SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

The Tol–Pal system of the Escherichia coli cell envelope is composed of five proteins. TolQ, TolR and TolA form a complex in the inner membrane, whereas TolB is a periplasmic protein interacting with Pal, the peptidoglycan-associated lipoprotein anchored to the outer membrane. This system is required for outer membrane integrity and has been shown to form a trans-envelope bridge linking inner and outer membranes. The TolA–Pal interaction plays an important role in the function of this system and has been found to depend on the proton motive force and the TolQ and TolR proteins. The Pal lipoprotein interacts with many components, such as TolA, TolB, OmpA, the major lipoprotein and the murein layer. In this study, six pal deletions were constructed. The analyses of the resulting Pal protein functions and interactions defined an N-terminal region of 40 residues, which can be deleted without any cell-damaging effect, and three independent regions required for its interaction with TolA, OmpA and TolB or the peptidoglycan. The analyses of the integrity of the cells producing the various Pal lipoproteins revealed strong outer membrane destabilization only when binding regions were deleted. Furthermore, a conserved polypeptide sequence located downstream of the peptidoglycan binding motif of Pal was required for the TolA–Pal interaction and for the maintenance of outer membrane stability.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

The cell envelope of Gram-negative bacteria acts as a physical and selective barrier against toxic compounds. The lipopolysaccharide (LPS) leaflet prevents the diffusion of toxic compounds through the outer membrane (Raetz, 1996), while porins and other specific receptors allow passive, selective or active transport of nutrients (Koebnik et al., 2000). The murein layer is important for bacterial resistance against osmotic and mechanical stresses. Many outer membrane proteins (OMPs) interact with this murein layer. The well-characterized major outer membrane lipoprotein (Lpp) (Braun and Rehn, 1969) is covalently bound to the peptidoglycan by its C-terminal lysine residue (Braun, 1975). Lpp exists as a trimer (Choi et al., 1986), and its three-dimensional structure has been solved by crystallography (Shu et al., 2000). This protein plays an important role in the stability of the outer membrane. Deletion or mutations of lpp result in numerous defects such as periplasmic leakage, increased susceptibility to many toxic compounds and the formation of outer membrane vesicles (Suzuki et al., 1978; Yem and Wu, 1978). Other peptidoglycan-associated proteins seem to play a role in maintaining the cell wall organization. These proteins are not covalently linked to the murein layer (Leduc et al., 1992) but interact with murein through a conserved, proposed α-helical motif (Koebnik, 1995). This motif is present in MotB, a component of the force-generating unit of the flagellar motor proteins (Blair et al., 1991), in two well-characterized proteins, OmpA and Pal, and in some proteins of unknown function (De Mot and Vanderleyden, 1994). Pal is a lipoprotein with a serine residue at position +2 that results in its localization in the outer membrane (Gennity and Inouye, 1991). After cleavage of the signal sequence by the signal peptidase II and acylation, Pal is targeted to the outer membrane by the LolABCDE system (Matsuyama et al., 1995; 1997; Yokota et al., 1999; Yakushi et al., 2000). The interaction of Pal with the peptidoglycan has been characterized by SDS extraction (Mizuno, 1979), chemical cross-linking (Leduc et al., 1992) or in vitro binding (Bouveret et al., 1999). The peptidoglycan binding sequence of Pal, located between residues 97 and 114, was mapped using PhoA insertions (Lazzaroni and Portalier, 1992), mutagenesis (Clavel et al., 1998) or competition with a synthetic peptide (Bouveret et al., 1999).

Besides the higher susceptibility to vancomycin observed in pal and tolA mutants (Cascales et al., 2000), each of the tol and pal cells presents outer membrane defects similar to lpp mutation (Lazzaroni and Portalier, 1992; Bernadac et al., 1998). Mutations in the tol–pal genes were initially found to make cells tolerant to group A colicins (Nagel de Zwaig and Luria, 1967; Nomura and Witten, 1967), and extensive biochemical results indicate that the Tol–colicin interactions are involved in colicin import through the outer membrane (for a review, see Lazdunski et al., 1998). The tol–pal cluster codes for seven proteins. TolQ, TolR and TolA are inner membrane proteins that interact together by their transmembrane helices (Derouiche et al., 1995; Lazzaroni et al., 1995; Germon et al., 1998; Journet et al., 1999). The periplasmic TolB and the Pal lipoprotein form another complex anchored to the outer membrane (Bouveret et al., 1995; Ray et al., 2000). TolB was previously found to interact in a Pal-dependent manner with OmpA and Lpp (Clavel et al., 1998), while Pal interacts with OmpA (Palva, 1979) and forms a homodimer in the cell envelope (Cascales et al., 2002) (see Fig. 9). Besides TolQ–R–A–B–Pal, the cytoplasmic YbgC and the periplasmic YbgF proteins are also encoded by the tol–pal cluster (Sun and Webster, 1987; Vianney et al., 1996) in most Gram-negative bacteria (Sturgis, 2001). However, deletion of either ybgC or ybgF does not confer a tol phenotype. In a systematic yeast two-hybrid screening, Walburger et al. (2002) have shown that TolA interacts with YbgF and that the N-terminal domain of TolB was able to interact with the C-terminal domain of TolA. Suppressor mutants of tolA mutations in the tolB gene have further confirmed the TolA–TolB interaction (Dubuisson et al., 2002). The C-terminal domain of TolA was also found to interact with Pal in a proton motive force-dependent manner (Cascales et al., 2000). Both TolQ and TolR, two proteins suspected to form ionic channels, were found to be required in this process (Cascales et al., 2001). Moreover, conformational changes of TolA dependent on its transmembrane residues, on TolQ and TolR and on the membrane potential have been shown (Germon et al., 2001). Thus, the Tol–Pal apparatus forms a trans-envelope bridge linking the inner and outer membranes to the peptidoglycan.

image

Figure 9. Schematic representation of the Tol–Pal system of the E. coli cell envelope. The topologies of the different proteins of the Tol–Pal system and of OmpA are represented. The various interactions are indicated by arrows.

Download figure to PowerPoint

In the present study, deletion-directed mutagenesis was originally performed to map the TolA binding sequence of Pal. In vivo cross-linking followed by TolA or Pal immunoprecipitations revealed three independent regions of Pal required for its interaction with TolA, TolB and OmpA, while a conserved motif located in the C-terminal region of Pal was found to be required for TolA binding. The analyses of outer membrane stability of the different pal mutants are discussed in relation with these interactions.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Production of truncated Pal lipoproteins

To characterize the important regions for the peptidoglycan-associated lipoprotein (Pal) function, we constructed six pal mutants, with deletions removing sequences coding for 15–32 residues (Fig. 1A). The truncated Pal proteins correspond to PalΔ3–17 (Pal1), PalΔ19–43 (Pal2), PalΔ44–62 (Pal3), PalΔ62–93 (Pal4), PalΔ94–121 (Pal5) and PalΔ123–152 (Pal6). pPal plasmids, expressing the truncated pal genes under the control of the bla promoter, were analysed after transformation into the pal strain JC892. The production of truncated Pal proteins was checked by immunoblotting using anti-Pal or anti-TolB–Pal polyclonal antibodies (Fig. 1B). Because some Pal constructs were poorly or not detected, a FLAG epitope (sequence DYKDDDDK) was inserted at the C-terminal end of each of the truncated proteins as well as of the native Pal (PalFLAG and PalFLAG1–PalFLAG6 correspond, respectively, to Pal and Pal1–Pal6). All the tagged proteins could be detected in pal or pal pTolA cells (used in the accompanying co-immunoprecipitation results) transformed with the pPalFLAG plasmids after immunodetections with the monoclonal anti-FLAG antibody (Fig. 1B), indicating that the absence of immunodetection of some constructs with the anti-Pal antibodies probably resulted from removal of anti-Pal epitopes.

image

Figure 1. A. Schematic representation of the truncated Pal lipoproteins. The residue number neighbouring the deletion is indicated together with the deletion length (in brackets). The residue content and the theoretical molecular mass (without the acyl chain) of the Pal derivatives are indicated. B. Immunodetections of the truncated Pal lipoproteins. About 2 × 108 cells were loaded on an SDS-12.5% PAGE, and the Pal derivatives harbouring the FLAG epitope were immunodetected using either the anti-Pal pAb or anti-FLAG mAb. Molecular weight markers are indicated in kDa.

