Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M706846200 on December 14, 2007

J. Biol. Chem., Vol. 283, Issue 8, 4993-5003, February 22, 2008
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
283/8/4993    most recent
M706846200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mora, L.
Right arrow Articles by de Zamaroczy, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mora, L.
Right arrow Articles by de Zamaroczy, M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Dual Roles of the Central Domain of Colicin D tRNase in TonB-mediated Import and in Immunity*

Liliana Mora, Mirjam Klepsch, Richard H. Buckingham, Valérie Heurgué-Hamard, Stephanie Kervestin, and Miklos de Zamaroczy1

From the CNRS, UPR 9073, Institut de Biologie Physico-Chimique, 75005 Paris, France

Received for publication, August 16, 2007 , and in revised form, December 10, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Colicin D import into Escherichia coli requires an interaction via its TonB box with the energy transducer TonB. Colicin D cytotoxicity is inhibited by specific tonB mutations, but it is restored by suppressor mutations in the TonB box. Here we report that there is a second site of interaction between TonB and colicin D, which is dependent upon a 45-amino acid region, within the uncharacterized central domain of colicin D. In addition, the 8th amino acids of colicin D (a glycine) and colicin B (a valine), adjacent to their TonB boxes, are also required for TonB recognition, suggesting that high affinity complex formation involves multiple interactions between these colicins and TonB. The central domain also contributes to the formation of the immunity complex, as well as being essential for uptake and thus killing. Colicin D is normally secreted in association with the immunity protein, and this complex involves the following two interactions: a major interaction with the C-terminal tRNase domain and a second interaction involving the central domain of colicin D and, most probably, the {alpha}4 helix of ImmD, which is on the opposite side of ImmD compared with the major interface. In contrast, formation of the immunity complex with the processed cytotoxic domain, the form expected to be found in the cytoplasm after colicin D uptake, requires only the major interaction. Klebicin D has, like colicin D, a ribonuclease activity toward tRNAArg and a central domain, which can form a complex with ImmD but which does not function in TonB-mediated transport.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Colicins are plasmid-encoded cytotoxins released by Escherichia coli under conditions of stress, which kill competing bacteria that are not immune to their effect. Colicinogenic strains are resistant to both exogenous and endogenous colicin molecules because of the constitutive production of an immunity protein (Imm) (1). Most colicins have a similar organization of functional domains as follows: an N-terminal part required for translocation across the outer membrane, a central region necessary for binding to the target cell surface receptor and a C-terminal domain encoding the killing function (2, 3). The existence of many natural chimeric colicins and the possibility of obtaining hybrid colicins by recombination suggest that colicin evolution is in part based on the exchange and mixing of similar domains from heterologous colicins or bacteriophages (2, 4-6). The major killing functions are either a nuclease activity (RNase or DNase), which functions in the cytoplasm, or a lethal pore-forming function that affects the inner membrane. Colicin B is an ionophoric toxin. Colicin D is an RNase, which cleaves specifically each of the four arginine isoaccepting tRNAs, thereby preventing protein synthesis (7). The crystal structure of the heterodimer complex formed between the colicin D cytotoxic domain and its cognate immunity protein showed that ImmD mimics the tRNA substrate backbone, thereby protecting colicin-producer cells against colicin molecules (8). Although they have different lethal activities, colicins B and D are 96% identical in their 313-residue-long N-terminal domains, which are required for entry into target cells (7, 9, 10). Both colicins parasitize the same outer membrane, high affinity iron-siderophore receptor FepA and use the energy-transducing TonB/ExbB-D pathway to enter the cell. Thus, they make use of natural functions of the target cell before killing it (11-13). The concept of a combined N-terminal domain, required for receptor binding and translocation, is supported by the crystal structure of colicin B (55 kDa), in which a single 74-Å-long helix separates the N-terminal domain and C-terminal killing domain (14). However, in comparison with colicin B, the 75-kDa colicin D possesses an additional specific central domain (CD)2 of 280 amino acids of unknown function (10). The recently reported phage-associated klebicin D, produced by Klebsiella oxytoca, was suggested to be a member of the family of tRNase bacteriocins on the basis of the similarity of its killing domain with that of the tRNase colicin D. Interestingly, the similarity includes its central region, which is reminiscent of the CD of colicin D (4).

The complex TonB-ExbB-D transduces conformationally stored potential energy, derived from the proton motive force of the inner membrane, to the outer membrane receptor FepA, which is responsible for the uptake of an iron-chelator complex, the ferric-enterobactin (15-18). Active transport is driven by the TonB protein interacting with the "TonB box" sequence, located near the N terminus of the FepA cork domain (19, 20). For several TonB-dependent outer membrane receptors, like BtuB, FhuA, and FecA, substrate binding enhances their interaction with TonB (21-23). A conformational change was shown to occur in the region of the TonB box, when the receptors were liganded with their substrates, probably as a result of exposing the TonB box to the periplasmic space (24-28). The crystal structure of FepA is compatible with a putative release of the flexible TonB box sequence toward the periplasm, upon the binding of ferric enterobactin, thus promoting the interaction with TonB (29). Deletions of the TonB box in FhuA receptor were shown to abolish all TonB-dependent transport functions, including the sensitivity to colicin M, a toxin inhibiting murein biosynthesis (30).

"TonB box-like" consensus sequences have been described at the N terminus of TonB-dependent colicins (12, 31). Certain mutations affecting the five-amino acid TonB box of colicins B or M completely abolished their toxicity. Suppressor mutations were found at or around residue 160 in TonB, which partially restored the toxicity to some colicin B or M derivatives, carrying a TonB-uncoupling mutation in the TonB box (32-34). Spontaneous tonB mutations, R158S or P161L (mutant strains D1 and K19, respectively), completely abolished colicin D toxicity but not that of colicin B (31) nor FepA-dependent siderophore uptake. These same amino acids have recently been shown to be required for the interaction with the TonB box of the outer membrane receptor FhuA or BtuB (35, 36). The sensitivity to colicin D of these tonB mutants was fully restored by specific suppressor mutations in the TonB box (HTMVV or HSIVV), thus demonstrating that the TonB box is the functional interaction site with TonB (31).

Here we present a functional study of the CD of colicin D, as it affects the TonB-dependent import mechanism, and with regard to the formation of the colicin D-ImmD immunity complex. We show that both the in vivo toxicity and the inhibition of the catalytic activity tested in vitro are specifically dependent on the CD of colicin D. We present evidence that the CD may be required both for the recognition of the energy transducer TonB, together with the N-terminal part of the molecule, and for formation of the complex with ImmD, together with the C-terminal catalytic domain.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Bacterial Strains, Growth, and Recombinant DNA Manipulations—E. coli strains D10 (37) and XAC were used as wild-type strains. Strains D1 and K19 carry spontaneous tonB mutations (R158S and P161L, respectively), which abolish colicin D toxicity, but do not affect either the sensitivity to colicin B or the FepA-dependent siderophore uptake (31). The medium used was Luria broth (LB) with ampicillin added, if necessary at 100 µg/ml. Standard procedures for recombinant DNA techniques and PAGE were performed as described by Miller (38).

