No description available.

Doctor of Philosophy (Medicine) / Background: Diabetic vascular complications affect both the macro- and microvasculature. Microvascular pathology in diabetes may be mediated by biochemical factors that precipitate cellular changes at both the gene and protein levels. In the diabetic retina, vascular pathology is found mainly in microvessels, including the retinal precapillary arterioles, capillaries and venules. Macular oedema secondary to breakdown of the inner blood-retinal barrier is the most common cause of vision impairment in diabetic retinopathy. Müller cells play a critical role in the trophic support of retinal neurons and blood vessels. In chronic diabetes, Müller cells are increasingly unable to maintain their supportive functions and may themselves undergo changes that exacerbate the retinal pathology. The consequences of early diabetic changes in retinal cells are primarily considered in this thesis. Aims: This thesis aims to investigate the effect of perivascular cells (Müller cells, RPE, pericytes) on retinal endothelial cell permeability using an established in vitro model. Methods: Immunohistochemistry, cell morphology and cell growth patterns were used to characterise primary bovine retinal cells (Müller cells, RPE, pericytes and endothelial cells). An in vitro model of the blood-retinal barrier was refined by coculturing retinal endothelial cells with perivascular cells (Müller cells or pericytes) on opposite sides of a permeable Transwell filter. The integrity of the barrier formed by endothelial cells was assessed by transendothelial electrical resistance (TEER) measurements. Functional characteristics of endothelial cells were compared with ultrastructural morphology to determine if different cell types have barrier-enhancing effects on endothelial cell cultures. Once the co-culture model was established, retinal endothelial cells and Müller cells were exposed to different environmental conditions (20% oxygen, normoxia; 1% oxygen, hypoxia) to examine the effect of perivascular cells on endothelial cell permeability under reduced oxygen conditions. Barrier integrity was assessed by TEER measurements and permeability was measured by passive diffusion of radiolabelled tracers from the luminal to the abluminal side of the endothelial cell barrier. A further study investigated the mechanism of laser therapy on re-establishment of retinal endothelial cell barrier integrity. Müller cells and RPE, that comprise the scar formed after laser photocoagulation, and control cells (Müller cells and pericytes, RPE cells and ECV304, an epithelial cell line) were grown in long-term culture and treated with blue-green argon laser. Lasered cells were placed underneath confluent retinal endothelial cells growing on a permeable filter, providing conditioned medium to the basal surface of endothelial cells. The effect of conditioned medium on endothelial cell permeability was determined, as above. Results: Co-cultures of retinal endothelial cells and Müller cells on opposite sides of a permeable filter showed that Müller cells can enhance the integrity of the endothelial cell barrier, most likely through soluble factors. Low basal resistances generated by endothelial cells from different retinal isolations may be the result of erratic growth characteristics (determined by ultrastructural studies) or the selection of vessel fragments without true ‘barrier characteristics’ in the isolation step. When Müller cells were co-cultured in close apposition to endothelial cells under normoxic conditions, the barrier integrity was enhanced and permeability was reduced. Under hypoxic conditions, Müller cells had a detrimental effect on the integrity of the endothelial cell barrier and permeability was increased in closely apposed cells. Conditioned medium from long-term cultured Müller cells and RPE that typically comprise the scar formed after lasering, enhanced TEER and reduced permeability of cultured endothelial cells. Conclusions: These studies confirm that bovine tissues can be used as a suitable model to investigate the role of perivascular cells on the permeability of retinal endothelial cells. The dual effect of Müller cells on the retinal endothelial cell barrier under different environmental conditions, underscores the critical role of Müller cells in regulating the blood-retinal barrier in health and disease. These studies also raise the possibility that soluble factor(s) secreted by Müller cells and RPE subsequent to laser treatment reduce the permeability of retinal vascular endothelium. Future studies to identify these factor(s) may have implications for the clinical treatment of macular oedema secondary to diseases including diabetic retinopathy.

