Дисертації з теми "Metabolomic analyses"

Metabolomics is an emerging field in functional plant biology that attempts to relate patterns in the molecular intermediates and products of metabolic pathways with genetic, gene expression, environmental and phenotypic traits - at the whole-tissue and/or whole-organism level. There is enormous potential for metabolomics tools to be applied in the study of tree species, and the demand for widespread application is promoting an ongoing evolution and refinement of newly-developed techniques. This body of research addresses the application of broad-scale, non-targeted metabolomics to questions of wood formation and quality in tree systems. Overall, it was shown that variation in metabolite profiles from developing xylem tissue was indeed correlated with the strength of specific phenotypic traits. Frequently, the strength of these relationships was such that phenotypic severity could be predicted accurately on the basis of metabolite profile data alone. The specific correlative patterns and metabolite/trait pairings observed in each study provided insight into the biological mechanisms by which these traits arise. Studies of secondary xylem development were conducted on breeding populations of Douglas-fir and radiata pine, as well as genetically modified hybrid poplar. In the Douglas-fir families studied, environment-induced variation in growth rate, fibre morphology and wood chemistry were correlated with metabolite profiles from developing xylem; metabolites involved in carbohydrate and lignin biosynthesis were primarily implicated in these relationships. Similarly, in juvenile trees from a series of radiata pine families, correlations were observed between metabolite profiles of developing xylem and the internal checking wood defect, a known heritable trait. In a different approach, two poplar hybrids, each modified separately with two exogenous gene constructs related to lignin biosynthesis, provided controlled model systems in which to investigate the interaction between genotype, metabolite profiles of developing xylem, and physico-chemical wood traits. Wood traits and metabolite profiles alike were altered by the genetic modifications, and it was found that the metabolic impact of the transgenes was not confined to pathways that were directly coupled to lignin biosynthesis. In fact, the scarcity of lignin-related metabolites in profiles from either the wild-type or modified genotypes suggested that metabolite channelling phenomena operate in the lignin biosynthetic pathway. Moreover, the analyses demonstrated that transgene-induced gradients in phenotypic traits could be associated with similar gradients within broad-scale metabolite profiles, and also that the wood-forming metabolisms of different poplar hybrids can respond similarly to the influences of genetic manipulation, at a global level. To conclude, the demonstrated associations between genotype, the metabolism of wood formation, and wood phenotype, as revealed by metabolite profiles, confirm the value of non-targeted metabolomics as a systems biology approach to understanding and modeling growth and secondary cell wall biosynthesis in trees.

Inborn errors of metabolism (IEMs) constitute a major class of genetic disorder. Most of IEMs are transmitted recessively, so consanguinity has a huge impact on disease prevalence, particularly in societies like Saudi Arabia, where consanguineous marriage is common. Understanding and treatment are very important in genetic diseases, and simple models would be helpful. Thus, the feasibility of applying the fruit fly, Drosophila melanogaster, as a model for a human renal genetic disease - xanthinuria - was investigated. Xanthinuria is a rare human genetic disease, caused by mutations in xanthine oxidase or molybdenum cofactor sulphurase; in Drosophila, the homologous genes are rosy (ry) and maroon-like (mal), respectively. The new Orbitrap technology of mass spectrometry has the potential to determine levels of many metabolites simultaneously by exact mass, and a major part of this thesis was to investigate the utility of Orbitrap technology in metabolomics of both wild-type and Drosophila mutant. Repeatable significant differences were identified between ry and wild-type flies, which recapitulated painstaking analytical biochemical determinations of the 1950s, but with greater precision. Additionally, completely novel impacts of the ry mutation (on pyrimidine metabolism, the urea cycle and osmolyte biosynthesis) were identified. As expected mal mutants showed more similar changes as ry, but with widespread metabolic perturbations. The online resource, FlyAtlas.org, provides detailed microarray-based expression data for multiple tissues and life-stages of Drosophila. Downstream genes, such as urate oxidase, are utterly tubule-specific. Accordingly, the utility of Orbitrap technology in elucidating tissue-specific metabolomes was also investigated. Additionally, genetic interventions using designed RNAi constructs were also made and validated by QPCR and metabolomics. As urate is a potent antioxidant, survival of urate oxidase knockdowns was tested in vivo, and a significant impact on survival identified. An Affymetrix microarray was performed, comparing ry506 mutant flies against wild-type and differences were identified in a second experiment, the anti-gout drug allopurinol was used to phenocopy the effects of ry. Overall, the thesis showed that Orbitrap technology was highly suitable for metabolomic analysis of both wild-type and mutant Drosophila, and had potential in the analysis of metabolomes of single tissues. The possibility of using Orbitrap-based metabolomicsin human diagnosis is discussed.

Glioblastoma multiforme (GBM) is the most common primary brain tumor. Given the current standard of care, the prognosis for patients diagnosed with this disease is still poor. There consequently exists a need to improve current treatments, as well as to develop new ones. Many obstacles however need to be overcome to facilitate this effort and one of these involves the development of improved methods to monitor treatment effects. At present, the effects of treatment are typically assessed by radiological means several months after its initiation, which is unsatisfactory for a fast growing tumor like GBM. It is however likely that treatment effects can be detected on a molecular level long before radiological response, especially considering many of the targeted therapies that are currently being developed. Biomarkers for treatment efficacy may be of great importance in the future individualization of brain tumor treatment. The work presented herein was primarily focused on detecting early effects of GBM treatment. To this end, we designed experiments in the BT4C rat glioma model in which we studied effects of both conventional radiotherapy and an experimental angiogenesis inhibitor, vandetanib. Brain tissue samples were analyzed using a high throughput mass spectrometry (MS) based screening, known as Surface Enhanced Laser Desorption/Ionization - Time of Flight - Mass Spectrometry (SELDI-TOF-MS). The vast amounts of data generated were subsequently analyzed by established multivariate statistical methods, such as Principal Component Analysis (PCA), Partial Least Squares (PLS), and Orthogonal Partial Least Squares (OPLS), developed for analysis of large and complex datasets. In the radiotherapy study we detected a protein spectrum pattern clearly related to tumor progression. We notably observed how this progression pattern was hampered by radiotherapy. The vandetanib study also revealed significant alterations of protein expression following treatment of different durations, both in tumor tissue and in normal brain contralateral to the tumor. In an effort to further elucidate the pathophysiology of GBM, particularly in relation to treatment, we collected extracellular fluid (ECF) samples from 11 patients diagnosed with inoperable GBM. The samples were collected by means of stereotactic microdialysis, both from within the contrast enhancing tumor and the brain adjacent to tumor (BAT). Samples were collected longitudinally from each patient in a time span of up to two weeks, during which the patient received the first five fractions of radiotherapy. The ECF samples were then analyzed by Gas Chromatography Mass Spectrometry (GC-MS) to screen them with respect to concentrations of low molecular weight compounds (metabolites). Suitable multivariate analysis strategies enabled us to extract patterns of varying metabolite concentrations distinguishing between samples collected at different locations in the brain as well as between samples collected at different time points in relation to treatment. In a separate study, we also applied SELDI-TOF-MS and multivariate statistical methods to unravel possible differences in protein spectra between invasive and non-invasive WHO grade I meningiomas. This type of tumor can usually be cured by surgical resection however sometimes it grows invasively into the bone, ultimately causing clinical problems. This study revealed the possibility to differentiate between invasive and non-invasive benign meningioma based on the expression pattern of a few proteins. Our approach, which includes sample analysis and data handling, is applicable to a wide range of screening studies. In this work we demonstrated that the combination of MS screening and multivariate analyses is a powerful tool in the search for patterns related to treatment effects and diagnostics in brain tumors.

Human African trypanosomiasis (HAT) is a disease that is in desperate need of new pharmacological agents active against the causative parasite, the flagellated protozoan Trypanosoma brucei. In this thesis, new metabolomics techniques have been developed to study pathways in response to drug action with the aim of defining the mode of action of current and future drugs. Eflornithine, a polyamine pathway inhibitor, was used as a proof of principle, revealing both expected changes that correlate well with the literature and unexpected changes that lead to pathways and metabolites not previously described in bloodstream form trypanosomes. One metabolite not previously described in trypanosomes is acetylornithine, whose levels correlate well with ornithine and whose production comes directly from ornithine transported from the medium. Nifurtimox and the nifurtimox- eflornithine combination therapy were assayed for changes to their metabolomes revealing changes in nifurtimox treatment that included alterations to sugar and purine levels. The combination therapy had reduced changes to some metabolites compared to each drug in isolation suggesting reasons for the combination‟s lack of synergy. Isotopically labelled metabolites were also of use in determining flux through the pathways identified as being affected by drug perturbation. These techniques, along with other biochemical techniques, were used to show arginase activity is absent in bloodstream form trypanosomes and that ornithine is not made from arginine when ornithine is present in the medium. Arginine can, however, be used to produce ornithine through an arginase-independent mechanism when exogenous ornithine is lacking. Evidence is also provided that parts of the pentose phosphate pathway, not thought to be active in bloodstream form trypanosomes, may still be active in in vitro grown cells. A mechanism of resistance to eflornithine involving the deletion of an amino acid transporter that is able to transport eflornithine is also described. It is hoped that simple PCR-based tests for this resistance mechanism will be of use in resistant foci in prescribing appropriate drugs to HAT patients.

Le profilage métabolomique global de matrices biologiques dans de larges séries d'échantillon est un enjeu majeur. Dans ce contexte, notre travail vise à développer des approches LC-HRMS et outils bioinformatiques pour les analyses métabolomique et lipidomique de larges cohortes. Dans un premier temps, nous avons développé puis évalué la pertinence de 4 méthodes LC-HRMS dans l'annotation du métabolome/lipidome sérique humain. Ainsi, une base de données spectrales a été implémentée à l'aide de spectres MS, MS/MS et les temps de rétention de composés de référence afin d'assurer l'annotation de jeux de données. La combinaison de méthodes RP, HILIC et PFPP-HRMS a permis l'identification de 266 métabolites et 706 espèces lipidiques sériques répartis sur 20 et 24 classes chimiques respectivement dont 27% d'espèces isomères. Ces outils ont été appliqués, dans un second temps, à la stratification de 78 patients diabétiques. Outre le syndrome métabolique marqué (perturbation du métabolisme énergétique), nos analyses ont montré l'impact délétère de facteurs physiologiques confondants -âge et IMC-. Nous en avons évalué l'influence sur une cohorte de 227 salariés du CEA. Les empreintes lipidomiques sont robustes, néanmoins l'impact de l'IMC est marqué pour les lipides neutres. L'effet du genre démontre un catabolisme masculin important. L'effet de l'âge se manifeste par des activités enzymatiques altérées. Ces études combinent une analyse globale métabolomique et lipidomique des mêmes échantillons humains. Elles visent à construire une base de données relationnelle incluant données spectrales et biologiques servant à la caractérisation de biomarqueurs dans le cas d'études cliniques
Global metabolomic profiling of biological media in large sample sets is a major challenge. In this context, our work aims to develop LC-HRMS approaches and data mining tools for metabolomics and lipidomics analysis of large cohorts. We have first developed and evaluated the reliability of four LC-HRMS methods in the annotation of human serum metabolome and lipidome. Thus, spectral database was implemented using MS spectra, MS/MS and retention times of reference compounds to further ensure datasets annotation. The combination of RP, PFPP and HILIC-HRMS methods allowed identification of 266 metabolites and 706 lipid species in human serum over 20 to 24 chemical classes respectively including 27% of isomeric species. These analytical tools were then applied for the stratification of 78 diabetic patients. Unsurprisingly, we highlighted a metabolic syndrome (energy metabolism disruption), moreover our analyses have shown the deleterious impact of confounding physiological factors on diabetes biomarker discovery –age and BMI-. We finally evaluated their influence on a cohort of 227 CEA employees. Lipidomic fingerprints are robust, however BMI impact is marked for neutral lipids. Gender effect shows significant male catabolism and age altered enzyme activities. These studies combine an overall metabolomics and lipidomics analyses of the same human samples. They aim to build up a relational database including spectral and biological data for biomarker characterization in clinical studies

