Search

La randomisation mendélienne est une méthode d’instrumentation utilisant des instruments de nature génétique afin d’estimer, via par exemple la régression des moindres carrés en deux étapes, une relation de causalité entre un facteur d’exposition et une réponse lorsque celle-ci est confondue par une ou plusieurs variables de confusion non mesurées. La randomisation mendélienne est en mesure de gérer le biais de confusion à condition que les instruments utilisés soient valides, c’est-à-dire qu’ils respectent trois hypothèses clés. On peut généralement se convaincre que deux des trois hypothèses sont satisfaites alors qu’un phénomène génétique, la pléiotropie, peut parfois rendre la troisième hypothèse invalide. En présence d’invalidité, l’estimation de l’effet causal de l’exposition sur la réponse peut être sévèrement biaisée. Afin d’évaluer la potentielle présence d’invalidité lorsqu’un seul instrument est utilisé, Glymour et al. (2012) ont proposé une méthode qu’on dénomme ici l’approche de la différence simple qui utilise le signe de la différence entre l’estimateur des moindres carrés ordinaires de la réponse sur l’exposition et l’estimateur des moindres carrés en deux étapes calculé à partir de l’instrument pour juger de l’invalidité de l’instrument. Ce mémoire introduit trois méthodes qui s’inspirent de cette approche, mais qui sont applicables à la randomisation mendélienne à instruments multiples. D’abord, on introduit l’approche de la différence globale, une simple généralisation de l’approche de la différence simple au cas des instruments multiples qui a comme objectif de détecter si un ou plusieurs instruments utilisés sont invalides. Ensuite, on introduit les approches des différences individuelles et des différences groupées, deux méthodes qui généralisent les outils de détection de l’invalidité de l’approche de la différence simple afin d’identifier des instruments potentiellement problématiques et proposent une nouvelle estimation de l’effet causal de l’exposition sur la réponse. L’évaluation des méthodes passe par une étude théorique de l’impact de l’invalidité sur la convergence des estimateurs des moindres carrés ordinaires et des moindres carrés en deux étapes et une simulation qui compare la précision des estimateurs résultant des différentes méthodes et leur capacité à détecter l’invalidité des instruments., Mendelian randomization is an instrumentation method that uses genetic instruments to estimate, via two-stage least squares regression for example, a causal relationship between an exposure and an outcome when the relationship is confounded by one or more unmeasured confounders. Mendelian randomization can handle confounding bias provided that the instruments are valid, i.e., that they meet three key assumptions. While two of the three assumptions can usually be satisfied, the third assumption is often invalidated by a genetic phenomenon called pleiotropy. In the presence of invalid instruments, the estimate of the causal effect of exposure on the outcome may be severely biased. To assess the potential presence of an invalid instrument in single-instrument studies, Glymour et al. (2012) proposed a method, hereinafter referred to as the simple difference approach, which uses the sign of the difference between the ordinary least squares estimator of the outcome on the exposure and the two-stage least squares estimator calculated using the instrument. Based on this approach, we introduce three methods applicable to Mendelian randomization with multiple instruments. The first method is the global difference approach and corresponds to a simple generalization of the simple difference approach to the case of multiple instruments that aims to detect whether one or more instruments are invalid. Next, we introduce the individual differences and the grouped differences approaches, two methods that generalize the simple difference approach to identify potentially invalid instruments and provide new estimates of the causal effect of the exposure on the outcome. The methods are evaluated using a theoretical investigation of the impact that invalid instruments have on the convergence of the ordinary least squares and two-stage least squares estimators as well as with a simulation study that compares the accuracy of the respective estimators and the ability of the corresponding methods to detect invalid instruments.

Les variations du nombre de copies (i.e., VNC, perte ou gain de matériel génétique de plus de 1 kilobase) figurent parmi les facteurs biologiques les plus associés aux troubles neurodéveloppementaux (TNDs), tels que les troubles du spectre autistique (TSAs) ou la psychose précoce. Les variants génétiques classés comme pathogéniques sont identifiés chez environ 20% des enfants avec des symptômes de TSA référés en génétique clinique. Actuellement, seules les VNCs les plus récurrentes (i.e., plusieurs individus non apparentés ont le même variant) ont été associées avec les TSAs et leurs tailles d’effets ont pu être décrites avec précision grâce à des études d'associations (i.e., cas-contrôles). Cependant, la plupart des VNCs identifiées dans les cliniques neurodéveloppementales et génétiques sont ultra-rares. À ma connaissance, aucune méthode n’a été développée afin d’estimer et de prédire de façon précise la contribution de tels variants aux phénotypes cliniques. De ce fait, l’impact de ces variants ultra-rares sur les risques d'avoir des TNDs, comme les TSAs ou la psychose précoce, reste incertain. Une étude récente de mon groupe de recherche a démontré que les tailles d'effet des délétions et duplications à travers le génome sur les capacités cognitives pouvaient être prédites statistiquement avec 78% de précision en utilisant des mesures d'intolérance à la perte de fonction. Le but de cette thèse est de développer des modèles similaires pour définir les tailles d'effet des VNCs à travers le génome sur les risques de TSA et de psychose précoce, ainsi que sur quelques traits cognitifs et comportementaux affectés dans ces troubles. J’ai analysé tous les VNCs ≥ 50 kilobases identifiées via les données de puces de génotypage et de séquençage sur génome entier chez 137 enfants et adolescents avec une psychose précoce (Boston Children’s hospital), 5,540 probands avec des TSAs (Simons Simplex Collection et MSSNG), et 17,471 personnes de la population générale (Lothian birth cohort, Generation Scotland, IMAGEN et Saguenay Youth Study). Les gènes codants totalement compris dans les VNCs ont été annotés avec neufs variables quantitatives, incluant le score d’intolérance à la perte de fonction et d’autres scores fonctionnels et génétiques. Des modèles statistiques incluant ces scores ont été testés afin de sélectionner celui qui explique le mieux l’effet des VNCs à travers le génome sur le risque de TSA et le quotient intellectuel (QI). Le meilleur modèle a été utilisé par la suite pour investiguer les tailles d’effets des VNCs sur d’autres traits cognitifs et comportementaux liés aux TSAs, ainsi que sur le risque de psychose précoce. Le score d’intolérance à la perte de fonction expliquait le mieux les effets des VNCs sur le risque de TSA et la cognition générale. Les modèles incluant ces scores ont démontré que les délétions et les duplications augmentaient les risques de psychose précoce et de TSA, même après ajustement pour le QI. Il n’y avait aucune différence de tailles d’effets des VNCs entre la psychose précoce et le TSA. La fréquence de loci associé précédemment avec des TNDs et des troubles neuropsychiatriques était également similaire entre dans les TSA et la psychose précoce, et le modèle estimait précisément la taille d'effet de la plupart de ces loci sur le risque de TSA en comparaison aux observation empiriques publiées précédemment. Les CNVs à travers le génome mesurés par le score d’intolérance à la perte de fonction diminuaient de façon similaire le QI dans les populations TSA et générale. Les effets des duplications étaient systématiquement plus faibles que les effets des délétions pour chacun de ces phénotypes, ce qui suggère un effet plus pathogénique des délétions. Les délétions et les duplications affectaient différentiellement la communication sociale, les comportements, et la mémoire phonologique, tandis qu'elles affectaient similairement les capacités motrices dans les populations TSA. L'enrichissement similaire des VNCs à travers le génome dans la psychose précoce et le TSA suggère un effet pléiotropique des VNCs dans ces différentes symptomatologies. Le dépistage routinier pour les VNCs doit être accessible dans les soins cliniques standards des jeunes avec une psychose précoce, comme il est recommandé pour les TSAs. Une telle pratique contribue à établir une médecine personnalisée et peut apporter des bénéfices médicaux comme la détection de comorbidités, la prédiction de la progression de la maladie, et faciliter la communication avec les parents à propos de la nature biologique du trouble. Les modèles appliqués dans ce projet, entraînés sur des VNCs incluant plus de 4,500 gènes, suggèrent des propriétés hautement polygéniques du dosage génique dans les TNDs. J’ai estimé que chaque VNC de 1 mégabase, incluant au moins un gène scorant pour l’intolérance à la perte de fonction, augmente le risque de TSA. La combinaison de ces résultats ouvre de nouvelles perspectives dans la compréhension des effets des VNCs à travers le génome sur les TNDs et les traits associés (e.g., QI ou symptômes comportementaux). Ces modèles ont été implémentés dans un outil en ligne qui a pour but d'aider les cliniciens à estimer les tailles d’effet des VNCs identifiés en clinique et à interpréter leur contribution au phénotype du patient., Copy number variants (CNVs; i.e., loss or gain of genetic material of over 1 kilobase) are robustly associated with neurodevelopmental disorders (NDDs), such as autism spectrum disorder (ASD) and early-onset psychosis (EOP). Genetic variants classified as pathogenic are identified in approximately 20% of children with ASD symptoms referred to genetic clinics. To date, only the most recurrent CNVs (i.e., similar variants across multiple unrelated individuals) were associated with ASD and their effect-sizes were characterized through association studies (i.e., case-controls). However, most of the CNVs routinely identified in neurodevelopmental and genetic clinics are ultra-rare. To my knowledge, no method was developed to accurately estimate and predict the contribution of such variants to clinical phenotypes. Therefore, the impact of these ultra-rare variants on risk for NDDs, such as ASD and EOP, remains undocumented. A recent study from my research group has shown that the effect-size of genome-wide deletions and duplications on cognitive ability can be statistically predicted with an 78% accuracy using measures of loss-of-function (LoF) intolerance. The aim of this thesis was to develop similar models to define the effect-size of genome-wide CNVs on ASD and EOP risk, as well as on several cognitive and behavioral traits altered in these disorders. I analyzed all CNVs ≥ 50 kilobases called from genotyping arrays and whole genome sequencing data from 137 children and adolescents with EOP (Boston Children’s hospital), 5,540 probands with ASD (Simons Simplex Collection and MSSNG), and 17,471 individuals from unselected populations (Lothian birth cohort, Generation Scotland, IMAGEN and Saguenay Youth Study). Coding genes fully encompassed by CNVs were annotated with nine quantitative variables, including the LoF intolerance score and other functional and genetic scores. Statistical models including these scores were tested to select the one that best explained the effects of genome-wide CNVs on ASD risk and IQ. The best model was subsequently used to investigate the effect-size of genome-wide CNVs on cognitive and behavioral domains related to ASD, as well as on EOP risk. The LoF intolerance score best explained the effect-sizes of genome-wide CNVs on ASD-risk and general cognition. Models including such scores demonstrated that deletions or duplications increased risks for EOP and for ASD, even after adjusting for IQ. There was no difference in effect-sizes between EOP and ASD. The frequency of loci previously associated with NDDs or neuropsychiatric disorders was also similar between EOP and ASD, and the model accurately estimated the effect-size of most of these loci on the risk for ASD comparing to previously published empirical observations. Genome-wide CNVs measured by LoF intolerance score also similarly decreased IQ in both ASD and unselected populations. The effect of duplications was smaller than the effect of deletion for all phenotypes investigated, suggesting a higher pathogenicity of deletions. Deletions and duplications were found to differentially affect social communication, behavior, and phonological memory, whereas both equally affected motor skills in the ASD population. The identical enrichment of genome-wide CNVs in EOP and ASD suggests a pleiotropic effect of CNVs in these different symptomatology. Routine screening for CNVs should be made available in the standard clinical care for EOP youth, as is recommended in ASD. Such practice contributes to the establishment of personalized medicine and may bring medical benefits as detecting medical comorbidities, prediction of the disease progression, and facilitating the communication with parents about the biological nature of the disorder. The models applied in this project, trained on CNVs encompassing more than 4,500 genes, suggest highly polygenic properties of gene dosage in NDDs. I estimated that any 1 megabase CNV, encompassing at least one gene scoring for intolerance to LoF, would increase ASD risk. Overall, these results open new avenues for understanding the effect of genome-wide CNVs on NDD risk and related traits (e.g., IQ or behavioral symptoms). These models were implemented in an online tool which aims to help clinicians estimate the effect-size of CNVs identified in the clinic and interpret their contribution to the patient’s phenotype.

Is the fate of a mutation predictable? The repeatability of evolution may indicate its predictability. If some change has emerged multiple times in the past, it may also occur in the future. Convergent evolution provides a natural model of repeated evolution and suggests the presence of certain constraints. Examination of the genetic processes, that have led to convergent adaptive traits, can contribute to the knowledge of the predictability of evolution. The aim of my work is to summarize the available knowledge about factors affecting gene reuse in adaptation. My work shows that the repeatability of evolution is biased by the genome characteristics, the population characteristics and also the trait encoding complexity. The likelihood of genomic region reuse increases with the amount of emerging mutations affecting the phenotype. Regarding the pleiotropic constraints of gene reuse, the most suitable is a mutation in the cis-regulatory region of the gene on the periphery of the functional pathway. Of the population traits, the greater effective size of the population, the higher standing genetic variation and the lower divergence between populations raise the chance of gene reuse. Despite the dependence of evolution on random processes, there are many factors that direct it and make it more...

Background: Inborn errors of immunity (IEI) present with a large phenotypic spectrum of disease, which can pose diagnostic and therapeutic challenges. Suppressor of cytokine signaling 1 (SOCS1) is a key negative regulator of cytokine signaling, and has recently been associated with a novel IEI. Of patients described to date, it is apparent that SOCS1 haploinsufficiency has a pleiotropic effect in humans. Objective: We sought to investigate whether dysregulation of immune pathways, in addition to STAT1, play a role in the broad clinical manifestations of SOCS1 haploinsufficiency. Methods: We assessed impacts of reduced SOCS1 expression across multiple immune cell pathways utilizing patient cells and CRISPR/Cas9 edited primary human T cells. Results: SOCS1 haploinsufficiency phenotypes straddled across the International Union of Immunological Societies classifications of IEI. We found that reduced SOCS1 expression led to dysregulation of multiple intracellular pathways in immune cells. STAT1 phosphorylation is enhanced, comparably with STAT1 gain-of-function mutations, and STAT3 phosphorylation is similarly reduced with concurrent reduction of Th17 cells. Furthermore, reduced SOCS1 E3 ligase function was associated with increased FAK1 in immune cells, and increased AKT and p70 ribosomal protein S6 kinase phosphorylation. We also found Toll-like receptor responses are increased in SOCS1 haploinsufficiency patients. Conclusions: SOCS1 haploinsufficiency is a pleiotropic monogenic IEI. Dysregulation of multiple immune cell pathways may explain the variable clinical phenotype associated with this new condition. Knowledge of these additional dysregulated immune pathways is important when considering the optimum management for SOCS1 haploinsufficient patients.

