No description available.

No description available.

While peripheral viral loads can be largely controlled by combination antiretroviral therapy, a significant number of people with HIV (PWH) suffer from HIV-associated neurocognitive disorders (HAND). Cocaine use is a major risk factor for becoming HIV infected, and data have shown that HIV+ cocaine user individuals have worse neurocognitive impairments than those observed from either disease alone. The HIV protein Tat and cocaine synergize to cause damage to neurons in the brain, but astrocytes in the central nervous system (CNS) are an understudied population in these individuals. Importantly, astrocytes support neurons metabolically by supplying key metabolites such as lactate and cholesterol to meet the large energy demands of neurons. Astrocyte-derived lactate is taken up by neurons to serve as a key substrate for ATP production. Additionally, the brain requires an intricate balance of cholesterol to support synaptodendritic communications. Disruption of cholesterol supply and energy deficits have been implicated in brain aging and many neurodegenerative diseases including Alzheimer’s disease. There have been reports of altered brain metabolite profiles in PWH indicative of changes in energy homeostasis. We hypothesized that cocaine and Tat work together to disrupt astrocyte energy metabolism. These changes in energy demand compromise the astrocyte’s ability to support neurons leading to neuronal dysfunction and worsening HAND. Our data show that exposure of astrocytes to cocaine and Tat significantly enhances aerobic glycolysis and decreases lactate production. Our findings also demonstrate that cocaine and Tat decrease astrocyte-derived cholesterol, pointing to a deficiency in cholesterol supply and efflux for use by neurons. These findings are supported in the Tat transgenic mouse model of chronic cocaine use. Taken together, these data uncover novel alterations in the bioenergetics pathway in astrocytes exposed to cocaine and the HIV protein, Tat. Results from these studies point to a new pathway in the CNS that may contribute to HAND in HIV+ cocaine users. / Biomedical Sciences

The blood-brain barrier (BBB) is regulated by factors that can be secreted by multiple cell types, including astrocytes, that maintain the BBB in health and promote repair after injury. However, astrocyte contributions to the BBB are largely assumed from transplantation studies in which astrocyte progenitor grafts conferred BBB-like properties to tissues that normally lack a BBB. To determine if astrocytes contribute an essential and non-redundant function in maintaining the healthy BBB, I genetically ablated a small number of astrocytes using a conditional, tamoxifen-inducible mouse model. Within 2 hours after induction, I observed sparse astrocyte death in the cortex and leakage of the small molecule Cadaverine and large plasma protein fibrinogen, which are normally contained by a functional BBB. Vessels within regions of ablated astrocytes showed reduced expression of the tight junction protein zonula occludens-1, indicating impairment of the physical barrier formed between endothelial cells. Cadaverine leakage persisted for weeks, a feature I also found in mice after mild concussive traumatic brain injury (cTBI), thus highlighting the potential for revealing astrocyte roles in post-injury repair. Unlike the genetic ablation model, astrocytes within Cadaverine leakage areas did not undergo cell death after cTBI and instead downregulated homeostatic proteins. Our preliminary results show this atypical phenotype appearing 10 minutes after cTBI, along with severe vessel rupture, BBB leakage, and disruption of endfoot and basement membrane proteins. This damage persists for months, suggesting that the BBB fails to repair in these areas. Our results provide direct in-vivo evidence for essential astrocyte roles in the maintenance of the healthy BBB. Maintenance and/or repair fail after mild concussive cTBI, possibly contributing to irreversible progression to neurodegenerative diseases. / Doctor of Philosophy / The blood-brain barrier (BBB) is a unique property of blood vessels in the Central Nervous System (CNS) different from other vessels in the body. The physically tight barrier of the BBB is formed by tight junction proteins between endothelial cells and limits paracellular diffusion. The metabolic barrier is formed by concentrations of glucose transporters that promote transport of essential nutrients to the brain. Lastly, a transport barrier limits the passage of molecules and cells across the endothelial cell layer, preventing the entry of non-essential molecules, including pathogens and immune cells found in the blood. The BBB is thought to be induced and maintained by factors secreted by nearby cells in the brain. Among these cells are astrocytes, a type of glial cell that nearly completely cover blood vessels with their processes called endfeet. This strategic positioning led the field to assume that astrocytes are responsible for generating the unique properties of the BBB. Yet little direct evidence exists to support this conclusion, and newer evidence calls into question if astrocytes are even needed for BBB functions. To test this, I used a genetic mouse model to induce death of small numbers of astrocytes in adult mice. This caused leakage of blood contents of various sizes into the brain. In addition, the tight junction proteins responsible for forming the physical BBB were disrupted. These effects remained for weeks, a feature I also found after mild concussive traumatic brain injury (cTBI). This suggests that astrocytes may have an additional function in repairing the injured BBB. Our results demonstrate an essential role for astrocytes in the maintenance of the healthy adult BBB. Maintenance and/or its repair fail after cTBI, possibly contributing to the cascade into irreversible progression to neurodegenerative diseases.

