91É«ÇéƬ

Some people’s talents are so well hidden that you overlook them completely.
Sometimes it is the same with cells. According to tradition in neuroscience,
brain cells fall into two broad groups – neurons and glial cells. Neurons
are the smart, fast-talking communicators that process information in the
brain; glial cells are the humble drudges that do little more than fill
in the space between neurons. Or so generations of biology students were
taught.

No longer. These days glial cells are fast shaking off their old image,
as neuroscientists uncover an exciting and hitherto hidden side to their
personality. Far from being dumb space-fillers, glial cells are highly communicative.
Not only can they pass messages to each other, they can also receive signals
from – and perhaps even broadcast signals to – their supposed superiors,
the neurons.

And important medical implications are emerging as well. Glial cells
turn out to play a vital part in brain development, in part because they
produce substances that help neurons to prosper and grow. They may also
hold the key to unlocking the biological ‘blood-brain barrier’ that stops
many drugs entering the brain. Already glial cells are drawing interest
from drug and biotechnology companies. In future, their medical importance
may mushroom as researchers attempt to exploit them to develop therapies
for illnesses such as Alzheimer’s disease.

The change of fortunes began in the 1980s, when it gradually dawned
on neuroscientists in a number of laboratories that glial cells have more
in common with neurons than they thought. In particular, researchers discovered
that the membranes of glial cells carry an impressive array of ‘signalling’
proteins similar to those found in neurons – namely, proteins which act
as channels for ions or receptors for neurotransmitters.

Similar types of protein enable neurons to fire off electrical impulses
in response to neurotransmitters – the nervous activity that lies behind
everything from flexing a finger to writing a symphony. The puzzle with
glial cells is that they stop short of producing such full-blown electrical
impulses. So why do they need ion channels and receptors?

Housekeeping duties

The search for a complete answer is in full swing, but several elements
are already clear. Researchers suspect that glial cells perform a number
of ‘housekeeping’ tasks in the brain. For example, when neurons fire they
pump out potassium ions. Glial cells can absorb these ions through some
of the specialised channels in their cell membranes, thereby preventing
a potentially dangerous build-up of high levels of potassium in the fluid
surrounding brain cells.

Glial cells also perform important duties at the brain’s synapses,
mopping up neurotransmitters such as glutamate after it has done its job
of triggering a burst of electrical activity. One beneficial result is that
the amount of glutamate around neurons is kept to a low level – a vital
task, for high concentrations of glutamate are toxic to neurons and are
a major cause of the damage sustained during strokes and similar insults.

The mopping up is done by a membrane protein called the glutamate uptake
carrier. David Attwell and his colleagues at University College London
have shown that this carrier conducts a kind of foreign trade, importing
glutamate and sodium ions across the cell membrane while simultaneously
exporting potassium and hydroxyl ions. Sometimes – during strokes, for example
– the carrier protein goes into reverse and exports glutamate, with disastrous
consequences.

These ‘housekeeping’ roles form only a part of the rich relationship
that exists between glial cells and neurons. Since glial cells have receptors,
they can respond to many of the neurotransmitters that neurons use for communication.
Glial cells, in other words, can ‘understand’ the chemical language of the
brain.

‘One thing that is increasingly clear now is that glial cells have the
capacity to sense neurotransmitters,’ says Barbara Barres, a neurobiologist
at Stanford University. That could mean that glial cells spend time ‘eavesdropping’
on signals passing between neurons. It may also mean that they conduct subtle
but influential dialogues of their own with neurons.

Either way, evidence from molecular biology suggests their vocabulary
could be anything but limited. The receptors found so far on glial cells
include ones for glutamate, gamma-aminobutyric acid and noradrenaline, all
important neurotransmitters with a wide range of functions in the brain,
including the processes underlying learning, anxiety and moods.

