Distinct amyloid-β and tau-associated microglia profiles in Alzheimer's disease

Abstract

Alzheimer's disease (AD) is the most prevalent form of dementia and is characterized by abnormal extracellular aggregates of amyloid-β and intraneuronal hyperphosphorylated tau tangles and neuropil threads. Microglia, the tissue-resident macrophages of the central nervous system (CNS), are important for CNS homeostasis and implicated in AD pathology. In amyloid mouse models, a phagocytic/activated microglia phenotype has been identified. How increasing levels of amyloid-β and tau pathology affect human microglia transcriptional profiles is unknown. Here, we performed snRNAseq on 482,472 nuclei from non-demented control brains and AD brains containing only amyloid-β plaques or both amyloid-β plaques and tau pathology. Within the microglia population, distinct expression profiles were identified of which two were AD pathology-associated. The phagocytic/activated AD1-microglia population abundance strongly correlated with tissue amyloid-β load and localized to amyloid-β plaques. The AD2-microglia abundance strongly correlated with tissue phospho-tau load and these microglia were more abundant in samples with overt tau pathology. This full characterization of human disease-associated microglia phenotypes provides new insights in the pathophysiological role of microglia in AD and offers new targets for microglia-state-specific therapeutic strategies.

Keywords: Alzheimer’s disease; Amyloid-β; Microglia; Single-nucleus RNA sequencing; Tau.

Fig. 1

Enrichment yields high numbers of microglia and astrocytes for snRNAseq. a Pathological hallmarks of donor groups. b Enrichment strategy for NEUN^neg and OLIG2^neg nuclei. (Brain Image courtesy of the Neurobiobank of the Institute Born-Bunge, Antwerp (Wilrijk), Belgium (NB190113)). c Donor information. Age, RIN and PMD are presented as mean ± SD. d UMAP depicting 482,472 nuclei derived from 36 human cortical brain samples. Colors indicate cell type clusters. e Heatmap depicting expression of selected cell type marker genes. f Dot plot depicting logFC per gene from the comparison CAM versus microglia nuclei. Size depicts significance level. g Heatmap depicting Chi-squared associations between subcluster distribution within each cell type and donor group per brain region. **: p < 0.01. OC Occipital Cortex; OTC Occipitotemporal Cortex; RIN RNA integrity number; PMD Postmortem delay; CTR non-demented controls; CTR+  non-demented controls with mild amyloid-β pathology; AD clinical and neuropathological Alzheimer’s disease

Fig. 2

Two groups of microglia subclusters are associated with AD. a UMAP of 148,606 microglia nuclei in 13 subclusters. b Heatmap depicting average expression of three most enriched genes per subcluster. c Violin plots depicting expression of selected genes per subcluster. *: significantly enriched genes for each subcluster compared to all other subclusters (logFC > 0.15, adjusted p-value < 0.05). d Bar plots depicting the percentage of microglia in each subcluster group by category. Representative marker genes are listed on the bottom. OC Occipital Cortex; OTC Occipitotemporal Cortex; CTR non-demented controls; CTR+  non-demented controls with mild amyloid-β pathology; AD clinical and neuropathological Alzheimer’s disease

Fig. 3

AD1-microglia subclusters gradually transition towards a phagocytic/activated profile. a UMAPs depicting trajectory analysis of homeostatic and AD1-subclusters. Color-scale indicates pseudotime, subclusters and expression of CX3CR1 (homeostasis) and MYO1E (AD1). b Density heatmaps depicting the distribution of nuclei over the UMAP for each sample group. c Violin plots depicting expression of selected genes per subcluster. *: genes significantly differentially expressed (Moran’s I test, q-value < 0.05). d Heatmap depicting all genes significantly differentially expressed over the trajectory *: Moran’s I test, q-value < 0.05). e Heatmaps depicting top 40 (non-ribosomal) DAM genes from [18] and top 37 (non-ribosomal) ARM genes from [30] over pseudotime. *: Moran’s I test, q-value < 0.05. f IBA1, P2RY12, and ITGAX co-expression in tissues from a CTR and an AD donor. Tissue core diameter = 1 mm. g Left: Microglia expressing P2RY12 and IBA1, but not ITGAX. Right: Microglia expressing IBA1 and ITGAX, but not P2RY12. Microglia are from the tissue section and exist next to each other. Scale bar = 20 μm. h Left: Gene expression of P2RY12 (green) and ITGAX (pink) along the trajectory. Right: Correlation between % P2RY12^pos area and % ITGAX^pos area per sample. OC Occipital Cortex; OTC Occipitotemporal Cortex; CTR non-demented controls; CTR+  non-demented controls with mild amyloid-β pathology; AD clinical and neuropathological Alzheimer’s disease; PD Parkinson’s disease; LB Lewy body dementia

Fig. 4

Microglia segregate into distinct amyloid-β and tau-associated profiles. a Amyloid-β and phospho-tau immunohistochemistry of an AD donor. b Heatmap depicting Pearson correlations of amyloid-β/tau load versus the percentage of microglia located in each subcluster. *: p ≤ 0,05; **: p ≤ 0.01; ***: p ≤ 0.001. c Four-way plots depicting differential gene expression of the indicated AD1 clusters (on x-axis) versus homeostasis (subclusters 0, 1, 5) and logFC of AD2 (subclusters 2, 3, 6) versus homeostasis (subclusters 0, 1, 5) on the y-axis. d Four-way plots depicting differential gene expression of AD1 (subclusters 7, 9, 10) on the x-axis versus homeostasis (subclusters 0, 1, 5) and logFC of the indicated AD2 clusters versus homeostasis (subclusters 0, 1, 5) on the y-axis. e GRID2 expression (brown) in AD samples with only amyloid-β or both amyloid-β and tau pathology. Cresyl violet was used to detect nuclei. f IBA1 (green), GRID2 (orange) and phospho-tau (magenta) colocalization in human AD brain tissue. OC Occipital Cortex; OTC Occipitotemporal Cortex