Schizophrenia Research Forum - A Catalyst for Creative Thinking

ICOSR 2011—Translating GxE in Schizophrenia into Etiology and Biology

As part of our ongoing coverage of the 2011 International Congress on Schizophrenia Research (ICOSR), 2-6 April, in Colorado Springs, Colorado, we bring you session summaries from some of the Young Investigator travel award winners. For this report, we thank Bart P.F. Rutten, of the Maastricht University Medical Centre in the Netherlands.


At the 2011 International Congress on Schizophrenia, Mikhail Pletnikov, John Hopkins School of Medicine, Baltimore, Maryland, chaired an afternoon session on Monday, 4 April, called, “Gene and Environment Interplay in Schizophrenia: Basic and Clinical Advancements.” He emphasized that schizophrenia research is now in the phase where researchers can commence investigations on gene-environment interplay in both etiology and pathophysiology. The work presented in this session provided excellent examples of observational human studies (i.e., genetic epidemiology) as well as experimental animal studies, and serves as an important starting point for further translational research.

In the session’s first presentation, entitled “Psychosis and Differential Sensitivity to Trauma,” my colleague Jim van Os, Maastricht University Medical Centre, Maastricht, The Netherlands, began with an overview of the current epidemiologic literature, including several recent reports on prospectively sampled cohorts with relatively good indices of trauma exposure, which converges onto the conclusion that trauma is probably associated with psychosis. He then reported unpublished findings from the Maastricht Trauma study of more than 750 subjects (schizophrenia patients, siblings of schizophrenia patients, and controls), in which trauma exposure was significantly associated with psychotic symptoms in cases, and with scores of schizotypy in the sibling and control groups. Next, van Os presented findings from the Dutch NEMESIS study that early adversities (such as physical punishment at age seven years) increased the cannabis effect size on psychosis scores, which is suggestive of synergism between early adversity and cannabis use on psychosis risk.

Using experience sampling methodologies in a large twin study, his group showed that the impact of childhood trauma exposure on stress sensitivity (i.e., the effect of daily life stressors on negative affect) was influenced by the level of genetic risk. Differential sensitivity to trauma was also observed on the neuroanatomical level in the Maastricht subsample of the Dutch G.R.O.U.P study: the effects of trauma on cortical thickness showed significant differences between the patient and the control group. Other analyses showed that the effects of childhood trauma exposure on lifetime presence of hallucinations also significantly differed between the sibling and the control group. Taken together, these findings presented by van Os indeed suggest the presence of differential sensitivity to trauma, as well as environmental synergism in psychosis.

John Waddington, Royal College of Surgeons in Ireland, Dublin, then presented his work on “Biological and Psychosocial Adversities Across the Developmental Trajectory of Schizophrenia: Gene X Environment Interactions in Mutant Mice.” He started his presentation by illustrating a theoretical framework in which psychosis is the consequence of interactions between genetic predisposition influencing brain morphogenesis and exposures to environmental factors throughout the lifespan. Waddington also pointed out the great challenge of translating schizophrenia-related phenotypes to relevant mouse models. Using the behavioral phenotypes of locomotor activity, working memory, recognition memory, sociability, social novelty preference, prepulse inhibition, and anxiety as outcome measures, he has focused his research on the interaction between schizophrenia candidate genes and three environmental exposures: 1) maternal infection during pregnancy, 2) psychosocial stress during adolescence and early adulthood, and 3) substance abuse during adolescence and early adulthood.

Waddington's group found that maternal immune activation (through administration of polyriboinosinic-polyribocytidillic acid (poly I:C) decreased sociability at adult age in offspring of wild-type (i.e., controls) dams. This immune-related increase in sociability in adulthood was, however, attenuated in offspring of dams transgenic for mutant neuregulin-1 (NRG1). He then showed evidence that repeated social defeat during adolescence impaired spatial working memory in NRG heterozygous mutant mice, but not in wild-type controls. Finally, Waddington showed data indicating that chronic administration of THC (the psychoactive component of cannabis) during adolescence did not affect spatial working memory in adult wild-type mice, but did affect such memory in COMT knockout mice—the knockout mice have enhanced spatial working memory, but it could be reversed by chronic THC administration. Thus, these findings from experimental animal studies illustrate nicely that psychosis-related phenotypes may be influenced by the interaction of risk genes with both biological and psychosocial environmental adversities operating at critical time points across development.

The last presentation of the session, entitled “Genetically Compromised Neurons and Astrocytes Shape Neuroimmune Dysfunctions in Schizophrenia,” was by Mikhail Pletnikov. He presented his findings on mutant human DISC1 mice, which feature enlargement of the lateral ventricles, decreased cortical volumes, fewer parvalbumin-positive cells in cortex, and an increased dendritic spine density in the hippocampal dentate gyrus. At the behavioral level, the mice show hyperactivity and enhanced responses to stimulants, increased aggression, and attenuated spatial memory. Interestingly, the effects of mutant human DISC1 expression on the behavioral outcome measures were different when expression of mutant human DISC1 occurred prenatally versus postnatally.

As maternal immune response have been proposed as a leading pathogenic factor in schizophrenia, Pletnikov's group tested whether an acute systemic challenge with poly I:C to pregnant dams would differentially impact mutant human DISC1 versus wild-type offspring. Pletnikov and colleagues found that mutant human DISC1 modulated basal and poly I:C-induced secretion of soluble inflammatory factors, and that prenatal interactions between mutant DISC1 and immune activation were required for some abnormal behavioral phenotypes, mainly alterations in affective phenotypes such as forced swim test, elevated plus maze test, and the sociability test. In addition, mutant human DISC1 and in utero immune challenge synergistically interacted to 1) attenuate lateral ventricle enlargement, 2) decrease volume of the amygdala and the periaqueductal gray, and 3) decrease dendritic spine density in the dentate gyrus of the hippocampus. These elegant studies are a good starting point to scrutinize the interactive effects of mutant DISC1 and maternal immune activation on psychosis-related phenotypes and neurobiology in mice. The last part of Pletnikov’s presentation focused on the role of DISC1 in astrocytes and gave future perspectives on animal models of toxoplasmosis infection and its relation to psychosis.

Together, these observational human and experimental animal studies provided a first glance at the wide possibilities and opportunities in translating epidemiologic research in schizophrenia, but they also underscored the major challenges in GxE research. In schizophrenia, for example, a tight focus is critical, as one can easily get lost in testing the plethora of possible GxE interactions, and it is important to employ multidisciplinary investigations combining appropriate timing of environmental exposures during development, as well as dynamic alterations in (mutant) gene expression with cross-species-valid measurements of psychosis-relevant phenotypes.—Bart P.F. Rutten.