Evidence from cell biology reveals a gift for getting messages from
one place to another. Biochemical responses to neurotransmitters may spread
like a wave from one glial cell to its neighbours, judging from observations
made by Stephen Smith and his colleagues, first at Yale University, then
at Stanford. The team worked with cultured astrocytes, one of the main types
of glial cell, from a part of the brain called the hippocampus. When they
added glutamate to the cells, the researchers noticed something quite extraordinary:
levels of calcium ions inside the cells began to rise and fall rhythmically.
Moreover, these ‘calcium waves’ rippled through the cells, passing from
one cell to the next where their membranes touched. Unlike neurons – which
communicate by releasing neurotransmitters across gaps called synapses –
astrocytes make direct contact with their neighbours. These specialised
connections allow the cells to exchange ions and small molecules, communicating
biochemical messages in the process.

Smith and his colleagues watched one calcium wave that flowed through
no fewer than 59 cells before disappearing. It was important to find out
whether such waves occurred in the living brain as well as in laboratory
cultures, so the researchers examined slices of tissue from the hippocampus.
The answer seemed to be yes. By electrically stimulating a nerve pathway
in the tissue, Smith and his colleagues found that they could trigger calcium
waves in astrocytes. Moreover, the nerve pathway in question was formed
by neurons that work by releasing glutamate. So it seemed that the calcium
waves were being set off by interactions between astrocytes and neurons.

Smith is optimistic about the implications. ‘There’s a fair weight of
evidence that we are talking about a phenomenon that occurs in intact brain
tissues,’ he says. If so, glial cells may do more than simply listen to
neurons. They could conceivably send out messages of their own over relatively
long distances. Being carried by calcium waves, as opposed to neurotransmitters,
such missives would be slow and stately compared with those dispatched by
neurons. But in the long run they might be no less influential in the brain
– a case perhaps of glial cells playing tortoise to the neuronal hare.

Glial cells listen to neurons – and talk to one another – but do they
talk back to neurons? The most direct way of communicating with neurons
is to release neurotransmitters. Can glial cells do this? The jury is still
out. But if it turns out that they do, the discovery will have far-reaching
implications for attempts to understand the workings of the brain. ‘It’s
an incredibly exciting possibility,’ says Barres, ‘because then you’d have
to include them in neural circuits.’

Chemical traffic

A less direct way to communicate with neurons is by influencing the
flow of chemical traffic at synapses. ‘Glial cells are clearly responding
to what’s being released into the synaptic area,’ says Sean Murphy of the
University of Iowa, ‘and they have the potential for releasing factors back
into that area which could modulate synaptic transmission.’ In theory this
could give them a role in the sorts of cellular changes that are widely
believed to underlie learning and memory.

If glial cells are as communicative as some researchers believe then
one medium for their messages could be nitric oxide (‘The body’s vital poison’,
New Scientist, 13 March 1993). Many neurons release nitric oxide, and it
has long been suspected that glial cells may be among their targets. But
until recently it seemed unlikely that glial cells could make the gas themselves.

Researchers such as Murphy and his colleagues have now begun to challenge
that view. ‘We have shown that some neurotransmitters will activate nitric
oxide production from astrocytes – and other groups have done the same,’
explains Murphy. Dahlia Minc-Golomb and Joan Schwartz at the US National
Institutes of 91É«ÇéƬ have found evidence that astrocytes can make the enzyme
nitric oxide synthase, the biochemical machine that generates nitric oxide.
Although some neuroscientists will want to see more evidence, these findings
hint that astrocytes could be as fluent as neurons in the language of nitric
oxide.

No one yet knows where astrocytes direct their nitric oxide, but one
ingenious speculation focuses on the brain’s small blood vessels. Astrocytes
make intimate contacts both with neurons and blood vessels in the brain.
They would be ideally placed to sense neural activity and respond by puffing
nitric oxide at blood vessels. Since blood vessels expand in response to
nitric oxide, here would be one mechanism by which neural activity in a
particular brain region could stimulate the blood supply to that region.

In a quite separate development, researchers have unearthed evidence
of another kind of signalling between astrocytes and the brain’s blood vessels.
In the brain, small blood vessels are not as leaky as they are elsewhere
in the body. As a result many bloodborne substances, including some drugs,
cannot enter the brain and spinal cord via the circulation. This blood-brain
barrier creates a haven in which the brain can practise its delicate chemistry,
free from outside disturbance. Because the barrier frustrates drug treatment
for brain disease, researchers are keen to understand how it originates.

