Clostridium difficile est une bactérie enteropathogène responsable d'infections intestinales dont les manifestations cliniques varient d’une simple diarrhée à une colite pseudomembraneuse parfois mortelle. L’une des problématiques majeures rencontrées dans la prise en charge des infections à C. difficile (ICD) est la survenue de récidives. Chez de nombreuses espèces bactériennes, la formation de biofilm est associée à la chronicité de l’infection. Le biofilm est un mode de vie dans lequel les bactéries sont engluées dans une substance polymérique qu'elles secrètent elles-mêmes. L’aptitude de C. difficile à former un biofilm in vitro a été clairement établie depuis 2012. Cependant aucune étude n’a montré jusqu’ici s’il est capable de former un biofilm in vivo. L’objectif de cette thèse était d’une part d’étudier différents paramètres impliqués dans la formation du biofilm in vitro et d’autre part de déterminer si C. difficile est capable de former un biofilm in vivo. Dans un premier temps, nous avons analysé la capacité de différentes souches de C. difficile (souches cliniques et souches de laboratoires modifiées génétiquement) à former un biofilm in vitro. Parmi ces dernières, la souche mutée pour le gène cwp84, codant une protéase de surface responsable de la maturation de la couche S de C. difficile, présente un biofilm particulièrement robuste et épais comparé à la souche parentale 630∆erm. L’activité protéolytique de Cwp84 est donc impliquée dans la formation du biofilm et module certaines propriétés de surface de la bactérie telles que l’hydrophobicité. Nous avons également étudié la composition en sucre de la matrice du biofilm, après une étape de mise au point qui a permis de déterminer les conditions permettant d’obtenir suffisamment de matériel. Les conditions retenues ont été les suivantes : formation de biofilm sur un support en verre, en présence de glucose et en milieu renouvelé. La présence d’un sucre qui possède un profil proche du PSII (polysaccharide associé à la surface des cellules planctoniques de C. difficile) dans la matrice du biofilm a été détecté par spectroscopie infrarouge. Dans un second temps, afin d’étudier l’aptitude de C. difficile à former un biofilm in vivo, différents modèles animaux ont été utilisés : un modèle de souris monoxéniques dans lequel plusieurs souches de C. difficile ont été testées (souches 630∆erm, mutant cwp84, R20291, P30) et un modèle de souris dixénique pour étudier la formation de biofilm mixte (C. difficile/Finegoldia magna et C. difficile/Clostridium scindens). Dans le modèle monoxénique, quelles que soient les souches testées, C. difficile est distribué de manière hétérogène tout au long de la surface du tissu intestinal. Les bactéries sont majoritairement retrouvées isolées sauf pour la souche R20291 qui forme le plus souvent des petits agrégats. Pour cette souche, différents marquages immunohistochimiques réalisés sur des coupes de cecum et de colon ont montré que la majorité des bactéries sont enchâssées dans de petites structures en 3 dimensions adhérentes à la couche du mucus. Le polysaccharide PSII est détecté en grande quantité à l'intérieur de cette structure. Ce composé étant présent dans la matrice de biofilm de C. difficile formé in vitro, ces résultats suggèrent que la souche R20291 pourrait s’organiser en biofilm dans le modèle de souris monoxénique. En modèle de souris dixéniques, nous avons montré que la présence de F. magna n’influe pas sur le niveau de colonisation de C. difficile, alors que l'association avec C. scindens semble être bénéfique aux deux bactéries puisqu'une augmentation de la population globale de deux espèces est observée comparé à la population présente dans chaque modèle mono-espèce. En conclusion, une souche productrice de biofilm in vitro semble être capable de s’organiser en une structure biofilm in vivo. Le rôle du biofilm dans l'étape de colonisation du colon par C. difficile et les rechutes des ICD devra être analysé. / Clostridium difficile is an enteropathogenic bacterium responsible for intestinal infections, the clinical symptoms vary from moderate diarrhea to pseudomembranous colitis, sometimes fatal. One of the major problems encountered in the management of C. difficile infections (DCI) is the occurrence of recurrences. In many bacterial species, biofilm formation is associated with the chronicity of infection. Biofilm is a way of life in which bacteria are entrapped in a polymeric substance secreted by the bacteria themselves. The ability of C. difficile to form an in vitro biofilm has been clearly established since 2012. However, no study has so far shown whether it is capable of forming a biofilm in vivo. The objective of this thesis was to analyze different parameters involved in the formation of biofilm in vitro and to determine whether C. difficile is able to form a biofilm in vivo.We first analyzed the ability of different strains of C. difficile to form a biofilm in vitro (clinical strains and strains genetically modified laboratories). Among the latter, the mutant strain for the cwp84 gene, encoding a surface-associated protease responsible for the maturation of the S layer of C. difficile, forms a particularly robust and thick biofilm compared to the 630Δerm parental strain. The proteolytic activity of Cwp84 is involved in the formation of biofilm and modulates certain surface properties of the bacterium such as hydrophobicity. We have also studied the polysaccharide composition of the biofilm matrix, after a development stage that allowed us to determine the optimal conditions for obtaining sufficient material. The conditions retained were the following: formation of biofilm on a glass support in the presence of glucose and in a renewed medium. We were able to determine by infrared spectroscopy the presence of a sugar which has a similar profile than the PSII (polysaccharide associated with the surface of C. difficile planktonic cells) in the matrix of the biofilm.Second, in order to study the ability of C. difficile to form a biofilm in vivo, different animal models were used: a monoxenic mouse model in which we tested several strains of C. difficile (strains 630Δerm, mutant cwp84, R20291, P30) and a dixenic mouse model to study the formation of mixed biofilm (C .difficile/Finegoldia magna and C. difficile/Clostridium scindens). In the monoxenic model, regardless of the strains tested, C. difficile is distributed heterogeneously throughout the intestinal tissue surface. The bacteria are mostly found isolated except for C. difficile R20291 which usually forms small aggregates. For this strain we have shown, thanks to various immunohistochemical labeling performed on cecum and colon sections, that the majority of the bacteria are embedded in a small 3-dimensional structures overlaying the mucus layer. The PSII polysaccharide is present in a large amount in this structure. As this compound has been detected in the in vitro C. difficile biofilm matrix, these results suggest strongly that the R20291 strain could be organized into biofilm structures in the monoxenic mouse model. In the dixenic mouse model, we have shown that the presence of F. magna does not influence the level of colonization of C. difficile, whereas the association with C. scindens seems to be beneficial to both bacteria since an increase of the global population is observed in this model compared to the population present in each single-species model.To conclude, an in vitro biofilm producing strain appears to be able to organize in a biofilm structure in vivo. The role of biofilm in the colonization step of the intestinal tract by C. difficile and in the occurrence of recurrences should be further analyzed.

Several factors can modulate the link between fetal disruptions and later-life illnesses. The main objective of this thesis was to determine, in a mouse model, the impact of prenatal stressor timing and offspring sex on prepubertal metabolic and mental health outcomes. C57BL/6 dams in the first or second trimester of pregnancy experienced a restraint stressor or were left undisturbed. Pups were weighed daily until postnatal day (PND) 21, at which time fat distribution was measured. Anxiety- and depressive-like behaviors were tested on PND19-20 in open field, elevated plus maze, splash and tail suspension tests. Second trimester stressed males gained more weight and had increased fat deposits surrounding the kidneys. Although anxiety- and depressive-like behaviors were not apparent in prenatally stressed offspring of either sex, females stressed in utero exhibited a hyperactive phenotype. This work is the first to show sex- and trimester-specific consequences of early pregnancy stressors in prepubertal offspring.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22882号 / 医博第4676号 / 新制||医||1047(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 髙折 晃史, 教授 濵﨑 洋子, 教授 羽賀 博典 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

The aim of this study was to establish a bacteria-induced mouse model of human IBD applicable for investigation of the role of commensal versus pathogenic bacteria in the onset and development of chronic intestinal inflammation. The mouse model introduced in the present work focused on chronic inflammatory reaction triggered by S. enterica infection. Furthermore, differences between the acute and chronic phase of infection were analyzed. Additionally, it was investigated, whether antibiotic treatment prior to infection has a significant impact on the course of the systemic or local mucosal immune response against S. enterica. Due to the significant increase of the IL-22 level reported from IBD patients, and the assumed crucial role of this immunoregulatory cytokine for human IBD a special focus of this study was to analyze the role of IL-22 in the established mouse model of bacteria-induced chronic colitis.

Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by the accumulation of amyloid-β peptide (Aβ) in the brain and intraneuronal hyperphosphorylated tau. Microglia in the brain adopt M1 (pro-inflammatory) or M2 (anti-inflammatory) phenotypes similar to peripheral monocytes. M1 microglia negatively affect neurogenesis and have reduced phagocytic capabilities whereas M2 microglia can enhance neurogenesis and support phagocytosis. Cluster of Differentiation-200 (CD200) is an anti-inflammatory glycoprotein physiologically expressed on neurons and lymphocytes, and its receptors (CD200R1 and CD200R3) are expressed on glia. Both AD patients and mouse models of AD show an age-related or Aβ-induced reduction in neural CD200 that may contribute to M1-skewing of microglia in AD. We hypothesize that CD200 skews microglia to an M2 phenotype, and that genetic over-expression of CD200 in transgenic mice expressing the Swedish familial AD mutation of human β-amyloid precursor protein (APP mice) can restore neurogenesis and enhance Aβ clearance in the hippocampus. In this study, we constructed a tetracycline-controlled transactivator-inducible adeno-associated virus serotype 2/1 expressing full-length CD200 (AAV2/1-CD200) or green fluorescent protein (AAV2/1- GFP). These were bilaterally injected into the hippocampi at 6 months of age, and mice were sacrificed at 12 months of age. AAV2/1-GFP-injected APP mice showed a reduction in number of proliferating neural stem cells (NSCs) by 65.0% and differentiating NSCs by 70.5% in the dentate gyrus compared to wild-type controls. AAV2/1-CD200 restored these neurogenic deficits to those of wild-type mouse levels. AAV2/1-CD200 reduced diffuse Aβ plaques in the hippocampal region by 65.5% compared to AAV2/1-GFP-injected APP mice, but did not alter thioflavin-S-positive compact plaques as measured by protein and immunohistochemical assays. In vitro studies demonstrated that CD200-stimulated microglia co-cultured in transwells increased differentiation and complexity of neural stem cells. CD200 also directly enhanced Aβ phagocytosis by microglia. CD200 enhanced expression of the adaptor protein TYRO protein tyrosine kinase binding protein (TYROBP), suggesting this may be the mechanism by which CD200 enhances phagocytosis of Aβ. Overall, the data presented here indicate that CD200 is a plausible therapeutic agent in patients with AD to enhance neural differentiation and microglial-mediated clearance of Aβ.

Chikungunya virus (CHIKV) is classified as an alphavirus in the Togaviridae family. This virus is known to rely on Aedes arthropod vectors for its dissemination. Human infection is characterized by rash, high fever, and severe chronic polyarthritis that can last for years. Recently, efforts in developing animal models have been made in an attempt to better understand CHIKV pathogenesis. CHIKV infection starts with a 7 to 10 day long febrile acute phase, in which most of the symptoms occur (rash, fever, and incapacitating pain in joints and muscle). Once the immune system clears most of the viral infection, a chronic phase starts in as many as 70% of the infected patients. Long term virus-related polyarthralgia is the hallmark of the CHIKV chronic phase. It is believed that CHIKV-infected macrophages infiltrate the joints during the acute phase, and CHIKV infects joint tissue and persists in it. Research into the effects of CHIKV infection in human and murine macrophages revealed that CHIKV-infected human macrophages produce high amounts of virions as well as induce the production of pro-inflammatory cytokines and monocyte recruiting chemokines. This contrasts with murine macrophage infection where low quantities of the virus were detected as well as lower production of pro-inflammatory cytokines. This may contribute to the lack of polyarthritis in murine animal models. Current literature suggests that CHIKV’s viral proteins bind and interact with human host cell machinery promoting viral replication more efficiently in humans than in mice. CHIKV-related neuropathology is not the most common outcome of the disease. However, recent outbreaks suggest that this pathology is becoming more prevalent, affecting as many as 30% of confirmed patients. The role of adaptive and innate immunity in CHIKV disease amelioration has been extensively, yet separately, explored. A RAG2-/-γc-/- Balb/c mouse model was used to study the role of these immune pathways and their associated immune cells in CHIKV infection. The mice in this study developed local arthritis at the site of inoculation as well as showed signs of viral invasion in the brain. This study added to the hypothesis that both innate and adaptive immune responses are necessary to ameliorate the disease and that the lack of adequately matured lymphocytes and STAT6-activation deficient macrophages may result in more severe pathologies.

Motivation: Alzheimer’s Disease (AD) is a highly prevalent, neurodegenerative disease which causes gradual cognitive decline. As documented in the literature, evi- dence has recently mounted for the role of metabolic dysfunction in AD. Metabolomic data has therefore been increasingly used in AD studies. Metabolomic disease studies often suffer from small sample sizes and inflated false discovery rates. It is therefore of great importance to identify algorithms best suited for the inference of metabolic networks from small cohort disease studies. For future benchmarking, and for the development of new metabolic network inference methods, it is similarly important to identify appropriate performance measures for small sample sizes. Results: The performances of 13 different network inference algorithms, includ- ing correlation-based, regression-based, information theoretic, and hybrid methods, were assessed through benchmarking and structural network analyses. Benchmark- ing was performed on simulated data with known structures across six sample sizes using three different summative performance measures: area under the Receiver Op- erating Characteristic Curve, area under the Precision Recall Curve, and Matthews Correlation Coefficient. Structural analyses (commonly applied in disease studies), including betweenness, closeness, and eigenvector centrality were applied to simu- lated data. Differential network analysis was additionally applied to experimental AD data. Based on the performance measure benchmarking and network analysis results, I identified Probabilistic Context Likelihood Relatedness of Correlation with Biweight Midcorrelation (PCLRCb) (a novel variation of the PCLRC algorithm) to be best suited for the prediction of metabolic networks from small-cohort disease studies. Additionally, I identified Matthews Correlation Coefficient as the best mea- sure with which to evaluate the performance of metabolic network inference methods across small sample sizes. / Graduate

Chlamydia trachomatis/HSV-2 vaginal co-infections are seen clinically, suggesting that these sexually transmitted pathogens may interact. We previously established an intravaginal Chlamydia muridarum/HSV-2 super-infection model and observed that chlamydial pre-infection protects mice from a subsequent lethal HSV-2 challenge. However, the mechanism of protection remains unknown. The type I interferon, IFN-β, binds to the type I interferon receptor (IFNR), elicits a host cellular antiviral response and inhibits HSV replication in vitro and in vivo. Previous studies have demonstrated that C. muridarum infection stimulates genital tract (GT) IFN-β production; therefore, we hypothesized that chlamydial pre-infection protects mice from HSV-2 challenge via the IFN-β/IFNR-induced antiviral response. To test this prediction, we quantified IFN-β levels in vaginal swab samples. Detection of IFN-β in C. muridarum singly infected, but not in mock-infected animals, prompted the use of the super-infection model in IFNR knockout (IFNR-/-) mice. We observed that C. muridarum pre-infection reduces HSV-2-induced mortality by 40% in wild-type mice and by 60% IFNR-/-mice. Severity of HSV-2 disease symptoms and viral shedding was also similarly reduced by C. muridarum pre-infection. These data indicate that, while chlamydial infection induces GT production of IFN-β, type I IFN-induced antiviral responses are likely not required for the observed protective effect.

