Dissertationen zum Thema „Dendritic arborization“

During development, the mammalian nervous system wires into a precise network of unrivaled complexity. The formation of this network is regulated by an assortment of molecular cues, both secreted molecules and cell-surface proteins. The ã-Protocadherins (ã-Pcdhs) are particularly good candidates for involvement in these processes. This family of adhesion molecules consists of 22 members, each with diverse extracellular adhesive domains and shared cytoplasmic domains. Thus, cellular interactions with varied adhesive partners can trigger common cytoplasmic responses. Here we investigated the functions of the ã-Pcdhs in two processes involved in neural network formation: dendrite arborization and synaptogenesis. We first asked how ã-Pcdhs regulate synaptogenesis in the spinal cord. We found that the ã-Pcdhs are differentially expressed by astrocytes as well as neurons. In astrocytes, the proteins localize to perisynaptic processes where they can mediate contacts between neurons and astrocytes. In an in vitro co-culture system in which either only astrocytes or only neurons were null for the ã-Pcdhs, we found that astrocytic ã-Pcdh is required for an early stage of synaptogenesis in a contact-dependent manner, while neuronal ã-Pcdh is sufficient for later stages. Conversely, if neurons lacked the adhesion molecules, very few synaptic contacts formed at all. By deleting the ã-Pcdhs from astrocytes in vivo, we demonstrated that these contacts are required for the normal progression of synaptogenesis. We also investigated the function of the ã-Pcdhs in the cerebral cortex. We found that cortical-restricted loss of the adhesion molecules resulted in a severe reduction in thickness of layer 1. By crossing the mutant mice to a line in which scattered layer 5 neurons express YFP, we saw that this thinning resulted from a reduced complexity in the apical tufts of dendrites from layer 5 neurons. Sholl analysis demonstrated that the arbor reduction existed throughout the cell, a phenotype that was recapitulated in vitro. Using the in vitro system, we found that the arborization defect was caused by hyperphosphorylation of the PKC substrate, MARCKS, indicating that the ã-Pcdhs may function by inhibiting PKC activity. Thus, we provide new information about the mechanisms through which the ã-Pcdhs influence neural network development.

CHMP2B est une sous-unité du complexe ESCRT-III, un complexe cytosolique très conservé, responsable du remodelage des membranes biologique, dans divers processus cellulaires. Des mutations de CHMP2B sont associées à une forme familiale de démence frontotemporale. Une étude précédente a mis en évidence que les mutants pathogènes de CHMP2B altèrent la morphologie des épines dendritiques, un phénomène potentiellement à l'origine de la maladie. Ce travail de recherche a pour objectif de décrire le rôle de CHMP2B, et du complexe ESCRT-III, dans la structure et le fonctionnement des épines dendritiques. Dans des lignées cellulaires, nous avons démontré que CHMP2B a la propriété de s'associer préférentiellement à la membrane plasmique, de se polymériser en filaments hélicoïdaux et de former de longs et fins tubes membranaires. Ce résultat indique que CHMP2B est directement impliqué dans le remodelage de la membrane plasmique. Dans les neurones, CHMP2B se concentre dans des régions sous-membranaires proches de la PSD. Une analyse biochimique a montré que CHMP2B et CHMP4B sont associées à d'autres sous-unités, pour former un complexe ESCRT-III postsynaptique particulièrement stable. Nous avons identifié par spectrométrie de masse que ce complexe interagit également avec des protéines d'échafaudage postsynaptiques et des protéines de remodelage du cytosquelette d'actine. La déplétion de CHMP2B par RNAi, dans des neurones en culture, affecte la complexité de l'arborisation dendritique, la morphologie des épines dendritiques et empêche le gonflement des épines associé à la LTP. Des expériences de récupération, avec des mutants pontuels, indiquent que le rôle de CHMP2B dans le maintien de l'arborisation dendritique est dépendant à la fois de de son association avec ESCRT-III et la bicouche phospholipidique. Nous proposons une nouvelle fonctionnalité pour un complexe ESCRT-III contenant CHMP2B, dans les processus de remodelage de la membrane postsynaptique associés à la maturation et à la plasticité des épines dendritiques
