FOR decades, neuroscientists laboured under the notion that the
brain鈥攐ur most precious organ, which defines our mind and controls our
every move鈥攊s utterly incapable of healing itself.
Then, a few years ago, the picture changed: we do produce new neurons,
researchers found. But this only seemed to be happening in paltry numbers in a
few small areas of the brain, so no one was ripping up the textbooks.
Until now. Our brains turn out to be brimming with cells that have the
potential to generate new neurons. As the San Diego conference heard last week,
glial cells, long thought to be merely menial labourers in the brain, could
become neuronal factories with just a bit of tinkering. This raises the exciting
prospect that brains might one day be made to heal themselves.
Advertisement
Doctors are eager to find a ready source of neurons to replace the damaged
brain cells found in disorders such as stroke, Alzheimer鈥檚 and Parkinson鈥檚
disease. Their search has led them to some distinctly unsavoury candidates:
neurons from aborted fetuses, which take us into an ethical quagmire; cancer
cells that can be coerced into becoming nerve cells; or pig neurons, with the
associated risks of rejection by the immune system or infection with animal
viruses.
Using a patient鈥檚 own cells to create replacement neurons would be a far
better solution. Patients could walk into hospital carrying their own
genetically matched home-made repair kit. 鈥淲e could poach cells from a patient鈥檚
brain, coax them to become the proper neurons and reimplant them,鈥� says Arnold
Kriegstein of Columbia University in New York City. 鈥淩ight now that鈥檚 sci-fi,
but this is a step in that direction.鈥�
The brain possesses two major classes of cells: neurons, which carry the
nerve impulses that convey our thoughts; and glia, which were believed to serve
much less glamorous roles. Then researchers isolated neural stem cells (NSCs),
which give birth to neurons. Most textbooks show NSCs as featureless spheres.
鈥淚t was like our picture of the Moon before we were there,鈥� says Magdalena
G枚tz of the Max Planck Institute of Neurobiology at Martinsried in Germany.
鈥淣ow these cells have an identity, a face.鈥�
At the conference, researchers revealed the latest evidence that some of the
cells that function as NSCs are two types of glia. The first are radial glia,
cells with long extensions that neurons migrate along during early brain
development. Kriegstein鈥檚 team tagged the glial cells with a fluorescent protein
and discovered they give birth to neurons, which in turn move along glial cells
into position.
Related work by G枚tz goes further. She has found that most neurons in
the cerebral cortex of mice are descendants of radial glia. The clear message is
that glia are a major source of neurons in the developing brain.
Meanwhile, Arturo Alvarez-Buylla and his team at the University of
California, San Francisco, have shown that another kind of glia, star-shaped
cells called astrocytes, generate neurons in certain regions of the brains of
adult mice. The researchers are confident that glia play the same role in
humans. Previously, astrocytes were thought to do little more than eat debris,
deliver nutrients and hold neurons in place.
Glia would be a hugely attractive source of new nerve cells, because they
outnumber our neurons 10 to 1. But exploiting them won鈥檛 be easy. Not all glia
naturally produce neurons. In fact, even in the parts of adult brains where
neurons are known to form, mice studies show that only about 1 in 100 glial
cells act as NSCs.
But Gordon Fishell of the New York University Medical Center has found a way
to improve the odds. When his team used an engineered virus to introduce more
genes for a protein receptor called Notch into astrocytes, the number capable of
neurogenesis increased 25-fold. And when he infected the cells with a protein
that poisons the Notch activity, the number of cells with NSC ability dropped to
1 in 300. 鈥淭hat鈥檚 strong proof Notch is a central player,鈥� he says. Similarly,
G枚tz鈥檚 team has shown that a protein called Pax6, a transcription factor
which activates other genes, pushes glial cells to produce neurons.
This is still a long way from treating disease, of course. To treat
Parkinson鈥檚, for instance, the cells would have to become dopamine-producing
neurons, which will probably require treating them with a sequence of different
factors. 鈥淚t will take a while to work out the recipe for that,鈥� says
Fishell.
But researchers are thrilled by the idea that glial cells may be the 鈥渕other
tissue鈥� of the brain. Kriegstein thinks that as well as clearing damage, they
could eventually generate neurons and then help guide them to their proper
position. And that might be just the beginning, according to Fiona Doetsch of
Harvard University, who has been working with Alvarez-Buylla. 鈥淕lia have been
the most mysterious cells in the brain,鈥� she says. 鈥淭heir role as NSCs is just a
hint there is much more to discover about them.鈥�
