Document Detail

Structural evidence that colicin A protein binds to a novel binding site of TolA protein in Escherichia coli periplasm.
Jump to Full Text
MedLine Citation:
PMID:  22493500     Owner:  NLM     Status:  MEDLINE    
The Tol assembly of proteins is an interacting network of proteins located in the Escherichia coli cell envelope that transduces energy and contributes to cell integrity. TolA is central to this network linking the inner and outer membranes by interactions with TolQ, TolR, TolB, and Pal. Group A colicins, such as ColA, parasitize the Tol network through interactions with TolA and/or TolB to facilitate translocation through the cell envelope to reach their cytotoxic site of action. We have determined the first structure of the C-terminal domain of TolA (TolAIII) bound to an N-terminal ColA polypeptide (TA(53-107)). The interface region of the TA(53-107)-TolAIII complex consists of polar contacts linking residues Arg-92 to Arg-96 of ColA with residues Leu-375-Pro-380 of TolA, which constitutes a β-strand addition commonly seen in more promiscuous protein-protein contacts. The interface region also includes three cation-π interactions (Tyr-58-Lys-368, Tyr-90-Lys-379, Phe-94-Lys-396), which have not been observed in any other colicin-Tol protein complex. Mutagenesis of the interface residues of ColA or TolA revealed that the effect on the interaction was cumulative; single mutations of either partner had no effect on ColA activity, whereas mutations of three or more residues significantly reduced ColA activity. Mutagenesis of the aromatic ring component of the cation-π interacting residues showed Tyr-58 of ColA to be essential for the stability of complex formation. TA(53-107) binds on the opposite side of TolAIII to that used by g3p, ColN, or TolB, illustrating the flexible nature of TolA as a periplasmic hub protein.
Chan Li; Ying Zhang; Mireille Vankemmelbeke; Oliver Hecht; Fadilah Sfouq Aleanizy; Colin Macdonald; Geoffrey R Moore; Richard James; Christopher N Penfold
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't     Date:  2012-04-09
Journal Detail:
Title:  The Journal of biological chemistry     Volume:  287     ISSN:  1083-351X     ISO Abbreviation:  J. Biol. Chem.     Publication Date:  2012 Jun 
Date Detail:
Created Date:  2012-06-04     Completed Date:  2012-08-20     Revised Date:  2013-06-26    
Medline Journal Info:
Nlm Unique ID:  2985121R     Medline TA:  J Biol Chem     Country:  United States    
Other Details:
Languages:  eng     Pagination:  19048-57     Citation Subset:  IM    
School of Molecular Medical Sciences, Centre for Biomolecular Sciences, University of Nottingham, University Park, Nottingham NG7 2RD, United Kingdom.
Data Bank Information
Bank Name/Acc. No.:
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Amino Acid Substitution
Binding Sites
Escherichia coli / genetics,  metabolism*
Escherichia coli Proteins / genetics,  metabolism*
Lipoproteins / genetics,  metabolism*
Mutation, Missense
Periplasm / genetics,  metabolism*
Protein Binding
Protein Structure, Secondary
Grant Support
//Biotechnology and Biological Sciences Research Council; //Wellcome Trust
Reg. No./Substance:
0/Cal protein, E coli; 0/Escherichia coli Proteins; 0/Lipoproteins; 0/tolA protein, E coli

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Full Text
Journal Information
Journal ID (nlm-ta): J Biol Chem
Journal ID (iso-abbrev): J. Biol. Chem
Journal ID (hwp): jbc
Journal ID (pmc): jbc
Journal ID (publisher-id): JBC
ISSN: 0021-9258
ISSN: 1083-351X
Publisher: American Society for Biochemistry and Molecular Biology, 9650 Rockville Pike, Bethesda, MD 20814, U.S.A.
Article Information
Download PDF
© 2012 by The American Society for Biochemistry and Molecular Biology, Inc.
Received Day: 18 Month: 1 Year: 2012
Revision Received Day: 22 Month: 3 Year: 2012
Print publication date: Day: 1 Month: 6 Year: 2012
Electronic publication date: Day: 9 Month: 4 Year: 2012
pmc-release publication date: Day: 9 Month: 4 Year: 2012
Volume: 287 Issue: 23
First Page: 19048 Last Page: 19057
ID: 3365938
PubMed Id: 22493500
Publisher Id: M112.342246
DOI: 10.1074/jbc.M112.342246

Structural Evidence That Colicin A Protein Binds to a Novel Binding Site of TolA Protein in Escherichia coli Periplasm*, Alternate Title:Binding Sites of E. coli TolA Hub Protein
Chan Li12
Ying Zhang1
Mireille Vankemmelbeke
Oliver Hecht§
Fadilah Sfouq Aleanizy
Colin Macdonald§
Geoffrey R. Moore§
Richard James
Christopher N. Penfold3
From the School of Molecular Medical Sciences, Centre for Biomolecular Sciences, University of Nottingham, University Park, Nottingham NG7 2RD and
the §Centre for Molecular and Structural Biochemistry, School of Chemical Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom
3 To whom correspondence should be addressed. Tel.: 44-1158467957; Fax: 44-115 8467951; E-mail:
1Both authors contributed equally to this work.
2Present address: School of Pharmacy, Centre for Biomolecular Sciences, University of Nottingham, Nottingham NG7 2RD, United Kingdom.


The Tol-Pal4 network of proteins in the periplasm of Gram-negative bacteria is intriguing as conservation of the sequence of genes of the tol-pal operon across a variety of bacterial orders (1) suggests important functional significance of the Tol proteins. Although their normal cellular function in Escherichia coli is still uncertain (1), they appear to play a role in maintaining the integrity of the cell envelope, transducing energy from the cytoplasmic membrane, and may form a dynamic subcomplex at constriction sites to promote the energy-dependent septal wall formation across invaginating peptidoglycan and inner membrane layers during cell division (24). The Tol-Pal system consists of five proteins, TolA, TolB, TolQ, TolR, and peptidoglycan-associated lipoprotein (Pal). TolA is a 44-kDa periplasmic protein organized into three domains separated by clusters of glycine residues. It is anchored to the cytoplasmic membrane via a single transmembrane region (TolAI)5 that is important for interactions with the TolQ and TolR proteins in the membrane (5), spans the periplasm via its extended central domain (TolAII), and binds to both TolB (6, 7) and Pal (8) via its C-terminal domain (TolAIII). TolQ and TolR are transmembrane proteins that are involved in the proton motif force-dependent activation of TolA (2), which shuttles energy from the inner to outer membrane through its association with Pal anchored to the outer membrane (9, 10). TolB is a periplasmic protein that is associated with the outer membrane via an interaction of its C-terminal β-propeller domain with Pal (11) and with TolA via its N-terminal domain (6).

ColA is a pore-forming colicin that forms pores in the cytoplasmic membrane of E. coli cells (12). The N-terminal translocation domain of ColA contains a binding region from residues 52–97 that interacts directly with TolAIII (13, 14). A TolA binding motif with a conserved tyrosine residue has been identified at two locations (57SYNT60 and 89PYGR92) within the TolA box region of ColA with good homology to a TolA binding motif in the C-terminal domain of Pal (15). TolA is also required for the entry of ColN (16) and for the import of filamentous bacteriophage DNA into E. coli cells (17) through an interaction with the gene 3 protein of bacteriophage fd (g3p), which is a minor coat protein located at the end of the phage capsid (18).

