Ecosystems comprise large groups of highly interdependent organisms. Cnidarians, such as sea anemones and corals, are keystone species in many marine ecosystems, especially coral reefs. Each individual cnidarian also constitutes an ecosystem unto itself, a "holo- biont", consisting of the host animal and accompanying microbial symbionts. To interro- gate cnidarian holobionts, I used computational approaches to analyze the transcriptomes of three cnidarians and build mechanistic models of their microbial symbionts. In par- ticular, I analyzed and annotated the transcriptomes of the cauliflower coral Pocillopora damicornis, the lined sea anemone Edwardsiella lineata, and the starlet sea anemone Ne- matostella vectensis, providing information about the molecular functions expressed by these organisms, and allowing development of a corresponding set of public databases: PocilloporaBase, EdBase, and an updated version of StellaBase, that facilitate access to the corresponding datasets. Additionally, I developed a method to infer the phylogenetic antiquity of transcripts. This method also allowed me to identify transcripts from other organisms (e.g., microbes) belonging to the anemone or coral holobiont. In parallel – in order better to understand the microbial symbionts that share envi- ronments with cnidarian hosts, I also developed new computer-simulation approaches for modeling metabolic interactions between different microbial species. These approaches are based on genome-scale stoichiometric reconstructions of metabolic networks and on Flux Balance Analysis (FBA). In addition to contributing to the development and testing of a new FBA-based platform for modeling communities in structured environments (Compu- tation Of Microbial Ecosystems in Time and Space, or COMETS), I used this platform for specific in silico experiments on microbial symbiosis. In particular, I computed all pairwise interactions between 582 different prokaryotic models, and identified global patterns of pu- tative positive (cross-feeding) vs. negative (food competition) interactions in this matrix of species pairs. I found that about 7% of the pairs yielded a greater biomass when grown together than when grown separately as monocultures. Despite existing challenges, such as the limitations of gap-filling steps in model construction and the need for a better knowl- edge of nutrient composition in natural environments, this approach could in the future help forecast shifts in the coral holobiont under likely scenarios of marine environmen- tal changes. In general, this work demonstrates how the integration of high-throughput sequencing technology and mechanistic systems-biology simulations, can provide unique tools to analyze interactions between microbes, and to mitigate or reverse adverse changes in marine ecosystems.

No description available.

Most vertebrates have separate sexes, and sex-specific traits that are regulated by genes with sex-biased expression patterns. In many species with genetic sex determination system, genetic recombination is suppressed in genomic regions linked to the master regulator of sex determination – the gene or set of linked loci that orchestrate sexual differentiation. Natural selection may favour alleles with sex-specific effects - including those with sexually antagonistic (SA) fitness effects (e.g., beneficial to females but harmful to males) – to become fixed in or be translocated to these non-recombining regions of sex chromosomes, because sex-specific or sex-biased modes of inheritance can resolve genomic conflict associated with SA. Sexually antagonism may also be resolve by sex-biased gene expression, and in theory these two mechanism (sex-linkage and sex-biased gene expression) could operate synergistically. However, there are relatively few empirical studies that test whether genes with sex biased expression patterns are indeed more abundant on sex chromosomes – and especially on newly evolved sex chromosomes. We explored this question with an African frog species Xenopus borealis, whose sex chromosome evolved within the last 25 million years (my) and have a large (~50Mbp) region of suppressed recombination, making it a young sex chromosome system compared to many other intensively studied systems, such as the sex chromosomes of mammals. We tested the possibility that a higher proportion of genes with sex-biased expression would be located on the sex-linked region of the sex chromosome of this species. By examining gene expression in adult liver and gonad and also tadpole gonad/mesenephros at two developmental stages, we found that the sex-linked region of these sex chromosome do have a higher proportion of