Directed evolution is a powerful method for engineering of specific affinity proteins such as antibodies and alternative scaffold proteins. For selections from combinatorial protein libraries, robust and high-throughput selection platforms are needed. An attractive technology for this purpose is cell surface display, offering many advantages, such as the quantitative isolation of high-affinity library members using flow-cytometric cell sorting. This thesis describes the development, evaluation and use of bacterial display technologies for the engineering of affinity proteins. Affinity proteins used in therapeutic and diagnostic applications commonly aim to specifically bind to disease-related drug targets. Angiogenesis, the formation of new blood vessels from pre-existing vasculature, is a critical process in various types of cancer and vascular eye disorders. Vascular Growth Factor Receptor 2 (VEGFR2) is one of the main regulators of angiogenesis. The first two studies presented in this thesis describe the engineering of a biparatopic Affibody molecule targeting VEGFR2, intended for therapeutic and in vivo imaging applications. Monomeric VEGFR2-specific Affibody molecules were generated by combining phage and staphylococcal display technologies, and the engineering of two Affibody molecules, targeting distinct epitopes on VEGFR2 into a biparatopic construct, resulted in a dramatic increase in affinity. The biparatopic construct was able to block the ligand VEGF-A from binding to VEGFR2-expressing cells, resulting in an efficient inhibition of VEGFR2 phosphorylation and angiogenesis-like tube formation in vitro. In the third study, the staphylococcal display system was evaluated for the selection from a single-domain antibody library. This was the first demonstration of successful selection from an antibody-based library on Gram-positive bacteria. A direct comparison to the selection from the same library displayed on phage resulted in different sets of binders, and higher affinities among the clones selected by staphylococcal display. These results highlight the importance of choosing a display system that is suitable for the intended application. The last study describes the development and evaluation of an autotransporter-based display system intended for display of Affibody libraries on E. coli. A dual-purpose expression vector was designed, allowing efficient display of Affibody molecules, as well as small-scale protein production and purification of selected candidates without the need for sub-cloning. The use of E. coli would allow the display of large Affibody libraries due to a high transformation frequency. In combination with the facilitated means for protein production, this system has potential to improve the throughput of the engineering process of Affibody molecules. In summary, this thesis describes the development, evaluation and use of bacterial display systems for engineering of affinity proteins. The results demonstrate great potential of these display systems and the generated affinity proteins for future biotechnological and therapeutic use. / <p>QC 20141203</p>

Enterohemorrhagic Escherichia coli (EHEC) are a subgroup of Shiga toxin producing E. coli that cause gastrointestinal disease with the potential for life-threatening sequelae. Cattle serve as the natural reservoir for EHEC and outbreaks occur sporadically as a result of contaminated beef products and other farming products. E. coli O157:H7 was the first EHEC strain described and has been responsible for hundreds of outbreaks in Canada, Europe, Japan and the U.S.A. since 1982. While certain EHEC virulence mechanisms have been extensively studied, the factors that mediate host colonisation are poorly defined. Autotransporter (AT) proteins have been identified in many Gram-negative pathogens and are unique in that their primary sequence is sufficient to direct their transport across the bacterial membrane system. Many characterised members are associated with virulence. Using conserved AT motifs as a search tool five putative AT proteins were identified in the EHEC O157:H7 EDL933 genome. The genes encoding these proteins (z0402/ehaA, z0469/ehaB, z3487/ehaC, z3948/ehaD and z5029/ehaG) were PCR amplified, cloned and expressed in an E. coli K-12 MG1655flu background. Characterisation revealed that ehaA, ehaB, ehaD and ehaG encode proteins associated with an increase in biofilm formation. EhaA, EhaB and EhaG were found to mediate biofilm formation under continuous flow conditions when expressed in E. coli K-12. Over-expression of either EhaA or EhaG in E. coli K-12 resulted in the formation of large cell aggregates. Three of the AT proteins were found to mediate adhesion when over-expressed in E. coli K-12. The EhaA AT protein mediated binding to primary epithelial cells of the bovine terminal rectum. EhaB promoted adhesion to the extracellular matrix (ECM) proteins laminin and collagen I and was recognised by IgA serum taken from calves challenged with E. coli O157:H7. EhaG is a member of the trimeric autotransporter adhesin (TAA) sub-group of AT proteins and mediated adhesion to colorectal adenocacinoma (Caco-2) epithelial cells. Our results suggest that EhaA, EhaB, EhaD and EhaG may contribute to adhesion, colonisation and biofilm formation by E. coli O157:H7. This study also used a bioinformatic approach to identify AT encoding genes in available E. coli genomes. We identified 156 AT encoding genes in 18 E. coli genomes queried. Alignment and analysis of these proteins identified three broad groups, the serine protease autotransporters of enterobacteriacae (SPATEs), the trimeric autotransporter adhesins (TAAs) and AIDA-I type AT vii proteins. The latter group consisted of a further ten sub-groups. The results demonstrated that E. coli strains encode multiple AT proteins, many of which may have some degree of functional redundancy.

