Targeting vascular NADPH oxidase 1 blocks tumor angiogenesis through a PPARα mediated mechanism.

Reactive oxygen species, ROS, are regulators of endothelial cell migration, proliferation and survival, events critically involved in angiogenesis. Different isoforms of ROS-generating NOX enzymes are expressed in the vasculature and provide distinct signaling cues through differential localization and activation. We show that mice deficient in NOX1, but not NOX2 or NOX4, have impaired angiogenesis. NOX1 expression and activity is increased in primary mouse and human endothelial cells upon angiogenic stimulation. NOX1 silencing decreases endothelial cell migration and tube-like structure formation, through the inhibition of PPARα, a regulator of NF-κB. Administration of a novel NOX-specific inhibitor reduced angiogenesis and tumor growth in vivo in a PPARα dependent manner. In conclusion, vascular NOX1 is a critical mediator of angiogenesis and an attractive target for anti-angiogenic therapies.

Mechanism of hcnA mRNA recognition in the Gac/Rsm signal transduction pathway of Pseudomonas fluorescens.

In the plant-beneficial bacterium Pseudomonas fluorescens CHA0, the expression of antifungal exoproducts is controlled by the GacS/GacA two-component system. Two RNA binding proteins (RsmA, RsmE) ensure effective translational repression of exoproduct mRNAs. At high cell population densities, GacA induces three small RNAs (RsmX, RsmY, RsmZ) which sequester both RsmA and RsmE, thereby relieving translational repression. Here we systematically analyse the features that allow the RNA binding proteins to interact strongly with the 5' untranslated leader mRNA of the P. fluorescens hcnA gene (encoding hydrogen cyanide synthase subunit A). We obtained evidence for three major RsmA/RsmE recognition elements in the hcnA leader, based on directed mutagenesis, RsmE footprints and toeprints, and in vivo expression data. Two recognition elements were found in two stem-loop structures whose existence in the 5' leader region was confirmed by lead(II) cleavage analysis. The third recognition element, which overlapped the hcnA Shine-Dalgarno sequence, was postulated to adopt either an open conformation, which would favour ribosome binding, or a stem-loop structure, which may form upon interaction with RsmA/RsmE and would inhibit access of ribosomes. Effective control of hcnA expression by the Gac/Rsm system appears to result from the combination of the three appropriately spaced recognition elements.

Insulin resistance in mice lacking neuronal nitric oxide synthase is related to an alpha-adrenergic mechanism

Rapport de synthèse : Le monoxyde d'azote (NO) joue un rôle important dans la régulation de l'homéostasie du système cardiovasculaire et du glucose. Les souris déficientes pour le gène codant l'isoforme neuronale de la synthase de monoxyde d'azote (nNOS) sont résistantes à l'insuline, mais les mécanismes sous-jacents sont inconnus. Le manque de NO produit par la nNOS pourrait être à l'origine d'une diminution de la perfusion du muscle squelettique et ainsi d'une diminution de l'apport de substrat. Alternativement, le déficit de nNOS normalement hautement exprimé dans le tissu musculaire squelettique pourrait directement y perturber la consommation de glucose. Finalement l'absence de l'action sympatholytique du NO neuronal pourrait diminuer la sensibilité à l'insuline. Afin de tester ces hypothèses nous avons étudié, chez des souris déficientes en nNOS et des souris-contrôle, la consommation corporelle totale de glucose et le flux musculaire squelettique pendant des clamps hyperinsulinémiques euglycémiques in vivo, ainsi que la consommation de glucose dans le muscle squelettique in vitro. De plus nous avons analysé les effets d'une inhibition alpha-adrénergique sur la consommation de glucose pendant les clamps hyperinsulinémiques euglycémiques in vivo. Le taux de perfusion de glucose pendant les clamps était grossièrement 15 pourcent plus bas (P<0.001) chez les souris déficientes en nNOS que chez les souris-contrôle. Cette résistance à l'insuline chez les souris déficientes en nNOS n'était due ni à une stimulation déficiente du flux sanguin musculaire par l'insuline ni à un défaut intrinsèque de la consommation de glucose du muscle (qui étaient comparables dans les deux groupes), mais à un mécanisme alpha-adrénergique, car l'administration de phentolamine rétablissait la sensibilité à l'insuline chez les souris déficientes en nNOS. Ces résultats suggèrent qu'une hyperactivité sympathique, potentiellement due à la perte de l'inhibition neuronale centrale du flux sympathique par le NO provenant de nNOS, contribue à la résistance à l'insuline des souris déficientes en nNOS. Par ailleurs ces résultats tendent à prouver qu'un défaut de production de NO provoquerait une résistance à l'insuline par des mécanismes différents selon l'isoforme de NO synthase déficiente (par exemple chez les souris déficientes pour la forme endothéliale de NO synthase, il a été montré que la résistance à l'insuline est due à un défaut de stimulation de la perfusion musculaire par l'insuline et à un défaut du signalling de l'insuline dans la cellule musculaire squelettique). Chez l'être humain il est établi que les états de résistance à l'insuline sont associés à une synthèse défectueuse et/ou une mauvaise biodisponibilité du NO, ainsi qu'à une hyperactivité sympathique. Nous spéculons que la perte d'inhibition centrale du flux sympathique représente un mécanisme contribuant à la résistance à l'insuline et ses complications cardiovasculaires chez l'être humain.

