|
There's more to the mind than neural networks. Messages also percolate through the soup-like fluid bathing the brain. Alison Mitchell reports
|
IF YOU WERE TOLD your brain looked distinctly foamy, you might, understandably, feel rather worried. But take heart--it's quite normal. Even in the healthiest people, an incredible 20 per cent of the brain is completely devoid of brain cells. Rather than forming a solid mass, the cells are interspersed with a convoluted network of fluid-filled spaces and cavities. Yet nobody is quite sure why. Is a whole fifth of the brain really going spare?
A small group of researchers is busy trying to convince the world that there's a perfectly good reason for all that seemingly wasted space. They believe it's part of a brain-wide communications system that complements the electrical impulses which hop from one neuron to the next via the synapses. There's a growing feeling that nerve cells can communicate with large regions of the brain by releasing chemicals into the watery spaces outside the cells. Not only does this soup of chemicals drifting around the brain seem perfectly suited to influencing such things as wakefulness and mood, they say, but it may explain things as diverse as the pleasure we get from eating chocolate, and why the symptoms of Parkinson's disease take so long to develop.
The conventional model of the brain is of individual neurons "wired up" to other neurons at junctions called synapses. These tiny gaps are 30 000 times smaller than a pinhead. The nerve signal jumps from one neuron to the next across the synapse by triggering the release of a
neurotransmitter, and the signal is rapidly picked up by receptor molecules on the other side of the gap.
But Kjell Fuxe from Sweden's Karolinska Institute in Stockholm and Luigi Agnati of the University of Modena, Italy, believe that not all nerve signals follow this traditional route. They now think that the nerve cells use the fluid-filled spaces to communicate in an entirely different way--a process they have christened "volume transmission". They imagine this alternative form of signalling as being less like a private conversation between pairs of neurons and more like a radio broadcast. Signals can go just about anywhere, and can be picked up by any properly tuned receiver.
First of all, says Fuxe, volume transmission neatly explains why neurotransmitter receptors aren't always where you'd expect them to be. In 1986 Fuxe and Agnati thought they had spotted receptor sites for neurotransmitters much further than a synapse-breadth away from where the neurotransmitter is released. This suggested to them that neurons might be able to chat with other neurons that they weren't directly wired up to.
They've gradually become more certain of this mismatch. Last year, Fuxe, Agnati and their colleagues used two types of fluorescent molecule to label a neurotransmitter called neuropeptide Y and one of its receptors so they could map their distributions in slices of rat brain. Amazingly, they found that the receptors were sometimes several millimetres, or a million times the width of a synapse, away from the source of the neuropeptide Y. There's no way that synaptic transmission could work effectively on this sort of scale, but Fuxe and his colleagues felt that volume transmission certainly could. Either the neurotransmitter would leak out of a synapse, they say, or it could be released from different parts of the neuron and travel around passing signals to other groups of nerve cells.
Weird positions
Receptors for the neurotransmitters serotonin and dopamine have also been spotted away from their most likely haunts. Last year, Feng Zhou from the Indiana University School of Medicine in Indianapolis and his collaborators found serotonin receptors on axons, the elongated part of a neuron that transports the electrical nerve impulse. You wouldn't expect to find receptors on an axon, they say, unless serotonin could travel there by volume transmission. And Anders Jansson, Fuxe's colleague at the Karolinska Institute, described last month in the journal Neuroscience (vol 89, p473) how he found two types of dopamine receptors well away from synapses.
But it's not just receptors in weird places that support the idea of volume transmission. There's also evidence from the transmitter molecules themselves. During the past ten years it's become clear that nitric oxide and carbon monoxide are active signalling chemicals in the central nervous system. As both are gases, it would be virtually impossible to confine them to the synapse, and so they seem perfect candidates for "volume transmitters".
Nitric oxide may travel a tenth of a millimetre in the five seconds or so that it's active. In this short space of time it would encounter somewhere between 10 and 50 neurons. But, as Fuxe points out, nitric oxide doesn't quite fit his general model of volume transmission, because it passes straight through cell membranes, rather than following the extracellular spaces. Even so, he believes that general acceptance of gas signalling in the brain has given the theory a substantial boost.
Unlike gases, the larger neurotransmitter molecules would have to stick to the convoluted spaces. But can they diffuse through the brain far enough and quickly enough to be of any practical use? Charles Nicholson from the New York University Medical School certainly thinks they can.
He and Eva Syková from the Institute of Experimental Medicine in Prague have been measuring the sizes of the spaces between the cells and the rate that neurotransmitters move across the brain. Because they diffuse in random directions, they're unlikely to get very far before they bump into neurons or other molecules. Nicholson and Syková estimate that an average signalling molecule would diffuse through the brain about two-and-a-half times more slowly than through water. In other words, it could cover about three millimetres per hour--much slower than nitric oxide but, according to Nicholson, easily fast enough to relay useful signals. "In that time a signal could cross a major part of a cortical layer," he says. "It could easily reach an entire brain region."
Whisked away
And there's yet another way signals could get around the brain that could extend the volume transmission route and carry signals even more widely and more rapidly. The brain and spinal cord are bathed in cerebrospinal fluid, or CSF. The fluid circulates from the three interconnected chambers in the centre of the brain, the ventricles, towards the base of the brain, and onwards into the spaces between the skull and cortical surface. It bathes the folds of the cortical surface, and is eventually absorbed by blood vessels near the front of the brain. Release a substance into the CSF and it could circulate throughout this system in a matter of minutes, says Nicholson. The process would be helped along by the pulsing of the blood vessels.
Molecules might congregate near the brain surface or pass out of the extracellular spaces and be whisked away by the CSF. The signals could then target sites well away from where they originate, explains Nicholson. Hormones in the bloodstream, CSF transmission and volume transmission could all be seen as logical extensions of one another, he says.
