Kidney failure results in fluid and toxin accumulation within body fluid compartments, contributing to the excess mortality seen in this condition. Such uremic toxins have been measured in plasma, with levels assumed to reflect extraplasmatic concentrations such as in interstitial fluid (ISF). ISF is separated from plasma by nanometre-order microvascular pores; toxins may not circulate “freely” between the two compartments. This work set out to characterise the ISF in uremic subjects, with the hypothesis that there may be differences with plasma. Any such difference may be clinically relevant, owing to the much larger size of the ISF compartment, its proximity to cell metabolic processes, and its expansion in renal impairment.We used a modified microdialysis technique to successfully sample subcorneal ISF of some the uremic toxins (urea, creatinine, urate, phosphate). Reverse iontophoresis (RI) was also used as a non-invasive technique to sample epidermal ISF of urea. Hollow microneedles were developed and their ability to extract ISF tested in CKD patients and controls. The mechanical properties (pressure, volume, permeability) and biochemical composition (proteomic and metabolomic profiles) of the interstitial compartment were also investigated.Microdialysis and RI performed very well as interstitial uremic toxin sampling techniques. Small differences were seen in steady states between ISF and plasma urea, creatinine, phosphate and urate, with slightly lower ISF levels. Dialysis seemed to enhance this difference, with a lag in the clearance of ISF toxins seen in some patients, most remarkable with phosphate. Metabolomic analysis identified several uremic toxins in ISF, whilst proteomics found some significant differences between the two compartments, with toxins like beta-2 microglobulin occurring in ISF only. Microneedle arrays successfully extracted ISF in 68.8% of patients with oedema. Successful extraction of ISF with microneedles occurred mainly in oedematous patients, who were found to have raised interstitial pressures (ISP) and volumes. ISP correlated significantly with body fluid volumes and seemed time-dependent, lower in more chronic oedema. ISP and volumes also correlated with the oedema depitting time (after thumb pressure), a potential novel parameter that probably relates to tissue hydraulic conductivity and hence volume status and fluid mobility within the interstitium.This study demonstrates that interstitial fluid may need to be considered as a separate active compartment in patients with renal dysfunction, with a different “uremic" composition and unique pathophysiological characteristics that cannot be explained by blood compartment based measurements alone. There is a need for more studies, to further characterise this compartment and elucidate its importance.

Introduction: Dupuytren's disease (DD) is an ill-defined fibroproliferative disorder affecting the palms of the hands of certain patient groups. Whether changes in DD fibroblasts are due to genetic alterations alone or related to metabolic dysregulation has not yet been investigated. Hypotheses: 1. DD is a disease of several networks rather than of a single gene. 2. DD may be investigated more effectively by employing systems biology. 3. Strict definition of cell passage number is important for the revelation of any DD phenotype. 4. Some of the differences between DD and healthy tissues reside in a difference in their respiratory metabolism. 5. Any such differences are akin the Warburg effect noted for tumour cells in the literature. Methods: We induced hypoxia in healthy and disease cells to test whether the difference in disease cell types and healthy is the same as the difference in control fibroblasts cultured in normoxia and hypoxia. We investigated both at the metabolic level (intracellular and extracellular) and at the transcript level. This study also employed Fourier transform infrared spectroscopy to permit profiling of cells: (1) DD cords and nodules against the unaffected transverse palmar fascia (internal control), (2) those (1) with carpal ligamentous fascia (external controls) (3) those in (1) against DD fat surrounding the nodule, and skin overlying the nodule. We then compared metabolic profiles of the above to determine the effect of serial passaging by assessment of reproducibility. Subsequently, a novel protocol was employed in carefully controlled culture conditions for the parallel extraction of the metabolome and transcriptome of DD-derived fibroblasts and control at normoxic and hypoxic conditions to investigate this hypothesis. Gas chromatography-mass spectrometry combined with microarrays was employed to identify metabolites and transcript characteristic for DD tissue phenotypes. The extracellular metabolome was also studied for a selected subset. The metabolic and transcriptional changes were then integrated employing a network approach. Results: Carefully controlled culture conditions combined with multivariate statistical analyses demonstrated metabolic differences in DD and unaffected transverse palmar fascia, in addition to the external control. Differences between profiles of the four DD tissue phenotypes were also demonstrated. In addition early passage (0-3) metabolic differences were observed where a clear separation pattern in clusters was observed. Subsequent passages (4-6) displayed asynchrony, losing distinction between diseased and non-diseased sample phenotypes. A substantial number of dysregulated metabolites involved in amino acid metabolism, carbohydrate metabolism and also metabolism of cofactors and vitamins including downregulated cysteine and aspartic acid have been identified from the integrative analyses. Metabolic and transcriptional differences were revealed between fibroblast cell samples (passage number 3) cultured in 1% and 21% oxygen. The hypothesis that the difference in disease and healthy cells maybe akin to the differences in healthy cells in normoxia and hypoxia was rejected as only a very small number of significant molecules from these studies coincided in perturbed fascia and disease samples. No lactic acid was observed and little difference in the pyruvate concentrations. Yet, upon perturbation several of these transcripts and metabolites involved in the afore-mentioned pathways were significantly dysregulated. Conclusion: Early, but not late, passage numbers of primary cells provide representative metabolic and transcript fingerprinting for investigating DD. A unique parallel analysis of transcript and metabolic profiles of DD fibroblasts and control, enabled a robust characterization of DD and correlation of parameters across the various levels of systemic description. The tools that should facilitate our understanding of these complex systems are immature, but the pleiotropy of the difference between healthy and DD tissue suggest the aetiology of a network-based disease.

