The peripheral immune system plays a role in delayed neural injury after stroke. This response originates from the spleen as splenectomy prior to middle cerebral artery occlusion (MCAO) in rats significantly reduces infarct volume in the brain. This research is based on the hypothesis that inhibiting the splenic response will reduce neurodegeneration after stroke. Studies in animals have implicated lymphocytes as the immune cell type that is detrimental following MCAO. Interferon gamma (IFNγ) has been identified as a pro-inflammatory cytokine that is also detrimental following stroke. IFNγ is important because it activates microglia and macrophages in a pro-inflammatory nature that increases neural injury following stroke. Therefore IFNγ was examined in the brain and the spleen following MCAO. IFNγ protein was elevated at 24 h in the spleen and at 72 h in the brain post MCAO. Microglia/macrophages become maximally activated at 72 h in the brain after MCAO. Splenectomy decreases the levels of IFNγ in the brain following MCAO. Systemic administration of IFNγ reversed the protective effects of splenectomy. The cellular response to MCAO was examined next because of the difference in time between the spike in IFNγ in the spleen and the delayed increase in the brain. The cellular response from the spleen was studied by labeling splenocytes five days prior to MCAO with a fluorescein dye. Tissues were examined 48 and 96 h post MCAO or sham MCAO for fluorescence. These cells were released from the spleen into circulation at 48 h post MCAO and migrated to the brain where the cells produced IFNγ at 96 h post MCAO. IFNγ appears to play a role in the splenic response to stroke. One protein that is up regulated by cells that have been activated by IFNγ, interferon-inducible protein 10 (IP-10) is part of the inflammatory cycle driven by IFNγ. IP-10 recruits more IFNγ producing T helper (Th) cells to the site of injury. IP-10 has the unique ability to attract Th1 cells, the pro-inflammatory Th cells, and inhibit Th2 cells, the anti-inflammatory Th cells. This leads to more IFNγ production as IFNγ is the signature cytokine of a Th1 response. IP-10 is significantly increased in the brain at 72 h post MCAO, similar to IFNγ expression. In the spleen IP-10 increased at 24 h and remained elevated out to 96 h following MCAO. IFNγ signaling was inhibited by utilizing an IFNγ neutralizing antibody administered beginning 24 h post MCAO. The IFNγ antibody treated group had decreased infarct volumes, IP-10 levels in the brain, and appeared to have decreased T cells in the ipsilateral hemisphere at 96 h post MCAO. Following ischemic stroke splenocytes are released into circulation and migrate to the brain. They release IFNγ to activate microglia/macrophages in a proinflammatory phenotype causing an increase in IP-10 levels. IP-10 then potentiates the Th1 driven inflammation which inhibits the Th2 response. The elevated levels of IFNγ increase neural injury following MCAO. Blocking IFNγ selectively blocks the inflammatory facet of the immune response to reduce stroke induced neurodegeneration. This leaves the other immune responses intact and able to contribute to tissue repair, regeneration, and able to respond to infections. Selectively inhibiting IFNγ signaling is a promising stroke therapeutic.

Les rétinopathies ischémiques prolifératives telles que la rétinopathie diabétique proliférative et la rétinopathie du prématuré sont les principales causes de la perte de la vision dans la population active et pédiatrique des pays industrialisés. Malgré le fait que les événements initiateurs sont différents et propres à chacune des pathologies, les rétinopathies ischémiques prolifératives sont le résultat d’un processus biphasique. On a d’abord une phase initiale de dégénérescence microvasculaire suivie d’une néovascularisation excessive et pathologique de la rétine hypoxique qui tente de réinstaurer l’apport en nutriments et en énergie. Mais au lieu d’aller revasculariser les zones avasculaires de la rétine, ces nouveaux vaisseaux sanguins sont mal orientés et se dirigent plutôt vers le vitré normalement avasculaire. Ceci provoque des tensions physiques dans la rétine et mène à long terme à son détachement et une perte de vision conséquente. Les traitements actuels ne viennent pas sans effets secondaires majeurs. Par exemple, la formation de la cataracte et l’augmentation de la pression intraoculaire avec l’utilisation des corticostéroïdes ou la perte de la vision partielle dans le cas du traitement au laser sont fréquemment observées. De même, la thérapie anti-VEGF (Vascular endothelial growth factor) apporte aussi son lot de complications, telles que la thromboembolie veineuse et l’augmentation de la neurotoxicité après un long usage, vu les propriétés neuro- et vaso-protectrices du VEGF. Le développement d’une nouvelle approche thérapeutique pour les rétinopathies ischémiques prolifératives est donc nécessaire afin de contrer ces limitations thérapeutiques. Dans notre première étude, nous mettons en évidence un nouveau mécanisme par lequel les cellules neuronales sous stress diabétique sont à l’origine d’une forte inflammation oculaire. Nos résultats démontrent que le co-récepteur multi-ligand Neuropiline-1, le VEGF et la Sémaphorine-3A agissent de concert afin d’attirer une sous-population particulière de phagocytes mononucléaires susceptibles d’activer le processus de croissance vasculaire pathologique dans la rétine diabétique. De plus, notre étude propose une base pour de futures recherches sur l’impact des phagocytes mononucléaires exprimant