Loss of insulin synthesis and beta cell survival induced by oxidized LDL require activation of reticulum endoplasmic stress: a mechanism countered by HDL

Elevated circulating concentrations in modified LDL-cholesterol particles (e.g. oxidised LDL) and low levels in HDL increase not only the risk for diabetic patients to develop cardiovascular diseases but also may contribute to development and progression of diabetes by directly having adverse effects on β-cells. Chronic exposure of β-cells to 2 mM human oxidised LDL-cholesterol (oxLDL) increases the rate of apoptosis, reduce insulin biosynthesis and the secretory capacity of the cells in response to nutrients. In line with the protective role, HDL efficiently antagonised the harmful effects of ox- LDL, suggesting that low levels of HDL would be inefficient to protect β-cells against oxLDL attack in patients. Activation of endoplasmic reticulum (ER) stress is pointed out to contribute to β-cell dysfunction elicited by environmental stressors. In this study we investigated whether activation of ER stress is required for oxLDL to mediate detrimental effects on β-cells and we tested the potential antagonist properties of HDL: The mouse MIN6 insulin-secreting cells were cultured with 2 mM of LDL-cholesterol preparation (native or in vitro oxidized) in the presence or absence of 1 mM of HDL-cholesterol or the ER stress inhibitor 4-phenylbutyrate (4-PBA): Prolonged exposure of MIN6 cells to 2 mM oxLDL-cholesterol for 48 hours led to an increase in expression of ER stress markers such as ATF4, CHOP and p58 and stimulated the splicing of XBP-1 whereas, induction of these markers was not observable in the cells cultured with native LDL. Treatment of the cells with the 4-PBA chemical chaperone molecule efficiently blocked activation of the ER stress markers induced by oxLDL. The latter mediates β-cell dysfunction and apoptosis by diminishing the expression of islet brain 1 (IB1) and Bcl2. The levels of these two proteins were preserved in the cells that were co-treated with oxLDL and the 4-PBA. Consistent with this result we found that blockade of ER stress activation alleviated the loss of insulin synthesis and abolished apoptosis evoked by oxLDL. However incubation of the cells with 4-PBA did not prevent impairment of insulin secretion elicited by oxLDL, indicating that ER stress is not responsible for the oxLDL-mediated defect of insulin secretion. Co-incubation of the cells with HDL mimicked the effects of 4-PBA on the expression of IB1 and Blc2 and thereby counteracted oxLDL attacks on insulin synthesis and cell survivals. We found that HDL efficiently inhibited activation of the ER stress mediated by oxLDL: These data highlight the contribution of the ER stress in the defects of insulin synthesis and cell survivals induced by oxLDL and emphasize the potent role of HDL to counter activation of the oxLDL-mediated ER-stress activation:

