Dataset Information

Building a schizophrenia genetic network: Transcription Factor 4 regulates genes involved in neuronal development and schizophrenia risk

ABSTRACT: The transcription factor 4 (TCF4) locus is a robust association finding with schizophrenia (SZ), but little is known about the genes regulated by the encoded transcription factor. Therefore, we conducted chromatin immunoprecipitation sequencing (ChIP-seq) of TCF4 in neural-derived (SH-SY5Y) cells to identify genome-wide TCF4 binding sites, followed by data integration with SZ association findings. We identified 11,322 TCF4 binding sites overlapping in two ChIP-seq experiments. These sites are significantly enriched for the TCF4 Ebox binding motif (>85% having ≥1 Ebox) and implicate a gene set enriched for genes down-regulated in TCF4 siRNA knockdown experiments, indicating the validity of our findings. The TCF4 gene set was also enriched among 1) Gene Ontology categories such as axon/neuronal development, 2) genes preferentially expressed in brain, in particular pyramidal neurons of the somatosensory cortex, and 3) genes down-regulated in post-mortem brain tissue from SZ patients (OR=2.8, permutation p<4x10-5). Considering genomic alignments, TCF4 binding sites significantly overlapped those for neural DNA binding proteins such as FOXP2 and the SZ-associated EP300. TCF4 binding sites were modestly enriched among SZ risk loci from the Psychiatric Genomic Consortium (OR=1.56, p=0.03). In total, 130 TCF4 binding sites occurred in 39 of the 108 regions published in 2014. Thirteen genes within the 108 loci had both a TCF4 binding site ±10kb and were differentially expressed in siRNA knockdown experiments of TCF4, suggesting direct TCF4 regulation. These findings confirm TCF4 as an important regulator of neural genes and point towards functional interactions with potential relevance for SZ.

ORGANISM(S): Homo sapiens

PROVIDER: GSE112704 | GEO | 2018/04/05

REPOSITORIES: GEO

ACCESS DATA

Shared Molecules

Only show the datasets with similarity scores above: 0.5

Threshold

0.5

Similar Datasets

Project description:Cell-based models of many neurological and psychiatric diseases, established by reprogramming patient somatic cells into human induced pluripotent stem cells (hiPSCs), have now been reported. While numerous reports have demonstrated that neuronal cells differentiated from hiPSCs are electrophysiologically active mature neurons, the âageâ of these cells relative to cells in the human brain remains unresolved. Comparisons of gene expression profiles of hiPSC-derived neural progenitor cells (NPCs) and neurons to the Allen BrainSpan Atlas indicate that hiPSC neural cells most resemble first trimester neural tissue. Consequently, we posit that hiPSC-derived neural cells may most accurately be used to model the early developmental defects that contribute to disease predisposition rather than the late features of the disease. Though the characteristic symptoms of schizophrenia SZ generally appear late in adolescence, it is now thought to be a neurodevelopmental condition, often predated by a prodromal period that can appear in early childhood. Postmortem studies of SZ brain tissue typically describe defects in mature neurons, such as reduced neuronal size and spine density in the prefrontal cortex and hippocampus, but abnormalities of neuronal organization, particularly in the cortex, have also been reported. We postulated that defects in cortical organization in SZ might result from abnormal migration of neural cells. To test this hypothesis, we directly reprogrammed fibroblasts from SZ patients into hiPSCs and subsequently differentiated these disorder-specific hiPSCs into NPCs. SZ hiPSC differentiated into forebrain NPCs have altered expression of a number of cellular adhesion genes and WNT signaling. Methods: We compared global transcription of forebrain NPCs from six control and four SZ patients by RNAseq. Results: Multi-dimensional scaling (MDS) resolved most SZ and control hiPSC NPC samples; 848 genes were significantly differentially expressed (FDR<0.01) Conclusions: The WNT signaling pathway was enriched 2-fold (fisher exact test p-value = 0.031). 1-2 independent differentiations (biological replicates) for each of four control and four schizophrenia patients were analyzed; samples were generated in parallel to neuron RNAseq data.

Project description:Cell-based models of many neurological and psychiatric diseases, established by reprogramming patient somatic cells into human induced pluripotent stem cells (hiPSCs), have now been reported. While numerous reports have demonstrated that neuronal cells differentiated from hiPSCs are electrophysiologically active mature neurons, the “age” of these cells relative to cells in the human brain remains unresolved. Comparisons of gene expression profiles of hiPSC-derived neural progenitor cells (NPCs) and neurons to the Allen BrainSpan Atlas indicate that hiPSC neural cells most resemble first trimester neural tissue. Consequently, we posit that hiPSC-derived neural cells may most accurately be used to model the early developmental defects that contribute to disease predisposition rather than the late features of the disease. Though the characteristic symptoms of schizophrenia SZ generally appear late in adolescence, it is now thought to be a neurodevelopmental condition, often predated by a prodromal period that can appear in early childhood. Postmortem studies of SZ brain tissue typically describe defects in mature neurons, such as reduced neuronal size and spine density in the prefrontal cortex and hippocampus, but abnormalities of neuronal organization, particularly in the cortex, have also been reported. We postulated that defects in cortical organization in SZ might result from abnormal migration of neural cells. To test this hypothesis, we directly reprogrammed fibroblasts from SZ patients into hiPSCs and subsequently differentiated these disorder-specific hiPSCs into NPCs. SZ hiPSC differentiated into forebrain NPCs have altered expression of a number of cellular adhesion genes and WNT signaling. Methods: We compared global transcription of forebrain NPCs from six control and four SZ patients by RNAseq. Results: Multi-dimensional scaling (MDS) resolved most SZ and control hiPSC NPC samples; 848 genes were significantly differentially expressed (FDR<0.01) Conclusions: The WNT signaling pathway was enriched 2-fold (fisher exact test p-value = 0.031).