Doctor of Philosophy (Medicine) / Background: Diabetic vascular complications affect both the macro- and microvasculature. Microvascular pathology in diabetes may be mediated by biochemical factors that precipitate cellular changes at both the gene and protein levels. In the diabetic retina, vascular pathology is found mainly in microvessels, including the retinal precapillary arterioles, capillaries and venules. Macular oedema secondary to breakdown of the inner blood-retinal barrier is the most common cause of vision impairment in diabetic retinopathy. Müller cells play a critical role in the trophic support of retinal neurons and blood vessels. In chronic diabetes, Müller cells are increasingly unable to maintain their supportive functions and may themselves undergo changes that exacerbate the retinal pathology. The consequences of early diabetic changes in retinal cells are primarily considered in this thesis. Aims: This thesis aims to investigate the effect of perivascular cells (Müller cells, RPE, pericytes) on retinal endothelial cell permeability using an established in vitro model. Methods: Immunohistochemistry, cell morphology and cell growth patterns were used to characterise primary bovine retinal cells (Müller cells, RPE, pericytes and endothelial cells). An in vitro model of the blood-retinal barrier was refined by coculturing retinal endothelial cells with perivascular cells (Müller cells or pericytes) on opposite sides of a permeable Transwell filter. The integrity of the barrier formed by endothelial cells was assessed by transendothelial electrical resistance (TEER) measurements. Functional characteristics of endothelial cells were compared with ultrastructural morphology to determine if different cell types have barrier-enhancing effects on endothelial cell cultures. Once the co-culture model was established, retinal endothelial cells and Müller cells were exposed to different environmental conditions (20% oxygen, normoxia; 1% oxygen, hypoxia) to examine the effect of perivascular cells on endothelial cell permeability under reduced oxygen conditions. Barrier integrity was assessed by TEER measurements and permeability was measured by passive diffusion of radiolabelled tracers from the luminal to the abluminal side of the endothelial cell barrier. A further study investigated the mechanism of laser therapy on re-establishment of retinal endothelial cell barrier integrity. Müller cells and RPE, that comprise the scar formed after laser photocoagulation, and control cells (Müller cells and pericytes, RPE cells and ECV304, an epithelial cell line) were grown in long-term culture and treated with blue-green argon laser. Lasered cells were placed underneath confluent retinal endothelial cells growing on a permeable filter, providing conditioned medium to the basal surface of endothelial cells. The effect of conditioned medium on endothelial cell permeability was determined, as above. Results: Co-cultures of retinal endothelial cells and Müller cells on opposite sides of a permeable filter showed that Müller cells can enhance the integrity of the endothelial cell barrier, most likely through soluble factors. Low basal resistances generated by endothelial cells from different retinal isolations may be the result of erratic growth characteristics (determined by ultrastructural studies) or the selection of vessel fragments without true ‘barrier characteristics’ in the isolation step. When Müller cells were co-cultured in close apposition to endothelial cells under normoxic conditions, the barrier integrity was enhanced and permeability was reduced. Under hypoxic conditions, Müller cells had a detrimental effect on the integrity of the endothelial cell barrier and permeability was increased in closely apposed cells. Conditioned medium from long-term cultured Müller cells and RPE that typically comprise the scar formed after lasering, enhanced TEER and reduced permeability of cultured endothelial cells. Conclusions: These studies confirm that bovine tissues can be used as a suitable model to investigate the role of perivascular cells on the permeability of retinal endothelial cells. The dual effect of Müller cells on the retinal endothelial cell barrier under different environmental conditions, underscores the critical role of Müller cells in regulating the blood-retinal barrier in health and disease. These studies also raise the possibility that soluble factor(s) secreted by Müller cells and RPE subsequent to laser treatment reduce the permeability of retinal vascular endothelium. Future studies to identify these factor(s) may have implications for the clinical treatment of macular oedema secondary to diseases including diabetic retinopathy.

