The Infection Connection in Schizophrenia

Adapted from image by I.I. Gottesman ©2001 GENES AND MORE: The risks of developing schizophrenia over a lifetime to the relatives of schizophrenia sufferers accord with a largely genetic explanation. Yet with 48% concordance for identical twins, environmental factors may play a role. It's a scary thought that one could develop a debilitating mental illness such as schizophrenia as easily as catching a cold. Well, it's more complicated than that, say advocates of the so-called infectious hypothesis, which states that viral and possibly bacterial infections occurring at critical points in brain development could increase the risk of mental illness. Everything from influenza to herpes simplex viruses to Toxoplasma gondii has been implicated in elevating risk for schizophrenia, perhaps indirectly through immunological reactions that change brain chemistry or wiring at key developmental stages. Raised, ridiculed, retracted, and rewrought, the hypothesis has morphed for nearly a century. Its current permutation may be gaining ground, though. With better epidemiological studies and animal work emerging, some former scoffers are beginning to accept the notion that, while not a direct cause, infections may factor into schizophrenia's complex gene/environment formula that dictates one's chances for landing somewhere among the less than 1% of people affected worldwide. The Stanley Research Institute's associate director of research, E. Fuller Torrey, has been a notorious proponent of infectious hypotheses since the 1970s. "I wouldn't say we're respectable, but it's no longer ridiculed, which seems like a step forward." BLAMING MOM Though most have eschewed the notion of a schizophrenogenic mother, a healthy environment does start in the womb. Various strands of evidence link increased propensity for schizophrenia to infections during gestation. Past studies uncovered a higher propensity in people born soon after known disease outbreaks such as the 1957 flu pandemic. Other data link increased risks to winter births or reported flu symptoms. Timing could be important, as infections in the second trimester appear to have an effect. But, of the 30 to 40 studies published, says S. Hossein Fatemi of the University of Minnesota, Minneapolis, as many as one-third produced negative results. Paul H. Patterson at California Institute of Technology says, "The problem with those epidemiological studies is that they're based on probabilities. They're sometimes called ecological studies." Researchers can only guess that a woman who was pregnant at the time would have been infected. Investigators are attempting to collect more direct evidence of infection. In 2001, Columbia University's Alan Brown discerned a 20% incidence of schizophrenia in more than 60 individuals whose mothers definitely had rubella during pregnancy.1 More recently, Brown has been presenting data on a large birth cohort in California, which followed roughly 20,000 pregnancies in the 1950s and 1960s. Blood sera were stored and detailed records kept. Brown searched these records to locate offspring displaying symptoms of schizophrenia, interviewed them, and then analyzed maternal samples. Against matched controls, he says, risk for schizophrenia tripled when the mother was exposed to influenza during early to middle pregnancy. Patterson, Fatemi, and others tested such phenomena in rodents. They gave mice respiratory influenza infections nine days into pregnancy and noted behavioral abnormalities in the offspring versus controls.2 But, an animal model for mental illness often raises doubts. "The core symptoms of [schizophrenia] is having something wrong with your thoughts," says Håkan Karlsson, neuroscientist at Karolinska Institutet in Sweden. "[Mice] won't answer whether they experience any of the core symptoms." So, Patterson and his group measured what they could. They charted a mouse exploring an open field or measured its reaction to an acoustic startle test. Normal mice and humans are generally less startled by a loud noise if it is preceded by a quieter tone, a quality called prepulse inhibition. PPI is diminished in mice born to infected mothers, indicating poor sensory motor gating. "Schizophrenics are well known to have trouble screening sensory information," says Patterson. While not necessarily indicative of clinically diagnosed schizophrenia in humans, endophenotypes such as sensory motor gating abnormalities can be modeled and tested in animals. Also, brain pathologies have been noted. A DOUBLE HIT Few infectious hypothesis proponents place the blame squarely on infections, which may not even reach the developing brain of a fetus. Neuroimmunologist Bradley Pearce at Emory University in Atlanta says, "If it were something as simple as a virus causing cell death, you would have more hard signs of a neurological disorder ... it's much more subtle than that." Rather, the mother's immunological reactions to infection may cause changes in brain chemistry and development, whose full effects aren't seen for years. John Gilmore, University of North Carolina, Chapel Hill, injected pregnant rats with lipopolysaccharide, a bacterial endotoxin that models infection. Recently, his group found that offspring had increased production of brain-derived neurotrophic factor and nerve growth factor in the neonatal brain and cortex, respectively.3 Altered neurotrophic factor levels have been associated with brain development abnormalities. Cytokines such as interleukin1-ß (IL1) and tumor necrosis factor-* are believed to regulate these factors. "There's definitely evidence that these cytokines can change the synaptic connectivity or the dendritic development of neurons," says Gilmore. In a study to gauge effects on the brain, Pearce infected neonatal rats with lymphocytic choriomeningitis virus. He says preliminary evidence points to IL1 as the critical cytokine involved in gradual hippocampal cell loss. Other groups, including Patterson's, have tried to establish immune effects as the critical blow in this so-called two-hit hypo-thesis by injecting pregnant rodents with the synthetic double-stranded RNA polyinosinic-polycytidylic acid, or Poly(I:C), which mimics infection.2 Again, behavioral abnormalities, including possible sensory motor gating problems, were noted in the offspring. 