The pneumococcus

Contacts: T.Vernet, A.M. Di Guilmi, A.Zapun, C.Durmort

Our group is interested in the biology of pneumococcus, an important human pathogen. We study the interactions of the pneumococcus with its host, the metabolism of its cell wall, its mechanism of division, as well as its resistance to beta-lactam antibiotics.

Pneumococcal infections are responsible for over a million deaths per year worldwide. The pneumococcus is a Gram-positive bacterium with the shape of a rugby ball that colonizes asymptomatically the rhino-pharynx, but which causes diseases when it invades other sites and tissues: otitis, sinusitis, pneumoniae, meningitis, and bacteremiae. Today in France, close to half of S. pneumoniae isolates are resistant to several antibiotics, particularly to beta-lactams (penicillins…). Current vaccines are quite effective but do not protect against all the virulent strains, and they are progressively losing their efficacy due to the selection of strains that are not covered.

The bacterial cell wall is of particular interest for several reasons:

The cell wall harbors numerous protein virulence factors that mediate interactions between the pneumococcus and human tissues. These surface proteins can be involved in the adhesion to human cells, their invasion, or in escaping the immune system. Understanding the role of surface proteins is central to the understand the origin of diseases caused by the pneumococcus. Virulence factors also constitute promising vaccinal antigens or targets for novel antibiotics.

The main constituent of the cell wall is the peptidoglycan, a polymer of disaccharides and peptides that surrounds the whole cell, confers its shape and protects the cell from its internal osmotic pressure. Several classes of antibiotics such as the beta-lactams (that include penicillins) or glycopeptides (vancomycin) act by preventing peptidoglycan synthesis. Enzymes responsible for the assembly of the peptidoglycan are the targets of the beta-lactams, the Penicillin-Binding Proteins (PBPs). Their mechanism remains mysterious, as is the organisation of the peptidoglycan itself. Another fundamental question is that of the regulation of the cell wall synthesis during cell division, in relationship with other processes such as membrane invagination and chromosome segregation.

Surface proteins and virulence

Many questions remain concerning the physiopathology of the pneumococcus. We are aiming to understand how the bacteria, which colonizes the nasopharynx asymptomatically, can invade sterile sites (lungs, blood and meninges) to cause diseases. In particular, it is necessary to understand how the pneumococcus identifies its environment in order to adapt its metabolism, to adhere to and to invade host tissues, and to escape to the innate immunity. Many proteins of the pneumococcal cell surface play important roles by interacting with host componants. Detailed structural and functional characterization of these interactions is paramount to understanding the molecular mechanisms of pneumococcal pathogenesis.

Environmental metabolic adaptation

The colonization of the nasopharynx, the invasion of other tissues by the pneumococcus, an dits adaptation to drastic environmental changes, are important steps in the virulence processes, whose central players are proteins that are at the surface of S. pneumoniae. We have thus characterized AdcAII, a zinc-binding surface lipoprotein. The crystal structure of the recombinant protein has confirmed the specificity for the binding of one zinc atom and the homology to ABC transporters (Loisel et al. 2008, LCCP). Moreover, the presence of zinc in a defined medium repress the expression of the adcaII gene in the R6 strain. We are exploring the mechanisms of zinx homeostasis and possible relationships between zinc control and the interactions with the human host.

Structure cristallographique de AdcAII qui lie un atome de zinc. (Loisel et al. 2008)

Screening for interactions between the pneumococcus and the host To better understand interactions between pneumococcus and its human host, our team identifies and investigates interactions between proteins from the bacterial cell surface and some human cells and proteins. An in vitro screen for molecular interactions using an extended set of recombinant pneumococcal surface proteins has revealed a host of new interactions with human proteins of the extracellular matrix and of the innate immunity. Antibodies against some of the surface proteins have been found in human sera (collaboration C.-A. Siegrist, Genève). These pneumococcal proteins are potential vaccinal antigens. We are investigating the biological significance of the interactions through detailed analysis of the binding and function in vitro and in vivo, and of the high resolution structure of participating pneumococcal proteins (Zhang et al. 2009).