The area of surface expression has gathered a lot of interest from research groups all over the world and much work is performed in the area. Autotransporters have been used for surface expression in Gram-negative bacteria. One of the more commonly used autotransporters is the Adhesin Involved in Diffuse Adherence (AIDA) of pathogenic Escherichia coli. The surface expression of enzymes and vaccine epitopes offer several advantages. Surface expressed enzymes gain similar properties to immobilised enzymes, mainly simplified handling and separation using centrifugation. Surface expressed vaccine epitopes can have longer half-lives inside the animal that is to be immunized and surface groups on the host cell can act as adjuvants, increasing the immune response and leading to a better immunisation.    However, while much basic research is directed towards mechanisms of surface expression using autotransporters there are few reports regarding production of surface expressed protein. Thus the aim of this work was the optimisation of the yield and productivity of surface expressed protein. Protein Z, an IgG-binding domain of Staphylococcal protein A, was used as a model protein for the investigation of which cultivation parameters influenced surface expression. The choice of cultivation medium gave the largest impact on expression, which was attributed to effects based on the induction of the native promoter of AIDA. The AIDA system was then used for the expression of two Salmonella surface proteins, SefA and H:gm, with potential for use as vaccine epitopes. SefA was verified located on the cell surface, and H:gm was found in the outer membrane of the host cell, though only in proteolytically truncated forms lacking the His6-tag used for detection. This proteolysis persisted in E. coli strains deficient for the outer membrane protease OmpT and was concluded to be dependent on other proteases. The removal of proteolysis and further optimisation of the yield of surface-expressed protein are important goals of further work. / QC 20111123 / Vinnova: BIO-AMINES / SIDA Vietnam: Production of viral proteins for vaccine development

Fusobacterium nucleatum is a Gram-negative, anaerobic bacterium that is a member of the human oral microbiota. Although it is a normal resident of the mouth, it is associated with a number of human diseases including: sepsis, inflammatory bowel disease (IBD), and colorectal cancer (CRC). Despite the important association of F. nucleatum with human health and disease, remarkably little is known about the molecular mechanisms underlying these infections. This knowledge gap can, in part, be attributed to a lack of molecular tools and experimental workflows. Creating the genetic tools to fill this knowledge gap is an imperative undertaking for the future development of treatments for diseases involving F. nucleatum. Previous work in the field has assigned functions to just a handful of Fusobacterium proteins (Fap2, FadA), and only two of those proteins have a well-defined role in the host-pathogen relationship. This dissertation contains work that lays the molecular and genetic foundation for future studies involving F. nucleatum by creating a unique gene deletion system while simultaneously establishing broadly applicable experimental workflows and molecular tools to study initial bacterial attachment and invasion processes crucial to Fusobacterium virulence. Marker-less gene deletions confirm the importance of Fap2 in host-cell attachment and invasion and suggest a lesser role in invasion for FadA, representing a significant revision to the Fusobacterium-host relationship. Also, our system allows for the overexpression and purification of virulence factors directly from Fusobacterium for the first time. This permits us to study aspects of Fusobacterium protein biology that were previously impossible and will provide further insights into the nature of Fusobacterium virulence. A custom suite of molecular tools was also developed to facilitate recombinant expression of these proteins in general laboratory settings using simple E. coli protein expression systems. We have used these new technologies to express and purify a number of potential Fusobacterium virulence factors as detailed in this dissertation. Also contained in this dissertation is the application of these breakthroughs to probe the function of a novel F. nucleatum outer membrane phospholipase, FplA. Phospholipases are important virulence factors in a number of well-studied human pathogens including Pseudomonas aeruginosa and Legionella pneumophila, where they