Molecular Neurodegeneration (Mar 2022)

Cerebrospinal fluid tau levels are associated with abnormal neuronal plasticity markers in Alzheimer’s disease

  • Pieter Jelle Visser,
  • Lianne M. Reus,
  • Johan Gobom,
  • Iris Jansen,
  • Ellen Dicks,
  • Sven J. van der Lee,
  • Magda Tsolaki,
  • Frans R. J. Verhey,
  • Julius Popp,
  • Pablo Martinez-Lage,
  • Rik Vandenberghe,
  • Alberto Lleó,
  • José Luís Molinuevo,
  • Sebastiaan Engelborghs,
  • Yvonne Freund-Levi,
  • Lutz Froelich,
  • Kristel Sleegers,
  • Valerija Dobricic,
  • Simon Lovestone,
  • Johannes Streffer,
  • Stephanie J. B. Vos,
  • Isabelle Bos,
  • ADNI,
  • August B. Smit,
  • Kaj Blennow,
  • Philip Scheltens,
  • Charlotte E. Teunissen,
  • Lars Bertram,
  • Henrik Zetterberg,
  • Betty M. Tijms

DOI
https://doi.org/10.1186/s13024-022-00521-3
Journal volume & issue
Vol. 17, no. 1
pp. 1 – 16

Abstract

Read online

Abstract Background Increased total tau (t-tau) in cerebrospinal fluid (CSF) is a key characteristic of Alzheimer’s disease (AD) and is considered to result from neurodegeneration. T-tau levels, however, can be increased in very early disease stages, when neurodegeneration is limited, and can be normal in advanced disease stages. This suggests that t-tau levels may be driven by other mechanisms as well. Because tau pathophysiology is emerging as treatment target for AD, we aimed to clarify molecular processes associated with CSF t-tau levels. Methods We performed a proteomic, genomic, and imaging study in 1380 individuals with AD, in the preclinical, prodromal, and mild dementia stage, and 380 controls from the Alzheimer’s Disease Neuroimaging Initiative and EMIF-AD Multimodality Biomarker Discovery study. Results We found that, relative to controls, AD individuals with increased t-tau had increased CSF concentrations of over 400 proteins enriched for neuronal plasticity processes. In contrast, AD individuals with normal t-tau had decreased levels of these plasticity proteins and showed increased concentrations of proteins indicative of blood–brain barrier and blood-CSF barrier dysfunction, relative to controls. The distinct proteomic profiles were already present in the preclinical AD stage and persisted in prodromal and dementia stages implying that they reflect disease traits rather than disease states. Dysregulated plasticity proteins were associated with SUZ12 and REST signaling, suggesting aberrant gene repression. GWAS analyses contrasting AD individuals with and without increased t-tau highlighted several genes involved in the regulation of gene expression. Targeted analyses of SNP rs9877502 in GMNC, associated with t-tau levels previously, correlated in individuals with AD with CSF concentrations of 591 plasticity associated proteins. The number of APOE-e4 alleles, however, was not associated with the concentration of plasticity related proteins. Conclusions CSF t-tau levels in AD are associated with altered levels of proteins involved in neuronal plasticity and blood–brain and blood-CSF barrier dysfunction. Future trials may need to stratify on CSF t-tau status, as AD individuals with increased t-tau and normal t-tau are likely to respond differently to treatment, given their opposite CSF proteomic profiles.

Keywords