¿ìè¶ÌÊÓƵ

Research on the three-pound ball of tissue that makes us conscious is
booming. The US government spends $1.2 billion per year on neuroscience,
and the European Community has just launched a ‘Decade of the Brain’ to
encourage Europe’s 12000 neuroscientists. To explain the excitement, we
here begin a four-part series examining how researchers are at last getting
to grips with the workings of the human brain

Decades after the arrival of modern medicine, the brain remained stuck,
stubbornly, beyond the reach of science. Even today it retains the aura
of an intimidating ‘black box’, an organ whose devastating malfunctions
we can rarely understand, still less remedy. Personality, memory and even
consciousness itself may all suffer terribly if only a fraction of the billions
of neurons from which the brain is intricately ‘wired’ die or lose their
ability to communicate normally. The question is: how can we intervene?

Never have neuroscientists been so anxious to find clues. Over the next
30 years there is likely to be a massive increase in the number of people
suffering from brain diseases associated with old age. The risk of developing
Alzheimer’s disease – a fatal condition in which neurons in many different
parts of the brain, but particularly the cortex, shrink or die – increases
sharply over the age of 65 and afflicts one in five people over 80. Already
there are estimated to be 4 million people with Alzheimer’s disease in the
US, and half a million in Britain. With the populations of the Western world
ageing, these figures could rise by 25 per cent in the next three decades.

But brain disorders are not the exclusive preserve of the old. Stroke,
which can leave people incontinent or unable to communicate, can afflict
people in their forties and fifties, and at least 1 in 200 people in Britain,
young and old alike, suffers from epilepsy . On top of this comes a new
threat to millions of young people worldwide: HIV which can infect brain
cells and cause a form of dementia. Some doctors fear that dementia will
become a common symptom of HIV infection as treatments for other aspects
of AIDS succeed in prolonging lives.

The search for treatments is beset with difficulties. One is the problem
of diagnosing brain diseases. Alzheimer’s disease, for example, is difficult
to distinguish from other types of dementia. Psychological tests and brain
scans can provide very strong clues, but ultimately a postmortem is the
only way of confirming the disease. Another difficulty is getting drugs
into the brain across the blood-brain barrier.

Inevitably, there are also political problems, such as the highly charged
debate in the US over the use of fetal tissue for research into restoring
degenerating brains. Fearing a backlash from the powerful anti-abortion
lobby the Bush administration banned the use of government money for such
research in 1988, although some private funding is available. John Sladek,
director of the Neuroscience Institute at the Chicago Medical School, has
called the ban ‘the biggest impediment to clinical progress that we can
imagine’. In Europe, there is no blanket ban on the use of fetal tissue,
although the approval of ethics committees is always required. Without the
influence of America’s Bible belt, there is less public controversy too.

But politics and morals aside, fetal neurons are uniikely to vanish
from the scientific agenda for treating brain disease. The reason is simple:
unlike most mature neurons, fetal neurons can divide and grow new fibres.

Another political hot potato is animal rights. Although the advent of
noninvasive techniques such as brain scanning has extended the range of
studies that can be done on human subjects, most neuroscience experiments
still involve procedures that can only be carried out on animals – making
surgical lesions, for example, and probing the behaviour of neurons using
microelectrodes. In the past, neuroscientists have often been singled out
by animal rights activitists, and many fear that future campaigns will hamper
their work even further.

BAFFLING JIGSAW

None of this is exactly reassuring. Yet there are many common brain
disorders for which research is now providing clues to new treatments. With
recent advances in molecular genetics have come the first pieces in the
baffling biological jigsaw of Alzheimer’s disease. And over the past two
years, improved scanning techniques have made it possible to pinpoint damage
in diseased brains, monitor recovery and even take ‘snapshots’ of blood
flow in the brain as it processes information.

Gene therapy – where ‘therapeutic’ genes are implanted into specific
cells within the body – may provide a new way of attacking brain tumours
and other disorders . And, at least in laboratory animals, it no longer
seems that damaged brains are beyond repair: transplants of fetal neurons
into the brains of adult animals with artificial lesions seem to restore
some functions.

