中国科学院昆明动物研究所机构知识库

精神分裂症是一种复杂异质性的涉及到行为和认知障碍的精神疾病，它是全球25大导致残疾的疾病之一。临床症状分为阳性、阴性以及阳性和阴性混合型。由于精神类疾病的种类众多，并且临床症状有交叉性，因此目前对于精神分裂症的鉴定存在一定的困难。其病理学特征有大脑灰质的减少、多巴胺能神经元和谷氨酸能神经元神经传导功能的异常。目前对于精神分裂症没有很好的治疗药物，已知的一些药物基本都是通过作用于多巴胺D2受体来发挥药效，并且都存在一定的副作用。精神分裂症的病因学目前还不是很清楚，可能是大脑在发育的过程中受到了遗传、环境以及遗传和环境共同影响的结果。精神分裂症的环境风险因素包括母亲在妊娠期间的感染、儿童时期生活环境的变化、脑损伤以及毒品的滥用等。它的遗传力在目前几乎所有已知的精神类疾病中最高，约为0.8，也就是说遗传因素贡献了其发病的80%。精神分裂症的遗传研究主要有连锁分析和关联研究。连锁分析从家族性的精神分裂症入手，找到了众多的稀有且效力值较大的位点。而关联分析则从散发性的患者和正常对照入手，找到了一些常见的但是效力值较小的位点。在最近的十年中，利用全基因组关联研究（Genome Wide Association Study，GWAS）、全基因组测序、外显子组测序等技术定位到了一系列与精神分裂症发病相关的风险位点，使精神分裂症的遗传研究取得了显著的进展。另外，需要注意的是一些大片段的拷贝数变异或者结构变异如22q11.2缺失等，虽然在人群中的频率较低，但是对精神分裂症的发病贡献却很大。虽然精神分裂症的GWAS鉴定到了众多的遗传风险位点，但是大部分的位点都位于非编码区，这些位于非编码区的位点是如何影响精神分裂症的发生，还需要做进一步的研究，这也是后GWAS时代亟待解决的问题。另一方面，随着各种生物测序技术的发展以及大数据时代的来临，在最近的这些年里，各种组学数据如甲基化测序、代谢组测序、蛋白组测序等层出不穷。整合GWAS与这些多组学数据能够从不同的方面解析精神分裂症的发病机制，为精神分裂症的治疗或者预防提供基础。本研究中，我们首先整合了精神分裂症的GWAS数据和大脑皮层的表达数量性状位点（Expression Quantitative Trait Loci, eQTL）信号。结果显示BTN3A2与精神分裂症风险显著相关，位于BTN3A2 3’UTR区域内的GWAS显著单核苷酸多态性（Single Nucleotide Polymorphisms，SNP）位点rs1979能显著地影响BTN3A2 mRNA的表达。使用不同的整合分析方法与eQTL数据集，我们同样鉴定到了该基因。查看大脑相关的eQTL数据集发现rs1979的G等位风险基因能够显著的上升BTN3A2基因的表达。进一步的研究发现，相比于正常对照，BTN3A2在精神分裂症患者的大脑皮层和诱导多能干细胞（induced Pluripotent Stem Cells，iPSC）诱导分化的神经元中都显著上升，这与之前的整合分析结果吻合。使用大鼠的海马脑片作为研究体系记录双细胞的电生理，我们发现过表达BTN3A2能够显著的降低α-氨基-3羟基-5甲基-4异恶唑受体（L-α-amino-3-hydroxy-5-methy-4-isoxazole propionic acid receptor，AMPAR）和N-甲基-D-天冬氨酸受体（N-methy-D-aspartate receptor，NMDAR）介导的兴奋性突触传递电流，并且过表达BTN3A2后，配对脉冲比例也增加了。这表明BTN3A2可以通过抑制突触前的释放，来降低兴奋性的突触传递电流。另外，过表达BTN3A2后对γ-氨基丁酸（gama-aminobutyric acid，GABA）受体介导的抑制性突触传递电流没有影响，这表明BTN3A2只能特异性的作用于兴奋性的突触传递电流。细胞表面结合实验证实BTN3A2可以与突触前的轴突蛋白相结合，但是不能与轴突后神经连接蛋白相结合。这些研究结果表明BTN3A2可以和突触前的轴突蛋白相互作用，进而影响突触前的释放，来特异性地作用于兴奋性的突触传递。这些通过整合分析和功能实验的结果表明，BTN3A2可能是通过调节兴奋性/抑制性的突触平衡来影响精神分裂症的发生。为了方便科研人员获得与精神分裂症相关的遗传研究大数据，我们搭建了精神分裂症的遗传数据库SZDB（www.szdb.org）。我们收集了来自连锁分析、关联研究、基因表达、表观遗传学、蛋白相互作用网络等各方面的数据，并对这些数据进行了深入的分析，将分析的结果整合到数据库中。通过对来自各方面的与精神分裂症相关的数据进行基因打分，形成了本数据库独具特色的基因打分系统。对得分较高的基因进行共表达分析与蛋白相互作用分析发现，这些基因会形成共表达网络与蛋白相互作用网络。本数据库的搭建为精神分裂症科研工作者提供了一个很好的基因检索、浏览、分析平台，方便了他们的研究。综上所述，本研究整合了精神分裂症的GWAS数据与多组学数据，并通过功能实验证实了BTN3A2可能是精神分裂症的一个潜在风险基因。另一方面，我们整合了多组学数据，并搭建了精神分裂症的遗传数据库，方便了相关科研工作者的研究。

Schizophrenia is a complex, heterogeneous and cognitive syndrome, which constitutes one of the twenty-five leading causes of disability. The clinical features of schizophrenia are classified as positive symptoms, negative symptoms and mixed schizophrenia. Because of many types of psychotic disorders and the overlap symptoms of different psychotic disordes, there are currently some difficulties in the diagnosis of schizophrenia. The pathological features of schizophrenia are the reduction of gray matter in the brain, abnormalities in the nerve conduction function of dopaminergic neurons and glutamatergic neurons. At the present, there is no good therapeutic drug for schizophrenia. Some known drugs exert their effects by targeting the dopamine D2 receptor, and all drugs have some side effects.The etiology of schizophrenia is not yet clear, and may be the result of the genetics, environment, and genetics and environment interplay effects during brain development. Environmental risk factors for schizophrenia include infections during pregnancy, changes in the living environment during childhood, brain damage, drug abuse, and so on. The heritability of schizophrenia is currently the highest of almost all known psychotic disorders, about 0.8, which means that genetic factors contribute to 80% of its incidence. Previous study of schizophrenia focused on linkage analysis and association studies. The linkage analysis started with familial schizophrenia and had identified numerous rare and potential risk loci. Association studies started with comaparing sporadic cases with normal controls and had identified some common with sites lower risk effect. During the last decade, technologies such as genome wide association study (GWAS), whole genome sequencing, and exome sequencing have been used to identify a range of risk loci associated with schizophrenia. In addition, it should be noted that some large fragments variants, like copy number variants or structural variants, such as 22q11.2 deletion, contribute largely to schizophrenia, although the frequency of these variants in population is low. Although schizophrenia GWASs have