Les représentations mentales, en particulier les représentations spatiales, sont étroitement associées à l'activation coordonnée de groupes de cellules dans l'hippocampe. Nous avons entrepris l'étude des propriétés et activités spontanées du réseau hippocampique (et plus particulièrement de la région CA3) afin de mieux en comprendre le fonctionnement, dans les situations normales et pathologiques. En effet, certaines pathologies neurodégénératives telle que la maladie de Parkinson serait potentiellement associées à des troubles cognitifs hippocampo-dépendants. Ainsi dans la première partie de ma thèse, nous avons caractérisé la dynamique temporelle des signaux excitateurs et inhibiteurs spontanés de l'hippocampe par une approche électrophysiologique in vitro sur tranches d'hippocampe mais aussi chez l'animal anesthésié grâce des enregistrements multi-unitaires multi-sites. Ces travaux nous a permis de mettre en évidence que les caractéristiques de la dynamique du réseau CA3 remplissent quelques critères essentiels au concept d'assemblées cellulaires. De plus, cette étude a mis en évidence les caractéristiques fonctionnelles de l'hippocampe chez l'animal normal. Ces résultats peuvent donc être utiles pour de futures études sur les pathologies hippocampo-dépendantes touchant le codage ou la mémoire spatiale telle que la maladie de Parkinson. Ainsi, dans la deuxième partie de ma thèse, nous avons étudié les altérations fonctionnelles du circuit hippocampique chez un modèle rat de la maladie de Parkinson. La maladie de Parkinson est une maladie neurologique qui affecte le système nerveux central et entraine des symptômes essentiellement moteurs. La cause est une dégénérescence des neurones dopaminergiques mais aussi noradrénergiques et sérotoninergiques. Cependant, en dehors des troubles moteurs, cette pathologie est aussi caractérisée par des troubles cognitifs notamment des déficits spatiaux. Notre projet a donc consisté à analyser les mécanismes par lesquels les déplétions monoaminergiques entraîneraient des troubles de l'apprentissage spatial. Ce travail a été réalisé chez le rongeur à l'aide d'une étude associant une approche comportementale et des enregistrements électrophysiologiques chez l'animal anesthésié mais aussi chez l'animal éveillé en comportement. Nous avons ainsi pu mettre en évidence des dysfonctionnements hippocampiques causés par des lésions contrôlées des différents systèmes mono-aminergiques (plus particulièrement dopaminergique et noradrenergique) impliqués dans la maladie de Parkinson. / Mental representations, especially spatial ones are closely related to correlated activity in cellular assembly in the hippocampus. In this work, we analyzed the properties and the spontaneous activity of the hippocampal network in order to unravel its functioning in normal and pathological conditions. Several neurodegenerative disorders such as Parkinson's disease seems to be also associated to cognitive disorder related to hippocampus dysfunction. We first characterized the temporal dynamic properties of spontaneous excitatory and inhibitory signal. We then studied the functional alteration of the hippocampal network in a rat model of Parkinson's disease using behavioral and electrophysiological investigations. Our work showed that controlled lesion of the various monoaminergic systems induced hippocampus dysfunction related to spatial disorientation.In the first part of my thesis, we characterized the temporal dynamic of excitatory and inhibitory signals with electrophysiological recordings in vivo on hippocampal slices but also in anesthetized animals with multi-units multi-sites recordings. These studies allowed us to highlight that dynamic of CA3 network meets the criteria of cells assembly concept. Moreover, we characterize the functional properties of hippocampus in physiological conditions. These results could be useful for further studies on hippocampo-dependant pathologies in the context of spatial coding and memory.Thus, in the second part of my work, we studied the functional alterations of hippocampal network in the context of Parkinson disease. This pathology is a neurodegenerative disease which affects the central nervous system and leads essentially to motor symptoms. The cause is the degeneration of dopamine neurons but also of noradrenalin and serotonin neurons. Nevertheless, this pathology is also associated to cognitive disorders notably a form of spatial disorientation. Our project consisted to analyze the mechanisms by which monoamines depletions led to spatial learning impairments. This work was realized on rats with a study combinating behavioral approach with electrophysiological recordings in anesthetized animals but also in awake animals. We showed that some monoamines depletions (and notably dopamine and noradrenalin depletions) led to spatial impairments in behavioral tasks correlated to a change in firing and coding of neurons of hippocampus.