Mild traumatic brain injury (mTBI)/ concussion accounts for 70-90% of all reported TBI cases in the United States and can cause long-term neurological outcomes that negatively impact quality of life. Previous studies revealed that increased blood-brain barrier (BBB) leakage is correlated with poor neurological outcomes after mTBI, yet the biological mechanisms linking BBB damage to the onset of neurological deficits after mTBI are not well understood. Previously, we found that astrocytes lose expression of homeostatic proteins after mTBI, characterizing the changes in astrocytic protein expression as an "atypical astrocyte response." Yet, the upstream mechanisms that induce this atypical astrocyte response after mTBI have yet to be elucidated. In models of more severe TBI, exposure to blood-borne factors triggers astrogliosis via upregulation in markers, such as glial fibrillary acidic protein (GFAP), but how exposure to blood-borne factors affects astrocyte protein expression in the context of mTBI is not well understood. Therefore, we hypothesized that mTBI-induced BBB damage causes atypical astrocytes via exposure to blood-borne factors. To test this hypothesis, we use a mTBI mouse model, two-photon microscopy, an endothelial cell-specific genetic ablation model, and serum-free primary astrocyte cultures. Here, we found that mTBI causes BBB damage through the loss of proteins involved in maintaining the BBB's physical and metabolic barriers, and BBB damage is sustained long-term after injury. Also, we demonstrated that leakage of blood-borne factors is sufficient to trigger atypical astrocytes, and plasma exposure triggers a similar response in vitro. Overall, these findings suggest that mTBI induces long-term BBB damage, and exposure to blood-borne factors triggers the loss of key homeostatic astrocytic proteins involved in maintaining healthy neuronal function. / Doctor of Philosophy / Mild traumatic brain injury (mTBI)/ concussion makes up 70-90% of all TBI cases reported in the United States and is commonly observed after car crashes, sports-related tackles, and blast exposure during military combat. People who experience mTBI develop debilitating long-term neurological consequences, such as sleep disturbances, depression, and dementia. Clinical data suggests mTBI causes damage to the barrier between the brain and blood, known as the blood-brain barrier (BBB). This damage has been correlated to the onset of poor neurological deficits, yet how damage to this barrier is causally linked to long-term neurological consequences remains to be fully understood. In our lab, we found that mTBI causes loss of proteins important for maintaining a healthy environment in the brain in specialized cells called astrocytes. However, the biological events that trigger the loss of protein expression in astrocytes after mTBI have yet to be fully investigated. Thus, we hypothesized that mTBI causes loss of these proteins via leakage of blood-borne factors. To test this hypothesis, we used a mTBI mouse model, two-photon microscopy, genetic manipulation, and cell cultures. In our studies, we found that mTBI triggers BBB damage via loss of proteins that make up its protective properties. Also, we demonstrated that leakage of blood-borne factors is sufficient to cause loss of astrocyte-specific proteins both in brain and cell cultures. Altogether, we show that a single mTBI is sufficient to cause loss of astrocyte-specific protein expression via exposure to blood-borne factors. These findings may point to targeting either the blood-borne factor(s) or their corresponding receptor pathways in astrocytes to halt the progression of long-term neurological deficits after mTBI.

