The difference between HIV infection and full-blown AIDS is, in large part, the massive die-off of the immune system’s CD4 T-cells. But researchers have only observed the virus killing a small portion of those cells, leading to a longstanding question: What makes the other cells disappear? New research shows that the body is killing its own cells in a little-known process. What’s more, an existing, safe drug could interrupt that self-destruction, thereby offering a way to treat AIDS.

The destructive process has caught scientists by surprise. “We thought HIV infects a cell, sets up a virus production factory and then the cell dies as a consequence of being overwhelmed by virus. But there are not enough factories to explain the massive losses,” says Warner Greene, director of virology and immunology at the Gladstone Institutes, whose team published two papers today in Science and Nature describing the work. (Scientific American is part of Nature Publishing Group.)

Greene suspects that researchers in the past failed to notice the process because they were looking in the wrong place. Instead of studying active CD4 T cells in the blood, his team examined spleen and tonsil tissue, where most cells are in a resting state. When HIV enters a resting cell, it transcribes its genes into DNA, but then hits a dead end: the cell’s machinery isn’t available to finish the replication process. This part of the story had been known for years, but Greene’s team discovered that something very surprising happens next. “Instead of that being the end of the story, cells detect that DNA in their cytoplasm and launch an immune response against it, and that immune response results in the death of those cells.”

The response is a self-destruct protocol called pyroptosis. In contrast to the better-known apoptosis, in which cells die quietly without triggering inflammation, pyroptosis is “not a bland, but a fiery death,” Greene says. These cells spew inflammation-causing chemicals as they die, attracting more T-cells that can then become infected themselves by the newly freed HIV. “In a bacterial infection, recruiting all these cells might be a good strategy for containing the infection,” Greene says, but with HIV a vicious cycle of infection results. Pyroptosis also explains why AIDS is associated with high levels of inflammation.

Experiments by Greene’s team showed that blocking a key component of pyroptosis could stop the cell death entirely; they also identified the protein that senses viral DNA to kick off the process. After studying the pathway in cultured spleen and tonsil tissue, they had an opportunity to confirm the findings in a patient’s freshly removed lymph node, stained for their target proteins and viewed under the microscope. It showed the traditional virus replication happening in cells at the center of the node and resting cells dying all around. To Greene, the sight was unbelievable: “We could see this pyroptotic pathway playing out like nobody’s business. In this one snapshot, we could see what we had been working on for eight years.”

There’s good news though: Greene estimates 95 percent of the cells that die in HIV infections are killed through pyroptosis, so the findings raise hope for a new type of treatment that could prevent HIV from progressing into AIDS. “Inhibiting activation of the immune system is not a new concept, but this gives us a new pathway to target,” says Robert Gallo, director of the Institute of Human Virology at the University of Maryland School of Medicine, who was not involved in the study. He warns that other pathways may be at work that are still unknown but that this one is promising if it truly accounts for the large percentage of T cell deaths.

And in fact, a drug already exists that can block pyroptosis. Known as VX-765, it was tested years ago by Vertex Pharmaceuticals as a treatment for chronic seizure disorder. A trial showed that it wasn’t effective enough against seizures, but it was safe for humans. “Now it’s just sitting on a shelf waiting for a disease to cure,” says Greene, who is trying to arrange a phase II trial to test the drug in HIV patients.”

This is solid, interesting, useful science in the field of HIV pathogenesis,” Gallo says, although he questions whether blocking the cells’ death might lead them to later replicate the virus. (Greene believes the cells will degrade the viral DNA and recover.)

Greene envisions the drug, if it proves effective, as a stopgap for patients in developing countries who don’t have easy access to antiretroviral drugs. It could also be an addition to such therapy. Because it targets inflammation, it could reduce the risk of aging-related diseases in patients with AIDS, who are at risk for heart attacks and kidney disease, for example, a decade earlier than noninfected peers.