DISC1, differential expression, and deconvolution; human and mouse RNA-Seq investigation of a t(1;11) translocation that predisposes to major mental illness
Item statusRestricted Access
Embargo end date27/06/2021
The t(1;11) translocation is a mutation unique to a Scottish pedigree, members of which have been diagnosed with schizophrenia, bipolar disorder, recurrent major depressive disorder and other related disorders. The translocation is significantly linked to increased risk of these diagnoses. It disrupts three genes, only one of which, DISC1, encodes a protein. A number of experiments have explored the function of DISC1 as a molecular scaffold and developmental regulator. DISC1 and its interactors have roles in processes of relevance to psychiatric disease. These include neuronal precursor proliferation, migration and integration in the developing and adult brain, neurite outgrowth, mitochondrial activity, which is particularly important in neurons due to their high energy demands, and intracellular trafficking, especially critical in neurons due to their highly elongated morphology. Although various DISC1 mutations have been investigated in the past, it is only with advances in technology that neural cells derived directly from translocation carriers, and therefore carrying the translocation plus their genetic background, have been generated and analysed. In addition a recently described mouse model mimics the effects of the translocation upon DISC1 expression. It does so by removing endogenous Disc1 exons corresponding to those distal to the breakpoint in translocation carriers, and fusing the remaining endogenous 5’ Disc1 genomic sequence to human chromosome 11 genomic sequence distal to the translocation breakpoint. The result is a chimeric gene with 5’ mouse Disc1 joined to a segment of human DISC1FP1, the non-coding fusion partner of DISC1 located on chromosome 11. This leads to loss of wild-type Disc1 and prediction of chimeric transcripts encoding aberrant C-terminally truncated forms of Disc1. This thesis builds on the work of previous researchers to characterise the RNAsequenced transcriptome of ‘cortical’ neurons derived from induced pluripotent stem cells from various members of the pedigree. Both heterozygous and homozygous mutant mice have also been utilised to generate RNA-sequencing data from the hippocampus and cortex. The thesis not only describes the differential expression of genes and exons, but also carries out a series of analyses to examine whether proportions of certain cell types are altered, as well as whether differentially expressed genes are highly associated with specific cell types. RNA-Seq data have been analysed for differential expression at the gene and individual exon level using DESeq2 and DEXSeq, respectively. This has revealed over 1,200 differentially expressed genes in human neurons carrying the translocation, which predict changes to functions relating to intracellular transport and synaptic activity. In addition, a number of genes have been verified by RT-qPCR as being differentially expressed in these neurons. These include genes of known relevance to schizophrenia such as DRD2, which encodes the D2 dopamine receptor, NTRK2, which encodes the BDNF receptor NTRK2, and BBS1 which encodes the DISC1 interactor and centrosomal protein BBS1. The human neurons also show significant overlap with previously published dysregulated genes in human neurons carrying other DISC1 mutations, as well as with genes associated with schizophrenia by large-scale genome wide association and copy number variation studies. Human neuron RNA-Seq data have also been examined for evidence of local effects of the translocation upon gene expression, and no obvious strong effect was found. The pattern of gene dysregulation in heterozygous mutant mouse cortex overlaps with that of the mutant human neurons. Gene expression changes in the mutant mouse cortex have also been verified by RT-qPCR in the genes Arc and Avp, and the list of implicated genes also shows overlap with genes associated with schizophrenia by large-scale genome wide association and copy number variation studies. An RNA-Seq deconvolution analysis was carried out to look for evidence of altered proportions of cell types at both the broad and more specific cell type level. This compared the observed expression of hundreds of genes in in the RNA-Seq samples against their expression in publically available RNA-Seq data of specific cell types. There does not appear to be any strong and consistent effect of the t(1;11) or mouse mutation on cell proportions. However, the data indicate greater than expected dysregulation of genes that are highly enriched in specific cell types. This includes certain subtypes of astrocyte. Mutant mouse cortex also shows dysregulation of genes associated with several subtypes of interneuron and pyramidal neuron, including parvalbumin positive interneurons. This indicates that, while the proportions of cell types appears to be unaffected by the translocation or mouse mutation, specialised cellular functions may be perturbed. To conclude, this thesis highlights a number of processes which appear to be disturbed by the translocation and mouse mutation. In all models, RNA-Seq evidence suggests signalling pathways of known relevance to psychiatric disease have been affected without significant alteration of cell proportions. This concurs with histological analyses of the mouse model by previous researchers. This thesis also describes the overlap between genes implicated in the study of this unique mutation as well as those implicated by studies seeking common or rare mutations predisposing to schizophrenia, supporting the hypothesis that different genomic risk variants and mutations converge upon certain molecular pathways that are especially important in this illness. The implication that the t(1;11) may alter the activities of certain cell types is also notable and future work can elucidate the cell-specific effects of the translocation.