Cholesterol enables nerve cells to connect.

While cholesterol has a bad reputation for clogging up arteries and causing heart disease, this fatty molecule is an essential part of all cell membranes. Scientists have now found to their surprise that cholesterol may also regulate when and where nerve cells in the brain form the vital junctions known as synapses.

Equally unforeseen, say investigators, is their finding that non-nerve cells called glia seem to provide the cholesterol that controls synapse building.

"We were definitely shocked," says Frank W. Pfrieger of the Max-Delbruck Center for Molecular Medicine in Berlin. He leads a French-German collaboration that reports the new findings at the Society for Neuroscience meeting in San Diego this week and in the Nov. 9 SCIENCE.

Glia make up 90 percent of the cells in the brain, but they have traditionally drawn less interest than have nerve cells, or neurons, which relay electrical signals by releasing chemicals at synapses. About 4 years ago, however, Pfrieger and a colleague noticed that certain nerve cells in the retina had little synaptic activity when grown in isolation from star-shaped glia called astrocytes.

In subsequent experiments, researchers added these glia back into the mix and found that synaptic activity increased 70-fold. The number of synapses per neuron increased by seven times in the presence of the astrocytes (SN: 4/7/01, p. 222).

Biologists subsequently determined that glia secrete a molecule that encourages synapse formation, or synaptogenesis. They rushed to hunt it down because there are few known molecules that regulate that process.

Initial studies by Pfrieger's group pointed to a protein called APOE, but added alone to retinal nerve cells, it didn't promote synapse formation. APOE typically ferries cholesterol around in blood, so the investigators tried adding cholesterol, without APOE, to purified retinal nerve cells. The cell's synaptic activity jumped dramatically. Moreover, when the investigators inhibited synthesis of cholesterol in glia, the cells promoted formation of fewer synapses than normal.

The brain doesn't take up cholesterol from the bloodstream, notes Pfrieger. Investigators had thought, however, that neurons make all the cholesterol that they need. The new findings indicate that glia provide what's needed for synapse formation.

The amount of cholesterol produced by glia "may be a limiting factor for all sorts of things in neurons," says Pfrieger.

Synaptogenesis investigator Craig C. Garner of the University of Alabama at Birmingham cautions that the new data don't prove that cholesterol initiates synapse formation. He notes that stripping nerve cells of cholesterol causes the collapse of synapses, which implies that cholesterol may be merely a structural element required for the junctions.

"The signaling function of [cholesterol] is not clear. It may just be building material," agrees Pfrieger.

He and his colleagues now plan to test cholesterol's importance to synapse formation in other kinds of nerve cells. To extend their test-tube experiments, they will also examine whether glia-derived cholesterol governs synaptogenesis in live animals.

Moreover, they intend to explore a link between APOE and Alzheimer's disease. Having the gene for one variant of APOE predisposes a person to the memory-robbing brain disorder (SN: 8/14/93, p. 108). That variant's interactions with cholesterol may make synapse formation more difficult and somehow cause a brain to become vulnerable to Alzheimer's disease, speculates Pfrieger.

In general, the discovery that glia provide vital cholesterol could lead to new ways of treating a variety of brain disorders, say researchers. "If you want to repair nervous systems, you need to know what critical factors control synapse formation," notes Philip G. Haydon of the University of Pennsylvania School of Medicine in Philadelphia.