Suicidal behaviors are strongly linked with mood disorders, but the specific neurobiological and functional gene-expression correlates for this linkage remain elusive. We performed neuroimaging-guided RNA-sequencing in two studies to test the hypothesis that imaging-localized gray matter volume (GMV) loss in mood disorders, harbors gene-expression changes associated with disease morbidity and related suicide mortality in an independent postmortem cohort. To do so, first, we conducted study 1 using an anatomical likelihood estimation (ALE) MRI meta-analysis including a total of 47 voxel-based morphometry (VBM) publications (i.e. 26 control versus (vs) major depressive disorder (MDD) studies, and 21 control vs bipolar disorder (BD) studies) in 2387 (living) participants. Study 1 meta-analysis identified a selective anterior insula cortex (AIC) GMV loss in mood disorders. We then used this results to guide study 2 postmortem tissue dissection and RNA-Sequencing of 100 independent donor brain samples with a life-time history of MDD (N = 30), BD (N = 37) and control (N = 33). In study 2, exploratory factor-analysis identified a higher-order factor representing number of Axis-1 diagnoses (e.g. substance use disorders/psychosis/anxiety, etc.), referred to here as morbidity and suicide-completion referred to as mortality. Comparisons of case-vs-control, and factor-analysis defined higher-order-factor contrast variables revealed that the imaging-identified AIC GMV loss sub-region harbors differential gene-expression changes in high morbidity-&-mortality versus low morbidity-&-mortality cohorts in immune, inflammasome, and neurodevelopmental pathways. Weighted gene co-expression network analysis further identified co-activated gene modules for psychiatric morbidity and mortality outcomes. These results provide evidence that AIC anatomical signature for mood disorders are possible correlates for gene-expression abnormalities in mood morbidity and suicide mortality.

自杀行为与情绪障碍有很强的联系，但是与这种联系相关的特定神经生物学和功能基因表达仍然难以捉摸。我们在两项研究中进行了神经影像引导的RNA测序，以检验以下假设：在独立的死后队列中，情绪障碍中影像定位的灰质体积（GMV）丢失，具有与疾病发病率和相关自杀死亡率相关的基因表达变化。为此，首先，我们使用解剖学似然估计（ALE）MRI荟萃分析进行了研究1，该研究包括总共47篇基于体素的形态计量学（VBM）出版物（即26篇对照与（vs）重大抑郁症（MDD）研究） ，以及2387名（生活中）参与者的21项对照vs双相情感障碍（BD）研究。研究1荟萃分析确定了情绪障碍中选择性前岛叶皮质（AIC）GMV丢失。然后，我们使用此结果指导研究2死后组织解剖和100个独立供体脑样本的RNA测序，这些样本的生命史为MDD（N = 30），BD（N = 37）和对照（N = 33）。在研究2中，探索性因素分析确定了代表Axis-1诊断次数（例如，物质使用失调/精神病/焦虑症等）的较高阶因素，在此称为发病率和自杀完成，称为死亡率。病例-病例对照和因素分析定义的高阶因素对比变量的比较显示，影像学鉴定的AIC GMV丢失子区域在高发病率和死亡率与高死亡率和死亡率之间存在差异的基因表达变化。免疫，炎症小体和神经发育途径的发病率和死亡率低。加权基因共表达网络分析进一步确定了用于精神病发病率和死亡率结果的共激活基因模块。这些结果提供了证据，表明情绪障碍的AIC解剖特征可能与情绪发病率和自杀死亡率的基因表达异常有关。