A comprehensive systematic review of CSF proteins and peptides that define Alzheimer’s disease

Abstract

Background: During the last two decades, over 100 proteomics studies have identified a variety of potential biomarkers in CSF of Alzheimer's (AD) patients. Although several reviews have proposed specific biomarkers, to date, the statistical relevance of these proteins has not been investigated and no peptidomic analyses have been generated on the basis of specific up-or down-regulation. Herein, we perform an analysis of all unbiased explorative proteomics studies of CSF biomarkers in AD to critically evaluate whether proteins and peptides identified in each study are consistent in distribution; direction change; and significance, which would strengthen their potential use in studies of AD pathology and progression. Methods: We generated a database containing all CSF proteins whose levels are known to be significantly altered in human AD from 47 independent, validated, proteomics studies. Using this database, which contains 2022 AD and 2562 control human samples, we examined whether each protein is consistently present on the basis of reliable statistical studies; and if so, whether it is over-or under-represented in AD. Additionally, we performed a direct analysis of available mass spectrometric data of these proteins to generate an AD CSF peptide database with 3221 peptides for further analysis. Results: Of the 162 proteins that were identified in 2 or more studies, we investigated their enrichment or depletion in AD CSF. This allowed us to identify 23 proteins which were increased and 50 proteins which were decreased in AD, some of which have never been revealed as consistent AD biomarkers (i.e. SPRC or MUC18). Regarding the analysis of the tryptic peptide database, we identified 87 peptides corresponding to 13 proteins as the most highly consistently altered peptides in AD. Analysis of tryptic peptide fingerprinting revealed specific peptides encoded by CH3L1, VGF, SCG2, PCSK1N, FBLN3 and APOC2 with the highest probability of detection in AD. Conclusions: Our study reveals a panel of 27 proteins and 21 peptides highly altered in AD with consistent statistical significance; this panel constitutes a potent tool for the classification and diagnosis of AD.

Pathophysiological subtypes of Alzheimer’s disease based on cerebrospinal fluid proteomics

“…Unbiased or large-scale targeted proteomic CSF analyses have potential to further refine which biological processes become disrupted in Alzheimer’s disease. So far, Alzheimer’s disease proteomic studies mostly focussed on finding novel biomarkers by comparing patients with Alzheimer’s disease with control subjects ( Maarouf et al , 2009 ; Meyer et al , 2018 ; for reviews see Pedrero-Prieto et al , 2020 ; Wesenhagen et al , 2020 ), and so it remains unclear whether pathophysiological subtypes within Alzheimer’s disease can be discovering with CSF proteomics. Furthermore, if genetic variance in Alzheimer’s disease risk genes contributes to interindividual variability in underlying disease mechanisms, it can be hypothesized that these should already be detectable in presymptomatic stages of Alzheimer’s disease.…”

Secreted Chaperones in Neurodegeneration

“…Additional rare polymorphisms have been identified in patients suffering from late onset AD, in which secreted clusterin levels are reduced due to folding abnormalities ( Bettens et al, 2015 ). Six independent proteomics studies show that CSF levels of clusterin are significantly increased in AD patients (see meta-analysis in Pedrero-Prieto et al, 2020 ); plasma levels also rise in ALS ( Xu et al, 2018 ). Additionally, while brain clusterin levels increase within the brain during AD progression, the levels of Abeta increase to a greater extent, resulting in a declining molar ratio between clusterin and Abeta specifically within regions of the brain with high Abeta deposition ( Miners et al, 2017 ).…”

“…Immunofluorescence studies have similarly shown that proSAAS co-localizes with aggregated proteins involved in neurodegenerative disease, namely tau tangles in dementia ( Kikuchi et al, 2003 ); Abeta plaques in AD ( Hoshino et al, 2014 ); and Lewy bodies in PD ( Helwig et al, 2013 ). Seven independent proteomics studies have shown that the level of proSAAS in CSF taken from AD and/or FTD patients is reduced as compared to controls, suggesting possible cellular retention within the brain ( Davidsson et al, 2002 ; Abdi et al, 2006 ; Finehout et al, 2007 ; Jahn et al, 2011 ; Choi et al, 2013 ; Holtta et al, 2015 ; Spellman et al, 2015 ; reviewed in Pedrero-Prieto et al, 2020 ). Two very recent CSF studies support this reduction ( Rotunno et al, 2020 ; Van Steenoven et al, 2020 ).…”

“…With regard to other neurodegenerative diseases, elevated serum levels of HspB1 have been reported during attacks in multiple sclerosis ( Ce et al, 2011 ). However, in a large proteomics meta-analysis of Alzheimer’s CSF biomarkers, no sHSPs were identified as differentially expressed in any of the over 40 collated studies ( Pedrero-Prieto et al, 2020 ) indicating that AD progression does not involve alterations in the secretion of sHsps into the CSF. The extracellular (blood) presence of HspB5 has been demonstrated by inference in the form of autoantibodies found in the sera of AD and PD patients ( Papuc et al, 2016 ).…”

“…In this review, we will summarize various secreted chaperones and describe the evidence for their involvement in neurodegenerative processes. We have necessarily omitted many secreted proteins which may in future turn out to contribute to proteostasis in neurodegenerative disease; examples include SPARC/osteonectin, which exhibits extracellular chaperone activity ( Chlenski et al, 2011 ) and has repeatedly been linked to neurodegeneration (reviewed in Chen S. et al, 2020 ; Pedrero-Prieto et al, 2020 ). However, its general chaperone activity against neurodegeneration-related aggregates has not yet been documented.…”