Unit: Molecular Microbial Pathogenesis
Director: Philippe SANSONETTI
Our Research Unit is involved in studying the molecular, cellular and tissular basis of the rupture, invasion and inflammatory destruction of the intestinal barrier by invasive bacteria, as well as the defense and protective mechanisms against these infections. Our major model is Shigella, the aetiological agent of bacillary dysentery. For this study, we apply a multidisciplinary approach that encompasses molecular genetics, functional genomics, cell biology, experimental medicine and immunology. The regulatory mechanisms that control expression of bacterial virulence genes are investigated. Their products (i.e. effectors of pathogenicity) modify the behaviour of both epithelial cells and phagocytic cells interacting with their cellular targets. These cross talks between bacteria and cells lead to bacterial internalisation, intracellular motility and cell to cell spread of bacteria. Interaction of invasive shigellae with phagocytic or epithelial cells elicits a cascade of pro-inflammatory signals that causes disruption of the epithelial barrier and, eventually, its destruction. We develop innovative imaging techniques to vizualize and analyze the initiation and development of infection and inflammation at cell and tissue levels. We also develop animal models of intestinal inflammation. We are also studying the mechanisms of induction and the effectors of the adaptive immune protection against Shigella and how the innate immune response, dominated by strong inflammation, influences the quality of this adaptive response. These studies are applied to the development of vaccine candidates against bacillary dysentery. Phase 1 and 2 clinical trials are currently underway.
Genetic analysis of the invasive phenotype of Shigella flexneri
Group leader: Claude Parsot. Post-doctoral scientists: Maria Mavris, Kaïs Jamoussi, Dong Wong Kim. PhD Student: Christophe Penno. Technician: Elisabeth Ageron.
Determinants of entry into and dissemination of bacteria within epithelial cells are encoded by a plasmid of 213 kb, pWR100, the sequence and annotation of which have been determined, in collaboration with the Laboratory of Genomics of Pathogenic Microorganisms. This plasmid carries a 30-kb region that encodes a type III secretion apparatus (the Mxi-Spa apparatus), proteins that are secreted by this apparatus (the IpaA-D, IpgB, and IpgD proteins), and cytoplasmic chaperones (IpgC, the chaperone for IpaB and IpaC, and IpgE, the chaperone for IpgD). The secretion apparatus is activated upon contact of bacteria with eukaryotic cells, which induces secretion of IpaA-D, IpgB, and IpgD. The plasmid also encodes approximately 20 other proteins that are secreted by the Mxi-Spa apparatus (the Osp and IpaH proteins), some of these proteins being produced only in conditions of active secretion.
Work is focussed on three main points: 1) the role of Osp proteins in the pathogenicity of S. flexneri is investigated using a genetic approach, by inactivating the corresponding genes and characterizing the phenotypes of the mutants both in vitro and in vivo. Two mutants coudl ne characterized. INterestingly, OspF shows an attenauted virulence pheonotype, whereas OspG shows a stongly exagerated viurlence phenotype. 2) Interactions between secreted proteins and chaperones is analyzed using the two hybrid system in the yeast, copurification assays in S. flexneri and structural analysis following cristallography. We have characterized the sites of interactions of IpgC on IpaB and IpaC and identified a new chaperone, Spa15, which is associated with IpaA, IpgB, and OspC3. Structure of these chaperons is currently studied in collaboration with cristallographers ; 3) The mechanism of transcription control of osp and ipaH genes by the activity of the Mxi-Spa secretion apparatus has been elucidated and involves a transcriptional activator of the AraC family, MxiE, the activity of which is modulated by the IpgC chaperone and responds to the presence of a consensus operator sequence upstream of these genes.
These results allow us to draw a general framework on the function of the type III secretion pathway in S. flexneri. At 37°C, the Mxi-Spa apparatus is produced and assembled in an inactive form. The IpaA-D, IpgB, and IpgD proteins are produced and stored in the cytoplasm, most of them in association with specific chaperones. Upon contact of bacteria with host cells, the Mxi-Spa apparatus is activated, allowing delivery of bacterial invasins within or beyond the membrane of eukaryotic cells. The IpgC chaperone then activates the transcriptional activator MxiE, leading to expression of a second set of secreted effectors, the function of which has now to be elucidated.
