In 2002 a clinical trial of an experimental Alzheimer's vaccine was halted when a few patients began experiencing brain inflammation, a result of the immune system mounting an attack against the body. Now some researchers claim that inducing a mild autoimmune reaction could actually protect the central nervous system from a spectrum of neurodegenerative conditions, from glaucoma and spinal cord injury to Parkinson's and Alzheimer's disease. "This is a hot-button issue right now," says Howard Gendelman of the University of Nebraska Medical Center in Omaha.

It all started with glaucoma. Once thought to result primarily from high pressure in the eyeball constricting the optic nerve, the disease has lately come to be seen as a form of neurodegeneration, propagating from the injured optic nerve to healthy cells in the brain. Before monkey studies had demonstrated as much, neuroimmunologist Michal Schwartz of the Weizmann Institute in Rehovot, Israel, observed in the late 1990s that crushing a small portion of a rat optic nerve creates a large zone of sickened cells. She and her team also found that T cells, the immune system's attackers, gathered at these wounds.

Curious if the small accumulation was helpful or hurtful, the researchers injected different types of T cells into rats with optic nerve injury. Surprisingly, rats given T cells specific to myelin, the fatty sheath coating neurons, retained three times as many functional retinal ganglion cells as rats injected with other T cells. In subsequent experiments, rats genetically engineered to lack T cells, as well as rats insensitive to myelin autoimmune reactions, fared worse in glaucoma models than normal rats did.

Introducing antimyelin T cells to people would most likely cause brain inflammation, so Schwartz looked for a compound that would induce a weaker reaction. Copaxone, a peptide drug approved for the treatment of multiple sclerosis, fit the bill because the body's immune response against it also weakly targets myelin. And indeed, rodents vaccinated with Copaxone after insults to their optic nerves retained more retinal ganglion cells than untreated animals did.

Schwartz argues that the effect exploits a natural "protective autoimmunity" and has championed it as a more general measure for protecting the brain from disease. Too much autoimmunity causes brain disease, but too little may exacerbate the gamut of neurodegenerative conditions, she asserts. "It's a beautiful hypothesis," remarks Hartmut Wekerle of the Max Planck Institute for Neurobiology in Martinsried, Germany, but one that has split neuroimmunologists. "I think Schwartz's theory is right because it's been shown in a number of animal models," says Howard Weiner of the Center of Neurologic Diseases at the Brigham and Women's Hospital in Boston. "There's a reasonable chance it'll work in humans." In further support, Gendelman's group reported in 2004 that transferring Copaxone-specific immune cells to mice protects neurons in a model of Parkinson's disease.

The evidence is mixed, however. Spinal cord researcher Phillip Popovich of Ohio State University has been unable to mimic results from Schwartz's lab, in which transferred T cells protect spinal cord tissue. "We get what the conventional wisdom would expect: we get more problems," Popovich reports. The discrepancy probably results from subtle differences in the models employed, which implies that the effect is not robust enough to treat spinal cord injuries, he contends.

Mice have been cured of their versions of many diseases that still afflict humans, notes neuropathologist V. Hugh Perry of the University of Southampton in England. And unlike lab rat strains, individual people vary in their immune responses, creating the risk that vaccination will cause harmful autoimmune reactions, as occurred in the interrupted Alzheimer's trial. Perry acknowledges, however, that in some cases, "the regulation of inflammation is not as precise as it might be. If you can induce T cells to produce anti-inflammatory molecules, that may be a good thing."

Gendelman sees obstacles ahead before the great potential of protective autoimmunity, as he describes it, can be exploited. "How this occurs is a big black box," he says. The positive evidence has piqued some biotech interest, though: Israel's Teva Pharmaceutical Industries is investigating Copaxone and a similar peptide in models of glaucoma and several other neurodegenerative conditions. If the company moves ahead with clinical trials, that black box may open up.