Take one person who has been driven insane by syphilis and infect them with
malaria. As a therapy, it sounds none too promising. But earlier this century, a
certain Vienna-based psychiatrist called Julius Wagner-Jauregg discovered
otherwise. Malaria curbed his patients鈥� insanity. And the reason for that, said
Wagner-Jauregg, was that it induced a series of beneficial fevers.
Wagner-Jauregg went on to claim further success treating mental disorders
such as depression and schizophrenia. In 1927 he became the first鈥攁nd
only鈥攑sychiatrist ever to win a Nobel for medicine.
Today, the notion of using fever to manipulate the mind would seem to come
from the outer fringes of weird medicine. But in fact, it is not as mad as it
sounds.
Advertisement
In labs around the world, researchers are discovering that the immune
responses which help to trigger fever have distinct effects on the brain,
altering sleep patterns and even your mood. If this surprises you it鈥檚 because
we are used to hearing about the other side of the coin. Time and again, we are
told that our mental state can affect our body鈥檚 defences against
disease鈥攊n other words, that the brain sends signals to the immune system.
But the communication can run both ways, and in some labs the familiar mantra of
鈥渕ind over body鈥� is being turned on its head.
Rallying defences
鈥淲e are starting to think of the immune system as a sense organ that gives
the brain information about things happening inside the body, much as the eyes
and ears give it external information,鈥� says Steven Maier, from the University
of Colorado in Denver.
The chief agents of this 鈥渟ense organ鈥� are the cytokine molecules that are
produced by immune cells when microbes enter the body. Researchers have known
for years that cytokines play a key role in rallying defence cells to bring
infections under control. Now it is clear that these messenger molecules also
influence the brain and behaviour in specific ways.
On the one hand, cytokines can make animals and humans socially withdrawn and
poor at concentrating. On the other, there is controversial evidence that
cytokines and the fevers they trigger help to ward off some of the symptoms of
depression. Wagner-Jauregg, in other words, lives on in spirit if not in name,
especially in Germany where, according to some estimates, thousands of depressed
people every year are being injected with bacterial proteins that boost cytokine
levels.
But the modern story of 鈥渇ever and the mind鈥� begins instead with sleep
research. Unlike moods and emotions, sleep is a fairly unambiguous mental state,
and healthy people do it daily. That鈥檚 why researchers have turned to it ahead
of mental illness for clues as to how immune responses鈥攁nd the microbes
that trigger them鈥攁ffect the brain.
鈥淓veryone knows that microbes are important for normal physiology,鈥� says
James Krueger, an expert on infection and sleep at the University of Tennessee,
Memphis. 鈥淲hat we are saying is that they are important for sleep. Sleep is
exclusive to the brain, so in some sense, microbes affect our
肠辞苍蝉肠颈辞耻蝉苍别蝉蝉.鈥�
Few would have expressed this humbling insight so forcefully in the past. Yet
the basic idea actually goes back decades. Researchers have known for more than
20 years that if you keep animals awake for long enough you can begin to detect
bacterial proteins in their spinal fluid. Krueger and others have suggested that
these bacterial proteins build up during waking hours and eventually trigger
sleep. This may even happen when there is no noticeable infection: in theory the
kilogram or so of bacteria that live peaceably in your guts could supply the
sleep-inducing proteins. Indeed there鈥檚 some evidence for this. Rats reared on a
diet containing antibiotics that lower their levels of gut microbes sleep for
fewer hours than rats on a normal diet, says Linda Toth, a former colleague of
Krueger鈥檚 who is now at the St Jude Children鈥檚 Research Hospital in Memphis.
When you up the microbe levels in animals, however, the story gets more
complicated鈥攁s Toth and other researchers discovered in the 1980s when
they began inoculating rats and rabbits. Sometimes the injections contained
whole bacteria or fungi, sometimes just protein components. But broadly
speaking, the results were usually the same. Animals injected with low doses at
first needed a larger than usual amount of slow-wave sleep (the deeper form of
sleep that takes up most of the animal鈥檚 sleeping time). But higher doses, or
infections left to take hold, had the opposite effect. They sharply reduced the
amount of slow-wave sleep and completely blocked the bursts of rapid eye
movement (REM) sleep that normally interrupt slow-wave sleep.
People with infections behave much the same way, although the effects on
their sleep patterns are less dramatic. Over the past few years, Thomas
Pollm盲cher, at the Max Planck Institute of Psychiatry in Munich, has found
that research volunteers need up to a quarter more slow-wave sleep than usual
after being injected with lipopolysaccharide, a component of bacteria. Again,
the effect is limited to low doses that do not raise the volunteer鈥檚
temperature. Larger doses wreak havoc, suppressing the regular bursts of REM
sleep and reducing deeper sleep by up to an hour in the first three hours of the
night.
