Characterizing the appearance and growth of amyloid plaques in APP/PS1 mice

Abstract

Amyloid plaques are primarily composed of extracellular aggregates of amyloid-beta (Abeta) peptide and are a pathological signature of Alzheimer's disease. However, the factors that influence the dynamics of amyloid plaque formation and growth in vivo are largely unknown. Using serial intravital multiphoton microscopy through a thinned-skull cranial window in APP/PS1 transgenic mice, we found that amyloid plaques appear and grow over a period of weeks before reaching a mature size. Growth was more prominent early after initial plaque formation: plaques grew faster in 6-month-old compared with 10-month-old mice. Plaque growth rate was also size-related, as smaller plaques exhibited more rapid growth relative to larger plaques. Alterations in interstitial Abeta concentrations were associated with changes in plaque growth. Parallel studies using multiphoton microscopy and in vivo microdialysis revealed that pharmacological reduction of soluble extracellular Abeta by as little as 20-25% was associated with a dramatic decrease in plaque formation and growth. Furthermore, this small reduction in Abeta synthesis was sufficient to reduce amyloid plaque load in 6-month-old but not 10-month-old mice, suggesting that treatment early in disease pathogenesis may be more effective than later treatment. In contrast to thinned-skull windows, no significant plaque growth was observed under open-skull windows, which demonstrated extensive microglial and astrocytic activation. Together, these findings indicate that individual amyloid plaque growth in vivo occurs over a period of weeks and may be influenced by interstitial Abeta concentration as well as reactive gliosis.

Figure 1.

Serial in vivo multiphoton microscopy demonstrates growth of individual amyloid plaques. A, B, Epifluorescence micrographs of 4 plaques labeled with methoxy-X04 and imaged over a 7 d interval in the brain of a 6-month-old APP/PS1 mouse. C, D, Multiphoton micrographs of those same 4 plaques illustrate growth of individual plaques over time. Scale bars, 20 μm.

Figure 2.

Amyloid plaques exhibit age- and size-related growth under thinned-skull window preparations. Using serial intravital multiphoton microscopy, individual plaques were labeled with methoxy-X04 and imaged at multiple time points. Plaque growth was expressed as fold increase in cross-sectional area relative to initial plaque size. A–C, In 6-month-old APP/PS1 mice, plaques exhibited significant growth over 7, 28, and 90 d intervals (paired t test, *p ≤ 0.05). D, Time course of plaque growth in 6- and 10-month-old mice. In 6-month-old animals, the average fold increase in plaque size over a 28 d period was significantly greater than that observed over a 7 d period. There was no difference between average fold increase over a 28 d period compared with that over a 90 d period (one-way ANOVA followed by Dunn's post hoc test). Plaques imaged in 10-month-old animals did not exhibit significant plaque growth at either the 7 or 90 d interval (paired t test). E, Average plaque growth is plotted as a function of initial plaque size. Regardless of animal age, smaller plaques grew at a greater rate compared with larger plaques. F, Plaques imaged in 6-month-old animals under open-skull window preparations did not exhibit significant growth over a 28 d interval.

Figure 3.

Open-skull cranial window preparations are associated with extensive gliosis. A, Low-magnification image of Iba-1 immunofluorescence (red) illustrates extensive microglial activation in cortex under the open-skull cranial window but not in contralateral control cortex. High-power micrographs colabeled for methoxy-X04-positive plaques (blue) reveal little microglial activation in control hemisphere (A1), but robust microglial activation under the open-skull window (A2). B, Low-magnification image of GFAP immunofluorescence (green) demonstrates extensive activation of astrocytes in cortex under the open-skull cranial window. High-power micrographs illustrate astrocytic activation only in the immediate vicinity of plaques in control cortex (B1), whereas robust astrocytic activation is present throughout cortex under the open-skull window (B2). C, Iba-1-positive microglia are not abundantly visible at low magnification under a thinned-skull cranial window preparation. High-power micrographs illustrate rare microglial activation in control cortex (C1) and under the thinned-skull window (C2). D, Astrocytic activation is not robust at low magnification under a thinned-skull window preparation. Astrocytic activation is rare in control cortex (D1) and under the thinned-skull window (D2). Scale bars: A–D, 200 μm; A1–D2, 50 μm.

Figure 4.

γ-Secretase inhibition suppresses plaque growth and new plaque formation. A, B, Representative multiphoton micrographs of plaques imaged at a 28 d interval in a vehicle-treated 6-month-old APP/PS1 mouse illustrate growth of existing plaques (arrows) and appearance of new plaques (arrowheads). C, D, Plaque growth was suppressed in mice treated with the γ-secretase inhibitor, Compound E (3 mg/kg, i.p). All images are collapsed z-stack images (100–120 μm thick). E, Plaques imaged in vehicle-treated mice exhibited significant growth over 7 and 28 d intervals, whereas plaque growth was attenuated in Compound E-treated mice. (n = 4 mice/group; Student's t test, *p ≤ 0.05). The dotted line indicates no growth. Scale bars, 20 μm.

Figure 5.

γ-Secretase inhibition reduces plaque load in 6-month-old APP/PS1 mice. Six- and 10-month-old APP/PS1 mice were treated daily with Compound E (3 mg/kg, i.p.) or vehicle for 28 d. Compound E reduced X-34-positive cortical plaque load by 23% in 6-month-old animals compared with vehicle-treated controls. No difference was present between groups in 10-month-old animals (n = 8 mice/group; Student's t test, *p ≤ 0.05).

Figure 6.

γ-Secretase inhibition results in modest decreases in ISF Aβ levels in cortex. Three-month-old APP/PS1 mice were treated daily with Compound E (3 mg/kg, i.p.) or vehicle for 7 d. In vivo microdialysis was performed to measure ISF Aβ_x-40 and Aβ_x-42 in cortex throughout the final 3 d of drug treatment. A, B, Compound E reduced ISF Aβ_x-40 and Aβ_x-42 levels by 42% and 44%, respectively, during the first 8 h after each treatment compared with vehicle-treated controls. ISF Aβ_x-40 and Aβ_x-42 exhibited more modest decreases during hours 8–16 after Compound E treatment and were no different from controls during hours 16–24 (n = 4/group, two-way ANOVA with post hoc Bonferroni tests; *p ≤ 0.05). C, Expressed as relative levels over 24 h, mice treated with Compound E for 7 d exhibited a 22% decrease in ISF Aβ_x-40 and a 25% decrease in Aβ_x-42 compared with controls (n = 4/group; paired t test; *p ≤ 0.05).