Construction of Chimeric or Internally Deleted Colicin Variants—Derivatives carrying internal deletions within colicin D were created by three-step PCR amplifications, employing plasmid pColD, a derivative of pUC18 carrying the whole colicin D operon (cda, cdi, and cdl) (31) as template. For example to create the deletion derivative, eliminating the central domain (Arg318-Met590), the first PCR used an upstream 113-mer oligonucleotide (containing the tac promoter and the cda ribosomal binding site, followed by 30 nucleotides of the cda coding sequence starting from the fMet codon) and a downstream oligonucleotide, complementary to cda and starting just before the beginning (Ala317) of the internal deletion. The second PCR amplification used an upstream oligonucleotide that is in part complementary to the downstream oligonucleotide used in the first PCR and is in-frame with the codon at the end of the deletion in cda (Leu591), together with a downstream oligonucleotide, complementary to the stop codon of cdi (encoding ImmD). A denaturation/renaturation step of the mixed first two PCR products led to an annealing between complementary strands of the common amplified central part, bridging the internal deletion. The third PCR amplification using the upstream oligo (1st PCR) and the downstream oligo (2nd PCR) led to the amplification of a linear double-stranded DNA, carrying the internal deletion, which was used as template for protein synthesis. The same three-step PCR amplification technique was used to create hybrids between colicin D/colicin B or colicin D/klebicin D, by employing the plasmid pColB (pUCB19), carrying the whole colicin B operon (cba and cbi) (31), or the plasmid pBS-KlebD, carrying the kdp kda and kdi genes of klebicin D operon (4) as templates. When necessary, plasmids carrying mutations in the TonB box of colicins D or B were used during the first PCR amplification. This allowed us to test in different N-terminal backgrounds the suppressor effect of mutations HTMVV or HSIVV in the TonB box motif, against the colicin D resistance of tonB mutants D1 (R158S) and K19 (P161L) (12, 31) when associated with other mutations (or modifications) in the colicin. The colicin D constructions were expressed in vitro, in a coupled transcription-translation, Zubay-S30 system from E. coli, optimized for linear DNA templates (Promega) at 37 °C for 60 min. The presence of ImmD was verified, and the yield of the wild-type or modified colicins was estimated as comparable (data not shown) in a similar transcription-translation assay performed in the presence of [35S]Met, followed by SDS-PAGE and phosphorimagery (GE Healthcare) (data not shown). The tRNA hydrolyzing activity of colicin D derivatives was directly measured in the S30 expression samples, by separating the intact and cleaved forms of the tRNAArg on PAGE (10% in the presence of 7 M urea), followed by Northern blot analysis using a 32P-labeled DNA probe specific for tRNAArg(CCG). In parallel, the same in vitro synthesized colicin D derivatives or hybrid colicin D-ImmD complexes were directly tested for their in vivo cytotoxic activity.

Construction of Colicins D and B Molecules Mutated around the TonB Box—Using as template the plasmids pColD(pUC18) or pColB(pUCB19), point mutations switching 1 to 4 amino acids within the N-terminal domain of the colicins were generated by QuikChange site-directed mutagenesis method (Stratagene) in the cda (encoding colicin D) or cba gene (encoding colicin B), in positions 4, 8, 27, and 29, flanking the TonB box, as well as in position 172 of the two colicin N-terminal domains. Two oligonucleotide primers containing the desired mutation were used, each complementary to opposite strands of the plasmid, to generate the mutant derivatives. The mutated DNAs were subsequently transformed into XL1-Blue super-competent cells, prepared in the presence of 10 mM RbCl. The mutated colicins were produced and then exported into the medium after induction with mitomycin C (200 µg/liter) of C600 strains. The supernatants of the induced cultures were concentrated by ultrafiltration (Amicon Ultra-4 10K NMWL; Millipore).

Construction of Colicins D Molecules Mutated in the Central Domain—Single or double point mutations were introduced into the central domain of colicin D by two-step PCR amplification as described previously (8), and they were expressed in vitro (Zubay-S30).


Figure 1
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 1.
Cytotoxicity test and tRNA hydrolyzing activity of internally deleted or truncated colicins D and B/D or D/B hybrid-colicin derivatives. A, domain structures of wild-type colicin D and B are shown. The location of residues delimiting each deletion ({Delta}) or the position of in-frame fusion points between domains of different colicins in the hybrids is given. In colicin B, Ser291-Tyr339 indicates the location of the long {alpha}-helix that connects the N-terminal domain to the C-terminal pore-forming domain. Chimeric colicins (Chi), resulting from various domain swapping experiments, are designated by two-letter code, indicating whether the domain comes from colicin B or D. The colicin domains are color-coded as follows: N-terminal domain (colicin D, gold; colicin B, black), colicin D central domain (blue), colicin D tRNase domain (red), immunity protein (ImmD, green), colicin B pore-forming domain (indigo). The sequence of the colicin TonB box peptide (amino acids 17-21) carried by each hybrid is indicated (substituted residues dependent on suppressor mutations, according to Ref. 31, are written in red). Full-sized wild-type (WT) or chimeric or internally deleted colicins (coexpressed with ImmD where indicated) were synthesized in a coupled E. coli S30 in vitro transcription/translation system. The "cytotoxicity" of in vitro synthesized bacteriocins was quantified by halo test, spotting aliquots of undiluted colicin and 4-fold serial dilution directly onto a lawn of sensitive wild-type (WT) or tonB-mutant cells (D1, carrying R158S, and K19, carrying P161L (31)). The numbers 1-4 indicate the last serial 4-fold dilution (the highest sensitivity) that resulted in a clear zone of growth inhibition. R indicates resistance (no clearing) to the undiluted colicin (numbered as 0). Values in parentheses indicate the dilution producing a turbid zone of marginal growth inhibition. The tRNA hydrolyzing activity (tRNase activity) is measured in the S30 samples by Northern blot analysis to determine the presence of intact and/or cleaved tRNAArg. Results are shown and estimated as follows: - (100% intact form, signifying 100% immunity), ++ (100% cleaved form, signifying no immunity), + (intermediate level, 40-60% cleaved form), and -/+ (weak rate, <15% of cleaved form). B, in vitro tRNase activity of truncated colicins D derivatives in the absence of the N-terminal domain. The first amino acid (following the fMet) of truncated colicins is indicated. All colicin D proteins were synthesized together with ImmD.

 
Killing Activity Test—The cytotoxicity of wild-type, hybrid, or internally deleted colicins D, as well as colicin D molecules carrying point mutations in the central domain, synthesized in vitro (Zubay-S30), was quantified in vivo by growth inhibition (halo) assay. Undiluted colicins and corresponding 4-fold serial dilutions (5 µl) were spotted onto a lawn of both sensitive wild-type D10 (or XAC) and tonB mutant cells (D1 and K19) on LB plates and then incubated overnight. (Mutated colicins, when produced in vivo (see above), are quantified through 10-fold serial dilutions.) To optimize the binding of colicin derivatives to FepA, the in vivo test was repeated in the presence of the iron chelator 2,2'-dipyridyl (200 µM), which increases FepA expression. Data were collected in triplicate and averaged.