The barrier owes its existence to the endothelial cells that line the
brain’s blood vessels; these cells cling tightly to one another, creating
a highly effective seal. Astrocytes may provide a chemical signal that
tells them to do so. This hypothesis was backed up when Robert Janzer and
Martin Raff, working at University College London, took clumps of cultured
astrocytes from rats and placed them inside hens’ eggs.

Clinging tightly

The blood vessels that invaded the clumps turned out to be unusually
resistant to leaks. Their endothelial cells had formed a tight seal, implying
that they had received some persuasive molecular message from the astrocytes
telling them to do so. Chick cells understood the message, even though it
came from rat astrocytes – an intriguing form of communication between species.

Researchers are now trying to find out more about this message. One
approach is to try and recreate the interaction of astrocyte and endothelial
cell in the test tube, using cultured cells. So far Lee Rubin and his colleagues
at the Eisai London Laboratories, University College, have succeeded in
persuading cultured endothelial cells to form non-leaky junctions. But they
have been unable to produce this effect with astrocytes alone. The endothelial
cells also have to be tweaked biochemically by increasing the amount of
a messenger molecule called cyclic AMP inside them.

Brain development is another area where glial cells make their presence
felt. In many parts of the brain, such as the cerebral cortex, they act
as guides for developing neurons. Neurons are born by cell division in a
‘proliferative zone’ – a kind of neuronal nursery – and must then migrate
into position without losing their way or their position relative to their
neighbours. Classic studies by Pasko Rakic at Yale University showed that
this orderly flow of neurons is accomplished in a remarkable manner. Strands
of glial cells define the route the neurons take and the neurons hug the
strands as they move, like acrobats shinning up ropes.

Glial cells may also influence the welfare of neurons in more subtle
ways. To survive during development, neurons need a steady supply of substances
called ‘neurotrophic factors’, the best known being nerve growth factor.
Deprived of their supply, neurons die by turning on a kind of suicide program.
Some researchers believe that a similar addiction persists in the adult
brain, and perhaps in all tissues. Cultured astrocytes produce a huge range
of substances, including many neurotrophic factors. If they secrete those
same chemicals in the brain, they could have an important role as nurturers
of neurons.

The implications for finding substances that could halt the destruction
of neurons in degenerative illnesses such as Alzheimer’s disease and Parkinson’s
disease are not lost on medical researchers. Last year Frank Collins and
his colleagues at Synergen, a biotechnology company in Boulder, Colorado,
discovered a previously unknown neurotrophic factor in a cultured line of
glial cells. This substance is especially good at nurturing dopamine-releasing
neurons, the kind that degenerate in Parkinson’s disease. Researchers are
hoping that the newly discovered substance will one day help doctors treat
the disease. Several other neurotrophic factors are being commercially developed
in the US. The hope is that these compounds will limit the loss of neurons
that takes place in conditions such as Alzheimer’s disease.

The demyelinating diseases could also benefit from other research into
glial cells. For example, Raff and his colleagues at University College
London, are looking at oligodendrocytes, glial cells that wrap nerve fibres
in a sheath of myelin. The story begins with precursor cells, which divide,
multiply and then mature as oligodendrocytes, starting at around the day
of birth. Raff and Barres from Stanford established that the precursor cells
can sense the level of electrical traffic in nearby nerve fibres. If the
fibres are busy, the precursor cells keep dividing. But if the fibres become
silent, they stop. This sensitivity to electrical activity could form part
of a mechanism that matches the number of oligodendrocytes to the task in
hand – coating fibres with myelin.

How do the precursor cells ‘know’ that nearby fibres are active? Barres
and Raff suspect the message travels via a third type of cell acting as
a go-between. That go-between is another glial cell, a type of astrocyte.
From this view, the astrocyte picks up a message from busy fibres and then
releases a chemical signal that tells the precursor cells to keep dividing.
All the signs are that this chemical signal is a growth factor called platelet-derived
growth factor. Whatever the exact mechanism, the end result is a skilfully
crafted biological tissue, for which glial cells must take a large share
of the credit.

