Au cours des dernières décennies, le régime alimentaire a subi une transition sans précédent, avec une augmentation de la consommation de protéines, de lipides et de glucides simples, et la diminution des apports en fibres. Par ailleurs, au-delà de la composition, la structure des aliments joue un rôle essentiel sur les cinétiques de digestibilité et la biodisponibilité des nutriments, modulant ainsi leur accessibilité pour microbiote dans le côlon. L’impact de la structure d’une matrice alimentaire complexe, formée de protéines et de lipides, sur le microbiote a été analysé de façon intégrée et a montré in vivo que la structure seule, dans le contexte d’un régime équilibré, pouvait altérer la composition du microbiote dans les zones distales et proximales que sont l’iléon et le cæcum. L’émulsion de protéines natives en phase liquide continue avec de fines gouttelettes protéolipidiques a arboré des protéines moins digestibles que l’émulsion de protéines dénaturées, en phase gélifiée, solide, avec de grandes gouttelettes. D’autre part, les lipides de l’émulsion solide étaient, à l’inverse, moins digestibles. Les protéines non digérées de l’émulsion liquide ont favorisé in vivo, les communautés de Lactobacillus et de Copprococcus tout en activant plus fortement les métabolismes de protéolyse. Inversement, les communautés de Bifidobacterium et d’Akkermansia muciniphila ont vu leurs abondances augmenter chez les rats consommant l’émulsion solide. Le deuxième objectif de ce travail de thèse a alors été d'analyser la capacité d'espèces prévalentes du microbiote intestinal humain à métaboliser des protéines non digérées. Nous avons montré, par le suivi des cinétiques de croissance et des productions de métabolites spécifiques, que les protéines du lait étaient une source d'énergie pour B. caccae, P. distasonis, B. longum et B. cocccoides en milieu pauvre ainsi qu'en milieu riche. Dans ces mêmes conditions, le transcriptome de B. caccae a montré la sur-expression de gènes codant pour des peptidases de specifités différentes, pour la production d'indoles, de GABA et de fimbriae. Ces travaux apportent des informations nouvelles sur l'impact de la structure sur l'écosystème digestif, et ouvre des portes pour le développement de nouveaux aliments. / Over the past decades, diet in developed countries has undergone an unprecedented transition, with increased intakes of protein, fat and high glycemic index carbohydrates. The first goal of this PhD work was to investigate how, beyond its composition, the food structure itself could play a part in nutrient digestibility and bioavailability, and consequently modulate the microbiota. We showed in vivo that the structure of proteino-lipidic emulsions modulated peptides transporters, and protein fermentation. The native proteins emulsion in a continuous liquid phase, with fine proteolipid droplets, was less digestible and led to more protein fermentation. It modified the gut microbiota composition in the distal and proximal intestinal sections and increased Lactobacillus and Coprococcus communities. A second in vivo study, using 15N labelled emulsions allowed us to disentangle the digestibility from the transit time effect. The second objective of the PhD was to characterize the capacity of prevalent human gut bacterial species to use undigested proteins as energy source. By monitoring growth kinetics and the production of specific metabolites, we showed that B. caccae, P. distasonis, B. longum et B. cocccoides could use whey protein as energy source. In addition we measured in B. caccae transcriptome, the over-expression of genes encoding for distinct peptidases, but also of GABA and indole pathways, and fimbriae biosynthesis. These data provide new insights on the relationships between food structure and the digestive ecosystem and could lead to the design of new functional food.

