Evolution of developmental genes is an important mechanism for plant morphological evolution. The LRL genes are an ancient group of bHLH transcription factors that positively regulate root hair development in angiosperms. Here I show that, in the moss Physcomitrella patens, two LRL genes are present and they positively regulate rhizoid and caulonema (a rhizoid-like cell type) development. GUS-transcriptional reporter plants show that both PpLRL1 and PpLRL2 are expressed in tissues giving rise to caulonemata. Loss-of-function mutants in either PpLRL1 or PpLRL2 led to defective rhizoid and caulonema development, and the Pplrl1 Pplrl2 double loss-of-function mutants completely lack rhizoids and caulonemata. Consistent with this, gain-of-function mutants show enhanced rhizoid and caulonema development. In addition, I show that the stimulatory effects of auxin and low phosphate on the development of rhizoids and/or caulonemata required PpLRL gene function. Together, these results show that LRL genes are conserved, positive regulators in tip-growing rooting cell development in land plants. To elucidate whether LRL genes belong to part of a conserved gene network, I use qRT-PCR to determine the transcriptional interaction between LRL genes and the Class I RSL genes, which is another group of conserved regulators of rhizoids and root hairs. Comparing the LRL-RSL network between P. patens and A. thaliana reveals that LRL and Class I RSL genes are transcriptionally independent of each other in P. patens but one LRL gene is transcriptionally downstream of Class I RSL genes in A. thaliana. This suggests that the gene network controlling tip-growing rooting cell development has changed since mosses and angiosperms last shared a common ancestor.

Plants colonised land 460 million years ago and charophytes represent the closest living relatives of land plants. The ability to live on land may depend on the presence of certain cell wall polysaccharides such as xyloglucan, a hemicellulose exclusively found in land plants (Popper and Fry, 2003). The cell walls of charophytes are poorly characterised. The aim of this project was to use biochemical techniques to characterise the cell wall polysaccharides of charophytic algae in relation to early land plant phylogeny. Hydrolysis of Coleochaete scutata and Chara vulgaris cell walls in 2 M trifluoroacetic acid yielded predominantly GalA, Gal, Glc and Man residues and also some Ara, Xyl and traces of Fuc and Rha. In addition, hydrolysis of Chara pectin revealed an abundance of an unusual monosaccharide, 3-O-methyl-D-galactose, which was structurally identified by a series of 1-D and 2D NMR spectroscopy by COSY, TOCSY, NOESY and HSQC. 3-O-Methyl-D-galactose is more commonly found in lycophyte cell walls where its presence has been suggested to be related to lycophytes’ evolutionarily isolated position (Popper et al., 2001). The newly discovered presence of 3-O-methyl-D-galactose in charophyte pectin suggests that this polymer may be more complex than previously thought. Coleochaete and Chara hemicellulose extracts were fractionated by anion-exchange chromatography into five classes. A strongly anionic fraction from Chara hemicellulose was found to be rich in Glc, Xyl, Gal and Fuc suggestive of a xyloglucan-like polysaccharide. However, XEG was unable to produce diagnostic xyloglucan oligosaccharides in either Coleochaete or Chara hemicelluloses. Xylanase and mannanase digestion of Coleochaete and Chara hemicelluloses gave xylan- and mannan-oligosaccharides. Furthermore, lichenase digestion of Coleochaete hemicellulose yielded an unusual octasaccharide composed of approximately equimolar xylose and glucose. My work has shown that charophyte cell walls are a source of undiscovered monosaccharides and potentially novel pectic and hemicellulosic domains which may have important functions in enabling the successful colonisation of land by plants.

Les espèces forestières sont particulières à bien des égards par rapport aux autres espècesdomestiquées. Les arbres forestiers ont de longues phases juvéniles, entrainant de long et couteuxcycles de sélection et nécessitant une sélection en plusieurs étapes indépendantes. Bien que cetteméthode soit efficace du point de vue opérationnel, elle reste couteuse en temps et en ressources,entrainant une dilution de l’intensité et de la précision de sélection. Au vu de ces contraintes,les arbres sont de bons candidats pour la mise en oeuvre de l’évaluation génomique. La sélectiongénomique (SG) repose sur le classement et la sélection d’individus à partir de l’informationcontenu dans leur génome sans utilisé une étape d’évaluation phénotypique et ainsi accélérerle processus de sélection.Ce travail visait à identifier les situations, les critères et les facteursdans lesquelles la SG pourrait être une option réalisable pour le peuplier. Notre étude a montréque les avantages de l’évaluation génomique dépendent du contexte. C’est dans des situationsles moins avantageuse que l’évaluation génomique se montre la plus performante, elle profiteégalement de la densification de l’information génétique de faible à moyenne suite à une étaped’imputation de haute qualité. La sélection génomique pourrait être une option intéressante àstade précoce, où la précision de la sélection est généralement faible et la variabilité génétiqueabondante. Notre travail a également montré qu’il est important d’évaluer les performancesavec des critères alternatifs, comme ceux liés au classement, notamment lorsque ces critèresrépondent au contexte opérationnel du programme d’élevage étudié. / Forest species are unique in many ways compared to other domesticated species. Forest trees have long juvenile phases, leading to long and costly selection cycles and requiring selection in several independent stages. Even if this method is operationally effective, it remains costly in terms of time and resources, resulting in a diluted intensity and accuracy of selection.In view of these constraints, trees are good candidates for the implementation of genomic evaluation. Genomic selection (SG) is based on the classification and selection of individuals from the information contained in their genome without using a phenotypic evaluation step and thus accelerating the selection process, in order to identify the situations, criteria and factors in which SG could be a feasible option for poplar. Our study showed that the benefits of genomic evaluation are context-dependent. Genomic evaluation is most effective in theless-advantageous situations, it also benefits from low to medium density genetic information following a high-quality imputation step. Genomic selection could be an interesting option at an early stage, when the accuracy of selection is generally low and genetic variability is abundant.Our work has also shown that it is important to evaluate performance with alternative criteria,such as those related to ranking, especially when these criteria fit the operational context of the breeding programme under study.

