Les plantes de la famille des légumineuses ont acquis la capacité d’accueillir au sein d’organes symbiotiques, les nodosités, des bactéries fixatrices d’azote appelées bactéroïdes. Cette symbiose permet aux plantes hôtes de satisfaire leurs besoins en azote. Certaines légumineuses produisent au sein des nodosités une grande famille de peptides antimicrobiens particuliers, les NCR (Nodule-specific Cysteine-Rich). Ils permettent à l’hôte de contrôler la population bactérienne intracellulaire via leurs activités antimicrobiennes, mais aussi d’imposer aux rhizobia une différenciation terminale des bactéroïdes. Durant ma thèse, j’ai participé à l’identification et à la caractérisation de la protéine bactérienne BclA. Ce transporteur ABC est nécessaire pour la formation de bactéroïdes différenciés lors de la symbiose Aeschynomene-Bradyrhizobium. Pour mieux comprendre sa fonction symbiotique, j’ai étudié la relation entre BclA et une enzyme de modification du peptidoglycane, la DD-carboxypeptidase 1. J’ai pu montrer que ces deux facteurs agissent de manière indépendante dans la mise en place de bactéroïdes différenciés. Une analyse fonctionnelle de BclA et une expérience d’évolution expérimentale avec le mutant bacA de Sinorhizobium, un orthologue de bclA, apportent une meilleure compréhension du rôle de ces transporteurs in vivo. L’ensemble des résultats obtenus pendant ma thèse suggère que BclA et BacA assurent la résistance bactérienne face aux NCR in planta, comme un prérequis pour la suite du processus, mais ne sont pas nécessaires à la différenciation per se. De plus, leurs activités d’import des NCR ne semblent pas être le mécanisme sous-jacent du processus de résistance. / Plants of the legume family have acquired the ability to host in specific symbiotic organs, the roots nodules, nitrogen fixing bacteria, called bacteroids. This symbiosis allows plants to fulfill all their nitrogen requirements. Some legume plants produce in their nodules a large family of antimicrobials peptides called the NCRs (Nodule-specific Cysteine-Rich). Their antimicrobial activities allow the host plant to control the intracellular bacterial population. NCRs peptides also govern terminal differentiation of the bacteroids. During my PhD work, I participated in the identification and characterization of BclA. This bacterial ABC transporter is involved in bacteroid differentiation during the Aeschynomene- Bradyrhizobium symbiosis. In order to better understand its symbiotic function, I studied the link between BclA and a peptidoglycan-modifying enzyme, the DD-carboxypeptidase 1. I was able to show that these two factors act in an independent manner in the establishment of bacteroid differentiation. A functional analysis of BclA and an experimental evolution on Sinorhizobium bacA mutant, an orthologue of bclA, conferred a better understanding of the in vivo role of these transporters. The results obtained during my thesis suggest that the BclA and BacA function is to ensure bacterial resistance to NCRs, as a prerequisite for the bacterial differentiation process, but is not needed for differentiation per se. Furthermore, their NCR uptake activities do not seem to be the mechanism underlying the resistance.

Phytonematodes are known to cause substantial losses in crop yields across the world. Since the middle of the last century, these pests have been adequately controlled by chemical nematicides. However, due to increasing public health concern, strict regulations in the EU and elsewhere have significantly reduced the usage of these environmentally not-so-safe chemicals. This has led us to look for reliable biological alternatives. The Pasteuria group of Gram-positive endospore-forming bacteria (phylum: Firmicutes) often associated with nematode-suppressive soils are potentially reliable nematode biocontrol agents. However, the highly specific interaction of Pasteuria to their nematode hosts poses a challenge to the management of heterogeneous populations of nematodes in the field; the mechanism behind this specificity remains unclear. One of the fundamental basis of host specificity is the attachment of Pasteuria endospores to the cuticle of their host nematodes which is the first and essential step in the infection process. Thus, understanding the molecular mechanisms that govern the attachment process is important in identifying suitable populations of Pasteuria for effective broad-range management of plant parasitic nematodes in soil. Previous studies suggest the presence of immunogenic collagen-like fibres and carbohydrates on the endospore coat of Pasteuria that may have a role in the initial interaction of the endospores with their nematode hosts. Published work on phylogeny relates Pasteuria to Bacillus spp. most of which have well annotated and characterized genomes while the genome of Pasteuria remains to be sequenced completely. In this thesis, I attempt to explore the endospore biology of obligate and fastidious Pasteuria spp. using the wide knowledgebase of well studied Bacillus endospores. The primary aim was to characterize the immunogeneic determinants that are possibly responsible for the attachment of Pasteuria endospores to the host nematode cuticle by a combination of computational and lab-based approaches. To approve the suggested phylogenetic closeness of Pasteuria to Bacillus, the first part of the study focused on phylogeny reconstruction of Pasteuria spp. amongst Bacillus spp. and other members of the phylum Firmicutes. This was followed by in silico studies to identify candidate collagen-like genes in P. penetrans; the putative functional proteins encoded by these candidate genes were then comparatively characterized with collagens from other organisms including the members of the genus Bacillus. The surface associated collagen-like proteins and other possible immunogens on the endospores of Pasteuria were characterized by protein immunoblotting, lectin blotting and immunofluorescence microscopy and comparisons were made with B. thuringiensis endospores. Lastly, endospore attachment assays were done to test the hypothesis that collagens and carbohydrates play a role in Pasteuria endospore attachment. The results of the computational analyses suggest a family of collagen coding putative genes in the Pasteuria genome, all of which are predicted to have varied biochemical properties and are seemingly of diverse evolutionary origin. The Western blot and microscopic analyses show that the endospores of P. penetrans and B. thuringiensis share some common immunodominant surface epitopes. The attachment assays confirm the involvement of collagens and at least one carbohydrate (N-acetylglucosamine) in the endospore attachment. However, the results also indicate possible involvement of other adhesins in the process; to support this, at the end of the thesis, I propose a new 'Multitype Adhesin Model' for initial interaction of Pasteuria endospores with the cuticle of their host nematodes. The outcomes of this project will help in identifying the molecular basis of the complex Pasteuria-nematode interaction. This will provide a basis to develop environmentally benign nematode bio-management strategies.