Electron microscopy analysis of astrocyte-synapse interactions shows altered dynamics in an Alzheimer's disease mouse model

Abstract

Introduction: Astrocyte-synapse bi-directional communication is required for neuronal development and synaptic plasticity. Astrocytes structurally interact with synapses using their distal processes also known as leaflets or perisynaptic astrocytic processes (PAPs). We recently showed that these PAPs are retracted from hippocampal synapses, and involved in the consolidation of fear memory. However, whether astrocytic synaptic coverage is affected when memory is impaired is unknown.

Methods: Here, we describe in detail an electron microscopy method that makes use of a large number of 2D images to investigate structural astrocyte-synapse interaction in paraformaldehyde fixed brain tissue of mice.

Results and discussion: We show that fear memory-induced synaptic activation reduces the interaction between the PAPs and the presynapse, but not the postsynapse, accompanied by retraction of the PAP tip from the synaptic cleft. Interestingly, this retraction is absent in the APP/PS1 mouse model of Alzheimer's disease, supporting the concept that alterations in astrocyte-synapse coverage contribute to memory processing.

Keywords: APP/PS1; glia; leaflet; memory consolidation; perisynaptic astrocytic processes; synapse; tripartite synapse.

FIGURE 1

Measuring synaptic ultrastructure and astrocyte interactions by 2D EM analysis in mouse hippocampus CA1. (A) Schematic illustration of the methods used. (B) Example 2D EM image of a synapse in mouse hippocampus CA1. The outline of the pre-synaptic and post-synaptic membranes is indicated with a dashed line. The astrocyte is pseudo-colored in blue. Orange lines indicate the membrane interaction between pre-synapse and post-synapse with the astrocyte. (C) Frequency distribution plot showing the percentages of presence of contact types between synapse and astrocyte. (D) Perimeter of the pre-synapse and post-synapse. (E) Contact between astrocyte and pre-synapse or post-synapse, relative to the pre-synapse or post-synapse perimeter, respectively. (F) Example of measurement of PAP tip length (G) and distance between the PAP-tip and PSD (H). The base of the protruding part of the PAP-tip is indicated with the dashed line. (G) Quantification of the PAP-tip to PSD distance. (H) Quantification of the astrocyte PAP tip length. Data is presented as mean SEM. ***P≤ 0.0001. PAP, perisynaptic astrocytic process; PSD, post-synaptic density. Scale bars are 300 nm n = 5/441. Statistical details are reported in Supplementary Table 1.

FIGURE 2

Contextual memory formation induced by delayed shock in WT and APP/PS1 mice does not affect the proportion of astrocyte-synapse contact or synapse size. (A) Freezing levels measured during contextual fear conditioning in WT and APP/PS1 mice at 3 months of age. (B) Frequency distribution plot of contact categories at 4 months of age. (C) Pre-synapse perimeter at 4 months of age. (D) Post-synapse perimeter at 4 months of age. Data is presented as mean ± SEM. *P = ≤ 0.05. (A) WT n = 12, APP/PS1 n = 12; (B) WT HC n = 5/441, APP/PS1 HC n = 5/376, WT DS n = 431, APP/PS1 DS n = 5/438; (C,D) WT HC n = 5/459, APP/PS1 HC n = 5/376, WT DS n = 5/431, APP/PS1 DS n = 5/438. Statistical details are reported in Supplementary Table 1.

FIGURE 3

Contextual fear conditioning causes PAP retraction after 4 h in WT but not in APP/PS1 mice. (A) Representative images showing WT home cage (HC), APP/PS1 HC, WT delayed shock (DS) and APP/PS1 DS conditions. Astrocyte is pseudo-colored in purple. Outline of the pre-synapse is indicated with the red dashed line, the outline of the post-synapse is indicated by the blue dashed line. Scale bar, 100 nm. (B) Shared contact between the astrocyte (Astro) and pre-synapse, relative (rel.) to the pre-synapse perimeter and expressed as percentages (%). (C) Shared contact between the astrocyte and post-synapse, relative to the post-synapse perimeter and expressed as percentages. (D) Quantification of the length of the PAP tip protruding toward the synaptic cleft. (E) Frequency distribution plots of WT HC and WT DS conditions of the data presented in panel (D). (F) Quantification of the distance between the PAP-tip and PSD. (G) Frequency distribution plots of WT HC and WT DS conditions of the data presented in panel (F). Data is presented as mean ± SEM. *P ≤ 0.05; ***P ≤ 0.001; ****P ≤ 0.0001. (B) WT HC n = 5/158, APP/PS1 HC n = 5/108, WT DS n = 5/164, APP/PS1 DS n = 5/156; c: WT HC n = 5/168, APP/PS1 HC n = 5/107, WT DS n = 5/173, APP/PS1 DS n = 5/169; d + f; WT HC n = 5/104, APP/PS1 HC n = 5/67, WT DS n = 5/94, APP/PS1 DS n = 5/98. Statistical details are reported in Supplementary Table 1.