A cana-de-açúcar (Saccharum spp.) é uma das principais gramíneas cultivadas do mundo e o Brasil é seu maior produtor, ela se tornou uma importante cultura devido às altas taxas de assimilação de carbono permitindo a síntese e acumulação de grandes quantidades de sacarose em seus entrenós. Com isso, faz-se necessário uma melhor compreensão dos mecanismos moleculares que regulam o acúmulo de sacarose nesta planta. Tais mecanismos têm sido estudados em vários níveis, tais como, identificação e localização de genes, identificação de lócus controladores de características quantitativas, transcriptoma, proteôma, caracterização e identificação de metabólitos. Com todos esses estudos é evidente a necessidade de uma abordagem holística para o entendimento global da planta durante o acúmulo de sacarose. Assim este trabalho teve por objetivo integrar dados de metabolômica e proteômica de tecidos da cana-de-açúcar da variedade SP80-3280 durante o desenvolvimento e o acúmulo de sacarose, utilizando a bioinformática para unir esses resultados por meio da análise de correlação canônica regularizada em uma abordagem de biologia sistêmica. Os resultados obtidos indicam diferenças no perfil metabólico e proteico da cana-açúcar durante o desenvolvimento e acúmulo de sacarose. Foram propostas classes de metabólitos que podem estar relacionados com o acúmulo de sacarose na cana-de-açúcar tais como glicerolipídeos, glicerofosfolipídeos, cumarinas e derivados, esteroides e derivados de esteroides e acil graxos. Também foram propostas proteínas que podem estar relacionadas com o acúmulo de sacarose, onde as histonas foram as que mais se destacaram. Nas redes biológicas de correlações também foram observadas correlações entre possíveis metabólitos e proteínas que podem estar correlacionadas com o acúmulo de sacarose na cana-de-açúcar / Sugarcane (Saccharum spp.) is one of the most important cultivated grasses of the world and Brazil is the largest producer, it has become an important crop due to high carbon assimilation rates allowing the synthesis and accumulation of large amounts of sucrose in their internodes. Thus, it is necessary a high understanding of the molecular mechanisms involved in the regulation of sucrose accumulation in this plant. These mechanisms have been studied at various levels, such as gene identification and localization, identification of quantitative trait locus controlling, transcriptome, proteome, characterization and metabolites identification. With all these studies is evident the necessity for a holistic approach to global understanding of the plant during the sucrose accumulation. Thus, this work aims to integrate metabolomics and proteomics data from tissues of sugarcane variety SP80-3280, during plant development and sucrose accumulation, using bioinformatics to link these results by regularized canonical correlation analysis in a systems biology approach. The results indicate differences in the metabolic and protein profile of sugarcane during development and sucrose accumulation. Metabolites classes have been proposed that may be related to sugarcane sucrose accumulation as glycerolipids, glycerophospholipids, coumarins and derivatives, steroids and steroid derivatives and fatty acyl. In addition, some proteins have been proposed that may be related to sucrose accumulation, where the most highlighted were the histones. In the biological correlations networks, have been also observed correlations between possible metabolites and proteins that can be correlated with the accumulation of sucrose in sugarcane

As plantas se defendem do ataque de patógenos através de um sistema imune composto por duas fases. A primeira delas é mediada por receptores localizados na membrana celular ou intracelularmente, os quais são conhecidos como receptores de reconhecimento padrão (do inglês, pattern recognition receptors - PRR). Esses receptores reconhecem moléculas derivadas de microrganismos, as quais são conservadas evolutivamente e são chamadas de padrões moleculares associados a patógenos (do inglês, pathogen-associated molecular patterns - PAMPs). Esse reconhecimento dispara uma resposta de defesa conhecida como PTI (do inglês, PAMP-triggerd immunity - PTI). Alguns patógenos foram aptos a sintetizar moléculas capazes de suprimir a PTI e essas moléculas são denominadas de efetores. A resposta que ocorre devido à ação dos efetores é chamada de susceptibilidade disparada por efetores (do inglês, effector-triggered susceptibility - ETS). Entretanto, plantas resistentes podem reconhecer os efetores através de proteínas de resistência localizadas intracelularmente, ativando a imunidade disparada por efetores (do inglês, effector-triggeredimmunity - ETI). De modo geral, as respostas advindas da PTI e da ETI são similares, mas a segunda é ativada mais rapidamente e é mediada por um único gene de resistência R. Por essa razão, a ETI é conhecida como uma resposta à doença qualitativa e as plantas não desenvolvem sintomas, caracterizando a interação incompatível. Por outro lado, a PTI é mediada por diversos genes e as respostas de defesa são