The hippocampal formation forms a circuit of cytoarchitectonically distinct subregions, and substantial evidence suggests each region makes unique computational contributions that support spatial and episodic memory. With aging, hippocampal subfields undergo unique neurobiological alterations, and primate in vivo work making use of both MR imaging and chronic neural recording devices has important links to changes seen in nonprimate animal models with aging (Thome et al., 2016; Yassa et al., 2011a; Yassa et al., 2010). While MRI offers a noninvasive way to study the hippocampal subfields, identifying hippocampal subregions without using post mortem histology is a challenge. When different research labs attempted to identify the hippocampal subregions using a single subject’s MRI, researchers showed significant disagreement in where to label different subregions (Yushkevich et al., 2015a). Alternatively, chronic neural recording devices offer an invasive solution to studying hippocampal subfields. However, it is currently not clear whether the mechanical trauma and foreign body response produced by neural recording devices disrupts neural circuits critical for behavior. Here, my colleagues and I address these issues with in vivo primate research. Chapter I provides a general introduction to the hippocampal circuits and changes observed in aging. Chapter II presents novel methods for construction of a histology-driven MRI atlas of nonhuman primate hippocampus that addresses accurate identification of hippocampal subfields in MR images. Chapter III presents empirical work that examines whether chronic neural recording devices targeted at the hippocampus affect recognition memory. Finally, Chapter IV provides a general discussion of both works in the context of the broader literature.

Alzheimer’s disease (AD) is widely accepted as being linked with abnormal atrophy of the hippocampus. In the nonhuman-focused literature, the hippocampus has been identified as one of the prominent regions of interest with mechanisms of adult neurogenesis from aerobic exercise. Several human studies over the past decade have shown the effect of exercise that improves cardiorespiratory fitness on the size and function of the hippocampus in participants. However, the size of hippocampal subfields, especially the dentate gyrus (DG), has not been examined in humans even though various animal studies have identified the DG subfield as the primary region of adult neurogenesis induced by aerobic exercise. The point of this investigation, therefore, was to investigate the effect of an exercise intervention on the size of the DG subfield and the related subfield of cornu ammonis (CA) 3. The hypothesis was that an endurance training intervention, designed to improve cardiorespiratory fitness, would increase the volume of the DG and CA3 subfields of the hippocampus more than a resistance training intervention, designed to increase strength, flexibility, and balance, and that improvement in cardiorespiratory fitness would positively correlate with the change in volumes of these subfields. For this investigation, 32 participants (young adults from age 20 to 33 with sedentary lifestyles) were selected from a data set collected for an ongoing study by the Brain Plasticity and Neuroimaging (BPN) Laboratory at Boston University School of Medicine (Boston, MA, USA). The fitness data and T1-weighted and T2-weighted structural magnetic resonance imaging (MRI) data were used in the analysis. FreeSurfer v6.0 software was used to extract volumetric data of the hippocampal subfields using a hippocampal subfield segmentation algorithm. Analysis of variance (ANOVA) with repeated measures and linear regression were used to analyze the statistical significance of the results. The change in volumes for the whole hippocampus, DG, and CA3 did not show any statistically significant differences after endurance training compared with after resistance training. The effect of exercise on the volume of the CA3 subfield appeared to be asymmetrical from left to right, with heavier impact on the left CA3 than on the right CA3. There was no statistically significant correlation between the change in cardiorespiratory fitness and the change in volume of any of the regions analyzed. However, the left whole hippocampus showed a slight trend (p = 0.078; R = 0.317) of weak positive correlation between its volume change and the cardiorespiratory fitness change of the participants. This result was consistent with the previous human literature. Although statistically not significant, most data showed that the endurance training group saw more preservation or increase in volume. This result is encouraging and should be explored further to validate the efficacy of cardiorespiratory exercise as a possible prevention mechanism against AD for young adults later in life.