Protein-protein interactions are crucial for the translocation of colicin domains into target cells, but the temporal and spatial relationships between Tol proteins and colicin domains are only just starting to be elucidated (19, 20). X-ray structures of members of the Tol-Pal assembly in complex with partner proteins have provided valuable information in deciphering the sequence of events that occur when group A colicins parasitize a sensitive cell (2124). A solution structure of TolAIII (Protein Data Bank (PDB) ID: 1S62) (25) and an x-ray structure of a fusion formed from TolAIII and g3p (PDB ID: 1Tol) (26) have been reported, but despite its pivotal role in the Tol assembly of proteins and availability for binding to foreign polypeptides, a structure of TolAIII in complex with other Tol proteins or with colicin domains has not previously been reported. The present work aims to address the role of TolAIII in the uptake of ColA through determination of the x-ray crystal structures of E. coli TolAIII on its own and in complex with the TolA binding region of ColA.

Overexpression and Purification

E. coli B834(DE3) was used as an expression host for plasmids pYZ48 (residues 302–421 of TolA (TolAIII) cloned into pET21a via NdeI/XhoI restriction sites with an in-frame N-terminal His6 tag) or pYZ69 (27) to purify TolAIII and TA1–1076 containing a thrombin recognition site after residue 52, respectively. An overnight culture of 10 ml of Luria Bertani medium (LB) was used to inoculate 0.5 liter of 2×YT broth supplemented with ampicillin (Melford Laboratories) to a final concentration of 100 μg/ml. Cultures were shaken in 2-liter baffled flasks at 180 rpm at 37 °C until they reached an A600 of 0.6–0.8; protein expression was induced for a period of 4 h after adding isopropyl-β-D-thiogalactopyranoside to a final concentration of 1 mM. For selenomethionine-labeled protein expression, E. coli 834 (DE3) harboring pYZ48 or pYZ69 was used as the expression host. The pellet from a 5-ml overnight culture grown in LB supplemented with ampicillin and chloramphenicol (34 μg/ml) was resuspended in 5 ml of sterile 0.9% NaCl and incubated in a shaker at 37 °C for 30 min, and the cells were spun down and washed again with 5 ml of sterile 0.9% NaCl before the pellet was transferred into 100 ml of Selenomethionine medium complete (Molecular Dimensions Ltd.) containing L-methionine and grown overnight at 37 °C. The overnight culture was washed twice in 0.9% sterile NaCl as described previously, and the pellet was transferred into 500 ml of selenomethionine medium complete containing L-selenomethionine before protein induction was carried out for 4 h at 37 °C with 1 mM isopropyl-β-D-thiogalactopyranoside when the cell A600 reached 0.6–0.8.

Once resuspended in a buffer containing 50 mM Tris-HCl, pH 8.0, 300 mM NaCl, 10 mM imidazole (lysis buffer), the cell pellets were lysed by sonication, and the lysate was centrifuged at 27,000 × g for 30 min. The supernatant was filtered into nickel-agarose beads (nickel-nitrilotriacetic acid, Qiagen), which were previously equilibrated with lysis buffer and gently mixed by circular rotation in a cold room for 1 h. The beads were washed with 50 mM Tris-HCl, pH 8.0, 300 mM NaCl, 10–30 mM imidazole (wash buffer), and finally, the proteins were eluted with 50 mM Tris-HCl, pH 8.0, 300 mM NaCl, 250 mM imidazole (elution buffer).

A region of TA1–107 containing a TolA binding site (TA53–107) was separated from the TolB box by digestion of TA1–107 with thrombin at a thrombin recognition site engineered into TA1–107 after residue 52. The protein concentration was adjusted to 1 mg/ml in 20 mM Tris, pH 8.0, 100 mM NaCl containing 5 units/mg of thrombin (Haematologic Technologies Inc., Essex Junction, VT), and the mixture was left on a circulating platform rotor at room temperature overnight. The thrombin was then removed using ρ-aminobenzamidine-agarose beads (Sigma) according to the manufacturer's instructions, and TA53–107 was purified using metal chelate chromatography and nickel-agarose beads in His tag lysis buffer as described previously (27). TA53–107 was finally loaded to a Superdex 75 gel filtration column (GE Healthcare) to remove any undigested TA1–107. The purity of protein samples was checked by running 15–17% SDS-PAGE gels. Pure fractions of TA53–107 were combined, concentrated using 5-kDa molecular mass cut off concentrators (Vivascience), and dialyzed to 20 mM Tris, pH 8.0, 100 mM NaCl. Protein concentration was determined by using absorption at 280 nm and extinction coefficients of 15,470, 8480, and 6085 M−1cm−1 for TA1–107 with a thrombin cleavage site after residue 52, TA53–107, and TolAIII, respectively.

NMR Spectroscopy

Isotopically labeled TolAIII comprising amino acids 302–421 of the E. coli TolA protein (Swiss-Prot accession number P19934) and TA1–52 and TA53–107 containing the TolB box and TolA binding region of ColA, respectively, were prepared as described previously (27). The B1–13 peptide representing the N-terminal sequence of TolB lacking its signal peptide and comprising residues 21–33 of TolB (AEVRIVIDSGVDS) was purchased from United Peptide Corp. at a purity of 95%. NMR spectra were acquired at the indicated temperatures using a Bruker Avance III 800-MHz spectrometer equipped with a triple resonance probe and pulse sequences incorporated into the Bruker Topspin 2.1 software. Resonance assignments of 13C/15N-labeled TA53–107 in complex with unlabeled TolAIII (1.2 mM) were obtained from 1H-15N HSQC, CBCA(CO)NH, HNCACB, and HNCA spectra at 27 °C. NMR spectra were processed using NMRPipe (28) and analyzed in NMRView and CCPN-Analysis as described previously (29).

Formation and Methylation of TA53–107-TolAIII Complex

TA53–107 and TolAIII in 20 mM Tris, pH 8.0, and 100 mM NaCl were mixed in a 1.1:1 molar ratio and incubated at 4 °C overnight. The mixed sample was loaded onto a Superdex 75 column equilibrated with 20 mM Tris, pH 8.0, 100 mM NaCl and run at a flow rate of 1 ml/min on an FPLC system (ÄKTA). The elution fractions were checked by running on 17% SDS-PAGE gels. The methylation of the TolAIII-TA53–107 complex was carried out using a published protocol (30, 31). Briefly, both TA53–107 and TolAIII were exchanged into 50 mM Hepes, pH 7.5, 100 mM NaCl and were mixed in a 1.1:1 molar ratio. Then the complex was adjusted to 1 mg/ml or 0.5 mg/ml for methylation of the selenomethionine-labeled complex. Methylation was carried out by adding borane dimethylamine complex (Sigma), which is a reducing agent, followed by formaldehyde (Sigma) before an overnight incubation at 4 °C (30, 31). The reaction was quenched by adding Tris buffer, pH 8, to 20 mM, and then the product was concentrated and buffer-exchanged into 20 mM Tris, pH 8, 100 mM NaCl for gel filtration to remove the small reactants and excess TA53–107.

Crystallization, X-ray Diffraction, and Model Refinement

The methylated TA53–107-TolAIII complex was crystallized by using the sitting-drop vapor diffusion method at 20 °C with 2 μl of the complex at 35 mg/ml in 20 mM Tris-HCl, pH 8.0, 100 mM NaCl mixed in equal volumes with the reservoir solution containing 0.2 M K2HPO4, 2.2 M ammonium sulfate. The complex crystals grew after 3–4 weeks. Before being frozen in a nitrogen stream, crystals were soaked in 0.2 M KH2PO4, 2.26 M ammonium sulfate, and 1.1 M disodium malonate for cryoprotection. Diffraction data sets were collected at the European Synchrotron Radiation Facility (ESRF), Grenoble, France, using a MAR 225 detector at a microfocus beamline of ID23-2 (λ = 0.8726 Å) for native crystals and an ADSC Quantum Q315r detector at a beamline of ID14-4 (λ = 0.9795 Å) for selenomethionine-labeled crystals. A wavelength of 0.9795 Å was used for collecting the single anomalous dispersion data sets for selenomethionine-labeled crystals. After failure to get the phase from molecular replacement using the available TolAIII structures PDB ID 1Tol (26) and PDB ID 1S62 (25) as search models, selenomethionine-labeled crystals of this complex were used to obtain the phase.

