Le segment initial de l’axone (SIA) est un sous-domaine fonctionnel du neurone localisé dans l’axone proximal, qui assure deux fonctions : l’initiation du potentiel d’action et le maintien de l’identité axonale. Le maintien et la stabilité du SIA sont des éléments fondamentaux de l’excitabilité du neurone et la nature dynamique de l’organisation fonctionnelle du SIA a été mise en évidence. Les objectifs de mes travaux de thèse ont été d’étudier les mécanismes responsables du maintien du SIA, en condition physiologique ou pathologique et d’identifier de nouveaux acteurs impliqués dans ces mécanismes. Dans un premier temps, nous avons identifié et caractérisé l’expression d’une nouvelle protéine au SIA : la protéine Scrib1. En utilisant une approche par ARN interférent nous avons montré que Scrib1 est nécessaire au maintien de la morphologie du SIA. Les conséquences fonctionnelles de l’absence de Scrib1 sont une diminution de l’excitabilité neuronale. Dans un second temps, nous nous sommes intéressés aux mécanismes pouvant être à l’origine de l’expression ectopique du canal Nav1.8 observée dans certaines pathologies démyélinisantes. Nous avons montré que Nav1.8 possède un site d’interaction à l’ankyrine G. Ce motif d’interaction est suffisant pour adresser un canal chimérique au SIA et perturber l’expression des Nav1 endogènes. A l’inverse des Nav1 du système nerveux central, l’interaction entre Nav1.8 et l’ankyrine G n’est pas régulée par la CK2. Cette interaction constitutive entre Nav1.8 et l’ankyrine G pourrait expliquer son expression ectopique dans le système nerveux central. / The axonal initial segment (AIS) is a unique sub-domain that plays a central role in the physiology of the neuron, as it orchestrates both electrogenesis and the maintenance of neuronal polarity. The maintenance and the stability of the AIS after assembly ensure a reliable generation of action potentials. However, new mechanisms affecting AIS protein-protein interaction and composition have been shown to modulate the electrogenesis of the neuron. Moreover, recent findings highlight that the AIS is capable of homeostatic plasticity through an activity–dependent change either in its location along the proximal axon or in its length. The objectives of my thesis were to study the mechanisms responsible for AIS maintenance in physiological or pathological condition and to identify new players involved in these mechanisms.First we identified and characterized the expression of a novel protein in AIS: the protein Scrib1. Using an shRNA approach we showed that Scrib1 is necessary to maintain the AIS morphology. The functional consequence of the absence of Scrib1 is a decreased of neuronal excitability.Second, we are interested in the mechanisms that cause the ectopic expression of Nav1.8 channel observed in demyelinating diseases. We found that Nav1.8 constitutively interacts with ankG in contrast to Nav1.2, which requires CK2 phosphorylation to bind ankG. Furthermore, when Nav1.8 ankyrin-binding domain was expressed in hippocampal neuron, it clustered at the AIS where it acted as a dominant negative for endogenous Nav1. This constitutive interaction between Nav1.8 and ankG could explain the ectopic expression of Nav1.8 in the central nervous system.

Lors des efflorescences de micro-algues productrices de toxines paralysantes (PST), les bivalves filtreurs peuvent bioaccumuler une grande quantité de toxines et devenir à leur tour toxiques, notamment pour l’homme. La quantité de toxines PST accumulée d’un individu à l’autre s’avère être très variable au sein même d’une population de bivalves. Ainsi, dans nos conditions expérimentales, la quantité de PST accumulées par des huîtres creuses, Crassostrea gigas, d’un même lot, exposées au dinoflagellé toxique Alexandrium minutum, variait d’un facteur 450. L’origine de cette variabilité est inconnue jusqu’alors mais l’une des hypothèses pour l’expliquer serait l’existence de plusieurs formes de canaux sodium voltage-dépendant (NaV), cible des PST, qui confèreraient aux bivalves des sensibilités différentes aux PST. L’objectif principal de cette thèse était de comprendre s’il existe une sensibilité individuelle aux PST différente entre les huîtres et si cette variabilité pouvait être due à des formes différentes de NaV.Une première partie a permis de caractériser le NaV chez C. gigas par une approche de biologie moléculaire. Deux gènes NaV ont été mis en évidence chez C. gigas : CgNaV1, codant un canal sodium et CgNaV2 codant un canal potentiellement sélectif du sodium et du calcium. L’épissage alternatif de CgNaV1 produits trois variants (A, B et C) avec des profils d’expression différents : au niveau des jonctions neuromusculaires pour CgNaV1A, dans les cellules nerveuses pour CgNaV1B et dans les deux pour CgNaV1C. L'acide aminé Q, observé dans le site de liaison aux PST (domaine II) de la séquence CgNaV1 pour les 3 variants et chez tous les individus des 4 populations étudiées, pourrait conférer aux huîtres une certaine résistance aux PST. Ainsi, les variants issus du génotypage/épissage de CgNaV1 ne seraient donc pas le point déterminant du niveau de bioaccumulation des huîtres.Une deuxième partie a permis d’étudier la sensibilité aux PST des nerfs de l’huître creuse C.gigas en relation avec l’accumulation de PST par une approche d’électrophysiologie. La sensibilité à la STX des nerfs cérébroviscéraux d'huîtres a été évaluée en étudiant leur potentiel d'action (CNAP).Il a été montré que les nerfs de C. gigas possédaient une sensibilité à la STX de l’ordre du micromolaire, ce qui leur confère une sensibilité intermédiaire parmi les bivalves. Cette sensibilité des nerfs peut varier selon la période à laquelle les huîtres ont été prélevées et potentiellement selon leur condition physiologique. Une pré-exposition des huîtres à A. minutum semble augmenter la résistance des nerfs à la STX. Cependant, aucune corrélation significative n'a été observée entre la sensibilité nerveuse à la STX et la charge en PST dans la glande digestive des huîtres.Il apparait donc que la variabilité de l’accumulation des PST par les huîtres résulterait plutôt d’une plasticité physiologique, en terme de filtration, d’ingestion et d’assimilation, que d’une sensibilité différentielle des NaV. / During bloom of microalgae producing paralytic shellfish toxins (PST), filtering bivalves can bio-accumulate a large quantity of toxins and become toxic for human consumption. The amount of accumulated PST can greatly vary from one individual to another within a bivalve population. Indeed, under our experimental conditions, the amount of accumulated PST by Pacific oysters, Crassostrea gigas, exposed to the toxic dinoflagellate Alexandrium minutum, varied by a factor of 450. To explain such variability we hypothesized the existence of several forms of voltage-gated sodium channel (NaV), target of the PST, resulting in different sensitivities to PST. The main objective of this thesis was to understand whether there are relationships between nerve sensitivity to PST, the different forms of NaV and the amount of accumulated PST.The NaV was first characterized in C. gigas by a molecular biology approach. Two NaV genes were reported in C. gigas: CgNaV1, encoding a sodium channel and CgNaV2 encoding a channel potentially selective for sodium and calcium. Alternative splicing of CgNaV1 produced three variants (A, B and C) with different expression profiles: at the neuromuscular junctions for CgNaV1A, in the nerve cells for CgNaV1B and in both for CgNaV1C. The amino acid Q observed in the binding site of PST (domain II), of the sequence CgNaV1 for the 3 variants and in all individuals from the 4 studied populations possibly provide some PST resistance to oysters. Thus, the variants resulting from the genotyping/splicing of CgNaV1 would not therefore be the determining factor of the level of bioaccumulation in oysters.A second part allowed studying the nerve sensitivity to PST of C. gigas oyster in relation to the accumulation of PST by an electrophysiology approach. The sensitivity to saxitoxin (STX, a PST) of the cerebro-visceral nerves from oysters was assessed by studying their action potential (CNAP). C.gigas nerves have been shown to have sensitivity to STX of the micromolar range, which gives them intermediate sensitivity among bivalves. This nerve sensitivity may vary depending on the period at which the oysters were collected and potentially according to their physiological condition. A preexposure of oysters to A. minutum appears to increase nerve resistance to STX. However, there was no significant correlation between STX nerve sensitivity and PST content in the oyster digestive gland.Overall, it appears that the variability of the PST accumulation by oysters would result rather from a physiological plasticity, in terms of filtration, ingestion and assimilation, than from a differential sensitivity of the NaV.

>Magister Scientiae - MSc / This study focused on the synthesis of a series of novel tricycloundecane derivatives and evaluation of these compounds for neuroprotection using the fluorescent ratiometric calcium assay that indicates the ability of the test compounds to inhibit NMDA receptors and VGCC. The cycloaddition reaction between p-benzoquinone and monomerised dicyclopentadiene yielded tricycloundeca- 4,9-diene-3,6-dione which was used as the base structure and further derivatised. These derivatives were conjugated with benzylamine to form a series of imines and amines. A total of 10 compounds were synthesised for evaluation of inhibition of calcium influx through NMDA receptor channels and voltage-gated calcium channels. The structures were confirmed using NMR, IR and MS. On the proton NMR, the characteristic AB-quartet system was observed in the region of 1-2 ppm for all the compounds and the aromatic moiety was observed between 6.5-7.5 ppm for the novel polycyclic amines. These, with other functional groups, were used to confirm the individual structures

Systematic investigations through several experimental techniques have revealed that hippocampal pyramidal neurons express voltage gated ion channels (VGICs) with well-defined gradients along their dendritic arbor. These actively maintained gradients in various dendritic VGICs effectuate several stereotypic, topographically continuous functional gradients along the topograph of the dendritic arbor, and have been referred to as intraneuronal functional maps. The prime goal of my thesis was to understand the emergence and homeostasis of the several coexistent functional maps that express within hippocampal pyramidal neurons. In the first part of the thesis, we focus only on spatial interactions between ion channels and analyzed the role of such interactions in the emergence of functional maps. We developed a generalized quantitative framework, the influence field, to analyze the extent of influence of a spatially localized VGIC cluster. Employing this framework, we showed that a localized VGIC cluster could have spatially widespread influence, and was heavily reliant on the specific physiological property and background conductances. Using the influence field model, we reconstructed functional gradients from VGIC conductance gradients, and demonstrated that the cumulative contribution of VGIC conductances in adjacent compartments plays a critical role in determining physiological properties at a given location. These results suggested that spatial interactions among spatially segregated VGIC clusters are necessary for the emergence of the functional maps. In the second part of the thesis, we assessed the specific roles of only kinetic interactions between ion channels in determining physiological properties by employing a single-compartmental model. In doing this, we analyzed the roles of interactions among several VGICs in regulating intrinsic response dynamics. Using global sensitivity analysis, we showed that functionally similar models could be achieved even when underlying parameters displayed tremendous variability and exhibited weak pair-wise correlations. These results suggested that that response homeostasis could be achieved through several non-unique channel combinations, as an emergent consequence of kinetic interactions among these channel conductances. In the final part of the thesis, we analyzed the combined impact of both spatial and kinetic interactions among ion channel conductances on the emergence and homeostasis of functional maps in a neuronal model endowed with extensive dendritic arborization. To do this, we performed global sensitivity analysis on morphologically realistic conductance-based models of hippocampal pyramidal neurons that coexpressed six functional maps. We found topographically continuous functional maps to emerge from disparate model parameters with weak pair-wise correlations between parameters. These results implied that individual channel properties need not be set at constant values in achieving overall homeostasis of several coexistent functional maps. We suggest collective channelostasis, where several channels regulate their properties and expression profiles in an uncorrelated manner, as an alternative for accomplishing functional map homeostasis. Finally, we developed a methodology to assess the contribution of individual channel conductances to the various functional measurements employing virtual knockout simulations. We found that the deletion of individual channels resulted in variable, measurement-and location-specific impacts across the model population.

