Losing sleep could lead to losing brain cells, a new study suggests.

Levels of a protein that forms the hallmark plaques of Alzheimer's disease increase in the brains of mice and in the spinal fluid of people during wakefulness and fall during sleep, researchers report online September 24 in Science. Mice that didn't get enough sleep for three weeks also had more plaques in their brains than well-rested mice, the team found.

Scientists already knew that having Alzheimer's disease was associated with poor sleep, but they had thought that Alzheimer's disease caused the sleep disruption.

"This is the first experimental study that clearly shows that disrupted sleep may contribute to the disease process," says Peter Meerlo, a neuroscientist at the University of Groningen in the Netherlands. "What makes it exciting for me is that it shows that chronic sleep loss, in the long run, changes the brain in ways that may contribute to disease." A vicious cycle could result if sleep loss leads to Alzheimer's disease and the disease leads to more sleep loss, he says.

Researchers led by David Holtzman, a neurologist and neuroscientist at Washington University in St. Louis, used a method called microdialysis to measure the levels of a protein known as amyloid-beta in the fluid between brain cells of mice. Amyloid-beta sometimes twists into a sticky form and clumps together, forming such plaques. Scientists don't yet understand how, but they think that clumping of amyloid-beta eventually leads to the death of neurons and the symptoms of Alzheimer's disease (SN: 8/16/08, p. 20).

Although levels of amyloid-beta in the brain tissue of the mice didn't seem to change, Holtzman's group found that levels of the protein released into brain fluid did rise and fall throughout the day. "We didn't know it would coordinate with sleep and wakefulness," Holtzman says. "We just knew the levels fluctuated."

Levels of the protein increased in mice during the night - when mice are mostly awake - and fell during the day when mice sleep. The longer the mice stayed awake, the more amyloid-beta levels increased, the team found. The team also measured amyloid-beta levels in the cerebral spinal fluid of some healthy young people and found the same pattern observed in the mice - amyloid-beta levels increase when people are awake and fall during sleep.

Giving mice a shot of a hormone called orexin, which promotes wakefulness, also caused amyloid-beta levels to increase. And blocking orexin's activity led to a decrease in the amount of protein released into the brain fluid. The researchers don't yet know whether orexin is directly responsible for helping release amyloid-beta into brain fluid or if orexin keeps animals awake, allowing more time for levels of the protein to build up.

For three weeks, Holtzman's team studied mice that were genetically predisposed to build Alzheimer's plaques, allowing some of the animals to sleep only four hours a day while others slept normally. Sleep-deprived mice made more plaques than well-rested mice, but a drug that blocks orexin's action was also able to stop plaque buildup, the researchers discovered.

Studies in people haven't shown a link between Alzheimer's disease and chronic sleep loss, but Holtzman speculates that lack of sleep, particularly in mid-life when plaques begin to form, could hasten onset of the disease in genetically susceptible individuals. Drugs that block orexin might also be used as a potential therapy for halting plaque development, he says.

Other researchers aren't so sure that's a good idea. "Treating patients chronically with orexin inhibitors is really not an option," says Masashi Yanagisawa, a Howard Hughes Medical Institute investigator at the University of Texas Southwestern Medical Center at Dallas. The drugs would likely make patients sleepy unless used at extremely low doses, he says. One such drug is in clinical testing as a treatment for insomnia.

It is also unclear whether orexin or some other aspect of the sleep and wake cycle regulates amyloid-beta levels, researchers say.

"Mechanistically we don't understand why [sleep] is manipulating amyloid-beta rhythms," says Sangram Sisodia, a molecular neurobiologist at the University of Chicago, "but we do know it's doing something good for the brain.... There's a clear message here about why it is so important to sleep."