Les recherches menées pour comprendre les troubles du spectre autistique (TSA) et la schizophrénie (SZ) ont communément utilisé une approche dite descendante, partant du diagnostic clinique pour investiguer des phénotypes intermédiaires cérébraux ainsi que des variations génétiques associées. Des études transdiagnostiques récentes ont remis en question ces frontières nosologiques, et suggèrent des mécanismes étiologiques imbriqués. L’approche montante propose de composer des groupes de porteurs d’un même variant génétique afin d’investiguer leur contribution aux conditions neuropsychiatriques (NPs) associées. Les variations du nombre de copies (CNV, perte ou gain d’un fragment d’ADN) figurent parmi les facteurs biologiques les plus associés aux NPs, et sont dès lors des candidats particulièrement appropriés. Les CNVs induisant un risque pour des conditions similaires, nous posons l’hypothèse que des classes entières de CNVs convergent sur des dimensions d’altérations cérébrales qui contribuent aux NPs. L’imagerie fonctionnelle au repos (rs-fMRI) s’est révélée un outil prometteur en psychiatrie, mais presqu’aucune étude n’a été menée pour comprendre l’impact des CNVs sur la connectivité fonctionnelle cérébrale (FC). Nos objectifs étaient de: 1) Caractériser l’effet des CNVs sur la FC; 2) Rechercher la présence des motifs conférés par ces signatures biologiques dans des conditions idiopathiques; 3) Tester si la suppression de gènes intolérants à l’haploinsuffisance réorganise la FC de manière indépendante à leur localisation dans le génome. Nous avons agrégé des données de rs-fMRI chez: 502 porteurs de 8 CNVs associées aux NPs (CNVs-NP), de 4 CNVs sans association établie, ainsi que de porteurs de CNVs-NPs éparses; 756 sujets ayant un diagnostic de TSA, de SZ, ou de trouble déficitaire de l’attention/hyperactivité (TDAH), et 5377 contrôles. Les analyses du connectome entier ont montré un effet de dosage génique positif pour les CNVs 22q11.2 et 1q21.1, et négatif pour le 16p11.2. La taille de l’effet des CNVs sur la FC était corrélée au niveau de risque psychiatrique conféré par le CNV. En accord avec leurs effets sur la cognition, l’effet des délétions sur la FC était plus élevé que celui des duplications. Nous avons identifié des similarités entre les motifs cérébraux conférés par les CNVs-NP, et l’architecture fonctionnelle des individus avec NPs. Le niveau de similarité était associé à la sévérité du CNV, et était plus fort avec la SZ et les TSA qu’avec les TDAH. La comparaison des motifs conférés par les délétions les plus sévères (16p11.2, 22q11.2) à l’échelle fonctionnelle, et d’expression génique, nous a confirmé l’existence présumée de relation entre les mutations elles-mêmes. À l’aide d’une mesure d’intolérance aux mutations (pLI), nous avons pu inclure tous les porteurs de CNVs disponibles, et ainsi identifier un profil d’haploinsuffisance impliquant le thalamus, le cortex antérieur cingulaire, et le réseau somato-moteur, associé à une diminution de mesure d’intelligence générale. Enfin, une analyse d’exploration factorielle nous a permis de confirmer la contribution de ces régions cérébrales à 3 composantes latentes partagées entre les CNVs et les NPs. Nos résultats ouvrent de nouvelles perspectives dans la compréhension des mécanismes polygéniques à l’oeuvre dans les maladies mentales, ainsi que des effets pléiotropiques des CNVs., Research on Autism Spectrum Disorder (ASD) and schizophrenia (SZ) has mainly adopted a ‘top-down’ approach, starting from psychiatric diagnosis, and moving to intermediate brain phenotypes and underlying genetic factors. Recent cross-disorder studies have raised questions about diagnostic boundaries and pleiotropic mechanisms. By contrast, the recruitment of groups based on the presence of a genetic risk factor allows for the investigation of molecular pathways related to a particular risk for neuropsychiatric conditions (NPs). Copy number variants (CNVs, loss or gain of a DNA segment), which confer high risk for NPs are natural candidates to conduct such bottom-up approaches. Because CNVs have a similar range of adverse effects on NPs, we hypothesized that entire classes of CNVs may converge upon shared connectivity dimensions contributing to mental illness. Resting-state functional MRI (rs-fMRI) studies have provided critical insight into the architecture of brain networks involved in NPs, but so far only a few studies have investigated networks modulated by CNVs. We aimed at 1) Delineating the effects of neuropsychiatric variants on functional connectivity (FC), 2) Investigating whether the alterations associated with CNVs are also found among idiopathic psychiatric populations, 3) Testing whether deletions reorganize FC along general dimensions, irrespective of their localization in the genome. We gathered rsfMRI data on 502 carriers of eight NP-CNVs (high-risk), four CNVs without prior association to NPs as well as carriers of eight scarcer NP-CNVs. We also analyzed 756 subjects with idiopathic ASD, SZ, and attention deficit hyperactivity disorder (ADHD), and 5,377 controls. Connectome-wide analyses showed a positive gene dosage effect for the 22q11.2 and 1q21.1 CNVs, and a negative association for the 16p11.2 CNV. The effect size of CNVs on relative FC (mean-connectivity adjusted) was correlated with the known level of NP-risk conferred by CNVs. Consistent with results on cognition, we also reported that deletions had a larger effect size on FC than duplications. We identified similarities between high-risk CNV profiles and the connectivity architecture of individuals with NPs. The level of similarity was associated with mutation severity and was strongest in SZ, followed by ASD, and ADHD. The similarity was driven by the thalamus, and the posterior cingulate cortex, previously identified as hubs in transdiagnostic psychiatric studies. These results raised questions about shared mechanisms across CNVs. By comparing deletions at the 16p11.2 and 22q11.2 loci, we identified similarities at the connectivity, and at the gene expression level. We extended this work by pooling all deletions available for analysis. We asked if connectivity alterations were associated with the severity of deletions scored using pLI, a measure of intolerance to haploinsufficiency. The haploinsufficiency profile involved the thalamus, anterior cingulate cortex, and somatomotor network and was correlated with lower general intelligence and higher autism severity scores in 3 unselected and disease cohorts. An exploratory factor analysis confirmed the contribution of these regions to three latent components shared across CNVs and NPs. Our results open new avenues for understanding polygenicity in psychiatric conditions, and the pleiotropic effect of CNVs on cognition and on risk for neuropsychiatric disorders.

In the last few years, next-generation sequencing has led to enormous progress in deciphering monogenic forms of intellectual disability. Autosomal dominant intellectual disability (ADID) and X chromosomal intellectual disability (XLID) have been the focus of research. Apart from metabolic disorders, autosomal recessive intellectual disability (ARID) is still behind, probably because it is more heterogeneous and less prevalent in industrial populations. The prevalence of ARID in a cohort of affected children of an outbred population is estimated to be about 10%, with an upward tendency in still unclarified cases. The risk for ARID in children of first cousins or closer is a magnitude higher than for children of unrelated parents. Taken together, it seems that children of related parents are at a 2 to 3 times higher risk for ID. There are no prevalent ARID genes, pathways, or protein complexes and the functions of the affected proteins are very diverse and limited not only to neurological aspects. Thus, in a regular case, there is no reasoning for picking a few genes for a first diagnostic step, and a genetic diagnosis of ID in general, and ARID specifically, is better made using large panels or exome sequencing. In addition, in the last few months, evidence has been growing that many ARID genes are pleiotropic and that the resulting phenotypes may have a broad spectrum. For an exhaustive deciphering of the genetics of ARID, we suggest research at the level of single genes rather than large meta-analyses.„Next generation sequencing“ (NGS) führte in den letzten Jahren zu enormen Fortschritten bei der Entschlüsselung monogener Formen der geistigen Behinderung. Bisher standen die autosomal-dominante geistige Behinderung (ADID) und die X‑chromosomale geistige Behinderung (XLID) im Mittelpunkt der Forschung. Bis auf die Stoffwechseldefekte ist die autosomal-rezessive geistige Behinderung (ARID) weniger erforscht, wahrscheinlich weil sie heterogener ist und in den Industriebevölkerungen eine geringere Prävalenz aufweist. Die Prävalenz von ARID in einer Kohorte betroffener Kinder einer Outbred-Population wird auf etwa 10 % geschätzt, wobei eine Tendenz nach oben in noch ungeklärten Fällen zu erwarten ist. Das Risiko für ARID bei Kindern von Cousins und Cousinen ersten Grades oder näher verwandt scheint um ungefähr den Faktor 10 höher zu sein als bei Kindern von nichtverwandten Eltern. Nach Berücksichtigung anderer Ätiologien ist das Gesamtrisiko für ID bei Kindern verwandter Eltern etwa 2- bis 3-fach höher. Es gibt keine Gene, Signalwege oder Proteinkomplexe, welche auffällig häufig bei ARID betroffen sind. Die Funktionen der betroffenen Proteine sind sehr vielfältig und nicht nur auf neurologische Aspekte beschränkt. Daher ist in der Regel die Auswahl einiger Gene für einen ersten diagnostischen Schritt nicht gerechtfertigt. Eine genetische Diagnose eines Falles mit ID im Generellen und von ARID im Besonderen ist durch große Panels oder Exom-Sequenzierung zu stellen. In den letzten Monaten hat sich zudem gezeigt, dass viele ARID-Gene pleiotrop sind und die resultierenden Phänotypen ein breites Spektrum aufweisen können. Für eine vollständige Entschlüsselung der Genetik von ARID sind Untersuchungen an einzelnen Genen und weniger die großen Meta-Analysen erforderlich.

In order to grow, survive or reproduce, all organisms need energy, usually acquired through diet. However, this food resource is present both in fluctuating and limited quantities in the environment, forcing living beings to compromise and thus to divide their energy between their different functions. These evolutionary compromises, visible at the scale of a population in the form of a negative relationship between traits, are called trade-off. Trade-offs have long been considered as the result of a differential resource allocation and as immutable. Therefore, allocating more energy to a trait such as survival necessarily reduces the amount that can be redistributed to other traits, such as fecundity or growth. It is noteworthy that the differential allocation of resources is a process regulated by an endocrine mechanism, itself genetically coded and thereby able to evolve. The aim of my PhD thesis was to understand, theoretically, (i) how the evolution of the endocrine mechanism impacts the shape of trade-offs and (ii) how the shape of trade-offs itself evolves.To do so, I first developed evolutionary models where the allocation of resources is governed by an endocrine system. This system can evolve under the effect of mutations that impact both the expression and the conformation of hormones and receptors constituting this endocrine system. Thanks to this model, I show that the negative relationships between traits can evolve and that their shape strongly depends on a parameter rarely considered: the cost of storage. In a second step, I studied the impact of temporal variability in food abundance on the endocrine mechanisms responsible for the differential allocation of resources.Lastly, my thesis project includes a component complementary to the theoretical part, which attempts to empirically test certain of the expressed predictions. I conducted an artificial selection experiment in which I controlled the topology of a fitness landscape, thus allowing to select combinations of traits not belonging to the phenotypic relationship usually observed. This experiment, implemented in Drosophila melanogaster for 10 generations, has shown that evolution can indeed occur in this context, thereby partially challenging our understanding of the mechanisms underlying the expression of phenotypic traits; Pour se développer, survivre ou se reproduire, les organismes ont besoin d’énergie, généralement acquise par l’alimentation. Or cette ressource alimentaire est présente en quantité fluctuante et limitée dans l’environnement, obligeant les êtres vivants à faire des compromis et à répartir leur énergie entre différentes fonctions. Ces compromis évolutifs sont visibles à l’échelle d’une population sous la forme d’une relation négative entre traits appelée trade-off. Les trade-offs ont longtemps été considérés comme étant uniquement la résultante d’une allocation différentielle des ressources et comme étant immuables. Ainsi, allouer davantage d’énergie à un trait, la survie par exemple, réduit nécessairement la part pouvant être redistribuée aux autres traits tel que la fécondité ou la croissance. Or l’allocation différentielle des ressources est un processus régulé par un mécanisme endocrinien, lui-même codé génétiquement et donc capable d’évoluer. L’objectif de cette thèse a été de comprendre, de manière théorique, comment l’évolution du mécanisme endocrinien impacte la forme des trade-offs et comment la forme des trade-offs elle-même évolue. Pour ce faire, j’ai développé des modèles évolutifs où l’allocation des ressources est régie par un système endocrinien. Ce système peut évoluer, sous l’effet de mutations impactant l’expression et la conformation des hormones et des récepteurs qui composent ce système endocrinien. J’ai ainsi pu montrer que les relations négatives entre traits peuvent évoluer et que leur forme dépend fortement d’un paramètre rarement considéré : le coût du stockage. Dans un second temps, j’ai étudié l’impact de la variabilité temporelle dans l’abondance d’une ressource alimentaire sur l’allocation différentielle et les mécanismes endocriniens sous–jacents. Mon projet de thèse comprend un troisième volet, complémentaire à la partie théorique, qui s’attache à tester certaines prédictions empiriquement. J’ai mené une expérience de sélection artificielle dans laquelle je contrôlais la topologie d’un paysage de fitness, permettant de sélectionner des combinaisons de trait n’appartenant pas à la relation phénotypique habituellement observée. Cette expérience, conduite chez Drosophila melanogaster durant 10 générations, a montré que l’évolution peut effectivement se produire dans ce contexte, remettant partiellement en cause notre compréhension des mécanismes sous-tendant l’expression des traits phénotypiques