Purpose. These experiments were designed to characterize tight junction formation by retinal pigment epithelial (RPE) cells in vitro and to compare the effects on this process of hormonally defined medium (HDM) and serum- containing medium. Methods. Formation of RPE tight junctions was analyzed in freshly isolated rat RPE cells maintained either in HDM or serum-containing medium. Junctions were evaluated functionally by measuring transepithelial electrical resistance (TER) and permeability and structurally by immunolocalization of the junction-associated actin microfilaments. Calcium dependency of the junction was determined by reducing media calcium concentration. Results. RPE cells cultured in serum-free HDM developed calcium-dependent tight junctions, which exhibited TER levels > 150 Ω · cm 2 and low paracellular permeability. Serum-containing media inhibited tight junction formation as indicated by significant reductions in TER and increases in permeability. Junction-associated actin microfilaments and cell density were unchanged. Conclusions. Tight junction formation by RPE cells is inhibited by serum. This activity may play an important role in responses of the RPE layer to injury, contributing to the pathologic progression of blood- retinal barrier dysfunction.

Purpose. In the genetically defective Royal College of Surgeons (RCS) rat model for retinal degeneration, a breakdown occurs in the retinal pigment epithelial (RPE) cell tight junctions just as the photoreceptors begin to degenerate. These experiments sought to determine the impact of the RPE genetic defect on this alteration in the RPE cell tight junctions. Methods. Retinal pigment epithelial cell cultures prepared from RCS and control rats were treated with hormonally defined medium (HDM), base medium conditioned by RCS or control retinas, or unconditioned base medium. The tight junctions formed by these cultures were assayed functionally by measuring transepithelial electrical resistance and permeability. Junction structure was evaluated by immunolocalization of the tight junction protein zonula occludens I and of the junction-associated actin microfilaments. Results. Retinal pigment epithelial cultures from dystrophic rats formed structurally and functionally normal tight junctions when maintained in hormonally defined medium. The junctions remained stable when the medium bathing the apical surface was switched to base medium preconditioned by normal retinas. In contrast, cultures treated with medium preconditioned by degenerating dystrophic retinas or with unconditioned medium exhibited a breakdown in their tight junctions. Conclusions. Retinal pigment epithelial cells isolated from dystrophic RCS rats can form tight junctions normally in vitro. Normal, but not dystrophic, retinas release factors that support RPE tight junctions. Therefore, the junctional abnormality seen in dystrophic rat RPE cells in vivo is probably caused by the loss of trophic factors normally provided by the healthy neural retina rather than by a direct effect of the genetic defect on the tight junctions.

La rétinopathie diabétique (RD) se compose d’une part d’une ischémie rétinienne périphérique et d’autre part d’une exsudation rétinienne responsable d’un œdème maculaire diabétique, première cause de cécité chez les moins 55 ans. Les traitements utilisés actuellement sont non spécifiques et traitent les complications tardives de la RD. Les phases précoces de la RD ne sont donc pas ciblées. L’hyperglycémie chronique entraine un stress oxydant et une activation des PKC qui participent à l’altération des BHR. L’objectif de ce travail a été 1°) d’étudier l’implication de la PKCζ et de la voie Rho/ROCK/Myosine II sur la physiopathogénie de la RD et 2°) de montrer l’effet bénéfique de leurs inhibiteur sur les BHR et sur la reperfusion des capillaires rétiniens. Nous avons confirmé l’hyperactivation de la PKCζ et de la voie Rho/ROCK/Myosine II chez les rats diabétiques et leur participation à la rupture de la BHR externe. Le traitement par leurs inhibiteurs respectifs normalise l’activation des deux enzymes et restaure l’intégrité anatomique et fonctionnelle de la BHR externe. De plus l'hyperactivation de ROCK altère la perfusion rétinienne par 1) constriction focale artériolaire, 2) protrusions membranaires endoluminales des cellules endothéliales (blebbing) et 3) vasoconstriction capillaire diffuse. Nous avons montré que l'ensemble de ces phénomènes étaient réversibles par