sex biased genes compared to the non-sex-linked region of the same sex chromosomes, compared to (i) a homeologous genomic region in the tetraploid genome of X. borealis, and also (ii) the autosomes of this species. We did not observe the same pattern in a closely related frog species, Xenopus laevis, which has sex chromosome that are not homologous to those of X. borealis and, unlike X. borealis, lacks a large region of suppressed recombination on its sex chromosome. Using Brownian Motion model, we found as well that expression divergence evolution of genes in the sex-linked region of X. borealis is faster compared to its non-sex-linked homeologs (within X. borealis), and also compared to orthologous regions that are also non-sex-linked. One possible explanation for these observations is that natural selection favoured an expansion of recombination suppression (via unknown mechanisms) on chromosome such that polymorphic regulatory regions became linked (or unlinked) to the sex determining locus in such a way to resolve SA. Alternatively, it is possible that these sex-biased expression pattern evolved rapidly after recombination suppression. / Thesis / Master of Science (MSc) / Sexual selection favours the evolution of distinctive traits in each sex in order to optimize the reproductive success of each one. However, because most of the genome is shared between the sexes, sexual selection may result in genomic conflict when mutations are beneficial to one sex but harmful to the other; this conflict is known as sexual antagonism. Genomic conflict associated with alleles with sexually antagonistic (SA) fitness effects can be resolved via the origin of sex-biased expression patterns and this may be catalyzed by genetic linkage to a sex-determining locus on a sex chromosome. Consequently, one might predict there to be an enrichment of genes with sex-biased expression patterns on the sex chromosome as compared to the autosomes. We tested this expectation in an African frog species Xenopus borealis, which has a relatively young sex chromosomes and a large region of recombination suppression on the female-specific W-chromosome. We found enrichment of sex-biased genes on the nonrecombining region of the sex chromosomes of this species in adult liver and gonad tissue and also tadpole mesenephros/gonads at two developmental stages. Additionally, we found that expression divergence of genes in the non-recombining region have a faster rate of evolution as compared to the rate of expression divergence of genes in other genomic regions. One possible explanation for these observations is that natural selection favours an expansion of recombination suppression (via unknown mechanisms) on sex chromosome such that polymorphic regulatory region become linked (or unlinked) to the sex determining locus in such a way as to resolve SA.

Ticks are important vectors of a wide variety of pathogens including protozoa, bacteria and viruses. Many of the viruses transmitted by ticks are of medical or veterinary importance including tick-borne encephalitis virus (TBEV) and Crimean- Congo hemorrhagic fever virus causing disease in humans, and African swine fever virus and Nairobi sheep disease virus affecting livestock. Although several studies have elucidated tick antimicrobial mechanisms including cellular immune responses such as nodulation, encapsulation and phagocytosis and humoral immune responses such as the JAK/STAT pathway, complement-like proteins, antimicrobial peptides, lectin like pattern-recognition molecules and lysozymes, very little is known about the innate immune response of ticks towards viral infection. This study therefore aimed to identify molecules that might be involved in the response of ticks to viral infection. The hypothesis was that TBEV infection leads to changes in the expression of immunity-related transcripts and proteins in Ixodes spp. tick cells and that at least some of these might be antiviral. Ixodes scapularis-derived cell lines IDE8 and ISE6 were chosen since I. scapularis is currently the only tick species with a sequenced genome and an Ixodes ricinus-derived cell line, IRE/CTVM19, was used because I. ricinus is the natural vector of TBEV. Basic parameters required to study the responses of tick cells to infection were determined, including levels of virus infection, kinetics of virus replication and production, formation of replication complexes and uptake of dsRNA or siRNA. The cell lines IDE8, ISE6 and IRE/CTVM19 were infected with either of two tick-borne flaviviruses, TBEV and Langat virus (LGTV), or with the mosquito-borne alphavirus Semliki Forest virus (SFV). Infection was characterised using techniques