ExlA Pore-Forming Toxin: Localization at the Bacterial Membrane, Regulation of Secretion by Cyclic-Di-GMP, and Detection In Vivo.

ExlA is a highly virulent pore-forming toxin that has been recently discovered in outlier strains from Pseudomonas aeruginosa. ExlA is part of a two-partner secretion system, in which ExlA is the secreted passenger protein and ExlB the transporter embedded in the bacterial outer membrane. In previous work, we observed that ExlA toxicity in a host cell was contact-dependent. Here, we show that ExlA accumulates at specific points of the outer membrane, is likely entrapped within ExlB pore, and is pointing outside. We further demonstrate that ExlA is maintained at the membrane in conditions where the intracellular content of second messenger cyclic-di-GMP is high; lowering c-di-GMP levels enhances ExlB-dependent ExlA secretion. In addition, we set up an ELISA to detect ExlA, and we show that ExlA is poorly secreted in liquid culture, while it is highly detectable in broncho-alveolar lavage fluids of mice infected with an exlA+ strain. We conclude that ExlA translocation is halted at mid-length in the outer membrane and its secretion is regulated by c-di-GMP. In addition, we developed an immunological test able to quantify ExlA in biological samples.

Exolysin (ExlA) from Pseudomonas aeruginosa Punctures Holes into Target Membranes Using a Molten Globule Domain.

Bacteria employ several mechanisms, and most notably secretion systems, to translocate effectors from the cytoplasm to the extracellular environment or the cell surface. Pseudomonas aeruginosa widely employs secretion machineries such as the Type III Secretion System to support virulence and cytotoxicity. However, recently identified P. aeruginosa strains that do not express the Type III Secretion System have been shown to express ExlA, an exolysin translocated through a two-partner secretion system, and are the causative agents of severe lung hemorrhage. Sequence predictions of ExlA indicate filamentous hemagglutinin (FHA-2) domains as the prevalent features, followed by a C-terminal domain with no known homologs. In this work, we have addressed the mechanism employed by ExlA to target membrane bilayers by using NMR, small-angle X-ray scattering, atomic force microscopy, and cellular infection techniques. We show that the C-terminal domain of ExlA displays a "molten globule-like" fold that punctures small holes into membranes composed of negatively charged lipids, while other domains could play a lesser role in target recognition. In addition, epithelial cells infected with P. aeruginosa strains expressing different ExlA variants allow localization of the toxin to lipid rafts. ExlA homologs have been identified in numerous bacterial strains, indicating that lipid bilayer destruction is an effective strategy employed by bacteria to establish interactions with multiple hosts.

Exolysin Shapes the Virulence of Pseudomonas aeruginosa Clonal Outliers.

Bacterial toxins are important weapons of toxicogenic pathogens. Depending on their origin, structure and targets, they show diverse mechanisms of action and effects on eukaryotic cells. Exolysin is a secreted 170 kDa pore-forming toxin employed by clonal outliers of Pseudomonas aeruginosa providing to some strains a hyper-virulent behaviour. This group of strains lacks the major virulence factor used by classical strains, the Type III secretion system. Here, we review the structural features of the toxin, the mechanism of its secretion and the effects of the pore formation on eukaryotic cells.

Response of CnrX from Cupriavidus metallidurans CH34 to nickel binding.

Resistance to high concentration of nickel ions is mediated in Cupriavidus metallidurans by the CnrCBA transenvelope efflux complex. Expression of the cnrCBA genes is regulated by the transmembrane signal transduction complex CnrYXH. Together, the metal sensor CnrX and the transmembrane antisigma factor CnrY control the availability of the extracytoplasmic function sigma factor CnrH. Release of CnrH from sequestration by CnrY at the cytoplasmic side of the membrane depends essentially on the binding of the agonist metal ion Ni(ii) to the periplasmic metal sensor domain of CnrX. CnrH availability leads to transcription initiation at the promoters cnrYp and cnrCp and to the expression of the genes in the cnrYXHCBA nickel resistance determinant. The first steps of signal propagation by CnrX rely on subtle metal-dependent allosteric modifications. To study the nickel-mediated triggering process by CnrX, we have altered selected residues, F66, M123, and Y135, and explored the physiological consequences of these changes with respect to metal resistance, expression of a cnrCBA-lacZ reporter fusion and protein production. M123C- and Y135F-CnrXs have been further characterized in vitro by metal affinity measurements and crystallographic structure analysis. Atomic-resolution structures of metal-bound M123C- and Y135F-CnrXs showed that Ni(ii) binds two of the three canonical conformations identified and that Ni(ii) sensing likely proceeds by conformation selection.