CHMP2B is a subunit of ESCRT-III, a highly conserved cytosolic protein machinery, responsible for membrane remodeling in diverse cellular mechanisms. Mutations in CHMP2B are responsible for a familial form of frontotemporal dementia. A previous study highlighted that FTD-related mutants of CHMP2B impair the morphological maturation of dendritic spines, a process that may underlie neurodegeneration in this disease. The goal of this research work id directed towards understanding the role of CHMP2B and ESCRT-III in dendritic spines structure and function. In cell lines, we demonstrated that CHMP2B associates preferentially with the plasma membrane, polymerizes in helical filaments and forms long and thin membrane protrusions. This result indicates that CHMP2B is directly involved in plasma membrane remodeling. In neurons, CHMP2B concentrates in specific sub-membrane microdomains close to the PSD. Biochemical analysis revealed that CHMP2B and CHMP4B associate with other subunits to form a remarkably stable postsynaptic ESCRT-III complex. Mass-spectrometry indicated that this complex also interacts with postsynaptic scaffolds and proteins involved in actin cytoskeleton remodelling. RNAi depletion of CHMP2B, in cultured neurons, alters stability of dendrite branching and morphology of dendritic spines, and impairs spine head growth, normally associated with LTP. Rescue experiments, with point mutants, indicated that CHMP2B activity in dendrite branching is dependent on its capacity to both bind phospholipids and oligomerization with ESCRT-III. We propose a novel functionality for an ESCRT-III complex containing CHMP2B, in maturation-dependent and plasticity-dependent processes of dendritic spine morphogenesis

The assembly and function of neural circuits depend on the patterned outgrowth, guidance and targeting of dendrites into appropriate territories during development. As a hallmark of any neuron, cell type-specific dendrite morphology has a crucial role in determining the sensory or synaptic input a neuron receives. Despite recent advances in exploring the molecular and cellular mechanisms that define dendritic architecture, our understanding of dendrite development is still far from complete. In this thesis research, I have taken a genetic screen approach and a candidate molecule approach to discover novel genes and mechanisms that are involved in dendrite morphogenesis, using Drosophila dendritic arborization (da) neurons as a model system. In three independent but related studies, my research led me to focus on 1) the nuclear receptor for the steroid hormone ecdysone (EcR), 2) the transcription factor Longitudinals Lacking (Lola) and 3) the cell surface recognition molecule Turtle (Tutl). I have found that these three different factors each regulate distinct aspects of dendritic arborization including dendrite branching, distribution and self-avoidance. Through their identification, expression patterns, and a characterization of their effects in loss-of-function and gain-of-function experiments, my findings provide novel insight into the regulatory networks that control dendrite morphogenesis.
L'élaboration et le fonctionnement harmonieux des circuits nerveux dépendent de la croissance des dendrites et de leur guidage et ciblage vers les territoires appropriés au cours du développement. La morphologie des dendrites sert de signe distinctif pour chaque neurone, et ainsi, joue un rôle crucial dans la détermination des différents influx (synaptiques ou sensoriels) que reçoit un neurone. Malgré de récentes avancées dans la compréhension des mécanismes moléculaires et cellulaires qui contrôlent l'architecture dendritique, notre connaissance du développement des dendrites reste encore incomplète. Mes travaux de recherche se sont attachés à découvrir de nouveaux gènes et mécanismes impliqués dans la morphogenèse dendritique. Dans ce but, j'ai choisi au cours de ma thèse deux méthodes d'étude: une approche par crible génétique et une approche par gènes candidats, que j'ai appliquées aux neurones appelés dendritic arborization (da) de la drosophile, mon modèle d'étude. Mes recherches m'ont permis de me concentrer sur trois molécules: 1) le récepteur nucléaire de l'hormone stéroïde ecdysone (EcR), 2) le facteur de transcription Longitudinals Lacking (Lola) et enfin, 3) la molécule de surface Turtle (Tutl). J'ai pu montrer que chacun de ces facteurs est implique dans des aspects distincts du processus de morphogenèse dendritique incluant le branchement, la distribution et l'auto-répulsion dendritiques. L'identification de ces molécules, la description de leurs patrons d'expression et la caractérisation des phénotypes associés à leurs pertes ou gains de fonctions, m'ont permis d'apporter de nouvelles connaissances des réseaux de régulation contrôlant la morphogenèse dendritique.