3 STILL IN THE GENES Because schizophrenia's causes are rooted in a complex mix of gene/environment interactions, the severity and the effects of an immune response to infection could be genetically determined. Mady Hornig of Columbia University works with various models, including rats infected with borna disease virus. Whether this neurotropic RNA virus actually infects humans is controversial, Hornig says, but some have speculated on connections between BDV and "everything from glioblastomas to obsessive compulsive disorder, multiple sclerosis, and autism." ©2003: Society for Neuroscience EXPLORATORY RESEARCH: The illustrations depict a 10-minute test of open field exploration by a mouse born to an influenza-infected mother (right) and one born to sham-infected mother (left). Sites of rearing on hindpaws are illustrated by circles. The mice responded very differently to the stress of this situation as well as to the introduction of foreign objects or other mice. (L. Shi et al., J Neurosci, 23:297-302, 2003.) Her group also mimics RNA virus infections using Poly I:C prenatally in two different strains of mice, one of which is highly sensitive to autoimmune disease. The cytokine and chemokine patterns produced are very different, she says, and behavioral effects differ slightly. As a path to brain changes, Hornig says that viruses and immune mediators may cause gene expression changes within developmental programs. Neurovirologist Robert H. Yolken, Johns Hopkins School of Medicine, may have found different genetic elements interacting at different time points in patients with schizophrenia. In 2001, he and collaborators, including Torrey and Karlsson, published on the discovery of human endogenous retroviral transcripts in the cerebrospinal fluid of patients with recent-onset schizophrenia.4 Ancient viral remnants, absorbed into the genome roughly 35 million years ago, were actively transcribing pol genes. "The way we think that it fits in with infections," says Yolken, "is that these retroviral sequences are susceptible to activation through cytokines and through infections with other viral agents, particularly the herpes virus family." Herpes viruses such as cytomegalovirus (CMV) have been associated with schizophrenia, and in trying to tease out a causal role, Yolken and collaborators tested the antiviral drug valacyclovir on patients with schizophrenia. Research in press, says Yolken, notes that a subset of individuals whose symptoms improved had antibodies to CMV. "The amount they got better was not dramatic, but it was statistically significant to a pretty high level." Now, Torrey and Yolken say a blinded study will follow the drug's effects. Some skeptics remain adamant. Timothy Crow of the University of Oxford, UK, says he bought into the idea originally, but was convinced that there was no infectious etiology after, among other things, his research failed to confirm horizontal transmission in macaques.5 "It's a universal disease in the sense that all populations have it. I don't think that would be true if it were a viral disease," he says. Other doubters acknowledge that viruses may play a role. "The viral infection might well belong in [the] wide array of things [that] can produce developmental impairment," says Harvard psychologist Brendan A. Maher,* "but, I don't see anything [that] would single out the viral explanation as any more important than any of the other developmental anomalies." Ultimately, schizophrenia's varied nature continues to confound attempts to find a root cause. Pearce says, "If anyone tells you they know what causes schizophrenia, don't believe them." Though an infectious etiology is attractive, some urge caution. "I'm asking for not rushing to judgment because of the heterogeneity involved in the causal chain or chains," says psychiatric geneticist Irving Gottesman, University of Minnesota. He says he and Torrey have often written opposite one another. "We have a difference of opinion as to the role of viruses, and we find ourselves coming closer together to agreement." Brendan A. Maher can be contacted at bmaher@the-scientist.com. * Harvard psychologist Brendan A. Maher is not related to this article's author. References 1. A.S. Brown et al., "Prenatal rubella, premorbid abnor- malities, and adult schizophrenia," Biol Psychiatry, 49:473-86, 2001. 2. L. Shi et al., "Maternal influenza infection causes marked behavioral and pharmacological changes in the offspring," J Neurosci, 23:297-302, 2003. 3. J.H. Gilmore et al., "Maternal infection regulates BDNF and NGF expression in fetal and neonatal brain and maternal-fetal unit of the rat," J Neuroimmunol, 138:49-55, May 2003. 4. H. Karlsson et al., "Retroviral RNA identified in the cerebrospinal fluids and brains of individuals with schizophrenia," Proc Natl Acad Sci, 98:4634-9, 2001. 5. H.F. Baker et al., "A re-investigation of the behavioural effects of intracerebral injection in marmosets of cytopathic cerebrospinal fluid from patients with schizophrenia or neurological disease," Psychol Med, 19:325-9, 1989. function sendData() { document.frm.pathName.value = location.pathname; result = false if (document.frm.score[0].checked) result = true; if (document.frm.score[1].checked) result = true; if (document.frm.score[2].checked) result = true; if (document.frm.score[3].checked) result = true; if (document.frm.score[4].checked) result = true; if (!result) alert("Please select") return result; } .myradio{background-color : #94bee5} Please indicate on a 1 - 5 scale how strongly you would recommend this article to your colleagues? Not recommended 1 2 3 4 5 Highly recommended Please register your vote

The Infection Connection in Schizophrenia

Interested in reading more?

Become a Member of