En la presente tesis se ha investigado el potencial de la cromatografía líquida acoplada a la espectrometría de masas de alta resolución en aproximaciones analíticas tanto de tipo target (análisis de compuestos diana) como non-target (análisis no dirigido). Las ventajas que este tipo de instrumentación ofrece, derivadas de la adquisición del espectro completo con alta resolución y medidas de masa exacta, la hacen idónea para el screening de contaminantes orgánicos así como para estudios metabolómicos tanto de denominación de origen de alimentos como de carácter biomédico. El trabajo presentado se ha estructurado en tres grandes bloques. En el primero de ellos se aborda el desarrollo y optimización de un método de screening “universal” de contaminantes orgánicos usando un acoplamiento instrumental LC-QTOF MS. Con este fin, se ha desarrollado una base de datos de compuestos, se ha optimizado la metodología analítica y, finalmente, se ha validado dicha metodología en matrices medioambientales. El segundo bloque muestra el potencial de la técnica para la investigación y diagnosis en exhalados pulmonares, los cuales presentan un bajísima concentración de metabolitos. En el tercer y último bloque, se lleva a cabo la búsqueda de compuestos que puedan utilizarse como marcadores de D.O. en muestras de naranja y vino, centrándose especialmente en aquellas pertenecientes a la Comunidad Valenciana. El primer bloque se inicia con la optimización de los parámetros instrumentales que puedan tener un efecto notorio en la creación de la base de datos y/o en el análisis de las muestras. Para ello, se han investigado los factores que afectan a la sensibilidad instrumental y a la exactitud de masa. Además, se estudia la fragmentación de los compuestos incluidos en la base de datos con el fin de facilitar de forma automática y simultánea la confirmación de la identidad de los compuestos detectados en las muestras. Finalmente se ha evaluado la influencia de la resolución de masa y cromatográfica en la exactitud de masa en el caso de matrices complejas. Posteriormente, se han creado dos bases de datos de contaminantes orgánicos: una teórica, la cual contiene alrededor de 1000 contaminantes orgánicos reportados en la literatura como analizables por LC-MS. Esta base de datos incluye tiempo de retención y fórmula empírica de fragmentos y aductos de aquellos compuestos de los que se dispone de patrones de referencia y que han sido inyectados de acuerdo con los parámetros previamente optimizados. La segunda base de datos se trata de una librería de espectros experimental que contiene los espectros a alta y baja energía de colisión (estos últimos proporcionando información sobre la fragmentación), así como tiempos de retención, de los compuestos con patrón disponible. Finalmente, se ha evaluado la eficacia de la librería en el análisis de muestras reales. En un segundo trabajo se evalúa el potencial de los dos principales procedimientos para la investigación de contaminantes orgánicos mediante métodos de screening: non-target y post-target. Para ello, se lleva a cabo un screening de muestras ambientales, alimentos y muestras de interés toxicológico mediante las dos metodologías y se comparan los resultados obtenidos. La primera aproximación (non-target), basada en la deconvolución del cromatograma para la búsqueda de componentes en la muestra, ha demostrado una notable dependencia de la intensidad del pico cromatográfico debido a la baja eficiencia del algoritmo de deconvolución. En el caso de la aproximación non-target, tras la búsqueda de los componentes, los espectros correspondientes son automáticamente comparados con la librería de espectros experimental creada anteriormente, así como con una librería teórica generada a partir de los compuestos incluidos en la base de datos de contaminantes. En cuanto a la segunda aproximación, tipo post-target, esta se basa en la búsqueda, después de la inyección de las muestras, de compuestos seleccionados (target) incluidos en la base de datos. Como se ha indicado anteriormente, ésta incluye información de fragmentación y tiempo de retención de aquellos compuestos con patrón de referencia disponible, es decir aquellos que están también presentes en la librería experimental de espectros. A la vista de los resultados, se concluye que la aproximación post-target resulta la más ventajosa para abordar un screening “universal” de un elevado número de compuestos. Además, los procesos de revisión de datos y los tiempos de procesamiento se reducen considerablemente. Sin embargo, la metodología non-target presenta una excelente capacidad de confirmación de la identidad de los contaminantes encontrados ya que facilita la comparación de los espectros de fragmentación de patrones con los obtenidos en la muestra. Mediante la aproximación post-target, se encontró un importante número de contaminantes en muestras ambientales y alimentarias, así como drogas de abuso y fármacos en las muestras de orina de voluntarios en tratamientos de desintoxicación. El primer bloque de la tesis finaliza con una validación cualitativa de la metodología desarrollada en muestras de agua subterránea, superficial y el efluente de una planta de tratamiento de aguas residuales. Se han evaluado dos tipos de relleno en la extracción en fase sólida aplicada a las muestras: Oasis HLB y MCX. El primero ha resultado más genérico, perdiéndose únicamente el fármaco Gabapentina en dicho proceso de preconcentración. Para la validación cualitativa, se fortifican 3 muestras independientes de cada tipo de agua analizada a dos niveles de concentración (0.1 y 1 µg/L) y se comprueba la capacidad del método para detectar (típicamente, usando la molécula protonada) e identificar (mediante al menos dos iones: molécula protonada y un fragmento) los contaminantes seleccionados como modelo: 146 compuestos entre los que se incluyen 52 pesticidas, 52 medicamentos (21 antibióticos), 13 drogas de abuso, 11 hormonas, 11 micotoxinas y 7 agentes de protección UV. El método desarrollado permite la detección de la gran mayoría de los compuestos ensayados y la identificación de un buen número de ellos. Posteriormente, se ha aplicado dicha metodología al análisis de muestras reales, identificando varios de los contaminantes seleccionados, incluso a niveles de concentración inferiores al más bajo validado. También ha sido posible la detección e identificación tentativa de varios contaminantes no incluidos en la validación del método, incluso sin patrón de referencia disponible, gracias a la valiosa información suministrada por el analizador QTOF-MS. En el segundo bloque se aprovechan las ventajas que ofrece el acoplamiento LC-MS, y en especial HRMS, para aplicaciones metabolómicas que requieren de una elevada sensibilidad instrumental. Este es el caso de los condensados de exhalados pulmonares, cuyas concentraciones de metabolitos son extremadamente bajas y, por tanto, requieren de técnicas más sensibles como son las LC-HRMS frente al típico análisis mediante NMR. En el trabajo realizado, en colaboración con el Instituto de Estudios Biofuncionales de la Universidad Complutense de Madrid (UCM, Madrid, Spain) y llevado a cabo en el Department of Biomolecular Medicine en Imperial College London (London, UK), se realiza un estudio preliminar sobre las capacidades de LC-HRMS y NMR para abordar la diferenciación de pacientes sanos de aquellos con enfermedades respiratorias, concretamente con obstrucción pulmonar crónica, mediante el análisis no invasivo de condensados de exhalados pulmonares. En base a los resultados obtenidos, se propone el uso de LC-HRMS como aproximación metabolómica estándar para este tipo de análisis. En el tercer y último bloque se investigan los marcadores que permiten la diferenciación de alimentos según su origen, especialmente en productos de interés para la Comunidad Valenciana y cuya calidad está directamente relacionada con su D.O. El primero de los trabajos incluidos en este capítulo se centra en el análisis de las diferencias a nivel químico presentes en naranjas de diferentes orígenes. Para ello, se han seleccionado muestras de la Comunidad Valenciana y de países del hemisferio sur (Sudáfrica y Argentina) de variedades de maduración tardía. Las muestras completas (piel y pulpa) se trituran, homogenizan y extraen con una mezcla agua:metanol, se diluyen con agua y finalmente se analizan mediante UHPLC-HRMS. Posteriormente, se procesan los datos mediante análisis multivariante para encontrar los que permitirían la distinción en función del origen. Finalmente, se ha ampliado el muestreo a la siguiente campaña, incluyendo también muestras de Brasil, con la finalidad de comprobar la validez de los marcadores encontrados y su posible aplicación a otros destinos. Se ha encontrado un marcador idóneo en ambos casos (primer y segundo muestreo) que corresponde a un compuesto de tiempo de retención 4.83 minutos, que finalmente y gracias a la información espectral en masa exacta ofrecida por LC-HRMS se ha identificado como Citrusin D. Finalmente, en el último trabajo presentado se ha realizado un estudio similar al de las naranjas. El vino es un producto muy preciado cuyo valor está extremadamente ligado a su origen, entre otros. En este artículo científico se realiza un comparación interlaboratorio para el estudio metabolómico de muestras de vino procedentes de tres importantes denominaciones de origen en España (Ribera del Duero, Rioja y Penedés). Los resultados de ambas plataformas ofrecen una clasificación óptima de las muestras en base a marcadores definidos en las rutas metabólicas de la vitis vinífera. Además, se discute sobre las ventajas y desventajas de ambas plataformas, no solo en el aspecto instrumental (TOF vs Orbitrap) sino también en el procesamiento de los datos y la selección de los marcadores. Como cabía esperar por ser geográficamente la más diversa, la denomiancion de origen Penedés era separada de las otras dos mediante un simple análisis no-supervisado PCA. Para separar completamente las 3 D.O. se analizaron mediante PLS-DA obteniéndose una clasificación correcta para todas las muestras. Se encontraron diversos marcadores de la familia de los polifenoles como la Catequina, Epicatequina y Galactocatequina, entre otros. Finalmente, los marcadores de cada plataforma fueron cuantificados en un modo target en la otra plataforma. Se demostró que, para ambas plataformas, los diferentes marcadores eran significativos y que por tanto, el tratamiento de datos había filtrado estos marcadores. El modelo estadístico aplicado haciendo uso exclusivamente de los marcadores fue capaz de separar perfectamente las diferentes denominaciones de origen mediante PCA.

Développer une agriculture durable et respectueuse de l’environnement, implique l’utilisation de biostimulants tels que les extraits de macro-algues marines dans le but d’améliorer la croissance des plantes ainsi que leur tolérance aux stress biotiques et abiotiques. Ces extraits commerciaux d’algues sont utilisés en agriculture afin de favoriser la nutrition des plantes, d’améliorer leur qualité nutritionnelle et d’accroitre leur rendement. Dans ce domaine quelques modes d’action ont été élucidés par le centre R&D des Laboratoires Goëmar-Arysta. Cependant, jusqu’à présent, les matières actives n’ont pas été identifiées via une approche classique de fractionnement bio-guidé. De ce fait, leurs mécanismes d’action restent non élucidés. L’objectif premier de ce projet de thèse était d’identifier ces molécules biostimulantes via une approche de fractionnement assistée par la métabolomique, réalisée sur des extraits d’algues commerciaux. Les analyses RMN et LC-MS réalisées sur ces extraits se sont révélées infructueuses dans l’identification de molécules candidates. Ainsi, un classique fractionnement bio-guidé a conduit à la purification d’une fraction favorisant la croissance des plantes. Les analyses U-HPLC-HR-MS réalisées sur cette fraction et ses sous-fractions ont permis d’identifier deux molécules candidates. Un procédé de fractionnement utilisé au cours de ce travail fait l’objet d’une procédure de dépôt de brevet, afin d’apporter une valeur ajoutée à ces extraits biostimulants et de valoriser de nouveaux produits. Le deuxième objectif de ce projet, était d’étudier les réponses physiologiques de la plante modèle Arabidopsis thaliana à l’aide d’analyse transcriptomique. Ceci afin d’élucider les voies métaboliques régulées suite à l’application d’un extrait d’algue produit par Goëmar et d’une fraction stimulante de croissance purifiée au cours de ce projet. L’analyse du transcriptome d’Arabidopsis thaliana révèle la régulation de voies métaboliques complétement différentes par l’extrait d’algues en comparaison de celles régulées par sa fraction purifiée. De plus, les gènes dérégulés par la fraction purifiée constituent des biomarqueurs potentiels de croissance chez les plantes qui pourront être utilisés pour assister l’isolement bio-guidé de molécules candidates. Finalement, ces deux approches combinant fractionnement bio-guidé et analyses métabolomiques sur l’extrait d’Ascophyllum nodosum ainsi que les analyses transcriptomiques réalisées apportent de nouvelles connaissances sur les structures et les modes d’action de molécules candidates
To further develop a sustainable agriculture, new bio-solutions include the use of biostimulants such as seaweed aqueous extracts to improve plant growth or/and alleviate the effect of biotic and abiotic stress. These commercial products aim to improve plant nutrition, in order to impact yield and quality parameters. In this domain, some modes of action have been proposed by the Goëmar-Arysta R&D center. However, the bioactive ingredients have not been identified so far, using classical methods of bioassay-guided fractionation. Therefore, their mechanisms of action remain also elusive. The aim of this thesis project was first to identify, using a strategy of metabolomic profiling of seaweed extracts, the bioactive compounds responsible for plant growth stimulation. The 1H-NMR-based profiling and LC-MS metabolomic analyses of commercial seaweed extracts were not suitable to identify candidate molecules that promote plant growth. A classical bioassay-guided fractionation achieved on a Goëmar extract provided a growth promoting purified fraction and further bioactive sub-fractions. The U-HPLC-HR-MS analyses of these sub-fractions highlighted two candidate molecules. A fractionation process used in this work should be patented in order to improve added-value of growth-promoting filtrate and valorize new by-products. In parallel, the physiological effects of these seaweed extracts were studied in the model plant Arabidopsis thaliana through transcriptomic approaches in order to decipher patterns of gene regulation in response to a crude commercial extract and its purified fraction. The transcriptome in response to the application of seaweed extract was completely different of those obtained using its purified fraction. Genes dysregulated by this purified fraction provided potential biomarkers of plant growth that could be used. to assist the bioactive molecule isolation. Finally these two approaches combining, metabolomics-guided and bioassay-guided fractionation of extracts from the brown seaweed Ascophyllum nodosum, and global transcriptomics in Arabidopsis provided several new insights into the nature and structure of different molecules that trigger different physiological responses in plants