During the last decades, the amount of available genetic data on populations has growndrastically. From one side, a refinement of chemical technologies have made possible theextraction of the human genome of individuals at an accessible cost. From the other side,consortia of institutions and laboratories around the world have permitted the collectionof data on a variety of individuals and population. This amount of data raised hope onour ability to understand the deepest mechanisms involved in the functioning of our cells.Notably, genetic epidemiology is a field that studies the relation between the geneticfeatures and the onset of a disease. Specific statistical methods have been necessary forthose analyses, especially due to the dimensions of available data: in genetics, informationis contained in a high number of variables compared to the number of observations.In this dissertation, two contributions are presented. The first project called PIGE (Pathway-Interaction Gene Environment) deals with gene-environment interaction assessments.The second one aims at developing variable selection methods for data which has groupstructures in both the variables and the observations.The document is divided into six chapters. The first chapter sets the background of this work,where both biological and mathematical notations and concepts are presented and gives ahistory of the motivation behind genetics and genetic epidemiology. The second chapterpresent an overview of the statistical methods currently in use for genetic epidemiology.The third chapter deals with the identification of gene-environment interactions. It includesa presentation of existing approaches for this problem and a contribution of the thesis. Thefourth chapter brings off the problem of meta-analysis. A definition of the problem and anoverview of the existing approaches are presented. Then, a new approach is introduced.The fifth chapter explains the pleiotropy studies and how the method presented in theprevious chapter is suited for this kind of analysis. The last chapter compiles conclusionsand research lines for the future.; Durant les dernières décennies, la quantité de données disponibles en génétique a consi-dérablement augmenté. D’une part, une amélioration des technologies de séquençage demolécules a permis de réduire fortement le coût d’extraction du génome humain. D’autrepart, des consortiums internationaux d’institutions ont permis la mise en commun de lacollecte de données sur de larges populations. Cette quantité de données nous permetd’espérer mieux comprendre les mécanismes régissant le fonctionnement de nos cellules.Dans ce contexte, l’épidémiologie génétique est un domaine cherchant à déterminer larelation entre des caractéristiques génétiques et l’apparition d’une maladie. Des méthodesstatistiques spécifiques à ce domaine ont dû être développées, en particulier à cause desdimensions que les données présentent : en génétique, l’information est contenue dans unnombre de variables grand par rapport au nombre d’observations.Dans cette dissertation, deux contributions sont présentées. Le premier projet appeléPIGE (Pathway-Interaction Gene Environment) développe une méthode pour déterminerdes interactions gène-environnement. Le second projet vise à développer une méthode desélection de variables adaptée à l’analyse de données provenant de différentes études etprésentant une structure de groupe de variables.Le document est divisé en six parties. Le premier chapitre met en relief le contexte,d’un point de vue à la fois biologique et mathématique. Le deuxième chapitre présente lesmotivations de ce travail et la mise en œuvre d’études en épidémiologie génétique. Le troi-sième chapitre aborde les questions relatives à l’analyse d’interactions gène-environnementet la première contribution de la thèse y est présentée. Le quatrième chapitre traite desproblématiques de méta-analyses. Le développement d’une nouvelle méthode de réductionde dimension répondant à ces questions y est présenté. Le cinquième chapitre met en avantla pertinence de la méthode dans des cas de pleiotropie. Enfin, le sixième et dernier chapitredresse un bilan du travail présenté et dresse des perspectives pour le futur.

During the last decades, the amount of available genetic data on populations has growndrastically. From one side, a refinement of chemical technologies have made possible theextraction of the human genome of individuals at an accessible cost. From the other side,consortia of institutions and laboratories around the world have permitted the collectionof data on a variety of individuals and population. This amount of data raised hope onour ability to understand the deepest mechanisms involved in the functioning of our cells.Notably, genetic epidemiology is a field that studies the relation between the geneticfeatures and the onset of a disease. Specific statistical methods have been necessary forthose analyses, especially due to the dimensions of available data: in genetics, informationis contained in a high number of variables compared to the number of observations.In this dissertation, two contributions are presented. The first project called PIGE (Pathway-Interaction Gene Environment) deals with gene-environment interaction assessments.The second one aims at developing variable selection methods for data which has groupstructures in both the variables and the observations.The document is divided into six chapters. The first chapter sets the background of this work,where both biological and mathematical notations and concepts are presented and gives ahistory of the motivation behind genetics and genetic epidemiology. The second chapterpresent an overview of the statistical methods currently in use for genetic epidemiology.The third chapter deals with the identification of gene-environment interactions. It includesa presentation of existing approaches for this problem and a contribution of the thesis. Thefourth chapter brings off the problem of meta-analysis. A definition of the problem and anoverview of the existing approaches are presented. Then, a new approach is introduced.The fifth chapter explains the pleiotropy studies and how the method presented in theprevious chapter is suited for this kind of analysis. The last chapter compiles conclusionsand research lines for the future.; Durant les dernières décennies, la quantité de données disponibles en génétique a consi-dérablement augmenté. D’une part, une amélioration des technologies de séquençage demolécules a permis de réduire fortement le coût d’extraction du génome humain. D’autrepart, des consortiums internationaux d’institutions ont permis la mise en commun de lacollecte de données sur de larges populations. Cette quantité de données nous permetd’espérer mieux comprendre les mécanismes régissant le fonctionnement de nos cellules.Dans ce contexte, l’épidémiologie génétique est un domaine cherchant à déterminer larelation entre des caractéristiques génétiques et l’apparition d’une maladie. Des méthodesstatistiques spécifiques à ce domaine ont dû être développées, en particulier à cause desdimensions que les données présentent : en génétique, l’information est contenue dans unnombre de variables grand par rapport au nombre d’observations.Dans cette dissertation, deux contributions sont présentées. Le premier projet appeléPIGE (Pathway-Interaction Gene Environment) développe une méthode pour déterminerdes interactions gène-environnement. Le second projet vise à développer une méthode desélection de variables adaptée à l’analyse de données provenant de différentes études etprésentant une structure de groupe de variables.Le document est divisé en six parties. Le premier chapitre met en relief le contexte,d’un point de vue à la fois biologique et mathématique. Le deuxième chapitre présente lesmotivations de ce travail et la mise en œuvre d’études en épidémiologie génétique. Le troi-sième chapitre aborde les questions relatives à l’analyse d’interactions gène-environnementet la première contribution de la thèse y est présentée. Le quatrième chapitre traite desproblématiques de méta-analyses. Le développement d’une nouvelle méthode de réductionde dimension répondant à ces questions y est présenté. Le cinquième chapitre met en avantla pertinence de la méthode dans des cas de pleiotropie. Enfin, le sixième et dernier chapitredresse un bilan du travail présenté et dresse des perspectives pour le futur.

Dans cette thèse, nous nous sommes intéressée à une mutation dans le gène SOCS2(Suppressor Of Cytokine Signaling 2) dont l’effet est pléiotrope, c’est-à-dire qu’il concerne plusieurs caractères d’intérêt chez la brebis laitière Lacaune. Cette mutation estdéfavorable pour la résistance aux mammites (inflammation de la mamelle), et favorable pour les caractères de croissance et de productionlaitière. Située dans le domaine de liaison très conservé de la protéine SOCS-2, cette mutation ponctuelle (SNP:Single Nucleotide Polymorphism) entraîne la perte de fonctionnalité de la protéine qui est impliquée dans une voie majeure de signalisation chez les mammifères, la voie JAK (Janus Kinase) / STAT (Signal Transducer and Activator of Transcription), régulant un large spectre de cytokines et de facteurs de croissance.Le premier objectif de cette thèse était de proposer la prise en compte de cette mutation dans le contexte de la sélection génomique dans la race Lacaune. Les méthodes d’évaluation pondérées, et l’ajout du SNP SOCS2parmi les marqueurs de la puce, ont apporté des gains de précisions sur les prédictions (respectivement +3,99% et +0,26% en moyenne), laissant envisager la possibilité d’unesélection plus efficace. La méthode Gene Contentétait également intéressante car elle a permis de dissocier la valeur génétique due au gène SOCS2de celle provenant des autres gènes (polygène). Ainsi nous avons montré que la sélection actuelle en race Lacaune permet de diminuer la fréquence de l’allèle défavorable SOCS2tout en améliorant la résistance aux mammites expliquée par le reste du génome.Le deuxième objectif était d’étudier les effets de SOCS-2 sur des caractères d’intérêt non évalués jusqu’à présent: statut infectieux des mamelles et reproduction chez la femelle, et croissance des mâles. Des études d’association avec une puce de 960 SNP, dont le SNP SOCS2, ont confirmé l’effet de SOCS2sur la production laitière et sur l’inflammation mammaire, et ont montré un effet direct sur l’infection à l’aide de nouveaux phénotypes fins (bactériologie, examens cliniques). Des modèles linéaires ont montréque la mutation SOCS2était associée à une augmentation de poids chez les jeunes mâles (+1,5%), comme cela avait été observé chez la brebis, et àune diminution du taux de réussite à l’insémination artificielle chez les brebis (+1,3 fois de risque d’échec), sans effet sur la prolificité. Ces résultats marquants et originaux sur la reproduction ouvrent la voie à des analyses complémentaires pour déterminer notamment à quelstade biologique (ovulation, fécondation, implantation embryonnaire, etc.) la protéine SOCS-2 pourrait intervenir.Le dernier objectif était de déterminer le rôle de la protéine SOCS-2 dans les mécanismes biologiques sous-jacents de l’immunité grâce à l’étude de la réponse de 14 brebis porteuses et non porteuses de la mutation à l’état homozygote dans le cadre d’une infection intra-mammaire par un Staphylococcus aureus. L’ensemble des brebis a présenté deux pics de recrutement de cellules immunitaires versla mamelle à 20 et 88 heures post-inoculation (T20 et T88). Une analyse différentielle des gènes exprimés a confirmé une modification du transcriptome à partir de T16 en lien avec l’activation du système immunitaire,des voies de réparation de l’ADN et d’apoptose cellulaire. Les brebis portant la mutation SOCS2se sont différenciées des brebis sauvages, par une réaction clinique accrue à T56 et une différence de recrutement de lymphocytes T au moment des pics inflammatoires. A T56, un total de 177 gènes était différentiellement exprimé entre les deux génotypes. Une analyse fonctionnelle de ces gènes, et l’intégration entre l’expression génique et les phénotypes, suggèrent que la mutation conduit à une sur activation des voies de transmission de signal et derégulation (interférons, STAT3) avec un impact délétère sur l’état clinique des brebis., In this PhD thesis, we investigated a mutation in the SOCS2gene (Suppressor Of Cytokine Signaling 2) with a pleiotropic effect, i.e.it affects several traits of interest in the Lacaune dairy sheep. This mutation is unfavorable for mastitis resistance (udderinflammation), and favorable for growth and milk production traits.This point mutation (SNP -Single Nucleotide Polymorphism), located in the highly conserved binding domain of the SOCS-2 protein,causes the loss of functionality of the protein that is involved in a major signaling pathway in mammals:the JAK(Janus Kinase) / STAT (Signal Transducer and Activator of Transcription) pathway, regulating a large spectrum of cytokines and growth factors.The first objective of this PhD thesis was to provide a basis for considering this mutation in the context of genomic selection in the Lacaune breed. Weighted evaluation methods, and the addition of the SOCS2SNP among the chip markers, brought gains in accuracy on predictions (+3.99% and +0.26% respectively on average), suggesting the possibility of more efficient selection. The Gene Contentmethod was also interesting because it alloweddissociatingthe genetic value due to the SOCS2gene from that of the other genes (polygene). Thus, we have shown that the current selection in Lacaune breed allows reducing the frequency of the unfavorable SOCS2allele while improving the resistance to mastitis explained by the remaining part of the genome.The second objective was to study the effects of SOCS2on traits of interest not yet investigated: udder infection status and reproduction in females, and growth in males. Association studies with a 960 SNP chip, including SOCS2SNP, confirmed the effect of SOCS2on milk production and mammary inflammation, and showed a direct effect on infection using new fine phenotypes (bacteriology, clinical examinations). Linear models showed that the SOCS2mutation was associated withan increase in weight in young males (+1.5%), as observed in ewes, and a decrease in the artificial insemination success rate in ewes (+1.3 times the risk of failure), with no effect on prolificacy. These striking and original results on reproduction pave the way for further analyses to determine in particular at what biological stage (ovulation, fecundation, embryonic implantation, etc.) the SOCS-2 protein could be involved.The last objective was to determine the role of the SOCS-2 protein in the underlying biological mechanisms of immunity by investigating the response of 14 homozygous carrier and non-carrier ewes to intramammary infection by Staphylococcus aureus. All ewes developed two peaks of immune cell recruitment to the udder at 20 and 88 hours post-inoculation (T20 and T88). Differential analysis of the expressed genes confirmed a modification of the transcriptome from T16 related to the activation of the immune system, the DNA repair and the cellular apoptosis pathways. Ewes carrying the SOCS2mutation were distinguished from wild ewes by an increased clinical response at T56 and a difference in T lymphocyte recruitment at the time of the inflammatory peaks. At T56, a total of 177 genes were differentially expressed between the two genotypes. A functional analysis of these genes, and the integration between gene expression and phenotypes, suggest that the mutation leads to an over activation of signal transmission and regulatory pathways (interferons, STAT3) with a deleterious impact on the clinical status of ewe.

