|
||||||||
|
NEUROBIOLOGY Immune Proteins Found Moonlighting in BrainDiscovery Alters Perspective on Dyslexia and Neurodegenerative Diseases A pair of fuzzy-brained mice is providing tenet-toppling clues to one of the great puzzles in neuroscience: How do the finely wrought neural patterns of the brain emerge from a hodgepodge of early connections?
"It has to be seriously considered now that in brain cells undergoing degeneration there is a possibility that the neurons are being targeted for destruction by the immune system," said Carla Shatz (left), shown with co-authors Gene Huh and Lisa Boulanger, research fellow in neurobiology. Photo by Pam Murray The mutant mice, studied by HMS researchers, are also offering tantalizing clues to neurological mysteries such as how might developmental disorders like dyslexia arise? What triggers the killing of neurons in Parkinson's disease and other neurodegenerative disorders? And might immune system proteins be involved? While the brain's early scaffolding of connections is determined strictly by genetic instructions, the refashioning that occurs during development—and in learning—is a product of both genes and the brain's own activity. Yet neuroscientists have come up mostly empty-handed in their quest to identify the molecular artisans responsible for this activity-dependent remodeling. Gene Huh, Carla Shatz, and their colleagues have recently pulled a pair of candidates from an unlikely pool—the immune system. They found that mice lacking one of two immune proteins, Class I MHC and CD3-zeta, failed to undergo remodeling in a visual area of the brain, the lateral geniculate nucleus. In the immune system, Class I MHC and CD3-zeta act as part of a lock and key system to recognize and rid the body of foreign invaders. In the brain, they may be part of a signaling system that recognizes and eliminates inappropriate connections, says Shatz, the Nathan Marsh Pusey professor and chair of neurobiology at HMS. The mouse findings, which are reported in the Dec. 15 Science, support such an interpretation. Normally, projections from the eye form a small tidy patch in the geniculate nucleus, but in the mutants, the connections created a larger and fuzzier profile, presumably because cells in the region lacked the molecular means for getting rid of unneeded connections. The mice also experienced abnormal functioning in a region of the brain associated with learning, the hippocampus. Brain PoliceThe discovery that immune system proteins play a role in the activity-dependent remodeling of the brain overturns a long-cherished dogma. For years, the brain was thought to be an immunologically privileged place—free from the immune system policing that occurs everywhere else in the body. Although neuroscientists have recently found evidence that the brain is subject to immune surveillance, few suspected that the brain produces its own immune molecules.In fact, Huh, a research fellow in neurobiology, Shatz, and their colleagues found that the mouse brain produces Class I MHC and CD3-zeta not just in the hippocampus and visual system but in many other regions. Intriguingly, production is especially high in primary sensory areas of the brain—precisely those areas that are thought to function abnormally in people with dyslexia.
In the immune system Class I MHC and CD3-zeta act as part of a lock and key system for recognizing and getting rid of foreign invaders. At left, Class I MHC (pink) presents foreign antigens to the T cell receptor (TCR), which contains CD3-zeta (red). In the brain, at right, Class I MHC may interact with an unknown receptor (circle), which contains CD3-zeta, part of a signaling system that recognizes and eliminates inappropriate connections. Adapted from original by Gene Huh Though evidence is still sketchy, Shatz believes defects in the Class I MHC and CD3-zeta remodelers could play a role in dyslexia. "If you do MRI and look at the brain of a person with dyslexia, it looks perfectly normal. There's nothing structurally weird about the brain," she said. "But it is possible that if you look at the detailed connections, they would be funny." Preliminary studies by British researchers of families with dyslexia suggest that some of them carry genetic defects on chromosome 6—the same chromosome that carries the Class I MHC genes. "It's very speculative at this point, but it remains certainly a possibility that this could in some way be related to their dyslexia," Shatz said. The presence of MHC Class I prompts another speculation: that neurodegenerative diseases such as multiple sclerosis and Parkinson's may be the consequence of a misguided attack by immune cells on Class I MHC-bearing neurons. "The idea that neurons would normally be expressing Class I MHC might help explain why certain neurons die or are attacked," Shatz said. "MHC Class I-bearing neurons could be the target for an abnormal immune response. I think that people need to start thinking about that." The discovery that immune proteins play a role in sculpting the brain is but the latest in a string of Shatz lab surprises. The first came about ten years ago when she and her colleagues discovered that well before birth and long before the eye actually sees, cells of the retina send spontaneous waves of signals to the geniculate nucleus that result in the region's characteristic layered patterning. About FaceThinking that the waves of activity might be turning on and off genes in the region, the researchers blocked the retinal signals in the hopes of seeing which local genes might be involved. Of the handful of genes they found to be affected, one was Class I MHC. "We couldn't believe this. We really thought it was wrong. Class I MHC was not supposed to be expressed in the brain," she said. Additional tests confirmed that Class I MHC was expressed not just in the geniculate nucleus but in the hippocampus. But the clincher came when a subunit of the immune cell receptor for Class I MHC, called CD3-zeta, was discovered in those very same brain regions. "We were totally blown away," Shatz said.To see whether the molecular pair actually plays a role in the brain, Huh, Shatz, and colleagues examined strains of mice lacking normal versions of Class I MHC or CD3-zeta. Not only were the patches of retinal connections in the lateral geniculate nucleus unusually large and coarse, they were surrounded by a scattering of retinal connections that regress in normal mice. Shatz believes the reason for this messiness is that the mutants lack a molecular mechanism for getting rid of unneeded connections. "We think Class I MHC acts like an anti-glue," said Shatz. Evidence from the mutants' hippocampus appears to support this hypothesis. To promote learning and memory, the hippocampus employs an ingenious system called long-term potentiation, in which the more active a neuron is the stronger and longer-lasting its synapses. Conversely, the less active a neuron, the weaker and more transient its connections, a phenomenon known as long-term depression. The researchers found that the hippocampus of the mutants exhibited very high levels of long-term potentiation and no long-term depression. In other words, all connections—active and inactive—were retained. "It appears that the mice can learn because they can form appropriate connections. But they cannot eliminate inappropriate ones," said Shatz. What effect this lack of molecular de-adhesive has on behavior is not clear since the mice have not yet been put through a maze or given behavioral tests of any kind. One possibility is that like people with dyslexia, the mutants may have problems processing sensory information. Though intriguing, a mouse model of dyslexia remains a thing of the future. "Even normal mice cannot read, I'm afraid," Shatz said. —Misia Landau |