Metabolomics as a field has been used to track changes and perturbations in the human body by investigating metabolite profiles indicating the change of metabolite levels over time and in response to different challenges. In this thesis work, the main focus was on applying multiplatform-metabolomics to study the human metabolome following exposure to perturbations, such as diet (in the form of a challenge meal) and exhaust emissions (air pollution exposure in a controlled setting). The cutting-edge analytical platforms used for this purpose were nuclear magnetic resonance (NMR), as well as gas chromatography (GC) and liquid chromatography (LC) coupled to mass spectrometry (MS). Each platform offered unique characterization features, allowing detection and identification of a specific range of metabolites. The use of multiplatform-metabolomics was found to enhance the metabolome coverage and to provide complementary findings that enabled a better understanding of the biochemical processes reflected by the metabolite profiles. Using non-targeted analysis, a wide range of unknown metabolites in plasma were identified during the postprandial stage after a well-defined challenge meal (in Paper I). In addition, a considerable number of metabolites were detected and identified in lung lavage fluid after biodiesel exhaust exposure compared to filtered air exposure (in Paper II). In parallel, using targeted analysis, both lung lavage and plasma fatty acid metabolites were detected and quantified in response to filtered air and biodiesel exhaust exposure (in Paper III and IV). Data processing of raw data followed by data analysis, using both univariate and multivariate methods, enabled changes occurring in metabolites levels to be screened and investigated. For the initial pilot postprandial study, the aim was to investigate the plasma metabolome response after a well-defined meal during the postprandial stage for two types of diet. It was found that independent of the background diet type, levels of metabolites returned to their baseline levels after three hours. This finding was taken into consideration for the biodiesel exhaust exposures studies, designed to limit the impact of dietary effects. Both targeted and non-targeted approaches resulted in important findings. For instance, different metabolite profiles were detected in bronchial wash (BW) compared to bronchoalveolar lavage (BAL) fluid with mainly NMR and LC-MS. Furthermore, biodiesel exhaust exposure resulted in different metabolite profiles as observed by GC-MS, especially in BAL. In addition, fatty acid metabolites in BW, BAL, and plasma were shown to be responsive to biodiesel exhaust exposure, as measured by a targeted LC-MS/MS protocol. In summary, the new analytical methods developed to investigate the responsiveness of the human plasma and lung lavage metabolome proved to be useful in an analytical perspective, and provided important biological findings. However, further studies are needed to validate these results.
Metabolomik har använts för att spåra förändringar och störningar i kroppens funktioner genom undersökning av metabolit-profiler. I detta avhandlingasarbete har huvudfokus varit på tillämpning av flera olika analytiska plattformar för metabolomikstudier av det mänskliga metabolomet efter exponering för olika kost och avgasutsläpp från biodieselbränsle. De sofistikerade analytiska plattformarna som användes för detta ändamål var kärnmagnetisk resonans (NMR), samt gaskromatografi (GC) och vätskekromatografi (LC) kopplat till masspektrometri (MS). Varje plattform erbjöd unika karakteriseringsmöjligheter med detektion och identifiering av specifika grupper av metaboliter. Användningen av multipattformmetabolomik förbättrade täckningen av metabolomet och genererade kompletterande resultat som möjliggjorde en bättre förståelse av de biokemiska processer som reflekteras av metabolitprofilerna. Med hjälp av breda analyser har ett stort antal okända metaboliter i plasma identifierats under den postprandial fasen efter en väldefinerad måltid (i Paper I). Dessutom har ett stort antal metaboliter påvisats och identifierats i lungsköljvätska efter exponering av biodieselavgaser jämfört med kontollexponering med filtrerad luft (i Paper II). Parallellt med dessa breda analyser har också riktade analyser genomförts av både lungsköljvätska och plasma. Därigenom har bioaktiva lipider detekterats och kvantifieras efter avgasexponering och resultaten har jämförts med filtrerad luft som kontrollexponering (Paper III och IV). Processning av rådata följt av dataanalys, med både univariata och multivariata metoder möjliggjorde screening och fördjupad undersökning av förändringen i metabolitnivåer. I den första pilotstudien av postprandiala nivåer var syftet att undersöka responsen i plasmametabolomet efter en väldefinierad måltid under den postprandiala fasen vid två olika typer av kost. Resultaten visade att oberoende av kosten, så återvände metabolitnivåerna till sina baslinjenivåer tre timmar efter måltiden. Detta togs i beaktande vid exponeringsstudierna för biodieselavgaser, som designades så att dietens inverkan minimerades. Både breda och riktade analyser resulterade i viktiga resultat. Exempelvis så detekterades olika metabolitprofiler i bronkiell sköljvätska (BW) jämfört med bronkoalveolär sköljvätska (BAL), speciellt med NMR och LC-MS. Dessutom resulterade avgasexponering i förändrade metabolitprofiler, observerade med GC-MS, särskilt i BAL. Dessutom uppvisade fettsyrametaboliter i BW, BAL och plasma förändrade halter efter avgasexponering, uppmätt genom en riktad LC-MS/MS-analys. Sammanfattningsvis så visade sig de nya metoderna som utvecklats för att undersöka  förändringar i metabolithalterna i plasma och lungsköljvätska fungera väl ur ett analytiskt perspektiv och resulterade i viktiga biologiska fynd. Fördjupade studier behövs dock för att validera resultaten.

This thesis aimed to investigate several fundamental and perplexing questions relating to the phloem loading and transport mechanisms of Cucurbita maxima, by combining metabolomic analysis with cell biological techniques. This putative symplastic loading species has long been used for experiments on phloem anatomy, phloem biochemistry, phloem transport physiology and phloem signalling. Symplastic loading species have been proposed to use a polymer trapping mechanism to accumulate RFO (raffinose family oligosaccharides) sugars to build up high osmotic pressure in minor veins which sustains a concentration gradient that drives mass flow. However, extensive evidence indicating a low sugar concentration in their phloem exudates is a long-known problem that conflicts with this hypothesis. Previous metabolomic analysis shows the concentration of many small molecules in phloem exudates is higher than that of leaf tissues, which indicates an active apoplastic loading step. Therefore, in the view of the phloem metabolome, a symplastic loading mechanism cannot explain how small molecules other than RFO sugars are loaded into phloem.

Most studies of phloem physiology using cucurbits have neglected the possible functions of vascular architecture in phloem transport. It is well known that there are two phloem systems in cucurbits with distinctly different anatomical features: central phloem and extrafascicular phloem. However, mistaken conclusions on sources of cucurbit phloem exudation from previous reports have hindered consideration of the idea that there may be important differences between these two phloem systems.

The major results are summarized as below:
1) O-linked glycans in C.maxima were structurally identified as beta-1,3 linked glucose polymers, and the composition of glycans in cucurbits was found to be species-specific. Inter-species grafting experiments proved that these glycans are phloem mobile and transported uni-directionally from scion to stock.
2) As indicated by stable isotopic labelling experiments, a considerable amount of carbon is incorporated into small metabolites in phloem exudates. However, the incorporation of carbon into RFO sugars is much faster than for other metabolites.
3) Both CO2 labelling experiments and comparative metabolomic analysis of phloem exudates and leaf tissues indicated that metabolic processes other than RFO sugar metabolism play an important role in cucurbit phloem physiology.
4) The underlying assumption that the central phloem of cucurbits continuously releases exudates after physical incision was proved wrong by rigorous experiments including direct observation by normal microscopy and combined multiple-microscopic methods. Errors in previous experimental confirmation of phloem exudation in cucurbits are critically discussed.
5) Extrafascicular phloem was proved to be functional, as indicated by phloem-mobile carboxyfluorescein tracer studies. Commissural sieve tubes interconnect phloem bundles into a complete super-symplastic network.
6) Extrafascicular phloem represents the main source of exudates following physical incision. The major transported metabolites by these extrafacicular phloem are non-sugar compounds including amino acids, O-glycans, amines.
7) Central phloem contains almost exclusively RFO sugars, the estimated amount of which is up to 1 to 2 molar. The major RFO sugar present in central phloem is stachyose.
8) Cucurbits utilize two structurally different phloem systems for transporting different group of metabolites (RFO sugars and non-RFO sugar compounds). This implies that cucurbits may use spatially separated loading mechanisms (apoplastic loading for extrafascicular phloem and symplastic loading for central phloem) for supply of nutrients to sinks.
9) Along the transport systems, RFO sugars were mainly distributed within central phloem tissues. There were only small amounts of RFO sugars present in xylem tissues (millimolar range) and trace amounts of RFO sugars in cortex and pith. The composition of small molecules in external central phloem is very different from that in internal central phloem.
10) Aggregated P-proteins were manually dissected from central phloem and analysed by both SDS-PAGE and mass spectrometry. Partial sequences of peptides were obtained by QTOF de novo sequencing from trypsin digests of three SDS-PAGE bands. None of these partial sequences shows significant homology to known cucurbit phloem proteins or other plant proteins. This proves that these central phloem proteins are a completely new group of proteins different from those in extrafascicular phloem. The extensively analysed P-proteins reported in literature to date are therefore now shown to arise from extrafascicular phloem and not central phloem, and therefore do not appear to be involved in the occlusion processes in central phloem.
Phloem transportiert ein ausgedehntes Spektrum an Molekülen zwischen Pflanzenorganen, um Wachstum und Entwicklung zu koordinieren. Folglich ist eine umfassende und unvoreingenommene Metabolom-Analyse notwendig, um unser Verständnis über den Transport von Stoffwechselprodukten sowie über Phloemtransport zu vertiefen. Phloemexsudate von Kürbispflanzen werden unter Verwendung der Metabolom-Analyse analysiert. Bei diesen Pflanzen wird angenommen, dass sie symplastische Beladungswege verwenden, um Photoassmilate als Ausgangsschritt des Phloemtransportes zu konzentrieren. Zwei neue Familien Callose-verwandter Substanzen, 1,3-Overknüpfte Glycane, sowie eine Reihe anderer kleinerer Metabolite werden in den Phloemexsudaten detektiert. Metabolom-Daten und physiologische Experimente widersprechen früher berichtetem Verständnis des Phloemexsudationsprozesses in Kürbispflanzen. Folglich bestätigt sich der Phloemexsudationsprozeß durch Kombination unterschiedlicher mikroskopischer Techniken. Kürbispflanzen besitzen zwei Phloemsysteme mit eindeutigen anatomischen Eigenschaften. Es zeigt sich, daß Phloemexsudate in Kürbissen hauptsächlich vom extrafaszikulären Phloem, nicht vom zentralen Phloem, stammen. In den letzten Jahrzehnten wurde gewöhnlich mißverstanden, daß Phloemexsudate vom zentralen Phloem stammen. Die eindeutigen metabolischen Profile der unterschiedlichen Phloemsysteme, die durch Metabolom-Analysen in der räumlichen Auflösung beobachtet werden, bestätigen die unterschiedlichen physiologischen Funktionen der zwei unterschiedlichen Phloemsysteme: das zentrale Phloem transportiert hauptsächlich Zucker, während das extrafaszikuläre Phloem ein ausgedehntes Spektrum von Metaboliten transportiert. Es kann auch ein unterschiedliches metabolisches Profil kleiner Moleküle zwischen internem und externem zentralem Phloem beobachtet werden. Von Strukturproteinen des zentralen Phloems wurden auch Proben genommen und mittels Massenspektrometrie analysiert. Diese Proteine erweisen sich als neuartige Proteine, die sich zu denen im extrafaszikulären Phloem unterscheiden. Dies bestätigt ferner den Funktionsunterschied der unterschiedlichen Phloemsysteme in Kürbispflanzen. Basierend auf diesen neuartigen Entdeckungen des Phloem-Metaboloms und dem vorhergehenden Wissen über den Phloemtransport in Kürbispflanzen, wird ein neues Modell vorgeschlagen, um den Mechanismus des Phloemtransports in der symplastischen Beladung zu verstehen.