But some of the biggest advances have come from studies of stroke and
head injury which have led to a new understanding of what happens to neurons
when they become short of oxygen. Since the mid-1980s, scientists have become
increasingly convinced that much of the harm is done by a cascade of biochemical
reactions.

A stroke can be caused either by a ruptured blood vessel or a clot that
blocks the blood flow in an artery. In the immediate area of damage, neurons
become starved of oxygen and die. But in the surrounding area, known as
the penumbra, damaged neurons may be rescued if they are treated in time.
Here, neurons respond to the sudden loss of oxygen by discharging massive
amounts of an amino acid called glutamate. A key ingredient in Chinese cuisine,
glutamate is also one of the brain’s main neurotransmitters, and is vital
to a whole range of brain functions, including the storage of memory

Normally neurons release glutamate at synapses in order to stimulate,
or ‘excite’, other neurons – hence the buzz afrer a Chinese meal. But when
excess glutamate is released, as it is during a stroke, the result is chaos.
Neighbouring neurons become overstimulated and respond by opening channels
in their membranes, which let calcium ions and sodium ions flood in. The
calcium wreaks havoc with the neurons’ internal ‘scaffolding’, and the damage
may be compounded by an abnormally large influx of chloride ions and water.
The process may also activate enzymes that rupture cell membranes and help
to unleash free radicals – reactive molecules that damage DNA and other
essential cellular components.

Recently, studies of excitotoxicity have mushroomed, and pharmaceuticals
companies are hunting for compounds that can block – or ‘antagonise’ – the
receptors that bind glutamate at the surfaces of neurons. A key hurdle has
been characterising the molecular architectures of these receptors. Neurons
in different parts of the brain tend to carry different types of glutamate
receptor each of which seems to be assembled from a distinctive set of smaller
proteins, or subunits.

Most interest centres on the so-called NMDA receptor a special type
of glutamate receptor that responds not only to glutamate but to a synthetic
analogue, N-methyl-D-aspartate. Widely considered to be one of the brain’s
most important molecular components, the NMDA receptor appears to act as
a trigger for a host of neural events, from the organisation of the visual
system through to memory and – after injury it seems – the poisoning of
cells. Last year Japanese researchers won the race to isolate the first
example of a gene encoding a subunit of an NMDA receptor.

STROKE DRUGS

Pharmaceuticals companies such as Merck Sharp & Dohme, Sandoz, Novo
Nordisk and Ciba-Geigy have developed a number of compounds that antagonise
NMDA receptors in animals and provide some protection after an artificial
‘stroke’. Some of these antagonists, such as Ciba-Geigy’s CGS1975, have
already begun safety trials in stroke patients and people with head injury
in the US and Britain. NMDA antagonists are also being tested as a possible
treatment for epilepsy . Other subtypes of glutamate receptor including
the so-called AMPA receptor have also been implicated in the stroke cascade.
Drugs to block these receptors are being tested in animals.

Another approach is to block the free radicals that may be released.
The pharmaceuticals company Upjohn is testing a free-radical ‘scavenger’
called Tirilazad in patients with head injury in the US. This drug belongs
to a class of substances known as lazaroids, afrer the Biblical character
Lazarus.

It is too early to say how effective these various compounds will be
as treatments for stroke. But Dennis Choi, a leading researcher in the field,
at Washington University St Louis, is cautious. The drugs are most likely
to work if only a small area is damaged, he says, and even then they will
only offer damage limitation. ‘Stroke is not a very subtle disease, it is
like a bomb going off in your brain,’ he adds. The damage may be messy and
widespread, calling for much broader repair than one or two receptor antagonists
can deliver.

A particularly important problem is knowing when to administer stroke
drugs. In rats that have suffered an artificial stroke, glutamate receptor
antagonists tend to fail unless they are given within two hours. But for
humans, we do not know whether the equivalent window is 4 hours or 24 hours.
In Britain, stroke is often not treated as an emergency in the same way
as heart attacks. Can our GPs, ambulance services and hospitals gear up
to treating stroke differently?