In Vitro Analysis of Internal Cleavage of the Mutated Colicin D-ImmD Complexes by the Leader Peptidase—Mutations were introduced at residues of ImmD helix {alpha}4, by two-step PCR amplification as described previously (8). The in vitro (Zubay-S30) synthesized [35S]Met-labeled, wild-type or mutated colicin D-ImmD complexes (3 µg) were incubated in a concentrated crude cell extract of E. coli XAC wild-type strain enriched with added purified LepB (0.025 µg) for 1 h at 37 °C, as described previously by de Zamaroczy et al. (39). Full-sized and cleaved colicin D forms were precipitated with acetone and separated by (8%) SDS-PAGE, and the labeled proteins were detected and quantified by phosphorimagery (GE Healthcare).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Construction of Internally Deleted or Hybrid Colicins—The analysis of various chimeric colicins that retain toxicity shows that it is possible to construct colicins by assembly of different functional domains. However, the synthesis of certain colicin hybrids in E. coli was found to produce aggregates in the cytoplasm, or in some cases the colicin derivatives were shown to be more susceptible to degradation when their release into the extracellular medium was impaired (40). In this study we have avoided these problems by the use of hybrid or internally deleted colicins expressed in vitro, which we have then directly tested for both toxicity (in vivo) and catalytic activity (in vitro). Moreover, the in vitro synthesis allows us to test colicin derivatives that are no longer neutralized by the presence of ImmD, and which could not have been expressed in vivo. Hybrid colicin derivatives or molecules carrying internal deletions in cda (the colicin encoding structural gene) were obtained by three-step PCR amplifications, as described under "Experimental Procedures."

The Central Domain of Colicin D Is Required for Toxicity—In a first set of experiments, deletions of the CD of colicin D (amino acids 314-606) were constructed with wild-type or mutated TonB box motifs within the N-terminal domain, as depicted in Fig. 1A. The deletion of the CD of colicin D produced a stable protein {Delta}1(Arg318-Met590), which had lost all in vivo toxicity, but which retained full tRNase activity in vitro even though the ImmD protein was still synthesized. These results imply that the CD is required for the killing function of colicin D and, moreover, for both the import step and immunity. Deletions within the N-terminal half of the CD, {Delta}2(Arg318-Met444) and {Delta}4(Arg318-Glu421), exhibited a decreased level of toxicity on the wild-type strain. In contrast, deletion of the C-terminal half of the CD, {Delta}3(Leu445-Met590), had a severe negative effect on colicin D toxicity, suggesting it is the C-terminal part of the CD that is more required for the import of colicin D into wild-type cells. Although deletions {Delta}2 and {Delta}4 are still permissive for colicin D toxicity on wild-type bacteria (Fig. 1A), they are unable to kill the tonB mutant D1 and K19, even in the presence of the suppressor-type TonB box, HTMVV (31), as shown for {Delta}4 (Fig. 1A). The K19 strain, however, has a weak sensitivity to {Delta}4, carrying the suppressor-type TonB box HSIVV. Two smaller deletions in the CD were also examined. Both {Delta}5(Arg318-Val376) and {Delta}6(Trp377-Glu421) are almost fully active on wild-type bacteria. In the presence of the TonB box HSIVV {Delta}5 restored the toxicity on tonB mutant K19, like the wild-type colicin D molecule. But in the presence of a TonB box HTMVV the toxicity on tonB mutants D1 and K19 was weaker than with the wild-type colicin D molecule. In contrast, {Delta}6, even carrying a suppressor-type TonB box, is unable to kill the tonB mutants (Fig. 1A). The comparison of deletions carried by {Delta}4, {Delta}5, and {Delta}6 implies that the restoration of colicin D toxicity to mutants D1 and K19 required precisely the presence of region Trp377-Glu421 within the CD, in addition to a suppressor-type TonB box. To validate the involvement of CD in colicin D uptake, 18 residues, located between positions 377 and 421, were mutated to alanine. However, none of them exhibited a toxicity defect on tonB mutants D1 or K19 in the presence of a suppressor-type TonB box (data not shown). This was not really surprising because changes to alanine at the main interaction site of colicin B with TonB (namely affecting the residues of or around the TonB box) did not reduce the toxicity (33). Consequently, we decided to introduce more dramatic changes. For example, pairs of hydrophobic residues were mutated to charged amino acids and vice versa. As shown in Fig. 2, we identified three couples of mutations F390C/F391Y, S404R/P405D, and E420R/E421R, which almost completely abolished the toxicity on D1 and K19 despite the presence of a TonB box HTMVV. Two different couples of mutations R409E/I410D and W416E/F417D reduced the toxicity by 2 or 3 orders of magnitude on K19 in the presence of a TonB box HSIVV. In contrast, their toxicity on the wild-type strain is comparable with that of the wild-type colicin D, seeming to confirm that this region becomes important for the interaction with mutated TonB proteins.


Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 2.
Site-directed mutagenesis of the Trp377-Glu421 peptide region belonging to the N-terminal part of the central domain of colicin D. The positions of the different couples of double point mutations are indicated. Wild-type (WT) TonB box motif of colicin D is underlined. Suppressor-type TonB box sequences, HTMVV or HSIVV, present in the doubly mutated colicins D are given. The cytotoxicity of the in vitro produced mutant colicins was determined by halo test, as in Fig. 1, and quantified by spotting undiluted colicins and 4-fold serial dilution aliquots directly onto a lawn of sensitive wild-type (WT) or tonB-mutant cells, D1 and K19.

 
Functional Comparison of the Conserved Central Domains of Klebicin D and Colicin D—Protein sequence comparisons suggested that the recently identified klebicin D belonged to the tRNase family of bacteriocins and showed that it carries a CD similar to that present in colicin D (4). As depicted in Fig. 3A, the immunity proteins and the C-terminal toxic domains are conserved between klebicin D and colicin D, carrying 63.6 and 84.6% identical residues, respectively. The conservation is particularly high around the endoproteolytic cleavage site, located in vitro at position 607, just upstream of the catalytic domain of colicin D (39), suggesting the klebicin D is processed similarly to colicin D (Fig. 3A). There is no significant similarity in the N-terminal domains of colicin D and klebicin D (4), and there is no obvious TonB box sequence in the N-terminal domain of klebicin D, suggesting that it does not appropriate TonB for its translocation into Klebsiella. The in vitro expressed klebicin D, with or without its cognate immunity protein (Fig. 3B), had no toxicity on E. coli wild-type and tonB mutant strains suggesting the need for a Klebsiella-specific receptor prior to translocation. In contrast, when the whole N-terminal domain of klebicin D was replaced by that of colicin D, to give Chi3A and Chi3B (including or not ImmK) (Fig. 3B), the toxicity was fully restored on the wild-type E. coli strain, implying that in these chimeras entry does occur via the FepA/TonB import system. However, the D/K/K chimera is more active in complex with the ImmK protein (Chi3B) than in its absence (Chi3A). This is in agreement with the empirical observation that the naturally secreted form of the nuclease type colicin-Imm complexes is more toxic than colicin molecules freed in vitro of their Imm proteins (41, 42). The C-terminal part of the klebicin D CD is 56.7% identical compared with colicin D (amino acids 422-607), whereas the N-terminal amino acids 314-421 in colicin D are only weakly similar (27.3% identical residues) (Fig. 3A). Significantly, none of the hybrids carrying the CD of klebicin D in the presence of its own (Chi3B) or the colicin D (Chi4 or Chi5) tRNase domain were toxic on the tonB mutants (Fig. 3B), except for a marginal toxicity observed with Chi3B and Chi4 on K19 in the presence of the HSIVV. These results show that even if the CD is conserved between colicin D and klebicin D, they are not functionally equivalent, and this lack of complementation of toxicity on tonB mutants correlates with the absence of the Trp377-Glu421 region peptide.


