Astroglial dysfunctions drive aberrant synaptogenesis in developing brain with lengthy general anesthesia

Lengthy use of general anesthetics (GAs) causes cognitive deficits in developing brain, which has raised significant clinical concerns such that FDA is warning on the use of GAs in children younger than 3 years. However, the molecular and cellular mechanisms for GAs-induced neurotoxicity remain largely unknown. Here we report that sevoflurane, a commonly used GA in pediatrics, causes compromised astrocyte morphogenesis, spatiotemporally correlated to the synaptic overgrowth with reduced synaptic function in developing cortex in a regional-, exposure-length- and age-specific manner. Sevoflurane disrupts astrocyte Ca2+ homeostasis both acutely and chronically, which leads to the down regulation of Ezrin, an actin-binding membrane protein, which we found is critically involved in astrocyte morphogenesis in vivo. Importantly, in normal developing brain, the genetic intervention of astrocyte morphogenesis is sufficient to produce the aberrant synaptic structure and function virtually identical to the ones induced by lengthy sevoflurane exposure. Our data uncover that astrocytes are unexpectedly central targets for GAs to exert toxic effects, and that astrocyte morphological integrity is crucial for synaptogenesis in the developing brain.


116
Lengthy Sevo exposure disrupts astrocyte morphogenesis with compromised 117 "tripartite synapse" maturation in developing somatosensory cortex 118 To evaluate how lengthy Sevo exposure affects developing mouse brain, we had P7 119 mice exposed to 2.5% Sevo (we termed hereafter Sevo group mice) or their littermates 120 exposed to carrier gas (30% O 2 /70% CO 2 ) (we termed hereafter Control group mice) for 4 121 h. This experimental setting was chosen on the basis of a number of clinical studies which 122 have shown that cognitive dysfunctions were observed in children/infants with GAs 123 exposure longer than 3 h (14). All mice survived after exposure, with both artery blood gas 124 parameters measured immediately after exposure (Supplementary Table1) and body 125 weight at P14 comparable to the Control group mice (Control group vs Sevo group: 6.55 ± 126 0.73 g vs 6.27 ± 0.73 g, P = 0.299, n = 16 mice per group, unpaired t test). 127 Then, we examined in detail how lengthy Sevo exposure affected astrocyte 128 morphogenesis during the critical period of the brain development, i.e., from P7 to P21 in 129 mice (10). To achieve this, we performed intracellular lucifer yellow iontophoresis in 130 lightly fixed tissue followed with confocal imaging and morphological 3D reconstructions 131 to analyze astrocyte morphology in situ (15) at P8, P14 and P21 (Fig. 1A). In 132 somatosensory cortex, astrocytes from Sevo group mice displayed significantly smaller 133 distal (away from soma) and proximal (closed to soma) fine process volume, with 134 decreased territory volume, measured at both P8 and P14. However, when at P21, there 135 was no more significance between the two groups ( Fig. 1B, C). No significant difference 136 was observed between the two groups in astrocyte soma and primary branches (branches 6 137 directly protruding from soma) at P8, P14 and P21 (Supplementary Fig. 1A, B). We also 7 138 quantified GFAP expressions using immunostaining, a well-known marker for astrogliosis, 8 139 and no significant difference between the two groups was found (Supplementary Fig. 1C, 140 D). Therefore, our data so far suggest that Sevo exposure induced compromised astrocyte 141 fine structure development without causing apparent astrogliosis in the mouse cortex. Sevo group mice at P14 (P = 0.038, unpaired t test) and P21 (P = 0.074, unpaired t test).

142
In contrast to the cortex, much less morphological deficits were observed in the CA1 143 stratum radiatum (CA1sr) or in the molecular layer of dentate gyrus (DG-mo) of the 144 hippocampus at P14 (Supplementary Fig. 2) Fig. 3) displayed morphological deficits. Together, our data reveal that lengthy, but not 152 short, Sevo exposure induces compromised astrocyte morphogenesis only in developing 153 but not in mature brain.

154
Astrocyte processes are extremely fine (<50 nm) and form "tripartite synapse" with 155 axonal boutons and dendritic spines (16). We then performed serial block face scanning 156 electron microscopy (SBF-SEM) to examine the ultrastructure of the tripartite synapse.

157
Lower astrocyte volume fraction (which reflects the absolute volume of astrocyte processes) 158 was observed in the somatosensory cortex of Sevo group mice at P14, but not at P21. From 159 P14 to P21, the surface-to-volume ratio (which reflects the fineness of astrocyte processes) 160 was markedly increased by ~50% in Control group mice, but remained unchanged in Sevo 161 group mice (Fig. 1D-F). Then we examined astrocyte-neuron contact. There was also an 162 increase of astrocyte perimeters enwrapping the synaptic cleft by ~30% from P14 to P21 163 in Control group mice, but remained unchanged in Sevo group mice (Fig. 1G). In addition, 164 there was also a more than one-fold increase of astrocyte fine processes insertion into 10 165 synaptic cleft from P14 to P21, as shown by the markedly increased number of astrocyte 166 contacted PSD in Control group mice but remained unchanged in Sevo group mice. The 167 number of astrocyte apposed PSD remained unchanged in both groups from P14 to P21 168 (Fig. 1H, I). 169 The light and electron microscopic data together demonstrate that lengthy Sevo 170 exposure disrupts astrocyte morphogenesis resulting in altered tripartite synaptic structure 171 in a regional-, exposure-length-and age-specific manner.

