Autoimmune attack of central nervous system (CNS) components is associated with devastating neurodegenerative diseases such as multiple sclerosis. Although autoimmune T cells are usually viewed as detrimental, Schwartz and colleagues report on page 49 of this issue the unexpected finding that they also can be neuroprotective. They administered T cells specific for myelin basic protein (MBP) to rats in which the optic nerve had been injured and, to their surprise, found that the immune cells protected the injured neurons from further damage.

"We were somewhat worried because all of these results are against the conventional wisdom of generations of immunologists," says Schwartz. "We have all been taught that the immune system is designed to be kept out of the CNS. These results prompt us to consider the possibility that T cell-mediated immune activity against self CNS components can do good for the immune system as well."

CNS injury is accompanied by changes in the concentration of extracellular ions, free radicals, and neurotransmitters, resulting in the gradual secondary loss of adjacent undamaged neurons. The photograph shows a whole-mounted retina excised after partial injury to the optic nerve. Each retrogradely labeled cell (green) represents a neuron that escaped the primary lesion and has not yet undergone secondary degeneration. Schwartz and co-workers demonstrate that MBP-specific T cells inhibit the eventual secondary degeneration of these neurons after the primary injury. These results are particularly interesting given their previous observation that macrophage-induced inflammation can promote neuron re-growth after axonal transection (Nature Med. 4, 814–821; 1998).

But how does the immune response help to preserve these neurons? The authors suggest that MBP-specific T cells cause a transient reduction in the electrophysiological activity of damaged neurons, which may prevent depletion of their energy supplies keeping them alive longer.

Although their recent investigation into 'benign autoimmunity' has caused them to diverge from their initial goal—to determine the potential of infiltrating T cells to deliver gene therapy vectors to the CNS—the Schwartz team still intends to pursue the notion of T cells as gene therapy vehicles. "At areas in which there is no lesion, T cells don't accumulate. Self-reactive T cells are the perfect gene delivery vehicle because of their specificity for CNS lesion sites," says Schwartz.