Neurophysiological effects of vasopressin and oxytocin in the central amygdala of the rat : a potential mechanism for the opposite modulation of anxiety and fear behavior

Abstract The amygdala is a group of nuclei in the temporal lobe of the brain that plays a crucial role in anxiety and fear behavior. Sensory information converges in the basolateral and lateral nuclei of the amygdala, which have been the first regions in the brain where the acquisition of new (fear) memories has been associated with long term changes in synaptic transmission. These nuclei, in turn, project to the central nucleus of the amygdala. The central amygdala, through its extensive projections to numerous nuclei in the midbrain and brainstem, plays a pivotal role in the orchestration of the rapid autonomic and endocrine fear responses. In the central amygdala a large number of neuropeptides and receptors is expressed, among which high levels of vasopressin and oxytocin receptors. Local injections of these peptides into the amygdala modulate several aspects of the autonomic fear reaction. Interestingly, their effects are opposing: vasopressin tends to enhance the fear reactions, whereas oxytocin has anxiolytic effects. In order to investigate the neurophysiological mechanisms that could underlie this opposing modulation of the fear behavior, we studied the effects of vasopressin and oxytocin on the neuronal activity in an acute brain slice preparation of the rat central amygdala. We first assessed the effects of vasopressin and oxytocin on the spontaneous activity of central amygdala neurons. Extracellular single unit recordings revealed two major populations of neurons: a majority of neurons was excited by vasopressin and inhibited by oxytocin, whereas other neurons were only excited by oxytocin receptor activation. The inhibitory effect of oxytocin could be reduced by the block of GABAergic transmission, whereas the excitatory effects of vasopressin and oxytocin were not affected. In a second step we identified the cellular mechanisms for the excitatory effects of both peptides as well as the morphological and biochemical mechanisms underlying the opposing effects, by using sharp electrode recordings together with intracellular labelings. We revealed that oxytocin-excited neurons are localized in the lateral part (CeL) whereas vasopressin excited cells are found in the medial part of the central amygdala (CeM). The tracing of the neuronal morphology showed that the axon collaterals of the oxytocin-excited neurons project from the CeL, far into the CeM. Combined immunohistochemical stainings indicated that these projections are GABAergic. In the third set of experiments we investigated the synaptic interactions between the two identified cell populations. Whole-cell patch-clamp recordings in the CeM revealed that the inhibitory effect of oxytocin was caused by the massive increase of inhibitory GABAergic currents, which was induced by the activation of CeL neurons. Finally, the effects of vasopressin and oxytocin on evoked activity were investigated. We found on the one hand, that the probability of evoking action potentials in the CeM by stimulating the basolateral amygdala afferents was enhanced under vasopressin, whereas it decreased under oxytocin. On the other hand, the impact of cortical afferents stimulation on the CeL neurons was enhanced by oxytocin application. Taken together, these findings have allowed us to develop a model, in which the opposing behavioral effects of vasopressin and oxytocin are caused by a selective activation of two distinct populations of neurons in the GABAergic network of the central amygdala. Our model could help to develop new anxiolytic treatments, which modulate simultaneously both receptor systems. By acting on a GABAergic network, such treatments can further be tuned by combinations with classical benzodiazepines. Résumé: L'amygdale est un groupe de noyaux cérébraux localisés dans le lobe temporal. Elle joue un rôle essentiel dans les comportements liés à la peur et l'anxiété. L'information