Stanford Report Online

Stanford Report, February 7, 2001
Researchers discover brain ‘filler’ has critical role in nerve cell communication


Stanford medical researchers have found that brain cells traditionally disparaged for having no important purpose are critical for forming the links that allow neurons to communicate with one another. These glial cells, which constitute 90 percent of the cells in an adult human brain, were thought to be housekeeping cells providing physical support and nourishment to neurons and mopping up after them.

"Every textbook says they don't do anything, even though they're the major cell type in the brain," said Ben Barres, MD, PhD, associate professor of neurobiology.

Barres and his team have found that far from just providing inert filler material the glia actually induce neighboring neurons to form synapses -- the tiny gap between the end of one neuron and the beginning of another. Electrical signals travelling along neurons cross synapses via the release of chemicals called neurotransmitters. Without glia, neurons make sevenfold fewer synapses and those that do form are immature and largely ineffectual.

The researchers initially were surprised by their own findings, which were published in the Jan. 26 issue of Science magazine. "We never thought glia would be involved," said Barres. "No one ever implicated glia in the formation of synapses."

"Everyone just thinks it's glue; that it doesn't do anything, just holds stuff together," said Erik Ullian, PhD, postdoctoral fellow in Barres' lab and lead author of the Science paper. The word glia means glue in Greek, Ullian pointed out.

It was a series of technical advances in the laboratory that allowed the team to separate the two different types of brain cells from each other and decipher what they are actually doing.

"It's been really hard to take neurons and glia apart, so people have been unable to answer the question [of what glia do]. Now we can study the neurons by themselves and ask what they need the glia for. And it turns out, quite a lot," said Barres.

The most recent findings by Ullian and his colleagues build on previous work in Barres' lab showing that neurons grown in cultures without glia had very little synaptic activity. When glia were re-introduced to the culture, synaptic activity was between 10 and 100 times greater.

Ullian's research aimed to find out if the increased activity resulted from more synapses being produced under the influence of the glial cells, or whether the synapses already there were more effective when glia were present. The answer was convincing.

"Every experiment showed that there's a decrease in synapses [if no glia are present]. Only a small number form without glia and they are very immature," said Ullian. Furthermore, the team found that glia are important for maintaining healthy synapses. If glia are removed the synapses quickly wither away.

The researchers believe the behavior of the cells in the lab mimics what they do inside the human brain. Scientists already knew that most synapses in the brain form at the same time glia are produced. But until now, the critical role of glia in neuronal development was unknown. The neurons are ready and waiting to begin communication but until glia are produced, the synapses that enable signals to be transmitted are lacking, Barres explained.

The team's results may have implications for how young brains adapt during long-term memory and learning, when synaptic activity increases, and also for how brains react to injuries from strokes or from neurodegenerative diseases like Alzheimer's or Lou Gehring's disease.

"Injured brains have way more glia -- the glia are larger and they get very angry-looking," said Barres.

According to the researchers, this gliosis, or change in glia, is well-described but debate continues about whether gliosis is merely a symptom of brain injury or if gliosis itself damages the brain. "Our findings suggest gliosis may induce overproduction of synapses, which leads to neuronal death. If neurons are overstimulated they die," said Barres. The distribution of synapses is very finely balanced, he added. "You need the right number and they need to be in the right place."

"You don't need to increase synapses by a very large number to get a profound increase in activity," said Ullian. "A sevenfold increase in number is a tremendous increase in terms of function of a neuron."

Ullian's work also has shown that glia do not have to be in physical contact with neighboring neurons to induce them to form synapses. The cells appear to secrete a protein that instructs the neurons in synapse formation. He and the other members of the team plan to probe the nature of this protein and they will also begin looking at other neurons in the central nervous system to see if they respond to glia in the same way as the retinal ganglion cells they have been studying.

"It took 10 years to get to the stage to start these experiments," said Barres. "It's a lot of fun now figuring out what the protein is and whether glia are doing the same thing in vivo. And no one else is working on it so we sleep well at night."

Karen Christopherson, PhD, postdoctoral fellow in Barres' lab, and Stephanie Sapperstein, PhD, a former postdoc in the lab, are the remaining co-authors of the Science paper.