tardias, possibilitando a disseminação do patógeno pelas células da planta e a ocorrência da doença, o que caracteriza a interação compatível. Nas respostas de defesa, moléculas como o óxido nítrico, as poliaminas e o ácido salicílico participam do processo de sinalização. O sistema antioxidante da planta é ativado de modo a mitigar os efeitos das espécies reativas de oxigênio e o metabolismo da planta é alterado. Dessa maneira, o estudo das respostas de defesa contra patógenos, pode ser uma ferramenta útil para estabelecer controles efetivos para as doenças de plantas / Plants defend themselves from pathogen attack through an active immunity system composed by two phases. The first is mediated by cell surface and intracellular pattern recognition receptors (PRR), which recognizes conserved molecules derived from microbes known as pathogen-associated molecular patterns (PAMPs). This recognition triggers a defense response called PAMP-triggered immunity (PTI). Throughout evolution, pathogens were able to synthesize molecules capable of suppressing PTI. These molecules are named effectors and they are responsible for effector-triggered susceptibility (ETS). However, resistant plants can recognize effectors by intracellular resistance (R) proteins, initiating effector-triggered immunity (ETI). In general, responses derived from PTI and ETI are the same, but the latter is activated faster and is mediated by a single R gene. For this reason, ETI-response is also known as qualitative disease response (QDR) and plants do not develop disease symptoms, characterizing the incompatible interaction. On the other hand, PTI is mediated by several genes and the defense response is delayed, enabling the pathogen to spread out and to cause disease. This interaction is known as compatible. In defense responses, molecules like nitric oxide, polyamines and salicylic acid can participate in signaling process. The antioxidant system can be activated to quench the ROS effects and the plant metabolism is altered. In this sense, studying defense responses against pathogens can help to develop tools to establish effective control methods for plant disease

Introdução: A apneia obstrutiva do sono (AOS) é uma condição clínica comum, embora subdiagnosticada na prática clínica, que está associada de forma independente com um aumento na morbimortalidade cardiovascular. Evidências recentes sugerem que o aumento do risco cardiovascular atribuído à AOS pode ser parcialmente explicado pela desregulação metabólica. No entanto, pouco se sabe sobre o perfil metabólico detalhado destes pacientes e se estes metabólitos podem servir como potenciais biomarcadores para a AOS. Objetivos: O objetivo primário do estudo foi avaliar o perfil metabólico de pacientes com AOS por meio de diferentes estratégias metabolômicas. Como objetivo secundário, avaliamos se o painel de metabólitos selecionados poderia agregar valor diagnóstico ao uso de questionários tradicionalmente empregados para a triagem da AOS. Métodos: Participantes do sexo masculino sem doença cardiovascular prévia ou uso de medicamentos foram submetidos à polissonografia noturna e divididos em 2 grupos pareados por idade e índice de massa corpórea (IMC): sem AOS (índice de apneia e hipopneia [IAH] < 15 eventos/h) e com AOS (IAH >= 15 eventos/h). Além da avaliação clínica, foi aplicado dois questionários para triagem da AOS usados na prática clínica (questionário Berlim e o escore NoSAS). A quantificação de aminoácidos (AA) no plasma foi realizada por cromatografia líquida acoplada à espectrometria de massas sequencial (LC-MS²) previamente desenvolvido e validado no Grupo Fleury. Para as análises metabolômicas e lipidômicas foram utilizados cromatografia gasosa acoplado a espectrometria de massas (GC-MS) e cromatografia líquida (LC) off-line com detecção por ionização/dessorção a laser auxiliada por matriz (MALDI-MS), respectivamente. Para avaliação dos marcadores que estariam associados à AOS foram utilizados teste t de student (p < 0,05) e VIP score > 2 (predição de variável de importância). A sensibilidade e especificidade foram avaliadas por meio da curva receiver operating characteristic (ROC) e modelos de regressão logística em associação com o melhor questionário de triagem para a AOS. Resultados: 53 participantes foram estudados (16 sem AOS e 37 com AOS). Como proposto, a idade média (36 ± 6 vs. 39 ± 7 anos) e o IMC (30,3 ± 3,5 vs. 30,6 ± 3,4 kg/m2) foram semelhantes entres os grupos sem e com AOS, respectivamente. A quantificação dos AA permitiu observar uma diferença significante nos níveis de ácido glutâmico, que foram maiores nos pacientes com AOS (83,5 ± 22,5 ?M) quando comparados ao grupo sem AOS (66,7 ± 24,5 uM), p=0,023. A avaliação do perfil metabolômico resultou em 28 analitos significantes. Seis desses analitos foram provenientes da análise por GC-MS: pacientes com AOS apresentaram maiores níveis de 6-deoxi-D-glicose; 2,6-difenil-1,7-diidrodipirrolo [2,3-b:3\',2\'-e] piridina, ácido 9 (Z)-hexadecenóico e ácido araquidônico e menores níveis de 5,5\'-bifitalato e glutamina quando comparados ao grupo sem AOS (p < 