Download figure to PowerPoint

Outer membrane complementation effects produced by truncated Pal proteins

pPal derivatives carrying the pal deletions were introduced into the Escherichia coli pal strain for complementation analyses. Cell envelope integrity was evaluated by the formation of outer membrane vesicles, the leakage of periplasmic RNase I and the susceptibilities to the detergent sodium dodecyl sulphate (SDS) and the antibiotic vancomycin. As shown in Table 1, different regions were found to be functionally important for maintaining cell envelope integrity as the corresponding truncated Pal proteins were unable to complement the pal null mutation. Conversely, expression of Pal, Pal1 or Pal2 was found to restore outer membrane integrity, establishing that the 40 first residues of Pal are functionally silent. However, an intermediate level of complementation was observed when Pal4 was produced. We also checked whether the FLAG epitope or the overexpression of TolA had some incidence on cell envelope integrity. The various pPalFLAG plasmids were transformed in either pal or pal pTolA cells, and the periplasmic release of RNase I was analysed. Similar results to those obtained in Table 1 were observed, thus indicating that the FLAG epitope or the overexpression of TolA does not affect the function of the various truncated Pal.

Table 1. . Outer membrane defects of pal deletion mutants.
StrainsOMVaRNasebSDScVanc
  • a . Amounts of outer membrane vesicles observed by electron microscopy after negative staining of cells grown on agar plates (classified according to Bernadac et al., 1998): ++, many vesicles on all the cells; +, some vesicles on most cells; 0, no vesicle on most cells.

  • b

    . RNase I leakage, ++ and + indicated that the clear zone of RNA hydrolysis was larger or smaller than the size of the cell colony respectively (0, no leakage).

  • c . SDS (% w/v) and vancomycin (µg ml −1) required for a 50% decrease in cell turbidity measured after 3 h of culture (average values from duplicate experiments ± 10%).

JC8056 pT700>2 175
pal pT7+++ 0.1  30
pal pPal00>2 200
pal pPal100>2>200
pal pPal200>2 180
pal pPal3++++ 0.1  35
pal pPal4++>2 115
pal pPal5++++ 0.1  30
pal pPal6++++ 0.1  35
pal pPalΔKNRR++ 0.4 110
pal pPalΔSYGK++++ 0.1  50

Localization and peptidoglycan affinity of the truncated Pal lipoproteins

We then analysed the membrane localization and the peptidoglycan-binding capacity of the different truncated Pal lipoproteins. In order to test whether the different truncated proteins were correctly targeted to the outer membrane, fractionation experiments were performed. After ultrasonic treatment of an exponentially growing culture of the pal strain harbouring pPalFLAG derivatives, inner and outer membranes were isolated as described by de Cock et al. (2001). All the truncated proteins were found to be associated with the outer membrane, whereas TolR and OmpA were used as inner and outer membrane references respectively (data not shown). These results are consistent with the presence, in the various truncated Pal proteins, of the first two amino acids of the processed form, responsible for post-translational acylation and targeting.

The peptidoglycan affinity of the truncated Pal proteins was analysed in the pal pTolA strain carrying pPalFLAG derivatives. PalFLAG, PalFLAG1 and PalFLAG2 proteins were immunodetected mostly in the peptidoglycan fraction, whereas PalFLAG3 and PalFLAG4 were recovered mostly in the solubilized fraction corresponding to non-associated protein, and the two C-terminal deletions were quite absent from the peptidoglycan fraction. As controls, OmpA was immunodetected in the peptidoglycan and solubilized fractions, while TolR was exclusively detected in the solubilized fraction (Fig. 2).

image

Figure 2. Peptidoglycan association analyses of the truncated PalFLAG lipoproteins. Cell envelope proteins solubilized in 2% SDS and peptidoglycan-associated proteins were analysed in pal pTolA cells harbouring the various pPalFLAG plasmids. Solubilized (S) and peptidoglycan-associated fractions (P) were loaded on SDS-12.5% PAGE, and immunodetections of PalFLAG, OmpA and TolR were performed.

Download figure to PowerPoint

Mapping of Pal regions required for the interactions with cell envelope proteins

We reported previously that Pal lipoprotein interacts with OmpA and Lpp after in vivo cross-linking of wild-type cells (Cascales et al., 2002). As a first step, we verified that the different Pal complexes observed previously were still present after formaldehyde (FA) treatment of pal, lpp and ompA strains expressing PalFLAG. The PalFLAG dimer, OmpA–PalFLAG and Lpp–PalFLAG complexes were specifically detected (Fig. 3), indicating no negative effects of the additional C-terminal FLAG sequence. We therefore explored the regions of Pal involved in these multiple interactions. JC892 carrying the various pPalFLAG plasmids were treated with disuccinimidyl tartrate (DST) or formaldehyde, and complexes formed with the PalFLAG derivatives were immunodetected with the monoclonal anti-FLAG antibody. The result indicates that all the Pal truncations, except PalFLAG6, retain the ability to form dimers (data not shown). However, as some complexes present weak immunodetection signals, attributing a specific region of Pal required for its interactions with OmpA and Lpp remains speculative.

image

Figure 3. PalFLAG dimerizes and interacts with Lpp and OmpA in vivo. Western blotting immunodetections of PalFLAG produced in the indicated mutants. FA-treated samples were heated at 37°C before separation on SDS-12.5% PAGE and further immunodetected with anti-FLAG mAb. Prestained mass markers (in kDa) and Pal complexes are indicated.

Download figure to PowerPoint

We also investigated the region of Pal involved in its interaction with the TolA protein. Immunoprecipitation with the anti-TolA antibody was carried out after in vivo FA cross-linking of pal pTolA cells harbouring the various pPalFLAG plasmids, as this technique was used previously to check the TolA–Pal interaction, which found to be enhanced by the overproduction of TolA (Cascales et al., 2000). Our findings indicate that all truncated Pal could be co-immunoprecipitated with TolA, except PalFLAG6 (Fig. 4). As a control, TolR was found to co-immunoprecipitate with TolA in the cells harbouring the various pPalFLAG.

image

Figure 4. The C-terminal region of Pal interacts with TolA. A pal strain harbouring pTolA and the various pPalFLAG plasmids were cross-linked in vivo with formaldehyde and immunoprecipitated further with pAb anti-TolAII–III. The Western blotting immunodetections of modified Pal, TolR and TolA were detected by anti-FLAG mAb, anti-TolR pAb and by the overlay technique using the purified N-terminal domain of colicin A (AT) and anti-1C11 mAb respectively.