There are other unanswered questions. What about the possible side effects
of these drugs? One of the first antagonists for NMDA receptors, MK-801,
developed by Merck Sharp and Dohme, received huge publicity initially. But
it has since been tarnished by the discovery that it may cause a rapid fall
in blood pressure if it is given intravenously. Perhaps more worryingly
some NMDA receptor antagonists trigger abnormal muscle movements in rats,
and from laboratory experiments at least, it looks as if the compounds might
interfere with processes involved in learning and memory.

To make the best use of these drugs in future, doctors will need to
be able to ‘see’ precisely which areas of the brain have been damaged. Fortunately
magnetic resonance imaging (MRI) and positron emission tomography (PET)
mean that the brain is no longer hidden from view by the skull. These two
‘artificial eyes’ complement each other with PET producing mainly images
of the brain’s activity while MRI reveals its structure. The most powerful
images come from combining the two types of so information. Both techniques
have improved so dramatically in recent years that damage can now be traced
to the territories of individual blood vessels.

PET is mainly used to measure the flow of blood through the brain. It
works on the principle that the greater the blood flow in a particular area,
the more active are the area’s resident neurons. The patient being scanned
is given an injection of a radioactively-labelled nutrient, normally glucose,
which can be tracked as it flows through the brain.

Richard Frackowiak, who heads a team specialising in scanning at the
Medical Research Council’s Cyclotron Unit at the Hammersmith Hospital, London
has been analysing PET images to monitor the way the brain reorganises itself
to compensate for lost circuits after a stroke. In one study the researchers
examined people who had suffered a stroke affecting their ability to move
one side of their body. Compared with healthy people, the stroke patients
had clearly suffered a loss of blood flow in certain areas of the brain
– areas that included the sites where tissue had been damaged. Further differences
emerged when the researchers scanned the subjects as they moved, or attempted
to move, each of their hands in turn. In the stroke patients, movement of
the hand that had initially been partially paralysed by the stroke activated
larger networks of neurons scattered across a wider area of the cortex than
movement of the same hand by the healthy subjects. Most dramatically the
PET images for movements of the ‘recovered’ hand reveal intense activity
in both hemispheres of the brain. Some of these areas, says Frackowiak,
are normally only activated when people perform relatively complex hand
movements.

All this strongly suggests that, faced with the need to learn a task
again, the adult brain can to a limited extent, ‘rewire’ itself. Or as neuroscientists
would say it is inherently ‘plastic’.

POWERFUL IMAGES

Techniques such as PET may appear to be liberating researchers from
the ‘dark ages’ of neuroscience, but there are still drawbacks. PET involves
introducing a radioactive compound into the brain. The radiation level is
low but the WHO recommends that people should have no more than one PET
scan every five years. The WHO recommendation has no legal status, however
and many countries including Britain do not follow it, allowing more frequent
scans on some patients. Another frequent criticism is that the pattern of
blood flow which PET detects provides only a relatively crude picture of
neural activity. Its spatial resolution, for example, is much too low to
reveal the activities of individual cortical columns, the basic functional
units of the cortex.

Nevertheless, the technique is proving useful – and not just for acute
disorders such as stroke and head injury Ben Freedman at the University
of Sydney in Australia is planning to use PET and other scanning techniques
to see where drugs against epilepsy and schizophrenia travel in the brain.
And there are diagnostic applications, too. PET images of people with Alzheimer’s
disease generally show a reduction in brain activity that is most severe
in the parieto-temporal region of the cortex. Sadly only a handful of specialist
hospitals have the equipment needed to exploit this discovery in diagnosing
Alzheimer’s disease.

TERRIFYING LOSS

Of all the degenerative conditions, AIzheimer’s disease is by far the
most common. Sufferers are faced with a terrifying loss of memory – particularly
for recent events – and become disoriented in time and space. They tend
to be confused, wander and, in severe cases, may become aggressive and paranoid.
One sufferer quoted by the British Alzheimer’s Disease Society described
her mind as being ‘like a dark thunderstorm’.

The brain of a person who has died from Alzheimer’s disease contains
two characteristic features: plaques, made up of a core of a protein called
amyloid and surrounded by dying nerve fibres; and tangles, made up of fibrous
strands known as paired helical filaments. Amyloid is formed when a larger
protein chain, the amyloid precursor protein (APP), is cut by enzymes in
a particular fashion. The paired helical filaments contain an aberrant form
of another protein, tau.