A key component of neural circuit formation is the elaboration of complex dendritic arbors, the pattern of which constrains inputs to the neuron and thus, the information it processes. As such, many neurodevelopmental disorders such as autism and Down, Rett, and Fragile X Syndromes are associated with reduced forebrain dendrite arborization. Identifying molecules involved in regulating dendrite arborization and neural circuitry formation therefore, is a start to understanding these disorders. Nearly 70 cadherin superfamily adhesion molecules are encoded by the Pcdha, Pcdhb, and Pcdhg gene clusters. These so-called clustered protocadherins (Pcdhs) are broadly expressed throughout the nervous system, with lower levels found in a few non-neuronal tissues. Each neuron expresses a limited repertoire of clustered Pcdh genes, a complicated process controlled by differential methylation and promoter choice. The clustered Pcdh proteins interact homophilically in trans as cis-multimers, which has the potential to generate a combinatorially explosive number of distinct adhesive interfaces that may give neurons unique molecular identities important for circuit formation. Functional studies of animals in which clustered Pcdhs have been deleted or disrupted demonstrate that these proteins play critical roles in neuronal survival, axon and dendrite arborization, and synaptogenesis. Additionally, they have been implicated in the progression of several cancers, suggesting that basic studies of their function and signaling pathways will have important future clinical applications. Recent work has shown that γ-Pcdhs can regulate the Wnt signaling pathway, a common tumorigenic pathway which also play roles in neurodevelopment, but the molecular mechanisms remain unknown. I determined that γ-Pcdhs differentially regulate Wnt signaling: the C3 isoform uniquely inhibits the pathway while 13 other isoforms upregulate Wnt signaling. Focusing on γ-Pcdh-C3, I show that the variable cytoplasmic domain (VCD) is critical for Wnt signaling inhibition. γ-Pcdh-C3, but not other isoforms, physically interacts with Axin1, a key component of the canonical Wnt pathway. The C3 VCD competes with Dishevelled for binding to the DIX domain of Axin1, which stabilizes Axin1 at the membrane and leads to reduced phosphorylation of Wnt co-receptor Lrp6. I also present evidence that the Wnt pathway can be modulated up (by γ-Pcdh-A1) or down (by γ-Pcdh-C3) in the cerebral cortex in vivo, using conditional transgenic alleles. Studies have implicated γ-Pcdhs as a whole, in many neurodevelopmental processes but little is known if distinct roles exists for individual isoforms. By using a specific C3-isoform knockout mouse line engineered in collaboration with Dr. Robert Burgess of The Jackson Laboratory, I was able to uncover a unique role for the C3-isoform in the regulation of dendrite arborization. Mice without γ-Pcdh-C3 exhibit significantly reduced dendrite complexity in cortical neurons. This phenotype was recapitulated in cultured cortical neurons in vitro, which can be rescued by reintroducing the C3-isoform. The ability of γ-Pcdh-C3 to promote dendrite arborization cell-autonomously was abrogated when Axin1 was depleted with an shRNA, indicating that this process by which γ-Pcdh-C3 regulates dendrite arborization is mediated by its interaction with Axin1, which I had previously demonstrated. Together, these data suggest that γ-Pcdh-C3 has unique roles distinct from other γ-Pcdhs, in the regulation of Wnt signaling and dendrite arborization, both of which are mediated by interaction with Axin1.

