York
HOW far would you go to improve your memory, or to preserve it as old age or
dementia take their toll? The more we understand about how memories form, the
more hopeful researchers become that memory-boosting drugs are feasible. But
it鈥檚 starting to appear that improving memory could come at quite a
price鈥攇reater sensitivity to pain. It seems the nervous system takes a
painful reminder rather too literally.
Wherever they look, researchers investigating how the nervous system responds
to pain have been discovering uncanny links with the way we create new memories.
Many now believe that primitive organisms鈥� ability to detect a particular
painful stimulus with increasing sensitivity each time represents the most basic
form of learning. Without memory, pain wouldn鈥檛 persist. But equally, they say,
without pain our memory might never have evolved. From this simple survival
technique developed all the complexities of our own learning systems.
Advertisement
The link between pain and learning is starting to produce a few surprises.
Not only will potential memory-boosting treatments for Alzheimer鈥檚 need a
rethink to avoid leaving patients with long-term pain. It might explain why some
people feel pain when there鈥檚 no sign of a cause. And now new studies are
uncovering the molecular details of how painful memories form, opening up a
whole new way of treating pain. It might be possible, quite simply, to unlearn
it.
It took a particularly clever strain of mouse to highlight how closely pain
and memory are linked. For several years, brain researchers have known that as
mice mature their learning machinery is gradually downgraded. They swap a
component of the brain鈥檚 NMDA receptors, which activate the neurons needed to
lay down new memories, for a more sluggish version. In 1999, Joe Tsien and his
colleagues at Princeton University showed that with a bit of genetic tinkering
they could boost the number of receptors that retained the youthful component,
and so 鈥渟mart mouse鈥� was born. These brainy rodents were faster learners, and
remembered things for longer. So why, asked Tsien, have mice evolved to give up
this superior brainpower? Just last year, the answer became clear.
Min Zhuo and his colleagues at Washington University School of Medicine in St
Louis reported that Tsien鈥檚 clever mice initially respond exactly the same way
to an injury, for example licking a wounded paw just as a normal mouse would.
But the pain seems to be much more persistent (New Scientist, 3
February 2001, p 6). At first, Tsien thought the mice were simply better at
remembering they鈥檇 been injured. Now it鈥檚 clear that the change in the NMDA
receptors also increases signalling in the pain centres of the brain. So
developing new memories has to be balanced against the accumulation of more
pain. By downgrading their memory molecules, old mice may not learn many new
tricks, but at least they can sleep comfortably in their cages.
Perhaps we shouldn鈥檛 have been surprised about the pain these mice
experienced. Almost twenty years ago, Clifford Woolf, now at Harvard Medical
School, showed that neurons in the spinal cord respond to sustained pain signals
by becoming hypersensitive. They relay pain signals from sensors to the brain
that they would previously have ignored鈥攚hich partly explains why bruises
ache and sunburnt skin is so tender. Sensitisation also ratchets up chronic pain
from diseases such as arthritis, cancer and diabetes, and indelibly alters pain
sensation after major injuries and surgery.
Although it happened in the spine and not the brain, the process of
sensitisation looked just like learning. For more than 50 years, boosting the
strength of neural signalling has been the cornerstone of theories about how we
store new memories. If an animal is exposed to an unpleasant stimulus, says
Woolf, it makes sense for it to become hypersensitive and avoid it in future.
This could have been the evolutionary driving force that led to the ability to
remember stimuli generally. Pain hypersensitivity and memory are just different
ways of storing information, he says.
Our ability to form pain 鈥渕emories鈥� means that pain is often boosted by what
we鈥檝e experienced in the past. For some people, disease or injury sets them on
an ascending escalator of pain, so that eventually they feel widespread pain
even when there鈥檚 no recognisable source. Overactive pain learning seems to be
at the heart of a condition known as fibromyalgia syndrome, finds Roland Staud
at the University of Florida College of Medicine (Pain, vol 91, p 165).
