Blanc SVSE 8 - Blanc - SVSE 8 - Biochimie, biologie moléculaire et structurale

Molecular mechanisms of signal transduction by BvgS, a paradigm for bacterial Venus-flytrap-domain receptors – MECA VENUS

Signaling by the sensor-kinase BvgS, a model for a family of bacterial receptors with Venus flytrap domains

MECA VENUS : Molecular mechanisms of signal perception and transduction by the sensor-kinase BvgS of the two-component system that controls the virulence regulon of the whooping cough agent Bordetella pertussis. BvgS is a model for a family of bacterial receptors with extracytoplasmic, bilobed Venus flytrap perception domains.

Signal transduction by a multi-domain transmembrane receptor : how does mechanical signaling regulate enzymatic activity ?

BvgS is a multi-domain, dimeric transmembrane protein. In each monomer, two periplasmic VFT domains, one transmembrane segment and a PAS domain precede the DHp and catalytic domains that compose the kinase. In its basal, default state BvgS phosphorylates the transcriptional activator BvgA. Phosphorylated BvgA transactivates the expression of the virulence regulon of Bordetella pertussis. BvgS shifts to a phosphatase mode of activity in the presence of chemical modulators –used in the laboratory but most likely not physiological- such as nicotinate ions. Dephosphorylation of BvgA abolishes the expression of the virulence genes. The signals that are perceived by the bacterium in the human respiratory tract and the mechanisms by which their perception determines the activity of BvgS are the focus of this program.<br />Our first objective is to understand how VFT domains perceive signals and transduce information across the membrane. We want to identify the natural signals for the system, and we also use chemical modulators currently available in the lab to determine their effect on the conformation and dynamics of the VFT domains. Typically, VFT domains close around specific ligands that bind to their inter-lobe cavity. Such a conformational change might initiate signaling. We thus also want to understand the consequences of ligand binding on the transmembrane segments.<br />Our second objective is to understand signal transmission between the various domains of BvgS. Alpha helices are predicted between the VFT domains and the PAS domain, and between the PAS and kinase domains, which might form coiled coils. This structural arrangement may allow for mechanical modes of transmission that we want to decipher.<br />The principles that we will unveil for BvgS will serve as bases to describe the molecular mechanisms of a large family of bacterial sensor-kinases homologous to BvgS and notably found in several important pathogens.

X-ray crystallography is used to determine the structure of the periplasmic moiety and of the PAS domain of BvgS. Characterization of the dynamics of the VFT domains is performed by molecular dynamic simulations and by the analysis of the normal modes of motion of the protein.
Potential BvgS ligands are searched for by in vivo and in vitro methods. A fluorescent reporter system is used to measure BvgS activity in response to a library of compounds added to bacteria in culture. Rapid screening for ligands is performed using an assay that measures a shift of the denaturation temperature of purified BvgS domains, in the presence of candidate molecules.
Further characterization of the interaction between BvgS and a ligand is performed by using biophysical techniques such as microcalorimetry, thermophoresis and electron paramagnetic resonance.
Site-directed mutagenesis is used to modify in a targeted manner the perception domains and the signaling helices of BvgS to determine the effect of the mutations. Enzymatic report systems measure the kinase activity and the response to chemical modulators of the BvgS mutants in Bordetella pertussis. Another method used to determine BvgS activity is the Phos-tagTM assay, which allows electrophoretic separation of the phosphorylated and non-phosphorylated forms of BvgA, followed by their detection by immuno-blotting.
Studies of the topology and the dynamics of the transmembrane segment and of the ? helices between the membrane and the PAS domains, and between the PAS and kinase domains are performed by cysteine scanning. The proportions of dimeric form in the basal state and after the perception of a modulator inform on the proximity of the two monomers at the targeted positions.
Finally, in silico analyses of the sequences of the connectors found in all BvgS homologs of known sequence are performed to identify common motifs involved in the signaling mechanism.

