User:Fadel A. Samatey/FlgE III/Intrinsically Disordered Flagellar Rod Stretch

From Proteopedia

< User:Fadel A. Samatey(Redirected from Samatey/5)
Jump to: navigation, search

Interactive 3D Complement in Proteopedia


BMC Biology an online-only, open access journal: bmcbiol


An intrinsically disordered linker controlling the formation and the stability of the bacterial flagellar hook.
Clive S. Barker, Irina V. Meshcheryakova, Alla S. Kostyukova, Peter L. Freddolino, and Fadel A. Samatey.
BMC Biology 15:97 (October 27, 2017) (doi.org/10.1186/s12915-017-0438-7)

The interactive Molecular Tour below assumes that you are familiar with the journal article[1].

In addition to empirical structures, this report includes some theoretical models, which should be treated with caution.

Contents

Introduction

The flagellar hook is the universal joint that transmits torque from the motor, via the flagellar rod, to the helical flagellar filament which propels motile bacteria. Crystallographic structures of the hook monomer protein[2][3] and a cryo-EM structure of the rod monomer protein[4] have been reported.

Flagellar hook and rod proteins have a segment that appears likely to be intrinsically disordered before assembly of the hook or rod. For reasons explained in the publication[1], we shall refer to this segment as the Intrinsically Disordered Rod Stretch, ID-Rod-Stretch (in both the hook and the rod). "Rod" is included in the name because this segment is more conserved in the rod (FlgG)[1]. A less conserved homolog occurs in the hook (FlgE). This segment is missing in several earlier hook and rod protein monomer structures[2][5][3][4].

Previously, we reported a complete structure of the Campylobacter jejuni hook by cryo-EM at 3.5 Å[6]. This provided the first structure of the ID-Rod-Stretch. We showed that after assembly of the hook, the ID-Rod-Stretch (earlier termed the L-stretch) contacts several adjacent monomer proteins, assuming a stable folded structure.

Evidence presented in the current report[1] demonstrates that the ID-Rod-Stretch is crucial to stabilization and strength of the hook and rod. Because of its dual nature, we refer to the ID-Rod-Stretch as the yin and yang of the flagellum.

(Please scroll down to continue)

Image:Fig1 Samatey 400px.jpg

The bacterial flagellum consists of a filament, a universal joint ("hook"), and a motor ("basal body") containing a "rod" that transmits torque to the hook. This is Figure 1 in [1].

Molecular Tour

Contents

This Molecular Tour does not recap all results described in the publication[1]. It offers visualizations that complement some of those results, as well as some visualizations not described in the paper.

Campylobacter jejuni hook: contacts to ID-Rod-Stretch

The contacts that the ID-Rod-Stretch makes with neighboring chains in the hook assembly are crucial to the stability of the Campylobacter jejuni hook. As detailed in the accompanying paper[1], mutations in the hook ID-Rod-Stretch caused the flagellum to fail to form, and shortening of the hook ID-Rod-Stretch loop caused flagella to break off at their base. Here we will visualize these crucial stabilizing contacts.

Here is a single monomer protein chain of FlgE from Campylobacter jejuni (FlgE-Cj), colored by domain (restore initial scene):

D0 helices, D0 ID-Rod-Stretch (L-stretch), D1, D2, D3, and D4

This monomer is taken from the complete 3.5 Å cryo-EM structure of the hook, so its intrinsically disordered segment (ID-Rod-Stretch) is held in a stable conformation by contacts with adjacent monomers (not shown). Zoom in on the hook ID-Rod-Stretch.