The X-ray structure of NccX from Cupriavidus metallidurans 31A illustrates potential dangers of detergent solubilization when generating and interpreting crystal structures of membrane proteins.

The x-ray structure of NccX, a type II transmembrane metal sensor, from Cupriavidus metallidurans 31A has been determined at a resolution of 3.12 Å. This was achieved after solubilization by dodecylphosphocholine and purification in the presence of the detergent. NccX crystal structure did not match the model based on the extensively characterized periplasmic domain of its closest homologue CnrX. Instead, the periplasmic domains of NccX appeared collapsed against the hydrophobic transmembrane segments, leading to an aberrant topology incompatible with membrane insertion. This was explained by a detergent-induced redistribution of the hydrophobic interactions among the transmembrane helices and a pair of hydrophobic patches keeping the periplasmic domains together in the native dimer. Molecular dynamics simulations performed with the full-length protein or with the transmembrane segments were used along with in vivo homodimerization assays (TOXCAT) to evaluate the determinants of the interactions between NccX protomers. Taken as a whole, computational and experimental results are in agreement with the structural model of CnrX where a cradle-shaped periplasmic metal sensor domain is anchored into the inner membrane by two N-terminal helices. In addition, they show that the main determinant of NccX dimerization is the periplasmic soluble domain and that the interaction between transmembrane segments is highly dynamic. The present work introduces a new crystal structure for a transmembrane protein and, in line with previous studies, substantiates the use of complementary theoretical and in vivo investigations to rationalize a three-dimensional structure obtained in non-native conditions.

The crystal structure of the anti-σ factor CnrY in complex with the σ factor CnrH shows a new structural class of anti-σ factors targeting extracytoplasmic function σ factors.

Gene expression in bacteria is regulated at the level of transcription initiation, a process driven by σ factors. The regulation of σ factor activity proceeds from the regulation of their cytoplasmic availability, which relies on specific inhibitory proteins called anti-σ factors. With anti-σ factors regulating their availability according to diverse cues, extracytoplasmic function σ factors (σ(ECF)) form a major signal transduction system in bacteria. Here, structure:function relationships have been characterized in an emerging class of minimal-size transmembrane anti-σ factors, using CnrY from Cupriavidus metallidurans CH34 as a model. This study reports the 1.75-Å-resolution structure of CnrY cytosolic domain in complex with CnrH, its cognate σ(ECF), and identifies a small hydrophobic knob in CnrY as the major determinant of this interaction in vivo. Unsuspected structural similarity with the molecular switch regulating the general stress response in α-proteobacteria unravels a new class of anti-σ factors targeting σ(ECF). Members of this class carry out their function via a 30-residue stretch that displays helical propensity but no canonical structure on its own.

Metal sensing and signal transduction by CnrX from Cupriavidus metallidurans CH34: role of the only methionine assessed by a functional, spectroscopic, and theoretical study.

