Le syndrome de l’intestin irritable (SII) est un trouble fonctionnel d’origine multifactorielle, impliquant des facteurs environnementaux tels que le stress, l’alimentation et met en jeu un dysfonctionnement de l’axe intestin-cerveau, une micro-inflammation, une dysbiose et une hyperperméabilité intestinale. Le rôle du protéasome dans la régulation de la barrière intestinale au cours du SII a été étudié. De plus, ces troubles fonctionnels intestinaux (TFI) ont également été décrits comme exacerbés chez des patients souffrant d’obésité, dont la physiopathologie est complexe. Néanmoins, les mécanismes impliqués dans cette association restent mal compris et ont donc été recherchés. Dans ce travail, des modèles murins « SII-like » comme le modèle de stress « water avoidance stress » ou WAS et le modèle post-inflammatoire « post-TNBS » ont été utilisés afin d’étudier l’impact d’une inhibition du protéasome sur la régulation de la perméabilité intestinale. L’inhibition pharmacologique du protéasome par le PR-957 ou l’utilisation de souris invalidées pour une sous unité β2i du protéasome limite l’hyperperméabilité intestinale. Une supplémentation orale en glutamine permet également de diminuer la perméabilité intestinale. Une étude protéomique au niveau colique des souris WAS et une étude de l’ubiquitome colique de patients souffrant de SII à profil diarrhéique confirment l’implication du protéasome dans la physiopathologie du SII. Nous avons ensuite cherché à comprendre le lien entre l’obésité et le SII en combinant des modèles d’obésité (génétique et induite par une alimentation riche en graisses ou HFD) et le modèle WAS. Seules les souris HFD présentent une exacerbation de l’hyperperméabilité intestinale et une corticostéronémie plasmatique élevée en réponse au modèle WAS. Des études complémentaires suggèrent que ces résultats sont indépendants de la leptine, de la glycémie et du microbiote intestinal. Nos travaux proposent donc de nouvelles pistes de prise en charge des patients souffrant de SII, par intervention nutritionnelle via la glutamine ou en utilisant le protéasome comme cible thérapeutique. Nous suggérons également un rôle de l’alimentation (riche en graisse) dans le développement des TFI au cours de l’obésité. / Irritable bowel syndrome (IBS) is a multifactorial functional disorder, involving environmental factors (stress and diet for instance), gut-brain-axis dysfunction, micro-inflammation, dysbiosis and an alteration of intestinal permeability. The role of the proteasome in the regulation of the intestinal barrier during IBS has been studied. In addition, these intestinal functional disorders have also been described in patients with obesity. Nevertheless, the mechanisms underlying an association of intestinal functional disorders in the obesity context, remain poorly understood and have therefore been investigated in this thesis. In this study, "IBS-like" mouse models such as water avoidance stress (WAS) and the post-inflammatory (post-TNBS) models, were used to study the impact of proteasome inhibition on the regulation of intestinal permeability. We found that the pharmacological inhibition of the proteasome (with PR-957) or the use of knock-out mice for a subunit of the proteasome (β2i -/-) limit intestinal hyperpermeability occured in IBS-Like models. Moreover, we found that oral supplementation with glutamine also reduces intestinal hyperpermeability, wich, thus, can be considered as a putative nutritional treatment for IBS. A colonic proteomic study of WAS mice and a study of colonic ubiquitoma in IBS patients with diarrheal profiles confirmed the involvement of proteasome in the pathophysiology of IBS. Therefore, the link between obesity and IBS was examined by combining models of obesity (ob/ob genetic and high-fat diet [HFD] models) with WAS model. Only HFD mice displayed enhanced intestinal hyperpermeability and higher plasma corticosterone levels in response to WAS. Further studies suggest that these results, themselve, are independent of leptin, glycaemia and gut microbiota. This study paves new ways of treating patients suffering from IBS, by nutritional intervention via glutamine or by using the proteasome as a therapeutic target. We also suggest a role of diet (high fat) in the development of intestinal functional disorders during obesity.