The structure of the periplasmic moity of BvgS shows an intricate dimer, with an extended dimeric interface. The N-terminal VFT1 domains are open and mobile according to our molecular dynamic simulations, while atypically the VFT2 domains are closed without a ligand and display very limited mobility. Forcing the VFT1 closed by engineering a disulfide bond between their lobes causes BvgS to shift to a phosphatase mode of activity. The dynamics of the VFT1 domain and its transmission to the rest of the protein via the VFT2 domains determine the kinase state of BvgS.
Binding of the negative modulator nicotinate to VFT2 alters the dynamics of the periplasmic moiety in a global manner, which causes the transition of BvgS to the phosphatase mode. Our search for physiological ligands has not led to the identification of more affine compounds as yet.
The linker between the PAS and kinase domains of BvgS adopts the topology of a parallel, two-helix coiled coil. In the kinase mode this coil is dynamic. Upon perception of a modulator, which triggers the shift to the phosphatase mode, the coiled coil adopts its hydrophobic interface, hence its reduced mobility. Our model of activity regulation based on a balance between two distinct states of dynamics and rigidity of this coiled coil is likely to apply to the entire family of sensor-kinases homologous to BvgS.
Chimeric BvgS proteins in which the PAS domain has been replaced with the linkers of PAS-less BvgS homologs have various phenotypes that are most likely determined by the length and the composition of the linker. Extensive in silico analyses of the sequences of BvgS homologs has shown the occurrence of two distinct coiled coil registers in those linkers. We hypothesize that perception of a modulating signal will determine which of the two registers predominates, with consequences on coiled coil dynamics. Regulation of the activity of the sensor-kinases would be based on this mechanism.

We have intiated analyses of the helical linker between the VFT2 and PAS domains of BvgS by site-directed mutagenesis and cysteine scanning. Our results suggest that modulator perception by the VFT domains triggers a vertical translation motion of the transmembrane segments. This will be tested by cysteine accessibility studies to determine whether residues buried in the membrane in the kinase state may become exposed in the phosphatase state and vice versa. We will modify the composition of VFT2 surface loops that may be in contact with the polar groups of the membrane lipids to hamper putative lever motions.
We will pursue the characterization of the PAS domain. We are currently solving its structure by X-ray crystallography. Next, we will try to co-crystallize this domain with putative ligands. We have initiated the analysis of the PAS dimeric interface by cysteine scanning. All these data will enable us to establish if the PAS domain mainly facilitates the transition between the two states of BvgS or if it also binds a cytoplasmic ligand.
We will pursue the study of the BvgS chimera with no PAS domain. We will modify the relative stability of the two coiled coil registers to test our model of regulation. We will analyze the linkers of one or two chimeras by cysteine scanning in the kinase and phosphatase states. The coiled coils will be modeled in silico, and molecular dynamic simulations will be performed. Based on all these data we hope to propose a general model applicable to the entire BvgS family.
Finally, the in vivo data on the PAS-less BvgS variants will help us build a model that encompasses the transmembrane segments, the coiled coils and the kinase domains. The most stable structures will serve to model the entire protein including the periplasmic VFT domains.

Dupre et al. (2015a) ‘Virulence reguation with Venus flytrap domains : structure and function of the periplamic moiety of the sensor-kinase BvgS’. PLoS Pathogens 11(3):e1004700

Dupre et al. (2015b) ‘Signal transduction by Bvg sensor-kinase : binding of modulator nicotinate affect conformation and dynamics of entire periplasmic moiety’. J Biol Chem 290, 23307-319.

Lesne et al., (2016) ‘Balance between coiled coil stability and dynamics regulates activity of BvgS sensor-kinase in Bordetella’. mBIO 7 :e02089.

de Ruyck et al. (2016) ‘Molecular docking as a popular tool in drug design an in silico travel’. Advances and Applications in Bioinformatics and Chemistry 9, 1-11.

Dupre et al. (Poster) ‘The periplasmic portion of the BvgS sensor-kinase, a new paradigm in signaling’. Sensory Transduction in Microorganisms Gordon Research Conference, 12-17 January 2014, Ventura, CA, USA.

Lesne et al. (Poster) ‘Role of the PAS domain of BvgS, the sensory-kinase that regulates virulence in Bordetella Pertussis.’ Symposium on Metabolism and Bacteria Pathogenesis, 6-9 April 2014, Osnabrueck, Germany.

Lesne et al. (Poster) ‘Signal transduction in BvgS, the sensor-kinase regulating Bordetella pertussis virulence’. ESF-EMBO Symposium « Bacterial networks-BACNET 15 », 9-14 May 2015, Sant Feliu de Guixols, Spain.