issue des aires sensorielles converge vers les noyaux amygdaliens latéraux et basolatéraux, qui sont les projections vers différents noyaux du tronc cérébral et de l'hypothalamus, joue un rôle clef premières régions dans lesquelles il a été démontré que l'acquisition d'une nouvelle mémoire (de peur) était associée à des changements à long terme de la transmission synaptique. Ces noyaux envoient leurs projections sur l'amygdale centrale, qui à travers ses propres dans l'orchestration des réponses autonomes et endocrines de peur. Le contrôle de l'activité neuronale dans l'amygdale centrale module fortement la réaction de peur. Ainsi, un grand nombre de neuropeptides sont spécifiquement exprimés dans l'amygdale centrale et un bon nombre d'entre eux interfère dans la réaction de peur et d'anxiété. Chez les rats, une forte concentration de récepteurs à l'ocytocine et à la vasopressine est exprimée dans le noyau central, et l'injection de ces peptides dans l'amygdale influence différents aspects de la réaction viscérale associée à la peur. Il est intéressant de constater que ces peptides exercent des effets opposés. Ainsi, la vasopressine augmente la réaction de peur alors que l'ocytocine a un effet anxiolytique. Afin d'investiguer les mécanismes neurophysiologiques responsables de ces effets opposés, nous avons étudié l'effet de la vasopressine et de l'ocytocine sur l'activité neuronale de préparations de tranches de cerveau de rats contenant entre autres de l'amygdale centrale. Tout d'abord, notre intérêt s'est porté sur les effets de ces deux neuropeptides sur l'activité spontanée dans l'amygdale centrale. Des enregistrements extracellulaires ont révélé différentes populations de neurones ; une majorité était excitée par la vasopressine et inhibée par l'ocytocine ; d'autres étaient seulement excités par l'activation du récepteur à l'ocytocine. L'effet inhibiteur de l'ocytocine a pu être réduit par l'inhibition de la transmission GABAergique, alors que ses effets excitateurs n'étaient pas affectés. Dans un deuxième temps, nous avons identifié les mécanismes cellulaires responsables de l'effet excitateur de ces deux peptides et analysé les caractéristiques morphologiques et biochimiques des neurones affectés. Des enregistrements intracellulaires ont permis de localiser les neurones excités par l'ocytocine dans la partie latérale de l'amygdale centrale (CeL), et ceux excités par la vasopressine dans sa partie médiale (CeM). Le traçage morphologique des neurones a révélé que les collatérales axonales des cellules excitées par l'ocytocine projetaient du CeL loin dans le CeM. De plus, des colorations immuno-histochimiques ont révélé que ces projections étaient GABAergiques. Dans un troisième temps, nous avons étudié les interactions synaptiques entre ces deux populations de cellules. Les enregistrements en whole-cell patch-clamp dans le CeM ont démontré que les effets inhibiteurs de l'ocytocine résultaient de l'augmentation massive des courants GABAergique résultant de l'activation des neurones dans le CeL. Finalement, les effets de l'ocytocine et de la vasopressine sur l'activité évoquée ont été étudiés. Nous avons pu montrer que la probabilité d'évoquer un potentiel d'action dans le CeM, par stimulation de l'amygdale basolatérale, était augmentée sous l'effet de la vasopressine et diminuée sous l'action de l'ocytocine. Par contre, l'impact de la stimulation des afférences corticales sur les neurones du CeL était augmenté par l'application de l'ocytocine. L'ensemble de ces résultats nous a permis de développer un modèle dans lequel les effets comportementaux opposés de la vasopressine et de l'ocytocine sont causés par une activation sélective des deux différentes populations de neurones dans un réseau GABAergique. Un tel modèle pourrait mener au développement de nouveaux traitements anxiolytiques en modulant l'activité des deux récepteurs simultanément. En agissant sur un réseau GABAergique, les effets d'un tel traitement pourraient être rendus encore plus sélectifs en association avec des benzodiazépines classiques.

