A Proteomic Discovery Study of Cerebrospinal Fluid After Aneurysmal Subarachnoid Hemorrhage

Abstract

Background: Proteomic analysis of cerebrospinal fluid (CSF) has the potential to provide insight into the pathophysiology of aneurysmal subarachnoid hemorrhage (aSAH) and target trials to improve outcome. The aim of this study was to perform a definitive proteomic analysis of CSF following aSAH to identify proteins associated with neurological injury.

Methods: This was a retrospective proteomic analysis of CSF collected at neurosurgical centers in the United Kingdom between 2013 and 2023 either from external ventricular drain or lumbar puncture on day 7 after aSAH. Adults with confirmed aSAH were included. Exclusions were pregnancy, severe comorbidities, inability to follow-up, and those not expected to survive 24 hours. Proteomic analysis was performed using mass spectrometry to identify CSF proteins differentially expressed between patients with good (modified Rankin Scale score of 0-2) and poor (modified Rankin Scale score of 3-6) outcomes at 6 months following aSAH. Controlling for CSF albumin (a marker of blood-brain interface permeability and volume of hemorrhage), differentially expressed proteins were identified. Differential pathway activity was explored using protein interaction, gene set enrichment and TopMD analyses.

Results: A total of 152 patients were included (101 good and 51 poor outcome), and 4952 unique proteins were identified across all samples. The CSF proteomic profile differed between good and poor outcome individuals as evidenced by clustering of individuals by outcome using topological data analysis. Controlling for CSF albumin 16 intracellular and secreted proteins were differentially expressed between good and poor outcome patients. Two cellular pathways were identified to have differential activity by all 3 pathway analysis approaches: the phosphoinositide 3-kinase-Akt signaling pathway and glycolysis/gluconeogenesis.

Conclusions: In this study, 16 proteins were differentially expressed between good and poor outcome aSAH patients. The proteomic evidence, both on an individual protein and pathway level highlights that inflammation and oxidative injury are associated with the pathophysiology of neurological injury following aSAH. These results support the exploration of treatments targeting these pathways to improve outcome after aSAH.

Keywords: aneurysmal subarachnoid hemorrhage; cerebrospinal fluid; humans; outcome assessment; proteomics.

Figure 1.

Topological data analysis using all protein included in primary analysis generated 2 clusters which differ in probability of poor outcome. Nodes are colored according to probability of poor outcome.

Figure 2.

Volcano plot for primary analysis comparing proteomic differences between individuals with good and poor outcome after aneurysmal subarachnoid hemorrhage. Red and green dots represent significantly upregulated and downregulated proteins (respectively) in the poor outcome cohort. Dotted lines represent thresholds for significance false discovery rate (FDR) corrected P<5% and absolute log₂(fold change) >1.

Figure 3.

Glycolysis/gluconeogenesis and phosphoinositide 3-kinase (PI3K)-Akt signaling Kyoto Encyclopedia of Genes and Genomes pathways demonstrating components included in primary analysis (10 proteins from glycolysis/gluconeogenesis and 17 proteins from PI3K-Akt signaling pathways). Red signifies upregulated and green signifies downregulated in the poor outcome cohort. Relevant enzyme commission components included in the figure are as follows: 4.1.2.13=ALDOA/C; 1.2.1.12=GAPDH; 5.3.1.1=TPI1; 5.4.2.4/11=BPGM; 4.2.1.11=ENO1 & ENO2; 2.7.1.40=PKLR & PKM; 1.1.1.27=LDHB; HSP90=HSP90AA1, HSP90AB1 & HSP90B1; 14-3-3=YWHAQ/B/E/G/H/Z; and ECM=COL1A2, COL4A2, COL6A1/3, FN1, LAMB1/2 & LAMC1.