Dupre et al. (Poster) ’Designing functional heterodimers to decipher signal transduction in BvgS’ 6th Congress of European Microbiologists - FEMS 2015 - June 7-11, Maastricht, NL.

Wodak, Lensink et al. ‘Prediction of protein-protein interactions in CAPRI: an increasingly integrative approach’, Intl conference on structural genomics, June 2015, Tel Aviv, Israel.

Lensink et al. ‘Assessment of predicted protein complexes’, CASP11 meeting, December 2014, Cancun, Mexico.

Antoine, R. (commun. orale) ‘Signaling by BvgS: a dynamic story’, 11th Intl Bordetella Symposium, April 5-8 2016, Buenos Aires, Argentina.

The predominant signal transduction systems in bacteria, two-component systems are essential to enable microorganisms to adapt to changes of their environment. They regulate important developmental programs including bacterial virulence. Typically, they are composed of a transmembrane sensor-kinase protein and a cytoplasmic response regulator. Perception of a chemical or physical signal by the sensor leads to kinase activation and autophosphorylation, and then transfer of the phosphoryl group to the response regulator. Thus activated, the latter mediates a specific, frequently transcriptional, cellular response. The whooping cough agent Bordetella pertussis colonizes the upper respiratory tract of humans. Its virulence regulon is controlled by the two-component system BvgAS. At 37°C and in laboratory growth conditions, the BvgAS system is activated, leading to the transcription of the virulence regulon, including genes for B. pertussis’s adhesins and toxins. The virulent Bvg+ phase of B. pertussis is necessary for infection. Switching to the avirulent Bvg- phase is triggered by the addition of negative modulators. Thus, BvgS might be active by default and inactivated by antagonists at specific stages of the bacterium’s life cycle. BvgS is a hybrid sensor-kinase harbouring several cytoplasmic domains that mediate a complex phospho-transfer cascade. It also contains two periplasmic ‘Venus flytrap’ (VFT) domains in tandem and a cytoplasmic PAS domain before the kinase. Ubiquitous in nature, VFT domains usually function along a clamshell model, with two lobes that enclose specific ligands between them. Conformational changes of VFTs between open and closed forms upon ligand binding are widely used for transport, or also for signalling such as in the ion-channel coupled Glu receptors of higher eukaryotes. BvgS is the prototype for a large family of multi-domain sensor-kinases that harbour periplasmic VFT perception domains, which represents a new paradigm of VFT receptors involved in signal transduction. The molecular mechanisms of signal perception and transduction by these proteins remain unexplored.
We have recently obtained the crystal structure of the entire periplasmic domain of BvgS, showing a novel architecture for VFT receptors. The periplasmic domain of BvgS is dimeric, with extensive interfaces between protomers. Our preliminary data have indicated that the conformations of the VFT domains determine BvgS activity and that the interprotomer interfaces are critical for function. This new structure and the tools that we have developed for the functional analysis of BvgS put us in a good position to decipher the molecular mechanisms of signalling by using an integrated approach that combines functional analyses, biochemistry, structural biology and in silico modelling. We will identify critical regions of the VFT domains that determine the active conformation of BvgS and the mechanisms by which negative signals are perceived and transmitted, using site-directed mutagenesis and a functional assay in bacterio. The conformation and dynamics of BvgS and its mutants will be investigated by crystallography and by modelling, using molecular dynamics simulations and coarse-grained approaches. Simulations will also be performed of the VFT domains with the transmembrane segment in a lipid environment. This will be combined with biochemical approaches to explore the topology and dynamics of the segment linking the periplasmic and cytoplasmic moieties, and of the linker between the PAS and kinase domains. The function and topology of the PAS domain will be characterized. Our program will enable us to propose a model that describes the molecular mechanisms of signal transduction in this family of VFT sensor-kinases. This will enlarge our knowledge on two-component-mediated signal transduction in bacteria, as well as lay the bases for new avenues of targeted therapeutic intervention.

Project coordination

Françoise JACOB-DUBUISSON (Institut Pasteur de Lille-Center for Infection and Immunity of Lille)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


IPL-CIIL Institut Pasteur de Lille-Center for Infection and Immunity of Lille
CNRS Institut de Recherche Interdisciplinaire

Help of the ANR 320,000 euros
Beginning and duration of the scientific project: December 2013 - 42 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter