Advertisement

Retraction of astrocyte leaflets from the synapse enhances fear memory

  • Aina Badia-Soteras
    Affiliations
    Department of Molecular and Cellular Neuroscience, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Tim S. Heistek
    Affiliations
    Department of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Mandy S.J. Kater
    Affiliations
    Department of Molecular and Cellular Neuroscience, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Aline Mak
    Affiliations
    Department of Molecular and Cellular Neuroscience, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Adrian Negrean
    Affiliations
    Department of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Michel C. van den Oever
    Affiliations
    Department of Molecular and Cellular Neuroscience, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Huibert D. Mansvelder
    Affiliations
    Department of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Baljit S. Khakh
    Affiliations
    Department of Physiology, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA 90095-1751, USA; Department of Neurobiology, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA 90095-1751, USA
    Search for articles by this author
  • Rogier Min
    Affiliations
    Department of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands

    Department of Child Neurology, Emma Children’s Hospital, Amsterdam University Medical Centers, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • August B. Smit
    Affiliations
    Department of Molecular and Cellular Neuroscience, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
  • Mark H.G. Verheijen
    Correspondence
    Correspondence:
    Affiliations
    Department of Molecular and Cellular Neuroscience, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, The Netherlands
    Search for articles by this author
Open AccessPublished:October 28, 2022DOI:https://doi.org/10.1016/j.biopsych.2022.10.013

      Abstract

      Background

      The formation and retrieval of fear memories depends on orchestrated synaptic activity of neuronal ensembles within the hippocampus and it is becoming increasingly evident that astrocytes residing in the environment of these synapses play a central role in shaping cellular memory representations. Astrocyte distal processes, known as leaflets, fine-tune synaptic activity by clearing neurotransmitters and limiting glutamate diffusion. However, how astroglial synaptic coverage contributes to mnemonic processing of fearful experiences remains largely unknown.

      Methods

      We used electron microscopy to observe changes in astroglial coverage of hippocampal synapses during consolidation of fear memory in mice. To manipulate astroglial synaptic coverage, we depleted Ezrin, an integral leaflet-structural protein, from hippocampal astrocytes using CRISRP/Cas9 gene editing. Next, a combination of FRET analysis, genetically encoded glutamate sensors and whole-cell patch-clamp recordings was used to determine whether the proximity of astrocyte leaflets to the synapse is critical for synaptic integrity and function.

      Results

      We found that consolidation of a recent fear memory is accompanied by a transient retraction of astrocyte leaflets from hippocampal synapses and increased activation of NMDA-receptors. Accordingly, astrocyte-specific depletion of Ezrin resulted in shorter astrocyte leaflets and reduced astrocyte contact with the synaptic cleft, which consequently boosted extrasynaptic glutamate diffusion and NMDA-receptor activation. Importantly, after fear conditioning, these cellular phenotypes translated to increased retrieval-evoked activation of CA1 pyramidal neurons and enhanced fear memory expression.

      Conclusion

      Together, our data show that withdrawal of astrocyte leaflets from the synaptic cleft is an experience-induced temporally-regulated process that gates the strength of fear memories.

      Key words

      Introduction

      Remembering experiences that involve fear, pain or trauma may lead to the development and progression of psychopathologies, such as depression, anxiety, posttraumatic stress disorder (PTSD) and substance abuse (
      • Izquierdo I.
      • Furini C.R.G.
      • Myskiw J.C.
      Fear Memory.
      ). Preclinical and clinical evidence has revealed that the hippocampus and the amygdala, among other regions, are key brain structures associated with maladaptive responses that underlie these psychiatric disorders (
      • Knight D.C.
      • Smith C.N.
      • Cheng D.T.
      • Stein E.A.
      • Helmstetter F.J.
      Amygdala and hippocampal activity during acquisition and extinction of human fear conditioning.
      ,
      • Goosens K.A.
      Hippocampal regulation of aversive memories.
      ,
      • Duvarci S.
      • Pare D.
      Amygdala Microcircuits Controlling Learned Fear.
      ). In particular, coordinated activity of hippocampal neuronal ensembles is thought to be critical in the formation and expression of the contextual information of recently acquired aversive memories (
      • Rao-Ruiz P.
      • Couey J.J.
      • Marcelo I.M.
      • Bouwkamp C.G.
      • Slump D.E.
      • Matos M.R.
      • et al.
      Engram-specific transcriptome profiling of contextual memory consolidation.
      ,
      • Liu X.
      • Ramirez S.
      • Pang P.T.
      • Puryear C.B.
      • Govindarajan A.
      • Deisseroth K.
      • Tonegawa S.
      Optogenetic stimulation of a hippocampal engram activates fear memory recall.
      ,
      • Tayler K.K.
      • Tanaka K.Z.
      • Reijmers L.G.
      • Wiltgen B.J.
      Reactivation of Neural Ensembles during the Retrieval of Recent and Remote Memory.
      ). Converging evidence over the past years indicate that changes in the structural and/or physiological properties of synapses, including interactions with their peri-synaptic environment, are required for the formation, storage and retrieval of recent memories (
      • Leuner B.
      • Falduto J.
      • Shors T.J.
      Associative Memory Formation Increases the Observation of Dendritic Spines in the Hippocampus.
      ,
      • Geinisman Y.
      • Berry R.W.
      • Disterhoft J.F.
      • Power J.M.
      • Van der Zee E.A.
      Associative Learning Elicits the Formation of Multiple-Synapse Boutons.
      ,
      • Rao-Ruiz P.
      • Visser E.
      • Mitrić M.
      • Smit A.B.
      • van den Oever M.C.
      A Synaptic Framework for the Persistence of Memory Engrams.
      ,
      • Govindarajan A.
      • Kelleher R.J.
      • Tonegawa S.
      A clustered plasticity model of long-term memory engrams.
      ). Thus, decoding the molecular and cellular framework that supports memory processes is of utmost importance for understanding memory-related disorders.
      Recently, evidence has emerged that hippocampal astrocytes are implicated in the formation and remodeling of aversive memories and stress-related disorders (
      • Adamsky A.
      • Kol A.
      • Kreisel T.
      • Doron A.
      • Ozeri-Engelhard N.
      • Melcer T.
      • et al.
      Astrocytic Activation Generates De Novo Neuronal Potentiation and Memory Enhancement.
      ,
      • Kol A.
      • Adamsky A.
      • Groysman M.
      • Kreisel T.
      • London M.
      • Goshen I.
      Astrocytes contribute to remote memory formation by modulating hippocampal–cortical communication during learning.
      ,
      • Li Y.
      • Li L.
      • Wu J.
      • Zhu Z.
      • Feng X.
      • Qin L.
      • et al.
      Activation of astrocytes in hippocampus decreases fear memory through adenosine A1 receptors.
      ,
      • Suzuki A.
      • Stern S.A.
      • Bozdagi O.
      • Huntley G.W.
      • Walker R.H.
      • Magistretti P.J.
      • Alberini C.M.
      Astrocyte-Neuron Lactate Transport Is Required for Long-Term Memory Formation.
      ,
      • Gao V.
      • Suzuki A.
      • Magistretti P.J.
      • Lengacher S.
      • Pollonini G.
      • Steinman M.Q.
      • Alberini C.M.
      Astrocytic β 2 -adrenergic receptors mediate hippocampal long-term memory consolidation.
      ). For example, activation of the Gq-coupled receptor hM3Dq in CA1 astrocytes during fear conditioning enhanced recent fear memory (
      • Adamsky A.
      • Kol A.
      • Kreisel T.
      • Doron A.
      • Ozeri-Engelhard N.
      • Melcer T.
      • et al.
      Astrocytic Activation Generates De Novo Neuronal Potentiation and Memory Enhancement.
      ), whereas activation of the Gi-coupled receptor hM4Di impaired remote recall (
      • Kol A.
      • Adamsky A.
      • Groysman M.
      • Kreisel T.
      • London M.
      • Goshen I.
      Astrocytes contribute to remote memory formation by modulating hippocampal–cortical communication during learning.
      ). Thus, it is evident that hippocampal-dependent memory processing of aversive events requires coordinated activity of astrocytes and neurons.
      Astrocytes are morphologically complex glial cells that contact synapses with their thinnest terminal processes, referred to as leaflets or perisynaptic astrocyte processes (PAPs), which together with pre– and postsynaptic neuronal elements are an integral feature of synapses throughout the CNS (
      • Khakh B.S.
      • Sofroniew M.V.
      Diversity of astrocyte functions and phenotypes in neural circuits.
      ,
      • Araque A.
      • Carmignoto G.
      • Haydon P.G.
      • Oliet S.H.R.
      • Robitaille R.
      • Volterra A.
      Gliotransmitters Travel in Time and Space.
      ,
      • Semyanov A.
      • Verkhratsky A.
      Astrocytic processes: from tripartite synapses to the active milieu.
      ). Astrocytic leaflets fine-tune local synaptic transmission and plasticity because they are ideally positioned proximally to the synaptic cleft (
      • Ghézali G.
      • Dallérac G.
      • Rouach N.
      Perisynaptic astroglial processes: dynamic processors of neuronal information.
      ) and contain the molecular machinery to modulate the extracellular space around synapses (
      • Bazargani N.
      • Attwell D.
      Astrocyte calcium signaling: the third wave.
      ,
      • Araque A.
      • Parpura V.
      • Sanzgiri R.P.
      • Haydon P.G.
      Tripartite synapses: glia, the unacknowledged partner.
      ). Astrocyte leaflets are enriched in actin-associated proteins of the Ezrin Radixin Moesin (ERM) family (
      • Derouiche A.
      • Frotscher M.
      Peripheral astrocyte processes: Monitoring by selective immunostaining for the actin-binding ERM proteins.
      ) that dynamically regulate astrocytic filopodia formation in culture (
      • Lavialle M.
      • Aumann G.
      • Anlauf E.
      • Prols F.
      • Arpin M.
      • Derouiche A.
      Structural plasticity of perisynaptic astrocyte processes involves ezrin and metabotropic glutamate receptors.
      ) and astrocyte morphogenesis in the developing somatosensory cortex (
      • Zhou B.
      • Chen L.
      • Liao P.
      • Huang L.
      • Chen Z.
      • Liao D.
      • et al.
      Astroglial dysfunctions drive aberrant synaptogenesis and social behavioral deficits in mice with neonatal exposure to lengthy general anesthesia.
      ). Interestingly, changes in the astroglial coverage of synapses has been observed in vivo in the hypothalamus during lactation (
      • Oliet S.H.R.
      Control of Glutamate Clearance and Synaptic Efficacy by Glial Coverage of Neurons.
      ,
      • Panatier A.
      • Oliet S.H.R.
      Neuron–glia interactions in the hypothalamus.
      ) and sleep/awake cycles (
      • Becquet D.
      • Girardet C.
      • Guillaumond F.
      • François-Bellan A.-M.
      • Bosler O.
      Ultrastructural plasticity in the rat suprachiasmatic nucleus. Possible involvement in clock entrainment.
      ), in the nucleus accumbens after self-administered psychostimulants (
      • Scofield M.D.
      • Li H.
      • Siemsen B.M.
      • Healey K.L.
      • Tran P.K.
      • Woronoff N.
      • et al.
      Cocaine Self-Administration and Extinction Leads to Reduced Glial Fibrillary Acidic Protein Expression and Morphometric Features of Astrocytes in the Nucleus Accumbens Core.
      ,
      • Siemsen B.M.
      • Reichel C.M.
      • Leong K.C.
      • Garcia-Keller C.
      • Gipson C.D.
      • Spencer S.
      • et al.
      Effects of Methamphetamine Self-Administration and Extinction on Astrocyte Structure and Function in the Nucleus Accumbens Core.
      ,
      • Kruyer A.
      • Scofield M.D.
      • Wood D.
      • Reissner K.J.
      • Kalivas P.W.
      Heroin Cue–Evoked Astrocytic Structural Plasticity at Nucleus Accumbens Synapses Inhibits Heroin Seeking.
      ), and in the lateral amygdala following threat conditioning (
      • Ostroff L.E.
      • Manzur M.K.
      • Cain C.K.
      • Ledoux J.E.
      Synapses lacking astrocyte appear in the amygdala during consolidation of pavlovian threat conditioning: Astrocyte-free synapses increase with learning.
      ). Similarly, synaptic activity and synaptic potentiation has been reported to induce remodeling of astrocytic leaflets on a short timescale (30-60 min after stimulation) in the hippocampus (
      • Henneberger C.
      • Bard L.
      • Panatier A.
      • Reynolds J.P.
      • Kopach O.
      • Medvedev N.I.
      • et al.
      LTP Induction Boosts Glutamate Spillover by Driving Withdrawal of Perisynaptic Astroglia.
      ,
      • Perez-Alvarez A.
      • Navarrete M.
      • Covelo A.
      • Martin E.D.
      • Araque A.
      Structural and Functional Plasticity of Astrocyte Processes and Dendritic Spine Interactions.
      ,
      • Bernardinelli Y.
      • Randall J.
      • Janett E.
      • Nikonenko I.
      • König S.
      • Jones E.V.
      • et al.
      Activity-Dependent Structural Plasticity of Perisynaptic Astrocytic Domains Promotes Excitatory Synapse Stability.
      ,
      • Lushnikova I.
      • Skibo G.
      • Muller D.
      • Nikonenko I.
      Synaptic potentiation induces increased glial coverage of excitatory synapses in CA1 hippocampus.
      ,
      • Wenzel J.
      • Lammert G.
      • Meyer U.
      • Krug M.
      The influence of long-term potentiation on the spatial relationship between astrocyte processes and potentiated synapses in the dentate gyrus neuropil of rat brain.
      ), however these studies were performed in fixed and acute slices, and whether this structural plasticity of astrocytic leaflets occurs in vivo and on a longer timescale (hours to days), such as in long-term memory processing, is an important and intriguing question that remains to be addressed. Accordingly, whether astrocytic Ezrin is required for the structural and functional integrity of hippocampal synapses and memory processes has remained unexplored.
      To this end, we examined astrocyte leaflet-synapse spatial interaction in the hippocampus during the formation and consolidation of fear memories and found a learning-induced transient retraction of astrocyte leaflets. Altogether, our data reveal that the apposition of astrocyte leaflets to the synaptic cleft influences contextual fear memory expression and gates neuronal activation in a context-dependent manner.

      Methods and materials

      Description of additional methods are available in Supplementary information.

      Animals

      Wild-type, male C57BL/6J mice were 8-10 weeks old at the start of experiments and were individually housed on a 12 h light/dark cycle with ad libitum access to food and water. Behavioral experiments were performed during the light phase. All experimental procedures were approved by The Netherlands central committee for animal experiments (CCD) and the animal ethical care committee (DEC) of the Vrije Universiteit Amsterdam (AVD1120020174287). Mice were randomly assigned to experimental groups.

      Constructs

      The pAAV- GfaABC1D::Cas9-HA-Ezr was generated by first replacing the CMV promoter from pX601-AAV-CMV::NLS-saCas9-NLS-3xHA-bGHpA;Bsal-sgRNA (Addgene plasmid #61591) with the GFAP promoter (GfaABC1D). Next, we inserted the designed sgRNA to target exon-1 of Ezrin (TGGCTGGTTGGTGGCTCTGCGTGGGT, Genscript: NM_001271663.1_T3; CCGTCGCCTCCGCCGTACAGCCGAAT, Genscript: NM_001271663.1_T2). Finally, we cloned the modified plasmid to an AAV2/5 vector. The control virus, pAAV- GfaABC1D::Cas9-HA, was generated following the same procedure but it lacks the sequence to target Ezrin.

      Neuron-astrocyte proximity assay (Napa) and FRET analyses

      FRET from acute slices was examined 4-6 weeks following AAV injections in vivo. We measured FRET by sensitized emission (SE-FRET) using PixFRET, and ImageJ Plug-in. Image processing was carried out as described previously (
      • Badia‐Soteras A.
      • Octeau J.C.
      • Verheijen M.H.G.
      • Khakh B.S.
      Assessing Neuron–Astrocyte Spatial Interactions Using the Neuron–Astrocyte Proximity Assay.
      ).

      Results

      Contextual fear learning causes transient retraction of astrocyte leaflets from the synaptic cleft

      We first assessed the spatial organization of astrocyte leaflets in apposition to the synaptic elements after contextual fear conditioning (CFC). Mice underwent CFC and astroglial synaptic coverage was determined using electron microscopy (EM) in the CA1 at successive time-points; 30 min, 5 days and 28 days, after training (Figure 1A). Mice that remained in their home-cage (HC) were used as controls. The number of synapses without astrocyte contact (no contact; Figure 1B) and synapses contacted by an astrocyte leaflet (only postsynapse, only presynapse and both; Figure 1B) did not differ between timepoints (Figure 1B). Also, compared to HC control, the extent of astrocyte contact with the pre- and postsynapse (Figure S1A,B), as well as the perimeter of both synaptic structures (Figure S1C,D) were unaltered during encoding and consolidation (30 min and 5 days). A decrease in postsynapse-leaflet contact was found at the remote time-point (28 days) (Figure S1B), which is likely related to the observed increased spine size (Figure S1D), as astroglial synapse coverage inversely correlates with spine size (
      • Herde M.K.
      • Bohmbach K.
      • Domingos C.
      • Vana N.
      • Komorowska-Müller J.A.
      • Passlick S.
      • et al.
      Local Efficacy of Glutamate Uptake Decreases with Synapse Size.
      ). Importantly, and consistent with the notion that encoding and retrieval of recent contextual memories are mediated by the hippocampus (
      • Liu X.
      • Ramirez S.
      • Pang P.T.
      • Puryear C.B.
      • Govindarajan A.
      • Deisseroth K.
      • Tonegawa S.
      Optogenetic stimulation of a hippocampal engram activates fear memory recall.
      ,
      • Denny C.A.
      • Kheirbek M.A.
      • Alba E.L.
      • Tanaka K.F.
      • Brachman R.A.
      • Laughman K.B.
      • et al.
      Hippocampal Memory Traces Are Differentially Modulated by Experience, Time, and Adult Neurogenesis.
      ), we observed a rapid (30 min) and persistent (5 days) shortening of astrocyte leaflets following CFC that was no longer present at a remote time-point (28 days) (Figure 1C,E, Figure S1E-G). These observations were complemented by an increased distance of the astrocyte leaflet to the synaptic cleft at 30 min and 5 days after CFC (Figure D,E, Figure SH-J). In order to examine whether the changes in astrocyte coverage were specific to associative learning, we analyzed a group at 5 days after immediate shock exposure (5 d IS). We found that the 5d IS group exhibited a slightly shorter leaflet tip than the HC group, whereas no differences were observed for the astrocyte leaflet distance to the synaptic cleft (Figure S1K,L). It is important to note that although there is a significant reduction of the astrocyte leaflet tip after IS, this difference is minor (8 nm, 13% of HC; Figure S1K) compared to the change observed 5 days after CFC (17 nm, 25% of HC; Figure 1C), indicative that the stress of an aversive stimulus and/or exposure to a new context may cause minor changes in the leaflet tip length, but this is not sufficient to cause a detectable increase in the distance of the leaflet tip from the synaptic cleft. Taken together, encoding and consolidation of contextual fear memory is accompanied by a transient retraction of astrocyte leaflets from the synaptic cleft.
      Figure thumbnail gr1
      Figure 1Contextual fear conditioning induces transient retraction of the astrocyte leaflet from the synaptic cleft. A) Experimental workflow: mice were sacrificed 30 min, 5 days and 28 days after conditioning. Home-caged mice were used as control. B) Frequency distribution for the different types of astrocyte leaflet-synapse contact: synapse without an astrocyte contact (no contact), astrocyte leaflet contacts only the presynaptic button (presynapse), astrocyte leaflet contacts only the postsynapse spine (postsynapse) and astrocyte leaflet contacts both synaptic elements (both) at the depicted timepoints after CFC. Chi square: x2 (9, 742) = 4.77, p = 0.85. n.s = not significant. C) Quantification of astrocyte leaflet tip length at the depicted timepoints after CFC (HC: 130 synapses from 6 mice; 30 min: 152 synapses from 6 mice; 5 days: 123 synapses from 6 mice; 28 days: 86 synapses from 6 mice); F (3, 487) = 11.6, p<0.0001 post-hoc Bonferroni test: HC vs 30min * p = 0.01, HC vs 5d ****p<0.0001, HC vs 28d p = 0.99, 30min vs 5d p = 0.07, 30min vs 28d p = 0.06, 5d vs 28d ****p<0.0001. D) Quantification of astrocyte leaflet distance from the post-synaptic density (PSD) at depicted time points after CFC (HC: 122 synapses from 6 mice, 30 min: 142 synapses from 6 mice, 5 days: 133 synapses from 6 mice, 28 days: 80 synapses from 6 mice); F (3, 117) = 9.32, p<0.0001 post-hoc Bonferroni test: HC vs 30min * p = 0.01, HC vs 5d **** p<0.0001, HC vs 28d p = 0.41, 30min vs 5d * p = 0.15, 30min vs 28d p = 0.62, 5d vs 28d * p = 0.01. (right:) Illustration depicting how the astrocyte leaflet distance from the PSD is measured E) Representative images of astrocyte leaflet-synapse interaction during fear memory processing. In light purple, an astrocyte leaflet contacting an excitatory synapse is depicted. Scale bar: 200 nm (upper row) and 100 nm (bottom row). Data is presented as mean ± SEM. HC, home-cage; 30min, 30 minutes; 5d, 5 days; 28d, 28 days.

      Structural manipulation of astrocyte leaflets by CRISPR/Cas9-mediated depletion of Ezrin

      To mimic the EM observed changes in leaflet structure and to subsequently study synaptic function and behavior in vivo, we specifically reduced Ezrin expression in CA1 astrocytes of adult mice by delivering an adeno-associated viral vector (AAV2/5) encoding the saCas9 enzyme under the control of the minimal astrocyte-specific glial fibrillary acidic protein (GFAP) promoter, and a single guide RNA (sgRNA) complementary to the Ezrin gene (exon-1), (AAV2/5-GfaABC1D::Cas9-HA-Ezr) (Figure 2A). Stereotaxic delivery of this vector was restricted to dorsal CA1 astrocytes (Figure S2A,B), with high transduction efficiency (82.7 ± 2.9%; mean ± SEM; of astrocytes expressed Cas9) and very high specificity (Figure 2B,C). Next, we examined Ezrin levels by combining RNA scope in situ hybridization and immuno-fluorescence (IF) analysis, 4-5 weeks after AAV delivery. We found strong reduction (72%) of Ezrin RNA levels in the dorsal CA1 of mice expressing saCas9 together with the sgRNA against Ezrin (Ezr-saCas9) compared to controls expressing only saCas9 (Control: 27.4 ± 2.7, Ezr-saCas9: 7.5 ± 1.2 RNA molecules per astrocyte) (Figure 2D,E). Similarly, Ezrin protein levels were found to be reduced in Ezr-saCas9 mice compared to controls (Ezr-saCas9: 27.1 ± 5.7% Ezrin signal a.u relative to control) (Figure 2F,G). To determine possible virus-induced reactive astrogliosis (
      • Ortinski P.I.
      • Dong J.
      • Mungenast A.
      • Yue C.
      • Takano H.
      • Watson D.J.
      • et al.
      Selective induction of astrocytic gliosis generates deficits in neuronal inhibition.
      ) we investigated the number of astrocytes, and GFAP protein levels, but did not find signs of gliosis (Figure S2C-E). Thus, the current approach enabled us to optimally and selectively reduce Ezrin expression in a large population of hippocampal astrocytes and to bypass any developmental role of this protein.
      Figure thumbnail gr2
      Figure 2CRISPR/saCas9 viral approach to target Ezrin in adult hippocampal astrocytes in vivo. A) Experimental workflow: CNT and Ezr-saCas9 AAVs were micro-injected bilaterally in the CA1. Immuno-fluorescence (IF) and in situ hybridization experiments were performed 5 weeks later. B) Representative images of the transduction and specificity of viral vectors assessed by immuno-fluorescence (IF). Scale bar: 50 μm. C) Quantification of penetrance (718 cells from 5 mice) and specificity (NeuN: 837 cells from 5 mice, Iba-1: 364 cells from 5 mice). D) RNA scope in situ hybridization representative images for control and Ezr-saCas9 mice. Scale bar: 10 μm. E) Quantification of the number of RNA molecules of Ezrin per astrocyte (Control: 128 cells from 4 mice, Ezr-saCas9: 179 cells from 4 mice). Nested t test: t305 = 23.69, **** p <0.00001. F) IF representative images for control and Ezr-saCas9 mice. Scale bar: 30 μm. G) Quantification of Ezrin protein levels per astrocyte (%) (Control: 110 astrocytes from 4 mice, Ezr-saCas9: 102 astrocytes from 4 mice). Nested t test: t82 = 15.09, *** p = 0.0001. Data are presented as mean ± SEM. CNT, control; IF, immuno-fluorescence.

      Astrocyte morphological complexity and astrocyte-neuron proximity are decreased in the hippocampus after Ezrin deletion

      Next, we determined whether Ezrin indeed shapes astrocyte morphology and astrocyte-neuron contacts in the adult hippocampus. To this end, we made use of a Napa-a viral vector (
      • Octeau J.C.
      • Chai H.
      • Jiang R.
      • Bonanno S.L.
      • Martin K.C.
      • Khakh B.S.
      An Optical Neuron-Astrocyte Proximity Assay at Synaptic Distance Scales.
      ) to visualize astrocyte territories in the CA1 region of the hippocampus in the presence and absence of CRISPR/saCas9-mediated deletion of Ezrin. We found that Ezr-saCas9 mice had smaller astrocyte territories compared to controls (Figure 3A-C). To determine whether this smaller territory in Ezr-saCas9 mice was accompanied by shorter astrocyte leaflets, we measured their proximity to synapses using the FRET-based Napa technique that captures spatial interactions within ∼10 nm range between astrocyte processes and synapses in acute brain slices (
      • Octeau J.C.
      • Chai H.
      • Jiang R.
      • Bonanno S.L.
      • Martin K.C.
      • Khakh B.S.
      An Optical Neuron-Astrocyte Proximity Assay at Synaptic Distance Scales.
      ). Astrocytic membranes in the CA1 stratum radiatum were labelled with the FRET donor GFP (Napa-a) and Schaffer Collateral presynaptic terminals were labelled with the FRET acceptor mCherry (Napa-n), to allow detection of FRET signals at the synaptic scale. We prepared acute hippocampal slices 4-5 weeks after viral injection, and using confocal imaging to detect FRET signals in the CA1 (Figure 3D,E) we found that the number of synapses contacted by an astrocyte was significantly reduced in Ezr-saCas9 mice compared to controls (Figure 3F,G). Importantly, this reduction in FRET was not due to differential expression of Napa-a or Napa-n across groups (Figure S3).
      Figure thumbnail gr3
      Figure 3Deletion of Ezrin in mature astrocytes reduces morphological complexity and neuron-astrocyte interaction. A) Experimental workflow to assess astrocyte morphology. B) Representative images of the astrocyte domain for control and Ezr-saCas9 mice. Astrocyte territory is outlined in white. Scale bar: 20 μm. C) Quantification of the astrocyte territory area (μm2) (Control: 108 astrocytes from 4 mice, Ezr-saCas9: 109 astrocytes from 4 mice). Nested t test: t191 = 3.78, ** p = 0.002. D) Experimental workflow to assess the proximity of CA1 astrocyte leaflets to Schaffer Collateral inputs using Napa. E) Left: representative expression of Napa-n (red) and Napa-a (green) in the CA3 and CA1, respectively. Scale bar: 500 μm. Right: representative image of CA1 astrocytes expressing Napa-a (green) together with HA-saCas9 (magenta), and presynaptic puncta from CA3 projecting neurons (red). Scale bar: 50 μm. F) Representative images of CA1 astrocytes from control and Ezr-saCas9 mice expressing Napa-a (green) in proximity to projections from CA3 neurons expressing Napa-n (red). The FRET ROIs are shown on a colour scale that reflects FRET efficiency. Scale bar: 20 μm. G) Quantification of the number of FRET signals (ROIs) per CA1 astrocyte (Control: 101 astrocytes from 5 mice, Ezr-saCas9: 141 astrocytes from 5 mice). Nested t test: t63 = 3.97, *** p = 0.0002. Data are presented as mean ± SEM. CNT, control; IF, immuno-fluorescence.

      Ezrin is required for close proximity of astrocyte leaflets to the synaptic cleft

      Next, we employed EM to closely evaluate the ultrastructure of synapses in the absence of astrocytic Ezrin. Ezr-saCas9 and Control virus were injected in the CA1 together with AAV-GfaABC1D-GFP to visualize the injection site and confirm sufficient viral expression in samples that were subject to EM analysis (Figure 4A, Figure S4A). Our manipulation did not change the number of synapses contacted by an astrocyte (Figure 4C), instead we found that the leaflet tip, which is in direct opposition of the synaptic cleft, was shorter in Ezr-saCas9 mice (Figure 4D,E). Consequently, we observed an increased distance between the astrocyte leaflet tip and the synaptic cleft in Ezr-saCas9 mice (Figure 4F,G). It should be noted that our manipulation slightly reduced astrocytic contact with the presynaptic button (Figure S4B) whereas it did not change contact of astrocytes with postsynaptic spines (Figure S4C), nor did it affect the perimeter of the pre- and postsynaptic elements (Figure S4D-E). Notably, the reduction of astrocytic synaptic coverage following deletion of Ezrin highly resembles the physiological reduction of astrocytic synaptic coverage observed at 5 days after fear conditioning, i.e., increased shorter leaflet tips and reduced longer leaflet-PSD distances (Figure S4F,G). Together these data show that the depletion of Ezrin in astrocytes is an adequate mimic of the leaflet retraction after fear conditioning.
      Figure thumbnail gr4
      Figure 4Deletion of Ezrin in hippocampal astrocytes increases the distance between astrocytic leaflets and the synaptic cleft. A) Experimental workflow. Control (CNT) and Ezr-Cas9 AAVs were injected in a cocktail with AAV-GfaABC1D-eGFP to visualize the injection site prior to EM analysis. B) Representative images of synapse structure from CNT and Ezr-saCas9 mice. In light purple, an astrocyte leaflet contacting an excitatory synapse is depicted. Scale bar: 200 nm. C) Frequency distribution for the different types of astrocyte leaflet-synapse contact: synapse without an astrocyte contact (no contact), astrocyte leaflet contacts only the presynaptic button (presynapse), astrocyte leaflet contacts only the postsynapse spine (postsynapse) and astrocyte leaflet contacts both synaptic elements (both). Chi square: x2 (4, 258) = 0.38, p = 0.98. n.s= not significant. D) Quantification of astrocyte leaflet tip length from CNT and Ezr-saCas9 mice (CNT: 181 astrocytes from 8 mice, Ezr-saCas9: 179 astrocytes from 9 mice). Nested t test: t352 = 7.17, **** p < 0.0001. E) Frequency distribution plot from data presented in D. Note that Ezr-saCas9 mice show a clear shift towards a smaller astrocyte leaflet tip. Mann Whitney test **** p<0.0001. F) Quantification of the distance between the astrocyte leaflet tip and the post-synaptic density (PSD) from CNT and Ezr-saCas9 mice (CNT: 176 astrocytes from 8 mice, Ezr-saCas9: 173 astrocytes from 9 mice). Nested t test: t161 = 5.1, *** p = 0.0001. G) Frequency distribution plot from data presented in F. Note that Ezr-saCas9 mice show a shift towards larger distances. Mann Whitney test **** p < 0.0001. Data are presented as mean ± SEM. CNT, control.

      Reduced astrocyte contact decreases synaptic glutamate levels

      To examine whether Ezr-saCas9-mediated manipulation of astrocyte leaflet structure has consequences for synaptic function, we analyzed excitatory synaptic transmission 5 weeks after virus injection in CA1 (Figure 5A). Whole-cell patch-clamp recordings from hippocampal pyramidal neurons revealed no change in sEPSC amplitude and frequency, suggesting that basal synaptic transmission was unaffected in Ezr-saCas9 mice (Figure 5B-D). Next, we measured evoked AMPAR-mediated EPSCs and synaptic glutamate levels in Ezr-saCas9 mice. For this, CA3-CA1 Schaffer Collaterals were stimulated with or without γ-D-glutamylglycine (γ-DGG), a low-affinity competitive antagonist of AMPARs, at a non-saturating concentration (1 mM) (
      • Liu G.
      • Choi S.
      • Tsien R.W.
      Variability of Neurotransmitter Concentration and Nonsaturation of Postsynaptic AMPA Receptors at Synapses in Hippocampal Cultures and Slices.
      ). At this concentration, effectiveness of the drug depends on synaptic glutamate levels. Evoked EPSC amplitude and decay kinetics were not changed prior to drug application (Figure S5A-C). However, we found that γ-DGG inhibition of evoked AMPAR EPSCs was more pronounced in pyramidal cells from Ezr-saCas9 mice than from controls (Figure 5E-H). Together, this shows that basal strength of synaptic transmission was unaffected in Ezr-saCas9 mice, but that evoked synaptic glutamate levels were reduced in Ezr-saCas9 mice compared to controls. While this reduction was apparently insufficient to affect spontaneous or evoked AMPAR EPSC amplitude, it altered the competition between γ-DGG and glutamate. To determine whether a reduction in presynaptic release probability (
      • Oliet S.H.R.
      Control of Glutamate Clearance and Synaptic Efficacy by Glial Coverage of Neurons.
      ) underlies the observed y-DGG effect in Ezr-saCas9 mice, we examined the paired-pulse ratio (PPR) and found no significant PPR differences between Ezr-saCas9 mice and controls (Figure 5I,J). Taken together, we show that Ezr-saCas9 mice have reduced synaptic glutamate levels after stimulation of glutamatergic terminals, which is likely not due to a presynaptic release defect.
      Figure thumbnail gr5
      Figure 5Reduced astroglial contact due to loss of Ezrin does not affect basal synaptic transmission but it decreases synaptic glutamate levels. A) Experimental workflow. Mice were injected with CNT or Ezr-saCas9 AAVs in a cocktail with AAV-GfaABC1D-eGFP for visualization purposes. Whole-cell recordings were performed five-to-seven weeks later. B) Representative traces of spontaneous AMPAR EPSCs from pyramidal neurons from Control and Ezr-saCas9 mice. C,D) Quantification of (C) interval and (D) amplitude of sEPSCs (Control: 13-14 cells from 7 mice, Ezr-saCas9: 14-15 cells from 7 mice). Unpaired t test: (interval) t25 = 0.45, p = 0.65; (amplitude) t27 = 0.81, p = 0.42; n.s = not significant. E) Upper: representative image of the dorsal hippocampus with an adjacently placed glass pipette (rec.). The stimulation glass pipette (stim.) was placed in the Stratum Radiatum. Scale bar: 200 μm. Bottom: representative image of a patched neuron filled with biocytin (red) surrounded by astrocytes expressing GFP and HA-saCas9. Scale bar: 100 μm. F) Representative traces of evoked AMPAR EPSCs (eEPSCs) from pyramidal neurons from Control and Ezr-saCas9 mice with and without γ-DGG (1 mM). J,H) Quantification of the (J) percentage of eEPSC amplitude inhibited by γ-DGG and (H) decay kinetics in presence of γ-DGG (Control: 15 cells from 6 mice, Ezr-saCas9: 13 cells from 6 mice). Unpaired t test: (amplitude) t26 = 2.087, * p = 0.04; (decay) t26 = 0.46, p = 0.64; n.s = not significant. I) Representative traces of paired pulse ratio (PPR) from Control and Ezr-saCas9 mice. J) Quantification of PPR from Control and Ezr-saCas9 mice (Control: 17 cells from 7 mice, Ezr-saCas9: 12 cells from 7 mice). Unpaired t test: t27 = 1.38, p = 0.17. Data are presented as mean ± SEM. CNT, control; y-DGG, γ-D-glutamylglycine .

      Increased glutamate spill over and NMDA receptor activation upon shortening of astrocyte leaflets

      Next, the temporal dynamics of extrasynaptic glutamate were investigated. For this, the glutamate sensor iGluSnFR (
      • Marvin J.S.
      • Borghuis B.G.
      • Tian L.
      • Cichon J.
      • Harnett M.T.
      • Akerboom J.
      • et al.
      An optimized fluorescent probe for visualizing glutamate neurotransmission.
      ) was expressed in CA1 astrocytes, together with either Ezr-saCas9 or control virus (Figure 6A). Two-photon imaging of iGluSnFR signals allowed us to study the time course of extrasynaptic glutamate following synaptic stimulation. Synaptic activity was evoked by focal electrical stimulation (10 pulses at 50 Hz) in the CA1 (Figure 6B). We did not observe significant changes in the magnitude of iGluSnFR signal between groups (Figure S6A,B). To examine temporal dynamics of glutamate levels, we fitted the decay of the averaged evoked glutamate transients (6 sweeps) with a single exponential and found that the glutamate transients in Ezr-saCas9 mice displayed an increased decay time following trains of 50 Hz stimulation (Figure 6C-E). The decay of the iGluSnFR transients was not influenced by the amplitude of the response (Figure S6C,D). Together, these data suggest increased dwelling of glutamate in the extrasynaptic space after deletion of Ezrin in astrocytes, which is in line with the observed decrease in astroglial coverage of excitatory synapses in Ezr-saCas9 mice.
      Figure thumbnail gr6
      Figure 6Decreased astroglial synaptic contact boosts glutamate spillover and NMDAR activation. A) Experimental workflow. Mice were injected with CNT and Ezr-saCas9 AAVs in a cocktail with AAV-GfaABC1D-iGluSnFr. Five weeks later glutamate and electrophysiological measurements were performed. B) Left: representative images of CA1 astrocytes expressing AAV-GfaABC1D::iGluSnFR and AAV-GfaABC1D::Cas9-HA-Ezr. Scale bar: 200 μm. Right: stimulation glass pipette was placed adjacently to an astrocyte expressing iGluSnFR and transients were measured and quantified from the region of interest (ROI). Scale bar: 20 μm. C,D) Upon synaptic stimulation (10 x 50 Hz), a robust increase in iGluSnFR signal was detected. C) Thick blue line represents the average of 6 responses. D) Thick black (control) and red (Ezr-saCas9) lines represent the single-exponential fit of the decay. E) Quantification of the decay kinetics (Control: 22 astrocytes from 7 mice, Ezr-saCas9: 31 astrocytes from 7 mice) following 50 Hz stimulation. Unpaired t test: t51 = 2.06, * p = 0.02). F) Upper: representative image of the dorsal hippocampus with an adjacently placed glass pipette (rec.). The stimulation glass pipette (stim.) was placed in the Stratum Radiatum. Scale bar: 200 μm. Bottom: representative image of a patched neuron filled with biocytin (red) surrounded by astrocytes expressing AAV-GfaABC1D::GFP and AAV-GfaABC1D::saCas9-HA-Ezr. Scale bar: 100 μm. G) Representative traces of the NMDAR EPSCs upon incremental short burst stimulation (
      • Izquierdo I.
      • Furini C.R.G.
      • Myskiw J.C.
      Fear Memory.
      ,
      • Knight D.C.
      • Smith C.N.
      • Cheng D.T.
      • Stein E.A.
      • Helmstetter F.J.
      Amygdala and hippocampal activity during acquisition and extinction of human fear conditioning.
      ,
      • Duvarci S.
      • Pare D.
      Amygdala Microcircuits Controlling Learned Fear.
      ,
      • Leuner B.
      • Falduto J.
      • Shors T.J.
      Associative Memory Formation Increases the Observation of Dendritic Spines in the Hippocampus.
      ) for control and Ezr-saCas9 mice. H) Quantification of the area under the curve (AUC) (Control: 17 cells from 7 mice, Ezr-saCas9: 27 cells from 9 mice). Repeated measures two-way ANOVA, interaction effect Pulses x Genotype; F (1.23,15.17) = 6.67, * p = 0.01. Post-hoc Bonferroni test control vs Ezr-saCas9: 1 pulse p = 0.97, 2 pulses p = 0.77, 4 pulses p = 0.35, 8 pulses * p = 0.03. I) Representative traces of the NMDAR EPSCs upon incremental short burst stimulation (
      • Izquierdo I.
      • Furini C.R.G.
      • Myskiw J.C.
      Fear Memory.
      ,
      • Knight D.C.
      • Smith C.N.
      • Cheng D.T.
      • Stein E.A.
      • Helmstetter F.J.
      Amygdala and hippocampal activity during acquisition and extinction of human fear conditioning.
      ,
      • Duvarci S.
      • Pare D.
      Amygdala Microcircuits Controlling Learned Fear.
      ,
      • Leuner B.
      • Falduto J.
      • Shors T.J.
      Associative Memory Formation Increases the Observation of Dendritic Spines in the Hippocampus.
      ) for home-caged (HC) and fear conditioned (CFC) mice. J) Quantification of the AUC (HC: 21 cells from 6 mice, CFC: 18 cells from 6 mice). Repeated measures two-way ANOVA, interaction effect Pulses x Genotype; F (1.24,18.28) = 4.96, * p = 0.03. Post-hoc Bonferroni test HC vs CFC: 1 pulse p = 0.12, 2 pulses # p = 0.05, 4 pulses * p = 0.04, 8 pulses # p = 0.08.. Data are presented as mean ± SEM. CNT, control; HC, home-cage; CFC, contextual fear conditioning; Ephys, electrophysiology.
      As astrocyte leaflets contain the high-affinity glutamate transporter GLT-1, we next tested whether the lack of Ezrin might slow glutamate clearance. We found that partial blockade of glutamate transporters (GluTs) with threo-beta-Benzyloxyaspartate (DL-TBOA) (10 μM) significantly similarly increased the decay kinetics of glutamate transients in both groups (Figure S6E,F). Thus, these data suggest that even though astrocyte leaflets are further away from the synaptic cleft in Ezr-saCas9 mice, the uptake of glutamate remains unchanged.
      Perisynaptic glutamate has been shown to increase the activation of extrasynaptic NMDA receptors (
      • Kullmann D.M.
      • Asztely F.
      Extrasynaptic glutamate spillover in the hippocampus: evidence and implications.
      ). We therefore stimulated CA1 pyramidal neurons with incremental short bursts (1-2-4-8 stimuli) of high frequency stimulation (100 Hz) to increase glutamate spillover (
      • Lozovaya N.A.
      • Grebenyuk S.E.
      • TSh Tsintsadze
      • Feng B.
      • Monaghan D.T.
      • Krishtal O.A.
      Extrasynaptic NR2B and NR2D subunits of NMDA receptors shape ‘superslow’ afterburst EPSC in rat hippocampus: NMDA receptor subunits in hippocampus.
      ), and measured NMDAR-mediated EPSCs (Figure 6A). Ezr-saCas9 mice showed a remarkable increase in NMDAR-mediated eEPSCs with increased stimulation frequency, resulting from a larger amplitude of events as well as slower decay (Figure 6F-H, Figure S7A-C). These results are consistent with the interpretation that reduced astrocyte leaflet-synapse interaction leads to enhanced synaptic cooperation (via activation of extrasynaptic NMDARs and/or NMDARs in neighboring synapses) (
      • Henneberger C.
      • Bard L.
      • Panatier A.
      • Reynolds J.P.
      • Kopach O.
      • Medvedev N.I.
      • et al.
      LTP Induction Boosts Glutamate Spillover by Driving Withdrawal of Perisynaptic Astroglia.
      ,
      • Arnth-Jensen N.
      • Jabaudon D.
      • Scanziani M.
      Cooperation between independent hippocampal synapses is controlled by glutamate uptake.
      ). Next, we determined whether enhanced NMDAR activation was an endogenous process occurring during fear memory consolidation as a result of astrocyte leaflet retraction (Figure 1). We found that mice that underwent CFC showed an increase in NMDAR-mediated eEPSCs (Figure 6I,J and Figure S7D-F), similar to the effects observed in Ezr-saCas9 mice. Together, we show that increased activation of hippocampal NMDARs upon CFC can be recapitulated by saCas9-induced astrocyte leaflet retraction through deletion of astrocytic Ezrin.