Long sequence duplications, repeats, and palindromes in HIV-1 gp120: length variation in V4 as the product of misalignment mechanism.

We have shown that indels in gp120 V4 are associated to the presence of duplicated and palindromic sequences, suggesting that they may be produced by strand-slippage misalignment mechanism. Indels in V4 involved region-specific duplications 9 to 15 bp long, and repeats of various lengths, associated to trinucleotides AAT. No duplications were found in V3 and C3. The frequency of palindromic sequences in individual genes was found to be significantly higher in gp120 (p < or = 3.00E-7), and significantly lower in Tat (p < or = 9.00E-7) than the average frequency calculated over the full genome. The finding of elements of misalignment in association with indels in V4 suggests that these mutations may occur in proviral DNA after integration of HIV into the host genome. It also implies that occurrence of large indels in gp120 is not random but is directed by the presence and distribution of elements of misalignment in the HIV genome.

Length variation in HIV-1 gp120 as the product of DNA misalignment mechanism

Background: To determine whether misalignment structures such as duplications, repeats, and palindromes are associated to insertions/deletions (indels) in gp120, indicating that indels are indeed frameshift mutations generated by DNA misalignment mechanism. Methods: Cloning and sequencing of a fragment of HIV-1 gp120 spanning C2-C4 derived from plasma RNA in 12 patients with early chronic disease and naïve to antiretroviral therapy. Results: Indels in V4 involved always insertion and deletion of duplicated nucleotide segments, and AAT repeats, and were associated to the presence of palindromic sequences. No duplications were detected in V3 and C3. Palindromic sequences occurred with similar frequencies in V3, C3 and V4; the frequency of palindromes in individual genes was found to be significantly higher in structural (gp120, p ≤ 3.00E-7) and significantly lower in regulatory (Tat, p ≤ 9.00E-7) genes, as compared to the average frequency calculated over the full genome. Discussion: Indels in V4 are associated to misalignment structures (i.e. duplications repeat and palindromes) indicating DNA misalignment as the mechanism underlying length variation in V4. The finding that indels in V4 are caused by DNA misalignment has some very important implications: 1) indels in V4 are likely to occur in proviral DNA (and not in RNA), after integration of HIV into the host genome; 2) they are likely to occur as progressive modifications of the early founder virus during chronic infection, as more and more cells get infected; 3) frameshift mutations involving any number of base pairs are likely to occur evenly across gp120; however, only those mutants carrying a functional gp120 (indels as multiples of three base pairs) will be able to perpetuate the virus cycle and to keep spreading through the population.

Trace element chemistry and U-Pb dating of zircons from oceanic gabbros and their relationship with whole rock composition (Lanzo, Italian Alps)

The U-Pb ages and the trace element content of zircon U-Pb along with major and trace element whole rock data on gabbroic dikes from the Lanzo lherzolitic massif, N-Italy, have been determined to constrain crustal accretion in ocean-continent transition zones. Three Fe-Ti gabbros were dated from the central and the southern part of the massif providing middle Jurassic ages of 161 +/- 2, 158 +/- 2 and 163 +/- 1 Ma, which argue for magmatic activity over few millions of years. Zircon crystals are characterized by high but variable Th/U ratios, rare earth element patterns enriched in heavy rare earths, pronounced positive Ce and negative Eu-anomalies consistent with crystallization after substantial plagioclase fractionation. The zircon trace element composition coupled with whole rock chemistry was used to reconstruct the crystallization history of the gabbros. A number of gabbros crystallized in situ, and zircon precipitated from trapped, intercumulus liquid, while other gabbros represent residual liquids that were extracted from a cumulus pile and crystallized along syn-magmatic shear zones. We propose a model in which the emplacement mechanism of gabbroic rocks in ocean-continent transition zones evolves from in situ crystallization to stratified crystallization with efficient extraction of residual liquid along syn-magmatic shear zones. Such an evolution of the crystallization history is probably related to the thermal evolution of the underlying mantle lithosphere.