For years, researchers studying Alzheimer’s disease quarrelled over
whether amyloid or tau was more important, and whether the amyloid deposits
were cause or effect. Then in February 1991 came a discovery that made a
causal role for amyloid seem far more likely. By studying a large family
with a rare form of inherited. Alzheimer’s disease, a team headed by John
Hardy and Alison Goate, then at St Mary’s Hospital Medical School, London,
found a mutation in the gene encoding APP on chromosome 21. The same mutation
has now been found in five more families with inherited Alzheimer’s disease,
and further mutations in the APP gene have been linked to the disease. ‘The
mutations tell you that amyloid is central to the disease process,’ says
Goate.

What can we make of this? Only a minority of the families with the inherited
form of the disease carry the mutations identified so far. And only a minority
of all cases of Alzheimer’s disease are inherited; the rest follow no obvious
pattern. One explanation is that in some people, amyloid accumulates gradually
over a lifetime as a result of continuous exposure to a mild toxin in the
environment. Other theories implicate viruses and head injury as triggers
for the build-up of amyloid.

As with virtually all theories about Alzheimer’s disease, however there
are catches. The biggest is whether amyloid really is toxic to neurons.
Bruce Yankner and his colleagues at Harvard University found last year that
amyloid injected into rats’ brains prompted neurons to break down, but teams
elsewhere have detected no such damage to cells.

Meanwhile, the pharmaceuticals industry is hotly pursuing any leads
it can find. With a market set to expand as far ahead as anyone cares, a
wonder drug would be an income for life. The first generation of drugs for
Alzheimer’s disease have yet to match the hype that heralded their arrival.
These drugs, such as tacrine, physostigmine and eptastigmine, are intended
to stop the decline in a person’s memory by boosting cholinergic neurons
(so called because they produce the neurotransmitter acetylcholine). The
drugs work by inhibiting an enzyme, acetylcholinesterase, which breaks down
acetylcholine.

The cholinergic neurons in the basal forebrain play a part in memory,
and initially scientists thought they were particularly badly affected in
Alzheimer’s disease. Now it is clear that many different populations of
neurons are affected. The acetylcholinesterase inhibitors do produce a slight
improvement in mental function, but it is short-lived, and the compounds
are toxic to the liver. At best, say critics, this strategy is like shutting
the door after the horse has bolted. In future, researchers agree, it would
be preferable to develop drugs that prevent damage rather than simply boosting
the effectiveness of one particular neurotransmitter. One way for example,
might be to block the build-up of amyloid.

EXPLOITING PROTEINS

Another is to exploit natural proteins known as nerve growth factor
(NGF) and brain-derived neurotrophic factor (BDNF). Experiments in animals
by Franz Hefti at the University of Southern California, Los Angeles, and
Michael Sofroniew at the University of Cambridge suggest that these growth
factors help cholinergic neurons to survive. Can they be exploited to prolong
the life of degenerating nerves?

Lars Olssen at the Karolinska Institute in Sweden has tested this notion
with an experiment on a patient suffering from Alzheimer’s disease. He infused
NGF into the ventricles of the patient’s brain using a small tube inserted
through a tiny hole in the skull. The treatment appeared to produce no serious
side effects and brought some improvements, but scientists agree that studies
will be necessary on a much bigger scale before we can conclude anything.
According to Hefti, a larger trial, involving between 50 and 100 people,
may be in the pipeline in the US. The trial would cost some $50 million,
he says, and the government may be prepared to fund part of it.

But the infusion of large amounts of growth factor into the brain is
a comparatively crude idea. What is more, growth factors may not be good
for all neurons at all times. In certain circumstances, they may actually
hasten the disintegration of brain tissue. Another possible problem is that
neurons might be stimulated to make unwanted connections. Hefti accepts
that this is ‘a concern’, but there is no evidence that it has yet happened
in any of his animal experiments.

His team is now looking for more refined ways of manipulating growth
factors. Recently, working on elderly rats, the researchers identified a
compound called K-252b, which inhibits NGF and BDNF but spares all other
growth factors.