And to date, no way has been found to unwind pain sensitisation. Conventional
painkillers numb pain but do not reverse sensitisation, so pain can return just
as strongly once they wear off. Woolf and others are looking for molecular
targets to force the nervous system either to forget pain, or not to learn it in
the first place. Unfortunately, that鈥檚 proving harder than they expected. There
are several different processes involved, and all of them may have to be
halted.
For a start, they have found that stimulating pain pathways activates enzymes
called kinases. Initially these temporarily modify the receptors on
pain-signalling nerves, boosting pain for perhaps tens of minutes. More intense
stimulation triggers these same enzymes to activate another chain of events that
switches on genes, leading to the production of more damage-sensing receptors
and more of the neurotransmitters that relay pain signals between neurons.
According to Woolf, as many as 1500 genes can be affected as pain signals surge
through the nervous system. The result is a sharp increase in pain signalling
that lasts for days or more.
But there is now the tantalising possibility of halting or reversing these
changes. Woolf and his colleagues recently showed that a kinase enzyme needed
for memory鈥攅xtracellular signal-regulated kinase鈥攊s also vital for
pain sensitisation. Woolf and Rurong Ji, also at Harvard Medical School, are now
looking at the role of ERK in a region of the rats鈥� spinal cord called the
鈥渄orsal horn鈥�, where pain signals from the body鈥攊n this case a sore
paw鈥攁re first received. ERK ratchets up the excitability of neurons in the
dorsal horn by regulating gene activity, says Woolf.
What鈥檚 more, they have found that blocking ERK in the spinal cord before any
pain is inflicted prevents the change in gene activity and pain sensitisation,
so no pain memory is stored. Even more dramatically, blocking ERK a day after
the pain is inflicted reverses the changes in gene activity and removes
sensitisation鈥攖he pain memory is 鈥渇orgotten鈥� (Journal of
Neuroscience, vol 22, p 478).
But other researchers have shown that blocking ERK in the
hippocampus鈥攖he part of the brain most closely associated with
memory鈥攑revents the formation of long-term memories. So in ERK they hold a
key to both pain and memory, which will constrain how they can use it. Soaking
the central nervous system in ERK inhibitors is obviously a non-starter. Yet
there are some settings in which ERK inhibitors may be used, the most likely
being the operating theatre.
In an attempt to prevent sensitisation and long-lasting post-operative pain,
it is now common to inject local anaesthetics via a catheter into the spinal
cord during some procedures, such as amputations. Unfortunately, the success
rate is low, suggesting that pain signals are still getting through during the
op. More than 70 per cent of amputees, for example, report 鈥減hantom limb鈥� pain
years after the operation, as a result of the sensitisation of pain pathways
that used to connect to the missing limb.
In the same way that the brain seems to convert short-term memories into
long鈥攂y creating permanent new branches and connections between
neurons鈥攄amage-sensing neurons can form new branches after experiencing
severe pain, creating permanently reinforced pathways. Fearing that we may never
be able to reverse this, the researchers are concentrating on prevention. 鈥淭he
best way to think about it is by analogy to memory,鈥� says Jay Yang, a researcher
at the University of Rochester Medical Centre in New York state. And so far, we
don鈥檛 have a drug that can wipe human memories.
Woolf and his colleagues used a catheter to apply an ERK inhibitor just to
the spinal cord of rats. Doing the same during and after an operation, together
with a local anaesthetic, may help prevent residual pain signals from causing
sensitisation. 鈥淐ertainly, there are a number of companies that are developing
kinase inhibitors,鈥� says Woolf.
Yet blocking ERK is unlikely to be the whole answer. Woolf and his colleagues
have found that ERK is only used for pain sensitisation by neurons in the two
outermost layers of the dorsal horn. Yet many other signals swamp pain pathways
during an operation. Sensitisation of neurons in the next layer is achieved by a
different pathway and a different kinase, called protein kinase C gamma or
PKC-&ggr;. 鈥淭he bottom line is that there are parallel inputs and many processes
acting simultaneously,鈥� says Woolf. 鈥淲e will almost certainly need to use
several approaches.鈥�
Yang and his colleagues have already had some success reducing pain
sensitisation in rats by targeting PKC-&ggr;. Even more hopefully, mice that lack
the gene suffer less from chronic pain but otherwise behave
normally鈥攕uggesting that PKC-&ggr; is required for pain but not for normal
memory.