0,05). A análise lipidômica (LC off-line e MALDI-MS) resultou em 22 lipídios significantes subdivididos nas seguintes classes: ceramidas (Cer), fosfatidiletanolamina (PE), lisofosfatidilcolina (LPC), e esfingomielina (SM) com níveis mais elevados em pacientes com AOS em comparação ao grupo sem AOS (p < 0,05). Diacilglicerol (DAG), fosfatidilcolina (PC) e ácido fosfatídico (PA) tiveram níveis significativamente diminuídos nos pacientes com AOS em comparação aos sem AOS. Em relação ao objetivo secundário, o questionário com melhor desempenho para triar a AOS foi o escore NoSAS com uma área sob a curva (AUC) de 0,724. A combinação dos 4 metabólitos (ácido glutâmico, glutamina, 6-deoxi-D-glicose e ácido araquidônico) ou dos 3 lipídios (LPE 35:1, SM d18:1/12:0 e LPC 27:1) selecionados pelo modelo de regressão logística ao escore NoSAS positivo resultou em uma AUC de 0,917 e 0,951, respectivamente, para a detecção da AOS. Conclusão: A aplicação da metabolômica permitiu a identificação de potenciais biomarcadores precoces para a AOS em homens jovens. Estes achados não são explicados por fatores de confusão como a idade, IMC e composição corporal. Este estudo confirma o achado de que os questionários habitualmente usados para a triagem da AOS não apresentam um bom desempenho. A combinação dos metabólitos selecionados aumentou a sensibilidade e especificidade para detecção da AOS em relação ao questionário isolado. Neste contexto, novos estudos são necessários para avaliar se estes biomarcadores poderão ser úteis para a predição do risco cardiovascular e melhora do diagnóstico da AOS / Introduction: Obstructive sleep apnea (OSA) is a common clinical condition, although frequently underdiagnosed in clinical practice. OSA is independently associated with an increased risk of cardiovascular morbidity and mortality. Recent evidence suggests that the cardiovascular risk attributed to OSA may be partially explained by metabolic dysregulation. However, little is known about the detailed metabolic profile of these patients and whether these metabolites may serve as potential biomarkers for OSA. Objectives: The primary objective of the study was to evaluate the metabolic profile of OSA patients through different metabolomics strategies. As a secondary objective, we evaluated whether the panel of selected metabolites could improve diagnostic value to the use of questionnaires traditionally used for OSA screening. Methods: Male participants with no previous cardiovascular diseases and under no medications were submitted to a nocturnal polysomnography and divided into 2 groups matched by age and body mass index (BMI): no OSA (apnea and hypopnea index [AHI] < 15 events/h) an OSA (AHI >= 15 events/h). In addition to the clinical evaluation, was applied two questionnaires to screening OSA in the clinical practice (Berlin questionnaire and the NoSAS score). The quantification of amino acids (AA) in plasma was performed by liquid chromatography coupled to sequential mass spectrometry (LC-MS/MS) previously developed and validated in the Fleury Group. To the metabolomics and lipidomics analysis, we used gas chromatography coupled to mass spectrometry (GC-MS) and off-line liquid chromatography (LC) with matrix-assisted laser ionization/desorption detection (MALDI-MS), respectively. Student\'s t test (p < 0.05) and VIP score >2 (significance variable prediction) were used to evaluate the markers associated with OSA. Sensitivity and specificity were evaluated using the receiver operating characteristic (ROC) curve and logistic regression models in association with the best screening questionnaire for OSA. Results: 53 participants were studied (16 with no OSA and 37 with OSA). As proposed, mean age (36 ± 6 vs. 39 ± 7 years) and BMI (30.3 ± 3.5 vs. 30.6 ± 3.4 kg/m2) were similar between the groups without and with OSA, respectively. The quantification of AA showed a significant difference in the glutamic acid levels, which were higher in patients with OSA (83.5 ± 22.5 ?M) when compared to the group with no OSA (66.7 ± 24.5 ?M), p=0.023. Metabolomics profile analysis resulted in 28 significant analytes. Six of these analytes came from GC-MS analysis: patients with OSA had higher levels of 6-deoxy-D-glucose, 2,6-diphenyl-1,7-dihydrodipyrrolo [2,3-b:3\',2\'-e] pyridine, 9-hexadecenoic acid (Z) and arachidonic acid and lower levels of 5,5\'-biphthalide and glutamine when compared to the no OSA group (p < 0.05). The lipidomics analysis (LC off-line and MALDI-MS) resulted in 22 significant lipids subdivided into the following classes: glycerophosphoethanolamine (PE), lysophosphosphatidylcholine (LPC), and sphingomyelin (SM) with higher levels in patients with OSA when compared to the no OSA group (p < 0.05). Diaglycerol (DAG), glycerophosphocholine (PC) and phosphatidic acid (PA) had significantly decreased levels in patients with OSA compared to those with no OSA. Regarding to the secondary endpoint, the best performing questionnaire for screening OSA was the NoSAS score with an area under the curve (AUC) of 0.724. The combination of the 4 metabolites (glutamic acid, glutamine, 