including plaque assay, luciferase assay, immunostaining and conventional, confocal and electron microscopy. Two time points for transcriptomics and proteomics analysis of TBEVinfected IDE8 and IRE/CTVM19 cells were selected: day 2 post-infection (p.i.) when virus production was increasing and day 6 p.i. when virus production was decreasing. RNA and protein were isolated from TBEV-infected and mock-infected tick cells at days 2 and 6 p.i. and RNA-Seq and mass spectrometric technologies were used to identify changes in, respectively, transcript and protein abundance. Differential expression of transcripts was determined using the data analysis package DESeq resulting in a total of 43 statistically significantly differentially expressed transcripts in IDE8 cells and 83 in IRE/CTVM19 cells, while differential protein representation using Χ2 test statistics with Bonferroni correction in IDEG6 software resulted in 76 differentially represented proteins in IDE8 cells and 129 in IRE/CTVM19 cells. These included transcripts and proteins which could affect stages of the virus infection, including virus entry, replication, maturation and protein trafficking, and also innate immune responses such as phagocytosis, RNA interference (RNAi), the complement system, the ubiquitin-proteasome pathway, cell stress and the endoplasmic reticulum (ER) stress response. After verification of sequencing data by qRT-PCR, the ability of several of the identified transcripts or proteins to affect virus infection was determined by knockdown experiments in IDE8 and IRE/CTVM19 cells using wild type LGTV, LGTV replicons or TBEV replicons. Knockdown of genes encoding proteins including the ER chaperone gp96 and the heat-shock protein HSP90 resulted in increased virus production in both cell lines, hinting at an antiviral role. In contrast, knockdown of calreticulin, another ER chaperone, resulted in a decrease in virus production in IRE/CTVM19 cells but not in IDE8 cells, implying a requirement for virus production. This functional genomics approach has identified possible novel genes/proteins involved in the interaction between flaviviruses and tick cells and also revealed that there might be antiviral innate immune pathways present in ticks additional to the exogenous RNAi pathway.

L’objectif de cette thèse a été d’enrichir nos connaissances sur les bactériesintracellulaires strictes et spécialement Coxiella burnetii, agent responsable de la fièvre Q.Pour ce faire, nous avons d’une part amélioré les techniques de préparation de l’ARN pourles études transcriptionnelles et d’autre part utilisé la technologie des puces à ADN pouranalyser le transcriptome ainsi que la diversité génomique de C. burnetii.L’utilisation d’échantillons cliniques dans les études transcriptionnelles est limitéepar la quantité de matériel disponible qui ne permet pas d’analyser simultanément les profilsdu pathogène et de son hôte au cours de l’infection. De ce fait, nous avons développé, sur unmodèle d’escarres obtenus à partir de malades de fièvre boutonneuse méditerranéenne, unestratégie basée sur l’hybridation soustractive pour séparer les ARN eucaryotes etprocaryotes dans le but d’entreprendre des hybridations de puces à ADN.C. burnetii est une bactérie hautement résistante aux stress environnementaux commele changement de pH, la dessiccation, mais aussi le changement de température. Nous avonsparticulièrement étudié la réponse précoce de C. burnetii lors d’une exposition à une hauteet faible température. L’analyse globale du profil transcriptionnel a montré que la réponsede C. burnetii était limitée et similaire pour les différents stress appliqués. Cependant,malgré cette faible réponse, il apparait clairement qu’une accumulation de ppGpp, un arrêtde la croissance et des modifications de la membrane et de la paroi cellulaire permettraient àC. burnetii de résister à ces stress. Toutes ces régulations géniques convergent vers unchangement d’état de la bactérie vers une forme pseudo-sporulée. De plus, nous avonsobservé une organisation spatiale des gènes différentiellement exprimés. Nos analyses bioinformatiquesont montré que ces clusters de régulation ne répondaient ni au paradigmepromoteur - facteur de transcription – opéron, ni à des réseaux biologiques. Ayant retrouvéce phénomène dans plusieurs autres études transcriptionnelles chez d’autres bactériesintracellulaires, nous spéculons que ces clusters de régulations pourraient être dus à unerégulation épigénétique qu’il reste à caractériser.Différentes méthodes de typage ont déjà été mises au point pour classer les isolats deC. burnetii dans le but d'explorer son pouvoir pathogène. Ici, nous présentons une méthodede génomotypage