There may be another way to unwind chronic pain. Tsien鈥檚 smart mice had
altered NMDA receptors. These act very early in the pathways that underlie both
sensitisation and memory, before the kinases kick in. But the NMDA receptor鈥檚
dual role means that any treatments that alter its activity will have to be
approached very cautiously.
There had been hopes that boosting NMDA receptors in the brains of
Alzheimer鈥檚 patients would combat memory loss鈥攂ut now it seems that this
might leave patients with a better memory but more pain. And targeting the same
molecule to combat pain sensitisation in people suffering from chronic pain runs
the risk of impairing memory. If you could somehow boost its activity only in
the hippocampus of Alzheimer鈥檚 patients鈥攁nd counter its effects only in
the regions of the forebrain that process pain鈥攜ou鈥檇 be onto a winner.
Zhuo is carefully optimistic. 鈥淚 think it鈥檚 clearly a good target for us to
consider in the future,鈥� he says.鈥�
Since all these treatments are at least several years away, researchers such
as Woolf suggest we should concentrate on how we use existing
painkillers鈥攏ot only giving them to ease pain, but also to slow pain
sensitisation. Simply giving painkillers early in the treatment of a disease may
delay the onset of intense, chronic pain. For many patients, however, this will
require a shift in attitude. 鈥淵ou have to ask what you feel the role of pain
is,鈥� says Staud. 鈥淚n many cultures, people feel that pain is an important part
of their lives.鈥�
While people in the West may view pain mostly in clinical terms, to complain
about mild or moderate pain is often considered childish. And before doctors can
recommend 鈥減roactive鈥� painkillers, they will need to know at what point pain
sensitisation begins to store up problems for the future. When do the risks of
chronic pain outweigh the side effects from some painkillers and the risk of
addiction? For the moment, researchers are only sure about the extremes. 鈥淭his
is most important for great pain experiences, for example, post-operative pain,鈥�
says Staud, 鈥渂ut if you just have a blister on your finger, I wouldn鈥檛 recommend
painkillers for that.鈥�
* * *
Growing pains
JUST as newborn babies, with their partially wired brains, need to learn how
to see and hear, they may also be learning how to feel pain. 鈥淎s the nervous
system develops it becomes more and more influenced by outside events,鈥� says
Maria Fitzgerald of University College London. The sensory world has a major
influence on how wiring proceeds in the brain.
Unfortunately, for babies born prematurely or requiring surgery, the sensory
world is filled with painful medical procedures. Research on infant rats has
shown that developing pain pathways are rewired much more readily than those in
adults, and that minor injuries to newborns can cause permanent hypersensitivity
in the adult. So does lifesaving neonatal care teach babies to feel pain too
strongly?
Newborn babies can be sensitive for many weeks in the heel used for routine
blood sampling, and there is preliminary evidence that circumcised baby boys
feel more pain when being given vaccinations months later. Unlike laboratory
rats and mice, however, our infancy is measured in months, not days, giving
time, perhaps, to 鈥渃orrect鈥� early rewiring. 鈥淧eople have got very excited that
these infants are going to be the chronic pain patients of their middle age,鈥�
says Fitzgerald. But for the moment evidence is lacking.
Fitzgerald hopes to find out. 鈥淲e are going to look at a cohort of
adolescents who have come through surgery and intensive care as infants,鈥� she
says. She will test for differences in pain reflexes and overall pain sensation.
Unless the group has greatly altered pain sensitivity, it鈥檚 likely that they
will have simply accepted what they feel as normal. 鈥淔or each individual, that鈥檚
all they鈥檝e ever known,鈥� says Fitzgerald.