All this fits well with everyday experience: anyone who has ever had flu will
recognise the initial overwhelming urge to sleep, and the tossing and turning of
a bad bout of fever. Research on people infected with HIV also backs up the
idea: in the initial months and years before the onset of illness, HIV-positive
people tend to need more slow-wave sleep than uninfected people, while in the
later stages of AIDS, their sleep is often severely disturbed.
By the 1990s, Krueger was beginning to realise that microbes influence sleep
by stimulating immune cells to release the cytokines interleukin-1 (IL-1) and
tumour necrosis factor (TNF). Both are produced very early in an infection.
What鈥檚 more, low levels of IL-1 tend to induce sleep while high levels suppress
it鈥攋ust as you would expect from the animal experiments.
Yet despite this rule of thumb, predicting the impact of infections on sleep
is no easy matter. David Dinges, an expert on body rhythms at the University of
Pennsylvania in Philadelphia, thinks too few researchers have checked what
happens to the body鈥檚 levels of cytokines over a normal 24-hour cycle. In fact,
levels fluctuate, because they are strongly influenced by the daily ebb and flow
of hormones involved in alertness and stress such as cortisol. These
fluctuations help to explain why the effects of infection on sleep vary from one
time of day to the next.
Pollm盲cher, for example, found that the sleep-inducing effects of low
doses of microbial proteins were weaker in the morning than in the evening when
levels of IL-1 and TNF are naturally higher. This fits with what we know about
children鈥檚 responses to vaccinations. Paediatricians know from experience to
avoid immunising babies late in the day. Given early, a vaccination will
probably just make a baby sleepy. Given late in the day, there may be fever and
disturbed sleep.
Why mammals have evolved to need extra sleep in the early stages of infection
isn鈥檛 clear, although withdrawing from active life may help to conserve scarce
energy for fighting the infection. Nor is there an obvious evolutionary reason
for the disturbed sleep that goes with more severe infection. But no matter. For
scientists like Krueger, the key question is how an infection picked up by
immune cells somewhere in the body, such as the gut or the lungs, can be
communicated to the brain, where sleepiness and wakefulness are controlled.
There is no single 鈥渟leep centre鈥� or 鈥渟leep chemical鈥� in the brain. But
researchers have identified key roles in the sleep cycle for at least two areas
of the brainstem and three clusters of neurons in the hypothalamus, a structure
at the base of the forebrain that regulates the hormonal activities of the
nearby pituitary gland. What鈥檚 more, both the pituitary and the brainstem have
receptors for IL-1, says Mark Opp, a sleep researcher at the University of Texas
in Galveston.
So one possibility is that the cytokines work by increasing or decreasing the
secretion of hormones that influence the brain. Cortisol, which makes us alert,
is one candidate. Another is growth hormone: Opp and Krueger have found that
IL-1 stimulates the pituitary to make this substance. And since levels of growth
hormone rise during slow-wave sleep, some researchers believe it may help to
deepen slumber.
Now comes the rub. Infections are usually dealt with in the gut, lungs or
some other part of the body鈥攏ot the brain. And it is at the site of the
infection that defence cells make their cytokines. Cytokine molecules are way
too big to cross the blood-brain barrier, so how do they reach receptors in the
hypothalamus? The answer, says Maier, is that the brain produces its own
cytokines. But how does it know when to do so?
Gut reaction
Maier鈥檚 team thinks it has the answer. Maier reasons that if the immune
system acts as a sense organ, albeit a highly diffuse one, it must communicate
with the brain via some peripheral nerve, just as the eye communicates via the
optic nerve. His prime suspect is the vagus nerve, which connects the brainstem
with the lungs, stomach, liver and intestines, where infectious agents would be
certain to appear. So what effect would fever-causing bacterial proteins have on
rats whose vagus nerve had been surgically cut? The results astonished the
team. 鈥淚f you cut the vagus, peripheral sickness doesn鈥檛 result in the brain
making its own cytokines,鈥� says Maier. Nor do the animals show the expected
disturbances in their sleep patterns or signs of fever. But the icing on the
cake came when Maier examined vagal nerve endings in the gut and spleen and
found receptors for IL-1.
If Maier is right, the vagus nerve has a key role in passing cytokine
messages from the immune system to the brain. And the existence of such a
communication channel could mean that sleep is just one of many states of
consciousness influenced by the immune system. Might cytokine levels influence
mood and attention span?