In dieser Doktorarbeit werden drei Metabolomics-Studien der KORA Kohorte behandelt. Das Ziel dieser Doktorarbeit war, ein besseres Verständnis der Rolle des Metabolismus von komplexen Phänotypen anhand von Unterschieden im Blutbild, des Geschlechts und anhand von Veränderungen des Metabolitenprofils bei multifaktoriellen Krankheiten wie Typ 2 Diabetes mellitus zu erhalten. Um Artefakte auszuschließen wurden strikte Qualitätskontrollen aller gemessenen Metaboliten durchgeführt. Durch die Analyse von Blutplasma und -serum von 377 Personen konnten wir zeigen, dass die Konzentrationen der Metaboliten in Blutplasma und -serum stark korrelieren und darüber hinaus eine hohe Reproduktionsrate zeigen, bei der Blutplasma besser abschneidet. Im Gegensatz dazu zeigt das Blutserum höhere Metabolitenkonzentrationen und könnte deswegen besser für den Nachweis von Konzentrationsunterschieden geeignet sein. Ein weiteres Ergebnis dieser Doktorarbeit war der Nachweis von signifikanten geschlechtsspezifischen Unterschieden der Konzentrationen von 102 der ausgewerteten 131 Metaboliten. Dabei wurden die Daten von mehr als 3300 Personen der KORA Kohorte verwendet und die Analysen einer konservativen Bonferroni-Korrektur unterzogen. Darüber hinaus identifizierten wir potentielle Biomarker für Prä-Diabetes durch die Analyse von 140 Metaboliten in nüchtern abgegebenen Blutseren von 4297 Personen der KORA Kohorte. Wir konnten zeigen, dass Personen mit gestörter Glukosetoleranz (IGT) signifikant unterschiedliche Konzentrationen von drei Metaboliten (Glycin, lysoPhosphatidylcholine (LPC) 18:2 und acetylcarnitine) im Vergleich zu gesunden Personen aufweisen. Darüber hinaus konnten wir nachweisen, dass geringere Konzentrationen der Metaboliten Glycin und LPC bei Probanden mit Typ 2 Diabetes oder IGT vorhanden sind. Die in dieser Studie identifizierten Metaboliten sind biologisch unabhängig von zuvor entdeckten Diabetes Risikofaktoren. Durch weitere Analysen und die Einbeziehung systembiologischer Ansätze entdeckten wir sieben Diabetesrisiko Susseptibilitätsgene, welche durch Expressionsdaten bestätigt wurden. Metabolomics welches auf der Analyse von Stoffwechselzwischen- und Endprodukten basiert, ist eine wertvolle Methode besonders in der biomedizinischen Forschung, um Krankheitsmechanismen aufzuklären. Nachdem angemessene Qualitätskontrollen etabliert und der Einfluss von komplexen Störfaktoren (z.B. das Geschlecht) aufgeklärt wurden, konnte der Zusammenhang zwischen Krankheit und Metabolismus weiter an Klarheit gewinnen. Die Entdeckungen in unserer T2D Studie zeigen, dass die Analyse von Konzentrationsprofilen helfen kann neue Krankheitsrisikomarker genauso wie neue Wirkungspfade zu identifizieren, die möglicherweise das Ziel zur Heilung einer Krankheit sein könnten.

Epigenetics and metabolomics are rapidly growing areas of research, in part due to recent advances in technology that have allowed for a wide coverage of the human genome. Metabolites are small compounds present in cell and body fluids, and are involved in biochemical processes of the cell. Quantitative trait loci associated with levels of individual metabolites (mQTLs) have been identified from numerous metabolome GWAS. Here, I analysed metabolite levels in twins with the aim of identifying genetic variants that influence metabolomic traits (mQTLs) using two different metabolomics platforms, and consequently compared the results to report stable metabolites on both technologies to ultimately enable combining metabolite profiles across these two platforms. DNA methylation is a biochemical process that is vital for mammalian development. It is present throughout the genome and is the most extensively studied epigenetic mark. Previous studies have explored the heritability of DNA methylation and have identified methylation QTLs (meQTL). Here, I identified meQTLs with the goal of assesing the evidence of genetic effects influence not only DNA methylation levels, but also variability by using MZ-twin discordance as a measure of variance. Epigenetic mechanisms and metabolomic profiles have both been shown to play a role in gene expression and susceptibility for complex human disease. Here, I analysed the association between type 2 diabetes and metabolomic and epigenetic datasets and combined the data to identify connections between these levels of biological data at genetic variants linked to type 2 diabetes to gain more insight into the disease susceptibility and progression. Overall, the results confirmed previous findings of strong genetic influences on metabolites and extend current knowledge about genetic effects underlying several biochemical pathways. Additionally, the results also showed genetic influences on DNA methylation, and give insights into mechanisms by which genetic impacts epigenetic processes. Lastly, the findings show that specific genetic susceptibility variants for type 2 diabetes can also impact epigenetic and metabolomics profiles, and can help improve our understanding of the disease etiology.

Metabolomics is the study of metabolites and metabolic processes. Due to the diversity of structures and polarities of metabolites, no single analytical technique is able to measure the entire metabolome - instead a varied set of experimental designs and instrumental technologies are used to measure specific portions. This has led to the development of many distinct data analysis and processing methods and software. There is hope that metabolomics can be utilized for clinical applications, in toxicology and to measure the exposome. However, for these applications to be realised data must be high quality, sufficiently standardised and annotated, and FAIR (Findable, Accessible, Interoperable and Reproducible). For this purpose, it is also important that standardised, FAIR software workflows are available. There has also recently been much concern over the reproducibility of scientific research, which FAIR and open data, and workflows can help to address. To this end, this thesis aims to assess current practices and standards of sharing data within the field of metabolomics, using metascientific approaches. The types of functions of software for processing and analysing metabolomics data is also assessed. Reporting standards are designed to ensure that the minimum information required to un- derstand and interpret the results of analysis are reported. However, poor reporting standards are ignored and not complied with. Compliance to the biological context Metabolomics Standards Initiative (MSI) guidelines was examined, in order to investigate their timeliness. The state of open data within the metabolomics community was examined by investigating how much publicly available metabolomics data there is and where has it been deposited. To explore whether journal data sharing policies are driving open metabolomics data, which journals publish articles that have their underlying data made open was also examined. However, open data alone is not inherently useful: if data is incomplete, lacking in quality or missing crucial metadata, it is not valuable. Conversely, if data are reused, this can demonstrate the worth of public data archiving. Levels of reuse of public metabolomics data were therefore examined. With greater than 250 software tools specific for metabolomics, practitioners are faced with a daunting task to select the best tools for data collection and analysis. To help educate researchers about what software is available, a taxonomy of metabolomics software tools and a GitHub pages wiki, which provides extensive details about all included software, have been developed.

The stream of new technological advancements and their integration into the field of microbiology have contributed significantly towards our understanding of life in the micro-scale world, making the fields of microbiology and biotechnology shine like never before. Since 1980, the recombinant protein-based therapeutics industry has become one of the fastest growing sectors in the biopharmaceutical market. Nearly 30% of commercially available recombinant proteins are produced in Escherichia coli, making this species one of the most commonly used bacterial expression systems for the production of recombinant biotherapeutics. However, when it comes to the production of enzymes and bioactive secondary metabolites (antibiotic, antifungal, antiviral and immunosuppressant), Streptomyces species remain the major producer within this sector. Meeting the high demand for such products requires a clear and in-depth understanding of the bioprocesses involved to achieve high yield and quality products, whilst keeping the process industrially attractive. It is generally accepted that the metabolome, as a down-stream process to the genome and proteome, may provide a clearer picture of a biological system. Thus, in this thesis a series of metabolomics approaches were adopted to obtain a deeper insight into the metabolic effects of recombinant protein production in E. coli and Streptomyces lividans. Furthermore, a Geobacter-based biomagnetite nanoparticle production system which displayed a prolonged lag phase upon scale-up was investigated by employing metabolic profiling and fingerprinting approaches combined with multivariate analysis strategies, to identify growth-limiting metabolites. The results of this analysis identified nicotinamide as the growth limiting metabolite. Nicotinamide-feeding experiments confirmed the above findings, leading to improved biomass yield whilst restoring the lag phase to bench-scale level. Raman and Fourier transform infrared spectroscopies combined with stable isotopic probing strategies were also employed to demonstrate the application of metabolic fingerprinting in providing detailed biochemical information for quantitative characterisation and differentiation of E. coli cells at community and single-cell levels. The single-cell approach proved promising, offering detailed biochemical information and perhaps accompanying other cultivation-free approaches such as metagenomics for further future investigations. It is hoped that the advances made in these studies have proved the potential applications of metabolomics strategies to aid the optimisation of microbially-driven bioprocesses.

Tato diplomová práce se zabývá rekonstrukcí dynamického modelu produkce polyhydroxyalkanoátů (PHA) termofilní bakterií Schlegelella thermodepolymerans. První kapitola poskytuje čtenářům krátký úvod do systémové biologie a matematické teorie grafů. Na ni navazuje druhá kapitola zabývající se různými přístupy v dynamickém modelování, včetně běžně používaných nástrojů pro dynamickou analýzu komplexních systémů. Třetí kapitola pak sleduje další pojmy a možnosti týkající se analýzy modelu. Následující kapitola se zaměřuje na metabolomiku a často používané laboratorní techniky a pátá kapitola je pak věnována polyhydroxyalkanoátům, zejména jejich chemické struktuře a vlastnostem. V kapitole šesté je navržen obecný booleovský model pro produkci PHA termofilními bakteriemi. Kapitola sedmá se poté zaměřuje na zdokonalení modelu se zaměřením na S. thermodepolymerans. Výsledný dynamický model je podroben analýze a výsledky jsou diskutovány.

With ageing comes a deterioration of metabolic and physiological changes that often manifest themselves as age-related comorbidities. Calorie restriction (CR) is a robust intervention that can prevent and reverse such changes, resulting in reduced ageassociated disease and increased lifespan across a wide range of species. Moreover, a link between the extent of restriction and increased lifespan has also been established. Though widely studied, the mechanisms behind the beneficial effects of CR have yet to be fully understood. Consequently, I investigated metabolomic changes in the liver, plasma, brown adipose tissue (BAT) and cerebellum in five month old male C57BL/6 mice undergoing three months of either 10, 20, 30 or 40% CR, in addition to 12 hour and 24 hour ad libitum fed groups. Behavioural, physiological and molecular data was collected on each individual mouse and I used this information, in addition to my own metabolomic data to determine associations between phenotypic changes with graded CR. My results indicate that increasing CR resulted in greater numbers of significantly differentiated metabolites across all four tissues, and these were related to changes across sphingolipids, carnitines, bile acids, vitamins and amino acids. Metabolic remodelling in the liver indicated a shift from lipogenesis to lipolysis and changes in the plasma indicated an increase in absorption of vitamins from the stomach and colon. Changes in neurotransmitters and their precursors suggested activity and temperature driven BAT activation, in addition to an increase in antioxidant power, this was also seen in the cerebellum where metabolites associated with signalling in the hypothalamus were increased in a graded fashion with CR. In all tissues changes were linked with behaviours that accompany hunger signalling such as increased food anticipatory activity and reduced body temperature. Together, these changes reflect multi-tissue beneficial effects of CR, which may function to alleviate age-related comorbidities.

