Abstract The synaptic input received by neurons in cortical circuits is in constant flux. From both environmental sensory changes and learning mechanisms that modify synaptic strengths, the excitatory and inhibitory signals received by a post-synaptic cell vary on a continuum of time scales. These variable inputs inherent in different sensory environments, as well as inputs changed by Hebbian learning mechanisms (which have been shown to destabilize the activity of neural circuits) serve to limit the input ranges over which a neural network can effectively operate. To avoid circuit behavior which is either quiescent or epileptic, there are a variety of homeostatic mechanisms in place to maintain proper levels of circuit activity. This article provides a basic overview of the biological mechanisms, and consider the advantages and disadvantages of homeostasis on a theoretical level. / text

No description available.

Le synaptic scaling est une forme de plasticité homéostatique par lequel les synapses ajustent leur efficacité pour compenser des variations normales ou pathologiques de l'activité neuronale notamment lors des maladies neurodégeneratives ou suite à la perte d’afférences sensorielles après une lésion. Dans un modèle expérimental classique, le traitement chronique des neurones primaires avec la tétrodotoxine (TTX) pour bloquer la propagation des potentiels d'action présynaptiques induit une augmentation significative de l'amplitude des courants miniatures excitateurs transmis par les récepteurs du glutamate AMPA postsynaptiques. Plusieurs voies de signalisation ont été proposées, dont celle impliquant les microARNs (miRs), de petits ARN non-codants qui inhibent la traduction des protéines en se liant aux ARN messagers cibles. Dans ce contexte, nous avons exploré l'hypothèse que le microARN, miR-124, fortement exprimé dans le cerveau, pourrait être un régulateur important de l'homéostasie synaptique en contrôlant l'expression de la protéine synaptopodine, une protéine structurante des épines dendritiques et indispensable à l'expression du synaptic scaling.En combinant des approches de RTq-PCR, d'immunocytochimie et d'électrophysiologie in vitro, nous avons montré dans un premier temps que la privation globale de l'activité des neurones primaires d’hippocampe diminuait le niveau d'expression de miR-124 et augmentait celui de la synaptopodine et des récepteurs AMPA dont la sous-unité GluA2 est une autre cible de miR-124. Par ailleurs, en rendant des synapses individuelles inactives via l’expression présynaptique de la toxine tétanique, nous avons observé que le recrutement synaptique des récepteurs AMPA et de la synaptopodine était spécifique de ces synapses, suggérant une régulation homéostatique locale. Dans un deuxième temps, nous avons trouvé que la surexpression de miR-124 ou l’inhibition de son interaction avec l’ARNm de la synaptopodine ou de GluA2 bloquaient la réponse synaptique homéostatique induite par le traitement TTX. Enfin, des expériences de FRAP ont suggéré que la synaptopodine influençait le trafic des récepteurs AMPA à la membrane probablement en les stabilisant à la synapse, ce qui expliquerait ainsi son rôle pendant la plasticité homéostatique. / Synaptic scaling is a form of homeostatic plasticity where synapses adjust their own efficacy to compensate for normal or pathological variations in neuronal activity such as neurodegenerative disorders or sensory deprivation after a lesion. In a well-established paradigm, the chronic application of tetrodotoxin (TTX) in primary neurons, to block presynaptic action potential propagation, induces a significant upscaling of miniature excitatory postsynaptic currents mediated-AMPA receptors. Numerous regulators of this plasticity have been identified including microRNAs (miR), which are small endogenous non-coding RNAs, inhibiting protein translation by binding to mRNA targets. This led us to hypothesize that the most highly expressed microRNA in the brain, miR-124, could be an important regulator of homeostatic scaling by controlling the expression of synaptopodin, a structural protein of dendritic spines playing a crucial role in homeostatic plasticity.By combining qRT-PCR, immunocytochemistry and in vitro electrophysiology approaches, first we showed that a global 48hrs TTX treatment in hippocampal primary neurons led to a decrease in miR-124 level and an increase in the expression of synaptopodin and synaptic AMPA receptors containing the GluA2 subunit which is another miR-124 target. Moreover, we observed that the synaptic accumulation of AMPA receptors and synaptopodin could be synapse-specific by expressing the tetanus toxin to block the activity of individual presynapses, which suggested a local homeostatic regulation. Importantly, we found that overexpressing miR-124 or inhibiting its interaction with synaptopodin or GluA2 mRNAs blocked the synaptic homeostatic response. In addition, FRAP experiments suggested that synaptopodin controlled AMPA receptor trafficking at the membrane by probably retaining them in dendritic spines, which could explain its role during homeostatic plasticity.

Optogenetics is a set of technologies that enable optically triggered gain or loss of function in genetically specified populations of cells. Optogenetic methods have revolutionized experimental neuroscience by allowing precise excitation or inhibition of firing in specified neuronal populations embedded within complex, heterogeneous tissue. Although optogenetic tools have greatly improved our ability manipulate neural activity, they do not offer control of neural firing in the face of ongoing changes in network activity, plasticity, or sensory input. In this thesis, I develop a feedback control technology that automatically adjusts optical stimulation in real-time to precisely control network activity levels. I describe hardware and software tools, modes of optogenetic stimulation, and control algorithms required to achieve robust neural control over timescales ranging from seconds to days. I then demonstrate the scientific utility of these technologies in several experimental contexts. First, I investigate the role of connectivity in shaping the network encoding process using continuously-varying optical stimulation. I show that synaptic connectivity linearizes the neuronal response, verifying previous theoretical predictions. Next, I use long-term optogenetic feedback control to show that reductions in excitatory neurotransmission directly trigger homeostatic increases in synaptic strength. This result opposes a large body of literature on the subject and has significant implications for memory formation and maintenance. The technology presented in this thesis greatly enhances the precision with which optical stimulation can control neural activity, and allows causally related variables within neural circuits to be studied independently.

Information processing in the brain relies on a functional balance between excitation and inhibition, the disruption of which leads to network destabilization and many neurodevelopmental disorders, such as autism spectrum disorders. One of the homeostatic mechanisms that maintains the excitatory and inhibitory balance is called synaptic scaling: Neurons dynamically modulate postsynaptic receptor abundance through activity-dependent gene transcription and protein synthesis. In the first part of my thesis work, I discuss our findings that a chromatin reader protein L3mbtl1 is involved in synaptic scaling. We observed that knockout and knockdown of L3mbtl1 cause a lack of synaptic downscaling of glutamate receptors in hippocampal primary neurons and organotypic slice cultures. Genome-wide mapping of L3mbtl1 protein occupancies on chromatin identified Ctnnb1 and Gabra2 as downstream target genes of L3mbtl1-mediated transcriptional regulation. Importantly, partial knockdown of Ctnnb1 by itself prevents synaptic downscaling. Another aspect of maintaining E/I balance centers on GABAergic inhibitory neurons. In the next part of my thesis work, we address the role of the scaffold protein Shank1 in excitatory synapses onto inhibitory interneurons. We showed that parvalbumin-expressing interneurons lacking Shank1 display reduced excitatory synaptic inputs and decreased levels of inhibitory outputs to pyramidal neurons. As a consequence, pyramidal neurons in Shank1 mutant mice exhibit increased E/I ratio. This is accompanied by a reduced expression of an inhibitory synapse scaffolding protein gephyrin. These results provide novel insights into the roles of chromatin reader molecules and synaptic scaffold molecules in synaptic functions and neuronal homeostasis.