Maier is inclined to think so. He points out that rats given
lipopolysaccharide or other bacterial components that raise their levels of IL-1
and TNF lose interest in things they normally find appealing鈥攕uch as sweet
food. 鈥淭hey also show reduced activity, reduced exploration and changes in
social interaction,鈥� he says.
Pollm盲cher agrees that sleep may be just the most obvious mental state
to be affected. 鈥淭his question extends far beyond sleep, into areas where very
complex brain functions such as mood, anxiety and stress interact.鈥�
Does this mean that Wagner-Jauregg鈥檚 belief in fever therapy was valid after
all?
That鈥檚 harder to answer, but a few brave souls are trying. Wagner-Jauregg鈥檚
ideas fell out of favour because their main practical benefit had been to treat
advanced syphilis. Once antibiotics became available there was no need to risk
giving patients malaria. Nevertheless, the ideas have not completely died out.
In Germany, where alternative medicine is all the rage, malaria is no longer an
option but there are a dozen or more clinics that offer bacterial toxins to
induce fevers in people suffering from depression. What鈥檚 more, people are
trying it in their thousands every year.
One psychiatrist, Joachim Bauer at the University of Freiburg, has studied
the alternative clinics鈥� 鈥渇ever induction鈥� treatment. He observed a group of
patients given a dose of bacterial toxins for 24 hours. Their temperatures rose,
they produced IL-1 and TNF, and their sleep patterns were disturbed, with all
REM activity vanishing. Next day, they felt better鈥攁nd scored better on
standard psychological tests of mood. But the improvement lasted no more than a
few hours and the next night their REM sleep bounced back, giving them another
disturbed night.
Feverish blues
Bauer doubts the treatment is of much benefit and warns that giving people
bacterial toxins could be dangerous without constant monitoring. But he does not
dismiss the wider idea that immune responses can in principle influence the
course of mental illnesses. In fact, Bauer has just finished a study looking at
the links between depression and fevers. In the decade before their first bout
of depression, patients reported fewer episodes of fever than a 鈥渃ontrol鈥� group
of healthy people matched for age and sex, even though the people with
depression reported more illness generally than their healthy counterparts.
Studies based on people鈥檚 memory of their health tend not to be as robust as
studies that actively track people for a decade. But Bauer, for one, finds the
result intriguing and has a couple of ideas to explain it. On the one hand,
fever might be unusually rare in depressed people because they tend to produce
excessive amounts of the stress hormone cortisol. And the hormone blocks the
cytokines that trigger fever in response to infection. Alternatively, the
symptoms of depression could be rare in people who have lots of bouts of fever.
Once a fever is induced, raised levels of cytokines could boost the functions of
brain cells that use the chemicals dopamine or noradrenaline, some of which are
thought to be sluggish in depressed people.
Whatever the answer, Pollm盲cher thinks it is too early to develop grand
theories of immunity and mood. Nevertheless, he too is intrigued by findings
that suggest cytokines can have mind-altering effects. In the 1980s, when
doctors gave massive doses of certain cytokines as an experimental cancer
treatment, many patients suffered from hallucinations. And high levels of
cytokines might explain why some people hallucinate during a severe fever,
suggests Pollm盲cher.
His team has turned its attention to clozapine, a drug commonly given to
people with schizophrenia. So far, researchers have assumed the drug acts
directly on the brain鈥檚 neurotransmitter systems, but Pollm盲cher鈥檚 findings
suggest another possibility: the drug may work because it stimulates the immune
system which, in turn, influences the brain. At the beginning of treatment more
than half of all people who are given clozapine develop a fever that lasts for a
few days. It turns out that the drug also boosts levels of the cytokine TNF,
which is always produced during a fever. And in some people the drug increases
non-REM sleep as well.
In Scotland, David Horrobin, research director of Scotia Pharmaceuticals, has
a further suggestion to explain how some symptoms of schizophrenia might
sometimes improve in a fever. Cytokines stimulated by infection and fever would
boost levels of prostaglandins in the brain, he says. This might help to offset
a shortage of these hormones in schizophrenics, itself caused by a lack of
arachidonic acid, the hormones鈥� chief building block.
So far no one has come close to providing definitive explanations for these
provocative ideas. But most are convinced that the links between the immune
response and mood are real. In this they have some illustrious forebears.
Wagner-Jauregg was fond of quoting Parmenides, the famous Greek philosopher from
the 5th century BC, who reputedly said: 鈥淕ive me the potential to induce fever,
and I will heal all diseases.鈥�
No modern researcher would make such lofty claims. But now that scientists
have begun to see the immune system as a sense organ feeding information to the
brain, at least some of the Austrian Nobel prize-winner鈥檚 ideas fall into
place. And one thing at least is clear: none of us should underestimate the
impact of the microbe on our minds ever again.