All data were processed with XDS (32), converted from XDS format to mtz format with Pointless (33), and scaled in SCALA (34) of the CCP4 suite (35). The initial phase was determined by using Phaser-EP (36) with the coordinates of four anomalous atoms found by Solve (37). The phase was improved by running the Density Modification (DM) program (38) and extended from 2.8 Å of selenomethionine crystals to 2.65 Å of native crystals using Cad and Scaleit of the CCP4 suite. The initial model was built by using Buccaneer (39) with 100 cycles and COOT (40), and then the model was used as a search model for molecular replacement through the use of Phaser (36) against the native data, rebuilt in COOT, and refined using restrained and TLS refinement in REFMAC5 (41). The optimal number of TLS groups was determined using the TLSMD server (42). The final model has excellent geometry and no Ramachandran outliers. The data processing, phasing, and model refinement statistics are summarized in Table 1. Surface area accessibility calculations and cation-π interaction identifications were carried out using the programs AREAIMOL (43) and CaPTURE (44), respectively. All structural figures were prepared using PyMOL (67).

Colicin Activity Assay

Spot test assays of biological activity of ColA and ColA mutants were conducted as described previously (45).

PCR Mutagenesis

Point mutations of the ColA gene were generated using the QuikChange II site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions. All constructs were verified by DNA sequencing (Source BioScience, Nottingham, UK).

Surface Plasmon Resonance

Surface plasmon resonance assays were conducted as described previously (24). TolAIII was immobilized to the matrix of a newly docked CM5 sensor chip, pre-equilibrated in HBS-EP running buffer (10 mM Hepes, 150 mM NaCl, 3 mM EDTA, 0.005% Tween) (BIAcore AB) via amine coupling to a level of immobilization of 200 response units. Two-minute injections of 10 μM of the ColA and ColA mutant proteins were performed across the immobilized TolAIII on the CM5 chip at a flow rate of 30 μl/min. After each analyte injection, the sensor chip surface was regenerated using a 2-min pulse of 10 mM glycine, pH 1.8. Global analysis using BIAevaluation software 3.1 was used to align all the binding curves. To determine the binding affinity data, 2-min injections of a range of concentrations of TA1–107 or TA53–107 were performed using the same TolAIII chip. Global analysis using BIAevaluation software 3.1 was used to fit corrected surface plasmon resonance responses to the theoretical 1:1 Langmuir binding model.

Crystallization of TA53–107-TolAIII Complex

Initially, we were unable to obtain crystals of TolAIII in complex with a polypeptide of ColA expressing residues 1–107 (TA1–107). Disorder analysis of the sequence of TA1–107, using the RONN program (46), showed that the N terminus of ColA containing the TolB box (residues 11–20) is disordered (4749), whereas a region containing residues 52–97, predicted to contain the TolAIII binding region of ColA (50), was ordered (supplemental Fig. S1). The TA53–107 polypeptide produced from the thrombin cleavage of TA1–107 (see “Experimental Procedures”) and lacking the TolB box was able to form a stable complex with TolAIII with a binding affinity of 1.34 ± 0.008 μM (data not shown). However, only tiny protein crystals grew from crystallization screens of the TA53–107-TolAIII complex. High entropy residue analysis of TolAIII was performed by using a surface entropy reduction prediction server (51), which predicted five high entropy surface lysine residues in four clusters (supplemental Fig. S2). The reductive methylation of lysine residues has previously been reported to be a valuable rescue strategy for obtaining diffraction-quality crystals from proteins found previously difficult to crystallize (30). Crystals of a complex of selenomethionine-labeled TolAIII and TA53–107 proteins, which was methylated before crystallization, appeared in the orthorhombic space group P21212, with one TolAIII and one TA53–107 polypeptide in the asymmetric unit and a solvent content of 58%.

Crystal Structure of TA53–107-TolAIII Complex

The structure consists of residues 329–421 of TolAIII and residues 57–105 of TA53–107, 26 water molecules, and one Tris ion (Fig. 1). Only Lys-351 of TolAIII was methylated in the TA53–107-TolAIII complex. The structure was refined to a resolution of 2.65 Å with the Rfactor and Rfree being 24.4 and 27.3%, respectively (Table 1). The structure of TolAIII consists of three α-helices and three β-strands with a short helix between the first α-helix and the first β-strand. There are two small β-sheets in TA53–107 located in its C and N terminus, respectively, each composed of two β-strands connected by a hairpin. Thus the structure of TA53–107 is like a skipping rope with two small β-sheets at the C terminus and N terminus acting as the two handgrips of the rope. The N-terminal β-sheet forms four intramolecular hydrogen bonds, whereas the C-terminal β-sheet forms eight, all of them formed between the backbone atoms. The β-sheet at the C terminus of TA53–107 forms an intermolecular antiparallel β-sheet with the three β-strands of TolAIII. The remainder of TA53–107, mainly between the two small β-sheets, forms a hairpin and loops. Therefore, the complex forms a five β-strand sheet, with the α-helices of TolAIII on one side and the N-terminal small β-sheet, together with the hairpin and loops of TolA box, on the other side of the five β-strand sheet of the complex (Fig. 1). There are no similar structures to that of TA53–107 available in the PDB database from search results using the Dali server.

Binding Interface between TA53–107 and TolAIII

The binding interface between TA53–107 and TolAIII buries a total surface area of 1303 Å with TolAIII and TA53–107 contributing 683 and 620 Å, respectively. The buried surface area is in the range of 1200–2000 Å that has been observed for most heterodimeric protein-protein complexes (52, 53). The calculated shape complementarity index, (Sc) (54) is 0.72, which is higher than that of high affinity protein-protein complexes such as the colicin E3 DNase-Im3 and colicin E9 DNase-Im9 complexes, indicating a high level of complementarity between TolAIII and TA53–107 in the complex. The interactions between TolAIII and TA53–107 encompass mainly electrostatic interactions and hydrophobic/apolar contacts. There are six direct hydrogen bonds between residues 92–96 of TA53–107 and residues 375–380 of TolAIII with the two protruding side chains Arg-92 and Arg-96 from TA53–107 forming hydrogen bonds with Pro-380 and Leu-375 of TolAIII (Fig. 2, supplemental Table S1). A water molecule mediates three hydrogen bonds between Phe-94 of TA53–107 and Asp-377 of TolAIII. The crystal structure of the TA53–107-TolAIII complex indicates that the minimum size of the TolAIII binding region of ColA consists of residues 58–96, rather than the previously described residues 52–97 (50), which is still larger than the TolAIII binding region of colicin N (55).

Mutational Analysis of Interface Residues of TA53–107-TolAIII Complex

Alanine-scanning mutagenesis of the five interface residues of ColA (Arg-92–Arg-96) that interact with TolAIII was performed, and the biological activity of the purified mutant proteins was assessed using a colicin plate assay (Fig. 3). Alanine mutations of any one of the five interface residues of ColA had little effect on biological activity, as can be seen for the F94A mutant of ColA in YZ106 (Fig. 3). The introduction of multiple alanine mutations of the interface residues revealed an additive effect that was observed as a reduction in biological activity of the mutant ColA. The presence of both the F94A and the R96A mutations (in YZ108) resulted in a further small reduction of biological activity when compared with YZ106 (Fig. 3). The further addition of a T95A mutation (in YZ109) resulted in a major reduction of biological activity. A complete loss of biological activity was seen with YZ110, in which all five of the interface residues were mutated to alanine.