Download figure to PowerPoint

Conversely, in vivo cross-linking with FA and immunoprecipitations with anti-FLAG antibody were performed on cells expressing the PalFLAG derivatives (Fig. 5). Under these conditions, TolA was not immunodetected in the absence of the C-terminal region of Pal. We therefore asked whether these truncated forms of Pal interact with TolB or OmpA. The amount of TolA, TolB or OmpA, estimated by Western blotting immunodetections, was found to be similar when Pal or truncated forms were produced (data not shown). Thus, the immunoprecipitated samples of the truncated PalFLAG derivatives were analysed for the presence of TolB and OmpA. Both TolB and OmpA were found to co-immunoprecipitate specifically with PalFLAG (Fig. 5). However, PalFLAG3 and PalFLAG5 do not stably associate with OmpA and TolB respectively. At the end of the immunoprecipitation studies, we concluded that the regions 44–62, 94–121 and 123–152 of Pal are required for the interaction with OmpA, TolB and TolA, respectively, and that the absence of one of the three interacting regions did not prevent the binding of the two other regions.

image

Figure 5. Three independent regions of Pal are involved in its interactions with TolA, TolB and OmpA. A pal strain harbouring pTolA and the various pPalFLAG plasmids were cross-linked in vivo with formaldehyde and immunoprecipitated further with anti-FLAG mAb. The Western blotting immunodetections of truncated Pal, TolA, OmpA and TolB proteins were detected with anti-FLAG mAb, anti-TolAIII pAb, anti-OmpA pAb and anti-TolBHis pAb respectively.

Download figure to PowerPoint

Pal interacts independently with peptidoglycan, OmpA, TolA and TolB

The association of Pal with the peptidoglycan was analysed further in the tolA, tolB and ompA–lpp strains to check whether the absence of a Pal partner could modify its binding to the cell wall. The results indicated that, in each mutant, PalFLAG was found to be associated with the peptidoglycan (data not shown). Because the different proteins interacting with Pal have also been shown to interact with each other, notably TolA with TolB and TolB with Lpp or OmpA, the cross-linking and co-immunoprecipitation experiments were checked in tolA, tolB, lpp or ompA strains harbouring pPalFLAG. PalFLAG was found to interact with TolA in tolB, lpp and ompA strains, with TolB in tolA, lpp and ompA strains, and with OmpA in tolA, tolB and lpp strains (data not shown). Thus, the different binding regions of Pal are a reflection of direct interactions occurring independently with TolA, OmpA, TolB and/or the peptidoglycan, and are not the result of the formation of separate complexes. We reported previously that the TolA–Pal interaction was dependent on the proton motive force (pmf) (Cascales et al., 2000). We investigated the effect of the dissipation of the pmf by the protonophore carbonyl cyanide m-chlorophenylhydrazone (CCCP) on the protein co-immunoprecipitation with PalFLAG. Besides the absence of interaction with TolA, as observed previously, the TolB–Pal and OmpA–Pal interactions remained unaffected by CCCP treatment (data not shown).

A TolA box is well conserved among the various Pal sequences

The TolA–Pal interaction has been shown previously to require the C-terminal domain of TolA (Cascales et al., 2000) and, here, we demonstrated that the 30 C-terminal residues of Pal are required for the interaction with TolA. As Pal sequences of Gram-negative bacteria present a well-conserved peptidoglycan binding motif (Koebnik, 1995), we also analysed the Pal sequence to identify other conserved residues. The sequence alignments of the conserved regions of various outer membrane proteins belonging to the whole subdivisions of the Gram-negative bacteria were performed (see Supplementary material, Fig. S1). Three groups of peptidoglycan-associated proteins were differentiated, which correspond to Pal, peptidoglycan binding proteins (Pbp) and MotB. The Pal group corresponds to defined and probable Pal proteins (outer membrane proteins of less than 190 residues harbouring a signal sequence followed by a cysteine and a +2 residue other than aspartate). The Pbp group refers to membrane proteins harbouring the peptidoglycan binding motif and excluding Pal and MotB protein sequences. In this group were included E. coli OmpA and OmpA homologues, YiaD, MotY and two hypothetical lipoproteins (Plp) found in the OmpA family proteins (family harbouring the OmpA-like domain; accession number PS01068). The MotB group contains the inner membrane peptidoglycan binding proteins including the Vibrio homologue PomB. The consensus sequences of the three groups were aligned and, as expected, the peptidoglycan binding motif described previously (Koebnik, 1995) could be easily identified (see Fig. 6). The C-terminal region revealed a second conserved motif formed with two adjacent Arg residues. Others conserved sequences, specific to each group, could also be defined. In the C-terminal region, a SYG(K/E) motif could be assigned in the Pal group, or GxG in the Pbp group. The region of Pal required for the interaction with OmpA was also found to contain the consensus sequence lYFxxD (l and x corresponding to aliphatic and not conserved residues). It is noteworthy to indicate that no consensus sequence could be found in the N-terminal region of the Pal family except the first cysteine residue of the processed form (see Supplementary material, Fig. S1).

image

Figure 6. Consensus sequences of the C-terminal region of various peptidoglycan binding proteins. Consensus residues of the Pal (26 proteins), Pbp (28 proteins) and MotB (11 proteins) groups were aligned after removal of the gaps present in the alignments. The peptidoglycan binding motif proposed by Koebnik (1995) is shown on the bottom line. Lower or upper case letters correspond, respectively, to residues present in the 60% or 90% consensus, whereas fully conserved residues are underlined (for more details, see Supplementary material, Fig. S1). The regions required for OmpA, TolB and TolA binding in the Pal group are shaded

Download figure to PowerPoint

The deletions of the two conserved C-terminal motifs, e.g. SYGK and KNRR, were performed on the E. coli Pal protein, and the resulting constructions were examined for outer membrane integrity and interactions. We showed that the production of PalΔSYGK or PalΔKNRR in the JC892 strain induces cells with high and intermediate levels of outer membrane defects respectively (Table 1). The PalΔKNRR and PalΔSYGK lipoproteins produced in pal cells were found to interact with the peptidoglycan, and in vivo cross-linking experiments indicated that each mutant forms dimers (data not shown). Finally, immunoprecipitations of in vivo cross-linked samples with anti-TolA antibodies demonstrated that the TolA–Pal interaction is abolished when the SYGK sequence is deleted (Fig. 7). Together, these results indicate that the SYGK motif in the C-terminal of Pal constitutes an important TolA binding sequence.

image

Figure 7. A TolA binding motif is located within the C-terminal region of Pal. A pal strain harbouring pTolA and the indicated pPal plasmids were cross-linked in vivo with formaldehyde and immunoprecipitated further with anti-TolA. The Western blotting immunodetections of Pal and TolA were performed with the anti-Pal pAb and by the overlay technique respectively.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

The Pal lipoprotein interacts with several components of the Tol–Pal system, including TolB (Bouveret et al., 1995; Ray et al., 2000), TolA (Cascales et al., 2000), or cell envelope proteins, such as OmpA or Lpp (Clavel et al., 1998; Cascales et al., 2002). In this study, we used directed deletion mutagenesis to investigate regions of the Pal protein required for its function in the maintenance of cell wall organization and for its interactions with other components of the cell envelope. Mutants were engineered to generate six defined deletions along the pal coding sequence. All six of the Pal deletion mutants possess the signal sequence and the two first amino acids of the mature protein in order to ensure their correct post-translational processing and localization. We subjected deleted Pal mutants to three tests: (i) phenotypic characterization; (ii) interaction with the murein layer; and (iii) interaction with other cell envelope proteins. A summary of the results of this study is proposed in Fig. 8. Four of the six constructs failed to complement the pal null mutation while two mutants with deletions encompassing residues 3–43 of Pal did not affect cell envelope stability. Preliminary crystallographic data for Pal indicated that the 43 N-terminal residues of Pal affect crystal growth (Abergel et al., 2001). Moreover, most secondary structure programs predict this region to be poorly structured (Fig. 8). Thus, we suggest that the N-terminal sequence acts as a linker between the acyl chains anchoring Pal to the outer membrane and the functional C-terminal region of Pal.

