麻豆传媒

AN ORE prospector might think he was nearing the mother lode. For buried
within the grey and white matter in your head, stashed safely inside nuggets of
protein, is a surprising quantity of metal. In fact, the human brain contains
about 6 milligrams of copper, enough to print a small circuit board. Zinc, iron
and manganese are just as plentiful.

While we鈥檝e known about metals in the body and brain for decades, it鈥檚 only
recently that scientists have stopped viewing them as just 鈥渢race elements鈥� and
started to ask what on earth they鈥檙e doing there. 鈥淐alling copper or zinc in the
brain trace metals is quite silly,鈥� says Ashley Bush, a research psychiatrist at
the Genetics and Aging Unit of Massachusetts General Hospital in Charlestown.
鈥淭he brain concentrates metals better than any other tissue in the body.鈥� But to
what end?

For starters, our brain doesn鈥檛 work properly without metals. Many neurons
release zinc, copper or iron to help transmit signals across synapses. In fact,
cyanide kills by soaking up copper at the synapses. And young rats fed diets low
in iron learn poorly, while elderly people with zinc deficiencies seem to be at
greater risk of developing senile dementia. Metals are also integral to the way
brain cells defend themselves. When bound to antioxidant proteins, copper and
iron neutralise dangerous free radicals by mopping up electrons. And cells
sometimes release zinc to help fight infections.

But in the past year, metals have revealed a more sinister side鈥攁 Yin
to their antioxidant Yang. A growing number of researchers, Bush among them, now
suspect that mishandling metal in the brain is responsible for neurological
disorders such as Alzheimer鈥檚, Parkinson鈥檚 and prion diseases. If he鈥檚 right, it
will completely alter the way scientists think about these diseases and how they
go about trying to cure them. 鈥淚 think there will be a sea change,鈥� says Bush.
But it won鈥檛 be easy to prove.

It all began early in Bush鈥檚 career, when he was studying one very visible
problem with the Alzheimer鈥檚 brain鈥攕enile plaques. These clumps of protein
accumulate outside diseased neurons in the parts of the brain that control
higher cognitive functions such as judgement and memory. He happened to spot
that the primary constituent of plaque, a small protein called A&bgr;, binds to zinc
and copper, and that the brains of patients who had died of Alzheimer鈥檚
contained three to four times as much copper, zinc and iron as normal, mostly
concentrated in the plaques.

To find out whether metal had anything to do with plaque formation, Bush and
his colleagues mixed A&bgr; protein with metal ions in the lab, and left it to form
clumps. Then they added a chelator, a chemical that soaks up metals. 鈥淭he stuff
dissolved almost immediately, as fast as we could measure it,鈥� Bush says. They
tried the same experiment on mashed post-mortem brain from an Alzheimer鈥檚
patient. Again the plaques vanished like sugar in warm water.

For years, plaques had seemed the most logical cause of Alzheimer鈥檚 dementia.
Dozens of labs showed that A&bgr; was toxic to neurons, and people genetically
predisposed to get Alzheimer鈥檚, such as those who carry the ApoE4 gene,
develop more plaques at a younger age. Mice engineered to produce large amounts
of human A&bgr; developed most of the symptoms of Alzheimer鈥檚鈥攊ncluding
plaques. And now Bush鈥檚 work suggested that metals might help the plaques form.
But it soon became clear to him that this wasn鈥檛 the whole story. In some cases,
patients with the greatest dementia turned out to have few plaques.

Back in 1986, Colin Masters, a pathologist at the University of Melbourne,
proposed that Alzheimer鈥檚 might be caused instead by oxidative
stress鈥攕urplus electrons knocking about and damaging cells. Six years
later, Miguel Pappolla鈥檚 group at the University of South Alabama in Mobile
found abnormal amounts of antioxidant proteins in the brains of Alzheimer鈥檚
patients, particularly around plaques. Now Masters, Pappolla and others see
plaques more as a monument to the brain鈥檚 battles with oxidative stress than as
a source of damage. 鈥淭he plaque is like a tombstone,鈥� says Pappolla.

Finally, in December, an international team of researchers including Bush and
Masters reported how the two lines of research might dovetail. They found that
when the A&bgr; protein binds to copper it can create oxidative havoc (Journal
of Biological Chemistry, vol 274, p 37111) producing lots of hydrogen
peroxide and killing cells.

This immediately begs a question. Our cells produce A&bgr; all the time, and
copper is readily available too. So if this tiny protein is so risky, why do we
have it at all鈥攁nd why doesn鈥檛 everybody get Alzheimer鈥檚? Bush believes
the answer is that it only becomes dangerous when chemical conditions in the
cell allow it to grab too much copper. Normally A&bgr; would be safe, or even
beneficial.