Figure 3
View larger version (25K):
[in this window]
[in a new window]

  		FIGURE 3.
Structural comparison of colicin D and klebicin D, cytotoxicity, and in vitro cleavage of tRNAArg by klebicin D and hybrid colicin D/klebicin D derivatives. A, comparison of the domain structures of colicin D and klebicin D. Each peptide sequence is depicted as an N-terminal domain (colicin D, gold; klebicin D, brown), central domain (colicin D, blue; klebicin D, gray), nuclease domain (colicin D, red; klebicin D, vertical red stripes), and immunity protein (ImmD, green; ImmK, blue). Homology between equivalent domains of colicin D and klebicin D is expressed in % of identical residues (in italics). The amino acid position at the junction between successive domains (or subdomains inside the central domain) is given. The vertical line in the CDs of colicin D and klebicin D separates their N- and C-terminal parts with low and high homology. Alignment of klebicin D with the peptide sequence around the in vitro cleavage site, located at the start of the minimal tRNase domain of colicin D, is shown. B, cytotoxicity of bacteriocins, spotted directly onto a lawn of sensitive wild-type (WT) or tonB-mutant (D1, K19) cells, was quantified by halo test, as in Fig. 1. Chimeric bacteriocins, resulting from various domain swapping experiments, are designated by the three-letter code, and the position of fusion points between bacteriocin domains and the total number of amino acid residues is indicated. The in vitro tRNA hydrolyzing activities (tRNase) are estimated as in Fig. 1. C, separation of the intact and cleaved forms of tRNAArg by PAGE (10%, in the presence of 7 M urea) and analysis by Northern blot and phosphorimagery, using a 32P-labeled DNA probe specific to tRNAArg(CCG), is shown for truncated colicin D derivatives and wild-type colicin D and klebicin D.

 
Role of the TonB Box in Toxicity of Colicin B/D and D/B Hybrids—The exchange of the entire N-terminal domain of colicin D with that of colicin B produced a chimeric derivative, Chi1, that was highly active against both wild-type and tonB mutant cells (Fig. 1A), irrespective of the TonB box sequence carrying or not a suppressor mutation (data not shown). Previously, we showed that it was impossible to restore colicin D toxicity to tonB mutants D1 and K19 by swapping the sequence of the TonB box (HSMVV) with that of colicin B (DTMVV; Fig. 1A) (31). Together these results imply that it is the sequence of the rest of the colicin B type N-terminal domain that is required for toxicity of Chi1 on tonB mutants.

The chimera Chi2, carrying the colicin D N-terminal domain followed by the colicin B pore forming domain, required obligatorily the suppressor-type TonB box HSIVV to kill mutant K19, as efficiently as colicin D (Fig. 1A). In contrast, no suppressor-type TonB box was able to reverse the colicin D resistance of the D1 mutant. This is most probably because of the absence in Chi2 of the additional peptide sequence (Trp377-Glu421), outside the N-terminal domain. However, we observed that the insertion of the colicin D CD within the colicin B molecule is incompatible with the pore forming activity (data not shown).

Identification of Key Residues for Colicin Import in the Flexible Peptides Flanking the TonB Box—It has been suggested that the access of TonB to the TonB box peptide (amino acids 17-21) of colicin B is mediated by the highly disordered, glycine-rich, flexible peptides, residues 1-10 and 29-43 (14), flanking the TonB box. There are just 13 amino acid differences in the 313-amino acid-long N-terminal domains of colicins B and D, and 6 of these are located either in the TonB box (two changes) or in the flanking sequences (Fig. 4). Because the changes inside the TonB box themselves were not sufficient to switch the toxicity of colicin D to the colicin B pattern (31), we wondered whether the different sensitivity of the tonB mutants to colicins B and D may be specifically dependent on amino acids belonging to the flexible peptides surrounding their TonB boxes. Consequently, point mutations were introduced in the flanking sequences of wild-type colicin B (Fig. 4) to replace the colicin B-specific amino acids with the amino acids normally present in colicin D and vice versa. The N4Y, R27L, and P29S exchanges, as well as the double mutant R27L/P29S, had no effect on toxicity of colicin B on tonB mutants (Fig. 4). In contrast, the replacement of Val8 of colicin B by a Gly residue, as found in colicin D, completely abolished the toxicity of colicin B on tonB mutants, regardless of the TonB box motif (DTMVV or HSMVV, corresponding to either colicin B or D). Thus, this single V8G change conferred a colicin D-type toxicity pattern to that of colicin B. Moreover, the introduction of a colicin D suppressor-type TonB box (HTMVV or still better HSIVV) in colicin B, in addition to the V8G change, restored toxicity on the tonB mutants.


Figure 4
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 4.
Site-directed mutagenesis of the flexible peptides flanking the TonB boxes of colicins B and D. The positions of the four amino acid differences between colicins B and D, adjacent to the N-terminal TonB box (positions 17-21), are indicated. These four residues belonging to the flexible peptides flanking the TonB box in colicin B (Asn4, Val8, Arg27, and Pro29) and residue Thr172 were exchanged for the corresponding amino acid in colicin D (Tyr4, Gly8, Leu27, Ser29, and Met172) and vice versa. Wild-type TonB box motifs of colicins B and D are underlined. Mutated TonB box sequences when changed in the wild-type and in some colicin B derivatives are given. The cytotoxicity of the in vivo produced mutant colicins was determined by halo test, as in Fig. 1, and quantified by spotting undiluted colicin and 10-fold serial dilution aliquots directly onto a lawn of sensitive wild-type (WT) and tonB-mutant cells.

 
The reciprocal mutation G8V in colicin D completely restored toxicity on the tonB mutants with the native colicin D TonB box peptide (Fig. 4). No restoration of toxicity was observed on tonB mutants in the presence of Y4N or L27R changes, but a weak level of toxicity was restored with the S29P change. The positive effect of the G8V mutation in colicin D was maintained even if one or more other changes were simultaneously introduced in the flexible flanking sequences (Fig. 4). These results indicate that the interaction of colicins B and D with TonB is crucially dependent on the identity of the amino acid at position 8, located on the N-terminal side of the TonB box. In the crystal structure of colicin B, the Arg170 residue in the center of the N-terminal domain forms a salt bridge with Asp17 of the colicin B TonB box (14). There is only one amino acid difference between colicins B and D located close to this position: Thr172 in colicin B is Met172 in colicin D. The T172M change in colicin B affected severely its toxicity on tonB mutants, whereas the reciprocal M172T change introduced into colicin D exhibited a positive effect on both tonB mutant strains (Fig. 4), which could be further evidence for a long range interaction inside the N-terminal domain affecting the TonB box. Apparently, a colicin B type T residue at position 172 is best suited to support tonB mutations, affecting the interaction with the colicin TonB box.

