Marching orders. Microglia (green) migrate to injured neurons (red, top) but fail to do so when they cannot detect ATP (bottom).

The Brain's Bat Signal

Microglia are the brain's resident security guards, surveilling the organ for damage and then crawling to the injury site to engulf dead neurons. Exactly how they detect problems was unclear, but researchers now show that they respond to an SOS signal from dying cells that is relayed throughout the brain. The finding may have implications for the treatment of Alzheimer's and other neurodegenerative diseases.

The study builds on previous work in zebrafish. Developmental biologist Francesca Peri of the European Molecular Biology Laboratory in Heidelberg, Germany, and colleagues created genetically engineered versions of the animals that produced microglia labelled with green fluorescent protein, a glowing compound frequently used in laboratory research. Zebrafish embryos have transparent brains, which allowed Peri and her team to track the microglia in real time under the microscope. The researchers reported in 2008 that the embryonic zebrafish brain is patrolled by about 20 of the cells.

Other researchers have shown that the ability of microglia to engulf dead neurons depends on adenosine triphosphate (ATP), a ubiquitous energy source and signalling molecule that is released from damaged cells. But ATP is rapidly degraded after being released from cells and cannot act as a long-range signal.

In the new study, Peri and her colleagues used lasers to destroy small numbers of neurons in the genetically modified zebrafish. In response, all the microglia migrated to the injury site, suggesting that they are indeed attracted by a long-range signal. But what could it be?

On the move. Watch microglia migrate to an injury site to clear up cellular debris.
Credit: EMBL/Francesca Peri

Further experiments revealed that injured cells initiate a wave of elevated calcium ion concentration that travels through the brain, and that the microglia only begin migrating when the wave reaches them. The wave sweeps through the brain at approximately 14 micrometers per second; the microglia turn their fingerlike appendages in the direction of the wave and then begin migrating toward the injury site within one minute. The spread of the wave was related to the extent of the damage, so killing fewer cells produced a smaller wave detectable only by microglia in the immediate vicinity. Adding a drug that blocked the calcium wave prevented microglia from migrating to the injury site, the team reports today in Developmental Cell.

In a final set of experiments, the researchers showed that the calcium waves are generated by a neurotransmitter called glutamate, which is released from damaged cells. Glutamate activates receptors on neighboring cells, causing them to propagate the wave and release ATP.

Microglial cells may malfunction in neurodegenerative disorders such as Alzheimer's disease, and there is some evidence that they worsen the damage caused by a stroke. The new findings, says Peri, implicate the calcium waves as potential targets for drug treatments. A drug could stop microglia in their tracks, for example, or reroute their migration.

The findings provide "a convincing explanation of how dying neurons attract microglia," says Frank Kirchhoff, a glia physiologist at the University of Saarland in Homburg, Germany, who was not involved in the study. He cautions, however, that the experiments should be repeated in mammals. Still, says neurophysiologist Alexej Verkhratsky of the University of Manchester in the United Kingdom, the microglial response to brain injury in zebrafish is "strikingly similar to that of mammals," suggesting that what Peri and her team observed is applicable to humans.

Kirchhoff is skeptical, however, that the calcium wave could be a drug target. Calcium is a "promiscuous" signal used by all cells for a wide variety of processes, he notes, and thus drugs that target it could have major side effects. And despite the utility of the zebrafish model for sussing out how microglia work, drug researchers are going to have to look elsewhere, says Verkhratsky. "[They're] limited as a disease model, because neurodegeneration does not occur in fish."

Posted in Brain & Behavior