SUPPLEMENTARY FIGURES. Structure of the cholera toxin secretion channel in its. closed state

Size: px

Start display at page:

Download "SUPPLEMENTARY FIGURES. Structure of the cholera toxin secretion channel in its. closed state"

Estella Ray
5 years ago
Views:

1 SUPPLEMENTARY FIGURES Structure of the cholera toxin secretion channel in its closed state Steve L. Reichow 1,3, Konstantin V. Korotkov 1,3, Wim G. J. Hol 1$ and Tamir Gonen 1,2$ 1, Department of Biochemistry and 2, Howard Hughes Medical Institute, University of Washington, Seattle WA 98195, USA. 3 These authors contributed equally to this work. $ To whom correspondence should be addressed T.G. (tgonen@u.washington.edu) or W.G.J.H. (wghol@u.washington.edu). 1

2 Supplementary Figure 1. Rotational symmetry analysis of VcGspD. (a) Montage of rotationally averaged top views of VcGspD. (upper left) Reference-free class average without applied symmetry identifies weak radial spoke densities. 12 spokes are clearly resolved with applied C12 symmetry as well as with lower common symmetries (e.g. C6, C4, C3 and C2). The spoke features become smeared when other symmetries are applied. (b) Rotational power spectra analysis of reference-free class average (upper left in (a)) using the program RotaStat 1 also identifies a strong correlation with C12 rotational symmetry. (c) Projection slices of VcGspD reconstruction revealed distinctive 12-fold features throughout the density map. Similar detailed features were not observed with other applied symmetries, such as C14. (d) Projection slice of periplasmic domain of VcGspD reconstructed with applied C12 (left) and C14 (right) symmetry. Together, the analysis strongly supports the conclusion that VcGspD assembles as a dodecameric channel. 2

3 Supplementary Figure 2. Fourier shell correlation (FSC) curve. The Fourier shell correlation (FSC) curve of the VcGspD final reconstruction suggested a resolution limit of 19 Å according to the 0.5 threshold criteria 2. 3

Supplementary Figure 3. Comparison of the VcGspD secretin to the cryoem reconstruction of the T2SS secretin PulD from Klebsiella oxytoca (KoGspD).

4 Supplementary Figure 3. Comparison of the VcGspD secretin to the cryoem reconstruction of the T2SS secretin PulD from Klebsiella oxytoca (KoGspD). (a) cryoem reconstruction of the T2SS secretin KoGspD (grey) and its proteolytically stable core domain (blue) reproduced with permission [ The American Society for Biochemistry and Molecular Biology] 3 The figure was resized for approximate comparison to the VcGspD cryoem reconstruction in (b). In both reconstructions the periplasmic gate and constriction are clearly observed. However, in the VcGspD reconstruction, the periplasmic vestibule is ~80 Å longer than in the KoGspD reconstruction and an extracellular cap or gate domain is also observed. (c) The additional periplasmic density allowed molecular modeling of the complete secretin periplasmic domain and identifies the constriction site as the conserved N3 domain that is cleaved by proteolysis (see also Supplementary Fig. 7). An asterisk indicates the structurally equivalent sites corresponding to the site of proteolysis described in the main text. 4

5 5

6 6

7 Supplementary Figure 4. Sequence comparison of T2SS and T3SS secretins. Sequences of the T2SS secretins EcGspG (ETEC), VcGspD (V. cholerae), KoGspD (Klebsiella oxytoca) and the T3SS secretins EcEscC (EPEC), StInvG (Salmonella typhimurium), SfMxiD (Shigella flexneri) were aligned using ClustalW2 4. Signal peptide sequences were omitted from the alignment. Domain boundaries are displayed as colored bars below sequences (coloring scheme is the same as in Fig. 3 in the main text). The alignments of N0 and N1 subdomains were based on crystal structure superpositions of N-terminal fragments of EcGspD ( 5, PDB code 3EZJ) and EcEscC ( 6, PDB code 3GR5) (see Supplementary Fig. 5). The secondary structure elements corresponding to the crystal structures of EcGspD and EcEscC are colored in black and displayed above and below the sequences, respectively. N0 and N1 subdomains of EcEscC are connected by a short helix whereas the respective subdomains of EcGspD are connected by a ~20 residues linker that was disordered in the crystal structure. Note that the N2 subdomain is absent in the T3SS secretins. The predicted secondary structure elements for N3 subdomains of EcGspD and EcEscC are colored in green and displayed above and below the sequences, respectively. N3 subdomains of T2SS and T3SS secretins are characterized by a longer loop indicated by a dashed green line and 2 additional predicted β-strands compared to N1 and N2 subdomains. The same features are present in T4PB and filamentous phage assembly secretins (results not shown). Residues of KoGspD 3 and StInvG 6 that are susceptible to limited trypsinolysis are highlighted in green. Notably, the long N3 subdomain loop is the site of trypsin proteolysis in both T2SS and T3SS secretins. The sequence conservation is higher within N3 subdomain than in N0 and N1 subdomain. Although the N-terminal parts preceding the N3 subdomain loop are difficult to align between secretins from different secretion systems, the C-terminal parts display 54-65% pairwise sequence identity within T2SS family secretins, 35-54% within T3SS family secretins and 20-30% across different families. The sequence alignment was rendered using ESPript 7. 7