Involvement of the Central Domain in the Formation of the Colicin D-ImmD Complex—The tRNase activity was detected in the S30 in vitro expression system by Northern blot analysis, using a probe specific for tRNAArg(CCG). As several of our chimeric constructs were not toxic in vivo, the detection of tRNase activity in vitro allowed us to verify that the colicin derivatives were really synthesized. More importantly, this analysis detected whether the wild-type or mutated ImmD proteins, co-synthesized with the deleted or chimeric colicin D derivatives, could effectively form an inhibitory complex. As expected, no tRNase activity was detected with the wild-type colicin D-ImmD complex (31) (Fig. 1A). However, most of the internally deleted colicin D derivatives exhibited a high (100%; {Delta}1 and {Delta}3) or intermediate (40-60%; {Delta}2, {Delta}4, and {Delta}6) level of tRNAArg cleavage, even though the ImmD protein was synthesized (Fig. 1A). But only a low level (<15%) of tRNAArg cleavage was detected with {Delta}5, carrying a deletion of the first 60 residues at the N terminus of CD. This suggested that the CD of colicin D and, in particular, the C-terminal part (starting from position 445) is necessary for the formation of the colicin D-ImmD complex. However, we cannot formally exclude that aberrant conformational changes introduced by the deletions are responsible for the defect in formation of the immunity complex. To further investigate the role of the CD, different charged residues of its C-terminal part (positions 464-585) were changed into alanine. Several mutations led to a partial loss of the immunity, whereas two of them, the D571A and E585A, provoked a severe loss of immunity, without affecting their in vivo cytotoxicity. In contrast, among the five double mutations, introduced at adjacent residues within the N-terminal region of CD (positions Trp377-Glu421), only the E420R/E421R change provoked an important loss of immunity, comparable with that observed with {Delta}6 (Trp377-Glu421) or {Delta}4 or {Delta}9 (Fig. 1A and data not shown). These observations support the conclusion, derived from deletion analysis above, that the N-terminal part of the CD is more important for TonB recognition than for immunity.


Figure 5
View larger version (41K):
[in this window]
[in a new window]

  		FIGURE 5.
Mutations introduced into helix {alpha}4 of ImmD, which abolish immunity. Ribbon representation of the complex between colicin D (yellow) and ImmD (blue) was obtained from the Protein Data Bank file 1V74 (8). MIS indicates the major interaction surface (large purple rectangle) formed by the active site of the tRNase domain of colicin D and helixes {alpha}2 and {alpha}3 of ImmD. SIS indicates the putative secondary interaction surface (small purple rectangle) formed by the central domain (CD) of colicin D and helix {alpha}4 of ImmD. The catalytic residue His611 of colicin D and the Glu56 of ImmD, in interaction at the MIS, are shown as green sticks (8). Helix {alpha}4 (amino acids 72-84), located at the opposite side of MIS, is colored in red. Residues whose mutation to alanine provoked full loss of immunity (detected in vitro as 100% tRNase hydrolyzing activity) are in boldface capital letters. Amino acids previously mutated to alanine are underlined (8), and those corresponding to an intermediate level of tRNAArg cleavage are written in capital letters (residues Gly73 and Leu82 were not tested).

 
Although the CD is required, the N-terminal domain is not necessary for the neutralization activity of ImmD. Indeed, deletions within the CD had the same affect on ImmD complex formation in the presence or absence of the N-terminal domain. Thus, in the presence (wild-type colicin D versus {Delta}7) or in the absence ({Delta}1 versus {Delta}10) of the CD, no activity and full tRNase activity were observed (Fig. 1, A and B, and Fig. 3C), respectively. Truncated constructions with truncations ending at positions 377 ({Delta}8) and 422 ({Delta}9) exhibited weak or intermediate levels of tRNase activity (Fig. 1B and Fig. 3C), similar to the equivalent internally deleted constructions carrying the N-terminal domain ({Delta}5 and {Delta}4) (Fig. 1A). Likewise, the hybrid Chi1 is fully neutralized by the ImmD (Fig. 1A), confirming that the colicin B N-terminal domain, like that of colicin D, has no effect on formation of the immunity complex.

We also investigated the tRNase activity of klebicin D and showed that free klebicin D cleaved tRNAArg, and the in vitro cleavage was inhibited by the simultaneous synthesis of its immunity protein ImmK (Fig. 3, B and C). As in the case of the colicin D-ImmD complex, the formation of the klebicin D-ImmK complex is not dependent on the klebicin D N-terminal domain, which could be replaced with that of colicin D, without producing any detectable tRNase activity (Chi3B, Fig. 3B). When the CD of colicin D is replaced by that of klebicin D (Chi4) an intermediate level of tRNase activity was observed, indicating the possibility of a heterologous functional replacement of the colicin D CD with that of klebicin D. The introduction of an N-terminal deletion into the Klebicin CD of Chi4 (Chi5, Fig. 3B) had no effect on the level of tRNase activity, in agreement with that observed in the case of {Delta}4 (Fig. 1A).

The minimal tRNase domain of colicin D, {Delta}11 (starting at position 607), is 100% inhibited by ImmD in vitro, but truncated colicin D molecules, with only 16 (or more) residues (e.g. {Delta}10) preceding the residue Lys607 (Fig. 1B), are resistant to the inhibition (39). It appears that the N-terminal extension destabilizes the tRNase-ImmD complex and only when it is about 200 amino acids, corresponding to the majority of the CD, that it confers a greatly improved immunity to colicin D-ImmD complex (e.g. {Delta}8 and {Delta}5) (Fig. 1, A and B). The colicin tRNase-ImmD crystal structure showed that the major interaction surface (MIS) involves one side of the immunity protein (helices {alpha}2 and {alpha}3). However, point mutations to alanine introduced into helix {alpha}4 of ImmD (amino acids Asn72-Lys84) located on the opposite side of ImmD compared with MIS (Fig. 5), led also to the loss of immunity (8). We proposed that {alpha}4 constitutes a second interacting region, necessary for formation of the immune complex in the presence of full-length colicin D (as compared with the minimal tRNase domain). The present results strongly suggest that this second site of interaction with {alpha}4 involves the colicin D CD. Other residues of {alpha}4 helix were also mutated to alanine, each of them presumably introducing only a minor structural perturbation. Results concerning 11 mutated residues (of the 13 amino acids that constitute {alpha}4) are summarized in Fig. 5. To our surprise, seven of them produced ImmD proteins, which were completely defective in neutralizing the catalytic activity of colicin D in vitro, whereas the four others led to only partial loss of immunity. To verify that mutations in {alpha}4 did not affect the MIS interaction between the active site of colicin D and ImmD, we tested the ability of three mutated ImmD, carrying K75A, R79A, or E83A, to inhibit in vitro the minimal tRNase domain. All fully inhibited the tRNase activity of {Delta}11, like the wild-type immunity protein (Fig. 1B and data not shown). We also verified that a mutant ImmD, affecting the helix {alpha}4 (R79A), when expressed in wild-type target cells, was completely active in protecting them against wild-type colicin D when applied externally (data not shown). This result demonstrates that the MIS-mediated interaction of the mutated ImmD (without the participation of CD) is sufficient to inhibit the processed minimal tRNase domain, which is expected to enter into the cytoplasm (39, 43, 44). From these experiments we deduce that inhibition of full size colicin D (present in the naturally secreted form) requires two interactions with the ImmD protein, the MIS requiring the minimal tRNase domain and a secondary interaction surface (SIS) involving the CD of colicin D and most probably the ImmD {alpha}4. The MIS interaction alone is sufficient to neutralize the single catalytic domain presumably cleaved from the CD during the uptake.