image

Figure 8. Schematic representation of the interacting and functional region of Pal. The E. coli Pal sequence is shown with the predicted secondary structures obtained from the psipred program (http://www.psipred.net/). α-helices and β-strands are indicated by rectangles and arrows respectively. The conserved residues (upper and lower case letters, as described in Fig. 6) of Pal sequences are indicated in bold. Binding regions of Pal required for OmpA, TolB/peptidoglycan and TolA interactions or dimerization are indicated. The phenotypes of the different mutants engineered in the study are also indicated. Pal mutations previously characterized by Clavel et al. (1998) are indicated by arrows together with their corresponding phenotypes. PG, peptidoglycan; n, not determined.

Download figure to PowerPoint

Pal interaction with the peptidoglycan

Pal deleted of residues 94–121 (Pal5) was expected to confer outer membrane defects because this region has been shown previously to interact with the murein layer and with TolB. A biochemical approach to purification of the peptidoglycan-associated fraction confirms that this region is responsible for the interaction with the cell wall, a result reported previously, by competition with a synthetic peptide (Bouveret et al., 1999) or by genetic techniques (Lazzaroni and Portalier, 1992; Clavel et al., 1998). However, besides PalFLAG5, PalFLAG6 was not found to be associated with the murein layer, whereas PalFLAG3 and PalFLAG4 interacted weakly with the murein. These results agree with previous data for Pal point mutations (G101D, R104C, S126F, G128D and R146H; Fig. 8), which were shown to prevent peptidoglycan association (Clavel et al., 1998). However, we found that the two deletions of four residues within the C-terminal domain of Pal (SYGK and KNRR) did not modify its interaction with the peptidoglycan. Thus, these point mutations (S126F, G128D and R146H), which strongly modify the electric charge, may disrupt the peptidoglycan interaction either directly or by modifying the tridimensional structure of at least part of the peptidoglycan binding region of Pal.

Pal interaction with TolB

The co-immunoprecipitation studies indicated that residues 94–121 of Pal were also involved in the interaction with TolB, as described previously (Clavel et al., 1998; Bouveret et al., 1999). Moreover, when the neighbouring residues 94–121 of both regions were removed (PalFLAG4 and PalFLAG6), the level of TolB co-immunoprecipitated was reproducibly found to be lower than with the whole Pal lipoprotein. However, a point mutation within the C-terminal sequence, E130K (Fig. 8), has been described previously to disrupt the interaction of Pal with TolB (Clavel et al., 1998). We cannot explain this discrepancy but may suggest a similar explanation to that proposed above. It has been shown that TolB interacts with OmpA in a Pal-dependent manner (Clavel et al., 1998). Our data indicate that the two interactions, Pal–OmpA and Pal–TolB, involve two different regions of Pal and are independent, so we can suggest that the Pal lipoprotein acts as a recruiting factor for TolB and OmpA.

Pal interaction with OmpA

As all the truncated Pal interact with OmpA, except PalFLAG3, the OmpA binding sequence of Pal is localized within residues 44–61. Although the peptidoglycan binding motif and the SYG and NRR motifs are conserved in all the Pal sequences available, we also observed that conserved residues are present in the OmpA binding sequence. The deletion of the VYF conserved motif within this region of Pal was also performed, and we found that cells producing PalΔVYF released periplasmic proteins and were hypersensitive to SDS. However, PalΔVYF dimerizes, interacts with the peptidoglycan and with OmpA (data not shown). Thus, the OmpA–Pal interaction may not be restricted to the VYF conserved sequence. Measurement of the affinity of the Pal–OmpA and PalΔVYF–OmpA interactions will establish further the effect of the deletion of this motif.

Pal interaction with TolA

One important result of this study reveals that the interaction between the C-terminal domain of TolA and Pal exclusively required residues 123–152 of Pal. It is noteworthy that the TolA protein has been shown previously to be stable in tolB or pal mutants whereas TolA is degraded in a tolB–pal strain (Germon et al., 2001). As the C-terminal domain of TolA interacts with both TolB and Pal (Cascales et al., 2000; Dubuisson et al., 2002; Walburger et al., 2002), these interactions together may stabilize TolA. Precise mapping of Pal and TolB interaction domains will establish whether these proteins can interact simultaneously with TolA. However, from the two C-terminal Pal motifs, SYG(K/E) and NRR, found in the Pal consensus and the results of co-immunoprecipitations, we demonstrated that the SYGK motif of Pal is required for TolA binding whereas the deletion of the KNRR motif did not prevent Pal–TolA interaction. At the phenotypic level, outer membrane defects similar to the pal strain were observed in the pal strain producing PalΔSYGK, whereas partial complementation was observed when PalΔKNRR was produced. Moreover, two point mutations conferring a pal phenotype (S126F and G128D) map within the SYGK binding motif (see Fig. 8; Clavel et al., 1998). Thus, we conclude that the TolA–Pal interaction is required for the maintenance of cell envelope integrity, and suggest that the SYGK motif represents a ‘TolA box’.

From our knowledge of the TolA interactions with the g3p protein of filamentous phage capsid (Riechmann and Holliger, 1997; Lubkowski et al., 1999) and the translocation domain of various group A pore-forming bacteriocins (Bouveret et al., 2002), we analysed these sequences for the presence of an SYG(K/E) motif. Indeed, the N-terminal domain of g3p contains a CYGT sequence in which YGT was shown to be in contact with the C-terminal domain of TolA (Lubkowski et al., 1999). Moreover, we found two motifs in the TolA binding region of the colicin A (SYNT and PYGR, between residues 52 and 97, previously mapped by Journet et al., 2001), whereas in the colicin N, alanine mutations indicated that each of the Y62, H63 and I64 residues (in the SYHI sequence) are involved in the TolA–colicin N interaction (Raggett et al., 1998; Gokce et al., 2000). It is tempting to suggest that group A pore-forming colicins parasitized the Tol–Pal system using a ‘TolA box’ similar to that found in the Pal lipoprotein. These motifs may represent the shorter sequences required for TolA binding during the import process. The action of some N-terminal domains of colicins could be to displace the TolA–Pal interaction in order to affect outer membrane integrity, thus allowing the passage of the C-terminal active domain as hypothesized previously (Deprez et al., 2002). The presence of such a conserved ‘TolA box’ in different interacting partners is reminiscent of the situation in group B colicins harbouring a TonB box that is also present in TonB-dependent receptors (Gudmundsdottir et al., 1989) and required for TonB binding (Cadieux and Kadner, 1999). We are currently analysing the role of these motifs in outer membrane integrity and in the Tol-dependent phage DNA and colicin import.