L'hippocampe est une région du cerveau située dans le lobe temporal médian. Avec d'autres structures limbiques, l'hippocampe est impliqué dans des processus d'apprentissage et de mémorisation et possède un rôle crucial dans le traitement spatial de l'information. Les synapses de l'hippocampe formées entre les fibres moussues (fm) originaires du gyrus denté et les neurones pyramidaux de CA3 ont reçu une attention particulière, compte tenu de la position stratégique occupée par le gyrus denté à l'entrée de l'hippocampe. En outre les synapses fm- CA3 sont distinctes de la plupart des autres synapses excitatrices du système nerveux central par leurs propriétés morphologiques et physiologiques uniques. Cela soulève la question de savoir si ces propriétés uniques reflètent aussi un rôle fonctionnel unique dans le traitement de l'information effectué par cette synapse au sein du microcircuit de l'hippocampe. Malheureusement nous ne savons que peu de choses sur la façon dont les cellules granulaires modulent l'activité des neurones de CA3 dans le réseau intact in vivo (Henze et al, 2002 ; Hagena et Manahan - Vaughan, 2010, 2011). Le manque d'information est dû au fait que la manipulation classique des circuits neuronaux par des approches électriques, pharmacologiques et génétiques manque de précisions spatiale et temporelle in vivo. L'utilisation de la stimulation extracellulaire de fibres moussues peut conduire à l'activation polysynaptique de cellules pyramidales de CA3, qui peuvent ensuite contaminer les réponses enregistrées. Par ailleurs, l'utilisation de critères trop conservateurs peut conduire à l'exclusion des réponses provenant des fibres moussues «purs» aux propriétés méconnues (Henze et al., 2000). Toutefois, le développement récent et rapide de l’optogénétique dans les neurosciences a fourni de nouveaux outils offrant une sélectivité spatiale élevée (activation optique spécifique de la cellule), et une grande précision temporelle (à l'échelle de la milliseconde), permettant la dissection et l'étude des circuits neuronaux in vivo. L'objectif de ma thèse était de mieux comprendre les mécanismes et les conséquences physiologiques de la plasticité synaptique à court terme se produisant à la synapse formée entre les fibres moussues et les neurones pyramidaux de CA3 dans le cerveau de souris intact. La présente thèse se compose de deux parties principales. Dans la première partie, j'ai exploré de nouveaux outils optogénétiques dans le but de contrôler l'activité des cellules granulaires à l’aide d’impulsions de lumière. La stimulation optogénétique repose sur l'activation du canal ionique channelrhodopsin - 2 - lumière fermée ( ChR2 ) par une lumière bleue et induit des potentiels d'action sur une large gamme de fréquences de stimulation. J'ai aussi observé que la stimulation optique peut être utilisée pour déclencher la plasticité à court terme au niveau des synapses fm-CA3.Dans la deuxième partie j'ai affiné la méthodologie de stimulation optogénétique in vivo pour la caractérisation non invasive du fonctionnement synaptique des synapses fm- CA3. La fiabilité de la stimulation optogénétique d'une population neuronale génétiquement ciblée ainsi que la résolution d'une seule cellule obtenue en utilisant des enregistrements de cellules entières sont des étapes importantes vers une meilleure compréhension du rôle fonctionnel des fibres moussues dans le réseau de l'hippocampe in vivo. / The hippocampus is a brain region located in the medial temporal lobe. Along with other limbic structures, the hippocampus is involved in learning and memory processes and has a crucial role in spatial information processing. Within the hippocampus synapses made between mossy fibers (mf) originating from the dentate gyrus and CA3 pyramidal neurons have received particular attention, given the strategic position occupied by the dentate gyrus at the entrance of the hippocampus. Moreover mf-CA3 synapses are distinct from most of other excitatory synapses in the central nervous system for their unusual morphological and physiological properties. This raises the question if these unique properties reflect a unique functional role in information processing carried out by this synapse within the microcircuit of the hippocampus. Unfortunately very little is known on how granule cells modulate the activity of CA3 neurons in the intact network in vivo (Henze et al., 2002; Hagena and Manahan-Vaughan, 2010, 2011). The paucity of information is due to the fact that classical manipulation of neuronal circuits using electrical, pharmacological and genetic approaches lack spatial and temporal precision in vivo. The use of bulk extracellular