Post-traumatic stress disorder (PTSD) is a psychological disorder that can cause great social/economic hardship. Progress towards treating PTSD has been slow due to a lack of understanding of its pathogenesis. This dissertation aimed to address this issue by investigating the involvement of the astrocytic glutamate reuptake transporter, GLT-1, and regional differences in expression of NMDA receptor subtypes in the development of a rat model of PTSD. We hypothesized that impaired astrocytic glutamate reuptake inhibits long-term memory processes, and that concurrent presence of glucocorticoids (GCs) during situational trauma selectively inhibits fear extinction memory processes in the prefrontal cortex, but not of conditioned fear memory processes in the amygdala, due to differences between these brain regions in expression of NMDA receptor subtypes. The effect of GLT-1 manipulation was studied in vivo. Utilizing the Single Prolonged Stress (SPS) model of PTSD, rats were either exposed to SPS or not. Within these groups, rats were administered a saline sham, a GLT-1 facilitator (ceftriaxone (CEF)), or a GLT-1 inhibitor (dihydrokainic acid (DHK)). Using Classical Fear Conditioning (CFC) and Fear Extinction (EXT) paradigms, retention of fear extinction memories was measured to determine the effect of GLT-1 manipulation on SPS-induced behavior (i.e., impaired fear extinction retention). From the brain of each rat, the amygdala, hippocampus, and prefrontal cortex (PFC) were collected and expression of GLT-1, p-CREB (a molecular indicator of long-term memory), and glucocorticoid receptor (GR, a molecular indicator of a PTSD-like state) were quantified. Analysis of the behavioral data showed that SPS exposure alone reduced the retention of extinction memories, but CEF and DHK both eliminated this effect. Analysis of the brain tissues revealed that SPS induced an increase in GR expression in the hippocampus. SPS also increased GLT-1 expression, but not p-CREB, in the PFC and amygdala. To evaluate the involvement of regional differences in NMDA receptor subtype expression ex vivo, tissue sections of amygdala, hippocampus, and PFC were taken from SPS and non-SPS exposed rats. Synaptic transmission was stimulated in these tissues using bicuculline in the presence of glucocorticoids, NVP-AA077 (a NR2A NMDA receptor subtype inhibitor), or Ro-25 (a NR2B NMDA receptor subtype inhibitor). P-CREB was measured in the tissues treated with GCs to determine if GCs exert greater inhibition of long-term memory in the PFC (a region reported to express high NR2A) than in the amygdala (a region reported to express high NR2B). P-CREB was also measured in the tissues treated with NVP or Ro-25 to determine if these reported receptor profile differences could be demonstrated, and if they changed following SPS exposure. Contrary to the stated hypothesis, analysis of non-SPS exposed rats revealed that GCs, NVP, and Ro-25 decreased p-CREB in all three regions with no differences between regions. However, in the SPS exposed group, p-CREB was not decreased in PFC and hippocampal tissues treated with GCs, amygdalar and PFC tissues treated with NVP, and PFC tissue treated with Ro-25. Overall, the results of the in vivo experiment did not convincingly demonstrate a role of glutamate spill-over in the pathogenesis of PTSD, but did show that modulation of glutamate reuptake can mitigate some of the behavioral consequences of exposure to situational trauma. The results of the ex vivo experiment did not reveal evidence that regional differences in NMDA receptor profiles exist across the three regions analyzed, nor did they show that GCs exert a region specific inhibition of long-term memory formation. However, it was demonstrated that SPS may affect long-term memory by altering expression of synaptic NMDA receptors. This study provides evidence that glial cells may play a role in the pathogenesis of PTSD, and thus may serve as targets for future therapy. / Ph. D.