traitement intravitréen de son inhibiteur le Fasudil. De manière importante le traitement par Fasudil induit également une diminution du VEGF rétinien responsable de la perméabilité des barrières et témoin indirect de l’ischémie rétinienne. Ces travaux éclairent la physiopathogénie de la RD et ouvre des perspectives thérapeutiques permettant de cibler les événements précoces de la RD. / Diabetic retinopathy (DR) mainly results from peripheral retinal ischemia and exudation leading to sight threatening complications such as retinal neovascularization or macular edema. This latter represents the main cause of visual loss among working age individuals. Current treatments address late complications of DR and are non-specific. Therefore, early events are currently not addressed. Chronic hyperglycemia increases oxidative stress and activates PKC leading to blood retinal barrier (BRB) breakdown. The aims of the present work were two fold. First, to assess the implication of PKCζ and the Rho/ROCK/Myosin II pathway on the pathogenesis of DR and second, to assess whether their specific inhibitors have the potential to restore the phenotype. Herein we have demonstrated the pathogenic role of PCKζ and ROCK hyperactivation on the development of diabetes induced external BRB breakdown. Furthermore their inhibitors restored the morphologic and functional aspect of the external BRB. We also found that ROCK hyperactivation was responsible for altered retinal perfusion through several mechanism namely 1) focal constriction of retinal arterioles; 2) endoluminal protrusions of the endothelial cell membrane (blebs) and 3) capillary diffuse vasoconstriction. We were able to demonstrate that all this aspects were reversible by Fasudil, a ROCK inhibitor, administrated into the vitreous. Of importance this treatment decreased also retinal VEGF that is a well-known factor responsible for barrier breakdown and a surrogate marker for retinal ischemia. To conclude the present findings not only shed light on the mechanisms of DR but also open new therapeutic avenues addressing the early events of DR a current unmet medical need.

Thesis (Ph. D.)--University of Oklahoma. / Bibliography: leaves 164-183.

Le système nerveux central est un organe privilégié et protégé, notamment grâce à l’existence des barrières histologiques entre le sang et les tissus nerveux. La barrière-hémato encéphalique (BHE) et la barrière hémato-rétinienne (BHR) séparent respectivement le parenchyme cérébral et la rétine des composés contenus dans l’espace vasculaire, grâce à l’expression de jonctions serrées et de transporteurs membranaires permettant une régulation spécifique des échanges entre le sang et le parenchyme nerveux. Ce travail a porté sur l’étude d’un nouveau transporteur de cations organiques mis en évidence fonctionnellement à la BHE de la souris. Ce transporteur appartenant très probablement à la superfamille des solute carrier (SLC), fonctionne comme un antiport proton. Actuellement, sa présence ne peut être démontrée que de façon fonctionnelle car son identité moléculaire est encore inconnue. Cet antiport proton constitue un nouvel acteur de la perméabilité cérébrale et ouvre une nouvelle voie d’accès au cerveau. Nous nous sommes tout d’abord attachés à approfondir les connaissances fonctionnelles de ce transporteur en étudiant de nouveaux substrats et tissus d’expression. Le transport cérébral de psychotropes a été étudié in vivo par la technique de perfusion carotidienne in situ chez la souris et in vitro grâce à une lignée de cellules endothéliales cérébrales humaines immortalisées (hCMEC/D3). Nous avons démontré que la haute perméabilité cérébrale de la cocaïne fait intervenir à la fois une diffusion passive et surtout une diffusion médiée par un antiport proton. La vitesse d’entrée des substances d’abus dans le cerveau est associée à un plus fort risque d’addiction et fait de ce transporteur un nouvel acteur critique de la régulation du passage cérébral. En effet, d’autres substances comme la nicotine et certaines amphétamines comme le MDPV et l'ecstasy sont également des substrats de cet antiport. Ce transporteur apparaît comme une cible pharmacologique potentielle dans la prise en charge de toxicomanies. Malgré la diversité chimique et pharmacologique d’interactions des composés avec cet antiport, les concentrations nécessaires pour l’inhiber dépassent celles retrouvées dans le sang. Pour aider l’identification d’inhibiteurs sélectifs et efficaces