Objetivo: Avaliar a influência da infecção por Pseudomonas aeruginosa formadora de biofilme na penetração pulmonar da tobramicina através da modelagem populacional dos dados de plasma e microdialisado em animais sadios e infectados. Metodologia: A pneumonia foi desenvolvida através de inoculação de P. aeruginosa (cepa PA14) pela via intratraqueal (109 UFC/mL) a ratos Wistar. Sete dias após a inoculação os animais infectados (n = 5) receberam tobramicina 10 mg/kg i.v. bolus. Animais saudáveis (n = 6) foram utilizados como controle. As concentrações livres pulmonares foram coletadas por microdiálise (sonda CMA/20). As sondas de microdiálise foram calibradas in vitro através de diálise e retrodiálise e in vivo utilizando retrodiálise. A ligação da tobramicina às proteínas plasmáticas foi determinada por microdiálise. As concentrações do fármaco nas amostras foram determinadas por cromatografia líquida em tandem com espectrometria de massas (CLAE-EM/EM) utilizando metodologia validada. Os parâmetros farmacocinéticos foram determinados por abordagem não-compartimental (Phoenix®) e modelagem populacional (popPK) (Monolix®). Resultados e Discussão: A recuperação relativa (RR) das sondas foi independente da concentração de tobramicina e inversamente proporcional ao fluxo de perfusão. A RR determinada in vivo foi de 27,64 % ± 7,70 para animais sadios e 24,47 % ± 1,66 para animais infectados. A ligação às proteínas plasmáticas foi de 11,3 ± 1,9%. A infecção com formação de biofilme não alterou a farmacocinética plasmática da tobramicina, entretanto reduziu em cerca de 70% a penetração pulmonar do fármaco. As concentrações plasmáticas e teciduais foram simultaneamente descritas por um modelo farmacocinético populacional de dois compartimentos, tanto em animais sadios como infectados. A infecção, utilizada como covariável categórica, permitiu descrever as alterações no volume do compartimento periférico e na constante de eliminação do compartimento central devido à infecção. Conclusões: As concentrações plasmáticas da tobramicina, utilizadas para ajuste posológico, superestimam as concentrações ativas no pulmão infectado. O modelo popPK descrito permite a previsão das concentrações livres pulmonares da tobramicina em pulmão infectado, podendo auxiliar na otimização da terapia de pneumonias com P. aeruginosa formadora de biofilme. / Objective: To evaluate the influence of biofilm-forming Pseudomonas aeruginosa infection on tobramycin lung penetration by population pharmacokinetic modeling of plasma and microdialysate data in healthy and infected rats. Methodology: The infection was developed by intratracheal inoculation (109 CFU/mL) of P. aeruginosa (PA14 strain) to Wistar rats. In order to determine plasma and tissue concentrations, seven days after the inoculation the infected animals (n = 5) received tobramycin 10 mg/kg i.v. bolus dose via femoral vein. A healthy group (n = 6) was used as control. Free lung concentrations were determined in microdialysate samples obtained using CMA/20 probes. Microdialysis probes were calibrated in vitro by dialysis and retrodialysis and in vivo by retrodialysis. Tobramycin plasma protein binding was determined by microdialysis. Plasma and tissue concentrations were quantified by a developed and validated liquid chromatography in tandem with mass spectrometry (LC-MS/MS) method. Compartmental and non-compartmental analyses were carried out by Monolix™ and Phoenix™ software, respectively. Results and Discussion: Microdialysis probes relative recovery was independent of the tobramycin concentration and is inversely proportional to the perfusion flow rate investigated. The in vivo probe recovery was 27.64 % ± 7.70 (healthy rats) and 24.47 % ± 1.66 (infected rats). The plasma protein binding was 11.3 ± 1.9%. The biofilm-forming lung infection did not alter tobramycin plasma pharmacokinetics, however, reduced lung penetration in about 70%. The plasma and tissue concentrations-time profiles were simultaneously described by a two compartment popPK model in healthy and infected animals. The infection process, used as categorical covariate allowed describing the changes observed in the volume of the peripheral compartment and in constant rate of elimination from the central compartment. Conclusions: Tobramycin plasma concentrations, used for dosing adjustments, overestimate active concentrations in infected lung. The described popPK model allows predicting free tobramycin lung concentrations in infected lung and could be useful to optimize the treatment of pneumonia caused by biofilm-forming P. aeruginosa with this drug.