6-deoxy-D-glucose and arachidonic acid) or the 3 lipids (LPE 35:1, SM d18:1/12:0 and LPC 27:1) previously selected by logistic regression model to the NoSAS positive score resulted in an AUC of 0.917 and 0.951, respectively, for the OSA detection. Conclusion: The application of metabolomics strategies allowed the identification of potential early biomarkers for OSA in young male subjects. These findings are not explained by confounding factors such as age, BMI and body composition. This study confirms previous findings that the questionnaires commonly used for screening OSA do not have a good performance. The combination of the selected metabolites increased the sensitivity and specificity to OSA detection in relation to the use of sleep questionnaire only. In this context, further studies are necessary to assess whether these biomarkers may be useful for predicting cardiovascular risk and improving the OSA diagnosis

L’alimentation représente une piste prometteuse pour la prévention des pathologies du vieillissement cérébral, en particulier la démence et sa principale cause, la maladie d’Alzheimer, pour lesquelles il n’existe aucun traitement étiologique. La recherche épidémiologique nutritionnelle a traditionnellement reposé sur une approche par nutriment candidat, ignorant les effets additifs et potentiellement synergiques entre composés alimentaires, pourtant consommés ensemble dans l’alimentation. La combinaison nutritionnelle optimale pour la prévention du déclin cognitif et de la démence reste encore inconnue. L’objectif principal de ce travail de thèse était d’identifier de nouvelles cibles préventives pour le vieillissement cérébral via une approche holistique de l’exposition nutritionnelle, combinant des méthodes semi-confirmatoires (par profils de nutriments candidats) et des approches plus exploratoires (profilage métabolomique) dans la cohorte de personnes âgées des Trois-Cités (3C), Bordeaux. Dans une première partie semiconfirmatoire, deux profils multi-nutriments associés au risque de démence au cours de 12 ans de suivi, ont été identifiés par une méthode de réduction de dimension supervisée : un profil d’apports alimentaires en polyphénols (combinant principalement des composés phénoliques du vin rouge, des agrumes et des noix) et une combinaison de biomarqueurs nutritionnels (vitamine D, caroténoïdes et acide gras polyinsaturés). Dans une seconde partie exploratoire, des profils en métabolomique non-ciblée, analysés par régression pénalisée, ont mis en évidence une relation entre le métabolome alimentaire (métabolites dérivés du café, des agrumes, du chocolat, du poisson, du vin rouge) et le déclin cognitif. Des altérations du métabolome endogène ont également été observées, suggérant l’implication du statut cardiométabolique (phospholipides, acylcarnitines, acides biliaires, triméthyllysine, glucose) et du microbiote (acides biliaires secondaires, triméthyllysine) dans le vieillissement cérébral. Enfin, une analyse du lipidome a permis d’identifier une signature lipidique du déclin cognitif, validée dans un échantillon externe issu de 3C Dijon. Cette étude suggérait des modifications précoces des lipides membranaires au cours du vieillissement cérébral. / Diet is a promising strategy for the prevention of brain aging, including dementia and its main form Alzheimer’s disease, for which no cure currently exists. Traditional nutritional epidemiology is based on a candidate-nutrient approach that ignores additive effects and potential synergies between dietary compounds, though consumed together in the diet. The optimal combination of nutrients for the prevention of cognitive decline and dementia remains unknown. The objective of this thesis was to identify new preventive targets for brain aging through a holistic approach of nutritional exposures, including semiconfirmatory (patterns of candidate-nutrients) and exploratory methods (metabolomics profiling) applied to older people from the Three-City (3C) study, Bordeaux. In a first part (semi-confirmatory), we investigated combinations of nutrients best associated with the risk of dementia over 12 years, using partial least square regression. We identified two multinutrient patterns: a pattern of polyphenols intake (mainly composed of polyphenols from red wine, citrus and nuts) and a nutrient biomarker pattern (including vitamin D, carotenoids and polyunsaturated fatty acids). In a second part (exploratory), we discovered, using untargeted metabolomics and penalized regression, associations between the food metabolome (metabolites derived from coffee, citrus, chocolate, fish and red wine) and cognitive decline. Other metabolic alterations involved the endogeneous metabolome, including metabolites related to cardiometabolic status (phospholipids, acylcarnitines, bile acids, trimethyllysine, glucose) and to gut microbiota (secondary bile acids, trimethyllysine). Finally, the study of the lipidome revealed a lipid signature of cognitive decline, validated in an external sample from 3C Dijon, which suggested early change in membrane lipids over the course of brain aging.