Dans le contexte des études observationnelles, l'estimation de l'effet d'une exposition sur une issue doit gérer l'effet de variables de confusion non mesurées qui affectent à la fois l'exposition et l'issue, sans quoi l'estimation de l'effet causal sera biaisée. En réponse à ce problème de biais de confusion, la discipline de l'économétrie a développé la méthode des variables instrumentales. Cette dernière permet d'inférer un effet causal lorsqu'il y a de la confusion en utilisant des variables ayant la propriété d'être « aléatoires » dans le modèle d'estimation, en plus d'être fortement associées à l'exposition. Ainsi peuvent-elles être utilisées comme des instruments pour randomiser l'exposition. Le contexte d'estimation devient alors comparable à celui d'un essai randomisé, reconnu comme le \textit{gold standard} pour inférer la causalité. La solution se traduit en génétique par la technique de la randomisation mendélienne. Cette technique tire avantage de l'allocation aléatoire des allèles à la naissance et utilise des polymorphismes nucléotidiques (SNPs) comme instruments pour obtenir un phénotype exposition « quasi randomisé ». Par contre, les SNPs doivent satisfaire certaines suppositions de validité qui ne sont pas toutes vérifiables dans un jeu de données. Avec des instruments multiples, l'estimation de l'effet se fait fréquemment par la méthode des moindres carrés en deux étapes. Or, cette méthode suppose que tous les SNPs sont valides, alors qu'il est probable qu'en réalité certains SNPs soient invalides: par exemple, dû au phénomène de la pléiotropie. Pour tenir compte de la possibilité que certains SNPs soient invalides, Kang et al. (2014) ont proposé l'estimateur sisVIVE, qui gère l'effet de l'invalidité des SNPs jusqu'à un seuil de 50% de SNPs invalides et offre une estimation ponctuelle de l'effet causal. Nous contribuons à la littérature de sisVIVE en explorant une méthode bootstrap pour construire des intervalles de confiance. Les résultats de l'étude obtenus par simulations ainsi qu'une application à une base de données de la Biobanque de l'Institut de Montréal seront présentés., In the observational data framework, estimation must account for unmeasured confounders, which affect both exposure and outcome. The reason for this is that if the effect of confounders is not considered, the causal effect estimate will suffer from confounding bias. The econometrics literature addresses this problem by using the instrumental variables method, which enables causal inference when confounding is present by using variables which have the property of being "random" in the posited model and strongly associated with the exposure. Hence, they can serve as instruments to randomize exposure, thus mimicking the setting of randomized control trials—the gold standard to establish causality. With genetic data, the issue can be approached similarly using Mendelian randomization. Mendelian randomization studies apply the instrumental variables approach to genetics by exploiting the natural randomness of allele allocation and using single-nucleotide polymorphisms (SNPs) as instrumental variables to obtain a "quasi-randomized" exposure phenotype. However, the choice of SNPs is not arbitrary; instead, it relies on a set of assumptions for validity, some of which cannot be verified in most empirical settings. When using multiple SNPs as instruments, one might be tempted to use the widely known two-stage least squares estimator, given its simplicity. Yet, the use of this estimator comes at the high price of assuming that all SNPs verify the set of assumptions, whereas in reality, SNPs can be invalid, for instance, due to pleiotropy. To include the possibility that some SNPs may be invalid, Kang et al. (2014) proposed the sisVIVE estimator, which mitigates the effect of invalid SNPs up to a threshold of 50% of invalid SNPs, thereby providing a point estimate of the causal effect. We add to their contribution by exploring a bootstrap method to construct confidence intervals for sisVIVE. Results obtained from simulations and an application to a real dataset from the Montreal Heart Institute's Biobank are presented.

Secondary sexual traits play an irreplaceable role in the reproduction of a range of animals and are used as quality and fitness sensors during pairing of individuals. Expression of these traits, ornamentes, can correlated with an individual's personal and behavioral strategies. In the case of melanin ornaments, not only the relationships with personal individuals are considered, but there is the possible connections with stress resistance and the level of stress responses too. This hypothesis is based on the pleiotropic effect of the melanocortin system, which can be used during melanogenesis but also in the production of hormones that contribute to range of stress responses. The aim of this work was to clarify the relationship between an individual's ornaments, his stress response and individuality in the barn swallows (Hirundo rustica rustica). In this work the relationship between selected ornaments and the stress reaction of the organism, stressed glucose levels measured 15 minutes after a stress stimulus, was tested. Both sexes showed a correlation between area of white tail spots and stress response. Only for males a relationship with the length of outermost tail feathers was found and a correlation with the color saturation of feathers on the throat was shown for females. The second part of...

Propionibacterium freudenreichii is a beneficial Gram-positive bacterium, traditionally used as a cheese-ripening starter, and currently considered as an emerging probiotic. As an example, the P. freudenreichii CIRM-BIA 129 strain recently revealed promising immunomodulatory properties. Its consumption accordingly exerts healing effects in different animal models of colitis, suggesting a potent role in the context of inflammatory bowel diseases. This anti-inflammatory effect depends on surface layer proteins (SLPs). SLPs may be involved in key functions in probiotics, such as persistence within the gut, adhesion to host cells and mucus, or immunomodulation. Several SLPs coexist in P. freudenreichii CIRM-BIA 129 and mediate immunomodulation and adhesion. A mutant P. freudenreichii CIRM-BIA 129ΔslpB (CB129ΔslpB) strain was shown to exhibit decreased adhesion to intestinal epithelial cells. In the present study, we thoroughly analyzed the impact of this mutation on cellular properties. Firstly, we investigated alterations of surface properties in CB129ΔslpB. Surface extractable proteins, surface charges (ζ-potential) and surface hydrophobicity were affected by the mutation. Whole-cell proteomics, using high definition mass spectrometry, identified 1,288 quantifiable proteins in the wild-type strain, i.e., 53% of the theoretical proteome predicted according to P. freudenreichii CIRM-BIA 129 genome sequence. In the mutant strain, we detected 1,252 proteins, including 1,227 proteins in common with the wild-type strain. Comparative quantitative analysis revealed 97 proteins with significant differences between wild-type and mutant strains. These proteins are involved in various cellular process like signaling, metabolism, and DNA repair and replication. Finally, in silico analysis predicted that slpB gene is not part of an operon, thus not affecting the downstream genes after gene knockout. This study, in accordance with the various roles attributed in the literature to SLPs, revealed a pleiotropic effect of a single slpB mutation, in the probiotic P. freudenreichii. This suggests that SlpB may be at a central node of cellular processes and confirms that both nature and amount of SLPs, which are highly variable within the P. freudenreichii species, determine the probiotic abilities of strains.

The yeast S. cerevisiae is the only species able to complete the alcoholic fermentation of grape must which is the main step of the wine-making process. Because of the high technological variability found in this species, selection work is carried out with the aim of using efficient yeasts for the industry. These strains show important differences both in their fermentation kinetics and their metabolite yield which affects the wines quality. In addition to these important phenotypic variations, the phenotypic response of yeasts varies considerably and not homogeneously against environmental variations. Understanding the genetic mechanisms that explain these differentiated responses to environmental variations is a non-trivial scientific question. This is of particular importance in oenology, where the winemaking conditions are highly variable (vintages, grape variety, terroirs, oenological practices...). In order to obtain yeasts ensuring the success of the fermentations under a wide range of conditions, we propose here to better understand these mechanisms of interaction Gene x Environment in an oenological context. The identification of genetic locus (Quantitative Trait Loci (QTL)) controlling quantitative characters is made possible by QTL mapping approaches. These approaches require the study of a large progeny in segregation that must be characterized genetically and phenotypically. The establishment of a statistical link between genotype and phenotype allows the localization of QTLs that have an impact on phenotyped characters. During this thesis a phenotyping method that allows us to follow the fermentations of several hundred individuals has been developed. Thanks to it, two progenies genotyped by whole genome sequencing have been phenotyped in different environmental conditions. This lead to the identification of many QTLs and to the estimation of their impact in strain robustness. The implication of the alleles of 3 genes showing a strong interaction with environment and pleiotropic effects linked to the SO2 metabolism has been proved molecularly. The results achieved are discussed in the context of QTL exploitation for the selection of more efficient yeast., La levure S. cerevisiae est la seule espèce capable de terminer la fermentation alcoolique du jus de raisin qui est l’étape principale de la vinification. A cause de la forte variabilité technologique retrouvée chez cette espèce, des travaux de sélection sont réalisés dans le but d’utiliser des levains performants pour l’industrie. Ces souches montrent des différences importantes à la fois dans leurs cinétiques fermentaires et leur bilans en métabolites, ce qui impacte la qualité des vins. La réponse phénotypique des levures varie également de manière considérable et non homogène face aux variations environnementales. La compréhension des mécanismes génétiques expliquant cette réponse différenciée est une question scientifique non triviale. Elle revêt une importance particulière en oenologie, où les conditions de vinifications sont très changeantes (millésimes, cépages, terroirs, conduites de vinifications…). Afin de pouvoir proposer des levains garantissant le succès des fermentations dans un large éventail de conditions, nous proposons ici de mieux comprendre ces mécanismes d’interaction Gène x Environnement dans un contexte oenologique. L’identification de locus génétiques (Quantitative Trait Loci (QTL)) contrôlant des caractères quantitatifs est rendue possible par des approches de cartographie de QTLs. Celles-ci nécessitent l’étude d’une vaste descendance en ségrégation qui doit être caractérisée sur le plan génétique et phénotypique. L’établissement d’un lien statistique entre des marqueurs génétiques et un phénotype permet la localisation de QTLs influant les caractères étudiés. Au cours de cette thèse, une méthode de phénotypage pour suivre les fermentations de plusieurs centaines d’individus a été mise au point. Grâce à elle, les performances fermentaires de deux descendances génotypées par séquençage à haut débit ont été mesurées en faisant varier les conditions de fermentations. Cela a permis l’identification de nombreux QTLs et d’estimer leur impact sur la robustesse des souches. L’implication des allèles de trois gènes qui montrent une forte interaction avec l’environnement et qui possèdent des effets pléiotropiques liés au métabolisme du SO2 a été prouvée moléculairement. Les résultats obtenus font l’objet d’une discussion générale sur l’utilisation de QTLs pour la sélection de levures plus performantes.

Cardiovascular diseases (CVD) are complex diseases where many environmental and genetic factors are involved. Although the genetic aetiology of the CVD has been extensively investigated the last two decades, alternative approaches are needed in order to keep advancing in the pathophysiology of CVD. In this thesis, we propose an integrative approach to discover new genetic associations potentially involved in CVD. We chose previous GWAS hits and we centred our efforts in studying the pleiotropic and gene-gender interaction effects. Finally, we focused on the implementation of personalized genome-based therapy of the results obtained. New pleiotropic effects were discovered in the IL-6R and ABO genes relating them with different inflammatory and lipid phenotypes. In addition, we studied the gene-gender interaction effects, finding some sex-specific associations in two of the genes studied (ABO and GNB3). Further, we centered our efforts in implementing the results obtained during the thesis at the clinical level. One SNP within the TREM-1 gene was associated with increased levels of its protein and could be used as a predictor or risk biomarker for different diseases. Due to the high potential of this SNP, we applied a European patent and we are planning to start clinical trials in patients. Also, one haplotype in the IL-6R gene could be used in the treatment of personalized medicine. During this thesis, we discovered new gene-phenotype associations involved in CVD and other diseases. Our results help to better understand how the studied genes are exerting their effects at the molecular level. Our results will hopefully be taken into account in future personalized treatments, Les maladies cardiovasculaires (MCV) sont d’une étiologie complexe et elles sont soumises à de nombreux facteurs environnementaux ainsi que génétiques. Malgré les succès obtenus, pendant la dernière décennie, et pour réduire la mortalité CV il est nécessaire l’identification de nouveaux biomarqueurs en utilisant des approches différentes. Cette thèse propose une approche intégrative pour découvrir de nouvelles associations génétiques associés avec les MCV. Nous avons d’abord réuni les résultats existants grâce à des GWAS précédents, puis nous avons recherché la pléiotropie de ces gènes et nous avons dirigé nos efforts vers une possible traduction des résultats obtenus dans l’application clinique. Nous avons détecté les effets pléiotropiques de différent gènes (IL-6R et ABO) avec différents phénotypes lipidiques et inflammatoires. Par ailleurs, nous avons trouvé quelques associations gène-genre intéressantes pour certains gènes étudiés (ABO et GNB3). Concernant l’implémentation clinique des connaissances obtenues par cette thèse, une SNP dans le gène TREM-1, pourrait être utilisé comme un marqueur de risque pour différentes maladies, et nous avons déposé un brevet Européen et nous envisageons de mener des essais cliniques de chez les patients. D’autre part, nous avons détecté une haplotype du gène IL6R qui pourrait être utilisés dans la médecine personnalisée. Nos résultats aident à mieux comprendre comment les gènes étudiés exercent leurs effets au niveau moléculaire, en influant finalement sur l’état des patients souffrant de MCV. Nous espérons que nos résultats vont être pris en compte pour faire progresser la médecine personnalisée