In parallel, the contribution of the five residues of TolAIII to the binding interface of the TolAIII-TA53–107 complex was also determined. Mutations were engineered into pTol, which expresses the TolA protein (56), and assayed for activity by complementation of the tolA mutant, E. coli JC7782. Individual alanine mutations of residues Leu-375, Leu-376, Asp-377, Ile-378, and Pro-380 of TolAIII were still able to complement the tolA mutant, resulting in the restoration of sensitivity to ColA, whereas a mutant containing alanine substitutions in all five residues that contribute to intermolecular contacts across the interface did not complement the tolA mutant (Table 2). Surprisingly, mutants of pTol containing alanine substitutions of Leu-376, Asp-377, and Pro-380 of TolAIII were still able to complement the E. coli tolA mutant, albeit at reduced levels (Table 2), despite the apparent loss of several important intermolecular polar contacts, particularly those contacts resulting from the side chain of Asp-377 (Fig. 2).

NMR and X-ray Structural Comparisons of TolAIII and TA53–107-TolAIII

The x-ray structure of free TolAIII was obtained by chance in the process of the attempted crystallization of the complex formed between TolAIII and a peptide representing the N-terminal region of TolB lacking its signal peptide (referred to here as B1–13). The complex was methylated, but the structure was solved with only TolAIII in the density without any extra density for B1–13. Mass spectrometry analysis of the solubilized crystals proved that no peptide was present in the crystallized sample. Therefore, the structure we have obtained is that of TolAIII on its own.

The free TolAIII structure was refined to a resolution of 2.15 Å with Rfactor and Rfree being 19.2 and 22.5%, respectively. The first 32 residues have no corresponding electron density and are omitted in the model. Only three lysine residues (Lys-345, Lys-351, and Lys-413) were identified as being methylated, with lysines 345 and 413 being dimethylated and Lys-351 being monomethylated (supplemental Fig. S3). In addition, there are two nitrate ions and 43 water molecules in the model. The free x-ray structure has a marked similarity to the NMR structure of TolAIII (PDB ID: 1S62) (25) with three α-helices and three β-strands (supplemental Fig. S4). However, subtle differences exist that prevented our determination of the x-ray structure of free TolAIII by molecular replacement using PDB IDs 1S62 and 1Tol as search models, suggesting that there are some conformational differences between our free x-ray TolAIII structure and both the NMR solution structure of TolAIII and the x-ray structure of TolAIII fused with g3p. Our TolAIII structure was solved by molecular replacement using the structure of TolAIII from the TA53–107-TolAIII complex as a search model. This is consistent with comparisons of the root mean square distance (r.m.s.d.) of the Cα atoms of the free TolAIII x-ray structure with other TolAIII structures; the r.m.s.d. values are 2.08, 1.549, 1.673, and 0.537 when compared with PDB IDs 1LRO, 1S62, 1Tol, and 3QDR (TolAIII in complex with TA53–107), respectively. Helix II runs from residues 385–370 in PDB ID 1S62 and from residues 390–370 in PDB ID 3QDP, whereas Helix III is the same in both structures and includes residues 410–413 (supplemental Fig. S4). Helix I is shorter in PDB ID 3QDP and might be a consequence of the methylation of Lys-345 (supplemental Fig. S3). However, in silico mutagenesis of Lys-345 to Ala did not produce any extra helicity to Helix I. β-Strands 1 and 2 in PDB ID 1S62 are linked by a short hairpin and are longer than their counterparts in PDB ID 3QDP, whereas three shorter β-strands in PDB ID 3QDP form an antiparallel β-sheet (supplemental Fig. S4). The partially ordered segment that connects helix I and β-strand 1 shows differences between the two structures with Cα atoms r.m.s.d. of 1.83 Å.

Cation-π Interactions in TA53–107-TolAIII Complex

Geometrical analysis for cation-π interactions in the interface of the complex shows that three putative cation-aromatic contacts have suitable geometry. The NZ atoms of Lys-368, Lys-379, and Lys-396 of TolAIII, none of which are methylated, align above the aromatic rings of Tyr-58, Tyr-90, and Phe-94 of TA53–107, respectively (Fig. 4, supplemental Table S1). The cation-π interactions were analyzed by using the CaPTURE server (44), which predicted that the three pairs of Tyr-58–Lys-368, Tyr-90–Lys-379, and Phe-94–Lys-396 were energetically significant with electrostatic energies of −3.62, −3.32, and −2.78 kcal/mol, respectively. These three cation-π interactions are located on the two sides of the β-sheet in the complex. The hydrophobic/apolar contacts mainly involve the side chains of Met-87 and Phe-94 from TA53–107 and Leu-375, Ile-378, Leu-392, and Lys-396 from TolAIII that form a hydrophobic patch between the β-sheet and the second α-helix.

Superposition of TolAIII in complex with the TolAIII binding region of ColA and the solution structure of TolAIII (PDB ID: 1S62) reveals some side chain rearrangements at the interface with TA53–107 (Fig. 5). A significant rearrangement occurs from the movement of the side chain of Lys-396 of TolAIII, which makes space for the side chain of Met-87 and positions the side chain of Lys-396 directly over the aromatic ring of Phe-94, thus forming a cation-π interaction between both residues (Fig. 5). Less dramatic rearrangements of the side chains of Lys-379 and Lys-368 result in the juxtaposition of each residue with Tyr-90 and Tyr-58, respectively, allowing the formation of two further cation-π interactions across the binding interface. Interestingly, the relocation of the side chain of Asp-377 of TolAIII toward TA53–107 leads to the breakage of a salt bridge between Asp-377 and Lys-368 (observed in the free TolAIII structure) and establishment of a hydrogen bond between Asp-377 and Thr-95. In addition, the side chain of Leu-376 of TolAIII moves away from the TolA binding region, liberating Tyr-58 for the cation-π interaction with Lys-368 (Figs. 4 and 5).

The significance of the three cation-π interactions on the formation of the TA53–107-TolAIII complex was investigated by engineering the mutations Y58A, Y90A, and F94A in the TolA box of ColA and assessing the impact of these substitutions on both binding and biological activity. Comparison by surface plasmon resonance showed that TA1–107 Y90A and TA1–107 F94A had almost identical binding affinities with TolAIII as TA1–107, whereas no interaction of TolAIII was detected with TA1–107 Y58A, indicating an important role for Tyr-58 in stabilizing the interaction of TA53–107 with TolAIII (supplemental Fig. S5A). Consistent with these binding data, Y90A and F94A had no effect on the biological activity of ColA, whereas Y58A had significantly reduced activity when compared with wild-type ColA (Fig. 3; supplemental Fig. S5B).


This work describes the first crystal structure of TolA in complex with a colicin translocation domain and also reveals for the first time the interaction of a Tol protein with a globular colicin domain. The interface of the complex of TolAIII with the TA53–107 region of ColA comprises polar interactions involving residues within β-strands of each protein, further augmented by intermolecular cation-π interactions. The β-strand residues involved in contact across the interface align in an antiparallel fashion, and those from both the TA53–107 protein (92–96) and TolA (375–380) exhibit a significant degree of surface exposure. The average relative solvent accessibilities for all side chain atoms of these residues, calculated using Naccess (68), are >64 and >46%, respectively. An exhaustive search of the PDB allowed Remaut and Waksman (57) to classify protein-protein and protein-peptide complexes involving the association of a β-strand from a ligand with a β-strand or a β-sheet in a binding partner into three distinct classes: (i) β-sheet augmentation, (ii) β-strand insertion and fold complementation, or (iii) β-strand zippering. The protein-protein interface described by the x-ray structure here would appear to represent the classical β-sheet augmentation where the interaction is mediated by a β-strand from one protein pairing to the edge of a β-sheet in the binding domain of another protein (Fig. 1).