interfere with host cellular signaling processes to increase intracellular bacterial survival. Our data show that FplA is a Class A1 phospholipase (PLA1) with robust catalytic activity capable of binding to and cleaving a number of lipid types. Additionally, we show that it has the ability to bind to important host signaling lipids including phosphatidylinositol 3, 5-bisphosphate and phosphatidylinositol 3, 4, 5-triphosphate. These data suggest FplA may play a role in manipulating the intracellular processes of host cells. Taken together, work in this dissertation provides tools and experimental frameworks for the future study of F. nucleatum pathogenesis while identifying and initially characterizing a new, potentially significant, virulence factor in FplA. / Doctor of Philosophy

Fusobacterium nucleatum is a Gram-negative bacterium that serves as a bridging organism in polymicrobial biofilms within the oral cavity. Although the bacterium is abundant in healthy gingival tissue, recent studies have found that F. nucleatum is associated with a wide-spectrum of human diseases which include periodontal disease, preterm birth, endocarditis, colorectal cancer, and pancreatic cancer. Previous studies of F. nucleatum virulence have uncovered two surface adhesins, Fap2 and FadA, that interact with the surface of human cells; however, the study of new virulence factors was previously limited as there was no gene deletion system available to functionally analyze F. nucleatum proteins. Interestingly, F. nucleatum has a diverse landscape of structurally unique surface adhesins called Type 5c secreted trimeric autotransporter adhesins (TAAs), which are a family of proteins that are historically known for their contributions to bacterial pathogenesis. This dissertation encompasses the use of recombinant protein expression systems and newly developed gene deletion technology to provide a foundational understanding of the contribution of Type 5c secreted proteins in F. nucleatum pathogenesis. Our results show that the presence of TAAs on the surface of F. nucleatum contribute to the bacterium's ability to bind and invade human cells, establishing the need to characterize other F. nucleatum surface proteins. Additionally, our studies analyzed the proinflammatory landscape induced by F. nucleatum through the identification of specific cytokines that are being secreted during in vitro infections of human cells. Cytokine signaling is a critical aspect of the host cell immune response as it promotes the recruitment of immune cells to the site of infection for efficient clearance of bacterial pathogens. While it has been well established that F. nucleatum modulates the secretion of IL-8, our studies identified that the bacterium also promotes the secretion of CXCL1, which is an important signaling protein that promotes tumor metastases. Overall, the work provided in this dissertation has delivered the initial characterization of TAAs in F. nucleatum virulence, a framework for future studies of Type 5c secreted proteins in Fusobacterium pathogenesis, and the role of Fap2 and FadA in promoting pro-inflammatory and pro-metastatic signaling from colorectal cancer cells. / Master of Science in Life Sciences / Fusobacterium nucleatum is a Gram-negative bacterium that serves as a bridging organism in polymicrobial biofilms within the oral cavity. Although the bacterium is abundant in healthy gingival tissue, recent studies have found that F. nucleatum is associated with a wide-spectrum of human diseases which include periodontal disease, preterm birth, endocarditis, colorectal cancer, and pancreatic cancer. Previous studies of F. nucleatum virulence have uncovered two surface adhesins, Fap2 and FadA, that interact with the surface of human cells; however, the study of new virulence factors was previously limited as there was no gene deletion system available to functionally analyze F. nucleatum proteins. Interestingly, F. nucleatum has a diverse landscape of structurally unique surface adhesins called Type 5c secreted trimeric autotransporter adhesins (TAAs), which are a family of proteins that are historically known for their contributions to bacterial pathogenesis. This dissertation encompasses the use of recombinant protein expression systems and newly developed gene deletion technology to provide a foundational understanding of the contribution of Type 5c secreted proteins in F. nucleatum pathogenesis. Our results show that the presence of TAAs on the surface of F. nucleatum contribute to the bacterium’s ability to bind and invade human cells, establishing the need