nous avons développé un modèle pharmacophorique d’inhibiteurs du transporteur à partir de données générées in vitro et de l’approche FLAPpharm. Ce modèle semble prédictif de nouveaux composés pouvant constituer de meilleurs inhibiteurs de ce transporteur. L’étude des échanges in vivo au niveau du tissu nerveux nous a menés à étudier l’impact de transporteurs ABC et de l’antiport-proton au niveau cérébral et rétinien à l’aide de substances spécifiques ou de substrats mixtes comme le vérapamil. L’antiport proton est fonctionnel au niveau de la BHR et transporte notamment la clonidine, le DPH et le vérapamil. Cependant, dans le cas d’un substrat mixte P-gp et SLC (ex : vérapamil), ce transport d’influx n’est visible à la BHE que lorsque la P-gp est neutralisée. Au contraire, à la BHR l’influx lié à cet SLC est visible naturellement. L’impact de la P-gp à la BHR étant 6.3-fois plus faible ce processus est probablement moins masqué. Cette étude illustre la difficulté actuelle de prédire l’impact fonctionnel d’un transporteur pour des substrats multi-spécifiques et l’existence d’une priorisation du transport. Enfin, nous avons essayé d’identifier l’antiport proton au niveau moléculaire par une méthode de photo-activation à l’aide d’un composé adapté. Cette méthode s’est avérée efficace pour fixer une molécule sur le transporteur, permettant par la suite de l’isoler plus facilement. En conclusion, ce travail a permis de mettre en évidence l’importance de l’antiport proton dans la distribution cérébrale de psychotropes et d’ouvrir de nouvelles perspectives dans l’addiction et la compréhension du transport de substrats multi-spécifiques. / The central nervous system is a privilege organ protected by histological barriers between the blood and the nervous tissue. The blood-brain barrier (BBB) and the blood-retinal barrier (BRB) separate cerebral parenchyma and retina from the circulating blood and both express tight junctions and membrane transporters, allowing a precise regulation of the exchanges between the blood and nervous tissues. We studied a new cationic transporter functionally evidenced at the mouse BBB. This molecularly unknown transporter belong to the solute carrier super family (SLC) and is a proton antiporter. It could constitute a new actor in the cerebral permeability and may be a new brain access pathway. First, we worked on the functional identification studying new substrates and new localization. Psychotropic brain transport was studied in vivo by brain in situ perfusion on mouse and in vitro with human immortalized endothelial cells (hCMEC/D3). We showed that cocaine brain entry depends on passive diffusion but also mainly on a proton antiporter. Brain entry rate of drugs of abuse is associated with modulation of addiction liability, making this transporter a new component of brain entry of cocaine, and also nicotine and some amphetamines such as ecstasy and MDPV. This proton antiporter appears to be a new potential target in addiction. Various chemical entities interact with this transporter; however concentrations used to inhibit the transporter are much higher than the one possibly found in the blood. In order to help find or design new selective and potent inhibitors, we developed a pharmacophore model of the proton antiporter inhibitors using in vitro data and the FLAPpharm approach. The model predicts well new possible inhibitors of this transporter. We also studied the impact of the ABC transporters and the proton antiporter at the BBB and the BRB using specific or multi-specific substrates such as verapamil. The proton antiporter is functionally expressed at the BRB and transports clonidine, DPH and verapamil. However, for the multi-specific (P-gp and SLC) compound verapamil, influx transport by the proton antiporter is visible at the BBB only when P-gp efflux is neutralized. On the contrary, at the BRB, the proton antiporter influx is always visible. This is certainly due to the lower impact (by 6.3 fold) of P-gp at the BRB compared to the BBB. These results show the difficulty to predict the functional impact of a transporter for multi-specific compounds and a probable transport prioritization. Finally we worked on the molecular identification of the proton antiporter using a photolabeling method. This work evidenced the importance of the proton antiporter in the brain distribution of psychotropic and drugs of abuse and opened toward new perspectives in addiction and transport comprehension.