<p>(From abstract) </p> <p>In the first study, the urine collection method was effectively applied for evaluation of intestinal permeability using Cr-EDTA, an indigestible oral marker, demonstrating the applicability of the procedure in 1-week-old and 6-week-old neonatal heifer calves (n=15 calves). Calf health observations were recorded during the entire urinary catheterization process and collection period to evaluate any negative health reactions to the procedure, or localized reactions. Proportion of localized reactions were analyzed, and the proportions did not exceed 20% for the calves catheterized at either 1 week or 6 weeks of age. </p> <p>In the second study, the developed catheterization procedure and urine collection method was applied using Cr-EDTA as an oral marker to investigate if L-GLN supplementation would offer improvement to intestinal permeability. In this larger study, 30 Holstein heifer calves [1.5 ± 0.5 days old; 37.1 ± 0.86 kg body weight (<strong>BW</strong>)] were blocked by serum total protein, BW, and age, and randomly assigned to 1 of 2 treatments: <strong>GLN</strong> [24% crude protein (<strong>CP</strong>)], 17% fat milk replacer (<strong>MR</strong>) +10 g L-GLN/kg MR powder) or <strong>NS</strong> (24% CP, 17% fat MR). MR was reconstituted to 12.5% solids with warm water and fed 3.8 L/calf/d until weaning. Calves were weaned at 56.4 ± 0.5 days of age, and had <em>ad libitum</em> grain (17% CP, 2% fat) and water access throughout the experimental period.</p> <p>During the preweaning period, calves were individually housed in hutches and health observations, which included respiratory and fecal scores, were assessed daily. Body weight was measured weekly, and grain and MR intake was assessed daily to calculate average daily gain (<strong>ADG</strong>), average daily feed intake [<strong>ADFI</strong>; grain intake (dry matter (<strong>DM)</strong> basis) + MR intake (DM basis)], and feed efficiency (<strong>G:F</strong>; ADG:ADFI). At weaning, calves were weighed, moved to pens (n = 3 pens/treatment, 4-5 calves/pen), provided free access to grain and grass hay, and then weighed 2 weeks post-weaning. Additionally, urinary catheters were placed at 1 and 6 weeks of age, and calves were orally dosed with 1 L Cr-EDTA in their MR. Urine samples were then collected over a 24-hr period for Cr output analysis as an <em>in vivo</em>biomarker of intestinal permeability. </p> <p>Blood was collected on study days 1, 2, 5, 7, 14, 21, 42, 56, and 70 to measure haptoglobin, serum amyloid A, leukocyte data, neutrophil: lymphocyte (<strong>N:L</strong>), glucose, non-esterified fatty acids, insulin, and cortisol. Two study periods were identified for data analysis representing greater (<strong>P1</strong>; weeks 1-3) and reduced (<strong>P2</strong>; weeks 4-8) enteric disease susceptibility. Data were analyzed using PROC GLIMMIX or PROC MIXED in SAS 9.4 with calf as the experimental unit. There was a decrease in total preweaning Cr output (<em>P</em> < 0.05) for GLN calves, and Cr output in 1 week old calves was decreased (<em>P</em> = 0.04) in GLN versus NS calves. The N:L was decreased overall (<em>P</em> = 0.03) and during P2 (<em>P</em> = 0.01) and P2 neutrophil count tended to be reduced (<em>P</em> = 0.07) in GLN versus NS calves. There were no MR treatment differences for ADFI, ADG, body measurements, post-absorptive metabolic biomarkers, disease scores, and therapeutic treatments (<em>P</em> > 0.10). In summary, L-GLN supplementation improved intestinal integrity and biomarkers of physiological stress in pre-weaned Holstein heifer calves managed under production-relevant conditions.  </p>

Ce mémoire rapporte l’optimisation et l’évaluation d’une nouvelle version du test PAMPA (Parallel Artificial Membrane Permeability Assay) appelée Néo-PAMPA. Ce test qui permet la prédiction de l’absorption intestinale de médicaments consiste en l’utilisation d’une membrane modèle de la paroi intestinale composée d’une bicouche lipidique déposée sur un coussin de polydopamine recouvrant un filtre poreux. En effet, nous nous sommes intéressés lors de ce projet à la mise en place d’une membrane artificielle qui serait plus représentative de la paroi intestinale humaine. Nous avons pu déterminer, suite à une étude comparative des propriétés de huit médicaments ainsi que les coefficients de perméabilité obtenus, que les filtres en polycarbonate présentaient le meilleur choix de support solide pour la membrane. Nous avons également vérifié la déposition du coussin de polydopamine qui apporte le caractère fluide à la bicouche lipidique. Les résultats des tests de perméabilité ont démontré que le coussin de polymère n’obstrue pas les pores du filtre après un dépôt de 4h. Nous avons par la suite étudié la déposition de la bicouche lipidique sur le filtre recouvert de polydopamine. Pour ce faire, deux méthodes de préparation de liposomes ainsi que plusieurs