Researchers Pry into Schizophrenia's Stubborn Secrets

Adapted from image courtesy of Lynn D. Selemon Stereologic cell counting of prefrontal cortical area 9. Nissl-stained neurons (shown as triangles and dots) were counted directly in a stack of three-dimensional boxes extending throughout all layers of the cortex. Could any major illness be more difficult to study than schizophrenia? Despite unprecedented advances in research over recent years, largely aided by improved neuroimaging technologies, little is known for sure about either its origins or its mechanisms. No responsible genes have been located, no environmental effects firmly implicated, no developmental cause identified, no neurodegeneration proven, and no biological processes established as leading to specific symptoms. Like many mental illnesses, schizophrenia appears to involve not only complex genetics, but also environmental contributions that interact both with the genes and independently of them, notes Daniel P. Weinberger, chief of the clinical brain disorders branch at the National Institute of Mental Health. He points out that the risk of heritability is similar in most psychiatric illnesses. "This raises the question that maybe psychiatric illnesses share heritability genes. What may be unique about these illnesses may not be the genes they manifest so much as the environments that interact with these genes." The search for genetic causes of the disease has been as frustrating as it has been encouraging. Susceptibility has been tentatively linked to chromosomes 6 and 8, and there is other, less compelling evidence of linkage to five more chromosomes, but none of these findings have been clearly confirmed.1 Weinberger believes that schizophrenia will not turn out to be a discrete genetic disorder; instead, it will be seen as a compilation of abnormalities resulting from a variety of elementary biological processes. Although hallucinations and delusions are its most infamous symptoms, the disease primarily involves more subtle impairments in cognition and perception. They include: memory disturbances, attentional problems, eye-movement disorders, and defects in sensory "gating" or filtering of stimuli. These deficits have inspired promising attempts to construct intermediate or "endophenotype" models. Each model focuses on a specific biological process that evidence indicated may underlie an endophenotype. The goals are to uncover a neuronal problem underlying the endophenotype, and to aid in pinpointing its genetic origins. Such problems could involve abnormal development and adaptation of neuronal circuitry, fundamental aspects of information processing, or the responsiveness of specific neuronal systems to events in the environment.2 "I think the exciting thing for all of us is the potential for interactions between genetic research and neurobiological research," says Robert Freedman, a psychiatry professor at the University of Colorado Health Sciences Center. He and colleagues have pioneered the endophenotype approach over the past decade, researching a sensory gating deficit in schizophrenic patients that involves response to sounds. The P50 response is a positive wave that occurs 50 milliseconds after an auditory stimulus. In normal subjects, inhibitory mechanisms are then activated, and when the sound is quickly repeated, the P50 response is suppressed. But schizophrenic patients take significantly longer to suppress the response than do normal subjects. Freedman and coworkers contend that this deficit, found not only in the patients but in many of their first-degree relatives in a manner consistent with possible autosomal dominance, likely reflects a genetically transmitted abnormality.3 Theoretically, sensory gating problems could account for the overwhelming stimuli of which schizophrenia victims complain, a flood of new and old input that they can't filter. As Freedman notes, "We all have a great ability to see, and a limited ability to understand what we see." Sensory gating deficits not only predispose schizophrenia sufferers to attentional and learning problems because they can't selectively admit new information, but they also could lead to delusional thinking, hallucinations, and social dysfunction. Courtesy of Neal R. Swerdlow Albino rat in an acoustic startle chamber used to test prepulse inhibition. Freedman's group found the gene for this gating deficit on chromosome 15, overlapping with the alpha7-nicotinic cholinergic receptor locus.4 Their theory is that altered expression and function of this receptor could cause the gating deficit. However, the vast majority of schizophrenic patients are heavy smokers, and Freedman acknowledges that the altered nicotinic receptor might convey a risk primarily related to the experience of smoking, rather than to schizophrenia. One attraction of gating studies is that some are amenable to animal modeling, never easy for psychiatric illnesses that are largely characterized by psychological symptoms. Neal R. Swerdlow, associate professor of psychiatry at the University of California at San Diego, studies prepulse inhibition (PPI), which is the normal suppression of the startle reflex when a weak stimulus precedes an intense stimulus.5 Virtually all mammals exhibit PPI, but it is diminished in schizophrenia. Again, the theory is that this defective trait is a measure of the loss of sensorimotor gating, which could lead to sensory flooding. Excessive dopamine transmission in rats affects PPI, paralleling the deficits observed in schizophrenia, but whether elevated dopamine activity in humans causes the problem is still to be determined. Swerdlow's group and others have found deficient PPI in patients with numerous psychiatric illnesses, showing that the condition results not from a specific disease but from abnormalities within the startle gating circuit. In rats, those deficits are a reaction to high doses of a dopamine agonist, whereas in schizophrenia, they derive from a complex combination of genetic and other factors. It's therefore important to show that startle gating in rats is sensitive to those same factors. Studies with rats reared in isolation indicate that they exhibit elevated dopamine activity and demonstrate diminished PPI compared to group-reared controls. Similar isolation of adult rats failed to produce the same PPI deficit, suggesting that being reared in isolation was the