Résumé : Les régimes hyperprotéiques sont couramment consommés mais les conséquences au niveau du gros intestin sont peu connues. L’objectif de la thèse était d’étudier les effets des régimes hyperprotéiques et des métabolites bactériens dérivés des acides aminés sur la muqueuse du gros intestinUne série d’expérimentations animales et in vitro a permis de montrer que deux métabolites bactériens dérivés des acides aminés (le sulfure d’hydrogène et le p-cresol) sont toxiques pour l’épithélium lorsqu’ils sont présents en concentration élevée. Les résultats obtenus lors d’une étude clinique montrent que la quantité et la qualité des protéines alimentaires n’ont pas d’effets marqués sur la composition du microbiote fécal mais modifient les concentrations fécales et urinaires en métabolites bactériens.Ces modifications de l’environnement luminal du gros intestin n’étaient pas associées à une augmentation de la cytotoxicité des eaux fécales in vitro. Néanmoins, dans la muqueuse rectale, l’augmentation de l’apport en protéines a régulé l’expression de gènes impliqués dans le maintien de l’homéostasie et ces effets étaient distincts en fonction de la source de protéines utilisée. Toutefois, le niveau d’apport en protéines n’avait pas d’effet sur les paramètres inflammatoires et histologiques dans la muqueuse. Ces résultats ont été complétés par une étude chez le rat montrant qu’un régime hyperprotéique modifie le transcriptome dans les colonocytes mais n’a pas d’effets délétères en termes d’intégrité de l’ADN, de renouvellement de l’épithélium et de fonction barrière. / Abstract: High-protein diets are frequently consumed but the consequences for the large intestine are not well described. The objective of this thesis was to evaluate the effects of high-protein diets and of amino-acid derived bacterial metabolites on the large intestine mucosa. Animal and in vitro studies showed that two amino acid derived bacterial metabolites (hydrogen sulfide and p-cresol) are toxic for the epithelium when present at high concentration. The results obtained in a clinical trial indicate that quantity and quality of dietary protein do not have major effects on the fecal microbiota composition but modify the fecal and urinary concentration of bacterial metabolites.These changes in luminal environment were not associated with an increase in fecal water cytotoxicity in vitro. Nevertheless, in the rectal mucosa, the increase in protein intake regulated the expression of genes implicated in homeostatic processes and these effects were modulated by the source of protein. However, the level of protein intake had no effect on immune and histological parameters in the mucosa. These results were completed with a study in rats showing a clear transcriptome profile in colonocytes induced by a high-protein diet but that was not associated with detrimental effects in terms of DNA integrity, epithelial renewal and barrier function.

De plus en plus d’évidences suggèrent que la composition du microbiote intestinal pourrait jouer un rôle dans certaines pathologies comme l’hypertension, l’obésité, le diabète et plusieurs autres. Le but de cette étude est de démontrer que le microbiote intestinal modulé par des diètes enrichies en acide gras oméga-3 (w-3) ou en acide gras oméga-6 (w-6) avec un supplément ou non en probiotiques, pourrait influencer différemment la taille de l’infarctus du myocarde. Pour ce faire, des transplantations de microbiote provenant de rats nourris avec une diète w-3 ou w-6 combiné avec la prise ou non probiotiques ont été effectuées pendant 10 jours sur des rats dont le microbiote a été supprimé antérieurement par antibiothérapie. Ensuite, l’artère coronaire antérieure a été occluse pendant 30 minutes sur les rats transplantés et la taille de leur infarctus a été mesurée après 24 heures de reperfusion. La résistance intestinale, la concentration plasmatique de LPS (lipopolysaccharides), l’accumulation myocardique des neutrophiles, l’activation de la voie NF-kB (nuclear factor-kappa B), l’activation de la voie de cardioprotection RISK (Reperfusion Injury Salvage Kinase) et la composition du microbiote sont les autres paramètres qui ont été mesurés. Nos résultats démontrent une taille d’infarctus plus importante chez les animaux transplantés avec le microbiote w-6 sans probiotiques comparativement aux autres groupes incluant le groupe ayant reçu le microbiote w-6 avec probiotiques. Ces résultats indiquent qu’un microbiote provenant d’une diète enrichie en w-6 produit des effets délétères qui augmentent la taille de l’infarctus du myocarde comparativement à un microbiote provenant d’une diète enrichie en w-3 et que la prise de probiotiques permet de diminuer la taille de l’infarctus. / A lot of evidences suggest that the composition of the gut microbiota may play a role in certain conditions such as hypertension, obesity, diabetes and many others. The aim of our experiment is to demonstrate that the intestinal microbiota modulated by diets enriched in omega-3 fatty acid (w-3) or in omega-6 fatty acid (w-6) with or without a supplement in probiotics, could influence differently the size of a myocardial infarction. To do this, microbiota transplants from rats fed a w-3 or w-6 diet combined or not with intake of probiotics were performed for 10 days on rats whose microbiota had previously been suppressed by antibiotic therapy. Then, the anterior coronary artery was occluded for 30 minutes in the transplanted rats and the size of their infarction was measured after 24 hours of reperfusion. Intestinal resistance, plasma LPS (lipopolysaccharides) concentration, myocardial accumulation of neutrophils, activation of the NF-kB (nuclear factor-kappa B) pathway, activation of the RISK (Reperfusion Injury Salvage Kinase) cardioprotection pathway and microbiota composition are the other parameters that were measured. Our results demonstrate a larger infarct size in animals transplanted with the w-6 microbiota without probiotics compared to other groups (including the w-6 microbiota with probiotics group. These results indicate that a microbiota from a diet enriched in w-6 produces deleterious effects that increase the size of the myocardial infarction compared to a microbiota from a diet enriched in w-3 and taking probiotics can reduce the size of the myocardial infarction.

