For possible AIDS drug, smaller is better.

Pull the trigger on a revolver and the hammer draws back for a split second before slamming forward. Imagine stopping a deadly gunshot by quickly inserting an impediment between the cocked hammer and the bullet.

AIDS researchers are now developing drugs that perform a comparable feat, halting HIV midway through its attempt to fuse with and infect a cell. One such drug, T-20, has already entered clinical trials. However, because it's a large, fragile molecule that's not readily absorbed in the digestive system, T-20 must be injected.

Investigators have now taken a key step toward creating drugs that act like T-20 but are small and hardy enough to be taken orally. They've designed bits of protein, or peptides, that bind to an indentation on a viral-surface protein, gp41, that HIV uses to enter cells.

Both T-20 and the newly discovered peptides interfere with the actions of gp41, but they attach to different regions of the protein. The pocket targeted by the peptides normally stays hidden. When the AIDS virus bumps into a cell, gp41 changes shape and exposes the pocket as it helps the virus infect the cell.

"When we saw the crystal structure of gp41 several years ago, the pocket was something that jumped out at us" as a drug target, notes Peter S. Kim of the Howard Hughes Medical Institute at the Whitehead Institute for Biomedical Research in Cambridge, Mass.

In the Oct. 1 CELL, Kim and his colleagues describe peptides--some as small as one-fourth the size of T-20--that fit snugly into the gp41 indentation. In test-tube experiments, the peptides stop HIV from infecting cells, the group reports.

"It's a very nice example of the basic science of structural biology getting translated into a potential drug," says Anthony S. Fauci, director of the National Institute of Allergy and Infectious Diseases in Bethesda, Md.

To find HIV-blocking peptides, Kim group created an artificial protein with an exposed gp41 pocket. The scientists screened a large library of peptides for ones that bound to the pocket. Using D amino acids, mirror-image replicas of natural amino acids, the researchers then made synthetic versions of the selected peptides. These forms are resistant to degradation, making them ideal for pills that must survive a bath in the digestive enzymes of a stomach.

Pills that block the ability of HIV to infect cells should complement current AIDS drugs, which target enzymes used by the virus once it's already inside a cell, says Fauci. Such new pills should make it more difficult for HIV to evolve drug resistance, he adds.

The test-tube experiments, however, suggest that Kim's peptides aren't as effective as T-20 at stopping the AIDS virus. Still, now that scientists have shown that peptides binding the gp41 pocket can thwart HIV, drug developers will rush to craft molecules that are even smaller and more potent than the ones his group identified, predicts Kim.

Moreover, his team plans to make the artificial-pocket protein freely available so that pharmaceutical and biotech companies can screen their stocks for other drugs that might stop HIV from shooting its way into cells.