Disarming AIDS

Elledge, who is also a Howard Hughes Medical Institute investigator at Brigham & Women's Hospital, may be onto a better strategy–using the virus' weakness against it.

That weakness is its need to use human proteins to multiply itself. Taking those tools away from HIV could turn out to be much harder for the virus to circumvent.

"HIV is a very small virus; it has only nine genes and encodes only 15 proteins," Elledge points out. "That's not enough to allow it to multiply, so it has to take advantage and hijack the human proteins and make them do its bidding."

Now in research published in the journal Science, Elledge, lead author

Abe Brass and their colleagues have identified those human proteins.

They did it using a new genetics technique called RNA interference to block individual proteins one at a time in human cells in the lab. RNA interference stops a protein from being made by a cell by turning off the gene that encodes that protein.

"The beauty of the RNA interference method…is that we can now, at a one-by-one case, basically turn off each gene in a mammalian cell and ask, 'is that gene important for HIV's ability to multiply?'" Elledge explains. "We can ask, 'which ones, when you take them away, prevent HIV from multiplying?'"

"So we did this for 21,000 different genes, and in that way we cover all of the possible genes and all of the possible–at least, known–proteins in human cells, one at a time," he says.

It turned out that most of them had no effect on HIV's ability to multiply. "But a very small set, 273, end up being absolutely essential for its ability to duplicate itself and spread," says Elledge.

That is a small set compared to humans, but it greatly expands the number of potential drug targets to stop the virus.

"One way to look at it is that the HIV virus makes only 15 proteins of its own, and a subset of those are potentially good drug targets," he says. "We now have 15 times as many possible targets, and so I think there's hope out there that there will be some targets that are going to be even better than the drugs that are currently used."

While blocking any one of the 273 human proteins should be able to stop the virus, Elledge says some of them might not make good targets because people need them, too.

"The good side is that if you have a drug that interacts with a human protein and inactivates it somehow, the virus is much less likely to be able to overcome that. It's not as simple as changing the binding site on one of its own proteins. It would have to evolve a whole new capability. That’s unlikely to happen," he says.

"So that's the upside. The downside of using human proteins is that, if we need them, it might make us sick if you inhibit them. That’s a real concern, but I would point out that all drugs that we take, besides antibiotics, target human proteins. And so it certainly is possible to interfere with some functions in human cells without having a very bad effect on people."

Elledge also notes that it typically can take a decade for new drugs to come out of new discoveries like this.

"It's nothing that’s right around the corner, it's not an immediate cure, but it gives us hope for a whole new class of possibilities," says Elledge.

And, importantly, the new method can be used to test any protein for whether a disease needs it to cause harm.

"I think that these methods are really revolutionary," Elledge says. "They allow us to ask questions that we could not ask before, because now we can turn off every gene and say, 'are you needed or not?' for whatever process we're interested in… I think that the methods that we've employed here will be applied to lots of other human pathogens and viruses."

In fact, Elledge's lab is now working on finding out what proteins cancer needs in order to proliferate.

This research was published in Science Express, January 10. 2008, and funded by: National Institutes of Health, Harvard Center for AIDS Research, Crohns and Colitis Foundation of America, Center for Computational and Integrative Biology, Howard Hughes Medical Institute.