stimulation may lead to polysynaptic activation of CA3 pyramidal cells, which can subsequently contaminate putative mossy fibers synaptic responses measured in CA3 pyramidal cells. The use of overly conservative criteria on the other side may lead to the exclusion of “pure” mossy fibers responses with unexpected properties (Henze et al., 2000).However the recent and fast growth of optogenetics in neuroscience has provided new tools with high spatial selectivity (cell specific optical activation) and temporal precision (at the millisecond scale), allowing the dissection and investigation of neuronal circuits in vivo. The aim of my thesis was to gain insight into the mechanisms and the physiological consequences of short-term synaptic plasticity occurring at mossy fibers to CA3 pyramidal neurons synapses in the intact mouse brain. The present thesis consists of two main parts. In the first part I explored new optogenetic tools to control the activity of granule cells with pulses of light. Optogenetic stimulation, which relies on the activation of the light-gated ion channel channelrhodopsin-2 (ChR2) by blue light reliably induced action potentials over a wide range of frequencies of stimulation. I also found that optical stimulation can be used to trigger short term plasticity at mf-CA3 synapses. In the second part I refined optogenetic stimulation methodology in vivo for non-invasive characterization of synaptic functioning of the mf-CA3 synapses. The reliability of optogenetic stimulation of a genetically targeted neuronal population together with the single cell resolution obtained using whole-cell recordings are important steps towards a better understanding of the functional role of the mossy fibers in the hippocampal network in vivo.

In hippocampal CA1 and CA3 regions, various properties of neuronal activity follow skewed, lognormal-like distributions, including average firing rates, rate and magnitude of spike bursts, magnitude of population synchrony, and correlations between pre- and postsynaptic spikes. In recent studies, the lognormal features of hippocampal activities were well replicated by a multi-timescale adaptive threshold (MAT) neuron network of lognormally distributed excitatory-to-excitatory synaptic weights, though it remains unknown whether and how other neuronal and network properties can be replicated in this model. Here we implement two additional studies of the same network: first, we further analyze its burstiness properties by identifying and clustering neurons with exceptionally bursty features, once again demonstrating the importance of the lognormal synaptic weight distribution. Second, we characterize dynamical patterns of activity termed neuronal avalanches in in vivo CA3 recordings of behaving rats and in the model network, revealing the similarities and differences between experimental and model avalanche size distributions across the sleep-wake cycle. These results show the comparison between the MAT neuron network and hippocampal readings in a different approach than shown before, providing more insight into the mechanisms behind activity in hippocampal subregions. / Nas regiões CA1 e CA3 do hipocampo, várias propriedades da atividade neuronal seguem distribuições assimétricas com características lognormais, incluindo frequência de disparo média, frequência e magnitude de rajadas de disparo (bursts), magnitude da sincronia populacional e correlações entre disparos pré- e pós-sinápticos. Em estudos recentes, as características lognormais das atividades hipocampais foram bem reproduzidas por uma rede de neurônios de limiar adaptativo (multi-timescale adaptive threshold, MAT) com pesos sinápticos entre neurônios excitatórios seguindo uma distribuição lognormal, embora ainda não se saiba se e como outras propriedades neuronais e da rede podem ser replicadas nesse modelo. Nesse trabalho implementamos dois estudos adicionais da mesma rede: primeiramente, analisamos mais a fundo as propriedades dos bursts identificando e agrupando neurônios com capacidade de burst excepcional, mostrando mais uma vez a importância da distribuição lognormal de pesos sinápticos. Em seguida, caracterizamos padrões dinâmicos de atividade chamados avalanches neuronais no modelo e em aquisições in vivo do CA3 de roedores em atividades comportamentais, revelando as semelhanças e diferenças entre as distribuições de tamanho de avalanche através do ciclo sono-vigília. Esses resultados mostram a comparação entre a rede de neurônios MAT e medições hipocampais em uma abordagem diferente da apresentada anteriormente, fornecendo mais percepção acerca dos mecanismos por trás da atividade em subregiões hipocampais.

