Microglia Purinergic Receptor-Mediated Neuroinflammation In Alzheimer’s Disease Neurodegeneration involves a complicated cascade of homeostatic dysfunction that converges on neuron loss and cognitive decline, involving complex immune, metabolic, and cell cell crosstalk pathways. The complicated interplay and heterogeneous nature of these factors in the brain make therapeutic development challenging. Recent advances have placed the immune system as an important driver of neurodegeneration both mechanistically and genetically. Microglia are the professional phagocytes that inhabit the brain and direct these inflammatory pathways, which can have reparative or destructive outcomes on the brain parenchyma. While various genetic risk factors for neurodegeneration reside in microglia, how these trigger and facilitate disease requires further investigation. In the present dissertation, I investigate inflammatory activation in microglia upon various damage or pathology-associated stimuli by utilizing a primary human monocyte-derived microglia-like cell (MDMi) model from a diverse donor cohort, which allows for the examination of genetically driven differences. I find that MDMi stimulated through ATP-mediated P2RX7 activation display reduced phagocytic function for amyloid beta uptake, and this pathway is also influenced by individual donors’ SPI1 genotype which has been associated with Alzheimer’s disease in previous computational studies. These experiments demonstrate functional outcomes related to AD genetics in immune cells. Previous computational studies have identified cognitive-decline associated gene modules expressed in human brain tissues from late-stage AD. I conducted in vitro follow up experiments to interrogate these genetic findings which is crucial for validating RNA sequencing data in a biological model. To interrogate differential MDMi inflammatory pathways, I treated cells with the toxic immunostimulatory molecule lipopolysaccharide (LPS), or its non-toxic derivative monophosphoryl lipid A (MPLA) which has positive immune properties currently utilized in vaccine adjuvants. My results indicated that individual gene expression in this module does not shift in a uniform manner upon LPS or MPLA challenge, suggesting more nuanced in vitro interrogation is required to identify conditions propagating this end stage disease phenotype. Microglia serve as the primary immune cells of the brain but also interact closely with astrocytes, large glial cells that facilitate neuronal homeostasis and are central players in AD due to their high apolipoprotein (APOE) production. Given the newly appreciated role of cellular crosstalk in neurological disease pathogenesis, I sought to optimize a protocol for isolation of primary mouse astrocytes for coculture with MDMi and investigation of non-direct cell contact interactions through astrocyte supernatants. Described in this dissertation is my optimized protocol for purified mouse astrocyte isolation from mice expressing humanized APOE2, APOE3, or APOE4. By developing this model, I was able to discern differential changes to MDMi gene expression in the presence of APOE2, 3, or 4 astrocyte supernatants. Verification of these tools allows further exploration of APOE genotype on glial crosstalk and downstream AD pathology. Overall, this work uncovers important mechanisms of human microglia activation through AD genetics and extracellular P2RX7 receptor behavior. By interrogating these scientific questions in a human microglia model derived from donors of various genetic and age backgrounds, we can assess how real biological variation modulates canonical inflammatory pathways. This adds powerful clinical relevance as AD and other neurodegenerative conditions can present a very heterogenous phenotype pathologically and therefore may require the nuance of more personalized medicine therapeutically.

Neurodegenerative disorders are characterised by the activation of brain-resident microglia cells and by the infiltration of peripheral T cells. However, their interplay in disease has not been clarified yet. It is difficult to investigate complex cellular dynamics in living animals, and simple two-dimensional (2D) cell culture models do not resemble the soft 3D structure of brain tissue. Therefore, we developed a biomimetic 3D in vitro culture system for co-cultivation of microglia and T cells. As the activation and/or migration of immune cells in the brain might be affected by components of the extracellular matrix, defined 3D fibrillar collagen I-based matrices were constructed and modified with hyaluronan and/or chondroitin sulphate, resembling aspects of brain extracellular matrix. Murine microglia and spleen-derived T cells were cultured alone or in co-culture on the constructed matrices. Microglia exhibited in vivo-like morphology and T cells showed enhanced survival when co-cultured with microglia or to a minor degree in the presence of glia-conditioned medium. The open and porous fibrillar structure of the matrix allowed for cell invasion and direct cell-cell interaction, with stronger invasion of T cells. Both cell types showed no dependence on the matrix modifications. Microglia could be activated on the matrices by lipopolysaccharide resulting in interleukin-6 and tumour necrosis factor-α secretion. The findings herein indicate that biomimetic 3D matrices allow for cocultivation and activation of primary microglia and T cells and provide useful tools to study their interaction in vitro.