Although methylation was essential to obtain our x-ray structures of TolAIII and the TA53–107-TolAIII complex, we do not believe that methylation itself introduced any significant differences that affected the interpretation of the results. Our evidence for this is partly because of the observed similarity in the 1H-15N HSQC NMR spectra of the complex of 15N-labeled TA53–107 bound to unlabeled TolAIII with and without N-methylation (supplemental Fig. S6) and the fact that different patterns of methylation were observed in both free and complexed TolAIII. Although differences do exist between the structures of TolAIII determined by x-ray (herein) and NMR methods (25), they are small, and the residues predicted to be important in the interaction of TolAIII with TA53–107 by NMR (27) are identical to those residues shown to be involved in the x-ray structure of the TA53–107-TolAIII complex (Fig. 2). The x-ray data of TolAIII alone or in complex with TA53–107 are also identical apart from the conformational changes exerted by interacting residues across the interface of the two proteins (Fig. 5).

The contribution of individual residues of both TolAIII and TA53–107 to the binding interface of the TA53–107-TolAIII complex was determined by alanine-scanning mutagenesis. No mutation of a single residue at the interface in either TA53–107 or TolAIII prevented the interaction with the partner protein (Fig. 3; Table 2). This is very different from the interaction observed for the TolB box of ColA and TolB whereby mutations to individual residues that contribute intramolecular bonds to the fold of the TolB box on binding TolB abolish the interaction with TolB (24, 45). As the hydrogen bonds between residues Phe-94 and Thr-95 of TA53–107 and Leu-375, Leu-376, and Ile-378 of TolAIII occur from backbone atoms, this mutagenesis result is perhaps unsurprising (Fig. 2). However, the hydrogen bonds between residues Arg-92 and Arg-96 of TA53–107 are from side chain atoms specific to each arginine residue, which might be expected to be essential for the formation of the interaction between the two polypeptides. However, no significant loss of activity of ColA was observed when the mutations R92A and R96A were engineered into ColA simultaneously (Fig. 3).

This is the first reported example of a colicin-ligand interaction in which cation-π interactions contribute to the interacting complex. The importance of each cation-π interaction was initially assessed from the x-ray structure (Fig. 4; PDB ID: 3QDR). Tyr-58–Lys-368 has much closer, better coordinated dipoles than the other two paired residues and would perhaps be expected to contribute stronger electrostatic interactions. Mutagenesis of all three aromatic residues of the TolA binding region that are involved in cation-π interactions, and analysis of the interaction of the mutated polypeptides with TolA by biological activity assays and surface plasmon resonance, confirmed that Tyr-58 was essential for complex formation, whereas Tyr-90 and Phe-94 were not essential (supplemental Fig. S5, Fig. 3). Pommier et al. (15) had previously identified Tyr-58 and Tyr-90 within two putative TolA binding regions of ColA. Cation-π interactions have been shown to be vital in the translocation of lethal factor through the protective antigen channel of the anthrax toxin by establishing a protein-protein tether that encourages unidirectional movement of the toxin into the cytosol of endocytosed cells by a Brownian ratchet motion (58).

ColA is unique in colicin biology in possessing different binding epitopes for two Tol proteins. A detailed model for the recruitment of TolB by the TolB box of ColE9 has been presented (59, 60). Such models have to be extended to explain the translocation of ColA to accommodate the additional interaction of the TolA binding site with TolA. It has been proposed that the translocation domain of ColA interacts firstly with TolB in the periplasm, via a low affinity interaction of its TolB box, and then with TolA via a higher affinity interaction with its TolA binding site (24). The structural and mutagenesis information in this study provides more detail of the latter interaction.

Filamentous phages infect E. coli in a two-step process that involves the interaction of two N-terminal domains of the g3p protein of phage fd (N2 and N1) that bind to cell pili and TolAIII, respectively, in a sequential manner (26). In the fully folded form of g3p, the N1 and N2 domains are tightly associated and without disassociation of both domains by the binding of N2 to the cell pilus, N1 is unable to interact with TolAIII, and the virus is noninfective. However, the structure of the TA53–107-TolAIII complex reveals that the TolA binding region of ColA binds to a different region of TolAIII than that to which g3p binds (26), with no single residue of TolAIII being involved in binding to both proteins (Figs. 6 and 7).

There is currently no published evidence to suggest that the TolA binding site and the TolB box of ColA interact with each other in a similar way to the N1 and N2 domains of g3p. However, we have preliminary NMR data showing that the chemical shifts of a mixture of the TA1–52 and TA53–107 peptides were slightly different for some residues of TA53–107 when compared with those of TA53–107 on its own.7 NMR data have shown that a truncated polypeptide expressing both the TolA binding region and the TolB box of ColA causes an unfolding of TolAIII in vitro, which is not apparent when the TolA binding region and TolB boxes are physically separated and added to TolAIII concurrently (27). If TolA binds TolB and ColA at different binding interfaces in the TA1–107-TolA-TolB trimeric complex, then it is possible that an unoccupied TolB box of ColA could interfere and cause localized unfolding of TolAIII in the TA1–107-TolAIII complex as unfolding of TolAIII is not observed on binding of the translocation domain of ColN, which does not possess a TolB box (27). Interestingly, the TolA binding region of ColN forms intramolecular contacts with globular regions of the protein as a means of enhanced protection against proteolytic cleavage rather than domain inactivation, as seen with phage fd (29).

The binding interface of the TolA binding region of ColA presented here shows a unique fold that binds on the opposite side of TolAIII to a conserved hydrophobic cavity that forms the interface with the g3p protein (26), which has been reported as influential for conformational changes of TolAIII important for interacting with different binding partners in E. coli (25) and Pseudomonas aeruginosa (61). Recent NMR data on the interaction of TolAIII with ColN have also predicted a similar binding interface to that of g3p-TolAIII, but the precise details of the interaction of ColN with the hydrophobic cavity of TolAIII have not yet been elucidated.7 It has recently been demonstrated that TolA interacts with TolB as a prerequisite to the translocation of the enzymatic colicin, ColE9 (60). The interaction is believed to involve the extreme N terminus of TolB, but the interacting region of TolA is currently unknown. However, the similarity of the 1H-15N HSQC NMR spectrum of TolAIII in the B1–13-TolAIII complex (supplemental Fig. S7) with that of ColN-TolAIII (29) suggests that B1–13 binds TolAIII in a similar manner to ColN and g3p. TolA has also been predicted to interact with ColE1 differently than other group A colicins (62) and with YbgF via TolAII (63), and therefore, TolA exhibits the properties of a promiscuous hub protein that interacts with both intrinsic and extrinsic proteins through several different interaction sites (64). Apart from interacting networks involved in peptidoglycan synthesis and cell envelope biogenesis (65), the only other well characterized set of interacting periplasmic proteins consists of those linked by the hub protein NDsbD that are involved in disulfide isomerization (66). Several colicins and bacteriophages have taken advantage of the promiscuity afforded them by TolA for unimpeded entry into an E. coli cell. Elucidation of structural data on the interaction of TolA with TolB will not only extend our understanding of the pivotal role of TolA in cell function and homeostasis of Gram-negative bacteria, but will enhance opportunities of using TolA as a drug target for antimicrobial treatments.


*This work was supported by the Biotechnology and Biological Sciences Research Council, the Wolfson Foundation, and The Wellcome Trust.

This article contains supplemental Figs. S1–S7 and Table S1.

The atomic coordinates and structure factors (codes 3QDP and 3QDR) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (

5The following designations are used for TolA domains throughout this manuscript: TolAI, domain one of TolA consisting of N-terminal residues, 1–47; TolAII, domain two consisting of residues 48–313; TolAIII, domain three of TolA consisting of C-terminal residues 314–421.

6The following designations are used for residues throughout this manuscript: TA53–107, residues 53–107 of the translocation domain of colicin A; TA1–107, residues 1–107 of the translocation domain of colicin A; TA1–52, residues 1–52 of the translocation domain of colicin A.

7G. M. Moore, unpublished data.

