Introdução: A polpa farinácea do jatobá-do-cerrado (Hymenaea stigonocarpa Mart.) apresenta alto teor de fibra alimentar, em média 60 g/100 g, que são importantes para a redução do risco e controle de doenças crônicas não transmissíveis (DCNT). A extrusão termoplástica neutraliza aromas intensos, proporciona a formação de amido resistente, aumenta a fibra alimentar solúvel e melhora a textura do produto final. Objetivo: Estudar o efeito das farinhas de jatobá-do-cerrado in natura (FIN) e extrusada (FE) no metabolismo lipídico e parâmetros fermentativos em hamsters, bem como verificar a resposta glicêmica em humanos após a extrusão. Métodos: Processo de extrusão: velocidade de 200 rpm; matriz com 4 mm de diâmetro; taxa de compressão 3:1; alimentação constante de 70 gramas/minuto; temperatura de 150 °C; proporção farinha de jatobá-do-cerrado e amido de milho: 70:30 por cento e umidade a 25 por cento . Foi realizado um experimento animal com hamsters durante 21 dias, em que se analisou alguns parâmetros do metabolismo lipídico e colônico (fermentativos) dos animais, divididos em quatro grupos experimentais, se diferenciando pela dieta. As dietas controle (GC), in natura (GFI) e extrusada (GFE) eram hipercolesterolemizantes (13,5 por cento de gordura de coco e 0,1 por cento de colesterol) e a dieta referência (GR) com óleo de soja como fonte lipídica, não. Todas as dietas apresentavam 15 por cento de fibra alimentar, sendo que as dietas GR e GC tinham como fonte de fibra a celulose, e as dietas GFI e GFE tiveram as próprias fibras como fonte. A resposta glicêmica em humanos foi verificada por meio do ensaio do índice glicêmico e carga glicêmica da FE, com dez voluntários saudáveis que consumiram 25 gramas de carboidratos disponíveis do alimento teste (farinha extrusada) ou do pão branco como alimento controle. Resultados: Não foi observada diferença significativa entre o peso final, ingestão diária média e total, ganho de peso e CEA entre os animais dos quatro grupos. A concentração de triglicerídeos foi menor em 41 por cento e 38 por cento nos animais que receberam as dietas GFI e GFE, em relação aqueles que receberam a dieta GC, assim como também para o colesterol total (55 por cento e 47 por cento ), LDL-c (70 por cento e 53 por cento ) e não-HDL-c (63 por cento e 49 por cento ) séricos, lipídeos totais hepáticos (39 por cento e 45 por cento ) e o peso dos fígados dos animais também foi menor (21 por cento em ambos os grupos). Não houve diferença no colesterol hepático e excretado nas fezes dos animais dos quatro grupos. Os animais do GFE excretaram 57 por cento mais ácidos biliares nas fezes que os animais do GC. Com relação aos parâmetros fermentativos, observou-se maior excreção de fibras (1,24 ± 0,08 e 1,52 ± 0,09 gramas) nos animais dos grupos GR e GC respectivamente, em relação aos do GFI e GFE (0,50 e 0,48 gramas), porém o escore fecal (3,50 ± 0,19 e 3,38 ± 0,18) e o grau de fermentação (54 e 52 por cento ) foi maior nos animais dos grupos GFI e GFE. Houve uma maior produção de AGCC no ceco dos animais dos grupos GFI e GFE (80 e 57,5 µmol/g de ceco respectivamente) e maior diminuição do pH no conteúdo cecal nos animais do grupo GFI (7,49 ± 0,10), em relação ao GC (8,06 ± 0,13). Os ácidos acético e propiônico, estiveram presentes em maior quantidade no ceco dos animais dos grupos GFI (58,5 e 6,1 µmol/g de ceco) e GFE (42,5 e 6,6 µmol/g de ceco) e os animais do GFI produziram mais ácido butírico (15 µmol/g de ceco), em relação aos demais grupos. Quanto à resposta glicêmica da farinha pós extrusão, não houve diferença entre a área de resposta glicêmica da farinha extrusada e do pão branco, o índice glicêmico da farinha extrusada (glicose como controle) foi classificado como moderado, e a carga glicêmica (na porção de 30 gramas), baixa. Conclusão: As FIN e FE favoreceram a redução do colesterol total, LDL-c, não-HDL-c e dos triglicerídeos séricos, além da diminuição do acúmulo de lipídeos hepáticos. Foi observado também aumento expressivo na formação de AGCC e no grau de fermentação. A FE proporcionou um aumento na excreção de ácidos biliares nas fezes e apresentou índice glicêmico moderado e baixa carga glicêmica. / Introduction: The mealy pulp of jatobá-do-cerrado (Hymenaea stigonocarpa Mart.) has high dietary fiber content, on average 60 g/100 g, which is important for reducing risk and control chronic non-communicable diseases (NCDs). The thermoplastic extrusion neutralizes intense flavors, formation of resistant starch, increases soluble dietary fiber and improves the texture of the final product. Objectives: Check the effect of jatobá-do-cerrado flour in natura (IF) and extruded (EF) on lipid metabolism and fermentation parameters in hamsters as well as the glycemic response in humans after extrusion. Methods: Extrusion cooking: speed of 200 rpm; 4 mm die diameter; compression ratio 3:1; constant feed rate 70 g/min; temperature 150 °C; ratio of jatobá-do-cerrado flour to corn starch: 70:30 per cent and 25 per cent moisture. A 21-day biological experiment with hamsters was conducted, in which parameters of the animals lipid metabolism and colonic (fermentation) were analyzed; they were divided into four groups, differing by diet. The control diets (CG), in natura (IFG) and extruded (EFG) were hypercholesterolemic (13.5 per cent coconut fat and 0.1 per cent cholesterol), and the reference diet (RG) with soybean oil as a lipid source was not. All diets had 15 per cent of dietary fiber, being that RG and CG diets had the cellulose fiber source, and IFG and EFG diets had the fibers themselves as a source. The glycemic response in humans was evaluated by the glycemic index and glycemic load test of EF, with ten healthy volunteers who consumed 25 grams of available carbohydrate food test (extruded) or white bread as food control. Results: There was no significant difference between final weight, average total and daily intake, weight gain and FER between animals of the four groups. The values of triglycerides were lower by 41 per cent and 38 per cent in the animals receiving the IFG and EFG diets, for those who received the CG diet, as well as for total cholesterol (55 per cent and 47 per cent ), LDL-c (70 per cent and 53 per cent ) and non-HDL-c levels (63 per cent and 49 per cent ) serum, liver lipids (39 per cent and 45 per cent ) and the animals\' livers weight was lower (21 per cent in both groups). There was no difference in hepatic cholesterol and excreted in feces of animals of the four groups. The animals of the EFG group excrete 57 per cent more bile acids in stool than animals CG. Regarding the fermentation parameters, we observed increased excretion of fibers (1.24 ± 0.08 and 1.52 ± 0.09 g) in animals of RG and CG respectively, in relation to IFG and EFG (0.50 and 0.48 grams), but the fecal score (3.50 ± 0.19 and 3.38 ± 0.18) and the degree of fermentation (54 and 52 per cent ) were higher in animal IFG and EFG groups. There was a greater production of SCFA in the caecum of animals IFG groups and EFG (80 to 57.5 µmol/g caecum respectively) and greater decrease in pH in the cecal contents of IFG animals group (7.49 ± 0.10) compared to the control group (8.06 ± 0.13). The acetic and propionic acids were present in higher amounts in the caecum of animals IFG groups (58.5 and 6.1 µmol/g caecum) and EFG (42.5 and 6.6 µmol/g caecum) and the animals of IFG group produced more butyric acid (15 µmol/g caecum), compared to other groups. Concerning the glycemic response of flour after extrusion, there was no difference between the glycemic response area of extruded flour and white bread, the glycemic index of extruded flour (glucose as a control) was classified as moderate and glycemic load (the portion of 30 grams) low. Conclusion: IF and EF aided in the reduction of total cholesterol, LDL-c, non-HDL-c and triglycerides, as well as decreased the accumulation of liver lipids. A significant increase in the SCFA and degree of fermentation was also observed. The EF increased the excretion of bile acids in stool, and presented moderate glycemic index and low glycemic load.