Communication between the nervous and immune system is crucial for development, homeostasis and response to injury. Before the onset of neurogenesis, microglia populate the central nervous system, serving as resident immune cells over the course of life. Here, we describe new roles of an uncharacterized transcript upregulated by neurogenic progenitors during mouse corticogenesis: 4931414P19Rik (hereafter named P19). Overexpression of P19 cell-extrinsically inhibited neuronal migration and acted as chemoattractant of microglial cells. Interestingly, effects on neuronal migration were found to result directly from P19 secretion by neural progenitors triggering microglia accumulation within the P19 targeted area. Our findings highlight the crucial role of microglia during brain development and identify P19 as a previously unreported player in the neuro-immune crosstalk.

Neurodevelopmental disorders, such as autism spectrum disorders, schizophrenia, and attention deficit hyperactivity disorder, are a heterogeneous set of developmental disorders affecting the central nervous system. Studies into their etiology remain challenging, as neurodevelopmental disorders frequently present with a wide range of biological, behavioral, and comorbid symptomologies. Increasing epidemiological reports of antibiotic use during pregnancy as a significant correlate of subsequent mental disorder diagnosis in children suggest a mechanism of influence via the maternal gut-fetal brain axis. Importantly, antibiotics cause dysbiosis of the gut microbiome and disrupt the delicate composition of the microbial inoculum transferred from mother to child, which is critical for development of the immune system and holds implications for long-term health outcomes. The research objective of this dissertation is to reveal a causal mechanism of maternal microbial influence on neurodevelopment by examining the brain's resident immune cells, microglia, and corresponding behavioral outcomes in a mouse model of antibiotics-driven maternal microbiome dysbiosis (MMD). We identify early gross motor deficits and social behavior impairments in offspring born to MMD dams, which paralleled hyperactivated microglia in brain regions specific to cognition and social reward. The MMD microglia also exhibited altered transcriptomic signatures reflective of premature cellular senescence that support evidence of impaired synaptic modeling found in MMD brains. We report that these deficits are rescued in the absence of Cx3cr1, a chemokine receptor expressed ubiquitously on microglia, to highlight a pathway in which maternal microbiota may signal to neonatal microglia to undergo appropriate neurodevelopmental actions. Finally, we characterize Lactobacillus murinus HU-1, a novel strain of an important gut bacterium found in native rodent microbiota, and demonstrate its use as a probiotic to restore microglial and behavioral dysfunction in MMD offspring. / Ph. D. / Population studies on neurodevelopmental disorders, such as autism spectrum disorders, schizophrenia, and attention deficit hyperactivity disorder, highlight antibiotic use during pregnancy as a major correlate of subsequent diagnoses in children. These findings support a growing body of evidence from animal and human studies that the microbial ecosystems (“microbiome”) found in and on our bodies play significant roles in mental health, including mood, cognition, and brain function. Importantly, antibiotics during pregnancy create an imbalance of the gut microbiome (“dysbiosis”) and disrupt the microbial inoculum transferred from mother to child, which is critical for maturation of the infant immune system and holds implications for long-term health outcomes. Thus, the research objective of this dissertation is to identify a mechanism of influence from the mother’s gut to the neonate’s brain by examining the brain’s resident immune cells (“microglia”) in a mouse model of antibiotics-driven maternal microbiome dysbiosis (MMD). We uncover autism-like behavioral deficits and dysfunctional microglia in MMD offspring, and characterize signaling cues specific to microglia by which improper neurodevelopment may be taking place. We also reveal that the detrimental effects of MMD are reversed in mice born to mothers pretreated with a probiotic candidate, Lactobacillus murinus HU-1, to suggest maternally-derived Lactobacillus may help to mediate proper neurodevelopment.

Traumatic brain injury (TBI) elicits an immediate neuroinflammatory response that involves resident glia and infiltrating peripheral immune cells that coordinate tissue damage and functional deficits. The activation of resident microglial has been associated with a change in their morphology from a branched-like ramified cell to an ameboid state. This activation is thought to initiate a pro-inflammatory response leading to the release of neurotoxic, immune chemoattractant, and antigen-presenting signals. Subsequently, peripheral-derived immune cells (PICs), such as neutrophils and monocytes, travel to the site of injury and help coordinate this response. However, little is known regarding whether PICs influence the progressive activation state of microglia in the acute and chronic phases of injury. Overactivation of microglia can lead to neuroinflammation-mediated tissue damage and death or dysfunction of healthy neurons. Therefore, understanding how microenvironmental cues may regulate the microglial response may aid in strategies to retool their activation state in the brain. EphA4 receptor tyrosine kinase has been identified as a potential cell-to-cell contact protein on PICs that could be involved in the inflammatory changes following TBI. While microglial activation changes have been described in TBI models, the mechanistic role of infiltrating peripheral-derived immune cell (PIC) recruitment on microglial fate and function is not well understood. The purpose of my project is to gain a better understating of the temporospatial influence that EphA4-expressing PICs, specifically monocyte/macrophages, have on microglial proliferation, survival, activation phenotype, and debris clean-up using bone marrow GFP chimeric mice and the cortical contusion injury TBI model. / Doctor of Philosophy / Traumatic brain injury (TBI) triggers an immediate response from the brain's immune system, involving both local glial cells and immune cells from outside the brain. These cells work together to mediate the initial injury but, in some cases, cause development of a secondary injury. Microglia, the brain's resident immune cell, change their shape and behavior when activated by a TBI, becoming more aggressive and releasing inflammatory proteins. At the same time, immune cells from the bloodstream, like neutrophils and monocytes, rush to the injury site to assist. Yet, it's unclear how these immune cells affect microglia over time during the injury's acute and chronic phases. If microglia become too active, they can cause further damage to brain tissue and harm healthy neurons. Therefore, understanding the signals that control microglial activity could help us develop therapies to manage brain inflammation. One protein of interest in this process is the EphA4 receptor found on immune cells, which might play a crucial role in inflammation following TBI. While we know that microglia change post-TBI, we don't fully understand how the recruitment of immune cells from outside the brain affects them. My research aims to clarify how EphA4-expressing immune cells, especially monocytes/macrophages, influence microglia in terms of growth, behavior, and their ability to mediate a TBI.

