New Scientist 6 November 2004 (UK)
by Janet Ginsburg
"Coughs and sneezes spread mind diseases"
Pick up an infection and you'd expect to feel under the weather. But mental illness? Catching schizophrenia or depression isn't as fanciful as it sounds, says Janet Ginsburg.
It was hell on earth. "I was in a concentration camp. I could smell the burning bodies," recalls Stephanie Diers. But it was 1975, the second world war was long over, and Diers was in Martinez, California, thousands of miles from Auschwitz. The 19-year-old was locked up, but in a mental ward.
Diers's psychotic break was just the latest in a string of illnesses that had plagued her and puzzled her parents for the better part of a decade. No one could figure out what had happened to the bright little girl who loved riding horses and playing in the woods. Her list of ailments was as baffling as it was extensive: headaches, dizziness, fatigue, pain in her teeth, tingling in her toes, sore throats, flu-like illnesses, chest pains, pains in her spleen and liver, poor balance, sensitivity to light and sound, memory lapses. And now psychosis.
After six months in hospital, Diers was released and struggled on with her life, eventually earning a college degree and working as a teacher. But it would take another 14 years of serial illness before any of her doctors thought to test for Lyme Disease, a tick-borne illness identified in the late 1970s.
Like its distant cousin syphilis, Lyme is caused by corkscrew-shaped bacteria called spirochaetes that are able to burrow into tissue, including the brain. There they can lie dormant and harmless for months, even years. But once active, they can stir up a devil's brew of symptoms, including conditions that match the clinical diagnoses for schizophrenia, bipolar disorder and depression.
Diers, of course, tested positive, and was put on a long course of antibiotics. "I had so many psychiatrists and psychologists tell me it was all in my head," she says. They just did not know how right they were.
Borrelia burgdorferi, the Lyme bacterium, is now widely accepted as a cause of psychiatric disease. And in recent years several other bugs have come to light that are able to trigger symptoms of mental illness, ranging from behavioural problems to depression and full-blown psychosis. In fact, so many potential "mind germs" have now been unearthed that some researchers are ready to challenge the conventional wisdom about the principal causes of mental illness.
When it comes to causes, genes have always been the prime suspects, with environmental triggers poorly understood and infectious disease an afterthought. Infectious disease is "maybe 5 per cent as important as genes", claims Richard Straub of the Genes, Cognition, and Psychosis Program at the US National Institute of Mental Health in Bethesda, Maryland. Straub, who has identified several genes tentatively associated with schizophrenia, thinks that among environmental causes, peer groups, drug experiences and other social factors are far more significant targets for research.
But according to evolutionary biologist Paul Ewald of the University of Louiseville in Kentucky, that ignores some obvious facts. "The biggest breakthrough in the history of psychiatry was recognising that syphilis causes insanity and that it can be prevented with antibiotics," he says. The same mistake is being repeated today with Lyme disease and other infectious agents, Ewald believes.
If diagnosed quickly, before the bacterium gets a foothold, Lyme disease can be cured with a round of antibiotics. But spotting it can be tricky. Not everyone gets the telltale "bull's eye" rash and antibody tests can be unreliable, especially early on. That presents a dilemma for doctors concerned about overprescribing antibiotics. If they wait too long - as little as a few months after infection - the spirochaetes begin to attack the central nervous system with devastating consequences. This "late stage" Lyme is very hard to cure, often requiring months or years of antibiotics.
At Columbia University in New York, neuropsychiatrist Brian Fallon is in the middle of a series of studies designed to see what actually goes on inside the brains of chronic Lyme patients. One finding so far is that patients show significant reduction in blood flow in brain regions associated with memory and visuospatial organisation. Fallon has also looked at children, who are especially at risk because they tend to play outside, where the ticks are. He found significant cognitive and psychiatric problems in children who had been diagnosed late, on average a year after infection. They scored low on memory and perception tests and were also depressed, with some having suicidal thoughts.
In 2003, more than 21,000 cases of Lyme disease were reported to the Centers for Disease Control in Atlanta, Georgia, making it the most common vector-borne disease in the US. Yet as few as 1 in 10 cases are documented, so the true tally might be more than 200,000. And in 1996, another bacterium, Borrelia lonestari, which is carried by the lone star tick common throughout the American south, was linked to Lyme-like symptoms. So far there are no tests for it, so no case statistics.
Lyme isn't just a problem in the US. "You really have a pandemic," says Raphael Stricker, a San Francisco physician specialising in chronic Lyme. "It's all over the US. It's all over Europe. It's in parts of Asia. It's everywhere."