Les canaux sodiques dépendants du voltage, ou canaux Nav, jouent un rôle capital dans l'excitabilité neuronale, dans la genèse et dans la propagation des potentiels d'action. Le canal Nav1.9 se distingue par une expression restreinte aux nocicepteurs et par des propriétés électrophysiologiques uniques qui, si elles excluent sa contribution à la phase dépolarisante du potentiel d'action, lui confèreraient un rôle dans la modulation de l'excitabilité des nocicepteurs. Ce travail de thèse vise à caractériser son implication dans la physiopathologie de la douleur par une approche comportementale, moléculaire et fonctionnelle. La première partie de ce travail consiste à étudier la contribution du canal Nav1.9 à la douleur inflammatoire. Nous avons donc réalisé différents tests comportementaux chez des souris knock-out (KO) et des rats traités par antisens (knock-down) modèles de douleur inflammatoire (aigu, subaigu, chronique). L'expression du canal ainsi que ses propriétés électrophysiologiques sont ensuite analysées chez ces mêmes modèles animaux. Notre premier constat est que le canal Nav1.9 n'est pas impliqué dans la réponse à une stimulation mécanique ou thermique chaude nociceptive chez des animaux sains. En revanche, l'hypersensibilité douloureuse thermique et mécanique induite par une inflammation subaiguë (carragénine intraplantaire) ou chronique (monoarthrite) est significativement réduite chez la souris KO Nav1.9. Un résultat similaire est obtenu par traitement antisens chez le rat, sur le modèle d'inflammation subaiguë. Chez la souris, suite à l'induction d'une inflammation subaiguë, une légère diminution suivie d'une forte augmentation de l'expression protéique du canal Nav1.9 est observée dans les ganglions rachidiens innervant la patte enflammée. Une augmentation de la quantité de canaux est également observée au niveau des troncs nerveux cutanés innervant cette même zone. Les canaux néosynthétisés ne contribuent pas au courant sodique enregistré en patch clamp dans les corps cellulaires des neurones des ganglions rachidiens, mais nos données suggèrent qu'ils sont exportés en direction des terminaisons nerveuses, où ils pourraient devenir fonctionnels et augmenter l'excitabilité cellulaire. La deuxième partie de ce travail de thèse consiste à caractériser l'implication de canal Nav1.9 dans la perception du froid et dans l'hypersensibilité au froid induite par l'oxaliplatine. Nous avons en effet observé de manière inattendue que les souris KO Nav1.9 présentent des seuils de douleur au froid (<10°C) plus élevés que les souris sauvages. Ce phénomène est confirmé par plusieurs tests comportementaux chez les souris KO et chez des rats traités par antisens anti-Nav1.9. L'oxaliplatine, prescrit dans le traitement des cancers colorectaux, est connu pour induire une hypersensibilité au froid invalidante chez la majorité des patients. Nous avons donc décidé d'étudier la contribution du canal Nav1.9 à ce symptôme. Suite à une injection unique d'oxaliplatine, une forte hypersensibilité au froid apparait chez les souris dès 20°C. Nous montrons que le KO Nav1.9 permet de supprimer l'hypersensibilité au froid aux températures normalement non douloureuses (20 et 15°C, allodynie), et de réduire l'hypersensibilité aux températures douloureuses (10 et 5°C, hyperalgie). Le même effet est observé chez le rat après traitement antisens. En conclusion, ce travail permet de mettre en évidence l'intérêt du canal Nav1.9 en tant que cible pharmacologique potentielle pour le traitement de douleurs inflammatoires et de l'hypersensibilité au froid induite par l'oxaliplatine. Il est de plus intéressant de constater que les seuils de réponse à des stimuli nociceptifs ne sont pas perturbés chez les souris KO Nav1.9 saines, à l'exception de la douleur provoquée par des températures froides extrêmes. Le blocage du canal Nav1.9 aurait donc des propriétés anti-hyperalgiques plutôt qu'antalgique, ce qui est conceptuellement intéressant. / Voltage-gated sodium channels, or Nav channels, play a key role in neuronal excitability and in the emission and propagation of action potentials. Among the different Nav isoforms, Nav1.9 is only expressed in nociceptors and shows atypical electrophysiological properties which, if they exclude a possible contribution to the depolarizing phase of the action potential, could be important for the modulation of nociceptors' excitability. This study aims to characterize the Nav1.9 implication in the pathophysiology of pain using behavioral, molecular and functional approaches. The first part of this work is to assess the Nav1.9 contribution to inflammatory pain. Therefore we have performed several behavioral tests in different inflammatory pain models (acute, subacute, chronic), using knock-out (KO) mice and rats treated with antisense oligodeoxynucleotides. Nav1.9 