172
Aberrant synaptic growth and functions spatiotemporally correlated with astrocyte 173 morphological deficits in Sevo group mice 174 Next, we tested whether lengthy Sevo exposure also leads to structural and functional 175 deficits in cortical pyramidal neurons at P21. By performing lucifer yellow iontophersis in 176 combine with morphological 3D reconstruction, we found that the pyramidal neurons in 177 Sevo group mice had higher total and mushroom basal dendritic spine density than those 178 in Control group mice at P21 ( Fig. 2A, B). Interestingly, the dendritic spine density was 179 similar in the hippocampal CA1sr (Supplementary Fig. 4) between the two groups, 180 recalling the unchanged astrocyte morphology in this region. Using SBF-SEM, we 181 identified structurally visible PSDs and reconstructed the corresponding dendritic spines 182 as indictors for excitatory synapses (Fig. 2C). We found a more than 50% higher total 183 dendritic spine density, whereas the mushroom spine density remained unchanged in the 184 primary somatosensory cortex in Sevo group mice (Fig. 2D). This dataset suggests that the 185 synaptic overgrowth was present in parallel with astrocytic structural deficits in the cortex 186 but not in the hippocampus of Sevo group mice, and SBF-SEM data suggest the synaptic 11 187 overgrowth largely resulted from an increase of synapses which are relatively immature or 188 less functional.
14 222 slices from WT mice at P14. Interestingly, we found that acute Sevo exposure led to a 223 gradual and persistent loss of basal Ca 2+ level in cortical developing astrocytes, and the 224 effect was not driven by action potential firing, as basal Ca 2+ level still decreased in the 225 presence of 300 nM tetrodotoxin (TTX) (Fig. 3B, C). Independently, we performed 226 GCaMP6f microinjections at P0, Sevo/Control exposure in vivo at P7, and we found that  Similar to previous studies in mature cortical astrocytes (20), developing cortical astrocytes 241 displayed waves (events area >10 μm 2 ) and microdomain signals (events area 1.5-10 μm 2 ).

242
Consistent with the previous work (17), acute Sevo exposure resulted in decreased 243 frequency and amplitude of Ca 2+ waves (Supplementary Fig. 7B). In addition, Sevo 244 exposure in vivo at P7 also resulted in altered spontaneous Ca 2+ signals properties measured 15 245 at P14, including amplitude, half width ( Fig. 3F-H). We also assessed neurotransmitter-246 evoked Ca 2+ signals and found the peak amplitude of mGluR2/3 agonist LY354740-evoked 247 Ca 2+ signals and mGluR1/5 agonist DHPG-evoked Ca 2+ signals were significantly lower 248 in Sevo group mice. While ATP-evoked and Noradrenaline (NE)-evoked Ca 2+ signals were 249 similar between these two groups (Fig. 3I, J). We further evaluated the expression of 250 mGluR3 and mGluR5 in cortical astrocytes at P14, by using RT-qPCR for magnetic cell 251 sorting (MACS) isolated cortical astrocytes (GLAST + cells) from Control and Sevo group 252 mice (see methods). We found that both mGluR3 and mGluR5 mRNA levels were 253 significantly lower in Sevo group mice (Fig. 3K). 254 Together, this dataset suggests that lengthy Sevo exposure disrupted both acutely and 255 chronically the basal Ca 2+ levels, spontaneous and neurotransmitter-evoked Ca 2+ transients 256 in developing cortical astrocytes.

