Synaptic dysfunction in early encephalopathies : from genes to function

Tese de doutoramento, Medicina (Neurologia), Universidade de Lisboa, Faculdade de Medicina, 2015The synapse is the functional unit for neuronal communication. Mutations in genes that encode relevant proteins for synaptic functions are being increasingly identified in neuropediatric disorders. The present work is focused on synaptic dysfunction in Rett Syndrome, which is caused by mutation in the MECP2 gene. This encodes for methyl-CpG-binding protein 2 (MeCP2), an essential epigenetic regulator in mammalian brain development. Rett syndrome is characterized by a period of cognitive decline, hand stereotypies, autistic traits and seizures, following an apparently normal early infancy. This disease can be classified as a synaptopathy, since it comprises simultaneously impairments in synaptogenesis, synaptic maturation and synaptic plasticity. There is evidence for the possibility of phenotypical rescue in Mecp2 deficient mice models, but current treatments for Rett Syndrome are primarily symptomatic therapies for epilepsy or behavioral disturbances, and not for correction of the underlying brain abnormalities. The molecule γ-amino butyric acid (GABA) is the main inhibitory (hyperpolarizing) neurotransmitter in the adult brain, but has excitatory (depolarizing) action in the developing brain, when it binds to GABA A receptors. This functional shift is dependent of neuronal maturational changes that include differences in the expression of cation chloride cotransporters, which regulate intracellular chloride concentration. One of the key molecular changes for this process is a perinatal neuronal membrane upregulation of potassium chloride cotransporter 2 (KCC2), an ion cotransporter that extrudes chloride form the cell, together with sodium potassium chloride cotransporter 1 (NKCC1), which transports choride into the cell. Deregulation of cation chloride cotransporters’ expression or function has been associated with neurodevelopmental disorders like neonatal seizures, fragile X, tuberous sclerosis, Down syndrome, underpinning the relevance of this balance for an adequate central nervous system postnatal development. In the present thesis, I describe a disturbed GABAergic maturational process in Rett Syndrome, regarding intracellular chloride regulation. Starting from the clinic, we searched for a method to detect synaptic proteins in the cerebrospinal fluid of pediatric patients. The earlier postnatal period is characterized by intense synaptogenesis and synaptic pruning, allowing the detection of these proteins in the cerebrospinal fluid, using immunoblot analysis. Since we were interested in GABAergic function maturation, and this process was dependent on the expression of NKCC1 and KCC2, abnormalities in the cerebrospinal fluid levels of these proteins were searched in Rett Syndrome patients and a decrease in KCC2 was observed. In order to obtain a model to understand the impact of this reduction of KCC2 in neuronal function, human iPSCs were reprogramed from Rett patients’ fibroblasts. Human skin biopsies were collected in accordance with European and National ethical regulation and induced pluripotent stem cells were generated from fibroblasts upon infection with a retroviral vector expressing the four canonical transcription factors (Oct4, Sox2, Klf4, and Myc). Neural commitment of patient specific induced pluripotent stem cells was induced under defined conditions. Neuronal cortical populations were then derived in a monolayer culture system using a protocol that mimicks human cortical development in vitro. Perforated patch recordings were performed in these neurons, and GABA-evoked postsynaptic currents were measured to evaluate GABA A receptor equilibrium potencial. Our preliminary data indicates that recordings from MECP2 mutant cells exhibit a GABA A receptor equilibrium potential that is more positive than in recordings from control cells, suggesting pathologic changes in chloride gradient, characteristic of an immature state. These results were complemented with experiments in the extensively characterized Mecp2 knock out mouse model (Mecp2tm1.1Bird/J). The level of KCC2 protein expression is lower in Mecp2-KO mice, when compared to control littermates, as addressed by western blot analysis of 6 week old hippocampi. Moreover, hippocampal electrophysiological recordings show reduction of membrane resting potential and threshold potential in the Mecp2 knock-out model, where synaptic transmission evaluated by Input/Output curves reveals an increased excitatory synaptic transmission. Brain derived neurotrophic factor (BDNF) is a neurotrophin relevant for synaptic function, neuronal maturation and neuronal survival. Several groups have reproduced experiments that consistently support a regulatory role of MeCP2 upon BDNF expression. There is also evidence supporting the interference of BDNF upon KCC2 expression. Since BDNF is highly relevant in Rett Syndrome pathology, several pre-clinical and clinical strategies are being designed and tested to improve Rett Syndrome, restoring BDNF levels and physiological actions. Based on the knowledge about a facilitatory effect of adenosine on BDNF actions, we hypothesized that the modulation of BDNF with adenosine receptor agonists would have a positive effect in synaptic function in the Rett Syndrome mouse model. Long-term potentiation (LTP) is accepted as a neurophysiological paradigm to test synaptic plasticity, the basic process underlying learning and memory. Adenosine is a neuromodulator that acts mainly through A1 and A2A receptors. The activation of A2A receptors (A2ARs) potentiates BDNF synaptic actions in healthy animals. Therefore, we explored whether the activation of A2ARs in the Rett Syndrome animal model facilitates BDNF action upon LTP. We found that BDNF facilitatory actions upon LTP are absent in the Rett Syndrome animal model, suggesting that, in addition to BDNF reduction, there is also impairment in its actions, even when it is administered exogenously. This dysfunction could be explained by a reduction in the levels of the main BDNF receptor (the full length tropomyosin-related kinase B), which we describe for the first time in the present study. When BDNF was combined with the selective A2AR agonist, CGS2168, the BDNF effect upon LTP was restored, similar to what was observed in hippocampal slices from wild type animals with BDNF alone. Together these data highlight A2ARs as new possible therapeutic targets to increase BDNF actions in Rett Syndrome. In conclusion, this work contributes to elucidate two significant downstream effects of MeCP2 impairment. The first is the abnormality of GABA postsynaptic actions upon GABAA receptors, suggested by a cerebrospinal fluid proteomic change and corroborated by findings in the Mecp2 knock out animal model and in human neurons, derived from induced pluripotent stem cells of Rett Syndrome patients. This system is now available for pharmacological screening of compounds that target the detected disturbances with direct evaluation of phenotypical rescue, at a cellular level. KCC2 reduction or impaired function appears to have also impact on synaptic structure and plasticity, and is a pathophysiological mechanism that contributes to several neurodevelopmental disorders. The second is the reduction of the main BDNF receptors in the hippocampi of the Rett Syndrome animal model. We have also shown that, using an adenosine A2ARs agonist, it is possible to restore BDNF actions upon LTP, a paradigm for synaptic plasticity. Adenosine, through A2ARs, positively modulates the intracellular signaling cascades activated by BDNF, bypassing the BDNFtropomyosin-related kinase B receptor impairment that occurs in Rett Syndrome.Fundação Calouste Gulbenkian - "The Programme for Advanced Medical Education