Indiana University-Purdue University Indianapolis (IUPUI) / Neurofibromatosis Type 1 (NF1) is a highly penetrant autosomal dominant genetic disorder where mutations in the tumor suppressor gene NF1 leads to decreased neurofibromin. The most debilitating manifestation is the presence of complex multilineage Schwann cell-derived plexiform neurofibromas (PN). Historically, little clinical success has been achieved targeting PN through surgery or chemotherapies. I performed an shRNA library screen of patient-derived Schwann cell lines to identify novel therapeutic targets to disrupt PN formation and progression. An shRNA library screen of human kinases and Rho-GTPases was performed in NF1-/- and paired NF1 competent immortalized Schwann cell lines. Following sequencing, candidates were identified. We previously developed a novel mouse model of NF1 wherein a neural crest specific Postncre targeted loxp-flanked Nf1 that replicated the PN found in patients. Additional cohorts of mice were generated with biallelic deletion of Rac1 (Nf1f/fRac1f/f Postn-Cre+; DKO ). Mice were aged for 9 months and peripheral nerves were harvested and fixed in formalin. Peripheral nerve size was measured and tumors were identified through blinded analysis of hematoxylin and eosin and Masson’s Trichrome (collagen) stained slides. Rho family members, including RAC1, were identified as candidates through an shRNA library screen. Genetic disruption of Rac1 in the Schwann cell lineage resulted in the prevention of tumor formation in DKO mice, as observed by peripheral nerve size and histological analysis. I observed an average of 14.8 +/- 2.65 tumors per mouse in the Nf1f/f Postnviii Cre+ cohort compared to 0 tumors in the DKO (p<0.0001). Following an shRNA library screen, RAC1 was identified as a candidate to modulate PN formation. Biallelic deletion of Rac1 in vivo prevented PN formation. I demonstrate that a candidate identified in an shRNA library screen can translate to an biological effect in a mouse model of PN.

Indiana University-Purdue University Indianapolis (IUPUI) / There is an urgent demand for effective medical countermeasures (MCM) in the event of high-dose radiation exposure ranging from nuclear plant disasters to potential nuclear warfare. Victims of lethal-dose radiation exposure face multi-organ injuries including the hematopoietic acute radiation syndrome (H-ARS) and the delayed effects of acute radiation exposure (DEARE) years after irradiation. Defective lymphocyte reconstitution and its subsequent immune insufficiency are some of the most serious consequences of H-ARS and DEARE. In order to investigate potential MCMs to protect or mitigate these radiation injuries, the prolonged tissue-specific immunosuppression at all levels of lymphocyte development in established murine H-ARS and DEARE models was defined, along with unique sex-related and age-related changes present in some tissues but not others. The “double hits” of irradiation and age-related stress on lymphopoiesis led to significant myeloid skew and long-term immune involution. Different kinds and different combinations of hematopoietic growth factors, some in combination with angiotensin converting enzyme inhibitor, were administered to lethally irradiated mice. These radiomitigators were found to significantly increase survival and enhance hematopoiesis in H-ARS, but they did little to alleviate the severity of DEARE including immune insufficiency. 16,16 dimethyl-prostaglandin E2 (dmPGE2), a long-acting formulation of PGE2 with similar biological effects as PGE2, was found to enhance survival and hematopoiesis in lethal-irradiated mice when used as radiomitigator or radioprotectant. The optimum time window for administration of radioprotectant and radiomitigator dmPGE2 was defined, which is -3hr to -15min prior to irradiation and +6hr to +30hr post irradiation. Significant survival efficacy of radioprotectant dmPGE2 was also demonstrated in pediatric and geriatric mice. Using specific PGE2 receptor (EP) agonists, the EP4 receptor was defined as the PGE2 receptor potentially responsible for dmPGE2 radioprotection. Radioprotectant dmPGE2 was also found to prevent radiation-induced thymic involution and to ameliorate the long-term immune suppression in radiation survivors in the DEARE phase via promoting hematopoietic stem cell differentiation towards to the lymphoid lineage. This is the first report of an effective MCM for H-ARS which also targets long-term thymic involution and lymphoid lineage reconstitution.