expression and electrophysiological properties are then analyzed within the same animal models. First, we observe that Nav1.9 channels do not contribute to pain perception in response to noxious heat or pressure in healthy animals. However, thermal and mechanical pain hypersensitivity induced by subacute (intraplantar carrageenan) or chronic (monoarthritis) inflammation is significantly lowered in Nav1.9 knock-out mice. Similar results are obtained on the subacute inflammation model using a knock-down strategy in rats. A weak reduction followed by a strong increase in Nav1.9 protein expression is observed in mice dorsal root ganglions innervating the inflamed paw during subacute inflammation. We also observe an increase in Nav1.9 immunolabeling in cutaneous nerve trunks innervating this zone. Whereas the newly produced channels do not contribute to the sodium current recorded in dorsal root ganglion cell bodies, as assessed by patch clamp, our data suggest that they are transported to nerve terminals where they could become functional and increase neuronal excitability. In the second part of this study, we aim to characterize the implication of Nav1.9 channels in cold perception and in oxaliplatin-induced cold hypersensitivity. Indeed, we surprisingly observed that Nav1.9 KO mice showed higher pain thresholds to intense cold (<10°C) than wild-type mice. This observation is confirmed by several behavioral tests in KO mice and in antisense-treated rats. As oxaliplatine (a platinum salt used to treat colorectal cancer) is known to induce cold pain hypersensitivity in most of the patients, we decided to study the Nav1.9 contribution to this symptom. Following acute oxaliplatin injection, a strong cold hypersensitivity is observed in wild-type mice at 20°C and below. We show that Nav1.9 KO results in a suppression of cold hypersensitivity to non-noxious temperatures (20 and 15°C, allodynia), and a reduction of hypersensitivity to noxious cold (10 and 5°C, hyperalgesia). A similar observation is made using Nav1.9 knock-down in rats. To conclude, our data shows that Nav1.9 could be potentially a good target to treat acute to chronic inflammatory pain, as well as oxaliplatin-induced cold hypersensitivity. Furthermore, as Nav1.9 is not involved in defining pain thresholds of healthy animals (except for noxious cold), its blockade would have anti-hyperalgesic rather than analgesic effects, which is conceptually interesting.

A natural phenomenon characterized by dense aggregations of unicellular photosynthetic marine organisms has been termed colloquially as red tides because of the vivid discoloration of the water. The dinoflagellate Karenia brevis is the cause of the Florida red tide bloom. K. brevis produces the brevetoxins, a potent suite of neurotoxins responsible for substantial amounts of marine mammal and fish mortalities. When consumed by humans, the toxin causes Neurotoxic Shellfish Poisoning (NSP). The native function of brevetoxin within the organism has remained mysterious since its discovery. There is a need to identify factors which contribute to and regulate toxin production within K. brevis. These toxins are produced and retained within the cell implicating a significant cellular role for their presence. Localization of brevetoxin and identification of a native receptor may provide insight into its native role as well as other polyether ladder type toxins such as the ciguatoxins, maitotoxins, and yessotoxins. In higher organisms these polyether ladder molecules bind to transmembrane proteins with high affinity. We anticipated the native brevetoxin receptor would also be a transmembrane protein. Photoaffinity labeling has become increasingly popular for identifying ligand receptors. By attaching ligands to these photophors, one is able to activate the molecule after the ligand binds to its receptor to obtain a permanent linkage between the two. Subsequent purification provides the protein with the ligand directly attached. A molecule that is capable of fluorescence is a fluorophore, which upon excitation is capable of re-emitting light. Fluorescent labeling uses fluorophores by attaching them covalently to biologically active compounds. The synthesis of a brevetoxin photoaffinity probe and its application in identifying a native brevetoxin receptor will be described. The preparation of a fluorescent derivative of brevetoxin will be described and its use in localizing the toxin to an organelle within K. brevis. In addition, the general utility of a synthesized photoaffinity label with other toxins having similar functionality will be described. An alternative synthetic approach to a general photoaffinity label will also be discussed whose goal was to accelerate the preparation and improve the overall synthetic yields of a multifunctional label.