Picture #1 : The 214 kb-virulence plasmid pWR100 of Shigella flexneri.
Molecules and signals involved in the entry and dissemination of Shigella in epithelial cells
Group leader: Guy Tran Van Nhieu. Post-doctoral scientist: Nalini Ramarao, Caroline Clair. Graduate student: Laurence Bougnères. Technician: Joëlle Mounier.
We are analysing bacterial signalling to the host cell that causes cytoskeletal rearrangements leading to Shigella macropinocytosis by epithelial cells. A particular emphasis is being put on pathways that involve the small GTPases of the Rho family, Cdc42 and Rac, and the tyrosine kinase p60c-src. These two signalling pathways are engaged simultaneously by Shigella during epithelial cell invasion, and we have obtained evidence that IpaC is directly implicated in the activation of these pathways, following its insertion in the epithelial cell membrane. Cdc42 and Rac activation are required for actin polymerisation mediated by the Arp2/3 complex, that lead to the formation of filopodia and membrane leaflets that engulf the bacterial body. The tyrosine kinase p60c-src regulates actin polymerisation mediated by these GTPases, at a distance from the bacterial-cell membrane interaction site, through the phosphorylation of cortactin, a 80 kDa actin-binding protein which, thereby gets recruited to the cell membrane and activates the Arp2/3 nucleator. The molecular mechanisms that lead to the activation of these pathways and their coordination during Shigella invasion are currently under study.
We have also demonstrated the essential role plaid by connexins in cell invasion. Connexin-mediated hemi-channels open in epithelial cells infected by Shigella and release ATP which, in a paracrine way activate the purinergic receptor of non-infected adjacent cells. This original mechanism of cell communication causes recurrent calcium fluxes in non-infected cells and is associated with a major increase in the permissivity of these adjacent cells to bacterial entry and cell to cell spread.
Figure#2: Induction of calcium responses in connexin 26 transfectants of Hela cells during Shigella invasion.
Molecular and cellular bases of the rupture, invasion and inflammatory destruction of the intestinal barrier by Shigella.
Group leaders: Philippe Sansonetti, Régis Tournebize, Laurence Arbibe. Post-doctoral scientist: Stephen Girardin, Irit Paz. PhD Student: Tien Meng-Tsung. Engineer: Thierry Pédron, Monique Singer.
In our ongoing approach at understanding how Shigella invade and cause the inflammatory destruction of the intestinal epithelial barrier, we have confirmed the essential role plaid by the infected epithelial cell itself in a process in which IL-8, IL-1β and TNFα play a key role. Our strategy combines in vitro and in vivo approaches. In collaboration with Dana Philpott's group, we have demonstrated that a cytosolic protein,Nod1/Card4, acts as an intracellular sensor of Peptidoglycan from Gram negative bacteria. In macrophages, its homologue, Nod2/Card15 senses PGN in general and Muramyl dipeptide in particular. Beyond their specific importance to the understanding of the pathogenesis of shigellosis, these results shed light on our understanding of Inflammatory Bowel Diseases such as Crohn's disease and Ulcerative Colitis. It appears that this newly recognised signalling system is a cytosolic equivalent of Toll-Like Receptors (TLR). Activation of Nod1 by oligomerization in the presence of PGN, via RICK, activates the NF-κ B and Jun-terminal kinase (JNK) pathways, thereby triggering pro-inflammatory properties of infected cells. Massive production of IL-8 responds to this activation process. We have carried out a global transcriptional analysis of epithelial cells infected by Shigella in the AFFYMETRIX system and defined a transcriptional profile that confirms the essentially pro-inflammatory orientation taken by invaded epithelial cells. This profile was confirmed in vivo by transcripional analysis of infected human intestinal tissue transplanted in SCID mice (collaboration with Sam Stanley, Washington University). On these bases, in collaboration with Jean-Yves Coppée's group (Genopole, Institut Pasteur), we have developped a macroarray including about 1000 human genes that allows routine transcriptome analysis of infected cells and tissues. We now need to connect in vitro and in vivo observation. For this, we are developing methods for real time analysis of infectious processes based on magnetic resonance imaging and intravital microscopy. We are also developing murine models of intestinal inflammation, including transgenic mice.