  • ID-Rod-Stretch shown solid (spacefilling atoms).
  • Monomer (colored by domain) decorated with non-covalently bonded atoms in contacting monomers in the hook assembly. The contacting atoms are colored by chain; that is, each chain is assigned a different color. This reveals that the hook ID-Rod-Stretch is in contact[7] with four adjacent chains.
  • Here, the hook non-covalently bonded atoms are colored by element. There are about eleven hydrogen bonds[7]. There are two salt bridges[8], and one uncertain cation-pi interaction[9], both involving Arg58 near the tip of the ID-Rod-Stretch. A little more than a dozen carbon-carbon van der Waals interactions are also present. (Analysis of bond type was done with FirstGlance in Jmol[10].)
Element color key: C, O, N
  • Here the hook domain colors are lighter in order to contrast with the next scene: Contacting atoms are colored by domain. This reveals that nearly all of the hook contacts to the ID-Rod-Stretch are from Domain 1 in neighboring chains, with a few contacts from a neighboring hook ID-Rod-Stretch and Domain 0 helices. Contacts to hook Domain 0 helices are mostly from neighboring Domain 0 helices, plus 4 atoms from Domain 1 of neighboring chains. It follows that Domain 1 has many contacts from hook ID-Rod-Stretches of neighboring chains. It also has a few from Domain 1 of neighboring chains.

Salmonella enterica hook and rod: ID-Rod-Stretch models

Previous empirical determinations of the structures of the hook and rod of Salmonella enterica did not report structures for the ID-Rod-Stretch[2][3][4]. Therefore we used the cryo-EM structure of the Campylobacter jejuni hook (including its long hook ID-Rod-Stretch[6]) as a template to homology model the ''Salmonella'' ID-Rod-Stretch in the hook and rod (color key below, same as in Fig. 1d[6]). The lengths of these ID-Rod-Stretches vary:

Taxon, protein
ID-Rod-Stretch
Model
Clashscore[11]		 Sequence (Uniprot)
Length Residues[12]
Campylobacter hook FlgE 54 Thr30 - Ser833.5 Å cryo-EM[6]
15 (50%)		 Q0P7Q2 (FLGE_CAMJE)
Salmonella hook FlgE 37 Thr28 - Thr64Homology[13]
78 (0%)		 P0A1J1 (FLGE_SALTY)
Salmonella rod FlgG 55 Thr30 - Asn84Homology[14][15]
75 (0%)		 P0A1J3 (FLGG_SALTY)

In an earlier report[6], we showed how the Campylobacter hook ID-Rod-Stretch (earlier termed the "L-stretch") extends from the otherwise cylindrical assembly of D0, acting like a "finger" firmly anchoring the hook D1 domains (see L-Stretch Fingers).

Here, we found that shortening the long hook ID-Rod-Stretch of Campylobacter by removing the 21 amino acids at the distal tip of the hook ID-Rod-Stretch, Gln46-Ile66[12], caused much lower motility, and many of the flagella broke off at the hook[6].

Salmonella enterica hook: Lys32 is required for assembly

Lys32 (K32) is one of a handful of highly conserved residues in the Salmonella hook ID-Rod-Stretch. The mutant K32A had severely decreased motility. While wild type averaged 12 flagella/cell, less than 10% of K32A cells had any flagella, and then only 1-2/cell. K32A cells produced twice as much mutant FlgE as wild type cells. Individual point mutations at 3 other highly conserved residues (T28, G30, F38) and 2 semi-conserved residues (F31, M41) were created. Each of these five others single-residue mutants had near-normal motility. Thus, K32 is uniquely important among conserved residues in the hook ID-Rod-Stretch.

Here is the largely-empirical model of the Salmonella hook monomer, including a homology model of the hook ID-Rod-Stretch[13], described above.

D0 helices, D0 ID-Rod-Stretch, D1, D2

This theoretical model suggests why Lys32 is so important. Lys32 (K32) appears to anchor the ID-Rod-Stretch to Domain 1 by forming a salt bridge with highly conserved Glu361 (E361). Lys32 appears also to interact with Asp62 (D62). Here is an interactive version of Fig. 8A in [1], showing these interactions.

C O N Drag with your mouse to rotate.

Bear in mind that rotamer positions and distances between residues are very approximate in this homology model. However, similar interactions are seen in the 3.5 Å cryo-EM structure of the Campylobacter hook. There the homologous residues are Lys34 interacting with Gln81 and Glu809 (not shown).

In addition to the interactions shown in Fig. 8A, Phe60 (F60), Phe81 (F81), and Asn364 (N364) are close to Lys32. The aromatic sidechains might form cation-pi interactions with either Arg95 or Lys32. Residues homologous to Salmonella hook Lys32, Arg95, and Glu361 are highly conserved in bacterial FlgE proteins.

Salmonella enterica hook assembly

A model of the Salmonella enterica hook assembly was constructed using the above-described monomer plus helical symmetry parameters[16] previously determined by cryo-EM[17]. This is an approximate model. While the empirical 3.5 Å cryo-EM structure of the Campylobacter hook has a clashscore of 17 (40th percentile)[11], the Salmonella hook assembly model has a clashscore[11] of 64 (1st percentile).

Here is a hook assembly model that is split, about half Campylobacter and half Salmonella for direct comparison.

Salmonella ID-Rod-Stretch 
 Campylobacter ID-Rod-Stretch

Recall that Campylobacter FlgE has 5 domains (D0-D4), while FlgE of Salmonella is smaller, with only 3 domains (D0-D2). The cryo-EM Campylobacter FlgE has a longer hook ID-Rod-Stretch (57 residues) that curves into an L shape, strengthening the hook for the high speed of rotation of Campylobacter flagella. As mentioned above, mutation of Campylobacter FlgE that shortens the end of the hook ID-Rod-Loop by 21 residues causes most flagella to break off at the hook. Here is a view down the axis of the hook.

The hook ID-Rod-Stretch in Salmonella is shorter (37 residues) and, if the homology model is correct, straighter, without the pronounced L shape seen in Campylobacter. These differences may be seen more clearly when only the hook ID-Rod-Stretches are visible.

--- The End ---
Drag the structure with the mouse to rotate

See Also

Notes and References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Barker CS, Meshcheryakova IV, Kostyukova AS, Freddolino PL, Samatey FA. An intrinsically disordered linker controlling the formation and the stability of the bacterial flagellar hook. BMC Biol. 2017 Oct 27;15(1):97. doi: 10.1186/s12915-017-0438-7. PMID:29078764 doi:http://dx.doi.org/10.1186/s12915-017-0438-7
  2. ↑ 2.0 2.1 2.2 Samatey FA, Matsunami H, Imada K, Nagashima S, Shaikh TR, Thomas DR, Chen JZ, Derosier DJ, Kitao A, Namba K. Structure of the bacterial flagellar hook and implication for the molecular universal joint mechanism. Nature. 2004 Oct 28;431(7012):1062-8. PMID:15510139 doi:http://dx.doi.org/10.1038/nature02997
  3. ↑ 3.0 3.1 3.2 Yoon YH, Barker CS, Bulieris PV, Matsunami H, Samatey FA. Structural insights into bacterial flagellar hooks similarities and specificities. Sci Rep. 2016 Oct 19;6:35552. doi: 10.1038/srep35552. PMID:27759043 doi:http://dx.doi.org/10.1038/srep35552
  4. ↑ 4.0 4.1 4.2 Fujii T, Kato T, Hiraoka KD, Miyata T, Minamino T, Chevance FF, Hughes KT, Namba K. Identical folds used for distinct mechanical functions of the bacterial flagellar rod and hook. Nat Commun. 2017 Jan 25;8:14276. doi: 10.1038/ncomms14276. PMID:28120828 doi:http://dx.doi.org/10.1038/ncomms14276
  5. ↑ Shaikh TR, Thomas DR, Chen JZ, Samatey FA, Matsunami H, Imada K, Namba K, Derosier DJ. A partial atomic structure for the flagellar hook of Salmonella typhimurium. Proc Natl Acad Sci U S A. 2005 Jan 25;102(4):1023-8. Epub 2005 Jan 18. PMID:15657146
  6. ↑ 6.0 6.1 6.2 6.3 6.4 6.5 Matsunami H, Barker CS, Yoon YH, Wolf M, Samatey FA. Complete structure of the bacterial flagellar hook reveals extensive set of stabilizing interactions. Nat Commun. 2016 Nov 4;7:13425. doi: 10.1038/ncomms13425. PMID:27811912 doi:http://dx.doi.org/10.1038/ncomms13425
  7. ↑ 7.0 7.1 "Contacting" is defined as likely hydrogen bonds, plus likely apolar interactions. Likely hydrogen bonds: oxygens or nitrogens within 3.5 Å of oxygens or nitrogens in a neighboring monomer. Apolar interactions: carbons or sulfurs within 4.0 Å of carbons or sulfurs in a neighboring monomer.
  8. ↑ Salt bridges to the ID-Rod-Stretch: One ID-Rod-Stretch Arg58 sidechain nitrogen is 2.3 Å from the closest oxygen of Glu802, and 1.8 Å from the closest oxygen of Asp105. The Glu and Asp are in different chains.
  9. ↑ Cation-pi interaction with the ID-Rod-Stretch: Phe133 is close enough to Arg58 in the ID-Rod-Stretch that an interaction is possible. In the cryo-EM model, the Phe133 ring is in the wrong orientation to make this energetically significant, but such an interaction may occur given the uncertainties in that model.
  10. ↑ The model Image:Hkcj d10.pdb.gz contains one chain "d" from the 55-monomer cryo-EM hook model, plus all atoms within 10 Å of that chain. Chain "d" was verified to have 268 contacting atoms, the maximum number present for any of the 55 chains. This "chain d + 10 Å" model was uploaded into FirstGlance in Jmol. There, in the Tools tab, the tool Contacts and Non-Covalent Interactions was employed to analyze the types of non-covalent bonds to the ID-Rod-Stretch.
  11. ↑ 11.0 11.1 11.2 Clashscores were determined by Molprobity as the number of serious steric atomic overlaps (> 0.4 Å) per 1000 atoms, after addition and optimization of hydrogen atoms. Zero percentile is worst (highest clashscore) and 100th percentile is the best, based on 1,724 reference structures chosen in 2004.
  12. ↑ 12.0 12.1 Sequence numbers start with 1 at the second residue in the genomic sequence, since the initial Met is believed to be removed by N-terminal methionine aminopeptidase.
  13. ↑ 13.0 13.1 The model of the Salmonella enterica hook was constructed using the 7.1 Å cryo-EM monomer 3a69 plus the ID-Rod-Stretch homology modeled using the Campylobacter jejuni 3.5 Å cryo-EM monomer 5jxl as template. 3a69 was constructed by adjusting the interdomain angles of the 1.8 Å X-ray structure of domains 1 and 2, 1wlg, for optimal fit into the cryo-EM map.
  14. ↑ UniProt sequence P0A1J3 (FLGG_SALTY) was submitted to Swiss Model, with the template specified as 5jxl. Two homology models were returned, one spanning residues 2-128 (42.7% sequence identity), and one spanning residues 175-260 (41.9% sequence identity). When combined, these provided a model of D0, and a partial model of D1. The monomer model was completed using the 7.4 Å cryo-EM model 5wrh.
  15. ↑ Biasini M, Bienert S, Waterhouse A, Arnold K, Studer G, Schmidt T, Kiefer F, Gallo Cassarino T, Bertoni M, Bordoli L, Schwede T. SWISS-MODEL: modelling protein tertiary and quaternary structure using evolutionary information. Nucleic Acids Res. 2014 Jul;42(Web Server issue):W252-8. doi: 10.1093/nar/gku340., Epub 2014 Apr 29. PMID:24782522 doi:http://dx.doi.org/10.1093/nar/gku340
  16. ↑ Helical symmetry parameters for Salmonella enterica: hook rotation 64.79 degrees, rise 4.12 Å; rod rotation 64.72 degrees, rise 4.13 Å.
  17. ↑ Fujii T, Kato T, Namba K. Specific arrangement of alpha-helical coiled coils in the core domain of the bacterial flagellar hook for the universal joint function. Structure. 2009 Nov 11;17(11):1485-93. PMID:19913483 doi:10.1016/j.str.2009.08.017

Proteopedia Page Contributors and Editors (what is this?)

Eric Martz

Retrieved from "http://proteopedia.org/wiki/index.php/User:Fadel_A._Samatey/FlgE_III/Intrinsically_Disordered_Flagellar_Rod_Stretch"
Personal tools