to characterize other F. nucleatum surface proteins. Additionally, our studies analyzed the proinflammatory landscape induced by F. nucleatum through the identification of specific cytokines that are being secreted during in vitro infections of human cells. Cytokine signaling is a critical aspect of the host cell immune response as it promotes the recruitment of immune cells to the site of infection for efficient clearance of bacterial pathogens. While it has been well established that F. nucleatum modulates the secretion of IL-8, our studies identified that the bacterium also promotes the secretion of CXCL1, which is an important signaling protein that promotes tumor metastases. Overall, the work provided in this dissertation has delivered the initial characterization of TAAs in F. nucleatum virulence, a framework for future studies of Type 5c secreted proteins in Fusobacterium pathogenesis, and the role of Fap2 and FadA in promoting pro-inflammatory and pro-metastatic signaling from colorectal cancer cells

Directed evolution is today an established strategy for generation of new affinity proteins. This thesis describes the development of a cell-display method using Escherichia coli for directed evolution of Affibody molecules. Further, the thesis describes rational design of Affibody-based tracers, intended for future patient stratification using medical imaging. Fusing recombinant proteins to various autotransporters is a promising approach for efficient surface display on the surface of E. coli, as well as for construction of high-complexity libraries. In paper I, we successfully engineered an expression vector for display of Affibody molecules using the autotransporter AIDA-I. In paper II, a large Affibody library of 2.3x109 variants was constructed and screening using FACS resulted in new specific binders in the nanomolar range. In paper III, we demonstrated Sortase-mediated secretion and conjugation of binders directly from the E. coli surface.  The three following studies describe rational design of Affibody-based tracers against two cancer-associated targets for molecular imaging. First, anti-HER3 Affibody molecules were labelled with 111In, and SPECT imaging showed that the conjugates specifically targeted HER3-expressing xenografts. Furthermore, labeling with 68Ga for PET imaging showed that tumor uptake correlated with HER3 expression, suggesting that the tracers have potential for patient stratification. The last study describes the development and investigation of anti-EGFR Affibody-based imaging agents. Labeled with 89Zr, the Affibody tracer demonstrated higher tumor uptake at 3 h post injection than the anti-EGFR antibody cetuximab at 48 h post injection.  In conclusion, this thesis describes new tools and knowledge that will hopefully contribute to the development of affinity proteins for biotechnology, therapy and medical imaging in the future. / <p>QC 20170904</p>

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal.

Les autotransporteurs monomériques, appartenant au système de sécrétion de type V, correspondent à une famille importante de facteurs de virulence bactériens. Plusieurs fonctions, souvent essentielles pour le développement d’une infection ou pour le maintien et la survie des bactéries dans l’organisme hôte, ont été décrites pour cette famille de protéines. Malgré l’importance de ces protéines, notre connaissance de leur biogenèse et de leur mécanisme d’action demeure relativement limitée. L’autotransporteur AIDA-I, retrouvé chez diverses souches d’Escherichia coli, est un autotransporter multifonctionnel typique impliqué dans l’adhésion et l’invasion cellulaire ainsi que dans la formation de biofilm et d’agrégats bactériens. Les domaines extracellulaires d’autotransporteurs monomériques sont responsables de la fonctionnalité et possèdent pratiquement tous une structure caractéristique d’hélice β. Nous avons mené une étude de mutagenèse aléatoire avec AIDA-I afin de comprendre la base de la multifonctionnalité de cette protéine. Par cette approche, nous avons démontré que les domaines passagers de certains autotransporteurs possèdent une organisation modulaire, ce qui signifie qu’ils sont construits sous la forme de modules fonctionnels. Les domaines passagers d’autotransporteurs peuvent être clivés et relâchés dans le milieu extracellulaire. Toutefois, malgré la diversité des mécanismes de clivage existants, plusieurs protéines, telles qu’AIDA-I, sont clivées par un mécanisme qui demeure inconnu. En effectuant une renaturation in vitro d’AIDA-I, couplée avec une approche de mutagenèse dirigée, nous