Innate and adaptive immunity in bacillary dysentery.
Group leader: Armelle Phalipon. Post-doctoral scientists: Maria-Isabel Fernandez-Martinez. PhD Student: Joao Gamelas Magalhaes. Technician: Audrey Thuizat.
We study the effectors involved in the eradication of primary infection and in the protection against re-infection as well as the influence of innate immunity on the orientation of the adaptive immune response. In collaboration with James Di Santo, we have shown, using a murine model, that NK and TCD4+(α /α ) lymphocytes are both involved in the eradication of primary infection via the production of interferon γ( INF-γ ). In addition, it seems that Shigella, at the very early stages of primary infection, modulate the INF-γ production. This may constitute a key step in the induction and orientation of the protective humoral response.
The humoral response specific for the polysaccharidic part of LPS (O-Ag) is crucial for protection against re-infection. We have shown that the anti-O-Ag IgG-mediated immune response may contribute to protection only if the response is induced at the mucosal level in order to ensure the production of the effectors at the site of bacterial infection. Concerning the secretory IgA (S-IgA)-mediated response required to protect the mucosal surface from bacterial infection, we have shown, in collaboration with Blaise Corthésy, that the secretory component is directly involved in the protective function of S-IgA by ensuring through its glycosylated residues the appropriate localization of IgA for optimal immune exclusion function. Moreover, we have reported that LPS internalized by epithelial cells activates NF-κ B with a slower kinetics than that observed with the invasive bacterium. Interestingly, LPS-mediated NF-κ B activation is inhibited by a monoclonal IgA specific for LPS that intercepts during, its transcytosis, the bacterial product. This is a newly and unexpected anti-inflammatory protective function of S-IgA. Therefore, S-IgA-mediated mucosal protection seems to occur via a double mechanism comprising immune exclusion of bacteria from the epithelial surface and intracellular neutralization of pro-inflammatory bacterial product.
photo3 : A specific dimeric IgA interferes with NF-κ B translocation induced by intracellular S. flexneri 5a LPS
Towards a live, attenuated vaccine against shigellosis.
In collaboration with the Walter Reed Army Institute of Research in the USA and the ICDDR,B in Bangladesh, we continue the phase I and II trials of strain SC602, a live attenuated mutant of S. flexneri 2a. With the support of DGA, we are conducting a phase 1 study of strain SC599, a live attenuated mutant of S. dysenteriae 1. Our aim is to develop a pentavalent vaccine comprising 3 serotypes of S. flexneri, S. dysenteriae 1 and S. sonnei. In addition, we are developping a sub-unit vaccine approach based on the chemical synthesis of polysaccharidic antigens.
Photo 1: The 214 kb-virulence plasmid pWR100 of Shigella flexneri.
Photo 2: Induction of calcium responses in connexin 26 transfectants of Hela cells during Shigella invasion.
Stable connexin 26 (Cx26) transfectants of HeLa cells were loaded with the Fura-2 calcium (Ca2+)probe and infected with invasive Shigella. Ruffle formation during bacterial invasion was monitored by time-lapse phase contrast microscopy, while simultaneously following Fura-2 fluorescence upon excitation at 340nm and 380 nm. Top Panels : Phase contrast aquisition : the arrows point at Shigella entry events. Middle and bottom panels : 340/380nm ratio images of the corresponding field : the arrowheads correspond to cells showing Ca2+ oscillations. Time from the start of infection is indicated in seconds.
Photo 3: A specific dimeric IgA interferes with NF-κ B translocation induced by intracellular S. flexneri 5a LPS
Confocal microscopy images at the focal level corresponding to the nucleus of murine polarized intestinal epithelial cells pretreated with a-S. flexneri 5a LPS dIgA for 6 hr followed by addition of S. flexneri 5a LPS for 2 hr. In LPS 5a (blue) positive cells (a), in which NF-k B p65 subunit (red) is translocated to the nucleus, note the intense red colour in the nucleus and the absence of dIgAC5 (in green). In contrast, NF-k B p65 subunit is not translocated in LPS 5a (blue) and dIgAC5 positive (green) double psitive cells (b). Scale bars: 5 microns.
Keywords: Shigella, intestinal epithelium, invasion, inflammation, immune response, vaccine