Kidney Transplantation is the gold standard treatment for patients with end-stage renal failure. Most kidneys used for transplantation are from deceased donors and ensuring successful outcomes depends on many factors. One of these is organ storage. Hypothermic Machine Perfusion (HMP) of deceased donor organs has been shown to have several benefits. However, it has not been widely adopted and the underlying mechanism is poorly understood. The first section of this thesis examines the introduction of HMP into clinical practice. HMP outcomes were similar to those of standard storage techniques but with the additional benefit of increasing safe storage times. This was likely due to inherent benefits of the machine itself, improved recipient preparation and better peri-operative conditions. The second part of this study analysed HMP perfusate using metabolomics (Nuclear Magnetic Resonance) to identify potential predictors of graft outcome. Differences were identified in the metabolic profiles of perfusate from kidneys with immediate and delayed graft function. These may have a future role in viability assessment. Improved understanding of metabolism during storage may help target optimization strategies for deceased donor organs. The final part of this study describes the development of a porcine model of transplantation to test future hypotheses.

Since the beginning of human civilization, medicinal plants have been used to treat a variety of infectious and non-infectious diseases. The therapeutic properties of phytochemicals have been recognized since ancient human history. The genus Helichrysum Mill. with its attractive flowers consist of an estimated 500‒600 species in the Asteraceae family. In South Africa and Namibia there are about 244‒250 species with tremendous morphological diversity. Several Helichrysum species are widely used by the indigenous population to treat various disorders such as wounds, infections, respiratory conditions, headaches, coughs, colds and fevers. Several of the Helichrysum species exhibit antiviral activity with the most relevant to this study being the discovery of anti-human immunodeficiency virus (anti-HIV) and anti-reverse transcriptase (anti-RT) activity of some species. Drug discovery and development, from the early stages of a promising compound to the final medication, is an intensive, expensive and incremental process. The ultimate goal is to identify a molecule with the desired effect in the human body and to establish its quality, safety and efficacy for treating patients. The ability to combine high-throughput analytical techniques like metabolomic and other experimental approaches with drug discovery will speed up the development of safer, more effective and better-targeted therapeutic agents. The rapidly emerging field of metabolomics and molecular docking analysis provides valuable information on drug activity, toxicity, customized drug treatments and can predict therapeutic outcomes. Extraction of the aerial parts of 32 Helichrysum species was done using polar [methanol (MeOH) 50%: distilled water (dH2O) 50%] and non-polar [hexane (Hex), dichloromethane (DCM) and acetone (Ace)] solvent systems. Anti-human immunodeficiency virus bioassays on the live HI virus revealed that polar extracts of H. mimetes and H. chrysargyrum at 2.5 μg/mL and 25 μg/mL, polar and non-polar extracts of H. infuscum at 25 μg/mL and polar and non-polar extracts of H. zeyheri, H. setosum, H. platypterum and H. kraussii at 2.5 and 25 μg/mL, had higher than 90% inhibitory activity. The polar extract of H. mimetes also exhibited reverse transcriptase (RT) inhibition as a possible indication of the mechanism of action. Proton nuclear magnetic resonance (1H NMR) spectra of the polar extracts exhibited the presence of aromatic compounds and carbohydrate moieties. Principal component analysis (PCA) of the polar extracts showed clustering related to the activity of the extracts with good predictability scores (Q2 > 0.5). However, orthogonal projections to latent structures discriminant analysis (OPLS-DA) predictability of the model was low based on the Q2 at approximately 0.25. Quinic acid (QA), isolated from H. mimetes showed promising anti-RT activity [50% inhibition concentration (IC50) = 53.82 μg/mL] which was comparable to the positive drug control, doxorubicin (IC50 = 40.31 μg/mL). The molecular docking study revealed the probable binding site and conformation of QA within cavity 4, with a docking score of -8.03. The docking score of doxorubicin within cavity 4 was -7.87. With this study, it was shown that metabolomic analysis as a tool to predict anti-HIV activity in Helichrysum species can be valuable to shorten the process. Moreover, the study of molecular docking revealed the mechanism action of quinic acid and doxorubicin against RT.
Thesis (PhD)--University of Pretoria, 2019.
Plant Production and Soil Science
PhD
Unrestricted

[ES] En las últimas décadas los avances tecnológicos han tenido como consecuencia la generación de una creciente cantidad de datos en el campo de la biología y la biomedicina. A día de hoy, las así llamadas tecnologías "ómicas", como la genómica, epigenómica, transcriptómica o metabolómica entre otras, producen bases de datos con cientos, miles o incluso millones de variables. El análisis de datos ómicos presenta una serie de complejidades tanto metodoló-gicas como computacionales que han llevado a una revolución en el desarrollo de nuevos métodos estadísticos específicamente diseñados para tratar con este tipo de datos. A estas complejidades metodológicas hay que añadir que, en la mayor parte de los casos, las restricciones logísticas y/o económicas de los proyectos de investigación suelen conllevar que los tamaños muestrales en estas bases de datos con tantas variables sean muy bajos, lo cual no hace sino empeorar las dificultades de análisis, ya que se tienen muchísimas más variables que observaciones. Entre las técnicas desarrolladas para tratar con este tipo de datos podemos encontrar algunas basadas en la penalización de los coeficientes, como lasso o elastic net, otras basadas en técnicas de proyección sobre estructuras latentes como PCA o PLS y otras basadas en árboles o combinaciones de árboles como random forest. Todas estas técnicas funcionan muy bien sobre distintos datos ómicos presentados en forma de matriz (IxJ). Sin embargo, en ocasiones los datos ómicos pueden estar expandidos, por ejemplo, al tomar medidas repetidas en el tiempo sobre los mismos individuos, encontrándonos con estructuras de datos que ya no son matrices, sino arrays tridimensionales o three-way (IxJxK). En estos casos, la mayoría de las técnicas citadas pierden parte de su aplicabilidad, quedando muy pocas opciones viables para el análisis de este tipo de estructuras de datos. Una de las técnicas que sí es útil para el análisis de estructuras three-way es N-PLS, que permite ajustar modelos predictivos razonablemente precisos, así como interpretarlos mediante distintos gráficos. Sin embargo, relacionado con el problema de la escasez de tamaño muestral relativa al desorbitado número de variables, aparece la necesidad de realizar una selección de variables relacionadas con la variable respuesta. Esto es especialmente cierto en el ámbito de la biología y la biomedicina, ya que no solo se quiere poder predecir lo que va a suceder, sino entender por qué sucede, qué variables están implicadas y, a poder ser, no tener que volver a recoger los cientos de miles de variables para realizar una nueva predicción, sino utilizar unas cuantas, las más importantes, para poder diseñar kits predictivos coste/efectivos de utilidad real. Por ello, el objetivo principal de esta tesis es mejorar las técnicas existentes para el análisis de datos ómicos, específicamente las encaminadas a analizar datos three-way, incorporando la capacidad de selección de variables, mejorando la capacidad predictiva y mejorando la interpretabilidad de los resultados obtenidos. Todo ello se implementará además en un paquete de R completamente documentado, que incluirá todas las funciones necesarias para llevar a cabo análisis completos de datos three-way. El trabajo incluido en esta tesis por tanto, consta de una primera parte teórico-conceptual de desarrollo de la idea del algoritmo, así como su puesta a punto, validación y comprobación de su eficacia; de una segunda parte empírico-práctica de comparación de los resultados del algoritmo con otras metodologías de selección de variables existentes, y de una parte adicional de programación y desarrollo de software en la que se presenta todo el desarrollo del paquete de R, su funcionalidad y capacidades de análisis. El desarrollo y validación de la técnica, así como la publicación del paquete de R, ha permitido ampliar las opciones actuales para el análisis
[CAT] En les últimes dècades els avançaments tecnològics han tingut com a conseqüència la generació d'una creixent quantitat de dades en el camp de la biologia i la biomedicina. A dia d'avui, les anomenades tecnologies "òmiques", com la genòmica, epigenòmica, transcriptòmica o metabolòmica entre altres, produeixen bases de dades amb centenars, milers o fins i tot milions de variables. L'anàlisi de dades 'òmiques' presenta una sèrie de complexitats tant metodolò-giques com computacionals que han portat a una revolució en el desenvolupament de nous mètodes estadístics específicament dissenyats per a tractar amb aquest tipus de dades. A aquestes complexitats metodològiques cal afegir que, en la major part dels casos, les restriccions logístiques i / o econòmiques dels projectes de recerca solen comportar que les magnituts de les mostres en aquestes bases de dades amb tantes variables siguen molt baixes, el que no fa sinó empitjorar les dificultats d'anàlisi, ja que es tenen moltíssimes més variables que observacions Entre les tècniques desenvolupades per a tractar amb aquest tipus de dades podem trobar algunes basades en la penalització dels coeficients, com lasso o elastic net, altres basades en tècniques de projecció sobre estructures latents com PCA o PLS i altres basades en arbres o combinacions d'arbres com random forest. Totes aquestes tècniques funcionen molt bé sobre diferents dades 'òmiques' presentats en forma de matriu (IxJ), però, en ocasions les dades òmiques poden estar expandits, per exemple, cuan ni ha mesures repetides en el temps sobre els mateixos individus, trobant-se amb estructures de dades que ja no són matrius, sinó arrays tridimensionals o three-way (IxJxK). En aquestos casos, la majoria de les tècniques mencionades perden tota o bona part de la seua aplicabilitat, quedant molt poques opcions viables per a l'anàlisi d'aquest tipus d'estructures de dades. Una de les tècniques que sí que és útil per a l'anàlisi d'estructures three-way es N-PLS, que permet ajustar models predictius raonablement precisos, així com interpretar-los mitjançant diferents gràfics. No obstant això, relacionat amb el problema de l'escassetat de mostres relativa al desorbitat nombre de variables, apareix la necessitat de realitzar una selecció de variables relacionades amb la variable resposta. Això és especialment cert en l'àmbit de la biologia i la biomedicina, ja que no només es vol poder predir el que va a succeir, sinó entendre per què passa, quines variables estan implicades i, si pot ser, no haver de tornar a recollir els centenars de milers de variables per realitzar una nova predicció, sinó utilitzar unes quantes, les més importants, per poder dissenyar kits predictius cost / efectius d'utilitat real. Per això, l'objectiu principal d'aquesta tesi és millorar les tècniques existents per a l'anàlisi de dades òmiques, específicament les encaminades a analitzar dades three-way, incorporant la capacitat de selecció de variables, millorant la capacitat predictiva i millorant la interpretabilitat dels resultats obtinguts. Tot això s'implementarà a més en un paquet de R completament documentat, que inclourà totes les funcions necessàries per a dur a terme anàlisis completes de dades three-way. El treball inclòs en aquesta tesi per tant, consta d'una primera part teorica-conceptual de desenvolupament de la idea de l'algoritme, així com la seua posada a punt, validació i comprovació de la seua eficàcia, d'una segona part empíric-pràctica de comparació dels resultats de l'algoritme amb altres metodologies de selecció de variables existents i d'una part adicional de programació i desenvolupament de programació en la qual es presenta tot el desenvolupament del paquet de R, la seua funcionalitat i capacitats d'anàlisi. El desenvolupament i validació de la tècnica, així com la publicació del paquet de R, ha permès ampliar les opcions actuals per a l'anàlis
[EN] In the last decades, advances in technology have enabled the gathering of an increasingly amount of data in the field of biology and biomedicine. The so called "-omics" technologies such as genomics, epigenomics, transcriptomics or metabolomics, among others, produce hundreds, thousands or even millions of variables per data set. The analysis of 'omic' data presents different complexities that can be methodological and computational. This has driven a revolution in the development of new statistical methods specifically designed for dealing with these type of data. To this methodological complexities one must add the logistic and economic restrictions usually present in scientific research projects that lead to small sample sizes paired to these wide data sets. This makes the analyses even harder, since there is a problem in having many more variables than observations. Among the methods developed to deal with these type of data there are some based on the penalization of the coefficients, such as lasso or elastic net, others based on projection techniques, such as PCA or PLS, and others based in regression or classification trees and ensemble methods such as random forest. All these techniques work fine when dealing with different 'omic' data in matrix format (IxJ), but sometimes, these IxJ data sets can be expanded by taking, for example, repeated measurements at different time points for each individual, thus having IxJxK data sets that raise more methodological complications to the analyses. These data sets are called three-way data. In this cases, the majority of the cited techniques lose all or a good part of their applicability, leaving very few viable options for the analysis of this type of data structures. One useful tool for analyzing three-way data, when some Y data structure is to be predicted, is N-PLS. N-PLS reduces the inclusion of noise in the models and obtains more robust parameters when compared to PLS while, at the same time, producing easy-to-understand plots. Related to the problem of small sample sizes and exorbitant variable numbers, comes the issue of variable selection. Variable selection is essential for facilitating biological interpretation of the results when analyzing 'omic' data sets. Often, the aim of the study is not only predicting the outcome, but also understanding why it is happening and also what variables are involved. It is also of interest being able to perform new predictions without having to collect all the variables again. Because all of this, the main goal of this thesis is to improve the existing methods for 'omic' data analysis, specifically those for dealing with three-way data, incorporating the ability of variable selection, improving predictive capacity and interpretability of results. All this will be implemented in a fully documented R package, that will include all the necessary functions for performing complete analyses of three-way data. The work included in this thesis consists in a first theoretical-conceptual part where the idea and development of the algorithm takes place, as well as its tuning, validation and assessment of its performance. Then, a second empirical-practical part comes where the algorithm is compared to other variable selection methodologies. Finally, an additional programming and software development part is presented where all the R package development takes place, and its functionality and capabilities are exposed. The development and validation of the technique, as well as the publication of the R package, has opened many future research lines.
Hervás Marín, D. (2019). Use of multivariate statistical methods for the analysis of metabolomic data [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/130847
TESIS