avons démontré que cette protéine se clive par un mécanisme autocatalytique qui implique deux acides aminés possédant un groupement carboxyle. Ces résultats ont permis la description d’un nouveau mécanisme de clivage pour la famille des autotransporteurs monomériques. Une des particularités d’AIDA-I est sa glycosylation par une heptosyltransférase spécifique nommée Aah. La glycosylation est un concept plutôt récent chez les bactéries et pour l’instant, très peu de protéines ont été décrites comme glycosylées chez E. coli. Nous avons démontré que Aah est le prototype pour une nouvelle famille de glycosyltransférases bactériennes retrouvées chez diverses espèces de protéobactéries. La glycosylation d’AIDA-I est une modification cytoplasmique et post-traductionnelle. De plus, Aah ne reconnaît pas une séquence primaire, mais plutôt un motif structural. Ces observations sont uniques chez les bactéries et permettent d’élargir nos connaissances sur la glycosylation chez les procaryotes. La glycosylation par Aah est essentielle pour la conformation d’AIDA-I et par conséquent pour sa capacité de permettre l’adhésion. Puisque plusieurs homologues d’Aah sont retrouvés à proximité d’autotransporteurs monomériques putatifs, cette famille de glycosyltranférases pourrait être importante, sinon essentielle, pour la biogenèse et/ou la fonction de nombreux autotransporteurs. En conclusion, les résultats présentés dans cette thèse apportent de nouvelles informations et permettent une meilleure compréhension de la biogenèse d’une des plus importantes familles de protéines sécrétées chez les bactéries Gram négatif. / Monomeric autotransporters, a family of proteins that use the type V secretion pathway, are important mediators of virulence for many bacterial pathogens. Many functions important for host colonization and survival have been described for these proteins. Despite the recognized importance of this family of proteins, the mechanisms that are required for the biogenesis and functionality of monomeric autotransporters still remain poorly understood. The Escherichia coli adhesin involved in diffuse adherence (AIDA-I) is a classical multifunctional autotransporter protein that mediates bacterial aggregation and biofilm formation, as well as adhesion and invasion of cultured epithelial cells. Extracellular domains of autotransporters are responsible for the protein function and fold into a characteristic β-helical structure. We performed a random mutagenesis of the AIDA-I passenger domain in order to identify regions involved in the various phenotypes associated with the expression of this protein. Our study suggests that the passenger domain of AIDA-I possesses a modular organization, which means that AIDA-I is built with individual functional modules. Autotransporter passenger domains can be cleaved from the β-domain and released into the extracellular milieu. However, despite the fact that diverse cleavage mechanisms have been previously described, many autotransporters, like AIDA-I, are cleaved by an unknown mechanism. By monitoring the in vitro refolding and cleavage following by site-directed mutagenesis, we showed that AIDA-I processing is an autocatalytic event that involves two acidic residues. Our results unveil a new mechanism of auto-processing in the autotransporter family. AIDA-I is one of the few glycosylated proteins found in Escherichia coli. Glycosylation is mediated by a specific heptosyltransferase encoded by the aah gene, but little is known about the role of this modification and the mechanism involved. Our findings suggest that Aah represents the prototype of a new large family of bacterial protein O-glycosyltransferases that modify various substrates recognized through a structural motif. Furthermore, we showed that glycosylation occurs in the cytoplasm by a cotranslational mechanism. These observations are unique in bacteria and represent a significant advance in our comprehension of prokaryotic glycosylation. We also showed that glycosylation is required to ensure a normal conformation of AIDA-I and, as a consequence, is necessary for its cell-binding function. The finding that other autotransporters or large adhesin-encoding genes are linked to Aah homologue-encoding genes suggests that glycosylation may be important, if not essential, for the function of these proteins, as for AIDA-I. In conclusion, the results presented in this thesis bring new information about the autotransporter family and also give new insight into the mechanisms that are important for different aspects of the biogenesis of monomeric autotransporters.