The advent of multi-photon microscopes and highly sensitive protein sensors enables the recording of astrocyte activities on a large population of cells over a long-time period in vivo. Existing tools cannot fully characterize these activities, both within single cells and at the population-level, because of the insufficiency of current region-of-interest-based approaches to describe the activity that is often spatially unfixed, size-varying, and propagative. Here, we present Astrocyte Quantitative Analysis (AQuA), an analytical framework that releases astrocyte biologists from the ROI-based paradigm. The framework takes an event-based perspective to model and accurately quantify the complex activity in astrocyte imaging datasets, with an event defined jointly by its spatial occupancy and temporal dynamics. To model the signal propagation in astrocyte, we developed graphical time warping (GTW) to align curves with graph-structured constraints and integrated it into AQuA. To make AQuA easy to use, we designed a comprehensive software package. The software implements the detection pipeline in an intuitive step by step GUI with visual feedback. The software also supports proof-reading and the incorporation of morphology information. With synthetic data, we showed AQuA performed much better in accuracy compared with existing methods developed for astrocytic data and neuronal data. We applied AQuA to a range of ex vivo and in vivo imaging datasets. Since AQuA is data-driven and based on machine learning principles, it can be applied across model organisms, fluorescent indicators, experimental modes, and imaging resolutions and speeds, enabling researchers to elucidate fundamental astrocyte physiology. / Doctor of Philosophy / Astrocyte is an important type of glial cell in the brain. Unlike neurons, astrocyte cannot be electrically excited. However, the concentrations of many different molecules inside and near astrocytes change over space and time and show complex patterns. Recording, analyzing, and deciphering these activity patterns enables the understanding of various roles astrocyte may play in the nervous system. Many of these important roles, such as sensory-motor integration and brain state modulation, were traditionally considered the territory of neurons, but recently found to be related to astrocytes. These activities can be monitored in the intracellular and extracellular spaces in either brain slices and living animals, thanks to the advancement of microscopes and genetically encoded fluorescent sensors. However, sophisticated analytical tools lag far behind the impressive capability of generating the data. The major reason is that existing tools are all based on the region-of-interest-based (ROI) approach. This approach assumes the field of view can be segmented to many regions, and all pixels in the region should be active together. In neuronal activity analysis, all pixels in an ROI (region of interest) correspond to a neuron and are assumed to share a common activity pattern (curve). This is not true for astrocyte activity data because astrocyte activities are spatially unfixed, size-varying, and propagative. In this dissertation, we developed a framework called AQuA to detect the activities directly. We designed an accurate and flexible detection pipeline that works with different types of astrocyte activity data sets. We designed a machine learning model to characterize the signal propagation for the pipeline. We also implemented a compressive and user-friendly software package. The advantage of AQuA is confirmed in both simulation studies and three different types of real data sets.

L’ischémie cérébrale est caractérisée par une interruption totale ou partielle de l’apport sanguine au cerveau, conduisant à une privation d’oxygène et de glucose pour les cellules du cerveau. La série de processus cellulaires qui sont déclenchées par une ischémie cérébrale sont nombreux et complexes. La réduction sévère d’oxygène et de glucose la diminution de la production d’ATP et un changement drastique de la concentration de K+, du pH intracellulaire et extracellulaire et de la production de lactate. La perturbation du métabolisme énergétique au sein des tissus ischémiés conduit rapidement à la dépolarisation membranaire et au relarguage de neurotransmetteurs dans le milieu extracellulaire. Dans le cervelet, l’impact d’un stress ischémique à largement été étudié sur les cellules de Purkinje, seule voie de sortie neuronale du cortex cérébelleux. Il a été montré que le glutamate, relargué par une surexcitation des fibres glutamatergique et par l’inversion des transporteurs du glutamate, est la cause principale de la dépolarisation anoxique des cellules de Purkinje. Cependant, la compréhension de la réponse astrocytaire et l’influence des astrocytes vis-à-vis de l’ischémie ne sont pas encore connu.La cellule de Bergmann est un astrocyte radiaire qui compose un réseau couplé électriquement, formant des interactions anatomiques et fonctionnelles complexes avec les neurones du cortex cérébelleux. En utilisant un modèle in vitro d’ischémie cérébrale, la privation d’oxygène et de glucose (OGD), plusieurs caractéristiques de base de la réaction astrocytaire à l'ischémie sont analysés. Des expériences en patch clamp et d’imagerie calcique sont réalisées sur tranche de cervelet adulte révélant la réponse de la glie de Bergmann à l’OGD par une dépolarisation progressive de la membrane, avec en parallèle une augmentation de calcium cytosolique soutenue. L’enregistrement apparié entre cellule de Purkinje et cellule de Bergmann révèle des différences importantes de réponse à l’OGD entre ces deux types cellulaires. De plus, nous avons mesuré les changements de la concentration de K+ extracellulaire durant l’OGD en utilisant des microélectrodes sensibles aux ions. Nos résultats montrent une corrélation importante entre la dynamique du K+ extracellulaire et la dépolarisation membranaire de la cellule de Bergmann au cours de l’OGD. / Cerebral ischemia is characterized by partial or total interruption of the blood supply to the brain resulting in glucose and oxygen deprivation to brain cells. The series of cellular processes that are unleashed by cerebral ischemia are complex. The severe reduction in oxygen and glucose induces decreases in ATP production and dramatic changes in extracellular K concentration, pH of intracellular and extracellular space and lactate production. The disruption of energy metabolism in the ischemic tissue rapidly lead to membrane depolarisation and neurotransmitters are released into the extracellular space. In the cerebellum, the impact of an ischemic stress has been extensively studied in Purkinje cells, the only neuronal output of the cerebellar cortex. It has been shown that glutamate released from overexcited fibers and from reversal of glutamate transporters, is the principal cause of the dramatic, anoxic depolarization in Purkinje cells. However a detailed understanding of the astrocytic response to cerebellar ischemia and the potential influence of astrocyte to ischemia outcome is still lacking.Bergmann glia (BG) are radial gial cells that form networks of electrically coupled cells underling complex anatomical and functional interactions with the neurons of the cerebellar cortex. Using an in vitro model of cerebral ischemia, the oxygen and glucose deprivation (OGD), several basic features of astrocytic reaction to ischemia are analyzed. Patch clamp and calcium imaging experiments performed in cerebellar slices from adult mice revealed that BG respond to OGD with a progressive membrane depolarisation that is paralleled with a sustained cytosolic calcium increase. Double patch-clamp recordings between Purkinje neurons and BG reveal different responses to OGD in these cell types. Furthermore, we measured extracellular potassium concentration changes during OGD by using ion-sensitive microelectrodes. Our results indicate an important correlation between the BG membrane depolarisation and the extracellular K dynamics during OGD.

Les astrocytes protoplasmiques de la matière grise corticale sont des cellules gliales dont les prolongements très fins et ramifiés sont en contact avec les éléments neuronaux pré- et post-synaptiques d’une part, et les vaisseaux sanguins d’autre part. Ils expriment plusieurs récepteurs des neurotransmetteurs, entre autres des récepteurs purinergiques dont l'activation facilite l’activité calcique astrocytaire et la libération de gliotransmitters (par exemple, le glutamate, le GABA, l'ATP, et la D sérine) qui régulent l’activité des neurones et des cellules gliales situées au voisinage. L’objectif de ma thèse était d’étudier in situ l’activité calcique des astrocytes et de leurs prolongements en réponse à l’application des agonistes purinergiques. Lors de ma thèse, j’ai tout d'abord testé la possibilité d’induire l’expression spécifique de gènes d’intérêt par les astrocytes corticaux de souris adultes par la technique de recombinaison Cre-LoxP. J’ai comparé les performances d’un virus adeno-associé de type 5 (AAV5) flexé (AAV5.FLEX.EGFP) et d’une souris qui exprime un indicateur calcique (GCaMP3) sous contrôle de la recombinase (souris Rosa-CAG-LSL-GCaMP3). L’injection d’AAV5.FLEX.EGFP dans le cortex d’une souris hGFAPcre n’a pas permis l’expression spécifique d’EGFP. La combinaison des souris exprimant le cre recombinase sous contrôle d’un promoteur sélectif des astrocytes (GLAST-CreERT2 et Cx30-CreERT2) avec le AAV5.FLEX.EGFP ou avec une lignée des souris Rosa-CAG-LSL-GCaMP3 permet l’expression spécifique des gènes d’intérêt (EGFP et GCaMP3) par les astrocytes corticaux. J’ai ensuite analysé l’activité calcique des astrocytes qui expriment GCaMP3. J’ai utilisé la microscopie biphotonique et enregistré l’activité calcique spontanée et évoquée par application d’agonistes purinergiques sur des tranches de cortex somatosensoriel primaire de souris adultes GLAST-CreERT2. L’activité calcique spontanée est complexe, généralement locale et désynchronisée, répartie dans les prolongements et la région somatique. Les régions actives ont été identifiées à partir d’une carte de corrélation temporale calculée en MATLAB, et leurs caractéristiques (amplitude, durée, position, fréquence) mesurées grâce à des routines établies sous