Beyond Lyme disease, the evidence linking infections to psychiatric disorders becomes a little more hazy. Ironically though, some of the most tantalising clues are found in the very same data used to prove the case for genetic links: twin studies.
Identical twins, who have 100 percent of their genes in common, are much more likely to both develop schizophrenia than are fraternal twins or full siblings, who share just 50 per cent of their genes. Clearly, then, genes are important. But the data is more complicated, explains Ewald. "People saw these associations without thinking about alternative explanations," he says. For example, when one identical twin develops schizophrenia, half the time the other does not. If schizophrenia were strictly genetic, concordance (meaning both twins get sick) should be 100 per cent. This holds true even if many genes are involved in the disorder, as geneticists now believe.
But when the environment inside the womb is taken into account, the story begins to shift. Fetuses develop inside two porous sacs, the inner chorion and the outer amnion. Nearly 70 per cent of identical twins share a chorion, a feature that can be determined after birth by the presence of subtle physical traits such as mirror-image fingerprints. Those twin pairs are nearly six times as likely to be concordant for schizophrenia as identical twins with separate chorions.
Meanwhile the concordance rate for non-identical twins is nearly twice that of full siblings, even though the genetic relationship is the same: 50 per cent. Fraternal twins rarely share sacs, but they do share a womb. That says Ewald, points to an environmental factor - though not necessarily an infection.
To look for evidence that a prenatal infection could be linked to the development of schizophrenia years later, a team of Columbia University epidemiologists led by Ezra Susser and Alan Brown sifted through the medical records of 20,000 women who were pregnant in Alameda county, California, between 1959 and 1966. The women, all patients at Kaiser-Permanente, the largest healthcare provider in the region, were part of a massive child health and development study. But the real stroke of luck was that most of their children have remained in the Kaiser-Permanente system. "We could follow them right up to ago 40," says Susser.
The researchers first determined which children had been diagnosed with schizophrenia spectrum disorders. Then they tested their mothers' blood samples for antibodies to the strains of influenza virus that had been circulating during their pregnancies, and compared the results with blood tests from a set of matched controls - the mothers of mentally healthy children from the same Kaiser-Permanente group.
The results, which are published in the Archives of General Psychiatry (vol 61, p774), are dramatic. Maternal exposure to flu during the first half of pregnancy tripled the child's risk of developing a schizophrenia spectrum disorder. "It's far higher than any single gene in terms of what we call relative risk," says Brown.
Brown point out, however, that some of the healthy children's mothers were exposed to flu during the first half of pregnancy, too, so exposure doesn't guarantee schizophrenia. What's more schizophrenia is rare, affecting just 1 per cent of the population. Still, if the results of the Columbia study can be duplicated, a significant number of cases - perhaps as many as 14 per cent, according to Brown - may turn out to be preventable.
So how might a virus cause schizophrenia? According to Paul Patterson at the California Institute of Technology in Pasadena, the virus may not be doing the damage directly. He injected pregnant mice with a molecular mimic of the flu virus, which generates an immune response without causing infection. Nevertheless, offspring mice developed behavioural abnormalities reminiscent of schizophrenia, suggesting that the stress of a maternal immune response alone may be enough to affect neurodevelopment.
More information on how viruses attack the brain is coming from work on Borna virus, which was first seen in horses in the late 1800s. Borna affects a number of bird and mammal species, including non-human primates, causing a broad range of movement and behaviour disorders. There are also tentative links to human psychiatric illnesses, most notably schizophrenia, bipolar disorder and depression, based largely on the presence of Borna antibodies in some patient's blood.
In rats infected with Borna in the lab, symptoms resemble autism, with delayed growth, learning disabilities and repetitive behaviours. And there are now some clues as to what the infection is doing to the brain. When rats are infected shortly after birth (the neurodevelopmental equivalent of a human prenatal infection), neurons critical for cognitive, emotional and motor development either die off or miss key developmental cues. If rats are infected during adolescence, when their brains are more developed, Borna appears to kill neurons both directly and through an overzealous immune response.
Yet another pathogen that has been linked to psychiatric conditions is the protozoan parasite Toxoplasma gondii. It infects everything from cats and cattle to sea otters and people. Human infection rates range from 15 percent in the US to more than 80 per cent in some countries. Hundreds of millions of people have it, usually having caught it from undercooked meat or contact with cat faeces. Most of the time, toxo causes no more than a mild-flu like illness, but, like Lyme spirochaetes, the parasites can burrow into tissue and lie dormant for long periods of time.