Introduction: One characteristic of Alzheimer’s disease is the formation of amyloid-b plaques, which are typically linked to neuroinflammation and surrounded by inflammatory cells such as microglia and infiltrating immune cells. Methods: Here, we describe nonneurogenic doublecortin (DCX) positive cells, DCX being generally used as a marker for young immature neurons, at sites of amyloid-b plaques in various transgenic amyloid mouse models and in human brains with plaque pathology. Results: The plaque-associated DCX1 cells were not of neurogenic identity, instead most of them showed coexpression with markers for microglia (ionized calcium-binding adapter molecule 1) and for phagocytosis (CD68 and TREM2). Another subpopulation of plaque-associated DCX1 cells was negative for ionized calcium-binding adapter molecule 1 but was highly positive for the panleukocyte marker CD45. These hematopoietic cells were identified as CD3-and CD8-positive and CD4-negative T-cells. Discussion: Peculiarly, the DCX1/ionized calcium-binding adapter molecule 11 microglia and DCX1/CD81 T-cells were closely attached, suggesting that these two cell types are tightly interacting and that this interaction might shape plaque pathology.

The placenta plays a vital role in a healthy pregnancy by supporting the intricacies of fetal development. Over 10% of pregnancies experience impaired placental function, resulting in the loss of critical neuroactive steroids the fetal brain cannot yet make, thus leaving them vulnerable to perinatal brain injury and abnormal neurodevelopment. However, this vulnerability is not always equal. Many neurodevelopmental disorders exhibit a sex bias in incidence and severity. I hypothesize that loss of placental support during pregnancy results in sex differences in both behavioral presentation as well as on the cellular and transcriptomic levels. Utilizing the akr1c14cyp19aKO (plKO) mouse model, which features placenta-specific allopregnanolone (ALLO) knockdown, I investigated the sex specific impact of placental hormones on cerebellar development. Here I show that placental ALLO is essential for cerebellar white matter development and inflammatory regulation via microglial function. Male mice without placental ALLO exhibit signs of placental inflammation, accelerated postnatal myelination, and defects in microglial phagocytosis of excess myelin. Alternatively, females seem to be more resilient with a progressive anti-inflammatory profile across development and reduced myelination. Additionally male plKO show autism-like behaviors such as deficits in social behavior and increased stereotyped behavior. The females do not exhibit this phenotype. My main goals were threefold; to investigate how male and female inflammatory profiles differ and where this difference originates, to investigate how this inflammation impacts microglia and thereby oligodendrocytes, and how I can alter microglial function in a way to improve plKO outcomes. Mechanistically, these changes appear to be in part due to baseline sex differences in response to inflammatory stimuli which prime microglia to differentially support the surrounding white matter. Together, this work supports a novel link between placental ALLO loss, microglial function, and sex specific presentation of neurodevelopmental disorders.

Die Invasion von Leukozyten in das zentrale Nervensystem (ZNS) ist ein wesentlicher Bestandteil bei der Pathogenese von Hirnverletzungen sowie akuten und chronischen Entzündungsvorgängen im Gehirn. Mikrogliazellen, die überwiegende Population immunkompetenter Zellen des ZNS, stellen die erste Verteidigungslinie im Hinblick auf Verletzungen und Erkrankungen des Gehirns dar. Im Rahmen vieler neurodegenerativer Erkrankungen wird die Zerstörung von Neuronen, aber auch die kollaterale Gewebsschädigung auf die Aktivierung der Mikrogliazellen zurückgeführt. Die vorliegende Arbeit beschreibt erstmalig einen regulatorischen Aktivierungszustand der Mikroglia (CD40dimCD86dimIL-10high), der zur Induktion regulatorischer Foxp3+ T-Zellen (Treg) führt. Die Stabilität und funktionelle Aktivität Mikroglia-induzierter Treg konnte sowohl in vitro als auch in vivo gezeigt werden. In vitro inhibierten sie die Proliferation antigen-spezifischer Effektorzellen, in vivo führte ein adoptiver Transfer der regulatorischen T-Zellen zur Abmilderung des Krankheitsverlaufes experimentell induzierter, autoimmuner Enzephalomyelitis (EAE). Mikrogliazellen unterstützten sowohl die Proliferation bereits ausgebildeter regulatorischer T-Zellen als auch deren Differenzierung aus naiven T-Zellen. Die Induktion regulatorischer T-Zellen durch Mikroglia war Major Histocompatibility Complex (MHC)-II-abhängig und antigenspezifisch. Für Untersuchungen zur in vivo Relevanz wurden MHC-II-chimäre Mäuse generiert und eine Läsion im entorhinalen Kortex gesetzt. Fehlte MHC-II in ZNS-residenten Zellen, wurden weniger regulatorische T-Zellen pro Leukozyt in die lädierten Hemispheren rekrutiert. Zusammenfassend demonstrieren diese Ergebnisse das Modulationspotential von Mikrogliazellen auf die CD4+ T-Zellantwort. Die Mikroglia-induzierte Differenzierung und Proliferation von Foxp3+ regulatorischen T-Zellen ist ein möglicher Mechanismus der Regulation von Entzündungsvorgängen im ZNS durch Mikrogliazellen. / The invasion of leukocytes into the central nervous system (CNS) is a key event in the pathogenesis of CNS injury and acute or chronic inflammatory neurological diseases. However, regulatory mechanisms of local innate immune responses that limit CNS inflammation are only poorly understood. Microglia are the predominant innate immune cells of the brain and present the first line of defence in CNS injury or disease. In the context of neurodegenerative disease, microglia activation accounts for collateral tissue damage and neurodestruction. This thesis for the first time describes a regulatory microglia phenotype (MHCII+CD40dimCD86dimIL-10high) that induced a strong Foxp3+ regulatory T cell (Treg) response. Microglia-induced Treg cells were stable and functionally active in vitro by inhibiting antigen-specific proliferation of effector T cells and in vivo, by attenuating experimental autoimmune encephalomyelitis (EAE) disease course after adoptive transfer. The data also suggested that regulatory microglia can mediate both, proliferation of Foxp3+ Treg cells and de novo differentiation from naive CD4+ T cells. Microglia-mediated Treg induction was proven to be MHCII and antigen-dependent. Using entorhinal cortex lesion (ECL) as a brain injury mouse model, diminished Foxp3+ Treg cell recruitment per infiltrated leukocyte in chimeric mice lacking MHCII specifically in the CNS was demonstrated, indicating in vivo relevance of antigen presentation by brain resident cells. Taken together, these findings demonstrate that microglial cells can directly modulate CD4+ T cell responses by regulating molecule levels for efficient antigen presentation and levels of secreted cytokines and chemokines. Microglia-mediated differentiation and proliferation of Foxp3+ Treg cells can be one of the mechanisms how microglia contribute to local immune homeostasis and limit CNS inflammation.

Streptococcus suis de type 2 est un microorganisme pathogène d’importance chez le porc. Il est la cause de différentes pathologies ayant comme caractéristique commune la méningite. C’est également un agent émergeant de zoonose : des cas cliniques humains ont récemment été rapportés en Asie. Cependant, la pathogénèse de S. suis n’est pas encore complètement élucidée. Jusqu’à présent, la réponse pro-inflammatoire initiée par S. suis n’a été étudiée qu’in vitro. L’étude du choc septique et de la méningite requiert toujours des modèles expérimentaux appropriés. Au cours de cette étude, nous avons développé un modèle in vivo d’infection chez la souris qui utilise la voie d’inoculation intra-péritonéale. Ce modèle a servi à l’étude de la réponse pro-inflammatoire associée à ce pathogène, tant au niveau systémique qu’au niveau du système nerveux central (SNC). Il nous a également permis de déterminer si la sensibilité aux infections à S. suis pouvait être influencée par des prédispositions génétiques de l’hôte. Le modèle d’infection par S. suis a été mis au point sur des souris de lignée CD1. Les résultats ont démontré une bactériémie élevée pendant les trois jours suivant l’infection. Celle-ci était accompagnée d’une libération rapide et importante de différentes cytokines pro-inflammatoires (TNF-α, IL-6, IL-12p40/p70, IFN-ɣ) et de chémokines (KC, MCP-1 and RANTES), qui ont entraîné un choc septique et la mort de 20 % des animaux. Ensuite, pour confirmer le rôle de l’inflammation sur la mortalité et pour déterminer si les caractéristiques génétiques de l’hôte pouvaient influencer la réponse inflammatoire et l’issue de la maladie, le modèle d’infection a été étendu à deux lignées murines consanguines différentes considérées comme résistante : la lignée C57BL/6 (B6), et sensible : la lignée A/J. Les résultats ont démontré une importante différence de sensibilité entre les souris A/J et les souris B6, avec un taux de mortalité atteignant 100 % à 20 h post-infection (p.i.) pour la première lignée et de seulement 16 % à 36 h p.i. pour la seconde. La quantité de bactéries dans le sang et dans les organes internes était similaire pour les deux lignées. Donc, tout comme dans la lignée CD1, la bactériémie ne semblait pas être liée à la mort des souris. La différence entre les taux de mortalité a été attribuée à un choc septique non contrôlé chez les souris A/J infectées par S. suis. Les souris A/J présentaient des taux exceptionnellement élevés de TNF-α, IL-12p40/p70, IL-1β and IFN- γ, significativement supérieurs à ceux retrouvés dans la lignée B6. Par contre, les niveaux de chémokines étaient similaires entre les lignées, ce qui suggère que leur influence est limitée dans le développement du choc septique dû à S. suis. Les souris B6 avaient une production plus élevée d’IL-10, une cytokine anti-inflammatoire, ce qui suppose que la cascade cytokinaire pro-inflammatoire était mieux contrôlée, entraînant un meilleur taux de survie. Le rôle bénéfique potentiel de l’IL-10 chez les souris infectées par S. suis a été confirmé par deux approches : d’une part en bloquant chez les souris B6 le récepteur cellulaire à l’IL-10 (IL-10R) par un anticorps monoclonal anti-IL-10R de souris et d’autre part en complémentant les souris A/J avec de l’IL-10 de souris recombinante. Les souris B6 ayant reçu le anticorps monoclonal anti-IL-10R avant d’être infectées par S. suis ont développé des signes cliniques aigus similaires à ceux observés chez les souris A/J, avec une mortalité rapide et élevée et des taux de TNF-α plus élevés que les souris infectées non traitées. Chez les souris A/J infectées par S. suis, le traitement avec l’IL-10 de souris recombinante a significativement retardé l’apparition du choc septique. Ces résultats montrent que la survie au choc septique dû à S. suis implique un contrôle très précis des mécanismes pro- et anti-inflammatoires et que la réponse anti-inflammatoire doit être activée simultanément ou très rapidement après le début de la réponse pro-inflammatoire. Grâce à ces expériences, nous avons donc fait un premier pas dans l’identification de gènes associés à la résistance envers S. suis chez l’hôte. Une des réussites les plus importantes du modèle d’infection de la souris décrit dans ce projet est le fait que les souris CD1 ayant survécu à la septicémie présentaient dès 4 jours p.i. des signes cliniques neurologiques clairs et un syndrome vestibulaire relativement similaires à ceux observés lors de méningite à S. suis chez le porc et chez l’homme. L’analyse par hybridation in situ combinée à de l’immunohistochimie des cerveaux des souris CD1 infectées a montré que la réponse inflammatoire du SNC débutait avec une augmentation significative de la transcription du Toll-like receptor (TLR)2 et du CD14 dans les microvaisseaux cérébraux et dans les plexus choroïdes, ce qui suggère que S. suis pourrait se servir de ces structures comme portes d’entrée vers le cerveau. Aussi, le NF-κB (suivi par le système rapporteur de l’activation transcriptionnelle de IκBα), le TNF-α, l’IL-1β et le MCP-1 ont été activés, principalement dans des cellules identifiées comme de la microglie et dans une moindre mesure comme des astrocytes. Cette activation a également été observée dans différentes structures du cerveau, principalement le cortex cérébral, le corps calleux, l’hippocampe, les plexus choroïdes, le thalamus, l’hypothalamus et les méninges. Partout, cette réaction pro-inflammatoire était accompagnée de zones extensives d’inflammation et de nécrose, de démyélinisation sévère et de la présence d’antigènes de S. suis dans la microglie. Nous avons mené ensuite des études in vitro pour mieux comprendre l’interaction entre S. suis et la microglie. Pour cela, nous avons infecté des cellules microgliales de souris avec la souche sauvage virulente (WT) de S. suis, ainsi qu’avec deux mutants isogéniques, un pour la capsule (CPS) et un autre pour la production d’hémolysine (suilysine). Nos résultats ont montré que la capsule était un important mécanisme de résistance à la phagocytose pour S. suis et qu’elle modulait la réponse inflammatoire, en dissimulant les composants pro-inflammatoires de la paroi bactérienne. Par contre, l’absence d’hémolysine, qui est un facteur cytotoxique potentiel, n’a pas eu d’impact majeur sur l’interaction de S. suis avec la microglie. Ces études sur les cellules microgliales ont permis de confirmer les résultats obtenus précédemment in vivo. La souche WT a induit une régulation à la hausse du TLR2 ainsi que la production de plusieurs médiateurs pro-inflammatoires, dont le TNF-α et le MCP-1. S. suis a induit la translocation du NF-kB. Cet effet était plus rapide dans les cellules stimulées par le mutant déficient en CPS, ce qui suggère que les composants de la paroi cellulaire représentent de puissants inducteurs du NF-kB. De plus, la souche S. suis WT a stimulé l’expression de la phosphotyrosine, de la PKC et de différentes cascades liées à l’enzyme mitogen-activated protein kinase (MAPK). Cependant, les cellules microgliales infectées par le mutant déficient en CPS ont montré des profils de phosphorylation plus forts et plus soutenus que celles infectées par le WT. Finalement, la capsule a aussi modulé l’expression de l’oxyde nitrique synthétase inductible (iNOS) induite par S. suis et par la production subséquente d’oxyde nitrique par la microglie. Ceci pourrait être lié in vivo à la neurotoxicité et à la vasodilatation. Nous pensons que ces résultats contribueront à une meilleure compréhension des mécanismes sous-tendant l’induction de l’inflammation par S. suis, ce qui devrait permettre, d’établir éventuellement des stratégies plus efficaces de lutte contre la septicémie et la méningite. Enfin, nous pensons que ce modèle expérimental d’infection chez la souris pourra être utilisé dans l’étude de la pathogénèse d’autres bactéries ayant le SNC pour cible. / Streptococcus suis serotype 2 is an important swine pathogen responsible for diverse infections, meningitis being its most striking feature. In addition, it is an emerging agent of zoonosis, which has gained worldwide attention due to important outbreaks in Asia. Understanding the pathogenesis of S. suis infections still represents a challenge. Up to present, the pro-inflammatory response due to S. suis has only been studied in vitro, and there is still a great need of appropriate experimental models for both septic shock and meningitis. In the present study, we successfully developed an in vivo model of S. suis infection in adult mice infected by the intraperitoneal route. This model served to investigate the pro-inflammatory events that take place at both the systemic and Central Nervous System (CNS) levels associated with this important pathogen. In addition, this model was useful to determine if susceptibility to S. suis infection may be influenced by the genetic background of the host. The mouse model of S. suis infection was standardized in CD1 mice. Results showed sustained bacteremia during the 3 days post-infection (p.i.), accompanied by a quick and substantial release of different pro-inflammatory cytokines (TNF-α, IL-6, IL-12p40/p70, IFN-ɣ) and chemokines (KC, MCP-1 and RANTES) that lead to septic shock and 20% mortality in mice. Once the hallmark of the septic phase of S. suis infection was established in CD1 mice, research continued with the objective to confirm the role of inflammation in mortality and to determine if the genetic background of the host may influence the inflammatory response toward this pathogen and the further outcome of the disease. For this, the mouse model of S. suis infection was used with two genetically different inbred mouse strains, this is, C57BL/6 (B6) and A/J mice, which are considered as the prototype of Th1-type and Th2-type mice, respectively. Results demonstrated a striking susceptibility to S. suis infection in A/J mice in comparison to B6 mice, with 100% mortality in the former mice strain at 20 h p.i., and 16 % mortality at 36 h p.i. for the latter. Very interestingly, and similarly to CD1 mice, bacteremia did not seem to be responsible for the death of mice, as both mice strains presented similar amounts of bacteria in blood and organs. Thus, it was postulated that the higher mortality in S. suis-infected A/J mice was due to uncontrolled septic shock. In fact, A/J mice presented very high levels of TNF-α, IL-12p40/p70, IL-1β and IFN-ɣ, that significantly exceeded those found in B6 mice. Remarkably, chemokine levels were similar between strains, suggesting their limited participation in the development of septic shock by S. suis. A greater survival of B6 mice was partially related to a better regulation of the pro-inflammatory cytokine cascade, as they showed a higher production of the anti-inflammatory cytokine IL-10 than A/J mice. The potential beneficial role of the IL-10 in mice infected with S. suis was confirmed using two approaches: the first, by blockage of the cell receptor of IL-10 (IL-10R) with an anti-mouse IL-10R monoclonal antibody (Mab) in B6 mice and the second by administrating recombinant mouse (rm)IL-10 (rmIL-10) to A/J mice. B6 mice that received the IL-10R MAb treatment before challenge with S. suis developed a clinical acute disease similar to that observed with A/J mice, with a striking and rapid increase in mortality and higher levels of TNF-α in comparison to those of infected mice that did not receive the treatment. Controversially, treatment with rmIL-10 significantly delayed the onset of septic shock in A/J mice infected with S. suis. These results show that survival from S. suis septic shock requires a tight regulation of pro- and anti-inflammatory mechanisms, and that the latter should be activated at the same time or soon after the onset of the pro-inflammatory response. This part of the study may represent a first step in the identification of host genes associated with resistance against S. suis. One of the most important achievements of the mouse model of infection described in this project is the development of distinct clinical signs of neurological disease in CD1 mice from 4 days p.i. Indeed, in CD1 mice that survived sepsis due to S. suis infection, clinical signs of neurological disease and vestibular syndrome, which are quite similar to those observed in clinical cases of S. suis meningitis in both pigs and humans, were observed. Studies of the brains of infected CD1 mice using in situ hybridization combined with immunocytochemistry, demonstrated that the CNS inflammatory response began with a significant increase in the transcription of Toll-like receptor (TLR)2 and CD14 initially in the brain microvasculature and choroid plexuses, suggesting that S. suis may use these structures as portals of entry to the brain. There also was activation of NF-κB (as indicated by transcriptional activation of IκBα as a reporter system) and TNF-α, IL-1β and MCP-1, mainly in cells identified as microglia and to a lesser extent in astrocytes. These signals reached different brain structures, mainly the brain cortex, corpus callosum, hippocampus, choroid plexuses, thalamus, hypothalamus and meninges. All of these pro-inflammatory events were associated with extensive areas of inflammation and necrosis, severe demyelination and presence of antigens of S. suis inside microglia. In vitro studies were conducted in order to better understand the interactions of S. suis and microglia. For this, mouse microglia were infected with a virulent wild type (WT) strain of S. suis. Two isogenic mutants deficient in capsule (CPS) or hemolysin production (suilysin, SLY) respectively, were also included for comparative purposes. The CPS was important for S. suis resistance to phagocytosis, and it also modulated the inflammatory response by hiding pro-inflammatory components from the bacterial cell wall. On the other hand, the absence of SLY, a potential cytotoxic factor, did not have a major impact on S. suis interactions with microglia. Studies with microglia helped to confirm previous findings in vivo in mice, as the WT S. suis strain induced the up-regulation of TLR2 and the production of several pro-inflammatory mediators, including TNF-α and MCP-1. As observed in mice, S. suis induced NF-kB translocation, which was more rapid for cells stimulated with the CPS-deficient mutant, suggesting that bacterial cell wall components are potent inducers of NF-kB. Moreover, WT S. suis promoted phosphotyrosine, PKC and different mitogen-activated protein kinase (MAPK) events. However, microglia infected with the CPS-deficient mutant showed overall stronger and more sustained phosphorylation profiles. Finally, the CPS also modulated S. suis-induced inducible nitrogen oxide synthase (iNOS) expression and further nitric oxide production in microglia, which could be related to neurotoxicity and vasodilatation in vivo. We are confident that our results may help to more fully understand the mechanisms underlying S. suis induction of inflammation, leading to the design of more efficient anti-inflammatory strategies for sepsis and meningitis. Finally, we believe this experimental model of infection in mice could also be useful for studying the pathogenesis of infections of the CNS, due to other bacteria.

A tuberculose (TB) é um grave problema de saúde pública. Somente no ano de 2004, cerca de 9 milhões de pessoas desenvolveram TB ativa e mais de 2 milhões de pessoas morreram da doença. O desenvolvimento de novos modelos experimentais de TB seriam de grande utilidade para para elucidar mecanismos fisiopatológicos da doença e testar esquemas terapêuticos para a prevenção e contenção da doença. Além disso, o desenvolvimento de novas vacinas torna-se indispensável como ferramenta de prevenção e controle da TB. A TB no sistema nervoso central (SNC), assim como em outros tecidos do organismo, promove a ativação de células inflamatórias. No SNC a micróglia desempenha este papel, sendo capaz de produzir ou ser influenciada por mediadores solúveis. Vários mediadores estão envolvidos nos mecanismos moleculares decorrentes da infecção e inflamação causados pela TB, entre eles: NFB, iNOS e VEGF. A ativação do NFB, um fator de transcrição citoplasmático que sob estímulo migra para o núcleo celular, tem íntima relação com a indução da iNOS e de VEGF. A resistência intracelular a patógenos, inclusive ao Mycobacterium tuberculosis, parece estar associada a expressão de iNOS em macrófagos. O óxido nítrico (NO) tem papel importante na comunicação intercelular, estimulando a síntese de mediadores inflamatórios, como as citocinas, e regulando sua própria produção endógena. Estas citocinas por sua vez também podem induzir a atividade do NFB e a expressão da iNOS e VEGF. O VEGF é um potente ativador de permeabilidade vascular e de angiogênese, envolvido na