In hippocampal CA1 and CA3 regions, various properties of neuronal activity follow skewed, lognormal-like distributions, including average firing rates, rate and magnitude of spike bursts, magnitude of population synchrony, and correlations between pre- and postsynaptic spikes. In recent studies, the lognormal features of hippocampal activities were well replicated by a multi-timescale adaptive threshold (MAT) neuron network of lognormally distributed excitatory-to-excitatory synaptic weights, though it remains unknown whether and how other neuronal and network properties can be replicated in this model. Here we implement two additional studies of the same network: first, we further analyze its burstiness properties by identifying and clustering neurons with exceptionally bursty features, once again demonstrating the importance of the lognormal synaptic weight distribution. Second, we characterize dynamical patterns of activity termed neuronal avalanches in in vivo CA3 recordings of behaving rats and in the model network, revealing the similarities and differences between experimental and model avalanche size distributions across the sleep-wake cycle. These results show the comparison between the MAT neuron network and hippocampal readings in a different approach than shown before, providing more insight into the mechanisms behind activity in hippocampal subregions. / Nas regiões CA1 e CA3 do hipocampo, várias propriedades da atividade neuronal seguem distribuições assimétricas com características lognormais, incluindo frequência de disparo média, frequência e magnitude de rajadas de disparo (bursts), magnitude da sincronia populacional e correlações entre disparos pré- e pós-sinápticos. Em estudos recentes, as características lognormais das atividades hipocampais foram bem reproduzidas por uma rede de neurônios de limiar adaptativo (multi-timescale adaptive threshold, MAT) com pesos sinápticos entre neurônios excitatórios seguindo uma distribuição lognormal, embora ainda não se saiba se e como outras propriedades neuronais e da rede podem ser replicadas nesse modelo. Nesse trabalho implementamos dois estudos adicionais da mesma rede: primeiramente, analisamos mais a fundo as propriedades dos bursts identificando e agrupando neurônios com capacidade de burst excepcional, mostrando mais uma vez a importância da distribuição lognormal de pesos sinápticos. Em seguida, caracterizamos padrões dinâmicos de atividade chamados avalanches neuronais no modelo e em aquisições in vivo do CA3 de roedores em atividades comportamentais, revelando as semelhanças e diferenças entre as distribuições de tamanho de avalanche através do ciclo sono-vigília. Esses resultados mostram a comparação entre a rede de neurônios MAT e medições hipocampais em uma abordagem diferente da apresentada anteriormente, fornecendo mais percepção acerca dos mecanismos por trás da atividade em subregiões hipocampais.