Metabolomics seeks to achieve a comprehensive quantitative analysis of the wide range of metabolites in biological samples. Analytical chemistry, and specifically 1H NMR, can provide spectra detailing the vast array of physio-chemical properties of the sample. The identification of characteristics in the spectra is known as metabolomic fingerprinting. However, the thousands of variables in the 1H NMR spectra, .relating to hundreds of metabolites, make the data sets extremely complex and difficult to interpret. The aim of this research was to develop analytical techniques to classify biological samples, which are able to identify the metabolites responsible for the differences between classes. Novel methods based on genetic programming have been developed which provide models that are able to classify samples at least as well as currently used chemometric pattern recognition approaches. Moreover these techniques have the advantage that they are significantly easier to interpret in terms of the original spectrum. Preprocessing the spectra using wavelet transforms has further increased the interpretability of the models. This adaptive binning method integrates the data in such a way that the bins represent peaks in the original spectra thereby reducing the dimensionality whilst maintaining the information. The preprocessing and genetic programming methods have been combined and developed in response to additio~al problems faced by time resolved metabolomics data, where both intra-individual (time dependent) and inter-individual variation are present. The methods were employed in the analysis of TSE infected sheep and cattle and, in both cases, enabled a diagnostic of the disease to be determined, allowing the specific compounds responsible for the disease to be identified.

Over the past decades, the emerging omics technologies have enabled scientists to take a step further in the investigation of biological systems. From food safety to stratified medicine, omics technologies are now an essential and powerful means to study biological processes. Omics technologies are however at different stages of maturity, and the most recent field of the omics family, metabolomics, is still in its infancy. Metabolomics attempts to catalogue, characterise and quantify all small molecules constitutive of a biological system. Liquid Chromatography - Mass Spectrometry (LCMS) is now the most commonly used technique to generate metabolomics data. The method allows the detection of hundreds of metabolites from a single sample and can provide a rapid assignment of formulae to detected masses using high accuracy mass spectrometers. While analytical methods are well developed, support for linking metabolites to detected features and interpreting the results of a data analysis in a biological context is still poorly developed. Significant challenges also arise from the additional steps required to export the data to third party environments to create a biological context. The study of integrated omics datasets as a single system has also shown to provide greater inferences than the study of each omics separately. Methods to integrate the different omics layers of biological systems are, however, at an early stage of development and no standard approach currently exists to provide a holistic view of organisms systems organisation. The objective of this thesis is to formalise, standardise and unify the data analysis of the metabolomics field, by providing to biologists the tools to support them from planning to analysis to biological impact reporting. The work presented here focuses particularly on untargeted LC-MS metabolomics approaches and attempts to assist non-expert users in performing their own analysis of metabolomics datasets. The project also aims to enable systematic biological interpretation of metabolomics datasets. The first part of the thesis focuses on creating the foundation of a unified environment for LC-MS metabolomics data analysis. Subsequently, the created environment will be expanded to integrate and support the latest technological advances in the field and provide better support for both designing studies and interpreting analysis results in a biological context. Finally, the last part of this thesis concentrates on integrating metabolomics data with other omics datasets in an attempt to provide a holistic view of a biological system.

Direct electrospray ionisation-mass spectrometry (direct ESI-MS), by omitting the chromatographic step, has great potential for application as a high-throughput approach for untargeted urine metabolomics analysis compared to liquid chromatography-mass spectrometry (LC-MS). The rapid development and technical innovations revealed in the field of ambient ionisation MS such as nanoelectrospray ionisation (nanoESI) chip-based infusion and liquid extraction surface analysis mass spectrometry (LESA-MS) suggest that they might be suitable for high-throughput metabolomics analysis. In this thesis, LC-MS and high-throughput direct ESI-MS methods using high resolution orbital trap mass spectrometer were developed and validated for untargeted metabolomics of human urine. Three different direct ESI-MS techniques were explored and compared with LC-MS: flow injection electrospray ionisation-MS (FIE-MS), chip-based infusion and LESA-MS of dried urine spots on a cell culture slide. A high-throughput sample preparation protocol was optimised using in-house artificial urine. Urine samples after consumption of green tea and healthy controls were used as a model to explore the performance and classification ability of the direct ESI-MS. High-throughput data pre-processing and multivariate analysis protocols were established for each method. The developed methods were finally applied for the analysis of clinical urine samples for biomarker discovery and to investigate the metabolic changes in osteoarthritis and malaria. Also, the methods were applied to study the effect of oligofructose diet on the gut microbial community of healthy subjects. The analytical performance of the methods for urine metabolomics was validated using quality control (QC) and principal component analysis (PCA) approaches. Rigorous validation including cross-validation, permutation test, prediction models and area under receiver operating characteristic (ROC) curve (AUC) was performed across the generated datasets using the developed methods. Analysis of green tea urine samples generated 4128, 748, 1064 and 1035 ions from LC-MS, FIE-MS, chip-based infusion and LESA-MS analysis, respectively. A selected set of known green tea metabolites in urine were used to evaluate each method for detection sensitivity. 15 metabolites were found with LC-MS compared to 8, 5 and 6 with FIE-MS, chip-based infusion and LESA, respectively. The developed methods successfully differentiated between the metabolic profiles of osteoarthritis active patients and healthy controls (Q2 0.465 (LC-MS), 0.562 (FIE-MS), 0.472 (chip-based infusion) and 0.493 (LESA-MS)). The altered level of metabolites detected in osteoarthritis patients showed a perturbed activity in TCA cycle, pyruvate metabolism, -oxidation pathway, amino acids and glycerophospholipids metabolism, which may provide evidence of mitochondrial dysfunction, inflammation, oxidative stress, collagen destruction and use of lipolysis as an alternative energy source in the cartilage cells of osteoarthritis patients. FIE-MS, chip-based infusion and LESA-MS increased the analysis throughput and yet they were able to provide 33%, 44% and 44%, respectively, of the LC-MS information, indicating their great potential for diagnostic application in osteoarthritis. Malaria samples datasets generated 9,744 and 576 ions from LC-MS and FIE-MS, respectively. Supervised multivariate analysis using OPLS-DA showed clear separation and clustering of malaria patients from controls in both LC-MS and FIE-MS methods. Cross-validation R2Y and Q2 values obtained by FIE-MS were 0.810 and 0.538, respectively, which are comparable to the values of 0.993 and 0.583 achieved by LC-MS. The sensitivity and specificity were 80% and 77% for LC-MS and FIE-MS, respectively, indicating valid, reliable and comparable results of both methods. With regards to biomarker discovery, altered level of 30 and 17 metabolites were found by LC-MS and FIE-MS, respectively, in the urine of malaria patients compared to healthy controls. Among these metabolites, pipecolic acid, taurine, 1,3-diacetylpropane, N-acetylspermidine and N-acetylputrescine may have the potential of being used as biomarkers of malaria. LC-MS and FIE-MS were able to separate urine samples of healthy subjects on oligofructose diet from controls (specificity/sensitivity 80%/88% (LC-MS) and 71%/64% (FIE-MS)). An altered level of short chain fatty acids (SCFAs), fatty acids and amino acids were observed in urine as a result of oligofructose intake, suggesting an increased population of the health-promoting Bifidobacterium and a decreased Lactobacillus and Enterococcus genera in the colon. In conclusion, the developed direct ESI-MS methods demonstrated the ability to differentiate between inherent types of urine samples in disease and health state. Therefore they are recommended to be used as fast diagnostic tools for clinical urine samples. The developed LC-MS method is necessary when comprehensive biomarker screening is required.

From the Helichrysum genus 600 species occur in Africa of which 244 species are found in South Africa. The most commonly used Helichrysum species for medicinal purposes are H. cymosum, H. odoratissimum, H. petiolare and H. nudifolium. The medicinal uses include the treatment of coughs, colds, fever, infection, headaches, menstrual pain and are very popular for wound dressing. Previous published research has shown that H. aureonitens has antiviral properties against Herpes simplex virus type 1 (HSV-1). In this study, further investigation into the Helichrysum species was undertaken, to establish the active constituents responsible for anti-HSV activity using a metabolomics approach. The cytotoxicity of 12 Helichrysum species was investigated and ranged from <3.125 μg/ml to 277.8 μg/ml on the vero cell line. The 12 Helichrysum species also showed various levels of antiviral activity against HSV, with both the water-methanol and chloroform extracts of H. adenocarpum subsp. adenocarpum being the most active extract at 25 μg/ml. In this study the activity of Helichrysum species against HIV-1 RT was also investigated. Helichrysum populifolium was the most active extract, inhibiting the HIV-1 RT enzyme by 63.78 % at 200 μg/ml. The bioactivity data and the spectral nuclear magnetic resonance (NMR) data of al the Helichrysum species from this study was analysed using the SIMCA-P software to discriminate between the different species on the basis of their bioactivity and chemical composition. The samples did not group well on Principal Component Analysis (PCA) but did separate well using the Orthogonal Projection to Latent Structure – Discriminate Analysis (OPLS-DA) on the basis of their activity and NMR spectra data. From the OPLS scoring plots analysis, contribution plots were created which indicated regions responsible for the difference between the species, with these regions being investigated to identify the bioactive constituents. It was thus possible to use metabolomics to discriminate between samples on the basis of their activity and show that it could probably be used in future as a tool to identify active ingredients in medicinal plants and accelerate drug discovery. Copyright
Dissertation (MSc)--University of Pretoria, 2009.
Plant Science
unrestricted

Crohn’s disease is an inflammatory disorder of the bowel, arising from the dysregulation of intestinal mucosal immunity. The interleukin-10-deficient (IL10-/-) mouse develops intestinal inflammation with similar characteristics to Crohn’s disease, due to the loss of immune tolerance towards intestinal microbiota. Metabolomic analysis is the study of small molecule metabolites, involving the measurement of large numbers of metabolites in biological samples. The aim of the research was to study the urinary metabolite profile of IL10-/- mice by gas chromatography-mass spectrometry metabolomic analysis. A metabolite profile of intestinal inflammation, consisting of 15 metabolites, was discovered by comparing the urinary metabolite profiles of IL10-/- and wildtype C57BL/6 mice. Xanthurenic acid and fucose were identified as the main urinary metabolites associated with the early stage of intestinal inflammation. Their levels were increased in IL10-/- mice relative to wildtype. Xanthurenic acid levels were attributed to increased tryptophan catabolism which produces kynurenine metabolites that may induce immune tolerance of T-cells towards intestinal microbiota. Plasma levels of kynurenine and 3-hydroxykynurenine were confirmed to be elevated in IL10-/- mice. The increased fucose levels may be due to abnormal fucosylation of plasma or intestinal mucosal proteins involved in leukocyte trafficking. Comparisons of the urinary metabolite profiles of IL10-/- and wildtype mice also revealed eleven metabolite differences that were unaffected by inflammation severity in IL10-/- mice. The main metabolites were glutaric acid, 2-hydroxyglutaric acid and 2-hydroxyadipic acid, which were decreased in IL10-/- mice. These eleven metabolite differences may be associated with residual genes from embryonic stem cells of the 129P2 mouse strain used to create the IL10-/- mouse, or novel functions of IL10 that are unrelated to inflammation. The metabolite profile of inflammation was not altered in IL10-/- mice fed with kiwifruit extracts, consistent with other measures of inflammation which showed that intestinal inflammation was not attenuated by the dietary intervention. The urinary levels of some kiwifruit metabolites differed between IL10-/- and wildtype mice, suggesting differences in absorption or intestinal microbial metabolism of these metabolites. Overall, the research demonstrates that metabolomic analysis of IL10-/- mice can identify potential biomarkers of intestinal inflammation and provide new insights into the metabolic effects of IL10-deficiency.