The research may have important implications for multiple sclerosis,
a disease in which oligo-dendrocytes die and nerve fibres consequently lose
their power to conduct impulses, leading to symptoms such as tingling, double
vision, vertigo and paralysis. Learning how oligodendrocytes and their precursors
are controlled – an enormously complex process – should define points where
medicine could intervene in the disease. ‘One is years away, I think, from
being able to do this,’ stresses Raff. ‘I like to think that what we’re
doing will make it possible to attack the problem rationally.’

* * *

Channels of communication

Neurons carry messages in the form of electrical pulses, which are created
by rapid flows of charged ions across their membranes. The ions cross membranes
via holes in special proteins – channels – which can be opened or shut as
required. Ion channels are at the heart of all nervous systems and researchers
around the world are assiduously documenting their properties, not least
because they are the target of many drugs.

Some channels open when they sense a change in the voltage across the
cell membrane. The sodium channels that help nerve cells conduct electrical
impulses fall into this category. In many ways, it came as a surprise when
researchers discovered that some glial cells also have sodium channels of
this kind – since they do not conduct electrical impulses in the manner
of neurons. One solution to this puzzle could be that glial cells have the
channel proteins because their role is to deliver them to neurons. Glial
cell ‘fingers’ touch neurons where the sodium channels are clustered, just
as one would expect if they acted as a delivery service.

Glial cells are also equipped with a range of ‘ligand-gated ion channels’
– channels that open when they sense the presence of a certain chemical.
Channels of this type are also essential components of neurons. Most neurons
are not in direct electrical contact and so they cannot pass their signals
on to one another in the same way as wires do on a circuit board. Instead,
they converse in a chemical language across narrow gaps – the synapses –
using neurotransmitters.

When a nerve impulse arrives at a synapse, it causes the release of
a neurotransmitter. On the far side of the gap are ‘receptors’ for the neurotransmitter
– proteins that recognise the chemical and react to it, often by changing
shape and creating ion channels. Receptors for some transmitters react more
subtly, sparking off slower, longer-lasting biochemical changes inside the
receiving cell, including effects on ion channels. The discovery that glial
cells carry many receptors for neurotransmitters suggests that they can
pick up signals from neurons – and blurs the traditional distinction between
the two types of cell.

* * *

Glial cells: the cast characters

Glial cells get their name from the Greek for glue. They account for
about half the brain’s volume and perhaps as many as nine-tenths of its
cells. They come in several varieties, the most important of which are
astrocytes and oligodendrocytes.

Astrocytes are elegant star-shaped cells from the brain and spinal
cord. They seem obsessed with neurons, occupying the spaces between them,
touching them with their long processes and surrounding their synapses.
They also make intimate contact with blood vessels and other astrocytes.
Astrocytes can be divided into various subtypes according to their appearances
and haunts.

Oligodendrocytes are the second main class of glial cells in the brain
and spinal cord. Their role is to provide an insulating sheath of myelin
around axons – the long fibres of nerve cells – and so make them into faster,
more efficient conductors. The sheath is made of several concentric layers
of cell membrane, which are wrapped around the axon a little like plastic
film around a cucumber. Axons elsewhere in the body have a similar sheath,
made in the same way by a different class of glial cells called Schwann
cells.

Because of the way axons work – by shuttling charged ions of sodium
and potassium across their membranes – their insulating coat cannot be continuous.
Every millimetre or so, there are gaps where the naked skin of the axon
is exposed to the outside world and where ions can surge in and out – the
so-called nodes of Ranvier. In the brain, astrocyte ‘fingers’ touch axons
at these delicate points.

Other glial cells include ependymal cells, which form the linings of
the brain’s internal cavities, and microglia – relatives of white blood
cells which swallow and remove debris.

Stephen Young is a freelance science writer based in Wales.