Assembly of the Inner Rod Determines Needle Length in the Type III Secretion Injectisome

Thomas C. Marlovits, Tomoko Kubori, María Lara-Tejero, Dennis Thomas, Vinzenz M. Unger, and Jorge E. Galán

Supplementary Methods

Needle complex sample preparation and biochemical analysis

Needle complex sample preparation for electron microscopy or biochemical analysis was carried out as previously described1 with some modifications. Briefly, overnight cultures of the S. typhimurium grown in LB broth supplemented with 0.3 M NaCl were harvested when they reached an OD600 of 1.0. The needle complex and base substructures were then separated by CsCl density-gradient centrifugation. Samples were prepared essentially as described previously (1) except that the pH was kept at 8.0 at all times. Briefly, harvested cells were resuspended in 150mM Tris-HCl (pH 8), 0.5M sucrose, 0.5mg/ml hen egg lysozyme, 5mM EDTA, and kept first on ice for 45min and then at 37°C for 15min. Cells were lysed with 0.3% lauryldimethylamine oxide (LDAO) before adding 10mM MgCl2 and 500mM NaCl to the lysate. Cell debris was removed by low-speed centrifugation and bases and needle complexes were pelleted by high-speed centrifugation (Beckman, 45Ti rotor, 35krpm, 4 hs, 12°C). The pellet was resuspended in 0.5% LDAO in 10mM Tris-HCl (pH 8), 0.5M NaCl, 5mM EDTA and adjusted to a final concentration of 30% w/v of CsCl. Two-milliliter samples were centrifuged for 12 hours at 50k rpm in a Beckman TLS-55 rotor. Half-milliliter aliquots were combined with 2.5ml CsCl-free buffer and samples were pelleted in a TLA-110 rotor at 100krpm for 30min. The complexes were resuspended in 0.1ml 0.2% LDAO, 10mM Tris-HCl (pH 8), 0.5M NaCl and stored at 4°C. Western-blot analysis of needle complex preparations was carried out as previously described2.

Electron microscopy, image processing, and determination of rotational symmetry

Electron microscopy, image processing and the determination of the rotational symmetry were carried out as previously described1. Briefly, samples were placed onto glow-discharged carbon-coated Cu/Rh-grids and vitrified by flash-freezing in liquid ethane. During data collection, the specimens were held at –175°C, and low-dose micrographs were recorded on Kodak SO163 film using a Tecnai F20 field emission gun electron microscope operated at 200kV, 50,000-fold magnification, and underfocus values of 1.2 -2.9 µm. Micrographs were digitized on a Zeiss Scanner using a 7 µm stepsize, which corresponds to 1.4Å sampling at the level of the specimen. Individual particle projections were extracted, combined into a dataset, and subjected to 2x2 pixel averaging in IMAGIC-53 (final box size of 200x200 pixel). The contrast reversals imposed by the contrast transfer function (CTF) of the objective lens were corrected for each particle projection using the mean underfocus value of the respective micrographs as determined by the program CTFFIND34. Subsequent image processing was done in SPIDER5 as well as IMAGIC-5. Volumes were rendered using CHIMERA6. The rotational symmetry was determined using a multi-reference method to both classify the symmetry and determine the alignment from side views, as previously described (D. R. Thomas, N. R. Francis, and D. J. DeRosier, unpublished results). Briefly, a single subunit from a three dimensional map of the M-ring of the flagellar basal body, which is analogous to IR1, was used to create initial three dimensional models for hypothetical 18 through 28-mers. An appropriate number of subunits were placed into the IR1 of cylindrically averaged maps whose diameters were adjusted to fit the different number of subunits. To model IR2, an appropriate number of spheres were placed into IR2 in register with the subunits in IR1. A multi-reference dataset was created by projecting all the different symmetry models using a limited number of angles about the long axis of the complex, including out of plane tilt. This dataset was then utilized to sort the individual particles into several subpopulations with different and well-defined symmetries. Particles that migrated between different rotational symmetries were discarded and only those particle images whose symmetries remained stable during iterations of the alignment were used for the final reconstructions. To generate 3D volumes, individual particles were back-projected using the Euler angles assigned during projection matching. To calculate density difference maps between fully assembled needle complex, and the wild type and ∆invJ base structures, the 3D maps of these structures were normalized (10 and filtered to a resolution of 20 Å. The difference map was then generated by subtraction of the normalized ∆invJ or wild type base maps from the wild-type base or fully assembled needle complex, as appropriate, utilizing the IMAGIC-5 software package.

References for Supplementary Methods

1.Marlovits, T. C. et al. Structural insights into the assembly of the type III secretion needle complex. Science 306, 1040-2 (2004).

2.Sukhan, A., Kubori, T., Wilson, J. & Galán, J. E. Genetic analysis of assembly of the Salmonella enterica serovar Typhimurium type III secretion-associated needle complex. J Bacteriol 183, 1159-67 (2001).

3.van Heel, M., Harauz, G., Orlova, E., Schmidt, R. & Schatz, M. A new generation of the IMAGIC image processing system. J. Struct. Biol. 116, 17-24 (1996).

4.Mindell, J. A. & Grigorieff, N. Accurate determination of local defocus and specimen tilt in electron microscopy. J. Struct Biol. 142, 334-47 (2003).

5.Frank, J. et al. SPIDER and WEB: processing and visualization of images in 3D electron microscopy and related fields. J Struct Biol. 116, 190-9. (1996).

6.Pettersen, E. F. et al. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605-12 (2004).

1