Clustering of plaques contributes to plaque growth in a mouse model of Alzheimer's disease

Abstract

Amyloid-β (Aβ) plaque deposition plays a central role in the pathogenesis of Alzheimer's disease (AD). Post-mortem analysis of plaque development in mouse models of AD revealed that plaques are initially small, but then increase in size and become more numerous with age. There is evidence that plaques can grow uniformly over time; however, a complementary hypothesis of plaque development is that small plaques cluster and grow together thereby forming larger plaques. To investigate the latter hypothesis, we studied plaque formation in APPPS1 mice using in vivo two-photon microscopy and immunohistochemical analysis. We used sequential pre- and post-mortem staining techniques to label plaques at different stages of development and to detect newly emerged plaques. Post-mortem analysis revealed that a subset (22 %) of newly formed plaques appeared very close (<40 μm) to pre-existing plaques and that many close plaques (25 %) that were initially separate merged over time to form one single large plaque. Our results suggest that small plaques can cluster together, thus forming larger plaques as a complementary mechanism to simple uniform plaque growth from a single initial plaque. This study deepens our understanding of Aβ deposition and demonstrates that there are multiple mechanisms at play in plaque development.

Fig. 1

Pre-existing and new plaques revealed with dual staining technique. a–c In vivo two-photon imaging of newly appearing amyloid plaques. Methoxy-XO4 labeled pre-existing plaques in green (a), anti-Aβ antibody labeled plaques after 4 months post-injection as shown in red (b) and the merged image of the two fluorescence channels with newly developed plaques depicted with white arrowheads (c). d–l Occurrence of new plaques demonstrated by sequential pre- and post-mortem staining. Plaques that existed at day 0 (3-month-old animals) are visualized with Methoxy-XO4 (d, g, j). Post-mortem plaque analysis with immunohistochemical staining against Aβ (e) reveals almost complete co-localization at 1-day post-injection (f). After 1 month (g–i) and 4 months post-injection (j–l), an increasing amount of newly formed plaques (white arrowheads), including ‘new in vicinity’ plaques (<40 μm away from a pre-existing plaque, yellow arrowheads) were detectable (i, l). MX Methoxy-XO4. Scale bar a–c 50 μm, d–l 100 μm

Fig. 2

‘New in the vicinity’ plaques occur more frequently than expected by chance. (a–c) High resolution images of immunofluorescent staining 1 day, 1 and 4 months post-Methoxy-XO4 injection reveal newly developed plaques in the vicinity (<40 μm) of a pre-existing plaque (c, f, i, yellow arrowheads). Methoxy-XO4 shown in green, Aβ 3552 antibody staining shown in red. Quantification of new plaques showed an increase in both single new plaques further away (>40 μm) from pre-existing plaques and new plaques in the vicinity (<40 μm) of a pre-existing plaque over time (j). Mean ± SEM, n = 5–6 per group; Kruskal–Wallis with Dunn’s post hoc test **p < 0.01. k Computer-simulated random locations of new plaques revealed a significantly lower fraction of ‘new plaques in vicinity’ compared to the original data set. Each data point represents one animal. Mean ± SEM, n = 20; Wilcoxon matched-pairs signed rank test **p < 0.01. Scale bar 25 μm

Fig. 3

Flower plaques—new plaques cluster around pre-existing plaques (a–c). Maximum intensity projections of an in vivo image observed with a two-photon microscope 4 months post-Methoxy-XO4 injection (depicted in green) combined with acute topical application of fluorescently labeled Aβ antibody 6E10 staining (depicted in red) (b). The yellow arrowheads (c, f and i) indicate new plaque ‘petals’ clustering around a common pre-existing plaque creating a ‘flower plaque’. A flower plaque observed with pre-mortem Methoxy-XO4 and post-mortem Aβ antibody staining 4 months post-injection (d–f). A flower plaque visualized with post-mortem Thiazin Red 4 months post-injection (g–i) illustrating that flower plaques can also be composed of multiple dense core plaques. Scale bar 25 μm

Fig. 4

Large plaques can contain multiple dense cores at all time-points investigated. Plaques with more than one initial dense plaque core as shown by in vivo Methoxy-XO4 injection, the so-called ‘multicore plaques’, were found after 1 day (a–c), 1 month (d–f) and 4 months (g–i) post-injection. Pre-mortem Methoxy-XO4 labeling (a, d and g) compared to post-mortem Aβ antibody staining (b, e and h) in the merged images (c, f and i) revealed multiple dense cores within one larger plaque. White arrows indicate dense cores and yellow arrowheads new plaques in the vicinity (<40 μm) of pre-existing plaques. j Quantification of multicored plaques reveals that they become more numerous with increasing post-Methoxy-XO4 injection times. Mean ± SEM, n = 5–6 Kruskal–Wallis test with Dunn’s post hoc test *p < 0.05. k Positive correlation between the number of cores and plaque size 4 months after Methoxy-XO4 injection (linear regression R ² = 0.07474; p < 0.0001, n = 687 plaques from 5 mice). Scale bar 25 μm

Fig. 5

Fate of multicore plaques. Post-mortem Thiazin Red staining of dense core plaques (b, f, j) compared to Methoxy-XO4 pre-mortem staining 4 months prior to killing (a, e, i). Depicted are three different possibilities: a–d an example of non-merging multicores; e–h dense cores merging to a single dense core and i–l an example of a new dense core appearing within the diffuse halo around a pre-existing single dense core. White arrows point to multicores, the white arrowhead in j points to a newly developed dense core plaque. Scale bar 25 μm

Fig. 6

Cluster of plaques merging over time observed with in vivo imaging. Long-term, two-photon in vivo imaging shows an initial cluster of multiple, separate, dense core plaques (white arrows, a) that can be followed over subsequent weeks (b–d) to finally merge together after several months (white arrowheads, e–g). Methoxy-XO4 was injected 24 h before each imaging session. Scale bar 20 μm

Fig. 7

Schematic illustrating the clustering of amyloid plaque development. The initial plaque formation (a) can be followed by the formation of a new plaque close by b. These two plaques can merge over time (c) to form one large plaque with a single plaque core (d) or rather remain separate (e) as part of a larger plaque with multiple cores (g). Alternatively, the initial plaque (a) is followed by the clustering of new plaques around the initial plaque (f, flower plaque) and ultimately this cluster can merge together giving rise to one large plaque with multiple cores (g). Solid arrows refer to events captured by our data. The dashed arrow represents hypothetical plaque development