Skip to main navigation Skip to search Skip to main content

Convergent molecular signatures across eating disorders and obsessive-compulsive disorder in the human brain

Research output: Contribution to journalArticlepeer-review

Abstract

Eating disorder (ED) and obsessive-compulsive disorder (OCD) exhibit clinical and genetic overlap, yet whether they converge at the molecular level in the human brain is unknown. We perform large-scale transcriptomic profiling of the dorsolateral prefrontal cortex (DLPFC) and caudate in postmortem tissue from 86 controls, 57 individuals with ED, and 27 with OCD. ED shows robust, region-specific transcriptional dysregulation (102 differentially expressed genes [DEGs] in DLPFC and 222 in caudate at FDR <1%) that replicates in an independent cohort. OCD shows no single-cohort DEGs, but meta-analysis across three datasets identifies 57 caudate-associated genes. Despite these differences, transcriptome-wide effects strongly correlate between ED and OCD (DLPFC r = 0.67; caudate r = 0.75), indicating shared molecular pathology. Joint ED + OCD analysis identifies 233 DEGs in DLPFC and 815 in caudate, implicating GABAergic signaling, neuroendocrine regulation, mitochondrial metabolism, and CHD8-associated networks. Genetically regulated expression analyses identifies five genes (WDR6, NCKIPSD, P4HTM, DALRD3, and SHISA5) with convergent risk associations across disorders and brain regions, all mapping to a gene-dense region on chromosome 3. These findings define a shared cortico-striatal transcriptional architecture and identify candidate genes for transdiagnostic intervention.

Original languageEnglish
Article number117514
JournalCell Reports
Volume45
Issue number6
DOIs
StatePublished - 23 Jun 2026

Keywords

  • anorexia nervosa
  • brain transcriptomics
  • cortico-striatal systems
  • genetics
  • obsessive-compulsive disorder

Fingerprint

Dive into the research topics of 'Convergent molecular signatures across eating disorders and obsessive-compulsive disorder in the human brain'. Together they form a unique fingerprint.

Cite this