Orientador: Mário Roberto Maróstica Junior / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia de Alimentos / Made available in DSpace on 2018-08-24T07:02:00Z (GMT). No. of bitstreams: 1 Cazarin_CinthiaBauBetim_D.pdf: 7421314 bytes, checksum: 3937f33ea2104648545ff81374161389 (MD5) Previous issue date: 2014 / Resumo: A doença inflamatória intestinal (DII) édoença crônica recidivante que atinge milhões de pessoas no mundo, englobando a Doença de Crohn (DC) e a Retocolite Ulcerativa (RCU). A diferença entre as duas éa sua localização, sendo a RCU específica da região do cólon e reto. O tratamento atual para estas patologias é realizado a base de corticosteróides, imunomoduladores ou anti -TNF-a, conhecido como terapia biol ógica, os quais apresentam diversos efeitos colaterais ao paciente. A patogênese desta doença está relacionada com fatores gen éticos, imunol ógicos e ambientais. Acredita-se que o desequilíbrio da microbiota, assim como a ruptura na barreira natural exercida pela mucosa intestinal seja o primeiro passo para o desencadeamento da resposta inflamatória. As fibras alimentares apresentam função sobre a modulação da microbiota, sendo utilizada como substrato para a formação de ácidos graxos de cadeia curta (AGCC), gerados por meio do processo de fermentação. Em adição, os compostos fenólicos presentes nos alimentos apresentam atividade antioxidante e anti -inflamatória que podem atuar na modulação do processo inflamatório. Sendo assim, o objetivo deste trabalho foi avaliar a utilização de um subproduto da indústria de alimentos, a casca do maracujá, como fonte de fibras e compostos fenólicos, na alimentação de ratos com colite induzida por TNBS e sua influência no processo inflamatório. Ratos Wistar foram alimentados com dieta AIN-93, sendo substituídos 50% da celulose da dieta padrão (AIN) por fibras da casca do maracujá (PFF), em dois ensaios biológicos: prevenção e tratamento. O dano causado àmucosa foi avaliado macro e microscopicamente, assim como a expressão de marcadores inflamatórios. Avaliação da microbiota e formação de AGCC foram realizadas no conteúdo cecal. Embora a avaliação macroscópica da mucosa tenha apresentado um escore maior para o grupo PFF no ensaio prevenção, a avaliação microscópica em ambos os ensaios não mostrou diferença no dano àmucosa entre os grupos. O ensaio tratamento mostrou diminuição da peroxidação lipídica do cólon, diminuição na contagem de enterobactérias e aeróbios totais, assim como aumento de ácido acético e butírico nas fezes do grupo PFF. Jáno ensaio prevenção foi observada modulação dos lactobacilos e bifidobactérias. Com relação aos marcadores inflamatórios, foram observadas modulações significativas da expressão de IKK?, COX-2 e iNOS nos animais alimentados com a dieta PFF. Estes resultados sugerem que a casca do maracujá Passiflora edulis pode modular a microbiota aumentando a produção de AGCC, assim como a expressão de marcadores inflamatórios observados na colite induzida por TNBS. Desta forma, a casca do maracujá poderia ser utilizada como coadjuvante na terapêutica da DII como fonte de fibras e polifenóis / Abstract: Inflammatory bowel disease (IBD) is a chronic relapsing disease that affects millions of people worldwide, encompassing Crohn's disease (CD) and ulcerative colitis (UC). UC is an inflammation specific to the region of the colon and rectum. Current treatments for these diseases are based on the use of corticosteroids, immunomodulators or biological therapy, which have various side effects to the patient. The pathogenesis of IBD is related to genetic, immunological and environmental factors. It is believed that the microbial imbalance as well as natural break in the barrier exerted by the intestinal mucosa is the first step in triggering the inflammatory response. Food dietary fiber presents capacity to modulate the microbiota and improve short chain fatty acids (SCFA) formation, by fermentation process. In addition, the phenolic compounds present in the food have antioxidant and anti-inflammatory activities that can modulate the inflammatory process. Thus, this study aimed to evaluate the use of a byproduct of the food industry, the passion fruit peel, as a source of fiber and phenolic compounds in the diet of rats with TNBS-induced colitis and its influence on the inflammatory process. Wistar rats were fed a modified AIN-93 (50% of cellulose was replaced by passion fruit peel PFF) to evaluate prevention and treatment of colitis induced by TNBS. The damage to the mucosa was evaluated macroscopically and microscopically, as well as the expression of inflammatory markers. Evaluation of the microbiota and formation of SCFA in cecal contents were performed. The macroscopic appearance of the mucosa damage in the group PFF was higher than AIN in prevention trial. However, the microscopic evaluation in both trials showed no difference in mucosal damage amongst the groups. Treatment trial showed that PFF could promote a decrease in lipid peroxidation of the colon, decrease in enterobacteria and total aerobics counts, as well as increase in acetic and butyric acid in the stool. On the other hand, the prevention trial showed that the ingredient could exert modulation on lactobacilli and bifidobacteria. The inflammatory markers showed significant modulation, mainly IKK?, COX-2 and iNOS in animals fed with PFF diet. These results suggest that the passion fruit peel, Passiflora edulis, can modulate the microbiota, increase the production of SCFA, and modulate the expression of inflammatory markers observed in TNBS-induced colitis. Passion fruit peel could be used in the treatment of IBD as a source of fiber and polyphenols / Doutorado / Nutrição Experimental e de Alimentos / Doutora em Alimentos e Nutrição

A utilização de aditivos em dietas com altas proporções de concentrado para bovinos em confinamento tem sido uma prática bastante usual, visando o aumento da eficiência do sistema produtivo. Porém, em virtude de algumas restrições ao uso de substâncias antibióticas, devido a possibilidade do surgimento de microrganismos resistentes, o uso de produtos alternativos tem sido alvo de diversas pesquisas. Nesse contexto, este trabalho foi desenvolvido para avaliar os efeitos da inclusão de enzimas exógenas, em dietas de elevada proporção de concentrado para bovinos Nelores confinados. Foram realizados 3 experimentos, no primeiros foram utilizados 96 bovinos nelore alimentados com 3 níveis de inclusão de enzimas fibrolíticas exógenas (EFE) e duas granulometrias de milho (fino e grosso). Não foi observada interação entre os dois fatores testados, a granulometria do milho não influenciou o desempenho, características de carcaça, qualidade de carne, fermentação ruminal e morfologia ruminal. A inclusão de EFE não alterou o desempenho, qualidade de carne, morfologia e fermentação ruminal. Porém foi observado efeito quadrático, para GRPI em % e kg e efeito para a espessura de gordura subcutânea entre a 12ª e 13ª costela (EGS) em relação aos níveis de EFE. Para a realização dos 2º e 3º experimentos foram adotados os mesmos tratamentos, sendo composta de 90% de concentrado e 10% de volumoso (bagaço de cana), contendo 2 níveis de enzimas amilolíticas exógenas (EAE; com e sem), e dois níveis de amido (alto e baixo). No segundo experimento foram utilizados 48 bovinos nelores confinados distribuídos em um delineamento totalmente casualizados em esquema fatorial 2 x 2 (nível de amido X nível de enzima). Foi observado interação entre os fatores para espessura de gordura subcutânea na garupa (EGG), sendo que os animais alimentados com dietas alto amido e com enzima apresentaram resultados superiores quando comparados com os demais. Ao observar o efeito da utilização das enzimas, os animais alimentados com dietas sem enzima apresentaram maior espessura de gordura subcutânea 1,83 vs 1,50 mm e maior área de olho de lombo 61,85 vs 60,07 cm² em relação aos animais alimentados com dietas com enzimas. Já o efeito da suplementação com diferentes teores de amido, os animais alimentados com dietas de alto amido apresentaram maior ingestão de matéria seca (IMS), menor eficiência alimentar (EA) e rendimento de carcaça (RC). O terceiro experimento foi realizado utilizando 8 bovinos Nelore canulados no rúmen distribuídos em dois quadrados latinos contemporâneos em esquema fatorial 2 x 2 (nível de amido X nível de enzima). Foi observado efeito de interação entre os tratamentos (EAE e nível de amido) para IMS e nitrogênio amoniacal (N-NH3). Os animais alimentados com enzima apresentaram menores concentrações de propionato e maior relação acetato/propionato (A/P) em comparação com aos animais alimentados sem enzima. As dietas com alto amido diminuíram a relação A/P em relação ao que animais alimentado com baixo amido. A adição de enzimas exógenas não causou grande impacto no desempenho, qualidade de carne, porém afetaram a deposição de gordura de acabamento nas carcaças, juntamente com os padrões fermentativos ruminais. / The use of additives in diets with high proportions of concentrate for cattle in confinement has been a very usual practice, aiming at increasing the efficiency of the productive system. However, due to some restrictions on the use of antibiotic substances, due to the possibility of resistant microorganisms, the use of alternative products has been the subject of several studies. In this context, this study was developed to evaluate the effects of the inclusion of exogenous enzymes in high concentrate diets for feedlot Nellore cattle. Three experiments were carried out. In the first one, 96 Nelore cattle were fed with 3 levels of inclusion of exogenous fibrolytic enzymes (EFE) and two corn granulometry (fine and crush). There was no interaction between the two factors tested, corn grain size did not influence performance, carcass characteristics, meat quality, ruminal fermentation and ruminal morphology. The inclusion of EFE did not alter the performance, meat quality, ruminal morphology and fermentation. However, a quadratic effect was observed for pelvic kidney and inguinal fat (PIKF) in% and kg and effect on the thickness of the subcutaneous fat between the 12th and 13th rib (EGS) in relation to EFE levels. For the 2nd and 3rd experiments, the same treatments were used, being composed of 90% of concentrate and 10% of sugarcane bagasse, containing 2 levels of exogenous amylolytic enzymes (EAE, with and without), and two levels of starch (high and low). In the second experiment were used 48 confined Nellore cattle distributed in a completely randomized design in a 2 x 2 factorial scheme (starch level X enzyme level). Interaction between the factors for subcutaneous fat thickness in the croup (EGG) was observed, and the animals fed with high starch and enzyme diets presented superior results when compared with the others. When observing the effect of the use of the enzymes, the animals fed with diets without enzyme had a greater thickness of subcutaneous fat 1.83 vs 1.50 mm and greater area of loin eye 61.85 vs 60.07 cm² in relation to fed animals With enzyme diets. On the other hand, the effect of supplementation with different starch contents, the animals fed with high starch diets presented higher dry matter intake (IMS), lower feed efficiency (AE) and carcass yield (CR). The third experiment was carried out using 8 rumen cannulated Nellore cattle distributed in two contemporary Latin squares in a 2 x 2 factorial scheme (starch level X enzyme level). It was observed interaction effect between the treatments (EAE and starch level) for IMS and ammoniacal nitrogen (N-NH3). The animals fed with enzyme had lower concentrations of propionate and higher acetate / propionate ratio (A / P) compared to animals fed without enzyme. High starch diets decreased the A / P ratio in relation to animals fed low starch. The addition of exogenous enzymes did not have a great impact on the performance, meat quality, however they affected the deposition of finishing fat in the carcasses, along with ruminal fermentative standards.