Pathogenic bacteria employ a number of mechanisms to induce infection and survive in host tissues, including toxin secretion and the formation of protective multicellular structures called biofilms. Type IV Pili (T4P) are highly conserved organelles essential for both the establishment of infection and biofilm maturation. The goal of this research is to gain a molecular level understanding of the function of the highly dynamic T4P of Pseudomonas aeruginosa. The pilMNOPQ operon encodes 5 members of a transmembrane complex that facilitates pilus function. While PilQ is the putative outer membrane secretin through which the pilus exits the cell, the roles of the PilM/N/O/P proteins are less well defined. Using both in vivo and in vitro techniques our characterization of PilP has provided significant insight into organization of the apparatus. PilP is an inner membrane lipoprotein essential for T4P function, but lipidation is dispensable, suggesting that its interactions with other T4P components are sufficient for PilP function. We showed that PilN/O/P form a stable heterotrimer when expressed in E. coli, and we suggest that they form a similar subcomplex in P. aeruginosa. Additionally we were able to show that PilP is also able to interact with a periplasmic fragment of the outer membrane pore protein PilQ. Structural and bioinformatics studies suggest that the organization of PilN/O/P/Q complex is similar to that of the transenvelope complex of another important Gram-negative virulence factor – the Type II Secretion System (T2SS). Our structural and functional characterization of PilP, the PilN/O/P complex and the striking similarities between the T4P and T2S systems, as well as important differences that make each molecular machine unique, will be presented.

Pathogenic bacteria employ a number of mechanisms to induce infection and survive in host tissues, including toxin secretion and the formation of protective multicellular structures called biofilms. Type IV Pili (T4P) are highly conserved organelles essential for both the establishment of infection and biofilm maturation. The goal of this research is to gain a molecular level understanding of the function of the highly dynamic T4P of Pseudomonas aeruginosa. The pilMNOPQ operon encodes 5 members of a transmembrane complex that facilitates pilus function. While PilQ is the putative outer membrane secretin through which the pilus exits the cell, the roles of the PilM/N/O/P proteins are less well defined. Using both in vivo and in vitro techniques our characterization of PilP has provided significant insight into organization of the apparatus. PilP is an inner membrane lipoprotein essential for T4P function, but lipidation is dispensable, suggesting that its interactions with other T4P components are sufficient for PilP function. We showed that PilN/O/P form a stable heterotrimer when expressed in E. coli, and we suggest that they form a similar subcomplex in P. aeruginosa. Additionally we were able to show that PilP is also able to interact with a periplasmic fragment of the outer membrane pore protein PilQ. Structural and bioinformatics studies suggest that the organization of PilN/O/P/Q complex is similar to that of the transenvelope complex of another important Gram-negative virulence factor – the Type II Secretion System (T2SS). Our structural and functional characterization of PilP, the PilN/O/P complex and the striking similarities between the T4P and T2S systems, as well as important differences that make each molecular machine unique, will be presented.

