abstract: LKB1/STK11 is a serine/threonine kinase first identified in C.elegans as a gene important for cell polarity and proliferation. Mutations in LKB1 are the primary cause of Peutz-Jegher’s cancer syndrome, an autosomal dominantly inherited disease, in which patients are predisposed to benign and malignant tumors. Past studies have focused on defining LKB1 functions in various tissue types, for example LKB1 regulates axonal polarization and dendritic arborization by activating downstream substrates in excitatory neurons of the developing neocortex. However, the implications of LKB1, specifically in the developing cortical inhibitory GABAergic interneurons is unknown. LKB1 deletion was achieved by using Cre-lox technology to induce LKB1 loss in cells localized in the medial ganglionic eminence (MGE) that express Nkx2.1 and generate cortical GABAergic neurons. In this research study it is suggested that LKB1 plays a role in GABAergic interneuron specification by specifically regulating the expression of parvalbumin during the development of fast-spiking interneurons. Preliminary evidence suggest LKB1 also modulates the number of Nkx2.1-derived oligodendrocytes in the cortex. By utilizing a GABAergic neuron specific LKB1 deletion mutant, we confirmed that the loss of parvalbumin expression was due to a GABAergic neuron autonomous function for LKB1. These data provide new insight into the cell specific functions of LKB1 in the developing brain. / Dissertation/Thesis / Masters Thesis Biology 2019

The Distal-less homeobox (Dlx) genes encode homeodomain transcription factors found in all animals of the phylum Chordata. These genes are involved in early vertebrate development of limbs, sensory organs, branchial arches and the forebrain (telencephalon and diencephalon). The mouse and human genomes each have six Dlx genes organized into convergently transcribed bigene clusters (Dlx1/2, Dlx3/4 and Dlx5/6). In the forebrain, Dlx1/2 and Dlx5/6 genes play essential roles in GABAergic neuron proliferation, migration and survival. Each bigene cluster includes a short intergenic region (~3.5-16kb) harboring cis-regulatory elements (CREs) that control expression of the Dlx genes. The Dlx1/2 intergenic region harbors the I12b/I12a CREs, while Dlx5/6 includes I56i/I56ii. In determining the regulatory roles of the CREs on Dlx activity and forebrain development, I have characterized the phenotypic changes that occur in mice that have an I56i enhancer deletion. I have also characterized mice with double deletions of I56i and I12b as well as mice that harbored an I12b deletion and have a SNP in the I56i enhancer (vI56i). Mutant mice with a single targeted deletion of I56i are viable, fertile and do not show obvious developmental defects. These mice have significant decreases in Dlx5/6, Gad1/Gad2 and Evf-2 expression in the forebrain and have defects related to GABAergic neuron development. The ΔI56i mutants demonstrate a behavioral phenotype related to anxiety and learning deficits. Mice that lack the I12b enhancer and have the vI56i do not show morphological abnormalities but have severely disrupted Dlx expression. When mice are homozygous for the I56i and I12b enhancer deletion, they do not survive past post natal day 5 and exhibit a dwarfed body size. These mice look weak and seem to have limited motor ability. In characterizing mice with targeted deletions of highly conserved Dlx enhancers, we will have a better understanding of forebrain development.

The Distal-less homeobox (Dlx) genes encode homeodomain transcription factors that play important roles in the development of limbs, sensory organs, branchial arches and the forebrain. In the forebrain, Dlx1 and Dlx2 are expressed in neuronal progenitor cells and play essential roles in GABAergic neuron differentiation and migration. In order to understand the fate of neuronal progenitor cells that express dlx1a/2a genes in the brain, we produced lines of Tg(dlx1a/2a:CreERT2) transgenic fish expressing the CreERT2 recombinase driven by regulatory elements from the dlx1a/2a locus. CreERT2 expression in these fish faithfully recapitulates that of dlx1a/2a genes in the forebrain. These fish were mated with Tg(ubi:Switch) reporter fish that express a loxP-flanked GFP gene followed by mCherry, driven by the ubiquitin promoter. Upon tamoxifen treatment, the double transgenic fish express mCherry in dlx1a/2a-expressing cells. Live imaging data showed that mCherry-expressing cells were observed first in the telencephalon and prethalamus, regions from which they migrated and populated the telencephalon, prethalamus and hypothalamus by 10dpf. Fate mapping of mCherry-expressing cells in double transgenic fish demonstrated that a majority of dlx1a/2a-expressing cells give rise to GABAergic neurons. Furthermore, as zebrafish produce new neurons throughout life, the role of dlx1a/2a during adult neurogenesis was examined. Our preliminary data showed that dlx1a/2a-expressing progenitor cells populate various domains of the forebrain during adult neurogenesis. Our lineage tracing system provides a powerful tool to investigate the origin of GABAergic neuron progenitors and the mechanisms by which they populate or repopulate the adult brain.