Pal interacts independently with TolA, TolB and OmpA

The immunoprecipitation studies with truncated Pal derivatives show that the interactions between Pal and its different partners are independent. This may indicate that some regions of Pal can be deleted without inducing strong modifications in the Pal structure. However, the peptidoglycan binding results suggest that the long potential alpha-helix of Pal is accessible only when the whole molecule, except the linker region, is present, or that additional residues present in the neighbouring regions of the alpha-helix stabilize its interaction with the peptidoglycan. Recent crystallographic results indicated that the VYF, SYGK and KNRR sequences of Pal are present within exposed loops or beta-sheet structures and that the peptidoglycan binding motif is contained, as expected, within a long alpha-helix (C. Abergel, A. Walburger, E. Bouveret and J.M. Claverie, in prep.). We also observed that the different deletions between residue 3 and 122 and of the VYF, SYGK and KNNR sequences had no effect on Pal dimerization, whereas homodimer was not detected upon deletion of the 30 C-terminal residues, suggesting that the dimerization region of Pal is located in the C-terminal sequence. Because the PalFLAG6 protein still interacts with OmpA and TolB, we can conclude that the dimerization of Pal is not required for the interaction of Pal with these two partners. Conversely, as PalFLAG3 or PalFLAG5 dimerizes, OmpA and TolB are not required for the dimerization process of Pal, a result obtained previously with null mutations (Cascales et al., 2002). The interaction of Pal with Lpp could be faintly immunodetected with PalFLAG (Fig. 3), PalFLAG1 and PalFLAG2 (data not shown) but was not mapped further in this study. Indeed, it remains to establish whether this interaction is not only dependent on the statistical interaction of two molecules, one being highly abundant (Cascales et al., 2002).

Concluding remarks

The Tol–Pal system forms a trans-envelope bridge linking inner and outer membranes and the peptidoglycan layer. This characteristic could be the result of the TolA–Pal interaction, an interaction that depends on the pmf, and of the TolQ and TolR inner membrane proteins (Cascales et al., 2000; 2001). Experiments to clarify the role of these two accessory proteins in the molecular mechanism of energy coupling are currently under way. We suggest two functions for the Tol–Pal apparatus. First, the Tol–Pal system may contribute to maintaining cell envelope integrity through the interactions between the TolB and Pal proteins with the murein layer and the OmpA and Lpp proteins. In support of this hypothesis, we observed in this study that Pal still interacts with TolB and OmpA after dissipation of the pmf (see Fig. 9). However, because of the location of the tol–pal cluster and of the ompA and lpp genes on the chromosome, it seems difficult to argue that the products of these genes can act in synergy. Moreover, we cannot restrict the function of the Tol–Pal proteins to a structural or architectural role. Alternatively, the Tol–Pal system might play a more dynamic function in cell envelope biogenesis. Consistent with this hypothesis, both TolA and TolB interact in vitro with outer membrane porins (Derouiche et al., 1996; Rigal et al., 1997), which requires de novo synthesis of LPS. In addition, TolA is required in the surface expression of O-antigen and LPS in E. coli (Gaspar et al., 2000). Moreover, the Tol proteins and LPS are required for gliding motility of Myxococcus xanthus (Bowden and Kaplan, 1998; Youderian et al., 2003). Thus, the Tol–Pal apparatus might function to bring inner and outer membranes in close proximity via the TolA C-terminal domain–Pal interaction. The central domain of TolA, which forms an extended amphipatic α-helix (Derouiche et al., 1999; Witty et al., 2002), could be required for the transport of outer membrane components through the aqueous periplasm to reach the inner leaflet of the outer membrane. Nevertheless, it remains to be determined whether the pmf-dependent energization of the Tol–Pal system is directly involved in maintaining the cell wall architecture or in some other process of outer membrane biogenesis.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Bacterial strains, growth conditions and chemicals

The source and relevant characteristics of the E. coli strains used in this study are listed in Table 2. Cells were usually grown at 37°C in L-broth medium. For the lpp–ompA strain (JC8963), 10 mM Mg2+ was added to the medium. When required, ampicillin (100 µg ml−1) and chloramphenicol (40 µg ml−1) were added. Cells harbouring the pTolA plasmid were induced with arabinose (0.5 mg ml−1) for 30 min. For pmf dissipation, CCCP (Sigma) was added to a final concentration of 10 µM.

Table 2. Bacterial strains and plasmids.
StrainsRelevant characteristicsReferences
  1. pPalFLAG1 to pPalFLAG6 (this study) correspond to pPal1 to pPal6 with pal deleted from codons indicated as in the corresponding pPal derivatives.

1292 glnV hsdS met gal lacY fhuA W. Wood
JC80561292 ΔlacU169 Clavel et al. (1998)
JC892JC8056 pal892 (stop after codon 41) Clavel et al. (1998)
JC89311292 ompA::TnlacZ Clavel et al. (1998)
JE5505JE5506 lpp5508 Suzuki et al. (1978)
JC8963JE5505 ompA::TnlacZ Clavel et al. (1998)
JC77821292 tolA (stop after codon 40) Bernadac et al. (1998)
JC34171292 tolB (stop after codon 329) Isnard et al. (1994)
Plasmids
 pTolApACY184 vector, tolA, CmR Bernadac et al. (1998)
 pT7pT7-7 vector, AmpR Tabor and Richardson (1985)
 pPalpT7-7 vector, pal, AmpR Cascales et al. (2002)
 pPal1pPal with pal deleted from codon 3 to 17This study
 pPal2pPal with pal deleted from codon 19 to 43This study
 pPal3pPal with pal deleted from codon 42 to 66This study
 pPal4pPal with pal deleted from codon 62 to 93This study
 pPal5pPal with pal deleted from codon 94 to 121This study
 pPal6pPal with pal deleted from codon 123 to 152This study
 pPalΔSYGKpPal with pal deleted from codon 126 to 129This study
 pPalΔKNRRpPal with pal deleted from codon 144 to 147This study
 pPalΔVYFpPal with pal deleted from codon 50 to 52This study
 pPalFLAGpPal, pal fused to FLAG coding sequenceThis study

Plasmids and DNA techniques

pPal (Cascales et al., 2002) was used as a template to generate a series of defined deletions in the pal coding sequence. The pPal2 plasmid was constructed by deleting the NaeI–PvuII DNA fragment of pPal. The other deletions were constructed by recombinant polymerase chain reaction (RPCR), as described previously (Ansaldi et al., 1996), using oligonucleotides complementary to the DNA sequences flanking the region to be deleted. The RPCR, using A3-17 and B3-17 primers, inserted an NaeI restriction site in pPal and, after deleting the NaeI DNA fragment, the ligation gave the pPal1 plasmid. pPal3, pPal4 and pPal5 were constructed by RPCR using the primers described in the Supplementary material (Table S1). pPal6 was constructed by insertion of a stop codon after residue 122 of Pal. The FLAG epitope (corresponding to the DYKDDDDK sequence) was inserted at the carboxy-terminus by RPCR using Aflag and Bflag primers and either pPal and pPal1–pPal5 as templates or Cflag and Dflag primers and pPal6 as template. pPalΔSYGK, pPalΔKNNR and pPalΔVYF were constructed by RPCR using the A- and B-specific primers, sygk, knnr and vyf. PCR and RPCR amplifications were performed using Expand PCR polymerases (Roche). All plasmid constructs were checked by DNA sequencing using the Thermosequenase radiolabelled terminator cycle kit (Amersham Life Science).

Permeability and vesicle assays

Periplasmic RNase I leakage was estimated from LB plates containing 1.5% yeast RNA (Sigma Chemical) after overnight growth, by the addition of 10% trichloroacetic acid, according to the method described previously (Lazzaroni and Portalier, 1981). Tolerance for SDS or vancomycin was tested in liquid medium as described elsewhere (Cascales et al., 2000). Outer membrane blebs were observed by electron microscopy after uranyl acetate staining (Bernadac et al., 1998).