Thesis (Ph. D.) -- University of Maryland, College Park, 2007.
Thesis research directed by: Chemical Engineering. Title from t.p. of PDF. Includes bibliographical references. Published by UMI Dissertation Services, Ann Arbor, Mich. Also available in paper.

Modern, high-throughput techniques for the acquisition of metabolomic data, combined with an increase in computational power, have provided not only the need for, but also the means to develop and use, methods for the interpretation of large and complex datasets. This thesis investigates the methods by which pertinent information can be extracted from nontargeted metabolomic data and reviews the current state of chemometric methods. The analysis of real-world data and research questions relevant to the agri-food industry reveals several problems for which novel solutions are proposed. Three LC-MS datasets are studied: Medicago, Alopecurus and aged Beef, covering stress resistance, herbicide resistance and product misbranding. The new methods include preprocessing (batch correction, data-filtering), processing (clustering, classification) and visualisation and their use facilitated within a flexible data-to-results pipeline. The resulting software suite with a user-friendly graphical interface is presented, providing a pragmatic realisation of these methods in an easy to access workflow.

Coronary heart disease (CHD) is the leading cause of death in the UK. Recent technological advances in metabolomics have the potential to contribute to further the understanding of CHD, especially because they are facilitating the collection of metabolomics data in large observational studies. However, the high dimensionality of this type of information and its strong interdependencies raise several analytical difficulties. These difficulties were investigated, motivated by the study of 228 metabolites acquired from blood samples as part of the British Womens Heart and Health Study (BWHHS). Issues regarding transformations of the metabolomics data and their reliability were examined. Analytical methods typically adopted with high-dimensional data were reviewed, and then a more recently developed method, differential networks, was examined in detail. When investigating differential networks using simulations of three alternative data generating scenarios, it was found that an edge between two nodes can be induced if the effect of one node on disease is modified by another node, or if the disease causes (or is associated with) a "breaking down" in the relationship between the two nodes. The simulations focused on simplified settings but exemplify the difficulties in interpreting differential networks and helped elucidate the sample sizes required. Further algebraic examination of likely data generating mechanisms identified the potential pitfalls of relying on partial correlations in building differential networks. This shows that, when important nodes influencing the correlation structure are not measured, irrelevant edges may be selected, while relevant ones may be missed. Analysis of the BWHHS metabolite data flagged a small number of metabolites that could potentially be associated with CHD, with small VLDL triglycerides being the strongest candidate. Comparisons were made with the results obtained using regression-based methods as these are more easily accessible to epidemiologists. The fact that there was little overlap in identified biomarkers is an indication of the complexity of this field of research.

Non-targeted metabolite profiling by liquid chromatography-mass spectrometry (LC-MS) was used to determine the metabolite responses of Arabidopsis roots to auxin or ethylene. Crosstalk between these hormones regulates many important physiological processes in plants, including the initiation of lateral root formation and the response to gravity. These occur in part through alterations in the levels of flavonoids, specialized plant metabolites that have been shown to act as negative regulators of auxin transport. However, much remains to be learned about auxin and ethylene responses at the level of the metabolome. LC-MS analysis showed that a number of ions changed in response to both hormones in seedling roots. Although classes of specialized metabolites such as flavonols and glucosinolates change in abundance in response to both auxin and ethylene, there was little overlap with regard to the specific metabolites affected. These data will be integrated with information from transcriptomic and proteomic experiments to develop framework models that connect phytohormones and specialized metabolism with specific physiological processes. Previous studies by imaging techniques have shown that flavonols increase in response to both auxin and ethylene in the root elongation zone, but LC-MS showed that flavonols decreased in abundance in response to these hormones. Therefore a method was developed for targeted metabolite profiling of flavonols in individual root tips by flow injection electrospray mass spectrometry. This method uncovered spatial differences in metabolic profiles that were masked in analyses of whole roots or seedlings, and verified that flavonols increase in response to these hormones in root tips.
Master of Science

The grape is arguably one of the oldest cultivated products in human history and the analysis of its main product, wine, reveals clues to trade and associations of previous civilizations. In ancient times, wine was stored in clay amphorae, which, if not properly sealed with resin or pitch allowed the wine to wick into clay matrices, dry, and polymerize producing insoluble, intractable materials that may remain within the matrix for several thousand years. Presently, identification of wine residue is based upon the extraction of these polymeric materials from the ceramic matrix and analyzing/identifying the chemical fingerprints. Two main biomarkers have historically been employed for the identification of wine residue: tartaric and syringic acids. In some cases, the presence of one of these biomarkers has been designated as the confirmatory signature of wine often leading to false positives as amphorae were re-used in antiquity. Herein, a novel approach utilizing metabolomics has been applied to archaeological objects in order to further mine possible biomarkers for a more accurate assessment of the original foodstuff. An untargeted metabolic profiling method was combined with a targeted analytical method resulting in the successful validation of eight representative biomarkers in two separate archaeological sites.

Lung cancer is the cancer form that has the highest mortality worldwide and inaddition the survival of lung cancer is very low. Only 15% of the patients are alivefive years from set diagnosis. More research is needed to understand the biologyof lung cancer and thus make it possible to discover the disease at an early stage.Early diagnosis leads to an increased chance of survival. In this thesis 179 lungcancer- and 116 control samples of blood serum were analyzed for identificationof metabolomic biomarkers. The control samples were derived from patients withbenign lung diseases.Data was gained from GC/TOF-MS analysis and analyzed with the help ofthe multivariate analysis methods PCA and OPLS/OPLS-DA. In this thesis it isinvestigated how to pre-treat and analyze the data in the best way in order todiscover biomarkers. One part of the aim was to give directions for how to selectsamples from a biobank for further biological validation of suspected biomarkers.Models for different stages of lung cancer versus control samples were computedand validated. The most influencing metabolites in the models were selected andconfoundings with other clinical characteristics like gender and hemoglobin levelswere studied. 13 lung cancer biomakers were identified and validated by raw dataand new OPLS models based solely upon the biomarkers.In summary the identified biomarkers are able to separate fairly good betweencontrol samples and late lung cancer, but are poor for separation of early lungcancer from control samples. The recommendation is to select controls and latelung cancer samples from the biobank for further confirmation of the biomarkers.NyckelordLung cancer is the cancer form that has the highest mortality worldwide and inaddition the survival of lung cancer is very low. Only 15% of the patients are alivefive years from set diagnosis. More research is needed to understand the biologyof lung cancer and thus make it possible to discover the disease at an early stage.Early diagnosis leads to an increased chance of survival. In this thesis 179 lungcancer- and 116 control samples of blood serum were analyzed for identificationof metabolomic biomarkers. The control samples were derived from patients withbenign lung diseases.Data was gained from GC/TOF-MS analysis and analyzed with the help ofthe multivariate analysis methods PCA and OPLS/OPLS-DA. In this thesis it isinvestigated how to pre-treat and analyze the data in the best way in order todiscover biomarkers. One part of the aim was to give directions for how to selectsamples from a biobank for further biological validation of suspected biomarkers.Models for different stages of lung cancer versus control samples were computedand validated. The most influencing metabolites in the models were selected andconfoundings with other clinical characteristics like gender and hemoglobin levelswere studied. 13 lung cancer biomakers were identified and validated by raw dataand new OPLS models based solely upon the biomarkers.In summary the identified biomarkers are able to separate fairly good betweencontrol samples and late lung cancer, but are poor for separation of early lungcancer from control samples. The recommendation is to select controls and latelung cancer samples from the biobank for further confirmation of the biomarkers.Nyckelord

Moderne Hochdurchsatzmethoden erlauben die Messung einer Vielzahl von komplementären Daten und implizieren die Existenz von regulativen Netzwerken auf einem systembiologischen Niveau. Ein üblicher Ansatz zur Rekonstruktion solcher Netzwerke stellt die Clusteranalyse dar, die auf einem Ähnlichkeitsmaß beruht.
Wir verwenden das informationstheoretische Konzept der wechselseitigen Information, das ursprünglich für diskrete Daten definiert ist, als Ähnlichkeitsmaß und schlagen eine Erweiterung eines für gewöhnlich für die Anwendung auf kontinuierliche biologische Daten verwendeten Algorithmus vor. Wir vergleichen unseren Ansatz mit bereits existierenden Algorithmen. Wir entwickeln ein geschwindigkeitsoptimiertes Computerprogramm für die Anwendung der wechselseitigen Information auf große Datensätze. Weiterhin konstruieren und implementieren wir einen web-basierten Dienst fuer die Analyse von integrierten Daten, die durch unterschiedliche Messmethoden gemessen wurden. Die Anwendung auf biologische Daten zeigt biologisch relevante Gruppierungen, und rekonstruierte Signalnetzwerke zeigen Übereinstimmungen mit physiologischen Erkenntnissen.
Recent high-throughput technologies enable the acquisition of a variety of complementary data and imply regulatory networks on the systems biology level. A common approach to the reconstruction of such networks is the cluster analysis which is based on a similarity measure.
We use the information theoretic concept of the mutual information, that has been originally defined for discrete data, as a measure of similarity and propose an extension to a commonly applied algorithm for its calculation from continuous biological data. We compare our approach to previously existing algorithms. We develop a performance optimised software package for the application of the mutual information to large-scale datasets. Furthermore, we design and implement a web-based service for the analysis of integrated data measured with different technologies. Application to biological data reveals biologically relevant groupings and reconstructed signalling networks show agreements with physiological findings.