Defects in Colicin-ImmD Complex Formation Analyzed in Vitro by LepB-dependent Proteolytic Cleavage of Colicin D—We previously showed (39) that ImmD prevents colicin D processing in vitro, thus explaining why colicin D is protected against LepB-mediated cleavage during export. The in vitro cleavage site of free uncomplexed colicin D (made in the absence of the immunity protein) was located just upstream of the catalytic domain (Fig. 3A) and was shown to liberate both a large cleaved fragment (65 kDa) and the 10-kDa catalytic domain (39). Here we have analyzed the resistance to the proteolytic cleavage of the colicin D-ImmD complexes, carrying a deletion or point mutation in the CD of colicin D and/or point mutations in the {alpha}4 helix of the ImmD protein (i.e. positions presumably involved in the SIS; see Fig. 5). The cleavage of in vitro synthesized, [35S]Met-labeled wild-type or mutated colicin D-ImmD complexes by a crude E. coli extract, enriched with added LepB (LepB/extract), was detected by SDS-PAGE. First we confirmed that under these experimental conditions the colicin D-ImmD complex was nearly 100% resistant, whereas the uncomplexed colicin D, freed of ImmD, was almost completely cleaved (data not shown). Under the same conditions, each of the single mutations (in either colicin D (D571A) or ImmD (K75A, R79A, or S80A) or the internal deletion {Delta}6(Trp377-Glu421) (see Fig. 1), when present in the colicin D-ImmD complex, allowed about 20-25% cleavage (Fig. 6A). This indicates that the stability of the mutated complex was decreased as compared with that of the wild-type. The combination of mutation D571A in colicin D and R79A in ImmD or {Delta}6 in colicin D combined with the ImmD mutation K75A led to clearly higher levels of cleavage of the mutated complexes (Fig. 6A). We conclude that mutations in the SIS region (in either the CD of colicinDor {alpha}4 helix of ImmD) lead to a partial loss of resistance to the proteolytic cleavage and presumably reflect a decrease in the stability of the colicin D-ImmD complex and moreover suggest that the defects in complex formation introduced by the mutations in each component are additive. As an additional control, we measured the cleavage of the mutated colicin D-ImmD complexes under conditions of limited substrate utilization. We tested several dilutions of LepB/extract to find the amount required for ~10% cleavage. We found (Fig. 6B) that although the colicin D-ImmD (R79A) required a 2.5-fold dilution of the LepB/extract to give 9 ± 3% cleavage, the doubly mutated colicin D (D571A)-ImmD (R79A) complex required a higher (4-fold) dilution for a similar cleavage level (14 ± 3%). As expected the wild-type complex was not cleaved. To obtain a similar low level of cleavage (8 ± 3%) of the free uncomplexed colicin D, it was necessary to dilute the LepB/extract by a factor of 20 (Fig. 6B). This shows that the free colicin D is much more easily cleaved than the mutated complexes, thus confirming that the mutated complexes are still formed but are presumably less stable than the wild-type complex and are accessible to proteolytic cleavage. These results strongly support the existence of a secondary interacting region inside the colicin D-ImmD complex, in addition to the MIS.


Figure 6
View larger version (42K):
[in this window]
[in a new window]

  		FIGURE 6.
In vitro proteolytic cleavage of wild-type (wt) and mutated colicins D-ImmD complexes in the presence of the leader peptidase LepB. A, in vitro synthesized and [35S]Met-labeled, mutated colicin D-ImmD complexes (3 µg) were incubated with (+) or without (-) a crude extract, to which purified LepB was added (0.025 µg) (LepB/extract), as described under "Experimental Procedures." Cleaved proteins were separated by 8% SDS-PAGE and analyzed by phosphorimagery. Arrows indicate full sized colicin D (75 kDa) and its large cleaved fragment (comprising the first 607 residues; 65 kDa), and the {Delta}6(Trp377-Glu421) construction (70 kDa) and its large cleaved fragment (60 kDa). The presence of mutations in either colicin D (D571A, located in the CD of colicin D) or ImmD (R79A, K75A or S80A in the {alpha}4 helix of ImmD) is indicated. B, cleavage assays of wild-type or mutated colicin D-ImmD complexes and of free wild-type (wt) colicin D (without the immunity protein) were performed under conditions of limited substrate utilization. The dilution factor of the LepB/extract (dilution) used for each assay is given.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
In this study we have determined two functions for the 295-amino acid-long central region of colicin D, which is not found in any of the other E. coli bacteriocins. The colicin D molecule, without the CD, was synthesized in vitro. It is not toxic in vivo but exhibited a wild-type level of tRNase activity in vitro, even in the presence of the co-synthesized immunity protein (Fig. 1A). These observations indicated that the CD is required for the killing activity and that it is involved in both colicin D immunity and import. The deletion constructions described here have allowed us to demonstrate that the CD, together with the catalytic tRNase domain, are necessary and sufficient to form the colicin D-ImmD complex (Fig. 1, A and B). Because mutations in the {alpha}4 helix of ImmD specifically eliminate inhibition of the tRNase activity of the full size colicin D but not the minimal tRNase domain, we propose that {alpha}4 helix of ImmD interacts with the CD of colicin D and thus they form the SIS. Because the helix {alpha}4 is distant from the MIS area (Fig. 5), the double interaction should produce a complex with ImmD sandwiched between the tRNase domain and the CD. This sandwich type immunity complex is resistant to proteolytic cleavage in vitro and should protect the producing cell against colicin D toxicity during secretion (39, 43). Nevertheless, our results (Fig. 6) demonstrated that mutations within the CD or in the {alpha}4 helix of ImmD are sufficient to weaken the interaction at this second site (SIS) to produce a significant level of proteolytic cleavage of colicin D-ImmD complex in vitro. These results clearly support the existence of the SIS, which is necessary for the formation of an immune complex during the export of the full-length colicin D by colicin producer cells. Our results imply that any stability defect introduced at the SIS (Fig. 6) provokes in turn a defect in the MIS, which leads to the appearance of the toxic activity (Figs. 1A and 5) in the absence of mutation affecting directly the MIS. On the other hand, during uptake of exogenous colicin D once freed of ImmD, the presumable processing of colicin D liberates the tRNase domain that enters the cytoplasm and kills nonimmune target cells (39, 43). If the target cell is a colicin producer, neutralization of the tRNase activity just requires the MIS-mediated interaction for complete immunity. Accordingly, colicin producer cells are efficiently protected by ImmD, irrespective of point mutations carried by its {alpha}4 helix. It is striking that although the minimal tRNase domain, from position 607 onward ({Delta}11), is completely inactivated by ImmD, the unusual presence of 16 additional amino acids upstream of the N terminus of the tRNase domain ({Delta}10) leads to a full loss of immunity, as shown in vitro, which is only restored when the complete central domain is present ({Delta}7) (Fig. 1B).

Import of colicin D into sensitive bacteria depends upon an interaction in the periplasm between the energy transducer TonB and the so-called TonB box, a pentamer amino acid sequence near the N terminus of the colicin. Although this interaction between colicin B and TonB is sufficient to allow import of colicin B (10, 33), it is not sufficient for wild-type colicin D, as judged by the loss of toxicity observed in the absence of the CD ({Delta}1, Fig. 1A). These results indicate that the interaction between TonB and the TonB box of colicin D is possibly weaker than that of colicin B, so that colicin D requires a further "handle" on TonB to allow it to enter the cell. Such a second contact point seems to be supplied by the CD of colicin D. Deletion analysis of the CD highlighted two regions necessary for cytotoxicity. On a wild-type host, deletion of the C-terminal half of the CD ({Delta}3, amino acids 445-590) eliminated cytotoxicity, whereas deletion of the N-terminal half ({Delta}2, amino acids 318-444) only reduced cytotoxicity. We show here that the effect of the N-terminal part of the CD, up to position 421, is intimately linked with the interaction between the TonB box of the colicin D and TonB. This N-terminal part of the CD, dispensable for a partial toxicity on wild-type cells, becomes essential for colicin D import on tonB mutants D1 and K19. More precisely, the 45-amino acid-long peptide sequence, between positions Trp377 and Glu421 (Fig. 1A), is necessary for the restoration of colicin D toxicity on these tonB mutants, irrespective of whether the colicin D molecule carries one of the previously identified suppressor-type TonB boxes. This result implies that there is a second point of contact of colicin D with mutated TonB, which is apparently distinct from the TonB box. Accordingly, we showed that several sets of double mutants, affecting neighbor residues inside this peptide sequence (Fig. 2), led to the almost complete loss of colicin D toxicity, in the presence of one or the other suppressor-type TonB box.

