says Charifson. Within a year, he had settled on a promising strategy. He would look for molecules that would knock out gyrase, an enzyme bacteria used in DNA replication.

The gyrase molecule is the target of fluoroquinolones, a best-selling class of antibiotics that includes Johnson & Johnson's (nyse: JNJ - news - people ) Levaquin and Bayer's (nyse: BAY - news - people ) Cipro. Gyrase is built a bit like a flashlight--one end does work essential to replicating bacterial DNA, while the other harnesses energy like an intracellular battery pack. Levaquin, Cipro and their peers knock out the light bulb, but Charifson had another idea--why not disconnect the batteries? He set about looking for molecules that would cut off gyrase's energy supply.

He had a stroke of luck when he found that another molecule similar to gyrase could be inhibited by the same molecule, making his drug candidates more toxic to bacteria. An added bonus--inhibiting both of these molecules makes it less likely that bacteria would be able to figure out a way to evade the new drug.

Aventis' (nyse: AVE - news - people ) Ketek, a new drug that has not yet been approved in the U.S., decreases resistance by hitting two different spots on a single enzyme. "They're expanding the contact zone in the same site," says Charifson. "We're targeting a completely different enzyme."

Charifson's strategy would knock out two completely different molecules, both of which the bacteria need. That could be like giving two drugs at once. In the lab, bacteria are having trouble coming up with ways to resist it. Now begins the long process of trying out versions of Charifson's molecules in animals and people.