Le système nerveux central est un organe privilégié et protégé, notamment grâce à l’existence des barrières histologiques entre le sang et les tissus nerveux. La barrière-hémato encéphalique (BHE) et la barrière hémato-rétinienne (BHR) séparent respectivement le parenchyme cérébral et la rétine des composés contenus dans l’espace vasculaire, grâce à l’expression de jonctions serrées et de transporteurs membranaires permettant une régulation spécifique des échanges entre le sang et le parenchyme nerveux. Ce travail a porté sur l’étude d’un nouveau transporteur de cations organiques mis en évidence fonctionnellement à la BHE de la souris. Ce transporteur appartenant très probablement à la superfamille des solute carrier (SLC), fonctionne comme un antiport proton. Actuellement, sa présence ne peut être démontrée que de façon fonctionnelle car son identité moléculaire est encore inconnue. Cet antiport proton constitue un nouvel acteur de la perméabilité cérébrale et ouvre une nouvelle voie d’accès au cerveau. Nous nous sommes tout d’abord attachés à approfondir les connaissances fonctionnelles de ce transporteur en étudiant de nouveaux substrats et tissus d’expression. Le transport cérébral de psychotropes a été étudié in vivo par la technique de perfusion carotidienne in situ chez la souris et in vitro grâce à une lignée de cellules endothéliales cérébrales humaines immortalisées (hCMEC/D3). Nous avons démontré que la haute perméabilité cérébrale de la cocaïne fait intervenir à la fois une diffusion passive et surtout une diffusion médiée par un antiport proton. La vitesse d’entrée des substances d’abus dans le cerveau est associée à un plus fort risque d’addiction et fait de ce transporteur un nouvel acteur critique de la régulation du passage cérébral. En effet, d’autres substances comme la nicotine et certaines amphétamines comme le MDPV et l'ecstasy sont également des substrats de cet antiport. Ce transporteur apparaît comme une cible pharmacologique potentielle dans la prise en charge de toxicomanies. Malgré la diversité chimique et pharmacologique d’interactions des composés avec cet antiport, les concentrations nécessaires pour l’inhiber dépassent celles retrouvées dans le sang. Pour aider l’identification d’inhibiteurs sélectifs et efficaces nous avons développé un modèle pharmacophorique d’inhibiteurs du transporteur à partir de données générées in vitro et de l’approche FLAPpharm. Ce modèle semble prédictif de nouveaux composés pouvant constituer de meilleurs inhibiteurs de ce transporteur. L’étude des échanges in vivo au niveau du tissu nerveux nous a menés à étudier l’impact de transporteurs ABC et de l’antiport-proton au niveau cérébral et rétinien à l’aide de substances spécifiques ou de substrats mixtes comme le vérapamil. L’antiport proton est fonctionnel au niveau de la BHR et transporte notamment la clonidine, le DPH et le vérapamil. Cependant, dans le cas d’un substrat mixte P-gp et SLC (ex : vérapamil), ce transport d’influx n’est visible à la BHE que lorsque la P-gp est neutralisée. Au contraire, à la BHR l’influx lié à cet SLC est visible naturellement. L’impact de la P-gp à la BHR étant 6.3-fois plus faible ce processus est probablement moins masqué. Cette étude illustre la difficulté actuelle de prédire l’impact fonctionnel d’un transporteur pour des substrats multi-spécifiques et l’existence d’une priorisation du transport. Enfin, nous avons essayé d’identifier l’antiport proton au niveau moléculaire par une méthode de photo-activation à l’aide d’un composé adapté. Cette méthode s’est avérée efficace pour fixer une molécule sur le transporteur, permettant par la suite de l’isoler plus facilement. En conclusion, ce travail a permis de mettre en évidence l’importance de l’antiport proton dans la distribution cérébrale de psychotropes et d’ouvrir de nouvelles perspectives dans l’addiction et la compréhension du transport de substrats multi-spécifiques. / The central nervous system is a privilege organ protected by histological barriers between the blood and the nervous tissue. The blood-brain barrier (BBB) and the blood-retinal barrier (BRB) separate cerebral parenchyma and retina from the circulating blood and both express tight junctions and membrane transporters, allowing a precise regulation of the exchanges between the blood and nervous tissues. We studied a new cationic transporter functionally evidenced at the mouse BBB. This molecularly unknown transporter belong to the solute carrier super family (SLC) and is a proton antiporter. It could constitute a new actor in the cerebral permeability and may be a new brain access pathway. First, we worked on the functional identification studying new substrates and new localization. Psychotropic brain transport was studied in vivo by brain in situ perfusion on mouse and in vitro with human immortalized endothelial cells (hCMEC/D3). We showed that cocaine brain entry depends on passive diffusion but also mainly on a proton antiporter. Brain entry rate of drugs of abuse is associated with modulation of addiction liability, making this transporter a new component of brain entry of cocaine, and also nicotine and some amphetamines such as ecstasy and MDPV. This proton antiporter appears to be a new potential target in addiction. Various chemical entities interact with this transporter; however concentrations used to inhibit the transporter are much higher than the one possibly found in the blood. In order to help find or design new selective and potent inhibitors, we developed a pharmacophore model of the proton antiporter inhibitors using in vitro data and the FLAPpharm approach. The model predicts well new possible inhibitors of this transporter. We also studied the impact of the ABC transporters and the proton antiporter at the BBB and the BRB using specific or multi-specific substrates such as verapamil. The proton antiporter is functionally expressed at the BRB and transports clonidine, DPH and verapamil. However, for the multi-specific (P-gp and SLC) compound verapamil, influx transport by the proton antiporter is visible at the BBB only when P-gp efflux is neutralized. On the contrary, at the BRB, the proton antiporter influx is always visible. This is certainly due to the lower impact (by 6.3 fold) of P-gp at the BRB compared to the BBB. These results show the difficulty to predict the functional impact of a transporter for multi-specific compounds and a probable transport prioritization. Finally we worked on the molecular identification of the proton antiporter using a photolabeling method. This work evidenced the importance of the proton antiporter in the brain distribution of psychotropic and drugs of abuse and opened toward new perspectives in addiction and transport comprehension.

Le système nerveux central est un organe privilégié et protégé, notamment grâce à l’existence des barrières histologiques entre le sang et les tissus nerveux. La barrière-hémato encéphalique (BHE) et la barrière hémato-rétinienne (BHR) séparent respectivement le parenchyme cérébral et la rétine des composés contenus dans l’espace vasculaire, grâce à l’expression de jonctions serrées et de transporteurs membranaires permettant une régulation spécifique des échanges entre le sang et le parenchyme nerveux. Ce travail a porté sur l’étude d’un nouveau transporteur de cations organiques mis en évidence fonctionnellement à la BHE de la souris. Ce transporteur appartenant très probablement à la superfamille des solute carrier (SLC), fonctionne comme un antiport proton. Actuellement, sa présence ne peut être démontrée que de façon fonctionnelle car son identité moléculaire est encore inconnue. Cet antiport proton constitue un nouvel acteur de la perméabilité cérébrale et ouvre une nouvelle voie d’accès au cerveau. Nous nous sommes tout d’abord attachés à approfondir les connaissances fonctionnelles de ce transporteur en étudiant de nouveaux substrats et tissus d’expression. Le transport cérébral de psychotropes a été étudié in vivo par la technique de perfusion carotidienne in situ chez la souris et in vitro grâce à une lignée de cellules endothéliales cérébrales humaines immortalisées (hCMEC/D3). Nous avons démontré que la haute perméabilité cérébrale de la cocaïne fait intervenir à la fois une diffusion passive et surtout une diffusion médiée par un antiport proton. La vitesse d’entrée des substances d’abus dans le cerveau est associée à un plus fort risque d’addiction et fait de ce transporteur un nouvel acteur critique de la régulation du passage cérébral. En effet, d’autres substances comme la nicotine et certaines amphétamines comme le MDPV et l'ecstasy sont également des substrats de cet antiport. Ce transporteur apparaît comme une cible pharmacologique potentielle dans la prise en charge de toxicomanies. Malgré la diversité chimique et pharmacologique d’interactions des composés avec cet antiport, les concentrations nécessaires pour l’inhiber dépassent celles retrouvées dans le sang. Pour aider l’identification d’inhibiteurs sélectifs et efficaces nous avons développé un modèle pharmacophorique d’inhibiteurs du transporteur à partir de données générées in vitro et de l’approche FLAPpharm. Ce modèle semble prédictif de nouveaux composés pouvant constituer de meilleurs inhibiteurs de ce transporteur. L’étude des échanges in vivo au niveau du tissu nerveux nous a menés à étudier l’impact de transporteurs ABC et de l’antiport-proton au niveau cérébral et rétinien à l’aide de substances spécifiques ou de substrats mixtes comme le vérapamil. L’antiport proton est fonctionnel au niveau de la BHR et transporte notamment la clonidine, le DPH et le vérapamil. Cependant, dans le cas d’un substrat mixte P-gp et SLC (ex : vérapamil), ce transport d’influx n’est visible à la BHE que lorsque la P-gp est neutralisée. Au contraire, à la BHR l’influx lié à cet SLC est visible naturellement. L’impact de la P-gp à la BHR étant 6.3-fois plus faible ce processus est probablement moins masqué. Cette étude illustre la difficulté actuelle de prédire l’impact fonctionnel d’un transporteur pour des substrats multi-spécifiques et l’existence d’une priorisation du transport. Enfin, nous avons essayé d’identifier l’antiport proton au niveau moléculaire par une méthode de photo-activation à l’aide d’un composé adapté. Cette méthode s’est avérée efficace pour fixer une molécule sur le transporteur, permettant par la suite de l’isoler plus facilement. En conclusion, ce travail a permis de mettre en évidence l’importance de l’antiport proton dans la distribution cérébrale de psychotropes et d’ouvrir de nouvelles perspectives dans l’addiction et la compréhension du transport de substrats multi-spécifiques. / The central nervous system is a privilege organ protected by histological barriers between the blood and the nervous tissue. The blood-brain barrier (BBB) and the blood-retinal barrier (BRB) separate cerebral parenchyma and retina from the circulating blood and both express tight junctions and membrane transporters, allowing a precise regulation of the exchanges between the blood and nervous tissues. We studied a new cationic transporter functionally evidenced at the mouse BBB. This molecularly unknown transporter belong to the solute carrier super family (SLC) and is a proton antiporter. It could constitute a new actor in the cerebral permeability and may be a new brain access pathway. First, we worked on the functional identification studying new substrates and new localization. Psychotropic brain transport was studied in vivo by brain in situ perfusion on mouse and in vitro with human immortalized endothelial cells (hCMEC/D3). We showed that cocaine brain entry depends on passive diffusion but also mainly on a proton antiporter. Brain entry rate of drugs of abuse is associated with modulation of addiction liability, making this transporter a new component of brain entry of cocaine, and also nicotine and some amphetamines such as ecstasy and MDPV. This proton antiporter appears to be a new potential target in addiction. Various chemical entities interact with this transporter; however concentrations used to inhibit the transporter are much higher than the one possibly found in the blood. In order to help find or design new selective and potent inhibitors, we developed a pharmacophore model of the proton antiporter inhibitors using in vitro data and the FLAPpharm approach. The model predicts well new possible inhibitors of this transporter. We also studied the impact of the ABC transporters and the proton antiporter at the BBB and the BRB using specific or multi-specific substrates such as verapamil. The proton antiporter is functionally expressed at the BRB and transports clonidine, DPH and verapamil. However, for the multi-specific (P-gp and SLC) compound verapamil, influx transport by the proton antiporter is visible at the BBB only when P-gp efflux is neutralized. On the contrary, at the BRB, the proton antiporter influx is always visible. This is certainly due to the lower impact (by 6.3 fold) of P-gp at the BRB compared to the BBB. These results show the difficulty to predict the functional impact of a transporter for multi-specific compounds and a probable transport prioritization. Finally we worked on the molecular identification of the proton antiporter using a photolabeling method. This work evidenced the importance of the proton antiporter in the brain distribution of psychotropic and drugs of abuse and opened toward new perspectives in addiction and transport comprehension.