The gut microbiota plays a major part in maintaining the health of a human host. Countless of crucial functions in the body, including immune responses, cell signaling and energy metabolism to name a few, are conducted by the gut microbiota and its metabolites. Accordingly, it is of interest to gain knowledge on what can alter the gut microbiota, as these alterations by extension can give rise to adverse health effects. In this study, the impact of polychlorinated biphenyl (PCB)- and organochlorine pesticide (OCP) exposure on tricarboxylic acid (TCA) cycle metabolites, short-chain fatty acids (SCFAs) and bile acids, as well as other polar and semi-polar metabolites, which are all related to the gut microbiota, were investigated. An in vitro fermentation of faecal samples exposed to a PCB/OCP mixture was performed, and liquid chromatography-time of flight mass spectrometry (LC-qToF-MS) targeted and non-targeted approaches were applied to the extracts. The results obtained suggested that PCBs and OCPs most likely have an effect on the levels of several features of the gut metabolome with either increased or decreased levels upon exposure. Bile acids and TCA metabolites appear to follow a trend of decreasing levels, while no apparent effects could be seen for the SCFAs. Furthermore, distinct concentrations of the PCB/OCP mixture appear to induce different changes in gut microbiota functioning, which highlights the importance of performing dose-response studies when exploring biological effects of these compounds. The identification of different metabolite profiles during fermentation also allows for the possibility of further investigation of potential biomarkers to assess PCB/OCP exposure.

While consumer awareness of benefits of adequate nutrition has noticeably surged in recent years, developing countermeasures against improper eating habits still represents a public health priority in view of the growing prevalence of diet-related diseases. Eating behaviours are complex phenomena driven by a spectrum of biological and environmental factors, wherein (chemo)sensory perception is reckoned amongst the most influential. Analogously, chemosensation is affected by a myriad of determinants, and this warrants the commonly observed large variation in how tastes and smells are perceived among individuals. Given how such variability intimately relates to dietary habits, deciphering its underlying mechanisms is paramount to promoting healthier food choices. In this vein, emerging evidence suggests that human eating behaviours can also be affected by interactions between the gastrointestinal microbiota and the chemosensory systems. Despite growing interest, the sensory-oriented microbiome field suffers from obvious limitations due to its recent emergence. As a result, little efforts has been devoted to elucidating: a) the associations between the oral microbiota and olfaction or known psychological mediators of sensory perception; b) the links between the distal gut microbiota and taste functioning; c) the consequences of interactions between chemosensation and the gastrointestinal microbiota on dietary intakes. Against this backdrop, this thesis aimed at expanding the current knowledge on the interplays between domains of sensory perception and the gastrointestinal microbiota and how these might mirror variations in habitual food habits. In detail, four studies probing the associations a) between a psychosocial correlate of sensory perception (food neophobia), olfaction (Chapter 2) and the oral microbiota (Chapter 3); and b) between distal gut (Chapter 4) or oral (Chapter 5) microbiota, taste functioning and dietary intakes are here presented. In Chapter 2 and 3, a healthy cohort of 83 individuals (57.8 % women; aged 22-68 yo) remotely filled out the common Food Neophobia Scale and the trait anxiety subscale of the State-Trait Anxiety Inventory prior to providing a salivary sample for