Objetivo: Avaliar a influência da infecção por Pseudomonas aeruginosa formadora de biofilme na penetração pulmonar da tobramicina através da modelagem populacional dos dados de plasma e microdialisado em animais sadios e infectados. Metodologia: A pneumonia foi desenvolvida através de inoculação de P. aeruginosa (cepa PA14) pela via intratraqueal (109 UFC/mL) a ratos Wistar. Sete dias após a inoculação os animais infectados (n = 5) receberam tobramicina 10 mg/kg i.v. bolus. Animais saudáveis (n = 6) foram utilizados como controle. As concentrações livres pulmonares foram coletadas por microdiálise (sonda CMA/20). As sondas de microdiálise foram calibradas in vitro através de diálise e retrodiálise e in vivo utilizando retrodiálise. A ligação da tobramicina às proteínas plasmáticas foi determinada por microdiálise. As concentrações do fármaco nas amostras foram determinadas por cromatografia líquida em tandem com espectrometria de massas (CLAE-EM/EM) utilizando metodologia validada. Os parâmetros farmacocinéticos foram determinados por abordagem não-compartimental (Phoenix®) e modelagem populacional (popPK) (Monolix®). Resultados e Discussão: A recuperação relativa (RR) das sondas foi independente da concentração de tobramicina e inversamente proporcional ao fluxo de perfusão. A RR determinada in vivo foi de 27,64 % ± 7,70 para animais sadios e 24,47 % ± 1,66 para animais infectados. A ligação às proteínas plasmáticas foi de 11,3 ± 1,9%. A infecção com formação de biofilme não alterou a farmacocinética plasmática da tobramicina, entretanto reduziu em cerca de 70% a penetração pulmonar do fármaco. As concentrações plasmáticas e teciduais foram simultaneamente descritas por um modelo farmacocinético populacional de dois compartimentos, tanto em animais sadios como infectados. A infecção, utilizada como covariável categórica, permitiu descrever as alterações no volume do compartimento periférico e na constante de eliminação do compartimento central devido à infecção. Conclusões: As concentrações plasmáticas da tobramicina, utilizadas para ajuste posológico, superestimam as concentrações ativas no pulmão infectado. O modelo popPK descrito permite a previsão das concentrações livres pulmonares da tobramicina em pulmão infectado, podendo auxiliar na otimização da terapia de pneumonias com P. aeruginosa formadora de biofilme. / Objective: To evaluate the influence of biofilm-forming Pseudomonas aeruginosa infection on tobramycin lung penetration by population pharmacokinetic modeling of plasma and microdialysate data in healthy and infected rats. Methodology: The infection was developed by intratracheal inoculation (109 CFU/mL) of P. aeruginosa (PA14 strain) to Wistar rats. In order to determine plasma and tissue concentrations, seven days after the inoculation the infected animals (n = 5) received tobramycin 10 mg/kg i.v. bolus dose via femoral vein. A healthy group (n = 6) was used as control. Free lung concentrations were determined in microdialysate samples obtained using CMA/20 probes. Microdialysis probes were calibrated in vitro by dialysis and retrodialysis and in vivo by retrodialysis. Tobramycin plasma protein binding was determined by microdialysis. Plasma and tissue concentrations were quantified by a developed and validated liquid chromatography in tandem with mass spectrometry (LC-MS/MS) method. Compartmental and non-compartmental analyses were carried out by Monolix™ and Phoenix™ software, respectively. Results and Discussion: Microdialysis probes relative recovery was independent of the tobramycin concentration and is inversely proportional to the perfusion flow rate investigated. The in vivo probe recovery was 27.64 % ± 7.70 (healthy rats) and 24.47 % ± 1.66 (infected rats). The plasma protein binding was 11.3 ± 1.9%. The biofilm-forming lung infection did not alter tobramycin plasma pharmacokinetics, however, reduced lung penetration in about 70%. The plasma and tissue concentrations-time profiles were simultaneously described by a two compartment popPK model in healthy and infected animals. The infection process, used as categorical covariate allowed describing the changes observed in the volume of the peripheral compartment and in constant rate of elimination from the central compartment. Conclusions: Tobramycin plasma concentrations, used for dosing adjustments, overestimate active concentrations in infected lung. The described popPK model allows predicting free tobramycin lung concentrations in infected lung and could be useful to optimize the treatment of pneumonia caused by biofilm-forming P. aeruginosa with this drug.