But in spite of the compelling nature of the ideas and evidence, not everybody is convinced that the foamy spaces are important signalling channels. Many are content to picture the extracellular space as completely passive--just a soup of nutrients, rather than a soup of meaningful messages. Why bother to wire up the brain with a precise network of synaptic connections, only to circumvent them by a slow and unpredictable method of signalling?
"I have difficulty envisaging a use for such signalling," says Robert Malenka, a neurobiologist from the University of California at San Francisco. "It would be hard to make it at all specific." But Malenka freely admits that his view of research is very "synaptocentric" and, according to Nicholson, this reaction is not uncommon. Nicholson believes that many researchers who concentrate on synapses and electrical signals in the nervous system interpret the chemicals that leak into the extracellular spaces as simply noise--noise that the central nervous system is designed to filter out and ignore.
Others accept that neurotransmitter molecules leak out of the synapse, but don't think that they can travel very far. David Attwell from University College London points out that the distance covered by the wandering neurotransmitters may depend on how far they get before other molecules mop them up and recycle them. This would mean they are unlikely to provide a reliable signalling channel.
To convince him that volume transmission actually takes place, Malenka says he would need to see it happen. In other words, some experiment that showed a transmitter substance being released from a specific point in the brain, then travelling several millimetres to a receptor in a quite different area. Although the neuropeptide Y studies hint at this, they are only a static snapshot. Nicholson admits that it will be hard to get more dynamic evidence because, at the moment, there is no way to watch molecules in action in real time across the whole brain. But there are some global patterns of activity for which it's hard to imagine a practical alternative to volume transmission.
Take Parkinson's disease, for example. Sufferers seem to lose neurons in a region of the brain called the substantia nigra, which normally supplies the neighbouring striatum region with dopamine--a neurotransmitter that initiates and controls signalling to the muscles. The strange thing is, patients don't develop symptoms until a massive 80 per cent of the dopamine-producing cells have been lost. How can signalling keep going for so long?
A group led by Jay Schneider at the Hahnemann University School of Medicine in Philadelphia has been trying to find out. They recreated the symptoms of Parkinson's disease in cats by injecting a chemical that inactivates some of the dopamine-generating neurons. They then stimulated dopamine release. Naturally, the inactivated neurons produced no dopamine but healthy neurons nearby became super-active--as if they were pumping out more dopamine to compensate for the crippled cells.
By exporting extra dopamine to the striatum, unaffected neurons in the substantia nigra could maintain the status quo for some time. But because most of the dopamine-producing neurons in the cats were inactive, Schneider's team thinks the dopamine must bypass the usual synaptic route to get to the receptors in the striatum. A case for volume transmission? This diffuse method of signalling could also explain why the symptoms of Parkinson's disease diminish when just a few healthy cells are transplanted into the diseased brain.
But signals can be even more widespread than this. Neuropeptide Y can stimulate appetite, alleviate anxiety and can even reduce the desire for a gin and tonic. This single transmitter has so many different effects because its receptors are found all over the brain. And the easiest way for neuropeptide Y to signal over such widespread regions would be by volume transmission.
The feel-good factor
Another widespread signal may alter our mood. Many people find that their mood improves when they eat chocolate. Why? Because chocolate contains an amino acid called tryptophan, which is the precursor for serotonin--a feel-good neurotransmitter. From the chocolate bar, via the stomach, tryptophan passes into the bloodstream and then into the brain's extracellular fluid. According to Dick Wurtman from Massachusetts Institute of Technology, tryptophan itself isn't active at synapses, so it cannot be classed as a volume transmitter. But the boosted levels of tryptophan rapidly increase the number of neurons that release serotonin, as well as the number of serotonin molecules released by each neuron. And there is, he adds, "abundant evidence" that the neurons are releasing this serotonin outside the synapse.
Sleep is another good example which, according to Nicholson, illustrates why the brain needs volume transmission. A molecule called prostaglandin D2 sends you off to sleep, whereas its counterpart prostaglandin E2 promotes wakefulness. The enzyme that produces prostaglandin D2 comes mainly from non-neuronal cells in the brain, and may reach its target by diffusing through the CSF and the extracellular fluid in which the neurons bathe.
Twisty logic
Nicholson contends that volume transmission is "largely involved in setting the thresholds of populations of cells over long periods of time". In other words, it's more likely to help regulate the level of activity in a neural circuit than determine what that activity is. This means the slow process of volume transmission could work happily alongside more rapid signals from synaptic communication.
But if volume transmission is so beautifully logical, why has it taken so long for it to come to the fore? In fact, almost the same controversy was raging at the turn of this century. Then, Santiago Ramón y Cajal, a Spanish neurobiologist, stated that the brain was made up of individual neurons, which communicated with one another only at specific points. But Camillo Golgi, an Italian anatomist, believed the brain was a continuous network, with impulses covering much longer distances. Golgi's theory was eventually crushed in the 1950s, when the electron microscope revealed the structure of the synapse.
Now, researchers are realising that synaptic transmission doesn't necessarily rule out a slower, more global alternative. The gradual build-up of evidence in favour of volume transmission takes us right back to Golgi's way of thinking. As Nicholson explains, "as research on synaptic transmission has become ever more detailed, evidence for an entirely different mode of information transmission has been quietly accruing in the background". Quietly maybe, but this message is becoming hard to ignore.
ALISON MITCHELL is an associate editor at Nature
Further reading:
"The emergence of the volume transmission concept", by Michele Zoli and others, Brain Research Reviews, vol 26, p136 (1998)
"Signals that go with the flow", by Charles Nicholson, Trends in Neurosciences, vol 22, p143 (1999)
Graphic: Mind the gap
From New Scientist, 13 March 1999
|