      Shortening of astrocyte leaflets enhances recent contextual fear memory retrieval and increases neuronal activity

      Next, mice were fear conditioned 5 weeks after viral delivery, and memory expression was measured at recent and remote time-points after learning. Interestingly, Ezr-saCas9 mice exhibited enhanced fear expression when compared to control mice 5 days after CFC (Figure 7B), a phenotype that was replicated using a second sgRNA that targets a different region of the Ezrin gene (Figure S8). No change in freezing levels were detected 28 days after CFC (Figure S9A). Furthermore, no differences were found between controls and Ezr-saCas9 mice in the open field, indicating that the observed phenotype was not driven by anxiety or locomotor deficits (Figure S9B-D). Furthermore, no effect on freezing levels were detected in an independent cohort of mice that was trained in the conditioning box but placed in a novel context 5 days later (Figure 7A,C), demonstrating that the enhanced fear memory was context-specific and time-dependent.
      Figure thumbnail gr7
      Figure 7Enhanced expression of recent contextual fear memory and c-Fos+ cells in Ezr-saCas9 mice. A) Experimental workflow. Mice were injected with CNT and Ezr-saCas9 AAVs in the CA1, and 5 weeks later fear conditioning or novel context was performed. Mice were sacrificed 90 minutes after retrieval. B) Freezing levels during retrieval. Data pooled from two independent cohorts of mice color coded in black and gray (Control: 13 mice and Ezr-saCas9: 15 mice). Unpaired t test: t26 = 3.16, ** p = 0.004. C) Freezing levels during novel context (Control and Ezr-saCas9: 8 mice). Unpaired t test: t14 = 0.96, p = 0.35). D) Quantification of c-Fos+ cells in the CA1 pyramidal layer during recent retrieval (Control: 8 slices per mouse from 6 mice, Ezr-saCas9: 8 slices per mouse from 7 mice). Samples correspond to the cohort of mice highlighted in black in panel B. Unpaired t test: t11 = 2.27, * p = 0.04. E) Representative examples of activated neurons (c-Fos+ cells) in the CA1 during recent retrieval. F) Quantification of c-Fos+ cells in the CA1 pyramidal layer during novel context exposure (Control: 8 slices per mouse from 8 mice, Ezr-saCas9: 8 slices per mouse from 8 mice). Unpaired t test: t14 = 0.52, p = 0.6. G) Representative examples of activated neurons (c-Fos+ cells in the CA1 during novel context. Scale bar: 50 μm. Data are presented as mean ± SEM.
      Next, we tested whether memory enhancement was accompanied by increased neuronal activity in a context-dependent manner. For this, we quantified the number of neurons that expressed c-Fos (an established marker for neuronal activation (
      • Cruz F.C.
      • Javier Rubio F.
      • Hope B.T.
      Using c-fos to study neuronal ensembles in corticostriatal circuitry of addiction.
      )) in CA1 of mice that underwent retrieval in the conditioning context, mice that were placed in novel context during retrieval, and naïve home-cage (HC) controls. In line with the enhanced freezing levels observed, Ezr-saCas9 mice had a greater number of c-Fos+ neurons after recent retrieval when compared to controls (Figure 7D,E). Notably, this increased neuronal activation was context- and memory-specific, as no differences were observed between groups after retrieval in a novel context (Figure 7F,G), nor in naïve HC controls (Figure S9E,F). Together, these data demonstrate that interaction between Ezrin-dependent astrocyte leaflets and synapses in the hippocampus influences recent memory retrieval and gate neuronal activation in a context-dependent manner.

      Discussion

      Our data reveal that processing of contextual fear memory induces structural remodeling of astrocyte leaflets in the hippocampus, and that the proximity of these specialized structures to the synaptic cleft determines the strength of recent memory expression. In line with the critical role of astrocyte leaflets in fear memory processing, we also show that genetic depletion of Ezrin enhanced glutamate spillover, NMDAR activation and recent memory expression.
      Previous studies using acute hippocampal slices have proposed that synaptic potentiation induces a robust retraction of astroglial processes from hippocampal synapses within the timeframe of 10-90 minutes after stimulation (
      • Henneberger C.
      • Bard L.
      • Panatier A.
      • Reynolds J.P.
      • Kopach O.
      • Medvedev N.I.
      • et al.
      LTP Induction Boosts Glutamate Spillover by Driving Withdrawal of Perisynaptic Astroglia.
      ,
      • Perez-Alvarez A.
      • Navarrete M.
      • Covelo A.
      • Martin E.D.
      • Araque A.
      Structural and Functional Plasticity of Astrocyte Processes and Dendritic Spine Interactions.
      ). We now demonstrate in vivo that astrocyte leaflet retraction occurs rapidly (30 min) and is maintained within the first 5 days after contextual fear conditioning. Notably, this retraction is no longer present 4 weeks after learning. It should be noted that Mazaré et al., (2020) reported unchanged leaflet-to-synapse distance after recent retrieval, however, an important difference with our study is that we did not include a retrieval session. In addition, hyper-thin structures previously observed with EM might remain elusive with super-resolution microscopy (
      • Mazaré N.
      • Oudart M.
      • Moulard J.
      • Cheung G.
      • Tortuyaux R.
      • Mailly P.
      • et al.
      Local Translation in Perisynaptic Astrocytic Processes Is Specific and Changes after Fear Conditioning.
      ,
      • Arizono M.
      • Inavalli V.V.G.K.
      • Panatier A.
      • Pfeiffer T.
      • Angibaud J.
      • Levet F.
      • et al.
      Structural basis of astrocytic Ca2+ signals at tripartite synapses.
      ,
      • Vicidomini G.
      • Bianchini P.
      • Diaspro A.
      STED super-resolved microscopy.
      ,
      • Heller J.P.
      • Rusakov D.A.
      The Nanoworld of the Tripartite Synapse: Insights from Super-Resolution Microscopy.
      ). Our data are in line with the theory that recent memories (<1-week-old) rely on orchestrated activity in the hippocampus (
      • Liu X.
      • Ramirez S.
      • Pang P.T.
      • Puryear C.B.
      • Govindarajan A.
      • Deisseroth K.
      • Tonegawa S.
      Optogenetic stimulation of a hippocampal engram activates fear memory recall.
      ,
      • Denny C.A.
      • Kheirbek M.A.
      • Alba E.L.
      • Tanaka K.F.
      • Brachman R.A.
      • Laughman K.B.
      • et al.
      Hippocampal Memory Traces Are Differentially Modulated by Experience, Time, and Adult Neurogenesis.
      ), whereas remote memories (≥1-month-old) depend on network-wide changes involving cortical regions (
      • Frankland P.W.
      • Bontempi B.
      The organization of recent and remote memories.
      ,
      • Matos M.R.
      • Visser E.
      • Kramvis I.
      • van der Loo R.J.
      • Gebuis T.
      • Zalm R.
      • et al.
      Memory strength gates the involvement of a CREB-dependent cortical fear engram in remote memory.
      ,
      • Kitamura T.
      • Ogawa S.K.
      • Roy D.S.
      • Okuyama T.
      • Morrissey M.D.
      • Smith L.M.
      • et al.
      Engrams and circuits crucial for systems consolidation of a memory.
      ). Whether astrocyte leaflets also retract from inhibitory synapses during fear memory consolidation thereby contributing to GABA spillover remains to be determined. The cause and functional relevance of the increased spine size after 4 weeks of learning are unclear and an important topic for future research.
      Recent studies have demonstrated a crucial role for astrocyte-neuron signaling in shaping neuronal circuits and behavior (
      • Adamsky A.
      • Kol A.
      • Kreisel T.
      • Doron A.
      • Ozeri-Engelhard N.
      • Melcer T.
      • et al.
      Astrocytic Activation Generates De Novo Neuronal Potentiation and Memory Enhancement.
      ,
      • Suzuki A.
      • Stern S.A.
      • Bozdagi O.
      • Huntley G.W.
      • Walker R.H.
      • Magistretti P.J.
      • Alberini C.M.
      Astrocyte-Neuron Lactate Transport Is Required for Long-Term Memory Formation.
      ,
      • Gao V.
      • Suzuki A.
      • Magistretti P.J.
      • Lengacher S.
      • Pollonini G.
      • Steinman M.Q.
      • Alberini C.M.
      Astrocytic β 2 -adrenergic receptors mediate hippocampal long-term memory consolidation.
      ,
      • Kol A.
      • Adamsky A.
      • Groysman M.
      • Kreisel T.
      • London M.
      • Goshen I.
      Astrocytes contribute to remote memory formation by modulating hippocampal–cortical communication during learning.
      ,
      • Martin-Fernandez M.
      • Jamison S.
      • Robin L.M.
      • Zhao Z.
      • Martin E.D.
      • Aguilar J.
      • et al.
      Synapse-specific astrocyte gating of amygdala-related behavior.
      ,
      • Nagai J.
      • Rajbhandari A.K.
      • Gangwani M.R.
      • Hachisuka A.
      • Coppola G.
      • Masmanidis S.C.
      • et al.
      Hyperactivity with Disrupted Attention by Activation of an Astrocyte Synaptogenic Cue.
      ,
      • Papouin T.
      • Dunphy J.M.
      • Tolman M.
      • Dineley K.T.
      • Haydon P.G.
      Septal Cholinergic Neuromodulation Tunes the Astrocyte-Dependent Gating of Hippocampal NMDA Receptors to Wakefulness.
      ,
      • Robin L.M.
      • Oliveira da Cruz J.F.
      • Langlais V.C.
      • Martin-Fernandez M.
      • Metna-Laurent M.
      • Busquets-Garcia A.
      • et al.
      Astroglial CB1 Receptors Determine Synaptic D-Serine Availability to Enable Recognition Memory.
      ,
      • Mederos S.
      • Sánchez-Puelles C.
      • Esparza J.
      • Valero M.
      • Ponomarenko A.
      • Perea G.
      GABAergic signaling to astrocytes in the prefrontal cortex sustains goal-directed behaviors.
      ), and much of this work is based on the assumption that astrocytes processes are located in close proximity to synapses. However, how this proximity, which we show is dynamic (Figure 1), affects cognitive performance has remained largely unexplored (
      • Zhou B.
      • Chen L.
      • Liao P.
      • Huang L.
      • Chen Z.
      • Liao D.
      • et al.
      Astroglial dysfunctions drive aberrant synaptogenesis and social behavioral deficits in mice with neonatal exposure to lengthy general anesthesia.
      ,
      • Pannasch U.
      • Freche D.
      • Dallérac G.
      • Ghézali G.
      • Escartin C.
      • Ezan P.
      • et al.
      Connexin 30 sets synaptic strength by controlling astroglial synapse invasion.
      ). We employed CRISPR/Cas9 gene editing to manipulate astrocyte processes by targeting Ezrin. It should be noted, however, that the Ezr-saCas9 vector did not lead to a complete absence of astrocytic Ezrin, therefore, some of the effects caused by Ezrin depletion are likely to be an underestimation. We show that in vivo reduction of Ezrin in adult hippocampal astrocytes affects astrocyte morphogenesis and increases the distance between the astrocyte leaflet and synaptic cleft. This implies that astrocytic Ezrin could link membrane-bound proteins with actin filaments, and may thereby be essential in the formation and motility of the fine glial processes upon specific stimuli (
      • Derouiche A.
      • Geiger K.D.
      Perspectives for Ezrin and Radixin in Astrocytes: Kinases, Functions and Pathology.
      ). This mechanism has been suggested for other actin binding proteins, such as Cofilin-1, that regulates leaflet shrinkage through NKCC1 (Na+, K+, 2Cl- cotransporter) activation (
      • Henneberger C.
      • Bard L.
      • Panatier A.
      • Reynolds J.P.
      • Kopach O.
      • Medvedev N.I.
      • et al.
      LTP Induction Boosts Glutamate Spillover by Driving Withdrawal of Perisynaptic Astroglia.
      ), and Profilin-1, that regulates leaflet outgrowth upon Ca2+ elevation (
      • Molotkov D.
      • Zobova S.
      • Arcas J.M.
      • Khiroug L.
      Calcium-induced outgrowth of astrocytic peripheral processes requires actin binding by Profilin-1.
      ). Based on previous studies performed in epithelial cells, Ezrin undergoes a C-terminal phosphorylation/dephosphorylation cycle that regulates its apical membrane localization and that is necessary for microvilli formation (
      • Viswanatha R.
      • Ohouo P.Y.
      • Smolka M.B.
      • Bretscher A.
      Local phosphocycling mediated by LOK/SLK restricts ezrin function to the apical aspect of epithelial cells.
      ,
      • Viswanatha R.
      • Bretscher A.
      • Garbett D.
      Dynamics of ezrin and EBP50 in regulating microvilli on the apical aspect of epithelial cells.
      ). Therefore, it would be of particular interest to determine whether a phosphorylation/dephosphorylation cycle drives Ezrin-dependent leaflet structural plasticity upon neuronal activity and during the processing of aversive memories.
      Astrocytes are intimately associated with synapses both structurally and functionally (
      • Araque A.
      • Parpura V.
      • Sanzgiri R.P.
      • Haydon P.G.
      Tripartite synapses: glia, the unacknowledged partner.
      ,
      • Perez-Alvarez A.
      • Navarrete M.
      • Covelo A.
      • Martin E.D.
      • Araque A.
      Structural and Functional Plasticity of Astrocyte Processes and Dendritic Spine Interactions.
      ,
      • Arizono M.
      • Inavalli V.V.G.K.
      • Panatier A.
      • Pfeiffer T.
      • Angibaud J.
      • Levet F.
      • et al.
      Structural basis of astrocytic Ca2+ signals at tripartite synapses.
      ,
      • Santello M.
      • Toni N.
      • Volterra A.
      Astrocyte function from information processing to cognition and cognitive impairment.
      ,
      • Perea G.
      • Navarrete M.
      • Araque A.
      Tripartite synapses: astrocytes process and control synaptic information.
      ,
      • Theodosis D.T.
      • Poulain D.A.
      • Oliet S.H.R.
      Activity-Dependent Structural and Functional Plasticity of Astrocyte-Neuron Interactions.
      ). Considering that our manipulation affected the spatial organization of leaflets at synapses, we examined excitatory synaptic transmission. First, we found that the lack of astrocytic Ezrin had no effect on basal synaptic transmission. In line with this, reduction of astrocyte territory and astrocyte-synapse interaction by deletion of the astrocyte transcription factor NFIAA in the hippocampus had no effect on basal synaptic function (
      • Huang A.Y.-S.
      • Woo J.
      • Sardar D.
      • Lozzi B.
      • Bosquez Huerta N.A.
      • Lin C.-C.J.
      • et al.
      Region-Specific Transcriptional Control of Astrocyte Function Oversees Local Circuit Activities.
      ). In fact, Ezrin-dependent astrocyte leaflet-synapse interaction becomes apparent only in response to strong activation of local synapses; e.g. incremental burst stimulation that leads to NMDARs activation (Figure 6) or learning-induced synaptic activation (Figure 7). Thus, the increased NMDAR-mediated responses we observe during fear memory consolidation and in Ezrin-saCas9 mice are likely a consequence of increased glutamate spillover during repetitive synaptic stimulation (