tailles de liposomes ont été testées. Aussi, la composition en phospholipides a été sujette à plusieurs changements. Tous ces travaux d’optimisation ont permis d’aboutir à des liposomes préparés selon la méthode du « film lipidique » à partir d’un mélange de dioléoylphosphatidylcholine (DOPC) et de cholestérol. Une dernière étape d’optimisation de la déposition de la bicouche reste à améliorer. Enfin, le test standard Caco-2, qui consiste à évaluer la perméabilité des médicaments à travers une monocouche de cellules cancéreuses du colon humain, a été implémenté avec succès dans le but de comparer des données de perméabilité avec un test de référence. / This essay reports the optimization and evaluation of a new version of the Parallel Artificial Membrane Permeability Assay (PAMPA), called Neo-PAMPA. This test, used for the prediction of the drugs’ intestinal absorption, involves the use of a model intestine cell membrane in which a lipid bilayer is supported on a polymeric cushion deposited on a filter. In this project, we are interested in the development of an artificial membrane that is more representative of the human intestinal wall. We determined, based on a comparative study of the properties of 8 drug candidates and their permeability coefficients obtained, that polycarbonate filters were the best choice of solid support for the membrane. We also verified the deposition time of the polydopamine cushion that brings the fluid nature to our membrane. The results from the permeability tests showed that the polymer cushion do not clog the pores of the filter after a 4h deposition. Then, we studied the deposition of the lipid bilayer on the polydopamine-coated filter. To do this, two methods of liposome preparation as well as several liposome sizes were tested. Moreover, several phospholipid compositions were tested. According to our evaluation, liposomes prepared according to the "lipid film" method from a mixture of DOPC and cholesterol are better suited. Nevertheless, a final optimization step for the bilayer deposition is still required. Finally, the gold standard Caco-2 assay used to assess the drug permeability across a monolayer of human Colon Colorectal adenocarcinoma cells was successfully implemented for permeability data comparison.

Métodos in vitro utilizando epitélio intestinal animal são importantes ferramentas para avaliar a permeabilidade de fármacos, propriedade que é um importante parâmetro de biodiosponibilidade. Considerando que o maior objetivo na indústria farmacêutica é desenvolver novos fármacos com boa biodisponibilidade oral, o projeto teve como objetivo padronizar o modelo de permeação com membrana de intestino de rã (Rana catesbeiana) em células de Franz, comparando seus resultados com ensaios de células Caco-2. Os fármacos modelo utilizados foram os antivirais zidovudina e aciclovir. A quantidade de fármaco permeado foi determinada por método de eletroforese capilar para o método com intestino de rã touro e por HPLC-UV para os ensaios com células Caco-2. O parâmetro de permeação foi o coeficiente de permeabilidade aparente (Papp) dos fármacos para ambos modelos experimentais. Para estabelecimento do protocolo experimental dos estudos de permeabilidade intestinal de rã, foi proposto um projeto fracionado 24-1 com 4 ensaios adicionais usando o software Minitab, e as variáveis foram: secção intestinal, pH da solução de Ringer e temperatura. A análise do planejamento experimental feita pela estimativa dos parâmetros da regressão obtidos através dos resultados do modelo fatorial possibilitou a determinação dos coeficientes da equação matemática que definiu a influência das variáveis sobre o coeficiente de permeabilidade aparente dos fármacos. Os efeitos das variáveis pH e temperatura interpretados conjuntamente apresentaram interferência leve, porém as variáveis fármaco e secção intestinal interpretados juntos tiveram interferência importante, mostrando maior permeação dos fármacos através da secção inicial do intestino da rã. Os resultados de Papp foram: para o metoprolol, pelo método das células Caco-2 foi de 28 x 10-6 cm/s, valor que está de acordo com demais dados de células Caco-2 na literatura e o valor obtido com células de Franz que mais se adequada a estes resultados e demais disponíveis provenientes de outras técnicas na literatura, foi de 28,1 x 10-6 cm/s, em uma das condições do planejamento estatístico utilizando segmento final da membrana epitelial da rã. No caso do aciclovir, o resultado de Papp de 0,48 x 10-6 cm/s também obtido em uma das condições envolvendo segmento intestinal final da rã foi exatamente igual a o valor 0,48 x 10 - 6 cm/s, encontrado com células Caco-2 no presente estudo e estão de acordo com outros valores disponíveis na literatura por Trapani e colaboradores, 2004 e também com células Caco-2. Para zidovudina o valor de Papp obtido em uma das condições utilizando segmento intestinal inicial da rã, de 