IGOR. La fréquence et l’amplitude de l’activité calcique paraissent augmenter lors de l’enregistrement, ce qui suggère une sensibilité significative et une photoactivation des astrocytes, en imagerie biphotonique. La durée des impulsions laser modulerait ce phénomène. En présence d'adénosine (1-100 µM) et d’ATP (100 µM), et de façon marginale en présence d’un agoniste P2X7 non sélectif (BzATP 50-100 µM), une activité calcique synchronisée accrue est visible dans le soma et les prolongements astrocytaires en présence de tétrodotoxine qui bloque les potentiels d'action et minimise l’activité synaptique. Le mécanisme de ces réponses synchronisées reste à étudier. Aucun effet significatif n’a été observé en présence d’un agoniste spécifique P2Y1 (MRS2365 50 uM). Mon travail a permis le développement : i) de modèles murins pour l’adressage sélectif de protéines d’intérêt au niveau des astrocytes protoplasmiques ; ii) d’outils d’analyse des signaux calciques astrocytaires au niveau sub-cellulaire. Il a mis en évidence des limites possibles des protocoles standards d'enregistrement de l’activité calcique des astrocytes en imagerie biphotonique. Il confirme l’importance de l’ATP et de l’adénosine pour la signalisation astrocytaire. / Grey matter protoplasmic astrocytes are compact glial cells with highly branched processes, enwrapping synapses, and one or two endfeet contacting the blood vessels. Several neurotransmitter receptors are expressed by astrocytes, among them purinergic receptors. Upon activation of these receptors, intracellular calcium (Ca2+) transients can be induced, that, in turn, trigger gliotransmitter release (e.g. glutamate, GABA, ATP, D-serine) and participate in astrocyte-to-astrocyte signaling as well as in the communication between astrocytes and neurons or other glia. During my PhD work, I first implemented and validated several approaches for targeting transgene expression specifically to cortical astrocytes and employed them to study purinergic signaling in astrocytes. To achieve astrocyte-specific transgene expression, I used either floxed adeno-associated viral (AAV) vectors or a Cre-dependent mouse line and several mouse lines expressing the Cre recombinase under astrocyte-specific promoters. Intracerebral injections of a Cre-dependent AAV serotype 5 containing the ubiquitous CAG promoter and an enhanced green fluorescent protein (AAV5.CAG.flex.EGFP) in adult mice expressing Cre recombinase under the human glial fibrillary protein (hGFAP) promoter resulted in a non-astrocyte specific expression in the cortex. Combining inducible mouse lines expressing Cre recombinase under the glutamate aspartate transporter (GLAST) promoter with the same AAV vector resulted in a virtually astrocyte-specific expression of the reporter gene. As an alternative approach for astrocyte-specific transgene expression, we used a Cre-dependent mouse line expressing the genetically encoded Ca2+ indicator GCaMP3. Crossing this mouse line with the above described GLAST-CreERT2 mouse line or a Connexin30 (Cx30)-CreERT2 line led to selective GCaMP3 expression in cortical astrocytes. Second, I investigated both spontaneous and agonist-evoked Ca2+ transients in astrocytic processes, the investigation of which has presented a major challenge in earlier studies, due to the unspecific and weak labeling by membrane-permeable chemical Ca2+ indicators. Using the strategy developed in the first part of my work allowing an astrocyte-specific expression of the genetically encoded Ca2+ indicator GCaMP3. Using two-photon excitation fluorescence (2PEF) imaging in acute slices of the primary somatosensory cortex, I recorded Ca2+ transients in the astrocytic soma and processes. By aid of a custom-made MATLAB routine based on a temporal Pearson correlation coefficient, active regions could be identified in an unbiased manner. Evoked Ca2+ transients were quantified using custom IGOR routines. Spontaneous desynchronized Ca2+ transients occurred in the processes and rarely in the soma. Ca2+ signals appeared localized in distinct microdomains. Their frequency appeared to increase during long recordings of several hundred images, suggesting that fine astrocytes are vulnerable to photodamage under imaging conditions routine in 2PEF microscopy. The possibility to minimize photodamage, by varying the length of the femtosecond laser pulses is under investigation. Bath application of adenosine (1-100 µM) and adenosine-triphosphate (ATP, 100 µM), as well as the application of the non-selective P2X7 receptor agonist (2'(3')-O-(4-Benzoylbenzoyl)adenosine-5'-triphosphate, BzATP, 50-100 µM), in the presence of tetrodotoxin to block neuronal action potentials, evoked synchronized Ca2+ rises in the soma and the processes of astrocytes. The effect of adenosine was dose-dependent. No significant effect of the specific P2Y1 agonist (MRS2365, 50 µM) was seen. Altogether, my work sets up a powerful and versatile toolbox for studying astrocytic Ca2+ signaling at the sub-cellular level. It also pinpoints possible limits of standard two-photon recording protocols to investigate the local Ca2+ signals in fine astrocytic processes.