Objetivo: Avaliar a influência da infecção por Pseudomonas aeruginosa formadora de biofilme na penetração pulmonar da tobramicina através da modelagem populacional dos dados de plasma e microdialisado em animais sadios e infectados. Metodologia: A pneumonia foi desenvolvida através de inoculação de P. aeruginosa (cepa PA14) pela via intratraqueal (109 UFC/mL) a ratos Wistar. Sete dias após a inoculação os animais infectados (n = 5) receberam tobramicina 10 mg/kg i.v. bolus. Animais saudáveis (n = 6) foram utilizados como controle. As concentrações livres pulmonares foram coletadas por microdiálise (sonda CMA/20). As sondas de microdiálise foram calibradas in vitro através de diálise e retrodiálise e in vivo utilizando retrodiálise. A ligação da tobramicina às proteínas plasmáticas foi determinada por microdiálise. As concentrações do fármaco nas amostras foram determinadas por cromatografia líquida em tandem com espectrometria de massas (CLAE-EM/EM) utilizando metodologia validada. Os parâmetros farmacocinéticos foram determinados por abordagem não-compartimental (Phoenix®) e modelagem populacional (popPK) (Monolix®). Resultados e Discussão: A recuperação relativa (RR) das sondas foi independente da concentração de tobramicina e inversamente proporcional ao fluxo de perfusão. A RR determinada in vivo foi de 27,64 % ± 7,70 para animais sadios e 24,47 % ± 1,66 para animais infectados. A ligação às proteínas plasmáticas foi de 11,3 ± 1,9%. A infecção com formação de biofilme não alterou a farmacocinética plasmática da tobramicina, entretanto reduziu em cerca de 70% a penetração pulmonar do fármaco. As concentrações plasmáticas e teciduais foram simultaneamente descritas por um modelo farmacocinético populacional de dois compartimentos, tanto em animais sadios como infectados. A infecção, utilizada como covariável categórica, permitiu descrever as alterações no volume do compartimento periférico e na constante de eliminação do compartimento central devido à infecção. Conclusões: As concentrações plasmáticas da tobramicina, utilizadas para ajuste posológico, superestimam as concentrações ativas no pulmão infectado. O modelo popPK descrito permite a previsão das concentrações livres pulmonares da tobramicina em pulmão infectado, podendo auxiliar na otimização da terapia de pneumonias com P. aeruginosa formadora de biofilme. / Objective: To evaluate the influence of biofilm-forming Pseudomonas aeruginosa infection on tobramycin lung penetration by population pharmacokinetic modeling of plasma and microdialysate data in healthy and infected rats. Methodology: The infection was developed by intratracheal inoculation (109 CFU/mL) of P. aeruginosa (PA14 strain) to Wistar rats. In order to determine plasma and tissue concentrations, seven days after the inoculation the infected animals (n = 5) received tobramycin 10 mg/kg i.v. bolus dose via femoral vein. A healthy group (n = 6) was used as control. Free lung concentrations were determined in microdialysate samples obtained using CMA/20 probes. Microdialysis probes were calibrated in vitro by dialysis and retrodialysis and in vivo by retrodialysis. Tobramycin plasma protein binding was determined by microdialysis. Plasma and tissue concentrations were quantified by a developed and validated liquid chromatography in tandem with mass spectrometry (LC-MS/MS) method. Compartmental and non-compartmental analyses were carried out by Monolix™ and Phoenix™ software, respectively. Results and Discussion: Microdialysis probes relative recovery was independent of the tobramycin concentration and is inversely proportional to the perfusion flow rate investigated. The in vivo probe recovery was 27.64 % ± 7.70 (healthy rats) and 24.47 % ± 1.66 (infected rats). The plasma protein binding was 11.3 ± 1.9%. The biofilm-forming lung infection did not alter tobramycin plasma pharmacokinetics, however, reduced lung penetration in about 70%. The plasma and tissue concentrations-time profiles were simultaneously described by a two compartment popPK model in healthy and infected animals. The infection process, used as categorical covariate allowed describing the changes observed in the volume of the peripheral compartment and in constant rate of elimination from the central compartment. Conclusions: Tobramycin plasma concentrations, used for dosing adjustments, overestimate active concentrations in infected lung. The described popPK model allows predicting free tobramycin lung concentrations in infected lung and could be useful to optimize the treatment of pneumonia caused by biofilm-forming P. aeruginosa with this drug.