Les états de veille sont composés d’états cérébraux distincts, corrélés avec différents comportements et caractérisés par des oscillations spécifiques observables dans le potentiel de champ local (Local Field Potential, LFP). Bien que les différents états cérébraux et leur signature dans le LFP aient été caractérisés, les mécanismes cellulaires sous-jacents restent à ce jour peu connus. Des changements des propriétés de neurones uniques seraient corrélés avec, et pourraient participer à la génération de ces changements d’états cérébraux. L’activité coordonnée et synchronisée de neurones facilite certains processus cognitifs tels que la mémoire. L’hippocampe joue un rôle essentiel dans les mémoires spatiale et épisodique, et dans l’hippocampe, CA3 est important pour la formation d’associations facilitant l’encodage rapide de la mémoire. De plus, les informations provenant du cortex entorhinal, du gyrus denté, et de CA3 même sont comparées et intégrées dans CA3 avant d’être transmises à CA1. Lors de périodes de repos, le LFP hippocampique présente une activité large et irrégulière (Large Irregular Activity, LIA), ponctuée par des oscillations plus rapides, les sharp-wave ripples, jouant un rôle dans la consolidation de la mémoire. Lors de périodes exploratoires, le LFP hippocampique oscille aux fréquences theta (6-12 Hz) et gamma (30-100 Hz). Les cellules pyramidales (CP) de CA3 jouent un rôle important dans chacun de ces états ; elles sont nécessaires pour les sharp wave lors de périodes de repos, et les oscillations gamma lors de comportements exploratoires. Dans le but d’étudier les modulations intracellulaires des CP de CA3, nous avons réalisé des enregistrements de patch-clamp en configuration cellule entière chez l’animal éveillé. Nous avons associé ces enregistrements avec des mesures du diamètre pupillaire et de la vitesse de locomotion de l’animal, ainsi qu’avec l’enregistrement de l’activité oscillatoire du LFP dans l’hippocampe. Nos résultats montrent que certaines CP de CA3 sont sensibles à la modulation intracellulaire lors de différents rythmes hippocampiques, et ont tendance à diminuer leur potentiel de membrane moyen, leur excitabilité, leur variance et leur décharge de potentiel d’action lors des oscillations theta par rapport aux périodes de LIA. De futures études permettront de déterminer si ces changements sont dus à des changements d’entrées synaptiques et/ou de neuromodulateurs. Ces modulations pourraient jouer un rôle dans l’émergence des rythmes oscillatoires du LFP, et permettre à CA3 de réaliser différentes fonctions mnésiques à différents moments. / Wakefulness is comprised of distinct brain states, correlated with different behaviors and characterized by specific oscillatory patterns in the local field potential (LFP). While much work has characterized different brain states and their LFP signatures, the underlying cellular mechanisms are less known. Changes in single cell properties are thought to correlate with and possibly result in these changes in brain state. Synchronized and coordinated activity among distributed neurons supports cognitive processes such as memory. The hippocampus is essential for spatial and episodic memory, and within the hippocampus, area CA3 is important for rapid encoding of one-trial memory. Additionally, CA3 is the site where information from the entorhinal cortex, dentate gyrus, and CA3 itself is compared and integrated before output to CA1. During quiet wakefulness, the hippocampal LFP displays large irregular activity (LIA) punctuated by sharp-wave ripples, which play a role in memory consolidation. During exploratory behaviors, hippocampal LFP oscillates at both theta and gamma frequencies. CA3 pyramidal cells (PCs) play an important role in each of these brain states; they are necessary for both sharp waves during quiet wakefulness and for gamma oscillations during exploratory behavior. We explored the changes that occur in the intracellular dynamics of CA3 PCs during changes in brain state, by using whole-cell patch-clamp recordings from CA3 PCs in awake head-fixed mice. We combined those recordings with measurements of pupil diameter, treadmill running speed and LFP recordings of oscillatory activity. Our findings show that some CA3 PCs are prone to intracellular modulation during brain rhythms, and tend to decrease their average membrane potential, excitability, variance and output firing during theta as compared to LIA. Future studies will demonstrate whether these effects are due to changes in synaptic and/or neuromodulatory inputs. This modulation at the single-cell level in CA3 could play a role in the emergence of oscillations, and underlie the ability of CA3 to perform different memory functions during different brain states.

