Deep Proteomics Profiling of Cerebrospinal Fluid in Idiopathic Normal Pressure Hydrocephalus and Mild Cognitive Impairment
Aida Kamalian1, Megha Patel1, Sara Ho1, Abhay Moghekar2
1Johns Hopkins University, 2Johns Hopkins Bayview Medical Center
Objective:
To discover novel cerebrospinal fluid (CSF) biomarkers relevant to Idiopathic Normal Pressure Hydrocephalus (iNPH)
Background:
iNPH is manifested with a triad of cognitive impairment, urinary incontinence, and gait abnormalities. The CSF proteomics of iNPH and its etiology remain unknown.
Design/Methods:
CSF samples were collected from 50 HC (68.6 ± 6.2 years), 29 iNPH patients (71.5 ± 5.1 years) that were found to benefit significantly from shunt surgery in subsequent follow-ups, and 38 Mild Cognitive Impairment (MCI) participants (75.0 ± 9.2 years) with positive CSF Alzheimer’s disease (AD) biomarkers. We used Olink Explore 3072 platform to map approximately 3000 proteins in the CSF samples and then performed Welch’s t-test, analysis of variance (ANOVA), and Kruskal-Wallis (KW) analyses for statistical analysis using R v4.3.1. Then, we performed over-representation analysis (ORA) and gene set enrichment analysis (GSEA) using R.
Results:
583 proteins were differentially expressed in iNPH compared to HC (416 down- and 167 upregulated). 1675 proteins were found to be differentially expressed based on ANOVA and KW testing between iNPH, MCI, and HC. The pathway analysis of the data demonstrated significant downregulation of pathways related to “Neuronal cell-cell adhesion” (NRXN3, CNTNAP2, DSCAM), “synaptic signaling” (GRIK2, RTN4R, NPTX1, NPTX2, NPTXR, TMEM25, SNAP25, SNAPIN, NRGN), and a known congenital hydrocephalus marker, L1CAM. In contrast, “lymphocytic chemotaxis” and “chemokine-related” pathways were found to be significantly upregulated in iNPH (CCL2, CXCL10, CXCL9, GBP2). Additionally, known AD protein markers such as MAPT, TREM2, GFAP, SCRN1 were not differentially expressed in iNPH. NEFL was significantly upregulated in both MCI and iNPH. Proteins constitutively secreted by choroid plexus (ENPP2, TTR) were not differentially expressed in iNPH.
Conclusions:
While synaptic and neuronal adhesion pathways are downregulated in iNPH, lymphocytic chemotaxis and neuroinflammatory markers are upregulated and these changes are not accounted for by the increased CSF volume of distribution.