The yeast S. cerevisiae is the only species able to complete the alcoholic fermentation of grape must which is the main step of the wine-making process. Because of the high technological variability found in this species, selection work is carried out with the aim of using efficient yeasts for the industry. These strains show important differences both in their fermentation kinetics and their metabolite yield which affects the wines quality. In addition to these important phenotypic variations, the phenotypic response of yeasts varies considerably and not homogeneously against environmental variations. Understanding the genetic mechanisms that explain these differentiated responses to environmental variations is a non-trivial scientific question. This is of particular importance in oenology, where the winemaking conditions are highly variable (vintages, grape variety, terroirs, oenological practices...). In order to obtain yeasts ensuring the success of the fermentations under a wide range of conditions, we propose here to better understand these mechanisms of interaction Gene x Environment in an oenological context. The identification of genetic locus (Quantitative Trait Loci (QTL)) controlling quantitative characters is made possible by QTL mapping approaches. These approaches require the study of a large progeny in segregation that must be characterized genetically and phenotypically. The establishment of a statistical link between genotype and phenotype allows the localization of QTLs that have an impact on phenotyped characters. During this thesis a phenotyping method that allows us to follow the fermentations of several hundred individuals has been developed. Thanks to it, two progenies genotyped by whole genome sequencing have been phenotyped in different environmental conditions. This lead to the identification of many QTLs and to the estimation of their impact in strain robustness. The implication of the alleles of 3 genes showing a strong interaction with environment and pleiotropic effects linked to the SO2 metabolism has been proved molecularly. The results achieved are discussed in the context of QTL exploitation for the selection of more efficient yeast., La levure S. cerevisiae est la seule espèce capable de terminer la fermentation alcoolique du jus de raisin qui est l’étape principale de la vinification. A cause de la forte variabilité technologique retrouvée chez cette espèce, des travaux de sélection sont réalisés dans le but d’utiliser des levains performants pour l’industrie. Ces souches montrent des différences importantes à la fois dans leurs cinétiques fermentaires et leur bilans en métabolites, ce qui impacte la qualité des vins. La réponse phénotypique des levures varie également de manière considérable et non homogène face aux variations environnementales. La compréhension des mécanismes génétiques expliquant cette réponse différenciée est une question scientifique non triviale. Elle revêt une importance particulière en oenologie, où les conditions de vinifications sont très changeantes (millésimes, cépages, terroirs, conduites de vinifications…). Afin de pouvoir proposer des levains garantissant le succès des fermentations dans un large éventail de conditions, nous proposons ici de mieux comprendre ces mécanismes d’interaction Gène x Environnement dans un contexte oenologique. L’identification de locus génétiques (Quantitative Trait Loci (QTL)) contrôlant des caractères quantitatifs est rendue possible par des approches de cartographie de QTLs. Celles-ci nécessitent l’étude d’une vaste descendance en ségrégation qui doit être caractérisée sur le plan génétique et phénotypique. L’établissement d’un lien statistique entre des marqueurs génétiques et un phénotype permet la localisation de QTLs influant les caractères étudiés. Au cours de cette thèse, une méthode de phénotypage pour suivre les fermentations de plusieurs centaines d’individus a été mise au point. Grâce à elle, les performances fermentaires de deux descendances génotypées par séquençage à haut débit ont été mesurées en faisant varier les conditions de fermentations. Cela a permis l’identification de nombreux QTLs et d’estimer leur impact sur la robustesse des souches. L’implication des allèles de trois gènes qui montrent une forte interaction avec l’environnement et qui possèdent des effets pléiotropiques liés au métabolisme du SO2 a été prouvée moléculairement. Les résultats obtenus font l’objet d’une discussion générale sur l’utilisation de QTLs pour la sélection de levures plus performantes.

Summary Plants need to control transpiration and photosynthesis tightly in order to grow under drought and high temperature. Contrasting genetic-by-environment interactions are exploited by Arabidopsis thaliana to improve stress tolerance., How genetic factors control plant performance under stressful environmental conditions is a central question in ecology and for crop breeding. A multivariate framework was developed to examine the genetic architecture of performance-related traits in response to interacting environmental stresses. Ecophysiological and life history traits were quantified in the Arabidopsis thaliana Ler×Cvi mapping population exposed to constant soil water deficit and high air temperature. The plasticity of the genetic variance–covariance matrix (G-matrix) was examined using mixed-effects models after regression into principal components. Quantitative trait locus (QTL) analysis was performed on the predictors of genotype effects and genotype by environment interactions (G×E). Three QTLs previously identified for flowering time had antagonistic G×E effects on carbon acquisition and the other traits (phenology, growth, leaf morphology, and transpiration). This resulted in a size-dependent response of water use efficiency (WUE) to high temperature but not soil water deficit, indicating that most of the plasticity of carbon acquisition and WUE to temperature is controlled by the loci that control variation of development, size, growth, and transpiration. A fourth QTL, MSAT2.22, controlled the response of carbon acquisition to specific combinations of watering and temperature irrespective of plant size and development, growth, and transpiration rate, which resulted in size-independent plasticity of WUE. These findings highlight how the strategies to optimize plant performance may differ in response to water deficit and high temperature (or their combination), and how different G×E effects could be targeted to improve plant tolerance to these stresses.

International audience

Melanins are the most common pigments incorporated into birds' feathers. Since melanin-based coloration is variable in the extent and intensity, its function is considered to be particularly cryptic and mechanical, hardening the feather structure. However, recently it has been found that melanin- based coloration also has a signaling function, due to the observed correlation between this type of coloration and many physiological and behavioral traits. These relationships follow from the pleiotropic effects of genes regulating the melanocortin system and the physiological effects of testosterone and corticosterone, which are involved in melanogenesis. This work describes genetic and physiological mechanisms of the relationship between melanin-based coloration and behavioral traits in birds and, based on available studies, supports the signaling function of this type of coloration. Melanin-based coloration has been shown to be positively correlated with aggressiveness and dominance, sexual activity, coping with stress, a fast strategy of exploratory behavior, parental care, and anti-predator behavior in many bird species. However, despite intense research interest in this topic in last decade, it is worth noting that more experiments focused on species with a distinct sexual dichromatism or...

Yield is a key trait in pepper and is often measured simultaneously with other traits over several environments. The understanding and genetic improvement of yield may benefit from the joint analysis of yield with its related traits simultaneously. Linear mixed models have emerged as a flexible approach that correctly model underlying variance-covariance structures among the traits and between environments simultaneously. ln this study, we applied four different QTL approaches based on linear mixed model on five yield related pepper traits measured across four environments. We evaluated the performance of the approaches in terms of the number of QTLs detected for each trait and their explained variance. The QTL models are a single-trait single-environment approach (STSE), a multi-trait approach (MT), a multi-environment approach (ME), and a multi-trait multienvironment approach (MTME). We further compared prediction accuracies between STSE and MT models. The predictions were subjected to a five-fold cross validation. Our results showed that multi-trait and/or multi-environment QTL analyses are more powerful and effective to map pleiotropic QTL and QTL by environment interactions th an performing STSE analysis. The multivariate models further showed improvement over STSE in terms of both number of QTLs and the explained variance. MTME clearly outperformed ail the other methods. With MTME, nine QTLs explaining 51% of genetic variation were identified for yield in the autumn trial in Spain as against three, three and six QTLs explaining 37%,29% and 43% from STSE, ME and MT analyses, respectively. The MT model for yield in SP2 had prediction accuracy of 0.53, against 0.42 from the STSE model. These results confirmed that multivariate analyses of traits have better capabilities to unravel complex traits than single trait approach. Our result showed that trait's prediction accuracy depends not only on prediction model of choice and traits genetic architecture but also on the environmen

QTL detection using the regression of phenotypes on transmission probability is largely used when large families phenotyped for Gaussian trait are available. The aim of this thesis from a methodological point of view, is to propose a method for detection of QTL that takes into account the small number of families on the one hand, and the existence of discrete traits on the other. Thus, we propose to answer the first question, an QTL detection approach, integrating in the calculation of genetic merit of genotyped individuals, the performances calculated over n generations of descendants. The use of a 'de-regressed proof' as a phenotype to be analysed, proposed by Weller et al. (1990) and Tribout et al. (2008) is generalized. Next, we present the results of comparisons of a model assuming normality of the data to a thresholds model assuming a continuous distribution underlying the observed distribution in the QTL detection of discrete traits. Here we demonstrate that the discrete model is more accurate and more powerful when the studied trait has three modalities distributed unevenly in the population.In the second part of the thesis, the data analysis of GENECAN protocol was performed. This is to identify genomic regions or quantitative trait locus (QTL) associated with interest traits measured on over-feed mule ducks. The mule duck is an hybrid duck from a female Common duck (Anas Platyrhynchos) and a Muscovy drake (Cairina moschata). Three hundred forty two common ducks designed by back-cross (BC) were generated by crossing a line of Kaiya duck and a heavy line of Pekin duck. These BC females were mated with Muscovy ducks to produce 1600 mules ducks which undergo measures of growth, metabolism during the growth and over-feeding periods, over-feeding, of breast muscle and fatty liver qualities. The phenotypic value of genotyped BC females was estimated for each trait as the average phenotypes of their offspring and weighted by a coefficient of determination (CD) function on the number of offspring and heritability of the studied trait. The genetic map comprised 91 microsatellite markers aggregated into 16 linkage groups (LG) and representing 778 cM. For the uni-trait analysis, twenty-two QTL significant at 1% threshold in chromosome-wide have been mapped. These QTLs are mostly involved in the variability of the breast muscle and fatty liver qualities. Chromosomal regions of interest identified in the framework of this study should be in the future be densified to markers to do the fine mapping., La recherche de QTL par régression des phénotypes sur les probabilités de transmission (modèle Haley-Knott) est une méthode très largement utilisée quand on dispose de grandes familles phénotypées par des caractères gaussiens. L'objectif de cette thèse d'un point de vue méthodologique, est de proposer une méthode de détection de QTL qui prend en compte des effectifs de familles petits d'une part, et l'existence de caractères discrets d'autre part. Ainsi, nous proposons, pour répondre à la première question, une approche de détection de QTL intégrant dans le calcul du mérite génétique des individus marqués, les performances calculées sur n générations de descendants. L'obtention d'un mérite génétique dérégressé comme substitut de phénotypes, proposé notamment par Weller et al (1990) et Tribout et al (2008), est donc généralisée. Ensuite, sont présentés les résultats de comparaisons d'un modèle supposant la normalité des données à un modèle à seuils faisant l'hypothèse d'une distribution continue sous jacente à la distribution observée dans la détection de QTL des caractères discrets. Nous démontrons ici que le modèle discret est plus précis et plus puissant quand le caractère étudié possède trois modalités distribuées de façon déséquilibrée dans la population.Dans la deuxième partie de la thèse, l'analyse des données du protocole GENECAN a été réalisée. Il s'agit d'identifier les régions du génome ou locus à caractère quantitatif (QTL), associées à des caractères d'intérêt mesurés sur des canards mulards gavés. Le canard mulard est un hybride interspécifique obtenu par croisement d'une cane commune (Anas platyrhynchos) et d'un canard de Barbarie (Cairina moschata). Trois cents quarante deux canes communes conçues en back-cross (BC) ont été générées par croisement d'une lignée de canard Kaiya et d'une lignée de canard Pékin lourd. Ces femelles BC ont été accouplées avec des canards de Barbarie pour produire 1600 canards mulards sur lesquels sont effectuées des mesures de croissance, de métabolisme au cours de la période de croissance et du gavage, d'aptitude au gavage et de qualités du magret et du foie gras. La valeur phénotypique des femelles BC marquées a été estimée, pour chaque caractère, comme étant la valeur moyenne des phénotypes de sa progéniture et pondérée par un coefficient de détermination (CD) fonction du nombre de descendants et de l'héritabilité du caractère étudié. Une carte génétique de 91 marqueurs microsatellites réparties sur 16 groupes de liaison (GL) et couvrant un total de 778 cM a été utilisée. Dans le cadre de l'analyse uni-caractère, vingt-deux QTL significatifs à 1% au niveau du chromosome ont été cartographiés. Ces QTLs sont pour la plupart impliqués dans la variabilité de la qualité du magret et du foie gras. Les zones chromosomiques d'intérêt, identifiées dans le cadre de cette étude devront dans le futur, être densifiées en marqueurs pour faire l'objet d'une cartographie fine.

Background / Purpose: Pleiotropic genes have been identified to control simultaneously many processes involved in key functions like growth and reproduction. Mutations in these central genes affect numerous traits, but their impact on response to selection in natural populations, in which the genetic background and environments vary, remains poorly understood. We developed a modeling approach to uncover the variation of a critical trait related to fitness, the vegetative rosette surface development of Arabidopsis thaliana, at different levels of organization and for different processes such as flowering, cellular proliferation and expansion. Main conclusion: Plant growth, phenology and leaf anatomy were analyzed under well-watered and water-deficient conditions. In the joint population, ERECTA gene had common effects on cell patterning. However, differential trade-off effects of ERECTA on whole rosette development were found in each population induced by genotype-dependent epistatic interactions, revealing cryptic genetic variations at different levels. In response to water deficit, ERECTA affected the coordinated response of rosette development, through interaction with the quantitative trait loci (QTL), depending on the population in which the mutation segregates

The development of molecular techniques has offered possibilities to identify quantitative trait loci (QTL). Studies in dairy cattle have mainly focused on milk production traits. This thesis first gives an overview of the main identified QTL for milk production traits. Subsequently, a study to detect QTL affecting 27 conformation traits and functional traits was performed. A granddaughter design consisting of 20 Holstein-Friesian grandsires and 833 sons was analyzed by multi-marker regression. This across-family analysis suggested the presence of 61 QTL. Ten of these QTL exceeded the genomewise significance threshold. These were mainly QTL affecting body size traits and udder conformation traits.When QTL-information is used to select for a certain trait, genetic progress in other traits may be influenced as well, due to pleiotropic effects of QTL, or due to closely linked QTL. A method was developed to identify regions affecting multiple traits. The method is based on the covariance between marker contrasts from single-trait regression analysis for different traits. Application of this method to data on fifteen traits (milk production, udder conformation, udder health and fertility traits) in our granddaughter design resulted in 59 multiple trait quantitative trait regions (MQR). Most MQR were found on BTA 6, 13, 14, 19, 22, 23 and 25.QTL-information can be used in breeding schemes (marker-assisted selection, MAS) to increase the rate of genetic improvement. A number of multi-stage dairy cattle breeding schemes was evaluated, studying the impact of increased preselection using QTL-information. Response in multi-stage MAS-schemes was 4.5% to 31.3% higher than response in corresponding schemes without QTL-information. In some of the MAS-schemes with a reduced number of progeny tested bulls, genetic progress was identical to or higher than genetic progress in the original schemes. The results indicate opportunities to improve current breeding schemes. The gains depend on the amount of genomic variation explained by QTL.Currently available pedigrees and methods offer excellent opportunities to identify more QTL, thus increasing the fraction of the genomic variation explained by QTL. New initiatives, like sequencing the bovine genome, will further facilitate the identification of genomic variation, and its use.