The Distal-less homeobox (Dlx) genes encode homeodomain transcription factors found in all animals of the phylum Chordata. These genes are involved in early vertebrate development of limbs, sensory organs, branchial arches and the forebrain (telencephalon and diencephalon). The mouse and human genomes each have six Dlx genes organized into convergently transcribed bigene clusters (Dlx1/2, Dlx3/4 and Dlx5/6). In the forebrain, Dlx1/2 and Dlx5/6 genes play essential roles in GABAergic neuron proliferation, migration and survival. Each bigene cluster includes a short intergenic region (~3.5-16kb) harboring cis-regulatory elements (CREs) that control expression of the Dlx genes. The Dlx1/2 intergenic region harbors the I12b/I12a CREs, while Dlx5/6 includes I56i/I56ii. In determining the regulatory roles of the CREs on Dlx activity and forebrain development, I have characterized the phenotypic changes that occur in mice that have an I56i enhancer deletion. I have also characterized mice with double deletions of I56i and I12b as well as mice that harbored an I12b deletion and have a SNP in the I56i enhancer (vI56i). Mutant mice with a single targeted deletion of I56i are viable, fertile and do not show obvious developmental defects. These mice have significant decreases in Dlx5/6, Gad1/Gad2 and Evf-2 expression in the forebrain and have defects related to GABAergic neuron development. The ΔI56i mutants demonstrate a behavioral phenotype related to anxiety and learning deficits. Mice that lack the I12b enhancer and have the vI56i do not show morphological abnormalities but have severely disrupted Dlx expression. When mice are homozygous for the I56i and I12b enhancer deletion, they do not survive past post natal day 5 and exhibit a dwarfed body size. These mice look weak and seem to have limited motor ability. In characterizing mice with targeted deletions of highly conserved Dlx enhancers, we will have a better understanding of forebrain development.

L’élément constitutif des synapses centrales est le site synaptique individuel, comprenant une zone active du côté présynaptique et une densité postsynaptique associée. Du fait de limitations techniques nos connaissances sur le mode de fonctionnement d’un site synaptique restent insuffisantes. Pour faire progresser cette question nous projetons d’effectuer des enregistrements en paires entre interneurones de la couche moléculaire du cervelet. Ces neurones forment des synapses qui ont des signaux élémentaires quantiques de grande taille, et les synapses comprennent parfois un seul site synaptique, ce qui fait qu’ils offrent des avantages décisifs pour ce projet. Les réponses postsynaptiques à des trains de potentiels d’action seront étudiées dans différentes conditions expérimentales. Les résultats seront interprétés par un modèle supposant que les vésicules synaptiques doivent se lier à un petit groupe de sites d’arrimage avant l’exocytose. / The unitary element of central synaptic transmission is a single synaptic site, with one active zone as presynaptic component and the postsynaptic density as postsynaptic partner. Due to technical limitations there is much uncertainty on the mode of functioning of a single synaptic site. To address this issue it is planned to perform paired recordings between interneurons of the molecular layers of the cerebellum. These neurons form synapses with a large quantal size, and occasionally displaying a single release site, and are thus favorable for this study. Postsynaptic responses will be studied in response to trains of presynaptic action potentials under various conditions. The results will be compared to a model supposing the obligatory binding of vesicles to a small complement of docking sites prior to exocytosis.

A blockade of GABAergic inhibition increases the receptive field(RF) size of most somatosensory cortex (SI) and some ventrobasal thalamus (VB) neurons. The results suggest RF size of cuneate neurons may be modulated through GABAa and GABAb receptors, independent of firing frequency.

Dans le cerveau adulte, la signalisation liée à l’adénosine facilite ou inhibe la libération vésiculaire de neurotransmetteurs suite à l’activation des récepteurs de l’adénosine de type 2A ou 1 (A2AR ou A1R), respectivement. Cependant, son rôle dans le développement est mal connu. Au cours de ma thèse, j’ai étudié le rôle de la signalisation adénosine dans la synaptogenèse GABAergiques de l’hippocampe. Nous avons mis en évidence (i) une sécrétion activité-dépendante accrue d’adénosine et d’ATP pendant la période de synaptogenèse, (ii) un pic d’expression de l’enzyme limitant la formation de l’adénosine à partir de l’ATP extracellulaire, l’ecto-5’-nucleotidase, aux synapses pendant cette période critique, et (iii) un pic d’expression péri/post- synaptique du A2AR concomitant de la période de synaptogenèse. Cette expression développementale des molécules clés de la signalisation adénosine dépendante du A2AR corrélait avec un rôle de ce récepteur dans la stabilisation des synapses GABAergiques naissantes, une régulation restreinte à la période de synaptogenèse. De plus, la suppression de A2AR par une approche shRNA dans des neurones isolés conduisait à une perte de synapses GABAergiques équivalente à celle observée après un blocage pharmacologique de l’activité du A2AR, signifiant que la stabilisation synaptique médiée par le A2AR est un processus « cellule autonome » indépendant de l’activité du réseau neuronal et qu’elle requiert l’activation du A2AR dans la cellule post-synaptique.L’ATP et l’adénosine sont secrétés par la glie et les neurones ; cependant, nous avons montré in vitro que la libération neuronale activité-dépendante suffit à stabiliser les synapses GABAergiques naissantes. En utilisant la vidéomicroscopie sur cellules vivantes, nous avons montré que la signalisation adénosine stabilise les synapses actives. Puis, nous avons caractérisé le mécanisme moléculaire sous-jacent. Nous rapportons la contribution de la cascade adénylate cyclase/adénosine monophosphate cyclique/protéine kinase A et nous avons identifié une ciblé clé, la géphrine, la molécule d’ancrage postsynaptique des récepteurs GABAA. Enfin, nous avons mis en évidence que la stabilisation de l’élément présynaptique requiert probablement le complexe trans-synaptique Slitrk3-PTPd.Puisque le GABA exerce une fonction similaire au cours du développement et que le GABA et l’adénosine sont co-libérés à certaines synapses, j’ai étudié l’interaction entre ces deux voies de signalisation. Mes résultats favorisent l’hypothèse que la signalisation GABA, en activant la calcium-calmoduline, converge vers la signalisation adénosine en potentiant les adenylates cyclases sensibles au calcium. Mon travail m’a permis de proposer que, au cours d’une période clé du développement, les A2ARs postsynaptiques agissent comme des senseurs de l’activité des terminaisons présynaptiques GABAergiques pour stabiliser les synapses actives. En absence d’activité et donc de libération d’adénosine/ATP, les synapses seraient éliminées. / In the adult brain, adenosine signaling facilitates or inhibits neurotransmitter vesicular release mainly through activation of type 2A or 1 adenosine receptors (A2AR or A1R), respectively. However, its role in development remains to be elucidated. During my PhD, I addressed the role of A2AR-mediated signalling in GABAergic synaptogenesis in the hippocampus. We found (i) a larger activity-dependent release of ATP and adenosine during the period of synaptogenesis in the hippocampus, (ii) a peak of expression of the ecto-5’-nucleotidase, the rate-limiting enzyme for the formation of adenosine from extracellular ATP in synapses during this critical period, and (iii) a peak of peri/post-synaptic expression of A2AR concomitant with the period of synaptogenesis. This developmental expression of the key molecules of the adenosine A2AR signalling pathway correlated with a role of A2AR in the stabilization of nascent GABA synapses, a regulation restricted to the period of synaptogenesis. Furthermore, suppressing A2AR with a shRNA approach in isolated neurons led to a loss of synapses equivalent to that seen upon A2AR activity blockade, reporting that the A2AR-mediated synapse stabilization is a cell autonomous process that requires A2AR activation in the postsynaptic cell. ATP/adenosine can be secreted by both glia and neurons; however, we found that activity-dependent release of neuronal adenosine is sufficient to stabilize newly formed GABA synapses in vitro. Using live cell imaging, we showed adenosine signalling stabilizes active synapses. We then characterized the molecular mechanism downstream postsynaptic A2AR. We report the contribution of adenylyl cyclase/cyclic adenosine monophosphate/protein kinase A signalling cascade and we identified a key target, the postsynaptic scaffolding molecule gephyrin. We further showed the A2AR-mediated stabilization of the presynaptic compartment most probably requires the trans-synaptic Slitrk3-PTPd complex. Since GABA exerts a similar function during development and GABA and adenosine are co-released at some synapses, I further investigated the interplay between these two pathways. My results support the hypothesis that GABA signalling converge onto the adenosine signalling pathway by potentiating calcium-sensitive adenylyl cyclases through the activation of calmodulin.Altogether these results let us propose that, during a key developmental period, postsynaptic A2ARs act as sensors of the activity of GABAergic presynaptic terminals to stabilize active nascent GABAergic synapses. In absence of activity and therefore secretion of adenosine/ATP, synapses will be eliminated.

Inhibitory neurons exert control the expression of many aspects of behaviour by regulating the effectiveness of excitatory neural function. By comparison with excitatory neural systems, relatively little is known about the development of inhibitory neurons and the influence which these neurons exert on the development of other neural systems. Two issues which relate to the development of inhibitory neurons are of particular interest. First, a paradox arises when inhibitory neurons are considered in terms of modern models of synaptic development which involve activity-dependent mechanisms of synaptic plasticity. Second, there is some evidence that inhibitory neurotransmitters may act in a special trophic manner during the early development of nervous systems. Investigations of these issues would be greatly facilitated in a neural system in which it was possible to experimentally control aspects of the development of individual pre- and postsynaptic cells. The aim of the results presented in this thesis was to characterise the normal development of one such system: the GABAergic inhibitory system of the Australian freshwater crayfish, Cherax destructor. The ontogeny of the inhibitory neurotransmitter GABA across the embryonic period of 30% to 100% development was investigated using immunohistochemical techniques. GABA-like immunoreactive cells and fibres were first detected in the embryonic brain region. The expression of GABA-like immunoreactivity progressed along a rostro-caudal gradient, with GABA-like immunoreactive cells being detected in the most anterior thoracic ganglia at 45% development and in all ganglia by 65% development. GABA-like immunoreactive fibres were evident in peripheral nerves as early as 55% development and ramified extensively throughout the neuropil of the nervous system by 65% development. By contrast, immunoreactivity to the primary excitatory neurotransmitter, glutamate, was not detected until 60-65% development. Glutamate-like immunoreactivity at 60-65% development was evident only in the form of punctate staining in the midline of the ventral nerve cord. Cell body staining was observed only at 90% development and was restricted to only a few cells on the periphery of the ventral nerve cord. Radio-labelled ligand binding methods and autoradiography were used to study the expression of putative GABA receptors in the Cherax embryos from 30% to 100% development. Specific binding was evident in the earliest embryos studies at 30% development. There was an initial increase in binding from 30% to 40% development, followed by a dramatic drop to almost zero binding at 50-55% development. This was followed by a gradual increase in binding levels with age, reaching a plateau at 85% development. Preliminary pharmacological evaluation of binding indicated that at least three GABA receptor types were expressed during embryonic development. Methods for culturing, dissociated neural tissues explanted form Cherax embryos at 85% development were established. The success of cultures was demonstrated by neurite extension, and neuronal networks in which neurons appeared to form connections with other neurons and with explanted muscle cells after two days in culture. Immunohistochemical studies demonstrated that some explanted neurons expressed GABA-like immunoreactivity within two days of explanting. These studies have provided a comprehensive description of the development of GABAergic neurons and their receptors in Cherax destructor embryos. The very early expression of GABA-like immunoreactivity, coupled with the early onset of specific GABA binding, strongly indicates that the GABAergic neurons are functional and able to exert an effect on other cells during much of the period of nervous system development in crayfish embryos. These results support the hypothesis that inhibitory neurons may play an important role as regulators of the overall process of assembly and maturation of the nervous system and provide a substantial basis for future experimental studies in which the specific action of inhibitory neurons on the development of discrete components of the crayfish nervous system may be investigated.

Abstract The Dlx genes play an important role in the differentiation and migration of gamma-aminobutyric acid (GABA) interneurons of mice. GABAergic interneurons are born in the proliferative zones of the ventral telencephalon and migrate to the cortex early during mouse development. Single Dlx mutant mice show only subtle phenotypes. However, the migration of immature interneurons is blocked in the ventral telencephalon of Dlx1/Dlx2 double mutant mice leading to reduction of GABAergic interneurons in the cortex. Also, Dlx5/Dlx6 expression is almost entirely absent in the forebrain, most probably due to cross-regulatory mechanisms. In zebrafish, the role of dlx genes in GABAergic interneuron development is unknown. By injecting Morpholino, we double knocked down dlx1 and dlx2 genes in wildtype zebrafish to investigate the function of the two genes in zebrafish GABAergic interneuron development. By comparing different subsets of GABAergic interneuron development in wildtype and dlx1/2 morphant zebrafish forebrain, we found out that at 3dpf, 4dpf and 7dpf, double knockdown of dlx1 and dlx2 genes in zebrafish remarkably reduced the number of Calbindin-, Somatostatin- and Parvalbumin-positive GABAergic neurons, whereas the development of Calretinin-positive neurons is slightly affected. These results suggest that in zebrafish, dlx1a and dlx2a genes are important for the development of certain subtypes of GABAergic interneurons (Calbindin-, Somatostatin- and Parvalbumin-positive neurons) and may have minor influence on Calretinin-positive neuron development. This also suggests that different regulatory mechanisms are involved in the development of the different subtypes of GABAergic interneurons.

Abstract The Dlx genes play an important role in the differentiation and migration of gamma-aminobutyric acid (GABA) interneurons of mice. GABAergic interneurons are born in the proliferative zones of the ventral telencephalon and migrate to the cortex early during mouse development. Single Dlx mutant mice show only subtle phenotypes. However, the migration of immature interneurons is blocked in the ventral telencephalon of Dlx1/Dlx2 double mutant mice leading to reduction of GABAergic interneurons in the cortex. Also, Dlx5/Dlx6 expression is almost entirely absent in the forebrain, most probably due to cross-regulatory mechanisms. In zebrafish, the role of dlx genes in GABAergic interneuron development is unknown. By injecting Morpholino, we double knocked down dlx1 and dlx2 genes in wildtype zebrafish to investigate the function of the two genes in zebrafish GABAergic interneuron development. By comparing different subsets of GABAergic interneuron development in wildtype and dlx1/2 morphant zebrafish forebrain, we found out that at 3dpf, 4dpf and 7dpf, double knockdown of dlx1 and dlx2 genes in zebrafish remarkably reduced the number of Calbindin-, Somatostatin- and Parvalbumin-positive GABAergic neurons, whereas the development of Calretinin-positive neurons is slightly affected. These results suggest that in zebrafish, dlx1a and dlx2a genes are important for the development of certain subtypes of GABAergic interneurons (Calbindin-, Somatostatin- and Parvalbumin-positive neurons) and may have minor influence on Calretinin-positive neuron development. This also suggests that different regulatory mechanisms are involved in the development of the different subtypes of GABAergic interneurons.