      Kullmann DM, Min M-Y, Asztely F, Rusakov DA (1999): Extracellular glutamate diffusion determines the occupancy of glutamate receptors at CA1 synapses in the hippocampus ((F. Clementi, R. Fesce, J. Meldolesi, & F. Valtorta, editors)). Phil Trans R Soc Lond B 354: 395–402.

      ,
      • Lozovaya N.A.
      • Kopanitsa M.V.
      • Boychuk Y.A.
      • Krishtal O.A.
      Enhancement of glutamate release uncovers spillover-mediated transmission by N-methyl-d-aspartate receptors in the rat hippocampus.
      ). This is consistent with previous findings in hippocampal acute slices, wherein LTP-induced leaflet withdrawal enhanced the activation of extrasynaptic NMDARs and facilitated NMDAR-mediated cross-talk among neighboring synapses (
      • Henneberger C.
      • Bard L.
      • Panatier A.
      • Reynolds J.P.
      • Kopach O.
      • Medvedev N.I.
      • et al.
      LTP Induction Boosts Glutamate Spillover by Driving Withdrawal of Perisynaptic Astroglia.
      ). It should be noted, however, that increase glutamate spillover has been found to activate presynaptic and postsynaptic (extrasynaptic) mGluRs, thereby regulating synaptic strength and neuronal excitability (
      • Scanziani M.
      • Salin P.A.
      • Vogt K.E.
      • Malenka R.C.
      • Nicoll R.A.
      Use-dependent increases in glutamate concentration activate presynaptic metabotropic glutamate receptors.
      ,
      • Mitchell S.J.
      • Silver R.A.
      Glutamate spillover suppresses inhibition by activating presynaptic mGluRs.
      ,
      • Ireland D.R.
      • Abraham W.C.
      Group I mGluRs Increase Excitability of Hippocampal CA1 Pyramidal Neurons by a PLC-Independent Mechanism.
      ). Nevertheless, our observation that PPR is not affected in Ezr-saCas9 mice suggests that presynaptic mGluRs on glutamatergic terminals are likely not involved (
      • Oliet S.H.R.
      Control of Glutamate Clearance and Synaptic Efficacy by Glial Coverage of Neurons.
      ). Furthermore, we show that the increased glutamate spillover and NMDAR activation in the Ezr-saCas9 mice likely reflects an increased diffusion in the extracellular space (
      • Piet R.
      • Vargová L.
      • Syková E.
      • Poulain D.A.
      • Oliet S.H.R.
      Physiological contribution of the astrocytic environment of neurons to intersynaptic crosstalk.
      ) as glutamate clearance does not seem to be impaired. Taken together, our data reveals that while a decrease in synaptic coverage does not affect basal synaptic transmission, it is sufficient to boost glutamate spillover and (extra)synaptic NMDAR activation under conditions of robust neuronal activation.
      Numerous studies provided evidence that altered communication between astrocytes and neurons facilitates the development of psychiatric disorders (
      • Leuner B.
      • Falduto J.
      • Shors T.J.
      Associative Memory Formation Increases the Observation of Dendritic Spines in the Hippocampus.
      ,
      • Geinisman Y.
      • Berry R.W.
      • Disterhoft J.F.
      • Power J.M.
      • Van der Zee E.A.
      Associative Learning Elicits the Formation of Multiple-Synapse Boutons.
      ,
      • Siemsen B.M.
      • Reichel C.M.
      • Leong K.C.
      • Garcia-Keller C.
      • Gipson C.D.
      • Spencer S.
      • et al.
      Effects of Methamphetamine Self-Administration and Extinction on Astrocyte Structure and Function in the Nucleus Accumbens Core.
      ,
      • Kruyer A.
      • Scofield M.D.
      • Wood D.
      • Reissner K.J.
      • Kalivas P.W.
      Heroin Cue–Evoked Astrocytic Structural Plasticity at Nucleus Accumbens Synapses Inhibits Heroin Seeking.
      ,
      • Nagai J.
      • Rajbhandari A.K.
      • Gangwani M.R.
      • Hachisuka A.
      • Coppola G.
      • Masmanidis S.C.
      • et al.
      Hyperactivity with Disrupted Attention by Activation of an Astrocyte Synaptogenic Cue.
      ,
      • Parpura V.
      • Heneka M.T.
      • Montana V.
      • Oliet S.H.R.
      • Schousboe A.
      • Haydon P.G.
      • et al.
      Glial cells in (patho)physiology: Glial cells in (patho)physiology.
      ). Yet, it remains poorly understood whether changes in the proximity of the astrocyte leaflet to the synapse underlie such maladaptive responses. Nonetheless, Pannasch et al (2014) showed that deletion of Connexin 30, a gap-junction protein, causes astroglial penetration of the synaptic cleft and subsequent impairment of synaptic strength and contextual fear memory retrieval (
      • Pannasch U.
      • Freche D.
      • Dallérac G.
      • Ghézali G.
      • Escartin C.
      • Ezan P.
      • et al.
      Connexin 30 sets synaptic strength by controlling astroglial synapse invasion.
      ). However, leaflet insertion into the cleft has so far not been observed under physiological and/or pathological conditions, nor in the conditions tested in the present study. Conversely, we demonstrate that when the astrocyte leaflet is further away from the synaptic cleft, mice exhibit increased glutamate spillover and enhanced fear memory expression. Interestingly, we observe that this increased recall is time and context specific, which is in line with previous studies demonstrating the importance of the hippocampus in encoding the contextual component of recent memories (
      • Liu X.
      • Ramirez S.
      • Pang P.T.
      • Puryear C.B.
      • Govindarajan A.
      • Deisseroth K.
      • Tonegawa S.
      Optogenetic stimulation of a hippocampal engram activates fear memory recall.
      ,
      • Frankland P.W.
      • Bontempi B.
      The organization of recent and remote memories.
      ). Furthermore, Ezr-saCas9 mice showed an increase in retrieval-evoked neuronal activity, suggesting that astrocytes show a tailored response to the activity of their surrounding neurons (
      • Adamsky A.
      • Kol A.
      • Kreisel T.
      • Doron A.
      • Ozeri-Engelhard N.
      • Melcer T.
      • et al.
      Astrocytic Activation Generates De Novo Neuronal Potentiation and Memory Enhancement.
      ). Accordingly, we think that learning-induced increased in glutamate spillover, due to reduced astroglial synaptic coverage, might promote the engagement of synapses that are in close proximity, and consequently, enhance synaptic connectivity to strengthen memory consolidation and/or retrieval (
      • Govindarajan A.
      • Kelleher R.J.
      • Tonegawa S.
      A clustered plasticity model of long-term memory engrams.
      ). Whether astrocyte leaflet retraction is a process that also occurs upon the formation and consolidation of neutral and/or positive memories, is an important unanswered question.
      To conclude, we discovered that fear learning induces transient retraction of astrocyte leaflets during memory formation and consolidation, and experimentally determined that the proximity of astrocyte leaflets to the synaptic cleft enhances NMDAR activation and the expression of contextual fear memory. To our knowledge, this study is the first to show that a selective and physiologically relevant manipulation of the structure of astrocyte leaflets leads to enhanced contextual fear memory retrieval. Our data support the proposition that the proximity of astrocyte leaflets with synapses in the adult brain has an important role in shaping the plasticity mechanisms that underlie the processing of aversive events.