Some studies have found that people with schizophrenia are three times as likely as the general population to be infected with toxo (New Scientist, 26 October 2002, p41). Meanwhile, using the same Kaiser-Permananete data as in the influenza study, Susser and Brown found a correlation between high maternal levels of anti-toxo antibodies and schizophrenia in the corresponding children. According to their calculations, toxo more than doubles the risk.
In the pantheon of mind-altering microbes, toxo is unique in its mission: it must alter the behaviour of its intermediate host, usually a rodent, to get itself back inside its reproductive host, a cat. In a series of experiments, Joanne Webster at the University of Oxford discovered that toxo-addled wild rats not only lost their natural fear of wandering into open spaces, but were actually attracted to cat smells. They practically delivered themselves for dinner.
Infected humans don't end up as cat food, but acute toxo can cause hallucinations and other psychotic behaviours. In fact, both rats and people may be tripping: studies from the 1950s and 60s suggest that toxo can trigger the production of LSD-like substances in the brain.
The most insidious mind germ of all could be an inside operator: a germ that behaves like a gene. Human endogenous retroviruses (HERVs) infect egg and sperm cells. Like genes, they are then copied into all the cells of the body, although they only become active under specific circumstances.
Virologist Robert Yolken at John Hopkins University in Baltimore has discovered one called HERV-W reproducing in the cerebrospinal fluid of some people with schizophrenia, but not in healthy controls. He suspects that a second, conventional pathogen, possibly a herpesvirus or Toxoplasma, may trigger HERV-W to switch on, and the combination somehow triggers symptoms. In a study published last year in The American Journal of Psychiatry (vol 160, p2234), Yolken's team reported improvement in the symptoms of people with schizophrenia who were treated for cytomegalovirus, a common herpesvirus.
The idea that you can catch a mental illness from something as innocuous as a bug bite or a sneeze may seem the stuff of nightmares, but it also holds hope for new ways to fight back. For one thing, it suggests that treating mental illness can sometimes be as simple as tackling an underlying infection. That's exactly what Borna researchers Liv Boden of the Robert Koch Institute and Hans Ludwig at the Free University, both in Berlin, Germany, found in clinical trials of the antiviral drug amantadine. A majority of Borna-positive patients diagnosed with major depression or bipolar disorder showed significant improvement in as little as seven weeks on the drug.
There are also hints that some existing psychiatric drugs work because they eradicate infections. For example, Yolken reports that the antipsychotic haloperidol and the mood stabiliser valproic acid appear to inhibit the growth of Toxoplasma, at least in cell culture.
And for those of us lucky enough to be free from mental illness, there's a lesson to be learned, too. Good mental health may be as basic - and as cheap - as avoiding cat litter and undercooked meat, and checking for ticks.
Janet Ginsburg is a science writer in Chicago
6 November 2004
This article is accompanied by an associated small piece not by the above author.
It is entitled "Obsessive-compulsive sore throat?"
It's not always infectious agents themselves that do the damage. More than a century ago doctors made the link between "strep throat" - a sore throat caused by the common bacterium Streptococcus - and the development of the heart condition rheumatic fever a few weeks later. The immune system gets confused, mistakes heart cells for bacteria, and attacks.
A similar autoimmune response seems to be causing behavioural problems ranging from Tourette-like tics and obsessive-compulsive dissorder to attention-deficit hyperactivity disorder and anorexia. The syndrome, called PANDAS (Paediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcus). strikes about once in 1000 cases of strep infection.
PANDAS is a strep-induced antoimmune attack on an area of the brain called the basal ganglia, which helps control numerous behaviours. High levels of anti-strep antibodies have been linked to enlargement of the basal ganglia in PANDAS patients. As the levels go down, the brain recovers and symptoms fade away.
In an experiment at the National Institutes of Health in Maryland, researcher Susan Swedo removed antibodies from the blood plasma of 30 children with PANDAS. More than a year later, 80 per cent remained symptoms-free.
Though PANDAS is technically a childhood condition, there are reports of
adults with similar abrupt-onset behavioural problems. And a team from the UK's Institute of Child Health in London found that a PANDAS-like condition may be behind the mystery of von Economo's disease, the "sleepy sickness" that devastated thousands around the time of the 1918 flu pandemic (New Scientist, 18 October 2003, p34). Of 20 contemporary patients with a similar illness, more than half had had a sore throat before developing the sickness, and 95 per cent tested positive for antibodies reactive against the basal ganglia.