The quest for an HIV vaccine has been given a bad prognosis recently, due to increasing agreement that the human immune system isn’t clever enough to outsmart the ever-changing surface of the virus. But now a new approach promises to solve the problem by sidestepping the immune system altogether, instead using gene therapy to produce immune molecules that neutralize the virus.

Reporting in Nature Medicine this week, Philip Johnson, a professor of pediatrics at the University of Pennsylvania, in Philadelphia, and his colleagues managed to protect monkeys from infection with the simian immunodeficiency virus (SIV), the animal model that is closest to HIV, by shuttling a gene into their muscles that produces antibody-like molecules that work against SIV.

With both SIV and HIV, the chameleon-like mutability of the virus changes its surface so quickly that most antibodies made by the immune system are soon rendered ineffective, explains Johnson. “I concluded about 10 years ago that traditional approaches weren’t going to work,” he says. “Successful vaccines all mimic the body’s natural immune response and this just does not work well in HIV-infected people. HIV is too good at hiding itself from the immune system.”

Some antibodies have been shown to neutralize SIV, although exactly how they do this is not well understood. Getting the human immune system produce the few known antibodies that show similar potential for neutralizing HIV has also been impossible.

To tackle this problem, Johnson turned to another virus for help. A modified version of the adeno-associated virus (AAV) has been proved to work as a convenient vector for delivering gene snippets safely into the human body for gene therapy, and it has been used successfully to treat hemophilia and congenital blindness by supplying patients with genes that are otherwise missing.

Johnson and his colleagues designed DNA sequences for particularly stable versions of two antibodies known to be effective against SIV. They used antibody-like molecules, called immunoadhesins, in which the functional part of an antibody is fused with a more stable section of another antibody.

The team engineered the immunoadhesin sequences into the AAV vector and injected the constructs into the muscles of nine rhesus macaques, where the muscle cells then started to produce the chimeric antibodies and secrete them into the bloodstream. After four weeks, the team infected the monkeys with SIV and monitored their health along with antibody and virus levels in their blood over one year. Six of the nine macaques showed no sign of SIV infection, and the remaining three did not develop AIDS during the course of the study. In contrast, six control monkeys all became infected, and four of them died before the experiment finished.

The results are “really encouraging,” says Andrew Sewell of Cardiff University’s School of Medicine in the U.K., who was not involved in the study. “I have not seen anything that worked this well before.”

“This has given us a really big green light in the monkey model, but of course we still need to show this also works in humans,” says Johnson. There are four potent antibodies that work against HIV, and it may be possible to use the AAV vector to deliver the necessary genetic sequences for the antibodies in humans. Johnson hopes to get permission to start clinical trials soon and is optimistic that the technique can be adapted to humans. If successful, he says, it could turn into an affordable way to protect against HIV.

“The manufacturing technology available today makes it possible to do this at a very accessible cost,” Johnson says. “Having said that, at this point, the field is looking for anything that works, no matter what the cost; we just need something that works, full stop.”

“The data from this study show the promise of vector-mediated gene transfer as an approach for immunization,” says Pat Fast, chief medical officer at the International AIDS Vaccine Initiative. “We plan to collaborate with Doctor Johnson to expand on the monkey studies and to support the optimization of the vector for eventual testing in humans.”

Sewell shares this optimism: “If you think of HIV as an Achilles that can only be attacked by its heel, this is a designer weapon that can go straight to the heel. The fact that it works so well in macaques is very promising.”

Johnson and Sewell both caution, however, that it will be important to find more antibodies to keep HIV in check. “We may want to put in two or more effective antibodies at the same time to put additional pressure on the virus, so the more good antibodies we can find, the better,” Johnson says.