Alzheimer's disease clinical variants show distinct neuroinflammatory profiles with neuropathology
Baayla D. C. Boon,Irene Frigerio,Danae de Gooijer,Tjado H. J. Morrema,John Bol,Yvon Galis ‐ de Graaf,Martijn Heymans,Netherlands Brain Bank,Melissa E. Murray,Normal Aging Brain Collection Amsterdam,Sven J. van der Lee,Henne Holstege,Wilma D. J. van de Berg,Laura E. Jonkman,Annemieke J. M. Rozemuller,Femke H. Bouwman,Jeroen J. M. Hoozemans
DOI: https://doi.org/10.1111/nan.13009
2024-10-15
Neuropathology and Applied Neurobiology
Abstract:Alzheimer's disease (AD) clinical variants including typical amnestic AD, behavioural/dysexecutive AD, and posterior cortical atrophy AD show distinct neuroanatomical distribution for activated myeloid cells, complement, and reactive astrocytes, in addition to that for phosphorylated tau. The correlation between neuroinflammatory markers and AD neuropathology was brain region– and clinical variant–specific. However, when a certain neuroinflammatory marker correlated with phosphorylated tau in those regions suspected to play an important role in the corresponding clinical variant, that marker also correlated with amyloid‐β. Aims Although the neuroanatomical distribution of tau and amyloid‐β is well studied in Alzheimer's disease (AD) (non)‐amnestic clinical variants, that of neuroinflammation remains unexplored. We investigate the neuroanatomical distribution of activated myeloid cells, astrocytes, and complement alongside amyloid‐β and phosphorylated tau in a clinically well‐defined prospectively collected AD cohort. Methods Clinical variants were diagnosed antemortem, and brain tissue was collected post‐mortem. Typical AD (n = 10), behavioural/dysexecutive AD (n = 6), posterior cortical atrophy (PCA) AD (n = 3), and controls (n = 10) were neuropathologically assessed for AD neuropathology, concurrent pathology including Lewy body disease, limbic‐predominant age‐related TDP‐43 encephalopathy neuropathologic change (LATE‐NC), and vascular pathology. For quantitative assessment, we analysed the corticolimbic distribution of phosphorylated tau, amyloid‐β, CD68, MHC‐II, C4b, and glial fibrillary acidic protein (GFAP) using digital pathology. Results Phosphorylated tau was distinctly distributed in each variant. In all variants, amyloid‐β was neocortical‐dominant, with a notable increase in the middle frontal cortex of behavioural/dysexecutive AD. Typical AD and PCA AD had no concurrent Lewy body disease, whereas three out of six cases with behavioural/dysexecutive AD did. LATE‐NC stage >0 was observed in three AD cases, two typical AD (stage 1/3), and one behavioural/dysexecutive AD (stage 2/3). Vascular pathology was present in each variant. In typical AD, CD68 and MHC‐II were hippocampal‐dominant. In behavioural/dysexecutive AD, C4b was elevated in the middle frontal and inferior parietal cortex. In PCA AD, MHC‐II was increased in the fusiform gyrus, and GFAP in parietal cortices. Correlations between AD neuropathology and neuroinflammation were distinct within variants. Conclusions Our data suggests that different involvement of neuroinflammation may add to clinical heterogeneity in AD, which has implications for neuroinflammation‐based biomarkers and future therapeutics.
pathology,neurosciences,clinical neurology