AIDA-I (adhesin involved in diffuse adherence) est une importante adhésine autotransporteur exprimée par certaines souches de Escherichia. coli impliquée dans la colonisation des porcelets sevrés causant la diarrhée post-sevrage et la maladie de l’œdème. Une précédente étude de notre laboratoire a identifié l’apolipoprotéine AI (ApoAI) du sérum porcin, la protéine structurale des lipoprotéines à haute densité, comme récepteur cellulaire putatif de AIDA-I. L’interaction entre ces deux protéines doit être caractérisée. Ici, nous montrons par ELISA que AIDA-I purifiée est capable d’interagir avec l’ApoAI humaine, mais également avec les apolipoprotéines B et E2. L’ApoAI est rencontrée sous deux formes, soit libre ou associée aux lipides. Nous montrons que la forme libre n’interagit pas avec les bactéries AIDA-I+ mais s’associe spécifiquement à l’ApoAI membranaire de cellules épithéliales HEp-2. Afin d’étudier le rôle de l’ApoAI dans l’adhésion des bactéries, nous avons infecté des cellules HEp-2 en présence d’anticorps dirigés contre l’ApoAI, mais l’adhésion des bactéries AIDA I+ n’a jamais été réduite. De plus, l’induction de l’expression de l’ApoAI par fénofibrate et GW7647 chez les cellules Caco 2 polarisée et Hep G2, n’a pas permis l’augmentation de l’adhésion cellulaire des E. coli exprimant AIDA-I. Notre étude suggère davantage que l’interaction entre AIDA-I et ApoAI n’intervient pas dans les mécanismes d’adhésion cellulaire. / The adhesin involved in diffuse adherence (AIDA-I) is an important autotransporter adhesin expressed by some strains of Escherichia coli and is involved in the intestinal colonisation of weaned piglets, causing the postweaning diarrhea and the edema disease. A previous study from our laboratory identified the apolipoprotein AI (ApoAI) from porcine serum, the structural protein of high density lipoproteins, as a putative receptor of AIDA-I. The interaction between these two proteins must be characterized. Here, we show that purified AIDA-I, using an ELISA assay, is able to bind the human ApoAI and the apolipoprotein B and E2. The ApoAI is found under two forms, either free or bound to lipid. We show that the free form of ApoAI does not interact with AIDA-I+ bacteria but specifically interact with membrane bound ApoAI on Hep-2 epithelial cells. To study the role of ApoAI in the adhesion of bacteria, we infected Hep-2 cells preincubated with antibodies to ApoAI. The adhesion of AIDA-I+ bacteria to the cells couldn’t be reduced. Additionally, the induction of ApoAI synthesis using fenofibrate and GW7647 on polarized Caco-2 or Hep G2 cells did not increase the adhesion of AIDA-I+ bacteria. Our study suggests that the interaction between AIDA-I and ApoAI is not involved in the cellular adhesion of the bacteria.

Les autotransporteurs monomériques représentent le système de sécrétion le plus simple et le plus utilisé chez les bactéries à Gram négatif. Les autotransporteurs monomériques sont des protéines modulaires qui contiennent toute l’information pour leur sécrétion dans leur séquence. Les phénotypes associés à l’expression d’un autotransporteur peuvent être très variés et, souvent, les autotransporteurs sont des protéines multifonctionnelles. C’est le cas notamment des autotransporteurs AIDA-I, TibA et Ag43 d’Escherichia coli qui promouvoient l’adhésion et l’invasion de cellules épithéliales, l’auto-agrégation des bactéries et la formation de biofilm. Ces trois autotransporteurs ont d’ailleurs été regroupés dans une même famille, appelée les autotransporteurs auto-associatifs (SAATs). À cause de leur fonctionnalité, les SAATs sont considérés comme étant d’importants facteurs de virulence d’Escherichia coli. Toutefois, il existe plusieurs différences entre les SAATs qui ne sont pas bien comprises, si bien que leur rôle pour les bactéries n’est toujours pas bien compris. Nous avons donc d’abord caractérisé TibA, le membre des SAATs le moins bien étudié à l’aide d’une étude structure-fonction. Nous avons observé que TibA était une protéine modulaire et que son domaine fonctionnel était composé de deux modules : un module d’auto-agrégation en N-terminal et un module d’adhésion en C-terminal. En comparant nos résultats avec ceux obtenus pour les autres SAATs, nous avons réalisé que l’organisation des trois SAATs était très variée, c’est-à-dire que