Neuropiline-1 dans les pathologies du système nerveux central caractérisées par une inflammation excessive. Nos résultats permettent aussi de mettre en lumière le caractère anti-inflammatoire potentiel des thérapies actuelles anti-VEGF (à cause du rôle de VEGF dans la mobilisation des phagocytes mononucléaires via Neuropiline-1) au niveau oculaire. Dans notre deuxième étude, nous mettons en évidence l’activation du facteur HIF1α dans les phagocytes mononucléaires présents dans la rétine hypoxique. L’utilisation d’une approche protéomique non biaisée de spectrométrie de masse en tandem nous a permis d’identifier les partenaires interagissant avec HIF1α dans un milieu déficient en oxygène. Nous avons pu ainsi déterminer pour la première fois l’association entre la voie d’HIF1α et celle d’IRE1α (un des trois senseurs de la voie de l’UPR « unfolded protein response ») dans le processus d’adaptation à l’oxygène des phagocytes mononucléaires. Nos résultats révèlent ensuite l’importance d’IRE1α (plus principalement son activité kinase) dans la production d’HIF1α. Nous démontrons finalement que la synergie entre les signalisations d’IRE1α et HIF1α pourrait être responsable du comportement pathogénique des phagocytes mononucléaires via leur libération de cytokines inflammatoires; ce qui participerait ainsi à la progression des rétinopathies. Collectivement, nos travaux ont permis d’identifier d’importants régulateurs de l’activité pathogénique des phagocytes mononucléaires. Nous montrons : 1) le rôle de Neuropiline-1 dans l’infiltration des phagocytes mononucléaires au niveau des zones endommagées de la rétine et 2) l’impact du mécanisme convergent entre les voies d’IRE1α et HIF1α sur leur sécrétion de facteurs pro-inflammatoires durant les rétinopathies. Nos résultats offrent une base pour le développement de nouvelles stratégies thérapeutiques (ciblant Neuropiline-1, IRE1α et HIF1α) dans le traitement de maladies oculaires et d’autres pathologies caractérisées par une inflammation excessive. / Proliferative ischemic retinopathies such as proliferative diabetic retinopathy (PDR) and retinopathy of prematurity (ROP) are the principal causes of vision loss in working age and pediatric populations of industrialized countries. Although they display different initial triggers, proliferative ischemic retinopathies are biphasic ocular diseases that affect retinal vessels. There is an initial degeneration of the microvasculature, followed by a hypoxic stress on the retina. This triggers a second phase of deregulated and destructive blood vessel growth within the retina. Given this sequence of events and prominent clinical features, the currently most widely used local ocular therapeutic interventions directly target pathological blood vessel growth, yet present a number of non-desirable off-target effects such as the destruction of the retina itself. In fact, currently available treatments for proliferative ischemic retinopathies present non-negligible side effects, such as cataract formation with intravitreal use of corticosteroid or reduced visual field with laser-based photocoagulation surgery. Similarly, the anti-VEGF (Vascular endothelial growth factor) therapy may be associated with thromboembolic events, neuronal toxicity and atrophy when used as frequent long-term treatment given the fact that VEGF serves a vaso- and neuro-protective factor in the retina. Overcoming these therapeutic limitations and exploring novel pharmacological avenues are therefore required to ameliorate the safety profiles of current interventions. In our first study, we describe a novel mechanism by which severely stressed neuronal cells in the diabetic retina provoke destructive inflammation in the eye. We demonstrate that the multi-ligand co-receptor Neuropilin-1, VEGF and Semaphorin3A act as potent attractants for a specialized population of immune cells (mononuclear phagocytes) that later promote the exaggerated pathological vessel growth associated with the disease progression. Importantly, we provide evidence for a novel pharmacological intervention that reduces the inflammation associated with pathological retinal vessel growth. Our findings also suggest that current anti-VEGF therapies (a popular treatment for ocular vascular diseases) may in part be effective by reducing destructive ocular inflammation. In our second study, we provide evidence that those mononuclear phagocytes activate HIF1α in the hypoxic and inflamed retina. After using the unbiased proteomic approach of tandem mass spectrometry, we were able to identify HIF1α partners and found a novel link between HIF1α and the UPR (unfolded protein response) sensor IRE1α. Our data next established the crucial role of IRE1α (precisely via its kinase activity) in HIF1α production. We also suggested that the synergy between IRE1α and HIF1α pathways may be responsible of the pathogenic activity of the hypoxic mononuclear phagocytes via their secretion of inflammatory cytokines, thus contributing to the progression of the retinopathy. Collectively, our work identifies important regulators of the pathogenic activity of mononuclear phagocytes. We show that: 1) Neuropilin-1 promotes the infiltration of mononuclear phagocytes in the retina and 2) the convergent mechanism between IRE1α and HIF1α pathways is responsible for their release of pro-inflammatory factors during retinopathy. Our results could be used as a basis for the development of alternative therapeutic strategies (targeting Neuropilin-1, IRE1α and HIF1α) to treat ocular diseases or other pathologies characterized by an excessive inflammation.