4The abbreviations used are:

Pal peptidoglycan-associated lipoprotein
ColA colicin A
g3p gene 3 protein of bacteriophage fd
r.m.s. root mean square
r.m.s.d. root mean square distance
HSQC heteronuclear single quantum correlation.


We thank Sitthanan Thanintranon for construction of single TolAIII mutants. We are grateful to the European Synchrotron Radiation Facility at Grenoble, France for data collection.

1.. Sturgis J.. (Year: 2001) Organization and evolution of the tol-pal gene cluster. J. Mol. Microbiol. Biotechnol.3, 113–12211200223
2.. Cascales E.,Lloubès R.,Sturgis J. N.. (Year: 2001) The TolQ-TolR proteins energize TolA and share homologies with the flagellar motor proteins MotA-MotB. Mol. Microbiol.42, 795–80711722743
3.. Goemaere E. L.,Cascales E.,Lloubès R.. (Year: 2007) Mutational analyses define helix organization and key residues of a bacterial membrane energy-transducing complex. J. Mol. Biol.366, 1424–143617222427
4.. Gerding M. A.,Ogata Y.,Pecora N. D.,Niki H.,de Boer P. A.. (Year: 2007) The trans-envelope Tol-Pal complex is part of the cell division machinery and required for proper outer-membrane invagination during cell constriction in E. coli. Mol. Microbiol.63, 1008–102517233825
5.. Germon P.,Ray M. C.,Vianney A.,Lazzaroni J. C.. (Year: 2001) Energy-dependent conformational change in the TolA protein of Escherichia coli involves its N-terminal domain, TolQ, and TolR. J. Bacteriol.183, 4110–411411418549
6.. Dubuisson J. F.,Vianney A.,Lazzaroni J. C.. (Year: 2002) Mutational analysis of the TolA C-terminal domain of Escherichia coli and genetic evidence for an interaction between TolA and TolB. J. Bacteriol.184, 4620–462512142433
7.. Walburger A.,Lazdunski C.,Corda Y.. (Year: 2002) The Tol/Pal system function requires an interaction between the C-terminal domain of TolA and the N-terminal domain of TolB. Mol. Microbiol.44, 695–70811994151
8.. Cascales E.,Gavioli M.,Sturgis J. N.,Lloubès R.. (Year: 2000) Proton motive force drives the interaction of the inner membrane TolA and outer membrane Pal proteins in Escherichia coli. Mol. Microbiol.38, 904–91511115123
9.. Lloubès R.,Cascales E.,Walburger A.,Bouveret E.,Lazdunski C.,Bernadac A.,Journet L.. (Year: 2001) The Tol-Pal proteins of the Escherichia coli cell envelope: an energized system required for outer membrane integrity?Res. Microbiol.152, 523–52911501670
10.. Cascales E.,Bernadac A.,Gavioli M.,Lazzaroni J. C.,Lloubès R.. (Year: 2002) Pal lipoprotein of Escherichia coli plays a major role in outer membrane integrity. J. Bacteriol.184, 754–75911790745
11.. Bouveret E.,Derouiche R.,Rigal A.,Lloubès R.,Lazdunski C.,Bénédetti H.. (Year: 1995) Peptidoglycan-associated lipoprotein-TolB interaction: a possible key to explaining the formation of contact sites between the inner and outer membranes of Escherichia coli. J. Biol. Chem.270, 11071–110777744736
12.. Baty D.,Lakey J.,Pattus F.,Lazdunski C.. (Year: 1990) A 136-amino-acid-residue COOH-terminal fragment of colicin A is endowed with ionophoric activity. Eur. J. Biochem.189, 409–4131692532
13.. Bouveret E.,Rigal A.,Lazdunski C.,Bénédetti H.. (Year: 1998) Distinct regions of the colicin A translocation domain are involved in the interaction with TolA and TolB proteins upon import into Escherichia coli. Mol. Microbiol.27, 143–1579466263
14.. Journet L.,Bouveret E.,Rigal A.,Lloubès R.,Lazdunski C.,Bénédetti H.. (Year: 2001) Import of colicins across the outer membrane of Escherichia coli involves multiple protein interactions in the periplasm. Mol. Microbiol.42, 331–34411703658
15.. Pommier S.,Gavioli M.,Cascales E.,Lloubès R.. (Year: 2005) Tol-dependent macromolecule import through the Escherichia coli cell envelope requires the presence of an exposed TolA binding motif. J. Bacteriol.187, 7526–753416237036
16.. Raggett E. M.,Bainbridge G.,Evans L. J.,Cooper A.,Lakey J. H.. (Year: 1998) Discovery of critical Tol A-binding residues in the bactericidal toxin colicin N: a biophysical approach. Mol. Microbiol.28, 1335–13439680221
17.. Webster R. E.. (Year: 1991) The tol gene products and the import of macromolecules into Escherichia coli. Mol. Microbiol.5, 1005–10111683466
18.. Riechmann L.,Holliger P.. (Year: 1997) The C-terminal domain of TolA is the coreceptor for filamentous phage infection of E. coli. Cell90, 351–3609244308
19.. Bonsor D. A.,Meenan N. A.,Kleanthous C.. (Year: 2008) Colicins exploit native disorder to gain cell entry: a hitchhiker's guide to translocation. Biochem. Soc. Trans.36, 1409–141319021565
20.. Kleanthous C.. (Year: 2010) Swimming against the tide: progress and challenges in our understanding of colicin translocation. Nat. Rev. Microbiol.8, 843–84821060316
21.. Housden N. G.,Wojdyla J. A.,Korczynska J.,Grishkovskaya I.,Kirkpatrick N.,Brzozowski A. M.,Kleanthous C.. (Year: 2010) Directed epitope delivery across the Escherichia coli outer membrane through the porin OmpF. Proc. Natl. Acad. Sci. U.S.A.107, 21412–2141721098297
22.. Bonsor D. A.,Grishkovskaya I.,Dodson E. J.,Kleanthous C.. (Year: 2007) Molecular mimicry enables competitive recruitment by a natively disordered protein. J. Am. Chem. Soc.129, 4800–480717375930
23.. Loftus S. R.,Walker D.,Maté M. J.,Bonsor D. A.,James R.,Moore G. R.,Kleanthous C.. (Year: 2006) Competitive recruitment of the periplasmic translocation portal TolB by a natively disordered domain of colicin E9. Proc. Natl. Acad. Sci. U.S.A.103, 12353–1235816894158
24.. Zhang Y.,Li C.,Vankemmelbeke M. N.,Bardelang P.,Paoli M.,Penfold C. N.,James R.. (Year: 2010) The crystal structure of the TolB box of colicin A in complex with TolB reveals important differences in the recruitment of the common TolB translocation portal used by group A colicins. Mol. Microbiol.75, 623–63619627502
25.. Deprez C.,Lloubès R.,Gavioli M.,Marion D.,Guerlesquin F.,Blanchard L.. (Year: 2005) Solution structure of the E. coli TolA C-terminal domain reveals conformational changes upon binding to the phage g3p N-terminal domain. J. Mol. Biol.346, 1047–105715701516
26.. Lubkowski J.,Hennecke F.,Plückthun A.,Wlodawer A.. (Year: 1999) Filamentous phage infection: crystal structure of g3p in complex with its coreceptor, the C-terminal domain of TolA. Structure7, 711–72210404600
27.. Hecht O.,Zhang Y.,Li C.,Penfold C. N.,James R.,Moore G. R.. (Year: 2010) Characterization of the interaction of colicin A with its co-receptor TolA. FEBS Lett.584, 2249–225220433837
28.. Delaglio F.,Grzesiek S.,Vuister G. W.,Zhu G.,Pfeifer J.,Bax A.. (Year: 1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR.6, 277–2938520220
29.. Hecht O.,Ridley H.,Boetzel R.,Lewin A.,Cull N.,Chalton D. A.,Lakey J. H.,Moore G. R.. (Year: 2008) Self-recognition by an intrinsically disordered protein. FEBS Lett.582, 2673–267718573254
30.. Walter T. S.,Meier C.,Assenberg R.,Au K. F.,Ren J.,Verma A.,Nettleship J. E.,Owens R. J.,Stuart D. I.,Grimes J. M.. (Year: 2006) Lysine methylation as a routine rescue strategy for protein crystallization. Structure14, 1617–162217098187
31.. Rayment I.. (Year: 1997) Reductive alkylation of lysine residues to alter crystallization properties of proteins. Methods Enzymol.276, 171–1799048376