But to a few doctors – and many more carers of people with Alzheimer’s
– the scientists’ pursuit of memory-enhancing drugs or growth factors may
be missing the point. According to several surveys, the symptoms that are
most difficult to live with are aggression and the tendency to wander, not
memory.

While research into Alzheimer’s disease has been in the limelight, scientists
have not forgotten Parkinson’s disease. In this condition, the neural degeneration
is confined to a small cluster of neurons buried deep in the brain, called
the substantia nigra. These neurons use the neurotransmitter dopamine to
communicate with other areas of the brain.

No one knows why these cells fail in people with Parkinson’s disease,
but a popular theory is that they are poisoned, slowly, by one or more weak
toxins in the environment. The theory holds that genetic defects in certain
individuals might make them relatively inefficient at detoxifying natural
compounds that are weakly toxic, such as certain sulphur compounds in vegetables.
Recent support for this theory has come from research showing that people
with a mutation in a gene encoding a ‘detoxifying’ enzyme face a slightly
higher risk than average of developing Parkinson’s disease.

But, with the causes of this disorder still elusive, Stephen Dunnett
at the University of Cambridge is concentrating on research that might lead
to treatments. He has transplanted dopamine-containing neurons from the
brains of monkey fetuses into adult monkeys with Parkinson-like symptoms,
whose dopamine cells had previously been destroyed. The rationale is that
grafts of fetal brain cells should continue to grow and develop, unlike
adult brain cells, and so replace the damaged or dead neurons.

To produce the symptoms of Parkinson’s disease, the researchers inject
a toxin which destroys most of the substantia nigra on one side of the brain,
leaving the other side unharmed. The animal’s ability to control its movements
is thus lost on only one side of the body and the unaffected side acts as
a good control for the experiment.

Dunnett’s early results seem very encouraging. Treated monkeys, for
example, were able to use their hands for dexterous tasks such as reaching
into a tube to pick up pieces of banana. Moreover, postmortem examinations
reveal that the transplanted cells grow new fibres and make synaptic contacts
with the surrounding neurons. And most importantly for the sceptics, says
Dunnett, the degree of recovery is proportional to the number of new nerve
connections.

Compared with previous experiments, this is success. The crucial factor,
maintains Dunnett, is to take the fetal grafts at the right stage of gestation.
This is the point where the neurons have just undergone their last division
and so are specialised, but before they have grown any fibres.

While animal studies like these continue to provide clues, a handful
of clinicians has gone further and used fetal cells as an experimental treatment
for patients with severe Parkinson’s disease. Although more than 100 patients
worldwide have received fetal grafts – in the US, Sweden, Cuba and Britain
– only a tiny minority of those can be said to have benefited. Olle Lindvall
and colleagues at the University Hospital in Sweden have reported treating
six patients, four of whom showed significant improvement.

Unlike the other groups attempting transplants, the Lund team has produced
data that convince at least some sceptics. Brain scans on their patients
showed they were producing significantly more dopamine than before the transplants,
and their symptoms were greatly reduced as long as 18 months later. Experimental
though this treatment may be, it is one of many that suggest the dark ages
of the brain may indeed be coming to an end.

* * *

1: Treating epilepsy

The term ‘epilepsy’ covers a number of conditions, some of them still
the subject of fierce debate among scientists. Although many sufferers lead
virtually normal lives with treatment, the disorder can be extremely debilitating.

During a fit, neurons fire in an abnormally synchronised and rhythmic
fashion, preventing the brain from functioning normally. There are two principal
types of seizures: generalised, in which both halves of the brain are affected
symmetrically, and partial, in which a limited area of the brain is affected.

The most common type of generalised fit is the so-called ‘tonic-clonic’
seizure, in which the sufferer may collapse with muscular spasms. This form
affects some 80 per cent of sufferers. The second most common generalised
form of epilepsy, usually found in children, is the type known as absence,
in which the sufferer appears to ‘go blank’.

Partial seizures, such as temporal lobe epilepsy, may disturb consciousness,
or cause muscle twitching if the motor cortex, which controls movement,
is affected. People who suffer from epilepsy have a variety of warning signals,
such as a sensation in the gut or a phantom smell.