Bilayer embedded proteins are affected by stress. This general affirmation is, in this thesis, embodied by two types of proteins: membrane spandex and voltage-gated sodium channels. In this work, we essentially explore, using methods from physics, the theoretical consequences of ideas drawn from experimental biology. Membrane spandex was postulated to exist and we study the theoretical implications and possible benefits for a cell to have such proteins embedded in its bilayer. There are no specific membrane spandex proteins, rather any protein with a transition involving a large enough area change between two non-conducting states could act as spandex. Bacterial cells have osmovalve channels which open at near-lytic tensions to protect themselves against rupture. Spandex expanding at tensions just below the osmovalves’ opening tension could relieve tension enough as to avoid costly accidental osmovalve opening due to transient bilayer tension excursions. Another possible role for spandex is a tension-damper: spandex could be used to maintain bilayer tension at a fixed level. This would be useful as many bilayer embedded channels are known to be modulated by tension. The Stress/shear experienced in traumatic brain injury cause an immediate (< 2 min) and irreversible TTX-sensitive rise in axonal calcium. In situ, this underlies an untreatable condition, diffuse axonal injury. TTX sensitivity indicates that leaky voltage-gated sodium (Nav) channels mediate the calcium increase. Wang et al. showed that the mammalian adult CNS Nav isoform, Nav1.6, expressed in Xenopus oocytes becomes “leaky” when subjected to bleb-inducing pipette aspiration. This “leaky” condition is caused by a hyperpolarized-shift (left-shift or towards lower potentials, typically 20 mV) of the kinetically coupled processes of activation and inactivation thus effectively degrading a well-confined window conductance into a TTX-sensitive Na leak. We propose experimental protocols to determine whether this left-shift is the result of an all-or-none or graded process and whether persistent Na currents are also left-shifted by trauma. We also use modeling to assess whether left-shifted Nav channel kinetics could lead to Na+ (and hence Ca2+ ) loading of axons and to study saltatory propagation after traumatizing a single node of Ranvier.

This thesis focuses on the structural and functional characterization of two integral membrane proteins; CfrA, an outer membrane TonB-dependent transporter (TBDT) from Campylobacter jejuni, and ELIC, a pentameric ligand-gated ion channel (pLGIC) from Erwinia Chrysanthemi. The spectroscopic characterization of CfrA revealed a fold consistent with the structural and biophysical properties observed for other TBDT. Both a homology model of CfrA and sequence alignments of CfrA with other ferric-enterobactin transporters suggested a unique mode of ligand binding, thus raising the possibility that C. jejuni can be specifically inhibited. To investigate the molecular determinates of binding to CfrA, I set out to crystallize CfrA. Hundreds of crystal trials led to crystals diffracting to 3.6 Å resolution, with a complete data set acquired at 5 Å resolution that led to a structural model of the CfrA β-barrel. In the second part of this thesis, I reconstituted ELIC into model membranes in order to test the role of intramembrane aromatic interactions in ELIC gating and lipid sensing. ELIC was reconstituted into both asolectin (aso-ELIC) and 1-palmitoyl-2-oleoyl phosphatidylcholine (PC-ELIC), membranes that stabilize the homologous nicotinic acetylcholine receptor (nAChR) in functional coupled versus non-functional uncoupled conformations, respectively. In both membrane environments, ELIC exhibits a mixed α-helical and β-sheet secondary structure, with a thermal denaturation intermediate between those of the nAChR and the close prokaryotic homolog, GLIC, in similar membranes. The data suggest that although ELIC has a decreased propensity to adopt an uncoupled conformation relative to the nAChR, its ability to undergo cysteamine-induced channel gating is sensitive to its lipid environment. The decreased propensity to uncouple may reflect an increased level of aromatics at the interface between the transmembrane α-helices, M1, M3, and M4. To test this hypothesis further, the level or aromatic residues at the M1, M3, and M4 interface in both GLIC and ELIC were varied, and in both cases the levels of intramembrane aromatic interactions correlated with the efficiency of coupling binding to gating. The data provide further evidence for a role of intramembrane aromatics in channel gating and in dictating the propensity of pentameric ligand-gated ion channels to adopt an uncoupled conformation.