Les astrocytes protoplasmiques de la matière grise corticale sont des cellules gliales dont les prolongements très fins et ramifiés sont en contact avec les éléments neuronaux pré- et post-synaptiques d’une part, et les vaisseaux sanguins d’autre part. Ils expriment plusieurs récepteurs des neurotransmetteurs, entre autres des récepteurs purinergiques dont l'activation facilite l’activité calcique astrocytaire et la libération de gliotransmitters (par exemple, le glutamate, le GABA, l'ATP, et la D sérine) qui régulent l’activité des neurones et des cellules gliales situées au voisinage. L’objectif de ma thèse était d’étudier in situ l’activité calcique des astrocytes et de leurs prolongements en réponse à l’application des agonistes purinergiques. Lors de ma thèse, j’ai tout d'abord testé la possibilité d’induire l’expression spécifique de gènes d’intérêt par les astrocytes corticaux de souris adultes par la technique de recombinaison Cre-LoxP. J’ai comparé les performances d’un virus adeno-associé de type 5 (AAV5) flexé (AAV5.FLEX.EGFP) et d’une souris qui exprime un indicateur calcique (GCaMP3) sous contrôle de la recombinase (souris Rosa-CAG-LSL-GCaMP3). L’injection d’AAV5.FLEX.EGFP dans le cortex d’une souris hGFAPcre n’a pas permis l’expression spécifique d’EGFP. La combinaison des souris exprimant le cre recombinase sous contrôle d’un promoteur sélectif des astrocytes (GLAST-CreERT2 et Cx30-CreERT2) avec le AAV5.FLEX.EGFP ou avec une lignée des souris Rosa-CAG-LSL-GCaMP3 permet l’expression spécifique des gènes d’intérêt (EGFP et GCaMP3) par les astrocytes corticaux. J’ai ensuite analysé l’activité calcique des astrocytes qui expriment GCaMP3. J’ai utilisé la microscopie biphotonique et enregistré l’activité calcique spontanée et évoquée par application d’agonistes purinergiques sur des tranches de cortex somatosensoriel primaire de souris adultes GLAST-CreERT2. L’activité calcique spontanée est complexe, généralement locale et désynchronisée, répartie dans les prolongements et la région somatique. Les régions actives ont été identifiées à partir d’une carte de corrélation temporale calculée en MATLAB, et leurs caractéristiques (amplitude, durée, position, fréquence) mesurées grâce à des routines établies sous IGOR. La fréquence et l’amplitude de l’activité calcique paraissent augmenter lors de l’enregistrement, ce qui suggère une sensibilité significative et une photoactivation des astrocytes, en imagerie biphotonique. La durée des impulsions laser modulerait ce phénomène. En présence d'adénosine (1-100 µM) et d’ATP (100 µM), et de façon marginale en présence d’un agoniste P2X7 non sélectif (BzATP 50-100 µM), une activité calcique synchronisée accrue est visible dans le soma et les prolongements astrocytaires en présence de tétrodotoxine qui bloque les potentiels d'action et minimise l’activité synaptique. Le mécanisme de ces réponses synchronisées reste à étudier. Aucun effet significatif n’a été observé en présence d’un agoniste spécifique P2Y1 (MRS2365 50 uM). Mon travail a permis le développement : i) de modèles murins pour l’adressage sélectif de protéines d’intérêt au niveau des astrocytes protoplasmiques ; ii) d’outils d’analyse des signaux calciques astrocytaires au niveau sub-cellulaire. Il a mis en évidence des limites possibles des protocoles standards d'enregistrement de l’activité calcique des astrocytes en imagerie biphotonique. Il confirme l’importance de l’ATP et de l’adénosine pour la signalisation astrocytaire. / Grey matter protoplasmic astrocytes are compact glial cells with highly branched processes, enwrapping synapses, and one or two endfeet contacting the blood vessels. Several neurotransmitter receptors are expressed