In a minority of cases, epilepsy has a clear genetic cause. For example,
juvenile myoclonic epilepsy, a form of the disease which accounts for more
than 10 per cent of affected children, is thought to be caused by a gene
located on chromosome 6.

A common cause of epilepsy is damage to a particular area of the brain,
from injury, a tumour, an abnormality in development or an infection such
as malaria. MRI images and PET scans show that during epileptic fits there
may be a marked increase in metabolic activity in the affected region of
the brain.

In animals with lesions that cause epilepsy, drugs known as NMDA antagonists
– which prevent neurons from firing in response to a neurotransmitter called
glutamate – appear to block seizures. Sandoz, a Swiss company, has developed
an NMDA antagonist called D-CPPene which is now in clinical trials in patients
with epilepsy. If such trials are successful, epilepsy sufferers could be
among the first patients to benefit from NMDA antagonists. However, scientists
need to know more about the side effects of the drugs in long-term use.

* * *

2: Gene therapy to heal the brain

Future treatments for brain diseases may rely not on conventional drugs
but on gene therapy. The technique is in its infancy and has several safety
hurdles to jump. But many neuroscientists see it as a powerful way of delivering
treatments to the site where it is needed in the brain.

Kenneth Culver and his team at the National Cancer Institute at Bethesda,
Maryland, have received approval to test a form of gene therapy in a small
group of patients with inoperable brain tumours. Most gene therapy aims
to introduce a healthy copy of a missing or malfunctioning gene into cells
affected by disease. However, the aim with gene therapy for cancer is to
destroy the cancer cells.

Culver’s strategy is to implant a new gene into brain tumour cells that
turns them into sitting targets for a drug. His team has used a retrovirus
that infects mice to carry the gene into the animals. The gene comes from
the herpes simplex virus and encodes an enzyme, thymidine kinase-the target
of ganciclovir, a drug used to treat herpes infections. By blocking the
enzyme, the drug disrupts the virus’s life cycle and kills both the infected
cell and the virus. Culver reasoned that if he could get both thymidine
kinase and ganciclovir into tumour cells in the brain, the cells should
die.

In many gene therapy experiments, scientists remove the cells they want
to alter and put them back into the body with the new gene added. But this
is impossible with brain cells. So in his experiments with rats, Culver
used mouse cells that had already been infected with the altered mouse retrovirus,
and injected these into the animals’ brains.

The injected cells produced new virus particles which went on to infect
the brain rumour cells. Healthy cells do not become infected because, unlike
tumour cells, they do not divide. When Culver used ganciclovir to treat
14 rats with tumours, the tumours disappeared in 11 of the animals.

As with all gene therapies, there are concerns about the safety of the
technique, such as the possibility that the mouse retrovirus could cause
disease many years later. But for patients who are terminally ill, the potential
benefits may outweigh the risks.

Gene therapy could also come to the aid of people with Alzheimer’s disease.
Fred Gage at the University of California, San Diego, is developing a strategy
to protect the cholinergic system, a population of neurons involved in memory
which deteriorates in people with Alzheimer’s disease.

Gage’s approach exploits a substance known as nerve growth factor which
nourishes neurons. He has found that skin cells taken from ageing animals
can be infected with a retrovirus carrying the gene for NGF and then implanted
near to the animals’ cholinergic system. The skin cells survive and secrete
NGF, encouraging the survival of neurons that normally die as the animals
age.

In the laboratory, Gage has shown that skin cells taken from elderly
people are equally able to take up the NGF gene and secrete the growth factor.
The technique could be used to treat patients in the foreseeable future,
says Gage. Its advantage over other methods of delivering NGF to the brain,
he says, is that the neurons are more precisely targeted. The alternative
– large doses of NGF infused into the brain’s ventricles – may have unwanted
effects. The researchers hope to test a similar form of gene therapy on
people with Parkinson’s disease. The goal is to insert a gene for the enzyme
tyrosine hydroxylase into skin cells and then implant these into the brain,
near to a region known as the basal ganglia. The enzyme should help the
cells to produce dopamine, the neurotransmitter that is deficient in people
with Parkinson’s disease. Already animal experiments have shown that implanted
cells can survive in the brain and make dopamine.