Effects of sleep and wake on astrocytes: clues from molecular and ultrastructural studies

Abstract

Background: Astrocytes can mediate neurovascular coupling, modulate neuronal excitability, and promote synaptic maturation and remodeling. All these functions are likely to be modulated by the sleep/wake cycle, because brain metabolism, neuronal activity and synaptic turnover change as a function of behavioral state. Yet, little is known about the effects of sleep and wake on astrocytes.

Results: Here we show that sleep and wake strongly affect both astrocytic gene expression and ultrastructure in the mouse brain. Using translating ribosome affinity purification technology and microarrays, we find that 1.4 % of all astrocytic transcripts in the forebrain are dependent on state (three groups, sleep, wake, short sleep deprivation; six mice per group). Sleep upregulates a few select genes, like Cirp and Uba1, whereas wake upregulates many genes related to metabolism, the extracellular matrix and cytoskeleton, including Trio, Synj2 and Gem, which are involved in the elongation of peripheral astrocytic processes. Using serial block face scanning electron microscopy (three groups, sleep, short sleep deprivation, chronic sleep restriction; three mice per group, >100 spines per mouse, 3D), we find that a few hours of wake are sufficient to bring astrocytic processes closer to the synaptic cleft, while chronic sleep restriction also extends the overall astrocytic coverage of the synapse, including at the axon-spine interface, and increases the available astrocytic surface in the neuropil.

Conclusions: Wake-related changes likely reflect an increased need for glutamate clearance, and are consistent with an overall increase in synaptic strength when sleep is prevented. The reduced astrocytic coverage during sleep, instead, may favor glutamate spillover, thus promoting neuronal synchronization during non-rapid eye movement sleep.

Fig. 1

ALDH1L1-eGFP expression is specific for astrocytes. Top panels: Double-labeling studies showing colocalization of ALDH1L1-eGFP (green) and the astrocytic marker GFAP (red). ALDH1L1 is expressed also in GFAP– astrocytes. Middle and bottom panels: Double-labeling studies showing the absence of colocalization between ALDH1L1-eGFP (green) and the oligodendrocyte marker CNP (red, arrows indicate cell bodies) or the neuronal marker NeuN (red). Scale bar = 15 μm

Fig. 2

Sleep/wake pattern and response to sleep deprivation in ALDH1L1 – eGFP-L10a mice. a Twenty-four hour sleep/wake patterns. b, c Twenty-four hour time course of NREM duration and SWA for baseline (BSL) and sleep deprivation (SD). *P < 0.05, significant increase during the first 2 hr of recovery sleep after SD relative to the first 2 hr of BSL (paired t test). For a-c, values are mean ± standard error of the mean. White and black bars indicate the light and dark periods, respectively

Fig. 3

Enrichment analysis of ALDH1L1 – eGFP-L10a IP samples. a qPCR expression (mean ± standard deviation, n = 6, two per group for IP, n = 6, two per group for UB) of the cell-specific marker for astrocytes (Gfap) is consistently enriched in the IP RNA across all groups (S, W and SD), whereas the negative controls (Mbp for oligodendrocytes, Syt1 for neurons) are consistently enriched in the UB samples. b Histograms showing IP/UB ratios expressed in log₂ fold change for S, W and SD samples. In all three experimental groups, the genes identified by [25] as specific for astrocytes (red) are enriched in IP samples, whereas the genes specific for neurons (blue), mature oligodendrocytes (MO, yellow), oligodendrocyte precursor cells (OPCs, orange), microglia (green), endothelial cells (light grey) and pericytes (dark grey) are enriched in S, W and SD UB samples. c Heat diagram (left) showing the expression intensities of several common astrocytic genes in each individual S, W and SD mouse. Arrows indicate gjb6 and slc1a3, the two astrocytic genes selected for Western blot analysis (right): Cx-30 (+19 ± 8 %, Mann–Whitney (MW), P = 0.02), GLAST (+49 ± 17 %, MW, P = 0.02) in SD (n = 4, grey bars) relative to S (n = 4, white bars). Representative bands from S and SD pools (n = 4 mice per pool) are depicted above each bar for Cx-30 and GLAST

Fig. 4

Functional characterization of genes differentially expressed in sleep (S) and wake (W + SD). Top: Heat diagrams show the probeset intensity for each individual animal in the three experimental conditions. Bottom: Functional annotation analysis (DAVID default settings, except for kappa = 4, similarly threshold = 0.7) for S (n = 55) and W + SD (n = 396) genes. The top eight functional annotation clusters in order of enrichment score are shown for S (left) and W + SD (right)