Introduction: Microbial fermentation of carbohydrates in forestomach of ruminants produces large amounts of short-chain fatty acids (SCFA, mainly acetic acid, propionic acid, and n-butyric acid). The majority of these substrates is taken up directly across the ruminal wall. After luminal uptake into the epithelial cells, SCFA mainly occur in the dissociated form due to the intracellular pH of ~7.4. Moreover, a big portion of SCFA is metabolised within the cytosol. Main end products of epithelial SCFA metabolism are ketone bodies (D-3-hydroxybutyric acid and acetoacetic acid) and lactic acid. Both intact SCFA and ketone bodies and lactate need to be efficiently extruded from the ruminal epithelial cells to prevent a lethal drop of intracellular pH and counteract osmotic load of the cytosol. All these substances are less lipophilic in comparison to the undissociated form of SCFA. Thus, dissociated SCFA (SCFA-) and their metabolites need Protein mediated mechanisms for the extrusion across the basolateral side of ruminal epithelium. One mechanism suggested to be involved in the extrusion of SCFA- across basolateral membrane of the ruminal epithelium is the monocarboxylate transporter 1 (MCT1). Functionally, MCT1 was first assumed to operate as proton-coupled transporter for monocarboxylates including SCFA. Nonetheless, a recent study found a bicarbonate dependent anion exchange mechanism which turned out to be sensitive to MCT1 Inhibitors at the basolateral side of the ruminal epithelium pointing to the ability of MCT1 to act as an anion exchanger. However, in these experiments the inhibition of MCT1 abolished bicarbonate dependent transport only by half. This suggests the involvement of further anion exchanger(s) in the transport of SCFA across the basolateral membrane of ruminal epithelium. Promising candidates to underlie this exchange are the putative Anion exchanger 1 (PAT1) and a transport protein designated „down-regulated in adenoma“ (DRA). Materials and Methods: Sheep rumen epithelium was mounted in Ussing Chambers under short-circuit conditions. Radioactively labelled acetate (ac) was added to the serosal side. Serosal to mucosal flux of ac (Jsm ac) was measured with or without anion Exchange inhibitors (50 mM NO3- or 1 mM DIDS) or the MCT1 inhibitor p-hydroxy mercuribenzoic acid (pHMB; 1.5 mM) in the serosal buffer solution. The inhibitors were added alone or in combination with each other. Furthermore, mucosal to serosal flux of radioactivelly labelled ac or butyrate (bu) (Jms ac, bu) was measured in the presence or absence of SO42-, Cl- or NO3- (50 mM respectively) as exchange substrate in the serosal buffer solution. Immunohistochemical staining was conducted to locate PAT1 and DRA by use of commercially available antibodies. Results: NO3- and pHMB significantly reduced Jsm ac by 57 % and 51 %, respectively. When pHMB was applied after pre-incubation with NO3- an additional inhibition of Jsm ac was observed. Vice versa, NO3- further inhibited Jsm ac when epithelia were pre-incubated with pHMB before. DIDS had no inhibitory effect on SCFA flux. Serosal presence of SO42- or Cl- enhanced Jms ac significantly. Regarding bu, Cl- or SO4 2- also enhanced Jms bu significantly. The different anions available in the serosal buffer solution numerically enhanced Jms in the order of SO4 2- > Cl- for both ac and bu, which corresponds to the known affinity sequence of PAT1 and DRA. Immunohistochemistry revealed localization of PAT 1 in the stratum basale, whereas DRA was not detectable using this method. Conclusions: Basically, this study supports the suggestion that MCT1 works as an Anion exchanger in ruminal epithelium. In addition, it clearly shows that there is at least one further anion exchanger involved in the basolateral extrusion of SCFA and their metabolites. The functional and immunohistochemical findings suggest that PAT1 holds a significant role in this respect.:1 Introduction 1 2 Literature Review 3 2.1 Importance of short-chain fatty acid production of ruminants 3 2.2 Apical uptake of short-chain fatty acids from the rumen 5 2.2.1 Apical uptake of undissociated SCFA from the rumen 6 2.2.2 Apical uptake of dissociated fatty acids from the rumen 8 2.3 Intraepithelial metabolism of short-chain fatty acids 9 2.4 Mechanisms for the basolateral discharge of the short-chain fatty acids 11 2.4.1 Basolateral extrusion of short-chain fatty acids in other gastrointestinal tract epithelia 12 2.4.2 Basolateral extrusion of short-chain fatty acids in ruminal epithelium 14 2.4.3 Further candidate proteins for extrusion of SCFA- in exchange for HCO3 - 19 2.4.3.1 Putative Anion transporter 1 (PAT1 = SLC26A6) 19 2.4.3.2 Down-regulated in adenoma (DRA = SLC26A3) 21 2.4.3.3 Anion exchanger 2 (AE2 = SLC4A2) 22 2.5 Literature implications for this study 23 3 Materials and Methods 24 3.1 Animals 24 3.2 Ussing chamber studies 24 3.2.1 Buffer solutions 24 3.2.2 Preparation of ruminal epithelium 25 3.2.3 Incubation 25 3.2.4 Electrophysiological parameters 26 3.3 Experimental procedure 27 3.3.1 Determination of the unidirectional SCFA flux rate 29 3.4 Experimental Setups 30 3.4.1 Sensitivity of Jsm ac to inhibitors 30 3.4.1.1 Effect of nitrate and pHMB on Jsm ac 30 3.4.1.2 Effect of DIDS, NO3 - and pHMB on Jsm ac 31 3.4.2 Effect of the basolateral replacement of the anions on the extrusion of SCFA 32 3.4.2.1 Effect of Cl- and NO3 - on Jms of acetate and butyrate 32 3.4.2.2 Effect of SO4 2- on Jms of acetate and butyrate 32 3.4.3 Effect of different anions available in the serosal solution on Jms of acetate and butyrate 33 3.5 Immunohistochemistry 34 3.5.1 Preparation of the samples. 34 3.5.2 Fixation and staining of the samples. 34 3.5.3 Evaluation 35 3.6 Statistical analysis 36 4 Results 37 4.1 Inhibitors sensitivity 37 4.1.1 Effect of nitrate and pHMB on Jsm ac 37 4.1.2 Effect of DIDS, pHMB and NO3 - on Jsm ac 41 4.2 Effect of Cl- and NO3 - on Jms of acetate and butyrate 43 4.2.1 Effect of SO4 2- on Jms of acetate and butyrate 44 4.3 Effect of Cl-, NO3 - or SO4 2- when present in the serosal solution for 150 min 49 4.4 Immunohistochemistry 52 5 Discussion 54 5.1Ussing chamber experiments 56 5.1.1 Effect of Cl- and NO3 - on Jms of acetate 56 5.1.2 Effect of nitrate and pHMB on Jsm of acetate 57 5.1.3 Effect of DIDS, pHMB or NO3 - on Jsm of acetat 58 5.1.4 Effect of SO4 2- on Jms of acetate 59 5.1.5 Comparison between different anions as exchange substrate for the basolateral extrusion of acetate 60 5.2 Immunohistochemistry 62 5.3 Comparison between basolateral extrusion of butyrate and acetate 62 5.4 Conclusions 64 6 Summary 66 7 Zusammenfassung 68 8 References 70 Ac Aknowledgements

Background: Diet may play an important role in the management of patients with short bowel syndrome who have colon in continuity. However, macronutrient absorption has not been well characterized, and the most appropriate dietary constituents have not been well defined. Objective: To define carbohydrate absorption characteristics in patients with short bowel syndrome and determine the potential role of pectin as a dietary substrate. Methods: The authors studied the effect of a custom pectin-based supplement in 6 subjects (3 male/3 female) aged 29-67 years with jejunocolonic anastomosis, 4 of whom required long-term parental nutrition. Small intestinal absorption capacity, macronutrient and fluid balance, gastrointestinal transit time, and energy consumption were measured. Results: Data showed that 53% nitrogen, 50% fat, and 32% total energy were malabsorbed. In contrast, the majority (92%) of total carbohydrate was utilized. Fecal short-chain fatty acids (SCFAs) were increased, an indication of increased fermentation. Although only 4% of starch was recovered in stool, it is indicative of considerable starch malabsorption, thus providing the main carbohydrate substrate, for colonic bacterial fermentation. In contrast, non-starch polysaccharide was a relatively minor fermentation substrate with only 49% utilized. Eighty percent of the pectin was fermented. Supplementation was associated with increased total SCFAs, acetate, and propionate excretion. There was a trend observed toward greater fluid absorption (-5.9% ± 25.2%) following pectin supplementation. Nonsignificant increases in gastric emptying time and orocolonic transit time were observed. Conclusion : Despite malabsorption, starch is the primary carbohydrate substrate for colonic bacterial fermentation in patients with short bowel syndrome, although soluble fiber intake also enhances colonic SCFA production.