ruptura da barreira hemato-encefálica. Neste estudo, mostramos a caracterização morfológica e imuno-histoquímica de um modelo murino de TB no SNC, com a indução da doença pela inoculação de BCG. Com este modelo experimental obtivemos importantes resultados que podem esclarecer mecanismos envolvidos na fisiopatologia da neuro-TB humana. A indução de meningite e tuberculomas foi possível através da inoculação de 104 cfu de BCG no cerebelo de camundongos, por estereotaxia, e esta indução foi dependente do tempo. A confirmação do diagnóstico foi feita pela detecção de bacilos álcool-ácido resistentes (BAAR), nas lesões tuberculosas. Observamos, ao longo do tempo (1 a 6 dias; 1, 2, 4 e 8 semanas) o recrutamento de diferentes populações gliais (micróglia e astrócitos) no sítio de injeção. Houve aumento de produção e ativação NFB nas lesões tuberculosas, caracterizada pela translocação da molécula do citoplasma para o núcleo celular. Houve expressão de iNOS restrita às lesões tuberculosas, além do aumento de expressão de VEGF nestas lesões. Além disso, camundongos imunizados com a vacina gênica hsp65, contra a TB, não expressam VEGF em suas lesões. Esta vacina parece conferir um efeito protetor em nosso modelo experimental, reduzindo a expressão de VEGF, e consequentemente reduzindo seu efeito angiogênico decorrente do processo inflamatório. O recrutamento glial, e a produção de mediadores solúveis (NFB, iNOS e VEGF) pelo hospedeiro, em resposta à invasão do patógeno no SNC, parecem estar envolvidos na fisiopatologia da neurotuberculose, como demonstrado neste modelo experimental. Nosso modelo permitirá investigar fatores possivelmente responsáveis pelo desenvolvimento e manutenção de lesões tuberculosas no SNC. O objetivo final seria elucidar a fisiopatologia desta grave doença e compreender eventos moleculares envolvidos na produção de lesões. O conhecimento gerado poderá permitir o delineamento de terapias específicas e efetivas. / Tuberculosis (TB) is a serious public health problem; in 2004, 9 million people developed active TB and the disease killed 2 million patients. Development of experimental models and new vaccines are essential both to elucidate physiopathological mechanisms and to control the disease. This infection in the central nervous system (CNS), as in other tissues of the organism, activates inflammatory cells. In CNS, this role is performed by the microglia, which is capable of producing or be influenced by soluble mediators. Several mediators are involved in the molecular mechanisms of the infection and inflammation by mycobacteria , such as NFB, iNOS and VEGF. NFB activation, a cytoplasmic transcriptional factor that migrates to the cellular nucleus under stimuli, is involved with the iNOS and VEGF induction of expression. The intracellular resistance to Mycobacterium tuberculosis has been associated with iNOS expression in macrophage cells. Nitric oxide (NO) is crucial in intercellular communication, modulating the synthesis of mediators of inflammation, such as cytokines, and modulation itself. These cytokines induces NFB activity, and induces iNOS and VEGF expression. VEGF is a potent activator of vascular permeability and of angiogenesis and it is a factor involved in the breakdown of the blood brain-barrier in tuberculous meningitis. In this study, we showed the morphologic and immunohistochemistry characterization of an experimental model of TB in the CNS, with inoculation of BCG in mice. In this model we elicited important outcome that can elucidate mechanisms involved in the physiopathology of human neuron-TB. Induction of meningitis and tuberculomas were possible with stereotaxic inoculation of 104 cfu of BCG in mice cerebellum, in a time-dependent way. Diagnostic was confirmed by detection of alcohol-acid resistant bacilli (BAAR), in tuberculous lesions. We observed, the time-course (1 to 6 days; 1, 2, 4 e 8 weeks) of the recruitment of different glial populations (microglia and astrocytes) in the injection site. There was increased production and activation of NFB in the tuberculous lesions, it was characterized by its nuclear translocation from cytoplasm. There was iNOS expression only in the tuberculous lesions, and expression increased of VEGF in these lesions. Furthermore, mice immunizated with vaccine DNA-hsp65 there was no expression of VEGF in its lesions. This vaccine seems confer a protector effect in our experimental model, reducing the expression of VEGF, and then reducing its angiogenic effect derived from inflammatory process. Glial recruitment, and the soluble mediators production (NFB, iNOS e VEGF) by the host, producing in response to invasion of the pathogen in the CNS, has been involved in the pathophysiology of the neuro-TB, such as demonstrated in this experimental model. Our model will allow investigate possible factors responsible for the development and maintenance of tuberculous lesions in the CNS. The final aim is to elucidate the physiopathology of this serious illness and understand the molecular events involved in the production of the lesions. The knowledge created may permit to pave the way to delineate specific and effective therapies.