Growth of a properly complex dendrite arbor is a vital step in neuronal differentiation and a prerequisite for normal neural circuit formation; likewise, overly dense or sparse dendrite arbors are a key feature of abnormal neural circuit formation and characteristic of many neurodevelopmental disorders. Thus, identifying factors involved in aberrant dendrite complexity and therefore aberrant circuit formation, are necessary to understanding these disorders. In my doctoral work I have elucidated both intracellular and extracellular aspects to the gamma-protocadherins (γ-Pcdhs) that regulate dendrite complexity. Loss of the 22 γ-Pcdhs, adhesion molecules that interact homophilically and are expressed combinatorially in neurons and astrocytes, leads to aberrantly high activity of focal adhesion kinase (FAK) and reduced dendrite complexity in cortical neurons. Little is known, however, about how γ-Pcdh function is regulated by other factors. Here I show that PKC phosphorylates a serine residue situated within the shared γ-Pcdh C-terminus; PKC phosphorylation disrupts the γ-Pcdhs’ inhibition of FAK. Additionally, γ-Pcdh phosphorylation or a phosphomimetic mutant reduce dendritic arbors, while blocking γ-Pcdh phosphorylation increases dendrite complexity. Together, these data identify a novel intracellular mechanism through which γ-Pcdh control of a signaling pathway important for dendrite arborization is regulated. Although specific interactions between diverse cell surface molecules are proposed to regulate circuit formation, the extent to which these promote dendrite growth and branching is unclear. Here, using transgenic mice to manipulate expression in vivo, I and my colleagues show that the complexity of a cortical neuron’s dendritic arbor is regulated by γ-Pcdh isoform matching with surrounding cells. Expression of the same single γ-Pcdh isoform leads to exuberant or minimal arbor complexity depending on matched expression of surrounding cells. Additionally, loss of γ-Pcdhs in astrocytes, or induced mis-matching between astrocytes and neurons, reduces dendrite complexity in a cell non-autonomous manner. Thus, these data support our proposal that γ-Pcdhs create a rare neuronal identity that, depending on the identities of surrounding cells, specifies the complexity of that neuron’s dendritic arbor.

Autism spectrum disorders (ASDs) are clinically characterized by decreased communication abilities, impaired social interaction, and the occurrence of repetitive behaviors, with high genetic heritability. Ubiquitin protein ligase E3A (UBE3A) is a gene located on human chromosome 15q11-13, a region that has been the focus of genetic studies of susceptibility to ASD AND Angelman syndrome. An increased UBE3A gene dosage and thus an elevated amount of E6AP, the protein product of UBE3A, is associated with ASD. However, the underlying cellular and molecular details remain poorly understood. Normal development of neuronal structure is critical for intercellular connectivity and overall brain function, and abnormal brain development is a commonality amongst ASDs. These studies therefore investigated the role of increased dosage of Ube3A/E6AP in dendritic arborization and synapse maturation during brain development. Increased E6AP expression in vitro led to significant reduction in dendritic arborization by thinning and fragmentation of the distal tip, along with a decrease in spine density and an increase in immature spines in hippocampal neurons. This morphological remodeling effect was mediated by the ubiquitination and subsequent degradation of the X-linked inhibitor of apoptosis protein (XIAP) by E6AP, which led to activation of caspase-3. Furthermore, activated caspases cleaved tubulin, leading to retraction of microtubules from the distal tip of dendrites, dendritic thinning and eventual disappearance. In vivo studies investigated the role of E6AP in ASD-related neuronal development in Ube3A 2X transgenic mice and found that, consistent with our in vitro studies, increased E6AP in the brain lead to decreased XIAP levels, increased active caspase-3, and enhanced tubulin cleavage in hippocampal tissue in Ube3A 2X mice. In accord, Ube3A 2X mice showed a reduction in dendritic growth and branching and spine density. This work elucidated an important role of Ube3A/E6AP in dendritic pruning and identified XIAP as a novel ubiquitination target of E6AP. These findings provide a new insight into the molecular pathways underlying neurodevelopmental defects in Ube3A/E6AP-associated ASDs.
2018-07-09T00:00:00Z

The molecular mechanisms that coordinate postnatal brain enlargement, synaptic properties and cognition remain an enigma. This study demonstrates that neuronal complexity controlled by p21-activated kinases (PAKs) is a key determinant for postnatal brain enlargement and synaptic properties. Double knockout (DK) mice lacking both PAK1 and PAK3 were severely impaired in postnatal brain growth, resulting in a dramatic reduction in brain volume at maturity. Remarkably, the reduced brain was accompanied by minimal changes in total cell count, due to a significant increase in cell density. However, the DK neurons have smaller soma, markedly simplified dendritic arbors/axons and reduced synapse density. Surprisingly, the DK mice were elevated in basal synaptic responses due to enhanced individual synaptic potency, but severely impaired in bi-directional synaptic plasticity. The PAK1/3 action is likely mediated by cofilin-dependent actin regulation because the activity of cofilin and the properties of actin filaments were specifically altered in the DK mice.