<p> The crystal structures of Ca3(VO4)2, Ca3(asO4)2, alpha-Zn3(VO4)2, alpha-Zn2V2O7, Mg2V2O7, VPO5, and whitlockite (Room temperature and 1200 degree C) have determined by X-ray diffraction methods. Tests on the existing theories on the prediction of bond lengths have been made on the vanadate structures determined in this work as well as other reported structures. The range of validity of these theories have been brought out. The importance of the difference of the structure of whitlockite from that of BetaCa3(PO4)2, have also been discussed in detail. </p> / Thesis / Doctor of Philosophy (PhD)

Sharp-wave ripples (SWR) generated in the CA3 subregion of the hippocampus (HC) during rest and sleep appear to coordinate memory consolidation to the neocortex (NC) by (1) reactivating small subsets of neurons (i.e. cell-assemblies) that encode recent waking experience and (2) propagating this information through the hippocampal formation. Although CA3 self-organizes SWRs in the absence of extrinsic inputs, cortical input to the HC conveyed by perforant path (PP) may influence SWR initiation nevertheless. Still, direct evidence that PP synaptic excitation can elicit SWRs is lacking, and it is unclear how this influence might compete or interact with self-organizing mechanisms. This dissertation tested the hypothesis that CA3's SWR pattern generator would self-organize its activity in the absence of PP input, but readily entrain to such input when present. Spontaneous SWRs (sSWR) occurred in slices prepared from the ventral portion of the mouse HC. Low-intensity electrical stimulation of PP afferents evoked short-latency field EPSPs in CA3 that were often followed by precisely timed evoked SWRs (eSWR). The network and single-cell characteristics of sSWRs and eSWRs were indistinguishable, indicative of a common patter generator. PP stimuli that followed sSWRs too closely usually failed to elicit eSWRs. Using a custom MATLAB/Simulink application to control PP stimulus timing during the ~250 ms sSWR refractory period revealed a statistically significant effect of stimulus delay (25, 50, 100, and 200 ms) on eSWR incidence, reaching a value of 0.72 (95% CI = [0.61, 0.81]) 200 ms after sSWR onset. In contrast, sSWR incidence at this time was much lower (95% CI = [0.015, 0.049]). Lesions targeting the direct PP input to CA3 substantially reduced eSWR incidence. In intact slices, eSWRs were readily evoked by stimulating the medial entorhinal cortex (MEC). In summary, PP input to CA3 from the MEC can initiate SWRs at times when self-organizing mechanisms generally cannot. Assuming sSWRs convey information to the NC, the ensuing refractory period might provide an opportunity for cortical feedback to reinforce the recently engaged cell-assembly. In the absence of such feedback, CA3 could revert to its default mode of self-organized replay.

L'encodage de la mémoire dépend de changements durables dans l'activité des circuits synaptiques dans un ensemble de neurones interconnectés. La région CA3 de l'hippocampe reçoit des informations directement ou indirectement (à travers le gyrus denté - GD) en provenance des structures corticales. Des données théoriques et comportementales ont montré que la région CA3 est importante pour l'encodage de la mémoire épisodique, en particulier au stade initial de l'acquisition, en développant vraisemblablement une représentation instantanée d'un contexte. Les neurones pyramidaux CA3 reçoivent une variété de connections afférentes, parmi lesquelles les fibres moussues (FM), les axones des cellules du gyrus denté. Ces connections synaptiqes ont attiré une attention par leurs propriétés morphologiques et fonctionnelles uniques. Malgré les nombreuses études comportementales et computationnelle, les liens entre plasticité des circuits CA3 et encodage de la mémoire ne sont pas bien compris.Le cadre général de ce projet de thèse se situe dans l'étude des mécanismes synaptiques de l'encodage de la mémoire épisodique dans des conditions physiologiques ainsi que dans un modèle de souris de la maladie d'Alzheimer (MA). En effet, la MA se caractérise à un stade précoce par une mémoire épisodique altérée, qui peut être associée à une dysrégulation de la plasticité des circuits CA3.