That might sound strange, but five years ago, mutations in a protein called
SOD1 were linked to familial amyotrophic lateral sclerosis鈥攁n inherited
motor neuron disease. SOD1 is one of a group of proteins called superoxide
dismutases, which are powerful antioxidants. SOD1 binds copper and zinc, and
uses the metals to mop up electrons to disarm superoxide (O₂^鈥�), a
dangerously reactive relative of oxygen. But it wasn鈥檛 immediately clear how the
mutant proteins could cause the disease. In lab tests some were just as good at
mopping up superoxide as the normal form.

Then last year, a team of biochemists led by Joseph Beckman of the University
of Alabama, Birmingham, found that normal SOD1 binds up to 50 times more zinc
than any of the five mutant versions they tested. And somehow, when zinc was
missing, the copper bound to SOD1 stole electrons from other chemicals in the
cell 3000 times as fast as normal SOD1 does. It then handed over the extra
electrons to make more superoxide. Beckman thinks this sequence of events could
cause disease because extra superoxide generates peroxynitrite, the same poison
used by damaged or superfluous motor neurons to commit suicide.

Bush thinks that A&bgr; could be just like SOD1: a protective antioxidant when
properly loaded with copper and zinc, but a dangerous pro-oxidant when zinc is
lacking. He鈥檚 found many striking parallels. A&bgr; breaks down superoxide, and like
SOD1 it normally binds more easily to copper than zinc. But Bush found that
adding extra zinc, mimicking its release by cells under threat, blocks the
oxidative havoc that A&bgr; causes. The metal entombs the protein in clumps that do
no further damage. Clumps which are very reminiscent of plaques.

All this fits nicely with the discovery of a second, more toxic form of A&bgr;
some years ago. Most A&bgr; in the body is 40 amino acids long (A&bgr;40), but brains of
Alzheimer鈥檚 patients are brimming with a longer form, A&bgr;42. This is more toxic
to neurons. And as it turns out, it also binds copper much more tightly than
A&bgr;40 and is much more efficient at making hydrogen peroxide.

But there is an ironic twist. With the proper zinc balance, A&bgr;42 is also a
better antioxidant. So, says Bush, the enzyme that makes A&bgr;42 may have evolved
to provide better protection from oxidation damage. But the same protein that
defends our brains through the reproductive years sometimes bashes them up later
on in life.

The picture of Alzheimer鈥檚 emerging from this is a snowball effect. We have
A&bgr; all over our bodies our entire lives, but at some stage it begins to
accumulate in the brain. Perhaps some of it fails to load zinc properly and does
some oxidative damage. In response, neurons make more antioxidants, including
A&bgr;. These, especially the long form, spread more oxidative injury
(see Diagram).
鈥淚 think there is a vicious cycle,鈥� Bush says. 鈥淥nce A&bgr; moves from being an
antioxidant to a pro-oxidant, it can then initiate its own generation.鈥�

But what starts the snowball rolling? One possibility is acid. Oxygen
starvation can produce a condition called acidosis, as anaerobic metabolism
kicks in. The same phenomenon causes sore muscles when you exercise hard. As a
person ages, events that can cause mild acidosis in the brain become more
frequent鈥攎ild strokes, temporary drops in heart output, infections and
even head injuries. And when the pH drops, zinc hardly binds to A&bgr; at
all. 鈥淭his is our number one hypothesis,鈥� says Bush.FIG-mg22534601.JPG

Genetic predispositions to Alzheimer鈥檚 could fit into this picture well.
ApoE, for example, is a copper-binding protein. Little is known about the
metal-binding role, except that it seems to help protect cells from oxidative
damage by excess copper. ApoE4, the form associated with familial Alzheimer鈥檚,
is the worst at binding copper.

The Jekyll and Hyde protein scenario is now reverberating throughout the
field of neurodegenerative disease. For example, in prion diseases such as BSE,
scrapie and CJD, an insoluble form of the prion protein (PrP) forms clumps in
the brain. What many researchers assume, but no one knows for sure, is that
these kill neurons.

Then last year, David Brown, a biochemist at Cambridge University, showed
that PrP can act as an antioxidant when copper is bound. Cells containing PrP
are known to resist oxidative stress more easily. These results suggest a
different picture of mad cow disease than the standard cruddy brain theory,
Brown says. 鈥淚 think what leads to the death of neurons in prion disease is the
loss of antioxidant activity.鈥�

Early this year, Brown found that PrP binds not only to copper, but to
manganese. And when it does, it becomes inactive. Enzymes that chew up normal
PrP can鈥檛 digest the manganese version, a hallmark of the disease form of PrP
(The EMBO Journal, vol 19, p 1180). So it looks as though metals could
have something to do with prion disease too.