Down-regulation of Ezrin expression with lengthy Sevo exposure 258
Ezrin is an actin-binding membrane-bound protein expressed mainly within astrocyte 259 processes in the central nervous system (21), and was found required for the structural 260 plasticity of astrocyte processes in culture (22). Since the loss of morphology in Sevo group 261 mice predominantly occurred in the fine processes ( Fig. 1), we hypothesized that Ezrin 262 may be a key structural determinant of astrocyte fine processes in vivo, and its dysfunction 263 may occur following lengthy general anesthesia. To this end, first, we found that, in the 264 somatosensory cortex of P21 mice, Ezrin was only partially colocalized with the astrocyte 265 marker S100β, which labeled mainly the soma and primary branches (Fig. 4A), whereas 266 most of the Ezrin signals seemed to be peripheral to S100β signals. To precisely quantify 16 267 the distribution of Ezrin within astrocyte territory, we fluorescently labeled cortical 268 astrocytes by stereotactic injection of AAV5·gfaABC1D·mCherry into the cortex of P0 269 mice and stained Ezrin at P9, P16 and P21 (Fig. 4B, C). We found that the volume fraction 270 of Ezrin was gradually increased in fine processes but decreased in soma and primary 271 branches of cortical astrocytes from P9 to P21 (Fig. 4C, D). At P21, more than 70% of 272 total Ezrin was localized within astrocyte fine processes. Interestingly, the expression of   283 We then asked whether the down-regulation of Ezrin by early Sevo exposure was due 284 to the disrupted astrocyte Ca 2+ signals (Fig. 3). Intracellular Ca 2+ chelation was achieved 285 by incubating cultured astrocytes with a membrane-permeant Ca 2+ chelator BAPTA-AM 286 (30 μM). The Ezrin expression in the primary cultured astrocytes was significantly 287 decreased after 1 h-incubation with BAPTA-AM (Fig. 5A, B). Interestingly, BAPTA-AM AAV5·gfaABC 1 D·eGFP was microinjected into the mouse cortex at P0 to sparsely label 293 astrocytes in vivo (Fig. 5F). Astrocytes fine process volume was decreased after BAPTA-294 AM incubation for 1 h in the presence of TTX (Fig. 5G, H). Thus, our data so far suggest 295 that reducing astrocyte intracellular Ca 2+ was sufficient to produce the loss of Ezrin and the 296 loss of astrocyte morphology.

301
(see methods), and performed microinjections into the cortex of P0 mice (Fig. 5I). Ezrin-302 miRNAi injection resulted in a significantly down-regulation of Ezrin expression (by ~26%) 303 in the somatosensory cortex of mice at P21, assessed by Ezrin immunostaining 304 (Supplementary Fig. 9A, B). The miRNAi expression is highly specific to astrocytes, as 305 94.4% of mCherry + cells colocalized with S100β and only 5.6% of which colocalized with 306 NeuN (Supplementary Fig. 9C, D). Furthermore, Ezrin-miRNAi or NC-miRNAi 307 injection into the cortex of P0 mice did not cause apparent astrogliosis at P21 308 (Supplementary Fig. 9E). As expected, Ezrin knocking-down resulted in decreased 309 astrocytes fine processes and territory volume in P21 mice ( Fig. 5J-L), whereas the soma 310 volume was slightly increased and primary branches volume and number were unchanged 311 (Supplementary Fig. 9F, G). 312 Together, our data suggest that suppressing intracellular Ca 2+ signals in astrocytes whereas in NC-miRNAi mice, no significant difference was found between the mCherry + 329 and the mCherryarea (Fig. 6A, B), suggesting Ezrin knock-down resulted in an increase 330 of the excitatory synapses within the astrocyte territory. We next recorded the mEPSCs 331 and eEPSCs in L 3-5 pyramidal neurons from Ezrin-miRNAi and NC-miRNAi mice. The 332 mEPSCs frequency was significantly 50% smaller in Ezrin-miRNAi mice compared to 333 NC-miRNAi mice (Fig. 6C, D), while the mEPSCs amplitude was similar between the two 334 groups (Fig. 6E). The AMPAR/NMDAR ratio was significantly 35% smaller in Ezrin-335 miRNAi mice (Fig. 6F, G). The similar extent of reductions in eEPSCs frequency and 336 AMPAR/NMDAR ratio suggest the decreased synaptic transmission are largely AMPA-337 dependent. Together, the structural and functional data suggest that astrocytic Ezrin knock-338 down was sufficient to produce an increase of the synaptic density, with reduced AMPA-339 mediated synaptic transmission, a phenotype virtually identical to the one found in the 340 cortex of Sevo group mice (Fig. 2).

342
There are several key findings from this study, some of which are schematized in

368
Our study indicates that the detailed morphological analysis of astrocytes in addition to 369 conventional staining for astrogliosis or cytoskeletal impairment is needed for a better 370 understanding of astrocyte biology in the context of diseases or injuries.

371
Astrocyte morphological deficits displayed exposure-length-, age-and brain    Sevoflurane exposure protocol P7 mouse littermates were randomly assigned to 2 groups.

528
The stubby spines were not separately classified from long thin spines. determined by fitting the decay to a single exponential by using pCLAMP10.7 software.

584
Ca 2+ imaging Acute brain slices from mice injected with GCaMP6f were prepared for Ca 2+

585
imaging. Slices were imaged using a Nikon A1R + multiphoton microscope with 40x (NA 586 0.8) water immersion objective (Nikon), using the 488 nm laser. Slices were continuously 587 superfused at 1-2 mL/min with oxygenated aCSF at room temperature. Images were 588 typically 512 × 512 pixels/frame with 1-2× optical zoom at a scan rate of 1 sec per frame.

676
Statistics All statistical analyses were performed using Origin 9 software (OriginLab).

677
Data were represented as means ± s.e.m; nonparametric data were represented as medians.

678
Data fitting a parametric distribution were tested for significance using analysis of paired 679 and unpaired Student's two-tailed t tests; Data fitting a nonparametric distribution were 680 tested for significance using two-tailed Mann-Whitney. Data with more than two groups 681 were tested for significance using one-way ANOVA test. Significance was defined as P < 682 0.05.