Dissertação de mestrado em Ciências da Saúde
Over the past years increasing amount of attention has been given to signaling lipids as well as to its modulating enzymes, such as phospholipases. Specifically, phospholipase D (PLD), that converts phosphatidylcholine to phosphatidic acid, has been shown to exhibit a role in neurological development and physiology. Several studies have been associating PLD1 and PLD2, the two mammalian PLD isozymes, to neurological events, including neurotransmitter release, dendritic branching, cognition, and brain development. Also, the hippocampus has been suggested as one of the brain regions showing the highest PLD activity, and neurodegenerative conditions (such as Alzheimer’s disease) associated pathways have been shown to be modulated by PLD signaling. Thus, the aim of this project is to better understand the potential role of PLD in hippocampal function in adult mice upon Pld1 or Pld2 genetic ablation. To achieve this, we performed a hippocampal related behavioral characterization of these animals and a structural analysis regarding dendritic morphology. Our behavioral data, specifically considering motor and exploratory activity, anxiety and memory, showed that behavior of PLD2 knockout mice is not altered when compared to their wild type littermates. Although most of this was also observed in the animals lacking PLD1, the results indicate an object recognition-dependent shortterm memory deficit. Regarding hippocampal dendritic arborization, the ablation of either PLD1 or PLD2 led to alterations in dendritic morphology, although with a different impact of each isozyme in the dorsal-ventral axis and in the trisynaptic circuitry of the hippocampus. In summary, our results suggest that the ablation of either PLD1 or PLD2 may have a different effect in the dorsal and ventral hippocampus.
Nos últimos anos tem-se verificado um aumento na atenção dada aos lípidos de sinalização e suas respetivas enzimas moduladoras, tais como as fosfolipases. Especificamente, a fosfolipase D (PLD), que converte fosfatidilcolina em ácido fosfatídico, tem sido demonstrada como desempenhando um papel fundamental no desenvolvimento e fisiologia neurológicos. Diversos estudos têm vindo a associar a PLD1 e a PLD2, as duas isoenzimas identificadas nos mamíferos, a eventos neurológicos, incluindo libertação de neurotransmissores, ramificação dendrítica, cognição e desenvolvimento cerebral. Além disso, o hipocampo tem sido sugerido como uma das regiões do cérebro com maior atividade da PLD, e vias metabólicas associadas a doenças neurodegenerativas (como a doença de Alzheimer) têm sido demonstradas como sendo moduladas por vias de sinalização que envolvem a PLD. Deste modo, o objetivo deste projeto é compreender melhor o potencial papel da PLD na função do hipocampo em ratinhos adultos com ablação genética de Pld1 ou Pld2. Para o alcançar, executámos uma caracterização comportamental destes animais, diretamente relacionada com o hipocampo, assim como uma análise estrutural da sua morfologia dendrítica. Os nossos dados comportamentais, especificamente no que diz respeito a atividade motora e comportamento exploratório, ansiedade e memória, demonstraram que o comportamento de ratinhos knockout para a PLD2 não se encontra alterado quando comparado com os ratinhos wild type. Apesar de que a maior parte destes resultados foi também observada nos ratinhos com ablação genética de PLD1, foi detetado um défice cognitivo, especificamente na memória a curto-prazo dependente do reconhecimento de objetos. Na análise da arborização dendrítica, a remoção de PLD1 ou PLD2 levou a alterações na sua morfologia, com um impacto diferencial de cada uma das isoenzimas no eixo dorsal-ventral e ao longo do circuito trisináptico hipocampal. Em suma, os nossos resultados sugerem que a ablação de PLD1 e PLD2 poderá ter um efeito distinto no hipocampo dorsal e ventral.