basée sur la présence ou l'absence de gènes à l'aide de puces à ADN.Nous avons testé notre stratégie de génomotypage sur 52 isolats provenant de différenteszones géographiques, de différents hôtes et de patients présentant différentes manifestationscliniques. L'analyse a révélé la présence de 10 génomotypes organisés en 3 groupes avecune topologie congruente à celle observée avec le Multi Spacer Typing. Nous avons aussidécouvert 4 génomotypes particulièrement associés à la fièvre Q aiguë, alors que tous lesgénomotypes étaient associés à la forme chronique. De plus, le génomotypage a révélé queles isolats retrouvés dans les tiques dures, y compris la souche de référence Nine Mileappartiennent au même genomotype.Globalement, les données que nous avons obtenues confirment le fait que les puces àADN sont un outil adapté pour l’analyse de la pathogénicité de C .burnetti mais aussi desautres bactéries intracellulaires strictes. Cependant, de nouvelles technologies plusrésolutives comme le DNA ou RNAseq semblent être plus prometteuses mais restent encoreà optimiser / The objective of this thesis was to increase knowledge of obligate intracellular bacteriaand specifically, the causative agent of Q fever C. burnetii. In this regard, we have improvedstrategies to purify RNA for the transcriptional studies. We also used the technologymicroarrays to analyze the transcriptome and genomic diversity of C. burnetii.The use of clinical samples in the transcriptional studies is limited by the amount ofmaterial available and thus the transcriptional profiles of the pathogen and its host duringinfection can not be simultaneously analyze. We developed, with a model of eschars obtainedfrom Mediterranean spotted fever patients, a strategy based on subtractive hybridization toseparate RNA eukaryotic and prokaryotic cells in order to perform microarray experiments.Analysis of the survival strategies used by this bacterium to adapt to new environmentalconditions is critical for our understanding of C. burnetii pathogenicity. Here, we report theearly transcriptional response of C. burnetii under temperature stresses. Our data show thatC. burnetii exhibited minor changes in gene regulation under short exposure to heat or coldshock. While small differences were observed, C. burnetii seemed to respond similarly to coldand heat shock. The expression profiles obtained using microarrays produced in-house wereconfirmed by quantitative RT-PCR. Under temperature stresses, 190 genes were differentiallyexpressed in at least one condition, with a fold change of up to 4. Globally, the differentiallyexpressed genes in C. burnetii were associated with bacterial division, (p)ppGpp synthesis,wall and membrane biogenesis and, especially, lipopolysaccharide and peptidoglycansynthesis. These findings could be associated with growth arrest and witnessed transformationof the bacteria to a spore-like form. Unexpectedly, clusters of neighboring genes weredifferentially expressed. These clusters do not belong to operons or genetic networks; theyhave no evident associated functions and are not under the control of the same promoters. Wealso found undescribed but comparable clusters of regulation in previously reportedtranscriptomic analyses of intracellular bacteria, including Rickettsia sp. and Listeriamonocytogenes. The transcriptomic patterns of C. burnetii observed under temperature stressespermits the recognition of unpredicted clusters of regulation for which the trigger mechanismremains unidentified but which may be the result of a new mechanism of epigenetic regulation.Different typing methods have been previously developed to classify C. burnetii isolatesin order to explore its pathogenicity. Here, we report a comprehensive genomotyping methodbased on presence or absence of genes using microarray. The genomotyping method was thentested on 52 isolates obtained from different geographic areas, different hosts and isolated frompatient with different clinical manifestations. The analysis reveals the presence of10 genomotypes organized in 3 groups with a topology congruent with that of Multi SpacerTyping. We also found out 4 genomotypes especially associated with acute Q fever whereas allthe genomotypes could be associated to chronic human infection. Serendipity, genomotypingreveals that hard ticks isolates including Nine Mile belong to the same genomotype.Overall, the data we obtained confirm that DNA microarrays are a suitable tool forexploring pathogenicity of C. Burnetti and other obligate intracellular bacteria. However newtechnologies such as DNAseq or RNAseq seem more promising but still need to optimize andalso are still expensive compared to microarray.