L’intestin est un organe majeur de l’organisme de par ses fonctions et sa localisation, établissant une barrière active avec un environnement complexe composé du microbiote intestinal, des aliments digérés et d’éléments sécrétés par l’hôte. Outre ses fonctions digestives, absorptives et immuno-modulatrices, l’intestin est également un important organe endocrinien, sécrétant une vingtaine d’hormones régulant des fonctions physiologiques telles que la prise alimentaire, le métabolisme énergétique ou la digestion et le transit intestinal. Ces hormones sont produites par une famille de cellules épithéliales, les cellules entéroendocrines, et sécrétées en réponse à l’activation de récepteurs reconnaissant des éléments du contenu intestinal. En particulier, les cellules entéroendocrines de type L sécrètent GLP-1 et Peptide YY (PYY), impliqués respectivement dans le contrôle de la sécrétion d’insuline et dans la régulation de la prise alimentaire ainsi que le contôle du transit intestinal. Elles sont majoritairement localisées dans l’iléon et le côlon, là où le microbiote intestinal est le plus dense. Le microbiote intestinal permet notamment la fermentation des fibres en acides gras à chaîne courte (AGCC), la production de vitamines, la maturation du système immunitaire de l’hôte et joue lui-même un rôle de barrière contre les pathogènes. Un dialogue entre le microbiote intestinal et l’hôte est nécessaire dans le maintien de l’homéostasie intestinale, nécessitant la reconnaissance par l’hôte de produits bactériens. En particulier, les récepteurs Toll-Like (TLR) permettent la reconnaissance de motifs moléculaires microbiens conservés et sont impliqués dans l’immunité innée de l’hôte. Certains produits bactériens ont également un rôle physiologique tels que les AGCC qui sont une source d’énergie importante pour les colonocytes, en plus d’activer des voies de signalisation. Il a été montré que des régimes riches en fibres, et donc permettant une production accrue d’AGCC, ou plus directement l’administration d’AGCC dans le colon, induit chez l’Homme ou la souris une augmentation des concentrations plasmatiques de PYY, par des mécanismes encore peu compris. En utilisant des lignées cellulaires humaines modèles de cellules entéroendocrines, nous avons caractérisé les effets des AGCC et des motifs bactériens reconnus par les TLR sur l’expression et la sécrétion de PYY et les réponses calciques dans ces cellules. Nous avons pu démontrer que les TLR sont exprimés de manière fonctionnelle, à l’exception de TLR4 et TLR8 dans ces cellules, et que le butyrate augmente leur expression et leur activité. De plus, la stimulation des TLR augmente l’expression de Pyy d’un rapport de 2, mais a peu d’effet sur la sécrétion dans ces cellules. Les AGCC ont des effets divers sur l’expression et la sécrétion de PYY. Alors que le butyrate et le propionate augmentent très fortement l’expression de Pyy, par des rapports respectivement de 120 et 40, par un mécanisme d’inhibition des déacétylases d’histone et de lysine, l’acétate augmente l’expression de Pyy plus modestement par l’activation des récepteurs aux AGCC FFAR2 et FFAR3. L’activation de FFAR2 par les AGCC induit une forte réponse calcique oscillatoire induisant la sécrétion de PYY alors que l’activation de FFAR3 et de GPR109a par le butyrate diminue la concentration calcique cellulaire et réduit les réponses sécrétoires. Ainsi, les AGCC augmentent la production de PYY et régulent sa sécrétion, mais avec et par des effets différents. Ces travaux ont permis de montrer le rôle des cellules entéroendocrines humaines de type L dans la reconnaissance de produits bactériens par l’expression de TLR et par leurs réponses aux AGCCs modulant l’expression et de la sécrétion de PYY. De plus, ces résultats ont déterminés en partie les mécanismes impliqués dans la réponse bénéfique de l’hôte à la consommation de fibres et l’augmentation de la production d’AGCC. / The human gut exerts major functions, mainly due to its localization and by forming an active barrier between a complex environment made of the gut microbiota, digested food products and secreted elements by the host. The main functions of the gut are digestion and absorption of nutrients and it is the first pool of immune cells and a barrier against pathogens, but the gut is also a main endocrine organ secreting more than twenty different hormones. These hormones regulate a wide range physiological functions including food intake, energy metabolism or digestion. Enteroendocrine cells, a sparse family of intestinal epithelial cells, produce and secrete these hormones in response to the activation of a variety of receptors that sense luminal content. Among them, L-cells secrete GLP-1 and Peptide YY (PYY) that are implicated in the regulation of insulin secretion, food intake and intestinal motility. They are mainly found in the distal ileum and in the colon where the microbiota is the densest. Gut microbiota ferments fibers into short chain fatty acids (SCFAs), produces vitamins, participates in regulation of host immune system and is a barrier against pathogens. The cross talk between microbiota and intestinal epithelium is important to maintain the local homeostasis, and is mediated by host receptors recognizing microbial products. Among them, Toll-like receptors (TLRs) recognize conserved microbial associate molecular patterns (MAMPs) and participate to the host innate immunity. Some microbial products also have important functions for the host such has SCFAs that are an important energy substrate for colonocytes and can also activate different signaling pathways. It was shown that fiber-rich diets, increasing production of SCFAs, as well as direct administration of SCFAs in the colon in humans or mice increased PYY plasma levels through mechanisms still undeciphered. Taking advantage of human cell lines as L-cell models, we assessed the different effects of SCFAs and TLR stimulation on PYY expression and secretion and calcium signaling in these cells. We showed that TLRs are functionally expressed in these cells at the exception of TLR4 and TLR8, and that butyrate, one of the three main SCFAs produced by the microbiota increases cell sensitivity to TLR stimulation by increasing their expression. Moreover, TLR stimulation increases Pyy expression by a fold of two but has little effect on secretion. SCFAs differently regulate Pyy expression. Propionate and butyrate highly increase Pyy expression by a fold of 40 and 120 respectively, and their effects are mainly mediated by inhibition of lysine/histone deacetylases whereas acetate increases expression of Pyy by a fold of 1.8 through stimulation of FFAR2 and FFAR3. SCFAs also induce a strong FFAR2-dependent oscillatory response monitoring PYY secretion whereas butyrate via FFAR3 and GPR109a decreases cytosolic calcium concentration and consequently reduces secretory responses. Thus, SCFAs differently increase PYY production and secretion depending of their chain length. Altogether, these results highlight the role human L-cells in microbiota-host crosstalk by sensing microbial products through expression of TLRs and their responses to SCFAs modulating PYY production and secretion. Furthermore, we deciphered some of the mechanisms implicated in beneficial host response to enriched fiber diets and increased production of SCFAs.

