Type III secretion systems of Gram-negative bacteria form injection devices that deliver effector proteins into eukaryotic cells during infection. They span both bacterial membranes and the extracellular space to connect with the host cell plasma membrane. Their extracellular portion is a needle-like, hollow tube that serves as a secretion conduit for effector proteins. The needle of Shigella flexneri is approximately 50-nm long and 7-nm thick and is made by the helical assembly of one protein, MxiH. We provide a 7-Å resolution 3D image reconstruction of the Shigella needle by electron cryomicroscopy, which resolves α-helices and a β-hairpin that has never been observed in the crystal and solution structures of needle proteins, including MxiH. An atomic model of the needle based on the 3D-density map, in comparison with that of the bacterial-flagellar filament, provides insights into how such a thin tubular structure is stably assembled by intricate intermolecular interactions. The map also illuminates how the needle-length control protein functions as a ruler within the central channel during export of MxiH for assembly at the distal end of the needle, and how the secretion-activation signal may be transduced through a conformational change of the needle upon host-cell contact.helical image analysis | Shigella pathogenesis | bacterial protein secretion

Type III secretion systems are found in many Gram-negative bacteria. They are activated by contact with eukaryotic cells and inject virulence proteins inside them. Host cell detection requires a protein complex located at the tip of the device's external injection needle. The Shigella tip complex (TC) is composed of IpaD, a hydrophilic protein, and IpaB, a hydrophobic protein, which later forms part of the injection pore in the host membrane. Here we used labelling and crosslinking methods to show that TCs from a ΔipaB strain contain five IpaD subunits while the TCs from wild-type can also contain one IpaB and four IpaD subunits. Electron microscopy followed by single particle and helical image analysis was used to reconstruct three-dimensional images of TCs at ∼20 Å resolution. Docking of an IpaD crystal structure, constrained by the crosslinks observed, reveals that TC organisation is different from that of all previously proposed models. Our findings suggest new mechanisms for TC assembly and function. The TC is the only site within these secretion systems targeted by disease-protecting antibodies. By suggesting how these act, our work will allow improvement of prophylactic and therapeutic strategies.

MxiG is a single-pass membrane protein that oligomerizes within the inner membrane ring of the Shigella flexneri type III secretion system (T3SS). The MxiG N-terminal domain (MxiG-N) is the predominant cytoplasmic structure; however, its role in T3SS assembly and secretion is largely uncharacterized. We have determined the solution structure of MxiG-N residues 6–112 (MxiG-N(6–112)), representing the first published structure of this T3SS domain. The structure shows strong structural homology to forkhead-associated (FHA) domains. Canonically, these cell-signaling modules bind phosphothreonine (Thr(P)) via highly conserved residues. However, the putative phosphate-binding pocket of MxiG-N(6–112) does not align with other FHA domain structures or interact with Thr(P). Furthermore, mutagenesis of potential phosphate-binding residues has no effect on S. flexneri T3SS assembly and function. Therefore, MxiG-N has a novel function for an FHA domain. Positioning of MxiG-N(6–112) within the EM density of the S. flexneri needle complex gives insight into the ambiguous stoichiometry of the T3SS, supporting models with 24 MxiG subunits in the inner membrane ring.

Difficulties associated with using X-ray crystallography for structural studies of large macromolecular complexes have made single particle cryo-electron microscopy (cryoEM) a key technique in structural biology. The efficient application of the single particle cryoEM approach requires the sample to be vitrified within the holes of carbon films, with particles well dispersed throughout the ice and adopting multiple orientations. To achieve this, the carbon support film is first hydrophilised by glow discharge, which allows the sample to spread over the film. Unfortunately, for transmembrane complexes especially, this procedure can result in severe sample adsorption to the carbon support film, reducing the number of particles dispersed in the ice. This problem is rate-limiting in the single particle cryoEM approach and has hindered its widespread application to hydrophobic complexes. We describe a novel grid preparation technique that allows for good particle dispersion in the ice and minimal hydrophobic particle adhesion to the support film. This is achieved by hydrophilisation of the carbon support film by the use of selected detergents that interact with the support so as to achieve a hydrophilic and neutral or selectively charged surface.

We have performed open cell transmission electron microscopy experiments through pure water vapor in the saturation pressure regime (>0.6 kPa), in a modern microscope capable of sub-Å resolution. We have systematically studied achievable pressure levels, stability and gas purity, effective thickness of the water vapor column and associated electron scattering processes, and the effect of gas pressure on electron optical resolution and image contrast. For example, for 1.3 kPa pure water vapor and 300kV electrons, we report pressure stability of ± 20 Pa over tens of minutes, effective thickness of 0.57 inelastic mean free paths, lattice resolution of 0.14 nm on a reference Au specimen, and no significant degradation in contrast or stability of a biological specimen (M13 virus, with 6 nm body diameter). We have also done some brief experiments to confirm feasibility of loading specimens into an in situ water vapor ambient without exposure to intermediate desiccating conditions. Finally, we have also checked if water experiments had any discernible impact on the microscope performance, and report pertinent vacuum and electron optical data, for reference purposes.

No abstract

No abstract

No abstract