Crossing the epithelial and endothelial barrier through intercellular junctions

The crystal structure of CBPE (choline binding proteine E) has been solved (Garau et al. 2005, LPM). We have shown that CBPE participes to the recruitment of human plasminogen at the bacterial surface. The binding of CBPE and plasminogen occurs via the Kringle domains of the latter. The surface bound plasminogen can be activated in plasmin. The proteolytic activity of plasmin can be used by the pneumococcus to invade tissus through the extracellular matrix and the inter-endothelial junctions (Attali et al. 2008).

Déstabilisation des jonctions intercellulaires par des pneumocoques porteurs de plasmine/plaminogène. (Attali et al. 2008)

Structure and function of the pilus in S. pneumoniae

Pili have been recently found at the surface of Gram-positive bacteria, including various strains of pneumococci, where they play a role in the adherence to the host tissues. The fibrillar structure of the pilus is formed by a polymer of the protein RrgB. The minor pilin RrgA and RrgC are attached to the RrgB fiber. Pilins are linked by covalent bonds that are formed enzymes termed sortases. We are interested in the mechanisms of assembly of the pilus. The crystal structure of the three sortases involved in the assembly of the pneumococcal pilus have been determined in collaboration with the team Bacterial pathogenicity directed by Andrea Dessen at the IBS (LPM) (Manzano et al. 2008). In order to decypher the assembly steps of the pilus, a co-expression system of the pilins and sortases has been set up. We have thus shown that the C-1 sortase catalyses the polymerization of RrgB. The minor pilin RrgA is responsible for the adherence of the pilus to the host cells. The recently solved crystal structure of RrgA will help to define interaction sites (Izoré et al. 2009). Every pilin contains unconventional internal covalent bonds between lysine and asparagine residues. The formation of these bonds is autocatalyzed by a glutamate or an aspartate. These intramolecular bonds have been demonstrated using site-directed mutagenesis and mass spectrometry. Their role in the stabilization of the pilins has been shown by limited proteolysis and thermal denaturation.

Localisation of the major pilin RrgB (orange) at the surface of TIGR4 pneumococcus. DNA is in blue.

Cell wall and division

Our studies of the mechanisms of resistance to beta-lactams over the last years has led us to investigate the synthesis of the peptidoglycan. The assembly of the peptidoglycan, the main constituent of the cell wall, is catalyzed by the PBPs (Penicillin-Binding Proteins), which are also the targets of beta-lactams. The reaction of the PBPs with beta-lactams with understood in some details, the mechanisms of resistance due to mutation in PBPs are also well understood, structurally and enzymatically(Carapito et al. 2006, Job et al. 2008, Zapun et al. 2008a, LPM). It is thus paradoxical that the physiological reactions of peptidoglycan synthesis catalyzed by the PBPs are nearly completely unknown. Understanding how PBPs that are altered to resist beta-lactams remain functional for cell wall synthesis should provide fresh insights to combat resistance.

Crystal structure of a variant of PBP1a that confers a high level of resistance to beta-lactams. (Job et al. 2008)

We are trying to reproduce peptidoglycan synthesis in vitro using recombinant pneumococcal membrane protein complexes, which include PBPs and accessory proteins. Membrane proteins are produced in E. coli with the expertise of the RoBioMol platform.

Division and cell wall synthesis are intimately related processes (Zapun et al. 2008b).We are investigating structurally and functionally complexes of division proteins, such as DivIB, DivIC and FtsL (Noirclerc et al. 2005), which could also play a role in peptidoglycan synthesis. The structure of this complex has been studied in solution by RMN, SAXS and SANS. It was found that one face of a domain of DivIB interacts with the C-terminal extremity of an elongated rof formed by the coiled-coil dimer of DivIC and FtsL.

Ab initio model from SANS data of a soluble form of the complex DivIB/FtsL/DivIC. (Masson et al. 2009)

The deletion of divIB in pneumococcus alters the division and the separation of daughter cells, and increases the susceptibility to beta-lactams (Le Gouellec et al. 2008). The localization of the division and cell wall assembly proteins in diverse conditions is studied by immunofluorescence, or using GFP-fusions.

Morphology of a wild type pneumococcus (R6) and the deletion mutant (deltadivIB), which does not produce the DivIB protein. (Le Gouellec et al. 2008, collab. O. Massidda, Università di Cagliari, Italie).

Localization of the fusion protein GFP-DivIB in pneumococcus.