La phospholipase A2 sécrétée de type IIA (sPLA2-IIA) est une protéine possédant une activité antibactérienne très élevée envers les bactéries à Gram positif. La mucoviscidose (MV) est une maladie génétique résultant de la mutation du gène Cftr. L'altération de la fonction du CFTR dans les poumons favorise la colonisation par des pathogènes bactériens, dont les plus retrouvés, S. aureus et P. aeruginosa, colonisent les voies aériennes de manière séquentielle: S. aureus est principalement retrouvé chez les jeunes patients et P. aeruginosa chez les adultes. Les mécanismes impliqués dans ce changement de prévalence restent cependant mal connus. Dans ce travail, nous démontrons que l'expression de la sPLA2-IIA est plus importante dans les poumons de patients MV que dans les poumons de patients non MV et augmente avec l'âge des patients. La sPLA2-IIA présente dans les expectorations de patients MV est capable de tuer spécifiquement S. aureus. L'utilisation de modèles animaux nous a permis de démontrer sa spécificité d'action contre S. aureus et l'induction de son expression par P. aeruginosa contribuant à l'élimination de S. aureus des voies respiratoires. Enfin, nous avons identifié les cellules épithéliales comme une source majeure de sPLA2-IIA chez les patients MV. Dans ces cellules, P. aeruginosa induit l'expression de la sPLA2-IIA par un mécanisme dépendant de l'injection de la toxine ExoS et du facteur de transcription KLF2. L'ensemble de ces résultats indique i) que la sPLA2-IIA induite par P. aeruginosa, participe à l'élimination de S. aureus dans les voies aériennes des patients MV et ii) qu'une bactérie élimine une autre bactérie en utilisant la défense de l'hôte. / The type IIA secretory phospholipase A2 (sPLA2-IIA) is a host defense protein endowed with antibacterial activity, especially against Gram positive bacteria. Cystic fibrosis (CF) is a genetic disease due to mutations of Cftr gene. In the lungs, CFTR mutation favored bacterial colonization by bacterial pathogens, of which S. aureus and P. aeruginosa are the most isolated. These two bacterial species sequentially colonized airways of CF patients: S. aureus is predominant in young patients and P. aeruginosa in adults. But the mechanisms involved in this switch of bacterial prevalence are still unknown. In this work we showed that sPLA2-IIA levels were increased in lungs of CF patients compared to lungs of non-CF patients and that sPLA2-IIA levels increased with age of patients. sPLA2-IIA recovered from CF patients expectorations was active and killed specifically S. aureus. Using animal models of lung infection, we demonstrated the selectivity of sPLA2-IIA against S. aureus and that P. aeruginosa induced sPLA2-IIA expression, the latter contributed to S . aureus elimination from the airways. Finally, we identified epithelial cells as a major source of sPLA2-IIA in CF airways. In these cells, P. aeruginosa induced sPLA2-IIA expression through injection of ExoS toxin and activation of KLF2 transcription factor. Taken together, these results indicate that i) P. aeruginosa-induced sPLA2-IIA expression in CF airways participated to S. aureus elimination and ii) a bacteria eliminate another bacteria by manipulating host innate immunity.

Pseudomonas aeruginosa is an important pathogen of the immunocompromised, causing both acute and chronic infections. In cystic fibrosis (CF) patients, P. aeruginosa causes chronic disease. The impressive sensory network of P. aeruginosa allows the bacterium to sense and respond to a variety of stimuli found in diverse environments. Transcriptional regulators, including alternative sigma factors and response regulators, integrate signals changing gene expression, allowing P. aeruginosa to cause infection. The two-component transcriptional regulator AlgR is important in P. aeruginosa pathogenesis in both acute and chronic infections. In chronic infections, AlgR and the alternative sigma factor AlgU activate the genes responsible for alginate production. Previous work demonstrated that AlgU controls rsmA expression. RsmA is a posttranscriptional regulator that is antagonized by two small RNAs, RsmY and RsmZ. In this work, we demonstrate that AlgR directly activates rsmA expression from the same promoter as AlgU. In addition, phosphorylation was not necessary for AlgR activation of rsmA using algR and algZ mutant strains. AlgU and AlgR appear to affect the antagonizing small RNAs rsmY and rsmZ indirectly. RsmA was active in a mucA22 mutant strain using leader fusions of two RsmA targets, tssA1 and hcnA. AlgU and AlgR were necessary for posttranscriptional regulation of tssA1 and hcnA. Altogether, our work demonstrates that the alginate regulators AlgU and AlgR are important in the control of the RsmA posttranscriptional regulatory system. These findings suggest that RsmA plays an unknown role in mucoid strains due to AlgU and AlgR activities.

Dans les dernières décennies, la plupart des pays du monde ont connu une pénurie d'eau et l’augmentation du taux de consommation. Aujourd'hui, tous les pays dans le monde essayent de trouver des alternatives pour remédier à cette pénurie. Une solution consiste en la réutilisation des eaux grises (GW) pour l'irrigation après traitement. Les GW correspondent aux eaux usées générée dans une maison à l'exception de l'eau des toilettes. Les risques associés à la réutilisation de ces eaux est la présence de microorganismes pathogènes qui peuvent infecter les humains, les animaux et les plantes. Dans cette thèse centrée sur l'étude de la survie des représentants d'agents pathogènes, comme E. coli, P. aeruginosa, et le bactériophage MS2 qui sont trouvés dans les eaux grises. Il a été étudié l’effet de quelques facteurs physico-chimiques tels que; température (6 ± 2,23 ± 2 et 42 ± 2 ° C), la salinité (1,75 and 3.5% de NaCl), de l'oxygène (aérobie et anaérobie), des éléments nutritifs (milieu riche et de milieux pauvres), la lumière avec la photocatalyse (lampes UV et visible) et filtre à sable lent (sable du désert égyptien et le sable piscine). Une combinaison de la température, la lumière du soleil et de haute photocatlysis sont principalement responsables de la baisse rapide des bactéries et du coliphage MS2. Le filtre à sable lent a une influence nettement moindre sur la survie des bactéries dans les eaux grises, mais il est efficace pour diminuer la turbidité et de la DCO. / In recent decades, most countries of the world have experienced a shortage of water and increase its rate of consumption. Today, every country in the world are interested in this problem by trying to find alternatives to address this shortage. One solution is reuse greywater (GW) for irrigation after treatment. GW is all water generated from Household except toilet water. The risks associated with the reuse of these waters are the presence of pathogens that can infect humans, animals and plants. In this thesis focused on studying treatment by slow sand filtration and the survival of representatives of pathogens, such as E. Coli, P. aeruginosa , E. Faecalis and Bacteriophage MS2 which could be found in the greywater. The study factors was a physico-chemicals factors such as; temperature (6±2,23±2,42±2°c), salinity (1.75 and 3.5% Nacl), oxygen (aerobic and anaerobic condition), nutrient ( rich media , 50%: 50% salt and poor media ), light with photocatalysis ( UV and Visible lights) and slow sand filter (Egyptian desert sand and swimming pool sand). A combination of high temperature, sunlight and photocatlysis are mainly responsible for the rapid decline of bacteria and MS2 coliphage. Slow sand filter have clearly less influence on the survival of bacteria in the greywater, but it effective to decline turbidity and COD for short times.

Les rejets urbains par temps de pluie dégradent l’état écologique des écosystèmes aquatiques et peuvent induire une exposition des populations humaines aux contaminants chimiques et microbiens (bactéries, virus, parasites). L’objectif de ce travail de thèse était d’évaluer les effectifs et de prédire le devenir de bactéries pathogènes introduites dans les milieux aquatiques par une source majeure comme les eaux usées rejetées par des dispositifs tels que les déversoirs d’orage (DO) et les lagunes d’épuration (WWTL). La répartition d’un agent pathogène fortement liée aux milieux hydriques, Pseudomonas aeruginosa, a été comparée avec celles observées pour des indicateurs de contaminations fécales (E. coli et les entérocoques intestinaux), mais également avec celle de l’espèce pathogène Aeromonas caviae. La dangerosité des formes retrouvées dans ces milieux a été évaluée par approches moléculaires (PFGE et MLST). Les résultats obtenus montrent un fort apport en P. aeruginosa via les eaux usées, avec un effet significatif sur les effectifs observés en fonction de l’intensité des pluies et des périodes de temps sec, et les fluctuations du régime hydrologique et des paramètres physico-chimiques. Une grande diversité infra-spécifique des P. aeruginosa, et la capacité de certains génotypes à s’installer durablement dans ces milieux (macrophytes et périphyton) ont été observées. Certaines souches ont par ailleurs montré une parenté avec des lignées d’infections