by astrocytes, among them purinergic receptors. Upon activation of these receptors, intracellular calcium (Ca2+) transients can be induced, that, in turn, trigger gliotransmitter release (e.g. glutamate, GABA, ATP, D-serine) and participate in astrocyte-to-astrocyte signaling as well as in the communication between astrocytes and neurons or other glia. During my PhD work, I first implemented and validated several approaches for targeting transgene expression specifically to cortical astrocytes and employed them to study purinergic signaling in astrocytes. To achieve astrocyte-specific transgene expression, I used either floxed adeno-associated viral (AAV) vectors or a Cre-dependent mouse line and several mouse lines expressing the Cre recombinase under astrocyte-specific promoters. Intracerebral injections of a Cre-dependent AAV serotype 5 containing the ubiquitous CAG promoter and an enhanced green fluorescent protein (AAV5.CAG.flex.EGFP) in adult mice expressing Cre recombinase under the human glial fibrillary protein (hGFAP) promoter resulted in a non-astrocyte specific expression in the cortex. Combining inducible mouse lines expressing Cre recombinase under the glutamate aspartate transporter (GLAST) promoter with the same AAV vector resulted in a virtually astrocyte-specific expression of the reporter gene. As an alternative approach for astrocyte-specific transgene expression, we used a Cre-dependent mouse line expressing the genetically encoded Ca2+ indicator GCaMP3. Crossing this mouse line with the above described GLAST-CreERT2 mouse line or a Connexin30 (Cx30)-CreERT2 line led to selective GCaMP3 expression in cortical astrocytes. Second, I investigated both spontaneous and agonist-evoked Ca2+ transients in astrocytic processes, the investigation of which has presented a major challenge in earlier studies, due to the unspecific and weak labeling by membrane-permeable chemical Ca2+ indicators. Using the strategy developed in the first part of my work allowing an astrocyte-specific expression of the genetically encoded Ca2+ indicator GCaMP3. Using two-photon excitation fluorescence (2PEF) imaging in acute slices of the primary somatosensory cortex, I recorded Ca2+ transients in the astrocytic soma and processes. By aid of a custom-made MATLAB routine based on a temporal Pearson correlation coefficient, active regions could be identified in an unbiased manner. Evoked Ca2+ transients were quantified using custom IGOR routines. Spontaneous desynchronized Ca2+ transients occurred in the processes and rarely in the soma. Ca2+ signals appeared localized in distinct microdomains. Their frequency appeared to increase during long recordings of several hundred images, suggesting that fine astrocytes are vulnerable to photodamage under imaging conditions routine in 2PEF microscopy. The possibility to minimize photodamage, by varying the length of the femtosecond laser pulses is under investigation. Bath application of adenosine (1-100 µM) and adenosine-triphosphate (ATP, 100 µM), as well as the application of the non-selective P2X7 receptor agonist (2'(3')-O-(4-Benzoylbenzoyl)adenosine-5'-triphosphate, BzATP, 50-100 µM), in the presence of tetrodotoxin to block neuronal action potentials, evoked synchronized Ca2+ rises in the soma and the processes of astrocytes. The effect of adenosine was dose-dependent. No significant effect of the specific P2Y1 agonist (MRS2365, 50 µM) was seen. Altogether, my work sets up a powerful and versatile toolbox for studying astrocytic Ca2+ signaling at the sub-cellular level. It also pinpoints possible limits of standard two-photon recording protocols to investigate the local Ca2+ signals in fine astrocytic processes.