When CnrX, the periplasmic sensor protein in the CnrYXH transmembrane signal transduction complex of Cupriavidus metallidurans CH34, binds the cognate metal ions Ni(II) or Co(II), the ECF-type sigma factor CnrH is made available in the cytoplasm for the RNA-polymerase to initiate transcription at the cnrYp and cnrCp promoters. Ni(II) or Co(II) are sensed by a metal-binding site with a N3O2S coordination sphere with octahedral geometry, where S stands for the thioether sulfur of the only methionine (Met123) residue of CnrX. The M123A-CnrX derivative has dramatically reduced signal propagation in response to metal sensing while the X-ray structure of Ni-bound M123A-CnrXs showed that the metal-binding site was not affected by the mutation. Ni(II) remained six-coordinate in M123A-CnrXs, with a water molecule replacing the sulfur as the sixth ligand. H32A-CnrXs, the soluble model of the wild-type membrane-anchored CnrX, was compared to the double mutants H32A-M123A-CnrXs and H32A-M123C-CnrXs to spectroscopically evaluate the role of this unique ligand in the binding site of Ni or Co. The Co- and Ni-bound forms of the protein display unusually blue-shifted visible spectra. TD-DFT calculations using structure-based models allowed identification and assignment of the electronic transitions of Co-bound form of the protein and its M123A derivative. Among them, the signature of the S-Co transition is distinguishable in the shoulder at 530 nm. In vitro affinity measurements point out the crucial role of Met123 in the selectivity for Ni or Co, and in vivo data support the conclusion that Met123 is a trigger of the signal transduction.

Spectroscopic characterization of the metal-binding sites in the periplasmic metal-sensor domain of CnrX from Cupriavidus metallidurans CH34.

CnrX, the dimeric metal sensor of the three-protein transmembrane signal transduction complex CnrYXH of Cupriavidus metallidurans CH34, contains one metal-binding site per monomer. Both Ni and Co elicit a biological response and bind the protein in a 3N2O1S coordination sphere with a nearly identical octahedral geometry as shown by the X-ray structure of CnrXs, the soluble domain of CnrX. However, in solution CnrXs is titrated by 4 Co-equiv and exhibits an unexpected intense band at 384 nm that was detected neither by single-crystal spectroscopy nor under anaerobiosis. The data from a combination of spectroscopic techniques (spectrophotometry, electron paramagnetic resonance, X-ray absorption spectroscopy) showed that two sites correspond to those identified by crystallography. The two extra binding sites accommodate Co(II) in an octahedral geometry in the absence of oxygen and are occupied in air by a mixture of low-spin Co(II) as well as EPR-silent Co(III). These extra sites, located at the N-terminus of the protein, are believed to participate to the formation of peroxo-bridged dimers. Accordingly, we hypothesize that the intense band at 384 nm relies on the formation of a binuclear µ-peroxo Co(III) complex. These metal binding sites are not physiologically relevant since they are not detected in full-length NccX, the closest homologue of CnrX. X-ray absorption spectroscopy demonstrates that NccX stabilizes Co(II) in two-binding sites similar to those characterized by crystallography in its soluble counterpart. Nevertheless, the original spectroscopic properties of the extra Co-binding sites are of interest because they are susceptible to be detected in other Co-bound proteins.

Structural basis for metal sensing by CnrX.

CnrX is the metal sensor and signal modulator of the three-protein transmembrane signal transduction complex CnrYXH of Cupriavidus metallidurans CH34 that is involved in the setup of cobalt and nickel resistance. We have determined the atomic structure of the soluble domain of CnrX in its Ni-bound, Co-bound, or Zn-bound form. Ni and Co ions elicit a biological response, while the Zn-bound form is inactive. The structures presented here reveal the topology of intraprotomer and interprotomer interactions and the ability of metal-binding sites to fine-tune the packing of CnrX dimer as a function of the bound metal. These data suggest an allosteric mechanism to explain how the complex is switched on and how the signal is modulated by Ni or Co binding. These results provide clues to propose a model for signal propagation through the membrane in the complex.

X-ray structure of the metal-sensor CnrX in both the apo- and copper-bound forms.

Both the X-ray structures of the apo- and the copper-bound forms of the metal-sensor domain (residues 31-148) of CnrX from Cupriavidus metallidurans CH34 were obtained at 1.74A resolution from a selenomethionine derivative. This four-helix hooked-hairpin is the first structure of a metal-sensor in an ECF-type signaling pathway. The copper ion is bound in a type 2-like center with a 3N1O coordination in the equatorial plane and shows an unprecedented remote fifth axial ligand with Met93 contributing a weak S-Cu bond. The signal onset cannot be explained by conformational changes associated with CnrX metallation.

Idiosyncratic features in tRNAs participating in bacterial cell wall synthesis.