Les gliomes de bas-grade représentent la tumeur cérébrale la plus fréquente chez l’enfant. Elles sont caractérisées par un large spectre de sous-types tumoraux, très hétérogènes. Leur définition actuelle est principalement basée sur des critères histologiques ce qui représente une limite importante car ces classifications souffrent d’un manque de précision. Les progrès récents de la génomique nous permettent d’approfondir considérablement les connaissances sur la biologie de ces tumeurs afin d’enrichir leur classification actuelle. Ce travail présente une analyse approfondie des altérations génomiques de l’ADN et l’ARN des gliomes de bas-grade pédiatriques. Le premier niveau d’analyse se base sur l’analyse du séquençage à haut débit de 169 gliomes de bas-grade de l’enfant. Bien que les mutations des gènes BRAF et FGFR1 sont les plus fréquemment décrites dans ces tumeurs, nous avons identifié pour la première fois le réarrangement chromosomique MYB-QKI majoritairement associé aux gliomes angiocentriques. Dans un deuxième temps ce travail décrit l’analyse du transcriptome de 151 gliomes de bas grade extraits à partir de tissu conservé en paraffine. Nous avons observé des différences moléculaires en fonction de leur sous-type histologique, de la localisation tumorale et de leur statut BRAF. Dans le dernier volet de ce travail, nous avons testé la faisabilité d’isoler par cytométrie en flux une cellule unique en les distinguant selon un marqueur de différenciation glial (A2B5+ et A2B5-) et d’effectuer une analyse transcriptomique à haut-débit en séquençant l’ARN à l’échelle d’une cellule unique. Cette technique nous a permis de décrire des différences moléculaires intéressantes entre des cellules A2B5+ et A2B5-. Ces résultats soulignent l’intérêt d’exploiter des nouvelles technologies de pointe pour servir de base à l’étude des caractéristiques biologiques des cellules tumorales. / Low-grade gliomas represent the most frequent brain tumor arising during childhood. They are characterized by a broad spectrum of tumor types.The definition of low-grade gliomas has been mainly based on morphology. This histological classification of pediatric low-grade gliomas (PLGG), suffers from the lack of reproducibility. The recent progress in molecular biology and genetics has brought new insights in the biology of those tumors and allows better understanding of their biology. This work provides a comprehensive analysis of two different genetic approaches in PLGGs. The first part is based on the description of somatic genetic alterations of the DNA. Using a large PLGG cohort, we have dissect the genome of those tumors and draw the landscape of their genetic alteration. Although BRAF and FGFR1 alterations are predominantly altered, we have discovered a new translocation, MYB-QKI, that is almost exclusively present in a specific histological subgroup; angiocentric gliomasThe second part of the thesis describes transcriptomic analysis of bulk PLGGs. This work describes molecular differences between PLGGs from distinct histologies and arising from different locations in the brain as well as different BRAF mutation status.We were also able to test single-cell expression analyses in three pilocytic astrocytomas (PAs) using RNA-sequencing. In this experimental work we have successfully tested the hypothesis that we can isolate single-cells from fresh PLGG tumors in order to analyze the trasncriptome at a large scale. We observed that single-cells expressing A2B5, a glial progenitor marker, isolated in pediatric PAs are characterized as a distinct biological population. These results underline the importance to improve the precision of the transcriptomic studies to capture the molecular signal of tumor cells and further understand the different pattern between normal cells and tumor cells.

