Multimodal Imaging: Genetic and Environmental Effects in Neuropsychiatry

为了更好地了解精神分裂症的遗传风险如何在认知，行为和神经系统水平上运作，我们最近在该项目下的工作重点是了解多基因疾病风险和基因与环境相互作用的生物学相关性。在这些领域正在进行的研究允许更好地了解精神分裂症的遗传，性状相关的异常，负责这种异常的潜在分子生物学，并解决一些疾病的异质性，使生物学研究如此具有挑战性的策略。 精神分裂症具有高度遗传性，但与其他精神疾病一样，其遗传结构相当复杂。该领域已经确定了大量的遗传标记，在大型病例对照全基因组关联研究中与疾病统计学相关，但其对精神分裂症风险生物学的贡献仍不清楚。需要大量的额外工作来将这些发现转化为对导致精神分裂症系统水平功能障碍的分子途径的更好理解。最早达到全基因组关联水平显著性的基因之一ZNF 804 a就是一个例子，它仍然是精神分裂症风险的重要位点，但该基因的变异如何赋予精神分裂症风险仍然未知。在ZNF 804 a介导的多巴胺受体调节的临床前支持之后，我们在一个大的健康志愿者队列中采用多巴胺受体PET成像，对ZNF 804 a风险位点进行基因分型，并确定了基因型和纹状体中D2多巴胺受体可用性之间的显著和特异性关联，提供了新的证据表明该基因与活人脑中的多巴胺系统有关（Hegarty等，2021年）。 目前的工作采用了单基因效应以外的各种策略，包括使用多基因评分。这些分数总结了整个基因组与给定疾病或特征的遗传关联，是调查疾病和行为的广泛遗传影响的新兴工具。例如，我们已经证明，在精神分裂症中，精神分裂症、ADHD、教育程度和认知能力的多基因评分分别预测了急性疾病发作前认知发展的不同轨迹。在最近的研究中，检查了一组跨基因组的古老DNA序列变体，我们已经发现，累积起来，这些变体与大脑结构和功能的细微差异相关，包括纹状体多巴胺合成减少，这可能与精神分裂症表型的表达减少以及疾病风险和严重程度降低有关（Gregory和Berman，2022; Gregory等人，2021年）。 最近的协作方法学实验已经帮助开发了分析神经成像遗传学问题的新的数据驱动方式，这将有助于发现更复杂的基因-大脑关系，尽管这些数据类型具有多维性质（Ghosal等人，2022年）。 结合精神分裂症潜在风险的遗传因素，环境影响已被确定为流行病学和双胞胎研究中的疾病风险因素，描述基因-环境相互作用可能对更好地了解精神分裂症的病因至关重要。最近的研究强调了两个与精神分裂症相关的环境变量：城市化和早期生活并发症。例如，我们已经证明，一个人是否在城市环境中长大，会显著改变多巴胺相关基因对前额叶皮层如何响应工作记忆需求的影响。我们在另外两个数据集中进一步复制了这一发现，并证实城市教养可能以一种与多巴胺能神经发生机制有意义的交叉方式改变大脑功能。其他研究表明，当有临床显著的产科并发症史时，最近的精神分裂症全基因组关联研究中最强关联的染色体位置的预测强度被大大放大。这些数据表明，传统的单变量方法必须得到支持的基因环境实验，以更充分地阐述精神分裂症的遗传风险架构。

In order to better understand how the genetic risk for schizophrenia operates at the cognitive, behavioral and neural systems level, our recent work under this project has focused on understanding the biological correlates of both polygenic illness risk and gene-by-environment interactions. Ongoing studies in these areas permit better understanding of heritable, trait-related abnormalities in schizophrenia, of the underlying molecular biology responsible for such abnormalities, and of strategies for resolving some of the illness heterogeneity that makes biological research so challenging. Schizophrenia is highly heritable but, like other psychiatric disorders, its genetic architecture is quite complex. The field has identified numerous genetic markers that are statistically associated with illness in large case-control genome-wide association studies but whose contributions to schizophrenia risk biology remain unclear. Considerable additional work is needed to translate these findings into a better understanding of the molecular pathways leading to system-level dysfunction in schizophrenia. This is exemplified by one of the first genes reaching genome-wide association level significance, ZNF804a, which has continued to be an important locus of schizophrenia risk, but it remains unknown how variation in this gene might confer schizophrenia risk. Following preclinical support for ZNF804a-mediated dopamine receptor regulation, we have employed dopamine receptor PET imaging in a large cohort of healthy volunteers genotyped for the ZNF804a risk locus and identified significant and specific associations between genotype and D2 dopamine receptor availability in the striatum, providing novel evidence that this gene has implications for the dopamine system in the living human brain (Hegarty et al., 2021). Current work employs various strategies beyond single-gene effects, including the use of polygenic scores. These scores, which summarize genetic associations across the genome with a given disorder or trait, are an emerging tool for investigations of broad genetic influences on illness and behavior. For instance, we have shown that in schizophrenia, a profile of polygenic scores separately derived for schizophrenia, ADHD, educational attainment, and cognitive ability predicts distinct trajectories of cognitive development prior to the onset of acute illness. In recent studies examining a set of ancient DNA sequence variants across the genome, we have found that cumulatively, these variants are associated with subtle differences in brain structure and function, including less striatal dopamine synthesis, that may be associated with diminished expression of schizophrenia phenotypes in addition to reduced illness risk and severity (Gregory and Berman, 2022; Gregory et al., 2021). Recent collaborative methodological experimentation has helped develop new data-driven ways of analyzing neuroimaging genetics questions, which will aid discovery of more complex gene-brain relationships despite the multidimensional nature of these data types (Ghosal et al., 2022). In conjunction with genetic factors underlying risk for schizophrenia, environmental influences have been identified as illness risk factors in epidemiological and twin studies, and delineating gene-environment interactions is likely crucial to better understanding the causes of schizophrenia. Recent research has highlighted two environmental variables that show association with schizophrenia: urbanicity and early life complications. For example, we have demonstrated that whether an individual was raised in an urban environment significantly modifies effects of dopamine-related genes on how the prefrontal cortex responds to working memory demands. We have further replicated this finding in two additional datasets and posit that urban upbringing may alter brain function in a way that meaningfully intersects with dopaminergic neurogenetic mechanisms. Other work suggests that the predictive strength of most strongly associated chromosomal location from the recent schizophrenia genome-wide association study is greatly amplified when there is a history of clinically significant obstetrical complications. These data suggest that traditional univariate approaches must be buttressed by gene-by-environment experimentation to more fully elaborate genetic risk architecture in schizophrenia.