32.. Leslie A. G.. (Year: 1992) Joint CCP4 and ESF-EACMB Newsletter on Protein Crystallography, Vol. 26, Collaborative Computational Project Number 4 in Protein Crystallography, Oxon, UK
33.. Evans P.. (Year: 2006) Scaling and assessment of data quality. Acta Crystallogr. D Biol. Crystallogr.62, 72–8216369096
34.. Evans P.. (Year: 1997) CCP4 Newsletter on Protein Crystallography, Vol. 33, pp. 22–24, Collaborative Computational Project Number 4 in Protein Crystallography, Oxon, UK
35.. Collaborative Computational Project, Number 4 (Year: 1994) The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D Biol. Crystallogr.50, 760–76315299374
36.. McCoy A. J.,Grosse-Kunstleve R. W.,Adams P. D.,Winn M. D.,Storoni L. C.,Read R. J.. (Year: 2007) Phaser crystallographic software. J. Appl. Crystallogr.40, 658–67419461840
37.. Terwilliger T. C.,Berendzen J.. (Year: 1999) Automated MAD and MIR structure solution. Acta Crystallogr. D Biol. Crystallogr.55, 849–86110089316
38.. Cowtan K.. (Year: 1999) Error estimation and bias correction in phase-improvement calculations. Acta Crystallogr. D Biol. Crystallogr.55, 1555–156710489450
39.. Cowtan K.. (Year: 2006) The Buccaneer software for automated model building. 1. Tracing protein chains. Acta Crystallogr. D Biol. Crystallogr.62, 1002–101116929101
40.. Emsley P.,Cowtan K.. (Year: 2004) Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr.60, 2126–213215572765
41.. Murshudov G. N.,Vagin A. A.,Dodson E. J.. (Year: 1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr. D Biol. Crystallogr.53, 240–25515299926
42.. Painter J.,Merritt E. A.. (Year: 2006) TLSMD web server for the generation of multi-group TLS models. J. Appl. Crystallogr.39, 109–111
43.. Lee B.,Richards F. M.. (Year: 1971) The interpretation of protein structures: estimation of static accessibility. J. Mol. Biol.55, 379–4005551392
44.. Gallivan J. P.,Dougherty D. A.. (Year: 1999) Cation-π interactions in structural biology. Proc. Natl. Acad. Sci. U.S.A.96, 9459–946410449714
45.. Hands S. L.,Holland L. E.,Vankemmelbeke M.,Fraser L.,Macdonald C. J.,Moore G. R.,James R.,Penfold C. N.. (Year: 2005) Interactions of TolB with the translocation domain of colicin E9 require an extended TolB box. J. Bacteriol.187, 6733–674116166536
46.. Yang Z. R.,Thomson R.,McNeil P.,Esnouf R. M.. (Year: 2005) RONN: the bio-basis function neural network technique applied to the detection of natively disordered regions in proteins. Bioinformatics21, 3369–337615947016
47.. Macdonald C. J.,Tozawa K.,Collins E. S.,Penfold C. N.,James R.,Kleanthous C.,Clayden N. J.,Moore G. R.. (Year: 2004) Characterization of a mobile protein-binding epitope in the translocation domain of colicin E9. J. Biomol. NMR30, 81–9615452437
48.. Tozawa K.,Macdonald C. J.,Penfold C. N.,James R.,Kleanthous C.,Clayden N. J.,Moore G. R.. (Year: 2005) Clusters in an intrinsically disordered protein create a protein-binding site: the TolB-binding region of colicin E9. Biochemistry44, 11496–1150716114886
49.. Collins E. S.,Whittaker S. B.,Tozawa K.,MacDonald C.,Boetzel R.,Penfold C. N.,Reilly A.,Clayden N. J.,Osborne M. J.,Hemmings A. M.,Kleanthous C.,James R.,Moore G. R.. (Year: 2002) Structural dynamics of the membrane translocation domain of colicin E9 and its interaction with TolB. J. Mol. Biol.318, 787–80412054823
50.. Deprez C.,Blanchard L.,Guerlesquin F.,Gavioli M.,Simorre J. P.,Lazdunski C.,Marion D.,Lloubès R.. (Year: 2002) Macromolecular import into Escherichia coli: the TolA C-terminal domain changes conformation when interacting with the colicin A toxin. Biochemistry41, 2589–259811851406
51.. Goldschmidt L.,Cooper D. R.,Derewenda Z. S.,Eisenberg D.. (Year: 2007) Toward rational protein crystallization: a Web server for the design of crystallizable protein variants. Protein Sci.16, 1569–157617656576
52.. Lo Conte L.,Chothia C.,Janin J.. (Year: 1999) The atomic structure of protein-protein recognition sites. J. Mol. Biol.285, 2177–21989925793
53.. Bahadur R. P.,Zacharias M.. (Year: 2008) The interface of protein-protein complexes: analysis of contacts and prediction of interactions. Cell. Mol. Life Sci.65, 1059–107218080088
54.. Lawrence M. C.,Colman P. M.. (Year: 1993) Shape complementarity at protein/protein interfaces. J. Mol. Biol.234, 946–9508263940
55.. Gokce I.,Raggett E. M.,Hong Q.,Virden R.,Cooper A.,Lakey J. H.. (Year: 2000) The TolA-recognition site of colicin N: ITC, SPR, and stopped-flow fluorescence define a crucial 27-residue segment. J. Mol. Biol.304, 621–63211099384
56.. Bernadac A.,Gavioli M.,Lazzaroni J. C.,Raina S.,Lloubès R.. (Year: 1998) Escherichia coli tol-pal mutants form outer membrane vesicles. J. Bacteriol.180, 4872–48789733690
57.. Remaut H.,Waksman G.. (Year: 2006) Protein-protein interaction through β-strand addition. Trends Biochem. Sci.31, 436–44416828554
58.. Krantz B. A.,Melnyk R. A.,Zhang S.,Juris S. J.,Lacy D. B.,Wu Z.,Finkelstein A.,Collier R. J.. (Year: 2005) A phenylalanine clamp catalyzes protein translocation through the anthrax toxin pore. Science309, 777–78116051798
59.. Papadakos G.,Housden N. G.,Lilly K. J.,Kaminska R.,Kleanthous C.. (Year: 2012) Kinetic basis for the competitive recruitment of TolB by the intrinsically disordered translocation domain of colicin E9. J. Mol. Biol.418, 269–28022310049
60.. Bonsor D. A.,Hecht O.,Vankemmelbeke M.,Sharma A.,Krachler A. M.,Housden N. G.,Lilly K. J.,James R.,Moore G. R.,Kleanthous C.. (Year: 2009) Allosteric β-propeller signaling in TolB and its manipulation by translocating colicins. EMBO J.28, 2846–285719696740
61.. Witty M.,Sanz C.,Shah A.,Grossmann J. G.,Mizuguchi K.,Perham R. N.,Luisi B.. (Year: 2002) Structure of the periplasmic domain of Pseudomonas aeruginosa TolA: evidence for an evolutionary relationship with the TonB transporter protein. EMBO J.21, 4207–421812169623
62.. Schendel S. L.,Click E. M.,Webster R. E.,Cramer W. A.. (Year: 1997) The TolA protein interacts with colicin E1 differently than with other group A colicins. J. Bacteriol.179, 3683–36909171417
63.. Krachler A. M.,Sharma A.,Cauldwell A.,Papadakos G.,Kleanthous C.. (Year: 2010) TolA modulates the oligomeric status of YbgF in the bacterial periplasm. J. Mol. Biol.403, 270–28520816983
64.. Patil A.,Kinoshita K.,Nakamura H.. (Year: 2010) Hub promiscuity in protein-protein interaction networks. Int. J. Mol. Sci.11, 1930–194320480050
65.. Hu P.,Janga S. C.,Babu M.,Díaz-Mejía J. J.,Butland G.,Yang W.,Pogoutse O.,Guo X.,Phanse S.,Wong P.,Chandran S.,Christopoulos C.,Nazarians-Armavil A.,Nasseri N. K.,Musso G.,Ali M.,Nazemof N.,Eroukova V.,Golshani A.,Paccanaro A.,Greenblatt J. F.,Moreno-Hagelsieb G.,Emili A.. (Year: 2009) Global functional atlas of Escherichia coli encompassing previously uncharacterized proteins. PLoS biol7, e9619402753
66.. Stirnimann C. U.,Grütter M. G.,Glockshuber R.,Capitani G.. (Year: 2006) nDsbD: a redox interaction hub in the Escherichia coli periplasm. Cell. Mol. Life Sci.63, 1642–164816786221
67.. DeLano W. L.. (Year: 2010) The PyMOL Molecular Graphics System, version 1.3r1. , Schrödinger, LLC, New York
68.. Hubbard S.,Thornton J.. (Year: 1996) Naccess, University of Manchester, Manchester, U.K.