The FemX(Wv) aminoacyl transferase of Weissella viridescens initiates the synthesis of the side chain of peptidoglycan precursors by transferring l-Ala from Ala-tRNA(Ala) to UDP-MurNAc-pentadepsipeptide. FemX(Wv) is an attractive target for the development of novel antibiotics, since the side chain is essential for the last cross-linking step of peptidoglycan synthesis. Here, we show that FemX(Wv) is highly specific for incorporation of l-Ala in vivo based on extensive analysis of the structure of peptidoglycan. Comparison of various natural and in vitro-transcribed tRNAs indicated that the specificity of FemX(Wv) depends mainly upon the sequence of the tRNA although additional specificity determinants may include post-transcriptional modifications and recognition of the esterified amino acid. Site-directed mutagenesis identified cytosines in the G1-C72 and G2-C71 base pairs of the acceptor stem as critical for FemX(Wv) activity in agreement with modeling of tRNA(Ala) in the catalytic cavity of the enzyme. In contrast, semi-synthesis of Ala-tRNA(Ala) harboring nucleotide substitutions in the G3-U70 wobble base pair showed that this main identity determinant of alanyl-tRNA synthetase is non-essential for FemX(Wv). The different modes of recognition of the acceptor stem indicate that specific inhibition of FemX(Wv) could be achieved by targeting the distal portion of tRNA(Ala) for the design of substrate analogues.

Deregulation of the SecYEG translocation channel upon removal of the plug domain.

Previous studies have shown that the SecY plug is displaced from the center of the SecYEG channel during polypeptide translocation. The structural and functional consequences of the deletion of the plug are now examined. Both in vivo and in vitro observations indicate that the plug domain is not essential to the function of the translocon. In fact, deletion of the plug confers to the cell and to the membranes a Prl-like phenotype: reduced proton-motive force dependence of translocation, increased membrane insertion of SecA, diminished requirement for functional leader peptide, and weakened SecYEG subunit association. Although the plug domain does not seem essential, locking the plug in the center of the channel inactivates the translocon. Thus, the SecY plug is important to regulate the activity of the channel and to confer specificity to the translocation reaction. We propose that the plug contributes to the gating mechanism of the channel by maintaining the structure of the SecYEG complex in a compact closed state.

Investigating the SecY plug movement at the SecYEG translocation channel.

Protein translocation occurs across the energy-conserving bacterial membrane at the SecYEG channel. The crystal structure of the channel has revealed a possible mechanism for gating and opening. This study evaluates the plug hypothesis using cysteine crosslink experiments in combination with various allelic forms of the Sec complex. The results demonstrate that the SecY plug domain moves away from the center of the channel toward SecE during polypeptide translocation, and further show that the translocation-enhancing prlA3 mutation and SecG subunit change the properties of channel gating. Locking the plug in the open state preactivates the Sec complex, and a super-active translocase can be created when combined with the prlA4 mutation located in the pore of the channel. Dimerization of the Sec complex, which is essential for translocase activity, relocates the plug toward the open position. We propose that oligomerization may result in SecYEG cooperative interactions important to prime the translocon function.

Structure-based site-directed mutagenesis of the UDP-MurNAc-pentapeptide-binding cavity of the FemX alanyl transferase from Weissella viridescens.

Weissella viridescens FemX (FemX(Wv)) belongs to the Fem family of nonribosomal peptidyl transferases that use aminoacyl-tRNA as the amino acid donor to synthesize the peptide cross-bridge found in the peptidoglycan of many species of pathogenic gram-positive bacteria. We have recently solved the crystal structure of FemX(Wv) in complex with the peptidoglycan precursor UDP-MurNAc-pentapeptide and report here the site-directed mutagenesis of nine residues located in the binding cavity for this substrate. Two substitutions, Lys36Met and Arg211Met, depressed FemX(Wv) transferase activity below detectable levels without affecting protein folding. Analogues of UDP-MurNAc-pentapeptide lacking the phosphate groups or the C-terminal D-alanyl residues were not substrates of the enzyme. These results indicate that Lys36 and Arg211 participate in a complex hydrogen bond network that connects the C-terminal D-Ala residues to the phosphate groups of UDP-MurNAc-pentapeptide and constrains the substrate in a conformation that is essential for transferase activity.