      Acknowledgments

      A.B.S received funding from the EU-FP7-PEOPLE progam (CognitionNET; grant 607509). M.S.J. from ZonMW Memorable (grant 7330508160). The authors would like to thank Yvonne Gouwenberg and Robbert Zalm (VU University, Amsterdam, The Netherlands) for AAV vector construct preparation, Rolinka van der Loo and Joke Wortel (VU University, Amsterdam, The Netherlands) for helping with animal perfusions, J Christopher Octeau for all the technical support during Napa experiments, and Priyanka Rao-Ruiz (VU University, Amsterdam, The Netherlands) for the critical revision of the manuscript. An earlier version of the current manuscript has been previously published in BioRxiv (doi.org/10.1101/2022.01.30.478393).

      References

        • Izquierdo I.
        • Furini C.R.G.
        • Myskiw J.C.
        Fear Memory.
        Physiological Reviews. 2016; 96: 695-750
        • Knight D.C.
        • Smith C.N.
        • Cheng D.T.
        • Stein E.A.
        • Helmstetter F.J.
        Amygdala and hippocampal activity during acquisition and extinction of human fear conditioning.
        Cognitive, Affective, & Behavioral Neuroscience. 2004; 4: 317-325
        • Goosens K.A.
        Hippocampal regulation of aversive memories.
        Current Opinion in Neurobiology. 2011; 21: 460-466
        • Duvarci S.
        • Pare D.
        Amygdala Microcircuits Controlling Learned Fear.
        Neuron. 2014; 82: 966-980
        • Rao-Ruiz P.
        • Couey J.J.
        • Marcelo I.M.
        • Bouwkamp C.G.
        • Slump D.E.
        • Matos M.R.
        • et al.
        Engram-specific transcriptome profiling of contextual memory consolidation.
        Nat Commun. 2019; 10: 2232
        • Liu X.
        • Ramirez S.
        • Pang P.T.
        • Puryear C.B.
        • Govindarajan A.
        • Deisseroth K.
        • Tonegawa S.
        Optogenetic stimulation of a hippocampal engram activates fear memory recall.
        Nature. 2012; 484: 381-385
        • Tayler K.K.
        • Tanaka K.Z.
        • Reijmers L.G.
        • Wiltgen B.J.
        Reactivation of Neural Ensembles during the Retrieval of Recent and Remote Memory.
        Current Biology. 2013; 23: 99-106
        • Leuner B.
        • Falduto J.
        • Shors T.J.
        Associative Memory Formation Increases the Observation of Dendritic Spines in the Hippocampus.
        J Neurosci. 2003; 23: 659-665
        • Geinisman Y.
        • Berry R.W.
        • Disterhoft J.F.
        • Power J.M.
        • Van der Zee E.A.
        Associative Learning Elicits the Formation of Multiple-Synapse Boutons.
        J Neurosci. 2001; 21: 5568-5573
        • Rao-Ruiz P.
        • Visser E.
        • Mitrić M.
        • Smit A.B.
        • van den Oever M.C.
        A Synaptic Framework for the Persistence of Memory Engrams.
        Front Synaptic Neurosci. 2021; 13661476
        • Govindarajan A.
        • Kelleher R.J.
        • Tonegawa S.
        A clustered plasticity model of long-term memory engrams.
        Nat Rev Neurosci. 2006; 7: 575-583
        • Adamsky A.
        • Kol A.
        • Kreisel T.
        • Doron A.
        • Ozeri-Engelhard N.
        • Melcer T.
        • et al.
        Astrocytic Activation Generates De Novo Neuronal Potentiation and Memory Enhancement.
        Cell. 2018; 174 (e14): 59-71
        • Kol A.
        • Adamsky A.
        • Groysman M.
        • Kreisel T.
        • London M.
        • Goshen I.
        Astrocytes contribute to remote memory formation by modulating hippocampal–cortical communication during learning.
        Nat Neurosci. 2020; 23: 1229-1239
        • Li Y.
        • Li L.
        • Wu J.
        • Zhu Z.
        • Feng X.
        • Qin L.
        • et al.
        Activation of astrocytes in hippocampus decreases fear memory through adenosine A1 receptors.
        eLife. 2020; 9e57155
        • Suzuki A.
        • Stern S.A.
        • Bozdagi O.
        • Huntley G.W.
        • Walker R.H.
        • Magistretti P.J.
        • Alberini C.M.
        Astrocyte-Neuron Lactate Transport Is Required for Long-Term Memory Formation.
        Cell. 2011; 144: 810-823
        • Gao V.
        • Suzuki A.
        • Magistretti P.J.
        • Lengacher S.
        • Pollonini G.
        • Steinman M.Q.
        • Alberini C.M.
        Astrocytic β 2 -adrenergic receptors mediate hippocampal long-term memory consolidation.
        Proc Natl Acad Sci USA. 2016; 113: 8526-8531
        • Khakh B.S.
        • Sofroniew M.V.
        Diversity of astrocyte functions and phenotypes in neural circuits.
        Nat Neurosci. 2015; 18: 942-952
        • Araque A.
        • Carmignoto G.
        • Haydon P.G.
        • Oliet S.H.R.
        • Robitaille R.
        • Volterra A.
        Gliotransmitters Travel in Time and Space.
        Neuron. 2014; 81: 728-739
        • Semyanov A.
        • Verkhratsky A.
        Astrocytic processes: from tripartite synapses to the active milieu.
        Trends in Neurosciences S0166223621001430. 2021;
        • Ghézali G.
        • Dallérac G.
        • Rouach N.
        Perisynaptic astroglial processes: dynamic processors of neuronal information.
        Brain Struct Funct. 2016; 221: 2427-2442
        • Bazargani N.
        • Attwell D.
        Astrocyte calcium signaling: the third wave.
        Nat Neurosci. 2016; 19: 182-189
        • Araque A.
        • Parpura V.
        • Sanzgiri R.P.
        • Haydon P.G.
        Tripartite synapses: glia, the unacknowledged partner.
        Trends in Neurosciences. 1999; 22: 208-215
        • Derouiche A.
        • Frotscher M.
        Peripheral astrocyte processes: Monitoring by selective immunostaining for the actin-binding ERM proteins.
        Glia. 2001; 36: 330-341
        • Lavialle M.
        • Aumann G.
        • Anlauf E.
        • Prols F.
        • Arpin M.
        • Derouiche A.
        Structural plasticity of perisynaptic astrocyte processes involves ezrin and metabotropic glutamate receptors.
        Proceedings of the National Academy of Sciences. 2011; 108: 12915-12919
        • Zhou B.
        • Chen L.
        • Liao P.
        • Huang L.
        • Chen Z.
        • Liao D.
        • et al.
        Astroglial dysfunctions drive aberrant synaptogenesis and social behavioral deficits in mice with neonatal exposure to lengthy general anesthesia.
        PLoS Biol. 2019; 17e3000086
        • Oliet S.H.R.
        Control of Glutamate Clearance and Synaptic Efficacy by Glial Coverage of Neurons.
        Science. 2001; 292: 923-926
        • Panatier A.
        • Oliet S.H.R.
        Neuron–glia interactions in the hypothalamus.
        Neuron Glia Biol. 2006; 2: 51-58
        • Becquet D.
        • Girardet C.
        • Guillaumond F.
        • François-Bellan A.-M.
        • Bosler O.
        Ultrastructural plasticity in the rat suprachiasmatic nucleus. Possible involvement in clock entrainment.
        Glia. 2008; 56: 294-305
        • Scofield M.D.
        • Li H.
        • Siemsen B.M.
        • Healey K.L.
        • Tran P.K.
        • Woronoff N.
        • et al.
        Cocaine Self-Administration and Extinction Leads to Reduced Glial Fibrillary Acidic Protein Expression and Morphometric Features of Astrocytes in the Nucleus Accumbens Core.
        Biological Psychiatry. 2016; 80: 207-215
        • Siemsen B.M.
        • Reichel C.M.
        • Leong K.C.
        • Garcia-Keller C.
        • Gipson C.D.
        • Spencer S.
        • et al.
        Effects of Methamphetamine Self-Administration and Extinction on Astrocyte Structure and Function in the Nucleus Accumbens Core.
        Neuroscience. 2019; 406: 528-541
        • Kruyer A.
        • Scofield M.D.
        • Wood D.
        • Reissner K.J.
        • Kalivas P.W.
        Heroin Cue–Evoked Astrocytic Structural Plasticity at Nucleus Accumbens Synapses Inhibits Heroin Seeking.
        Biological Psychiatry. 2019; 86: 811-819
        • Ostroff L.E.
        • Manzur M.K.
        • Cain C.K.
        • Ledoux J.E.
        Synapses lacking astrocyte appear in the amygdala during consolidation of pavlovian threat conditioning: Astrocyte-free synapses increase with learning.
        J Comp Neurol. 2014; 522: 2152-2163
        • Henneberger C.
        • Bard L.
        • Panatier A.
        • Reynolds J.P.
        • Kopach O.
        • Medvedev N.I.
        • et al.
        LTP Induction Boosts Glutamate Spillover by Driving Withdrawal of Perisynaptic Astroglia.
        Neuron. 2020; 108 (e11): 919-936
        • Perez-Alvarez A.
        • Navarrete M.
        • Covelo A.
        • Martin E.D.
        • Araque A.
        Structural and Functional Plasticity of Astrocyte Processes and Dendritic Spine Interactions.
        Journal of Neuroscience. 2014; 34: 12738-12744
        • Bernardinelli Y.
        • Randall J.
        • Janett E.
        • Nikonenko I.
        • König S.
        • Jones E.V.
        • et al.
        Activity-Dependent Structural Plasticity of Perisynaptic Astrocytic Domains Promotes Excitatory Synapse Stability.
        Current Biology. 2014; 24: 1679-1688
        • Lushnikova I.
        • Skibo G.
        • Muller D.
        • Nikonenko I.
        Synaptic potentiation induces increased glial coverage of excitatory synapses in CA1 hippocampus.
        Hippocampus. 2009; 19: 753-762
        • Wenzel J.
        • Lammert G.
        • Meyer U.
        • Krug M.
        The influence of long-term potentiation on the spatial relationship between astrocyte processes and potentiated synapses in the dentate gyrus neuropil of rat brain.
        Brain Research. 1991; 560: 122-131
        • Badia‐Soteras A.
        • Octeau J.C.
        • Verheijen M.H.G.
        • Khakh B.S.
        Assessing Neuron–Astrocyte Spatial Interactions Using the Neuron–Astrocyte Proximity Assay.
        Current Protocols in Neuroscience. 2020; 91https://doi.org/10.1002/cpns.91
        • Herde M.K.
        • Bohmbach K.
        • Domingos C.
        • Vana N.
        • Komorowska-Müller J.A.
        • Passlick S.
        • et al.
        Local Efficacy of Glutamate Uptake Decreases with Synapse Size.
        Cell Reports. 2020; 32108182
        • Denny C.A.
        • Kheirbek M.A.
        • Alba E.L.
        • Tanaka K.F.
        • Brachman R.A.
        • Laughman K.B.
        • et al.
        Hippocampal Memory Traces Are Differentially Modulated by Experience, Time, and Adult Neurogenesis.
        Neuron. 2014; 83: 189-201
        • Ortinski P.I.
        • Dong J.
        • Mungenast A.
        • Yue C.
        • Takano H.
        • Watson D.J.
        • et al.
        Selective induction of astrocytic gliosis generates deficits in neuronal inhibition.
        Nat Neurosci. 2010; 13: 584-591
        • Octeau J.C.
        • Chai H.
        • Jiang R.
        • Bonanno S.L.
        • Martin K.C.
        • Khakh B.S.
        An Optical Neuron-Astrocyte Proximity Assay at Synaptic Distance Scales.
        Neuron. 2018; 98 (e9): 49-66
        • Liu G.
        • Choi S.
        • Tsien R.W.
        Variability of Neurotransmitter Concentration and Nonsaturation of Postsynaptic AMPA Receptors at Synapses in Hippocampal Cultures and Slices.
        Neuron. 1999; 22: 395-409
        • Marvin J.S.
        • Borghuis B.G.
        • Tian L.
        • Cichon J.
        • Harnett M.T.
        • Akerboom J.
        • et al.
        An optimized fluorescent probe for visualizing glutamate neurotransmission.
        Nat Methods. 2013; 10: 162-170
        • Kullmann D.M.
        • Asztely F.
        Extrasynaptic glutamate spillover in the hippocampus: evidence and implications.
        Trends in Neurosciences. 1998; 21: 8-14
        • Lozovaya N.A.
        • Grebenyuk S.E.
        • TSh Tsintsadze
        • Feng B.
        • Monaghan D.T.
        • Krishtal O.A.
        Extrasynaptic NR2B and NR2D subunits of NMDA receptors shape ‘superslow’ afterburst EPSC in rat hippocampus: NMDA receptor subunits in hippocampus.
        The Journal of Physiology. 2004; 558: 451-463
        • Arnth-Jensen N.
        • Jabaudon D.
        • Scanziani M.
        Cooperation between independent hippocampal synapses is controlled by glutamate uptake.
        Nat Neurosci. 2002; 5: 325-331
        • Cruz F.C.
        • Javier Rubio F.
        • Hope B.T.
        Using c-fos to study neuronal ensembles in corticostriatal circuitry of addiction.
        Brain Research. 2015; 1628: 157-173
        • Mazaré N.
        • Oudart M.
        • Moulard J.
        • Cheung G.
        • Tortuyaux R.
        • Mailly P.
        • et al.
        Local Translation in Perisynaptic Astrocytic Processes Is Specific and Changes after Fear Conditioning.
        Cell Reports. 2020; 32108076
        • Arizono M.
        • Inavalli V.V.G.K.
        • Panatier A.
        • Pfeiffer T.
        • Angibaud J.
        • Levet F.
        • et al.
        Structural basis of astrocytic Ca2+ signals at tripartite synapses.
        Nat Commun. 2020; 11: 1906
        • Vicidomini G.
        • Bianchini P.
        • Diaspro A.
        STED super-resolved microscopy.
        Nat Methods. 2018; 15: 173-182
        • Heller J.P.
        • Rusakov D.A.
        The Nanoworld of the Tripartite Synapse: Insights from Super-Resolution Microscopy.
        Front Cell Neurosci. 2017; 11: 374
        • Frankland P.W.
        • Bontempi B.
        The organization of recent and remote memories.
        Nat Rev Neurosci. 2005; 6: 119-130
        • Matos M.R.
        • Visser E.
        • Kramvis I.
        • van der Loo R.J.
        • Gebuis T.
        • Zalm R.
        • et al.
        Memory strength gates the involvement of a CREB-dependent cortical fear engram in remote memory.
        Nat Commun. 2019; 10: 2315
        • Kitamura T.
        • Ogawa S.K.
        • Roy D.S.
        • Okuyama T.
        • Morrissey M.D.
        • Smith L.M.
        • et al.
        Engrams and circuits crucial for systems consolidation of a memory.
        Science. 2017; 356: 73-78
        • Kol A.
        • Adamsky A.
        • Groysman M.
        • Kreisel T.
        • London M.
        • Goshen I.
        Astrocytes contribute to remote memory formation by modulating hippocampal–cortical communication during learning.
        Nat Neurosci. 2020; 23: 1229-1239
        • Martin-Fernandez M.
        • Jamison S.
        • Robin L.M.
        • Zhao Z.
        • Martin E.D.
        • Aguilar J.
        • et al.
        Synapse-specific astrocyte gating of amygdala-related behavior.
        Nat Neurosci. 2017; 20: 1540-1548
        • Nagai J.
        • Rajbhandari A.K.
        • Gangwani M.R.
        • Hachisuka A.
        • Coppola G.
        • Masmanidis S.C.
        • et al.
        Hyperactivity with Disrupted Attention by Activation of an Astrocyte Synaptogenic Cue.
        Cell. 2019; 177: 1280-1292.e20
        • Papouin T.
        • Dunphy J.M.
        • Tolman M.
        • Dineley K.T.
        • Haydon P.G.
        Septal Cholinergic Neuromodulation Tunes the Astrocyte-Dependent Gating of Hippocampal NMDA Receptors to Wakefulness.
        Neuron. 2017; 94 (e7): 840-854
        • Robin L.M.
        • Oliveira da Cruz J.F.
        • Langlais V.C.
        • Martin-Fernandez M.
        • Metna-Laurent M.
        • Busquets-Garcia A.
        • et al.
        Astroglial CB1 Receptors Determine Synaptic D-Serine Availability to Enable Recognition Memory.
        Neuron. 2018; 98 (e5): 935-944
        • Mederos S.
        • Sánchez-Puelles C.
        • Esparza J.
        • Valero M.
        • Ponomarenko A.
        • Perea G.
        GABAergic signaling to astrocytes in the prefrontal cortex sustains goal-directed behaviors.
        Nat Neurosci. 2021; 24: 82-92
        • Pannasch U.
        • Freche D.
        • Dallérac G.
        • Ghézali G.
        • Escartin C.
        • Ezan P.
        • et al.
        Connexin 30 sets synaptic strength by controlling astroglial synapse invasion.
        Nat Neurosci. 2014; 17: 549-558
        • Derouiche A.
        • Geiger K.D.
        Perspectives for Ezrin and Radixin in Astrocytes: Kinases, Functions and Pathology.
        IJMS. 2019; 20: 3776
        • Molotkov D.
        • Zobova S.
        • Arcas J.M.
        • Khiroug L.
        Calcium-induced outgrowth of astrocytic peripheral processes requires actin binding by Profilin-1.
        Cell Calcium. 2013; 53: 338-348
        • Viswanatha R.
        • Ohouo P.Y.
        • Smolka M.B.
        • Bretscher A.
        Local phosphocycling mediated by LOK/SLK restricts ezrin function to the apical aspect of epithelial cells.
        Journal of Cell Biology. 2012; 199: 969-984
        • Viswanatha R.
        • Bretscher A.
        • Garbett D.
        Dynamics of ezrin and EBP50 in regulating microvilli on the apical aspect of epithelial cells.
        Biochemical Society Transactions. 2014; 42: 189-194
        • Arizono M.
        • Inavalli V.V.G.K.
        • Panatier A.
        • Pfeiffer T.
        • Angibaud J.
        • Levet F.
        • et al.
        Structural basis of astrocytic Ca2+ signals at tripartite synapses.
        Nat Commun. 2020; 11: 1906
        • Santello M.
        • Toni N.
        • Volterra A.
        Astrocyte function from information processing to cognition and cognitive impairment.
        Nat Neurosci. 2019; 22: 154-166
        • Perea G.
        • Navarrete M.
        • Araque A.
        Tripartite synapses: astrocytes process and control synaptic information.
        Trends in Neurosciences. 2009; 32: 421-431
        • Theodosis D.T.
        • Poulain D.A.
        • Oliet S.H.R.
        Activity-Dependent Structural and Functional Plasticity of Astrocyte-Neuron Interactions.
        Physiological Reviews. 2008; 88: 983-1008
        • Huang A.Y.-S.
        • Woo J.
        • Sardar D.
        • Lozzi B.
        • Bosquez Huerta N.A.
        • Lin C.-C.J.
        • et al.
        Region-Specific Transcriptional Control of Astrocyte Function Oversees Local Circuit Activities.
        Neuron. 2020; 106 (e9): 992-1008
      1. Kullmann DM, Min M-Y, Asztely F, Rusakov DA (1999): Extracellular glutamate diffusion determines the occupancy of glutamate receptors at CA1 synapses in the hippocampus ((F. Clementi, R. Fesce, J. Meldolesi, & F. Valtorta, editors)). Phil Trans R Soc Lond B 354: 395–402.