Buzz-pollination or pollination by vibration occurs in several families of angiosperms including some important commercial crops such as potatoes and tomatoes. Buzz-pollinated flowers release pollen via small pores or slits on the anther’s tip that require the use of vibrations by specialized pollinators, usually bees, to remove the pollen. Some buzz-pollinated species have elaborate floral morphologies including dimorphic anthers within the same flower (heteranthery), and mirror-image flowers (enantiostyly) where the style is reciprocally deflected to either the left or right side of the floral axis. The complex floral morphology and buzz-pollination syndrome seen in these species require a close physical interaction between the sexual organs of the flowers and the bodies of insect visitors. Despite the broad taxonomic distribution of buzz-pollination in angiosperms (more than 60 families are buzz-pollinated) relatively few studies have described the pollination ecology of these species under natural conditions. The main goal of the present work was to characterize the pollination biology, mating patterns and antagonistic interactions (e.g. pollen theft) in natural populations of a buzz-pollinated species. For this purpose, I studied Solanum rostratum (Solanaceae), a buzz-pollinated, self-compatible, annual weed with complex floral morphology (both enantiostylous and heterantherous flowers). This species usually grows in disturbed areas in its native range (Mexico) and has become invasive around the world. My research was divided into three components. First, I characterized the pollination and reproductive biology of natural populations in Mexico. I performed floral manipulations in six natural populations of S. rostratum to estimate fruit and seed set. In three of these populations, I carried out 115 hours of pollinator observations and quantified the incidence of pollinators versus pollen thieves. I also measured the efficiency of a subset of floral visitors in triggering fruit set after single visits. Second, I investigated whether morphological correspondence between the size of the pollinator’s body and floral morphology influences pollen transfer. In experimental arrays, I exposed flowers of S. rostratum that varied in the distance between their sexual organs, to bumblebees (Bombus terrestris) of different sizes, and recorded pollen deposition and fruit and seed production. Finally, I determined the mating system (i.e. the proportion of self- and cross-fertilized offspring) of natural populations in Mexico and of introduced populations in the United States of America, using newly developed microsatellite markers. My results show that S. rostratum is visited by a wide range of bees of different sizes (0.9–9.8 mm in thorax diameter), but that only a small subset of these visitors act as pollinators. Most visitors act as pollen thieves, consuming pollen while effecting little or no pollination. I also found that correspondence between a pollinator’s size and the separation of the S. rostratum sexual organs determines pollen deposition and fruit production; pollen deposition decreased when bees were small relative to the distance between the sexual organs visited the flowers. My genetic analyses show that natural populations of S. rostratum maintain a relatively high outcrossing rate (tm = 0.75 ± 0.03) across the native and introduced range. Furthermore, genetic diversity is reduced in invasive populations, but this is not accompanied by changes in mating system. My work shows that the morphological fit between the pollinator and the flowers is important in determining the dynamics of pollen transfer and fruit production in this buzz-pollinated plant. Distinguishing between pollinators and pollen thieves in buzz-pollinated plants is essential for understanding the evolution of buzz-pollination, as pollen theft could be a major selective force for these species.