The development of molecular techniques has offered possibilities to identify quantitative trait loci (QTL). Studies in dairy cattle have mainly focused on milk production traits. This thesis first gives an overview of the main identified QTL for milk production traits. Subsequently, a study to detect QTL affecting 27 conformation traits and functional traits was performed. A granddaughter design consisting of 20 Holstein-Friesian grandsires and 833 sons was analyzed by multi-marker regression. This across-family analysis suggested the presence of 61 QTL. Ten of these QTL exceeded the genomewise significance threshold. These were mainly QTL affecting body size traits and udder conformation traits.When QTL-information is used to select for a certain trait, genetic progress in other traits may be influenced as well, due to pleiotropic effects of QTL, or due to closely linked QTL. A method was developed to identify regions affecting multiple traits. The method is based on the covariance between marker contrasts from single-trait regression analysis for different traits. Application of this method to data on fifteen traits (milk production, udder conformation, udder health and fertility traits) in our granddaughter design resulted in 59 multiple trait quantitative trait regions (MQR). Most MQR were found on BTA 6, 13, 14, 19, 22, 23 and 25.QTL-information can be used in breeding schemes (marker-assisted selection, MAS) to increase the rate of genetic improvement. A number of multi-stage dairy cattle breeding schemes was evaluated, studying the impact of increased preselection using QTL-information. Response in multi-stage MAS-schemes was 4.5% to 31.3% higher than response in corresponding schemes without QTL-information. In some of the MAS-schemes with a reduced number of progeny tested bulls, genetic progress was identical to or higher than genetic progress in the original schemes. The results indicate opportunities to improve current breeding schemes. The gains depend on the amount of genomic variation explained by QTL.Currently available pedigrees and methods offer excellent opportunities to identify more QTL, thus increasing the fraction of the genomic variation explained by QTL. New initiatives, like sequencing the bovine genome, will further facilitate the identification of genomic variation, and its use.

The formation of germ layers during gastrulation and the specification and patterning of the body axes are important events in the development of the embryo. The investigations described in this thesis aimed to isolate and characterize the distribution of transcripts of genes, in particular novel genes, that are expressed during the formation of mesoderm and the patterning of the anteroposterior axis in the carp (Cyprinus carpio), a cyprinid teleost fish. Such studies may contribute to a better insight in the molecular mechanisms underlying the above processes, in two respects. Firstly, the characterization of a gene's expression pattern is one of the first steps towards the elucidation of its function. Especially the characterization of novel genes may provide a key to new insights in development. Furthermore, in studies of development, it is of importance to have markers that identify specific cell types, for example early mesodermal precursor cells. The isolation of genes that are specifically expressed in certain cell types provides such markers.Two molecular approaches were chosen for gene isolation. Firstly, we specifically searched for homeobox genes, which encode transcription factors with important regulatory functions during development. A particular class of homeobox genes, the Hox genes, provides cells along the anteroposterior axis with positional information and the expression patterns of members of this class are excellent markers of position on this axis. Our second approach was a subtractive hybridization strategy. It was applied to isolate genes that are differentially expressed between the oocyte and the early segmentation stage, a period during which mesoderm is induced.For the identification of homeobox-containing sequences in a carp early segmentation stage cDNA library, we used a probe that was composed of a mixture of homeobox fragments, produced by PCR. The PCR primers were designed against the most conserved regions of the homeobox. This approach yielded a number of different genes of which two are described in this thesis. The gene cdx1(Chapter 2) is a member of the caudal class of homeobox genes and is expressed in ventrolateral cells of the embryo prior to gastrulation. During gastrulation, transcripts of this gene accumulate in the posterior half of the embryo. The functions of caudal class genes of Drosophila, mouse and Xenopus indicate that genes of this class mediate the specification of posterior positional values in the embryo. Because of their characteristic distribution, cdx1 transcripts are useful markers of (ventro) posterior position in the embryo and have been used as such in the studies of mesoderm formation and anteroposterior patterning in blastoderm explants, performed in our laboratory.A second gene isolated in the search for homeobox genes was the Hoxb-1 gene, described in Chapter 3. This gene belongs to the class of Hox genes, whose members are organized into clusters in the genome of many species. With expression reaching into the hindbrain, the Hoxb-1 gene is one of the most anteriorly expressed Hox genes. Its most prominent expression, especially during segmentation, is found in rhombomere 4. In the late segmentation stage embryo, Hoxb-1 expression is a valuable marker of this rhombomere and the neural crest cells at that level of the hindbrain.Chapter 4 describes the cloning of genes on the basis of their differential gene expression between the oocyte and the early segmentation stage, using a subtractive hybridization strategy. Fifteen genes, identified from the oocyte stage cDNA library, are expressed in early development, when mesoderm induction occurs, and their expression disappears before the beginning of segmentation. From the early segmentation stage cDNA library, 26 genes were selected whose expression was activated during segmentation but not yet in early development, coinciding with the differentiation of the mesoderm and the patterning of the anteroposterior axis. In total 27 genes appeared to code for novel proteins and are therefore candidates for further studies and may provide a better insight into molecular mechanisms underlying developmental processes. Also in light of the large scale mutagenesis screens of zebrafish that have recently been undertaken in a number of laboratories and for which the affected genes yet need to be molecularly identified, it is important that the search for novel genes continues for only few candidate sequences are available so far. The subtractive hybridization strategy described in this thesis appears a worthwhile technique to obtain such candidate sequences.Further investigations of these novel genes were restricted to a detailed characterization of the expression of one gene: cth1.Chapter 5 gives a description of the distribution of the mRNA transcripts of this gene during cleavage, blastula and gastrula stages. Whereas maternal cth1 mRNA is ubiquitously distributed in the blastomeres, the embryonically transcribed cth1 mRNA is expressed after the late blastula stage, in cells at the blastoderm margin which have mesodermal and endodermal fates. The cth1 transcripts disappear around midgastrulation, coinciding with the commitment of cells to the mesodermal (or endodermal) fate. In the cth1 protein, motifs containing three cysteines and one histidine (C3H) are present that are most likely zinc fingers, structures involved in the regulation of expression of target genes. The cth1 mRNA expression pattern and the gene's homology to the TIS11 family, a family of primary response genes whose expression is activated after treatment with for example growth factors, suggest a function for the cth1 gene of maintenance of the cellular potential in cells with mesodermal (and endodermal) fates, and in cells of cleavage stages. By inhibiting the expression of certain target genes, cth1 could prevent or delay the selection of certain differentiation pathways, such as for example the commitment to a mesodermal fate before midgastrulation.Proteins containing C3H motifs are expressed in a number of species.For example in C.elegans, the PIE-1 protein is required to keep germlineblastomeres totipotent during early development, most likely by suppressing the transcription in these blastomeres. In Chapter 6 the literature on the C3H class of proteins is shortly reviewed and the hypothesis is proposed that they may be widely involved in preserving cellular potency in specification events during development.In Chapter 7 the results presented in previous chapters are discussed, with emphasis on the proposed role for cth1 and how its activity could affect the fate of the cells expressing this gene.

Plant viruses are interesting pathogens because they can not exist without their hosts and exploit the plant machinery for their multiplication. Fundamental knowledge on viral processes is of great importance to understand, prevent and control virus infections which can cause drastic losses in crops. In this thesis, cowpea mosaic virus (CPMV) was studied. This virus consists of two, icosahedral particles that each carry a distinct single stranded RNA molecule of positive polarity. Several years of research have revealed much information on the genomic organisation, the strategy of gene expression and the multiplication processes of CPMV, which are described in Chapter 1, but also many aspects remain to be elucidated.To study individual viral processes, like replication, encapsidation or cell to cell movement, transgenic plants can be generated that express individual viral genes like the replicase, coat protein or movement protein gene. A prerequisite in this approach is the presence of an efficient and reliable plant regeneration and transformation system. (CPMV) 5 natural host is the tropical grain legume cowpea, Vigna unguiculata, a plant species that is recalcitrant at regeneration. Although in experiments described in Chapter 2 fertile plants could be regenerated from nodal thin cell layer segments, the explants were not competent for Agrobacterium-mediated transformation. Possibly in further studies, these nodal explants could prove suited for another transformation method.Therefore, tobacco, which is also a host for CPMV and highly competent for regeneration and transformation, was preferred as the species to generate transgenic plants carrying CPMV specific genes. Especially the CPMV movement proteins (MP) genes appealed to us for overexpression studies. CPMV cell to cell movement is enabled by the CPMV MPs that act to modify plasmodesmata. They are assumed to channel plasmodesmata with MP-containing tubular structures and through or with these tubules virus particles are transported to adjacent cells. To obtain more information on the plasmodesmatal modifications brought about by the MPs, transgenic tobacco plants were generated that carried the MP gene under the control of either a constitutive or an inducible 35S promoter. However, in none of these plants the MPs were expressed to detectable levels (Chapter 3). Using the potato virus X (PVX)-based expression vector, accumulation of CPMV MPs was observed in the form of tubular structures extending from the surface of infected protoplasts into the medium. These PVX-derivatives look promising for providing effective tools in future studies on the effects of the CPMV MPs in plants.Studies on MP functioning could involve complementation experiments with a CPMV mutant that is defective in cell to cell movement. In experiments described in Chapter 4 is was analysed by a molecular approach whether the CPMV mutant N123, that was first described in 1976, could be used to this effect. As the basis of the N123 specific phenotype was found not only to rest in the movement protein gene but also in one of the two coat protein genes, this mutant seemed not very suitable for complementation studies. Presumably a recently developed CPMV mutant in which the MP gene has been replaced by the fluorescent marker protein GFP (green fluorescent protein), will be a more appropriate tool.Transgenic Nicotiana benthamiana plants that were expressing either the CPMV MP or the replicase gene under the control of a constitutive promoter, were found to exhibit a resistant phenotype when inoculated with CPMV (Chapter 5). Protoplast studies revealed that the resistance occurred as full immunity and was maintained in the cell. Resistance was specific to viruses highly homologous to CPMV, and in addition it was found to be specifically directed against the replication of the CPMV segment of which the transgene was derived (Chapter 5). Pathogen derived resistance can be mediated either by the protein encoded by the transgene or by the transcribed mRNA. Protein -mediated resistance generally offers moderate protection against a broad range of viruses, while RNA-mediated resistance results in immunity at the cellular level. Resistance obtained in transgenic plants transformed with defective genes confirmed that an RNA-based mechanism was underlying the highly specific transgenic resistance against CPMV (Chapter 6).Specifically in the resistant lines, the transgene mRNA steady state levels were low compared to the relative transgene nuclear transcription rates (Chapter 6). This indicated that resistance occurs from a specific, cytoplasmic RNA turnover mechanism. This process can be regarded as a post- transcriptional gene-silencing process, that is primarily induced on the transgene mRNAs but to which also incoming, homologous CPMV genomes fall victim. In addition, heterologous RNA molecules, like PVX genomes, that contain the sequences corresponding to the transgene, are eliminated (Chapter 6). By inserting sequences homologous to only parts of the transgene in the genome of PVX and studying the fate of these recombinant genomes, it was shown that the degradation process is primarily targeted to a defined region of the transgene mRNA, the 3' region. Further analyses revealed that degradation can occur at various sites within this 3' region and that not a specific sequence or structure is of predominant importance. We observed that small inserts, like of only 60 nucleotides, can tag recombinant PVX molecules for the elimination process, albeit with reduced efficiency, which suggested that the RNA turnover process carries quantitative features.On the intruiging question why post-transcriptional gene-silencing is induced in only some of the transgenic lines, we revealed (Chapter 6) that the organisation of integrated transgene sequences has an important role. Transformation with a transgene containing a directly repeated MP gene, increased the frequency at which resistant lines arise to 60%, compared to 20% of resistant lines that occur upon transformation with a transgene with a single MP gene. Thus, the resistance process seems influenced by qualitative features of the integrated transgenes. Also, it was observed that resistance concurred with extensive methylation at the transcribed transgene sequences (Chapter 6), which could indicate an essential role of methylation at transcribed sequences in obtaining RNA-mediated pathogen derived resistance.From these observations and from data described in literature, a model for RNA-mediated virus resistance was made and presented in Chapter 6. In Chapter 7, the post-transcriptional gene-silencing phenomenon is discussed in more details and in addition an approach is presented by which the process could be exploited to efficiently engineer virus resistance or study plant gene expression.