        • Lozovaya N.A.
        • Kopanitsa M.V.
        • Boychuk Y.A.
        • Krishtal O.A.
        Enhancement of glutamate release uncovers spillover-mediated transmission by N-methyl-d-aspartate receptors in the rat hippocampus.
        Neuroscience. 1999; 91: 1321-1330
        • Scanziani M.
        • Salin P.A.
        • Vogt K.E.
        • Malenka R.C.
        • Nicoll R.A.
        Use-dependent increases in glutamate concentration activate presynaptic metabotropic glutamate receptors.
        Nature. 1997; 385: 630-634
        • Mitchell S.J.
        • Silver R.A.
        Glutamate spillover suppresses inhibition by activating presynaptic mGluRs.
        Nature. 2000; 404: 498-502
        • Ireland D.R.
        • Abraham W.C.
        Group I mGluRs Increase Excitability of Hippocampal CA1 Pyramidal Neurons by a PLC-Independent Mechanism.
        Journal of Neurophysiology. 2002; 88: 107-116
        • Piet R.
        • Vargová L.
        • Syková E.
        • Poulain D.A.
        • Oliet S.H.R.
        Physiological contribution of the astrocytic environment of neurons to intersynaptic crosstalk.
        PNAS. 2004; 101: 2151-2155
        • Parpura V.
        • Heneka M.T.
        • Montana V.
        • Oliet S.H.R.
        • Schousboe A.
        • Haydon P.G.
        • et al.
        Glial cells in (patho)physiology: Glial cells in (patho)physiology.
        Journal of Neurochemistry. 2012; 121: 4-27