Crystal structures of Weissella viridescens FemX and its complex with UDP-MurNAc-pentapeptide: insights into FemABX family substrates recognition.

Members of the FemABX protein family are novel therapeutic targets, as they are involved in the synthesis of the bacterial cell wall. They catalyze the addition of amino acid(s) on the peptidoglycan precursor using aminoacylated tRNA as a substrate. We report here the high-resolution structure of Weissella viridescens L-alanine transferase FemX and its complex with the UDP-MurNAc-pentapeptide. This is the first structure example of a FemABX family member that does not possess a coiled-coil domain. FemX consists of two structurally equivalent domains, separated by a cleft containing the binding site of the UDP-MurNAc-pentapeptide and a long channel that traverses one of the two domains. Our structural studies bring new insights into the evolution of the FemABX and the related GNAT superfamilies, shed light on the recognition site of the aminoacylated tRNA in Fem proteins, and allowed manual docking of the acceptor end of the alanyl-tRNAAla.

Crystallization and preliminary X-ray analysis of Weissella viridescens FemX UDP-MurNAc-pentapeptide:l-alanine ligase.

Synthesis of the cell-wall peptidoglycan of firmicutes involves a unique family of peptide-bond-forming enzymes that use amino-acyl-tRNAs as substrates and are referred to as Fem proteins as they are factors essential for methicillin resistance in Staphylococcus aureus. The FemX UDP-MurNAc-pentapeptide:l-alanine ligase of Weissella viridescens was overexpressed, purified and crystallized. Native data were collected to 1.7 A resolution. The crystals belong to space group P2(1), with unit-cell parameters a = 42.03, b = 99.92, c = 45.84 A, beta = 116.02 degrees. The asymmetric unit contains one molecule. A selenium-derivative data set has been collected to 2.1 A resolution at the peak wavelength of the selenium absorption edge. Six strong selenium positions were visible in the anomalous Patterson map. Three additional weaker Se atoms have been identified by anomalous Fourier synthesis.

Differential stability and fusion activity of Lyssavirus glycoprotein trimers.

The oligomeric structure and the fusion activity of lyssavirus glycoprotein (G) was studied by comparing G from Mokola virus (GMok) and rabies virus (PV strain) (GPV), which are highly divergent lyssaviruses. G expressed at the surface of BSR cells upon either plasmid transfection or virus infection are shown to be mainly trimeric after cross-linking experiments. However, solubilization by a detergent (CHAPS) and analysis in sucrose sedimentation gradient evidenced that GMok trimer is less stable than GPV trimer. A chimeric glycoprotein (G Mok-PV) associating the N-terminal half of GMok to the C-terminal half part of GPV formed trimers with an intermediate stability, indicating that the G C-terminal domain is essential in trimer stability. A cell to cell fusion assay revealed that GMok (and not G Mok-PV) was able to induce fusion at a higher pH (0.5 pH unit) than GPV. Such differences in the oligomeric structure stability and in the fusion activity of lyssavirus glycoproteins may partly account for the previously reported differences of their immunogenic and pathogenic properties.

Rabies virus glycoprotein can fold in two alternative, antigenically distinct conformations depending on membrane-anchor type.

Rabies virus glycoprotein (G) is a trimeric type I transmembrane glycoprotein that mediates both receptor recognition and low pH-induced membrane fusion. We have previously demonstrated that a soluble form of the ectodomain of G (G(1-439)), although secreted, is folded in an alternative conformation, which is monomeric and antigenically distinct from the native state of the complete, membrane-anchored glycoprotein. This has raised questions concerning the role of the transmembrane domain (TMD) in the correct native folding of the ectodomain. Here, we show that an ectodomain anchored in the membrane by a glycophosphatidylinositol is also folded in an alternative conformation, whereas replacement of the TMD of G by other peptide TMDs results in correct antigenicity of G. However, mutants with an insertion of a hydrophilic linker between the ectodomain and the TMD also fold in an alternative conformation. The influence of the membrane-anchor type on G ectodomain trimerization and folding is discussed.