This is welcome news to Mark Purdey, a Somerset farmer with a background in
biochemistry, who over the past 10 years has analysed soil, water and vegetation
for more than a dozen metals at sites of prion disease clusters around the
world, including Iceland, Slovakia and Colorado. Only one metal was consistently
more abundant in disease-ridden areas: manganese. 鈥淚 was really blown out by
Brown鈥檚 work,鈥� says Purdey. 鈥淎fter I had done my soil tests, I read all his
papers and thought, `Wow, the two together explain what is going on.'鈥�

Purdey is testing the soil in Leicestershire near the town of Queniborough,
where a cluster of CJD cases was recently reported
(麻豆传媒, 22 July, p 3).
The tests are not complete, but he has already found a possible
source of metal in a nearby rocky outcrop that he says contains manganese oxide
in Precambrian strata. Is it possible that chronic exposure to manganese could
raise the risk of prion diseases? 鈥淚t鈥檚 not absolutely conclusive,鈥� says Brown.
鈥淭here is CJD everywhere in the world. But if there is some factor to do with
pipes or water, it could be very focal.鈥�

All this is bound to make people wonder if they should get rid of their
copper pots and avoid foods that are high in manganese. And it鈥檚 sure to
revitalise the debate about aluminium and dementia (see 鈥淲hat鈥檚 to blame?鈥�). But
that鈥檚 not necessary, Bush says. Experiments on lab animals show that a meal
high in iron or zinc has no effect on the amount of these metals in the brain,
despite high levels in the blood. The blood-brain barrier keeps them out.

What might make a difference, Bush and Brown agree, is long-term exposure to
high levels of metal, but there is currently little evidence for or against
this. If it鈥檚 true, the risk is slight. Even in the regions of Slovakia where
people were continually exposed to high levels of manganese, the rate of disease
was still only one in a thousand.

But the theory does not mean that preventing and treating neurological
diseases might be as easy as soaking up metals. A copper chelator made by Prana
Biotechnology in Melbourne is now in the first phase of clinical trials, after
promising results with mice. Antioxidants such as vitamin E are already thought
to slow the onset of Alzheimer鈥檚, but chelators go to the source of the disease.
They have potential in amyotrophic lateral sclerosis and vCJD as well.

Evidence linking metals to several other diseases is also beginning to
trickle in. Iron deposits have been found in the nigral cells that die in
Parkinson鈥檚 disease. And iron seems to cause the Parkinson鈥檚 protein a-synuclein
to aggregate. The protein mutated in Friedrich鈥檚 ataxia fails to export iron
from mitochondria, the powerhouses of the cell, leaving neurons short of energy.
And cataracts form when the lens protein alpha-crystallin becomes crosslinked,
possibly by copper-generated superoxide.

Ironically, if Bush is right, approaches to Alzheimer鈥檚 treatment based on
the idea that plaques cause the disease won鈥檛 work. For example, the antibody
against A&bgr; made by Elan Pharmaceuticals near San Francisco makes plaques in
mouse brains vanish. But if they鈥檙e just a tombstone that may not matter. Still,
Bush realises his work isn鈥檛 going to change these strategies overnight. 鈥淚 hope
Elan鈥檚 approach works,鈥� he says. 鈥淚t鈥檚 logical, whereas the idea that an
antioxidant can also be bad for you is counterintuitive.鈥�

A SUSPECTED link between aluminium in the diet and Alzheimer鈥檚 disease
prompted some people in the late 1980s to discard their aluminium pans and have
their water checked. And worries about an increase in neurological disease in
Camelford in Cornwall after the 1988 release of aluminium sulphate into the
water supply still reverberate. But the link remains inconclusive. 鈥淲e know one
thing for sure, aluminium is toxic,鈥� says Stephen Bondy, of the University of
California at Irvine.

Aluminium seems to depend on iron for its toxic effects, Bondy says.
Aluminium forms particles in the brain, which provoke the brain鈥檚 immune
defences. Cells release iron in an attempt to oxidise and disarm the particles,
but they persist and the cells end up self-destructing. Whether this has
anything to do with Alzheimer鈥檚 is the unanswered question. 鈥淲e are exposed to
aluminum our whole lives,鈥� Bondy says. 鈥淚t鈥檚 not likely to be the sole factor in
Alzheimer鈥檚 disease, but I believe it may be one of several.鈥