communautaires, ou encore des clones épidémiques majeurs (PA14 et C). Des études en microcosme ont été effectuées pour valider les interactions observées avec certains macrophytes, et identifier des propriétés d’adhérence bactérienne (dont les lectines) impliquées dans ces interactions. Ces travaux ont impliqué une analyse de la distribution des gènes lecA et lecB, codant des lectines chez P. aeruginosa, et une étude de leurs ligands. Le gène lecA a été localisé dans une zone de forte plasticité génomique. Ces travaux ont permis la description d’une nouvelle structure de l’adhésine LecB / Urban wet-weather discharges degrade the ecological status of aquatic ecosystems and may expose human populations to chemical and microbial contaminants (bacteria, viruses, parasites). The aim of this thesis was to evaluate the numbers and predict the fate of pathogenic bacteria introduced into aquatic ecosystems by a major source like wastewater from devices such as combined sewer overflows (CSO) and wastewater treatment lagoons (WWTL). The distribution of a human pathogen closely linked to hydric environments, Pseudomonas aeruginosa, was compared with those observed for fecal indicators (E. coli and intestinal enterococci), but also with that of Aeromonas caviae pathogenic species. Dangerousness of strains found in these environments was evaluated by molecular approaches (PFGE and MLST). Obtained results showed a high contribution of wastewater in P. aeruginosa release, with a significant effect of rainfall intensity and preceding dry periods, in addition to changes in hydrological regime and physico-chemical parameters on recorded data. A large infra-specific diversity was observed within P. aeruginosa and the ability of some genotypes to colonize permanently aquatic surfaces (macrophytes and periphyton) were observed. Some strains showed a kinship with lineages of community infections or major epidemic clones (PA14 and C). Microcosm studies were performed to validate observed interactions with macrophytes, and to identify bacterialadhesion properties (including lectins) involved in these interactions. These investigations involved analysis of the distribution of lectin encoding loci lecA and lecB within P. aeruginosa, and a study of their ligands. lecA was located in a highly unstable genomic region. This work allowed the description of a new structure of the adhesin LecB

La mucoviscidose est une pathologie pulmonaire causée par un dysfonctionnement du canal CFTR et caractérisée par un mucus visqueux, une susceptibilité accrue aux infections chroniques et une inflammation excessive. Une première partie de ma thèse a eu pour objectif d’étudier les mécanismes inflammatoires impliqués dans le développement de la pathologie. Nous avons plus particulièrement analysé le rôle de l’IL-1β et de l’IL-17 dans la réponse à l’infection par Pseudomonas aeruginosa, dans le modèle murin ΔF508 de mucoviscidose. La seconde partie de ma thèse a porté sur l’étude de la malaria pulmonaire et cérébrale, une complication létale de l’infection à P. falciparum. Nous avons mis en évidence l’importance de trois voies d’activation des lymphocytes T CD8+ cytotoxiques dans le développement de la neuropathologie induite par Plasmodium berghei ANKA chez la souris : la protéine PKC-θ, la sous-unité β2 du récepteur à l’IL-12 et le récepteur des IFN de type I, mais qui ne semblent pas impliquées dans l’inflammation pulmonaire associée. / Cystic fibrosis is a pulmonary pathology, caused by the CFTR channel dysfunction, and characterized by high mucus viscosity, increased sensitivity to chronic infections and excessive inflammation. The aim of my thesis was first to study the inflammatory mechanisms involved in this lung pathology. Indeed, we analyzed the role of IL-1β and IL-17 in response to Pseudomonas aeruginosa infection, in the ΔF508 mouse model of cystic fibrosis. In the second part of my thesis, I studied pulmonary and cerebral malaria, a lethal complication of P. falciparum infection. We showed the importance of three pathways implicated in cytotoxic CD8+ T lymphocytes activation during the Plasmodium berghei ANKA-induced neuropathology development in mice: PKC-θ protein, β2 subunit of IL-12 receptor and type I IFN receptor, which did not seem essential for the associated lung inflammation.