Plant viruses are interesting pathogens because they can not exist without their hosts and exploit the plant machinery for their multiplication. Fundamental knowledge on viral processes is of great importance to understand, prevent and control virus infections which can cause drastic losses in crops. In this thesis, cowpea mosaic virus (CPMV) was studied. This virus consists of two, icosahedral particles that each carry a distinct single stranded RNA molecule of positive polarity. Several years of research have revealed much information on the genomic organisation, the strategy of gene expression and the multiplication processes of CPMV, which are described in Chapter 1, but also many aspects remain to be elucidated.To study individual viral processes, like replication, encapsidation or cell to cell movement, transgenic plants can be generated that express individual viral genes like the replicase, coat protein or movement protein gene. A prerequisite in this approach is the presence of an efficient and reliable plant regeneration and transformation system. (CPMV) 5 natural host is the tropical grain legume cowpea, Vigna unguiculata, a plant species that is recalcitrant at regeneration. Although in experiments described in Chapter 2 fertile plants could be regenerated from nodal thin cell layer segments, the explants were not competent for Agrobacterium-mediated transformation. Possibly in further studies, these nodal explants could prove suited for another transformation method.Therefore, tobacco, which is also a host for CPMV and highly competent for regeneration and transformation, was preferred as the species to generate transgenic plants carrying CPMV specific genes. Especially the CPMV movement proteins (MP) genes appealed to us for overexpression studies. CPMV cell to cell movement is enabled by the CPMV MPs that act to modify plasmodesmata. They are assumed to channel plasmodesmata with MP-containing tubular structures and through or with these tubules virus particles are transported to adjacent cells. To obtain more information on the plasmodesmatal modifications brought about by the MPs, transgenic tobacco plants were generated that carried the MP gene under the control of either a constitutive or an inducible 35S promoter. However, in none of these plants the MPs were expressed to detectable levels (Chapter 3). Using the potato virus X (PVX)-based expression vector, accumulation of CPMV MPs was observed in the form of tubular structures extending from the surface of infected protoplasts into the medium. These PVX-derivatives look promising for providing effective tools in future studies on the effects of the CPMV MPs in plants.Studies on MP functioning could involve complementation experiments with a CPMV mutant that is defective in cell to cell movement. In experiments described in Chapter 4 is was analysed by a molecular approach whether the CPMV mutant N123, that was first described in 1976, could be used to this effect. As the basis of the N123 specific phenotype was found not only to rest in the movement protein gene but also in one of the two coat protein genes, this mutant seemed not very suitable for complementation studies. Presumably a recently developed CPMV mutant in which the MP gene has been replaced by the fluorescent marker protein GFP (green fluorescent protein), will be a more appropriate tool.Transgenic Nicotiana benthamiana plants that were expressing either the CPMV MP or the replicase gene under the control of a constitutive promoter, were found to exhibit a resistant phenotype when inoculated with CPMV (Chapter 5). Protoplast studies revealed that the resistance occurred as full immunity and was maintained in the cell. Resistance was specific to viruses highly homologous to CPMV, and in addition it was found to be specifically directed against the replication of the CPMV segment of which the transgene was derived (Chapter 5). Pathogen derived resistance can be mediated either by the protein encoded by the transgene or by the transcribed mRNA. Protein -mediated resistance generally offers moderate protection against a broad range of viruses, while RNA-mediated resistance results in immunity at the cellular level. Resistance obtained in transgenic plants transformed with defective genes confirmed that an RNA-based mechanism was underlying the highly specific transgenic resistance against CPMV (Chapter 6).Specifically in the resistant lines, the transgene mRNA steady state levels were low compared to the relative transgene nuclear transcription rates (Chapter 6). This indicated that resistance occurs from a specific, cytoplasmic RNA turnover mechanism. This process can be regarded as a post- transcriptional gene-silencing process, that is primarily induced on the transgene mRNAs but to which also incoming, homologous CPMV genomes fall victim. In addition, heterologous RNA molecules, like PVX genomes, that contain the sequences corresponding to the transgene, are eliminated (Chapter 6). By inserting sequences homologous to only parts of the transgene in the genome of PVX and studying the fate of these recombinant genomes, it was shown that the degradation process is primarily targeted to a defined region of the transgene mRNA, the 3' region. Further analyses revealed that degradation can occur at various sites within this 3' region and that not a specific sequence or structure is of predominant importance. We observed that small inserts, like of only 60 nucleotides, can tag recombinant PVX molecules for the elimination process, albeit with reduced efficiency, which suggested that the RNA turnover process carries quantitative features.On the intruiging question why post-transcriptional gene-silencing is induced in only some of the transgenic lines, we revealed (Chapter 6) that the organisation of integrated transgene sequences has an important role. Transformation with a transgene containing a directly repeated MP gene, increased the frequency at which resistant lines arise to 60%, compared to 20% of resistant lines that occur upon transformation with a transgene with a single MP gene. Thus, the resistance process seems influenced by qualitative features of the integrated transgenes. Also, it was observed that resistance concurred with extensive methylation at the transcribed transgene sequences (Chapter 6), which could indicate an essential role of methylation at transcribed sequences in obtaining RNA-mediated pathogen derived resistance.From these observations and from data described in literature, a model for RNA-mediated virus resistance was made and presented in Chapter 6. In Chapter 7, the post-transcriptional gene-silencing phenomenon is discussed in more details and in addition an approach is presented by which the process could be exploited to efficiently engineer virus resistance or study plant gene expression.

The formation of germ layers during gastrulation and the specification and patterning of the body axes are important events in the development of the embryo. The investigations described in this thesis aimed to isolate and characterize the distribution of transcripts of genes, in particular novel genes, that are expressed during the formation of mesoderm and the patterning of the anteroposterior axis in the carp (Cyprinus carpio), a cyprinid teleost fish. Such studies may contribute to a better insight in the molecular mechanisms underlying the above processes, in two respects. Firstly, the characterization of a gene's expression pattern is one of the first steps towards the elucidation of its function. Especially the characterization of novel genes may provide a key to new insights in development. Furthermore, in studies of development, it is of importance to have markers that identify specific cell types, for example early mesodermal precursor cells. The isolation of genes that are specifically expressed in certain cell types provides such markers.Two molecular approaches were chosen for gene isolation. Firstly, we specifically searched for homeobox genes, which encode transcription factors with important regulatory functions during development. A particular class of homeobox genes, the Hox genes, provides cells along the anteroposterior axis with positional information and the expression patterns of members of this class are excellent markers of position on this axis. Our second approach was a subtractive hybridization strategy. It was applied to isolate genes that are differentially expressed between the oocyte and the early segmentation stage, a period during which mesoderm is induced.For the identification of homeobox-containing sequences in a carp early segmentation stage cDNA library, we used a probe that was composed of a mixture of homeobox fragments, produced by PCR. The PCR primers were designed against the most conserved regions of the homeobox. This approach yielded a number of different genes of which two are described in this thesis. The gene cdx1(Chapter 2) is a member of the caudal class of homeobox genes and is expressed in ventrolateral cells of the embryo prior to gastrulation. During gastrulation, transcripts of this gene accumulate in the posterior half of the embryo. The functions of caudal class genes of Drosophila, mouse and Xenopus indicate that genes of this class mediate the specification of posterior positional values in the embryo. Because of their characteristic distribution, cdx1 transcripts are useful markers of (ventro) posterior position in the embryo and have been used as such in the studies of mesoderm formation and anteroposterior patterning in blastoderm explants, performed in our laboratory.A second gene isolated in the search for homeobox genes was the Hoxb-1 gene, described in Chapter 3. This gene belongs to the class of Hox genes, whose members are organized into clusters in the genome of many species. With expression reaching into the hindbrain, the Hoxb-1 gene is one of the most anteriorly expressed Hox genes. Its most prominent expression, especially during segmentation, is found in rhombomere 4. In the late segmentation stage embryo, Hoxb-1 expression is a valuable marker of this rhombomere and the neural crest cells at that level of the hindbrain.Chapter 4 describes the cloning of genes on the basis of their differential gene expression between the oocyte and the early segmentation stage, using a subtractive hybridization strategy. Fifteen genes, identified from the oocyte stage cDNA library, are expressed in early development, when mesoderm induction occurs, and their expression disappears before the beginning of segmentation. From the early segmentation stage cDNA library, 26 genes were selected whose expression was activated during segmentation but not yet in early development, coinciding with the differentiation of the mesoderm and the patterning of the anteroposterior axis. In total 27 genes appeared to code for novel proteins and are therefore candidates for further studies and may provide a better insight into molecular mechanisms underlying developmental processes. Also in light of the large scale mutagenesis screens of zebrafish that have recently been undertaken in a number of laboratories and for which the affected genes yet need to be molecularly identified, it is important that the search for novel genes continues for only few candidate sequences are available so far. The subtractive hybridization strategy described in this thesis appears a worthwhile technique to obtain such candidate sequences.Further investigations of these novel genes were restricted to a detailed characterization of the expression of one gene: cth1.Chapter 5 gives a description of the distribution of the mRNA transcripts of this gene during cleavage, blastula and gastrula stages. Whereas maternal cth1 mRNA is ubiquitously distributed in the blastomeres, the embryonically transcribed cth1 mRNA is expressed after the late blastula stage, in cells at the blastoderm margin which have mesodermal and endodermal fates. The cth1 transcripts disappear around midgastrulation, coinciding with the commitment of cells to the mesodermal (or endodermal) fate. In the cth1 protein, motifs containing three cysteines and one histidine (C3H) are present that are most likely zinc fingers, structures involved in the regulation of expression of target genes. The cth1 mRNA expression pattern and the gene's homology to the TIS11 family, a family of primary response genes whose expression is activated after treatment with for example growth factors, suggest a function for the cth1 gene of maintenance of the cellular potential in cells with mesodermal (and endodermal) fates, and in cells of cleavage stages. By inhibiting the expression of certain target genes, cth1 could prevent or delay the selection of certain differentiation pathways, such as for example the commitment to a mesodermal fate before midgastrulation.Proteins containing C3H motifs are expressed in a number of species.For example in C.elegans, the PIE-1 protein is required to keep germlineblastomeres totipotent during early development, most likely by suppressing the transcription in these blastomeres. In Chapter 6 the literature on the C3H class of proteins is shortly reviewed and the hypothesis is proposed that they may be widely involved in preserving cellular potency in specification events during development.In Chapter 7 the results presented in previous chapters are discussed, with emphasis on the proposed role for cth1 and how its activity could affect the fate of the cells expressing this gene.