subsequent metataxonomic analysis (16S rRNA gene sequencing). Next, volunteers were tested for orthonasal olfactory functioning via the Sniffin’ Sticks battery, and monitored for retronasal aroma release while consuming a strawberry jelly candy by nose-space analysis (Selected-Ion Flow-Tube Mass Spectrometry). In Chapter 4 and 5, instead, 100 young adult volunteers (52 % women; aged 18-30 yo) attended a 7-day lasting remote protocol where responsiveness to genetically-mediated bitterness of 6-n-propylithiuracil (PROP), hedonics and intensity of oral sensations elicited by ten commercially-available food products, a battery of food-related psychological traits, a 4-day food record, and one salivary and one stool sample (sequenced by targeting the 16s rRNA gene) were collected. Overall, results substantially strengthen past evidence suggesting: a) that pronounced neophobic tendencies translate into higher levels of (negative) emotional activation or arousal towards foods; b) the existence of homogenous groups of individuals with generalized hypergeusia towards oral stimulations; c) that hyperresponsiveness to a peculiar taste quality is a barrier to the intake of foods evoking such sensation; d) that habitual consumption of dietary fibers and simple carbohydrates can shape both the gut and oral microbial ecology, respectively. Intriguingly, food neophobia and poor olfaction were positively associated with oral microbial markers of dysbiosis (e.g., Porphyromonas gingivalis), whilst a Clostridia-enriched salivary microbiota co-occurred with low responsiveness to alarming oral sensations (astringency, bitter, sour) elicited by real foods. Similarly, an ample panel of commensal gut bacterial genera mainly allocated to the families Lachnospiraceae and Ruminococcaceae was found to be enriched in individuals exhibiting lower acuity to both tastes (bitter, salty, sour, sweet) and trigeminal sensations (astringent, pungent). Besides taxonomically annotating a range of microbial taxa tied to sensory perception, putative metabolic pathways used by salivary and gut microbial communities to modulate taste perception were inferred and discussed. To conclude, this thesis supports the notion that the gastrointestinal microbiota is an additional candidate to explain interindividual variations in taste and smell perception, and provides novel important insights into the aetiology of eating behaviours. More importantly, this work also offers methodological cues to robustly assess the associations between chemosensation and host-related non genetic factors, and paves the way for future interventional studies targeting the efficacy of sensory-related microbial taxa as potential modulators of dietary habits.

No description available.

No description available.

No description available.

Changes in the human gastrointestinal microbiome are associated with several diseases. To infer causality, experiments in representative models are essential, but widely used animal models exhibit limitations. Here we present a modular, microfluidics-based model (HuMiX, human-microbial crosstalk), which allows co-culture of human and microbial cells under conditions representative of the gastrointestinal human-microbe interface. We demonstrate the ability of HuMiX to recapitulate in vivo transcriptional, metabolic and immunological responses in human intestinal epithelial cells following their co-culture with the commensal Lactobacillus rhamnosus GG (LGG) grown under anaerobic conditions. In addition, we show that the co-culture of human epithelial cells with the obligate anaerobe Bacteroides caccae and LGG results in a transcriptional response, which is distinct from that of a co-culture solely comprising LGG. HuMiX facilitates investigations of host-microbe molecular interactions and provides insights into a range of fundamental research questions linking the gastrointestinal microbiome to human health and disease.