13 x 10-6 cm/s, foi a que mais se assemelhou ao obtido pela técnica de células Caco-2, 13,6 x 10-6 cm/s, e também está de acordo com demais dados da literatura. / There are lots of different in vitro technics in the literature using animal intestinal epithelium to estimate permeability of drugs, property that is an important parameter of bioavailabity. Considering that the main objective of Pharmaceutical Companies is the development of new drugs with good oral bioavailabity, the aim of this work was to standardize the permeability of antivirals using in vitro/ex vivo method of intestinal epithelium of Rana catesbeiana in Franz cells and compare these results to those obtained from studies using Caco-2 cells. Zidovudine and Acyclovir were selected as model drugs. The amount of drug permeated will be determined by the method of capillary electrophoresis for assays using Rana Catesbeiana and HPLC-UV for studies with Caco-2 cells. The permeation parameter determined was the apparent permeability coefficient (Papp) of drugs for both experimental models. To establish the experimental protocol to studies of intestinal permeability of frog, it was proposed a fractional 24-1 design with 4 additional tests using Minitab software and the variables were: intestinal section, pH of Ringer solution and temperature. The analysis of the experimental design made by the estimate of the regression parameters obtained from the factorial model results allowed the determination of the coefficients of the mathematical equation that defines the influence of the variables on the apparent permeability coefficient of acyclovir and zidovudine. The effects of pH and temperature interpreted jointly presented a slight interference, but the variables drug and intes tinal section interpreted together had major interference, showing greater permeation of drugs through the initial section of the intestine of the frog . The results of Papp were: for metoprolol, with the method of Caco-2 cells was 28 x 10-6 cm/s, a value which is consistent with other data of Caco-2 cells provided in the literature and the condition obtained with Franz cells that are most suitable for these and other results obtained from other techniques available in the literature, was 28.1 x 10-6 cm/s, provided with the final intestinal segment using frog epithelial membrane. In the case of acyclovir, the result of Papp of 0.48 x 10-6 cm/s obtained in one condition with final frog intestinal segment was exactly equal to the value of 0.48 x 10-6 cm/s, found with Caco-2 cells in the present study and are in agreement with other values available in the literature for Trapani and colegues, 2004 and also with Caco-2 cells. The Papp value for zidovudine obtained with the initial gut segment of frog, 13 x 10-6 cm /s was which more resembled that obtained by the technique of Caco-2 cells, 13.6 x 10-6 cm/s and is also consistent with other literature data.

A injúria ao epitélio intestinal é uma das mais importantes complicações da sepse, associada à perda da integridade da barreira epitelial intestinal pela alteração da expressão de proteínas constituintes das tight junctions (TJ). Os dois subtipos de receptores de estrógeno são normalmente expressos na mucosa intestinal, sendo responsável pela manutenção da arquitetura do epitélio intestinal. Além disso, diversos modelos experimentais fisiopatológicos têm atribuído um papel imunomodulador ao estradiol. O objetivo do presente estudo foi avaliar a participação do estradiol na modulação da resposta inflamatória e na proteção da barreira epitelial intestinal durante a inflamação sistêmica induzida por lipopolissacarídeo (LPS; 1,5 mg/kg, i.v.) em ratas. As ratas foram ovariectomizadas e mantidas para recuperação durante 10-12 dias antes do experimento. Por três dias consecutivos, as ratas foram tratadas com cipionato de estradiol (50 ou 100 µg/kg, s.c.) ou óleo. Após 6h da indução da endotoxemia, foram avaliadas a permeabilidade intestinal pela injeção de dextrana FITC no íleo ou cólon, a translocação bacteriana nos linfonodos mesentéricos e as citocinas no plasma e na mucosa intestinal. Adicionalmente, a infiltração de mastócitos e neutrófilos foi avaliada no íleo e no cólon, a integridade das TJ e junções aderentes (JA) foi determinada por microscopia eletrônica de transmissão, e expressão das proteínas (ocludina, claudina-1, JAM-A, E-caderina) bem como suas localizações. Nossos resultados demonstraram que o estradiol reduziu a permeabilidade intestinal bem como preveniu a translocação bacteriana nos linfonodos mesentéricos induzidas pela administração de LPS. Em ratas endotoxêmicas tratadas com estradiol, as concentrações das citocinas pró-inflamatórias (TNF?, IL-6, IFN-? e IL-1?), migração de neutrófilos (atividade da mieloperoxidase) e degranulação dos mastócitos no íleo e no cólon foram reduzidas. O estradiol também reverteu a disfunção da barreira epitelial induzida pelo LPS, aumentando