La métabolomique est une approche de phénotypage présentant des perspectives prometteuses pour le diagnostic et le suivi de plusieurs pathologies. La technique d'observation la plus utilisée en métabolomique est la spectrométrie de masse (MS). Des développements technologiques récents ont considérablement accru la taille et la complexité des données. Cette thèse s'est concentrée sur deux verrous du traitement de ces données, l'extraction de pics des données brutes et l'annotation des spectres. La première partie de la thèse a porté sur le développement d'un nouvel algorithme de détection de pics pour des données d'analyse par injection en flot continue (Flow Injection Analysis ou FIA), une technique haut-débit. Un modèle dérivé de la physique de l'instrument de mesure prenant en compte la saturation de l'appareil a été proposé. Ce modèle inclut notamment un pic commun à tous les métabolites et un phénomène de saturation spécifique pour chaque ion. Ce modèle a permis de créer une workow qui estime ce pic commun sur des signaux peu bruités, puis l'utilise dans un filtre adapté sur tous les signaux. Son efficacité sur des données réelles a été étudiée et il a été montré que proFIA était supérieur aux algorithmes existants, avait une bonne reproductibilité et était très proche des mesures manuelles effectuées par un expert sur plusieurs types d'appareils. La seconde partie de cette thèse a porté sur le développement d'un outil de détection des similarités structurales d'un ensemble de spectre de fragmentation. Pour ce faire une nouvelle représentation sous forme de graphe a été proposée qui ne nécessite pas de connaître la composition atomique du métabolite. Ces graphes sont de plus une représentation naturelle des spectres MS/MS. Certaines propriétés de ces graphes ont ensuite permis de créer un algorithme efficace de détection des sous graphes fréquents (FSM) basé sur la génération d'arbres couvrants de graphes. Cet outil a été testé sur deux jeux de données différents et a prouvé sa vitesse et son interprétabilité comparé aux algorithmes de l'état de l'art. Ces deux algorithmes ont été implémentés dans des package R, proFIA et mineMS2 disponibles à la communauté. / Metabolomics is a phenotyping approach with promising prospects for the diagnosis and monitoring of several diseases. The most widely used observation technique in metabolomics is mass spectrometry (MS). Recent technological developments have significantly increased the size and complexity of data. This thesis focused on two bottlenecks in the processing of these data, the extraction of peaks from raw data and the annotation of MS/MS spectra. The first part of the thesis focused on the development of a new peak detection algorithm for Flow Injection Analysis (FIA) data, a high-throughput metabolomics technique. A model derived from the physics of the mass spectrometer taking into account the saturation of the instrument has been proposed. This model includes a peak common to all metabolites and a specific saturation phenomenon for each ion. This model has made it possible to create a workflow that estimates the common peak on well-behaved signals, then uses it to perform matched filtration on all signals. Its effectiveness on real data has been studied and it has been shown that proFIA is superior to existing algorithms, has good reproducibility and is very close to manual measurements made by an expert on several types of devices. The second part of this thesis focused on the development of a tool for detecting the structural similarities of a set of fragmentation spectra. To do this, a new graphical representation has been proposed, which does not require the metabolite formula. The graphs are also a natural representation of MS/MS spectra. Some properties of these graphs have then made it possible to create an efficient algorithm for detecting frequent subgraphs (FSM) based on the generation of trees covering graphs. This tool has been tested on two different data sets and has proven its speed and interpretability compared to state-of-the-art algorithms. These two algorithms have been implemented in R, proFIA and mineMS2 packages available to the community.

