PT  - JOURNAL ARTICLE
AU  - Kevin M Anderson
AU  - Meghan A Collins
AU  - Ru Kong
AU  - Kacey Fang
AU  - Jingwei Li
AU  - Tong He
AU  - Adam M Chekroud
AU  - B.T. Thomas Yeo
AU  - Avram J Holmes
TI  - Convergent molecular, cellular, and neural signatures of major depressive disorder
AID  - 10.1101/2020.02.10.942227
DP  - 2020 Jan 01
TA  - bioRxiv
PG  - 2020.02.10.942227
4099  - http://biorxiv.org/content/early/2020/02/11/2020.02.10.942227.short
4100  - http://biorxiv.org/content/early/2020/02/11/2020.02.10.942227.full
AB  - Major depressive disorder emerges from the complex interactions of biological systems that span across genes and molecules through cells, circuits, networks, and behavior. Establishing how neurobiological processes coalesce to contribute to the onset and maintenance of depression requires a multi-scale approach, encompassing measures of brain structure and function as well as genetic and cell-specific genomic data. Here, we examined anatomical (cortical thickness) and functional (functional variability, global brain connectivity) correlates of depression and negative affect across three population-imaging datasets: UK Biobank, Genome Superstruct Project, and ENIGMA (combined N≥23,723). Integrative analyses incorporated measures of cortical gene expression, post-mortem patient transcriptional data, depression GWAS, and single-cell transcription. Neuroimaging correlates of depression and negative affect were consistent across the three independent datasets. Linking ex-vivo gene downregulation with in-vivo neuroimaging, we found that genomic correlates of depression-linked neuroimaging phenotypes tracked gene downregulation in post-mortem cortical tissue samples of patients with depression. Integrated analysis of single-cell and Allen Human Brain Atlas expression data implicated somatostatin interneurons and astrocytes as consistent cell associates of depression, through both in-vivo imaging and ex-vivo cortical gene dysregulation. Providing converging evidence for these observations, GWAS derived polygenic risk for depression was enriched for genes expressed in interneurons, but not glia. Underscoring the translational potential of multi-scale approaches, the genomic correlates of depression-linked brain function and structure were enriched for known and novel disorder relevant molecular pathways. These findings bridge across levels to connect specific genes, cell classes, and biological pathways to in-vivo imaging correlates of depression.Key FindingsMajor depressive disorder and negative affect are associated with replicable profiles of cortical anatomy and function across independent population-level neuroimaging datasets (combined N≥23,723).Somatostatin interneurons are consistent spatial transcriptional associates of in-vivo depression-linked imaging phenotypes.Integrative single-cell gene expression analysis associate somatostatin interneurons and astrocytes with both in-vivo depression-linked imaging and ex-vivo gene downregulation in independent MDD cortical tissue samples.Transcriptional correlates of in-vivo depression imaging phenotypes selectively capture gene downregulation in post-mortem tissue samples from patients with depression, but not other psychiatric disorders.Indicating that some cell classes are preferentially sensitive to inherited disease liability, genome-wide risk for depression is enriched among interneurons, but not glia.Gene associates of depression-linked anatomy and function identify specific neurotransmitter systems, molecular signaling pathways, and receptors, suggesting possible targets for pharmaceutical intervention.