[Figure ID: F1]

Structure of TA53–107-TolAIII complex. A stereo image of the secondary structure of TA53–107 (cyan) in complex with TolAIII (green) displaying a classical β-sheet augmentation pattern of complex formation is shown.

[Figure ID: F2]

Binding interface of TA53–107-TolAIII complex. A ball and stick representation of the interface between TA53–107 (pink) and TolAIII (green) is shown. Intermolecular hydrogen bonds either direct or mediated by a water molecule (gray sphere), between interacting residues are shown as hyphenated lines.

[Figure ID: F3]

Effects of TolA box mutations of ColA on activity of ColA are cumulative.A, individual (pYZ106), double (pYZ107 and pYZ108), and multiple mutations of the TolA box of ColA (pYZ109 and pYZ110) were engineered using PCR mutagenesis, and the mutant polypeptides were purified and spotted onto a lawn of sensitive E. coli cells. B, zones of clearing indicative of cell killing were reduced in mutants containing successively more mutations and disappeared in all mutants containing three or more mutations.

[Figure ID: F4]

Cation-π interactions across TA53–107-TolAIII complex. Electron-rich aromatic tyrosine (Tyr-58, Tyr-90) and phenylalanine (Phe-94) residues of the TolA box stack perpendicular to the positively charged side chains of lysine residues (Lys-368, Lys-379, Lys-396), forming cation-π interactions that strengthen and stabilize the interaction at the interface between the two polypeptides.

[Figure ID: F5]

Conformational changes to lysine residues of TolAIII on binding TA53–107. Lysine residues, Lys-368 and Lys-396, are shifted on binding the TolA box of ColA and realign with the aromatic residues Tyr-58 and Phe-94, respectively, forming stabilizing cation-π interactions. Residues Leu-376 and Asp-377 are favorably shifted to accommodate the formation of the two cation-π interactions. Asp-377 forms a hydrogen bond (dashed line) with Thr-96 across the binding interface of the two polypeptides. Arrows indicate the size and direction of the conformational shifts.

[Figure ID: F6]

Primary sequence of TolAIII. Helical (white boxes) and β-sheet (gray arrow) regions are shown above the primary sequence of TolAIII. Residue numbers in TolA are shown to the left. Residues interacting with the TolA binding region of ColA (*) are very different from residues that interact with g3p (+). A His6 tag with the sequence MAKHHHHHHM was engineered at the N terminus of TolAIII to promote efficient purification and avoid the important C terminus of TolAIII.

[Figure ID: F7]

Comparison of structures of TolAIII in complex with TA53–107 and g3p. Structures of TA53–107 (blue) in complex with TolAIII (green) and g3p (red) in complex with TolAIII (red) were overlaid using the PyMOL program (67), showing that both ligands form β-strand addition pairs to opposing β-sheets of TolAIII. The interactions of both do not cause any major conformational changes to the tertiary structure of TolAIII.

[TableWrap ID: T1] TABLE 1 

Data collection and refinement statistics

TolAIII TA53–107-TolAIII
Native Se-Met
Data collection
    Space group P31 2 1 P21 21 2 P21 21 2
    Cell dimensions a, b, c (Å) 70.28, 70.28, 53.82 47.50, 119.62, 30.15 47.10,118.69, 29.85
    Wavelength, Å 0.9792 0.8726 0.9795
    Resolution range, Å 40.32–2.15 (2.27–2.15) 44.14–2.65 (2.79–2.65) 43.81–2.8 (2.95–2.8)
    No. of unique reflections 8641 (1234) 5238 (757) 4356 (585)
    Total no. of observations 84,194 (12,226) 18,767 (2757) 50,579 (3176)
    Redundancy 9.7 (9.9) 3.6 (3.6) 11.6 (5.4)
    Mean I/σ(I) 25.7 (5.0) 10.5 (3.8) 21.9 (5.7)
    Completeness, % 100 (100) 97.4 (98.2) 97.3 (92.4)
    Rmerge, % 5.6 (43.3) 9.8 (34.4) 9.1 (30.4)
    Anomalous completeness, % 96.5 (87.2)
    Anomalous redundancy 6.5 (3.0)
    Phasing statistics
        FOMa after PHASER 0.335
        FOM after DM 0.528
        FOM after Refmac5 0.797

Refinement statistics
    Resolution range, Å 60.87–2.15 59.81–2.65
    Rwork/Rfree % 19.2/22.5 24.4/27.3
    r.m.s. bond lengths, Å 0.0286 0.0137
    r.m.s. angles, ° 2.371 1.867
    Average B-factor (protein, Å2) 20.3 26.4
    No. of protein atoms 685 1059
    No. of water and ion atoms 52 34
    PDB accession code 3QDP 3QDR

TF1-1a FOM, figure of merit.

[TableWrap ID: T2] TABLE 2 

Complementation of the TolA mutant strain of E. coli, JC7782 with TolA and TolA containing various mutations expressed from the plasmid, pTol

Strain TolAIII mutation Complementationa
JC7782 TolA >250 nMb
JC7782 (pTolA) None 10 nM
JC7782 (pST1) L375A 10 nM
JC7782 (pST2) L376A 50 nM
JC7782 (pST3) D377A 10 nM
JC7782 (pST4) I378A 10 nM
JC7782 (pST5) P380A 10 nM
JC7782 (pNP341) L375A, D377A, P380A 50 nM
JC7782 (pNP342) L375A, L376A, D377A, I378A, P380A >250 nMb

TF2-1a Complementation was assessed as the lowest concentration of colicin A that was able to produce a visible zone of inhibition on LB agar plates containing an overlay of the strain indicated in the table.

TF2-2b Zones of inhibition were not seen in these strains at any concentration of ColA tested.

Article Categories:
  • Protein Structure and Folding

Keywords: Bacterial Toxins, Microbiology, Protein Structure, Protein-Protein Interactions, X-ray Crystallography, Tol, Bacteriocin, Colicin.

Previous Document:  Receptor signaling lymphocyte-activation molecule family 1 (Slamf1) regulates membrane fusion and NA...
Next Document:  Stretch-induced fetal type II cell differentiation is mediated via ErbB1-ErbB4 interactions.