Es gibt keine bekannte Lungenerkrankung, die eine so frühe, chronische und intensive Entzündungsreaktion hervorruft, wie sie in den Atemwegen von Patienten mit Mukoviszidose (CF) auftritt. CF ist die häufigste tödliche autosomal-rezessiv vererbte Krankheit in der kaukasischen Bevölkerung, die durch eine Mutation im CFTR-Gen (Cystic Fibrosis Transmembrane Conductance Regulator) verursacht wird, dass für das CFTR-Protein kodiert. Defekte in diesem Protein führen zu einer epithelialen Dysfunktion und betreffen mehrere Organe, aber die Lungenpathologie ist für über 85% der Morbidität und Mortalität bei CF verantwortlich. Die CF-Lungenpathologie konzentriert sich auf die Wechselwirkungen zwischen Wirt und Erreger, wobei die CFTR-Dysfunktion Infektionen begünstigt und die Infektionen in Verbindung mit einer dysfunktionalen Immunantwort einen anhaltenden Entzündungskreislauf in Gang setzen. Dieser Teufelskreis aus Infektion und Entzündung führt schließlich zu Lungenschäden, Atemversagen und letztlich zum Tod. Die vorherrschende Infektion bei Mukoviszidose ist die durch P. aeruginosa, wobei im europäischen Durchschnitt 41% der erwachsenen Patienten infiziert sind. Bemerkenswert ist, dass das Mikrobiom der Atemwege bei Mukoviszidose polymikrobieller Natur ist. Da frühere Studien einen positiven Zusammenhang zwischen einer hohen Mikrobiom-Diversität und einer verbesserten Lungenfunktion bei Mukoviszidose festgestellt haben, wurde die Hypothese aufgestellt, dass bestimmte Kommensalen vor einer Infektion mit P. aeruginosa in den CF-Atemwegen schützen könnten. Um dies genauer zu untersuchen wurden 105 kommensale Isolate aus 32 verschiedenen Arten von Sputumproben von Patienten mit Mukoviszidose isoliert und mit einem fluoreszierenden P. aeruginosa PA01-mcherry-Stamm auf antagonistische Wirkungen bei direkten Erreger-Kommensalen-Interaktionen untersucht. Diese Isolate wurden zusätzlich auf immunmodulatorische Effekte bei Kommensal-Wirt-Pathogen-Interaktionen, unter Verwendung von BEAS-2B-Bronchialepithelzelllinien, untersucht. Ausgewählte Isolate mit schützender Wirkung wurden anschließend auf immunmodulatorische Effekte unter Verwendung von CFBE41o ΔF508-Zellen und einem natürlicheren Lungen-Präzisionsschnitt-Modell (PCLS) untersucht und die produzierten Zytokine mittels ELISA sowie mit einem Multiplex-Zytokin-Assay gemessen. Genexpressionsanalysen wurden zudem mittels qRT-PCR durchgeführt. Die zugrundeliegenden Mechanismen wurden mittels transkriptomischer Analysen, Vergleiche der gesamten Genomsequenz (WGS) und mechanischer Studien einschließlich Stoffwechselanalysen mittels Hochleistungsflüssigkeitschromatographie untersucht. Es konnte gezeigt werden, dass ausgewählte Streptokokken-Kommensal-Isolate, insbesondere Vertreter von S. mitis, S. oralis, S. cristatus, S. gordonii, S. sanguinis und S. parasanguinis, starke antagonistische Effekte auf das Wachstum von P. aeruginosa in direkten Kokulturen haben. Ausgewählte Vertreter von S. mitis, S. oralis und S. cristatus verhinderten zudem das Wachstum anderer klinischer und nicht-klinischer Isolate von P. aeruginosa, sowie anderer typischer Mukoviszidose-Erreger wie Staphylococcus aureus, Burkholderia spp., Achromobacter xylosoxidans, Proteus mirabilis, Haemophilus influenzae, Stenotrophomonas maltophilia, Enterococcus faecalis und Klebsiella pneumoniae. Eine wirksame Mukoviszidose-Therapie sollte nicht nur die Infektion, sondern auch die damit einhergehende bösartige Entzündung bekämpfen. Im Gegensatz zu den Mitgliedern der gram-negativen Neisseria spp., die die IL-8-Produktion bei einer Monoinfektion signifikant stimulierten, taten dies alle gram-positiven Kommensalen-Isolate nicht. Ausgewählte Kommensalen regulierten auch die P. aeruginosa PA01- und LPS-induzierte Produktion mehrerer entzündlicher Zytokine in menschlichen Atemwegsepithelzellen (BEAS-2B sowie CFBE41o ΔF508 und korrigierte wtCFTR) und in PCLS der Maus. Diese Ergebnisse wurden auch durch Genexpressionsanalysen bestätigt, was darauf hindeutet, dass die Immunmodulation möglicherweise durch eine veränderte TLR-Signalübertragung vermittelt wird. Transkriptomische Analysen nach Koinfektion von S. mitis Isolat 4 (SM4) und PA01 auf PCLS zeigten eine signifikante Runterregulation von Entzündungsreaktionen wie mTOR und Toll-like-Rezeptor-Signalen. Ein WGS-Vergleich zeigte, dass mehr als die Hälfte der am stärksten angereicherten Genfunktionen bei hemmenden Streptokokken-Isolaten für den Kohlenhydrat-Transport und -Stoffwechsel verantwortlich waren, während sie bei den nicht hemmenden Streptokokken-Isolaten unter den am stärksten angereicherten Genfunktionen fehlten. Mechanische Untersuchungen zeigten, dass der glykolytische Signalweg für die antipseudomonische Wirkung entscheidend ist und dass hemmende Kommensalen hemmende Wirkungen vermitteln, indem sie den pH-Wert ihrer Wachstumsmedien < 5,0 senken und Acetat > 0,2 mg/ml produzieren. Es wurde nachgewiesen, dass Acetat signifikante immunmodulatorische Effekte gegen PA01- und LPS-induzierte Entzündungsreaktionen in BEAS-2B und PCLS vermittelt. Zusammenfassend lässt sich sagen, dass ausgewählte kommensale Bakterien Schutzwirkungen in den Atemwegen von Mukoviszidose-Patienten herbeiführen, indem sie Acetat produzieren, das antipseudomonale und immunmodulatorische Wirkungen vermittelt. Einserseits direkt, indem es durch Bakterien- und Wirtszellen diffundiert und so unmittelbare Auswirkungen hat, als auch indirekt, indem es Wirtszellen dazu anregt, Bakterien effizient zu beseitigen und Entzündungen zu kontrollieren. Da die Verwendung ganzer Bakterien als Probiotika bei immungeschwächten Patienten beispielsweise bei Mukoviszidose mit einigen Herausforderungen verbunden ist, stellt die Verwendung von bakteriellen Metaboliten wie Acetat eine sicherere, einfachere und praktischere Alternative dar.:List of Abbreviations (i) Table of Contents (iv) 1. SUMMARY (1) 1.1 Zusammenfassung (1) 1.2 ABSTRACT (3) 2. INTRODUCTION (5) 2.1 Cystic fibrosis (5) 2.2 Development of the CF lung pathology (6) 2.3 The immune response (8) 2.3.1 Innate and adaptive immunity (8) 2.3.2 Toll-like receptors (TLRs) (9) 2.4 Inflammation in CF (11) 2.4.1 Neutrophils in CF (11) 2.4.2 Macrophages in CF (12) 2.4.3 Eicosanoid metabolites in CF (12) 2.4.4 Chemokines in CF (12) 2.5 Airway sampling for microbiome studies (13) 2.6 CF airway microbiome (14) 2.6.1 The healthy lung microbiome (14) 2.6.2 Pathogenic bacterial members of the CF microbiome and pulmonary exacerbations (15) 2.6.3 Pseudomonas aeruginosa in CF (16) 2.6.4 Anaerobic CF microbiota (17) 2.6.5 Fungal CF microbiota (17) 2.6.6 Virus CF microbiota (18) 2.6.7 Commensal-pathogen interactions in CF (18) 2.7 CFTR modulators (18) 2.8 Human epithelial cell lines and murine precision-cut lung slices (PCLS) as in vitro model systems (19) 2.9 Next-generation sequencing (NGS) in CF microbiome studies (20) 2.10 Objectives of this study (21) 3. MATERIALS AND METHODS (23) 3.1 Materials (23) 3.1.1 Devices and Instruments (23) 3.1.2 Software (24) 3.1.3 Consumables (25) 3.1.4 Chemicals, Reagents, Media, and Antibiotics (26) 3.1.5 Kits (29) 3.1.6 Buffers, Media, and Solutions (30) 3.1.7 qPCR Primers (32) 3.1.8 Cell lines (33) 3.1.9 Mouse strains (33) 3.1.10 Bacteria isolates (34) 3.2 Methods (37) 3.2.1 Isolation, identification, and storage of isolates (37) 3.2.2 Pathogens-Commensals direct cocultures (38) 3.2.3 HPLC of conditioned media from bacterial isolates (40) 3.2.4 Cell-Pathogen-Commensal cocultures (41) 3.2.5 PCLS cocultures (42) 3.2.6 RNA extraction, cDNA preparation, and quantitative RT-PCR (43) 3.2.7 RNA Sequencing and Transcriptome analysis (45) 3.2.8 Bacteria DNA extraction and Whole Genome Sequencing (46) 3.2.9 Biochemistry (47) 3.2.10 Statistical analyses (49) 4. RESULTS (50) 4.1 Analysis of direct commensal-pathogen interactions (50) 4.1.1 Several streptococcal isolates inhibit the growth of P. aeruginosa with inter- and intra-species variability in the antipseudomonal effect (51) 4.1.2 Further commensal isolates that do not inhibit the growth of P. aeruginosa (54) 4.1.3 The lack of antipseudomonal effect by noninhibitory isolates is not due