Les canaux potassiques (Kv) dépendants du voltage sont des protéines transmembranaires qui permettent le flux passif d’ions potassium à travers une membrane plasmique lorsque celle-ci est dépolarisée. Ils sont constitués de quatre domaines périphériques sensibles au voltage et un domaine central, un pore, qui délimite un chemin hydrophile pour le passage d’ions. Les domaines sensibles à la tension (VSD) et le pore sont couplés, ce qui signifie que l’activation des VSD déclenche l’ouverture du pore, et qu’un pore ouvert favorise l’activation des VSD. Le phosphatidylinositol-4,5-bisphosphate (PIP2) est un lipide mineur du feuillet interne de la membrane plasmique. Ce lipide fortement chargé négativement module le fonctionnement de plusieurs canaux ioniques, y compris les membres de la famille Kv. En particulier, l’application de ce lipide à Kv1.2 et Kv7.1, deux canaux homologues, augmente leur courant ionique. Cependant, alors que Kv1.2 est capable de s’ouvrir en l’absence de PIP2, dans le cas de Kv7.1, ce lipide est absolument nécessaire pour l’ouverture du canal. En outre, dans Kv1.2, PIP2 induit une perte de fonction, qui est manifesté par un mouvement retardé des VSD. Jusqu’à présent, les mécanismes sous-jacents à de telles modulations des canaux Kv par PIP2 restent inconnus. Dans ce travail, nous tentons de faire la lumière sur ces mécanismes en utilisant des simulations de dynamique moléculaire (DM) combinées avec une approche expérimentale, entreprise par nos collaborateurs. En utilisant des simulations de DM sans contrainte, nous avons identifié les sites potentiels de liaison du PIP2 au Kv1.2. Dans l’un de ces sites, PIP2 interagit avec le canal de sorte à former des ponts salins dépendants de l’état du canal, soit avec le VSD soit avec le pore. Sur la base de ces résultats, nous proposons un modèle pour rationaliser les données expérimentales connues. En outre, nous avons cherché à évaluer quantitativement la perte de fonction induite par la présence de PIP2 au voisinage du VSD du Kv1.2. En particulier, nous avons calculé l’énergie libre des deux premières transitions le long de l’activation du VSD en présence et en l’absence de ce lipide. Nous avons constaté que PIP2 affecte à la fois la stabilité relative des états du VSD et les barrières d’énergie libre qui les séparent. Enfin, nous avons étudié les interactions entre PIP2 et un autre membre de la famille Kv, le canal Kv7.1 cardiaque. Dans le site de liaison de PIP2 que nous avons identifié pour ce canal, l’interaction entre les résidus positifs de Kv7.1 et le lipide sont dépendants de l’état du VSD, comme dans le cas de Kv1.2. On montre que cette interaction est importante pour le couplage entre les VSD et le pore, couplage qui est par ailleurs affaibli à cause de la répulsion électrostatique entre quelques résidus positifs. Ces résultats et prédictions ont été vérifiés par les données expérimentales obtenues par nos collaborateurs / Voltage-gated potassium (Kv) channels are transmembrane proteins that enable the passive flow of potassium ions across a plasma membrane when the latter is depolarized. They consist of four peripheral voltage sensor domains, responding to the applied voltage, and a central pore domain that encompasses a hydrophilic path for passing ions. The voltage sensors and the pore are coupled, meaning that the activation of the voltage sensors triggers the pore opening, and the open pore promotes the activation of the voltage sensors. Phosphatidylinositol-4,5-bisphosphate (PIP2) is a minor lipid of the inner plasma membrane leaflet. This highly negatively charged lipid was shown to modulate the functioning of several ion channels including members of the Kv family. In particular, application of this lipid to Kv1.2 and Kv7.1, two homologous channels, enhances their ionic current. However, while Kv1.2 is able to open without PIP2, in the case of Kv7.1, this lipid is absolutely required for opening. Additionally, in Kv1.2, PIP2 induces a loss of functioning, which is manifested by delayed motions of the voltage sensors. So far, the mechanism underlying the Kv