Tumor sensitizing function and mechanism of action of a RasGAP derived peptide

Cancer is the second cause of death after cardio-vascular diseases in economically developed countries. Two of the most commonly used anti-cancer therapies are chemo and radiotherapy. Despite the remarkable advances made in term of delivery and specificity of these two anti-tumor regimens, their toxicity towards healthy tissue remains a limitation. A promising approach to overcome this obstacle would be the utilization of therapeutic peptides that specifically augment the sensitivity of tumoral cells to treatments. Lower therapeutical doses would then be required to kill malignant cells, limiting toxic effects on healthy tissues. It was previously shown in our laboratory that the caspase-3 generated fragment N2 of RasGAP is able to potentiate the genotoxin-induced apoptosis selectively in cancer cells. In this work we show that fragment N2 strictly requires a cytoplasmic localization to deliver its pro-apoptotic effect in genotoxin-treated cancer cells. The tumor sensitizing capacity of fragment N2 was found to reside within the 10 amino acid sequence 317-326. Our laboratory earlier demonstrated that a peptide corresponding to amino acids 317 to 326 of RasGAP fused to the TAT cell permeable moiety, called TAT-RasGAP317.326, is able to sensitize cancer cells, but not normal cells, to genotoxin-induced apoptosis. In the present study we describe the capacity of TAT-RasGAP 317.326 to sensitize tumors to both chemo and radiotherapy in an in vivo mouse model. The molecular mechanism underlying the TAT-RasGAP 317.326-mediated sensitization starts now to be elucidated. We demonstrate that G3BP1, an endoribonuclease binding to amino acids 317-326 of RasGAP, is not involved in the sensitization mechanism. We also provide evidence showing that TAT-RasGAP3 17-326 potentiates the genotoxin-mediated activation of Bax in a tBid-dependent manner. Altogether our results show that TAT-RasGAP 317.326 could be potentially used in cancer therapy as sensitizer, in order to improve the efficacy of chemo and radiotherapy and prolong the life expectancy of cancer patients. Moreover, the understanding of the TAT-RasGAP317.326 mode of action might help to unravel the mechanisms by which cancer cells resist to chemo and radiotherapy and therefore to design more targeted and efficient anti-tumoral strategies.

Correlations of glycogen synthase and phosphorylase activities with glycogen concentration in human muscle biopsies. Evidence for a double-feedback mechanism regulating glycogen synthesis and breakdown.

The purpose of this study was to verify in man the relationships of muscle glycogen synthase and phosphorylase activities with glycogen concentration that were reported in animal studies. The upper level of glycogen concentration in muscle is known to be tightly controlled, and glycogen concentration was reported to have an inhibitory effect on synthase activity and a stimulatory effect on phosphorylase activity. Glycogen synthase and phosphorylase activity and glycogen concentration were measured in muscle biopsies in a group of nine normal subjects after stimulating an increase of their muscle glycogen concentration through either an intravenous glucose-insulin infusion to stimulate glycogen synthesis, or an Intralipid (Vitrum, Stockholm, Sweden) infusion in the basal state to inhibit glycogen mobilization by favoring lipid oxidation at the expense of glucose oxidation. Phosphorylase activity increased from 71.3 +/- 21.0 to 152.8 +/- 20.0 nmol/min/mg protein (P < .005) after the glucose-insulin infusion. Phosphorylase activity was positively correlated with glycogen concentration (P = .005 and P = .0001) after the glucose-insulin and Intralipid infusions, respectively. Insulin-stimulated glycogen synthase activity was significantly negatively correlated with glycogen concentration at the end of the Intralipid infusion (P < .005). In conclusion, by demonstrating a negative correlation of glycogen concentration with glycogen synthase and a positive correlation with phosphorylase, this study might confirm in man the double-feedback mechanism by which changes in glycogen concentration regulate glycogen synthase and phosphorylase activities. It suggests that this mechanism might play an important role in the regulation of glucose storage.