Chemical FA cross-linking

In vivo paraformaldehyde (FA) cross-linking was performed as described by Bouveret et al. (1995). Briefly, exponentially growing cells were washed and suspended in 10 mM sodium phosphate buffer (pH 6.3) to an A600 of 0.5. After the addition of 1% formaldehyde (Merck) and incubation for 20 min at room temperature, the pellet was solubilized in Laemmli buffer containing benzonase (FA; Merck) and incubated further for 20 min at 37°C. In vivo DST cross-linking was performed as described previously (Cascales et al., 2002) with some modifications. Exponentially growing cells were washed, resuspended in 20 mM sodium phosphate buffer (pH 7.5), 150 mM NaCl to an A600 of 0.4, and 5 mM DST (Pierce) was also added. Samples were incubated at room temperature for 60 min without shaking. Cross-linked cells were harvested by centrifugation, suspended in loading buffer and heated for 10 min at 96°C before SDS-PAGE analyses.

Cellular fractionation

Cell envelopes were isolated by centrifugation at 200 000 g after ultrasonic disintegration of exponential cultures. Inner and outer membranes were prepared according to the method of de Cock et al. (2001). Exponentially growing cells were harvested on ice and suspended in K buffer (50 mM triethanolamine acetate, pH 7.5, 1 mM EDTA, 1 mM dithiothreitol, 250 mM sucrose) and then sonicated further. After 5 min of 105 000 g ultracentrifugation, the pellet and supernatant constitute, respectively, the outer and inner membrane fractions. Then, the inner membrane was recovered by 105 000 g centrifugation for 90 min.

Purification of peptidoglycan-associated proteins

The procedure described by Mizuno (1979) was used in this study with some modifications. Cells were suspended in SDS extraction buffer (10 mM Tris-HCl, pH 8.0, 100 mM NaCl, 10% glycerol, 2% SDS) and then incubated at 37°C for 60 min with shaking (300 r.p.m.). After centrifugation (1 h at 125 000 g), solubilized fractions were TCA precipitated. Insoluble pellets were washed with the same buffer at room temperature and, as for the peptidoglycan-precipitated fractions, were suspended in loading buffer and heated for 15 min at 96°C before SDS-PAGE analyses.

Immunoprecipitation experiments

Samples, treated with formaldehyde, were solubilized at 37°C for 15 min in TES (10 mM Tris-HCl, pH 7.5, 5 mM EDTA, 1% SDS) supplemented with protease inhibitors (Complete EDTA-free; Roche). Then, samples were diluted 15-fold in TNE (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA) supplemented with 1% Triton X-100. After incubation at room temperature with shaking, the extract was centrifuged for 15 min at 18 000 g in order to remove unsolubilized material. Then, the supernatant was incubated for 60 min at room temperature with protein A–Sepharose CL-4B (Pharmacia) in order to remove unspecific binding components. The unbound samples were recovered by centrifugation and incubated further for 60 min at room temperature with antibodies coupled to protein A–Sepharose CL-4B.

Protein G–agarose beads coupled to the monoclonal anti-FLAG-M2 antibody (Kodak) were equilibrated by four cycles of TBS buffer (pH 8.0) and 100 mM glycine (pH 3.5) as recommended by the manufacturer. Beads were washed once with TBS buffer, once with TNE (0.1% Triton X-100) and twice with TNE (1% Triton X-100) before incubation with solubilized cross-linked samples for 60 min at room temperature. Beads were then washed twice with TNE supplemented with 1% Triton X-100 and once with TNE supplemented with 0.1% Triton X-100. Immunoprecipitates were suspended in loading buffer, heated at 96°C for 20 min and separated on 12.5% SDS-polyacrylamide gels.

Antibodies

Anti-TolAII–III (Derouiche et al., 1995), anti-TolAIII (Derouiche et al., 1996), anti-Pal (Bouveret et al., 1999), anti-TolB–Pal (Bouveret et al., 1995), anti-OmpA (Cascales et al., 2002) and anti-TolR (Journet et al., 1999) polyclonal antibodies and anti-1C11 monoclonal antibody (Cavard et al., 1986) have been described previously. Anti-FLAG monoclonal antibody was purchased from Kodak.

Miscellaneous

SDS-PAGE, Western blotting and immunodetections were performed as described previously (Cascales et al., 2000). Alkaline phosphatase-conjugated secondary antibodies were revealed using bromo-chloro-indolyl phosphate (BCIP) and nitroblue tetrazolium (NBT). Horseradish peroxidase-conjugated secondary antibodies were revealed by the chemiluminescent system (Pierce). The overlay technique, using the N-terminal domain of colicin A and the 1C11 monoclonal antibody, was used for the immunodetection of TolA (Bénédetti et al., 1991). Prestained protein markers (See-Blue, Novex) were used.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

We are grateful to Alain Bernadac for electron microscopic analyses, Stephanie Pommier and Marthe Gavioli for technical help, Dr Emmanuelle Bouveret, Professor James Sturgis and Dr Peter J. Christie for useful discussions and critical reading of the manuscript. This work was supported by grants from the Centre National de la Recherche Scientifique, Vaincre la Mucoviscidose and a Ministère de l’Enseignement de la Recherche et de la Technologie fellowship to E.C.

Supplementary material

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Fig. S1. Sequence alignments of Pal, peptidoglycan-binding proteins (Pbp) and MotB.