Fig. 5

Astrocytic coverage of cortical synapses increases after CSR. a–d Examples of electron microscope images showing PAP⁻ (a) and PAP⁺ (c) spines, with their relative 3D reconstructions (b, d). In (a) and (c), PAPs are depicted in light blue and spine heads (S) are in yellow. B = presynaptic bouton; A = axon; D = dendrite. The ASI is traced in red and the apposed astrocytic surface on the spine head is in green. Scale bar = 250 nm. In (d), the PAP embracing the spine (*) has been made transparent to allow a view of the spine head underneath. Note that the apposed astrocytic surface (green) contacts the ASI (red) on one side. e Percentage of PAP⁻ and PAP⁺ spines per dendrite in layer II cortical fields for S (n = 9 dendrites; PAP⁻, 17.95 ± 6.54 %; PAP⁺, 82.05 ± 6.53 %), SD (n = 9 dendrites; PAP⁻, 19.16 ± 5.18 %; PAP⁺, 80.84 ± 5.18 %) and CSR (n = 8 dendrites; PAP⁻, 18.37 ± 7.07 %, PAP⁺, 81.62 ± 7.07 %). PAP⁻, Kruskal–Wallis, P = 0.96; PAP⁺, Kruskal–Wallis, P = 0.96. f Correlation between area of the apposed astrocytic surface and spine head volume in S (n = 250, green, r = 0.61, P < 0.0001), SD (n = 300, red, r = 0.6, P < 0.0001) and CSR (n = 254, brown, r = 0.69, P < 0.0001). g Apposed astrocytic surface area in S (n = 250, 0.15 ± 0.17 μm²), SD (n = 300, 0.16 ± 0.17 μm²) and CSR (n = 254, 0.23 ± 0.29 μm²). SD vs CSR, Mann-Whitney (MW), *P < 0.0001; S vs CSR, MW, *P < 0.0001. h Apposed astrocytic surface area to spine head surface area ratio in S (n = 250, 10.77 ± 8.59 %), SD (n = 300, 10.65 ± 8.38 %) and CSR (n = 254, 13.89 ± 9.84 %). SD vs CSR, MW, *P = 0.003, S vs CSR, MW, *P = 0.0005. i Apposed astrocytic surface area to spine head surface area ratio for spines with small (head volume between 0th and 50th percentiles), medium (50th to 75th percentiles) and large (>75th percentile) spine head volume in S (green), SD (red) and CSR (brown). Note that CSR showed higher ratios than S and SD for medium (MW; SD vs CSR, *P = 0.03; S vs CSR, *P = 0.027) and large spines (MW; SD vs CSR, *P = 0.0003; S vs CSR, *P = 0.0002). j Example of spine head including the spine apparatus (*). PAP indicates the astrocytic process, S the spine head and B the presynaptic bouton. Scale bar = 300 nm. k Apposed astrocytic surface area to spine head surface area ratio for spines with and without the spine apparatus in S (green), SD (red) and CSR (brown). Note that CSR showed higher ratios than S and SD for spines with spine apparatus (MW; SD vs CSR, *P = 0.03; S vs CSR, *P = 0.0003). All values are mean ± standard deviation

Fig. 6

Wake brings astrocytic processes closer to the synaptic cleft. a–d Examples of electron microscope images showing PAP^+ASI– (a) and PAP^+ASI+ (c) spines, with their relative tridimensional reconstructions (b, d). PAPs are depicted in light blue and spine heads (S) are in yellow. B = presynaptic bouton, A = axon, D = dendrite. The ASI is traced in red and the apposed astrocytic surface on the spine head is in green. Scale bar = 250 nm. In (b), the PAP embracing the spine (*) has been made transparent to allow a view of the spine head underneath. Note that the apposed astrocytic surface (green) does not reach the ASI (red). In (d), the orientation of the structure shows the contact between the apposed astrocytic surface (green) and the ASI. The line drawing on the right shows the ASI perimeter (red) and the astrocytic perimeter (red) in 3D. e Percentage of PAP^+ASI– and PAP^+ASI+ spines per dendrite in layer II cortical fields for S (n = 9 dendrites; PAP^+ASI–, 31.1 ± 9.6 %; PAP^+ASI+, 51.09 ± 7.88 %), SD (n = 9 dendrites; PAP^+ASI–, 17.38 ± 9.45 %; PAP^+ASI+, 62.93 ± 10.55 %) and CSR (n = 8 dendrites; PAP^+ASI–, 17.51 ± 7 %; PAP^+ASI+, 64.53 ± 9.34 %). PAP^+ASI–, S vs SD, MW, * P = 0.01; S vs CSR, MW, *P = 0.001; PAP^+ASI+, S vs SD, MW, * P = 0.02; S vs CSR, MW, *P = 0.0033. f Astrocytic perimeter in S (n = 148, 0.35 ± 0.32 μm), SD (n = 207, 0.42 ± 0.4 μm) and CSR (n = 171, 0.59 ± 0.63 μm). SD vs CSR, MW, *P = 0.002; S vs CSR, MW, *P < 0.0001. g Astrocytic perimeter to ASI perimeter ratio in S (n = 148, 18.66 ± 14.83 %), SD (n = 207, 19.77 ± 15.79 %) and CSR (n = 171, 25.25 ± 17.08 %). SD vs CSR, MW, *P = 0.0001; S vs CSR, MW, *P = 0.0005. h, i Example of portion of neuropil occupied by PAP (depicted in light blue) and its relative 3D reconstruction. Scale bar = 300 nm. j PAP surface-to-volume ratio in S (n = 314, 27 ± 13 per μm, S vs CSR, MW, *P < 0.0001), SD (n = 365, 27 ± 12 per μm, SD vs CSR, MW, *P < 0.0001) and CSR (n = 307, 33 ± 13 per μm). All values are mean ± standard deviation