The monocarboxylate transporter MCT4 : mechanism of regulation in astrocytes and its putative role in cerebral ischemia

Despite its small fraction of the total body weight (2%), the brain contributes for 20% and 25% respectively of the total oxygen and glucose consumption of the whole body. Indeed, glucose has been considered the energy substrate par excellence for the brain. However, evidence accumulated over the last half century revealed an important role for the monocarboxylate lactate in fulfilling the energy needs of neurons. This is particularly true during physiological neuronal activation and in pathological conditions. Lactate transport into and out of the cell is mediated by a family of proton-linked transporters called monocarboxylate transporters (MCTs). In the central nervous system, only three of them have been well characterized: MCT2 is the predominant neuronal isoform, while the other non¬neuronal cell types of the brain express the ubiquitous isoform MCT1. Quite recently, the MCT4 isoform has been described in astrocytes. Due to its high transport capacity compared to the other two isoforms, MCT4 is particularly adapted for glycolytic cells. Because of its recent discovery in the brain, nothing was known about its regulation in the central nervous system. Here we show that MCT4 is regulated by oxygen levels in primary cultures of astrocytes in a time- and concentration-dependent manner via the hypoxia inducible factor-la (HIF-la). Moreover, we showed that MCT4 expression is essential for astrocyte survival under low oxygen conditions. In parallel, we investigated the possible implication of the pyruvate kinase isoform Pkm2, a strong enhancer of glycolysis, in its regulation. Then we showed that MCT4 expression, as well as the expression of the other two MCT isoforms, is altered in a murine model of stroke. Surprisingly, neurons started to express MCT4, as well as MCT1, under such conditions. Altogether, these data suggest that MCT4, due to its high transport capacity for lactate, may be the isoform that enables cells to operate a major metabolic adaptation in response to pathological situations that alter metabolic homeostasis of the brain. -- Le cerveau représente 2% du poids corporel total, mais il contribue pour 20% de la consommation totale d'oxygène et 25% de celle de glucose au repos. Le glucose est considéré comme le substrat énergétique par excellence pour le cerveau. Néanmoins, depuis un demi- siècle maintenant, de plus en plus de travaux ont démontré que le lactate joue un rôle majeur dans le métabolisme cérébral et est capable du subvenir aux besoins énergétiques des neurones. Le lactate est tout particulièrement nécessaire pendant l'activation neuronale ainsi qu'en situation pathologique. Le transport du lactate à travers la barrière hématoencéphalique ainsi qu'à travers les membranes cellulaires est assuré par la famille des transporteurs aux monocarboxylates (MCTs). Dans le système nerveux central, uniquement trois d'entre eux ont été décrits: MCT2 est considéré comme le transporteur neuronal, alors que les autres types cellulaires qui constituent le cerveau expriment l'isoforme ubiquitaire MCT1. Récemment, l'isoforme MCT4 a été rapportée sur les astrocytes. Dû à sa grande capacité de transport pour le lactate, MCT4 est tout particulièrement adapté pour soutenir le métabolisme des cellules hautement glycolytiques, comme les astrocytes. En raison de sa toute récente découverte, les aspects comprenant sa régulation et son rôle dans le cerveau sont pour l'instant méconnus. Les résultats exposés dans ce travail démontrent dans un premier temps que l'expression de MCT4 est régulée par les niveaux d'oxygène dans les cultures d'astrocytes corticaux par le biais du facteur de transcription HIF-la. De plus, nous avons démontré que l'expression de MCT4 est essentielle à la survie des astrocytes quand le niveau d'oxygénation baisse. En parallèle, des résultats préliminaires suggèrent que l'isoforme 2 de la pyruvate kinase, un puissant régulateur de la glycolyse, pourrait jouer un rôle dans la régulation de MCT4. Dans la deuxième partie du travail nous avons démontré que l'expression de MCT4, ainsi que celle de MCT1 et MCT2, est altérée dans un modèle murin d'ischémie cérébrale. De façon surprenante, les neurones expriment MCT4 dans cette condition, alors que ce n'est pas le cas en condition physiologique. En tenant compte de ces résultats, nous suggérons que MCT4, dû à sa particulièrement grande capacité de transport pour le lactate, représente le MCT qui permet aux cellules du système nerveux central, notamment les astrocytes et les neurones, de s'adapter à de très fortes perturbations de l'homéostasie métabolique du cerveau qui surviennent en condition pathologique.