22 May 2008
23 May 2008. Microdeletions at chromosome 22q11 that are associated with schizophrenia in humans alter the production of gene-regulating microRNAs in mice, according to a new report in the May 11 issue of Nature Genetics. The work, from Joseph Gogos and Maria Karayiorgou of Columbia University, New York, zeroes in on deletion of the microRNA-processing gene Dgcr8 as responsible for changes in miRNA expression and contributing to the behavioral and neuronal phenotypes of 22q11 deletion in mice. The work is the first time that microRNAs have been implicated in cognitive dysfunction associated with schizophrenia, and may provide a new handhold on the complex molecular basis of the disease.
In a separate paper, a human genetics study ties alleles of GNB1L, another gene deleted in 22q11 deletion syndrome (22q11DS, encompassing velocardiofacial syndrome and DiGeorge syndrome, among others), to the risk of schizophrenia in several case-control studies. The 22q11 region has been genetically linked to schizophrenia in the general population, and the new results suggest that the alleles, which alter expression of GNB1L, may account for at least some of that linkage. The work, from Nigel Williams and colleagues at Cardiff University in Scotland, was published online last November, and appears in the April print edition of Human Molecular Genetics.
Microdeletions of a 1.5 to 3 Mbase span of DNA at 22q11.2 result in a highly variable syndrome—cardiac or facial/palatal abnormalities are seen in most cases, but there is a long list of other health issues that may affect a given individual with the deletion (see SRF related news story, as well as GeneReviews). The finding of schizophrenia or schizoaffective disorder in about one-third of cases has prompted researchers to comb the region for schizophrenia genes. The work has led to several candidates, identified by their effects on cognitive phenotypes in animals or genetic association studies in humans, or both. In addition to the genes discussed here, recent association studies have cast suspicion on other genes within the 22q11DS region, including the myelin associated gene PIK4CA (Jungerius et al., 2007), the Nogo-66 receptor gene RTN4R (Hsu et al., 2007), and HTF9C, a gene for an RNA binding protein (Liu et al., 2007).
To further study the deletion phenotype, Gogos and Karayiorgou made a mouse model of 22q11DS by removing a 1.5 Mb region of mouse chromosome 16 corresponding to most of the functional genes in the human deletion. First authors Kimberly Stark and Bin Xu headed up the team that found that the mutant mice showed several behavioral phenotypes associated with the cognitive alterations seen in people with schizophrenia. This included a defect in sensory gating detected by the paired pulse inhibition test, as well as effects on learning.
Transcriptional profiling of the hippocampus and prefrontal cortex of the mice revealed a number of changes, and not just in genes that were part of the hemizygous deletion. "One striking finding of our analysis was the divergence in the transcriptional responses in the prefrontal cortex and hippocampus of the mutant mice, which suggests that different pathways (energy metabolism versus synaptic transmission) may be affected or activated (as compensations) in these two brain regions," Gogos told SRF.
In particular, the researchers noticed alterations in levels of microRNA-related transcripts. The researchers reasoned that upregulation of these transcripts might be due to the deletion of the Dgcr8 gene, the product of which is involved in processing of pre-microRNAs to their mature microRNAs. To test this idea, they directly measured levels of pre-microRNAs in hippocampus and PFC, and found that several were indeed elevated. A comprehensive analysis of 386 mature miRNAs revealed 30 that were downregulated in hippocampus, and 59 in prefrontal cortex.
In support of the idea that downregulation of a subset of miRNAs stemmed from loss of Dgcr8, the researchers found similar changes in a Dgcr8 heterozygous knockout mouse. When that mouse was tested for behavioral and cognitive changes, it replicated some but not all of the deficits observed in the deletion mouse, including the paired pulse inhibition defect and a deficit in spatial working memory in a T maze. Thus, loss of Dgcr8 explains part of the mouse deletion phenotype, consistent with the idea that different genes in the region make individual contributions to the overall phenotype.
The researchers also looked at the effect of the genetic manipulations on neuronal morphology. In the 22q11 deletion model, they found a reduction in dendritic spine density and width in hippocampal neurons, as well as alterations in dendritic complexity. The authors note unpublished data showing that deficiency of another gene in the deletion, Zdhhc8, affects dendritic structure and complexity. However, Dgcr8 also had some impact, with the knockout mice revealing decreased dendritic spine width and dendritic complexity. The latter, found in both the Dgcr8 and Zdhhc8 models, as well as the full deletion model, may signal a developmental effect, the authors note. (Interestingly, the first report of a postmortem study of neuropathology of three patients with 22q11 deletion syndrome and schizophrenia reveals evidence for neuronal migration defects in one patient and vascular abnormalities in two others, suggesting both developmental and ongoing processes may play a role in this syndrome. That work, published online in Cerebral Cortex on May 14, comes from Anne Bassett and colleagues at the University of Toronto in Ontario, Canada.)
The 22q11 deletion phenotype in humans is highly variable and clearly involves multiple interacting genes. The new data implicating miRNAs, each of which controls multiple target genes in the brain, adds yet another layer of complexity to the story. At the same time, understanding whether and which miRNAs are affected in people with the 22a11 deletion syndrome and/or schizophrenia could offer a new path to discovery of an additional set of genes, the changing expression of which may contribute to disease.
The second study focuses on other genes in the 22q11 region that have been tied in one way or another to schizophrenia. In mice, hemizygous deletion of two adjacent 22q11 genes, those for the transcription factor Tbx1 and the G-protein b-subunit-like Gnb1L, causes defects in prepulse inhibition, leading Williams and colleagues to look into a possible genetic link between human TBX1 and GNB1L alleles and schizophrenia. By single nucleotide polymorphism (SNP) association studies, he and his colleagues identified a psychosis-associated marker. The association, held up in three different case-control studies, but only for males. A fourth study, a parent-proband study, showed allelic association but not at that marker. The most significant SNPs were also associated with psychosis in males in a sample of 22q11 deletion syndrome patients. Full sequencing of the TBX1 and GNB1L exons showed no changes that could account for the allelic linkage, but the psychosis-related alleles were associated with changes in GNB1L expression, but not TBX1 in human postmortem tissue. The authors conclude that the data provide significant evidence for association between schizophrenia and GNB1L genotype, and that abnormal gene expression could be a possible mechanism. However, that is not the only possible explanation for their data, and more work will be needed to sort out effects of the alleles on disease.—Pat McCaffrey.
Stark KL, Xu B, Bagchi A, Lai WS, Liu H, Hsu R, Wan X, Pavlidis P, Mills AA, Karayiorgou M, Gogos JA. Altered brain microRNA biogenesis contributes to phenotypic deficits in a 22q11-deletion mouse model. Nat Genet. 2008 May 11. [Epub ahead of print] Abstract
Williams NM, Glaser B, Norton N, Williams H, Pierce T, Moskvina V, Monks S, Del Favero J, Goossens D, Rujescu D, Giegling I, Kirov G, Craddock N, Murphy KC, O'Donovan MC, Owen MJ. Strong evidence that GNB1L is associated with schizophrenia. Hum Mol Genet. 2008 Feb 15;17(4):555-66. Epub 2007 Nov 13. Abstract
Kiehl TR, Chow EW, Mikulis DJ, George SR, Bassett AS. Neuropathologic Features in Adults with 22q11.2 Deletion Syndrome. Cereb Cortex. 2008 May 14. [Epub ahead of print] Abstract