<p>Postmortem aging has been extensively practiced as value-adding process due to the beneficial impacts on meat palatability. Meat tenderization occurred through proteolytic fragmentation of myofibrillar structural proteins via endogenous protease systems, which is considered as the primary drive to enhance major palatability attributes including tenderness, juiciness, and flavor. Recent theoretical framework proposes apoptosis, or programmed cell death, as the preceding step that initiates postmortem proteolysis. Whereas small heat shock proteins have been consistently recognized as meat quality biomarkers, probably due to their protective activities against proteolysis through anti-stress, anti-apoptotic, and chaperoning functionalities. To shed light on detailed mechanisms controlling postmortem proteolysis and consequential impacts on the development of fresh meat quality characteristics, postmortem proteolytic changes of small heat shock proteins, apoptotic factors, and myofibrillar structural proteins were profiled in postmortem skeletal muscles under different metabolic backgrounds and across species. </p> <p>In beef, three muscles, <i>longissimus lumborum</i> (LL), <i>semimembranosus</i> (SM), and <i>psoas major</i> (PM), have been selected to represent glycolytic, intermediate, and oxidative muscle types. Tenderness and water - holding capacity were determined, and proteolysis, apoptotic features, and small heat shock proteins were measured in 8 beef carcasses at 1, 2, 9, 16, and 23 days of aging. PM exhibited limited aging potential in quality developments shown by lower extents of shear force, water-holding capacity, and proteolytic changes, including calpain 1 autolysis, troponin T, and HSP27 compared to LL and SM. Conversely, LL had an increase in tenderization and water-holding capacity, which was accompanied with more extended calpain 1 autolysis, proteolysis and HSP27 degradation, compared with other muscles. The results of this study suggest that postmortem proteolytic changes of myofibrillar proteins, small HSPs and apoptotic factors occur in a muscle-specific manner, which is likely attributed to different rate and extent of meat quality developments of each muscle during aging. </p> <p>Callipyge lambs are a unique genetic background showing calpastatin over-expression, muscle hypertrophy in loin and hindquarter area, substantially compromised meat tenderization potential, and a shift of muscle fiber composition towards fast-glycolytic directions. Proteome and metabolome changes in muscles from callipyge mutation (+/C) and non-callipyge phenotype (+/+, C/+, and C/C) lambs were profiled to provide insight into the biochemical changes affecting meat quality attributes. M. longissimus thoracis from lambs with all four possible callipyge genotype (n = 4, C/+, C/C, +/C, and +/+) were collected after 3d aging and analyzed using mass-spectrometry based platforms. Among identified proteomes, cytochrome c (pro-apoptotic protein) was detected with significantly lower abundances in +/C. Anti-apoptotic HSP70, BAG3, and PARK7 were over-abundant in +/C, which could result in delayed apoptosis and possibly attributed to tougher meat in callipyge lambs. Eight glycolysis enzymes were overabundant in +/C lambs, whereas 3 enzymes involved in TCA cycle were overabundant in non-callipyge ones (C/C and/or C/+). Twenty-five metabolites were affected by genotypes (P < 0.05), including metabolic co-factors, polyphenols, and AA/short peptides.</p> <p>Pig production is facing increased public pressure regarding antibiotic usage restriction. Recently, dietary L-glutamine at cost effective level (0.2%) was identified as an effective antibiotic alternative in post-transport nursery pig diets. To evaluate carcass and meat quality characteristics in market-ready pigs when 0.2% dietary L-glutamine was applied as for early-life post-weaning and transport recovery, pigs (N=480) were weaned and transported in two replication trials in SPRING (April of 2017) vs. SUMMER (July of 2016), fed 3 different diets (Non: no antibiotic, Anti: 441 ppm chlortetracycline and + 38.6 ppm tiamulin, Gln: 0.20% L-glutamine) for 14 days after transport, and fed basal diet until reaching market weight. Pairs of <i>longissimus dorsi</i> (LD) and <i>psoas major</i> (PM) muscles from each carcass (n=10/diet/trial) were separated at 1 d and 7 d postmortem, respectively. Carcass yield and meat physical and quality attributes were evaluated. Overall impacts of Gln on physical attributes of carcasses and porcine muscles were minimal. No dietary effects were found in carcass, proximate composition, water-holding capacity, or shear force. Significant difference between trials were found in terms of productivity and pork/carcass qualities, where SPRING replicates showed increased body weight, faster pH decline, paler surface color, higher intra-muscular fat deposition, and improved tenderness and water-holding capacity as indicated by lower shear force values, thaw-purge loss, and cooking loss (P < 0.05).</p> <p>The pork and carcass quality results give rise to a postulation that different metabolism and animal growth might have been occured between the two production trials, consequentially differentiated meat quality development. In this regard, myofibrillar proteolysis, small heat shock proteins, and apoptotic factors were characterized during 7 d postmortem aging in porcine LD and PM muscles from both seasonal trials, combined with metabolomics profiles of 1d samples using the GC-TOF-MS/MS platform. Compared to SUMMER counterparts, SPRING muscles showed concurrence of more extended apoptosis, further calpain 1 autolysis, and increased structural protein degradation (P<0.05). SPRING muscles showed more ATP catabolism compounds and increase in carbohydrates, branched-chain amino acids, and 16-18 carbon fatty acids, which could be chemistry fingerprints of increased cellular oxidative stress, consequentially favoring onset of apoptosis and proteolysis. Meanwhile, SUMMER pigs showed increased stress-defending metabolites, such as ascorbic acid, antioxidant amino acids, and decreased inhibitory neuro-transmitter GABA, which may indicate elevated stress-defending activity in SUMMER pigs that possibly inhibited apoptosis and proteolysis. </p>

