Core Bioinformatics and Statistics Team, College of Biomedical and Life Sciences, Cardiff University, Cardiff, UK.
MRC Centre for Neuropsychiatric Genetics and Genomics, School of Medicine, Cardiff University, Cardiff, UK.
Transl Psychiatry. 2018 Jan 10;8(1):7. doi: 10.1038/s41398-017-0053-y.
Genome-wide association studies have identified over 100 robust risk loci for schizophrenia with thousands of variants mediating genetic heritability, the majority of which reside in non-coding regions. Analytical approaches have shown this heritability is strongly enriched at variants within regulatory elements identified from human post-mortem brain tissue. However, bulk post-mortem brain tissue has a heterogeneous cell composition, making biological interpretations difficult. We sought to refine the cell types mediating schizophrenia heritability by separating neuronal and glial signals using data from: (1) NeuN-sorted post-mortem brain and (2) cell culture systems. Schizophrenia heritability was partitioned using linkage disequilbrium (LD) score regression. Variants within genomic regions marked by H3K4me3 (marker of active promoters) from NeuN-positive (neuronal) and NeuN-negative (non-neuronal) cells explained a significant amount of schizophrenia heritability (P = 1.38 × 10 and P = 7.97 × 10). However, variants located in H3K4me3 sites specific to NeuN-positive (neuronal) cells were enriched (P = 3.13 × 10), while those specific to NeuN-negative (non-neuronal) cells were not (P = 0.470). Data from cell culture systems mimicked this pattern of association. We show the previously observed enrichment of heritability from variants at brain H3K4me3 sites is mediated by both neuronal and non-neuronal brain cell types. However, only neuronal cell populations showed a unique contribution driven by cell-type specific regulatory elements. Cell culture systems recapitulate disease relevant gene-regulatory landscapes, validating them as a tool for future investigation of genetic mechanisms underlying schizophrenia. Identifying the cell types in which risk variants operate will greatly increase our understanding of schizophrenia pathobiology and aid in the development of novel model systems and therapies.
全基因组关联研究已经确定了 100 多个与精神分裂症相关的稳健风险位点，这些风险位点由数千个变体介导，遗传可遗传性很大，其中大多数位于非编码区域。分析方法表明，这种遗传可遗传性在人类死后脑组织中鉴定的调控元件内的变体中强烈富集。然而，大量的死后脑组织具有异质的细胞组成，使得生物学解释变得困难。我们试图通过使用以下数据从神经元和神经胶质信号中分离出调节精神分裂症遗传力的细胞类型：（1）NeuN 分选的死后脑组织和（2）细胞培养系统。使用连锁不平衡（LD）分数回归对精神分裂症遗传力进行分区。NeuN 阳性（神经元）和 NeuN 阴性（非神经元）细胞中基因组区域内的变体（H3K4me3 的标记，活跃启动子的标记）解释了大量精神分裂症遗传力（P = 1.38×10 和 P = 7.97×10）。然而，位于 NeuN 阳性（神经元）细胞特异性 H3K4me3 位点的变体被富集（P = 3.13×10），而位于 NeuN 阴性（非神经元）细胞特异性 H3K4me3 位点的变体则没有（P = 0.470）。细胞培养系统模拟了这种关联模式。我们表明，先前观察到的在大脑 H3K4me3 位点的变体中遗传力的富集是由神经元和非神经元脑细胞类型介导的。然而，只有神经元细胞群显示出由细胞类型特异性调节元件驱动的独特贡献。细胞培养系统再现了疾病相关的基因调控景观，验证了它们作为未来研究精神分裂症遗传机制的工具。确定风险变体作用的细胞类型将极大地提高我们对精神分裂症病理生物学的理解，并有助于开发新型模型系统和疗法。