The present study aimed to develop a fermented soy beverage containing fruit by-products and probiotics and to evaluate the impact of this product on the composition and metabolic activity of the human intestinal microbiota using an in vitro simulation model of the intestinal conditions (TIM-2). Therefore, the present study was divided into three stages. Stage I was based on obtaining, processing and physical-chemical, microbiological and functional characterization of fruit by-products (acerola, orange, mango, and passion fruit) and soybean (okara), as well as amaranth flour. Additionally, the ability to use these vegetable by-products and amaranth flour by probiotic and non-probiotic strains was evaluated. The results showed that the acerola byproduct presented the highest dietary fibre content (48.46 g/100 g) among the by-products tested, as well as amaranth flour. Orange and passion fruit by-products were the substrates that most promoted the growth of bacterial populations, including strains of Escherichia coli and Clostridium perfringens. On the other hand, the acerola by-product was the substrate that showed the highest selectivity for beneficial bacteria. Also, in this stage, ten probiotic strains (seven lactobacilli and three bifidobacteria) and three starter strains (Streptococcus thermophilus) were tested for their ability to deconjugate bile salts and for proteolytic activity against milk and soy proteins. The results showed that none of the tested strain showed proteolytic ability against milk and soybean proteins. In addition, the probiotic strains Lactobacillus acidophilus LA-5 and Bifidobacterium longum BB-46 deconjugated more types of bile acids tested, and the strains of S. thermophilus tested showed no ability to deconjugate bile salts. Next, the acerola by-product (ABP) and the probiotic strains LA-5 and BB-46 were selected to continue stage II of the study (development of a fermented soy beverage). For this purpose, a 23 factorial design was used, in a total of 8 trials with three replicates of each one, and the effects of the probiotic strains and the acerola by-product on the physical-chemical, microbiological, and sensory characteristics of these fermented soy beverages were evaluated. At the same time, probiotic viability and survival under in vitro gastrointestinal (GI) simulated conditions were evaluated in fermented soy beverage (FSB). The results showed that the presence of BB-46 and ABP affected the sensory acceptability of FSB negatively. ABP also led to significant differences in the texture profile of the FSB (P<0.05). Populations of probiotic strains ranged from 7.0 to 8.2 log CFU equivalent/mL during 28 days of refrigerated storage (4° C) of FBS, and the co-culture (LA-5+BB-46) and the ABP did not affect the viability of both microorganisms significantly (P> 0.05). However, ABP increased the survival of BB-46 under in vitro simulated GI conditions significantly. For stage III, a 22 experimental design was performed. To evaluate the impact of these FBS on the composition and metabolic activity of the intestinal microbiota of lean and obese humans, a validated in vitro model called TIM-2 was used, available at the Maastricht University (Venlo, The Netherlands), which simulates normal conditions of the lumen of the proximal colon, with all parameters controlled by a computer. Samples were collected from TIM-2 to quantify probiotic microorganisms (LA-5 and BB-46), Lactobacillus spp., Bifidobacterium spp., and total bacteria, using the quantitative PCR method (qPCR) and the intestinal microbiota profile was determined using an Illumina Mysec Next Generation Sequencing (NGS) method. Concentrations of shortchain fatty acids and branched-chain fatty acids and lactate produced by the different microbiotas during fermentation in TIM-2 were also determined. The results showed that the lean microbiota presented the high production of acetate and lactate than the microbiota of obese individuals. Significant reductions in Bifidobacterium populations in the lean microbiota were observed at 0 and 48 h of an assay for all experimental meals, except for the meal that had the probiotic combination (LA-5 and BB-46) and the ABP supplementation, which showed an increased total Bifidobacterium and Lactobacillus populations throughout the experimental period for both microbiotas tested. The FSB supplemented with ABP presented the best characteristics regarding the modulation of the obese microbiota, with an increase in Bifidobacterium spp. and Lactobacillus spp. Additionally, after 48 hours of intervention in TIM-2, the obese microbiota was apparently similar to the lean microbiota, showing a beneficial modulation of this microbiota. The results suggest that the fermented soy beverage supplemented with the acerola by-product and the probiotic strains may present beneficial health effects. However, clinical studies are required to complement and confirm the results observed in the in vitro assays. / O presente trabalho visou desenvolver uma bebida fermentada de soja adicionada de resíduos de frutas e suplementada com cepas probióticas e avaliar o impacto desse