<p>Within the last decade, the popularity and interest in dry-aging have constantly increased among both consumers and producers. Dry-aging is a natural value-adding process where meat is exposed to a controlled refrigerated environment without any protective barrier during the aging process. This process leads to the development of unique flavors in the final meat product. Although the prevalence of this process is increasing, there are inconsistent reports regarding the impacts of dry-aging on meat sensory attributes, especially on the flavor aspect. Given that flavor generation is dependent on the composition and availability of flavor precursors, the presence or absence of these precursors may contribute to the inconsistency observed. Thus the main objective of the research described here was to characterize the flavor precursors in dry-aged meat and elucidate potential factors or mechanisms favoring to their production.</p> <p> To achieve this objective, metabolomics analysis was conducted in conjunction with various chemical analyses (free amino acids, fatty acids, sugar content and volatile analysis), microbiome profiling and meat quality analysis (tenderness, water holding capacity, color stability, oxidative stability, microbial attributes and sensory analysis) to identify the essential flavor precursors and their production process. In addition, similar analyses were conducted using multiple meat sources (grass-fed beef loins, cull cow beef loins and pork loins) aged by wet-aging (WA), conventional dry-aging (DA), dry-aging in bag (DWA) and UV-light dry-aging (UDA) to elucidate the impact of the different aging treatments on meat quality, sensory attributes and flavor precursor availability.</p> <p>Regardless of the meat source, the results demonstrated that dry-aging altered the meat flavor precursor compositions, primarily by increasing the presence of protein-derived precursors (e.g., free amino acids and dipeptides), especially glutamine and glutamate compounds. Additionally, nucleotide and carbohydrate-derived compounds such as adenosine and reducing sugars were greatly increased after the dry-aging process. While the fatty acid profile was minimally affected, metabolomics analysis revealed a decrease in sterol and terpenoid lipids following dry-aging, which could potentially reduce off-flavors development in the meat. Other compounds such as vitamin B and vitamin C were also detected in the dry-aged product, which potentially could contribute to the flavor development.</p> <p>Analysis of the liberation mechanisms demonstrated that dehydration played a role in increasing the concentration of the flavor precursors in the dry-aged product, potentially promoting greater (e.g., Maillard reaction) during cooking. Furthermore, microorganisms might be responsible for further increasing the availability of flavor precursors in dry-aged meat, especially free amino acids, along with the dehydration process. Microbiome profiling found that <i>Pseudomonas</i> spp. are the most prominent bacterial species in microbial communities found on dry-aged meat which could affect the precursor release in dry-aged meat. Metabolomics analysis also indicated increased glutathione metabolism during dry-aging, which could lead to the liberation of glutamine-related compounds. The analysis also identified other compounds such as porphyrin rings (iron-related) and shikimic acid (bacterial metabolism), providing further examples of how metabolomics can identify dry-aged flavor precursors and reveal other potential mechanisms related to flavor development mechanisms.</p> <p>These outcomes demonstrate that dry-aging alters meat flavor precursor composition, mainly by increasing the availability of protein-, nucleotide- and carbohydrate-derived compounds. Such results indicate that the Maillard reaction is likely be the main mechanism in flavor generation in dry-aged meat. The current results provided more insights into the dry-aging flavor development, especially highlighting important flavor precursor such as glutamate and glutamine containing products, likely to contribute to the dry-aged flavor. Future study to identify the impact of different microorganism (especially mold and yeast) on dry-aging flavor development would be of interest. Additionally, impact of different cooking process should also be studies to maximize the dry-aged flavor potential from the product.</p>

This thesis is focused on the possibility of analysis of volatile organic compounds produced by monocytes during sepsis. Method of comprehensive two-dimensional gas chromatography with mass spectrometric detection was chosen for this purpose. Content of the first part was the optimization of the method of two-dimensional gas chromatography for the determination of volatile organic compounds. In this part were gradually adjusted parameters of the gas chromatography method to achieve the maximum efficiency. Further were adjusted conditions of samples preparation. Content of the second part was the usage of already optimized method for the analysis of the samples set of monocytes. Samples were subjected to the action of different inhibitors of the immune system and stimulators simulating bacterial or yeast infection. Based on this analysis were identified some compounds, which are produced by monocytes under condition simulating the infection.