Table S1. Oligonucleotide sequences.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information
  • Abergel, C., Walburger, A., Chenivesse, S., and Lazdunski, C. (2001) Crystallization and preliminary crystallographic study of the peptidoglycan-associated lipoprotein from E. coli. Acta Crystallogr 0: 317319.
  • Ansaldi, M., Lepelletier, M., and Méjean, V. (1996) Site-specific mutagenesis by using an accurate recombinant polymerase chain reaction method. Anal Biochem 234: 110111.
  • Bénédetti, H., Lazdunski, C., and Lloubès, R. (1991) Protein import into E. coli: colicins A and E1 interact with a component of their translocation system. EMBO J 10: 19891995.
  • Bernadac, A., Gavioli, M., Lazzaroni, J.C., Raina, S., and Lloubès, R. (1998) Escherichia coli tol-pal mutants form outer membrane vesicles. J Bacteriol 180: 48724878.
  • Blair, D., Kim, D., and Berg, H. (1991) Mutant MotB proteins in Escherichia coli. J Bacteriol 173: 40494055.
  • Bouveret, E., Derouiche, R., Rigal, A., Lloubès, R., Lazdunski, C., and Bénédetti, H. (1995) Peptidoglycan associated lipoprotein–TolB interaction. J Biol Chem 270: 1107111077.
  • Bouveret, E., Bénédetti, H., Rigal, A., Loret, E., and Lazdunski, C. (1999) In vitro characterization of peptidoglycan-associated lipoprotein (Pal)–peptidoglycan and Pal–TolB interactions. J Bacteriol 181: 63066311.
  • Bouveret, E., Journet, L., Walburger, A., Cascales, E., Bénédetti, H., and Lloubès, R. (2002) Analysis of the E. coli Tol-Pal and TonB systems by periplasmic production of Tol, TonB, colicin, or phage capsid soluble domains. Biochimie 84: 413421.
  • Bowden, M., and Kaplan, H. (1998) The Myxococcus xanthus lipopolysaccharide O-antigen is required for social motility and multicellular development. Mol Microbiol 30: 275284.
  • Braun, V. (1975) Covalent lipoprotein from the outer membrane of E. coli. Biochim Biophys Acta 415: 335377.
  • Braun, V., and Rehn, K. (1969) Chemical characterization, spatial distribution and function of a lipoprotein (murein lipoprotein) of the Escherichia coli cell wall. The specific effect of trypsin on the membrane structure. Eur J Biochem 10: 426438.
  • Cadieux, N., and Kadner, R. (1999) Site-directed disulfide bonding reveals an interaction site between energy-coupling protein TonB and BtuB, the outer membrane cobalamin transporter. Proc Natl Acad Sci USA 96: 1067310678.
  • Cascales, E., Gavioli, M., Sturgis, J.N., and Lloubès, R. (2000) Proton motive force drives the interaction of the inner membrane TolA and outer membrane Pal proteins in E. coli. Mol Microbiol 38: 904915.
  • Cascales, E., Lloubès, R., and Sturgis, J.N. (2001) The TolQ–TolR proteins energise TolA and share homologies with the flagellar motor proteins MotA–MotB. Mol Microbiol 42: 795807.
  • Cascales, E., Bernadac, A., Gavioli, M., Lazzaroni, J.C., and Lloubès, R. (2002) Pal lipoprotein of E. coli plays a major role in outer membrane integrity. J Bacteriol 184: 754759.
  • Cavard, D., Crozel, V., Gorvel, J.P., Pattus, F., Baty, D., and Lazdunski, C. (1986) A molecular, genetic and immunological approach to the functioning of colicin A, a pore-forming protein. J Mol Biol 187: 449459.
  • Choi, D.S., Yamada, H., Mizuno, T., and Mizushima, S. (1986) Trimeric structure and localization of the major lipoprotein in the cell surface of E. coli. J Biol Chem 261: 89538957.
  • Clavel, T., Germon, P., Vianney, A., Portalier, R., and Lazzaroni, J.C. (1998) TolB protein of E. coli K-12 interacts with the outer membrane peptidoglycan-associated proteins Pal, Lpp and OmpA. Mol Microbiol 29: 359367.
  • De Cock, H., Pasveer, M., Tommassen, J., and Bouveret, E. (2001) Identification of phospholipids as new components that assist in the in vitro trimerization of a bacterial pore protein. Eur J Biochem 268: 865875.
  • De Mot, R., and Vanderleyden, J. (1994) The C-terminal sequence conservation between OmpA-related outer membrane proteins and MotB suggests a common function in both gram-positive and gram-negative bacteria, possibly in the interaction of these domains with peptidoglycan. Mol Microbiol 12: 333334.
  • Deprez, C., Blanchard, L., Guerlesquin, F., Gavioli, M., Simorre, J.P., Lazdunski, C., et al. (2002) Macromolecular import into E. coli: the TolA C-terminal domain changes conformation when interacting with the colicin A toxin. Biochemistry 41: 25892598.
  • Derouiche, R., Bénédetti, H., Lazzaroni, J.C., Lazdunski, C., and Lloubès, R. (1995) Protein complex within E. coli inner membrane. J Biol Chem 270: 1107811084.
  • Derouiche, R., Gavioli, M., Bénédetti, H., Prilipov, A., Lazdunski, C., and Lloubès, R. (1996) TolA central domain interacts with E. coli porins. EMBO J 15: 64086415.
  • Derouiche, R., Lloubès, R., Sasso, S., Bouteille, H., Oughideni, R., Lazdunski, C., and Loret, E. (1999) Circular dichroism and molecular modeling of the E. coli TolA periplasmic domains. Biospectroscopy 5: 189198.
  • Dubuisson, J.F., Vianney, A., and Lazzaroni, J.C. (2002) Mutational analysis of the TolA C-terminal domain of E. coli and genetic evidence for an interaction between TolA and TolB. J Bacteriol 184: 46204625.
  • Gaspar, J., Thomas, J., Marolda, C., and Valvano, M. (2000) Surface expression of O-specific lipopolysaccharide in E. coli requires the function of the TolA protein. Mol Microbiol 38: 262275.
  • Gennity, J.M., and Inouye, M. (1991) The protein sequence responsible for lipoprotein membrane localization in E. coli exhibits remarkable specificity. J Biol Chem 266: 1645816464.
  • Germon, P., Clavel, T., Vianney, A., Portalier, R., and Lazzaroni, J.-C. (1998) Mutational analyses of the E. coli K-12 TolA N-terminal region and characterization of its TolQ-interacting domain by genetic suppression. J Bacteriol 180: 64336439.
  • Germon, P., Ray, M.C., Vianney, A., and Lazzaroni, J.C. (2001) Energy-dependent conformational change in the TolA protein of E. coli involves its N-terminal domain, TolQ and TolR. J Bacteriol 183: 41104114.
  • Gokce, I., Raggett, E., Hong, Q., Virden, R., Cooper, A., and Lakey, J. (2000) The TolA-recognition site of colicin N. ITC, SPR and stopped-flow fluorescence define a crucial 27-residue segment. J Mol Biol 304: 621632.
  • Gudmundsdottir, A., Bell, P., Lundrigan, M., Bradbeer, C., and Kadner, R. (1989) Point mutations in a conserved region (TonB box) of Escherichia coli outer membrane protein BtuB affect vitamin B12 transport. J Bacteriol 171: 65266533.
  • Isnard, M., Rigal, A., Lazzaroni, J.C., Lazdunski, C., and Lloubès, R. (1994) Maturation and localization of the TolB protein required for colicin import. J Bacteriol 176: 63926396.
  • Journet, L., Rigal, A., Lazdunski, C., and Bénédetti, H. (1999) Role of TolR N-terminal, central and C-terminal domains in its dimerisation and interaction with TolA and TolQ. J Bacteriol 181: 44764484.
  • Journet, L., Bouveret, E., Rigal, A., Lloubès, R., Lazdunski, C., and Bénédetti, H. (2001) The translocation domain of colicin A interacts with at least three different proteins of the Tol system by three distinct regions. Mol Microbiol 42: 331344.
  • Koebnik, R. (1995) Proposal for a peptidoglycan-associating alpha-helical motif in the C-terminal regions of some bacterial cell-surface proteins. Mol Microbiol 16: 12691270.
  • Koebnik, R., Locher, K., and Van Gelder, P. (2000) Structure and function of bacterial outer membrane proteins: barrels in a nutshell. Mol Microbiol 37: 239253.
  • Lazdunski, C., Bouveret, E., Rigal, A., Journet, L., Lloubès, R., and Bénédetti, H. (1998) Colicin import into Escherichia coli cells. J Bacteriol 180: 49935002.
  • Lazzaroni, J.C., and Portalier, R. (1981) Genetic and biochemical characterization of periplasmic-leaky mutants of E. coli K-12. J Bacteriol 145: 13511358.
  • Lazzaroni, J.C., and Portalier, R. (1992) The excC gene of E. coli K-12 required for cell envelope integrity encodes the peptidoglycan-associated lipoprotein (Pal). Mol Microbiol 6: 735742.
  • Lazzaroni, J.C., Vianney, A., Popot, J.L., Bénédetti, H., Samatey, F., Lazdunski, C., et al. (1995) Transmembrane α-helix interactions are required for the functional assembly of the E. coli Tol complex. J Mol Biol 246: 17.
  • Leduc, M., Ishidate, K., Shakibai, N., and Rothfield, L. (1992) Interactions of E. coli membrane lipoproteins with the murein sacculus. J Bacteriol 174: 79827988.
  • Lubkowski, J., Hennecke, F., Plückthun, A., and Wlodawer, A. (1999) Filamentous phage infection: crystal structure of g3p in complex with its coreceptor, the C-terminal domain of TolA. Struct Fold Des 7: 711722.
  • Matsuyama, S., Tajima, T., and Tokuda, H. (1995) A novel periplasmic carrier protein involved in the sorting and transport of Escherichia coli lipoproteins destined for the outer membrane. EMBO J 14: 33653372.
  • Matsuyama, S., Yokota, N., and Tokuda, H. (1997) A novel outer membrane lipoprotein, LolB (HemM), involved in the LolA (p20)-dependent localization of lipoproteins to the outer membrane of Escherichia coli. EMBO J 16: 69476955.
  • Mizuno, T. (1979) A novel peptidoglycan-associated lipoprotein found in the cell envelope of Pseudomonas aeruginosa and Escherichia coli. J Biochem (Tokyo) 86: 9911000.
  • Nagel de Zwaig, R., and Luria, S. (1967) Genetics and physiology of colicin-tolerant mutants of E. coli. J Bacteriol 94: 11121123.
  • Nomura, M., and Witten, C. (1967) Interaction of colicins with bacterial cells. 3. Colicin-tolerant mutations in Escherichia coli. J Bacteriol 94: 10931111.
  • Palva, E. (1979) Protein interactions in the outer membrane of E. coli. Eur J Biochem 93: 495503.
  • Raetz, C. (1996) Bacterial lipopolysaccharides: a remarkable family of bioactive macroamphiphiles. In Escherichia coli and Salmonella: Cellular and Molecular Biology. Neidhardt, F.C. (ed.). Washington, DC: American Society for Microbiology Press.
  • Raggett, E., Bainbridge, G., Evans, L., Cooper, A., and Lakey, J. (1998) Discovery of critical Tol A-binding residues in the bactericidal toxin colicin N: a biophysical approach. Mol Microbiol 28: 13351343.
  • Ray, M.C., Germon, P., Vianney, A., Portalier, R., and Lazzaroni, J.-C. (2000) Identification by genetic suppression of Escherichia coli TolB residues important for TolB–Pal interaction. J Bacteriol 182: 821824.
  • Riechmann, L., and Holliger, P. (1997) The C-terminal domain of TolA is the coreceptor for filamentous phage infection of E. coli. Cell 90: 351360.
  • Rigal, A., Bouveret, E., Lloubès, R., Lazdunski, C., and Bénédetti, H. (1997) The TolB protein interacts with the porins of E. coli. J Bacteriol 179: 72747279.
  • Shu, W., Liu, J., Ji, H., and Lu, M. (2000) Core structure of the outer membrane lipoprotein from Escherichia coli at 1.9 Å resolution. J Mol Biol 299: 11011112.
  • Sturgis, J.N. (2001) Organization and evolution of the tol–pal gene cluster. J Mol Microbiol Biotechnol 3: 113122.
  • Sun, T., and Webster, R. (1987) Nucleotide sequence of a gene cluster involved in entry of E colicins and single-stranded DNA of infecting filamentous bacteriophages into E. coli. J Bacteriol 169: 26672674.
  • Suzuki, H., Nishimura, Y., Yasuda, S., Nishimura, A., Yamada, M., and Hirota, Y. (1978) Murein-lipoprotein of E. coli: a protein involved in the stabilization of bacterial cell envelope. Mol Gen Genet 167: 19.
  • Tabor, S., and Richardson, C. (1985) A bacteriophage T7 RNA polymerase/promoter system for controlled exclusive expression of specific genes. Proc Natl Acad Sci USA 82: 10741078.
  • Vianney, A., Muller, M., Clavel, T., Lazzaroni, J.C., Portalier, R., and Webster, R. (1996) Characterization of the tol-pal region of E. coli K-12: translational control of tolR expression by TolQ and identification of a new open reading frame downstream of pal encoding a periplasmic protein. J Bacteriol 178: 40314038.
  • Walburger, A., Lazdunski, C., and Corda, Y. (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: 695708.
  • Witty, M., Sanz, C., Shah, A., Grossmann, J., Mizuguchi, K., Perham, R., and Luisi, B. (2002) Structure of the periplasmic domain of Pseudomonas aeruginosa TolA: evidence for an evolutionary relationship with the TonB transporter protein. EMBO J 21: 42074218.
  • Yakushi, T., Masuda, K., Narita, S., Matsuyama, S., and Tokuda, H. (2000) A new ABC transporter mediating the detachment of lipid-modified proteins from membranes. Nature Cell Biol 2: 212218.
  • Yem, D., and Wu, H. (1978) Physiological characterization of an E. coli mutant altered in the structure of murein lipoprotein. J Bacteriol 133: 14181426.
  • Yokota, N., Koruda, T., Matsuyama, S., and Tokuda, H. (1999) Characterization of the LolA-LolB system as the general lipoprotein localization mechanism of E. coli. J Biol Chem 274: 3099530999.
  • Youderian, P., Burke, N., White, D., and Hartzell, P. (2003) Identification of genes required for adventurous gliding motility in Myxococcus xanthus with the transposable element mariner. Mol Microbiol 49: 555570.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary material
  9. References
  10. Supporting Information

Table S1. Oligonucleotide sequences. Fig. S1. Sequence alignments of Pal, peptidoglycan binding protins (Pbp) and MotB. The sequence alignments of the conserved regions of outer membrane proteins from various Gram-negative bacteria were obtained using ClustalW multiple alignment (http://npsa-pbil.ibcp.fr) and Consensus (http://www.bork.embl-heidelberg.de/cgi/consensus) programs. Consensus sequences at the 60% and 90% level and conserved sequences (100%) are shown. Lower case letters correspond to: o alcohol, l aliphatic, a aromatic, and c charged residues present in the consensus. The alignments of the C-terminal regions of Pbp and MotB proteins are shown. Protein accession numbers are indicated. The protein sequences were from the following bacteria: Gram-negative bacteria, Borrelia burgdorferi, Chlamydia trachomatis, C. muridarum, Chlamydophila pneumoniae, Treponema pallidium. Various proteobacteria were used: a subdivision, Agrobacterium tumefaciens, Brucella melitensis, Caulobacter crescentus, Rickettsiae conorii, R. prowazekii, Rhodospirillum centenum, Sinorhizobium meliloti; b subdivision: Bordetella avium, Neisseria gonorrhoeae, N. meningitidis, Ralstonia solanacearum; de subdivision, Campylobacter jejuni, C. coli, Helicobacter pylori, H. hepaticus; g subdivision, Actinobacillus pleuropneumoniae, Citrobacter freundii, Enterobacter aerogenes, Erwinia chrysanthemi, Escherichia coli (K-12 or O157), E. hermanii, Haemophilus influenzae, H. ducreyi, Klebsiella pneumoniae, Legionella pneumophila, Pasteurella multocida, Pseudomonas aeruginosa, P. fluorescens, P. putida, P. syringae, Salmonella typhi and S. typhimurium, Serratia marcescens, Vibrio cholerae, V. parahaemolyticus, Xylella fastidiosa, Yersinia pestis.

FilenameFormatSizeDescription
MMI_3881_sm_TableS1.doc21KSupporting info item
MMI_3881_sm_FigS1.doc47KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.