It should be noted that the role of the C-terminal half of the CD has not been defined. One possibility is that in the absence of the C-terminal part, the N-terminal part of the CD does not fold properly to interact with TonB. However, the fact that the Chi2 hybrid (with pore forming activity) does not require a CD on wild-type cells could be evidence that the C-terminal part of the CD is required by colicin D for some later step during its import, not necessary for a pore former, e.g. for the colicin processing step in the periplasm and/or the translocation across the inner membrane.

The recently obtained crystal structure of BtuB:TonB is compatible with a possible interaction of BtuB with two TonB molecules (36), which is consistent with the previously observed dimerization of TonB in vivo (45). Because there is no obvious TonB box-like motif inside the colicin D sequence between positions 377 and 421, a putative binding of colicin D to a dimer of TonB mediated by this site must require an atypical interaction with TonB. TonB may indeed be capable of such alternative or complementary modes of interaction, because phage display recently identified new regions interacting with TonB on the periplasm-exposed surfaces of the cork and barrel domains of the receptor FhuA (46). Other TonB-dependent outer membrane receptors were suggested to act with different parts of the C-terminal domain of TonB (45, 47, 48).

It is still unclear how TonB mediates the energy transduction process required for the penetration of colicin B or D into the cell. It is proposed that colicin binding to the FepA receptor induces a conformational change that makes the FepA TonB box become available to bind TonB in the periplasm (22). This event may be coupled to the release of the TonB box peptide from the receptor-penetrating colicin, together with its flanking flexible residues (14), which thus become available to bind TonB. Even if the TonB boxes of FepA and colicin B or D are considered to be the main sites of interaction with TonB, our in vivo study shows that the upstream flanking sequence of the TonB box of both colicins B and D becomes crucial for the interaction with TonB, when TonB carries point mutations affecting its major canonical interaction with colicin D. This is best illustrated by the single mutation at position 8 of colicin B or D just to the N-terminal side of the TonB box. A single change of Val to Gly at position 8 in colicin B abolished all toxic activity against the tonB mutants, whereas the reciprocal Gly to Val replacement in colicin D restored toxicity against tonB mutants (Fig. 4). We conclude that the Val residue at position 8 in colicin B is compatible with a low stringency recognition by TonB, whereas a Gly residue at the same position in colicin D requires a specific suppressor-type TonB box for toxicity on tonB mutants D1 and K19. These results suggest that the 8th residue of colicin B or D facilitates the access of the TonB box to the periplasmic space during import and imply that the access of the mutated TonB to the colicin TonB box could be dependent on subtle conformational rearrangements affecting the two participants.

As suggested in the case of the BtuB receptor (49), multiple transient interactions may be required before formation of a stable colicin-TonB complex. Thus, the changes in TonB of the mutants D1 or K19 presumably weaken the interaction with the colicin D TonB box and necessitate a compensatory change in the TonB box sequence (31) or its flanking arms. It should be noted that the first amino acid of the colicin D TonB box is an unusual His17, instead of a more conserved Asp17 found in colicin B. This latter amino acid was shown in the crystal structure of colicin B to form a salt bridge with the side chain of Arg170 (14). The equivalent residue of the BtuB receptor, Asp7, makes a salt bridge with the residue Arg158 of TonB during their interaction (36). If Asp17 of colicin B makes a similar long distance interaction with Arg158 of TonB, its replacement by His in colicin D must modify any equivalent interaction and implies that colicin D employs an alternative strategy to facilitate its import. Together these observations are consistent with the idea that a set of sequential structural transitions are required for colicin translocation across the outer membrane dependent on a cascade of interactions between TonB and the TonB box of both the receptor and the colicin itself.

A CD similar to that of colicin D was recently identified in two RNase klebicins C and D (4), but no other examples are known. The complete replacement of the CD of colicin D, with that of klebicin D, did not alter the in vivo toxicity against wild-type cells, but the hybrid colicin D-ImmD complex is only in part inactivated (Fig. 3B). We showed that the in vitro catalytic activity of klebicin D corresponds to the hydrolysis of tRNAArg, like colicin D. The first 120 residues of the CD of klebicin D are different from those of colicin D, and thus, this region cannot functionally substitute for toxicity on tonB mutants. Colicin D and klebicin D should be considered as "naturally chimeric" proteins. They are most probably derived from a nuclease expressing bacteriocin ancestor, as recently suggested (4), but carry a different N-terminal domain required for the translocation. Consequently, we observed more divergence in the N terminus of CD than in the C-terminal part, which is involved in the interaction with the immunity protein. Undoubtedly, the CD is necessarily implicated in the evolutionary adaptation to the different import processes, parasitized for entry of the toxin into E. coli and Klebsiella, and also in the efficient formation of the immunity complex prior to the toxin secretion.


   FOOTNOTES
 
* This work was supported by CNRS (UPR 9073) and Université Paris 7. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 To whom correspondence should be addressed: IBPC, CNRS, UPR 9073, 13 Rue Pierre et Marie Curie, 75005 Paris, France. Tel.: 331-58-41-51-54; Fax: 331-58-41-50-20; E-mail: zamaroczy{at}ibpc.fr.

2 The abbreviations used are: CD, central domain; MIS, major interaction surface; SIS, secondary interaction surface. Back