channels modulation by PIP2 remains unknown. In the present manuscript, we attempt to shed light on this mechanism using molecular dynamics (MD) simulations combined with experiments, which was undertaken by our collaborators. Using unconstrained MD simulations, we have identified potential PIP2 binding sites in Kv1.2. In one of these sites, PIP2 interacts with the channel in a state-dependent manner forming salt bridges either with the voltage sensor or with the pore. Based on these findings, we propose a model rationalizing the known experimental data. Further, we aimed to estimate the loss of functioning effect induced by PIP2 on the Kv1.2 voltage sensors. In particular, we have calculated the free energy of the first two transitions along the activation path in the presence and absence of this lipid. We found that PIP2 affects both the relative stability of the voltage sensor states and the free energy barriers separating them. Finally, we studied the interactions between PIP2 and another member of the Kv family, the cardiac channel Kv7.1. In the PIP2 binding site that we have identified for this channel, the interaction between positive residues of Kv7.1 and the lipid was state-dependent, as in the case of Kv1.2. This state-dependent interaction, however, is prominent for coupling between the voltage sensors and the pore, which is otherwise weakened due to electrostatic repulsion of some positive residues. These findings are in a good agreement with the experimental data obtained by our collaborators

Pentameric ligand-gated ion channels (pLGICs) are membrane-embedded receptors found extensively in pre- and post-synaptic membranes throughout the nervous system where they play an important role in neurotransmission. The function of the prototypic pLGIC, the nicotinic acetylcholine receptor (nAChR) is highly sensitive to changes in its lipid environment, while other pLGICs display varying lipid sensitivities. This thesis presents a multidisciplinary investigation into the features of the transmembrane domain (TMD) that determine the unique functional and physical traits of different pLGICs. Using two prokaryotic homologues of the nAChR, ELIC and GLIC, as models, I focus on the outermost, lipid-exposed α-helix, M4, which, despite being distant from the primary allosteric pathway coupling agonist binding to channel gating, exercises significant control over channel function. Here, I present evidence that M4 acts as a lipid sensor, detecting changes in the surrounding lipids and transmitting these changes to the channel pore via contacts with the adjacent TMD α-helices, M1 and M3, and/or with structures in the extracellular domain. Using ELIC and GLIC chimeras, I first show that the TMD is the main driver of pLGIC thermal stability. I then demonstrate that the M4 α-helices in each channel play different roles in channel maturation and function, which suggests a divergent evolutionary path. Following this, I show that the M4 C-terminus is essential to both maturation and function in GLIC, while in ELIC its role is less defined, again showcasing possible evolutionary differences. Building on these findings, I examined the role of aromatic residues at the M4 – M1/M3 interface, and found that they predictably determine the interactions between M4 and M1/M3. Notably, the addition of aromatic residues to enhance M4-M1/M3 interactions in ELIC promotes channel function, while the elimination of aromatic residues at the M4-M1/M3 interface in GLIC is detrimental to channel function. Furthermore, I show that these same aromatics alter the strength of pLGIC lipid sensing and the sensitivity to certain disease-causing mutations, both indicating that aromatic residues are key players in channel function, stability and modulation. Finally, I and my collaborators identified and characterized a novel desensitization-linked lipid binding site in ELIC. Extensive mutagenesis studies coupled with biophysical measurements allowed us to develop a model describing how lipid binding influences the rates of ELIC desensitization to shape the agonist-induced response.