MAPPING THE CEREBROSPINAL FLUID PROTEOME IN BIPOLAR DISORDER.

Abstract

Background: Bipolar disorder (BD) is a severe psychiatric condition with unclear etiology and no established biomarkers. Here, we aimed to characterize the cerebrospinal fluid (CSF) proteome in euthymic BD individuals to identify potential protein biomarkers.

Methods: We employed nano-flow liquid chromatography coupled to high-resolution mass spectrometry to quantify over 2,000 CSF proteins in 374 individuals from two independent clinical cohorts (n=164+89 and 66+55 cases and controls, respectively). A subset of the cases was followed longitudinally and reexamined after a median of 6.5 years.

Results: Differential abundance analysis revealed 41 proteins with robust case-control association in both cohorts. These included lower levels of synaptic proteins (e.g., APP, CLSTN1, NPTX2, NRXN1), axon guidance and cell adhesion molecules (e.g., NEO1, NCAM1, SEMA7A), higher levels of blood-brain-barrier integrity proteins (e.g., VTN, SERPIN3), and complement components (e.g., C1RL, C3, C5). The findings were consistently driven by the BD type 1 subtype. Comparing BD type 1 with controls increased discoverability, revealing 86 replicated associations despite a loss of statistical power. Moreover, longitudinal analyses of co-expression modules revealed dynamic changes in the CSF proteome composition that correlated with clinical outcomes, including disease severity, future manic episodes, and symptom improvement. Finally, we conducted association analyses of CSF proteins with genetic risk loci for bipolar disorder and schizophrenia.

Conclusions: This study represents the first large-scale untargeted profiling of the CSF proteome in BD, unveiling potential biomarkers and providing in vivo support for altered synaptic and brain connectivity processes, impaired neurovascular integrity, and complement activation in the pathology of BD.