a expressão das proteínas das TJ, reduzindo a abertura das TJ e JA e atenuando os danos histológicos. Em conjunto, os resultados sugerem um papel protetor do estradiol, prevenindo a disfunção da barreira epitelialintestinal induzida pela inflamação sistêmica, possivelmente modulando a resposta inflamatória e a liberação de proteases de mastócitos. / Intestinal injury is one of the most important complications of sepsis, associated with the loss of integrity of the intestinal epithelial barrier due to the alteration of the expression of proteins that constitute the tight junctions (TJ). The two subtypes of estrogen receptors are normally expressed in the intestinal mucosa, being responsible for maintaining the architecture of intestinal epithelium. Moreover, several experimental pathophysiological models have been attributed the immunomodulatory role for the estradiol. The aim of the present study was to evaluate the role of estradiol in the modulation of the inflammatory response and the protection of the intestinal epithelial barrier during systemic inflammation induced for lipopolysaccharide (LPS, 1.5 mg / kg, i.v.) in rats. The female rats were ovariectomized and allowed to recover for 10-12 days before the experiment. For three consecutive days, rats were pretreated with estradiol cypionate (50 or 100 µg/kg, subcutaneous) or corn oil. At 6h after of endotoxemia induction, were evaluated the intestinal permeability by injecting FITC dextran into the ileum or colon, bacterial translocation in the mesenteric lymph nodes and plasma and intestinal mucosa cytokines levels. In addition, the infiltration of mast cells and neutrophils was evaluated in the ileum and colon, the integrity of the TJ and adherent junctions (JA) integrity was determined by transmission electron microscopy, and the protein expression (occludin, claudin-1, JAM-A, E-cadherin) as well as their localization. Our results demonstrated that estradiol reduced intestinal permeability as well as prevented bacterial translocation in the mesenteric lymph nodes induced by the LPS administration. In the endotoxemic rats treated with estradiol, the concentration of proinflammatory cytokines (TNF?, IL-6, IFN-? e IL-1?), neutrophil infiltration (myeloperoxidase activity), and mast cells degranulation were reduced in the ileum and colon. Estradiol also reverted the LPS-induced epithelial barrier dysfunction, increasing the expression of the TJ proteins, reducing TJ and AJ opening and attenuating the histological damages. Together, these results suggest a protective role for estradiol, attenuating damage to the intestinal epithelium induced by systemic inflammation, possibly due to modulation of the inflammatory response and the release of mast cells proteases.

Les cellules à mucus intestinales sécrètent des mucines, principalement MUC2, qui forment un gel protecteur. Malgré la place importante des acides gras à chaîne longue (AGCL) dans l'alimentation et leurs liens avec des pathologies où les cellules à mucus sont altérées il n'existe pas d'étude sur leurs impacts sur ces cellules. Mes travaux ont consisté à mettre en lumière ces interactions et à étudier leurs conséquences. In vitro, nous montrons que les AGCL saturés augmentent l'expression de MUC2 et son relarguage alors que les AGCL insaturés les inhibent et que l'acide palmitique favorise la différenciation cellulaire. In vivo, chez des ratons gavés avec de l'huile de palme, la production de MUC2 est augmentée et la perméabilité paracellulaire colique diminuée. Les huiles de colza ou de tournesol ne modifient pas la production de MUC2 et la perméabilité est identique aux contrôles malgré l'augmentation de l'expression de l'occludine. Dans un modèle de syndrome de l'intestin irritable, nous montrons que le gavage avec l'huile de palme, malgré la diminution de l'occludine, restaure la perméabilité et augmente le nombre de cellules à mucus. L'huile de colza et l'huile de tournesol ne corrigent pas le défaut de perméabilité. Parallèlement, deux études chez la souris ont mis en évidence l'impact de certains régimes sur le nombre des cellules à mucus. Ce travail a permis d'identifier les AGCL comme des nutriments capables de moduler les cellules à mucus, de faire émerger une nouvelle piste thérapeutique pour restaurer les populations de cellules à mucus et la perméabilité et d'ouvrir la réflexion sur les impacts de l'augmentation des populations de cellules à mucus chez l'individu sain / Intestinal goblet cells secrete mucins, mainly MUC2, which form a protective mucus gel. Despite the important place of long chain fatty acids (LCFA) in the diet and their links with pathologies where goblet cells are altered, there is no study dealing with their impact on goblet cells. The aim of my work was to highlight these interactions and to study their consequences. In vitro, we show that saturated LCFA increase MUC2 expression and release whereas unsaturated