La famille des facteurs de transcription lkaros, est composée des protéines lkaros, Helios, Aiolos et Eos. Elles sont exprimées pendant le développement et régulent la différenciation des lymphocytes B et T. Ces protéines présentent une forte homologie de leurs séquences nucléiques et protéiques et sont toutes impliquées dans l'apparition de leucémies lymphoblastiques T ou B. Ces facteurs présentent toutefois de très fortes divergences dans leurs profils d'expression, leurs fonctions et leurs gènes cibles. Ce travail à partir d'une lignée cellulaire T immature déficiente pour lkaros à permis d'étudier les différences fonctionnelles et moléculaires des membres de la famille. La réexpression d'lkaros, d'Aiolos et d'Helios permet la différenciation et la diminution de la prolifération de ces cellules. Cette expérience a montré que les différents membres de la famille avaient des capacités distinctes pour activer ou réprimer certains gènes cibles. Un échange des séquences des domaines de liaison à l'ADN de Aiolos et d'lkaros, a permis de montrer que la spécificité fonctionnelle est en partie déterminée par le domaine de liaison à l'ADN, mais aussi par les autres régions d'lkaros et d'Aiolos. / The lkaros transcription factor family is made of the proteins lkaros, Helios, Aiolos and Eos. They are expressed during the development and regulate the differentiation of lymphocytes B and T. These proteins present a strong homology between their nucleic and protein sequences and are involved the appearance of T or B lymphoblastic leukaemia. However these factors present strong differences in their profiles of expression, their functions and their target genes. An immature T cell line, deficient for lkaros, allows us to study the functional and molecular differences of members of the family. There-expression of lkaros, Aiolos and Helios allows the differentiation and the decrease of the proliferation of these cells. I also showed that the various members of the family had different capacities to activate or repress certain target genes. An exchange of the protein sequences coding for the DNA binding domain (DBD), shows that the functional specificity is partially determined by the DBD domain, but also by the other regions of lkaros and Aiolos.

La Spondyloarthrite (SpA) est un rhumatisme inflammatoire chronique fréquent, avec une prévalence de 0,43 % en France. Elle consiste en une atteinte prédominante du squelette axial, mais aussi des articulations périphériques, et peut conduire à une immobilité du rachis et des articulations sacro-iliaques. Des atteintes extra-articulaires sont fréquentes, telles qu'une uvéite, un psoriasis ou une maladie inflammatoire chronique de l'intestin. Les traitements actuels ne sont que symptomatiques, ciblant principalement les manifestations inflammatoires. L'étiologie de la SpA est multifactorielle avec une composante génétique dominée par l'association forte et bien connue avec l'allèle HLA-B27. Cependant, ce facteur génétique n'est clairement pas suffisant pour induire le développement de la maladie. L'objectif de ce projet de thèse était donc d'identifier d'autres facteurs génétiques à l'origine du développement de la SpA.Mon travail a porté sur l'analyse de deux jeux de données complémentaires, dans une perspective de biologie des systèmes. Dans une première partie, j'ai conduit une analyse de liaison dans 210 familles atteintes de la maladie représentant 1310 personnes génotypées avec des puces Affymetrix 250k. Une nouvelle région significativement liée à la SpA a été détectée en 13q13, avec un intervalle de 1,3 Mb défini par des haplotypes recombinants chez les patients.Ensuite, une analyse transcriptomique des cellules dendritiques dérivées des monocytes de 23 patients HLA-B27+, 23 témoins sains HLA-B27+ et 21 témoins sains HLA-B27-, et stimulées ou non par du LPS, a tenté de distinguer les gènes dont l'expression est modifiée par la maladie de ceux influencés par l'allèle HLA-B27 seul. L'annotation fonctionnelle et une analyse par réseau de gènes ont mis en évidence l'inhibition chez les patients des étapes précoces de la biosynthèse du cholestérol. / Spondyloarthritis is a frequent chronic inflammatory rheumatism, with a prevalence of 0.43 % in France. This disease presents axial skeleton injuries, but also on peripheral joints, and can results in a total spinal and sacro-iliac motility loss. Extra-articular features including uveitis, psoriasis and inflammatory bowel disease are frequent. Current SpA treatments are only symptomatic, relieving inflammatory symptoms. SpA etiology is largely multifactorial with a genetic component dominated by the long-known strong association with the HLA-B27 allele. This allele, however, is not sufficient for the disease to occur. This thesis project objective was then to identify other genetic factors in the origin of SpA.My work was mainly divided in two complementary data analyses, in a way to get a systems biology approach. The first one consisted in proceed linking analyses on data from Affymetrix genotyping chips gathered from DNA of 1310 people grouped in 210 families. This study allowed notably to detect a new significantly linked region to SpA : 13q13, with an interval of 1.3 Mb. This part of genome is currently being sequenced to allow a better causal SNP identification.Secondly, an Affymetrix HumanGene 1.0 st transcriptomic chips analysis was performed on MD-DCs extracted from 68 people, stimulated or not by LPS during 6 or 24 hours. This cohort was grouped between 23 patients HLA-B27+, 23 healthy controls HLA-B27+ and 21 healthy controls HLA-B27-. I could notice that HLA-B27 allele is farly enough to considerably affect cell transcriptomic profiles, which encourages to include HLA-B27+ healthy controls. Otherwise, a gene network analysis allowed me to highlight on an inhibition of early steps of cholesterol biosyntthesis.