8 Supplementary Figure 5. Comparison of N0 and N1 subdomains of EcGspD and EcEscC. The N0 and N1 subdomains of EcGspD ( 5, PDB code 3EZJ) and EcEscC ( 6, PDB code 3GR5) were superimposed individually using EBI SSM server 8. (a) Stereo view of N0 subdomain of EcGspD (dark blue) superimposed with N0 subdomain of EcEscC (light orange). Root-mean-square deviation (rmsd) is 2.05 Å for 70 amino acids with 13% identity. Other domains are omitted for clarity. There is an additional N- terminal α-helix in the EcEscC structure (see also the alignment in Supplementary Fig. 4). (b) Stereo view of the N1 subdomain of EcGspD (light blue) superimposed onto the N1 subdomain of EcEscC (orange). The rmsd is 1.37 Å for 59 amino acids with 13% identity. N0 subdomains are also displayed to demonstrate that, although the structures of individual N0 and N1 subdomains of EcGspD and EcEscC are similar, the domaindomain contacts are very different. 8

Supplementary Figure 6. Side-by-side comparison of the outer membrane channels of three protein secretion systems: T2SS (left), T3SS (middle) and T4aSS (right). Upper row: outside shapes.

9 Supplementary Figure 6. Side-by-side comparison of the outer membrane channels of three protein secretion systems: T2SS (left), T3SS (middle) and T4aSS (right). Upper row: outside shapes. Lower row: Vertical cross sections. The cryoem reconstruction of T2SS VcGspD (left) appears structurally similar to the closed outer membrane secretin of the Salmonella typhimurium T3SS (middle) 9 (see also Fig. 3 in the main text). The outer membrane pore forming channel in the E. coli T4aSS secretion system (right) 10,11 is not related to the secretins from the T2SS and T3SS at the amino acid sequence level, and appears structurally unique from VcGspD. In addition to differences in shape, the periplasmic windows of the T4aSS proteins are absent in the T2SS and T3SS secretins, and the pronounced periplasmic gate in the T2SS and T3SS secretin dodecamers (b above, left and middle) are absent in the T4aSS 14-meric structure at the right. 9

10 Supplementary Figure 7. Construction of the VcGspD secretin N0-N3 periplasmic domain model. (a) The VcGspD cryoem density map contoured at 3.4σ was used for placing ring models of the secretin periplasmic domains N0-N3 generated using the SymmDock 12. The lowest energy ring models for the N0/N1 domain, the N2 domain and the N3 domains were initially placed at the indicated positions in a head-to-tail fashion with the N-termini pointed downward (toward the periplasm). These initial fits were then optimized by automated translational and rotational fitting procedures in Chimera 13. A single model for each domain was then selected based on a best fit to the experimental map. (b) A calculated volume from the final periplasmic model gives a cross-correlation value of 0.65 to the experimental VcGspD density map. 10

11 Supplementary References 1. Kocsis, E., Cerritelli, M.E., Trus, B.L., Cheng, N. & Steven, A.C. Improved methods for determination of rotational symmetries in macromolecules. Ultramicroscopy 60, (1995). 2. Stewart, P.L., Chiu, C.Y., Haley, D.A., Kong, L.B. & Schlessman, J.L. Review: resolution issues in single-particle reconstruction. J Struct Biol 128, (1999). 3. Chami, M. et al. Structural insights into the secretin PulD and its trypsin-resistant core. J Biol Chem 280, (2005). 4. Larkin, M.A. et al. Clustal W and Clustal X version 2.0. Bioinformatics 23, (2007). 5. Korotkov, K.V., Pardon, E., Steyaert, J. & Hol, W.G. Crystal structure of the N- terminal domain of the secretin GspD from ETEC determined with the assistance of a nanobody. Structure 17, (2009). 6. Spreter, T. et al. A conserved structural motif mediates formation of the periplasmic rings in the type III secretion system. Nat Struct Mol Biol 16, (2009). 7. Gouet, P., Robert, X. & Courcelle, E. ESPript/ENDscript: Extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res 31, (2003). 8. Krissinel, E. & Henrick, K. Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions. Acta Crystallogr., Sect. D: Biol. Crystallogr. 60, (2004). 9. Marlovits, T.C. et al. Structural insights into the assembly of the type III secretion needle complex. Science 306, (2004). 10. Fronzes, R. et al. Structure of a type IV secretion system core complex. Science 323, (2009). 11. Chandran, V. et al. Structure of the outer membrane complex of a type IV secretion system. Nature 462, (2009). 12. Schneidman-Duhovny, D., Inbar, Y., Nussinov, R. & Wolfson, H.J. PatchDock and SymmDock: servers for rigid and symmetric docking. Nucleic Acids Res 33, W363-7 (2005). 13. Pettersen, E.F. et al. UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem 25, (2004). 11

Supplementary Figure 1. Fourier shell correlation curves for sub-tomogram averages and

Supplementary Figure 1. Fourier shell correlation curves for sub-tomogram averages and comparisons to other published in situ T3SS structures. a, Resolution estimates after applying Fourier shell correlation