LCFA inhibit these process and that palmitic acid promotes cellular differentiation. In vivo, rat pups receiving oral administration of palm oil show a decrease of colonic paracellular permeability and an increase of MUC2 production. On the opposite, rapeseed and sunflower oils do not change MUC2 production and, intestinal permeability is the same as controls despite the increase of occludin expression. In an animal model of irritable bowel syndrome, palm oil effects are found again. When subjected rat pups to a maternal deprivation stress a few days after birth, they develop an intestinal hyperpermeability and a goblet cell depletion. We show that oral administration of palm oil concomitantly to the stress is able to, despite a decrease of occludin expression, restore the permeability at the level of non-stressed animals and to increase goblet cell number. Rapeseed oil and sunflower oil are not able to correct the increase of intestinal permeability. In parallel, studies in mice show that some types of diets can be associated with an increase of the number of proximal and distal goblet cells. To conclude, this work (i) helped to identify LCFA as nutrients able to modulate intestinal goblet cell number and physiology, (ii) put forward a new therapeutic track, via palmitic acid use, to restore goblet cell populations and intestinal permeability in pathologies associated with these kind of alterations and finally (iii) offered new tracks of reasoning about physiological and pathophysiological impacts of the increase of intestinal goblet cell number in a healthy subject

La maladie de Crohn (MC) est une maladie inflammatoire chronique du tube digestif caractérisée par un état d’hyperactivation du système immunitaire intestinal. Les données cliniques et expérimentales montrent que l’étiologie de la MC serait une réponse immunitaire aberrante à des facteurs environnementaux et/ou infectieux chez un hôte génétiquement prédisposé. Les patients atteints de MC présentent une perméabilité intestinale anormalement élevée pouvant expliquer la stimulation du système immunitaire intestinal par les bactéries et antigènes du microbiote. La muqueuse iléale des patients atteints de MC est anormalement colonisée par des souches de Escherichia coli ayant la propriété d’adhérer et d’envahir les cellules épithéliales intestinales, de survivre et de se multiplier dans les macrophages en entraînant la sécrétion de TNF-α (Tumor Necrosis Factor-alpha). Un pathovar de E. coli associé à la MC et dénommé AIEC pour « Adherent-Invasive E. coli » a été défini. Les souches AIEC adhérent via l’adhésine FimH des pili de type 1 aux résidus mannose de la glycoprotéine CEACAM6, anormalement exprimée au niveau de l’épithélium iléal des patients atteints de MC. Au cours de la maladie, l’inflammation intestinale peut être contrôlée par les traitements médicamenteux ou la chirurgie sans pour autant obtenir de rémission complète et définitive.Le but du travail était d’analyser les conséquences de l’infection par des bactéries AIEC in vivo dans un modèle murin reproduisant l’interaction AIEC/CEACAM6 et de proposer des stratégies pour éliminer ces bactéries de l’intestin. Nous avons montré que les AIEC avaient la capacité d’altérer la fonction de barrière de l’épithélium intestinal chez des souris transgéniques CEABAC10 exprimant CEACAM6. Cette altération est associée à une forte induction de l’expression de la protéine de jonction Claudine-2, comme observée chez les patients atteints de MC. Nous avons ensuite démontré que la levure S. cerevisiae CNCM I-3856 et des produits de levures étaient capables de maintenir l’intégrité de la barrière intestinale des souris en prévenant la colonisation du tractus digestif par les bactéries AIEC. Dans une deuxième partie, nous nous sommes intéressés au développement de nouvelles molécules antagonistes de l’adhésine FimH. Les thiazolylaminomannosides ont montré un puissant effet inhibiteur de l’adhésion des bactéries AIEC à des cellules épithéliales intestinales. Les heptyl-mannosides sont également de puissants inhibiteurs de l’adhésion des AIEC et leur présentation en multivalence sur des corps polymériques ou des cyclodextrines potentialisent l’effet anti-adhésif. De manière intéressante, certains heptyl-mannosides diminuent fortement la colonisation de l’intestin par les bactéries AIEC et préviennent la colite chez des souris CEABAC10. Toutefois, la multivalence n’apporte pas d’efficacité supplémentaire dans ce contexte. En conclusion, deux stratégies anti-adhésives ont été étudiées : les levures et les mannosides. Elles pourraient être proposées aux patients atteints de MC fortement colonisés par les AIEC afin d’éliminer ces bactéries du tractus digestif pour espérer diminuer l’inflammation. Les perspectives à ce travail seront la mise au point de techniques de détection simples et rapides des personnes colonisées ou susceptibles d’être colonisées par ces souches