La Fusariose de l’épi (FHB) est une maladie importante des céréales et en particulier du blé tendre. Elle est causée par deux genres fongiques, le genre Fusarium et le genre Microdochium. L’espèce Fusarium graminearum est l’agent principal de cette maladie. Elle affecte non seulement les rendements et la qualité des grains chez le blé, mais elle cause un sérieux problème sanitaire via la production des mycotoxines. L’établissement de cette maladie requiert l’expression de gènes végétaux (facteurs de sensibilité) qui restent encore méconnus. Afin d’étudier les événements moléculaires qui participent à la mise en place de cette maladie sur un grain de blé tendre en développement et sensible au FHB, une cinétique d’infection de cinq points correspondant aux principaux stades développementaux du grain a été réalisée. Ensuite, deux approches, la protéomique comparée et la transcriptomique comparée à l’aide de puces à ADN, ont été utilisées pour répondre à ces questions. L’analyse protéomique a permis d’identifier 73 protéines différentiellement régulées appartenant à 5 grands groupes fonctionnels alors que l’approche transcriptomique a mis en évidence 1309 gènes répartis dans 16 groupes fonctionnels différents. Ces deux approches ont montré que l’infection ne bloque pas le développement du grain, mais elle induit des changements importants dans le métabolisme primaire notamment sur la synthèse de l’amidon et des protéines de réserve. Elle a également montré que la réponse du grain au FHB est liée au stade développemental du grain. Cette étude apporte de nouveaux éléments nécessaires à la compréhension de la sensibilité du blé tendre à la Fusariose de l’épi. Cette sensibilité du grain de blé se caractérise principalement par l’induction des mécanismes de détoxification des mycotoxines, la mise en place des mécanismes du détournement du métabolisme carboné de l’hôte par l’agent pathogène et le contrôle de la mort cellulaire programmée des cellules végétales. Enfin, l’étude a permis d’établir une liste d’au moins 100 gènes candidats potentiellement impliqués dans la sensibilité du blé au FHB. / Fusarium head blight (FHB) is an important disease of cereals, particularly of wheat. It is caused by two fungal genera, the genus Fusarium and genus Microdochium. The species Fusarium graminearum is the principal agent of this disease. This disease affects not only yield and grain quality in wheat, but it causes serious health problem through the production of mycotoxins. The establishment of this disease requires the expression of plant genes (susceptibility factors) that are still unknown. To study the molecular events involved in the development of this disease in the susceptible wheat grain during its development, time course infection of five points corresponding to the main developmental stages of grain was performed. Then two approaches, proteomics and transcriptomics using DNA microarrays were used to answer to this question. Proteomic analysis identified 73 differentially regulated proteins belonging to five major functional groups while the transcriptomics data revealed 1309 genes involved in 16 distinct functional groups. Both approaches have shown that infection does not interrupt grain development, but induces significant changes in primary metabolism, mainly on the synthesis of starch and storage proteins. It also showed a link between FHB response and the grain development. This study provides new evidence necessary to understand the susceptibility response of wheat to FHB. The wheat grain susceptibility is mainly characterized by the induction of detoxifying mechanisms of mycotoxins, the diversion ofcarbon metabolism of the host by the pathogen and control of Programmed Cell Death (PCD) of plant cells. Finally, the study allowed the establishment of a list of at least 100 candidate genes potentially involved in the wheat susceptibility to FHB.