produto sobre a composição e a atividade metabólica da microbiota intestinal humana, utilizando um modelo de simulação in vitro das condições intestinais (TIM-2). Para tanto, o presente trabalho foi dividido em três etapas. A etapa I foi baseada na obtenção, processamento e caracterização físico-química, microbiológica e funcional de subprodutos de frutas (acerola, laranja, manga e maracujá) e soja (okara), bem como da farinha de amaranto. Adicionalmente, a capacidade de utilização desses subprodutos vegetais e da farinha de amaranto por cepas probióticas e não probióticas foi avaliada. Os resultados mostraram que o subproduto de acerola apresentou o maior conteúdo de fibras alimentares totais (48,46 g/100 g) dentre os subprodutos testados, bem como a farinha de amaranto. Os subprodutos de laranja e maracujá foram os substratos que mais promoveram a multiplicação das populações bacterianas, incluindo das cepas de Escherichia coli e Clostridium perfringens. Por outro lado, o subproduto de acerola foi o substrato que apresentou a maior seletividade para bactérias benéficas. Ainda nessa etapa, dez cepas probióticas (sete lactobacilos e três bifidobacterias) e três cepas starter (Streptococcus thermophilus) foram testadas quanto à sua capacidade de desconjugação de sais biliares e atividade proteolítica frente às proteínas do leite e da soja. Os resultados revelaram que nenhuma cepa testada apresentou capacidade de proteólise das proteínas do leite e da soja. Adicionalmente, as cepas probióticas Lactobacillus acidophilus LA-5 e Bifidobacterium longum BB-46 desconjugaram a maior quantidade de ácidos biliares testados e as cepas de S. thermophilus testadas não apresentaram capacidade de desconjugação de sais biliares. Após a análise dos resultados da etapa I, o resíduo de acerola (ABP) e as cepas probióticas LA-5 e BB-46 foram selecionadas para dar continuidade à etapa II do estudo(desenvolvimento de uma bebida fermentada a base de soja). Para esse fim, foi utilizado um delineamento experimental do tipo fatorial 23, totalizando 8 ensaios com três repetições de cada, e foram avaliados os efeitos das cepas probióticas e do subproduto de acerola sobre as características físico-químicas, microbiológicas e sensoriais dessas bebidas fermentadas de soja. Paralelamente, foram realizadas análises da sobrevivência das cepas probióticas frente às condições gastrintestinais simuladas in vitro nas bebidas fermentadas de soja (FSB). Os resultados mostraram que a presença de BB-46 e ABP afetaram negativamente a aceitabilidade sensorial das FSB. O ABP também levou a diferenças significativas no perfil de textura das FSB (P<0,05). As populações das cepas probióticas nas diferentes formulações de FSB variaram de 7,0 a 8,2 log de UFC equivalente/mL durante os 28 dias de armazenamento (4 ºC) e a co-cultura (LA-5+BB-46) e o ABP não afetaram (P>0,05) a viabilidade de ambos os microrganismos. No entanto, ABP aumentou significativamente a sobrevivência de BB-46 frente às condições gastrintestinais sumuladas in vitro. Para a etapa III do presente estudo, um delineamento experimental fatorial 22 foi realizado. Para a avaliação do impacto dessas FSB sobre a composição e atividade metabólica da microbiota intestinal de humanos eutróficos e obesos, foi utilizado um modelo in vitro TIM-2 na Maastricht University (Venlo, Holanda), que simula as condições normais do lúmen do cólon proximal, com todos os parâmetros controlados por um computador. Amostras foram coletadas do TIM-2 para a quantificação dos microrganismos probióticos (LA-5 e BB-46), Lactobacillus spp., Bifidobacterium spp. e bactérias totais, utilizando o método de PCR quantitativo (qPCR), e o perfil da microbiota intestinal foi determinado utilizando Next-Generation Sequencing (NGS) Illumina Mysec. A concentração de ácidos graxos de cadeia curta e de cadeia ramificada e lactato produzidos pelas diferentes microbiotas durante a fermentação no TIM-2 também foi determinada. Os resultados mostraram que a microbiota de humanos eutróficos apresentou uma alta produção de acetato e lactato em comparação com a microbiota de obesos. Reduções significativas das populações de Bifidobacterium na microbiota de eutróficos foram observadas entre 0 e 48 h de ensaio para todas as refeições experimentais, exceto para a refeição que apresentou a combinação probiótica (LA-5 e BB-46) e a suplementação com ABP, que apresentou aumento de Bifidobacterium e Lactobacillus totais durante todo o período de análise para ambas as microbiotas testadas. As FSB suplementadas com ABP apresentaram os melhores resultados em relação à modulação da microbiota de humanos obesos, com o aumento Bifidobacterium spp. e Lactobacillus spp. Adicionalmente, após 48 horas de intervenção no TIM-2, a microbiota de obesos foi aparentemente similar à microbiota de eutróficos, mostrando uma modulação benéfica dessa microbiota. Os resultados sugerem que as bebidas fermentadas de soja suplementadas com o subproduto de acerola e cepas probióticas podem apresentar efeitos benéficos à saúde. No entanto, estudos clínicos são necessários para complementar e confirmar os resultados observados nos ensaios in vitro.