de E. coli pour cibler la population à traiter par les probiotiques levures et les molécules anti-adhésives. / Crohn’s disease (CD) is an inflammatory bowel disease (IBD) with a multifactorial etiology, resulting from an exacerbated inflammatory response to intestinal microbes and/or microbial components in genetically susceptible hosts. CD patients present an increased intestinal permeability which can favor the overstimulation of the intestinal immune system by bacteria or antigens from microbiota. Ileal mucosa from CD patients is abnormally colonized by Escherichia coli strains sharing the ability to adhere to and to invade intestinal epithelial cells, to survive and to replicate within macrophages, inducing high secretion of TNF-α (Tumor Necrosis Factor-). These strains associated with CD are grouped in a pathovar of E. coli named AIEC for « Adherent-Invasive E. coli ». AIEC bacteria adhere via the adhesin FimH localized at the tip of the type 1 pili, to mannose residues exposed on the glycoprotein CEACAM6 abnormally expressed at the ileal mucosa of CD patients. Currently, intestinal inflammation can be controlled with drugs or intestinal surgery but total remission cannot be yet achieved. The aim of the present work was to investigate consequences of AIEC infection in a murine model mimicking the AIEC/CEACAM6 interaction and to test different strategies to eradicate these bacteria from the gut. We showed that AIEC bacteria altered barrier function of the intestinal epithelium in transgenic CEABAC10 mice expressing human CEACAM6. The overexpression of the pore-forming tight junction protein claudin-2 was correlated with the increase intestinal permeability, as observed in CD patients. We demonstrated that the yeast strain S. cerevisiae CNCM I-3856, as well as some yeast products, were able to prevent increase of intestinal permeability in decreasing AIEC gut colonization. In a second part, we investigated another strategy targeting AIEC bacteria using antagonists to FimH adhesin. Thiazolylaminomannosides molecules exerted a strong inhibitory effect on the ability of AIEC bacteria to adhere to intestinal epithelial cells. Heptyl-mannosides (HM) also shared high inhibitory properties in vitro and their efficacy can be potentiated when HM are harbored in multiple copies on polymeric or cyclodextrin cores. Interestingly, some HM molecules strongly decreased AIEC gut colonization and the signs of colitis in vivo, in AIEC LF82-infected CEABAC10 mice. In that context, multivalency did not improve inhibitors efficacy.To conclude, two different strategies were studied: probiotic yeasts and anti-adhesive mannosides. These treatments should be proposed in CD patients highly colonized by AIEC bacteria in order to eliminate these bacteria from the gut and to decrease intestinal inflammation. Future works will focus on the development of quick and easy detection methods to determine people colonized or susceptible to be colonized by AIEC bacteria to treat this subpopulation of CD patients.

No description available.

No description available.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / NÃveis subterapÃuticos de drogas antimicobacterianas tÃm sido observados no curso do tratamento de pacientes com tuberculose e nos co-infectados TB/HIV e podem facilitar o surgimento de cepas de M. tuberculosis resistentes Ãs drogas. A mal-absorÃÃo intestinal inclue-se entre as provÃveis causas e tem sido descrita em tuberculosos desnutridos, alcoÃlatras, diabÃticos, aidÃticos ou com patologias gastrointestinais associadas. Estudos da permeabilidade intestinal permitem a avaliaÃÃo da funÃÃo intestinal em vÃrias doenÃas, mas tÃm sido escassos na tuberculose. O teste da lactulose / manitol tem sido utilizado como critÃrio de lesÃo com dÃficit absortivo da mucosa intestinal e pode ser de utilidade no prognÃstico da absorÃÃo intestinal de drogas em pacientes com tuberculose. Foram estudados 41 pacientes (30H e 11M). A coleta dos dados foi realizada no Hospital de MaracanaÃ, Fortaleza-CE, em 2001. Procedeu-se descriÃÃo clÃnica, social e laboratorial do grupo de pacientes e estudo piloto de observaÃÃo utilizando o teste da lactulose / manitol em 40 pacientes com tuberculose pulmonar ativa, comparando com grupo controle de 28 voluntÃrios sadios objetivando estudar a permeabilidade intestinal. Testes de sensibilidade Ãs drogas antituberculose, pelo mÃtodo das proporÃÃes indireto, foram realizados em 20 pacientes com cultura de escarro positivas. O grau de nutriÃÃo avaliado atravÃs do Ãndice de Massa CorpÃrea (IMC) e o nÃvel sÃrico das drogas foram correlacionados com os valores da permeabilidade intestinal. CinqÃenta e nove por cento (24 pacientes) foram considerados desnutridos pelo IMC<18,5 kg/m2, sendo vinte e dois por cento (9 pacientes) considerados severamente desnutridos (IMC<16 kg/m2). Em 18 pacientes, apÃs 2 h da tomada de 600mg de rifampicina (R) e 400 mg de isoniazida (H) os nÃveis sÃricos de R (CRM 2h) e H (CINH 2h) foram analisados por HPLC. A taxa de excreÃÃo urinÃria de manitol (mÃdia  desvio padrÃo) foi significativamente menor p<0,001 (9,52  5,70) nos pacientes que dos controles (20,14  10,84). A taxa de excreÃÃo de lactulose foi significativamente maior p<0,05 nos pacientes (0,59  1,79) que nos controles (0,52  0,47) e a razÃo L/M foi aumentada de forma consistente, embora nÃo significativa p>0,05 (0,05  0,10 pacientes contra 0,02  0,02 controles). Considerando a faixa terapÃutica da R (8-24 mcg/mL ) e da H (3-6 mcg/mL) observamos que as CRM2h e CINH2h em 16/18 (88,8%) pacientes nÃo atingiram nÃveis sÃricos adequados para as duas drogas. 