Rapid appearance and local toxicity of amyloid-beta plaques in a mouse model of Alzheimer's disease

Abstract

Senile plaques accumulate over the course of decades in the brains of patients with Alzheimer's disease. A fundamental tenet of the amyloid hypothesis of Alzheimer's disease is that the deposition of amyloid-beta precedes and induces the neuronal abnormalities that underlie dementia. This idea has been challenged, however, by the suggestion that alterations in axonal trafficking and morphological abnormalities precede and lead to senile plaques. The role of microglia in accelerating or retarding these processes has been uncertain. To investigate the temporal relation between plaque formation and the changes in local neuritic architecture, we used longitudinal in vivo multiphoton microscopy to sequentially image young APPswe/PS1d9xYFP (B6C3-YFP) transgenic mice. Here we show that plaques form extraordinarily quickly, over 24 h. Within 1-2 days of a new plaque's appearance, microglia are activated and recruited to the site. Progressive neuritic changes ensue, leading to increasingly dysmorphic neurites over the next days to weeks. These data establish plaques as a critical mediator of neuritic pathology.

Figure 1. Appearance of a novel plaque is a rapid process

a–c, Low-magnification images provide an overview of the areas of potential plaque formation. The angiogram (red, Texas red), amyloid deposition (blue, methoxy-XO4) and neurons (green, YFP) are easily identified on the initial day of surgery (a) as well as one week (b) and two weeks later (c), allowing reimaging of the same sites over different imaging sessions. A new parenchymal amyloid deposition was identified one week (b) and two weeks (c) after the first imaging at this site. The new plaque appearing is indicated by an arrow (b and c). d–i, Sequential daily imaging of a potential plaque-formation area revealed that plaques can form rapidly in a short time interval from one day to another (e). The initial plaque was joined by a novel plaque marked with arrows over six consecutive days (e–i). Note the formation of a dystrophy indicated by an arrowhead (i). A line diagram was used in this figure to visualize plaque size over time. The mean area of new plaques was 88.7 ± 69.3 µm² (mean ± s.d.) (n = 18). This is comparable to the mean plaque area of 93.7 ± 74.8 µm² (mean ± s.d.) (n = 153) (Student’s t-test not significant, P = 0.77) measured post mortem in a subset of these animals by using the Bioquant image analysis system. Each individual plaque that appeared either after one day or seven days is represented by a different colour (j). Scale bars, 30 µm (a–c) and 20 µm (d–i).

Figure 2. Microglia recruitment follows plaque formation

a–f, Green fluorescent protein-positive microglia cells were imaged in PDAPP^+/−CX3CR1^+/− mice before and after plaque formation in daily imaging sessions. b, e, Two plaques appeared within 24 h (arrows) as well as microglia around these newly formed plaques. Individual microglia remained stable but new microglia surrounding plaques were also evident (arrowhead). Scale bars, 20 µm.

Figure 3. Plaque formation has no immediate effect on neuritic curvature

a–d, Neurites were observed before and after plaque formation. A new plaque labelled with methoxy-XO4 is shown in c and indicated by an arrow. Images from the green channel were further analysed and neurite curvature was measured. Arrowheads in c and d denote the increased neurite curvature after a plaque formed. Three-dimensional reconstruction as an example of a developed plaque (blue) surrounded by deforming neurites (green) (b, d). Scale bar, 50 µm. e, Neurite curvature was measured one week before and after plaque development. Neurites (n = 29) measured in B6C3-YFP mice (n = 6) less than 50 µm from a plaque, neurites more than 50 µm from a plaque (n = 27) and neurites (n = 40) in control mice (n = 5) are depicted as black, grey and white bars, respectively. There was a statistically significant increase in neurite curvature after plaques form compared with the initial imaging session one week earlier. This increase persists one week after formation (two weeks from initial time point) (asterisk, ANOVA P < 0.0001). f, Daily assessment of neurite curvature changes (n = 12) from five new plaques revealed no significant differences at the day of plaque occurrence (data from all new plaques normalized to show plaque appearance at day 2). However, there was a tendency towards increased curvature, which became statistically significant at the third imaging day, one day after plaque appearance (asterisk, ANOVA P = 0.002). At imaging day 6, the neurite curvature ratio reached the highest level (asterisk, ANOVA P = 0.0002) and was comparable to that seen one week after plaque formation (day 7). Data are shown as mean ± standard deviation.