determining factor.6 "With animal models," Swerdlow concludes, "as long as you define the hypothesis and maintain healthy limits in interpretation of results, you can actually ask a question and get an answer." Reduced Neuropil Hypothesis Lynn Selemon Another way to get answers is through postmortem studies. In 1995 neurobiologist Lynn D. Selemon and colleagues at the Yale University School of Medicine reported increased neuronal density in the cortex of schizophrenic brains compared to normal control brains.7 Although most neurologic diseases are associated with cell loss, the Yale group's study did not support that conclusion. Neuroimaging data suggest a smaller than normal cortex in schizophrenia, and from this and other information, the team concluded that the distance between neurons in the schizophrenic cortex is diminished, while the number of neurons probably remains the same. The space between the cells, called the neuropil, contains neuronal processes and synaptic contacts. They hypothesized that reduced neuropil is responsible for cognitive disturbances in schizophrenia, either due to atrophy of neuronal processes and connections, or a failure of them to develop. They measured neuropil reductions of 10 to 21 percent in different brain areas, a relatively subtle pathology, not sufficient to kill neurons."This is really distinct from the kinds of degenerative processes seen in Alzheimer's or Huntington's disease," Selemon notes. She thinks the relative sparsity of connections caused by reduced neuropil leads to the deficiencies in learning and memory found in schizophrenic patients. Much work remains to determine the precise nature of the neuropil alterations. Electron microscopy, the best technology for postmortem studies of synaptic contacts, is limited in the human brain because brains must be chemically fixed within two hours of death to ensure adequate preservation, she says. The longstanding question of whether, ultimately, schizophrenia is a neurodevelopmental or a neurodegenerative disease, or both, remains unresolved. The clinical significance of this debate is that if there is neurodegeneration, early intervention might help. "There's excellent evidence for developmental risks and factors causing schizophrenia," Selemon says, "but there's also other evidence that has to be taken into account." Weinberger, a formulator of the widely held theory that an early "lesion" or insult to the fetus causes a neurodevelopmental problem,8 states, "I think the evidence for neurodegeneration in schizophrenia is just not there." Swerdlow suggests that an early trauma "could set up a condition to be vulnerable to other insults." Freedman finds the dual hypothesis plausible, offering, "I think the theme is likely to be played out with other complex illnesses also." Weinberger applauds the lateral thinking in the many pathophysiological models being pursued: "A lot of these directions will turn out to be blind alleys, but the critical thing is that the science be reviewed and examined. You don't want to micromanage science too much, because you might be wrong." Steve Bunk (sbunk@uswest.net) is a contributing editor for The Scientist. S. Moldin, "Genetics and mental disorders: summary of research," Biological Psychiatry, 45:573-602, March 1, 1999. D.R. Weinberger, "Schizophrenia: new phenes and new genes," Biological Psychiatry, 46:3-7, July 1, 1999. L.E. Adler et al., "Elementary phenotypes in the neurobiological and genetic study of schizophrenia," Biological Psychiatry, 46:8-18, July 1, 1999. R. Freedman et al., "Linkage of a neurophysiological deficit in schizophrenia to a chromosome 15 locus," Proceedings of the National Academy of Sciences, 94:587-92, 1997. N.R. Swerdlow, M.A. Geyer, "Using an animal model of deficient sensorimotor gating to study the pathophysiology and new treatments of schizophrenia," Schizophrenia Bulletin, 24:285-301, 1998. L.S. Wilkinson et al., "Social isolation in the rat produces developmentally specific deficits in prepulse inhibition of the acoustic startle response without disrupting latent inhibition," Neuropsychopharmacology, 10:61-72, 1994. L.D. Selemon et al., "Abnormally high neuronal density in the schizophrenic cortex: a morphometric analysis of prefrontal area 9 and occipital area 17," Archives of General Psychiatry, 52:805-18, 1995. D.R. Weinberger, "Implications of normal brain development for the pathogenesis of schizophrenia," Archives of General Psychiatry, 44:660-9, 1987. "A fortuitous accident" is how neurobiologist Anthony-Samuel LaMantia describes the genesis of a model he constructed to explain the symptoms of schizophrenia, which are thought to arise from aberrant forebrain function. In recent years he has studied the mechanisms of normal forebrain development in mice, particularly the induction of proper cell organization and the role of a signaling molecule, the steroidlike hormone retinoic acid. After describing forebrain induction in a talk to a group of psychiatrists, he was approached by University of North Carolina (UNC) Medical School psychiatry professor Jeffrey A. Lieberman. He asked if LaMantia was aware that the minor structural abnormalities in the head and face that accompany developmental miscues in this process are found in schizophrenia. LaMantia was intrigued, and later read a paper providing genetic evidence to suggest that retinoidal dysfunctions cause schizophrenia.1 His own model, implicating the retinoid-mediated induction process, soon followed.2 LaMantia, who was at Duke University, moved to UNC, where he now is collaborating with Lieberman and others. They are interested in the hypothesis that a mechanism conferring plasticity on forebrain circuitry might be defective in schizophrenia, possibly due to a neurodevelopmental "lesion" in the fetus. A progressive problem could then ensue, involving modulation of plasticity rather than neurodegeneration. Noting that about 1 percent of populations across the world suffer from schizophrenia, LaMantia comments, "This suggests to me there's probably something that's biological." But finding concrete evidence is immensely difficult. "There's an old joke about schizophrenia being the graveyard of neuropathology," he chuckles. "I'm wondering if it will open up a new grave for neurobiology." --Steve Bunk

Researchers Pry into Schizophrenia's Stubborn Secrets

Interested in reading more?

Become a Member of