Co-localization of amyloid beta and tau pathology in Alzheimer's disease synaptosomes

Abstract

The amyloid cascade hypothesis proposes that amyloid beta (Abeta) pathology precedes and induces tau pathology, but the neuropathological connection between these two lesions has not been demonstrated. We examined the regional distribution and co-localization of Abeta and phosphorylated tau (p-tau) in synaptic terminals of Alzheimer's disease brains. To quantitatively examine large populations of individual synaptic terminals, flow cytometry was used to analyze synaptosomes prepared from cryopreserved Alzheimer's disease tissue. An average 68.4% of synaptic terminals in the Alzheimer's disease cohort (n = 11) were positive for Abeta, and 32.3% were positive for p-tau; Abeta and p-tau fluorescence was lowest in cerebellum. In contrast to synaptic p-tau, which was highest in the entorhinal cortex and hippocampus (P = 0.004), synaptic Abeta fluorescence was significantly lower in the entorhinal cortex and hippocampus relative to neocortical regions (P = 0.0003). Synaptic Abeta and p-tau fluorescence was significantly correlated (r = 0.683, P < 0.004), and dual-labeling experiments demonstrated that 24.1% of Abeta-positive terminals were also positive for p-tau, with the highest fraction of dual labeling (39.3%) in the earliest affected region, the entorhinal cortex. Western blotting experiments show a significant correlation between synaptic Abeta levels measured by flow cytometry and oligomeric Abeta species (P < 0.0001). These results showing overlapping Abeta and tau pathology are consistent with a model in which both synaptic loss and dysfunction are linked to a synaptic amyloid cascade within the synaptic compartment.

Figure 1

Flow cytometry analysis of Aβ and p-tau in AD synaptosomes. A–E: representative samples showing Aβ labeling in synaptosomes from (A) AD frontal cortex, (B) aged normal frontal cortex, (C) aged Tg2576 mouse, and (D) wild-type mouse. The forward scatter parameter on the abscissa is proportional to size. E: Size standards used to draw the size analysis gate for analysis of particles in the range of ∼0.5 to 1.5 microns. F: Uranyl acetate-stained electron micrograph image of synaptosomes that were sorted based on the size gate; inset from an ultrathin section illustrates internal structure. G: Size of positive fraction within the AD cohort for the synaptic and damage markers indicated. Value shown is mean percentage positive across brain regions (A7, A9, A39, A40, A28, and hippocampus, n = 77 samples from 11 cases). H: Correlation between synaptic Aβ and synaptic p-tau fluorescence (Pearson’s product-moment correlation coefficient, P < 0.004).

Figure 2

Regional distribution of synaptic (A) Aβ, (B) p-tau, (C) free cholesterol, and (D) ganglioside GM1. For each measure, the Estimate statement was used to test for grouped differences between hippocampus and entorhinal cortex versus neocortical regions (A7, A9, A39, and A40).

Figure 3

Co-localization of Aβ and p-tau. A–B: Flow cytometry quadrant analysis for a representative sample of AD frontal cortex (A), and aged Tg2576 mouse (B). Only Aβ-positive synaptosomes were analyzed; particles positive for Aβ only are in the upper left quadrant and particles positive for both Aβ and p-tau are in the upper right quadrant, (C) regional variation in dual labeled fraction (n = 29 samples from 6 cases), (D) quadrant analysis of size-gated synaptosomes illustrating size of fraction positive only for p-tau (lower right quadrant).

Figure 4

Confocal analysis of Aβ and p-tau co-localization. A–D: Confocal analysis of washed P-2 fraction from an 87 y/o AD case (superior parietal cortex) dual labeled for Aβ (A, green) and p-tau (B, red). C: Overlay image with yellow color indicating co-localization (arrows). E–H: Co-localization SNAP-25 (A, green) and p-tau (B, red) indicating co-localization in synaptosomes (G). D, H: Differential interference contrast images for each field. I: 1.53-μm size standards.

Figure 5

Synaptic Aβ is correlated with oligomers in AD samples. A: Western blot of washed P-2 samples from AD cases previously examined by flow cytometry showing multiple Aβ species and APP identified by the 6E10 antibody; p-tau was identified by the p422s antibody. B: Correlation of synaptic Aβ measured by flow cytometry (Table 1) with a hexameric oligomer species (P < 0.0001 with one outlier removed). C: Correlation of synaptic tau measured by flow cytometry with p-tau (P < 0.0008).

Figure 6

Aβ and p-tau are co-localized in neuronal processes in AD brain. Confocal images of section from AD case showing (A) p-tau (green) identified by the AT100 antibody, (B) Aβ (red) identified by the 10G4 antibody, and (C) nuclei (blue) labeled with 4′,6-diamidino-2-phenylindole. D: Overlap shows co-localization (yellow) in multiple dystrophic neurites. E: Merged image with 6E10 antibody also shows extensive overlap (yellow) with AT100-positive neurites. F: Merged image of C-terminal APP labeling does not show co-localization with p-tau in AT100-positive neurites; Bar = 20 μm.