les trois SAATs sont composés de modules différents. Nous avons par ailleurs observé cet arrangement en modules lorsque nous avons analysé plusieurs séquences d’aidA, suggérant qu’un mécanisme d’échange et d’acquisition de modules était à la base de l’évolution des SAATs. Sans surprise, nous avons aussi observé que la famille des SAATs ne se limitait pas à AIDA-I, TibA et Ag43 et ne se limitait pas à Escherichia coli. La comparaison a aussi révélé l’importance du phénotype d’auto-agrégation dans la fonctionnalité des SAATs. Nous avons donc entrepris une étude du mécanisme d’auto-agrégation. Nos résultats on montré que l’auto-agrégation était le résultat d’une interaction directe SAAT/SAAT et ont mis en évidence un mécanisme similaire à celui utilisé par les cadhérines eucaryotes. De plus, nous avons observé que, comme les cadhérines, les SAATs étaient impliqués dans des interactions homophiliques; un SAAT interagit donc spécifiquement avec lui-même et non avec un différent SAAT. Finalement, les SAATs font parties des quelques protéines qui sont glycosylées chez Escherichia coli. Nous avons déterminé que le rôle de la glycosylation de TibA était de stabiliser la protéine et de lui donner la flexibilité nécessaire pour moduler sa conformation et, ainsi, être pleinement fonctionnelle. Globalement, nos résultats suggèrent que les SAATs sont des molécules « cadhérines-like » qui permettent la reconnaissance de soi chez les bactéries. Une telle habilité à discriminer entre le soi et le non-soi pourrait donc être utilisée par les bactéries pour organiser les communautés bactériennes. / Autotransporters are versatile virulence factors of Gram-negative bacteria and use one of the simplest and most widespread secretion system in bacteria. The name autotransporter originate from the observation that all the information needed for the secretion of the protein is encoded in its own sequence, meaning that autotransporters do not need a specialized secretion apparatus. Many autotransporters are multifunctional proteins and can perform a large variety of functions. The self-associating autotransporters (SAATs), represented by AIDA-I, TibA and Ag43, are such multifunctional proteins and can mediate the adhesion and invasion of epithelial cells, the auto-aggregation of bacteria and the formation of biofilm. Because of these functionalities, SAATs are considered important virulence factors of Escherichia coli. However, there are many differences between the SAATs and we still do not know their exact role for the bacteria. Therefore, we have realized a structure-function study of TibA, the least studied SAAT. Our study showed that TibA is a modular protein and that the functional domain of TibA is composed of two modules: an N-terminal module responsible for auto-aggregation and a C-terminal module responsible for adhesion. Our results showed that the organization of AIDA-I, TibA and Ag43 is different and that the SAATs represent different assemblies of modules. We also observed the modular organization when we analyzed various sequence of aidA, suggesting that the SAATs have evolved by a mechanism of domain shuffling. Not surprisingly, we have found new SAATs in Escherichia coli and in other proteobacteria. Our results also highlighted the importance of auto-aggregation in the functionality of the SAATs. We therefore assessed the mechanism of SAAT-mediated auto-aggregation of bacteria. Our results showed that SAATs mediate auto-aggregation of bacteria through direct SAAT/SAAT interactions and that these interactions were reminiscent of the interactions made by cadherin molecules in eukaryotes. We further observed that the SAATs were involved in homophilic interactions, as it is the case with cadherin molecules. SAATs are part of the few proteins that are glycosylated in Escherichia coli. We therefore characterized the glycosylation of TibA and found that glycosylation of TibA stabilized the protein and allowed the protein to modulate its conformation, resulting in a fully functional protein. Taken together, our results suggest that the SAATs may be cadherin-like molecules by bacteria in order to discriminate between self and non-self. Such an ability to discriminate self from non-self is rarely evoked in bacteria, but could play a role in the organization of multicellular communities.