Biomarkers disclose associations between Neuroinflammation and Synaptic Depletion in AD.
Nesrine Rahmouni, Marcel Woo, Cécile Tissot, Stijn Servaes, Joseph Therriault, Jenna Stevenson, Andrea Lessa Benedet, Nicholas J. Ashton, Henrik Zetterberg, Kaj Blennow, Tharick Ali Pascoal, Eduardo R Zimmer, Pedro Rosa‐Neto- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
While in healthy conditions, synaptic function is sustained by the interplay between neurons and glial cells. The presence of Alzheimer’s disease (AD) pathophysiology might impose synaptic alterations via neuroinflammatory responses from microglia and astrocytes. Imaging and fluid biomarkers allow for testing this conceptual framework in living individuals. Here, we aim at uncovering the associations between neuroinflammatory and synaptic markers in aging and AD.
Method
Participants were recruited from the Translational Biomarker for aging and dementia cohort (TRIAD). We analyzed cognitively unimpaired young (CUY, N = 20), cognitively unimpaired older adults (CU; N = 48) individuals, patients with mild cognitive impairment (MCI; N = 24), AD dementia patients (ADD; N = 16) participants. The endpoints were CSF 14‐3‐3 sigma‐delta (ζδ), CSF Growth Associated Protein 43 (GAP43) and neurogranin as markers of synaptic dysfunction. We modeled associations between these synaptic biomarkers with astrocyte and microglial related inflammatory markers as well as tau, amyloid and neurodegeneration biomarkers.
Result
According to ROC analyses contrasting CSF biomarkers, 14‐3‐3 ζδ was able to discriminate amyloid‐ β pathology in cognitively impaired individuals (AUC = 0.88) (Figure 1). The amyloid‐induced 14‐3‐3 ζδ increase was mediated by inflammatory astrogliosis measured by GFAP, whereas the tau induced 14‐3‐3 ζδ pathology was mediated by anti‐inflammatory cytokines (Figure 2). CSF levels of GAP43 and neurogranin were positively associated with TSPO measured with [11C]PBR28(Figure 3.1) and GFAP (Figure 3.2). [11C]PBR28 cluster‐based SUVRs was positively correlated to concentrations of (A) Neurogranin (p = 5.8e‐08) and (B) GAP‐43(p = 6.8e‐10).
Conclusion
The present observations support the framework in which glia‐related neuroinflammatory responses contribute to synaptic alterations in carriers of AD pathophysiology. We will discuss the implications and limitations of this framework.