Les variétés traditionnelles de vigne nécessitent de très nombreux traitements phytosanitaires pour lutter contre les maladies cryptogamiques qui touchent leurs parties herbacées, comme le mildiou et l’oïdium, causés par Plasmopara viticola et Erysiphe necator respectivement. Ces traitements, coûteux et préjudiciables pour l'environnement, pourraient être réduits par l’emploi de variétés résistantes. La vigne cultivée européenne (Vitis vinifera, 2n=38) est très sensible au mildiou et à l'oïdium. En conséquence, la résistance doit être introduite à partir d’autres Vitaceae ayant un niveau de résistance plus élevé à ces maladies. Plusieurs origines de résistance ont déjà été observées et inventoriées, en particulier chez l’espèce d’origine américaine Muscadinia rotundifolia (2n=40). Ces facteurs sont d'un intérêt majeur pour la sélection de variétés résistantes. Cependant, lors du processus d'introgression, des difficultés à obtenir des pépins viables en F1 ainsi que des anomalies phénotypiques dans les descendances en rétrocroisement ont été constatées. Afin d’optimiser la gestion des résistances provenant de cette espèce dans les programmes d’amélioration variétale, il est nécessaire de comprendre l’organisation génétique et génomique de M. rotundifolia, et de compléter la connaissance des facteurs de résistance issus de cette espèce. Dans ce contexte, les objectifs de la thèse sont : (i) de réaliser une analyse comparative des génomes de V. vinifera et M. rotundifolia et (ii) d’identifier de nouveaux facteurs de résistance chez M. rotundifolia utilisables à terme en sélection. Pour cela, une carte génétique de M. rotundifolia a été développée à partir d’une population de 200 individus issue de l’autofécondation de M. rotundifolia cv. Regale. Parallèlement, la même population a été testée pour son niveau de résistance au mildiou et à l’oïdium. Une carte génétique couvrant 950 cM a été réalisée. Elle comprend 191 marqueurs microsatellites répartis sur les 20 chromosomes de M. rotundifolia, et permet de conclure à un niveau de macrosynténie très élevé avec V. vinifera. Le groupe de liaison 20 de M. rotundifolia correspondrait à la partie inférieure du groupe de liaison 7 de V. vinifera. Par ailleurs, un QTL de résistance au mildiou a été détecté sur le groupe de liaison 18 de M. rotundifolia, au niveau d’une région riche en gènes de type NBS-LRR, et un nouveau QTL de résistance majeur à la l’oïdium a été mis en évidence sur le groupe de liaison 14 de M. rotundifolia. Ce QTL, nommé Ren5 pour ‘Resistance to Erysiphe necator 5’, montre une action précoce dans l’arrêt de la croissance du mycélium du pathogène, dès l’établissement des premiers stades de biotrophie du champignon. De plus, le QTL Ren5 a été confronté à deux souches supplémentaires d’E. necator, appartenant aux deux groupes d’oïdium retrouvés dans vignobles européens, contre lesquelles il reste efficace. Les données de cartographie génétique générées pour M. rotundifolia dans ce travail, ainsi que la mise en évidence de Ren5 et de son mode d’action, permettront d’améliorer la gestion des facteurs de résistance issus de cette espèce pour la sélection de variétés résistantes au mildiou et à l’oïdium. / Grapevine requires numerous harmful plant-health treatments, in particular to control downy and powdery mildews, caused by Plasmopara viticola and Erysiphe necator, respectively. One way to reduce the use of fungicides in viticulture is to create resistant grapevine cultivars. European cultivated grapevine (Vitis vinifera, 2n=38) is susceptible to mildews, whereas related species such as Muscadinia rotundifolia (2n=40) exhibit high levels of resistance. In breeding programs, resistance factors from these related species are introduced into the susceptible elite varieties. Nevertheless, difficulties in obtaining viable seeds in the siblings of Vitis x Muscadinia crosses are observed. In order to optimize the management of resistance factors from M. rotundifolia in breeding programs, it is necessary to understand the genomic organisation of this species, and to complete the knowledge of these factors. Thus, the main objectives of this work are : (i) making a comparative analysis of V. vinifera and M. rotundifolia genomes and (ii) identifying new resistance factors from M. rotundifolia. For this purpose a framework genetic map of M. rotundifolia cv. Regale has been created, using a 200 individual S1 population. This population has also been screened for its resistance to downy and powdery mildew. A 950 cM genetic map has been generated, including 191 SSR markers distributed across the 20 chromosomes of M. rotundifolia. A high level of macrosynteny has been observed between the Muscadinia map and the genetic maps available for V. vinifera. Linkage group (LG) 20 of M. rotundifolia matches with the lower part of V. vinifera LG7. Furthermore, a QTL for resistance to downy mildew has been identified on M. rotundifolia LG18, and a major QTL for resistance to powdery mildew has been mapped on LG14. The latest, called Ren5 for ‘Resistance to Erysiphe necator 5’, acts during an early stage to stop E. necator’s mycelium growth, since the first stages of biotrophy have been established for the fungus. Moreover, Ren5 has been confronted to two additional powdery mildew strains, belonging to the two European groups of E. necator, and it remained efficient. Gathering knowledge about the genetic organization of M. rotundifolia, and the mechanism and spectrum of action of newly identified resistance factors such as Ren5, will be useful to optimize the management of M. rotundifolia resistance traits in breeding programs aiming to create new resistant varieties to downy and powdery mildews