A transcriptomic signature of late‐life depression. (20th December 2022)

• Record Type: Journal Article
• Authors: Matan‐Lithwick, Stuart A; Bennett, David A; Wang, Yanling; Tripathy, Shreejoy; Felsky, Daniel
• Is Part Of: Alzheimer's & dementia. Volume 18(2022)Supplement 6
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-20
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.069357
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Late‐onset Alzheimer's disease (AD) is commonly accompanied by symptoms of depression. While genes associated with pathological AD diagnosis and mid‐life depression have been identified, the transcriptomic signature of late‐life depression has not been described. We hypothesized that neocortical gene expression would be associated with symptoms of late‐life depression proximal to death, independent of pathologic AD diagnosis and medication status. Method: Bulk tissue RNA sequencing data from frontal cortex of 1, 001 elderly brain donors were analyzed (mean age at death: 89.7yrs [range: 67‐108]). All donors had genotype, demographic, and clinical assessments within one year prior to autopsy, as well as detailed postmortem neuropathological characterization. Differential expression analysis was performed with robust linear modeling including technical and demographic covariates, cell type proportions, educational attainment, APOE e4 genotype, and medication status for AD‐ and depression‐prescribed drugs. We then included main effects of depressive symptom burden proximal to death (CESD‐sum) and pathologic diagnosis of AD (NIA‐Reagan criteria). Result: After false discovery rate correction (FDR q <0.05), only one gene, Prader Willi/Angelman region RNA1 ( PWAR1 ), was associated with depressive symptoms (t=5.2, q =0.005); greater PWAR1 abundance was linked to higher symptom burden. At q <0.1, 14 genes showed association. Strikingly, a majority of these genes (e.g. CTDSPL2, ACR2B‐AS1, ADGRE2, MRM1, IRF8, COL19A1 ) have been previously implicated in unipolar depression or bipolar disorder. Gene Set Enrichment Analysis revealed upregulation of processes related to energy metabolism (top: "oxidative phosphorylation", p =2.1x10‐8) and downregulation of DNA modifying processes (top: "DNA conformation change", p =2.0x10‐7), among others. Results were not changed when controlling for smoking and alcohol consumption, and no significant associations were observed for depression medication status (top q >0.23). Conclusion: Here we describe a cortical transcriptomic signature of late‐life depressive symptoms. Among our top signals are genes with known links to mid‐life depression, as well as biological processes relevant to energy metabolism, synaptic plasticity, DNA modification, antigen presentation, and response to pH. Our work provides a depression‐specific expression signature for elderly frontal cortex and lays the foundation for investigation of specific mechanisms toward precision therapeutics.