S. Giovanoli et al., Science/AAAS

Starry night. Active immune cells (yellow) light up the brain tissue of mice exposed to both prenatal infection and stress (right), compared with the quieter control (left).

One-Two Punch of Infection, Stress May Lead to Schizophrenia

Emily is a staff writer at Science.

No single cause has yet been discovered for schizophrenia, the devastating neuropsychiatric syndrome characterized by hallucinations, disordered thoughts, and other cognitive and emotional problems, typically beginning in early adulthood. Although schizophrenia runs in families, in many cases no genetic risk is apparent, leading many researchers to look for environmental explanations. Now, research in mice provides support for a long-held hypothesis: that the syndrome, and other neurological disorders, can emerge when multiple environmental insults such as prenatal infection and adolescent trauma combine.

Environmental stressors such as infection and abuse were long ago shown to be risk factors for schizophrenia. Large studies of children whose mothers were infected with influenza during the last months of their pregnancy, for example, have a roughly twofold increase in risk of developing the syndrome compared with the general population. That doesn't explain why a few people who are exposed to an infection in the womb go on to develop schizophrenia while most don't, however, says Urs Meyer, a behavioral neurobiologist at the Swiss Federal Institute of Technology in Zurich and co-author of the study reported online today in Science.

One long-held hypothesis, he says, is that early infection creates a latent vulnerability to schizophrenia that is only "unmasked" by later insults, such as physical injury or psychological trauma. Such stressors are thought to be particularly damaging during critical periods of brain development such as early puberty, he says. Although the "multiple-hit" hypothesis has been prominent in the literature for some time, it is difficult to test the idea with human epidemiology studies, he says. "You need huge, huge data sets to see anything."

Lacking that epidemiological data in the field, Meyer and his colleagues decided to explore the multiple-hit hypothesis in mice. There is no way to create a "schizophrenic" mouse, Meyer emphasizes, but certain key cognitive deficits and behaviors can be measured in both animals and humans and are thought to rely on similar brain regions and neural circuits. In one group of mice, he and his colleagues dosed their mothers with a synthetic compound that simulates a mild viral infection during late pregnancy; when their offspring hit early puberty at about 6 weeks of age, the young mice were exposed to unpredictable stress, such as being restrained, deprived of water, or given electric foot shocks. The second and third groups were exposed to infection or stress alone, while the fourth group grew up relatively unperturbed as a control.

After the final stressors were administered, the researchers gave the young mice a round of behavioral tests and then waited until the mice reached adulthood, at about 10 weeks old, to see if their different upbringings had affected their cognitive abilities or brain morphology. To the researchers' surprise, the mice subjected to both prenatal infection and stress during puberty showed far greater behavioral deficits and cellular changes in brain regions relevant to schizophrenia than those that had been exposed to infection or stress alone. Rather than merely a cumulative effect, Meyer says, the two conditions appeared to act in synergy, increasing anxiety behaviors in the mice as well as damaging their performance on tests for associative memory and other basic cognitive skills associated in people with a range of neurological disorders, including schizophrenia.

One striking example, Meyer says, was a neurological test called prepulse inhibition (PPI) in which animals are startled by an unexpected noise such as an air horn. If they hear a quiet tone milliseconds before the blast, both mice and humans typically automatically diminish their startle response to a second, louder noise. When mice exposed to both infection and stress were tested for PPI, however, they responded to the second loud noise with the same intensity even if they had been exposed to a previous tone—the same deficit seen in humans with schizophrenia.

Additional evidence that the "double-hit" had affected the animals beyond a simple additive effect could be seen in the animals' brain tissue, Meyer says. Microglia, the brain's immune cells, are normally slender and threadlike; however, they bulk up like the Hulk when activated by a threat. Exposure to infection or stress alone did not significantly change the shape or quantity of these cells, the authors say. The double-hit, however, doubled or tripled the activation of microglia in the hippocampus and prefrontal cortex, two areas of the brain associated with schizophrenia. People with schizophrenia also tend to have more activated microglia in these areas, Meyer says.

Neuroscientists are increasingly recognizing the role of the immune system and inflammation in brain disease, notes James Koenig, a neurobiologist at the University of Maryland School of Medicine in Baltimore. One possibility is that the overzealous cells, primed by early infection and set off by later trauma, might be endangering what they are trying to protect.

Although the multiple-hit hypothesis has been around for a long time, "this is really a new twist on things," Koenig says. Studies have shown similar effects in disorders such as depression, he says, but this is the first time it has been related directly to cognitive deficits found in schizophrenia. It would be "ludicrous" to think that we can model all aspects of schizophrenia in animals, Koenig cautions. However, he says, studies like this can "give us some clue" about what can result from multiple insults to the developing brain because it shows experimentally that prenatal and postnatal impacts can work together to have synergistic effects on later behavior, he says. Given that, he says, the study "makes a pretty significant leap forward."

Posted in Brain & Behavior