Originally created 01/26/01

Mystery cell has role in wiring the brain



WASHINGTON - Stanford University scientists have filled an important gap in understanding how the brain works, discovering what prompts nerve cells to build the vital connections they need to communicate.

Glial cells, long thought to be just some passive scaffolding for the brain's all-important neurons, are directly responsible for how many connections neurons form so they can talk to each other, the scientists report in Friday's edition of the journal Science.

The surprise discovery could lead to better understanding of how memory forms, and perhaps shed new light on what causes certain brain diseases such as epilepsy or Lou Gehrig's disease.

"I'll bet money that there is going to be some disease that is a breakdown in this regulation," said Dr. Charles Stevens, a neurobiologist who wasn't involved in the new research and calls it a major finding.

More immediately important is the basic understanding of how glial cells affect those vital neuron connections called synapses, he said.

"If you want to understand how the brain computes, you have to understand how they form," explained Stevens, who is with the Salk Institute for Biological Studies in La Jolla, Calif.

Glial cells make up most of the brain's cells - for every one neuron there are 10 glia, says Stanford lead researcher Dr. Ben Barres. The scientific dogma was that they only supported neurons, perhaps by providing nutrition, but nobody really knew.

So Barres and postdoctoral student Erik Ullian set out to uncover the function of a main glial cell called an astrocyte.

Neurons are nerve cells that send and receive messages by swapping chemical signals, such as signals that say you're suffering pain or move that leg to walk or retrieve that memory. To do that communicating, neurons first must form synapses.

Scientists once thought neurons were wired to simply build as many synapses as needed. Not so, Barres' team discovered - young neurons form only a few immature synapses when there are no astrocytes nearby, he said.

But add astrocytes to neurons in laboratory dishes and suddenly they form seven times more synapses, and strong, healthy ones, Barres said. Ullian confirmed the finding with another experiment in which he took astrocytes away and the synapses promptly started shriveling.

"People really have not had a good feel for how the brain controls the number of synapses - is the neuron just born with it or are there environmental signals?" Barres said. "Our results show absolutely clearly that environmental signals can have a profound effect on how many synapses neurons can have."

What does it mean for brain research? "We're very interested in the possible disease implications," he said.

Under the microscope, numerous brain diseases show "gliosis," an abnormal accumulation of glia in the brain-injured area. Perhaps glia overreact to an injury, causing neurons to form too many synapses and thus triggering the overfiring that means an epileptic seizure, Barres theorized.

Or consider degenerative diseases like Lou Gehrig's, in which neurons initially die in just one area before the disease spreads. Could overreacting glia kill those additional neurons by overstimulating them?

Then there's the question of memory. Many scientists believe memories are stored by building or strengthening synapses. Astrocytes signal neurons to build synapses by secreting a protein. If Barres could identify that protein - he has experiments under way to try - then scientists could test both the disease and memory theories.