Les plantes sont capables de percevoir dans leur environnement la disponibilité en nitrate (NO3-), un macro-nutriment essentiel. Chez Arabidopsis thaliana, le transporteur de NO3- NRT1.1 constitue un système de perception qui active de nombreuses réponses au NO3-, notamment la régulation de l'expression de gènes et le développement des racines latérales. Dans ce dernier cas, un mécanisme de transduction du signal a été proposé. Celui-ci met en jeu une activité de transport d'auxine par NRT1.1 qui est inhibée par le NO3-. Cependant, le(s) mécanisme(s) moléculaire(s) permettant à NRT1.1 de contrôler un large panel de réponses au NO3- reste(nt) largement inconnu(s). L'objectif de ce travail était donc d'approfondir nos connaissances sur les voies de signalisation du NO3- dépendantes de NRT1.1. Grâce à l'analyse de mutants et de lignées transgéniques exprimant des versions de NRT1.1 présentant des mutations ponctuelles, nous avons pu découpler certaines des réponses NRT1.1-dépendantes et montré que cette protéine peut percevoir/transduire le signal NO3- au travers d'au moins trois mécanismes distincts, possédant des bases structurales différentes au sein de la protéine. D'autre part, ce travail a permis de valider l'hypothèse selon laquelle NRT1.1, en intervenant comme transporteur d'auxine, contrôle directement le développement des racines latérales, et ce indépendamment des autres transporteurs d'auxine qui y sont exprimés. Enfin, nous avons montré qu'en plus de sa régulation transcriptionnelle déjà connue, NRT1.1 est soumis à une puissante et complexe régulation post-transcriptionnelle. En effet, le transcrit NRT1.1 est stabilisé en présence de NO3- dans la racine alors que l'accumulation de la protéine NRT1.1 est réprimée par le NO3- spécifiquement au niveau des primordia de racines latérales. Les résultats obtenus au cours de ce travail ont permis d'élaborer un modèle cohérent du rôle de signalisation joué par NRT1.1, et ouvrent de nombreuses perspectives pour comprendre comment, chez les plantes, un « transcepteur » (transporteur/senseur) peut contrôler une vaste gamme de réponses adaptatives aux facteurs de l'environnement. / Plants are able to sense the external availability of nitrate (NO3-), a major macro-nutrient. In Arabidopsis thaliana, the NO3- transporter NRT1.1 acts as a sensor that triggers many different adaptive responses, including the regulation of gene expression and lateral root development. In the latter case, a transduction mechanism that involves a NO3--inhibited auxin transport activity dependent of NRT1 has been proposed. However, the molecular mechanism(s) allowing NRT1.1 to control such a large palette of NO3- responses is still largely unknown. Thus the aim of this work was to better understand and characterize the NRT1.1-dependent NO3- signaling pathway(s). Using mutants and transgenic lines expressing point mutated forms of NRT1.1, we uncoupled several of the NRT1.1-dependent responses and thus demonstrated that NR1.1 can sense/transduce NO3- signal through at least three distinct mechanisms at the protein level. This work also largely confirmed the hypothesis that NRT1.1 directly controls lateral root development through its auxin transport activity regardless of the other auxin transporters expressed in lateral root primordia. Finally, we showed that, besides the already well characterized transcriptional NO3--dependent regulation of NRT1.1, this gene is also subjected to complex post-transcriptional regulations. Indeed, on the one hand, NRT1.1 mRNA is stabilized by NO3- in roots whereas, on the other hand, protein accumulation is specifically repressed by NO3- in lateral root primordia. Altogether, these results allowed us to build a comprehensive model of the complex NRT1.1 signaling and open many perspectives to understand how plant “transceptors” (transporter/sensor) can monitor a large variety of adaptive responses to environmental factors.