to insufficient cell numbers (54) 4.1.4 Fungal CF isolates in this study do not possess antipseudomonal effects (56) 4.1.5 SCAPEs (Selected Commensals with strong Anti-Pseudomonal Effects) also inhibit other P. aeruginosa strains (58) 4.1.6 SCAPEs inhibit other non-pseudomonal pathogenic CF isolates (60) 4.1.7 Inhibitory effects mediated by SCAPEs do not extend to the fungal CF isolates in this study (63) 4.2 Analysis of commensal-host-pathogen interactions using human bronchial epithelial cell lines (63) 4.2.1 Some commensal isolates are able to modulate PA01-induced IL-8 release in BEAS-2B cells (64) 4.2.2 Commensal-mediated IL-8 modulation in BEAS-2B cells is not due to PA01 growth inhibition (67) 4.2.3 Selected commensal isolates also modulate LPS-induced IL-8 release in BEAS-2B cells (68) 4.2.4 Selected S. mitis isolates also modulate IL-8 release in BEAS-2B cells induced by other CF P. aeruginosa isolates (68) 4.2.5 Selected commensal isolates modulate PA01-induced IL-8 release in CFBE41o cells (70) 4.2.6 Protective commensals need to be metabolically active to exert immunomodulatory effects (72) 4.2.7 Hydrogen peroxide produced by peroxide-producing Streptococcus spp. affects the viability of human bronchial epithelial cells (72) 4.2.8 Selected peroxide-producing Streptococcus spp. possess immunomodulatory activity when peroxide-induced cell death is prevented (75) 4.3 Analysis of commensal-host-pathogen interactions using mouse PCLS (80) 4.3.1 PCLS is more resilient against peroxide-induced loss of viability (80) 4.3.2 Selected S. mitis isolates modulate PA01-induced inflammatory response in mouse PCLS (82) 4.3.3 Immunomodulation of PA01-induced response by SM4 in PCLS is not due to active PA01 growth inhibition (84) 4.4 Analysis of the underlying mechanisms behind the streptococcal-mediated effects via transcriptome and whole genome sequencing (84) 4.4.1 Transcriptomic analyses show that SM4 downregulates signalling pathways involved in PA01-induced inflammatory responses in mouse PCLS (84) 4.4.2 Whole genome sequence comparison shows that in inhibitory commensals, most of their genes are involved in carbohydrate transport and metabolism (87) 4.5 Uncovering the mechanisms behind the observed streptococcal-mediated antipseudomonal effects (89) 4.5.1 Conditioned medium (CM) from SCAPEs inhibits the growth of P. aeruginosa and other typical CF pathogens (89) 4.5.2 Inhibitory activity of SCAPEs CM is neither heat sensitive nor proteinaceous (91) 4.5.3 Iron competition and the arginolytic pathway are not responsible for the observed inhibitory effects (91) 4.5.4 Peroxide production may contribute but does not play a major role in the antipseudomonal effects (94) 4.5.5 Several members of Streptococcus spp. mediate antipseudomonal effects via the glycolytic pathway (94) 4.5.6 Low pH plays a major role in the observed inhibition (97) 4.5.7 SCAPEs and other selected commensal isolates can mediate antipseudomonal effects by simultaneously lowering the pH and secreting acetate (98) 4.5.8 Extracellular addition of 0.5 mg/ml acetate at pH 5.0 inhibits the growth of P. aeruginosa (100) 4.5.9 Other SCFAs like propionate and butyrate at pH 5.0 also inhibit P. aeruginosa isolates (102) 4.5.10 Acetate has better antipseudomonal activity than propionate and butyrate (103) 4.6 Commensals may mediate their protective effects via acetate production (104) 4.6.1 SCFAs modulate PA01- and LPS-induced IL-8 release in BEAS-2B cells (104) 4.6.2 SCFA levels used are well below cell toxicity levels (105) 4.6.3 Acetate modulates PA01 and LPS-induced immune response in mouse PCLS (107) 5. DISCUSSION (110) 5.1 SCAPEs mediate inhibitory effects in direct commensal-pathogen interactions against P. aeruginosa and other typical CF pathogens (110) 5.1.1 Members of Streptococcus spp. mediate inter- and intra-species variability in their antipseudomonal effects (111) 5.1.2 SCAPEs inhibit other clinical and nonclinical P. aeruginosa strains as well as other typical CF pathogens (113) 5.2 Selected commensals modulate PA01- and LPS-induced inflammatory response in human airway epithelial cells and mouse PCLS (115) 5.2.1 The gram-positive commensal isolates in this study do not significantly stimulate inflammatory response in human bronchial epithelial cells and mouse PCLS (115) 5.2.2 Selected gram-positive commensal isolates modulate P. aeruginosa-triggered inflammatory response in BEAS-2B cells with inter- and intra-species variation (117) 5.2.3 Selected commensal isolates modulate P. aeruginosa-triggered inflammatory response in CFBE41o ΔF508 (120) 5.2.4 Selected S. mitis isolates modulate P. aeruginosa-induced inflammatory response in mouse PCLS (121) 5.3 Commensals exert protective effects against P. aeruginosa infection via acetate production (124) 5.3.1 Conditioned medium (CM) from selected commensal isolates need to be acidic to mediate inhibition of growth of P. aeruginosa and other typical CF pathogens (125) 5.3.2 The glycolytic pathway is important for streptococcal-mediated antipseudomonal effects (127) 5.3.3 Commensal bacteria mediate growth inhibitory effects by simultaneously lowering the pH and producing acetate (128) 5.3.4 Acetate modulates PA01- and LPS-induced inflammation in bronchial epithelial cells and PCLS (131) 5.4 Conclusions and Outlook (134) 6. DECLARATIONS (158) 6.1 Statement of Authorship (158) 6.2 Declaration of compliance (160) 7. Acknowledgements (161) / There is no known lung disease that causes such a very early, chronic, and intense inflammatory reaction as seen in the airways of patients with cystic fibrosis (CF). CF is the most common lethal autosomal recessive genetic condition in the Caucasian population caused by a mutation in the cystic fibrosis transmembrane conductance regulator (CFTR) gene encoding for the CFTR protein. Defects in this protein result into epithelial dysfunction and affect several organs but lung pathology accounts for over 85% of CF morbidity and mortality. The CF lung pathology centers on the host-pathogen interactions where CFTR dysfunction predisposes to infections and the infections coupled with a dysfunctional immune response drive a sustained inflammatory cycle. This vicious cycle of infection and inflammation ultimately results in lung damage, respiratory failure and then, death. The most predominant infection in CF is by P. aeruginosa with an overall European average of 41.0% of adult patients infected. Of note, airway microbiome in CF is of polymicrobial nature. Given that previous studies have established positive correlations between a high microbiome diversity and improved lung function in CF, it was hypothesized that certain commensals may be protective against P. aeruginosa infection in the CF airways. Therefore, 105 commensal isolates from 32 different species were isolated from sputum samples of patients with CF and screened for antagonistic effects in direct pathogen-commensal interactions using a fluorescent P. aeruginosa PA01-mcherry strain. These isolates were also screened for immunomodulatory effects in commensal-host-pathogen interactions using BEAS-2B bronchial epithelial cell lines. Selected isolates with protective effects were subsequently screened for immunomodulatory effects using CFBE41o ΔF508 cells and a more natural precision-cut-lung-slices (PCLS) model and the produced cytokines were measured via ELISA as well as via a multiplex cytokine assay. Gene expression analyses were also conducted via qRT-PCR. Underlying mechanisms were explored via transcriptomic analyses, whole genome sequence (WGS) comparisons, and mechanistic studies including metabolic analyses via high-performance liquid chromatography. It was demonstrated that selected streptococcal commensal isolates, especially members belonging to S. mitis, S. oralis, S. cristatus, S. gordonii, S. sanguinis, and S. parasanguinis, mediate potent antagonistic effects against the growth of P. aeruginosa in direct cocultures. Selected members from S. mitis, S. oralis, and S. cristatus also prevented the growth of other P. aeruginosa clinical and nonclinical isolates as well as other typical CF