8/18 (44,4%) pacientes nÃo alcanÃaram nÃveis sÃricos para ambas as drogas simultaneamente. Na anÃlise das drogas isoladas verificamos que em 12/18 (66,7%) pacientes a CRM2h o nÃvel sÃrico mÃnimo nÃo foi alcanÃado (mÃdia R =6,47 mcg/mL) e 13/18 (72,2%) a CINH2h tambÃm foi reduzida (mÃdia H =2,17 mcg/mL). Quanto ao perfil de resistÃncia do M. tuberculosis em cultura de escarro, observamos 4/20 (20%) multirresistentes (3 à R+H e 1 à todas as drogas) e 2/20 (10%) monorresistentes à H sendo 14/20 (70%) amostras consideradas sensÃveis Ãs drogas testadas. Os resultados observados sugerem uma intensa reduÃÃo na Ãrea de absorÃÃo e lesÃo da mucosa intestinal nos indivÃduos estudados. Os dados sÃo consistentes com a reduÃÃo na biodisponibilidade de rifampicina e isoniazida e com o estado nutricional destes pacientes. Os dados preliminares recomendam estudos adicionais na avaliaÃÃo completa da biodisponibilidade das drogas antimicobacterianas em pacientes com tuberculose / Subtherapeutic antimycobacterial drugs levels have been observed during treatment of patients with tuberculosis and in HIV coinfected. It may facilitate the development of M. tuberculosis resistant strains. Malabsorption may be one of the underlyne cause. It has been documented in tuberculosis patients associated to malnutrition, alchoolics, diabetes, AIDS and gastrointestinal symptoms. Intestinal permeability studies have been conducted in order to evaluate the intestinal function in some diseases but are scant in tuberculosis. Lactulose / manitol ratio in urine a widely used measure of malabsorption and intestinal permeability may be useful to assesss the drug absorption area in tuberculosis. In order to study the intestinal permeability in tuberculosis, we conducted in a pilot study 40 patients and 28 healthy volunteers, using the urinary excretion of ingested lactulose and manitol as respective markers of barrier disruption and overall villous surface area. The Maracanaà Hospital in Fortaleza-CE, was choosen and the data were collected in 2001. Eigthteen patients receiving 600 mg rifampicin (R) and 400 mg isoniazid (H) were evaluated through venous blood analysis obtained at 2 hours after directly observed ingestion. The serum samples were analysed by HPLC at the Infectious Disease Pharmacokinetics Laboratory / National Jewish Center / Denver, USA. Nutritional status by body mass index (BMI) and serum drug levels were correlated with intestinal permeability data. M. tuberculosis isolates from 20 clinical specimens were tested to drug susceptibility by standard proportion method (APM ) using Lowestein-Jensen medium. Forty-one patients (30 M, 11 F) were studied. Fifty nine per cent (24) patients were malnourish (BMI<18,5 kg/m2); 22% (9) were severely malnourish (BMI<16 kg/m2). The urinary excretion of mannitol (mean  standard deviation) showed a significant decrease p < 0,001 in patients (9,52  5,70) compared to controls (20,14  10,84). The excretion of lactulose was significant increased p < 0,05 in patients (0,59  1,79), than controls (0,52  0,47) and the L/M ratio shown consistent increased (0,05  0,10) in patients compared to controls (0,029  0,0,02) p > 0,05. Considering the reported target range for R (8-24 mcg/mL ) and H (3-6 mcg/mL) we observed that 2-h serum levels were below the lower limit for R and H in 88,8% (16/18) patients. Forty four per cent ( 8/18) of patients had reduced levels for the two drugs. Analysing single drugs we documented that 66,7% (12/18) patients had levels below the target limit for R (mean R =6,47 mcg/mL) and 72,2% (13/18) for H (mean H =2,17mcg/mL). Concerning to M. tuberculosis susceptibility profile in sputum culture, we observed 20% (4/20) multidrug resistants strains (3 for R + H and 1 for all drugs). Ten per cent (2/20) were resistant only for H and 70% (14/20) strains were sensitive. These results suggests an important decrease in the funcional absorptive surface of the intestine and damage of the intestine in patients studied. The data are consistent with reduced antituberculosis drugs bioavailability and compromissed nutritional status of these patients. The preliminary results indicate the necessity of new approaches to accurate portrait of drugs bioavailability in tuberculosis patients