Brystkreft er den vanligste kreftformen blant kvinner i Norge. Prognose og overlevelse avhenger av type kreft, tumorstørrelse, lymfeknutemetastaser og reseptorstatus (østrogen, progesteron og HER2). Basert på genekspresjonanalyser av tumorvev kan brystkreft deles inn i fem grupper; luminal A, luminal B, basal-like, HER2 positive og normal-like. Prognosene til pasientene i de ulike gruppene varierer der pasienter med basal-like brystkreft har den dårligste prognosen, mens pasienter med luminal A brystkreft har de beste prognosene. Metabolismen i kreftceller og normale celler er svært forskjellig. Et kjennetegn ved kreftceller er endret glykolytisk aktivitet. Kreftcellene kan forbruke glukose og omdanne dette til laktat til tross for at det er rikelig med oksygen til stede. Denne effekten er ofte omtalt som Warburgeffekten. En annen karakteristisk egenskap ved kreftceller er endringer i kolinmetabolismen. High resolution magic angle spinning MR spektroskopi (HR-MAS MRS) er en metode som er egnet for å studere biokjemiske forbindelser, kalt metabolitter, i vev. Glykolyseog kolinmetabolitter i kreftvev kan derfor studeres med denne teknikken. Proton (1H) MRS gir et spekter med informasjon om hvilke metabolitter som finnes i vev. 13C MRS er velegnet til å studere metabolsk omsetning i celler, dyr eller mennesker. Ved administrering av 13C-merkede metabolitter, kan man derfor kartlegge metabolske reaksjonsveier. Siden MRS er en kvantitativ metode kan den brukes til å beregne metabolittkonsentrasjoner i vev. Ved bruk av multivariate dataanalyser kan flere metabolitter i HR-MAS MR-spektrene analyseres samtidig. Denne metoden er derfor egnet til å studere metabolske forskjeller mellom ulike brystkreftgrupper. Siden vevet er intakt etter HR-MAS MRS kan det brukes til andre analyser etterpå, som for eksempel histopatologi eller genekspresjonsanalyser. Genekspresjonsanalyse er en egnet metode for å kartlegge hele eller deler av genomet. Med denne metoden kan man undersøke de genetiske forandringene som oppstår i kreftceller. Denne doktorgraden består av tre studier. Målet med det første studiet var å kartlegge prognostiske faktorer i brystkreftvev ved bruk HR-MAS MRS og multivariate dataanalyser. Tre ulike typer multivariate metoder ble benyttet for å undersøke om HRMAS MR spektrene inneholder informasjon som kan brukes til å prediktere østrogenreseptorstatus, progesteronreseptorstatus og lymfeknutestatus. Resultatene viste at det finnes metabolske forskjeller mellom tumorer som har positiv og negativ hormonreseptorstatus. I det andre studiet ble 13C HR-MAS MRS og genekspresjonsanalyser brukt til å kartlegge den glykolytiske aktiviteten i to ulike brystkreft musemodeller som representerer luminal-like og basal-like brystkreft. 13C-merket glukose ble injisert i de to modellene og tumorvev samlet 10 eller 15 minutter etter injeksjon. HR-MAS MRSanalysene av tumorvevet viste at glukose/laktat (Glc/Lac) og glukose/alanin (Glc/Ala)- ratioene var større i de raskt voksende basal-like svulstene sammenlignet med den luminal-like modellen. De fleste glykolytiske genene var dessuten oppregulert i den luminal-like modellen. Disse resultatene indikerer at den luminal-like modellen har større glykolytisk aktivitet enn den basal-like modellen, og at tumorvekst ikke nødvendigvis er en avgjørende faktor for glykolytisk aktivitet. Hensikten med det tredje studiet var å beskrive den metabolske profilen til et større utvalg av brystkreft musemodeller som representerer både luminal A, luminal B, basallike og HER2 positiv brystkreft. Resultatene viste at luminal B-svulstene hadde en større fosfokolin/glyserofosfokoline (PCho/GPC)-ratio enn de fleste basal-like svulstene. I tillegg var kolin, PCho og GPC korrelert til andre gener i kolinmetabolismen i luminal B- svulstene enn i de basal-like svulstene. Dette kan bety at reguleringen av kolinmetabolismen er ulik i de to undergruppene av brystkreft. Det var i tillegg god overensstemmelse mellom både metabolitt- og genekspresjonsprofiler mellom xenograftprøvene og brystkreftprøver fra pasienter i de to undergruppene. Resultatene fra studiet viser at dette panelet av xenograftmodeller er representativt for brystkreft hos mennesker, og betyr at modellene kan brukes til å identifisere nye behandlingsregimer ved bruk av HR-MAS MRS og genekspresjonsanalyser. Studiene beskrevet i denne avhandlingen har vist at HR-MAS MRS og genekspresjonsanalyser reflekterer ulike karakteristikker i brystkreft og at disse metodene derfor kan brukes til å utvikle prognostiske verktøy for brystkreftpasienter.

The 21st Century has seen a rise in incidence of complex diseases such as cancer and in the quest to develop essential new therapeutic options, the study of drug-cell interactions can yield powerful information. Acute myeloid leukaemia (AML) is an aggressive cancer that causes life-threatening deficits of functional blood cells in humans for which current treatment options are highly toxic and often poorly tolerated. A combination of two existing drugs, bezafibrate and medroxyprogesterone acetate in a drug redeployment situation has shown promise in vitro and in vivo and further investigations are crucial to elucidate the mode of action of this treatment. This project investigated the mechanistic action of BaP at a cellular level. Orthogonal spectroscopic and mass spectrometric platforms were employed to probe the biochemical composition of two AML cell lines, HL60 and K562 in the presence and absence of this combined drug treatment. Analysis was performed on single living cells, dehydrated cells, fixed cells and cell extracts to give a large and detailed data set. A consideration of the main spectral differences obtained by Synchrotron-FTIR and ATR-FTIR in conjunction with multivariate statistical analysis revealed a significant change to the cellular lipid composition with drug treatment; furthermore, this response was not caused by cell apoptosis. In particular, the ratio of CH2:CH3 was observed to increase with BaP treatment and this was determined to be a significant change in both cell lines (p <0.05). An overall increase in lipid unsaturation suggests that BaP targets cellular lipid biosynthesis. Raman microspectroscopy added a further dimension to the spectroscopic study by providing spatial information of lipid distribution which suggested that BaP-induced saturation change is uniform across a single cell. UHPLC-MS was employed for global metabolomics analysis of AML cell extracts and revealed a number of biochemical pathways that were indicated as targets of BaP therapy in both cell lines. Univariate and multivariate analysis determined statistically significant metabolites for which putative identifications were made. Pyrimidine metabolism was the most significant pathway identified for changes consistent in both HL60 and K562 cell lines. The complementarity of ToF-SIMS and UHPLC-MS provided large coverage of the lipidome of AML cells through untargeted and targeted approaches. For data derived by both techniques, a general increase in polyunsaturated species for BaP treated cell extracts was observed which correlated well with findings from spectroscopic investigations. Adopting a multi-disciplinary approach to cell analysis can afford a powerful insight into understanding drug mode of action at a cellular level and novel information regarding BaP mechanistic action in AML cell lines was revealed. This analytical approach could be extended to the future study of drug-cell interactions for other oncological systems.

In questa tesi vengono studiate alcune caratteristiche dei network a multiplex; in particolare l'analisi verte sulla quantificazione delle differenze fra i layer del multiplex. Le dissimilarita sono valutate sia osservando le connessioni di singoli nodi in layer diversi, sia stimando le diverse partizioni dei layer. Sono quindi introdotte alcune importanti misure per la caratterizzazione dei multiplex, che vengono poi usate per la costruzione di metodi di community detection . La quantificazione delle differenze tra le partizioni di due layer viene stimata utilizzando una misura di mutua informazione. Viene inoltre approfondito l'uso del test dell'ipergeometrica per la determinazione di nodi sovra-rappresentati in un layer, mostrando l'efficacia del test in funzione della similarita dei layer. Questi metodi per la caratterizzazione delle proprieta dei network a multiplex vengono applicati a dati biologici reali. I dati utilizzati sono stati raccolti dallo studio DILGOM con l'obiettivo di determinare le implicazioni genetiche, trascrittomiche e metaboliche dell'obesita e della sindrome metabolica. Questi dati sono utilizzati dal progetto Mimomics per la determinazione di relazioni fra diverse omiche. Nella tesi sono analizzati i dati metabolici utilizzando un approccio a multiplex network per verificare la presenza di differenze fra le relazioni di composti sanguigni di persone obese e normopeso.

Biological processes in living organisms consist of a vast number of different molecular networks and interactions, which are complex and often hidden from our understanding. This work is focused on recovery of such details for two quite distant examples: acclimation to extreme freezing tolerance in Siberian spruce (Picea obovata) and detection of proteins associated with prostate cancer. The first biological system in the study, upon P. obovata, is interesting by this species ability to adapt and sustain extremely low temperatures, such as -60⁰C or below. Despite decades of investigations, the essential features and mechanisms of the amazing ability of this species still remains unclear. To enhance knowledge about extreme freezing tolerance, the metabolome and proteome of P. obovata’s needles were collected during the tree’s acclimation period, ranging from mid August to January, and have been analyzed. The second system within this study is the plasma proteome analysis of high risk prostate cancer (PCa) patients, with and without bone metastases. PCa is one of the most common cancers among Swedish men, which can abruptly develop into an aggressive, lethal disease. The diagnostic tools, including PSA-tests, are insufficient in predicting the disease’s aggressiveness and novel prognostic markers are urgently required. Both biological systems have been analyzed following similar steps: by two-dimensional difference gel electrophoresis (2D-DIGE) techniques, followed by protein identification using mass spectrometry (MS) analysis and multivariate methods. Data processing has been utilized for searching for proteins that serve as unique indicators for characterizing the status of the systems. In addition, the gas chromatography-mass spectrometry (GC-MS) study of the metabolic content of P.obovata’s needles, from the extended observation period, has been performed. The studies of both systems, combined with thorough statistical analysis of experimental outcomes, have resulted in novel insights and features for both P. obovata and prostate cancer. In particular, it has been shown that dehydrins, Hsp70s, AAA+ ATPases, lipocalin and several proteins involved in cellular metabolism etc., can be uniquely associated with acclimation to extreme freezing in conifers. Metabolomic analysis of P. obovata needles has revealed systematic metabolic changes in carbohydrate and lipid metabolism. Substantial increase of raffinose, accumulation of desaturated fatty acids, sugar acids, sugar alcohols, amino acids and polyamines that may act as compatible solutes or cryoprotectants have all been observed during the acclimation process. Relevant proteins for prostate cancer progression and aggressiveness have been identified in the plasma proteome study, for patients with and without bone metastasis. Proteins associated with lipid transport, coagulation, inflammation and immune response have been found among them.

Enteric (or typhoid) fever is a systemic infection mainly caused by Salmonella Typhi and Salmonella Paratyphi A. The disease is common in areas with poor water quality and insufficient sanitation. Humans are the only reservoir for transmission of the disease. The presence of asymptomatic chronic carriers is a complicating factor for the transmission. There are major limitations regarding the current diagnostic methods both for acute infection and chronic carriage. Metabolomics is a methodology studying metabolites in biological systems under influence of environmental or physiological perturbations. It has been applied to study several infectious diseases, with the goal of detecting diagnostic biomarkers. In this thesis, a mass spectrometry-based metabolomics approach, including chemometric bioinformatics techniques for data analysis, has been used to evaluate the potential of metabolite biomarker patterns for diagnosis of enteric fever at different stages of the disease. In Paper I, metabolite patterns related to acute enteric fever were investigated. Human plasma samples from patients in Nepal with culture-confirmed S. Typhi or S. Paratyphi A infection were compared to afebrile controls. A metabolite pattern discriminating between acute enteric fever and afebrile controls, as well as between the two causative agents of enteric fever was detected. The strength of using a panel of metabolites instead of single metabolites as biomarkers was also highlighted. In Paper II, metabolite patterns for acute enteric fever, this time focusing only on S. Typhi infections, were investigated. Human plasma from patients in Bangladesh with culture-positive or -negative but clinically suspected S. Typhi infection were compared to febrile controls. Differences were found in metabolite patterns between the culture-positive S. Typhi group and the febrile controls with a heterogeneity among the suspected S. Typhi samples. Consistencies in metabolite patterns were found to the results from Paper I. In addition, a validation cohort with culture-positive S. Typhi samples and a control group including patients with malaria and infections caused by other pathogens was analysed. Differences in metabolite patterns were detected between S. Typhi samples and all controls as well as between S. Typhi and malaria. Consistencies in metabolite patterns were found to the primary Bangladeshi cohort and the Nepali cohort from Paper I. Paper III focused on chronic Salmonella carriers. Human plasma samples from patients in Nepal undergoing cholecystectomy with confirmed S. Typhi or S. Paratyphi A gallbladder carriage were compared to non-carriage controls. The Salmonella carriage samples were distinguished from the non-carriage controls and differential signatures were also found between the S. Typhi and S. Paratyphi A carriage samples. Comparing metabolites found during chronic carriage and acute enteric fever (in Paper I) resulted in a panel of metabolites significant only during chronic carriage. This work has contributed to highlight the potential of using metabolomics as a tool to find diagnostic biomarker patterns associated with different stages of enteric fever.