DISC1 has emerged as one of the strongest candidate genes for mental illness. It was first identified in a large Scottish family, in which a balanced translocation that disrupts the DISC1 gene segregated with major mental illness. Although this chromosomal translocation appears to be limited to the Scottish family, numerous subsequent association studies provided further evidence for a potential role of DISC1 in schizophrenia and affective disorders. There has been, however, no single polymorphism or even gene region consistently associated in various studies and populations, and molecular mechanisms of the association with mental disorders are still unclear. This is why mouse models may be particularly useful in elucidating pathological involvement of DISC1 in brain development and function.

One of the strongest clues about the type of pathology in DISC1 expected to occur in psychiatric patients is provided by the initial finding of translocation that could potentially result in abnormal truncation of DISC1 protein. However, because the truncated DISC1 protein, a product of translocation, has not yet been found in any individual, it has been assumed that the causative problem in the Scottish family, and possibly in other individuals with mental illness, is not a truncation itself but haploinefficiency and reduced expression of the full-length DISC1 protein. The evidence for this is also lacking, however.

Despite these caveats, several groups of investigators attempted to mimic putative truncation in transgenic mouse models. This paper provides another example of such a model, in which truncated DISC1 protein is overexpressed on the background of normally expressed endogenous full-length DISC1. The design of the transgene expression appears perhaps more biologically relevant in this study than in other models, because it mimics endogenous DISC1 both in terms of the temporal and spatial expression patterns. These DISC1 transgenic mice show several morphological abnormalities, including dilation of lateral ventricles, thinning of the prefrontal cortex, and agenesis of the corpus callosum. The changes are quite dramatic and one has to wonder whether they indeed mimic changes in schizophrenia or whether this model is more relevant for more severe cases of mental retardation.

The authors also demonstrate molecular changes in these mice, including reduced parvalbumin immunoreactivity in certain layers of the prefrontal cortex. With regard to schizophrenia, it would be interesting to investigate more specific changes in the GABA system, including one of the most replicable findings in schizophrenia, i.e., reduced levels of GAD67 mRNA expression. The authors also evaluated the behavioral phenotype of their transgenic model by assessing latent inhibition responses and measures of depressive-like behaviors. With regard to schizophrenia, however, one would like to see whether the DISC1 transgenic mice also exhibit working memory problems, believed to be core features of schizophrenia, as well as deficits in social interactions, increased anxiety, and prepulse inhibition deficits. At the end, the authors provide an excellent summary of findings on DISC1 transgenic mouse models in the table that can be used as a source of reference. It illustrates similarities between various transgenic models and thus suggests that many different problems in the function of the DISC1 gene may result in a similar behavioral/molecular phenotype, much the same as predicted for the human mental illness.

View all comments by Barbara K. Lipska
View all comments by Erin Fink