pathogens including Staphylococcus aureus, Burkholderia spp., Achromobacter xylosoxidans, Proteus mirabilis, Haemophilus influenzae, Stenotrophomonas maltophilia, Enterococcus faecalis, and Klebsiella pneumoniae. An effective CF therapy should not only address infection but the accompanying vicious inflammation as well. Unlike the members of the gram-negative Neisseria spp. which significantly stimulated IL-8 production in mono-infection, all the gram-positive commensal isolates did not. Selected commensals also modulated P. aeruginosa PA01- and LPS-induced production of several inflammatory cytokines in human airway epithelial cells (BEAS-2B as well as CFBE41o ΔF508 and corrected wtCFTR) and in mouse PCLS. These findings were also confirmed via gene expression analyses indicating that the immunomodulation may be mediated by altered TLR signalling. Transcriptomic analyses after co-infection of S. mitis isolate 4 (SM4) and PA01 on PCLS revealed a significant downregulation of inflammatory responses such as mTOR and toll-like receptor signalling. WGS comparison showed that over half of the most enriched gene functions in inhibitory streptococcal isolates were responsible for carbohydrate transport and metabolism but were absent among the most enriched gene functions for the noninhibitory streptococcal isolates. Mechanistic investigations demonstrated that the glycolytic pathway was essential for antipseudomonal effects and that inhibitory commensals mediate inhibitory effects by lowering the pH of their growth media < 5.0 and producing acetate > 0.2 mg/ml. Acetate was demonstrated to mediate significant immunomodulatory effects against PA01- and LPS-induced inflammatory response in BEAS-2B and PCLS. In conclusion, selected commensal bacteria induce protective effects in the CF airway by producing acetate, which mediates antipseudomonal and immmunomodulatory activities both directly by diffusing across bacterial and host cells to mediate direct effects as well as indirectly by stimulating host cells to clear bacteria efficiently and control inflammation. Given that the use of whole bacteria as probiotics in immunocompromised patients like in CF possesses several challenges, the use of bacterial metabolites like acetate presents a safer, easier, and more practical alternative.:List of Abbreviations (i) Table of Contents (iv) 1. SUMMARY (1) 1.1 Zusammenfassung (1) 1.2 ABSTRACT (3) 2. INTRODUCTION (5) 2.1 Cystic fibrosis (5) 2.2 Development of the CF lung pathology (6) 2.3 The immune response (8) 2.3.1 Innate and adaptive immunity (8) 2.3.2 Toll-like receptors (TLRs) (9) 2.4 Inflammation in CF (11) 2.4.1 Neutrophils in CF (11) 2.4.2 Macrophages in CF (12) 2.4.3 Eicosanoid metabolites in CF (12) 2.4.4 Chemokines in CF (12) 2.5 Airway sampling for microbiome studies (13) 2.6 CF airway microbiome (14) 2.6.1 The healthy lung microbiome (14) 2.6.2 Pathogenic bacterial members of the CF microbiome and pulmonary exacerbations (15) 2.6.3 Pseudomonas aeruginosa in CF (16) 2.6.4 Anaerobic CF microbiota (17) 2.6.5 Fungal CF microbiota (17) 2.6.6 Virus CF microbiota (18) 2.6.7 Commensal-pathogen interactions in CF (18) 2.7 CFTR modulators (18) 2.8 Human epithelial cell lines and murine precision-cut lung slices (PCLS) as in vitro model systems (19) 2.9 Next-generation sequencing (NGS) in CF microbiome studies (20) 2.10 Objectives of this study (21) 3. MATERIALS AND METHODS (23) 3.1 Materials (23) 3.1.1 Devices and Instruments (23) 3.1.2 Software (24) 3.1.3 Consumables (25) 3.1.4 Chemicals, Reagents, Media, and Antibiotics (26) 3.1.5 Kits (29) 3.1.6 Buffers, Media, and Solutions (30) 3.1.7 qPCR Primers (32) 3.1.8 Cell lines (33) 3.1.9 Mouse strains (33) 3.1.10 Bacteria isolates (34) 3.2 Methods (37) 3.2.1 Isolation, identification, and storage of isolates (37) 3.2.2 Pathogens-Commensals direct cocultures (38) 3.2.3 HPLC of conditioned media from bacterial isolates (40) 3.2.4 Cell-Pathogen-Commensal cocultures (41) 3.2.5 PCLS cocultures (42) 3.2.6 RNA extraction, cDNA preparation, and quantitative RT-PCR (43) 3.2.7 RNA Sequencing and Transcriptome analysis (45) 3.2.8 Bacteria DNA extraction and Whole Genome Sequencing (46) 3.2.9 Biochemistry (47) 3.2.10 Statistical analyses (49) 4. RESULTS (50) 4.1 Analysis of direct commensal-pathogen interactions (50) 4.1.1 Several streptococcal isolates inhibit the growth of P. aeruginosa with inter- and intra-species variability in the antipseudomonal effect (51) 4.1.2 Further commensal isolates that do not inhibit the growth of P. aeruginosa (54) 4.1.3 The lack of antipseudomonal effect by noninhibitory isolates is not due to insufficient cell numbers (54) 4.1.4 Fungal CF isolates in this study do not possess antipseudomonal effects (56) 4.1.5 SCAPEs (Selected Commensals with strong Anti-Pseudomonal Effects) also inhibit other P. aeruginosa strains (58) 4.1.6 SCAPEs inhibit other non-pseudomonal pathogenic CF isolates (60) 4.1.7 Inhibitory effects mediated by SCAPEs do not extend to the fungal CF isolates in this study (63) 4.2 Analysis of commensal-host-pathogen interactions using human bronchial epithelial cell lines (63) 4.2.1 Some commensal isolates are able to modulate PA01-induced IL-8 release in BEAS-2B cells (64) 4.2.2 Commensal-mediated IL-8 modulation in BEAS-2B cells is not due to PA01 growth inhibition (67) 4.2.3 Selected commensal isolates also modulate LPS-induced IL-8 release in BEAS-2B cells (68) 4.2.4 Selected S. mitis isolates also modulate IL-8 release in BEAS-2B cells induced by other CF P. aeruginosa isolates (68) 4.2.5 Selected commensal isolates modulate PA01-induced IL-8 release in CFBE41o cells (70) 4.2.6 Protective commensals need to be metabolically active to exert immunomodulatory effects (72) 4.2.7 Hydrogen peroxide produced by peroxide-producing Streptococcus spp. affects the viability of human bronchial epithelial cells (72) 4.2.8 Selected peroxide-producing Streptococcus spp. possess immunomodulatory activity when peroxide-induced cell death is prevented (75) 4.3 Analysis of commensal-host-pathogen interactions using mouse PCLS (80) 4.3.1 PCLS is more resilient against peroxide-induced loss of viability (80) 4.3.2 Selected S. mitis isolates modulate PA01-induced inflammatory response in mouse PCLS (82) 4.3.3 Immunomodulation of PA01-induced response by SM4 in PCLS is not due to active PA01 growth inhibition (84) 4.4 Analysis of the underlying mechanisms behind the streptococcal-mediated effects via transcriptome and whole genome sequencing (84) 4.4.1 Transcriptomic analyses show that SM4 downregulates signalling pathways involved in PA01-induced inflammatory responses in mouse PCLS (84) 4.4.2 Whole genome sequence comparison shows that in inhibitory commensals, most of their genes are involved in carbohydrate transport and metabolism (87) 4.5 Uncovering the mechanisms behind the observed streptococcal-mediated antipseudomonal effects (89) 4.5.1 Conditioned medium (CM) from SCAPEs inhibits the growth of P. aeruginosa and other typical CF pathogens (89) 4.5.2 Inhibitory activity of SCAPEs CM is neither heat sensitive nor proteinaceous (91) 4.5.3 Iron competition and the arginolytic pathway are not responsible for the observed inhibitory effects (91) 4.5.4 Peroxide production may contribute but does not play a major role in the antipseudomonal effects (94) 4.5.5 Several members of Streptococcus spp. mediate antipseudomonal effects via the glycolytic pathway (94) 4.5.6 Low pH plays a major role in the observed inhibition (97) 4.5.7 SCAPEs and other selected commensal isolates can mediate antipseudomonal effects by simultaneously lowering the pH and secreting acetate (98) 4.5.8 Extracellular addition of 0.5 mg/ml acetate at pH 5.0 inhibits the growth of P. aeruginosa (100) 4.5.9 Other SCFAs like propionate and butyrate at pH 5.0 also inhibit P. aeruginosa isolates (102) 4.5.10 Acetate has better antipseudomonal activity than propionate and butyrate (103) 4.6 Commensals may mediate their protective effects via acetate production (104) 4.6.1 SCFAs modulate PA01- and LPS-induced IL-8 release in BEAS-2B cells (104) 4.6.2 SCFA levels used are well below cell toxicity levels (105) 4.6.3 Acetate modulates PA01 and LPS-induced immune response in mouse PCLS (107) 5. DISCUSSION (110) 5.1 SCAPEs mediate inhibitory effects in direct commensal-pathogen interactions against P. aeruginosa and other typical CF pathogens (110) 5.1.1 Members of Streptococcus spp. mediate inter- and intra-species variability in their antipseudomonal effects (111) 5.1.2 SCAPEs inhibit other clinical and nonclinical P. aeruginosa strains as well as other typical CF pathogens (113) 5.2 Selected commensals modulate PA01- and LPS-induced inflammatory response in human airway epithelial cells and mouse PCLS (115) 5.2.1 The gram-positive commensal isolates in this study do not significantly stimulate inflammatory response in human bronchial epithelial cells and mouse PCLS (115) 5.2.2 Selected gram-positive commensal isolates modulate P. aeruginosa-triggered inflammatory response in BEAS-2B cells with inter- and intra-species variation (117) 5.2.3 Selected commensal isolates modulate P. aeruginosa-triggered inflammatory response in CFBE41o ΔF508 (120) 5.2.4 Selected S. mitis isolates modulate P. aeruginosa-induced inflammatory response in mouse PCLS (121) 5.3 Commensals exert protective effects against P. aeruginosa infection via acetate production (124) 5.3.1 Conditioned medium (CM) from selected commensal isolates need to be acidic to mediate inhibition of growth of P. aeruginosa and other typical CF pathogens (125) 5.3.2 The glycolytic pathway is important for streptococcal-mediated antipseudomonal effects (127) 5.3.3 Commensal bacteria mediate growth inhibitory effects by simultaneously lowering the pH and producing acetate (128) 5.3.4 Acetate modulates PA01- and LPS-induced inflammation in bronchial epithelial cells and PCLS (131) 5.4 Conclusions and Outlook (134) 6. DECLARATIONS (158) 6.1 Statement of Authorship (158) 6.2 Declaration of compliance (160) 7. Acknowledgements (161)

The gut microbiota and its metabolites have vast impacts on the human digestive system, immune system, and health outcomes. Short chain volatile fatty acids (SCVFAs) present in feces can be representative of the interactions of the microbiota present in the gut. Low microbiota diversity in the human gut is highly associated with obesity and adverse health outcomes. Furthermore, the maternal microbiome has a direct impact on neonatal microbiota through various pathways such as environment, skin flora, breast milk composition, and vaginal secretions. This study is aimed to further understand the associations between various factors (maternal adiposity, gestational time, length of life, delivery mode, and race/ethnicity ) and neonatal microbiome and its metabolites, SCFA. Data (pre-pregnancy BMI, gestational time, length of life at time of sample collection, delivery mode, race/ethnicity, SCVFA profiles, fecal fermentation profiles, and 16s rRNA sequences, n=75) was obtained from 75 mother-infant dyads. Qiagen CLC Genomics Workbench was used to process 16s RNA data, generate quantitative and qualitative measures of alpha and beta diversity, and generate an analysis of the composition of microbiomes for differential abundances. Multiple metrics were analyzed for alpha and beta diversity and no significant differences were found for acetic acid (A), propionic acid (P), butyric acid (B), or APB combined. Shannon diversity index, a measure of Alpha diversity, showed no significant difference between groups in each subset. BMI differences were significant for no c-section vs. c-section and Black vs. White race/ethnicity. There were no significant differences found in PERMANOVA, a measure of beta diversity, or found in differential abundances among the groups.

Preterm birth is the number one cause of death in neonates, accounting for 35% of neonatal mortality. The preterm birth rate in the U.S. in 2021 was 10.5%, disproportionally affected by race and ethnicity. Obese women have an increased risk of preterm pregnancy, and if delivered before 37 weeks of gestation, the offspring have higher rates of complications that extend from the neonatal period into life-long metabolic and immune adverse outcomes. In the early months after delivery, preterm infants have higher rates of adverse health outcomes than term infants, including infections, extrauterine growth restrictions, respiratory, metabolic, and neurological complications, necrotizing enterocolitis (NEC), and bronchopulmonary dysplasia (BPD). Diet for the preterm infant is crucial for infection prevention, and maternal breast milk is most beneficial when given in the first few days after birth. Expectant and breastfeeding mothers should consume appropriate food and supplements to optimize their nutrition. In addition to nutrients, bioactive components, vitamins, and minerals found in breast milk (BM), there is evidence that microbes (microbiome) are a significant factor in infant development, contributing to protection against pathogens and playing a role in the development of the immune and nervous systems. Maternal BM composition and microbiome are affected by many factors, including maternal body mass index (BMI), maternal health, antibiotics use, mode of delivery, maternal parity, gestational age, and time and duration of lactation. Maternal body composition, however, and not maternal nutritional status, is associated with breast milk nutritional composition. Altogether, these maternal factors may modify the premature infant gut microbiome. We examined the role of maternal adiposity and how it impacts the composition of human breast milk, specifically hormones, nutrient composition, short and long chain fatty acids, and microbiome. We also examined the role of maternal adiposity and how it impacts the short-chain fatty acids and microbiome in infant fecal samples. We found that maternal adiposity affects breast milk hormones, potentially modulating infant metabolism. Additionally, we found that maternal adiposity does not alter the nutrient composition of breast milk; however, differences in both short and long chain fatty acids and maternal adiposity were detected. In our small cross-sectional cohort of preterm infants, we did not observe differences in short-chain fatty acids in the preterm infant stool samples compared to maternal adiposity. Concerning maternal adiposity and its impact on the microbiomes of breast milk and infants, we observed differences in phyla and genera between the maternal BMI groups on the outcome of breast milk and preterm infant microbiomes but no statistical significance in alpha and beta diversities between the groups. Thus, our results indicate that maternal adiposity impacts hormonal, microbial composition, and short-chain fatty acid profiles in breast milk, which tremendously influences infant growth and development.

The large intestinal epithelium is confronted with the necessity to adapt quickly to varying levels of oxygenation. In contrast to other tissues, it meets this requirement successfully and remains unharmed during (limited) hypoxic periods. The large intestine is also the site of bacterial fermentation producing short-chain fatty acids (SCFA). Amongst these SCFA, butyrate has been reported to ameliorate many pathological conditions. Thus, we hypothesized that butyrate protects the colonocytes from hypoxic damage. We used isolated porcine colon epithelium mounted in Ussing chambers, incubated it with or without butyrate and simulated hypoxia by changing the gassing regime to test this hypothesis. We found an increase in transepithelial conductance and a decrease in short-circuit current across the epithelia when simulating hypoxia for more than 30 min. Incubation with 50 mM butyrate significantly ameliorated these changes to the epithelial integrity. In order to characterize the protective mechanism, we compared the effects of butyrate to those of iso-butyrate and propionate. These two SCFAs exerted similar effects to butyrate. Therefore, we propose that the protective effect of butyrate on colon epithelium under hypoxia is not (only) based on its nutritive function, but rather on the intracellular signaling effects of SCFA.