The only way to confirm that someone has Alzheimer's disease, which afflicts an estimated five million Americans, is by peering into their brains and seeing plaque, nestled between the twisted endings of affected nerve cells. Unfortunately, these markers can only be viewed during postmortem investigations.

To date, researchers are unsure whether these bundles (made of a protein fragment called amyloid beta) cause all—if any—of the symptoms of Alzheimer's dementia. Previous work has led to different theories on when plaque forms in the Alzheimer's-ravaged brain: One view holds that they are the cause of the disease, protein deposits develop and disrupt the functions of axons and dendrites (projections originating at the cell body that send and receive messages, respectively). The counter theory posits that they develop as a result of it; affected neurons pump out a surfeit of amyloid precursor protein (the full protein strip from which amyloid beta hails), changing the shape of axons and dendrites (collectively known as "neurites"), eventually leading to disruptive plaque deposits.

Researchers report in Nature that they conducted a study, employing a new visualization technique, on a mouse cortex (the outer layers of the brain) that indicates the plaque may in fact precede disruptions to the neurons' messengers. In addition, they say, it shows that plaque amasses quickly, becoming significant enough within 48 hours to block normal neuronal function and to trigger inflammation (the immune system's first line of defense) within a week.

"When we observed the plaque formation, we didn't see these dystrophic neurites before the plaque," says study co-author Bradley Hyman, director of the Alzheimer's Unit at Massachusetts General's Institute for Neurodegenerative Disease. "Clearly the whole disease, if you think about the entire cortex, plays out really slowly, but every plaque forming event plays out very rapidly."

The researchers also observed that within a day of plaque formation—glial cells (nonneuronal nervous system cells that help support and maintain neurons) race to fill space in order to prevent additional protein buildup. Scientists had previously been unsure of the role of glial cells, that is, whether they slowed or accelerated plaque formation.

To make these observations, Hyman and his colleagues employed a sophisticated new visualization tool called "multiphoton laser confocal microscopy." Essentially, the technology allows the team to section the brain with long-wave light and then drill down and look deep into it without causing physical changes. This way, they were able to track plaque as it formed in mice that were genetically engineered to develop it.

Larry Goldstein, a professor of cellular and molecular medicine at the University of California, San Diego, says that no one suspected that plaque forms so swiftly yet remains so stable (without being cleaned out by the glia). His research and that of others, however, supports the idea of chemical-transporting neurites preceding plaque formation. "I don't think that [the new work] necessarily disagrees with what we're saying. … The regions that we were seeing anomalies in were not regions that normally have plaques," he says. "There are other important cellular changes that happen in Alzheimer's disease–like models that don't have anything to do with plaque, but have to do with transport."

Hyman plans to continue to probe how plaque forms so fast (at least in the cortex) and if there are smaller, plaque complexes that join up to create them. He also plans to explore what happens immediately after plaque accumulates, specifically, how—or if—they disrupt normal neuronal activity. "If you understood what it was that caused that initial precipitation event," Hyman says, "I think that would be a therapeutic target."