À l'aide de techniques d'enregistrement électrophysiologique, nous avons d'abord exploré les modifications dans les circuits CA3 peu de temps (quelques heures) après conditionnement de la peur contextuelle chez les souris adultes C57Bl6j. Nous avons observé une augmentation de la fréquence des IPSC spontanés accompagnée de changements mineurs dans le nombre de filopodia issus des boutons synaptiques des FM, tandis que les EPSCs et les plasticités à court terme de ces synapses ne sont pas modifiés. Cependant, cette augmentation n'est peut observée 24 heures après l'apprentissage contextuel. Nous avons également tenté de modéliser de manière simplifiée les réseaux neuronaux GD-CA3, afin d'étudier si et dans quelle mesure les interneurones locaux dans la région CA3 contribuent à la précision de l'encodage de la mémoire. [...]Dans l'ensemble ce travail a révélé que la transmission inhibitrice des circuits locaux CA3 de l'hippocampe pourrait être importante dans l'encodage de la mémoire épisodique. Dans le modèle murin de la MA avec déficit de mémoire, il y a une réduction de la transmission GABAergique et des courants médiés par les KAR réduits cellules pyramidales de CA3. Finalement, avons observé une modification transcriptionnelle d'un certain nombre de gènes dans CA3, à des stades précoces de développement de la pathologie dans notre modèle de MA. Notre étude pourrait contribuer à la compréhension des mécanismes pathologiques précoces de la MA, au niveau synaptique ainsi qu'au niveau transcriptionnel, et fournir des idées nouvelles sur les mécanismes sous-jacents au codage rapide de la mémoire contextuelle. / Memory encoding is thought to proceed from durable changes in the activity of synaptic circuits to the storage of patterns of electrical events in a sparsely distributed ensemble of neurons. Located at the entry level of hippocampal circuitry, the CA3 region of hippocampus is thought to be important for episodic memory encoding, especially at the initial stage of acquisition, by presumably developing an instant representation of a context. CA3 pyramidal neurons receive a variety of diverse inputs, among which the mossy fiber (MF) inputs draw special attention for its peculiar structure and unique synaptic properties. However, the links between the plasticity of CA3 circuits and memory encoding are not well understood.This thesis project aimed to address the synaptic mechanisms of episodic memory encoding in physiological conditions as well as in a mouse model of Alzheimer's disease (AD).Using electrophysiological recording techniques, we first explored the changes in CA3 circuits shortly after one-trial contextual fear conditioning in adult C57Bl6j mice. We show that despite hardly any changes in filopodia number of MF terminals, an increase in spontaneous IPSC frequency can be registered, while the EPSCs and short-term plasticities of theses synapses are unaltered. However, this increase cannot be seen anymore 24 hours after the contextual learning. We also tried to do simplified computational modeling of the DG-CA3 neuronal networks, to investigate if and to what extent the local interneurons in CA3 region contribute to memory encoding precision.AD is characterized at an early stage by impaired episodic memory, which may involve dysregulation of the plasticity of CA3 circuits.In the next step, we searched for synaptic deficits in CA3 local circuit in the early stage of AD pathology, taking advantage of a familial AD mouse model: 6-month male APP/PS1 mice. We report that there is a reduction in spontaneous IPSC frequency in CA3 neurons together with decreased inhibitory charges of evoked events at MF-CA3 synapses, whereas the short-term plasticity of these synapses and intrinsic properties of CA3 neurons remain unaffected. Furthermore, there is a robust reduction in Kainate receptor (KAR) mediated currents at MF-CA3 synapses, and the same results can be obtained from PSKO mice too, suggesting that disturbed function of γ-secretase and NCad processing pathways might underlie the dysfunction of KARs at MF-CA3 synapses.Finally, to screen for changes on a transcriptome level, we performed RNA-seq with dissected CA3 tissue from APP/PS1 mice and found a list of up- and down-regulated genes at this early stage of AD. Moreover, we carried out ChIP-seq for a histone modification marker: H3K4me3, which has been shown to be directly related to one-trial contextual memory, and here we report that there is a concrete decrease in H3K4me3 levels at the promoter areas of various genes in CA3 neurons. However, these genes are not overlapping much with the down-regulated genes from RNA-seq result, suggesting that other epigenetic mechanisms might play more important roles in expressing early deficits in this AD mouse model.Taken together, we show that inhibitory innervations of hippocampal CA3 local circuits might be important for episodic memory encoding, and in early AD mouse model with memory deficits, there is reduced GABAergic transmission and reduced KAR-mediated currents in CA3 neurons, together with many active transcriptional regulations across the genome. Our study might contribute to the understanding of early AD pathologies at synaptic level as well as transcriptional level, and provide novel insights into the mechanisms underlying rapid encoding of contextual memory.