This thesis describes a molecular study of the genetics ofL-arabinan degradation in Aspergillus niger and Aspergillus nidulans. These saprophytic hyphal fungi produce an extracellular hydrolytic enzyme system to depolymerize the plant cell wall polysaccharideL-arabinan. Chapter 1 surveys the occurrence, properties and applications ofL-arabinanolytic enzymes (arabinases). The A.niger system, which constitutes an endolytic endo-1,5-α-L-arabinase (ABN A) and two distinctα-L-arabinofuranosidases (ABF A and ABF B), has been a frequent subject of investigation in the past and represents the best characterizedL-arabinanolytic system to date. These three enzymes are all glycosylated. Current knowledge on the induction of fungal arabinase expression is summarized in this Chapter. Furthermore, the structure of the polysaccharide substrate and its function in the plant cell wall matrix are introduced.In Chapters 2 to 5, the cloning and characterization of the structural genes coding for the three glycosyl hydrolases from the A. nigerL-arabinan-degrading complex are described. A. niger abf A and abf B ar e the first eukaryotic ABF-encoding genes to be isolated and sequenced, abn A is the first ABN-encoding gene published. Chapter 2 reports on the isolation of the abf A gene encoding ABF A, the minor extracellular ABF. This gene could be cloned by utilizing ABF Aspecific cDNA as the probe. This cDNA was immunochemically identified from a cDNA library generated fromL-arabitol-induced myceliurn of an A. nigerD-xylulose kinase mutant. This mutant is unable to grow onL-arabitol and features enhanced expression of all three arabinases when transferred to medium containing this pentitol as sole carbon source. In Chapter 3 , the cloning of the ABN A-encoding gene (abn A) is described. This gene was isolated following the same strategy as with abf A, although a second cDNA library had to be generated first. The induction process was immunochemically monitored in order to establish the proper induction conditions for the new library. The abn A gene and the gene product were characterized by DNA sequence analyses of the cloned genomic DNA and the cDN A. The N-terminal amino acid sequences of ABN A and a CNBr-derived peptide were determined. Several transcription initiation sites and one polyadenylation site could be identified. The structural region codes for a protein of 321 amino acids and is interrupted by three introns. Extracellular ABN A consists of 302 amino acid residues with a deduced molecular weight of 32.5 kDa and a theoretical pl of 3.5. For the protein, an apparent pl of 3.0 and an apparent molecular weight of 43 kDa, determined upon SDS-PAGE, were previously reported. Chapter 4 documents the isolation and characterization of the abf B gene, coding for the major extracellular ABF. The determination of N-terminal amino acid sequences from ABF B and CNBr-generated peptides allowed the design of deoxyoligonucleotide mixtures which enabled the cloning of abf B. When utilized as primers in a polymerase chain reaction (PCR), ABF B-specific amplification products emerged, one of which was used to probe the gene. The abf B gene and the gene product were characterized by DNA sequence analyses of the cloned genomic DNA and of ABF B- specific cDNA isolated from the library described in Chapter 3. Several transcription initiation sites and one polyadenylation site could be identified. The structural region is a single open reading frame and codes for a protein of 499 amino acids. The mature enzyme consists of 481 amino acid residues with a deduced molecular weight of 50.7 kDa and a theoretical pl of 3.8. An apparent pl of 3.5 and an apparent molecular weight of 67 kDa, determined upon SDS-PAGE, were previously reported. The abf B gene product was suggested to be identical to the ABF purified and characterized by Kaji and Tagawa (Biochim Biophys Acta 207 : 456-464 (1970)). Considering the non-amino acid content of the latter protein, a molecular weight of 64 kDa could be deduced for ABF B. In Chapter 5 , the abf A gene and its gene product were characterized by DNA sequence analyses of the genomic DNA and of the cDNA for which the isolation was described in Chapter 2. The N-terminal amino acid sequences of ABF A and a CNBr-derived peptide were determined. One transcription initiation site and two polyadenylation sites could be identified. The structural region is interrupted by seven introns and codes for a protein of 628 amino acids. Mature ABF A consists of 603 amino acid residues with a deduced molecular weight of 65.4 kDa and a theoretical pl of 3.7. For this ABF, an apparent pi of 3.3 and an apparent molecular weight of 83 kDa, determined upon SDS-PAGE, were previously documented.Although the three enzymes are all active against (1->5)-α-glycosidic bonds betweenL-arabinofuranosides, ABF A, ABF B and ABN A are genetically unrelated. ABF A was found to be N -glycosylated whereas ABF B and ABN A were not - these enzymes are only O -glycosylated. For each gene, arabinaseoverproducing strains were generated by introducing multiple gene copies in A.niger or in A.nidulans uridine auxotrophic strains through co-transformation. Transformants were isolated upon primary selection for uridine prototrophy. Subsequent overproduction of the genes introduced was demonstrated in these recombinant strains upon growth on sugar beet pulp, both immunochemically and by assaying enzyme activity. abf A was shown to be expressed in the heterologous host A.nidulans, despite the absence of an abf A gene equivalent in this organism. High-copy number A.niger abf B transformants featured impaired secretion of other extracellular proteins upon growth on sugar beet pulp. ABN A overproduction was found to be limited to approximately five times the wild-type level in A.niger abn A transformants, but not in A.nidulans transformants. Such a limitation was not observed in case of the ABFs.In Chapters 5 and 6, the regulation ofL-arabinan degradation is addressed. The structural genes seem to be regulated mainly at the transcriptional level. Additional copies of either A13F-encoding gene in A.niger were shown to result in a reduction, but not in total silencing of the expression of the wild-type ABN Aencoding gene upon induction with either sugar beet pulp orL-arabitol ( Chapter 5 ). The reduction of the expression level of abn A correlated with the abf gene dosage. The repression effected by extra abf B gene copies was more stringent and more persistent than that elicited by additional abf A copies. Although observed with both inducers, these phenomena were more outspoken and more persistent on sugar beet pulp. Similar, but more moderate effects were observed towards the expression of the other abf gene in multiple copy abf A- and abf B-transformants. It was proposed that the abf genes titrate two distinct gene activators both involved in coordination of arabinase gene expression. However, the three genes were shown to respond differently upon a mycelial transfer toL-arabitol-containing medium, indicating that gene-specific factors are also involved. Four distinct sequence motifs were found in common in the promoter regions of the three genes. One of these elements is identical to the A.nidulans CREA-motif, which has been shown to mediate carbon catabolite repression on several A.nidulans enzyme systems. Arabinase expression in A.niger is known to be repressed in the presence ofD-glucose. Two other motifs are highly homologous to cAMP-responsive elements described in other organisms. For the fourth motif no functional analogues could be found, but the element was found to be present in several other fungal genes which are not involved inL-arabinan degradation at all. It is therefore likely that none of these common elements confer system-specific regulation.The presumed involvement ofL-arabitol in the induction process of fungal arabinases was further emphasized by the induction characteristics of an A. nidulans mutant unable to grow on the end-product ofL-arabinan degradation,L-arabinose, nor onL-arabitol ( Chapter 6).L-Arabitol is an intermediate ofL-arabinose catabolism in Aspergilli. This mutant was shown to lack NAD +-dependentL-arabitol dehydrogenase activity resulting inL-arabitol accumulation, both intracellularly and in the culture medium, wheneverL-arabinose is present. Upon submerged growth on various carbon sources in the presence ofL-arabinose, the mutant featured enhanced expression of the enzymes involved in extracellularL-arabinan degradation, and of those of the intracellularL-arabinose catabolism. The co-substrates on which the mutant secreted large amounts of arabitol simultaneously exhibited high arabinase expression and featured reduced growth.L-Arabitol secretion and enzyme production were also observed on a mixed carbon source ofD-glucose andL-arabinose, resulting in normal growth. Hence, in the presence ofL- arabinose, the carbon catabolite repression conferred byD-glucose in the wild-type, is overruled in the mutant.In Chapter 7 , ABN A is shown to have remote sequence similarity with four bacterial xylanolytic glycosyl hydrolases (threeβ-D-xylosidases and an endo-1,4-β-D-xylanase), three of which feature activity against para -nitrophenyl-α-L-arabinofuranoside. This synthetic compound is commonly utilized to assay potential ABF activity, whereas it is known to be an inhibitor of the fourth enzyme. The homology became evident only after multi pie-sequence alignments and hydrophobic cluster analysis. It was proposed that these enzymes share a binding site for a terminal non-reducingα-linkedL-arabinofuranosyl residue and that they all belong to glycosyl hydrolase family 43. Implications from these suggestions were discussed. The ABFs could not be assigned to an established glycosyl hydrolase family.Based on theL-arabinolytic system of the brown-rot fungus Monilinia fructigena, the sequence similarity found amongst ABF A and bacterial pullulan-degrading enzymes, and ABF expression levels under carbon starvation conditions and onD-glucose as the carbon source, distinct functions inL-arabinan and plant cell-wall degradation were proposed for ABF A and ABF B. ABF A would be essentially cell-wall associated and act to degradeL-arabinan fragments generated by ABN A. ABF B activity would be important for the primary release of small amounts ofL-arabinose which initiate induction of various endolytic systems to degrade plant cell walls, and thus function in substrate sensing. In line with these considerations, the involvement of other, not yet identified glycosyl hydrolases inL-arabinan degradation by A.niger was suggested.Induction and repression of arabinase gene expression are further discussed in Chapter 7 . The results of the studies in A.niger (Chapter 5) and A.nidulans (Chapter 6) were interpreted in a mutual context. The identity of the lowmolecular-weight compound directly responsible for induction of arabinase gene expression, was addressed. BothL-arabinose andL-arabitol are likely candidates to fulfil such a role. However, it was not possible to weigh the actual inductive capacities ofL-arabinose andL-arabitol due to their in vivo convertibility and the carbon catabolite repression elicited by the pentose. Competition for such a compound provides an alternative explanation for the phenomena observed in Chapter 5. The involvement of the transcriptional repressor CREA in arabinase gene expression is not limited to the direct repression of structural and regulatory genes of theL-arabinan-degrading system. It also plays a role in inducer exclusion and end-product repression, two processes shown to be eminently involved in the regulation ofL-arabinan degradation in wild-type A.nidulans. Fungal growth rate was suggested to be related to derepression of theL-arabinan-degrading system. The possible involvement of cAMP in arabinase gene expression, as suggested by the presence of potential cis -acting cAMP-responsive elements in the structural genes, was considered. Various ways by which cAMP might modulate arabinase synthesis were surveyed.

In this thesis, gene-expression and gene silencing were examined for Granule Bound Starch Synthase (GBSS) which catalyses the formation of amylose and Branching Enzyme (BE) which catalyses the formation of amylopectin. The (GBSS) deficient, with iodine, red staining amylose-free (amf) potato mutant was used in order to facilitate the experiments; GBSS-gene expression and BE-gene silencing resulted in blue staining starch.A dosage effect of the wild-type GBSS-allele on GBSS-activity and amylose content was found in a tetraploid dosage population. The presence of amylose had a distinct influence on the physico-chemical properties of the starch.Insertion of the wild-type GBSS-gene in the amf -mutant resulted in fully and partially complemented plants. The minimum number of independently segregating active GBSS-inserts was estimated by genetic analysis after microspore staining. The complemented phenotype was normally transmitted to the F1 for the fully complemented, but not for the partially complemented plants. For one plant the partial complementation was correlated with the presence of a block of five inserts. This block was also capable of inhibiting the endogenous GBSS-gene in a wild- type. This indicates that partial complementation can be caused by co-suppression, besides low transgene expression. Co-suppression was also obtained when the full size (GBSS) sequence or (GBSS) cDNA were introduced in a wild-type potato.Introduction of the distal 1.5 kb cDNA coding for BE in both sense and antisense orientation resulted in several transgenic plants with a small blue core in these starch granules. This could indicate the presence of loosely branched amylopectin. in the core of the starch granules. The expression of the endogenous BE-gene was largely or fully inhibited as judged by the absence of BE mRNA and protein. This did not result in a measurable effect on the branching degree, but resulted in altered physico-chemical properties of the starch compared to amf -starch.It is concluded that this study shows that the amf -mutant is successfully used as a model plant to examine different aspects of gene expression.

Bacteria of the genera Rhizobium, Bradyrhizobium or Azorhizobium secrete lipo-oligosaccharide signal molecules, which play a pivotal role in the induction of early steps of root nodule formation on legumes. In these nodules the bacteria are hosted and a proper environment is provided for the bacteria to fix atmospheric nitrogen into ammonia, making the plant for its growth independent of nitrogen compounds in the soil.The goal of the research described in this thesis is to provide insight in the mechanism by which the Nod factors secreted by Rhizobium bacteria initiate root nodule formation. Such kind of study requires a plant-bacteria system in which both the morphological and molecular changes induced by the Nod factor can be examined. Working on the Nod factors secreted by Rhizobium leguminosarum bv. viciae the choice of the host plant is limited to the genera Pisum, Lathyrus, Lens and Vicia. We decided that Viciasativa (vetch) would be the most useful plant for our studies, since the plant is small and various morphological changes, like root hair deformation and the formation of nodule primordia, that are induced by the Nod factors can easily be observed (Chapter 1).Nod factors can induce the expression of early nodulin genes. These genes are expressed during different developmental stages of root nodule formation and the expression of these genes can therefore be used as molecular markers of root nodule development and Nod factor induced processes. To study Nod factor induced plant responses in V.sativa on the molecular level, the homologues of the early nodulin genes of pea (Pisum sativum) were isolated from V. sativa and their expression pattern was studied by in situ hybridization during root nodule development (Chapter 2).To obtain information about the pathway from Nod factor to early nodulin gene expression, the mechanisms controlling ENOD12 expression were studied. With a promoter analysis in transgenic V.hirsuta root nodules we have identified that the 200 bp immediately upstream of the transcription start are sufficient to induce nodule specific and Nod factor induced expression (Chapter 3). For the isolation of transcription factors involved in controlling ENOD12 expression an expression library was screened and a preliminary characterization of cDNA clones encoding polypeptides that bind to the PsENOD12 promoter is described in Chapter 4.In Chapter 5 early nodulin gene expression during Nod factor induced morphological changes, like root hair deformation and nodule primordia induction, was studied. Furthermore we examined whether RNA and protein synthesis are required for root hair deformation and for the activation of the early nodulin genes. These studies provided new insights about the mode of action of Nod factors.In Chapter 6 is discussed to what extent V. sativa is a suitable host plant to study the mode of action of Nod factors and in which way the studies reported in this thesis have contributed to elucidate the mechanism by which Nod factors induce a diversity of plant responses.

In this thesis, gene-expression and gene silencing were examined for Granule Bound Starch Synthase (GBSS) which catalyses the formation of amylose and Branching Enzyme (BE) which catalyses the formation of amylopectin. The (GBSS) deficient, with iodine, red staining amylose-free (amf) potato mutant was used in order to facilitate the experiments; GBSS-gene expression and BE-gene silencing resulted in blue staining starch.A dosage effect of the wild-type GBSS-allele on GBSS-activity and amylose content was found in a tetraploid dosage population. The presence of amylose had a distinct influence on the physico-chemical properties of the starch.Insertion of the wild-type GBSS-gene in the amf -mutant resulted in fully and partially complemented plants. The minimum number of independently segregating active GBSS-inserts was estimated by genetic analysis after microspore staining. The complemented phenotype was normally transmitted to the F1 for the fully complemented, but not for the partially complemented plants. For one plant the partial complementation was correlated with the presence of a block of five inserts. This block was also capable of inhibiting the endogenous GBSS-gene in a wild- type. This indicates that partial complementation can be caused by co-suppression, besides low transgene expression. Co-suppression was also obtained when the full size (GBSS) sequence or (GBSS) cDNA were introduced in a wild-type potato.Introduction of the distal 1.5 kb cDNA coding for BE in both sense and antisense orientation resulted in several transgenic plants with a small blue core in these starch granules. This could indicate the presence of loosely branched amylopectin. in the core of the starch granules. The expression of the endogenous BE-gene was largely or fully inhibited as judged by the absence of BE mRNA and protein. This did not result in a measurable effect on the branching degree, but resulted in altered physico-chemical properties of the starch compared to amf -starch.It is concluded that this study shows that the amf -mutant is successfully used as a model plant to examine different aspects of gene expression.

Bacteria of the genera Rhizobium, Bradyrhizobium or Azorhizobium secrete lipo-oligosaccharide signal molecules, which play a pivotal role in the induction of early steps of root nodule formation on legumes. In these nodules the bacteria are hosted and a proper environment is provided for the bacteria to fix atmospheric nitrogen into ammonia, making the plant for its growth independent of nitrogen compounds in the soil.The goal of the research described in this thesis is to provide insight in the mechanism by which the Nod factors secreted by Rhizobium bacteria initiate root nodule formation. Such kind of study requires a plant-bacteria system in which both the morphological and molecular changes induced by the Nod factor can be examined. Working on the Nod factors secreted by Rhizobium leguminosarum bv. viciae the choice of the host plant is limited to the genera Pisum, Lathyrus, Lens and Vicia. We decided that Viciasativa (vetch) would be the most useful plant for our studies, since the plant is small and various morphological changes, like root hair deformation and the formation of nodule primordia, that are induced by the Nod factors can easily be observed (Chapter 1).Nod factors can induce the expression of early nodulin genes. These genes are expressed during different developmental stages of root nodule formation and the expression of these genes can therefore be used as molecular markers of root nodule development and Nod factor induced processes. To study Nod factor induced plant responses in V.sativa on the molecular level, the homologues of the early nodulin genes of pea (Pisum sativum) were isolated from V. sativa and their expression pattern was studied by in situ hybridization during root nodule development (Chapter 2).To obtain information about the pathway from Nod factor to early nodulin gene expression, the mechanisms controlling ENOD12 expression were studied. With a promoter analysis in transgenic V.hirsuta root nodules we have identified that the 200 bp immediately upstream of the transcription start are sufficient to induce nodule specific and Nod factor induced expression (Chapter 3). For the isolation of transcription factors involved in controlling ENOD12 expression an expression library was screened and a preliminary characterization of cDNA clones encoding polypeptides that bind to the PsENOD12 promoter is described in Chapter 4.In Chapter 5 early nodulin gene expression during Nod factor induced morphological changes, like root hair deformation and nodule primordia induction, was studied. Furthermore we examined whether RNA and protein synthesis are required for root hair deformation and for the activation of the early nodulin genes. These studies provided new insights about the mode of action of Nod factors.In Chapter 6 is discussed to what extent V. sativa is a suitable host plant to study the mode of action of Nod factors and in which way the studies reported in this thesis have contributed to elucidate the mechanism by which Nod factors induce a diversity of plant responses.