identified numerous genetic risk SNPs, most of these risk SNPs are located in non-coding regions. How does these SNPs located in non-coding regions contribute to schizophrenia remains to be investigated and there is also an urgent need to do that in the post-GWAS era. On the other hand, with the development of various biological sequencing technologies and the advent of the big data era, various omics data, such as methylation sequencing, metabonomics sequencing, and proteomics sequencing have emerged in recent years. Integrating GWAS with these multi-omics data can parse the pathogenesis of schizophrenia from different aspects and provide a basis for the treatment or prevention of schizophrenia.In this study, we first integrated the GWAS data of schizophrenia and the expression quantitative trait loci (eQTL) signal of the prefrontal cortex. The results showed that BTN3A2 was significantly associated with schizophrenia, and the GWAS significant single nucleotide olymorphisms (SNP) rs1979, which locates in the 3'UTR region of BTN3A2, can significantly affect the expression of the BTN3A2 gene. Using different integrative analyses method and eQTL datasets, we could also confirm association this. It was found that the risk G allele of rs1979 can significantly increase the expression of BTN3A2 in brain related eQTL datasets. Further investigation showed that the expression of BTN3A2 was significantly elevated in prefrontal cortex and neurons differentiated from schizophrenia induced pluripotent stem cells (iPSC) compared to normal controls, which is consistent with previous integrative analyses. We used hippocampal CA1 pyramidal neurons as the study system and dual-whole cell recordings as the electrophysiological analyses. It was found that overexpression of BTN3A2 decreased both AMPAR (L-α-amino-3-hydroxy-5-methy-4-isoxazole propionic acid receptor) and NMDAR (N-methy-D-aspartate receptor) mediated synaptic transmission compared with the control neuron. Furthermore, BTN3A2 increased the paired-pulse ratio, the parameter for presynaptic release probability. These results suggest that BTN3A2 regulates excitatory synaptic transmission through decreasing the presynaptic release. Consistently, BTN3A2 expression did not change the ratio of AMPAR and NMADR-mediated excitatory postsynaptic currents (EPSCs) related to neighboring wild-type neurons. In contrast, BTN3A2 had no effect on GABA receptor-mediated inhibitory postsynaptic transmission, suggesting that the BTN3A2-mediated regulation is specific to excitatory synapses. Cell surface binding experiments confirmed that BTN3A2 can interact with presynaptic neurexins, but can not interact with postsynaptic neuroligins. These findings suggest that BTN3A2 decreased the presynaptic release probability, most likely through its interaction with the presynaptic adhesion molecule neurexins. These results from integrative analyses and functional characterization suggest that BTN3A2 may confer risk of schizophrenia through regulating the excitatory/inhibitory synaptic balance.To facilitate researchers to obtain genetic data of schizophrenia genetic research, we have established the genetic database SZDB (www.szdb.org) for schizophrenia. We collected data from linkage analysis, association studies, gene expression, epigenetics, protein-protein interaction networks, etc., and conducted in-depth analysis of these data. Then we integrated the results of the analysis into the database. By scoring data related to schizophrenia from various aspects, we prioritized schizophrenia genes. We further showed that genes implicated in schizophrenia are highly co-expressed in human brain, and proteins encoded by the prioritized schizophrenia risk genes are significantly interacted. The user-friendly SZDB provides high-confidence candidate variants and genes for further functional characterization. More important, SZDB provides convenient online tools for data search and browse, data integration, and customized data analysis.Taken together, through integrative analyses of schizophrenia GWAS data and multi-omics data and functional characterization, we identified BTN3A2 as a potential risk gene for schizophrenia. More over, we have integrated multi-omics data and built a genetic database of schizophrenia to facilitate schizophrenia research.