   ACKNOWLEDGMENTS
 
We are grateful to J. Plumbridge for many helpful comments and in-depth discussions. We thank M. Chavan for kindly providing plasmids carrying the klebicins C or D operons.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Pugsley, A. P. (1984) Microbiol. Sci. 1, 168-175[Medline] [Order article via Infotrieve]
  2. Braun, V., Pilsl, H., and Gross, P. (1994) Arch. Microbiol. 161, 199-206[Medline] [Order article via Infotrieve]
  3. Lazdunski, C. J., Bouveret, E., Rigal, A., Journet, L., Lloubes, R., and Benedetti, H. (1998) J. Bacteriol. 180, 4993-5002[Free Full Text]
  4. Chavan, M., Rafi, H., Wertz, J., Goldstone, C., and Riley, M. A. (2005) J. Mol. Evol. 60, 546-556[CrossRef][Medline] [Order article via Infotrieve]
  5. Kageyama, M., Kobayashi, M., Sano, Y., and Masaki, H. (1996) J. Bacteriol. 178, 103-110[Abstract/Free Full Text]
  6. Jakes, K. S., Davis, N. G., and Zinder, N. D. (1988) J. Bacteriol. 170, 4231-4238[Abstract/Free Full Text]
  7. Tomita, K., Ogawa, T., Uozumi, T., Watanabe, K., and Masaki, H. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 8278-8283[Abstract/Free Full Text]
  8. Graille, M., Mora, L., Buckingham, R. H., Van Tilbeurgh, H., and de Zamaroczy, M. (2004) EMBO J. 23, 1474-1482[CrossRef][Medline] [Order article via Infotrieve]
  9. Roos, U., Harkness, R. E., and Braun, V. (1989) Mol. Microbiol. 3, 891-902[CrossRef][Medline] [Order article via Infotrieve]
  10. de Zamaroczy, M., and Buckingham, R. H. (2002) Biochimie (Paris) 84, 423-432
  11. Pugsley, A. P., and Reeves, P. (1977) Biochem. Biophys. Res. Commun. 74, 903-911[CrossRef][Medline] [Order article via Infotrieve]
  12. Braun, V., Patzer, S. I., and Hantke, K. (2002) Biochimie (Paris) 84, 365-380
  13. Devanathan, S., and Postle, K. (2007) Mol. Microbiol. 65, 441-453[CrossRef][Medline] [Order article via Infotrieve]
  14. Hilsenbeck, J. L., Park, H., Chen, G., Youn, B., Postle, K., and Kang, C. (2004) Mol. Microbiol. 51, 711-720[CrossRef][Medline] [Order article via Infotrieve]
  15. Faraldo-Gomez, J. D., and Sansom, M. S. (2003) Nat. Rev. Mol. Cell Biol. 4, 105-116[CrossRef][Medline] [Order article via Infotrieve]
  16. Wandersman, C., and Delepelaire, P. (2004) Annu. Rev. Microbiol. 58, 611-647[CrossRef][Medline] [Order article via Infotrieve]
  17. Postle, K., and Kadner, R. J. (2003) Mol. Microbiol. 49, 869-882[CrossRef][Medline] [Order article via Infotrieve]
  18. Braun, V. (2003) Front. Biosci. 8, S1409-S1421[CrossRef][Medline] [Order article via Infotrieve]
  19. Skare, J. T., Ahmer, B. M., Seachord, C. L., Darveau, R. P., and Postle, K. (1993) J. Biol. Chem. 268, 16302-16308[Abstract/Free Full Text]
  20. Schramm, E., Mende, J., Braun, V., and Kamp, R. M. (1987) J. Bacteriol. 169, 3350-3357[Abstract/Free Full Text]
  21. Cadieux, N., and Kadner, R. J. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 10673-10678[Abstract/Free Full Text]
  22. Moeck, G. S., and Letellier, L. (2001) J. Bacteriol. 183, 2755-2764[Abstract/Free Full Text]
  23. Ogierman, M., and Braun, V. (2003) J. Bacteriol. 185, 1870-1885[Abstract/Free Full Text]
  24. Ferguson, A. D., Hofmann, E., Coulton, J. W., Diederichs, K., and Welte, W. (1998) Science 282, 2215-2220[Abstract/Free Full Text]
  25. Ferguson, A. D., Chakraborty, R., Smith, B. S., Esser, L., van der Helm, D., and Deisenhofer, J. (2002) Science 295, 1715-1719[Abstract/Free Full Text]
  26. Locher, K. P., Rees, B., Koebnik, R., Mitschler, A., Moulinier, L., Rosenbusch, J. P., and Moras, D. (1998) Cell 95, 771-778[CrossRef][Medline] [Order article via Infotrieve]
  27. Cadieux, N., Phan, P. G., Cafiso, D. S., and Kadner, R. J. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 10688-10693[Abstract/Free Full Text]
  28. Buchanan, S. K., Lukacik, P., Grizot, S., Ghirlando, R., Ali, M. M., Barnard, T. J., Jakes, K. S., Kienker, P. K., and Esser, L. (2007) EMBO J. 26, 2594-2604[CrossRef][Medline] [Order article via Infotrieve]
  29. Buchanan, S. K., Smith, B. S., Venkatramani, L., Xia, D., Esser, L., Palnitkar, M., Chakraborty, R., van der Helm, D., and Deisenhofer, J. (1999) Nat. Struct. Biol. 6, 56-63[CrossRef][Medline] [Order article via Infotrieve]
  30. Endriss, F., Braun, M., Killmann, H., and Braun, V. (2003) J. Bacteriol. 185, 4683-4692[Abstract/Free Full Text]
  31. Mora, L., Diaz, N., Buckingham, R. H., and de Zamaroczy, M. (2005) J. Bacteriol. 187, 2693-2697[Abstract/Free Full Text]
  32. Bell, P. E., Nau, C. D., Brown, J. T., Konisky, J., and Kadner, R. J. (1990) J. Bacteriol. 172, 3826-3829[Abstract/Free Full Text]
  33. Mende, J., and Braun, V. (1990) Mol. Microbiol. 4, 1523-1533[CrossRef][Medline] [Order article via Infotrieve]
  34. Pilsl, H., Glaser, C., Gross, P., Killmann, H., Olschlager, T., and Braun, V. (1993) Mol. Gen. Genet. 240, 103-112[Medline] [Order article via Infotrieve]
  35. Pawelek, P. D., Croteau, N., Ng-Thow-Hing, C., Khursigara, C. M., Moiseeva, N., Allaire, M., and Coulton, J. W. (2006) Science 312, 1399-1402[Abstract/Free Full Text]
  36. Shultis, D. D., Purdy, M. D., Banchs, C. N., and Wiener, M. C. (2006) Science 312, 1396-1399[Abstract/Free Full Text]
  37. Pfennig, P. L., and Flower, A. M. (2001) Mol. Genet. Genomics 266, 313-317[CrossRef][Medline] [Order article via Infotrieve]
  38. Miller, J. H. (1992) A Short Course in Bacterial Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  39. de Zamaroczy, M., Mora, L., Lecuyer, A., Geli, V., and Buckingham, R. H. (2001) Mol. Cell 8, 159-168[CrossRef][Medline] [Order article via Infotrieve]
  40. Frenette, M., Benedetti, H., Bernadac, A., Baty, D., and Lazdunski, C. (1991) J. Mol. Biol. 217, 421-428[CrossRef][Medline] [Order article via Infotrieve]
  41. De Graaf, F. K., and Klaasen-Boor, P. (1977) Eur. J. Biochem. 73, 107-114[Medline] [Order article via Infotrieve]
  42. Oudega, B., Klaasen-Boor, P., Sneeuwloper, G., and De Graaf, F. K. (1977) Eur. J. Biochem. 78, 445-453[Medline] [Order article via Infotrieve]
  43. Shi, Z., Chak, K. F., and Yuan, H. S. (2005) J. Biol. Chem. 280, 24663-24668[Abstract/Free Full Text]
  44. Duche, D. (2007) J. Bacteriol. 189, 4217-4222[Abstract/Free Full Text]
  45. Sauter, A., Howard, S. P., and Braun, V. (2003) J. Bacteriol. 185, 5747-5754[Abstract/Free Full Text]
  46. Carter, D. M., Gagnon, J. N., Damlaj, M., Mandava, S., Makowski, L., Rodi, D. J., Pawelek, P. D., and Coulton, J. W. (2006) J. Mol. Biol. 357, 236-251[CrossRef][Medline] [Order article via Infotrieve]
  47. Ghosh, J., and Postle, K. (2004) Mol. Microbiol. 51, 203-213[CrossRef][Medline] [Order article via Infotrieve]
  48. Larsen, R. A., Foster-Hartnett, D., McIntosh, M. A., and Postle, K. (1997) J. Bacteriol. 179, 3213-3221[Abstract/Free Full Text]
  49. Cadieux, N., Bradbeer, C., and Kadner, R. J. (2000) J. Bacteriol. 182, 5954-5961[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
283/8/4993    most recent
M706846200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mora, L.
Right arrow Articles by de Zamaroczy, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mora, L.
Right arrow Articles by de Zamaroczy, M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2008 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement